From f90b3ab54f2e24ed74ed0714ed297631ab5e9968 Mon Sep 17 00:00:00 2001 From: internationalwe Date: Sat, 2 Mar 2024 06:34:49 +0000 Subject: [PATCH 1/8] Feat : MLFlow Dockerfile & Mini IO, MySQL, MLFlow Docekr Compose - - #3 --- docker/mlflow/DockerFile_mlflow | 14 ++++++ docker/mlflow/docker-compose_mlflow.yaml | 58 ++++++++++++++++++++++++ 2 files changed, 72 insertions(+) create mode 100644 docker/mlflow/DockerFile_mlflow create mode 100644 docker/mlflow/docker-compose_mlflow.yaml diff --git a/docker/mlflow/DockerFile_mlflow b/docker/mlflow/DockerFile_mlflow new file mode 100644 index 0000000..4eb5baf --- /dev/null +++ b/docker/mlflow/DockerFile_mlflow @@ -0,0 +1,14 @@ +FROM amd64/python:3.9-slim + +RUN apt-get update && apt-get install -y \ + git \ + wget \ + && rm -rf /var/lib/apt/lists/* + +RUN pip install -U pip &&\ + pip install boto3==1.26.8 mlflow==1.30.0 psycopg2-binary + +RUN cd /tmp && \ + wget https://dl.min.io/client/mc/release/linux-amd64/mc && \ + chmod +x mc && \ + mv mc /usr/bin/mc diff --git a/docker/mlflow/docker-compose_mlflow.yaml b/docker/mlflow/docker-compose_mlflow.yaml new file mode 100644 index 0000000..5d4a5cc --- /dev/null +++ b/docker/mlflow/docker-compose_mlflow.yaml @@ -0,0 +1,58 @@ +version: "3" + +services: + mlflow-backend-store: + image: postgres:14.0 + container_name: mlflow-backend-store + environment: + POSTGRES_USER: mlflowuser + POSTGRES_PASSWORD: mlflowpassword + POSTGRES_DB: mlflowdatabase + healthcheck: + test: ["CMD", "pg_isready", "-q", "-U", "mlflowuser", "-d", "mlflowdatabase"] + interval: 10s + timeout: 5s + retries: 5 + + mlflow-artifact-store: + image: minio/minio:RELEASE.2024-01-18T22-51-28Z + container_name: mlflow-artifact-store + ports: + - 9000:9000 + - 9001:9001 + environment: + MINIO_ROOT_USER: minio + MINIO_ROOT_PASSWORD: miniostorage + command: server /data/minio --console-address :9001 + healthcheck: + test: ["CMD", "mc", "ready", "local"] + interval: 5s + timeout: 5s + retries: 5 + + mlflow-server: + build: + context: . + dockerfile: mlflowDockerFile + container_name: mlflow-server + depends_on: + mlflow-backend-store: + condition: service_healthy + mlflow-artifact-store: + condition: service_healthy + ports: + - 5001:5000 + environment: + AWS_ACCESS_KEY: AKIA3FLD32HPRN22NJQ7 + AWS_SECRET_ACCESS_KEY: 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 + MLFLOW_S3_ENDPOINT_URL: http://mlflow-artifact-store:9000 + command: + - /bin/sh + - -c + - | + mc config host add mlflowminio http://mlflow-artifact-store:9000 minio miniostorage && + mc mb --ignore-existing mlflowminio/mlflow + mlflow server \ + --backend-store-uri postgresql://mlflowuser:mlflowpassword@mlflow-backend-store/mlflowdatabase \ + --default-artifact-root s3://mlflow/ \ + --host 0.0.0.0 From 87160ad6a1df375f7469c170ee1d0d94a72e9586 Mon Sep 17 00:00:00 2001 From: internationalwe Date: Mon, 4 Mar 2024 00:03:34 +0000 Subject: [PATCH 2/8] Feat : MLFlow & Wandb Sf2f model registry - - #3 --- .gitignore | 5 + docker/mlflow/docker-compose_mlflow.yaml | 2 +- mlflow/sf2f/GETTING_STARTED.md | 120 +++ .../sf2f/data/\353\205\271\354\235\214.wav" | Bin 0 -> 1138800 bytes mlflow/sf2f/datasets/__init__.py | 10 + mlflow/sf2f/datasets/build_dataset.py | 64 ++ mlflow/sf2f/datasets/utils.py | 158 ++++ mlflow/sf2f/datasets/vox_dataset.py | 465 +++++++++ mlflow/sf2f/infer.py | 121 +++ mlflow/sf2f/model_registry.py | 65 ++ mlflow/sf2f/models/__init__.py | 19 + mlflow/sf2f/models/attention.py | 223 +++++ mlflow/sf2f/models/crn.py | 152 +++ mlflow/sf2f/models/discriminators.py | 110 +++ mlflow/sf2f/models/encoder_decoder.py | 87 ++ mlflow/sf2f/models/face_decoders.py | 583 ++++++++++++ mlflow/sf2f/models/fusers.py | 136 +++ mlflow/sf2f/models/inception_resnet_v1.py | 439 +++++++++ mlflow/sf2f/models/layers.py | 257 +++++ mlflow/sf2f/models/model_collection.py | 38 + mlflow/sf2f/models/model_setup.py | 99 ++ mlflow/sf2f/models/networks.py | 203 ++++ mlflow/sf2f/models/perceptual.py | 111 +++ mlflow/sf2f/models/voice_encoders.py | 474 ++++++++++ mlflow/sf2f/options/__init__.py | 0 mlflow/sf2f/options/data_opts/vox.yaml | 7 + mlflow/sf2f/options/opts.py | 92 ++ mlflow/sf2f/options/vox/baseline/v2f.yaml | 50 + .../sf2f/options/vox/sf2f/sf2f_1st_stage.yaml | 52 + mlflow/sf2f/options/vox/sf2f/sf2f_fuser.yaml | 62 ++ .../options/vox/sf2f/sf2f_mid_1st_stage.yaml | 58 ++ .../sf2f/options/vox/sf2f/sf2f_mid_fuser.yaml | 67 ++ .../sf2f_1st_stage.yaml | 52 + .../sf2f_1st_stage.yaml | 52 + .../sf2f_1st_stage.yaml | 52 + .../sf2f_1st_stage.yaml | 52 + .../sf2f_1st_stage.yaml | 52 + .../sf2f_1st_stage.yaml | 52 + .../sf2f_1st_stage.yaml | 52 + .../sf2f_1st_stage.yaml | 52 + .../sf2f_1st_stage.yaml | 52 + .../sf2f_1st_stage.yaml | 52 + .../sf2f_1st_stage.yaml | 52 + .../sf2f_1st_stage.yaml | 52 + .../sf2f_1st_stage.yaml | 52 + .../sf2f_1st_stage.yaml | 52 + .../sf2f_1st_stage.yaml | 52 + .../sf2f_1st_stage.yaml | 52 + .../sf2f_1st_stage.yaml | 52 + .../sf2f_1st_stage.yaml | 52 + .../sf2f_1st_stage.yaml | 52 + .../sf2f_1st_stage.yaml | 52 + .../sf2f_1st_stage.yaml | 52 + .../sf2f_1st_stage.yaml | 52 + mlflow/sf2f/scripts/DScore/__init__.py | 0 mlflow/sf2f/scripts/DScore/models/__init__.py | 0 .../sf2f/scripts/DScore/models/base_model.py | 60 ++ .../sf2f/scripts/DScore/models/dist_model.py | 326 +++++++ mlflow/sf2f/scripts/DScore/models/models.py | 11 + .../scripts/DScore/models/networks_basic.py | 268 ++++++ .../DScore/models/pretrained_networks.py | 181 ++++ mlflow/sf2f/scripts/DScore/util/__init__.py | 0 mlflow/sf2f/scripts/DScore/util/html.py | 66 ++ mlflow/sf2f/scripts/DScore/util/util.py | 452 +++++++++ mlflow/sf2f/scripts/DScore/util/visualizer.py | 216 +++++ mlflow/sf2f/scripts/build_demo_set.py | 85 ++ .../sf2f/scripts/compute_diversity_score.py | 28 + mlflow/sf2f/scripts/compute_fid_score.py | 230 +++++ .../sf2f/scripts/compute_inception_score.py | 229 +++++ mlflow/sf2f/scripts/compute_mel_mean_var.py | 55 ++ mlflow/sf2f/scripts/compute_vggface_score.py | 291 ++++++ mlflow/sf2f/scripts/convert_wav_to_mel.py | 216 +++++ mlflow/sf2f/scripts/create_split_json.py | 88 ++ .../sf2f/scripts/download_vggface_weights.sh | 4 + mlflow/sf2f/scripts/install_requirements.py | 50 + mlflow/sf2f/scripts/print_args.py | 22 + .../sf2f/scripts/sample_mel_spectrograms.py | 52 + mlflow/sf2f/scripts/strip_checkpoint.py | 53 ++ mlflow/sf2f/scripts/strip_old_args.py | 55 ++ mlflow/sf2f/scripts/watch_data.py | 111 +++ mlflow/sf2f/train.py | 854 +++++++++++++++++ mlflow/sf2f/train_registry.py | 885 ++++++++++++++++++ mlflow/sf2f/utils/__init__.py | 1 + mlflow/sf2f/utils/bilinear.py | 296 ++++++ mlflow/sf2f/utils/box_utils.py | 133 +++ mlflow/sf2f/utils/common.py | 192 ++++ mlflow/sf2f/utils/evaluate.py | 67 ++ mlflow/sf2f/utils/evaluate_fid.py | 82 ++ mlflow/sf2f/utils/filter_pickle.py | 35 + mlflow/sf2f/utils/logger.py | 93 ++ mlflow/sf2f/utils/losses.py | 144 +++ mlflow/sf2f/utils/metrics.py | 50 + mlflow/sf2f/utils/mlflow_wandb.py | 15 + mlflow/sf2f/utils/pgan_utils.py | 45 + mlflow/sf2f/utils/s2f_evaluator.py | 449 +++++++++ mlflow/sf2f/utils/training_utils.py | 250 +++++ mlflow/sf2f/utils/utils.py | 89 ++ mlflow/sf2f/utils/vad_ex.py | 174 ++++ mlflow/sf2f/utils/visualization/__init__.py | 2 + mlflow/sf2f/utils/visualization/html.py | 65 ++ mlflow/sf2f/utils/visualization/plot.py | 199 ++++ mlflow/sf2f/utils/visualization/tsne.py | 40 + mlflow/sf2f/utils/visualization/vis.py | 219 +++++ mlflow/sf2f/utils/wav2mel.py | 104 ++ 104 files changed, 13104 insertions(+), 1 deletion(-) create mode 100644 mlflow/sf2f/GETTING_STARTED.md create mode 100644 "mlflow/sf2f/data/\353\205\271\354\235\214.wav" create mode 100644 mlflow/sf2f/datasets/__init__.py create mode 100644 mlflow/sf2f/datasets/build_dataset.py create mode 100644 mlflow/sf2f/datasets/utils.py create mode 100644 mlflow/sf2f/datasets/vox_dataset.py create mode 100644 mlflow/sf2f/infer.py create mode 100644 mlflow/sf2f/model_registry.py create mode 100644 mlflow/sf2f/models/__init__.py create mode 100644 mlflow/sf2f/models/attention.py create mode 100644 mlflow/sf2f/models/crn.py create mode 100644 mlflow/sf2f/models/discriminators.py create mode 100644 mlflow/sf2f/models/encoder_decoder.py create mode 100644 mlflow/sf2f/models/face_decoders.py create mode 100644 mlflow/sf2f/models/fusers.py create mode 100644 mlflow/sf2f/models/inception_resnet_v1.py create mode 100644 mlflow/sf2f/models/layers.py create mode 100644 mlflow/sf2f/models/model_collection.py create mode 100644 mlflow/sf2f/models/model_setup.py create mode 100644 mlflow/sf2f/models/networks.py create mode 100644 mlflow/sf2f/models/perceptual.py create mode 100644 mlflow/sf2f/models/voice_encoders.py create mode 100644 mlflow/sf2f/options/__init__.py create mode 100644 mlflow/sf2f/options/data_opts/vox.yaml create mode 100644 mlflow/sf2f/options/opts.py create mode 100644 mlflow/sf2f/options/vox/baseline/v2f.yaml create mode 100644 mlflow/sf2f/options/vox/sf2f/sf2f_1st_stage.yaml create mode 100644 mlflow/sf2f/options/vox/sf2f/sf2f_fuser.yaml create mode 100644 mlflow/sf2f/options/vox/sf2f/sf2f_mid_1st_stage.yaml create mode 100644 mlflow/sf2f/options/vox/sf2f/sf2f_mid_fuser.yaml create mode 100644 mlflow/sf2f/output/sf2f_1st_stage-20240303-215640/sf2f_1st_stage.yaml create mode 100644 mlflow/sf2f/output/sf2f_1st_stage-20240303-220349/sf2f_1st_stage.yaml create mode 100644 mlflow/sf2f/output/sf2f_1st_stage-20240303-221106/sf2f_1st_stage.yaml create mode 100644 mlflow/sf2f/output/sf2f_1st_stage-20240303-225403/sf2f_1st_stage.yaml create mode 100644 mlflow/sf2f/output/sf2f_1st_stage-20240303-225533/sf2f_1st_stage.yaml create mode 100644 mlflow/sf2f/output/sf2f_1st_stage-20240303-225656/sf2f_1st_stage.yaml create mode 100644 mlflow/sf2f/output/sf2f_1st_stage-20240303-225950/sf2f_1st_stage.yaml create mode 100644 mlflow/sf2f/output/sf2f_1st_stage-20240303-230032/sf2f_1st_stage.yaml create mode 100644 mlflow/sf2f/output/sf2f_1st_stage-20240303-230142/sf2f_1st_stage.yaml create mode 100644 mlflow/sf2f/output/sf2f_1st_stage-20240303-230232/sf2f_1st_stage.yaml create mode 100644 mlflow/sf2f/output/sf2f_1st_stage-20240303-230355/sf2f_1st_stage.yaml create mode 100644 mlflow/sf2f/output/sf2f_1st_stage-20240303-230514/sf2f_1st_stage.yaml create mode 100644 mlflow/sf2f/output/sf2f_1st_stage-20240303-230631/sf2f_1st_stage.yaml create mode 100644 mlflow/sf2f/output/sf2f_1st_stage-20240303-231044/sf2f_1st_stage.yaml create mode 100644 mlflow/sf2f/output/sf2f_1st_stage-20240303-231201/sf2f_1st_stage.yaml create mode 100644 mlflow/sf2f/output/sf2f_1st_stage-20240303-231308/sf2f_1st_stage.yaml create mode 100644 mlflow/sf2f/output/sf2f_1st_stage-20240303-231542/sf2f_1st_stage.yaml create mode 100644 mlflow/sf2f/output/sf2f_1st_stage-20240303-231831/sf2f_1st_stage.yaml create mode 100644 mlflow/sf2f/output/sf2f_1st_stage-20240303-232325/sf2f_1st_stage.yaml create mode 100644 mlflow/sf2f/output/sf2f_1st_stage-20240303-233127/sf2f_1st_stage.yaml create mode 100644 mlflow/sf2f/output/sf2f_1st_stage-20240303-233311/sf2f_1st_stage.yaml create mode 100644 mlflow/sf2f/output/sf2f_1st_stage-20240303-233619/sf2f_1st_stage.yaml create mode 100644 mlflow/sf2f/scripts/DScore/__init__.py create mode 100644 mlflow/sf2f/scripts/DScore/models/__init__.py create mode 100644 mlflow/sf2f/scripts/DScore/models/base_model.py create mode 100644 mlflow/sf2f/scripts/DScore/models/dist_model.py create mode 100644 mlflow/sf2f/scripts/DScore/models/models.py create mode 100644 mlflow/sf2f/scripts/DScore/models/networks_basic.py create mode 100644 mlflow/sf2f/scripts/DScore/models/pretrained_networks.py create mode 100644 mlflow/sf2f/scripts/DScore/util/__init__.py create mode 100644 mlflow/sf2f/scripts/DScore/util/html.py create mode 100644 mlflow/sf2f/scripts/DScore/util/util.py create mode 100644 mlflow/sf2f/scripts/DScore/util/visualizer.py create mode 100644 mlflow/sf2f/scripts/build_demo_set.py create mode 100644 mlflow/sf2f/scripts/compute_diversity_score.py create mode 100644 mlflow/sf2f/scripts/compute_fid_score.py create mode 100644 mlflow/sf2f/scripts/compute_inception_score.py create mode 100644 mlflow/sf2f/scripts/compute_mel_mean_var.py create mode 100644 mlflow/sf2f/scripts/compute_vggface_score.py create mode 100644 mlflow/sf2f/scripts/convert_wav_to_mel.py create mode 100644 mlflow/sf2f/scripts/create_split_json.py create mode 100644 mlflow/sf2f/scripts/download_vggface_weights.sh create mode 100644 mlflow/sf2f/scripts/install_requirements.py create mode 100644 mlflow/sf2f/scripts/print_args.py create mode 100644 mlflow/sf2f/scripts/sample_mel_spectrograms.py create mode 100644 mlflow/sf2f/scripts/strip_checkpoint.py create mode 100644 mlflow/sf2f/scripts/strip_old_args.py create mode 100644 mlflow/sf2f/scripts/watch_data.py create mode 100644 mlflow/sf2f/train.py create mode 100644 mlflow/sf2f/train_registry.py create mode 100644 mlflow/sf2f/utils/__init__.py create mode 100644 mlflow/sf2f/utils/bilinear.py create mode 100644 mlflow/sf2f/utils/box_utils.py create mode 100644 mlflow/sf2f/utils/common.py create mode 100644 mlflow/sf2f/utils/evaluate.py create mode 100644 mlflow/sf2f/utils/evaluate_fid.py create mode 100644 mlflow/sf2f/utils/filter_pickle.py create mode 100644 mlflow/sf2f/utils/logger.py create mode 100644 mlflow/sf2f/utils/losses.py create mode 100644 mlflow/sf2f/utils/metrics.py create mode 100644 mlflow/sf2f/utils/mlflow_wandb.py create mode 100644 mlflow/sf2f/utils/pgan_utils.py create mode 100644 mlflow/sf2f/utils/s2f_evaluator.py create mode 100644 mlflow/sf2f/utils/training_utils.py create mode 100644 mlflow/sf2f/utils/utils.py create mode 100644 mlflow/sf2f/utils/vad_ex.py create mode 100644 mlflow/sf2f/utils/visualization/__init__.py create mode 100644 mlflow/sf2f/utils/visualization/html.py create mode 100644 mlflow/sf2f/utils/visualization/plot.py create mode 100644 mlflow/sf2f/utils/visualization/tsne.py create mode 100644 mlflow/sf2f/utils/visualization/vis.py create mode 100644 mlflow/sf2f/utils/wav2mel.py diff --git a/.gitignore b/.gitignore index 68bc17f..d20afce 100644 --- a/.gitignore +++ b/.gitignore @@ -152,9 +152,14 @@ dmypy.json # Cython debug symbols cython_debug/ +#Model weight +*.pt # PyCharm # JetBrains specific template is maintained in a separate JetBrains.gitignore that can # be found at https://github.com/github/gitignore/blob/main/Global/JetBrains.gitignore # and can be added to the global gitignore or merged into this file. For a more nuclear # option (not recommended) you can uncomment the following to ignore the entire idea folder. #.idea/ +wandb/ +*.pkl +events.out* \ No newline at end of file diff --git a/docker/mlflow/docker-compose_mlflow.yaml b/docker/mlflow/docker-compose_mlflow.yaml index 5d4a5cc..f284193 100644 --- a/docker/mlflow/docker-compose_mlflow.yaml +++ b/docker/mlflow/docker-compose_mlflow.yaml @@ -33,7 +33,7 @@ services: mlflow-server: build: context: . - dockerfile: mlflowDockerFile + dockerfile: DockerFile_mlflow container_name: mlflow-server depends_on: mlflow-backend-store: diff --git a/mlflow/sf2f/GETTING_STARTED.md b/mlflow/sf2f/GETTING_STARTED.md new file mode 100644 index 0000000..1241f5a --- /dev/null +++ b/mlflow/sf2f/GETTING_STARTED.md @@ -0,0 +1,120 @@ +## Environment Setup +Create python env: +``` +conda create -n sf2f python=3.6 +conda activate sf2f +``` +Our repo is developed with python 3.6, CUDA 10.1, and PyTorch 1.7.1. We suggest you to install the PyTorch version which is suitable for your machine, according to [PyTorch Official Website](https://pytorch.org/). +``` +# This is the installation command for our environment +conda install pytorch==1.7.1 torchvision==0.8.2 torchaudio==0.7.2 cudatoolkit=10.1 -c pytorch + +``` + +Next, install other python dependencies: +``` +pip install -r requirements.txt +``` + +## Prepare Dataset + +**Face Data.** Please download the high-quality face data from [HQ-VoxCeleb](https://github.com/BAI-Yeqi/HQ-VoxCeleb). + +**Speech Data.** Please download speech data from VoxCeleb1 and VoxCeleb2 on their [Offical Website](https://www.robots.ox.ac.uk/~vgg/data/voxceleb/). + +The speech data need to be converted to mel-spectrograms via STFT with 30ms frame duration, please process the data using our script: +``` +python ./scripts/convert_wav_to_mel.py --n_jobs 5 +``` +`convert_wav_to_mel.py` used parallel processes to speed up the mel extraction. Feel free to adjust `n_jobs` to adapt to the configuration of your machine. + +**Train-val Split.** To create train-validation split: +``` +python ./scripts/create_split_json.py +``` + +**Dataset Structure.** Refer to the following picture for ```data/``` folder structure. + + + +## Download VGGFace2 Resnet Checkpoint +We developed VGGFace Score (VFS) with pretrained ResNet model on VGGFace2 Dataset (https://github.com/cydonia999/VGGFace2-pytorch). + +To make sure VGGFace Score can be evaluated, download this https://drive.google.com/file/d/1A94PAAnwk6L7hXdBXLFosB_s0SzEhAFU/view to ```./scripts/weights/resnet50_ft_weights.pkl```. + + + +## Launch Training + +To train SF2F 1st-stage, where only encoder & decoder is learned: +``` +python train.py \ + --path_opt options/vox/sf2f/sf2f_1st_stage.yaml \ + --batch_size 256 \ + --visualize_every 10 \ + --epochs 12000 \ + --eval_epochs 15 +``` + +To train SF2F 2nd-stage, where fuser is learned while encoder & decoder are frozen: +``` +python train.py \ + --path_opt options/vox/sf2f/sf2f_fuser.yaml \ + --batch_size 256 \ + --visualize_every 10 \ + --epochs 50 \ + --eval_epochs 1 \ + --eval_mode_after 100 \ + --train_fuser_only True \ + --pretrained_path \ + output/{sf2f_1st_stage_experiment_pretrained_model}.pt \ + --learning_rate 1e-4 +``` + +To train baseline voice2face: +``` +python train.py \ + --path_opt options/vox/baseline/v2f.yaml \ + --batch_size 256 \ + --visualize_every 10 \ + --epochs 12000 \ + --eval_epochs 15 +``` + +## Visualize Training Process + +The training process is visualized with [tensorboard](https://github.com/yunjey/pytorch-tutorial/blob/master/tutorials/04-utils/tensorboard). + +Start the **tensorboard** server: +``` +tensorboard --logdir='./output' --port=8097 +``` +(Optional) If working on a remote server, mapping the remote tensorboard server to local: +``` +ssh -N -L localhost:8000:localhost:8097 user@your_remote_server +``` +Visualize the training process by opening `localhost:8097` (from local) `localhost:8000` (from remote) + + +## Test +To evaluate 1st-stage SF2F and baseline models: +``` +python test.py \ + --path_opt xxxx.yaml \ + --batch_size 1 \ + --checkpoint_start_from output/xxxx/best_with_model.pt \ + --recall_method cos_sim +``` + +To evaluate SF2F with fuser: +``` +python test.py \ + --path_opt options/vox/sf2f/sf2f_fuser.yaml \ + --batch_size 1 \ + --checkpoint_start_from \ + output/{sf2f_fuser_experiment_name} \ + --recall_method cos_sim \ + --face_gen_mode naive \ + --train_fuser_only True \ + --checkpoint L1 cos R10 epoch_2 epoch_4 epoch_6 epoch_8 epoch_10 epoch_12 +``` diff --git "a/mlflow/sf2f/data/\353\205\271\354\235\214.wav" "b/mlflow/sf2f/data/\353\205\271\354\235\214.wav" new file mode 100644 index 0000000000000000000000000000000000000000..dab0b2994b02d24dd9c38c485f72b7e45874879d GIT binary patch literal 1138800 zcmeF)1(+qrku~g|8OzLT21#aSW|^58CEGGHGs(=%C^IvIWSN;6B$=5RH2Tlm>evnE zR^7fm8hQ8sw(ohKiG)b3tcZ-rte$IK@p6~D?Ow-Qx&9@uciCIt>7KhEcV%T|%gWZ3 zhrTa?UvFJG(#mmHu5!g|U+1EUUh!&|yZRL`eXYyvlJL3DdG7N{x#U%kk=QRD`JP|C z@Cm!0@!V%h+O;otont5MDmTCPtDyHL&P-cI zf{g7gDuZ|YO{I-I|I_h9^Beon(|Wl0r1f8=PtnddhPNO3uaN(*Dci>)tMQR%N8*PZ z`tO7OZKSn6p0Qlzjp;{3{GW#|kE!+-mHx+}|M+jUR~^0F6kpGu3s+r5@l|+^_PXYH zb|ii3`7e5omDlOA?e;Po4NtyoYl+_NMfsIiGN9s%=v1Bh{6F!V>dVgP#}QR;^{wXn zh@jQRvUIKwz5EDc+I7vSp_D=K8+OeBi0_Q*~GWM!L(&ayatJe@B)4?>R^QZ?zUrm6@Ys|E9*$ z(?WUHgs0kFRPTDSsAT+ld>hHHI*X2F z=hBO>@^^%m-X3Y+lTPhv`yPBMJeHZ_S(Mhx_rA=9i+@p_mB&^oRN5S^dgtg|xw-Ue zcZ&Y~(0Tmr(0|1|=i;2Hc9%(;8&6N4o1?k(UUnnlx%zs#b<*P%o?Aat>E4N+UUlqj z+S@py!u|ao+vs^y;q~N7m6_tJyq!UJcKuWJ{clOz8S~ZasB!hQ^}*-@&=c~W%R(}^zyuw9`N4W}D9$LR%?=s=Bk2@9rA*Hvj~FIQn>8Q0W~#;5)cZm+B6t!U4yusZ5*`qp0q7MET; zQ?%-wqBYh|i?7o9JXN^z@K&3`{U^QpCQsQ~>z)JNu`OxJ8P{KfzXYhil3Zi0dFXR6 zRloMsx9Epw(N9@*k7JSE%Zd9{+T1#;{KL5aYy342=z71ZqiAnSxPFyhb3RgkL%sh^ zcj5kb)yF*X{+bbdebp~xtLN+adA1#%_n(75HHd?}%I|4?eOMz^FP>gc zg?n9NeN#NWf3*&cq1K$S8Gntt=0%!cJx}4zSMyo_U%*jT<}vm^|b0& z-plXxu`r;n!2jqqGNm24`pgD$F}5+KmGDt#%J!i7kXX~SN;_3 zd=)M}?e(yD%@1nLbc}fw@8c8ptGqdS7}vl4i+B8u^E4N)=Tpz4p5a)Z_*|bV9ZzpZ z+^_QI=wV#{%2#PqbgCYFz3rVHHZL8${+_-g;a*pbx2ILO>K@-G@yhG8%B@GI%2Yqh zb*<$-?>$ZZy-(F2>Ai00g^~j$n>ww$&MSYBJnVU+eC@KfhfCI$T&;FH)rUR?VQC{z z=j-{xe&)00Pk1%4{nOiM$^Y55dH|oFI{bvIPsLMIAN#xs_kNGzUVrsRTIZ{}dw!*L zTKPqLo^YjiTH~^|$k#=3xAI164>0b}@RYpn^;X%QN0yg7uQgt@o?m|_iXV^k{xhEO zXS~K=&!(^Y{;XSi)$FX1Hz3;v5v0Q&Q`lIzENp@%GdLfRy^Zc@BG!K^40%(9u+TsJX1e))HCYON4TEb z996E$7SC8m#VfzcjO8l7=(2qHJMTuqwMJ_mm!VZ=j?UH5=YHhrdBPjb$FDzqPHW!# zbF8`PeIMJZc-2>BP-*=)#Z!DePq@&_3mb2(_j(`13*&oG>$>RpeoUnoPvv!5?Tl34 zs?Q@e##*O+{o$*;n)jLy_0?~zua~R(q}9(HP5tEey2SC-So%0CTzRNp)uEs3^%bvj zRaedHs1te|Pj5>d^{a7@?TVX+TEl&=D-G@Cs;;T|lQ-6Z9iKZCI!qb{%h)N|b^9mi4SE8W~2QNMBS zHaZsN{Z=33+CSCT$`h}#^sz~=a>D&r=|$yL9i4A3j8|K=CW_W`^$gh#M_mVatc_YH zXswmX8}mC)@k-Nn&2RO)#@A_oM=E`azo(7wsQ0nI_x=5y${Tsdc)jbrZ_9+M+(>)b z8e>lr_8ZfBdX?J|T60%x3iYeHs@;ljlve#agHE+w{VO`fvm%d)*riR~@LbJw6p4`>-fJ zRers+vH$C>6Zw9{S8IA%x{MA{SJglEPkg;GEt@}8cTdxYs%z1_%bQw@wT9O#JGN1A zeHrIL-g@=*{Hm+dKE|nVwK>w_?R8bS@`_eET4_~Z#XGIMIa<8E-U?Tqvb7H9)>-A_ z^{YBY+UIZ#*K-{yKELWu(O!3@P0<>=KI(t(>o~SW`!(fieA?;tj&;h{zj2Jxh~0_u@&vloo=jn<&Sk#Ta{jX%GG{l zv=2txK$)pu)x9Hhsy=Djs=n3us?SBM?AUIlFQQckT0iv&ORM?pw94SAHiiAFo=WfZ zFpk6it-5PJGmfo~+5AtTY`3CQXHD`deY7Pjy~jMXn`=A)inpS#Y#C_I+0x}+B>t<%aIskW+r<5((Q`Eyhq zRd2Of@lNsi^|azgd-{6e!)iM0-;uZ4Q0Mqgjq!T_dizxlZ%@~b_?Ta5BgH@d%wf$d zTE9x`>#VO$>0`X}R(Oi1>hA50{Sw!gq7^Q_agN8aPQ|;t>fw@YHTOMlQFyA2o>qOV zKBJX}ufo!6-p6rPp1B(5ywB}ic+vh&nR-q|tDcHiyK`+9U#I=~_GesUt8w?a>S6g) zzbaR>^Q>3KbFAmwpJ{Km!d3rBtG!Cc-`lNch_85rE4`>RzlwJ%y~-AqR=-M{qs7;0 z=k4hgmS1C^qrHx0_^X~N-m$K!c$FWYTcuBxIlO6ebys?))mF{nGPKI5r?0^tt~J_a zQEzt)tFzLoo?gdPxcI6ZT79c=Sd-&>+~4sEm%Jz{eJodTbyodFrPZ(aidK53%1!mP z(rZk`Q}Lpej`lR=wNq_(T4SxW-k(ZaL_2?ltG~TJ^3+r9jkMaRIx1eY(yN|I@3iw& zSpL}7RNOPBPx@JG>6hZKw$yEmTzt+^>Gy?(-#K1HYIr0Oj?mK)2=)cA_Pq7N%aqZdW$USD(CZz?Gga5f)8jQyQ?$>0cdE@k5g-? z`ZkskpZayaF<)M_Rpo13)z?n@SjKSY8^a@S#fz`nL@TX$&`K*Rt*>qAJMtUr>uoL* zuJYRMb6b6>{GydU(#oHrczfN#y?oEB^r?EL(iV-e^H1^heBqI&o=@MeRa()WCtT?x zt-8lH#dq}A>#92XT-JNz{V3VqX}ve~j#k;Kt7va$3>VL`R2$W=Ia>Xy_gR_s>c_L* zuj-iMtGqdC-1Q9mv#sYpbtY1IMSD5nN*`&RnMkiP)pqYkrQxaY9IfYOd=>7zz20hb znR<)2>Z-b{zRI7XeSA~-Q)$)CdbG+GpZ?7KdY`AlRj$@e#dkKX`ulq!ZEUCb=4gL! zrg(b&W4QB=?e=)_?Pz&*RUN(Fsc=0*&s2Eo8TGWOI;8dXHo~vm)SQlEuXxF?QPz(6 z#aFcWYi}-IdCK;%c>neIIR4Hz=85;RcstJ+H*fuU)O*%x%U&E`cqV%aP@bN))*_l>eNn!7tv}D_3QB-uKG7htKP8>H4pWiE5El{X(O$&a}@ug zyvnb((J4&TQDuwb8OzROn$K6yQ@^G=0puH;p?<{PbNRe4p1Hs->;{W09j$Q#Gq z<0ThLeiW5ne8p3AQTo`IF|Kc8+FZQa8ENNNU(H+1skHtp{$7u?9v3#g8viIiD}O9A z#?{-~u0HiXRu~`t`t@?lhSgQ;2Q4|%Y0s~4&+p;NM|(f3Kb1B`i)W-G@0efXN5}cB zd9Srs&!^%eohn~xopD4#X z*y<--^HFlD$47ZpJVo(Vxae3$Twc}N%k;FKUUiN6BhMILRBq%Q`#F}acv0nRT?qI1 zIH$$eY0Y`jdNxHXec3Xe{oJqC#Uh?6TkRH=Ug1Tw`niZ_uCAWm!&O(+Jwv?xNu9Y#4x7X3b`cgk*^{e=L{Vcx2N=^UUpLKBCZcI~6XvsE>V| zeeR~_XH4tyKG)-1_V%kiX%#No>!>ue%J*=k8DoVP%}>qYdUUL#=5UIRyyEzZ&Xwuu z`abuoadn;^u5wfT=xK9&eJ;9AqPu#$>#0gtb_~zeSA3mT8>J_UPU*|yRlfLoT|HcV z=;N=jJip>Wr|Rtaou{``{V3Y=gvUCj;=Nzh&REYF$LCkiw6BpKmfzz&T<=y-+h};I z{i=8BJyn1G^uOj~42zGy;vH$9n|cmYwDVORV;kbtzeTi4h|iUi-k*gy{`!g6uk(%J;u-14 z*Vp^pn(t+*zUogOSD&BC?{wtt@#3pGiuUx*ztM2jvomPT!A9t!cBjTN@>g5^Sxkk; z=h@@MUuk`>(!+gEG~OqeiWh&cqlYWMXysKuI;}o-x?X2uEe9y16 zdLColop&nS``+6s-l?=|v$wTgJyqWv?RC~zicYP+F|ByfMR`+vU3bmN_%Xr)h4byPc3bnbcebmhih)ith*Mdf;#;;Xc(zs68wE83sU z)V^a;S}!YHynX*WhKpyUBVWZwzOl`zade)_A8E};&Cf_{{x(9#I+lr7KNitm&&XGK zQ~j*9P;}&}^pTc)E_uBpwCX^2q^^E-#AICJtYkn5du}sC4sqh@#k$TswtMWH8 z-jTQFtO+I_LT|rjKLm;~vAs zJJzu?;>*_E=WA-5V}6fM%}=#EM{5q|@)xD|dFlO_3wQoS^;TZzUoTu`IvvZ5{TlPf z_{e{F<754E@jiwgu4lCo+RN{(aJ64DsngO6hllq1gw<7cO`U;Pd?W3Q^^Wlx*D`c0 zTm9_)DIWEWb3gU$7x9dF;){M8;h$R9%hp|d>Zow1W8IZEmtOf(blEW#Up>c)7p?T7 zl|DzuzRdAedew_o+K$p**BG|G#j`?G}_~!av`BQYPtH-DKtBv)>xhTKtC_2Ym&#UhPD-Esv+#D_5x%w-; z_?D@&`ch?!_B!Um#k(GzDqmyld2{tF;^}$9Q@@d?+Fs8_FTO>z>R7}hulS12@mKyx z=lFa2IL?aWsqobN%%#kE?+VX7 z8|j{D|4sF&#?joo;+vx*zxWjI9N$#> zn6~KoEt}W*>OCLV2A;|*{_0D`cNVR7itb3gi~3*nPVw}%YJ5efo}IK>Cum>4WwVq` z)oJC;(W!doczb#;SK;d4*#AnOqU-UF`D5Ii^|Yxr=F+O)Xz}#AgvVdcSEk384Og9g z4b)ocdU6bpy0*thUh$s4Y`E%~>QALj^}F&$s=l7q%g%*2%CDYYW(+UWjxzPD=To%O z$9gMX`Dm3{mM&T|%hrK!qkWiax7wS^lQ#ETukFeo%U62MW6?@4{<;2Cdhu1cDOz=} zN5`^L{3Fj8?`7(JT0}el&I;?(SXb}onAhWDUwhuN;ojC5o~vVyuhQqr9iH?W>s%d` zkIwPTrB}bFXsy{=!;9Xp;;nirj#l}e*3&D0s{WoghR6DT8eX9C)l=gP!!ejg5HRtQmsj^dP>gmr^xc?UAb-qR6YPWcLot3tTR(Z7eDs67= zq}Q_-BxTc^LDjo>irdyggn#Bkepryk5U6zn7T`7av;l)zg*^@saj&HKsYb4DS?APuoa%t{>8C-ivnLO4|sn_i?0EM^Bel@shcn&gE75 zSZ6OQTzqKd^}W7urH^!~>>SV7k6v~RtE1AC>+>zV=(+Yaw}@|1UazCVwYH0PUu=ad zZ;n=7Q?&P|=U4hT2Nkb*oukEv&hgBpSJ~=A(cVAdN}r-5?^Im5!}F_l$2w|UwU(@_ zTHCcwD{qcg9cc0NvwLnUcdl+c zW4ZO>bNYKxdhG$wMSOF4W1sri#&C_b=$K#e%3qe&bMAE6vWxhtKT~v3xkY(XbxG^L z8b?nPu5@(F>+!iZ#`Iohqv7hyM#@#0MYLquBA&|YZC6^SJ%5>SFI(YWuI8_&jo~Uc z(#}_Tl~(nlQ@_fe$~(MiRrexV@Ax7fdBulL$%`&)b|gGi=Xz(QzU zoob_}9ns-&-gZWNW1Um+YNyY|&Inh%)n>)3?usue}9xtBi<4F75jCqUVRjzodosm}lB3gCrOj>K8(_Vj1@8RO ze>qwn?WWj7M8@Kjz0NC@ny@_-vrty-ALQLFMXUf4?S(Y zaFs!ex2FmBXE)|e#f!h^E&I&p`q1mzk?^wZSKk-$cb+3EY>XSJyX4BUbeX!Btqb3j zJY1#?JYze{mYd@n>-*mnpX+l^pIYBNUw9GEm^YT)+3~6Qt@SWRccjc*U57V)sxRxM z^)@?i&23NHnc=Fx(~Z>M%k}WMW~Sm({D((gtFPLrdeHIL+uf0|_A>psEfel#=fb^z z6)xV2FGH&gI`ymksXS?8e#O_LV||^!&r$8Kd)j*8DpR!c_jblUR2)zLjccRF`x;s{ z+{-NzY?snFqyk+dSUYGD#rt%lHgJ;aI_fLA^ z%kbB;U61a}y4Gu}^1FN$K0G?9mmTY!idP>-IyVpfdG&R(OnT?(;iSU9R+SFDJdn$M962YCc9HgnN67>gn&>TzD)$#W(Wwc<;+pxOgkCo=3$;TKPMJ_O&ooU(c`b6yJ`dP1RRv zMHkgO&dXGMQTfW#R)u$j?nt{+W#-aqeaz8uJ-?jPWXi_MbM#Iqvg^ll{eL1uc!Fd zqhme2j*W)r)|>S1XYYUObVtIyt{n+iKYQ6qo1*xp(&y5~c^mo0eDRTIE6e)Y$R)B|$>K&iOnC=;l&v$N};G4?p`V~mgO|2yNoFFU*4sporm z)8^KAuXjhru_NWH%%b-C{8o5w?Nt60ohwu6z09KiR9@#7uJlE8EHf1!%TMtvlUC!} z5xPkBPt~`Kewiw_v(su``W#joD!s>rs}Ey(#jCDbt8=v8tGWD2ulg3z;zbwzs*gn% z@lMsbsNABwMeWF|vS{b+>B2Sl=-e9WWjcQkcb*>ZZI9v3r>>3u#`@;sQ}yB<)BAXq z4Ug~XRJoDo@Wy)^;~1vmbMsQ^wJzu0pGq%U{GC=_(Vo}C)#ei<;QdUMe8cUE|+ zAC)%MM&%Wq;+aY-{$=P?d1=e6vAMF9UhP-BQ+Yiutj#fPj8~nDXswODZiL6*m_HY< zHHpsgtd~BvJ;v87KbK#9o64`eWoX%1MaMFgj#l1Oo8!2ac}`=Q9`EJHwb0|8Zz{Zq zS9@cb&M$m;e|=09t~x6|NBcZ?o=UIhQMA%K?_9We(f;e{Jv_EidD8nD>FcD*bXw)m z{_E+>gnQXV;hO7>&_(qvo2R|v9miVfMaS`0I@2>ZP-peftSJ~p}{p?|DYwp)&!^l&4qh9!*j8~tgXzzdJFQdnL zxt$fR_UGt&b2wLSNBTF)pGE6os{XNjUt4qGUS^J`r!V5~c{>vB?W`AGRDaFQA|82D zzxB$D`C~ul^2hY4xc*F~OY6Upr{@We`D6K!zv6Ric}y>UwCDBosc`3AHeB!D$Ujy8 zdTGnf%M{;KnrA93xmxR0XO*pZ<@bKc>*-xK^>E3qMYQt^7w_D+ zij_Y{$8tU1%Z%+F-guvn9l4RcB9IHr(s%;VNHsPtoG(bc$~(t7_Q~aG)*`8M6v2T^Wh*o`_;_vY>+}G$b;q`P& z>9)={Rqxo3G2Y9Jc{>uHnzvrhRQ^<2FH_;-8{?IZRvy~ZJ?HV)c`MxKV=i30ozC(0 z^mNd}#yfwFZHi8<)hWJSUk{J%ZY198Sri`Y ztT`+?%89A@r~47W2-i+Y}H%wqLq$TTBnuYpH&Z6{#d@+ulE=&pKzz; zTSkxEaFmt9aMYEft{iO*M_)Pm$}t*_xpEkewQ{VLV+Y4+IPS`S;dm>@TRDF4AHfNO z69*>=PPT@Vubh146u~Liugi~Tz2KME0+r{A6#J#6-EnqinsG$apf>n*-qtLsX<(R)j3l2kEnj4`fUV8 zWsJsp@xZvj9GDmLW1h^{`2%xj4$Z0gHP7b!EP-d>S$HPk8F^lwo#*Jedd?>etbr2+ zCkU(y>%%&+hO8%R&Dygrtz~kLJ@wb%&%qyq-v_@9eiQsM_<8Wt;77p^f`fyDg6{_3 zS;Myu{nnx12)-74Irw7m`QS6br-Dxe9}7Mbd?+{|*gyC{@V?+Z!MlRD2X77D6udro zZSd;gRRdmm=wW!pq5p-KANuk`Ul!~eyf}DKL&aZs=nD^hUhwST8Nt(nr#6T`eGT%T zxdvhBQ~c7#a?d*SS*vYm3tkZD&r5=r2Ks6Q#$|kk*Bp8njNAC%7`!=&5N4+rMhTz@(+_n!;C5O^k@jpyXKeJ}7le-`{I_+9X)z&fxV=uWcF8X?z6 zE%J(-vW~1L>&hCl)~vabJ=P*QV{MWx)~j_)epuhuy0z~;D4F0L@veApyhGl*D+gB( zt`l58xM6Uk;KspCgFSw0=uR5g3Q(oEI zHmIW@{dU3agF7@x!=r8e&>#J}Nq`o_wXOd*2(BAkJ21AZ1mq2QMShh0EBUcUlOg74 z3SF+4SMr1`A*0CI#0(2dr_--9Vx@#{l>y(%ky zap-@6>?l}gKMW2GtV8Saqk*++9ls;6zOD1O1lGK@?;Y?)cvrmJ-XHIh_t!i3#NbiE zLxTGTcMEPG+%mXXaKqrb!L@^HuHhPo9)_zQdi6uE8eB2BY_Lag;oyA1*@H6%rw&dL zoG3V6aIE0y!I6V)!Afw*_CvP+J@|9*hu}BCFN2>4KMj5y{4n@laA5H5;2Xi$g0BQ$ z4$!X#(!UjaH#jIbI8g5U0a_6MLGYvCC&AAG?SZf~dD;bikp8RScY(hCGtmECf};h; z4o(o9EI3tghQK_WH@HY}so)BMdAnY4)4)8N@4E)~2_6(YJa}C2lz<$3e&D&0dE}gD z>iLpo?+x}3tcA}8Uk&IM>xC}zvku7_Yu28{dZr((bMJ$9=W2oX%DZ-pz&m@V;O@bF zg9itX44xP~GkAXRvf#DBn}fFp?+V@%yg%4K_;B!v;B&!Og6{-B2!0m)Cir9U*Wi#~ z+m>xxjv5>%I8ktl;MBnxg0lwa2+kdxH#mQAq2QvyC4x%^mkBNxTp_qpaMj@I!8L78y@uOvISjYka=R^e2<{l%DL_Hm zh}&Od)Rs$axp;8l z;C#V3gR=!^3CTzq8l2FQ+@{Q*zdi{c0a(|3#9P%(S2tFL_ z7rZ-oYw)_@Wx)%AX9Q0Q9veI=czE!z;Gt`H(DnyyzkhJ=;O@blgWCtU4)zXi9_$(1 zB)DO4{op#mwSsE|R}ZebhO2Db#+%+(sdjs|JF7Z|o*@5A=Kp^C@3;Rc_($OR9wj(V z;JJJLXAS5*>w+G-O1$Phb9Xud-Sn!zO$pP7KZPMG; z=^Fz2{oMh*ZynpGeku57z^-6tus^&9-U<6mHVqrcUXm?D_S@fhm+T+ufIWgs2KG>W z?^rrv@9fd;9NZ^(Xz-Z8`}^#``|Q2;etXXk2tFQ=0p!6~12TfVAUntrGG%*kq%B9< za*W`30l9y&V7K6O!I=YcpS&mQ$vm=&<0Z1AYy;lV?L z2M6Z+zQH|%I|sK3=uUd=dI5b#Z_&>e3_MG^hhE<`pmRk>3f>pkhrB*`MexGFUgZhFBLe#6o&h_AUb$vKr`QMC=hG={ zE_#Iz$X?#Q$vb2Z#fGcBy!XpK{ttn@yZsh>&^!3;;46W5ldbN3_3pB(*}(R9_Y7_m z+$6BSF8xAgTr#*&aIU}}diS8~m#v3xUEMR=f6_C53+zYf8v9lNMC6~!?vK($4pf?EUjjILZ{w0Hp2ImXT7MwoVJve!A(%{6w z34`MY#|n-Tc;Wwc$lnh6L-4EMC&3Sb?*-ovei%smS@82U6wkqj9DK+@!GQrPuADMI z4t^T^GWc!q`#>AN3w|9)|HT@FwOxFFI^<7>{3ZB%5HsYEZQHkPKU#3yz!-K5&Jdg< zIDc@lfDWV+=|}THhoW@h<$^te3kBy6&K{g4I8(!!x1V|YSp#+AQSU{9O9b?)dgy0! zPe;;+_X!>nJUV!MU{CPm!1H@r@a(`deRc4rfX;hg@WJ3i!AAo6Pu!ZY-w>Am$>4JV zUC5>?8_HgVtz|v2qpU4@kj-PSV_lvzxL9D_UM(m+XI=JjrQ&*(R{G(VjEAAc;p#*a#;(Pw-he391(_zl=J zcMRyShXrH;+3=!(tawB4&Vb!Qca_aT)*KYr@7nwRDX@@_w&iH6`(?81?7>9>`*XI2 zz5QJRGLrnHhv=Yp26PQQLQelM_(?!tkmKYz`TqA{+tzJcj~*O1I6-jY;3UCG*Kmrh zr`WoCaK_*~!5+a?gX;!23HA*33X0xh>n*n4D!6@cm*DQfeS!xD4+|a{JT`btApFok z-u(mgA;H6gM+J`#9ueSqKyc6Cj=|o+&4L@RLHL%zZG$@n+J!p@cL;7B=$HOpKe$$K z_26oOv0N#*Y;dvQT)}CA69wkQZ~K<*TmBLJE%;0Dr+_Y{FU_a9wo=gWU`rz zCacf2hVyJc4Cmc`-c=n)58}CCK=xlaK&6)+peN~rlKFV(&}xfar61P|^o=ca&wzc$ z&LjUH7VHzS|BUgzfwAE$TL~{ddY!IdA3ZNHXY_yRhxcxO@2W1K1K31tQuFUQ06oCB zCfzeZ#ijGZ@rnI5un+N!`S(r~oOliPEV~BQ!a0Hq1{V+PZ!WzCc2WmlBJU#oafjgk zfwjjz=Lcg8S)c46=N|OK#{)il^do`wynnz(LO&Qt!^>u}=AEzDr~M@GjyQ*;OZeKo zQ~Y7}i1vzi4eT5572G@Eqo70hF6a>Z%zcBG25cek?6ZQW1y2p09y~AbKEEU27h@Nd zZu#EI_g440rDN=WkG$o`tNY?pZ8_Dd@8$wqF0ks;;oITsLCF}l5&sTbiOqD{fKLb5 zQS2$UmHPg<^3PR24n3wVdhMG5`;7laoPEaEQ$9I+a&qf9!SMrftZX!Gm5p|;E$3R* zgM3Qzj~V!}^GeSBWaTHTtg%fc^zxY4Sc3(8r$*J{{1{ z_WS&d_WQ33%sroyIcJxWnNJGXsSgS67Lc3Q3;2@QxNKcEul+)u0kDU`K7#-M)PZvv z^3=Y=SxhpZ58o5JoZkt)9ymK9d--U^*$CDsUoLl&={ z4tpYY1p2CA-@rTZoPe+3xdDH{OM(}#feh#GcZNomJFDYcw(nwxp!Qx*3hEq>jpE!6 zzq-*nD^M1k8{n5tKS;Bmv&XwjaI4@(f&HL;p>u+31wee|H&-^^=OMP#XgfSg>A*30>XUj{CW8H3dk1q0)LD?NY_r)ss0Y_ zR9yRa44jpy-`@CmU^1@)>fFF&o=taQ`Ccws#hwl}^%L8%sKRU2~zFVLl&QZ!w#E*Bq z0KIW=)1drE&OhimXFcZ$=s9C^mf`$j_rUn+B|7f70Y4I5=iI{H`rv?`&*TfoLg9P&i(B@=n!|TtV?@#>(@PlI=i4lgy|6XMyzM|72X!u7m)!ar^y9= zWU_*-?R_Lu$Y!#%@86wi*u!@lo{S=^yz};I{87$3_@q$#eAK?~%z-@~KlpCJNdsr0 z{8aY*&T{zagEslklPOsX3eG<9ANoZI@qcKMeTv^$!6m?rc~t`^Nab z7x1C*sW@Z!X29mbYtQV=*m?0OgVO|O3eFbrshm47zo>twt2LMPD$^d<xphu zmcFo#*n-xEw)OA)0o|?~9}u189*?~Y`ENhZ{v_+|&+U)wvAiSRo4T*%EWI7914(A;@v~ zwj&3-1U8j_+xoYyzYBgI{3!Tg@Y4YA);Nof8=Q0v!pbV=cbwog>>cM4#3kK&3&K{g&4QJeT#%(7KywKkdJ|4U&ctP-l;6VYO4u8}2f@=g<4^Y0G zD+Ct_&Jyez@MW2glLe;?__U549J1w*Rlkk%l>-7k8Ty0nxMkp6oKB3@#V&EASojEAu7uEuStpW$+)tF@s|S#|cgxNave}T?27AMZj0jN6+US zciMJ-@IME9=f4Q}=f4?zCiqbB-r$|V+k>|TZwcNIye4?%fL9#yibM7dUKl(tcz*EW z;H7Jj{*2)9!NY?G2KNf?65KktNpRiZS^;|FV9zxOUoW^$aQ)zx!Civ~1&<1z7^qvj z+Iw~I+CV>FwT9On^14Ib7Q8PwAoytT;Xru5-~$csKjbh-lmD^cv%!~xuLs`=z85&V zaP~l19d-Lr|MMKdygNtWKjKfiRB)N#l56KEo+;G%!j8z_WLzeusch>)`?4 zzw_=F2m1!@3-F7*Hn5iHJ?rd%fZpT#rzgJ@xEp0X{yLyrtyO18be=P9_m}7oz8j!N z=pXw4djj&G>?ilxJ@yrBN$&_-@wovCb$7|W-rgOxcW1kj|MvXS?fJ_tVGFyT@yUSS zpB?SJB)`3@_DlBYbi!EzcOu=}v(NHQ1AE>c>;{27n!UO6KYkngH~txZ1M-63mH)#& z(fuTQaC_~G1@@zN4;~Tl4ZlD5ad5n?$6NJDUuNrNR^{BD@wDG7czW>m;IqMj0U1qh zlbt8rcEW9E4#?Px1{Vs*JG2Y(BWa-^di>FB|cgKfbs!7+p52geDr1h)M&_*(F}K-upG z2M1pdwD-0^KVG?peYfqq?b*TOf=2`o4;~dfGPr+mtKcfZ1%fjLy9e~iQ3Lwxo52CW zn}e4IPYWIr+$NwuE)bk9pf~vb2`cB|2M3=G-W~AW^V{>;KPb3o;FcEOJR6A5jDBWI zoIPMqxpT%w@Z9;L>fRFH6MM}Xv0khl`vm%9@4z14ojL2w*)dpyrw{Do>5{sGWdGqF z()WYfuUp$}Tej~z19q)RA4`@$$ z!oXhe_S^-#xfrz`u<0qmxgU z56CmL?gBZ#ba!ZvfV=~;jlcW+f%DJP1f`2mXUEQr=`QE6hvY|&(o0(d;Xejstn*?J z{!LJ4y+6tM@ekL|k){1J*k$`J+m9Tm=O2gsWA(h(IkfuZ9X)VHecV9%Cksv$oHlT- zO>U#~EdM>9f64GGZNJj0UUok2jul;9pu78BC}$3Iy8BRf4(=a3I&gmLEdKey3j=<7 za-0q0Ik0E=?a6p~uL`6&<8?mb`IT(<#l&vF= z-6Nl!qOL!#q4?A%U0fT=93!AV*nVsqX>^LZ#NGQvtw+4h39MUoj5W>o59Mp(H!D9G z%C;!~8Gl*%p!nbJ7d$xNhk95bT)rsp&VvH!^i-WWd#}7_^c8(nz9%*b{}aCw9~GP8 z<7;OO_J!;XHU@tao7_3t*8?^Szmt6=pD;ajnt<)kr_ArXS3s7~Mb7k`Lz6Sk58PYf z&wh9C-rxfP+rwGG_kuqJ?o06ju|4((_)VNgI-}&j;j`g;amUBKA2QNiqg}S{vU-k8 zj$S-qgFHBRUhu}?0|A|7fA3{G*|w9d?*8d0x`}O2`sps)?y@S^AGGa3+a4G^EO>nI z{NPo=>w;GV&k5+R#|HAB6g(loPsiOjV2fNnQ2!pmxr37jY!|xo>%r%PPX-?h*d5Og zoKrtw?dfpY;rzQpeapAYyUouyho_BQrF{Iu)_@|(Yx zKlO5fGgCIWcfh;g9=UhQzKq@AJr+jEZtpgKf-^dB4u|qzuq&M5xlhWk@qxgZDY;>f z==_h2v1eps*grlhAj|D3!C4}iBuqY$UF18vje7;kt5f=A17+xh3k3GYY??C$UDnFe4qJy@X0xzs&W+~M8ss{e ze!PHOXA_b0!t{*#)v4dwbvA1pZ?p?7}}c!r*t`&l4-aKLB8@8ynZ`MStW_9VN~w*-7cd^Oe$zue8%d^OHI z$iO=LAPXN7kawuKJ-V|QaqH0@3ePr$bM=26yc69y=g2j~?*Ht5aR}QsT<( zzIM-tPu^W0?_z!Hhl-zX4f4+sun*h~s_*~o(Y(84v+s2HW3{baLWf_e_G{$1@f))< z1>=_H+?fBBpMVd_T%Z+Zs~i%T6LVn!x(8c6rag0)=Mus0f%#s_~pDjyR6_*Vk< zyS?Bm1AD}h=k1DYdF#N6RrA-$W!w6NP)dI*=pa;=V701e{HY+`{1|1uY=mRAGCIV^GApL zXw}DL|L#uaPXl}UgWKIs{-^H-UkE-Nyer^)dUvp2@R5Ko=HtO91AGSt+Oxms*E)7U zKLGiEf}r#VDlWbF=nQGb%q}q}_WktG#REQ@^54)E^xMq?{uzF|y9MU@A;D7v`}O?- z_h;wuxoD!2j;f8kv3G-~wykyqE8$>}qnJ{4U?WyBI(Z&?n9Uv`L51C)xpdbkD8< zO4jR(zLLkz54@|=^ivvLL;gF1Q=hWhDd;o(a*2Rn=}HauhSv{#JICK7k1vW2qL0XW zcTJo*umg-)-h~2bS6=&0*Y~XIa!2K6fp25^9Qdy32zOZAJ$iV+ZXk#1`wM559}f63 z>|xy@a^J@Xmd-<+kJ1hF$=QN)2Imbf7F;>-EycZqCk4)L`BXj@IKLn(-LZ5R^$Y=j z%q0T8ud4(%4(=Mb7y6vw6#-w)e!*9R-v-Cnc8qPi2B%*Go#K4qV!;)ID+iYe_{P`( z=w*Xz2GBDS?Snf9cMNV5+#eK60EU&;Nl8wYfV_rd)G`viL{`wj1>yKAoryw~)IJ22jV_VJ|x=X(DU z{4MsoGe@?7b58n#zkq*XpTIpx`+EKd`*8buvdn$NlK*r7x$kU%ytJ1l!+*YpU*`UN z$?G5I-aT3V&EQ)BIn4)7et$doO7QuBpB%_$ac2pi2>9EbDeND-CwO!4s^BHTO9TFT za-Q5L-|-(1d@vAJr#8MCd@fLzdYnnTDR^V>`k?A`{s8KBX5lPD-TLyWKpkHX=pf_d zKcYvBlYjpkYY;9SMb{X&dEk5cQ^59crb2Hzb1|QE2>XSdLEo}H_>q7OQoesSk?#rk z{P~OM6TUY3+dWRtlbu1Id@S${Av=OC=xmw2XdUp~@UgHT=#x_g><2zWYle@79^prE z?};zpy0S-c=F66#Ti6c#D8PPTXSh%0-V!s5turADhJ4$L9mK%DaPC2QLVo z9y~eV&)O%@=H7umTp`#aIDc@4;CR7bw*F=H4Cqr^KehGU!7GC&1$Pdv6!23WC-7kY zly4V)68JXcodLb_(134`ZuZ^dIf7#a?&!J$%U-k(;7_z?*dwr3t-+%N&K~UNJ`}t? z@LswngLz(%#bD5_dID5s({yiC@|r3>io!))_Usfqpq8KYsecUk~}~s_bPi*#DEYWGtOR*L*GbQZVWk z=_Q-VaOJvuRu@_Bd;lNaLFUs-bV!v|jyzWfy+@||G3uQkh|`JAd4>P72IuB%p0a!B z54H~~Oq;PdG`3^s(!N_@K$|viMH7*>O-p!=7Jt%i_jtZgoj>av&dt^6wIS}Hs^d*bizRa zn~m+}e1g6u3+M#;f&TetKrfXIr+)1iC+MGk@@*KOdEvhTb5>{X_N~G|x6;$@+LKFs zB=-p(9JrIuf8vgnvyJx!&N@){t$Y{Y4lLV+@5SEvNbzYoi+8`u8LqQje=EaX(Ypjs z2wofPA2JdgiIYzQJ1pb^x2;7Xe+v zzjT~{ujzEb*#mc@*c7J;oKv`S#jZJRz^1`>%Ao8Gdf)_sv;X4XK>t0*wicbg5d0rYkJ$QIvPOcxYLFnb<2T`@F zccjcOy}?#=-|d0Ht%9ou?oW88-W|S9=Ud)scdYq{`4;R2-F?4)z<)(=usz5U-~YG= z$-nXW;G=;(DnIyZgBJ#TAx~JtQ}W%aIQjK}z*#b1JDpKysp8HGoFn^IihQI)PQ8ZR z@@=lOTKa_l$-xyFOAeE{>=gTV5H49vZqqURL(ZAWW4^5NXGyDc_3{hx3E9(YtIib4 zUnGsaWPi^`bcuj}h_A@LUfKoMz6%p3-;KpthcgG~;m#e3m&})Tv4HG1e&B23v%6Ms z!?m-Rn?~0P^FcYIF@J2kI%lRc%q=^D&UjcrU(?&q4EP6~CDR$sDn7o3Pv%6~bENyf zz4i^TXX}ovvtYV}uZoY#`IY^HJ8|v}TO(`)w#2Oh_QmZ2`s2v~d*KCv^9T2o`K+9i zu^-;FhS%mk5S#Zcf&I&u*8au?pOkY~>ytdQHzv#cYL87(+f>m3i zp1hMrW|Q$`9yzBRDxDr6zsYl9JY=Q%$xiRgY6tRnd8 z^o2P#*L)DpgPei$A@U`%=bVGE-{=}T29M{%J}DhvdP%+7@H0+h0&`=G^bH+DKaxdk zv$98=k??^zGk;XzO!ldPZ|YtikZ1IWdprIXg!}#O>i!^bzsEOd$H+HqrwuL~*h|wP z&dzTYxL?Xnd42HiV87s_fqSU_&W5uS=jIm-E*J0%-6?R+@S?z(^m~H?f=>ot3)m~< ze6Cn+JNg>fF2ctS=mdIY_khkgbHJtncci49I^erHML>VJFGVl$r~NXZr|LfNx3+$3 z)tB|boRRZmy)Mu6 zM#z-Y1$^Nr2#y`lGkhEFvivr1KJv|gZ~apNp9tM!|4XL5BXF;Y58l}c>RjZ-!83x# z1`i3GeY0Qq#`s>yTJoD5=C2{I-2>v!aV7xtQ1_S719b*MZ#aLZBZQrkuy5E3Yz8{Y z83}z=b`Sf4-f@PaEPdlXk~0_7nS?a@2P)q=1^u&Y!*0=yYzlrob_Bb^SV|8$bEl`+ zW|t01&#>3%6Z*q>_g#Z~1f@HkoO=W2j!u7L@P+_=cR***6?6of1KeG3p3SD*H8@Lf z&cM9~&)7F|bOc}1LxOz*cXS>dJS})h;G0KxamXI$YHSSWul6aN`Og1i-P+5NLl+NFekAq?l#G(czt2XIhDZF8f%Q*@ z38Va0mkp%jVHaqRE@1D_L;6eyklDuNogvfdG%}TpCbR2Inp`EnQM!=q<`=T>H%H=? zMvnU%f#kUJA9@vp-ErcpL)mff3+(^Q`+I^91p5c5djf9^yubW?b#K>mlEyBiKj?kULA*J+QC8X~6IG;NX#gy+0dZ|G?c?a{iA2 zee%!1Hyy_cP7=@`rwP~}bcXX4x`d5Fm#|qnX+%+uQ~FM8GC_ zZJ-p*Kj`KOwGGx&*VL3^IG@z9qz1HdoK?7wtV+yugRb7+g|4+ zzHz%)K=!!L&IdxDxT8ouxo`E|;EMq{Mt}JG2yYJTk)Ia0tGQ3`@PI#Fm~Y;>uYI_E zxqD4Oj)Hyr9s$`)4!d_mkC4m4K!?yL^ap(bWILHI4?nwyOlR-Zxq`C31A(f?{iu@r zbe1*&4?U(IY>kTtWWVo8^q1Xm%Ydyw?i-W40>aL+`I>6X4_y1(8pdueo*1y@*mmp- zcck!`uV)43?`3OuS5dlyUHIYP)4}J~Q1-_6*UmBAg`(%(LG{~~yQK7py#Sljb1eUr z`=jpFu|IsD<9jsUu{o#UyW+#5V}zZTTO&^ieA@&3SOOIJo*8L$tx=TI25dKW zoHV+c+;>iCe%(o6-zhImn11(e@+Ei&GpVNT3b%_03_uK3T)8$HWkNw=`C*i&>2`^El(9mn@) zKVYu_WE5Rb&y)}BzWFAcuf>_Kzn|;xgScC2@BG1lPlmp6cK_qRoh~ww@9S8>NdxD} z=L_}-$WG_Xd|%NY~< z?d5^H3=a(0G3M%W!8wAH1V;(@7yWIL?+50e?qO3tBDh~+r(kXy$J*f0$B)e9QAq(R30CZL1JM79fm{_BF5 z2hR`4STdDtbp}8-^Y6f&0%`U4faJNqYcJ1-hSE`dL+l&ABKLF2f8TAecjzm30^IF& zhnKGj@X=#ayGuCguKJkYHj|!d?(5=prP~ewxrm=6p zW_$G-*lxn+gw5f81N)Bcv0uP{Rp;Sf%6BbnLOMm5jrfDWH`;#;oQL?)|HllRi@1N~ zTTbU-WuIIje$Fekze(a7guQ~>2Yg)Q6?(UT{CZ?S9=FM5%_Fa|Q${@*yj9^V-0R^vAhV1Aq-IZGnP$#$~ddn1i}H>YGh z9YOxnL%@GXC(u8>+0YI>tWV^8`3=2`e2L^go5Q?ommboe7X{MkdJxuMWzgtS(4a)vx@0E>NenqyUyJhYHT{UnI46YU2 zF!0?Odxek1y)Sa=C4u`yZw{QN`2N}7xUiRY2kcKlOybqQ-LzZo5}!R_w_G}Kj>5Nd z`+)!J@qs=1zJc=;e-rP3;PZjMbL&jycR`LXw;XfpF;~CaKIhhRuHKKjM!tY#!>|qR z6Yz_Xy7wPi11uqI77u+LYZ?Zj|lX?Dpw`_X;3-7J<)a`2T zkv)@rl53OTevt?G*eov&l&61`VZXdRC?0+=X?&0L4L{hyYhcf~ zOF{SiDWLcHG2P9gb9M{(IX%y_2mGG&k8}Bp2cEn0dAf)l3v^Q1J2%T+Kzofl1*MDV zBdGnd{kFSFF*qoa|&LLCJFRm^=r2eD#pC z zdXw%cy{U}0=$g_gbUYi+ISc*52gc{+$KQFJfZvOL;rAkA>MVsl!biDBP(Dca-LAIw z|9v=Hp<8_8Mpiu_aK_?}DnE?xpx+&^Q$8B-(fC%{`HH{i$5wH^QufO6bKdNnM=zcPiK`?uEN6?%ud(>zVUs^OaGQ>=m{OU1J^F zW7;d(i+XqJ{DmB`zjY6YZ{Bx`bPjpMemQ+`ih%6$J?}07zYLj1-~2TAR$%YVj^Rsl zCiB67e0+Vto*_?P7`z~OZs072{&`%$#^G0DBk(ctKLMG}cA%Te&MBMc!SPQi1LQtk z0OIO=%G%$_etPVTZeNi1dEH+X-re?VulGst<=*!Zu$9vl?V zvj+y`KZyS#_+wyxejm^S(%5^_{;>u&3H<={INQ+t(iL#D;Ml z|Dd2d><=yyTq2-5=nMCKovZowh>wh2N^jgE*n17mVLX55_UzY^v(`rG4KkRYg)Ajg z#mQnIugAPfqa)~#@*9!)Ri`p^5V@>g?Xdy)d%B#rPU#Ok%9l*1!%Fwy(N^)QyYvgb z(ocBAN1YWVf-v36jF~1jma$xYi zfDg0;BMH#!Dz#f z)og*A1Af2+$qEoUR8Tj&kGF}BSQ0_XGmVEkeHUwmV1 z6%gh(bEd;~@eLz?ns4vvBQ_FS$(=CwpM6W??l9d$&(U{x58MsAc5vB%EyEw?`$oPn z_ZIlV*fDPnUJ~$!vGwQ}_a819uwzag&@+du9I|@W z&)mHY`h_iHZCRK0GuE*8g@1%Tp-cSzAhO9Gm+!vL_4&idH+m#jvR1#1`$@ic|3*NE z91y%OAdBA-ydl^(Aj|0r@}B&sFMxg+{bT%L&a$2H^Odn_ZWMGIhAq=k{xEh;`Ot2h zd(Agl!%gGkVwc<|P-orErK{K=beHpx#|HEd9rv7o&U@L~ImnyW{>BI018)zE)0jUI ze75~RMRW{(^2>mZ_-lavHu!abvi-cU{8~VV@Ga6OV1C(&@_b+J9v8jBemrjAY~@4& zo06X47n4rk`^HF`{6e>1=pg6&1!pt#k-z=t?*W!AV^2XJ`Gy^Y*|+Q)ap!e62$ACQrcL`8>kv^a& z*cJ2>+e$iHtNek^9@*~ntMw1s(x!gsbB!6T`C!u-hyLFzpp)5k&auiaqL)EGO24ve z=+Lf{#MwDzvz2b4o21$My)@`PFk!Zg_n1zh8~D8V!<>IfL0o%YGEcVL11O6y- z+g69V8mKO(f4*P|@@%n(iai2?^E_zKMon8Q8>B=bk!a#lMWv4hxxmVz<=Z?Xx zgFS=m1y@`9`;6?C3k7EjP8l3M`1_W>ul}DT2W>fM)eq<{1AWR~ahC6IY|$mx3oaF$ zEjU?lw7^{nJ~3x2?8UNKd{5+gIoo%qkNxSJE_N!vsJ#WBsdG0zXgY;GY|mq_WZ&i8 zvp4iUddF@Yc&F(X-_o;F&JgStkW=&vSx4^i`P*;XcaxWVV?PR<4e-<0!?R!918|nW zenFih0DWZN&lcD(K-o4Z8{ypny8-AbcExMgd`K_JH~B9O_?cc4ye>fblwK7G;{`nU z=?Z5*D6o~-FLh4BuZ6#K2m1x|!}&Vf#NVE!JKW_3di6U2UGl@g`1z;kT=v_~0`tN~ zVXK&@(l>O7`71p_f6yJr3CuHHB8_cGXZTx!e3am91?B6a8%i&DX6(yzul+4relK^+ z*)VwcI`Mn1^({Qx2H2)}QO}{8Id1d{6vV-Z!=dnNQvWJA*FpKJp_R7?Ahy{Wb7Sf%o>H zHOS+ya!-q1pbuY@LM?#;UlAG+%cn9-WE8c*+2MT z;LM-h!N#Fq-Wbp^^v}M5Z~WOI&UyI1=(9Tq^xF-Bs|2M_&XvECdB))60iE*aeEa$R zz`4=~19y&J8I(?OKkVj#d7_6e9b7P=r%x2n;m)<(>*f!m=f4p6o6Tqg28}=c;39M6Ro_wv=Jssm6;XAOu@*RWyAX!2VkQ4T^An%31B)ebXN{1O38>#~xub0Q%9=Mm2OiEYD9Q7^lPy`mi`n~9ymPv)$~SrG7n(Rsj5 zd3r!sJvX5LUKFrTUKMyZouSZEeAn(GuuK%V;O}x)%-_XEu}`*NK5DF!{|KBb^KqRcaOaEs19plte0iq}&K&TA*?Y5B z_6(d)J7=L&?i)C7*(dOw1UZ-F3czj^MkZ*Z8lT#UCrs zha(05%$;fetb+sJ$GcPPZUTG6xzQT}zCJd|9RlYkmk7=o(7z`S__ye0KE)pdY{d5m z&QaJTzN5c?U?1TB1@KJWy>JKPN`XCt^>Nam?33?huk(fAfWVq`hG(y4uVpQ>$@#Lq z4|@jQDesbfq5UKILe6;Ko&A=*LXPo=k&|R29rA;~-LM0Lg9CeaVKSKALY6A8er@W5 zeIA>CpTN5=k4z#<$PfEndW}r;UXvl@6x)d(hzurA$xZTB{bVqkiaaG(^;HVb(l(UT5;oYV4)>#pna~ z=AgoKk8x?M>l!|={;qgmpjrNg18*^m;TZ@ z^bDO+=O}cEvlRZWuLXQwzy|ti;C>uAz`prMz@`CVf5)ZHP@JF8IleU_M|>}I#lYR5 zdjxzUzN2GXI1`~4*cIJxbY2d{1NUEF7x?zVJ*M{u z{8T>-j+}eIXAQ0yklE#ndPu&9e?V~Gfb70!Q0F1O|8NGuPWV*deB9YMTY>$+N5xJM zW;-}D_x*^oblUe9wujNiMTZ68lgPgCQ zByhIwTN<_z8;QN-T>S;XV*>NSrr0}RQ?Mu4VstB;Vz7* zqw?qyHh}U~k9Oz)Jmm{h#$KarkFqtiNiXT2b2+*imG8Vp9o7%o3&M1T`aoD4WOeb_ zN7NpN+=l8$r7LSp#;f1fo_h>*zcmQ_h5d}sd_&oUrE9HI&w~TT^YmP;56{@X$-1%M zuvX}5JoLCS_ARx>Ju|T0Jx}?bKVEgSUCZAoU;CiEK7hTFcf;PL{G2tH#>{qNFVR)~ zEXjNAedf#i|M+}16A0EH||9(9l_qEUKy3Xr7kK=v3kK;VQXNQ=fC!)sCKj%aF z8&vm&{)PCjCRi!t0oPZE_Xhp%I1v2_I{EjZ4}3v5*4wd&^u9jv0zrH8Y~%s)_Re@( z$EhzFFOAp4it*mqIJSyyV(a)+Y%*c<8Jo}eWPBnvh_%COJ{+HpJ>rn~Mf^Mth;PJK zV!QZKd_Fdcw}oRE7mTNenP0~!>aEKQdf|>oQ0jdkI? zmgacB9p;_&!<{49x>6tM@#tm94SE=NHAAglv)=hj{6F*7a)X(x?Sf~xVek^whReot zLrz#A_?6}(9JlDPMdF!3BhUtXRQ{7$Lh)N)gEpdP216s!M0{rZJTD63xqaQfvoE*Vv|@W^zc6vYscv4Z_s(&KN(*LpJn&hBY0?*p>J<~*sL0#&86G! zAG2y^5WMEA;q`Ez@9MMg;COF7>k%Q}sYiHyyg%#9I#ruk@8XpFsV0(}{1b!ZCV6(m z8JFrW@et}SzwdZlY!W}^ATeEDuq}wew7j}u)E@NwfR|&Nb7>nK{bxFU=P*;n%h7Ar zkJG!LJ7|5{dZ!7y)Ti@Yk9m&gd!1P4evj*&^N75cnnry=vrMl$vD@c zICPCXCWqP9#c12NZF_jJd|9X|iF=9Qv<39V3y4-JMIbJ6( zjapCKx6Yj#$94H%z3sD&9+%uI7pR}~;(r}#4jNHUgeu`>>fgW}*gDt;MfguhL7S^PP!j%(t&xG`>z zyTkHk%sFGu`Qmx8a4Zoa%DD|%O$ zRp1}#ZErF8+hh9OJBMR<_=NKMav{g-b3OjV<5TDU{7+t`8bbb8Z>mMr0`h|G-u7cV z69?!~+p29%E>J_Oh3&8QC;Jf1L1WP%wn4pAUW7~ABkkq>;rZxg@HX_n`62&vT+aGG zJpnpejN*~d(fk|tiDhP&j-Sjk=|7(q?jJo4Ivji${c15$to&?Ep`pDSMNj2hU*^NHZ}`$UMt1RWAS)iJSP?i*B6S#Vu^6MaJc`u z;rg=is(4eZ7;lL;#%p7_cyV~G*E}Pf!)30RGiHlf{d+VRN*-GYxz*-h~*Bq#U{V$sBOe4+m4u~KHRWwhxB2|$U-_hZ z-j>Oap^5boBawr{QolLI8pg<10EmEudie-Vft*ezJj>}eFoQ84t=OBh$(N8A0+16 zm&9^@C+$Pu({J>@x?R1j_EeklTxc9x$8C8*4Xxg{4~-rj-_){RA8J(llh@Ee_8r^u z^f*4QAC=Y?e{B1-v3-UPrpe7M*_XtAy4JoWZ;10_KNIWe;hkcepv$+M9KSBkZ#cnq z_lx^q4*TPmLR_>TVmxMl6(4sD&#}x;gxv9tSU#4D`C|4Ee;=Bhf9u}P2lIuZx&K{%qvWg!=k>Qr zJm!5ZcmB@v@`89nye&Q&p9yisIuk3ziRp1oTv2_szsj zzqgz~$I`U6BN~Ao75l|;abE1+e{wtme^Cz5TjDJay{?9Uyuoi&yN=u;M~LxyQMTs; z!}91|_Ya1T$%B_O;JWuybErXR9~z&wm$%eA!|$QP$MG~H2l{>k&*SCzuIA6|7vimY z$#cYOdPd%mQ^jnW#d?=l^4c)V^;2(W|tt*tW%Kz6wo4 zXNsv}?{*W!VETIKowoL-KiwDNwD?L#)8C5*ZM|f;Ts$m?b{CUrYcX5gJwHy4qvMeHVSFpT8au>Ru}OR= z)`}J5O|e3(8EePJ;dx()Z^V9aa2y#&#Sw8(d_TS$-wBudhWotdkHTwy634^|@$2|a z$OUr4^>KHo51t*%ggoFFlr6&g<1_hP%x^(XI6UZ1xs`uJ-$L9Hv(5;Q$p!N9i9v6w z#XN7sGyWgHoA)ku@eXZk@&JE`cj!K~f$h&F4WcKYt{mRK=h7(Rzxq>dumnqh{@Au&(*{k$-D!v-$1nn&*i@}S8n9rl(&+iawf>*}y z;hm$y=W-6)XNBu>jq`7Ke_C9>)-eb!-M%6m>*l;3uM6J(2SRT^j?iE5efdt#Bjx*y zdaN@Z>{UPC;xBB+G#kS#h9kqaCR^AeN#LI+U0HaF2 zzay@SbK)0aTe5%fUwJk>P5QyMEZ*BU`8zJ%Ru9`g;W_ex$N9T_9sDFP23)i0dR~IJ-NgjOH3VG?0ggYs(Pn-pfojo?UIka zPw?Kq9J>YmduaSDjtt%kADn)FQ_#ZBqhz_xa{;>FvBt}E{-k%u7h>Pgd!@yBFIU7R zaaQnfejN1n*Mo1PFZ;!yxjza?n5X|J9bHcG_=7St(pf+*LnLY!}eO#Owym>KQ z{^8m4?q46PhUKmqD}|o#{GoRFW2k$+9-oYNhhBzrYWT46*+oVSB_~@wd1wZU}kd;xHHV+n^=zn+f_vGzX1jKj4+p zU-E)Fls3}y`d87 zbwl$udrJR1Jpa{`bJoAKzeysGy0q@}EWGBu zF?uXKH90~LTff?GdQ9+?E)L$H9H3@YFDyNQ?&hiJC+IKeJ@6rIduC+4#y@oi9iq4D zw#W2dZJRU%&t>>qH|3%0salrTSiYGcx`sY~Z>VYcnY=3b#(p4I@RRHp^pxBnzsMi@ z(((qquLkfMb%31W`81fkaCO)}?+$t7p|Bt8&&ebDvGRi%PTsC#%j5|@{hl#;wBm>U ztT`0kgB{Wg3{T^?aZ#KfC&$5I2FCmgZ(PjZJH&gv*IxyHMIQKIED?8gZV54!#ujVE z+y5qG#)px3g!6{Y%9^`fC%X#J55p z!MT{;7atDC6z>?$!E{16r>lJAv-!N*cN(D`cx zkDecPbI7A+I~^}4e!nADjg`Y}C;!ZSYAct{(eCj#gyWMPH?c%46)z6QRIC)@JY4rY zUjKhOZv2XH{2!hF>F|3y9>jdl!%sXs)vLdBa*Q%P!RMr7=o@vSZEv^Wsq!*uB-wnsIo?Uc^hF=#4UN?onKR@bV%X?^n?W5&To%KtR`sK&SN9u)h9_%Fus zXLucQfu5+?zf|afig9AoYl60a{REo$6(KK(iHn9lEzf*-NVnx5(dBeH?JW-TWB5I^ zJl)UB*(P?5y@R&*JTd(6@VXO%Z)7IkTHRfXE5*mYUv3x#`I@h+YhOmFw7xbaUbe@rT&1(^}>HaI@x}c$k zwi3VjsZeu^gM@tH7RsRyfHO3#buVmUwOAMt3+Gv_>WE)a{y3n%dU^;|ti zUie;Y9xH}E!J~6LIyJw~gOPVm4YNZ0s0+i4KTnEJb#wT9Uh|C3=E}>W^TquS#m6Ve zWvhici0`9sp_Xvo3Z4~z$MIn1b@aM#4Sm^5u;9B&EneSw%e{Ci@y zCuZa4xIKTU$Cix6Vc)Vpy)o<);yw@5ypBHDGGV_F zr)hWlulP@wi_u5L5n=y7G!BVzDL?#Ra?CJqpN}xSAf5qD$Zyb3j_49+~e@fWV#5b{DU-Y5yPw^-A^Lae>*Z6bL*u0YCLmZ{$U5ce*vByr0>9JYd zroH*@d&M^=xc=>+zr}f)oc12j>h7n_JrDB5(CW1Nc-^714xQ5aJ_p_Y%a9l3gqdMk zL)Y_zctc{nJZBw<=i;&)MtAdf=y&rzABc6s`nA5TfBozo;|n3Dis7Csp1(QdLpt92 zRxd3bZdou=;Sk^TL~T=iMBXA#(f0P5u$_wa zw#&!<``DCrAN^LE19St-AJFy>2K_!O{PQ|>B_C3K;Boh<2W;oIWAC+c@IL8#`w(43 z4~hM>)R;x0>*WE*0;`iRj*G%|HM3gzjQGQ3o|2}&FZer-{arA`e*5qWq5tYqF5vyp z^}JiNe2(Qc^Zigb9}rL3ajW8#nDxT`-i}ztC)q206g(U~%wyuDI5#ef>*L1wQ=A#* zCC$)_vrh@1u<*tubt!{-(0jk?@96vq2IuZ zax9n6PuHIp9zQtLE$fH=$2=j%pA)}`LxXp-PwX7dm+{e9C6t-naa{Lu?V##TD^FJQ7#L54g^IasOMxbNB@O zXs`| zOm?4{bot!&K%?6>{GKDP+g|*(^82XgemR*fr1jNydc)pJjP_5@lz;!ypd-Y2K0Z%n z=oDJryowmC=H#=8#d@i>b^TO6zPW?-!t=#!+TQHTn?o%39^(4q;XTdC*oJMt^p9Bo zuCVV|zGaK`{F?nkFPPVPUhrl98g#tlEbWi>PhQN@;e0^LhJ9C`jaTEGLH1qGbDw?M zthM9!7mX!?k7J*=?>o1Hb2{+x9qTFPUmx`1CxhN}P6A%Xz9GKd6#B1?u1{(7h5LJ+ z^zL%8a;zOZ{_TSoBS-K&-X5OA2j_S2>cx1+&YIJ4d3K1I^uF14`rmW)TD>2ShhH+@ zPfwSZ;yv~YdFQIQD;^8+p8j|H>X3irD0zanv~8#%HVb_PeP20-XT$H;_tnEygQ#=V z0>6thV&nw9-V?*;mmhflW{&g$_+Vx<#eDNRpNO?%r7(y2{BRtve~&b~$=km=_&tZk z9w7(N`Er1HoJC^3cxucDTK|FgYdBB#8F6&z$9yN`0)1NN=rDUp$E*F`6J~Mb19{)J zFy=RHTedNp!FKmx$7QH9X%5@xLBU6~f2eisQ}TfAU0kMvXf6BU@Za@M?VCJLzWW^^ zPk1h`R4qUksfR5`PH`#L(_-?=vw|nzSAK%75=(9m^}F|?_q>NX|0iJ{hgY*n=rO(_ zcr`o({=$fbyoV77c?)zVz3I8U8a@qwMy_~ExOAK6>{9-aJLp!Pklf+YObhM1Qizeu zh8Qach_T{5A6&g4){3!zit~cD=Cjb-v^fq4{tZuEy)gV4zP;C696a``~h)kKCS#G59*uK-g|}h&2tl@#pUyZ4xo|6 zdA^!Uv0r?a4|w0Uhs(q9b#wslo&Uy*5x;E{wgwtLDcM2Z5!1BIGc+ zMIKl@%qI!9`pxo^_vQQO)%vXT|IF~&z8D{m)h08RPiqE;mve16Ci#@G4(a`4Cj6}P z6Ir+4jU8jN(35>#EEfElhpNTyiob;OWQg~!-yD~Q<8*m8y9LkQIpE~>SBJb$6FfEi zd+dqFrgX*M`J9IkDPOp#q*Pfcz`G1d%xNd$M(y6EYvdE~H_L(Up8-8$iQ`QBCL z+IAuSzAD69+W1W|JiHMDU5bBn@rcWOLb_Tk6${PK*#6uWAI)E`8Dh6NFxPFH9$$Yl zW3Ja^`xUpv?0w^aI4sN%h@pH-+dIs~oD^c?xf5T-Ji_%MuKq3V37#kJira%er-!`e ztWY2G=Iu-L`@ozHPli|H`54#5M;>|(B z@on_!R*TmL{brw+2jqaig!j5)avm1f)%NQBH-xxPFPh)}M?A0dPU<0kBAio(9{o*R z7I%fZ;I*+%d^ywyhlPHuI4c3vy&{$l zy?_NmKSHiiho}kEFLIRcwq?9Ec=u0O6B-&VJ>TcYkUBUmk8-e}4$;@`%_w zc=;a>eOz-lj(uA)77Bhn{2tJwHIx5n{=GcFxBpfAICwo@oE&RQ->($55!;IT&$jcl zuuZ8gFAG}2EHW)%pP_ryrDo~XyJl_70PEY18rZ(6Cu`qT-##zQL-007JQojWdR{&4 zrcb60=I3~g`{*~?j&7sx`94nxbvti>2IL9Rh_wBgaeS!%XNK6%^QAlKPxF_%g!ugB z5VL92jY6E~?eHnYJzhRfpI^ftk~h4LmNolnd3wE;DHigpw+PFUL%epxdHP;_7uUsG zUJD;atTv;`Ke;L{jI)EU&s#q(c@ zn`;sC`Q?1_>6%%Lw~f=wV)7+1J-&$%DSg_$|%bNvBxl#an#EBHcYIKLF)yJM-;24er7;aC8DgB^ko zBsZ8vayhQ^hSU{u%Zoz|zz33x)GWTU*~|sQ0R6q4uk*IJIxdd0Co`I~y>+LK+AzfV zR|jqHJZy{r9?pGnPy93f7PrQAae15{yq#YMe@Bf*@Bbj2vyIokYkWS|kGBPnpZ}*O zq+{fM+l%dp&KK)xd@~F7L%QC+%46Zlz&sjX-@Xj{wKz|Bs^k9>H-tVL@7AT7pZ`P4 ziv{|V%fB%Wm&Eog7*Jui2GvEu0fNF-88xw?K7|HA53&{aycAgpH zv!1Y-cV5h$ljCl0oOR<=Jil_*l~ew{IM27y7uHL?A#Mt@k~|?EKYvJH!1CyQJ?CeH z`eEV8EGd79XXP`9`$O~7`;H?J+v$Cul~*Osk3NUIM*GW!;{C{nav@Jg&K38^{FHor zZSZRRPWq$rwH(0X@y~B>Tap9hc-xoXTP-=XyI9VP=OfblV!CZ|Xpb!?9=sU?@$}-W@DBNbMB5`r<)wdLQC^TT_66a z=kVV7qU`_zrclGwQk|+1)DHTG`^Bjt=J1wjP#y&T zd9N^cqEC5U{44yum&HO2@U~C~Y#yJDFT{31>+cX>3Gq@6PzR_L^g{Jd>3ce#hfedu zby{C-aAj`#B3KJ$Lz`)v~MiMPay;yK}%+B-uIFr%-hyK^`u!hAiA|H7c_ zt?S2Tdu-~wr~Dmp|Ms{c%we7rj=wUOxKQ19f-id!Dk(FJG8!BATJEurr$lsbH#HSpVlAR zU%ju^xBsgjyPOK>Sosi2JV$Jp;W~ zahW;F!x6j1bTOW;|83Cz{ChtB;qjAc^OmPH_jB6hm?^cwITO5=$FDB9EG*Nqc|~(~ zZU$bD_%G&*_q6=mL#%&asDF51G(C^YrSITU?|RcvD~bDZqngL}wa$EZHI{X|e^}Sn zxpglmoD#e~GdKD>SBBiczvK6*;p7JY^3L_YVVL(kH)v2kgBayFA@Tofad4a# z*M#`*d}3xm7mg(+EYjaQc~;C7kInwr)ZYbj+^K`n^f~oAmyK72c)5D$1HkMf&wS+x zuFESQjEzHG;rJ?Ek=Yu@%kpE)BCQhQJH4*w%**Hd)A}_3cfyRwep&E(<_h}%+W37q*6OgJ@A>-rxMs5)bL;pE@txM!zjuyi^F4h3ab6zB z?$IW6jyVF`O`)c|Og@-&s85f$6os_mT%z3%;(rGdy0sGyOU~vRvZ*M*P2^7^A=Y^ALm70^+?` zFHZ4nMyxmEDc0{CJH?K{#~3w%IstBb%yYd)%%giP&vNw_z7>}1IbK5}i<#o5_{vj~ z2gF|*eDPQy`0+`>DL>ya0b({ye@oE#YJr=>eY~B!!t?$W^2a0L^=b)mp06ao%pYR@ zVj=Gg-OuaMYv%K>5G#j11h2%_a3v)$v66J>Jz!b=Wu)Y{Z@#7mO-CA6kbQ?dkzgR zUzza-?e0FG&u!n&=aqweKgfe}k=*9C_p@wukUB<;Fc+d{`1deV;#g2|%JHJd#d$&N zn@J&NoHO}lA>PsYbpDP(&wnR)IuN6IJmMt1&->B0^*qN=d5-?=;lcl*rN#8)L(f1> zK=-@O8{z};T%7k|o*DB@`7rz){=V7IeZ%p=zYa5}mxkGYGp9Gj4dIvsb3qr!IdO6v z9f!vELvG*&(f#YhsvLel)*xM402;AslzR zQHcM%H1VGYv_|N$tPsn`OTz3G-_5`ObS#!0ulf87;fw#K z3VPl1hDU2Z=HH6%?jOiAuG`mnPIRE0qc)IpuJpr#%-&nR<=PSY2vfsHJxUHuFa~S$@`W*BcO>aI`znfpC zezKqP&s@6C^Ssvjvaic|^EAh|Qfw3a<6FY@XYZ{5~_o{HI!g&(9yC{Vm^d0CR+M zHt>1$v3=Ihg;}N72d`gkb3>dNhX;Rb+c3BDkr3Oh*H^`hVyWQuFC1zs>;36L?+YFe z#l0cc^L@;5>g)3S)nsP{?_d0vvyTn_KfmXz;T#>l*Kxrf4DZ76NPL)$g0Icr7XS2I&Ez{a z>bAHg#ATky*}=P~jSmj~$o_F)I8TEfFJ^{gyZHX%zI-7+cpo){9H4G+OqKZ_@5gr$ zi_QDLEX?GX&7sfbfGtD6mR~O)e0jollXFUVoUgxe@OktF)D6qW65-ghyE;$9W#K#c zzI>hc$1CHRp`UwsoESX(oq~tY!&xm}8{)gTK3`}MKRVl^oh|uaQ$5_jHlyi$YPU|1 zb6w}}j~PzSf!pJ`H}y9%^?mvO&Z#N(J1+OAY3J1B2Yn%UK>QxLotL*@=mGIM)r7Vk z+sW}^+a7*Dzx7k`WIeHN$G_MQ`9fkeT_#5JYWXx{pXJjJ?|$^`%okWDT<1A%8azmP zg}-mP;=X0^e`tUCMn2Jd<_++Y9{um5Q*)wp<2AucI4i_{K0P0r_T*EGYo7=`OnP;_ z5DWP;;-H)$Zj1Z7gqSxj#dJA9FF=fcGUB4gq2he@vIha z3$=|LG5r5^s%5Mn-(ka$FZ46y3^kKnVSQR>+r(IJ*3q8v-LUTE0Xe|$amM6a3Uc(# zPx5;7h5W8`13f`k$no-l-<@`l7t8|C0rG)u#P(v_u`SWQTL&#n8`J7^`gHGm*J*m1 z2kyDHaaxD(?e(s^&tn+Jb*Xp9HtSODG2V|Rmw&vM+#`R;9W=d5xn$&(ebe_ex96!9 zT#j6ISha~~ zGvncSdB+SnXT#1hGt8e~8g~XAt6%x@;I;EL=yvC&Wn-Gb9sK4!x1mdWa`Nt zI>B{wF?zK89De!e;mQq;g`%Ztes#jm;nHn+!RyU_n*X^q9t_7{InSFpOtU>=_X@$M zUo*_*KwN)otR5a)IXv%;LD!q*TqZ2bKVH&R;r)EBZG)H3$G1NIJ^R1^&v95sc3x_8 z`C`9b$ok=!2;L3v{uQxI$N_ZyJTXV;>C*K)ocqJ&Bf->=C!P{|zjKEB{8sXY{_l-p z=7;vD_r-l4&o^WH*gV91$89YWW;$toJ+Lv$Z(A9$-u5p>)8XPcf1Oqrd+By{IPGs= z9NJ&+R!oNd*8b|46@HDHK^_ptEz|x=zsU(BUx@GaXI_ted-yz~#t;K6n>M8N>BQQD zQ!_*Vn4FvZ(q>G=mSf}K5PN7<8lH|7uf#2}UkuwK#4mA6Uhv$#VrW^qR;}PZxbD88 zcgJ;)$pi9)nt>-SZqnPl5^;Btm^T===6mi8@%^f}Jj8WA(1pRj5#z=DtK)`{6TIf{ zaZm7%hVJJnnIT;=%<`x$#B;IyL&1}h-^B01$a|x&JDwv4T1Wr4dO&S7P%Bt})}x$& z&0?%m>sy~h9_&BYZYg@J5dM~*~ zo|)cf81Y`LS0{+KG?aZyjiKHU%f(vX+*ApXk<=7zkEk2`s~z2K33DTbdf zuk!TQjrC%K34RCn@q*L}?+*Fm&A}6TLGXi~AI}MXKR-w=P(R!qasbcYxtI7ndcV%i z>R1hVUe8EhM=#5^VH>mk9Ubad^)W4?Zl+<>()J&|kNr!{{m!u8h|%_qvG3St>_2eb z{v=;`j(v>3BewJR;X2$ua*KVNH^eKl-|GvD@e&|y|L-s}&MPo0Vs7NDI4<-Tc``Jp zxW;3lX=z-sm-ZDK^;li%e>%pA7Uqw5-ZKX98b?lJIKXnC<+fAuM$ z|4NIC?U%;+aaIg(-z?9u@$=x_yM+0kai9C$=IcAI_tdbA^Fp4W`FTSR#2o*fW2&b> z=ldK>OpZnNIemBE<>fJEJFOQwpT^hYkPpOdIZH0I4#jRBnA`He$d_XMrZM8ZOZUkI z{4#k~eKb6Ny8rO-8=V?@ycfo0LHmpS)8|3&&%1jt9*Nod-OUF)JM?$tfBl{j|5wiw zqD|xgH2}nQh|@0JXPc!hY`3;Kue05azLM<^Zre6J*K7X&mg_O^&6DyP%XZy;UhBQ( zC3Ue}q&~1u=}YkQ>3;r_93d~z`FtaqUhJmH_&WC0o5Ku|zOeqXelPwLVmsWgpKNCD zb-~;HcJPCF#%spz@!R-E%-1>Ioln3_4INL*ZW3lN`RRJA^e!)bk!B^ubGq3a5aPWe!e~*=+sb8+#Jp$ z#D8(#H1Fy2ne9JwGV?EAIG2vvV;oOx-qX1!9EU4^$Y0mQxuKWLL(-pJFTB5bOzXxB zl6vQ=_Dw`fvsTC5^2YTtSC4mu*vKOnA3qwdgD3KVkYA?H@5m)1SI9Ao1|Oa#f6jyj@_WQ+ zm(LEhz}zuc%m{H@PximF{yXdOpzYy0PiNphHH7&dvqC(7@3BzGQ$CA)HfB0~AM-}O zi+KH+*fw?y^@?@EuM)F)`D!0}T}`t`i03rC`o?4W90!J+J2TW-ydAjxqtFAejPZPr znZ5OU=*i0gemi}ddxGyT-amDs`xok*fAl_mA-0O=yhSmU4sly-r^#(2V!HS}`bf4N zI(_=KB(8hh_B85C+m#wsUV*xneo=eMGjxpG7&U>~gZ`0Ae72E;d@i+xZP~tHd*=!5 z5cV6#WgQguDSChCe8&OPW1~O7v*-1T{pK%deA?g44$o$eW{MV%HNvr*dVBW-zy2$6 zb~Ii$J8xox&?n{3n3H#IF>!RR8FNkPbFrK^zFQm+`klvzS(hKj*J6v%N9Btw84HAS zR>Co^^T)HpY$Z)E&hq}nV!GLR8u;})Oy>IOXnsFWh)3jnZ8m;h%S+-dA=kVwc<^Q| zq0ZptZxow{&!q-akQ06w2QCWkn$n$INe%XRwGYlOPTT&kHIn*P?f zC{7AJ?jOdt!+Z{0*B5cW`ib5@Y{JYuvmeBs;oJ;cg?zDL(D?Gkd&4~E8$v&Tu9v6z zIr_Q3i(`V0-zD^GHwkgxG246`G2U_Aj^Xf+CNTd+V-9M^M8+z@8(dAolJkJ-P?gX-1l z5f3ki&Nt&B2AJL9<7n?F5hZP4rP$OZD6+$H|g>0-KAZ~c8g#B^HypwO$8Gx;>)wRQc=;He!QJRJE{4iM+% z4f#XgL(JC;mlMXZ5nf|?C&q6>4WKvUceSnD9sWJ5WBK@owlQ@hT}?|5&v1J0XlRV- zzWcU0F?5MbUcJ~$f2d*2zj(c6iPvJcZCfmt3*;G?FLU3h349LQzt3dulm>xqe|Vdg zMX&fwVm%L4JuSY|@%pv51np%Gloq4Y7muZ5si4#J;T8ywxxHLC7H6w4x2MOqbmw%` zlXXljeQLIZAF*MGXL_Y#J^e2J-W0qSTHTyFf9BJn|LJ^d=Gs3T{yxe=G0tI3SJUuj zD)~P@i6evOE_R#O^nU9H{cg_l`Qh9(=KOg{=KA?0Zwj-IpAEWRKCmoxf|_CYJUpKP zeLiXjdY;yI&VHEf7FU*Nl&Gm`Th8}N!)AoMJFGsua+_W z_4JZ`hqkic%Oj8f_wlKCz!T8JbzBgihsGB-X!_&BJg3;B$NK4@ZOtHx|0_)xdY-<2 zd(goy&0~5@-;*c8BjM%nNq9iRAEBA$lOsuTSND#cjB5Zq5DTxaZLKa)IsFb}d)fel6GXeLnB4pUbZ~E$Aq*oi~#NdO}}K z4Cl9s;Y)^nRW0rqVaHY4hwaZa+$y2}x?$`X&Zp^E9mug(>1Vf^~%go2)*!^3eNhM8f;YMaUb#{?2&NH4ZZH1Yie`*uKPSj!}DLnYWuRdt!}nFHTdX}*>}WuxNe_R zf74ZJiKT9usRxV%*G|HW&uTuhfAydOGWo>sTnhMwLmkNF+CTD-UYtvA7S+lOst=<$_`*>r?U+lp<9$KpA*A=rM! zcYe?G=MPWDcIJ6BjQk>g(=arMeE_tG-mN-8u2K8)XZWAIPc`xJ@tdHP%wW<~j>$47 zH|K0~PU&>}=Fuob)vi0rw$L0jh4f3DqYJNE# zZnj*Hb(P?|e=PVNy994We0PrGyC;7;<>46*&-i;>6ym>GN3Rt-ori$;aa@R*W-iSy zemfjzC9jy%UpvI~W#SpZM{=HFmwfRT2ED&&n9)}k(BSm>?ZIz~L8`9zn5{<8TXpUZN;5On-I!kp>i@zij<)b;WEaGr#NgCDa~Y#p10 zcucR;^p2f+Uwkk=8ax8O=MIx*Xm- z{mzsBdGO=6kM%-4m-8J*#h0PIX>+}Ln!)U@SzcZ|eeQTYeH`1CzK+=(UXJ^{R$S)+ z()2tWueA*II2ImVid1~2A=M1@Hb8feul^R7-HIH zLHDa0d>%D|_gyr69zEWNXFWWXSN_~_RkS;QMy&o}d?)sfed1fOZ^#9Q#WC^gI5YS@ zSH_JYNAZN@Cpm8M5QlxI5r6qNpz(b+b&B~-yqn0CT9@*hzPJ3K#xWP<()t~^ zZGH00)Gp$^9O$>O{M`g4Lm1dn@+_V6CIGx>w(KlHj-FTQ^<91p;kQCEyh zwS&51&rky&6{m)MQ;g@+IQG>(JmUMiXM6XQUejCMC^in4wEG6(`X__7+dGa9$6=b) z5R1er$4I?vj(1II`2RE!%b-iWwf1=koOB4%a>IxoU%J zT>!w>p$(DU-esxfBu`8Ny1(}MWCFT{Tyy?X6y!JA(#=mF=4@Yko;_>k5cD2EEVOjj) zLqflAm!J5U=rC+Js)ATWpK!L^(yx zNQ1~Jd>XZZnu6!U-`^rWALeX!o*b9UFZ$X9J=nd1&Oa#B$7244L5J~hW}j{Lsk&O+ z=eN@Cw7mGfMd-tc?_Z9ugkvf93q9CtV*dVi-p4{r5rftZ{YJ5Ox&D@k*vxkqKWD}< z@!QGYx)t;9iN~VyV$R2BJU(@s#+MuR3&$ww*)A4pgzMsr;JN5&nyWFJ{^@XB7GLIl zK}Yj!-X44;UdfiBH|#ho{tln#)-Wr>qtVkf=ecyeDD+Z&2l1OPq7Icz3K8Z;983-n3(;%plSB&kXVVZ(-&}T-JL%VS?*?oD1Ue_;WaK z!qstEoEN-(=f^)N^k;XAZDWftclnOs%`Y8hC+Ydy;<7j^P7FEwh&Uv8@_ySrW2YEB zSe}i#Zg@84t@wK4`tbMo_N&Kx!g+S&2j|%TOz`+^PwGX_S0jk~=LQ|mW3(-+U&VRe zr1-now7JL`Pt17Yq0V>gn918h9H!rCcl)M&$z%2*dWlY=+vy_OM~`Y|m_^`ay7b<1 z&EAvqz}lD8%Ik)C3-$JLLC4Dz-qV~7zn5PuZkU@9Yw1?noTkKdeJ&r+v_oUk{@Vv1 z!Mr{#et7&m=yUzYQ$qaVDZBK(^t+zq5#clJ9r^+MYM+JA!PB9UX=5=}+}F1qKK%8~ zXo{!2_RA*d^@^!-l$a_$^Y6uCvD9+p5V6$tdE*(OHV{{7abAk=Ev{N0^g7fgqetO! zx?2n$^Qm6HQ1F=KI$GXia-8oW$Bn**!f=7>W-F(AO;#XnT#!Q6z z+HsG(dRlD7uwR>*rt#@}dVZ@A>-Pz<;q2h)@^{}EW&*`5p2-)3H_z`d2fsr&*Ub5G zOWYsrx-*_K=TrI%ga2dB{yjS5(HZ}YKgSv2HG9VwV}r>|y8i7lAujXM`6uGPxa}BN zez=($eOsS_e!n?re)^yPLd)~p&GN`2H2EfBK7Pm8Bk1<;PR@bmGq}%V&V69kX1y?* z!&{OE77gFmjE&hM=h@Vs)xR~r$#bIT4~XvuziH>NuFQJsCz$(OD^`eC#j>$b@SN!O z;X|pP?u)xZuDCt?T^ae{!jJ>Z>zKXyRJ=W28sa_OetldN;yn-U)CsQtCggx$#Nn}j zd@JO5zRu@j(^x0o7kcw+#42Ii5%cwR)Qirk@UqbR(a%>8@N%vQa{+2jy(`->uV%pz ztLY!J(e^bljvvlbrN8WpW-83m**DE!fbLe0>y6qs=`#D4x>`MLmY#oO-*V}_+~+ZM zxB9`kw&V}>!7su*2>qw8M?;G>;<>msybv*vzN9z#@w6vsc5#;e6dU7c=dXnybbk$<-l6nCb!5V3xqr%XNmo4963S`P)7~#OHLi@QLUri+A8E?xzsFR~?jI{dj zZx*S$q-GSy^;Y@$eEWA!=Hq#Wv^$S}%m|3*+s1COPy8Tgb=u_c35Vs+9}@J-fkCr~ z^K|`pW8e5ucpeX*e=pwiRp;+Gs8_^WgN_?9e#_2h{Q3A&>>4y5KiA*H`&oz$dU*c~ ze$Jc2v7J8+UIH!q`4H>Orqj99Pxd={$?RR zzcR#lG5wS9SPYhObi1`bJyz$a_#RNJY`oE5m)oayz<*^`E?z5Ts{7i`P>JC1Rd8Jjt<;o%EFCQn$JA6JCqBjf-v z|IjdRxmWBG+XlTZ<~ui;1f_?|p~;T5=T&XZ>)K6yROO3Ug$@+ibTpG}Mu|HWD{_wLY#6^m(ldj8@Nui?7f zV75m5=Am30;fo{g}C=&3$J2oG<0P znIq?^QOxO>yRp8;?BvXHyY(g49~M6j@mnm1>*Bb_#Bp_!bxSMI0K7gr{i3j*txsD0 zrm%fEcEa{&JFpGeR%|Cc`6c3|Aztg_@YL4}ahZS4OZjH-QfO|qBwuo1TNl6i?qiTMk&)cVYI5hoku}6sa-;E!KUfxM@X7G5=3+Gh%W0=ix?v)2(&e`Xj zn(2E*$Me4@^bO?$@ro!}taIIqYdjZo=H_F@?EA5u=|G$|%W`ixZ-E)j#e+uo@1M=z zn4!_P<*AsBH1~2*$O$(Ez0E&=FwEWG9(u9z1^sQ_-<**CujM)Z>9$ZeShiU1_^R=I zuW>#FGnV=U`ni_teH?@9v*-={G!6-xUf$t7IaZjLhbb`j*5L^yZB_R8~pjz<27NvQa)x#-y1i_B|*<06LkDx z;d}_kg&CZa!|#hz;eD)A z`^2zq+s5e_`-=Tx(=hj7zoLcY1of@H7=LG>;IZ?2)V_LKD^7TGGZg08;C}V6*ALwG z9Q!VxhIjvr;1kIyVgc_&Eg>fB<$WX6{p$b!O#uGyf1bcL!Aqej#VMD3VfT5=Gj|Jd zOFSB0!RejDQ5_-v$XnvOyv2{wGoU5i7LVk(^W&5d%X}uei2hd>h?%Q~I7G{1?YQ z(B|qpnwbnTolc*dpRwKY z0zVm_4%@OBOmUnRKPG+|zYg2J>nFr34JSe0R?|EzXM5!)^D|`Sd<*@AAW(X6((!@pkN=aQ(@+fAO#%^KZ@I(RgMu&GGPo z%5v}EL_s$boeJ@-Ov-@$@Bl6 zQ_akg{;&BP9ujTOPf}}`(HFzTa`(#>W;!2_c{|?Av0&yn`9C}+^?^KbL3qB;?6qnZ zIl=6z9Ki$98}M1okuDq02z_1k)Xg!x{Xg`#!_Erpj<^5SFz55ucuDBF%n@|I9PszJ zJ#Gs5`l^r<R>mc8|^t^We5!=`U!PB`h#C+T6 zk3+AP2Kq#NJnTbea`^dTzx`?Buup9m>x9S6&**3I+CLngXP+}0O@G<%`1kyHGnM*Z za)F-MQXzM!y~TXqj~YR|6@SGJF_x~}H|R|su-OhV-7!I8KF^1TE+@F8y?F^@5?@|C zIyz1W%hPMNY%xVFQBz>V8$N>jU3!igPq9o)<6Di-qGsU#i1~bMx>?+(tHoh)R^M`- zAYS=RV)$R<*2&E0%^jb0XPC$2NqMf<=oj;zEJq%p&-o_&c)DEAf^WWF@SGeQ%!5*M z490EqMDF`ksCCp#UdN|eKg`IB@0LT~yWBkFGq0PTN5^{iJ6O+Rzj|lHd|I9_L%&<6 za=0F=IWD#0@X<&8$X}=H>3Kd3|K){23-B0+p69oihqO)EPW4J{vwRb=T)h2l(BzI4 z5npc&@l~J7HZRWpJIpMof6YVMH_SemO%`Vtn;^a}5oVcR6ymHJTC8=vy8TNpO|Mnl zHADH4P;-mF>j(Y*iQvWW5Hve&MzcH5fq9#+hZv1-1n>5Hp@-+veyy$-%lSRLAGLw^ zQzv-u_lCG{hR^%y`O$tmg<63ZY}wvhE;u;E5_#Z)5O3~|+55LZEE~?>qVKp;tQBT~ zHjK~4_OVm!7_NUlwhTQ>deqLordO%|1lib zYS#4Optbod^9KFBOuQ!aX&tY%dB`byg_tcD&75{#PMV&dFCWP>JbrOryqB}&4!v(N zfBz8k&GYbsEPK<~EL?6Fa)Cbeo8v|C+?X@?K!1(l^E=m$xqir{=fnwdK=$q)7|Gd;8oox>lZq4o|n@=h`81o>jbcs<~khrSq3)xK>X zpC@>+dIvlmx{$6rGsIYZR9c^h;Bw~w1@}*Xj{d?=LyT81i19od@D|*L_wYF1;i#|- z@r%#w{k)e;o`v`(-id`HK8l~>qu9w85tI2RG&Ak|sxYTWlk2V0=|ij2-Zc3#L4S+y zJRu&H*NXEr__!_Ri}!L1pK9>Y&g~#(>pN^0a^CJC=j|CjyS!(eSeGv4E!XA2lj8KS zF2rko`Y*$!{*Cn`PaYq98a0yj%6F3s>2O>eZp)?O{g}a(yQh2bVtl3Ul*h;G^ZWB| zczg2yvY~EVG4%WC2%2G=*dzGlVmz&Fd$e7eA*QEovu2T2#bIo2m7s`)9&8|1=e29$|PVL`jwblD_fgT>u$93}^<8}Ogv0mTTXOL^? zKmNbxS+@8uAFUGdmFx13m@nS{E1au^w&WY|k;A|eFt0@ocy##30|xZ63DoZRT8`@&~bF@Tq+VpVR$5lXc_y*3XD(zsTc0CvJ&Z;oOtY zjc0`Se{8nLrq-L!D^`l12F{Fz88fMx6u4 z55GTnT<@OLG~*Z_{uF;foZ@roMMFO0e_`P;@6C_UpBC?kOnFXSL*MePXy%ID(6nOEnp@j%uH#}n~F)MhT-cK@oe$^?1Z z{bKHT++#dB+M6e5hQ+d98)7tFE~mq7&k?uBa`}40=d*lup<2;8S3lD5>O(r-EP>zb z+o1;a+mHB8f7nj=>om&Uaep}8gnv%Y-xtF-(bob!Kej#FpB>kI9uwc`e3#62Itb`kc0=v*|5botBpa=rD0r9f&3sYh`l}^edx36 z8-6EzPIZBNMN1A}!Q)<|rtltej!XRr@8!C>!+YsZ^O@ff)|9B`5i|6<38UJvC?<*+VT61 z`@L37l`rK~`2D*-OWWI8Vg0#n z&eFD~ru2I2Zal|hKm9{ICL`e;8;X{>ZxIQXHUaz8g9gVwT0KIEzCfO z`RWTj2>mxRa$^2&p;yOS5=+DsI?*{(^!Yq4?z=6{h&y~I&k_5*)||DtBWJ*UVvzSc zFzC!t3)*?bmy z*E-Q-w4Qj#JZb9`uDkztAx2&nJP%q}A78$OzNGu)T{WGY>vtZE-}{V?yS6;P6t3h|%+e?N=;*MTo<% z2^z(Gn(h3xp+_oat7C2V{22R%IA{NHege<)I+}NAb^Db57cYm$L;qWjxbOXzkCnr| zMN^AkUTgV2hn(R4FN8QPUWiThW&5c(C=T#y^C!^F2xivO5C*!vBxs(-}IPT z!~2L8@_;zvyZM~*7fmL1iiegZR{GB4dx{l4j~Zp@x6xA;KYS;MOLD}>DQ3I9M(hwH z{5E2S^(2PqW6QsCu9-k{B-XFb>v!-Rxl`=2j)vD@-A@0Fd>7x7zVthI%yYc2_~d;p zU(E9R$kRh((xJ8k@l($5J$<)XPt2M+$DHlmcI7rrAkNv=d~UaWAIq?v(FeA>@g02@ z?>GH>+9rKZ+l+PT^U<)jY5SFWa%`hMC#?bNY{YM`zcmkX?7Q{}a{_dWWn2>H$K~NT zZO6vi@5EiY`1bHQtrP3szT|$-_q}{3kMVIVTYMIi?Z4uIbt2aDLhY~Oha ze#J5^o8}ku#2hij<37X4Lt=_`N6V>K{9b-n&-MAtcZk#Sl=V1#E&1%=7+N%5{5Y0S?C{TLvz{!={zJ>EXT>?ON?i87Vzih?V~Wq>K3&Bpq0PL;ZOga6 zKRv`KpL1v{aY!5$|7i3>CVmOsDDH~oVxrhA2D?pXirvQt|APmsUgy861H>hHz@=Ev ze;@jij-;<0cS&3GXuU=*v3%dvb3Koy@#3f#^-&^cdyZcSVIysdl_nGw={l0!r`=xzd z{IQK$*FL9yiJWE5$#y0#h(oq3`?lAKNA?@>LL9K2TjtOLwhg|Veb8sKozfYes|FMU z=wRE2^&pqnuI=OWAnj<`_J>j9js00n7IVZN`e=`ng( z4Nd>jkz$k2<$Wx}YrG%L%0IOpXmjgA%=Z0gc3PQ6<`3{c=w|&i+Fl(_Ytp(yd%JYo z{gx#Mh_&LJ@9niT?JmKyu^i8X$LV_dn@;Ajm?h%5@MqKy;=TA|IeZwt3XjG0{logV z4aqC=70-b;!KdK^SubOz+WXV$VzQj*`?~J6KDTYswl3zG+wfZP-S*|5*$Qz?EhPub zJ=XDvleP=j#Yc~e!D66g+NQ>DO;6b;Hw>D^GCf~By*d20_BHWOoV1^cb@oZ|l{Ouo zhW(J{am=>3=XtKvRQ6AJFM3i;5j*YYp67emx9rbiv{-7N5m)pv#T6c&&u=^PeZ_O1 zN1XLs@97v*`?*{uH`zzUTk%Yc^cmzT%d$U;6|}BcYG1I9e1D(EZQqB+lT+=}mPf04 zops>5(NUHqe!HJ;R;P=dJWl;q>z5DWw%lajqeU%)-jz3eNBh6HWVxPiJ^5Yik7A(p zColXo#9uMl?;xL8hV`j;%Fm|r`P^c?^=N;#uD!o_%o8{({vA*6Se{2_dt_>!!~Oge zaoy|X5HVkDrNezT{(;BLdz%B{KluJ)BfVuk`R#o_9+B@a56FM=i+tmI@O^E|zMJ1h zZgrpBVf~Baa;a^`=XaZyfc)(?oo>DPjcl*BQ;&I#@9VRORWyxQDE`>Syr1nw9qfbYD>^oY9eP2%WoKZ{J zAH-xa*7;_{Tk+5~KlGBkZr>9t#Y4~c9mOb`QOq-QBW8+s2Z!&b7O);H%Wdy3&YH#H zZM)<@_?~_PaoqRjgZN&wv-m0>T2FolpI@%=+xTAIOD++sE!$`JTZsYUf!}z{TKJ7@ z6Sg_mtt0D@H@$M0G2-p3Q!K;#Ssp*d@~tD6e0=ZgdA^srL|*dR;TiKKtRrz-PM16M zrR6)nkKf93_%U+D5@Du9J>c=7ziltpp}1swvK{&7_11yUV4JmV+b+dYaarxkxA5=g z;`#7J#83MSJuZIHwYFowmHJ5z_FUV!fAWW#)BZJbx&24Hp-aU$`;_NgZKtW%#wzVn%_Q+dn!@t*Xde?EttrZ-8S zi}hlub>Q>p<-Mcs;%4!}P&0TR`?owIF6+7a+ZdL?m*yu~ru8b9`h0SL&#GRb1Jr!J zhxqTew|>kALVoaFtvhvqW$~bX71oVq*arPBVzS>>UX;^q8}ci!Lrzo|c&*>odiPwq zUJSP0y|;Dm{anf^?vpF5E8kU)k;~j?dsBzey>z^Pwlx~W>*Z>%lXI+p`=H0gkY@(X z;5++$>~nrk&l}zo?QGi<>(%aZi5MuZxNU#2KhoCXtJp5Ch%hl%9G*pi^X<=da@=692A;P#(m5E{uB+>-pA3$5J@pwnSYI>7@ucx%EYd^f z1lw{Awd>HuzGrUnw|Pc?&`0z|b0t18o{UL))b{joV?rBK_#4Et_DfTVd-OHO zf)BKdF|H3fKV#0hYDZ%h=lVXLpfv$U;G5y~<6QBBoG2}f$N4Tp|C<-g8+;P{Xx{UE z^CnKP9dnhr5_kf*2bVafF~^wiY|ASj1TW$s9A_JzpQdeH8|#=eajNrl&0G)Lbo}Nh z98Cjj1N`L`@6n=Q`*i!*j&{Jew&{D~DscYh5Bt&f`kD4}T-J3gw(l6}0-zhMi@WXD zKD4dx;A7i^!L#OZoKNTDXSrDM4-ZT|xwuB$fb+GzHZ~8Lo6Jk{7TVUlxKprPjA=fz zzSzsT*blGB{GbgSlYO}+);Dk2j(Ej6(pts=zSob;k@^ zcZeUySReWdPwW5q!>hiD%WF!?WVG;4K`E6U95mzE^x@ zKEZ9q8-6f%@c?iM{-Ixu!{Nyqhxoz1=|bb49-(uLci+P!#;7*YXY?<9!FaW<_pZ0= ztRLt@L-)~aw3T)>2DGRCq#uYIT^rk=-L;7^i-%|x^D16+T#nhRVZ`?Kq2xpURN z+QBx=50?jRsJ*nIcDL=ZzIayO*RRdJAkL*dO-Z_NSvmHspOwDZzuTxZ`mHqA}?x@)3shW@k6wX&_jwbsFXxR?faEzBXf zMjyblxQ?cEO^gv-A{Mtj+Y!s)EZpgO==b2d)2!xR@$K-!=;h(J=mWUjwFKObj~%nQ z!Ygj`nK9saY+Ju{p5{t@A17!xeVAt}o;9bL`*68)(LeP=F{E~MeXXb6@Q-#ijvW&} z${dN?^(!6hm z^ow(HZ2GJ7rys4{k9Q@s}}z*IXB4(OC8!{gr0d zfAxQ{t7F57t}&j%Nr1ncr}-W1U;o4HK676Bz?iG(4g18^w3TtDKbVWfSmrj@+qG~F z9S0uMzRumTI5*eF*bz_gZLI6qaQT+O+>fv6Pd*2&2#$+BbA9w({Xw4}no$3Ap5}$Y zcfNyH>3r9H#7nmCSk0~GRsG4i=!346Yww?Ns;$R+*A55a2=fPib3CqrYinKer_U^R&i?6NmOE~3Wm~R`?K)@6 z@t<|*Q~P$V`kgVXjq##!WIXD_Vs`q)7s{l>n0 z*IcWO^pW9ByUxRh#sSXTyfpfh@7srdiLZ@S@rxMEJb=gbUHxhBratTYuDO1$?Oboi z;QX8?u61plx9j8@()Zd?``b4D_Ud@dAA_IuVdGOjb`6Y2pWCnP8H=v9>ugT7Ki9$W z*oOIE|8cz5w+-j#{Kwo%=Xme@@iDH`r)|&qYA4@!?OaD(DpLV^pqif^*j4$nMeaE2vv?X3QHuOF1?O1((#PP;D-qY6hX&d&h@8ej@ z9j|`lpU=lQat@9W$GFzGf5eBD=`TYQX+Q00yZV^6bnc@)#UsXnHt-$Wc3gPVHto~c z!adqvAGL4opienhTx?$So_?k+XbHMPJK-8*M7~rXa4e2fo0)I5jq7H)_S7ew7e3Ip zX&r5DIj+Kmw&$AJo;J1*<3nF_oP+OZGv_)qfj%_Nk3>Hm(&^*5gm zXiwWRrt#9~+cXAl*MIQ1KC7?kJ2=An+SGN_p4xfDAv&S*01p!-|m1(Tx(z2 z*8aeGYg^m!IOD#=lGrHzkAO0Gw13vT2nm7m!&1>e}0Mmo2Tu^+-%OV zf7{2`mbw1=!|3DAS^K*#uCH<9y1L%ZU4L{O+FARLKJMCTYwc_d_*`GoHojxt8GXw# z=Z_Ed0pD}I0T1YB+FD;Re(l3QpXpEfpMGu((gViq+~E4*4dd5x>kOaCoaJxV7gyQ0 zV{omsr(<&5BmU7wqrHtg?K1kD^|hmGq5nBH?PuTir_YK9=|;I794gMBB^{T(oSjSc5$|NeF?xLzCK1+gBUGCoH?cWsSD zoQ&J_2m8`4{>H72Pn%m$e4%Y|jhGexd9Phv6R=P7m-9Dfy*e&)fc@Fl*rsbd&fnOv zPp?Clxla0med9TEiSxA$eNmgyLyjLWI4{S5ck~_Wjr|$Z`k8&;X5ZC+^;d0boLk3t z?FaAcD~?Y)*oWgC?LFGjGS|a+alP~(*y@z{s9aDKjrx2&tb>1+NC zzO;XR#rO0v*W7!32G=r=+1eLGft&@!J}SD$eF zKF3k|{rKE?(Ei#Kk2p5{!}WK~wV~s8oVH`E*tcUbX1op!ZGS!kZDoCYr9JdV%N@UM z`JT44Pv-}&9q2F4+kQuzY45=exB{2y3yud*fpfBLZHVu+ul?BX^szY>V^N#oRew7d z`?4QnMBj4^`i$+sxCYwPIoXE(GPvBf=zSW7k7Le|7c^JladW|#Cvby)hJPHlc5*!C z40$>6(!7Af##q+Q*7aT6@tOCotG>r?<;~G`beTTsIJLR!t8H9w+r#C)>wEgS?P){Y zF|M0+Y{T}gJN7l=cm2rr^zFI9^>JOaziVT;W%`kL5vOTi>sTL`;TOPp`n)mhJcpJP ztKeV|tJ{WgH?FU1VqaceBmHMwNBzY5w2ZOtZ`aoTaSa`W>+tUA?~cQ|xK7`9{MPe5 z$1VmL+=9>aQT=%IX~(Tk<3PONy6U6mD#z!z^(p+|Jhk5#6Z)wBq<#4=bb;~gn)_#8 zj=}z&(};QSv*UNH#*S_3llrG?txb)uF&}%?PX6i3j?uN&7rpx1HXN_E)`qrkf7(j> z0F5u67Vo%*<`Hw1c}rV6&e7(^k~YEB`k3=HUhvY`FV1jGq_%Pa)CNoZqHJ*x`kMTy|`P}$%4R{vb`~H{%2H&}^ z&cnF1k1ZH{GH?AV&e)2jQM7aU;HsNAMVn3yc)}{>lmZ9J$($?(KR=I z^*!Hl9S67TKjt6fOdAH z@LkuztNmDq*3g&uq|<%Mcr(uQ6>X-i?8mFW9kb(i?Tz1Y-A8{g9_@ehy`i7Bj?J;!hU@Rzx!$-` z|973V&6o>)XKc@_cK1DP=vvq>p3v^LWxU#seyFco?)roG*0+7v&GES=-n+hD?az5x zKH6L#vt6&Y=^EIF{>q~;7W6gScHEB9wj8H+^UwCQnR9o&M!(a~TqoCLXl%S8R(I_k zkK-N3Y#WyAH@4v%9H)Kj7x=_^diA$!=v{t^{0q;Rx6jx2s=kRA zEEi>6{N?)FmvQBsVYK%chsKz5*ADo}b(@B3@7VE`@#WQd={vq_43BH)Z*8a#x#r_q zTL)a%ajxUsEpv@*!#di*cO8?qfbq}ua&265$LhU*_Un7v5^U49hBk2={y8sw+VMMI z$Lcfv*fxwI%fS9^-&k|b+Q2zGR(%N1(mS|PyK2vI4r5$cf6S+ji{{V=X%@igL&M-q zfBSsscbqye&{Jvz)MI&1kIBiYw;H)QUf;-3*yflk9HVU-w>aG~7{AWJa+=!lk7L(Y zeC~VZ6UX3q^c&YqI~mK)aa_A`{fuMVH`Z*|KFt@}W?cWV?zq;rrQh4GzG}N*xv}Cp z>4UC;*h;^i1`TBV(PJaNoSQy#-SmHB(yKA$y5o9%$Fc$IxK^$?xMucg4B}z@J78ZU zhk*ZS3%bH`W7RPjuVXo`7&>lfN#A!4<}&A}U*jOhqi@^4zNo!HJ2)TwZy)rNc>+H< zC&#K^(-UKU)ZfRr(60Klu`__}C<`Ul>wUxeSyV{p;Z8=y+ zJ&#&A^|Ii-;}M%^YjcR}0NUR^!E(pqSRJ2j`Hpio4>^DR14f%_8-2lba@|}H(8fN~ zZ%03N%?788>*#ZwV%?z=^i!YX4E@-68@}z(+{Qc3ru6`)i?N3u$HUX{w>iN0GjA9( zVu@)DXB&J_v4iod9?000UoftD1j8q_AKSMt+qG}8GJPk<@187qF?A%qZ+o=p`hnjt z>S5F#$T6uSlj~C}<{3h2B-JFi_dpG%x=Hn9{swiE@`CCuw+ud0gQn(3ZKeAj)O4z) z^?VxlD0rTa`^h|O$lsnR=GklG*+iamqn1Zar}|F!52#sF59+ya>F@*sr=c=dXUtcAXc<$2mXnUb~Lk8||xh%Q2|m z(2n5w5TKpL_Z*+^T3=0uzkLr}uaRTrH|PWUmg_v`m47RkHD>yhYp*}*KVU8Z^Qn2> zyg24TbB#F%EH|g%S!2&w#8WhxF%OpedyG-b@Ubz7+vq;y)40bCG$&6`E=kS2+Ee!# zfP9ZyRrf>rTYZ`ODmDM^`*Q!A8WZ;j_ziaOUX8t4O!pRBM?JlJ&(tn?){W;yc#fL; zxW;pD+>7j9V)wPVw^v=J=Yn|_m}iN3hMH%Ctrt9F%)Ptr!-DbbFrRt9Q5esZcJHtI z<=i{wxzV0CW?RFJp0Y_*xb*yYVb@rzk}l0n&a7X z;~boy7PPFO3f8(tJ% z7G4p&TK3fN&~WFhJ7--J{5FfYvSNYDDqz;9Pl6xkB?N9dB+Hhu~3vo3Cly5j){ff8$>JHm{mr&Aa9V_Zxui zj@lL4-1g1Yv^z~qQ`0N79^FOz(K>3X-Rq>^t9KsH!B+$Ay=N|{IX5=cKMyX&k+{*= z@!N>_3s-sWvF8`yAmi+xJ^kF6J|+Ahd_C+Ib_|}U{*mCB>z*&Xb$EC1{9&K@eUVQG z&%5{B`SE*sU%Brq|M{JzUHe^*j|I;tertHs1aIg$@tcKBf@jCSCTtNt5N!KXVeI?u zJ^$S?*v@Ce*TT2MeiQ89^N~GI_<-=uut)HH&zg6UY`Mngs5$QK_{eJty!uZ`>zp>?aTzVYte||sm=6-|gui?+(f^cED zG~5{c^X%v4f@eK@w)6Vo)nUuv+4_!shp8`&Cta;C(=T+B zd!_t7klzFH`$67&eyp*4X~3Zy1)S)a<9@$m`F^{4{qW|=?`eO$-{Zhvp9(vLuLr-~ z;kT5Q?RTAC8a@`j9sKV1Md9W!YtdPYE*TyX9v&VVmI_{%36Bns3Cjk{{JVF~y>tE* zZVXok+d4O#9QPr!&HV!ExhnJc?@{6Fe*4 zZo)5MVc zy%Bqv2Zt9ZMx*iQJ@JkBM_l8({?Ulyb@QQX%1cl)WJHI17f5H5j-yFQ8->W?< z92349z8d~FY#BBR&kn1F=Y}dBO4hBRGE7&~-KMnWLP)ImaCQ`{26J*Yvt;L%(bP zJHp+8MiwtD8$9pWy-@Bc<`szp-4p5eRdMXs!*2tQU3B)Mtg+dCx8g71Hvtb_9_GxR zGot}NIQxS$zd!NYe)HOI+}#we4!;h5ukPTmf1o`t4v+4)_TC|)tW?_r) zrtp$r+2g`PLWi(u-E4-Z7h80(MHdhF_sVcw_)OS1;P@N+UGCq9Urumgzq4oAgXcVW z=J#MRn zA}F^q%Kn-gSKZ41S8|O?XcI zm4VjAr)z~x0zTb3ygP_jJ{mq9b`5?5<$&4eNp51>% z@XUS3h6^S1DoFgO}_<7u4ZcQEkseGBfJ`7P`} zPJS2f{C?~7;y_#dEBK95Jnr{0UL3Xx^sr;04{6Asg)_n*1FpR_i0N^m@jPPl5s%~M zGs6j!-@x#CZV+?(4*umQ@E1I9fWN`};E#Z3+R$b0>+&pl&m|as1+Mm+r1+cXARedX z=>M~WW5vnl5ba7IYfst_N10zf9?W-iJUy$umJfa->fQzS&irQ7b^VqUzB@M@9ljB^ z58H%ShPA@e!sEkZg4b2Tx&ag8VC&-GcZc_fj|F_bOTfi@gs%jgd}#Pdz{&XblyGV| zEu0a~p8VeUIWxaH{_o!%|9!t>cxAXf(A9n)<%QwRfrb}D(E8eh2cnI5FWSsprVV*J z+LQMort%viuJ2`mufmJ*ok!jOs2Lr>8^Z_2D2?;^ApU>a?5E9sb$Cnt1Q9NBBgbHMb1g2Fp$eeoO8pb6zrM-#{O~V9^&W`mwN2I52qK zF>Dn!3{MRHT|4L6Ip>6*g#E*p!}j6b0ne@xaQ|ZA?%8+GzCGL(=JXGDKO<}swhg<6 zuZAy%j|AF%>F|esd)4nAy(&B`_?@~R1RR8uwhTOkmj`@E#|UXmP&`#$9!_(pOd_*#CR zem0@{ea^?|ytu6gd zigq*q@x*bpc!Rd)lktt^*Z9C<3i%xE$7dBA(vyD*#|IoNp2QnC7x!)(EW?#z_Q5k> z?l(~BYCMXst^e|Xn}I&WJ@hWTG0@!43F28A8zEVyiSoE;p1@b!@p8e=^>pd&*kom@V*m?UcL45m$uw}r-s|9fZe}F!|C0rBCSNIP< z?iD^1ULKwlW``^0Uororuz&bW_-Ob*_*mFEd@1Y`_6_@mgTnX155f<_QQ=47*f1~r zCfMdp;a}ml@Q-kP_-nW}{5@FC*nY@@hb&k=tPyDPcZHq8-r4h-Tn9?pq@myM7A60Q#BTXQchEk5C0n={>$@`u3B)9)9Xwb-nk z!yN%X^ILxt{uF){{C+c^;>F=n;qF;?_bY4vnzh*M#b)&8=Ce1S`Q1=l{O9n`P#86v zSFl1@Gi)Bd6!0yt;&lNppBw%W?h6aTP2msW$Kmtgt%0_FaJX*Hb#s0hcocgFocpQp ziQsj|@Y%3O*fUsXzwrIQ+o0cn8h#us<8k~?!1eh2cm2NgXToy>zxwBa=lia}(!Wuzr&(rWbN#g&D10JGFaSSbOKGfDc4tfxW%A@gf)fdP=sb`Wq+BO(t8-|g) zG4|wqc%Q}>u2z5UI$jcRygV?EL+!!bzze|F{1f}*t0_EqY^;yqSKN%3@vS`J@XWOVcoqUa z#@RmStJBeM3w#a!InB*Kr{{PjIK+HH=ktJf3eJh=jf=iI0WE2{cHn`F73fXAtyqF@ z!AJOZplv@8coWYGONYDX-#x>hm(9Oy{)ORJ;q-vFPY5T5Q^T3z-0<7*r*K2a9+|() zf@KywCGbGj3_K8gYx}E*=Z3Yzi-P^~Oz3SI`=Ic%Ks)~_+z{w|982jc`*WGpD z(S0ji6!5h^?>Elg*Kd5vPsyWwD9DfbT}?T%cLZK6ZOsEQF3$xk^l!o)fzQEP=4HQ_m7r9eCLDHabjx@}z= z%qjAF{Nv9B$G=T@d-zEBZ15YPhlZoVPbTo&(tH_l!5hPh;rdzE&-zi=KKT7Px{|Ld z(JWrSdh*-Omn?G0A{Pao0`1ImRwJQaLGFrwiYrGAnY^3)8=qslpr*nz(D>3m;~Nic z&4d14kgs(;J*Q`n@XheUp#Gn4g_~&V+rrS*YVc@H``$9B#o)#8KzJ{X+jA|Q%Z(SxqVZ*RR*fAUsXy*%qwla4=X!e6B`w+?hPuUbrx zpNwZ(+p%pI56;C zP7izz$1Nx5JmeYV6Q2;C7Q_u}gf)Zs%5|o#^$Gm#y6QXTb$Wn)R=YvGt6vc#xmH^Hg9`%9j)UU$_bcT_0 z0^j73{jTxL!~X{SjFZm}XN7se-{*xJ!g6z#oAbuN%iwLSyy(h{ZW%rsz7h5gJBO{q zx?#C+_nf=u{5hNy4hTDkcZ8RRX9wK>nDB`3@IWWe4bKc4hBpM>^d|#7&i|mjY5iRS zZ-h7V$RM6DPnmmg%oVe)m~~9x#r|*L&jO9OL*N07 z&l7_8%LJaS-~ANh^Ci{~4+?4zv+NAoRs8oUH`bi8TY%zHj0 zc$P5##dG4-bMS<5_VIxiyk8jK;W1dBmwip(RXa}SV%~aZ;5$3#F9f~^jeXDrv@Wimh&c4s|+kl(v5@J0R=m{Jd!^N=}D4lfGt47-KH!};O1@bKz7o)T6I zUj2RhoZIL8A@Hle5acA+4vz}+XV0I>W7>!9FF9w)Igbr127K*34yMN+7R`Xz)`2LoD>-sMN2Z`;+ zGlq97KgNT=ue@aZx>r!QMsvR?s7br){<~&kpOdO@^sK?b+57gan4`iu;fAo-B8$z` ztc#KO9_s`(9iH`}CXR0ZV8DU&{P2bGGj66$%@dy6&R_UlxITy@%q9HgXN4Dpjl!nk z&0))+J%CT7&+w!V4s`P+;ihoc1p0@0otKTb_X|6QZ31n*Zr~~LN9g%VeP;c=Jog2` zTq6hTSmiz)`xk>Y0;aShF@eq&oKBQHTuj64G$%Ut@V33yZdPcfl6m&f%#!T84y#<(#n=KNj| zyW;6@2mJekuzQdhk-u-LilKSOHfmPz+Ra5xl@Q83tQT*{?&+wig*T~PsXU`7uj$WS>mJ0t` z8lVs3+bc(B7X9`%dz9>On3Ie#?p1{^Vf(G1t)jkDEYDK=+!Tc=oRg z;)b^cnixOt621~>>YoN4wS2~7XFqnPW?X-GbJ#Sj7gh;AduZUPSyvwN?7;t~x8*N4 z2s{zBH9rdGHC{h%_8w>Z-4Pt_^`IcO;Q5%NHVE3}p8Nm(%-=JsZ#*W5Pvp|v_u;+^ ze8oecb@_QAne+ab`uYLWSq?+W669sr)kwbsLBCj{D<&&`wGFW_-Cz1F9zaU9U_ zG#K8)@wB~t(qrZb-^JaunQe|3S}m7FUuz&b^ z*g3o>@Q>FHtA-~Ayt-WALGvy67>tNJ!#~5dfme;kPY5{t$RMYP)8_@B@xA2_eb07q zGu@7x`P$D3YlaO2jlFH4@g3g@fp<+?Kdjnm-p49|2fsoPGtk^P+~;!0;xBWT*i5|S zdg=@2Pu>QO=6#5Z=mcCXkNoVw&lc<8;b@s=tlh${fgeh1yGM#oDsQz& z|M(-~=m&M5gxL4b0Uv)X@TCtAUU?t%_X>fI;v3LaH-yD!FFxa68nf;J;az_}91#u; z2ZV2h?*#RHYW`^7=gxWVOl|Waa}JsF`*3ThBxF(bdjG0U`hsw1_(Z_Zj|*x)E)TyB zr-oky{>$lsem*Ch5qR5b^3;8Z0n~Z(xM91%ci%1S70gk*?k@(Pi!GK3m(99trcPwL zS=-Ie^W$346VJ_J6-HF9NYmhsO4fd;2_`4M;u52=yh z%dQ@D9Pa)jbCR4Z9;c6a{QCx;2S1yxp4~rlIp2E4uu{-Z%&z%$YR-KS!#@CIm3z9A0dXX8+s8NcFOni*%~VR<#a01YZQ1 z!1uu6rw2OQvY&=S!k2?w-$eYb zdV>=WuKrpriP~#y)r} zaIect1AQtthPwrr^eY~ew;R0u_u_lGyeq=UQJ%NRc{906wZ?LN*1t21@8c=nHeD>o zKwrK;yf*Mo)(bTCW`Q23L%$qwn0evCp!R~lpiUF-%H83?R|I-lzK>SkE6~b3XK}=7 zf%cWpI4S%j92$5YJR+J}zL`e8C*bFm!@5Boq#C2Q1bKjs!v^66L2UNC@a&+U;By?# zfBIXX(~k{b4ceF9b}!9s{m!d7?`MHGzhih$n9l3c)_i>)pPcI*-G_KVps(cswhM0# zFAVPMSTQ_0JUq~>c+A-1JJOSC_ju59hjNMHUNw3=>dgZl$6vHEZHnW02SbbU5acX< zjtB8S-d4v=pZm^X0q@i4#`l1~2XE8-C56;}>&aOyv9ul`t_9uMTy z;QkSL#czgv0)N^h1;%)_pcwl)X4Hu@AEmh)9}vu=1%*f^;Dl26z=yeAB8?U@4K4u=H1thRs$ zq7R=M{uJ&CbF2M&Wl-yNQjiBTuI1H^3V#ZB2R;ce%IXe9umAB80hVTg1QYk%Im{J=R9=A-;kU9P*A5Q&!}GHJz>+}o)$UAJ7(W8 zqp|rev^0%r7>A-uQEKa5sN6ygP zDGz5EZ$qw)UrRglY=NeR$`&U}CHIv_7C_WE?d^?L+-|D7P0sK4M`oPJFBQ8+yK?7P9duwQWBzntZl!|uVh z^f$hU{3cJvahdB*3)S2%PNV@jmVj$?apldt`jfWP?{d=Z+MHoh_NL#_%}1YCSwxFy^l=;jpyZ-v(( z_sHvbd!XUfqI@gxXa*PG*gfC!q3XgP6P6F^T5vP}!+nS}GvCMD%+Jye<}`gm-}p^1 z=eahn=cU1X>6-JY)al3-tF`1!i|6G(;k3U_a9#KN;W@Fb7+0MM z9$zicU9SlETa6uU%ah=j@K$&fv^VeM2Vq{|JLBS|yN6}vux@xoz|~&;{p9fA;QmRz zGp_c1p2nAgIzHN-hJJtG$@5g#3F^&ow)lacN;^9j=XBKs&T)7>7tFd~MswnA+=I7w z3?B?zgiQnf#xqX~%Llr4so;JZ9`*1}E}qn9Y z)gZe@JeI-7=l2{BJWR*CXPc+>xZoaeKE|3sZjWa5Tot_hvOsIUCdeh?=#PY*gZt&b z5pe4Pfwp$uKb+jJcxt)D^8=rV5299&*P*V)+&%mcbE06^JgAT8JLW%Ho(FC|a2)ji zB7sk%EqFxg)p`Z5?Ddd&VyL2{vZ5=+vK0%yqe(=gm9-8jtJQVp-+~K^;54hKP@=e7L zY7cNbel|DZ6!XQ%v)a$dE%I^10^$oEzFeu?A}zgG;Cs-XHwXDeyhn%fcdfH*z}Gs& zZ3}Lj@eux!pFn&6D)1Oi3cL{<${(R`zZhsr)N zJ9w{+4hlaE=4qZ1U2QI<0r(tpit>rHv^w`shrPn#;nYA+TovTT_1{}27|+uCbI$_f zd;BI?_nN@7esU0Fe=fN1?v|i7;MIYSGH&+|M}(h+GXn0C^PeB^GEM!QKvRo{X<6JX zCiXhw<1KoQks6Hm1)cu(!vJ?`AvMylfEci5#$K33)cpI$DabPsjaUH+c^*lY81NaHJ9-s4TOjPEQ#|ECY8V!7o$8q>8C*W0Y z7TyxngYfS^8-}K)nZ+F-SB>9&rY+59>K(PIIY1qWxqr8y?&cFg9puJAoFYcKw4A86 z=$rDXI9eRIOn6}M@6PVIzbW8ou_BH>InXb33XSr)@X4@kcw2Z)kariK%KNxC zghr(^`N24kMppBOSNRaQ)Y$auZ_7^$bm(t`u}b^mLHvv()ywh5#lK=-9wbkO)|W#9 z`6ab0?+;>QbBcN{@w4wQ8|YwK8V{)Fz_)ym+k*Y68KLFn5_!_*3Gnwn0zIoilvR%jN0HIbIyj4`&CQJbde4 zR>No6xq-g+-ZJapQ@o8^_X~T4-2xBUz2~0{H1Y?7+#oK0U*KnK7x24v-Sfh0k>5Nq z(DHu`_XfE^bM5m3zg>Ncd6{OGkKosvYjHJx)&}ygt~Fnu2gOgq$>!RTcc7^+3;K)p z7n|G@%#-rjw6*>vro-8;?}x(nVYjenpq=*)^t7CmetSx|Fwh@rd-;3354BMHg)@V^ z82?qCX}!R^7MB`l_zYi*bLR#A0Iud+(o;N;r-W6*8i7ZND_1d z;b?q;Bb*yQL)@*Vj1T?d@WOzz)Uc`f5MR)yqh^l=TrRku183i}$UQUqm$$(Gz}4Cl zPx2%1^+AEJ;d!NEm%W4co`c~t>l_u14(^4dwQ(%{Oh=mot_*mV#=bq!>NNYq0)JOt zkfv9!Lf_NUYlU$R>VBMq&z+NN{IcNuw+u8ce}lKNZ!kCWvFYI7hKs{r!li*ew(PXP z?=b)IHq3v%kB4oGZzAr{zS`5Ape_aJh#UC!;)`R$_k($c=cBIWt$~hyMo{A@mx>70 z$y^rB4dO{2AN~C0uygoG*e0mm!;>!#G|mfxu_K=|yaF7}zXs!r7R71C94^Fl^pYH( zSKLT1<4)`2c{R=(26_Hh1$9_&4etozU-uH?V?6EH@G{<351_`@+_6y@$M?+cZQ*zF zG#uBMLk72+8$5^8T!Le1aeAIU=E>09Jd8a9A6m>n`}1#rm!s|sXVI7H(WZG=9oyC6 z!pZ$D^U4<h9qy!7>_|zd_sX8R%v3-a6C3%htoOzH6Ob1OEMZ z_(a$-d?DB-4Ni+68));hg1O7_n@D}KDhe_!F|ZdOkqCq;wtGU+|q>gHg);G4O@IL1-pVdM5g!FUq~8?(O%bg9}|^&jF{euo+m zoX7`$TENphhL;EY{GQ<1A-jY!Z!l10=!qR!MBhHq>aZqd=bldCGwn`h2dey zZSm68fs4n@A^cJvDQ}cd;T&k>tpZM>8F9v_pH^RNe)13ZEevf$f6{yECtzr5e2F*B zaWu6YB2LAv{9zsi9vZ;C13vprus;5!&F>97WcMuLGj-hRG2{l#75EkCU)RfWT2$>I zzr%Oj6UhIt4VoE`yT%U*&J%oQx!kCIn49r8?som@T-tf(KtJyqb_nJ-c}X=J`iS=2 zD3}w}j5rS4)27;q7sI!q=k+@}|KKpZ4e`O|;nhJchkm+2SUWsB&=ZdjO9y@jUz&#I z+x;>8I-D2I5B|pA`aF$&b+|l;C217voE3P4M+O?ky}dLI?M>hC2<3Tb9yL?O(k_7x z=A+S{`s!+NvoVX?%_DqG_Z*r(=4)O$%$xP(|KkDAelf`9@-O85%u!+k{7W;75#)q@N36ixQRics zISgN0Z`ZJU;B&|k%SYgA`CwZ6kYFyv)yDg{ye_yu@ht&AZxQ%E{GAU4I+~vNNZ@;@J>&_= zq1c|h;<|yxULiaz$S?k@=k4KVe2j~a4?hV<1moj-ffpi2GtJGsLoq0BR@(~RcGobH+W7D{VnH5Yw@pXbh#-W3T?($ zIU)>Aj4yD97+oD5URIN^PIyM3gL%Vv700S2101_BMx8KDiI96WwV*{VU{9>M2H*6T_ zesd8oMlMr3;9B`k?ZR(1cd04FyLcIgiyv^Y^>8%cXZ0MTcAig->jys%{{ttx7PPb2 zg|{!3lCO0wHwzyN=J=n5i^JXFG2OBD{Gh+88;}$C>>Fd|?qKZFG-CI+2KO%DFYzc& z6Q9Znieqt}IQH@&kM@s18@qt&UzZIlhLwZbs%M2~hv$ap1-{0b3jr_VV_FydZQa$w zssaDva``4Yc+r4|N1o|dvwk&GW2-)l?`hs}{N@qQFTp9e2KV5Q6$37#b$P|O*L;L` z={H=9OT6M=JdAhc^JrKceOd4wT9|J^`_Z9*qJeSh8v`!oS*vRu-ox-6=5`N*8dzSj@hLB- zrh*SC7&UHjEw56{EA~A;sNbTy#LQynQO8H?@k421{IYpaKS3+vB)Llbq?W>~zt;&H z1YWUM>(R{W$Of!Oclyq=C*Wtm>S5eNj|1I*w@ko|2K?}w@h8s3rL?bnARf044Q_pN z0NsmIZG--`JvmqR=-apB;|0sN%GuGcxDl7)Q0Kl=F#l)^{74T^Ygnfko%;(m(E&Dh**tcWAv$)bde7F;5!eIfQ;$GZ}fAK6Y9Ov?Y zdE;LFjbnYb{{+?@HGgp{9kz*&uqfKYN%g3xnGx_b^}whmhavG>~oP7!~LvFSkT zcyF0?w+a3p@%eVuTG-|;friuvz1kJkis6!yu7(2wa|-h}n|#M8$xeH`AuFz}C! zfoF!5!xIA^Ld>To$LKqtw!FCXwQzn6X*brb4c<#~+ze})@_oc*YoxVE|p zIXn3vbMo0vDoJb_2e6$10jJ9MemDF){3@tZ(?=fJZ>Z@{ye;Ez z=RiY<-S8@HBOb*axPp$8|G+i8M=>XVP#wOwQ+&xw6l?PD#GX7~o*>Thj0oN!4&#yX zA!tXuyH$995X;lYbS5sOIS&freYtZ`=Y)rUFahtvGV592_jt3@`E}Y`Z7&`E_P{H6 zb9ikqC#ZokKhW#^12t!km8Rt-jAuj8x8iB%=R9%n@Ilo&@K#68iZ@FTiM4UFxE=p$ zPo9i;pNtG_ppAzPrf_2(tFE%NBl1DMK8*IJvrbm zvAkD*KQ)N&SDnCn%d9gu$d^4fJTBn8Ck5wWyUx$?4bJrHZ+z+8o)v7*epU?hs%_y- zx`K9HHn?A8aOj*x=FI#q*zG-Yk(WTn@_3CK;3-@i@ax~gjT5N-w64#rCwE8t%Hv%S z@bK>fEvybyUB;P#C&9=2Vc-?Z-{Ix|4YaWOh_{BJgYoZrVel^vOiQU##k=avMqS3u z)erL@@aA-F27P>Mz{&DE-s5I<8@L2~Z@}MbJO(Vs-?-YnC+>++qorO8SJ8ZYO!`ke z&fBE3=|)--|IwMeQ~Zfr<-zD+wb;B@euf+%50+-#Wb({DehB?K;Cs9T_YU%CaW!tn z<#a7C7RTGCeLDu))N%0;)GQns@GDN`i)%Oc!iX)z8}4Q1weyJNRq?60bY75$b-nKj z;voH8KQumRAY)e?CZ03K#hYTwprQ}|{4GC?Pw^ArDxCV# zKr8YbaNOI&J0{pFj-@4OU7V-R6z_ReZ)`bE^u3X%@f~>^TG@ASr2PzT#Ow4mejM=J zHatmtn-k0rG^(7G{1q-Vm*B(UpUQ9f3}0Fg%rRaqqv`pbzNf8*e#NIS?G5k*#M|O^ zZ7oi~sbXYxt>RlcQ5`Ssxn;nkbQ(Q|yTrIS@(BS~xULTf^p%O${0Z}eza8YFVX@%e zcHmKZ^}PoL{6eeYAKS%CI1HcaFZMn1RgMLx(~suog*76M1^n!IaUMR#x6cWVU3`u| z9sl6BC3}{?`Y`a`F*!Hz-m(V=$Lw63@BLH%#(9o&8r$&QrNa2Gw(w8;$W0g<>V>_+ z^fDgqZGn%AOYtg>y&~XH8rCZ>0S`}e@8DeZ5tmPDBk-&`ieHA)0)AB^fn#|Oc$Nk} zAjq|SG3*>@;P(Z!qG}rGUYhr1;l+XXAfL8YP$NwHs##h&$c;Zf;NYc$XC#PmaR|tt zgL+~d?R9*vHe$hj3ubDC@h?4mS-3dxE^w<_w%>%IXV2`p8m9ztwcH*4!eca}+&*rj zA-@&8f@S;;unu@Xy{>y9a5Z?pFh{D}ph4+R$3R;S$=iDCDPRl#WKbot| zTjntBOQVWAb_{&@L&EXF9H}4ZZ_5Q*g|@+|d?{m?Hu`QbzUe39UH(Q4D?cJHvQ)r9 zG}~$cR}MZ^8#J)N;z~UD&OkT5cLFgvPQ{IH3)aJ%)4;#H1iXn?<)zg0 z<4^m<-S|>{k^H&boMWY_Xa1E{%|Kc@V^zeXd={x<0hMXH#47itu#E)WW^R?p` z@HhQwos|O3h(m49@!&Z78*>4V3Wtfg<)COnJcW0{af5c1`!oTYfcxuKLeW{jG{I3>x8l&DoPK?G8 zH`1Xr4=t)*j6bVRSB{J}rHSOq+!JLi^9~bH zAIu}p#~h<(n)Vk5XnS>ha{sQS>#rY~@63_rQ1Od?uRqfSyr!*!oGY)4|9@Z*=hH{V zIX|?+V*afJ5;qe|5_+ z_!EZ$?TJ@?Z)i_kj63a*ckt}6YIthEpDTqGg5&ipF!8r~CFddbrZIV(Vs3nh<8YaC zS4SiVLzl}54fy;22l;UxsPD@$$>*FBifs@_gdyTnueG z{L@c&znXcD|LNKdeJaMM1;yzjHXpig+8@Mm;^k$-(t$P`n$LU7@GDNkwO(;1?zG=h1-bUB%FJow)hD za83|ApAv8=&O9pMxxsOAw)CB`qTXtN?}^*w{Ea2rgHMLjXb~E4+kn^9#;UI}_T(t| zqHhU&(NXh?C5bR8EgyOkWNTy(9H(`}SqOJZR@J z^rrTt>0M9PTE0=gP&57GpdayO%-doU{hU5nFYxWv7Sc4vuAB#+FxKfTyn;K$wQ7yz zzVQ)#Ctjy1=|G&PPRV_}o)IIDt{$D=IquEmdww;jZBm!z{vwcjQXBB~z}r+epoYMA zcbUL_0b@I(rog&t3;gXf`;^n-wbHO+d;CiC@&Yyr8wAHqqvC!%>wI{;;&{SKygu|O zy?sRxcgw?#9K75)Uc{C1+3w%sY05+6J-~zGc{b)GZF^{t_ogf9d)jh*j~A-e!o5$Q z4d%0*!uCN-&u^hQ-xb95YCgr(;%jmC;7D_$>x+wM!@)yzrV|va;-FDKq`njX{2|bD z7lpqBzRz_*9BO?K!}^;J zI28PikFDdobTAFTm(Z6-e0@vzq&o)3=$Hn-(x~Dsd`gQN58~tf11*R*@#P-j^FjS7 zO^7!?9cV)L3($gibDA&3(jb=)_;d3x;%m8ITqv#xNBEUE zV0e)DL##aF*5aHrJNESM*rav+Z%e!_USR2(S2@s4CJ)@jc4xbpY?pl zHf#@{P3N$@w?E766KG9;kDQkrmiLz9MBXD!IQWczz)zrOwS`ylG;NFPe6DR^cn#W1 zJ%sgf;lTLr@E~l%wtbGnwLNWazM(nkdVJ~HixtEa`iNM@ysWSDd1wfpym2PKEA}&1 z`EGQO7*cF5_rbrT%a$lVLaVJ9Rtq=>2Z@{cmN*Jm$_Wjw92_UNgC}v~LBYKPo{w-! zI48)*yKmr+6UepyE?EBSpqBK!;QjgG_rbbqUcr0oTj%s}e2}9b_t@iMoch@?IQ0YF z!|Avkzw>ZD_>(@J=1w^@J|+*7F6IB4L)0&c(Z%K3Oid6z#C^EZZ!?UXx4bv_o4-sq zdi6J6G*|s^kT0SmaiTd)J&)Xx{JQIeA9;We$A8xudTsiUnctZF~ ztW784Y4JBN7N`1$R~;*E#f6TYzFe4Nr+o$SJpJjr&e3-8r*pM!`@p64XTQ^4rSmrz zI9})R%W!INp3c{KYHP&U;$vFq5dmK~F&Yf_T^{7i>8+E4+zl^LZKQhv_76NlKH@hf*k>kt zvd@1!LZ9s$_=)lD{yLKO_?Hx6N_i{sW@b+Onc_QB#@VDA5&#O)!54Si`m8SbR=ELN1fJ?@>zTUe2xq0SK63YAA7}7+VZ3#y&jIifWPMd4|RmQ>$ylopd9f{C{zzTBU7+>w+WS9`NMW z6Ntn4&+fM$yvfg$hmmiV%aOwt2hR<(+VC<*433L%4UWNAwBX1~B)A#Wv&ZZZfmvc~;k27(;nBI6Xckr|Lp~leQO}^&$rtYQsX1|4sALVYu z@H7di?pb`I^I;{Z!C+^<#)u>gEwg-yh-oLmD8MflitKB zyg}ZdIvw1E1La%fyVXakiB#L8_Gi?Vstr;jq*h4nky=pq22H~<`Ej4CD;n^(<<^zo zepiq~-zKO>+9GToMxBz}`hewI2J5~vyf=Ix*!Ff|)ZVL8b=+!^@cXDm`bqbR;#Rzh zQ-2-IAAbs$1-h0NHowr%^r)I7UZ#7Oco@UW^jx4X1u?kmZSK^EjSKxsZ4zCgW^boJ z2Z<%c-0BnMM&y=xfX@v)NwuB$krwk@4KXtA+drriqTz58zN6{9`uphMGj;B?B<@ms zZ@F4eHSlyGZHWtMOZD-W1kd@nGF%hXcHTIt`&5^ISGYIK51yOny=AvfJx3OY;!j!> z_uAL7Velso7sreB)iH_NohQ$gZeBjP-#}fKyqB6yb)WRMybr$%IvAOw8T&@Q5{Q z7<=S@#P+xs$Khd|jhFE=uHzZdw$r(w;a}3qxDL-5540eE%J`uD2EXwrJre_uiKlsy zBj19*#M1mn+Qhv|G@x-uv+%-b88t2RAWzWPqzgxUNdL&4sN?jxm~g<~JihTAvE_6e zI$iI{j}uSRm||P8w-_8(h|lrn@c7P}sk-^U|39<*$r<6aK%2^S%d^mUxR$o0BZ2+{ z%fNg36Ccx)2Tk(gc$toMFZ{wa`d_Vf|0_WaraB>*-I&Gjike zy2pU#bo@B-oZx)V4;O?#2fhHFNwp;LwJa}aprS^oQk?0`RXl$_?0Fb`EGd|o+Up0NbtN2 z+KquG`S~DDR?|N6)j0FRVcMS*Gs{=wPqlq&QRJ*~sQ8&q z5krq03mz5o;nIb{F9XZvYdx1rz8c@skjC}$0e>3j{5D>lLTY(mno=CfKjZ~o5%49R zpsOwlv?lJ91HmJ_O}v6NaSc5tFHM`dZyZO8hjHUH2Ody8rZ|1N?&q7;`H1Op8oi49 zaNW+qeL$ZKbSjNVchZZrr~6Z49gY-1jfpQ$2=pHwn-@+EbSGUn;_-94-$+c(!yNE8ZE8K=aZi!$+SefgM^Af}J9dAO z9F$xTF9FBP%Wn~!r<%|W0$u&IFgNfp)hLP0{TsDOazS!m?t>rDo|cK>c^T8sELptz zrh&y~Er!Ntv@72Aosn0%GGElZHFEay>f-(3i_)$%D?RGEy2nS%jyGQs)(G0#wbz&48Iyj8B!(-q*UK5}D9PiSA;5+{f z<8$l!4!y`vv_1Y2eiSG3L!WemQvGd4*O%Q$^_Y#Q} z<&X-T{&w?t3s_k$NlozMZ*f71=>j`s+k33ySyQ$Cx2DE}!A!*Q<(cx+l9ju@N% z8_y#hH8`{i9vgA9xEjRJFk)<D?|fUjQLdODNjDn%_>e{v zZ;Cx};k+l%u*1^9tCcVgmq#fO}1KtCE9(vOEX;b`-|HSLGtK*?>#p8TR`5zjY@4$yJ zhiFsRhd;vi6%&XV%#reS`j!6A4|4w)ZscF7-IH(NtI;#!MVv?*ZWZ`)yhgbdT92+1 z`_gRkBJ#<@C&WjC=kOGs!(HNFoVI8X-->e|5grqk4^If5+41!7%&ht(;dK*uziHScyfnNptQA%dw)^Zey02nKa<8EwUnRiUU1rw{^ya%y@{#iY}N?8MsWgacdW)o9PJr6#)^CjJ)x!tFVcXx(OANd zG#*`Ne5vQ*I~s5Dsq(74L!2hpDmS2(hWS%!<-U8uodK`i5@g4DRJ+(zAHZ@jGtE>DUK{kMpw*;PL@}0Dsi^ zX_Hd}PuBdQefbjRANr9GLgTysUk?0feU66o&$v;4%+vP_20Y2z61N%io;UfFpcY1r zA+G#Tpc$tu{^aLwK!O94~;i;<<&hy`G#;)xIM_d zO!MnQCcfg*J!=lnE*o$y9-JHS?UN>0v1i>qE-Vvxl0EWjMqlzW<;5KfO)1v@MxZC1 zBj40H^Bv`<)(UEZvoRJjjCr?Z0!tlUoIPQC?fEf}G6Lg4*u6|0S=jjz)fkAm{&? zKWVwan{?mcOY-f$nX= zW&XAtza1WMq3_y;ZQ)CEK6rmzz@NOvQL|$oyhhs@UgNkoX7}pHX$P^@&H-1BddN|m zgC}{6{6*fPc$cR*otII!z3_c4?s2iqXMjh6#sqbe<9qm3UPrx(*m)YfN41o6C|!wr z<+AC>d+)nEodeJs;Dm{od@g*&Yr{#X|9sb6Zj>oINaVRbo zKg$XEp5w#6pAS4xpfUFh_!o!L&gK>Q?h$LdR^~m|d#9l9>9=AZ-ndwM{jg=&DX3`{ z$I(CPA&qyMa^-*{NG-Db4Y#ki}R zvbE)zU%PPagdgdb^*J^!)0m97jF_%1+Z%8GB5)plxNtzPbzbA(c!Z6PFU61Par8C3 zbxJUD6K=ytxQ_1+|2;q83*@r|IX%A13&NLTF>GgcIL^gobV__jm!!{@AC-#}m5Z>I2I z=#z3*=qqBe^ccF0`OhC`E_>$%{puOPet|fK7`ClA4Sul|1Aaju?@;UmU*bu*HF6{y zvym6c*8OB4KZtI}hUY`%n`E%yIGl+0?it)CV3Thhu%{moJUFOb&ejGty8T-YzekTN z`2qMqai#bU9ansaKCEtHOD_)&44N;==DvS`zi=fElPl^o{u8&cg}Voif%AHfYvmB( zK|F~QL2TT;aj0u?EBPe;L#L#d(nH0_`HIE2`6u-id;Ele{cT+iopQgRaT)!~4@s|N zdv6+)6Xg)H!ReFYGuK^z-6R%GpS&!nKM?ov1@br2DcR!eVq8Xd#Bp>ydK|rv9yi+H z_TjFAUG6x2vbou3L~rz&?&x^0v90YazJmMskLa9@MYEOYhjc^wfZX85TiDUYKRbG# z;F;6)Fi%PBq%mlDJor#9NOO5`+{kZ@N8`5gnyv46ul(sepL-te6PKnh(hu3@w)Dok z29AxoCp{BaIxcUBk5cZBcr0C#Us8+)XPO)o9^6@bo{n_o0C607FZ547x&FCEJlwgL z2RN5Lic{&Bbj#ZEe3oonx+T5r?LoPbK1eSd^+`O56UBGHcRFObksXe|=zsEp=#OlC z{EH_~4s3n5-|^!D=PLUxf%4r)`N~#~cMvU`=3_SZ5=Y}^98PD&v$*%HfNb#Wd?VUh z-?JacPcb|)ob2bLvPMJx@e2Yze>r{h6l?J0O`D_UD;>!^r~m8}>=wv{!H@Wm4knI6 z$1KD_@Xg@?zb1}kqvInwq}Ve5Ci@#V@)Oea=zRQ_w+r|b`4{;j_X?gh{q50%<`0@b zICxR;g5Y^mJU_onYM*0=1kQVIuy^pdfS;2+PA?Vrk!OjMb2(4+%%xNGGam(cf(O~? zX>(5VdgzO|u{k^C(9y2Og^Mdd?l$2-yof)YBgY4PX0PMUu3H>+Nj^>-S8kIR$ZqJBsbl9q}Q3j{b$0sxRUs@fUVBeGm`gpnZbI zxOZRJeG=z>%H-FVZul?u?U*>YbN37In%Io^4Lh06cYGlBf{*F0>cO_goA?q}vaiKz zYX7Sfo)&k1Za{zCH&`C18}3w=I@4eAD<6V;jms-`r@Mku^K#y{|*mcAG|GiXK-wA%na;*d+_Gq$l#^H zfx*)Qy!+5VyohaH-{o8qZp54ZGJO~EMTx)DJNW|Wy6klI=7;4sV6Tg7(L?zY#LNB0 zFdJUXpM2nVAS>wAazx1!ZNdL3PDIu$540N&rn}NP>7v@2|C0}-xtG?{ixGJ*iT;s0 z**tb}AbRE}1M7a?ADkM9{m?tv=Wh#M8;IRm`zZGFj9~Y`7?4-VPTwV9tM3?W7swUE zvHYC34b}}fe2R3@ZwK_$3j+C%?0EMApT)V^ z_05gM*T5#aC=g4>+c=v)R5^H#-rnxXoxAS0BT71^ovS9C7iMfbylIA`ww$I0!H)ANWx?#=^)`v&(3 z?it9-k)PANolWze_Kky$f(-*(-!}`KCx=JQ&o;q!fjpsI0y#WS4CEBy*K+J0d3PnA z-E)9_*jVn7>*O6hCQwiJl8fY+`qM8*e@x7tap=e$7s(l6=k@)O8Fxmv~y&g8!_ ze%VsSB|THF$kKp+%R3XTJEpgYxr)Jh2M!*Uw?g;)O(51w{}h9u7qQp*^5{_t1b&x?k{&04F~_*eTdPV865T+5MXZ8w7MxG5feM=J`r_P^34K1N%@j|AM9;82<&b7McSPo86Wab zl4)|vPYF&8aG~6C{>mc)e8^{MougbyaXY^kCEnw`LVT9n1pJkJG1eE*E$Nl~*z&^J z;&=-`$;rV@c#y0;Gr&*c%ydb598m&s7aEJ?xGfI0 zpZ(uByz=qm`GI=yZ_+*SBt4TJdvw4jdHD2w=W-p{;B?8S2KdrDcj$?9$mV`mN2DLp z_2_@}J-T609np8T@~9{JY@cIvMCbbK9`5J9>~%g%Hh%q@w!TAso1Zs z@inf+v21d>=V+I+$^RDoA)r(8NwTlmoudy^?m1l(eCF@u%aeD0et?_lo;U?p(?{LI z`TU;XI=)e{BG)+o>|k|Z+w0{mg70i#IqdX1oF%qRujLo@U7peiar!al?Ofl*p5%Y> ztBP-bEl^MO7pL+p>Z^`?v~+CxINyXnZAYgh3&p~0xpIT3Os*em#x zon$=YP`ZNof*95h%qe;(d(M0__sm0c^2k6Oc2&SW#H)O2bRY3_ylM>(9Z8%Qr;3xX zXT?v%iB?Q?Py8a*B5w!p{4fyX!AtTd>4KXFe3jb;xQ)Ju=i~@IIoKn3X0U&N58361 z1uqF~>8R}Y0|MtP57_Fi!JBOU-2?aAC3s}OhTkesc6k%O<4rl7`2J6Uy3$F-jrd^Y zRM9`_n*0);Ra~7u>D_QU1pFIpa{8q9!I!@e*x&rea*J@KHBs^=@g)DUHh*0poC5yl_pJZb*1qb}&ci!m_p|3a z7hF&Ob{_q_d85h)y7#R9t&Zda`0O~HT{)hUe8X|tiVV_^(9hb7ABCLMuiC$TpzWQ< z-z?5X_5nGiJ^8P-x$E^=;a>WWuf}*V29^b6w7J2);g2$AjcGatU4yJ=U(p5VB>xKN zE9^IO)O_VX!?kR_rv`MM0|I^koa;SBIG653k78e*8^}lWz6tgxoeP)n>(Oa(3?Cq$ zp&U^0Cvh2mM)ocn`1OHYAa*ppj}L&Klul^PBAyd-zdT@zivjTq@NFg}6joek{$_zZ zCGmK9nevY0i@J`#mF@nE09VSX=ZnRe=LC3D{=Ik>``vm?xdZYD)Q^pRNWez7-c#(1 z4$42w-@)%87DUGIgVxU}UrCJI-&J{EC||SRV&mh~W;j-Edf&-`W96mOL)i}2>fl*1 zA^ovaKlt)<##$|3S5t0?Z?f9@AOifZR@+ST|Lr&L$r|97pg+@zo#R>5zipttqd&H>EOk&e;g-I8a@W+g-D>|_U7gp2S8_)a$9l2JyC{$Fwc$KiUQ=Z~t0nB=3zW{ca2yKjx4zV*D60a!Bwf+v?)LoD)Bwzy3WCQxK08U!bt@4Vb5N zAMZurFB8Y56WuLfL&~p~Lr8bpIl#gE2;KuB-mx;k3v5+>JU%^o?>PZ~pWGX|@5KS0 zvA$3~d0d#2JV`tx#&w$jFIl7L9l-dRe%u@(Ti@*$$8OH0c#U|rKRG}0Dc%2?8PFvQ z$LQeB6d_|Y-`1o3#|nV%Iu;zm4;mz7su zR7cOihw8aJAOC^pr7vg$?V)Y>c(s$UuFZh_k%tA^l769G`9k{VyZz2_?G=Ic<*%4+ zQ+{J@*|sz93T?_yY!0#;FA7|*uI8fq=`VE#WfY#Fu+Ou(&a;x6?N9et_9!37Ir30n zkYUiuhhTJCl7I%`xpn6VaA3r(sAN>9CHp?>N?}l7%@Keg|->_&K%K)Bj=e(8 z)fS*)smR%^XrND)o2W)@y)cM}m>RSFa@Y%kN0{3z4mVxO0XO9oWL#&g=Cv;qK5^-nn**?dei*Llm#7^ALefbON z$=)kMXU0E`qwpC%K4=Vu9tU{-(ZlZ};;IqiF2%TCtec+hdiFMc6O-X9kuRhzwhqV@ z?W3Kv&5l$0-9} z8Hb>3FpdjcW=@Q-ug9G~UT=T(!!_x&F36`L1<8_oH!Tvk`-f_)eSptJU9CcU&>V`FSUz5`XRp5Zli992ivFm zQ9tW*eLwbp`+n2p2jWFzkA8_SjYZp$GhJ&8=u7=f=fp?2M$A`E4Zj|Lt2`TdQ*sme zUV%-DQ|%WY!FPBT=eh3b0o@A+(_`^oP~Y7CmO`8ghwz5~rrKn9KUdI2hy9`r25-6L{l)lAYvqb*;+L%2oU6X6)A5 zg>(hBVe^H^{6gHtesB&yC0lZ|FWH>#2kt$q`}t1)Yh7?FPE~);;dwk)Z5m^@JV%!@ z56l;w2j)hhjlkS7Z|O^PF8Wq^kxpn#vwO5R`L8eWf^lhlj=a?Ij7P|-ksI+I7%OB!ZxFzq%J@9@8`yA`|8acDvrySge_X>Pg4s?HG)V;-W`NYf(avCagADP@-$DOC# za<;PF%e{?pbxP56BeuiuPnf8t3Df#3BIbD>K#_z@lzQS9coh<2CsrcXj>U& zu9H*b9QaOlkz@28I$z~&eR}jiwg6pk@ z_+V`9TNyZ~*L6F4z4Hp^J6@Y=(ZX8%@A{V8vXu|^JMK*O89S|Z7(bBQO12y>#=p3i zjfcxeE~Q)2JK56Ohn-chtqWYM9qA}^hw3Ng-0CLf)R9l=3v>l?3a6USwXMukbCkZ& zxoHlXYvz}6R9g$DRCk!Qw>sXdKXlytUUUxNug(M9R?ak*abES z2G}%goYCH~kM01@={4iF*N@!XIcNS?KKKkHN6d1CVE zU!n7{KMM1%wum{&7O3B;eia;wC!yTR7N9G#Lux1ENj5Bdmd%Sd##k~PjUIIL5RVQ^* z@A9YT@mzG>>aNvS>zf>7H_dn*wY&6LK^D?y%dhnhvODSgwZHfx z@hV-2?quB1rPvAdSL4d~!m%U2)(2%Q;@Rpf#wWc6@NRWo96HOrbU~ok81JKgY93U_ zF+a?can6k2?O&W(i+W8zcRyan(Ky>2t3Ij>pY2mtIlX+2(_y@}buy>vGSzvupR#5- zww&ATnt0bWj?eOi`c~I3caSNC=Tf%oT#v)?eK{Iu&GKJ)u-lOj*{B6>#D9*Li>vP| zV+7d%l_CA!t#ff>`I3BU-ISrMmTNomvg0k^b$GY;a$jX=)7m`kYwgwk)>rNC+Cj#e zvA8BXW8_o11^WV*`ZG`X^!Y64Bjyj?L@a<#YED%ztskA;X#RouSbutTPJCKlH66)3 zr%$cPUyVcQo%n<9iCb_CT@ByJ?EyRr(v{WAcjq?#SpIYj*E`p<@B`HTx8)N6&#n#VrP@T>(96jk z?WZm453BzSZ{kq)zxKto^?^E8{ggZd{W5Diuov{Be$~ffEc#t6$ha^*_*K{|{*0S) z>8P(7XZRCWlIyq+7a6M~%1JoLxHjL-HFJTUV=j;Sj5*``Qvz{QITQR;){9v;CU=d$ zpWnWDzK%JEpNr4SJd|hQJ%DoC__x4k``p{S#0lo0`B%QhMf5Z0@PF0!k4tcXbMczF zSU-Sg_ssTtM$hWFIjVg|x#By!xcU!y#3rW$6^oYo4YGFFEC^!*!luO3gV)IdO4BfDLBpw2sR9o0Nc*=fUh~sc3PIGS@_VuaHeE#%N+{u?h z#_3yJNoL~6`f}L!y~h3MuFl1+#swawgR=4Kdo$1Jq;%E#uf?~GC$=Djg$ ze7cWvoy*SQZ(s)*qwZzgx;I@HXN%Q1?itB%x^(@+o{=8ye(qcUgKL%VdUL>M_r+7@ z3!XF2a3ubw&)`lz1#=76sW06bN8y@sD$b?TnX`BZXYn7pKV4mU^cC~boW-}|4>-wI zndWu5RT;{&rQhN)d}6(^cQja&b4g$wvK*6@f%n$nJomwso&~3oA!5Vwq`hC(@3e{o zTR(?eaEJcWZ~BH@!mMl`<@<4ldR`p0)_us=%0B(4AFIdWRNu7`S*(1} zciK#!>0h!)+3uyj>PBX|PWkR(Yi#Hr^aOAx~Vzgvo<8d zt?7{ifd7?GHX1kPK$Q>DpR7oXxs`Q38&}xOAkRN;>bWU=St)ZQa5pzagc^|WWW+R$2_^`YvCP6+MH~LggM%O3M zC2$7r)F1R?ei%L+bum87CAxq)N*|z~;UIj)59HbC%5-V{Q4Ys>_}h8v%vXR<>C2vr z{H%P^Z`wv*lc{8Kb$32y_5+!wKjcZurR+K>`Q&mj$u-wj*C)&6LCW`Td+Qsy;AATK ztS{;F>dZ%_zvu~UB({os$|aRMsUJM6zA^W;i@9OG8vEMaJYu)GA0G<7rGIN@?~Xk= zkPGw1z&`DXJIsH4fPW4Q^piXr{Umo!4j<0*{u|fgR&~G;VD7k&cj73+HMm)xG~Hb} z>VdoP3A+|ws5_p+hpxq^&7-DQ8XNe;d?%l^m9gPD$rI(OHx9&edbnA>M7QyRW{YquBBhI^WzgKg>__(r4H4PvQt%NN>}Z#({pIOW-Gb z?0kABZly!&6EQb`#vb{_?qF-^Pr8?S)3wyO-0WIqnw#W=cJiBWek)FnuXZvf)PXHc z9(e}Ws~?$W{5l73=^y<;H(>wZMDv3Fqz|=QWA)^kYxOG`Ypxp?)f?Q8Zl!E}LB=h5 zZ+2G^4T z#)Z1;2W5F?y3oo%pV9ryM{~@arz5MQKEo+=KXcRh@(s;dJc*0wFZ!zOhg;1jdaUtp zj?tg+=V4Qjjr1^n5YLL6^s71Vx$zJVqD$#lW#I?o5~tDk=;_Yaeq^$C*4Ns-x{J1P z4H@Y^>S_M!Q*CIT;ZNgO|9Vz((e=hp^F2I2em3s(H(t{xd}nkZwyXAa%=4Rf#x@SY z1;&Ol__uIp?Kkt^yv5)AQR;5|d4Bz592#TnHeAgnB>%`kd|*x*&-GtvJN<_PU2FU~ zhiqqm>r*ffjbCkNezTeIJ%5V!@tiop^>V?CQ~aT>z`k@p7jxWpgEv^G{(_fRk8 zdluIiN9sl=)DQaH*t2yHGS>b1R>^C=0D7Q4(QbTU><0WxS7BQi+seZi){6lB%NQaT z%JcZvoH8fPZ90NJaVn!JiHAk1=cR7^lXOvT?IHXe`r7=oq)p`jWc_>jv}=Iz5h6CcTJmY+UPm_8q>l z#$%nVR@o|eP_RMZUhH1HMi$Up$ztn$jd5*{+q}~uQ)+VU)_IrhjRJZjxkkPi(;b7J z*BHaSbSb*1a;%&2UJiOD9a8)H!!bDB_;D|7fQQ-N*6>-+XT6I0d)J5lr4z6#^aXyz zakvGy(^H+}J#?!B>#gaxa?kY_ey9K7W#e3%`OMdWYw$a6_KscaF1`Q9+IPAdJyMy* z6kE+aG6r!5n^)iSt@1z6$F1c&C7>shZQ4XRWCM;crnIv*Ge`ZAJ?cnb(Y|Dr=kjOV zvZKsR+~hvSq`81oaW38@Gs$i8*Yg-V+Le4Vmeia5rF`?(yfVLVi)SPI$QL@hv2NU& zXXH11&=_GWSr12V!%OB0*`m&5zxwktk-Ov>8<#C*9h`M@?t$m@KYIvI;|u+#JkOz? z+K2Dax@$JHwQlrNb3>Ws8#WfspmWo6>6iF=MX)STUw%_-+|&i1(qqXs9HqTn>wdDk>H&b8Zm2Rrt&2xRkk8B?Dv$FAV9-9OIsJHXvA?g?PC8M=H zdsiLV{@w>dR^fHy6n~rN`cqx`iEyN6@GNABy73+3bas_7ZjRt^&q+VkPWah4qf@lL ze9vqtx;#Gvn+$KNuW`Vp5$CZk1CQZt$MBW$qprq;HHqv8vI76;S98ibXZ9+6UOlvd zamc^T2gBahzc|d;HU{vx`CfmcI`ezzU;2i=wDwTF)xlh#>zYT#RNKi|^sMwAW$Fvh zr!3CgNDb|5~+0r<-NH$QNJdphPk{cKDb7dRGA8Z+9OUZf6Wt}@uZ?gQkG{dmUw z_blq{xzq!1wC-e*J_qAjJ?KlGS-Wd%V-z1~L)W`j8-jARn{lZ=(* zkA9_G<3@X_GoHdX?!oTV?)psK$P;z0Z_M~M){GtF5#M-LbIbMgAmfIf<{Wj@r`lIP z*^mGAD}HB}v0dTx~w%A-rjxYIpNa+Zd<%fZWqI z>^ZWVe~QnFKc8=moTqb{wERkVorgOgWt`6p#HdS}`pf}(+ z>hAvPPUoaYXg9K*%yLilRd0RjdgbdcI*EIdr|Q6mLC@vS&~ECaEP6qG@8%}%(HD-< z<&{alrC)=-AoH|eWtTqFcdpZx>Z1PYr)@l|`C^>oVdKrY=DhODnZ7%&-rB&k(eJgF zHgb)5Yi=7~I7`{)z47At>9hPa>S3;^7cMcLjZJd_4;Z87x%-*VxXL|@2V3<}FLMrG z;2?9mc^uVS@r$wFdf)-qsT`^uw8k4v_)kK?lq6_I9X);;TPi(KO0}}L1r0?u4S7VANXGT;%@tlEn~wNRhIE! z40?8RR9Wg~Yri@fJ7l*pk5BO*`@uZ4#WU3f$aTIMbKP~FyN;Vf>g^up0J-8^ZLd6i zVZ3<;_3=E$J{jzp$wK9sx45}F7%s*|^iXYLJ_BCBgXTJUfLod)#CG+$x;>t!|KW-1 zDauuE_rx>uhj1hACa>HRkJ|6Jdni+T;28XbANc{u2JtbmVth+Z(LWu>g=83+S9mu2 zajEibor90nhdjbfbw2;Q`Yuh`v^r=z{5$4~kyoJ4@t8-L;#(v~7PXm(0;0$}^_S1!I1e zUpRWShB%d11dlV^|%WuTJz? zW0x+7Bk%>T!E?Bb?PN~k3-SVws~c38tSrEZxB}0RFKjXVtPW%Y_-tRV#YL{8*WoX| zI{qTuWJ^Yja2(G%7e@fzP;d2guIrpJ>A1W&26avk?hJzi)jq0!7$($7px$L`3ndqkH@>82fzS4-lltKzby4ri5uAjZ+yn2G`|NXnbx?0|M%}fAcGbSx z+Oy(q`Xu>M-E@>Il@+cT~Z=QZm?z3;yfJ1=CrjB zrJY~6;t$lPz{Do)vS?Hx~cJP@kj~Dqh z@N0EXd`wsM-PU(p$PV|6?oV#uejJ5!p)#Ruqb};|8NuiBpl7L!pl`aTXCViar{2oO z)7r{$=a1ZJjOagG{e>TmfsTjrpZ;_X{jKAnx}1J>9?sRb+8fA{%E69Df$91M?pb;2J3Bl~Zw=dC$JXgZRTaIHd3$Z{Y^lIe)yy zdH4zk;xT+jXT`hpRD8%UISc!tx-8vxM0H@iT{+_ZbWR+Jm#a_ujyp&0pVhP7n;cNL z)dAgE9o2yiz#b4Mw{0G~ch&1Q76{s+Eib(?QFHB{rZO7AvZ_Z zr(N}l^T|1StnJwL?ol{bKRTx}qg>B-1Z&b&JD%vP)rpPY&L8@UV;yUi)s;iWCn(Fl zmSap7)sF|xYPfzy=j-t`UO zd3FQ6nQZXgeO#;j^1JiZgB+q~lc7Mi(B)mz`l(N0pSpnS?RRhRnS8G;pnP@deMgyB zdEq|JuN))m$YZij-IdodM;2*I&tKieGqr#9zxL24=7#?1JTg9w7h{y}qQBdIhEN#80<=p1~UZX7??;LTSdn%KRpvU2W3j;hsmo^WKWpz>?b+J`< zbEmieeT-G+N$zj-CF04 zH`mzTb~2vmclCE_E9I)Qay`GYjREZqK5G;G13v4o){p$harmY(oNU3J_L+~`P+hdW zXRvKuNBzU~Fka_8*R=25r)82O_^3QqPOI&OcYC~Y2?yfg$|jtL+wde_b-p^_S5Ss$ zaWC*0oKtzKzTmraaQ7(FaBgLrdZ=f82EDJXvR&&K^f{DOIaC>^4GMgx9UTLo9W!pU zpZ3;%#)7f0|9sZ2cnY67f4q--_-?;>;Lp9lUpZLcI>**N_fuBqw6SMg*&4T<$IhGF zbb4R&({<|CacAqh{mS(`EwA-joEIukjhzbibC} za(j-ttEW28*4sIS^R%mHZ@U;*pl!61KmDN%wU4^g~ zZkyY7&d-j2eWD(HZqMHN>KgTMZ{t{bn+zzwsna5hExy0ecgLJd&&9Xu>mJ6I=QReD zg`3*W+68>aDf$I`wr`YGJ;(L>s(t5K%-fDJ-#w4x?Nj}(&x|o+gI-a+$7j%2w#Jw7 zr$5Cd=o{oJkm+P98E-tg7dh?T#*1+S%Imhz=eg9U?OpC^JOzQ(ZgaF~0+c>Or$eICy(pBuZan>nT~%2&4b zR;IQq*3_2vX>Xk6uk+oO?oeLwEaUt!Ki35H=zLedvHjcb`a++m`&e&n(t6*})Bo=$ z&(g6|*{QDnX6viXw23lWm&!a_bt`<=*XoFaj160Lbx!N)p88Kewl3s;YuY}LM<9+Q<}day4z@gyk17r( zKWl{CUb$j&a^<*MXCzlu?y5Cf)^`>5Ip!Q|ujIvA4{GbXb-!|ctw)pdYVCqG?}gv> zwr^$7yx5*&{i-!UeJ_@^Lw*ATe3rZHn0#jIujCS2Bjn!h3qIS|{nj!oM}5@G8VdDP zXSr&9?(rGq&N*j1uldj7hjJ?9Vi(P`(9VweF4sf*izn(2{i1)G_oWZXmC7`IOn#9W z#)`3HOc+n(r13}psm^3u{ml3T-^p0A*4QKZuY!oa+xKQv~~_anzi zyr%h5{6PFajRlGI$k&&fE7l{2c2yvM-}(@F^m0+<@)y?Efi*MMPswTZ?g{y?a!ReC zvu;Ulsr5Adtu61F0QtGrKzZMicP)8ulN?#^t*}n1zcVR!7OWvEeDCj1S`%eWm3M%jV(!hUDDcYhPpd!!2G+HY-?^W4j|$^)O> zv%js_a_m<(zu)LL8vEO4-qYdt8o}rBch9^xrN5Esx;0^G)@N-WxPO0p&6*5@}(NfW2}>&Vyu%<)lbM`^8^2r`((O5^U2(zyRnVg z6Zo6HX)g1li`UDy!VNeSKYGW6^?XRtJ82^4uxDGWV6q_p|5Do*Z-BXXg~o>G|i) zEyDL+bJpBhb6*PlhNj=N{P#eap9wx4sKdvC4+ZZFP73^1u-}tc{}sVYf)@tQ4)zZA z2%Zq^6g)h*e{ip0v*3=wt%4f|e%tNu!JmWQ1=`rVw5%VtHrD!7>wv9w^)31Ko>o?BtGR0=sZ#H@7()hdQ{T+1gVsf5y3g3IK-`xf0`^|L6HV*tQuHV%4 z+uVMm`^LeN;9m>>n*4tE6$@8Pe#`sPg-aKH6I>YlBKT?Wqu>X@_owjLKF6H%tKg!* zb^jgwF1RfCV{m!!*Wk+Fs^FRdS1(+>@b};^!5;$Gy4R0`?*`upz80Jpd?i>NoEdyE z_&!Q>HHMQ)LyI3L$r>q=hxO-yTEaJ#WgAHe7I2zev8y!kBV z2|Z+WV7-p`lQ~3R!Q=E8Iy`HKoj_N@*P8@%XS$bsE;<{%j;?4;oc9P>Uymnn?bgAr z!E*xK^X>pw;yJ(5`-fm|es2Eu!99be!EV7m!Ha^I2P=cu1g{U?5WF#XQ}EW{sNkK! z@xiL#jNs$J=YlT;Uk=Wk;;i{aoIQW`{JFvT!MB4S2EPb?9b6LpF1R%KZE!*Gli<6- z*8=zZbnv0z)IhoK2-NFUfjX=ElY&PC_YF1ROllf=JYPJ;HiayO=<9{O8>CyGaRBtBx>CJ3I z_OG$Uj^*p1JIF~Amo*NJOSxFqK3FG4C-FWzW15aim+`JVx|2CT*Yd8k`v-no8+ZC$ zalbciZp{bREnGKYCp?9(06+a>fMe$8p936+FU{MFgI@=Du<*V6%cpqrI{~i2D_;)& zBft^xxhZhhrvp5O*G>yg2;LFAAvipEX>f4xyx=*(e!=qK8Nt(nCkML*j|#R6wh6Wl z?j77CxO-sh`+b841P=}#9&8^xGT0&5af+Q5c3OCJ@Yvw-!EV7G!Ct{LgT1Fv{+_{8 zgC_>+;5zql&qoBdzPsmRf?Wdlb?)PWCj{=ld+_vNpWwN{A;GHyZSc>v*ucP^QyFZFrl9NA?o6uuiDoWFAZ>cGUl@me=t>lVRng7t!12R94UV_|+_{_nvb zgYx*#=6^Q-qu_hNHv)3tGr{`;GUV04a{}D`(BN*t`oWC?zeVqN>Msbs7tqIX_Nw5h z;8nr%f@cKPrI+S6;~yGq6VR!*4DJ!IJ?<3HtJx)fKl$>&`)IwR=BGjRY44}4KJ8sS z>>cl)!P(x!!|zian*W7AYRt**nv>=~^K9_;rN5jNUsEZ?mL|h z@CB}+>*0{@{dZot^TK8US#Y0#e0X5sUJnVj4IUOO4e&Lt#?5#i=RYfW zen1W#7Q8GVmkQrs5#a3C1aAt+Fwcd%J>%)Y`vN+1w`VMzvGCzQTabz5B;8pXYx~+D zKUw(6gp+^1@bifdP3B$_;9s)Z-`dx&J<*xhyMDci4Rpur?>OmeTVNM?FYfw*t?$Od zor4VnAefKuD+}C~Fn+(AD_X=>mF-upW zv#=YDbv)0m=D0l^c47X)t!J{Wv8_*HO4Ft^s+gp=_seumoy z<=tDYb*l;g`o4bPm~(u0?4|)e$J_r3u9@Px`RnHA*P5Rkcg;ot`LK1cZSaU-r(l=h zvB9oW`26T#$6#sT-1`OO%O-)c*A0~O`+yFAZt$_--2t28z~ITjcELRY?XY%0*3r?~ zCtnZvay}899(<%ebn9FANvvb!3%53iO=gXY^+dM5+izXD zbLsKcck`u-W6#OW4A_eD7ro<;eva4ixv}is zT67t{$ATXj=*ImXcnr5W&$Vw4 zUK_kTctNmlfY0f@^kUqL!|^wsUN=B0e+w=PE)ISf{51GU@Z%|b|9SAM;Mc)#gG&R~ z{3*C1xGMNp5VdFFrU98iM%*R1S3rK?eSA-@upQ{=Yr@V8J5S_6?E&(L&i<4cutUIp zHi2hlBY1A|jO@@BWZdfl^6seMU4gbgHDCjgjdXG{;D-S_i>}R1!^3me&rSSVBj@7x za_y*Z`;31ba}M4o^Kmub$}NI!eZF~s^Nk@s8D)_zxZWRM8ad+}at6mMgRQxBU@Y?a z^Yt5_#$QKlCL2N_x9_#V^CpUfunY#QDuZF1oYLGE6B8rjEB7uSwmiH3%z$r}FJ`^D_2%vyJT~Cler3QXd}=_C<%|ASz%Ox0 z@Yf(=ulZXB{}yZ!@MSzWczCdVz`p@}HV+Hv%(jm080->|DNhad44x6}9q^Cr8|)t( z5b&A2G~namOF1R@M8I!DPp^OG!ubp5FA8kQrC$a=59sx55W4;MgYN{cJ1=mL&jue3 zP7dh#{7C#t&kvRdWWdgW`fn9%9&8ZYI#@R_DCqx}2bTuF3Vs-TBVY&czw@{A-5edf zHXsw8AMme>DeM?LD4_rI(cu4d=iHS`u3U0y;QgE53-CYRjCWnuZ+_yE6DPkpLEk6W zUlkk{@VoJwFAMnL#7xKrv1;qg`RC{Z)?6AV=| zCY{5WJUjSOK)>gwp_7Ocd9ToiXJA~5V|w3^an29PAJ4DDUVV2!w>dg^d%%``OJJU` zZTtIM-oYb&X)e)^*v90Hc}6bqf%1il|FIp!{meffRF)A7jx@&T8T8}wiP3q1Fh;Ee(9J2Jq7uL@on;KdR2 zY2WG8;2c}~s&l_&;U$yLJQ{2{^qmDYY-ok=fEc~pM`x7TLr#@W590T(p$y+#Qs(fFa(|dz=1#b>k1TPNALUNMaykD?UaP#1rxoaje=f`tD zp8Hzx`QU@WiGdgaIm1_{?Y0l@8Hl@Fx8%A>jQ0CWzCY0eyzf{aiMt*e>>rR9{6hME zBUw#-qnEK((S98}qHn+tk`Q7>2 z#Sn2qxmK*~X8|3sF+7~YuU?G)-THa$a~}V1{WQMY?>_Y@@bTg$u}!=VKHH~UJdeBh zfj~XL_26FWp&ULS{^QXHgd_Q>$pqZVKA_*?Qg#Af#jkWB0sjj> zJ0JP1&Rt!bzU?#p`$+-c4Iewd(PILBd7#VF@#*{gJW$==ap$@QTwnjZdsm-dwyqA^l&8RLX(rf>7# z@ul#akMU0&FYZp~CjZ%#dj`t_HY1*Yalqe5k1$?wJiX(60bZxWoE6At8RMRpCad;O z0e=2_aCN}<`Okn49n2Gdc$@wt?$sP@K2*78czTlo=9pZ0JnkHMX!2lOH=B#*ejuji zm~)&gO6c+ zKL~#U-}8&6@6MO2yWP zIA8f}B=!eC)HVSdguS(KAm(z5fbh9`?&?Xroc-~g;LPA70e>Ew<3)is+98nFr2TFZ zT)pJ#iR=}Zm!HGF@Y_JA1V;ztxE!l}1O6cWN=Gm@|KhBzD1znKOm5xYOh`G@l`FP|~{647M z5VynSxMbvSd|o}2U&uA|PkMgizsjJ0+R{bsSDvzzBNm98ltEUIA(c;+1!4k#+sFkr zg7|=#<5__?C7qe?h&t*?JSgxdpA>!Ed1MCgNx7yr2CjX5urk2GxcEf@ ze#X=InmieKn_sHD&E_C`*bnTJ+754uFQ4t;{=k-DC#aY5`MYpCImafz?P4X`L!0As zIzQRowyyu2!d8C@8;pO3PsJGEQ=!MR+xYqD@cb$CcRGeK#5dmjMf&@Z0b9{;&9EQY zlE(%o2CD*G&n~6E)7#mus{=Ya-k;Uk={)rIA5Zfc>FsnO>n(m4kPCkc$O7Ci*8!j6 zUG{W&ANS*XTX_=pJC39Avukj2b0%i_-Z9+mJQ!sJJ3(#*`9l_&pXAh)fx3auYzxPn z~KSF&ScTJz!^LbYG2pQr%1)~gMSCARrnc$rkd=dN-{Ly^T z`vmgj<;LS8`RH=g#W|bbdG?aCmz*DnQ{%A<0`b!3c&01nZW`P^kOyqTu%dAK10QI`iwyuNUCsdk1`A^zm0o@w-4wDJdzHsEQpK^U&qhoW${KiI(!#=7;*{esJM(z zhaNAEOP8m=*56;9e$?l2AztSH$IIn%d?_Ak3%K1ee2?4lD1Xt&@9rf&0sKj80vm$- zD8yA8TV-SPyYpP{o@9q|l&9P=<}t=T93ykUxt^=_Yu#EW_itTVzIGwsv=iAt-f2s+ zj|?OW*-K<2ImuQdGuTyRYHb0sn%vdDWHI^7AH+w`XHRC-|0B*XuZ9nZpHB|JY_0%Z zgT25$ZO-KA7h*4vE%M9whQwa^>?>dBF?^Co1^ku#L}W{Ih!cb;op|MS2--J9*CEOwOj2G%o>lb;OKNuDZyoOp+L z_&Wo(*lPlDr9%RCn)tWa$D@L612V;$0y3Iia_y39CwWg7FS&Tqu3w+AP5jO_xx&{- zKUfy9PmGZV1#}GKnmueDu*uD*%L4O`@6}usuRb&2yZvY&4q*<{75%o-%0PU9>^Wc> z!zwO!aR z3 ziOVY|D=WnZ3Vd(BbDZls_i#T@ma@rSbyH7r*)x#Up4IcSwX_i%PusD_$ow_g5oCfG zFS#HWN56;-uqng`#D4fB$p-zneL%M$3#`RzPR?j!kO6#0bduKw>u8$NG`*ia)HQAVNbe2!)9^d{KwKN=%YU-Ij6AaDjQv0^$RlGvkrS{wu(i)I z=Q!WB;x``-&IsJ!`Z??9`0(BlyfsjtBZF55FAEM1o*V2FJS}))uv74`;C_K;-Y}5g zux_Azt_awMY!&%$?8Y+#a$S3$6wm?4mRAPWJ3K3(U$9B(LdFC=$rv&Q%{B9ojA2`w zyB7!MH{UV6&;0*6H%$DnZy=V9L{NA^DJI&ouM)!Fp&*m8`1K0;* z{p5gr9WexQ+gd8}K>Ly9TLs!wo~HH{W6;*}dbE4j{OSvOfVISOIrURx4{TL^xL2S* zmk07Y#hc{-$yt!QPrnc^tN%(~1Nm}fK*ks&^bxTNV}^brWQKgb}u z3?G;A#+H_cVf=k+noIP>#Ho$Ta|3#i_ac5Ppcfgp^7+^*bR}bXj8*e{ky*w$xdp!s z*fV69I0-q%wxM&)%Co=5zA;y7*ZgDp9tu7w_6_i(k#*Mwop)63Y^3{pWrH<-K2CFxD%uZvkkconby7F=^+yNr%xpV;Kr)}4BuASMfTzegIx+di-$sU!TV$3vK3huu zC>x5;&3XZL62A=| zPw;?1KHRcE`A-eV7sNt!Y+jq= z{A+BU#%S;Z-r&E$FRuycw_>FH9^?_u!e8`ZdNKW&-(^yPD}1-#Ib>G3xc9Jc_CC(Fuji9@$`vQT-|B(G$-|Myak%=5Bd9xlpF9;)=(`}u zSUfI<>^Z$(!87vRPz%k$(rKN7Cjz90sM_sM^`NBm3L zUwjPc1}g^mZa*2}T-UHs-NU_lUt8rUPr2lbdZ?4S$)_fV)D_0M`|g-?oJ&@X&(i0q z-0Cx^!z?_P`W5O6-QRk*uI{PM>hGBAw1?+rj~kzriEqf-eEv5!n7CqPqp?gT%B7W; zLKYh5>@vA3;y&zsHk$YkJ53CTttJj+ew-DMk07^^56|2oFU5#PnfdLkQN^ zm{H?J<{_Dh_sjQ{6O|Q}ndAxEkIcdOm4W!49IJe*3}pYA|Mly!do~X6!iPMa)5V^_E#!gxP``y_Jph?d{~B9nS-_`8KFBd;&yW#(X|E1&mpr1{H2nXK^H)}! zJ&o7+?Q*}JF2*lr!*7Ns*)x1+xO1%}T-_|dsdozK^W+8}+Fb+k1NYuLxNks?JS4!) zd`ofz#4YePf75OOnZoXY-2-uPvSr_ZZwmJx6p#gE!b=0Pgq+}uV}BeTtO(diY!Wf{ z*G<9KWxwOjcb$8De*tHS*DOdT#QCAMuvKI)Oj( zm5kwY)P{VJKMKeh_RTqgxIQ~l{+)aavir1vy(9OFT-RshJlj+rAAhO1|6T!mmY%?O z%U{a|GKNMO#3tu^Fed3$^d>sYD1-Pu=$LFG{t)v@zJR$$FEuaetmF~sI1F$Y&cZF_D*S>I9T!KID}?jOf~7%a16e?yXNQpi;In;X zg>$>kT$};Ks@V!;47tK4V8iuYJ-(Blt|gD0PhM4Sky&J_dsJ4jG03Xgf6ZZLqqvsb ztL7w36yVE<~9KA-mGm(`Dj@2&@*M>)@}PLWsOd+j*qbWHOd{{LVO z7~f|Pd@~Sx;SVD}aAD&vw@xlGAK2XjJ|VG|M+f8q&c*xuN4OvGu~@hL;5*+K znQ&Nu_wl@#hBY!`GW=6V1$T4 z>pFP|d~D!5-xyz&d^#D za+l2}md{5f_Ts%D)*h18Y!>fs`g}l+Sc7tEU=0f2rQ8hf38Vwi2gs1;1@r~F16_h| zR6LLj;Zvo9@TKv=vQv!-a)cc#mxOQI*zsp9HJ^n&Oh2M0(V6%kE)4i6_}1xP-w2FZ zW0#yUj_GyA^nV2GALE@K$UkCU&=u*7d?~=6U!RKk1HL;}xkE0|LCtUe8air$CxHHr zH}D6Y7O&8!@edwiTRnXWvdH(n19pbv;5#_(-0?ie`3&jsiNoHuehIkkJr|g zaXnkcx!_**wM;p*IJs*yaVeg~s|5}wXK;DtD6S^sthXTp$R?<)Bm>AVGGX@pRb8vi zR|#Vbhuz432R`%Xb2==`(KHKMSHm^-wu+QT8$Wn1BvKRO!!LxWK^4K$yyX121H383Qh7?}(Ql(nC4Z4zgeMDb0uh~BQappg{L1wUL$O$q6kMPOk75su{#8}uP;`;Pi@f3c2 z@p>FaMu-8>o5e^f8`utfIs8xJB5VTwA9_Dd7mE<{5bF@{ZhS*dG5!|wuoc_*THrk( zzYN6E{}TK?uuf2}mCy3W{xrp(=l(qRx8PsFjplDOSyy?7tT*2{*gUvfU@K>9gFxQa zt%F+xHxF#B#k8;cZ<2gwf7Xa7M=W3dnj8W3J3A1AQE$&9Kl)<<*|RDjyF51=hTLPb zkTYasZ56VHeZ|h&KG-(6e;~I(%!!;Kf5lz|zyA@CG2{!`!cHN#^^2VEu5Hv`CkJE- ze;AvD?B^3ZG+=YGJ>_(WtvxMZuU4+GX~`71hGLE4eQaT|KfW@tK|a#aR-v=-=d#O< zLHW$aCLcJPh2FznAw$G*YO5IAeC^^n#=0#TL5_^Fg1tgtqo?t)(HrT4)e&v`*_`6< zpmWkkaXIeB`D_ooi*Io)8>D(YJHz}P<3fE$iY*Qnx1dAQo6U2mO=0W1`vCg{Z~?oa zHiYuP_hQ^fSI7UZb)S}@O!X^#FYIHhs5{#PAK@kSZfvV^V&pkoSpOh?^SSVS1UmZyghQa&yL}E@hCDye2h#X2W)*OXB-pDY8=aFu`{u> z#@WVmo#&onT8(F^OJ$?@7TBl0>f#wIL)IiOJ)iozu500I)2W~OfX^*o*~)OO_T;A} zzsZyOSB*1$sbBTK@xYhLZYE3U7REL|ApMBF#2zw7YOk0}qwkKM#x4=JqSNtZvqi+F zE(+*@<}e$C?ZH>ZFJ`Wr|7;98Cm$Dox;(K11Ds;rG@fB!i2w3g@ynAN-wMbLJ}I$v zKB)Ty{7jDw=-PZt{7YgLIFf&dzlR@)eIOsoIuv@oI62OR#>>eOv2iv5j>gp&2IL3+ zCPS_WYzyDX4RS>;GQM9oAP;U6tRLKN3NmB8;8wv+19E0zZeen5*RIR;lIJDI>%2gH z$Op249U|{c4uHI1vSDSwul0g}FN+_mHVC=E{-|7NJbt6B>lBaYyJ~EN-|EtU?@HVV zw6_?&HHzNdD^|jP$lj3qC13B=0a-6bPww+cT4$tR*^P7qGC`gOyOQ6NTrdts@Q>0f z$OmyV@qF<_@innFek{5O_}myJzpSw)7AR)N7H+J?xa8+DMy(~F^Y9t-A=813VRFQH zW`od=jBoz0f^F_I+oW=ajN$j9TbV2LEcQoz4*U%K1Pul_4Cg#5ta z<|nSk$K(SJ2J^VU$L2R4rhDR6oPsyYleWHBXLTG0;TZga<8T|jn0|asKvxEz3*Y5* zS=)1YAWshD{mS>UM%+2_>HJkA_u0O29jUcCeNUgQ@4c6M7xu{~lQ-TxybF_qcTphU zOn#m`J=eQN-PEIy(`TRURRQ|}-+N}yjx%xcqNG`rKV(Sx7q3F?1INnW+7T+dv@?#! z{m$>Tc-;LQ-zacR<7(^zx5v=`_d0 zbj&w?X8N0XXl{zj(Dld+b_Tz-_z52bKEM&;tYiib**U1(z&B*ZTLb%_HxaobfIh&qm(-Aoy+|ub9jsBjtbb zr-_}2^OG%NH}!3?Irz577jl^W#pUTc!&O6 z7CbqSU%oV`e`?FT4MCiPpOU{)Y>f|7E)z8A;KbGGi|e${{J#?YKHpPK_#ebo2r z)qF5GpFF_3@F@PpxqK+#^C-{iH&UP0d#r2sk2-yG z<{YC}dvNKHzMM7LIrg!MM&A&=FZ8ox&T%jLyZhRb6TQ~?{BxBD z__?45kPpTb89~n{CyX;Tq%lXIF#c*E+WHQoZCIHy+KA+fdlh7Y`?s92EZ4J7-Pf3> z)37o4DcQ^91>J{Fl+0jru&eoD`D)EMah3Y1=wRk2|1f#MZ%kf@SBmZPEz|MX8gxH- zWa9383S~{w`J?ix_`~J@vNh!VvNQP6#b(b5*cx(#_(%Q{@SE_V z$So2J_AZqArub#(;BN}Z5$i2h1>y+wb}QNvB7qMI;aa7$H%o+Kn7kFuzS24{x^ZODST~wT6}VRSf>SK>gc}`zY)h- z7Vuw*|FGNm`)2)CcbL1wB!7#KuBcQjes5UiJ5n!%Xg?PB2L9G#RlP*+%!x$3VU|WzQYzeu7aw{(geir;N_-??SAW!Hs#wlHg|A`Ed6KXv3 zE3v5?ZxFX+3y5h}p3tH83Cs`kM7)(>pFANy%vbZud^X=kJr5@o)#1kYFxi2-=%Tm{ zFB*Kae)3W0FLz&&RpIpOO9FcLF9K^mz8`!iu&(9n!PkQGf^&ki0&7S9Bl!1W4a#S- z-sIE4CxVX#9}M0foElia`0n87z`Itw7wffw_hcP0VC7^5{mNm_4M(kuUlW{Rzk9hS zSU(Rf!|@ftD}$E?FAfe44w}Me`wk1N-+y)Bn%4(!4&D~1yEUWMY@Rq>mw8fFWv&XW zzkFYCM&P-vz0?NULYru(F9xfFbA$7PZwB8Deh~a5(Dv+~i--4gv7gvU>>qtq8>qGz z!qVU5Vr3c`OP;ZN$g;{CIt3ggUrKxyHzALk z2J)li2JyRF_aGjUvBj^!WXGZK67?kxH| zz5Xcyp586+{c%$~Depz6|35omYrG*iJ~$)zMDUp@`rZ@v#c2WCf}LSa`&Xw9J*2ufhse$*ic()fhvuCh#@Q{E^xleHKpxe98-+f{? z@t5(3S^xO`KpaQRojqpFrI_l>(FH4^v96URqBNz{bxYknwHVM5#JVV@qUrMY(?Ckpi*#Ke{Kv$u&&|&zN z_)?AM!vl76W6yM>#+KO?;<d!`FE2~kCszH#~*|ct!Fb#I=qHnj6tAJ-qq6@+jm` z@#nCK=KYS8=d z7(6=IKHwkoE-m*5pC3Gh@5)vWv6CAI1xi3fOw$;p{(tD)t~dLk#MSU{z3ml{mODKzCRg&?hzv*r5DU z^bNUFVi?AoIEArjj2gejxp_dJq1VWFGQZ?BiM@3lh(_?)&io598&NgX+xoePsHM5*+>EnfJJp3Eqth?+D0+Q3t1o zlV9Y3bDeWkKz7*rPPRGjT(SbZQ-w@(%sI|~>lDfw%Wga!xv(WJ=&Q;>nA`fCBOJs z*~R>=^rPmBkFu=mGDlfPwy_Chp5%&{iJ9U&PcH+Pgl?2`2YYXs}J5+fDo z*dZWu#@nG+;M4k56ic!13BaVN9cVL^}0^5U%|eG zIfFmn3wSzMcYiS46Uv3-eG~j$bkX*~dqezO*3uR8n*glTt##3SsSv%KX*;4jc)wjiV)7SNRwjDp0@$kc7FO@ZKW6PMchotY2 zInr6+=j}dQ^BjHv_A8mdwxzG?v#zaM-NygG?@XVum93eW53!&5$w3y?2jA;t7T;Ib z6|9GV{Y@Xv>^hU}9D*=N6vbJyJCTd`)~`K}#w%|dwj=JSX1 zKHt1IjP)P7TX|p0@6SI2b3^rF&Y+HFmbv7vdQ_+G)sNr0wmu*X-wnCkcQeT6;#zTy z?P0ykx)=YBc?Y!VFumS;-REr2Gx$2);~VN4n)!@=fu`?Vnx*&nM!V-uKT-}j?|OC1 z(`Ur#IqMT_CE?jm>PJ5t#x?Z5bNy635ATiZe9d+D)pMU!zn<3y(yEWF`*`U(I!l7X^N>s{%dLH4^?XdtZ!ay2w0;?y(2|EVzb??B}b4c(G2>s;yR(xuX z-ubW=ME&FoK%jnIS z*(L9;crV0zj@~Jx2lxf)U%Fs{pf2VO{8aWss~>emXM2xEU8>W+gzJO3$rVAJ+wb5V z9P<>uENklD58eY|BhV4_1HYCw<;${H*Bbl#2Ya*D&3hpBSeW}R8s-gjgn5fKk(Yz{ z3!kcf>>+8t?|aEzFJT9= z33m$Q4tZdFlL0WUE2sSKy1VgbOd6lYYS&AL*G%05x#-A#ZCLWPbpMs3Yh*C_+;=Ig z)7rx`V@J>bzVA+scOOpIX&hZEeh=@zA!FpFGDR6sU0GS~H@Q_=Rhi=m*WK^)m2nM6 zHj$Yi6QS;g>$Gye{wDQL)|27ndZ}#JMqtnLX_9lb>HAI$xmWsLyY&FFur@@MbxTHiCV2Jd}B^O&=OwH9+3GL3w@ELeA86Z50>JC9_d z`HgiL{xr6lLy3YplC^$Dyum&UyJy!f4zAhf|iB}rZw4a$Cp%dv6 zI?{e;`P?|H7v%pP!MtvnV4gR;{-G}H_vLeZCYX0j4)!{zQ+s@`9ePjQUSDlN&szth z|JecdO_+atK4{b3gLbyoPN!@d_{8)JdoirHEf)A%XCLg>qEqxm{nGxwyMr}6eVV`5 zxG?`Pj)vzP#xA|VS8aYz9++p47i0(-L++4IY)keZc}B+7o+A&*P;!<$C(FrT>z`ye z-+cGz)K+JocYeX<=L<2<06V{XgZv)$l?aQO=PR}MrDI+*y#G^PhHZcPrn-x6ntAWd zu#Ks#WHxzSJ6&Cn&+4VVmHJisPBq;-EL-_g>R+Lw&GUw3tUV57to1B7*Dho*U(2v8 zCWHH}EOK^c`8q6P`9R57b^<@lk3zp0;^;TI>mJBq2#@~mHL_ZKWOse5mF@r6QU0uX z$SdfF;bw%x@VZFh&r#HGr!p^scwz`7jT-QpfBkYBtCyZ5og4fa&_Fd5x<`WAA+nqsa%bm22Jw%X zv+$Aa9}Ws%4(2b`XZTONkH|LmehFW`xeWgbyV+cZyyJ7Bf6fnNAU)(=Yj!q$WL@k<{IF9_COz+Uzn5837#wB6d&CwE$PYw zz3RPmd8Kb39=s=_49eo$G5lA`=zB8CO;@WE^}{E5d6;?5qMp09qAj!wzvj__p5f~< zhoNh%x7pXVYw$fTa~OSOqrmz~Ye!qPp zZfOqJk$Qu?$s;{2@5;gkqLb-eeGmVa4%SEY)z{Kp)mg*y8hT2fH-Djn=pW;O-r;L9 zM(nHan!K^oHTljvs#koQ#CWqW%ic;pgw8?e6$??FfACA2gBY*oA^g|YQ|AxnBgV76 z|9sms>l$_@?C&gO0rNY z^v)iE{~hKzWE5S*HYT&odAuJ-FFE>M{N6Y4i4z z9-U{+>Hot0#^ITUr!Rfe#L=_DH2$XHntQ^T1JbpSN4iCR=@;b~)-lRRzbLbLrFu{| z^uDxi1A`>L`OXk=qWZFd(Jpx-+8x%y;pr@&csGER_QHk zl;%);a&%bNabC;bG|q)_9A?-K6<2*1^;-vy( zz&TuZe|TRVn-=cD`EZ!tYkYj*>WxgA8TaV^rfL3(K0AC}yz)WLDT^|xdomFGGTK4e z`E0a7eWBzio1aWY{Y9MvxbNIO^BeOTbB5u$gS}SfJz#^7+2$X$LCAJ<6LyJveby1K z^}6S!(KtJ&@*BhRD(Ayt*+89-)#qi7uFoj1KG%7Qy0Cvs-L#H+PH(73J}Y&u9kh@6 zN!v-g!Ca+l>c)gMb>qX)Z|m&lFYuebsjji_%ed-X##l4v*nxD9BaB6Qi1BNDcCC~y z0)Kev8mjSZTpQzj8^${w}5y3JUVe0YXatMa4w9ne1DF@+OJ?fM|W0l zb{^+9eQKTu=WKQFL+rI?v`H5qfE}F?-}xq+!Ic` zu2-IkyK_`Js6Kw@(jhDB%W)(xU9TRSp-Jx;{oUuQ!>W`09>#U+3v^u98KkRB;;zp0 zyE>6xr_=p*edcsscV9Y=uci0p*|YLlT$3hUt&Wrt)f39A4qA8GNk7sr^a)40!u7VD zwh%^tF&;Wbj04xSk7pe9CHIXx^4^|KN8{2M?AS9F`4>8W5oQ?knaY3T-MFU*PM_8~y9O(o7kG%HhT>o>u_x*OSvb^%19?)Nf zrEi3%Q(QM@$R_f~kqjm0J4creJ>wdgOuh zKd76As}3PYT&s>%m*SI#H07=G#&6|@xZP84)i2c#%I2At_qFs?bxZAPd(XuqU#(N; zu92se?ZPXc_~l2vir@3(apgD}E^PBuouz)nDLna1UpP)4>_rlvIuV}SmM)x=(`~D^ zi}urJ+@}YmDSmlVm-?~3uDv`jJ$ZAKhIZD+jU^aEWV$@ayRxe*58_ScU-{K)V$O=cljYs$4PaNb7c||@L`{Fhhg^`v%MLx(AnM2l- zgJc!mq2DQ^XXKZDA+IX$CPx0yA9RPgh%nNYN3w@r5>MwZ(t+RdtiIa6$qV&TJw@gz zgD~QCUtI1>m&_EuKY26;w3jfR6~P4B!B>7a{%71mP z4BAFmd2^IM*Q6zGamhc#;huDbl^4JDnT8QwI5;J})Jb9KZQ4VRVBXuA@Q2Eqqy9YN9`?6?b*I){OL>T2g;|M;;pTtPbq_TB}4Ud&xuc& z$v*9-?BdlohWmm3#+GDPic>q1|NKgPNW$tX=Jfmo{0!<-{hGh_obY?~L!x8ozPD&b78+eVYu^{?=QB<4^O9c3=Z5gF4Y3`tGE_?`Gbnoz<)Su+_D< zalrp3Z^FwHyOn*dt+bo?JHJz>$|5~=Bs>`<9{N{1dPYB@kJOj`s2t+)yfI;ZTp-6L2Yo{SQFnB#aMD$;l^gOT@A9P%wV!bMo_NS9d6zF^S|4{`IQ>|^ z7Qg=d2u8=ik!=xZ?bu8b~ly`kuU+0S!Cq1Ve z(ojcaqCTUK$v@wPGO&Bqzq+!|jZe~8C12g+_g7!)!gx@AVdYEu%H&V})UkeO%t%vv zYA5yYc{ZuKBD0ON&V`K+bw*A})0j6-^fB#YUagNg_gw8N{i^y`IpvL9()Z<2yNOqy zGH&d#u?I)r)=$VX>FGE6z458;)jQc{E!1=Jp!b3f%-7! zr2e&K<$!T6U1NhT;OkYd%0jMLJFx~Kf65|^y+himeMp1Wn>$xQXc z_e6K7V|_qA$uW5$L&#;i!Wx%xrcH%W7yK*wgmI-$kpq=c+FTjP5aIP}WhF~G-tq(?8%}NwYl=z=dAC@8=b06JTGlyO8&_=_vwGv zwWoPz`@TM`F7-cQ zWaAsw18fy_V|OffaLe8ioH#>%a~>NCcGqkf>A>W`k5J~=MGopZ~-G31YqA?wIi_3WPUtRKj$amsE~ z_xiUq$YX1dWVgIoH?jX=m!K^AhrVsiP#>1AIOWHE@`mlG|B=P|v@*++>tvxe)A!Yr zJW5wsD6_m6PwJd}R!`D2hRHnnR_A1=cwIN9>?M{bVYHRJlWF>-{hs!8k?EdM&)P@Y z;#NQEolLXugH1`6NncwCtIf4j$E$n#hH*kp&|hdv=MP{+!q&h$lNR-B$QHjHcSFj0`Xa$Z2bJ>;-)S>W40&$BZX&XlMP-en9&@?DNpi z$OU_g?TccY)AeM&JskGw&^6?~HX%>chdLzlm0cStzj4BzP$oJ-{oC(jPmaAh+DpH* z)@_a5c+{`#+wtC&J@U%0eYGRqLf81yKJ{N4gZhJV>9_ifwy@X49!Gmw>0|k@$HHC> z{gqr-clHL^gD+h=P7~@cO0rmD4!(N8T!nu<}aQYIl9Z*ff^3yMApS zt+v%pEx$dE<}S*ppO963KV4X5Q4V^*bH)^#i=L=VQU~&=?d4OO=_BGYPK|x>lN)Ra zbwEETi#=8LM%gQ6zm<8F{;5w%M|nI;&TCusWY4I18u?AeE2p~axDk#XrTdi49#7+3 z-3q54>D%&Ly+;PHX~{s(=@Z(341>BQyU0=bQtxCK`DET_|AxKsbS~X1&-8(Dr%u$3 zK0+Vb*JppAy@2Z9-tytS7}BRt$wcL^eJgMFqT2uO9SD0n?SEvyR7bFp{kDI^yH)ld z_WxnC=TLb2EbYUz@3Zf)8gpb9eIae*!v0Y2CVQX3o>TGkeE{L@cNB-~_H^3o>9@V7 z_Mq+@q~m=9>Dp^)ucSPY*UCvIsz2!o({H1ApV+%;_GNmPLAu^uP;cG?@xGiqNyENe zbqf0^TShWp-5Td?7jtjAOTO+WG-DmChQpO)9d?T!dBnVBV@WU<5@BrWRZCc*&;sv6?(z9 zLhX|jmp%XX1^Nb@J?CT!`C{*;{hG>WZ>2JNPtE(#@}e!&4LK|>dkyVvx8Kn3zI)q! zot}d|hGdsK*n8-mgr0lOI|ue)+K>6;Agp}T)sE`QI|=qidS}c2P5Dw*`xDvU(o~=J zdwRD){N9UD*UDz!qjwyvk4$x1@QqP+gvsL`j?`#tNoLpxU(|w%}39GL4d%BR#!>?tG=(qHlamt<|D~xyY zkG&>M`oVZK_Ki&#t7I}6&z@te@F%)XugbHzoqW^%#*cANmb013Xkpn@@XYK6L5Rr}8C#@=KmcPyWe6aVaBPksKux=?MCO45u5_u{nf1sXsb^uJFFK z_qgeX%3`*H{qXkHdq}4{ob|o8?{!(uCW^G0F zXl{a#QEMmv{9w~c^jq9Bswk3OzjAG-6M|~S}`hfV%4?uPrAFk6;{10TJ z@t}T;BjbyVG~ZALc3#WLo?}n5o%s;RI&o-skm1?^+FD(TpKMmQWQuFjS603sa+ZB7 zEi#+DsqesVcDZ^|7i29vQ$472`Rq8M2kBG!^}KPX&rcroef>baibwtL68u&+NLQJ) zt2q!m(QmjXpL|#P1Q}}_sVn2(xiTx4H1$38qOZ~!wVBPYJWFn?e|6)Uz3`jj0>`ojvZT{L)ib#+3Rt$MJrd zK4zYz?#+wTfjnv_z984Nn|T;r&hN(dR#tUp&f$6em9Eupd|}?fYTq^&)28~8c9Fii zpm+3FzC|_#%**(llvNqkkvdU->KM++^tOe1R^R#=IiX(2dGgNQfYSpRN@f^0^%ojD z=(ntBfK0#Nm1R)%!MVSvrv{6OVjKlRlwy<*Txs zzU52Tr`d|ylFl-I%w6;kN6*tOkS}Es7ddW@LFQ|Fb1(Br?Z~I5pOTYgyz$9K(w4@w zwDm9gjE@VfAvsXo4#esvNidBx~4!k&@XhizRZ^QZZiAd zw=;axgHO!7LcQ_d@G4=c-?I}hdx?JF^AfN3fPEK_FN}{$yzCrD?-ctch41J2wubL+c!!uR#OK9F z#vbxr5APTIMu)hr3H*4%^u206Lh;IT-$Awp#g8TqVc}d@deJwnJP+?2dmmeQ)QR}z z!}G%P$(e)FZ^~#6u8iJOrWe_Y`T@JqeQm@pVSi{3`c}VX`_*otGlli6c_-AN_SVkA zYP0HC*F5V;_p&Fotv;hK(f#xR{ix3w3v>k7QS2UL#CX!zJj3pC?i$_VzN2e&Et{8a zV+YYQwTb%Me5&%F{-MX{E(lAvb#CYATJ;>e(K&rYR+x{_vvd#L>fD@y{^f%rFUVW+ znk?b7B9DY~-+aKFjGm_J$aZ>9eEd}OF`Ar z=EHnEp7ji04xL9H3qwA8j=dpW`aoL3iIbfI`S+WSp&QLlpn$wqw8J zOOKJq(xQXJ!=EJI>RkTZ(>}_jjE?$&{-dq^roTF`H)pQjux+dF)MjE2 z2t)7b*YqSiiharUbf52mU%~Iro%L<@o@b0HW6WIBnBf!XGwc!LtiA>FCg;MliP<7e zQ=V%ZvWtZ=COgkF?=ZhH=Q6*bm*^yVmwo6S8=dW49ZJ5CHLlg}rL*a8b}gODr&K$( zzA^X3Yo5Z!HwU91>1q0so@T$w3)#~-k8{@>U+rac0ycuVfON?!d3KZz`NjrjE127p zf8?BaCkAC;I|x_#NA{7O>=|v@u(B!RS(L^u7P%Q%{}>34*H61!p<<)Q%<^18;FyBk)F1wZ$KW~KGITN zdC>;qu6~k+a+<5se{?bb0e^|#Y#eJ{^2`Pl-VxeGoZ5?y)Zh3qY9H#a=0yCr`ZeF1 z_EazOt)Ao;>Q~&_s`+dlwW0dcr#gm=PkmIKvGdfK_|1FSzQ&i|;!rQrqSN)s_CtA5 z-}MRd709!G595Fzf{rBv=wiR=SmVanH|OO;sr>MaF|AFFIdPIB#ubch&(|g=i^vbM zhuomI>*FWu#7$1QE+3AN7te`Ln(!x%%2IPp`rAC0oFl`!rba%}{qDQwp68_3{p?MX z+*Ces2y5<329k%;U<0rz*aPAfXYZ5IYz>gr}$hqnt@}Jx#U*Vi=BvaW7j>3@XmGO>rN5gxjI=%Xa zKC6$Id?l+JH$6i)i_<-Fo-P5MLnf2W(xJmV>*yNYHe;4GZcU;5HD zb>*DTLL2e5XkYDC9jYHeo6>9ANxqa>S)ttOlb)Bh-|C$`B~E#k*Xla`NgvmDJD;SF z^mqMRzeR05ebqCyM{1{uvuWs`!s)|~@~cnM%f^tjI$xEh@y4e0XFM5uwQt2GEn|rt zVf@U{YBTnT7;Eapbz{#QSKP)mIY1s&r;rJ3alb1Ej9dEB81DLx^#XMuF89a_a>z4e z4JtR-`(zPWNUoB3u8}9wbL=_!LU&fCHIJ1)yvptw@ycWE3Tcan z?xWl263{X7E>8KZtRA){-1EHr3ghUvG}H&0SLgDo4D^%o(8tnn&vWXi@q3PbgnY}h z_>~V$OStN4@eGGkZokE$j#_upQ7`Rx`h)bv1_aTx2>y?$3O*8f)7;z(AImE;q-LoQTa zbq?j8Jd#7wB5UYf@`=19AI($6BTh(*?5u1jdn=pCCih(Ptn-OMI^>o7st=G=ev4BW z&xmto$Wc0=G+n>RZMuV=pu_1Jc~zg{cdUG_yqAviqI|U-+^Y7F(dxAxOFl|!E#PDfhCf-zP;PZraGL? zl`r+ta=7mq@o9_N8S(_>ly_mo<9X$y2PcQu3NL@vch$SfB`$eV&t2Pa?po_mIC-Hb z#V=j)$*=tAC)&&N`oQq|NZZFz7=1zC>{xNG4~s_~G=6oU9ppvd5vTfVo|LhDNL{F3 z`S6S*gqNN^I^5UA<-TW{{~4|R>}a`r-lMvZcX-Y-o^@Z|^?&hKcN%xq-^PyoD~J3z z8W+Y(|D4MwoRbOuoC~W?#o;$OL9URuWDvdWTE}5+i zEMx@vR#~Z>{wf!gwPA*3o^#K@^YSUa+WXF3BX{9k-YT2)k;+x~)U!WFc@S^sJ7l`B z?zK$XM&4W#FT_*1PEI%e=3!WNt6O<+&vmk%exX|&)xYblYx#z{RnM(!_uJpaC4J%C zSI6?GE?ie<$^z-hU!Q6Hr~`cg&ZQ@fmO(rXD-Ch?PhCh`erC4$8qRSTad<`=XxqI`$IEVXW$Gm>2zx1#6ZDoyc+F4)J?(VBkf6a^Et*_yGz2C4c zPs<^!_{3Yp)4#^sIMspj4Y&XBZ}ECoobuZ|yVkUuC+Rd^*BW=@(+28pxL>Jf;XDU% zw$H0$_dVDA3F}W<@}jQQrTn?>zBt6y`g6T)C~k2z9%0=Z&YL{8zj$^yO!MM-h)*BZ z=iBe)+ja3aotCe8QLc_t;e~BIYd3jsUCiilGx}Wf-~4v`w4NF-`^{K{@!2si9e=GK z=hA6;9EFjNxZvD%;h_GckCwA#Z8&9724&GU&iyqGN59?Ij@nRL8XKriHfH=5MjZ0% zIdKY4H&xF$7rxi3_vl~epwsDTwg4SmTS9vB0O^T?uB11GsjVgs=~cf;PdH)3Nyj_- z?K=Hg9qbzWqc+HJJmQlkq_2GHQeC&r{qA4uRvoAtf5Pc2`h;`W+-u#5OP|v>YTs*L zZPEVe8Exr$`}J^meN{ZdXn*p!^HuW{{Y@IeIr>fSm@|PM>b#J?@QgW5=T*HgPS4Vp z(vX+-H+gG0^ljs=`I0y1uCXhZGd z8F`kz@Say+?ep40JL^mOrat1DKhKL#-(bVJ*S_X@&)cS+Q8sZm|9-pgS!w8po|RYm z=$Q44KXK@9!bnq}hddY$^5og}Ibr<`m&Z7eo;W(@qj2h67-NHPK>rt?_D}|Y`jr0XPdWW{ zyfuDtsT*mE!_n`~0X*9=pg*WL_32L^@#nXExhMYiXYJm05>9`2Up&$`emYiMcg>%3 zK8*Tkj9qo_zN2yFnsnr?W3A<#F&h6@E*WF)NmriaTNrf%>4{ffTGozt&%pEK40!`} zHJrBm$TK;`uS7mMR)%?&P3_srGxwbLn%})9{>l$&D?^`mtjv;s`Idj>(k{k=aw?-d zimNhGJ}W2P*9M*&F2DONkNl9!^68qo5+2f)7v|{6xcR9bM0Hzqn={ zL^%E=>q6E&tWjCpw|~O1Ya!B*7Qd2bT(?$I`kZwd=kANM*R2~typGm)>{)Q`8DZr^ zc-Pz$rg>=o#4COIS7!NY`Q@$kFudN=@GX1GEbr2je%G72Z^QlzdnEiu_fgopVQ-`T z8r?r)U&raeo=8V~O6}#am&Kk)=fc<wZ(^fIT*@iObO*6?<`tvUU%Sdg>k=>vEHV zqc&5Q>Q+1GtMxrrmpjsJZ0zbdwg}x~ycl2g+1I~sJa;XDEU3(|hGCuE9ysgq*5>UM zw`XscV85RCr|b`~X78T8-`+X!4uJh{-FI*Qx%U-ZbI)Epd*r+e=9>}TXRr_6{`WJ2 zcMt93_bx#H?=9buu;<-A_`VBZFa0&4|9_^V?_l_*hW~%letGW&z`1w2y*J?h0P_xj z?`7EYZ%@7V48+~A(&)D-?BDnOC~aXT&L-|6ff0Jm;Gm^5B^n zU$pqaXnB;?I|)rs9P%Sh?*;UI0QF=qzV}1D2V#G}eg57V^{zno>8odTTfDDfPolbq z-`d6dp}v`6zo+)I?{daY!uO2s{phq+lOv%xV_u^);5!wmdA(!>Qrq@YQg5IAp+=#?0WLF$az2ZKwP{^P7j&!ZKl@@WwD-ctdcUHz?oh zgTEJNd2#gpD(_mUkN!W6WVrv!(Ysl2?i!g*UX!C_B)LhB_WxrhSIJGXpNtodELP`q z2|0awAX6c{d&1KN{=Z0gPm8?wO(Jr<@sa<+$*2E^vj0!Byvw8i_tJB|eIy_BNOelz zN%Jn3en589ePngtS*BCTVSC`|TDp}^b&dWdd+qJBN6kI}kiYf?u;s~N>+sg&+2!UP z<`(9t?A*zL&A_&0!`X|&*05LAdqL#m4}*74$T@P&|5xj~I`Dtd`c2Lm6RuUZ`TmUm z514#2hKwgi?|F7CdH2)1rM@NS-B0g%l6T}A*#__4zLtzLwvF>eg0W8?kQdVikS}Bn zne(qe#*j&5%){Z{aC^8V+#GHVcZGYx-Qlm{FX2z&kKv|pL-<{|I$S>Bs_9ovzc&1S zz%|panSN!sH2gYT7X0?iq#*2dL7cY*X-ezaxs_8^}vj2avcdxv^>^(K_MbXRlbeo@4wwkY*+nMi~N1Bt`uW77!@6dMu z*h$76`Fe98UmpzK;UsG&jF~WIfv{Y7XIL+65w;I|g#!W^cT6}goDjYhP7CLTOT%x& zAH(h8-tbU(JUkH|3w}Q^;NdY3k6!=Rn179VHarvl8HDldo#D^nhCud;pUnPgAfJB_ z$aS*!ut5HQD(oIU61EDP2IW{itP++FONS-G!hsB)J^1r~virZ`{Xg;Q#Q(|uK=6OB zlVxNZd)j-g>{Q=@Bd^%zzMIFs^j`6yfeq;!7yZ9(cg~!^J2icm$GL0d7t9s>E`Il3 zkKU6Z>&QO)jpo?7Rgn<1M_3@ls#adsQIk<96OXuH9zz&A-j=W)#ocmeRsfk zAWw}AGSXP_{=G58ZZw8m>;JD#mKsa!N@Ghn|4)1Wul7FUIq{Iq#-y>y)@1*%Dal(n zC$l?t`~SO(7sJn!+3N;J_LH>aLw@AXcMODQUkNAv+N+M_y!539VO%HU*{_R*1p?c2 z&cHTxB=^CNWfKwzY|7^YJCdw_B#`&y8XZ8ky%>a}AK03%HN3F*2lDp5@KEraIE7`$ zJ`w&A+^0u`_xrIx*NB(BDqn1Bd0!|j8rY|moxU5fa_~I^<<>5~uSyQ~TXMcR+W#Zo zb^XLQNzFZcv%%bj?de?;{mWdZ-w<)F@4T2N@e|PZ{O|nz=2o3w@e}Zkk*nrjd}m~< zzHbg^tkmyke_`ix{EF_eJJ`2;1IDO#>Adr8+>*)l39wuff*8~Fm~ z=KgFnvd7%nyY+t!WE5FNW_c(7(eUs9*V)c5hw)>^kJ^}Q&&9%`;Z0$lKnAiqF%Dov>lpI5_(4IvaS6;J1744O@o~h7W~p z!-oer-zIDsgnQ3SY%*q((YV(hv-X%(!%E@JVW}Wb$|!#ehWP_qTe*~Z?x38n3+iTM z06o}xaL;e|F>7c&x$c?%jT)Q$bZ})2ho89*k=$_j4zPG^7;kyRJNRg-=64up7aL$ zna!+!*PpG=(*^uvAp8C9HU4-0v#t+Vhp3;~n6l1AAMm^JGqbhX+58v$s{Eh)cXR`P z9^ZxWYV7h2_MN=?Ech?^A^9GR{l3rMcg_3WJ-f*}`QDc&E69)P74|x+TYOulx`j+< ztCK_31D!9MJDW$Z6xe)Z*ur7{FzhoRlZPSy_zxz8&ezFBb|pKV46ZC@y9?);J}2yO zIQ|9pJXu_MOx9MGyGAyX*_GksI2lb=KM>p_r-g%aGN1f*kG`Nc$b7P#%;($i#}CoG zyVkVLHRvAsZ5h-HpNo1T|M^+?DcI6<6&)rFpM`M3v_6I9dr@cN9gf>|_4E2rKaTV2 zMD_2yzvq2*-qCoQjw9$Jb0<2fNA0?7;5*?P;U}S+=wf|>{-(!$qlbQHbNEiT{>2}{ zmf;IKH28J`{MJX=@x%UbbIba;`Tf~P-aRoU*iC;6lLJ4BwQ&Bw=YlcDUuf)Eck4P> z*TRg`B?FtzSf-bZaqoB0Pw2bL-dST)(ou9(ZOgvTMplp+5Dvo8XYjkeP&TRO#l>zl zcW3+Z^?1LWu45zfu{!dlH8Fqh$1;FsVFAusyO<-rJ^ZN4nhrkZuv#H;PE~0;Y6Nlc>mi#u**1nCSt=rz#H_VOnkNRz_r_ep- zEBrX-ulk;TC!GF8PYGk*!dFhum~&Wfc8%WhEZ_Tz!5XQ#i1~*#B7Sh=i~Yy;WA~{{ zeo1TE^wiyf?~+fPe&N?NH=%FXm)-}vCHyteU3}`sGhM>w=L2BN^4qXqjelcZeAdl; z$AmsIAE0CEzxFJDkof_>D*IMAvV`t3|1p;l-duoOB9qKX=q)z#mjmCFJd-#4?tD?c zpDaGQuyVk*?{P1+t{jXbmVt*zJ+w+FJ4{Pdh}!&sY;hU?l` z`)gk^($R17R6k%(oEDVBI*K(xHjFu~d9SqWbKoOX=^(<#(-F)4A-2C93ck!J)5X`Hd8&EyMhBpr* zb3M-&er<3hYsp`B{hfhdoUC=9-`-j%{}O+?H2KLL>u2}7-*RLBTMxG8$>#(%nEN$i)ZvdebV^S|Hutvh`-*L zg7HK)8TsA);r^##I*CM~k{`&f>3&?vuBkKt4OTYPs=ohvsTY0Hq zK2m=ZA2S~kJC(iKxeDLW;(>k3XYITC*88j{nA4cguzA_xY;E%kYXs&34}?d<^T9pq zqvByVv$3aT{$kCNFP<-*Ki?XHxrRA~HBj@52LoUJZ^JLcPr?twDdF_+^T1y(jJXG2 zlsS&)?+?}=tT9|1E(nfS1b+NG!sH+y@@>xID1F!e8ucvg^2VV4wSzSR?Pcyze?MD6`|1bg z6>I>yQ-9Nk=}KdR>?9+NLoyHSZ}QSuG=7a~vd_98`od1%7|O?+JYQw*}#>58M9}JHM>jc&Y?hEcqV^X*}I9?N+UluM7=LKuG)@#oS-woCT4i5VbunybbA0PAa zQ6JN0V>TOI8+g~4ca7G`o8vd)&zd8!v)S0zoZ0thhv_TyDnBK8%qAvV3*F5oBLCP6>;`sjZ9BdgHVj+GHMUl5S!R9*_~vpudp7t-_%P}x;>&Q2UxX}i&vU-*OJ18>IPy!7`{XrQUR}Vy z;Tg}9$^1#r26AfHuR|u1kNo^(t8?Bm&-^d$ zk$3bHJwfitr=#_Y`dD037P`k=##+d_LH+Tsm>;QIb*^6d<*eCQBd@- zBl?xjW(Tv4>1jG0)&#ni&+Zk5K4(wckILRtG8RgXY1;DcjCiwr7JZB`*&T zv72Ff%%R!t61o237`|#edL0BWK9F`7m2paD3KsNt6kfGlUhlYKE^3elUM9^vTEmGEtBqISAx7^KlcFt#x$c`nO$vLu=Ol5mmKebkAoEdM% zUH7Z-cd{kfE!OC|myG?`{bTOA?mjtH-#8h-UL^a-IWmqsVJERM$sBeVTak^%eq?7| z8GaQm3~W8N;ZMU^;neV*@XheGa8x)nd?D-`J{|T7yN2z5S9uHhPlJ6 zAybiQFNJ@HC&DA)fpCAgFWejM33m_h+dX0a5rls{{4+Q{8eI3xU&Afo#&CVOF5D1Y zyDNx8n$L#+gcrlhAtHR*`03-P&lTiVo>vUYux{8iygzIcwhKFjJ;L7M^Wl(ibofS4 zU*8XG?Q+`m(?;u4z5XKnCaC|~w)6m7mM)+x%mdi<+J_x(KhGsWJF({>o{NHZJ1?9Y z#Q(G4x9h@O8pI(k_O<;s^py1jy2HGMoTrQR-}{1b$*;`UX#ULh+90s=?A>Eyj?LfO z!!F_L;j-{Rm@qP77j4Uy-V3;SoE^v&zH1?&jPlfx#E#ZprlW9Sp`9|n;Ha33< zJ!Q?Iz7%`N^)c)4!t!_Xf%hBuYGzbTN(C1H#^6x3E*#F*xoJwhiwK8-&%u3SpVBL|7y&7~T{d7Y-A{vSIn~ z=CD*)D9jh;2(t~CecJ5P<_zhs}=@L>3RxINq&9PbMEheyJb;om{{7lUU<>(uzxrun7fCEi;c`1+<#*16Ju`=zX=zFABVHT8R3*bRF^OqlWglcn|UZ(wz916 zgmhkO9&7$>9&V1$uh%s>&$6BQ1LO}|g<<*?G3ql`7QO@ig6<)aZ}ujC1)oGQKadae z1Mv_0VgH;x+xCD_VjO0 zKRz5AjtP#(g|7$Cvz>)y^IjdU3DRQw{v=ednU8!jY#TNVZ1zINm<~}C}`|Q;N-@G}U^-%LjYc1x% z=HTSB^?zf1t?;REYWPF=cUW-jf@9YXWEfupc}Mo06Ugfyg&zjj4-cOVTL-df*+4eE zDJ&Eg3w$APeSt7gbhc!}bitJ4GF#Gq2 zU_H>0T|8sO@N1*z{{KS0Xlt?d7qU6+-SHk5e-B@^XV}&D@$h4#_V)1q-9)E&-^kh} z8<*YPHA(iZJcxg{z=pNIWBFh%NV!{QwkG)W5PW!k+auupc58)qg&PB33tPq7H$Mnp zGM}OS6MTtu30ce+X>T&WC)s#sFwV&g^5iRl4RmlgDC{5h3A+SilE1;Y=W|#mtQFP> z?+kAXl>_GV3kPF+wm=r}Lp&WOhd+l&;iB-PaBBEo_-^=iI60gUjt*ZA2ZsFv8~30A z`%T+#+9$)VfsMO;_-NQ=z|PZVu*bALrtK3v`^6xhPlS(ztp|AS)8WYQ{cvvhbzqDC zCdljW!ym)#fq&zU;CF3ycDmnx4`f~aCCnfkM@5l&{2Hi z^cKI?YGJ#;-q4p$4%QgJw`^UQfA6-SPg>6=gLet`Oz|a~|9&Bu+dmh`OSUn)^7L>| zVAJj#P6@Y!>0y;|tBl({d_UYAUI=rJpL6`YVO)56+|%Q33s;AW!g=As@T+h|m=vxL ze-3{O4}?d<T97Kgaktg~h@mVV*F3-1Kpegg=BIhap|EIJIxHWW4yX0wBae^V7Je5l3+IK?!|{Pn=#zneXs@u(0Kc~jTZDIq&BE4U z`|#ngY2X9m<9TfCV`Kjmt_&9jI`jMCD}nESi?B+d%jiG4?8(5t^NT<~d?tK2kn{BO zI|4rn-}`*wwbmYc{gqi>8TCQ%OV~f6pOMRa0c>tV0_n4M zm@hmw^|7f}h7-dl!zN*=@amLTr`#7Ng|oxg!hvDW@WHTtcuSZgJo(C#uiOxR5{?U> z4cmsb!^AKyJpJ<1FW(dX6s`{Ehwp|%!(QP7VZE?Qm>A{`gupXla`Bg1#XdEvS+Iq+HVzpNfU82Bpaz@G5I5@w%u_E{GRD}*(}#$ofYMc62;8Ws(V#78GQI^mXZ zRk%1@7%mCdhTDU9{~N~5I&Rjv!h&I`@YeA5uzGk`STDRMY#Tln4hTnv%pUw%41YP$V^4(p!=J-d;hgZz@P)8j*eYxgRu3zMrNaEdWQY&yFX5Nr z2f^44$pq0IyMwvP#W z1wLYHrV9sau34rV-9yWk_RYJ$g+GOBf;C3|8{>u@&8Om>M|LwkVGkFd0bA|Na6(`& zv*)%9Y+-he#{!wSRUjW%4axNAvv6LxI9w5gyETa8-(i+1vrL&QNNb6(Y@jPw39E&*!^YwLL0NVQ%A)+r zU6fto8@dbl858LkVrgge4r;odMgJQAJ`FNU#G$4;Fm zEE(1aTZhjFdf@87M$qWsC6Xq-CH#T99gil3&<@NnRKI{SEn>jm>AHNUc zd^GTbj~^L7GGE~Pep^^SY#DY4djz&GoV&)>-6UA=v>v*2SU8A-%{y)Ew6XsV4~M(L zP2s9=abW+R7QP+65!8|Ld^*?@LvGoZqHdLiFNi;z{F)Sg64=D%NcCBKRr3Iam|4*2Y)intPrR z-+96Dtnlr?Uvg0J&cFRFJA|!+{plNocLnyuTf*YO-jcDwUJ~zf@sHc*as9wA?)whj z6|nE0Z~R++II+**<)r=F;{^ zY!%iDZwZTpc>}wNU$VS@`s+>YM)yd(GVPU7KP8#{WOy|2Q~WJhpR@jE&FzA4c3|g| z>Fl%P!ZE=b+@XPg;Li4W3e}$U+vPsK6oUI&%P;r-ukoud)<2u#;WmqMBqc^^Ib1k3nVMZ#1{kkvR3#= zFfad3_*LNlcr>uFUO(>jxrQn=>*K2$i?+P1)%>p}+-{C#MHTRbcbB0$&UKx2TNcYwt&uqZk z!{5V0;prgUtTC$Pdx`Ltz`wCp;N#dlY!yBfwhtc-JBFQtqiY`s^1%Ng?|de6g_vKX zWmH$n%;#}KV2551P77ZRp9_2&>xSjRJb@FEJbxtI7w9N^Rp^mRgZ(Ne2m4j_4D=4Y zL?6x*#ss>Q9Dgi4815PD1Gk6Vo^k7xmxT+0y*u9zCxoMey)0h}hs?wm^L;PAeg2V; z1$$fQaNpM2c)+IFtN6a~!SIo=WB6#;HVDW5wU2TB;LkfdlY>1RY~iZ{y+O9~S05hs z3EPE@!YX0Wz&3v6)fxYH?*rMF$G?1QxH;Su_*3lLpEP(^i##_*$Zz|Ot!H}Q!@8mU zwD!YU`!fFQgEL+)3GC*thR=o%1+seSKt_-i{|fx3Y#=g(zqOFbe5{V$KXGIBpiK(D z2tN+w5_!x=yk{VX-y7Bs?+nWZ`xK($r#(CM*{S!0KZIX}ABFFRuZN?;m&1X9t@i1# zclcP?HJI1$5VjBI`PIcIIhS+B6Hlx>s^_)0%344T(2l0O@u@&Otc7h<$_deKmY5oDujWE)Bm8<{Nhe_48n0 zqdpT}2>v((?AfsQgYCLVSUfBd77Ggna~XR(W)12)uC-BL2Y&~}z;HFk5oUp~WH1-{ zpRjSD*ZAsp5A@pp!CdOAf!?8`=w@@Rhk|tm{b<>+PM~M_*}oU)2mX;2v#MGLk8%0{cINlOx5A0xo59g<*G~;+26nMEOvmo& zsC|0e*yBcRRCelD1DnU9Kkz}5r+f!|(qyV<%l6qP zNVdAaRS@>Q!M>m3ntQ^0C~Oyo<8dx-`PppnUXJz56~fZNp2B$pTi*s9^SA#7{qNy$ zXShCG63z_P55E{b9(on_cK9`cU3z99f3 zX4*E>whHeLY;Lx;BR$~S)`6dO+d#hX%UDZfx9=6mnooz%hXcaF;fTQAKRKKh>?5Od z*t=aHl+W9SxOLjCqwAfwWM4EH1?LTSN7fPPG5YMDa9?;Zn1k3i@qBnWSVNsHyfG{m zmJe)cGR~f9z9+J7$FN)AkJ>v}zqNOquH83~g$IQ%g~P&;fsdgaG5v_?hX-wDy+c_0 zFu+!~KCDgc9Wm#k%g+z`gncab)YunkuZ=a7;k`1}UhJn|H`q6{bMTJR5ka3kJMh0* z(`J863Eo+^ztjFt>pu3uP8c&`R6ZY+ci=7x4+s12)(!TpkuT)pFT!=f{QdD@uiR`S zvyCho_$Jp0_C>SD`30<#u^smf`-cOW83nRXB!x_OG;e>E(IC_9Qd?S22oDzgPJN$TnbL)TK z5Ayo8AnykUW!pRO)v}S_A2teW1nYf^hk1iPvi;G($8%$#D}El-?eT%FY`>qqe(w(N z2#W`M_}I$-2)6|LOU@1K;ja$fWBgS1gV+N?M)H5KW64Rfn|yZe8o5i(vVDamo5@yo zGMtmWmDj?1p8j4Vye-({y;QK)XdlU&1~{kdUANDh4;jvdnIqV1LMGoEZVcoxzY070 zsBmELj;ilaZ5cKUtA=I6{K3TSxqR1VaxjPFzhgI_7ETOb35N{!=~=I_SMQ@?n?N3~ z6O0+_`quKthX3UKVSCZ;3AYaRk4?(AM=uTDMK~v%8oH;$*tKu#qk){Thr{~OieZ_s zcvvXBA+Vju8GEw~-WS3@1KGsR<{L22|3wG+Q-2+< z3cnAxguBC|!QNoDCOdQCuv~asU}y5L@T0IdHyOYm!#{7%uy(MAn4jMK;q76SuyR-_ zIEw2X!T#!Z1@+;1dglXS*oHMPIV5}~93SWt{uMqHz8U)Mtiac#uFYxansWo6&BZ}G zuzSsWeiyC}+Q%Lg*i$~UuAyh{4eVroE^Yg2;1ioaEE)8P{|R*SdxQDeEffU1~@-6 z{3x)^&kMr-DqK3mCE3G$aX3E+rwxA)wC8bycYo+~z7Xxr9`~(8Hut*&na+l1%fB=Eo!;rrestQGjb`BT{9^M==lS;9-xUmCTq>}`HHu({ah{Nn7etHMQreMS$k_1J=J z#0|n*16wlcZMuDVbzsx}ERaKdzvkKOZT@3&|Leg%7q+k?$T9x#ZwB(rGvwJ> z!5p1m_>yo{Ag}m`*~aW%ex64HUoqd$e*&AAY~rKmtDh$<5Ecm&1HJX;pd4h@%3+n@ zxKeQ4oI;r80)GIU!*w>aa9ANU9`g~lH66(}rM~R>**feH%wf!B*v3Z$@=zUrH=Gj4 z#WTWL;p||ZWFPg{3|a9vYdTyjgl|L+9g@ecd~t8^OJM@?PNJQ__biYkbES|$x6Nve%|wgHN?xq zHNmQGcxL#&z7h@({MX&jO-5@=`k9QTtJe(V za&2z*^?ZS!gOagU;@f_F)c*_*h5G`3hknBb_}$>$LALh!fxUf7I4O|RhlT^fK7swf zA7Wi{?ZB74RNx2EUkQhJva{dCo1pWv1ApZzkksWC-Gr3JBlg;EdIqgV}+h<@+ z?+4+uK;E-Atqqd@=LK?`J|MfhPDqBcJAWJ4on*N@urKKm{u=W9@xaccU!DnMyCc7b zybHtE!6u~#?5(ypTzr2B%5-gzmNfZUE)SOt-j8Gh%h#`hbxipj#&vo2th{>1Qa+_6 zUHXQ9g(I4tVF%m0^QN#^SSrvH^a7or4fw0p4eVn&p?;U0hTc=>YdIkBpU@BX=kbTp z5#I{c6MScMg!!5^Y<_~ik;1!w#rxIVxa#@uZKe#1Qi zAdvg?JeTKbe;BI$8(6pN6Dn<2MBI$-0q z3H!c0xGiRDT=x=oCrNNg2Hu}VXoi+~G>ca!}`uzdjT6|}$&`S0Q_zXJ- z&kc;Fe+&L4kc*`C8Dkp<;zRY_8wP4k9vVC-Fb>)HHG{hcp3l5(Kx?@nzPX$Z-J5UD ze>p#pyA*qVDfnz~OmIXXCN<{Seg2T#|Goj+@A-<_eshFQZoWJvVDH7YY`td@#k%~m zRReiX^N=66Ecij7ZsGL6yf)9xdo>U54cG>nDj!`fA+21jy=TCdjCjRgUoofSvM>Bk zHpYH9gWmp*pmv5n%hnWPcITtfJ14tjKl@kfb#C0_{Pbei;GEz(_`URGdVu;3I`gi< zP62(1w*Tw^Z#^@x6mwY5ccxF%6Lt&m+lzwg)Q;_Z&hI+t&AmT5^=L=klV@f=9vl~p zel;yZd*fGr^(Da-0nO}J!8L*Y@$5?^uJqQ5cYELbZ!7-qv0FvE5NE9x;BNL)-{Nv{ z7(1#S?}33j5-}Q^`lx^}E`P!g@P2h;L>$LAkXv~|K-Z&x8%xI2F0=Q5)lU}>)<@@y z95Q=nDnI>0!N&r=`fQKH|FrcAcLo);}d+qdZF}XG*JA zL&0BV-`O_x6?BMoX3s{l@sAGJMtX(k#MyZo24A0@-y`7f(>nGKJd48SzAd0l^QqZk z`BF=;?i~|!yZ&Zh$B>J4Og7v(i!XA(P32z$e8i47-@9(=M%eGw13vnV zd8xtgf}aIq8`{b_vz(@O(q6QXcLi*=+CW^nYe4s=Yi|+IwDrS%gL?;_Z`Viu#9RF8 zD+50D4}$Z8GXu8zvw@nyCBX*+&m|rl7zbiFKJ_aCel;7tTd?Eo-Fxbt*y*hTv7fQj z+~cCz|M$p8Fy7LwbAGp&@prR#OI?+BM_nFxcHa9{4(!ld&@<4oZ zzQ~fmxTlxDEjS=}L*Ti1HCu9+I|k1Qo))O-dVIjgF(=m#)(iMRs|Vs}zW8qgzBnI> zcBwu^9*YkqW*5t|rR zPiqJMCeSxVs9Cgr+|KXb?bKREx`=m*@Lf&T8}oO4@W<@B!F;wq9=2 z*b@_q4dvCJ8+hNNXK{^Jeg*$R4xZ2PuHXZ~VZqUXasS2OjNsdWxb@0_UlVC~?m1yT zkU992V5h*mm5-ujn&V##)PC{V_~gG0#NKkJ>JHcrHDhA+O#=BgTI-GhKT%B@J47Q` z63}efayILN;PQ#quEv~?e*ZuoP;Cp_{j{JOjc4lV?Ry8KMssMMC8W>rr}^3ZZvOXa zf%*|S+;ajN%6EeA1$1~dCKm=j3R;&=xA~ty-jW7W_zr$s9{f14Ki%i7;FN$a!=I-Y z;e!JLcHeU;I|OXKJm+HrS`5CT!+7SAf9^d4_>FH)cj1@Q!H*8)KE*uy1@ai`RkStj zw-Ol7YFP( zP5apaU7J?I*XCz$6tLrJTGtEIka!OLcEK$J&e3(juLEO{Z(%(0sm}{&DSVGpf|CRO zi1Ew6J~E)I7~96Uv2M&8`{KM^+3(i{`v!72{BynuABC^NXZcLPhL3*vWqD`AZvua-=RKYLCix|?FF)&*!P^4Q zWqdC9dT?PN{{C}tn>;1W7Z<0?jXo}r7gHZWTk(DkwuOzUzH)rt+40rj+rhHnx53Tw zM^m9Q@Ux#BJU{UK%Dw^H_pv~&2fh02;0FP_NeicWtEax@Dz}`g30z~9HRkF@?vwWf zJ}~e+%?5#I03REudEa#Qe!|)FEm$>> zFI9uMERZL~9dsJpLR+V+)70_N;^5)I+5!KZ?ba?hjb{FhfF>m;K_~Z|tXzeh#ZH0t zrggK|4-D1{?h@1w_kZiC8To0TZbTjS=!btMZ|xIv(MZ_rQPY0u?El!@A@-WC&0dSy zo)}ac!TU6kM+7vH^#k#pniIOnT7i7!8o`}|+XuG}yl{|~cTFItV?46ud~mVjnE~DJ zq}j7Nba6h(2Ltx|pkVL7_*XwGZsl{c-QHit2cj>s+iL`TD07vMCEhm2&2@UEm|Scw zMpskFHpq*qD-_e$_dY1^`lJD{J-Y||NIoUI?VW9G7{8O9WAoU)hX$kX&8D)KY$(54 zjSJs-m4E|SpSk`rKe-z9RX47{m2(w&SI%kH|C8GED+XLX_g2))=QQPId2WcFeetYr z{erwFRgE8QTfUS2{M{LRx1Mfp`%8l5f#d%ya1H!+*TtXbr_=rU>aO`O0iWGH@c&Ip zv4EI>mhC>p5boRkYX^K&O^8qT?(y*Ut)Y$Ycb}HeYX750pCV!YM#jT)5ZS~+-jv; z&E-TNlK+cri{LqdIm{O~zs210S^RMJK<F?2kiVKg9itAEEa24T7H!P6#|#b7;Vp3tp~asY#$g$+s^*KgD2(x6Ee2=>+C+8&(`zR*>~EvhO74d z%h_}7be8`P=qqE+^Mw4rNVIQb{Otifh305Zh*SCFG!%N}qXRyZcL|Ao#lmV3%teNW zzxJCzyv?SY+vYlZ&gWy#<+lzC_=fyMKI7g2JHjvLU+xm{$vwMSJ%ue|)A`kWXg)C= zf?ZP|9@_pZJIFrLG}ufzMfR8-y)=-IWLrHmWXU#LFaO7ev$bryedPn$-0CCMORi4- z()#(}H`}=e_L)s)i`iiuL$6_f`ReY0kB*;4AAPjB^~bdVeydMTYhr_G?BW+ZNOz~9 z<3)P>O9L7@uB27TCF~!(CE%NXFrcZZML98GqvaIk6WL+?M?3j>fU8@7!XFoRsY3z2 zxxV+uM=$6oVlHxwG-z4`O@u#7|4^So!%%0Ue$ewew+r~eYXxj6o65HGySE6~-KPdm3!WJ~JJ>FG z?hHP|a{|9jVV~j|t2|?sZ3D+rJI~fT$72G%`8vU!gPR5HI9=u&v*-RloM(<`>a=@W zzI&yY(^2S0{^U5X3Vs~OZ=MmzZ5|WIE2vk&iLVHD4*25QnvS(@pw>hoa+Q3B$gFc?PM%V2%1;7*1zknn-PmNeX)NaiX9s5m;>oWD zV$074#|DQ7azpeL^WkNIxub?$Y|Dm=fB8)2q&cdtj4rx77&VkHrRIPSC#QB~z_(MA z!frQ5DKAI6rKi+x(@^;0>g8x5Y#qIVhD?{HHS?tTxaSAw1g8b+Gt@JyYgXTUY@mMS z6T#7erFDK++wp}!{fAl*`=1f06Hzas{#oq^_zceDw{HYr57Zu>8CcrSdfTa$=8NMA z+~Hc;b$)lc>P9|wZ8;vIx$vXK8MWthAbdp^q7P{cnv3`Z=tr~{zO`p|s=0_|o)NI| z@}xTi+I2pye4peY-W2E$zBilx(SZMrXX#&TyFBxkgRcba{8xk10=&#GrNGX_dh1MZ1Hb&20sU?C$?4@Zx@s2$-r1HL-l<*mWH12wfr1)mAb4fAGMz(1#B(l*US^VIw{@98UQ z&1kA@#VuC8#he`%%h!&J?>7s!3aYWt#$OkxtvDj^Jm~p>oTgmny1{0FniTm-dB>vz zdN!^3lHlrq4}PmvZZ-F;4c~X&!1Fi!Z~ie~o1NW0*fsFpzE=nP1)l%$T+fGs!-AuN zPX@;aCj_4gK0Slah2Ly*Z19QT$iQ<R(DnG;KMy?LAm0f-TQ|bGD+BpZc3&R!n*qP%q=3)DhcUj5 zd2>MgDxOsrzDY3FotSs*Iz5zL%2%VM%6-vVl?pHsdozR~)4cK@*$iDO8pBnJtwJB}A^)Pax+L{l~ z&hy{-?do3GcwGCA-~++oGoKwV>$_6}{m53=PSaE9X|xo&x}5f@U4CRRYU6x!+PHWS*l0E8#v<*n^(%6l#;Ke<8+}kv+q@*2_$LE-q0i0! z-!gXEc$edp_ZOd<1D>G}x0)Y+48*qbQkw^2W%JiuX9wsh?DjeV{e*VnnZSP!)J~{N z;U`+YYzCicgL-KWe@mu9x-46)Vbi6q5T#Z_78j74DpO$ZX$6(EX4$a@D zO|#wn;7tPB$yNa$`B}kEf#)t?6&w(}EBI(|OmI^0)!@wF?BLwsn}Ovwg0BUq2d4xl z2cHetaW;HO@ZsQnf#*E&!(PD)0=~NIz$<)m@dEoz&sPIUQ<3AW{oW>ZiFBk5gY^Sk zMqA;VYnK~VzF|&xTAuntHe6nTMoyn%zxm@&3LYNNQskQHUHVBLL*7zO1=j{w2cGXS4*BP_ay|unO*7}K)6J``7|(K_M+YAaYP;#==7?GV z`lR`#UeKIVQ$|Y>FN>vVDf}xrG4}fGfbaIHfUSO4ux}t2$9C|!X|?>$`vvS2AC+BZ zkJ(~=EL+I_(wo^_ww3+-O7QvMGr=*zQNc%p4+Lt)-x};6ygqnU@bciL!Ha|cn8D|l z1b%yE@Y>)F!9Kx$!GXb>f;R^TO>l7Do8=sb1n&&~Yp6$Ga^sRY|6fd?rk&sKdR;d^ zp1l`GTokxJ8V=r}!?<_6grnGXHk}<`GpOG#PbyB)cI-EPH*l~3${BJo=xpICtfBQkf!-LkP@EMH1G2dBR&gW*!=95BMtb{Omm6h1Q}bQM^ijG>7;}v`RWT|4Iz~=HRG+J)o&v8_-kk9;k0q z=f($T+tnS)cdA$3B2cfgZLmwgp7SBuqO$|~!K$ene`xTG;MKtg19c^GgX%!|vbV{* zOuW}*?cn|aJO1#1b|UAvX+Zy`Z_`Bh;jakT@An3LZqN9CCs-EzI`~U)^Za>d*X@GU zf?EeG2Y*`SPjlN`k#}WX6xff=|8j6*z%S>^bbdLVg$;jIz~`oYt4G;7Fdozuvfm(9 zD;}RZ6*hgd;Bmo5L%qtPS>53s^X|b_gKFUX1~y#H3Vmhjlk>}sNxI(^!SX>${*UevkxE$EhK#YD=piYH8 zORJ@Y^95^p)EBZ_WN+vFaZDWxn$W z_SAa5HciB~d~rTCu5b+gx_6fF-TByfgfB0zDR;>R^YLlxY;kpU+~YZ)>gu#|+B#mk zCip`@zo)Obmp*@k$N1XS%GqeS$?7R=G(TJ0KxgH@WeFga6g<2Qw zFCRfmXV=x()6m}*yfQ2xYgVe`+D}`QGxmEnQ3)r#{}j*d!SxL{Ju`Wj)dMzut$^K>cciWC5bPNo9I&_S@7IG1gDZn;gX;s%R48_p8w2+Hn&7J7$H8|4 zzWJ8|dJBL2@Y#D;-k*1~&{W?Hka` zcMi4RbDR@cn%m8PqQ@L9pUQ(Z;o{;^P*TlVS zH?3TbQ@x5j=h=ZgC%>FO{-HoW)5r0*9OaV&c3CaT`oaAJwW;)O-yat6z1eNg^FVF5 zZH=Ql1Z+3Gj%^nIvBP4*(}J%A#-^q3#wHt0Bd3?M+ZP6MIX?5t#geB7#`Q4)8~(O{ zo?=e$b;Pl14r!2LUw#w6T)v8)$*+>UQK&Y45n-=~6I{gI~&|TzDcM8~e_FaxtZBg}fTG*jOEejj3CW&@VZ~12M{eWF(yZPl;2Y(3s ziQ_z9VEn8#t2^}U!3II=Q|R1u6m^C)lxe<`&8B-Bqc0Em8^-K@!NGy&_1SN}_Hlun zCVMSMWG?Vy_%yW0YXWoXPr>yA_~L+gSIVE**Bmq##mIavdZ_vZ8mjq!exQc@wt1dk z{ULkKhCe6ZgTE;F&w$QCW0CJvYvH*dwn!a0yG}3ve!%xsv+O;2I(=`Gb* zmge1CeDTi(d~w?!8nEMU4cPG42YBN@0~*Wo0=68du;(-&JVP78Jq+8mc{jl2ff`D+ z?BbA51ohQlo%rO1vv(nGndgW+|FmWxhM}8Z5zt%YjQQ#B4dg)g2zCs%3APM23i=!; zzg?}Ue5g8dcAu8cho`6GXm%c7(_cOj(9GF)&jQg|_~~+-&2R3U94FmH?o%H0#R2Yz zZfo4gf2wJv!QcgaaL>RPll%0iR;6RoSoA%v&)@vb1sf}^Jg3K4e_cRd=d&B%-wD`u ze#_y3Ik8WmE`iR`6Z=CMPb^AmGf&QHu~FxIoV z?7!{c(-Zj3cJ@0epp_gx!1p5qzuVS1PYL+@t@xRqg@hf%>MjOumrlYXq{P2APvE+dPo2~|nmcmD2*X8u( zf!KBWBE9j-KupVKv)5{R?;4m_>cH4;OLf22Sz5O!;Aio})Q|Dk<-5#r{v7?4-^X9S zN1oSbL!KV6=dTam8hDnRPJUP*=c%5KE%#iDx(v^qtI=2*uv6-&{~X*o?`)9^q(8hg zczfWv`LhCbA>Nm=a*8!>AE-m5!>k*y-2xx$0Kv>Te9*6|mXt z_Zh*rg9`$_`Xzzo4+DNX+fTc}2TKBai&%ht=cBXja<5Mh)HKq8HV!rjT!U*`Be-pV zlW^4Xfd4L6;Qcb*3-I26#;$I0r{Jl9y2bkiboJ|3zJAWW%RhWB(9ZZ0k9xj#tKgBr zdcj=+n!389-vn0$%L3ZE+~_$${q@@OgHzW+b78~vJx$zrFm`Sp{64u$aowfC4`-S- z`^vYrl#h|0{a9cuza?Okje9yGe@6WlZE=KoV!fJN=r-&) z-JC5K<8B(T<+K%ACma6zya&--7ITZi`DvdEjtLGA)JD8JU@zVr>=)2k*p}8Eiv8)i za&z)^n+Nn@c5JJssuLfGsQ0fo*KdUfYkajWfXK(dVX#;|2EJeg(ep8(rV_v>w;unp_({oLv`p&|uhd zdbzqqa9{GI*8e`R4Y2!`>Ui9@ed@!nnj9+sTx^2}qoT!}1d4W1}_Ws8Kjootxv~&4Sj)q)0U!85g zcVJxgnSMU|Lj#(Nd=1@Y%!wM0v=((L{B?eZx+}Ka*f$408K_4&IXEStCz>bVbH6W* zjc+__K8a!Fo7j4O(&_;_PB*84dInqGix2kFK%E*ty)f6+9Q%!b#xLWay*v=N)7(8H zdR%a3K(}R2Xan8@qR!>T!T!Mq0{-TS!IuL5Cwwtr@8wNDJ%i6D2Xq^CjeJ(N?M``r z$@+m>zg>g91Agxb!P0SH0h@o1;0}S>9eR^| zFb#@tkJmmQ)VDu4^_#B?;0fX_fj z=X3uy;Bzbs=rQa)9gwy{n=w}TCUh1YU<@1Q{FdW_BLcZ0nhPG-GhpA@bK7bAXd>~KCU;ldW?EtSY5BTad7IwSZIE`h^VBO&U0h>+3tKH_G8%K{0 z#DT^a&D@xyyYMaeA;zWg`Idm~{%CM?a6)imz^`G$;R`ePEWb&Uq?e0DX_e-J`N2oJ zJorVxzjEO;bF~KMn0l561=U8?D9}(h3$_U8FY;mXV(h(mdb>b-dmY1lI%vPdPs|9G;8WAlNie z+kqdR6YLP|8oV&rJ*f6W_o3szF*q=gQ~Y4?@c@sU9*7Z^&7MUigx>yUx1Y0x54_m} z=W0(D=lM)}KfU|80l(QhZ2u#8S+Hjyzxl?1Z>>g$R(@!3WPn>}NciQP0LT0=P``6& za9Kc)VYe;$?dmcw4lW98%bt73fPK|%s=ZhDcy>U;;p3kgd@)d;&)0YUcLe(duF*9= zE5Ln^4e%gcr#An(ydzyrSA@{_<^=d*n)VAI(En(NYlU;pJm zO|!fyyTOjo-46@c8?`ZP5FcMHMtyubFFl^!r~3k120pWGY#dw1=7|O54B12R0KXM4 z(3sVEtA(ULv%&Oeey`X7NAQCSy0siFn|+5_UE!VcHWwNZ8*YhP{AN4*vGKS=jDb_w zef%-PdcWJw_Riz{<2iaB=ji#I)AiD7?h%X}f+J{H?vI`&U*Vq15B&AfSEqsD&K-t2 z_4@1bANvOSLhN;DKu^P{SBnR5%^3k+DCDU9 zX4`Uva~=|igK*gnfw*AP;2{Biz=w*5N?dSOAP)0R2W|7tVBbJ|uzf%) zQhTZGJp+XsX#Dz+Kk&}U{~@HtMlO~fbdNx-t+6g|B!|T3Hz)Wya;Wl7VqDMNc^;46 z$$u9gKOo?zi>c|TYS7JLaku#U;^4mn_ML6#=kfR0^Wy_G6}9D4&4!Q3uJGl@_Y%_O zfvxu&e{>=E-FCJY4-{%Vtn)j)*}2#+dIKBgIUC&ao%oN!;X`w8)%wt!O=g$rC?5$v7Koi_ zD*SbQfnV8d{ljkKUp&lyt9dyqptZ<}>Qnx@8kq9~x(nNm*ZJ>aHC#`Bp|=x`j1fA! zJbQil_42Hcam0tGyU<+t75ob|={}td1bK3JQ_(L2p!aBe4 z2W{uL_Hk_c7ryscwzZ!;pzCp6ILbBhDfgKNe%n7#-*-^pSsTZ2PUmp_IE!zJr^F?n z3g{B@5BNnKf}`lz_~p{Tdnn~9e;3fG=p8ub4uKj`+~QfShX&&_xBTRtg8vBciCE;I z;N1bAS}gLZ;N(Eg;+#NEnw`Zja?^bCJW8{&XS!%u_X_SCJT%xSV6UGLJSBK~@a*7u z0edc9*(Ja^yAAN&dbK<|1~j4V1N%NR*d}nit%5Cr#|4iJ#0|Ku&r=b8)cpJ+xHNF> zxPdNrZ19ml+_g_27cWlJCip=+;bHA}tKhomN9yy{ZR!Kh#nFk3gEs`mtg$Sg%iqBj zYXtJR;)S0D^dVXhP4hE>c_)_UdwEXVJXI@+D>e-77nt+n0l7DJfsT(O)K~BiPY>7} z^))mC+5y|dPO)2T9-GOIva#}meE+=zx({29Gw=bsKXQcieitX$md&@{2i93EZz`k6}CePw(ILxGvX< z^W;6~H}tT=cktWDi}T%|V`)om(R*}W*HXUJFE|LVh|%ah@-q5Uzv7MNV{rukUkzmW z;fBB7FjsWv9RhMaVo2OT{(f_;>J|7!>jrAs)UnB-^7o$-Y#W#xe0+L5T}I5!x0hdK7uiELneDAD zQt6SL`TJKK+|*aY{7zwns6LURpjLbL~Hkt2{PX9MHei+B`ehG3aw^WrQ2;3nYr!uAFK+A)f*@Aodd-qJezIlt=w_Sbb4 z^cZlxu3PNJKD!t9ixbs9A^)oHY=_GQ1dIlbl(R zFV#!ZbUd@f&h!1Z37#G7F!=p69%Do9RXq*gUoDui!`?3mJ`vFJKNpAr==JRWIe{7n zHIs6#mjpC>gKnvmri3`>a?h`yPpbOy%z7nnwAK(lA z6V6bBK_g=K*hRLEon$X}3D{2cIqayIfo3$;*_mS?j@V-cHn`|IzuS(@#7oe8uw$~- zjt|aN(5al?b(qU+FkAfG!1c1z=C{0|dy$`TUr!F)`$Y*E+I+;A9nCY>B z_Mtt9qrkHSG->sM4-dpx+EU(IZjv3DLJnnw-)(Cju)pItp5r>c^EjXLvrp<_T-Vr7 zbn@zW?py!p8*QZxwXruFVWoSZ;f}nC-U-u zt;J9BFnsgB1Sy4FS^eg1XB?#tsQMavOIVo|WHsh;^bMZU>gU_MejazetuVDNcYx+;W>U+60bJF}=El}fhV>ssz z!PUV}1JAOkX)^zD&uM|2p4jD+0lOjR@!kLj@gw<@Yz>=3a}d+)7|=fLC27`|g_Ac=4QjX#FHD5Er{oIUqLQ zGfi~DwF2$I?&EH4M9Y`=pXP40r#eeqAcw1+@jw4x{%PykJA0pzeAKgo=LGx&Tp=z{ zk4djoTc!@}Wr4h}+WKi-rkM530WQ(^p8KOG;gk=}o^zE$I4Uq!j-Ne0B~L7mAO=1o zFs{xG#20*=*3gNSan168#%erX6^Ntdi*b)}ihuY?>VRlc#xE}Vb3mWMNjC(>wvO`` zGqDok*qEnX;VPOg4ND!7c;=45on|mc_+@GV?-t;*y9Z*Pd(LV|~FX$P*)b-G@E?!o=IFZW1WxGr$tcvw4V4|OS^ePI0D zb-&JYpZ(nJ?c4Trytb)xXjgE~v3d7psRTnjD42BQCqrdAcu43V2$8j!J>eD$1!3AF%b?Zx0F-1%rgj29G>s+fF4!&?l;@n=gEOs%dv129>P733UH3N?0$jka(!YpnxJbJ%iWmg9Nd@O zlzY7_aR1st-b>rabM^h58>p${SLp}+!B-Qz>MQ;Y|4v`>H9X_6*6jU2 z#=G%t%otb3qB*`SxG-Q3*amij&ET7|Bl41J2jyAb9ndat&%S}$ar{H;_MR15hj@s6 z+c9`vfQ#6_Z38h9pP6Qko7BDVquEV596QRM;;Y5M<}+-Tm}=92&7FeoqLzj(5AywZ zPX1>h_}zB4w=XyjIKK0NYj8cZe%HswyWY{pyC?TXOB4&xecZSEubtN>Y`q$V))wPy zc3-=RBiVmC5iZxBfDdSuYW1}h{Kwv z;x~)o_|R;c7*D*{_y*@RMtge^3!n;{ZX9rnV@+cHdfsL4Nl#G z+Elq6S`wSwScjedeV~p27m0V+aq+`y0oyL7kh7r8tsSfz;3WAUytICRn?ODYPk?-Q z;k(~#^WeZf_H_(=;&_hh_|7Ai!GU)T?iAofF_^rGxQ3%FKEpM1GMtKUa4fEMpL9p} z?f%D_T->8xOMRa96EA90+^(HR{t7QyM`z6nB~@fJ4ipxU_t_H1!5J z75I8Z*ZF+}4qga7uH|^H^0s5+ZgW_T1MV(wo8NdFXa6X`)%aQ615RE@?-ZYjZ*CWq zdm7i^9QFq1usyUkc4&-kaE>}=`XIfKc8G82H*`kNrr?@a2J#1No_d4k%h*G?s>1`h zEOpFmX8C2T8<3mue=LmLa`j4A&)Mb1IOP?&4f+(DK8|;>L25 z7|wmFm%ypo0MAZiAbA{hVeG{27vjSa_-*9J^5nQI584kG7LNPRpxbxpoP}eL=O4M59ydbl2J}+! z+3$_#rq}8IwUIW{?)pX_=_@{j{#z)fD}P!R`oDb2CmBEcZj6kvpLh?C;z}G@{;NJ{ ztd-l0zv_c{Obmy+#BO-Wm^F6!RXBqkH>TO`ug%_n+L((k*0_s)*k^G)b5)odo;8`q zWAueFCZjbrE@SgSTn6Tv_zVy53C+L8XyUV*0zYyD4uNr<{e3sL8;^;@8h_E?#8u)c zKB{;~9K<&8U1^u>1)Y-ppfyf;M}9|~Lr;@qqNk0V!%p#=#W!>uIkCnzt@1x8j;V1Y!jGns^59u-9~} zbpm`NF2OUn08h}r=wW@%<^}n`fYh>i-^VM1*9ZFq2Ly8Z2M2Ep4hhtKjj-M`{I;>J zXL9xn-Wa?#a6HHT&p`n>w`*>j40Uj6js!qfnY8|iRrV&u@oKr~PKo48aTidofJ;u<_7--lns!N!R3 zV(ie7#XI7ETEoaG<_y2cTr=;?LGy8pclen6O&W~3%vYt=&~423#yb38c0ruOFK186 zFKiQDVXxROc8slMgW1l;GVEV%B7LiNylLy%Qg6_wh^bXi!?5_ud##c z#D}iAc@8lfznw-kpYM)S3-`{iXZLK6XX#vH%;)&*COgWe($>T@Vw{_b``pXeKIIhm ziyz7(ZI`xFm(@DugQ|7X5b2zKzLefbhoniW8u>m(>ds;^wesa^vvd~Xt6X|*Xv%~Tl>@fm-F;%IZte9ES2xtui95TYj5o? zmcvo{qxma+X55OqjAQ=Nn7b0+i+$*Y{424A_(U9Gj+iTrAI2QksP~mi$|JaBM0o|b z6y=ur_<>H*T+OJ>&G)%nGX1SOo@F(>F~?Kg4(IseG_xIQJa7*mwfrOYnJ=f)dJlC4 zY8l3y4xgBBOsi4rp^kvgwp%d9J@VSs*QPZDY$3l~eg_Bf)!9n9a{jygPUD`1@;fv& zHaI22|C`?#V;?aD9uil`tKcHGosDPf@e$iEFM|tk0&Y+{wo`yFUKpq+^WFf@qTrKP z1bfZ{>(wON4(HepzwHvJsq-9*`a0*rgJP-812I@R$F;5%h<)xD%$N7^OgtS<9l7V? z<$gT(M2p1Vo^6^>gHz8Y{t<`j7r8e+gZda8q%ZX=KSR7D<~0r$1!CjX1LMb^F=kxy zi_8ap7~hP~#-}R>iQ(x9<{*87c7cOv|2RlJHJwD9gNNuS)!nAH7$>m-Ywx6bd@k9tqBI|BgAC`s`h<@ZA6Jzx{1R z24}^6Ifiq&hB3AiJCujTF>Ef|%Lc0-V3*~anipfcn+s#t<*?+q9u=rTX7@p@0v;G) zU11yBgMH`Qe>?_W@%#&vYw%6iS&rP}c(@0zHJ^s>+=u&dUpUge@;mU0cuITdL+zx^ zv@9;+BZzkv;t}I-As!Lm(8op&5y#*U?~V}D)11wNvDS{C#!ip10Bu07eQJ;8Lc|1j z7-|{#eq&xlJpzn1405w#2lEux3HXjQEcyBQ;5p9`wy{0fw{R@SbbRNbcZ_ur?7|4K z%zUs-ubYnVSdQyFuA^ETZAIL}_K1Ddvf-Xh0x=NXO@3OAfWItXjfWP>`HXy29ZpO{ zhignk8)~fqF6uL;V~&S?^`6L^iiPM>@+LQ*lyX6G54c>T;8V|o4PhN=I+1m zVNu(r9z$IQ{ysjSWy+hWH&A;o)}giG8uglVwsK8t4dn3n7<5+Sz?d_>jr&nkhkobZPfibVp zb<-Sz{p=5p1&%ufzjnm<`*goe68AWsV>`F&E$8yT=|%kU zDdRGzKoD0buT(*SYP7$36RV|<9K#5nA_+6ZxlI7i(CyD#p*Ta9(_miVN+Q!Xlh zmBVmQxyrV&nUNuW$=smOib(A-9vdsOj@47%eWfXXdhfNjajs_c(e6( zBiD$@X=!|<#w&C*8d~e^TBm?-_*y>>@D1H-tVL+tfqQ6DVh+CAs7Xz8_Be@OhnM6= zZaK@*H&4H6-kpS>09OEbgJ%Px+PI!B6pY!~(5%Td2lB9Kp`Z2hp+0LHCKa zb^oEp;K9kmJT!P%fHUN1rm()iTeh{2eb*0g%mae^2J6joIpP_bpqPe+=6b|wVjG-u zvp^i@`pY@qztCE0_e@ixNvdy?8>3g^Y55&(C!TDLn`gZC3-JA3ff$F*rq=zYY7F>5 z&F9=PuS+v;?4#y@W`TpmKkB7gd(b!t zC$Sl9ID0+X>e^}Cz*d)^%29ZzT!p8^O~5We^W>J|se+wnzkt1Oofx||1;2Veh{>9V z5X&?+D_1mT;-lk?#xv|GO^9u6T*IfQG2)QwXkr}tSv5v>8h^AV5Pz`aI7GZq4jE$~ zwqI;9jeQ!kh(*SjWsFnCSY^sD)yy1w+^cv_8;C>4I7a*H zpOIT|HU6$1CXO+l#MQ=_T#cNf*qa~3AEJkqU+7`YsZZmXv7XzQr)|l3;FRi5d@_0f zZMu3Bn_N3Q>P`5<+^XH>?h$voUtEe)$2z=z)+T+yUkJIP0Hz4emWGd=ux&-Add)#JP$=O4vD zeWquOfo_*~t+Ds)zIcNsC!Zn)x!+8edq`pvobsq(qoF2YjDy5DVj=O)2dRGY(3>bCi+Yzls2Uuf=PCOk!h2f7`U zv)H4Nx5`~?75Ls5WQ)0|K_ZH$FlBT+t^FE@k-Ml@g4x(8eV;wSvqW%EeQr}=I>`#auPBQfSpBYT5e+H{!o`^Qr_v`d|)7Ht2H49^m)MtX^ zjM(ZQ1@a@_`^k>evHIL8old;L0ie~1LBt}sKz<)z(d=**h+6=EiJ!zWBXHUjez&dt zXn*1_$8~(?aX!3ri$MH?Gu4OF-PH7{Z*#qPNQ|fMLe59r=f32`)CAJv@UncFIPvs= zZi&0ak>W|Ug<>Reruq+juYJ7-Nt??BwH8j_h+SI?Cnr>{qPtFO;nXScON_3Gf&WYkZ_Fn&;78GHzl_T%>Kab7Lg&upYwU`jg**>+wI|01xqP_&I7~ z_&)p~9E5*bL%?Sm`KR^BYT8sOzn4{%VEDk z3<3RYX?`_+;U7+8nAQ_mI{t|2gnr{!PO;Fw3&lj!IH)m@Yi#_(m#yB&PVjNXZQ>s> zA`asJii6mdYIEfxelpEQ&OluBkeNm&Z}ga1Y{b9z?gloD-_8HtCBQ@S5%js%DKsYX zoQgUvdQ#tYAg@vG5gW;`v$5)o`%LP5F%o-CryFbL%0KKr`+u9jvq0q@aS7lcz#Vu9 zkAT=^iH}BC=OD;NUswwwI*Aid3M zH__+W0y(Ft_9t&V);(1JQ{TW6=ZIFTOOMV%KmFjjweP;t{zB zHj!<_D~(6^_iQH{T1|~@ZQWku32_9JPs$vOeZ4rk2S zs5=<-H?;yqwYX|@maS80uE+6PJAiXsn`@R|axc>wHL=)Ev)Z)gcC=Yzp5}M(cHe0is*~~63hVsVeXH5A$+gRk4cKH_1>etS@dA5YP0sm7?Sk#c zQTV_;xEC>t`?B=C_h%b%i#bb!H-}xLW4fm5eb%{W@eF=(T(J>-o)&_S%~$i2{vjuf z%f*zSKDK(O`D@LPYE%j)%OL$*n+1$%|7)YN(A zM4Y6Sn;m2m`^?bDN3GwcJK-dH9X}s0(WDv|ot{{LEfy206O&I72Z=9m4?B-n@PK^H zg9G^|5eX#K!%4-Fm>(Bx=yj;&U~@p??_ zoY(p7CwC-<5;KXD#78(tJmi@l+#|-rJL=RM|EPbH9})wpgTp^~8oYx?`)DWaC3mC^ z)gj6ssgIx5EZ`zMgoE@C-dP-oclj9)3&cQlJG{dm;g>85_$gv%zKi*w-_1MoizaT& znylI7&mMz7)Sh~h2ZxQ_7UG3V~$yUx^+9WK=Hk6GT!P(xn{nU>vtdSr}@;z z`qh3L=U3;&(?0X38Vi=M`DBiXv)d+?&NZH=nnE>Z=K|k>F6}zT7}K#G$319UZR(!2 zpE)RoG%sl)I1K-Y3&m?V(R}wuyJJW2G~0rA*&FspJoLyweuy5YK0%E(eTSbd)?w?s z=U_>IM?9y}|2v2O-WqH*U!M7AcRlw*qgrL z>w^6P&+7R954^YYqrs8ECxa6L?-4yIIC%!|`z-w506trOIrvI&O5p#LIXyTd@Sf1K z0?RW4>y`#z4eWd3?A@Us&wEQh7JMLZEpG|-58e>GI#9>`l3@2>w?OUoa|7@6QM>&w zfxMIalzbHq#926WQ6QcYw~6V*ck)W*E;=6lPE7;dj&3Q2RLel8!|mcuF{nC4x~Mt^ z{(yK@9m6gGKBCRFh7KQjHdQQ)cZ>!8QGGAmf>ZcYa*{ZK&s7c1I1>XIb7CORL*NBt zkw2ua!5lTu>qFrHV^@q*9gbd79jrn#8LKeV;RKQy)0q|wjBJZz?TKwKj}5YMz;fZb((M;;Pa;2$=-+%vxa z&b#2e6VCtZ#_oI1vvwEU>i;EPc{w)IR z^v^oK_5F92<2&%|?|YA7+iqWe^X}8z2ky~*<2F2u+r0k{zxn@?v=g3F_pRMA{} z;W}-M)6{I>F}W)~0v;2q(nIwjKSIBX!RV-TKKUkNz<4k&#KOi3t~0L0&1~?PYbxIv zqwFwFxhc-WNBEBZ$5t1WMhd+`^Ug1Tf#G+-aS;*gO}F z!dx|H`E&d_^BGq(hT;$63v=HeUsJA!4~jGRrs6k#D?j!T0WC!?hrS|@!|xs6-Npvt z5!xBsq^?cQ#`7igGFln`8kf+3-=VW`=1*)CSEG%_J!92Q{7GO2d_-u z0jDm67TSCdjmL%ub6~?V(b}%VvVsb;o_OT zX57-%#2m&tzlRTL9`P;tn&ynTNh8A_v@w2N>k#U9;S~N>d8L||d^0=19^eQzqPdsm zU&#&*ZCwuDu4)D%VwwDcNi`irxz;?6w;uSGUV<7R*)xj@v%FRO=YeDYDsZ0P1kNeW5@Wd@ zf36SL^}VXD9Y^9#_v3x%jhEbGdBrL81P-2+J2KY#A0lUOLu~VKq zWw-DVokp#%`Zjzdx9$0!4+d*FX|7aJ`Opm9Ampp%IaaF%!oZ?WsR>v@5gVwd0r0Uqn~rTA=*8SsMG3BSE6 z5S!o&v6S~Dc+bJ!0UojRT@2%Q+u6r4JdY|4^DOJG0q(>zo^^FiPYYb*R)P497DwB2 zPhuLmCHWz_A+-|Pqw&nB#o>1~+G58(6ExOmtBvz63b7|`(=$QpK=9Abf&SF5-n+g~ ze#e+FHtrj!IWS(vI=R*2*We?%+BD`d_UO3gj%QZoU(E&M)|fXZ`Bug`e+(zJ)&MWn zuft9JMe}{s1K0z8CV#Ur4BNpE#Yb|`?27s^%R>HOpMXYFp6YWpcxcSw;FB?rBbH&y zcA9{VYkfd-JI(L#t;c$4{x*2_W_PXIg{Z8f(Afzvhm{=Xk1>RU;F_l?$5hX&nLX#lzx3Z8~55Kw}F1#h0LGibMGr z`WGMQZ?Ubum*?TDw3f}-pidbm&^SYUTun_ZU|fkC#M7-Wz!~%~zL7Y z=1#c+5AeT?_2yNK|K|6nb;-2j5n={5dOnCHKn|p!S^G2R#(ao)gN-*IIF&$zN<{Aux)ytP~cA6x7tA4JdLcW*L#?v)?DMSuhO9uhItTF@JF4En|2gD&OFT(8+AYAB^6NNL495tiT*ee1Y0M=V-{zBX ztH#c}G>`dM&9AZnd_z7YpNwzFUhvhL|KZ2+>DU?erZJhkDDI>20eeN;z=74nr*#a| zT01t9-%Jn3jWoW3J_C*0s`K%^3w9S*ww?}OLcfpPi9g5Zn98G#@y6vCZw`)u8Gz3=0`j^TUPHCDtujV~=*TkZOQ-c|T6M(ut0JgyVlx@WO& zp*^&b_R@~Hj4z^HwXvE7?XMrIzv?G_hTFv6%@N@-z7H*H)Um3K@tN>e;~lZW7$@Ma zzH6tkg0a|lx$(&wH{dXSnz6eOm*FvE-I&Kk=D?JP*mAbNHoW#5SKtn@2OHjJlGt(c zj2&0kz{j-gXY+9FfPDv_t+QQWTl=>r!ZB&`=C1QNXL)Nr&T@_An{rOM(wqnUX5Wzy z9diVJaxU7=* zcZJHsh+A-{n6vyT_EgKRo_fCkk9yv>JSyL&jzb)TJH)1VL_Hi%RfB+AaO;u)zp72Z zI|aT4@hqrc0Q^+=?ze7hU&nGx$94|qa!%)V4O8HGpc%SWHRJSMzJS~my|qDSG)VawMLz>b$t9IdNiMj?}Y!14dcX^X^pw@YJ3`Z zW4u=#lJ84-NT)J@!(r7w6$S zF#=mkp9J<*Oem*yQ?=!6^Av10+g%%8`(2*wQmlzb#huPmu<_3A8eEI(a(&HHH%E=X z3vmtfyQO=if7#x0jDNI6fj87sYbX3Mg3mDKU>YM?@+-78AEP-FJ_lX6@e*G|%+!2K z{Sw<1Vyt4UUueA-&Z5uh>((%cyFgB+Xe>q(z-#g_v^?X5UQi8;W=2NmQoA_Sc3e(mgmr8@Z44*i2^@nn@eyunOjz#f z5{H$?a2fdScYKFa@gFY4k2td^Uyj^p-!bl?>&=H1?7f0BN8ZGpt{cjqzjg}Sd29bU1!<#I1uOYf5v(Qnm}_t z#==5qoL7w!*Nyt5oRIhq$I&9hci=Nv58}UK+;-$exuo$KBROZ^0ayARZyN7~@79+~ z@uqo!C(RM@V>xo5#W_v0+-OemHAfCK@5Er`KXVh$;UN4~9xIobzvaO$t?Raph0OQH zL**IEat*#&2<57g@5E8%9lo`AY9ZA3wh#Xsd>+yDBaim_TsPo7z{4Z{5x?SP$L%@B z^SXyZd{D5J;Mzx6XB+!C&h&am-gaHCv;N1}clu``*slF%|4}<||6@Puld)ggw&OkZ zJ{IbqalLz;-lO|;T=!aN)57`NR~=97uin3VHHJGzjq54=R-er0@%Qf6_4vvhDPNVd zs%PM?V$>Yy2Y{mix$h)v2)eCs(Cq9;n?#0WBz{{7q~Xp39flW z*IT+L_t$%M@2wH54b-+Hn&%nY+df5)F&?vRI&RZ-U3Pod)a&Z8^bNGHdTotgYLmxU z#plK=mer;#ji322ZZm%VNBei%j>~?x>@~O^f8~>oTlX?;OKqVI#_?vH@!P<71D_l7 z*;ZeUZPs>j&bEhZa-C!RMwhFOIMw~k!|IRbV&|iI2#(XX?EQ}YFn81c;Gee1LYvKZ zyzXPU(0=o+@3F={?e)8t-cRqvcHO5-$Ngu|-+j95HTC{#itzfJe)`8$R>e)N6E zPUpGt0Ua~-i`c9BNgYeI8{=3q?~EgVuBYQi-?PW#*kQ+^I&{BV8dt`c?K_6Fb=%oJ zjrYS=^nOMgT>H@Z(K*ul?b7;gYk#PXaoiELOSLhbH?9Z#xjy|e&Yf!E)wkVO=SunO zrj~7s&ZBYcX_xl@xO5%k-}sx)9XstO$82AX*W1s2>-pVV`@^x^_t+;ZtnlCeZM${Z zUyR4G4cmT>Ip1}yuquE5x7XI?c#Y$jaXf$Hbvkc(qVuBHXW4!?7W#So{iZ(KzQ6Zg z``yyI@!s*~bPTnRw8Pjp9XlN_Q+wPw)Ol39Y(MQY?YpsE+9s~gb-Itvh0ZPWrRcmF zm*e~y?{&Q2&XxA}T<6UGao@f7j!{eP&@oXvJe_0aiTka8SdJQP&1iG_d)v}DTBwa& zp6mWcZ8YBheB+?wz4zXG8}G$6bWH2Fe(v>kJdE2-xAEH)w)y+)`MV98w%c?}wVk?b z`?lTe-}AdR_tL)8m;O5U+J57F=;v`hww=a#(KhS&=$M!;+m>x7%YGi)tZmV2w``kr zoo(D7Ot-hK<2wI%zw_PmIHx+Vj1}m78ab)!x-`c+w(LKSt+re3h<0)foiF3P_x`$E z(K0LUVQk0w?nghi?K^I}Pq*)PeNc3rW!Lq*mR;9l_E^pZ{yKi#W5-bM`=*w*9nl!2 zHpsDiUe_^STXihVx9k|O?|7g6?AX1B-iLb`|2oE}*LG9ejQ7<0FV}Qwz3tj&<8r)z z*EhC-dz#+sKbi+)JG3pvd!OE8uYY>&wFT2<$9$K5AMbO#*PhFAyyv!W*Y(`vz6<@< zb{vo4vvcUz@o#L;j;GoVc4V}h)@$pwU+07I)ApN=ALn#Ez5kn9YA0Tj1SM8-A*mAbMF_(CXo!1w(b?xrU{koQp8N6%k zxTf*^J+FK0IdDb&vXM8&zVmtPyYagJnf0!-`}7{WU)#I)Jw2D-{noMB@!9P=E-meA zdw=?D?04VCzv*xG=~(Fg`mE^ombKYkwr{)5xQ}!8xGQ>mg)Qhg97EsfD}Bay;(Ly@ zR=!)`ey!itc-Vf%U~NIq?;OtAw$kSQT&KVO);83BJ{u3lnXzX~SnnPjx7XTn?!3mB zbN9H;VS9byoZwvk9HZCS_VT@B*ztS3-iP18_Kx9ry;e)#ozL&~b&SIGxz@IsW!p*{ z*k`;Z$M1D_EOi^(`P}cWc|LT`jlcCCx}4sdZM&as{An}y*8a2f9bDT8>$C;fuI)W{ zpxHmY?%v0IZQ1=T9bdcl7>?QTVQd&Djr)ujcB*<%H6Y7s*>q*!?dRC8LqE+o_Kj`h zTHpEI(zf<-O|GSV-ZpiA?#cC9r=Pp8vWX?=fAQKJ2x$ zKRXUBozK40<2gp}&3&13v{>y1KDWKKo!|AtI1bwXoy+5Ijya;+I!=$Lzq^g`qd)a! z^P1I$N9}K1XFK@?xd}OlWr5s+b7(txCG`nvHq=6>*YFI1S`zQ7lFO3A(T6>^rFQX~ zniTbzjw5%Z@8w41K0Fub-0q#-l$TQLpgzU344!FFyCR<@r{o;CRNK0L=a5(PyE+)T zEZ3qPzvE+!1=Y2 zes-^YzO>^@e#>v#$Fa04un~@H>}fOO)cDl?`rJCd+sByE2kHbmmMop4-<{j|vYqk5 z<~gTh>5JCXIfrdPKj{<4_g9-|2pS^69N&QGwd z{;U7sv%c0}^0xB1=1_CH`p9_qyS%JzE%l+CHTbOGI)+@EbGq)fh5KZ8U86p9J?51* z)@HR2+Op3Dt50m(`dyo8JN=|TjDIzaeTKx`FprD_<6pbmPd~IT%uBT)YK_3>juYF0 zF$3*8=d$0jz;W@Ev15+95B;EhjA?U4TWDwZshzc%cIq6meQT&JwU>S{?j29t>8l>M z{i2q@IUL{iwz0qdaO^R+?mBD>^6ciOZMDDg@6VXi2OS%()pa|rYp`8yz42Tw=$z3X zK6h?8Ci}0Q^n>=*Kl&5cFYElKU$wEhr%$!DerSH$Qvd27`>7SPjxB3HwSV-jYcp3| zPi?aOj0NAF!+t)StHzLRyN_#WAM0=5dt7a1UU=`3+C6pj>eAKNtHW1&sP4}1xJ3P= zcGm|!tHHx(>ipE|sXb@+^ojOn$Jix(!A9aTwR?S5PQ9b&w(u3+!e7Rgwsy^~*F4jI zYAB31b5DD!7gj5$_DjvQdNOrpYNyqHEe-IMxu&hvTdC(#4{e)nuii|}9L~aXI8Pty zBlAx!mpU@fo;p|GGp4Rg-Ltwb=V3Elmwg?>SkSk=w@+|j+k)*;li1o)%RZy1c2PZ} z_Eq1A%dO)xs5#XaY8*X}G4A8}H~ZMv_G%v;*Jt}^XEl-5`E1+m1J4iWhs8mC5_4By zxL)&xO#pL(EjMr6mws};>KyqBfWyrXW3;w~%`)!lN9kK_r2bodoc?hf$J36TBaUOd z>kECVmald0`rp2-r}M0V<2jaN>vv0Ig)gEu&u7$g>x_Hzz-QZeZb{pFk56lu)mf|8R&T7YwX<4c zbA?S)EAD)J3fE-buzPsOHM3cbS$!@y;YHhm`(Pv4xbm5Oj7MYDTw}YzXV7oPr@qr? z#~_^!>hA=?Jp z-FIzVU(9+wMQg~lv3c0|NI$lg*nISxZS`M$iq2nsqp#TDv0wCq>vsKY#e8GXJ>qcw zQs)=ntnJJ18|yO7Nx#|8cE)i1L-r8tW1Zs~1KQBI2cNsH`NxpPN<`)o{T zb2dSHg3pd^edmZ{@ZB88(lxub+LDel>%o5Rn_aI>@Yy=o=ziU!d943kkL&Zh{d;cT z-J9*}qjXGYL+95%?3OX3eYKT8ZD1RHqu)UP8Uy;xxF|37z8wQk8V}~Kv0*IfYuow^ z;;xQC=K%XUj^(&+ye{qPuh-UbWgG43zTJQO*ggC1{+865ZLYyN9i#rC`Cxm0J-*)^OFO$q;~TVP=Z~@E8jK&;?wai5 zcktP9T$}A|Q~0h=dmiVuy<^&@cE}iVf3|1uoWq~~&`0KucDK%cY?$_QPS;?&&THG+ z$2@C4xnFSq`q4ZqXR_(@r^5y!c= zk@wuQd&PZ)@76C1e73K4cmIyBy&V^vQ+peS#-Mv`oAe$nwUK+NeRBQn-?o=+#y0GD zF%}$i$~*4eIm;QwoU!M6jCFlfXb;z|@5Dc~(~X7Pi{IU&`EP8o*Y3SzO&jQc+d7{4 z=RB6$q5B(qj^lf`>m1To+RAlyKijy@+Cjhd&s^~5-U`Ps_T0bm)wZ$JRvmMe=9=}L zgB@4))o#wCy8t1y}ac#ykPI2wMKYi|=YNNFYdo;q>VN>}?#!&5SeQEo)P4ta+aV`F| zO~Q8D^0+E=@~fA?ujwEx_1$B1>V*LC-Pdynqh z*f1w-*E!H}W?kor<2jFeW@kre1J|J)^u>IC+NR^g{4P4?+i&)FoWgNO=vU`+9j?c` zfa$f_23)75{d!DG$Lu=S>)MQ8eP*2LH+I`{KKKpvgL`4e9LI5;(=}&s7&D-AAbxgPy+k*S3{8FTj;SncSm2y_Ks&8?dKfYx^u~RaUHIyyod+7U&opAbvwuBS1t;~ zfBett1MEfZo_nl)a6HGwtA+C!&(5JeYY!~lm-D$N_s+ID2Mx!vT+QybjV#$-*X5dM zKQtm*5kJ{JuFo-@(_hDur9bV_Ytg6r)7&v0jF*CaHCFVowsfpszhfDf=3mED$Cz>9 znCyW4j1Svd-*M+0qi^J&Z_Nh(MXSVgbL;WM_-x`ollm&I|i@`=CC<7&KL93*l(QJ`O~>XC#9Rx!R038 zBIG#aI-2k3dpNvH1$?KYH<#6XmYhiUSv;rIFP_`Zx^dfXZ@uFo%;^az6-6n7U$}Hx*oY1`2qO@@8a^V7fave807Kf z7v%E1qhxvyuFEw!pL4T2+FX0HN0!=}E$a2TZu+=;0Q=Io$M@E_PHoh7avt|$`_`v8 zCR^7&9_=HW$lmfH*xK3!k6cC2TgZNnym&&HE+WxV-o9nskvxQc7FQa^EjS;orfK1PT#VcIy*&Em)E4Z%cIcX zmj?1z(|3Q!HG(`0`0V{5^3UEaB5ygqf5bb{yl15E4zaYY9Gg6|T$vnN--GNuFLK=8 zV=}(C%V+Dn59a^zbtmxJRdfHxJLkSD6)I_v=FmV&GDNcm$&f^7o+V9^QtECnlX(_R zglI5^aF-!v$e0YNR8kr!Bz2$D|NFkq^?mH;{PUd8>vi_N_OSNaYp?a2*4k_9BRzfG z7iZ~wec!X^0MnnUe_ML46ze2yBGPBf2 z3*`rWiOO-N#Ag0?elYu>43)Doavk}RFZa?sy6@(3SeDM4jq7zDX#HdQIQ65nKdbu5 zGuK}R>-r?s3H5BKThuf4afoen4%uHDaj5IH3GJutsk(op5XS$snc5RQ*3veIv<=ZC z)e+j5Kk2o7YJ+GyCFA*L+BW-#e9`tRTgZh6gNy|EQ~Y*p<>gQwStq0Hcf4c$Ca1_c z@|axjoLOWj8Cmq18!uwKq<(b$;+->6pR75s%!_3nEn^reb1jW)aE`v*&Yfl1IfM15 zJ5Jg?)_&tzjMFq<)^MJz&TnN~=gcw(mbv%M?QG19`5P_GQ)x~tb77f(tMhPmF0RhQ z)%mxYw(YL#yv}`T>Gy--SZNM1ZpS=Y=Fn0;4>o^R`@Hpgcb)Fi2ku#TCPRHek5*TD zrl1?EH|Z+6Y1j`-XV3%mLSqE#p!(kWtIpQWC7bJGYA@urHq_X$cBfrxr)049OU`KX zWPv%d$Sq?F$UgE8#`ch#=87T*jYsI*i=9`>c!kPG>2y9Vve0(tI@VYrb3K|H%RE=s zJ7>Ky5XK*wFU!2Z=1n%2aKCwDX^vrY2itExE%R!bL(6Y-3!9U;b0v2!W53O#+;n_f z#XQKq|JE_=K=vSc+$Fip zJ|!Q>Nix%zJ#vxk^Nsm_3&z~H=EL>9OY_zH?xOG0Tb>`xW8ZmnJEwh*F&DgVG@4hp zY1rpmCzphNOXjMDt{Q!7+4p1a4^eF+d%j>?>gITMS7#luoz*|2=Dd}wgY z-Qm{YTPwE(zkS2mw`nYWkJE8WjKAl+;>3!hZ`3TEZzn%&%rf~N<;vkPVcoDncv{#z zY&~X^e9v>xwP zrpgwwr}G+i-MonOL%*X|*>~djdo|{oCm+%I=*fNGY9w#_-5K8_hVMQ5mJAucb+D|= z-#lN1@f}Ftb@5G?O~Q=y-(vLLvwj2GG5rqo`VXu>`nKe153KgU%42+o(>Kw4f6_O@ zd=t(0(0o_ZcgB2Utlt;&J!aqMgl)bz*6)lpowXlWd-RQI->mcD#=I%`-m>r4`|b#t=ldkrg}(=SV{&3Ll75MaM~C&o)5nnMPY6#5&kcKr zgTsMgkMZ}$j?Q<%J`{Wl`Rn2QaC-1PX5Y)S&vz<65k3=qE7P}>eN)*tm!0Pu&6fn< z$@GoP?+4o*e?hR%H#ME(8<>utAABqGn&3N`zW01;I63$Rvvd2sOxJ&7czv++%}n2l zmOsZI9$pb%6|5f;b`Lv*t-?0p`C-REx6x_Jiaw-cE$KC7NeA5$l)LXTdj|ck%;TWwH*!uK&n?WKGC#iO%0Hri*d=t1_?uWBUqn3;;}^d- zhLjEhAD7?3&uGj?T~lXaUA^Sfivd}-4e&YCVetwv1%AfRC*T*b8Tk`*0Qv8|0(Lzc zm))y9vypvQ%=b#z@3#Bai8cq{1fCmZE~JabKX zpa;k>a+EBu{*ZR}IcJqXp4m>n_+44-`^UYe?Y>n;PWr}~`?gK`PY!gRZ+cqyU1Z;~ zeNpY|&+`y4|j`VOdXfcE>-#Fa2kZ2K{d5g`k=>}zQ2SbJH0W`C1g@J%e=&tiv@W4_r~y}=K7e0Xx$Jdm9) z3FPR$fzCKIye_;cyf=``9}gc1)fXR}_~67x$B@r&4JU?qq31c?G17QTAj3}z_WKqx z-WJ{*obNa~>C*{e`QlZERuh#AGz^TaVK#o@hJUX47bLfTKDwJ@f-9V&;$I0_ASy4^Z|dJjUz5i zC-Cd(1no#nK|3V>eN)6YqQrsN_3U}Jytoe8-&ji9ymn3wh@0#htdjxG>3MX7{p1FH zLMAxR@xDJv4#7Hk#5W+XhO&)a=$nsZC)?Zifop$0{(;AjY)Ez`xk_GJvQ1%~JO(*W z=JzvA>-)>dV;S&XIQdleNVZ1#&Pl>FG~d9Fs{8!AJ8w|Pp{A~?D>AP*uA)> z`<qug+=S+4N+8zjI0c`#t>LF}Y5Eko$D4vh24!Eq(i#jHl0)xiaQ&(akRiws{`# zd_5~m&&hKKo7!`<&3DPz3FIyMy;62H`?~Y;lka|uk&*voH2KV~SJ(O-8^?4me&;k! zNmq!Ku}{czdVv3`?vmsDUVSVOo8k|PwboDOBa`n?|KS*ZfOjC<_d_3O7kmzWb7N;> zV0>wPVD`12i_guzx;F6Bt1m3YgT(-}TlO3OLOjeLe}WI8eRoWm-;UL<0_(PkozWj+ z(PC(Hh<*<7Y%y-{@9B5Y*Py>vAA|S@yHiYrPvIL=zS(qo5U(PqK!1=`Ah*P>z8vfm z^XfOE#Ky$L$T@z9_!fC*$rqtl#B{`R*w5r9IVz4NZnQ?ATiD%C3QrAN1UC6D;bq~) z!Ebtpo&NH$UwCDpd-e)14P?0GZeyIgQ+Qr*zH}UWP~hwE3-~?s7_3Vdj&mJ(u(Xe! zfPFiM?E{_h%pl+VtJMOZ!Tr%0b3$GdogfW1yyIZK-ypWM-*MzNokEXP2a)N8j;j2Z7jX^x%XRn~miF5&AN|I$ zYsy>WUcIJ!D*OZIN>7}|`HtrU@CDqj^rS;yC?kC^^lkf7ls_F#*Q5R&y-h!m|8#+U zw$%<`^Lxg0f8#IqJKsJ!q~EmEU*y}AbOJws?LpRq?Jn*}=CkeD?)t^p@3Z^J*#D@H zKu4&Pd|LI>dHOwu{Tap$h>?=zWPbbk`2wALv2=`M`4N6s z*R*YESGHNUjp?_aeQxcwZfDc+_vseCN&SCgvc!(HYxbbna{ZC|BHDg+3O&Rpp;zc0 z@rI7g;hUIWM4t=&vPIZ7RLAIFC2#aqn=j$eKvvN=Vp{r)$*_JS*pePHuQMOcO5-jzNvXO`i;uY zx1otgji#~T_X!$bgEx*NhllF6HdwqA9EvG$bcd*Y`CiWMb zj4m>cN1N7nS-Xv2(>`@_fPX{I&`U=L_A7ZrE|E{QdD*@6PU9MM6J15uh-s0NbW!ye z-DI6kqq`axk%oQr6dB7d7q79dpIdB442B%8T~EGSrxWNMdZhjUd0u9wlMs<*1Ue0SON@AsAIBmEHlezH2k zzu^OeuCj0D%g{4p8rU?B13wv#Rmb=_{G0kR;#PDUou)pjpZXv8F&(4EAK}kj66hsM zI*Omu?^OEj7`}-5Z+mIqpY+<_C*Cz%*RkpJ;nR0)y!uDIQWpYVGKS}(k->)Z&^&{|`s>_DH6Q4qV`*K15zy5lC_YVpD z4)8hnAN}*&F^;F}_$jM~#<=JN>$Z#ctsV3aJvr#>-!{~bqHpLxx`xlge=E*&zI0s2 zHDH{9{uKRD^s=~_ekzDJ!a5yp9Dw^bM!;`lBZLw@f9k3_ zYpj|v@LDAV?nhOZOCs+?MJ&6bFkEowJZCyVQrZm%bum@ z=sWn$&h1z+>-I~>zV`K#J@jDVud%zyC{#wZ&tFWP&K$;Q$TV@9#&0U)I_^o@(v_Bd zjq8Z#xR&D@!|(SNYxCPLUGYaTCBA|4<-z&#XuoZZA=&3T;!ET;IbNAw`7D0f?fS3^ z`A-kf4fK8e7(R?NoFk2W#Xk&atiN zDRq$SrxW-|WV`yV@5GoCwi!9!aqeV2KZ)E&x3lxC^OcNgah~|JG}wD$57O6WhB}JR z#4dDA$Mu@haV_x-?Q%GFnVw-|>aU?^7Rq0BP5q&c>)|K)t4^|nSf2NW>SJ}xvNUby ze|3Dm*ox!&u2OXySt1@ows;S#eZ6#+zFzSgV~oiq{+B)hvWmQRor+gC3z<#NGH6q{-*eN9H>G%WntwA_l5|t$jK44u43zPFc|Nu-@Nvk+S0RvJs&C*a*t9 z@TJ%h_Sr^vu{AscV?y{)dxe*c$EfkS`z$?g^?>f;OQ|pFjry}{Fm8Y@Q?J-Uo%^Eo z%dw8L-S~NWZU`Go{nRG318qdzSC@zSu4B)%L+zjKsGSb=oprhn!#SVm#l}0d`?h_0 z6XXD2&yq~h-krlntt_YxWp~n#)rVwX#9qud!>o$SL@^` z+2|Z;y9SwAd22h}N57HH^cB6t52zlf{;-W6BgcCT-9fj|BcM0v0lJ6ZK^I7;dVv0; zgB&~58P2J`>)30*tD}1S(3hd-98>+|ntsb~b+CN8=Fk?Pi`*|g#AfKcl#Smh53ynU zm5KA+yELl5rR%rrxPSNPzIu$~+_U4@2cDa9_e|&{HWNFF4K+WgZ*+$G!v9bgE!89S zmEKUd)I)KE)-QDoj-}uDIeeI=DLr-FIrW$5De0-Zt(q)BQ8|o>2S>Ef=$3ai&``9aVE=qkM z;*enD`(2xVxU}7VdXHXoonF)R=tS3{5BY+w<9!n{Sf7h!-#@W$?_e7_t$)TiKUmjK zLzn7*Gv3d&-3Ptwo}r(gEa#8Nk9FJROPRpBW983r@<$iAws8ZlZJdbw4fE!j%F40s zN7LTourg7cs z{OV)c28P@0qdTErOUuh9%i0jq38IRXIr4l80np)7fBD=)`8*u%4_?sW<0h3^m)0~@EX02 zbk(V0T~YqE_msD1)aTS^WdHD(+0W3v;kix2Ggm(v_f&6Mchs+9Z=FoB4`hqq{NVbi{8c`zzuFT^491(VBiIXU2tEUONe+|YWOx13 z`lkF6&>^3}UzZFmYRbkbE~|?8Ol~c3XdB-~ZyP@<;iqe8Nit|Iqtf{84>p;<(0z z8~ehi{B55Ce9;_sH%G`ubqKzlZOH+2&m>$Gdh+%Lz;4-ff?KPT%J;uE%>>?#F$( zKljQf=R@<=J1(ek0P%s2FFIp9MxGC?-_W>LD1Sb+XW^N6HlC4wM$gTA#m3Bg=4>lA zn7YF5V~40$>=rtoj;5R0#p<;Bt*=~r(!R(%?NvJ^UoHt`16@QIV*sb7XWsp}bNbHdo5OYCH{n;|=i$om z!|=WE&G41*#c<)6^QO<6{&e_wcyD-XI3=7AjtPf_{lYF`hp<)HBs?uVIXp3}6;=08>qMp!dg z_87b7UX;Du+1#Co8E5vIpKNXh2h2FWns^7U^pbaIvf!W56Z*(QQ`GL zSt+x3h4+V#hO@%C;Y;D`;oIRm!87jO6aMSbDd@_p5(GlEzHpSAsC^fH}F2h;EBEg3HMLN+&MNEY+O*vf2p^^sSIl#sO%@-n}$#Lf;V?W1OD;o0Wrc5bJ~ugYgE>3f>=j zQ84aq|6shstHW!ge}6e!q#K9ncQY_yRbueL3rU9>(39} z_TtGGPwo<48Z37XJBH_l=Y(em=~((Lz3s!cVXLrZu=Lw8mYavCgeQc@h2GD4lk1J{ z!S=Pnnqk%O=fWh4!k@=jzdQUR{4LxTZVopE%Ui-(L_;qSpYmQf8;OHMC2lAmPa zV*@#jCxp!dnYVk`H@qT{bteb1>{HbFeHcDYeF**o8rO%Uw;@wCu_yc z*c9@lZnG`K-`LIUX!piOXREtkb)K!y#^7TpE0FX2T(M*^V{uY(Onsn@5sUq56JoOZ z0nM9X3?)0mxFlm(j3qL*#rPNFT{@1(SQvIl$G!A@v5uFqolf%H?Q^`bGSZa3dY~_| zeWBWXW74wnH4(`C__sBs)u4=sR_t z_6sX3@p7?XG4s|_dFZ?AeSge#*mV3s&w;(im*b~Z*Yb5+N5x&}Tk#roQ{B^#qHd@w zbZui}Vs>I*WT2QIe}kOVpQG=`*bla?_d(4UVJw63^c~yCpU~&Te=r7w9j*UL9~D1= zkHGF`$Fs|g19@nmpH~f!3r`5-x$)@F2%80C1HI?^obbZ%vanz9?$c4>m|*jPcw zeJNuTy@y3#pA*gt^a;IkNuax}e?43rz7wttzX^2FZ^KW6?R35Kr6awM1-5{9dA*PMde&^xyH{ z^x@S9@h%p>qdtb`qz|tBZ^rxR|Kv;ZHN{o=rDB-;SFuJhQgwkmAus4W@|Cs*ANLP3_mYy(H-FrzeW_{K1dr0hC znzJ55ej_!bar~AcZ#75|e#=x{gx^1FoUE)=X5=HeNTwOnNUs^+ZhWNt8fVXj=vexWtM}V}$88dfn`F0JH{Q|xk&pDF zzBBz_{60(lXZoe&NB4=XTz;fgj!_9}j;FHwHEo zTT?so&Y$+iCRP7E1G<*(^lUm_i0)KR^o45Yp1b&>`XKfo2En&u8|x3?+tD>-hH+>5 z7xb}_FKiXEM&Ab6qd$huFzRR{#fN%X*f$&!jtOrH9}1rh z7lm(zZ-g&}^TP$iQ81-{$;6Zij*?{>q)4HLf)zYISL^y-(w+2P&cgm73mFz~BA5APi882H|+ z2hWYJyCH}-ib0A^8lz%tBwOPXK|GCaW0UBI6l>+9tsV3SFA~@}Vq$zUGKg(2_H|BR zi_uT)DE;^J6aSz-8oSe&>zPsH+mGh!AuqKHejC|JcZh*$Gv0mmj;y!|pXBZ_f6vS} z*8djBEnmjxLneHQ^7z zx^`~6rMCauAT9glNgnSF{|?e@p5@>D_@h(nW6?MC43tC5z@8#K9n((cD0k!Wz0cM6!jwOo!G1WGF5E11>^}WQ&sEoLk&*m#rF7&~KFwXx z_nYYzc7%IWHp)!7Sb8SP?2aHVCZ-AN%3j%!osO~H@yb&C({oUkY>|YWAMhTvI#Rns zyxztbQ`_TE?2YXJN0lhAMtR2zy?N8FS*nsSu>jJ+}|6KcnY~u^_DcO$Xy*_Mx z@%mLZ58AHr4(w5OX~!Ap0}Ixk2P=ZCEf z{T6IzeG>Ws^h4-R&<|i9Ieh~97?uib@P~&*$9Vr$oY2^Da_VV;U$SbTW0nsq1@WdQ zhK&M$W&N;PV9(P@uH_o5gpGrk)y{zr)vlhB`a z@IUk8*~SN;H_3tTh98ajUi<*|AKynTg4{4RkdGwR!Dk>dDksDwE(!bq>4}TDPxmVg z{snnse`82|A!GFT3GSclGLBq-1HXm;qD9PSJ5`LRKnz9bwGP7R+6SB8HCvF`1{5rMBbKd2jG2;MW~->BnsmDno1 zBaR}*B3>m1qpygJJ|&17-4s@sTw&CIa>nEtlYb7Y%voj5reV9VLwIR8I-D7P5dIKu z2sehmgvq&+b5{?Wh3AHC!-iqGFg0gt&aL6v;Jll{9pP`m`VYf7;SJ#xVVCgS@T{Z`Ec}!3k+4tLAUq`8HutuN#HxXNUKMR|n5v`Ecjtos+)~ zKMI!y{@B~Y-a$Xw%0XY;!-M`Pzt;|YKK)sij|t0%Wx^_fFSJwGH|!rQ4-WjSi^E@o zGF>Sw8y*%O8P*P4h8KpN!n1?=^0e^Ouu*V~bEUyg>hN%I*e#G>PYLS>GVFQbm4Tf2RQSpmvhmH~z_3$ze%K?BmG2544Ic`w zcTylrjtbHuQ?&0@16g<1g1Z*{DO?-K#h(ZA;d9}g;mEK@*g3o?SiUf99iAAL4RgYR z`xo3lkNezQ^B$C3MYh9!kOVm zL7EI)_dGW&71jyQ4*Le>_4@GU@Ud`FP^Q;~>%;HD?*siq_pB105Y`K8gcZW00-Zyr z?HxSBPlT_AABNwBn*)7$f1pp74r_*Igck+6SBy@KO}uTNpbyhnj~&Av;qV|PC$9IN zV62ESBw}NHFTN}PhkwR5Vh5OmihaP|VNZ!qva$F#><;l-cFeWm{-A$XJW5+K23oxQ z_)s6@w7i$ewxb{T75bI=5XLFd6KrPv(3=LaC;b%SPkRP&4}BH81>0T_b_~WFm?ME5 zu0O-?=Z0;A{toGEANUL6CP#;NhEIoY2YnNN2!9U$2vIE~nMIau6vVHL71%i(7Q~6f zl*lV_BylG4>*VmxAP#k9xHa4p{uTZ{=HHY5o}8RBIp-l^Zt(lIaAWvWAakAbk6_ys z;k@wvaAG(j>=}$JAlJxOa&@&}jDS3o!wZKSGalfwa8YodSBJj`_vi2KiMtbZx?wbC zz;-g)n2D!{4T5xSGtNS~#xA@lh~bHE8AIWC*P{nqlOs;nN?(j?sWFR8E;72F7}pO1 zUrX%pli@?*+K{_Q{VM~rdT&-S9gKQey(S%H7V zS9oOLBg_r@kJVBAxcX#3-Wzw$w=iBoI^GZ1DOfk&&^2BkJ{Y{SMsM(M_&WSIb=kN- z@2OoB^xx^b6N~$IFiwv@OYgWhezAJbKQb2Zy5PNT{9|LT9^>z)<#%V`(J<%mCiH+W|M z3dY=s>8>BN3-v@D*gmKa+Q;*PZBGvC1;=e0#6*t@rvx$I-l;APb}3tfxirw1oG>%L97I425}Vqf$E=nv{g_KAx^mayTs*A4peh5;hDS3uqi4U31UWJyU-Tw}(H4Uj;S}ABUY|`NhER z`AB#}V4EBl-W1p^*8e#dcWQdu*fGu!c_+|KM_(ixXSpG1~4_^xxhbscx;-NMTQ z`_QvGGkh(4Kd?8h2tNpxzYI5qe+EALBEj;$sr#n>6|6rptQ*)O>|tyc)F*5jwhrnT z8-(p~UifOz52;U2pW_FE`Y9eP&MYpodSHW?UsKF?gCGv0k5z2P7)dsWb|D5L?%}r> ztC*HpnD)1QFg}Ca*T+FGc>joPVr;i|Os?x6^PaD`Cz&j+WgN8e1;)jRUt+^CWb9r+ zpP0Dd8NpZyVi=1F%<7Y?kH#u)I=Sh{A3l8Y@R5up`zrtRJ-s`e z6}}OE7OoHcX0`*HfsMhQ;Q#PjR}L!$%f|(N((}R|;efDj*fs1J_vUTb4^A8Hg1^)O6;Z=c8 zF3l~%=0X0p56+SP{$Zc+;=nKDgRd7XpAj|&db;7!VU;pGWj$Jb-qvgUf;o)J? z;BT?;sNkGUg0vj3jFq|aXRqBA!~v$ly@Bt^Zo4XcFR%~K4W9~UhmQw+i1grV!>fZO z+m4<)F6g*AgQH^v#&b}_d9vOq7g&(03|67~*!Y3~6H#LSqd)9}ge@eg9R>AlMV_Ep!`t$VbHx7CB^x31hHDWTWa_i2iJEyJ>R|T;ye)}bX&qb$+>5-d!Z@xGg`N8n^ zAZB-BAb(E|;*+lnbl$u`&b~8{%i?`|hh2lXAK%<|dhzH$Zj50Yd8(tj5 zJ+}+u1!4qZb-M+=y=@x=ekDIr%+l{o0)LXO-6*UR)(rIS`oVVTz97&Qj^TsS9pVMz z4B{5__0GX@FAA=^Ul5NKj}XtO9{Rx42Szbl_e{5lZz_w^gEAEdRhG)O_Y3*~wOM`(yOytVMtFBHE>Yi(J{QN?W_kK}PGIlq=$Fwi^|?TA zvV|Q#FX)FnB)mE3AG$V>75aY6^Yr6D$DSXYYx#xo93lFC_)&DOK1B~$|Cl_|$>fVP zFAnVT8v@^uZ?$4rF3bt8C`dIZ%@fAIT`ilBjrEyju7ugx+`#3r{#u)h%gR*x|`oH8&KIOe- zUedauq| zs)No~*VTXbDGg;W9ov<;vTvDtX6lA?)f4^d_KOq9=P^M$&^DBrG8ezlPV7H2D0gLJ zo3deDO(y`>h@aG79>Wjc=c`(9HK;$u0ZIIAa)KJj2a_ZwoSzugkYJ z)}Kssy?+Ji*-lon+g;Cgvd=lvGEU<9;I}+DR^0KLa8rNx$lWKg%IYu!2Oyx5Am zE^q3eXQMon+c$zT?ekFX>VjwQx&AI(9+Z)1z;^X+!N-I1l)JRl6VKW6_Y9@ON7c7) zsm#^08-qHdE=s@k#5QH9-YQdN+`94Oj1hI7GI4M8zjAUvo{N0AhA{%#i1JgO=aM8+kE4K{>IB%$+Ar zWgNy^!|CJk4eGu;>syen@$ScmSA{);F$ikZ_J#uaN{bO{^N`&8nVj|tvuctS7+jb36` zinlvYd(@`YQT0vxXUn!tYOC64WrKFBo~qBVUtQH+jc3x<{ea4qM0Hl7i^&+e+8X9GxQ(;!Xbyye$*8gG_I&x5U? z--0d1zF;2>?FjeFj@zkPKZZTWW)rjESFq#Q*K9tvI$P0LJT^Xmgm1v+H=c!ktv`UhEe+?ei`j>K zr?-aMrP`hQt}X5UpC8;O8(2Ab-s%Q_&2gTUGV$!Y-g5KYhwW6mP`|YWeOz<_IUu&l zF4T7DLG43Zah$fJu4wD(vV5p}uE{oc4Q*u2pq;yhdZ*oK1Jc&c=wIzYzecaMYTk3Q zU!2uFx>xt8E=k+{sq@AVx}J8S&Z*mNht{R(zTAsExRx~Ap51F@2suKA_)X4Ol96l- z_9pqsUZL;kc>0*0BYW95^jO~$q=VQ)bTqk4KN{O){2+b6&o|D`d1N@9PG^ztbWwE# zJ)|#&PV&yOvBJj08%JzRwWa;^TEBxqH#*;vPLm$}VM%AwHDcb)hb6s457Ciyld)Lz zSkt6)?U!b^Nt#x}p76zppZM%lV&Ww(29 zebd&a%OwambE4YSs zz(L&!$=RvFK(Qci5L+AnlJU9H^UTxBIKc7QTc9?nrF(sC{tFMV>^((z;~ z*-h?SvL)opz0pN%0904eSAM$(X}GTZSa*-)J9+Oo_f8l3?cUilbdS71Ugg30?m?L- z1NTNBuxr(Eb-wjhT~;5J3)_^=a*u4)-p9%rAx`J2n&++o$T5M{|=E;7~+r6^`V7)q4nyxKUSvt$kvol7@$t#t z$|o|JjI!Oa;x^=!-~45=ioGC?;XO<~Jl{)QuoTa+pFeC}dg4st8jy~>zIp+r7+0 zs;>DTA747+JYr&S9dUr6Z%^ORf8GaVN70R96m%ooNo+`;hP=pwG^8cnB3>bHuGeMP z9pjq47x%1Ol$GsGtM}GtrEEPL>GXWp_nQv#9K~B2bLsO`7Rsn))AZDb)*W%3J~wd~ zd6O??-MrZ@@2)2nB+b?p^~ZI|6Lq!qL(EbgRA9c|`Y<8``?INS4C7_RapF%gJT3 zOn%*mwn|o#g&-Gd%k=)7M}Ck^!}woq8FIru$FPUU7w1~~&3=&2UPE3SFW+R4yjTA8 z9_R#q5qVSDOvYA5+fGKgN5@){yTxy5Nx!iQ&&)m7?rGdZdG;P13wc&9c(8jXCzUIi z=Xl4_6_v5d*|j|HmbWsuUVYShAz!r{ok!nz&W^K`PW7qd;IFz(x@fxgNfWLi4f|_r zs(WOex~9%mhpLOiy088@$GP$#UC&inxnFg~y?d^nX`gS)&$05T{%TLH>z0-8t>2Du zJ$b9HXxp=|bzge@8VrJEUwL!>usOR=O-o1IQ%BRoHHTx`wOWV6oKAhvveanYyyB~Gi zb=eBNuBG4d()_T2hV}q|hOOc_?NfQVx7G`FuJum+aQ)f!KpTX5=J~XK%XjmvY@I)P zsQKp}Y?B@w=e{jHbH@(l_Ewrr_fv9HhCy(urxpmMSIIlMRdcOU98TIXzU zy|8SZbWC-Gb@g+2?C=@(+1l3j>e$Ma;dzd2`Lv9bPigy9rpiPehQHS32YXh-c1vfI zscaj18`djhE7#hdJriZ74And3tZvmdRhQVDwL5DEtBa0vyz`vz8M+5~RK~7Fp15}N zBrnYy+X(ds9Oqt?J=q`~W!7?02F12v8P0wVeHMMbEqnDs+0t{pXLU_``M=@;>~C#L zTT=#(RYuCd`Rb&0CT;oh+}(%!nX$5F;;fzzS|&3ZaOS`5%B#;v9daMibHBqjGpZE% z*S6ysx+gN^zsa+qF4vCe&FXRYVBPlrCQsEB`A}b#r@Xn(>dxwr-fx$Erj_&3@Ygy% z%)55%xl{(Xjoa@2hV4NfJm2b(nb!UP|5q0s#1ovO26|5NO_nvk%0!-r&qQ5UPvoUKO8&G7xSz@g`^$`lxa_FSXg9FU zerZY9b9X)0^-TO#mdn?Ft8bMrm3hM@d1HUu7u}|9swtkAG~3Qx%X4a7uv94Nown%SoUhI*M|pIe+RxRo zo`ZT=JHvXPjbl7(&qe;U*|q`YrVQKehj}C4Ag}Tx?~Zkj`|?-4wT;H=U%BX*pP2K0WJbNzMoHmWx0+!u2 zd+C0ir*27KTAs7={BO^r&%%9aPn8kFHrDNqL7$uURC!(*uY7HzuZR1{dgY?dsvl4; z=xlZG4oSIeBFm*iEsD6b)o zN6XH-d-EK%YtO9Lt!$8=%0y)W$MzY^)3DxncG8s>Wke6t$3vamIJqSq-F|`UaK~OZ z?$>uO^2pYxtdXZ)*R|nKp4I8Li`I49l%MnI4P_&Ly}oO)D~jLUCU5#|hWU4{G^C{* zT-P<#H}$7=X;|-UBfDxR)Gi<&)K__PpPr>MalMv@CA+4yZ>fDK?e~yp=S#;kk}vIm zY><~;zw&`xC3CBj=p|I=INrJHs6T0X4({7?ZaW>eRrO8TsAH{%>OX%a$%we%h)sl%A#EJ=cC|_&q#MeJIU`XIPm%`y6H2S=HyuBeg)M^~Yo9vhK8Me*b*T2`@VU~h)u*$4Ic29Tm4$1#kKv#DQx@nx z?Mxk4KP~(0+>>;qC*9gN)#0v<)-5&-8^>4yJzzdDUb3ulm){wmtRPy&%rfX#^6pqzmzSpBW9*X;bwXXJ{bb!UYMp62a@_FQwaz%+K54lh*Xy;0 zb=o%hDDolCo{_)N!^l7Dwzer3Ng6XZUu+dlMt=>mF$4wh!^l%b6>W5vwuuPsyi#`)EI^3Z&^hP?PqM>%Gw zH;3!>zlU~7Z4c#Gsw?XI)rRS@wo60a-HUwp-g-|lRRnL>ZbZPrP#0%JF zo@r(Cf0KJ;UTp;H?#=aD7M6aiGt!c#wCQ0wdiK~bzm17A2F{p2V;qg^Ge&)gF>M{! zX3TrXt2^F$$INxiC>y}_+?)G!zx0**DsS#j+4MPjF3Pa7yLDnHzlSnh{V1MWpAG*> z-LxD&Pv;hCdajiZu4TRWtBdJH z*Hk)=j0~>)?XvMsGE#bNgLIF39oEO*!-Fl|bL+hFa;|(!r)`|PA{WTuq3=cq(MR%Y zOxA3SA2SYZI3~?8?yF+}XEfY@zME@%k#@(7^_(8xYfEo`=$pEOgFLW|0-z zIk~S5uNJf?V>-20_O&)^td%jO#;R(E+NO5ealjq#ZA`Oqz%W+XSY+dkhwqDfN633g z#%X)M%2@6bgZIaaca)6j_HK^%-HrYBzLIy?Vcj)@}2C53GCl-h1fY z?>2^9`Y?WcNVj*NyvUb4dOxc8J*4kW8H4T}aqn^+8r-k(<;GPT(+*>@jjc8wd#|8; z`mUd`^u|*gS8d$2G1jo&-yMIx#X?(*PtMPH3TYV(YvT~AsCmYFdvZg)* zdE~FUNZXyQ$Jt8it2T@4f^B4vdh0iPPFvD0v#mJA*koysP6q9`AK}&%=9~-uv*bjrVN43*%iG@4>)(G2VUgUZ{6xyh{S> zaJ+ry)OH-K+wM5;+IYvt(r?>)oa=Z$XNY~Xuhnzp#d|J&Uvlv8}IYa~;J(%h=010?<`(mg(KCZGQD)|N@jQ+R-iPvT zulK*b8|EGUzV}V0s^i*$*aCT`-Fff9J0RqSc}UDnVQw+s0hl{6cl7Opjq7T-1!kxkQKh1M!*>!V!n$y#Krsf@kb;sQvZVuOn z-v!?a{aN^N_+c<7W5VI#z_3TyIqVRg z6`m0`3{MDahsTA-gw?|;VZ~q`%O%6&Vc`&yW#ayY?jOw|W3J0PgZW{UpE(-M^0)Ax+b>*D=P?{<10^parS7iDk$5ATP1AM}hs{+mmLTs98}J<@l5yg%)IMY6W< zPt!r3t9Mhpvr2|qlCR!Z`ek$Ubt9%<~?c_X-aS4Nm<-NVlgSOLm`@E}X&Ls21YiExQ=JeC{%&Fx2 z3(Ev;UK=GJrt^1yxHsGr{t^Bb?g+PrKZWbVHR0;;i*RN5QMf$(AY3-)d-)F7cf)tW zrL&knJ%98))URhQO!E+$2kHyq{9rC9^Lw8e%#r+&@c!`5@YZljI3XMx%(?uU@ak|_ z@EwF#hW*1{;pM@+hc5^_glC6sf;kem2=^L`s{3DaSP%n7=2aBuGj-wd9Q=j%Pp(*wP&PP0?IOKtwAtumL~JA%3D z9+mkIUmiXjeir7=nLC=leUHoycuF`cd_7zqt_(j3-wzjsbAoy5j}Hfi7lqBjI$`Nx z9>m`U-w`__yfGXR_6@H2(lNVD?l!sCm_0H#t$i;F&kx&=zt{Zi%rm<~*fG3#{N3lB zGsmrMn}&76vLREFjbyR!U6Rq?3S>6fO-{P+W5fPo$MN^N)}2^)G~e~InOAbD@UXB% zAomsxlfm3pWFA>(t}AnCnSaXi`fy_~f42Q(;vK<&*n=I`1}p=95RnB$j{+JdDq6f+jK7-%&sHXysOK;GY5xv zvdv-RU2n&G56HRvGkz-h$=BhBi_P#=)Kzj>T!9=WgZcFARd#4)E!)$3KkQHMH?cX{ zoAN{MlDYC>4ioPVdr#c^QS4K)+I>h*Ops4JM4tGy?w!Buk3a0Uee7e`XLEQ@olIxD zdXJfX!Y*X*vGKh7%!XSxkhAPGGM0^GN#>Hhs|7NdjptZ$+LAqFzB}K>Ts1s8^t+y8F?*n;!MIdj#5zMW3{#d4dcKWj;xq5cyZ>((naOQsXJsGm(P2r?) zd@zqa`TF|cw`JumS^KJRKp=;93oi>V2`?S9>-4UpIq&zJ-g9(}rQ@)Ba2^?bXduJM zbn;z(bzmLE1_OHxSsVwde%Hx)BWB5b(P59OLn{;%P zviVy0d^jVxf7gF)I4H>9PGS47Wgu_OVYqfM$K{IQk%3NFats}@T%b1|7tG^Kzw8+H z33Sw3!Y9H-!F~KYuXLW(#DVyCd`fNIycbJnK5X`>?_IG&?+NQ=ZtwlVN5U21 z?_q^GD~!Iy@q*0zx=%PP92?9b`{wYrF>jgkmXVx4X3jCAxnEzLIb6x`$A*W7e`N0a zYr+r0m&2LCeBj501H(>Xvq1i?5Ecvh6h$oFyGL^P#>{(vO)%%~PXask%Yi)pV0cS- zeK;b#GVB}JsxJ@h)#rt6!ZX5?!aCzQ+u5mW1a>R=y`ZLH`=Q>Ir6bHKMMuyN^8z`^FYi1N z{9^X)aLx(-AK9l(@N>yXz9PT3^MCq=B^#Hm%kC{?F8Gbolr}qhufV@=JiyX#wzfGt zI#0m_gmpB;le6Rk;?C#${F!1A>~b>3_bb?l>_W1P+#?6YAx;S&3ZDzUpYV%tQxHd(GdXAUWsj9I z5B1}M?|?idJTq(&whCK^?E-t6&CIrbVPH$YEYt?4N67V8goA>m-}?vq*zWs`**o8h z*ekHvoiCkT0=t}EkiKQFv1h)OazHpF92t%Y^vLVS9GmY=yl#wr)@>_(+y2_|T-~qA zH-6=5_u!h(3)=+#fcsYtYlh{+LxS&SC^PXDaTGe}y+N$x(6Cz&A9-eYe0X%=2eAFw z_2!#0*RA-4XJ&pX^I);R$$he3{F^*xbBjNiN5q^CV(aE^pwIPbtvHS^=))4f_s0$v z>t~Dd+sIqCYU2vt?KjVjd2rM(@zaxn`ey!|&LK1t6+LtmQxNMEOC@u~RUON|5OXDu z*~aFx?3{?^6CktM?B>4{*A?Gozq7a5z3h1J*TXtl&klF2xUcw!xp+8RH?~q}1lvpL1#^U3eNyJ-ds4Qu+fNR@<4)(iEwHEA zwjT`iQDykK)8~$K7Tv{1Aj{>0JpWk`*C8+EmrP|-iwC(M@nZ5_yvTjhjcDAMOy;BT zdB}gVk}g+9Vo&^?#(aGzMSP0B755dlGS}4X*p}zvxp+RFndj#@^0z!!Hbr&3dgb}6 zAKHnz!_-4_w`xOY2eyg&s?MwL>bW+@u3;1Dmt%+P4`ws*MVAQdDE({NF}tdL@oSF9 zvaz=cU$${ZKCR>Vx{nC_?Dkue0s3h43G+4e_lY_3Mfu?R1N8~>#ld#-+jenDvO_+} z5$Q-z?2^6b8uU;5_GkAC>QiDTij|8Ah^4Y4$uaVbTw}NDXETSgd7Rn5V!G^P@d+~0 zHzwKB?CoX3N`a03q_9bNc3{`<8QA+r1@c$i+BZ6Ux8oDxli}PT&Mn^kjlka#&!PV; z$?&fS`+Xltoag-T`9M!PUyS_I;Zs4Z{Nuqj$bGS(H-(b|A4I(THGwZe=N=N|QT%+r zuvgeM>@xmV1K-5=8nzCb1wP7C!Ulo<5Knji%3-mfT$GR45MA$^RpvA_pE>`ceucRV z=@)TvI_9)M&*&4SYs`JkC*TXvIeZ7t$NZw6mH0UOLd?e;j$%EH@#sq&#=q6qRfF*r z#(U^LR7X0_MeIX;QV+EQ^-dgU*cV6-n3HCH(2m4ev@P?jig{~y?0#*L9M+e&doYI` zdCLwYTlI61rDOvcs(*-$)o+^*Whhzn=fI{TN7-v+DXjDL*?RUn?zX^=W0(Fjhz~SQ zVBI!0-VnzQe-F=dE;-01yf288vpwYz{o^Yz>W2Mp^Z~iah9o!H$8?2Q5IIRkDg$;U znMl`=hjb8Kv|UhcWS=<=_XzYF{ib}`0iExbo>ca9uKou4m+tpW^g*y8KsSrsiI<6I ziRZBmW=%0z|6^D1`^4VZ0%rtE@j3F>KBy<`1ajH++D{_w_LH~= z_fmb&4k+C=v;pXTWiZ4x4DAEI)k}5=J45-gMU*A`Lw}62)z8a+RMz@sl)HWzI*6}A zFKrRjWAVPBeZWsquRTxt4q|(rEgy69z~*3+uy@?6a@2=po4equOIYVvM1_8kk{;y_EFORYz20KaYpvBR_#^O5rN z`SNT!bza@q9>_UyaWQRnA~{GNl8fw0GSavKSnn7EdV#!jp7Yt>Y)CqUtYkN8kFKe0 zf}Zi49ObWTqw5Fx5ND!uq{)VLJu;gtC+Fz_`h(v2OQ2K46zCZGMl3<>;EJH1^V>nZ znqB?HATD%45Pt{zUR?XL;nP7M=O=@B_gTTxHv7*D`a|`LN>3lDxbYC%?f=p&9&B5W zX?oIO`|CgCf6Nc!>|*2kYxx6W?&9U*B<@qJhi^d74&x+rsWQ`l>{;;z_yGJN_P;Xr zy!2(%5Af`Uet>7<*^r~2)6fT?|HSq@OYtGTfVxF~(uqYZ2IMpPyW7N>#GBO3LZ6c9 z>=N-nF++ZZI?T7A1Ly;>OL~G|%`c@x=pu5Pj-|8s8Sf8tojxgYod2xfiLGv2D4WaN zj@tU3fgQ)bCcoL%!|&>dVUyXL2ln$p;rKvikzZsPx%S;a){%K+A9-!;$$i0qu!Y0o zVaf1_uw2jwOopxz)(-j;HVkAsdAnI4d$D=gBy1e)cU)!hWAgpmRfF`F2@em81|6EN z-#&pqPW*9{Klzkj`Ii6F!~23h0r$ohKRD1$%le{v|(?-^PDaXZW9FJ!c}O<;Yj$Ck|!) z^%-lbo{{%te8;g`LLaO#t&QFAdC7Qj9*8%otL%Emvpwh#HoraszA+u)y!N}Af08aZ zGtdX@x(@~R9Q#naV0Rhkq7AX7*gM)7Tgp1ys{T70%&}})_L+0pfBbklhRsUfuoY{Q zYIF59v`g(&n%Xy6!8Q}K5qBVm=o&hPjZSXe5yXqdJM_tkfBZ0z=VT)pONNulVl$_Q z)53>>*ti&r7>#i$;xu9~^Md%yaRbEHPYCuo#<5~F?+)TK9}9GV>(CqeC+QP1Se!__ zNIzy}uRPO9jZ@QCWH9~3*BJU4VlwHpVxz`;31hZx+t@xz)p_fh-gEcz0l9>l4*K>YL|ebv{EoF?VwA zs4w-Q`PL*k%C;r1jd3AgR|;e{tlP#OC)bw<`eE6^meO#({p@7NLR{Ro%4z!H-idoh z;}Yo&;~K9Yk4rRGh`cr)@sjYhAU5-va7Oq@AiL?KjzJVxIUyKtb98XeM+UaFc>Ak@ z_zGPH>$dYlUK`lqCx3Z)9r-ru$w!3~3*L+=&j&%13bcb<|d>-*A zx`K|--y$Z%em90u->-2!?0CLY$1obdNI!JkBLB+RMc=B>2P{_O`xE-91wq z(-r37-ad%0k@?~-#@UFqkm+J9#+Il{WV2X`xFnxhThSJ@G4hb@LAJ54`KDy4y3bZo z|3RK=6OJQ4$xQZ$IKOn+BaQD{vg>`fmmD`wfkpZq4bwJ)|bn_FAb zezYaa%9O@)wI%J33`S!a(O%h)Y)Z0QY(*@__>ZCNzB1zw z$xC)DJDknSj{Zy_TPwH4*2Tri-cy6IFXXrJ8RYUCgLwN%L2SMFZM*%BCBL0Z2S`hr zbOPNljJLD<}X{X?B|clreQM|3FN197nji;WSK!+6+me7cyK z7@Am`n3{CN+@$aM(s8aUkMibTl#Q}b27D7J2ix4Y3$F?KTE#oWI@sH6@Rb93u78*PY}|Oqmj89)KZtXX*Yt<}O)}fK^xp<@yW`2P zn7CpT10lb~)5So<$IlGpcVpt@ExAknlEdUExy%+93ulk-9vTOK{>1Yqo*Rry65ro$ zjCK8~_VZa@8l0yb#a5&>Bl!=ScttoM82=;At}oW{I|XTrQ|Xr#TYqe*{5EdKm>pxz zJ*)Y__alw(7jMzOE?!O+v)>O5VrAaTVXNz-6(iT*@ti4WK*8n!Z7{@u`V;cNymP_mUJM82|-;kP(S5fgPBA5uJ> zY!-WAqf7U?z?Yl~#!oI4Drwss@Cxv^<>5}7QXL%xf<@YgHf8-I}|pPe1HS#T^{lb(>CHpx~cLyXxVXUHw` zO+1B+6o(Lx5StLAV2@uM$WivOn8=q0@ek{^+3#5AIUh}H_?wMiSxlC=HdzYt#5TTw z{8*ARjv;I1l`N9)%2F~Ce+;&fo5k;98yQ8$N`su19;`QA=_ngM%HIRoSQ}qof>?{V zndc;qCf=ei!P0Nr8&~5G@C%0T0FjO8pR_BB$z%BK&v9gU@!Ro^kuQ3{vy@N$-Ns7s zO~`Qh*B9en^eMZqYlHiA&&spqt&EksXF#9PUygO2wCH{HM!e40FmZPAJ$5=jQG4OD z>Sxuy$pPc{$a8&>;tS*-So_pHFOc6ig+=BpGG~qOw6Jy9K5QE{4NnX!g+;?% z`S$$p!)1Y-C*w~I#Q+L#7o3TT+cX3vRj;ntTv|Sr(?WZB!)wd-xzKQw+3?l zFTuKPWWJ?s-X)?Rj0s}Li_^opaiZccuIG4qLLBD0Ag*ueTE7oh2l0OIAc>h*U+8Bw zc2i%grMQj$SADUL5zFbjN{zE$Fmb`CFW381d;|8mepY>}^vFwsn2zTmwywXFT`s2X zeWG84?}RS}ea`wqy&wAapbxY$bbX)3x;KVSrjzC5IG>CxFI(oDGx|-Q9>ie8UifbM zR4O~!&+A&9AnkFcAKIaklfbuv}_!&2Xk@n7QPwY9&K7)#?S zV(#>Vv6puQeg1Tb_uKgf`e?=djUAF#zwP!rUmnF?8qaxG#@x6L+rIvQF*PTT;R8UP#CYWIm|%RT{JW># zqvfl^(ZTV23%&yXk>ym|HBF31A zONdE$AJbBdqGMddx{Yx$HiZs?82EX?I2Q4h`9ZveEcR|FJwWc0;odd1B=>I)L%W|0 zr#r-y=n>=YFAK)ckln7Se^*|_+D{DPCE_z|{lml1-e<3iS+V1{3Sua01^vimwSHyq zIvLXxK2NgxkRWEx?)D7HH*&b&*V3omI5<1IGI0`N$DSmY$SZQobzIMNU7tK+S2ji<)-vp$YRbp2}Lhi=U*gduY@1%cMb9 zOOGuqZTdi&DjQ|3Y|wJAUXU01epr_B=vk7_o~Lp|b%TA?4^7W?T+j3Ky!aLJB){&1 zJm&+F=jt7Qlf9x&8&hy~xHo7g>^1Xih^2amTp!-kgZO|k8kYt#Z|TXUN8><@4|#QX zM>sopM@hWpitxk09{*-IFMKq-As9FA{ZixO9~-2l@I+L^PRMa~EE#MpA)O-TOrDeF zVk7KWvQzxod!^(z`5#_-aU{$3p{_W&;{W68 zKEURxs_uc4Mi4{wVr&$Ey$6Pg-9r z!2R+E+8y7^AH)pDNq;1-iqFHV*BdFmh%@%L?R^oOjtb~<9u3ch-=e-slUtjLK;O)o zR6cq6EfcezpWotqxps>j3(tQy*fXHXc`aK9n+N5$@LF&n{#!Gk$$2dWk3Eko+ih0P zRZ(gUzy8_<0(DZc0B`a;@H#I3TX07(v0I09S*}IGN1-KnEIb$WDR5uCvb+}C;@t^> z&%p0bck5N(iO<9H5qo^Maf`bb=|I1$;fhE6ef8S6l0}F$4+M0Ko&df+KF72A5xyGW zSA4*C;fu)Yr+2^nZ~sg63(D*92Y$f&Vt_gxZxioV4>jKA%T(*pK26@Dy!^HzUCT<5)Rvc+qii4CGKa84u!U$J#k{&fM3(PMy#4)6^e8zTtOR$L94q*YZX0^;zPo^DLLJPxe#3 zXRPw_czbH3#hW7W zu)b%v0M}a&bX0Iea6s^_!0W38`d4`7IG3-(zo4gaIK6A#EUmk5z<(YU@R|A38wDE% zZZ{9`pl#GF)iTu|+g=+Sds~2W>1@6U9|srnR`9>PfM-+JhT*~RXA1b-Z!JBK*URJa z-r{#2jhN#4FusqMuWn4!<9I%cS~5=O=j(H?n(WDgPVVhLFJF$dT|mpP7C83mq4Nd$ zX2jbJ-Cn=Ut24aXt6i&!;(K1qZNY@#Pl4K}^LRnPYghN=&G6h?=%I1#E;)esz_&Rn z(C;K45F^Bg4Fi17Z|8p)FY^a zPnnyh`2I}95uBYi;7vXckl7ZMH z-&ioHo*j9B-{&{W3uX-*1Dc<&B#!xh-jCSgw~IMq5AK(v@Z#xvxe9Lw|Cj&d`0#b` zzF5a^$Ng&u_@D3Y*zj~5t6H}t9?TKU6il08+6-?6u4}{NvDWLc;Gy9D;I817pyI-1 zGhEh-31Y#qfx4?)fHyDxuMzOGtvREc>EzuzhV z`y71WJ^tYdoIz{iXS%fDvHkaY@rm#AU4Dn}!Qpg2ZFqgaH=%>+Yx@>5f7KM}W5?a^q`C30_{(>ZhgIyu-};r@;&1h9anCmR+ZZUXMm+WzY9|$k z5rp&LSUNvd-{wi#52!k_eZtwsf^mfVz1R8RlSAcK;(+ssx5ajPhsY=5l%(?!w^qfG?m2Th9mo zm|udUc`5WPj-<=+9$w_l^Uzlaz8owQ;78oaBcYG!cRXjEKJNvO?HJ%~9FEVI4{(>Y zIUef`w-3ebyQH6lmwiN_FA*>DKGZE~Xq-x8)8P1A-2nfp&&ogSGk6U4O)LU^miW51 z!{1onyt7(&xAe}=>-qEQsdPNfzBOEU z9_-YtE!`ko&ciVt)GB%S#>awzvBDo=8qvxx1(O3jZfw!!G_x8iFI``X`^KKFP_K;d`|Il z={@_>YBu_`KF|Aordv9`@`1`V%1go+yuHx5n!QzJlkyPVdv^bho&nH>U+1@A*P$dj5*{`Yc+VKVe_&6JKL;AU270;-&NB zcZp|?vtx^&Y1pp@&h4gwnnv5hZmk;Bbtt^HLqOZp`T7KDU>Y9xiwpRleis9(_qAex zWAFQ0pGThOJXB5HIj67XSU%JK@yP9?@3b~aPO)`hUyM(^ylOMnxgH(h?b8C>{o~-g z!1$Jfy8ec~1bnXlz`VLOxG#7#FlXei^b(y!AIW{?xoS~7*Byex1NA%Fj`o{5>qqtA z@kb5`^kMTJ_~m*H@fv>~pVF?@VS@Xf6ASp&yb{`9A0o}F_mKa*N8q#RQ+$rw!Pntq%~6RUaRa-K>gUypz3Yo7X2!HU5efqu=60-n+4!6w1_-L+#@&e|qD zI(#QOT|Dv`a|Lq;?t8AhdwQO4U%LL&>Dkm{PtS`1_}@RdjrihNhy#vA#ezxcz2V(E z9%2B$N^RKne)#hGF!v314jh|xf)xYa{Jep4^8O6(_q-aq{Eh&(^X2(7deZSaPo_K= z>re5!7|xG7C~(b4Jik}4N3dHUkK3`kjyaE}dM5Q#tr4i*(((KfwM#Wmdg0~37^B~5 z>bnCzIu17u0jJ}39DhTAr^Nx|RbP#|D<7U$ZcL90tSh3&e-Iqo^~4u{t8LOfdcbIN zz63ABe8iLZnC9h;;6U7s!|^%(C|ykx@>*%@0^J6lFL;b6X=8lEnJ?Aysc@KUC99Zxer?#It5a-tW=v_Qo-Ux2Rx3n-{goanYyfDxc z&*wfP&rL4PoMMO`SQNo zKJa;XHsXN&_PZPxI+?dUDRAuZJzXqrIgj`r*NTJUC@%`{UAKRw- zm&Dx~Be>dUikBeXdMqCI$8s^&d&OMwrme$>J7R`j7XFF#8F<|K4YhXK+Za|0Jw4zX z;&S8roZv@I|F5xc4wwtBuVH@3QOz6kNz4){fmKILLVm{UnOv>&g*%v z^I~u{p2p|&IxRnAFf{N_>?jaPxZC&ge_YdD-PLcEJ2)Qdt@cx$RV?#ct6r>5%+h*Y8fdi}S_<4ZmujzFPe_c%An>UqBo1PxvLTbp8x(zc;|; z^g8a==gt$Pw~f&&13XTL7`HUI-pC7s^8&iXc;_e5GUfuW#d;+^qcxv=7wZgZv|9Tl zr?tL`mZQ`9y;pjTjyLb=G#X8wPIsBBxPnGA=lkJ#Ft-b~ulwe^d>hy3i}blZ-}m4b z-@9s{=iS_=AHjXG$!mC_>apTf_3z*i`&W5@eUm$gU6l{W0owRQyW>N>9AX%+iN92F zj4y;cdEDX@A4ja>Iq)86YB5M}j2I+F;a)L|hHp`Er`6*=N1c$bAzu1!-i5e|qj@X# zxj5AC@w*&PvAMXDe%3GISQd9XMvfX_<(qLBy`D!^}`p6j{dcAf>^9~-Dg@?pdk zHGmre@q!Lk2f(#6#e?CCj|_O|G&!$ml|Y@Bjt8}2>!oP><+|&_@Rw+J@qr)Dx54i? zod%bS_#QmYS66=!vv@JqxblVMGkQ7s>v&yVnr}pxt0h-`l|Q4dT-Orw`|!Pbf*9xb zt9dxCYXm$R$9ut`pY#721wScxB)BKQ`{mK7wYDD3>G``yCk9nZJv7%17SEk;HB}m( zUqi$5Y$^`u!PzkQMzBUuzK#Aw8h*}zHy&G{=Zo|IXbN76F^K1JyjrJm$}^L7vrb$H43<~MLlmj@1rsKwLS}cE`J2i;x*p|?t9K_{C3~#Gvo&R34Sha#F2Ow zzv51uTH3rcx!R-}rCMk88q&r3T`mmtvGB7?qw7 z^_wIj@Ni1@RNS58SW!^Jzd_rUj}e3ipe{VhZ0yERj3%hvbpvx$w-5 zV{874&y96so*t)X%mq0s|B@c3hj^U#2M-6l$jJfD=7YW#;B8t8DVcBj#PGLXG8|5O z;c>N8T&{0!yTF{L^Tq^ndESim=ky_Ne=@L^o$)$*fM@tMs|0GEn+CK!Uc%>m1wG%m zncnBCw7_LRm-A;v2V(+tMcTId+Vw%=LcEA0X@0)U#DKqDK0EIN598!d1+@4V0-Ak| z;2XgP0Z)cMqfgWQ4TJT9uLo-e;(+VVEFIAQ#rJ#{zWb0s9N`7=obbH9_sGE-Nt~sNZ5~r)7?i%3p?SpMPUB6kb8n#hD*W-1Xezia! zjWJ{_=}AE4+xO100Cs`OPO4c5Kt4w%5z8s6HUZI_~(|xxmqQ7pL;A zoj>Q7CdQ-k96B3tYCk;R=0jq?_W@lFKCA9~-ZoI5S(=|uF7GS-ZCqN%OoQ{ut!1d| z(%^1mnQwSzK#%`8puGq3%;m4HiD6FAJU0jC2wuN0c(AKKsgcr0yiRkeI9`tYX25?5 z758#yJq~kq{leC`=y5Yg%~!RwEdn}?W;5Rp4)DBM9KH9WK)uzP8hter0=3r%173mp z9`3guYHFbHsGsknf$N9O6{riU)8Zy;c+?B^$MAe`7Y^GdpzCRNJf~-}`rI$c+9sNP zLU3omx8?=uQPCGOEPXJ%nGpfq{-uCk|7L*8t)Chd>>BJI>>2D8xaOn&&Ygnof-M8Q z&#V7hcTIPE&*!1<iFq75^r6DY*_H_4Da@?i7Y1L`l}_%Rm; z`2FI(62k|7YnfFcC>M8$ax28w0 zNUO8;w(zR;l+eX`OwP~z@LS~=>HuPaSfy?uhVhBYV;2h@b9IvH734$l+RGnvuFAtv zdv^Y+rY&B$U+|b;<~;Ly+!j2R1Mq)Z@7`E3=8QkFgx|7d*VlsU`GNcvJWr<>&pgD_ z`s$+mM&63LD9z4`q}zFwe017f{YmcuzXfpqZ2{glf9?$Mzy5>!gTDp*n10V)4;#+c zXGU-7C)Ve-WI&JcSl0^ZHFKPn@Obcd&1;f1pPn^zQE$c+ECD2e^KVVBaYt3{Cc%l zv8DW+{LKIF=kVplptfh8pQMMVydB&x7vb-yP2hgKuePCo65sQ7)LqqG4acV6US*9XoS-v-zJIM4&nui@428#w3Uy|^#_^W*vOG(lbO zS3iw>u+|gJ5g*=KSoQj;|C!q3c{wHh@kn578n<){|6LCNkA~lPRzUyILFKQv*VLo$ z#Q=UyU0aODJ}M9sX#Rq0bh+<&uUk{vqMmcRLtxvj19Ml-uxX&jL5!i} z9Krk-BX9z)kPF}*Ubs8}H}Sv41o5C^0q=$1jnn8`@j$OPj~m~W2ZR6U@bX~(n9dY2 z0!lZ#?|FP#@VHRh+Med7zo)D5|NURB73Wsmsn{d_xE2A3LoGaKJMZy+pYQwat9_O? z@ZgV1@y+Ob{H`u8ziR88xE|yHYM%vzzy5!FiP%()P}G`7XwX@uIe49J}-okVBJYIh&FrV-{|7KD^-|Ly7?e!STq5U%# z`S0`>|FqVR(D<$yMYEaLdM3>EodZ5C&Bw!`19`jZdw&UT2<{B%`p1LGfxiKY=hX)F z^SOqN-fSGNX1i#>=g~8R-_;m*2yhp@%BOL?M7`nq3#|>+Gh;mwZ$v)|{ft}jyPlb) z1HL-Hg$~EjxEjyz5YXy$Jl$UK7_aXb*v7WBJnhfd5i3>?_&sesc7a?IXWl@cjhd|e z&}&(BSzJGD(6qSs@AvAk_7~r~MkSuFnzP?T&*OA`Njx2X4o(-}^ilG$`16j#>E`R-DK+CH&oe&haSM8LxS38v# z<97XRa%J3Y&e8hC$GDg;Uo|uHQ{HW^%FE>d{2FtazUKv)=d>GbXa3`8x{yzgt8qlj zA#EQHk9b=C;UeA)Po4(FUAzh0hUaK^Y>E&0;p2kx!fEu@3;$mF#>WJFZTgs2Zm-3% zPxqJd_D)~A`+8XR4QxYW+rHKCKEq>r+iSGAZ3p^HpH-;u8R)kEJAFR=epJ`{F3*r} z=)=*E!;9zpTf6Fa|16NF$X8s4K%GF%K9J986MjU)9s2HHW;ds9u@U@`o8*(h? zgZAgKAqM@YVuN!@=hNx(JRT7L#~7i_^(@i`bORo5uMG(yziUlP~zad136ulGls(jST2)fMg0)FF-ncV1jp4@wx`Td~bIL#Eq_hlA-ux)Or+@IhJeRLYAE~#>gR6&`es?V} z+FRclj;Fgf4mRjGo{!3l<+Yl_@@yUry`~4Fzc-Hd!vQ>DS`jDEjr2h&i z9%h_W9kjhBKflEo#N{; zbHJ_Vy{?`r7TnYa&v~u1eR(os1nu6!b2Pmepm3(-5bH#v%tcHlQ=u^Co z->mKARj8lpt5?&UIsJ;I(fQ%H9M2Y))7!K+PR7rCmqUXigE7I;!S{k=g0X>XDI69Y z9JJgn4)AGwPHB36%{GDWP@ml-pzHO|uN&BhuLY|H`gQCdugL!Lj%fJ0&W@fRe9mw8 zTY2pCxmxqSUBA2DnGFKmEw@=V_-r69^0sCRW(;^R)i)_;;=wo{a;gi1T8pYrhR3dk zTKb&tB6j0#aooA)tuGX)nHmE;m*Q{!o-sl*=#Qb(jUnUB7&LZ^!zXk-?fSU*j6lPi z1Gt^usXj^i#~i`+<_o?zcYYSARh<)@8JrfJ6qsk`8(n1{nwxmP?Ta*z&1d?Ihx%b0 z&%>qHORM7#K7d|me1cbSJ$}dQZwCJgo)2&nzQT2U2potz@g%;**F0x@Otp9!f`{2IkeHM=!e70Mf*=v;>9Gfngu|eepbUYpY?O^v{*I?&h z#|}H?Z*T9?{GDw&-h24@qnloKpI_hKes}#oub(*+<8Mj;G-rC;DAH5liGg zJRN?HSfH2a>7e*uT*d#T)y3TEwRGO(UZvG>ymK$!^W10xIiNK~_?&i7P>=>x0x@P?9 z&3aa~YPk@-PFy<1;;JN(F^wNnf3FGWZyWI7i zTpEbB5kJfaalza&&rS?R2RPsS#rNj(CIRn;XG6#FSoPbO_k35qZy;x{`)*q=M_xZ7 zz#aOv@jZUw8S@T!$!&eLy!46Td^)<-&0+vlkBir;uZ8w*ds|9hm)9BbJ-z$fIZ8&C9vd=hXv@Z0$`^ahSUJCJ+wZ0HjGk5x}KK7SWn z5#ar+0(@RPj>~Bv!0BQF{x`m%VgsJ1^J)8Ax;5X$_vS=9UmneRCwa5^HaXz!@N>*b zbCXUpU#&x@<7hhl&g!oGS^X9#1>*wV4DGJ>n%AzL>TgR788oElv*R9~aDhLgr)JK8 zmw==6($J{u1v>?5ht@pl8`o>X$FTnMW4U4g|3hz#dM7U3D!{Eg6g~`i%ulgK74O@w z;IZfV>$JHzfVb)U6N9sYp9B{MzYH$vaPi=ad)LWyefl2-X9On%-Y16a6NoE5mybiw ztHTF@I%B+cd6qS+2M78fd2D*wt8S`?iiTf3ST$HBaIRMf)ZCW}%6Fe9-W@$Z zD8S>qc4NrcGUklUKL(cvw74-&kLw?BOW#rH=NO5H*9W{Ju|W>_b=UH5=1%X&62Yp0e#wIa z{qMYKUiBLRo~Mh|P-$ge37+SlxW(VDLqON_%GFud2*d<+Qe2P!@$rEHzvj%~=fM@h z_`o%e9}FG~o(!G}+&&&W9NZV!=B8jma7}P^;5x}Z!*!8=6NnMxf^&l(1pFcKL_NTM z+Bf|;`f$`{`B1oDOuIkeQCS1aud;4eAH6kR)3PCrR(KyD+Xe{b5FDL@{ASwoMxb}<&E;g z$`R>uI-CZlDfl_Y9=Pv08r^oh9eRb9G1iS;v7&|fVBCAnt;fc>@z2+xp?Ep8m%Li9 zvgCIE~Rb8!xd2BwfADG|jvevkW0rWfXmq$M>|L{Ehj#KzGJmiAMe0i_E z6o>~e2k!qP@cPpMUgAabvdg!pYsH9G)0f`&yclw8(E8wf|ANoLdpr(a2R;;2@T53E zU(@9LkwS56)jV;lZ3`aL?>v7TZ2R`TVoKEm#3OtxR{0Hb3Vdu|!2Z{7^t;-a*2X|F zQZB;xSF@(^d18*M<4(`xZ~DD@Y;bq$ukqM`wpU*P_tnbryZDXYYppRqr|rMT^ZY%x zw1V*fwdTTouc`IO4{F`29spW{cW4Y6i^cKf-P1e9u5pahOV|IU=VVUzGCe0xEzWOq z0NafZXnTDO@?P9db3Gf-U*;!IwXWNz7s9;71^D7;0gcZu;2V@@^L%>dc?sgcV}ZJ?{z|Cr^xcRLv_6f`tKU7a7Km@Hmxlf= z|Asc!m#*H5FKOlxf&23X`que7z!z5o6(`h3R}aJqUdtiDDZvH7Rl%(RPJcS^_euU0 zyc)a~xb^tOz-!M3&jz-g983x-E<7~&p~3eAw{+LW|9$?($wh&D;G}>Tv~QrlhsQ4# z>GQOIJSqEKK999ku4AUBQtZ=Tg!2y#tRdJq5F=@PF;hOpr+3_WXO6SH%UTBglCDL; zv)4Dt(`ox`)LU@A{+cZVn%{YzFA(dUe`5(>)AGgB*G11)?&-drpJ^uCt4F4M7W2!z(<6hAdG2y>oUO-()&loEXa0G=`N*H4*KoJq zIa*G?jQLOR;f{B@Ix1eq$&&+|taiv7#yR*@{~x}^skaCC7nk8<{DoWbBVMIhPYdW@ z`uK=|WL`8*)~IAW_<%+q&n#S3-a z*8;b7U%kZ`;0vk0R6d|4BW|eMh!N_#azDLUxF6T+?V|6ET^a+=^Yi7K#`X^aTBP-N z>b{P@&9lh?F2>{Vbb#0CC0a_Ywfs%mt2iHD<8(Z)uEt-TD=?QA z2>7pycGuokhto%a*X8c;$-raVE*9vwSS@gk6a1jJl4n3qULWx4`SWV9H2ts+_{d}R zLmL0Bz_l!^Rx2M^AoxBKZ$$@!bKFD!#JT26& zbEUt{A)MX%>-1J>Ew>(v2mBh_;eYSL_dcU^yw5Q&y+^+dFUPzs-lzTWKG8wHo9k&s z>-FJz-kj$F(2@nu+5Y`_#(Mf_cOJ9YKx2yo^et|pd2trdv3#4->#Y}oqv%+5NBt>c z4E~~VeI9-6eLmCY_}muw5HI>3{X7*LN~5>_xp)L(MsX~jZBcq&Ou^N-`|JRZL;3mQ z1Mc>D;*i*6zuMnfd5Lua@)5tSfS;`y@VuA`9(&#K6muOHy!~R}c;jUHUd)zzIhSHN zF6PbSWBz>AUU+%6?%8w3foHN5Qu9`H`VF2Sw= zt#3^Y{^tp(;qv(PVbYFzwbcmKa>oa>r#yh)jQjiJ0NIwl8rpMJ} zj}G|zG`qau#^Bzd^?=;JKaex#)X4Qi@!TA9 zy|B*9T><}}AH@IJComrL>Cgu1vg$j=sv2yo`zJ)-Umei$#=JSfw=`GGnd+CMhj2aL zlQuWU^hL^*`6)bfb6LKww*tS)=RsZcgTTDU*)+Rt<>~kwtbNoAZ@mnz#?dq!JxA-& ze86*gCBP+KtM8!sd{3+U!~*+*^ZCGhgzEL<71$5^D@NGXXZUP>MD<_ta&WpFf(EUg zlG=wivOdVZ(4ROC*NF#skdCEic_sLt|3UZSdK{16%fF|ai>q-oO^>JXHlC)p@i~;< zw!PPBbh*js;E+J?&3+yB?|Sul|8zcmZ(F&5oW}Rm_nqEz>!<%;-!FF1{Ps^AlCy{r zVi_N5d>~%>jp7_{$8Q!N{eCgku@DO!Kj+VJc3$Xl=Rs}O7!U*Mx-@b;*Zt+=h!4hx z?OfLgMg;mZy-pX11@c2NKu&3V(h=eSO<$0c8o$-&OLy=HJ@)*lKpoh4R|BSb=zj4+ zY$&bIr@S(to33h5JvrAVPrg3To1ow3fk51#>uG;|5OQ!a1l%__UkKcKj)z0D@pE1a z%wM^Ct*0W$_2q~?G@kr`k2_1CmOEQ8doX7pwunb!kURk|@P6C4qAn<===reTwql8R z0^-XiftX`EwF0+li+TjbmIDIbv+FL>r@RVYl3vfh1$-m-)hzW0i7&400uzGU0<~o^ zg!bisuMqG@_6Uv$aHVxhJQQnDU31L3RB=RIb$sBu1$u=hcWZL_@E^%KDY?ahf&O~= zL|t=1PN9yhe}|`|zenE>zCSKFK5*;t*a3P$)g9ym`ii_RKByO;9Pon6nv9DobNjshOQ5k3oV@OUt=(X3Jx!9K?RwXK~Itu&U*Wh4LS@=Tic{@6mzZYdgI~>VV$MC&oFrWSgM) zr|N}nago<>4xX~kz_F1Jl~3jv*?+&owIlGN8a3~tVr2E0;!|-*3=)@sFXFk1k7A|g z#8AH2Q-RnfPq`z&+jtzO7jXN;pm^OH1o?pN@jAUP2M{0Io=fpsErr$>$8mjIkEQ*c zU+3CyagLpTyeD%K4>E$%n7Ys%Q@(uY(`A6c0E4L*yo2 zhyEPpbGd$&{t*yYmJH-BVvX;xA95qH$?vIS;Wyc5-{Ujw-*N#y5{Kj~`v?BUhS@XqyuxiO-p4i7<`gP#@U6ak)2-jxdHQ8VNR({{Yf!Lr< zuOHXnE9CvDDToW_1w28=&v6tZ_)E@#oN-j(ysRC3IWU%t72{BSil=#Vzz;RAc&G9R za}h_HuQ(Ui$=m5RIer1}$@OVEF@i3{eYjsNpc`=ttxtQ?lYlv@!0dWQ%ewM@Gl;&n1Pqo z7u?Pqcuh=cW0Cry*YI`a7d&QphQ~A`KKIYQ=o_>TctI_Ljs*8T=e2g zpGI$#nyVTtJ$-Y)Cv$x{^%HB9hYlLryY7S>h9<|o);3t1ykoFuV0|+Et`01Qz~Rkx zM8~Fv^8J9$*C)y=!tvsZJjZn$#0Yu8GrQ506KhT6nX3h~f!vj5p+}6r(iCEl^DkGHH=8HM zo1D7locUOD%=}WPQx7qx^6f5;o$_e;fK+n77-+P{S^IRat${}ccc}972p67i{3LX#s7RV9!K=KE*^!w@w_KB^s0;S&UjL49~TA}1mlA91AbNOm3izn+p6pP zEcF(CobQs`=n1gj_TTT2C;6>31`Q$Ks_XMR=9RZp&k>DMbqwcG4kve^6-pOUu&c=ns z*Z3Zn`59&UTiB6^Q@iE>}4%Tep zey_u+w`=GcaDeOEx!^YmJ2=?$VcR0auP99y<07bXZ3Wz4|E=H44hYguj`UPy~DY0 zkaVNt-%8UuExc@ARY{!2R@L;^_<*E9KdPnBkD3@g&4z=7jN8xo*B118XjEhy~_>C zw{Jar-kiAQ^L%!juT{*DpW#=%w4QV8vHi!r;)&x4?ia)%s64N7KCk0#aSMmLogCQC zXFCq|!|$|jye4@JFA`U)9f-xeIJrpWdDWjFC&c4oyL?0cgRvn8le@_+#;i9xq&fe`_dLZ!lxgpNbOO=02&tRtKF!GVgW88XN=zpFDqsPQ4xrw7u_;L-|g5&m&z9T2$R#J%kqm>Lc(pLeO(S=O;$mCgTsQu1G&YRfOjSC z9UtH|xrTTr*5SH}b>bb+$v+8PV@>a_x;aNGwTHG3!ew< zljrP@nzHwZ#nu{#!?i#9kJ^24-*e)!?QDN?;4}O#{vEhqupb56`R%~tgL?1Sczf~z zy$at6^gGGj9P^C>yiZ5a6LbZQL2JP3f%7LW=%M&hV7-g@Ade_-QZ6BnEnP%688`As z`Lcf>B+oXkj4xwazHe-rGq(oj6tBP>(-+8Fum)62!2#yJ9u)qwIXWg#YvAkfk#NhX zfX^XU;F0jg_$KlK{5w#+g!ZTP`Rh2>eL7W~Dh@9`r(JO~xbJnGUmP#*z~MBl@5Jr$ z619l7w!%}wm;4!d4sPVbr~~n%{5HQ4UllLv6~>FWP>x0C%DH$ed_#VUd5sJ3gJUgj zJ125@=Zoh^XY!i$G&-N=sB^t;pbyR*G)9aS^G>g$aco?gFRokV?@7>$yic`C^+-BH zeNvrLy^?mM8)*@GQ7%iL(1`Tkp!}m-tkd~Oz*C+r(5s=Bo3_!@#|IcO06gz?H9Oa8 z@t#iv9}i{?eCCXS?@$wko@t!_{fo@+^@CJzroK#CR==SB5c|)Y@SEh>dPVr*uBCZl zP+Wm83dJXNOIx>Gf=A3_yoJZ`8-JKz;&{v1dBuE{odV}V4o{!q$E^eN)_F2_)s)R$ zbGGK_^103g?ln)HYq^D7gO)Q#akzQu08C2*hj~fc8818~xsW+(43+mJ&a^&&xhr;< zziRYi-<<(pk0$`Pb#UMF;-tCG^YcE}Fc7oENVh)Eb1gjQb)Y$YZ~21uN&Z%`(LR@N z$H(BI`AzaVzpp$pzqv<9`Oh&hM{raNaZikEeLG%><5zhg&Qq6>8w19J9yL5}ob>monZIBx@#Ksv z*Cpc3fw2Z+tg&aEo3ZG+Q|7!_0Um?bz;vV$NEV{wCry<-HJeCJLR(P!9nfSFz;JDVjaJ-!pT)1FUU#q#{+{Xg# z`Ju*;9ui~Bc*Cv6r!j2I8{_&_%KyU2;uVcraY~JW-sNG4QECff6n)IYmm7&w`oZXG ze!n;(wuniU1ISHic3S=G7|?iP!2N;z$UMU7 zd^mbgyb*i&&^Vcf#Ib7I<|>ZFje4hW0S?41as?blf780OFa1wTw|a&DQT`Twg(1>1V;p;gQElW zkb=LhR`=anYiNB1{|WeIY7{&(^>FohzL_=G{3`X73xkV-i-Su7H4e27-WH$i@}T<5 z_+EbwcwpBA{Il@^U+ntehCrQOjRo}h7Wi!HHEJ;YG<6$r?Wel~&+CDxT_)5V0)Z=X|8B=ON#uvB`iv$aFz0B%Evj?;?@9(35@oL=a zZ{`E?T=hBgXX#mG3z^gJJ}Ka9aLVzXjh;?B+M>^%b|(X?T%hoSF>3(ppGs3C9GZgTn*ybML_Q1Gfn_ z4K@hA8LSF+SF-N-w)mm-VWXjUJL#e{3Cci zc(%hcQ=geSIe0R7Ja{yCWB@!g^`WT`2KNPb1-At^1vdl}g6o361b+&y41OQ{uEFI~ zFQ59`;Ia9QvS=bNtJbMuS! z(ij)U2QPvi(Ep*QMqdXVuxhi;OOK@*m>wG*fw5P;HsZgyFXr<;^lbqDK;Mh8E4K49 zjcGoDn9a{L=F3AcFZlN6j=4m~ir40vyv_U*ujOlUG;>rQ*ydj5v)oEfDM#W3n*Ws> zsUygR#BV&Ya)2{vbNnHm)9y6B92dvP2gG$b3+|C~$d`F*@@KvmjuP+X()=xa#h3HC z=Xr6J!<07QkMY%jPl)UI8oa;@0)I=VAkVkH^LGL5aAklm^)1}c{S6I0VAk=gwWyQP z6F3+D;$hwbO%Vy%yY4!!E`Nijm?xksXbRlFY#NV5s__njr9p)Cg&^=Yq+BxW$j*BRRfqJqEA0&%58k zHqMLl<2>n~p!+-r?u&_Tc~!ReexK{UYsYwA@1ytfgY-hGks1TQ_qsl~HsBE%C*VH4 zsejV#RYARG9C-|NO#k>&0I@qz=R5lw05`z9;U&V{I>*_}i|TZF3&=MR6G)sx!);@TOc! z4}rW&-=5lI<<)XF94x=)Kfr-SX<%P$SoSWgb< z7P&ruLH;jxZW(+nST0yJm^YX`m?an*&?9sR{UKh{7!L&Z2W~-JzdjKAe-r#NxG1;MB^I)3}?r+jzi>X`mUfXo)rc*Z#Jil3B zn~j3CgH?i+0=H`go?9*WQm}NeOt5mWZs2`e1@#?ncMc8+jtNfcaNN}6dXJBudi2y| zgVTZ^2R{$|UOI?oqJ!MhK6eJs1+*M(Hzc5OrUWxio3W<@SDd!uwDp3G0(xgZu#3 z&;V)1{Vk8gNK8E1aAaW0*0__A$&YoAXp|?E7&C1y2F-3 zwj8oeut#uoaDMQo;D!#@4Y_W}^}&O|Yr*V8XCM06VEN$7f!nVHUk^qFhX-c{7X`l! zeji*JToc?Ayd2CjY?fh*1WO0Y1a6lP)(`dx&Io=RTpe5yT-xE0OEONhaKD7r*(&|JM6Q;+`+7Y-!MFwGIYw&=YpF8zxkA4bZ|)U-C(a^t6-&I?%=&4 z?+y7!@b}>5;MKtWyMkW@-wXB+_6-i|;M}YqEFL(Y&hNj1mxC9A2ZKKZrw88+b_})& zwhlH9jGcvo|3t*=^D)+V71sxR48Di8bvzt7JSv-N!7=$isOeRE(A|1L16#|Ai= zKf8M%r*SK%l<+}sCb8+ya;0FP&IwJUXfIC+Y@ZaLWLczSj zM}xOMc`!~HmF}N=9+Ree;!GpoWfzSC{;IZ3B1K;7hUJ70d-U{9grUd^9Or%+Zd4dta z!olLflEJdUSAw;IZ*=(jl<8P+%6e0_2u20F2YUo=VejCe;OOAw;LPBM!HK1kVKj4E`Ow8N3?^F2g?<{=poh=1V z~3BQIMGUJX{wuVSB^S3QqDSEEyRqvdIP@lZZ+ zO~9+WCy=K+AIL4`QR?4vMLC?<$d5cA;OFvb`L8@w*M!j5E+&eH^9P>|z7nh->=5i9 z921C#7X-fz{v6y8+z~twJQh3?ycE10Fp`Gk-|%3TK%AT__*5`oFmK@gCjtREQI-e=liFq#s&jh~n zzTnn?&bTVLG`KMMVQ^9~Iyf-+cHsAJ8~FW>!y18O^o782bNrnL$I*Fm>}L)9L!9pf zZw9XiuLj~Lz2sVe&vkIWh4^Z_R|4-7d+Ytik=LIMjH{=Er-DarQwhyqKi}v7Ki#OW+^>f^Wg|F*kV?yax3p^H$z3kJksHkL-j%o^Gy}4%i|! zZaMjy0d1^4t=@$%^la!iPz$3E@X5;oe$gB8a6l{Y>u3c&9#8MQfM($RjSlqA%b{uY z-2*u@Ex}XdFY+1b3cfZ?Pha3YIU^2~Q_7p^fBjK%Po|a}R{d0di#z`(cqrg;-xu&g z@#-A`J#uY;Yt^dxroRa;4%Bt)Oztho?T=%bUarzVtQ0Re`)pPI+D+?}BrKvx3tDT1QUi{u#lU z9nPM5c25hPKlS|H`^6yn8K3$O0iT-p&6~a_@cmB)Ja+pe@1?QiZ!}kV(o0NRV%jpn z3c*){wE}tE7Qv{1ZlmSo+3NX@mzZ{Ja6)imcTI7%fBBt00LS<2u4mdUc&^R;&d9Y* zY%4Yv>alZkeBe9kdwf24{+M8N2m8V&w;$dI?iakrbEh^v06cXykaIeC3?A3>wa@m= z@2bzWKlH77$-#l+BF?Lu>>BX#Hw)BFz8$ zTc~l|9;k1O57Z!xA?IA(;@5$(bV2a5z&J8ajHQCd#*ndOTp3@d2B&mz-*aB~9v+7= z>GQ?~d=LHyE$;h_J7c)&BkC-)y76uNsgK+c*k@xAZVl9F{1#pUj|GfVzp?5l>ON{I z#w|<^oxK>{x7>(~i(*|XP51TeqV4Tkud?GMkKOHO_d?r{r z_)_qdfL>GIGVax4whr`|$XlFWV_q+babR4WALtoTkKs}255Yr^2fR4_9y~Xm9(~9A z;*F`j@C|7^Yh!7=!vp+zZg6q%$KZzG{y-m)x(98iM`6K$zoFMiKdl}n{C-5B*1)s9 zEZ`AayT?}+Gt>z9;bO{S0pFkBzgZv-sW%@JoEfOMtEZ0-?g}0ctW|tF7@U8zb$^3( z@%;VMF9mA^8wJ}3dj{g==-`CF^}WsueigVr?_Yx(194aU6^q4Z@&1Eg=#Zg3jW_#{ z*?W46W~0-V43-V(h}D9%1KMJvVDo@CvO};_uuHI8uzNs{>=zsm&|QZIM+8R&V}h{( zKLy+`cub$k^S?g;Jnwbe(>^})u;9?Z_wWRKC%sob!QMmm9f| zXK-^cKA?ZD3@!^U3Vs$icP9tm502`tp|*Ffr@3q3TyGihExs8T11kn!2xy`OgHHt? z4`vMLqBjHAiF-bHB6uLMX6d@XIw=0exZv!7=XPXZTJf#7pxW-zn=|wcKT-Y z_8Q}^Z>M%sYm#nA-Q?=v(m>yVoJWtb`kuZ*{)k$IK0Y-HnrPvG9#WHt72oR>(lc4OM}V7gHO-(#j?Z!X zFM{&|aYQ^28%_*(FZ|-80v^m^!NI|IgYN`986G%4eAj^QyhE^kAb)mS_j%!bY90=3 z*};AOH}I})vu)sgK5JC4Q?N_0Tj0BVr`+6qw|fTmWuIW*V83AhfCp}W4-M=)93J>R ze%JQ`zmYe7Lcl9;W1<);?{{1(Hj24ouj3{bi-%&fSa@wf^NEAvpSZ`)X;+3J1|EqsMjm;Yj`kLfp{F#kAY?^*UeaLdPtpc@FF`jp` zW3Wr0|5@Et+}9VqU)NWy2jP(5uz()GadN<80(}f}LVXCQ1!pvCmaGZVmtg&X{88;! zZ^EwvyvZN>L-41r&P-cO2yme%Yq&j$L!tjW;(=vrrtR!Qct zK<$3E0EhFkM|6FUdI0nV=vDh-@Z~@cn;cmmfqYs2qrOM|ZhWb2g6*3=$KCU{4fYK5 zEb67(H;{Mpt_}(g4i0Uu>!9~e%o!bw36AdYy{?C``WE#o0x$mL;FLgKua6H->%-~! z+iCS&ZR0)O?{maDpZ)#7clvJoU|;Oh;eq|MzxI8??g^loe% z=mq34tQ&lzTh~nctr>`=Ukz3ctohX=$b)da#aOEcw`OLJu4iM`X3dPA4P!y?iukRDVSS9TVvP*#YW+9QLtLL6@ILfZ z7-PIjH4n8sV~-zlXW*I@dM73Z`VZT_2;-SX{&k@D=I6orfjOWrQJ;>!-jf1#6!XVi zIyA60L7$Md2<}kzNU24#YKm71ldE->qq$oHfZ$1!5nc>Cr$ed@#5_D6h$S0NU;5 zfahebz{G%-6GyKLs(0b)^k0gvG~OS&YmD=z#9jW>Wx=HZpNa+)lhwrw9=FfAZ#&z2 zug_6?yCU%UzRP-c>lk=p_Q(FI*WDV}@4EsytKaoV;J4DG)@eHy{|Fo($4cyfC*Z^R zHz-9)I>2#uyd8J?cJ9FW_;euWb*^e1g8oD65cIO@KU|~v8zXK1;dbe%Pzp)@+_1&<=8F<6D1#@lHehDX^aUy1<;UPC(uB_P|-wSNV#c5A@QO>@*-l`CYam(;5_k7rIZAKP8MFo7(c`#oKv$^wZqluPrYE-U)^nAX z(EC_g0-wo;^gq%OxQ>P>{eTDYpB?0Yld7#hi%HWSd zX$bWuygWXjCGa#oabrMJ@U(F_ZDCy`eSz13uR&|5RpEQRa=8D2zGe24Gzy}sYR*eCmFU+uU3_dE3J`E7ou-zx^mMXMLjF>!3fBgaYI zUR-kAc=_U!;|R1^3vtVFhKoAXehX5&cr?EkM`pofpZGhrMK(Ft#fhS z;j9i~B`pc=gV$_hdt=0B_$;64^L?N1ZO4q?V88uVzuhr#JRBRxsQO~*P%xIDdK>H3 zYmO^;kN396xQ=rj>;B{I7~7Bb7}oK1EZgHrC)Y8of4E0*;7x@%)hB?fykTb|1)(*^nTtIKE6yOWF3NHhf$YbO))=bE6 zXb;>oZ=hdF4)lotN6{fT3wO~XxQsxPLs`E&rR&esYt`x!yx6Zx@TmL@ui{s_q+MS@ zm+0N<*CqJ-7l9tmalv^3pG7|>o$|wgR#A7NSHu7w%&CD~QjZw@azb!i@cn>4L&uEi zAdZx#5m)FMc`0op*2q)E9saB}8U6ZZr_|^9Jamm-dAf$@L(}LD{bqOFF!lU31G+}c zqie)JzR(JRcu3=@fh`w^kp=lQ-2=R1P$%mLUBqJ&XXztYyu)JYzZ8QP4rr#40o}xp zqMhao#A`J-eVO9-oB=Im?V+B}f*22aH)joa0Ajye@SNA6w)I}Rz~}kgxdPweyL>1A zz&_9m_RId!4ffSOs|nhFsNYn-(=Fej;21a-&>kbl%`tR+2FKQMcDx;X=fL?`CE!~) zSA0A5NxGCibxxgIUWPg*JxiacZPF-oFP&nn)Vc^`$+$Acj5*`ax<+Ht_%v3H*)s!U znTB!w7dnO>H|A+`dWKFnCwL-s&2Iu4-r7gku%K_Q3TPbjNqu&FpoYLNkqel6asxTS zLxG&Zy2#Qv&t+W{eWT_}+qAl7YSvWY0R9P2)Ossxm`l&dVQ@+5nUU%1#xpdHdRX4-OTk&Krom@;jh2y5;W|2oFTxYyhl~pPy%73Eex~Mqcpzu9J_EnvS-vms zmD^Rk5~IrhC>?{>#YbF^`^8o9Ky0XZEH+o1605`!F-3eCC}vfxqJL-rTHvo8N(<2n zGz0BGN0hHZW6&Ek4t+!0l%^>yL%+z0p9s`^K%U&9p0~~ag<7vb|J3);Kfaf?vJXHD z4TO5V|JJtkKEI`Ol-s&b*ZIBlp5IRc0!>xAidLi>`B1HKI470gJCCJz#A#Zj z;<9ldE*lg5ahYdUYaWav%%mDX>sG( zxTe+VcKW?yIGxXDWb6l9lVv56zYdOQF)6v{m-Cc zwRMm%+SvwrxokL{esvc{_{56dR*JJx8e`Y06r6V>HYDD ze^UDF zv%a6^lA}An{3JT9jdipfO($ok@haBQeZLGU&dK48D|vk7^2QrK%Njyskd8DS<@Ls< z++N%>R>eN`0^`>jLu0ve{fdLey;%5UU@n*wmGhe?=8Jj5e-j&BJAnRuH!#oW;>!Jb zcS8f(xoQpKC0#9M(%7?ipt;Kr5JSytag;A;o`c@30ofG%HtsA89T<7aR$a5KZ-9U7c?dd#;7<`W7ho| zw{BZ_kN4MS^uOkH&(&wSt!;d6<)7wC3-d>e-r~_`e$y_z~*_uV~`Ky0C3;yp7)*>KF4SJ?D~9<{qxxOir;Prg6(QQ`aj=y z`tJ7or|;eL{qWts!{^iR?QeD*94p7gF|K3lmT!r(inrW4FP?Yne6`Qjd%QONUDLO( zb7}i_{L~mTUW`2)Sg6N1(bzOb3+>}}{MxqO?|z}ib&bt|Zu`Hl|2tYP#=phIc(*vY z&jO#@zYldT%*S@V_n%|u zzEE?potwr{&As-vopZ)u!JHdt+)tk+|2+?DOxGN2Z)@z={OjM=>*lCygMir0w-?8O zPL}hE?ewyG8;JJ>8kz>6qrv@x*9x`0+j_45*8Akz>Us1<3wik8135VzQgBTK^$7Z8 zQb4QFEI_{l9RsE3X&XKqxbOMWIo@CJc|N!Q^Xv7x^ z(rxiX@rGOQ-jSV#q$8mom!_mEpu8QLgSMe5>6`-XQ++cu(XxT(ylz|XfmQ>xx~J7X zzN;UlZE9a=p6LqefAvynqxN%KJ>@gK-)H%K^?MyV$HTF5j2&-rqvc02&iScxg$E1u zoZB{L*8Sqpf!<#niZjLPmLp3$lol}7T0ZONG~*Vh6_@q%kjKV&@m+gcJZ78XIrGB& zkPj7inLGH(JgQss${Z`MG9!zViWl&K9IViC1fHnJ?QP33{oD8)zQuz9?}FDU*OT+f z{p5grQ28M6GYaxV@Yrj%tL^K4n?JUnTh9;l9raoDy*}UfsDsE)`yp@bhiz={eLl~A zRUYd%`n`U8aVs9h(T;h`fzCytIIwsS7rGB#E7bONTkk90taIMx$sfj__|kaAi~iBM z#e1bcTfUPY&?2Q7TmCD}*xt6fuK2Mul{ric755d_72lcrH~=5u2YfM*U#+%Qd7>P! z@ZOZhuA{z;DUyhrV{{8z`LJXpt7E!8ovTB^LCKTtlb z7$?3MC&rAiWNaCeG_rAP42$8FM~d;rIj*i8%^c|0n>ZK8;#Qp6>Ob?5X2G2_l6hXf zEWW`9_@VV=d1E}XD+4v6>i25vs{A(ITh&(ivphL9DB#tqNmZSdcL#jDw&u#)7 zJ|#c9zb0oLB+t7aJWS7dou}Df(^luJWA3~DFE}T*t@@ySa_+_j_NjbR=eYLA_w(bO z{{lbNeb0khV(q8*+E2c{&+|Kt6YxDyk9}ty3)^~+_ttm&T)wTkfcm~MS#?S4C&W1M zOe_;i`{P*U0>-|WBRJ7#NL@@VPmS*B;P1hULH*4mJ+XRYtpoP=j`Yh~bZT*IteqYP<=@~yPi;SjS{NP- z>o7d)Wo8Oy4rU2H7R(mR-r*DJ$DS*gyTd%am1Fb#|HodljqSbfqk+$xG4M~1xBb<} z+pbym8`b{wOzF>c92^flI_jzFt$KshWc3oM$*J8Q9XQ77ZhB)Je=)>4!S7-kKF8(a zra0@IiqT@WJOqb}@5X`gKqJr?#)>iHAmV8nnr6ny#u{y7{Lw%(m9a@{RgKK}EiNuy zrv_o1PYBc})GG8J7B{Qen;ZDqJTWKRxo95Me5^TXjWTB zNq-+OPQ;P;Q%y%rPOV2RPM?EXklqJ%HZ>zPrK+V>?M$6YZ>K(r^#Xk{RXbDLs{R;t zvQ@kKm>!wxkI|>0XG8CXo(?@8Z4a=1k?Lj9OQM&h`dIXu=s9UoeJE`&O52CxvwUXt zp6H+PeR@rNw|!{)N%ZIp)Kg;rtFOfGQsb-Ip5O0SI3|vdV^zH+j^U_4&9drmYHwS;**`VbstH!# zOVtC7l`#YKz0e1AgdQ$2&6pEcj6>sAe4<;_mDTjb33JPQ69>#k`pI0Sxy)mHPOF*s z^c`MMqt{=DfANZ50lfv*q_|!&j>26F1ubFgi&Y4FWp^_(E*P)mYSkrq!aC+ePSw}J| z*f3Z*_-ydWKx`0ithdmYfukKmoQ#WIrpVN(Gz^VtJhWV+zZ>5eKWfUxlC=i~k3Fx4&o;L89@-9H(VTRg{y>___tf~pwe$!+ z#?SI`{LN#KE6Zo?J6$1HHO^>yK8{!+$E5w`01F0k#`2x`Rjp^`HQQ9a;Jq{nU!3nJ z*6?TOe%~v$(D#m)aW8I&6S&*{(|(Q>ZLFV6&D}bTx@I))&;wxZ;#ECu_!DR8pTli> z@AT)n-lQJ>_ouzzyLRxubM3_!f@gv!gU5ou1@{Jb2Db(`c5580CAh8+6SMB&hTxXq z_TcW|f#8wA=RF&|7`zs|8xSt8?V(Q)U+ej+HH5fRJP<2zkDdkn2zU+8wCe|*bH_tJ z2~H_K5hsl^8n@N~h_|=}+y~=P47QDJy+=%Syz!FHYsaYkLT`g{XFTdpskJb;5GUeB z9Eof3CJx2<&WTuHU5g$>aYSs<>!=sganLKNkJ7pde+yz-{?$2pCF?L8!{>v^!4tuw z!9(43P@JoKg1Z9etl+Wp;r08wzg^*dbzc6N^)Jqg@9~|!-#8Hm#Xa#&EEB(+6a6N7 zNW^_H3Fn)S&INuq=Nvma!+evA(*EWfeMIxq`?$fFr>*3@^gQm6H=A?3QrevtYmU+E zYVCRp=sey9uE$68p6;+z?o|?r#EE!~RHMJ^Rza--DNeR|D(a-wmb&(*m(#Xn>z*31$oC z2<8sv4MqeD2a5+w1~`4`4(@wyiD1!SWZ*scUi=vrOdT|}_nl(XJmf-5ZZ}@R= zLU3rXN3d0}POyC7n9Lc-5nN};x=qJVKcbw)ah20J&hj4K5D$fCB7frR)7IiKo^<~B z9(eSn;KO2pF~W!7GnmKb7G9?_^%v1NG>smlT2r<|I+gl+xjZi2WDepAnn<6LIVxs| z6UM0cfJf!ia#)%e7l==M6rKug&vW5(iU;xmV-sf^pH)luJF9*!M$jZU-Z3cel0RfD z%QI{%*YJMFhj;I@`Rj69bHSW2H_Q?HB6gZTVgZh-SRlTMn>b1QbKJxP{8-1%{A_v2 zvBRhICT=SINB5cA`poEZy=>+?-HNls5*&cL+MadsrJ#?!da7+_ds@DF-ugALI7R>J zyTd1Ph<2TLoi}~^au8mP-^+i(_0FHz&XeFl$b;~=cLom zj>_I4B&4Efun{UlG*KBg4WyJLqQN#dsuX32Z7dF_QJG>JLzzn%-_G^@?ceo3>;Hc0 zy*{6Jt=}4+;ePJty6@+Hp0$=Sfw2OcyIf#QV!XlD`R*UPhiBP9{QhL{MDTckgC7mp z$bSTX5B?T-{_ddwPZxeW=Ni|$S2|IvNw!@c-Z#P4d#A3o`fNWNV{D{c><*r{&Ymse zE8u%=sr|H>^@qj+))KO9{0(ix2f_FJ5;jvGqNjWp^Amik^1JzMI^KT7w^C4jiI1f} zK-a!F?!5ly+Tq;2J_C+9?>c|(*SE}@>5KZGxIf>N?XW(K-4J6IL*UCAA6UD6VIT%` zd*Ho?c+m68(}R}=uMA!nc%Irb=I;qSOU_<=GWcx3uRbO)X5eR^8hED(zuP=jRlrsnABY2qAw87$ z=ieL{Q(1#;ZM?Pg*4ncVWNfZtM!+^`1OBfVxj$n#elLDCSB2~N27F#V8GX_>=@&ht zZ@7#Nz+d_v&cajp1>fKy&kBnb44;{?Pp9=wd{jTCz9m1#7@Ez&qw+L1#k2fEc7aZt zXA=XW$8?gO8S_jJ#Foqzm;#v1RtUHB*=ro<`CuFb_8r4R}Vs;==GhvWFW0)GDWfpHW+-}^Y%Q#aK77|*!5#rS1{an zjvqzm`Ex@!#y&ako|}UDdbo)#(Wmg1__6+s)5QC|XKha4nI6xmcwWXBgCT27$@pSs zAhzPY9D4+8#UTMZ;a#Ao1m6pA?)icE(v<<*;GJv66L$yf!ehZi{@8^73ThK@`D(#> z!A8NB!FItefjH9v!C}GC!JNSRO}&ThtiU_J;HSZlg6{?=1z!n17l=*m8L%N@O=|~Y zOzeiS&5ePv&}jktw`agUt{E6-8FLv&c$UYQ%(K6qr!hzT_Q2QyXOgjb8nGj>cmB3{ zPwmMrX#e`a#*)UA9XIH6=JND29L_)C*YaW61pbWiHLl@*;UEqJA4@+mrvA(L9T55( z7-Nfv@U8e{`j6)rJ*Q|qOuvog>9jtrf3q9LFyao%NN?$uIV*KCPQcUZ+Hvg6tdJ)I zc4F$Be(geSgY#^OYx~?U`(U1$ezLLRplkzs)cI+7vb*dQ*f0F1$pdfWH1$F+KVT2=Ho0qWe!=XZ;{&$4;{$%_@c}!|?>EltxWLwLykjijTJ5Jkc*z*R zSW&vRemmau%lEeObDwj@Ct#l~Fh=7`496*S#2Ccb({UWgX0R8=Za9)1W2?xMUt=DE zJjI2`ke-U+h|B#n_-1fia71uOaBy%yu-_Q=9s6`}Sn&Da$l&O}eZGZyQt<8I2Z1!_ z2GSQZx-JkK)Hl4_QoCtKV-D@A&5UFGY3q(Z^fzT@_lo zF~L5;CxR~oUkgqP@cwzhrNLFf?}BTB>w+7C>w{|p$DO|N-8A z*Yw%-^cBysF~$~R)_hDp9$!*lrmy-p{-LY*lV2j%!!F=ldS*Q?{jwg`^H}DsX9VW4 z&1>O)wt#%-o%wFC-+6E21~!3yT5BsUYuV_y@dCR*kLATa(0BcUJ)rYqSH?8fFZr`h zk1bG6{mD9LYeJ3N^f9q2ZDKrOEMXjhYw;mY7Q5r~kvn_IC&c4ysCYkns^8-QvS%mA zchqsyd9?+97+1K?SV4@!7(^S2SF!cm2illDV2iaoF6E=+VDWoCCLN(GY!^S1o%YN; zm=jWdz6IOBzrfS#!#>~}wgjX_*USwVql$NHW4ed$@sq9J@^($ z1>X9^zhlIz?j9dN&PH%@H3 z(-~t6K7c-K3`Gw7J+?tV_q?zeuQ(z3Yom@Uw6)k2J@sBR{uvn8`)zK*Iq!!uH;7y4 zh5P6RRA2OC*IY23pTnW?1eD`JF#l8dhn`%UDzTJ zH)0peM}I6hBrrxeIXELYKe#;jb@2OuP548=4tRI-FMV~Hbe8}O6GJ=g;_-CPmhTKj1W{gp4J zpXqz}-iN9g}JhAy%N#-QeY%>(0Ib>yeuZQ}%SANBz3nit>}aej#3G>@J3c zE7cQk;Vir?216$LFBy}S7zE#s{BVFiPrk+l#$@6de+ux8`A)wbXP5B|4$)51lP-I} zR}_~QBP93EOVR^&SR3%o=?Gn+Gx(Pe(U=1c0RF|Wj;>Kxb#~l&ey4G%EiRBoQ{lL4*a2l^OUaTwRUh?Y1B?&Um7WcK8uBnsz@KDAZu0{EySUMB z1M|SHnG@f&bV^|e1CC$ z@fk1<0Q=%XGXwFPt%F&CIf4%d;ya%VJ`)@s91(mu;MW@){5Zhz{C?ws+k^XpzXtyZ z9tj=}9t`ddZVG-AxW{*$jtveD-WR+nurA8jWx?RiX?IS$BCxjV(BR#{YXa+!$mQz5 zyK=lw?V#X&ftZM~0e=>!;VizizNcNaHNW3DQ5+ZF^AY%;{3ZQ_{@{JFA%3J-rgbOc zp!80h$QnWXLXWGrbeKNVUA{m4buAy34mXBvTw%^ko?;Zr!XD|*{7c+fe}RseQ#NOX z_xb0>mwbP5A@#Mk)fm&*$TJon4y>&Fg6&D3ye+3nr&h`tkGZ-{uul&5Svpc{h{+V{8v1Sv-srf;BXAU z-jccguAkQykZ*m#jse(NzM?PcJYSCJx)pH-_VC7FUSLj0pBE$JmypY^25W+GLw!tXk`+6_rj%Rg zFZ)p5=f8Fwi`UHOcbymR=f~hdaZcj^HjpgHlZ=d|_~&fqox#lk|NRevc+Qo9wOY7; zI7awx-a%`OV0>_5FlP+=#tEkd#u3I7X9ed47YEjSu`9m|t`4pV*c^GXKeq*U1?DKs zTeD4ksrvi+nLbvq6Z)Vzk>NPo{E&7w&SnG5$7^Fgp1x-6$KN_IU<^sI58_x~La%?r5&EC_0`on(E1N~dS(XYg5^*_%FcrT>) zREiUsZ}#46>+h{a6(89&ct>DNaA3e5d?PTI+kTHSy(F%ci1979(0=9#l;IF?f5I15g#Qk>;j1fK<>>C&>d^8aA**g$da*b=< zW1iWXRB2kr^sK;K_I>##+8+XQ;3ov14t5L7Ew2+SA1o1&h4+24kH$;Z9(f+XcmS{A zF>|D1-qu!%UE?KV0ZPlFNhbId$YDdjM2DK3<&q*^5Ob%9FObi zA?{%hV5)B7erfSR^+~ZNehL4V?$IlKTYu*(;sJSOGk|}jJ1@fV+T5H zorN*O{Xt_t_|E(cZo{4Wg1VXu5r^UXv%ff5pCxnlk*utJ6n``C2<$=M+bBlFHskb} zfw^a62%L@s`FebM{Zjij)`HJLAJw1rZTzlW`X~EqOpo9BZeYLT80Xmmu#cmMex6tX z{o?0|dExlNZ^!$bFP{<*kc~FyC(2J=(l+ieR#H~H#8%>Xu?+Dnb1CI_dS#qSX66BK zJR4~2#4ZlUO@B^&-na?h;(Tj7#e?uZ-(Osa|IUxUAmG>kB>3_8o#bca{oy|fa67v& zH()2+XZ&zez@D%xI9%S1*C>xM9eb`X8B4QC%B|h4z0jUwfqXITDu!p=27ENUu8-n& z{+oV8m+(2iz8r20hpV-}W94ht6zqIC+}3aYAwFPJ>5(-KbjmXp{5AH(mhNf`{tMgU zTCoJSMO)EvdD1=Yi2JoSU%;RKvq8Z3GUqAA?HL4XjP4FR!}YX0H}!&G-2e|-lQ}yO zFUR%g1^8SX-Tc73z&Jq6$Gq_Uf%T>M)jCqK9(=w*z~8nmi0}QO0B?JyK>YtxL4E$C z@{K2R0jCDOe<`+qad1Vz?+0-p$DDJGv4Jsyd(9K`@!1Bpz_SUSdr%JCZ{AwG>Vn|J z!2CD6Cg!zKApXPdEfp*nSXU%Qc4gqXFmgLA@Sa7yhp)sta1>wFJPi&N(_jN|5xAad$c*PNOc?4*?tCxA_k~d7mE;%PGW%+-tm0I47nrU1JG$Q7m7c zV)({N{1LGxI?7gxJ>f50rQFKTm%*L-kbbAWe0ThfgF0uzCo@hmpU2n417ZN?&G~ud zAHLr$ff$ZBfw=KwfqRUl#47Lv{x>c#M^KKy4Qz)Phcajr?SmiL^7d16Eroup@ACa^ z9Tzjv_WHbjtIgPn&I$0n*$4WfACp7lJb1gjPgdr@*>78%&$sXRhcD00o9{KgVEge! z<2K4e@6-dgcTU9k#rTu&!FT5i^Nab%e0X!>xE4BZPL9_4@TbLi$X6_ey#E;RyN6%{ z_~T+WYym%=Junu4`t0n5w4^IPu{OSswq|$uN_-}quD!ZWRD2JgYbX6lJK$D6kbXiI zLwjBOtPL8MA$#_gE|R&irM|?b+Q+YC-M9=NkG|qx{hZvjYi&dQc)HAADCgsIWy9@Y zpMI3%afR#I3Nb@%$xh3MZ@`}GKm0fIX5u^6XL!fKp#jdaCek~5t$*>&BhSV@H?Wp; zX7H9^m%zNJ=M9bwtQ$RLye3He-}46N2Dp=t?|B1r0F4ECo`6rkTwvan&uwie{x=sZ z<|Ec`o#_@eEa+B zf#(W76wC@X57rA-4VDSSxE2X0wK)I%fp0>2)|ox@e&4eL?^ZT8VlVl$o-r_o!awDk zifx;>$8|UkmztxrP6em&MUB<Xtt@x}P%px@W$V?XGiSdCZ* z+rq}MHReg56(|RO#?xX_fX_iW%j@Q0L0o?2z+5cbr4H6>N>e@92zl80J(0O@9FEKB zne{sEVe_olai6h|czwBDtX^D#eqJ+hI$iI%3;>zuij@sym%i;$z}3{9zo)rr>ewkknBO#@NQV#aNIWj1k#6 zHV+SrxAV98*5WJQ2#yY{!#N~4X#73lkK{Xjdk4D(?+JDuVAs*tsdpWH3v&0#-ACW6 z-fMEN$qxn}3O*dTf8SugV0Q4S;B&!I!PkRRgC7K9{TBts0>2FKKl{Lb;D2+x);+N6 zVlu|T_@6)Dd02eUS0@WTk@ySyN?!O_U(=s12-rz|iO<1D6E71x1N-_|b(ZhM*5ZAf zuPu$~rKN4OwYFh9hMX^7oGzoJ)nX!I)p)c%vMo-9A;?Sq8^!hsVU*mC?Q8E8Z=%MPRMI_&who zUt4>Yq2NFH-+0}+VlfoH_=5qvRo-U5^b5LA7x_K>@~LrS_KNnd z=2!C_^MxX%lV|kXS>9Zm6I9rQX`RZWbF@6=)7E6atOH=!4D{~;n2ey7|Lp-hh`DpH8uM2*S^x1j+#@viG z{;n81ZXCk87Vm|? zo8mHJGS+OG>-7BYjG!@fKKOP4UmF*T(TJb#9*EQ6bMYDLD)G6s`(ie@ewRRu9d}Dt z`u`R1%{?>raKJymH85vve!z3f>M%RlHP|u`Z(lKZW`L*bmwQ)_cMjbW`0mG5f#>4b zJ{-+Ax0cCTO&o|bhu^FcZ#Gvc)+}Z{8fq{O9N6Y78s(6P#_U8Zx;~st+UdHRE2j=XK5577EfA7TLTLGUBM_m}0tHb5| z(%Z-D`ReQ8g~ny9#W2?A&+~8lEDwGV7vukmi;2nL3%tn}#G%IjotF~>=$M*+Xj~0= z7}$r#2FQR<&j#rKVBax*KU?8?Fj%m_9P-Wa?sm=)lBKK@68g9E-i`(g|sZhu)Io+BnNUSodB zm`&ZypWt&ox^bTN=$a_<=MS@owY}tvDbE@oHpH9j_8GHxpHvlxl< zt}_Pc^>`j<#7&gV&LEQXv!M=f*(OZKz1Z))ir&TN8_~4-bwD*cR`96EnLg5D)TPvG-4~L*hj0ad#kg#Ab^lJvxSc?<;yBxF--> za=m#3&ls3XmX`PIc@~Wxv1`7+#SV~%nB6e}Ik0JL*$#m+E4w;~^Qjlz^sKG1qPUQF zFb>yW@SHX8Vglm2#?1J~z|y)0{DZssW!BvB+4$qf1cwLshi}IZFZhw42%O{d9TD*R z@I1cfbMpVJ-H_fH0d6=i!1vZ~vlZs=a6bEi_i={!1b-g?i{*&R^K-@b#WJSO-QiaA z=jQ8ht+^4rkAKAgtWg(F1^Z&}c*i;ZwHUZL1u^`_!Ns1;4|eUM*om^3OSm-gK)Y!x zbFB2j+=;jv8I;G_GIBS@GcF->vV+=SZ7`ghqKAhB>}7fVn9=BJ{ySa8-M~4=alSDn z9TvL*`;Iv$UcxpMeml=D7;Ev##cITGaJbm4xF`DB z_|SU@Z764i`K!5omk6`r}^p9z|$)R<~7Ajti}9mU_Mq{UR>wwz<5DS z{7V75WE^pDa9|)tPrdI-l|<66(l;p|(-@Y`|M;B{k@^1E|?3~)LbTpaX! zfWD=O*R37p=lkZd@{8A)i#IN@c8HB)zr@(h<9NPIyj;wMFGH{SMPk1EOZ^Mq;l3Gx zc}o7dSSbHk?AID)eiz>4JM%roN5s#KwZWPq@nLpBtj2l{>tEap_E!$XRGgPiufxUG zzu^L9;qR7r&0`s>cOFZ*t$h>k)W6l|^g!K=$2$f$c6N*}$7gHYWEgwl(}8P@!%qru z0^iYiMZU^mETCL`PBOyj zHG`K1D+c%#uk)#w2zc7_z3%_}zwqS3Bc6Za$tRxtdvJencYx<_3vB-!JQVnI=Q9Gn z`O3j70{p#Uz$f26*gIf{zZU#B5I<*Ajmyl<8mEbu8$)S(@fCJWe5~u4_y*>-@ws_m z<0(EmE@u~qZECKP&n~uZ{fTzP?Z!XGQXN0dk8~e48XGyrzwdc|K3&G${P|M@aWuZT z_>EY*7`*YRmA2jU8R($@#L3^ofk2{sM* ztm4PwRKRccoDrfF z$My`ZScrG)cveb0#9A72snV8@bvA4bPX2JPM<7OG9saDqdw$*?Y#VGHSfisp;@__h z#Q*DeKR0u`<~bJ&6v8@2oW&Q#RoRP=-^~XWPZ4M46XPkiQf!%@jK{@P z_|4+sxVhk;dJm`b;?U-w-DB<)cS}p0MjXSt+r&cfAsdVzaW`LET;3Q#j9$AcGv2ih z-+P?Z$DD^)5!km@M^M5$9@z21_24(ZoD8hzWV6im;%mM)+re)6W-h(zctja}ifp-1wB9>jPrrd`>*6pYbzsHjeFczxY^uMSr!< z(>x}wGL9At;SYm-KACIrw|KCAjkEP_cA&O^J#frcT$TUU_o2Kt^P<+$;BZ@Iat!R( zzs3!?RGbB`vOn(%@WT+|Eyng@F2xQRf4J6tcttw=b-W`Vu^eS(XUy@clQn$!*_;86 z#%ab2{BGlixdA^_tfcb?N9O$j#w>i<&jhx`p_7MBelj>P*f+qNVk$cZZx8sx;@?{a zn+KZ){B|*r^@6nmagmpgvBu;YBQ6$;c|+iS{{H&{F?RWUHkj&Di;es#uwDi?Yu`Ro zs{Y!DJgmhLb0cs1LWX#l{tV+-<_+*Ezgdi$on$}7z3HH_ARkza&wQfwU&e@@nZ~c? ze#`-R?&I;mph0|)A0i&gZ#K3TFEtOzhc<_`ZeSe*e-0<$3jCq(c!r+;#dp%*^vlKb zK6bHy7X@nvuM6H1>=5i0>>Yd{u-!cnS8xtDRlf7Qz&*bWaI|?+ezdhEVmf?jT+om0OF#?{=-N->z4FzW0X$YY?rsH0I!c<9N1-&BE`m3RVm1 zgD*L4$r112JAODhn zYyG=P|?Sk=|>N7KEDh+(e z&pkE3n{xtjdF2!Hk+-e6R_9zNj$@t3*@5)LsiiMpd}e(gj<)Vk?1qf2y#z5P94^ju zV?bVFO`g#i%IZ6LhDTl18E4{h{K>Z!f5NAYKiTJ>_n7#FxnePC+zw(ej=LA{<9~kr zP{#66PrkE!%}q3RMdoDH*wy)Y)|AZ6ud+kxqE6;7#j5Crbw_NncvWpOPG@^@I(x_V znV+C@#z^KOjElrYjA7`evCTz+d3SU0I2%vXZ+cGO#W=0?(kIpmUK^PIF`m|!^iBO( zj6+N>F;?*jOC@Xf$G#B7J)qZ=i+O8PJeJTZsyDT zHWuHWkB_6pWvnsXG{Do=L#{J^<{w{+vx`sTZR=t1HqJJd<4f=>{8@v-XU5n36l*5U z!{BN$JmdT01M4#m4vatb9b;2Iw6Q(E)%f1p2=NF$6rbAjoLv_#=4~9zch*m>BNhX< zZp8WnoF(>*$M8BX#Z$OW-#1r@1I=ri>%tXy1E=F|TqX94ufRUv6UXBme1_Xy4>-0w zD{hX1*$G>aHf|cC*A2M}XNuVZUdN+&3)hPW@R83MKP%5}unA%W?1nmu@#1gm?9^Qx zzVMr!12KBIIdE;^caK@y1J1b?KZ+ldE1qN<3cuM=F@OGWWB(82U0&>lc)Rw|Rvq&g z6BO(*-?`6Yic5)IiSIOiB|aq%W^B<|2Ytc8#v$`Gh!Dw@s(4cjK&Je zTTUEuCf>x6>bpW2Do7jb5Owivj!>b#z|!@puw))#s|r+G#3j>Z^H$U2a(2j(5cw0vVqY-B1L zm*JQ5$#JuocsW@N#`8#Ej=HgQar8F?;uSLj9F3Rx4dxDw=lBuJ1?Ca?6-x!yls`2v z4m2h-KIDTKH(E1N|JwRle#)i6Ie|Eyc&r%j;enW+7@_`T4$i!r^&~q4VhHAr%pn=; zTHhw#FpOv8a`TprMeE;uD!#dxiFm>A9SQs|e1SV23;1R@hi{6{@DtACC-dd_v3z%& z2sr=B06&TgyADtCb@|oKxdzmoWn0Q<{@|72x1RnKaR7l z#zo=~+6&lTaU5~tVXT;6E6$^BwJ}cZJS&^+IN)68=!5YWADE7??_wE81m*(SeCu$; z&Kf)BC)2Gtf%W*-DT%S+Wyd;C$Y*dKSMwt-3Ouv(t3XWT>M`)P=M(tY#*Y5zJ?<9Q z~*7+;G|j_3L1IKh~l zUxxE94m@v&PjCyK!9_R+KbgNa1Q+}5ymZV5T4N3NrC|)guf@&NgJ%TFAw~=QSaDmiX*_Lx2+jue zPSe1=Y@XxNbgy%s=OGKeur(Rx`OHlgev?_}?8uNjt>+L2Hg{)B=3>vb zV&q+)p-nCt!#>On8c!*w<8|vdw5wy<7{9aaY&@NqJ%+f*!DHB`Gan21-t?&NOQBo* zZeu&Kk<9}-$d@qg!|UFi(r4xIxVc0=wiq5?WWj)LKNiq)JWl6t4sf_<8!j2Imz|qu zQ!Wf_ar&8oF{v$-M#?E*Pm*F%qZZ-hy^YL*WJ`_7Q_GSa{CSTE3?7T5_wt>$twv21}{pQK=t#jpC z;{`q@|C22-UXYeC#PB$dc8z1;8tK^fm~u!%9`g2EnHt-#y>iZVEr&Qbt`|!Habk0~ zowHSM^(Pziv-sb2&a;W^qvvDUP(CnO@`>3~Ho)<|*FYSe{A-7`0iU^X3+<-;0ROYu zpzXx_@jE|Se3}o9*W2#e**&&6AD>%SiO2DI`JKPO$F@d~j+x)JMh|~q5YSJ42E7Dx z6mXoMZj8!5(y#QdjRSmN&d2x0 zz51cJHy8_xTZ)0}v;27w=hT<^Yao^?R;rJ)1^j$GQGf8X1x_2SDLgS_@%n_;!nqEo z@FCd*9AzyU{xOz!zgVuVF@teEuqDR)bH~6_&WrgrUpB|J@!s`{FhFrvAzoGyl{zxVDb*%dE`;{+T!@-;Ukj^I1Otd_Xn>*bi}0YsPvV zKM{YJ|FVwHK90dP;zHK<@hi!uB9jR#ayG_9%ZZFLadpR??vbbK zZ2jSFT;4Uk*7rUmpufhb^j122cXLDZ-ZjRvbYEGVufCV}^&xYb#>CP#-^pik4_}I3 z#Rkj_#5P^uGU#8%2Ysf`b^0&+!Il_T*T(3RuMhZgIKef_$WPPn*dSvJHbvZ(P0{CZ zg7~U_EkDm4iuL0Y;|rW5mZW?y46I4u7h0=e-VNuNual?yW(4>`nrxbM*d#n?y|MWN zX_1HeTJOMb#mDRbf0mD`p6X3L*6PWpb^$-CgL2>@^2KS6HP&T|BOQ0%b?UF&%7J%{ znZ%zRCo8g-renAkFZ;&1`jfr-(-G}X{%mz^v2<~#_O0HzR(j;TW1xM+Md%1$nay^NoIb~Amxgwf zzdmJc6zISD9sZWLIU)Ur55$kxH-8pf7MT0wO(0B_)MoPziDcl}?vmBnx4?sAi{xjM2N%EE>yr+dwJ@%Q+8>Sx@6!|)7T z9Pj~MhpU{EM(eEnWQP}VqWi6#)OL77ozxdU;z{$qKM2?@T%=Cqsm{*fWAO)^NM`a; zce2Ev)~n%F@n31+JiI8MtH$s7As_W|J$b8#I!ljyl}CM~qfUHUww2#14d=;E8tUU7 zGL?>N$&Nm`Ub^ygWmrUdpPyaH~3yhw{(~=N#uxtC#fW1-19` zs_!jreg$5_htiZc9j@%~C@z;Cd9wpJ7`OJEc_7@6%W<-DYeTxGt@S_gOZ^u&;v4rm z$Cj`e`iuTzq8gwoutVx6IZpxo9-3Q z$6Nd|{yyK|H~^pGPWg+oy52GIT0F(?#IHC9N61SV_<#6E{c!X~f%I{&Jj4RPy|@bB zIuG{c&6mXGuEF)LvyWrRjvY}h<;Uyd2#&jttr0ioU&@;=sr=+-U)^!HF^uc^rPgM! zedLW7>#NGQYvjmSJ>@AK>5EOZ4rpCP(~3 z$BYBr<67q$Gtp*z2;7M;>7w7_eCz-l+;L>j^%-uOWzu(W0-p~DSi9o9zCq91pUcnqLH|@ezmeGw>nB+n<=-t%SRdH$quNib2w8P@>gH7(EfCftZxa}2{AEa z0QQhf{Ly89_!9rBFI(mw^DFYEcFm!^LEePwAm^IMelX0H^ZL=)FEh z2gwP)^OJBB{#GCQB2Rk8FU5y^ndT)Qeyp@`hrXrH@NaRh>%l&L)nD{0*YnFje|Jn- zZS`aK;78|iIR0eI?VSIJ!2q`@x~FU8~T*XClOINrVdR=mfi zkh8i=1J~fa#&y^Y{;%?|Rj$W{u5pfBonzaaS9Y9+!eNCI-2j}rRJK{VUk^}$3 zHRR!0a=ht0E>v#i*SB3uew7z_kiGn?+vC3plIb4R@Z26D4gZ&^2J{=x1hF8b>cJi-(i9`6_{7Ss%H=9TP zbelf1A^2R316R{CZN%o*N0W#7V7_PVnKIxG@>W-}F;}O&Yy%FX3-rUa?i1(1>FO*G z`l>#9@t`iP95|S`bb~M-55r^hP?Ip`l$L3dsLgQ9O|mv^c>ge zYj~56i1&c@(~t0_m?W98J#2W#9sC{KqfcpDWnn|n;s(B>{wyClL$>5a26Rgq z@C!c`NAThF!N%qF1)Quu%a3i~6XH~|P+#L+{e}G9%NFr%`R4AWqvW7m^vs%NF=L#^ zAJv8(-_iyAz%Q-cFkee=^i#a4uIh}p#0r$(zCZm;e^n-&MUGl*jr zcQ(j1>@a=Op4yqtxDHQgTXuo%!J%D?fnT&O8Iuj!%gg=B zL+08OpNkKflcxXlO24JYc$Yn)8{#|0Aj*O7wX1TFAKQuR)yH{#h!3gF`S#k5?Q@NC z&{yZ(Pgm4e+T=pU5QnLsd+{w@Q)c?%9{z~*ocE``%0~|J)COdXhn>Sg zohx;X^58N0fwQ#>4#H>TtKZ@qvg3D z`JCr}&@E-6+w8CW)Q|0FOZ0KFQXleA2Wiqpd~S?kzJkrsZ}lmCfxqBh<0vvvW_7|P z`ZhgbAI08XCtrF(#-M)s1UsVdYg^nWuB7bhj6bDEx0F{IfUal{vcz{dmu}V{!S8%u zdgi+60iTK-?30!9ursbxFSd$*if{BAWyB@8LOaO^7vObi;$rpTlaLuX%L6Ydul}ZP zbVmKOmG*Qm9nxmXPmc6Y+4V)~;~V3DKDv99#d+5%hkhm>I!{jR=lDrklvn>zH|?Mw zIH&x!%0yoHfGnh;tnzlAoY-*3hx|?#wXtiILwU3t8R?I7RNC%WH}`92urE!#M=$Xh zK69KbaG|Yo>DSUF5BJa~{aiX^?p`uhUgyYGUFoy>Iz~@xqxC5^fjz|OI0494{k5h0 z@T$J6y>PH&WJ$hwzI9^5=`{UTPWRw(dCFTq!yPyhuei7NP}|dG_iH0*8&lAA?TKsj zb7j>(w4J)sgW3*hXfL|wS{$pNNmm->1ox4#d}?=9goc6$X z{BUt}e{3FmtIw-DE@Xqqxm?NC_^n^5D|ygc^1>Zt?_T;!he6u(i$1X5%0{nnk-U|S z4C%HpvNpgo{A2o0F3QbM!zc4%qnxKh+CjT07hO|ldcrR0%j&HCaH)LAU;gDboT2Tt z6CNi=b;WOVQGd}_^ojNp9Pb=?E3>l6lT64!{oSX1)SJ9))dQEQhkflkH(*x^xPE(+|J(L3S0F(L?=LndPlb=negsmh|b8bIPD>@~)5Joa1;v({*H@FH1YBx}JGSr52McetU z47h?UlvBT^2d<|d`V5)7k1q1>>5)AAk(;p(8_qYv@8Z97ogDRHw$FLx*QRU)9if-n z2;Z>7?p1!*&=WouZj=U{_vaqGNLKWXf2Z%Ozy3hi+^en06*p^ZJgYwLQCIzo9Mp}D z;681^N8sns8C)VJsh?^$Tp||kJQ+z#JChA=kgohIQ{~Ygv>jP%JL3R#&=0k%vdfDO z;Y0QvN0Q$T0h>jK_%!@h*RiY8bDy--L%X?_p6P4aTpWVzYQxmoJ>=KpY(JTk6Q0rk z@SptHf{x#vQ!dw$wff_F*DEvmk%2ZRA8kmV<7}@Z%=!AZTOY}8=_=K*@pKX-Caws#I>!Z@v&+&?W1@h9a`hvRXlgdLM`NrDT zHOeb5f&}BM9*KmiJr}-K6Q5NabXMIaq>4oxWU$W7cm4mLkULLsI zADxoFdfIAp_m)?+t-jzs^>IDD1ZATqpS4V_V1A zIB8Ti%BVl!0A*A+`ba)}4Ej!PxXnB@&QKPdWqgls)rror!P-H3>f<;W`O~kBy~*64 zzN>xdiI{?Mv+*q6HhwVncd^Ov6XtU^Z0^ZsXG~vo%`q^eWgG80~x6w zzlnb&9-yuCtF|xwGKM7sa;87zChc;ve6=MxlLfuNgWxya!-;sw{bZmof%}w=-!J|v zADk~u{Ggra2M)58HvQul;2=7x9O~3^&^25^-p=D1Tp&GmPg`m$e4&2wkcP6jPyYTK z!_WGPeC5wxN`ua*kG`YL^;O5|AKjt<>W!D!8L}mN^3X?=MVi`|PSYv!&?e;oZNmom zO{dk-ADsp5K*r9gv$BZEn3Ki-@@Fg9EcVZ}^w6LD=?;$IM=LWuaZb6kKOLsu^bU8C zuRh23!%w)$zCNMu+E`nvCs~mv|BF2NGv{-^|OzDv}cMo|vN6xqdv?Cj+JjMmK_(eV3 z=MQh|56Y>JOP7tb&)*_f{6lZtD{s0XFJ-Q;!HxlWx>tI}Z@5Pu%F3SUoAP&!b95h1 z(Mk7^pLWp)Dr54+1Rd4yTN7hy7SH^#S%ZGo5 z)9H@BhC8*ZI^r^VQT@VQWTvfLgMV<5_E&zoqn*?X=%7E>tDAm-1GF(+^jrJk)TY6n z(+xV{I`x^C@fjV~*U6dgsTbXsAO6G<^g>?xihQ+$eoUA3Cw5F)y;d8zR-W3*Z)H&@ zeNX-A51p?5HU_HA$J=9rR9rKcPWl9e8h^&=ady|1YG>nJnZ(_R=5={Xl!+EWfpnW3I&) zxC4*SQP(*~-s}Xs&AzeoY#13kFArP&Okc$_^oLyNrDNAkwr5$!0S)~}qS|FQ2tN7R7~JGP*M(xRu1lLw9_8(f0N$sJ$eDQ)dK zeX2U;9^9dixz~O4LLIcHzTzHs*tP0aKaqU>;XCza8?~$UkdF3sTs_p2zN@3_q^ZrM zNhj$q*|LeQBPSf8F4CdH+RAzD#QxB4{fgeWhhHTh`MKUX`8n^HcG3p!(|*dP-|9E6 zb51#2-#$a%a5eo0vLkOYQg%E`@5qI`s&CrYwd?{}t26y0EB#BF&Pj`o;4pfxe>qM* zK*oM+4|3H;@|734_)|vd%Txa3+5FU{`c8k@K>bO5)nA_S*58y54>?aK`ImT%+~~PK zWp^Fj1!>?4X$&DfZRi;L>Hg}8d)!OL(*s2D z>PfEpJ00^|`)E^tWKDKJ7VH6i*$qPkN`1)>qZ={jGOUf?IX(SD%4=p&mtEryf9_uAyJ8mp)17 z9VZ9JwUx4I3;LnoYfJ4){*C9b;h+s11O1E~hc>s@xgU>#V{EWCr2n;}j*~ywmzK7* z)y{OC{wi~MgzZy*_0mpY-!W~EV{kIrYK!WDt@G-sP4s#9IL_vfJ=;M~wFUWWcjsNF zAGn9CTrYj+rPaRToMW}wj=Rs+ar&XZf&KOY<-#-WbDiJX0@T-UY5Ofb^1!uZLwCu{ zK0awV>6qVi#I@SHx~Gll2R)G%y=>X(mwM6>eXLwkUqKo`?!)sqivDSfwhf&G*Xj?- zFIyUtsl|T%G3RhtDisjtAlb&Publ&fAnrt&IN^^85F&-4s`;42)4tBe(Desxi{ z%0`*mrsS?Ju2)C0a7-Qj2K$bYTV?1tdHAiK%CBxvnUjUO((T$jy5YQY$^^8IR0jQtoaqF)ldWU;sN7ebp;Pi~Us9%)OM2?79OMI)MP*c- zt*l&AS+%b;4aeQ*w{tx%O{nh6*Ll}AKXp|`^tO45`=_>-CjC?f zvK;D2%Y~=#B0mCm(y4M;^_*To`-w73SG&kheLMNOsT|eC z%2GaLUHI*s>uu?vKXoS``UdU)?PGL;J);kfDX;s;NIB>OopDZ?q^-VWraj2fJ@T;~ z>aY9hk~%5F&<|oS0e8_&eW4twzsT2p(o-JoO&)Ya8RX9{P7kz?cB>s9dl3nfVw(PM%7_uQbzjFws1`T;Cl6tj&>Z%Px{qC=lyoQ``Ul( zw=Kv`ndRNK)xP>v`K@j582GJ@%0LFn2<0SsmW!%mbe)dl0QdRR9_1(HpeuBk9^o8% z)iTT1IebLFoOiwZMv0BnBX8{{-}Zm~xVj{N$LV>?r0mX^QY#rZ-4$<${$<}!?EXO;qEb2!F%H~hnFYQUM z>69{RS2C;|%U|wQ=R$hqNfzo{+o~MOBY$NlL#PgrwR@CHy<3*%O{R{?R~hIWeRG~n z{AnBMOQ&rr56897uq^Zn90O%_oq9-5UO->LF|r{S{flkUhJGuHa*!3fQy-B`m7{f= zTBdT7ytS8dw=dW#Pk*b2vbCMaqcTxG{fpe%4))uw&R3@#qf^o)tLls2@*qds_Cwbw zzu*3vhjvn4=atiOI->4mqMcmdKBVsQRt{zLyEWcQ$revrLm5blXN(Szgj`}Do_>&L*o<0U&gx{iEpv?NBYrwwqbj%=?gb>*i__LaFln5}EvN9N>P9g%+9q%yAVD~J8+nXR_1-25glZKclg1^dcD zj`Gsh+EAY0kG{E2eH@c6`zo$MCgn^rEa;>%YIAA&bFVT;Pu}Wkj3q84{y z#HEbko#%g+qm-3w%B6lQC)u>lLw{478LKo7ql{#Quk;V+rU!IQ8tN%+Z2pVLa*qyvR0lQS1FTy^6i-2b?%b}{gS4#w*TQ) z`L(=wlx`_cV{OW4-?_?7yR@H@tMnWrCv7W#u&=yeUztI<^i4X3kF>S)$_@0ZI;U(c zU)x%{m+RaoZ-4GrMsQr(?F07xRbR_l`eDz}GxCuRNW-6d?Ay{u_mHXjs)KfthBWDe z!`6&w?A_w}VFVqH+6IqZ+xqvQe6J-YX_qVdji!LjZ zeR+cNwIARz<#t@%K>4(#Hq@T(Z5eG{TbXDR`I0BOlZ*Rp$+6`wcWHC-S6=s&ck2iB z{HQ_3Z6Y0i@@g5oZ5yc*kayd<*ZZwJ2iX)%pPaAp>Q@W947h z(-nWx)b8|B9+ipX%BfuV9B?^(Z+~cc+}HB8Uup|-QbuK&s-x;gCcRHu?geF0rpiU$ zcpLa#wULhF&~jz_jx@Ow}!rX1Ru zyoUZ@{d+c>9N0l^*EV+_U6G!0fqc3p8-L`Zt)M!tT=rd~EG>hraw)5{rluzi@*RTi zw}0yQ{LGGb%2j+$`yDs(9om=4s(ss5pOwBcXans*B2mvej4h z1NtW&z#C)=_N755q5JYtcJ&}vb&;p*=n`d@oZ@8^$r{%_3Voc_^R^S`q5>?TOhb^4gK2iC|~S7MIVysi0L>jbPbv+l%t zAnW0*_naA6^JLw+wV&3pduGO3DC?wb9dnKAtwFSQ)YiPO`Co9ZYpktn>6~%PV11Bv zE$(wIxL;abqvV{lltrE`gSGU^qFgOY%iZf-qz!w0NR2~tYI>5uHCvGP1}KXw7v;{PHlhs zq1}PcXFJrH@@&u3J9<`BZ`|t~&ZVdNj((u8lpE zWPRCkX7j*Z^73aZ4SlZM-E|}S|B#=of3z+UM_D^(eVnaz(2nCTzpa(SVc;6;BEi~5 zT;sTX>CFwSxAUxM;W-h{_<26Xy8*qk%f9E6!1GOxIp=wK*IFy+nO|!xyMD6kvz0-< zplp7(jCc>a?$R1^>&&GoFKau!Tg5TY%vdXEjiCI!k84ig9h0`+FQ%OC@r+BKcafHI z_S-tWe*T-R=KpuhbIZQx;9ZE`b?AEz;N9)sCF8wb-meJV_kADFM0y@q9(|5W+2yOe z%C9`usFF1~S%a_5$f)h$yn8(F>3(%457)MB$qDFy``zoDV|0b?uu1GC)Rxxnu|@nl zKBzbw9jm`wo5)VFMeGwhWglv@9GBO;KwfM`{Rn+R->AKIymo@EpzG`n*yoq%5Bis5 zwLSK0Ak#xWbOEbk2Y#(sfk zrUy6-=i@l(f_=b;czwv3cp67{y(3Pv-`}`VJ@BA z)*ySH+i~lX!Eamd^mUKt%Dm64?~IhTXDw@M+J4GTUbVMf7fU9}s@%@mZyU2G+E`oD z7x&D!NNbI*_nW-uTc8(@u5q3A1^e|WhUwBjy27t1=mxMs9pAGv1s|F&vp2ZEc7X3% zU(GT7M1Rmf_+9!J+n}Eg`z=4QHlm!4{|lUt8}NjEJZr3}@3v3--SsheL7$cf4k)^o zwfFiBewVw;W4Iq`t7-%AJt&(pu@lyo&ka16=lMG5F&tS~ zJjemxvk{;UpxkT$o1mVc&g?LoQ`<~N?k7j-u^XUV%8b9u`Pvwaap)G#W-DqZ9HV!5 z7>DC?TyKm+C+Vy|dg?r#tQ8zybVE?;`M> z2j5ljT^rvP^Bo%R?Do!A->31t9N(Dix9EJ!!S{8(75H|8?+y8Wj^Do91KzLbdpO|R zJia63yE?uF;hPYy2m8K-<6AwpzNOG>)mVXx9dS({SKbuLH{9+7rfT zS(>JDN(0>QxMTJ~y1s4Xdquv>*>4E>9+7L^Cm-dKwr>Ub{*rHERA#RhGyFk)h93ytpMp#Hj>B)z@a$#y4$z z-^O=yymNe3F#L{Wzs0luKi409_s#dpe51_w%KH5`-!oe-@SQW?z4XmD-#PQmGy9G^ z@4INejXD3TY0nz%_uH6``=*+2skv7=;Fxd6`Tm>lvia7g`&}<>`>yxfw>F3Q`>kxg zN9Wt5w!X)vEY7)C+P?YbJ93lx%kaf!^qu3sk!8yTM!4HG)1m6t4 z6`UUU7WPkqvjgA8w(olPOZTeaw}JZH5vZf@-uYg$_8}wRqw~#fZBHNQig5v79>0qp z(Pcg)ABA6|fBBw^?|k8hD+1r{T`1p$Tp?ILcw4Yr@X_Ft!54!QgC7UJ{eFGmJMxbO zi%l#x@vLC^V5Q(if$fT8tUR&u#HztdgVlmHgS7(t&OI+!B1kbO9-j8_=)3g3TYpn< zZQwiizY2Ww{*vJA;5)(BgD(Wiw_mVN@ZsQt!Jfe`!8-%rO5Qx!IG7o{GI&|=!r=MA zvx8>_O9qPu93tNZ^^IuX!Mr!{4a_?N-!1lyHs5RWy-VMC^PPBG-z|0w>|YgJ9^gyg zB=xO19EvA#B%4v-dEbN^ekaZQWW689H_rYQdd_#-*eE`)cjkC63e?wS-}t-z{uy4y zqc~HXhVRZ#=i~BE`Lujbek-2jV~Sm{tHwZLYqpIW+ZS^acVLT+qu6Tk!p1Mllk7ep zgkQk7z=__A;{E^LC+*$M#=G9H>fL|FxA@Vx*!xy}FU+^fz`pMaxW{?d;nQ-f(Y1GK z8h0DVyB>_Ajl;dG#k&AI_U<@a+PIq!#3$m{+46&Y<3M@k=bhN)Si{}Et)}n#KJX?1?)E*&eyiCxhw-6rv*Xt#0{rURn)ncWs}mYcW2~Xjx(}I(NuLs8kUkm)kXWtLL z7r5@^!0}T8_naEo2j^Xj3vrx0aN@TE=WL}TUt7PMe(yPV^4!TYgC7La!k>=eUEB>g z8E@ikHsJRG4*gYt&(-^e;I`oI;Ll^+GkMSCeZhmlU&p|=j|RRaigVc_ZQPzSQ!fOOu1n&;^4Gs(N z+z*3GgX@C(f{FYs9pKM(gH402gLehH1bYS_2@VVn3yutq4UP{^4o)58q=}Ozz7ZS~ zd?7d_aL+!0GJz9H|8(;Rzf$z)vMnu0!gs%(VaVtOXzrg;Zcb@yk z5iaKAid%?h@K?pV`LRR(6^{`E;G>Ekn+M~Y8rOjNj{2kwr+w)7Ufj#r$T-UQ$=K@l zz?j`w(3s4)(pZguYMhFTaU2fAk#zY}0lxBm0^f`DZAIT)^v#%lE7CV&fKQ6oz;|rH zH;Y}%ca_e^gM9+&NypZA32dd;^zplEebdr=q22E~X}Q*K=L^^Q?ijxr&ni3q_6~pV zl(Ti-zHcB%Q+Zk@ezk8)`tD`F&+Z$OzFDb{`<}c0?pwZV2mCJIPv*1MKlROEJmH(e z_?fTDU*n_V=D!E{<=z1A;GZD}-8FgF=)3l~>E-}WT^C#v{2{RQ+wpmUYxujie%~1I z6?^@)lh^*gYjHmtz>mB$kT1J{_mt=U;GqEDu^ntmznxs)9QXMaa=(A<+sEq!>j%C8 zN={n^z60tzp1%FbCnZzg3$^u~WAK}P(3RbTevh>6PdDt-AJ_RdG9AOSuBVIig)Y!1 z?KnH&4;0#*4$ues%s*h4*>Jkvc#Sws;~io&#%$svVoc&v#(Cym3^>eP>W};)oNi3X zU&8JDDRD&ae&dJv&Vg?#;0iGTu>qXJ_rp=X+w0rH#?pL6KI0z(+=l~i4ScJf&*?ks zwvPGk`rW}_0{-b=1ODpc!IMEgr@g?m{5>V$&;DodjNm!JvH>4|rQpQ@Kll~FY5`tm zf7TAx4b}^6`OA)hea8ydygJ}lZy0P8ye{BtZyLNIV9#*>Rsmn!*n&OSK6qDPf7{@# z!8XC0gEt172hxz1blHyA1j?b@D+bF1e0OzRBw$M(UEt9L{ucOF{awLr!3_bwoIg(H zzU|Czuo3KoI0bIU>0&?PKH85@jkC3>IJ5TUI}dqV+l#%6D~aXOi^db!{?3WxX>$&Z ztD2t|*D|l)_yONo4A9uE{4A!1qs77UoAYjqA$Thvr zdB<(L?|0$)!ZBsW)#`+=)lpq>HuX+0C(efe6KvXn=N1$*avYXwt~%ACs-?Bb8z{pfjY5GI9*I;sbGl!ulsv!fyYL+ zuH%Fo7r1dGGq#f)&kgwhV&UY?pVtPy7b-@=|JF|8AYx$uj?0Jhd}7CJJWl6}(+BYo z@w8`;*K>$}m}ldo1KqKXj=%n|z#afSo9fG(*XaD5SPDO$z2M8!)6Q@6NyJpl%iR>1 zFQealdvjg%erP}R6~0a9bo8&TX~5n3-Vp3a=X}cf-LeDh0{c*#p?`~ah=J();w(6U zFIN8@fAHsUFCG{F;49)){I)@W3tt(Cfrwp;gER)R(mz)k#f16X;vvT4(0wsczA7{( zJ=``9Efy`7!e1Ao7DEwN5wqsA8}B!M^2Fp5Bj250{Et99MO*;vJN8K6JRjfn97TTr zB7t!LKYFP^47$13+q>tXYCojW-pD! z*v_dS|DmtV&T1FvSj<=m_|=8q#$W#m?lUecq{Vla5C6Vp81A=xw#r;vtX^!7dP40w zU16VW3%|(-*sy}VhT1mg-6JhBBOh{5cl9!MFgB#K;*j*1Uej|v4BNv8Gq0n6naAX_ z8B<$xVjhxjr%&Snb^}K&AK(sgcVqhv0_zyeWtnfZ{y;29?Dn9*_v3LEpH!TfpNiYA z5BU6iT0F?t6}Mg`;0xndyerm?lRpsP>rV&x{Hwvqfta~C(piBR`nkc+0&$xQgNw$n zKR0mv{4rd2PT<}%g42WV2Hy@&3r-5Y5qu?JOO6P{<_k9GbAfo)p<}Qw$|z36KCx5g z8^rP1EoeNyHfPsqyN+zmj?;D=*`4jCZ8vS}fX!iRW(2Pa*q1c}Hge@)`QW+1RGad| z0#A&_P7f^bz-X*s+;vUx%fPry8?e>J3dV273Fd&=6=OT?#^#$x-Zk*uX5WIarm^#o z=A8Kt)*R9awu8;U|F|Bv^F{DDKZTy*YaC4n=_8+tf2^ie zJTqf>N~}SgKs?~rfw`-T1M^ejsba3;wI>8(yI%>u6dWFWIyf-+Xz=0S1Ht=)-GX-q z?+msH-V|zS#TIk|bT*%)t_oDau(e3QvdCN~N;2wod(7;GGD8oWNR1!=hZje)%0 zJOyu^eCy=fgIR%c?Har{*ga5QW#2D2I5;FYEcjgT`QXUlm_Xf+555tIb2|q1=LBT* z-9YSF4El^fY+4L@ZqWHZay6$1_Q~3OBH4@Mh|!3_i}Q%ZXdf{c@n>wu?Pu`&JCt$i}7D&p3iT#ev31IFHWk5#VN=XPm`vH}=AL;v#H1yHD@B zmctqou|o6O)^J$Ab6p^|O<(yRxR5VGhr144+)>QL8V=yY06*K>BEF4319ovs@l5>5 zcM@OWYw0uAe~GJ@voa1BOZB}O~Jjvm25SW?1xpA19+-Gw;*Y@< z!P&ug1LYRS|7@^-ut)HYV6$Mo;Kjl719PoUPJ43NJ%Ksa-v*Zl_}=_3K0h`v1~?%2 zP{6L>T7EtMebvA?L|kd{V8Ps)lVTiC1r3Hap3IO1GC4EW{82jVQ^>Fk|Yi@6YS z^Zf$fRNgz-b1;vo-PuYyfLF!K*;3;$9BXThrS*60Eso`<<5{{yr})|a=pA(Zn=PH> zSJPj5Y|fH@ZER>uzuA1Y-&!jEbz?MhGY;iHi)$F~8Uyor%}w$z_%(c7<8OW~f0qBs z7c`HT(D10Q>-&wxHa;v?izmc$&1C`p@Y}qWIW4?m-V4WE5?mJeZBFmUfw?vFcE<)s zk8#xGQ6t{_h4y z3f2zR3T$EZ;Kjl6!Lx%UgT;fzf<=O-1RCbC;NjrG;NIY_;Ev$7;MTy_@7sf0f*XS0 z1y=@_1s4V92j>Pq34Rp(AUHkvesD%`W^h*Eysh7k_d3_wcl^BI(%`D#>fridULb#E zyC=9mcpy;bzXr2T(g4Kdo2b%*9FR|%+}4YO*s0wV?1Z#Iis~T&zyMXXwLN+6VJ$z=Zvm1 z_xrNIz3_^_*4mnNgVzM>2h#U@?ZDcb)dSbA8kozK-jczh!9sz-?<2tjfps`H1Xl;v z;IN}t1eXNH1V0MC71aMfZrX9vjtRt)*pE*IYzV)eFTZynb|h{jX0%lxzAr8$7Q|O) zSH$$ieym9{pCR_sn7*}2=0ePs7$5P|FAgpY`0E#p!9TEX?)b7m>|d;{etqWv_!8!Y z&C!U}9T#-HB42~A-#G!kzI8P_1niL*p!ooExond)i1e4w;&~_b$=ohK#ypW{kJu(- zL}SIpgW#`MYgSax#Ru;?}Nus!Ps#u@mk;|k*pTnqLc$7RMQ_}A9&@*KNnj50H@b)RvGx@;TZX|~dM z0|$}|IpJD-+_40io)X|jvhMhS4QIb=!;L4L8;;RVoIG*F<6oWp>S$bo-=_lax4rPN za(q0%@gEF~;n@7bZ#G5yxmLS=F_4FKICR3A8oFaFV%?z_m2m{W$9Pqol5ShyV(!cu zH|yl|-_-(he8&1>1~}v|0bhKX0B5~6*dlmq@UCE|VAo)`;Qaya9AdAD`EU;Q4EVb^ z-aYWXp!>G(4Wub;dAu!PAO4rK`vCf^surrtPC)tE->(If>iVL((uH%K<12qHqmTfgu7ui<##*ELh` zTUGr}H=KL!IeYEB_CB|r_ld#sVCUe$f!akj_zr=;Yf_hZd2n997P?fC;tH-Y=i|Nl);4Mk4ljG*?JEvC`aQ|R=@h?wBGfZb(tTYn=ykc(Ak&kvxB zvCAJ1d}f!`-q2sv=2w4Vm*r&C*ECNnK9IK&AE>Dj7trP9V%5g157=9AZtG#bo}3eX zT&>^n!6yT`N-?t>7!8r_rCrO}sR>bIy-h%0rYV~%{2jFx>Ne;u=8`#OZm|peEAx{7 zO}F4%u_x>io5+5#f%gpV81Rw#$5#bc1bpH1gP#X{;O__D4%P-=5BR?SJNQEI+2Hu# zxWMb@0^k30@Gmp??0eg+367cJGaKfydc)1=@1NfA=?$L@{xSGq@ZR8E!8?M(gTsPV zf%iuQzCSW}XJEVHs15Ttdc)Ck$9VsS_iy-c@QJ`#wT=JIU+0s%#*+g+|91lZ{||y6 z1wRXZ5%AN+3cn8c=vN2)a(|0%ToljzmN(uu*dn+`fM;;W&H=s<)8MWH0{6-N!h?Kh zc3gZ#3sTR>&&8$R4{+-kuZh{jZoZ=fv8ToZThH#F7}#IjhKEP^9zTogP783h<2&a1 z;Icr?qCYX^T?6BgFJNraf%ppi24h&xRlfAXfbX$=z;3HCmY-y|ZN>4Z_8i4Ah7@LCfF^YUF3`kI|sK9 zZWRy|ZdiQ7;%kFH1iudW+kEd|1V0V<+TROK5BS{t_iqGW3FtS+2WlSQ7rZ^tW*qU- zfG)xZXaAoO;DDzFw2~(UVh6tY;{v?EkLFK5Cg4wd?|Z-F4*S@*`}=N1U^|?`=iV`3 z%a;UfJfEAbzHZTVbJ~>JPMq@PU`@a#)4k(pOJrDHml*_ORH_OhK8<6Ulp^` zTjdn3eWAlPXSmr-o85HFK;AFm%ABq<*8I_SE(y$6zPEUo4KSbSHs-%M&)@rMz~`ec zvn_maP#5i${_Ho~usbwnc4M>=*0?x6jfkyK_e?WlGw9Ff1#$v3XgZU&ff(KQez%={ z?5}p3-C?uXxN1($$v#={qV3wFU240v71yzm{N3Yc_2FKNN@r)zj_SVSR_|^_hF1uh@96v^s6}n|7ZwtK)PJ^nLRHt;1Y=)8aY3 zU)%W>>WKK=az6(KG)80FIR8w*SNUc@o0OY0uO1xi70_7uUFxglz?KJWJAIs=$`|G5 zvQKOmKlz=(u>mdPwBSdNx;WcUJ5f_kJ3lnwTdPBN40e*<@W6ntx>Zm=m0x|U zAZgP%U-jBW*Us^aSK0&rHT1lQk?WtE;CsF}-Ic+X!pw%B+6 z172X0@c{o>%@yB*Z8jF!XMXe-gMSYA(#ER#KjT(yF3tSN;D}&Vz*ej0RyQI~WnD#U z0uM+%fqWLsrzpt-9So7TuSo4@8R8^KpHU)c^1wx?y=MIe(bQEAU{xS|ABxuazs#Fm1f$QyD>H&k}l8~o6TeA*)_J5y zkKJX%y#n9T_h1_~1&6RNBah$&>uhj|bux}mZ(xhr(%N6|edjmVV_VmzE$q4Du8gp{Gwd>+UK83N%@1WzUy`=3K|L+Tq3Cth!iZ8}bGM~k`)~B!; zv{7~2;`U<$+5($KS9w9ODp(Vo5YV!J6Kn`J-(>SSf1A(kJstjPiytt@7wjY3&PKAEG#wf?TMs__&33d4x-(yx&H^-Sxk{(d~yf32^`1y!2M?5#c9d)1t;(9eEe0drNf1a;>`+&~O_m&H#JL3Vr zd-d5i#+E<6H$M1vV9c!#tN!QS)9)tamI1-9R3 z-{S)Hgyn=!r+!fW^j!ga&%fC_kULecERIz>&0jNLH`+ot}52F}K_bKt%2{3fTidR8C8hO(V(iDR-= z{5$+&?GGJZTiPZzix260;7fY-o!@Lb!f)&to5z+puIs>C1s-D;U8C)aa-QGIajw6< zKAxjbh!toQxTgGqlk`t*BpV0pqwm;ee8irztB%R1beXpz zdju~G)HbSR#Ph9Jd0lFr@W1%Y>%jpVsiwvDjpH24G5Ph?;+&726|Wf|t@Y#=7%TLW z!e`&N#>Vz?D8^QwSs?zCzc=pGqte1%kMpU)GB(AHd=mBj;>xE6YPHnX(_zG*Y7Y56 z<^*5K920w+$NW0>!J2Zl&vIjGxoNfPG}yg6F23WO4drLcCsM=N>GuX74L%WkGLYl^ zUxE4-8u)VpwIJ*<{hc3uZa^Pq=j8Ng)?R7WY69uTbZgp#^IM1i-N2ep_L0BM*S<91 z5A(12?tJ7;12&wEzACT{-&`)yah-=&{YL4o>~mjx>Wemon`ho{e| z8^!nA1Zr96);hE{`Zv)=E}PY@{3>-SeDVtxfbac&Nnm@n{dd8Y3;6e!r{?9NfXzQE z_+G%LR};pzTl=F{Y)!zI?;8D&CXdgCe;}Z{u;s0VF^A3ReFO7=U2oktAKqLscOMmu zHeN1stHAtYKg>}Y3qOx-Ve{B9c9!j`KgW;br_&?&XS4`@E`L=W1l#@LfX~X;0zN$5`c1(*g7*X;3_cS4Q}FS?D_j48;9bER0vgV<0`{Jcfiu*};14#G zHUexZ8_MskA59D46Vp4udp(2<7~pP!eQ7LwZX6^Q zP?thSr@_nr;xqYK8okf*4S3FR)oqGv9M`_}<;4wR4!-~G0)xgitqX7)o3B<~E`zSV zD$w6-{(b?w&!2xv@YtX_`yI39ilA#P7M)#g$XD|3 z`Sxl@*n9l`;^4Uf9iCRR%j|iG#s~kMU%pwu&a?4sI{&*qH^2MLfbR11;QWBiX1i%L zwd;SFTQ&68|BhM?9-!qo{x1W^{B~g6iaYu0Y&_d;OdlP*BVgBQHPvY3me_bc+kSyK z_IZJrmS4ucS5v36^8dixp}F$s%{B9``BQqjxygTJ=gr^q12%>o@CxR3zoWhKKh1ku zE1S&jo)pwy#SQ#b8j9`sy^g_8=Ns29@Vhqz?2opv4}3#5jbA9w#y14-eODh7PtlE= z4{#hdkze2V!@1Za{H6WcLEq4B?bA+N!M2J^@D1BWU&k9_4K@$2(R0}1#w+3z{J}M&a32n2FKa6u-}%HVY-ROy*Dr_ddil|GaXLAlmOtwr^TE}*^2^yhv752LS2rH` z>GCb*ePe>16w^H}5Yrh4o-4qoW_#sTaEG{oeHO>K|MC>>5zuh-U1J@w5U}gj^4NLs z*>Cz?OeII)`0k~+fWB_r;UZ(ubsCSxC0}Hhz&Z;)i|cr8VC=s?SRJUDH($;O#K!zA z@w43Oy#i~)&2=_k?5`fex_z2~nu>jb!vbq({x#qO^M8L7@Rj+!Y;^SwzAf;l`M+$q zXJql;zZj_9R(E$uz_zKq*)w=S;9PVf{;st+?4fl*?4ulS>r(2=;{;r!u7v%jnXtX= zu-s zK%Bw%*FXA7Tq1takDi-jEh}69x`1ZG&U*&IlLB#-+9Y)>cMEPE(9-z%>w~ib&z@o9 z)v>efe0+6-oU(I22cOi6cvxmSAEd;;ORB=FU zCR@%2XUo_~wvCNr7uh^MHLZpXLXG~_qgHTn#@%lGEj@|ouq^?S|b zB7IjoD3?j^;MdbiXfkKbKtG{ntGT4Z(7D-BcAPEVJ=iNyBd?B99#@V05rJ4jydka- zZ+tAEE2;fi9mp$q^_k`*2Ki7xb2=(`dq9_ZbMVFiVj1wAx*X?xb?}N{??4V&4UW2X zxdGQL?<)W6I;vTnkTpxzqsq1ZW1vrQ8;*NXP%Vc)zcjd0Fs)T&=lS#ecXr)9TN9`` zRD;NluL|(@Apvb({`KX-zQK!v)`EL>xN%@T(!&F_HSGF*0{YFJ0&$>Pm)dv!_!R-a zVf4X`J@)!b!M_B?C4a*hl|SW^e=y*e8^dfk|BFAy2UjP+PvdKuo7Gp$34RSfyr?fu zzZ~mb#N*~J|JK}P7slsWHMhpro2P7jeOf*>8%>)5?|~1`*5d^8oS(@jV|S)=nm=e> zYlFA|*VNCx!%cUX^T}!bd_}sr^Wz7+z>ntl(}37-cACDzMpye`E9pOMHOSx6enu_G z`+`2?ef1%0WNAq3EqGrnq}SNDHkzh`D{v7zE#FYToVGzD;{UVPw2U7FG%kJU8ucIF zxOQ8tAx5Eb>mxSNGke)zHjxbl zGUkVQ@|D@z)l>59adulgdv?GVqJ4^``Aq5_ZV2SN_*!Zr(d`KyDs z2K@4m1-OAfPZy_&)7{Su*mX9Yzl$^2dpd_WKs^k-n?EczQ160ES0@?u&QiW#1mqbY2mNb)FV14|WRp z_;(IA4Lob|il9FJ8L3gH-|*|zq|@bSH*|Zo=I$l?Uq7DRmy^Zyv>W4to!>gRXTZi= z+s~*OYxEW4k52QWz_|QB!M_J=y0L6*tAR1@&4c>jV&&E+nydUT&kYfy@~vo)=C^sy zHqyw=|AK!6KC_qJ(q?tdEVm^zMtQ;0bj6ta+B=3bI^F%Nxm(=oyO5;7SO{F z4GswS&~$8T7Om&8=16V#vf%H7M+aVa39Lo(+qfo(9%}=qL9H2k@c! z@BCw&NV{=Q7t*b815TyuRLjN@#Wc4m?gRYM=higGIr0blUmGtk%f{1**m-qZ>^x0~ zUL*&~z8@Xb{}z*)6Ld#D_`e0_$$tcVq5lkMEhh%22B!z|Q|$at19R-W;KG2tm+KM_ z10Sn-FFu#Kd3B&hna;`|lUq{nd>;*Xb)XjH&jd|<_xjtxE=d@lHW z(AO_4{=(d6zlj6H24V!U!iNKSiuVWd(&7PJfkO@tb`NmEBZ3`+?PkxA#0}!J z8?(;%>fqM_z5kqm-tV5#d(`h(lO&Ie4`}&#KpoCQgXU?~)2L-PzKp|j0=B<83_nHv z3;*R~f$`7B;p?ml)IC;1Rqs%UTWK_OS8=R4MUynY%r$-)e++MoFfaLd=IXA2m{>d@ zR_F8M6Z$J2;b-FpJW)-DJ)wi+2eCVwf-}Sd;Qdq^cMO~~#^$Y?U|(uu=tcP(YCYiv=g}* zv7$IZ{K%hY+l{fk1G>IgQ_Ly$l<$}K|3Dx%{rK#;k84sVz^1DiFy`5GxuvfM{GgM9 zZw36K+H!dHtdK@4}Z2hKnoc!Q&fY^Y4EFQQpxFondpqI2xM$UjQ zO=FTnpzqUELVJs=H;EJIE$ZLJ3T!Yfp4QJ#w`PhTFGq_%=tgWeU*D_GFA3Okem_6I zYptsB?-p%Gu0RYz<)clgzpO9^UH0&G04TS z*^XNsm%m#0&hPfWJdhhubI;Fb2l>7H{IvlMhvv?P^7U!<)!At_?D1g%8$Gr2d~_N* zyUyOvE9*D=@qT0L`Rw??I!}DT&eM6Sr;9I~w|N5fIy5D1;=9WgfSdrZ^cYd>t>-}+ni(HF#iY4%X^NSIzwUG~aYOqhB){p{w$ZundvpNk*967Y>r z4EV|FVC0nL2Ixk#BKZcKATPky(%9Kv^)zAzS`0faX7Id(eS!l58V}u`<|A&P5qYi7 zPY0qQ(SUFQKA`L4jcPuQ`J{mEvun_rA^Bo;`bz^j=i3GT@D)DOcGvN(K<;F!1K|Mu zrVsZGo*g_s5MR-Otbbb^c;@4};0FPXP99f`#=rOcW;q;MzkCjzPfg~Ez`dpSEDi1) z7-zeQmwuC-W=?8hYzp@B)l=FRjt1v#{6p z7c;1%VVBj`IQ|L2NrCg7640;MX+An=%ZPq&8yK~(>Hgn|mUBkH*Qe#vbnu*O#d)q9 z|4r>b+h2``&#w>lr+&uy?i<~PHq&|)d3)o5FTt;<_QLNFSJGbI9*8+<>c%7+Z#_nJ zbiTPU%+9ls)mQ3&v*GOMXt(*}muv)=&g}WT8Z)oQ^u1ShzM#|4>iPBNmD&b14s7_f zvu8;1)!ALB-;Nt>Yd^=-HhiPaI7$ru{nP@9^|AH%tneLs|J}g$=CfnCF8N>Q z7hB*OKA_y5=ZfqUsN<6(cwS)bj=GOG2S){bN*o}^cv3)DI4{5jH0j88bFqN6{pwNf z6^ILHL+lo<;(oycf*pc~1Z*1a*fqc(YFqXWUKzYTz!`W#?C{xut&~r|3H)d=L+khD z4Vp8cvxo=i@AA%K0erwux32f`fmq-f!E*xseZju_PVH&^{n~k4vOM7TI}T0i5rG(^ zbw1msR`ou?U4q(vaf-HyYpkK0ec={!G69j+X(^;&p<{y%E}hvxrFJTP!i#RBp_PYRX?_`tJDX+Ly9^*iE9 zK8kT@j)-B6J7bG}DLxPfcl{Gykk2v?<+sGo;s$dQPmJqiSIzVx8ftmM{N~G97c0M} z)x~1L!p1J@~%A^?hHvy&4@FJ>OqV%?NSD>jSX?h%tb^@A+A;3Ft&L|5r`mJHOk` zK91wq+NaHO!i_QLMYzhfiY?R_wQdJ5eJ8*h`dj`)j36!$58zjvjKAqSG(8#){RSWK z8RT782I5XSyjYaaaabV#gb$22{)w?DezkVbI5l3?#PDg>2K?gs#Ek*!|JT>A?=L1` zgGczz?`&&%1V`Wiwx8XvPCsAacJsgNd^H%p*M`6!U!Tv<7Z(H2W5Bw{jiLHJUyi;{ z*XP^aH!!cQ$zs>#`e-?PLHR#EIQJZ}u-ebsq;Rv>@hnjbktHlIyVKX_`u?w=pf znB{}jqp;Bf(sh>rcN+5h)J+oo~TJj4l}wa(s)1?b}J z^n_|w!^1iIfM7j^Wp+|-KPY& zg1u+!A2xdq&V93PfPe3N&dbi80SOqC-_t5PuEYGSGD`rP<<>IZ9g5Ao+>_|9mxr>`Ca=s z+WhK9>^|)X%=dDD=Yg{Ua>%qKF+l6{mP8L`Gt~Fc@oD@Uq7D5y81u3GP@vls{jT5h z<#EQS3GwZLtt$LpeP}!u&go~mhizz07X{UZ@C=UOL!S}wyZP%lhPI-{kksCD~@gF~buYi^#Zj>)#*ZKHjOm$+$S8Haf$J2D=S=9vd`^Bop z=+^>v{)B*ke0sotp?&aWjBEZs|C-IN|ITk`zkT*SpIJ=M7-9O_c)_;j%Xpl|0JxyB zK;wb(f_cWz5(D6X+J7QMNh>x0W@|Nj$y zkdtJ)ZWnA3hz-;ai3fHJ9ue#ksA=c_vxT&eHwFA*+Bj{Xk(`iQd&TqD%DXG=>XZ;yp-f`t|theAK-2A8vRH7Hv0Te&a-`21djt`xlo#kIKMnZD;t%}527hk`?|m=N$}ZcMZN?||1@FQ4 zVj0`0IkLUug7@9uHue((iHr1y>(w^gJor_cD|htY|Q`LiauTe$%`idPUPhz08ZUyxh?ZomQT zKHs0Wf9*``Vc%~EkflF%UW{$Njrm>novr7GvETF=Hk@y!Hjh@bB-kntudDOiZU*nC zuuZpP`yUX{e&qnw`_%Tc2V#U}0sG*Yvp7J^z-Gt=(2wZ`v>`R)bY<~_9OL0ZIYGR@ z7L^m!r_-seA)r~yS#Ah6S+vPq4)eCDC#7f5wdFN&1G|SC=tH>S$pLP7aqxwXku_rSuK45B0Ec)+gPbh8?K$ysv~0OrO}PbrJT4J$ zu=ifUcYb64ZAaI~17Z_Apq|#X zijUNT?h!m=c%HGC3U}(KZ36L^e%v&;?iSb0#R(VY|J#)t#?3gw^BuZAz&bC_U{VVt zUcl`K1mcA~0{7jR;3M2Gz!755#Q`0&@xhs?T{DOHTWa0ZzO4@A-ONS)o7miZ<+nLT6I@C3Wf z*0Jljp}G)Xmp;LUvYqTLTg(pgi)k2qWPbZXdikl*+gA*L(5%`TLH?*KZub=XV}{|Clq-7VV;Yxn3NB zOXy^HL4WBx{DsS|531#lT+kXFc^Mo~-=CeI)?CW(9TV~;LbzyG9-)aCpvIy(DJEt@;S_kOn>zuZ2+R`c1+N4DE5jNhr@ z| z-~)9t;srW7{fJ$~1I;C~|7(NVe$Vh3pU-nqp3TFat3UN@Oj=Ikg*(iiC(&5p?sHF& zxcgkpaM$FUeXs6F%-|VrYJbEHVh#G#CIQ1a#t(P|ci;-#gNMWo_(5J+yKxy_6E|2J zc32>^Vxb&ejdBe@1yVV z1?dF*#O8NtJnRZzb6m$IUf`3`d)TAv0=B8PexbOaHlP20e$YHXa{%?9>qGOAp}w}? z3*U7+zBZf5$Hp1;;p@lq-@OOF731%0Z}|m>&|`4J2?1M87pgYIrqh4e_-a0MAHW|V zN34bi@B)62JC>8c8MvkVf@2Eb`Q5hD8X)|k4e}v4i*BWjc&yxq_gn|QET8BXeI(w{ z=Q!8BpeNDY#B=T)&UQcD*VgOecew%gej0D!8F9u~FC^}iH^ME|miA73YTWTTXiQzN zWn8KO#V_g@-W9wr_;B!vfG=ZAi+>w8tWExauJrBTjDWUe9*7-&8sHf5f*1kMY%D(b zbF?3Efxj^(5Em5m9kBufVw>^LcYd>t?QvG~2I2>G|6&MpR~%8O@dWvT#uj)Chv6^W zAXX43;1^tkqu76bA>04lV6T8S#BR`<`HwUQ93baNgIE<Adl>9$? z-wXJu>x0G#v?CXc3uqsA4rt+92Xe@seSiz(6y!+v4AdUcRo)OBF{wSp2cHi-TR~jV zH3%o?xf7lZ$M({0noq_Dv?Mwb4M|R!9lv$JzO(b}J-g5T7v+I1@~k2HQMtf=lp55#@5O`ZmCq8&f(D(q)yYIN4MkrpS9r8Vl8L=rJ$@nuK zjqPg#n&;(#Tp8XlN8}d17l^;rFUT>_Zt1vmUGq-t{>eaY;X{F3@=?J%0=Z<|V*Z-P z>i^7Z&uWt+h<9i|Bg8*qp#n$RzIvHsR3oDkIfrvOr}&B;ul6PO!pjTUbJs^d za^3ns?bfIv=|BBRGol@eADTO$6NwXPjI^TK`EgyYdv6@5F_0&&jhBm}8Hpo|B{6FC zBI8Ug)QuXK-wbN|s~decbvEoj?Fa`L(}mCAxAAv*h;jwK0Gd(O z+}~VrzV8I)%xM7*(iT1#PAYu%z2CvM_Ob6mz+)qvYrHOPF0Yxt>u32BbNb5Q>VVFL zqs;r;1>z~adAqNN6oXJBscA#0XP3j!wJbw_#HHaTB z4*b7O;s@IJW{WnPvvcg-cEN*##td|JJVAFqFgP@zx6obaNOBGQZ8lb}cH|0c6;29p zg`6&~uwH@gudat?gfHks?72L!I704$od@rIhfi!{Tl?6T-5>QL`3*I%&L<|22NxI6 zjqrfA!((jVI$jshjPQW``V)i41i144f!Kh?Mkm6n;)81f{U=7SrUMtB7^w65Lhz{o zcdHFh3#^U^2dE?92Rtzlv+@(fn_^6{v3qYEh!>0vx+vSvFESU5P4kJZH`myC{uaMW z++dEf@1tg9?HIehZ@`{6-%vX)hEOZcezO&9H+#*7u+MCBZL+!{cDeAmnA+>{`3-Uw zwbfmpz-EuNrfeI`SDDWL>oc81O(}o2b*1dOIHqgV=876P2Q%=4N6@y|Qy4X_F~=bO z`9tvg;L1Q8G;#vFeM!KUUl5cJs(m#UYHTycKm2#U!<0Ap^&@ad`Gd}Oae$9TzR=eC z{jR5ezkV6l=IP(A{Vrc@nc7kOif`Rtyi7BcZyNInI9@%$7z<4Kp>;(33jT#L_<`A4 zu*Ly>R)bn1oFL{sAz*9S)E@+FHamWHKxbkr8v_{sG^dLKIb*qIxncHGOnuF)K9n8J zZJw*sWKZko^Yshg6~5#DPjd-k0bq+;N7R@=EHK6Z)wM<~Ys@pykmy8gI6tp?5uHh# zFy<53_A!tAv{~+;F@yZU%Aj$BXHU={4hgI?Yc0~d^MB-hDELI6rtx0`&pDuFpC0@; zkOO7A)S=3Yj`@Yn^Z%+>M`HJu1osWp*6@w_$b98z2c9$cvOw+`FUT*5A?Pl+fvzOa zV7&r;iS6}FYg|F6q2@}@PTtWM<)`TzC1vrDv2k(97H#I}e7qI{8T;-C* zDJKSwi;LtCoJUUFIh|Xcc}+0J2XD+;0x=QJYpz+$^r!$g>W?jhy9RhcpZOCb=uf%) zbwT3<9PGJq@;qt))QX7_#Cq1*Uwq&CB0j%)FDK2<=R2w?|XtP8VD_m|I9XC6I>mz zl?6W=YUf8Af5ngoXLIQ+>@TqQ)xOwg_L|+VEoax+_}YFskZM}Asxc2DUqI^t?`b~O zduU+v9$J}mIIp&_{dmdsxJEoS+I)S58}%K&6bnstA3W_IxOeP5t{1miAKsjTd)|0} z_G3KIe#8W7(2Ny%1@U8J0ydxjp>~w+Bex(1;G=ZCFaO%uzk z+v(33r=7|t$f2>pY%iNz&4qtlO}@6ac9yN3f*od$?PFgWt$F2`j_n-IWsb=!;|3go zGyLI~8z$hZf#3YuZq%H*4}LOd#r1y*_;>j04*@Rw-2n62Z?+rtF72ROO*uK*A6YF}`M+EqFe&fv@bXTWyJKYS~YgTNoZ2;@omz^!MM zci1}MH{%F#1Af>)puh9A;z|E(dV8MEtbK7ED_=i4{0%ntEE*(C6u|IVZ5i&nd^Ux5RB&t|b- zo?+u=(?QsA_F5j54_@2M4`-XbvemxlpWB8lw>|&8Hom^R-xkt$*!Ok8`N7o2v+r#3 z)V|9{$Whe3H`iSoFCTGour`p3kUy3Skb{tOkSCB|Xl`N5Ny|gXQ`p|VCk2k@9L|el zv`4#e&j{bmXJF>*nY{NMzH)tdMPK0+eJc(UN9glDKgfM>&%_Y!i~B=!bFXM|G&%Rb zu>?&Iuh0O+lX3~-2il>y6PL(i;StY?#v@}*lK4Rmfv;kWb`1h8iH@{3pvja&=qh8Z zur6x~3b6zDY#zuZh#}zOKpa7X5nr1tAf^yo;3M1EhhHW3Hjl(8VsPh&#w~ucjq{07 z9G5LGpRxNRXZXynv+uTZPVI3{`+McnX@@r90^2)AIZ<3fE8q_nau&X$Ay$h*tGizCEQIALiZhEUJt4=3PMoS>i8AjvV{V*hXY z>Oek74H8~ZGl0WAlSq7s8{`+32kIZh4oikSv@v8nnFo9@+NF7=21KrohHLK0LyH&a zz3Pzodbq-T^{jv~ZlEoxJ+bCgjsS18j&+wj8yIi!HR%iD2d_S}6JiS6$YBG2lm=m+ z#v1GjU7M|8d&&{GMI3Nb94tJ+|1W$V!DnXUaEVxe&BF`q-{k>Yh(B=22)r`FHqh_z z56%z^u&;b~K0Hp~*N^o|?010^GhTp|)%_X=W?u&aQS1c}U&8a#h z_kZtzA5lBswPrL){)SkA?_qp(t=X~pKb%{SG-^b~<+lQwIvaaVAYNcE*-n0rKmKsz z1h$Ut;|q=OUB71^*-2=vDBoXBm%m*2?EBhP$L3%2z1d-Z8G7;jZ-(~xp3&fo@yXd? z=WcG9U+&!6*JC!uu)p|+&rjEKo#KPv&)~i9aEsrs4fKP3%-J>LJApbSCyme!6kPiP6K=&Ug;$xxu9Y z-CPapor0}{odWhy{#71YPFB5+=O6K_KNHZ2Q`w`YLmnZ9|^<@a?o_Ml>seH-a*dgLBTcw4z>P6Fey)?|IP`17~o|6jHlg$ zu3f<6FAm(d6~VL?$^CZ!jURpr|H#~;!_tjBqsv^R-OBT=3d~{inGdJ#h;3gMu;cOs zG-S4d{?B$a9^f}t&zU}Fls>~o@x|FvHjF(R?I)YaHnNf53fRk203VR`L$z&@7gZ=Iga>%ZD-ekO=r(*%SYe6@AXH)ZwEYcZg5_( zE)cUc2Eto70GEkx#6f@)^fQQw@SeV>gW(6fh#zQ5{QvSobtJq{pTD_gT9SKCPZA5# zl*EPn|E^6iMmAPcL{k!T(v$cTe2eNyG)iMn9zpGbcmP-MOO6S|1!^273_%^Ex!~0H2#*&gRz6vd3&Jdv8KkgTXu1UA+1Xa(HYy z8;cA4nRkxC|GzG9O!k`{rv*6|`+j909=T$O3$$4rBgSwoxbT;OYZH5T1>cKVE(pXD zm(5}b$8>ygg!5_xPAC_+Chc@x+O7S#4mUJzpqJ6p~_N#el>#@{Y(#~ZoHyIpMj{^g6GfHJkuiZ z0Xw56$ul+S73z|jUr}*66k+1TWgiXb>#+{;4=c;@OQzeBN<0G z#<>15Fh_8LoSwPV=Utd<;(C5tIe~wtUWE2L<{RW2&3jrxpI6cK3}dc=JyBCMu776# z<%QY$seP};!=BN3*f#z!J6FAj-DCIp&`|q7wc~94h}w4Bv7z=WdaUua$F=QvY{x7= zu*(Z^1iQ}eiy6+GK@34B0&#`-f_@~{Xsj}gPsB_`oxJeI2LqisLU7uarYv zi|Z73=v)0PzG!SgdvaggANR@qlA{(EmN!PtsksFF@80tV_ysgUIR&+9%~h*KmRqnU zW@Rwt3gb*{Ay>UBC`XJnO2+4>Cu^N)P~zZe?drFadpIMY{ht}|gBFS}_}zS^%LDNQ zU!MP6?FN^?H3RycIR&=GJNC1`m_$6_n2vq%ELT7e6Bpnt^YGjNhlvUBnKm|Wf#Up% zoQ>*EY|XOZNdZpbqrNmaD0qFqMsD6jxXqy?-Dtu_~ZD`9HMTSGpwP&RT9bK8tuVnV~!f< z(agk1VkL3K+XL&Q4+zw|%17)PJUqB>VBH|iDH8Eq9n(d5Ql3~%+}7vh9h?6T0hiO9 zmez67aLdM|iR8BWN`^a=Jc(9w&B^(`A2w#?#u`*4N6{7wq*E z{mrXf|5$hAn5{jc8#Q0d#y4l&*ueF;F4w6K*nB!tV+6Ke&cVHN59vubp#S#_9#Eg= zo_8%YUqT*29fNqYxd`!tytB0~c){~9tZl#z#ueYB{4mB4%~RkAaYUb?#J}|*s9wwzrzC)j(o|4$Q*26;w4nwvlLq-r#ye?Rs8ZDU(8ih0N8 zIu6Z;1|_Di{^Wf4hkY*Z{4V;}ly~@gbhC1gIn2HT9s<{ej|$g1!uPhZZF$mh^#P8< zckKhQmNvTvd?DtMHxaMk3^iQzD7kFBL6dqyuzNt4;=|$&n)vGiwo1*VSVRtWb+9r1 zptXxZZwb1VT5XfMCYryzu>7#PRCx;P6)p{|U9;YS z&o57L+hB`8u0re~KG-IZe-MA{7|@$QECP5zjzUb)JcWEV9STp#Q{W~#6s{2GTous7 zE(qk*o4=-^$+O5&;K1gupy(6wqM%rhu@RtebEd1q#K%)nG49))P zp{|v$E!Lsw^SxXTRHSj5+KtR%x8Eu^5AHwr&<75zlQ1{5h6B#gop{yoR=Dn|5lmH5uw=TxYq2 zof3!a8ypZE9I#`!L><5UE1$T}>Nq)h4z*HhWUaGf7p<|Q(bMW__OyGsZ?^PF!!!J> zuX{tlPgf&lO#&T@U8euYamaa#Mfm!13A7^hP;5IJFTWu#T`sw2{^w}*QFuXYLVJSk zX23B^1Dt_(#{4!;!cF2FafY>bp4qf6prhfmYEx>sv{jqm8yp^ptE{biLGZMovBv%K zToSRyZ32BW<}K8kjj_f@Zt;=1web3!uDCL=PQv}6MV33n8T3$N26ynS&J7x4;12WA z^UU~i@{00|{64t~ejz{6bJNUuI<)#l_Mp$?>9z~K_I%ETr?dsM3#auO zaG~o1?_n-jHjo0>-}M~&MD9i(fp}vYTZmC`us$7YQoFXT@dX?2ez5cIjo1mdv-fzO z&3FIAfURd@_r(-qMjB+-yw&!rZJWjsIH1p0;&-TR!UJmNtcCcept_RvZ`Nw!gXXK{ zA8-O4MJ|n|gcp7moHNr-_{jVpnh#%`Z_d_D{bFMtPtfz#@vyOM?WpJTmGK4}$3D|@ z&6}=4lhZA%IbzcRzX1DgTkFXG|->sKJXt4HN1uI{BB#Im)X~Sa6HGy z(_$7}&8E=w+zV~he%Isr8k;Oh4yAe)E?E}f)n^8K1+Q>@MW8HW*hm2b>w(;qHyZ6SN@$ub&4r(6Lip_7FV&2Plh)1lm!z+zPs#nn_ zJhP2nReq^fr5={vU7d;sP7|p{MI)(B#gC<*uya69sYb*GZaIVZ^rD5pP8NRee(WfS zg#f=*W9s|yH~w{XC)!gnYEV->$UeY!LqF4rihk$Sd7X3Gwzi|~w6EiBtUcO+H{>6R zsWygdHil|k_UZnP*Xwg`ZPl0V2M%)2+()@@yyHHPyhF#LVbL-B>{jbBX;#J$txCLM zOm&?#uJH_xa!u=K`8U<3`fNX3GuB42d&V)_$-c3vbRDtCs7HxO*tz=BY&Kt-e{OE@ zr>A*uIt@Nz57~e9vtFmD?{9k?R^Q)oM~>jX^W(JvPn0A0c+HP!v$nei*FtA9SNVS< zN3iSV27Q1BXmsW?;9`*5D5l@h_KI%P&tkm3w;v7Aam9T&-ucQm`!y53n&DQy`-PV1p;q}~&g`O)&7qiaKdhb44 zkS68%uj;pGWjKjeHNtm(vyE-#$!TkN%=4e*Yw*e10MFq&=U)?OleOt$EbF{q9cVu- z@96<;PYzwK2*H^$v_zMgR~?Yg4DTwE2au3*_53r_i{l{(pV{ahrbEJcF33aO~P+ zwwMhTgMqf;A@*3iNB$UXvpBBt!C1$tZ<>2xtDAd}gKn%a<{#9w;dF7M*g?O`MaWNs zm=g95#1!(??vdDnZ@}I+PvIU;^AxLw9Cn|=CvC( zeX8Bm-ZxhhEw0Hm^6~M9>t~P0I77aYtrlb8 z5&fqB#0?{t;1jkU^z#(Hw~hWECJhzZ8M)UtF@pH{)Z zmUIv3lXOaPrW}}Dn4B0N0sqi!#ql%&8UYY7N`i>qC~zMxI%8!@ANmUsc*=s$cTro=P)wj3h|KJtup7uNEWYt-Rw zEH934+)K}P70-4}hI`%rC&Dv})oL^^-sk_}AX=B$7!RrK@Z2ZxECO0APFfp`bwy&7 z0$1^;`Pjza=tr}kQ`=el$u3vFVXtdH`Pj9g?C1#JH&&>fpXy2cb>{#!e<59|JW$)u z|JU|0PH;`Gjeek>%Ju69{h?p<56;$C_*>tJtB0h0y+2&*qqp4bJ7f zVnglF@7g7PY3zc#%2m_2MgQpEauXh6Gx!KNhd%=_twatS&2IZWk$@{B~!Z{BO)E~3qG`HP?@{a$DP7J~3;~p`E+y_3OUE!dE13V!w zf{XBmd=E|nTmtxMgzx-rd;8)td3nctRY31^zI_7RhjXkclylKWc^B(O@XmvR`v+nh zYi(R7j>Wa|_|~9b8R#4Q^YcK>uo^J^=>N|9Qh?Lf1UO!fMqPz_q0U0Rg&1_?Aoo$u zhR#+FqLZrSpj-1lz?&H9t+y7j{*|9=L1$Nu3V zJ~)4z-7E*Ot!!+$2k-Fv`Tt#)KGu=3ZrW}K>^%pd@XJ=M`0OqJzA={=NSq};YEGP%ZGOpH_^X~r-1zVQ=60C;<2X||J~$UN zj-%Zz6wCSUe}Xac8@G4d!nt6)&R%0ROR)}rm%X7u%2Ux{_`*vAelafMBg=uQbD*)Q zsbIgH^VuSm3y9wDUM0`%U*`*V%nspzra6 z_Tmrsz;(DL*H^xG{f#$qp?=!fJ;ALw7T1o?x5vq&4}ibR*?paUXZsP3QBD^Z7Gr$q z9=8tKm~mgLL%Nslq3d#e?i+4(-}Hg|B@Wj=CDTl436jF#_F^`0BhyWKH z9mu^{qiGF4uCP`Uca(c@3U0CfMopQu@oEA_{70VM>$BfHhtan93K!uq>s@f09F%j? z?&PJ!P+})_c4`UKr9CH*%d#fuv4I$A`@l8h-7NxjYW}RF6fdglv(5+)(J(z5Q@?&W z!1MSX?|(SJMa>7{qJ09s11@q;-Cs31TLt1_eu;b#KZaIoE|@doY+OVK5wBN^qqT^E zXmNZ{zG-tpWBgp@1gU_zA}V z+gh&aZ~e~p<($S9)vvUvIT5ytpWmDaJI{s{_3M4cK|bT8@__wufoo?AUAN=AMz8w7 zckPeH9zHkj@anVQy08Ay7p}pz;a2>CcX2U}VE^fb>NM0>s9jLYATK3fHs;D&XWo37 z99Qe!)FOC4zuIR0Iluj_+mHFR=FjBST5HHr_qQRqVOEEJU26B>FM;pX>#56=mveq~ zjM}Q*a)GYJHMurAC9V`JyT`@={?RAKjeczngSgCn8J`y~mZBNs8*!8IOH;-%e5}?k zoDf|MztGU=QFN#=XSFV|5Zy@}WUjE=V~&H(t-T%n{?^K|xoWn|KXt$$2H+bOKKp*_ z8Eh+7u&=fA<}DqarsX+b^ez6R`UvM1d#HJ1`?W{R)Hum|ey87UXM6J>=po>}?``AH zG3<|9@C+Wpdm~@+Ut2HN{E*M>Cp>3Ad?lXp=REpItpT4{T;v*EyL=E%8tcxj-H=mQ z9>@>jC-vmC9vro2@S@ps>tB{vKmGPRD+W)|?Py4zWu-1bj-&ry$=Uh;tu6>~ z6J07M_`hNz@xgt9Z3DFl;-Uu!VhA>zPDk@nCx)-YN^HD(Yq}V_F9z8=z+K`JIvwsR zZ^`$_19@JTwV~oBoC4naPCSKUY=h(MgKKDZ)$7Dz;xXrNF6YFVYT49Ks5_w3Ef2I= z+aDIt9JdM7N|%RjnYs*hZhs1X8{ls1`o)p7&E|?47vZDN1^fW|sOP{O74Q@A5}(1^ zTJbA>qS^5)+_x75xXS&MKjMdoiPbNd6MPn9-ME%hh<2!n;nCH$rjM7*#=r5pVYchzN(mv7N|agZK2cPM7GBJ`nPP3?OLbH z_H(9Nay>3mq&-en5sHtc@MQbV? z3wJp#l*h(;y4Jlr#=R!?~_VhezHCm^kHbdWU6E4nTJtU8{U z2ye;v8}sEZV_x1#OjNrphGCZ*Z#1@O%;7ili*2``d7)0gah-!cjw`i8+s%zJe?)UM zj>lY3wZ}1rYu-m3=h&UQ+HSu3&h_v`%|)@F{q>W+(Ra=b%{BQ=pU_fpxcQ8iT#xIc z@8QC7khrKEO`F5f_!lSB=WtQ$*ZI)6NNxcS$s5Ue)9K`n_6pcETAlpSYXUZpM#L8K z(>^cmiLj>)vMu@4@%-O>{q8ChucEoPysRhsL%L#BRZBA}YTq!P6<6%9N{uCRj!=}^W zBlQb7Nt}e2)MmR6_=$Gvz8oIlC$a5;0Utt5gBldRM(gQlwPI#|89mRqsV^p;5@++r z#N6{9)nbvCI@fLf9UuZRSGCU)`W7k?|fiuJ@v@SYO^+vH+b6V;N#3|5P zHou8qif(HgS`&VuebK1cW%YsO7`9s+BMuRF)b5vO#1t?$xlQmH5{mrSpYOWe_&UE@ zXnwtrI0k&T&^EsJdwC4U;32V7pGlxD5YLEb@XT1B?Yd}keeQr5Zdp*Bt-K@NtHy?R z^f!*iIjyB|Kg6Q^0GcH|Q%ueeaZlajat=R(Um@4eW*b|4o95x!*oA7w*kQIG2e703 zdi+3#5*PRrGgwE;N2|`nr#DB~RJm7RTbItz*v5B$vz-{nehaPXaSr~U+e+sWOVH58 zC9T72Ena!d{3@!w;XYj8I?OvU9o=JH;2P^omV0m%Zq-kC1^?jP!e`&(CwM^VXW%wlPxc%^e%pfODy*r=?AweWShdK;@t2dt9d&4_AwSa5O%r zxix>+IzT$AI8wifi^Pi~KUGI{U)&=cC66YbCN63&Xsq?)i%er=W6wAgJFu~AF}rNM zo*%HQZ0e}Rv6D1ywh~wPqh+z1Q2mPrRvpZ|7%~3V_kFehh#uEDoXdI5G2AHD;OpTh zcAP!eR`z}5q3UPmqF8OL4Ro!zq_yB)ea2I^!!dxn^oe8OX}kkI`yLz{Y}?nq@4mJ# z?BDti`7mv6U$Y_QB-gJ0_`Q4pJ~1B0>+~CO6g`i95drU-9zbpnxCF z2GWh_N9dMrY;U)2u0liQXmPMYzg& z#aP;a7jY!cS{BHUsZGbHYSp(6=#+9vjiJPi{^Z*5vYK~Vo|3+R4?4 zM{&LSJ29&IM(d97RpTfzl=z9R%75TT@F(yU{{mN8i}aLWMNmDj`6rxYjNl}`j5%c9 z7*FPkd19jT6?>@PdKEEsu* zy=J%BZ#<*+n|_5`*nF`DF2E=FLaZbv5ub=v#4u+E;uhZH#lbopW%YQHWvq%1eXQ&|4s0_;P zQsTJgfEqWs4%dUz!281Ik>7Bve37_{E{U6Il=xaaDXtV>>f6RvG(Gpi{c+F4Q(`H3 zBllX|JAHm9Uxn^#{Lp2MDdURwFUn z6|R@o=C@w6>-N5JlfL0s>L*<8)n{|ozKxl18@|JBUO^kYvH{wzo?5ay*YFp3FK6X@@E)h( zLA;2g)}_|3f}kg(!^KzxAk}OO=2qir0>L*X9nsQagx{*$J6;5Pl=_} z;M4VRk$d#2fG=T9;Mcc4P@Ka~s_COg$%PcH3G|s=Wnbx6cu1_n{)#n5JMFzVXUyHu z#%OnPKe&kf$46d^>uA&zTr#n*ktkNP;{tYeKv+_7f7`D`0JHP!v(QVMps>!G~*zIL767jxk* zW1T$)@1eOGzkAGrjqW12T?zt_TN=PYmuj?fiknz^3u9tw-{#1@*`LadD8G2YpFx7~h?Z zR4c}IvZ3QM;%IZ~y{pTK8J5f-ht$vP^bUa>%`O4Gj20&s)HO$RHu@V*qO&zN63>Wp z@WiU1yrWjma}m_z;hOgZxa#o08dvq>UhSin&VFMpgql9}4}MqoSN>_-Bz9^JOMBWT z{3MpbrMOm3%(ZP5(D-nW{!jx?=aXy0M|8gapJ@8f|4WT~)PakM=y%O4c|IQ*s?DuVo=EPfpv|yDbQ(IHToE6cK8L&HyvI7}g=(bBQ}XG! zNbI9NPyB^cLkQuDyJsp%3gd1dcj9sEOZSn#eu%)~CzjnqG=Tb7@vE%D*SQEJ9%PM+I< z$HZ0iJp85pSv(-;FTPR((Dk(JGks5tC3a|SIo(SR%DNtWNAshLf%m>wTQTOT#AWI# z@PHZ%ynr77KiF1`LQA8m5w85<39+45yy1MdwZ=*;C?7;$>vhm4#WeD2^h>pfI9rV4 zJM{~=MVzU=UTmW;#iII{)!d3&z2+aezvoJb5DG5+jKiz z=UmS1xVX`|*g-zJ^_ljyjbrtAj!9o+BW-U#?UM(S6Qe`=-Z7lhd2Fi>`0Q+^>u*kt zU2ScZ{`0+l(N}mK*Xw8d+Rv-LcW>BLnu~3BA#5uKV8hr|{&Dji#*6RSPv>G^Yk%3h+BmjtlUr;uXQPZcHm+mL@#w?s7<*;`c8(9;x`O^D&bN*I`0KV&ufYC+d^(?f^xu7UeA>0|ZDZan4Sdg* znkSBFUdeUv_4szypjY!28}wXkE_-aw`QEYUR1Xd0P2^CVx3G=AX^zP`#TWBR*EVz4 z=~({czi8|23)kH1wjcQ1ZM0kec=d;8_?PZu^)~z95%H3XB*JOX^)pq@F%;>xJz3VDEzwBop*R9{R*ErCIjz#y%eU>kE&m7PGj%7RJ z%2;gek$d4d#)IFCBR&wj!LKv6j6vIV%sQt(pY@mVx-O`n##T9ZZG!RQxco!Ev0vVc zx2kbCmp}XTo*UbISGLl2Z4ckF+6mjJ9rkfQU8n1J9^0^&&cg@9wZ^|W=34cU?;P8< z+N7PX)t~EfPWF`_tsNsrYn!&{FXz-=$8$dYp>3U??H_R@AD-Rkqw)vX2(Qksowfh= zb^cz9HUOTe{)1n%-@fjZc7XTx)jr1r{jMEtpYQAkwrhL!jj>`5yC?3yYjjMu%xB}w zIMSE;(LTnYvFRQfb3S)Y8RKmB(!jl9E8HuuzBg~#VDHTn@ZLOR%h>~S${goc^S^wD z8)yJt`P%FQPH{Z?w7JV)qZQDv=|bv&_};Xl?q^@``FH$0$HYIa0i%Cg3rK(9yZijK zfF|J;PuZ5u;4|94{=U!Jt8MHDU(!C>VgKcU@7W^uf{n7hV>zaFv0dzm`Oog?JN;~1 z{i6N$am|k9vsZoX9vEN7zH!Y?xesiQ`|CcsSI%!geeQF|vfrJTZF66Z8yX0IO5A2l zxbNnlV{0G##5Op$dF1@Yo$<@>q4o2(>~D@5v(9N>Hj9lh&YYVqVV@ksc;}<74fs{& zrg^EpiodBY%9z|Y&sU}GS+iuE@a39Y=7-Wn`BVHfF}8WFZ(P57r!Sn>9CYs5Xmdyo zv)Z$L&0qa&ANR#R=9%`ILwpeT0)M!_j?1r98%)EIbKq0DcJZLNgoddf%uD_2Ub$!b zlkLz3v4FX2F51pD^4;}|b6L}RY#=x1e)ygLt^dqt{mtI#KgZH0_rcsTo{e#1)f_U$ zj5Wt$fB3)JtS#=dIcl7CjNnsT*LjTt8zY+s#+bg+PtL2&?zwxVrbYbay?!;eI%f5s zz9?V%t#vY9@vMGyKJ&pGG|yd!wrGR4)$X`wjn`{0%(Y&xG3~gntMki!V4v`sfFYkdxaZtbNSW&+Ql}wKG&nY zY=isDHrvPeV84wO-)p;efn&QD{14-m?KjT&2=-?;oYVg1vfpf@4ccV<+s^$nW*y6z z@SXFz$J$^!bJKWsJbYsuI*$F^7ss_9?qYvzXB*$y*ZB0=9CN(3$2Ebr*-o3yKijh- z+Uc{l;8VZ5R`)*tw{`oC6Qq-nHmo=h8mE+uu2AEA)py+t`*J zVgK~Ed84iFx4+J5+qv(?n)kIQI9%W9gWi9i-5>T;o3+b*&>o-XPCh%oKF}UxP#avI zYjyqYH+^L6+Rpy^N&9@S9qf}i*WX>QapwLRzqZ%D_NQaGc7NL9IE|CE-9E_jj@7-VHSKruHdt8s>Xot4d4jN1Prt`ve+Q%5uUhS}tKCusY z@0#2v_rv-0wf5*w+qm}jwcmYTd!vo^w@vr8zdy&;9;glRy<_=o9@xfr#)h$@z52~I z&aF?hAM~@|Y@<)iLHCMX(wDZ?F0Zz8AKb@sv3sNKj??41Z@%;Doca!IV}I?^7d~sV zesZn$@w;Q%*YDoDm)c`L-@88WyYI|3zjqF}pN{AEw$10-O4s2U>}wzW;CTAn9I%~s zw~yKfzPFuM-)WnEE0@+zIG62QSC6R={qBBfuW_Vp&gWj*);;sR-`N6vqz%repTM=b zhxT<1^)Xz3$Gr1so3Z0s+Xv(QcP`i7esCSmXFvU>z0T{k=W=}a$8maXezzanV6J=B zuRV@_ay)I)Pp-|pGuE`-zWyB3vAp{1dws#~^xn6BT|YRk^JtT6wXgovHorTD{lR!O zcJ!xxI;LE=KliNH>O1GQjXrYy;QYPc?d#56zqhZQ!?7GwJN((s{*I@;+UC_h-Cq0s zIhI%Znm4|;uQut6w%IwrpX0j5w%0MC?RGA|wY|>Uz8in@yYFqQjcup%`>ns*r(?&f z-?c+qdmTN#V|nkK_HRFU)jsX4?Qk#JkKX&Uk9HVW`n~6~uQ33&>oqy9^R*wftH&9C zv!8SOvu(H6e*M}p=6ip&1Aeow>(U<3AHH+H%$wf#?%Qj$O|Qp&bSc!#-`F z@7w3yr+unzz4qQqzxnHVI}d&5__hbf1>5(1U;XZ{$MEVqkVFTttQlH#E4o=7x(hHgQvu3>DFElQ9jNRE9#C;wps-4Wf)u z(bZKwp&~_dRL^<(f8OVu&$a&N-~I2;>vi_tYp?bFe%E(eYp=b}Z<9wKYCEzj$4`4a z$>n%#i!b}x@Y5dX1Wmlqf{{O&(XyedzPeo)n(SdR6dE4lV zcGa1EX|)`0H0)<3o%JcQlBsfNM_xKa*LmqxXRbT$xw^{KLy!IGot*d^r;XR<0qWw9 zC!6pob#3L%kD+!{zjC6fp6i`|;Gw)eg%(*Vqhp&owT(K}pY~|lYM=T``IT1nq3yUi zM~BR0Q(u|(w(7x{Om63NuK&$_^hUnQj#lZYTfO4zT6Kz+cJ(WD)u}x6kFWaLkq6(7 zA7o^|=4$7S8MO4d^RCL@`H()=mUQM=8G7_kzaWotwDd{2JsE4C+87NoxP}J#)wj7e z`W87{*Dv(R>pERkU+B74o#}&i__ls^rcB-HQ$B0n!&^Doplz$G`iycLdDSOJ^KtWI z>DbSwqYgAS)B*ic-toq(5-4V^KDmC-T% zns?UU=o2mcZRoHk8;Aw=Ys{cs|3C}W)i-psM+7#>Y zb%Phat%HUQPy5P(e%mNlPan`z$6kN%YrNvA9If_4UG4C6T^n`ug^uM2{(zUd)zK)I z`fWezj9lI719+>$R$WI_hfkSXn>*lD`O9Zk^Z)RZ+KbH971_wBZf&l8W$Sf)W24bl z)x&-DpB~Vxp3rn&8tCamWx|h4?9I;{SEuEvjSXL$f@Lcyw(< zpUDN}#1kCjtK3Eo$LP7QxBjEAWM~`p^_zdy$CSBVJ>fU2z1hFjO?j)QKAwC+-@q|m zP+4m$ZSl2UWgT5EQYLR&OVD3@$zpRn%Hy0J)bBkvd#GE##H+R;ADOIujcZ`!D(&_` zJ-SerTx2i~TQBe&#WP>)0r;{B9%QM`oV)LJjm+rjC%)`}C)x02U)QYnf&b&97rJPi zcb(NzH_1Y_<^sA?*Ir-LM++RI(YQgMPRWO_zEltT z0QgiVj+LR`96&b5zA$iv3ISxhiuyD1Gug|`Se5o_!mB8u%Xw!w2ZEs z%k&3b{lj11@PqcX*U`hj{c3-zNAlo>CwgRWee#hNUH%4LFQ`2Hy1rN)(=-0g?QQs= zN00c~(AGbCp$DLAb?Z0Uk_k<4+%kQy53)0mmH&0FtWRj+V_$ud3;*h+HdYUR<>azg zrY@jCetWXuKdbp!-8)Cyo-N2hM~w|OBm-LXj21fDkY79fbiI7&N}pS1U*6Sm^`gJz zF(+t#&|V+#Q;%F^k}KFwU2VwVdijyFwA9fTGAb`Wa;ZAQ!o{ zXXTPr$W%L{S3koal+~V%ZMIbpEp5qzPIb`!jMvm5gF5y- zuRgRX(4}*Hn@8|tQ+3#aZuu|erY@ZuU(o5iqwSUB;ThxjKX;zB zL!Z@=r%-16BsV(bsy^E{eCU_%_zGFnsejNjH23H?ImxR{n;^9k~gKBP?F%(e92`#*jEg$*kMI|Dn5N&Pa`q3`*+_cy#R0NywA zj({@nC3x4!duQHT@(!h#B$K{~cRJBG{!Ko)MeMO9*<4pwye|$)8z21q|Azky@V{H$ zN%Q{&_RhUC=Y2DEymRF}3h=)L+LE#Friep)=oK%qiXV9$dw$cG0f$WxC@_Y$`T@uNphrXFjLA{=jFH@ohR$pI@*MA7(%HX-q0Nx97*+ zZxKh0vA!4QeYL*l>HRC{JB;4>1n~fU2UPs@ohP)#4*t;hwv;(;S>p~J{Jh(S5B|#d zotPAN1wP)nL1&CZI?{i7qQmBCegyo8j_5~RY1BVgui}e;$HnGs<@N8%M{cy5SJcst?}H6pTY1d(ZkE2>(lL&&@m2mN@6_k` ztvn!Byz9d^?Y*}wN4O4rUp+8>8JE!Lm?5t@$A|xlcRnp9)b}1Kx{X8cn4S~ZiwSY& zI`sW%Wz7-ZH9|++h!eE&X`kc?eFc5QA8+m2mdxt6F4;%9t0ULNw){uW&4Y3?&?~#q zxxKpDK-VtVWVD~SENV~w=U91jxv`YZwFN!}d;`CZ>G&#ZZs4!tS}chxaIV}K;#}-_ zcT#MLCGoC4Kj!OV2(*(=-j7{6lWlg#FQ9wUY|GSIJOrn#twVq2%3#KyzvqP zqu=tsj{ElH0zOLCjyKwM%n?^(jED(+(MDVTP6zbCuj!5c zH=%)6b&dznE1#xgJjq)9x~46j<`lq|;sjlBKqi|ye3e|y&GKexv#WFU(Z!>=iayzu z4#l!-&MRN@Y4z=#4E0m7!Ot4gbitSTrak{GTqB?Qor5|@t910G{T6%l-1zDkDYpFo zl=tGj-wghb+5bX#PaC`|u1?3({vSl+wqvMxax72Si)V1&_npxY_xO6J8ejjvfvI4CKXCUF;Qu!KzcKW8eEcp?{}0*q*75%z z>N)rS&iyTuUT?YoPjk&VI{wE6Mw;lLSDIugO?x!_pNZe6(4USmjSo8M+>egUn>vp1 zRbx=mTv-3*XZ5x2I}X+F#Egx9iC^Q0oGEXEoap}1_#sb80Y2# z@>u5v+Q_|thC0o2<`8llIF{?$*14P~-^puorDO9axfdKOb8W1nojDMCym1$QbK+AdbA6;=+O@;~&sOwM{^gC=$d4Yzc`V=3cVn$_w&N|m z6w1g*w%VpXr>}UCy}E5b@KuJk@tRNbZR1z{8-Ldd=e?#(+tMWiS(J%~+N}GwH3#Y3 zR!rH%mwXI1F()@04~#9H*EJ^P>t1u+v9gY*_RZzZ=ks&DI>pGVJhbcek(M%>F{|T_ zF~!`j`lG|<33_5vx~X5WB|RyptLjT#IYryf7tuAI8prS(d3IijpLXb=i`NJ?s@?1R z?4>Td;8#8!w>x%sY_5)~XSO3FzN3GjSD!6kzJ{NEyM`b7Hu|h>=-&8mZXsuzZ9&f>Xsg5K*JPmG<$7h|XYm+JUwjjyog z=9)29edDfsAmgm~_v(#fd+)8rv7KznyWUY=<%bUXU4I|v zA)Ski^OkXLV*aR|v6EiFG0@dm?wT@f%yZ~hY_gx&WzWV1`_3OP>t|#9t1q7T0pH>S zXz~MN@i-5}$0i2Q7Z?0iPHLRVi~MXpl#4h1$#40&O>B?z&dv!sAF$U?Aj|x@?ihpQ zK!1z}@gY9RNoIPgpQASdFaBEl)OO0$qZc|tgWjqGGN7+tbc`Q3A9YP{^rz2cpx==n zn~I~x7e6w~RZo5TXD9rj8(Bb3px5dIe{v~vT^qdhVa(rXH;3b)PI=+MPPO?+zd4U> z<(fk7tG&k@D?d7y_grrsIf>m0_EXlnd}+*|#@WL7Iv;YZ>&D+jbMrv^&OMCP=z-ig z&MS=FoimwxneUlb-51Qcig6C6o%xpe2wG^C|0q-EFX|M^)HhctI-hWDypEQB>VNwr z&$9y?j4@b!(>*)TF@3Y4xT(+b-$Gf(7xw5lIks!v`TP1_$4>s&93zICU--7;5z30j zXY+z{bFm0YGS|p~Z+SMi+beSoEqQ?K%|p&-x&7n*R^HH0{T^dhY|}$^AaAN`Pbc)}+NkeQ z$FrGswyrv*FLyK#$OY^vH#DEjT445XbBSxsC9a@{e*0BjDQjQIuRjG^Q2M3qI)5rE zhrZYLbXk4(o^;f2?KH|$nb7Fis-OBSzcz>C@49;El!pC$y7O&cTPtT9bjVMq;zcZV zp5NFU+tZ{-q`?;=OvQ;l+Y(JD~k3QOTI>uV#t8p{>vg6T*wY60?G}&dGGvFca#+VkX_=p+4 zGx{EX;@|R2{oHkJ@UD;Xf%3yso-c3wN3`6W!hDC!=!ipo=ezdK+b4dkZs!W+#WrkD zr}Z=AR>$v-Vf80Zq%ov5JlPS;zX3pE5pd-0axNceTNvZtzA| z|L)7)au+?+zTKZ-&%%jzjee;TX=&^0b4YmaK9mfJ?#!O*Jjy@$ zP?sJUCVBOZa ze!5;h)%iFk)CcL4-T1&5^J2Ju#V`2^pC0pNv`UF=I@lrN5&;HBOY14;=Sgo5nI9MGswer2q2cU&=Zz z$~)>p?bdU$>j#~UeqLFsXLag-`pFMVw>G7h>ZUO?-!6FQ*O&C9$9}BCPwS)l zRo~HfwCRjp3b{fVpQ;S@&a2~I*B+hrNuL`Z_SJ#D(b?Fq`Lsv+WFU{c2#)bAKRQKQ z-^6BPUi=o#1&x1on;+-PMV9J;kJp#Syj(lzhdx7nliul7pR4cYP5d2$YolKDgYN8Y z?1m2c!0`xW>|Gt#2gZJ(QJ*8@7$2iv>2$uW$GkDx6n*m3^Oz%hUm;g2cU?X8$*NCm zPY$%3r||yY^sL@}=~ln#k3QM9bItq8)9TWLK9o1#)n7K9pQq%i-W!mMFOY-m=;#l= zW8aMt^T!Ew=J%by@GrdaHRtR6(mjS_d9w36yw$HB$Y7&S{ikDnrhDU`@pv2u$)Nw* z=zD#;{cdh12fpT+m8WYqu920@+M`34bWuCVDYITZJ3qjeY=!e)CwKKCKh;0=&9O3m zUcW|Ldv>OCww`Zq<*uuze&dRaw(^iS$q$X!k!SmZ4w`J*xo77box^tiVg6&zNAGN* z&t$0H@nQGoSGw=qtA3`8PqK5zlKFDc8QqP#cC4Jf)YmUI5>t4vDSPs*u1D6-dS9#_ z{`EDovIUxG)2qJer+Kz{E;y!>_MfiQB@_Q7kG{w`Vx3I=WKa)(ZR_K; zJ)7e}7W(czfMYfi1M1K}Jyizu%D;0KeFpLb4j*85 zay7pBTk~h_-o0|y^`SWcU%J7UAK=Nx`0;CcurL4mLa)0vUe}gh$yoc)w{7%U$2R4) z+ql=i`X!mr{QvnQy%g%*7hJDwXxqrZf5E1V9OYj>K~F#VM)N`MPsY47<|g&{6CKhQ z{`ykgkOA<+n;qCv?66PaoR1Xh0Nah=3uxH%!{%7u>L+Z^ALK+ia9n4Ro9E~1mf7=j z;}<{X`)G)T`c(bOeuRGL1DVM|99lg#JJ+s$-L`BF zwOextKjK^RhIv(=oi1813v;f{x8xJ`z?=v8zV`gN^N-QTYB%ygxB8xaW3TdQU-|eh zAFkg!7ax4jroZBwjp$cCFsG9r%<=mC&-&)#z`kRke4~xFXwV+0i;4?hwI_1o^1xu#C}>mQx-1$su4 zj@fdwFT3&&V}^R_<7s0{ybJxRjw%bij`K=-(GCsvLz6A_2M=(p%r(%@`55z!zL24D zPO8vzi)a$s(hJ3iTlsg(fWZ{G4W>dQD zm{OgIq3W$;#hAlguMWE(W36_rFVYpB;+p*Y&E7d3*|Z@CzQznWj6aR<%=iD1`aZn( z=)6a-4qwwoZquI4$)tZaJnZNDmvedm$K>FDBkG4^e&aWz@6`vZLwbSHzsNurVoLwn zn5{dW)aGJR9d*fpW_8l%WaGPq`Y-16-<(jtwX?Bvbx-enu<>0!asj_^PAHSpi zs?q~rbgxZ6p#StIPp~~1=$(J@JwCH2@D7M~E%+4Pu2pWlZET}GAC~7jZxb`_O)pkK_oxrmk2M2YApa`Mf9M|0(?csQ)kR|ErQ4UwWcH?fIlW zi$TDL&6SV#B%gNT$9pHe{a*^f=};xFXX zNBU%w#u&Tu!TNji=IAT@h~DX&T;4q#|L=p`eRqc~$csNY$c$%mfqcjwpfCK3oyi93 z>jN2UKmJ$yv7H#>cl>YsUn)M?ShVMtivweV_RZP#33Ta!Y_{fF`e=;u`Nl9EDd*qp zi>5aEDVCtQRQ{y<#wXwB)A+P+?N@UHzOxx;HV5HRIUMT;blzod#n+U{F>J@4=m9^W zLwaTZ%4AK?dr$n^|3mxVA-?VXD0Xy>--$BPp|NFUV;Qv7VKdAo=)jxj9XYmwsd{YeS z7rN}lx5=lj{`VbQ$7uLpQo2Ed9AwmgHlcrZkoWjs|6hRLu?5}fSH~dj#ijcVJV7oa z16{CFb)xP#W;JgsQwR7JIMy$8pksz>`lzoqzRqXaMXd2j^4X)W9zVvLzmUhqmi$RR z(R;tcR`T&#?+yAtJ93g2Z*c6om{DK6*S@YdpOd>b zW-B@ZvXIYMDfY!tb1*+6v+Ha?Ryx(U&KvR5Pa7J1a-1XbN&m~@yV?Fv)Nj8yH@BAC z@Y9A5gB-_}=+!6KRnBa_W>Y>ze|*PzeM#TQ#8&!m?_59mfj-H(^(Qirw{s^tzz^;E zw?6A5`=G~n*xhmG&W$IwV~5Ts=vZ86&&K3HPdOc`%YT5a*-_k&_GVxFT-PVZ_%xT0 zwRs=y>V^LB>zt>uk-7HM2fWFl9y*OJzKb3>ZhYzo8hnqO^0$1~|Nk^+Q6_ikk8xXn z8w=IHvB9RCoIrMuH9H2E#v(wDNvN^@52qb+^d=t^FoKe<2~ebRSj>?haMcJ}xx zQ;!{$0si*Er)9b^=19FhbcQ^f>gtoqYh~`$os{-H2^bJ4rbU9bQ$O_~k zpN;M4jy~yKU3M2I&DZ$#TcG`xd&j2!<^=ip{(a>JbjC3RA2C4p&1qtRY?V{FSfE=u zwSG)izGc2^-9tWao;L3@|B^e+b>&jN#@}a0z}a;lb|-^56aDC$c@jOdIeEnb8Q2mH zwl%*)*Z(N`-${GCI{!8IaZWB{ivM9;9FW)ijXY#khL+g2=^s8~T#SHz>N7owXK<_^ z^?mwLk8H*a|EuXY1pMxR-!pLRcbxnmCwbi)80R{s(2F<_XXq~qn#b`Lb7I+esNZP# zoi^Wr_bpA|jQ4v7__|IH`X%1=LF|dG+MnMk)8~%i_HqI~=8*Os@5rVtUwc9D+F++S zyl%nk{^vJT{EkY$yD852Om&T(es=tGO+MxKKMEvl9#`mU~J8u)bc@E{Lp6}r9 z5WGEjU+|&esNlrll;B&zH-pmx$KMIg418PlpgDoTQ z!NL0jb;$AW!3Tmb2Hy_8790_%`<~!~!Lh;Tf=>m92FgAX937BToBs?>4!#_GDL5(k zWN=7;-@d{A!G8qr4)B#<=$ic71#b)J;Z1?O<2$5uLzW|hV}g$cdj`I#`qscTdeASl z(a@e=9KSQDKG=c%mSA;XpdJ1KMg>cF>ceJ7S|bV%3g>yN(R zBOk~^;!qyvE8 z7j5~IylDAuGWz}PlvD0KW#)UjC*6C}%x^_1R~HYm<3}&x|Jgg=rweV#=r<_XUV3@(z2J=C=fU;C-+~*1%YqAnp9EJ1%KjAmGWbUDnc(Q)$l$QxgTelRy=!L& zX#6%n8;?_h?*_jP(87-lc>FB5I=CmOE|$G_7OUO6+Px1C$i5)Z_PRiS9}_$$SU*^r z?We(i26&O-2f??3PX&hrWW!HC@zI8iKMcr4N6!o1Jcljs-SXbef|mum2Zsl~k9|b& z-eBjzH?emO_6j}}d^+&kK77wu%NIWt92#hUT)?N^6TB)Q|FeRZ1lt6!49LKT;9r9s zgZ~K7K~HRoMX|=Woty8N$H;YZv)m@9nfu9~#t7fD-ZhZlMn^Tf>)zYEY^b#m3o=LT;H-WTi{ z>>9i?I5M~(xH*`Znwa|gdv}|O7vF3bSI-EZ9&8vqE)ak6fPC@!;H2R6;N0Ll z!54y$1@8{t6}&f)a}EwZ9vl^XGB`0fIgktN(K%tjaZ8U|dQgDIYl9aA&kWWN)(Tb& z76i9W-#UF`a6|Ct;19un1;3cX<UmyG}u-zSKqwM#=mBFur9|dOwUkW}M95shy zrjMC9SLXQW;Mm}V;N;-+;LJc<$MC%X|LX$$7A{@5bShXfy=3~XAgO(NrKKw^T_adG zcuerPV1wX!!Rvx|1p5W=59Dn)FrX89JT&-Fpbj`bBA~Zpg2U%<{Lo*BrU^0~3b zSoMhD1%VuAOg=l13q6y*Dlo@4XEHA`f9pN1`&{?Dau;9ZW9Ep?&410o3v)JjL!hqQ zCO4S_{~-8fa8+&?Ngb8x&v@cQ5l!H&V(g8hQSf)fJq zcwTTx@Q2{0;LbpvlcTN=eji*B{8u1H{4DrUAm5!IXfMyn72ga_4Za?HCHO*cd~jTF zbZ~ee_Kyla8GJU7C)N90AUBZf}ws zq3*}c2hAam3HayhgBJ%I1?CgxVK*(kY3Y9lmj>oB=0N5N7X|Xr*Mctx_U1!h59Akd zvTN|F;Kjl71G(jef%BIIFArWC{A;jTuyOF*z_nKg+XiBE%Ru{Q2O9=Y3jQfrGk9dM zb|7CpJ$P2|oB%ESUJ<-0cxSLrAinMQ2zCj!3$_ln33dqN!4C)KbjJqbTVMVokY7I$ zd^IqCGp{-|kY_&->=|hPo?zeLeSz~g2ipfb2D=5y(UiYG639IV1nA2-=4kTtR|34~ z^qT>FpBJ1Tn5SJBnCJZ}Fjo9EFwZqlTqk%+z?K^ao6Nx+-JI%0!REoMgMSb93_ci` zm+=Sld~>@)gM$Nm^Zw(5uLbUr+*g3#kTe(LOJ@b{v*bDFy@zl=q0IAE>pq@&nWq^i z!MSzo`m8lX>uv58+>%-$Sryk>GCVI( zP6o1i>#xM$*?iCcp|2iF92XD)krK$r6Ww}Y<+Ck`-Q?i}y^6Yrmy z-|arJ`^>1sM*kkL*9(IUf<=K`y<%XlCtu$h$kYE7TohaoSZlJb^Xb6+&OC1Kz`TC9 zz_@CS=6Qi}nBVo8$hfv>P14$;XK%)DjLp7 zHlGx%7d$#xGk8eg-gd>nT^}dDYw2A}Zw|~Wt_l7S{4Tg6FyD{^+~3JZX9e!_%=hGw zPY32Ia@1$$AitQq$Z?$`%HMLCeDk@Wa~t`{y)ASuB)9Dyn0ta;`u0G+ms8&o$aAn$ z@R~rLH|P47z!)IME0d!)4dix^yL&8$%ln%L^7HEh$I6TuAm^iRT)}Ir zK>mGeuzNs0V}iUaKg+@NB;TGEn8(Sn#yDf1d-LA}j_r+qivwe!-0MDCefV+ki@^E- zyU5eZe-vo{$KdwB5q`iY_y<3EV&MM%$-%ROmj~MgZw=lS>=Nt{ z@GoQewt@TQodVa~x4$cp>-PwHFXF!Bu)uwa`xW;s-cyjP`wpct`P!PlwK!{j){m`O zTNAuJ=r!wo^0#$Jab&Jyty{Ht=EcO_vPkE z;uowjyKem!+|yatbw6ibrOaAuVZLQ8*SU7uyO)y(@N-WmKcM5f_1FXFpj>-= z$%K}@wPI@r*0-${7>nTR!CAq%!P&w0g6{|Bhvt>$jz*_H2fqn^5nLKv8OUQ-2hO#z zevY;^b26a+`M}z>b?>hRr_4b;b<`yj{yz|Gj1nQWdqU&A(Z}$tXyGQ;;aC$)I z3j^)p$8&IO4aDBsh<5gPToRz=T)*9q;7N|31ah;zetkE<LZ%^_3PmB059#7lN~+u@wc}QL`T}&tHajrz4#p; ztS_4LTW7G2Z{D(P!2iwhox67v6XrM8o5YCuj(LtbKNz3f=e;h_77h0V>X;u{%MxFn z*ErrR=<_?VxF|5c^6b@`I#{!}W?{Q0Png88y|sRO_oto}nX_5v{%zpi;lBgt)}(Kk zdmqIbb>DBYMlFBH8}{CJvNk8bfHAYr3a#CB95tUbhxGi+zTaEX)_t?<-wj*?>x-Vd z8FQ@(n!8%BJ|R%{#lYUY^%H@0>kkChuJ;SxAAC6YNT7V*xiMRgloy?=Yya-xoxwW- z`P22%Fn9iF@bSRh8RS%S4-Bk_y7rmCoLH_kkGAISnWTQ0FDv^(KyLjqx2A)013IzZ ze{!HbpoOM+`^N%vb@O(5I5r@MbL*_m>EW}1y>s>G4NtVc5u6&3i#*?(dv1caI_d-7 z?7_ZlZO%^?djE2uo%>66W+yVU10MRq=6pzd$DXh72jdNT)^t6?=5L;P@IQX$nX59- zY}K__SDU`$i6-Av*RvG&7uw)=W$=f8Up?}f1M>yXm&jK+=!yQ(Z!Gf- z{q1uo&!T#a2A{)EY={B$wAV(N82Nl4X5<9rWJO1N_nYLY?$G9Y>e`c+ujo77ILAjD zGS-h=XAganf5`2cd{_I?sW$Av7wLrGsl(3JgzpH2 z8~gA@IoWehb+mE+K_B?BvG&$}pYc4m&r#)m3kUY@Z;T7xK|!nU z0~s^4QQqgd?vLD~8(YlpJcoAg)OAeHRE$H$G|yg)TV2oUGiGC&+cB6+dS9zOvm_%wz1S0`>w#=^KmEf3c>R4-Y4`|M6 z-OO`z>uttZeX#dlgfZG)TYK*-8EcKX-dDoYbA#6h#`3;rYz|BxWW`fm>l^rZ7mvQs z^z0oE^w3s@H(K=T`8mCl6%V|~#}4ezPM)uO$Jsj~`rw)}_H&&**_Dm0A?maDy1Xxk zMt#xVx#t_MTdVebjgRqFer@AdU5~`WeS+(JkH4Bv@L%r=jNo(H%e(6GTk{Zg+-LAp zGAO$%5C`2xzvbJ2o^l%*>dR>1Z=DHE{!LDDw_F|X*tvi4ZUTDf;7?z!S!4BXLS^&L z0y-@dM`R^~dnDJ<(SPx7Y?3$JGm!;vdJ^yQs9Yldid*@)V~4hKfxIGL$@_AS93V%@ z0md-b<)O}#O@@-b?TpcA~oJVQT#{>fE)uvcxyhwIC1#xCqD*7%Ruam@C> zuNDRJjhxDtns51(_i6Z}I?kKF>(}0$sV~ai{JMVXSR6G@T=!0nI2F@kMNE;;_`z4T zL680Q1^6ys5Kk@VYhqtJbon|y`i`Eyp`$O_IagPievpwY`Xe5}vHpA4)SiswQLlZX zKYd0QToX>%^b-%V7|oP_zU~(wlyeU)}0SUzM>d8QB(W=+$qur5n7^RiEy~ z4?U8H{`8w}@KmQZp?hPFz8e43>sag9b#(EdH`mnfd>SqKX9u#;P3Qgi;?L$bGA<5` z8^#GdjfG$wagAQVb>oox7I2?cp7cxa=oq(*QSL{KUHG#lo1tec?fs7X7UQ!q%v=D_ zFup1`wt{;g_Y>s6hpp%yUpmpwb>k(SxyDxFhJWBo=20$stKG$?xdA@zr$BwOsK-ax zhhNc+dgd1TiXWY`Kb^BDIrY+_W9Rqoc5gm`*(-@zO6MdIKKR{h$j&aO=kue5cx*?nT#yR&$?wxvFzwkv@nL5TO zW1O<`G=>_Fi{1+v?~I%F+87gEGuFAcGIn;2Ra@g5I99LYGCA4axIkaVKYF$C4}1Dl z-fMI$ro;sM>kGZnp+3uh^v!qaULAVJ%QZeO-sA=L#INH8zZ3^z91Zl)#RDC=x%rE2 zT?hHqxjOuY>~bAn=ksJ?e{if^PV9Ig$MegM7xJoDYOFX%lf2sNJ17%tVon|ceNdl2 z@>O!^k9;aG;M;MG4qT@ba%#UgkTcYEovzsdZ+yrsKe`8yljR?B;^CUSW35Q8lf&dh z{ODCa)=qA|i4!+0aLq4CtvZf1_zF%QblqePe(eFQ2M!yr5V4 zOdI)A8)FIidJSDP+|w9y$RkJMtv)Dk8~%jGIe~RubX}t}If$L)Z9KK3CwW!hI;Qgh zy3mgP!Lc&7WN&(+XY$~O4ju71zQe!pVoN^2U&Rmq2eH;1%iqR1nmo>z#iBZ7L#J~c z*E=U|4s9GdcMT7FF-0EnATAmk`idtx#hjQIUykXD9{CD8sMq+?&e+3lu89Hil#jS} zjt=?dMzT1TyX79a%wFDWKh%?Jhp)u2<(o*b}v4(Ms8Zs*T(tX$w) z>B$SA50z0HIo4P}Hhrak_Mj)%@seZ2ns}7oK{K^=LK{Nz)QuIXHUWCQ&1z~6QHpmRPi#`!9JC?}6T z(RcGAA7EQD;!D2jj10yhGNA)(Ne1H(*~nF$t5=`0=R16f|Kh1kALS^q3UUUYqlv3{j4J6Cq}>4`q*O+ET{9ep4d z-P4`x^u$N#*g1OY={I`%E|jw2)a^aq?*Z?!WzXo~}SXCF34n;iO~4ttQ_ zhCiLxxAXzXZ?9i$jt=ODjh*!YEq$T~wAe+im;3cWUA)jEKRxq5_2nLNl8*=A9c8YKxkFj|Om@Iq9d$t;=tqC_!5GlKtE-Lr=%`aZ{Je3e zo<5QT*j9aXl^ch|9R1LPzR+!R+^p7Te;biL^d0c7jP~sty^s^gmt*Kr-{?Th=oek-2j9{^^z~WZPzMeArX%IX24(p3Bfckh z$ZhH^%6Qv*I{H9U9ea9ny>lJy^>d76^5`cW@L9)vg{1^#{Dkh!)!Vs1E-ZOYHA@^~x9UagA-X zkr$w21^SK`1@0ZhD;wZNR{GHvZ91W6V-LDwvvEa6{D3a`w|k_n<)GV~&pz6bpTDDz zHlGn^`Y$KdSLuQ*WZ<{X*@YZzPgXXjcRtU?WS|@R(8t=B%%ydH+&0BP# zZ){zl!T$K_FCCGOT>3#K`XR5jc#zBZz)w1dBAfQ?#~0ZN4d55;vpV>ZTR+&BTu@u% zOb+6emlg_oF1NJ5>Jyajcj0I4=lGpX}rkC2^IK=PFt>_zHI^qlL!7hA<4xq8E zTwn13^ywXK{b4IK$f{51lHHyk>#OVXk3NWJxzC=QY{5T495~mmW0r9Xy~af2%AQ^M z9^WS?U9hECVMF>-M|*w?YzdC#B%lv<=t`gTk8Hq~^-Vt<;~^he&ythG8vRsH+R%-@ zuqz)YqdMBMn>u9XM?g2)kd?nEQ;)vc)b+-gJ)P@69rH;#l4H%a*tz2Zn%eS7eb7I0 z(WQR6R=v>|-H8Ky`2;@l5MKJP4O`k=LlcyfyK#>n-eRZilrF3KVS!b zNH%(=6ExH<@WcxrvXZBHNI4%D_r^GM*+c(9xtwFqclCpu;ui2nzx^Q-zocVvM9=8Z z1>2#sC=h3Ss(lz^5?$wfLBD$c%?_P=px0c6K00(sK3jE3~1t4J@YO4 zp(}lE9L*5UA2jhRUwTz$uZ+GLua5Z?+4uxq;mJ1m@N@pFZv|P%M7Lz-&*W*J8#j$p zyu=Y*8-tWLA83Q8KI3WM{_+)LCH=}Z=2`NBc`9GxS7aeC`;vuB)!*n>Y(dYiYXj=h z89maCJ}Bd#`rh{ZvHGKL0)^~2us2tCXDuG>0BjPZzvK9hq^ z=!cE*r5k+e%le5nx*e;GTl|}^xvuYY#b5c2yl4(5*E`1>FOU<&Eg5QO{nHOI%6|Ya zJkTdgeFZN%ufFl6C%Qw=o}bcJeT@z20bl)AAC2aGdPGNieGsGClm_1A!+wodw9wQi zu?gj&4q4D)Gxm^w0dG2XtudvaKo@L)HhYj89jHH7pLp2n58C5HH)JjkZS{@p`mQh4 zXZuDL&<2ghGdqZD`_9!C&l|V2tA6Q#FUcW%i*DGO4C?8xy>tC^PJX^r|K)Gh5qa5z zF7#VIGX5?K2qmRZJ zK6Fn1`hh1tWY@m>Xl^BQ`*B}$BJCRIXjf)^BCBi4@Wc!7rbionk-d3b46+3|?C~Dw zXZkwY2oJi!n=jBEoA3#FfStr9{%pxt(QY0wrqCrh8Y|=YR{M~b?DR;V=0ZTn%J8i} zt4l66MxV@l3m-Dj4ZVm5=X7YRuC&)5dmuZ1aI9^|0`$a+a~qzv9`l3RTU*!phknu_ zT|oOw-uim?Xrrk;pVIe^nZ`@`j}EG5x~B^=lNBBG@gXn%HrK#L2l&tfeb=UA-}Qrj z)FHnZ1GK9HZK3q=DqZcsxqgyGd-2HsI`?%=dvcMZayGx}8^2=v#dEpEIosQsm(k}_ z)v0|!2DIr&3~E2-tLl_ri52mqKDp2*vwYIE0`EIo8yBBqhOVmo#=67rFRY^Oo_O9>~c5=$|jx z^pPy%m`3hGdF@JGz6f}e1@FqoHex~>I+P#eqUJ$416{h|b7<(RxFnZ+L)MNz`rufZ zYvgBZ_4$%M^G|%qL=N=yjh*O1Uz;!Owa06|9<(QuzE&5;pRphEC^YX?H*Bll%?J8d z8m)siIrW*GHu|r;^o#$-ss}TH(e8h>d^x}+M+{;^oA#$+4D=WH0yD*_n+!2 zhvu#7b4ICifAqh(r(+wviCghnAG3E(w~Z0?$cz;E@Y45=r|jP`wd47sx%o-QUVYF` z``V*Sb<8}c80>zLg>IoB_64EhR=wP{?TgC6^$Y;yn+0&!>7n&^a6Z(MRn`C;C7q{G$D69vE#T4>T9JXOIuT z@n}b7`q>=8|LF)1yc!=J)0EXm)UnlvW=`^_^6?=uilO>6+Gwy9UBMXRweS2r-D)HE zww-;?`8eNhPNBDsC*l#wAhURE46uK7SUooP^j_ZmJdEd#XXWh}>J9x_aY{Z(Xxf;JjgAd^^FY1T&D)R^LMfl$f z&nLY@<-Jqq-fJ%r^og8QKxLh z2MWBJH)b=!Yz`EI$^mcZ%?;>}wrK9w7y54A=Fj@YH*MPKA3E|5{db(BSAOOlO>pcQ zU8}F1{;)sW<0G%odEb{af1_V|_CK27eYEOZzN&9%&vw;4+T@c9=#zcT8=6nYJSo4h zjacD7WCw9U_P#gCf9ga0$g%f4?fH^Ec8;o_=;FynbSC!suyKZe_4xuH8}n^*Zr{nF z2iMi>dbRfY-}q?mXnwJmW5i+eGo6UTG4HWOeS&WIj66_Xh$ZiP+MBy5rz`r?KhO>h zzALu%wfU1AWMCVpOl;0pWCJ-*4y{i$xAAK}Ek5WH##o_S_SG*NKSm4f=Ggji$9j9< zGx7-eMtQX7mmO2|hwcDus|uBKzY)^};W@9Y06ygv+%2 z|J0*@_d|3}*W{sJxl-;G7ve*n^8UO3QTBhveuu>WDBJs9ga2Rl-o5uaJv%esk@MI` zjO!nL(mmUXYwXeWpjU zk&&!y$M?m7coHMxM~?P>s{bpn*I&8Yo^Je)fNSchPY-gBxT1^Zadu{ZzJ;%Gmkr59 zzRGUS!lux2eb6uGu8SLaLEFYmeVJeDr}B;s`iq}F!|3N^n)Srl`!&9xjL-64e%Sb` zFRQPfzH}@uP37WG`>~#KI^bt?Vr-{tK4U(}=j17K#O5iv%CR!=?sK2-h&i;-wCST< z!aw*9pW^p@pWgfP{-@agMcX^~zrfzDS4Um`C->5^JS5K8iazw6kKij#j2HH7+&PB6 zu)W;sSbpw&OuLSMVw}&>tv<*dy{~cJ99BK*4$u9ee|=|bK2zVf?=hW>^E7>w zhv`Wi>UZZb{EQwuw$p>$KI(!j^uaI0x*W)sivqC&oy)lg5KG2lwsXyRV~n=%*wa{{ zo8~?K#c$>9`gHx;Imlu9P432t^D#Ea-uUGe7CrS0CadjYV>ItNdipKk@+CIu_bj z=E`l~^7f%Ll*=!Cqu?LSJ#+-6@0f1v(M6A3Waj_;r{kS{$9J*7r|3bP@h$!Wj+MEt z9({O}_epb|c7a8|}zPL2T9ruD4$EDw~6PAhz{ge><Ywb1KnjL@+*nwYbiw56puBp8C%3RYIoBHKZpF$5^ zdZbVB(XmNv@)JJ82goz>El+x+ljcIUZH{ex(1HGo0eZVHdy}nw*GD$353>8{r>?87 zU*uq$(VpYjMZe@O$^(6UZywTbd7$d`n&;=X@U+eV+~Wv-!}N$<{EAJ?c~s zWFNQ2u06Yqu{hdMoBCjNULQhR8Jnns)_q~V-RVR>#hgADbfnBRI#X}7Lw%2KY;=Nu zbAID&j5XKzNYT1InyM*W~!KO6l|SwZ*UI_Ptu zys?2F9_<5}oR79~tX!SyW_})??-$PNYj_lmyY5@Ae{|9QRt{zApx5|d&yGFj6?|SS zjN`3wcG28i&O8pB8+*l!Jj_q{N8`A<6Qh8ijs4h@425&o>6D)t(;AQHk7GMNqc4)b zz?(hrpac3f2F;H{F%b6H$Sc)cCTKcOj{c|Y|J;W$I*o-pJ85r+U^9tqWQ&80Y6?A&36rJNjdNM}BGiS2xv< zTzFsiK)uIwUE9*;^Ym?RbIkA9o6m@Kxd$98yRUdDjoOW_*oiF0g6deA{@M4sYaK7_ z$;!5!%UWlI-qUoArS@h6x@3>8O?J()YqQ#*fqv~=oz+L_Mj8E9@74MIn60n6PM6iE za&gjFqN~Pa%jwW~Bo4)2W6a()ZXx#;V>Mqoo~M;8>aQOIu^hqQIJ)wKr?i)>y6Ad+z8t zn04#^c8)cD>-&C>6#SN{b8zhUO#54|e%ICSwW{Ygd$jd?rS_f`fO2j7TdJj}&Za@v z)UBawtDfKYfztJxK>Zya>+K_y7w9>*Mz3%BX-(PMqjh6z^ZKi=_ySt$L(kFiTmxV5 z?1Q}e0_0Se4!a)BRX$F}`>vB-eX>1lF)Y?)Qvd-uLXQzkAcXFHgBQr!#sWtA5K51sin? zc1$<=p%2aRWF?n8Xf55?W6bd^!EXk6zTo#w{r)NVZPNZOsoyd68}@#$-fy@2-FWD? znEcMW-<9{f^?rBW@6-Fedf)iiHrOV3Mc~_lTLoJLzBl9BPrlFUyHNd>i0_Fw2HzEN z&9^pvi?H90^1UeE1w+gCqVVWuK9P5PlhOAP{NA(QkoH^9 z{q}{T_q*e%~2A-){E%)8LzJevi7}y*?q|lk`mx z-`DV+4c`XwjfnFC-^}o>kKYG>3H}}|%(vw3A3P{{Nbt~Ljo?wiy21LvhQYIg&4T9z zzF}p*aq#p&oqr0}2^Iy~uNF)NzJvLvpx>tREezkT@EsH1JMj$`-$BtAzlBW~@)@1W zk?ujnwb<&?&eaZo$?!Z>AboHeF4@1Z&71u+ zs==(Qz z*d=&H@SI@%V4dKR!NY?!0>{c8A3QmDQt;Sd%|IRd#{^Fao*6tlcwVr1uw(G<;Gp2B z;JD!U;N;-^;M!p2r7JIeO0Y%Xn^3-Q_o;w>uL!0C-@5W`BRNTq>G$Y-zo_5-m7jda znXPUNwEbz&Zx5c3Zw>k`mG4o}yYC44&gTOHAI`mb`sV2?f(wFg1YZc~?90Jv!FPi5 zf}aGJ1=k1KJv4Y?;QLg*+jmHCVc=WTa;a|+`<-dIogd2?pA8J$+XucI=UZso2B!rp zPpv$)b8ud;>cUkQz9Kj@xG-2I+vdSnf(I}A;AKA){55#Sa?e<9>);i^#=&EQHRrI# za%(KNFt~ErE0;Yz_;m2O;H==M!5P6`!GnU+mN{*i_XN8K2L)daP7Xd2;Jb2g@xqH2 z9v|!#yfWA@SS7f6>guU41iJ@Y1+NL-ADj}L8=M&I5b|>W8nIp0oh+4Y!a*y+%kE~@9&h>YF3@KIB7#Yw~TmF9yEl=G$z(rTL6N?vkf`OZUcn1K4+b z#nX3#uLeg3dj~rN{}#M3*eH-+RtoYt-X*r>rk9(3P$2iL9jq5TE6`@Q;J^T_6N58@ zUj;V>zUhniior6$WS|TrlG_5`7``<4VW91ofXDP!M=fU<@*8qvO7QY%|G9}_HAopwph6}xZl0^o0-2FKg^ANKgpcieBONBdnV?E zzVqgLZNBT}yK3fNE95(&j}A5qUK#8ZyeBv$_+sE2puY`n4OW<3VdguU&!2q$%(vE$ zm^@aPb^|IW>z*rY@PfEck73 zRj_yt^6>A2YlG{8YXaBR{YCKO;QZjs0FUE>g97<+uRsno{%jj;9mtWd3Y?Q~-{7#| zv%xol?+55!9{eu&bMUvop!1Mm-Qb^sHG|ax+ux`DKK0jtY`+P978vi&2#gn>4-N_5 z6_DYd17n5!enoJ4@Zn&a;PJuq#B{zKf9cFOLO(L`k(qC_zI5WH6Hg8v7?{ugHt;<* z^J?F|^DVr7lh616e1pxmm3_z0cLRNc*>?uJcWmCfU*NlgzALj!uwCF=H109n>w7QQ zJ8152-4D74^*)r}*XC#Lb3PUL-j;hH^8?>kbg$$)jhh4;1oE}{j2zEr{}}u}_+4NR ze+cAa*UWit32qPW4DJfdouaatZ^Rn|9uPb@FrPwOneRucdvoxo;M!nu@Lz$kuP z$+w9g9*~*r@W8-#i_OdR&luNv`T6;NrtcqrJ~%k=on>=9`N#P5*ub3GJXwyoE8l)M zCj2!pUV&p}8H%QtTe{rLOxyhX@qv8i8>BA@UKyZm&g$EyzSI1Pz_&s5^JjrM=$`}j z@O@TuLf^_>C6I%hvyTN1eW8oHgF6DcVUOzqd*^=-YER$%WXDB;xv#O^_v78eZyoR_ zjzkeMa!Y;FW>- z=MKS6fqZLTx_=-?9~FEi_+oHs@SWg?!B6J!gQ*|Pj2ZIs*MhGG_UfJ8a_NI_lmO{3W;{_)}m! zGBz0}j9|u~(}LrJj|T4uUK{*NuwLLE;m*lBCod1q3XTc(2(}1(mwLtE`ibi&ej0o| zI6QcFzy{9^9viF_7_)u9Q_h#)-yX=>qqE_;B!%;G@C8!QsI%!Lh;7!6Ctc!M?%!2JF3L z?=WSgj-c1}TC&#+{{xmBFtgWo=U}_w)xp-mtAba~!MVBkj=@fI z*nY|OOI{O<^|o5F)y&*>lO>xhc}if;zRn!hU9#@X(V9!vT=IzE;eq2v2kQlo4;()# zcxdpDz#Mw5!1WD+jf0m4WYNd>2ksS43{DTu4=xF=4*nASEm%0caQXqkTEWu-_d%}- z+=H;+5rOrGZwA(Geh^qMam?PmPQ5h)Yf$FU#!Peh^@Aq`)@nAG!&8?&b>^PWxac0xT>r&^Z*0FY zcxSMGz?VN67#qJA@OQp%3>O#T#aLyn`ozFkVLtA;pD{>$%IBV+Y#;0&_%^fe`uYy9 zZ}46h$jP_OgKxYmGY6EfSD9F4V!hxc!7hP(eN^zN;P}A)nBe1qx#atU0|Lj#1*Zl- z3*?^VCzqdmZ19p`x8R`Qgy5^e*8}SY^8IM84lA_*@_#UJ=Nh z^68@BnZf44w!tpJ+k?G>4+cjECk5XKz8Cy7xH7mcSZSG+mRS@$DR^e^>^W?-%tp&> z5Ud@n7~HV%hM9J!EIegqO$W`Rf@6YD1V;vk1o+{*Yp`{&QDAO=%hWA1W5$=JzBKj0 zz*@>?ficb6%srF$O#VGs5@dN}26s)~HTma2UcWGKKXh_%d~jH>fAF?o``}f<=D~}C z=LQ=E8_r?l$&F|9{@TgcPQEGFDcCkJ)@~X+C0I9jNU&1i9>G1?4Z(H6wZZR#-vrid z&IrB`938Or`-0tr9fDT`*3O?6tQjm5T)W`fnQuN@w?8uQ-DdX)?hkxZlApRK@PAL{ z8;b(>1bo?C#+=D~$9S_iFpuhUHuIyMbjS-oKPWCj_n^5LgR9%RRnh2c*>2|66C;jmaa3iF17R0oo8~;?n`%{ z$w&HSotEs20y)n*k^2Gb60Z#o4b0y!nZvnD&z-rSIdbWdOFt6m>p{U0!Dj;ZLe}5@ z8n~zOOw79PR)M*|OM>SIn+$US&(EyidFHTNV1DkIp685L2K>NWch!K;cus1r+vlpz ztp|G6Y`*Gwlka|;xB8YTKRhb%z4=1|`Erlooq^|0`^HcraGd<7W+S=c?i$#g3SXt2&`@VTOcp} zYp_A!`JJ^XN!Ig0Fy5K}UKY^HH-l3Hduv+X3BDD0?r1IKW5Ig^&-M6&_cN@6+4BkC zj`i%4FWGxm>iH_Wm`C$h&tTnuDdRt$hk6&o9xZth4>|L5!KVWD<9FVn@ca& zHm`!s7a{2XoH113Md8wJ*6_>*|S!#f@B`}mso7_JVy+hyKi zjYbZ&_pIIg-}4)O>)H9{f%r9-w>BZq+Pfb&r+3dUXZ&aI<>2eVnZenCy|s};f_DU4 z2hRWCf<5Q(o;)kuH`qUTckssG`GGlXZu_Qh zS#nG6C~jJESFrr_^3$sa*3ms1w0`fop}B6K5gJz@>a8%(DZse=s;YI4SsgV4i(uU~c{6 zfUa&2j1R^M&k4R82iC)^d9nKif%TwY2iDfQCNy>L z)XZ9zbuMGO@!-wDfr0x>ekVuve@z`*JAwM8TzXh=Kwz!M-n!!F0{5-v5q)>@!aM^< z!+q>Q!LEVvss8+^dmlCPeuw$^+k#^P`Pq7mca1%V+BGoOIU)FZAcuHA;L6~hz&i!6 z4&<+21*=W2HZy;7?|yjj`M`Y7-1zL^;^6vV`Kje+?)zUh^|GnA1qTNw1>X*=Ygmu4 zMq&Nns=(Uz9lpbT&0+x>&703Y*eW8C?{ z*80Do*S$mct_XmaokP&k5vo z$MX8Ifwh@kgI5P~z`q3Uf!!NFE^z;BKI~l$^UD{+yx{4<o)lvKMefF0Ff&Avcv&HonuD@{S;6s6RfRlnB1@{C`UgpWmY#qEU*e5tVI3u_* zSZmp}mVIHcO|W&aMex$#MZspl;{wU#vSlt?=Kt_@C-9ff_5c5knL$dlsf0nBw5U{+ zokS@uq(!z=qO{Onw$di;5@jbXN_$#lDMg~blG4JAB@{*WWhTGJ{nL4W|K~rq`p)Ba zoAWv6T<1F1_I@qT*Xw_#PxHopL2E`8;8V4 zg6;og&?#sp4~rXT-#DW+)4cVcs42cX_Kvs3dxQ4;{@`=i>nFs6VouO5u8a9WuVBAv z(t1}`4|W}E1r2}e&|ktYyg97X-}?j&|C2$xKO^|D8$xpe?VMksr{6ao5FYQEeb?-Z zgT8W5*yjD?r;sn*7T3l6P!Iokd?(aM#N9iEe00N5OL=n8?c|T@>h}tETTHH> zQa%AWfzRr>6ko_;^-OLPVs-1FpLkTudFmvdcRhP9Uyw778qRZy`(GAnvmWJ)@@2UR ze<7D8MyM^xJH`D+hS=$?K|j$8vs;Mw=_BHN^}5^tb^A=cRV>FRi08$7;=VV>E8_)W zJkwz5vh0EQl1~4nI40PVeS)U)v{)nbSv)kvv|?v{G2?apF?TGyW8pP%Y5XbX$AVz9 zE(o@tEo0xLU39OI!^mB3i_+$q z+(RDn_;^(CyNebsn(?>iE(ra0&xk+UHGJ2XYiIXYs~${uqNUSS(7MH|G~4Hd zSd#WZ8=o5<$KRaUZE<5}rU=U<3V1dVp@cyrKD z*fBcF0YTTN-O%*E6=w#0=4YWtM~r$wTo(TfF|3$)m3U078(Rk7xLer2cg6eSgYo{@ zH`I0C9Pfw^gdQL9{So1BJ%eKXcg7yEZ8+|1u=v||(x-j@9kcG3b#44EY~Q!TzVg%d z#Xi~3*M>UhOXKCSQ@k>E4Y9Uf!xx5r!!2U{7_m3aOI>v7STfl4yW-9$w3_KdTxQlX zDco~sd45A&6_>{qLHG8Y571L{Y5Xl}Z&*SM|LAyfJUuoGzGS!HTj*Vf1f7fD@qE8{ zOY9N0ozAs!JT~qhw=Z$~%y+2a$}`WKK+mG-9TlI9PlQK$5WQ%xpaH#P!gkFu+A8=A zV}kahmw;c=?;wAmXVNR>*YwK6Le4<ye+N&B#|4f3 zVez22U(kx~6ZCie%KAEJM-HObSB#Zo)!=tLiskghsF&d}@%Zqawq?6eC-7+d#`f5T zN5@LxdzTM=HrDmnpgr6_W`%9IEA*n=9k;~agI?j$bIUA@Tn2W7pV4P4HV`vU_XF6M?}YuL zuY5iZjgQ6$Lob(OU{1li;(%a>T}Sdy*OmQVqJQ*AF_UY0m07FIi~)5hxjUP}20vi- z17`0Udg7Lxv*euZS*pfd7yYj~gv;ejT-oH5hCw>!_ z^;x>pva^<*(G?!i{1^6TlThQOk+X+vrAN=%q^)A35a;^Liox#DAAAq{LZ6l|i6b`% z@uS*@IV1R~^Q z@5EQ*#Q0p$bKf8D33=xmgFd@U>=60}{Jm$0C+W2G1Ap%obkAMGvfIaY@sgn5THlA` z@HjSn=NIFY5UrO4!GIAGnhSOs7pU7*}TtWA6 zz5Ft4i|z7n`F}0{uZM-}Mo#3qQD>!{(aT*cdMWL*hlXl7{<+;;tgT^d?eIRz7>w8c?{|& z#;o~Ou3fs0o)`P@`S^42%lilawQ2RQQ(}$!13w$$;5WoML1Wl4z7TYP_3HI_dz=(M ziC@J9ac!@QcMLjT3zx`U zI@rbyWBu4Do*7=JGg(Id?~y;A#`}DVxA8^U+xiai38%p@v)$7 z9Tvxi<8x-%C&%Kda6IS^_SZ2t55ar{vCt*q*!?`jMUJCeJT9zK)f6 z=I!D9==_3H zz53HZQ~zk}7q5xULQkDu>F>t-;zjZBu-@Os_u`m1INl!g+^yp&LAOX&Uy_z_Y5Y3a zEznf{95k0jL2ppIn-`Cmu=0{C&(!knoOS1{E5m0_k5hvVLyw`GoF8g@zC*pwI&X-- z$6vzh-oJ1H`*~T&1=I+M4)+c9!&PJTkTdvxHBJ38ovN92&Yv}Z*8c>1N&`|;JT~;* z92jp1`Uj1JzCkazWrKpo6jtI3wdb0EO!q_}$0CK+v#ZsYOXdVO`B|oE^sTrE9 zaLT0b``CMqozaMosFot1)3bF*&ilcAkG=>s_y357#dLdR!b6M@{%#_e}Uq=(l`-n5StQUKg(k{>ZlRhkTXm zl8<}g1l#Vlmrw4g@apDByfOBRPlR#vk;2Xr|3x_dt?4E3iU#A?mO*55~y7XSV@ zR-3)rOr7m_&3}3HoJY@jLwr51jg{uEGGJ>DCiio@fu_-MQv03;P43D1wBrNll@VSHIow0j}&Fv$-X{A^?Zf%bL<#9p$AWjT%`u-s&k-t1Y zHVZL3U-sajKkPA`kLrQ zPYZoc;&1g)`qTA6H~Lka6{iM$S`Ul9RPpl(6a4LUwNw46=Y-GEwycB3{`a6o(dMn6 zKJPg_p59K2r;+_S&Wx|e=fiwe{!XvVCZX>6fDoJC5Ld+SgOBr@1*gTggWi2qd@}en z8m1cPt}){84e!}-M&Fj>&Wja;UM=nx6Pv5Z9`I-14*pC0_`cXDjA^;PespmxA0x-7 zb%^=IWa2*gKA*Ev(2i&&dTQt;PmZU?&gXc}^TIX^o}7od-^x9m9Ey=80|{Lm|b1cdRGbG=ezkvK5eIaj$_}`(` zTu`r({iI?1Cin%9G#r|Z`9+rJr|CU>Bfqw9>>V^m-^W*N9sH_ev2Hvi)Fb&^J)G*L z`a*3pf2?-OKf9m8ZPi!*7WVg#@%#8)c)TS39FEWR;rKb84-3b8i*PRNFCQn~U!gfn ze1dpg9%7!v(cz!I0%K9nnflW&LX7$H_*`5ZY#}}1Rq??%A=s!hgMDVtXby+R+vE8$ zVsbUmQ{vO0{yoZHasCb(#691}qE{H#cm|TpjXPjM94`rin3xkSYBhLC{b1r^`#(>;vM>p$_+= z*gWVYY7-BOA2!`3G&{_&7YoP#+3;0ozR{cwVrzG<~*}rYnbDIoRy` z1p6w_UMl=evz5c&z3AWgTmE2qww;D*UF)nx{i6_9!2rVTf5b3p#;(oeyB| z#j|{a?GfL~QzYv*#Qb36#PaeFF*|MdfRJ;%B}U9H-hOGYaU)hAf3vsGo8Uchy4Za8 zP-n3oc1d1wWXKWp-=7)h1zSaz5QjS^G>Ca&yX}iQkz=fWB8Jt==NzzWG^+2!>62b^ zF}OTX4F1Ph5LZr^pXMY#q+MCYvU&o=@N&yzL%!!;W&E-_6&;=iLw{1YVNdyJ=Yg-) zm$-WH&(53LjD5Is(Va6oi~NE%LzAb&oF6|4d6c|K{zN~g*;w|bC@h=t@#{2)RD6F> z$Te*LeY-D(>-W9kI#-)j&o@43ox6rujeTQB*q48X_;uG1m(ozg*kWz-Db+R~5o=9& z4(I}LTt*$v7ZOV zdnUZEo)OC(7Ig2UgKhm{d?S7oKM7j+nL&^Cy8WV&d;D7PHG}b7AH6x0*1_KTtnIVz zqr!LUyZ=PkHrr`?KN55#f7@=$eKrma?}@S1q4liL@YMnMP5L+=_w-=5^|7oG^c%iT z&ksL$dC;KF2zLDVpndNjVrli*X9c~R)-4aCSJCO1FM^xUR&;_nX`|mwkmg?{abPxNoc&^J2|OkD<>G zo?MUPda+?_6y`0h9eNe5-@5cY(hmO~^rf@n%W-J1+pmw8#xvu|v0CU8q3O_y=r*(> zYJ2RvIyk+AjfZ%7;JNqdD3)it4-4xOL$h5!3%VAK%Qna% z)TuWJ`=H;GX3kz88EhDP>s-<3_*?oMP41!b=x~ndgPVkNzEwOg`0edOzO`+bUw556IP~?<{f54%2Jz3J5AqBA1HGPK;P>e_^gnve8S$eyHTW3YPqz^Z zoE-EU8lgx2!*wTCI5YSb8jyPi`Of8vG4vH}9CXlk#OLDl&|CH3dU%Y-eM2lHKBJ@4 zalan4Yx7Uk;ppjC#)D@+c=nb-*Ell7!n4!BWMw{6`BX#gdRdW(N`_*rA34LH0PAA{=wKU^clV}#N4k8I=h}i z^G!F5Cx?3B!$J+xzni+(!sT&ch;iv3$HvFQIXACVKeo7;ouo~uAF{Fj-Q4_N^};{I z1@V&@^W4}=8vNlQ-qmOJ!T3PXD-H^_RR5Lh?kyo6enY%2Xn(te9yw!6O@~guMQjvL zi^subv)bS8dA}qI1&VjT4%tI7-}N{Ii>E8+%5JSS!S$;ym%Ld$zD2w}jdh zJEoUJd@62YW8~*z#k1qI5I=q+*rJnyHbFC&=d)969q1M24(S!515ER=dW5|W%Z%kb z`m8+Px(*Na)B5#J&=Gv6I9A+D_qHuKC8oD&*j*a5xSDSA`Y<;}JS}dfSY^0seyK;rQBb=RiGULC}ljHFPQVU+%G0@Q3QA;%G57 zU7K#sp3=cwGi)y#={mYO)MNN{k2eN8J?b)IcApckv%{{l&rf{3da;~L+=oQO;Ii%*0+^SfaV=^x|z z@K4X=V?&H3=8`v?+bvdO2gGu6^Jm1)aZr3Q&Wr1VR!fu6JO1hrce>Y;IG2W?51qC^ z3$8_a_Hw}|49%Kt7Wad#WT)2%S~qQb`Jg2)5qB+l*UWlu>Rz)qhMd8A?+){Q=;5?y z>k#`tEBFF)qTU_y3i-h|gO>fq|oK=IL*(6>VyT%)1zc@J5e2_$tOeJ;`qNk-Vm<}K4qKOG=|1bLlfuIplDea1YhF39v05=KCxSD z8|t%94_eLY;d;2wM(@Mn1bQsIiO~lRo4lQBhpq21NVr<`|b|$wMi|-WE@xI(&51YqPd$SDNrq1T` z)@8lsE7I(36a9icR1>E6&|KVqMVzjmlHNfVxFl%JH0X1J*8JlTU(?)a&g?CJqb|(% z;DpedD83&)NWAZJZ1b7HSN6_r|{Q zj?f3=`k2OR-cuW8@6~46cAx)j*uIk{eM@3;+a|83{kR_0W7v0nNot@=hZz0-@sLk-MRaJ(Cc@b!gH_h6nu+*fAOQw&|?Q)_r7Ic6?*;I0JeZG@R2wy z)Zf^VpT_S(9gSYh9_g(U+wKzYj8DbMq4(snxGiY$YsSW*R!qk>w?{339&>uARhl0) z=1vLx^kDKW4-EcAefg2WU-4T{47mfJBj%Swc=Y^9p(e_fo7c(jS%>~nzD-`hcljRk zwDjp;A6Lg^!GGGOvw{{+7vf*p`ger;X1zGnP@fPh$Gt+mRIWh7G51QZ<7dO1owo+t zE+3%7%i-pQJfO>L#-8iv*Q2jj>W3j0_(BZb=0n|McfT+kyO4(P=%_VST>GKYeIo zG#Y|AJNg;u`+O1&;W=Ty=?8Le<94UmIsCmtxGyY?fiF`h8^?zoubS`pJ$lYQ(rg^}XUCQymvr5+q1HnO@#t84k6pFS7liZVoa`FA zhimt3v0sQiJ|1#iHuj|WT6{b7r2RP9+n>iT;)3{}xG2Oje~twqwh^ZrJ9^x5&;R?o z(B?TATeMG_=R?DITWvDW+W2!1C)%by0^{S^bN`X7B_ndbPc9BkI+@#1)GyerI4Iyz_~>W_L4=Z8E+U!<7Vy;szA>E5)e4MQDu z+t?}S<#h9X;{!o+r>)cCPY8Oxn(BArhjB*yJk(5o6XJ7ro6e)Ys&9YfDjq%OTht@e zVtvM(Eb+WrfOwui^1bg3HQm<+?f$u8d)fI_Lru9d%&dQe{j|UK`&)5hd?x5Yj?G>{ z2YN|7H_Yv%-SeAz$>~7%3-e0#tK1MYeEsA6t$VN?7v|-e_a;7nQP9ktKRW!vdlt_0 zio2fWnn#DRK*xDS7(Z;Dm__diTh2DJS!)0Kk=2j%e6e4&PBv8z&JOsSUdhI?336_> zj-92KvhQ~V8{q;IGr@Q0^%m>O1k( z_(Hg5UBC7j#<+bK%Y7s0y*~}P1f5CUK%Q>D_z>UywBTpt0`3=bKzusB816@=r$YRx zzEIoEj9pr-`XZaZeu(p56g$T5LFayZ>>nQpzvV&4;0tKu`p(%z8plsUAJK0@51|-Y z&x*N1cg370>awd33H^v_q~`l?8GOXf!AI;J?}`J0pHPccBWAa0F>1DYa1IT1=7Hzl zx2zamZJDomThMIW=Viz6U1|Yxm8S)b&Gwilp!Z)cqb{H?iEsO7TpRP_;;@fD4z=6k zLT$kD;RnztLaw&2Zl?c3$o#9&r6RhtB6O?1p%h9XW2&S1dNAw|LEa>Q$$NdX{*QCL^u7`28H^d=j;?4Wm)z}E+cNGKbcdY&!2A^KG3$xc=IW=3CEQW zb{wbuuVXfRBwfwEP3shn8C%TWvd`@FfcftSX>Kli&OhuI;&}6KKOA~Qk+ILFurF*48z5#B_p#H% zezV=|0&Qkm0}vB}Uc+X<-(DM$AFLUJ-*NPvOnw^n`RqryZw{1(6soT zX?==sSR6SCF7B4gSl+tjN&c~)jw4OQb~rDN9X-jhW5d~9 zIF7d2v2-pSPx~>A*Su#L%Ug%_S+Dct+_KBVX4_xeGWOZ_*-rZ-|AK8-d!^CYXMSe* zX4`BVeJ7v6zsLpT2+oNZUQQ6SINa+-yi9w4 ze8{!=ig_U~;6H2!TVBif|7tUImdoQ0L0hszA>VpX*Qy0pG+{nzS@=GDk4=-BjZe&)K+W6oc!7>^3RL7kJH^qSB= zLpwPlXy<%_xki7EzlVODdp9p&rFe9(y&DIA@Z#7x=r(-8JLAA$zuA5^pQc3Tp)2XF zKX$^=^_HsV>gBW?yRSy!J`HjXdOu&o=5G<}1}*D>;ol8)3E4*R|99e;Fb9gwmoIG< z>%^+TSBm@D{tM%b_*&5T4~}<*qb^UAXUZe>3b@YYjq-sz@40g(-l6SX65^qsP39$y zyz#_-6NbjYSExN1d-4V~DL#Y7v0czs*;96xJvNTTiu?z=&9<;B({Ui1zqtPxK4bCN zkgZ`m=tRSIvMKB*n==i2Gi(qW%9gUL^dd33*mr)&AqFFtunarQc3ID~&1K`+dUk8X zq-+-(Y9HA!{^8$h#lQa_dWHCTK)aZRUm1Aaaz1Tc`V3bIdB-?@>^r+`JJ^25j9rG~ zMw?-y+1X*A9W%$s@o)^}0CFt3om_0xOKJEggnFrb;?wc*p!KWGs-5bC@#wjlsG5ZQ z;}cRFSg0J{pu(fQX7|D2`^RSb2d^$Pg2Y;RX z-X0r0@EW{tIiK-)Hhg|quk{;azU!j!y%$bt?Ov0Mh|l@1%ffb8A6xy;5X13f{!Qm1 zY`0o39d~hm#^3Ne{LkY4jGq$cv+d^83_SO`_xU}(OkTsEu_JO3e$k`n@E$*DIm`R( zw7=vh@D z@LWwn&#dLlO*k@WMq+=t1C2;sg7zflrw`GjXj6}md2zpxr>J$Cr)G|t-k#G#O@t1l zF0Q|aUbbhvB3=+`DeHz_{?+3_!9O(JXGVvUoBbg!2)&`|GG7YwuJnh#H~3-aL!QdV zJ9p~_pCMimD=iapVf7=|x@+4tE4H8;j($;nJV%H4Oq`}i)VO?WsI~DMJBM1SJ_qAj zJx?u<4-xzEBjP|YI6axhz$UT_;#je%xKs|oCJ%czVp4X`-)tbe;(fN!vOdFhvZ2<) z-dHCa%hs~FzKi`~m-#VvXy`?@OP*mHZL4jz?e;-^QOwFG+Bdbvf#=?{41cMn!lsQJ zWaJ_87O;7~Lry}MqEpFtY|A(vwwJagZ{i2!1>$jjkso7U)nn9P)WOep35-5;+7-J!@Eo)y{fwi2K-c3#tef^lW72oYm-3-(`@-O#*!OAQY5Vy(eou}i zZlK9IzK(~wjN|IK$xnBR-GZL)T2iNB_ucc6w!dGf!MNr=6k?W-h5C%X8G4Yu1u+dR zTn&dl#7D@9)dBblHDC1r@%40U&4-AMJWk^^y*S2zxcl@Fvx~iHu57)rGVN#h2;+!- zpROmkVCO6`=YMr3ak(62UdZR!bK_EuqAnm#XWQX#ud(-P4)PgwW%-8n$T#R$;y-?b zKe;_9W#5HvJ72LB}flwZ-x z_zQMM{v@Yhd*l*t2=+*AK`cvmk`L$~V!!k#ULH3EU5UMWWYEOfLcK9N$LoWwd|!M# zGQN0w)rvPK0oTWv>G*M_1x!8*t+`w(fh^pdU-Yu z%d+q0YOObg=g$a_<2`zmdiJOjsB!aEwxKranfd@fcFE-T@FyFvTPWe$(FLOY%cr69?%8ENc} zvHteD&sm3cStt9aW}u!S)|5BMBiJOdGF`*Av7yttlWkr+z8B9g?i2W^Q9~H|6Ppds z=}802v2W@M!;V`gpToC_@o6Hqg{|bT9J3L}J9c7udCACqMy_Mu#P=iDvA>S*K;GlE zt0&sh;<}RjYT9QyHojku!PoMY!%xb0Y&XqDeZ>BV)9tVGuJ1-KJ0B-zXT$aF@b$wt z>WP9HEyM$Cyw`@Wv^;<5v#zOu=U(^zK)rXsU%+>P54P^npQY|44pFCnQ;1FYj{`!@ z-}S5yNDjegjC$$t6Er72f^VR=u={MjIEZazC&l{43tOtLU<}cq=ufmCcA1TS(xlf% z-28vV#MvGPULP^K*qkp=FL-qD8*IPt^^fK+cM-q+HrVsC!}i+`*?fV}m@FCvk6UMQidC6V)HruJrZ6D)#B zLd``Gt%f64I6LS`eDD52&wEku3y%uhTyd>9d&Ky|4i5dbstgIGUaEeeCx5 zj2M2}Ua=`7-c(n7RnVYm4Qz%-_QQK}8T;e!@!V(GDEnvK({a9S@%^@$?c@iOmX4|O%r7|)d=VeQAE_^3wzVX5&Ro8KbUD$FZC^YhI}DdK9lu_l2rWc+I%u zL)dV61AEH0vd8?+h{f6H5v$X>=(^rFR_BKpUpz;nvi;t(jAeaJ3{S7}IO2Z$YMC=a zeSzPjr}>O!X;pFvF+XTlv;w&eJ7F6gGoRs4Z0q=5>x3M_vAiyP*71a6Ck`Do_fbDF z#~*TwQA=_BLz5051#Or#H=7pR=p1@YWBh-+O2wKva;hq-r<1b;(0Q^0;vN_lb;*a6p zK9+}j0sTJy6yDQ^dUm*Pg}Maa!B5b&_$4{XBZCHZdH04mD{QxY*f-=1eB+j3t{9)7 zuJVvjqc>OMSE26D=kmWaJ3c|)CI;aX9BV#Zoa5X(hiY)HjS*weUd2e_2-g$*&371A zbX$FT#-TWnj}X(*SJ-c3nN6n+u&?4ewtCoE_)-R@IYuI}0R#Rn{yf40`W3V-_ zt$d1pboEZ(=ezg?Ih60)IPAAaz3ukdzKLb+FW=*{_HWcn9ZNZxb$c{N+cBnxm`_AU zu@2ksQ9Z{x#MZvY_wz@NiS6fe=s0Zcs5|?ceWvyNHAdZ6E+XfUf9d7=P0;-1dLBov zr!PsK_J;}H_ul9Q^LbhlKQeq3d(XdkKK#UZZ}>s)%a?r4GPYff#yJ{3*!f>IcAUOiXOroPOfN3pf+rJUljpd(qI&$*AFbr@q_H|C5zwiv_Cv#a8Au|NMq2jWlY zx}F=e{0cj(4lFlOKd=m+<#X0+8TQp{)<0?^KJRtT*n8~0<;DB{wybUAPvjf6d-#Fj zuN)`)A_s8{jt+!*}Y-lUUeXs$a)ngA;y0JUrHk%|c8~cX?^>6|bGlZD42Bq1jzE3Hm;r zpRv48nD<2k(u?%gP-j+0K4!uRli%$%Z{Y{wQLRCZK_9BVA~^!PKd_Et!}7lhKEqrG zbLI3X`HVg%ISU^}nZNAlkp&oivd^PCk z*2(6G)om9WPcLB;*(CK^wuQgoulOf6n|4iK(9_|%Gfs^W<4n9{ta_cMDo!#s=&;6= zfAV=_liw8=$)m-QV$R`%UCVR;{Tlkh9DA{l<87Ss*R+8#R;{mI$jIY=A2h{ z-@4dv>lrzUNBS7s&7Th2?AVGA$QL>f{5oGS@-G^lT$I1%>*OTY z2ftxi8WG#i_tMcm8T-Vig3pq(^3&o2{$76M`{k)#=hOLEaf`e|d?P1j@5JYPgTJ^X%n_sQtIKU1Vz2jv`JLwQvqxgLb;I1Cd9iXBlbePZZ|BfUvRkMnvPt4X ze2EK zIiYjJZmDC_O!O%IIj#-I$T?BFrn9qwKMo1D+N?e(+s)o=5XOjmVv5z-YW~1HFEt;1 zZu}LE%sehO^`zj7%oS6=p3_`p^N-zs=<(g};qBcIM9-GqU3GM^yu3zl5#8xI-K*l; z@$pzK#$#Uo$Ji}M)eaHK5^mR@$L9~JaN$z zXYP~p)9!>jpgG}aZH#eYs`N*#00Ks^+k4uZF2q5!rAk~gFRMnvp;HBZ0tdC zPCUB%G|3m`h>rVvgU?(mo)Ppo{k{CIc;JfWw8$&vV*Xu|mbvt-rDyKtaZ>ljrb$}Y ztHU*~28k&D=eCmF_eAsyHO}jc3GF-Ann{I3&Ivx5bv-zxjjl<@io~ zH_nI!@v!bSw`Zs$`^Qhv#pVS)haGji`B%P249~Xn;rfg8DX}l?41LUb5+{fs^sTY4 zv;;AseoXqf`l&H2CKNC7lQbIVlKw&47gvh^cM3LJe#fp~Q(gV!ct`9SZx6?hEn@F{ zH$NnoV1w8-KEZy`vYZ3=2>ENjf%f%q&3z+&9am3&5ACw<-Em%=7OvIHx$~{Hi z9rjs|l^(3O#B1aAVSMPrIxO^Ty)AYL*EDvHH-tFGJ@cGzu@k>5x0JKWN9Cb%9X``> z6n{IG>K>tG7vxZ_e5NAup9Dd*pwOCAp|$O7rEbX@;ZsOB_J! zbv!Q%y=mqb*=OhbJK;6Qao;e<{VDO}*f?CT8^v1T+InO>FZ8YPC(eO5MD90gSFRaj zMUEkNqjT90F_5@RUZqx{Z;4-gTyqkPWxq*vukP6(hBhbQytpB5jVt1;&=X`nfSyo2 zqI8;bg1$4me~%8f_;vA-_)7dN{t#?2u8M!gefw?lr^SxpH_P^qcgH(ozxa5t-G{~- zW1YCE`>_5bPKi_FcX3y^x6uaivUp2;Fg_Xw#h$TcSmu`5x6HmM&Ws<2-xF96x5n(8 z&az><9uzCb8sTxNm_K{|%=)16*J-zYBVq4&PwXG>iZ{f|V*R*( z(8j(PZ;CBq^vXKkch9P*?nVvbK|}eau2m7`37y( zn3tQ%IcQI=ao4<9()blKiXYe;PW$TWrZft%o=5R5ok9$2uI@+T)$#PWZ~Sf1-)8PB zcY60fJ~t!RxSam$;K%)jA z*SS2{aQ25?G$w89_5ErfU;Fd;QLuH#$0x(Fcy(+OPl|iTWqFow#K&X*I3V61?~OOc zw(*QuD;^Q6#tL!U-M8I+ar_{@6vxG}aZDT)pNr3j=O@I8;eD@tHVzNZzZyS`U&LkM zhIXs%kH-1M<%*M$( z(Hht?toq{pZav-4^W^d7mUnuT7_=ZBbC|BhO%xcaC# zEA;BnO7x_NrFRKAi1Uwg-kh3RvZH=I{>*>^6 zqb{qT(!KgG4?2c^P~Rbj|5DH}^aP3L)q*|J!fls)XW=#T)p86mp_qgIByLwf74yF{-W`t70daVUgWbDf?dD0F-=?QkekK+i z$L6NyIoqdOLcHjaEnGWx3EIE9zyrJ2(N@8pejpByljH1QkB8k}e$Mj!LfXS;?EKs2 zylu`2aYl&Ye;D72)8dkl3#`y@>fAfz1HX-P;-dIREHihRxets7$I9`T*f?GgFN#g$ zQL$89J?H8=G}JZR0tyUaS;1cF%CW=o7JfY!Q!$S+Ssc z3E%p?_)^fpc8G@tO-n8O(BLcdOo^j64tcUT+c{A0a6aTsv=UmT_=c`4UpEHj?&3S~ z9(~^!ql3^Y*g%>9ZGasVV~anY8S=bm#Kz${Td-k>UBoNwI=i|4geP>rG%?J*V{R-H z_lp%n9<*%Sx#-TBxbb)0fy6Guq7!A=dW3SWGXdeZDwoHsUL~`(NUJV#I3VH!-Q$)3MnpY?J+c zOspC9_laSfw@@uEZ05lY~wn?t~;;K3dfYD zVIRc<2Z#KM#$nC`dw+O%^tv3vv3OT}Amm*>=ecXgHS&mX?)i~FhhuVGIPax^GxxT= zwcj?lEee?zE%mRZW`4I>93iJTJH!@O#4T}0h%Ii2g~2!ROU}ou!Zos8xL1^G!FHb* zVu{n^?2u2oHf{_)i$3$fpmou1)K}EL_$T8<+|QS&KN&-Gd-{;vdek+<*lGiE3o)6x z^dCbUu4XLP_IWuOU#IS3UE&rwh39hi&xPDW9Dh`ZZ+zF6L!D7>!jByr;vYWKXZ5Ij zEDnk{g`7o=ZIf7kQrnX=tQ7pSTyxc+HSw?BlUL9wjSq1b-)_E(`j_!TR}f2CpBkHE zqmDuAu)lOyI`Czo_D=s*<2UysiL~m)oinv{wSRxhU-=jMt(bP{u$=^bL)(MI%Uo&b3PUCiv#0B;rTJ~%{VXq9{kq&v0c0*HVZzCUpqIx z7DvZn;q#vek4MCD@r5`s4v+W891z**rGr-I{!wZ@uZ#8Ko`h{N zk38lHaYgrW5>uEn;~cvV=o7|`>)Q1%_7l&!uIM7{uv*4`L1&m1Y=B(j)A6|we;*fz z#s}lQad5D6N5$vEvL6eNmSz8b5I+oYJU>IP5aaVNPYwQS)!?&+&v`|^L$F)y67DO< z-m!n*4nBk?3}ZwNp{7IM*IT5v?C%j5)7XtQF)TaqsW7gMAMvR%tzXlzVF!+!^q`p! zAhz`xu`XZE1`HiTzEA6;ZKzE`P0(^|gJVpy_&}T-^zFX|jqRo|MvY%u$Tvc~Oh0;C zh{b6g8;2T+JWB2)R**lk4?hYT5g+ozs1e5?2^mQ)SUIO>Sq-fv*U6k`m;K)deA2EqIhkv;m5|w z;ZgngW$}bqD(2TCJGM`~TfZs2n&!=({U+ou`l8IU-YK3JtA={Gd%d0+azESqpzd>R z8`h7Pgg)4x$9=n(>vo}LLDQmb&Fi-%wuo26&XZi?NpqewV}JiO`(LyF5m(2*m9#5o z|1EA0y49m&>yZD*M}8K6hy^i!!j*HbocaCHzs~vVoU7u}@ZODazq$9D`;6Ez_KCN| ztK!+QX51?*cX?P3-^AZ+5WB=X=E0<ySn45Tf930M}^SNSN*YDNobv`)gWNK9MSMzH=8}j!3!yFv* z_0^8mk<=Y&=JJ01C}J1!o*3V?&9-k8j|%z3^6`LJEsTYAL%h6t$Xmpsf1Pks@u(Pg zUKr=}Z0l6RH=j(LNo!gqo6AI7vfW&=Sw`Y`+Ns|e8XVF*#O!fUD>@>#nJXd%*|Kwn{Nwt=2;=PcucGk z%f%fF@0j^5%AYO#*}{{9rnGw3FAdQMXIrv)l1piK)7U#r8!OzPrXNV}I*+EB_SoH#yz) zaeLeo-oJmW9&5*X6XbM{jkRJNgMIqVA-?AKA#QQK?iqY8&6KVGahxAw5j#DvI+R>r zzlNORk}#i5Z5c2pHx$%S?ZgS`+qg~KmHB+V~&ZVL*D6g#ZY^OoK2i3&f+t_KjF0c$!L>uMse~3o3m_g z4b6Y|7+Ra$#{CM_(RT{xQ9dPhV;iiSefUTi-N%^(0 zF)wJc`p>qEX9xY-`Pw6#Grw!0|Ie7!`~0)`T^PIiJM=N!7`KGqhxT|=m|sa3p4(i@ z2gbugU(N=huD@;U9AdP+;`Q;GkW;@TUKnCKi2bm2i20rwo5sfBm~9@<4m~?Bj+e&{ z;nDNw#zr9~)X%La<&Gu({ktD>ep)B=k?1v`af>+}?+e2|yC#hn*RSjNeX)OtPmL+} z4>&ON)bAboAN1L85$f&wn)H~_B=n`-5Z48rLqbEk}F(c6j_j$X|XQbSK*jI-PZm&(iQ{1HRYy+Ya{HHmlp&FYyd5 z5_HS4pK>92sTwD(kk%;%pl#DWY1->eeoIB1Cr%I(c%8mTixAuKx%5(xp3^wxwjb!4h_fui1>W)F(<@Vf(||WiDT#7I?wm5hh)`|Z|TF+=b-kb-Y@@p zWvF$jhtdedf37z^(lvNiToC*|-|qV56U^&bF7#aTzkI~T!Po1}+d0gUP!H0lXWXzU z&>u9$jj;pIKM?xrXg*&JHjvFRKGi17X;iCtOgtslkHzhP`VsrWuF#RxI*t#$yZU(D zJJ)^H2}AtH>XUn-u*=U6wWNJQ&c%kG6mki@#I(D!CY;gz(|;7-k5ea%W#nqVjz7c| z@wdQ{A-^-F^O<+m}83wa9NaL3pxo*EC2`-gfTjqxubN3lau>FL*EIW}^$U`J`)yN39RR{!oW4jczL&~b5c z=*#0z`B%F9rEzW0z{H;7;uS)k$nNXsQyVs~cdHPuZaewy8NOERMBCIiCl)hic8T4B zUl;4XA&eohFP~%Vd^n5~_bqVGIQUz9#z%<@j3?ucFW1Yj_G+Eh@4NXZKHqkU1H@}$ zf}xd*;lvd7$9~;?&)qX)pADePI|gFUFNI?yM^u+!PxPzP`)HMNR>xf~aeeSFt`~X6 zDdD`ay$6JIEsv)C$&c7m`4mm{fgx`;aBnYHQir-a=z)I>In8e;TwK4!C81t)MW`L! z9HoVemaeWw2cw74*w|^dna1e5>9gW=8ZFz(zx*uZ;p%|;6!dD@FFtq6*eT=^a_OVu z+u``QKG;kD*vgfHj_nwa>q6``eOqq1TubH^ zyda#@O~QAu%WH<1Pfnm`L5<|rxPCH+azXd~Qgiw91ThJl0P!S#A8InoUFvo5{ktAt1Az;*3?-(`89`**f@@0q`?*L&sz_&(d=ys??a^yons zJCD9sV^uyRPo*QNAJLTbP@EV1nD|nRuZLopeh)-Ux>jr$FN)oR&!Qoo6h97m4INX@ z*Ye#voloE!wv3ksZA*QLMmc)y=#+dCU!^DSH=#FzK6ZKVKYxjB_(bM^Vk-9lcrNjxFU>A6?jKI`_Go)$eQ`Z&HA?&HN@iTTCtPmk5( zULk)Fd;cW#tbH=rJvPg+Hg22~*RNcL4Q7w!t!f^l&he^xIK?;ecXnP5t#)GE-u%xxcq~dQEbZ=iBtJ6K1D3c4v1~}G4UT?_NJ+Omg_m?qx6)*d+dj}jlbGI#E{ly zonkiMK`;1juz_rgI91(>ZKNyE0BC-6e0Fp0J#%Ml=%_L4OXG*tQ^fZ)d2t;dO()>5 z*(Aq7zZ5?%wsY@FHDtd1fH*MtImeBzpk_leG(VE==a?QIpPST@*f#OOF%$f)#^ilI zz-Q=<)?eJ`Om09 zuU;L5J}E{~x43`k+tZ&yyHR^`O!aAq3;q^j1oz01>&Q1~x^fP=gBqP!V&ooj4S9sz zgI?ykaXumTkjr@8`b1jN-4&Lry;Q4n1n(AAM=^m6wFLNG!&m^1*y7KTH$hi}_ssnE&O6 z#bs{|HtO9WK9kFc(ZpZ}hnS2VJ2dE+;&5|e`qP z`s{uh=L9Y2H*rbOhiGTC8QSXfz37){f>rHdGdTe}yWHeH^Yj@$NKPeZpw+Ay^pdqh z5AoAty;whJDKr;44sAvKWV@+3pER6ZW48(Fa?{9B241(E<>hPoBI!2rHyRP&@SNBp z&}eRCwdSaPrV*P?;U^BfaU3O6XSXy2KAaczR%EJSVoJ zAFu^#1&%4$1N}Sy4ewjdXMEoJd@q|Mo@1+QpLs2|S&n8uoCmQkJwt3NR;6(`K91Md z<14{d(!%vX@NIk|zb4P&JLQP_8~85w-yMEy_xgwVUHK#*tjAb>DIVdw*lPVMViWnN zx&!}Bn;G+EXf^!%=snkq&abik@V6L1tzRx{S)btp%@gDs#69K)@)2?|-^EwRSA0JW z$hL?V^kPk8I|t?gs5NdE_JI%PYxraPMnl>q_-gx2OL7eOaL0sC=a1A6`Fj3eFPxab zF%*x85vCEJVE-6#idaIOa>Oe7)#O`HuY~h+NXXm8G5-ni$#~sz@R`AQZM<%s*6sTs zu7T^ow!n6QzqOqM@zwZyI?l4q*6TZLk2jFy5ol!Sm_&T&qWfxJ;Zz zyHK}tUs@UlP5GPQT4y_4_iO??U@Wjh@-;c1abmo%U+mcQ_^+%sGxm)&^bm1BKdkn|4-Y+r@16F$bgiMq(-((6qJAM4<7{&ZTL6^r{*Jpp_NpDJ%#9ItzSe(=3~jNFAUmh1gC_-6M^;HTw)YMg31 z(|FB$mboDKD4!X4KJdQfrq^e?N4-c*O%2t4xzDlu)4tMD?YG*Wy6Wg{R6}xH)L6wL z>Tr&g$7#eXI3gTRag2C|zob3UqWDe6orcSQii5;M?+U#X&e^nPv1@ZB#98z&ao0;j z3^wrG>)s!li0eQsW<6psx(H2#pLRXK-@a?Fu&mGc9_thDx&CNf{I}SXKjORS4{Fw~ zb3Ttw!AJ0i{38D-PUb5Io_k&FY~094X(aro@k8$bU+U3ov@^?Dr}4%I^Y8pMzr*Jk z!)yba!bb8j>?u3Tm$1!S2OlEu<3rea^=p5tS+n(WB)JkB&Th-2_$BsPUdT4H%WN`V zCP!nNJ$lYw^KEodcAbsq)7bvu&#d3J@F~7;FnkRE!hhKAQE!l=_-^^1{bMKDR5q9` zWYgrGY?!>oF;NF&$M|Kn36BH4EwIdByf&7nQ?h?wTkO7-BL{L$M-DV>ADxQMAZ~D7 zh#}ZOInmKyl*-CI353qbFI_5>54SY zv5kC!?T{OKgxADp#v3dPpA$n`hkaEmwO;?knT~~+({Zug(|FDPSjMvcO>dWYd}xEy zd8_Ri$I!832S&~+rk9^Co}21-0pB)q)RCvMC)0nA=U&73%vgs!7oIQvxcIu=*Lpht zmFKcye5#ro{r;Ij_f>!6Tj{`ZTROkogkoU5i(|FB$mb3gW z6WG>)b@JU{w+G`j?@xcPbyye1_o`9QwfXwR^X*~h)imVX^j7{wt|kUhmlp>NJO@7n z%UIUu_$2G&Kll>g#b1as?5l0Et&SDn;TSrW>Nk!n-IqVnU*mW??hxBJ4{B`AiG0!d z*<^zArC-Cj(>EiRbUwvN&avypH6ibj+qe$JRWudXsqtiN8#Bhs7}Lgw{sQBaJu#N` z80g88ck)5}8UMup=;7vr7oX#`V3gt@TNj8#5($jd-`T%7Tq<)$@D>#|Nh!goyj6We85hmWvd z_ED@Xr{QXL75udrg&p7%`9Z#!{~Lap&tpf}6WR+OM@wOI*d8_D>3oeHW7pIu**$iU zO=R=gZ1$2iI&w0$m8~899P$Erg1kXaL2ILNj$F(=(x!8

OIFl^e-*-U#6UF^a7|y%88v1zL5XpGx<>dPYyM5CBD}=<(K(v@q+X1yz_PX;Q2Ow zO)k%m@n5ufxxP4uhEJDsT@Bww?-QGf|6GS+y`j&E*~D{VOL4QfbNH<*`VA-=1ie&F zMVDZM*p=b4hE{5vu&u@o`)mvuOMKI?BkTv;Z|w06{06&VywYLhAG8_T4!wr{M4wVi z8TrttQ^}v?F{4f;Hv^L*8 z^fucfN0h7CHrgA^+4;YiUPev^`zY_5*6rkK(>bDJLc<&N#G!RLZt~W}bvk+%9OI#} z4IR$oc+I0_EIV{JxuNx0H$9LxC~uUj+9un!_<3<$93wgFI5+wx9ap)qb2H8f?Ngpg z50%e?K05yPIJDCLq?ZoubaBn}-)H^*``B!xk3;L;Z;M#TU_$I!`HST(M?HdRDt1%+>;8XZ!{zT5pm&kYd5qYlM%0KxSA0lTH zf6A4H)@Q8CU*st4fS6XSJ9NIqdt2BS@x6GP?V;JQM|{TUZPLs1nV8n?)N1$%`j-9x z_L5&@P_QfviN7gSmtA9(yed2zxVyqPEMytmoLcQy;@q|!YdP+Ps){AF^-VAlG zO=8p7JhlimuxH0}L#;~x#*5>np$B=V*ky8`4>d2n9eafPO}{?g5O0h(O?dOHH_zM; zL_hd6Ubl>8^-KG_^;n6)X#jk+SxW?+idFwVSgSUkBWJr=I40aE0zrXDC&j$ z7oWwa^0SVw+he(xw z!@9<{`0i1|;$zgp_-FoF>?HRw4&^y=9Qm18Yw_I1*qzR8Xboyla$UKiywrFf8mt&_ zI=`WB&>-o3v=26e7D88)+YH|%{uGPyOY9B3McgWerQL{alW&L22&1c~+ULNc> zd(O_kGxiVjo(_pm#}RRC=$-pod^^lFJ~MtE=Z79aec|RoT^{B@ndf>#nDcyF+!cBV zHP9|mVcfqtF?TV%Haq;YoaKE^FQ0Y&J+6%_;?nqC{5sAJ+w}eTM%dnC;?OYH)_(D8 zuMNJDpW;Iu6TWwiaLksS`t446oZ~DHbnN9o&WAiltx3!v@1#>YpL96-nse^_%X4YI zc7)a~N0v{Bcjzl}2D!rUC9Yf7vG~cgZVVW!>;ap^RjbNXQQ)ADV#}a&VeY4XMc+T}wWF6&PowR%-L~I7PxQ9x2ehAhKkT>t*HfYvC~gsBOve%8 zhyTQgF{a}npB2|wkGkThHJS_Jd-R$3&N2T?4~lKkkK(x7HvK7Z4o1Jqc+LBxcg6fQ zy(@ZJ^~3m%F(=nHIM24*coXZm4qOi-PclB`e8wtYJ;szVJjT?4`A+$%>rbxFf4M$g ztH!@vY4|PUz!>4P=(p?zotE}MhoDK!4>=RW9~L9~6R#VNrD>K#8Q>=!@svaru`9LIsb zT`u@xnjHUZ{*(O0aiy8bEA=SQznlm8iI~7SlE=zN#1e8|c?h2&2XVf|DXs%{lI>&L zTtjR%yXBg54YKjBPuD8DWX$Nra$Osv`nJdTFh*d^c-=U%oNJ#wlNazO)+df<2iZhE z$#u;~(MH%+zJcwfE3@6~yc&i60va-(C&scbVl28Ef2htqVysbOs(Xv^oDX_7UE4Md zt%|k;RIhTDJInKyuJdW2q(iqv-#q~A$-t_*9kH)^!&TN}~vrqQJb!}`6-@?EA zH!L%qFOPiMdK?SK%ew#D{u?*W%h>m6KVkozQ|Hz><|m9Pe!{pizFiZpjm7gD*N^LH z)D1_?kiT%fscq3K)Vkz3uE#M)Nh+|89vwnEUy=wkFDeueI(4|}^1 zx3jx@$9qDa^6`+PoEY*G@&8ZbSK;1D@&q}9{DIiSx3Kps#Tv0@JR#NzKIR$0&uko< z1fSv2bFaH6lI1M#bB~VI;=ys>kiXnH>&{vK2)Pjd_KTSIBj(H=5Og)Zl72_y<3sq> zrGth@Q*^BOV|5obEcv0H0osIBu}y5L^XXi(zv5?gN;aB35$78}u1WcW z>sh?UH;9ksg?Lm9DjpOkE^5w-xrgS2{4@R$*TvuBuW@a7^xW&-x18ng49ziOOmU?+ z*Zf8CFrUNM*akMr7&g9*2iqzal5fzD+46UX3xa#DW4x zxps}%1?8CW-TVE8!OHWzdmp@?`}{wjJ?5Hy%`)1Y>)n3tBeXv8b>l++;TJYe?m{>} z>MYTnlzo=Sjw6%Z{cxtpN9=40*yx3&xvL@XKC^P)#9a+~R{q9!ox2y#sGLu^Q{k>S z-Aq62Cw>Hb%$XDaw{um#Mtjx%_0H@n{tWxy9Fhy{6=Q^dm`nVW9*XaoQ~U^j!n??S zV@y}3-_molOX$HzyTts{@qNcc-}lWj2#Yg!(&b|ZxF4P^puVF%O_#R5=jdw@hU3p_vB zO6JfR`P|4Gb}*kBdxRd>{YNqe|6q&I=lCng7J4YXhE0-eaks+Wbf!bLxGTV)=G>I6 za=b6#o{+sur=nkFx6sY_Q0!S}|Mrx9#z%(#u~*0x{usQ``L*+S@`N9ZjdF*g^IY>Iaouua${?2iu{e8k{m2A?qa)WK&C{5_sW4<0-44J+#Tk%LbkeA3|K1|K@&vORK{N{kX`O|?PndNWI z*k}9`_#OUZU)T@!0oxo8Wrx$ft#j+!dZxRRFZgltWz?T5cVL|ru@}e|Yt&s8<3tC_ zx7!^R_xH$Y_xY^fEnCQtsRZ@3Zx2d|JMMkE_T0sUP3pa>1ES{!BIkTOoTwf1R=9hr#cT zXA9jKaJB%Ry}R=V`STC*j~WB^ul24i6qQRdH{%G)f1NzwiJNVUsJ@qStXAgdH z!5$I^zdW#?ZW{dhfDT80Q>VJM!%wJfK0$pW&&VV4-h0q9%%M3qH|#rpRAWY$v!3|6 z=%V%;yM;|`4YE->hao??9e(=ah_j@vUBbMaJOoe|x}AdGi4o!v-M-_`mKq z;QQkD`p^M+^N_*A24u&-7(8Nt2G7NjO%EC{_Qn;j~|>qN3M}?>=t~9+_Tp43G$H5;#`FOZf)ba>=nG2zQb1I zW3m?AedPap&w;xN_&!^Ny+Iy3XL5(G@44NtqXV(6eG~Qg!3_((=O)jcA4+#m!dm8g z^i7)YrTBg6QtqmolTRD4E7>(<4;e=0*c>Ob z^poK=wj9{`=c`_f1{#vC;A!v$$F?!h&4XMxURI-G==H1Mh zyBYjs#>ASz5BQp`Cw?!sGWkJn(B1fA$q9R|@2JRyzJtPN_-@IIARn1RwkV4afx5^Z za)>R{-F9JZC6|uAGct&M!^UUp`{sxJ@9t5y4L(TUw)Y-8u($1B_ndxU;2g#ow0lu_ zsKg-g|&LYaye^l6M%q)!-fj za*W@N|BVgx7q|M0D|ZFRz-JEh>l+4S3)_Vp&2}MU?>cbyWDM9Z{%$Y(g>Mm$bFPA4 z;$hBJ=*zxq!DF2l;8D&O%(Hp5R@gCQ3_Aut!iVfP`<|bH&PKM-aqJs<4}FaOMXu11 z><4n>e+>Tn;3o$^KJa%|esu7|gJ%u??~A{^A`d;xp0SVUYGjRlX5ZNl>=yfgPdc3n zzoK{2N%5H{4Cw82C_I)r?Vy4EcpEWF22{wc44>h z<+4-qf4S>meCbDI2Dm%PE|=yzru<~)z`EvN^Y=~EuO01?`_@1GOP|aSpV(^;+|Sl; zx`6#kme2**Wn>7Qk)2|`$PoO~obhAOGwBA#h`hk#@ee*H=kocWy7Pv|IO}p|Y<}tC z##}!6&3k8Nohzg4Nj@J`IsBE*ahCz=nLaPgJX$xM@w09r9?)B(_f=FW4Y_$!rpSX?GjMk!QxxJeoIi zXztrT%-tFEGUgKPy9dEBV z{{VY|-OsMDpM=GI(!ic^|5zFI*ca;U+q@{f3lHPRrYG9J^f2GafIU>4eNA7}A9jpy zVd!`CJo+B}o1Vzu^NfM-`F~;Xs{{PBzYE8%dA-4126wx-r+m-(J8|xiv2)&k@Q?vN z+Gh^__2Bac&iS2@^QSrM|CT}aj=N&`JYSmsi(<^DD`zp}%WDl@cko7ow;7Nf&hFVV z?3Rxm@SzE_S@^`*EFUl+Z^$8XhF|QB2V{(T{j^24k#o+5*e>oOe9pj~H+SIPb>Qsf zbq236AZOisAX~^`dN$c&Jk77U{D1)-YCTvN{`BiI-mVUccaUYDb^-k zhu%Y;(1q}RV6)J@=t^u6KChn`kSFL*4Sss?-v`!e{5 zeL%l4FGeJ6zF=e+rY2ZG$^DX{de>cKiyJrpDb@V+_f2-XcNcSjwljI)$jRR@j4W|eA z8x!v6-7wI`za7{E_8eJe4?c9DpZd$@vTyD_Ab;$2J_Ei^x&qyhK1k2w)23sRMdp?- zobQZX(|3Z#&oiA=8r>qB$2xF+L(j*1$Sr=T^!#`<-ju%$ucGhc?f5M|OD`rL$SU)S z=UGQ&3_eP?$5ZM0cr89FA0Le~0_DjPd>cQv9>rN3c)T|9vt>82vGjp#AY0i)WG$Hs z{A+9`azhz)tCy^`*0rq<>~pe!9L1+maoHw(U%?p@o{sn9_xL=Wi~T@XqIhdp)hU4w5QeAD0?2VZ{y;cp&%$KZPg|9S93gJ%xxKXvkz zX-m8O*Y=xzOJ=XNq!;O!?T_C!V^UQm! zZ(aHxpImUym>%U!o_#^rqf-G}y|BGY7gQ&^-S0(z{h}*2iuF$B{o%q3u6WU2a3!8E zjXbnJ*!_5(Hpw9VX#5ag#2?*lq8s`fEc_PsJimi)1o?sN)9)L&7exj+yK%=1-*v~# z`NS&^?l5p4fRF9X2Y+|)E`xtOxc|U?fJY2IWx&@)_ONlB<2-%f4#4vU&PT|b*BIP+ z@MZ&d%Gfc^^Y1fwj{&=dJYmo9h5f?;+2YQbdjRa2cN~y2&VAJ9yq_K8j)5~6GLCPL z+|w8L6Wmq!=mA;ddmZniY=GkrP}dxOnlZ~Vx>8vowGe;Qc(ApD&Jd&OS(fx$Bd z_L2S7K491X<-p%JwO{RDdMq7`{es72yWn@u^!Y>ZN_>>8#9!%M)}OUcp7JqS)8q|( zh>qkekX>P&(V@r}elfhBENN}gZ+y?n|L6_{8=6dar^0u*?&n!=e>O1obRY14M~HLZ z(Kov8fGg+QUv-g3>~i0Oku!X1=7|5yd5->(i*&X(86;EKWA-*#!Wa0A0lUrp9kvWP zY+T3&x&j%1zw#;KxBH$KJ{tUA`O@R@V&hKU=O>He(Re&LKtIN(@hrBF^+H!b z0sm$znM?VssqQ-Qee$)bi+tj1(+1mFAGLq%<5dgeCFdpb(TmifUOKyW*;i~N>y~X` z?OJ1atu>7A@=4+Sd`|ejvqHQ+ed)*Me_vpO2Wy=jP7YYhaO2?VgQs0U7SWIl|Cxnj4_ z*{lsV@8bqf7|=WTnC<@`9=JP_EMY(6>3BR||MCNT-#G~WVQrEJ?sMViK*|eg|Fao&mhhk^*0%uBhq`wrM`1z%aHqQ7`BnM z&hF%I>uiP|z`mz%@o&ECfRBJ4NH4H|y8~cf(bsM>Xm8q+_MvfLuY#~WDb0L`_ONt1 zDSdGNoi6;P`_IoCKRaKY@pk9O*$;k{tz>T8OC|HzM&`?%BK(Ve^uz%h)ES96rN;n2 z|CbEdP3Ak>i5@|p#uGae$@ZXcuxH3E=OAPVTY^1oov|af?0DYzVSiyjZjcQ>IH32i z1K0(RAAHr|-!FXS{69G2;HwwD=E`4i|C%d*YhAe~T>M`{-!T7|!;=S38KBu5Pn`dk zNFI6NYX^nD?&AM(d(!-0LCUaIzIE^)2jomLhO8pb$Qg2vU%(xi=M0?Xvys?KWG24= ze*;^QtzxgSSIFCZ&Bl+PkzD3qG`_}}uaN)H*gK~(7tXCZ!!kcd$m4uZetIcech4KS zMn53`*-G>Y_NP6+sXhU_Qa=2EUSrPbM|c98+M2PRtT%ck9|k*~T;f|MW9XB7H}*OI z8$FV2q09N(l=d~-1fL{V>}R~Qvlsl7T)|u2sd~r3K70Ad`8FOuW|KI}*fxo8N_;EH z59a$w-$gnLdCkGyF8(i|e=ukKY>^Kbe8_-&c0iI`UI#=O0aX0Ke z2K*-OhtU`C9J&DCJw8H@CpYi}K2>*-|MS4Ux7X<})(#)EwZ;Ezy_s8cZ~t14_AuGP zCl1yr9rfM=x+ot5y-`_pxT8ci;TiNJchr7pz{a)~oQ*h>g5jo(jxauVMqN7O^6sFQwWz44i82Q@zAjx!nQKnIiG_$f=bQjamE zZ|Rr%+5Xx>#nm3)06U4SR=0N8PUgW656AD>?|8U*!Nk2S>~XAJ12 z#>ZIdgL%_8eKwaj4V-zA_3ocOePHkN!MT6I&a@WrAiRs8$~t+_fW6|LS@s5h6rULz z)_DRSEIoyN@hgL$9qt{@*sw z4;(yn@bJOM3_fx2S%ZHw_~HSXLPoH|$s_Hux!E~vhrbx`iLn{TwsfX@&%Wrq15d>t zy)QeGJ&E_)bM^?{%;t3N`K1^47VJy%*BwBz*garyMnXQ{aPfU2JjIIpIUyUXSK zcQyilkb6^qYw$XQyA0f?dW(Vc4d0L8+5AcGK43e%@8AP3?hHTt)(^jO|JQxtM-M)4 z@TG&V9DLp28wcMy_|C!i4}NfPfAy z=OfPFy$hM;@4B!b_}%>YOq_-IyI<}Ok(cyid)S_}SMe@;lx~f0;3J#Q&>I7JER6RA zc{d+QFHKLyo7l4QEARdW`NC(!Z;B6~p6UAZQu?X9_?-IjIQ8n^d}ey(`wKi755{Nk zR6G_xV-MhE_!7H-zOW&VFHRmZfh;f%csG8fUb;738sF6yIS&WXs@XAbaQGM4Vo|G>{>9_%e+Z`|o%)<-@Fa**62 zzmk3Aih1CBFrLZ1qwFIml_T@WJM(E?jkPv^Z9v|W3(jo#NAR9xf^%PEV2s$Yd}jD1 z9TTshcksdBtN0Awg4cHTOg69~oIyATa2`P4{)GWK@n6>e-Sf;jJ0LUoSowsUDUcs@ zcs?Wgyt4@M1cb#&<7bke4B)H!;Q?9jlLNjk?Xh3nS7cx7Q|ICAY5pWKjX#zx)naKp93GBapVi$`kS)OqSOPz z^g8{+uhpqc-z({V_oLY@`~=Rr$m;G5um#2fKbpC=?)eOE9N-()4qb+S z*|{ZOGXIEk@!Q_&wpad-wD+CAY301w8N+uDoGtLl`2K?K?(CWlj<4hC^mja+j{X4) zp6@r1=DrkO?3_UvXAP)l<)zE}fWbc-(EA@Y;9qhk{t1Im9(?-XGY5~nfH41(^hXTv ze0@;=+Ya7vz<8Z1yZZ zYH!E$>_a@j7`snrKjIC11n$m3cjSO>J!X~2ofoJj3 zoi{5(HmJwGAcN?H_~VuV>45sht|ANe-+dtyyayiZo%&7;-zDqF3A(v+58$IS7Gxn` zoo{aVjGVF4$$shr!p2P=eB9WnQ~lZ_kBk?cRiC|s_wjz6ozsEyFW~Qd>i7XZ&xXX$ z=$7YrTIbYwkh5xhh)*ea2C@R=IYN48+~mpTU%rdtlX17no%PV)V3CHm;cLn{U(>e! z;H~&7{^>pNXzzw+2ljw>btg+c`SEIJg6ne@Kfb|HwXN$UlC#&Q!=ka#P)8qBgZn z9{%?M8|`-oe%^=dG2hm`y~H0$f3$z;`R*>#@qGvPigRA@b_2Zca|itYKQnM2*8L>< zGM@b40X_P2246AwhQWUv(8HfLc-G)Y20ymoS$ul=_s;*h{2haD89aILq`?gXy8YJ- z=<8p#;930F4jw;{FNl-=xWShUK7a6-0o%cyDKg{z2JY*!Kiso&rr`T@ekkYU-#B2) zkZbL z?rfCZe}ucJ(()0I#lqwZy_>Gh2ebVHcbl^)e3%Z&H-QJ!Fa7XzekgwJYzVwQ{|0-4 zJ&--YmS8`SA7l-^5zoP=@GiP8zQ%TC`_doq9lHDG(Re04^lu0FG=9pi#jo)u_dh^b zT>ew}H_ye-fc~xi;2BCUk2pNqI*06ayjox5)%wooPTsKJ@onz}h2!Jw6Ym_d<=OXz zv+0#lpSrbm!+_18-|PtQ_JsrQ%ukC)lRab=c|=C>&1_#jp3S$%4~&P=J@IJzA$`;S zrf0fW6g-QkH_{*FrBimN_ms|=zNrj;s(d_u>#^0D9$UTDNrr6T$9OIoLx1M0qj4>d?tqW_xm>E0s2>=zEy`yU2R9pLlqE%)!q zT5H1^q6g3w?SJ+-9^}r^GY0N)+;RT4-@6PRHgH$fcM#5YeRua)1NTMg@bu$%9pITC zGT?W2M*DFC=e3_W!2dtx0>a|y;~zPA$bha+-~K0q_ZYZ4#HY_Ta8~RNk^8D349JI7=|ATz}~^8|lO-Z>^a!T%@L z{|n}P-W<5E>Ab%G^YVtdpXr-4XV2~@yE8_1`1Xyg!1MW&+5PlxdN%!>{}*4!zxj*! zcKCSsSLpTpKJ@O*%juGMIX?xx8}GvJ=-c=szG+|CcX+dX$p1ob*!OSvjPWFX#O}Pa zkGnsKKiU8ILAER&;Ta#m1MGi1O+D)4cgR1B|Jlph?ykQ5%^!qc^54+Yk2*iU_P#s1 zb$|7RzTegkWDm%%%?|;EIPlGN;-+Rj7@dG^M3kIJvzyp9^96!NV_}c&70AIj&zJ7o|eA@s|qAz~$K)rYd z-wDtc@h@$5UP^a#pY6X6_}8C3!2jrx_@es>bYkzr-#}l+Px&XjI~|+9#2F?(oN+at z=8S)d%(J%e@cd_FFL{}KJ<3vNR@NtZN?w1%fUc1YARE|%>TbQ7V={n0P+y#n(tWKt zYtlL-Gs!cygL&c`@|`R{kvkiFTfXsb&d39ImGWzuGk&cf9gqe5T6-*|8AJXk^X*;- zn}w}mpYT!IV{`^~ynW2yN5`}G@dEcB@D_UKUFUnRcN;u>fM>hA|6_w+8}ME5VeB(g zz7+lvessM1!v=gGKo9=s1HAd44)|gCSDc^X&v-9>i~r-d^nG_k-SxzK?>TU9)Zcpw z`x$Q*k2;5SH;C_FA8$9{9C+caUU=m$Dw#v~b{6c8k-uB@gn_?Bf?t2!z}?pOA2?Tb z=D;?2%fVX=++{*Ni+{(#KNz^z^p6Mro)mun;RAJj#=zgoa<|EyA%7c}zsL6wPaE7g z;QxQt1w8vc(D!)XGI-L!_k+Igc+}wI2m1HkgLfLR6<&96yMeou>;-beofvoY&8f2k zzN7BrJ5P46>mvr;#rG|NvqHK)-tUYR|L0p`v+>FB8Rze%*YmUB^<+OEv@>S=i|$01 zvM=0Wupg|!eb(oE$k~}SY7ek~`@0J6F9AKtzL3WMOMk-O?lPcLv7xPRdy5{~*|c`- zJ^X=u06LjIpyJu&?M=ERd%ix|m)Ya&S$f)^4A|uKANM+(1=7>-EV`P#$M^i-Eie?`u=mpg-4~HBUC@5k#?KAZjc?0qCO(mU}T`l$E9r|~F#caM_a8_&VR%$vCk zc)0nL4ji|})g3kO?VWqy=gv0-#_X2{WEEDj6A~2-2G0E z^{o=$QQsOFFSudg-VA={F1l}(jHAD$>pa1IK(fu9|5q9CvEZF|8NBh}%`W~wkbBMF zKh&4E9=z?~Z3e>P>F;+N(BYMTrNPS$+*_x^3llr z=P#r=iY<28Mw>h ze88C~Ka4fU*V5e}dO5uS?|#yN?|{9VKZUNtS3=*R2h(@yv-ETP-QKdt>@_}6=S!QX zTL1JnzE--<{u>8)D}HG2;3=Ee(P`+2_#OR`PKoF7XMWSby}oQ;Iw{?h?uviGza8LX zj~n1+@;-4uXH}N%48l(uNPEJ-JwW$Ll>3r_y6{H!^2ZDwGWfv3`wZS~@b?FAGq~5_ zo`W|V+J zZZeci1@e^)lqR2YWGNY7P2v0eQ|?5w+pJ?}9PHCO%v~@3%LfnKC3e3KUl|tua`rvF z&EINsN6+_5?vvoJe?Fi;(k1=<1-dC7htIy(fS&=c{fNQ8960xT)PR5b3kL3u@Kf^> z^D{qX@a==|8}Ld0)Zmu~zdrb%gFhb-khikRHRxXa*;E^xQo z-0e26Ke*%I)dsg6yztg9y!CGlerE7P13qs3aPOTwpnLO!;L+~L;LpE4=&mAvG~Y$K z=0oTIsQADEeUpy+&VzRt@MX|{@kYE7Z=@U3N8fbtMgukkevVh-v+z2D*BZR~;8h3o zR(=o*4-X~t_&@kH@JRg8H>UU@o{#s@W$`}$R}`JroVq9Z=)q?V$WV93-P`{7i~rx1 zjRVe3KW0Gx1!3_YIr#7aeHj0IzX2aOo%lWjywtrq`f>N`q{;I>1Nnjfq-XcU$#mZ( z@vS^)ppB0jX!~Ccd|#%YzDLKu^-q8N4F~y9&-y7{8Tl)t9{Jo&cP6ZD_xjue)NLk7I+fWS(Bf9{;pk^{~K`IgCmCk*g^=biL;KAC-{_rURV@}ttR=}7F_yAAMn zIv;(I4rGtoulP7#Kv(5sz_akGe>reZ4=;Vs!9N(_F@I-(-@MV_^#^#+-x}akw;s$9 z*^6H^__M(u5B_lQg25kMKpc1${|AHLAN=m%w+FvD`1Qf_2fucKn_hg=i=Q|6mBF(I zzcl#8!7mJce(-Y_5SA|gZw%C_jXxdy%Xp0_g%pN)5Z(t|ffAF3IXNmk1bXYzMJUPA9nXAA3?K@0&?CGcY zEj`p7e7dH;4Pb7+cHn+Ke(QV7M-BK8@Fe=CvqrqnIU-*9P6Iv!x@UK_?{(#7@x8A2 z6x{c5et4Gw-@@w-+}k6k-RHogK={=MuW|`@yYt~b3Llqy zAMP;x*x;E1XNliGaHjUHgM52mH9lx(d5^n*Fdqh=AK&5Q2m4+K-pH@$EXX+{9?9S6 zOvoKS{$e@^pDsPwxti}T`9b&|=*M^_y_ijjXVPQn!E_>e5`D+_Bz%1Qsm^-nrgTy| zuD#2acXsm$1H2UV>>f0ojLu4Dq?;*&@6u=ac-8Y2gD<~;IQkpiQhxk->$@K`cc|&R z^uMKk8xQ@x7ysUi|73uF(nIkt zb;i5i^yW9cqFcZ1&2M}2I}XT+`wZ|h{Q+Tdbn15=(8*DdAEe1cAO0r;I`@MHA3J!| zz*w*?_@wEm=7nF~`6piiI|whdo}NEo;Jc&rjsxF?vB8|H&{h1uRnHyxcI7s=y3G|^ zkuLOx19!W!HSt8c6hGs;4D4C=zvx=;Hoz-ESRC+6zR!T4k$?7+2JQ|&Zh$A?FZdE3 z<*ppQqf>VPedfBHtn*_*r4uRM5-0X^6G7K#Vnen9_y`GLu$EB@)gnU^~ZzPk^< zGk33lV{p^pmj*vQaJP-G;+X?{3_rswK^Q;t_nzh9TmF#&Kd8P-|Ng=E4)A~P#BWMJ zcRvIlv>^}O@Iy$*a>Tj`q*9sc~V1xF6yhto**p ztLM!I*0eUX=d929-c18Ml^&3;_>?&x`PzXy%lMzOV0WN92l4%Yvo^jpcG-sw_zE5} zc<2Cq@IXH2dVFv0Pw~t0-#OR1`{2$4zF&8;C>DID{Hgq|`~%-VU~{tB@j1K?KV$_@(UI^S-)z$x{cR9@(7iG|kuMU@0bz0S z_|E&egXa(MGxv;tb3g~gL-93yRo!$n?ch!EB)(0cx8hH}>v;SCZ+3=5$Hk*QWkBzJ z#Nc6re?EA?;JpU;|KA(@odKPV-R!;qza{@A{hF^4zcOxoj=woDj${tshC2bq7|$}^ z{EB#%^B|z(^|zd)KXV{2@R@?Y0R_VODgDiV{7+eR@KLhM`Ch+l@NWjbQDGZc1ME@! z&bKObK;M4e6 zb?zkKU3^e{pZ6HJEBg8aXIyM!XIS=@zol`90iGt!nHI3U#P2f(wzRVZ=OpyQ;~A8F z%U9-X2rqovfbY#6Pd-h1o!=84KfojDIsAuwiRqDaJ>SmJh446(ACZ4AA0%B8f8}ds zH}a!u=RXekW!aHW9k>_joZ6YS{_*+YE&AU17XHoO%a_MDxX-uwd%Xv~?hdLp_-6H; zj>;d)-$}ovxB2!@zr2_ClwV)?9?1wgwJ~%CW$ek5w;4DWbpONrnse*Pc_wCSjj~_Z zG;}*Y7&gyi2465B>*;!IBWoXT`__T|z`xkvL(5k$@8bsiR^V@@D(60@y<~s+8#iod z=LI(n>}@_V{xNnmK1f&mM}rR2zxBb-N55ka;C0T;UT;3Y$LIpS=XryHZ+`GF=kn|$ zXT6-(>{mR>+4W5WXSjTS{7U@)_M$y)U)s}n27jP)9A_{$44l#6E!m*#Ph`Q+$g~=DS`x8~x`a2l!a_F@1`p7PH zzsX%GeMi~lcpd+yvkLF(tit<(Fy7`28>MH-@BXGbK%ID}IAeq7Z61oRE02fb|06DM zcHY5v;2xB@!^8L-+)E<2=xN_QAUDZHGLYYnZtt6LK5ly17Z045;#cH48;8w9e_#jE zFYF5T<@-`NKpj;-mg z=ko^6gx%L91NhtM$@q%DJp=r!>`A_VHXVJ+xr2LK!p;}`oo6x%|MATJa~1%e*@SGv z?)K2r`Ec((z@yla_>^y|oDt-QyWQOBX9K?QuU~lOt|$G9PWH3`9~xhXwf2pJuN!>% z0Dr=h*nxP^)h}fqeuW2#!>8Pt1a^hEj~(D!?$CMWcgDBkSL{Q;x12o)<6ZPF5RR9< z_uTJcw~+667QM@PGrQ596O{89rCT|F26m6TD|njwD)`wS4Z4fdnI*mDxdZ+=cX8aq zv!|hZjr?}@tvqzI_yWI@JU=kNPv|uKZulG<&0QRJD?WyY@zb+eosaNSu{*auMhD{4 zqxa;?^Q=Dh9$txWN{gRq6TjVh8=m7n(K814G=5h9@o;(?z3tBjjTv4f&Nr zJNb+$*+5p%b={+3*VDWB1*{G8Z7tBjtRMc?Ck(8o8wSp!$U%PGUmNg8@cmk&cOAGl zV=q|iWH4DwKl|zd`Tc_fJ{&$B`jo#j=0lmYWJvvZ&3q1<* z@r%bx=q~uxRjcn?uJ9OsD*6}y*Ut~|tmh1(_!oZ}KiZ!RQ2+movj%=OzB!B383g~^ zZ3mPZ{yKLr-8&b5r-A!6&Mur`y!YTe2mff``F$_&?(c8w>ePjc7)`5rOr|ewzEZ&J{;*VcEKtE@IC(=jp`gAqt!|ZnFKer#)bj}-oW8hv4 zU6+metU)#^+m!A_uf_)-IdJC5o;1gJAO3}B#j8+0E4C$H+C2yOTRe_$LVEexp3WEW zH~JWTpAGtF1Gb`j?e4Sl@#1;xMc=$ygHIhiao`>}x$8S^HXlDR`_7#YGu zJMar zDFZqceuhUI7kmwjm+>=4{D^oJo=>lJ2h_b#a)vB=@B(tzDGAwmj9hU-u|bH z@!{V%P@cbx&c$cX|D_&$fM1JG>%oJ6I-vj2gXly5aPYnZ@gF*n2haG}!6y$seekJ+ zzL}v*vGed~`sKb)LYKnB?>2ag0shT5=zQ5-J>x=8b@pN`>9o%nu;t7F8S&V`XAj5> z_eI>L_=o}j_yY!f^ey_-`wYlvJ}rBpZ)DUd48qE2gFfb5(_Ikvv7K## zu)74}`xXzcWLw_(!l&zpxJ&txQD1zqZu2V~&y57>+b5>Ml| zKyN>IlL7xe`;p&--NHXi$MUzH=~;ZyY#h9c&6wQ3X?$AjOFm|M5#M9~xHH1f^=j$!QaE6U*Jb{4nCoeAH-Ym5;hkI^PfxOS7*cViQ!A|l>_zQi~R8ToIL8p|M;l* z#^`N$9{)Z(V}R$;fB44mIQ%XCg>SJV@lt1&%HWsyB0hbHjWv99);vUE=46M;#AMlsqg?Qk% z4YEVs-{33zvVrx_j`)TFz3nRpcqE$x-^0J~J$9u#%=a7MJ@_A__(D4Fma)_8~tP-c^14 zbZkoX(zp1>w8w^I2kSQ*H$NBt3UplhD0o(ePf-7VYTzBf`{UJn;_u{@MpNu=8Y!!Ds`E5RZ@aO?wvHPmd0r-;L<#*2t z@-aW@%58!tUGYJS}h+~nS*pbZQlhu1B?gLfBx15@M33&_^o{M zv%flbV3Xl3{C?~-em{n%^7K!@htwmDZ;)U8R|j|zKE#La9q}N%$2kdqIbS3F*EsNJ zedfSDUACROx5g3Qx%a?WyANR=?04sc?tz&PcO=+te=@U&clbXT=u7+pbSHY*Zw=^Q z^rfF0bdHGc(7(ii=kFTOoA8Q%Kfpu8@j0N@lXNz~QwHvj{loz8s(#<`i02Rc`oWV1 z%Hc_%9KQzt>=zBR^UVW(G<~BpvN6@?9DpwQlLKjZob{`{?R(>Yd&WS#_qU$;Md+u- ziJZ1>@lJBy83elof0ah=yBo|_#&6$pfal+D@ZfiIfHF(X+7vb+PY^C;x%9XAbZf zcV*~LWF7m!d|GEuADA!m%0948Uo92=FyxOPx`kxgqsHTzIil<_?EKPfVIR<#IMLP>xT`A&#?FC^mqtfL&qWe zQ2giH2hM-oWq9(%*`c(@52VwDzIwn{Pkz#Y_&eAO_#l1<@+kL&0p5n^s1t7!Cruu_ z5r4zil*i}3c<_aTz8U-M`G$bc?DGchR7m6Vl@7wt9a?$axsawTTlGN$_9@vVOuxq8{^_PcYqZLbSIbMXHRq|2vn&*oZwVfLEpT$xiUV++4v`StOM!p zA3tWmr)~_~Z?S&(vAo+C49Ft$sLy;`=GJ^F$4_9~<@H|rW$ffN*UGvRX3V4;AN}qg z&G*jV*CXE=2QrN=CQrQ3vvrT}u{H4-Kl~IQwMWo+B0j~}W}o1#()nt|lYRI%-u7Ap zG7!H<#YtDMviKbTiajRI{$zXL=gQ)Bs5>BJB|C(^$Mzvx*d%zXeJ>9gLHEO}@Gkkt zR&s>w!3W7LyzRjQb&<`|-QQF{*-O5XHTVj-EsyfrL#4@!pR=p*TC!JJY30YGo%M;| z_wd!BU)m%GjL#hh-og8bBiHCy#?qLP5qOv}G%hGv+1fXE=7G$xMyv(v)S9!e_#MnM zoyi{g@qv9uzK~buF}=s0}{}C?PYf&=w|eM`$%5(SR3{us&4hhTg|m~ zl&rSy>^c3hHpyf%(2w3}9LP@R<>r&FL#Hr)bSHC8rn>JTkLS)(=?K!u2|9_jVcpPe zL*$0oK3p+wV@Bv9yK5%t#fhmi6e{2XLbyGCi%=? z@x%c;#+@7Cue*RU?&Q!FvmMk$egk=J>^^tE-jE0N{OEx(0by~{flQ}=D6c+t3VTdj z&J4+a^TKB16Q*ZqM?8NQdyC!X9X#``&|lQ6E#tuEqoaT^FdpW`*q9^zbXT%_MEant z_mWS3>%sWbg~7a-2lew==ig}!q4X5}pi6mI_Lq2N^u?T6lXSAa)u9vSU$O7;0(u^P zf*)D)^29UlI(|ayT$%2z)1~bj`-_fa@7YiG-nM1fTwAZickvd$`{}v(2^|T42KJq2 zz_-<Qg^I5B!e--zt+&T&wV;$2?$RqNS zY_*s0V7wWxN88)@L_9S9gcsl?((HZVzW>FC<;PR-e!PS&D!RJP<$TA6Cxas&FppJ~`@cDt{0E5?>LY5x+nGPriJ)vJe``yiy=03*r2mDde_^MuOaJRwT2lp7f`M~+!y#{YKaOU~y15P}?EdO`5 zdfbQSkNT?t|J3aV{8e`vI4gYB0spzPK^NQH}87Wz&pz4I|ca~ z3-9av!Wpjd=7V$JgbXtm_<4NX{FyiNMULX_4?@2J=XaQ zyOQsVorqd%*4|aC^XlK!mWMul)R*y7dJ@pF=wa4zdKsQ64#g|+$m~se9Xkkr%`L~iczcR{z&_!NXSaYb+m9W{50AInbJ8lqHeuVaVb~$|Eql+oCLXMeyvo?O z@~cZ*_HO<+`xk$=m-PusV+VN;eZ~ir%YKx$eaQAZ`C+Vna$tPOGO`Hp0prC6V%yMz z*>t7LPew@t`2r+YG0kc zrI7AfUi!Ma)q~QF)vs;HKC)iOaplQevX*^kT^J8zp#RF~qq<%^U`Npr*dugd`UHP2 zdxL&qEdzan{Q>-L*&^9Z`5Dn!N$M^U% z^p#!+bi{yH$VXqqQ|OU546b_7zV3=oN*rI3bYLGVV?W^y`l0W}<;DTu34VfC#CNX#PyFw_$~ z+k^I>a|=EZ-zU<;to8gV(($V78SS?J>=k>`JJ=iSXZ{;$N2`G)8Q^n{NYh?AE;Ngih^+SGUc2k&Jp>{b1bH{vgNPGe-8@Ka;u zJ&dREhQ`)==hMJ*`6}|;d$)8|=lb3Y*#6x!M>_-HmttGGE1(|WI|1?DA>S3hgYhv= zd}Px3NX(UY@*d6ieh2eoUgCS+1D_6fS12wyU>)OCbS8WazbF6j1bf83u@CJn_GEI@ z9>Xu}9eDtcv2X23HbVLl9SI-7C&VZB(fqJ@2YwVZ*&l{!JLxico%;H z;8*<0=AJAi1MwZt((p0-kF2RIoewHR|51`8IcIj{-Y0gl=lVU^e4R6J9!U%2k$E%-fs-> z1M-AE2g3R18z=MNy^RTe=N<8La$(Daz9RvcY|&K6HqmVIDAceL-}Fv-0Yl(ogodd9g0pAm0{3@M97+*5q zX!d~gV7=J$)~9s})~B^<|JWPs3HA50j?%~Ev8U88-dfXd@6fs|UYdA(Rh!nZ_Usq^ z&`<9T+SJB=hwK`0-bI`0GiJu6@zGCp^uDOF@>=i43qMz`aO)(0iLp(eRhP2fLsA!gGc<=h^orKeg8z1$QW^BzfnL@VI2k%2~PR}mfv$^jbdJpgAr%r8v`S5=B zjP+*yYe$@)@luaIfM4rLANAc&{`#bCao)!`X}2{gEKVI8>M>{5lr@8T7GIip<7#~f zTPyMyTWi)hi!-L)t9JEG-}R}!SFUv>-n;3$G3;He9nZ!;|Cq4%=^d4mk3I;(()3$j z8fSfI9*wa!#2F`fxkK|Nbnt$XjTf6}yJ9Mso4%j z)%z;j*hy^{g-C!z)yVh;vKZhd1y zx93pL;-y>D@++slps(^7%f`%@>%08qf%46za{8_x`YW%#dPnuBulF)`p5?K&T3^bl z-@B`uehR|M_^Gq?qF(P_yQP(`AC0;9Q+MxYTs?!ZI@D*3tpVfYz4c8$$ZS|Z2DWYjzBlYOg=bP{pem#&O@z2{yB zytUSK^}T(??~s8_(Bj2&Btn%Xf?R>)K+`5Ucw8!ib@z9>J zF2o0A@XO?1GSJ=w`@|k0`?4YEWkDQ}kMgNUIg}p8reGuRSFt1bp!1KgPk@ai54(yj zk$tHw+m4-~9N!9mvv9s=_77W)-3I(y{6O-sUA8^Q=dBK9^To4e*mu>zHatST`Gwq# zVMnrq)R&FPK2i_ck^d>%lAp-4yEAN0{e!FCrI&tIRz48U_SBxRKB$YW%73Sy@~KZc z)Ia^%eN%@tdG0gFZrpjilX&ANzA-fZY(bQ-9XxN_FQ4XJ$M)U&x^)BANBVqzJyc%$ zs&$vHYOUfu_!mBB-M1&u_KLL+rP)gXpGdag5$q-VHa_#bv6t9RY$~=Ru#Ln?lNV)E zO4~LgTa5k5U&U9&AC_OeGbZ*MNasf@5C0K=JpZ}7uATGoaq+Wnh}$@(b#K!7<@Sy7 zSvi;NY?A+qFAJPI@^zIiEDn?_%~_}O$$V_iJ%O)Gobvg`)O+=o+tq(Bt52Pkm9Ktn z$`8l?#t(PH!23D-^uEC}?6f_GzFXisbT^h?l0VY9h;i>c*gTjIbLR}jJee=P&Kn1O zlje}W&s_7B@>#+@SLI7!Lw9D=T4GaMKgl2Ki43v^$tG)(?6N+|tmI90bvm&1PnWSD z(h)X4r+Wmt50C-bCiI;4AN$t6!*}soyqB-YUXBO1*YRm+&*P`@&CU_=3;YBRihsq6 zk2KyEmCm2e?}ne@mHcsh^3EOj?wvLCorJr6M@Z{Dg1_Irlnrs61MIq-OFVes+yf2b zt5Z4saDJgr&J9B62imHi>JQ>-d)K3UK1c8C{9vPV1$p*aMeSBr8M46mB`?U7_kl7|Ha?QhmB-e@|D~)?|44BY@Ou)U}Ur4Vc zFYNKoDed_UGKJp~PjL3J!4Kyg06G)sc~kk6Q%CiP+vVji?i}51mp`h^ZlktP{r4Ul z-uL*1Fi`JeGLu0WT@3yn8^m_G<#G&iC7GKX~N=Gc6jcltG*n%!YdWGkRS z{5h1znzFXG?oH3|qib7_*5;}8X^r-iKOIF_nzH!#IbPmgh|<&dzDO_HbtDtyRnDGD z4$!HZ}&U9V)w5)jDdG=uhP+uNRKl%evRAJoqly3 z#qWB|iMcUX=Fwaq=exN>Hd_!4l2Jf(bj{L*gYJIiF z`;ZJ|BZF}E_NL-Pwzy|y$WeZ>XkpJmTu?^1`l{z@53Z(_S9!KPkl7%7thci2+Wp?) z-QKBu#YL4p@~J(p#&Pqrt6c5szvsq0ey5)Fjpi|)*>m&WbF{U9*KQshzq6L&TYG&S z`RLAnEVIkU3!|lZwvLZ`LHhp5bNp2%+Ma2zp~Y=lezf>qCcb*DeXD%G`nt!;*zED? zohoy#yfW?GQikNu76j1cK+WjA9r8k8?CE&Nwof|SzER5o$F&`-t(rV zMfbckpF6*Bd5`p1w(`e(o~yIDX}pf>z2|5ncpR^uclr)X_gubcd7`z`IO><5=hDi1 zb+oVkTiTA>;j6X0^tqLH-|>Zybl3B;gzFos@0%9B%9UR7Z{xFR@4f5OAHU|&ydHne z^>OXAUiQ9k&6Kw3sdZG`vEAL~4wt`rwM}oJLj=$Dsr=E(9jWx~-22My+XUe6Rguc<+V6N4m?rWM^GHqw^e3Khk*nrt;7w=pS2GOt;yzJg4%N zb}oPV=uY3^W1ZEv_1PUR9o_L;FY4Wji;8ayjMpB|V_155PmO2euUzwFj*t9nuZhCe z#QuqI4YjUXXYGw>YxpwS+HGAQ_h)ua>3i=~HoEgRSYeAjGi(|O{h;OXpC#UAMIJ9|gK6)-s&(Y#rKSx^n zrsZkvwHDFiuQc)2aQZ_syuEPTPuKI@o;vQm_HVR(eB76w%X2Nhv+dt)6K~#B9&wxR z2=9FR`IIfV>)&;rs&}`u+o;?*ddk*3malI+Z*fPvRXe-!I}{{w_#)RBj6eze!Hb$#SB*SkO4PkRrE+vU&YDL$&q-pBc@@>?DEyXV^3X`2U}3-A2@ zUx(`x+PgJA@smyC^+*2FTsDuV=Df5`H}4_e;y;_li*}rF@u=th+j&pL@3O7u)_1gZ zkDl5mr}m$?_F4PA=koX59xdN~E*$TO7GIg_LU-D!@Xjk-e=egt-yVbgp5@u;g&V`h zG-}M719R0JY`W*L`Avt9$Mk%P|CHyr7JEkbS`>~SB`c%ik34CoRW4e+Jzp=a&X^+-=LTA9AzrMX?549%igE?*tGmRuJB$X z*3o`04|=Sld^^v%@Ud+9#24PQ{HLfn+2!=3^^f-3ys)@a`K0Z<$GzBddF0>q9NRBV zeD72qW&6n|extHSy7Ls4$FF{!qw?(fDs%3=q@Vg#$8Ntq?EK{`u71VWJeO8pw0bu! zU0mU4b)2H*sV?(Sc-K|g`gkqcn4O~W_~@xTdkn7CR(`rKN&jP{Po`1P*EUyCYN zzjj~iOKGUM>*?Fx!}WdlL)^}@_jdV@wEXeso{#x@-t9Fu`oG_2&)0sx-f`C5-DhF_m9Jm#c}r+zkMvl_ zo`apXC!;)>3mw z&lOJJ-LyRC(&bHWK1I{5(bB8GXSBFYkLkNFyFbOBde7rncrIV%Hr?$PF1>P_mWGzU zeqZ+P@~fx*najevkGrj%Ubr>T8bQVF=UWSzt%hMaZ$@Jm~IMZEt$2AJ-dK{nu*adfGeodDmHge?w`Dp1qU+?>t&p7pK%uwn3`53Nz?M917#nr#w^+-#v ze6;w>XnE1yZhh^&PEl*-xQ==*?bNzD)>U5XcJGP3o_BoVoqi0LC;g@8BRyA6zMXa~ z*K_rsqUlK0)jr$r-}p3cyUoJKw$z(nX17tkYtj0u>@7iUDu1et`in}JuljeHov-kz z`l_RP&|Q!8$~S(|#*s!tYlYkI*W1>yZT<32`yF?9uYu<6WudK$=J^z@-7AAU|1@UJ)qcm`qp{m*$}8(v zzD;+TW4YQ!oA2VT)_C>b>N{7L^qr-itAm^(Z(d&8oCy>KJlx3?fJk@(elWD7%GKW^mA>Qlm=s?fN4ncNhRa(WyWY~2DU5pFW71eiQ~t6! zlwMotuJ>;e?!D?4y65ZE+Agj=aXobJ*}}^0x{62pm42l4amN>Te3$Z6XLPrB49DN% zHLA& z*IQ+-rFZT0%3O=Kp6l~|kHW@ekJqVi<9j`{@{M)vp!$K9R^B7k&M~g%>*<3$drj?i zwaZuMWmI3!{k%_ef33MLU72&gQ|+9}TV2ueRK91lJgvK)OWW(V^-`Y76kq*l<;&~2 z^qnT`SK5)b7LM`P^Q>O&?|$xZ`=hg`&^mIM$3m57cK2vowdJd z>$G)?7Psjxv%@=2;a%6RulS>Wa(pgd^i=vWuIJ0jlor+B{Vu}2!=}ePjdygXSFScT zEv@{OKhoy$Nad@KyWYapziDk9X?ZFeZ5)rZ^va>?^wZDQT(rLTj2`!oecGOF52NBv z?crU%dTZy{2ha7bw9CpL>$~z|>!0KJ?dRI5j!m_FtzY@&tI6u`89l}w^Y6JS+`O63 zZ1SUDCp&$|7q{)@WB9m6E_<#Y^~<~5lHW1^e)m&jQ{D3U8RH|r7GJ)`7Ojt)9^2n- z9@Et8*Suavujc$e|K0s6e2(tEp9-HVC;tCz`6XjgdgYIG7mt>9Dz3EScox6Q6fP~g z%kHrB>-|;NIeM*nb~$URh3jk2^7Y&03zt^@%C~>6msSp~t|Q%f&gI*Aj^*`L zzuQ}Te-*#izBpx$^?2St`TWk6-F*^1j>)-qsSR}(K2m+D{#|czr|9v0j&*9UU+-kB z$T{*ZJvO>!Vc~N!QGNZ6?Vftx?|155dS|@6@TSR;cztO-CnHXkFV88e+%dko&e2nC z=+C+M`Wp5APZS?+o ztc#bgGW>L(kQf9)a$$RZ+&CTMPmTSl zxLxj;F74EBmk~ae^IX46t4z+P02=d;?j;HmPpwW&V%)u$uf{T9B~@0fp=5pG->_nwb**CVa`@i))$-FP2b zT={PaJ@%op*Hh=U%ATv^SngWS`gE;4*E4p#+os2QcRi=(-MTrhCHa)=C(r&Bzp351u!n;!kDJs-cd?i=ZI>BaB5(?!wU#~ojImlH02({puHw!HG4 z8=vDC?tG`xii?lzIoaVox2M8;o`p}%wYX!x{k+%VURyhlaP1%aCrvr)yY=3FydGK^ z^i=-R+duVJyZwrb79TwqFTJ#LRG!AQF+htit#K)B*C%g3>HAl?)&g1_dhWOL6h7uH z{z!Me`i0i#;&wjabMeP~$LHGb9eXZc&!tD(C!01tm90-`{Xa#^r`)Ob>~DL0@B7Ug z3h%xYcP!_*w4ML5_bUJ9H^ptLzUuKFg?BqUPwBn$_0q=oNcT9mCtp_DJgM_kJA2Ng z@4lCR$K8@}pa_(7v z<(jMGT*sep360NJMxE-fZt>-bZawr=xO_*t+o=BNZs$~;J8suool*4_j>^BwAAJuy zt*|ncuaB3}<^uo3=L4;`3LX`nxaE%2(N*Ptjvt zm;GI$Z0ix6ykyX4gv_zdawvd6iZ^)Nkjx60!c-X~HY%WU#t-qH?X5Ekn-9 zfm1Rh*>I%l+o&CMw|k7gl6cXtWBJmKw0Dbo$DOA(PSq!G<*tXGoAcJjUPt9Ux1RRe zDqKC4MeEn`y*&H%u8pl{@s*3JYxiY`YyTW=A4bb}s=xKUv|~Fv{<84#-My1v?{lQJ zd#>Eo(&zs@*0-N`-RHu`cWI2G$MpSNo@nEY7Ps@C>x206U6$u>nzqNLusRDLM|rPN z>vZqMmnGbu-n4p-yl6ki+ls3Uy5p+1es6kwx1M|VO{MjG9CPuf;>usWyX|w~+NnNu z^i!sAR9x?Pq{lI7%(CmU4UN~%doEo4+Pt1$`KwPm^^Max^+BQ5btc4t~(sB5+|SXr{1-GMR(uj-(~hY z_rB3xPF*{1;T?Z0S6i3S`nc(3`A>~m<8ZFru6u`jhwG)@uV3#PZNIf|&~%)w_ngvw z%7<27ebrr?(ftm>#g|{%>Ops1<+&bOz0qTt;_FY(=kiIH)>s|s?sNS<<~jAe+p7;~ z{n}$x`c5m{`mvt&K0GDs#O0GdMLQ#l?mWdsciO3N`#!pPf^dDQzG&xq{e9)m^|Vo4 zX#6Ce(jIF(PK{sV*x2`sHlF9oRVI2Yv&&W2?n~*V^;~_Y=(X0(sWRerx$+d(dhJ>I z@pr90mcP1B(dw_gBQ5<@*=xmBHoEJ^kB+=+KgYL}sr;t3ed-;HiP7ch?f7%y@<(^RW1F>oEvnqE=Tx}%qNnogcfVFx z`C}e&>TfOae-;*hs?AgNT~{| z$ltF%UJu>xyz}q)!qRJF(^Gw|-=!(59^=_OoSN^Q_OgVJ>!CFlmA=c7QD>&qSsaiwp%>p2x} z9@@{F-;{<{wz_*pi@O$;w>o!w!toW9E_)6~{rQ{@ebkdr>E^|ybMG>z+9@vmTmD_P z^75VQ|GD`6{=$vV9{W9hjiWff)7HhN z?JsMuH7wjZ+j)g|eLH;2*BY(<;~GBJzw3Hg!n^%*m zci-uiN9*sYu`!R$?`5<#VxQ&PNiLuvxpAcB&9~EjjqbH}Tw6UiUOjJm?0adI*|fB1 zX|0JpPQ~pxX-pgM@3d$K6K~V z@jJZp2w&@0{_2m4_g=^HJFE$I<7^_zw6w4qwsD+c$d2+VP&fmpV)dt@zMB6&(Y$ubECzdqVlBM)aHJV z-cK8a&r$E|XWsT)@A+>1Zfb3}=8v>>zv(fL=bc{Ndrj^B6kk4cmp@hij+3{t>N-_- zZERYZ+S+yQaOveot7pfR??`uDh2t%J0sF3W;o{Hv3Y0JHNOv8_G|#n-)}Ls7-|tnp zv5gP3wt7adEV%X0PCJJ8I&3^q?diw4K9yH_Kh3qbPwj>F+v01d_d=U5<6wL) z`!z$+CC z_Dgjnw~|NIS-ztjD}K}6w;k4(jnc&JU+;dTm*tVBp6c9m_f1$@{n&L1YqPj>d5W)% zU4L;$TAAwZ-S)aFt-R>*cTB6EX!W&Mk9)c2_H_Bt;-cr`#Gm`^`U+Ry_0VIP-Bx2! z+tPNOJy%Cmer5OA2{%W3{@O=7Zm*fjpy%o!`_bZ}JO2(#-{lG) z%k4hx@UBaEr=<%v7S|f1-Zk2|?efQQKlOZSZZBIG@}H`wxMROBdse3U(AHJ-O5)+4 z()97N{j%5nEeY>^bSkVaZTH*V6W6nF< zS0CCMD6Tx!+cR38{cgu_d3L$#LU(*&@w@+x^`@oEd)ZIEU2cc>Iyx8L`3s-R`~P=* z{fVAy-yVq{ANlm=)$&zGb)tK$%~5lDql~snc(2D_WcB>hf2|k?yujs~zQQL!H$hO)l*B ztIgudb1dJuT}F31g-_94$DWtMo9?{A*ZP%zx4YlzTzKcdo^b8dA8XEDIQJ_bS|0gN z?ZxAsD*rjU>pq6-=VkP=KGnbabgB=>_lVD+$?(mm@T=m>TRWR4skdM6b?!a(oXN9) z>gf4geY<>Xq&>0sRB6{stK*n|wE@SM!A$dvW{c8FqV>N1=DDP;tFz}#>+>!rzdl(z@tz~k z>RI{@Z~mnYzuK=}ajntnk5*T-_|~zq=YG{+d){-uUt_dsPIdWR$3?{r!Do$p-zyFKB4#<%d19@8o(zBxV4Rq^Jsc}0(NTU*C|-IC|t zb<<;cX^kD)GkVOs-}jj2`Lds~{p73s?pxvd(ev(0ah2b7i7WkD^!U!F@}H}3m%S`} ztn*ZTy~ioqcs6d=T046Ul@~2ODz0DodX66B>kBGR&!_rb-0}T;mfo-SHa*qn`hITD z)Zb0jxu4I4t28M9q`jIMM8~c5Py|4E+2CeTs4vj}tnd+)vJx7o2?KT?g?*o zA82KF8{&?2R9|(qucF1DqvhN5)cUfXT4S3Y*Y|!dUu)5N?N{8jX!+04Q{|6&cb@u$ z7BByP=Uu)$mGj(BnZi3?;Y}-VdORuih?d@3 z>G?9MZ0VJ|9$J~&-c;ItZVZY$mhE}d@aPifBOX;zEbE+P{-ClX4r_yS_Jf~>o zj%mf88~4&td0I2&-|51;{?c~2UH+JVm+hTeyFH`DZMxG7@A?Xt7A?NMUJvcvkMCt4 zZ9aWV!pb%_o7PtSs=ugr*kf=EZ+^RZ%g%Qkhh6^;m+ur+Uu_+A9nYmjuP2W>YpZ_r zylHK#^LVbnkDp(E`}Fh8Z@T+=>|gyxciN5@Zd`Z&go{7YUB0wq+5Ox&9O-dw>@vdp z_rGZS6S)7^Zm#>^`?_+MkU~m{E0VDZw@8`F6h&@RiHs#vDPuAwG80llB$*}iJVivP zsGD$;Br=t;bYFYl>-pRJOxv{doM=F|1>)b*y6@>o`82^E~&)Tiot= z-FKJnLTGo4?i|rGwx7ruQcT~EHZOF~;VkRlbK3LDF4VqC=a$*{t3JB)dz4S#?s{u@ z(!I`Sxp>;KSh>C{TD)!j|2xX&cbs)&_d0cd6<4j(uh#Um+9~g8 zJ-@otJo1Y=4&&3I?1?ve)zRV3sbT)ThSI679lirfqnEJPWte0oU>Ai;a&+a?R z#fCT$Go5G3#nh&|zZk1+(_4MDqnlrO_bI108>8m1ruVCg#WNpOzD4t6>u`=a`{Rr5 zb=MaEEqxi<8mQw<-u-Cu@1m!SY&_em*I(RLuRpqVYQtwtaUCt*yPx7o<`qr;rdyZt z?o;0V+t>LQ&Bte~?DhvK7jwnpNgp(3lY=Isw(Q9M>e1S-=*g2m@=ba|b_a*OY zJ?>i9Jx)29r!Cs!J=eSX)>B(!u4uZp?9I3PcX`{Ef1|svG1@Ahbn{=8<7d~NQ|(u> zujFK3F@^TTpGu)aY`|`!k{mR`}+&j8+DAykUdH+b>E!wzEm8-_vK+(R`Rs(sAXdY}+`bMtL`D(`vM)|l#>=PbuVXMVn-)}r=p zUu~%Np5$(tf3+p!9M|}hZaHUP^<A8BvwrrLY8(Oz3x^1l7xKrPio_HF2 zi{d-&EmuU7A0L|h-FB;N{9SS43^@DKc|~vO$)=NybG6-`Eb)lXVKyRvor?ddHy z*W{z=IAykeCoR_Yousn(RPK(lX*{*D+yd zLzA5k(AuKA@2b50dG_nO+84#=rh5(VDmSJ$+O)oN{9WzxtMYYfo(X|2uS~ z^R!Jn8~S$kp!AmA>ABTUe>|xAp6mVk_qtVX{GMAnc3XVW#-mql_3xZe+otvJn6B-l zo98O-i~T(=UD=4%7TtZy=e*;K#=rgDnmuR?#MgILbuwC4veSd68{O?sPkGPps(ef4 z74K>19?RCAyVh>6<>qVb72W>WKB>JiTV7@Q-s=vDU7U#Y1_uurmMdRCh zW$%|{8Q=edJz0&dj@H&*YtQa|Cw|X0+kU?)$B&-R?z<|-f38*Q)w);*ZJ~8J+w0%* zb~$|~-L|!dYumhA)-K1h+mvfZ>tkb2{nhoE^R4^SRcvlwRL&3oD>UCGbJJULn+Lji z8oTG#UGeIET3Du)chjj`WbdTi5=+jP&pvi@7o_Mh@y@wEGj+vx5)%kiAFyl{1&sc-L-bRUO6Gqo4llIQ_os*%^O*wUc$qPtTQYXWzCf{#&%Mbnd72TjRElJ?>Vy zxt%or+Vcfk+e!Bre8qBcQcM+R43C;}f&b?Rn zv!Q71SMS%g?Q^GceNl4A^tW=q_&EBzIaJ?+ywfKyn!e^y`(3m#`)te?_1&U-OzV?R z`Tbnuv)%J`sGUZ_Dl2Kw7H`_`NOZ3MsoDUx5t(rJ1Jhn z9e?A|+RpLG+U>q8@9)gJ+;{0;|In}hm2cFwoisUpr;eWT&ia-^Pr9PJe~+zv>cM|C zwsw8bbO#zsezMVgceM_CUGn{=)}+|7XY8N7pW1ty?tbNR&Ruz@eT}#A?bT@Q(ObN` zf7|u+l(P#x$Jck}s9kT|l}(LD_n6A%!TaUQ&dw`p>{WYh`gfk)Hf8&s_i5|N&-Uzo z5PHrnyEctyw_TO@`Q90iukm+Le2s5RerX=YpL&hyei2Phw05+%O?Usvv+dervV%?8 zSKORyaPqXS#pFrbZx2HEIeMN;JD=~Ia+Uvgj_Ew4KbiDY?u@!6pRS$P&aynX^W#}A zUp8*b+w<7%joGs0;m&|^_cK5A%nv=w0c}?(*Pmn5^^v*do-!IA-Tc~4{yV9S{TYhqW!$ZH}p=VuC-`IW1`WlOF-d&C-yPLOe&AaEY%gwhn z-)FFx5KB8oD%V~N#;d(pD-IiT()#1KFGBmOy@R%YHjO_SXlIrgLdxpPP7#BST= zTm8FT*%(xwy2@>5e#3KDZogk&xiq@->{-6zIrH}~Jo*cd{`}+kxkn#Eb$vU>lUIJe z%HP}eTl3sL+C0flhWeh*IhW=U&lP{Rrss-BTRNlJSReXCkWJZmi^g}-&A9hVERq%lgWB<-5jfziVu{c<1PHyE7qZi>EOsWz%kpKf3!W?|Gc% zZFl_1I%)l<4ZD5Iul=OtYKxxz^_}{w=Q}jtJqYbQzI8^Eam9mQ`>7+|vpxH?SB`Gp z#-rI(+aBAP`cB%|X#Lf@t5o0pmD87xH@#Y)tG25#;-S6MzUiJ|Zz`XU_RU*=`&+-K z{kLPd_`9;NHg>iin`S@0>?Zpb&U}!a*bQSM}PLw|MobP_2Etb)|I|fws!p8kFIDs>DW5y zt}PnxmZPoysIl3b&5wERY5yi8nVWx)Ip^7&o+jyOrFgd=l=J;QJBx?TPVpdCsz-~T z9aHTe^n7+-<&&p+{dduLqs4u+_GDC#){fpam)&=k^Huxj$~U#$MYr5-SLJMOZfI?2 ze9`)@_-iY-?HqEeTyBXbW7GP`*!l8SZqvT=R=PKTZSn2;E^pnty>j-HpUc0yUfa&g zJJ*-HcW&?8yrP|Z(JP*PHdNkv&hoZ5+qS)@PxH%3t(&&(pZu<^@`KWC%Pwzw&+?wz z>H96CIc4jWees_ghJ_>M5@DC41Y| zJT|?ucaPb+k{{hR)FxJoYcX7Gw}*CK+j+bEx%+py@waGt?nlq@o4-A`_uH*`#k=X2 zwe{?IRIa}jZa-ZtH@oV;jyPI~sUd#~3S!_}|$?`~6$ z7fmMl+lJFm^=&)09CGEATPXjOr>dX)j&fB}_Ej#yI&T{8`bocMJbFTSw^D7(EodH$XzHMs` zrw^*%qWNa~R=M~QXT|UtbJef<8;_n({G0AQR@u1Zuqj(ln(f**U+bKo@=>(*yJ)=m zullC-p{H%tvt^H~+_?OH`m6f(*IDMX@=^Xn&nH>8X#B}5r^&U>59e*~e`x3Lrsp|R zo{lCHt<9N`{8Pu)x651SS?J&U$CqwR-H@9K*$9(2pEobGM+F4rIbN$Zc+udbZ_ z)1LUyp7ged@>BF|Z*BHEZQkwQtuLFdXnfg(Zo9|~+T&|nwDxE+(8fPeYHaeNTV8X& zqF1`&%a13m|B9ZxPo&=d+GA?lYpH+N*;yrdww1wb6gtrcM88Lu2ov#u@J+w>`D@d*zeHdy8i0W6~e& zU-tBDySGnw`&n)*s;_ayYV|0do-C~XMCg^SbVj$%?AEs9WA7v9S?Awf6koA*rDN-= zKbz3nvgOL2)0V5VcH5r(E&KFQ{Mldqq}#^i#JX)rl+i3Z;kjbzqY=&Xnd_lG+A5zS&j#7T=H(w9V3m|fe3JJ?Xfj__mMtys!G=ZGGM_=nJk%3!_D5^KqW1>QKilr))^|S||2g04dyntEtH1h5 z+v}U2JjuT$M|-yIIpl|1{;mIt=F@0A>bK(MDrSsJ2I|>1R?hyD)_*_xM8<92?b|E= z)_*m|I@SL~>0R>l?MHpvc1u^|?m7ARXRkr)x7#Y~JM&5WSM7~Muf{f}e16jMJ-Yj9 zcYf5DEOhhk_OqPrSGLu+<)!zeTi;bVe)gW zj-T>O_olvA)m}`eGd-KucNg6;dgilqY@Yj-)5VTG$!*tzmXoz<`fGboS@AUvP1Y5S zuW{zmTDOO)>%*7K?Z;&6^CV}-d2w2-ow3>8(jV_#`zO1i?Vqjl9KYoj*Ujsm;gA25 zU!wI@k7{qO>^PrnJL$GHJ5HXne#)+%ZBgyUo_=WVXlrnnuiN|CM{YWrf9sBF^WfW0 zW9qv#Fa4+7Vyt|C)^?@yl#x8`=X3U*HrL-A)t|_-`IBw#x1RiRtN%gUlXL3YwzI8y zMR!|1Mz8v{;SXDOc52%)8?$*&9{ToLpX_y{^HYT4K7k@k_Z4BPVb-qSx-|~{XZ9B{HY`XO%_oV6F-{txl6LLG?R(5_-!>_y_oVSPHhDV+_Pq7i zx8v`On|RK#$=`Bz+it&#YcacH`79Uzx9G`Vee>7W-f4ecwZ~H*IomH=ZteM|`877W z$0*m1o_*@4+~h~=zeVGVZoAUA$DVC_+~zyW@wMLSo2UL;bdS5@sejYrx;VdUFW4LU zJ;``_Y&1SS?SFL3Psb^rKd$Dc&2uI9)X^Tt3s+C`yG5_YWb+=A{bFLrseRG;)|pow z)wX|MrLy+B{-z~9^h)mDk7t}#KWQ&TK{fSE(ewKw0X9^t&?8$tGVrITS#1Pr7w%`Md19-Ti<+JJIBB8}~X?zOwZ}+sNS8 z?PujPhIYMpM6dFm_RYV`>7jE!cg>}~=pJ{LlfQK;*M8D-{Qc_2pR#vdzb9K!^~PQK zZoeh1oW2^m8yhdw3`5&}ye6s9^y`OgZ)}Gn>CY{-o z+)eA-v~hdh`uDiKSGDi)%3B6G`IWz~9y}i7lC||+>4@)sG+WU9Y+YN%)~kKjD{J?d z%Z{PS+VyYUr+)n&y!%N8dKJTGn>ya}Nsji%G(GL#c(vTJvG(no%KWh7vv{Z-6eO4%IJJ+jmt!wL6`<55aU3Bwb zl^a_hyY@R_cGiE=#@~+`w`F^$J@x1h_{rS!-*M6U6bCy87JuU6T;KSwVkQ|ox1M*B zE&G&TjGQr1JG$*r?`dx5I_Ho3(foD#EdQPBbFSlE^~Rn&w{)fR7B$!Aoo}jN(R_oF z>4|UqG+8H&cT+ORG6(az@^A6k{RCfgHr{j2QQOv|tZhHLpRGNcM_Kzhm$Tm3TQr&3 zSv|VP)fVrqbLZXL~S=fUDE-B)wG(p%e}OES_^ebeTa{n^c5J5QbEa%cH+=ftzkSn|?= z(zA6WZ_{+PcALMx=;o{qcz^t zI$GD(dVe?GW&N%BmV3tLj*-g6?k&3Q*|K+e^X&2!ng-zqmIUzmH}yziphM-N)g z*L2Z)%ckaLJu6>P3pJNtKAGPm3-{>E${G$wtw=*~mR@=k5W+Nf~Cw)*^$+@D*OjdTD^saea$!)E#`c9p6_nda5mw);8 z{HA@rFXF#tyS}|nr>y-u+4b%9+`7_}y;t;{)8?t~7TvsemG}57U)Q$zDkt~!m-hXf z^<-|^dhfWcKUw5#UFom?q{fpcf5FYMU#a#$N1{o?OTs>bKUV*xjwXKkJ$b@MLX&|-1N zbg|po+>eq?58b!y(%znn=j6}Fn|8kJ^XGiOb{56g*qv8ui_)J@^ABns`6s{RtL+1P zS5LB%jV3q#cpii{X45@x>)GXWX0tg|M$c!gd$vTBPrs=e~2+HcvQ&r^SM&Oglu&nZ)z=Ny|1 z?T>Lh_OE(<(R^3^YK*r1#2eL@o?AB7kKV5A(^rWlx z+kR`V@!T7*{>iSI@SSp6vuN`1?PtrVy|!C=$T+d(@4niz$wma@AK`>vKhmwS(r@mWfRp-#A<%Uy6Hs3`XlMJ-B2cZwbpZ?Q#)$6}S@#Sab_)hz3 zNAY=#vlcrhw%_ATe)O@?Pt&z!S5ANRE1H~>Uh$r7#$^9F2lHvK>^RT<;&0Epw)@d! zoU}P)XZ4fTkLIJ3#*lQao6M7=NBE#;iTuBsy|VGZ~x}IWL(Lt4{hvTv)VRa z<@(~Uj`q-bu1~(+^z_H>tBj{7`6s3Ovv+z(@t$NUCrjCLPtp1(yRYnQ&qbS8cBA__WtFGyNAt~HG#l>XIs1&+em?Ey zkJFEQn$Na>&vHCh^yIB>4a*JDVr*0VavF0#dOsakw$tI+va>tiZlC94Je7?-b#Hl(X?SeDm80<`uQlKD;zje> zZma!4=11;E`cdtBPN$CQ*>J`{{aaQ(yo>HRv!N$>Xl>i}?7b!1x^+$}pM0(Brg*jw z()&c{wmV-v2;K9f@5&e1ar$A`-p|=r9Z%&envA>j;o1FXxj9GAe)Z;BIo{f{17&9qer0q&joWRN8@uOwOUAat z+EgbiInm^u`?U6G{MoZYGyt9m#>|~i|XVZSC+hu<@u>Tu6XL^nE z(#}D$vSsU0PM5LCGq3%aucx-hYQNfF`E65v+vAMiWAkfmXk&I;W#h9E&CcDA=hhP+ znr?L4tbDfF7gz5Td%v7A&U*T?p*rf>Hs9i@{fh4K{Bq*-Tm5@na(A0@?Pz_C-*V4# zJkiE(d1pDEr-!D$eNg?Rr#*DlwrS&%w`u#m`{<_awf5GzhskMgp7d(3_TSvPl{3<_ zZK%9$X}@p&^zXTC`;v1-ldUf~CnYOCqUqUfyS@HB?y7B%NoO+m*xL5G-dgX?QyH%q zJ@;#ItN(n?KC-d{Ro_qi=+<+VRBl)E3oO8TF9i**bQazRsA=AK%a_?>K0Vd;DY4 zpZ4!Pr4R7W(?fcgfoFl6kIMI(C0;>DX&VwkKViKGyTJf3qz+lC$N~wd<9$AzJ%M)A2N| zPqVG{u?Cg%fA!thT*%rsXG8R~LH){(+BV&GDyMVPY}>T{J-)VR^7gpm?^au5>&Jh_ zetpr#T+wacl|9M0qF3^?CA0pkvAZq4^jF_JmGjTJ_aCIbW9O7teN%R@sdy-EPCM0m z;(bth6`NOW$xF`_&2J~g+gy~Bb){#wo%_eQ2YHfz7yp%wcgeo8 z>AlLi)VG<=s?Mj))A`l-XgrP4p04cO@pjHHIVVj9nw;HMd-CFmZk^iBvG=QQIjy6$ zYu$GYDc4>+G-lgT+f@vl60yY_ig*O4VB~F^6Bk4bsCqa~&8b3eIf%%?Y5xB9a?J)72-tf!YI=dL+yo+}#~yZKIg>NCD_ zG?}eqwDvuwHQ4m@Uvg1T^3dvN{mMOi>|N#f(d5>)^_?*q&!(rI-M7nI?paPxdv4SE zqK$nJ*_)?w+sC(j zm%lg7pN%Igxv08Ff4p~%Pfql)(T~a8w(s)RSNVSZC*Std*|%lo*Vgo;_3tstJzenqf zH=2wq8qcPiZ?`8unqKr)Ox(q@$85gJyT9CU-pQV5y}!pe^U=;tyS(|+#g=T#)|1x1 zX>&N~=DACbKEC&~MrWL8-)nZO-1sYgI!-&%o4h@K%cwoOP;tHMmE*ld(}fmuryaF# zUgcXdYd`sS{g#e&oxVBi@g*~zXmda5=D8}zUthAfuKLjU(eg_9=B_*z@2xdU?)`Kn zC!N(#xyflho7NZIeNS&WS!i?Fww>j8(Er`fod5kbBWe|7sbn0`S?NopMQ-% zpRND2ee=c_)u)_2+m`h07|}*P-|;2tiI*&O{`9Cf&-(H=-KSpP@%Hz1W0Rd7eP6xm zWB>jZxXZ~sY4Ye{Bb;`Tn?C(L@!z7^ev6U``8_@9YTPNm_Gtatbw%T8OtkiB?b#Ns zebdH7Z}si=2Q8-;%|ClRTete6dt75{dr;k*ueqV;v-{5SmZ7}I-`gCI|5NTc^H}?& z{!JUd+s<wHttG)a?f$e-e+O&H7AXyF{iBbZy#jK?o&24Ih`HVx7@}h zXZtk&6&I*D+VyzQo@Al5ZN3L7Zym~wN#0!)Z?;~={g$2F=w0KkWL>qj_fDGc=(>8g z-zuMc*?B*@=XjRyH+JjH4zyU=@l&~&+x7TQ8c*?v7W?gqtM35yq494S$-Sbd4X3>i zDks^SUs?N=j{LWMdivk`?6%FToQx~^yDh#=>)&l>Ii5}L;yL@Ov(-GFNW2;2xl6|O z&0Vs0-yXBek7;|FK6>}OD`(IBX#D71kN&IJsV$luc4l|;s*cvS>8)IJ)m|=Ye6p)= zT3>p1J%84BYg~LM-LlT|=G}dCG^aDis@I3Fami|p_PX6wjz1c2^^?|rKiYb4dNqdp zY>#f8jUlHzq+I(+8+*1lzPx%rx^*>&ofB)@dTP6(du-*+x8*4BzFXzwYI z8t-nay!*DU-CsG{nCRKQ{dB)_evIz?Q(H7y>5XbXdE(tP{->8V&rP?_&hn1U^Zq49 z!;Vj~uk16A^LbD^*wvGa?K8g1@0*_IgFd|Zl&s`#nvT2Z=C2(8mak3!?mKO$FWUI% z=CAFEP~%VJr_)DQXXq`v*s!1caEr!!_ScT?dF^@YJIAT-nB8$$%ox{mHNHM{_iwpp znVz$4*RvzqK8aq9k2iYCGxmPxQu428a<}g0n-8|{E9x(v?9y*$5YiJ$LwNY;kTc_>!Y~DI<$v1Axs+|7hvg`B#nNRd-zIRb}-||`Q zd+lzO@sf3yuI#_!*?pDKdyaVgpWXid&vN$NZ$9{LJ=^YcEt7xJD_QZ>ms~MY+!RNf z7RTK|(b}TBZ}*cE&;GrRlm8r({>EPMZ+Xg_XZPKbosLcO1Dfs0&6dr7zj9;Hc(;zb z%J}OqzeHQR))L+8Tlu7$f0s93`KLWw>|XKLmOWST8PC>HTjR=8)o;<}A3r%;?o~Nm z=$1=H{zbQJ_10^z!!BR(@Aa-+{1lV@o${vjUD3ur2u&xt`FDHs+H*hWd&P5W?yW_> z&EM5eKcBizJJh#tZ`YhQh-cII%RifTXXw2d?cEuz9bY=P?8@8D=2sk?YrT1u zlM$^yIh)oO-Fg6`ne{p&;FZ@%B@-Jm8{k- zJJ9SrX?@L!KPtzoo}X^f)<1g2!Iqo+TNLk>ldMxl_BGGyTW)RWes+6( z+u3*4N-s6eJ zn~Xi?Dz4*4TT^!3Z~X30M&nOPUUNgYFIvytR=F|8drrCct3NjCG2N9TYXpk@on25)E@GBww%iPvM;_(iYOk1(~s64|E61CXT?ePd2&@g z`IDKR>L)$NRX_RE>)TX&I+`O|8@eC$WZgB7E4le|#~InlW7pNVt-tbJ_SF~f6+PFK z-di~RwQbBFx5jGU;~%7amtW8ExAf+NE4ux0_V2!1<<{|vo?}j)z4p7WcC?(d+cs}H zvR%FYJ!Y5dk0t|6PV~yB_3d$YmFe02jeiij$L(^qqkC@3J@Ma)yB%9%skySNa#UN- zX?N{teT|FO{vb4Ydv4-jpNYHNUFM|m6hmmfB0D|N-S!}Mn%fnh_V(uXaqk-Ud$(w8 z^vZ{~bZmWBv~sa=Nv}U%$SdbVqAP zYuo(h8&Cb!@A5@`_oLgl`Inw--}I_In{Lr;-Tk#~+q7*S<@4G6%G>s<{I%Q4b60u2 zw!Yt^SNVR+yQ`e8P4D90^DwXGefmJ3C;q!+?!NeM(Z{4d_U{!>vdkx6MYYu*-Sgh% z#^m4X(bNChcUxucTkbBip*ie%x*8=J;yW-@7b5#=X#uR7+>_3ocd3C zCI42NvB}@_)3%>oPuAXBmG^#Z-`5ro>e=-tQcm8vzqgERJZU_;y?N$`P3w!__^n^L zd6wVKdUhrQP3|qaZMaq5W8*p3P{H*i z?ESfUlYJMxvTwH)Gv~TLz4hXwd%O3Lz90Br(0{YNZwLDD;NR~om6Ma)EBZujO-AGL z!S>5pzRSn@PJZ=0X7}&*TVvzTwrK5WZO!AHSM6xLS2UT~RsD*_6TRZO)z+9|SiSO& zcV+EQS3e^5EuoPVlEwQqmkDz^rk;@$PLT#OgWkJ_-hNQpxbl4R%J*nCvHN_^xzt~KW20NfRU9<_7Cq%|JF^u} z`q0g@+mm(Dt$W+3eb<$DY?lw72;F(<>f2D?@J@=iwtiLKWR4acCxD9Pj3D3qdl!ZdbW$9{WRvJTdp#Dqm@sd>Zh#gTkmPd z?!#ZbnBV(hm+OBJdNnRRo2I|{v{swe7j2H4*4Mb=^)4#bZ}s2mYk%y0bc;7zvhAe# z!94WgXAio0YtPm#ck_}%X6v)ppgf4ypI+2sY&0G+li1-Wu^wYu3%a!#OU^G1etz%^ z1AiltY;@;ZW&QYnaiC6ay2%w|jiJv&mp=9S$<~&gc*r83Y&!62O9meE$&beB-}>m!ezXCC~9!7~l^yFWSkfN$|W>)_c2e5CKc4Ssf@Y(0HT z@YoYSJ5cEFOO1 zBXaTZ5834Dw;p_fA0_jr20t?RF9SZ~AF_=h?XH_&EQu@jnmrF?F03(MuvdJG1AZb)JzZ?)Q$G8z1M?(@KD_$AfAD<+Wq#H#PS`Czw6oJZ`O!GK z+Q00h-@NJh(E)yYhTS0E&G)+p|7Ktg{9LXuR=@S*8}p=}Kg0>W#U4AXp}4gU`poBf z2W+>F=2{Gq!5?h&n3H~E$!41|;*I`n(9cG;Sx05E@-tcNFjsx-FlXQ+zUEK5f7^hK z=1Z=%V}o(@gE`av!vlGhF7ex1vD<#bBewKg1A7g`Io<3tAM)4@VpS}#*&_x~J}&O@ zn+rXBq0Cm$hmQ`qO)jB}9pV(MNB%J{{udAYXWa7)*lT^oo4BJ(8~>2UPW^ORFS=fE zzz6j6nfasUBi8MI;AeivN58cpOZ*lq`tb0#wK0G3l-=5`8~w%_N0xP?gP-vj$A@^? zV85_e-F*3l|H=_$Xx9eTgpb96_5OhYADS0?ign}hnG4zU%d>0|yX4S?kB#Ere&kd9 z#engAEw|9~g9B@$UEJa~5B^D}K0ek4>@sJRZZ;Z|&-Bp;>_N>{ESo2}_@6S66RbD? znmc>+o1Zw*M}KQ=t;B%+KtJ2X58shdY-gXf_#FdrN~ZNRZ@TzFJihRN9p++R>EnO8 zongk4$F}bpWFy(d8@oK_ONaTg$C{~U4{H6jiz|BAQ7q`=FL6Sicoe5%x_oW^?6zL! z%U|%A#~<^xSeHl3+4}jOum08G-wo8s=1X%m-d@$txAr36*mq)yK986;Ut_h22@jji zpDuGUxAw3x{7bI&WS2ht;)j0n%%A>pw%8Su`in0z_(h!aOMA|E?R0riJK8ZZ7OJj*V=vSH+0+cE@(VAdf8mMD?@VI5zQt^~o3F z$z9Ig`@aWl5hwPTxiw#FV6Xk2!S5UBXDj>UPyS-V?;V)4^)1g>7weCgo^KxT1>3Bn zyFXuJw+y1eDFk`vkb#nYUc|uLts?vG$3(zt~YWM!x)y1My(a z;!Yf}oh|IM#`@^-SQBHqlZsh3lfh2DW_P*EJnT2V0%blWSMCsBd|)p0(`|3@ofyL_ ze(+lZ(&svC!>}8v}d-Z=BSV!~VrxzH!$iP`gzj^Wj`Ft-Ah!OE+5714AxD~(jvV#wd zVW0J5JK5dOS_^#Uqt98+AMGb=#4bM5W>1S<0DWO6V|PLqRz)Xe8GM(W9``}=aDOBt+yCvi~aCY zgBKq>WuT25YvthsupZ=?pR?Xv>~T4OUDh8JGx+%5T-a~^e_-&cgFiN~&UnOtxr+y} zZr$+nukqH%y0gu<1F^{O*4*AU2Rv+Mqj6;AcYWH`|Ha@7ABWu2yPdK2jx{td{^cw3 zzkTqX1O8%{y+fwQx{5FHWRI~;Kf8++J`~GhPfj#%^WiJ)e95ooE}ziDR%<7Zn>UD2 zHqv9BD2)_@SJ^RyNaT z@072_A72%V>SD9~#&7&=FUSwBr08eENq5V)%~_USVK8$z!j% zfxYMvBW%~k|8gOpI4kU3KHyh=w0F(NSZ!q5OJWM2b+cyt_-zZ{^60nR^DiGOgV_G2 zfwpfQv>x_8y?kxo@HN?Vv!8xzU`>=%>@N(fsb$H z%D(xDVL6C@|L_2P$$>kDeWPFf{~U;Q_YZrjy<~0qnQz@G#G2SOKXX9!H+L~+ZR8tg zkX&GWjJHn?_MMqi1B|i__V?24*uCdY>H!h%o@AjbeEFfyw6w*yr})bo*wqd z8*&W0JgB<0&~IPzFJJtT!OIQ!qjQa)?Mr^n$L`O3^)iD$I(W4K%D*o$;7j>J-r)bv zSCo#|9K6OrpMAof_AB4?f+XLcS z-sp}fzQnTJ$oKMqHf6EQ-|Yqd2kpl2{qG;}J^#wj7yc+o6x;1n`+)~wD3-O`AE>&0gy-xtmTrAwKwguRf7U!k9c>nUj7Y)R^d&`#&zGfi4%k|bt-5Bd>O{{tP=m)yRh3B5} z^N+LDT*~3%OTYQpzhazU_z&fq{3kE-xx9duOYm9;KE{t)Gko@ewX!a)lYV=`{&>ZK zeSwDz>tHY1OY)h!2OG^p%!n~@%WraOKH&>>PzEv0=X@_0SQBf7$_dtr@6d9EvUs&8 zI&-ZLIbz*8BfgE3TiTCui@4xt=Mw+Q3+$IG_`G|$`-nDu@&-6Jo&DmUzsujR^w29k z^eT@-nZKL?&PzGmI#?I6ZX7w*QCzaAyOlhlO|AfCcTm1%i#*@`r#lk6iu>|QF=p;N zrup5P$S36jYh}Olm3++S@(`Z|b?fSkAWuA7H@-)$Lq69pkB}$6t)+)w^Jn>ZQ~THa z%m=kL#+W}}h<$e)?VX`~0P?;Z;m&7Giu29`{ML=%$){H?BP+@__vmai#+hOt*kfcX zlW7fhzA;C!Y3_Vm+{@8&z&8w>iE_l(4!UdbqcQyM+^}xqv-df9S6z;xH+m z{A3@BGk*85f$!+S*L%Np;!F0ht9h}X?ee$0;fyNRc5je#`Kr9jcgpeuzwy7=vo>;U z=cRtJEe6Rg{_F{mPuNn9vYzrrzG_dKk9Y+>6kFzlk8gLr>m8V1#1uc2gT-S07k};{ z&c-huxP!Y3Iw!?C-@D7WgScPtHQ%zEj`p5%b@Sr~{xDx>jrinq>%-UVH!m|v3_J(zUOOq8vf;H=U(sp@4h0J$VFn^dVq2{UO(T9XKTj> zWA|AsUihqd6|;QD=lQ=lHfQDP@&|uHea4W5YLipMYVRj}F1K`7>5ftkLd7-6<;K%z zO|3hB_YNR_I#=`4W7+1J@5R3O=VNilcjjF#Mb)*-DUg5VDE{IGpe9(DWj!-9; zFQI#ca%;yn`-vVoAzk7!-*z_cJl;Es`8&_-``%I1<=}k6pZw|^v==&O>7Y+8V1s=Q z;FZ9dEpKpUfg~#ci867%_8Z85~XDwR^=r|2q46FPAUm59?r!?IZHcgHAr-i}t)a z-R4!UGY9L#&u2_oL-8ff$kva~BR9}f&M7z0CAP({{2=z_V)r;_q4#a)AvpKE`^eey zw{iT$KJvv!=brd7Pv?z0f^*6II|Jnjed5s`uwU8CUh!s~Jmquq+WCfjaa{fu<8mzj zk`2yX>x0jFv5%eN!~QR>P<3k@wDG;YEC$R|S^JJJ^2CqzF&2+Jnom2o${)4M8GQJI z1O9fG;Ct<2fE<1mpL}ro)qKq@ROf?z#^+zLA#T{pPVv^cZccQypXjtd#OApMaxH(- z$G`k6{;jcmF3;E(*2;SFIT`$9kI`dp;?DftVa1v63n0e&jw#3D<9q94Je$meY%x>r z?>rY5psvrn#guW@$Xw~SruJ7**Pc)4V=J9}L8pCh9mu3xOo6#Vb$#u{y${(zMz%Yv z#6EvPw)5{s^N};miJpARw)QnYi9InVj+{??4EwCI&-qmjFlTe+o8~ERwBOjr7JJ#6 zgR-`LPRTLihArgrBl&do&BQy5`*h!1t-C()Ze5J$XFfJZ^64qode3&>a<|&~lwZV} z_$JG`mfP$(<$b@oYpuyKe=<9tQR~GneSCz^BewVsl!N+?dGl=F_f3v}*u@_E(mEIe z%GJrDPh7FZzP9)6cY98Zuvc8j7wxyk?>(Kp+Vsn3Vja}!YL6A`<}7yYWjuUr&sbx= z7JJ&PKe=?Uqj+!M@E<$vPjir~eq)C^o4^|D+h>>j#zwY^NBi58AK1eVanG09p}j!A7;HbH>c)vzd4WuJ zWRHAOobe&>AwG6yvpOGn#0(p(g*Bm<4(n=7%pWf~<)vcSx_a=i2Os_AF?D_7YOjU3 zHFrA9g&k})mTf>EnZ`e+~ z%(wiC*BImTZRiU`uWLx z?Kghs|9#e_SG&D0H{k0$5Z}g&FFM78JS4ySuHico^qoU)3i@kTHpW;o*k?}WCtm2J z8_XZr#J+)_{-szBLz} z?O`!Rx4KxsV-MMb)`=h4$1XOARkFpiar(tn_w4R6<+^mSW6L)N4}EltJH9awzH*lG zrS<1GbCR3dE6v}0#X|AGCwSOPC;QBwZu4Mw>)86Cayb3QnkPQAy^xRduRij{NB1iC zXK>ziuE_o78q~Y{@G}{7+Dpz|-)Q^{xSwbGcI#V>ceSSs_@sTX_aOhU**wV?*WM4k zE4s`3*^D#T`p|1#JaQx7n~!znSH2*dK6ksm9kET>ozFW49=z-)hdeP+JXfa6I`aYB zx)Zy9$xnPDKKVk-W{X%SmmGW0I+C+2dzZ8J$Dg?2GcOIBhTR zk#*;X;+&arMEa21InP?{t0|WzI8~4^*U_J|An-5kXny?EJeF5-$Dezf2AH;V3+PnnqS9?urG zi3Q`VyE#~Udh&B|CYJYkD+bG<>|qD{`IxNkTH=jA_(i+5U@xDULwPfwlgr0qTki30 z4q}ll)|xNOoj&sHQS#|D7y8W0{PrWhvBw8|RKCkLYrw|jX;(B;uo9hxuqTPkv{+_y@L{gZ5&;*!G&TvHT{k*uB@xTCuYj*>PBWkPF2+dD(}0 zS{G}w_lYqcb1+Bi%V+HKu*n*Bj-t)OnEcDH#y5ssV?B7-FNg7McO&s8x1BLYmh~wI z>$eWt*hRlMu=edaF`&)-(v7m+e2W`<$NHRc!uICP=DqLmnzuGSKliV_WL-kJjoosH zSn5t=&Y*7JHFxn@jM2e&^tF%JU=E#Y=56mgLs9kiese3H$OAEedRlLKv)5i=4=OKf z=kxXss;=F7S|3!KxFM4c>r>i?%KoVO%|Erfp3ck`;za(&h~42*IwSf zHJ%*)U>6(gP3INgxvM#sozL$68$URArtWgan1lIwte4zS9u?p8T5n(w5AYbh8g>N?I%om`n z?L1%D+Zx#qV#ECC<9{*kj_6$Vt+4ae-nV9K6=!0}SU#~v9x-aI+Pid_Ka@-L;V}yQ|3y#gX`@A1~X<;Iq!Z z_E+!U?!|tl;C;l;vfP!uFa6*^9&?{zuXSb=pC#6ua4Ft@yCLnSJzU7u)g3 zY1S9`R9p6<5HG*;zcm5t(Yu`SWRa0fYlt7>(?_PY;*0hr3g)7{K0cR&_FbSk^Ru~S zAKOEAu(>;lc$PDr9XofjiHzcCubVPki+g^z7VW*(t2JS#bqU#R&w{;UZsgLNZTU1m zv0Hy@$Uaa9J}jR2M;r8i&8%-6z9XeSv&!TL*r;`tqCK?cjgE-`nq? z^I1QS+`rq<_g8CeEz5r=t*?D1miC=N+=y#&B!*FO6yMH~;!*zIC|BE0_M2Gm&gHJ< z8v~ z9Y#NA@Y#dU9Qd2Fe>C`qgHK*i|I~rM_k(!;!Tf#NCk{S-@Nt8W9em8-qX%eE|H#3I z4nA=3euMWK{I$Wm4c_%}C~N!c1O1?U^1b_`@BZj}4&Hb00fP@2kpH&_bbZv|BL><( zXz;fl#|J<9!H<5}VDlP-7vJ9)oOWdseeXSZp8=X}Y{3)QdfG-lo9F^%HfYZlzWcj_ zzd!h-0sr&wCk+1H;O`8KZJe>{+Ks{A8nr&!tPdHey7jZ(esA30Q+?&&s~?BIVfMF& z{>D)J==1j!;>+Lt_+0_tHhVvCe-pFX+~-hr?+wlvu`b8k6P>ZnC-=JclszZky4$(S z`MpiKvA@Zb>jwPuRfDe>@Y$CQzGT3M@I?bY{rth_4nAw}j|ZPJ_{72A9em8- zBL^Qk_*;YbAH4U#dcDiQ8ou4&FAUyd@Fs&lGjM)6r>wuT%HN|qH^hju$oV6NoWagx zXM?jl z;N=E?WblUvFFW`{gO?fn!NE)2!2kK^V|d9&U-HqH7`*u4_YZ#G;P(t(Z15t37aIKT z!S5P8-{5x+o_Fv&2ET3aoP*~W{HDRP4xV}NjD!DU@M{K79z4A8q(`50uT8uDXBhn2 zficFC@oWP!fAfI+psxMd2Y7-y=xf|A8~0lV&oy}N!Sf8BGBBSjJpZH5e{a6#ZT>Gh z_`QQaFyOV99`G@rzx?2p2G-(L2Y-Cjr8i)bsj5EjC=?v}6bH+L!W?Z#kANK zC9tf6(Bg2KN6a z4gS&Kvj_ilAcuVIK(3K{{q?4?cJB1%odgeC6P42j4jOmch3TzJ2f= zkK;QZ{?*W~|6dO9{)@+9>^DD-Z+iHf?#cUxhri+BuX`L{{qR>m{8fXm7|{DggMT_; z%V!Keb?}LUj~jgC;DZN$WAN7p?=mp2w;8jvL2 z_@==(559Hq?Sp?c`0fE;{lMUd2R}CWsR4iTEB`uIe%*lYt%3C@-#f45^cNW1%IDVH z+RNk4*4G&P$w4{X+4}|qIonzMMgwu=Y<|;$b6R}8_26v>V(uLW?>z91AeVdJ@NObT z-Q~SQ?Yq49F*)3OTlxD-=g!{U+?_-0``b@<5r6;bo-UXB&g#DI{n4JWclvpTZ)Uz> z^=-%Avj@G`b{F?O!uz%NdHdIW+qVOI+a23}m(M?BAQyb{z#Y%MPY!oy{5J!+L{5=g z+&$fApK0)%gQpB$Wbl%Mmm9p=;B^LXJopQPcO1OOK(79%!6yvl=r0_6%|L$r?!gZY zeq!)TgC{@f$xnKg!EYJ-&cTZgUUKk92CqE$6NA?sywTvz25&uhyTLmS-evID2Jba^ zzrhC#K6vn9gO3<|)Zn8BA2aya0jmDD2l_u`V9fgt-ed62gTFNRi-Wfqyz$`m2d_1F zwZSV4US{wI1}`*tp22ero^kNV!;d`tvx6TUeDA=S;+*&=gMT>q*ue)4oB@A%@D_tN z9K7bh`S(%-`Tlta&pwdn-T%FFI8Uw3HxAs{KV$Im1NZRv9=LbE<-qqO-sYX9LPWaVIY6YNB@5Ce-FNE@STHi8_4DInLIAn z$>E_; zGU&sD_ag?z?z~SweSuwaI2%7{AeTF1*uL}m=RNXyzmm_t{E;uem&=`D<}ZJL$H4ja z?*{zDUq3Y9M}FmJKL5Fa_3(Hv@IK&O;5QDOt?umdx7_W$L>zdp@vd`~v&D~GEvDq= zw;cS1fp;qJRy!xVUyD=UgWR#@<_{nE?y|q($k!X*;k?WF9wXj;6LSag4aXkv-Q_a| z-N*emZN7N$Wdq+_+{wK^`W7RvyPwPL-Yvm*J^RX?z5lkE?|;4r`nza%cze%xLO*Bo z&Ct)?%Kh&1zOTDi`5x)xfV-D`@10n#_f6S1SMS#H{M!!Rec-*^o!z}puK2=%_kTIw zJ@E$yz9aa);a>jCgWo!EC;tP3KQeG1f4#w*59F)&9Q>_;Jod?h&l!CAz&-ij4t{v> zUkCEwGd<~_h&136eu{`kSC3_fG3m_=EvHA2fLHfn5K#gEt+#-r$c9 z*y-$f;lb}1{KkRv=@%aUg@=D^@O=aC3*Is0^-mgn_~5+;?(}au_)~*FHjvL>Xz<$x z?(O{S+miR9?;U*m;OhpyxwxZyzj8Lq=YMtJ8;b8_?&Q|kdzx?QzK{5Ms-K0(>te#U z5bt_^zEW=Y`_t|gVo!{UUEhwywRcZ(Z;#kB-cR3u;QOq5&U*}eBn zhue$xq}*?hx{LR1!+W#4*Z#gC|9juI=Y6a2okAYicXBzxJH9)(v%opv+d}v6Uw-75 zANi$$T=O#nd0q~Z&;Or+{Pb@J?wfL#duTcQiyrx+$#kE0FMrG3a-+MhJSwlsx9>N2 zufe+w-f8fc2X8lc>%m(M-hA-q25&rgqrsmZy#C<-8oc)4PYnL};ME4NJoux7R~o#+ z;N=G|cR-uIR~+E`V}n;27*7s)J!VyAR%T@IHh0A2>VA z%lYEGaXxgXcLq4eyofg$Gsbf>)s8#bGqxe^N4-l_3Q(8A$!8T$h(3&l6~X7wVx}z&;0!1 zZw~w{!=1{$bH;q;!1o>Jgun6op@Cd)FZ-PY`~2kx?(lNKUmy7P>KnLsd%4`V7T>jf zD|SbhYhHNZo64&UUUwjWyO+QBK(2Bp|Ez&;CEq%5SN+d}pC9-L?Jh2Nzrf%n2JXD> zzwW|sKX}i?!_@?%if_X2~b44!N7Yys}!jpb{@B@SI9Q?DvmkikRi39ihzd3Nm zcvtvygV!0n%HX93-ZA9&-!$<2+`~Wj@Q)0>7bjSI-r+R>mMzqVH&yy`$+2mL#@zriDKaPLfU zzP$C|?FY`DcOLw;ft>yU1M_g_cgJ_9@7s)A?%hN__kHFI246bZIoOY?7p3W^#x_`aW6RVeck$t1K-_VZcu)gD<5U<`rgOK3*x(NjUSsfQ2EK>*4)Xp3-#g^L&m72y zUq5g+{@%fV9Q@=!?tEnMYoGjU@8#y_e)4la`Gp2AHF)L0>ki(0@D77_9sKpddkx-m z@Ye?KJb1^!+YjD$@HT_LIC$&9TMgc9@Mj0FH+ZeVs}5d%@X~`9AISI5KakU(Wgt2H z*MYNxtzR|xoPqPjcbxYbywkuuok8;Ws}A_UyTywPo^SBm2hTq6Eyks>oc>)8f7iXQ z_MiRmXWz@`@B8rkKK%9r-(K9){S3mn>F2QWxcj-Esrq@w*AIMe@g3dIK|gNb8@l%^ zKlA)613BHdmp2{wT^`@j{fyha{WTu{y&Qid=6kx|gXzEJ`iyF_NDup zd)o^S>~lH5_ZdIm^Zn%)ANj?5cW?RGH4joym3z1FYfj3Q@+o*o7v;;b-vB|Ii;WTxjV~0z7xyAzBhdL;C~FB z=gH6WA&AQSAOV8KlG%3JNV{-?>7H<;2!RM!T0nx z7`*zxHx}Q}|KPw~{0|LYc7VRr;P($+aNw+a)`7>niSO#)Gw`!nKXdg>{X++TWne8{ zcJRCdzf0qr`M(?ZS*Y(T`yT$T^ZP|_J@BpE&pPF0zi;AqOyuHc9mvPv=bhfg`aO2P zkAl~`neT6YJ}RegXvgCl-E$87?A`D7y2Jan=Vw0d2EHfC?S5wDn~Xby`sWS4aPTDq z`QLZOuU>HH58BCsc#JXD&!47re)NueX=n5A8SK zQ2cz)ebT+sJ=@O$|%k6UKw-4mh&tEzHBcA#ZPyN7w{QTaJ<9(j`K2QA{gAX43 z?ZGDwK7H_ogRdFJ5-QM>bIotOfFvk5K@9PbGH}d_)8OT@iH{W?bk)Qj%FGu@+<2#P; zQuyS1?*_fA_!b0wOQv@Ra=p{|E`;AZgm)d^w4A?xeej--gUojw;Im%V>MuSHa=afI z<9uP4Z%Cgs=w}GtwVW5~L0o|Iu|=1Ko<@DfjciF z-(~QogI5{2LqF5t=brR)Px|43Zxim^zJ)x;lb_?se`w%-{E>q%8<73q2LE&L%Y&aD ze9z#k2hN2L82r_Nd$s$uJGV1LK4!zG4$u!7{N=%G4}R~!_m-zV>8Vfp`N1y@e$A7A z&6A&V@FD|uaCh;S9r)5Bhr3j<^IHer8GL{D9o~CHa2Nl$f%gkPU-h$A-&Xt#_Wz^o zKA`2S%CCVlb3^ZhCXs{y(yI~_1QN<0=|~e0NCE;%QBee?CUgXmCcOouOYa~hp(98a zq$Yq72tD-V&hYKuaG(F0m5;T)^<}Tc%$<45dCz;!e)hAU^A7rNw}8j8eXvy^H(5Iv z5-c9zaUS|}fxO0=7#iGqA8TEf4VDO&4tOi82Wtji>GN*|#HY2K4+S{=vcOs-uhuJy z%hv`M1t$c*3HU5S0$N-A#_xC;=UeY&?Mtn5I()jrJ06?;tX<^0S<9}TZVik$FjGLI zPYjGH8ePo?M^{Y<7yC@@N4Hbxs5jC2qXS$m@1*VN zd1KsoS6AbM(?R$dC-VjH2F^GlFxSm(ynuJ`6t9`SUMawFH0R&||IQKMLcI8D@OW@< zaC?A5uMY6(1py6AQ|}-ABG@t5CfKqM-f!0j?>7t935Enq2MY)D2mOP2gN1^{g5`rB z1h(Bc*gDuD*frQIpwE9F92^`L92L;!borUVg#j&oYoP7j8vOOv>#a9~_X9e7YT$Zl zar_T62K@ptNS^UoAlLXfu#b4*bN>{&dII~~Z`NS0V1Yo+qF=?5MT3FCf`RXy$G*L- z-cB1F%W>>Ke_;C+f*%HA2i-4*;?JW3>vV7|eSTvgHq+?xFWiZf@wD+kr+}tE zGdMjsAvi23-rlum*Ut9)#WQg{oQ>a~4c-hs48#EX{KbIAzdfMA&kr2?&%qx98vKCZ zSHXw?-_znl0$%yxpnUS#dS>gHA&9lqGqpLj6NBDwzTfZRj?n}f>&^;fl6 zdR={X=Kzj;e3V16YH9l%QA6UdxwQ=!9u5%92hE1Ly8t(Ag>13o!j{;%Mk z09UJfUKr5l#pAdgaJuhK2}TE=5!fYA6Xmn;T=ognORa-BHQ=@LXMCm>zE!Y#pk7MX z`-i_L2cAi@UXd<;Gr+$u1WyDH1?u_5>-09i-FL49*4WbY;sL*Y*1j67`s-`KV}W%7 zw7scf+^g*YGdt7{04qnC_Nu^HTUI^ffN0>-f_&G(Ae6@<$d5XjlA! zi*O50`7-z>Kb!y#6}aBN!Rr z_$`7>1026qAQt!@Pm2+_`P_g{^0$DFzd5)g7#r|xX#A%G?RqH?f5rzBf_n8{`)T)g zg2@3c<_XOb%pKrf8Xn)`QT)4jP&|y2LC#V9EC=Dch>5ENs|I4E*C7Ev#&;lR@w#%L z|9w}mukGbF?vI>6&O`Ga65veUigh+PQ+-vQ#lzsG@iWA8_uu!{Qu5VsY;n6dZ(KMI z{f&ch_jv)n$KB4O?hD@Ie7tTu`(6{Mhtt?&18aSFIlLAgj{1unkWbTyk!l^!d;!nI zI4#cvmoFROZGMUR5bbSTTQ4TxJTJiCcLj8iIY#UB&$gbbLNbK+ikxN8xtE31{Vj12l#tfuy`;>VBLy!pw=RKKT9xgz%z&Wf|&x} ztASeE@Vak>{r1%|IOR+}8O4eplo6oDf|O-t*f$i(oy>?*i-9M+6%N z>ZZmKKF7(lHqPar(9p&m9|ecg;Zr!M`lt2e8FF8}6KC8!Dot)a9v;ZYCA4eiNn7L zJ`LzrKJs(H!vU}P^5E=%H;rqHyZ34B)6wN4S|d7~yg_S&)>;Amy+W{DuxuY_ZlA5# z2Y>r+nLhaYyTK2G9|bh|&jRt_h=A83Cy*Ol6I>tg!tuPnuMBX%ZM0RK`AgvStYA!F z-)jT8#(jaz2cE03mgKNN9TE@oxo-{l(F^ojpxeZqPrVnI8*i z{G$S%3f^2XP|KVvz?n2~|9}U-LcoXrNuXY-rpY%sBjB~*ZN51loFC3R7i(S%=yS!n z*#hT>*#n>PY~&gD1^n}K0y%(o(d#@Meh;69rspBi_paUcG(K%F&yZt?4XXsaq(uX3 zU&I-(erCR4ra&xsIk-CzC#`=U8PMw1E?T!ZwK265OJ8if*m?TNGf{U1yz}b=&na08 z#6$l>uuHIU@cm%PfR|3U4-VYRRRSLS&|r;#Pctx33$X_2sQ{m=mD2OHJAcMA2G<4F zpn9eA&kBwR_6Tq}e`Wf>8W$eORe`$834!rwT&lD4^mrnB1ad_DYaCk}rk<&;NmHxg z?;Y63Smvh~$DY;Th128esJ{(tyGdZ(fjm{si2hM)x;3D!_$fS59xiRSbU<(7XZ*lt zz{ghvxJGRd&JWZf@iczHhhA;xbKHZ2_}Rw>JdFbay89QwP62(rZm?>Amlq8d4tOSo zZ@0eP`D~HaA{`%P$<~r#?d4mm1ZxI(d#iwESLZw=;G5uMZM!AFn@hO0V{RJ;2%a=dsKl$W!=DG&HTuTL)U(-(n))W0|1p0`^h+ zrSIi6^uGA0ZI1tTAQzD1@M>_hcxoRxfcu22-7}uOwKn``K05AnU-9gj0bag3pwDSy z_aBG;CzuqxA9%*#qkzxBfBQ6`lRbwpHPC?i=lztx_x9tD(EIegx(c4QUWb-1{VoP* zi@#}g{?7dYe+8Gz5%IXO$`7Qq<&!4|mjt&2w9UT)c__b8PAXUR&$FQB8Q^Z>&OfgR zO&+Ue^__q|-!!1hjtu1L)nMHbCJ#PBu^v(4G98Z7Jqr8|+0_#NZAT4}J zK*!_0s{@?K6FED;lSc$R^`8g)7QQ-P9uLyw^t+tE+P{gxTY+=A80!O<2;>XSOPAw% z+FMPPM#uZz?|C>!1~?dRYXg7Xdu_Il_B+;Q!O(!;BA$q2Vh*joZm?yrbHJbZePCaH zO?g2(^z6{HX+Zn06j%efP#{K$ReXA0&4bmxZZQ7H2wU+T!A$)ev9D4z*?o} zgMS2f2IBmA0q^Ax0Y8RrH*Watyb(SNtw4L<7VtyP2#yJ?owATAeB)hGK03k059SO4*TuD}{Y^Ub_8&*ZVz4LlUw9mt{Oz2?*f0Y8*ajsw)w ztV3Bdz&kuz-YkvzLV(-(6s0He-Iuo4z%DYtBF%pyhG==E2T^bMwV%U9pJ% zrpakrJnxmC?=$g;UoIY98qmR|(`ob*0>^S)*2Flr*d)L3DhF`AqXY3v9J4k?-|*rc z)3NzJ&f|RY0{MZq{5=>KsLhH!;s?!NK0U4%f7}mogf~BPK!@{s_&Dkqyc)HQZjKkP z)jw(Hqk}&O_|^TtJW$WRGk7?#rsUP&tpLB{bK{ARZJj&b^(x2XiF}pyCI02Mb$B1o ze;HVR0yy8R&wOV)``Di^K6hYU6%U+$Vl6G-d&gkUz}zrLP77%7>jSwb4`qBndpn>S z(!zl?WioZ_pvJ{JJs>zbI3rL4<-wZIp7W&P%>VBN8w5KBzX^^BE(kp5 zZtVq4X|0R;-tU8p0z8G2@h?uKv+@2hSMr zZ)kbWl6c0Oe>(6C6Yjq-_*1ZdpuW0MV2#WY!2*Fbq5ioq>Z|Vu*2#EI>G6Qxx7Ocp zCL9v%9;ms}?=(HWR}1IO<9W{~@o8QSo(}wW49$K{K(F)E)l`e)t!?4w{Wo|%kQ?&K zc`p|PboL>EG41&vYl7vV3kEX=<{7R1MnD@`o6G}+=K}K*k3Sxmr`DhGR&EP;qVyX5 zM7y0I$fI#PO(z#uzrq7M1seyvSsIjI{#EO%j+gLa)($@0=cnK*x}Bdwi|-fg8Q}1r z1e*l3I4^{@rm>d_)Gh}G^!R)MUY{$NJ(w++HSjt|;JXC_+b$_qMkkF2=)m2KNSd_=&)}#nSRL zKCeYQqPbq4hpOvIVLzRP;>NpY;a^SI-qxL2Zshew-)@Y;IiPB07pI( zycWD0SZ5)XW?@n=NP9WQwF{2=gNyrAD{cic_aTljcXM^t+_aeYTd}nweVzai3;CIq*ypzi!`v4?`;$ zOP)1dEm$rX7|b2W4L^_1{!Z{}@KkVrAm21@c`AGw-pmexxv_jOe?X|HYvS^8!F7T8 zeQIFNoBO{FtUcqATkn7$tbJG~SRlX|h*MirTOS2< zjZdu?LZvhtJC#j1Fv1(lSbyR z(8D}tT9^)YPBl;dww&UrfM+89@KC(^o0si-e9Tkg7x9X$Z=tV`3AEEW4(fyVK07-2 zQ$YKRWA?-IG`}^J;+($FC;ts_IUeV&Si@pXI&Ccn0&9YN#(%-_K!4+U{jJuj2JPMq z4K@vS2!0Xl7x1R};qEn`Tb#ZkFa~Z7aJn4r$w2IXJ)j5ZXXA!e#_6=O^&`gA_XC`6 ztskDInQ<-;+nN!c-(dmXJvLB>{8PY7q(`hN;g@uC@ws{K|H9zXfcJPqaC<;=^HbFQ zp9@|I{u7u79|g3J^;{Zj!kH)5cFBEtDb@yJZM}tHh466E*KxY6?_m_pULOIfinej1UPX(AP-q4_-f18i4zPdi5Ue5AgVdfwk`4@VRXq zhY#ksj*suJ39bk(3C0B0+nf^A?-7WN)=_R9xRwuz=SuUhsm%47J_XbgfQ})|h!b;OX=1PYDhR ztc~EL|@qk8>ldtlQ%|E8>3v*dz2mL{pWZrbzOwZn^9JGq-6|&FS~~Z^059U@ zTl;`>`4VDveD{!GeD$ z2mEaH-B*K(1i?bN$W|B7x|*+jp+&cn-1rD|2Ux2e;SMoJi|4)4iTjtq_uP7Y2F{AQ%zWIR8R!+`hVknecrey`ECr}e>mpVz(@ z<~JuV39blm@U?+l$nOR4$oMT{*}Vas4&M9Bcee*O2KIFv?QxD10~|goP>aReV(iYr z4#8HzCc*y%YXx{){dL(u&NeXc`#t>wYj$Q1^rakN%5+mY{mtilGnf#_32^&;0gYjd zTo_nC`NsgyTX!QT*fdx(@LUtGk*0}t(pdx9oeSI+f8QDb{C`eBUzvk{42}+t2u25^ z0&`T|idO5!fqAag+A(!rbJ=s&>bUA$a%*`z4p=J?OROEnA)X200jSZv7Vrl=KQcDx z_Wm!-N(`K3t%SJY_1u6bd~x78`qHL%r*3e6@MwU;_)#>fnjo+0)xfJ-gU@}hX2`3; zd((PVf6LnJO9DO>?&Nh?-=PM1XrLBheaEi@YX))eZh?B`$Y6M|Qy2VgJ*e-+Jeu2n z_IDi5_TcPMfi+0Dn3yj z{}=r2^LqliOgzB7e&fsh6-OK&KjUfh*zZ36I*{YbGe!i~Pj4S=7ifRM-*qh`v-ZyR z-*i3V#jydesp6NMMSKxw_zdzF*CZD49`KAd-xavNv3+no1$^Ya?`*@n(jQ`rW9XlX zajxC5{p~y3+1`HRg8Ku$hkB*DHUG;Rm1hF+$uk7}u$KZ__w_)|^iCjFiBr}zR%}uu z;gyL`fMeB1@oXM55WYHmn=)0L^xm-Ul+ON@qltNrDF!;eJWgy>S9Sgu18X9+^N~Pb zxE9yt8hLpY%jt1(*F7H-_-%am{x5yM&tCWc;5^a9-{IREo9?;Vkn#M(fJRv=7!=Sj zvjr~Vi?6=u)R^AN{}aHR;X#@|)=rvBcLlV*xg}S=z6<`Qvu+M<4a5c8@-(d<0~!px z_qlCsYd`y68_2^i4|t>IGLK2_t}Z5yh!f_#90Lc45A+}(mNqPp*BUn-!Onp=Vta8# zjKLS>6^K)O0_Qj-pdHJn+tGtpR50gg|-zhwb8yjwt{%-Gv9$eDcI-G{}0&SzK&PEiN27>Y5y6) z*@0`SK2Vn|=o8oMJKN|l8o)lFAM~Z;=vUgnvGog$aD2e0DfrvxbuZj|$JDnt-54=$ z?hTB|XM&f4Hv--a9X~bb$-)pCo=%sm4h%dCV$9Ps#{Vk8s)3r;54s=*6uLjJ_r9a= z*9a<3Y>?Wk_(1P(9pF=Ofj_Sfi%ZQ#+DeR2m-XuJX|ZR1SH$-AGk1pvxEWWAJ#ujK z*tyMR*T7d=DGl*CuI>bD_)X+YxllsECfgHxO@~%%D^y+Wt z_gz7(1pC+*)(*7Maltv951d~+v`PE4)qU}N{W5_(%Du8a@H>HfH#k@%z*+7mP0CyQ zRv-tPH*mk@Y}4ZFjP2|HRvu@2+-KdBG2%Fm2lBakUw``yoOjyqYa8DcY(rycoAWrk zoYJ_QF)&7XD7}H)6ZaPA6yuoQCaV1N-gdl49;ETE{#rb3F34@IqZ}LHSlp_vbX`EB zTf>gCa2PH#e{1fVzqrnP_PIHZ*KB)5U|+{@EXNcB@VE1b3wL)t|H?a66IRDkGq9%K z+B9Bj!Drg+dmfTC4%%-&`@0T!IPjV8Y{Qeak9{5U-@%K4e13fJT7VDK&h*c;;8}@p z;QjyWd&l7u=x66p?{;qOz(Lwq?e*T8C)aHoZPO^5 zV#=c{jW9Ef9Y(CT9=C(O*uABcj3#Z`;Jc4iN zZtKkHS3V?u#Gy0}--6!eVE~V%UcIM92%&hj_v~OPVZU+NyE#J=_~Nw zni^W$evYLbw7+)Ydb*qMLrdfFkpV7u4fwn?I8GNkTqpj-!MGAH7rOZwuNF7s;o@SS z`3~3GPmIE`I3NG>ZS}4ApvH$|yZP00#gRC9bbvqk2y$2JR^+oh5uOI$nvYOEczFqO zVqOBs`RjeX`kvpw7vK}fo%tk=<(T}I^4Iy~HT-*OYqy9f= zVuIKpPKW_wi@4)=rugRirTRo)l%ChGaszdN@?!KmhzZle=d{20=N=V!P}8Emvn`ar zQ-Ay3S`T@O`|kepw!|LeqjCykN4%iPjVm!)9KiMBc(>LT-|_t{fpLq|yPq>C?Mipx zclwnUDULT^%p>!{9HHsVF2=^Cb8ey96TwUIl-X zhr#dUeeyuny{uug=8Q+GF2*k{d{dpw`Zd0(HE-SCDWAl7cIt!o1)r&@c`i_0O*{P# zo_1@$>u^2dkLwhVcoE&0wh#1$ZEeG!70W8euQeI`0N~wr_o;pKq4SDW+M*_)UEqDS zPYe{hoUdY*^N3sEJ&(cp#j|PQbKh6nv_pINL3|M21usI}<7e&BFL zCGHrj#%1LK#x@;8&veHO`duzkdYxtme?zyvul;=HJMlyQU@m#)LL8AVc6;1ngLzB` z(vftcT)x}O#v{cG-JDSQzwLc)8+=5=(zN0NAA=6Z>$s3M#*x;W^M=dk#;N$3Kg;Xn zzp6{sdKBKNMFFu>{d`HI84!kiSoK2wDUa4V7QQJD*E)9Isk)ZB?YRE=ySqkaA9c&}WckGY-4R_byu~i(u#fmtzHfKz63?`uJaF+!tkb_ful>X>e!TX$4^{j1J1LdFd{g|YJVs0x z@5L?}K-?0u#1dnNjxd(!3x1CA#mh0)csNT0;)Yy-Cqk=;A26+7QeH`UIrNQORBVw$ zh$qIi@ok*TY0Uxip*v@$vkG)qq4I``EAj#QQCy()@Vt4AycXuMtP^>8pQ zc3h>P~;M>z)$);8Dcy0r`6;&ibC?|Su_zQM^jTwa7{9Yg=P zR{5F!#g)MK$DP{coce;+#?#J)%dPje?nhh|7rm|#$bYOAk>iMCwikQcGw0O)x_;-@ zH~PT!d8N_$N&1f;#jg_o#Tao-&LdxQT(4jZcok3N8#ID_?C*GVg)vZZ0Eff)fG+8- zRaCxfdwH(BwmdC4tiIAt{X(CJQI&7_TfQpxxM%L4^N2OZnmiou(^b56nhc-ufsJ{e znM>x5+?_{DWAUKG8)Ml0F-9D}?pf6{#3ep0uSz^Jhj_}4k1NeF`cS+umvOAm&E-0` z_Gzc~YPY$kpQ`quze>NUeYid`)wVnYUV?fUU!mJOb1i%`?V=~~1|7=B(8liisD8l} z+9Y=36@D8pLx0rVb&l@y_)aX=fBF$W>0jQxYZj|;C~m`z{2|;3_*{G{9v2V9GckiF zQ@khk=uf&DkLf?YKE16!@h}b+E5#Ad5Ae~f4WapgZ$vd#8)v2-->6& zi&dj_E-_KOqxW$$4-juVmzW~=kju!Ez=r7ZE~ITh1evX$%*I*=dAqC+EaQ$-X)gl5Aj}`T?38e_{J^1tD2;-Mu*{L zW6D4C!I*JBUAvg>{PJ9J)^)h2-SYlk|9`48Tk zU-Y9n=$St{PEE|cchBiAaD4NMjuDGXrx`!Spm8l08l$esTs6O(gO^=Ai=)N2;%Xd2 z*NPW72tVs1^Uho|FT`DA-tpyW`itM-JYt@NJV{5JfB44yuK8Kqfv<2Bj@N#%n{Jmw@Tue#VyKu^@ruvoGcg^f%U8r# z`30S?R)GH;8|S!x>Q}ghUU1#+5&vEv;ePSPzI1_DA%5}ghV`NHFdm=lcV7LE*To|J zt?%3~9D~D*UEJjr-?$GngfT2$h+h{3bb-DVhaKCsl}4d2+}ny%JSqJuhvp%b-(-H2 z=j6J1v9yBN=6*oczP#!q=dUb~?Z-p3!^xk;^G#kX=NoR0&=CGEB^PIEr< zkjBNGp9b3MGv~lTViiuJ?H!lyuUyBgTtIA;yWmzaiMJ@{u&)?^he157nCNe@(KVF5 z*FF$40AF_3$9*;|kORp5)IVaSz7mhb1#1-C!{R~rL4NH% zi*3FW$HV~f+j(7=_*c2MYZb%A4(IUdZ|4$AplU?!fi}|@u1%bFKb*rkjSbobN0n~D zkN$~$J~uwZSRBrmqlfgdKGb(&i~FTr`h$LPjLd7>R zP3$qYjX`5a-xx>QZcT!FtzTWgF;Hvcw83Zkz;(N3`x$$#$@nhrr}0ZC<6wH8A1@}D zuja2gVJ_fO*X!OJd-84?$+~RdU&vLp!#&eZ{f8?nhTTlG2D6`L2 z)tZbOIWr9+-vrk*DiGIQ8(jkW-u~9*Ij6kQJqMaiF6}-U7ln%7+93XxHsZ6E{*(7= zlRB4c^EdE8U8C!;#ztOdT!FsS&*p`GHpYxupBaDJAufyiuHF6do*zIn7|Ysge!Azz z2`yq?8*`vm%Srb0Y2?zD}5@SS7hqM8RZa&cnCXz@j^ zYMkOLbHSWY*QhmAa!NcZ&YL4%#d`6;w&s&|Sg}XmfbZmgVhdeQzl&31 z6wbGun1Oe3gX23F?c!A%w8gpbIG)6rxJ3<44kj1#41}DC539x`PZWQ(0q=;XH7~Ut z=M^8|2Yq5Yaf&Vw%ggKY?8%sbpIhrEtmBY-%Kzk{{^mDYN5SK?zu1pw<)(Z}`77U2 zj_MqEQGTqC!~*@Ky?7aK+Fu)Zr9e~2|D2cKS+%fQ!y!N9d#c5$>#0M^9X)ShTikDd z{-rz}Pvd3zuX`f?=wtU${v~FK$?lW*C{EH1@?C31T)Td8{qk{rU@c13_2kgvj=Wr6 z?K+JEV?_QaHjDA%w3vmDX+dL}-ttO6iZ9}Stv&WyV?tZ$jq+-F&pccGs=bxFx(2#N zoN_$#g>I(Xd20Axeu49g2k{oZ;I)Z8=Ba(n9mlHvDIeJwrs2#f8W~UEFPdGRhRfwh zo)J^;^4kyUW9nc$Ky@RrMqBVGexQJ6JR-;op z^DLp?67uYt`kLPo;x)?O% z@>2P}`zOBXOZVKdyo$f#oc4-OV!z|)Z|~J-#SJ-~eA2avEza%wjU#c+cVdm4TAu?= zrmgP3SS(hGWAuPAOB)#P;)JmzHqeP)E4GM3;)?B!Te_CUqi<^6k$C6$?!ET8H|CIh z(|9+3E5`72Xl-*47vepf!Q(2Q4hP~q+u>(9hJWH*X>M_oZy^5^8(jlkjxTTr9>FK( zx4B-mC-DJSsoUT?+>4v|Xx2J-wT{l(1^lRP_i(`P^BGPRUvQmc$d&Llc#l7QXB)gL zE{H=sJ8Lv11=fGmZ_bHJ&f}b7mRc3~+qtdr@I1M7BK;cu8a;tE9GPWh_hlsrkychBTXcn`0;*ZN(5+fSSl7pgXA9Y^)QwKCc-PK%fJ6CczRD-W$2 zCa!gl#Ux{eUJ#GuuI_`eCBB0ACT7Y%c%;UcJkzm?o8>yjyKye}p!LiZx=G$Z8&u9F zK8h{kiF+?*iqHB%JLwskpH9Qow4OPFyUc6zU+zTb)1vr8JTaf?LGg~Zp@YPG{bo$* z6W8Xv>eFJSIg1DRZ}=6reV*Urd@Xo3csh77m>l?xDtsyyXotC~jzS+g4oyg3ng4Pq z?Zz$iur}jUe5MAqdLU+5bAeaYr`8YTD%RoET0GosZ2}GiaRs-Ec|1bDFXwpn^QwJ* z6GNQvx47ZF@=dQe)iH2-d8u+D+brK@KNBmSn79{tQ)b` zL|gcy`Vo)fPw__F#obk(bZ*z~cY8l+f71DXHcrZKxqKXa7JLxA8N3^aXX1-`omeV{ ziG^YleNp_3=W)K6p}*Amtbq}u#Aw&>Mc_AD#VEfYVjIWRr!-XGdNftCiaVgd~Ld> z*3;-$*QReBmo5=|Rt*7p7KGq!OotaZ31M}OQs@ekG!{e!#AV&h;sl1FvG>)}d zet|>qAl*rq;^wcbMPILB!%1Qgjwo-!xofQjohn}8c3k2(VuyWvW{%<(`-)%o!w-0? z)<#rq%r;_&=V;7lyo~e23C~62TUm{sr@U8fyn5gFSm~Cr(scA4-~V+S#fegV7;R@c|KiV&KmgV-1avv+!tQ3d&W<7 zeq&w!A?L0du=ddX@?2bPA8}V-iErYtdn2C9xzr1_Putvg9(#u|iHOM=&?c6JCVl;Sjvzx8=-#^Au0; zXKK#UTwIn%YnXZf7$tclSlG9 zCU*z-1pf$L3B*jX!umseELOP|@u1?i>vm2)l^CHd{vTo==J!FK4;<^E!14bXJQeu- zjX=DR`^d-S9NHz0IfnRHexGydH+(7{g1^O4aaSHAw)=nI)Nf6Rg%z9Ki=lxW$a+NF zjO+EUW8#13bZ+riAF0SxE1OFFM}>NEFT zOx0iV1#v(gg3paj+L7O2J{iB}1FkXmaXO#feXV>)&Sb0@v-ANyC5{?Lw2S$mHY5MQ z8)69FFdyhf{2^bjXGOesZk`v-M+ebHVx0FhB8}v`(n{u%{GUgKb8)ZZ;sdd(bf(W; zA3o!sh!J!-eL5zI6{tN%>cgTADff4dg;@YZ@;4LwS?v?+@H+V|U zg;R0RxJ(bBZDQZjoSE zuw5`b*gSAf{W(KGg!)Ysu|-VMH#CFTMjyzT=pH)7{d5nVPptMCt>u2|U;DTR`bYeb z8;buRrZ~3oL^ts(9ov{QE_uT8ak-5-(OnlYex1|&aNXvGc_%ND^MLWD54_seKgUyF zGAG3XuVN8i@SgwVyvC+@Y5ww}t(l%`oAGHnViV5-y zIRJf5r{iF;2?vWCUinF4hnxn#<9V@%4`uDTHpx}KOdRSB?B^K1x3Bf?zW=oE|G%T{ zuBq}CYwqO<&h2}z{{!+)#as1hT@9PwKG<`c?`wAFXwjeBA3 zAFoM1fotV7@-Mld+BrP0B_>o1!2e=NX_p!Yj%gjDSS?4<_wrQR<9b@e zcXAgoQH(NP#4BED<)Gpg{bJ0}I5d=aV2l{6#-6;UG?sBC9?@6@zMl5*(!@UAme@m+ z8RK4gYH}>EyeOW5v2E=d#ah*>j?$Q^4!(1D`2nWlX#9lc9Kg>0VCjzT; z$&Ho?Rtwe)#1rRtoqn6xb@Q+)hvd_U=X5`K?;7NLINYnXg!+P)CXVSJ&n`KRb$))g zi!bNC%$VQZ;>Rr+d@uMxAg0O<<%rtKpW~s4Ep&zJc8zu2#s(c!{VRX6owkYf;A z>O7!t-9tHq8i*Q-d#KN}OU|Q~qp$V7cwN7L<~z?_i5>QLKKTd^r!BOdr`3H<_rSey zd~J7K`p*3nWAveI9GF%qqh)aj&XG&VW%zeE+23*r zoF@OkgSZsWi6!D84!6Df1wcfn_w_pq zd^qO;aX@U4>&dP7hvJxc$xq`!YBP_?dE{uGHFJFDg&a&avR4JA9;ED zO`o|waZBH*ajC<|i}VNIObtv77AI|MKmCEP9UHG3bH;-)U@W-5_*6bYPuNE;OA~nh zSu9t-5~nto>v+a0orDu{e66K5Msa1;7>#>lN?vWvoHjX^xaz*@C-<5^>)Pd0 z@+um}xHi7U5y!0cOzJS^q4`CZ$$4l8W72zbr$EEVbHo&L*xZ#D(UyCXKYIoYZLCkY4{6Y%1^{M{EEZzub71Eotr1_A0H5E%}>pLZO~pi zj_$J7xaL1@!u|MMo`cW+9o8Qb{V-?&d+RIndzqZxC@)x;*IqKeOk3Q1xj%jRp55}6D#C$cr z90&NqjkT+MIJv{bWw)E7xB0 zlBY-5if6chUxr`tJFeoL@CtE?ctd~UF&G5Hj~$2o8muJW04iA&;! z_#u|)GuOihwRVp0DPH17JSisf;_$whB?gFzVwn6yZ9^U@_Bn_9B5qgwc78Fy=dQ!O z!p*go+xc8KKN084L%f2xSZg1|2Dzm?!M={AFSOe+^d(&&*3j(6fIjrO?VXcO5UZU_ ztgC#h*1gdWyi02WDsQ4UXphPljRDsxw(wKISdsrazWA^IteMk3nnf%W8^kfdzy79+ zF>7qPKJ&)8w8y;X zr<68UpSC~#ko%aYIMDXuwBu>N^YdGrSIlr6*Q_l*uec(nIG*}?#ZBCd3-K*ZtD1qh zfty`N)uJ5NxwOID#ber4`q**h7TWK)jv-F+tmFVtv**z z)Ia*$eKOCCL;s9p?G%swGjG&btImUq%xUutZ{aEPO^lKY;Q)D$d4dznG1o#j$ph(5 zeg~e$2RIui;BY<hZwT>Qd!w5^yWe{#LJfrlYRIF`63Pr-*E*7=*~;#fQ^^>R9$ z-sjng8~hpZ5#Q1K;;l6s;-^?BzZ2K-i1WH0^ByPSSDY)qk(Nl@)TCpEb%a61hKjLNWr}G^Lv>ShFufB6!oUMlA+_tg3{d^AU@!G2$ z{0@Kq zxHQI{%Y3Azy?1|okJF81*X=&JE-_Giq$Tu)@t{5CGL2;385?4Q`C^{p6Jy;RbsjOw zez;!Tb+3&9amt*c`;2RI$FcQ;ed$qrPcP!O`V60xrW9Y~N4Q_kA`akXTxma=)phDq zZRTg<1nV5cQQyo3UB^&d)Gz#6$ZU z3+4j-V{V8|#-9A$Ty#wD%?+*>5ll(tN)K~of z8(#@{WB89QcWi!*?TZijIJnx{Yi&{Mm^14I77msVHVl3l{3$piI6nAIU`?62(2l`w z!LGs9!5V%4FYv)xm#|1MAW-M97R)*VHFaylW=QQ`y~Xy{3ak@&-6+^9*gn`U*fy|M zZtvjWV05s5@bh3~;Jn%g+F~8VpkS83wOCuHt|i`zgX(gX`|#e>r1XzI6_?~Jo@uhb zb%FoW_y5RRI<=trgSmr+g5`p$SJ`gyV2;3bIv?Lqe53)~JMmG@gxke)ItRBHJ2bfX zAg+kLa#M9B+$>iye#9qn&e)Tu$cKy%dO?g3m&_029yj7dUIJeM-|_eHgjxx`MYGe^ z_#5|{S7Iw4L|pYxyui)IH6FG--<~I=Hg7GI`aYgh)5a6@GVO{_9UK3fgXXw)i3zkH zzHvPB8_(bt>k)7UuSuKLWN2=hpH{{p@(RayZH_HxF!#h1niB^&HVuz=z<%0|D|trz zC-oESr*IdJR@=dc)^p)^^#~k@gZOdy8IRNM+J&d}qrb%g>jM0a9{vUOe_on35Y`&l z7hm(bTptd1Kh*W{srS4_am8`e`T20juGU%m{~TI-t?$)`#1h_@zS2g=#LKQvy~;6K z;rjQ2mxD)xJA!M1EBpT6WOugj?5v-AA^*>@iNS|~wHDSDSO@4jtS?Y=62r7nKe|s) z>%#S)broNksfu1CzX9?`lC*Q1|mU7vde&R?}Y_suoA z2lm$&;=F6-L28fO%s4lf=~AAIe3}+BE{#$8!ZM?FKW!#Y~e zy6}m_4D}N=6>-XX1~qJTdBCx_ylUB5BG6frHDhY(YUhIju|OQapI%o8HVO6$@cGe! z*L{QCgAoDFhv9+0cL{zG>=PUsoETgj+#Wm@{401NcrJJ{cqI6Ha7l1n@Y}$##S#0T z5}Y5L7l>={*Wm8p)!_4>(QGt(8@-M9gV%z82hRqN2V;XPfYMw6K*Vg?;AhY%Okh0S%GZ7y27^;uv|m^D`B zF76&0(7$*UFFU8Un3LKgm!Zk{@!q=@o{Jns4q#5=HMM=``Lg|G=Q$So8E@iG=Pa(O zHNbenxpBHVPH)ngrDt7-bI`luu-3jvv7dd!Ff|OC9Cy+7d>>qD?V0=nr;BO0(S9%Hxh65?5v<1!`*cmphd9poip9as)YNc?NhA_tE=2b)F85W?u1O==Aa{%pG|> z@VL!Qbzk$-YvGW1=zMGQmG5HS@%7D5n%CMVI^V1Rf2!Y!#+NwM8dG}Iv-IY$wL+o) z&NCZ}<+&7k(^_6y^Sc4gpwV%NXBW+L$F{E;f@cXmQ&OOLwddAA&Y)(i&aADjORd2( z8R`R7m(U)^#7VeCeM0*jTb$rC@j0q*=v~(@N1;#oYBVoTOyBa@XxZ`{)EDSu9A}Li z4~$>J%b|7oU+xoMM%=?Gw5oe)Zqui7Hr^Yb&vE%?AWw6DOUL4*s>ksi#85t^e6767 zs@XXvO>V5v)p9vpX@5G|IKn6LM|miJvyNA7#C`Y;@S8l>IP}W1)qeZ&_gts;is7z- z7gM=_&-8(K&m&S7;G^JTo`k>oD>zS{479)V<8If+!*ZQ`BFCW*#DSrKaWDSU3w$Yk zt^Ogeb#3_5`6|EG_VR#S7sw6N6ZDO~qkZ(192?i_6X*7M)#l~f?kV4j-z4vFo!a5P z$(_y3L4nV!UZVYU5v?U(p^5apxP*Jf8k(Js!Xxq#{x?1mYxIZn8W+xOEcmAl&dV30 zjhxqK#tk2gH-t-Solxahat#pM+*|Qe9l$?h$TiSz`qpQTB^DkU;7Tr>+`#%ce2*i=BN!F%`+bhL=~Q(O zxtSQR-@NMA@?_*-;;ilXZQ5a+>l1k;O<=7S&F+~=_rm91jZgPU?Sv2G+TA0$BTa2z z?G>NI6md<=qs!%Hw6l0YdpN%w&AF={;<)CS<2nz%&=>M-bH)8LF7&QTae}dJyc-MUX^KZR|Gk2FV;^(HyfGHtOLMn! zPrAbR;$4}8#)9i`ZpSyrjSKS|ANY(<8U7QZd3@#3eXSd{TQ5@2%JJp2yB-Q`Z4A71T2-hKi}`4&s{q9h;9QwyBkfgZcmm%gcC)>fg@GgHk`> z$?@!9Odw|S_^b^8z8=p`4avGLev^327gUo_o8W2DEA$DksLt;^{6+bmXI<2c9E&z^ zJb#1t&dDPa`>iGOoQ?c~uCb=fcFwDfw1ahXJVnPA(_IU#;rQ0OO$g+X;5emutPLC+ z*iOD_eVyN}pe@`7_k%X8nvM25zH`_5Jo{Ut4f@FWXd&P0NBg*T_u2YC^%?ig=ju7u zi|8x&P5=4McFs-L@CNmfwS|0C*XVm59L-~28qj<9S+2%6R=c4UjSKD2PhK6559{7l zea?BRCRW#~y~Y~ zF^2`>Fdblyo6EcgTw&fgw|$Ij=jR<5%e0I+Peb4q@m2iCUHD8~<89z;94D5EUpNM* zI2R7Uq4);R;Ya*~cf>Vu9U}PmfjI*mk}%A3%&W zo}I&cu}_?&x1Ez_p{uo%?ot=?DyED3_A^K1AdX`{^G-aa?dfdoz9*nR^f#U0`qcRR ztxxPL{>pP`YVqE@G+*_HZD z`pkTG{pLNLOn+;S&&^5g(0-c5b!&?@dUa1+i#hE6@dJzn`cE4j*SU-f`_n$gfcvN| z?w9-N7|vsVzJg=vANQ3m(Kc?B~3vOIUba^LFb4sVirB) zI`pgdx(4gK^ojn%zuM>;^`$XYeNnnm91ugqD{<8^#U^@^&a-YoelLF7PW&=H={n=p zvkKx$<#_Tg@kL)$e;8Yi;~3&HJ)r&8r-(W9ocLty8AHykerElmm?<{#9&Ar{iOKv0 z`i(X$51dvsrfJPuvq>Lr8r1p=F;a}>ThO24q4S$}w4u+mSHIA6j$?jkgMH}Q8+tCetSNcMaO@eDX=xYu;(MbC>?*CAb$f06oIjsoxExm()OM zJKBR*qg!YKV}mE8jl3MX#r<@@&3R)}twA61bezNWYLhWWH_)}}k+{J&`i^#Y|7m6V z!|{1YJTEl_-jemM{5HD6_OxEX5~#1Xt9MpTZ;6T%-jQ2ge}@_A_* zo-{3DeT>>$Jtt1vsDG*7(G7fWJ~#bCzc?q&K`W_|(J)m{Ro|17tF`i>9|$fF{ubO5 z@TngQ9uA%l)FJ5>z6))^18{u$$!mE;@_YUWO{2!D-Cpg>W2UWWEzbbZTDAL;|zQk}v5 zQUl=4xPE=fbD$}VWxOT_!#m~zZsgh0FZ2%YhAyP@ctEtax{Me{+tDB5l32(a5F5+| zzw2Yp;aU73Zi=bmtiDuZp_kQ9>UYZ2($#s@NYq$pSelmBFRzVm5mV&(=90N1?v4u7 zh^Dlsw4VuX3oZ)&5}X;_8mOP^3wZ?HO9zy`6i@k=`o7iznuqSQSRx*ozs@K2(_gON zoTB~cU3$Vj)?db!`Hd?p7q*5-?Vx|25C2{;G*~ql6f^^Qx18H=TB!}-J@H5Wt(NFn z1kb6<^{v-YPw*X1#aVKBz_W67-_`RA_|s?dS}{kPZ0q>yAC3H*BT%!@7HgE$M${L? zLTytM>DT!B_qW7{s%QA#K1&5_1g>rFz&>gcuE(*QWBy>VVCBI2osq%s0`=FQ2I?o; z;QSK;?Zf%FTYtJYxKrGfm-8uR&GXtz2OIXShdQZoQm6iWcH`_$ZF$A~7K(l|M`;S$ zTTY|iWvzj{%e>^Jx}S6?T_u*$BXp8jWUjfN=C?5}zvij%S;P(=f;fuvjTtorbIo|e zRo1XLm$f!{Uz~G~)SIkn(tqmDTLc>iD+DtJ`oSFKJDH>SMr;yO^+DB<#8quIhw%np zP*-wq77xA?d^=bqaG$KxvSv)Z-91ph@*8Z9VZNFhVvQUW57O)U&z$B(h`+dqAH;*Q zzRDUmHFD>dhlyACPX39H)Fyiab6cKA8;YO!g1?7Tai;lYK8Y>j$}6dBiW`pvPXw<9 z*5Fw8qQ)&xptIE@<=*n~H?pn;4_p!47~lo5M*Omm>vMn9v$gN~_Vw)n!6~g%T1y9$ zdMEWhAG{x|*jll5MliTNxIHdVm!`kzDm+IcxaVRpUK`UM)BdFONo)Jyt=_kKFAfe3 zwhxvJCQY3*_1@sIV7}h@djA|O*jli4a`0BLXnWE27Qv`M>{Qz@&j$V5{o6kX)(h4MRtlC6 z{C#jBR`Wc>WBv@7KfDV1Lu{cbcqZbS+JZPMPZ5`l&uy|E(b|ozn_D-p2%Zcc4$ca8 z2$l&J4u%B7gWm_c28#v%X#AtIuF#mpAL^<2&avfd#);U*hY+9FPM+Xw+VLLxQsdtar+Kcnjn~DPhl8Irf7U!8c)sy`gZFIR*M+SMTaWZU()-t7w$^N|tAd5w z3%3^vE^A%Znj^To_wHV=Gq-1M^YG~Ni^Hm)1rG&34(N4lr71qn?|=R*_;2gKoiRCI zYrdA_FBBX$^{}avr%ax5QgBr8S@6l!Pp1B|^~)A-&9zNzOl%y|JfwL}aAa^q@IWxE zXIRgD!QB1k?l)6#UC(trLjwKw$H2868hG6zNXp(kt#Mk1OT{HIYG7kvBH20I2HG(Kot9-I*1LThP{4Gs@R2iCRj8HiQZ0ExTuGr9>Ei=#N)c+xjlHLq$e z*|TKN^1;o`o15bs;~TF}_j+f&COv#X`-1k+AX2UMXn<=r3O?=qw8Ihm^zPI9>%d(0 z`I7A=+tzQ3J0~|zZk*UWvAIRh7Co;7L;DTwH##^wI4BqtoY`|`&s@Px&6}F124@6g zg42S7gFS=a1g8Y!g2{ol9v-xU5UFaNN#^{UYlPKWM+-w#HA~Jq4IU1D9(>UIL5D}@>7A$dlBt(WT|T&B$_-OC3kLLI z=#-&Tei@t<+#h@rd}r!+rv4~cKNt|aH07l!*93P2GfbUf>K}sFf+c#F=-nbXKKLrQ zx^;DHul8Q7t+u=c-{M)`DnFjSryF=Ea`rI+?Pjje9Plm7eZGOZ6Ad*azmHEs)xdHd0L<|ch3zWX7r*|M6DhIJl(0N;on zpf!9(Z>T-dnsk6Xnr@P-m@o2vx<~zlFH-)6xUU|CvwxD`Mx{xv32q9m5AfV|!R^7g z;P1g%!IpvFQ!o#CO0*fCqc>?Dy3m?P@qlL3Z*nO4KAnuW={7zAEhYbVEZ&N-D&LU@ z%W18$YG-Fu8Yf?}5Rb z!F^Njn|gNe=ivC@h`>1cZE!$vQgCJP_u!7;^5Dc^|6pXWaj<$YBv>ohJlHPSBv>w( zHE0Ik3XDtpT@*YKIRD%mW3KW*I;%+j8vqX{-?Z_-{TSTK08^U)#+_QFI3?$yPu2;x4t^UL-}eO8wSu^74eSHK%fXbOf3tt{JHhI~&cV6Cgkb)j z`FoZP77mENSAyq)w}Ux*=IrqJo;`bZG|AOHSNDt!9tiFX{C#n7W#F@qg2nnR)^E$; zcYzpjR&YUZZg6riD%dC3BiK9mO&@mew|l>>gAIcp2fjN!xG$I#%r^aO)2|S08ypkd z7(5lc5IhmwA3PMi7R)}w>@%z$>=^t!7#?gGtQgE2d_4Wf(?1-X9Bdsd7`)c+wSLzJ z;^Zd5e1X_`d2m3mVbDK#zWIDddx(!e3cedG9?Tz1AAH>SxYOr%H16nVykYs>%gNJC z?%Y2ykVhjn@oabnyaPFu`p>%o4@kbi7vS|Fc^(VzM6SbUQ5W(zABe8x4e_#!J!?6Q zL-`Xuesdq5&2QY}9RA(90e{1`UuJFT@G@Zj{|ir|Len&8r4OmInXU2tn3 zuJ66R_iugh{2*w331)m45*|p>4qXIeNL%|EdtHJo-%>bYA2xwq+KDDo^J9@9iB~I`R9}IYl z?jfIwAI8JwTF*dj>5F=@wF~JeRZh?5Yd9XvUSFm3n9K&;68wHMuht>{O50(w)4g3xs z@5wU=JSXc})O%=vFXD;Hz4<73lHMy%Ry~?`EN?nJ*ge3pc=?~Le|B=--CDb~wh4;s zcW>?9`f;#)Fjp{h@a zd;Wq(zBjlcI6XKcps75Q@ntY;b5`kNbA0j@`+F|qSHaJMo%>+y89QqR@@1YCElBSf z3p^=0UA@CtpEcmK<2Q9tUeIZQwFmrCHC=to|E_g~G{1i4uedjSdvTQy%-8XqabZZnp`qzj+uEAHJf$3zvnVd&HWA0R4(pa!gNBzclQ=|D$ z`#&AO>Gt;RoomG3*97MVCj_|swBV}XAHlnUH6`{H3usn3w(EPn{d(tEbj=xo{rFV3 z1dj*4e?QPKV#UP3XKw_!kgfxt^W*qwj|S>HATGTWIEH`xK7PaJ`Tw!_OxwgH`JI@> z>z^Y~Thiy&rtk`RIl}{bnpTucs$-~Ws22qw)3J72|R~nKHL(RFMkXC&0Bdhn7KJ~bE#mhVAFuV z#rN4M7#Zvl$hnRR{u0nr^#ggCT${#QBJkd}YxH5& z=BmwQ1Kyb&c!q!oZw1Z#n&AB>l8`vBeM{I%SJb_o&LbGnr2Yui=77At$ z_&AP_pJ*;I%yVVhjpum${?^V>!9KyrKx`DdRtwZt#7JC4U;E8HUfRonXRzFJzRtD* z&F#IsSKc^RAokk+-hgJ8Q}ce*%J^*R5~l?6)(e6w139Gce1^-#NBO4lW^E!rPp-aD zALh*S+|vc0#5c;Bg|@UrCNJV5I%<+JM3@?KmbXSMb~4N*Og-c-Y* zJ%Jyqw((q`4reWmdIt@I*KoA?BKBD$hI6f-18dUcI&vPoXio9E)s}A!%(Vjp^Ok1i z0h!Z$cwVVFFYdk<6vy}*if?Ex9EPVa4K5CF9S(P$?Ie+cAfc>Rb#&Uj#e_lE>> zI62|BfsSI)4<=8 zgBja1wwDOj4CMW51$5Px0d4jBK;Lg03=8DkY6t2D{@y6qCE$6;{oP0R%)PRo&v_&A zme&IDMXaIC_yA%(&qeI_%&;0h?XCW7TpkQUnd0Vu+I4|a_ z)0wN*A-$8@vH8d6mouvA@yTdEIuRMz4zJ)SH9;J~H^48ri1$yc^9Z&H_y+16!voK} z^0MhPg6`ea6=+4gDQ4Xq;Qa-H83J`!b-?`t99Z1vb^X9`vHW5Aq#;DDT=e>Tvzs)9C5Mh?J-s{1`k> zgS$68&G7*q_B>xwhtB^;L|&%mqc7FParCo+dbf2`YSX-bpQ+`j|MRVRwmeJU(I8_2 zn%`JZN0wVz_oU`*oY9KLit+U8z#0?z0?wC5(A>7OM#MO@{zy(?tja;GiBhkYXVRPe zbkCO=yS5cSj8|h&T;Ww&8zrByuVc`$qXM;FIf?e!PJ9Dv3izpF9Pd{gw2gk0AGmfN ztodNgjP}!E@?E~LV>s7_0YBM#E7v5xX(O+=)=p}->y%@uKRc%NT>4LpHb=$<;k;qmFVAs)ZMD7+FU%WQ-$}pA_s!)+ z0(A}F&D{rnw>-fbVX+w(JQd)Z;u$#v-++fCzpy6EKRE|K3FO;627c1Jfmr`qp#J%E zP{-s&;2g*Hop>#u!FfC;>r>@m)~$>VaH$$D9_D+?9pujRw%-u4){4enGoZI=b88#q zX>xygi~ZHW#eQ{TI>36c)dD*G?e^Q9{Nc-ZRGy*cTh(`R1aW~b5EE#7`aphS4bPoH z)&Kd$cL(CkS%LcalYv^Kyo1Rm-w9X&BywYoTez1$BFDT^(qp%BO4k@0}gNFrTul0Av24B+nkiQuRa!%vI_O_)H`C?B7JTv;!`sEJ-W7t|MdexjU z#y`&b{8s~W!MNp*8s`%NW0glBzH1AwX;L8fb{yN&q|RwApZCCLFea^Cv+v=-IRWp1 z*0l!3TzE7nU2H9hwLo+N?cf;ngKe$t8rQd8o&Gg{=#tVW)~eqf&=T6Ac1MHIEw(#1 z;DgX1-qRuK39hZ4wQzlW61rO7>5s<)eaVM#PW?f%xJKtZE8ttXR{L9nK@XXe;F(R| zxi8lARF17y;rUa3hkAu;ww6>Md0%~?z1oc{^qW57>%14Zm$=CN)rUM#+UcVJZ}CPz z&6n=+>Ym|5{KymKnOkFyGw~(w$e8fncjB2?AYRZ$xLYi;UYTDiZsL74P&t(NDz1y~ zI9;r^c0}GKAN*m!3*9N81^!QKcOGMRRmB0E=`0mcAj;Izlu}ekM609{D#+5Z*jURV zAc6=pK&TohbXpe#ZKs5=3P?kON^FY?)re70f@pBc7HDOsEV8r^OCX>v6)7<9z0vRI z<(-6>_@6yV-^_cz_q+F=<$J#8-22|B`|P8--X0|GBpmrXv^ybvo-iM#zKKgXtZ~s%z!9F*xWxnR^ z%!=Ne+TEk)oN>Ver$zKJ^yzRP91c&!5A}f0ikuU{AD#R9cI4cS;76Xrk^Jm=JnB0U z*Ac(#jVz9s6FxaNG9#jgf(zo6ex4YSOVzaWheNW1*JnkXbJK$`rZrmY&5>0R`(L=M z-h}auqd(17wMJ{kJFV3{opE!X(wPso#+7}C1J30vquSp*)Ys;uzW1GCIGW$_KXcYw z-93WeO^)E|*ml>0VE?};|cl(uF(&byKn$`^oNnfk=oBv|Ev4evG%a->#Adqi>P4_ zi|7rjSM`SV3YsC3Qy4U+~o!|G0=oQ$Xw-#$9es*8;qK}0UT#S8jE@v(Dnw=@QBcgv$ zT*vteJqCSo*Wh;iir|5K3ird4oIl`S{_#KiKRz2N7s{P-oLnlm>5Jh6@-C6<^a5~j zJYUWdKlXLRkbLDeeKy=1$969~!XA>?us`^GL|$x0#v=L(@~7YBMSLGF=u^W1=f3m~ z;fmVP8AW&nYn;DR=fEBqq+Zdpl(+O4omaDOC;zEk)ud_}HJLg|omS%h?&wK%PyB2@628lC)n|6Ud6DBI>h!}R z@|NG_H{-zn`4P{}jO-JU2P%i?`^pXSx*Slrum4f~RJ~MtX!3*Ju$*;TL_V{B>Kvf? znWvujg%NwC_<^;%ul@&)Ag9oaeBn`VS$(OVaJV&qc4q-?AWPMz9=!yti zI-^eabgrKuhsif*NA%(KaLiR6v==N7?H_T*$-JCLa1Zx47r&dI-iEb%&AgoDRF6Lx zaji$=>vgD=aV+(?UY5_V+q=@sR`1*A^LaR2%ev%E{>TS?M!~&^wL7zJ57XHy=f2eF zYHxd_^3%Z)wJv@yHuRy@r|MUGZ0cFHt$(}ZOs@Q;_OE`0+_xa2XCdFNjL5zEWY(ew zuQwonilu!c#WU@}%ZctKN4l5uZ1S+aiTrC{uJW*JtEZxm;&W}T)yI&t1j$B+BZ&sJX0)UMVi~Y$FeGWXqoUF}xv`Zq^DwjJer|*DAIBz1C!z4Wb zd0WkLMdZrJH4$}(+>PI>Q`8&wusyFfQ71U#FUQN>^17T}JptF?{puYUXf61H{e$8P zcmvGC8Q`e8$G&c1Dt_SJ#!#E6L;S9eQ^UYVHI{L43->GR+=#=Z{GBAZp0q5I#*4mKBEs<>fPCaGqYCgJAkLi7@rS(qmC*QT>9LgaP zy>B(0T2|g-CwkMn=2v=0#q;!#RG;`kA84OPE^n9-QaF&{rVnL zx?w-_R13?K#uU5O>fUUj2DD~=qrP+>{T;b>AVRm_jM&fDD^i=QKdr@l+=Jcp%ABFn z7x0`~n0}4zTJN*hWZv%Y43zm-FT!hlp5MBMz8jHu`GkC}zaS1_j?YV~$;AiT;Jx;= z=-d5$ho8?#8qYp8JF1Ue>pE+pD{E1Mvkx7(PQ2mG;*(GCIX+BRaDmO#zj}%O(WTfD z8|Kem{F~2-bN(fE#M$>E{D!~!oqfz*T-&#^Pr?p9ui$eU#OFupk1g2M{Mp1gR^L~P z6L}3Th*hz{N7+aGi3Q`cKRhshePuXHU?>cO@$d|{(HC$JzHh$HdeE79`7A5nw&(90 zxp|ut{8Cehtp_63?LOj<-l~rvkBEEzqKDu!&hnN#peHRK$ye%lxzRZ$oXOwV@w&XG zzM2*B8F=}p^t&XVEi_;ADILQD^AZ1WTfE5?@Y!CET&e$E`^I|X`V?}L{<$8%JX-x} zy?(uAIIqv{^H5)o$S0>qIZ|0^9> zzj_!qp$D8mpWA2J%*Vd9Jpde3ZdAWIqk{A5*}1=UA0KhOwOOwmY(GQKUCwtuXT#aT z{Pk?`PxHWa@KZf^oKu}{PESS5iQVW*Zg(%7L_Rkk`BvVgbGqkeJ{Kw{x`uA;n-!nM z(=as?PTXxu;x@Cs-eGPu#&6YIV8SbG!bRxRx*IN%A2A!}K_Jk2vqe zmbi*NCF`s{1b?(2q<8f~#C{PSIv>cV@X)(%FCWmqS7sz2=idmlV8 zXX_FVKj;X5;zRaU*wwuF9iB$V-UA=yGPzi;)wd<`20Sz7k)6+6!**xnozu~mR)4AA z^^f5vjDbz^tX{Cbo*2b1>4T5p!uDu9$M3{|eHJn4bAn^Km5cE#CEO^*-kf*@8O|-qsd74d!C$y+p7dib=A}pNHF~uUugR0<2!nAZ0vF*ZjCO8{?u{dN z6a9Ssd&S`9)WG&!*kn?^bK4#cj$@wgVXfwktKm0r?WqWx%j;_+dNCmp-LJt}!wEe*r*+v42vh~ME2{K1o5=X$v3 zy|@qml#k3GSCBi{%i0JVnX~*Q7t6!eCl|}N{1d;xS6w6Dc~0(jANdQn5d-qDf4H~L ze!>W~0ZdZg=*P(g@`{=YkN4eT>LI;B=RolhvCXFZPH&z~@DStRJn+40hpqWsvNM(P zj_1T^>3|=y9bI~ld(#t~qATyEAH4?n4aekZ_-*f4?t$U3NsTC9nYS8AKSG~F|Gv&J z;LyghkLjLruyOD!DE_!OT}Gulsb#rz8i>>X~8y`mzGyGG2mJjrf|JwEv%`7zw13ae_Rjefe0QdRp4HK8?Rjg)5A+Ri z5q&A=&GoME1%99(CSUOpcEsWB1;S!=sr*9JIaL?Z7hU0`;*wt3)L8B(_tA~`@oPkI3 zE$phf(>d;JOmV0OP#g$1;Pd#M9)xxIED9aNA3l$Ziz~P(=J`0TSh>qOVFcgem&SvI z<|4M$VPYRga4&PO7=_V%4ey3MFcOA2yDdliY?qqVzA(}Mm#5?(xlvBzKWZD82Ft_; z-KZxDqng+IPnnL{)H3#8m^n0yi;5sZov(HCx@7a+DvYdPjGBLqqk8#95pwd%--^;pN%Dk z@iZ7Fcfm1wf@ADuZFq!y1~;71!Smr1JXRyXQge~#C9#I?KtAH^{ohaI@Fy2^Zf&V=6R2S3sWG+&&bPUSN5lrw5C zUEaYv(#}CClJ-`SU&d1=W_sHqS@SU;x z#d0-WTBr4RPX4rS3qRRi4a^5%sD8V=3ftg1-O;5!gZblG{My{)cJp;cdSgUiT3uf} zfp4jw`MmMz#Twu&o5NV^b8oMU3wl)3v2F1P>mu-w9`PsVz&s}>=$q4}8W>+OzPg*O z@IdyWQ$28J*XT%oG!I-A?-PHpf?nAX#`-58!B|+~TCo6w3*XE4*vNd?LT!w5!6mhi zc=kRq!zXcJ!cWbg4=jnOSK+^00GIJ9I*|{|Lk|Y8V|U~8eQ_gynLFLmt6rC0UGaU` zXf5~vUE?g~fjlQYz$0}*)xhN}6@3wMJ>xHUfE5r5Jfac?}2o!|@1V*_{&-`Iiwz&v%o-kyGd z{eOAIz4adWs@g}LEQivo*tT9=#koSC6;cz61Dw^Knz`UGxEXtkQ@jlxxz_%!I#`XX z7U7?2ChJ!-TPuvGclu>#J|jQKo7Tn#xD9>tUp#>>@l5x^4RHqRa)wb3!fW;YtWEzw z4em2adJD#YQ<9+?`L40M+UNgf=CcxX$Da13_!GYpPq0T^>aXF(xT^ji9CN+cmdE5a zy3ixQYs|;zqncyQv2p(%*FLAGpMd|kzIZxrU_LP2*f0Y(kn8Yk{)A)UR@Nit_!R!n zhxCd0Hy$J2#V5R$Ct({Lkq51b|Ek$uh`<@VgAY!L(62ZU=X{R;(K}ya1AYf*VI{xt zoEirwhIPe(?fJoe^@6-cf96{;0Z(8b9Ku^*pSc+m_ZLTeTa3aRc7%I;oS1_eS>Mh+ zr+N)0!c=i#UG|*KVKBl+*+dTsmdoXE2j6#Y7Js7yI&!b!2pzDC@rt*xx!2?lK47oj z8s!;!H9pR+UtoRS4};+k{=@&+4p!rU&O7^lPkG)v;jkE{U-py3VLm*^?d(yxrs_9q z^%}k?_F$I!SWXs0cr2fxcl<={3Cs8pUZ&?kzt+On*nBRTuV&hM7VwNW^)!(>S%dJ?5Q6JTZwD_Vo!kGU;)0bXJ(%Iv@nl; z*_p1{h#l!n%&JM*0)LmQ%}0KLH?WHBtY1EYGdPrWT9cZCf9Q*eb@^H>tFb(fkI1WH zl}{Cq)R)ua&_5^U2}jt1ec?YI#kTxbp3oEZSunUvkLHKR>N&H!+<~L0)!;P$g~9ef zU?46mXR7tBMco7+U{RdefC}4li%Yf?2K2{=WSt+9uyAgK5$T- z?cBMVUA=}Yi8VfLPeuQR&%hoy4U2JTy&~A-JPo^9qu1pi&%+S<;lIYp@vruoFtG-! zTn{(o5&D4D>PDOiCeXilQ@hxclWXusdX<;p7+)8ka1&PZ9oQ|Ve3s2v>@9Z14PEei zm|(7Qt(s39z&gH6*K(4&MJI13^iXY(rFid=id;Czms}cC6`@mJc0+-;jGyBdJ z-~u=+e6w!2!1hEwQ4_!xk$*jJ{(gs1UY9puhB=5)bE;fzeey9ZsLziK zgnPQ)YhoSG6SFW<&%pi-d{+a)^okejF;}|qvse;;{KH(tg?xqki(g~FPda2T_zJIJ zAUoq*;vZjhEj-2Z_#E4K6jLyVkHA)30q>+=*oo^BcIUTVr&}?nSHMPWg}-{uTExAc zzw6~2HLMyO_b?B8Brw_g`KegKkJ$qU&L-ZKE)pJhu&eQJg#4$KRq+TPvM!|E`Qly$1QLiJq_dPkvfOwoV;GMd*I3P z3LVOM;u@yI{eR~3C~zANs8`fLVxP~zRqxZQlK1cyxW?a%CFXo)dbgbK)5qqM_=k8f zmOR1N;TYTN`M_=Z;XSF5_`)_BKjO~4Q3I)vk~ky z9{*5_RcP~EWdOm^+^8qmq55yjxAhzTSTtRMUfO_SqVT5&vSJj*aJixS(3WXXy00@z5m^9D;v~A0n>QYV|oidVq7dEFUl@ z`9Ln`V|+lJ!#ChK-ST()&3qb9^4)gEGC%s&Po`^+)-MLw7Cx#;ha>u2a9mGP9MHdh zpS@7{Z;(|GLYo@xc>R$hqMw^GmOiQx8f zDtuy3I7UDEuHq10(-}XK1Nk{lC04~8PJ`RY3u41__^5jtmp{=PoAFb5+-H(-AD=0) z4`@FX{<@Yfa5H;7&d}+{!Cmi@GhHK3hygL^XL(BPPHLpyD2uz#(gJfANO{ z!2&r+yvRd*7>DKG;tJ=ne_~DS0Ke%>UgH1!&d+y7>&-p`K+^Yp2L~pIK9g6{8L=>QFC@bn2k5VbU6rCI=}66Saho161Q+rjN=wCmoGU( zq2F;+G>Koj} z_v|?1pw1PiYFm7m&*~MagYXji;V1k>j>dcOWW7dQ5RZi$@;_g79c3%9-(Z*DaUy#X_W#tU;!f=?fB5{1Si(W&d18$)$$jMn`NF-$owf2=HJiFe zoal3{h}363%?am|f8ZJa)PJbHK0ko-)+koQgxn`~;5hu?yUwuHcaX|Aa;j_Kk9*M% z%;JM=i*Mk!e3{+(f_OD|KEY4;gt`qEQy0s#@(5k1Y2BaGVdq5p}og~~8e`I`QO}@`vyu%8)8NTXwzypr?G(S6v0-ONk>1!ll&V~H(&G5g*)hTbB)FmG$+r()b3 zjVGV+KWn0MJ}W-t1Q>uj@liR$8pRXrvUYeNo?#VVz%|_y2H*tNYwhv{{>oqBr+!lP z6WEph#fV&1Jy%#(y*cBVHx58we)mWZFxdU@eVmc~s&?cHFp@vuVEkIX71Qt*M$@nL z;fZR0^{IK9GyCAW9{FH#M!KSJeA<4BTt;tlDeT0@*c4vKNyd>E?BCKU+o~DNi{AWP zIPE-|9)~{Za74cMewf78aKPN@T0E*Dd=>*AwBM(XWe>u7=|n9kZ(BG1jDvWuSh6NL z5cZfKpAv^So_NJ8*qN=&!!_z7`#kCk^%{H2vGNl;;K||=zk&JkGT(u(Vn|#O?<>A3 zFYrxzP|wmMf8d+&AD4%T*1#X}wOTtr!58V9zrZcIT^)`S!VvfKK01V-eArsV8hjRS z@Dm=IFKn?s_oQEI!kb|}+=8k46Z9sx&?jHyGak)PeBnX-QGUY@@DN;>Px1@%7k@BXYR&2-I>vjQ=XJId z2eKCkv()8m!lvwt#~4#RV_bCuovO_|&z@}OzUnOe)%k#x5jDJZ*wbWFHMV{!4uIdn z2Use;;FeewpY+PkbS!?v6v6vpi=4-X_$wULpVKdf=lUj9Pq3>zfS0J1;3S;D8N@YP zi8pl`ys$P{!Efo^JX~A4GzYm=9>jC1M&Kj*H08VE8t(B)F^_8$9J-oHLHt zFqVAi{pPJ-V836lAFt70l8eQ#Yph?q<3DPB>u^4leXN6@@^2g)PoWQZNw@q$O{cz; zvvFXb-F6n)c{bzff2wQYw0M9Y^lNQ6Abwc2hrA-!vw=sp^SXJ7VOUmi$7WRzsT;+H zxWi|h4N$A84Xh8g(TzCcC+@>XUjy{lvGqIoBfi5f#;YCyKZ7eULw?ooEb*K%VKDrtJ}Dk0?&KHt5tsZ9cBpIk2yEjs z&ScSn+{res#UJn!+*vNgyZEM>+kS)o3w|V4)B@^6_HYd!<-a%(-eWJ!`F-ch@L7C< zUc@b5!8fcOr{GWY#`laR_t7su6vMck7{mqTK3pD-iy8P<^&3vY*V!B{tMB-Ye8)H0 z+3R{iVqHAIeesS@AD>{y+Ixay zat8io4sZ(ZgrWE^d%y>};y+>;|G`VxkX_AB-WD4$4gbdh?DfjudQJF_dBQe)Rm~;7 za2!79oJM_DfWD%Bmc1N3pu+aSe7COW)WUqC_IlN#Vu3I4OLk`yoRf{6fz&f29X= z!av{wALIx8ksqpQ;5HnGX{?nmS% z@g^_81Xw`-_9L9-(gW5Hh85-`KC3pMBlm)*^du+94KP<;$BWBG;s8F|m(fSko5hE4 zF#5v(_$d8~Cp?j!;4l7)f14X0)L-H|aK+r^4K{Q=yIQ-L69@La^vHZJ+PO=dOdry7 z_BQzqT)~}SsJZhAy#ul4^RzHa?-6dPeLOF3&=C$Qj^Jy}LC)g$;vH7to3KEv@D=fB zjxZI@@hg~XjrhO3gjd0D{RWR}AkWzYH>SQL5s!R;#(c@xGEKf@|>_UJkJ)tJ@;Tg9k#)ju{hJ|AUcI<9_;c!3%0 zz!&U=u8jDZ?&uH?fp6l+7=d*LoWL9Tz{Ws z3prnogE8jfXZnJ9<|oh4vvsfo>`;SSCtsu!;}KXQUagx=;D9ka$^|fiUSUPWDjzi- zJ>%)(0mjHv{0eUJ1Nfm=V10bjdhrd<<1p+mZ;DU)$0_7her8Q#-g?w*o-;3c)_bJ; zs*TKxA9zjdxR(Cu4o87Mex?iSl-u~d=V7co?`QZ)&vZwJ^01hwewf@TudC1Ok>FS2 z1%4Wn59=4xBR%kc_s3(c-QGAql()q#TsJRsrDriB_jnW|#Lw*RdH1JpI97RwZDF)n zrgOd`FNsaL!0+PAy^P_#cq?p#8*mJt=L2}5x#Ab>1@HMS?B|bc1wYt^56G<^@i?*K zk>J4c3eHEL^0GOoxr$?Wo}Uyq@;iT}3v+}qaFSlxl#S&Jc5z?2@W?O40AJz%RomlD zFhMRhA2?=?=7m?eKi`laD(2{s&5TP|#Ls+&k9wU>YtPG?*qPtE&YW;dbAkh&Y0O+Jc_(HoOMU_8hwy+qmXTU&euvxFY}K`!GQbB38X#!WZ!=>*C|a zs@lcae4eiO8SHfr^(xoK9h(*z+9T z#s}C(OtJw^>7IO%{>pFR0v*94@x*88mOsEi>w?+(q@_=)$ppL;$6MEsl)?R#;yz=n>cBFH0 zEaw>$2C_BZE-W@b`Ve2{10!AM8gqt&HGlWCX8Z#m5KnMVY|t0|I6oja;g9l#o{Rnm z{n4iwat-k_TUO41-|WcW_#WS&6E@;o<_I^$zPVbT`N9=*_B%V^J@`1meLN5E=+phg ziGTFO$M_Xq2RF?fp6auRVY%G;_y(K93OEQ8#7WJ~{N!2jz_;1Z{Mpg(#nI@?&&KtL z+Zfxr{qFwOsfJ}UYqMrJAy(x^H3;Ej)@^LIVkh>3ch<`1{my1yH&>j-ysg0_d|_)m z0ItG>!hX6Xd=)OwjJS>;d!K)BjEEm`!577D;h`90YjFxc#Hu(Izv7ghrI@pp=niK5g=XzN5#xCr@ulT4rh!r;VUOM;4z7;q6 z#P&z+N#Y6mPI}5}OTBurP9Mgh3+p#0e#@qEjF@0oy5M8t7GGi;amUVl6JF}4)PBlf zr_Uqqo`YTf!DIf;uJEPehOh7|`Ij%j0DcVP*$9th6P!XE zz*74592=>{^pDi+Y$Q%_Q86qZu}g6n`rvQ$!-sK=>dWvYn1plaHPNHF@hj`48-B~4 za-Do@J|5klkFl3^8pB%Qq_N~|dC%J5n%JNNuhEBV#fjKpZ+O6O*-qXu9(~X!pJ6W; z%Xa))T#Io&Wv<4+ZQPfh#R@(p=dr7C+1tIm*EnTAxCdY4F)_hUjLo*}s^81T=45W( zZ@$*1hH-xw;Afb~*9$AmjSkI~jm3$%dR@MvC%U0)Ig4| z_@LKdg?<@PFMFO|@h7gujUu>&IQ( zM}BuLJC|SZ1^bkGBKo@W4?nTbA^u>dSQG1fVQPdf&5!P3jrps+@I#NaCN_aRbjv5` zRR6?h0_<1o{noia-(%ymhWf?6qfhTx4;TKcCHWAJCqKY`*a(~G%6y2Q&4*s;8;373 z4}JsF%#Xgs7oUJ5@JBsJ$9M^>qi1nRcdjYUC)Vhcp5d8X34e(D!a!?h@A7wc<#%vX zys8KIpqwI~n+u)Nu{pWcb$B?t625Dl@Xp$64BU=B|64Z|Nf_#pbZw^NIuAXktoom%aFjN4ba2 zt(gt*J33`EIOiI^kNfK3=pBhSKZ^~qV?AQU&*I8F;VK{EGwv;}`3QR(v;34FvK>F6 zGuY)d>w$rVjpoHh=ICDU&UN@g@%_r>avwhz2VzqW(nE%wVi8|7ZrR?v#Io^?>mUCz zmfX*0;fOe3bGfH{MGU|hI)=H{%|7&wW6}p*sm<+M$VF^wFONR?2A{@{=m0jFv$?Z{ zNAsiuv29L#s(6FAFg{)L1@odS>mcIMJ**#Yhz0pv{?=2&_wfQTA#ckwbmMxlBu?Zy z97(-IuP}uU#hH2ow#k9GgnPTM9=@|3_FkPghIJnOUf((C^GxQZ@@n1Phq0H z0R3jY=jz?#!!>WO*(-pN_7B|K`6c7n3--BM=UIIphvMJt%ucKI7^9*tYpDs{9#w8%#ylRAGtd9-V^>&3{ukzYpqZKaDMOC!UP zC6V(Y3nTL)Cq!mPW<>Uk>>TNdyt&PrZ9B8Q@7=NAW*=X^fHCp|d%`bujJTvLXCwGE zj^X=Ve233#`CC7aMtm;ohRAi16_H;@Mk0?#{ubF1@y~bolzrH@KjORY#F71NXKH=E zz<161Ogp=?t?%RE6T3ycuQAqW|NXwZ@_jqEM^;B3j64>(KVtmvM;1oD8krwCKXPeg zS>)QtPa?mHnCFv`%@Kbm&)+XJ4;%|`f=}{1?35eTit?{q>+?!t#djZc<=>8x$&rso z_{R~E6C(p1Il2AUzD{oc&6U&hH@;4bc>VAQ8~Mzdd)gnBD`2TJ2F|{|GxkpV@8TSl zzpK-1HCwkt&Wju!**mgx|Q z>5=0iiy~J?ZjZbk*=KB@_TLJ+FWxjsABR54hMEu?96C!+PV?^G8gjFsgbRXt&PVcw?wXvoE@1H`DkRvNNcn;x+U^z zRZBF{wLitN

OIFl^e-*-U#6UF^a7|y%88v1zL5XpGx<>dPYyM5CBD}=<(K(v@q+X1yz_PX;Q2Ow zO)k%m@n5ufxxP4uhEJDsT@Bww?-QGf|6GS+y`j&E*~D{VOL4QfbNH<*`VA-=1ie&F zMVDZM*p=b4hE{5vu&u@o`)mvuOMKI?BkTv;Z|w06{06&VywYLhAG8_T4!wr{M4wVi z8TrttQ^}v?F{4f;Hv^L*8 z^fucfN0h7CHrgA^+4;YiUPev^`zY_5*6rkK(>bDJLc<&N#G!RLZt~W}bvk+%9OI#} z4IR$oc+I0_EIV{JxuNx0H$9LxC~uUj+9un!_<3<$93wgFI5+wx9ap)qb2H8f?Ngpg z50%e?K05yPIJDCLq?ZoubaBn}-)H^*``B!xk3;L;Z;M#TU_$I!`HST(M?HdRDt1%+>;8XZ!{zT5pm&kYd5qYlM%0KxSA0lTH zf6A4H)@Q8CU*st4fS6XSJ9NIqdt2BS@x6GP?V;JQM|{TUZPLs1nV8n?)N1$%`j-9x z_L5&@P_QfviN7gSmtA9(yed2zxVyqPEMytmoLcQy;@q|!YdP+Ps){AF^-VAlG zO=8p7JhlimuxH0}L#;~x#*5>np$B=V*ky8`4>d2n9eafPO}{?g5O0h(O?dOHH_zM; zL_hd6Ubl>8^-KG_^;n6)X#jk+SxW?+idFwVSgSUkBWJr=I40aE0zrXDC&j$ z7oWwa^0SVw+he(xw z!@9<{`0i1|;$zgp_-FoF>?HRw4&^y=9Qm18Yw_I1*qzR8Xboyla$UKiywrFf8mt&_ zI=`WB&>-o3v=26e7D88)+YH|%{uGPyOY9B3McgWerQL{alW&L22&1c~+ULNc> zd(O_kGxiVjo(_pm#}RRC=$-pod^^lFJ~MtE=Z79aec|RoT^{B@ndf>#nDcyF+!cBV zHP9|mVcfqtF?TV%Haq;YoaKE^FQ0Y&J+6%_;?nqC{5sAJ+w}eTM%dnC;?OYH)_(D8 zuMNJDpW;Iu6TWwiaLksS`t446oZ~DHbnN9o&WAiltx3!v@1#>YpL96-nse^_%X4YI zc7)a~N0v{Bcjzl}2D!rUC9Yf7vG~cgZVVW!>;ap^RjbNXQQ)ADV#}a&VeY4XMc+T}wWF6&PowR%-L~I7PxQ9x2ehAhKkT>t*HfYvC~gsBOve%8 zhyTQgF{a}npB2|wkGkThHJS_Jd-R$3&N2T?4~lKkkK(x7HvK7Z4o1Jqc+LBxcg6fQ zy(@ZJ^~3m%F(=nHIM24*coXZm4qOi-PclB`e8wtYJ;szVJjT?4`A+$%>rbxFf4M$g ztH!@vY4|PUz!>4P=(p?zotE}MhoDK!4>=RW9~L9~6R#VNrD>K#8Q>=!@svaru`9LIsb zT`u@xnjHUZ{*(O0aiy8bEA=SQznlm8iI~7SlE=zN#1e8|c?h2&2XVf|DXs%{lI>&L zTtjR%yXBg54YKjBPuD8DWX$Nra$Osv`nJdTFh*d^c-=U%oNJ#wlNazO)+df<2iZhE z$#u;~(MH%+zJcwfE3@6~yc&i60va-(C&scbVl28Ef2htqVysbOs(Xv^oDX_7UE4Md zt%|k;RIhTDJInKyuJdW2q(iqv-#q~A$-t_*9kH)^!&TN}~vrqQJb!}`6-@?EA zH!L%qFOPiMdK?SK%ew#D{u?*W%h>m6KVkozQ|Hz><|m9Pe!{pizFiZpjm7gD*N^LH z)D1_?kiT%fscq3K)Vkz3uE#M)Nh+|89vwnEUy=wkFDeueI(4|}^1 zx3jx@$9qDa^6`+PoEY*G@&8ZbSK;1D@&q}9{DIiSx3Kps#Tv0@JR#NzKIR$0&uko< z1fSv2bFaH6lI1M#bB~VI;=ys>kiXnH>&{vK2)Pjd_KTSIBj(H=5Og)Zl72_y<3sq> zrGth@Q*^BOV|5obEcv0H0osIBu}y5L^XXi(zv5?gN;aB35$78}u1WcW z>sh?UH;9ksg?Lm9DjpOkE^5w-xrgS2{4@R$*TvuBuW@a7^xW&-x18ng49ziOOmU?+ z*Zf8CFrUNM*akMr7&g9*2iqzal5fzD+46UX3xa#DW4x zxps}%1?8CW-TVE8!OHWzdmp@?`}{wjJ?5Hy%`)1Y>)n3tBeXv8b>l++;TJYe?m{>} z>MYTnlzo=Sjw6%Z{cxtpN9=40*yx3&xvL@XKC^P)#9a+~R{q9!ox2y#sGLu^Q{k>S z-Aq62Cw>Hb%$XDaw{um#Mtjx%_0H@n{tWxy9Fhy{6=Q^dm`nVW9*XaoQ~U^j!n??S zV@y}3-_molOX$HzyTts{@qNcc-}lWj2#Yg!(&b|ZxF4P^puVF%O_#R5=jdw@hU3p_vB zO6JfR`P|4Gb}*kBdxRd>{YNqe|6q&I=lCng7J4YXhE0-eaks+Wbf!bLxGTV)=G>I6 za=b6#o{+sur=nkFx6sY_Q0!S}|Mrx9#z%(#u~*0x{usQ``L*+S@`N9ZjdF*g^IY>Iaouua${?2iu{e8k{m2A?qa)WK&C{5_sW4<0-44J+#Tk%LbkeA3|K1|K@&vORK{N{kX`O|?PndNWI z*k}9`_#OUZU)T@!0oxo8Wrx$ft#j+!dZxRRFZgltWz?T5cVL|ru@}e|Yt&s8<3tC_ zx7!^R_xH$Y_xY^fEnCQtsRZ@3Zx2d|JMMkE_T0sUP3pa>1ES{!BIkTOoTwf1R=9hr#cT zXA9jKaJB%Ry}R=V`STC*j~WB^ul24i6qQRdH{%G)f1NzwiJNVUsJ@qStXAgdH z!5$I^zdW#?ZW{dhfDT80Q>VJM!%wJfK0$pW&&VV4-h0q9%%M3qH|#rpRAWY$v!3|6 z=%V%;yM;|`4YE->hao??9e(=ah_j@vUBbMaJOoe|x}AdGi4o!v-M-_`mKq z;QQkD`p^M+^N_*A24u&-7(8Nt2G7NjO%EC{_Qn;j~|>qN3M}?>=t~9+_Tp43G$H5;#`FOZf)ba>=nG2zQb1I zW3m?AedPap&w;xN_&!^Ny+Iy3XL5(G@44NtqXV(6eG~Qg!3_((=O)jcA4+#m!dm8g z^i7)YrTBg6QtqmolTRD4E7>(<4;e=0*c>Ob z^poK=wj9{`=c`_f1{#vC;A!v$$F?!h&4XMxURI-G==H1Mh zyBYjs#>ASz5BQp`Cw?!sGWkJn(B1fA$q9R|@2JRyzJtPN_-@IIARn1RwkV4afx5^Z za)>R{-F9JZC6|uAGct&M!^UUp`{sxJ@9t5y4L(TUw)Y-8u($1B_ndxU;2g#ow0lu_ zsKg-g|&LYaye^l6M%q)!-fj za*W@N|BVgx7q|M0D|ZFRz-JEh>l+4S3)_Vp&2}MU?>cbyWDM9Z{%$Y(g>Mm$bFPA4 z;$hBJ=*zxq!DF2l;8D&O%(Hp5R@gCQ3_Aut!iVfP`<|bH&PKM-aqJs<4}FaOMXu11 z><4n>e+>Tn;3o$^KJa%|esu7|gJ%u??~A{^A`d;xp0SVUYGjRlX5ZNl>=yfgPdc3n zzoK{2N%5H{4Cw82C_I)r?Vy4EcpEWF22{wc44>h z<+4-qf4S>meCbDI2Dm%PE|=yzru<~)z`EvN^Y=~EuO01?`_@1GOP|aSpV(^;+|Sl; zx`6#kme2**Wn>7Qk)2|`$PoO~obhAOGwBA#h`hk#@ee*H=kocWy7Pv|IO}p|Y<}tC z##}!6&3k8Nohzg4Nj@J`IsBE*ahCz=nLaPgJX$xM@w09r9?)B(_f=FW4Y_$!rpSX?GjMk!QxxJeoIi zXztrT%-tFEGUgKPy9dEBV z{{VY|-OsMDpM=GI(!ic^|5zFI*ca;U+q@{f3lHPRrYG9J^f2GafIU>4eNA7}A9jpy zVd!`CJo+B}o1Vzu^NfM-`F~;Xs{{PBzYE8%dA-4126wx-r+m-(J8|xiv2)&k@Q?vN z+Gh^__2Bac&iS2@^QSrM|CT}aj=N&`JYSmsi(<^DD`zp}%WDl@cko7ow;7Nf&hFVV z?3Rxm@SzE_S@^`*EFUl+Z^$8XhF|QB2V{(T{j^24k#o+5*e>oOe9pj~H+SIPb>Qsf zbq236AZOisAX~^`dN$c&Jk77U{D1)-YCTvN{`BiI-mVUccaUYDb^-k zhu%Y;(1q}RV6)J@=t^u6KChn`kSFL*4Sss?-v`!e{5 zeL%l4FGeJ6zF=e+rY2ZG$^DX{de>cKiyJrpDb@V+_f2-XcNcSjwljI)$jRR@j4W|eA z8x!v6-7wI`za7{E_8eJe4?c9DpZd$@vTyD_Ab;$2J_Ei^x&qyhK1k2w)23sRMdp?- zobQZX(|3Z#&oiA=8r>qB$2xF+L(j*1$Sr=T^!#`<-ju%$ucGhc?f5M|OD`rL$SU)S z=UGQ&3_eP?$5ZM0cr89FA0Le~0_DjPd>cQv9>rN3c)T|9vt>82vGjp#AY0i)WG$Hs z{A+9`azhz)tCy^`*0rq<>~pe!9L1+maoHw(U%?p@o{sn9_xL=Wi~T@XqIhdp)hU4w5QeAD0?2VZ{y;cp&%$KZPg|9S93gJ%xxKXvkz zX-m8O*Y=xzOJ=XNq!;O!?T_C!V^UQm! zZ(aHxpImUym>%U!o_#^rqf-G}y|BGY7gQ&^-S0(z{h}*2iuF$B{o%q3u6WU2a3!8E zjXbnJ*!_5(Hpw9VX#5ag#2?*lq8s`fEc_PsJimi)1o?sN)9)L&7exj+yK%=1-*v~# z`NS&^?l5p4fRF9X2Y+|)E`xtOxc|U?fJY2IWx&@)_ONlB<2-%f4#4vU&PT|b*BIP+ z@MZ&d%Gfc^^Y1fwj{&=dJYmo9h5f?;+2YQbdjRa2cN~y2&VAJ9yq_K8j)5~6GLCPL z+|w8L6Wmq!=mA;ddmZniY=GkrP}dxOnlZ~Vx>8vowGe;Qc(ApD&Jd&OS(fx$Bd z_L2S7K491X<-p%JwO{RDdMq7`{es72yWn@u^!Y>ZN_>>8#9!%M)}OUcp7JqS)8q|( zh>qkekX>P&(V@r}elfhBENN}gZ+y?n|L6_{8=6dar^0u*?&n!=e>O1obRY14M~HLZ z(Kov8fGg+QUv-g3>~i0Oku!X1=7|5yd5->(i*&X(86;EKWA-*#!Wa0A0lUrp9kvWP zY+T3&x&j%1zw#;KxBH$KJ{tUA`O@R@V&hKU=O>He(Re&LKtIN(@hrBF^+H!b z0sm$znM?VssqQ-Qee$)bi+tj1(+1mFAGLq%<5dgeCFdpb(TmifUOKyW*;i~N>y~X` z?OJ1atu>7A@=4+Sd`|ejvqHQ+ed)*Me_vpO2Wy=jP7YYhaO2?VgQs0U7SWIl|Cxnj4_ z*{lsV@8bqf7|=WTnC<@`9=JP_EMY(6>3BR||MCNT-#G~WVQrEJ?sMViK*|eg|Fao&mhhk^*0%uBhq`wrM`1z%aHqQ7`BnM z&hF%I>uiP|z`mz%@o&ECfRBJ4NH4H|y8~cf(bsM>Xm8q+_MvfLuY#~WDb0L`_ONt1 zDSdGNoi6;P`_IoCKRaKY@pk9O*$;k{tz>T8OC|HzM&`?%BK(Ve^uz%h)ES96rN;n2 z|CbEdP3Ak>i5@|p#uGae$@ZXcuxH3E=OAPVTY^1oov|af?0DYzVSiyjZjcQ>IH32i z1K0(RAAHr|-!FXS{69G2;HwwD=E`4i|C%d*YhAe~T>M`{-!T7|!;=S38KBu5Pn`dk zNFI6NYX^nD?&AM(d(!-0LCUaIzIE^)2jomLhO8pb$Qg2vU%(xi=M0?Xvys?KWG24= ze*;^QtzxgSSIFCZ&Bl+PkzD3qG`_}}uaN)H*gK~(7tXCZ!!kcd$m4uZetIcech4KS zMn53`*-G>Y_NP6+sXhU_Qa=2EUSrPbM|c98+M2PRtT%ck9|k*~T;f|MW9XB7H}*OI z8$FV2q09N(l=d~-1fL{V>}R~Qvlsl7T)|u2sd~r3K70Ad`8FOuW|KI}*fxo8N_;EH z59a$w-$gnLdCkGyF8(i|e=ukKY>^Kbe8_-&c0iI`UI#=O0aX0Ke z2K*-OhtU`C9J&DCJw8H@CpYi}K2>*-|MS4Ux7X<})(#)EwZ;Ezy_s8cZ~t14_AuGP zCl1yr9rfM=x+ot5y-`_pxT8ci;TiNJchr7pz{a)~oQ*h>g5jo(jxauVMqN7O^6sFQwWz44i82Q@zAjx!nQKnIiG_$f=bQjamE zZ|Rr%+5Xx>#nm3)06U4SR=0N8PUgW656AD>?|8U*!Nk2S>~XAJ12 z#>ZIdgL%_8eKwaj4V-zA_3ocOePHkN!MT6I&a@WrAiRs8$~t+_fW6|LS@s5h6rULz z)_DRSEIoyN@hgL$9qt{@*sw z4;(yn@bJOM3_fx2S%ZHw_~HSXLPoH|$s_Hux!E~vhrbx`iLn{TwsfX@&%Wrq15d>t zy)QeGJ&E_)bM^?{%;t3N`K1^47VJy%*BwBz*garyMnXQ{aPfU2JjIIpIUyUXSK zcQyilkb6^qYw$XQyA0f?dW(Vc4d0L8+5AcGK43e%@8AP3?hHTt)(^jO|JQxtM-M)4 z@TG&V9DLp28wcMy_|C!i4}NfPfAy z=OfPFy$hM;@4B!b_}%>YOq_-IyI<}Ok(cyid)S_}SMe@;lx~f0;3J#Q&>I7JER6RA zc{d+QFHKLyo7l4QEARdW`NC(!Z;B6~p6UAZQu?X9_?-IjIQ8n^d}ey(`wKi755{Nk zR6G_xV-MhE_!7H-zOW&VFHRmZfh;f%csG8fUb;738sF6yIS&WXs@XAbaQGM4Vo|G>{>9_%e+Z`|o%)<-@Fa**62 zzmk3Aih1CBFrLZ1qwFIml_T@WJM(E?jkPv^Z9v|W3(jo#NAR9xf^%PEV2s$Yd}jD1 z9TTshcksdBtN0Awg4cHTOg69~oIyATa2`P4{)GWK@n6>e-Sf;jJ0LUoSowsUDUcs@ zcs?Wgyt4@M1cb#&<7bke4B)H!;Q?9jlLNjk?Xh3nS7cx7Q|ICAY5pWKjX#zx)naKp93GBapVi$`kS)OqSOPz z^g8{+uhpqc-z({V_oLY@`~=Rr$m;G5um#2fKbpC=?)eOE9N-()4qb+S z*|{ZOGXIEk@!Q_&wpad-wD+CAY301w8N+uDoGtLl`2K?K?(CWlj<4hC^mja+j{X4) zp6@r1=DrkO?3_UvXAP)l<)zE}fWbc-(EA@Y;9qhk{t1Im9(?-XGY5~nfH41(^hXTv ze0@;=+Ya7vz<8Z1yZZ zYH!E$>_a@j7`snrKjIC11n$m3cjSO>J!X~2ofoJj3 zoi{5(HmJwGAcN?H_~VuV>45sht|ANe-+dtyyayiZo%&7;-zDqF3A(v+58$IS7Gxn` zoo{aVjGVF4$$shr!p2P=eB9WnQ~lZ_kBk?cRiC|s_wjz6ozsEyFW~Qd>i7XZ&xXX$ z=$7YrTIbYwkh5xhh)*ea2C@R=IYN48+~mpTU%rdtlX17no%PV)V3CHm;cLn{U(>e! z;H~&7{^>pNXzzw+2ljw>btg+c`SEIJg6ne@Kfb|HwXN$UlC#&Q!=ka#P)8qBgZn z9{%?M8|`-oe%^=dG2hm`y~H0$f3$z;`R*>#@qGvPigRA@b_2Zca|itYKQnM2*8L>< zGM@b40X_P2246AwhQWUv(8HfLc-G)Y20ymoS$ul=_s;*h{2haD89aILq`?gXy8YJ- z=<8p#;930F4jw;{FNl-=xWShUK7a6-0o%cyDKg{z2JY*!Kiso&rr`T@ekkYU-#B2) zkZbL z?rfCZe}ucJ(()0I#lqwZy_>Gh2ebVHcbl^)e3%Z&H-QJ!Fa7XzekgwJYzVwQ{|0-4 zJ&--YmS8`SA7l-^5zoP=@GiP8zQ%TC`_doq9lHDG(Re04^lu0FG=9pi#jo)u_dh^b zT>ew}H_ye-fc~xi;2BCUk2pNqI*06ayjox5)%wooPTsKJ@onz}h2!Jw6Ym_d<=OXz zv+0#lpSrbm!+_18-|PtQ_JsrQ%ukC)lRab=c|=C>&1_#jp3S$%4~&P=J@IJzA$`;S zrf0fW6g-QkH_{*FrBimN_ms|=zNrj;s(d_u>#^0D9$UTDNrr6T$9OIoLx1M0qj4>d?tqW_xm>E0s2>=zEy`yU2R9pLlqE%)!q zT5H1^q6g3w?SJ+-9^}r^GY0N)+;RT4-@6PRHgH$fcM#5YeRua)1NTMg@bu$%9pITC zGT?W2M*DFC=e3_W!2dtx0>a|y;~zPA$bha+-~K0q_ZYZ4#HY_Ta8~RNk^8D349JI7=|ATz}~^8|lO-Z>^a!T%@L z{|n}P-W<5E>Ab%G^YVtdpXr-4XV2~@yE8_1`1Xyg!1MW&+5PlxdN%!>{}*4!zxj*! zcKCSsSLpTpKJ@O*%juGMIX?xx8}GvJ=-c=szG+|CcX+dX$p1ob*!OSvjPWFX#O}Pa zkGnsKKiU8ILAER&;Ta#m1MGi1O+D)4cgR1B|Jlph?ykQ5%^!qc^54+Yk2*iU_P#s1 zb$|7RzTegkWDm%%%?|;EIPlGN;-+Rj7@dG^M3kIJvzyp9^96!NV_}c&70AIj&zJ7o|eA@s|qAz~$K)rYd z-wDtc@h@$5UP^a#pY6X6_}8C3!2jrx_@es>bYkzr-#}l+Px&XjI~|+9#2F?(oN+at z=8S)d%(J%e@cd_FFL{}KJ<3vNR@NtZN?w1%fUc1YARE|%>TbQ7V={n0P+y#n(tWKt zYtlL-Gs!cygL&c`@|`R{kvkiFTfXsb&d39ImGWzuGk&cf9gqe5T6-*|8AJXk^X*;- zn}w}mpYT!IV{`^~ynW2yN5`}G@dEcB@D_UKUFUnRcN;u>fM>hA|6_w+8}ME5VeB(g zz7+lvessM1!v=gGKo9=s1HAd44)|gCSDc^X&v-9>i~r-d^nG_k-SxzK?>TU9)Zcpw z`x$Q*k2;5SH;C_FA8$9{9C+caUU=m$Dw#v~b{6c8k-uB@gn_?Bf?t2!z}?pOA2?Tb z=D;?2%fVX=++{*Ni+{(#KNz^z^p6Mro)mun;RAJj#=zgoa<|EyA%7c}zsL6wPaE7g z;QxQt1w8vc(D!)XGI-L!_k+Igc+}wI2m1HkgLfLR6<&96yMeou>;-beofvoY&8f2k zzN7BrJ5P46>mvr;#rG|NvqHK)-tUYR|L0p`v+>FB8Rze%*YmUB^<+OEv@>S=i|$01 zvM=0Wupg|!eb(oE$k~}SY7ek~`@0J6F9AKtzL3WMOMk-O?lPcLv7xPRdy5{~*|c`- zJ^X=u06LjIpyJu&?M=ERd%ix|m)Ya&S$f)^4A|uKANM+(1=7>-EV`P#$M^i-Eie?`u=mpg-4~HBUC@5k#?KAZjc?0qCO(mU}T`l$E9r|~F#caM_a8_&VR%$vCk zc)0nL4ji|})g3kO?VWqy=gv0-#_X2{WEEDj6A~2-2G0E z^{o=$QQsOFFSudg-VA={F1l}(jHAD$>pa1IK(fu9|5q9CvEZF|8NBh}%`W~wkbBMF zKh&4E9=z?~Z3e>P>F;+N(BYMTrNPS$+*_x^3llr z=P#r=iY<28Mw>h ze88C~Ka4fU*V5e}dO5uS?|#yN?|{9VKZUNtS3=*R2h(@yv-ETP-QKdt>@_}6=S!QX zTL1JnzE--<{u>8)D}HG2;3=Ee(P`+2_#OR`PKoF7XMWSby}oQ;Iw{?h?uviGza8LX zj~n1+@;-4uXH}N%48l(uNPEJ-JwW$Ll>3r_y6{H!^2ZDwGWfv3`wZS~@b?FAGq~5_ zo`W|V+J zZZeci1@e^)lqR2YWGNY7P2v0eQ|?5w+pJ?}9PHCO%v~@3%LfnKC3e3KUl|tua`rvF z&EINsN6+_5?vvoJe?Fi;(k1=<1-dC7htIy(fS&=c{fNQ8960xT)PR5b3kL3u@Kf^> z^D{qX@a==|8}Ld0)Zmu~zdrb%gFhb-khikRHRxXa*;E^xQo z-0e26Ke*%I)dsg6yztg9y!CGlerE7P13qs3aPOTwpnLO!;L+~L;LpE4=&mAvG~Y$K z=0oTIsQADEeUpy+&VzRt@MX|{@kYE7Z=@U3N8fbtMgukkevVh-v+z2D*BZR~;8h3o zR(=o*4-X~t_&@kH@JRg8H>UU@o{#s@W$`}$R}`JroVq9Z=)q?V$WV93-P`{7i~rx1 zjRVe3KW0Gx1!3_YIr#7aeHj0IzX2aOo%lWjywtrq`f>N`q{;I>1Nnjfq-XcU$#mZ( z@vS^)ppB0jX!~Ccd|#%YzDLKu^-q8N4F~y9&-y7{8Tl)t9{Jo&cP6ZD_xjue)NLk7I+fWS(Bf9{;pk^{~K`IgCmCk*g^=biL;KAC-{_rURV@}ttR=}7F_yAAMn zIv;(I4rGtoulP7#Kv(5sz_akGe>reZ4=;Vs!9N(_F@I-(-@MV_^#^#+-x}akw;s$9 z*^6H^__M(u5B_lQg25kMKpc1${|AHLAN=m%w+FvD`1Qf_2fucKn_hg=i=Q|6mBF(I zzcl#8!7mJce(-Y_5SA|gZw%C_jXxdy%Xp0_g%pN)5Z(t|ffAF3IXNmk1bXYzMJUPA9nXAA3?K@0&?CGcY zEj`p7e7dH;4Pb7+cHn+Ke(QV7M-BK8@Fe=CvqrqnIU-*9P6Iv!x@UK_?{(#7@x8A2 z6x{c5et4Gw-@@w-+}k6k-RHogK={=MuW|`@yYt~b3Llqy zAMP;x*x;E1XNliGaHjUHgM52mH9lx(d5^n*Fdqh=AK&5Q2m4+K-pH@$EXX+{9?9S6 zOvoKS{$e@^pDsPwxti}T`9b&|=*M^_y_ijjXVPQn!E_>e5`D+_Bz%1Qsm^-nrgTy| zuD#2acXsm$1H2UV>>f0ojLu4Dq?;*&@6u=ac-8Y2gD<~;IQkpiQhxk->$@K`cc|&R z^uMKk8xQ@x7ysUi|73uF(nIkt zb;i5i^yW9cqFcZ1&2M}2I}XT+`wZ|h{Q+Tdbn15=(8*DdAEe1cAO0r;I`@MHA3J!| zz*w*?_@wEm=7nF~`6piiI|whdo}NEo;Jc&rjsxF?vB8|H&{h1uRnHyxcI7s=y3G|^ zkuLOx19!W!HSt8c6hGs;4D4C=zvx=;Hoz-ESRC+6zR!T4k$?7+2JQ|&Zh$A?FZdE3 z<*ppQqf>VPedfBHtn*_*r4uRM5-0X^6G7K#Vnen9_y`GLu$EB@)gnU^~ZzPk^< zGk33lV{p^pmj*vQaJP-G;+X?{3_rswK^Q;t_nzh9TmF#&Kd8P-|Ng=E4)A~P#BWMJ zcRvIlv>^}O@Iy$*a>Tj`q*9sc~V1xF6yhto**p ztLM!I*0eUX=d929-c18Ml^&3;_>?&x`PzXy%lMzOV0WN92l4%Yvo^jpcG-sw_zE5} zc<2Cq@IXH2dVFv0Pw~t0-#OR1`{2$4zF&8;C>DID{Hgq|`~%-VU~{tB@j1K?KV$_@(UI^S-)z$x{cR9@(7iG|kuMU@0bz0S z_|E&egXa(MGxv;tb3g~gL-93yRo!$n?ch!EB)(0cx8hH}>v;SCZ+3=5$Hk*QWkBzJ z#Nc6re?EA?;JpU;|KA(@odKPV-R!;qza{@A{hF^4zcOxoj=woDj${tshC2bq7|$}^ z{EB#%^B|z(^|zd)KXV{2@R@?Y0R_VODgDiV{7+eR@KLhM`Ch+l@NWjbQDGZc1ME@! z&bKObK;M4e6 zb?zkKU3^e{pZ6HJEBg8aXIyM!XIS=@zol`90iGt!nHI3U#P2f(wzRVZ=OpyQ;~A8F z%U9-X2rqovfbY#6Pd-h1o!=84KfojDIsAuwiRqDaJ>SmJh446(ACZ4AA0%B8f8}ds zH}a!u=RXekW!aHW9k>_joZ6YS{_*+YE&AU17XHoO%a_MDxX-uwd%Xv~?hdLp_-6H; zj>;d)-$}ovxB2!@zr2_ClwV)?9?1wgwJ~%CW$ek5w;4DWbpONrnse*Pc_wCSjj~_Z zG;}*Y7&gyi2465B>*;!IBWoXT`__T|z`xkvL(5k$@8bsiR^V@@D(60@y<~s+8#iod z=LI(n>}@_V{xNnmK1f&mM}rR2zxBb-N55ka;C0T;UT;3Y$LIpS=XryHZ+`GF=kn|$ zXT6-(>{mR>+4W5WXSjTS{7U@)_M$y)U)s}n27jP)9A_{$44l#6E!m*#Ph`Q+$g~=DS`x8~x`a2l!a_F@1`p7PH zzsX%GeMi~lcpd+yvkLF(tit<(Fy7`28>MH-@BXGbK%ID}IAeq7Z61oRE02fb|06DM zcHY5v;2xB@!^8L-+)E<2=xN_QAUDZHGLYYnZtt6LK5ly17Z045;#cH48;8w9e_#jE zFYF5T<@-`NKpj;-mg z=ko^6gx%L91NhtM$@q%DJp=r!>`A_VHXVJ+xr2LK!p;}`oo6x%|MATJa~1%e*@SGv z?)K2r`Ec((z@yla_>^y|oDt-QyWQOBX9K?QuU~lOt|$G9PWH3`9~xhXwf2pJuN!>% z0Dr=h*nxP^)h}fqeuW2#!>8Pt1a^hEj~(D!?$CMWcgDBkSL{Q;x12o)<6ZPF5RR9< z_uTJcw~+667QM@PGrQ596O{89rCT|F26m6TD|njwD)`wS4Z4fdnI*mDxdZ+=cX8aq zv!|hZjr?}@tvqzI_yWI@JU=kNPv|uKZulG<&0QRJD?WyY@zb+eosaNSu{*auMhD{4 zqxa;?^Q=Dh9$txWN{gRq6TjVh8=m7n(K814G=5h9@o;(?z3tBjjTv4f&Nr zJNb+$*+5p%b={+3*VDWB1*{G8Z7tBjtRMc?Ck(8o8wSp!$U%PGUmNg8@cmk&cOAGl zV=q|iWH4DwKl|zd`Tc_fJ{&$B`jo#j=0lmYWJvvZ&3q1<* z@r%bx=q~uxRjcn?uJ9OsD*6}y*Ut~|tmh1(_!oZ}KiZ!RQ2+movj%=OzB!B383g~^ zZ3mPZ{yKLr-8&b5r-A!6&Mur`y!YTe2mff``F$_&?(c8w>ePjc7)`5rOr|ewzEZ&J{;*VcEKtE@IC(=jp`gAqt!|ZnFKer#)bj}-oW8hv4 zU6+metU)#^+m!A_uf_)-IdJC5o;1gJAO3}B#j8+0E4C$H+C2yOTRe_$LVEexp3WEW zH~JWTpAGtF1Gb`j?e4Sl@#1;xMc=$ygHIhiao`>}x$8S^HXlDR`_7#YGu zJMar zDFZqceuhUI7kmwjm+>=4{D^oJo=>lJ2h_b#a)vB=@B(tzDGAwmj9hU-u|bH z@!{V%P@cbx&c$cX|D_&$fM1JG>%oJ6I-vj2gXly5aPYnZ@gF*n2haG}!6y$seekJ+ zzL}v*vGed~`sKb)LYKnB?>2ag0shT5=zQ5-J>x=8b@pN`>9o%nu;t7F8S&V`XAj5> z_eI>L_=o}j_yY!f^ey_-`wYlvJ}rBpZ)DUd48qE2gFfb5(_Ikvv7K## zu)74}`xXzcWLw_(!l&zpxJ&txQD1zqZu2V~&y57>+b5>Ml| zKyN>IlL7xe`;p&--NHXi$MUzH=~;ZyY#h9c&6wQ3X?$AjOFm|M5#M9~xHH1f^=j$!QaE6U*Jb{4nCoeAH-Ym5;hkI^PfxOS7*cViQ!A|l>_zQi~R8ToIL8p|M;l* z#^`N$9{)Z(V}R$;fB44mIQ%XCg>SJV@lt1&%HWsyB0hbHjWv99);vUE=46M;#AMlsqg?Qk% z4YEVs-{33zvVrx_j`)TFz3nRpcqE$x-^0J~J$9u#%=a7MJ@_A__(D4Fma)_8~tP-c^14 zbZkoX(zp1>w8w^I2kSQ*H$NBt3UplhD0o(ePf-7VYTzBf`{UJn;_u{@MpNu=8Y!!Ds`E5RZ@aO?wvHPmd0r-;L<#*2t z@-aW@%58!tUGYJS}h+~nS*pbZQlhu1B?gLfBx15@M33&_^o{M zv%flbV3Xl3{C?~-em{n%^7K!@htwmDZ;)U8R|j|zKE#La9q}N%$2kdqIbS3F*EsNJ zedfSDUACROx5g3Qx%a?WyANR=?04sc?tz&PcO=+te=@U&clbXT=u7+pbSHY*Zw=^Q z^rfF0bdHGc(7(ii=kFTOoA8Q%Kfpu8@j0N@lXNz~QwHvj{loz8s(#<`i02Rc`oWV1 z%Hc_%9KQzt>=zBR^UVW(G<~BpvN6@?9DpwQlLKjZob{`{?R(>Yd&WS#_qU$;Md+u- ziJZ1>@lJBy83elof0ah=yBo|_#&6$pfal+D@ZfiIfHF(X+7vb+PY^C;x%9XAbZf zcV*~LWF7m!d|GEuADA!m%0948Uo92=FyxOPx`kxgqsHTzIil<_?EKPfVIR<#IMLP>xT`A&#?FC^mqtfL&qWe zQ2giH2hM-oWq9(%*`c(@52VwDzIwn{Pkz#Y_&eAO_#l1<@+kL&0p5n^s1t7!Cruu_ z5r4zil*i}3c<_aTz8U-M`G$bc?DGchR7m6Vl@7wt9a?$axsawTTlGN$_9@vVOuxq8{^_PcYqZLbSIbMXHRq|2vn&*oZwVfLEpT$xiUV++4v`StOM!p zA3tWmr)~_~Z?S&(vAo+C49Ft$sLy;`=GJ^F$4_9~<@H|rW$ffN*UGvRX3V4;AN}qg z&G*jV*CXE=2QrN=CQrQ3vvrT}u{H4-Kl~IQwMWo+B0j~}W}o1#()nt|lYRI%-u7Ap zG7!H<#YtDMviKbTiajRI{$zXL=gQ)Bs5>BJB|C(^$Mzvx*d%zXeJ>9gLHEO}@Gkkt zR&s>w!3W7LyzRjQb&<`|-QQF{*-O5XHTVj-EsyfrL#4@!pR=p*TC!JJY30YGo%M;| z_wd!BU)m%GjL#hh-og8bBiHCy#?qLP5qOv}G%hGv+1fXE=7G$xMyv(v)S9!e_#MnM zoyi{g@qv9uzK~buF}=s0}{}C?PYf&=w|eM`$%5(SR3{us&4hhTg|m~ zl&rSy>^c3hHpyf%(2w3}9LP@R<>r&FL#Hr)bSHC8rn>JTkLS)(=?K!u2|9_jVcpPe zL*$0oK3p+wV@Bv9yK5%t#fhmi6e{2XLbyGCi%=? z@x%c;#+@7Cue*RU?&Q!FvmMk$egk=J>^^tE-jE0N{OEx(0by~{flQ}=D6c+t3VTdj z&J4+a^TKB16Q*ZqM?8NQdyC!X9X#``&|lQ6E#tuEqoaT^FdpW`*q9^zbXT%_MEant z_mWS3>%sWbg~7a-2lew==ig}!q4X5}pi6mI_Lq2N^u?T6lXSAa)u9vSU$O7;0(u^P zf*)D)^29UlI(|ayT$%2z)1~bj`-_fa@7YiG-nM1fTwAZickvd$`{}v(2^|T42KJq2 zz_-<Qg^I5B!e--zt+&T&wV;$2?$RqNS zY_*s0V7wWxN88)@L_9S9gcsl?((HZVzW>FC<;PR-e!PS&D!RJP<$TA6Cxas&FppJ~`@cDt{0E5?>LY5x+nGPriJ)vJe``yiy=03*r2mDde_^MuOaJRwT2lp7f`M~+!y#{YKaOU~y15P}?EdO`5 zdfbQSkNT?t|J3aV{8e`vI4gYB0spzPK^NQH}87Wz&pz4I|ca~ z3-9av!Wpjd=7V$JgbXtm_<4NX{FyiNMULX_4?@2J=XaQ zyOQsVorqd%*4|aC^XlK!mWMul)R*y7dJ@pF=wa4zdKsQ64#g|+$m~se9Xkkr%`L~iczcR{z&_!NXSaYb+m9W{50AInbJ8lqHeuVaVb~$|Eql+oCLXMeyvo?O z@~cZ*_HO<+`xk$=m-PusV+VN;eZ~ir%YKx$eaQAZ`C+Vna$tPOGO`Hp0prC6V%yMz z*>t7LPew@t`2r+YG0kc zrI7AfUi!Ma)q~QF)vs;HKC)iOaplQevX*^kT^J8zp#RF~qq<%^U`Npr*dugd`UHP2 zdxL&qEdzan{Q>-L*&^9Z`5Dn!N$M^U% z^p#!+bi{yH$VXqqQ|OU546b_7zV3=oN*rI3bYLGVV?W^y`l0W}<;DTu34VfC#CNX#PyFw_$~ z+k^I>a|=EZ-zU<;to8gV(($V78SS?J>=k>`JJ=iSXZ{;$N2`G)8Q^n{NYh?AE;Ngih^+SGUc2k&Jp>{b1bH{vgNPGe-8@Ka;u zJ&dREhQ`)==hMJ*`6}|;d$)8|=lb3Y*#6x!M>_-HmttGGE1(|WI|1?DA>S3hgYhv= zd}Px3NX(UY@*d6ieh2eoUgCS+1D_6fS12wyU>)OCbS8WazbF6j1bf83u@CJn_GEI@ z9>Xu}9eDtcv2X23HbVLl9SI-7C&VZB(fqJ@2YwVZ*&l{!JLxico%;H z;8*<0=AJAi1MwZt((p0-kF2RIoewHR|51`8IcIj{-Y0gl=lVU^e4R6J9!U%2k$E%-fs-> z1M-AE2g3R18z=MNy^RTe=N<8La$(Daz9RvcY|&K6HqmVIDAceL-}Fv-0Yl(ogodd9g0pAm0{3@M97+*5q zX!d~gV7=J$)~9s})~B^<|JWPs3HA50j?%~Ev8U88-dfXd@6fs|UYdA(Rh!nZ_Usq^ z&`<9T+SJB=hwK`0-bI`0GiJu6@zGCp^uDOF@>=i43qMz`aO)(0iLp(eRhP2fLsA!gGc<=h^orKeg8z1$QW^BzfnL@VI2k%2~PR}mfv$^jbdJpgAr%r8v`S5=B zjP+*yYe$@)@luaIfM4rLANAc&{`#bCao)!`X}2{gEKVI8>M>{5lr@8T7GIip<7#~f zTPyMyTWi)hi!-L)t9JEG-}R}!SFUv>-n;3$G3;He9nZ!;|Cq4%=^d4mk3I;(()3$j z8fSfI9*wa!#2F`fxkK|Nbnt$XjTf6}yJ9Mso4%j z)%z;j*hy^{g-C!z)yVh;vKZhd1y zx93pL;-y>D@++slps(^7%f`%@>%08qf%46za{8_x`YW%#dPnuBulF)`p5?K&T3^bl z-@B`uehR|M_^Gq?qF(P_yQP(`AC0;9Q+MxYTs?!ZI@D*3tpVfYz4c8$$ZS|Z2DWYjzBlYOg=bP{pem#&O@z2{yB zytUSK^}T(??~s8_(Bj2&Btn%Xf?R>)K+`5Ucw8!ib@z9>J zF2o0A@XO?1GSJ=w`@|k0`?4YEWkDQ}kMgNUIg}p8reGuRSFt1bp!1KgPk@ai54(yj zk$tHw+m4-~9N!9mvv9s=_77W)-3I(y{6O-sUA8^Q=dBK9^To4e*mu>zHatST`Gwq# zVMnrq)R&FPK2i_ck^d>%lAp-4yEAN0{e!FCrI&tIRz48U_SBxRKB$YW%73Sy@~KZc z)Ia^%eN%@tdG0gFZrpjilX&ANzA-fZY(bQ-9XxN_FQ4XJ$M)U&x^)BANBVqzJyc%$ zs&$vHYOUfu_!mBB-M1&u_KLL+rP)gXpGdag5$q-VHa_#bv6t9RY$~=Ru#Ln?lNV)E zO4~LgTa5k5U&U9&AC_OeGbZ*MNasf@5C0K=JpZ}7uATGoaq+Wnh}$@(b#K!7<@Sy7 zSvi;NY?A+qFAJPI@^zIiEDn?_%~_}O$$V_iJ%O)Gobvg`)O+=o+tq(Bt52Pkm9Ktn z$`8l?#t(PH!23D-^uEC}?6f_GzFXisbT^h?l0VY9h;i>c*gTjIbLR}jJee=P&Kn1O zlje}W&s_7B@>#+@SLI7!Lw9D=T4GaMKgl2Ki43v^$tG)(?6N+|tmI90bvm&1PnWSD z(h)X4r+Wmt50C-bCiI;4AN$t6!*}soyqB-YUXBO1*YRm+&*P`@&CU_=3;YBRihsq6 zk2KyEmCm2e?}ne@mHcsh^3EOj?wvLCorJr6M@Z{Dg1_Irlnrs61MIq-OFVes+yf2b zt5Z4saDJgr&J9B62imHi>JQ>-d)K3UK1c8C{9vPV1$p*aMeSBr8M46mB`?U7_kl7|Ha?QhmB-e@|D~)?|44BY@Ou)U}Ur4Vc zFYNKoDed_UGKJp~PjL3J!4Kyg06G)sc~kk6Q%CiP+vVji?i}51mp`h^ZlktP{r4Ul z-uL*1Fi`JeGLu0WT@3yn8^m_G<#G&iC7GKX~N=Gc6jcltG*n%!YdWGkRS z{5h1znzFXG?oH3|qib7_*5;}8X^r-iKOIF_nzH!#IbPmgh|<&dzDO_HbtDtyRnDGD z4$!HZ}&U9V)w5)jDdG=uhP+uNRKl%evRAJoqly3 z#qWB|iMcUX=Fwaq=exN>Hd_!4l2Jf(bj{L*gYJIiF z`;ZJ|BZF}E_NL-Pwzy|y$WeZ>XkpJmTu?^1`l{z@53Z(_S9!KPkl7%7thci2+Wp?) z-QKBu#YL4p@~J(p#&Pqrt6c5szvsq0ey5)Fjpi|)*>m&WbF{U9*KQshzq6L&TYG&S z`RLAnEVIkU3!|lZwvLZ`LHhp5bNp2%+Ma2zp~Y=lezf>qCcb*DeXD%G`nt!;*zED? zohoy#yfW?GQikNu76j1cK+WjA9r8k8?CE&Nwof|SzER5o$F&`-t(rV zMfbckpF6*Bd5`p1w(`e(o~yIDX}pf>z2|5ncpR^uclr)X_gubcd7`z`IO><5=hDi1 zb+oVkTiTA>;j6X0^tqLH-|>Zybl3B;gzFos@0%9B%9UR7Z{xFR@4f5OAHU|&ydHne z^>OXAUiQ9k&6Kw3sdZG`vEAL~4wt`rwM}oJLj=$Dsr=E(9jWx~-22My+XUe6Rguc<+V6N4m?rWM^GHqw^e3Khk*nrt;7w=pS2GOt;yzJg4%N zb}oPV=uY3^W1ZEv_1PUR9o_L;FY4Wji;8ayjMpB|V_155PmO2euUzwFj*t9nuZhCe z#QuqI4YjUXXYGw>YxpwS+HGAQ_h)ua>3i=~HoEgRSYeAjGi(|O{h;OXpC#UAMIJ9|gK6)-s&(Y#rKSx^n zrsZkvwHDFiuQc)2aQZ_syuEPTPuKI@o;vQm_HVR(eB76w%X2Nhv+dt)6K~#B9&wxR z2=9FR`IIfV>)&;rs&}`u+o;?*ddk*3malI+Z*fPvRXe-!I}{{w_#)RBj6eze!Hb$#SB*SkO4PkRrE+vU&YDL$&q-pBc@@>?DEyXV^3X`2U}3-A2@ zUx(`x+PgJA@smyC^+*2FTsDuV=Df5`H}4_e;y;_li*}rF@u=th+j&pL@3O7u)_1gZ zkDl5mr}m$?_F4PA=koX59xdN~E*$TO7GIg_LU-D!@Xjk-e=egt-yVbgp5@u;g&V`h zG-}M719R0JY`W*L`Avt9$Mk%P|CHyr7JEkbS`>~SB`c%ik34CoRW4e+Jzp=a&X^+-=LTA9AzrMX?549%igE?*tGmRuJB$X z*3o`04|=Sld^^v%@Ud+9#24PQ{HLfn+2!=3^^f-3ys)@a`K0Z<$GzBddF0>q9NRBV zeD72qW&6n|extHSy7Ls4$FF{!qw?(fDs%3=q@Vg#$8Ntq?EK{`u71VWJeO8pw0bu! zU0mU4b)2H*sV?(Sc-K|g`gkqcn4O~W_~@xTdkn7CR(`rKN&jP{Po`1P*EUyCYN zzjj~iOKGUM>*?Fx!}WdlL)^}@_jdV@wEXeso{#x@-t9Fu`oG_2&)0sx-f`C5-DhF_m9Jm#c}r+zkMvl_ zo`apXC!;)>3mw z&lOJJ-LyRC(&bHWK1I{5(bB8GXSBFYkLkNFyFbOBde7rncrIV%Hr?$PF1>P_mWGzU zeqZ+P@~fx*najevkGrj%Ubr>T8bQVF=UWSzt%hMaZ$@Jm~IMZEt$2AJ-dK{nu*adfGeodDmHge?w`Dp1qU+?>t&p7pK%uwn3`53Nz?M917#nr#w^+-#v ze6;w>XnE1yZhh^&PEl*-xQ==*?bNzD)>U5XcJGP3o_BoVoqi0LC;g@8BRyA6zMXa~ z*K_rsqUlK0)jr$r-}p3cyUoJKw$z(nX17tkYtj0u>@7iUDu1et`in}JuljeHov-kz z`l_RP&|Q!8$~S(|#*s!tYlYkI*W1>yZT<32`yF?9uYu<6WudK$=J^z@-7AAU|1@UJ)qcm`qp{m*$}8(v zzD;+TW4YQ!oA2VT)_C>b>N{7L^qr-itAm^(Z(d&8oCy>KJlx3?fJk@(elWD7%GKW^mA>Qlm=s?fN4ncNhRa(WyWY~2DU5pFW71eiQ~t6! zlwMotuJ>;e?!D?4y65ZE+Agj=aXobJ*}}^0x{62pm42l4amN>Te3$Z6XLPrB49DN% zHLA& z*IQ+-rFZT0%3O=Kp6l~|kHW@ekJqVi<9j`{@{M)vp!$K9R^B7k&M~g%>*<3$drj?i zwaZuMWmI3!{k%_ef33MLU72&gQ|+9}TV2ueRK91lJgvK)OWW(V^-`Y76kq*l<;&~2 z^qnT`SK5)b7LM`P^Q>O&?|$xZ`=hg`&^mIM$3m57cK2vowdJd z>$G)?7Psjxv%@=2;a%6RulS>Wa(pgd^i=vWuIJ0jlor+B{Vu}2!=}ePjdygXSFScT zEv@{OKhoy$Nad@KyWYapziDk9X?ZFeZ5)rZ^va>?^wZDQT(rLTj2`!oecGOF52NBv z?crU%dTZy{2ha7bw9CpL>$~z|>!0KJ?dRI5j!m_FtzY@&tI6u`89l}w^Y6JS+`O63 zZ1SUDCp&$|7q{)@WB9m6E_<#Y^~<~5lHW1^e)m&jQ{D3U8RH|r7GJ)`7Ojt)9^2n- z9@Et8*Suavujc$e|K0s6e2(tEp9-HVC;tCz`6XjgdgYIG7mt>9Dz3EScox6Q6fP~g z%kHrB>-|;NIeM*nb~$URh3jk2^7Y&03zt^@%C~>6msSp~t|Q%f&gI*Aj^*`L zzuQ}Te-*#izBpx$^?2St`TWk6-F*^1j>)-qsSR}(K2m+D{#|czr|9v0j&*9UU+-kB z$T{*ZJvO>!Vc~N!QGNZ6?Vftx?|155dS|@6@TSR;cztO-CnHXkFV88e+%dko&e2nC z=+C+M`Wp5APZS?+o ztc#bgGW>L(kQf9)a$$RZ+&CTMPmTSl zxLxj;F74EBmk~ae^IX46t4z+P02=d;?j;HmPpwW&V%)u$uf{T9B~@0fp=5pG->_nwb**CVa`@i))$-FP2b zT={PaJ@%op*Hh=U%ATv^SngWS`gE;4*E4p#+os2QcRi=(-MTrhCHa)=C(r&Bzp351u!n;!kDJs-cd?i=ZI>BaB5(?!wU#~ojImlH02({puHw!HG4 z8=vDC?tG`xii?lzIoaVox2M8;o`p}%wYX!x{k+%VURyhlaP1%aCrvr)yY=3FydGK^ z^i=-R+duVJyZwrb79TwqFTJ#LRG!AQF+htit#K)B*C%g3>HAl?)&g1_dhWOL6h7uH z{z!Me`i0i#;&wjabMeP~$LHGb9eXZc&!tD(C!01tm90-`{Xa#^r`)Ob>~DL0@B7Ug z3h%xYcP!_*w4ML5_bUJ9H^ptLzUuKFg?BqUPwBn$_0q=oNcT9mCtp_DJgM_kJA2Ng z@4lCR$K8@}pa_(7v z<(jMGT*sep360NJMxE-fZt>-bZawr=xO_*t+o=BNZs$~;J8suool*4_j>^BwAAJuy zt*|ncuaB3}<^uo3=L4;`3LX`nxaE%2(N*Ptjvt zm;GI$Z0ix6ykyX4gv_zdawvd6iZ^)Nkjx60!c-X~HY%WU#t-qH?X5Ekn-9 zfm1Rh*>I%l+o&CMw|k7gl6cXtWBJmKw0Dbo$DOA(PSq!G<*tXGoAcJjUPt9Ux1RRe zDqKC4MeEn`y*&H%u8pl{@s*3JYxiY`YyTW=A4bb}s=xKUv|~Fv{<84#-My1v?{lQJ zd#>Eo(&zs@*0-N`-RHu`cWI2G$MpSNo@nEY7Ps@C>x206U6$u>nzqNLusRDLM|rPN z>vZqMmnGbu-n4p-yl6ki+ls3Uy5p+1es6kwx1M|VO{MjG9CPuf;>usWyX|w~+NnNu z^i!sAR9x?Pq{lI7%(CmU4UN~%doEo4+Pt1$`KwPm^^Max^+BQ5btc4t~(sB5+|SXr{1-GMR(uj-(~hY z_rB3xPF*{1;T?Z0S6i3S`nc(3`A>~m<8ZFru6u`jhwG)@uV3#PZNIf|&~%)w_ngvw z%7<27ebrr?(ftm>#g|{%>Ops1<+&bOz0qTt;_FY(=kiIH)>s|s?sNS<<~jAe+p7;~ z{n}$x`c5m{`mvt&K0GDs#O0GdMLQ#l?mWdsciO3N`#!pPf^dDQzG&xq{e9)m^|Vo4 zX#6Ce(jIF(PK{sV*x2`sHlF9oRVI2Yv&&W2?n~*V^;~_Y=(X0(sWRerx$+d(dhJ>I z@pr90mcP1B(dw_gBQ5<@*=xmBHoEJ^kB+=+KgYL}sr;t3ed-;HiP7ch?f7%y@<(^RW1F>oEvnqE=Tx}%qNnogcfVFx z`C}e&>TfOae-;*hs?AgNT~{| z$ltF%UJu>xyz}q)!qRJF(^Gw|-=!(59^=_OoSN^Q_OgVJ>!CFlmA=c7QD>&qSsaiwp%>p2x} z9@@{F-;{<{wz_*pi@O$;w>o!w!toW9E_)6~{rQ{@ebkdr>E^|ybMG>z+9@vmTmD_P z^75VQ|GD`6{=$vV9{W9hjiWff)7HhN z?JsMuH7wjZ+j)g|eLH;2*BY(<;~GBJzw3Hg!n^%*m zci-uiN9*sYu`!R$?`5<#VxQ&PNiLuvxpAcB&9~EjjqbH}Tw6UiUOjJm?0adI*|fB1 zX|0JpPQ~pxX-pgM@3d$K6K~V z@jJZp2w&@0{_2m4_g=^HJFE$I<7^_zw6w4qwsD+c$d2+VP&fmpV)dt@zMB6&(Y$ubECzdqVlBM)aHJV z-cK8a&r$E|XWsT)@A+>1Zfb3}=8v>>zv(fL=bc{Ndrj^B6kk4cmp@hij+3{t>N-_- zZERYZ+S+yQaOveot7pfR??`uDh2t%J0sF3W;o{Hv3Y0JHNOv8_G|#n-)}Ls7-|tnp zv5gP3wt7adEV%X0PCJJ8I&3^q?diw4K9yH_Kh3qbPwj>F+v01d_d=U5<6wL) z`!z$+CC z_Dgjnw~|NIS-ztjD}K}6w;k4(jnc&JU+;dTm*tVBp6c9m_f1$@{n&L1YqPj>d5W)% zU4L;$TAAwZ-S)aFt-R>*cTB6EX!W&Mk9)c2_H_Bt;-cr`#Gm`^`U+Ry_0VIP-Bx2! z+tPNOJy%Cmer5OA2{%W3{@O=7Zm*fjpy%o!`_bZ}JO2(#-{lG) z%k4hx@UBaEr=<%v7S|f1-Zk2|?efQQKlOZSZZBIG@}H`wxMROBdse3U(AHJ-O5)+4 z()97N{j%5nEeY>^bSkVaZTH*V6W6nF< zS0CCMD6Tx!+cR38{cgu_d3L$#LU(*&@w@+x^`@oEd)ZIEU2cc>Iyx8L`3s-R`~P=* z{fVAy-yVq{ANlm=)$&zGb)tK$%~5lDql~snc(2D_WcB>hf2|k?yujs~zQQL!H$hO)l*B ztIgudb1dJuT}F31g-_94$DWtMo9?{A*ZP%zx4YlzTzKcdo^b8dA8XEDIQJ_bS|0gN z?ZxAsD*rjU>pq6-=VkP=KGnbabgB=>_lVD+$?(mm@T=m>TRWR4skdM6b?!a(oXN9) z>gf4geY<>Xq&>0sRB6{stK*n|wE@SM!A$dvW{c8FqV>N1=DDP;tFz}#>+>!rzdl(z@tz~k z>RI{@Z~mnYzuK=}ajntnk5*T-_|~zq=YG{+d){-uUt_dsPIdWR$3?{r!Do$p-zyFKB4#<%d19@8o(zBxV4Rq^Jsc}0(NTU*C|-IC|t zb<<;cX^kD)GkVOs-}jj2`Lds~{p73s?pxvd(ev(0ah2b7i7WkD^!U!F@}H}3m%S`} ztn*ZTy~ioqcs6d=T046Ul@~2ODz0DodX66B>kBGR&!_rb-0}T;mfo-SHa*qn`hITD z)Zb0jxu4I4t28M9q`jIMM8~c5Py|4E+2CeTs4vj}tnd+)vJx7o2?KT?g?*o zA82KF8{&?2R9|(qucF1DqvhN5)cUfXT4S3Y*Y|!dUu)5N?N{8jX!+04Q{|6&cb@u$ z7BByP=Uu)$mGj(BnZi3?;Y}-VdORuih?d@3 z>G?9MZ0VJ|9$J~&-c;ItZVZY$mhE}d@aPifBOX;zEbE+P{-ClX4r_yS_Jf~>o zj%mf88~4&td0I2&-|51;{?c~2UH+JVm+hTeyFH`DZMxG7@A?Xt7A?NMUJvcvkMCt4 zZ9aWV!pb%_o7PtSs=ugr*kf=EZ+^RZ%g%Qkhh6^;m+ur+Uu_+A9nYmjuP2W>YpZ_r zylHK#^LVbnkDp(E`}Fh8Z@T+=>|gyxciN5@Zd`Z&go{7YUB0wq+5Ox&9O-dw>@vdp z_rGZS6S)7^Zm#>^`?_+MkU~m{E0VDZw@8`F6h&@RiHs#vDPuAwG80llB$*}iJVivP zsGD$;Br=t;bYFYl>-pRJOxv{doM=F|1>)b*y6@>o`82^E~&)Tiot= z-FKJnLTGo4?i|rGwx7ruQcT~EHZOF~;VkRlbK3LDF4VqC=a$*{t3JB)dz4S#?s{u@ z(!I`Sxp>;KSh>C{TD)!j|2xX&cbs)&_d0cd6<4j(uh#Um+9~g8 zJ-@otJo1Y=4&&3I?1?ve)zRV3sbT)ThSI679lirfqnEJPWte0oU>Ai;a&+a?R z#fCT$Go5G3#nh&|zZk1+(_4MDqnlrO_bI108>8m1ruVCg#WNpOzD4t6>u`=a`{Rr5 zb=MaEEqxi<8mQw<-u-Cu@1m!SY&_em*I(RLuRpqVYQtwtaUCt*yPx7o<`qr;rdyZt z?o;0V+t>LQ&Bte~?DhvK7jwnpNgp(3lY=Isw(Q9M>e1S-=*g2m@=ba|b_a*OY zJ?>i9Jx)29r!Cs!J=eSX)>B(!u4uZp?9I3PcX`{Ef1|svG1@Ahbn{=8<7d~NQ|(u> zujFK3F@^TTpGu)aY`|`!k{mR`}+&j8+DAykUdH+b>E!wzEm8-_vK+(R`Rs(sAXdY}+`bMtL`D(`vM)|l#>=PbuVXMVn-)}r=p zUu~%Np5$(tf3+p!9M|}hZaHUP^<A8BvwrrLY8(Oz3x^1l7xKrPio_HF2 zi{d-&EmuU7A0L|h-FB;N{9SS43^@DKc|~vO$)=NybG6-`Eb)lXVKyRvor?ddHy z*W{z=IAykeCoR_Yousn(RPK(lX*{*D+yd zLzA5k(AuKA@2b50dG_nO+84#=rh5(VDmSJ$+O)oN{9WzxtMYYfo(X|2uS~ z^R!Jn8~S$kp!AmA>ABTUe>|xAp6mVk_qtVX{GMAnc3XVW#-mql_3xZe+otvJn6B-l zo98O-i~T(=UD=4%7TtZy=e*;K#=rgDnmuR?#MgILbuwC4veSd68{O?sPkGPps(ef4 z74K>19?RCAyVh>6<>qVb72W>WKB>JiTV7@Q-s=vDU7U#Y1_uurmMdRCh zW$%|{8Q=edJz0&dj@H&*YtQa|Cw|X0+kU?)$B&-R?z<|-f38*Q)w);*ZJ~8J+w0%* zb~$|~-L|!dYumhA)-K1h+mvfZ>tkb2{nhoE^R4^SRcvlwRL&3oD>UCGbJJULn+Lji z8oTG#UGeIET3Du)chjj`WbdTi5=+jP&pvi@7o_Mh@y@wEGj+vx5)%kiAFyl{1&sc-L-bRUO6Gqo4llIQ_os*%^O*wUc$qPtTQYXWzCf{#&%Mbnd72TjRElJ?>Vy zxt%or+Vcfk+e!Bre8qBcQcM+R43C;}f&b?Rn zv!Q71SMS%g?Q^GceNl4A^tW=q_&EBzIaJ?+ywfKyn!e^y`(3m#`)te?_1&U-OzV?R z`Tbnuv)%J`sGUZ_Dl2Kw7H`_`NOZ3MsoDUx5t(rJ1Jhn z9e?A|+RpLG+U>q8@9)gJ+;{0;|In}hm2cFwoisUpr;eWT&ia-^Pr9PJe~+zv>cM|C zwsw8bbO#zsezMVgceM_CUGn{=)}+|7XY8N7pW1ty?tbNR&Ruz@eT}#A?bT@Q(ObN` zf7|u+l(P#x$Jck}s9kT|l}(LD_n6A%!TaUQ&dw`p>{WYh`gfk)Hf8&s_i5|N&-Uzo z5PHrnyEctyw_TO@`Q90iukm+Le2s5RerX=YpL&hyei2Phw05+%O?Usvv+dervV%?8 zSKORyaPqXS#pFrbZx2HEIeMN;JD=~Ia+Uvgj_Ew4KbiDY?u@!6pRS$P&aynX^W#}A zUp8*b+w<7%joGs0;m&|^_cK5A%nv=w0c}?(*Pmn5^^v*do-!IA-Tc~4{yV9S{TYhqW!$ZH}p=VuC-`IW1`WlOF-d&C-yPLOe&AaEY%gwhn z-)FFx5KB8oD%V~N#;d(pD-IiT()#1KFGBmOy@R%YHjO_SXlIrgLdxpPP7#BST= zTm8FT*%(xwy2@>5e#3KDZogk&xiq@->{-6zIrH}~Jo*cd{`}+kxkn#Eb$vU>lUIJe z%HP}eTl3sL+C0flhWeh*IhW=U&lP{Rrss-BTRNlJSReXCkWJZmi^g}-&A9hVERq%lgWB<-5jfziVu{c<1PHyE7qZi>EOsWz%kpKf3!W?|Gc% zZFl_1I%)l<4ZD5Iul=OtYKxxz^_}{w=Q}jtJqYbQzI8^Eam9mQ`>7+|vpxH?SB`Gp z#-rI(+aBAP`cB%|X#Lf@t5o0pmD87xH@#Y)tG25#;-S6MzUiJ|Zz`XU_RU*=`&+-K z{kLPd_`9;NHg>iin`S@0>?Zpb&U}!a*bQSM}PLw|MobP_2Etb)|I|fws!p8kFIDs>DW5y zt}PnxmZPoysIl3b&5wERY5yi8nVWx)Ip^7&o+jyOrFgd=l=J;QJBx?TPVpdCsz-~T z9aHTe^n7+-<&&p+{dduLqs4u+_GDC#){fpam)&=k^Huxj$~U#$MYr5-SLJMOZfI?2 ze9`)@_-iY-?HqEeTyBXbW7GP`*!l8SZqvT=R=PKTZSn2;E^pnty>j-HpUc0yUfa&g zJJ*-HcW&?8yrP|Z(JP*PHdNkv&hoZ5+qS)@PxH%3t(&&(pZu<^@`KWC%Pwzw&+?wz z>H96CIc4jWees_ghJ_>M5@DC41Y| zJT|?ucaPb+k{{hR)FxJoYcX7Gw}*CK+j+bEx%+py@waGt?nlq@o4-A`_uH*`#k=X2 zwe{?IRIa}jZa-ZtH@oV;jyPI~sUd#~3S!_}|$?`~6$ z7fmMl+lJFm^=&)09CGEATPXjOr>dX)j&fB}_Ej#yI&T{8`bocMJbFTSw^D7(EodH$XzHMs` zrw^*%qWNa~R=M~QXT|UtbJef<8;_n({G0AQR@u1Zuqj(ln(f**U+bKo@=>(*yJ)=m zullC-p{H%tvt^H~+_?OH`m6f(*IDMX@=^Xn&nH>8X#B}5r^&U>59e*~e`x3Lrsp|R zo{lCHt<9N`{8Pu)x651SS?J&U$CqwR-H@9K*$9(2pEobGM+F4rIbN$Zc+udbZ_ z)1LUyp7ged@>BF|Z*BHEZQkwQtuLFdXnfg(Zo9|~+T&|nwDxE+(8fPeYHaeNTV8X& zqF1`&%a13m|B9ZxPo&=d+GA?lYpH+N*;yrdww1wb6gtrcM88Lu2ov#u@J+w>`D@d*zeHdy8i0W6~e& zU-tBDySGnw`&n)*s;_ayYV|0do-C~XMCg^SbVj$%?AEs9WA7v9S?Awf6koA*rDN-= zKbz3nvgOL2)0V5VcH5r(E&KFQ{Mldqq}#^i#JX)rl+i3Z;kjbzqY=&Xnd_lG+A5zS&j#7T=H(w9V3m|fe3JJ?Xfj__mMtys!G=ZGGM_=nJk%3!_D5^KqW1>QKilr))^|S||2g04dyntEtH1h5 z+v}U2JjuT$M|-yIIpl|1{;mIt=F@0A>bK(MDrSsJ2I|>1R?hyD)_*_xM8<92?b|E= z)_*m|I@SL~>0R>l?MHpvc1u^|?m7ARXRkr)x7#Y~JM&5WSM7~Muf{f}e16jMJ-Yj9 zcYf5DEOhhk_OqPrSGLu+<)!zeTi;bVe)gW zj-T>O_olvA)m}`eGd-KucNg6;dgilqY@Yj-)5VTG$!*tzmXoz<`fGboS@AUvP1Y5S zuW{zmTDOO)>%*7K?Z;&6^CV}-d2w2-ow3>8(jV_#`zO1i?Vqjl9KYoj*Ujsm;gA25 zU!wI@k7{qO>^PrnJL$GHJ5HXne#)+%ZBgyUo_=WVXlrnnuiN|CM{YWrf9sBF^WfW0 zW9qv#Fa4+7Vyt|C)^?@yl#x8`=X3U*HrL-A)t|_-`IBw#x1RiRtN%gUlXL3YwzI8y zMR!|1Mz8v{;SXDOc52%)8?$*&9{ToLpX_y{^HYT4K7k@k_Z4BPVb-qSx-|~{XZ9B{HY`XO%_oV6F-{txl6LLG?R(5_-!>_y_oVSPHhDV+_Pq7i zx8v`On|RK#$=`Bz+it&#YcacH`79Uzx9G`Vee>7W-f4ecwZ~H*IomH=ZteM|`877W z$0*m1o_*@4+~h~=zeVGVZoAUA$DVC_+~zyW@wMLSo2UL;bdS5@sejYrx;VdUFW4LU zJ;``_Y&1SS?SFL3Psb^rKd$Dc&2uI9)X^Tt3s+C`yG5_YWb+=A{bFLrseRG;)|pow z)wX|MrLy+B{-z~9^h)mDk7t}#KWQ&TK{fSE(ewKw0X9^t&?8$tGVrITS#1Pr7w%`Md19-Ti<+JJIBB8}~X?zOwZ}+sNS8 z?PujPhIYMpM6dFm_RYV`>7jE!cg>}~=pJ{LlfQK;*M8D-{Qc_2pR#vdzb9K!^~PQK zZoeh1oW2^m8yhdw3`5&}ye6s9^y`OgZ)}Gn>CY{-o z+)eA-v~hdh`uDiKSGDi)%3B6G`IWz~9y}i7lC||+>4@)sG+WU9Y+YN%)~kKjD{J?d z%Z{PS+VyYUr+)n&y!%N8dKJTGn>ya}Nsji%G(GL#c(vTJvG(no%KWh7vv{Z-6eO4%IJJ+jmt!wL6`<55aU3Bwb zl^a_hyY@R_cGiE=#@~+`w`F^$J@x1h_{rS!-*M6U6bCy87JuU6T;KSwVkQ|ox1M*B zE&G&TjGQr1JG$*r?`dx5I_Ho3(foD#EdQPBbFSlE^~Rn&w{)fR7B$!Aoo}jN(R_oF z>4|UqG+8H&cT+ORG6(az@^A6k{RCfgHr{j2QQOv|tZhHLpRGNcM_Kzhm$Tm3TQr&3 zSv|VP)fVrqbLZXL~S=fUDE-B)wG(p%e}OES_^ebeTa{n^c5J5QbEa%cH+=ftzkSn|?= z(zA6WZ_{+PcALMx=;o{qcz^t zI$GD(dVe?GW&N%BmV3tLj*-g6?k&3Q*|K+e^X&2!ng-zqmIUzmH}yziphM-N)g z*L2Z)%ckaLJu6>P3pJNtKAGPm3-{>E${G$wtw=*~mR@=k5W+Nf~Cw)*^$+@D*OjdTD^saea$!)E#`c9p6_nda5mw);8 z{HA@rFXF#tyS}|nr>y-u+4b%9+`7_}y;t;{)8?t~7TvsemG}57U)Q$zDkt~!m-hXf z^<-|^dhfWcKUw5#UFom?q{fpcf5FYMU#a#$N1{o?OTs>bKUV*xjwXKkJ$b@MLX&|-1N zbg|po+>eq?58b!y(%znn=j6}Fn|8kJ^XGiOb{56g*qv8ui_)J@^ABns`6s{RtL+1P zS5LB%jV3q#cpii{X45@x>)GXWX0tg|M$c!gd$vTBPrs=e~2+HcvQ&r^SM&Oglu&nZ)z=Ny|1 z?T>Lh_OE(<(R^3^YK*r1#2eL@o?AB7kKV5A(^rWlx z+kR`V@!T7*{>iSI@SSp6vuN`1?PtrVy|!C=$T+d(@4niz$wma@AK`>vKhmwS(r@mWfRp-#A<%Uy6Hs3`XlMJ-B2cZwbpZ?Q#)$6}S@#Sab_)hz3 zNAY=#vlcrhw%_ATe)O@?Pt&z!S5ANRE1H~>Uh$r7#$^9F2lHvK>^RT<;&0Epw)@d! zoU}P)XZ4fTkLIJ3#*lQao6M7=NBE#;iTuBsy|VGZ~x}IWL(Lt4{hvTv)VRa z<@(~Uj`q-bu1~(+^z_H>tBj{7`6s3Ovv+z(@t$NUCrjCLPtp1(yRYnQ&qbS8cBA__WtFGyNAt~HG#l>XIs1&+em?Ey zkJFEQn$Na>&vHCh^yIB>4a*JDVr*0VavF0#dOsakw$tI+va>tiZlC94Je7?-b#Hl(X?SeDm80<`uQlKD;zje> zZma!4=11;E`cdtBPN$CQ*>J`{{aaQ(yo>HRv!N$>Xl>i}?7b!1x^+$}pM0(Brg*jw z()&c{wmV-v2;K9f@5&e1ar$A`-p|=r9Z%&envA>j;o1FXxj9GAe)Z;BIo{f{17&9qer0q&joWRN8@uOwOUAat z+EgbiInm^u`?U6G{MoZYGyt9m#>|~i|XVZSC+hu<@u>Tu6XL^nE z(#}D$vSsU0PM5LCGq3%aucx-hYQNfF`E65v+vAMiWAkfmXk&I;W#h9E&CcDA=hhP+ znr?L4tbDfF7gz5Td%v7A&U*T?p*rf>Hs9i@{fh4K{Bq*-Tm5@na(A0@?Pz_C-*V4# zJkiE(d1pDEr-!D$eNg?Rr#*DlwrS&%w`u#m`{<_awf5GzhskMgp7d(3_TSvPl{3<_ zZK%9$X}@p&^zXTC`;v1-ldUf~CnYOCqUqUfyS@HB?y7B%NoO+m*xL5G-dgX?QyH%q zJ@;#ItN(n?KC-d{Ro_qi=+<+VRBl)E3oO8TF9i**bQazRsA=AK%a_?>K0Vd;DY4 zpZ4!Pr4R7W(?fcgfoFl6kIMI(C0;>DX&VwkKViKGyTJf3qz+lC$N~wd<9$AzJ%M)A2N| zPqVG{u?Cg%fA!thT*%rsXG8R~LH){(+BV&GDyMVPY}>T{J-)VR^7gpm?^au5>&Jh_ zetpr#T+wacl|9M0qF3^?CA0pkvAZq4^jF_JmGjTJ_aCIbW9O7teN%R@sdy-EPCM0m z;(bth6`NOW$xF`_&2J~g+gy~Bb){#wo%_eQ2YHfz7yp%wcgeo8 z>AlLi)VG<=s?Mj))A`l-XgrP4p04cO@pjHHIVVj9nw;HMd-CFmZk^iBvG=QQIjy6$ zYu$GYDc4>+G-lgT+f@vl60yY_ig*O4VB~F^6Bk4bsCqa~&8b3eIf%%?Y5xB9a?J)72-tf!YI=dL+yo+}#~yZKIg>NCD_ zG?}eqwDvuwHQ4m@Uvg1T^3dvN{mMOi>|N#f(d5>)^_?*q&!(rI-M7nI?paPxdv4SE zqK$nJ*_)?w+sC(j zm%lg7pN%Igxv08Ff4p~%Pfql)(T~a8w(s)RSNVSZC*Std*|%lo*Vgo;_3tstJzenqf zH=2wq8qcPiZ?`8unqKr)Ox(q@$85gJyT9CU-pQV5y}!pe^U=;tyS(|+#g=T#)|1x1 zX>&N~=DACbKEC&~MrWL8-)nZO-1sYgI!-&%o4h@K%cwoOP;tHMmE*ld(}fmuryaF# zUgcXdYd`sS{g#e&oxVBi@g*~zXmda5=D8}zUthAfuKLjU(eg_9=B_*z@2xdU?)`Kn zC!N(#xyflho7NZIeNS&WS!i?Fww>j8(Er`fod5kbBWe|7sbn0`S?NopMQ-% zpRND2ee=c_)u)_2+m`h07|}*P-|;2tiI*&O{`9Cf&-(H=-KSpP@%Hz1W0Rd7eP6xm zWB>jZxXZ~sY4Ye{Bb;`Tn?C(L@!z7^ev6U``8_@9YTPNm_Gtatbw%T8OtkiB?b#Ns zebdH7Z}si=2Q8-;%|ClRTete6dt75{dr;k*ueqV;v-{5SmZ7}I-`gCI|5NTc^H}?& z{!JUd+s<wHttG)a?f$e-e+O&H7AXyF{iBbZy#jK?o&24Ih`HVx7@}h zXZtk&6&I*D+VyzQo@Al5ZN3L7Zym~wN#0!)Z?;~={g$2F=w0KkWL>qj_fDGc=(>8g z-zuMc*?B*@=XjRyH+JjH4zyU=@l&~&+x7TQ8c*?v7W?gqtM35yq494S$-Sbd4X3>i zDks^SUs?N=j{LWMdivk`?6%FToQx~^yDh#=>)&l>Ii5}L;yL@Ov(-GFNW2;2xl6|O z&0Vs0-yXBek7;|FK6>}OD`(IBX#D71kN&IJsV$luc4l|;s*cvS>8)IJ)m|=Ye6p)= zT3>p1J%84BYg~LM-LlT|=G}dCG^aDis@I3Fami|p_PX6wjz1c2^^?|rKiYb4dNqdp zY>#f8jUlHzq+I(+8+*1lzPx%rx^*>&ofB)@dTP6(du-*+x8*4BzFXzwYI z8t-nay!*DU-CsG{nCRKQ{dB)_evIz?Q(H7y>5XbXdE(tP{->8V&rP?_&hn1U^Zq49 z!;Vj~uk16A^LbD^*wvGa?K8g1@0*_IgFd|Zl&s`#nvT2Z=C2(8mak3!?mKO$FWUI% z=CAFEP~%VJr_)DQXXq`v*s!1caEr!!_ScT?dF^@YJIAT-nB8$$%ox{mHNHM{_iwpp znVz$4*RvzqK8aq9k2iYCGxmPxQu428a<}g0n-8|{E9x(v?9y*$5YiJ$LwNY;kTc_>!Y~DI<$v1Axs+|7hvg`B#nNRd-zIRb}-||`Q zd+lzO@sf3yuI#_!*?pDKdyaVgpWXid&vN$NZ$9{LJ=^YcEt7xJD_QZ>ms~MY+!RNf z7RTK|(b}TBZ}*cE&;GrRlm8r({>EPMZ+Xg_XZPKbosLcO1Dfs0&6dr7zj9;Hc(;zb z%J}OqzeHQR))L+8Tlu7$f0s93`KLWw>|XKLmOWST8PC>HTjR=8)o;<}A3r%;?o~Nm z=$1=H{zbQJ_10^z!!BR(@Aa-+{1lV@o${vjUD3ur2u&xt`FDHs+H*hWd&P5W?yW_> z&EM5eKcBizJJh#tZ`YhQh-cII%RifTXXw2d?cEuz9bY=P?8@8D=2sk?YrT1u zlM$^yIh)oO-Fg6`ne{p&;FZ@%B@-Jm8{k- zJJ9SrX?@L!KPtzoo}X^f)<1g2!Iqo+TNLk>ldMxl_BGGyTW)RWes+6( z+u3*4N-s6eJ zn~Xi?Dz4*4TT^!3Z~X30M&nOPUUNgYFIvytR=F|8drrCct3NjCG2N9TYXpk@on25)E@GBww%iPvM;_(iYOk1(~s64|E61CXT?ePd2&@g z`IDKR>L)$NRX_RE>)TX&I+`O|8@eC$WZgB7E4le|#~InlW7pNVt-tbJ_SF~f6+PFK z-di~RwQbBFx5jGU;~%7amtW8ExAf+NE4ux0_V2!1<<{|vo?}j)z4p7WcC?(d+cs}H zvR%FYJ!Y5dk0t|6PV~yB_3d$YmFe02jeiij$L(^qqkC@3J@Ma)yB%9%skySNa#UN- zX?N{teT|FO{vb4Ydv4-jpNYHNUFM|m6hmmfB0D|N-S!}Mn%fnh_V(uXaqk-Ud$(w8 z^vZ{~bZmWBv~sa=Nv}U%$SdbVqAP zYuo(h8&Cb!@A5@`_oLgl`Inw--}I_In{Lr;-Tk#~+q7*S<@4G6%G>s<{I%Q4b60u2 zw!Yt^SNVR+yQ`e8P4D90^DwXGefmJ3C;q!+?!NeM(Z{4d_U{!>vdkx6MYYu*-Sgh% z#^m4X(bNChcUxucTkbBip*ie%x*8=J;yW-@7b5#=X#uR7+>_3ocd3C zCI42NvB}@_)3%>oPuAXBmG^#Z-`5ro>e=-tQcm8vzqgERJZU_;y?N$`P3w!__^n^L zd6wVKdUhrQP3|qaZMaq5W8*p3P{H*i z?ESfUlYJMxvTwH)Gv~TLz4hXwd%O3Lz90Br(0{YNZwLDD;NR~om6Ma)EBZujO-AGL z!S>5pzRSn@PJZ=0X7}&*TVvzTwrK5WZO!AHSM6xLS2UT~RsD*_6TRZO)z+9|SiSO& zcV+EQS3e^5EuoPVlEwQqmkDz^rk;@$PLT#OgWkJ_-hNQpxbl4R%J*nCvHN_^xzt~KW20NfRU9<_7Cq%|JF^u} z`q0g@+mm(Dt$W+3eb<$DY?lw72;F(<>f2D?@J@=iwtiLKWR4acCxD9Pj3D3qdl!ZdbW$9{WRvJTdp#Dqm@sd>Zh#gTkmPd z?!#ZbnBV(hm+OBJdNnRRo2I|{v{swe7j2H4*4Mb=^)4#bZ}s2mYk%y0bc;7zvhAe# z!94WgXAio0YtPm#ck_}%X6v)ppgf4ypI+2sY&0G+li1-Wu^wYu3%a!#OU^G1etz%^ z1AiltY;@;ZW&QYnaiC6ay2%w|jiJv&mp=9S$<~&gc*r83Y&!62O9meE$&beB-}>m!ezXCC~9!7~l^yFWSkfN$|W>)_c2e5CKc4Ssf@Y(0HT z@YoYSJ5cEFOO1 zBXaTZ5834Dw;p_fA0_jr20t?RF9SZ~AF_=h?XH_&EQu@jnmrF?F03(MuvdJG1AZb)JzZ?)Q$G8z1M?(@KD_$AfAD<+Wq#H#PS`Czw6oJZ`O!GK z+Q00h-@NJh(E)yYhTS0E&G)+p|7Ktg{9LXuR=@S*8}p=}Kg0>W#U4AXp}4gU`poBf z2W+>F=2{Gq!5?h&n3H~E$!41|;*I`n(9cG;Sx05E@-tcNFjsx-FlXQ+zUEK5f7^hK z=1Z=%V}o(@gE`av!vlGhF7ex1vD<#bBewKg1A7g`Io<3tAM)4@VpS}#*&_x~J}&O@ zn+rXBq0Cm$hmQ`qO)jB}9pV(MNB%J{{udAYXWa7)*lT^oo4BJ(8~>2UPW^ORFS=fE zzz6j6nfasUBi8MI;AeivN58cpOZ*lq`tb0#wK0G3l-=5`8~w%_N0xP?gP-vj$A@^? zV85_e-F*3l|H=_$Xx9eTgpb96_5OhYADS0?ign}hnG4zU%d>0|yX4S?kB#Ere&kd9 z#engAEw|9~g9B@$UEJa~5B^D}K0ek4>@sJRZZ;Z|&-Bp;>_N>{ESo2}_@6S66RbD? znmc>+o1Zw*M}KQ=t;B%+KtJ2X58shdY-gXf_#FdrN~ZNRZ@TzFJihRN9p++R>EnO8 zongk4$F}bpWFy(d8@oK_ONaTg$C{~U4{H6jiz|BAQ7q`=FL6Sicoe5%x_oW^?6zL! z%U|%A#~<^xSeHl3+4}jOum08G-wo8s=1X%m-d@$txAr36*mq)yK986;Ut_h22@jji zpDuGUxAw3x{7bI&WS2ht;)j0n%%A>pw%8Su`in0z_(h!aOMA|E?R0riJK8ZZ7OJj*V=vSH+0+cE@(VAdf8mMD?@VI5zQt^~o3F z$z9Ig`@aWl5hwPTxiw#FV6Xk2!S5UBXDj>UPyS-V?;V)4^)1g>7weCgo^KxT1>3Bn zyFXuJw+y1eDFk`vkb#nYUc|uLts?vG$3(zt~YWM!x)y1My(a z;!Yf}oh|IM#`@^-SQBHqlZsh3lfh2DW_P*EJnT2V0%blWSMCsBd|)p0(`|3@ofyL_ ze(+lZ(&svC!>}8v}d-Z=BSV!~VrxzH!$iP`gzj^Wj`Ft-Ah!OE+5714AxD~(jvV#wd zVW0J5JK5dOS_^#Uqt98+AMGb=#4bM5W>1S<0DWO6V|PLqRz)Xe8GM(W9``}=aDOBt+yCvi~aCY zgBKq>WuT25YvthsupZ=?pR?Xv>~T4OUDh8JGx+%5T-a~^e_-&cgFiN~&UnOtxr+y} zZr$+nukqH%y0gu<1F^{O*4*AU2Rv+Mqj6;AcYWH`|Ha@7ABWu2yPdK2jx{td{^cw3 zzkTqX1O8%{y+fwQx{5FHWRI~;Kf8++J`~GhPfj#%^WiJ)e95ooE}ziDR%<7Zn>UD2 zHqv9BD2)_@SJ^RyNaT z@072_A72%V>SD9~#&7&=FUSwBr08eENq5V)%~_USVK8$z!j% zfxYMvBW%~k|8gOpI4kU3KHyh=w0F(NSZ!q5OJWM2b+cyt_-zZ{^60nR^DiGOgV_G2 zfwpfQv>x_8y?kxo@HN?Vv!8xzU`>=%>@N(fsb$H z%D(xDVL6C@|L_2P$$>kDeWPFf{~U;Q_YZrjy<~0qnQz@G#G2SOKXX9!H+L~+ZR8tg zkX&GWjJHn?_MMqi1B|i__V?24*uCdY>H!h%o@AjbeEFfyw6w*yr})bo*wqd z8*&W0JgB<0&~IPzFJJtT!OIQ!qjQa)?Mr^n$L`O3^)iD$I(W4K%D*o$;7j>J-r)bv zSCo#|9K6OrpMAof_AB4?f+XLcS z-sp}fzQnTJ$oKMqHf6EQ-|Yqd2kpl2{qG;}J^#wj7yc+o6x;1n`+)~wD3-O`AE>&0gy-xtmTrAwKwguRf7U!k9c>nUj7Y)R^d&`#&zGfi4%k|bt-5Bd>O{{tP=m)yRh3B5} z^N+LDT*~3%OTYQpzhazU_z&fq{3kE-xx9duOYm9;KE{t)Gko@ewX!a)lYV=`{&>ZK zeSwDz>tHY1OY)h!2OG^p%!n~@%WraOKH&>>PzEv0=X@_0SQBf7$_dtr@6d9EvUs&8 zI&-ZLIbz*8BfgE3TiTCui@4xt=Mw+Q3+$IG_`G|$`-nDu@&-6Jo&DmUzsujR^w29k z^eT@-nZKL?&PzGmI#?I6ZX7w*QCzaAyOlhlO|AfCcTm1%i#*@`r#lk6iu>|QF=p;N zrup5P$S36jYh}Olm3++S@(`Z|b?fSkAWuA7H@-)$Lq69pkB}$6t)+)w^Jn>ZQ~THa z%m=kL#+W}}h<$e)?VX`~0P?;Z;m&7Giu29`{ML=%$){H?BP+@__vmai#+hOt*kfcX zlW7fhzA;C!Y3_Vm+{@8&z&8w>iE_l(4!UdbqcQyM+^}xqv-df9S6z;xH+m z{A3@BGk*85f$!+S*L%Np;!F0ht9h}X?ee$0;fyNRc5je#`Kr9jcgpeuzwy7=vo>;U z=cRtJEe6Rg{_F{mPuNn9vYzrrzG_dKk9Y+>6kFzlk8gLr>m8V1#1uc2gT-S07k};{ z&c-huxP!Y3Iw!?C-@D7WgScPtHQ%zEj`p5%b@Sr~{xDx>jrinq>%-UVH!m|v3_J(zUOOq8vf;H=U(sp@4h0J$VFn^dVq2{UO(T9XKTj> zWA|AsUihqd6|;QD=lQ=lHfQDP@&|uHea4W5YLipMYVRj}F1K`7>5ftkLd7-6<;K%z zO|3hB_YNR_I#=`4W7+1J@5R3O=VNilcjjF#Mb)*-DUg5VDE{IGpe9(DWj!-9; zFQI#ca%;yn`-vVoAzk7!-*z_cJl;Es`8&_-``%I1<=}k6pZw|^v==&O>7Y+8V1s=Q z;FZ9dEpKpUfg~#ci867%_8Z85~XDwR^=r|2q46FPAUm59?r!?IZHcgHAr-i}t)a z-R4!UGY9L#&u2_oL-8ff$kva~BR9}f&M7z0CAP({{2=z_V)r;_q4#a)AvpKE`^eey zw{iT$KJvv!=brd7Pv?z0f^*6II|Jnjed5s`uwU8CUh!s~Jmquq+WCfjaa{fu<8mzj zk`2yX>x0jFv5%eN!~QR>P<3k@wDG;YEC$R|S^JJJ^2CqzF&2+Jnom2o${)4M8GQJI z1O9fG;Ct<2fE<1mpL}ro)qKq@ROf?z#^+zLA#T{pPVv^cZccQypXjtd#OApMaxH(- z$G`k6{;jcmF3;E(*2;SFIT`$9kI`dp;?DftVa1v63n0e&jw#3D<9q94Je$meY%x>r z?>rY5psvrn#guW@$Xw~SruJ7**Pc)4V=J9}L8pCh9mu3xOo6#Vb$#u{y${(zMz%Yv z#6EvPw)5{s^N};miJpARw)QnYi9InVj+{??4EwCI&-qmjFlTe+o8~ERwBOjr7JJ#6 zgR-`LPRTLihArgrBl&do&BQy5`*h!1t-C()Ze5J$XFfJZ^64qode3&>a<|&~lwZV} z_$JG`mfP$(<$b@oYpuyKe=<9tQR~GneSCz^BewVsl!N+?dGl=F_f3v}*u@_E(mEIe z%GJrDPh7FZzP9)6cY98Zuvc8j7wxyk?>(Kp+Vsn3Vja}!YL6A`<}7yYWjuUr&sbx= z7JJ&PKe=?Uqj+!M@E<$vPjir~eq)C^o4^|D+h>>j#zwY^NBi58AK1eVanG09p}j!A7;HbH>c)vzd4WuJ zWRHAOobe&>AwG6yvpOGn#0(p(g*Bm<4(n=7%pWf~<)vcSx_a=i2Os_AF?D_7YOjU3 zHFrA9g&k})mTf>EnZ`e+~ z%(wiC*BImTZRiU`uWLx z?Kghs|9#e_SG&D0H{k0$5Z}g&FFM78JS4ySuHico^qoU)3i@kTHpW;o*k?}WCtm2J z8_XZr#J+)_{-szBLz} z?O`!Rx4KxsV-MMb)`=h4$1XOARkFpiar(tn_w4R6<+^mSW6L)N4}EltJH9awzH*lG zrS<1GbCR3dE6v}0#X|AGCwSOPC;QBwZu4Mw>)86Cayb3QnkPQAy^xRduRij{NB1iC zXK>ziuE_o78q~Y{@G}{7+Dpz|-)Q^{xSwbGcI#V>ceSSs_@sTX_aOhU**wV?*WM4k zE4s`3*^D#T`p|1#JaQx7n~!znSH2*dK6ksm9kET>ozFW49=z-)hdeP+JXfa6I`aYB zx)Zy9$xnPDKKVk-W{X%SmmGW0I+C+2dzZ8J$Dg?2GcOIBhTR zk#*;X;+&arMEa21InP?{t0|WzI8~4^*U_J|An-5kXny?EJeF5-$Dezf2AH;V3+PnnqS9?urG zi3Q`VyE#~Udh&B|CYJYkD+bG<>|qD{`IxNkTH=jA_(i+5U@xDULwPfwlgr0qTki30 z4q}ll)|xNOoj&sHQS#|D7y8W0{PrWhvBw8|RKCkLYrw|jX;(B;uo9hxuqTPkv{+_y@L{gZ5&;*!G&TvHT{k*uB@xTCuYj*>PBWkPF2+dD(}0 zS{G}w_lYqcb1+Bi%V+HKu*n*Bj-t)OnEcDH#y5ssV?B7-FNg7McO&s8x1BLYmh~wI z>$eWt*hRlMu=edaF`&)-(v7m+e2W`<$NHRc!uICP=DqLmnzuGSKliV_WL-kJjoosH zSn5t=&Y*7JHFxn@jM2e&^tF%JU=E#Y=56mgLs9kiese3H$OAEedRlLKv)5i=4=OKf z=kxXss;=F7S|3!KxFM4c>r>i?%KoVO%|Erfp3ck`;za(&h~42*IwSf zHJ%*)U>6(gP3INgxvM#sozL$68$URArtWgan1lIwte4zS9u?p8T5n(w5AYbh8g>N?I%om`n z?L1%D+Zx#qV#ECC<9{*kj_6$Vt+4ae-nV9K6=!0}SU#~v9x-aI+Pid_Ka@-L;V}yQ|3y#gX`@A1~X<;Iq!Z z_E+!U?!|tl;C;l;vfP!uFa6*^9&?{zuXSb=pC#6ua4Ft@yCLnSJzU7u)g3 zY1S9`R9p6<5HG*;zcm5t(Yu`SWRa0fYlt7>(?_PY;*0hr3g)7{K0cR&_FbSk^Ru~S zAKOEAu(>;lc$PDr9XofjiHzcCubVPki+g^z7VW*(t2JS#bqU#R&w{;UZsgLNZTU1m zv0Hy@$Uaa9J}jR2M;r8i&8%-6z9XeSv&!TL*r;`tqCK?cjgE-`nq? z^I1QS+`rq<_g8CeEz5r=t*?D1miC=N+=y#&B!*FO6yMH~;!*zIC|BE0_M2Gm&gHJ< z8v~ z9Y#NA@Y#dU9Qd2Fe>C`qgHK*i|I~rM_k(!;!Tf#NCk{S-@Nt8W9em8-qX%eE|H#3I z4nA=3euMWK{I$Wm4c_%}C~N!c1O1?U^1b_`@BZj}4&Hb00fP@2kpH&_bbZv|BL><( zXz;fl#|J<9!H<5}VDlP-7vJ9)oOWdseeXSZp8=X}Y{3)QdfG-lo9F^%HfYZlzWcj_ zzd!h-0sr&wCk+1H;O`8KZJe>{+Ks{A8nr&!tPdHey7jZ(esA30Q+?&&s~?BIVfMF& z{>D)J==1j!;>+Lt_+0_tHhVvCe-pFX+~-hr?+wlvu`b8k6P>ZnC-=JclszZky4$(S z`MpiKvA@Zb>jwPuRfDe>@Y$CQzGT3M@I?bY{rth_4nAw}j|ZPJ_{72A9em8- zBL^Qk_*;YbAH4U#dcDiQ8ou4&FAUyd@Fs&lGjM)6r>wuT%HN|qH^hju$oV6NoWagx zXM?jl z;N=E?WblUvFFW`{gO?fn!NE)2!2kK^V|d9&U-HqH7`*u4_YZ#G;P(t(Z15t37aIKT z!S5P8-{5x+o_Fv&2ET3aoP*~W{HDRP4xV}NjD!DU@M{K79z4A8q(`50uT8uDXBhn2 zficFC@oWP!fAfI+psxMd2Y7-y=xf|A8~0lV&oy}N!Sf8BGBBSjJpZH5e{a6#ZT>Gh z_`QQaFyOV99`G@rzx?2p2G-(L2Y-Cjr8i)bsj5EjC=?v}6bH+L!W?Z#kANK zC9tf6(Bg2KN6a z4gS&Kvj_ilAcuVIK(3K{{q?4?cJB1%odgeC6P42j4jOmch3TzJ2f= zkK;QZ{?*W~|6dO9{)@+9>^DD-Z+iHf?#cUxhri+BuX`L{{qR>m{8fXm7|{DggMT_; z%V!Keb?}LUj~jgC;DZN$WAN7p?=mp2w;8jvL2 z_@==(559Hq?Sp?c`0fE;{lMUd2R}CWsR4iTEB`uIe%*lYt%3C@-#f45^cNW1%IDVH z+RNk4*4G&P$w4{X+4}|qIonzMMgwu=Y<|;$b6R}8_26v>V(uLW?>z91AeVdJ@NObT z-Q~SQ?Yq49F*)3OTlxD-=g!{U+?_-0``b@<5r6;bo-UXB&g#DI{n4JWclvpTZ)Uz> z^=-%Avj@G`b{F?O!uz%NdHdIW+qVOI+a23}m(M?BAQyb{z#Y%MPY!oy{5J!+L{5=g z+&$fApK0)%gQpB$Wbl%Mmm9p=;B^LXJopQPcO1OOK(79%!6yvl=r0_6%|L$r?!gZY zeq!)TgC{@f$xnKg!EYJ-&cTZgUUKk92CqE$6NA?sywTvz25&uhyTLmS-evID2Jba^ zzrhC#K6vn9gO3<|)Zn8BA2aya0jmDD2l_u`V9fgt-ed62gTFNRi-Wfqyz$`m2d_1F zwZSV4US{wI1}`*tp22ero^kNV!;d`tvx6TUeDA=S;+*&=gMT>q*ue)4oB@A%@D_tN z9K7bh`S(%-`Tlta&pwdn-T%FFI8Uw3HxAs{KV$Im1NZRv9=LbE<-qqO-sYX9LPWaVIY6YNB@5Ce-FNE@STHi8_4DInLIAn z$>E_; zGU&sD_ag?z?z~SweSuwaI2%7{AeTF1*uL}m=RNXyzmm_t{E;uem&=`D<}ZJL$H4ja z?*{zDUq3Y9M}FmJKL5Fa_3(Hv@IK&O;5QDOt?umdx7_W$L>zdp@vd`~v&D~GEvDq= zw;cS1fp;qJRy!xVUyD=UgWR#@<_{nE?y|q($k!X*;k?WF9wXj;6LSag4aXkv-Q_a| z-N*emZN7N$Wdq+_+{wK^`W7RvyPwPL-Yvm*J^RX?z5lkE?|;4r`nza%cze%xLO*Bo z&Ct)?%Kh&1zOTDi`5x)xfV-D`@10n#_f6S1SMS#H{M!!Rec-*^o!z}puK2=%_kTIw zJ@E$yz9aa);a>jCgWo!EC;tP3KQeG1f4#w*59F)&9Q>_;Jod?h&l!CAz&-ij4t{v> zUkCEwGd<~_h&136eu{`kSC3_fG3m_=EvHA2fLHfn5K#gEt+#-r$c9 z*y-$f;lb}1{KkRv=@%aUg@=D^@O=aC3*Is0^-mgn_~5+;?(}au_)~*FHjvL>Xz<$x z?(O{S+miR9?;U*m;OhpyxwxZyzj8Lq=YMtJ8;b8_?&Q|kdzx?QzK{5Ms-K0(>te#U z5bt_^zEW=Y`_t|gVo!{UUEhwywRcZ(Z;#kB-cR3u;QOq5&U*}eBn zhue$xq}*?hx{LR1!+W#4*Z#gC|9juI=Y6a2okAYicXBzxJH9)(v%opv+d}v6Uw-75 zANi$$T=O#nd0q~Z&;Or+{Pb@J?wfL#duTcQiyrx+$#kE0FMrG3a-+MhJSwlsx9>N2 zufe+w-f8fc2X8lc>%m(M-hA-q25&rgqrsmZy#C<-8oc)4PYnL};ME4NJoux7R~o#+ z;N=G|cR-uIR~+E`V}n;27*7s)J!VyAR%T@IHh0A2>VA z%lYEGaXxgXcLq4eyofg$Gsbf>)s8#bGqxe^N4-l_3Q(8A$!8T$h(3&l6~X7wVx}z&;0!1 zZw~w{!=1{$bH;q;!1o>Jgun6op@Cd)FZ-PY`~2kx?(lNKUmy7P>KnLsd%4`V7T>jf zD|SbhYhHNZo64&UUUwjWyO+QBK(2Bp|Ez&;CEq%5SN+d}pC9-L?Jh2Nzrf%n2JXD> zzwW|sKX}i?!_@?%if_X2~b44!N7Yys}!jpb{@B@SI9Q?DvmkikRi39ihzd3Nm zcvtvygV!0n%HX93-ZA9&-!$<2+`~Wj@Q)0>7bjSI-r+R>mMzqVH&yy`$+2mL#@zriDKaPLfU zzP$C|?FY`DcOLw;ft>yU1M_g_cgJ_9@7s)A?%hN__kHFI246bZIoOY?7p3W^#x_`aW6RVeck$t1K-_VZcu)gD<5U<`rgOK3*x(NjUSsfQ2EK>*4)Xp3-#g^L&m72y zUq5g+{@%fV9Q@=!?tEnMYoGjU@8#y_e)4la`Gp2AHF)L0>ki(0@D77_9sKpddkx-m z@Ye?KJb1^!+YjD$@HT_LIC$&9TMgc9@Mj0FH+ZeVs}5d%@X~`9AISI5KakU(Wgt2H z*MYNxtzR|xoPqPjcbxYbywkuuok8;Ws}A_UyTywPo^SBm2hTq6Eyks>oc>)8f7iXQ z_MiRmXWz@`@B8rkKK%9r-(K9){S3mn>F2QWxcj-Esrq@w*AIMe@g3dIK|gNb8@l%^ zKlA)613BHdmp2{wT^`@j{fyha{WTu{y&Qid=6kx|gXzEJ`iyF_NDup zd)o^S>~lH5_ZdIm^Zn%)ANj?5cW?RGH4joym3z1FYfj3Q@+o*o7v;;b-vB|Ii;WTxjV~0z7xyAzBhdL;C~FB z=gH6WA&AQSAOV8KlG%3JNV{-?>7H<;2!RM!T0nx z7`*zxHx}Q}|KPw~{0|LYc7VRr;P($+aNw+a)`7>niSO#)Gw`!nKXdg>{X++TWne8{ zcJRCdzf0qr`M(?ZS*Y(T`yT$T^ZP|_J@BpE&pPF0zi;AqOyuHc9mvPv=bhfg`aO2P zkAl~`neT6YJ}RegXvgCl-E$87?A`D7y2Jan=Vw0d2EHfC?S5wDn~Xby`sWS4aPTDq z`QLZOuU>HH58BCsc#JXD&!47re)NueX=n5A8SK zQ2cz)ebT+sJ=@O$|%k6UKw-4mh&tEzHBcA#ZPyN7w{QTaJ<9(j`K2QA{gAX43 z?ZGDwK7H_ogRdFJ5-QM>bIotOfFvk5K@9PbGH}d_)8OT@iH{W?bk)Qj%FGu@+<2#P; zQuyS1?*_fA_!b0wOQv@Ra=p{|E`;AZgm)d^w4A?xeej--gUojw;Im%V>MuSHa=afI z<9uP4Z%Cgs=w}GtwVW5~L0o|Iu|=1Ko<@DfjciF z-(~QogI5{2LqF5t=brR)Px|43Zxim^zJ)x;lb_?se`w%-{E>q%8<73q2LE&L%Y&aD ze9z#k2hN2L82r_Nd$s$uJGV1LK4!zG4$u!7{N=%G4}R~!_m-zV>8Vfp`N1y@e$A7A z&6A&V@FD|uaCh;S9r)5Bhr3j<^IHer8GL{D9o~CHa2Nl$f%gkPU-h$A-&Xt#_Wz^o zKA`2S%CCVlb3^ZhCXs{y(yI~_1QN<0=|~e0NCE;%QBee?CUgXmCcOouOYa~hp(98a zq$Yq72tD-V&hYKuaG(F0m5;T)^<}Tc%$<45dCz;!e)hAU^A7rNw}8j8eXvy^H(5Iv z5-c9zaUS|}fxO0=7#iGqA8TEf4VDO&4tOi82Wtji>GN*|#HY2K4+S{=vcOs-uhuJy z%hv`M1t$c*3HU5S0$N-A#_xC;=UeY&?Mtn5I()jrJ06?;tX<^0S<9}TZVik$FjGLI zPYjGH8ePo?M^{Y<7yC@@N4Hbxs5jC2qXS$m@1*VN zd1KsoS6AbM(?R$dC-VjH2F^GlFxSm(ynuJ`6t9`SUMawFH0R&||IQKMLcI8D@OW@< zaC?A5uMY6(1py6AQ|}-ABG@t5CfKqM-f!0j?>7t935Enq2MY)D2mOP2gN1^{g5`rB z1h(Bc*gDuD*frQIpwE9F92^`L92L;!borUVg#j&oYoP7j8vOOv>#a9~_X9e7YT$Zl zar_T62K@ptNS^UoAlLXfu#b4*bN>{&dII~~Z`NS0V1Yo+qF=?5MT3FCf`RXy$G*L- z-cB1F%W>>Ke_;C+f*%HA2i-4*;?JW3>vV7|eSTvgHq+?xFWiZf@wD+kr+}tE zGdMjsAvi23-rlum*Ut9)#WQg{oQ>a~4c-hs48#EX{KbIAzdfMA&kr2?&%qx98vKCZ zSHXw?-_znl0$%yxpnUS#dS>gHA&9lqGqpLj6NBDwzTfZRj?n}f>&^;fl6 zdR={X=Kzj;e3V16YH9l%QA6UdxwQ=!9u5%92hE1Ly8t(Ag>13o!j{;%Mk z09UJfUKr5l#pAdgaJuhK2}TE=5!fYA6Xmn;T=ognORa-BHQ=@LXMCm>zE!Y#pk7MX z`-i_L2cAi@UXd<;Gr+$u1WyDH1?u_5>-09i-FL49*4WbY;sL*Y*1j67`s-`KV}W%7 zw7scf+^g*YGdt7{04qnC_Nu^HTUI^ffN0>-f_&G(Ae6@<$d5XjlA! zi*O50`7-z>Kb!y#6}aBN!Rr z_$`7>1026qAQt!@Pm2+_`P_g{^0$DFzd5)g7#r|xX#A%G?RqH?f5rzBf_n8{`)T)g zg2@3c<_XOb%pKrf8Xn)`QT)4jP&|y2LC#V9EC=Dch>5ENs|I4E*C7Ev#&;lR@w#%L z|9w}mukGbF?vI>6&O`Ga65veUigh+PQ+-vQ#lzsG@iWA8_uu!{Qu5VsY;n6dZ(KMI z{f&ch_jv)n$KB4O?hD@Ie7tTu`(6{Mhtt?&18aSFIlLAgj{1unkWbTyk!l^!d;!nI zI4#cvmoFROZGMUR5bbSTTQ4TxJTJiCcLj8iIY#UB&$gbbLNbK+ikxN8xtE31{Vj12l#tfuy`;>VBLy!pw=RKKT9xgz%z&Wf|&x} ztASeE@Vak>{r1%|IOR+}8O4eplo6oDf|O-t*f$i(oy>?*i-9M+6%N z>ZZmKKF7(lHqPar(9p&m9|ecg;Zr!M`lt2e8FF8}6KC8!Dot)a9v;ZYCA4eiNn7L zJ`LzrKJs(H!vU}P^5E=%H;rqHyZ34B)6wN4S|d7~yg_S&)>;Amy+W{DuxuY_ZlA5# z2Y>r+nLhaYyTK2G9|bh|&jRt_h=A83Cy*Ol6I>tg!tuPnuMBX%ZM0RK`AgvStYA!F z-)jT8#(jaz2cE03mgKNN9TE@oxo-{l(F^ojpxeZqPrVnI8*i z{G$S%3f^2XP|KVvz?n2~|9}U-LcoXrNuXY-rpY%sBjB~*ZN51loFC3R7i(S%=yS!n z*#hT>*#n>PY~&gD1^n}K0y%(o(d#@Meh;69rspBi_paUcG(K%F&yZt?4XXsaq(uX3 zU&I-(erCR4ra&xsIk-CzC#`=U8PMw1E?T!ZwK265OJ8if*m?TNGf{U1yz}b=&na08 z#6$l>uuHIU@cm%PfR|3U4-VYRRRSLS&|r;#Pctx33$X_2sQ{m=mD2OHJAcMA2G<4F zpn9eA&kBwR_6Tq}e`Wf>8W$eORe`$834!rwT&lD4^mrnB1ad_DYaCk}rk<&;NmHxg z?;Y63Smvh~$DY;Th128esJ{(tyGdZ(fjm{si2hM)x;3D!_$fS59xiRSbU<(7XZ*lt zz{ghvxJGRd&JWZf@iczHhhA;xbKHZ2_}Rw>JdFbay89QwP62(rZm?>Amlq8d4tOSo zZ@0eP`D~HaA{`%P$<~r#?d4mm1ZxI(d#iwESLZw=;G5uMZM!AFn@hO0V{RJ;2%a=dsKl$W!=DG&HTuTL)U(-(n))W0|1p0`^h+ zrSIi6^uGA0ZI1tTAQzD1@M>_hcxoRxfcu22-7}uOwKn``K05AnU-9gj0bag3pwDSy z_aBG;CzuqxA9%*#qkzxBfBQ6`lRbwpHPC?i=lztx_x9tD(EIegx(c4QUWb-1{VoP* zi@#}g{?7dYe+8Gz5%IXO$`7Qq<&!4|mjt&2w9UT)c__b8PAXUR&$FQB8Q^Z>&OfgR zO&+Ue^__q|-!!1hjtu1L)nMHbCJ#PBu^v(4G98Z7Jqr8|+0_#NZAT4}J zK*!_0s{@?K6FED;lSc$R^`8g)7QQ-P9uLyw^t+tE+P{gxTY+=A80!O<2;>XSOPAw% z+FMPPM#uZz?|C>!1~?dRYXg7Xdu_Il_B+;Q!O(!;BA$q2Vh*joZm?yrbHJbZePCaH zO?g2(^z6{HX+Zn06j%efP#{K$ReXA0&4bmxZZQ7H2wU+T!A$)ev9D4z*?o} zgMS2f2IBmA0q^Ax0Y8RrH*Watyb(SNtw4L<7VtyP2#yJ?owATAeB)hGK03k059SO4*TuD}{Y^Ub_8&*ZVz4LlUw9mt{Oz2?*f0Y8*ajsw)w ztV3Bdz&kuz-YkvzLV(-(6s0He-Iuo4z%DYtBF%pyhG==E2T^bMwV%U9pJ% zrpakrJnxmC?=$g;UoIY98qmR|(`ob*0>^S)*2Flr*d)L3DhF`AqXY3v9J4k?-|*rc z)3NzJ&f|RY0{MZq{5=>KsLhH!;s?!NK0U4%f7}mogf~BPK!@{s_&Dkqyc)HQZjKkP z)jw(Hqk}&O_|^TtJW$WRGk7?#rsUP&tpLB{bK{ARZJj&b^(x2XiF}pyCI02Mb$B1o ze;HVR0yy8R&wOV)``Di^K6hYU6%U+$Vl6G-d&gkUz}zrLP77%7>jSwb4`qBndpn>S z(!zl?WioZ_pvJ{JJs>zbI3rL4<-wZIp7W&P%>VBN8w5KBzX^^BE(kp5 zZtVq4X|0R;-tU8p0z8G2@h?uKv+@2hSMr zZ)kbWl6c0Oe>(6C6Yjq-_*1ZdpuW0MV2#WY!2*Fbq5ioq>Z|Vu*2#EI>G6Qxx7Ocp zCL9v%9;ms}?=(HWR}1IO<9W{~@o8QSo(}wW49$K{K(F)E)l`e)t!?4w{Wo|%kQ?&K zc`p|PboL>EG41&vYl7vV3kEX=<{7R1MnD@`o6G}+=K}K*k3Sxmr`DhGR&EP;qVyX5 zM7y0I$fI#PO(z#uzrq7M1seyvSsIjI{#EO%j+gLa)($@0=cnK*x}Bdwi|-fg8Q}1r z1e*l3I4^{@rm>d_)Gh}G^!R)MUY{$NJ(w++HSjt|;JXC_+b$_qMkkF2=)m2KNSd_=&)}#nSRL zKCeYQqPbq4hpOvIVLzRP;>NpY;a^SI-qxL2Zshew-)@Y;IiPB07pI( zycWD0SZ5)XW?@n=NP9WQwF{2=gNyrAD{cic_aTljcXM^t+_aeYTd}nweVzai3;CIq*ypzi!`v4?`;$ zOP)1dEm$rX7|b2W4L^_1{!Z{}@KkVrAm21@c`AGw-pmexxv_jOe?X|HYvS^8!F7T8 zeQIFNoBO{FtUcqATkn7$tbJG~SRlX|h*MirTOS2< zjZdu?LZvhtJC#j1Fv1(lSbyR z(8D}tT9^)YPBl;dww&UrfM+89@KC(^o0si-e9Tkg7x9X$Z=tV`3AEEW4(fyVK07-2 zQ$YKRWA?-IG`}^J;+($FC;ts_IUeV&Si@pXI&Ccn0&9YN#(%-_K!4+U{jJuj2JPMq z4K@vS2!0Xl7x1R};qEn`Tb#ZkFa~Z7aJn4r$w2IXJ)j5ZXXA!e#_6=O^&`gA_XC`6 ztskDInQ<-;+nN!c-(dmXJvLB>{8PY7q(`hN;g@uC@ws{K|H9zXfcJPqaC<;=^HbFQ zp9@|I{u7u79|g3J^;{Zj!kH)5cFBEtDb@yJZM}tHh466E*KxY6?_m_pULOIfinej1UPX(AP-q4_-f18i4zPdi5Ue5AgVdfwk`4@VRXq zhY#ksj*suJ39bk(3C0B0+nf^A?-7WN)=_R9xRwuz=SuUhsm%47J_XbgfQ})|h!b;OX=1PYDhR ztc~EL|@qk8>ldtlQ%|E8>3v*dz2mL{pWZrbzOwZn^9JGq-6|&FS~~Z^059U@ zTl;`>`4VDveD{!GeD$ z2mEaH-B*K(1i?bN$W|B7x|*+jp+&cn-1rD|2Ux2e;SMoJi|4)4iTjtq_uP7Y2F{AQ%zWIR8R!+`hVknecrey`ECr}e>mpVz(@ z<~JuV39blm@U?+l$nOR4$oMT{*}Vas4&M9Bcee*O2KIFv?QxD10~|goP>aReV(iYr z4#8HzCc*y%YXx{){dL(u&NeXc`#t>wYj$Q1^rakN%5+mY{mtilGnf#_32^&;0gYjd zTo_nC`NsgyTX!QT*fdx(@LUtGk*0}t(pdx9oeSI+f8QDb{C`eBUzvk{42}+t2u25^ z0&`T|idO5!fqAag+A(!rbJ=s&>bUA$a%*`z4p=J?OROEnA)X200jSZv7Vrl=KQcDx z_Wm!-N(`K3t%SJY_1u6bd~x78`qHL%r*3e6@MwU;_)#>fnjo+0)xfJ-gU@}hX2`3; zd((PVf6LnJO9DO>?&Nh?-=PM1XrLBheaEi@YX))eZh?B`$Y6M|Qy2VgJ*e-+Jeu2n z_IDi5_TcPMfi+0Dn3yj z{}=r2^LqliOgzB7e&fsh6-OK&KjUfh*zZ36I*{YbGe!i~Pj4S=7ifRM-*qh`v-ZyR z-*i3V#jydesp6NMMSKxw_zdzF*CZD49`KAd-xavNv3+no1$^Ya?`*@n(jQ`rW9XlX zajxC5{p~y3+1`HRg8Ku$hkB*DHUG;Rm1hF+$uk7}u$KZ__w_)|^iCjFiBr}zR%}uu z;gyL`fMeB1@oXM55WYHmn=)0L^xm-Ul+ON@qltNrDF!;eJWgy>S9Sgu18X9+^N~Pb zxE9yt8hLpY%jt1(*F7H-_-%am{x5yM&tCWc;5^a9-{IREo9?;Vkn#M(fJRv=7!=Sj zvjr~Vi?6=u)R^AN{}aHR;X#@|)=rvBcLlV*xg}S=z6<`Qvu+M<4a5c8@-(d<0~!px z_qlCsYd`y68_2^i4|t>IGLK2_t}Z5yh!f_#90Lc45A+}(mNqPp*BUn-!Onp=Vta8# zjKLS>6^K)O0_Qj-pdHJn+tGtpR50gg|-zhwb8yjwt{%-Gv9$eDcI-G{}0&SzK&PEiN27>Y5y6) z*@0`SK2Vn|=o8oMJKN|l8o)lFAM~Z;=vUgnvGog$aD2e0DfrvxbuZj|$JDnt-54=$ z?hTB|XM&f4Hv--a9X~bb$-)pCo=%sm4h%dCV$9Ps#{Vk8s)3r;54s=*6uLjJ_r9a= z*9a<3Y>?Wk_(1P(9pF=Ofj_Sfi%ZQ#+DeR2m-XuJX|ZR1SH$-AGk1pvxEWWAJ#ujK z*tyMR*T7d=DGl*CuI>bD_)X+YxllsECfgHxO@~%%D^y+Wt z_gz7(1pC+*)(*7Maltv951d~+v`PE4)qU}N{W5_(%Du8a@H>HfH#k@%z*+7mP0CyQ zRv-tPH*mk@Y}4ZFjP2|HRvu@2+-KdBG2%Fm2lBakUw``yoOjyqYa8DcY(rycoAWrk zoYJ_QF)&7XD7}H)6ZaPA6yuoQCaV1N-gdl49;ETE{#rb3F34@IqZ}LHSlp_vbX`EB zTf>gCa2PH#e{1fVzqrnP_PIHZ*KB)5U|+{@EXNcB@VE1b3wL)t|H?a66IRDkGq9%K z+B9Bj!Drg+dmfTC4%%-&`@0T!IPjV8Y{Qeak9{5U-@%K4e13fJT7VDK&h*c;;8}@p z;QjyWd&l7u=x66p?{;qOz(Lwq?e*T8C)aHoZPO^5 zV#=c{jW9Ef9Y(CT9=C(O*uABcj3#Z`;Jc4iN zZtKkHS3V?u#Gy0}--6!eVE~V%UcIM92%&hj_v~OPVZU+NyE#J=_~Nw zni^W$evYLbw7+)Ydb*qMLrdfFkpV7u4fwn?I8GNkTqpj-!MGAH7rOZwuNF7s;o@SS z`3~3GPmIE`I3NG>ZS}4ApvH$|yZP00#gRC9bbvqk2y$2JR^+oh5uOI$nvYOEczFqO zVqOBs`RjeX`kvpw7vK}fo%tk=<(T}I^4Iy~HT-*OYqy9f= zVuIKpPKW_wi@4)=rugRirTRo)l%ChGaszdN@?!KmhzZle=d{20=N=V!P}8Emvn`ar zQ-Ay3S`T@O`|kepw!|LeqjCykN4%iPjVm!)9KiMBc(>LT-|_t{fpLq|yPq>C?Mipx zclwnUDULT^%p>!{9HHsVF2=^Cb8ey96TwUIl-X zhr#dUeeyuny{uug=8Q+GF2*k{d{dpw`Zd0(HE-SCDWAl7cIt!o1)r&@c`i_0O*{P# zo_1@$>u^2dkLwhVcoE&0wh#1$ZEeG!70W8euQeI`0N~wr_o;pKq4SDW+M*_)UEqDS zPYe{hoUdY*^N3sEJ&(cp#j|PQbKh6nv_pINL3|M21usI}<7e&BFL zCGHrj#%1LK#x@;8&veHO`duzkdYxtme?zyvul;=HJMlyQU@m#)LL8AVc6;1ngLzB` z(vftcT)x}O#v{cG-JDSQzwLc)8+=5=(zN0NAA=6Z>$s3M#*x;W^M=dk#;N$3Kg;Xn zzp6{sdKBKNMFFu>{d`HI84!kiSoK2wDUa4V7QQJD*E)9Isk)ZB?YRE=ySqkaA9c&}WckGY-4R_byu~i(u#fmtzHfKz63?`uJaF+!tkb_ful>X>e!TX$4^{j1J1LdFd{g|YJVs0x z@5L?}K-?0u#1dnNjxd(!3x1CA#mh0)csNT0;)Yy-Cqk=;A26+7QeH`UIrNQORBVw$ zh$qIi@ok*TY0Uxip*v@$vkG)qq4I``EAj#QQCy()@Vt4AycXuMtP^>8pQ zc3h>P~;M>z)$);8Dcy0r`6;&ibC?|Su_zQM^jTwa7{9Yg=P zR{5F!#g)MK$DP{coce;+#?#J)%dPje?nhh|7rm|#$bYOAk>iMCwikQcGw0O)x_;-@ zH~PT!d8N_$N&1f;#jg_o#Tao-&LdxQT(4jZcok3N8#ID_?C*GVg)vZZ0Eff)fG+8- zRaCxfdwH(BwmdC4tiIAt{X(CJQI&7_TfQpxxM%L4^N2OZnmiou(^b56nhc-ufsJ{e znM>x5+?_{DWAUKG8)Ml0F-9D}?pf6{#3ep0uSz^Jhj_}4k1NeF`cS+umvOAm&E-0` z_Gzc~YPY$kpQ`quze>NUeYid`)wVnYUV?fUU!mJOb1i%`?V=~~1|7=B(8liisD8l} z+9Y=36@D8pLx0rVb&l@y_)aX=fBF$W>0jQxYZj|;C~m`z{2|;3_*{G{9v2V9GckiF zQ@khk=uf&DkLf?YKE16!@h}b+E5#Ad5Ae~f4WapgZ$vd#8)v2-->6& zi&dj_E-_KOqxW$$4-juVmzW~=kju!Ez=r7ZE~ITh1evX$%*I*=dAqC+EaQ$-X)gl5Aj}`T?38e_{J^1tD2;-Mu*{L zW6D4C!I*JBUAvg>{PJ9J)^)h2-SYlk|9`48Tk zU-Y9n=$St{PEE|cchBiAaD4NMjuDGXrx`!Spm8l08l$esTs6O(gO^=Ai=)N2;%Xd2 z*NPW72tVs1^Uho|FT`DA-tpyW`itM-JYt@NJV{5JfB44yuK8Kqfv<2Bj@N#%n{Jmw@Tue#VyKu^@ruvoGcg^f%U8r# z`30S?R)GH;8|S!x>Q}ghUU1#+5&vEv;ePSPzI1_DA%5}ghV`NHFdm=lcV7LE*To|J zt?%3~9D~D*UEJjr-?$GngfT2$h+h{3bb-DVhaKCsl}4d2+}ny%JSqJuhvp%b-(-H2 z=j6J1v9yBN=6*oczP#!q=dUb~?Z-p3!^xk;^G#kX=NoR0&=CGEB^PIEr< zkjBNGp9b3MGv~lTViiuJ?H!lyuUyBgTtIA;yWmzaiMJ@{u&)?^he157nCNe@(KVF5 z*FF$40AF_3$9*;|kORp5)IVaSz7mhb1#1-C!{R~rL4NH% zi*3FW$HV~f+j(7=_*c2MYZb%A4(IUdZ|4$AplU?!fi}|@u1%bFKb*rkjSbobN0n~D zkN$~$J~uwZSRBrmqlfgdKGb(&i~FTr`h$LPjLd7>R zP3$qYjX`5a-xx>QZcT!FtzTWgF;Hvcw83Zkz;(N3`x$$#$@nhrr}0ZC<6wH8A1@}D zuja2gVJ_fO*X!OJd-84?$+~RdU&vLp!#&eZ{f8?nhTTlG2D6`L2 z)tZbOIWr9+-vrk*DiGIQ8(jkW-u~9*Ij6kQJqMaiF6}-U7ln%7+93XxHsZ6E{*(7= zlRB4c^EdE8U8C!;#ztOdT!FsS&*p`GHpYxupBaDJAufyiuHF6do*zIn7|Ysge!Azz z2`yq?8*`vm%Srb0Y2?zD}5@SS7hqM8RZa&cnCXz@j^ zYMkOLbHSWY*QhmAa!NcZ&YL4%#d`6;w&s&|Sg}XmfbZmgVhdeQzl&31 z6wbGun1Oe3gX23F?c!A%w8gpbIG)6rxJ3<44kj1#41}DC539x`PZWQ(0q=;XH7~Ut z=M^8|2Yq5Yaf&Vw%ggKY?8%sbpIhrEtmBY-%Kzk{{^mDYN5SK?zu1pw<)(Z}`77U2 zj_MqEQGTqC!~*@Ky?7aK+Fu)Zr9e~2|D2cKS+%fQ!y!N9d#c5$>#0M^9X)ShTikDd z{-rz}Pvd3zuX`f?=wtU${v~FK$?lW*C{EH1@?C31T)Td8{qk{rU@c13_2kgvj=Wr6 z?K+JEV?_QaHjDA%w3vmDX+dL}-ttO6iZ9}Stv&WyV?tZ$jq+-F&pccGs=bxFx(2#N zoN_$#g>I(Xd20Axeu49g2k{oZ;I)Z8=Ba(n9mlHvDIeJwrs2#f8W~UEFPdGRhRfwh zo)J^;^4kyUW9nc$Ky@RrMqBVGexQJ6JR-;op z^DLp?67uYt`kLPo;x)?O% z@>2P}`zOBXOZVKdyo$f#oc4-OV!z|)Z|~J-#SJ-~eA2avEza%wjU#c+cVdm4TAu?= zrmgP3SS(hGWAuPAOB)#P;)JmzHqeP)E4GM3;)?B!Te_CUqi<^6k$C6$?!ET8H|CIh z(|9+3E5`72Xl-*47vepf!Q(2Q4hP~q+u>(9hJWH*X>M_oZy^5^8(jlkjxTTr9>FK( zx4B-mC-DJSsoUT?+>4v|Xx2J-wT{l(1^lRP_i(`P^BGPRUvQmc$d&Llc#l7QXB)gL zE{H=sJ8Lv11=fGmZ_bHJ&f}b7mRc3~+qtdr@I1M7BK;cu8a;tE9GPWh_hlsrkychBTXcn`0;*ZN(5+fSSl7pgXA9Y^)QwKCc-PK%fJ6CczRD-W$2 zCa!gl#Ux{eUJ#GuuI_`eCBB0ACT7Y%c%;UcJkzm?o8>yjyKye}p!LiZx=G$Z8&u9F zK8h{kiF+?*iqHB%JLwskpH9Qow4OPFyUc6zU+zTb)1vr8JTaf?LGg~Zp@YPG{bo$* z6W8Xv>eFJSIg1DRZ}=6reV*Urd@Xo3csh77m>l?xDtsyyXotC~jzS+g4oyg3ng4Pq z?Zz$iur}jUe5MAqdLU+5bAeaYr`8YTD%RoET0GosZ2}GiaRs-Ec|1bDFXwpn^QwJ* z6GNQvx47ZF@=dQe)iH2-d8u+D+brK@KNBmSn79{tQ)b` zL|gcy`Vo)fPw__F#obk(bZ*z~cY8l+f71DXHcrZKxqKXa7JLxA8N3^aXX1-`omeV{ ziG^YleNp_3=W)K6p}*Amtbq}u#Aw&>Mc_AD#VEfYVjIWRr!-XGdNftCiaVgd~Ld> z*3;-$*QReBmo5=|Rt*7p7KGq!OotaZ31M}OQs@ekG!{e!#AV&h;sl1FvG>)}d zet|>qAl*rq;^wcbMPILB!%1Qgjwo-!xofQjohn}8c3k2(VuyWvW{%<(`-)%o!w-0? z)<#rq%r;_&=V;7lyo~e23C~62TUm{sr@U8fyn5gFSm~Cr(scA4-~V+S#fegV7;R@c|KiV&KmgV-1avv+!tQ3d&W<7 zeq&w!A?L0du=ddX@?2bPA8}V-iErYtdn2C9xzr1_Putvg9(#u|iHOM=&?c6JCVl;Sjvzx8=-#^Au0; zXKK#UTwIn%YnXZf7$tclSlG9 zCU*z-1pf$L3B*jX!umseELOP|@u1?i>vm2)l^CHd{vTo==J!FK4;<^E!14bXJQeu- zjX=DR`^d-S9NHz0IfnRHexGydH+(7{g1^O4aaSHAw)=nI)Nf6Rg%z9Ki=lxW$a+NF zjO+EUW8#13bZ+riAF0SxE1OFFM}>NEFT zOx0iV1#v(gg3paj+L7O2J{iB}1FkXmaXO#feXV>)&Sb0@v-ANyC5{?Lw2S$mHY5MQ z8)69FFdyhf{2^bjXGOesZk`v-M+ebHVx0FhB8}v`(n{u%{GUgKb8)ZZ;sdd(bf(W; zA3o!sh!J!-eL5zI6{tN%>cgTADff4dg;@YZ@;4LwS?v?+@H+V|U zg;R0RxJ(bBZDQZjoSE zuw5`b*gSAf{W(KGg!)Ysu|-VMH#CFTMjyzT=pH)7{d5nVPptMCt>u2|U;DTR`bYeb z8;buRrZ~3oL^ts(9ov{QE_uT8ak-5-(OnlYex1|&aNXvGc_%ND^MLWD54_seKgUyF zGAG3XuVN8i@SgwVyvC+@Y5ww}t(l%`oAGHnViV5-y zIRJf5r{iF;2?vWCUinF4hnxn#<9V@%4`uDTHpx}KOdRSB?B^K1x3Bf?zW=oE|G%T{ zuBq}CYwqO<&h2}z{{!+)#as1hT@9PwKG<`c?`wAFXwjeBA3 zAFoM1fotV7@-Mld+BrP0B_>o1!2e=NX_p!Yj%gjDSS?4<_wrQR<9b@e zcXAgoQH(NP#4BED<)Gpg{bJ0}I5d=aV2l{6#-6;UG?sBC9?@6@zMl5*(!@UAme@m+ z8RK4gYH}>EyeOW5v2E=d#ah*>j?$Q^4!(1D`2nWlX#9lc9Kg>0VCjzT; z$&Ho?Rtwe)#1rRtoqn6xb@Q+)hvd_U=X5`K?;7NLINYnXg!+P)CXVSJ&n`KRb$))g zi!bNC%$VQZ;>Rr+d@uMxAg0O<<%rtKpW~s4Ep&zJc8zu2#s(c!{VRX6owkYf;A z>O7!t-9tHq8i*Q-d#KN}OU|Q~qp$V7cwN7L<~z?_i5>QLKKTd^r!BOdr`3H<_rSey zd~J7K`p*3nWAveI9GF%qqh)aj&XG&VW%zeE+23*r zoF@OkgSZsWi6!D84!6Df1wcfn_w_pq zd^qO;aX@U4>&dP7hvJxc$xq`!YBP_?dE{uGHFJFDg&a&avR4JA9;ED zO`o|waZBH*ajC<|i}VNIObtv77AI|MKmCEP9UHG3bH;-)U@W-5_*6bYPuNE;OA~nh zSu9t-5~nto>v+a0orDu{e66K5Msa1;7>#>lN?vWvoHjX^xaz*@C-<5^>)Pd0 z@+um}xHi7U5y!0cOzJS^q4`CZ$$4l8W72zbr$EEVbHo&L*xZ#D(UyCXKYIoYZLCkY4{6Y%1^{M{EEZzub71Eotr1_A0H5E%}>pLZO~pi zj_$J7xaL1@!u|MMo`cW+9o8Qb{V-?&d+RIndzqZxC@)x;*IqKeOk3Q1xj%jRp55}6D#C$cr z90&NqjkT+MIJv{bWw)E7xB0 zlBY-5if6chUxr`tJFeoL@CtE?ctd~UF&G5Hj~$2o8muJW04iA&;! z_#u|)GuOihwRVp0DPH17JSisf;_$whB?gFzVwn6yZ9^U@_Bn_9B5qgwc78Fy=dQ!O z!p*go+xc8KKN084L%f2xSZg1|2Dzm?!M={AFSOe+^d(&&*3j(6fIjrO?VXcO5UZU_ ztgC#h*1gdWyi02WDsQ4UXphPljRDsxw(wKISdsrazWA^IteMk3nnf%W8^kfdzy79+ zF>7qPKJ&)8w8y;X zr<68UpSC~#ko%aYIMDXuwBu>N^YdGrSIlr6*Q_l*uec(nIG*}?#ZBCd3-K*ZtD1qh zfty`N)uJ5NxwOID#ber4`q**h7TWK)jv-F+tmFVtv**z z)Ia*$eKOCCL;s9p?G%swGjG&btImUq%xUutZ{aEPO^lKY;Q)D$d4dznG1o#j$ph(5 zeg~e$2RIui;BY<hZwT>Qd!w5^yWe{#LJfrlYRIF`63Pr-*E*7=*~;#fQ^^>R9$ z-sjng8~hpZ5#Q1K;;l6s;-^?BzZ2K-i1WH0^ByPSSDY)qk(Nl@)TCpEb%a61hKjLNWr}G^Lv>ShFufB6!oUMlA+_tg3{d^AU@!G2$ z{0@Kq zxHQI{%Y3Azy?1|okJF81*X=&JE-_Giq$Tu)@t{5CGL2;385?4Q`C^{p6Jy;RbsjOw zez;!Tb+3&9amt*c`;2RI$FcQ;ed$qrPcP!O`V60xrW9Y~N4Q_kA`akXTxma=)phDq zZRTg<1nV5cQQyo3UB^&d)Gz#6$ZU z3+4j-V{V8|#-9A$Ty#wD%?+*>5ll(tN)K~of z8(#@{WB89QcWi!*?TZijIJnx{Yi&{Mm^14I77msVHVl3l{3$piI6nAIU`?62(2l`w z!LGs9!5V%4FYv)xm#|1MAW-M97R)*VHFaylW=QQ`y~Xy{3ak@&-6+^9*gn`U*fy|M zZtvjWV05s5@bh3~;Jn%g+F~8VpkS83wOCuHt|i`zgX(gX`|#e>r1XzI6_?~Jo@uhb zb%FoW_y5RRI<=trgSmr+g5`p$SJ`gyV2;3bIv?Lqe53)~JMmG@gxke)ItRBHJ2bfX zAg+kLa#M9B+$>iye#9qn&e)Tu$cKy%dO?g3m&_029yj7dUIJeM-|_eHgjxx`MYGe^ z_#5|{S7Iw4L|pYxyui)IH6FG--<~I=Hg7GI`aYgh)5a6@GVO{_9UK3fgXXw)i3zkH zzHvPB8_(bt>k)7UuSuKLWN2=hpH{{p@(RayZH_HxF!#h1niB^&HVuz=z<%0|D|trz zC-oESr*IdJR@=dc)^p)^^#~k@gZOdy8IRNM+J&d}qrb%g>jM0a9{vUOe_on35Y`&l z7hm(bTptd1Kh*W{srS4_am8`e`T20juGU%m{~TI-t?$)`#1h_@zS2g=#LKQvy~;6K z;rjQ2mxD)xJA!M1EBpT6WOugj?5v-AA^*>@iNS|~wHDSDSO@4jtS?Y=62r7nKe|s) z>%#S)broNksfu1CzX9?`lC*Q1|mU7vde&R?}Y_suoA z2lm$&;=F6-L28fO%s4lf=~AAIe3}+BE{#$8!ZM?FKW!#Y~e zy6}m_4D}N=6>-XX1~qJTdBCx_ylUB5BG6frHDhY(YUhIju|OQapI%o8HVO6$@cGe! z*L{QCgAoDFhv9+0cL{zG>=PUsoETgj+#Wm@{401NcrJJ{cqI6Ha7l1n@Y}$##S#0T z5}Y5L7l>={*Wm8p)!_4>(QGt(8@-M9gV%z82hRqN2V;XPfYMw6K*Vg?;AhY%Okh0S%GZ7y27^;uv|m^D`B zF76&0(7$*UFFU8Un3LKgm!Zk{@!q=@o{Jns4q#5=HMM=``Lg|G=Q$So8E@iG=Pa(O zHNbenxpBHVPH)ngrDt7-bI`luu-3jvv7dd!Ff|OC9Cy+7d>>qD?V0=nr;BO0(S9%Hxh65?5v<1!`*cmphd9poip9as)YNc?NhA_tE=2b)F85W?u1O==Aa{%pG|> z@VL!Qbzk$-YvGW1=zMGQmG5HS@%7D5n%CMVI^V1Rf2!Y!#+NwM8dG}Iv-IY$wL+o) z&NCZ}<+&7k(^_6y^Sc4gpwV%NXBW+L$F{E;f@cXmQ&OOLwddAA&Y)(i&aADjORd2( z8R`R7m(U)^#7VeCeM0*jTb$rC@j0q*=v~(@N1;#oYBVoTOyBa@XxZ`{)EDSu9A}Li z4~$>J%b|7oU+xoMM%=?Gw5oe)Zqui7Hr^Yb&vE%?AWw6DOUL4*s>ksi#85t^e6767 zs@XXvO>V5v)p9vpX@5G|IKn6LM|miJvyNA7#C`Y;@S8l>IP}W1)qeZ&_gts;is7z- z7gM=_&-8(K&m&S7;G^JTo`k>oD>zS{479)V<8If+!*ZQ`BFCW*#DSrKaWDSU3w$Yk zt^Ogeb#3_5`6|EG_VR#S7sw6N6ZDO~qkZ(192?i_6X*7M)#l~f?kV4j-z4vFo!a5P z$(_y3L4nV!UZVYU5v?U(p^5apxP*Jf8k(Js!Xxq#{x?1mYxIZn8W+xOEcmAl&dV30 zjhxqK#tk2gH-t-Solxahat#pM+*|Qe9l$?h$TiSz`qpQTB^DkU;7Tr>+`#%ce2*i=BN!F%`+bhL=~Q(O zxtSQR-@NMA@?_*-;;ilXZQ5a+>l1k;O<=7S&F+~=_rm91jZgPU?Sv2G+TA0$BTa2z z?G>NI6md<=qs!%Hw6l0YdpN%w&AF={;<)CS<2nz%&=>M-bH)8LF7&QTae}dJyc-MUX^KZR|Gk2FV;^(HyfGHtOLMn! zPrAbR;$4}8#)9i`ZpSyrjSKS|ANY(<8U7QZd3@#3eXSd{TQ5@2%JJp2yB-Q`Z4A71T2-hKi}`4&s{q9h;9QwyBkfgZcmm%gcC)>fg@GgHk`> z$?@!9Odw|S_^b^8z8=p`4avGLev^327gUo_o8W2DEA$DksLt;^{6+bmXI<2c9E&z^ zJb#1t&dDPa`>iGOoQ?c~uCb=fcFwDfw1ahXJVnPA(_IU#;rQ0OO$g+X;5emutPLC+ z*iOD_eVyN}pe@`7_k%X8nvM25zH`_5Jo{Ut4f@FWXd&P0NBg*T_u2YC^%?ig=ju7u zi|8x&P5=4McFs-L@CNmfwS|0C*XVm59L-~28qj<9S+2%6R=c4UjSKD2PhK6559{7l zea?BRCRW#~y~Y~ zF^2`>Fdblyo6EcgTw&fgw|$Ij=jR<5%e0I+Peb4q@m2iCUHD8~<89z;94D5EUpNM* zI2R7Uq4);R;Ya*~cf>Vu9U}PmfjI*mk}%A3%&W zo}I&cu}_?&x1Ez_p{uo%?ot=?DyED3_A^K1AdX`{^G-aa?dfdoz9*nR^f#U0`qcRR ztxxPL{>pP`YVqE@G+*_HZD z`pkTG{pLNLOn+;S&&^5g(0-c5b!&?@dUa1+i#hE6@dJzn`cE4j*SU-f`_n$gfcvN| z?w9-N7|vsVzJg=vANQ3m(Kc?B~3vOIUba^LFb4sVirB) zI`pgdx(4gK^ojn%zuM>;^`$XYeNnnm91ugqD{<8^#U^@^&a-YoelLF7PW&=H={n=p zvkKx$<#_Tg@kL)$e;8Yi;~3&HJ)r&8r-(W9ocLty8AHykerElmm?<{#9&Ar{iOKv0 z`i(X$51dvsrfJPuvq>Lr8r1p=F;a}>ThO24q4S$}w4u+mSHIA6j$?jkgMH}Q8+tCetSNcMaO@eDX=xYu;(MbC>?*CAb$f06oIjsoxExm()OM zJKBR*qg!YKV}mE8jl3MX#r<@@&3R)}twA61bezNWYLhWWH_)}}k+{J&`i^#Y|7m6V z!|{1YJTEl_-jemM{5HD6_OxEX5~#1Xt9MpTZ;6T%-jQ2ge}@_A_* zo-{3DeT>>$Jtt1vsDG*7(G7fWJ~#bCzc?q&K`W_|(J)m{Ro|17tF`i>9|$fF{ubO5 z@TngQ9uA%l)FJ5>z6))^18{u$$!mE;@_YUWO{2!D-Cpg>W2UWWEzbbZTDAL;|zQk}v5 zQUl=4xPE=fbD$}VWxOT_!#m~zZsgh0FZ2%YhAyP@ctEtax{Me{+tDB5l32(a5F5+| zzw2Yp;aU73Zi=bmtiDuZp_kQ9>UYZ2($#s@NYq$pSelmBFRzVm5mV&(=90N1?v4u7 zh^Dlsw4VuX3oZ)&5}X;_8mOP^3wZ?HO9zy`6i@k=`o7iznuqSQSRx*ozs@K2(_gON zoTB~cU3$Vj)?db!`Hd?p7q*5-?Vx|25C2{;G*~ql6f^^Qx18H=TB!}-J@H5Wt(NFn z1kb6<^{v-YPw*X1#aVKBz_W67-_`RA_|s?dS}{kPZ0q>yAC3H*BT%!@7HgE$M${L? zLTytM>DT!B_qW7{s%QA#K1&5_1g>rFz&>gcuE(*QWBy>VVCBI2osq%s0`=FQ2I?o; z;QSK;?Zf%FTYtJYxKrGfm-8uR&GXtz2OIXShdQZoQm6iWcH`_$ZF$A~7K(l|M`;S$ zTTY|iWvzj{%e>^Jx}S6?T_u*$BXp8jWUjfN=C?5}zvij%S;P(=f;fuvjTtorbIo|e zRo1XLm$f!{Uz~G~)SIkn(tqmDTLc>iD+DtJ`oSFKJDH>SMr;yO^+DB<#8quIhw%np zP*-wq77xA?d^=bqaG$KxvSv)Z-91ph@*8Z9VZNFhVvQUW57O)U&z$B(h`+dqAH;*Q zzRDUmHFD>dhlyACPX39H)Fyiab6cKA8;YO!g1?7Tai;lYK8Y>j$}6dBiW`pvPXw<9 z*5Fw8qQ)&xptIE@<=*n~H?pn;4_p!47~lo5M*Omm>vMn9v$gN~_Vw)n!6~g%T1y9$ zdMEWhAG{x|*jll5MliTNxIHdVm!`kzDm+IcxaVRpUK`UM)BdFONo)Jyt=_kKFAfe3 zwhxvJCQY3*_1@sIV7}h@djA|O*jli4a`0BLXnWE27Qv`M>{Qz@&j$V5{o6kX)(h4MRtlC6 z{C#jBR`Wc>WBv@7KfDV1Lu{cbcqZbS+JZPMPZ5`l&uy|E(b|ozn_D-p2%Zcc4$ca8 z2$l&J4u%B7gWm_c28#v%X#AtIuF#mpAL^<2&avfd#);U*hY+9FPM+Xw+VLLxQsdtar+Kcnjn~DPhl8Irf7U!8c)sy`gZFIR*M+SMTaWZU()-t7w$^N|tAd5w z3%3^vE^A%Znj^To_wHV=Gq-1M^YG~Ni^Hm)1rG&34(N4lr71qn?|=R*_;2gKoiRCI zYrdA_FBBX$^{}avr%ax5QgBr8S@6l!Pp1B|^~)A-&9zNzOl%y|JfwL}aAa^q@IWxE zXIRgD!QB1k?l)6#UC(trLjwKw$H2868hG6zNXp(kt#Mk1OT{HIYG7kvBH20I2HG(Kot9-I*1LThP{4Gs@R2iCRj8HiQZ0ExTuGr9>Ei=#N)c+xjlHLq$e z*|TKN^1;o`o15bs;~TF}_j+f&COv#X`-1k+AX2UMXn<=r3O?=qw8Ihm^zPI9>%d(0 z`I7A=+tzQ3J0~|zZk*UWvAIRh7Co;7L;DTwH##^wI4BqtoY`|`&s@Px&6}F124@6g zg42S7gFS=a1g8Y!g2{ol9v-xU5UFaNN#^{UYlPKWM+-w#HA~Jq4IU1D9(>UIL5D}@>7A$dlBt(WT|T&B$_-OC3kLLI z=#-&Tei@t<+#h@rd}r!+rv4~cKNt|aH07l!*93P2GfbUf>K}sFf+c#F=-nbXKKLrQ zx^;DHul8Q7t+u=c-{M)`DnFjSryF=Ea`rI+?Pjje9Plm7eZGOZ6Ad*azmHEs)xdHd0L<|ch3zWX7r*|M6DhIJl(0N;on zpf!9(Z>T-dnsk6Xnr@P-m@o2vx<~zlFH-)6xUU|CvwxD`Mx{xv32q9m5AfV|!R^7g z;P1g%!IpvFQ!o#CO0*fCqc>?Dy3m?P@qlL3Z*nO4KAnuW={7zAEhYbVEZ&N-D&LU@ z%W18$YG-Fu8Yf?}5Rb z!F^Njn|gNe=ivC@h`>1cZE!$vQgCJP_u!7;^5Dc^|6pXWaj<$YBv>ohJlHPSBv>w( zHE0Ik3XDtpT@*YKIRD%mW3KW*I;%+j8vqX{-?Z_-{TSTK08^U)#+_QFI3?$yPu2;x4t^UL-}eO8wSu^74eSHK%fXbOf3tt{JHhI~&cV6Cgkb)j z`FoZP77mENSAyq)w}Ux*=IrqJo;`bZG|AOHSNDt!9tiFX{C#n7W#F@qg2nnR)^E$; zcYzpjR&YUZZg6riD%dC3BiK9mO&@mew|l>>gAIcp2fjN!xG$I#%r^aO)2|S08ypkd z7(5lc5IhmwA3PMi7R)}w>@%z$>=^t!7#?gGtQgE2d_4Wf(?1-X9Bdsd7`)c+wSLzJ z;^Zd5e1X_`d2m3mVbDK#zWIDddx(!e3cedG9?Tz1AAH>SxYOr%H16nVykYs>%gNJC z?%Y2ykVhjn@oabnyaPFu`p>%o4@kbi7vS|Fc^(VzM6SbUQ5W(zABe8x4e_#!J!?6Q zL-`Xuesdq5&2QY}9RA(90e{1`UuJFT@G@Zj{|ir|Len&8r4OmInXU2tn3 zuJ66R_iugh{2*w331)m45*|p>4qXIeNL%|EdtHJo-%>bYA2xwq+KDDo^J9@9iB~I`R9}IYl z?jfIwAI8JwTF*dj>5F=@wF~JeRZh?5Yd9XvUSFm3n9K&;68wHMuht>{O50(w)4g3xs z@5wU=JSXc})O%=vFXD;Hz4<73lHMy%Ry~?`EN?nJ*ge3pc=?~Le|B=--CDb~wh4;s zcW>?9`f;#)Fjp{h@a zd;Wq(zBjlcI6XKcps75Q@ntY;b5`kNbA0j@`+F|qSHaJMo%>+y89QqR@@1YCElBSf z3p^=0UA@CtpEcmK<2Q9tUeIZQwFmrCHC=to|E_g~G{1i4uedjSdvTQy%-8XqabZZnp`qzj+uEAHJf$3zvnVd&HWA0R4(pa!gNBzclQ=|D$ z`#&AO>Gt;RoomG3*97MVCj_|swBV}XAHlnUH6`{H3usn3w(EPn{d(tEbj=xo{rFV3 z1dj*4e?QPKV#UP3XKw_!kgfxt^W*qwj|S>HATGTWIEH`xK7PaJ`Tw!_OxwgH`JI@> z>z^Y~Thiy&rtk`RIl}{bnpTucs$-~Ws22qw)3J72|R~nKHL(RFMkXC&0Bdhn7KJ~bE#mhVAFuV z#rN4M7#Zvl$hnRR{u0nr^#ggCT${#QBJkd}YxH5& z=BmwQ1Kyb&c!q!oZw1Z#n&AB>l8`vBeM{I%SJb_o&LbGnr2Yui=77At$ z_&AP_pJ*;I%yVVhjpum${?^V>!9KyrKx`DdRtwZt#7JC4U;E8HUfRonXRzFJzRtD* z&F#IsSKc^RAokk+-hgJ8Q}ce*%J^*R5~l?6)(e6w139Gce1^-#NBO4lW^E!rPp-aD zALh*S+|vc0#5c;Bg|@UrCNJV5I%<+JM3@?KmbXSMb~4N*Og-c-Y* zJ%Jyqw((q`4reWmdIt@I*KoA?BKBD$hI6f-18dUcI&vPoXio9E)s}A!%(Vjp^Ok1i z0h!Z$cwVVFFYdk<6vy}*if?Ex9EPVa4K5CF9S(P$?Ie+cAfc>Rb#&Uj#e_lE>> zI62|BfsSI)4<=8 zgBja1wwDOj4CMW51$5Px0d4jBK;Lg03=8DkY6t2D{@y6qCE$6;{oP0R%)PRo&v_&A zme&IDMXaIC_yA%(&qeI_%&;0h?XCW7TpkQUnd0Vu+I4|a_ z)0wN*A-$8@vH8d6mouvA@yTdEIuRMz4zJ)SH9;J~H^48ri1$yc^9Z&H_y+16!voK} z^0MhPg6`ea6=+4gDQ4Xq;Qa-H83J`!b-?`t99Z1vb^X9`vHW5Aq#;DDT=e>Tvzs)9C5Mh?J-s{1`k> zgS$68&G7*q_B>xwhtB^;L|&%mqc7FParCo+dbf2`YSX-bpQ+`j|MRVRwmeJU(I8_2 zn%`JZN0wVz_oU`*oY9KLit+U8z#0?z0?wC5(A>7OM#MO@{zy(?tja;GiBhkYXVRPe zbkCO=yS5cSj8|h&T;Ww&8zrByuVc`$qXM;FIf?e!PJ9Dv3izpF9Pd{gw2gk0AGmfN ztodNgjP}!E@?E~LV>s7_0YBM#E7v5xX(O+=)=p}->y%@uKRc%NT>4LpHb=$<;k;qmFVAs)ZMD7+FU%WQ-$}pA_s!)+ z0(A}F&D{rnw>-fbVX+w(JQd)Z;u$#v-++fCzpy6EKRE|K3FO;627c1Jfmr`qp#J%E zP{-s&;2g*Hop>#u!FfC;>r>@m)~$>VaH$$D9_D+?9pujRw%-u4){4enGoZI=b88#q zX>xygi~ZHW#eQ{TI>36c)dD*G?e^Q9{Nc-ZRGy*cTh(`R1aW~b5EE#7`aphS4bPoH z)&Kd$cL(CkS%LcalYv^Kyo1Rm-w9X&BywYoTez1$BFDT^(qp%BO4k@0}gNFrTul0Av24B+nkiQuRa!%vI_O_)H`C?B7JTv;!`sEJ-W7t|MdexjU z#y`&b{8s~W!MNp*8s`%NW0glBzH1AwX;L8fb{yN&q|RwApZCCLFea^Cv+v=-IRWp1 z*0l!3TzE7nU2H9hwLo+N?cf;ngKe$t8rQd8o&Gg{=#tVW)~eqf&=T6Ac1MHIEw(#1 z;DgX1-qRuK39hZ4wQzlW61rO7>5s<)eaVM#PW?f%xJKtZE8ttXR{L9nK@XXe;F(R| zxi8lARF17y;rUa3hkAu;ww6>Md0%~?z1oc{^qW57>%14Zm$=CN)rUM#+UcVJZ}CPz z&6n=+>Ym|5{KymKnOkFyGw~(w$e8fncjB2?AYRZ$xLYi;UYTDiZsL74P&t(NDz1y~ zI9;r^c0}GKAN*m!3*9N81^!QKcOGMRRmB0E=`0mcAj;Izlu}ekM609{D#+5Z*jURV zAc6=pK&TohbXpe#ZKs5=3P?kON^FY?)re70f@pBc7HDOsEV8r^OCX>v6)7<9z0vRI z<(-6>_@6yV-^_cz_q+F=<$J#8-22|B`|P8--X0|GBpmrXv^ybvo-iM#zKKgXtZ~s%z!9F*xWxnR^ z%!=Ne+TEk)oN>Ver$zKJ^yzRP91c&!5A}f0ikuU{AD#R9cI4cS;76Xrk^Jm=JnB0U z*Ac(#jVz9s6FxaNG9#jgf(zo6ex4YSOVzaWheNW1*JnkXbJK$`rZrmY&5>0R`(L=M z-h}auqd(17wMJ{kJFV3{opE!X(wPso#+7}C1J30vquSp*)Ys;uzW1GCIGW$_KXcYw z-93WeO^)E|*ml>0VE?};|cl(uF(&byKn$`^oNnfk=oBv|Ev4evG%a->#Adqi>P4_ zi|7rjSM`SV3YsC3Qy4U+~o!|G0=oQ$Xw-#$9es*8;qK}0UT#S8jE@v(Dnw=@QBcgv$ zT*vteJqCSo*Wh;iir|5K3ird4oIl`S{_#KiKRz2N7s{P-oLnlm>5Jh6@-C6<^a5~j zJYUWdKlXLRkbLDeeKy=1$969~!XA>?us`^GL|$x0#v=L(@~7YBMSLGF=u^W1=f3m~ z;fmVP8AW&nYn;DR=fEBqq+Zdpl(+O4omaDOC;zEk)ud_}HJLg|omS%h?&wK%PyB2@628lC)n|6Ud6DBI>h!}R z@|NG_H{-zn`4P{}jO-JU2P%i?`^pXSx*Slrum4f~RJ~MtX!3*Ju$*;TL_V{B>Kvf? znWvujg%NwC_<^;%ul@&)Ag9oaeBn`VS$(OVaJV&qc4q-?AWPMz9=!yti zI-^eabgrKuhsif*NA%(KaLiR6v==N7?H_T*$-JCLa1Zx47r&dI-iEb%&AgoDRF6Lx zaji$=>vgD=aV+(?UY5_V+q=@sR`1*A^LaR2%ev%E{>TS?M!~&^wL7zJ57XHy=f2eF zYHxd_^3%Z)wJv@yHuRy@r|MUGZ0cFHt$(}ZOs@Q;_OE`0+_xa2XCdFNjL5zEWY(ew zuQwonilu!c#WU@}%ZctKN4l5uZ1S+aiTrC{uJW*JtEZxm;&W}T)yI&t1j$B+BZ&sJX0)UMVi~Y$FeGWXqoUF}xv`Zq^DwjJer|*DAIBz1C!z4Wb zd0WkLMdZrJH4$}(+>PI>Q`8&wusyFfQ71U#FUQN>^17T}JptF?{puYUXf61H{e$8P zcmvGC8Q`e8$G&c1Dt_SJ#!#E6L;S9eQ^UYVHI{L43->GR+=#=Z{GBAZp0q5I#*4mKBEs<>fPCaGqYCgJAkLi7@rS(qmC*QT>9LgaP zy>B(0T2|g-CwkMn=2v=0#q;!#RG;`kA84OPE^n9-QaF&{rVnL zx?w-_R13?K#uU5O>fUUj2DD~=qrP+>{T;b>AVRm_jM&fDD^i=QKdr@l+=Jcp%ABFn z7x0`~n0}4zTJN*hWZv%Y43zm-FT!hlp5MBMz8jHu`GkC}zaS1_j?YV~$;AiT;Jx;= z=-d5$ho8?#8qYp8JF1Ue>pE+pD{E1Mvkx7(PQ2mG;*(GCIX+BRaDmO#zj}%O(WTfD z8|Kem{F~2-bN(fE#M$>E{D!~!oqfz*T-&#^Pr?p9ui$eU#OFupk1g2M{Mp1gR^L~P z6L}3Th*hz{N7+aGi3Q`cKRhshePuXHU?>cO@$d|{(HC$JzHh$HdeE79`7A5nw&(90 zxp|ut{8Cehtp_63?LOj<-l~rvkBEEzqKDu!&hnN#peHRK$ye%lxzRZ$oXOwV@w&XG zzM2*B8F=}p^t&XVEi_;ADILQD^AZ1WTfE5?@Y!CET&e$E`^I|X`V?}L{<$8%JX-x} zy?(uAIIqv{^H5)o$S0>qIZ|0^9> zzj_!qp$D8mpWA2J%*Vd9Jpde3ZdAWIqk{A5*}1=UA0KhOwOOwmY(GQKUCwtuXT#aT z{Pk?`PxHWa@KZf^oKu}{PESS5iQVW*Zg(%7L_Rkk`BvVgbGqkeJ{Kw{x`uA;n-!nM z(=as?PTXxu;x@Cs-eGPu#&6YIV8SbG!bRxRx*IN%A2A!}K_Jk2vqe zmbi*NCF`s{1b?(2q<8f~#C{PSIv>cV@X)(%FCWmqS7sz2=idmlV8 zXX_FVKj;X5;zRaU*wwuF9iB$V-UA=yGPzi;)wd<`20Sz7k)6+6!**xnozu~mR)4AA z^^f5vjDbz^tX{Cbo*2b1>4T5p!uDu9$M3{|eHJn4bAn^Km5cE#CEO^*-kf*@8O|-qsd74d!C$y+p7dib=A}pNHF~uUugR0<2!nAZ0vF*ZjCO8{?u{dN z6a9Ssd&S`9)WG&!*kn?^bK4#cj$@wgVXfwktKm0r?WqWx%j;_+dNCmp-LJt}!wEe*r*+v42vh~ME2{K1o5=X$v3 zy|@qml#k3GSCBi{%i0JVnX~*Q7t6!eCl|}N{1d;xS6w6Dc~0(jANdQn5d-qDf4H~L ze!>W~0ZdZg=*P(g@`{=YkN4eT>LI;B=RolhvCXFZPH&z~@DStRJn+40hpqWsvNM(P zj_1T^>3|=y9bI~ld(#t~qATyEAH4?n4aekZ_-*f4?t$U3NsTC9nYS8AKSG~F|Gv&J z;LyghkLjLruyOD!DE_!OT}Gulsb#rz8i>>X~8y`mzGyGG2mJjrf|JwEv%`7zw13ae_Rjefe0QdRp4HK8?Rjg)5A+Ri z5q&A=&GoME1%99(CSUOpcEsWB1;S!=sr*9JIaL?Z7hU0`;*wt3)L8B(_tA~`@oPkI3 zE$phf(>d;JOmV0OP#g$1;Pd#M9)xxIED9aNA3l$Ziz~P(=J`0TSh>qOVFcgem&SvI z<|4M$VPYRga4&PO7=_V%4ey3MFcOA2yDdliY?qqVzA(}Mm#5?(xlvBzKWZD82Ft_; z-KZxDqng+IPnnL{)H3#8m^n0yi;5sZov(HCx@7a+DvYdPjGBLqqk8#95pwd%--^;pN%Dk z@iZ7Fcfm1wf@ADuZFq!y1~;71!Smr1JXRyXQge~#C9#I?KtAH^{ohaI@Fy2^Zf&V=6R2S3sWG+&&bPUSN5lrw5C zUEaYv(#}CClJ-`SU&d1=W_sHqS@SU;x z#d0-WTBr4RPX4rS3qRRi4a^5%sD8V=3ftg1-O;5!gZblG{My{)cJp;cdSgUiT3uf} zfp4jw`MmMz#Twu&o5NV^b8oMU3wl)3v2F1P>mu-w9`PsVz&s}>=$q4}8W>+OzPg*O z@IdyWQ$28J*XT%oG!I-A?-PHpf?nAX#`-58!B|+~TCo6w3*XE4*vNd?LT!w5!6mhi zc=kRq!zXcJ!cWbg4=jnOSK+^00GIJ9I*|{|Lk|Y8V|U~8eQ_gynLFLmt6rC0UGaU` zXf5~vUE?g~fjlQYz$0}*)xhN}6@3wMJ>xHUfE5r5Jfac?}2o!|@1V*_{&-`Iiwz&v%o-kyGd z{eOAIz4adWs@g}LEQivo*tT9=#koSC6;cz61Dw^Knz`UGxEXtkQ@jlxxz_%!I#`XX z7U7?2ChJ!-TPuvGclu>#J|jQKo7Tn#xD9>tUp#>>@l5x^4RHqRa)wb3!fW;YtWEzw z4em2adJD#YQ<9+?`L40M+UNgf=CcxX$Da13_!GYpPq0T^>aXF(xT^ji9CN+cmdE5a zy3ixQYs|;zqncyQv2p(%*FLAGpMd|kzIZxrU_LP2*f0Y(kn8Yk{)A)UR@Nit_!R!n zhxCd0Hy$J2#V5R$Ct({Lkq51b|Ek$uh`<@VgAY!L(62ZU=X{R;(K}ya1AYf*VI{xt zoEirwhIPe(?fJoe^@6-cf96{;0Z(8b9Ku^*pSc+m_ZLTeTa3aRc7%I;oS1_eS>Mh+ zr+N)0!c=i#UG|*KVKBl+*+dTsmdoXE2j6#Y7Js7yI&!b!2pzDC@rt*xx!2?lK47oj z8s!;!H9pR+UtoRS4};+k{=@&+4p!rU&O7^lPkG)v;jkE{U-py3VLm*^?d(yxrs_9q z^%}k?_F$I!SWXs0cr2fxcl<={3Cs8pUZ&?kzt+On*nBRTuV&hM7VwNW^)!(>S%dJ?5Q6JTZwD_Vo!kGU;)0bXJ(%Iv@nl; z*_p1{h#l!n%&JM*0)LmQ%}0KLH?WHBtY1EYGdPrWT9cZCf9Q*eb@^H>tFb(fkI1WH zl}{Cq)R)ua&_5^U2}jt1ec?YI#kTxbp3oEZSunUvkLHKR>N&H!+<~L0)!;P$g~9ef zU?46mXR7tBMco7+U{RdefC}4li%Yf?2K2{=WSt+9uyAgK5$T- z?cBMVUA=}Yi8VfLPeuQR&%hoy4U2JTy&~A-JPo^9qu1pi&%+S<;lIYp@vruoFtG-! zTn{(o5&D4D>PDOiCeXilQ@hxclWXusdX<;p7+)8ka1&PZ9oQ|Ve3s2v>@9Z14PEei zm|(7Qt(s39z&gH6*K(4&MJI13^iXY(rFid=id;Czms}cC6`@mJc0+-;jGyBdJ z-~u=+e6w!2!1hEwQ4_!xk$*jJ{(gs1UY9puhB=5)bE;fzeey9ZsLziK zgnPQ)YhoSG6SFW<&%pi-d{+a)^okejF;}|qvse;;{KH(tg?xqki(g~FPda2T_zJIJ zAUoq*;vZjhEj-2Z_#E4K6jLyVkHA)30q>+=*oo^BcIUTVr&}?nSHMPWg}-{uTExAc zzw6~2HLMyO_b?B8Brw_g`KegKkJ$qU&L-ZKE)pJhu&eQJg#4$KRq+TPvM!|E`Qly$1QLiJq_dPkvfOwoV;GMd*I3P z3LVOM;u@yI{eR~3C~zANs8`fLVxP~zRqxZQlK1cyxW?a%CFXo)dbgbK)5qqM_=k8f zmOR1N;TYTN`M_=Z;XSF5_`)_BKjO~4Q3I)vk~ky z9{*5_RcP~EWdOm^+^8qmq55yjxAhzTSTtRMUfO_SqVT5&vSJj*aJixS(3WXXy00@z5m^9D;v~A0n>QYV|oidVq7dEFUl@ z`9Ln`V|+lJ!#ChK-ST()&3qb9^4)gEGC%s&Po`^+)-MLw7Cx#;ha>u2a9mGP9MHdh zpS@7{Z;(|GLYo@xc>R$hqMw^GmOiQx8f zDtuy3I7UDEuHq10(-}XK1Nk{lC04~8PJ`RY3u41__^5jtmp{=PoAFb5+-H(-AD=0) z4`@FX{<@Yfa5H;7&d}+{!Cmi@GhHK3hygL^XL(BPPHLpyD2uz#(gJfANO{ z!2&r+yvRd*7>DKG;tJ=ne_~DS0Ke%>UgH1!&d+y7>&-p`K+^Yp2L~pIK9g6{8L=>QFC@bn2k5VbU6rCI=}66Saho161Q+rjN=wCmoGU( zq2F;+G>Koj} z_v|?1pw1PiYFm7m&*~MagYXji;V1k>j>dcOWW7dQ5RZi$@;_g79c3%9-(Z*DaUy#X_W#tU;!f=?fB5{1Si(W&d18$)$$jMn`NF-$owf2=HJiFe zoal3{h}363%?am|f8ZJa)PJbHK0ko-)+koQgxn`~;5hu?yUwuHcaX|Aa;j_Kk9*M% z%;JM=i*Mk!e3{+(f_OD|KEY4;gt`qEQy0s#@(5k1Y2BaGVdq5p}og~~8e`I`QO}@`vyu%8)8NTXwzypr?G(S6v0-ONk>1!ll&V~H(&G5g*)hTbB)FmG$+r()b3 zjVGV+KWn0MJ}W-t1Q>uj@liR$8pRXrvUYeNo?#VVz%|_y2H*tNYwhv{{>oqBr+!lP z6WEph#fV&1Jy%#(y*cBVHx58we)mWZFxdU@eVmc~s&?cHFp@vuVEkIX71Qt*M$@nL z;fZR0^{IK9GyCAW9{FH#M!KSJeA<4BTt;tlDeT0@*c4vKNyd>E?BCKU+o~DNi{AWP zIPE-|9)~{Za74cMewf78aKPN@T0E*Dd=>*AwBM(XWe>u7=|n9kZ(BG1jDvWuSh6NL z5cZfKpAv^So_NJ8*qN=&!!_z7`#kCk^%{H2vGNl;;K||=zk&JkGT(u(Vn|#O?<>A3 zFYrxzP|wmMf8d+&AD4%T*1#X}wOTtr!58V9zrZcIT^)`S!VvfKK01V-eArsV8hjRS z@Dm=IFKn?s_oQEI!kb|}+=8k46Z9sx&?jHyGak)PeBnX-QGUY@@DN;>Px1@%7k@BXYR&2-I>vjQ=XJId z2eKCkv()8m!lvwt#~4#RV_bCuovO_|&z@}OzUnOe)%k#x5jDJZ*wbWFHMV{!4uIdn z2Use;;FeewpY+PkbS!?v6v6vpi=4-X_$wULpVKdf=lUj9Pq3>zfS0J1;3S;D8N@YP zi8pl`ys$P{!Efo^JX~A4GzYm=9>jC1M&Kj*H08VE8t(B)F^_8$9J-oHLHt zFqVAi{pPJ-V836lAFt70l8eQ#Yph?q<3DPB>u^4leXN6@@^2g)PoWQZNw@q$O{cz; zvvFXb-F6n)c{bzff2wQYw0M9Y^lNQ6Abwc2hrA-!vw=sp^SXJ7VOUmi$7WRzsT;+H zxWi|h4N$A84Xh8g(TzCcC+@>XUjy{lvGqIoBfi5f#;YCyKZ7eULw?ooEb*K%VKDrtJ}Dk0?&KHt5tsZ9cBpIk2yEjs z&ScSn+{res#UJn!+*vNgyZEM>+kS)o3w|V4)B@^6_HYd!<-a%(-eWJ!`F-ch@L7C< zUc@b5!8fcOr{GWY#`laR_t7su6vMck7{mqTK3pD-iy8P<^&3vY*V!B{tMB-Ye8)H0 z+3R{iVqHAIeesS@AD>{y+Ixay zat8io4sZ(ZgrWE^d%y>};y+>;|G`VxkX_AB-WD4$4gbdh?DfjudQJF_dBQe)Rm~;7 za2!79oJM_DfWD%Bmc1N3pu+aSe7COW)WUqC_IlN#Vu3I4OLk`yoRf{6fz&f29X= z!av{wALIx8ksqpQ;5HnGX{?nmS% z@g^_81Xw`-_9L9-(gW5Hh85-`KC3pMBlm)*^du+94KP<;$BWBG;s8F|m(fSko5hE4 zF#5v(_$d8~Cp?j!;4l7)f14X0)L-H|aK+r^4K{Q=yIQ-L69@La^vHZJ+PO=dOdry7 z_BQzqT)~}SsJZhAy#ul4^RzHa?-6dPeLOF3&=C$Qj^Jy}LC)g$;vH7to3KEv@D=fB zjxZI@@hg~XjrhO3gjd0D{RWR}AkWzYH>SQL5s!R;#(c@xGEKf@|>_UJkJ)tJ@;Tg9k#)ju{hJ|AUcI<9_;c!3%0 zz!&U=u8jDZ?&uH?fp6l+7=d*LoWL9Tz{Ws z3prnogE8jfXZnJ9<|oh4vvsfo>`;SSCtsu!;}KXQUagx=;D9ka$^|fiUSUPWDjzi- zJ>%)(0mjHv{0eUJ1Nfm=V10bjdhrd<<1p+mZ;DU)$0_7her8Q#-g?w*o-;3c)_bJ; zs*TKxA9zjdxR(Cu4o87Mex?iSl-u~d=V7co?`QZ)&vZwJ^01hwewf@TudC1Ok>FS2 z1%4Wn59=4xBR%kc_s3(c-QGAql()q#TsJRsrDriB_jnW|#Lw*RdH1JpI97RwZDF)n zrgOd`FNsaL!0+PAy^P_#cq?p#8*mJt=L2}5x#Ab>1@HMS?B|bc1wYt^56G<^@i?*K zk>J4c3eHEL^0GOoxr$?Wo}Uyq@;iT}3v+}qaFSlxl#S&Jc5z?2@W?O40AJz%RomlD zFhMRhA2?=?=7m?eKi`laD(2{s&5TP|#Ls+&k9wU>YtPG?*qPtE&YW;dbAkh&Y0O+Jc_(HoOMU_8hwy+qmXTU&euvxFY}K`!GQbB38X#!WZ!=>*C|a zs@lcae4eiO8SHfr^(xoK9h(*z+9T z#s}C(OtJw^>7IO%{>pFR0v*94@x*88mOsEi>w?+(q@_=)$ppL;$6MEsl)?R#;yz=n>cBFH0 zEaw>$2C_BZE-W@b`Ve2{10!AM8gqt&HGlWCX8Z#m5KnMVY|t0|I6oja;g9l#o{Rnm z{n4iwat-k_TUO41-|WcW_#WS&6E@;o<_I^$zPVbT`N9=*_B%V^J@`1meLN5E=+phg ziGTFO$M_Xq2RF?fp6auRVY%G;_y(K93OEQ8#7WJ~{N!2jz_;1Z{Mpg(#nI@?&&KtL z+Zfxr{qFwOsfJ}UYqMrJAy(x^H3;Ej)@^LIVkh>3ch<`1{my1yH&>j-ysg0_d|_)m z0ItG>!hX6Xd=)OwjJS>;d!K)BjEEm`!577D;h`90YjFxc#Hu(Izv7ghrI@pp=niK5g=XzN5#xCr@ulT4rh!r;VUOM;4z7;q6 z#P&z+N#Y6mPI}5}OTBurP9Mgh3+p#0e#@qEjF@0oy5M8t7GGi;amUVl6JF}4)PBlf zr_Uqqo`YTf!DIf;uJEPehOh7|`Ij%j0DcVP*$9th6P!XE zz*74592=>{^pDi+Y$Q%_Q86qZu}g6n`rvQ$!-sK=>dWvYn1plaHPNHF@hj`48-B~4 za-Do@J|5klkFl3^8pB%Qq_N~|dC%J5n%JNNuhEBV#fjKpZ+O6O*-qXu9(~X!pJ6W; z%Xa))T#Io&Wv<4+ZQPfh#R@(p=dr7C+1tIm*EnTAxCdY4F)_hUjLo*}s^81T=45W( zZ@$*1hH-xw;Afb~*9$AmjSkI~jm3$%dR@MvC%U0)Ig4| z_@LKdg?<@PFMFO|@h7gujUu>&IQ( zM}BuLJC|SZ1^bkGBKo@W4?nTbA^u>dSQG1fVQPdf&5!P3jrps+@I#NaCN_aRbjv5` zRR6?h0_<1o{noia-(%ymhWf?6qfhTx4;TKcCHWAJCqKY`*a(~G%6y2Q&4*s;8;373 z4}JsF%#Xgs7oUJ5@JBsJ$9M^>qi1nRcdjYUC)Vhcp5d8X34e(D!a!?h@A7wc<#%vX zys8KIpqwI~n+u)Nu{pWcb$B?t625Dl@Xp$64BU=B|64Z|Nf_#pbZw^NIuAXktoom%aFjN4ba2 zt(gt*J33`EIOiI^kNfK3=pBhSKZ^~qV?AQU&*I8F;VK{EGwv;}`3QR(v;34FvK>F6 zGuY)d>w$rVjpoHh=ICDU&UN@g@%_r>avwhz2VzqW(nE%wVi8|7ZrR?v#Io^?>mUCz zmfX*0;fOe3bGfH{MGU|hI)=H{%|7&wW6}p*sm<+M$VF^wFONR?2A{@{=m0jFv$?Z{ zNAsiuv29L#s(6FAFg{)L1@odS>mcIMJ**#Yhz0pv{?=2&_wfQTA#ckwbmMxlBu?Zy z97(-IuP}uU#hH2ow#k9GgnPTM9=@|3_FkPghIJnOUf((C^GxQZ@@n1Phq0H z0R3jY=jz?#!!>WO*(-pN_7B|K`6c7n3--BM=UIIphvMJt%ucKI7^9*tYpDs{9#w8%#ylRAGtd9-V^>&3{ukzYpqZKaDMOC!UP zC6V(Y3nTL)Cq!mPW<>Uk>>TNdyt&PrZ9B8Q@7=NAW*=X^fHCp|d%`bujJTvLXCwGE zj^X=Ve233#`CC7aMtm;ohRAi16_H;@Mk0?#{ubF1@y~bolzrH@KjORY#F71NXKH=E zz<161Ogp=?t?%RE6T3ycuQAqW|NXwZ@_jqEM^;B3j64>(KVtmvM;1oD8krwCKXPeg zS>)QtPa?mHnCFv`%@Kbm&)+XJ4;%|`f=}{1?35eTit?{q>+?!t#djZc<=>8x$&rso z_{R~E6C(p1Il2AUzD{oc&6U&hH@;4bc>VAQ8~Mzdd)gnBD`2TJ2F|{|GxkpV@8TSl zzpK-1HCwkt&Wju!**mgx|Q z>5=0iiy~J?ZjZbk*=KB@_TLJ+FWxjsABR54hMEu?96C!+PV?^G8gjFsgbRXt&PVcw?wXvoE@1H`DkRvNNcn;x+U^z zRZBF{wLitN window_length else 0 + + log_mel = log_mel[:, start:] + mel_length = log_mel.shape[1] + # Calulate the number of windows that can be generated + num_window = 1 + (mel_length - window_length) // stride_length + # Sliding Window + for i in range(0, num_window): + start_time = i * stride_length + segment = log_mel[:, start_time:start_time + window_length] + segments.append(segment) + seg_count = seg_count + 1 + if seg_count == 20: # 20 + segments = torch.stack(segments) + return segments + segments = torch.stack(segments) + return segments + + def set_image_transform(self): + print('Dataloader: called set_image_size', self.image_size) + image_transform = [T.Resize(self.image_size), T.ToTensor()] + if self.image_random_hflip and self.split_set == 'train': + image_transform = [T.RandomHorizontalFlip(p=0.5),] + \ + image_transform + if self.image_normalize_method is not None: + print('Dataloader: called image_normalize_method', + self.image_normalize_method) + image_transform.append(imagenet_preprocess( + normalize_method=self.image_normalize_method)) + self.image_transform = T.Compose(image_transform) + + def set_mel_transform(self): + mel_transform = [T.ToTensor(), ] + print('Dataloader: called mel_normalize_method', + self.mel_normalize_method) + if self.mel_normalize_method is not None: + mel_transform.append(imagenet_preprocess( + normalize_method=self.mel_normalize_method)) + mel_transform.append(torch.squeeze) + self.mel_transform = T.Compose(mel_transform) + + def load_split_dict(self): + ''' + Load the train, val, test set information from split.json + ''' + with open(self.split_json) as json_file: + self.split_dict = json.load(json_file) + + def list_available_names(self): + ''' + Find the intersection of speech and face data + ''' + self.available_names = [] + # List VoxCeleb1 data: + for sub_dataset in (['vox1']): #, 'vox2' + mel_gram_available = os.listdir( + os.path.join(self.data_dir, sub_dataset, 'mel_spectrograms')) + face_available = os.listdir( + os.path.join(self.data_dir, sub_dataset, self.face_dir)) + available = \ + set(mel_gram_available).intersection(face_available) + for name in available: + if name in self.split_dict[sub_dataset][self.split_set]: + self.available_names.append((sub_dataset, name)) + + self.available_names.sort() + + def load_mel_gram(self, mel_pickle): + ''' + Load a speech's mel spectrogram from pickle file. + + Format of the pickled data: + LogMel_Features + spkid + clipid + wavid + + Inputs: + - mel_pickle: Path to the mel spectrogram to be loaded. + ''' + # open a file, where you stored the pickled data + file = open(mel_pickle, 'rb') + # dump information to that file + data = pickle.load(file) + # close the file + file.close() + log_mel = data['LogMel_Features'] + #log_mel = np.transpose(log_mel, axes=None) + + return log_mel + + def crop_or_pad(self, log_mel, out_frame): + ''' + Log_mel padding/cropping function to cooperate with collate_fn + ''' + freq, cur_frame = log_mel.shape + if cur_frame >= out_frame: + # Just crop + start = np.random.randint(0, cur_frame-out_frame+1) + log_mel = log_mel[..., start:start+out_frame] + else: + # Padding + zero_padding = np.zeros((freq, out_frame-cur_frame)) + zero_padding = self.mel_transform(zero_padding) + if len(zero_padding.shape) == 1: + zero_padding = zero_padding.view([-1, 1]) + log_mel = torch.cat([log_mel, zero_padding], -1) + + return log_mel + + def collate_fn(self, batch): + min_nframe, max_nframe = self.nframe_range + assert min_nframe <= max_nframe + np.random.seed() + num_frame = np.random.randint(min_nframe, max_nframe+1) + #start = np.random.randint(0, max_nframe-num_frame+1) + #batch = [(item[0], item[1][..., start:start+num_frame], item[2]) + # for item in batch] + + batch = [(item[0], + self.crop_or_pad(item[1], num_frame), + item[2]) for item in batch] + return default_collate(batch) + + def count_faces(self): + ''' + Count the number of faces in the dataset + ''' + total_count = 0 + for index in range(len(self.available_names)): + sub_dataset, name = self.available_names[index] + # Face Image + image_dir = os.path.join( + self.data_dir, sub_dataset, self.face_dir, name) + cur_count = len(os.listdir(image_dir)) + total_count = total_count + cur_count + print('Number of faces in current dataset: {}'.format(total_count)) + return total_count + + def count_speech(self): + ''' + Given a id, return all the speech segments of him as a batch tensor, + with shape (N, C, L) + ''' + total_count = 0 + for index in range(len(self.available_names)): + sub_dataset, name = self.available_names[index] + window_length, stride_length = self.mel_seg_window_stride + # Mel Spectrogram + mel_gram_dir = os.path.join( + self.data_dir, sub_dataset, 'mel_spectrograms', name) + mel_gram_list = os.listdir(mel_gram_dir) + cur_count = len(mel_gram_list) + total_count = total_count + cur_count + print('Number of speech in current dataset: {}'.format(total_count)) + return total_count + + +if __name__ == '__main__': + ''' + from utils import imagenet_deprocess, deprocess_and_save + + # Config + image_size = (256, 256) + #image_normalize_method = 'imagenet' + image_normalize_method = 'standard' + mel_normalize_method = 'vox_mel' + test_case_dir = os.path.join('./data', 'test_cases') + os.makedirs(test_case_dir, exist_ok=True) + + # Dataset + vox_dataset = VoxDataset( + data_dir=VOX_DIR, + image_size=image_size, + image_normalize_method=image_normalize_method, + mel_normalize_method=mel_normalize_method) + np_mel = vox_dataset.load_mel_gram(mel_pickle= \ + './data/VoxCeleb/vox1/mel_spectrograms/A.J._Buckley' + \ + '/id10001_J9lHsKG98U8_00026.pickle') + print('np_mel.shape:', np_mel.shape) + + print('dataset length:', len(vox_dataset)) + + # Try out 1 case + # image, log_mel, mel_len = vox_dataset[220] + image, log_mel, human_id = vox_dataset[220] + print('image shape:', image.shape) + print('log_mel shape:', log_mel.shape) + print('human_id:', human_id) + # print(mel_len) + deprocess_and_save(image, image_normalize_method, + os.path.join(test_case_dir, 'vox_dataset_image_test_1.jpg')) + + log_mel = np.array(log_mel) + print(log_mel) + print(np.max(log_mel), np.min(log_mel)) + + # Unit test after train test split update: + print('### Testing splitted dataset ###') + print('Test identities:', vox_dataset.split_dict['vox1']['test']) + ''' + + for split_set in ['train', 'val', 'test']: + vox_dataset = VoxDataset( + data_dir=VOX_DIR, + image_size=(64, 64), + nframe_range=(300, 600), + face_type='masked', + image_normalize_method='imagenet', + mel_normalize_method='vox_mel', + split_set=split_set, + split_json=os.path.join(VOX_DIR, 'split.json')) + print('Length of {} set: {}'.format(split_set, len(vox_dataset))) + vox_dataset.count_faces() + vox_dataset.count_speech() + + # Collate Function dataloader test + loader_kwargs = { + 'batch_size': 16, + 'num_workers': 8, + 'shuffle': False, + "drop_last": True, + 'collate_fn': vox_dataset.collate_fn, + } + val_loader = DataLoader(vox_dataset, **loader_kwargs) + for iter, batch in enumerate(val_loader): + images, log_mels, human_ids = batch + print('log_mels.shape:', log_mels.shape) + if iter > 10000: + break + + for e in range(1000): + for iter, batch in enumerate(val_loader): + images, log_mels, human_ids = batch + print('log_mels.shape:', log_mels.shape) + if iter > 10000: + break + + ''' + + #### Test utils imagenet_deprocess_batch and fast_imagenet_deprocess_batch + imgs_de = imagenet_deprocess_batch( + images, + rescale=False, + normalize_method='imagenet') + imgs_de_fast = fast_imagenet_deprocess_batch( + images, + normalize_method='imagenet') + + print('imgs_de:', imgs_de[0]) + print('imgs_de_fast:', imgs_de_fast[0]) + + #### Test utils fast_mel_deprocess_batch + log_mels_de = fast_mel_deprocess_batch( + log_mels, + normalize_method='vox_mel') + + print('log_mels_de:', log_mels_de[0]) + + # Test DataLoader Deepcopy + from copy import deepcopy + val_loader_copy = deepcopy(val_loader) + val_loader_copy.dataset.set_length(100) + print('length of copied val dataset: {}'.format( + len(val_loader_copy.dataset))) + print('length of val dataset: {}'.format( + len(val_loader.dataset))) + + # Test get all faces + faces = val_loader.dataset.get_all_faces_of_id(3) + print('get_all_faces_of_id:', faces.shape) + + # Test get all mel segments + segments = val_loader.dataset.get_all_mel_segments_of_id(3) + print('get_all_mel_segments_of_id:', segments.shape) + + # Test return many segments mode + val_loader.collate_fn = default_collate + val_loader.dataset.return_mel_segments = True + val_loader.dataset.mel_seg_window_stride = (125, 60) + for iter, batch in enumerate(val_loader): + images, log_mels, human_ids = batch + print('log_mels.shape (many segments mode):', log_mels.shape) + if iter > 10000: + break + + ''' diff --git a/mlflow/sf2f/infer.py b/mlflow/sf2f/infer.py new file mode 100644 index 0000000..a795e8e --- /dev/null +++ b/mlflow/sf2f/infer.py @@ -0,0 +1,121 @@ +''' +author: Bai Yeqi +email: yeqi.bai@yitu-inc.com +''' +''' +Reload the generator checkpoint, inference with provided audio, in form of .wav + or .m4a + +We infer with different policies: + 1. Generate the face conditioned on each segment + 2. Generate the face conditioned on the whole speech + 3. Generate the face with the fuser +''' + + +import os +import shutil +import glob +import pyprind +import glog as log +import torch +import numpy as np +from imageio import imwrite + +import models +from utils.wav2mel import wav_to_mel +from datasets import imagenet_deprocess_batch, set_mel_transform, \ + deprocess_and_save, window_segment +from options.opts import args, options + + +torch.backends.cudnn.benchmark = True + + +def main(): + global args, options + print(args) + device = torch.device('cuda') + float_dtype = torch.cuda.FloatTensor + long_dtype = torch.cuda.LongTensor + + image_normalize_method = options["data"]["data_opts"].get( + 'image_normalize_method', 'imagenet') + print('image_normalize_method:', image_normalize_method) + mel_normalize_method = options["data"]["data_opts"].get( + 'mel_normalize_method', 'vox_mel') + mel_transform = set_mel_transform(mel_normalize_method) + + # Model + log.info("Building Generative Model...") + print(options["generator"]) + model, _ = models.build_model( + options["generator"], + image_size=options["data"]["image_size"], + checkpoint_start_from=args.checkpoint_start_from) + model.cuda().eval() + + exp_dir = os.path.dirname(args.checkpoint_start_from) + result_dir = os.path.join(exp_dir, 'infer_result') + try: + shutil.copytree(args.input_wav_dir, result_dir) + except: + shutil.rmtree(result_dir) + shutil.copytree(args.input_wav_dir, result_dir) + voice_path = os.path.join(result_dir, '*.wav') + voice_list = glob.glob(voice_path) + for filename in voice_list: + result_sub_dir = filename.replace('.wav', '') + os.makedirs(result_sub_dir, exist_ok=True) + # Load mel_spectrogram + log_mel = wav_to_mel(filename) + log_mel = mel_transform(log_mel).type(float_dtype) + #print(log_mel) + log_mel_segs = window_segment( + log_mel, window_length=125, stride_length=63) + log_mel = log_mel.unsqueeze(0) + + # image generated by directly feed in mel gram + with torch.no_grad(): + img_d_fed, others = model(log_mel) + if isinstance(img_d_fed, tuple): + img_d_fed = img_d_fed[-1] + img_d_fed = img_d_fed.cpu().detach() + img_d_fed = imagenet_deprocess_batch( + img_d_fed, normalize_method=image_normalize_method) + + for j in range(img_d_fed.shape[0]): + img_np = img_d_fed[j].numpy().transpose(1, 2, 0) + img_path = os.path.join(result_sub_dir, 'd_fed_%d.png' % j) + imwrite(img_path, img_np) + + # image generated by each segment + with torch.no_grad(): + imgs_from_segs, others = model(log_mel_segs) + if isinstance(imgs_from_segs, tuple): + imgs_from_segs = imgs_from_segs[-1] + imgs_from_segs = imgs_from_segs.cpu().detach() + imgs_from_segs = imagenet_deprocess_batch( + imgs_from_segs, normalize_method=image_normalize_method) + for j in range(imgs_from_segs.shape[0]): + img_np = imgs_from_segs[j].numpy().transpose(1, 2, 0) + img_path = os.path.join(result_sub_dir, 'seg_%d.png' % j) + imwrite(img_path, img_np) + + # image generated by fuser + if args.fuser_infer: + with torch.no_grad(): + imgs_fused, others = model(log_mel_segs.unsqueeze(0)) + if isinstance(imgs_fused, tuple): + imgs_fused = imgs_fused[-1] + imgs_fused = imgs_fused.cpu().detach() + imgs_fused = imagenet_deprocess_batch( + imgs_fused, normalize_method=image_normalize_method) + for j in range(imgs_fused.shape[0]): + img_np = imgs_fused[j].numpy().transpose(1, 2, 0) + img_path = os.path.join(result_sub_dir, 'fused_%d.png' % j) + imwrite(img_path, img_np) + print(f"Done. file name : {filename}") + +if __name__ == '__main__': + main() diff --git a/mlflow/sf2f/model_registry.py b/mlflow/sf2f/model_registry.py new file mode 100644 index 0000000..cb82b57 --- /dev/null +++ b/mlflow/sf2f/model_registry.py @@ -0,0 +1,65 @@ +from options.opts import options +import models +import torch +import os, glob +import mlflow +from utils.wav2mel import wav_to_mel +from datasets import imagenet_deprocess_batch, set_mel_transform, \ + deprocess_and_save, window_segment + +os.environ["MLFLOW_S3_ENDPOINT_URL"] = "http://localhost:9000" +os.environ["MLFLOW_TRACKING_URI"] = "http://localhost:5001" +os.environ["AWS_ACCESS_KEY_ID"] = "minio" +os.environ["AWS_SECRET_ACCESS_KEY"] = "miniostorage" +mlflow.set_experiment("new-exp") +model, _ = models.build_model( + options["generator"], + image_size=[128,128], + checkpoint_start_from="/home/hojun/Documents/project/boostcamp/final_project/mlops/voice2face-modeling/sf2f/data/best_IS_with_model.pt") +model.cuda().eval() + +voice_path = os.path.join("/home/hojun/Documents/project/boostcamp/final_project/mlops/voice2face-modeling/sf2f/data", '*.wav') +voice_list = glob.glob(voice_path) +filename = voice_list[0] +print(filename) + + + +#데이터 전처리 +mel_transform = set_mel_transform("vox_mel") +image_normalize_method = 'imagenet' +log_mel = wav_to_mel(filename) +float_dtype = torch.cuda.FloatTensor +log_mel = mel_transform(log_mel).type(float_dtype) + +log_mel_segs = window_segment( +log_mel, window_length=125, stride_length=63) +log_mel = log_mel.unsqueeze(0) + +with torch.no_grad(): + imgs_fused, others = model(log_mel_segs.unsqueeze(0)) + print("dksldklwkdlw : ",type(imgs_fused)) +if isinstance(imgs_fused, tuple): + imgs_fused = imgs_fused[-1] +imgs_fused = imgs_fused.cpu().detach() +imgs_fused = imagenet_deprocess_batch( + imgs_fused, normalize_method=image_normalize_method) +for j in range(imgs_fused.shape[0]): + img_np = imgs_fused[j].numpy().transpose(1, 2, 0) # 64x64x3 +# img_np = torch.from_numpy(img_np) +# log_mel_segs = log_mel_segs.unsqueeze(0).cpu().numpy() +print(type(img_np)) +print(type(log_mel_segs)) +signature = mlflow.models.signature.infer_signature(model_input=log_mel_segs.unsqueeze(0).cpu().numpy(), model_output=img_np) +input_sample = log_mel_segs.unsqueeze(0).cpu().numpy() + + + +with mlflow.start_run(): + mlflow.pytorch.log_model( + pytorch_model=model, + artifact_path = "sf2f_pytorch", + signature= signature, + input_example = input_sample, + pip_requirements = "rec.txt" + ) \ No newline at end of file diff --git a/mlflow/sf2f/models/__init__.py b/mlflow/sf2f/models/__init__.py new file mode 100644 index 0000000..694bee5 --- /dev/null +++ b/mlflow/sf2f/models/__init__.py @@ -0,0 +1,19 @@ +from .model_setup import * +from .encoder_decoder import EncoderDecoder + +# Lower level encoder decoders +from .voice_encoders import V2F1DCNN +from .face_decoders import V2FDecoder + +# Progressive GAN +# from .pgan import PGan_Net +# from .pgan_dis import PGAN_Discriminator + +# Discriminator +from .discriminators import PatchDiscriminator +from .discriminators import AcDiscriminator +from .discriminators import AcCropDiscriminator +from .discriminators import CondPatchDiscriminator + +# Facenet for Perceptual Loss and Evaluation +from .inception_resnet_v1 import InceptionResnetV1, fixed_image_standardization diff --git a/mlflow/sf2f/models/attention.py b/mlflow/sf2f/models/attention.py new file mode 100644 index 0000000..e74b2af --- /dev/null +++ b/mlflow/sf2f/models/attention.py @@ -0,0 +1,223 @@ +''' +Attenion implementation +Adapted from torchnlp: + https://github.com/PetrochukM/PyTorch-NLP/tree/master/torchnlp/nn +''' +import torch +import torch.nn as nn +import torch.nn.functional as F +from math import sqrt +try: + from .layers import get_activation, get_normalization_2d +except: + from layers import get_activation, get_normalization_2d + + +class AttnBlock(nn.Module): + def __init__(self, + attn_dim_in, + attn_dim_out, + attn_num_query=16, + attn_type='general', + normalization='none'): + super(AttnBlock, self).__init__() + self.attn_layer = AttnLayer( + attn_dim_in, + attn_dim_out, + attn_num_query, + attn_type, + normalization) + + def forward(self, cat_tensor, context): + N, C, H, W = cat_tensor.shape + attn_map = self.attn_layer(context.transpose(1, 2)) + #print('attn_map 1:', attn_map.shape) + attn_map = F.interpolate(attn_map, (H, W)) + #print('attn_map 2:', attn_map.shape) + map_out = torch.cat([cat_tensor, attn_map], 1) + + return map_out + + def return_attn_weights(self): + return self.attn_layer.attn_weights + + +class AttnLayer(nn.Module): + def __init__(self, + dim_in, + dim_out, + num_query=16, + attention_type='general', + normalization='none'): + super(AttnLayer, self).__init__() + self.num_query = num_query + self.query = nn.Parameter( + torch.randn((1, num_query, dim_in), + requires_grad=True)) + self.attention = Attention( + dimensions=dim_in, + dim_out=dim_out, + attention_type=attention_type) + self.norm_layer = get_normalization_2d( + dim_out, normalization) + # Cache Attenion Weights for visualization + self.attn_weights = None + + def forward(self, context): + N, L, D_in = context.shape + # Repeat query layer over batch dim + query = self.query.repeat(N, 1, 1) + # **output** (:class:`torch.FloatTensor` + # [batch size, output length, dim_out] + attn_out, attn_weights = self.attention(query, context) + self.attn_weights = attn_weights + N, Q_L, D_out = attn_out.shape + # [batch size, dim_out, output length] + attn_out = torch.transpose(attn_out, 1, 2) + # [batch size, dimensions, H, W] + H = W = int(sqrt(self.num_query)) + attn_map = attn_out.view( + N, D_out, H, W) + if self.norm_layer is not None: + attn_map = self.norm_layer(attn_map) + + return attn_map + + +class Attention(nn.Module): + """ Applies attention mechanism on the `context` using the `query`. + + **Thank you** to IBM for their initial implementation of :class:`Attention`. Here is + their `License + `__. + + Args: + dimensions (int): Dimensionality of the query and context. + attention_type (str, optional): How to compute the attention score: + + * dot: :math:`score(H_j,q) = H_j^T q` + * general: :math:`score(H_j, q) = H_j^T W_a q` + + Example: + + >>> attention = Attention(256) + >>> query = torch.randn(5, 1, 256) + >>> context = torch.randn(5, 5, 256) + >>> output, weights = attention(query, context) + >>> output.size() + torch.Size([5, 1, 256]) + >>> weights.size() + torch.Size([5, 1, 5]) + """ + + def __init__(self, + dimensions, + dim_out, + attention_type='general', + linear_out_layer=True, + ignore_tanh=False): + super(Attention, self).__init__() + + if attention_type not in ['dot', 'general']: + raise ValueError('Invalid attention type selected.') + + # Attention Setup + self.attention_type = attention_type + if self.attention_type == 'general': + self.linear_in = nn.Linear(dimensions, dimensions, bias=False) + self.softmax = nn.Softmax(dim=-1) + + # Output Setup + self.linear_out_layer = linear_out_layer + if self.linear_out_layer: + self.dim_out = dim_out + self.linear_out = nn.Linear(dimensions * 2, dim_out, bias=False) + + self.ignore_tanh = ignore_tanh + if not ignore_tanh: + self.tanh = nn.Tanh() + + def forward(self, query, context): + """ + Args: + query (:class:`torch.FloatTensor` [batch size, output length, + dimensions]): Sequence of queries to query the context. + context (:class:`torch.FloatTensor` [batch size, query length, + dimensions]): Data overwhich to apply the attention mechanism. + + Returns: + :class:`tuple` with `output` and `weights`: + * **output** (:class:`torch.LongTensor` [batch size, output length, + dimensions]): + Tensor containing the attended features. + * **weights** (:class:`torch.FloatTensor` [batch size, + output length, query length]): + Tensor containing attention weights. + """ + batch_size, output_len, dimensions = query.size() + query_len = context.size(1) + + if self.attention_type == "general": + query = query.reshape(batch_size * output_len, dimensions) + query = self.linear_in(query) + query = query.reshape(batch_size, output_len, dimensions) + + # TODO: Include mask on PADDING_INDEX? + + # (batch_size, output_len, dimensions) * \ + # (batch_size, query_len, dimensions) -> + # (batch_size, output_len, query_len) + # batch matrix-matrix product + attention_scores = torch.bmm( + query, context.transpose(1, 2).contiguous()) + + # Compute weights across every context sequence + attention_scores = attention_scores.view( + batch_size * output_len, query_len) + attention_weights = self.softmax(attention_scores) + attention_weights = attention_weights.view( + batch_size, output_len, query_len) + + # (batch_size, output_len, query_len) * \ + # (batch_size, query_len, dimensions) -> + # (batch_size, output_len, dimensions) + mix = torch.bmm(attention_weights, context) + + if self.linear_out_layer: + # Concatenate the attention output with the context and process the + # concatenated feature with a fc output layer + # concat -> (batch_size * output_len, 2*dimensions) + combined = torch.cat((mix, query), dim=2) + combined = combined.view(batch_size * output_len, 2 * dimensions) + # Apply linear_out on every 2nd dimension of concat + # output -> (batch_size, output_len, dimensions) + output = self.linear_out(combined).view( + batch_size, output_len, self.dim_out) + else: + # Directly output the attention output + output = mix + + if not self.ignore_tanh: + output = self.tanh(output) + + return output, attention_weights + + +if __name__ == '__main__': + attention = Attention(256, 100, linear_out_layer=False) + query = torch.randn(5, 1, 256) + context = torch.randn(5, 5, 256) + cat_tensor = torch.randn(5, 170, 128, 128) + output, weights = attention(query, context) + print('output.shape:', output.shape) + ''' + attn_layer = AttnLayer(256, 110, 16) + attn_map = attn_layer(context) + print('attn_map.shape:', attn_map.shape) + for param in attn_layer.parameters(): + print(type(param.data), param.size()) + + attn_block = AttnBlock(256, 110, 16, normalization='batch') + cat_map = attn_block(cat_tensor, context.transpose(1, 2)) + print('cat_map.shape:', cat_map.shape) + ''' diff --git a/mlflow/sf2f/models/crn.py b/mlflow/sf2f/models/crn.py new file mode 100644 index 0000000..33cd9a3 --- /dev/null +++ b/mlflow/sf2f/models/crn.py @@ -0,0 +1,152 @@ +import torch +import torch.nn as nn +import torch.nn.functional as F + +from .layers import get_normalization_2d +from .layers import get_activation + + +""" +Cascaded refinement network architecture, as described in: +Qifeng Chen and Vladlen Koltun, +"Photographic Image Synthesis with Cascaded Refinement Networks", +ICCV 2017 +""" + + +class RefinementModule(nn.Module): + def __init__(self, layout_dim, input_dim, output_dim, + normalization='instance', activation='leakyrelu'): + super(RefinementModule, self).__init__() + + layers = [] + layers.append(nn.Conv2d(layout_dim + input_dim, output_dim, + kernel_size=3, padding=1)) + layers.append(get_normalization_2d(output_dim, normalization)) + layers.append(get_activation(activation)) + layers.append(nn.Conv2d(output_dim, output_dim, + kernel_size=3, padding=1)) + layers.append(get_normalization_2d(output_dim, normalization)) + layers.append(get_activation(activation)) + layers = [layer for layer in layers if layer is not None] + for layer in layers: + if isinstance(layer, nn.Conv2d): + nn.init.kaiming_normal_(layer.weight) + self.net = nn.Sequential(*layers) + + def forward(self, layout, feats): + _, _, HH, WW = layout.size() + _, _, H, W = feats.size() + assert HH >= H + if HH > H: + factor = round(HH // H) + assert HH % factor == 0 + assert WW % factor == 0 and WW // factor == W + layout = F.avg_pool2d(layout, kernel_size=factor, stride=factor) + net_input = torch.cat([layout, feats], dim=1) + out = self.net(net_input) + return out + + +class RefinementNetwork(nn.Module): + def __init__(self, dims, normalization='instance', + activation='leakyrelu', + use_tanh=False, + use_deconv=False, + multi_resolution=False,): + super(RefinementNetwork, self).__init__() + layout_dim = dims[0] + self.refinement_modules = nn.ModuleList() + self.upsample_modules = nn.ModuleList() + self.multi_resolution = multi_resolution + for i in range(1, len(dims)): + input_dim = 1 if i == 1 else (dims[i] if use_deconv else dims[i-1]) + output_dim = dims[i] + mod = RefinementModule(layout_dim, input_dim, output_dim, + normalization=normalization, activation=activation) + self.refinement_modules.append(mod) + if use_deconv: + if i == 1: + mod = nn.Sequential() + else: + mod = nn.Sequential( + nn.ConvTranspose2d(dims[i-1], dims[i], + kernel_size=2, stride=2,), + get_normalization_2d(dims[i], normalization), + get_activation(activation) + ) + self.upsample_modules.append(mod) + output_conv_layers = None + if not multi_resolution: + output_conv_layers = [ + nn.Conv2d(dims[-1], dims[-1], kernel_size=3, padding=1), + get_activation(activation), + nn.Conv2d(dims[-1], 3, kernel_size=1, padding=0), + ] + if use_tanh: + ## added according to Pix2Pix-HD + output_conv_layers.append(nn.Tanh(),) + nn.init.kaiming_normal_(output_conv_layers[0].weight) + nn.init.kaiming_normal_(output_conv_layers[2].weight) + else: + output_conv_layers_low = [ + nn.Conv2d(dims[-2], dims[-2], kernel_size=3, padding=1), + get_activation(activation), + nn.Conv2d(dims[-2], 3, kernel_size=1, padding=0), + ] + output_conv_layers_high = [ + nn.Conv2d(dims[-1], dims[-1], kernel_size=3, padding=1), + get_activation(activation), + nn.Conv2d(dims[-1], 3, kernel_size=1, padding=0), + ] + if use_tanh: + ## added according to Pix2Pix-HD + output_conv_layers_low.append(nn.Tanh(),) + output_conv_layers_high.append(nn.Tanh(),) + + if output_conv_layers: + self.output_conv = nn.Sequential(*output_conv_layers) + else: + assert output_conv_layers_low + assert output_conv_layers_high + self.output_conv_low = nn.Sequential(*output_conv_layers_low) + self.output_conv_high = nn.Sequential(*output_conv_layers_high) + + def forward(self, layout): + """ + Output will have same size as layout + """ + # H, W = self.output_size + N, _, H, W = layout.size() + self.layout = layout + + # Figure out size of input + input_H, input_W = H, W + for _ in range(len(self.refinement_modules)): + input_H //= 2 + input_W //= 2 + + assert input_H != 0 + assert input_W != 0 + + feats = torch.zeros(N, 1, input_H, input_W).to(layout) + if not self.multi_resolution: + for i, mod in enumerate(self.refinement_modules): + if i == 0 or len(self.upsample_modules) == 0: + feats = F.interpolate(feats, scale_factor=2, mode='nearest') + else: + feats = self.upsample_modules[i](feats) + feats = mod(layout, feats) + out = self.output_conv(feats) + return (out, feats) + elif self.multi_resolution: + for i, mod in enumerate(self.refinement_modules): + if i == 0 or len(self.upsample_modules) == 0: + feats = F.interpolate(feats, scale_factor=2, mode='nearest') + else: + feats = self.upsample_modules[i](feats) + feats = mod(layout, feats) + if i == len(self.refinement_modules) - 2: + out_low = self.output_conv_low(feats) + out_high = self.output_conv_high(feats) + return (out_high, out_low) diff --git a/mlflow/sf2f/models/discriminators.py b/mlflow/sf2f/models/discriminators.py new file mode 100644 index 0000000..f0f276d --- /dev/null +++ b/mlflow/sf2f/models/discriminators.py @@ -0,0 +1,110 @@ +import torch +import torch.nn as nn +import torch.nn.functional as F + +from utils.bilinear import crop_bbox_batch, uncrop_bbox +from utils.box_utils import box_union, box_in_region +from .layers import GlobalAvgPool, Flatten, get_activation, build_cnn +import models + + +class PatchDiscriminator(nn.Module): + def __init__(self, arch, normalization='batch', activation='leakyrelu-0.2', + padding='same', pooling='avg', input_size=(128, 128), + layout_dim=0): + super(PatchDiscriminator, self).__init__() + input_dim = 3 + layout_dim + arch = 'I%d,%s' % (input_dim, arch) + cnn_kwargs = { + 'arch': arch, + 'normalization': normalization, + 'activation': activation, + 'pooling': pooling, + 'padding': padding, + } + self.cnn, output_dim = build_cnn(**cnn_kwargs) + self.classifier = nn.Conv2d(output_dim, 1, kernel_size=1, stride=1) + + def forward(self, x): + return self.classifier(self.cnn(x)) + + +class AcDiscriminator(nn.Module): + def __init__(self, arch, normalization='none', activation='relu', + padding='same', pooling='avg', num_id=None): + super(AcDiscriminator, self).__init__() + # self.vocab = vocab + + cnn_kwargs = { + 'arch': arch, + 'normalization': normalization, + 'activation': activation, + 'pooling': pooling, + 'padding': padding, + } + cnn, D = build_cnn(**cnn_kwargs) + self.cnn = nn.Sequential(cnn, GlobalAvgPool(), nn.Linear(D, 1024)) + self.num_id = num_id + + self.real_classifier = nn.Linear(1024, 1) + self.id_classifier = nn.Linear(1024, num_id) + + def forward(self, x, y): + if x.dim() == 3: + x = x[:, None] + + vecs = self.cnn(x) + real_scores = self.real_classifier(vecs) + id_scores = self.id_classifier(vecs) + ac_loss = F.cross_entropy(id_scores, y) + return real_scores, ac_loss + + +class AcCropDiscriminator(nn.Module): + def __init__(self, arch, normalization='none', activation='relu', + object_size=64, padding='same', pooling='avg'): + super(AcCropDiscriminator, self).__init__() + self.discriminator = AcDiscriminator(arch, normalization, + activation, padding, pooling) + self.object_size = object_size + + def forward(self, imgs, objs, boxes, obj_to_img, + object_crops=None, **kwargs): + if object_crops is None: + object_crops = crop_bbox_batch(imgs, boxes, obj_to_img, self.object_size) + real_scores, ac_loss = self.discriminator(object_crops, objs) + return real_scores, ac_loss, object_crops + + +# Conditional Discriminator +class CondPatchDiscriminator(nn.Module): + def __init__(self, arch, normalization='batch', activation='leakyrelu-0.2', + padding='same', pooling='avg', input_size=(128, 128), + cond_dim=0): + super(CondPatchDiscriminator, self).__init__() + input_dim = 3 + arch = 'I%d,%s' % (input_dim, arch) + cnn_kwargs = { + 'arch': arch, + 'normalization': normalization, + 'activation': activation, + 'pooling': pooling, + 'padding': padding, + } + self.cnn, output_dim = build_cnn(**cnn_kwargs) + self.joint_conv = nn.Sequential( + nn.Conv2d(output_dim+cond_dim, output_dim, \ + kernel_size=3, stride=1, padding=1, bias=False), + nn.BatchNorm2d(output_dim), + nn.LeakyReLU(0.2, inplace=True) + ) + self.classifier = nn.Conv2d(output_dim, 1, kernel_size=1, stride=1) + + def forward(self, x, cond_vecs): + if len(cond_vecs) == 2: + cond_vecs = cond_vecs.view(-1, cond_vecs.size(1), 1, 1) + x = self.cnn(x) + cond_vecs = cond_vecs.expand(-1, cond_vecs.size(1), x.size(2), x.size(3)) + x = torch.cat([x, cond_vecs], dim=1) + x = self.joint_conv(x) + return self.classifier(x) diff --git a/mlflow/sf2f/models/encoder_decoder.py b/mlflow/sf2f/models/encoder_decoder.py new file mode 100644 index 0000000..d564d86 --- /dev/null +++ b/mlflow/sf2f/models/encoder_decoder.py @@ -0,0 +1,87 @@ +''' +Flexible Encoder-Decoder Framework +''' + + +import torch +import torch.nn as nn +import functools +from torch.autograd import Variable +import numpy as np + +try: + from .model_collection import model_collection +except: + from model_collection import model_collection +#from .voice_encoders import V2F1DCNN +#from .face_decoders import V2FDecoder + + +class EncoderDecoder(nn.Module): + def __init__(self, + encoder_arch, + encoder_kwargs, + decoder_arch, + decoder_kwargs, + image_size): + super(EncoderDecoder, self).__init__() + # Initialize Encoder Decoder + self.encoder = getattr(model_collection, encoder_arch)(**encoder_kwargs) + self.decoder = getattr(model_collection, decoder_arch)(**decoder_kwargs) + #self.model = nn.Sequential( + # self.encoder, + # self.decoder + #) + + def forward(self, x, average_voice_embedding=False): + ''' + average_encoder_embedding: an option for evaluation, not needed for + training + ''' + others = {} + if self.encoder.return_seq: + emb, seq_emb = self.encoder(x) + imgs_pred = self.decoder(emb, seq_emb) + else: + emb = self.encoder(x) + if average_voice_embedding: + emb = torch.mean(emb, 0, keepdim=True) + imgs_pred = self.decoder(emb) + others['cond'] = emb + return imgs_pred, others + + def train_fuser_only(self): + print('Training Attention Fuser Only') + for name, param in self.named_parameters(): + if 'attn_fuser' in name: + param.requires_grad = True + else: + param.requires_grad = False + + +if __name__ == '__main__': + # Demo Input + log_mels = torch.ones((16, 40, 127)) + + # Test V2F 1D CNN + v2f_id_cnn_kwargs = { + 'input_channel': 40, + 'channels': [256, 384, 576, 864], + 'output_channel': 64, + } + # Test V2F 1D CNN + v2f_decoder_kwargs = { + 'input_channel': 64, + 'channels': [1024, 512, 256, 128, 64], + 'output_channel': 3, + } + + baseline_v2f = EncoderDecoder( + encoder_arch='V2F1DCNN', + encoder_kwargs=v2f_id_cnn_kwargs, + decoder_arch='V2FDecoder', + decoder_kwargs=v2f_decoder_kwargs + ) + + print(baseline_v2f) + print('baseline_v2f Output shape:', baseline_v2f(log_mels).shape) diff --git a/mlflow/sf2f/models/face_decoders.py b/mlflow/sf2f/models/face_decoders.py new file mode 100644 index 0000000..a05a8b7 --- /dev/null +++ b/mlflow/sf2f/models/face_decoders.py @@ -0,0 +1,583 @@ +''' +Implementations of different Face Decoders +''' + + +import torch +import torch.nn as nn +from torch.autograd import Variable +import torch.nn.functional as F +import functools +import numpy as np +from collections import OrderedDict +try: + from .attention import AttnBlock +except: + from attention import AttnBlock +try: + from .layers import get_activation, get_normalization_2d +except: + from layers import get_activation, get_normalization_2d +try: + from .networks import upBlock, GLU +except: + from networks import upBlock, GLU + + +class DeConv2DBLK(nn.Module): + def __init__(self, + input_channel, + output_channel, + kernel_size, + stride, + padding, + activation='relu', + normalization='none'): + super(DeConv2DBLK, self).__init__() + self.layers = [] + self.layers.append( + nn.ConvTranspose2d( + input_channel, + output_channel, + kernel_size, + stride, + padding, + bias=True)) + self.layers.append(get_activation(activation)) + self.layers.append( + get_normalization_2d( + output_channel, + normalization)) + self.layers = [layer for layer in self.layers if \ + layer is not None] + self.model = nn.Sequential(*self.layers) + + def forward(self, x): + x = self.model(x) + return x + + +class ExtraFeatureBlock(nn.Module): + def __init__(self, + dim_in, + dim_hidden, + normalization='none'): + super(ExtraFeatureBlock, self).__init__() + self.projection = nn.Linear(dim_in, dim_hidden, bias=True) + self.norm_layer = get_normalization_2d( + dim_hidden // 16, normalization) + + def forward(self, embedding, cat_tensor): + ''' + Transform the embedding, then concat it to the hidden tensor map + ''' + # Projected Embedding + #print('embedding:', embedding.shape) + N, D_in, _, _ = embedding.shape + proj_emb = self.projection(embedding.view(N, -1)) + N, D_hid = proj_emb.shape + feat_map = proj_emb.view(N, -1, 4, 4) + if self.norm_layer is not None: + feat_map = self.norm_layer(feat_map) + N, C, H, W = cat_tensor.shape + feat_map = F.interpolate(feat_map, (H, W)) + map_out = torch.cat([cat_tensor, feat_map], 1) + + return map_out + + +class V2FDecoder(nn.Module): + def __init__(self, + input_channel, + channels, + output_channel, + normalization='none'): + super(V2FDecoder, self).__init__() + output_size = 2 ** (len(channels) + 1) + print('V2FDecoder: According to your channels list, ' + \ + 'the output will be generated with shape ({}, {})'.format( + output_size, output_size)) + #torch.nn.ConvTranspose2d( + #in_channels, out_channels, kernel_size, stride=1, padding=0, + #output_padding=0, groups=1, bias=True, + #dilation=1, padding_mode='zeros') + self.layers = [] + # First Layer + layer_id = 0 + self.layers.append( + DeConv2DBLK(input_channel, + channels[0], + kernel_size=4, + stride=1, + padding=0, + activation='relu', + normalization=normalization)) + # Hidden Layers + for layer_id in range(1, len(channels)): + self.layers.append( + DeConv2DBLK(channels[layer_id - 1], + channels[layer_id], + kernel_size=4, + stride=2, + padding=1, + activation='relu', + normalization=normalization)) + # Output Layer + layer_id = layer_id + 1 + self.layers.append( + DeConv2DBLK(channels[layer_id - 1], + output_channel, + kernel_size=1, + stride=1, + padding=0, + activation='none', + normalization='none')) + self.model = nn.Sequential(*self.layers) + + def forward(self, x): + x = self.model(x) + return x + + +class MRDecoder(nn.Module): + def __init__(self, + input_channel, + base_block_channels, + mid_block_channel=32, + high_block_channel=None, + output_channel=3, + normalization='none', + use_up_block=False, + extra_feature_map=False): + super(MRDecoder, self).__init__() + output_sizes = [] + output_sizes.append( + 2 ** (len(base_block_channels) + 1)) + output_sizes.append( + 2 ** (len(base_block_channels) + 2)) + if high_block_channel is not None: + self.build_high_img = True + output_sizes.append( + 2 ** (len(base_block_channels) + 3)) + else: + self.build_high_img = False + print('MRDecoder: According to your channels list:') + for i, output_size in enumerate(output_sizes): + print('\tOutput {} will be generated with shape ({}, {})'.format( + i, output_size, output_size)) + + self.input_channel = input_channel + self.base_block_channels = base_block_channels + self.mid_block_channel = mid_block_channel + self.high_block_channel = high_block_channel + self.output_channel = output_channel + self.normalization = normalization + self.use_up_block = use_up_block + self.extra_feature_map = extra_feature_map + + # Output Layer + self.build_base_block() + self.build_low_block() + self.build_mid_block() + if self.build_high_img: + self.build_high_block() + + def build_base_block(self): + #torch.nn.ConvTranspose2d( + #in_channels, out_channels, kernel_size, stride=1, padding=0, + #output_padding=0, groups=1, bias=True, + #dilation=1, padding_mode='zeros') + self.base_block_layers = [] + # First Layer + layer_id = 0 + self.base_block_layers.append( + DeConv2DBLK(self.input_channel, + self.base_block_channels[0], + kernel_size=4, + stride=1, + padding=0, + activation='relu', + normalization=self.normalization)) + # Hidden Layers + for layer_id in range(1, len(self.base_block_channels)): + self.base_block_layers.append( + DeConv2DBLK(self.base_block_channels[layer_id - 1], + self.base_block_channels[layer_id], + kernel_size=4, + stride=2, + padding=1, + activation='relu', + normalization=self.normalization)) + self.base_block = nn.Sequential(*self.base_block_layers) + + def build_low_block(self): + self.low_output_layer = \ + DeConv2DBLK(self.base_block_channels[-1], + self.output_channel, + kernel_size=1, + stride=1, + padding=0, + activation='none', + normalization='none') + + def build_mid_block(self): + extra_channel = 0 + if self.extra_feature_map: + # Mid-resolution ExtraFeatureBlock + self.mid_ef_block = ExtraFeatureBlock( + dim_in=self.input_channel, + dim_hidden=self.input_channel, + normalization=self.normalization) + extra_channel = self.input_channel // 16 + if self.use_up_block: + self.mid_hidden_layer = \ + upBlock(self.base_block_channels[-1] + extra_channel, + self.mid_block_channel) + else: + self.mid_hidden_layer = \ + DeConv2DBLK(self.base_block_channels[-1] + extra_channel, + self.mid_block_channel, + kernel_size=4, + stride=2, + padding=1, + activation='relu', + normalization=self.normalization) + self.mid_output_layer = \ + DeConv2DBLK(self.mid_block_channel, + self.output_channel, + kernel_size=1, + stride=1, + padding=0, + activation='none', + normalization='none') + + def build_high_block(self): + extra_channel = 0 + if self.extra_feature_map: + # High-resolution ExtraFeatureBlock + self.high_ef_block = ExtraFeatureBlock( + dim_in=self.input_channel, + dim_hidden=self.input_channel, + normalization=self.normalization) + extra_channel = self.input_channel // 16 + if self.use_up_block: + self.high_hidden_layer = \ + upBlock(self.mid_block_channel + extra_channel, + self.high_block_channel) + else: + self.high_hidden_layer = \ + DeConv2DBLK(self.mid_block_channel + extra_channel, + self.high_block_channel, + kernel_size=4, + stride=2, + padding=1, + activation='relu', + normalization=self.normalization) + self.high_output_layer = \ + DeConv2DBLK(self.high_block_channel, + self.output_channel, + kernel_size=1, + stride=1, + padding=0, + activation='none', + normalization='none') + + def forward(self, x): + low_code = self.base_block(x) + low_imgs = self.low_output_layer(low_code) + #print('low_code:', low_code.shape) + if self.extra_feature_map: + low_code = self.mid_ef_block(x, low_code) + #print('low_code:', low_code.shape) + mid_code = self.mid_hidden_layer(low_code) + mid_imgs = self.mid_output_layer(mid_code) + if self.build_high_img: + #print('mid_code:', mid_code.shape) + if self.extra_feature_map: + mid_code = self.high_ef_block(x, mid_code) + #print('mid_code:', mid_code.shape) + high_code = self.high_hidden_layer(mid_code) + high_imgs = self.high_output_layer(high_code) + return (low_imgs, mid_imgs, high_imgs) + else: + return (low_imgs, mid_imgs) + + +class AttnV2FDecoder(nn.Module): + def __init__(self, + input_channel, + cnn_channels, + attn_channels, + output_channel, + seq_in_dim, + normalization='none'): + super(AttnV2FDecoder, self).__init__() + output_size = 2 ** (len(cnn_channels) + 1) + print('V2FDecoder: According to your channels list, ' + \ + 'the output will be generated with shape ({}, {})'.format( + output_size, output_size)) + #torch.nn.ConvTranspose2d( + #in_channels, out_channels, kernel_size, stride=1, padding=0, + #output_padding=0, groups=1, bias=True, + #dilation=1, padding_mode='zeros') + self.layers = [] + self.layer_names = [] + # First Layer + layer_id = 0 + self.layers.append( + DeConv2DBLK(input_channel, + cnn_channels[0], + kernel_size=4, + stride=1, + padding=0, + activation='relu', + normalization=normalization)) + self.layer_names.append('DeConv2DBLK{}'.format(layer_id)) + # Hidden Layers + for layer_id in range(1, len(cnn_channels)): + self.layers.append( + DeConv2DBLK( + cnn_channels[layer_id - 1]+attn_channels[layer_id - 1], + cnn_channels[layer_id], + kernel_size=4, + stride=2, + padding=1, + activation='relu', + normalization=normalization)) + self.layer_names.append('DeConv2DBLK{}'.format(layer_id)) + attn_channel = attn_channels[layer_id] + if attn_channel > 0: + self.layers.append( + AttnBlock(attn_dim_in=seq_in_dim, + attn_dim_out=attn_channel, + attn_num_query=16, + normalization=normalization)) + self.layer_names.append('AttnBlock{}'.format(layer_id)) + # Output Layer + layer_id = layer_id + 1 + self.layers.append( + DeConv2DBLK(cnn_channels[layer_id - 1]+attn_channels[layer_id - 1], + output_channel, + kernel_size=1, + stride=1, + padding=0, + activation='none', + normalization='none')) + self.layer_names.append('DeConv2DBLK{}'.format(layer_id)) + + for layer_id, layer in enumerate(self.layers): + self.add_module(self.layer_names[layer_id], layer) + + def forward(self, x, seq_feat): + for layer, name in zip(self.layers, self.layer_names): + if 'AttnBlock' in name: + x = layer(x, seq_feat) + else: + x = layer(x) + #print(x.shape) + return x + + def return_attn_weights(self): + for layer, name in zip(self.layers, self.layer_names): + if 'AttnBlock' in name: + return layer.return_attn_weights() + + +class FaceGanDecoder(nn.Module): + def __init__(self, + image_size=(64, 64), + normalization='batch', + activation='leakyrelu-0.2', + mlp_normalization='none', + noise_dim=128, + **kwargs): + + super(FaceGanDecoder, self).__init__() + if len(kwargs) > 0: + print('WARNING: Model got unexpected kwargs ', kwargs) + + self.dim = noise_dim + self.image_size = image_size + + self.fc = nn.Sequential( + nn.Linear(self.dim, self.dim * 4 * 4 * 2, bias=False), + nn.BatchNorm1d(self.dim * 4 * 4 * 2), + GLU()) + + up_num, up_layers = [], [] + num_in, num_out = 1, 1 + up_num = range(1, int(np.log2(image_size[0]))-1) + + for i, num in enumerate(up_num): + if i == 0: + num_in = num + num_out = 2**num + up_layers.append(upBlock(self.dim//num_in, self.dim//num_out)) + num_in = num_out + + self.up_layers = nn.Sequential(*up_layers) + + gnet_layers = [ + nn.Conv2d(self.dim//num_out, self.dim//num_out, kernel_size=3, padding=1), + get_activation('leakyrelu'), + nn.Conv2d(self.dim//num_out, 3, kernel_size=1, padding=0), + ] + self.gnet = nn.Sequential(*gnet_layers) + + def forward(self, x): + """ + Required Inputs: + - placeholder: This is a placeholder. + + Optional Inputs: + - placeholder: This is a placeholder. + """ + + vecs = self.fc(x.view(-1, 512)) + vecs = vecs.view(x.size()[0], -1, 4, 4) + + img_code = self.up_layers(vecs) + img_pred = self.gnet(img_code) + + return img_pred + + +class FaceGanDecoder_v2(nn.Module): + def __init__(self, + image_size=(64, 64), + normalization='batch', + activation='leakyrelu-0.2', + mlp_normalization='none', + noise_dim=128, + **kwargs): + + super(FaceGanDecoder_v2, self).__init__() + if len(kwargs) > 0: + print('WARNING: Model got unexpected kwargs ', kwargs) + + self.dim = noise_dim + self.image_size = image_size + + self.fc = nn.Sequential( + nn.Linear(self.dim, self.dim * 4 * 4 * 2, bias=False), + nn.BatchNorm1d(self.dim * 4 * 4 * 2), + GLU()) + + up_num, up_layers = [], [] + num_in, num_out = 1, 1 + up_num = range(1, int(np.log2(image_size[0]))-1) + + for i, num in enumerate(up_num): + if i == 0: + num_in = num + num_out = 2**num + up_layers.append(upBlock(self.dim//num_in, self.dim//num_out)) + num_in = num_out + + self.up_low_layers = nn.Sequential(*up_layers) + self.up_mid_layer = upBlock(self.dim//num_out, self.dim//num_out//2) + + gnet_layers = [ + nn.Conv2d(self.dim//num_out, self.dim//num_out, kernel_size=3, padding=1), + get_activation('leakyrelu'), + nn.Conv2d(self.dim//num_out, 3, kernel_size=1, padding=0), + ] + self.gnet_low = nn.Sequential(*gnet_layers) + + gnet_layers = [ + nn.Conv2d(self.dim//num_out//2, self.dim//num_out//2, kernel_size=3, padding=1), + get_activation('leakyrelu'), + nn.Conv2d(self.dim//num_out//2, 3, kernel_size=1, padding=0), + ] + self.gnet_mid = nn.Sequential(*gnet_layers) + + def forward(self, x): + """ + Required Inputs: + - placeholder: This is a placeholder. + + Optional Inputs: + - placeholder: This is a placeholder. + """ + + vecs = self.fc(x.view(-1, 512)) + vecs = vecs.view(x.size()[0], -1, 4, 4) + + img_code_low = self.up_low_layers(vecs) + img_pred_low = self.gnet_low(img_code_low) + + img_code_mid = self.up_mid_layer(img_code_low) + img_pred_mid = self.gnet_mid(img_code_mid) + + return (img_pred_low, img_pred_mid) + + +if __name__ == '__main__': + # Demo Input + voice_embeddings = torch.ones((16, 512, 1, 1)) + seq_voice_embeddings = torch.ones((16, 512, 270)) + + # Test V2F 1D CNN + mr_decoder_kwargs = { + 'input_channel': 512, + 'base_block_channels': [1024, 512, 256, 128, 64], + 'mid_block_channel': 32, + 'high_block_channel': 32, + 'output_channel': 3, + 'normalization': 'batch', + 'use_up_block': True, + 'extra_feature_map': True, + } + mr_decoder = MRDecoder(**mr_decoder_kwargs) + print(mr_decoder) + for i, imgs in enumerate(mr_decoder(voice_embeddings)): + print('MRDecoder Output {} shape:'.format(i), imgs.shape) + + # Test V2F Decoder + v2f_decoder_kwargs = { + 'input_channel': 512, + 'channels': [1024, 512, 256, 128, 64], + 'output_channel': 3, + 'normalization': 'batch', + } + v2f_decoder = V2FDecoder(**v2f_decoder_kwargs) + print(v2f_decoder) + print('V2FDecoder Output shape:', v2f_decoder(voice_embeddings).shape) + print('V2F Sequential:') + print(v2f_decoder.model[5]) + for name, param in v2f_decoder.model[5].named_parameters(): + print(name) + #for module in v2f_decoder.model: + # print(module) + + v2f_decoder_kwargs = { + 'input_channel': 512, + 'cnn_channels': [1024, 512, 256, 128, 64], + 'attn_channels': [0, 0, 128, 0, 32], + 'output_channel': 3, + 'seq_in_dim': 512, + 'normalization': 'none', + } + attn_v2f_decoder = AttnV2FDecoder(**v2f_decoder_kwargs) + print(attn_v2f_decoder) + print('AttnV2FDecoder Output shape:', \ + attn_v2f_decoder(voice_embeddings, seq_voice_embeddings).shape) + print('Return Attention Weights:', + attn_v2f_decoder.return_attn_weights().shape, + attn_v2f_decoder.return_attn_weights()[0][0][0:10]) + + dimLatentVector = 512 + depthScale0 = 512 + + pgandecoder = PganDecoder(dimLatent=dimLatentVector, + depthScale0=depthScale0, + initBiasToZero=True, + leakyReluLeak=0.2, + normalization=True, + generationActivation=None, + dimOutput=3, + equalizedlR=True, + depthOtherScales=[512,512,512,256]) + + img = pgandecoder(voice_embeddings) + diff --git a/mlflow/sf2f/models/fusers.py b/mlflow/sf2f/models/fusers.py new file mode 100644 index 0000000..ecc8043 --- /dev/null +++ b/mlflow/sf2f/models/fusers.py @@ -0,0 +1,136 @@ +''' +Implementations of different Fusers +''' + + +import os +import torch +import torch.nn as nn +import torch.optim as optim +import torch.nn.functional as F +import numpy as np +import math + +try: + from attention import Attention +except: + from .attention import Attention + + +class AttentionFuserV1(nn.Module): + def __init__(self, + dimensions, + dim_out, + attention_type='general', + linear_out_layer=True, + ignore_tanh=False): + ''' + 1 Layer Attnetion Fuser + ''' + nn.Module.__init__(self) + self.attention = Attention( + dimensions, + dim_out, + attention_type=attention_type, + linear_out_layer=linear_out_layer, + ignore_tanh=ignore_tanh) + + def forward(self, embeddings): + ''' + seq_in: tensor with shape [N, D, L] + ''' + # Embeddings: [B, N, C] + embeddings, attention_weights = \ + self.attention(embeddings, embeddings) + # [B, C, N] + embeddings = embeddings.transpose(1, 2) + #print('Transposed Shape:', embeddings.shape) + embeddings = F.avg_pool1d( + embeddings, embeddings.size()[2], stride=1) + embeddings = embeddings.view(embeddings.size()[0], -1, 1, 1) + #print('Attention Fuser Output Shape:', embeddings.shape) + return embeddings + + +class AttentionFuserV2(nn.Module): + def __init__(self, + dimensions, + attention_type='general'): + ''' + 2-Layer Attention + ''' + nn.Module.__init__(self) + self.attention_1 = Attention( + dimensions, + dimensions, + attention_type=attention_type, + linear_out_layer=True, + ignore_tanh=False) + self.attention_2 = Attention( + dimensions, + dimensions, + attention_type=attention_type, + linear_out_layer=True, + ignore_tanh=True) + + def forward(self, embeddings): + ''' + seq_in: tensor with shape [N, D, L] + ''' + # Embeddings: [B, N, C] + embeddings, attention_weights = \ + self.attention_1(embeddings, embeddings) + embeddings, attention_weights = \ + self.attention_2(embeddings, embeddings) + # [B, C, N] + embeddings = embeddings.transpose(1, 2) + #print('Transposed Shape:', embeddings.shape) + embeddings = F.avg_pool1d( + embeddings, embeddings.size()[2], stride=1) + embeddings = embeddings.view(embeddings.size()[0], -1, 1, 1) + #print('Attention Fuser Output Shape:', embeddings.shape) + return embeddings + + +class AttentionFuserV3(nn.Module): + def __init__(self, + dimensions, + attention_type='general'): + ''' + 2-Layer Attention with Residual Connection + ''' + nn.Module.__init__(self) + self.attention_1 = Attention( + dimensions, + dimensions, + attention_type=attention_type, + linear_out_layer=True, + ignore_tanh=False) + self.attention_2 = Attention( + dimensions*2, + dimensions, + attention_type=attention_type, + linear_out_layer=True, + ignore_tanh=True) + + def forward(self, input): + ''' + seq_in: tensor with shape [N, D, L] + ''' + # Embeddings: [B, N, C] + hidden, attention_weights = \ + self.attention_1(input, input) + hidden = F.normalize(hidden, dim=2) + # Residual Connection + combined = torch.cat((hidden, input), dim=2) + # 2nd Attention + embeddings, attention_weights = \ + self.attention_2(combined, combined) + # [B, C, N] + embeddings = embeddings.transpose(1, 2) + #print('Transposed Shape:', embeddings.shape) + embeddings = F.avg_pool1d( + embeddings, embeddings.size()[2], stride=1) + embeddings = embeddings.view(embeddings.size()[0], -1, 1, 1) + #print('Attention Fuser Output Shape:', embeddings.shape) + return embeddings diff --git a/mlflow/sf2f/models/inception_resnet_v1.py b/mlflow/sf2f/models/inception_resnet_v1.py new file mode 100644 index 0000000..f95a1f8 --- /dev/null +++ b/mlflow/sf2f/models/inception_resnet_v1.py @@ -0,0 +1,439 @@ +''' +Reference: https://github.com/timesler/facenet-pytorch/ \ + blob/master/models/inception_resnet_v1.py + +Both pretrained models were trained on 160x160 px images, +so will perform best if applied to images resized to this shape. +For best results, images should also be cropped to the face using +MTCNN (see below). +''' + + +import torch +from torch import nn +from torch.nn import functional as F +import requests +from requests.adapters import HTTPAdapter +import os + + +class BasicConv2d(nn.Module): + + def __init__(self, in_planes, out_planes, kernel_size, stride, padding=0): + super().__init__() + self.conv = nn.Conv2d( + in_planes, out_planes, + kernel_size=kernel_size, stride=stride, + padding=padding, bias=False + ) # verify bias false + self.bn = nn.BatchNorm2d( + out_planes, + eps=0.001, # value found in tensorflow + momentum=0.1, # default pytorch value + affine=True + ) + self.relu = nn.ReLU(inplace=False) + + def forward(self, x): + x = self.conv(x) + x = self.bn(x) + x = self.relu(x) + return x + + +class Block35(nn.Module): + + def __init__(self, scale=1.0): + super().__init__() + + self.scale = scale + + self.branch0 = BasicConv2d(256, 32, kernel_size=1, stride=1) + + self.branch1 = nn.Sequential( + BasicConv2d(256, 32, kernel_size=1, stride=1), + BasicConv2d(32, 32, kernel_size=3, stride=1, padding=1) + ) + + self.branch2 = nn.Sequential( + BasicConv2d(256, 32, kernel_size=1, stride=1), + BasicConv2d(32, 32, kernel_size=3, stride=1, padding=1), + BasicConv2d(32, 32, kernel_size=3, stride=1, padding=1) + ) + + self.conv2d = nn.Conv2d(96, 256, kernel_size=1, stride=1) + self.relu = nn.ReLU(inplace=False) + + def forward(self, x): + x0 = self.branch0(x) + x1 = self.branch1(x) + x2 = self.branch2(x) + out = torch.cat((x0, x1, x2), 1) + out = self.conv2d(out) + out = out * self.scale + x + out = self.relu(out) + return out + + +class Block17(nn.Module): + + def __init__(self, scale=1.0): + super().__init__() + + self.scale = scale + + self.branch0 = BasicConv2d(896, 128, kernel_size=1, stride=1) + + self.branch1 = nn.Sequential( + BasicConv2d(896, 128, kernel_size=1, stride=1), + BasicConv2d(128, 128, kernel_size=(1,7), stride=1, padding=(0,3)), + BasicConv2d(128, 128, kernel_size=(7,1), stride=1, padding=(3,0)) + ) + + self.conv2d = nn.Conv2d(256, 896, kernel_size=1, stride=1) + self.relu = nn.ReLU(inplace=False) + + def forward(self, x): + x0 = self.branch0(x) + x1 = self.branch1(x) + out = torch.cat((x0, x1), 1) + out = self.conv2d(out) + out = out * self.scale + x + out = self.relu(out) + return out + + +class Block8(nn.Module): + + def __init__(self, scale=1.0, noReLU=False): + super().__init__() + + self.scale = scale + self.noReLU = noReLU + + self.branch0 = BasicConv2d(1792, 192, kernel_size=1, stride=1) + + self.branch1 = nn.Sequential( + BasicConv2d(1792, 192, kernel_size=1, stride=1), + BasicConv2d(192, 192, kernel_size=(1,3), stride=1, padding=(0,1)), + BasicConv2d(192, 192, kernel_size=(3,1), stride=1, padding=(1,0)) + ) + + self.conv2d = nn.Conv2d(384, 1792, kernel_size=1, stride=1) + if not self.noReLU: + self.relu = nn.ReLU(inplace=False) + + def forward(self, x): + x0 = self.branch0(x) + x1 = self.branch1(x) + out = torch.cat((x0, x1), 1) + out = self.conv2d(out) + out = out * self.scale + x + if not self.noReLU: + out = self.relu(out) + return out + + +class Mixed_6a(nn.Module): + + def __init__(self): + super().__init__() + + self.branch0 = BasicConv2d(256, 384, kernel_size=3, stride=2) + + self.branch1 = nn.Sequential( + BasicConv2d(256, 192, kernel_size=1, stride=1), + BasicConv2d(192, 192, kernel_size=3, stride=1, padding=1), + BasicConv2d(192, 256, kernel_size=3, stride=2) + ) + + self.branch2 = nn.MaxPool2d(3, stride=2) + + def forward(self, x): + x0 = self.branch0(x) + x1 = self.branch1(x) + x2 = self.branch2(x) + out = torch.cat((x0, x1, x2), 1) + return out + + +class Mixed_7a(nn.Module): + + def __init__(self): + super().__init__() + + self.branch0 = nn.Sequential( + BasicConv2d(896, 256, kernel_size=1, stride=1), + BasicConv2d(256, 384, kernel_size=3, stride=2) + ) + + self.branch1 = nn.Sequential( + BasicConv2d(896, 256, kernel_size=1, stride=1), + BasicConv2d(256, 256, kernel_size=3, stride=2) + ) + + self.branch2 = nn.Sequential( + BasicConv2d(896, 256, kernel_size=1, stride=1), + BasicConv2d(256, 256, kernel_size=3, stride=1, padding=1), + BasicConv2d(256, 256, kernel_size=3, stride=2) + ) + + self.branch3 = nn.MaxPool2d(3, stride=2) + + def forward(self, x): + x0 = self.branch0(x) + x1 = self.branch1(x) + x2 = self.branch2(x) + x3 = self.branch3(x) + out = torch.cat((x0, x1, x2, x3), 1) + return out + + +class InceptionResnetV1(nn.Module): + """Inception Resnet V1 model with optional loading of pretrained weights. + + Model parameters can be loaded based on pretraining on the VGGFace2 or CASIA-Webface + datasets. Pretrained state_dicts are automatically downloaded on model instantiation if + requested and cached in the torch cache. Subsequent instantiations use the cache rather than + redownloading. + + Keyword Arguments: + pretrained {str} -- Optional pretraining dataset. Either 'vggface2' or 'casia-webface'. + (default: {None}) + classify {bool} -- Whether the model should output classification probabilities or feature + embeddings. (default: {False}) + num_classes {int} -- Number of output classes. If 'pretrained' is set and num_classes not + equal to that used for the pretrained model, the final linear layer will be randomly + initialized. (default: {None}) + dropout_prob {float} -- Dropout probability. (default: {0.6}) + """ + def __init__(self, + pretrained=None, + classify=False, + num_classes=None, + dropout_prob=0.6, + device=None, + auto_input_resize=True, + return_pooling=False): + super().__init__() + + # Set simple attributes + self.pretrained = pretrained + self.classify = classify + self.num_classes = num_classes + self.auto_input_resize = auto_input_resize + + if pretrained == 'vggface2': + tmp_classes = 8631 + elif pretrained == 'casia-webface': + tmp_classes = 10575 + elif pretrained is None and self.num_classes is None: + raise Exception('At least one of "pretrained" or "num_classes" must be specified') + else: + tmp_classes = self.num_classes + + if self.auto_input_resize: + self.input_resize = nn.UpsamplingBilinear2d(size=(140, 140)) + self.return_pooling = return_pooling + # Define layers + self.conv2d_1a = BasicConv2d(3, 32, kernel_size=3, stride=2) + self.conv2d_2a = BasicConv2d(32, 32, kernel_size=3, stride=1) + self.conv2d_2b = BasicConv2d(32, 64, kernel_size=3, stride=1, padding=1) + self.maxpool_3a = nn.MaxPool2d(3, stride=2) + self.conv2d_3b = BasicConv2d(64, 80, kernel_size=1, stride=1) + self.conv2d_4a = BasicConv2d(80, 192, kernel_size=3, stride=1) + self.conv2d_4b = BasicConv2d(192, 256, kernel_size=3, stride=2) + self.repeat_1 = nn.Sequential( + Block35(scale=0.17), + Block35(scale=0.17), + Block35(scale=0.17), + Block35(scale=0.17), + Block35(scale=0.17), + ) + self.mixed_6a = Mixed_6a() + self.repeat_2 = nn.Sequential( + Block17(scale=0.10), + Block17(scale=0.10), + Block17(scale=0.10), + Block17(scale=0.10), + Block17(scale=0.10), + Block17(scale=0.10), + Block17(scale=0.10), + Block17(scale=0.10), + Block17(scale=0.10), + Block17(scale=0.10), + ) + self.mixed_7a = Mixed_7a() + self.repeat_3 = nn.Sequential( + Block8(scale=0.20), + Block8(scale=0.20), + Block8(scale=0.20), + Block8(scale=0.20), + Block8(scale=0.20), + ) + self.block8 = Block8(noReLU=True) + self.avgpool_1a = nn.AdaptiveAvgPool2d(1) + self.dropout = nn.Dropout(dropout_prob) + self.last_linear = nn.Linear(1792, 512, bias=False) + self.last_bn = nn.BatchNorm1d(512, eps=0.001, momentum=0.1, affine=True) + self.logits = nn.Linear(512, tmp_classes) + + if pretrained is not None: + load_weights(self, pretrained) + + if self.num_classes is not None: + self.logits = nn.Linear(512, self.num_classes) + + self.device = torch.device('cpu') + if device is not None: + self.device = device + self.to(device) + + def forward(self, x): + """ + Calculate embeddings or probabilities given a batch of input image + tensors. + + Arguments: + x {torch.tensor} -- Batch of image tensors representing faces. + + Returns: + torch.tensor -- Batch of embeddings or softmax probabilities. + """ + if self.auto_input_resize: + x = self.input_resize(x) + x = self.conv2d_1a(x) + x = self.conv2d_2a(x) + x = self.conv2d_2b(x) + x = self.maxpool_3a(x) + x = self.conv2d_3b(x) + x = self.conv2d_4a(x) + x = self.conv2d_4b(x) + x = self.repeat_1(x) + x = self.mixed_6a(x) + x = self.repeat_2(x) + x = self.mixed_7a(x) + x = self.repeat_3(x) + x = self.block8(x) + x = self.avgpool_1a(x) + if self.return_pooling: + pooled = x.view(x.shape[0], -1) + pooled = F.normalize(pooled, p=2, dim=1) + return pooled + x = self.dropout(x) + x = self.last_linear(x.view(x.shape[0], -1)) + x = self.last_bn(x) + x = F.normalize(x, p=2, dim=1) + if self.classify: + x = self.logits(x) + return x + + def get_hidden_states(self, x): + """ + Calculate embeddings or probabilities given a batch of input image + tensors. + + Arguments: + x {torch.tensor} -- Batch of image tensors representing faces. + + Returns: + torch.tensor -- Batch of embeddings or softmax probabilities. + """ + if self.auto_input_resize: + x = self.input_resize(x) + conv2d_1a = self.conv2d_1a(x) + conv2d_2a = self.conv2d_2a(conv2d_1a) + conv2d_2b = self.conv2d_2b(conv2d_2a) + maxpool_3a = self.maxpool_3a(conv2d_2b) + conv2d_3b = self.conv2d_3b(maxpool_3a) + conv2d_4a = self.conv2d_4a(conv2d_3b) + conv2d_4b = self.conv2d_4b(conv2d_4a) + repeat_1 = self.repeat_1(conv2d_4b) + mixed_6a = self.mixed_6a(repeat_1) + repeat_2 = self.repeat_2(mixed_6a) + mixed_7a = self.mixed_7a(repeat_2) + repeat_3 = self.repeat_3(mixed_7a) + block8 = self.block8(repeat_3) + avgpool_1a = self.avgpool_1a(block8) + dropout = self.dropout(avgpool_1a) + last_linear = self.last_linear(dropout.view(x.shape[0], -1)) + last_bn = self.last_bn(last_linear) + x = F.normalize(last_bn, p=2, dim=1) + if self.classify: + x = self.logits(x) + out = [repeat_1, repeat_2, repeat_3, x] + return out + + +def load_weights(mdl, name): + """Download pretrained state_dict and load into model. + + Arguments: + mdl {torch.nn.Module} -- Pytorch model. + name {str} -- Name of dataset that was used to generate pretrained state_dict. + + Raises: + ValueError: If 'pretrained' not equal to 'vggface2' or 'casia-webface'. + """ + if name == 'vggface2': + features_path = \ + 'https://drive.google.com/uc?export=download&' + \ + 'id=1cWLH_hPns8kSfMz9kKl9PsG5aNV2VSMn' + logits_path = \ + 'https://drive.google.com/uc?export=download' + \ + '&id=1mAie3nzZeno9UIzFXvmVZrDG3kwML46X' + elif name == 'casia-webface': + features_path = 'https://drive.google.com/uc?export=download&id=1LSHHee_IQj5W3vjBcRyVaALv4py1XaGy' + logits_path = 'https://drive.google.com/uc?export=download&id=1QrhPgn1bGlDxAil2uc07ctunCQoDnCzT' + else: + raise ValueError('Pretrained models only exist for "vggface2" and "casia-webface"') + + model_dir = os.path.join(get_torch_home(), 'checkpoints') + os.makedirs(model_dir, exist_ok=True) + + state_dict = {} + for i, path in enumerate([features_path, logits_path]): + cached_file = os.path.join( + model_dir, '{}_{}.pt'.format(name, path[-10:])) + #print('Cashed File Location:', cached_file) + if not os.path.exists(cached_file): + print('Downloading parameters ({}/2)'.format(i+1)) + s = requests.Session() + s.mount('https://', HTTPAdapter(max_retries=10)) + r = s.get(path, allow_redirects=True) + with open(cached_file, 'wb') as f: + f.write(r.content) + state_dict.update(torch.load(cached_file)) + + mdl.load_state_dict(state_dict) + + +def get_torch_home(): + torch_home = os.path.expanduser( + os.getenv( + 'TORCH_HOME', + os.path.join(os.getenv('XDG_CACHE_HOME', '~/.cache'), 'torch') + ) + ) + return torch_home + + +def fixed_image_standardization(image_tensor): + ''' + This is the default preprocess function for this pretrained model + ''' + processed_tensor = (image_tensor - 127.5) / 128.0 + return processed_tensor + + +if __name__ == '__main__': + model = InceptionResnetV1(pretrained='vggface2').eval() + print(model) + demo_imgs = torch.zeros(16, 3, 64, 64) + print('Output size 1:', model(demo_imgs).shape, model(demo_imgs)[0]) + + model_2 = InceptionResnetV1( + pretrained='vggface2', + auto_input_resize=False).eval() + demo_imgs = torch.zeros(16, 3, 140, 140) + print('Output size 2:', model(demo_imgs).shape, model_2(demo_imgs)[0]) diff --git a/mlflow/sf2f/models/layers.py b/mlflow/sf2f/models/layers.py new file mode 100644 index 0000000..001e3eb --- /dev/null +++ b/mlflow/sf2f/models/layers.py @@ -0,0 +1,257 @@ +import torch +import torch.nn as nn +import torch.nn.functional as F +## We use the batchnorm in pytorch 0.3.x to avoid unexpected errors +## Otherwise, some in-place modification will bring some errors +# from models.utils.batchnorm import BatchNorm1d, BatchNorm2d + + + +def get_normalization_2d(channels, normalization): + if normalization == 'instance': + return nn.InstanceNorm2d(channels) + elif normalization == 'batch': + return nn.BatchNorm2d(channels) + elif normalization == 'none': + return None + else: + raise ValueError('Unrecognized normalization type "%s"' % + normalization) + + +def get_activation(name): + kwargs = {} + if name.lower().startswith('leakyrelu'): + if '-' in name: + slope = float(name.split('-')[1]) + kwargs = {'negative_slope': slope} + name = 'leakyrelu' + activations = { + 'relu': nn.ReLU, + 'leakyrelu': nn.LeakyReLU, + 'none': None, + } + if name.lower() not in activations: + raise ValueError('Invalid activation "%s"' % name) + if activations[name.lower()] is not None: + return activations[name.lower()](**kwargs) + + +def weights_init(m): + raise NotImplementedError + classname = m.__class__.__name__ + if classname.find('Conv') != -1: + m.weight.data.normal_(0.0, 0.02) + m.bias.data.fill_(0) + elif classname.find('BatchNorm2d') != -1: + m.weight.data.normal_(1.0, 0.02) + m.bias.data.fill_(0) + elif classname.find('BatchNorm1d') != -1: + m.weight.data.normal_(1.0, 0.02) + m.bias.data.fill_(0) + elif classname.find('Linear') != -1: + m.weight.data.normal_(0.0, 0.02) + m.bias.data.fill_(0) + + +class Flatten(nn.Module): + def forward(self, x): + return x.view(x.size(0), -1) + + def __repr__(self): + return 'Flatten()' + + +class Unflatten(nn.Module): + def __init__(self, size): + super(Unflatten, self).__init__() + self.size = size + + def forward(self, x): + return x.view(*self.size) + + def __repr__(self): + size_str = ', '.join('%d' % d for d in self.size) + return 'Unflatten(%s)' % size_str + + +class GlobalAvgPool(nn.Module): + def __init__(self, pooling="avg"): + super(GlobalAvgPool, self).__init__() + self.pooling = pooling + def forward(self, x): + N, C = x.size(0), x.size(1) + if self.pooling == "avg": + return x.view(N, C, -1).mean(dim=2) + elif self.pooling == "max": + return x.view(N, C, -1).max(dim=2)[0] + else: + assert False, "Unrecognized pooling: {}".format(self.pooling) + + +class ResidualBlock(nn.Module): + def __init__(self, channels, normalization='batch', activation='relu', + padding='same', kernel_size=3, init='default'): + super(ResidualBlock, self).__init__() + K = kernel_size + P = _get_padding(K, "same") + C = channels + self.padding = P + layers = [ + get_normalization_2d(C, normalization), + get_activation(activation), + nn.Conv2d(C, C, kernel_size=K, padding=P), + get_normalization_2d(C, normalization), + get_activation(activation), + nn.Conv2d(C, C, kernel_size=K, padding=P), + ] + layers = [layer for layer in layers if layer is not None] + # for layer in layers: + # _init_conv(layer, method=init) + self.net = nn.Sequential(*layers) + + def forward(self, x): + P = self.padding + shortcut = x + y = self.net(x) + return shortcut + self.net(x) + + +def _get_padding(K, mode): + """ Helper method to compute padding size """ + if mode == 'valid': + return 0 + elif mode == 'same': + assert K % 2 == 1, 'Invalid kernel size %d for "same" padding' % K + return (K - 1) // 2 + + +def build_cnn(arch, normalization='batch', activation='relu', padding='same', + pooling='max', init='default', non_linear_activated=False): + """ + Build a CNN from an architecture string, which is a list of layer + specification strings. The overall architecture can be given as a list or as + a comma-separated string. + + All convolutions *except for the first* are preceeded by normalization and + nonlinearity. + + non_linear_activated: whether to add [non-linear activation / normalization] + Default is [False] (do not add these layers) + + All other layers support the following: + - IX: Indicates that the number of input channels to the network is X. + Can only be used at the first layer; if not present then we assume + 3 input channels. + - CK-X: KxK convolution with X output channels + - CK-X-S: KxK convolution with X output channels and stride S + - R: Residual block keeping the same number of channels + - UX: Nearest-neighbor upsampling with factor X + - PX: Spatial pooling with factor X + - FC-X-Y: Flatten followed by fully-connected layer + + Returns a tuple of: + - cnn: An nn.Sequential + - channels: Number of output channels + """ + if isinstance(arch, str): + arch = arch.split(',') + cur_C = 3 + if len(arch) > 0 and arch[0][0] == 'I': + cur_C = int(arch[0][1:]) + arch = arch[1:] + + first_conv = True + flat = False + layers = [] + for i, s in enumerate(arch): + if s[0] == 'C': + if not first_conv: + layers.append(get_normalization_2d(cur_C, normalization)) + layers.append(get_activation(activation)) + first_conv = False + vals = [int(i) for i in s[1:].split('-')] + P = None + if len(vals) == 2: + K, next_C = vals + stride = 1 + elif len(vals) == 3: + K, next_C, stride = vals + elif len(vals) == 4: + # Add padding + K, next_C, stride, P = vals + # K, next_C = (int(i) for i in s[1:].split('-')) + if P is None: + P = _get_padding(K, padding) + conv = nn.Conv2d(cur_C, next_C, kernel_size=K, + padding=P, stride=stride) + layers.append(conv) + # _init_conv(layers[-1], init) + cur_C = next_C + elif s[0] == 'R': + norm = 'none' if first_conv else normalization + res = ResidualBlock(cur_C, normalization=norm, activation=activation, + padding=padding, init=init) + layers.append(res) + first_conv = False + elif s[0] == 'U': + factor = int(s[1:]) + layers.append(nn.Upsample(scale_factor=factor, mode='nearest')) + elif s[0] == 'P': + factor = int(s[1:]) + if pooling == 'max': + pool = nn.MaxPool2d(kernel_size=factor, stride=factor) + elif pooling == 'avg': + pool = nn.AvgPool2d(kernel_size=factor, stride=factor) + layers.append(pool) + elif s[:2] == 'FC': + _, Din, Dout = s.split('-') + Din, Dout = int(Din), int(Dout) + if not flat: + layers.append(Flatten()) + flat = True + layers.append(nn.Linear(Din, Dout)) + if i + 1 < len(arch): + layers.append(get_activation(activation)) + cur_C = Dout + else: + raise ValueError('Invalid layer "%s"' % s) + if non_linear_activated: + layers.append(get_normalization_2d(cur_C, normalization)) + layers.append(get_activation(activation)) + layers = [layer for layer in layers if layer is not None] + #for layer in layers: + # print(layer) + + model = nn.Sequential(*layers) + return model, cur_C + + +def build_mlp(dim_list, activation='relu', batch_norm='none', + dropout=0, final_nonlinearity=True, + start_nonlinearity=False,): + layers = [] + for i in range(len(dim_list) - 1): + dim_in, dim_out = dim_list[i], dim_list[i + 1] + if start_nonlinearity and i == 0: + if batch_norm == 'batch': + layers.append(nn.BatchNorm1d(dim_in)) + # elif batch_norm == 'instance': + # layers.append(InstanceNorm1d(dim_in)) + if activation == 'relu': + layers.append(nn.ReLU()) + elif activation == 'leakyrelu': + layers.append(nn.LeakyReLU()) + layers.append(nn.Linear(dim_in, dim_out)) + final_layer = (i == len(dim_list) - 2) + if not final_layer or final_nonlinearity: + if batch_norm == 'batch': + layers.append(nn.BatchNorm1d(dim_out)) + if activation == 'relu': + layers.append(nn.ReLU()) + elif activation == 'leakyrelu': + layers.append(nn.LeakyReLU()) + if dropout > 0: + layers.append(nn.Dropout(p=dropout)) + model = nn.Sequential(*layers) + return model diff --git a/mlflow/sf2f/models/model_collection.py b/mlflow/sf2f/models/model_collection.py new file mode 100644 index 0000000..bf3e4ba --- /dev/null +++ b/mlflow/sf2f/models/model_collection.py @@ -0,0 +1,38 @@ +''' +A collection of all the models, for the convenience of internal calling +''' + + +# Lower level encoder decoders +try: + from .voice_encoders import V2F1DCNN, TransEncoder +except: + from voice_encoders import V2F1DCNN, TransEncoder + +try: + from .face_decoders import V2FDecoder, AttnV2FDecoder, MRDecoder, \ + FaceGanDecoder, FaceGanDecoder_v2 +except: + from face_decoders import V2FDecoder, AttnV2FDecoder, MRDecoder, \ + FaceGanDecoder, FaceGanDecoder_v2 + +try: + from .attention import Attention +except: + from .attention import Attention + + +class ModelCollection(): + def __init__(self): + self.V2F1DCNN = V2F1DCNN + self.V2FDecoder = V2FDecoder + self.AttnV2FDecoder = AttnV2FDecoder + self.TransEncoder = TransEncoder + self.MRDecoder = MRDecoder + self.FaceGanDecoder = FaceGanDecoder + self.FaceGanDecoder_v2 = FaceGanDecoder_v2 + # self.PganDecoder = PganDecoder + self.Attention = Attention + + +model_collection = ModelCollection() diff --git a/mlflow/sf2f/models/model_setup.py b/mlflow/sf2f/models/model_setup.py new file mode 100644 index 0000000..94ef21a --- /dev/null +++ b/mlflow/sf2f/models/model_setup.py @@ -0,0 +1,99 @@ +import torch +import models +from copy import deepcopy +from utils import update_values, load_model_state +import glog as log + +def build_model(opts, image_size, checkpoint_start_from=None): + if checkpoint_start_from is not None: + log.info("Load checkpoint as initialization: {}".format( + checkpoint_start_from)) + checkpoint = torch.load(checkpoint_start_from) + # kwarg aka keyword arguments + #kwargs = checkpoint['model_kwargs'] + kwargs = deepcopy(opts["options"]) + kwargs["image_size"] = image_size + model = getattr(models, opts["arch"])(**kwargs) + raw_state_dict = checkpoint['model_state'] + state_dict = {} + for k, v in raw_state_dict.items(): + if k.startswith('module.'): + k = k[7:] + state_dict[k] = v + #print("state_dict: ", state_dict.keys()) + model = load_model_state(model, state_dict, strict=False) + else: + kwargs = deepcopy(opts["options"]) + kwargs["image_size"] = image_size + model = getattr(models, opts["arch"])(**kwargs) + return model, kwargs + +def build_ac_discriminator(opts): + d_kwargs = deepcopy(opts["generic"]) + discriminator = [] + if "identity" in opts.keys(): + d_kwargs = update_values(opts["identity"], d_kwargs) + dis = models.AcDiscriminator(**d_kwargs) + discriminator.append(dis) + if "identity_low" in opts.keys(): + d_kwargs = update_values(opts["identity_low"], d_kwargs) + dis_low = models.AcDiscriminator(**d_kwargs) + discriminator.append(dis_low) + if "identity_mid" in opts.keys(): + d_kwargs = update_values(opts["identity_mid"], d_kwargs) + dis_mid = models.AcDiscriminator(**d_kwargs) + discriminator.append(dis_mid) + if "identity_high" in opts.keys(): + d_kwargs = update_values(opts["identity_high"], d_kwargs) + dis_high = models.AcDiscriminator(**d_kwargs) + discriminator.append(dis_high) + return discriminator, d_kwargs + +def build_img_discriminator(opts): + discriminator = [] + if opts.get("pgan_dis") is not None: + d_kwargs = deepcopy(opts["pgan_dis"]) + pgan_dis = models.PGAN_Discriminator(**d_kwargs) + discriminator.append(pgan_dis) + else: + d_kwargs = deepcopy(opts["generic"]) + if "image" in opts.keys(): + d_kwargs = update_values(opts["image"], d_kwargs) + dis = models.PatchDiscriminator(**d_kwargs) + discriminator.append(dis) + else: + if "image_low" in opts.keys(): + d_kwargs = update_values(opts["image_low"], d_kwargs) + discriminator_low = models.PatchDiscriminator(**d_kwargs) + discriminator.append(discriminator_low) + if "image_mid" in opts.keys(): + d_kwargs = update_values(opts["image_mid"], d_kwargs) + discriminator_mid = models.PatchDiscriminator(**d_kwargs) + discriminator.append(discriminator_mid) + if "image_high" in opts.keys(): + d_kwargs = update_values(opts["image_high"], d_kwargs) + discriminator_high = models.PatchDiscriminator(**d_kwargs) + discriminator.append(discriminator_high) + return discriminator, d_kwargs + +def build_cond_discriminator(opts): + discriminator = [] + d_kwargs = deepcopy(opts["generic"]) + if "condition" in opts.keys(): + d_kwargs = update_values(opts["condition"], d_kwargs) + dis = models.CondPatchDiscriminator(**d_kwargs) + discriminator.append(dis) + else: + if "cond_low" in opts.keys(): + d_kwargs = update_values(opts["cond_low"], d_kwargs) + discriminator_low = models.CondPatchDiscriminator(**d_kwargs) + discriminator.append(discriminator_low) + if "cond_mid" in opts.keys(): + d_kwargs = update_values(opts["cond_mid"], d_kwargs) + discriminator_mid = models.CondPatchDiscriminator(**d_kwargs) + discriminator.append(discriminator_mid) + if "cond_high" in opts.keys(): + d_kwargs = update_values(opts["cond_high"], d_kwargs) + discriminator_high = models.CondPatchDiscriminator(**d_kwargs) + discriminator.append(discriminator_high) + return discriminator, d_kwargs diff --git a/mlflow/sf2f/models/networks.py b/mlflow/sf2f/models/networks.py new file mode 100644 index 0000000..28f3333 --- /dev/null +++ b/mlflow/sf2f/models/networks.py @@ -0,0 +1,203 @@ +import torch +import torch.nn as nn +import functools +from torch.autograd import Variable +import numpy as np +import torch.nn.functional as F +import math +from numpy import prod + + + +class VGGLoss(nn.Module): + def __init__(self): + super(VGGLoss, self).__init__() + self.vgg = Vgg19().cuda() + self.criterion = nn.L1Loss() + self.weights = [1.0/32, 1.0/16, 1.0/8, 1.0/4, 1.0] + + def forward(self, x, y): + x_vgg = self.vgg(x) + # we do not store gradients for y + with torch.no_grad(): + y_vgg = self.vgg(y) + loss = 0 + for i in range(len(x_vgg)): + loss += self.weights[i] * self.criterion(x_vgg[i], y_vgg[i]) + return loss + +from torchvision import models +class Vgg19(torch.nn.Module): + def __init__(self, requires_grad=False, ignore_last=False): + super(Vgg19, self).__init__() + vgg_pretrained_features = models.vgg19(pretrained=True).features + self.slice1 = torch.nn.Sequential() + self.slice2 = torch.nn.Sequential() + self.slice3 = torch.nn.Sequential() + self.slice4 = torch.nn.Sequential() + if ignore_last: + self.slice5 = None + else: + self.slice5 = torch.nn.Sequential() + for x in range(2): + self.slice1.add_module(str(x), vgg_pretrained_features[x]) + for x in range(2, 7): + self.slice2.add_module(str(x), vgg_pretrained_features[x]) + for x in range(7, 12): + self.slice3.add_module(str(x), vgg_pretrained_features[x]) + for x in range(12, 21): + self.slice4.add_module(str(x), vgg_pretrained_features[x]) + if self.slice5 is not None: + for x in range(21, 30): + self.slice5.add_module(str(x), vgg_pretrained_features[x]) + if not requires_grad: + for param in self.parameters(): + param.requires_grad = False + + def forward(self, X): + h_relu1 = self.slice1(X) + h_relu2 = self.slice2(h_relu1) + h_relu3 = self.slice3(h_relu2) + h_relu4 = self.slice4(h_relu3) + if self.slice5 is None or h_relu4.size(3) < 3: + return [h_relu1, h_relu2, h_relu3, h_relu4,] + h_relu5 = self.slice5(h_relu4) + out = [h_relu1, h_relu2, h_relu3, h_relu4, h_relu5] + return out + + +class GLU(nn.Module): + def __init__(self): + super(GLU, self).__init__() + + def forward(self, x): + nc = x.size(1) + assert nc % 2 == 0, 'channels dont divide 2!' + nc = int(nc/2) + return x[:, :nc] * F.sigmoid(x[:, nc:]) + + +def upBlock(in_planes, out_planes): + block = nn.Sequential( + nn.Upsample(scale_factor=2, mode='nearest'), + nn.Conv2d(in_planes, out_planes * 2, kernel_size=3, + stride=1, padding=1, bias=False), + nn.BatchNorm2d(out_planes * 2), + GLU()) + return block + +# The defination of Progressive GAN + +class NormalizationLayer(nn.Module): + + def __init__(self): + super(NormalizationLayer, self).__init__() + + def forward(self, x, epsilon=1e-8): + return x * (((x**2).mean(dim=1, keepdim=True) + epsilon).rsqrt()) + + +def Upscale2d(x, factor=2): + assert isinstance(factor, int) and factor >= 1 + if factor == 1: + return x + s = x.size() + x = x.view(-1, s[1], s[2], 1, s[3], 1) + x = x.expand(-1, s[1], s[2], factor, s[3], factor) + x = x.contiguous().view(-1, s[1], s[2] * factor, s[3] * factor) + return x + + +def getLayerNormalizationFactor(x): + r""" + Get He's constant for the given layer + https://www.cv-foundation.org/openaccess/content_iccv_2015/papers/He_Delving_Deep_into_ICCV_2015_paper.pdf + """ + size = x.weight.size() + fan_in = prod(size[1:]) + + return math.sqrt(2.0 / fan_in) + + +class ConstrainedLayer(nn.Module): + r""" + A handy refactor that allows the user to: + - initialize one layer's bias to zero + - apply He's initialization at runtime + """ + + def __init__(self, + module, + equalized=True, + lrMul=1.0, + initBiasToZero=True): + r""" + equalized (bool): if true, the layer's weight should evolve within + the range (-1, 1) + initBiasToZero (bool): if true, bias will be initialized to zero + """ + + super(ConstrainedLayer, self).__init__() + + self.module = module + self.equalized = equalized + + if initBiasToZero: + self.module.bias.data.fill_(0) + if self.equalized: + self.module.weight.data.normal_(0, 1) + self.module.weight.data /= lrMul + self.weight = getLayerNormalizationFactor(self.module) * lrMul + + def forward(self, x): + + x = self.module(x) + if self.equalized: + x *= self.weight + return x + + +class EqualizedConv2d(ConstrainedLayer): + + def __init__(self, + nChannelsPrevious, + nChannels, + kernelSize, + padding=0, + bias=True, + **kwargs): + r""" + A nn.Conv2d module with specific constraints + Args: + nChannelsPrevious (int): number of channels in the previous layer + nChannels (int): number of channels of the current layer + kernelSize (int): size of the convolutional kernel + padding (int): convolution's padding + bias (bool): with bias ? + """ + + ConstrainedLayer.__init__(self, + nn.Conv2d(nChannelsPrevious, nChannels, + kernelSize, padding=padding, + bias=bias), + **kwargs) + + +class EqualizedLinear(ConstrainedLayer): + + def __init__(self, + nChannelsPrevious, + nChannels, + bias=True, + **kwargs): + r""" + A nn.Linear module with specific constraints + Args: + nChannelsPrevious (int): number of channels in the previous layer + nChannels (int): number of channels of the current layer + bias (bool): with bias ? + """ + + ConstrainedLayer.__init__(self, + nn.Linear(nChannelsPrevious, nChannels, + bias=bias), **kwargs) diff --git a/mlflow/sf2f/models/perceptual.py b/mlflow/sf2f/models/perceptual.py new file mode 100644 index 0000000..c22015f --- /dev/null +++ b/mlflow/sf2f/models/perceptual.py @@ -0,0 +1,111 @@ +import torch +import torch.nn as nn +import functools +from torch.autograd import Variable +import numpy as np +from torchvision import models +try: + from .inception_resnet_v1 import InceptionResnetV1 +except: + from inception_resnet_v1 import InceptionResnetV1 + + +class FaceNetLoss(nn.Module): + def __init__(self, cos_loss_weight=0.0): + super(FaceNetLoss, self).__init__() + self.facenet = InceptionResnetV1(pretrained='vggface2').eval().cuda() + self.criterion = nn.L1Loss() + self.weights = [1.0/16, 1.0/8, 1.0/4, 1.0] + self.cos_loss_weight = cos_loss_weight + if self.cos_loss_weight > 0.0: + self.cos_emb_loss = torch.nn.CosineEmbeddingLoss() + self.cos_target = torch.ones((1,), dtype=torch.int8).cuda() + + def forward(self, x, y): + x_facenet = self.facenet.get_hidden_states(x) + # we do not store gradients for y + with torch.no_grad(): + y_facenet = self.facenet.get_hidden_states(y) + loss = 0 + for i in range(len(x_facenet)): + #print(x_facenet[i].shape) + #print(y_facenet[i].shape) + loss += self.weights[i] * \ + self.criterion(x_facenet[i], y_facenet[i]) + if self.cos_loss_weight > 0.0: + loss += self.cos_loss_weight * \ + self.cos_emb_loss( + x_facenet[-1], + y_facenet[-1], + self.cos_target) + return loss + + +class VGGLoss(nn.Module): + def __init__(self): + super(VGGLoss, self).__init__() + self.vgg = Vgg19().cuda() + self.criterion = nn.L1Loss() + self.weights = [1.0/32, 1.0/16, 1.0/8, 1.0/4, 1.0] + + def forward(self, x, y): + x_vgg = self.vgg(x) + # we do not store gradients for y + with torch.no_grad(): + y_vgg = self.vgg(y) + loss = 0 + for i in range(len(x_vgg)): + loss += self.weights[i] * self.criterion(x_vgg[i], y_vgg[i]) + return loss + + +class Vgg19(torch.nn.Module): + def __init__(self, requires_grad=False, ignore_last=False): + super(Vgg19, self).__init__() + vgg_pretrained_features = models.vgg19(pretrained=True).features + self.slice1 = torch.nn.Sequential() + self.slice2 = torch.nn.Sequential() + self.slice3 = torch.nn.Sequential() + self.slice4 = torch.nn.Sequential() + if ignore_last: + self.slice5 = None + else: + self.slice5 = torch.nn.Sequential() + for x in range(2): + self.slice1.add_module(str(x), vgg_pretrained_features[x]) + for x in range(2, 7): + self.slice2.add_module(str(x), vgg_pretrained_features[x]) + for x in range(7, 12): + self.slice3.add_module(str(x), vgg_pretrained_features[x]) + for x in range(12, 21): + self.slice4.add_module(str(x), vgg_pretrained_features[x]) + if self.slice5 is not None: + for x in range(21, 30): + self.slice5.add_module(str(x), vgg_pretrained_features[x]) + if not requires_grad: + for param in self.parameters(): + param.requires_grad = False + + def forward(self, X): + h_relu1 = self.slice1(X) + h_relu2 = self.slice2(h_relu1) + h_relu3 = self.slice3(h_relu2) + h_relu4 = self.slice4(h_relu3) + if self.slice5 is None or h_relu4.size(3) < 3: + return [h_relu1, h_relu2, h_relu3, h_relu4,] + h_relu5 = self.slice5(h_relu4) + out = [h_relu1, h_relu2, h_relu3, h_relu4, h_relu5] + return out + + +if __name__ == '__main__': + demo_imgs = torch.ones((4, 3, 224, 224)) + facenet = InceptionResnetV1(pretrained='vggface2').eval() + facenet_out = facenet.get_hidden_states(demo_imgs) + for i, feature in enumerate(facenet_out): + print('FaceNet Feature Shape {}:'.format(i), feature.shape) + + vgg_19 = Vgg19() + vgg_out = vgg_19(demo_imgs) + for i, feature in enumerate(vgg_out): + print('VGG Feature Shape {}:'.format(i), feature.shape) diff --git a/mlflow/sf2f/models/voice_encoders.py b/mlflow/sf2f/models/voice_encoders.py new file mode 100644 index 0000000..f26acfb --- /dev/null +++ b/mlflow/sf2f/models/voice_encoders.py @@ -0,0 +1,474 @@ +''' +Implementations of different Voice Encoders +''' + + +import os +import torch +import torch.nn as nn +import torch.optim as optim +import torch.nn.functional as F +import numpy as np +import math + +try: + from fusers import AttentionFuserV1, AttentionFuserV2, AttentionFuserV3 +except: + from .fusers import AttentionFuserV1, AttentionFuserV2, AttentionFuserV3 + + +class EncoderModelCollection(): + def __init__(self): + self.AttentionFuserV1 = AttentionFuserV1 + self.AttentionFuserV2 = AttentionFuserV2 + self.AttentionFuserV3 = AttentionFuserV3 + +encoder_model_collection = EncoderModelCollection() + + +class TransposeLayer(nn.Module): + def __init__(self, dim_1, dim_2): + ''' + Transpose Layer + ''' + nn.Module.__init__(self) + self.dim_1 = dim_1 + self.dim_2 = dim_2 + + def forward(self, x): + ''' + seq_in: tensor with shape [N, D, L] + ''' + x = x.transpose(self.dim_1, self.dim_2) + return x + + +class CNN1DBlock(nn.Module): + def __init__(self, + channel_in, + channel_out, + kernel=3, + stride=2, + padding=1): + ''' + 1D CNN block + ''' + nn.Module.__init__(self) + self.layers = [] + self.layers.append( + nn.Conv1d(channel_in, + channel_out, + kernel, + stride, + padding, + bias=False)) + self.layers.append( + nn.BatchNorm1d(channel_out, affine=True)) + self.layers.append( + nn.ReLU(inplace=True)) + self.model = nn.Sequential(*self.layers) + + def forward(self, seq_in): + ''' + seq_in: tensor with shape [N, D, L] + ''' + seq_out = self.model(seq_in) + return seq_out + + +class Inception1DBlock(nn.Module): + def __init__(self, + channel_in, + channel_k2, + channel_k3, + channel_k5, + channel_k7): + ''' + Basic building block of 1D Inception Encoder + ''' + nn.Module.__init__(self) + self.conv_k2 = None + self.conv_k3 = None + self.conv_k5 = None + + if channel_k2 > 0: + self.conv_k2 = CNN1DBlock( + channel_in, + channel_k2, + 2, 2, 1) + if channel_k3 > 0: + self.conv_k3 = CNN1DBlock( + channel_in, + channel_k3, + 3, 2, 1) + if channel_k5 > 0: + self.conv_k5 = CNN1DBlock( + channel_in, + channel_k3, + 5, 2, 2) + if channel_k7 > 0: + self.conv_k7 = CNN1DBlock( + channel_in, + channel_k3, + 7, 2, 3) + + def forward(self, input): + output = [] + if self.conv_k2 is not None: + c2_out = self.conv_k2(input) + output.append(c2_out) + if self.conv_k3 is not None: + c3_out = self.conv_k3(input) + output.append(c3_out) + if self.conv_k5 is not None: + c5_out = self.conv_k5(input) + output.append(c5_out) + if self.conv_k7 is not None: + c7_out = self.conv_k7(input) + output.append(c7_out) + #print(c2_out.shape, c3_out.shape, c5_out.shape, c7_out.shape) + if output[0].shape[-1] > output[1].shape[-1]: + output[0] = output[0][:, :, 0:-1] + output = torch.cat(output, 1) + #print(output.shape) + return output + + +class PositionalEncoding(nn.Module): + + def __init__(self, d_model, dropout=0.1, max_len=10000): + super(PositionalEncoding, self).__init__() + self.dropout = nn.Dropout(p=dropout) + pe = torch.zeros(max_len, d_model) + position = torch.arange(0, max_len, dtype=torch.float).unsqueeze(1) + div_term = torch.exp(torch.arange(0, d_model, 2).float() * \ + (-math.log(10000.0) / d_model)) + pe[:, 0::2] = torch.sin(position * div_term) + pe[:, 1::2] = torch.cos(position * div_term) + pe = pe.unsqueeze(0).transpose(0, 1) + self.register_buffer('pe', pe) + + def forward(self, x): + x = x + self.pe[:x.size(0), :] + return self.dropout(x) + + +class PosEmbLayer(nn.Module): + def __init__(self, max_length=2000, pos_embedding_dim=5): + ''' + Postional embedding block + ''' + nn.Module.__init__(self) + + self.max_length = max_length + self.pos_embedding_dim = pos_embedding_dim + # Position Embedding + # Each position index in range (0, 2 * max_length - 1) have its own embedding + self.pos_embedding = nn.Embedding( + max_length, pos_embedding_dim, padding_idx=0) + + def forward(self, seq_in): + ''' + seq_in: tensor with shape [N, L, D] + ''' + N, L, D = seq_in.shape + # (L, ) + pos_idx = torch.tensor(np.arange(0, L)) + # (1, L) + pos_idx = pos_idx.unsqueeze(0) + # (N, L) + pos_idx = pos_idx.repeat(N, 1).cuda() + # (N, L, D_pos) + pos_emb = self.pos_embedding(pos_idx).type(seq_in.type()) + #pos_emb = torch.transpose(pos_emb, 1, 2) + seq_out = torch.cat([seq_in, pos_emb], 2) + # (N, L, D + D_pos) + return seq_out + + +class TransEncoder(nn.Module): + def __init__(self, + input_channel=40, + cnn_channels=[512, ], + transformer_dim=512, + transformer_depth=3, + return_seq=True, + sin_pos_encoding=False, + pos_embedding_dim=8, + add_noise=False, + normalize_embedding=False): + super(TransEncoder, self).__init__() + self.layers = [] + + if pos_embedding_dim > 0: + # Postional Embedding Layer + # (N, L, D) + self.layers.append(TransposeLayer(1, 2)) + # (N, D, L) + self.layers.append( + PosEmbLayer(2000, pos_embedding_dim)) + # (N, L, D) + self.layers.append(TransposeLayer(1, 2)) + + # CNN Layer for dimension conversion + self.layers.append( + CNN1DBlock(input_channel+pos_embedding_dim, cnn_channels[0])) + for cnn_layer_id in range(1, len(cnn_channels)): + self.layers.append(CNN1DBlock( + cnn_channels[cnn_layer_id - 1], + cnn_channels[cnn_layer_id])) + + self.layers.append(TransposeLayer(1, 2)) + # Sin Positional Encoding + if sin_pos_encoding: + self.layers.append(PositionalEncoding(transformer_dim)) + + # Transformer + trans_layer = nn.TransformerEncoderLayer( + transformer_dim, + 8, + transformer_dim) + transformer_encoder = nn.TransformerEncoder( + trans_layer, transformer_depth) + self.layers.append(transformer_encoder) + self.layers.append(TransposeLayer(1, 2)) + + self.model = nn.Sequential(*self.layers) + self.add_noise = add_noise + self.normalize_embedding = normalize_embedding + self.return_seq = return_seq + + def forward(self, seq_in): + seq_out = self.model(seq_in) + #for layer in self.layers: + # print(layer) + # x = layer(x) + # print(x.shape) + embeddings = F.avg_pool1d(seq_out, seq_out.size()[2], stride=1) + embeddings = embeddings.view(embeddings.size()[0], -1, 1, 1) + + if self.normalize_embedding: + embeddings = F.normalize(embeddings) + if self.add_noise: + noise = 0.05 * torch.randn( + embeddings.shape[0], embeddings.shape[1], 1, 1) + noise = noise.type(embeddings.type()) + embeddings = embeddings + noise + if self.normalize_embedding: + embeddings = F.normalize(embeddings) + + if self.return_seq: + return embeddings, seq_out + else: + return embeddings + + +class V2F1DCNN(nn.Module): + def __init__(self, + input_channel, + channels, + output_channel, + add_noise=False, + normalize_embedding=False, + return_seq=False, + inception_mode=False, + segments_fusion=False, + normalize_fusion=False, + fuser_arch='Attention', + fuser_kwargs=None): + super(V2F1DCNN, self).__init__() + if inception_mode: + self.model = nn.Sequential( + Inception1DBlock( + channel_in=input_channel, + channel_k2=channels[0]//4, + channel_k3=channels[0]//4, + channel_k5=channels[0]//4, + channel_k7=channels[0]//4), + Inception1DBlock( + channel_in=channels[0], + channel_k2=channels[1]//4, + channel_k3=channels[1]//4, + channel_k5=channels[1]//4, + channel_k7=channels[1]//4), + Inception1DBlock( + channel_in=channels[1], + channel_k2=channels[2]//4, + channel_k3=channels[2]//4, + channel_k5=channels[2]//4, + channel_k7=channels[2]//4), + Inception1DBlock( + channel_in=channels[2], + channel_k2=channels[3]//4, + channel_k3=channels[3]//4, + channel_k5=channels[3]//4, + channel_k7=channels[3]//4), + nn.Conv1d(channels[3], output_channel, 3, 2, 1, bias=True), + ) + else: + self.model = nn.Sequential( + nn.Conv1d(input_channel, channels[0], 3, 2, 1, bias=False), + nn.BatchNorm1d(channels[0], affine=True), + nn.ReLU(inplace=True), + nn.Conv1d(channels[0], channels[1], 3, 2, 1, bias=False), + nn.BatchNorm1d(channels[1], affine=True), + nn.ReLU(inplace=True), + nn.Conv1d(channels[1], channels[2], 3, 2, 1, bias=False), + nn.BatchNorm1d(channels[2], affine=True), + nn.ReLU(inplace=True), + nn.Conv1d(channels[2], channels[3], 3, 2, 1, bias=False), + nn.BatchNorm1d(channels[3], affine=True), + nn.ReLU(inplace=True), + nn.Conv1d(channels[3], output_channel, 3, 2, 1, bias=True), + ) + self.add_noise = add_noise + self.normalize_embedding = normalize_embedding + self.return_seq = return_seq + self.output_channel = output_channel + + self.segments_fusion = segments_fusion + self.normalize_fusion = normalize_fusion + if segments_fusion: + #self.attn_fuser = Attention( + # output_channel, + # output_channel, + # ignore_tanh=True) + self.attn_fuser = \ + getattr(encoder_model_collection, fuser_arch)(**fuser_kwargs) + + def forward(self, x): + # In case more than one mel segment per person is passed + if len(x.shape) == 4: + fusion_mode = True + B, N, C, L = x.shape + #print('Fusion Mode On! Input Shape:', x.shape) + x = x.view(B*N, C, L) + else: + fusion_mode = False + B, C, L = x.shape + x = self.model(x) + embeddings = F.avg_pool1d(x, x.size()[2], stride=1) + embeddings = embeddings.view(embeddings.size()[0], -1, 1, 1) + + if self.normalize_embedding: + embeddings = F.normalize(embeddings) + if self.add_noise: + noise = 0.05 * torch.randn(x.shape[0], x.shape[1], 1, 1) + noise = noise.type(embeddings.type()) + embeddings = embeddings + noise + if self.normalize_embedding: + embeddings = F.normalize(embeddings) + + # Restore Tensor shape + if fusion_mode: + #print(embeddings.shape) + C_emb = embeddings.shape[1] + embeddings = embeddings.view(B, N, C_emb) + # Attention fusion + embeddings = self.attn_fuser(embeddings) + if self.normalize_fusion: + embeddings = F.normalize(embeddings) + + if self.return_seq: + return embeddings, x + else: + return embeddings + + def print_param(self): + print('All parameters:') + for name, param in self.named_parameters(): + print(name) + + def print_trainable_param(model): + print('Trainable Parameters:') + for name, param in model.named_parameters(): + if param.requires_grad: + print(name) + + def train_fuser_only(self): + print('Training Attention Fuser Only') + for name, param in self.named_parameters(): + if 'attn_fuser' in name: + param.requires_grad = True + else: + param.requires_grad = False + + def init_attn_fusion(self): + self.attn_fuser = Attention( + self.output_channel, + self.output_channel, + ignore_tanh=True) + +if __name__ == '__main__': + import time + # Demo Input + log_mels = torch.ones((16, 40, 151)) + pos_log_mel = torch.ones((16, 48, 150)).transpose(1, 2) + + # Test Inception1DBlock + incept_block = Inception1DBlock( + channel_in=40, + channel_k2=64, + channel_k3=64, + channel_k5=64, + channel_k7=64) + incept_block(log_mels) + + # Test V2F 1D CNN + log_mel_segs = torch.ones((16, 20, 40, 150)) + v2f_id_cnn_kwargs = { + 'input_channel': 40, + 'channels': [256, 384, 576, 864], + 'output_channel': 64, + 'segments_fusion': True, + 'inception_mode': True, + } + v2f_id_cnn_fuse = V2F1DCNN(**v2f_id_cnn_kwargs) + print(v2f_id_cnn_fuse) + print('V2F1DCNN Output shape:', v2f_id_cnn_fuse(log_mel_segs).shape) + v2f_id_cnn_fuse.print_param() + v2f_id_cnn_fuse.train_fuser_only() + v2f_id_cnn_fuse.print_trainable_param() + + # test a simple transformer layer + trans_layer = nn.TransformerEncoderLayer(48, 8, + dim_feedforward=48, dropout=0.1, activation='relu') + print(trans_layer(pos_log_mel).shape) + + + # Transformer + trans_kwargs = { + 'input_channel': 40, + 'cnn_channels': [512, 512], + 'transformer_dim': 512, + 'transformer_depth': 2, + 'return_seq': True, + 'pos_embedding_dim': 0, #8 + 'sin_pos_encoding': True, #False + } + trans_encoder = TransEncoder(**trans_kwargs).cuda() + print(trans_encoder) + emb, seq_emb = trans_encoder(log_mels.cuda()) + print('TransEncoder Output shape:', emb.shape, seq_emb.shape) + + # Test V2F 1D CNN + v2f_id_cnn_kwargs = { + 'input_channel': 40, + 'channels': [256, 384, 576, 864], + 'output_channel': 64, + } + v2f_id_cnn = V2F1DCNN(**v2f_id_cnn_kwargs) + print(v2f_id_cnn) + print('V2F1DCNN Output shape:', v2f_id_cnn(log_mels).shape) + + # Test V2F 1D CNN Sequential Return + v2f_id_cnn_kwargs = { + 'input_channel': 40, + 'channels': [256, 384, 576, 864], + 'output_channel': 64, + 'return_seq': True, + } + v2f_id_cnn = V2F1DCNN(**v2f_id_cnn_kwargs) + print(v2f_id_cnn) + emb, seq_emb = v2f_id_cnn(log_mels) + print('V2F1DCNN Output shape:', emb.shape, seq_emb.shape) diff --git a/mlflow/sf2f/options/__init__.py b/mlflow/sf2f/options/__init__.py new file mode 100644 index 0000000..e69de29 diff --git a/mlflow/sf2f/options/data_opts/vox.yaml b/mlflow/sf2f/options/data_opts/vox.yaml new file mode 100644 index 0000000..d0d3207 --- /dev/null +++ b/mlflow/sf2f/options/data_opts/vox.yaml @@ -0,0 +1,7 @@ +root_dir: /home/data/VoxCeleb +face_type: masked +image_normalize_method: imagenet #standard +mel_normalize_method: vox_mel +nframe_range: (100, 150) +split_set: train +split_json: /home/data/VoxCeleb/split.json diff --git a/mlflow/sf2f/options/opts.py b/mlflow/sf2f/options/opts.py new file mode 100644 index 0000000..988372f --- /dev/null +++ b/mlflow/sf2f/options/opts.py @@ -0,0 +1,92 @@ +import os +import os.path as osp +import argparse +import yaml +from utils import update_values +from utils import int_tuple, float_tuple, str_tuple, bool_flag + +parser = argparse.ArgumentParser() + +# Optimization hyperparameters +parser.add_argument('--batch_size', default=256, type=int) +parser.add_argument('--epochs', default=200, type=int) +parser.add_argument('--learning_rate', default=5e-4, type=float) +# by default, it is disabled +parser.add_argument('--decay_lr_epochs', default=10000, type=float) +parser.add_argument('--beta1', type=float, default=0.9, + help='momentum term of adam') +parser.add_argument('--eval_epochs', default=1, type=int) +parser.add_argument('--eval_mode_after', default=10000, type=int) +parser.add_argument('--disable_l1_loss_after', default=10000000, type=int) +parser.add_argument('--path_opts', type=str, + default='options/vox/sf2f/sf2f_1st_stage.yaml', help="Options.") + +parser.add_argument('--workers', default=12, type=int) + +# Output options +parser.add_argument('--print_every', default=100, type=int) +parser.add_argument('--visualize_every', type=int, + default=1, help="visualize to visdom.") +parser.add_argument('--timing', action="store_true") +parser.add_argument('--output_dir', type=str) +parser.add_argument('--log_suffix', type=str) +parser.add_argument('--checkpoint_name', default='checkpoint') +parser.add_argument('--checkpoint_start_from', default=None) +parser.add_argument('--checkpoint', nargs='+') +parser.add_argument('--resume', type=str) +parser.add_argument('--evaluate', action='store_true', + help="Set to evaluate the model.") +parser.add_argument('--evaluate_train', action='store_true', + help="Set to evaluate the training set.") + +# For inference in run_model.py +parser.add_argument('--output_demo_dir', default='output/results') +parser.add_argument('--samples_path', default='samples') + +# For evaluation +parser.add_argument('--recall_method', default='cos_sim') +parser.add_argument('--facenet_return_pooling', default=False) +# parser.add_argument('--face_gen_mode', default='average_facenet_embedding') +parser.add_argument('--face_gen_mode', nargs='+') + +# For test.py +parser.add_argument('--get_faces_from_different_segments', default=False) + +# Attention Fuser Mode +parser.add_argument('--train_fuser_only', default=False) +# Train the fuser and decoder +parser.add_argument('--train_fuser_decoder', default=False) +parser.add_argument('--pretrained_path', default=None) +parser.add_argument('--freeze_discriminators', default=False) + +# For Inference +parser.add_argument('--input_wav_dir', type=str, default='data/example_audio') +parser.add_argument('--fuser_infer', default=False) +parser.add_argument('--seed', type=int, default=311) +args = parser.parse_args() + +options = { + "data": { + "batch_size": args.batch_size, + "workers": args.workers, + "data_opts": {}, + }, + "optim": { + "lr": args.learning_rate, + "epochs": args.epochs, + "eval_epochs": args.eval_epochs, + }, + "logs": { + "output_dir": args.output_dir, + }, +} + +with open(args.path_opts, "r") as f: + options_yaml = yaml.full_load(f) +with open(options_yaml["data"]["data_opts_path"], "r") as f: + data_opts = yaml.full_load(f) + options_yaml["data"]["data_opts"] = data_opts + +options = update_values(options, options_yaml) +if args.log_suffix: + options["logs"]["name"] += "-" + args.log_suffix diff --git a/mlflow/sf2f/options/vox/baseline/v2f.yaml b/mlflow/sf2f/options/vox/baseline/v2f.yaml new file mode 100644 index 0000000..cd406bf --- /dev/null +++ b/mlflow/sf2f/options/vox/baseline/v2f.yaml @@ -0,0 +1,50 @@ +# The implementation of NeurIPS 19 V2F baseline model +# Log-related settings +logs: + name: v2f + output_dir: output/ +# data-related settings +data: + dataset: vox + data_opts_path: options/data_opts/vox.yaml + image_size: [64, 64] +# model related settings +generator: + arch: EncoderDecoder + options: + encoder_arch: V2F1DCNN + encoder_kwargs: + input_channel: 40 + channels: [256, 384, 576, 864] + output_channel: 512 + add_noise: True + normalize_embedding: True + decoder_arch: V2FDecoder + decoder_kwargs: + input_channel: 512 + channels: [1024, 512, 256, 128, 64] + output_channel: 3 +discriminator: + generic: + normalization: none + # padding mode is set to unused because we are specifying padding value + padding: unused + activation: leakyrelu-0.2 + # image discrimintor is implemented in AC Discriminator + identity: + # 'C{kernel_size}-{channel_out}-{stride}-{padding}' + arch: 'C1-32-1-0,C4-64-2-1,C4-128-2-1,C4-256-2-1,C4-512-1-0' + num_id: 0 # will be updated in train.py +optim: + # Discriminator Loss Weights + d_loss_weight: 1.0 + d_img_weight: -1 + ac_loss_weight: 1.0 + # Generator Loss Weights + gan_loss_type: 'gan' + l1_pixel_loss_weight: 100.0 + # Perceptual Loss +eval: + facenet: + deprocess_and_preprocess: True + crop_faces: True diff --git a/mlflow/sf2f/options/vox/sf2f/sf2f_1st_stage.yaml b/mlflow/sf2f/options/vox/sf2f/sf2f_1st_stage.yaml new file mode 100644 index 0000000..c627b1d --- /dev/null +++ b/mlflow/sf2f/options/vox/sf2f/sf2f_1st_stage.yaml @@ -0,0 +1,52 @@ +# Optimal final version of encoder-decoder and Loss Function +# Encoder: Inceptional 1D CNN +# Decoder: Upsampling + CNN +logs: + name: sf2f_1st_stage + output_dir: output/ +# data-related settings +data: + dataset: vox + data_opts_path: options/data_opts/vox.yaml + image_size: [64, 64] +# model related settings +generator: + arch: EncoderDecoder + options: + encoder_arch: V2F1DCNN + encoder_kwargs: + input_channel: 40 + channels: [256, 384, 576, 864] + output_channel: 512 + normalize_embedding: True + inception_mode: True + decoder_arch: FaceGanDecoder + decoder_kwargs: + noise_dim: 512 + mlp_normalization: none + normalization: batch + activation: leakyrelu-0.1 +discriminator: + generic: + normalization: batch + padding: valid + activation: leakyrelu-0.1 + image: + arch: 'C4-64-2,C4-128-2,C4-256-2' + identity: + arch: 'C4-64-2,C4-128-2,C4-256-2' + num_id: 0 # will be updated in train.py +optim: + # Discriminator Loss Weights + d_loss_weight: 1.0 + d_img_weight: 1.0 #0.5 + ac_loss_weight: 0.05 + # Generator Loss Weights + gan_loss_type: 'gan' + l1_pixel_loss_weight: 10.0 + # Perceptual Loss + perceptual_loss_weight: 100.0 +eval: + facenet: + deprocess_and_preprocess: True + crop_faces: True diff --git a/mlflow/sf2f/options/vox/sf2f/sf2f_fuser.yaml b/mlflow/sf2f/options/vox/sf2f/sf2f_fuser.yaml new file mode 100644 index 0000000..a414da5 --- /dev/null +++ b/mlflow/sf2f/options/vox/sf2f/sf2f_fuser.yaml @@ -0,0 +1,62 @@ +# Optimal final version of encoder-decoder and Loss Function +# Encoder: Inceptional 1D CNN +# Decoder: Upsampling + CNN +# Fuser: Vanilla 1-layer attention 'general' +logs: + name: sf2f_fuser + output_dir: sf2f/output/ +# data-related settings +data: + dataset: vox + data_opts_path: /home/hojun/Documents/project/boostcamp/final_project/mlops/voice2face-modeling/sf2f/options/data_opts/vox.yaml + image_size: [64, 64] +# model related settings +generator: + arch: EncoderDecoder + options: + encoder_arch: V2F1DCNN + encoder_kwargs: + input_channel: 40 + channels: [256, 384, 576, 864] + output_channel: 512 + normalize_embedding: True + inception_mode: True + segments_fusion: True + normalize_fusion: True + # Initialize Attention Fuser as a submodule of encoder + fuser_arch: AttentionFuserV1 + fuser_kwargs: + dimensions: 512 + dim_out: 512 + attention_type: general + ignore_tanh: True + decoder_arch: FaceGanDecoder + decoder_kwargs: + noise_dim: 512 + mlp_normalization: none + normalization: batch + activation: leakyrelu-0.1 +discriminator: + generic: + normalization: batch + padding: valid + activation: leakyrelu-0.1 + image: + arch: 'C4-64-2,C4-128-2,C4-256-2' + identity: + arch: 'C4-64-2,C4-128-2,C4-256-2' + num_id: 0 # will be updated in train.py +optim: + # Discriminator Loss Weights + d_loss_weight: 1.0 + d_img_weight: 1.0 #0.5 + ac_loss_weight: 0.05 + # Generator Loss Weights + gan_loss_type: 'gan' + l1_pixel_loss_weight: 10.0 + # Perceptual Loss + perceptual_loss_weight: 100.0 +eval: + facenet: + deprocess_and_preprocess: True + crop_faces: True diff --git a/mlflow/sf2f/options/vox/sf2f/sf2f_mid_1st_stage.yaml b/mlflow/sf2f/options/vox/sf2f/sf2f_mid_1st_stage.yaml new file mode 100644 index 0000000..6f0e407 --- /dev/null +++ b/mlflow/sf2f/options/vox/sf2f/sf2f_mid_1st_stage.yaml @@ -0,0 +1,58 @@ +# Optimal final version of encoder-decoder and Loss Function +# Encoder: Inceptional 1D CNN +# Decoder: Upsampling + CNN +# Mid resolution: with multi-reso training +logs: + name: sf2f_mid_1st_stage + output_dir: output/ +# data-related settings +data: + dataset: vox + data_opts_path: sf2f/options/data_opts/vox.yaml + image_size: [128, 128] +# model related settings +generator: + arch: EncoderDecoder + options: + encoder_arch: V2F1DCNN + encoder_kwargs: + input_channel: 40 + channels: [256, 384, 576, 864] + output_channel: 512 + normalize_embedding: True + inception_mode: True + decoder_arch: FaceGanDecoder_v2 + decoder_kwargs: + noise_dim: 512 + mlp_normalization: none + normalization: batch + activation: leakyrelu-0.1 +discriminator: + generic: + normalization: batch + padding: valid + activation: leakyrelu-0.1 + image_low: + arch: 'C4-64-2,C4-128-2,C4-256-2' + image_mid: + arch: 'C4-64-2,C4-128-2,C4-256-2,C4-512-2' + identity_low: + arch: 'C4-64-2,C4-128-2,C4-256-2' + num_id: 0 # will be updated in train.py + identity_mid: + arch: 'C4-64-2,C4-128-2,C4-256-2,C4-512-2' + num_id: 0 # will be updated in train.py +optim: + # Discriminator Loss Weights + d_loss_weight: 1.0 + d_img_weight: 1.0 #0.5 + ac_loss_weight: 0.05 + # Generator Loss Weights + gan_loss_type: 'gan' + l1_pixel_loss_weight: 100.0 + # Perceptual Loss + perceptual_loss_weight: 100.0 +eval: + facenet: + deprocess_and_preprocess: True + crop_faces: True diff --git a/mlflow/sf2f/options/vox/sf2f/sf2f_mid_fuser.yaml b/mlflow/sf2f/options/vox/sf2f/sf2f_mid_fuser.yaml new file mode 100644 index 0000000..ed021dc --- /dev/null +++ b/mlflow/sf2f/options/vox/sf2f/sf2f_mid_fuser.yaml @@ -0,0 +1,67 @@ +# Optimal final version of encoder-decoder and Loss Function +# Encoder: Inceptional 1D CNN +# Decoder: Upsampling + CNN +# Mid resolution: with multi-reso training +logs: + name: sf2f_mid_fuser + output_dir: output/ +# data-related settings +data: + dataset: vox + data_opts_path: sf2f/options/data_opts/vox.yaml + image_size: [128, 128] +# model related settings +generator: + arch: EncoderDecoder + options: + encoder_arch: V2F1DCNN + encoder_kwargs: + input_channel: 40 + channels: [256, 384, 576, 864] + output_channel: 512 + normalize_embedding: True + inception_mode: True + segments_fusion: True + normalize_fusion: True + # Initialize Attention Fuser as a submodule of encoder + fuser_arch: AttentionFuserV1 + fuser_kwargs: + dimensions: 512 + dim_out: 512 + attention_type: general + ignore_tanh: True + decoder_arch: FaceGanDecoder_v2 + decoder_kwargs: + noise_dim: 512 + mlp_normalization: none + normalization: batch + activation: leakyrelu-0.1 +discriminator: + generic: + normalization: batch + padding: valid + activation: leakyrelu-0.1 + image_low: + arch: 'C4-64-2,C4-128-2,C4-256-2' + image_mid: + arch: 'C4-64-2,C4-128-2,C4-256-2,C4-512-2' + identity_low: + arch: 'C4-64-2,C4-128-2,C4-256-2' + num_id: 0 # will be updated in train.py + identity_mid: + arch: 'C4-64-2,C4-128-2,C4-256-2,C4-512-2' + num_id: 0 # will be updated in train.py +optim: + # Discriminator Loss Weights + d_loss_weight: 1.0 + d_img_weight: 1.0 #0.5 + ac_loss_weight: 0.05 + # Generator Loss Weights + gan_loss_type: 'gan' + l1_pixel_loss_weight: 100.0 + # Perceptual Loss + perceptual_loss_weight: 100.0 +eval: + facenet: + deprocess_and_preprocess: True + crop_faces: True diff --git a/mlflow/sf2f/output/sf2f_1st_stage-20240303-215640/sf2f_1st_stage.yaml b/mlflow/sf2f/output/sf2f_1st_stage-20240303-215640/sf2f_1st_stage.yaml new file mode 100644 index 0000000..c627b1d --- /dev/null +++ b/mlflow/sf2f/output/sf2f_1st_stage-20240303-215640/sf2f_1st_stage.yaml @@ -0,0 +1,52 @@ +# Optimal final version of encoder-decoder and Loss Function +# Encoder: Inceptional 1D CNN +# Decoder: Upsampling + CNN +logs: + name: sf2f_1st_stage + output_dir: output/ +# data-related settings +data: + dataset: vox + data_opts_path: options/data_opts/vox.yaml + image_size: [64, 64] +# model related settings +generator: + arch: EncoderDecoder + options: + encoder_arch: V2F1DCNN + encoder_kwargs: + input_channel: 40 + channels: [256, 384, 576, 864] + output_channel: 512 + normalize_embedding: True + inception_mode: True + decoder_arch: FaceGanDecoder + decoder_kwargs: + noise_dim: 512 + mlp_normalization: none + normalization: batch + activation: leakyrelu-0.1 +discriminator: + generic: + normalization: batch + padding: valid + activation: leakyrelu-0.1 + image: + arch: 'C4-64-2,C4-128-2,C4-256-2' + identity: + arch: 'C4-64-2,C4-128-2,C4-256-2' + num_id: 0 # will be updated in train.py +optim: + # Discriminator Loss Weights + d_loss_weight: 1.0 + d_img_weight: 1.0 #0.5 + ac_loss_weight: 0.05 + # Generator Loss Weights + gan_loss_type: 'gan' + l1_pixel_loss_weight: 10.0 + # Perceptual Loss + perceptual_loss_weight: 100.0 +eval: + facenet: + deprocess_and_preprocess: True + crop_faces: True diff --git a/mlflow/sf2f/output/sf2f_1st_stage-20240303-220349/sf2f_1st_stage.yaml b/mlflow/sf2f/output/sf2f_1st_stage-20240303-220349/sf2f_1st_stage.yaml new file mode 100644 index 0000000..c627b1d --- /dev/null +++ b/mlflow/sf2f/output/sf2f_1st_stage-20240303-220349/sf2f_1st_stage.yaml @@ -0,0 +1,52 @@ +# Optimal final version of encoder-decoder and Loss Function +# Encoder: Inceptional 1D CNN +# Decoder: Upsampling + CNN +logs: + name: sf2f_1st_stage + output_dir: output/ +# data-related settings +data: + dataset: vox + data_opts_path: options/data_opts/vox.yaml + image_size: [64, 64] +# model related settings +generator: + arch: EncoderDecoder + options: + encoder_arch: V2F1DCNN + encoder_kwargs: + input_channel: 40 + channels: [256, 384, 576, 864] + output_channel: 512 + normalize_embedding: True + inception_mode: True + decoder_arch: FaceGanDecoder + decoder_kwargs: + noise_dim: 512 + mlp_normalization: none + normalization: batch + activation: leakyrelu-0.1 +discriminator: + generic: + normalization: batch + padding: valid + activation: leakyrelu-0.1 + image: + arch: 'C4-64-2,C4-128-2,C4-256-2' + identity: + arch: 'C4-64-2,C4-128-2,C4-256-2' + num_id: 0 # will be updated in train.py +optim: + # Discriminator Loss Weights + d_loss_weight: 1.0 + d_img_weight: 1.0 #0.5 + ac_loss_weight: 0.05 + # Generator Loss Weights + gan_loss_type: 'gan' + l1_pixel_loss_weight: 10.0 + # Perceptual Loss + perceptual_loss_weight: 100.0 +eval: + facenet: + deprocess_and_preprocess: True + crop_faces: True diff --git a/mlflow/sf2f/output/sf2f_1st_stage-20240303-221106/sf2f_1st_stage.yaml b/mlflow/sf2f/output/sf2f_1st_stage-20240303-221106/sf2f_1st_stage.yaml new file mode 100644 index 0000000..c627b1d --- /dev/null +++ b/mlflow/sf2f/output/sf2f_1st_stage-20240303-221106/sf2f_1st_stage.yaml @@ -0,0 +1,52 @@ +# Optimal final version of encoder-decoder and Loss Function +# Encoder: Inceptional 1D CNN +# Decoder: Upsampling + CNN +logs: + name: sf2f_1st_stage + output_dir: output/ +# data-related settings +data: + dataset: vox + data_opts_path: options/data_opts/vox.yaml + image_size: [64, 64] +# model related settings +generator: + arch: EncoderDecoder + options: + encoder_arch: V2F1DCNN + encoder_kwargs: + input_channel: 40 + channels: [256, 384, 576, 864] + output_channel: 512 + normalize_embedding: True + inception_mode: True + decoder_arch: FaceGanDecoder + decoder_kwargs: + noise_dim: 512 + mlp_normalization: none + normalization: batch + activation: leakyrelu-0.1 +discriminator: + generic: + normalization: batch + padding: valid + activation: leakyrelu-0.1 + image: + arch: 'C4-64-2,C4-128-2,C4-256-2' + identity: + arch: 'C4-64-2,C4-128-2,C4-256-2' + num_id: 0 # will be updated in train.py +optim: + # Discriminator Loss Weights + d_loss_weight: 1.0 + d_img_weight: 1.0 #0.5 + ac_loss_weight: 0.05 + # Generator Loss Weights + gan_loss_type: 'gan' + l1_pixel_loss_weight: 10.0 + # Perceptual Loss + perceptual_loss_weight: 100.0 +eval: + facenet: + deprocess_and_preprocess: True + crop_faces: True diff --git a/mlflow/sf2f/output/sf2f_1st_stage-20240303-225403/sf2f_1st_stage.yaml b/mlflow/sf2f/output/sf2f_1st_stage-20240303-225403/sf2f_1st_stage.yaml new file mode 100644 index 0000000..c627b1d --- /dev/null +++ b/mlflow/sf2f/output/sf2f_1st_stage-20240303-225403/sf2f_1st_stage.yaml @@ -0,0 +1,52 @@ +# Optimal final version of encoder-decoder and Loss Function +# Encoder: Inceptional 1D CNN +# Decoder: Upsampling + CNN +logs: + name: sf2f_1st_stage + output_dir: output/ +# data-related settings +data: + dataset: vox + data_opts_path: options/data_opts/vox.yaml + image_size: [64, 64] +# model related settings +generator: + arch: EncoderDecoder + options: + encoder_arch: V2F1DCNN + encoder_kwargs: + input_channel: 40 + channels: [256, 384, 576, 864] + output_channel: 512 + normalize_embedding: True + inception_mode: True + decoder_arch: FaceGanDecoder + decoder_kwargs: + noise_dim: 512 + mlp_normalization: none + normalization: batch + activation: leakyrelu-0.1 +discriminator: + generic: + normalization: batch + padding: valid + activation: leakyrelu-0.1 + image: + arch: 'C4-64-2,C4-128-2,C4-256-2' + identity: + arch: 'C4-64-2,C4-128-2,C4-256-2' + num_id: 0 # will be updated in train.py +optim: + # Discriminator Loss Weights + d_loss_weight: 1.0 + d_img_weight: 1.0 #0.5 + ac_loss_weight: 0.05 + # Generator Loss Weights + gan_loss_type: 'gan' + l1_pixel_loss_weight: 10.0 + # Perceptual Loss + perceptual_loss_weight: 100.0 +eval: + facenet: + deprocess_and_preprocess: True + crop_faces: True diff --git a/mlflow/sf2f/output/sf2f_1st_stage-20240303-225533/sf2f_1st_stage.yaml b/mlflow/sf2f/output/sf2f_1st_stage-20240303-225533/sf2f_1st_stage.yaml new file mode 100644 index 0000000..c627b1d --- /dev/null +++ b/mlflow/sf2f/output/sf2f_1st_stage-20240303-225533/sf2f_1st_stage.yaml @@ -0,0 +1,52 @@ +# Optimal final version of encoder-decoder and Loss Function +# Encoder: Inceptional 1D CNN +# Decoder: Upsampling + CNN +logs: + name: sf2f_1st_stage + output_dir: output/ +# data-related settings +data: + dataset: vox + data_opts_path: options/data_opts/vox.yaml + image_size: [64, 64] +# model related settings +generator: + arch: EncoderDecoder + options: + encoder_arch: V2F1DCNN + encoder_kwargs: + input_channel: 40 + channels: [256, 384, 576, 864] + output_channel: 512 + normalize_embedding: True + inception_mode: True + decoder_arch: FaceGanDecoder + decoder_kwargs: + noise_dim: 512 + mlp_normalization: none + normalization: batch + activation: leakyrelu-0.1 +discriminator: + generic: + normalization: batch + padding: valid + activation: leakyrelu-0.1 + image: + arch: 'C4-64-2,C4-128-2,C4-256-2' + identity: + arch: 'C4-64-2,C4-128-2,C4-256-2' + num_id: 0 # will be updated in train.py +optim: + # Discriminator Loss Weights + d_loss_weight: 1.0 + d_img_weight: 1.0 #0.5 + ac_loss_weight: 0.05 + # Generator Loss Weights + gan_loss_type: 'gan' + l1_pixel_loss_weight: 10.0 + # Perceptual Loss + perceptual_loss_weight: 100.0 +eval: + facenet: + deprocess_and_preprocess: True + crop_faces: True diff --git a/mlflow/sf2f/output/sf2f_1st_stage-20240303-225656/sf2f_1st_stage.yaml b/mlflow/sf2f/output/sf2f_1st_stage-20240303-225656/sf2f_1st_stage.yaml new file mode 100644 index 0000000..c627b1d --- /dev/null +++ b/mlflow/sf2f/output/sf2f_1st_stage-20240303-225656/sf2f_1st_stage.yaml @@ -0,0 +1,52 @@ +# Optimal final version of encoder-decoder and Loss Function +# Encoder: Inceptional 1D CNN +# Decoder: Upsampling + CNN +logs: + name: sf2f_1st_stage + output_dir: output/ +# data-related settings +data: + dataset: vox + data_opts_path: options/data_opts/vox.yaml + image_size: [64, 64] +# model related settings +generator: + arch: EncoderDecoder + options: + encoder_arch: V2F1DCNN + encoder_kwargs: + input_channel: 40 + channels: [256, 384, 576, 864] + output_channel: 512 + normalize_embedding: True + inception_mode: True + decoder_arch: FaceGanDecoder + decoder_kwargs: + noise_dim: 512 + mlp_normalization: none + normalization: batch + activation: leakyrelu-0.1 +discriminator: + generic: + normalization: batch + padding: valid + activation: leakyrelu-0.1 + image: + arch: 'C4-64-2,C4-128-2,C4-256-2' + identity: + arch: 'C4-64-2,C4-128-2,C4-256-2' + num_id: 0 # will be updated in train.py +optim: + # Discriminator Loss Weights + d_loss_weight: 1.0 + d_img_weight: 1.0 #0.5 + ac_loss_weight: 0.05 + # Generator Loss Weights + gan_loss_type: 'gan' + l1_pixel_loss_weight: 10.0 + # Perceptual Loss + perceptual_loss_weight: 100.0 +eval: + facenet: + deprocess_and_preprocess: True + crop_faces: True diff --git a/mlflow/sf2f/output/sf2f_1st_stage-20240303-225950/sf2f_1st_stage.yaml b/mlflow/sf2f/output/sf2f_1st_stage-20240303-225950/sf2f_1st_stage.yaml new file mode 100644 index 0000000..c627b1d --- /dev/null +++ b/mlflow/sf2f/output/sf2f_1st_stage-20240303-225950/sf2f_1st_stage.yaml @@ -0,0 +1,52 @@ +# Optimal final version of encoder-decoder and Loss Function +# Encoder: Inceptional 1D CNN +# Decoder: Upsampling + CNN +logs: + name: sf2f_1st_stage + output_dir: output/ +# data-related settings +data: + dataset: vox + data_opts_path: options/data_opts/vox.yaml + image_size: [64, 64] +# model related settings +generator: + arch: EncoderDecoder + options: + encoder_arch: V2F1DCNN + encoder_kwargs: + input_channel: 40 + channels: [256, 384, 576, 864] + output_channel: 512 + normalize_embedding: True + inception_mode: True + decoder_arch: FaceGanDecoder + decoder_kwargs: + noise_dim: 512 + mlp_normalization: none + normalization: batch + activation: leakyrelu-0.1 +discriminator: + generic: + normalization: batch + padding: valid + activation: leakyrelu-0.1 + image: + arch: 'C4-64-2,C4-128-2,C4-256-2' + identity: + arch: 'C4-64-2,C4-128-2,C4-256-2' + num_id: 0 # will be updated in train.py +optim: + # Discriminator Loss Weights + d_loss_weight: 1.0 + d_img_weight: 1.0 #0.5 + ac_loss_weight: 0.05 + # Generator Loss Weights + gan_loss_type: 'gan' + l1_pixel_loss_weight: 10.0 + # Perceptual Loss + perceptual_loss_weight: 100.0 +eval: + facenet: + deprocess_and_preprocess: True + crop_faces: True diff --git a/mlflow/sf2f/output/sf2f_1st_stage-20240303-230032/sf2f_1st_stage.yaml b/mlflow/sf2f/output/sf2f_1st_stage-20240303-230032/sf2f_1st_stage.yaml new file mode 100644 index 0000000..c627b1d --- /dev/null +++ b/mlflow/sf2f/output/sf2f_1st_stage-20240303-230032/sf2f_1st_stage.yaml @@ -0,0 +1,52 @@ +# Optimal final version of encoder-decoder and Loss Function +# Encoder: Inceptional 1D CNN +# Decoder: Upsampling + CNN +logs: + name: sf2f_1st_stage + output_dir: output/ +# data-related settings +data: + dataset: vox + data_opts_path: options/data_opts/vox.yaml + image_size: [64, 64] +# model related settings +generator: + arch: EncoderDecoder + options: + encoder_arch: V2F1DCNN + encoder_kwargs: + input_channel: 40 + channels: [256, 384, 576, 864] + output_channel: 512 + normalize_embedding: True + inception_mode: True + decoder_arch: FaceGanDecoder + decoder_kwargs: + noise_dim: 512 + mlp_normalization: none + normalization: batch + activation: leakyrelu-0.1 +discriminator: + generic: + normalization: batch + padding: valid + activation: leakyrelu-0.1 + image: + arch: 'C4-64-2,C4-128-2,C4-256-2' + identity: + arch: 'C4-64-2,C4-128-2,C4-256-2' + num_id: 0 # will be updated in train.py +optim: + # Discriminator Loss Weights + d_loss_weight: 1.0 + d_img_weight: 1.0 #0.5 + ac_loss_weight: 0.05 + # Generator Loss Weights + gan_loss_type: 'gan' + l1_pixel_loss_weight: 10.0 + # Perceptual Loss + perceptual_loss_weight: 100.0 +eval: + facenet: + deprocess_and_preprocess: True + crop_faces: True diff --git a/mlflow/sf2f/output/sf2f_1st_stage-20240303-230142/sf2f_1st_stage.yaml b/mlflow/sf2f/output/sf2f_1st_stage-20240303-230142/sf2f_1st_stage.yaml new file mode 100644 index 0000000..c627b1d --- /dev/null +++ b/mlflow/sf2f/output/sf2f_1st_stage-20240303-230142/sf2f_1st_stage.yaml @@ -0,0 +1,52 @@ +# Optimal final version of encoder-decoder and Loss Function +# Encoder: Inceptional 1D CNN +# Decoder: Upsampling + CNN +logs: + name: sf2f_1st_stage + output_dir: output/ +# data-related settings +data: + dataset: vox + data_opts_path: options/data_opts/vox.yaml + image_size: [64, 64] +# model related settings +generator: + arch: EncoderDecoder + options: + encoder_arch: V2F1DCNN + encoder_kwargs: + input_channel: 40 + channels: [256, 384, 576, 864] + output_channel: 512 + normalize_embedding: True + inception_mode: True + decoder_arch: FaceGanDecoder + decoder_kwargs: + noise_dim: 512 + mlp_normalization: none + normalization: batch + activation: leakyrelu-0.1 +discriminator: + generic: + normalization: batch + padding: valid + activation: leakyrelu-0.1 + image: + arch: 'C4-64-2,C4-128-2,C4-256-2' + identity: + arch: 'C4-64-2,C4-128-2,C4-256-2' + num_id: 0 # will be updated in train.py +optim: + # Discriminator Loss Weights + d_loss_weight: 1.0 + d_img_weight: 1.0 #0.5 + ac_loss_weight: 0.05 + # Generator Loss Weights + gan_loss_type: 'gan' + l1_pixel_loss_weight: 10.0 + # Perceptual Loss + perceptual_loss_weight: 100.0 +eval: + facenet: + deprocess_and_preprocess: True + crop_faces: True diff --git a/mlflow/sf2f/output/sf2f_1st_stage-20240303-230232/sf2f_1st_stage.yaml b/mlflow/sf2f/output/sf2f_1st_stage-20240303-230232/sf2f_1st_stage.yaml new file mode 100644 index 0000000..c627b1d --- /dev/null +++ b/mlflow/sf2f/output/sf2f_1st_stage-20240303-230232/sf2f_1st_stage.yaml @@ -0,0 +1,52 @@ +# Optimal final version of encoder-decoder and Loss Function +# Encoder: Inceptional 1D CNN +# Decoder: Upsampling + CNN +logs: + name: sf2f_1st_stage + output_dir: output/ +# data-related settings +data: + dataset: vox + data_opts_path: options/data_opts/vox.yaml + image_size: [64, 64] +# model related settings +generator: + arch: EncoderDecoder + options: + encoder_arch: V2F1DCNN + encoder_kwargs: + input_channel: 40 + channels: [256, 384, 576, 864] + output_channel: 512 + normalize_embedding: True + inception_mode: True + decoder_arch: FaceGanDecoder + decoder_kwargs: + noise_dim: 512 + mlp_normalization: none + normalization: batch + activation: leakyrelu-0.1 +discriminator: + generic: + normalization: batch + padding: valid + activation: leakyrelu-0.1 + image: + arch: 'C4-64-2,C4-128-2,C4-256-2' + identity: + arch: 'C4-64-2,C4-128-2,C4-256-2' + num_id: 0 # will be updated in train.py +optim: + # Discriminator Loss Weights + d_loss_weight: 1.0 + d_img_weight: 1.0 #0.5 + ac_loss_weight: 0.05 + # Generator Loss Weights + gan_loss_type: 'gan' + l1_pixel_loss_weight: 10.0 + # Perceptual Loss + perceptual_loss_weight: 100.0 +eval: + facenet: + deprocess_and_preprocess: True + crop_faces: True diff --git a/mlflow/sf2f/output/sf2f_1st_stage-20240303-230355/sf2f_1st_stage.yaml b/mlflow/sf2f/output/sf2f_1st_stage-20240303-230355/sf2f_1st_stage.yaml new file mode 100644 index 0000000..c627b1d --- /dev/null +++ b/mlflow/sf2f/output/sf2f_1st_stage-20240303-230355/sf2f_1st_stage.yaml @@ -0,0 +1,52 @@ +# Optimal final version of encoder-decoder and Loss Function +# Encoder: Inceptional 1D CNN +# Decoder: Upsampling + CNN +logs: + name: sf2f_1st_stage + output_dir: output/ +# data-related settings +data: + dataset: vox + data_opts_path: options/data_opts/vox.yaml + image_size: [64, 64] +# model related settings +generator: + arch: EncoderDecoder + options: + encoder_arch: V2F1DCNN + encoder_kwargs: + input_channel: 40 + channels: [256, 384, 576, 864] + output_channel: 512 + normalize_embedding: True + inception_mode: True + decoder_arch: FaceGanDecoder + decoder_kwargs: + noise_dim: 512 + mlp_normalization: none + normalization: batch + activation: leakyrelu-0.1 +discriminator: + generic: + normalization: batch + padding: valid + activation: leakyrelu-0.1 + image: + arch: 'C4-64-2,C4-128-2,C4-256-2' + identity: + arch: 'C4-64-2,C4-128-2,C4-256-2' + num_id: 0 # will be updated in train.py +optim: + # Discriminator Loss Weights + d_loss_weight: 1.0 + d_img_weight: 1.0 #0.5 + ac_loss_weight: 0.05 + # Generator Loss Weights + gan_loss_type: 'gan' + l1_pixel_loss_weight: 10.0 + # Perceptual Loss + perceptual_loss_weight: 100.0 +eval: + facenet: + deprocess_and_preprocess: True + crop_faces: True diff --git a/mlflow/sf2f/output/sf2f_1st_stage-20240303-230514/sf2f_1st_stage.yaml b/mlflow/sf2f/output/sf2f_1st_stage-20240303-230514/sf2f_1st_stage.yaml new file mode 100644 index 0000000..c627b1d --- /dev/null +++ b/mlflow/sf2f/output/sf2f_1st_stage-20240303-230514/sf2f_1st_stage.yaml @@ -0,0 +1,52 @@ +# Optimal final version of encoder-decoder and Loss Function +# Encoder: Inceptional 1D CNN +# Decoder: Upsampling + CNN +logs: + name: sf2f_1st_stage + output_dir: output/ +# data-related settings +data: + dataset: vox + data_opts_path: options/data_opts/vox.yaml + image_size: [64, 64] +# model related settings +generator: + arch: EncoderDecoder + options: + encoder_arch: V2F1DCNN + encoder_kwargs: + input_channel: 40 + channels: [256, 384, 576, 864] + output_channel: 512 + normalize_embedding: True + inception_mode: True + decoder_arch: FaceGanDecoder + decoder_kwargs: + noise_dim: 512 + mlp_normalization: none + normalization: batch + activation: leakyrelu-0.1 +discriminator: + generic: + normalization: batch + padding: valid + activation: leakyrelu-0.1 + image: + arch: 'C4-64-2,C4-128-2,C4-256-2' + identity: + arch: 'C4-64-2,C4-128-2,C4-256-2' + num_id: 0 # will be updated in train.py +optim: + # Discriminator Loss Weights + d_loss_weight: 1.0 + d_img_weight: 1.0 #0.5 + ac_loss_weight: 0.05 + # Generator Loss Weights + gan_loss_type: 'gan' + l1_pixel_loss_weight: 10.0 + # Perceptual Loss + perceptual_loss_weight: 100.0 +eval: + facenet: + deprocess_and_preprocess: True + crop_faces: True diff --git a/mlflow/sf2f/output/sf2f_1st_stage-20240303-230631/sf2f_1st_stage.yaml b/mlflow/sf2f/output/sf2f_1st_stage-20240303-230631/sf2f_1st_stage.yaml new file mode 100644 index 0000000..c627b1d --- /dev/null +++ b/mlflow/sf2f/output/sf2f_1st_stage-20240303-230631/sf2f_1st_stage.yaml @@ -0,0 +1,52 @@ +# Optimal final version of encoder-decoder and Loss Function +# Encoder: Inceptional 1D CNN +# Decoder: Upsampling + CNN +logs: + name: sf2f_1st_stage + output_dir: output/ +# data-related settings +data: + dataset: vox + data_opts_path: options/data_opts/vox.yaml + image_size: [64, 64] +# model related settings +generator: + arch: EncoderDecoder + options: + encoder_arch: V2F1DCNN + encoder_kwargs: + input_channel: 40 + channels: [256, 384, 576, 864] + output_channel: 512 + normalize_embedding: True + inception_mode: True + decoder_arch: FaceGanDecoder + decoder_kwargs: + noise_dim: 512 + mlp_normalization: none + normalization: batch + activation: leakyrelu-0.1 +discriminator: + generic: + normalization: batch + padding: valid + activation: leakyrelu-0.1 + image: + arch: 'C4-64-2,C4-128-2,C4-256-2' + identity: + arch: 'C4-64-2,C4-128-2,C4-256-2' + num_id: 0 # will be updated in train.py +optim: + # Discriminator Loss Weights + d_loss_weight: 1.0 + d_img_weight: 1.0 #0.5 + ac_loss_weight: 0.05 + # Generator Loss Weights + gan_loss_type: 'gan' + l1_pixel_loss_weight: 10.0 + # Perceptual Loss + perceptual_loss_weight: 100.0 +eval: + facenet: + deprocess_and_preprocess: True + crop_faces: True diff --git a/mlflow/sf2f/output/sf2f_1st_stage-20240303-231044/sf2f_1st_stage.yaml b/mlflow/sf2f/output/sf2f_1st_stage-20240303-231044/sf2f_1st_stage.yaml new file mode 100644 index 0000000..c627b1d --- /dev/null +++ b/mlflow/sf2f/output/sf2f_1st_stage-20240303-231044/sf2f_1st_stage.yaml @@ -0,0 +1,52 @@ +# Optimal final version of encoder-decoder and Loss Function +# Encoder: Inceptional 1D CNN +# Decoder: Upsampling + CNN +logs: + name: sf2f_1st_stage + output_dir: output/ +# data-related settings +data: + dataset: vox + data_opts_path: options/data_opts/vox.yaml + image_size: [64, 64] +# model related settings +generator: + arch: EncoderDecoder + options: + encoder_arch: V2F1DCNN + encoder_kwargs: + input_channel: 40 + channels: [256, 384, 576, 864] + output_channel: 512 + normalize_embedding: True + inception_mode: True + decoder_arch: FaceGanDecoder + decoder_kwargs: + noise_dim: 512 + mlp_normalization: none + normalization: batch + activation: leakyrelu-0.1 +discriminator: + generic: + normalization: batch + padding: valid + activation: leakyrelu-0.1 + image: + arch: 'C4-64-2,C4-128-2,C4-256-2' + identity: + arch: 'C4-64-2,C4-128-2,C4-256-2' + num_id: 0 # will be updated in train.py +optim: + # Discriminator Loss Weights + d_loss_weight: 1.0 + d_img_weight: 1.0 #0.5 + ac_loss_weight: 0.05 + # Generator Loss Weights + gan_loss_type: 'gan' + l1_pixel_loss_weight: 10.0 + # Perceptual Loss + perceptual_loss_weight: 100.0 +eval: + facenet: + deprocess_and_preprocess: True + crop_faces: True diff --git a/mlflow/sf2f/output/sf2f_1st_stage-20240303-231201/sf2f_1st_stage.yaml b/mlflow/sf2f/output/sf2f_1st_stage-20240303-231201/sf2f_1st_stage.yaml new file mode 100644 index 0000000..c627b1d --- /dev/null +++ b/mlflow/sf2f/output/sf2f_1st_stage-20240303-231201/sf2f_1st_stage.yaml @@ -0,0 +1,52 @@ +# Optimal final version of encoder-decoder and Loss Function +# Encoder: Inceptional 1D CNN +# Decoder: Upsampling + CNN +logs: + name: sf2f_1st_stage + output_dir: output/ +# data-related settings +data: + dataset: vox + data_opts_path: options/data_opts/vox.yaml + image_size: [64, 64] +# model related settings +generator: + arch: EncoderDecoder + options: + encoder_arch: V2F1DCNN + encoder_kwargs: + input_channel: 40 + channels: [256, 384, 576, 864] + output_channel: 512 + normalize_embedding: True + inception_mode: True + decoder_arch: FaceGanDecoder + decoder_kwargs: + noise_dim: 512 + mlp_normalization: none + normalization: batch + activation: leakyrelu-0.1 +discriminator: + generic: + normalization: batch + padding: valid + activation: leakyrelu-0.1 + image: + arch: 'C4-64-2,C4-128-2,C4-256-2' + identity: + arch: 'C4-64-2,C4-128-2,C4-256-2' + num_id: 0 # will be updated in train.py +optim: + # Discriminator Loss Weights + d_loss_weight: 1.0 + d_img_weight: 1.0 #0.5 + ac_loss_weight: 0.05 + # Generator Loss Weights + gan_loss_type: 'gan' + l1_pixel_loss_weight: 10.0 + # Perceptual Loss + perceptual_loss_weight: 100.0 +eval: + facenet: + deprocess_and_preprocess: True + crop_faces: True diff --git a/mlflow/sf2f/output/sf2f_1st_stage-20240303-231308/sf2f_1st_stage.yaml b/mlflow/sf2f/output/sf2f_1st_stage-20240303-231308/sf2f_1st_stage.yaml new file mode 100644 index 0000000..c627b1d --- /dev/null +++ b/mlflow/sf2f/output/sf2f_1st_stage-20240303-231308/sf2f_1st_stage.yaml @@ -0,0 +1,52 @@ +# Optimal final version of encoder-decoder and Loss Function +# Encoder: Inceptional 1D CNN +# Decoder: Upsampling + CNN +logs: + name: sf2f_1st_stage + output_dir: output/ +# data-related settings +data: + dataset: vox + data_opts_path: options/data_opts/vox.yaml + image_size: [64, 64] +# model related settings +generator: + arch: EncoderDecoder + options: + encoder_arch: V2F1DCNN + encoder_kwargs: + input_channel: 40 + channels: [256, 384, 576, 864] + output_channel: 512 + normalize_embedding: True + inception_mode: True + decoder_arch: FaceGanDecoder + decoder_kwargs: + noise_dim: 512 + mlp_normalization: none + normalization: batch + activation: leakyrelu-0.1 +discriminator: + generic: + normalization: batch + padding: valid + activation: leakyrelu-0.1 + image: + arch: 'C4-64-2,C4-128-2,C4-256-2' + identity: + arch: 'C4-64-2,C4-128-2,C4-256-2' + num_id: 0 # will be updated in train.py +optim: + # Discriminator Loss Weights + d_loss_weight: 1.0 + d_img_weight: 1.0 #0.5 + ac_loss_weight: 0.05 + # Generator Loss Weights + gan_loss_type: 'gan' + l1_pixel_loss_weight: 10.0 + # Perceptual Loss + perceptual_loss_weight: 100.0 +eval: + facenet: + deprocess_and_preprocess: True + crop_faces: True diff --git a/mlflow/sf2f/output/sf2f_1st_stage-20240303-231542/sf2f_1st_stage.yaml b/mlflow/sf2f/output/sf2f_1st_stage-20240303-231542/sf2f_1st_stage.yaml new file mode 100644 index 0000000..c627b1d --- /dev/null +++ b/mlflow/sf2f/output/sf2f_1st_stage-20240303-231542/sf2f_1st_stage.yaml @@ -0,0 +1,52 @@ +# Optimal final version of encoder-decoder and Loss Function +# Encoder: Inceptional 1D CNN +# Decoder: Upsampling + CNN +logs: + name: sf2f_1st_stage + output_dir: output/ +# data-related settings +data: + dataset: vox + data_opts_path: options/data_opts/vox.yaml + image_size: [64, 64] +# model related settings +generator: + arch: EncoderDecoder + options: + encoder_arch: V2F1DCNN + encoder_kwargs: + input_channel: 40 + channels: [256, 384, 576, 864] + output_channel: 512 + normalize_embedding: True + inception_mode: True + decoder_arch: FaceGanDecoder + decoder_kwargs: + noise_dim: 512 + mlp_normalization: none + normalization: batch + activation: leakyrelu-0.1 +discriminator: + generic: + normalization: batch + padding: valid + activation: leakyrelu-0.1 + image: + arch: 'C4-64-2,C4-128-2,C4-256-2' + identity: + arch: 'C4-64-2,C4-128-2,C4-256-2' + num_id: 0 # will be updated in train.py +optim: + # Discriminator Loss Weights + d_loss_weight: 1.0 + d_img_weight: 1.0 #0.5 + ac_loss_weight: 0.05 + # Generator Loss Weights + gan_loss_type: 'gan' + l1_pixel_loss_weight: 10.0 + # Perceptual Loss + perceptual_loss_weight: 100.0 +eval: + facenet: + deprocess_and_preprocess: True + crop_faces: True diff --git a/mlflow/sf2f/output/sf2f_1st_stage-20240303-231831/sf2f_1st_stage.yaml b/mlflow/sf2f/output/sf2f_1st_stage-20240303-231831/sf2f_1st_stage.yaml new file mode 100644 index 0000000..c627b1d --- /dev/null +++ b/mlflow/sf2f/output/sf2f_1st_stage-20240303-231831/sf2f_1st_stage.yaml @@ -0,0 +1,52 @@ +# Optimal final version of encoder-decoder and Loss Function +# Encoder: Inceptional 1D CNN +# Decoder: Upsampling + CNN +logs: + name: sf2f_1st_stage + output_dir: output/ +# data-related settings +data: + dataset: vox + data_opts_path: options/data_opts/vox.yaml + image_size: [64, 64] +# model related settings +generator: + arch: EncoderDecoder + options: + encoder_arch: V2F1DCNN + encoder_kwargs: + input_channel: 40 + channels: [256, 384, 576, 864] + output_channel: 512 + normalize_embedding: True + inception_mode: True + decoder_arch: FaceGanDecoder + decoder_kwargs: + noise_dim: 512 + mlp_normalization: none + normalization: batch + activation: leakyrelu-0.1 +discriminator: + generic: + normalization: batch + padding: valid + activation: leakyrelu-0.1 + image: + arch: 'C4-64-2,C4-128-2,C4-256-2' + identity: + arch: 'C4-64-2,C4-128-2,C4-256-2' + num_id: 0 # will be updated in train.py +optim: + # Discriminator Loss Weights + d_loss_weight: 1.0 + d_img_weight: 1.0 #0.5 + ac_loss_weight: 0.05 + # Generator Loss Weights + gan_loss_type: 'gan' + l1_pixel_loss_weight: 10.0 + # Perceptual Loss + perceptual_loss_weight: 100.0 +eval: + facenet: + deprocess_and_preprocess: True + crop_faces: True diff --git a/mlflow/sf2f/output/sf2f_1st_stage-20240303-232325/sf2f_1st_stage.yaml b/mlflow/sf2f/output/sf2f_1st_stage-20240303-232325/sf2f_1st_stage.yaml new file mode 100644 index 0000000..c627b1d --- /dev/null +++ b/mlflow/sf2f/output/sf2f_1st_stage-20240303-232325/sf2f_1st_stage.yaml @@ -0,0 +1,52 @@ +# Optimal final version of encoder-decoder and Loss Function +# Encoder: Inceptional 1D CNN +# Decoder: Upsampling + CNN +logs: + name: sf2f_1st_stage + output_dir: output/ +# data-related settings +data: + dataset: vox + data_opts_path: options/data_opts/vox.yaml + image_size: [64, 64] +# model related settings +generator: + arch: EncoderDecoder + options: + encoder_arch: V2F1DCNN + encoder_kwargs: + input_channel: 40 + channels: [256, 384, 576, 864] + output_channel: 512 + normalize_embedding: True + inception_mode: True + decoder_arch: FaceGanDecoder + decoder_kwargs: + noise_dim: 512 + mlp_normalization: none + normalization: batch + activation: leakyrelu-0.1 +discriminator: + generic: + normalization: batch + padding: valid + activation: leakyrelu-0.1 + image: + arch: 'C4-64-2,C4-128-2,C4-256-2' + identity: + arch: 'C4-64-2,C4-128-2,C4-256-2' + num_id: 0 # will be updated in train.py +optim: + # Discriminator Loss Weights + d_loss_weight: 1.0 + d_img_weight: 1.0 #0.5 + ac_loss_weight: 0.05 + # Generator Loss Weights + gan_loss_type: 'gan' + l1_pixel_loss_weight: 10.0 + # Perceptual Loss + perceptual_loss_weight: 100.0 +eval: + facenet: + deprocess_and_preprocess: True + crop_faces: True diff --git a/mlflow/sf2f/output/sf2f_1st_stage-20240303-233127/sf2f_1st_stage.yaml b/mlflow/sf2f/output/sf2f_1st_stage-20240303-233127/sf2f_1st_stage.yaml new file mode 100644 index 0000000..c627b1d --- /dev/null +++ b/mlflow/sf2f/output/sf2f_1st_stage-20240303-233127/sf2f_1st_stage.yaml @@ -0,0 +1,52 @@ +# Optimal final version of encoder-decoder and Loss Function +# Encoder: Inceptional 1D CNN +# Decoder: Upsampling + CNN +logs: + name: sf2f_1st_stage + output_dir: output/ +# data-related settings +data: + dataset: vox + data_opts_path: options/data_opts/vox.yaml + image_size: [64, 64] +# model related settings +generator: + arch: EncoderDecoder + options: + encoder_arch: V2F1DCNN + encoder_kwargs: + input_channel: 40 + channels: [256, 384, 576, 864] + output_channel: 512 + normalize_embedding: True + inception_mode: True + decoder_arch: FaceGanDecoder + decoder_kwargs: + noise_dim: 512 + mlp_normalization: none + normalization: batch + activation: leakyrelu-0.1 +discriminator: + generic: + normalization: batch + padding: valid + activation: leakyrelu-0.1 + image: + arch: 'C4-64-2,C4-128-2,C4-256-2' + identity: + arch: 'C4-64-2,C4-128-2,C4-256-2' + num_id: 0 # will be updated in train.py +optim: + # Discriminator Loss Weights + d_loss_weight: 1.0 + d_img_weight: 1.0 #0.5 + ac_loss_weight: 0.05 + # Generator Loss Weights + gan_loss_type: 'gan' + l1_pixel_loss_weight: 10.0 + # Perceptual Loss + perceptual_loss_weight: 100.0 +eval: + facenet: + deprocess_and_preprocess: True + crop_faces: True diff --git a/mlflow/sf2f/output/sf2f_1st_stage-20240303-233311/sf2f_1st_stage.yaml b/mlflow/sf2f/output/sf2f_1st_stage-20240303-233311/sf2f_1st_stage.yaml new file mode 100644 index 0000000..c627b1d --- /dev/null +++ b/mlflow/sf2f/output/sf2f_1st_stage-20240303-233311/sf2f_1st_stage.yaml @@ -0,0 +1,52 @@ +# Optimal final version of encoder-decoder and Loss Function +# Encoder: Inceptional 1D CNN +# Decoder: Upsampling + CNN +logs: + name: sf2f_1st_stage + output_dir: output/ +# data-related settings +data: + dataset: vox + data_opts_path: options/data_opts/vox.yaml + image_size: [64, 64] +# model related settings +generator: + arch: EncoderDecoder + options: + encoder_arch: V2F1DCNN + encoder_kwargs: + input_channel: 40 + channels: [256, 384, 576, 864] + output_channel: 512 + normalize_embedding: True + inception_mode: True + decoder_arch: FaceGanDecoder + decoder_kwargs: + noise_dim: 512 + mlp_normalization: none + normalization: batch + activation: leakyrelu-0.1 +discriminator: + generic: + normalization: batch + padding: valid + activation: leakyrelu-0.1 + image: + arch: 'C4-64-2,C4-128-2,C4-256-2' + identity: + arch: 'C4-64-2,C4-128-2,C4-256-2' + num_id: 0 # will be updated in train.py +optim: + # Discriminator Loss Weights + d_loss_weight: 1.0 + d_img_weight: 1.0 #0.5 + ac_loss_weight: 0.05 + # Generator Loss Weights + gan_loss_type: 'gan' + l1_pixel_loss_weight: 10.0 + # Perceptual Loss + perceptual_loss_weight: 100.0 +eval: + facenet: + deprocess_and_preprocess: True + crop_faces: True diff --git a/mlflow/sf2f/output/sf2f_1st_stage-20240303-233619/sf2f_1st_stage.yaml b/mlflow/sf2f/output/sf2f_1st_stage-20240303-233619/sf2f_1st_stage.yaml new file mode 100644 index 0000000..c627b1d --- /dev/null +++ b/mlflow/sf2f/output/sf2f_1st_stage-20240303-233619/sf2f_1st_stage.yaml @@ -0,0 +1,52 @@ +# Optimal final version of encoder-decoder and Loss Function +# Encoder: Inceptional 1D CNN +# Decoder: Upsampling + CNN +logs: + name: sf2f_1st_stage + output_dir: output/ +# data-related settings +data: + dataset: vox + data_opts_path: options/data_opts/vox.yaml + image_size: [64, 64] +# model related settings +generator: + arch: EncoderDecoder + options: + encoder_arch: V2F1DCNN + encoder_kwargs: + input_channel: 40 + channels: [256, 384, 576, 864] + output_channel: 512 + normalize_embedding: True + inception_mode: True + decoder_arch: FaceGanDecoder + decoder_kwargs: + noise_dim: 512 + mlp_normalization: none + normalization: batch + activation: leakyrelu-0.1 +discriminator: + generic: + normalization: batch + padding: valid + activation: leakyrelu-0.1 + image: + arch: 'C4-64-2,C4-128-2,C4-256-2' + identity: + arch: 'C4-64-2,C4-128-2,C4-256-2' + num_id: 0 # will be updated in train.py +optim: + # Discriminator Loss Weights + d_loss_weight: 1.0 + d_img_weight: 1.0 #0.5 + ac_loss_weight: 0.05 + # Generator Loss Weights + gan_loss_type: 'gan' + l1_pixel_loss_weight: 10.0 + # Perceptual Loss + perceptual_loss_weight: 100.0 +eval: + facenet: + deprocess_and_preprocess: True + crop_faces: True diff --git a/mlflow/sf2f/scripts/DScore/__init__.py b/mlflow/sf2f/scripts/DScore/__init__.py new file mode 100644 index 0000000..e69de29 diff --git a/mlflow/sf2f/scripts/DScore/models/__init__.py b/mlflow/sf2f/scripts/DScore/models/__init__.py new file mode 100644 index 0000000..e69de29 diff --git a/mlflow/sf2f/scripts/DScore/models/base_model.py b/mlflow/sf2f/scripts/DScore/models/base_model.py new file mode 100644 index 0000000..36f4d83 --- /dev/null +++ b/mlflow/sf2f/scripts/DScore/models/base_model.py @@ -0,0 +1,60 @@ +import os +import torch +import scripts.DScore.util.util as util +from torch.autograd import Variable +from pdb import set_trace as st +from IPython import embed + +class BaseModel(): + def __init__(self): + pass; + + def name(self): + return 'BaseModel' + + def initialize(self, use_gpu=True): + self.use_gpu = use_gpu + self.Tensor = torch.cuda.FloatTensor if self.use_gpu else torch.Tensor + # self.save_dir = os.path.join(opt.checkpoints_dir, opt.name) + + def forward(self): + pass + + def get_image_paths(self): + pass + + def optimize_parameters(self): + pass + + def get_current_visuals(self): + return self.input + + def get_current_errors(self): + return {} + + def save(self, label): + pass + + # helper saving function that can be used by subclasses + def save_network(self, network, path, network_label, epoch_label): + save_filename = '%s_net_%s.pth' % (epoch_label, network_label) + save_path = os.path.join(path, save_filename) + torch.save(network.state_dict(), save_path) + + # helper loading function that can be used by subclasses + def load_network(self, network, network_label, epoch_label): + # embed() + save_filename = '%s_net_%s.pth' % (epoch_label, network_label) + save_path = os.path.join(self.save_dir, save_filename) + print('Loading network from %s'%save_path) + network.load_state_dict(torch.load(save_path)) + + def update_learning_rate(): + pass + + def get_image_paths(self): + return self.image_paths + + def save_done(self, flag=False): + np.save(os.path.join(self.save_dir, 'done_flag'),flag) + np.savetxt(os.path.join(self.save_dir, 'done_flag'),[flag,],fmt='%i') diff --git a/mlflow/sf2f/scripts/DScore/models/dist_model.py b/mlflow/sf2f/scripts/DScore/models/dist_model.py new file mode 100644 index 0000000..d11eee1 --- /dev/null +++ b/mlflow/sf2f/scripts/DScore/models/dist_model.py @@ -0,0 +1,326 @@ +from __future__ import absolute_import + +import sys +sys.path.append('..') +sys.path.append('.') +import numpy as np +import torch +from torch import nn +import os +from collections import OrderedDict +from torch.autograd import Variable +import itertools +from .base_model import BaseModel +from scipy.ndimage import zoom +import fractions +import functools +import skimage.transform +from IPython import embed + +from . import networks_basic as networks +# from PerceptualSimilarity.util import util +from scripts.DScore.util import util + +class DistModel(BaseModel): + def name(self): + return self.model_name + + def initialize(self, model='net-lin', net='alex', pnet_rand=False, pnet_tune=False, model_path=None, colorspace='Lab', use_gpu=True, printNet=False, spatial=False, spatial_shape=None, spatial_order=1, spatial_factor=None, is_train=False, lr=.0001, beta1=0.5, version='0.1'): + ''' + INPUTS + model - ['net-lin'] for linearly calibrated network + ['net'] for off-the-shelf network + ['L2'] for L2 distance in Lab colorspace + ['SSIM'] for ssim in RGB colorspace + net - ['squeeze','alex','vgg'] + model_path - if None, will look in weights/[NET_NAME].pth + colorspace - ['Lab','RGB'] colorspace to use for L2 and SSIM + use_gpu - bool - whether or not to use a GPU + printNet - bool - whether or not to print network architecture out + spatial - bool - whether to output an array containing varying distances across spatial dimensions + spatial_shape - if given, output spatial shape. if None then spatial shape is determined automatically via spatial_factor (see below). + spatial_factor - if given, specifies upsampling factor relative to the largest spatial extent of a convolutional layer. if None then resized to size of input images. + spatial_order - spline order of filter for upsampling in spatial mode, by default 1 (bilinear). + is_train - bool - [True] for training mode + lr - float - initial learning rate + beta1 - float - initial momentum term for adam + version - 0.1 for latest, 0.0 was original + ''' + BaseModel.initialize(self, use_gpu=use_gpu) + + self.model = model + self.net = net + self.use_gpu = use_gpu + self.is_train = is_train + self.spatial = spatial + self.spatial_shape = spatial_shape + self.spatial_order = spatial_order + self.spatial_factor = spatial_factor + + self.model_name = '%s [%s]'%(model,net) + if(self.model == 'net-lin'): # pretrained net + linear layer + self.net = networks.PNetLin(use_gpu=use_gpu,pnet_rand=pnet_rand, pnet_tune=pnet_tune, pnet_type=net,use_dropout=True,spatial=spatial,version=version) + kw = {} + if not use_gpu: + kw['map_location'] = 'cpu' + if(model_path is None): + import inspect + # model_path = './PerceptualSimilarity/weights/v%s/%s.pth'%(version,net) + model_path = os.path.abspath(os.path.join(inspect.getfile(self.initialize), '..', '..', 'weights/v%s/%s.pth'%(version,net))) + + if(not is_train): + # print('Loading model from: %s'%model_path) + self.net.load_state_dict(torch.load(model_path, **kw)) + + elif(self.model=='net'): # pretrained network + assert not self.spatial, 'spatial argument not supported yet for uncalibrated networks' + self.net = networks.PNet(use_gpu=use_gpu,pnet_type=net) + self.is_fake_net = True + elif(self.model in ['L2','l2']): + self.net = networks.L2(use_gpu=use_gpu,colorspace=colorspace) # not really a network, only for testing + self.model_name = 'L2' + elif(self.model in ['DSSIM','dssim','SSIM','ssim']): + self.net = networks.DSSIM(use_gpu=use_gpu,colorspace=colorspace) + self.model_name = 'SSIM' + else: + raise ValueError("Model [%s] not recognized." % self.model) + + self.parameters = list(self.net.parameters()) + + if self.is_train: # training mode + # extra network on top to go from distances (d0,d1) => predicted human judgment (h*) + self.rankLoss = networks.BCERankingLoss(use_gpu=use_gpu) + self.parameters+=self.rankLoss.parameters + self.lr = lr + self.old_lr = lr + self.optimizer_net = torch.optim.Adam(self.parameters, lr=lr, betas=(beta1, 0.999)) + else: # test mode + self.net.eval() + + if(printNet): + print('---------- Networks initialized -------------') + networks.print_network(self.net) + print('-----------------------------------------------') + + def forward_pair(self,in1,in2,retPerLayer=False): + if(retPerLayer): + return self.net.forward(in1,in2, retPerLayer=True) + else: + return self.net.forward(in1,in2) + + def forward(self, in0, in1, retNumpy=True): + ''' Function computes the distance between image patches in0 and in1 + INPUTS + in0, in1 - torch.Tensor object of shape Nx3xXxY - image patch scaled to [-1,1] + retNumpy - [False] to return as torch.Tensor, [True] to return as numpy array + OUTPUT + computed distances between in0 and in1 + ''' + + self.input_ref = in0 + self.input_p0 = in1 + + if(self.use_gpu): + self.input_ref = self.input_ref.cuda() + self.input_p0 = self.input_p0.cuda() + + self.var_ref = Variable(self.input_ref,requires_grad=True) + self.var_p0 = Variable(self.input_p0,requires_grad=True) + + self.d0 = self.forward_pair(self.var_ref, self.var_p0) + self.loss_total = self.d0 + + def convert_output(d0): + if(retNumpy): + ans = d0.cpu().data.numpy() + if not self.spatial: + ans = ans.flatten() + else: + assert(ans.shape[0] == 1 and len(ans.shape) == 4) + return ans[0,...].transpose([1, 2, 0]) # Reshape to usual numpy image format: (height, width, channels) + return ans + else: + return d0 + + if self.spatial: + L = [convert_output(x) for x in self.d0] + spatial_shape = self.spatial_shape + if spatial_shape is None: + if(self.spatial_factor is None): + spatial_shape = (in0.size()[2],in0.size()[3]) + else: + spatial_shape = (max([x.shape[0] for x in L])*self.spatial_factor, max([x.shape[1] for x in L])*self.spatial_factor) + + L = [skimage.transform.resize(x, spatial_shape, order=self.spatial_order, mode='edge') for x in L] + + L = np.mean(np.concatenate(L, 2) * len(L), 2) + return L + else: + return convert_output(self.d0) + + # ***** TRAINING FUNCTIONS ***** + def optimize_parameters(self): + self.forward_train() + self.optimizer_net.zero_grad() + self.backward_train() + self.optimizer_net.step() + self.clamp_weights() + + def clamp_weights(self): + for module in self.net.modules(): + if(hasattr(module, 'weight') and module.kernel_size==(1,1)): + module.weight.data = torch.clamp(module.weight.data,min=0) + + def set_input(self, data): + self.input_ref = data['ref'] + self.input_p0 = data['p0'] + self.input_p1 = data['p1'] + self.input_judge = data['judge'] + + if(self.use_gpu): + self.input_ref = self.input_ref.cuda() + self.input_p0 = self.input_p0.cuda() + self.input_p1 = self.input_p1.cuda() + self.input_judge = self.input_judge.cuda() + + self.var_ref = Variable(self.input_ref,requires_grad=True) + self.var_p0 = Variable(self.input_p0,requires_grad=True) + self.var_p1 = Variable(self.input_p1,requires_grad=True) + + def forward_train(self): # run forward pass + self.d0 = self.forward_pair(self.var_ref, self.var_p0) + self.d1 = self.forward_pair(self.var_ref, self.var_p1) + self.acc_r = self.compute_accuracy(self.d0,self.d1,self.input_judge) + + # var_judge + self.var_judge = Variable(1.*self.input_judge).view(self.d0.size()) + + self.loss_total = self.rankLoss.forward(self.d0, self.d1, self.var_judge*2.-1.) + return self.loss_total + + def backward_train(self): + torch.mean(self.loss_total).backward() + + def compute_accuracy(self,d0,d1,judge): + ''' d0, d1 are Variables, judge is a Tensor ''' + d1_lt_d0 = (d1 %f' % (type,self.old_lr, lr)) + self.old_lr = lr + + + +def score_2afc_dataset(data_loader,func): + ''' Function computes Two Alternative Forced Choice (2AFC) score using + distance function 'func' in dataset 'data_loader' + INPUTS + data_loader - CustomDatasetDataLoader object - contains a TwoAFCDataset inside + func - callable distance function - calling d=func(in0,in1) should take 2 + pytorch tensors with shape Nx3xXxY, and return numpy array of length N + OUTPUTS + [0] - 2AFC score in [0,1], fraction of time func agrees with human evaluators + [1] - dictionary with following elements + d0s,d1s - N arrays containing distances between reference patch to perturbed patches + gts - N array in [0,1], preferred patch selected by human evaluators + (closer to "0" for left patch p0, "1" for right patch p1, + "0.6" means 60pct people preferred right patch, 40pct preferred left) + scores - N array in [0,1], corresponding to what percentage function agreed with humans + CONSTS + N - number of test triplets in data_loader + ''' + + d0s = [] + d1s = [] + gts = [] + + # bar = pb.ProgressBar(max_value=data_loader.load_data().__len__()) + for (i,data) in enumerate(data_loader.load_data()): + d0s+=func(data['ref'],data['p0']).tolist() + d1s+=func(data['ref'],data['p1']).tolist() + gts+=data['judge'].cpu().numpy().flatten().tolist() + # bar.update(i) + + d0s = np.array(d0s) + d1s = np.array(d1s) + gts = np.array(gts) + scores = (d0s 0: + with self.doc.head: + meta(http_equiv="reflesh", content=str(reflesh)) + + def get_image_dir(self): + return self.img_dir + + def add_header(self, str): + with self.doc: + h3(str) + + def add_table(self, border=1): + self.t = table(border=border, style="table-layout: fixed;") + self.doc.add(self.t) + + def add_images(self, ims, txts, links, width=400): + self.add_table() + with self.t: + with tr(): + for im, txt, link in zip(ims, txts, links): + with td(style="word-wrap: break-word;", halign="center", valign="top"): + with p(): + with a(href=os.path.join(link)): + img(style="width:%dpx" % width, src=os.path.join(im)) + br() + p(txt) + + def save(self,file='index'): + html_file = '%s/%s.html' % (self.web_dir,file) + f = open(html_file, 'wt') + f.write(self.doc.render()) + f.close() + + +if __name__ == '__main__': + html = HTML('web/', 'test_html') + html.add_header('hello world') + + ims = [] + txts = [] + links = [] + for n in range(4): + ims.append('image_%d.png' % n) + txts.append('text_%d' % n) + links.append('image_%d.png' % n) + html.add_images(ims, txts, links) + html.save() diff --git a/mlflow/sf2f/scripts/DScore/util/util.py b/mlflow/sf2f/scripts/DScore/util/util.py new file mode 100644 index 0000000..1a84327 --- /dev/null +++ b/mlflow/sf2f/scripts/DScore/util/util.py @@ -0,0 +1,452 @@ +from __future__ import print_function + +import numpy as np +from PIL import Image +import inspect +import re +import numpy as np +import os +import collections +import matplotlib.pyplot as plt +from scipy.ndimage.interpolation import zoom +from skimage.measure import compare_ssim +import torch +from IPython import embed +# import cv2 +from datetime import datetime + +def datetime_str(): + now = datetime.now() + return '%04d-%02d-%02d-%02d-%02d-%02d'%(now.year,now.month,now.day,now.hour,now.minute,now.second) + +def read_text_file(in_path): + fid = open(in_path,'r') + + vals = [] + cur_line = fid.readline() + while(cur_line!=''): + vals.append(float(cur_line)) + cur_line = fid.readline() + + fid.close() + return np.array(vals) + +def bootstrap(in_vec,num_samples=100,bootfunc=np.mean): + from astropy import stats + return stats.bootstrap(np.array(in_vec),bootnum=num_samples,bootfunc=bootfunc) + +def rand_flip(input1,input2): + if(np.random.binomial(1,.5)==1): + return (input1,input2) + else: + return (input2,input1) + +def l2(p0, p1, range=255.): + return .5*np.mean((p0 / range - p1 / range)**2) + +def psnr(p0, p1, peak=255.): + return 10*np.log10(peak**2/np.mean((1.*p0-1.*p1)**2)) + +def dssim(p0, p1, range=255.): + # embed() + return (1 - compare_ssim(p0, p1, data_range=range, multichannel=True)) / 2. + +def rgb2lab(in_img,mean_cent=False): + from skimage import color + img_lab = color.rgb2lab(in_img) + if(mean_cent): + img_lab[:,:,0] = img_lab[:,:,0]-50 + return img_lab + +def normalize_blob(in_feat,eps=1e-10): + norm_factor = np.sqrt(np.sum(in_feat**2,axis=1,keepdims=True)) + return in_feat/(norm_factor+eps) + +def cos_sim_blob(in0,in1): + in0_norm = normalize_blob(in0) + in1_norm = normalize_blob(in1) + (N,C,X,Y) = in0_norm.shape + + return np.mean(np.mean(np.sum(in0_norm*in1_norm,axis=1),axis=1),axis=1) + +def normalize_tensor(in_feat,eps=1e-10): + # norm_factor = torch.sqrt(torch.sum(in_feat**2,dim=1)).view(in_feat.size()[0],1,in_feat.size()[2],in_feat.size()[3]).repeat(1,in_feat.size()[1],1,1) + norm_factor = torch.sqrt(torch.sum(in_feat**2,dim=1)).view(in_feat.size()[0],1,in_feat.size()[2],in_feat.size()[3]) + return in_feat/(norm_factor.expand_as(in_feat)+eps) + +def cos_sim(in0,in1): + in0_norm = normalize_tensor(in0) + in1_norm = normalize_tensor(in1) + N = in0.size()[0] + X = in0.size()[2] + Y = in0.size()[3] + + return torch.mean(torch.mean(torch.sum(in0_norm*in1_norm,dim=1).view(N,1,X,Y),dim=2).view(N,1,1,Y),dim=3).view(N) + +# Converts a Tensor into a Numpy array +# |imtype|: the desired type of the conve + +def tensor2np(tensor_obj): + # change dimension of a tensor object into a numpy array + return tensor_obj[0].cpu().float().numpy().transpose((1,2,0)) + +def np2tensor(np_obj): + # change dimenion of np array into tensor array + return torch.Tensor(np_obj[:, :, :, np.newaxis].transpose((3, 2, 0, 1))) + +def tensor2tensorlab(image_tensor,to_norm=True,mc_only=False): + # image tensor to lab tensor + from skimage import color + + img = tensor2im(image_tensor) + # print('img_rgb',img.flatten()) + img_lab = color.rgb2lab(img) + # print('img_lab',img_lab.flatten()) + if(mc_only): + img_lab[:,:,0] = img_lab[:,:,0]-50 + if(to_norm and not mc_only): + img_lab[:,:,0] = img_lab[:,:,0]-50 + img_lab = img_lab/100. + + return np2tensor(img_lab) + +def tensorlab2tensor(lab_tensor,return_inbnd=False): + from skimage import color + import warnings + warnings.filterwarnings("ignore") + + lab = tensor2np(lab_tensor)*100. + lab[:,:,0] = lab[:,:,0]+50 + # print('lab',lab) + + rgb_back = 255.*np.clip(color.lab2rgb(lab.astype('float')),0,1) + # print('rgb',rgb_back) + if(return_inbnd): + # convert back to lab, see if we match + lab_back = color.rgb2lab(rgb_back.astype('uint8')) + # print('lab_back',lab_back) + # print('lab==lab_back',np.isclose(lab_back,lab,atol=1.)) + # print('lab-lab_back',np.abs(lab-lab_back)) + mask = 1.*np.isclose(lab_back,lab,atol=2.) + mask = np2tensor(np.prod(mask,axis=2)[:,:,np.newaxis]) + return (im2tensor(rgb_back),mask) + else: + return im2tensor(rgb_back) + +def tensor2im(image_tensor, imtype=np.uint8, cent=1., factor=255./2.): +# def tensor2im(image_tensor, imtype=np.uint8, cent=1., factor=1.): + image_numpy = image_tensor[0].cpu().float().numpy() + image_numpy = (np.transpose(image_numpy, (1, 2, 0)) + cent) * factor + return image_numpy.astype(imtype) + +def im2tensor(image, imtype=np.uint8, cent=1., factor=255./2.): +# def im2tensor(image, imtype=np.uint8, cent=1., factor=1.): + return torch.Tensor((image / factor - cent) + [:, :, :, np.newaxis].transpose((3, 2, 0, 1))) + +def tensor2vec(vector_tensor): + return vector_tensor.data.cpu().numpy()[:, :, 0, 0] + +def diagnose_network(net, name='network'): + mean = 0.0 + count = 0 + for param in net.parameters(): + if param.grad is not None: + mean += torch.mean(torch.abs(param.grad.data)) + count += 1 + if count > 0: + mean = mean / count + print(name) + print(mean) + +def grab_patch(img_in, P, yy, xx): + return img_in[yy:yy+P,xx:xx+P,:] + +# def load_image(path): +# if(path[-3:] == 'dng'): +# import rawpy +# with rawpy.imread(path) as raw: +# img = raw.postprocess() +# # img = plt.imread(path) +# elif(path[-3:]=='bmp' or path[-3:]=='jpg' or path[-3:]=='png'): +# import cv2 +# return cv2.imread(path)[:,:,::-1] +# else: +# img = (255*plt.imread(path)[:,:,:3]).astype('uint8') +# +# return img + + +def resize_image(img, max_size=256): + [Y, X] = img.shape[:2] + + # resize + max_dim = max([Y, X]) + zoom_factor = 1. * max_size / max_dim + img = zoom(img, [zoom_factor, zoom_factor, 1]) + + return img + +def resize_image_zoom(img, zoom_factor=1., order=3): + if(zoom_factor==1): + return img + else: + return zoom(img, [zoom_factor, zoom_factor, 1], order=order) + +def save_image(image_numpy, image_path, ): + image_pil = Image.fromarray(image_numpy) + image_pil.save(image_path) + + +def prep_display_image(img, dtype='uint8'): + if(dtype == 'uint8'): + return np.clip(img, 0, 255).astype('uint8') + else: + return np.clip(img, 0, 1.) + + +def info(object, spacing=10, collapse=1): + """Print methods and doc strings. + Takes module, class, list, dictionary, or string.""" + methodList = [ + e for e in dir(object) if isinstance( + getattr( + object, + e), + collections.Callable)] + processFunc = collapse and (lambda s: " ".join(s.split())) or (lambda s: s) + print("\n".join(["%s %s" % + (method.ljust(spacing), + processFunc(str(getattr(object, method).__doc__))) + for method in methodList])) + + +def varname(p): + for line in inspect.getframeinfo(inspect.currentframe().f_back)[3]: + m = re.search(r'\bvarname\s*\(\s*([A-Za-z_][A-Za-z0-9_]*)\s*\)', line) + if m: + return m.group(1) + + +def print_numpy(x, val=True, shp=False): + x = x.astype(np.float64) + if shp: + print('shape,', x.shape) + if val: + x = x.flatten() + print( + 'mean = %3.3f, min = %3.3f, max = %3.3f, median = %3.3f, std=%3.3f' % + (np.mean(x), np.min(x), np.max(x), np.median(x), np.std(x))) + + +def mkdirs(paths): + if isinstance(paths, list) and not isinstance(paths, str): + for path in paths: + mkdir(path) + else: + mkdir(paths) + + +def mkdir(path): + if not os.path.exists(path): + os.makedirs(path) + + +def rgb2lab(input): + from skimage import color + return color.rgb2lab(input / 255.) + + +def montage( + imgs, + PAD=5, + RATIO=16 / 9., + EXTRA_PAD=( + False, + False), + MM=-1, + NN=-1, + primeDir=0, + verbose=False, + returnGridPos=False, + backClr=np.array( + (0, + 0, + 0))): + # INPUTS + # imgs YxXxMxN or YxXxN + # PAD scalar number of pixels in between + # RATIO scalar target ratio of cols/rows + # MM scalar # rows, if specified, overrides RATIO + # NN scalar # columns, if specified, overrides RATIO + # primeDir scalar 0 for top-to-bottom, 1 for left-to-right + # OUTPUTS + # mont_imgs MM*Y x NN*X x M big image with everything montaged + # def montage(imgs, PAD=5, RATIO=16/9., MM=-1, NN=-1, primeDir=0, + # verbose=False, forceFloat=False): + if(imgs.ndim == 3): + toExp = True + imgs = imgs[:, :, np.newaxis, :] + else: + toExp = False + + Y = imgs.shape[0] + X = imgs.shape[1] + M = imgs.shape[2] + N = imgs.shape[3] + + PADS = np.array((PAD)) + if(PADS.flatten().size == 1): + PADY = PADS + PADX = PADS + else: + PADY = PADS[0] + PADX = PADS[1] + + if(MM == -1 and NN == -1): + NN = np.ceil(np.sqrt(1.0 * N * RATIO)) + MM = np.ceil(1.0 * N / NN) + NN = np.ceil(1.0 * N / MM) + elif(MM == -1): + MM = np.ceil(1.0 * N / NN) + elif(NN == -1): + NN = np.ceil(1.0 * N / MM) + + if(primeDir == 0): # write top-to-bottom + [grid_mm, grid_nn] = np.meshgrid( + np.arange(MM, dtype='uint'), np.arange(NN, dtype='uint')) + elif(primeDir == 1): # write left-to-right + [grid_nn, grid_mm] = np.meshgrid( + np.arange(NN, dtype='uint'), np.arange(MM, dtype='uint')) + + grid_mm = np.uint(grid_mm.flatten()[0:N]) + grid_nn = np.uint(grid_nn.flatten()[0:N]) + + EXTRA_PADY = EXTRA_PAD[0] * PADY + EXTRA_PADX = EXTRA_PAD[0] * PADX + + # mont_imgs = np.zeros(((Y+PAD)*MM-PAD, (X+PAD)*NN-PAD, M), dtype=use_dtype) + mont_imgs = np.zeros( + (np.uint( + (Y + PADY) * MM - PADY + EXTRA_PADY), + np.uint( + (X + PADX) * NN - PADX + EXTRA_PADX), + M), + dtype=imgs.dtype) + mont_imgs = mont_imgs + \ + backClr.flatten()[np.newaxis, np.newaxis, :].astype(mont_imgs.dtype) + + for ii in np.random.permutation(N): + # print imgs[:,:,:,ii].shape + # mont_imgs[grid_mm[ii]*(Y+PAD):(grid_mm[ii]*(Y+PAD)+Y), grid_nn[ii]*(X+PAD):(grid_nn[ii]*(X+PAD)+X),:] + mont_imgs[np.uint(grid_mm[ii] * + (Y + + PADY)):np.uint((grid_mm[ii] * + (Y + + PADY) + + Y)), np.uint(grid_nn[ii] * + (X + + PADX)):np.uint((grid_nn[ii] * + (X + + PADX) + + X)), :] = imgs[:, :, :, ii] + + if(M == 1): + imgs = imgs.reshape(imgs.shape[0], imgs.shape[1], imgs.shape[3]) + + if(toExp): + mont_imgs = mont_imgs[:, :, 0] + + if(returnGridPos): + # return (mont_imgs,np.concatenate((grid_mm[:,:,np.newaxis]*(Y+PAD), + # grid_nn[:,:,np.newaxis]*(X+PAD)),axis=2)) + return (mont_imgs, np.concatenate( + (grid_mm[:, np.newaxis] * (Y + PADY), grid_nn[:, np.newaxis] * (X + PADX)), axis=1)) + # return (mont_imgs, (grid_mm,grid_nn)) + else: + return mont_imgs + +class zeroClipper(object): + def __init__(self, frequency=1): + self.frequency = frequency + + def __call__(self, module): + embed() + if hasattr(module, 'weight'): + # module.weight.data = torch.max(module.weight.data, 0) + module.weight.data = torch.max(module.weight.data, 0) + 100 + +def flatten_nested_list(nested_list): + # only works for list of list + accum = [] + for sublist in nested_list: + for item in sublist: + accum.append(item) + return accum + +def read_file(in_path,list_lines=False): + agg_str = '' + f = open(in_path,'r') + cur_line = f.readline() + while(cur_line!=''): + agg_str+=cur_line + cur_line = f.readline() + f.close() + if(list_lines==False): + return agg_str.replace('\n','') + else: + line_list = agg_str.split('\n') + ret_list = [] + for item in line_list: + if(item!=''): + ret_list.append(item) + return ret_list + +def read_csv_file_as_text(in_path): + agg_str = [] + f = open(in_path,'r') + cur_line = f.readline() + while(cur_line!=''): + agg_str.append(cur_line) + cur_line = f.readline() + f.close() + return agg_str + +def random_swap(obj0,obj1): + if(np.random.rand() < .5): + return (obj0,obj1,0) + else: + return (obj1,obj0,1) + +def voc_ap(rec, prec, use_07_metric=False): + """ ap = voc_ap(rec, prec, [use_07_metric]) + Compute VOC AP given precision and recall. + If use_07_metric is true, uses the + VOC 07 11 point method (default:False). + """ + if use_07_metric: + # 11 point metric + ap = 0. + for t in np.arange(0., 1.1, 0.1): + if np.sum(rec >= t) == 0: + p = 0 + else: + p = np.max(prec[rec >= t]) + ap = ap + p / 11. + else: + # correct AP calculation + # first append sentinel values at the end + mrec = np.concatenate(([0.], rec, [1.])) + mpre = np.concatenate(([0.], prec, [0.])) + + # compute the precision envelope + for i in range(mpre.size - 1, 0, -1): + mpre[i - 1] = np.maximum(mpre[i - 1], mpre[i]) + + # to calculate area under PR curve, look for points + # where X axis (recall) changes value + i = np.where(mrec[1:] != mrec[:-1])[0] + + # and sum (\Delta recall) * prec + ap = np.sum((mrec[i + 1] - mrec[i]) * mpre[i + 1]) + return ap diff --git a/mlflow/sf2f/scripts/DScore/util/visualizer.py b/mlflow/sf2f/scripts/DScore/util/visualizer.py new file mode 100644 index 0000000..3609705 --- /dev/null +++ b/mlflow/sf2f/scripts/DScore/util/visualizer.py @@ -0,0 +1,216 @@ +import numpy as np +import os +import time +from . import util +from . import html +# from pdb import set_trace as st +import matplotlib.pyplot as plt +import math +# from IPython import embed + +def zoom_to_res(img,res=256,order=0,axis=0): + # img 3xXxX + from scipy.ndimage import zoom + zoom_factor = res/img.shape[1] + if(axis==0): + return zoom(img,[1,zoom_factor,zoom_factor],order=order) + elif(axis==2): + return zoom(img,[zoom_factor,zoom_factor,1],order=order) + +class Visualizer(): + def __init__(self, opt): + # self.opt = opt + self.display_id = opt.display_id + # self.use_html = opt.is_train and not opt.no_html + self.win_size = opt.display_winsize + self.name = opt.name + self.display_cnt = 0 # display_current_results counter + self.display_cnt_high = 0 + self.use_html = opt.use_html + + if self.display_id > 0: + import visdom + self.vis = visdom.Visdom(port = opt.display_port) + + self.web_dir = os.path.join(opt.checkpoints_dir, opt.name, 'web') + util.mkdirs([self.web_dir,]) + if self.use_html: + self.img_dir = os.path.join(self.web_dir, 'images') + print('create web directory %s...' % self.web_dir) + util.mkdirs([self.img_dir,]) + + # |visuals|: dictionary of images to display or save + def display_current_results(self, visuals, epoch, nrows=None, res=256): + if self.display_id > 0: # show images in the browser + title = self.name + if(nrows is None): + nrows = int(math.ceil(len(visuals.items()) / 2.0)) + images = [] + idx = 0 + for label, image_numpy in visuals.items(): + title += " | " if idx % nrows == 0 else ", " + title += label + img = image_numpy.transpose([2, 0, 1]) + img = zoom_to_res(img,res=res,order=0) + images.append(img) + idx += 1 + if len(visuals.items()) % 2 != 0: + white_image = np.ones_like(image_numpy.transpose([2, 0, 1]))*255 + white_image = zoom_to_res(white_image,res=res,order=0) + images.append(white_image) + self.vis.images(images, nrow=nrows, win=self.display_id + 1, + opts=dict(title=title)) + + if self.use_html: # save images to a html file + for label, image_numpy in visuals.items(): + img_path = os.path.join(self.img_dir, 'epoch%.3d_cnt%.6d_%s.png' % (epoch, self.display_cnt, label)) + util.save_image(zoom_to_res(image_numpy, res=res, axis=2), img_path) + + self.display_cnt += 1 + self.display_cnt_high = np.maximum(self.display_cnt_high, self.display_cnt) + + # update website + webpage = html.HTML(self.web_dir, 'Experiment name = %s' % self.name, reflesh=1) + for n in range(epoch, 0, -1): + webpage.add_header('epoch [%d]' % n) + if(n==epoch): + high = self.display_cnt + else: + high = self.display_cnt_high + for c in range(high-1,-1,-1): + ims = [] + txts = [] + links = [] + + for label, image_numpy in visuals.items(): + img_path = 'epoch%.3d_cnt%.6d_%s.png' % (n, c, label) + ims.append(os.path.join('images',img_path)) + txts.append(label) + links.append(os.path.join('images',img_path)) + webpage.add_images(ims, txts, links, width=self.win_size) + webpage.save() + + # save errors into a directory + def plot_current_errors_save(self, epoch, counter_ratio, opt, errors,keys='+ALL',name='loss', to_plot=False): + if not hasattr(self, 'plot_data'): + self.plot_data = {'X':[],'Y':[], 'legend':list(errors.keys())} + self.plot_data['X'].append(epoch + counter_ratio) + self.plot_data['Y'].append([errors[k] for k in self.plot_data['legend']]) + + # embed() + if(keys=='+ALL'): + plot_keys = self.plot_data['legend'] + else: + plot_keys = keys + + if(to_plot): + (f,ax) = plt.subplots(1,1) + for (k,kname) in enumerate(plot_keys): + kk = np.where(np.array(self.plot_data['legend'])==kname)[0][0] + x = self.plot_data['X'] + y = np.array(self.plot_data['Y'])[:,kk] + if(to_plot): + ax.plot(x, y, 'o-', label=kname) + np.save(os.path.join(self.web_dir,'%s_x')%kname,x) + np.save(os.path.join(self.web_dir,'%s_y')%kname,y) + + if(to_plot): + plt.legend(loc=0,fontsize='small') + plt.xlabel('epoch') + plt.ylabel('Value') + f.savefig(os.path.join(self.web_dir,'%s.png'%name)) + f.clf() + plt.close() + + # errors: dictionary of error labels and values + def plot_current_errors(self, epoch, counter_ratio, opt, errors): + if not hasattr(self, 'plot_data'): + self.plot_data = {'X':[],'Y':[], 'legend':list(errors.keys())} + self.plot_data['X'].append(epoch + counter_ratio) + self.plot_data['Y'].append([errors[k] for k in self.plot_data['legend']]) + self.vis.line( + X=np.stack([np.array(self.plot_data['X'])]*len(self.plot_data['legend']),1), + Y=np.array(self.plot_data['Y']), + opts={ + 'title': self.name + ' loss over time', + 'legend': self.plot_data['legend'], + 'xlabel': 'epoch', + 'ylabel': 'loss'}, + win=self.display_id) + + # errors: same format as |errors| of plotCurrentErrors + def print_current_errors(self, epoch, i, errors, t, t2=-1, t2o=-1, fid=None): + message = '(ep: %d, it: %d, t: %.3f[s], ept: %.2f/%.2f[h]) ' % (epoch, i, t, t2o, t2) + message += (', ').join(['%s: %.3f' % (k, v) for k, v in errors.items()]) + + print(message) + if(fid is not None): + fid.write('%s\n'%message) + + + # save image to the disk + def save_images_simple(self, webpage, images, names, in_txts, prefix='', res=256): + image_dir = webpage.get_image_dir() + ims = [] + txts = [] + links = [] + + for name, image_numpy, txt in zip(names, images, in_txts): + image_name = '%s_%s.png' % (prefix, name) + save_path = os.path.join(image_dir, image_name) + if(res is not None): + util.save_image(zoom_to_res(image_numpy,res=res,axis=2), save_path) + else: + util.save_image(image_numpy, save_path) + + ims.append(os.path.join(webpage.img_subdir,image_name)) + # txts.append(name) + txts.append(txt) + links.append(os.path.join(webpage.img_subdir,image_name)) + # embed() + webpage.add_images(ims, txts, links, width=self.win_size) + + # save image to the disk + def save_images(self, webpage, images, names, image_path, title=''): + image_dir = webpage.get_image_dir() + # short_path = ntpath.basename(image_path) + # name = os.path.splitext(short_path)[0] + # name = short_path + # webpage.add_header('%s, %s' % (name, title)) + ims = [] + txts = [] + links = [] + + for label, image_numpy in zip(names, images): + image_name = '%s.jpg' % (label,) + save_path = os.path.join(image_dir, image_name) + util.save_image(image_numpy, save_path) + + ims.append(image_name) + txts.append(label) + links.append(image_name) + webpage.add_images(ims, txts, links, width=self.win_size) + + # save image to the disk + # def save_images(self, webpage, visuals, image_path, short=False): + # image_dir = webpage.get_image_dir() + # if short: + # short_path = ntpath.basename(image_path) + # name = os.path.splitext(short_path)[0] + # else: + # name = image_path + + # webpage.add_header(name) + # ims = [] + # txts = [] + # links = [] + + # for label, image_numpy in visuals.items(): + # image_name = '%s_%s.png' % (name, label) + # save_path = os.path.join(image_dir, image_name) + # util.save_image(image_numpy, save_path) + + # ims.append(image_name) + # txts.append(label) + # links.append(image_name) + # webpage.add_images(ims, txts, links, width=self.win_size) diff --git a/mlflow/sf2f/scripts/build_demo_set.py b/mlflow/sf2f/scripts/build_demo_set.py new file mode 100644 index 0000000..127a76e --- /dev/null +++ b/mlflow/sf2f/scripts/build_demo_set.py @@ -0,0 +1,85 @@ +''' +Build a demo inference dataset +''' + + +import os +import shutil +import json +import pandas as pd +from PIL import Image +from pydub import AudioSegment + + +DATA_DIR = './data' +vox2_face_dir = os.path.join(DATA_DIR, 'VoxCeleb', 'vox2', 'masked_faces') +vox2_wav_dir = os.path.join(DATA_DIR, 'VoxCeleb', 'raw_wav', 'vox2') +vox2_meta_csv = os.path.join(DATA_DIR, 'VoxCeleb', 'vox2', 'full_vox2_meta.csv') +vox_split_json = os.path.join(DATA_DIR, 'VoxCeleb', 'split.json') +demo_dir = os.path.join(DATA_DIR, 'VoxCeleb', 'demo_data') + +def main(): + try: + shutil.rmtree(demo_dir) + except: + pass + os.makedirs(demo_dir, exist_ok=True) + #print(os.listdir(vox2_face_dir)) + #print(os.listdir(vox2_wav_dir)) + with open(vox_split_json) as json_file: + split_dict = json.load(json_file) + # A list of names in the vox2 test set (in out splition) + test_list = split_dict['vox2']['test'] + # meta_csv + meta_df = pd.read_csv(vox2_meta_csv) + print(meta_df.head(5)) + name2id = {} + name2set = {} + for i in range(len(meta_df)): + name = meta_df['Name'][i] + id = meta_df['VoxCeleb2ID'][i] + split_set = meta_df['Set'][i] + name2id[name] = id + name2set[name] = split_set + #print(name2id) + available_names = set(test_list).intersection( + set(os.listdir(vox2_face_dir))) + available_names = list(available_names) + for name in available_names[:50]: + id = name2id[name] + split_set = name2set[name] + face_dir = os.path.join(vox2_face_dir, name) + #print(os.listdir(face_dir)) + wav_dir = os.path.join(vox2_wav_dir, split_set, id) + # Avoid PermissionError + try: + #print(os.listdir(wav_dir)) + wav_sub_dir_list = os.listdir(wav_dir) + except PermissionError: + continue + save_dir = os.path.join(demo_dir, name) + os.makedirs(save_dir, exist_ok=True) + for i, jpg in enumerate(os.listdir(face_dir)): + face_jpg = os.path.join(face_dir, jpg) + img = Image.open(face_jpg) + for reso, w in [('low', 64), ('mid', 128), ('high', 256)]: + save_path = os.path.join(save_dir, '{}_{}.jpg'.format(i, reso)) + cur_img = img.resize((w, w)) + cur_img.save(save_path) + # select 2 wavs + for j, wav_sub_dir in enumerate(wav_sub_dir_list[:2]): + #print(wav_sub_dir) + wav_sub_dir = os.path.join(wav_dir, wav_sub_dir) + wav_path = os.path.join(wav_sub_dir, os.listdir(wav_sub_dir)[0]) + wav_save_dir = os.path.join(save_dir, 'audio_{}'.format(j)) + os.makedirs(wav_save_dir, exist_ok=True) + try: + track = AudioSegment.from_file(wav_path, 'm4a') + wav_save_path = os.path.join(wav_save_dir, '{}.wav'.format(j)) + file_handle = track.export(wav_save_path, format='wav') + except: + print("ERROR CONVERTING " + str(wav_path)) + print(name) + +if __name__ == '__main__': + main() diff --git a/mlflow/sf2f/scripts/compute_diversity_score.py b/mlflow/sf2f/scripts/compute_diversity_score.py new file mode 100644 index 0000000..52e338c --- /dev/null +++ b/mlflow/sf2f/scripts/compute_diversity_score.py @@ -0,0 +1,28 @@ +from scripts.DScore.models import dist_model as dm +from scripts.DScore.util import util +import numpy as np +from pyprind import prog_bar + +def compute_diversity_score(images_A, images_B, use_gpu): + assert(type(images_A) == list) + assert(type(images_B) == list) + assert(len(images_A) == len(images_B)) + + img_pair_num = len(images_A) + + ## Initializing the model + model = dm.DistModel() + model.initialize(model='net-lin', net='alex', use_gpu=use_gpu) + + distance = [] + for i in prog_bar(range(img_pair_num), title="Calculating Diversity Scores", width=50): + img_A = util.im2tensor(images_A[i]) + img_B = util.im2tensor(images_B[i]) + dist = model.forward(img_A, img_B) + distance.append(dist) + + return np.mean(distance), np.std(distance) + + +if __name__ == '__main__': + print("please Call from other file.") diff --git a/mlflow/sf2f/scripts/compute_fid_score.py b/mlflow/sf2f/scripts/compute_fid_score.py new file mode 100644 index 0000000..d7a7212 --- /dev/null +++ b/mlflow/sf2f/scripts/compute_fid_score.py @@ -0,0 +1,230 @@ +# Code derived from https://github.com/openai/improved-gan/tree/master/inception_score +from __future__ import absolute_import +from __future__ import division +from __future__ import print_function + +import os.path +import sys +import tarfile + +import numpy as np +from six.moves import urllib +import tensorflow as tf +import glob +import scipy.misc +from scipy import linalg +import math +import sys +from pyprind import prog_bar + +tf.compat.v1.disable_v2_behavior() + + +try: + MODEL_DIR = '/mnt/.tensorflow/imagenet'#'/mnt/imagenet' + os.listdir(MODEL_DIR) +except: + MODEL_DIR = './.cache/.tensorflow/imagenet' +DATA_URL = 'http://download.tensorflow.org/models/image/imagenet/inception-2015-12-05.tgz' +softmax = None +last_layer = None + +config = tf.compat.v1.ConfigProto() +config.gpu_options.per_process_gpu_memory_fraction = 0.3 + + +def inception_forward(images, layer): + assert (type(images[0]) == np.ndarray) + assert (len(images[0].shape) == 3) + assert (np.max(images[0]) > 10) + assert (np.min(images[0]) >= 0.0) + # bs = 100 + bs = 1 + # images = images.transpose(0, 2, 3, 1) + #with tf.Session(config=config) as sess: + preds = [] + n_batches = int(math.ceil(float(len(images)) / float(bs))) + # for i in range(n_batches): + for i in prog_bar(range(n_batches), title="Calculating FID Scores", width=50): + # sys.stdout.write(".") + # sys.stdout.flush() + inps = images[(i * bs):min((i + 1) * bs, len(images))] + for inp in inps: + pred = fw_sess.run(layer, {'ExpandDims:0': [inp]}) + preds.append(pred) + preds = np.array(preds) + # preds = np.concatenate(preds, 0) + return preds + + +def get_mean_and_cov(images): + before_preds = inception_forward(images, last_layer) + m = np.mean(before_preds, 0) + cov = np.cov(before_preds, rowvar=False) + return m, cov + + +def get_fid(images, ref_stats=None, images_ref=None, splits=10): + before_preds = inception_forward(images, last_layer) + if ref_stats is None: + if images_ref is None: + raise ValueError('images_ref should be provided if ref_stats is None') + m_ref, cov_ref = get_mean_and_cov(images_ref) + fids = [] + for i in range(splits): + part = before_preds[(i * before_preds.shape[0] // splits):((i + 1) * before_preds.shape[0] // splits), :] + m_gen = np.mean(part, 0) + cov_gen = np.cov(part, rowvar=False) + fid = np.sum((m_ref - m_gen) ** 2) + np.trace( + cov_ref + cov_gen - 2 * scipy.linalg.sqrtm(np.dot(cov_ref, cov_gen))) + fids.append(fid) + return np.mean(fids), np.std(fids) + + +# Call this function with list of images. Each of elements should be a +# numpy array with values ranging from 0 to 255. +def get_inception_score(images, splits=10): + preds = inception_forward(images, softmax) + scores = [] + for i in range(splits): + part = preds[(i * preds.shape[0] // splits):((i + 1) * preds.shape[0] // splits), :] + kl = part * (np.log(part) - np.log(np.expand_dims(np.mean(part, 0), 0))) + kl = np.mean(np.sum(kl, 1)) + scores.append(np.exp(kl)) + return np.mean(scores), np.std(scores) + + +def get_inception_pred(images): + preds = inception_forward(images, softmax) + return preds + + +def get_fid_pred(images): + preds = inception_forward(images, last_layer) + return preds + +def close_sess(): + fw_sess.close() + + +# This function is called automatically. +def _init_inception(): + global softmax + global last_layer + if not os.path.exists(MODEL_DIR): + os.makedirs(MODEL_DIR) + filename = DATA_URL.split('/')[-1] + filepath = os.path.join(MODEL_DIR, filename) + if not os.path.exists(filepath): + def _progress(count, block_size, total_size): + sys.stdout.write('\r>> Downloading %s %.1f%%' % ( + filename, float(count * block_size) / float(total_size) * 100.0)) + sys.stdout.flush() + + filepath, _ = urllib.request.urlretrieve(DATA_URL, filepath, _progress) + print() + statinfo = os.stat(filepath) + print('Succesfully downloaded', filename, statinfo.st_size, 'bytes.') + tarfile.open(filepath, 'r:gz').extractall(MODEL_DIR) + with tf.io.gfile.GFile(os.path.join( + MODEL_DIR, 'classify_image_graph_def.pb'), 'rb') as f: + graph_def = tf.compat.v1.GraphDef() + graph_def.ParseFromString(f.read()) + _ = tf.import_graph_def(graph_def, name='') + # Works with an arbitrary minibatch size. + with tf.compat.v1.Session(config=config) as sess: + pool3 = sess.graph.get_tensor_by_name('pool_3:0') + ops = pool3.graph.get_operations() + for op_idx, op in enumerate(ops): + for o in op.outputs: + shape = o.get_shape() + shape = [s.value for s in shape] + new_shape = [] + for j, s in enumerate(shape): + if s == 1 and j == 0: + new_shape.append(None) + else: + new_shape.append(s) + # o.set_shape(tf.TensorShape(new_shape)) + o.set_shape = tf.TensorShape(new_shape) + w = sess.graph.get_operation_by_name("softmax/logits/MatMul").inputs[1] + last_layer = tf.squeeze(pool3) + logits = tf.matmul(tf.expand_dims(last_layer, 0), w) + softmax = tf.nn.softmax(logits[0]) + + global fw_sess + fw_sess = tf.compat.v1.Session(config=config) + + +if softmax is None: + _init_inception() + + + +def calculate_activation_statistics(act): + """Calculation of the statistics used by the FID. + Params: + -- act : Numpy array of dimension (n_images, dim (e.g. 2048)). + Returns: + -- mu : The mean over samples of the activations of the pool_3 layer of + the inception model. + -- sigma : The covariance matrix of the activations of the pool_3 layer of + the inception model. + """ + mu = np.mean(act, axis=0) + sigma = np.cov(act, rowvar=False) + return mu, sigma + + +def calculate_frechet_distance(mu1, sigma1, mu2, sigma2, eps=1e-6): + """Numpy implementation of the Frechet Distance. + The Frechet distance between two multivariate Gaussians X_1 ~ N(mu_1, C_1) + and X_2 ~ N(mu_2, C_2) is + d^2 = ||mu_1 - mu_2||^2 + Tr(C_1 + C_2 - 2*sqrt(C_1*C_2)). + Stable version by Dougal J. Sutherland. + Params: + -- mu1 : Numpy array containing the activations of a layer of the + inception net (like returned by the function 'get_predictions') + for generated samples. + -- mu2 : The sample mean over activations, precalculated on an + representive data set. + -- sigma1: The covariance matrix over activations for generated samples. + -- sigma2: The covariance matrix over activations, precalculated on an + representive data set. + Returns: + -- : The Frechet Distance. + """ + + mu1 = np.atleast_1d(mu1) + mu2 = np.atleast_1d(mu2) + + sigma1 = np.atleast_2d(sigma1) + sigma2 = np.atleast_2d(sigma2) + + assert mu1.shape == mu2.shape, \ + 'Training and test mean vectors have different lengths' + assert sigma1.shape == sigma2.shape, \ + 'Training and test covariances have different dimensions' + + diff = mu1 - mu2 + + # Product might be almost singular + covmean, _ = linalg.sqrtm(sigma1.dot(sigma2), disp=False) + if not np.isfinite(covmean).all(): + msg = ('fid calculation produces singular product; ' + 'adding %s to diagonal of cov estimates') % eps + print(msg) + offset = np.eye(sigma1.shape[0]) * eps + covmean = linalg.sqrtm((sigma1 + offset).dot(sigma2 + offset)) + + # Numerical error might give slight imaginary component + if np.iscomplexobj(covmean): + if not np.allclose(np.diagonal(covmean).imag, 0, atol=1e-3): + m = np.max(np.abs(covmean.imag)) + raise ValueError('Imaginary component {}'.format(m)) + covmean = covmean.real + + tr_covmean = np.trace(covmean) + + return (diff.dot(diff) + np.trace(sigma1) + + np.trace(sigma2) - 2 * tr_covmean) diff --git a/mlflow/sf2f/scripts/compute_inception_score.py b/mlflow/sf2f/scripts/compute_inception_score.py new file mode 100644 index 0000000..85992e2 --- /dev/null +++ b/mlflow/sf2f/scripts/compute_inception_score.py @@ -0,0 +1,229 @@ +# Code adapted from +# https://github.com/openai/improved-gan/blob/master/inception_score/model.py +# which was in turn derived from +# tensorflow/tensorflow/models/image/imagenet/classify_image.py +from __future__ import absolute_import +from __future__ import division +from __future__ import print_function + +import argparse +import os +import sys +import tarfile +import h5py + +import numpy as np +from six.moves import urllib +import tensorflow as tf +import glob +# from scipy.misc import imread, imresize +from cv2 import resize as imresize +from imageio import imread + +from imageio import imwrite +import math +from pyprind import prog_bar + +tf.compat.v1.disable_v2_behavior() + +os.environ['TF_CPP_MIN_LOG_LEVEL'] = '2' + +parser = argparse.ArgumentParser() +parser.add_argument('--input_npy_file', default=None) +parser.add_argument('--input_image_dir', default=None) +parser.add_argument('--input_image_dir_list', default=None) +parser.add_argument('--input_image_superdir', default=None) +parser.add_argument('--image_size', default=None, type=int) + +# Most papers use 50k samples and 10 splits but I don't have that much +# data so I'll use 3 splits for everything +parser.add_argument('--num_splits', default=3, type=int) +parser.add_argument('--tensor_layout', default='NHWC', choices=['NHWC', 'NCHW']) + +IMAGE_EXTS = ['.png', '.jpg', '.jpeg'] + +def main(args): + got_npy_file = args.input_npy_file is not None + got_image_dir = args.input_image_dir is not None + got_image_dir_list = args.input_image_dir_list is not None + got_image_superdir = args.input_image_superdir is not None + inputs = [got_npy_file, got_image_dir, got_image_dir_list, got_image_superdir] + if sum(inputs) != 1: + # raise ValueError('Must give exactly one input type') + images = read_images() + mean, std = get_inception_score(images) + print('Inception mean: ', mean) + print('Inception std: ', std) + + + if args.input_npy_file is not None: + images = np.load(args.input_npy_file) + images = np.split(image, images.shape[0], axis=0) + images = [img[0] for img in images] + mean, std = get_inception_score(args, images) + print('Inception mean: ', mean) + print('Inception std: ', std) + elif args.input_image_dir is not None: + images = load_images(args, args.input_image_dir) + import pdb; pdb.set_trace() + print(type(images)) + import pdb; pdb.set_trace() + exit() + mean, std = get_inception_score(args, images) + print('Inception mean: ', mean) + print('Inception std: ', std) + elif got_image_dir_list: + with open(args.input_image_dir_list, 'r') as f: + dir_list = [line.strip() for line in f] + for image_dir in dir_list: + images = load_images(args, image_dir) + mean, std = get_inception_score(args, images) + print('Inception mean: ', mean) + print('Inception std: ', std) + print() + elif got_image_superdir: + for fn in sorted(os.listdir(args.input_image_superdir)): + print(fn) + if not fn.startswith('result'): continue + image_dir = os.path.join(args.input_image_superdir, fn, 'images') + if not os.path.isdir(image_dir): continue + images = load_images(args, image_dir) + mean, std = get_inception_score(args, images) + print('Inception mean: ', mean) + print('Inception std: ', std) + print() + +def load_images(args, image_dir): + print('Loading images from ', image_dir) + images = [] + args.tensor_layout = 'NHWC' + for fn in os.listdir(image_dir): + ext = os.path.splitext(fn)[1].lower() + if ext not in IMAGE_EXTS: + continue + img_path = os.path.join(image_dir, fn) + img = imread(img_path) + if args.image_size is not None: + img = imresize(img, (args.image_size, args.image_size)) + images.append(img) + print('Found %d images' % len(images)) + return images + +def read_images(): + img_dir = './data/vg/images/' + vg_dataset = [] + print("Begin to load test...") + test_file = "./data/visual_genome/val.h5" + test_set = h5py.File(test_file, 'r') + test_set_name = np.array(test_set['image_paths']) + for iter, img_name in enumerate(test_set_name): + img_path = os.path.join(img_dir, img_name) + img = imread(img_path) + if len(img.shape) != 3: + continue + vg_dataset.append(img) + print('Found %d images' % len(vg_dataset)) + return vg_dataset +try: + MODEL_DIR = '/mnt/.tensorflow/imagenet'#'/mnt/imagenet' + os.listdir(MODEL_DIR) +except: + MODEL_DIR = './.cache/.tensorflow/imagenet' +DATA_URL = 'http://download.tensorflow.org/models/image/imagenet/inception-2015-12-05.tgz' +softmax = None + +# Call this function with list of images. Each of elements should be a +# numpy array with values ranging from 0 to 255. +def get_inception_score(images): + splits = 3 + layout = 'NHWC' + + assert(type(images) == list) + assert(type(images[0]) == np.ndarray) + assert(len(images[0].shape) == 3) + # assert(np.max(images[0]) > 10) + # assert(np.min(images[0]) >= 0.0) + inps = [] + for img in images: + img = img.astype(np.float32) + inps.append(np.expand_dims(img, 0)) + bs = 1 + # bs = 100 + with tf.compat.v1.Session() as sess: + preds = [] + n_batches = int(math.ceil(float(len(inps)) / float(bs))) + n_preds = 0 + for i in prog_bar(range(n_batches), title="Calculating Inception Scores", width=50): + # import pdb; pdb.set_trace() + inp = inps[(i * bs):min((i + 1) * bs, len(inps))] + + inp = np.concatenate(inp, 0) + if layout == 'NCHW': + inp = inp.transpose(0, 2, 3, 1) + # print(inp.shape) + pred = sess.run(softmax, {'ExpandDims:0': inp}) + preds.append(pred) + n_preds += pred.shape[0] + # print('Ran %d / %d images' % (n_preds, len(images))) + # import pdb; pdb.set_trace() + preds = np.concatenate(preds, 0) + scores = [] + for i in range(splits): + part = preds[(i * preds.shape[0] // splits):((i + 1) * preds.shape[0] // splits), :] + kl = part * (np.log(part) - np.log(np.expand_dims(np.mean(part, 0), 0))) + kl = np.mean(np.sum(kl, 1)) + scores.append(np.exp(kl)) + return np.mean(scores), np.std(scores) + +# This function is called automatically. +def _init_inception(): + global softmax + if not os.path.exists(MODEL_DIR): + os.makedirs(MODEL_DIR) + filename = DATA_URL.split('/')[-1] + filepath = os.path.join(MODEL_DIR, filename) + if not os.path.exists(filepath): + def _progress(count, block_size, total_size): + sys.stdout.write('\r>> Downloading %s %.1f%%' % ( + filename, float(count * block_size) / float(total_size) * 100.0)) + sys.stdout.flush() + filepath, _ = urllib.request.urlretrieve(DATA_URL, filepath, _progress) + print() + statinfo = os.stat(filepath) + print('Succesfully downloaded', filename, statinfo.st_size, 'bytes.') + tarfile.open(filepath, 'r:gz').extractall(MODEL_DIR) + with tf.io.gfile.GFile(os.path.join( + MODEL_DIR, 'classify_image_graph_def.pb'), 'rb') as f: + # + graph_def = tf.compat.v1.GraphDef() + graph_def.ParseFromString(f.read()) + _ = tf.import_graph_def(graph_def, name='') + # Works with an arbitrary minibatch size. + with tf.compat.v1.Session() as sess: + pool3 = sess.graph.get_tensor_by_name('pool_3:0') + ops = pool3.graph.get_operations() + for op_idx, op in enumerate(ops): + for o in op.outputs: + shape = o.get_shape() + shape = [s.value for s in shape] + new_shape = [] + for j, s in enumerate(shape): + if s == 1 and j == 0: + new_shape.append(None) + else: + new_shape.append(s) + # o._shape = tf.TensorShape(new_shape) + o.set_shape = tf.TensorShape(new_shape) + w = sess.graph.get_operation_by_name("softmax/logits/MatMul").inputs[1] + pool3 = tf.squeeze(pool3) + pool3 = tf.reshape(pool3, (1, 2048)) + # logits = tf.matmul(tf.squeeze(pool3), w) + logits = tf.matmul(pool3, w) + softmax = tf.nn.softmax(logits) + +if softmax is None: + _init_inception() + +if __name__ == '__main__': + args = parser.parse_args() + main(args) diff --git a/mlflow/sf2f/scripts/compute_mel_mean_var.py b/mlflow/sf2f/scripts/compute_mel_mean_var.py new file mode 100644 index 0000000..d730e17 --- /dev/null +++ b/mlflow/sf2f/scripts/compute_mel_mean_var.py @@ -0,0 +1,55 @@ +# -*- coding: utf-8 -*- +''' +Compute the mean and variance of VoxCeleb dataset's mel spectrograms +''' + + +import os +import numpy as np +import sys +import torch +from torch.utils.data import DataLoader +sys.path.append('./') +print(sys.path) +from datasets.vox_dataset import VoxDataset + + +def mean(): + epoch_num = 100 + vox_dataset = VoxDataset( + data_dir='./data/VoxCeleb', + image_size=(64, 64), + image_normalize_method=None, + mel_normalize_method=None) + loader_kwargs = { + 'batch_size': 128, + 'num_workers': 8, + 'shuffle': False, + "drop_last": True, + } + print(vox_dataset) + vox_loader = DataLoader(vox_dataset, **loader_kwargs) + + total_mean = torch.tensor(0.0) + total_std = torch.tensor(0.0) + min_length = 10000000 + for epoch in range(epoch_num): + for iter, batch in enumerate(vox_loader): + #images, log_mels, mel_length = batch + images, log_mels = batch + total_mean = total_mean + torch.mean(log_mels) + total_std = total_std + torch.std(log_mels) + #cur_min_length = torch.min(mel_length) + #if cur_min_length < min_length: + # min_length = cur_min_length + print(epoch) + + avg_mean = (total_mean / epoch_num) / (iter + 1) + avg_std = (total_std / epoch_num) / (iter + 1) + + print('Dataset Mel Spectrogram Mean:', avg_mean) + print('Dataset Mel Spectrogram Standard Deviation:', avg_std) + print('Dataset minimum mel spectrogram length:', min_length) + +if __name__ == '__main__': + mean() diff --git a/mlflow/sf2f/scripts/compute_vggface_score.py b/mlflow/sf2f/scripts/compute_vggface_score.py new file mode 100644 index 0000000..8868c2b --- /dev/null +++ b/mlflow/sf2f/scripts/compute_vggface_score.py @@ -0,0 +1,291 @@ +import argparse +import os +import sys +import torch +import torch.nn as nn +import math +import pickle +import numpy as np +import torch.nn.functional as F +import requests +from requests.adapters import HTTPAdapter + + +__all__ = ['ResNet', 'resnet50'] + +def conv3x3(in_planes, out_planes, stride=1): + """3x3 convolution with padding""" + return nn.Conv2d(in_planes, out_planes, kernel_size=3, stride=stride, + padding=1, bias=False) + + +class BasicBlock(nn.Module): + expansion = 1 + + def __init__(self, inplanes, planes, stride=1, downsample=None): + super(BasicBlock, self).__init__() + self.conv1 = conv3x3(inplanes, planes, stride) + self.bn1 = nn.BatchNorm2d(planes) + self.relu = nn.ReLU(inplace=True) + self.conv2 = conv3x3(planes, planes) + self.bn2 = nn.BatchNorm2d(planes) + self.downsample = downsample + self.stride = stride + + def forward(self, x): + residual = x + + out = self.conv1(x) + out = self.bn1(out) + out = self.relu(out) + + out = self.conv2(out) + out = self.bn2(out) + + if self.downsample is not None: + residual = self.downsample(x) + + out += residual + out = self.relu(out) + + return out + + +class Bottleneck(nn.Module): + expansion = 4 + + def __init__(self, inplanes, planes, stride=1, downsample=None): + super(Bottleneck, self).__init__() + self.conv1 = nn.Conv2d(inplanes, planes, kernel_size=1, stride=stride, bias=False) + self.bn1 = nn.BatchNorm2d(planes) + self.conv2 = nn.Conv2d(planes, planes, kernel_size=3, stride=1, padding=1, bias=False) + self.bn2 = nn.BatchNorm2d(planes) + self.conv3 = nn.Conv2d(planes, planes * 4, kernel_size=1, bias=False) + self.bn3 = nn.BatchNorm2d(planes * 4) + self.relu = nn.ReLU(inplace=True) + self.downsample = downsample + self.stride = stride + + def forward(self, x): + residual = x + + out = self.conv1(x) + out = self.bn1(out) + out = self.relu(out) + + out = self.conv2(out) + out = self.bn2(out) + out = self.relu(out) + + out = self.conv3(out) + out = self.bn3(out) + + if self.downsample is not None: + residual = self.downsample(x) + + out += residual + out = self.relu(out) + + return out + + +class ResNet(nn.Module): + + def __init__(self, block, layers, num_classes=1000, include_top=True): + self.inplanes = 64 + super(ResNet, self).__init__() + self.include_top = include_top + + self.conv1 = nn.Conv2d(3, 64, kernel_size=7, stride=2, padding=3, bias=False) + self.bn1 = nn.BatchNorm2d(64) + self.relu = nn.ReLU(inplace=True) + self.maxpool = nn.MaxPool2d(kernel_size=3, stride=2, padding=0, ceil_mode=True) + + self.layer1 = self._make_layer(block, 64, layers[0]) + self.layer2 = self._make_layer(block, 128, layers[1], stride=2) + self.layer3 = self._make_layer(block, 256, layers[2], stride=2) + self.layer4 = self._make_layer(block, 512, layers[3], stride=2) + self.avgpool = nn.AvgPool2d(7, stride=1) + self.fc = nn.Linear(512 * block.expansion, num_classes) + + for m in self.modules(): + if isinstance(m, nn.Conv2d): + n = m.kernel_size[0] * m.kernel_size[1] * m.out_channels + m.weight.data.normal_(0, math.sqrt(2. / n)) + elif isinstance(m, nn.BatchNorm2d): + m.weight.data.fill_(1) + m.bias.data.zero_() + + def _make_layer(self, block, planes, blocks, stride=1): + downsample = None + if stride != 1 or self.inplanes != planes * block.expansion: + downsample = nn.Sequential( + nn.Conv2d(self.inplanes, planes * block.expansion, + kernel_size=1, stride=stride, bias=False), + nn.BatchNorm2d(planes * block.expansion), + ) + + layers = [] + layers.append(block(self.inplanes, planes, stride, downsample)) + self.inplanes = planes * block.expansion + for i in range(1, blocks): + layers.append(block(self.inplanes, planes)) + + return nn.Sequential(*layers) + + def forward(self, x): + x = self.conv1(x) + x = self.bn1(x) + x = self.relu(x) + x = self.maxpool(x) + + x = self.layer1(x) + x = self.layer2(x) + x = self.layer3(x) + x = self.layer4(x) + + x = self.avgpool(x) + + if not self.include_top: + return x + + x = x.view(x.size(0), -1) + x = self.fc(x) + return x + + def resnet50(**kwargs): + """Constructs a ResNet-50 model. + """ + model = ResNet(Bottleneck, [3, 4, 6, 3], **kwargs) + return model + + +def load_state_dict(model, fname): + """ + Set parameters converted from Caffe models authors of VGGFace2 provide. + See https://www.robots.ox.ac.uk/~vgg/data/vgg_face2/. + Arguments: + model: model + fname: file name of parameters converted from a Caffe model, assuming the file format is Pickle. + """ + with open(fname, 'rb') as f: + weights = pickle.load(f, encoding='latin1') + + own_state = model.state_dict() + for name, param in weights.items(): + if name in own_state: + try: + own_state[name].copy_(torch.from_numpy(param)) + except Exception: + raise RuntimeError('While copying the parameter named {}, whose dimensions in the model are {} and whose '\ + 'dimensions in the checkpoint are {}.'.format(name, own_state[name].size(), param.size())) + else: + raise KeyError('unexpected key "{}" in state_dict'.format(name)) + + +def transform(img): + img = img[:, :, ::-1] # RGB -> BGR + img = img.astype(np.float32) + img -= self.mean_bgr + img = img.transpose(2, 0, 1) # C x H x W + img = torch.from_numpy(img).float() + return img + +def download_weight_for_small_file(weight_file): + ''' + Not working for large files + ''' + weight_url = 'https://drive.google.com/uc?export=download' + \ + '&id=1A94PAAnwk6L7hXdBXLFosB_s0SzEhAFU' + if not os.path.exists(weight_file): + print('Downloading VGGFace resnet50_ft_weight.pkl (1/1)') + s = requests.Session() + s.mount('https://', HTTPAdapter(max_retries=10)) + r = s.get(weight_url, allow_redirects=True) + with open(weight_file, 'wb') as f: + f.write(r.content) + +def download_weight(weight_file): + ''' + Reference: + https://medium.com/@acpanjan/ + download-google-drive-files-using-wget-3c2c025a8b99 + ''' + if not os.path.exists(weight_file): + cmd = "bash scripts/download_vggface_weights.sh" + print(cmd) + os.system(cmd) + +def get_vggface_score(imgs): + N_IDENTITY = 8631 + # include_top = True if args.cmd != 'extract' else False + include_top = True + weight_file = 'scripts/weights/resnet50_ft_weight.pkl' + # mean_bgr = np.array([91.4953, 103.8827, 131.0912]) # from resnet50_ft.prototxt + model = ResNet.resnet50(num_classes=N_IDENTITY, include_top=include_top) + download_weight(weight_file) + load_state_dict(model, weight_file) + model = model.cuda() + model.eval() + mean_rgb = np.array([131.0912, 103.8827, 91.4953]) + # mean_rgb = torch.Tensor([131.0912, 103.8827, 91.4953]).cuda() + preds = [] + for img in imgs: + # img = img.permute(1, 2, 0) - mean_rgb + img = img - mean_rgb + img = torch.Tensor(img).cuda().permute(2, 0, 1) + img = nn.UpsamplingBilinear2d(size=(224, 224))(img.view(1, 3, img.size(1), img.size(2))) + pred = model(img) + pred = F.softmax(pred, dim=1) + preds.append(pred.data.cpu().numpy()) + preds = np.concatenate(preds, 0) + # preds = model(imgs) + + np.random.shuffle(preds) + splits = 5 + scores = [] + for i in range(splits): + part = preds[(i * preds.shape[0] // splits):((i + 1) * preds.shape[0] // splits), :] + kl = part * (np.log(part) - np.log(np.expand_dims(np.mean(part, 0), 0))) + kl = np.mean(np.sum(kl, 1)) + scores.append(np.exp(kl)) + + return np.mean(scores), np.std(scores) + + +def get_vggface_act(imgs): + N_IDENTITY = 8631 + # include_top = True if args.cmd != 'extract' else False + include_top = True + weight_file = 'scripts/weights/resnet50_ft_weight.pkl' + # mean_bgr = np.array([91.4953, 103.8827, 131.0912]) # from resnet50_ft.prototxt + model = ResNet.resnet50(num_classes=N_IDENTITY, include_top=include_top) + load_state_dict(model, weight_file) + model = model.cuda() + model.eval() + mean_rgb = np.array([131.0912, 103.8827, 91.4953]) + # mean_rgb = torch.Tensor([131.0912, 103.8827, 91.4953]).cuda() + preds = [] + for img in imgs: + # img = img.permute(1, 2, 0) - mean_rgb + img = img - mean_rgb + img = torch.Tensor(img).cuda().permute(2, 0, 1) + img = nn.UpsamplingBilinear2d(size=(224, 224))(img.view(1, 3, img.size(1), img.size(2))) + pred = model(img) + pred = F.softmax(pred, dim=1) + preds.append(pred.data.cpu().numpy()) + preds = np.concatenate(preds, 0) + return preds + + +def main(): + N_IDENTITY = 8631 + # include_top = True if args.cmd != 'extract' else False + include_top = False + weight_file = 'scripts/weights/resnet50_ft_weights.pkl' + model = ResNet.resnet50(num_classes=N_IDENTITY, include_top=include_top) + load_state_dict(model, weight_file) + model = model.cuda() + +if __name__ == '__main__': + main() diff --git a/mlflow/sf2f/scripts/convert_wav_to_mel.py b/mlflow/sf2f/scripts/convert_wav_to_mel.py new file mode 100644 index 0000000..ff334fa --- /dev/null +++ b/mlflow/sf2f/scripts/convert_wav_to_mel.py @@ -0,0 +1,216 @@ +''' +Convert raw_wavs in VoxCeleb dataSet to mel_gram + + 1. Build two id2name mapping for vox1 & vox2 + 2. Create a folder for each identity with his/her name + 3. for each raw wav file, convert it to mel_gram, + save it under the identity folder +''' + + +import warnings +warnings.filterwarnings("ignore") +import argparse +import os +import shutil +import gc +import time +import pickle +import numpy as np +import pandas as pd +import multiprocessing as mp +from tensorflow.io import gfile +import sys +sys.path.append('./') +from utils.wav2mel import wav_to_mel +from concurrent.futures import ProcessPoolExecutor, as_completed +from utils.filter_pickle import filtering_pickle + + +VOX_DIR = os.path.join('/home/data/VoxCeleb') +vox1_raw = os.path.join(VOX_DIR, 'raw_wav', 'vox1') +# vox2_raw = os.path.join(VOX_DIR, 'raw_wav', 'vox2') +vox1_meta_csv = os.path.join(VOX_DIR, 'vox1', 'vox1_meta.csv') +# vox2_meta_csv = os.path.join(VOX_DIR, 'vox2', 'full_vox2_meta.csv') + +parser = argparse.ArgumentParser() + +# arguments: +#parser.add_argument('--output_folder', +# default=os.path.join(PROC_DIR, 'VoxCeleb1', 'faces'), +# help='folder that contains the outputs') + +args = parser.parse_args() + +class WavConvertor: + def __init__(self): + self.load_metadata() + self.get_id2name() + self.get_wav_dirs() + self.create_output_dirs() + + def load_metadata(self): + self.vox1_df = pd.read_csv(vox1_meta_csv, sep='\t') + # self.vox2_df = pd.read_csv(vox2_meta_csv, sep=',') + #print(self.vox1_df.head()) + #print(self.vox1_df.columns) + #print(self.vox2_df.head()) + #print(self.vox2_df.columns) + + def get_id2name(self): + self.vox1_id2name = dict( + zip(self.vox1_df['VoxCeleb1 ID'], self.vox1_df['VGGFace1 ID'])) + # self.vox2_id2name = dict( + # zip(self.vox2_df['VoxCeleb2ID'], self.vox2_df['Name'])) + #print((self.vox2_id2name)) + + def convert_identity(self, wav_dir, mel_home_dir, dataset): + ''' + Create a mel_gram folder for the given identity + + Arguments: + 1. wav_dir (str): path to the identity's raw_wav folder + 2. mel_home_dir (str): path to the identity's raw_wav folder + 3. dataset (str): 'vox1' or 'vox2' + ''' + spkid = wav_dir.split('/')[-1] + # In case the input path ends with '/' + if spkid == '': + spkid = wav_dir.split('/')[-2] + if dataset == 'vox1': + name = self.vox1_id2name[spkid] + # elif dataset == 'vox2': + # name = self.vox2_id2name[spkid] + else: + raise ValueError("Invalid dataset argument") + #print(name) + + # Create mel_gram directory of the speaker + mel_dir = os.path.join(mel_home_dir, name) + gfile.mkdir(mel_dir) + + clipids = os.listdir(wav_dir) + for clipid in clipids: + clip_dir = os.path.join(wav_dir, clipid) + wav_files = os.listdir(clip_dir) + for wav_file in wav_files: + # Read and process the wav + wav_path = os.path.join(clip_dir, wav_file) + try: + wavid = wav_file.replace('.wav', '').replace('.m4a', '') + pickle_name = "{}_{}_{}.pickle".format( + spkid, clipid, wavid) + pickle_path = os.path.join( + mel_dir, pickle_name) + if os.path.exists(pickle_path): + # Skip if exists + continue + # Vox1 use .wav format, Vox2 use .m4a format + log_mel = wav_to_mel(wav_path) + pickle_dict = { + 'LogMel_Features': log_mel, + 'spkid': spkid, + 'clipid': clipid, + 'wavid': wavid + } + pickle.dump( + pickle_dict, + open(pickle_path, "wb") + ) + except IndexError: + pass + gc.collect() + + def get_wav_dirs(self): + ''' + Generate a list containing paths to the wav_dir of all the speakers + ''' + # The original VoxCeleb dataset comes with test set and dev set + vox1_dev = os.path.join(vox1_raw, 'dev') + vox1_test = os.path.join(vox1_raw, 'test') + # vox2_dev = os.path.join(vox2_raw, 'dev') + # vox2_test = os.path.join(vox2_raw, 'test') + # Get Vox1 wav_dir paths + vox1_wav_dirs = [] + for wav_dir in os.listdir(vox1_dev): + vox1_wav_dirs.append(os.path.join(vox1_dev, wav_dir)) + for wav_dir in os.listdir(vox1_test): + vox1_wav_dirs.append(os.path.join(vox1_test, wav_dir)) + # Get Vox2 wav_dir paths + # vox2_wav_dirs = [] + # for wav_dir in os.listdir(vox2_dev): + # vox2_wav_dirs.append(os.path.join(vox2_dev, wav_dir)) + # for wav_dir in os.listdir(vox2_test): + # vox2_wav_dirs.append(os.path.join(vox2_test, wav_dir)) + self.vox1_wav_dirs = vox1_wav_dirs + # self.vox2_wav_dirs = vox2_wav_dirs + return 0 + + def create_output_dirs(self): + self.vox1_mel = os.path.join(VOX_DIR, 'vox1', 'mel_spectrograms') + # self.vox2_mel = os.path.join(VOX_DIR, 'vox2', 'mel_spectrograms') + if not gfile.exists(self.vox1_mel): + gfile.mkdir(self.vox1_mel) + # gfile.mkdir(self.vox2_mel) + + def _worker(self, job_id, infos): + for i, info in enumerate(infos): + self.convert_identity(info[0], info[1], info[2]) + print('job #{} prcess {} {} done'.format(job_id, i, info[0])) + shutil.rmtree(info[0]) # remove complete file + + def convert_wav_to_mel(self, n_jobs=1): + + infos = [] + for wav_dir in self.vox1_wav_dirs: + infos.append((wav_dir, self.vox1_mel, 'vox1')) + print(len(infos)) + # l1 = len(infos) + # for wav_dir in self.vox2_wav_dirs: + # infos.append((wav_dir, self.vox2_mel, 'vox2')) + # print(len(infos) - l1) + + n_wav_dirs = len(infos) + n_jobs = n_jobs if n_jobs <= n_wav_dirs else n_wav_dirs + n_wav_dirs_per_job = n_wav_dirs // n_jobs + process_index = [] + for ii in range(n_jobs): + process_index.append([ii*n_wav_dirs_per_job, (ii+1)*n_wav_dirs_per_job]) + if n_jobs * n_wav_dirs_per_job != n_wav_dirs: + process_index[-1][-1] = n_wav_dirs + + futures = set() + with ProcessPoolExecutor() as executor: + for job_id in range(n_jobs): + # future = executor.submit(_worker, process_index[job_id][0], process_index[job_id][1]) + future = executor.submit(self._worker, job_id, infos[process_index[job_id][0]:process_index[job_id][1]]) + futures.add(future) + print('submit job {}, {}-{}'.format(job_id, process_index[job_id][0], process_index[job_id][1])) + for future in as_completed(futures): + pass + + print("Done.") + + + +def main(): + + import argparse + parser = argparse.ArgumentParser() + parser.add_argument('--n_jobs', '-n_jobs', type=int, default=5) + args = parser.parse_args() + + wav_convertor = WavConvertor() + wav_convertor.convert_wav_to_mel(args.n_jobs) + #gfile.mkdir('./data/test') + #wav_convertor.convert_identity( + # 'data/VoxCeleb/raw_wav/vox1/dev/id10001/', + # './data/test', 'vox1') + + dir_path = './sf2f/data/VoxCeleb/vox1/mel_spectrograms' # wav_convertor.vox1_mel + filtering_pickle(dir_path) + + +if __name__ == '__main__': + print(os.listdir(VOX_DIR)) + main() diff --git a/mlflow/sf2f/scripts/create_split_json.py b/mlflow/sf2f/scripts/create_split_json.py new file mode 100644 index 0000000..3bb5b8f --- /dev/null +++ b/mlflow/sf2f/scripts/create_split_json.py @@ -0,0 +1,88 @@ +''' +This script generates a train/test/val split json for VoxCeleb dataset + +vox1_meta.csv could be download from VoxCeleb official website: + https://www.robots.ox.ac.uk/~vgg/data/voxceleb/meta/vox1_meta.csv +''' + + +import os +import json +import pandas as pd + +VOX_DIR = os.path.join('/home/data/VoxCeleb') +vox1_meta_csv = os.path.join(VOX_DIR, 'vox1', 'vox1_meta.csv') +# vox2_meta_csv = os.path.join(VOX_DIR, 'vox2', 'full_vox2_meta.csv') +# output: +split_json = os.path.join(VOX_DIR, 'split.json') + + +def main(): + vox1_df = pd.read_csv(vox1_meta_csv, sep='\t') + # vox2_df = pd.read_csv(vox2_meta_csv, sep=',') + + print(vox1_df.head()) + # print(vox2_df.head()) + + split_dict = { + 'vox1': {'train':[], 'val':[], 'test':[]}, + # 'vox2': {'train':[], 'val':[], 'test':[]} + } + + for i in range(len(vox1_df)): + name = vox1_df['VGGFace1 ID'].iloc[i] + if i % 10 == 8: + split_dict['vox1']['test'].append(name) + elif i % 10 == 9: + split_dict['vox1']['val'].append(name) + else: + split_dict['vox1']['train'].append(name) + + # for i in range(len(vox2_df)): + # name = vox2_df['Name'].iloc[i] + # if i % 10 == 8: + # split_dict['vox2']['test'].append(name) + # elif i % 10 == 9: + # split_dict['vox2']['val'].append(name) + # else: + # split_dict['vox2']['train'].append(name) + + ''' + for i in range(len(vox1_df)): + name = vox1_df['VGGFace1 ID'].iloc[i] + set = vox1_df['Set'].iloc[i] + if set == 'test': + split_dict['vox1']['test'].append(name) + if set == 'dev': + if i % 10 == 9: + split_dict['vox1']['val'].append(name) + else: + split_dict['vox1']['train'].append(name) + + for i in range(len(vox2_df)): + name = vox2_df['Name'].iloc[i] + set = vox2_df['Set'].iloc[i] + if set == 'test': + split_dict['vox2']['test'].append(name) + if set == 'dev': + if i % 10 == 9: + split_dict['vox2']['val'].append(name) + else: + split_dict['vox2']['train'].append(name) + ''' + + print(len(split_dict['vox1']['train']), + len(split_dict['vox1']['val']), + len(split_dict['vox1']['test'])) + # print(len(split_dict['vox2']['train']), + # len(split_dict['vox2']['val']), + # len(split_dict['vox2']['test'])) + + with open(split_json, 'w') as outfile: + json.dump(split_dict, outfile) + + return 0 + + +if __name__ == '__main__': + main() diff --git a/mlflow/sf2f/scripts/download_vggface_weights.sh b/mlflow/sf2f/scripts/download_vggface_weights.sh new file mode 100644 index 0000000..2d59ada --- /dev/null +++ b/mlflow/sf2f/scripts/download_vggface_weights.sh @@ -0,0 +1,4 @@ +mkdir './scripts/weights/' +wget --load-cookies /tmp/cookies.txt \ + "https://docs.google.com/uc?export=download&confirm=$(wget --quiet --save-cookies /tmp/cookies.txt --keep-session-cookies --no-check-certificate 'https://docs.google.com/uc?export=download&id=1A94PAAnwk6L7hXdBXLFosB_s0SzEhAFU' -O- | sed -rn 's/.*confirm=([0-9A-Za-z_]+).*/\1\n/p')&id=1A94PAAnwk6L7hXdBXLFosB_s0SzEhAFU" \ + -O 'scripts/weights/resnet50_ft_weight.pkl' && rm -rf /tmp/cookies.txt \ No newline at end of file diff --git a/mlflow/sf2f/scripts/install_requirements.py b/mlflow/sf2f/scripts/install_requirements.py new file mode 100644 index 0000000..b531402 --- /dev/null +++ b/mlflow/sf2f/scripts/install_requirements.py @@ -0,0 +1,50 @@ +import os +from multiprocessing import Process + + +def pip_install(package): + cmd = 'pip install {}'.format(package) + print(cmd) + try: + os.system(cmd) + return 0 + except: + print('###########################') + print('Exception') + print('###########################') + pip_install(package) + + +def txt2list(txt_path): + ''' + Read a txt file and return a list + Arguments: + txt_path (str): path to save the text file + ''' + output = [] + f = open(txt_path, 'r') + for line in f.readlines(): + output.append(line.replace('\n', '')) + f.close() + print('{} loaded.'.format(txt_path)) + return output + + +def install_once(): + packages = txt2list('./requirements.txt') + procs = [] + + # instantiating process with arguments + for package in packages: + proc = Process(target=pip_install, args=(package,)) + procs.append(proc) + proc.start() + + # complete the processes + for proc in procs: + proc.join() + + +if __name__ == "__main__": # confirms that the code is under main function + for i in range(100): + install_once() diff --git a/mlflow/sf2f/scripts/print_args.py b/mlflow/sf2f/scripts/print_args.py new file mode 100644 index 0000000..5ccaa26 --- /dev/null +++ b/mlflow/sf2f/scripts/print_args.py @@ -0,0 +1,22 @@ +""" +Tiny utility to print the command-line args used for a checkpoint +""" + + +import argparse +import torch + + +parser = argparse.ArgumentParser() +parser.add_argument('checkpoint') + + +def main(args): + checkpoint = torch.load(args.checkpoint, map_location='cpu') + for k, v in checkpoint['args'].items(): + print(k, v) + + +if __name__ == '__main__': + args = parser.parse_args() + main(args) diff --git a/mlflow/sf2f/scripts/sample_mel_spectrograms.py b/mlflow/sf2f/scripts/sample_mel_spectrograms.py new file mode 100644 index 0000000..deb63a8 --- /dev/null +++ b/mlflow/sf2f/scripts/sample_mel_spectrograms.py @@ -0,0 +1,52 @@ +''' +This file samples several mel spectrograms from the VoxCeleb Dataset, and + visualize them as png images +''' + + +import os +import PIL + +import sys +sys.path.append('./') +#print(sys.path) +from datasets import VoxDataset, fast_mel_deprocess_batch +from utils.visualization.plot import plot_mel_spectrogram #get_np_plot, +from tensorflow.io.gfile import mkdir + + +VOX_DIR = os.path.join('./data', 'VoxCeleb') +output_dir = os.path.join('./output', 'sampled_mel_spectrograms') +mkdir(output_dir) + + +def main(): + vox_dataset = VoxDataset( + data_dir=VOX_DIR, + image_size=(64, 64), + nframe_range=(300, 600), + face_type='masked', + image_normalize_method='imagenet', + mel_normalize_method='vox_mel', + mel_seg_window_stride=(125, 63), + split_set='test', + split_json=os.path.join(VOX_DIR, 'split.json')) + + log_mels = vox_dataset.get_all_mel_segments_of_id(5) + + log_mels_de = fast_mel_deprocess_batch(log_mels, 'vox_mel') + + for i in range(len(log_mels_de)): + log_mel = log_mels_de[i].cpu().detach().numpy() + buf = plot_mel_spectrogram( + log_mel, + colorbar=False, + label=False, + coordinate=False, + remove_boarder=True) + log_mel_img = PIL.Image.open(buf) + log_mel_img.save(os.path.join(output_dir, '{}.png'.format(i))) + + +if __name__ == '__main__': + main() diff --git a/mlflow/sf2f/scripts/strip_checkpoint.py b/mlflow/sf2f/scripts/strip_checkpoint.py new file mode 100644 index 0000000..f91831b --- /dev/null +++ b/mlflow/sf2f/scripts/strip_checkpoint.py @@ -0,0 +1,53 @@ +""" +Checkpoints saved by train.py contain not only model parameters but also +optimizer states, losses, a history of generated images, and other statistics. +This information is very useful for development and debugging models, but makes +the saved checkpoints very large. This utility script strips away all extra +information from saved checkpoints, keeping only the saved models. +""" + + +import argparse +import os +import torch + + +parser = argparse.ArgumentParser() +parser.add_argument('--input_checkpoint', default=None) +parser.add_argument('--output_checkpoint', default=None) +parser.add_argument('--input_dir', default=None) +parser.add_argument('--output_dir', default=None) +parser.add_argument('--keep_discriminators', type=int, default=1) + + +def main(args): + if args.input_checkpoint is not None: + handle_checkpoint(args, args.input_checkpoint, args.output_checkpoint) + if args.input_dir is not None: + handle_dir(args, args.input_dir, args.output_dir) + + +def handle_dir(args, input_dir, output_dir): + for fn in os.listdir(input_dir): + if not fn.endswith('.pt'): + continue + input_path = os.path.join(input_dir, fn) + output_path = os.path.join(output_dir, fn) + handle_checkpoint(args, input_path, output_path) + + +def handle_checkpoint(args, input_path, output_path): + input_checkpoint = torch.load(input_path) + keep = ['args', 'model_state', 'model_kwargs'] + if args.keep_discriminators == 1: + keep += ['d_img_state', 'd_img_kwargs', 'd_obj_state', 'd_obj_kwargs'] + output_checkpoint = {} + for k, v in input_checkpoint.items(): + if k in keep: + output_checkpoint[k] = v + torch.save(output_checkpoint, output_path) + + +if __name__ == '__main__': + args = parser.parse_args() + main(args) diff --git a/mlflow/sf2f/scripts/strip_old_args.py b/mlflow/sf2f/scripts/strip_old_args.py new file mode 100644 index 0000000..578cece --- /dev/null +++ b/mlflow/sf2f/scripts/strip_old_args.py @@ -0,0 +1,55 @@ +""" +This utility script removes deprecated kwargs in checkpoints. +""" + + +import argparse +import os +import torch + + +parser = argparse.ArgumentParser() +parser.add_argument('--input_checkpoint', default=None) +parser.add_argument('--input_dir', default=None) + + +DEPRECATED_KWARGS = { + 'model_kwargs': [ + 'vec_noise_dim', 'gconv_mode', 'box_anchor', 'decouple_obj_predictions', + ], +} + + +def main(args): + got_checkpoint = (args.input_checkpoint is not None) + got_dir = (args.input_dir is not None) + assert got_checkpoint != got_dir, "Must give exactly one of checkpoint or dir" + if got_checkpoint: + handle_checkpoint(args.input_checkpoint) + elif got_dir: + handle_dir(args.input_dir) + + +def handle_dir(dir_path): + for fn in os.listdir(dir_path): + if not fn.endswith('.pt'): + continue + checkpoint_path = os.path.join(dir_path, fn) + handle_checkpoint(checkpoint_path) + + +def handle_checkpoint(checkpoint_path): + print('Stripping old args from checkpoint "%s"' % checkpoint_path) + checkpoint = torch.load(checkpoint_path) + for group, deprecated in DEPRECATED_KWARGS.items(): + assert group in checkpoint + for k in deprecated: + if k in checkpoint[group]: + print('Removing key "%s" from "%s"' % (k, group)) + del checkpoint[group][k] + torch.save(checkpoint, checkpoint_path) + + +if __name__ == '__main__': + args = parser.parse_args() + main(args) diff --git a/mlflow/sf2f/scripts/watch_data.py b/mlflow/sf2f/scripts/watch_data.py new file mode 100644 index 0000000..8cd319c --- /dev/null +++ b/mlflow/sf2f/scripts/watch_data.py @@ -0,0 +1,111 @@ +import os +import numpy as np +import random +import pickle +import json +import PIL +from PIL import Image + +DIR = './data/VoxCeleb/' + + +def main(): + dir_1 = os.path.join(DIR, 'vox1/masked_faces') + names = os.listdir(dir_1) + print("Length of names: ", len(names)) # 1187 + img_num_1 = 0 + all_W1, all_H1 = 0, 0 + num1_50, num1_100, num1_200, num1_300, num1_400, num1_600 = 0, 0, 0, 0, 0, 0 + for i, name in enumerate(names): + path = os.path.join(dir_1, name) + images = os.listdir(path) + for j, image in enumerate(images): + image_path = os.path.join(path, image) + with open(image_path, 'rb') as f: + with PIL.Image.open(f) as image: + WW, HH = image.size + img_num_1 += 1 + all_W1 += WW + all_H1 += HH + # print("PIL Image:", np.array(image)) + if (WW >= 50 and WW < 100) or (HH >= 50 and HH < 100): + num1_50 += 1 + elif (WW >= 100 and WW < 200) or (HH >= 100 and HH < 200): + num1_100 += 1 + elif (WW >= 200 and WW < 300) or (HH >= 200 and HH < 300): + num1_200 += 1 + elif (WW >= 300 and WW < 400) or (HH >= 300 and HH < 400): + num1_300 += 1 + elif (WW >= 400 and WW < 600) or (HH >= 400 and HH < 600): + num1_400 += 1 + elif (WW >= 600) or (HH >= 600): + num1_600 += 1 + + print("======== Vox1 =========") + print("Number of vox1: ", img_num_1) + print("Average W: ", all_W1/img_num_1) + print("Average H: ", all_H1/img_num_1) + print("50~100: ", num1_50) + print("100~200: ", num1_100) + print("200~300: ", num1_200) + print("300~400: ", num1_300) + print("400~600: ", num1_400) + print("600+: ", num1_600) + print("\n") + + dir_2 = os.path.join(DIR, 'vox2/masked_faces') + names = os.listdir(dir_2) + print("Length of names: ", len(names)) # 2850 + img_num_2 = 0 + all_W2, all_H2 = 0, 0 + num2_50, num2_100, num2_200, num2_300, num2_400, num2_600 = 0, 0, 0, 0, 0, 0 + for i, name in enumerate(names): + path = os.path.join(dir_2, name) + images = os.listdir(path) + for j, image in enumerate(images): + image_path = os.path.join(path, image) + with open(image_path, 'rb') as f: + with PIL.Image.open(f) as image: + WW, HH = image.size + img_num_2 += 1 + all_W2 += WW + all_H2 += HH + # print("PIL Image:", np.array(image)) + if (WW >= 50 and WW < 100) or (HH >= 50 and HH < 100): + num2_50 += 1 + elif (WW >= 100 and WW < 200) or (HH >= 100 and HH < 200): + num2_100 += 1 + elif (WW >= 200 and WW < 300) or (HH >= 200 and HH < 300): + num2_200 += 1 + elif (WW >= 300 and WW < 400) or (HH >= 300 and HH < 400): + num2_300 += 1 + elif (WW >= 400 and WW < 600) or (HH >= 400 and HH < 600): + num2_400 += 1 + elif (WW >= 600) or (HH >= 600): + num2_600 += 1 + + print("======== Vox2 =========") + print("Number of vox2: ", img_num_2) + print("Average W: ", all_W2/img_num_2) + print("Average H: ", all_H2/img_num_2) + print("50~100: ", num2_50) + print("100~200: ", num2_100) + print("200~300: ", num2_200) + print("300~400: ", num2_300) + print("400~600: ", num2_400) + print("600+: ", num2_600) + print("\n") + + print("======== ALL DATASET ========") + print("Number of images: ", img_num_1 + img_num_2) + print("Average W: ", (all_W1 + all_W2) / (img_num_1 + img_num_2)) + print("Average H: ", (all_H1 + all_H2) / (img_num_1 + img_num_2)) + print("50~100: ", num1_50 + num2_50) + print("100~200: ", num1_100 + num2_100) + print("200~300: ", num1_200 + num2_200) + print("300~400: ", num1_300 + num2_300) + print("400~600: ", num1_400 + num2_400) + print("600+: ", num1_600 + num2_600) + +if __name__ == '__main__': + main() diff --git a/mlflow/sf2f/train.py b/mlflow/sf2f/train.py new file mode 100644 index 0000000..ab8304e --- /dev/null +++ b/mlflow/sf2f/train.py @@ -0,0 +1,854 @@ +import functools +import os +import json +import math +from collections import defaultdict +import random +import time +import pyprind +import glog as log +from shutil import copyfile + +import numpy as np +import torch +import torch.optim as optim +import torch.nn as nn +import torch.nn.functional as F +from torch.utils.data.dataloader import default_collate +import mlflow + +from datasets import imagenet_deprocess_batch +import datasets +import models +import models.perceptual +from utils.losses import get_gan_losses +from utils import timeit, LossManager +from options.opts import args, options +from utils.logger import Logger +from utils import tensor2im +from utils.utils import load_my_state_dict +# losseds need to be modified +from utils.training_utils import add_loss, check_model, calculate_model_losses +from utils.training_utils import visualize_sample +from utils.evaluate import evaluate +from utils.evaluate_fid import evaluate_fid +from utils.s2f_evaluator import S2fEvaluator +torch.backends.cudnn.benchmark = True + + +def main(): + global args, options + print(args) + print(options['data']) + # Set random seed, deterministic + torch.cuda.manual_seed(args.seed) + torch.manual_seed(args.seed) + np.random.seed(args.seed) + random.seed(args.seed) + torch.backends.cudnn.deterministic = True + torch.backends.cudnn.benchmark = False + #visualize_attn=options['eval'].get('visualize_attn', False) + #print('########### visualize_attn: {} ###########'.format(visualize_attn)) + float_dtype = torch.cuda.FloatTensor + long_dtype = torch.cuda.LongTensor + log.info("Building loader...") + train_loader, val_loader, test_loader = \ + datasets.build_loaders(options["data"]) + # Fuser Logic + if args.train_fuser_only: + train_loader.collate_fn = default_collate + train_loader.dataset.return_mel_segments = True + train_loader.dataset.mel_segments_rand_start = True + val_loader.collate_fn = default_collate + val_loader.dataset.return_mel_segments = True + s2f_face_gen_mode = 'naive' + else: + s2f_face_gen_mode = 'average_facenet_embedding' + # End of fuser logic + #s2f_val_evaluator = S2fEvaluator(val_loader, options) + s2f_val_evaluator = S2fEvaluator( + val_loader, + options, + extraction_size=100, #[100,200,300], + hq_emb_dict=True, + face_gen_mode=s2f_face_gen_mode) + + # [:len(checkpoint[term_name]['id_classifier.bias'])] class 맞추기 + # if restore_path is not None and os.path.isfile(restore_path): + train_loader.dataset.available_names = train_loader.dataset.available_names[:729] + + # Get total number of people in training set + num_train_id = len(train_loader.dataset) + # Initialize softmax neuron value + for ac_net in ['identity', 'identity_low', 'identity_mid', 'identity_high']: + if options['discriminator'].get(ac_net) is not None: + options['discriminator'][ac_net]['num_id'] = num_train_id + + log.info("Building Generative Model...") + model, model_kwargs = models.build_model( + options["generator"], + image_size=options["data"]["image_size"], + checkpoint_start_from=args.checkpoint_start_from) + model.type(float_dtype) + # Fuser logic + if args.train_fuser_only: + # Hardcode to freeze batchnorm layers + model.eval() + if args.train_fuser_decoder: + model.encoder.train_fuser_only() + else: + model.train_fuser_only() + # End of fuser logic + print(model) + + optimizer = torch.optim.Adam( + filter(lambda x: x.requires_grad, model.parameters()), + lr=args.learning_rate, + betas=(args.beta1, 0.999),) + + if (options["optim"]["d_loss_weight"] < 0 or \ + options["optim"]["d_img_weight"] < 0): + # Ignore image discriminator + img_discriminator = None + d_img_kwargs = {} + log.info("Ignoring Image Discriminator.") + else: + img_discriminator, d_img_kwargs = models.build_img_discriminator( + options["discriminator"]) + log.info("Done Building Image Discriminator.") + + if (options["optim"]["d_loss_weight"] < 0 or \ + options["optim"]["ac_loss_weight"] < 0): + # Ignore AC discriminator + ac_discriminator = None + ac_img_kwargs = {} + log.info("Ignoring Auxilary Classifier Discriminator.") + else: + ac_discriminator, ac_img_kwargs = models.build_ac_discriminator( + options["discriminator"]) + #print('AC Discriminator:', ac_discriminator) + log.info("Done Building Auxilary Classifier Discriminator.") + + if (options["optim"]["d_loss_weight"] < 0 or \ + options["optim"].get("cond_loss_weight", -1) < 0): + # Ignore Conditional discriminator + cond_discriminator = None + cond_d_kwargs = {} + log.info("Ignoring Conditional Discriminator.") + else: + cond_discriminator, cond_d_kwargs = models.build_cond_discriminator( + options["discriminator"]) + #print('Conditional Discriminator:', cond_discriminator) + log.info("Done Building Conditional Discriminator.") + + perceptual_module = None + if options["optim"].get("perceptual_loss_weight", -1) > 0: + ploss_name = options.get("perceptual", {}).get("arch", "FaceNetLoss") + ploss_cos_weight = options["optim"].get("cos_percept_loss_weight", -1) + perceptual_module = getattr( + models.perceptual, + ploss_name)(cos_loss_weight=ploss_cos_weight) + log.info("Done Building Perceptual {} Module.".format(ploss_name)) + if ploss_cos_weight > 0: + log.info("Perceptual Cos Loss Weight: {}".format(ploss_cos_weight)) + else: + log.info("Ignoring Perceptual Module.") + + gan_g_loss, gan_d_loss = get_gan_losses(options["optim"]["gan_loss_type"]) + + optimizer_d_img = [] + if img_discriminator is not None: + for i in range(len(img_discriminator)): + img_discriminator[i].type(float_dtype) + img_discriminator[i].train() + print(img_discriminator[i]) + optimizer_d_img.append(torch.optim.Adam( + filter(lambda x: x.requires_grad, + img_discriminator[i].parameters()), + lr=args.learning_rate, + betas=(args.beta1, 0.999),)) + + optimizer_d_ac = [] + if ac_discriminator is not None: + for i in range(len(ac_discriminator)): + ac_discriminator[i].type(float_dtype) + ac_discriminator[i].train() + print(ac_discriminator[i]) + optimizer_d_ac.append(torch.optim.Adam( + filter(lambda x: x.requires_grad, + ac_discriminator[i].parameters()), + lr=args.learning_rate, + betas=(args.beta1, 0.999),)) + + optimizer_cond_d = [] + if cond_discriminator is not None: + for i in range(len(cond_discriminator)): + cond_discriminator[i].type(float_dtype) + cond_discriminator[i].train() + print(cond_discriminator[i]) + optimizer_cond_d.append(torch.optim.Adam( + filter(lambda x: x.requires_grad, + cond_discriminator[i].parameters()), + lr=args.learning_rate, + betas=(args.beta1, 0.999),)) + + restore_path = None + if args.resume is not None: + restore_path = '%s_with_model.pt' % args.checkpoint_name + restore_path = os.path.join( + options["logs"]["output_dir"], args.resume, restore_path) + + if restore_path is not None and os.path.isfile(restore_path): + log.info('Restoring from checkpoint: {}'.format(restore_path)) + checkpoint = torch.load(restore_path) + model.load_state_dict(checkpoint['model_state']) + optimizer.load_state_dict(checkpoint['optim_state']) + + if img_discriminator is not None: + for i in range(len(img_discriminator)): + term_name = 'd_img_state_%d' % i + img_discriminator[i].load_state_dict(checkpoint[term_name]) + term_name = 'd_img_optim_state_%d' % i + optimizer_d_img[i].load_state_dict(checkpoint[term_name]) + + if ac_discriminator is not None: + for i in range(len(ac_discriminator)): + term_name = 'd_ac_state_%d' % i + ac_discriminator[i].load_state_dict(checkpoint[term_name]) + term_name = 'd_ac_optim_state_%d' % i + optimizer_d_ac[i].load_state_dict(checkpoint[term_name]) + + if cond_discriminator is not None: + for i in range(len(cond_discriminator)): + term_name = 'd_img_state_%d' % i + cond_discriminator[i].load_state_dict(checkpoint[term_name]) + term_name = 'd_img_optim_state_%d' % i + optimizer_cond_d[i].load_state_dict(checkpoint[term_name]) + + t = checkpoint['counters']['t'] + 1 + if 0 <= args.eval_mode_after <= t: + model.eval() + else: + model.train() + # Reset epoch here later + start_epoch = checkpoint['counters']['epoch'] + 1 + log_path = os.path.join(options["logs"]["output_dir"], args.resume,) + lr = checkpoint.get('learning_rate', args.learning_rate) + best_inception = checkpoint["counters"].get("best_inception", (0., 0.)) + best_vfs = checkpoint["counters"].get("best_vfs", (0., 0.)) + best_recall_1 = checkpoint["counters"].get("best_recall_1", 0.) + best_recall_5 = checkpoint["counters"].get("best_recall_5", 0.) + best_recall_10 = checkpoint["counters"].get("best_recall_10", 0.) + best_cos = checkpoint["counters"].get("best_cos", 0.) + best_L1 = checkpoint["counters"].get("best_L1", 100000.0) + options = checkpoint.get("options", options) + else: + t, start_epoch, best_inception, best_vfs = 0, 0, (0., 0.), (0., 0.) + best_recall_1, best_recall_5, best_recall_10 = 0.0, 0.0, 0.0 + best_cos, best_L1 = 0.0, 100000.0 + lr = args.learning_rate + checkpoint = { + 'args': args.__dict__, + 'options': options, + 'model_kwargs': model_kwargs, + 'd_img_kwargs': d_img_kwargs, + 'train_losses': defaultdict(list), + 'checkpoint_ts': [], + 'train_batch_data': [], + 'train_samples': [], + 'train_iou': [], + 'train_inception': [], + 'lr': [], + 'val_batch_data': [], + 'val_samples': [], + 'val_losses': defaultdict(list), + 'val_iou': [], + 'val_inception': [], + 'norm_d': [], + 'norm_g': [], + 'counters': { + 't': None, + 'epoch': None, + 'best_inception': None, + 'best_vfs': None, + 'best_recall_1': None, + 'best_recall_5': None, + 'best_recall_10': None, + 'best_cos': None, + 'best_L1': None, + }, + 'model_state': None, + 'model_best_state': None, + 'optim_state': None, + 'd_img_state': None, + 'd_img_best_state': None, + 'd_img_optim_state': None, + 'd_ac_state': None, + 'd_ac_optim_state': None, + } + + log_path = os.path.join( + options["logs"]["output_dir"], + options["logs"]["name"] + "-" + time.strftime("%Y%m%d-%H%M%S") + ) + + ### Fuser Logic + if args.pretrained_path is not None and \ + os.path.isfile(args.pretrained_path): + # Load + log.info('Loading Pretrained Model: {}'.format(args.pretrained_path)) + pre_checkpoint = torch.load(args.pretrained_path) + #model.load_state_dict(pre_checkpoint['model_state']) + load_my_state_dict(model, pre_checkpoint['model_state']) + #optimizer.load_state_dict(pre_checkpoint['optim_state']) + + if img_discriminator is not None: + for i in range(len(img_discriminator)): + term_name = 'd_img_state_%d' % i + img_discriminator[i].load_state_dict(pre_checkpoint[term_name]) + #term_name = 'd_img_optim_state_%d' % i + #optimizer_d_img[i].load_state_dict(pre_checkpoint[term_name]) + + if ac_discriminator is not None: + for i in range(len(ac_discriminator)): + term_name = 'd_ac_state_%d' % i + ac_discriminator[i].load_state_dict(pre_checkpoint[term_name]) + #term_name = 'd_ac_optim_state_%d' % i + #optimizer_d_ac[i].load_state_dict(pre_checkpoint[term_name]) + ### End of fuser logic + logger = Logger(log_path) + log.info("Logging to: {}".format(log_path)) + # save the current config yaml + copyfile(args.path_opts, + os.path.join(log_path, options["logs"]["name"] + '.yaml')) + + model = nn.DataParallel(model.cuda()) + if ac_discriminator is not None: + for i in range(len(ac_discriminator)): + ac_discriminator[i] = nn.DataParallel(ac_discriminator[i].cuda()) + if img_discriminator is not None: + for i in range(len(img_discriminator)): + img_discriminator[i] = nn.DataParallel(img_discriminator[i].cuda()) + if cond_discriminator is not None: + for i in range(len(cond_discriminator)): + cond_discriminator[i] = nn.DataParallel(cond_discriminator[i].cuda()) + perceptual_module = nn.DataParallel(perceptual_module.cuda()) if \ + perceptual_module else None + + if args.evaluate: + assert args.resume is not None + if args.evaluate_train: + log.info("Evaluting the training set.") + train_mean, train_std, train_vfs_mean, train_vfs_std = \ + evaluate(model, train_loader, options) + log.info("Inception score: {} ({})".format(train_mean, train_std)) + log.info("VggFace score: {} ({})".format( + train_vfs_mean, train_vfs_std)) + log.info("Evaluting the validation set.") + val_mean, val_std, vfs_mean, vfs_std = evaluate( + model, val_loader, options) + log.info("Inception score: {} ({})".format(val_mean, val_std)) + log.info("VggFace score: {} ({})".format(vfs_mean, vfs_std)) + fid_score = evaluate_fid(model, val_loader, options) + log.info("FID score: {}".format(fid_score)) + return 0 + + + got_best_IS = True + got_best_VFS = True + got_best_R1 = True + got_best_R5 = True + got_best_R10 = True + got_best_cos = True + got_best_L1 = True + others = None + for epoch in range(start_epoch, args.epochs): + if epoch >= args.eval_mode_after and model.training: + log.info('[Epoch {}/{}] switching to eval mode'.format( + epoch, args.epochs)) + model.eval() + if epoch == args.eval_mode_after: + optimizer = optim.Adam( + filter(lambda x: x.requires_grad, model.parameters()), + lr=lr, + betas=(args.beta1, 0.999),) + if epoch >= args.disable_l1_loss_after and \ + options["optim"]["l1_pixel_loss_weight"] > 1e-10: + # + log.info('[Epoch {}/{}] Disable L1 Loss'.format(epoch, args.epochs)) + options["optim"]["l1_pixel_loss_weight"] = 0 + start_time = time.time() + for iter, batch in enumerate(pyprind.prog_bar( + train_loader, + title="[Epoch {}/{}]".format(epoch, args.epochs), + width=50)): + # An iteration + if args.timing: + print("Loading Time: {} ms".format( + (time.time() - start_time) * 1000)) + t += 1 + ######### unpack the data ######### + imgs, log_mels, human_ids = batch + imgs = imgs.cuda() + log_mels = log_mels.type(float_dtype) + human_ids = human_ids.type(long_dtype) + ################################### + with timeit('forward', args.timing): + #model_out = model(imgs) + model_out = model(log_mels) + """ + imgs_pred: generated images + others: placeholder for other output + """ + imgs_pred, others = model_out + + if t % args.visualize_every == 0: + #print('Performing Visualization...') + training_status = model.training + model.eval() + samples = visualize_sample( + model, + imgs, + log_mels, + options["data"]["data_opts"]["image_normalize_method"], + visualize_attn=options['eval'].get('visualize_attn', False)) + model.train(mode=training_status) + logger.image_summary(samples, t, tag="vis") + + with timeit('G_loss', args.timing): + #model_boxes = None + # Skip the pixel loss if not using GT boxes + skip_pixel_loss = False + + # calculate L1 loss between imgs and imgs_self + total_loss, losses = calculate_model_losses( + options["optim"], skip_pixel_loss, imgs, imgs_pred,) + + if img_discriminator is not None: + for i in range(len(img_discriminator)): + # TODO: Here we need to choose the two methods: + # 1. determine whether imgs_pred is tuple or not + # 2. change the return (imgs_pred) of model to be tuple + if isinstance(imgs_pred, tuple): + scores_fake = img_discriminator[i](imgs_pred[i]) + else: + scores_fake = img_discriminator[i](imgs_pred) + weight = options["optim"]["d_loss_weight"] * \ + options["optim"]["d_img_weight"] + loss_name = 'g_gan_img_loss_%d' % i + total_loss = add_loss(total_loss, + gan_g_loss(scores_fake), losses, loss_name, weight) + + if ac_discriminator is not None: + for i in range(len(ac_discriminator)): + if isinstance(imgs_pred, tuple): + scores_fake, ac_loss = ac_discriminator[i]( + imgs_pred[i], human_ids) + else: + scores_fake, ac_loss = ac_discriminator[i]( + imgs_pred, human_ids) + weight = options["optim"]["d_loss_weight"] * \ + options["optim"]["ac_loss_weight"] + loss_name = 'ac_loss_%d' % i + total_loss = add_loss(total_loss, + ac_loss.mean(), + losses, + loss_name, + weight) + loss_name = 'g_gan_ac_loss_%d' % i + total_loss = add_loss(total_loss, + gan_g_loss(scores_fake), + losses, + loss_name, + weight) + + if cond_discriminator is not None: + for i in range(len(cond_discriminator)): + # TODO: check whether condition is in others or not ? + cond_vecs = others['cond'] + if isinstance(imgs_pred, tuple): + scores_fake = cond_discriminator[i]( + imgs_pred[i], cond_vecs) + else: + scores_fake = cond_discriminator[i]( + imgs_pred, cond_vecs) + weight = options["optim"]["d_loss_weight"] * \ + options["optim"]["cond_loss_weight"] + loss_name = 'g_cond_loss_%d' % i + total_loss = add_loss(total_loss, + gan_g_loss(scores_fake), losses, loss_name, weight) + + if options["optim"].get("perceptual_loss_weight", -1) > 0: + if isinstance(imgs_pred, tuple): + # Multi-Resolution Pixel Loss + for i, img_pred in enumerate(imgs_pred): + loss_name = 'img_perceptual_loss_%d' % i + img = imgs + while img.size()[2] != img_pred.size()[2]: + img = F.interpolate( + img, scale_factor=0.5, mode='nearest') + if s2f_val_evaluator.do_deprocess_and_preprocess: + img_pred_in = \ + s2f_val_evaluator.deprocess_and_preprocess( + img_pred) + img_in = \ + s2f_val_evaluator.deprocess_and_preprocess( + img) + if s2f_val_evaluator.crop_faces: + img_pred_in = \ + s2f_val_evaluator.crop_vgg_box(img_pred_in) + img_in = s2f_val_evaluator.crop_vgg_box(img_in) + perceptual_loss = perceptual_module( + img_pred_in, img_in) + perceptual_loss = perceptual_loss.mean() + weight = options["optim"]["perceptual_loss_weight"] + total_loss = add_loss(total_loss, perceptual_loss, + losses, loss_name, + weight) + else: + if s2f_val_evaluator.do_deprocess_and_preprocess: + imgs_pred_in = \ + s2f_val_evaluator.deprocess_and_preprocess( + imgs_pred) + imgs_in = \ + s2f_val_evaluator.deprocess_and_preprocess(imgs) + if s2f_val_evaluator.crop_faces: + imgs_pred_in = \ + s2f_val_evaluator.crop_vgg_box(imgs_pred_in) + imgs_in = s2f_val_evaluator.crop_vgg_box(imgs_in) + perceptual_loss = perceptual_module( + imgs_pred_in, imgs_in) + perceptual_loss = perceptual_loss.mean() + weight = options["optim"]["perceptual_loss_weight"] + total_loss = add_loss(total_loss, perceptual_loss, + losses, "img_perceptual_loss", + weight) + + losses['total_loss'] = total_loss.item() + if not math.isfinite(losses['total_loss']): + log.warn('WARNING: Got loss = NaN, not backpropping') + continue + + optimizer.zero_grad() + with timeit('backward', args.timing): + total_loss.backward() + optimizer.step() + + total_loss_d = None + ac_loss_real = None + ac_loss_fake = None + d_losses = {} + + with timeit('D_loss', args.timing): + if img_discriminator is not None: + d_img_losses = LossManager() + for i in range(len(img_discriminator)): + if isinstance(imgs_pred, tuple): + imgs_fake = imgs_pred[i].detach() + else: + imgs_fake = imgs_pred.detach() + imgs_real = imgs.detach() + while imgs_real.size()[2] != imgs_fake.size()[2]: + imgs_real = F.interpolate( + imgs_real, scale_factor=0.5, mode='nearest') + + scores_fake = img_discriminator[i](imgs_fake) + scores_real = img_discriminator[i](imgs_real) + + d_img_gan_loss = gan_d_loss(scores_real, scores_fake) + d_img_losses.add_loss( + d_img_gan_loss, 'd_img_gan_loss_%d' % i) + + for i in range(len(img_discriminator)): + optimizer_d_img[i].zero_grad() + d_img_losses.total_loss.backward() + for i in range(len(img_discriminator)): + optimizer_d_img[i].step() + + if ac_discriminator is not None: + d_ac_losses = LossManager() + for i in range(len(ac_discriminator)): + if isinstance(imgs_pred, tuple): + imgs_fake = imgs_pred[i].detach() + else: + imgs_fake = imgs_pred.detach() + + imgs_real = imgs.detach() + while imgs_real.size()[2] != imgs_fake.size()[2]: + imgs_real = F.interpolate( + imgs_real, scale_factor=0.5, mode='nearest') + + scores_real, ac_loss_real= ac_discriminator[i]( + imgs_real, human_ids) + scores_fake, ac_loss_fake = ac_discriminator[i]( + imgs_fake, human_ids) + + d_ac_gan_loss = gan_d_loss(scores_real, scores_fake) + d_ac_losses.add_loss( + d_ac_gan_loss, 'd_ac_gan_loss_%d' % i) + d_ac_losses.add_loss( + ac_loss_real.mean(), 'd_ac_loss_real_%d' % i) + + for i in range(len(ac_discriminator)): + optimizer_d_ac[i].zero_grad() + d_ac_losses.total_loss.backward() + for i in range(len(ac_discriminator)): + optimizer_d_ac[i].step() + + if cond_discriminator is not None: + cond_d_losses = LossManager() + for i in range(len(cond_discriminator)): + if isinstance(imgs_pred, tuple): + imgs_fake = imgs_pred[i].detach() + else: + imgs_fake = imgs_pred.detach() + imgs_real = imgs.detach() + cond_vecs = others['cond'].detach() + while imgs_real.size()[2] != imgs_fake.size()[2]: + imgs_real = F.interpolate( + imgs_real, scale_factor=0.5, mode='nearest') + + scores_fake = cond_discriminator[i]( + imgs_fake, cond_vecs) + scores_real = cond_discriminator[i]( + imgs_real, cond_vecs) + + cond_d_gan_loss = gan_d_loss(scores_real, scores_fake) + cond_d_losses.add_loss( + cond_d_gan_loss, 'cond_d_gan_loss_%d' % i) + + for i in range(len(cond_discriminator)): + optimizer_cond_d[i].zero_grad() + cond_d_losses.total_loss.backward() + for i in range(len(cond_discriminator)): + optimizer_cond_d[i].step() + + # Logging generative model losses + for name, val in losses.items(): + logger.scalar_summary("loss/{}".format(name), val, t) + if img_discriminator is not None: + for name, val in d_img_losses.items(): + logger.scalar_summary("d_loss/{}".format(name), val, t) + if ac_discriminator is not None: + for name, val in d_ac_losses.items(): + logger.scalar_summary("d_loss/{}".format(name), val, t) + if cond_discriminator is not None: + for name, val in cond_d_losses.items(): + logger.scalar_summary("d_loss/{}".format(name), val, t) + start_time = time.time() + + if epoch % args.eval_epochs == 0: + log.info('[Epoch {}/{}] checking on val'.format( + epoch, args.epochs) + ) + val_results = check_model( + args, options, epoch, val_loader, model) + val_losses, val_samples, val_inception, val_vfs = val_results + # call evaluate_s2f_metrics() here + val_facenet_L2_dist, val_facenet_L1_dist, val_facenet_cos_sim, \ + val_recall_tuple, val_ih_sim = \ + s2f_val_evaluator.get_metrics( + model, recall_method='cos_sim', get_ih_sim=True) + val_recall_at_1, val_recall_at_2, val_recall_at_5, \ + val_recall_at_10, val_recall_at_20, \ + val_recall_at_50 = val_recall_tuple + # Update the best of metrics + if val_inception[0] > best_inception[0]: + got_best_IS = True + best_inception = val_inception + if val_vfs[0] > best_vfs[0]: + got_best_VFS = True + best_vfs = val_vfs + if val_recall_at_1 > best_recall_1: + got_best_R1 = True + best_recall_1 = val_recall_at_1 + if val_recall_at_5 > best_recall_5: + got_best_R5 = True + best_recall_5 = val_recall_at_5 + if val_recall_at_10 > best_recall_10: + got_best_R10 = True + best_recall_10 = val_recall_at_10 + if val_facenet_cos_sim > best_cos: + got_best_cos = True + best_cos = val_facenet_cos_sim + if val_facenet_L1_dist < best_L1: + got_best_L1 = True + best_L1 = val_facenet_L1_dist + checkpoint['counters']['best_inception'] = best_inception + checkpoint['counters']['best_vfs'] = best_vfs + checkpoint['val_samples'].append(val_samples) + # checkpoint['val_batch_data'].append(val_batch_data) + for k, v in val_losses.items(): + checkpoint['val_losses'][k].append(v) + logger.scalar_summary("ckpt/val_{}".format(k), v, epoch) + logger.scalar_summary("ckpt/val_inception", val_inception[0], epoch) + logger.scalar_summary("ckpt/val_facenet_L2_dist", + val_facenet_L2_dist, epoch) + logger.scalar_summary("ckpt/val_facenet_L1_dist", + val_facenet_L1_dist, epoch) + logger.scalar_summary("ckpt/val_facenet_cos_sim", + val_facenet_cos_sim, epoch) + logger.scalar_summary("ckpt/val_recall_at_1", + val_recall_at_1, epoch) + logger.scalar_summary("ckpt/val_recall_at_2", + val_recall_at_2, epoch) + logger.scalar_summary("ckpt/val_recall_at_5", + val_recall_at_5, epoch) + logger.scalar_summary("ckpt/val_recall_at_10", + val_recall_at_10, epoch) + logger.scalar_summary("ckpt/val_recall_at_20", + val_recall_at_20, epoch) + logger.scalar_summary("ckpt/val_recall_at_50", + val_recall_at_50, epoch) + logger.scalar_summary("ckpt/val_ih_sim", + val_ih_sim, epoch) + logger.scalar_summary("ckpt/val_vfs", + val_vfs[0], epoch) + # Add speech2face metrics here.. + # + logger.image_summary(val_samples, epoch, tag="ckpt_val") + # log.info('[Epoch {}/{}] val iou: {}'.format( + # epoch, args.epochs, val_avg_iou)) + log.info('[Epoch {}/{}] val inception score: {} ({})'.format( + epoch, args.epochs, val_inception[0], val_inception[1])) + log.info('[Epoch {}/{}] best inception scores: {} ({})'.format( + epoch, args.epochs, best_inception[0], best_inception[1])) + log.info('[Epoch {}/{}] val vfs scores: {} ({})'.format( + epoch, args.epochs, val_vfs[0], val_vfs[1])) + log.info('[Epoch {}/{}] best vfs scores: {} ({})'.format( + epoch, args.epochs, best_vfs[0], best_vfs[1])) + log.info('[Epoch {}/{}] val recall at 5: {}, '.format( + epoch, args.epochs, val_recall_at_5) + \ + 'best recall at 5: {}'.format(best_recall_5)) + log.info('[Epoch {}/{}] val recall at 10: {}, '.format( + epoch, args.epochs, val_recall_at_10) + \ + 'best recall at 10: {}'.format(best_recall_10)) + log.info('[Epoch {}/{}] val cosine similarity: {}, '.format( + epoch, args.epochs, val_facenet_cos_sim) + \ + 'best cosine similarity: {}'.format(best_cos)) + log.info('[Epoch {}/{}] val L1 distance: {}, '.format( + epoch, args.epochs, val_facenet_L1_dist) + \ + 'best L1 distance: {}'.format(best_L1)) + + checkpoint['model_state'] = model.module.state_dict() + + if img_discriminator is not None: + for i in range(len(img_discriminator)): + term_name = 'd_img_state_%d' % i + checkpoint[term_name] = \ + img_discriminator[i].module.state_dict() + term_name = 'd_img_optim_state_%d' % i + checkpoint[term_name] = \ + optimizer_d_img[i].state_dict() + + if ac_discriminator is not None: + for i in range(len(ac_discriminator)): + term_name = 'd_ac_state_%d' % i + checkpoint[term_name] = \ + ac_discriminator[i].module.state_dict() + term_name = 'd_ac_optim_state_%d' % i + checkpoint[term_name] = \ + optimizer_d_ac[i].state_dict() + + if cond_discriminator is not None: + for i in range(len(cond_discriminator)): + term_name = 'cond_d_state_%d' % i + checkpoint[term_name] = \ + cond_discriminator[i].module.state_dict() + term_name = 'cond_d_optim_state_%d' % i + checkpoint[term_name] = \ + optimizer_cond_d[i].state_dict() + + checkpoint['optim_state'] = optimizer.state_dict() + checkpoint['counters']['epoch'] = epoch + checkpoint['counters']['t'] = t + checkpoint['counters']['best_inception'] = best_inception + checkpoint['counters']['best_vfs'] = best_vfs + checkpoint['counters']['best_recall_1'] = best_recall_1 + checkpoint['counters']['best_recall_5'] = best_recall_5 + checkpoint['counters']['best_recall_10'] = best_recall_10 + checkpoint['counters']['best_cos'] = best_cos + checkpoint['counters']['best_L1'] = best_L1 + checkpoint['lr'] = lr + checkpoint_path = os.path.join( + log_path, + '%s_with_model.pt' % args.checkpoint_name) + log.info('[Epoch {}/{}] Saving checkpoint: {}'.format( + epoch, args.epochs, checkpoint_path)) + torch.save(checkpoint, checkpoint_path) + if got_best_IS: + copyfile( + checkpoint_path, + os.path.join(log_path, 'best_IS_with_model.pt')) + got_best_IS = False + if got_best_VFS: + copyfile( + checkpoint_path, + os.path.join(log_path, 'best_VFS_with_model.pt')) + got_best_VFS = False + if got_best_R1: + copyfile( + checkpoint_path, + os.path.join(log_path, 'best_R1_with_model.pt')) + got_best_R1 = False + if got_best_R5: + copyfile( + checkpoint_path, + os.path.join(log_path, 'best_R5_with_model.pt')) + got_best_R5 = False + if got_best_R10: + copyfile( + checkpoint_path, + os.path.join(log_path, 'best_R10_with_model.pt')) + got_best_R10 = False + if got_best_L1: + copyfile( + checkpoint_path, + os.path.join(log_path, 'best_L1_with_model.pt')) + got_best_L1 = False + if got_best_cos: + copyfile( + checkpoint_path, + os.path.join(log_path, 'best_cos_with_model.pt')) + got_best_cos = False + + if epoch > 0 and epoch % 1000 == 0: + print('Saving checkpoint for Epoch {}.'.format(epoch)) + copyfile( + checkpoint_path, + os.path.join(log_path, 'epoch_{}_model.pt'.format(epoch))) + # Fuser Logic + elif args.train_fuser_only and epoch > 0 and epoch % 1 == 0: + print('Saving checkpoint for Epoch {}.'.format(epoch)) + copyfile( + checkpoint_path, + os.path.join(log_path, 'epoch_{}_model.pt'.format(epoch))) + # End of fuser logic + + if epoch >= args.decay_lr_epochs: + lr_end = args.learning_rate * 1e-3 + decay_frac = (epoch - args.decay_lr_epochs + 1) / \ + (args.epochs - args.decay_lr_epochs + 1e-5) + lr = args.learning_rate - decay_frac * (args.learning_rate - lr_end) + for param_group in optimizer.param_groups: + param_group["lr"] = lr + if img_discriminator is not None: + for i in range(len(optimizer_d_img)): + for param_group in optimizer_d_img[i].param_groups: + param_group["lr"] = lr + # for param_group in optimizer_d_img.param_groups: + # param_group["lr"] = lr + log.info('[Epoch {}/{}] learning rate: {}'.format( + epoch+1, args.epochs, lr)) + + logger.scalar_summary("ckpt/learning_rate", lr, epoch) + + # Evaluating after the whole training process. + log.info("Evaluting the validation set.") + is_mean, is_std, vfs_mean, vfs_std = evaluate(model, val_loader, options) + log.info("Inception score: {} ({})".format(is_mean, is_std)) + log.info("VggFace score: {} ({})".format(vfs_mean, vfs_std)) + + +if __name__ == '__main__': + main() diff --git a/mlflow/sf2f/train_registry.py b/mlflow/sf2f/train_registry.py new file mode 100644 index 0000000..f9785b3 --- /dev/null +++ b/mlflow/sf2f/train_registry.py @@ -0,0 +1,885 @@ +import functools +import os +import json +import math +from collections import defaultdict +import random +import time +import pyprind +import glog as log +from shutil import copyfile + +import numpy as np +import torch +import torch.optim as optim +import torch.nn as nn +import torch.nn.functional as F +from torch.utils.data.dataloader import default_collate +import mlflow, wandb + +from datasets import imagenet_deprocess_batch +from datasets import window_segment +import datasets +import models +import models.perceptual +from utils.losses import get_gan_losses +from utils import timeit, LossManager +from options.opts import args, options +from utils.logger import Logger +from utils import tensor2im +from utils.utils import load_my_state_dict +# losseds need to be modified +from utils.training_utils import add_loss, check_model, calculate_model_losses +from utils.training_utils import visualize_sample +from utils.evaluate import evaluate +from utils.evaluate_fid import evaluate_fid +from utils.s2f_evaluator import S2fEvaluator +from utils.mlflow_wandb import mlflow_wandb_logging + + +torch.backends.cudnn.benchmark = True + + +def main(): + global args, options + print(args) + print(options['data']) + # Set random seed, deterministic + torch.cuda.manual_seed(args.seed) + torch.manual_seed(args.seed) + np.random.seed(args.seed) + random.seed(args.seed) + torch.backends.cudnn.deterministic = True + torch.backends.cudnn.benchmark = False + #visualize_attn=options['eval'].get('visualize_attn', False) + #print('########### visualize_attn: {} ###########'.format(visualize_attn)) + float_dtype = torch.cuda.FloatTensor + long_dtype = torch.cuda.LongTensor + log.info("Building loader...") + train_loader, val_loader, test_loader = \ + datasets.build_loaders(options["data"]) + # Fuser Logic + if args.train_fuser_only: + train_loader.collate_fn = default_collate + train_loader.dataset.return_mel_segments = True + train_loader.dataset.mel_segments_rand_start = True + val_loader.collate_fn = default_collate + val_loader.dataset.return_mel_segments = True + s2f_face_gen_mode = 'naive' + else: + s2f_face_gen_mode = 'average_facenet_embedding' + # End of fuser logic + #s2f_val_evaluator = S2fEvaluator(val_loader, options) + s2f_val_evaluator = S2fEvaluator( + val_loader, + options, + extraction_size=100, #[100,200,300], + hq_emb_dict=True, + face_gen_mode=s2f_face_gen_mode) + + # [:len(checkpoint[term_name]['id_classifier.bias'])] class 맞추기 + # if restore_path is not None and os.path.isfile(restore_path): + train_loader.dataset.available_names = train_loader.dataset.available_names[:729] + + # Get total number of people in training set + num_train_id = len(train_loader.dataset) + # Initialize softmax neuron value + for ac_net in ['identity', 'identity_low', 'identity_mid', 'identity_high']: + if options['discriminator'].get(ac_net) is not None: + options['discriminator'][ac_net]['num_id'] = num_train_id + + log.info("Building Generative Model...") + model, model_kwargs = models.build_model( + options["generator"], + image_size=options["data"]["image_size"], + checkpoint_start_from=args.checkpoint_start_from) + model.type(float_dtype) + # Fuser logic + if args.train_fuser_only: + # Hardcode to freeze batchnorm layers + model.eval() + if args.train_fuser_decoder: + model.encoder.train_fuser_only() + else: + model.train_fuser_only() + # End of fuser logic + print(model) + + optimizer = torch.optim.Adam( + filter(lambda x: x.requires_grad, model.parameters()), + lr=args.learning_rate, + betas=(args.beta1, 0.999),) + + if (options["optim"]["d_loss_weight"] < 0 or \ + options["optim"]["d_img_weight"] < 0): + # Ignore image discriminator + img_discriminator = None + d_img_kwargs = {} + log.info("Ignoring Image Discriminator.") + else: + img_discriminator, d_img_kwargs = models.build_img_discriminator( + options["discriminator"]) + log.info("Done Building Image Discriminator.") + + if (options["optim"]["d_loss_weight"] < 0 or \ + options["optim"]["ac_loss_weight"] < 0): + # Ignore AC discriminator + ac_discriminator = None + ac_img_kwargs = {} + log.info("Ignoring Auxilary Classifier Discriminator.") + else: + ac_discriminator, ac_img_kwargs = models.build_ac_discriminator( + options["discriminator"]) + #print('AC Discriminator:', ac_discriminator) + log.info("Done Building Auxilary Classifier Discriminator.") + + if (options["optim"]["d_loss_weight"] < 0 or \ + options["optim"].get("cond_loss_weight", -1) < 0): + # Ignore Conditional discriminator + cond_discriminator = None + cond_d_kwargs = {} + log.info("Ignoring Conditional Discriminator.") + else: + cond_discriminator, cond_d_kwargs = models.build_cond_discriminator( + options["discriminator"]) + #print('Conditional Discriminator:', cond_discriminator) + log.info("Done Building Conditional Discriminator.") + + perceptual_module = None + if options["optim"].get("perceptual_loss_weight", -1) > 0: + ploss_name = options.get("perceptual", {}).get("arch", "FaceNetLoss") + ploss_cos_weight = options["optim"].get("cos_percept_loss_weight", -1) + perceptual_module = getattr( + models.perceptual, + ploss_name)(cos_loss_weight=ploss_cos_weight) + log.info("Done Building Perceptual {} Module.".format(ploss_name)) + if ploss_cos_weight > 0: + log.info("Perceptual Cos Loss Weight: {}".format(ploss_cos_weight)) + else: + log.info("Ignoring Perceptual Module.") + + gan_g_loss, gan_d_loss = get_gan_losses(options["optim"]["gan_loss_type"]) + + optimizer_d_img = [] + if img_discriminator is not None: + for i in range(len(img_discriminator)): + img_discriminator[i].type(float_dtype) + img_discriminator[i].train() + print(img_discriminator[i]) + optimizer_d_img.append(torch.optim.Adam( + filter(lambda x: x.requires_grad, + img_discriminator[i].parameters()), + lr=args.learning_rate, + betas=(args.beta1, 0.999),)) + + optimizer_d_ac = [] + if ac_discriminator is not None: + for i in range(len(ac_discriminator)): + ac_discriminator[i].type(float_dtype) + ac_discriminator[i].train() + print(ac_discriminator[i]) + optimizer_d_ac.append(torch.optim.Adam( + filter(lambda x: x.requires_grad, + ac_discriminator[i].parameters()), + lr=args.learning_rate, + betas=(args.beta1, 0.999),)) + + optimizer_cond_d = [] + if cond_discriminator is not None: + for i in range(len(cond_discriminator)): + cond_discriminator[i].type(float_dtype) + cond_discriminator[i].train() + print(cond_discriminator[i]) + optimizer_cond_d.append(torch.optim.Adam( + filter(lambda x: x.requires_grad, + cond_discriminator[i].parameters()), + lr=args.learning_rate, + betas=(args.beta1, 0.999),)) + + restore_path = None + if args.resume is not None: + restore_path = '%s_with_model.pt' % args.checkpoint_name + restore_path = os.path.join( + options["logs"]["output_dir"], args.resume, restore_path) + + if restore_path is not None and os.path.isfile(restore_path): + log.info('Restoring from checkpoint: {}'.format(restore_path)) + checkpoint = torch.load(restore_path) + model.load_state_dict(checkpoint['model_state']) + optimizer.load_state_dict(checkpoint['optim_state']) + + if img_discriminator is not None: + for i in range(len(img_discriminator)): + term_name = 'd_img_state_%d' % i + img_discriminator[i].load_state_dict(checkpoint[term_name]) + term_name = 'd_img_optim_state_%d' % i + optimizer_d_img[i].load_state_dict(checkpoint[term_name]) + + if ac_discriminator is not None: + for i in range(len(ac_discriminator)): + term_name = 'd_ac_state_%d' % i + ac_discriminator[i].load_state_dict(checkpoint[term_name]) + term_name = 'd_ac_optim_state_%d' % i + optimizer_d_ac[i].load_state_dict(checkpoint[term_name]) + + if cond_discriminator is not None: + for i in range(len(cond_discriminator)): + term_name = 'd_img_state_%d' % i + cond_discriminator[i].load_state_dict(checkpoint[term_name]) + term_name = 'd_img_optim_state_%d' % i + optimizer_cond_d[i].load_state_dict(checkpoint[term_name]) + + t = checkpoint['counters']['t'] + 1 + if 0 <= args.eval_mode_after <= t: + model.eval() + else: + model.train() + # Reset epoch here later + start_epoch = checkpoint['counters']['epoch'] + 1 + log_path = os.path.join(options["logs"]["output_dir"], args.resume,) + lr = checkpoint.get('learning_rate', args.learning_rate) + best_inception = checkpoint["counters"].get("best_inception", (0., 0.)) + best_vfs = checkpoint["counters"].get("best_vfs", (0., 0.)) + best_recall_1 = checkpoint["counters"].get("best_recall_1", 0.) + best_recall_5 = checkpoint["counters"].get("best_recall_5", 0.) + best_recall_10 = checkpoint["counters"].get("best_recall_10", 0.) + best_cos = checkpoint["counters"].get("best_cos", 0.) + best_L1 = checkpoint["counters"].get("best_L1", 100000.0) + options = checkpoint.get("options", options) + else: + t, start_epoch, best_inception, best_vfs = 0, 0, (0., 0.), (0., 0.) + best_recall_1, best_recall_5, best_recall_10 = 0.0, 0.0, 0.0 + best_cos, best_L1 = 0.0, 100000.0 + lr = args.learning_rate + checkpoint = { + 'args': args.__dict__, + 'options': options, + 'model_kwargs': model_kwargs, + 'd_img_kwargs': d_img_kwargs, + 'train_losses': defaultdict(list), + 'checkpoint_ts': [], + 'train_batch_data': [], + 'train_samples': [], + 'train_iou': [], + 'train_inception': [], + 'lr': [], + 'val_batch_data': [], + 'val_samples': [], + 'val_losses': defaultdict(list), + 'val_iou': [], + 'val_inception': [], + 'norm_d': [], + 'norm_g': [], + 'counters': { + 't': None, + 'epoch': None, + 'best_inception': None, + 'best_vfs': None, + 'best_recall_1': None, + 'best_recall_5': None, + 'best_recall_10': None, + 'best_cos': None, + 'best_L1': None, + }, + 'model_state': None, + 'model_best_state': None, + 'optim_state': None, + 'd_img_state': None, + 'd_img_best_state': None, + 'd_img_optim_state': None, + 'd_ac_state': None, + 'd_ac_optim_state': None, + } + + log_path = os.path.join( + options["logs"]["output_dir"], + options["logs"]["name"] + "-" + time.strftime("%Y%m%d-%H%M%S") + ) + + ### Fuser Logic + if args.pretrained_path is not None and \ + os.path.isfile(args.pretrained_path): + # Load + log.info('Loading Pretrained Model: {}'.format(args.pretrained_path)) + pre_checkpoint = torch.load(args.pretrained_path) + #model.load_state_dict(pre_checkpoint['model_state']) + load_my_state_dict(model, pre_checkpoint['model_state']) + #optimizer.load_state_dict(pre_checkpoint['optim_state']) + + if img_discriminator is not None: + for i in range(len(img_discriminator)): + term_name = 'd_img_state_%d' % i + img_discriminator[i].load_state_dict(pre_checkpoint[term_name]) + #term_name = 'd_img_optim_state_%d' % i + #optimizer_d_img[i].load_state_dict(pre_checkpoint[term_name]) + + if ac_discriminator is not None: + for i in range(len(ac_discriminator)): + term_name = 'd_ac_state_%d' % i + ac_discriminator[i].load_state_dict(pre_checkpoint[term_name]) + #term_name = 'd_ac_optim_state_%d' % i + #optimizer_d_ac[i].load_state_dict(pre_checkpoint[term_name]) + ### End of fuser logic + logger = Logger(log_path) + log.info("Logging to: {}".format(log_path)) + # save the current config yaml + copyfile(args.path_opts, + os.path.join(log_path, options["logs"]["name"] + '.yaml')) + + model = nn.DataParallel(model.cuda()) + if ac_discriminator is not None: + for i in range(len(ac_discriminator)): + ac_discriminator[i] = nn.DataParallel(ac_discriminator[i].cuda()) + if img_discriminator is not None: + for i in range(len(img_discriminator)): + img_discriminator[i] = nn.DataParallel(img_discriminator[i].cuda()) + if cond_discriminator is not None: + for i in range(len(cond_discriminator)): + cond_discriminator[i] = nn.DataParallel(cond_discriminator[i].cuda()) + perceptual_module = nn.DataParallel(perceptual_module.cuda()) if \ + perceptual_module else None + + if args.evaluate: + assert args.resume is not None + if args.evaluate_train: + log.info("Evaluting the training set.") + train_mean, train_std, train_vfs_mean, train_vfs_std = \ + evaluate(model, train_loader, options) + log.info("Inception score: {} ({})".format(train_mean, train_std)) + log.info("VggFace score: {} ({})".format( + train_vfs_mean, train_vfs_std)) + log.info("Evaluting the validation set.") + val_mean, val_std, vfs_mean, vfs_std = evaluate( + model, val_loader, options) + log.info("Inception score: {} ({})".format(val_mean, val_std)) + log.info("VggFace score: {} ({})".format(vfs_mean, vfs_std)) + fid_score = evaluate_fid(model, val_loader, options) + log.info("FID score: {}".format(fid_score)) + return 0 + + + got_best_IS = True + got_best_VFS = True + got_best_R1 = True + got_best_R5 = True + got_best_R10 = True + got_best_cos = True + got_best_L1 = True + others = None + for epoch in range(start_epoch, args.epochs): + if epoch >= args.eval_mode_after and model.training: + log.info('[Epoch {}/{}] switching to eval mode'.format( + epoch, args.epochs)) + model.eval() + if epoch == args.eval_mode_after: + optimizer = optim.Adam( + filter(lambda x: x.requires_grad, model.parameters()), + lr=lr, + betas=(args.beta1, 0.999),) + if epoch >= args.disable_l1_loss_after and \ + options["optim"]["l1_pixel_loss_weight"] > 1e-10: + # + log.info('[Epoch {}/{}] Disable L1 Loss'.format(epoch, args.epochs)) + options["optim"]["l1_pixel_loss_weight"] = 0 + start_time = time.time() + for iter, batch in enumerate(pyprind.prog_bar( + train_loader, + title="[Epoch {}/{}]".format(epoch, args.epochs), + width=50)): + # An iteration + if args.timing: + print("Loading Time: {} ms".format( + (time.time() - start_time) * 1000)) + t += 1 + ######### unpack the data ######### + imgs, log_mels, human_ids = batch + imgs = imgs.cuda() + log_mels = log_mels.type(float_dtype) + human_ids = human_ids.type(long_dtype) + ################################### + # log_mel_segs = window_segment(log_mels[0], window_length=125, stride_length=63) + ################################### + with timeit('forward', args.timing): + #model_out = model(imgs) + model_out = model(log_mels) + """ + imgs_pred: generated images + others: placeholder for other output + """ + imgs_pred, others = model_out + + if t % args.visualize_every == 0: + #print('Performing Visualization...') + training_status = model.training + model.eval() + samples = visualize_sample( + model, + imgs, + log_mels, + options["data"]["data_opts"]["image_normalize_method"], + visualize_attn=options['eval'].get('visualize_attn', False)) + model.train(mode=training_status) + logger.image_summary(samples, t, tag="vis") + + with timeit('G_loss', args.timing): + #model_boxes = None + # Skip the pixel loss if not using GT boxes + skip_pixel_loss = False + + # calculate L1 loss between imgs and imgs_self + total_loss, losses = calculate_model_losses( + options["optim"], skip_pixel_loss, imgs, imgs_pred,) + + if img_discriminator is not None: + for i in range(len(img_discriminator)): + # TODO: Here we need to choose the two methods: + # 1. determine whether imgs_pred is tuple or not + # 2. change the return (imgs_pred) of model to be tuple + if isinstance(imgs_pred, tuple): + scores_fake = img_discriminator[i](imgs_pred[i]) + else: + scores_fake = img_discriminator[i](imgs_pred) + weight = options["optim"]["d_loss_weight"] * \ + options["optim"]["d_img_weight"] + loss_name = 'g_gan_img_loss_%d' % i + total_loss = add_loss(total_loss, + gan_g_loss(scores_fake), losses, loss_name, weight) + + if ac_discriminator is not None: + for i in range(len(ac_discriminator)): + if isinstance(imgs_pred, tuple): + scores_fake, ac_loss = ac_discriminator[i]( + imgs_pred[i], human_ids) + else: + scores_fake, ac_loss = ac_discriminator[i]( + imgs_pred, human_ids) + weight = options["optim"]["d_loss_weight"] * \ + options["optim"]["ac_loss_weight"] + loss_name = 'ac_loss_%d' % i + total_loss = add_loss(total_loss, + ac_loss.mean(), + losses, + loss_name, + weight) + loss_name = 'g_gan_ac_loss_%d' % i + total_loss = add_loss(total_loss, + gan_g_loss(scores_fake), + losses, + loss_name, + weight) + + if cond_discriminator is not None: + for i in range(len(cond_discriminator)): + # TODO: check whether condition is in others or not ? + cond_vecs = others['cond'] + if isinstance(imgs_pred, tuple): + scores_fake = cond_discriminator[i]( + imgs_pred[i], cond_vecs) + else: + scores_fake = cond_discriminator[i]( + imgs_pred, cond_vecs) + weight = options["optim"]["d_loss_weight"] * \ + options["optim"]["cond_loss_weight"] + loss_name = 'g_cond_loss_%d' % i + total_loss = add_loss(total_loss, + gan_g_loss(scores_fake), losses, loss_name, weight) + + if options["optim"].get("perceptual_loss_weight", -1) > 0: + if isinstance(imgs_pred, tuple): + # Multi-Resolution Pixel Loss + for i, img_pred in enumerate(imgs_pred): + loss_name = 'img_perceptual_loss_%d' % i + img = imgs + while img.size()[2] != img_pred.size()[2]: + img = F.interpolate( + img, scale_factor=0.5, mode='nearest') + if s2f_val_evaluator.do_deprocess_and_preprocess: + img_pred_in = \ + s2f_val_evaluator.deprocess_and_preprocess( + img_pred) + img_in = \ + s2f_val_evaluator.deprocess_and_preprocess( + img) + if s2f_val_evaluator.crop_faces: + img_pred_in = \ + s2f_val_evaluator.crop_vgg_box(img_pred_in) + img_in = s2f_val_evaluator.crop_vgg_box(img_in) + perceptual_loss = perceptual_module( + img_pred_in, img_in) + perceptual_loss = perceptual_loss.mean() + weight = options["optim"]["perceptual_loss_weight"] + total_loss = add_loss(total_loss, perceptual_loss, + losses, loss_name, + weight) + else: + if s2f_val_evaluator.do_deprocess_and_preprocess: + imgs_pred_in = \ + s2f_val_evaluator.deprocess_and_preprocess( + imgs_pred) + imgs_in = \ + s2f_val_evaluator.deprocess_and_preprocess(imgs) + if s2f_val_evaluator.crop_faces: + imgs_pred_in = \ + s2f_val_evaluator.crop_vgg_box(imgs_pred_in) + imgs_in = s2f_val_evaluator.crop_vgg_box(imgs_in) + perceptual_loss = perceptual_module( + imgs_pred_in, imgs_in) + perceptual_loss = perceptual_loss.mean() + weight = options["optim"]["perceptual_loss_weight"] + total_loss = add_loss(total_loss, perceptual_loss, + losses, "img_perceptual_loss", + weight) + + losses['total_loss'] = total_loss.item() + if not math.isfinite(losses['total_loss']): + log.warn('WARNING: Got loss = NaN, not backpropping') + continue + + optimizer.zero_grad() + with timeit('backward', args.timing): + total_loss.backward() + optimizer.step() + + total_loss_d = None + ac_loss_real = None + ac_loss_fake = None + d_losses = {} + + with timeit('D_loss', args.timing): + if img_discriminator is not None: + d_img_losses = LossManager() + for i in range(len(img_discriminator)): + if isinstance(imgs_pred, tuple): + imgs_fake = imgs_pred[i].detach() + else: + imgs_fake = imgs_pred.detach() + imgs_real = imgs.detach() + while imgs_real.size()[2] != imgs_fake.size()[2]: + imgs_real = F.interpolate( + imgs_real, scale_factor=0.5, mode='nearest') + + scores_fake = img_discriminator[i](imgs_fake) + scores_real = img_discriminator[i](imgs_real) + + d_img_gan_loss = gan_d_loss(scores_real, scores_fake) + d_img_losses.add_loss( + d_img_gan_loss, 'd_img_gan_loss_%d' % i) + + for i in range(len(img_discriminator)): + optimizer_d_img[i].zero_grad() + d_img_losses.total_loss.backward() + for i in range(len(img_discriminator)): + optimizer_d_img[i].step() + + if ac_discriminator is not None: + d_ac_losses = LossManager() + for i in range(len(ac_discriminator)): + if isinstance(imgs_pred, tuple): + imgs_fake = imgs_pred[i].detach() + else: + imgs_fake = imgs_pred.detach() + + imgs_real = imgs.detach() + while imgs_real.size()[2] != imgs_fake.size()[2]: + imgs_real = F.interpolate( + imgs_real, scale_factor=0.5, mode='nearest') + + scores_real, ac_loss_real= ac_discriminator[i]( + imgs_real, human_ids) + scores_fake, ac_loss_fake = ac_discriminator[i]( + imgs_fake, human_ids) + + d_ac_gan_loss = gan_d_loss(scores_real, scores_fake) + d_ac_losses.add_loss( + d_ac_gan_loss, 'd_ac_gan_loss_%d' % i) + d_ac_losses.add_loss( + ac_loss_real.mean(), 'd_ac_loss_real_%d' % i) + + for i in range(len(ac_discriminator)): + optimizer_d_ac[i].zero_grad() + d_ac_losses.total_loss.backward() + for i in range(len(ac_discriminator)): + optimizer_d_ac[i].step() + + if cond_discriminator is not None: + cond_d_losses = LossManager() + for i in range(len(cond_discriminator)): + if isinstance(imgs_pred, tuple): + imgs_fake = imgs_pred[i].detach() + else: + imgs_fake = imgs_pred.detach() + imgs_real = imgs.detach() + cond_vecs = others['cond'].detach() + while imgs_real.size()[2] != imgs_fake.size()[2]: + imgs_real = F.interpolate( + imgs_real, scale_factor=0.5, mode='nearest') + + scores_fake = cond_discriminator[i]( + imgs_fake, cond_vecs) + scores_real = cond_discriminator[i]( + imgs_real, cond_vecs) + + cond_d_gan_loss = gan_d_loss(scores_real, scores_fake) + cond_d_losses.add_loss( + cond_d_gan_loss, 'cond_d_gan_loss_%d' % i) + + for i in range(len(cond_discriminator)): + optimizer_cond_d[i].zero_grad() + cond_d_losses.total_loss.backward() + for i in range(len(cond_discriminator)): + optimizer_cond_d[i].step() + + # Logging generative model losses + for name, val in losses.items(): + logger.scalar_summary("loss/{}".format(name), val, t) + if img_discriminator is not None: + for name, val in d_img_losses.items(): + logger.scalar_summary("d_loss/{}".format(name), val, t) + if ac_discriminator is not None: + for name, val in d_ac_losses.items(): + logger.scalar_summary("d_loss/{}".format(name), val, t) + if cond_discriminator is not None: + for name, val in cond_d_losses.items(): + logger.scalar_summary("d_loss/{}".format(name), val, t) + start_time = time.time() + + if epoch % args.eval_epochs == 0: + log.info('[Epoch {}/{}] checking on val'.format( + epoch, args.epochs) + ) + val_results = check_model( + args, options, epoch, val_loader, model) + val_losses, val_samples, val_inception, val_vfs = val_results + # call evaluate_s2f_metrics() here + val_facenet_L2_dist, val_facenet_L1_dist, val_facenet_cos_sim, \ + val_recall_tuple, val_ih_sim = \ + s2f_val_evaluator.get_metrics( + model, recall_method='cos_sim', get_ih_sim=True) + val_recall_at_1, val_recall_at_2, val_recall_at_5, \ + val_recall_at_10, val_recall_at_20, \ + val_recall_at_50 = val_recall_tuple + if log_mels.dim() == 2: + log_mel_segs = window_segment(log_mels, window_length=62, stride_length=31) + elif log_mels.dim() == 3: + log_mel_segs = window_segment(log_mels[0], window_length=62, stride_length=31) + + signature = mlflow.models.signature.infer_signature(model_input=log_mel_segs.unsqueeze(0).cpu().numpy(), model_output=model_out[0].detach().cpu().numpy()) + + log_dict = {"val_losses":val_losses['L1_pixel_loss'], "val_inception":val_inception[0], "val_vfs":val_vfs[0], "val_facenet_L2_dist":val_facenet_L2_dist, "val_facenet_L1_dist":val_facenet_L1_dist, "val_facenet_cos_sim":val_facenet_cos_sim, "val_ih_sim":val_ih_sim, "val_recall_at_1":val_recall_at_1, "val_recall_at_2":val_recall_at_2, "val_recall_at_5":val_recall_at_5, "val_recall_at_10":val_recall_at_10, "val_recall_at_20":val_recall_at_20, "val_recall_at_50":val_recall_at_50} + mlflow_wandb_logging(log_dict,model,signature,log_mel_segs.unsqueeze(0).cpu().numpy(),val_samples['samples']) + # Update the best of metrics + if val_inception[0] > best_inception[0]: + got_best_IS = True + best_inception = val_inception + if val_vfs[0] > best_vfs[0]: + got_best_VFS = True + best_vfs = val_vfs + if val_recall_at_1 > best_recall_1: + got_best_R1 = True + best_recall_1 = val_recall_at_1 + if val_recall_at_5 > best_recall_5: + got_best_R5 = True + best_recall_5 = val_recall_at_5 + if val_recall_at_10 > best_recall_10: + got_best_R10 = True + best_recall_10 = val_recall_at_10 + if val_facenet_cos_sim > best_cos: + got_best_cos = True + best_cos = val_facenet_cos_sim + if val_facenet_L1_dist < best_L1: + got_best_L1 = True + best_L1 = val_facenet_L1_dist + + checkpoint['counters']['best_inception'] = best_inception + checkpoint['counters']['best_vfs'] = best_vfs + checkpoint['val_samples'].append(val_samples) + # checkpoint['val_batch_data'].append(val_batch_data) + for k, v in val_losses.items(): + checkpoint['val_losses'][k].append(v) + logger.scalar_summary("ckpt/val_{}".format(k), v, epoch) + logger.scalar_summary("ckpt/val_inception", val_inception[0], epoch) + logger.scalar_summary("ckpt/val_facenet_L2_dist", + val_facenet_L2_dist, epoch) + logger.scalar_summary("ckpt/val_facenet_L1_dist", + val_facenet_L1_dist, epoch) + logger.scalar_summary("ckpt/val_facenet_cos_sim", + val_facenet_cos_sim, epoch) + logger.scalar_summary("ckpt/val_recall_at_1", + val_recall_at_1, epoch) + logger.scalar_summary("ckpt/val_recall_at_2", + val_recall_at_2, epoch) + logger.scalar_summary("ckpt/val_recall_at_5", + val_recall_at_5, epoch) + logger.scalar_summary("ckpt/val_recall_at_10", + val_recall_at_10, epoch) + logger.scalar_summary("ckpt/val_recall_at_20", + val_recall_at_20, epoch) + logger.scalar_summary("ckpt/val_recall_at_50", + val_recall_at_50, epoch) + logger.scalar_summary("ckpt/val_ih_sim", + val_ih_sim, epoch) + logger.scalar_summary("ckpt/val_vfs", + val_vfs[0], epoch) + # Add speech2face metrics here.. + # + logger.image_summary(val_samples, epoch, tag="ckpt_val") + # log.info('[Epoch {}/{}] val iou: {}'.format( + # epoch, args.epochs, val_avg_iou)) + log.info('[Epoch {}/{}] val inception score: {} ({})'.format( + epoch, args.epochs, val_inception[0], val_inception[1])) + log.info('[Epoch {}/{}] best inception scores: {} ({})'.format( + epoch, args.epochs, best_inception[0], best_inception[1])) + log.info('[Epoch {}/{}] val vfs scores: {} ({})'.format( + epoch, args.epochs, val_vfs[0], val_vfs[1])) + log.info('[Epoch {}/{}] best vfs scores: {} ({})'.format( + epoch, args.epochs, best_vfs[0], best_vfs[1])) + log.info('[Epoch {}/{}] val recall at 5: {}, '.format( + epoch, args.epochs, val_recall_at_5) + \ + 'best recall at 5: {}'.format(best_recall_5)) + log.info('[Epoch {}/{}] val recall at 10: {}, '.format( + epoch, args.epochs, val_recall_at_10) + \ + 'best recall at 10: {}'.format(best_recall_10)) + log.info('[Epoch {}/{}] val cosine similarity: {}, '.format( + epoch, args.epochs, val_facenet_cos_sim) + \ + 'best cosine similarity: {}'.format(best_cos)) + log.info('[Epoch {}/{}] val L1 distance: {}, '.format( + epoch, args.epochs, val_facenet_L1_dist) + \ + 'best L1 distance: {}'.format(best_L1)) + + checkpoint['model_state'] = model.module.state_dict() + + if img_discriminator is not None: + for i in range(len(img_discriminator)): + term_name = 'd_img_state_%d' % i + checkpoint[term_name] = \ + img_discriminator[i].module.state_dict() + term_name = 'd_img_optim_state_%d' % i + checkpoint[term_name] = \ + optimizer_d_img[i].state_dict() + + if ac_discriminator is not None: + for i in range(len(ac_discriminator)): + term_name = 'd_ac_state_%d' % i + checkpoint[term_name] = \ + ac_discriminator[i].module.state_dict() + term_name = 'd_ac_optim_state_%d' % i + checkpoint[term_name] = \ + optimizer_d_ac[i].state_dict() + + if cond_discriminator is not None: + for i in range(len(cond_discriminator)): + term_name = 'cond_d_state_%d' % i + checkpoint[term_name] = \ + cond_discriminator[i].module.state_dict() + term_name = 'cond_d_optim_state_%d' % i + checkpoint[term_name] = \ + optimizer_cond_d[i].state_dict() + + checkpoint['optim_state'] = optimizer.state_dict() + checkpoint['counters']['epoch'] = epoch + checkpoint['counters']['t'] = t + checkpoint['counters']['best_inception'] = best_inception + checkpoint['counters']['best_vfs'] = best_vfs + checkpoint['counters']['best_recall_1'] = best_recall_1 + checkpoint['counters']['best_recall_5'] = best_recall_5 + checkpoint['counters']['best_recall_10'] = best_recall_10 + checkpoint['counters']['best_cos'] = best_cos + checkpoint['counters']['best_L1'] = best_L1 + checkpoint['lr'] = lr + checkpoint_path = os.path.join( + log_path, + '%s_with_model.pt' % args.checkpoint_name) + log.info('[Epoch {}/{}] Saving checkpoint: {}'.format( + epoch, args.epochs, checkpoint_path)) + torch.save(checkpoint, checkpoint_path) + if got_best_IS: + copyfile( + checkpoint_path, + os.path.join(log_path, 'best_IS_with_model.pt')) + got_best_IS = False + if got_best_VFS: + copyfile( + checkpoint_path, + os.path.join(log_path, 'best_VFS_with_model.pt')) + got_best_VFS = False + if got_best_R1: + copyfile( + checkpoint_path, + os.path.join(log_path, 'best_R1_with_model.pt')) + got_best_R1 = False + if got_best_R5: + copyfile( + checkpoint_path, + os.path.join(log_path, 'best_R5_with_model.pt')) + got_best_R5 = False + if got_best_R10: + copyfile( + checkpoint_path, + os.path.join(log_path, 'best_R10_with_model.pt')) + got_best_R10 = False + if got_best_L1: + copyfile( + checkpoint_path, + os.path.join(log_path, 'best_L1_with_model.pt')) + got_best_L1 = False + if got_best_cos: + copyfile( + checkpoint_path, + os.path.join(log_path, 'best_cos_with_model.pt')) + got_best_cos = False + + if epoch > 0 and epoch % 1000 == 0: + print('Saving checkpoint for Epoch {}.'.format(epoch)) + copyfile( + checkpoint_path, + os.path.join(log_path, 'epoch_{}_model.pt'.format(epoch))) + # Fuser Logic + elif args.train_fuser_only and epoch > 0 and epoch % 1 == 0: + print('Saving checkpoint for Epoch {}.'.format(epoch)) + copyfile( + checkpoint_path, + os.path.join(log_path, 'epoch_{}_model.pt'.format(epoch))) + # End of fuser logic + + if epoch >= args.decay_lr_epochs: + lr_end = args.learning_rate * 1e-3 + decay_frac = (epoch - args.decay_lr_epochs + 1) / \ + (args.epochs - args.decay_lr_epochs + 1e-5) + lr = args.learning_rate - decay_frac * (args.learning_rate - lr_end) + for param_group in optimizer.param_groups: + param_group["lr"] = lr + if img_discriminator is not None: + for i in range(len(optimizer_d_img)): + for param_group in optimizer_d_img[i].param_groups: + param_group["lr"] = lr + # for param_group in optimizer_d_img.param_groups: + # param_group["lr"] = lr + log.info('[Epoch {}/{}] learning rate: {}'.format( + epoch+1, args.epochs, lr)) + + logger.scalar_summary("ckpt/learning_rate", lr, epoch) + + # Evaluating after the whole training process. + log.info("Evaluting the validation set.") + is_mean, is_std, vfs_mean, vfs_std = evaluate(model, val_loader, options) + log.info("Inception score: {} ({})".format(is_mean, is_std)) + log.info("VggFace score: {} ({})".format(vfs_mean, vfs_std)) + + +if __name__ == '__main__': + os.environ["MLFLOW_S3_ENDPOINT_URL"] = "http://localhost:9000" + os.environ["MLFLOW_TRACKING_URI"] = "http://localhost:5001" + os.environ["AWS_ACCESS_KEY_ID"] = "minio" + os.environ["AWS_SECRET_ACCESS_KEY"] = "miniostorage" + mlflow.start_run() + mlflow.set_experiment("sf2f_voxceleb") + wandb.init( + project="Voice2Face", + config={ + "architecture": "sf2f", + "dataset": "Voxceleb", + "notes": "test", + }, + name="sf2f_voxceleb_1", + ) + main() diff --git a/mlflow/sf2f/utils/__init__.py b/mlflow/sf2f/utils/__init__.py new file mode 100644 index 0000000..55e5f84 --- /dev/null +++ b/mlflow/sf2f/utils/__init__.py @@ -0,0 +1 @@ +from .common import * diff --git a/mlflow/sf2f/utils/bilinear.py b/mlflow/sf2f/utils/bilinear.py new file mode 100644 index 0000000..7a12100 --- /dev/null +++ b/mlflow/sf2f/utils/bilinear.py @@ -0,0 +1,296 @@ +import numpy as np +import torch +import torch.nn.functional as F +from .common import timeit + + +""" +Functions for performing differentiable bilinear cropping of images, for use in +the object discriminator +""" + +def crop_bbox_batch(feats, bbox, bbox_to_feats, HH, WW=None, backend='cudnn'): + """ + Inputs: + - feats: FloatTensor of shape (N, C, H, W) + - bbox: FloatTensor of shape (B, 4) giving bounding box coordinates + - bbox_to_feats: LongTensor of shape (B,) mapping boxes to feature maps; + each element is in the range [0, N) and bbox_to_feats[b] = i means that + bbox[b] will be cropped from feats[i]. + - HH, WW: Size of the output crops + + Returns: + - crops: FloatTensor of shape (B, C, HH, WW) where crops[i] uses bbox[i] to + crop from feats[bbox_to_feats[i]]. + """ + if backend == 'cudnn': + return crop_bbox_batch_cudnn(feats, bbox, bbox_to_feats, HH, WW) + N, C, H, W = feats.size() + B = bbox.size(0) + if WW is None: + WW = HH + dtype, device = feats.dtype, feats.device + crops = torch.zeros(B, C, HH, WW, dtype=dtype, device=device) + for i in range(N): + idx = (bbox_to_feats == i).nonzero() + if idx.dim() == 0: + continue + idx = idx.view(-1) + n = idx.size(0) + cur_feats = feats[i].view(1, C, H, W).expand(n, C, H, W).contiguous() + cur_bbox = bbox[idx] + cur_crops = crop_bbox(cur_feats, cur_bbox, HH, WW) + crops[idx] = cur_crops + return crops + + +def _invperm(p): + N = p.size(0) + eye = torch.arange(0, N).type_as(p) + pp = (eye[:, None] == p).nonzero()[:, 1] + return pp + + +def crop_bbox_batch_cudnn(feats, bbox, bbox_to_feats, HH, WW=None): + N, C, H, W = feats.size() + B = bbox.size(0) + if WW is None: + WW = HH + dtype = feats.type() + + feats_flat, bbox_flat, all_idx = [], [], [] + for i in range(N): + idx = (bbox_to_feats == i).nonzero() + if idx.dim() == 0: + continue + idx = idx.view(-1) + n = idx.size(0) + cur_feats = feats[i].view(1, C, H, W).expand(n, C, H, W).contiguous() + cur_bbox = bbox[idx] + + feats_flat.append(cur_feats) + bbox_flat.append(cur_bbox) + all_idx.append(idx) + + feats_flat = torch.cat(feats_flat, dim=0) + bbox_flat = torch.cat(bbox_flat, dim=0) + crops = crop_bbox(feats_flat, bbox_flat, HH, WW, backend='cudnn') + + # If the crops were sequential (all_idx is identity permutation) then we can + # simply return them; otherwise we need to permute crops by the inverse + # permutation from all_idx. + all_idx = torch.cat(all_idx, dim=0) + eye = torch.arange(0, B).type_as(all_idx) + if (all_idx == eye).all(): + return crops + return crops[_invperm(all_idx)] + + +def crop_bbox(feats, bbox, HH, WW=None, backend='cudnn'): + """ + Take differentiable crops of feats specified by bbox. + + Inputs: + - feats: Tensor of shape (N, C, H, W) + - bbox: Bounding box coordinates of shape (N, 4) in the format + [x0, y0, x1, y1] in the [0, 1] coordinate space. + - HH, WW: Size of the output crops. + + Returns: + - crops: Tensor of shape (N, C, HH, WW) where crops[i] is the portion of + feats[i] specified by bbox[i], reshaped to (HH, WW) using bilinear sampling. + """ + N = feats.size(0) + assert bbox.size(0) == N + assert bbox.size(1) == 4 + if WW is None: + WW = HH + if backend == 'cudnn': + # Change box from [0, 1] to [-1, 1] coordinate system + bbox = 2 * bbox - 1 + x0, y0 = bbox[:, 0], bbox[:, 1] + x1, y1 = bbox[:, 2], bbox[:, 3] + X = tensor_linspace(x0, x1, steps=WW).view(N, 1, WW).expand(N, HH, WW) + Y = tensor_linspace(y0, y1, steps=HH).view(N, HH, 1).expand(N, HH, WW) + if backend == 'jj': + return bilinear_sample(feats, X, Y) + elif backend == 'cudnn': + grid = torch.stack([X, Y], dim=3) + return F.grid_sample(feats, grid) + + +def uncrop_bbox(feats, bbox, H, W=None, fill_value=0): + """ + Inverse operation to crop_bbox; construct output images where the feature maps + from feats have been reshaped and placed into the positions specified by bbox. + + Inputs: + - feats: Tensor of shape (N, C, HH, WW) + - bbox: Bounding box coordinates of shape (N, 4) in the format + [x0, y0, x1, y1] in the [0, 1] coordinate space. + - H, W: Size of output. + - fill_value: Portions of the output image that are outside the bounding box + will be filled with this value. + + Returns: + - out: Tensor of shape (N, C, H, W) where the portion of out[i] given by + bbox[i] contains feats[i], reshaped using bilinear sampling. + """ + N, C = feats.size(0), feats.size(1) + assert bbox.size(0) == N + assert bbox.size(1) == 4 + if W is None: + W = H + + x0, y0 = bbox[:, 0], bbox[:, 1] + x1, y1 = bbox[:, 2], bbox[:, 3] + ww = x1 - x0 + hh = y1 - y0 + + x0 = x0.contiguous().view(N, 1).expand(N, H) + x1 = x1.contiguous().view(N, 1).expand(N, H) + ww = ww.view(N, 1).expand(N, H) + + y0 = y0.contiguous().view(N, 1).expand(N, W) + y1 = y1.contiguous().view(N, 1).expand(N, W) + hh = hh.view(N, 1).expand(N, W) + + X = torch.linspace(0, 1, steps=W).view(1, W).expand(N, W).to(feats) + Y = torch.linspace(0, 1, steps=H).view(1, H).expand(N, H).to(feats) + + X = (X - x0) / ww + Y = (Y - y0) / hh + + # For ByteTensors, (x + y).clamp(max=1) gives logical_or + X_out_mask = ((X < 0) + (X > 1)).view(N, 1, W).expand(N, H, W) + Y_out_mask = ((Y < 0) + (Y > 1)).view(N, H, 1).expand(N, H, W) + out_mask = (X_out_mask + Y_out_mask).clamp(max=1) + out_mask = out_mask.view(N, 1, H, W).expand(N, C, H, W) + + X = X.view(N, 1, W).expand(N, H, W) + Y = Y.view(N, H, 1).expand(N, H, W) + + grid = torch.stack([X, Y], dim=3) + out = F.grid_sample(feats, grid) + out[out_mask] = fill_value + return out + + +def bilinear_sample(feats, X, Y): + """ + Perform bilinear sampling on the features in feats using the sampling grid + given by X and Y. + + Inputs: + - feats: Tensor holding input feature map, of shape (N, C, H, W) + - X, Y: Tensors holding x and y coordinates of the sampling + grids; both have shape shape (N, HH, WW) and have elements in the range [0, 1]. + Returns: + - out: Tensor of shape (B, C, HH, WW) where out[i] is computed + by sampling from feats[idx[i]] using the sampling grid (X[i], Y[i]). + """ + N, C, H, W = feats.size() + assert X.size() == Y.size() + assert X.size(0) == N + _, HH, WW = X.size() + + X = X.mul(W) + Y = Y.mul(H) + + # Get the x and y coordinates for the four samples + x0 = X.floor().clamp(min=0, max=W-1) + x1 = (x0 + 1).clamp(min=0, max=W-1) + y0 = Y.floor().clamp(min=0, max=H-1) + y1 = (y0 + 1).clamp(min=0, max=H-1) + + # In numpy we could do something like feats[i, :, y0, x0] to pull out + # the elements of feats at coordinates y0 and x0, but PyTorch doesn't + # yet support this style of indexing. Instead we have to use the gather + # method, which only allows us to index along one dimension at a time; + # therefore we will collapse the features (BB, C, H, W) into (BB, C, H * W) + # and index along the last dimension. Below we generate linear indices into + # the collapsed last dimension for each of the four combinations we need. + y0x0_idx = (W * y0 + x0).view(N, 1, HH * WW).expand(N, C, HH * WW) + y1x0_idx = (W * y1 + x0).view(N, 1, HH * WW).expand(N, C, HH * WW) + y0x1_idx = (W * y0 + x1).view(N, 1, HH * WW).expand(N, C, HH * WW) + y1x1_idx = (W * y1 + x1).view(N, 1, HH * WW).expand(N, C, HH * WW) + + # Actually use gather to pull out the values from feats corresponding + # to our four samples, then reshape them to (BB, C, HH, WW) + feats_flat = feats.view(N, C, H * W) + v1 = feats_flat.gather(2, y0x0_idx.long()).view(N, C, HH, WW) + v2 = feats_flat.gather(2, y1x0_idx.long()).view(N, C, HH, WW) + v3 = feats_flat.gather(2, y0x1_idx.long()).view(N, C, HH, WW) + v4 = feats_flat.gather(2, y1x1_idx.long()).view(N, C, HH, WW) + + # Compute the weights for the four samples + w1 = ((x1 - X) * (y1 - Y)).view(N, 1, HH, WW).expand(N, C, HH, WW) + w2 = ((x1 - X) * (Y - y0)).view(N, 1, HH, WW).expand(N, C, HH, WW) + w3 = ((X - x0) * (y1 - Y)).view(N, 1, HH, WW).expand(N, C, HH, WW) + w4 = ((X - x0) * (Y - y0)).view(N, 1, HH, WW).expand(N, C, HH, WW) + + # Multiply the samples by the weights to give our interpolated results. + out = w1 * v1 + w2 * v2 + w3 * v3 + w4 * v4 + return out + + +def tensor_linspace(start, end, steps=10): + """ + Vectorized version of torch.linspace. + + Inputs: + - start: Tensor of any shape + - end: Tensor of the same shape as start + - steps: Integer + + Returns: + - out: Tensor of shape start.size() + (steps,), such that + out.select(-1, 0) == start, out.select(-1, -1) == end, + and the other elements of out linearly interpolate between + start and end. + """ + assert start.size() == end.size() + view_size = start.size() + (1,) + w_size = (1,) * start.dim() + (steps,) + out_size = start.size() + (steps,) + + start_w = torch.linspace(1, 0, steps=steps).to(start) + start_w = start_w.view(w_size).expand(out_size) + end_w = torch.linspace(0, 1, steps=steps).to(start) + end_w = end_w.view(w_size).expand(out_size) + + start = start.contiguous().view(view_size).expand(out_size) + end = end.contiguous().view(view_size).expand(out_size) + + out = start_w * start + end_w * end + return out + + +if __name__ == '__main__': + import numpy as np + # from scipy.misc import imread, imsave, imresize + from cv2 import imresize + from imageio import imread + from imageio import imwrite as imsave + + cat = imresize(imread('cat.jpg'), (256, 256)) + dog = imresize(imread('dog.jpg'), (256, 256)) + feats = torch.stack([ + torch.from_numpy(cat.transpose(2, 0, 1).astype(np.float32)), + torch.from_numpy(dog.transpose(2, 0, 1).astype(np.float32))], + dim=0) + + boxes = torch.FloatTensor([ + [0, 0, 1, 1], + [0.25, 0.25, 0.75, 0.75], + [0, 0, 0.5, 0.5], + ]) + + box_to_feats = torch.LongTensor([1, 0, 1]).cuda() + + feats, boxes = feats.cuda(), boxes.cuda() + crops = crop_bbox_batch_cudnn(feats, boxes, box_to_feats, 128) + for i in range(crops.size(0)): + crop_np = crops.data[i].cpu().numpy().transpose( + 1, 2, 0).astype(np.uint8) + imsave('out%d.png' % i, crop_np) diff --git a/mlflow/sf2f/utils/box_utils.py b/mlflow/sf2f/utils/box_utils.py new file mode 100644 index 0000000..898e543 --- /dev/null +++ b/mlflow/sf2f/utils/box_utils.py @@ -0,0 +1,133 @@ +import torch +import numpy as np +""" +Utilities for dealing with bounding boxes +""" + + +def box_union(box1, box2): + assert isinstance(box1, torch.Tensor) + return torch.cat( + [torch.min( + torch.stack([box1[:, :2], box2[:, :2],], dim=0), + dim=0, + keepdim=False, + )[0], + torch.max( + torch.stack([box1[:, 2:], box2[:, 2:],], dim=0), + dim=0, + keepdim=False, + )[0],] + , dim=1) + +def box_in_region(boxes, regions): + region_w = regions[:, 2] - regions[:, 0] + region_h = regions[:, 3] - regions[:, 1] + assert (region_w > 0).all() and (region_h > 0).all() + result = torch.zeros_like(regions) + result[:, 0::2] = (boxes[:, 0::2] - regions[:, 0::2]) / region_w + result[:, 1::2] = (boxes[:, 1::2] - regions[:, 1::2]) / region_h + return result + + +def apply_box_transform(anchors, transforms): + """ + Apply box transforms to a set of anchor boxes. + + Inputs: + - anchors: Anchor boxes of shape (N, 4), where each anchor is specified + in the form [xc, yc, w, h] + - transforms: Box transforms of shape (N, 4) where each transform is + specified as [tx, ty, tw, th] + + Returns: + - boxes: Transformed boxes of shape (N, 4) where each box is in the + format [xc, yc, w, h] + """ + # Unpack anchors + xa, ya = anchors[:, 0], anchors[:, 1] + wa, ha = anchors[:, 2], anchors[:, 3] + + # Unpack transforms + tx, ty = transforms[:, 0], transforms[:, 1] + tw, th = transforms[:, 2], transforms[:, 3] + + x = xa + tx * wa + y = ya + ty * ha + w = wa * tw.exp() + h = ha * th.exp() + + boxes = torch.stack([x, y, w, h], dim=1) + return boxes + + +def invert_box_transform(anchors, boxes): + """ + Compute the box transform that, when applied to anchors, would give boxes. + + Inputs: + - anchors: Box anchors of shape (N, 4) in the format [xc, yc, w, h] + - boxes: Target boxes of shape (N, 4) in the format [xc, yc, w, h] + + Returns: + - transforms: Box transforms of shape (N, 4) in the format [tx, ty, tw, th] + """ + # Unpack anchors + xa, ya = anchors[:, 0], anchors[:, 1] + wa, ha = anchors[:, 2], anchors[:, 3] + + # Unpack boxes + x, y = boxes[:, 0], boxes[:, 1] + w, h = boxes[:, 2], boxes[:, 3] + + tx = (x - xa) / wa + ty = (y - ya) / ha + tw = w.log() - wa.log() + th = h.log() - ha.log() + + transforms = torch.stack([tx, ty, tw, th], dim=1) + return transforms + + +def centers_to_extents(boxes): + """ + Convert boxes from [xc, yc, w, h] format to [x0, y0, x1, y1] format + + Input: + - boxes: Input boxes of shape (N, 4) in [xc, yc, w, h] format + + Returns: + - boxes: Output boxes of shape (N, 4) in [x0, y0, x1, y1] format + """ + xc, yc = boxes[:, 0], boxes[:, 1] + w, h = boxes[:, 2], boxes[:, 3] + + x0 = xc - w / 2 + x1 = x0 + w + y0 = yc - h / 2 + y1 = y0 + h + + boxes_out = torch.stack([x0, y0, x1, y1], dim=1) + return boxes_out + + +def extents_to_centers(boxes): + """ + Convert boxes from [x0, y0, x1, y1] format to [xc, yc, w, h] format + + Input: + - boxes: Input boxes of shape (N, 4) in [x0, y0, x1, y1] format + + Returns: + - boxes: Output boxes of shape (N, 4) in [xc, yc, w, h] format + """ + x0, y0 = boxes[:, 0], boxes[:, 1] + x1, y1 = boxes[:, 2], boxes[:, 3] + + xc = 0.5 * (x0 + x1) + yc = 0.5 * (y0 + y1) + w = x1 - x0 + h = y1 - y0 + + boxes_out = torch.stack([xc, yc, w, h], dim=1) + return boxes_out diff --git a/mlflow/sf2f/utils/common.py b/mlflow/sf2f/utils/common.py new file mode 100644 index 0000000..2248793 --- /dev/null +++ b/mlflow/sf2f/utils/common.py @@ -0,0 +1,192 @@ +import os +import time +import inspect +import subprocess +from contextlib import contextmanager + +import torch +from PIL import Image +import numpy as np +from collections import OrderedDict + +def int_tuple(s): + return tuple(int(i) for i in s.split(',')) + + +def float_tuple(s): + return tuple(float(i) for i in s.split(',')) + + +def str_tuple(s): + return tuple(s.split(',')) + + +def set_trainable(model, requires_grad): + set_trainable_param(model.parameters(), requires_grad) + +def set_trainable_param(parameters, requires_grad): + for param in parameters: + param.requires_grad = requires_grad + + +def update_values(dict_from, dict_to): + for key, value in dict_from.items(): + if isinstance(value, dict) and key in dict_to.keys(): + update_values(dict_from[key], dict_to[key]) + elif value is not None or key not in dict_to.keys(): + dict_to[key] = dict_from[key] + return dict_to + + +def params_count(model): + count = 0 + for p in model.parameters(): + c = 1 + for i in range(p.dim()): + c *= p.size(i) + count += c + return count + + +def bool_flag(s): + if s == '1': + return True + elif s == '0': + return False + msg = 'Invalid value "%s" for bool flag (should be 0 or 1)' + raise ValueError(msg % s) + + +def lineno(): + return inspect.currentframe().f_back.f_lineno + + +def get_gpu_memory(): + torch.cuda.synchronize() + opts = [ + 'nvidia-smi', '-q', '--gpu=' + str(0), '|', 'grep', '"Used GPU Memory"' + ] + cmd = str.join(' ', opts) + ps = subprocess.Popen( + cmd, shell=True, stdout=subprocess.PIPE, stderr=subprocess.STDOUT) + output = ps.communicate()[0].decode('utf-8') + output = output.split("\n")[1].split(":") + consumed_mem = int(output[1].strip().split(" ")[0]) + return consumed_mem + +# Converts a Tensor into an image array (numpy) +# |imtype|: the desired type of the converted numpy array + + +def tensor2im(input_image, imtype=np.uint8): + if not isinstance(input_image, torch.Tensor): + return input_image + image_numpy = input_image.numpy() + if image_numpy.ndim == 3: + if image_numpy.shape[0] == 1: + image_numpy = np.tile(image_numpy, (3, 1, 1)) + elif image_numpy.shape[0] == 3: + image_numpy = np.transpose(image_numpy, (1, 2, 0)) + elif image_numpy.ndim == 4: + if image_numpy.shape[1] == 1: + image_numpy = np.tile(image_numpy, (1, 3, 1, 1)) + elif image_numpy.shape[1] == 3: + image_numpy = np.transpose(image_numpy, (0, 2, 3, 1)) + else: + raise ValueError("Only support images or batches") + return image_numpy.astype(imtype) + + +def save_image(image_numpy, image_path): + image_pil = Image.fromarray(image_numpy) + image_pil.save(image_path) + + +def mkdirs(paths): + if isinstance(paths, list) and not isinstance(paths, str): + for path in paths: + mkdir(path) + else: + mkdir(paths) + + +def mkdir(path): + if not os.path.exists(path): + os.makedirs(path) + + +@contextmanager +def timeit(msg, should_time=True): + if should_time: + torch.cuda.synchronize() + t0 = time.time() + yield + if should_time: + torch.cuda.synchronize() + t1 = time.time() + duration = (t1 - t0) * 1000.0 + print('%s: %.2f ms' % (msg, duration)) + + +class LossManager(object): + def __init__(self): + self.total_loss = None + self.all_losses = {} + + def add_loss(self, loss, name, weight=1.0): + cur_loss = loss * weight + if self.total_loss is not None: + self.total_loss += cur_loss + else: + self.total_loss = cur_loss + + self.all_losses[name] = cur_loss.data.cpu().item() + + def items(self): + return self.all_losses.items() + + +def load_model_state(model, state_dict, strict=True): + # sometimes if the state_dict is saved from DataParallel model, the keys + # has a prefix "module.", so we remove them before Loading + new_state_dict = OrderedDict() + if not isinstance(model, torch.nn.DataParallel): + for k, v in state_dict.items(): + k = k[7:] if k.startswith("module") else k + new_state_dict[k] = v + else: + new_state_dict = state_dict + if strict: + # requires the model's state_dict to be exactly same as the new_state_dict + model.load_state_dict(new_state_dict) + else: + # only load parameters that are same size and type, and print warnings + # for parameters that don't match + own_state = model.state_dict() + for name, param in new_state_dict.items(): + if name in own_state: + if isinstance(param, torch.nn.Parameter): + # backwards compatibility for serialized parameters + param = param.data + if own_state[name].size() == param.size(): + own_state[name].copy_(param) + else: + print( + "Warning: While copying the parameter named {}, " + "whose dimensions in the model are {} and " + "whose dimensions in the checkpoint are {}.".format( + name, own_state[name].size(), param.size() + ) + ) + else: + print( + "Warning: Parameter named {} is not used " + "by this model.".format(name) + ) + for name, _ in own_state.items(): + if name not in new_state_dict: + print( + "Warning: Parameter named {} in the model " + "is not initialized.".format(name) + ) + return model diff --git a/mlflow/sf2f/utils/evaluate.py b/mlflow/sf2f/utils/evaluate.py new file mode 100644 index 0000000..81f27cc --- /dev/null +++ b/mlflow/sf2f/utils/evaluate.py @@ -0,0 +1,67 @@ +import pyprind +import numpy as np +import torch +import torch.nn.functional as F +import glog as log + +from datasets import imagenet_deprocess_batch +from scripts.compute_inception_score import get_inception_score +from scripts.compute_vggface_score import get_vggface_score + + +VGG_BOX = [0.235, 0.195, 0.765, 0.915] +def crop_vgg_box(imgs): + # with correct cropping & correct processing + left, top, right, bottom = VGG_BOX + # = [0.235015, 0.19505739, 0.76817876, 0.9154963] + N, C, H, W = imgs.shape + left = int(left * W) + right = int(right * W) + top = int(top * H) + bottom = int(bottom * H) + imgs = imgs[:, :, top:bottom+1, left:right+1] + return imgs + +def evaluate(model, data_loader, options): + ''' + Evaluate the current model... + ''' + normalize_method = options["data"]["data_opts"].get( + 'normalize_method', 'imagenet') + model.eval() + log.info("Evaluating with Inception Scores.") + images = [] + imgs_gt = [] + for iter, batch in enumerate(pyprind.prog_bar(data_loader, + title="[Generating Images]", + width=50)): + ######### unpack the data ######### + imgs, log_mels, human_ids = batch + imgs = imgs.cuda() + imgs_gt.append(imgs[0]) + ################################### + with torch.no_grad(): + float_dtype = torch.cuda.FloatTensor + log_mels = log_mels.type(float_dtype) + model_out = model(log_mels) + """ + imgs_pred: generated images + others: placeholder for other output + """ + imgs_pred, others = model_out + img = imagenet_deprocess_batch( + imgs_pred, normalize_method=normalize_method) + img = crop_vgg_box(img) + for i in range(img.shape[0]): + img_np = img[i].numpy().transpose(1, 2, 0) + images.append(img_np) + log.info("Computing inception scores...") + is_mean, is_std = get_inception_score(images) + log.info("Computing VF scores...") + vfs_mean, vfs_std = get_vggface_score(images) + + return is_mean, is_std, vfs_mean, vfs_std + + +if __name__ == '__main__': + evaluate() diff --git a/mlflow/sf2f/utils/evaluate_fid.py b/mlflow/sf2f/utils/evaluate_fid.py new file mode 100644 index 0000000..90da53d --- /dev/null +++ b/mlflow/sf2f/utils/evaluate_fid.py @@ -0,0 +1,82 @@ +import pyprind +import numpy as np +import torch +import torch.nn.functional as F +import glog as log + +from datasets import imagenet_deprocess_batch +from scripts.compute_fid_score import calculate_activation_statistics, calculate_frechet_distance +import scripts.compute_fid_score as inception_score +from scripts.compute_vggface_score import get_vggface_act + +VGG_BOX = [0.235, 0.195, 0.765, 0.915] +def crop_vgg_box(imgs): + # with correct cropping & correct processing + left, top, right, bottom = VGG_BOX + # = [0.235015, 0.19505739, 0.76817876, 0.9154963] + N, C, H, W = imgs.shape + left = int(left * W) + right = int(right * W) + top = int(top * H) + bottom = int(bottom * H) + imgs = imgs[:, :, top:bottom+1, left:right+1] + return imgs + +def evaluate_fid(model, data_loader, options): + ''' + Evaluate the current model... + ''' + normalize_method = options["data"]["data_opts"].get( + 'normalize_method', 'imagenet') + model.eval() + log.info("Evaluating with Inception Scores.") + images = [] + imgs_gt = [] + for iter, batch in enumerate(pyprind.prog_bar(data_loader, + title="[Generating Images]", + width=50)): + ######### unpack the data ######### + imgs, log_mels, human_ids = batch + imgs = imgs.cuda() + ################################### + with torch.no_grad(): + float_dtype = torch.cuda.FloatTensor + log_mels = log_mels.type(float_dtype) + model_out = model(log_mels) + """ + imgs_pred: generated images + others: placeholder for other output + """ + imgs_pred, others = model_out + img = imagenet_deprocess_batch( + imgs_pred, normalize_method=normalize_method) + img = crop_vgg_box(img) + for i in range(img.shape[0]): + img_np = img[i].numpy().transpose(1, 2, 0) + images.append(img_np) + + img = imagenet_deprocess_batch( + imgs, normalize_method=normalize_method) + img = crop_vgg_box(img) + for i in range(img.shape[0]): + img_np = img[i].numpy().transpose(1, 2, 0) + imgs_gt.append(img_np) + + log.info("Computing FID scores...") + acts_set = inception_score.get_fid_pred(imgs_gt) + fake_acts_set = inception_score.get_fid_pred(images) + real_mu, real_sigma = calculate_activation_statistics(acts_set) + fake_mu, fake_sigma = calculate_activation_statistics(fake_acts_set) + fid_score = calculate_frechet_distance(real_mu, real_sigma, fake_mu, fake_sigma) + + acts_set = get_vggface_act(imgs_gt) + fake_acts_set = get_vggface_act(images) + real_mu, real_sigma = calculate_activation_statistics(acts_set) + fake_mu, fake_sigma = calculate_activation_statistics(fake_acts_set) + fvfd_score = calculate_frechet_distance(real_mu, real_sigma, fake_mu, fake_sigma) + + return fid_score + + +if __name__ == '__main__': + evaluate() diff --git a/mlflow/sf2f/utils/filter_pickle.py b/mlflow/sf2f/utils/filter_pickle.py new file mode 100644 index 0000000..ccf5646 --- /dev/null +++ b/mlflow/sf2f/utils/filter_pickle.py @@ -0,0 +1,35 @@ +import pickle +import os + + +def check_pickle(mel_pickle): + """ + Load a mel pickle file and remove the file if it is empty + + Args: + mel_pickle: str, path to the mel pickle file + Returns: + None + """ + file = open(mel_pickle, 'rb') + try: + data = pickle.load(file) + except: + print('Error loading:', mel_pickle) + os.remove(mel_pickle) + + file.close() + +def filtering_pickle(dir_path): + for dir in os.listdir(dir_path): + data_path = os.path.join(dir_path, dir) + for id in os.listdir(data_path): + data = os.path.join(data_path, id) + check_pickle(data) + print("Filtering Done.") + + +# if __name__=="__main__": +# dir_path = './data/VoxCeleb/vox1/mel_spectrograms' # wav_convertor.vox1_mel +# filtering_pickle(dir_path) + diff --git a/mlflow/sf2f/utils/logger.py b/mlflow/sf2f/utils/logger.py new file mode 100644 index 0000000..c21f383 --- /dev/null +++ b/mlflow/sf2f/utils/logger.py @@ -0,0 +1,93 @@ +# Code referenced from https://gist.github.com/gyglim/1f8dfb1b5c82627ae3efcfbbadb9f514 +import tensorflow as tf +import numpy as np +# import scipy.misc +from PIL.Image import fromarray + +try: + from StringIO import StringIO # Python 2.7 +except ImportError: + from io import BytesIO # Python 3.x + +tf.compat.v1.disable_v2_behavior() + + +class Logger(object): + + def __init__(self, log_dir): + """Create a summary writer logging to log_dir.""" + self.writer = tf.compat.v1.summary.FileWriter(log_dir) + # self.writer = tf.compat.v1.summary.create_file_writer(log_dir) + + def scalar_summary(self, tag, value, step): + """Log a scalar variable.""" + summary = tf.compat.v1.Summary( + value=[tf.compat.v1.Summary.Value(tag=tag, simple_value=value)]) + self.writer.add_summary(summary, step) + + def image_summary(self, images, step, tag="vis"): + """Log a list of images.""" + + img_summaries = [] + if isinstance(images, list): + for i, img in enumerate(images): + _tag = '%s/%d' % (tag, i) + # Create a Summary value + img_summaries.append(self.add_image(img, _tag)) + elif isinstance(images, dict): + for key, img in images.items(): + _tag = '%s/%s' % (tag, key) + # Create a Summary value + img_summaries.append(self.add_image(img, _tag)) + else: + raise ValueError("Either a list or a dict is supported.") + + # Create and write Summary + summary = tf.compat.v1.Summary(value=img_summaries) + self.writer.add_summary(summary, step) + + def histo_summary(self, tag, values, step, bins=1000): + """Log a histogram of the tensor of values.""" + + # Create a histogram using numpy + counts, bin_edges = np.histogram(values, bins=bins) + + # Fill the fields of the histogram proto + hist = tf.HistogramProto() + hist.min = float(np.min(values)) + hist.max = float(np.max(values)) + hist.num = int(np.prod(values.shape)) + hist.sum = float(np.sum(values)) + hist.sum_squares = float(np.sum(values**2)) + + # Drop the start of the first bin + bin_edges = bin_edges[1:] + + # Add bin edges and counts + for edge in bin_edges: + hist.bucket_limit.append(edge) + for c in counts: + hist.bucket.append(c) + + # Create and write Summary + summary = tf.compat.v1.Summary( + value=[tf.compat.v1.Summary.Value(tag=tag, histo=hist)]) + self.writer.add_summary(summary, step) + self.writer.flush() + + def add_image(self, img, tag): + # Write the image to a string + try: + s = StringIO() + except: + s = BytesIO() + # scipy.misc.toimage(img).save(s, format="png") + fromarray(img).save(s, format="png") + + + # Create an Image object + img_sum = tf.compat.v1.Summary.Image(encoded_image_string=s.getvalue(), + height=img.shape[0], + width=img.shape[1]) + + return tf.compat.v1.Summary.Value(tag=tag, image=img_sum) diff --git a/mlflow/sf2f/utils/losses.py b/mlflow/sf2f/utils/losses.py new file mode 100644 index 0000000..55fcb70 --- /dev/null +++ b/mlflow/sf2f/utils/losses.py @@ -0,0 +1,144 @@ +import torch +import torch.nn.functional as F + + +def get_gan_losses(gan_type): + """ + Returns the generator and discriminator loss for a particular GAN type. + + The returned functions have the following API: + loss_g = g_loss(scores_fake) + loss_d = d_loss(scores_real, scores_fake) + """ + if gan_type == 'gan': + return gan_g_loss, gan_d_loss + elif gan_type == 'wgan': + return wgan_g_loss, wgan_d_loss + elif gan_type == 'lsgan': + return lsgan_g_loss, lsgan_d_loss + else: + raise ValueError('Unrecognized GAN type "%s"' % gan_type) + + +def bce_loss(input, target): + """ + Numerically stable version of the binary cross-entropy loss function. + + As per https://github.com/pytorch/pytorch/issues/751 + See the TensorFlow docs for a derivation of this formula: + https://www.tensorflow.org/api_docs/python/tf/nn/sigmoid_cross_entropy_with_logits + + Inputs: + - input: PyTorch Tensor of shape (N, ) giving scores. + - target: PyTorch Tensor of shape (N,) containing 0 and 1 giving targets. + + Returns: + - A PyTorch Tensor containing the mean BCE loss over the minibatch of + input data. + """ + neg_abs = -input.abs() + loss = input.clamp(min=0) - input * target + (1 + neg_abs.exp()).log() + return loss.mean() + + +def _make_targets(x, y): + """ + Inputs: + - x: PyTorch Tensor + - y: Python scalar + + Outputs: + - out: PyTorch Variable with same shape and dtype as x, but filled with y + """ + return torch.full_like(x, y) + + +def gan_g_loss(scores_fake): + """ + Input: + - scores_fake: Tensor of shape (N,) containing scores for fake samples + + Output: + - loss: Variable of shape (,) giving GAN generator loss + """ + if scores_fake.dim() > 1: + scores_fake = scores_fake.view(-1) + y_fake = _make_targets(scores_fake, 1) + return bce_loss(scores_fake, y_fake) + + +def gan_d_loss(scores_real, scores_fake): + """ + Input: + - scores_real: Tensor of shape (N,) giving scores for real samples + - scores_fake: Tensor of shape (N,) giving scores for fake samples + + Output: + - loss: Tensor of shape (,) giving GAN discriminator loss + """ + assert scores_real.size() == scores_fake.size() + if scores_real.dim() > 1: + scores_real = scores_real.view(-1) + scores_fake = scores_fake.view(-1) + y_real = _make_targets(scores_real, 1) + y_fake = _make_targets(scores_fake, 0) + loss_real = bce_loss(scores_real, y_real) + loss_fake = bce_loss(scores_fake, y_fake) + return loss_real + loss_fake + + +def wgan_g_loss(scores_fake): + """ + Input: + - scores_fake: Tensor of shape (N,) containing scores for fake samples + + Output: + - loss: Tensor of shape (,) giving WGAN generator loss + """ + return -scores_fake.mean() + + +def wgan_d_loss(scores_real, scores_fake): + """ + Input: + - scores_real: Tensor of shape (N,) giving scores for real samples + - scores_fake: Tensor of shape (N,) giving scores for fake samples + + Output: + - loss: Tensor of shape (,) giving WGAN discriminator loss + """ + return scores_fake.mean() - scores_real.mean() + + +def lsgan_g_loss(scores_fake): + if scores_fake.dim() > 1: + scores_fake = scores_fake.view(-1) + y_fake = _make_targets(scores_fake, 1) + return F.mse_loss(scores_fake.sigmoid(), y_fake) + + +def lsgan_d_loss(scores_real, scores_fake): + assert scores_real.size() == scores_fake.size() + if scores_real.dim() > 1: + scores_real = scores_real.view(-1) + scores_fake = scores_fake.view(-1) + y_real = _make_targets(scores_real, 1) + y_fake = _make_targets(scores_fake, 0) + loss_real = F.mse_loss(scores_real.sigmoid(), y_real) + loss_fake = F.mse_loss(scores_fake.sigmoid(), y_fake) + return loss_real + loss_fake + + +def gradient_penalty(x_real, x_fake, f, gamma=1.0): + N = x_real.size(0) + device, dtype = x_real.device, x_real.dtype + eps = torch.randn(N, 1, 1, 1, device=device, dtype=dtype) + x_hat = eps * x_real + (1 - eps) * x_fake + x_hat_score = f(x_hat) + if x_hat_score.dim() > 1: + x_hat_score = x_hat_score.view(x_hat_score.size(0), -1).mean(dim=1) + x_hat_score = x_hat_score.sum() + grad_x_hat, = torch.autograd.grad(x_hat_score, x_hat, create_graph=True) + grad_x_hat_norm = grad_x_hat.contiguous().view(N, -1).norm(p=2, dim=1) + gp_loss = (grad_x_hat_norm - gamma).pow(2).div(gamma * gamma).mean() + return gp_loss diff --git a/mlflow/sf2f/utils/metrics.py b/mlflow/sf2f/utils/metrics.py new file mode 100644 index 0000000..f337aaf --- /dev/null +++ b/mlflow/sf2f/utils/metrics.py @@ -0,0 +1,50 @@ +import torch + +def accuracy(output, target, topk=(1,)): + """Computes the accuracy over the k top predictions for the specified values of k""" + with torch.no_grad(): + maxk = max(topk) + batch_size = target.size(0) + + _, pred = output.topk(maxk, 1, True, True) + pred = pred.t() + correct = pred.eq(target.view(1, -1).expand_as(pred)) + + res = [] + for k in topk: + correct_k = correct[:k].view(-1).float().sum(0, keepdim=True) + res.append(correct_k.mul_(100.0 / batch_size).item()) + return res + + +def intersection(bbox_pred, bbox_gt): + max_xy = torch.min(bbox_pred[:, 2:], bbox_gt[:, 2:]) + min_xy = torch.max(bbox_pred[:, :2], bbox_gt[:, :2]) + inter = torch.clamp((max_xy - min_xy), min=0) + return inter[:, 0] * inter[:, 1] + + +def jaccard(bbox_pred, bbox_gt): + inter = intersection(bbox_pred, bbox_gt) + area_pred = (bbox_pred[:, 2] - bbox_pred[:, 0]) * (bbox_pred[:, 3] - + bbox_pred[:, 1]) + area_gt = (bbox_gt[:, 2] - bbox_gt[:, 0]) * (bbox_gt[:, 3] - + bbox_gt[:, 1]) + union = area_pred + area_gt - inter + iou = torch.div(inter, union) + return torch.sum(iou) + + +def get_total_norm(parameters, norm_type=2): + if norm_type == float('inf'): + total_norm = max(p.grad.data.abs().max() for p in parameters) + else: + total_norm = 0 + for p in parameters: + try: + param_norm = p.grad.data.norm(norm_type) + total_norm += param_norm ** norm_type + total_norm = total_norm ** (1. / norm_type) + except: + continue + return total_norm diff --git a/mlflow/sf2f/utils/mlflow_wandb.py b/mlflow/sf2f/utils/mlflow_wandb.py new file mode 100644 index 0000000..5ce4f04 --- /dev/null +++ b/mlflow/sf2f/utils/mlflow_wandb.py @@ -0,0 +1,15 @@ +import mlflow +import wandb + +def mlflow_wandb_logging(log_dict,model,signature,input_sample,sample_image): + #mlflow Log metric & model + mlflow.log_metrics(log_dict) + mlflow.pytorch.log_model( + pytorch_model=model, + artifact_path = "sf2f_pytorch", + signature= signature, + input_example = input_sample, + # pip_requirements = "rec.txt" + ) + log_dict['sample'] = [wandb.Image(sample_image, caption=f"Sample")] + wandb.log(log_dict) diff --git a/mlflow/sf2f/utils/pgan_utils.py b/mlflow/sf2f/utils/pgan_utils.py new file mode 100644 index 0000000..efb4e52 --- /dev/null +++ b/mlflow/sf2f/utils/pgan_utils.py @@ -0,0 +1,45 @@ +import os +import time +import json +import math +import torch + + +def num_flat_features(x): + size = x.size()[1:] # all dimensions except the batch dimension + num_features = 1 + for s in size: + num_features *= s + return num_features + + +def miniBatchStdDev(x, subGroupSize=4): + r""" + Add a minibatch standard deviation channel to the current layer. + In other words: + 1) Compute the standard deviation of the feature map over the minibatch + 2) Get the mean, over all pixels and all channels of thsi ValueError + 3) expand the layer and cocatenate it with the input + Args: + - x (tensor): previous layer + - subGroupSize (int): size of the mini-batches on which the standard deviation + should be computed + """ + size = x.size() + subGroupSize = min(size[0], subGroupSize) + if size[0] % subGroupSize != 0: + subGroupSize = size[0] + G = int(size[0] / subGroupSize) + if subGroupSize > 1: + y = x.view(-1, subGroupSize, size[1], size[2], size[3]) + y = torch.var(y, 1) + y = torch.sqrt(y + 1e-8) + y = y.view(G, -1) + y = torch.mean(y, 1).view(G, 1) + y = y.expand(G, size[2]*size[3]).view((G, 1, 1, size[2], size[3])) + y = y.expand(G, subGroupSize, -1, -1, -1) + y = y.contiguous().view((-1, 1, size[2], size[3])) + else: + y = torch.zeros(x.size(0), 1, x.size(2), x.size(3), device=x.device) + + return torch.cat([x, y], dim=1) diff --git a/mlflow/sf2f/utils/s2f_evaluator.py b/mlflow/sf2f/utils/s2f_evaluator.py new file mode 100644 index 0000000..9341cc0 --- /dev/null +++ b/mlflow/sf2f/utils/s2f_evaluator.py @@ -0,0 +1,449 @@ +''' +This file contains the utils for evaluation, and is used to cooperate with \ + training_utils.py and test.py +''' + + +import os +import json +import math +from collections import defaultdict +import time +from copy import deepcopy + +import numpy as np +import torch +import torch.optim as optim +import torch.nn as nn +import torch.nn.functional as F +from scipy.stats import entropy +from torchvision.models import inception_v3 +from pyprind import prog_bar +from tensorflow.io import gfile +from imageio import imwrite + +from datasets import fast_imagenet_deprocess_batch, imagenet_deprocess_batch +import datasets +import models +from models import InceptionResnetV1, fixed_image_standardization + + +# left, top, right, bottom +VGG_BOX = [0.235, 0.195, 0.765, 0.915] + + +class S2fEvaluator: + def __init__(self, + loader, + options, + nframe_range=None, + extraction_size=100, + hq_emb_dict=True, + face_gen_mode='naive', + facenet_return_pooling=False): + ''' + Inputs: + loader + options + ''' + self.loader = deepcopy(loader) + # This makes sure same group of meg_spectrograms is used for fuser + self.loader.dataset.shuffle_mel_segments = False + if nframe_range is not None: + self.loader.dataset.nframe_range = nframe_range + self.facenet = InceptionResnetV1( + pretrained='vggface2', + auto_input_resize=True, + return_pooling=facenet_return_pooling).cuda().eval() + self.float_dtype = torch.cuda.FloatTensor + self.long_dtype = torch.cuda.LongTensor + self.options = options + self.image_normalize_method= \ + self.options["data"]["data_opts"]["image_normalize_method"] + self.do_deprocess_and_preprocess = \ + self.options["eval"]["facenet"]["deprocess_and_preprocess"] + # crop faces according to VGG average bounding box + self.crop_faces = \ + self.options["eval"]["facenet"]["crop_faces"] + self.extraction_size = extraction_size + self.hq_emb_dict = hq_emb_dict + self.face_gen_mode = face_gen_mode + + self.get_dataset_embeddings() + + # Implement Evaluation Metrics + self.L2_dist = \ + nn.PairwiseDistance(p=2.0, eps=1e-06, keepdim=False) + self.L1_dist = \ + nn.PairwiseDistance(p=1.0, eps=1e-06, keepdim=False) + self.cos_sim = nn.CosineSimilarity(dim=-1, eps=1e-08) + + def deprocess_and_preprocess(self, imgs): + ''' + For a batch of real / generated image, perform 'imagenet' or 'standard' + deprocess, and FaceNet Preprocess + ''' + #print('Begin:', imgs[0]) + imgs = fast_imagenet_deprocess_batch( + imgs, + normalize_method=self.image_normalize_method) + #print('Our distribution:', imgs[0]) + imgs = fixed_image_standardization(imgs) + #print('fixed_image_standardization:', imgs[0]) + return imgs + + def crop_vgg_box(self, imgs): + # with correct cropping & correct processing + left, top, right, bottom = VGG_BOX + # = [0.235015, 0.19505739, 0.76817876, 0.9154963] + N, C, H, W = imgs.shape + left = int(left * W) + right = int(right * W) + top = int(top * H) + bottom = int(bottom * H) + imgs = imgs[:, :, top:bottom+1, left:right+1] + return imgs + + def get_dataset_embeddings(self): + with torch.no_grad(): + # To avoid influence to the running_mean/var of BatchNorm Layers + embedding_batches = [] + if self.hq_emb_dict: + for i in prog_bar( + range(len(self.loader.dataset)), + title="[S2fEvaluator: " + \ + "Preparing FaceNet Embedding Dictionary]", + width=50): + # Loop logic + ######### unpack the data ######### + imgs = self.loader.dataset.get_all_faces_of_id(i) + imgs = imgs.cuda() + ################################### + if self.do_deprocess_and_preprocess: + imgs = self.deprocess_and_preprocess(imgs) + if self.crop_faces: + imgs = self.crop_vgg_box(imgs) + embeddings = self.facenet(imgs) + embeddings = torch.mean(embeddings, 0, keepdim=True) + embedding_batches.append(embeddings) + else: + for batch in prog_bar( + self.loader, + title="[S2fEvaluator: " + \ + "Preparing FaceNet Embedding Dictionary]", + width=50): + # Loop logic + ######### unpack the data ######### + imgs, log_mels, human_ids = batch + imgs = imgs.cuda() + ################################### + if self.do_deprocess_and_preprocess: + imgs = self.deprocess_and_preprocess(imgs) + if self.crop_faces: + imgs = self.crop_vgg_box(imgs) + embeddings = self.facenet(imgs) + embedding_batches.append(embeddings) + self.dataset_embedding = torch.cat(embedding_batches, 0) + print("S2fEvaluator: dataset_embedding shape:", + self.dataset_embedding.shape) + + def get_pred_img_embeddings(self, model): + ''' + This function focus on evaluating the speech-to-face specific metrics, + Including: + 1. Deep feature (FaceNet Feature) Similarity + 2. Extraction Recall@K + 3. Landmark Correlation + ''' + training_status = model.training + model.eval() + pred_img_embedding_batches = [] + with torch.no_grad(): + # To avoid influence to the running_mean/var of BatchNorm Layers + if self.face_gen_mode == 'naive': + for batch in prog_bar( + self.loader, + title="[S2fEvaluator: " + \ + "Getting FaceNet Embedding for Predicted Images]", + width=50): + # Loop logic + ######### unpack the data ######### + imgs, log_mels, human_ids = batch + imgs = imgs.cuda() + log_mels = log_mels.type(self.float_dtype) + human_ids = human_ids.type(self.long_dtype) + ################################### + # Run the model as it has been run during training + model_out = model(log_mels) + imgs_pred, others = model_out + # Multi-Reso Case + if isinstance(imgs_pred, tuple): + imgs_pred = imgs_pred[-1] + if self.do_deprocess_and_preprocess: + imgs_pred = self.deprocess_and_preprocess(imgs_pred) + if self.crop_faces: + imgs_pred = self.crop_vgg_box(imgs_pred) + pred_img_embeddings = self.facenet(imgs_pred) + pred_img_embedding_batches.append(pred_img_embeddings) + elif self.face_gen_mode == 'average_facenet_embedding': + for i in prog_bar( + range(len(self.loader.dataset)), + title="[S2fEvaluator: " + \ + "Getting FaceNet Embedding for Predicted Images, " + \ + "with average Facenet embedding policy]", + width=50): + # Loop logic + ######### unpack the data ######### + log_mels = self.loader.dataset.get_all_mel_segments_of_id(i) + log_mels = log_mels.type(self.float_dtype) + ################################### + # Run the model as it has been run during training + model_out = model(log_mels) + imgs_pred, others = model_out + if isinstance(imgs_pred, tuple): + imgs_pred = imgs_pred[-1] + if self.do_deprocess_and_preprocess: + imgs_pred = self.deprocess_and_preprocess(imgs_pred) + if self.crop_faces: + imgs_pred = self.crop_vgg_box(imgs_pred) + pred_img_embeddings = self.facenet(imgs_pred) + pred_img_embeddings = torch.mean( + pred_img_embeddings, 0, keepdim=True) + pred_img_embedding_batches.append(pred_img_embeddings) + elif self.face_gen_mode == 'average_voice_embedding': + for i in prog_bar( + range(len(self.loader.dataset)), + title="[S2fEvaluator: " + \ + "Getting FaceNet Embedding for Predicted Images, " + \ + "with average voice embedding policy]", + width=50): + # Loop logic + ######### unpack the data ######### + log_mels = self.loader.dataset.get_all_mel_segments_of_id(i) + log_mels = log_mels.type(self.float_dtype) + ################################### + # Run the model as it has been run during training + model_out = model(log_mels, average_voice_embedding=True) + imgs_pred, others = model_out + if isinstance(imgs_pred, tuple): + imgs_pred = imgs_pred[-1] + if self.do_deprocess_and_preprocess: + imgs_pred = self.deprocess_and_preprocess(imgs_pred) + if self.crop_faces: + imgs_pred = self.crop_vgg_box(imgs_pred) + pred_img_embeddings = self.facenet(imgs_pred) + #pred_img_embeddings = torch.mean( + # pred_img_embeddings, 0, keepdim=True) + pred_img_embedding_batches.append(pred_img_embeddings) + pred_img_embedding = torch.cat(pred_img_embedding_batches, 0) + #print("S2fEvaluator: pred_img_embedding shape:", + # pred_img_embedding.shape) + model.train(mode=training_status) + + return pred_img_embedding + + def L1_query(self, x, y): + ''' + Given x, extract top K similar features from y, based on L1 distance + x in shape (N_x, D) + y in shape (N_y, D) + ''' + # (N_x, D) --> (N_x, 1, D) + x = x.unsqueeze(1) + # Initialize: (N_x, ) + x_ids = torch.tensor(np.arange(x.shape[0])).cpu() + # (N_y, D) --> (1, N_y, D) + y = y.unsqueeze(0) + # Output: (N_x, N_y, D) + L1_table = torch.abs(x - y) + # (N_x, N_y, D) --> (N_x, N_y) + L1_table = torch.mean(L1_table, dim=-1) + L1_table = torch.neg(L1_table) + # Top K: (N_x, K) + top_1_vals, top_1_indices = torch.topk(L1_table, 1, dim=-1) + top_5_vals, top_5_indices = torch.topk(L1_table, 5, dim=-1) + top_10_vals, top_10_indices = torch.topk(L1_table, 10, dim=-1) + top_50_vals, top_50_indices = torch.topk(L1_table, 50, dim=-1) + + recall_at_1 = self.in_top_k(top_1_indices.cpu(), x_ids) + recall_at_5 = self.in_top_k(top_5_indices.cpu(), x_ids) + recall_at_10 = self.in_top_k(top_10_indices.cpu(), x_ids) + recall_at_50 = self.in_top_k(top_50_indices.cpu(), x_ids) + + return recall_at_1, recall_at_5, recall_at_10, recall_at_50 + + def cos_query(self, x, y): + ''' + Given x, extract top K similar features from y, based on L1 distance + x in shape (N_x, D) + y in shape (N_y, D) + ''' + # (N_x, D) --> (N_x, 1, D) + x = x.unsqueeze(1) + # Initialize: (N_x, ) + x_ids = torch.tensor(np.arange(x.shape[0])).cpu() + # (N_y, D) --> (1, N_y, D) + y = y.unsqueeze(0) + # Output: (N_x, N_y) + cos_table = self.cos_sim(x, y) + + # Top K: (N_x, K) + top_1_vals, top_1_indices = torch.topk(cos_table, 1, dim=-1) + top_2_vals, top_2_indices = torch.topk(cos_table, 2, dim=-1) + top_5_vals, top_5_indices = torch.topk(cos_table, 5, dim=-1) + top_10_vals, top_10_indices = torch.topk(cos_table, 10, dim=-1) + top_20_vals, top_20_indices = torch.topk(cos_table, 20, dim=-1) + top_50_vals, top_50_indices = torch.topk(cos_table, 50, dim=-1) + + recall_at_1 = self.in_top_k(top_1_indices.cpu(), x_ids) + recall_at_2 = self.in_top_k(top_2_indices.cpu(), x_ids) + recall_at_5 = self.in_top_k(top_5_indices.cpu(), x_ids) + recall_at_10 = self.in_top_k(top_10_indices.cpu(), x_ids) + recall_at_20 = self.in_top_k(top_20_indices.cpu(), x_ids) + recall_at_50 = self.in_top_k(top_50_indices.cpu(), x_ids) + + recall_tuple = (recall_at_1, recall_at_2, recall_at_5, recall_at_10, \ + recall_at_20, recall_at_50) + + return recall_tuple + + def cal_ih_sim(self, x, y): + ''' + For an existing face embedding distribution, calculate the inter-human + similarity + + Arguments: + x in shape (N_x, D) + x in shape (N_y, D) + ''' + # (N_x, D) --> (N_x, 1, D) + y = y.unsqueeze(0) + x = x.unsqueeze(1) + # Output: (N_x, N_y) + cos_table = self.cos_sim(x, y) + cos_table = cos_table.detach().cpu().numpy() + ih_sum = 0.0 + for i in range(cos_table.shape[0]): + for j in range(cos_table.shape[1]): + if i != j: + ih_sum = ih_sum + cos_table[i, j] + ih_sim = ih_sum / float(cos_table.shape[0] * (cos_table.shape[1] - 1)) + return ih_sim + + def in_top_k(self, top_k_indices, gt_labels): + results = [] + for i, top_k_id in enumerate(top_k_indices): + gt_label = gt_labels[i] + #print(gt_label, top_k_id) + if gt_label in top_k_id: + results.append(1.0) + else: + results.append(0.0) + #print('results:', results) + return np.mean(results) + + def get_metrics(self, model, recall_method='cos_sim', get_ih_sim=False): + ''' + Arguments: + recall_method: the similarity metric to use for retrival + get_ih_sim: get the inter-human similarity + ''' + pred_img_embedding = self.get_pred_img_embeddings(model) + L2_dist = self.L2_dist(self.dataset_embedding, pred_img_embedding) + L2_dist = torch.mean(L2_dist).item() + L1_dist = self.L1_dist(self.dataset_embedding, pred_img_embedding) + L1_dist = torch.mean(L1_dist).item() + cos_sim = self.cos_sim(self.dataset_embedding, pred_img_embedding) + cos_sim = torch.mean(cos_sim).item() + + # We might use only 100 id for extraction + if self.extraction_size is None: + pred_emb_to_use = pred_img_embedding + data_emb_to_use = self.dataset_embedding + elif isinstance(self.extraction_size, list): + pred_emb_to_use = [pred_img_embedding[0:self.extraction_size[0]], \ + pred_img_embedding[self.extraction_size[0]:self.extraction_size[1]], \ + pred_img_embedding[self.extraction_size[1]:self.extraction_size[2]]] + data_emb_to_use = [self.dataset_embedding[0:self.extraction_size[0]], \ + self.dataset_embedding[self.extraction_size[0]:self.extraction_size[1]], \ + self.dataset_embedding[self.extraction_size[1]:self.extraction_size[2]]] + else: + pred_emb_to_use = pred_img_embedding[0:self.extraction_size] + data_emb_to_use = self.dataset_embedding[0:self.extraction_size] + + if recall_method == 'L1': + if isinstance(pred_emb_to_use, list): + recall_temp = [] + for i, pred_emb in enumerate(pred_emb_to_use): + recall_temp.append(self.L1_query(pred_emb, data_emb_to_use[i])) + recall_tuple = tuple(np.mean(np.array(recall_temp), axis=0)) + else: + recall_tuple = self.L1_query(pred_emb_to_use, data_emb_to_use) + elif recall_method == 'cos_sim': + if isinstance(pred_emb_to_use, list): + recall_temp = [] + for i, pred_emb in enumerate(pred_emb_to_use): + recall_temp.append(self.cos_query(pred_emb, data_emb_to_use[i])) + recall_tuple = tuple(np.mean(np.array(recall_temp), axis=0)) + else: + recall_tuple = self.cos_query(pred_emb_to_use, data_emb_to_use) + + if get_ih_sim: + ih_sim = self.cal_ih_sim(pred_img_embedding, self.dataset_embedding) + return L2_dist, L1_dist, cos_sim, recall_tuple, ih_sim + else: + return L2_dist, L1_dist, cos_sim, recall_tuple + + def get_faces_from_different_segments(self, model, output_dir): + #gfile.MkDir(output_dir) + os.makedirs(output_dir, exist_ok=True) + temp_loader = deepcopy(self.loader) + for i in prog_bar( + range(len(self.loader.dataset)), + title="[S2fEvaluator: " + \ + "Generating Faces from Different Speech Segments]", + width=50): + ######### unpack the data ######### + imgs = temp_loader.dataset.get_all_faces_of_id(i) + log_mels = temp_loader.dataset.get_all_mel_segments_of_id(i) + imgs = imgs.cuda() + log_mels = log_mels.type(self.float_dtype) #cuda() + #human_ids = human_ids.type(long_dtype) + ################################### + with torch.no_grad(): + model_out = model(log_mels) + imgs_pred, _ = model_out + if isinstance(imgs_pred, tuple): + imgs_pred = imgs_pred[-1] + imgs = imagenet_deprocess_batch( + imgs, normalize_method=self.image_normalize_method) + imgs_pred = imagenet_deprocess_batch( + imgs_pred, normalize_method=self.image_normalize_method) + #print(imgs.shape, imgs_pred.shape) + identity_dir = os.path.join(output_dir, str(i)) + gfile.MkDir(identity_dir) + + for j in range(imgs.shape[0]): + img_np = imgs[j].numpy().transpose(1, 2, 0) + img_path = os.path.join(identity_dir, 'origin_%d.png' % j) + imwrite(img_path, img_np) + + for k in range(imgs_pred.shape[0]): + img_np = imgs_pred[k].numpy().transpose(1, 2, 0) + img_path = os.path.join(identity_dir, 'pred_%d.png' % k) + imwrite(img_path, img_np) + + def L2_distances(self, x, y=None): + ''' + Input: x is a Nxd matrix + y is an optional Mxd matirx + Output: dist is a NxM matrix where dist[i,j] is the square norm between x[i,:] and y[j,:] + if y is not given then use 'y=x'. + i.e. dist[i,j] = ||x[i,:]-y[j,:]||^2 + ''' + if y is not None: + differences = x.unsqueeze(1) - y.unsqueeze(0) + else: + differences = x.unsqueeze(1) - x.unsqueeze(0) + distances = torch.sum(differences * differences, -1) + return distances diff --git a/mlflow/sf2f/utils/training_utils.py b/mlflow/sf2f/utils/training_utils.py new file mode 100644 index 0000000..b917b7f --- /dev/null +++ b/mlflow/sf2f/utils/training_utils.py @@ -0,0 +1,250 @@ +import os +import json +import math +from collections import defaultdict +import time + +import numpy as np +import torch +import torch.optim as optim +import torch.nn as nn +import torch.nn.functional as F +from scipy.stats import entropy +from torchvision.models import inception_v3 +from pyprind import prog_bar + +from datasets import imagenet_deprocess_batch, fast_mel_deprocess_batch +import datasets +import models +from utils.metrics import jaccard +from utils import tensor2im +from utils.visualization.plot import get_np_plot, plot_mel_spectrogram, \ + plot_attention + +from scripts.compute_vggface_score import get_vggface_score + + +def add_loss_with_tensor(total_loss, + curr_loss, + loss_dict, + loss_name, + weight=1): + curr_loss = curr_loss * weight + loss_dict[loss_name] = curr_loss + if total_loss is not None: + total_loss += curr_loss + else: + total_loss = curr_loss + return total_loss + +def add_loss(total_loss, curr_loss, loss_dict, loss_name, weight=1): + curr_loss = curr_loss * weight + loss_dict[loss_name] = curr_loss.item() + if total_loss is not None: + total_loss += curr_loss + else: + total_loss = curr_loss + return total_loss + + +def visualize_sample(model, + imgs, + log_mels, + image_normalize_method, + mel_normalize_method='vox_mel', + visualize_attn=False): + ''' + Prepare data for tensorboard + ''' + samples = [] + # add the ground-truth images + samples.append(imgs[:1]) + + with torch.no_grad(): + model_out = model(log_mels) + imgs_pred, others = model_out + # add the reconstructed images + if isinstance(imgs_pred, tuple): + if len(imgs_pred) >= 2: + low_imgs_pred = imgs_pred[0] + # Rescale + while low_imgs_pred.size()[2] != imgs_pred[-1].size()[2]: + low_imgs_pred = F.interpolate( + low_imgs_pred, scale_factor=2, mode='nearest') + samples.append(low_imgs_pred[:1]) + if len(imgs_pred) == 3: + mid_imgs_pred = imgs_pred[1] + while mid_imgs_pred.size()[2] != imgs_pred[-1].size()[2]: + mid_imgs_pred = F.interpolate( + mid_imgs_pred, scale_factor=2, mode='nearest') + samples.append(mid_imgs_pred[:1]) + imgs_pred = imgs_pred[-1] + samples.append(imgs_pred[:1]) + + log_mels_de = fast_mel_deprocess_batch(log_mels, mel_normalize_method) + log_mel = log_mels_de[0].cpu().detach().numpy() + + if visualize_attn: + attn_weights = model.module.decoder.return_attn_weights() + attn_weight = attn_weights[0].cpu().detach().numpy() + + samples = torch.cat(samples, dim=3) + samples = { + "samples": tensor2im( + imagenet_deprocess_batch( + samples, + rescale=False, + normalize_method=image_normalize_method + ).squeeze(0) + ), + #"mel_spectrogram": get_np_plot( + # plot_mel_spectrogram( + # log_mel) + #) + } + if visualize_attn: + samples['attention_weights'] = get_np_plot( + plot_attention(attn_weight)) + + # Draw Scene Graphs + #sg_array = draw_scene_graph(objs[obj_to_img == 0], + # triples[triple_to_img == 0], + # vocab=vocab) + #samples["scene_graph"] = sg_array + + return samples + + +def check_model(args, + options, + t, + loader, + model): + training_status = model.training + model.eval() + float_dtype = torch.cuda.FloatTensor + long_dtype = torch.cuda.LongTensor + num_samples = 0 + all_losses = defaultdict(list) + inception_module = nn.DataParallel( + inception_v3( + pretrained=True, + transform_input=False).cuda()) + inception_module.eval() + preds = [] + images = [] + with torch.no_grad(): + # To avoid influence to the running_mean/var of BatchNorm Layers + for batch in prog_bar( + loader, + title="[Validating Inception Score and Pixel Loss]", + width=50): + # Loop logic + ######### unpack the data ######### + imgs, log_mels, human_ids = batch + imgs = imgs.cuda() + log_mels = log_mels.type(float_dtype) + human_ids = human_ids.type(long_dtype) + ################################### + # Run the model as it has been run during training + model_out = model(log_mels) + imgs_pred, others = model_out + + skip_pixel_loss = False + total_loss, losses = calculate_model_losses( + options["optim"], + skip_pixel_loss, + imgs, + imgs_pred, + get_item=True) + + if isinstance(imgs_pred, tuple): + imgs_pred = imgs_pred[-1] + image = imagenet_deprocess_batch( + imgs_pred, + normalize_method=options["data"]["data_opts"]["image_normalize_method"]) + for i in range(image.shape[0]): + img_np = image[i].numpy().transpose(1, 2, 0) + images.append(img_np) + + # check inception scores + x = F.interpolate(imgs_pred, (299, 299), mode="bilinear") + x = inception_module(x) + preds.append(F.softmax(x).cpu().numpy()) + + for loss_name, loss_val in losses.items(): + all_losses[loss_name].append(loss_val) + num_samples += imgs.size(0) + + samples = visualize_sample( + model, + imgs, + log_mels, + options["data"]["data_opts"]["image_normalize_method"], + visualize_attn=options['eval'].get('visualize_attn', False)) + mean_losses = {k: np.mean(v) for k, v in all_losses.items()} + + # calculate the inception scores + splits = 5 + preds = np.concatenate(preds, axis=0) + # Now compute the mean kl-div + split_scores = [] + N = preds.shape[0] + for k in range(splits): + part = preds[k * (N // splits): (k+1) * (N // splits), :] + py = np.mean(part, axis=0) + scores = [] + for i in range(part.shape[0]): + pyx = part[i, :] + scores.append(entropy(pyx, py)) + split_scores.append(np.exp(np.mean(scores))) + inception_score = (np.mean(split_scores), np.std(split_scores)) + + vf_score = get_vggface_score(images) + + out = [mean_losses, samples, inception_score, vf_score] + model.train(mode=training_status) + return tuple(out) + + +def calculate_model_losses(opts, + skip_pixel_loss, + imgs, + imgs_pred, + get_item=False): + if get_item: + add_loss_fn = add_loss + else: + add_loss_fn = add_loss_with_tensor + total_loss = torch.zeros(1).to(imgs) + losses = {} + + l1_pixel_weight = opts["l1_pixel_loss_weight"] + if skip_pixel_loss: + l1_pixel_weight = 0 + + if isinstance(imgs_pred, tuple): + # Multi-Resolution Pixel Loss + for i, img_pred in enumerate(imgs_pred): + loss_name = 'L1_pixel_loss_%d' % i + img = imgs + while img.size()[2] != img_pred.size()[2]: + img = F.interpolate(img, scale_factor=0.5, mode='nearest') + l1_pixel_loss = F.l1_loss(img_pred, img) + total_loss = add_loss_fn( + total_loss, + l1_pixel_loss, + losses, + loss_name, + l1_pixel_weight) + else: + # Single Resolution + l1_pixel_loss = F.l1_loss(imgs_pred, imgs) + total_loss = add_loss_fn( + total_loss, + l1_pixel_loss, + losses, + 'L1_pixel_loss', + l1_pixel_weight) + + return total_loss, losses diff --git a/mlflow/sf2f/utils/utils.py b/mlflow/sf2f/utils/utils.py new file mode 100644 index 0000000..0157a82 --- /dev/null +++ b/mlflow/sf2f/utils/utils.py @@ -0,0 +1,89 @@ +import time +import inspect +import subprocess +from contextlib import contextmanager + +import torch +from torch import nn + +def int_tuple(s): + return tuple(int(i) for i in s.split(',')) + + +def float_tuple(s): + return tuple(float(i) for i in s.split(',')) + + +def str_tuple(s): + return tuple(s.split(',')) + + +def bool_flag(s): + if s == '1': + return True + elif s == '0': + return False + msg = 'Invalid value "%s" for bool flag (should be 0 or 1)' + raise ValueError(msg % s) + + +def lineno(): + return inspect.currentframe().f_back.f_lineno + + +def get_gpu_memory(): + torch.cuda.synchronize() + opts = [ + 'nvidia-smi', '-q', '--gpu=' + str(0), '|', 'grep', '"Used GPU Memory"' + ] + cmd = str.join(' ', opts) + ps = subprocess.Popen( + cmd, shell=True, stdout=subprocess.PIPE, stderr=subprocess.STDOUT) + output = ps.communicate()[0].decode('utf-8') + output = output.split("\n")[1].split(":") + consumed_mem = int(output[1].strip().split(" ")[0]) + return consumed_mem + + +@contextmanager +def timeit(msg, should_time=True): + if should_time: + torch.cuda.synchronize() + t0 = time.time() + yield + if should_time: + torch.cuda.synchronize() + t1 = time.time() + duration = (t1 - t0) * 1000.0 + print('%s: %.2f ms' % (msg, duration)) + + +class LossManager(object): + def __init__(self): + self.total_loss = None + self.all_losses = {} + + def add_loss(self, loss, name, weight=1.0): + cur_loss = loss * weight + if self.total_loss is not None: + self.total_loss += cur_loss + else: + self.total_loss = cur_loss + + self.all_losses[name] = cur_loss.data.cpu().item() + + def items(self): + return self.all_losses.items() + + +def load_my_state_dict(model, state_dict): + print('Loading my state_dict') + own_state = model.state_dict() + for name, param in state_dict.items(): + if name not in own_state: + continue + if isinstance(param, nn.Parameter): + # backwards compatibility for serialized parameters + param = param.data + own_state[name].copy_(param) + print('{} Loaded!'.format(name)) diff --git a/mlflow/sf2f/utils/vad_ex.py b/mlflow/sf2f/utils/vad_ex.py new file mode 100644 index 0000000..f8b489e --- /dev/null +++ b/mlflow/sf2f/utils/vad_ex.py @@ -0,0 +1,174 @@ +import collections +import contextlib +import sys +import wave +from pydub import AudioSegment +import webrtcvad + +# Modified https://github.com/wiseman/py-webrtcvad/blob/master/example.py + +def read_wave(path): + """Reads a .wav file. + Takes the path, and returns (PCM audio data, sample rate). + """ + with contextlib.closing(wave.open(path, 'rb')) as wf: + num_channels = wf.getnchannels() + #assert num_channels == 1 + sample_width = wf.getsampwidth() + assert sample_width == 2 + sample_rate = wf.getframerate() + #assert sample_rate in (8000, 16000, 32000) + pcm_data = wf.readframes(wf.getnframes()) + return pcm_data, sample_rate + +def read_libri(path): + mf = AudioSegment.from_file(path, "wav") + sample_rate = mf.frame_rate + pcm_data = mf.raw_data + return pcm_data, sample_rate + + +def read_m4a(path): + mf = AudioSegment.from_file(path, "m4a") + sample_rate = mf.frame_rate + pcm_data = mf.raw_data + return pcm_data, sample_rate + + +def write_wave(path, audio, sample_rate): + """Writes a .wav file. + Takes path, PCM audio data, and sample rate. + """ + with contextlib.closing(wave.open(path, 'wb')) as wf: + wf.setnchannels(1) + wf.setsampwidth(2) + wf.setframerate(sample_rate) + wf.writeframes(audio) + + +class Frame(object): + """Represents a "frame" of audio data.""" + def __init__(self, bytes, timestamp, duration): + self.bytes = bytes + self.timestamp = timestamp + self.duration = duration + + +def frame_generator(frame_duration_ms, audio, sample_rate): + """Generates audio frames from PCM audio data. + Takes the desired frame duration in milliseconds, the PCM data, and + the sample rate. + Yields Frames of the requested duration. + """ + n = int(sample_rate * (frame_duration_ms / 1000.0) * 2) + offset = 0 + timestamp = 0.0 + duration = (float(n) / sample_rate) / 2.0 + while offset + n < len(audio): + yield Frame(audio[offset:offset + n], timestamp, duration) + timestamp += duration + offset += n + + +def vad_collector(sample_rate, frame_duration_ms, + padding_duration_ms, vad, frames): + """Filters out non-voiced audio frames. + Given a webrtcvad.Vad and a source of audio frames, yields only + the voiced audio. + Uses a padded, sliding window algorithm over the audio frames. + When more than 90% of the frames in the window are voiced (as + reported by the VAD), the collector triggers and begins yielding + audio frames. Then the collector waits until 90% of the frames in + the window are unvoiced to detrigger. + The window is padded at the front and back to provide a small + amount of silence or the beginnings/endings of speech around the + voiced frames. + Arguments: + sample_rate - The audio sample rate, in Hz. + frame_duration_ms - The frame duration in milliseconds. + padding_duration_ms - The amount to pad the window, in milliseconds. + vad - An instance of webrtcvad.Vad. + frames - a source of audio frames (sequence or generator). + Returns: A generator that yields PCM audio data. + """ + num_padding_frames = int(padding_duration_ms / frame_duration_ms) + # We use a deque for our sliding window/ring buffer. + ring_buffer = collections.deque(maxlen=num_padding_frames) + # We have two states: TRIGGERED and NOTTRIGGERED. We start in the + # NOTTRIGGERED state. + triggered = False + + voiced_frames = [] + for frame in frames: + is_speech = vad.is_speech(frame.bytes, sample_rate) + + #sys.stdout.write('1' if is_speech else '0') + if not triggered: + ring_buffer.append((frame, is_speech)) + num_voiced = len([f for f, speech in ring_buffer if speech]) + # If we're NOTTRIGGERED and more than 90% of the frames in + # the ring buffer are voiced frames, then enter the + # TRIGGERED state. + if num_voiced > 0.9 * ring_buffer.maxlen: + triggered = True + #sys.stdout.write('+(%s)' % (ring_buffer[0][0].timestamp,)) + # We want to yield all the audio we see from now until + # we are NOTTRIGGERED, but we have to start with the + # audio that's already in the ring buffer. + for f, s in ring_buffer: + voiced_frames.append(f) + ring_buffer.clear() + else: + # We're in the TRIGGERED state, so collect the audio data + # and add it to the ring buffer. + voiced_frames.append(frame) + ring_buffer.append((frame, is_speech)) + num_unvoiced = len([f for f, speech in ring_buffer if not speech]) + # If more than 90% of the frames in the ring buffer are + # unvoiced, then enter NOTTRIGGERED and yield whatever + # audio we've collected. + if num_unvoiced > 0.9 * ring_buffer.maxlen: + #sys.stdout.write('-(%s)' % (frame.timestamp + frame.duration)) + triggered = False + yield b''.join([f.bytes for f in voiced_frames]) + ring_buffer.clear() + voiced_frames = [] + #if triggered: + # sys.stdout.write('-(%s)' % (frame.timestamp + frame.duration)) + #sys.stdout.write('\n') + # If we have any leftover voiced audio when we run out of input, + # yield it. + if voiced_frames: + yield b''.join([f.bytes for f in voiced_frames]) + + +def main(args): + if len(args) != 2: + sys.stderr.write( + 'Usage: example.py \n') + sys.exit(1) + audio, sample_rate = read_wave(args[1]) + #audio, sample_rate = read_m4a(args[1]) + #audio, sample_rate = read_libri(args[1]) + vad = webrtcvad.Vad(int(args[0])) + frames = frame_generator(30, audio, sample_rate) + frames = list(frames) + segments = vad_collector(sample_rate, 30, 300, vad, frames) + total_wav = b"" + for i, segment in enumerate(segments): + total_wav += segment + + + path = 'test.wav' + write_wave(path, total_wav, sample_rate) + + +if __name__ == '__main__': + #main(sys.argv[1:]) + import os + WAV_DIR = os.path.join('./data', 'VoxCeleb', 'raw_wav') + wav_path = 'vox1/test/id10270/zjwijMp0Qyw/00001.wav' + wav_path = os.path.join(WAV_DIR, wav_path) + + audio, sample_rate = read_wave(wav_path) + print(audio, sample_rate) diff --git a/mlflow/sf2f/utils/visualization/__init__.py b/mlflow/sf2f/utils/visualization/__init__.py new file mode 100644 index 0000000..2b54423 --- /dev/null +++ b/mlflow/sf2f/utils/visualization/__init__.py @@ -0,0 +1,2 @@ +from .vis import * +# from .visualizer import Visualizer diff --git a/mlflow/sf2f/utils/visualization/html.py b/mlflow/sf2f/utils/visualization/html.py new file mode 100644 index 0000000..1e71687 --- /dev/null +++ b/mlflow/sf2f/utils/visualization/html.py @@ -0,0 +1,65 @@ +import dominate +from dominate.tags import * +import os + + +class HTML: + def __init__(self, web_dir, title, reflesh=0): + self.title = title + self.web_dir = web_dir + self.img_dir = os.path.join(self.web_dir, 'images') + if not os.path.exists(self.web_dir): + os.makedirs(self.web_dir) + if not os.path.exists(self.img_dir): + os.makedirs(self.img_dir) + # print(self.img_dir) + + self.doc = dominate.document(title=title) + if reflesh > 0: + with self.doc.head: + meta(http_equiv="reflesh", content=str(reflesh)) + + def get_image_dir(self): + return self.img_dir + + def add_header(self, str): + with self.doc: + h3(str) + + def add_table(self, border=1): + self.t = table(border=border, style="table-layout: fixed;") + self.doc.add(self.t) + + def add_images(self, ims, txts, links, width=400): + self.add_table() + with self.t: + with tr(): + for im, txt, link in zip(ims, txts, links): + with td(style="word-wrap: break-word;", halign="center", valign="top"): + with p(): + with a(href=os.path.join('images', link)): + img(style="width:%dpx" % + width, src=os.path.join('images', im)) + br() + p(txt) + + def save(self): + html_file = '%s/index.html' % self.web_dir + f = open(html_file, 'wt') + f.write(self.doc.render()) + f.close() + + +if __name__ == '__main__': + html = HTML('web/', 'test_html') + html.add_header('hello world') + + ims = [] + txts = [] + links = [] + for n in range(4): + ims.append('image_%d.png' % n) + txts.append('text_%d' % n) + links.append('image_%d.png' % n) + html.add_images(ims, txts, links) + html.save() diff --git a/mlflow/sf2f/utils/visualization/plot.py b/mlflow/sf2f/utils/visualization/plot.py new file mode 100644 index 0000000..7f24e9d --- /dev/null +++ b/mlflow/sf2f/utils/visualization/plot.py @@ -0,0 +1,199 @@ +import matplotlib +matplotlib.use('Agg') +import matplotlib.pyplot as plt +import io +import tensorflow as tf +import PIL +import numpy as np + + +def plot_attention(attn_mat, path=None, info=None): + num_query_head, seq_len = attn_mat.shape + fig, ax = plt.subplots() + fig.set_size_inches(10, 10) + + querys = ['query {}'.format(i) for i in range(num_query_head)] + steps = ['steps {}'.format(j) for j in range(seq_len)] + + im, cbar = heatmap(attn_mat, querys, steps, ax=ax, + cmap="YlGn", cbarlabel="Attention Weight") + annotate_heatmap(im) + + fig.tight_layout() + if path is not None: + plt.savefig(path, format='png') + buf = io.BytesIO() + plt.savefig(buf, format='png') + buf.seek(0) + plt.clf() + plt.cla() + plt.close(fig) + return buf + + +def plot_mel_spectrogram(log_mels, + path=None, + info=None, + colorbar=True, + label=True, + coordinate=True, + remove_boarder=False): + fig, ax = plt.subplots() + if not coordinate: + #ax.format_coord = lambda x, y: '' + plt.axis('off') + fig.set_size_inches(5, 3) + im = ax.imshow( + log_mels, + aspect='auto', + origin='lower', + interpolation='none') + im.set_cmap('viridis') + if colorbar: + fig.colorbar(im, ax=ax) + xlabel = 'Time' + if info is not None: + xlabel += '\n\n' + info + if label: + plt.xlabel(xlabel) + plt.ylabel('Frequency') + plt.tight_layout() + if path is not None: + plt.savefig(path, format='png') + buf = io.BytesIO() + if not remove_boarder: + plt.savefig(buf, format='png') + else: + plt.savefig(buf, format='png', bbox_inches='tight') + buf.seek(0) + plt.clf() + plt.cla() + plt.close(fig) + return buf + + +def get_np_plot(plot_buf): + image = PIL.Image.open(plot_buf) + WW, HH = image.size + + return np.array(image) + + +def heatmap(data, row_labels, col_labels, ax=None, + cbar_kw={}, cbarlabel="", **kwargs): + """ + Create a heatmap from a numpy array and two lists of labels. + + Parameters + ---------- + data + A 2D numpy array of shape (N, M). + row_labels + A list or array of length N with the labels for the rows. + col_labels + A list or array of length M with the labels for the columns. + ax + A `matplotlib.axes.Axes` instance to which the heatmap is plotted. If + not provided, use current axes or create a new one. Optional. + cbar_kw + A dictionary with arguments to `matplotlib.Figure.colorbar`. Optional. + cbarlabel + The label for the colorbar. Optional. + **kwargs + All other arguments are forwarded to `imshow`. + """ + + if not ax: + ax = plt.gca() + + # Plot the heatmap + im = ax.imshow(data, **kwargs) + + # Create colorbar + cbar = ax.figure.colorbar(im, ax=ax, **cbar_kw) + cbar.ax.set_ylabel(cbarlabel, rotation=-90, va="bottom") + + # We want to show all ticks... + ax.set_xticks(np.arange(data.shape[1])) + ax.set_yticks(np.arange(data.shape[0])) + # ... and label them with the respective list entries. + ax.set_xticklabels(col_labels) + ax.set_yticklabels(row_labels) + + # Let the horizontal axes labeling appear on top. + ax.tick_params(top=True, bottom=False, + labeltop=True, labelbottom=False) + + # Rotate the tick labels and set their alignment. + plt.setp(ax.get_xticklabels(), rotation=-30, ha="right", + rotation_mode="anchor") + + # Turn spines off and create white grid. + for edge, spine in ax.spines.items(): + spine.set_visible(False) + + ax.set_xticks(np.arange(data.shape[1]+1)-.5, minor=True) + ax.set_yticks(np.arange(data.shape[0]+1)-.5, minor=True) + ax.grid(which="minor", color="w", linestyle='-', linewidth=3) + ax.tick_params(which="minor", bottom=False, left=False) + + return im, cbar + + +def annotate_heatmap(im, data=None, valfmt="{x:.2f}", + textcolors=["black", "white"], + threshold=None, **textkw): + """ + A function to annotate a heatmap. + + Parameters + ---------- + im + The AxesImage to be labeled. + data + Data used to annotate. If None, the image's data is used. Optional. + valfmt + The format of the annotations inside the heatmap. This should either + use the string format method, e.g. "$ {x:.2f}", or be a + `matplotlib.ticker.Formatter`. Optional. + textcolors + A list or array of two color specifications. The first is used for + values below a threshold, the second for those above. Optional. + threshold + Value in data units according to which the colors from textcolors are + applied. If None (the default) uses the middle of the colormap as + separation. Optional. + **kwargs + All other arguments are forwarded to each call to `text` used to create + the text labels. + """ + + if not isinstance(data, (list, np.ndarray)): + data = im.get_array() + + # Normalize the threshold to the images color range. + if threshold is not None: + threshold = im.norm(threshold) + else: + threshold = im.norm(data.max())/2. + + # Set default alignment to center, but allow it to be + # overwritten by textkw. + kw = dict(horizontalalignment="center", + verticalalignment="center") + kw.update(textkw) + + # Get the formatter in case a string is supplied + if isinstance(valfmt, str): + valfmt = matplotlib.ticker.StrMethodFormatter(valfmt) + + # Loop over the data and create a `Text` for each "pixel". + # Change the text's color depending on the data. + texts = [] + for i in range(data.shape[0]): + for j in range(data.shape[1]): + kw.update(color=textcolors[int(im.norm(data[i, j]) > threshold)]) + text = im.axes.text(j, i, valfmt(data[i, j], None), **kw) + texts.append(text) + + return texts diff --git a/mlflow/sf2f/utils/visualization/tsne.py b/mlflow/sf2f/utils/visualization/tsne.py new file mode 100644 index 0000000..62f8791 --- /dev/null +++ b/mlflow/sf2f/utils/visualization/tsne.py @@ -0,0 +1,40 @@ +''' +Visualize embeddings via TSNE +''' + + +import numpy as np +import matplotlib.pyplot as plt +from mpl_toolkits.mplot3d import Axes3D +from sklearn.manifold import TSNE + + +def visualize_distribution( + embeddings, + num_class, + sample_per_class, + class_names=None): + ''' + Inputs: + embeddings: numpy array with shape + [num_class * sample_per_class, emb_dim] + ''' + emb_2d = TSNE(n_components=2).fit_transform(embeddings) + emb_2d = emb_2d.reshape([num_class, sample_per_class, 2]) + fig, axes = plt.subplots() + fig.set_size_inches(18.5, 10.5) + if class_names is None: + class_names = [' ' for i in range(num_class)] + color_list = ['b', 'g', 'r', 'c', 'm', 'y', 'k', \ + 'orange', 'purple', 'pink'] + for i, class_name in enumerate(class_names): + x = emb_2d[i, :, 0] + y = emb_2d[i, :, 1] + axes.scatter(x=x, y=y, c=color_list[i], label=class_name) + + axes.set_xlabel("Component 1") + axes.set_ylabel("Component 2") + axes.legend(loc='upper left') + axes.set_xlim(-300, 300) + axes.set_ylim(-300, 300) + plt.show() diff --git a/mlflow/sf2f/utils/visualization/vis.py b/mlflow/sf2f/utils/visualization/vis.py new file mode 100644 index 0000000..de412bd --- /dev/null +++ b/mlflow/sf2f/utils/visualization/vis.py @@ -0,0 +1,219 @@ +import tempfile +import os +import torch +import numpy as np +import matplotlib.pyplot as plt +from matplotlib.patches import Rectangle +from imageio import imread +import time + + +""" +Utilities for making visualizations. +""" + + +def draw_layout(vocab, objs, boxes, masks=None, size=256, + show_boxes=False, bgcolor=(0, 0, 0)): + if bgcolor == 'white': + bgcolor = (255, 255, 255) + + cmap = plt.get_cmap('rainbow') + colors = cmap(np.linspace(0, 1, len(objs))) + + with torch.no_grad(): + objs = objs.cpu().clone() + boxes = boxes.cpu().clone() + boxes *= size + + if masks is not None: + masks = masks.cpu().clone() + + bgcolor = np.asarray(bgcolor) + bg = np.ones((size, size, 1)) * bgcolor + plt.imshow(bg.astype(np.uint8)) + + plt.gca().set_xlim(0, size) + plt.gca().set_ylim(size, 0) + plt.gca().set_aspect(1.0, adjustable='box') + + for i, obj in enumerate(objs): + name = vocab['object_idx_to_name'][obj] + if name == '__image__': + continue + box = boxes[i] + + if masks is None: + continue + mask = masks[i].numpy() + mask /= mask.max() + + r, g, b, a = colors[i] + colored_mask = mask[:, :, None] * np.asarray(colors[i]) + + x0, y0, x1, y1 = box + plt.imshow(colored_mask, extent=(x0, x1, y1, y0), + interpolation='bicubic', alpha=1.0) + + if show_boxes: + for i, obj in enumerate(objs): + name = vocab['object_idx_to_name'][obj] + if name == '__image__': + continue + box = boxes[i] + + draw_box(box, colors[i], name) + + +def draw_box(box, color, text=None): + """ + Draw a bounding box using pyplot, optionally with a text box label. + + Inputs: + - box: Tensor or list with 4 elements: [x0, y0, x1, y1] in [0, W] x [0, H] + coordinate system. + - color: pyplot color to use for the box. + - text: (Optional) String; if provided then draw a label for this box. + """ + TEXT_BOX_HEIGHT = 10 + if torch.is_tensor(box) and box.dim() == 2: + box = box.view(-1) + assert box.size(0) == 4 + x0, y0, x1, y1 = box + assert y1 > y0, box + assert x1 > x0, box + w, h = x1 - x0, y1 - y0 + rect = Rectangle((x0, y0), w, h, fc='none', lw=2, ec=color) + plt.gca().add_patch(rect) + if text is not None: + text_rect = Rectangle( + (x0, y0), w, TEXT_BOX_HEIGHT, fc=color, alpha=0.5) + plt.gca().add_patch(text_rect) + tx = 0.5 * (x0 + x1) + ty = y0 + TEXT_BOX_HEIGHT / 2.0 + plt.text(tx, ty, text, va='center', ha='center') + + +def draw_scene_graph(objs, triples, vocab=None, **kwargs): + """ + Use GraphViz to draw a scene graph. If vocab is not passed then we assume + that objs and triples are python lists containing strings for object and + relationship names. + + Using this requires that GraphViz is installed. On Ubuntu 16.04 this is easy: + sudo apt-get install graphviz + """ + output_filename = kwargs.pop('output_filename', + 'graph-{}.png'.format(time.strftime("%Y%m%d-%H%M%S"))) + orientation = kwargs.pop('orientation', 'V') + edge_width = kwargs.pop('edge_width', 6) + arrow_size = kwargs.pop('arrow_size', 1.5) + binary_edge_weight = kwargs.pop('binary_edge_weight', 1.2) + ignore_dummies = kwargs.pop('ignore_dummies', True) + + if orientation not in ['V', 'H']: + raise ValueError('Invalid orientation "%s"' % orientation) + rankdir = {'H': 'LR', 'V': 'TD'}[orientation] + + if vocab is not None: + # Decode object and relationship names + assert torch.is_tensor(objs) + assert torch.is_tensor(triples) + objs_list, triples_list = [], [] + for i in range(objs.size(0)): + objs_list.append(vocab['object_idx_to_name'][objs[i].item()]) + for i in range(triples.size(0)): + s = triples[i, 0].item() + p = vocab['pred_idx_to_name'][triples[i, 1].item()] + # p = triples[i, 1].item() + o = triples[i, 2].item() + triples_list.append([s, p, o]) + objs, triples = objs_list, triples_list + + # General setup, and style for object nodes + lines = [ + 'digraph{', + 'graph [size="5,3",ratio="compress",dpi="300",bgcolor="transparent"]', + 'rankdir=%s' % rankdir, + 'nodesep="0.5"', + 'ranksep="0.5"', + 'node [shape="box",style="rounded,filled",fontsize="48",color="none"]', + 'node [fillcolor="lightpink1"]', + ] + # Output nodes for objects + for i, obj in enumerate(objs): + if ignore_dummies and obj == '__image__': + continue + lines.append('%d [label="%s"]' % (i, obj)) + + # Output relationships + next_node_id = len(objs) + lines.append('node [fillcolor="lightblue1"]') + for s, p, o in triples: + if ignore_dummies and p == '__in_image__': + continue + lines += [ + '%d [label="%s"]' % (next_node_id, p), + '%d->%d [penwidth=%f,arrowsize=%f,weight=%f]' % ( + s, next_node_id, edge_width, arrow_size, binary_edge_weight), + '%d->%d [penwidth=%f,arrowsize=%f,weight=%f]' % ( + next_node_id, o, edge_width, arrow_size, binary_edge_weight) + ] + next_node_id += 1 + lines.append('}') + + # Now it gets slightly hacky. Write the graphviz spec to a temporary + # text file + ff, dot_filename = tempfile.mkstemp() + with open(dot_filename, 'w') as f: + for line in lines: + f.write('%s\n' % line) + os.close(ff) + + # Shell out to invoke graphviz; this will save the resulting image to disk, + # so we read it, delete it, then return it. + output_format = os.path.splitext(output_filename)[1][1:] + os.system('dot -T%s %s > %s' % + (output_format, dot_filename, output_filename)) + os.remove(dot_filename) + try: + img = imread(output_filename) + os.remove(output_filename) + except: + print("####################################") + print("Fail to load image: {}".format(output_filename)) + print("####################################") + img = np.zeros((400, 400, 3)) + + return np.array(img) + + +if __name__ == '__main__': + o_idx_to_name = ['cat', 'dog', 'hat', 'skateboard'] + p_idx_to_name = ['riding', 'wearing', 'on', 'next to', 'above'] + o_name_to_idx = {s: i for i, s in enumerate(o_idx_to_name)} + p_name_to_idx = {s: i for i, s in enumerate(p_idx_to_name)} + vocab = { + 'object_idx_to_name': o_idx_to_name, + 'object_name_to_idx': o_name_to_idx, + 'pred_idx_to_name': p_idx_to_name, + 'pred_name_to_idx': p_name_to_idx, + } + + objs = [ + 'cat', + 'cat', + 'skateboard', + 'hat', + ] + objs = torch.LongTensor([o_name_to_idx[o] for o in objs]) + triples = [ + [0, 'next to', 1], + [0, 'riding', 2], + [1, 'wearing', 3], + [3, 'above', 2], + ] + triples = [[s, p_name_to_idx[p], o] for s, p, o in triples] + triples = torch.LongTensor(triples) + + draw_scene_graph(objs, triples, vocab, orientation='V') diff --git a/mlflow/sf2f/utils/wav2mel.py b/mlflow/sf2f/utils/wav2mel.py new file mode 100644 index 0000000..35a822d --- /dev/null +++ b/mlflow/sf2f/utils/wav2mel.py @@ -0,0 +1,104 @@ +''' +Convert *.wav or *.m4a into mel spectrogram\ + +for unit test, run: + python utils/wav2mel.py +''' + + +import random +import numpy as np +import argparse +import time +from _thread import start_new_thread +import queue +from python_speech_features import logfbank +import webrtcvad +try: + import vad_ex +except: + from utils import vad_ex + + +def vad_process(path): + # VAD Process + if path.endswith('.wav'): + audio, sample_rate = vad_ex.read_wave(path) + elif path.endswith('.m4a'): + audio, sample_rate = vad_ex.read_m4a(path) + else: + raise TypeError('Unsupported file type: {}'.format(path.split('.')[-1])) + + vad = webrtcvad.Vad(1) + frames = vad_ex.frame_generator(30, audio, sample_rate) + frames = list(frames) + segments = vad_ex.vad_collector(sample_rate, 30, 300, vad, frames) + total_wav = b"" + for i, segment in enumerate(segments): + total_wav += segment + # Without writing, unpack total_wav into numpy [N,1] array + # 16bit PCM 기준 dtype=np.int16 + wav_arr = np.frombuffer(total_wav, dtype=np.int16) + #print("read audio data from byte string. np array of shape:" + \ + # str(wav_arr.shape)) + return wav_arr, sample_rate + + +def wav_to_mel(path, nfilt=40): + ''' + Output shape: (nfilt, length) + ''' + wav_arr, sample_rate = vad_process(path) + #print("sample_rate:", sample_rate) + logmel_feats = logfbank( + wav_arr, + samplerate=sample_rate, + nfilt=nfilt) + #print("created logmel feats from audio data. np array of shape:" \ + # + str(logmel_feats.shape)) + return np.transpose(logmel_feats) + + +if __name__ == '__main__': + import os + VOX_DIR = os.path.join('./data', 'VoxCeleb') + WAV_DIR = os.path.join('./data', 'VoxCeleb', 'raw_wav') + #wav_path = 'vox1/test/id10270/zjwijMp0Qyw/00001.wav' + wav_path = 'vox1/dev/id10001/J9lHsKG98U8/00026.wav' + wav_path = os.path.join(WAV_DIR, wav_path) + # m4a_path = 'vox2/test/id04253/dfbCPe2xOPA/00257.m4a' + # m4a_path = os.path.join(WAV_DIR, m4a_path) + + #wav_arr, sample_rate = vad_process(wav_path) + #print(wav_arr, sample_rate) + #wav_arr, sample_rate = vad_process(m4a_path) + #print(wav_arr, sample_rate) + + print(wav_to_mel(wav_path).shape) + # print(wav_to_mel(m4a_path).shape) + + # Compare with preprocessed pickles + import sys + sys.path.append('./') + from datasets.vox_dataset import VoxDataset + # Config + image_size = (256, 256) + #image_normalize_method = 'imagenet' + image_normalize_method = 'standard' + mel_normalize_method = 'vox_mel' + test_case_dir = os.path.join('./data', 'test_cases') + os.makedirs(test_case_dir, exist_ok=True) + + # Dataset + vox_dataset = VoxDataset( + data_dir=VOX_DIR, + image_size=image_size, + image_normalize_method=image_normalize_method, + mel_normalize_method=mel_normalize_method) + np_mel = vox_dataset.load_mel_gram(mel_pickle= \ + './data/VoxCeleb/vox1/mel_spectrograms/A.J._Buckley' + \ + '/id10001_J9lHsKG98U8_00026.pickle') + print('Load from pickle:', np_mel.shape) + print('Calculated by program:', wav_to_mel(wav_path).shape) + + print(np_mel - wav_to_mel(wav_path)) From fc4fd2c9f998268416819493034a7a3820092130 Mon Sep 17 00:00:00 2001 From: internationalwe Date: Tue, 5 Mar 2024 08:12:46 +0000 Subject: [PATCH 3/8] Feat: add inference_fuser file - - #3 --- .gitignore | 11 +++- mlflow/sf2f/inference_fuser.py | 92 ++++++++++++++++++++++++++++++++++ 2 files changed, 102 insertions(+), 1 deletion(-) create mode 100644 mlflow/sf2f/inference_fuser.py diff --git a/.gitignore b/.gitignore index d20afce..8729547 100644 --- a/.gitignore +++ b/.gitignore @@ -162,4 +162,13 @@ cython_debug/ #.idea/ wandb/ *.pkl -events.out* \ No newline at end of file +events.out* +grafana.db +000000* +lock +queries.active +tombstones +prometheus/prometheus-volume/data/* +index +0000* +meta.json \ No newline at end of file diff --git a/mlflow/sf2f/inference_fuser.py b/mlflow/sf2f/inference_fuser.py new file mode 100644 index 0000000..aa7fa89 --- /dev/null +++ b/mlflow/sf2f/inference_fuser.py @@ -0,0 +1,92 @@ +''' +author: Bai Yeqi +email: yeqi.bai@yitu-inc.com +''' +''' +Reload the generator checkpoint, inference with provided audio, in form of .wav + or .m4a + +We infer with different policies: + 1. Generate the face conditioned on each segment + 2. Generate the face conditioned on the whole speech + 3. Generate the face with the fuser +''' + + +import os +import shutil +import glob +import pyprind +import glog as log +import torch +import numpy as np +from imageio import imwrite + +import models +from utils.wav2mel import wav_to_mel +from datasets import imagenet_deprocess_batch, set_mel_transform, \ + deprocess_and_save, window_segment +from options.opts import args, options + + +torch.backends.cudnn.benchmark = True + + +def main(): + global args, options + print(args) + device = torch.device('cuda') + float_dtype = torch.cuda.FloatTensor + long_dtype = torch.cuda.LongTensor + + image_normalize_method = options["data"]["data_opts"].get( + 'image_normalize_method', 'imagenet') + print('image_normalize_method:', image_normalize_method) + mel_normalize_method = options["data"]["data_opts"].get( + 'mel_normalize_method', 'vox_mel') + mel_transform = set_mel_transform(mel_normalize_method) + + # Model + log.info("Building Generative Model...") + print(options["generator"]) + model, _ = models.build_model( + options["generator"], + image_size=options["data"]["image_size"], + checkpoint_start_from=args.checkpoint_start_from) + model.cuda().eval() + + # voice_path = os.path.join(args.input_wav_dir, '*.wav') + # voice_list = glob.glob(voice_path) + # filename = voice_list[0] + filename = os.path.join(args.input_wav_file) + assert os.path.exists(filename), "File not found: {}".format(filename) + + # for filename in voice_list: + # result_sub_dir = filename.replace('.wav', '') + # os.makedirs(result_sub_dir, exist_ok=True) # 결과 폴더 생성 + # Load mel_spectrogram + log_mel = wav_to_mel(filename) + log_mel = mel_transform(log_mel).type(float_dtype) + #print(log_mel) + log_mel_segs = window_segment( + log_mel, window_length=125, stride_length=63) + log_mel = log_mel.unsqueeze(0) + + # image generated by fuser + # if args.fuser_infer: + with torch.no_grad(): + imgs_fused, others = model(log_mel_segs.unsqueeze(0)) + if isinstance(imgs_fused, tuple): + imgs_fused = imgs_fused[-1] + imgs_fused = imgs_fused.cpu().detach() + imgs_fused = imagenet_deprocess_batch( + imgs_fused, normalize_method=image_normalize_method) + for j in range(imgs_fused.shape[0]): + img_np = imgs_fused[j].numpy().transpose(1, 2, 0) # 64x64x3 + # img_path = os.path.join(result_sub_dir, 'fused_%d.png' % j) # 이미지 저장할 파일 + # imwrite(img_path, img_np) # 이미지 저장 + + return img_np.tobytes() + +if __name__ == '__main__': + main() From 020d27a42b891df2357f8c536223760ce08172c9 Mon Sep 17 00:00:00 2001 From: internationalwe Date: Wed, 6 Mar 2024 06:37:56 +0000 Subject: [PATCH 4/8] Style : Change mlflow folder structure - - #3 --- {mlflow => train/mlflow}/sf2f/GETTING_STARTED.md | 0 .../mlflow/sf2f/data/\353\205\271\354\235\214.wav" | Bin {mlflow => train/mlflow}/sf2f/datasets/__init__.py | 0 .../mlflow}/sf2f/datasets/build_dataset.py | 0 {mlflow => train/mlflow}/sf2f/datasets/utils.py | 0 .../mlflow}/sf2f/datasets/vox_dataset.py | 0 {mlflow => train/mlflow}/sf2f/infer.py | 0 {mlflow => train/mlflow}/sf2f/inference_fuser.py | 0 {mlflow => train/mlflow}/sf2f/model_registry.py | 0 {mlflow => train/mlflow}/sf2f/models/__init__.py | 0 {mlflow => train/mlflow}/sf2f/models/attention.py | 0 {mlflow => train/mlflow}/sf2f/models/crn.py | 0 .../mlflow}/sf2f/models/discriminators.py | 0 .../mlflow}/sf2f/models/encoder_decoder.py | 0 .../mlflow}/sf2f/models/face_decoders.py | 0 {mlflow => train/mlflow}/sf2f/models/fusers.py | 0 .../mlflow}/sf2f/models/inception_resnet_v1.py | 0 {mlflow => train/mlflow}/sf2f/models/layers.py | 0 .../mlflow}/sf2f/models/model_collection.py | 0 {mlflow => train/mlflow}/sf2f/models/model_setup.py | 0 {mlflow => train/mlflow}/sf2f/models/networks.py | 0 {mlflow => train/mlflow}/sf2f/models/perceptual.py | 0 .../mlflow}/sf2f/models/voice_encoders.py | 0 {mlflow => train/mlflow}/sf2f/options/__init__.py | 0 .../mlflow}/sf2f/options/data_opts/vox.yaml | 0 {mlflow => train/mlflow}/sf2f/options/opts.py | 0 .../mlflow}/sf2f/options/vox/baseline/v2f.yaml | 0 .../sf2f/options/vox/sf2f/sf2f_1st_stage.yaml | 0 .../mlflow}/sf2f/options/vox/sf2f/sf2f_fuser.yaml | 0 .../sf2f/options/vox/sf2f/sf2f_mid_1st_stage.yaml | 0 .../sf2f/options/vox/sf2f/sf2f_mid_fuser.yaml | 0 .../sf2f_1st_stage.yaml | 0 .../sf2f_1st_stage.yaml | 0 .../sf2f_1st_stage.yaml | 0 .../sf2f_1st_stage.yaml | 0 .../sf2f_1st_stage.yaml | 0 .../sf2f_1st_stage.yaml | 0 .../sf2f_1st_stage.yaml | 0 .../sf2f_1st_stage.yaml | 0 .../sf2f_1st_stage.yaml | 0 .../sf2f_1st_stage.yaml | 0 .../sf2f_1st_stage.yaml | 0 .../sf2f_1st_stage.yaml | 0 .../sf2f_1st_stage.yaml | 0 .../sf2f_1st_stage.yaml | 0 .../sf2f_1st_stage.yaml | 0 .../sf2f_1st_stage.yaml | 0 .../sf2f_1st_stage.yaml | 0 .../sf2f_1st_stage.yaml | 0 .../sf2f_1st_stage.yaml | 0 .../sf2f_1st_stage.yaml | 0 .../sf2f_1st_stage.yaml | 0 .../sf2f_1st_stage.yaml | 0 .../mlflow}/sf2f/scripts/DScore/__init__.py | 0 .../mlflow}/sf2f/scripts/DScore/models/__init__.py | 0 .../sf2f/scripts/DScore/models/base_model.py | 0 .../sf2f/scripts/DScore/models/dist_model.py | 0 .../mlflow}/sf2f/scripts/DScore/models/models.py | 0 .../sf2f/scripts/DScore/models/networks_basic.py | 0 .../scripts/DScore/models/pretrained_networks.py | 0 .../mlflow}/sf2f/scripts/DScore/util/__init__.py | 0 .../mlflow}/sf2f/scripts/DScore/util/html.py | 0 .../mlflow}/sf2f/scripts/DScore/util/util.py | 0 .../mlflow}/sf2f/scripts/DScore/util/visualizer.py | 0 .../mlflow}/sf2f/scripts/build_demo_set.py | 0 .../mlflow}/sf2f/scripts/compute_diversity_score.py | 0 .../mlflow}/sf2f/scripts/compute_fid_score.py | 0 .../mlflow}/sf2f/scripts/compute_inception_score.py | 0 .../mlflow}/sf2f/scripts/compute_mel_mean_var.py | 0 .../mlflow}/sf2f/scripts/compute_vggface_score.py | 0 .../mlflow}/sf2f/scripts/convert_wav_to_mel.py | 0 .../mlflow}/sf2f/scripts/create_split_json.py | 0 .../sf2f/scripts/download_vggface_weights.sh | 0 .../mlflow}/sf2f/scripts/install_requirements.py | 0 {mlflow => train/mlflow}/sf2f/scripts/print_args.py | 0 .../mlflow}/sf2f/scripts/sample_mel_spectrograms.py | 0 .../mlflow}/sf2f/scripts/strip_checkpoint.py | 0 .../mlflow}/sf2f/scripts/strip_old_args.py | 0 {mlflow => train/mlflow}/sf2f/scripts/watch_data.py | 0 {mlflow => train/mlflow}/sf2f/train.py | 0 {mlflow => train/mlflow}/sf2f/train_registry.py | 0 {mlflow => train/mlflow}/sf2f/utils/__init__.py | 0 {mlflow => train/mlflow}/sf2f/utils/bilinear.py | 0 {mlflow => train/mlflow}/sf2f/utils/box_utils.py | 0 {mlflow => train/mlflow}/sf2f/utils/common.py | 0 {mlflow => train/mlflow}/sf2f/utils/evaluate.py | 0 {mlflow => train/mlflow}/sf2f/utils/evaluate_fid.py | 0 .../mlflow}/sf2f/utils/filter_pickle.py | 0 {mlflow => train/mlflow}/sf2f/utils/logger.py | 0 {mlflow => train/mlflow}/sf2f/utils/losses.py | 0 {mlflow => train/mlflow}/sf2f/utils/metrics.py | 0 {mlflow => train/mlflow}/sf2f/utils/mlflow_wandb.py | 0 {mlflow => train/mlflow}/sf2f/utils/pgan_utils.py | 0 .../mlflow}/sf2f/utils/s2f_evaluator.py | 0 .../mlflow}/sf2f/utils/training_utils.py | 0 {mlflow => train/mlflow}/sf2f/utils/utils.py | 0 {mlflow => train/mlflow}/sf2f/utils/vad_ex.py | 0 .../mlflow}/sf2f/utils/visualization/__init__.py | 0 .../mlflow}/sf2f/utils/visualization/html.py | 0 .../mlflow}/sf2f/utils/visualization/plot.py | 0 .../mlflow}/sf2f/utils/visualization/tsne.py | 0 .../mlflow}/sf2f/utils/visualization/vis.py | 0 {mlflow => train/mlflow}/sf2f/utils/wav2mel.py | 0 103 files changed, 0 insertions(+), 0 deletions(-) rename {mlflow => train/mlflow}/sf2f/GETTING_STARTED.md (100%) rename "mlflow/sf2f/data/\353\205\271\354\235\214.wav" => "train/mlflow/sf2f/data/\353\205\271\354\235\214.wav" (100%) rename {mlflow => train/mlflow}/sf2f/datasets/__init__.py (100%) rename {mlflow => train/mlflow}/sf2f/datasets/build_dataset.py (100%) rename {mlflow => train/mlflow}/sf2f/datasets/utils.py (100%) rename {mlflow => train/mlflow}/sf2f/datasets/vox_dataset.py (100%) rename {mlflow => train/mlflow}/sf2f/infer.py (100%) rename {mlflow => train/mlflow}/sf2f/inference_fuser.py (100%) rename {mlflow => train/mlflow}/sf2f/model_registry.py (100%) rename {mlflow => train/mlflow}/sf2f/models/__init__.py (100%) rename {mlflow => train/mlflow}/sf2f/models/attention.py (100%) rename {mlflow => train/mlflow}/sf2f/models/crn.py (100%) rename {mlflow => train/mlflow}/sf2f/models/discriminators.py (100%) rename {mlflow => train/mlflow}/sf2f/models/encoder_decoder.py (100%) rename {mlflow => train/mlflow}/sf2f/models/face_decoders.py (100%) rename {mlflow => train/mlflow}/sf2f/models/fusers.py (100%) rename {mlflow => train/mlflow}/sf2f/models/inception_resnet_v1.py (100%) rename {mlflow => train/mlflow}/sf2f/models/layers.py (100%) rename {mlflow => train/mlflow}/sf2f/models/model_collection.py (100%) rename {mlflow => train/mlflow}/sf2f/models/model_setup.py (100%) rename {mlflow => train/mlflow}/sf2f/models/networks.py (100%) rename {mlflow => train/mlflow}/sf2f/models/perceptual.py (100%) rename {mlflow => train/mlflow}/sf2f/models/voice_encoders.py (100%) rename {mlflow => train/mlflow}/sf2f/options/__init__.py (100%) rename {mlflow => train/mlflow}/sf2f/options/data_opts/vox.yaml (100%) rename {mlflow => train/mlflow}/sf2f/options/opts.py (100%) rename {mlflow => train/mlflow}/sf2f/options/vox/baseline/v2f.yaml (100%) rename {mlflow => train/mlflow}/sf2f/options/vox/sf2f/sf2f_1st_stage.yaml (100%) rename {mlflow => train/mlflow}/sf2f/options/vox/sf2f/sf2f_fuser.yaml (100%) rename {mlflow => train/mlflow}/sf2f/options/vox/sf2f/sf2f_mid_1st_stage.yaml (100%) rename {mlflow => train/mlflow}/sf2f/options/vox/sf2f/sf2f_mid_fuser.yaml (100%) rename {mlflow => train/mlflow}/sf2f/output/sf2f_1st_stage-20240303-215640/sf2f_1st_stage.yaml (100%) rename {mlflow => train/mlflow}/sf2f/output/sf2f_1st_stage-20240303-220349/sf2f_1st_stage.yaml (100%) rename {mlflow => train/mlflow}/sf2f/output/sf2f_1st_stage-20240303-221106/sf2f_1st_stage.yaml (100%) rename {mlflow => train/mlflow}/sf2f/output/sf2f_1st_stage-20240303-225403/sf2f_1st_stage.yaml (100%) rename {mlflow => train/mlflow}/sf2f/output/sf2f_1st_stage-20240303-225533/sf2f_1st_stage.yaml (100%) rename {mlflow => train/mlflow}/sf2f/output/sf2f_1st_stage-20240303-225656/sf2f_1st_stage.yaml (100%) rename {mlflow => train/mlflow}/sf2f/output/sf2f_1st_stage-20240303-225950/sf2f_1st_stage.yaml (100%) rename {mlflow => train/mlflow}/sf2f/output/sf2f_1st_stage-20240303-230032/sf2f_1st_stage.yaml (100%) rename {mlflow => train/mlflow}/sf2f/output/sf2f_1st_stage-20240303-230142/sf2f_1st_stage.yaml (100%) rename {mlflow => train/mlflow}/sf2f/output/sf2f_1st_stage-20240303-230232/sf2f_1st_stage.yaml (100%) rename {mlflow => train/mlflow}/sf2f/output/sf2f_1st_stage-20240303-230355/sf2f_1st_stage.yaml (100%) rename {mlflow => train/mlflow}/sf2f/output/sf2f_1st_stage-20240303-230514/sf2f_1st_stage.yaml (100%) rename {mlflow => train/mlflow}/sf2f/output/sf2f_1st_stage-20240303-230631/sf2f_1st_stage.yaml (100%) rename {mlflow => train/mlflow}/sf2f/output/sf2f_1st_stage-20240303-231044/sf2f_1st_stage.yaml (100%) rename {mlflow => train/mlflow}/sf2f/output/sf2f_1st_stage-20240303-231201/sf2f_1st_stage.yaml (100%) rename {mlflow => train/mlflow}/sf2f/output/sf2f_1st_stage-20240303-231308/sf2f_1st_stage.yaml (100%) rename {mlflow => train/mlflow}/sf2f/output/sf2f_1st_stage-20240303-231542/sf2f_1st_stage.yaml (100%) rename {mlflow => train/mlflow}/sf2f/output/sf2f_1st_stage-20240303-231831/sf2f_1st_stage.yaml (100%) rename {mlflow => train/mlflow}/sf2f/output/sf2f_1st_stage-20240303-232325/sf2f_1st_stage.yaml (100%) rename {mlflow => train/mlflow}/sf2f/output/sf2f_1st_stage-20240303-233127/sf2f_1st_stage.yaml (100%) rename {mlflow => train/mlflow}/sf2f/output/sf2f_1st_stage-20240303-233311/sf2f_1st_stage.yaml (100%) rename {mlflow => train/mlflow}/sf2f/output/sf2f_1st_stage-20240303-233619/sf2f_1st_stage.yaml (100%) rename {mlflow => train/mlflow}/sf2f/scripts/DScore/__init__.py (100%) rename {mlflow => train/mlflow}/sf2f/scripts/DScore/models/__init__.py (100%) rename {mlflow => train/mlflow}/sf2f/scripts/DScore/models/base_model.py (100%) rename {mlflow => train/mlflow}/sf2f/scripts/DScore/models/dist_model.py (100%) rename {mlflow => train/mlflow}/sf2f/scripts/DScore/models/models.py (100%) rename {mlflow => train/mlflow}/sf2f/scripts/DScore/models/networks_basic.py (100%) rename {mlflow => train/mlflow}/sf2f/scripts/DScore/models/pretrained_networks.py (100%) rename {mlflow => train/mlflow}/sf2f/scripts/DScore/util/__init__.py (100%) rename {mlflow => train/mlflow}/sf2f/scripts/DScore/util/html.py (100%) rename {mlflow => train/mlflow}/sf2f/scripts/DScore/util/util.py (100%) rename {mlflow => train/mlflow}/sf2f/scripts/DScore/util/visualizer.py (100%) rename {mlflow => train/mlflow}/sf2f/scripts/build_demo_set.py (100%) rename {mlflow => train/mlflow}/sf2f/scripts/compute_diversity_score.py (100%) rename {mlflow => train/mlflow}/sf2f/scripts/compute_fid_score.py (100%) rename {mlflow => train/mlflow}/sf2f/scripts/compute_inception_score.py (100%) rename {mlflow => train/mlflow}/sf2f/scripts/compute_mel_mean_var.py (100%) rename {mlflow => train/mlflow}/sf2f/scripts/compute_vggface_score.py (100%) rename {mlflow => train/mlflow}/sf2f/scripts/convert_wav_to_mel.py (100%) rename {mlflow => train/mlflow}/sf2f/scripts/create_split_json.py (100%) rename {mlflow => train/mlflow}/sf2f/scripts/download_vggface_weights.sh (100%) rename {mlflow => train/mlflow}/sf2f/scripts/install_requirements.py (100%) rename {mlflow => train/mlflow}/sf2f/scripts/print_args.py (100%) rename {mlflow => train/mlflow}/sf2f/scripts/sample_mel_spectrograms.py (100%) rename {mlflow => train/mlflow}/sf2f/scripts/strip_checkpoint.py (100%) rename {mlflow => train/mlflow}/sf2f/scripts/strip_old_args.py (100%) rename {mlflow => train/mlflow}/sf2f/scripts/watch_data.py (100%) rename {mlflow => train/mlflow}/sf2f/train.py (100%) rename {mlflow => train/mlflow}/sf2f/train_registry.py (100%) rename {mlflow => train/mlflow}/sf2f/utils/__init__.py (100%) rename {mlflow => train/mlflow}/sf2f/utils/bilinear.py (100%) rename {mlflow => train/mlflow}/sf2f/utils/box_utils.py (100%) rename {mlflow => train/mlflow}/sf2f/utils/common.py (100%) rename {mlflow => train/mlflow}/sf2f/utils/evaluate.py (100%) rename {mlflow => train/mlflow}/sf2f/utils/evaluate_fid.py (100%) rename {mlflow => train/mlflow}/sf2f/utils/filter_pickle.py (100%) rename {mlflow => train/mlflow}/sf2f/utils/logger.py (100%) rename {mlflow => train/mlflow}/sf2f/utils/losses.py (100%) rename {mlflow => train/mlflow}/sf2f/utils/metrics.py (100%) rename {mlflow => train/mlflow}/sf2f/utils/mlflow_wandb.py (100%) rename {mlflow => train/mlflow}/sf2f/utils/pgan_utils.py (100%) rename {mlflow => train/mlflow}/sf2f/utils/s2f_evaluator.py (100%) rename {mlflow => train/mlflow}/sf2f/utils/training_utils.py (100%) rename {mlflow => train/mlflow}/sf2f/utils/utils.py (100%) rename {mlflow => train/mlflow}/sf2f/utils/vad_ex.py (100%) rename {mlflow => train/mlflow}/sf2f/utils/visualization/__init__.py (100%) rename {mlflow => train/mlflow}/sf2f/utils/visualization/html.py (100%) rename {mlflow => train/mlflow}/sf2f/utils/visualization/plot.py (100%) rename {mlflow => train/mlflow}/sf2f/utils/visualization/tsne.py (100%) rename {mlflow => train/mlflow}/sf2f/utils/visualization/vis.py (100%) rename {mlflow => train/mlflow}/sf2f/utils/wav2mel.py (100%) diff --git a/mlflow/sf2f/GETTING_STARTED.md b/train/mlflow/sf2f/GETTING_STARTED.md similarity index 100% rename from mlflow/sf2f/GETTING_STARTED.md rename to train/mlflow/sf2f/GETTING_STARTED.md diff --git "a/mlflow/sf2f/data/\353\205\271\354\235\214.wav" "b/train/mlflow/sf2f/data/\353\205\271\354\235\214.wav" similarity index 100% rename from "mlflow/sf2f/data/\353\205\271\354\235\214.wav" rename to "train/mlflow/sf2f/data/\353\205\271\354\235\214.wav" diff --git a/mlflow/sf2f/datasets/__init__.py b/train/mlflow/sf2f/datasets/__init__.py similarity index 100% rename from mlflow/sf2f/datasets/__init__.py rename to train/mlflow/sf2f/datasets/__init__.py diff --git a/mlflow/sf2f/datasets/build_dataset.py b/train/mlflow/sf2f/datasets/build_dataset.py similarity index 100% rename from mlflow/sf2f/datasets/build_dataset.py rename to train/mlflow/sf2f/datasets/build_dataset.py diff --git a/mlflow/sf2f/datasets/utils.py b/train/mlflow/sf2f/datasets/utils.py similarity index 100% rename from mlflow/sf2f/datasets/utils.py rename to train/mlflow/sf2f/datasets/utils.py diff --git a/mlflow/sf2f/datasets/vox_dataset.py b/train/mlflow/sf2f/datasets/vox_dataset.py similarity index 100% rename from mlflow/sf2f/datasets/vox_dataset.py rename to train/mlflow/sf2f/datasets/vox_dataset.py diff --git a/mlflow/sf2f/infer.py b/train/mlflow/sf2f/infer.py similarity index 100% rename from mlflow/sf2f/infer.py rename to train/mlflow/sf2f/infer.py diff --git a/mlflow/sf2f/inference_fuser.py b/train/mlflow/sf2f/inference_fuser.py similarity index 100% rename from mlflow/sf2f/inference_fuser.py rename to train/mlflow/sf2f/inference_fuser.py diff --git a/mlflow/sf2f/model_registry.py b/train/mlflow/sf2f/model_registry.py similarity index 100% rename from mlflow/sf2f/model_registry.py rename to train/mlflow/sf2f/model_registry.py diff --git a/mlflow/sf2f/models/__init__.py b/train/mlflow/sf2f/models/__init__.py similarity index 100% rename from mlflow/sf2f/models/__init__.py rename to train/mlflow/sf2f/models/__init__.py diff --git a/mlflow/sf2f/models/attention.py b/train/mlflow/sf2f/models/attention.py similarity index 100% rename from mlflow/sf2f/models/attention.py rename to train/mlflow/sf2f/models/attention.py diff --git a/mlflow/sf2f/models/crn.py b/train/mlflow/sf2f/models/crn.py similarity index 100% rename from mlflow/sf2f/models/crn.py rename to train/mlflow/sf2f/models/crn.py diff --git a/mlflow/sf2f/models/discriminators.py b/train/mlflow/sf2f/models/discriminators.py similarity index 100% rename from mlflow/sf2f/models/discriminators.py rename to train/mlflow/sf2f/models/discriminators.py diff --git a/mlflow/sf2f/models/encoder_decoder.py b/train/mlflow/sf2f/models/encoder_decoder.py similarity index 100% rename from mlflow/sf2f/models/encoder_decoder.py rename to train/mlflow/sf2f/models/encoder_decoder.py diff --git a/mlflow/sf2f/models/face_decoders.py b/train/mlflow/sf2f/models/face_decoders.py similarity index 100% rename from mlflow/sf2f/models/face_decoders.py rename to train/mlflow/sf2f/models/face_decoders.py diff --git a/mlflow/sf2f/models/fusers.py b/train/mlflow/sf2f/models/fusers.py similarity index 100% rename from mlflow/sf2f/models/fusers.py rename to train/mlflow/sf2f/models/fusers.py diff --git a/mlflow/sf2f/models/inception_resnet_v1.py b/train/mlflow/sf2f/models/inception_resnet_v1.py similarity index 100% rename from mlflow/sf2f/models/inception_resnet_v1.py rename to train/mlflow/sf2f/models/inception_resnet_v1.py diff --git a/mlflow/sf2f/models/layers.py b/train/mlflow/sf2f/models/layers.py similarity index 100% rename from mlflow/sf2f/models/layers.py rename to train/mlflow/sf2f/models/layers.py diff --git a/mlflow/sf2f/models/model_collection.py b/train/mlflow/sf2f/models/model_collection.py similarity index 100% rename from mlflow/sf2f/models/model_collection.py rename to train/mlflow/sf2f/models/model_collection.py diff --git a/mlflow/sf2f/models/model_setup.py b/train/mlflow/sf2f/models/model_setup.py similarity index 100% rename from mlflow/sf2f/models/model_setup.py rename to train/mlflow/sf2f/models/model_setup.py diff --git a/mlflow/sf2f/models/networks.py b/train/mlflow/sf2f/models/networks.py similarity index 100% rename from mlflow/sf2f/models/networks.py rename to train/mlflow/sf2f/models/networks.py diff --git a/mlflow/sf2f/models/perceptual.py b/train/mlflow/sf2f/models/perceptual.py similarity index 100% rename from mlflow/sf2f/models/perceptual.py rename to train/mlflow/sf2f/models/perceptual.py diff --git a/mlflow/sf2f/models/voice_encoders.py b/train/mlflow/sf2f/models/voice_encoders.py similarity index 100% rename from mlflow/sf2f/models/voice_encoders.py rename to train/mlflow/sf2f/models/voice_encoders.py diff --git a/mlflow/sf2f/options/__init__.py b/train/mlflow/sf2f/options/__init__.py similarity index 100% rename from mlflow/sf2f/options/__init__.py rename to train/mlflow/sf2f/options/__init__.py diff --git a/mlflow/sf2f/options/data_opts/vox.yaml b/train/mlflow/sf2f/options/data_opts/vox.yaml similarity index 100% rename from mlflow/sf2f/options/data_opts/vox.yaml rename to train/mlflow/sf2f/options/data_opts/vox.yaml diff --git a/mlflow/sf2f/options/opts.py b/train/mlflow/sf2f/options/opts.py similarity index 100% rename from mlflow/sf2f/options/opts.py rename to train/mlflow/sf2f/options/opts.py diff --git a/mlflow/sf2f/options/vox/baseline/v2f.yaml b/train/mlflow/sf2f/options/vox/baseline/v2f.yaml similarity index 100% rename from mlflow/sf2f/options/vox/baseline/v2f.yaml rename to train/mlflow/sf2f/options/vox/baseline/v2f.yaml diff --git a/mlflow/sf2f/options/vox/sf2f/sf2f_1st_stage.yaml b/train/mlflow/sf2f/options/vox/sf2f/sf2f_1st_stage.yaml similarity index 100% rename from mlflow/sf2f/options/vox/sf2f/sf2f_1st_stage.yaml rename to train/mlflow/sf2f/options/vox/sf2f/sf2f_1st_stage.yaml diff --git a/mlflow/sf2f/options/vox/sf2f/sf2f_fuser.yaml b/train/mlflow/sf2f/options/vox/sf2f/sf2f_fuser.yaml similarity index 100% rename from mlflow/sf2f/options/vox/sf2f/sf2f_fuser.yaml rename to train/mlflow/sf2f/options/vox/sf2f/sf2f_fuser.yaml diff --git a/mlflow/sf2f/options/vox/sf2f/sf2f_mid_1st_stage.yaml b/train/mlflow/sf2f/options/vox/sf2f/sf2f_mid_1st_stage.yaml similarity index 100% rename from mlflow/sf2f/options/vox/sf2f/sf2f_mid_1st_stage.yaml rename to train/mlflow/sf2f/options/vox/sf2f/sf2f_mid_1st_stage.yaml diff --git a/mlflow/sf2f/options/vox/sf2f/sf2f_mid_fuser.yaml b/train/mlflow/sf2f/options/vox/sf2f/sf2f_mid_fuser.yaml similarity index 100% rename from mlflow/sf2f/options/vox/sf2f/sf2f_mid_fuser.yaml rename to train/mlflow/sf2f/options/vox/sf2f/sf2f_mid_fuser.yaml diff --git a/mlflow/sf2f/output/sf2f_1st_stage-20240303-215640/sf2f_1st_stage.yaml b/train/mlflow/sf2f/output/sf2f_1st_stage-20240303-215640/sf2f_1st_stage.yaml similarity index 100% rename from mlflow/sf2f/output/sf2f_1st_stage-20240303-215640/sf2f_1st_stage.yaml rename to train/mlflow/sf2f/output/sf2f_1st_stage-20240303-215640/sf2f_1st_stage.yaml diff --git a/mlflow/sf2f/output/sf2f_1st_stage-20240303-220349/sf2f_1st_stage.yaml b/train/mlflow/sf2f/output/sf2f_1st_stage-20240303-220349/sf2f_1st_stage.yaml similarity index 100% rename from mlflow/sf2f/output/sf2f_1st_stage-20240303-220349/sf2f_1st_stage.yaml rename to train/mlflow/sf2f/output/sf2f_1st_stage-20240303-220349/sf2f_1st_stage.yaml diff --git a/mlflow/sf2f/output/sf2f_1st_stage-20240303-221106/sf2f_1st_stage.yaml b/train/mlflow/sf2f/output/sf2f_1st_stage-20240303-221106/sf2f_1st_stage.yaml similarity index 100% rename from mlflow/sf2f/output/sf2f_1st_stage-20240303-221106/sf2f_1st_stage.yaml rename to train/mlflow/sf2f/output/sf2f_1st_stage-20240303-221106/sf2f_1st_stage.yaml diff --git a/mlflow/sf2f/output/sf2f_1st_stage-20240303-225403/sf2f_1st_stage.yaml b/train/mlflow/sf2f/output/sf2f_1st_stage-20240303-225403/sf2f_1st_stage.yaml similarity index 100% rename from mlflow/sf2f/output/sf2f_1st_stage-20240303-225403/sf2f_1st_stage.yaml rename to train/mlflow/sf2f/output/sf2f_1st_stage-20240303-225403/sf2f_1st_stage.yaml diff --git a/mlflow/sf2f/output/sf2f_1st_stage-20240303-225533/sf2f_1st_stage.yaml b/train/mlflow/sf2f/output/sf2f_1st_stage-20240303-225533/sf2f_1st_stage.yaml similarity index 100% rename from mlflow/sf2f/output/sf2f_1st_stage-20240303-225533/sf2f_1st_stage.yaml rename to train/mlflow/sf2f/output/sf2f_1st_stage-20240303-225533/sf2f_1st_stage.yaml diff --git a/mlflow/sf2f/output/sf2f_1st_stage-20240303-225656/sf2f_1st_stage.yaml b/train/mlflow/sf2f/output/sf2f_1st_stage-20240303-225656/sf2f_1st_stage.yaml similarity index 100% rename from mlflow/sf2f/output/sf2f_1st_stage-20240303-225656/sf2f_1st_stage.yaml rename to train/mlflow/sf2f/output/sf2f_1st_stage-20240303-225656/sf2f_1st_stage.yaml diff --git a/mlflow/sf2f/output/sf2f_1st_stage-20240303-225950/sf2f_1st_stage.yaml b/train/mlflow/sf2f/output/sf2f_1st_stage-20240303-225950/sf2f_1st_stage.yaml similarity index 100% rename from mlflow/sf2f/output/sf2f_1st_stage-20240303-225950/sf2f_1st_stage.yaml rename to train/mlflow/sf2f/output/sf2f_1st_stage-20240303-225950/sf2f_1st_stage.yaml diff --git a/mlflow/sf2f/output/sf2f_1st_stage-20240303-230032/sf2f_1st_stage.yaml b/train/mlflow/sf2f/output/sf2f_1st_stage-20240303-230032/sf2f_1st_stage.yaml similarity index 100% rename from mlflow/sf2f/output/sf2f_1st_stage-20240303-230032/sf2f_1st_stage.yaml rename to train/mlflow/sf2f/output/sf2f_1st_stage-20240303-230032/sf2f_1st_stage.yaml diff --git a/mlflow/sf2f/output/sf2f_1st_stage-20240303-230142/sf2f_1st_stage.yaml b/train/mlflow/sf2f/output/sf2f_1st_stage-20240303-230142/sf2f_1st_stage.yaml similarity index 100% rename from mlflow/sf2f/output/sf2f_1st_stage-20240303-230142/sf2f_1st_stage.yaml rename to train/mlflow/sf2f/output/sf2f_1st_stage-20240303-230142/sf2f_1st_stage.yaml diff --git a/mlflow/sf2f/output/sf2f_1st_stage-20240303-230232/sf2f_1st_stage.yaml b/train/mlflow/sf2f/output/sf2f_1st_stage-20240303-230232/sf2f_1st_stage.yaml similarity index 100% rename from mlflow/sf2f/output/sf2f_1st_stage-20240303-230232/sf2f_1st_stage.yaml rename to train/mlflow/sf2f/output/sf2f_1st_stage-20240303-230232/sf2f_1st_stage.yaml diff --git a/mlflow/sf2f/output/sf2f_1st_stage-20240303-230355/sf2f_1st_stage.yaml b/train/mlflow/sf2f/output/sf2f_1st_stage-20240303-230355/sf2f_1st_stage.yaml similarity index 100% rename from mlflow/sf2f/output/sf2f_1st_stage-20240303-230355/sf2f_1st_stage.yaml rename to train/mlflow/sf2f/output/sf2f_1st_stage-20240303-230355/sf2f_1st_stage.yaml diff --git a/mlflow/sf2f/output/sf2f_1st_stage-20240303-230514/sf2f_1st_stage.yaml b/train/mlflow/sf2f/output/sf2f_1st_stage-20240303-230514/sf2f_1st_stage.yaml similarity index 100% rename from mlflow/sf2f/output/sf2f_1st_stage-20240303-230514/sf2f_1st_stage.yaml rename to train/mlflow/sf2f/output/sf2f_1st_stage-20240303-230514/sf2f_1st_stage.yaml diff --git a/mlflow/sf2f/output/sf2f_1st_stage-20240303-230631/sf2f_1st_stage.yaml b/train/mlflow/sf2f/output/sf2f_1st_stage-20240303-230631/sf2f_1st_stage.yaml similarity index 100% rename from mlflow/sf2f/output/sf2f_1st_stage-20240303-230631/sf2f_1st_stage.yaml rename to train/mlflow/sf2f/output/sf2f_1st_stage-20240303-230631/sf2f_1st_stage.yaml diff --git a/mlflow/sf2f/output/sf2f_1st_stage-20240303-231044/sf2f_1st_stage.yaml b/train/mlflow/sf2f/output/sf2f_1st_stage-20240303-231044/sf2f_1st_stage.yaml similarity index 100% rename from mlflow/sf2f/output/sf2f_1st_stage-20240303-231044/sf2f_1st_stage.yaml rename to train/mlflow/sf2f/output/sf2f_1st_stage-20240303-231044/sf2f_1st_stage.yaml diff --git a/mlflow/sf2f/output/sf2f_1st_stage-20240303-231201/sf2f_1st_stage.yaml b/train/mlflow/sf2f/output/sf2f_1st_stage-20240303-231201/sf2f_1st_stage.yaml similarity index 100% rename from mlflow/sf2f/output/sf2f_1st_stage-20240303-231201/sf2f_1st_stage.yaml rename to train/mlflow/sf2f/output/sf2f_1st_stage-20240303-231201/sf2f_1st_stage.yaml diff --git a/mlflow/sf2f/output/sf2f_1st_stage-20240303-231308/sf2f_1st_stage.yaml b/train/mlflow/sf2f/output/sf2f_1st_stage-20240303-231308/sf2f_1st_stage.yaml similarity index 100% rename from mlflow/sf2f/output/sf2f_1st_stage-20240303-231308/sf2f_1st_stage.yaml rename to train/mlflow/sf2f/output/sf2f_1st_stage-20240303-231308/sf2f_1st_stage.yaml diff --git a/mlflow/sf2f/output/sf2f_1st_stage-20240303-231542/sf2f_1st_stage.yaml b/train/mlflow/sf2f/output/sf2f_1st_stage-20240303-231542/sf2f_1st_stage.yaml similarity index 100% rename from mlflow/sf2f/output/sf2f_1st_stage-20240303-231542/sf2f_1st_stage.yaml rename to train/mlflow/sf2f/output/sf2f_1st_stage-20240303-231542/sf2f_1st_stage.yaml diff --git a/mlflow/sf2f/output/sf2f_1st_stage-20240303-231831/sf2f_1st_stage.yaml b/train/mlflow/sf2f/output/sf2f_1st_stage-20240303-231831/sf2f_1st_stage.yaml similarity index 100% rename from mlflow/sf2f/output/sf2f_1st_stage-20240303-231831/sf2f_1st_stage.yaml rename to train/mlflow/sf2f/output/sf2f_1st_stage-20240303-231831/sf2f_1st_stage.yaml diff --git a/mlflow/sf2f/output/sf2f_1st_stage-20240303-232325/sf2f_1st_stage.yaml b/train/mlflow/sf2f/output/sf2f_1st_stage-20240303-232325/sf2f_1st_stage.yaml similarity index 100% rename from mlflow/sf2f/output/sf2f_1st_stage-20240303-232325/sf2f_1st_stage.yaml rename to train/mlflow/sf2f/output/sf2f_1st_stage-20240303-232325/sf2f_1st_stage.yaml diff --git a/mlflow/sf2f/output/sf2f_1st_stage-20240303-233127/sf2f_1st_stage.yaml b/train/mlflow/sf2f/output/sf2f_1st_stage-20240303-233127/sf2f_1st_stage.yaml similarity index 100% rename from mlflow/sf2f/output/sf2f_1st_stage-20240303-233127/sf2f_1st_stage.yaml rename to train/mlflow/sf2f/output/sf2f_1st_stage-20240303-233127/sf2f_1st_stage.yaml diff --git a/mlflow/sf2f/output/sf2f_1st_stage-20240303-233311/sf2f_1st_stage.yaml b/train/mlflow/sf2f/output/sf2f_1st_stage-20240303-233311/sf2f_1st_stage.yaml similarity index 100% rename from mlflow/sf2f/output/sf2f_1st_stage-20240303-233311/sf2f_1st_stage.yaml rename to train/mlflow/sf2f/output/sf2f_1st_stage-20240303-233311/sf2f_1st_stage.yaml diff --git a/mlflow/sf2f/output/sf2f_1st_stage-20240303-233619/sf2f_1st_stage.yaml b/train/mlflow/sf2f/output/sf2f_1st_stage-20240303-233619/sf2f_1st_stage.yaml similarity index 100% rename from mlflow/sf2f/output/sf2f_1st_stage-20240303-233619/sf2f_1st_stage.yaml rename to train/mlflow/sf2f/output/sf2f_1st_stage-20240303-233619/sf2f_1st_stage.yaml diff --git a/mlflow/sf2f/scripts/DScore/__init__.py b/train/mlflow/sf2f/scripts/DScore/__init__.py similarity index 100% rename from mlflow/sf2f/scripts/DScore/__init__.py rename to train/mlflow/sf2f/scripts/DScore/__init__.py diff --git a/mlflow/sf2f/scripts/DScore/models/__init__.py b/train/mlflow/sf2f/scripts/DScore/models/__init__.py similarity index 100% rename from mlflow/sf2f/scripts/DScore/models/__init__.py rename to train/mlflow/sf2f/scripts/DScore/models/__init__.py diff --git a/mlflow/sf2f/scripts/DScore/models/base_model.py b/train/mlflow/sf2f/scripts/DScore/models/base_model.py similarity index 100% rename from mlflow/sf2f/scripts/DScore/models/base_model.py rename to train/mlflow/sf2f/scripts/DScore/models/base_model.py diff --git a/mlflow/sf2f/scripts/DScore/models/dist_model.py b/train/mlflow/sf2f/scripts/DScore/models/dist_model.py similarity index 100% rename from mlflow/sf2f/scripts/DScore/models/dist_model.py rename to train/mlflow/sf2f/scripts/DScore/models/dist_model.py diff --git a/mlflow/sf2f/scripts/DScore/models/models.py b/train/mlflow/sf2f/scripts/DScore/models/models.py similarity index 100% rename from mlflow/sf2f/scripts/DScore/models/models.py rename to train/mlflow/sf2f/scripts/DScore/models/models.py diff --git a/mlflow/sf2f/scripts/DScore/models/networks_basic.py b/train/mlflow/sf2f/scripts/DScore/models/networks_basic.py similarity index 100% rename from mlflow/sf2f/scripts/DScore/models/networks_basic.py rename to train/mlflow/sf2f/scripts/DScore/models/networks_basic.py diff --git a/mlflow/sf2f/scripts/DScore/models/pretrained_networks.py b/train/mlflow/sf2f/scripts/DScore/models/pretrained_networks.py similarity index 100% rename from mlflow/sf2f/scripts/DScore/models/pretrained_networks.py rename to train/mlflow/sf2f/scripts/DScore/models/pretrained_networks.py diff --git a/mlflow/sf2f/scripts/DScore/util/__init__.py b/train/mlflow/sf2f/scripts/DScore/util/__init__.py similarity index 100% rename from mlflow/sf2f/scripts/DScore/util/__init__.py rename to train/mlflow/sf2f/scripts/DScore/util/__init__.py diff --git a/mlflow/sf2f/scripts/DScore/util/html.py b/train/mlflow/sf2f/scripts/DScore/util/html.py similarity index 100% rename from mlflow/sf2f/scripts/DScore/util/html.py rename to train/mlflow/sf2f/scripts/DScore/util/html.py diff --git a/mlflow/sf2f/scripts/DScore/util/util.py b/train/mlflow/sf2f/scripts/DScore/util/util.py similarity index 100% rename from mlflow/sf2f/scripts/DScore/util/util.py rename to train/mlflow/sf2f/scripts/DScore/util/util.py diff --git a/mlflow/sf2f/scripts/DScore/util/visualizer.py b/train/mlflow/sf2f/scripts/DScore/util/visualizer.py similarity index 100% rename from mlflow/sf2f/scripts/DScore/util/visualizer.py rename to train/mlflow/sf2f/scripts/DScore/util/visualizer.py diff --git a/mlflow/sf2f/scripts/build_demo_set.py b/train/mlflow/sf2f/scripts/build_demo_set.py similarity index 100% rename from mlflow/sf2f/scripts/build_demo_set.py rename to train/mlflow/sf2f/scripts/build_demo_set.py diff --git a/mlflow/sf2f/scripts/compute_diversity_score.py b/train/mlflow/sf2f/scripts/compute_diversity_score.py similarity index 100% rename from mlflow/sf2f/scripts/compute_diversity_score.py rename to train/mlflow/sf2f/scripts/compute_diversity_score.py diff --git a/mlflow/sf2f/scripts/compute_fid_score.py b/train/mlflow/sf2f/scripts/compute_fid_score.py similarity index 100% rename from mlflow/sf2f/scripts/compute_fid_score.py rename to train/mlflow/sf2f/scripts/compute_fid_score.py diff --git a/mlflow/sf2f/scripts/compute_inception_score.py b/train/mlflow/sf2f/scripts/compute_inception_score.py similarity index 100% rename from mlflow/sf2f/scripts/compute_inception_score.py rename to train/mlflow/sf2f/scripts/compute_inception_score.py diff --git a/mlflow/sf2f/scripts/compute_mel_mean_var.py b/train/mlflow/sf2f/scripts/compute_mel_mean_var.py similarity index 100% rename from mlflow/sf2f/scripts/compute_mel_mean_var.py rename to train/mlflow/sf2f/scripts/compute_mel_mean_var.py diff --git a/mlflow/sf2f/scripts/compute_vggface_score.py b/train/mlflow/sf2f/scripts/compute_vggface_score.py similarity index 100% rename from mlflow/sf2f/scripts/compute_vggface_score.py rename to train/mlflow/sf2f/scripts/compute_vggface_score.py diff --git a/mlflow/sf2f/scripts/convert_wav_to_mel.py b/train/mlflow/sf2f/scripts/convert_wav_to_mel.py similarity index 100% rename from mlflow/sf2f/scripts/convert_wav_to_mel.py rename to train/mlflow/sf2f/scripts/convert_wav_to_mel.py diff --git a/mlflow/sf2f/scripts/create_split_json.py b/train/mlflow/sf2f/scripts/create_split_json.py similarity index 100% rename from mlflow/sf2f/scripts/create_split_json.py rename to train/mlflow/sf2f/scripts/create_split_json.py diff --git a/mlflow/sf2f/scripts/download_vggface_weights.sh b/train/mlflow/sf2f/scripts/download_vggface_weights.sh similarity index 100% rename from mlflow/sf2f/scripts/download_vggface_weights.sh rename to train/mlflow/sf2f/scripts/download_vggface_weights.sh diff --git a/mlflow/sf2f/scripts/install_requirements.py b/train/mlflow/sf2f/scripts/install_requirements.py similarity index 100% rename from mlflow/sf2f/scripts/install_requirements.py rename to train/mlflow/sf2f/scripts/install_requirements.py diff --git a/mlflow/sf2f/scripts/print_args.py b/train/mlflow/sf2f/scripts/print_args.py similarity index 100% rename from mlflow/sf2f/scripts/print_args.py rename to train/mlflow/sf2f/scripts/print_args.py diff --git a/mlflow/sf2f/scripts/sample_mel_spectrograms.py b/train/mlflow/sf2f/scripts/sample_mel_spectrograms.py similarity index 100% rename from mlflow/sf2f/scripts/sample_mel_spectrograms.py rename to train/mlflow/sf2f/scripts/sample_mel_spectrograms.py diff --git a/mlflow/sf2f/scripts/strip_checkpoint.py b/train/mlflow/sf2f/scripts/strip_checkpoint.py similarity index 100% rename from mlflow/sf2f/scripts/strip_checkpoint.py rename to train/mlflow/sf2f/scripts/strip_checkpoint.py diff --git a/mlflow/sf2f/scripts/strip_old_args.py b/train/mlflow/sf2f/scripts/strip_old_args.py similarity index 100% rename from mlflow/sf2f/scripts/strip_old_args.py rename to train/mlflow/sf2f/scripts/strip_old_args.py diff --git a/mlflow/sf2f/scripts/watch_data.py b/train/mlflow/sf2f/scripts/watch_data.py similarity index 100% rename from mlflow/sf2f/scripts/watch_data.py rename to train/mlflow/sf2f/scripts/watch_data.py diff --git a/mlflow/sf2f/train.py b/train/mlflow/sf2f/train.py similarity index 100% rename from mlflow/sf2f/train.py rename to train/mlflow/sf2f/train.py diff --git a/mlflow/sf2f/train_registry.py b/train/mlflow/sf2f/train_registry.py similarity index 100% rename from mlflow/sf2f/train_registry.py rename to train/mlflow/sf2f/train_registry.py diff --git a/mlflow/sf2f/utils/__init__.py b/train/mlflow/sf2f/utils/__init__.py similarity index 100% rename from mlflow/sf2f/utils/__init__.py rename to train/mlflow/sf2f/utils/__init__.py diff --git a/mlflow/sf2f/utils/bilinear.py b/train/mlflow/sf2f/utils/bilinear.py similarity index 100% rename from mlflow/sf2f/utils/bilinear.py rename to train/mlflow/sf2f/utils/bilinear.py diff --git a/mlflow/sf2f/utils/box_utils.py b/train/mlflow/sf2f/utils/box_utils.py similarity index 100% rename from mlflow/sf2f/utils/box_utils.py rename to train/mlflow/sf2f/utils/box_utils.py diff --git a/mlflow/sf2f/utils/common.py b/train/mlflow/sf2f/utils/common.py similarity index 100% rename from mlflow/sf2f/utils/common.py rename to train/mlflow/sf2f/utils/common.py diff --git a/mlflow/sf2f/utils/evaluate.py b/train/mlflow/sf2f/utils/evaluate.py similarity index 100% rename from mlflow/sf2f/utils/evaluate.py rename to train/mlflow/sf2f/utils/evaluate.py diff --git a/mlflow/sf2f/utils/evaluate_fid.py b/train/mlflow/sf2f/utils/evaluate_fid.py similarity index 100% rename from mlflow/sf2f/utils/evaluate_fid.py rename to train/mlflow/sf2f/utils/evaluate_fid.py diff --git a/mlflow/sf2f/utils/filter_pickle.py b/train/mlflow/sf2f/utils/filter_pickle.py similarity index 100% rename from mlflow/sf2f/utils/filter_pickle.py rename to train/mlflow/sf2f/utils/filter_pickle.py diff --git a/mlflow/sf2f/utils/logger.py b/train/mlflow/sf2f/utils/logger.py similarity index 100% rename from mlflow/sf2f/utils/logger.py rename to train/mlflow/sf2f/utils/logger.py diff --git a/mlflow/sf2f/utils/losses.py b/train/mlflow/sf2f/utils/losses.py similarity index 100% rename from mlflow/sf2f/utils/losses.py rename to train/mlflow/sf2f/utils/losses.py diff --git a/mlflow/sf2f/utils/metrics.py b/train/mlflow/sf2f/utils/metrics.py similarity index 100% rename from mlflow/sf2f/utils/metrics.py rename to train/mlflow/sf2f/utils/metrics.py diff --git a/mlflow/sf2f/utils/mlflow_wandb.py b/train/mlflow/sf2f/utils/mlflow_wandb.py similarity index 100% rename from mlflow/sf2f/utils/mlflow_wandb.py rename to train/mlflow/sf2f/utils/mlflow_wandb.py diff --git a/mlflow/sf2f/utils/pgan_utils.py b/train/mlflow/sf2f/utils/pgan_utils.py similarity index 100% rename from mlflow/sf2f/utils/pgan_utils.py rename to train/mlflow/sf2f/utils/pgan_utils.py diff --git a/mlflow/sf2f/utils/s2f_evaluator.py b/train/mlflow/sf2f/utils/s2f_evaluator.py similarity index 100% rename from mlflow/sf2f/utils/s2f_evaluator.py rename to train/mlflow/sf2f/utils/s2f_evaluator.py diff --git a/mlflow/sf2f/utils/training_utils.py b/train/mlflow/sf2f/utils/training_utils.py similarity index 100% rename from mlflow/sf2f/utils/training_utils.py rename to train/mlflow/sf2f/utils/training_utils.py diff --git a/mlflow/sf2f/utils/utils.py b/train/mlflow/sf2f/utils/utils.py similarity index 100% rename from mlflow/sf2f/utils/utils.py rename to train/mlflow/sf2f/utils/utils.py diff --git a/mlflow/sf2f/utils/vad_ex.py b/train/mlflow/sf2f/utils/vad_ex.py similarity index 100% rename from mlflow/sf2f/utils/vad_ex.py rename to train/mlflow/sf2f/utils/vad_ex.py diff --git a/mlflow/sf2f/utils/visualization/__init__.py b/train/mlflow/sf2f/utils/visualization/__init__.py similarity index 100% rename from mlflow/sf2f/utils/visualization/__init__.py rename to train/mlflow/sf2f/utils/visualization/__init__.py diff --git a/mlflow/sf2f/utils/visualization/html.py b/train/mlflow/sf2f/utils/visualization/html.py similarity index 100% rename from mlflow/sf2f/utils/visualization/html.py rename to train/mlflow/sf2f/utils/visualization/html.py diff --git a/mlflow/sf2f/utils/visualization/plot.py b/train/mlflow/sf2f/utils/visualization/plot.py similarity index 100% rename from mlflow/sf2f/utils/visualization/plot.py rename to train/mlflow/sf2f/utils/visualization/plot.py diff --git a/mlflow/sf2f/utils/visualization/tsne.py b/train/mlflow/sf2f/utils/visualization/tsne.py similarity index 100% rename from mlflow/sf2f/utils/visualization/tsne.py rename to train/mlflow/sf2f/utils/visualization/tsne.py diff --git a/mlflow/sf2f/utils/visualization/vis.py b/train/mlflow/sf2f/utils/visualization/vis.py similarity index 100% rename from mlflow/sf2f/utils/visualization/vis.py rename to train/mlflow/sf2f/utils/visualization/vis.py diff --git a/mlflow/sf2f/utils/wav2mel.py b/train/mlflow/sf2f/utils/wav2mel.py similarity index 100% rename from mlflow/sf2f/utils/wav2mel.py rename to train/mlflow/sf2f/utils/wav2mel.py From 9eb9182a3e240f3c2ad1aade8de6899e926494f2 Mon Sep 17 00:00:00 2001 From: internationalwe Date: Fri, 8 Mar 2024 14:15:52 +0000 Subject: [PATCH 5/8] Rename : model train folder structure change - - #3 --- .gitignore | 4 +- .../Swimswap/checkpoints/people/iter.txt | 2 + .../Swimswap/checkpoints/people/loss_log.txt | 11208 ++++++++++++++++ .../Swimswap/checkpoints/people/opt.txt | 72 + .../Swimswap/insightface_func}/__init__.py | 0 .../face_detect_crop_multi.py | 100 + .../face_detect_crop_single.py | 97 + .../utils/face_align_ffhqandnewarc.py | 159 + mlflow/registry/Swimswap/model_registry.py | 66 + mlflow/registry/Swimswap/models/__init__.py | 4 + .../Swimswap/models/arcface_models.py | 163 + mlflow/registry/Swimswap/models/base_model.py | 138 + mlflow/registry/Swimswap/models/config.py | 28 + mlflow/registry/Swimswap/models/fs_model.py | 242 + .../registry/Swimswap/models/fs_networks.py | 215 + .../Swimswap/models/fs_networks_512.py | 232 + .../Swimswap/models/fs_networks_fix.py | 172 + mlflow/registry/Swimswap/models/models.py | 177 + mlflow/registry/Swimswap/models/networks.py | 845 ++ .../Swimswap/models/pix2pixHD_model.py | 304 + .../Swimswap/models/projected_model.py | 122 + .../Swimswap/models/projectionhead.py | 14 + mlflow/registry/Swimswap/models/ui_model.py | 347 + .../registry/Swimswap/options/base_options.py | 104 + .../registry/Swimswap/options/test_options.py | 38 + .../registry/Swimswap/parsing_model/model.py | 283 + .../registry/Swimswap/parsing_model/resnet.py | 109 + mlflow/registry/Swimswap/pg_modules/blocks.py | 325 + .../registry/Swimswap/pg_modules/diffaug.py | 76 + .../pg_modules/projected_discriminator.py | 191 + .../registry/Swimswap/pg_modules/projector.py | 158 + .../registry/Swimswap/util/add_watermark.py | 118 + mlflow/registry/Swimswap/util/gifswap.py | 129 + mlflow/registry/Swimswap/util/html.py | 63 + mlflow/registry/Swimswap/util/image_pool.py | 31 + mlflow/registry/Swimswap/util/json_config.py | 15 + mlflow/registry/Swimswap/util/logo_class.py | 44 + mlflow/registry/Swimswap/util/norm.py | 25 + mlflow/registry/Swimswap/util/plot.py | 37 + .../Swimswap/util/reverse2original.py | 176 + mlflow/registry/Swimswap/util/save_heatmap.py | 57 + mlflow/registry/Swimswap/util/util.py | 100 + mlflow/registry/Swimswap/util/videoswap.py | 121 + .../Swimswap/util/videoswap_multispecific.py | 146 + .../Swimswap/util/videoswap_specific.py | 130 + mlflow/registry/Swimswap/util/visualizer.py | 131 + .../train}/sf2f/GETTING_STARTED.md | 0 .../sf2f/data/\353\205\271\354\235\214.wav" | Bin .../train}/sf2f/datasets/__init__.py | 0 .../train}/sf2f/datasets/build_dataset.py | 0 .../train}/sf2f/datasets/utils.py | 0 .../train}/sf2f/datasets/vox_dataset.py | 0 {train/mlflow => mlflow/train}/sf2f/infer.py | 0 .../train}/sf2f/inference_fuser.py | 0 .../train}/sf2f/model_registry.py | 10 +- .../train}/sf2f/models/__init__.py | 0 .../train}/sf2f/models/attention.py | 0 .../train}/sf2f/models/crn.py | 0 .../train}/sf2f/models/discriminators.py | 0 .../train}/sf2f/models/encoder_decoder.py | 0 .../train}/sf2f/models/face_decoders.py | 0 .../train}/sf2f/models/fusers.py | 0 .../train}/sf2f/models/inception_resnet_v1.py | 0 .../train}/sf2f/models/layers.py | 0 .../train}/sf2f/models/model_collection.py | 0 .../train}/sf2f/models/model_setup.py | 0 .../train}/sf2f/models/networks.py | 0 .../train}/sf2f/models/perceptual.py | 0 .../train}/sf2f/models/voice_encoders.py | 0 .../train/sf2f/options}/__init__.py | 0 .../train}/sf2f/options/data_opts/vox.yaml | 0 .../train}/sf2f/options/opts.py | 0 .../train}/sf2f/options/vox/baseline/v2f.yaml | 0 .../sf2f/options/vox/sf2f/sf2f_1st_stage.yaml | 0 .../sf2f/options/vox/sf2f/sf2f_fuser.yaml | 2 +- .../options/vox/sf2f/sf2f_mid_1st_stage.yaml | 0 .../sf2f/options/vox/sf2f/sf2f_mid_fuser.yaml | 0 .../sf2f_1st_stage.yaml | 0 .../sf2f_1st_stage.yaml | 0 .../sf2f_1st_stage.yaml | 0 .../sf2f_1st_stage.yaml | 0 .../sf2f_1st_stage.yaml | 0 .../sf2f_1st_stage.yaml | 0 .../sf2f_1st_stage.yaml | 0 .../sf2f_1st_stage.yaml | 0 .../sf2f_1st_stage.yaml | 0 .../sf2f_1st_stage.yaml | 0 .../sf2f_1st_stage.yaml | 0 .../sf2f_1st_stage.yaml | 0 .../sf2f_1st_stage.yaml | 0 .../sf2f_1st_stage.yaml | 0 .../sf2f_1st_stage.yaml | 0 .../sf2f_1st_stage.yaml | 0 .../sf2f_1st_stage.yaml | 0 .../sf2f_1st_stage.yaml | 0 .../sf2f_1st_stage.yaml | 0 .../sf2f_1st_stage.yaml | 0 .../sf2f_1st_stage.yaml | 0 .../sf2f_1st_stage.yaml | 0 .../train/sf2f/scripts/DScore}/__init__.py | 0 .../sf2f/scripts/DScore/models}/__init__.py | 0 .../sf2f/scripts/DScore/models/base_model.py | 0 .../sf2f/scripts/DScore/models/dist_model.py | 0 .../sf2f/scripts/DScore/models/models.py | 0 .../scripts/DScore/models/networks_basic.py | 0 .../DScore/models/pretrained_networks.py | 0 .../sf2f/scripts/DScore/util/__init__.py | 0 .../train}/sf2f/scripts/DScore/util/html.py | 0 .../train}/sf2f/scripts/DScore/util/util.py | 0 .../sf2f/scripts/DScore/util/visualizer.py | 0 .../train}/sf2f/scripts/build_demo_set.py | 0 .../sf2f/scripts/compute_diversity_score.py | 0 .../train}/sf2f/scripts/compute_fid_score.py | 0 .../sf2f/scripts/compute_inception_score.py | 0 .../sf2f/scripts/compute_mel_mean_var.py | 0 .../sf2f/scripts/compute_vggface_score.py | 0 .../train}/sf2f/scripts/convert_wav_to_mel.py | 0 .../train}/sf2f/scripts/create_split_json.py | 0 .../sf2f/scripts/download_vggface_weights.sh | 0 .../sf2f/scripts/install_requirements.py | 0 .../train}/sf2f/scripts/print_args.py | 0 .../sf2f/scripts/sample_mel_spectrograms.py | 0 .../train}/sf2f/scripts/strip_checkpoint.py | 0 .../train}/sf2f/scripts/strip_old_args.py | 0 .../train}/sf2f/scripts/watch_data.py | 0 {train/mlflow => mlflow/train}/sf2f/train.py | 0 .../train}/sf2f/train_registry.py | 0 .../train}/sf2f/utils/__init__.py | 0 .../train}/sf2f/utils/bilinear.py | 0 .../train}/sf2f/utils/box_utils.py | 0 .../train}/sf2f/utils/common.py | 0 .../train}/sf2f/utils/evaluate.py | 0 .../train}/sf2f/utils/evaluate_fid.py | 0 .../train}/sf2f/utils/filter_pickle.py | 0 .../train}/sf2f/utils/logger.py | 0 .../train}/sf2f/utils/losses.py | 0 .../train}/sf2f/utils/metrics.py | 0 .../train}/sf2f/utils/mlflow_wandb.py | 0 .../train}/sf2f/utils/pgan_utils.py | 0 .../train}/sf2f/utils/s2f_evaluator.py | 0 .../train}/sf2f/utils/training_utils.py | 0 .../train}/sf2f/utils/utils.py | 0 .../train}/sf2f/utils/vad_ex.py | 0 .../sf2f/utils/visualization/__init__.py | 0 .../train}/sf2f/utils/visualization/html.py | 0 .../train}/sf2f/utils/visualization/plot.py | 0 .../train}/sf2f/utils/visualization/tsne.py | 0 .../train}/sf2f/utils/visualization/vis.py | 0 .../train}/sf2f/utils/wav2mel.py | 0 149 files changed, 17323 insertions(+), 7 deletions(-) create mode 100644 mlflow/registry/Swimswap/checkpoints/people/iter.txt create mode 100644 mlflow/registry/Swimswap/checkpoints/people/loss_log.txt create mode 100644 mlflow/registry/Swimswap/checkpoints/people/opt.txt rename {train/mlflow/sf2f/options => mlflow/registry/Swimswap/insightface_func}/__init__.py (100%) create mode 100644 mlflow/registry/Swimswap/insightface_func/face_detect_crop_multi.py create mode 100644 mlflow/registry/Swimswap/insightface_func/face_detect_crop_single.py create mode 100644 mlflow/registry/Swimswap/insightface_func/utils/face_align_ffhqandnewarc.py create mode 100644 mlflow/registry/Swimswap/model_registry.py create mode 100644 mlflow/registry/Swimswap/models/__init__.py create mode 100644 mlflow/registry/Swimswap/models/arcface_models.py create mode 100644 mlflow/registry/Swimswap/models/base_model.py create mode 100644 mlflow/registry/Swimswap/models/config.py create mode 100644 mlflow/registry/Swimswap/models/fs_model.py create mode 100644 mlflow/registry/Swimswap/models/fs_networks.py create mode 100644 mlflow/registry/Swimswap/models/fs_networks_512.py create mode 100644 mlflow/registry/Swimswap/models/fs_networks_fix.py create mode 100644 mlflow/registry/Swimswap/models/models.py create mode 100644 mlflow/registry/Swimswap/models/networks.py create mode 100644 mlflow/registry/Swimswap/models/pix2pixHD_model.py create mode 100644 mlflow/registry/Swimswap/models/projected_model.py create mode 100644 mlflow/registry/Swimswap/models/projectionhead.py create mode 100644 mlflow/registry/Swimswap/models/ui_model.py create mode 100644 mlflow/registry/Swimswap/options/base_options.py create mode 100644 mlflow/registry/Swimswap/options/test_options.py create mode 100644 mlflow/registry/Swimswap/parsing_model/model.py create mode 100644 mlflow/registry/Swimswap/parsing_model/resnet.py create mode 100644 mlflow/registry/Swimswap/pg_modules/blocks.py create mode 100644 mlflow/registry/Swimswap/pg_modules/diffaug.py create mode 100644 mlflow/registry/Swimswap/pg_modules/projected_discriminator.py create mode 100644 mlflow/registry/Swimswap/pg_modules/projector.py create mode 100644 mlflow/registry/Swimswap/util/add_watermark.py create mode 100644 mlflow/registry/Swimswap/util/gifswap.py create mode 100644 mlflow/registry/Swimswap/util/html.py create mode 100644 mlflow/registry/Swimswap/util/image_pool.py create mode 100644 mlflow/registry/Swimswap/util/json_config.py create mode 100644 mlflow/registry/Swimswap/util/logo_class.py create mode 100644 mlflow/registry/Swimswap/util/norm.py create mode 100644 mlflow/registry/Swimswap/util/plot.py create mode 100644 mlflow/registry/Swimswap/util/reverse2original.py create mode 100644 mlflow/registry/Swimswap/util/save_heatmap.py create mode 100644 mlflow/registry/Swimswap/util/util.py create mode 100644 mlflow/registry/Swimswap/util/videoswap.py create mode 100644 mlflow/registry/Swimswap/util/videoswap_multispecific.py create mode 100644 mlflow/registry/Swimswap/util/videoswap_specific.py create mode 100644 mlflow/registry/Swimswap/util/visualizer.py rename {train/mlflow => mlflow/train}/sf2f/GETTING_STARTED.md (100%) rename "train/mlflow/sf2f/data/\353\205\271\354\235\214.wav" => "mlflow/train/sf2f/data/\353\205\271\354\235\214.wav" (100%) rename {train/mlflow => mlflow/train}/sf2f/datasets/__init__.py (100%) rename {train/mlflow => mlflow/train}/sf2f/datasets/build_dataset.py (100%) rename {train/mlflow => mlflow/train}/sf2f/datasets/utils.py (100%) rename {train/mlflow => mlflow/train}/sf2f/datasets/vox_dataset.py (100%) rename {train/mlflow => mlflow/train}/sf2f/infer.py (100%) rename {train/mlflow => mlflow/train}/sf2f/inference_fuser.py (100%) rename {train/mlflow => mlflow/train}/sf2f/model_registry.py (86%) rename {train/mlflow => mlflow/train}/sf2f/models/__init__.py (100%) rename {train/mlflow => mlflow/train}/sf2f/models/attention.py (100%) rename {train/mlflow => mlflow/train}/sf2f/models/crn.py (100%) rename {train/mlflow => mlflow/train}/sf2f/models/discriminators.py (100%) rename {train/mlflow => mlflow/train}/sf2f/models/encoder_decoder.py (100%) rename {train/mlflow => mlflow/train}/sf2f/models/face_decoders.py (100%) rename {train/mlflow => mlflow/train}/sf2f/models/fusers.py (100%) rename {train/mlflow => mlflow/train}/sf2f/models/inception_resnet_v1.py (100%) rename {train/mlflow => mlflow/train}/sf2f/models/layers.py (100%) rename {train/mlflow => mlflow/train}/sf2f/models/model_collection.py (100%) rename {train/mlflow => mlflow/train}/sf2f/models/model_setup.py (100%) rename {train/mlflow => mlflow/train}/sf2f/models/networks.py (100%) rename {train/mlflow => mlflow/train}/sf2f/models/perceptual.py (100%) rename {train/mlflow => mlflow/train}/sf2f/models/voice_encoders.py (100%) rename {train/mlflow/sf2f/scripts/DScore => mlflow/train/sf2f/options}/__init__.py (100%) rename {train/mlflow => mlflow/train}/sf2f/options/data_opts/vox.yaml (100%) rename {train/mlflow => mlflow/train}/sf2f/options/opts.py (100%) rename {train/mlflow => mlflow/train}/sf2f/options/vox/baseline/v2f.yaml (100%) rename {train/mlflow => mlflow/train}/sf2f/options/vox/sf2f/sf2f_1st_stage.yaml (100%) rename {train/mlflow => mlflow/train}/sf2f/options/vox/sf2f/sf2f_fuser.yaml (96%) rename {train/mlflow => mlflow/train}/sf2f/options/vox/sf2f/sf2f_mid_1st_stage.yaml (100%) rename {train/mlflow => mlflow/train}/sf2f/options/vox/sf2f/sf2f_mid_fuser.yaml (100%) rename {train/mlflow => mlflow/train}/sf2f/output/sf2f_1st_stage-20240303-215640/sf2f_1st_stage.yaml (100%) rename {train/mlflow => mlflow/train}/sf2f/output/sf2f_1st_stage-20240303-220349/sf2f_1st_stage.yaml (100%) rename {train/mlflow => mlflow/train}/sf2f/output/sf2f_1st_stage-20240303-221106/sf2f_1st_stage.yaml (100%) rename {train/mlflow => mlflow/train}/sf2f/output/sf2f_1st_stage-20240303-225403/sf2f_1st_stage.yaml (100%) rename {train/mlflow => mlflow/train}/sf2f/output/sf2f_1st_stage-20240303-225533/sf2f_1st_stage.yaml (100%) rename {train/mlflow => mlflow/train}/sf2f/output/sf2f_1st_stage-20240303-225656/sf2f_1st_stage.yaml (100%) rename {train/mlflow => mlflow/train}/sf2f/output/sf2f_1st_stage-20240303-225950/sf2f_1st_stage.yaml (100%) rename {train/mlflow => mlflow/train}/sf2f/output/sf2f_1st_stage-20240303-230032/sf2f_1st_stage.yaml (100%) rename {train/mlflow => mlflow/train}/sf2f/output/sf2f_1st_stage-20240303-230142/sf2f_1st_stage.yaml (100%) rename {train/mlflow => mlflow/train}/sf2f/output/sf2f_1st_stage-20240303-230232/sf2f_1st_stage.yaml (100%) rename {train/mlflow => mlflow/train}/sf2f/output/sf2f_1st_stage-20240303-230355/sf2f_1st_stage.yaml (100%) rename {train/mlflow => mlflow/train}/sf2f/output/sf2f_1st_stage-20240303-230514/sf2f_1st_stage.yaml (100%) rename {train/mlflow => mlflow/train}/sf2f/output/sf2f_1st_stage-20240303-230631/sf2f_1st_stage.yaml (100%) rename {train/mlflow => mlflow/train}/sf2f/output/sf2f_1st_stage-20240303-231044/sf2f_1st_stage.yaml (100%) rename {train/mlflow => mlflow/train}/sf2f/output/sf2f_1st_stage-20240303-231201/sf2f_1st_stage.yaml (100%) rename {train/mlflow => mlflow/train}/sf2f/output/sf2f_1st_stage-20240303-231308/sf2f_1st_stage.yaml (100%) rename {train/mlflow => mlflow/train}/sf2f/output/sf2f_1st_stage-20240303-231542/sf2f_1st_stage.yaml (100%) rename {train/mlflow => mlflow/train}/sf2f/output/sf2f_1st_stage-20240303-231831/sf2f_1st_stage.yaml (100%) rename {train/mlflow => mlflow/train}/sf2f/output/sf2f_1st_stage-20240303-232325/sf2f_1st_stage.yaml (100%) rename {train/mlflow => mlflow/train}/sf2f/output/sf2f_1st_stage-20240303-233127/sf2f_1st_stage.yaml (100%) rename {train/mlflow => mlflow/train}/sf2f/output/sf2f_1st_stage-20240303-233311/sf2f_1st_stage.yaml (100%) rename {train/mlflow => mlflow/train}/sf2f/output/sf2f_1st_stage-20240303-233619/sf2f_1st_stage.yaml (100%) rename {train/mlflow/sf2f/scripts/DScore/models => mlflow/train/sf2f/scripts/DScore}/__init__.py (100%) rename {train/mlflow/sf2f/scripts/DScore/util => mlflow/train/sf2f/scripts/DScore/models}/__init__.py (100%) rename {train/mlflow => mlflow/train}/sf2f/scripts/DScore/models/base_model.py (100%) rename {train/mlflow => mlflow/train}/sf2f/scripts/DScore/models/dist_model.py (100%) rename {train/mlflow => mlflow/train}/sf2f/scripts/DScore/models/models.py (100%) rename {train/mlflow => mlflow/train}/sf2f/scripts/DScore/models/networks_basic.py (100%) rename {train/mlflow => mlflow/train}/sf2f/scripts/DScore/models/pretrained_networks.py (100%) create mode 100644 mlflow/train/sf2f/scripts/DScore/util/__init__.py rename {train/mlflow => mlflow/train}/sf2f/scripts/DScore/util/html.py (100%) rename {train/mlflow => mlflow/train}/sf2f/scripts/DScore/util/util.py (100%) rename {train/mlflow => mlflow/train}/sf2f/scripts/DScore/util/visualizer.py (100%) rename {train/mlflow => mlflow/train}/sf2f/scripts/build_demo_set.py (100%) rename {train/mlflow => mlflow/train}/sf2f/scripts/compute_diversity_score.py (100%) rename {train/mlflow => mlflow/train}/sf2f/scripts/compute_fid_score.py (100%) rename {train/mlflow => mlflow/train}/sf2f/scripts/compute_inception_score.py (100%) rename {train/mlflow => mlflow/train}/sf2f/scripts/compute_mel_mean_var.py (100%) rename {train/mlflow => mlflow/train}/sf2f/scripts/compute_vggface_score.py (100%) rename {train/mlflow => mlflow/train}/sf2f/scripts/convert_wav_to_mel.py (100%) rename {train/mlflow => mlflow/train}/sf2f/scripts/create_split_json.py (100%) rename {train/mlflow => mlflow/train}/sf2f/scripts/download_vggface_weights.sh (100%) rename {train/mlflow => mlflow/train}/sf2f/scripts/install_requirements.py (100%) rename {train/mlflow => mlflow/train}/sf2f/scripts/print_args.py (100%) rename {train/mlflow => mlflow/train}/sf2f/scripts/sample_mel_spectrograms.py (100%) rename {train/mlflow => mlflow/train}/sf2f/scripts/strip_checkpoint.py (100%) rename {train/mlflow => mlflow/train}/sf2f/scripts/strip_old_args.py (100%) rename {train/mlflow => mlflow/train}/sf2f/scripts/watch_data.py (100%) rename {train/mlflow => mlflow/train}/sf2f/train.py (100%) rename {train/mlflow => mlflow/train}/sf2f/train_registry.py (100%) rename {train/mlflow => mlflow/train}/sf2f/utils/__init__.py (100%) rename {train/mlflow => mlflow/train}/sf2f/utils/bilinear.py (100%) rename {train/mlflow => mlflow/train}/sf2f/utils/box_utils.py (100%) rename {train/mlflow => mlflow/train}/sf2f/utils/common.py (100%) rename {train/mlflow => mlflow/train}/sf2f/utils/evaluate.py (100%) rename {train/mlflow => mlflow/train}/sf2f/utils/evaluate_fid.py (100%) rename {train/mlflow => mlflow/train}/sf2f/utils/filter_pickle.py (100%) rename {train/mlflow => mlflow/train}/sf2f/utils/logger.py (100%) rename {train/mlflow => mlflow/train}/sf2f/utils/losses.py (100%) rename {train/mlflow => mlflow/train}/sf2f/utils/metrics.py (100%) rename {train/mlflow => mlflow/train}/sf2f/utils/mlflow_wandb.py (100%) rename {train/mlflow => mlflow/train}/sf2f/utils/pgan_utils.py (100%) rename {train/mlflow => mlflow/train}/sf2f/utils/s2f_evaluator.py (100%) rename {train/mlflow => mlflow/train}/sf2f/utils/training_utils.py (100%) rename {train/mlflow => mlflow/train}/sf2f/utils/utils.py (100%) rename {train/mlflow => mlflow/train}/sf2f/utils/vad_ex.py (100%) rename {train/mlflow => mlflow/train}/sf2f/utils/visualization/__init__.py (100%) rename {train/mlflow => mlflow/train}/sf2f/utils/visualization/html.py (100%) rename {train/mlflow => mlflow/train}/sf2f/utils/visualization/plot.py (100%) rename {train/mlflow => mlflow/train}/sf2f/utils/visualization/tsne.py (100%) rename {train/mlflow => mlflow/train}/sf2f/utils/visualization/vis.py (100%) rename {train/mlflow => mlflow/train}/sf2f/utils/wav2mel.py (100%) diff --git a/.gitignore b/.gitignore index 8729547..e0f7aa3 100644 --- a/.gitignore +++ b/.gitignore @@ -171,4 +171,6 @@ tombstones prometheus/prometheus-volume/data/* index 0000* -meta.json \ No newline at end of file +meta.json +*.tar +*.pth \ No newline at end of file diff --git a/mlflow/registry/Swimswap/checkpoints/people/iter.txt b/mlflow/registry/Swimswap/checkpoints/people/iter.txt new file mode 100644 index 0000000..f05eb86 --- /dev/null +++ b/mlflow/registry/Swimswap/checkpoints/people/iter.txt @@ -0,0 +1,2 @@ +519 +4062 diff --git a/mlflow/registry/Swimswap/checkpoints/people/loss_log.txt b/mlflow/registry/Swimswap/checkpoints/people/loss_log.txt new file mode 100644 index 0000000..0585ac8 --- /dev/null +++ b/mlflow/registry/Swimswap/checkpoints/people/loss_log.txt @@ -0,0 +1,11208 @@ +================ Training Loss (Sun Apr 19 12:26:39 2020) ================ +(epoch: 1, iters: 400, time: 0.075) G_GAN: -0.149 G_GAN_Feat: 2.780 G_ID: 1.016 G_Rec: 0.992 D_GP: 0.019 D_real: 0.583 D_fake: 1.174 +(epoch: 1, iters: 800, time: 0.063) G_GAN: -0.172 G_GAN_Feat: 2.544 G_ID: 0.748 G_Rec: 1.044 D_GP: 0.031 D_real: 0.281 D_fake: 1.176 +(epoch: 1, iters: 1200, time: 0.063) G_GAN: 0.689 G_GAN_Feat: 2.089 G_ID: 1.005 G_Rec: 0.932 D_GP: 0.028 D_real: 1.355 D_fake: 0.362 +(epoch: 1, iters: 1600, time: 0.064) G_GAN: 0.376 G_GAN_Feat: 1.997 G_ID: 0.609 G_Rec: 0.903 D_GP: 0.028 D_real: 1.281 D_fake: 0.633 +(epoch: 1, iters: 2000, time: 0.063) G_GAN: 0.175 G_GAN_Feat: 1.587 G_ID: 0.981 G_Rec: 0.788 D_GP: 0.017 D_real: 0.824 D_fake: 0.841 +(epoch: 1, iters: 2400, time: 0.063) G_GAN: 0.010 G_GAN_Feat: 1.704 G_ID: 0.563 G_Rec: 0.708 D_GP: 0.034 D_real: 0.683 D_fake: 0.991 +(epoch: 1, iters: 2800, time: 0.063) G_GAN: 0.335 G_GAN_Feat: 1.430 G_ID: 0.991 G_Rec: 0.668 D_GP: 0.022 D_real: 1.130 D_fake: 0.671 +(epoch: 1, iters: 3200, time: 0.064) G_GAN: 0.127 G_GAN_Feat: 1.707 G_ID: 0.592 G_Rec: 0.958 D_GP: 0.050 D_real: 0.967 D_fake: 0.883 +(epoch: 1, iters: 3600, time: 0.063) G_GAN: -0.091 G_GAN_Feat: 1.385 G_ID: 0.983 G_Rec: 0.648 D_GP: 0.020 D_real: 0.696 D_fake: 1.091 +(epoch: 1, iters: 4000, time: 0.063) G_GAN: 0.290 G_GAN_Feat: 1.444 G_ID: 0.584 G_Rec: 0.722 D_GP: 0.040 D_real: 0.886 D_fake: 0.713 +(epoch: 1, iters: 4400, time: 0.063) G_GAN: -0.082 G_GAN_Feat: 1.536 G_ID: 0.997 G_Rec: 0.635 D_GP: 0.044 D_real: 0.781 D_fake: 1.082 +(epoch: 1, iters: 4800, time: 0.064) G_GAN: -0.138 G_GAN_Feat: 1.452 G_ID: 0.654 G_Rec: 0.626 D_GP: 0.053 D_real: 0.733 D_fake: 1.138 +(epoch: 1, iters: 5200, time: 0.064) G_GAN: 0.011 G_GAN_Feat: 1.153 G_ID: 1.004 G_Rec: 0.532 D_GP: 0.031 D_real: 0.769 D_fake: 0.990 +(epoch: 1, iters: 5600, time: 0.063) G_GAN: -0.038 G_GAN_Feat: 1.194 G_ID: 0.619 G_Rec: 0.553 D_GP: 0.036 D_real: 0.926 D_fake: 1.040 +(epoch: 1, iters: 6000, time: 0.063) G_GAN: -0.102 G_GAN_Feat: 1.214 G_ID: 0.981 G_Rec: 0.904 D_GP: 0.056 D_real: 0.589 D_fake: 1.102 +(epoch: 1, iters: 6400, time: 0.064) G_GAN: 0.025 G_GAN_Feat: 1.220 G_ID: 0.544 G_Rec: 0.605 D_GP: 0.037 D_real: 0.987 D_fake: 0.975 +(epoch: 1, iters: 6800, time: 0.063) G_GAN: 0.309 G_GAN_Feat: 1.110 G_ID: 1.040 G_Rec: 0.548 D_GP: 0.027 D_real: 1.106 D_fake: 0.695 +(epoch: 1, iters: 7200, time: 0.064) G_GAN: -0.014 G_GAN_Feat: 1.094 G_ID: 0.531 G_Rec: 0.562 D_GP: 0.036 D_real: 0.868 D_fake: 1.015 +(epoch: 1, iters: 7600, time: 0.064) G_GAN: 0.036 G_GAN_Feat: 1.040 G_ID: 1.029 G_Rec: 0.685 D_GP: 0.026 D_real: 0.821 D_fake: 0.964 +(epoch: 1, iters: 8000, time: 0.064) G_GAN: 0.114 G_GAN_Feat: 1.039 G_ID: 0.552 G_Rec: 0.505 D_GP: 0.035 D_real: 0.922 D_fake: 0.888 +(epoch: 1, iters: 8400, time: 0.063) G_GAN: 0.206 G_GAN_Feat: 1.051 G_ID: 0.967 G_Rec: 0.502 D_GP: 0.034 D_real: 1.008 D_fake: 0.796 +(epoch: 2, iters: 192, time: 0.063) G_GAN: 0.444 G_GAN_Feat: 0.948 G_ID: 0.473 G_Rec: 1.017 D_GP: 0.013 D_real: 1.357 D_fake: 0.582 +(epoch: 2, iters: 592, time: 0.063) G_GAN: 0.275 G_GAN_Feat: 1.016 G_ID: 0.997 G_Rec: 0.490 D_GP: 0.030 D_real: 1.145 D_fake: 0.727 +(epoch: 2, iters: 992, time: 0.064) G_GAN: -0.031 G_GAN_Feat: 1.141 G_ID: 0.524 G_Rec: 0.510 D_GP: 0.041 D_real: 0.697 D_fake: 1.032 +(epoch: 2, iters: 1392, time: 0.064) G_GAN: 0.327 G_GAN_Feat: 1.060 G_ID: 1.025 G_Rec: 0.422 D_GP: 0.053 D_real: 1.016 D_fake: 0.677 +(epoch: 2, iters: 1792, time: 0.064) G_GAN: 0.208 G_GAN_Feat: 0.843 G_ID: 0.505 G_Rec: 0.653 D_GP: 0.013 D_real: 1.128 D_fake: 0.803 +(epoch: 2, iters: 2192, time: 0.064) G_GAN: 0.256 G_GAN_Feat: 0.990 G_ID: 0.992 G_Rec: 0.692 D_GP: 0.039 D_real: 0.982 D_fake: 0.745 +(epoch: 2, iters: 2592, time: 0.064) G_GAN: 0.086 G_GAN_Feat: 1.025 G_ID: 0.453 G_Rec: 0.583 D_GP: 0.048 D_real: 0.910 D_fake: 0.914 +(epoch: 2, iters: 2992, time: 0.064) G_GAN: 0.137 G_GAN_Feat: 1.077 G_ID: 0.993 G_Rec: 0.510 D_GP: 0.073 D_real: 0.690 D_fake: 0.865 +(epoch: 2, iters: 3392, time: 0.064) G_GAN: -0.145 G_GAN_Feat: 0.977 G_ID: 0.442 G_Rec: 0.431 D_GP: 0.046 D_real: 0.611 D_fake: 1.145 +(epoch: 2, iters: 3792, time: 0.064) G_GAN: 0.089 G_GAN_Feat: 0.892 G_ID: 1.006 G_Rec: 0.590 D_GP: 0.020 D_real: 1.005 D_fake: 0.915 +(epoch: 2, iters: 4192, time: 0.064) G_GAN: 0.187 G_GAN_Feat: 0.993 G_ID: 0.551 G_Rec: 0.589 D_GP: 0.040 D_real: 1.031 D_fake: 0.817 +(epoch: 2, iters: 4592, time: 0.064) G_GAN: 0.390 G_GAN_Feat: 0.996 G_ID: 0.995 G_Rec: 0.517 D_GP: 0.061 D_real: 1.158 D_fake: 0.614 +(epoch: 2, iters: 4992, time: 0.064) G_GAN: 0.342 G_GAN_Feat: 1.101 G_ID: 0.477 G_Rec: 0.843 D_GP: 0.041 D_real: 1.131 D_fake: 0.661 +(epoch: 2, iters: 5392, time: 0.064) G_GAN: 0.102 G_GAN_Feat: 0.819 G_ID: 0.987 G_Rec: 0.398 D_GP: 0.034 D_real: 0.857 D_fake: 0.898 +(epoch: 2, iters: 5792, time: 0.064) G_GAN: 0.028 G_GAN_Feat: 0.885 G_ID: 0.463 G_Rec: 0.489 D_GP: 0.022 D_real: 0.821 D_fake: 0.973 +(epoch: 2, iters: 6192, time: 0.064) G_GAN: 0.201 G_GAN_Feat: 0.875 G_ID: 1.003 G_Rec: 0.398 D_GP: 0.045 D_real: 0.997 D_fake: 0.800 +(epoch: 2, iters: 6592, time: 0.064) G_GAN: -0.040 G_GAN_Feat: 0.904 G_ID: 0.454 G_Rec: 0.434 D_GP: 0.031 D_real: 0.754 D_fake: 1.040 +(epoch: 2, iters: 6992, time: 0.064) G_GAN: 0.115 G_GAN_Feat: 0.948 G_ID: 0.934 G_Rec: 0.439 D_GP: 0.054 D_real: 0.975 D_fake: 0.891 +(epoch: 2, iters: 7392, time: 0.064) G_GAN: 0.322 G_GAN_Feat: 1.029 G_ID: 0.501 G_Rec: 0.452 D_GP: 0.061 D_real: 0.894 D_fake: 0.682 +(epoch: 2, iters: 7792, time: 0.064) G_GAN: 0.180 G_GAN_Feat: 0.882 G_ID: 1.016 G_Rec: 0.476 D_GP: 0.039 D_real: 0.923 D_fake: 0.820 +(epoch: 2, iters: 8192, time: 0.064) G_GAN: 0.004 G_GAN_Feat: 0.991 G_ID: 0.541 G_Rec: 0.780 D_GP: 0.045 D_real: 0.800 D_fake: 0.996 +(epoch: 2, iters: 8592, time: 0.064) G_GAN: -0.001 G_GAN_Feat: 0.880 G_ID: 1.019 G_Rec: 0.499 D_GP: 0.029 D_real: 0.743 D_fake: 1.001 +(epoch: 3, iters: 384, time: 0.064) G_GAN: 0.170 G_GAN_Feat: 1.037 G_ID: 0.490 G_Rec: 0.696 D_GP: 0.038 D_real: 0.917 D_fake: 0.837 +(epoch: 3, iters: 784, time: 0.064) G_GAN: 0.151 G_GAN_Feat: 0.865 G_ID: 0.990 G_Rec: 0.409 D_GP: 0.049 D_real: 0.909 D_fake: 0.849 +(epoch: 3, iters: 1184, time: 0.064) G_GAN: -0.022 G_GAN_Feat: 0.814 G_ID: 0.442 G_Rec: 0.457 D_GP: 0.024 D_real: 0.774 D_fake: 1.022 +(epoch: 3, iters: 1584, time: 0.064) G_GAN: 0.229 G_GAN_Feat: 0.892 G_ID: 0.994 G_Rec: 0.673 D_GP: 0.037 D_real: 1.000 D_fake: 0.773 +(epoch: 3, iters: 1984, time: 0.064) G_GAN: -0.190 G_GAN_Feat: 0.847 G_ID: 0.463 G_Rec: 0.375 D_GP: 0.026 D_real: 0.596 D_fake: 1.190 +(epoch: 3, iters: 2384, time: 0.064) G_GAN: 0.437 G_GAN_Feat: 0.909 G_ID: 0.992 G_Rec: 0.464 D_GP: 0.032 D_real: 1.266 D_fake: 0.577 +(epoch: 3, iters: 2784, time: 0.064) G_GAN: 0.361 G_GAN_Feat: 0.958 G_ID: 0.524 G_Rec: 0.490 D_GP: 0.068 D_real: 1.151 D_fake: 0.641 +(epoch: 3, iters: 3184, time: 0.064) G_GAN: 0.120 G_GAN_Feat: 0.720 G_ID: 0.971 G_Rec: 0.545 D_GP: 0.013 D_real: 1.015 D_fake: 0.881 +(epoch: 3, iters: 3584, time: 0.064) G_GAN: 0.117 G_GAN_Feat: 0.940 G_ID: 0.439 G_Rec: 0.497 D_GP: 0.041 D_real: 0.908 D_fake: 0.883 +(epoch: 3, iters: 3984, time: 0.064) G_GAN: 0.032 G_GAN_Feat: 0.903 G_ID: 0.975 G_Rec: 0.459 D_GP: 0.043 D_real: 0.848 D_fake: 0.969 +(epoch: 3, iters: 4384, time: 0.064) G_GAN: 0.162 G_GAN_Feat: 0.898 G_ID: 0.422 G_Rec: 0.481 D_GP: 0.035 D_real: 0.877 D_fake: 0.838 +(epoch: 3, iters: 4784, time: 0.064) G_GAN: 0.080 G_GAN_Feat: 1.111 G_ID: 0.994 G_Rec: 0.512 D_GP: 0.060 D_real: 0.630 D_fake: 0.920 +(epoch: 3, iters: 5184, time: 0.064) G_GAN: 0.272 G_GAN_Feat: 0.968 G_ID: 0.462 G_Rec: 0.464 D_GP: 0.066 D_real: 0.938 D_fake: 0.728 +(epoch: 3, iters: 5584, time: 0.064) G_GAN: 0.041 G_GAN_Feat: 1.039 G_ID: 0.982 G_Rec: 0.621 D_GP: 0.054 D_real: 0.672 D_fake: 0.960 +(epoch: 3, iters: 5984, time: 0.064) G_GAN: 0.221 G_GAN_Feat: 0.823 G_ID: 0.424 G_Rec: 0.445 D_GP: 0.035 D_real: 0.974 D_fake: 0.779 +(epoch: 3, iters: 6384, time: 0.064) G_GAN: 0.307 G_GAN_Feat: 0.915 G_ID: 1.007 G_Rec: 0.432 D_GP: 0.078 D_real: 1.004 D_fake: 0.696 +(epoch: 3, iters: 6784, time: 0.064) G_GAN: 0.216 G_GAN_Feat: 1.005 G_ID: 0.473 G_Rec: 0.534 D_GP: 0.046 D_real: 1.028 D_fake: 0.785 +(epoch: 3, iters: 7184, time: 0.064) G_GAN: 0.264 G_GAN_Feat: 0.943 G_ID: 0.988 G_Rec: 0.523 D_GP: 0.039 D_real: 1.083 D_fake: 0.738 +(epoch: 3, iters: 7584, time: 0.064) G_GAN: 0.320 G_GAN_Feat: 0.810 G_ID: 0.444 G_Rec: 0.395 D_GP: 0.021 D_real: 1.115 D_fake: 0.681 +(epoch: 3, iters: 7984, time: 0.064) G_GAN: 0.067 G_GAN_Feat: 0.844 G_ID: 0.998 G_Rec: 0.445 D_GP: 0.029 D_real: 0.793 D_fake: 0.933 +(epoch: 3, iters: 8384, time: 0.064) G_GAN: 0.103 G_GAN_Feat: 0.967 G_ID: 0.420 G_Rec: 0.580 D_GP: 0.051 D_real: 0.858 D_fake: 0.897 +(epoch: 4, iters: 176, time: 0.064) G_GAN: 0.051 G_GAN_Feat: 0.797 G_ID: 0.993 G_Rec: 0.374 D_GP: 0.036 D_real: 0.758 D_fake: 0.949 +(epoch: 4, iters: 576, time: 0.064) G_GAN: 0.193 G_GAN_Feat: 0.879 G_ID: 0.526 G_Rec: 0.431 D_GP: 0.037 D_real: 1.062 D_fake: 0.807 +(epoch: 4, iters: 976, time: 0.064) G_GAN: -0.062 G_GAN_Feat: 0.940 G_ID: 0.975 G_Rec: 0.481 D_GP: 0.042 D_real: 0.681 D_fake: 1.062 +(epoch: 4, iters: 1376, time: 0.064) G_GAN: 0.017 G_GAN_Feat: 0.884 G_ID: 0.461 G_Rec: 0.519 D_GP: 0.022 D_real: 0.939 D_fake: 0.983 +(epoch: 4, iters: 1776, time: 0.064) G_GAN: -0.003 G_GAN_Feat: 0.931 G_ID: 1.008 G_Rec: 0.499 D_GP: 0.036 D_real: 0.566 D_fake: 1.003 +(epoch: 4, iters: 2176, time: 0.064) G_GAN: 0.132 G_GAN_Feat: 0.685 G_ID: 0.473 G_Rec: 0.459 D_GP: 0.013 D_real: 0.998 D_fake: 0.870 +(epoch: 4, iters: 2576, time: 0.064) G_GAN: 0.250 G_GAN_Feat: 0.933 G_ID: 0.978 G_Rec: 0.546 D_GP: 0.046 D_real: 0.989 D_fake: 0.751 +(epoch: 4, iters: 2976, time: 0.064) G_GAN: 0.249 G_GAN_Feat: 0.849 G_ID: 0.517 G_Rec: 0.383 D_GP: 0.034 D_real: 1.061 D_fake: 0.753 +(epoch: 4, iters: 3376, time: 0.064) G_GAN: 0.077 G_GAN_Feat: 0.875 G_ID: 1.008 G_Rec: 0.442 D_GP: 0.057 D_real: 0.765 D_fake: 0.924 +(epoch: 4, iters: 3776, time: 0.064) G_GAN: 0.539 G_GAN_Feat: 0.901 G_ID: 0.517 G_Rec: 0.412 D_GP: 0.051 D_real: 1.254 D_fake: 0.484 +(epoch: 4, iters: 4176, time: 0.064) G_GAN: 0.154 G_GAN_Feat: 0.827 G_ID: 0.974 G_Rec: 0.446 D_GP: 0.056 D_real: 0.958 D_fake: 0.849 +(epoch: 4, iters: 4576, time: 0.064) G_GAN: 0.172 G_GAN_Feat: 0.848 G_ID: 0.480 G_Rec: 0.459 D_GP: 0.041 D_real: 1.059 D_fake: 0.829 +(epoch: 4, iters: 4976, time: 0.064) G_GAN: 0.527 G_GAN_Feat: 0.867 G_ID: 1.031 G_Rec: 0.513 D_GP: 0.030 D_real: 1.342 D_fake: 0.478 +(epoch: 4, iters: 5376, time: 0.064) G_GAN: 0.073 G_GAN_Feat: 1.050 G_ID: 0.511 G_Rec: 0.814 D_GP: 0.044 D_real: 0.877 D_fake: 0.928 +(epoch: 4, iters: 5776, time: 0.064) G_GAN: 0.397 G_GAN_Feat: 0.932 G_ID: 0.976 G_Rec: 0.472 D_GP: 0.052 D_real: 1.108 D_fake: 0.607 +(epoch: 4, iters: 6176, time: 0.064) G_GAN: 0.083 G_GAN_Feat: 0.963 G_ID: 0.456 G_Rec: 0.391 D_GP: 0.068 D_real: 0.705 D_fake: 0.917 +(epoch: 4, iters: 6576, time: 0.064) G_GAN: 0.440 G_GAN_Feat: 0.844 G_ID: 0.984 G_Rec: 0.406 D_GP: 0.041 D_real: 1.212 D_fake: 0.563 +(epoch: 4, iters: 6976, time: 0.064) G_GAN: -0.016 G_GAN_Feat: 0.989 G_ID: 0.425 G_Rec: 0.501 D_GP: 0.042 D_real: 0.573 D_fake: 1.016 +(epoch: 4, iters: 7376, time: 0.064) G_GAN: 0.610 G_GAN_Feat: 0.742 G_ID: 1.012 G_Rec: 0.383 D_GP: 0.023 D_real: 1.441 D_fake: 0.592 +(epoch: 4, iters: 7776, time: 0.064) G_GAN: 0.235 G_GAN_Feat: 1.001 G_ID: 0.431 G_Rec: 0.631 D_GP: 0.126 D_real: 0.979 D_fake: 0.770 +(epoch: 4, iters: 8176, time: 0.064) G_GAN: 0.130 G_GAN_Feat: 0.808 G_ID: 0.982 G_Rec: 0.436 D_GP: 0.023 D_real: 0.929 D_fake: 0.870 +(epoch: 4, iters: 8576, time: 0.064) G_GAN: 0.206 G_GAN_Feat: 0.811 G_ID: 0.495 G_Rec: 0.444 D_GP: 0.030 D_real: 1.010 D_fake: 0.794 +(epoch: 5, iters: 368, time: 0.064) G_GAN: 0.093 G_GAN_Feat: 0.792 G_ID: 1.018 G_Rec: 0.649 D_GP: 0.034 D_real: 0.835 D_fake: 0.907 +(epoch: 5, iters: 768, time: 0.064) G_GAN: 0.301 G_GAN_Feat: 0.759 G_ID: 0.467 G_Rec: 0.508 D_GP: 0.025 D_real: 1.117 D_fake: 0.699 +(epoch: 5, iters: 1168, time: 0.064) G_GAN: 0.803 G_GAN_Feat: 0.758 G_ID: 0.983 G_Rec: 0.466 D_GP: 0.016 D_real: 1.650 D_fake: 0.218 +(epoch: 5, iters: 1568, time: 0.064) G_GAN: -0.000 G_GAN_Feat: 0.901 G_ID: 0.407 G_Rec: 0.419 D_GP: 0.054 D_real: 0.693 D_fake: 1.000 +(epoch: 5, iters: 1968, time: 0.064) G_GAN: -0.033 G_GAN_Feat: 0.799 G_ID: 0.979 G_Rec: 0.488 D_GP: 0.037 D_real: 0.805 D_fake: 1.033 +(epoch: 5, iters: 2368, time: 0.064) G_GAN: 0.176 G_GAN_Feat: 0.997 G_ID: 0.475 G_Rec: 0.546 D_GP: 0.074 D_real: 0.752 D_fake: 0.825 +(epoch: 5, iters: 2768, time: 0.064) G_GAN: 0.077 G_GAN_Feat: 0.872 G_ID: 1.023 G_Rec: 0.393 D_GP: 0.042 D_real: 0.813 D_fake: 0.923 +(epoch: 5, iters: 3168, time: 0.064) G_GAN: 0.224 G_GAN_Feat: 0.864 G_ID: 0.402 G_Rec: 0.476 D_GP: 0.031 D_real: 1.038 D_fake: 0.777 +(epoch: 5, iters: 3568, time: 0.064) G_GAN: -0.044 G_GAN_Feat: 0.685 G_ID: 1.025 G_Rec: 0.601 D_GP: 0.019 D_real: 0.826 D_fake: 1.046 +(epoch: 5, iters: 3968, time: 0.064) G_GAN: 0.006 G_GAN_Feat: 0.925 G_ID: 0.448 G_Rec: 0.388 D_GP: 0.056 D_real: 0.642 D_fake: 0.995 +(epoch: 5, iters: 4368, time: 0.064) G_GAN: 0.306 G_GAN_Feat: 0.798 G_ID: 0.937 G_Rec: 0.651 D_GP: 0.028 D_real: 1.142 D_fake: 0.694 +(epoch: 5, iters: 4768, time: 0.064) G_GAN: 0.128 G_GAN_Feat: 1.041 G_ID: 0.532 G_Rec: 0.514 D_GP: 0.072 D_real: 0.750 D_fake: 0.872 +(epoch: 5, iters: 5168, time: 0.064) G_GAN: 0.512 G_GAN_Feat: 0.925 G_ID: 0.989 G_Rec: 0.415 D_GP: 0.043 D_real: 1.188 D_fake: 0.525 +(epoch: 5, iters: 5568, time: 0.064) G_GAN: 0.149 G_GAN_Feat: 0.895 G_ID: 0.465 G_Rec: 0.568 D_GP: 0.037 D_real: 0.891 D_fake: 0.851 +(epoch: 5, iters: 5968, time: 0.064) G_GAN: 0.019 G_GAN_Feat: 0.611 G_ID: 0.964 G_Rec: 0.388 D_GP: 0.015 D_real: 0.985 D_fake: 0.981 +(epoch: 5, iters: 6368, time: 0.064) G_GAN: 0.155 G_GAN_Feat: 0.893 G_ID: 0.479 G_Rec: 0.403 D_GP: 0.035 D_real: 0.899 D_fake: 0.845 +(epoch: 5, iters: 6768, time: 0.064) G_GAN: 0.345 G_GAN_Feat: 0.910 G_ID: 0.979 G_Rec: 0.648 D_GP: 0.033 D_real: 0.962 D_fake: 0.656 +(epoch: 5, iters: 7168, time: 0.064) G_GAN: 0.373 G_GAN_Feat: 0.862 G_ID: 0.464 G_Rec: 0.401 D_GP: 0.053 D_real: 1.161 D_fake: 0.630 +(epoch: 5, iters: 7568, time: 0.064) G_GAN: 0.095 G_GAN_Feat: 0.737 G_ID: 0.993 G_Rec: 0.439 D_GP: 0.023 D_real: 0.973 D_fake: 0.905 +(epoch: 5, iters: 7968, time: 0.064) G_GAN: -0.164 G_GAN_Feat: 1.144 G_ID: 0.462 G_Rec: 0.829 D_GP: 0.102 D_real: 0.504 D_fake: 1.164 +(epoch: 5, iters: 8368, time: 0.064) G_GAN: 0.319 G_GAN_Feat: 0.947 G_ID: 1.013 G_Rec: 0.484 D_GP: 0.031 D_real: 1.347 D_fake: 0.682 +(epoch: 6, iters: 160, time: 0.064) G_GAN: 0.095 G_GAN_Feat: 0.823 G_ID: 0.357 G_Rec: 0.430 D_GP: 0.024 D_real: 0.856 D_fake: 0.905 +(epoch: 6, iters: 560, time: 0.064) G_GAN: 0.648 G_GAN_Feat: 0.995 G_ID: 1.027 G_Rec: 0.461 D_GP: 0.046 D_real: 1.335 D_fake: 0.393 +(epoch: 6, iters: 960, time: 0.064) G_GAN: -0.214 G_GAN_Feat: 0.687 G_ID: 0.403 G_Rec: 0.383 D_GP: 0.027 D_real: 0.612 D_fake: 1.214 +(epoch: 6, iters: 1360, time: 0.064) G_GAN: 0.780 G_GAN_Feat: 0.840 G_ID: 1.001 G_Rec: 0.714 D_GP: 0.017 D_real: 1.650 D_fake: 0.497 +(epoch: 6, iters: 1760, time: 0.064) G_GAN: -0.160 G_GAN_Feat: 0.776 G_ID: 0.490 G_Rec: 0.380 D_GP: 0.054 D_real: 0.604 D_fake: 1.160 +(epoch: 6, iters: 2160, time: 0.064) G_GAN: 0.186 G_GAN_Feat: 0.840 G_ID: 0.999 G_Rec: 0.646 D_GP: 0.040 D_real: 0.909 D_fake: 0.814 +(epoch: 6, iters: 2560, time: 0.064) G_GAN: 0.353 G_GAN_Feat: 1.001 G_ID: 0.480 G_Rec: 0.542 D_GP: 0.041 D_real: 1.250 D_fake: 0.655 +(epoch: 6, iters: 2960, time: 0.064) G_GAN: 0.355 G_GAN_Feat: 0.836 G_ID: 1.003 G_Rec: 0.374 D_GP: 0.042 D_real: 1.029 D_fake: 0.646 +(epoch: 6, iters: 3360, time: 0.064) G_GAN: 0.270 G_GAN_Feat: 0.821 G_ID: 0.502 G_Rec: 0.326 D_GP: 0.045 D_real: 1.002 D_fake: 0.735 +(epoch: 6, iters: 3760, time: 0.064) G_GAN: 0.174 G_GAN_Feat: 0.902 G_ID: 0.987 G_Rec: 0.426 D_GP: 0.028 D_real: 0.842 D_fake: 0.827 +(epoch: 6, iters: 4160, time: 0.064) G_GAN: 0.174 G_GAN_Feat: 0.954 G_ID: 0.518 G_Rec: 0.426 D_GP: 0.057 D_real: 1.052 D_fake: 0.826 +(epoch: 6, iters: 4560, time: 0.064) G_GAN: 0.314 G_GAN_Feat: 0.848 G_ID: 1.003 G_Rec: 0.456 D_GP: 0.036 D_real: 1.084 D_fake: 0.686 +(epoch: 6, iters: 4960, time: 0.064) G_GAN: 0.345 G_GAN_Feat: 0.948 G_ID: 0.395 G_Rec: 0.406 D_GP: 0.053 D_real: 0.904 D_fake: 0.655 +(epoch: 6, iters: 5360, time: 0.064) G_GAN: 0.086 G_GAN_Feat: 0.872 G_ID: 0.989 G_Rec: 0.381 D_GP: 0.057 D_real: 0.877 D_fake: 0.915 +(epoch: 6, iters: 5760, time: 0.064) G_GAN: 0.513 G_GAN_Feat: 0.791 G_ID: 0.422 G_Rec: 0.381 D_GP: 0.028 D_real: 1.343 D_fake: 0.584 +(epoch: 6, iters: 6160, time: 0.064) G_GAN: 0.234 G_GAN_Feat: 0.814 G_ID: 0.961 G_Rec: 0.363 D_GP: 0.043 D_real: 0.964 D_fake: 0.766 +(epoch: 6, iters: 6560, time: 0.064) G_GAN: 0.067 G_GAN_Feat: 0.902 G_ID: 0.409 G_Rec: 0.356 D_GP: 0.079 D_real: 0.765 D_fake: 0.934 +(epoch: 6, iters: 6960, time: 0.064) G_GAN: 0.162 G_GAN_Feat: 0.794 G_ID: 1.007 G_Rec: 0.437 D_GP: 0.030 D_real: 0.954 D_fake: 0.838 +(epoch: 6, iters: 7360, time: 0.064) G_GAN: 0.234 G_GAN_Feat: 0.825 G_ID: 0.403 G_Rec: 0.483 D_GP: 0.036 D_real: 1.022 D_fake: 0.766 +(epoch: 6, iters: 7760, time: 0.064) G_GAN: 0.102 G_GAN_Feat: 0.910 G_ID: 1.008 G_Rec: 0.529 D_GP: 0.050 D_real: 0.708 D_fake: 0.898 +(epoch: 6, iters: 8160, time: 0.064) G_GAN: -0.009 G_GAN_Feat: 0.811 G_ID: 0.552 G_Rec: 0.405 D_GP: 0.029 D_real: 0.735 D_fake: 1.010 +(epoch: 6, iters: 8560, time: 0.064) G_GAN: 0.160 G_GAN_Feat: 0.815 G_ID: 0.996 G_Rec: 0.488 D_GP: 0.038 D_real: 0.932 D_fake: 0.840 +(epoch: 7, iters: 352, time: 0.064) G_GAN: 0.265 G_GAN_Feat: 0.824 G_ID: 0.382 G_Rec: 0.375 D_GP: 0.043 D_real: 0.980 D_fake: 0.735 +(epoch: 7, iters: 752, time: 0.064) G_GAN: -0.008 G_GAN_Feat: 1.127 G_ID: 0.990 G_Rec: 0.421 D_GP: 0.075 D_real: 0.556 D_fake: 1.008 +(epoch: 7, iters: 1152, time: 0.064) G_GAN: -0.149 G_GAN_Feat: 0.800 G_ID: 0.477 G_Rec: 0.490 D_GP: 0.039 D_real: 0.623 D_fake: 1.149 +(epoch: 7, iters: 1552, time: 0.064) G_GAN: -0.040 G_GAN_Feat: 0.750 G_ID: 0.952 G_Rec: 0.363 D_GP: 0.043 D_real: 0.831 D_fake: 1.040 +(epoch: 7, iters: 1952, time: 0.064) G_GAN: 0.112 G_GAN_Feat: 0.568 G_ID: 0.497 G_Rec: 0.390 D_GP: 0.014 D_real: 1.012 D_fake: 0.888 +(epoch: 7, iters: 2352, time: 0.064) G_GAN: 0.066 G_GAN_Feat: 0.783 G_ID: 0.985 G_Rec: 0.380 D_GP: 0.057 D_real: 0.715 D_fake: 0.934 +(epoch: 7, iters: 2752, time: 0.064) G_GAN: 0.492 G_GAN_Feat: 0.944 G_ID: 0.549 G_Rec: 0.444 D_GP: 0.068 D_real: 1.279 D_fake: 0.513 +(epoch: 7, iters: 3152, time: 0.064) G_GAN: 0.359 G_GAN_Feat: 0.846 G_ID: 0.973 G_Rec: 0.498 D_GP: 0.054 D_real: 1.202 D_fake: 0.642 +(epoch: 7, iters: 3552, time: 0.064) G_GAN: 0.399 G_GAN_Feat: 0.756 G_ID: 0.423 G_Rec: 0.434 D_GP: 0.036 D_real: 1.160 D_fake: 0.602 +(epoch: 7, iters: 3952, time: 0.064) G_GAN: 0.367 G_GAN_Feat: 0.951 G_ID: 0.975 G_Rec: 0.395 D_GP: 0.046 D_real: 0.926 D_fake: 0.639 +(epoch: 7, iters: 4352, time: 0.064) G_GAN: -0.003 G_GAN_Feat: 0.828 G_ID: 0.453 G_Rec: 0.400 D_GP: 0.061 D_real: 0.746 D_fake: 1.003 +(epoch: 7, iters: 4752, time: 0.064) G_GAN: 0.224 G_GAN_Feat: 1.070 G_ID: 0.940 G_Rec: 0.408 D_GP: 0.147 D_real: 0.538 D_fake: 0.778 +(epoch: 7, iters: 5152, time: 0.064) G_GAN: 0.293 G_GAN_Feat: 1.008 G_ID: 0.382 G_Rec: 0.390 D_GP: 0.055 D_real: 0.757 D_fake: 0.707 +(epoch: 7, iters: 5552, time: 0.064) G_GAN: 0.408 G_GAN_Feat: 1.161 G_ID: 0.964 G_Rec: 0.361 D_GP: 0.076 D_real: 1.254 D_fake: 0.609 +(epoch: 7, iters: 5952, time: 0.064) G_GAN: 0.400 G_GAN_Feat: 0.850 G_ID: 0.422 G_Rec: 0.444 D_GP: 0.039 D_real: 1.139 D_fake: 0.600 +(epoch: 7, iters: 6352, time: 0.064) G_GAN: 0.224 G_GAN_Feat: 0.854 G_ID: 0.981 G_Rec: 0.423 D_GP: 0.034 D_real: 0.950 D_fake: 0.776 +(epoch: 7, iters: 6752, time: 0.064) G_GAN: 0.143 G_GAN_Feat: 0.887 G_ID: 0.557 G_Rec: 0.399 D_GP: 0.045 D_real: 0.905 D_fake: 0.858 +(epoch: 7, iters: 7152, time: 0.064) G_GAN: 0.252 G_GAN_Feat: 1.094 G_ID: 0.967 G_Rec: 0.373 D_GP: 0.099 D_real: 0.565 D_fake: 0.749 +(epoch: 7, iters: 7552, time: 0.064) G_GAN: -0.019 G_GAN_Feat: 0.901 G_ID: 0.361 G_Rec: 0.404 D_GP: 0.079 D_real: 0.719 D_fake: 1.019 +(epoch: 7, iters: 7952, time: 0.064) G_GAN: 0.192 G_GAN_Feat: 0.677 G_ID: 0.997 G_Rec: 0.613 D_GP: 0.022 D_real: 1.013 D_fake: 0.808 +(epoch: 7, iters: 8352, time: 0.064) G_GAN: -0.009 G_GAN_Feat: 0.767 G_ID: 0.407 G_Rec: 0.358 D_GP: 0.024 D_real: 0.750 D_fake: 1.009 +(epoch: 8, iters: 144, time: 0.064) G_GAN: -0.207 G_GAN_Feat: 0.785 G_ID: 0.997 G_Rec: 0.381 D_GP: 0.048 D_real: 0.512 D_fake: 1.207 +(epoch: 8, iters: 544, time: 0.064) G_GAN: 0.179 G_GAN_Feat: 0.886 G_ID: 0.424 G_Rec: 0.372 D_GP: 0.048 D_real: 0.835 D_fake: 0.823 +(epoch: 8, iters: 944, time: 0.064) G_GAN: -0.169 G_GAN_Feat: 0.695 G_ID: 0.988 G_Rec: 0.387 D_GP: 0.028 D_real: 0.694 D_fake: 1.169 +(epoch: 8, iters: 1344, time: 0.064) G_GAN: 0.385 G_GAN_Feat: 0.547 G_ID: 0.543 G_Rec: 0.533 D_GP: 0.009 D_real: 1.281 D_fake: 0.615 +(epoch: 8, iters: 1744, time: 0.064) G_GAN: -0.024 G_GAN_Feat: 0.543 G_ID: 0.960 G_Rec: 0.428 D_GP: 0.018 D_real: 0.886 D_fake: 1.024 +(epoch: 8, iters: 2144, time: 0.064) G_GAN: 0.019 G_GAN_Feat: 0.643 G_ID: 0.377 G_Rec: 0.350 D_GP: 0.056 D_real: 0.778 D_fake: 0.981 +(epoch: 8, iters: 2544, time: 0.064) G_GAN: 0.419 G_GAN_Feat: 0.679 G_ID: 0.968 G_Rec: 0.401 D_GP: 0.026 D_real: 1.266 D_fake: 0.582 +(epoch: 8, iters: 2944, time: 0.064) G_GAN: 0.003 G_GAN_Feat: 0.913 G_ID: 0.405 G_Rec: 0.436 D_GP: 0.085 D_real: 0.566 D_fake: 0.997 +(epoch: 8, iters: 3344, time: 0.064) G_GAN: 0.160 G_GAN_Feat: 1.116 G_ID: 0.990 G_Rec: 0.471 D_GP: 0.078 D_real: 0.797 D_fake: 0.840 +(epoch: 8, iters: 3744, time: 0.064) G_GAN: 0.351 G_GAN_Feat: 0.842 G_ID: 0.383 G_Rec: 0.394 D_GP: 0.028 D_real: 1.033 D_fake: 0.649 +(epoch: 8, iters: 4144, time: 0.064) G_GAN: 0.105 G_GAN_Feat: 0.814 G_ID: 0.990 G_Rec: 0.360 D_GP: 0.027 D_real: 0.779 D_fake: 0.895 +(epoch: 8, iters: 4544, time: 0.064) G_GAN: -0.019 G_GAN_Feat: 0.580 G_ID: 0.418 G_Rec: 0.501 D_GP: 0.019 D_real: 0.810 D_fake: 1.019 +(epoch: 8, iters: 4944, time: 0.064) G_GAN: 0.234 G_GAN_Feat: 0.641 G_ID: 0.969 G_Rec: 0.309 D_GP: 0.038 D_real: 0.999 D_fake: 0.766 +(epoch: 8, iters: 5344, time: 0.064) G_GAN: 0.176 G_GAN_Feat: 0.743 G_ID: 0.434 G_Rec: 0.368 D_GP: 0.068 D_real: 1.013 D_fake: 0.824 +(epoch: 8, iters: 5744, time: 0.064) G_GAN: 0.191 G_GAN_Feat: 0.949 G_ID: 0.951 G_Rec: 0.431 D_GP: 0.079 D_real: 0.675 D_fake: 0.811 +(epoch: 8, iters: 6144, time: 0.064) G_GAN: 0.128 G_GAN_Feat: 0.962 G_ID: 0.364 G_Rec: 0.389 D_GP: 0.076 D_real: 0.698 D_fake: 0.872 +(epoch: 8, iters: 6544, time: 0.064) G_GAN: 0.143 G_GAN_Feat: 0.974 G_ID: 0.984 G_Rec: 0.757 D_GP: 0.088 D_real: 0.624 D_fake: 0.857 +(epoch: 8, iters: 6944, time: 0.064) G_GAN: 0.408 G_GAN_Feat: 1.090 G_ID: 0.408 G_Rec: 0.428 D_GP: 0.090 D_real: 1.143 D_fake: 0.606 +(epoch: 8, iters: 7344, time: 0.064) G_GAN: 0.151 G_GAN_Feat: 0.806 G_ID: 0.952 G_Rec: 0.353 D_GP: 0.036 D_real: 0.814 D_fake: 0.849 +(epoch: 8, iters: 7744, time: 0.064) G_GAN: 0.004 G_GAN_Feat: 0.896 G_ID: 0.420 G_Rec: 0.344 D_GP: 0.124 D_real: 0.733 D_fake: 0.996 +(epoch: 8, iters: 8144, time: 0.064) G_GAN: 0.121 G_GAN_Feat: 0.627 G_ID: 0.988 G_Rec: 0.394 D_GP: 0.016 D_real: 1.048 D_fake: 0.879 +(epoch: 8, iters: 8544, time: 0.064) G_GAN: 0.228 G_GAN_Feat: 0.891 G_ID: 0.428 G_Rec: 0.387 D_GP: 0.069 D_real: 0.870 D_fake: 0.774 +(epoch: 9, iters: 336, time: 0.064) G_GAN: 0.058 G_GAN_Feat: 0.899 G_ID: 0.986 G_Rec: 0.346 D_GP: 0.107 D_real: 0.599 D_fake: 0.942 +(epoch: 9, iters: 736, time: 0.064) G_GAN: 0.082 G_GAN_Feat: 0.970 G_ID: 0.424 G_Rec: 0.330 D_GP: 0.095 D_real: 0.495 D_fake: 0.918 +(epoch: 9, iters: 1136, time: 0.064) G_GAN: 0.142 G_GAN_Feat: 0.625 G_ID: 0.986 G_Rec: 0.412 D_GP: 0.014 D_real: 1.022 D_fake: 0.858 +(epoch: 9, iters: 1536, time: 0.064) G_GAN: 0.357 G_GAN_Feat: 0.768 G_ID: 0.380 G_Rec: 0.401 D_GP: 0.048 D_real: 1.134 D_fake: 0.643 +(epoch: 9, iters: 1936, time: 0.064) G_GAN: 0.217 G_GAN_Feat: 0.872 G_ID: 0.963 G_Rec: 0.369 D_GP: 0.089 D_real: 0.799 D_fake: 0.783 +(epoch: 9, iters: 2336, time: 0.064) G_GAN: -0.044 G_GAN_Feat: 0.710 G_ID: 0.437 G_Rec: 0.372 D_GP: 0.036 D_real: 0.717 D_fake: 1.044 +(epoch: 9, iters: 2736, time: 0.064) G_GAN: -0.028 G_GAN_Feat: 1.063 G_ID: 0.975 G_Rec: 0.479 D_GP: 0.114 D_real: 0.497 D_fake: 1.028 +(epoch: 9, iters: 3136, time: 0.064) G_GAN: 0.247 G_GAN_Feat: 0.662 G_ID: 0.463 G_Rec: 0.388 D_GP: 0.028 D_real: 1.080 D_fake: 0.753 +(epoch: 9, iters: 3536, time: 0.064) G_GAN: 0.011 G_GAN_Feat: 0.589 G_ID: 0.967 G_Rec: 0.339 D_GP: 0.018 D_real: 0.929 D_fake: 0.989 +(epoch: 9, iters: 3936, time: 0.064) G_GAN: 0.082 G_GAN_Feat: 0.876 G_ID: 0.369 G_Rec: 0.400 D_GP: 0.114 D_real: 0.674 D_fake: 0.918 +(epoch: 9, iters: 4336, time: 0.064) G_GAN: 0.108 G_GAN_Feat: 0.798 G_ID: 0.935 G_Rec: 0.468 D_GP: 0.050 D_real: 0.829 D_fake: 0.892 +(epoch: 9, iters: 4736, time: 0.064) G_GAN: 0.088 G_GAN_Feat: 1.019 G_ID: 0.445 G_Rec: 0.372 D_GP: 0.068 D_real: 0.558 D_fake: 0.912 +(epoch: 9, iters: 5136, time: 0.064) G_GAN: 0.262 G_GAN_Feat: 1.389 G_ID: 0.990 G_Rec: 0.432 D_GP: 0.351 D_real: 0.388 D_fake: 0.740 +(epoch: 9, iters: 5536, time: 0.064) G_GAN: 0.056 G_GAN_Feat: 0.671 G_ID: 0.442 G_Rec: 0.450 D_GP: 0.030 D_real: 0.795 D_fake: 0.944 +(epoch: 9, iters: 5936, time: 0.064) G_GAN: 0.302 G_GAN_Feat: 0.817 G_ID: 0.935 G_Rec: 0.442 D_GP: 0.077 D_real: 0.890 D_fake: 0.698 +(epoch: 9, iters: 6336, time: 0.064) G_GAN: -0.064 G_GAN_Feat: 0.747 G_ID: 0.453 G_Rec: 0.358 D_GP: 0.057 D_real: 0.649 D_fake: 1.064 +(epoch: 9, iters: 6736, time: 0.064) G_GAN: 0.317 G_GAN_Feat: 0.612 G_ID: 0.959 G_Rec: 0.324 D_GP: 0.019 D_real: 1.150 D_fake: 0.683 +(epoch: 9, iters: 7136, time: 0.064) G_GAN: 0.718 G_GAN_Feat: 0.713 G_ID: 0.485 G_Rec: 0.537 D_GP: 0.019 D_real: 1.583 D_fake: 0.345 +(epoch: 9, iters: 7536, time: 0.064) G_GAN: -0.217 G_GAN_Feat: 0.831 G_ID: 0.973 G_Rec: 0.392 D_GP: 0.050 D_real: 0.735 D_fake: 1.217 +(epoch: 9, iters: 7936, time: 0.064) G_GAN: 0.082 G_GAN_Feat: 1.072 G_ID: 0.352 G_Rec: 0.420 D_GP: 0.061 D_real: 0.520 D_fake: 0.918 +(epoch: 9, iters: 8336, time: 0.064) G_GAN: 0.366 G_GAN_Feat: 0.801 G_ID: 0.971 G_Rec: 0.339 D_GP: 0.052 D_real: 1.052 D_fake: 0.634 +(epoch: 10, iters: 128, time: 0.064) G_GAN: 0.328 G_GAN_Feat: 0.941 G_ID: 0.393 G_Rec: 0.391 D_GP: 0.078 D_real: 0.831 D_fake: 0.673 +(epoch: 10, iters: 528, time: 0.064) G_GAN: 0.652 G_GAN_Feat: 1.110 G_ID: 0.961 G_Rec: 0.417 D_GP: 0.083 D_real: 1.016 D_fake: 0.461 +(epoch: 10, iters: 928, time: 0.064) G_GAN: 0.313 G_GAN_Feat: 0.717 G_ID: 0.400 G_Rec: 0.595 D_GP: 0.030 D_real: 1.115 D_fake: 0.688 +(epoch: 10, iters: 1328, time: 0.064) G_GAN: 0.127 G_GAN_Feat: 0.801 G_ID: 0.946 G_Rec: 0.391 D_GP: 0.069 D_real: 0.901 D_fake: 0.873 +(epoch: 10, iters: 1728, time: 0.064) G_GAN: 0.549 G_GAN_Feat: 0.736 G_ID: 0.434 G_Rec: 0.341 D_GP: 0.023 D_real: 1.345 D_fake: 0.466 +(epoch: 10, iters: 2128, time: 0.064) G_GAN: 0.415 G_GAN_Feat: 0.791 G_ID: 0.983 G_Rec: 0.350 D_GP: 0.045 D_real: 1.103 D_fake: 0.585 +(epoch: 10, iters: 2528, time: 0.064) G_GAN: 0.294 G_GAN_Feat: 0.660 G_ID: 0.447 G_Rec: 0.382 D_GP: 0.018 D_real: 1.143 D_fake: 0.708 +(epoch: 10, iters: 2928, time: 0.064) G_GAN: 0.586 G_GAN_Feat: 0.867 G_ID: 0.984 G_Rec: 0.452 D_GP: 0.058 D_real: 1.290 D_fake: 0.476 +(epoch: 10, iters: 3328, time: 0.064) G_GAN: 0.440 G_GAN_Feat: 0.872 G_ID: 0.392 G_Rec: 0.349 D_GP: 0.053 D_real: 1.103 D_fake: 0.609 +(epoch: 10, iters: 3728, time: 0.064) G_GAN: 0.007 G_GAN_Feat: 0.709 G_ID: 0.947 G_Rec: 0.383 D_GP: 0.029 D_real: 0.783 D_fake: 0.993 +(epoch: 10, iters: 4128, time: 0.064) G_GAN: 0.551 G_GAN_Feat: 0.736 G_ID: 0.465 G_Rec: 0.338 D_GP: 0.031 D_real: 1.343 D_fake: 0.554 +(epoch: 10, iters: 4528, time: 0.064) G_GAN: 0.406 G_GAN_Feat: 0.800 G_ID: 0.992 G_Rec: 0.582 D_GP: 0.051 D_real: 1.077 D_fake: 0.599 +(epoch: 10, iters: 4928, time: 0.064) G_GAN: 0.586 G_GAN_Feat: 0.979 G_ID: 0.418 G_Rec: 0.478 D_GP: 0.041 D_real: 1.163 D_fake: 0.492 +(epoch: 10, iters: 5328, time: 0.064) G_GAN: -0.007 G_GAN_Feat: 1.012 G_ID: 1.000 G_Rec: 0.497 D_GP: 0.099 D_real: 0.990 D_fake: 1.007 +(epoch: 10, iters: 5728, time: 0.064) G_GAN: 0.452 G_GAN_Feat: 0.857 G_ID: 0.398 G_Rec: 0.361 D_GP: 0.029 D_real: 1.090 D_fake: 0.548 +(epoch: 10, iters: 6128, time: 0.064) G_GAN: 0.468 G_GAN_Feat: 0.707 G_ID: 0.932 G_Rec: 0.407 D_GP: 0.015 D_real: 1.310 D_fake: 0.543 +(epoch: 10, iters: 6528, time: 0.064) G_GAN: 0.384 G_GAN_Feat: 0.672 G_ID: 0.488 G_Rec: 0.470 D_GP: 0.031 D_real: 1.157 D_fake: 0.616 +(epoch: 10, iters: 6928, time: 0.064) G_GAN: 0.193 G_GAN_Feat: 0.719 G_ID: 0.967 G_Rec: 0.395 D_GP: 0.039 D_real: 0.996 D_fake: 0.807 +(epoch: 10, iters: 7328, time: 0.064) G_GAN: 0.263 G_GAN_Feat: 1.028 G_ID: 0.369 G_Rec: 0.438 D_GP: 0.113 D_real: 0.618 D_fake: 0.738 +(epoch: 10, iters: 7728, time: 0.064) G_GAN: 0.052 G_GAN_Feat: 0.922 G_ID: 0.971 G_Rec: 0.517 D_GP: 0.140 D_real: 0.553 D_fake: 0.948 +(epoch: 10, iters: 8128, time: 0.064) G_GAN: 0.308 G_GAN_Feat: 0.673 G_ID: 0.437 G_Rec: 0.476 D_GP: 0.018 D_real: 1.176 D_fake: 0.692 +(epoch: 10, iters: 8528, time: 0.064) G_GAN: 0.130 G_GAN_Feat: 0.720 G_ID: 0.994 G_Rec: 0.320 D_GP: 0.071 D_real: 0.860 D_fake: 0.870 +(epoch: 11, iters: 320, time: 0.064) G_GAN: -0.044 G_GAN_Feat: 0.757 G_ID: 0.403 G_Rec: 0.407 D_GP: 0.042 D_real: 0.674 D_fake: 1.044 +(epoch: 11, iters: 720, time: 0.064) G_GAN: 0.401 G_GAN_Feat: 0.901 G_ID: 0.927 G_Rec: 0.421 D_GP: 0.087 D_real: 0.864 D_fake: 0.600 +(epoch: 11, iters: 1120, time: 0.064) G_GAN: 0.441 G_GAN_Feat: 0.961 G_ID: 0.398 G_Rec: 0.409 D_GP: 0.123 D_real: 0.835 D_fake: 0.571 +(epoch: 11, iters: 1520, time: 0.064) G_GAN: 0.411 G_GAN_Feat: 0.761 G_ID: 0.972 G_Rec: 0.473 D_GP: 0.043 D_real: 1.176 D_fake: 0.589 +(epoch: 11, iters: 1920, time: 0.064) G_GAN: 0.413 G_GAN_Feat: 0.707 G_ID: 0.472 G_Rec: 0.334 D_GP: 0.033 D_real: 1.243 D_fake: 0.588 +(epoch: 11, iters: 2320, time: 0.064) G_GAN: 0.397 G_GAN_Feat: 1.027 G_ID: 0.970 G_Rec: 0.354 D_GP: 0.048 D_real: 0.865 D_fake: 0.603 +(epoch: 11, iters: 2720, time: 0.064) G_GAN: 0.272 G_GAN_Feat: 0.907 G_ID: 0.327 G_Rec: 0.364 D_GP: 0.059 D_real: 0.959 D_fake: 0.731 +(epoch: 11, iters: 3120, time: 0.064) G_GAN: 0.334 G_GAN_Feat: 0.763 G_ID: 0.975 G_Rec: 0.399 D_GP: 0.029 D_real: 1.084 D_fake: 0.669 +(epoch: 11, iters: 3520, time: 0.064) G_GAN: 0.237 G_GAN_Feat: 0.696 G_ID: 0.405 G_Rec: 0.418 D_GP: 0.025 D_real: 1.009 D_fake: 0.763 +(epoch: 11, iters: 3920, time: 0.064) G_GAN: -0.078 G_GAN_Feat: 0.918 G_ID: 0.940 G_Rec: 0.385 D_GP: 0.094 D_real: 0.494 D_fake: 1.078 +(epoch: 11, iters: 4320, time: 0.064) G_GAN: 0.263 G_GAN_Feat: 0.698 G_ID: 0.422 G_Rec: 0.413 D_GP: 0.026 D_real: 1.147 D_fake: 0.737 +(epoch: 11, iters: 4720, time: 0.064) G_GAN: 0.508 G_GAN_Feat: 0.892 G_ID: 0.946 G_Rec: 0.418 D_GP: 0.031 D_real: 1.174 D_fake: 0.498 +(epoch: 11, iters: 5120, time: 0.064) G_GAN: 0.446 G_GAN_Feat: 0.983 G_ID: 0.399 G_Rec: 0.364 D_GP: 0.104 D_real: 0.880 D_fake: 0.565 +(epoch: 11, iters: 5520, time: 0.064) G_GAN: 0.214 G_GAN_Feat: 0.489 G_ID: 0.971 G_Rec: 0.320 D_GP: 0.016 D_real: 1.151 D_fake: 0.786 +(epoch: 11, iters: 5920, time: 0.064) G_GAN: -0.065 G_GAN_Feat: 0.613 G_ID: 0.418 G_Rec: 0.520 D_GP: 0.022 D_real: 0.700 D_fake: 1.065 +(epoch: 11, iters: 6320, time: 0.064) G_GAN: -0.042 G_GAN_Feat: 0.618 G_ID: 0.939 G_Rec: 0.347 D_GP: 0.038 D_real: 0.823 D_fake: 1.042 +(epoch: 11, iters: 6720, time: 0.064) G_GAN: 0.234 G_GAN_Feat: 0.647 G_ID: 0.353 G_Rec: 0.312 D_GP: 0.033 D_real: 1.007 D_fake: 0.766 +(epoch: 11, iters: 7120, time: 0.064) G_GAN: 0.555 G_GAN_Feat: 0.910 G_ID: 0.973 G_Rec: 0.477 D_GP: 0.042 D_real: 1.348 D_fake: 0.587 +(epoch: 11, iters: 7520, time: 0.064) G_GAN: 0.026 G_GAN_Feat: 0.664 G_ID: 0.399 G_Rec: 0.484 D_GP: 0.034 D_real: 0.873 D_fake: 0.974 +(epoch: 11, iters: 7920, time: 0.064) G_GAN: 0.280 G_GAN_Feat: 1.114 G_ID: 0.997 G_Rec: 0.433 D_GP: 0.140 D_real: 0.488 D_fake: 0.727 +(epoch: 11, iters: 8320, time: 0.064) G_GAN: 0.034 G_GAN_Feat: 0.838 G_ID: 0.351 G_Rec: 0.308 D_GP: 0.057 D_real: 0.595 D_fake: 0.966 +(epoch: 12, iters: 112, time: 0.064) G_GAN: 0.244 G_GAN_Feat: 0.763 G_ID: 0.950 G_Rec: 0.623 D_GP: 0.036 D_real: 1.038 D_fake: 0.756 +(epoch: 12, iters: 512, time: 0.064) G_GAN: 0.278 G_GAN_Feat: 0.668 G_ID: 0.428 G_Rec: 0.398 D_GP: 0.023 D_real: 1.084 D_fake: 0.722 +(epoch: 12, iters: 912, time: 0.064) G_GAN: 0.499 G_GAN_Feat: 0.929 G_ID: 0.956 G_Rec: 0.461 D_GP: 0.078 D_real: 0.982 D_fake: 0.527 +(epoch: 12, iters: 1312, time: 0.064) G_GAN: 0.032 G_GAN_Feat: 0.977 G_ID: 0.431 G_Rec: 0.608 D_GP: 0.058 D_real: 0.514 D_fake: 0.968 +(epoch: 12, iters: 1712, time: 0.064) G_GAN: 0.312 G_GAN_Feat: 0.752 G_ID: 0.962 G_Rec: 0.310 D_GP: 0.060 D_real: 1.078 D_fake: 0.689 +(epoch: 12, iters: 2112, time: 0.064) G_GAN: -0.177 G_GAN_Feat: 0.622 G_ID: 0.445 G_Rec: 0.379 D_GP: 0.022 D_real: 0.647 D_fake: 1.177 +(epoch: 12, iters: 2512, time: 0.064) G_GAN: 0.066 G_GAN_Feat: 1.198 G_ID: 0.928 G_Rec: 0.403 D_GP: 0.115 D_real: 0.388 D_fake: 0.934 +(epoch: 12, iters: 2912, time: 0.064) G_GAN: 0.169 G_GAN_Feat: 0.576 G_ID: 0.482 G_Rec: 0.400 D_GP: 0.015 D_real: 1.049 D_fake: 0.834 +(epoch: 12, iters: 3312, time: 0.064) G_GAN: 0.078 G_GAN_Feat: 0.845 G_ID: 1.003 G_Rec: 0.604 D_GP: 0.082 D_real: 0.745 D_fake: 0.922 +(epoch: 12, iters: 3712, time: 0.064) G_GAN: 0.196 G_GAN_Feat: 1.101 G_ID: 0.395 G_Rec: 0.653 D_GP: 0.152 D_real: 0.473 D_fake: 0.807 +(epoch: 12, iters: 4112, time: 0.064) G_GAN: 0.520 G_GAN_Feat: 0.851 G_ID: 0.952 G_Rec: 0.467 D_GP: 0.039 D_real: 1.284 D_fake: 0.484 +(epoch: 12, iters: 4512, time: 0.064) G_GAN: 0.185 G_GAN_Feat: 0.796 G_ID: 0.415 G_Rec: 0.349 D_GP: 0.047 D_real: 0.826 D_fake: 0.815 +(epoch: 12, iters: 4912, time: 0.064) G_GAN: 0.267 G_GAN_Feat: 0.435 G_ID: 0.953 G_Rec: 0.391 D_GP: 0.011 D_real: 1.190 D_fake: 0.733 +(epoch: 12, iters: 5312, time: 0.064) G_GAN: 0.200 G_GAN_Feat: 0.456 G_ID: 0.359 G_Rec: 0.340 D_GP: 0.012 D_real: 1.059 D_fake: 0.800 +(epoch: 12, iters: 5712, time: 0.064) G_GAN: 0.189 G_GAN_Feat: 0.491 G_ID: 0.946 G_Rec: 0.366 D_GP: 0.015 D_real: 1.098 D_fake: 0.811 +(epoch: 12, iters: 6112, time: 0.064) G_GAN: 0.036 G_GAN_Feat: 0.615 G_ID: 0.389 G_Rec: 0.421 D_GP: 0.019 D_real: 1.009 D_fake: 0.968 +(epoch: 12, iters: 6512, time: 0.064) G_GAN: -0.270 G_GAN_Feat: 0.532 G_ID: 0.948 G_Rec: 0.310 D_GP: 0.030 D_real: 0.566 D_fake: 1.270 +(epoch: 12, iters: 6912, time: 0.064) G_GAN: 0.098 G_GAN_Feat: 0.603 G_ID: 0.375 G_Rec: 0.318 D_GP: 0.052 D_real: 0.875 D_fake: 0.902 +(epoch: 12, iters: 7312, time: 0.064) G_GAN: 0.311 G_GAN_Feat: 0.696 G_ID: 0.906 G_Rec: 0.388 D_GP: 0.064 D_real: 1.057 D_fake: 0.690 +(epoch: 12, iters: 7712, time: 0.064) G_GAN: 0.270 G_GAN_Feat: 0.674 G_ID: 0.363 G_Rec: 0.345 D_GP: 0.065 D_real: 1.027 D_fake: 0.731 +(epoch: 12, iters: 8112, time: 0.064) G_GAN: 0.394 G_GAN_Feat: 0.585 G_ID: 0.960 G_Rec: 0.344 D_GP: 0.021 D_real: 1.261 D_fake: 0.606 +(epoch: 12, iters: 8512, time: 0.064) G_GAN: 0.240 G_GAN_Feat: 0.770 G_ID: 0.412 G_Rec: 0.500 D_GP: 0.066 D_real: 0.880 D_fake: 0.760 +(epoch: 13, iters: 304, time: 0.064) G_GAN: -0.080 G_GAN_Feat: 0.658 G_ID: 0.917 G_Rec: 0.354 D_GP: 0.049 D_real: 0.674 D_fake: 1.080 +(epoch: 13, iters: 704, time: 0.064) G_GAN: 0.379 G_GAN_Feat: 0.798 G_ID: 0.411 G_Rec: 0.471 D_GP: 0.069 D_real: 1.007 D_fake: 0.626 +(epoch: 13, iters: 1104, time: 0.064) G_GAN: 0.390 G_GAN_Feat: 0.740 G_ID: 0.947 G_Rec: 0.411 D_GP: 0.043 D_real: 1.147 D_fake: 0.612 +(epoch: 13, iters: 1504, time: 0.064) G_GAN: 0.582 G_GAN_Feat: 1.012 G_ID: 0.453 G_Rec: 0.466 D_GP: 0.127 D_real: 0.883 D_fake: 0.436 +(epoch: 13, iters: 1904, time: 0.064) G_GAN: -0.050 G_GAN_Feat: 0.735 G_ID: 0.946 G_Rec: 0.349 D_GP: 0.061 D_real: 0.624 D_fake: 1.050 +(epoch: 13, iters: 2304, time: 0.064) G_GAN: 0.566 G_GAN_Feat: 0.902 G_ID: 0.459 G_Rec: 0.346 D_GP: 0.143 D_real: 1.003 D_fake: 0.510 +(epoch: 13, iters: 2704, time: 0.064) G_GAN: 0.289 G_GAN_Feat: 0.679 G_ID: 0.972 G_Rec: 0.384 D_GP: 0.032 D_real: 1.033 D_fake: 0.711 +(epoch: 13, iters: 3104, time: 0.064) G_GAN: 0.175 G_GAN_Feat: 0.842 G_ID: 0.386 G_Rec: 0.571 D_GP: 0.085 D_real: 0.778 D_fake: 0.825 +(epoch: 13, iters: 3504, time: 0.064) G_GAN: 0.359 G_GAN_Feat: 0.767 G_ID: 0.920 G_Rec: 0.365 D_GP: 0.063 D_real: 0.998 D_fake: 0.641 +(epoch: 13, iters: 3904, time: 0.064) G_GAN: 0.793 G_GAN_Feat: 0.705 G_ID: 0.459 G_Rec: 0.349 D_GP: 0.025 D_real: 1.559 D_fake: 0.277 +(epoch: 13, iters: 4304, time: 0.064) G_GAN: 0.453 G_GAN_Feat: 0.734 G_ID: 0.954 G_Rec: 0.367 D_GP: 0.048 D_real: 1.102 D_fake: 0.557 +(epoch: 13, iters: 4704, time: 0.064) G_GAN: 0.088 G_GAN_Feat: 0.709 G_ID: 0.393 G_Rec: 0.370 D_GP: 0.028 D_real: 0.854 D_fake: 0.912 +(epoch: 13, iters: 5104, time: 0.064) G_GAN: 0.552 G_GAN_Feat: 0.694 G_ID: 0.933 G_Rec: 0.357 D_GP: 0.023 D_real: 1.362 D_fake: 0.499 +(epoch: 13, iters: 5504, time: 0.064) G_GAN: 0.157 G_GAN_Feat: 0.594 G_ID: 0.384 G_Rec: 0.450 D_GP: 0.024 D_real: 0.920 D_fake: 0.843 +(epoch: 13, iters: 5904, time: 0.064) G_GAN: 0.249 G_GAN_Feat: 0.664 G_ID: 0.949 G_Rec: 0.341 D_GP: 0.050 D_real: 0.960 D_fake: 0.751 +(epoch: 13, iters: 6304, time: 0.064) G_GAN: 0.246 G_GAN_Feat: 0.725 G_ID: 0.416 G_Rec: 0.339 D_GP: 0.079 D_real: 0.984 D_fake: 0.757 +(epoch: 13, iters: 6704, time: 0.064) G_GAN: 0.041 G_GAN_Feat: 0.644 G_ID: 0.938 G_Rec: 0.403 D_GP: 0.045 D_real: 0.741 D_fake: 0.959 +(epoch: 13, iters: 7104, time: 0.064) G_GAN: 0.415 G_GAN_Feat: 0.824 G_ID: 0.363 G_Rec: 0.347 D_GP: 0.096 D_real: 0.893 D_fake: 0.587 +(epoch: 13, iters: 7504, time: 0.064) G_GAN: 0.060 G_GAN_Feat: 0.850 G_ID: 0.921 G_Rec: 0.431 D_GP: 0.026 D_real: 0.712 D_fake: 0.940 +(epoch: 13, iters: 7904, time: 0.064) G_GAN: 0.239 G_GAN_Feat: 0.787 G_ID: 0.478 G_Rec: 0.415 D_GP: 0.073 D_real: 1.030 D_fake: 0.761 +(epoch: 13, iters: 8304, time: 0.064) G_GAN: 0.004 G_GAN_Feat: 0.784 G_ID: 0.950 G_Rec: 0.438 D_GP: 0.064 D_real: 0.609 D_fake: 0.996 +(epoch: 14, iters: 96, time: 0.064) G_GAN: 0.523 G_GAN_Feat: 0.987 G_ID: 0.483 G_Rec: 0.440 D_GP: 0.047 D_real: 1.045 D_fake: 0.485 +(epoch: 14, iters: 496, time: 0.064) G_GAN: 0.723 G_GAN_Feat: 0.740 G_ID: 0.919 G_Rec: 0.377 D_GP: 0.040 D_real: 1.437 D_fake: 0.411 +(epoch: 14, iters: 896, time: 0.064) G_GAN: 0.220 G_GAN_Feat: 0.934 G_ID: 0.408 G_Rec: 0.391 D_GP: 0.044 D_real: 0.695 D_fake: 0.780 +(epoch: 14, iters: 1296, time: 0.064) G_GAN: 0.168 G_GAN_Feat: 0.866 G_ID: 0.941 G_Rec: 0.367 D_GP: 0.035 D_real: 1.034 D_fake: 0.832 +(epoch: 14, iters: 1696, time: 0.064) G_GAN: 0.102 G_GAN_Feat: 0.824 G_ID: 0.402 G_Rec: 0.526 D_GP: 0.041 D_real: 0.929 D_fake: 0.898 +(epoch: 14, iters: 2096, time: 0.064) G_GAN: -0.040 G_GAN_Feat: 0.887 G_ID: 0.923 G_Rec: 0.574 D_GP: 0.067 D_real: 0.415 D_fake: 1.040 +(epoch: 14, iters: 2496, time: 0.064) G_GAN: 0.454 G_GAN_Feat: 0.688 G_ID: 0.455 G_Rec: 0.353 D_GP: 0.038 D_real: 1.145 D_fake: 0.553 +(epoch: 14, iters: 2896, time: 0.064) G_GAN: -0.044 G_GAN_Feat: 0.753 G_ID: 0.932 G_Rec: 0.371 D_GP: 0.038 D_real: 0.729 D_fake: 1.044 +(epoch: 14, iters: 3296, time: 0.064) G_GAN: 0.219 G_GAN_Feat: 0.665 G_ID: 0.454 G_Rec: 0.384 D_GP: 0.025 D_real: 0.971 D_fake: 0.781 +(epoch: 14, iters: 3696, time: 0.064) G_GAN: -0.050 G_GAN_Feat: 0.984 G_ID: 0.944 G_Rec: 0.341 D_GP: 0.130 D_real: 0.384 D_fake: 1.050 +(epoch: 14, iters: 4096, time: 0.064) G_GAN: 0.294 G_GAN_Feat: 0.821 G_ID: 0.302 G_Rec: 0.367 D_GP: 0.038 D_real: 1.029 D_fake: 0.706 +(epoch: 14, iters: 4496, time: 0.064) G_GAN: 0.228 G_GAN_Feat: 0.533 G_ID: 0.931 G_Rec: 0.337 D_GP: 0.015 D_real: 1.137 D_fake: 0.772 +(epoch: 14, iters: 4896, time: 0.064) G_GAN: 0.074 G_GAN_Feat: 0.739 G_ID: 0.430 G_Rec: 0.615 D_GP: 0.041 D_real: 0.924 D_fake: 0.926 +(epoch: 14, iters: 5296, time: 0.064) G_GAN: 0.581 G_GAN_Feat: 0.805 G_ID: 0.908 G_Rec: 0.390 D_GP: 0.055 D_real: 1.223 D_fake: 0.422 +(epoch: 14, iters: 5696, time: 0.064) G_GAN: 0.206 G_GAN_Feat: 0.681 G_ID: 0.456 G_Rec: 0.385 D_GP: 0.042 D_real: 0.937 D_fake: 0.795 +(epoch: 14, iters: 6096, time: 0.064) G_GAN: -0.085 G_GAN_Feat: 0.681 G_ID: 0.943 G_Rec: 0.359 D_GP: 0.039 D_real: 0.773 D_fake: 1.085 +(epoch: 14, iters: 6496, time: 0.064) G_GAN: 0.057 G_GAN_Feat: 0.766 G_ID: 0.521 G_Rec: 0.381 D_GP: 0.074 D_real: 0.746 D_fake: 0.943 +(epoch: 14, iters: 6896, time: 0.064) G_GAN: 0.028 G_GAN_Feat: 0.819 G_ID: 0.953 G_Rec: 0.352 D_GP: 0.053 D_real: 0.640 D_fake: 0.972 +(epoch: 14, iters: 7296, time: 0.064) G_GAN: 0.014 G_GAN_Feat: 1.235 G_ID: 0.351 G_Rec: 0.343 D_GP: 0.088 D_real: 0.499 D_fake: 0.986 +(epoch: 14, iters: 7696, time: 0.064) G_GAN: -0.037 G_GAN_Feat: 0.774 G_ID: 0.926 G_Rec: 0.504 D_GP: 0.049 D_real: 0.661 D_fake: 1.037 +(epoch: 14, iters: 8096, time: 0.064) G_GAN: 0.178 G_GAN_Feat: 0.720 G_ID: 0.438 G_Rec: 0.382 D_GP: 0.024 D_real: 0.844 D_fake: 0.822 +(epoch: 14, iters: 8496, time: 0.064) G_GAN: 0.533 G_GAN_Feat: 0.687 G_ID: 0.893 G_Rec: 0.352 D_GP: 0.027 D_real: 1.317 D_fake: 0.489 +(epoch: 15, iters: 288, time: 0.064) G_GAN: 0.306 G_GAN_Feat: 0.894 G_ID: 0.399 G_Rec: 0.495 D_GP: 0.104 D_real: 0.831 D_fake: 0.695 +(epoch: 15, iters: 688, time: 0.064) G_GAN: 0.036 G_GAN_Feat: 0.802 G_ID: 0.955 G_Rec: 0.341 D_GP: 0.040 D_real: 0.642 D_fake: 0.964 +(epoch: 15, iters: 1088, time: 0.064) G_GAN: 0.170 G_GAN_Feat: 0.952 G_ID: 0.429 G_Rec: 0.414 D_GP: 0.055 D_real: 0.640 D_fake: 0.830 +(epoch: 15, iters: 1488, time: 0.064) G_GAN: 0.090 G_GAN_Feat: 0.986 G_ID: 0.951 G_Rec: 0.325 D_GP: 0.103 D_real: 0.576 D_fake: 0.911 +(epoch: 15, iters: 1888, time: 0.064) G_GAN: 0.184 G_GAN_Feat: 0.703 G_ID: 0.413 G_Rec: 0.387 D_GP: 0.037 D_real: 0.862 D_fake: 0.816 +(epoch: 15, iters: 2288, time: 0.064) G_GAN: 0.092 G_GAN_Feat: 0.752 G_ID: 0.919 G_Rec: 0.403 D_GP: 0.041 D_real: 0.823 D_fake: 0.908 +(epoch: 15, iters: 2688, time: 0.064) G_GAN: 0.134 G_GAN_Feat: 0.580 G_ID: 0.486 G_Rec: 0.344 D_GP: 0.018 D_real: 1.006 D_fake: 0.866 +(epoch: 15, iters: 3088, time: 0.064) G_GAN: 0.353 G_GAN_Feat: 0.708 G_ID: 0.935 G_Rec: 0.370 D_GP: 0.057 D_real: 1.000 D_fake: 0.648 +(epoch: 15, iters: 3488, time: 0.064) G_GAN: 0.202 G_GAN_Feat: 0.739 G_ID: 0.400 G_Rec: 0.606 D_GP: 0.060 D_real: 0.911 D_fake: 0.818 +(epoch: 15, iters: 3888, time: 0.064) G_GAN: 0.174 G_GAN_Feat: 1.123 G_ID: 0.932 G_Rec: 0.356 D_GP: 0.154 D_real: 0.459 D_fake: 0.826 +(epoch: 15, iters: 4288, time: 0.064) G_GAN: 0.355 G_GAN_Feat: 0.927 G_ID: 0.419 G_Rec: 0.420 D_GP: 0.045 D_real: 0.804 D_fake: 0.645 +(epoch: 15, iters: 4688, time: 0.064) G_GAN: 0.451 G_GAN_Feat: 1.369 G_ID: 0.959 G_Rec: 0.369 D_GP: 0.298 D_real: 0.493 D_fake: 0.565 +(epoch: 15, iters: 5088, time: 0.064) G_GAN: 0.016 G_GAN_Feat: 0.663 G_ID: 0.421 G_Rec: 0.459 D_GP: 0.034 D_real: 0.744 D_fake: 0.984 +(epoch: 15, iters: 5488, time: 0.064) G_GAN: 0.003 G_GAN_Feat: 0.826 G_ID: 0.912 G_Rec: 0.361 D_GP: 0.094 D_real: 0.455 D_fake: 0.997 +(epoch: 15, iters: 5888, time: 0.064) G_GAN: 0.526 G_GAN_Feat: 0.940 G_ID: 0.423 G_Rec: 0.705 D_GP: 0.032 D_real: 1.128 D_fake: 0.523 +(epoch: 15, iters: 6288, time: 0.064) G_GAN: -0.014 G_GAN_Feat: 1.008 G_ID: 0.931 G_Rec: 0.368 D_GP: 0.170 D_real: 0.409 D_fake: 1.014 +(epoch: 15, iters: 6688, time: 0.064) G_GAN: 0.328 G_GAN_Feat: 0.762 G_ID: 0.365 G_Rec: 0.344 D_GP: 0.031 D_real: 1.202 D_fake: 0.672 +(epoch: 15, iters: 7088, time: 0.064) G_GAN: 0.077 G_GAN_Feat: 0.672 G_ID: 0.944 G_Rec: 0.373 D_GP: 0.017 D_real: 0.902 D_fake: 0.923 +(epoch: 15, iters: 7488, time: 0.064) G_GAN: 0.122 G_GAN_Feat: 0.755 G_ID: 0.332 G_Rec: 0.412 D_GP: 0.064 D_real: 0.798 D_fake: 0.878 +(epoch: 15, iters: 7888, time: 0.064) G_GAN: 0.250 G_GAN_Feat: 0.930 G_ID: 0.930 G_Rec: 0.441 D_GP: 0.128 D_real: 0.689 D_fake: 0.754 +(epoch: 15, iters: 8288, time: 0.064) G_GAN: 0.068 G_GAN_Feat: 0.594 G_ID: 0.412 G_Rec: 0.315 D_GP: 0.016 D_real: 0.948 D_fake: 0.932 +(epoch: 16, iters: 80, time: 0.064) G_GAN: 0.021 G_GAN_Feat: 0.790 G_ID: 0.924 G_Rec: 0.410 D_GP: 0.066 D_real: 0.744 D_fake: 0.979 +(epoch: 16, iters: 480, time: 0.064) G_GAN: 0.334 G_GAN_Feat: 1.100 G_ID: 0.361 G_Rec: 0.364 D_GP: 0.169 D_real: 0.652 D_fake: 0.669 +(epoch: 16, iters: 880, time: 0.064) G_GAN: 0.023 G_GAN_Feat: 0.729 G_ID: 0.922 G_Rec: 0.549 D_GP: 0.038 D_real: 0.772 D_fake: 0.977 +(epoch: 16, iters: 1280, time: 0.064) G_GAN: 0.310 G_GAN_Feat: 0.755 G_ID: 0.392 G_Rec: 0.320 D_GP: 0.061 D_real: 1.005 D_fake: 0.698 +(epoch: 16, iters: 1680, time: 0.064) G_GAN: -0.661 G_GAN_Feat: 1.123 G_ID: 0.931 G_Rec: 0.370 D_GP: 0.111 D_real: 0.061 D_fake: 1.661 +(epoch: 16, iters: 2080, time: 0.064) G_GAN: 0.237 G_GAN_Feat: 1.046 G_ID: 0.379 G_Rec: 0.377 D_GP: 0.334 D_real: 0.503 D_fake: 0.768 +(epoch: 16, iters: 2480, time: 0.064) G_GAN: 0.001 G_GAN_Feat: 0.892 G_ID: 0.903 G_Rec: 0.457 D_GP: 0.113 D_real: 0.884 D_fake: 1.000 +(epoch: 16, iters: 2880, time: 0.064) G_GAN: 0.521 G_GAN_Feat: 0.875 G_ID: 0.375 G_Rec: 0.420 D_GP: 0.045 D_real: 1.224 D_fake: 0.536 +(epoch: 16, iters: 3280, time: 0.064) G_GAN: 0.404 G_GAN_Feat: 0.820 G_ID: 0.907 G_Rec: 0.607 D_GP: 0.055 D_real: 1.017 D_fake: 0.596 +(epoch: 16, iters: 3680, time: 0.064) G_GAN: -0.031 G_GAN_Feat: 0.915 G_ID: 0.353 G_Rec: 0.479 D_GP: 0.075 D_real: 0.821 D_fake: 1.031 +(epoch: 16, iters: 4080, time: 0.064) G_GAN: 0.256 G_GAN_Feat: 0.984 G_ID: 0.915 G_Rec: 0.369 D_GP: 0.028 D_real: 0.850 D_fake: 0.744 +(epoch: 16, iters: 4480, time: 0.064) G_GAN: 0.340 G_GAN_Feat: 0.957 G_ID: 0.377 G_Rec: 0.448 D_GP: 0.070 D_real: 0.789 D_fake: 0.665 +(epoch: 16, iters: 4880, time: 0.064) G_GAN: 0.200 G_GAN_Feat: 0.598 G_ID: 0.922 G_Rec: 0.455 D_GP: 0.015 D_real: 1.075 D_fake: 0.800 +(epoch: 16, iters: 5280, time: 0.064) G_GAN: 0.478 G_GAN_Feat: 0.837 G_ID: 0.404 G_Rec: 0.397 D_GP: 0.027 D_real: 1.247 D_fake: 0.544 +(epoch: 16, iters: 5680, time: 0.064) G_GAN: 0.188 G_GAN_Feat: 0.788 G_ID: 0.924 G_Rec: 0.358 D_GP: 0.057 D_real: 0.869 D_fake: 0.812 +(epoch: 16, iters: 6080, time: 0.064) G_GAN: 0.322 G_GAN_Feat: 0.840 G_ID: 0.338 G_Rec: 0.373 D_GP: 0.052 D_real: 0.934 D_fake: 0.679 +(epoch: 16, iters: 6480, time: 0.064) G_GAN: 0.205 G_GAN_Feat: 0.694 G_ID: 0.881 G_Rec: 0.457 D_GP: 0.061 D_real: 0.966 D_fake: 0.795 +(epoch: 16, iters: 6880, time: 0.064) G_GAN: 0.019 G_GAN_Feat: 0.823 G_ID: 0.408 G_Rec: 0.355 D_GP: 0.078 D_real: 0.782 D_fake: 0.981 +(epoch: 16, iters: 7280, time: 0.064) G_GAN: -0.045 G_GAN_Feat: 1.056 G_ID: 0.942 G_Rec: 0.450 D_GP: 0.176 D_real: 0.262 D_fake: 1.045 +(epoch: 16, iters: 7680, time: 0.064) G_GAN: 0.092 G_GAN_Feat: 1.206 G_ID: 0.426 G_Rec: 0.366 D_GP: 0.274 D_real: 0.496 D_fake: 0.909 +(epoch: 16, iters: 8080, time: 0.064) G_GAN: 0.378 G_GAN_Feat: 0.804 G_ID: 0.960 G_Rec: 0.446 D_GP: 0.057 D_real: 1.057 D_fake: 0.624 +(epoch: 16, iters: 8480, time: 0.064) G_GAN: 0.508 G_GAN_Feat: 0.951 G_ID: 0.418 G_Rec: 0.341 D_GP: 0.069 D_real: 1.102 D_fake: 0.493 +(epoch: 17, iters: 272, time: 0.064) G_GAN: 0.325 G_GAN_Feat: 0.850 G_ID: 0.894 G_Rec: 0.388 D_GP: 0.051 D_real: 0.903 D_fake: 0.676 +(epoch: 17, iters: 672, time: 0.064) G_GAN: 0.448 G_GAN_Feat: 0.897 G_ID: 0.372 G_Rec: 0.426 D_GP: 0.059 D_real: 0.971 D_fake: 0.552 +(epoch: 17, iters: 1072, time: 0.064) G_GAN: 0.624 G_GAN_Feat: 0.794 G_ID: 0.946 G_Rec: 0.402 D_GP: 0.048 D_real: 1.279 D_fake: 0.380 +(epoch: 17, iters: 1472, time: 0.064) G_GAN: 0.530 G_GAN_Feat: 0.656 G_ID: 0.447 G_Rec: 0.423 D_GP: 0.031 D_real: 1.319 D_fake: 0.471 +(epoch: 17, iters: 1872, time: 0.064) G_GAN: 0.118 G_GAN_Feat: 0.845 G_ID: 0.934 G_Rec: 0.436 D_GP: 0.041 D_real: 0.710 D_fake: 0.882 +(epoch: 17, iters: 2272, time: 0.064) G_GAN: -0.163 G_GAN_Feat: 0.713 G_ID: 0.461 G_Rec: 0.348 D_GP: 0.024 D_real: 0.706 D_fake: 1.164 +(epoch: 17, iters: 2672, time: 0.064) G_GAN: 0.471 G_GAN_Feat: 0.838 G_ID: 0.896 G_Rec: 0.330 D_GP: 0.030 D_real: 1.107 D_fake: 0.530 +(epoch: 17, iters: 3072, time: 0.064) G_GAN: 0.476 G_GAN_Feat: 0.832 G_ID: 0.404 G_Rec: 0.342 D_GP: 0.053 D_real: 1.197 D_fake: 0.545 +(epoch: 17, iters: 3472, time: 0.064) G_GAN: 0.356 G_GAN_Feat: 0.851 G_ID: 0.902 G_Rec: 0.425 D_GP: 0.043 D_real: 1.127 D_fake: 0.655 +(epoch: 17, iters: 3872, time: 0.064) G_GAN: 0.316 G_GAN_Feat: 0.923 G_ID: 0.436 G_Rec: 0.382 D_GP: 0.026 D_real: 0.931 D_fake: 0.684 +(epoch: 17, iters: 4272, time: 0.064) G_GAN: 0.295 G_GAN_Feat: 0.857 G_ID: 0.920 G_Rec: 0.520 D_GP: 0.041 D_real: 0.956 D_fake: 0.705 +(epoch: 17, iters: 4672, time: 0.064) G_GAN: 0.319 G_GAN_Feat: 0.497 G_ID: 0.400 G_Rec: 0.337 D_GP: 0.021 D_real: 1.138 D_fake: 0.681 +(epoch: 17, iters: 5072, time: 0.064) G_GAN: 0.239 G_GAN_Feat: 0.704 G_ID: 0.909 G_Rec: 0.431 D_GP: 0.047 D_real: 0.917 D_fake: 0.761 +(epoch: 17, iters: 5472, time: 0.064) G_GAN: 0.012 G_GAN_Feat: 0.874 G_ID: 0.426 G_Rec: 0.348 D_GP: 0.120 D_real: 0.553 D_fake: 0.989 +(epoch: 17, iters: 5872, time: 0.064) G_GAN: 0.337 G_GAN_Feat: 0.883 G_ID: 0.886 G_Rec: 0.363 D_GP: 0.312 D_real: 0.708 D_fake: 0.665 +(epoch: 17, iters: 6272, time: 0.064) G_GAN: 0.076 G_GAN_Feat: 0.851 G_ID: 0.341 G_Rec: 0.374 D_GP: 0.054 D_real: 0.519 D_fake: 0.924 +(epoch: 17, iters: 6672, time: 0.064) G_GAN: 0.325 G_GAN_Feat: 1.015 G_ID: 0.854 G_Rec: 0.343 D_GP: 0.179 D_real: 0.682 D_fake: 0.677 +(epoch: 17, iters: 7072, time: 0.064) G_GAN: 0.354 G_GAN_Feat: 0.689 G_ID: 0.486 G_Rec: 0.440 D_GP: 0.027 D_real: 1.171 D_fake: 0.646 +(epoch: 17, iters: 7472, time: 0.064) G_GAN: 0.302 G_GAN_Feat: 0.867 G_ID: 0.894 G_Rec: 0.353 D_GP: 0.079 D_real: 0.957 D_fake: 0.698 +(epoch: 17, iters: 7872, time: 0.064) G_GAN: 0.186 G_GAN_Feat: 0.621 G_ID: 0.461 G_Rec: 0.378 D_GP: 0.018 D_real: 0.991 D_fake: 0.814 +(epoch: 17, iters: 8272, time: 0.064) G_GAN: 0.256 G_GAN_Feat: 0.831 G_ID: 0.915 G_Rec: 0.437 D_GP: 0.052 D_real: 0.827 D_fake: 0.744 +(epoch: 18, iters: 64, time: 0.064) G_GAN: -0.060 G_GAN_Feat: 0.583 G_ID: 0.418 G_Rec: 0.453 D_GP: 0.020 D_real: 0.810 D_fake: 1.060 +(epoch: 18, iters: 464, time: 0.064) G_GAN: 0.406 G_GAN_Feat: 0.730 G_ID: 0.892 G_Rec: 0.340 D_GP: 0.046 D_real: 1.128 D_fake: 0.594 +(epoch: 18, iters: 864, time: 0.064) G_GAN: 0.672 G_GAN_Feat: 1.016 G_ID: 0.456 G_Rec: 0.387 D_GP: 0.254 D_real: 1.252 D_fake: 0.362 +(epoch: 18, iters: 1264, time: 0.064) G_GAN: 0.286 G_GAN_Feat: 0.610 G_ID: 0.903 G_Rec: 0.353 D_GP: 0.024 D_real: 1.103 D_fake: 0.714 +(epoch: 18, iters: 1664, time: 0.064) G_GAN: 0.035 G_GAN_Feat: 0.764 G_ID: 0.418 G_Rec: 0.436 D_GP: 0.062 D_real: 0.817 D_fake: 0.966 +(epoch: 18, iters: 2064, time: 0.064) G_GAN: 0.670 G_GAN_Feat: 0.836 G_ID: 0.907 G_Rec: 0.700 D_GP: 0.036 D_real: 1.427 D_fake: 0.339 +(epoch: 18, iters: 2464, time: 0.064) G_GAN: 0.311 G_GAN_Feat: 0.656 G_ID: 0.421 G_Rec: 0.382 D_GP: 0.036 D_real: 1.046 D_fake: 0.689 +(epoch: 18, iters: 2864, time: 0.064) G_GAN: 0.215 G_GAN_Feat: 0.901 G_ID: 0.875 G_Rec: 0.400 D_GP: 0.107 D_real: 0.648 D_fake: 0.785 +(epoch: 18, iters: 3264, time: 0.064) G_GAN: 0.401 G_GAN_Feat: 0.563 G_ID: 0.333 G_Rec: 0.560 D_GP: 0.020 D_real: 1.323 D_fake: 0.599 +(epoch: 18, iters: 3664, time: 0.064) G_GAN: 0.447 G_GAN_Feat: 0.770 G_ID: 0.932 G_Rec: 0.422 D_GP: 0.056 D_real: 1.163 D_fake: 0.577 +(epoch: 18, iters: 4064, time: 0.064) G_GAN: 0.245 G_GAN_Feat: 1.142 G_ID: 0.406 G_Rec: 0.463 D_GP: 0.168 D_real: 0.481 D_fake: 0.755 +(epoch: 18, iters: 4464, time: 0.064) G_GAN: -0.035 G_GAN_Feat: 0.798 G_ID: 0.922 G_Rec: 0.424 D_GP: 0.055 D_real: 0.631 D_fake: 1.035 +(epoch: 18, iters: 4864, time: 0.064) G_GAN: 0.054 G_GAN_Feat: 0.667 G_ID: 0.360 G_Rec: 0.397 D_GP: 0.022 D_real: 0.807 D_fake: 0.946 +(epoch: 18, iters: 5264, time: 0.064) G_GAN: 0.119 G_GAN_Feat: 0.662 G_ID: 0.909 G_Rec: 0.358 D_GP: 0.041 D_real: 0.907 D_fake: 0.881 +(epoch: 18, iters: 5664, time: 0.064) G_GAN: -0.071 G_GAN_Feat: 0.711 G_ID: 0.430 G_Rec: 0.337 D_GP: 0.071 D_real: 0.592 D_fake: 1.071 +(epoch: 18, iters: 6064, time: 0.064) G_GAN: 0.442 G_GAN_Feat: 0.889 G_ID: 0.872 G_Rec: 0.386 D_GP: 0.090 D_real: 0.869 D_fake: 0.559 +(epoch: 18, iters: 6464, time: 0.064) G_GAN: 0.383 G_GAN_Feat: 0.674 G_ID: 0.389 G_Rec: 0.371 D_GP: 0.027 D_real: 1.155 D_fake: 0.617 +(epoch: 18, iters: 6864, time: 0.064) G_GAN: 0.303 G_GAN_Feat: 0.676 G_ID: 0.888 G_Rec: 0.414 D_GP: 0.020 D_real: 1.275 D_fake: 0.697 +(epoch: 18, iters: 7264, time: 0.064) G_GAN: 0.170 G_GAN_Feat: 0.726 G_ID: 0.363 G_Rec: 0.338 D_GP: 0.048 D_real: 0.834 D_fake: 0.830 +(epoch: 18, iters: 7664, time: 0.064) G_GAN: 0.119 G_GAN_Feat: 0.853 G_ID: 0.881 G_Rec: 0.444 D_GP: 0.080 D_real: 0.592 D_fake: 0.881 +(epoch: 18, iters: 8064, time: 0.064) G_GAN: 0.170 G_GAN_Feat: 0.848 G_ID: 0.351 G_Rec: 0.336 D_GP: 0.073 D_real: 0.890 D_fake: 0.830 +(epoch: 18, iters: 8464, time: 0.064) G_GAN: 0.749 G_GAN_Feat: 0.847 G_ID: 0.882 G_Rec: 0.514 D_GP: 0.037 D_real: 1.372 D_fake: 0.353 +(epoch: 19, iters: 256, time: 0.064) G_GAN: 0.373 G_GAN_Feat: 0.766 G_ID: 0.425 G_Rec: 0.405 D_GP: 0.056 D_real: 0.997 D_fake: 0.627 +(epoch: 19, iters: 656, time: 0.064) G_GAN: 0.048 G_GAN_Feat: 0.746 G_ID: 0.883 G_Rec: 0.359 D_GP: 0.067 D_real: 0.695 D_fake: 0.952 +(epoch: 19, iters: 1056, time: 0.064) G_GAN: 0.564 G_GAN_Feat: 0.877 G_ID: 0.413 G_Rec: 0.410 D_GP: 0.071 D_real: 1.042 D_fake: 0.449 +(epoch: 19, iters: 1456, time: 0.064) G_GAN: 0.392 G_GAN_Feat: 0.663 G_ID: 0.873 G_Rec: 0.317 D_GP: 0.045 D_real: 1.110 D_fake: 0.609 +(epoch: 19, iters: 1856, time: 0.064) G_GAN: 0.306 G_GAN_Feat: 0.874 G_ID: 0.460 G_Rec: 0.330 D_GP: 0.117 D_real: 0.739 D_fake: 0.694 +(epoch: 19, iters: 2256, time: 0.064) G_GAN: 0.132 G_GAN_Feat: 1.014 G_ID: 0.894 G_Rec: 0.388 D_GP: 0.067 D_real: 0.453 D_fake: 0.868 +(epoch: 19, iters: 2656, time: 0.064) G_GAN: 0.251 G_GAN_Feat: 0.857 G_ID: 0.410 G_Rec: 0.355 D_GP: 0.115 D_real: 1.046 D_fake: 0.750 +(epoch: 19, iters: 3056, time: 0.064) G_GAN: -0.001 G_GAN_Feat: 0.785 G_ID: 0.885 G_Rec: 0.373 D_GP: 0.055 D_real: 0.583 D_fake: 1.001 +(epoch: 19, iters: 3456, time: 0.064) G_GAN: 0.020 G_GAN_Feat: 0.887 G_ID: 0.396 G_Rec: 0.366 D_GP: 0.061 D_real: 0.521 D_fake: 0.980 +(epoch: 19, iters: 3856, time: 0.064) G_GAN: 0.223 G_GAN_Feat: 0.720 G_ID: 0.858 G_Rec: 0.404 D_GP: 0.015 D_real: 1.081 D_fake: 0.777 +(epoch: 19, iters: 4256, time: 0.064) G_GAN: -0.039 G_GAN_Feat: 0.914 G_ID: 0.362 G_Rec: 0.340 D_GP: 0.067 D_real: 0.444 D_fake: 1.039 +(epoch: 19, iters: 4656, time: 0.064) G_GAN: 0.438 G_GAN_Feat: 0.627 G_ID: 0.906 G_Rec: 0.374 D_GP: 0.022 D_real: 1.281 D_fake: 0.563 +(epoch: 19, iters: 5056, time: 0.064) G_GAN: 0.403 G_GAN_Feat: 0.851 G_ID: 0.355 G_Rec: 0.333 D_GP: 0.037 D_real: 1.045 D_fake: 0.597 +(epoch: 19, iters: 5456, time: 0.064) G_GAN: 0.525 G_GAN_Feat: 0.805 G_ID: 0.898 G_Rec: 0.366 D_GP: 0.036 D_real: 1.182 D_fake: 0.478 +(epoch: 19, iters: 5856, time: 0.064) G_GAN: 0.282 G_GAN_Feat: 0.647 G_ID: 0.399 G_Rec: 0.388 D_GP: 0.033 D_real: 1.014 D_fake: 0.718 +(epoch: 19, iters: 6256, time: 0.064) G_GAN: 0.344 G_GAN_Feat: 0.705 G_ID: 0.894 G_Rec: 0.340 D_GP: 0.050 D_real: 1.087 D_fake: 0.656 +(epoch: 19, iters: 6656, time: 0.064) G_GAN: -0.214 G_GAN_Feat: 0.677 G_ID: 0.339 G_Rec: 0.327 D_GP: 0.045 D_real: 0.536 D_fake: 1.214 +(epoch: 19, iters: 7056, time: 0.064) G_GAN: 0.395 G_GAN_Feat: 0.696 G_ID: 0.903 G_Rec: 0.344 D_GP: 0.027 D_real: 1.227 D_fake: 0.605 +(epoch: 19, iters: 7456, time: 0.064) G_GAN: 0.394 G_GAN_Feat: 0.458 G_ID: 0.440 G_Rec: 0.413 D_GP: 0.015 D_real: 1.362 D_fake: 0.606 +(epoch: 19, iters: 7856, time: 0.064) G_GAN: 0.282 G_GAN_Feat: 0.535 G_ID: 0.883 G_Rec: 0.355 D_GP: 0.018 D_real: 1.121 D_fake: 0.718 +(epoch: 19, iters: 8256, time: 0.064) G_GAN: 0.217 G_GAN_Feat: 0.655 G_ID: 0.348 G_Rec: 0.558 D_GP: 0.054 D_real: 0.967 D_fake: 0.783 +(epoch: 20, iters: 48, time: 0.064) G_GAN: 0.100 G_GAN_Feat: 0.851 G_ID: 0.884 G_Rec: 0.379 D_GP: 0.122 D_real: 0.541 D_fake: 0.900 +(epoch: 20, iters: 448, time: 0.064) G_GAN: 0.162 G_GAN_Feat: 0.736 G_ID: 0.402 G_Rec: 0.396 D_GP: 0.048 D_real: 0.860 D_fake: 0.838 +(epoch: 20, iters: 848, time: 0.064) G_GAN: 0.398 G_GAN_Feat: 0.635 G_ID: 0.879 G_Rec: 0.429 D_GP: 0.048 D_real: 1.136 D_fake: 0.602 +(epoch: 20, iters: 1248, time: 0.064) G_GAN: 0.187 G_GAN_Feat: 0.682 G_ID: 0.341 G_Rec: 0.331 D_GP: 0.038 D_real: 0.875 D_fake: 0.814 +(epoch: 20, iters: 1648, time: 0.064) G_GAN: 0.232 G_GAN_Feat: 0.769 G_ID: 0.858 G_Rec: 0.357 D_GP: 0.071 D_real: 0.812 D_fake: 0.772 +(epoch: 20, iters: 2048, time: 0.064) G_GAN: 0.226 G_GAN_Feat: 0.580 G_ID: 0.444 G_Rec: 0.297 D_GP: 0.029 D_real: 1.030 D_fake: 0.774 +(epoch: 20, iters: 2448, time: 0.064) G_GAN: -0.072 G_GAN_Feat: 0.886 G_ID: 0.848 G_Rec: 0.377 D_GP: 0.058 D_real: 0.324 D_fake: 1.072 +(epoch: 20, iters: 2848, time: 0.064) G_GAN: 0.219 G_GAN_Feat: 0.672 G_ID: 0.447 G_Rec: 0.368 D_GP: 0.038 D_real: 0.940 D_fake: 0.781 +(epoch: 20, iters: 3248, time: 0.064) G_GAN: 0.400 G_GAN_Feat: 0.885 G_ID: 0.864 G_Rec: 0.393 D_GP: 0.065 D_real: 0.967 D_fake: 0.642 +(epoch: 20, iters: 3648, time: 0.064) G_GAN: 0.619 G_GAN_Feat: 0.804 G_ID: 0.374 G_Rec: 0.368 D_GP: 0.033 D_real: 1.300 D_fake: 0.431 +(epoch: 20, iters: 4048, time: 0.064) G_GAN: 0.482 G_GAN_Feat: 1.070 G_ID: 0.848 G_Rec: 0.365 D_GP: 0.096 D_real: 0.782 D_fake: 0.518 +(epoch: 20, iters: 4448, time: 0.064) G_GAN: 0.005 G_GAN_Feat: 0.760 G_ID: 0.411 G_Rec: 0.345 D_GP: 0.054 D_real: 0.551 D_fake: 0.995 +(epoch: 20, iters: 4848, time: 0.064) G_GAN: 0.120 G_GAN_Feat: 0.672 G_ID: 0.890 G_Rec: 0.342 D_GP: 0.021 D_real: 1.045 D_fake: 0.880 +(epoch: 20, iters: 5248, time: 0.064) G_GAN: 0.238 G_GAN_Feat: 0.732 G_ID: 0.312 G_Rec: 0.347 D_GP: 0.028 D_real: 1.055 D_fake: 0.763 +(epoch: 20, iters: 5648, time: 0.064) G_GAN: 0.163 G_GAN_Feat: 0.791 G_ID: 0.868 G_Rec: 0.350 D_GP: 0.036 D_real: 1.002 D_fake: 0.837 +(epoch: 20, iters: 6048, time: 0.064) G_GAN: 0.213 G_GAN_Feat: 0.883 G_ID: 0.374 G_Rec: 0.337 D_GP: 0.060 D_real: 0.823 D_fake: 0.788 +(epoch: 20, iters: 6448, time: 0.064) G_GAN: 0.378 G_GAN_Feat: 0.761 G_ID: 0.824 G_Rec: 0.365 D_GP: 0.038 D_real: 1.163 D_fake: 0.622 +(epoch: 20, iters: 6848, time: 0.064) G_GAN: 0.074 G_GAN_Feat: 0.901 G_ID: 0.392 G_Rec: 0.551 D_GP: 0.077 D_real: 0.670 D_fake: 0.926 +(epoch: 20, iters: 7248, time: 0.064) G_GAN: 0.011 G_GAN_Feat: 0.819 G_ID: 0.828 G_Rec: 0.368 D_GP: 0.077 D_real: 0.687 D_fake: 0.989 +(epoch: 20, iters: 7648, time: 0.064) G_GAN: -0.041 G_GAN_Feat: 0.801 G_ID: 0.395 G_Rec: 0.362 D_GP: 0.065 D_real: 0.703 D_fake: 1.041 +(epoch: 20, iters: 8048, time: 0.064) G_GAN: 0.235 G_GAN_Feat: 0.840 G_ID: 0.835 G_Rec: 0.361 D_GP: 0.041 D_real: 0.968 D_fake: 0.766 +(epoch: 20, iters: 8448, time: 0.064) G_GAN: 0.260 G_GAN_Feat: 1.023 G_ID: 0.479 G_Rec: 0.378 D_GP: 0.076 D_real: 0.750 D_fake: 0.742 +(epoch: 21, iters: 240, time: 0.064) G_GAN: 0.397 G_GAN_Feat: 0.861 G_ID: 0.831 G_Rec: 0.343 D_GP: 0.044 D_real: 1.051 D_fake: 0.603 +(epoch: 21, iters: 640, time: 0.064) G_GAN: 0.259 G_GAN_Feat: 1.022 G_ID: 0.368 G_Rec: 0.449 D_GP: 0.111 D_real: 1.136 D_fake: 0.743 +(epoch: 21, iters: 1040, time: 0.064) G_GAN: 0.385 G_GAN_Feat: 0.983 G_ID: 0.853 G_Rec: 0.399 D_GP: 0.063 D_real: 0.908 D_fake: 0.616 +(epoch: 21, iters: 1440, time: 0.064) G_GAN: 0.233 G_GAN_Feat: 0.867 G_ID: 0.350 G_Rec: 0.409 D_GP: 0.069 D_real: 0.898 D_fake: 0.769 +(epoch: 21, iters: 1840, time: 0.064) G_GAN: 0.343 G_GAN_Feat: 0.905 G_ID: 0.815 G_Rec: 0.357 D_GP: 0.038 D_real: 0.951 D_fake: 0.658 +(epoch: 21, iters: 2240, time: 0.064) G_GAN: 0.283 G_GAN_Feat: 0.762 G_ID: 0.352 G_Rec: 0.327 D_GP: 0.044 D_real: 1.094 D_fake: 0.717 +(epoch: 21, iters: 2640, time: 0.064) G_GAN: 0.174 G_GAN_Feat: 0.693 G_ID: 0.865 G_Rec: 0.349 D_GP: 0.032 D_real: 0.918 D_fake: 0.826 +(epoch: 21, iters: 3040, time: 0.064) G_GAN: 0.383 G_GAN_Feat: 0.959 G_ID: 0.499 G_Rec: 0.417 D_GP: 0.078 D_real: 0.715 D_fake: 0.617 +(epoch: 21, iters: 3440, time: 0.064) G_GAN: 0.253 G_GAN_Feat: 1.118 G_ID: 0.878 G_Rec: 0.399 D_GP: 0.138 D_real: 0.589 D_fake: 0.748 +(epoch: 21, iters: 3840, time: 0.064) G_GAN: 0.227 G_GAN_Feat: 0.643 G_ID: 0.382 G_Rec: 0.417 D_GP: 0.022 D_real: 1.010 D_fake: 0.773 +(epoch: 21, iters: 4240, time: 0.064) G_GAN: 0.254 G_GAN_Feat: 0.765 G_ID: 0.883 G_Rec: 0.513 D_GP: 0.069 D_real: 0.975 D_fake: 0.746 +(epoch: 21, iters: 4640, time: 0.064) G_GAN: 0.186 G_GAN_Feat: 0.994 G_ID: 0.423 G_Rec: 0.373 D_GP: 0.093 D_real: 0.548 D_fake: 0.816 +(epoch: 21, iters: 5040, time: 0.064) G_GAN: -0.040 G_GAN_Feat: 0.632 G_ID: 0.854 G_Rec: 0.431 D_GP: 0.023 D_real: 0.769 D_fake: 1.040 +(epoch: 21, iters: 5440, time: 0.064) G_GAN: 0.280 G_GAN_Feat: 0.735 G_ID: 0.429 G_Rec: 0.370 D_GP: 0.036 D_real: 1.028 D_fake: 0.720 +(epoch: 21, iters: 5840, time: 0.064) G_GAN: 0.516 G_GAN_Feat: 0.769 G_ID: 0.833 G_Rec: 0.368 D_GP: 0.030 D_real: 1.263 D_fake: 0.485 +(epoch: 21, iters: 6240, time: 0.064) G_GAN: 0.692 G_GAN_Feat: 0.502 G_ID: 0.379 G_Rec: 0.391 D_GP: 0.019 D_real: 1.636 D_fake: 0.337 +(epoch: 21, iters: 6640, time: 0.064) G_GAN: 0.320 G_GAN_Feat: 0.536 G_ID: 0.835 G_Rec: 0.457 D_GP: 0.016 D_real: 1.192 D_fake: 0.680 +(epoch: 21, iters: 7040, time: 0.064) G_GAN: 0.135 G_GAN_Feat: 0.487 G_ID: 0.366 G_Rec: 0.478 D_GP: 0.018 D_real: 1.017 D_fake: 0.865 +(epoch: 21, iters: 7440, time: 0.064) G_GAN: 0.202 G_GAN_Feat: 0.545 G_ID: 0.845 G_Rec: 0.376 D_GP: 0.027 D_real: 1.000 D_fake: 0.798 +(epoch: 21, iters: 7840, time: 0.064) G_GAN: 0.357 G_GAN_Feat: 0.565 G_ID: 0.463 G_Rec: 0.373 D_GP: 0.035 D_real: 1.184 D_fake: 0.643 +(epoch: 21, iters: 8240, time: 0.064) G_GAN: 0.224 G_GAN_Feat: 0.634 G_ID: 0.801 G_Rec: 0.347 D_GP: 0.059 D_real: 0.973 D_fake: 0.776 +(epoch: 22, iters: 32, time: 0.064) G_GAN: 0.186 G_GAN_Feat: 0.768 G_ID: 0.390 G_Rec: 0.398 D_GP: 0.056 D_real: 0.782 D_fake: 0.815 +(epoch: 22, iters: 432, time: 0.064) G_GAN: 0.240 G_GAN_Feat: 0.669 G_ID: 0.887 G_Rec: 0.433 D_GP: 0.043 D_real: 1.042 D_fake: 0.760 +(epoch: 22, iters: 832, time: 0.064) G_GAN: 0.272 G_GAN_Feat: 0.821 G_ID: 0.403 G_Rec: 0.357 D_GP: 0.067 D_real: 0.795 D_fake: 0.732 +(epoch: 22, iters: 1232, time: 0.064) G_GAN: 0.219 G_GAN_Feat: 0.794 G_ID: 0.808 G_Rec: 0.652 D_GP: 0.070 D_real: 0.748 D_fake: 0.781 +(epoch: 22, iters: 1632, time: 0.064) G_GAN: 0.384 G_GAN_Feat: 0.734 G_ID: 0.425 G_Rec: 0.356 D_GP: 0.067 D_real: 1.018 D_fake: 0.616 +(epoch: 22, iters: 2032, time: 0.064) G_GAN: 0.209 G_GAN_Feat: 0.731 G_ID: 0.862 G_Rec: 0.339 D_GP: 0.061 D_real: 0.920 D_fake: 0.791 +(epoch: 22, iters: 2432, time: 0.064) G_GAN: 0.250 G_GAN_Feat: 1.068 G_ID: 0.413 G_Rec: 0.420 D_GP: 0.102 D_real: 0.496 D_fake: 0.750 +(epoch: 22, iters: 2832, time: 0.064) G_GAN: 0.228 G_GAN_Feat: 0.706 G_ID: 0.837 G_Rec: 0.344 D_GP: 0.026 D_real: 0.959 D_fake: 0.772 +(epoch: 22, iters: 3232, time: 0.064) G_GAN: 0.519 G_GAN_Feat: 0.773 G_ID: 0.412 G_Rec: 0.377 D_GP: 0.065 D_real: 1.115 D_fake: 0.499 +(epoch: 22, iters: 3632, time: 0.064) G_GAN: 0.242 G_GAN_Feat: 0.480 G_ID: 0.856 G_Rec: 0.330 D_GP: 0.016 D_real: 1.147 D_fake: 0.758 +(epoch: 22, iters: 4032, time: 0.064) G_GAN: 0.020 G_GAN_Feat: 0.553 G_ID: 0.332 G_Rec: 0.406 D_GP: 0.029 D_real: 0.906 D_fake: 0.980 +(epoch: 22, iters: 4432, time: 0.064) G_GAN: 0.905 G_GAN_Feat: 0.584 G_ID: 0.846 G_Rec: 0.438 D_GP: 0.023 D_real: 1.698 D_fake: 0.215 +(epoch: 22, iters: 4832, time: 0.064) G_GAN: 0.284 G_GAN_Feat: 0.628 G_ID: 0.325 G_Rec: 0.452 D_GP: 0.049 D_real: 0.983 D_fake: 0.717 +(epoch: 22, iters: 5232, time: 0.064) G_GAN: 0.385 G_GAN_Feat: 0.653 G_ID: 0.857 G_Rec: 0.425 D_GP: 0.054 D_real: 1.065 D_fake: 0.616 +(epoch: 22, iters: 5632, time: 0.064) G_GAN: 0.245 G_GAN_Feat: 0.570 G_ID: 0.414 G_Rec: 0.338 D_GP: 0.041 D_real: 1.065 D_fake: 0.757 +(epoch: 22, iters: 6032, time: 0.064) G_GAN: 0.122 G_GAN_Feat: 0.709 G_ID: 0.848 G_Rec: 0.353 D_GP: 0.076 D_real: 0.758 D_fake: 0.878 +(epoch: 22, iters: 6432, time: 0.064) G_GAN: 0.408 G_GAN_Feat: 0.785 G_ID: 0.388 G_Rec: 0.401 D_GP: 0.106 D_real: 1.011 D_fake: 0.596 +(epoch: 22, iters: 6832, time: 0.064) G_GAN: 0.195 G_GAN_Feat: 0.671 G_ID: 0.822 G_Rec: 0.421 D_GP: 0.066 D_real: 0.917 D_fake: 0.805 +(epoch: 22, iters: 7232, time: 0.064) G_GAN: 0.194 G_GAN_Feat: 1.041 G_ID: 0.399 G_Rec: 0.378 D_GP: 0.193 D_real: 0.479 D_fake: 0.806 +(epoch: 22, iters: 7632, time: 0.064) G_GAN: 0.326 G_GAN_Feat: 0.737 G_ID: 0.818 G_Rec: 0.329 D_GP: 0.042 D_real: 1.092 D_fake: 0.674 +(epoch: 22, iters: 8032, time: 0.064) G_GAN: 0.368 G_GAN_Feat: 0.731 G_ID: 0.391 G_Rec: 0.381 D_GP: 0.046 D_real: 1.178 D_fake: 0.633 +(epoch: 22, iters: 8432, time: 0.064) G_GAN: 0.160 G_GAN_Feat: 0.703 G_ID: 0.811 G_Rec: 0.468 D_GP: 0.063 D_real: 0.884 D_fake: 0.840 +(epoch: 23, iters: 224, time: 0.064) G_GAN: 0.237 G_GAN_Feat: 0.809 G_ID: 0.400 G_Rec: 0.412 D_GP: 0.089 D_real: 0.695 D_fake: 0.763 +(epoch: 23, iters: 624, time: 0.064) G_GAN: 0.318 G_GAN_Feat: 0.728 G_ID: 0.839 G_Rec: 0.418 D_GP: 0.036 D_real: 1.034 D_fake: 0.683 +(epoch: 23, iters: 1024, time: 0.064) G_GAN: 0.425 G_GAN_Feat: 0.582 G_ID: 0.381 G_Rec: 0.395 D_GP: 0.022 D_real: 1.328 D_fake: 0.579 +(epoch: 23, iters: 1424, time: 0.064) G_GAN: 0.512 G_GAN_Feat: 0.611 G_ID: 0.798 G_Rec: 0.431 D_GP: 0.024 D_real: 1.329 D_fake: 0.490 +(epoch: 23, iters: 1824, time: 0.064) G_GAN: 0.221 G_GAN_Feat: 0.681 G_ID: 0.467 G_Rec: 0.410 D_GP: 0.046 D_real: 1.020 D_fake: 0.779 +(epoch: 23, iters: 2224, time: 0.064) G_GAN: 0.091 G_GAN_Feat: 0.639 G_ID: 0.834 G_Rec: 0.427 D_GP: 0.039 D_real: 0.776 D_fake: 0.913 +(epoch: 23, iters: 2624, time: 0.064) G_GAN: 0.283 G_GAN_Feat: 0.852 G_ID: 0.429 G_Rec: 0.376 D_GP: 0.056 D_real: 0.799 D_fake: 0.717 +(epoch: 23, iters: 3024, time: 0.064) G_GAN: 0.328 G_GAN_Feat: 1.085 G_ID: 0.835 G_Rec: 0.382 D_GP: 0.167 D_real: 0.587 D_fake: 0.675 +(epoch: 23, iters: 3424, time: 0.064) G_GAN: 0.042 G_GAN_Feat: 0.760 G_ID: 0.417 G_Rec: 0.346 D_GP: 0.077 D_real: 0.567 D_fake: 0.960 +(epoch: 23, iters: 3824, time: 0.064) G_GAN: 0.151 G_GAN_Feat: 0.648 G_ID: 0.824 G_Rec: 0.409 D_GP: 0.035 D_real: 0.948 D_fake: 0.849 +(epoch: 23, iters: 4224, time: 0.064) G_GAN: 0.395 G_GAN_Feat: 0.823 G_ID: 0.461 G_Rec: 0.451 D_GP: 0.068 D_real: 0.856 D_fake: 0.606 +(epoch: 23, iters: 4624, time: 0.064) G_GAN: 0.302 G_GAN_Feat: 0.757 G_ID: 0.801 G_Rec: 0.342 D_GP: 0.071 D_real: 0.830 D_fake: 0.701 +(epoch: 23, iters: 5024, time: 0.064) G_GAN: 0.084 G_GAN_Feat: 0.829 G_ID: 0.381 G_Rec: 0.405 D_GP: 0.040 D_real: 0.625 D_fake: 0.916 +(epoch: 23, iters: 5424, time: 0.064) G_GAN: 0.066 G_GAN_Feat: 0.692 G_ID: 0.868 G_Rec: 0.361 D_GP: 0.040 D_real: 0.796 D_fake: 0.935 +(epoch: 23, iters: 5824, time: 0.064) G_GAN: 0.140 G_GAN_Feat: 1.091 G_ID: 0.367 G_Rec: 0.384 D_GP: 0.086 D_real: 0.455 D_fake: 0.860 +(epoch: 23, iters: 6224, time: 0.064) G_GAN: 0.200 G_GAN_Feat: 0.621 G_ID: 0.817 G_Rec: 0.357 D_GP: 0.024 D_real: 1.015 D_fake: 0.800 +(epoch: 23, iters: 6624, time: 0.064) G_GAN: 0.546 G_GAN_Feat: 0.837 G_ID: 0.341 G_Rec: 0.379 D_GP: 0.107 D_real: 1.114 D_fake: 0.455 +(epoch: 23, iters: 7024, time: 0.064) G_GAN: 0.292 G_GAN_Feat: 1.085 G_ID: 0.798 G_Rec: 0.397 D_GP: 0.105 D_real: 0.643 D_fake: 0.727 +(epoch: 23, iters: 7424, time: 0.064) G_GAN: 0.612 G_GAN_Feat: 0.521 G_ID: 0.412 G_Rec: 0.445 D_GP: 0.018 D_real: 1.482 D_fake: 0.393 +(epoch: 23, iters: 7824, time: 0.064) G_GAN: 0.126 G_GAN_Feat: 0.536 G_ID: 0.777 G_Rec: 0.383 D_GP: 0.020 D_real: 0.997 D_fake: 0.874 +(epoch: 23, iters: 8224, time: 0.064) G_GAN: 0.190 G_GAN_Feat: 0.503 G_ID: 0.437 G_Rec: 0.324 D_GP: 0.025 D_real: 1.040 D_fake: 0.811 +(epoch: 24, iters: 16, time: 0.064) G_GAN: 0.407 G_GAN_Feat: 0.548 G_ID: 0.858 G_Rec: 0.339 D_GP: 0.031 D_real: 1.196 D_fake: 0.594 +(epoch: 24, iters: 416, time: 0.064) G_GAN: 0.163 G_GAN_Feat: 0.530 G_ID: 0.429 G_Rec: 0.309 D_GP: 0.031 D_real: 1.005 D_fake: 0.837 +(epoch: 24, iters: 816, time: 0.064) G_GAN: 0.440 G_GAN_Feat: 0.601 G_ID: 0.829 G_Rec: 0.367 D_GP: 0.042 D_real: 1.186 D_fake: 0.567 +(epoch: 24, iters: 1216, time: 0.064) G_GAN: 0.292 G_GAN_Feat: 0.648 G_ID: 0.337 G_Rec: 0.367 D_GP: 0.038 D_real: 1.063 D_fake: 0.709 +(epoch: 24, iters: 1616, time: 0.064) G_GAN: 0.187 G_GAN_Feat: 0.672 G_ID: 0.821 G_Rec: 0.377 D_GP: 0.055 D_real: 0.846 D_fake: 0.813 +(epoch: 24, iters: 2016, time: 0.064) G_GAN: 0.151 G_GAN_Feat: 0.613 G_ID: 0.377 G_Rec: 0.336 D_GP: 0.060 D_real: 0.917 D_fake: 0.849 +(epoch: 24, iters: 2416, time: 0.064) G_GAN: 0.060 G_GAN_Feat: 0.614 G_ID: 0.823 G_Rec: 0.316 D_GP: 0.063 D_real: 0.854 D_fake: 0.940 +(epoch: 24, iters: 2816, time: 0.064) G_GAN: 0.373 G_GAN_Feat: 0.802 G_ID: 0.403 G_Rec: 0.462 D_GP: 0.046 D_real: 1.095 D_fake: 0.654 +(epoch: 24, iters: 3216, time: 0.064) G_GAN: 0.185 G_GAN_Feat: 1.155 G_ID: 0.808 G_Rec: 0.372 D_GP: 0.227 D_real: 0.507 D_fake: 0.817 +(epoch: 24, iters: 3616, time: 0.064) G_GAN: 0.090 G_GAN_Feat: 0.616 G_ID: 0.358 G_Rec: 0.472 D_GP: 0.044 D_real: 0.885 D_fake: 0.910 +(epoch: 24, iters: 4016, time: 0.064) G_GAN: 0.474 G_GAN_Feat: 0.740 G_ID: 0.790 G_Rec: 0.530 D_GP: 0.030 D_real: 1.249 D_fake: 0.528 +(epoch: 24, iters: 4416, time: 0.064) G_GAN: 0.552 G_GAN_Feat: 0.692 G_ID: 0.360 G_Rec: 0.366 D_GP: 0.066 D_real: 1.223 D_fake: 0.453 +(epoch: 24, iters: 4816, time: 0.064) G_GAN: 0.499 G_GAN_Feat: 0.877 G_ID: 0.834 G_Rec: 0.381 D_GP: 0.050 D_real: 1.068 D_fake: 0.523 +(epoch: 24, iters: 5216, time: 0.064) G_GAN: -0.163 G_GAN_Feat: 0.733 G_ID: 0.419 G_Rec: 0.388 D_GP: 0.084 D_real: 0.455 D_fake: 1.164 +(epoch: 24, iters: 5616, time: 0.064) G_GAN: 0.282 G_GAN_Feat: 0.861 G_ID: 0.789 G_Rec: 0.387 D_GP: 0.042 D_real: 1.048 D_fake: 0.722 +(epoch: 24, iters: 6016, time: 0.064) G_GAN: 0.479 G_GAN_Feat: 0.684 G_ID: 0.385 G_Rec: 0.419 D_GP: 0.041 D_real: 1.225 D_fake: 0.522 +(epoch: 24, iters: 6416, time: 0.064) G_GAN: 0.096 G_GAN_Feat: 0.729 G_ID: 0.795 G_Rec: 0.389 D_GP: 0.039 D_real: 0.780 D_fake: 0.906 +(epoch: 24, iters: 6816, time: 0.064) G_GAN: 0.584 G_GAN_Feat: 0.724 G_ID: 0.413 G_Rec: 0.463 D_GP: 0.025 D_real: 1.323 D_fake: 0.418 +(epoch: 24, iters: 7216, time: 0.064) G_GAN: 0.346 G_GAN_Feat: 0.694 G_ID: 0.798 G_Rec: 0.473 D_GP: 0.024 D_real: 1.140 D_fake: 0.654 +(epoch: 24, iters: 7616, time: 0.064) G_GAN: 0.723 G_GAN_Feat: 0.692 G_ID: 0.372 G_Rec: 0.351 D_GP: 0.040 D_real: 1.416 D_fake: 0.303 +(epoch: 24, iters: 8016, time: 0.064) G_GAN: 0.487 G_GAN_Feat: 0.668 G_ID: 0.813 G_Rec: 0.334 D_GP: 0.028 D_real: 1.284 D_fake: 0.514 +(epoch: 24, iters: 8416, time: 0.064) G_GAN: 0.026 G_GAN_Feat: 0.897 G_ID: 0.326 G_Rec: 0.504 D_GP: 0.114 D_real: 0.510 D_fake: 0.975 +(epoch: 25, iters: 208, time: 0.064) G_GAN: 0.234 G_GAN_Feat: 0.655 G_ID: 0.834 G_Rec: 0.341 D_GP: 0.051 D_real: 0.936 D_fake: 0.766 +(epoch: 25, iters: 608, time: 0.064) G_GAN: 0.338 G_GAN_Feat: 0.835 G_ID: 0.417 G_Rec: 0.336 D_GP: 0.108 D_real: 0.977 D_fake: 0.683 +(epoch: 25, iters: 1008, time: 0.064) G_GAN: 0.161 G_GAN_Feat: 1.075 G_ID: 0.757 G_Rec: 0.405 D_GP: 0.326 D_real: 0.554 D_fake: 0.840 +(epoch: 25, iters: 1408, time: 0.064) G_GAN: 0.218 G_GAN_Feat: 1.259 G_ID: 0.407 G_Rec: 0.430 D_GP: 0.367 D_real: 0.306 D_fake: 0.782 +(epoch: 25, iters: 1808, time: 0.064) G_GAN: 0.233 G_GAN_Feat: 0.793 G_ID: 0.824 G_Rec: 0.465 D_GP: 0.083 D_real: 0.757 D_fake: 0.768 +(epoch: 25, iters: 2208, time: 0.064) G_GAN: 0.158 G_GAN_Feat: 0.636 G_ID: 0.390 G_Rec: 0.354 D_GP: 0.024 D_real: 1.040 D_fake: 0.842 +(epoch: 25, iters: 2608, time: 0.064) G_GAN: 0.584 G_GAN_Feat: 0.760 G_ID: 0.782 G_Rec: 0.336 D_GP: 0.070 D_real: 1.186 D_fake: 0.422 +(epoch: 25, iters: 3008, time: 0.064) G_GAN: 0.230 G_GAN_Feat: 0.589 G_ID: 0.351 G_Rec: 0.326 D_GP: 0.035 D_real: 1.026 D_fake: 0.770 +(epoch: 25, iters: 3408, time: 0.064) G_GAN: 0.664 G_GAN_Feat: 0.724 G_ID: 0.786 G_Rec: 0.385 D_GP: 0.034 D_real: 1.442 D_fake: 0.377 +(epoch: 25, iters: 3808, time: 0.064) G_GAN: 0.399 G_GAN_Feat: 0.882 G_ID: 0.333 G_Rec: 0.409 D_GP: 0.040 D_real: 1.010 D_fake: 0.601 +(epoch: 25, iters: 4208, time: 0.064) G_GAN: 0.395 G_GAN_Feat: 1.238 G_ID: 0.777 G_Rec: 0.521 D_GP: 0.239 D_real: 0.593 D_fake: 0.613 +(epoch: 25, iters: 4608, time: 0.064) G_GAN: 0.470 G_GAN_Feat: 0.855 G_ID: 0.359 G_Rec: 0.328 D_GP: 0.076 D_real: 1.008 D_fake: 0.530 +(epoch: 25, iters: 5008, time: 0.064) G_GAN: 0.482 G_GAN_Feat: 0.744 G_ID: 0.835 G_Rec: 0.351 D_GP: 0.040 D_real: 1.247 D_fake: 0.534 +(epoch: 25, iters: 5408, time: 0.064) G_GAN: 0.522 G_GAN_Feat: 0.580 G_ID: 0.351 G_Rec: 0.347 D_GP: 0.022 D_real: 1.313 D_fake: 0.480 +(epoch: 25, iters: 5808, time: 0.064) G_GAN: 0.033 G_GAN_Feat: 0.881 G_ID: 0.819 G_Rec: 0.385 D_GP: 0.082 D_real: 0.444 D_fake: 0.967 +(epoch: 25, iters: 6208, time: 0.064) G_GAN: 0.233 G_GAN_Feat: 0.725 G_ID: 0.348 G_Rec: 0.372 D_GP: 0.059 D_real: 0.869 D_fake: 0.767 +(epoch: 25, iters: 6608, time: 0.064) G_GAN: 0.064 G_GAN_Feat: 0.830 G_ID: 0.796 G_Rec: 0.354 D_GP: 0.047 D_real: 1.040 D_fake: 0.937 +(epoch: 25, iters: 7008, time: 0.064) G_GAN: 0.120 G_GAN_Feat: 0.949 G_ID: 0.323 G_Rec: 0.494 D_GP: 0.095 D_real: 0.573 D_fake: 0.881 +(epoch: 25, iters: 7408, time: 0.064) G_GAN: 0.285 G_GAN_Feat: 0.631 G_ID: 0.774 G_Rec: 0.407 D_GP: 0.026 D_real: 1.057 D_fake: 0.715 +(epoch: 25, iters: 7808, time: 0.064) G_GAN: 0.287 G_GAN_Feat: 1.059 G_ID: 0.431 G_Rec: 0.440 D_GP: 0.110 D_real: 0.699 D_fake: 0.726 +(epoch: 25, iters: 8208, time: 0.064) G_GAN: 0.653 G_GAN_Feat: 0.834 G_ID: 0.800 G_Rec: 0.370 D_GP: 0.096 D_real: 1.109 D_fake: 0.365 +(epoch: 25, iters: 8608, time: 0.064) G_GAN: 0.388 G_GAN_Feat: 0.560 G_ID: 0.421 G_Rec: 0.384 D_GP: 0.024 D_real: 1.190 D_fake: 0.612 +(epoch: 26, iters: 400, time: 0.064) G_GAN: 0.157 G_GAN_Feat: 1.221 G_ID: 0.802 G_Rec: 0.499 D_GP: 0.127 D_real: 0.442 D_fake: 0.843 +(epoch: 26, iters: 800, time: 0.064) G_GAN: 0.363 G_GAN_Feat: 1.063 G_ID: 0.365 G_Rec: 0.446 D_GP: 0.058 D_real: 1.194 D_fake: 0.642 +(epoch: 26, iters: 1200, time: 0.064) G_GAN: 0.388 G_GAN_Feat: 1.135 G_ID: 0.776 G_Rec: 0.353 D_GP: 0.092 D_real: 1.084 D_fake: 0.660 +(epoch: 26, iters: 1600, time: 0.064) G_GAN: 0.130 G_GAN_Feat: 0.686 G_ID: 0.368 G_Rec: 0.500 D_GP: 0.039 D_real: 0.814 D_fake: 0.871 +(epoch: 26, iters: 2000, time: 0.064) G_GAN: 0.390 G_GAN_Feat: 0.748 G_ID: 0.783 G_Rec: 0.364 D_GP: 0.043 D_real: 1.035 D_fake: 0.610 +(epoch: 26, iters: 2400, time: 0.064) G_GAN: 0.407 G_GAN_Feat: 0.584 G_ID: 0.334 G_Rec: 0.376 D_GP: 0.017 D_real: 1.270 D_fake: 0.593 +(epoch: 26, iters: 2800, time: 0.064) G_GAN: 0.314 G_GAN_Feat: 0.783 G_ID: 0.789 G_Rec: 0.398 D_GP: 0.040 D_real: 1.142 D_fake: 0.687 +(epoch: 26, iters: 3200, time: 0.064) G_GAN: 0.069 G_GAN_Feat: 0.947 G_ID: 0.386 G_Rec: 0.475 D_GP: 0.425 D_real: 0.524 D_fake: 0.931 +(epoch: 26, iters: 3600, time: 0.064) G_GAN: 0.575 G_GAN_Feat: 0.703 G_ID: 0.747 G_Rec: 0.374 D_GP: 0.028 D_real: 1.354 D_fake: 0.438 +(epoch: 26, iters: 4000, time: 0.064) G_GAN: 0.371 G_GAN_Feat: 0.837 G_ID: 0.394 G_Rec: 0.388 D_GP: 0.070 D_real: 0.965 D_fake: 0.649 +(epoch: 26, iters: 4400, time: 0.064) G_GAN: 0.247 G_GAN_Feat: 0.749 G_ID: 0.784 G_Rec: 0.431 D_GP: 0.038 D_real: 0.883 D_fake: 0.753 +(epoch: 26, iters: 4800, time: 0.064) G_GAN: 0.663 G_GAN_Feat: 0.804 G_ID: 0.396 G_Rec: 0.530 D_GP: 0.038 D_real: 1.429 D_fake: 0.356 +(epoch: 26, iters: 5200, time: 0.064) G_GAN: 0.298 G_GAN_Feat: 0.658 G_ID: 0.762 G_Rec: 0.356 D_GP: 0.033 D_real: 1.037 D_fake: 0.702 +(epoch: 26, iters: 5600, time: 0.064) G_GAN: 0.409 G_GAN_Feat: 0.739 G_ID: 0.355 G_Rec: 0.356 D_GP: 0.051 D_real: 1.091 D_fake: 0.592 +(epoch: 26, iters: 6000, time: 0.064) G_GAN: 0.249 G_GAN_Feat: 0.751 G_ID: 0.751 G_Rec: 0.378 D_GP: 0.051 D_real: 0.970 D_fake: 0.764 +(epoch: 26, iters: 6400, time: 0.064) G_GAN: 0.355 G_GAN_Feat: 1.151 G_ID: 0.366 G_Rec: 0.355 D_GP: 0.361 D_real: 0.296 D_fake: 0.670 +(epoch: 26, iters: 6800, time: 0.064) G_GAN: 0.456 G_GAN_Feat: 0.756 G_ID: 0.798 G_Rec: 0.376 D_GP: 0.051 D_real: 1.095 D_fake: 0.547 +(epoch: 26, iters: 7200, time: 0.064) G_GAN: 0.232 G_GAN_Feat: 0.633 G_ID: 0.396 G_Rec: 0.361 D_GP: 0.024 D_real: 0.970 D_fake: 0.771 +(epoch: 26, iters: 7600, time: 0.064) G_GAN: 0.157 G_GAN_Feat: 0.693 G_ID: 0.769 G_Rec: 0.322 D_GP: 0.051 D_real: 0.861 D_fake: 0.843 +(epoch: 26, iters: 8000, time: 0.064) G_GAN: 0.414 G_GAN_Feat: 1.083 G_ID: 0.355 G_Rec: 0.425 D_GP: 0.239 D_real: 0.565 D_fake: 0.588 +(epoch: 26, iters: 8400, time: 0.064) G_GAN: 0.328 G_GAN_Feat: 0.521 G_ID: 0.783 G_Rec: 0.337 D_GP: 0.022 D_real: 1.211 D_fake: 0.672 +(epoch: 27, iters: 192, time: 0.064) G_GAN: 0.306 G_GAN_Feat: 0.554 G_ID: 0.374 G_Rec: 0.358 D_GP: 0.026 D_real: 1.152 D_fake: 0.694 +(epoch: 27, iters: 592, time: 0.064) G_GAN: 0.166 G_GAN_Feat: 0.578 G_ID: 0.771 G_Rec: 0.345 D_GP: 0.040 D_real: 0.937 D_fake: 0.835 +(epoch: 27, iters: 992, time: 0.064) G_GAN: 0.477 G_GAN_Feat: 0.661 G_ID: 0.330 G_Rec: 0.367 D_GP: 0.045 D_real: 1.215 D_fake: 0.523 +(epoch: 27, iters: 1392, time: 0.064) G_GAN: 0.521 G_GAN_Feat: 0.677 G_ID: 0.725 G_Rec: 0.350 D_GP: 0.044 D_real: 1.247 D_fake: 0.479 +(epoch: 27, iters: 1792, time: 0.064) G_GAN: 0.348 G_GAN_Feat: 0.927 G_ID: 0.343 G_Rec: 0.353 D_GP: 0.061 D_real: 0.773 D_fake: 0.652 +(epoch: 27, iters: 2192, time: 0.064) G_GAN: 0.314 G_GAN_Feat: 0.740 G_ID: 0.758 G_Rec: 0.386 D_GP: 0.039 D_real: 0.976 D_fake: 0.691 +(epoch: 27, iters: 2592, time: 0.064) G_GAN: 0.559 G_GAN_Feat: 0.904 G_ID: 0.390 G_Rec: 0.431 D_GP: 0.118 D_real: 0.838 D_fake: 0.523 +(epoch: 27, iters: 2992, time: 0.064) G_GAN: 0.465 G_GAN_Feat: 1.200 G_ID: 0.780 G_Rec: 0.403 D_GP: 0.243 D_real: 0.500 D_fake: 0.545 +(epoch: 27, iters: 3392, time: 0.064) G_GAN: 0.363 G_GAN_Feat: 0.778 G_ID: 0.362 G_Rec: 0.599 D_GP: 0.028 D_real: 1.114 D_fake: 0.638 +(epoch: 27, iters: 3792, time: 0.064) G_GAN: 0.604 G_GAN_Feat: 0.776 G_ID: 0.793 G_Rec: 0.378 D_GP: 0.091 D_real: 1.221 D_fake: 0.495 +(epoch: 27, iters: 4192, time: 0.064) G_GAN: 0.331 G_GAN_Feat: 0.826 G_ID: 0.318 G_Rec: 0.345 D_GP: 0.095 D_real: 0.823 D_fake: 0.670 +(epoch: 27, iters: 4592, time: 0.064) G_GAN: 0.432 G_GAN_Feat: 0.825 G_ID: 0.733 G_Rec: 0.423 D_GP: 0.029 D_real: 1.032 D_fake: 0.569 +(epoch: 27, iters: 4992, time: 0.064) G_GAN: -0.152 G_GAN_Feat: 0.825 G_ID: 0.391 G_Rec: 0.472 D_GP: 0.088 D_real: 0.591 D_fake: 1.152 +(epoch: 27, iters: 5392, time: 0.064) G_GAN: 0.451 G_GAN_Feat: 0.644 G_ID: 0.801 G_Rec: 0.363 D_GP: 0.028 D_real: 1.231 D_fake: 0.549 +(epoch: 27, iters: 5792, time: 0.064) G_GAN: 0.048 G_GAN_Feat: 0.734 G_ID: 0.353 G_Rec: 0.391 D_GP: 0.039 D_real: 0.727 D_fake: 0.953 +(epoch: 27, iters: 6192, time: 0.064) G_GAN: 0.153 G_GAN_Feat: 0.696 G_ID: 0.752 G_Rec: 0.410 D_GP: 0.023 D_real: 0.866 D_fake: 0.847 +(epoch: 27, iters: 6592, time: 0.064) G_GAN: 0.553 G_GAN_Feat: 0.787 G_ID: 0.360 G_Rec: 0.405 D_GP: 0.045 D_real: 1.326 D_fake: 0.451 +(epoch: 27, iters: 6992, time: 0.064) G_GAN: 0.624 G_GAN_Feat: 0.634 G_ID: 0.790 G_Rec: 0.447 D_GP: 0.020 D_real: 1.534 D_fake: 0.425 +(epoch: 27, iters: 7392, time: 0.064) G_GAN: 0.279 G_GAN_Feat: 1.169 G_ID: 0.390 G_Rec: 0.394 D_GP: 0.103 D_real: 0.490 D_fake: 0.721 +(epoch: 27, iters: 7792, time: 0.064) G_GAN: 0.497 G_GAN_Feat: 0.840 G_ID: 0.699 G_Rec: 0.383 D_GP: 0.049 D_real: 1.191 D_fake: 0.529 +(epoch: 27, iters: 8192, time: 0.064) G_GAN: 0.733 G_GAN_Feat: 0.737 G_ID: 0.330 G_Rec: 0.343 D_GP: 0.042 D_real: 1.375 D_fake: 0.275 +(epoch: 27, iters: 8592, time: 0.064) G_GAN: 0.425 G_GAN_Feat: 0.871 G_ID: 0.764 G_Rec: 0.353 D_GP: 0.057 D_real: 1.072 D_fake: 0.576 +(epoch: 28, iters: 384, time: 0.064) G_GAN: 0.138 G_GAN_Feat: 0.629 G_ID: 0.309 G_Rec: 0.358 D_GP: 0.035 D_real: 0.997 D_fake: 0.862 +(epoch: 28, iters: 784, time: 0.064) G_GAN: 0.412 G_GAN_Feat: 0.719 G_ID: 0.798 G_Rec: 0.372 D_GP: 0.063 D_real: 1.086 D_fake: 0.590 +(epoch: 28, iters: 1184, time: 0.064) G_GAN: 0.313 G_GAN_Feat: 1.014 G_ID: 0.352 G_Rec: 0.432 D_GP: 0.261 D_real: 0.895 D_fake: 0.701 +(epoch: 28, iters: 1584, time: 0.064) G_GAN: 0.300 G_GAN_Feat: 0.924 G_ID: 0.712 G_Rec: 0.426 D_GP: 0.076 D_real: 1.048 D_fake: 0.711 +(epoch: 28, iters: 1984, time: 0.064) G_GAN: 0.436 G_GAN_Feat: 0.817 G_ID: 0.299 G_Rec: 0.425 D_GP: 0.058 D_real: 1.172 D_fake: 0.564 +(epoch: 28, iters: 2384, time: 0.064) G_GAN: -0.048 G_GAN_Feat: 0.914 G_ID: 0.763 G_Rec: 0.392 D_GP: 0.155 D_real: 0.513 D_fake: 1.049 +(epoch: 28, iters: 2784, time: 0.064) G_GAN: 0.676 G_GAN_Feat: 1.238 G_ID: 0.367 G_Rec: 0.409 D_GP: 0.060 D_real: 1.449 D_fake: 0.447 +(epoch: 28, iters: 3184, time: 0.064) G_GAN: 0.513 G_GAN_Feat: 0.903 G_ID: 0.699 G_Rec: 0.358 D_GP: 0.121 D_real: 1.125 D_fake: 0.488 +(epoch: 28, iters: 3584, time: 0.064) G_GAN: 0.347 G_GAN_Feat: 0.568 G_ID: 0.361 G_Rec: 0.358 D_GP: 0.025 D_real: 1.173 D_fake: 0.653 +(epoch: 28, iters: 3984, time: 0.064) G_GAN: 0.389 G_GAN_Feat: 0.729 G_ID: 0.725 G_Rec: 0.356 D_GP: 0.034 D_real: 1.168 D_fake: 0.612 +(epoch: 28, iters: 4384, time: 0.064) G_GAN: 0.189 G_GAN_Feat: 0.567 G_ID: 0.344 G_Rec: 0.415 D_GP: 0.030 D_real: 0.979 D_fake: 0.812 +(epoch: 28, iters: 4784, time: 0.064) G_GAN: -0.116 G_GAN_Feat: 0.717 G_ID: 0.715 G_Rec: 0.452 D_GP: 0.048 D_real: 0.472 D_fake: 1.116 +(epoch: 28, iters: 5184, time: 0.064) G_GAN: 0.117 G_GAN_Feat: 0.584 G_ID: 0.345 G_Rec: 0.482 D_GP: 0.031 D_real: 0.906 D_fake: 0.883 +(epoch: 28, iters: 5584, time: 0.064) G_GAN: 0.123 G_GAN_Feat: 0.824 G_ID: 0.717 G_Rec: 0.357 D_GP: 0.047 D_real: 0.630 D_fake: 0.877 +(epoch: 28, iters: 5984, time: 0.064) G_GAN: 0.563 G_GAN_Feat: 0.769 G_ID: 0.383 G_Rec: 0.373 D_GP: 0.045 D_real: 1.135 D_fake: 0.477 +(epoch: 28, iters: 6384, time: 0.064) G_GAN: 0.220 G_GAN_Feat: 0.699 G_ID: 0.711 G_Rec: 0.399 D_GP: 0.033 D_real: 1.115 D_fake: 0.780 +(epoch: 28, iters: 6784, time: 0.064) G_GAN: 0.527 G_GAN_Feat: 0.680 G_ID: 0.352 G_Rec: 0.380 D_GP: 0.022 D_real: 1.307 D_fake: 0.478 +(epoch: 28, iters: 7184, time: 0.064) G_GAN: 0.238 G_GAN_Feat: 0.570 G_ID: 0.675 G_Rec: 0.356 D_GP: 0.028 D_real: 1.028 D_fake: 0.762 +(epoch: 28, iters: 7584, time: 0.064) G_GAN: 0.464 G_GAN_Feat: 0.589 G_ID: 0.358 G_Rec: 0.354 D_GP: 0.037 D_real: 1.243 D_fake: 0.536 +(epoch: 28, iters: 7984, time: 0.064) G_GAN: 0.079 G_GAN_Feat: 1.147 G_ID: 0.688 G_Rec: 0.435 D_GP: 0.548 D_real: 0.331 D_fake: 0.921 +(epoch: 28, iters: 8384, time: 0.064) G_GAN: 0.023 G_GAN_Feat: 0.887 G_ID: 0.352 G_Rec: 0.377 D_GP: 0.067 D_real: 0.457 D_fake: 0.977 +(epoch: 29, iters: 176, time: 0.064) G_GAN: 0.685 G_GAN_Feat: 0.642 G_ID: 0.733 G_Rec: 0.383 D_GP: 0.035 D_real: 1.471 D_fake: 0.327 +(epoch: 29, iters: 576, time: 0.064) G_GAN: 0.187 G_GAN_Feat: 0.594 G_ID: 0.248 G_Rec: 0.327 D_GP: 0.029 D_real: 0.915 D_fake: 0.813 +(epoch: 29, iters: 976, time: 0.064) G_GAN: 0.304 G_GAN_Feat: 0.781 G_ID: 0.787 G_Rec: 0.349 D_GP: 0.095 D_real: 0.862 D_fake: 0.696 +(epoch: 29, iters: 1376, time: 0.064) G_GAN: 0.505 G_GAN_Feat: 0.708 G_ID: 0.335 G_Rec: 0.440 D_GP: 0.040 D_real: 1.152 D_fake: 0.496 +(epoch: 29, iters: 1776, time: 0.064) G_GAN: 0.287 G_GAN_Feat: 1.163 G_ID: 0.743 G_Rec: 0.426 D_GP: 0.072 D_real: 0.345 D_fake: 0.714 +(epoch: 29, iters: 2176, time: 0.064) G_GAN: 0.853 G_GAN_Feat: 0.822 G_ID: 0.304 G_Rec: 0.378 D_GP: 0.047 D_real: 1.446 D_fake: 0.482 +(epoch: 29, iters: 2576, time: 0.064) G_GAN: 0.634 G_GAN_Feat: 0.781 G_ID: 0.711 G_Rec: 0.429 D_GP: 0.046 D_real: 1.331 D_fake: 0.496 +(epoch: 29, iters: 2976, time: 0.064) G_GAN: 0.088 G_GAN_Feat: 0.692 G_ID: 0.333 G_Rec: 0.360 D_GP: 0.029 D_real: 0.825 D_fake: 0.912 +(epoch: 29, iters: 3376, time: 0.064) G_GAN: 0.345 G_GAN_Feat: 0.715 G_ID: 0.722 G_Rec: 0.463 D_GP: 0.041 D_real: 0.977 D_fake: 0.655 +(epoch: 29, iters: 3776, time: 0.064) G_GAN: 0.433 G_GAN_Feat: 0.794 G_ID: 0.360 G_Rec: 0.395 D_GP: 0.051 D_real: 0.942 D_fake: 0.568 +(epoch: 29, iters: 4176, time: 0.064) G_GAN: 0.396 G_GAN_Feat: 0.706 G_ID: 0.702 G_Rec: 0.391 D_GP: 0.042 D_real: 1.098 D_fake: 0.604 +(epoch: 29, iters: 4576, time: 0.064) G_GAN: 0.235 G_GAN_Feat: 0.802 G_ID: 0.354 G_Rec: 0.360 D_GP: 0.049 D_real: 1.097 D_fake: 0.766 +(epoch: 29, iters: 4976, time: 0.064) G_GAN: 0.226 G_GAN_Feat: 0.840 G_ID: 0.726 G_Rec: 0.382 D_GP: 0.041 D_real: 0.831 D_fake: 0.774 +(epoch: 29, iters: 5376, time: 0.064) G_GAN: 0.089 G_GAN_Feat: 0.757 G_ID: 0.326 G_Rec: 0.381 D_GP: 0.071 D_real: 0.664 D_fake: 0.911 +(epoch: 29, iters: 5776, time: 0.064) G_GAN: -0.004 G_GAN_Feat: 0.820 G_ID: 0.686 G_Rec: 0.381 D_GP: 0.129 D_real: 0.469 D_fake: 1.004 +(epoch: 29, iters: 6176, time: 0.064) G_GAN: 0.366 G_GAN_Feat: 0.538 G_ID: 0.316 G_Rec: 0.348 D_GP: 0.019 D_real: 1.269 D_fake: 0.635 +(epoch: 29, iters: 6576, time: 0.064) G_GAN: 0.148 G_GAN_Feat: 0.602 G_ID: 0.669 G_Rec: 0.358 D_GP: 0.019 D_real: 0.991 D_fake: 0.853 +(epoch: 29, iters: 6976, time: 0.064) G_GAN: 0.147 G_GAN_Feat: 0.734 G_ID: 0.406 G_Rec: 0.472 D_GP: 0.040 D_real: 0.756 D_fake: 0.853 +(epoch: 29, iters: 7376, time: 0.064) G_GAN: 0.781 G_GAN_Feat: 1.135 G_ID: 0.689 G_Rec: 0.478 D_GP: 0.079 D_real: 1.044 D_fake: 0.379 +(epoch: 29, iters: 7776, time: 0.064) G_GAN: 0.184 G_GAN_Feat: 0.805 G_ID: 0.359 G_Rec: 0.360 D_GP: 0.081 D_real: 0.721 D_fake: 0.818 +(epoch: 29, iters: 8176, time: 0.064) G_GAN: 0.117 G_GAN_Feat: 1.066 G_ID: 0.779 G_Rec: 0.365 D_GP: 0.194 D_real: 0.518 D_fake: 0.885 +(epoch: 29, iters: 8576, time: 0.064) G_GAN: 0.472 G_GAN_Feat: 1.011 G_ID: 0.362 G_Rec: 0.422 D_GP: 0.076 D_real: 0.634 D_fake: 0.529 +(epoch: 30, iters: 368, time: 0.064) G_GAN: 0.341 G_GAN_Feat: 1.532 G_ID: 0.704 G_Rec: 0.360 D_GP: 0.287 D_real: 0.491 D_fake: 0.660 +(epoch: 30, iters: 768, time: 0.064) G_GAN: 0.806 G_GAN_Feat: 0.725 G_ID: 0.338 G_Rec: 0.384 D_GP: 0.034 D_real: 1.540 D_fake: 0.427 +(epoch: 30, iters: 1168, time: 0.064) G_GAN: 0.187 G_GAN_Feat: 1.095 G_ID: 0.739 G_Rec: 0.541 D_GP: 0.156 D_real: 0.422 D_fake: 0.813 +(epoch: 30, iters: 1568, time: 0.064) G_GAN: 0.470 G_GAN_Feat: 0.681 G_ID: 0.327 G_Rec: 0.379 D_GP: 0.031 D_real: 1.217 D_fake: 0.535 +(epoch: 30, iters: 1968, time: 0.064) G_GAN: 0.518 G_GAN_Feat: 0.820 G_ID: 0.696 G_Rec: 0.374 D_GP: 0.086 D_real: 1.064 D_fake: 0.495 +(epoch: 30, iters: 2368, time: 0.064) G_GAN: 0.180 G_GAN_Feat: 0.683 G_ID: 0.351 G_Rec: 0.378 D_GP: 0.030 D_real: 0.996 D_fake: 0.820 +(epoch: 30, iters: 2768, time: 0.064) G_GAN: 0.377 G_GAN_Feat: 0.737 G_ID: 0.755 G_Rec: 0.388 D_GP: 0.027 D_real: 1.113 D_fake: 0.623 +(epoch: 30, iters: 3168, time: 0.064) G_GAN: 0.222 G_GAN_Feat: 0.686 G_ID: 0.335 G_Rec: 0.418 D_GP: 0.051 D_real: 0.879 D_fake: 0.779 +(epoch: 30, iters: 3568, time: 0.064) G_GAN: 0.401 G_GAN_Feat: 0.595 G_ID: 0.704 G_Rec: 0.368 D_GP: 0.024 D_real: 1.221 D_fake: 0.600 +(epoch: 30, iters: 3968, time: 0.064) G_GAN: 0.426 G_GAN_Feat: 1.174 G_ID: 0.318 G_Rec: 0.448 D_GP: 0.237 D_real: 0.539 D_fake: 0.577 +(epoch: 30, iters: 4368, time: 0.064) G_GAN: -0.119 G_GAN_Feat: 0.937 G_ID: 0.725 G_Rec: 0.371 D_GP: 0.158 D_real: 0.546 D_fake: 1.119 +(epoch: 30, iters: 4768, time: 0.064) G_GAN: -0.039 G_GAN_Feat: 0.692 G_ID: 0.354 G_Rec: 0.490 D_GP: 0.032 D_real: 0.683 D_fake: 1.039 +(epoch: 30, iters: 5168, time: 0.064) G_GAN: 0.184 G_GAN_Feat: 0.686 G_ID: 0.679 G_Rec: 0.451 D_GP: 0.027 D_real: 1.020 D_fake: 0.816 +(epoch: 30, iters: 5568, time: 0.064) G_GAN: 0.353 G_GAN_Feat: 0.666 G_ID: 0.338 G_Rec: 0.433 D_GP: 0.032 D_real: 1.109 D_fake: 0.647 +(epoch: 30, iters: 5968, time: 0.064) G_GAN: 0.471 G_GAN_Feat: 0.716 G_ID: 0.661 G_Rec: 0.378 D_GP: 0.035 D_real: 1.195 D_fake: 0.532 +(epoch: 30, iters: 6368, time: 0.064) G_GAN: 0.282 G_GAN_Feat: 0.682 G_ID: 0.310 G_Rec: 0.354 D_GP: 0.048 D_real: 0.934 D_fake: 0.718 +(epoch: 30, iters: 6768, time: 0.064) G_GAN: 0.574 G_GAN_Feat: 0.973 G_ID: 0.666 G_Rec: 0.403 D_GP: 0.173 D_real: 0.867 D_fake: 0.437 +(epoch: 30, iters: 7168, time: 0.064) G_GAN: 0.235 G_GAN_Feat: 0.856 G_ID: 0.272 G_Rec: 0.436 D_GP: 0.048 D_real: 0.773 D_fake: 0.765 +(epoch: 30, iters: 7568, time: 0.064) G_GAN: 0.302 G_GAN_Feat: 0.558 G_ID: 0.697 G_Rec: 0.351 D_GP: 0.018 D_real: 1.201 D_fake: 0.698 +(epoch: 30, iters: 7968, time: 0.064) G_GAN: 0.526 G_GAN_Feat: 0.598 G_ID: 0.332 G_Rec: 0.384 D_GP: 0.028 D_real: 1.389 D_fake: 0.477 +(epoch: 30, iters: 8368, time: 0.064) G_GAN: 0.413 G_GAN_Feat: 0.774 G_ID: 0.694 G_Rec: 0.366 D_GP: 0.055 D_real: 1.054 D_fake: 0.588 +(epoch: 31, iters: 160, time: 0.064) G_GAN: 0.491 G_GAN_Feat: 0.841 G_ID: 0.321 G_Rec: 0.387 D_GP: 0.054 D_real: 1.006 D_fake: 0.509 +(epoch: 31, iters: 560, time: 0.064) G_GAN: 0.558 G_GAN_Feat: 1.008 G_ID: 0.713 G_Rec: 0.392 D_GP: 0.115 D_real: 0.739 D_fake: 0.450 +(epoch: 31, iters: 960, time: 0.064) G_GAN: 0.561 G_GAN_Feat: 0.823 G_ID: 0.360 G_Rec: 0.393 D_GP: 0.044 D_real: 1.211 D_fake: 0.442 +(epoch: 31, iters: 1360, time: 0.064) G_GAN: 0.647 G_GAN_Feat: 0.861 G_ID: 0.677 G_Rec: 0.362 D_GP: 0.045 D_real: 1.428 D_fake: 0.470 +(epoch: 31, iters: 1760, time: 0.064) G_GAN: 0.055 G_GAN_Feat: 0.854 G_ID: 0.371 G_Rec: 0.356 D_GP: 0.083 D_real: 0.605 D_fake: 0.945 +(epoch: 31, iters: 2160, time: 0.064) G_GAN: 0.544 G_GAN_Feat: 0.522 G_ID: 0.709 G_Rec: 0.391 D_GP: 0.018 D_real: 1.444 D_fake: 0.487 +(epoch: 31, iters: 2560, time: 0.064) G_GAN: 0.260 G_GAN_Feat: 0.542 G_ID: 0.340 G_Rec: 0.374 D_GP: 0.022 D_real: 1.078 D_fake: 0.740 +(epoch: 31, iters: 2960, time: 0.064) G_GAN: 0.208 G_GAN_Feat: 0.591 G_ID: 0.651 G_Rec: 0.326 D_GP: 0.034 D_real: 1.009 D_fake: 0.792 +(epoch: 31, iters: 3360, time: 0.064) G_GAN: 0.340 G_GAN_Feat: 0.731 G_ID: 0.350 G_Rec: 0.381 D_GP: 0.092 D_real: 0.909 D_fake: 0.661 +(epoch: 31, iters: 3760, time: 0.064) G_GAN: 0.145 G_GAN_Feat: 0.742 G_ID: 0.713 G_Rec: 0.405 D_GP: 0.054 D_real: 0.903 D_fake: 0.855 +(epoch: 31, iters: 4160, time: 0.064) G_GAN: 0.527 G_GAN_Feat: 0.675 G_ID: 0.301 G_Rec: 0.402 D_GP: 0.036 D_real: 1.278 D_fake: 0.475 +(epoch: 31, iters: 4560, time: 0.064) G_GAN: 0.336 G_GAN_Feat: 0.740 G_ID: 0.648 G_Rec: 0.396 D_GP: 0.034 D_real: 1.047 D_fake: 0.664 +(epoch: 31, iters: 4960, time: 0.064) G_GAN: 0.448 G_GAN_Feat: 0.807 G_ID: 0.353 G_Rec: 0.360 D_GP: 0.064 D_real: 0.996 D_fake: 0.553 +(epoch: 31, iters: 5360, time: 0.064) G_GAN: 0.539 G_GAN_Feat: 0.905 G_ID: 0.668 G_Rec: 0.375 D_GP: 0.108 D_real: 1.019 D_fake: 0.467 +(epoch: 31, iters: 5760, time: 0.064) G_GAN: 0.188 G_GAN_Feat: 0.745 G_ID: 0.322 G_Rec: 0.325 D_GP: 0.063 D_real: 0.905 D_fake: 0.812 +(epoch: 31, iters: 6160, time: 0.064) G_GAN: 0.688 G_GAN_Feat: 0.725 G_ID: 0.736 G_Rec: 0.382 D_GP: 0.031 D_real: 1.440 D_fake: 0.395 +(epoch: 31, iters: 6560, time: 0.064) G_GAN: 0.544 G_GAN_Feat: 0.766 G_ID: 0.296 G_Rec: 0.371 D_GP: 0.073 D_real: 1.034 D_fake: 0.461 +(epoch: 31, iters: 6960, time: 0.064) G_GAN: 1.075 G_GAN_Feat: 0.714 G_ID: 0.665 G_Rec: 0.389 D_GP: 0.023 D_real: 1.808 D_fake: 0.085 +(epoch: 31, iters: 7360, time: 0.064) G_GAN: 0.538 G_GAN_Feat: 0.658 G_ID: 0.356 G_Rec: 0.449 D_GP: 0.030 D_real: 1.357 D_fake: 0.465 +(epoch: 31, iters: 7760, time: 0.064) G_GAN: 0.702 G_GAN_Feat: 0.788 G_ID: 0.661 G_Rec: 0.396 D_GP: 0.056 D_real: 1.351 D_fake: 0.336 +(epoch: 31, iters: 8160, time: 0.064) G_GAN: 0.345 G_GAN_Feat: 0.688 G_ID: 0.359 G_Rec: 0.359 D_GP: 0.037 D_real: 1.049 D_fake: 0.655 +(epoch: 31, iters: 8560, time: 0.064) G_GAN: 0.360 G_GAN_Feat: 0.635 G_ID: 0.672 G_Rec: 0.390 D_GP: 0.029 D_real: 1.207 D_fake: 0.643 +(epoch: 32, iters: 352, time: 0.064) G_GAN: 0.425 G_GAN_Feat: 0.698 G_ID: 0.342 G_Rec: 0.365 D_GP: 0.034 D_real: 1.164 D_fake: 0.576 +(epoch: 32, iters: 752, time: 0.064) G_GAN: 0.129 G_GAN_Feat: 1.048 G_ID: 0.693 G_Rec: 0.486 D_GP: 0.117 D_real: 0.361 D_fake: 0.871 +(epoch: 32, iters: 1152, time: 0.064) G_GAN: 0.379 G_GAN_Feat: 0.628 G_ID: 0.389 G_Rec: 0.408 D_GP: 0.019 D_real: 1.277 D_fake: 0.636 +(epoch: 32, iters: 1552, time: 0.064) G_GAN: 0.331 G_GAN_Feat: 0.567 G_ID: 0.683 G_Rec: 0.360 D_GP: 0.017 D_real: 1.183 D_fake: 0.669 +(epoch: 32, iters: 1952, time: 0.064) G_GAN: 0.343 G_GAN_Feat: 0.548 G_ID: 0.310 G_Rec: 0.394 D_GP: 0.022 D_real: 1.381 D_fake: 0.658 +(epoch: 32, iters: 2352, time: 0.064) G_GAN: 0.282 G_GAN_Feat: 0.603 G_ID: 0.619 G_Rec: 0.369 D_GP: 0.029 D_real: 1.087 D_fake: 0.730 +(epoch: 32, iters: 2752, time: 0.064) G_GAN: 0.270 G_GAN_Feat: 0.594 G_ID: 0.399 G_Rec: 0.366 D_GP: 0.029 D_real: 1.116 D_fake: 0.730 +(epoch: 32, iters: 3152, time: 0.064) G_GAN: 0.072 G_GAN_Feat: 0.683 G_ID: 0.733 G_Rec: 0.355 D_GP: 0.058 D_real: 0.818 D_fake: 0.928 +(epoch: 32, iters: 3552, time: 0.064) G_GAN: 0.290 G_GAN_Feat: 0.753 G_ID: 0.311 G_Rec: 0.358 D_GP: 0.067 D_real: 1.017 D_fake: 0.712 +(epoch: 32, iters: 3952, time: 0.064) G_GAN: 0.339 G_GAN_Feat: 0.981 G_ID: 0.685 G_Rec: 0.395 D_GP: 0.065 D_real: 0.816 D_fake: 0.662 +(epoch: 32, iters: 4352, time: 0.064) G_GAN: 0.197 G_GAN_Feat: 0.738 G_ID: 0.331 G_Rec: 0.341 D_GP: 0.037 D_real: 0.831 D_fake: 0.803 +(epoch: 32, iters: 4752, time: 0.064) G_GAN: 0.553 G_GAN_Feat: 0.789 G_ID: 0.657 G_Rec: 0.479 D_GP: 0.061 D_real: 1.165 D_fake: 0.478 +(epoch: 32, iters: 5152, time: 0.064) G_GAN: 0.458 G_GAN_Feat: 0.731 G_ID: 0.352 G_Rec: 0.409 D_GP: 0.034 D_real: 1.105 D_fake: 0.543 +(epoch: 32, iters: 5552, time: 0.064) G_GAN: 0.514 G_GAN_Feat: 0.903 G_ID: 0.634 G_Rec: 0.365 D_GP: 0.083 D_real: 0.911 D_fake: 0.489 +(epoch: 32, iters: 5952, time: 0.064) G_GAN: 0.209 G_GAN_Feat: 0.618 G_ID: 0.278 G_Rec: 0.437 D_GP: 0.022 D_real: 1.033 D_fake: 0.791 +(epoch: 32, iters: 6352, time: 0.064) G_GAN: 0.189 G_GAN_Feat: 0.660 G_ID: 0.642 G_Rec: 0.411 D_GP: 0.033 D_real: 1.046 D_fake: 0.811 +(epoch: 32, iters: 6752, time: 0.064) G_GAN: 0.330 G_GAN_Feat: 0.732 G_ID: 0.330 G_Rec: 0.455 D_GP: 0.049 D_real: 1.001 D_fake: 0.672 +(epoch: 32, iters: 7152, time: 0.064) G_GAN: 0.314 G_GAN_Feat: 0.775 G_ID: 0.679 G_Rec: 0.335 D_GP: 0.043 D_real: 0.975 D_fake: 0.686 +(epoch: 32, iters: 7552, time: 0.064) G_GAN: 0.268 G_GAN_Feat: 0.847 G_ID: 0.346 G_Rec: 0.432 D_GP: 0.042 D_real: 0.750 D_fake: 0.733 +(epoch: 32, iters: 7952, time: 0.064) G_GAN: 0.437 G_GAN_Feat: 0.605 G_ID: 0.640 G_Rec: 0.325 D_GP: 0.028 D_real: 1.266 D_fake: 0.565 +(epoch: 32, iters: 8352, time: 0.064) G_GAN: 0.425 G_GAN_Feat: 0.870 G_ID: 0.328 G_Rec: 0.363 D_GP: 0.099 D_real: 0.898 D_fake: 0.577 +(epoch: 33, iters: 144, time: 0.064) G_GAN: -0.015 G_GAN_Feat: 0.844 G_ID: 0.682 G_Rec: 0.354 D_GP: 0.160 D_real: 0.530 D_fake: 1.015 +(epoch: 33, iters: 544, time: 0.064) G_GAN: 0.176 G_GAN_Feat: 0.924 G_ID: 0.323 G_Rec: 0.341 D_GP: 0.058 D_real: 0.568 D_fake: 0.824 +(epoch: 33, iters: 944, time: 0.064) G_GAN: 0.355 G_GAN_Feat: 1.024 G_ID: 0.693 G_Rec: 0.470 D_GP: 0.136 D_real: 0.581 D_fake: 0.646 +(epoch: 33, iters: 1344, time: 0.064) G_GAN: 0.759 G_GAN_Feat: 1.042 G_ID: 0.324 G_Rec: 0.391 D_GP: 0.075 D_real: 0.951 D_fake: 0.285 +(epoch: 33, iters: 1744, time: 0.064) G_GAN: 0.307 G_GAN_Feat: 0.489 G_ID: 0.650 G_Rec: 0.362 D_GP: 0.017 D_real: 1.209 D_fake: 0.693 +(epoch: 33, iters: 2144, time: 0.064) G_GAN: 0.256 G_GAN_Feat: 0.494 G_ID: 0.303 G_Rec: 0.326 D_GP: 0.020 D_real: 1.077 D_fake: 0.744 +(epoch: 33, iters: 2544, time: 0.064) G_GAN: 0.152 G_GAN_Feat: 0.546 G_ID: 0.703 G_Rec: 0.398 D_GP: 0.022 D_real: 0.943 D_fake: 0.848 +(epoch: 33, iters: 2944, time: 0.064) G_GAN: 0.421 G_GAN_Feat: 0.560 G_ID: 0.311 G_Rec: 0.446 D_GP: 0.027 D_real: 1.281 D_fake: 0.594 +(epoch: 33, iters: 3344, time: 0.064) G_GAN: 0.121 G_GAN_Feat: 0.570 G_ID: 0.644 G_Rec: 0.410 D_GP: 0.029 D_real: 0.859 D_fake: 0.879 +(epoch: 33, iters: 3744, time: 0.064) G_GAN: 0.257 G_GAN_Feat: 0.648 G_ID: 0.361 G_Rec: 0.377 D_GP: 0.039 D_real: 0.953 D_fake: 0.745 +(epoch: 33, iters: 4144, time: 0.064) G_GAN: 0.024 G_GAN_Feat: 0.654 G_ID: 0.683 G_Rec: 0.393 D_GP: 0.055 D_real: 0.723 D_fake: 0.976 +(epoch: 33, iters: 4544, time: 0.064) G_GAN: 0.386 G_GAN_Feat: 0.960 G_ID: 0.382 G_Rec: 0.382 D_GP: 0.120 D_real: 0.701 D_fake: 0.616 +(epoch: 33, iters: 4944, time: 0.064) G_GAN: 0.427 G_GAN_Feat: 0.800 G_ID: 0.605 G_Rec: 0.370 D_GP: 0.051 D_real: 1.115 D_fake: 0.574 +(epoch: 33, iters: 5344, time: 0.064) G_GAN: 0.728 G_GAN_Feat: 0.789 G_ID: 0.301 G_Rec: 0.416 D_GP: 0.061 D_real: 1.268 D_fake: 0.444 +(epoch: 33, iters: 5744, time: 0.064) G_GAN: 0.385 G_GAN_Feat: 0.572 G_ID: 0.648 G_Rec: 0.353 D_GP: 0.020 D_real: 1.289 D_fake: 0.615 +(epoch: 33, iters: 6144, time: 0.064) G_GAN: 0.356 G_GAN_Feat: 0.643 G_ID: 0.319 G_Rec: 0.373 D_GP: 0.032 D_real: 1.178 D_fake: 0.650 +(epoch: 33, iters: 6544, time: 0.064) G_GAN: 0.337 G_GAN_Feat: 0.816 G_ID: 0.639 G_Rec: 0.376 D_GP: 0.049 D_real: 1.008 D_fake: 0.668 +(epoch: 33, iters: 6944, time: 0.064) G_GAN: 0.221 G_GAN_Feat: 0.872 G_ID: 0.331 G_Rec: 0.505 D_GP: 0.055 D_real: 0.700 D_fake: 0.780 +(epoch: 33, iters: 7344, time: 0.064) G_GAN: 0.279 G_GAN_Feat: 0.657 G_ID: 0.664 G_Rec: 0.405 D_GP: 0.038 D_real: 0.973 D_fake: 0.721 +(epoch: 33, iters: 7744, time: 0.064) G_GAN: 0.422 G_GAN_Feat: 0.729 G_ID: 0.270 G_Rec: 0.429 D_GP: 0.056 D_real: 1.037 D_fake: 0.579 +(epoch: 33, iters: 8144, time: 0.064) G_GAN: 0.291 G_GAN_Feat: 0.684 G_ID: 0.704 G_Rec: 0.373 D_GP: 0.041 D_real: 1.096 D_fake: 0.709 +(epoch: 33, iters: 8544, time: 0.064) G_GAN: 0.150 G_GAN_Feat: 0.675 G_ID: 0.367 G_Rec: 0.364 D_GP: 0.019 D_real: 0.964 D_fake: 0.850 +(epoch: 34, iters: 336, time: 0.064) G_GAN: 0.176 G_GAN_Feat: 0.647 G_ID: 0.618 G_Rec: 0.367 D_GP: 0.034 D_real: 0.807 D_fake: 0.825 +(epoch: 34, iters: 736, time: 0.064) G_GAN: 0.474 G_GAN_Feat: 0.866 G_ID: 0.294 G_Rec: 0.351 D_GP: 0.122 D_real: 0.740 D_fake: 0.541 +(epoch: 34, iters: 1136, time: 0.064) G_GAN: 0.118 G_GAN_Feat: 0.738 G_ID: 0.651 G_Rec: 0.448 D_GP: 0.028 D_real: 0.836 D_fake: 0.882 +(epoch: 34, iters: 1536, time: 0.064) G_GAN: 0.446 G_GAN_Feat: 0.753 G_ID: 0.346 G_Rec: 0.345 D_GP: 0.040 D_real: 1.145 D_fake: 0.555 +(epoch: 34, iters: 1936, time: 0.064) G_GAN: 0.655 G_GAN_Feat: 0.730 G_ID: 0.644 G_Rec: 0.369 D_GP: 0.036 D_real: 1.423 D_fake: 0.354 +(epoch: 34, iters: 2336, time: 0.064) G_GAN: 0.419 G_GAN_Feat: 0.994 G_ID: 0.330 G_Rec: 0.375 D_GP: 0.107 D_real: 1.108 D_fake: 0.584 +(epoch: 34, iters: 2736, time: 0.064) G_GAN: 0.180 G_GAN_Feat: 0.770 G_ID: 0.648 G_Rec: 0.359 D_GP: 0.058 D_real: 0.891 D_fake: 0.821 +(epoch: 34, iters: 3136, time: 0.064) G_GAN: 0.281 G_GAN_Feat: 0.726 G_ID: 0.290 G_Rec: 0.404 D_GP: 0.031 D_real: 1.027 D_fake: 0.719 +(epoch: 34, iters: 3536, time: 0.064) G_GAN: 0.667 G_GAN_Feat: 0.794 G_ID: 0.642 G_Rec: 0.351 D_GP: 0.046 D_real: 1.411 D_fake: 0.348 +(epoch: 34, iters: 3936, time: 0.064) G_GAN: -0.065 G_GAN_Feat: 0.723 G_ID: 0.332 G_Rec: 0.356 D_GP: 0.031 D_real: 0.641 D_fake: 1.065 +(epoch: 34, iters: 4336, time: 0.064) G_GAN: 0.374 G_GAN_Feat: 0.762 G_ID: 0.607 G_Rec: 0.348 D_GP: 0.047 D_real: 0.974 D_fake: 0.626 +(epoch: 34, iters: 4736, time: 0.064) G_GAN: 0.563 G_GAN_Feat: 1.106 G_ID: 0.295 G_Rec: 0.354 D_GP: 0.084 D_real: 0.690 D_fake: 0.449 +(epoch: 34, iters: 5136, time: 0.064) G_GAN: 0.100 G_GAN_Feat: 0.575 G_ID: 0.590 G_Rec: 0.353 D_GP: 0.030 D_real: 0.971 D_fake: 0.900 +(epoch: 34, iters: 5536, time: 0.064) G_GAN: 0.343 G_GAN_Feat: 0.716 G_ID: 0.358 G_Rec: 0.409 D_GP: 0.052 D_real: 1.086 D_fake: 0.659 +(epoch: 34, iters: 5936, time: 0.064) G_GAN: 0.341 G_GAN_Feat: 0.712 G_ID: 0.634 G_Rec: 0.389 D_GP: 0.056 D_real: 1.066 D_fake: 0.660 +(epoch: 34, iters: 6336, time: 0.064) G_GAN: 0.366 G_GAN_Feat: 0.783 G_ID: 0.320 G_Rec: 0.400 D_GP: 0.088 D_real: 0.850 D_fake: 0.636 +(epoch: 34, iters: 6736, time: 0.064) G_GAN: 0.053 G_GAN_Feat: 0.995 G_ID: 0.643 G_Rec: 0.486 D_GP: 0.086 D_real: 0.387 D_fake: 0.947 +(epoch: 34, iters: 7136, time: 0.064) G_GAN: 0.329 G_GAN_Feat: 0.784 G_ID: 0.325 G_Rec: 0.356 D_GP: 0.046 D_real: 0.974 D_fake: 0.672 +(epoch: 34, iters: 7536, time: 0.064) G_GAN: 0.777 G_GAN_Feat: 0.729 G_ID: 0.572 G_Rec: 0.393 D_GP: 0.036 D_real: 1.523 D_fake: 0.393 +(epoch: 34, iters: 7936, time: 0.064) G_GAN: 0.155 G_GAN_Feat: 0.541 G_ID: 0.275 G_Rec: 0.314 D_GP: 0.028 D_real: 1.003 D_fake: 0.846 +(epoch: 34, iters: 8336, time: 0.064) G_GAN: 0.092 G_GAN_Feat: 0.702 G_ID: 0.681 G_Rec: 0.415 D_GP: 0.032 D_real: 0.776 D_fake: 0.909 +(epoch: 35, iters: 128, time: 0.064) G_GAN: 0.305 G_GAN_Feat: 0.668 G_ID: 0.358 G_Rec: 0.500 D_GP: 0.059 D_real: 0.983 D_fake: 0.695 +(epoch: 35, iters: 528, time: 0.064) G_GAN: 0.537 G_GAN_Feat: 0.852 G_ID: 0.592 G_Rec: 0.420 D_GP: 0.091 D_real: 1.163 D_fake: 0.515 +(epoch: 35, iters: 928, time: 0.064) G_GAN: 0.438 G_GAN_Feat: 0.790 G_ID: 0.284 G_Rec: 0.419 D_GP: 0.066 D_real: 1.076 D_fake: 0.564 +(epoch: 35, iters: 1328, time: 0.064) G_GAN: 0.522 G_GAN_Feat: 0.569 G_ID: 0.653 G_Rec: 0.345 D_GP: 0.021 D_real: 1.390 D_fake: 0.479 +(epoch: 35, iters: 1728, time: 0.064) G_GAN: 0.310 G_GAN_Feat: 0.619 G_ID: 0.303 G_Rec: 0.452 D_GP: 0.035 D_real: 1.120 D_fake: 0.690 +(epoch: 35, iters: 2128, time: 0.064) G_GAN: 0.486 G_GAN_Feat: 0.624 G_ID: 0.609 G_Rec: 0.381 D_GP: 0.039 D_real: 1.149 D_fake: 0.515 +(epoch: 35, iters: 2528, time: 0.064) G_GAN: 0.254 G_GAN_Feat: 0.788 G_ID: 0.311 G_Rec: 0.474 D_GP: 0.087 D_real: 0.708 D_fake: 0.746 +(epoch: 35, iters: 2928, time: 0.064) G_GAN: 0.233 G_GAN_Feat: 0.746 G_ID: 0.612 G_Rec: 0.347 D_GP: 0.086 D_real: 0.875 D_fake: 0.767 +(epoch: 35, iters: 3328, time: 0.064) G_GAN: 0.218 G_GAN_Feat: 0.923 G_ID: 0.302 G_Rec: 0.352 D_GP: 0.072 D_real: 0.565 D_fake: 0.783 +(epoch: 35, iters: 3728, time: 0.064) G_GAN: 0.326 G_GAN_Feat: 0.855 G_ID: 0.623 G_Rec: 0.352 D_GP: 0.153 D_real: 0.774 D_fake: 0.674 +(epoch: 35, iters: 4128, time: 0.064) G_GAN: 0.703 G_GAN_Feat: 0.697 G_ID: 0.331 G_Rec: 0.452 D_GP: 0.029 D_real: 1.490 D_fake: 0.313 +(epoch: 35, iters: 4528, time: 0.064) G_GAN: 0.391 G_GAN_Feat: 1.183 G_ID: 0.644 G_Rec: 0.370 D_GP: 0.301 D_real: 0.466 D_fake: 0.613 +(epoch: 35, iters: 4928, time: 0.064) G_GAN: 0.458 G_GAN_Feat: 1.097 G_ID: 0.362 G_Rec: 0.430 D_GP: 0.219 D_real: 0.451 D_fake: 0.571 +(epoch: 35, iters: 5328, time: 0.064) G_GAN: 0.190 G_GAN_Feat: 0.564 G_ID: 0.686 G_Rec: 0.450 D_GP: 0.019 D_real: 1.137 D_fake: 0.810 +(epoch: 35, iters: 5728, time: 0.064) G_GAN: 0.175 G_GAN_Feat: 0.569 G_ID: 0.325 G_Rec: 0.356 D_GP: 0.027 D_real: 1.020 D_fake: 0.826 +(epoch: 35, iters: 6128, time: 0.064) G_GAN: 0.181 G_GAN_Feat: 0.516 G_ID: 0.636 G_Rec: 0.451 D_GP: 0.021 D_real: 1.043 D_fake: 0.819 +(epoch: 35, iters: 6528, time: 0.064) G_GAN: 0.249 G_GAN_Feat: 0.553 G_ID: 0.290 G_Rec: 0.353 D_GP: 0.024 D_real: 1.162 D_fake: 0.761 +(epoch: 35, iters: 6928, time: 0.064) G_GAN: 0.150 G_GAN_Feat: 0.571 G_ID: 0.641 G_Rec: 0.409 D_GP: 0.034 D_real: 0.961 D_fake: 0.850 +(epoch: 35, iters: 7328, time: 0.064) G_GAN: 0.066 G_GAN_Feat: 0.629 G_ID: 0.276 G_Rec: 0.348 D_GP: 0.059 D_real: 0.870 D_fake: 0.935 +(epoch: 35, iters: 7728, time: 0.064) G_GAN: 0.221 G_GAN_Feat: 0.649 G_ID: 0.606 G_Rec: 0.365 D_GP: 0.056 D_real: 0.973 D_fake: 0.782 +(epoch: 35, iters: 8128, time: 0.064) G_GAN: 0.245 G_GAN_Feat: 0.654 G_ID: 0.291 G_Rec: 0.430 D_GP: 0.036 D_real: 1.039 D_fake: 0.755 +(epoch: 35, iters: 8528, time: 0.064) G_GAN: 0.441 G_GAN_Feat: 0.732 G_ID: 0.649 G_Rec: 0.366 D_GP: 0.059 D_real: 1.098 D_fake: 0.561 +(epoch: 36, iters: 320, time: 0.064) G_GAN: 0.103 G_GAN_Feat: 0.891 G_ID: 0.360 G_Rec: 0.368 D_GP: 0.100 D_real: 0.506 D_fake: 0.897 +(epoch: 36, iters: 720, time: 0.064) G_GAN: 0.456 G_GAN_Feat: 0.705 G_ID: 0.603 G_Rec: 0.357 D_GP: 0.074 D_real: 1.050 D_fake: 0.546 +(epoch: 36, iters: 1120, time: 0.064) G_GAN: 0.582 G_GAN_Feat: 0.710 G_ID: 0.325 G_Rec: 0.333 D_GP: 0.060 D_real: 1.233 D_fake: 0.450 +(epoch: 36, iters: 1520, time: 0.064) G_GAN: 0.278 G_GAN_Feat: 0.719 G_ID: 0.609 G_Rec: 0.406 D_GP: 0.034 D_real: 1.041 D_fake: 0.722 +(epoch: 36, iters: 1920, time: 0.064) G_GAN: 0.406 G_GAN_Feat: 0.827 G_ID: 0.298 G_Rec: 0.385 D_GP: 0.093 D_real: 0.938 D_fake: 0.631 +(epoch: 36, iters: 2320, time: 0.064) G_GAN: 0.406 G_GAN_Feat: 0.651 G_ID: 0.568 G_Rec: 0.370 D_GP: 0.030 D_real: 1.197 D_fake: 0.598 +(epoch: 36, iters: 2720, time: 0.064) G_GAN: 0.193 G_GAN_Feat: 0.796 G_ID: 0.328 G_Rec: 0.374 D_GP: 0.103 D_real: 0.712 D_fake: 0.808 +(epoch: 36, iters: 3120, time: 0.064) G_GAN: 0.321 G_GAN_Feat: 0.781 G_ID: 0.599 G_Rec: 0.390 D_GP: 0.071 D_real: 0.805 D_fake: 0.679 +(epoch: 36, iters: 3520, time: 0.064) G_GAN: 0.227 G_GAN_Feat: 0.897 G_ID: 0.242 G_Rec: 0.400 D_GP: 0.110 D_real: 0.473 D_fake: 0.773 +(epoch: 36, iters: 3920, time: 0.064) G_GAN: 0.125 G_GAN_Feat: 0.689 G_ID: 0.664 G_Rec: 0.361 D_GP: 0.042 D_real: 0.819 D_fake: 0.875 +(epoch: 36, iters: 4320, time: 0.064) G_GAN: 0.202 G_GAN_Feat: 0.652 G_ID: 0.292 G_Rec: 0.409 D_GP: 0.032 D_real: 0.955 D_fake: 0.798 +(epoch: 36, iters: 4720, time: 0.064) G_GAN: 0.404 G_GAN_Feat: 0.810 G_ID: 0.636 G_Rec: 0.328 D_GP: 0.056 D_real: 1.225 D_fake: 0.596 +(epoch: 36, iters: 5120, time: 0.064) G_GAN: 0.508 G_GAN_Feat: 0.722 G_ID: 0.312 G_Rec: 0.338 D_GP: 0.042 D_real: 1.134 D_fake: 0.499 +(epoch: 36, iters: 5520, time: 0.064) G_GAN: 0.326 G_GAN_Feat: 0.796 G_ID: 0.619 G_Rec: 0.342 D_GP: 0.049 D_real: 0.950 D_fake: 0.678 +(epoch: 36, iters: 5920, time: 0.064) G_GAN: 0.215 G_GAN_Feat: 0.764 G_ID: 0.296 G_Rec: 0.480 D_GP: 0.029 D_real: 0.940 D_fake: 0.785 +(epoch: 36, iters: 6320, time: 0.064) G_GAN: 0.308 G_GAN_Feat: 0.895 G_ID: 0.602 G_Rec: 0.372 D_GP: 0.083 D_real: 0.768 D_fake: 0.693 +(epoch: 36, iters: 6720, time: 0.064) G_GAN: 0.369 G_GAN_Feat: 0.633 G_ID: 0.296 G_Rec: 0.332 D_GP: 0.025 D_real: 1.207 D_fake: 0.632 +(epoch: 36, iters: 7120, time: 0.064) G_GAN: 0.223 G_GAN_Feat: 0.623 G_ID: 0.628 G_Rec: 0.326 D_GP: 0.026 D_real: 1.114 D_fake: 0.777 +(epoch: 36, iters: 7520, time: 0.064) G_GAN: 0.123 G_GAN_Feat: 0.669 G_ID: 0.321 G_Rec: 0.374 D_GP: 0.020 D_real: 1.009 D_fake: 0.877 +(epoch: 36, iters: 7920, time: 0.064) G_GAN: 0.103 G_GAN_Feat: 0.791 G_ID: 0.625 G_Rec: 0.415 D_GP: 0.032 D_real: 0.815 D_fake: 0.897 +(epoch: 36, iters: 8320, time: 0.064) G_GAN: 0.242 G_GAN_Feat: 0.721 G_ID: 0.279 G_Rec: 0.416 D_GP: 0.049 D_real: 1.027 D_fake: 0.759 +(epoch: 37, iters: 112, time: 0.064) G_GAN: 0.283 G_GAN_Feat: 0.802 G_ID: 0.592 G_Rec: 0.370 D_GP: 0.073 D_real: 1.043 D_fake: 0.720 +(epoch: 37, iters: 512, time: 0.064) G_GAN: 0.617 G_GAN_Feat: 0.774 G_ID: 0.327 G_Rec: 0.400 D_GP: 0.046 D_real: 1.309 D_fake: 0.401 +(epoch: 37, iters: 912, time: 0.064) G_GAN: 0.088 G_GAN_Feat: 0.738 G_ID: 0.615 G_Rec: 0.336 D_GP: 0.057 D_real: 0.741 D_fake: 0.914 +(epoch: 37, iters: 1312, time: 0.064) G_GAN: 0.474 G_GAN_Feat: 0.732 G_ID: 0.273 G_Rec: 0.346 D_GP: 0.035 D_real: 1.197 D_fake: 0.526 +(epoch: 37, iters: 1712, time: 0.064) G_GAN: -0.008 G_GAN_Feat: 1.181 G_ID: 0.656 G_Rec: 0.374 D_GP: 0.154 D_real: 0.289 D_fake: 1.008 +(epoch: 37, iters: 2112, time: 0.064) G_GAN: 0.241 G_GAN_Feat: 0.738 G_ID: 0.304 G_Rec: 0.419 D_GP: 0.031 D_real: 0.976 D_fake: 0.762 +(epoch: 37, iters: 2512, time: 0.064) G_GAN: 0.185 G_GAN_Feat: 0.605 G_ID: 0.607 G_Rec: 0.331 D_GP: 0.022 D_real: 1.075 D_fake: 0.816 +(epoch: 37, iters: 2912, time: 0.064) G_GAN: 0.485 G_GAN_Feat: 1.518 G_ID: 0.284 G_Rec: 0.461 D_GP: 0.080 D_real: 0.462 D_fake: 0.516 +(epoch: 37, iters: 3312, time: 0.064) G_GAN: 0.504 G_GAN_Feat: 0.602 G_ID: 0.635 G_Rec: 0.401 D_GP: 0.024 D_real: 1.335 D_fake: 0.499 +(epoch: 37, iters: 3712, time: 0.064) G_GAN: 0.087 G_GAN_Feat: 0.611 G_ID: 0.252 G_Rec: 0.366 D_GP: 0.034 D_real: 0.904 D_fake: 0.913 +(epoch: 37, iters: 4112, time: 0.064) G_GAN: 0.433 G_GAN_Feat: 0.636 G_ID: 0.626 G_Rec: 0.358 D_GP: 0.027 D_real: 1.218 D_fake: 0.568 +(epoch: 37, iters: 4512, time: 0.064) G_GAN: 0.274 G_GAN_Feat: 0.895 G_ID: 0.333 G_Rec: 0.360 D_GP: 0.075 D_real: 0.752 D_fake: 0.726 +(epoch: 37, iters: 4912, time: 0.064) G_GAN: 0.367 G_GAN_Feat: 0.842 G_ID: 0.557 G_Rec: 0.435 D_GP: 0.045 D_real: 0.964 D_fake: 0.633 +(epoch: 37, iters: 5312, time: 0.064) G_GAN: 0.427 G_GAN_Feat: 0.735 G_ID: 0.290 G_Rec: 0.363 D_GP: 0.062 D_real: 1.030 D_fake: 0.576 +(epoch: 37, iters: 5712, time: 0.064) G_GAN: 0.611 G_GAN_Feat: 0.787 G_ID: 0.620 G_Rec: 0.395 D_GP: 0.044 D_real: 1.289 D_fake: 0.409 +(epoch: 37, iters: 6112, time: 0.064) G_GAN: 0.471 G_GAN_Feat: 0.786 G_ID: 0.292 G_Rec: 0.437 D_GP: 0.033 D_real: 1.254 D_fake: 0.536 +(epoch: 37, iters: 6512, time: 0.064) G_GAN: 0.463 G_GAN_Feat: 0.752 G_ID: 0.615 G_Rec: 0.363 D_GP: 0.047 D_real: 1.223 D_fake: 0.571 +(epoch: 37, iters: 6912, time: 0.064) G_GAN: 0.371 G_GAN_Feat: 0.850 G_ID: 0.279 G_Rec: 0.375 D_GP: 0.053 D_real: 0.939 D_fake: 0.629 +(epoch: 37, iters: 7312, time: 0.064) G_GAN: 0.154 G_GAN_Feat: 0.781 G_ID: 0.583 G_Rec: 0.377 D_GP: 0.040 D_real: 0.730 D_fake: 0.846 +(epoch: 37, iters: 7712, time: 0.064) G_GAN: 0.206 G_GAN_Feat: 0.767 G_ID: 0.318 G_Rec: 0.381 D_GP: 0.070 D_real: 0.899 D_fake: 0.794 +(epoch: 37, iters: 8112, time: 0.064) G_GAN: 0.540 G_GAN_Feat: 1.050 G_ID: 0.577 G_Rec: 0.463 D_GP: 0.195 D_real: 0.670 D_fake: 0.532 +(epoch: 37, iters: 8512, time: 0.064) G_GAN: 0.510 G_GAN_Feat: 0.697 G_ID: 0.278 G_Rec: 0.359 D_GP: 0.030 D_real: 1.238 D_fake: 0.492 +(epoch: 38, iters: 304, time: 0.064) G_GAN: 0.171 G_GAN_Feat: 1.157 G_ID: 0.573 G_Rec: 0.359 D_GP: 0.165 D_real: 0.457 D_fake: 0.832 +(epoch: 38, iters: 704, time: 0.064) G_GAN: 0.176 G_GAN_Feat: 0.777 G_ID: 0.328 G_Rec: 0.348 D_GP: 0.044 D_real: 0.764 D_fake: 0.824 +(epoch: 38, iters: 1104, time: 0.064) G_GAN: 0.574 G_GAN_Feat: 0.707 G_ID: 0.620 G_Rec: 0.379 D_GP: 0.030 D_real: 1.345 D_fake: 0.459 +(epoch: 38, iters: 1504, time: 0.064) G_GAN: 0.469 G_GAN_Feat: 0.661 G_ID: 0.265 G_Rec: 0.363 D_GP: 0.029 D_real: 1.320 D_fake: 0.532 +(epoch: 38, iters: 1904, time: 0.064) G_GAN: 0.439 G_GAN_Feat: 1.124 G_ID: 0.573 G_Rec: 0.455 D_GP: 0.052 D_real: 0.677 D_fake: 0.565 +(epoch: 38, iters: 2304, time: 0.064) G_GAN: 0.484 G_GAN_Feat: 0.634 G_ID: 0.324 G_Rec: 0.381 D_GP: 0.022 D_real: 1.338 D_fake: 0.516 +(epoch: 38, iters: 2704, time: 0.064) G_GAN: 0.185 G_GAN_Feat: 0.825 G_ID: 0.650 G_Rec: 0.377 D_GP: 0.042 D_real: 0.733 D_fake: 0.815 +(epoch: 38, iters: 3104, time: 0.064) G_GAN: 0.177 G_GAN_Feat: 0.848 G_ID: 0.289 G_Rec: 0.434 D_GP: 0.075 D_real: 0.711 D_fake: 0.826 +(epoch: 38, iters: 3504, time: 0.064) G_GAN: 0.012 G_GAN_Feat: 0.842 G_ID: 0.617 G_Rec: 0.404 D_GP: 0.039 D_real: 0.585 D_fake: 0.988 +(epoch: 38, iters: 3904, time: 0.064) G_GAN: 0.271 G_GAN_Feat: 0.845 G_ID: 0.331 G_Rec: 0.395 D_GP: 0.039 D_real: 0.857 D_fake: 0.729 +(epoch: 38, iters: 4304, time: 0.064) G_GAN: 0.486 G_GAN_Feat: 0.894 G_ID: 0.587 G_Rec: 0.378 D_GP: 0.040 D_real: 1.047 D_fake: 0.515 +(epoch: 38, iters: 4704, time: 0.064) G_GAN: 0.582 G_GAN_Feat: 0.554 G_ID: 0.304 G_Rec: 0.367 D_GP: 0.020 D_real: 1.494 D_fake: 0.420 +(epoch: 38, iters: 5104, time: 0.064) G_GAN: 0.325 G_GAN_Feat: 0.509 G_ID: 0.617 G_Rec: 0.352 D_GP: 0.021 D_real: 1.240 D_fake: 0.675 +(epoch: 38, iters: 5504, time: 0.064) G_GAN: 0.435 G_GAN_Feat: 0.550 G_ID: 0.291 G_Rec: 0.596 D_GP: 0.022 D_real: 1.294 D_fake: 0.565 +(epoch: 38, iters: 5904, time: 0.064) G_GAN: 0.286 G_GAN_Feat: 0.574 G_ID: 0.581 G_Rec: 0.405 D_GP: 0.022 D_real: 1.142 D_fake: 0.714 +(epoch: 38, iters: 6304, time: 0.064) G_GAN: 0.231 G_GAN_Feat: 0.645 G_ID: 0.303 G_Rec: 0.369 D_GP: 0.032 D_real: 1.081 D_fake: 0.769 +(epoch: 38, iters: 6704, time: 0.064) G_GAN: 0.419 G_GAN_Feat: 0.703 G_ID: 0.609 G_Rec: 0.431 D_GP: 0.040 D_real: 1.157 D_fake: 0.583 +(epoch: 38, iters: 7104, time: 0.064) G_GAN: 0.371 G_GAN_Feat: 0.660 G_ID: 0.310 G_Rec: 0.349 D_GP: 0.050 D_real: 1.110 D_fake: 0.629 +(epoch: 38, iters: 7504, time: 0.064) G_GAN: 0.440 G_GAN_Feat: 0.808 G_ID: 0.626 G_Rec: 0.406 D_GP: 0.040 D_real: 1.207 D_fake: 0.565 +(epoch: 38, iters: 7904, time: 0.064) G_GAN: 0.592 G_GAN_Feat: 0.710 G_ID: 0.270 G_Rec: 0.392 D_GP: 0.030 D_real: 1.269 D_fake: 0.411 +(epoch: 38, iters: 8304, time: 0.064) G_GAN: 0.591 G_GAN_Feat: 0.711 G_ID: 0.575 G_Rec: 0.374 D_GP: 0.035 D_real: 1.357 D_fake: 0.462 +(epoch: 39, iters: 96, time: 0.064) G_GAN: 0.310 G_GAN_Feat: 0.741 G_ID: 0.311 G_Rec: 0.505 D_GP: 0.076 D_real: 0.981 D_fake: 0.690 +(epoch: 39, iters: 496, time: 0.064) G_GAN: 0.278 G_GAN_Feat: 0.708 G_ID: 0.643 G_Rec: 0.347 D_GP: 0.064 D_real: 1.029 D_fake: 0.722 +(epoch: 39, iters: 896, time: 0.064) G_GAN: 0.441 G_GAN_Feat: 0.783 G_ID: 0.341 G_Rec: 0.358 D_GP: 0.077 D_real: 1.099 D_fake: 0.567 +(epoch: 39, iters: 1296, time: 0.064) G_GAN: 0.380 G_GAN_Feat: 0.770 G_ID: 0.622 G_Rec: 0.373 D_GP: 0.069 D_real: 1.013 D_fake: 0.621 +(epoch: 39, iters: 1696, time: 0.064) G_GAN: 0.398 G_GAN_Feat: 0.672 G_ID: 0.276 G_Rec: 0.382 D_GP: 0.035 D_real: 1.086 D_fake: 0.605 +(epoch: 39, iters: 2096, time: 0.064) G_GAN: 0.328 G_GAN_Feat: 0.646 G_ID: 0.545 G_Rec: 0.354 D_GP: 0.051 D_real: 1.117 D_fake: 0.683 +(epoch: 39, iters: 2496, time: 0.064) G_GAN: 0.614 G_GAN_Feat: 0.832 G_ID: 0.261 G_Rec: 0.428 D_GP: 0.061 D_real: 1.080 D_fake: 0.452 +(epoch: 39, iters: 2896, time: 0.064) G_GAN: 0.272 G_GAN_Feat: 1.056 G_ID: 0.602 G_Rec: 0.362 D_GP: 0.137 D_real: 0.461 D_fake: 0.729 +(epoch: 39, iters: 3296, time: 0.064) G_GAN: 0.522 G_GAN_Feat: 0.888 G_ID: 0.267 G_Rec: 0.410 D_GP: 0.049 D_real: 1.179 D_fake: 0.481 +(epoch: 39, iters: 3696, time: 0.064) G_GAN: 0.227 G_GAN_Feat: 0.980 G_ID: 0.566 G_Rec: 0.412 D_GP: 0.064 D_real: 0.568 D_fake: 0.774 +(epoch: 39, iters: 4096, time: 0.064) G_GAN: 0.519 G_GAN_Feat: 1.037 G_ID: 0.289 G_Rec: 0.383 D_GP: 0.318 D_real: 0.700 D_fake: 0.525 +(epoch: 39, iters: 4496, time: 0.064) G_GAN: 0.705 G_GAN_Feat: 1.442 G_ID: 0.637 G_Rec: 0.478 D_GP: 0.615 D_real: 0.428 D_fake: 0.445 +(epoch: 39, iters: 4896, time: 0.064) G_GAN: 0.491 G_GAN_Feat: 0.706 G_ID: 0.283 G_Rec: 0.379 D_GP: 0.054 D_real: 1.123 D_fake: 0.509 +(epoch: 39, iters: 5296, time: 0.064) G_GAN: 0.447 G_GAN_Feat: 0.689 G_ID: 0.556 G_Rec: 0.351 D_GP: 0.034 D_real: 1.175 D_fake: 0.557 +(epoch: 39, iters: 5696, time: 0.064) G_GAN: 0.540 G_GAN_Feat: 0.643 G_ID: 0.265 G_Rec: 0.363 D_GP: 0.020 D_real: 1.345 D_fake: 0.465 +(epoch: 39, iters: 6096, time: 0.064) G_GAN: 0.184 G_GAN_Feat: 0.656 G_ID: 0.538 G_Rec: 0.407 D_GP: 0.031 D_real: 0.949 D_fake: 0.816 +(epoch: 39, iters: 6496, time: 0.064) G_GAN: 0.217 G_GAN_Feat: 1.075 G_ID: 0.311 G_Rec: 0.395 D_GP: 0.058 D_real: 0.474 D_fake: 0.784 +(epoch: 39, iters: 6896, time: 0.064) G_GAN: 0.301 G_GAN_Feat: 0.823 G_ID: 0.567 G_Rec: 0.395 D_GP: 0.034 D_real: 0.967 D_fake: 0.700 +(epoch: 39, iters: 7296, time: 0.064) G_GAN: 0.474 G_GAN_Feat: 0.788 G_ID: 0.284 G_Rec: 0.495 D_GP: 0.067 D_real: 1.087 D_fake: 0.526 +(epoch: 39, iters: 7696, time: 0.064) G_GAN: 0.290 G_GAN_Feat: 0.745 G_ID: 0.590 G_Rec: 0.412 D_GP: 0.034 D_real: 1.015 D_fake: 0.711 +(epoch: 39, iters: 8096, time: 0.064) G_GAN: 0.759 G_GAN_Feat: 0.898 G_ID: 0.273 G_Rec: 0.354 D_GP: 0.049 D_real: 1.493 D_fake: 0.267 +(epoch: 39, iters: 8496, time: 0.064) G_GAN: 0.418 G_GAN_Feat: 0.670 G_ID: 0.586 G_Rec: 0.378 D_GP: 0.019 D_real: 1.218 D_fake: 0.583 +(epoch: 40, iters: 288, time: 0.064) G_GAN: 0.193 G_GAN_Feat: 0.550 G_ID: 0.307 G_Rec: 0.362 D_GP: 0.025 D_real: 1.007 D_fake: 0.807 +(epoch: 40, iters: 688, time: 0.064) G_GAN: 0.009 G_GAN_Feat: 0.607 G_ID: 0.588 G_Rec: 0.307 D_GP: 0.031 D_real: 0.897 D_fake: 0.991 +(epoch: 40, iters: 1088, time: 0.064) G_GAN: 0.274 G_GAN_Feat: 0.739 G_ID: 0.274 G_Rec: 0.434 D_GP: 0.040 D_real: 0.909 D_fake: 0.727 +(epoch: 40, iters: 1488, time: 0.064) G_GAN: 0.177 G_GAN_Feat: 0.697 G_ID: 0.578 G_Rec: 0.381 D_GP: 0.034 D_real: 1.072 D_fake: 0.823 +(epoch: 40, iters: 1888, time: 0.064) G_GAN: 0.072 G_GAN_Feat: 0.829 G_ID: 0.353 G_Rec: 0.380 D_GP: 0.277 D_real: 0.459 D_fake: 0.930 +(epoch: 40, iters: 2288, time: 0.064) G_GAN: 0.489 G_GAN_Feat: 1.009 G_ID: 0.620 G_Rec: 0.361 D_GP: 0.127 D_real: 0.573 D_fake: 0.513 +(epoch: 40, iters: 2688, time: 0.064) G_GAN: 0.126 G_GAN_Feat: 0.710 G_ID: 0.319 G_Rec: 0.400 D_GP: 0.041 D_real: 0.946 D_fake: 0.874 +(epoch: 40, iters: 3088, time: 0.064) G_GAN: 0.752 G_GAN_Feat: 1.033 G_ID: 0.526 G_Rec: 0.421 D_GP: 0.110 D_real: 1.194 D_fake: 0.294 +(epoch: 40, iters: 3488, time: 0.064) G_GAN: 0.114 G_GAN_Feat: 0.652 G_ID: 0.335 G_Rec: 0.347 D_GP: 0.022 D_real: 0.964 D_fake: 0.886 +(epoch: 40, iters: 3888, time: 0.064) G_GAN: 0.335 G_GAN_Feat: 0.883 G_ID: 0.605 G_Rec: 0.434 D_GP: 0.068 D_real: 1.105 D_fake: 0.666 +(epoch: 40, iters: 4288, time: 0.064) G_GAN: 0.023 G_GAN_Feat: 0.744 G_ID: 0.260 G_Rec: 0.374 D_GP: 0.055 D_real: 0.743 D_fake: 0.977 +(epoch: 40, iters: 4688, time: 0.064) G_GAN: 0.521 G_GAN_Feat: 1.020 G_ID: 0.575 G_Rec: 0.456 D_GP: 0.178 D_real: 0.675 D_fake: 0.480 +(epoch: 40, iters: 5088, time: 0.064) G_GAN: 0.451 G_GAN_Feat: 0.726 G_ID: 0.255 G_Rec: 0.426 D_GP: 0.021 D_real: 1.249 D_fake: 0.560 +(epoch: 40, iters: 5488, time: 0.064) G_GAN: 0.308 G_GAN_Feat: 0.644 G_ID: 0.572 G_Rec: 0.351 D_GP: 0.028 D_real: 1.092 D_fake: 0.692 +(epoch: 40, iters: 5888, time: 0.064) G_GAN: 0.494 G_GAN_Feat: 0.607 G_ID: 0.277 G_Rec: 0.363 D_GP: 0.030 D_real: 1.312 D_fake: 0.511 +(epoch: 40, iters: 6288, time: 0.064) G_GAN: 0.507 G_GAN_Feat: 0.769 G_ID: 0.568 G_Rec: 0.382 D_GP: 0.044 D_real: 1.220 D_fake: 0.496 +(epoch: 40, iters: 6688, time: 0.064) G_GAN: 0.339 G_GAN_Feat: 0.802 G_ID: 0.297 G_Rec: 0.393 D_GP: 0.071 D_real: 0.971 D_fake: 0.662 +(epoch: 40, iters: 7088, time: 0.064) G_GAN: 0.349 G_GAN_Feat: 0.669 G_ID: 0.555 G_Rec: 0.397 D_GP: 0.029 D_real: 1.244 D_fake: 0.651 +(epoch: 40, iters: 7488, time: 0.064) G_GAN: 0.254 G_GAN_Feat: 0.754 G_ID: 0.293 G_Rec: 0.352 D_GP: 0.081 D_real: 0.932 D_fake: 0.747 +(epoch: 40, iters: 7888, time: 0.064) G_GAN: 0.012 G_GAN_Feat: 1.500 G_ID: 0.559 G_Rec: 0.404 D_GP: 1.063 D_real: 0.541 D_fake: 0.992 +(epoch: 40, iters: 8288, time: 0.064) G_GAN: 0.332 G_GAN_Feat: 0.859 G_ID: 0.273 G_Rec: 0.444 D_GP: 0.075 D_real: 0.813 D_fake: 0.668 +(epoch: 41, iters: 80, time: 0.064) G_GAN: 0.441 G_GAN_Feat: 0.745 G_ID: 0.553 G_Rec: 0.360 D_GP: 0.036 D_real: 1.219 D_fake: 0.583 +(epoch: 41, iters: 480, time: 0.064) G_GAN: 0.278 G_GAN_Feat: 0.708 G_ID: 0.356 G_Rec: 0.353 D_GP: 0.043 D_real: 0.949 D_fake: 0.722 +(epoch: 41, iters: 880, time: 0.064) G_GAN: 0.217 G_GAN_Feat: 0.886 G_ID: 0.614 G_Rec: 0.371 D_GP: 0.050 D_real: 0.831 D_fake: 0.783 +(epoch: 41, iters: 1280, time: 0.064) G_GAN: 0.212 G_GAN_Feat: 0.948 G_ID: 0.309 G_Rec: 0.342 D_GP: 0.095 D_real: 0.825 D_fake: 0.789 +(epoch: 41, iters: 1680, time: 0.064) G_GAN: 0.566 G_GAN_Feat: 0.756 G_ID: 0.557 G_Rec: 0.433 D_GP: 0.028 D_real: 1.328 D_fake: 0.441 +(epoch: 41, iters: 2080, time: 0.064) G_GAN: 0.313 G_GAN_Feat: 0.989 G_ID: 0.291 G_Rec: 0.402 D_GP: 0.042 D_real: 0.763 D_fake: 0.687 +(epoch: 41, iters: 2480, time: 0.064) G_GAN: 0.494 G_GAN_Feat: 0.743 G_ID: 0.501 G_Rec: 0.341 D_GP: 0.053 D_real: 1.188 D_fake: 0.507 +(epoch: 41, iters: 2880, time: 0.064) G_GAN: 0.054 G_GAN_Feat: 0.706 G_ID: 0.305 G_Rec: 0.346 D_GP: 0.040 D_real: 0.788 D_fake: 0.946 +(epoch: 41, iters: 3280, time: 0.064) G_GAN: 0.564 G_GAN_Feat: 0.873 G_ID: 0.553 G_Rec: 0.367 D_GP: 0.184 D_real: 0.965 D_fake: 0.439 +(epoch: 41, iters: 3680, time: 0.064) G_GAN: 0.329 G_GAN_Feat: 0.642 G_ID: 0.333 G_Rec: 0.385 D_GP: 0.033 D_real: 1.080 D_fake: 0.671 +(epoch: 41, iters: 4080, time: 0.064) G_GAN: 0.264 G_GAN_Feat: 0.927 G_ID: 0.531 G_Rec: 0.420 D_GP: 0.099 D_real: 0.766 D_fake: 0.736 +(epoch: 41, iters: 4480, time: 0.064) G_GAN: 0.183 G_GAN_Feat: 0.605 G_ID: 0.268 G_Rec: 0.339 D_GP: 0.023 D_real: 0.940 D_fake: 0.817 +(epoch: 41, iters: 4880, time: 0.064) G_GAN: 0.384 G_GAN_Feat: 0.792 G_ID: 0.538 G_Rec: 0.384 D_GP: 0.037 D_real: 0.934 D_fake: 0.617 +(epoch: 41, iters: 5280, time: 0.064) G_GAN: 0.151 G_GAN_Feat: 0.759 G_ID: 0.257 G_Rec: 0.384 D_GP: 0.069 D_real: 0.885 D_fake: 0.849 +(epoch: 41, iters: 5680, time: 0.064) G_GAN: 0.252 G_GAN_Feat: 0.716 G_ID: 0.579 G_Rec: 0.379 D_GP: 0.047 D_real: 1.044 D_fake: 0.749 +(epoch: 41, iters: 6080, time: 0.064) G_GAN: 0.482 G_GAN_Feat: 0.926 G_ID: 0.308 G_Rec: 0.376 D_GP: 0.092 D_real: 0.827 D_fake: 0.548 +(epoch: 41, iters: 6480, time: 0.064) G_GAN: 0.382 G_GAN_Feat: 0.611 G_ID: 0.595 G_Rec: 0.347 D_GP: 0.025 D_real: 1.286 D_fake: 0.618 +(epoch: 41, iters: 6880, time: 0.064) G_GAN: 0.192 G_GAN_Feat: 0.564 G_ID: 0.289 G_Rec: 0.394 D_GP: 0.026 D_real: 1.018 D_fake: 0.808 +(epoch: 41, iters: 7280, time: 0.064) G_GAN: -0.129 G_GAN_Feat: 0.623 G_ID: 0.539 G_Rec: 0.398 D_GP: 0.039 D_real: 0.642 D_fake: 1.129 +(epoch: 41, iters: 7680, time: 0.064) G_GAN: 0.230 G_GAN_Feat: 0.701 G_ID: 0.272 G_Rec: 0.389 D_GP: 0.049 D_real: 0.887 D_fake: 0.770 +(epoch: 41, iters: 8080, time: 0.064) G_GAN: 0.446 G_GAN_Feat: 0.729 G_ID: 0.547 G_Rec: 0.340 D_GP: 0.045 D_real: 1.183 D_fake: 0.556 +(epoch: 41, iters: 8480, time: 0.064) G_GAN: 0.349 G_GAN_Feat: 0.777 G_ID: 0.270 G_Rec: 0.357 D_GP: 0.092 D_real: 0.812 D_fake: 0.651 +(epoch: 42, iters: 272, time: 0.064) G_GAN: 0.163 G_GAN_Feat: 0.884 G_ID: 0.574 G_Rec: 0.351 D_GP: 0.043 D_real: 0.659 D_fake: 0.838 +(epoch: 42, iters: 672, time: 0.064) G_GAN: 0.010 G_GAN_Feat: 0.725 G_ID: 0.334 G_Rec: 0.379 D_GP: 0.045 D_real: 0.777 D_fake: 0.990 +(epoch: 42, iters: 1072, time: 0.064) G_GAN: 0.188 G_GAN_Feat: 0.828 G_ID: 0.619 G_Rec: 0.430 D_GP: 0.074 D_real: 0.809 D_fake: 0.812 +(epoch: 42, iters: 1472, time: 0.064) G_GAN: 0.378 G_GAN_Feat: 0.867 G_ID: 0.230 G_Rec: 0.412 D_GP: 0.041 D_real: 0.912 D_fake: 0.623 +(epoch: 42, iters: 1872, time: 0.064) G_GAN: 0.375 G_GAN_Feat: 0.787 G_ID: 0.592 G_Rec: 0.426 D_GP: 0.029 D_real: 1.134 D_fake: 0.626 +(epoch: 42, iters: 2272, time: 0.064) G_GAN: 0.139 G_GAN_Feat: 0.782 G_ID: 0.286 G_Rec: 0.348 D_GP: 0.079 D_real: 0.671 D_fake: 0.861 +(epoch: 42, iters: 2672, time: 0.064) G_GAN: 0.464 G_GAN_Feat: 0.790 G_ID: 0.581 G_Rec: 0.352 D_GP: 0.055 D_real: 1.134 D_fake: 0.545 +(epoch: 42, iters: 3072, time: 0.064) G_GAN: 0.666 G_GAN_Feat: 0.796 G_ID: 0.293 G_Rec: 0.368 D_GP: 0.039 D_real: 1.467 D_fake: 0.340 +(epoch: 42, iters: 3472, time: 0.064) G_GAN: 0.458 G_GAN_Feat: 0.918 G_ID: 0.569 G_Rec: 0.365 D_GP: 0.046 D_real: 1.139 D_fake: 0.542 +(epoch: 42, iters: 3872, time: 0.064) G_GAN: 0.289 G_GAN_Feat: 0.918 G_ID: 0.270 G_Rec: 0.384 D_GP: 0.088 D_real: 0.642 D_fake: 0.711 +(epoch: 42, iters: 4272, time: 0.064) G_GAN: 0.379 G_GAN_Feat: 0.811 G_ID: 0.530 G_Rec: 0.391 D_GP: 0.040 D_real: 1.056 D_fake: 0.621 +(epoch: 42, iters: 4672, time: 0.064) G_GAN: 0.458 G_GAN_Feat: 0.725 G_ID: 0.242 G_Rec: 0.467 D_GP: 0.029 D_real: 1.151 D_fake: 0.543 +(epoch: 42, iters: 5072, time: 0.064) G_GAN: 0.347 G_GAN_Feat: 0.593 G_ID: 0.477 G_Rec: 0.378 D_GP: 0.021 D_real: 1.218 D_fake: 0.653 +(epoch: 42, iters: 5472, time: 0.064) G_GAN: 0.756 G_GAN_Feat: 0.670 G_ID: 0.245 G_Rec: 0.485 D_GP: 0.028 D_real: 1.460 D_fake: 0.285 +(epoch: 42, iters: 5872, time: 0.064) G_GAN: 0.386 G_GAN_Feat: 0.673 G_ID: 0.538 G_Rec: 0.379 D_GP: 0.036 D_real: 1.222 D_fake: 0.614 +(epoch: 42, iters: 6272, time: 0.064) G_GAN: 0.254 G_GAN_Feat: 0.784 G_ID: 0.270 G_Rec: 0.374 D_GP: 0.073 D_real: 0.743 D_fake: 0.747 +(epoch: 42, iters: 6672, time: 0.064) G_GAN: 0.449 G_GAN_Feat: 0.800 G_ID: 0.612 G_Rec: 0.356 D_GP: 0.040 D_real: 1.085 D_fake: 0.552 +(epoch: 42, iters: 7072, time: 0.064) G_GAN: 0.350 G_GAN_Feat: 0.783 G_ID: 0.272 G_Rec: 0.375 D_GP: 0.044 D_real: 0.992 D_fake: 0.650 +(epoch: 42, iters: 7472, time: 0.064) G_GAN: 0.643 G_GAN_Feat: 0.717 G_ID: 0.571 G_Rec: 0.395 D_GP: 0.037 D_real: 1.357 D_fake: 0.359 +(epoch: 42, iters: 7872, time: 0.064) G_GAN: -0.068 G_GAN_Feat: 0.970 G_ID: 0.309 G_Rec: 0.383 D_GP: 0.114 D_real: 0.531 D_fake: 1.068 +(epoch: 42, iters: 8272, time: 0.064) G_GAN: 0.222 G_GAN_Feat: 0.813 G_ID: 0.546 G_Rec: 0.379 D_GP: 0.038 D_real: 0.775 D_fake: 0.779 +(epoch: 43, iters: 64, time: 0.064) G_GAN: 0.276 G_GAN_Feat: 0.949 G_ID: 0.274 G_Rec: 0.371 D_GP: 0.158 D_real: 0.368 D_fake: 0.724 +(epoch: 43, iters: 464, time: 0.064) G_GAN: 0.279 G_GAN_Feat: 0.866 G_ID: 0.559 G_Rec: 0.412 D_GP: 0.029 D_real: 1.099 D_fake: 0.721 +(epoch: 43, iters: 864, time: 0.064) G_GAN: 0.023 G_GAN_Feat: 0.892 G_ID: 0.326 G_Rec: 0.349 D_GP: 0.142 D_real: 0.408 D_fake: 0.978 +(epoch: 43, iters: 1264, time: 0.064) G_GAN: 0.431 G_GAN_Feat: 1.030 G_ID: 0.534 G_Rec: 0.378 D_GP: 0.051 D_real: 0.789 D_fake: 0.570 +(epoch: 43, iters: 1664, time: 0.064) G_GAN: 0.391 G_GAN_Feat: 1.053 G_ID: 0.245 G_Rec: 0.387 D_GP: 0.066 D_real: 0.578 D_fake: 0.612 +(epoch: 43, iters: 2064, time: 0.064) G_GAN: 0.312 G_GAN_Feat: 0.696 G_ID: 0.551 G_Rec: 0.395 D_GP: 0.024 D_real: 1.132 D_fake: 0.688 +(epoch: 43, iters: 2464, time: 0.064) G_GAN: 0.232 G_GAN_Feat: 0.592 G_ID: 0.249 G_Rec: 0.381 D_GP: 0.028 D_real: 1.130 D_fake: 0.768 +(epoch: 43, iters: 2864, time: 0.064) G_GAN: 0.157 G_GAN_Feat: 0.670 G_ID: 0.523 G_Rec: 0.401 D_GP: 0.036 D_real: 0.943 D_fake: 0.843 +(epoch: 43, iters: 3264, time: 0.064) G_GAN: 0.449 G_GAN_Feat: 0.747 G_ID: 0.278 G_Rec: 0.383 D_GP: 0.049 D_real: 1.077 D_fake: 0.553 +(epoch: 43, iters: 3664, time: 0.064) G_GAN: 0.157 G_GAN_Feat: 0.843 G_ID: 0.551 G_Rec: 0.405 D_GP: 0.097 D_real: 0.691 D_fake: 0.844 +(epoch: 43, iters: 4064, time: 0.064) G_GAN: 0.116 G_GAN_Feat: 0.832 G_ID: 0.284 G_Rec: 0.337 D_GP: 0.045 D_real: 0.635 D_fake: 0.884 +(epoch: 43, iters: 4464, time: 0.064) G_GAN: 0.253 G_GAN_Feat: 0.720 G_ID: 0.612 G_Rec: 0.357 D_GP: 0.039 D_real: 0.978 D_fake: 0.747 +(epoch: 43, iters: 4864, time: 0.064) G_GAN: 0.300 G_GAN_Feat: 0.806 G_ID: 0.320 G_Rec: 0.349 D_GP: 0.050 D_real: 0.758 D_fake: 0.701 +(epoch: 43, iters: 5264, time: 0.064) G_GAN: 0.576 G_GAN_Feat: 0.740 G_ID: 0.599 G_Rec: 0.365 D_GP: 0.028 D_real: 1.379 D_fake: 0.458 +(epoch: 43, iters: 5664, time: 0.064) G_GAN: 0.430 G_GAN_Feat: 0.707 G_ID: 0.301 G_Rec: 0.376 D_GP: 0.032 D_real: 1.169 D_fake: 0.570 +(epoch: 43, iters: 6064, time: 0.064) G_GAN: 0.043 G_GAN_Feat: 0.771 G_ID: 0.579 G_Rec: 0.387 D_GP: 0.025 D_real: 0.894 D_fake: 0.957 +(epoch: 43, iters: 6464, time: 0.064) G_GAN: 0.420 G_GAN_Feat: 0.847 G_ID: 0.276 G_Rec: 0.363 D_GP: 0.066 D_real: 0.928 D_fake: 0.601 +(epoch: 43, iters: 6864, time: 0.064) G_GAN: 0.421 G_GAN_Feat: 0.790 G_ID: 0.541 G_Rec: 0.398 D_GP: 0.027 D_real: 1.122 D_fake: 0.579 +(epoch: 43, iters: 7264, time: 0.064) G_GAN: 0.493 G_GAN_Feat: 0.879 G_ID: 0.263 G_Rec: 0.401 D_GP: 0.070 D_real: 0.990 D_fake: 0.517 +(epoch: 43, iters: 7664, time: 0.064) G_GAN: 0.377 G_GAN_Feat: 0.800 G_ID: 0.567 G_Rec: 0.382 D_GP: 0.026 D_real: 1.146 D_fake: 0.624 +(epoch: 43, iters: 8064, time: 0.064) G_GAN: 0.783 G_GAN_Feat: 0.805 G_ID: 0.304 G_Rec: 0.427 D_GP: 0.039 D_real: 1.600 D_fake: 0.322 +(epoch: 43, iters: 8464, time: 0.064) G_GAN: 0.543 G_GAN_Feat: 1.147 G_ID: 0.551 G_Rec: 0.381 D_GP: 0.101 D_real: 0.519 D_fake: 0.459 +(epoch: 44, iters: 256, time: 0.064) G_GAN: -0.319 G_GAN_Feat: 0.838 G_ID: 0.274 G_Rec: 0.414 D_GP: 0.044 D_real: 0.506 D_fake: 1.319 +(epoch: 44, iters: 656, time: 0.064) G_GAN: 0.300 G_GAN_Feat: 0.640 G_ID: 0.565 G_Rec: 0.494 D_GP: 0.029 D_real: 1.141 D_fake: 0.700 +(epoch: 44, iters: 1056, time: 0.064) G_GAN: 0.277 G_GAN_Feat: 0.762 G_ID: 0.270 G_Rec: 0.357 D_GP: 0.089 D_real: 1.047 D_fake: 0.725 +(epoch: 44, iters: 1456, time: 0.064) G_GAN: 0.368 G_GAN_Feat: 0.888 G_ID: 0.542 G_Rec: 0.366 D_GP: 0.047 D_real: 0.921 D_fake: 0.634 +(epoch: 44, iters: 1856, time: 0.064) G_GAN: 0.130 G_GAN_Feat: 0.972 G_ID: 0.250 G_Rec: 0.341 D_GP: 0.158 D_real: 0.333 D_fake: 0.870 +(epoch: 44, iters: 2256, time: 0.064) G_GAN: 0.047 G_GAN_Feat: 0.906 G_ID: 0.545 G_Rec: 0.348 D_GP: 0.179 D_real: 0.440 D_fake: 0.954 +(epoch: 44, iters: 2656, time: 0.064) G_GAN: 0.743 G_GAN_Feat: 0.981 G_ID: 0.305 G_Rec: 0.373 D_GP: 0.064 D_real: 1.328 D_fake: 0.395 +(epoch: 44, iters: 3056, time: 0.064) G_GAN: 0.516 G_GAN_Feat: 0.892 G_ID: 0.584 G_Rec: 0.356 D_GP: 0.056 D_real: 0.951 D_fake: 0.562 +(epoch: 44, iters: 3456, time: 0.064) G_GAN: 0.489 G_GAN_Feat: 0.856 G_ID: 0.249 G_Rec: 0.381 D_GP: 0.045 D_real: 1.105 D_fake: 0.533 +(epoch: 44, iters: 3856, time: 0.064) G_GAN: 0.254 G_GAN_Feat: 0.659 G_ID: 0.560 G_Rec: 0.380 D_GP: 0.022 D_real: 1.135 D_fake: 0.748 +(epoch: 44, iters: 4256, time: 0.064) G_GAN: 0.318 G_GAN_Feat: 0.613 G_ID: 0.339 G_Rec: 0.330 D_GP: 0.030 D_real: 1.084 D_fake: 0.682 +(epoch: 44, iters: 4656, time: 0.064) G_GAN: 0.372 G_GAN_Feat: 0.859 G_ID: 0.530 G_Rec: 0.441 D_GP: 0.115 D_real: 0.945 D_fake: 0.629 +(epoch: 44, iters: 5056, time: 0.064) G_GAN: 0.526 G_GAN_Feat: 0.904 G_ID: 0.276 G_Rec: 0.366 D_GP: 0.044 D_real: 1.178 D_fake: 0.482 +(epoch: 44, iters: 5456, time: 0.064) G_GAN: 0.630 G_GAN_Feat: 1.709 G_ID: 0.619 G_Rec: 0.512 D_GP: 1.071 D_real: 0.515 D_fake: 0.386 +(epoch: 44, iters: 5856, time: 0.064) G_GAN: 0.218 G_GAN_Feat: 0.576 G_ID: 0.268 G_Rec: 0.462 D_GP: 0.024 D_real: 1.046 D_fake: 0.782 +(epoch: 44, iters: 6256, time: 0.064) G_GAN: 0.305 G_GAN_Feat: 0.716 G_ID: 0.534 G_Rec: 0.380 D_GP: 0.052 D_real: 1.075 D_fake: 0.696 +(epoch: 44, iters: 6656, time: 0.064) G_GAN: 0.512 G_GAN_Feat: 0.831 G_ID: 0.265 G_Rec: 0.392 D_GP: 0.060 D_real: 0.983 D_fake: 0.505 +(epoch: 44, iters: 7056, time: 0.064) G_GAN: 0.292 G_GAN_Feat: 0.688 G_ID: 0.564 G_Rec: 0.410 D_GP: 0.035 D_real: 1.057 D_fake: 0.709 +(epoch: 44, iters: 7456, time: 0.064) G_GAN: 0.119 G_GAN_Feat: 0.759 G_ID: 0.249 G_Rec: 0.337 D_GP: 0.054 D_real: 0.747 D_fake: 0.881 +(epoch: 44, iters: 7856, time: 0.064) G_GAN: 0.374 G_GAN_Feat: 1.039 G_ID: 0.544 G_Rec: 0.386 D_GP: 0.216 D_real: 0.652 D_fake: 0.631 +(epoch: 44, iters: 8256, time: 0.064) G_GAN: 0.367 G_GAN_Feat: 0.800 G_ID: 0.284 G_Rec: 0.517 D_GP: 0.064 D_real: 0.956 D_fake: 0.634 +(epoch: 45, iters: 48, time: 0.064) G_GAN: 0.296 G_GAN_Feat: 0.946 G_ID: 0.527 G_Rec: 0.361 D_GP: 0.063 D_real: 0.752 D_fake: 0.705 +(epoch: 45, iters: 448, time: 0.064) G_GAN: 0.277 G_GAN_Feat: 1.196 G_ID: 0.282 G_Rec: 0.395 D_GP: 0.291 D_real: 0.517 D_fake: 0.723 +(epoch: 45, iters: 848, time: 0.064) G_GAN: 0.453 G_GAN_Feat: 0.697 G_ID: 0.518 G_Rec: 0.358 D_GP: 0.027 D_real: 1.217 D_fake: 0.548 +(epoch: 45, iters: 1248, time: 0.064) G_GAN: 0.403 G_GAN_Feat: 0.773 G_ID: 0.272 G_Rec: 0.372 D_GP: 0.054 D_real: 0.952 D_fake: 0.599 +(epoch: 45, iters: 1648, time: 0.064) G_GAN: 0.392 G_GAN_Feat: 0.729 G_ID: 0.547 G_Rec: 0.378 D_GP: 0.034 D_real: 1.132 D_fake: 0.608 +(epoch: 45, iters: 2048, time: 0.064) G_GAN: 0.253 G_GAN_Feat: 0.727 G_ID: 0.242 G_Rec: 0.339 D_GP: 0.057 D_real: 0.894 D_fake: 0.747 +(epoch: 45, iters: 2448, time: 0.064) G_GAN: 0.216 G_GAN_Feat: 0.581 G_ID: 0.521 G_Rec: 0.373 D_GP: 0.020 D_real: 1.097 D_fake: 0.784 +(epoch: 45, iters: 2848, time: 0.064) G_GAN: 0.333 G_GAN_Feat: 0.805 G_ID: 0.274 G_Rec: 0.383 D_GP: 0.048 D_real: 0.923 D_fake: 0.668 +(epoch: 45, iters: 3248, time: 0.064) G_GAN: 0.046 G_GAN_Feat: 1.211 G_ID: 0.516 G_Rec: 0.395 D_GP: 0.298 D_real: 0.423 D_fake: 0.955 +(epoch: 45, iters: 3648, time: 0.064) G_GAN: 0.524 G_GAN_Feat: 0.915 G_ID: 0.293 G_Rec: 0.399 D_GP: 0.131 D_real: 0.935 D_fake: 0.494 +(epoch: 45, iters: 4048, time: 0.064) G_GAN: 0.395 G_GAN_Feat: 0.843 G_ID: 0.525 G_Rec: 0.414 D_GP: 0.025 D_real: 1.214 D_fake: 0.608 +(epoch: 45, iters: 4448, time: 0.064) G_GAN: 0.347 G_GAN_Feat: 0.670 G_ID: 0.238 G_Rec: 0.374 D_GP: 0.029 D_real: 1.127 D_fake: 0.653 +(epoch: 45, iters: 4848, time: 0.064) G_GAN: 0.466 G_GAN_Feat: 0.800 G_ID: 0.537 G_Rec: 0.401 D_GP: 0.031 D_real: 1.296 D_fake: 0.540 +(epoch: 45, iters: 5248, time: 0.064) G_GAN: 0.578 G_GAN_Feat: 0.753 G_ID: 0.235 G_Rec: 0.348 D_GP: 0.070 D_real: 1.248 D_fake: 0.426 +(epoch: 45, iters: 5648, time: 0.064) G_GAN: 0.393 G_GAN_Feat: 0.842 G_ID: 0.543 G_Rec: 0.365 D_GP: 0.070 D_real: 0.910 D_fake: 0.615 +(epoch: 45, iters: 6048, time: 0.064) G_GAN: 0.408 G_GAN_Feat: 0.840 G_ID: 0.285 G_Rec: 0.386 D_GP: 0.043 D_real: 1.093 D_fake: 0.595 +(epoch: 45, iters: 6448, time: 0.064) G_GAN: 0.201 G_GAN_Feat: 0.708 G_ID: 0.583 G_Rec: 0.372 D_GP: 0.030 D_real: 0.958 D_fake: 0.799 +(epoch: 45, iters: 6848, time: 0.064) G_GAN: 0.420 G_GAN_Feat: 0.624 G_ID: 0.240 G_Rec: 0.309 D_GP: 0.025 D_real: 1.266 D_fake: 0.580 +(epoch: 45, iters: 7248, time: 0.064) G_GAN: 0.390 G_GAN_Feat: 0.876 G_ID: 0.562 G_Rec: 0.349 D_GP: 0.078 D_real: 0.929 D_fake: 0.610 +(epoch: 45, iters: 7648, time: 0.064) G_GAN: 0.240 G_GAN_Feat: 0.751 G_ID: 0.227 G_Rec: 0.391 D_GP: 0.032 D_real: 1.075 D_fake: 0.760 +(epoch: 45, iters: 8048, time: 0.064) G_GAN: 0.280 G_GAN_Feat: 0.790 G_ID: 0.537 G_Rec: 0.338 D_GP: 0.035 D_real: 1.024 D_fake: 0.720 +(epoch: 45, iters: 8448, time: 0.064) G_GAN: 0.414 G_GAN_Feat: 0.977 G_ID: 0.288 G_Rec: 0.408 D_GP: 0.111 D_real: 0.693 D_fake: 0.589 +(epoch: 46, iters: 240, time: 0.064) G_GAN: 0.460 G_GAN_Feat: 0.746 G_ID: 0.590 G_Rec: 0.378 D_GP: 0.038 D_real: 1.212 D_fake: 0.540 +(epoch: 46, iters: 640, time: 0.064) G_GAN: 0.401 G_GAN_Feat: 0.710 G_ID: 0.262 G_Rec: 0.334 D_GP: 0.032 D_real: 1.159 D_fake: 0.600 +(epoch: 46, iters: 1040, time: 0.064) G_GAN: 0.524 G_GAN_Feat: 0.614 G_ID: 0.548 G_Rec: 0.388 D_GP: 0.024 D_real: 1.389 D_fake: 0.476 +(epoch: 46, iters: 1440, time: 0.064) G_GAN: 0.409 G_GAN_Feat: 0.785 G_ID: 0.293 G_Rec: 0.366 D_GP: 0.065 D_real: 0.979 D_fake: 0.594 +(epoch: 46, iters: 1840, time: 0.064) G_GAN: 0.221 G_GAN_Feat: 0.885 G_ID: 0.500 G_Rec: 0.363 D_GP: 0.148 D_real: 0.852 D_fake: 0.779 +(epoch: 46, iters: 2240, time: 0.064) G_GAN: 0.085 G_GAN_Feat: 0.725 G_ID: 0.291 G_Rec: 0.364 D_GP: 0.035 D_real: 0.771 D_fake: 0.915 +(epoch: 46, iters: 2640, time: 0.064) G_GAN: 0.560 G_GAN_Feat: 0.840 G_ID: 0.519 G_Rec: 0.390 D_GP: 0.054 D_real: 1.165 D_fake: 0.443 +(epoch: 46, iters: 3040, time: 0.064) G_GAN: 0.330 G_GAN_Feat: 0.737 G_ID: 0.312 G_Rec: 0.387 D_GP: 0.036 D_real: 0.983 D_fake: 0.670 +(epoch: 46, iters: 3440, time: 0.064) G_GAN: 0.319 G_GAN_Feat: 0.832 G_ID: 0.587 G_Rec: 0.373 D_GP: 0.049 D_real: 0.857 D_fake: 0.681 +(epoch: 46, iters: 3840, time: 0.064) G_GAN: 0.428 G_GAN_Feat: 1.066 G_ID: 0.217 G_Rec: 0.396 D_GP: 0.310 D_real: 0.677 D_fake: 0.592 +(epoch: 46, iters: 4240, time: 0.064) G_GAN: 0.321 G_GAN_Feat: 0.789 G_ID: 0.510 G_Rec: 0.369 D_GP: 0.041 D_real: 0.889 D_fake: 0.681 +(epoch: 46, iters: 4640, time: 0.064) G_GAN: 0.212 G_GAN_Feat: 0.673 G_ID: 0.279 G_Rec: 0.400 D_GP: 0.033 D_real: 0.994 D_fake: 0.789 +(epoch: 46, iters: 5040, time: 0.064) G_GAN: 0.302 G_GAN_Feat: 0.957 G_ID: 0.529 G_Rec: 0.402 D_GP: 0.057 D_real: 0.687 D_fake: 0.698 +(epoch: 46, iters: 5440, time: 0.064) G_GAN: 0.379 G_GAN_Feat: 0.902 G_ID: 0.279 G_Rec: 0.371 D_GP: 0.075 D_real: 0.806 D_fake: 0.654 +(epoch: 46, iters: 5840, time: 0.064) G_GAN: 0.483 G_GAN_Feat: 0.819 G_ID: 0.551 G_Rec: 0.368 D_GP: 0.042 D_real: 1.148 D_fake: 0.520 +(epoch: 46, iters: 6240, time: 0.064) G_GAN: 0.362 G_GAN_Feat: 0.694 G_ID: 0.250 G_Rec: 0.338 D_GP: 0.039 D_real: 1.129 D_fake: 0.639 +(epoch: 46, iters: 6640, time: 0.064) G_GAN: 0.256 G_GAN_Feat: 0.940 G_ID: 0.495 G_Rec: 0.397 D_GP: 0.102 D_real: 0.549 D_fake: 0.744 +(epoch: 46, iters: 7040, time: 0.064) G_GAN: 0.619 G_GAN_Feat: 1.309 G_ID: 0.266 G_Rec: 0.388 D_GP: 0.107 D_real: 0.531 D_fake: 0.392 +(epoch: 46, iters: 7440, time: 0.064) G_GAN: 0.475 G_GAN_Feat: 0.832 G_ID: 0.544 G_Rec: 0.361 D_GP: 0.058 D_real: 1.193 D_fake: 0.530 +(epoch: 46, iters: 7840, time: 0.064) G_GAN: 0.531 G_GAN_Feat: 0.784 G_ID: 0.246 G_Rec: 0.346 D_GP: 0.036 D_real: 1.198 D_fake: 0.476 +(epoch: 46, iters: 8240, time: 0.064) G_GAN: 0.182 G_GAN_Feat: 0.892 G_ID: 0.519 G_Rec: 0.395 D_GP: 0.044 D_real: 0.696 D_fake: 0.818 +(epoch: 47, iters: 32, time: 0.064) G_GAN: 0.193 G_GAN_Feat: 1.384 G_ID: 0.272 G_Rec: 0.360 D_GP: 0.099 D_real: 0.215 D_fake: 0.807 +(epoch: 47, iters: 432, time: 0.064) G_GAN: 0.341 G_GAN_Feat: 0.855 G_ID: 0.567 G_Rec: 0.339 D_GP: 0.067 D_real: 0.971 D_fake: 0.660 +(epoch: 47, iters: 832, time: 0.064) G_GAN: 0.577 G_GAN_Feat: 0.820 G_ID: 0.280 G_Rec: 0.409 D_GP: 0.047 D_real: 1.259 D_fake: 0.447 +(epoch: 47, iters: 1232, time: 0.064) G_GAN: 0.062 G_GAN_Feat: 0.738 G_ID: 0.516 G_Rec: 0.440 D_GP: 0.034 D_real: 0.795 D_fake: 0.938 +(epoch: 47, iters: 1632, time: 0.064) G_GAN: 0.379 G_GAN_Feat: 0.632 G_ID: 0.284 G_Rec: 0.347 D_GP: 0.026 D_real: 1.184 D_fake: 0.621 +(epoch: 47, iters: 2032, time: 0.064) G_GAN: 0.423 G_GAN_Feat: 0.737 G_ID: 0.554 G_Rec: 0.361 D_GP: 0.054 D_real: 1.182 D_fake: 0.578 +(epoch: 47, iters: 2432, time: 0.064) G_GAN: 0.127 G_GAN_Feat: 0.811 G_ID: 0.230 G_Rec: 0.440 D_GP: 0.045 D_real: 0.663 D_fake: 0.873 +(epoch: 47, iters: 2832, time: 0.064) G_GAN: 0.841 G_GAN_Feat: 1.052 G_ID: 0.490 G_Rec: 0.380 D_GP: 0.080 D_real: 1.423 D_fake: 0.273 +(epoch: 47, iters: 3232, time: 0.064) G_GAN: 0.470 G_GAN_Feat: 0.610 G_ID: 0.284 G_Rec: 0.351 D_GP: 0.022 D_real: 1.406 D_fake: 0.532 +(epoch: 47, iters: 3632, time: 0.064) G_GAN: 0.351 G_GAN_Feat: 0.635 G_ID: 0.523 G_Rec: 0.369 D_GP: 0.022 D_real: 1.206 D_fake: 0.651 +(epoch: 47, iters: 4032, time: 0.064) G_GAN: 0.282 G_GAN_Feat: 0.572 G_ID: 0.263 G_Rec: 0.374 D_GP: 0.024 D_real: 1.142 D_fake: 0.719 +(epoch: 47, iters: 4432, time: 0.064) G_GAN: 0.123 G_GAN_Feat: 0.650 G_ID: 0.561 G_Rec: 0.420 D_GP: 0.037 D_real: 0.874 D_fake: 0.877 +(epoch: 47, iters: 4832, time: 0.064) G_GAN: 0.036 G_GAN_Feat: 0.655 G_ID: 0.280 G_Rec: 0.417 D_GP: 0.041 D_real: 0.771 D_fake: 0.964 +(epoch: 47, iters: 5232, time: 0.064) G_GAN: 0.444 G_GAN_Feat: 0.696 G_ID: 0.493 G_Rec: 0.358 D_GP: 0.042 D_real: 1.145 D_fake: 0.560 +(epoch: 47, iters: 5632, time: 0.064) G_GAN: 0.178 G_GAN_Feat: 0.751 G_ID: 0.288 G_Rec: 0.423 D_GP: 0.067 D_real: 0.772 D_fake: 0.822 +(epoch: 47, iters: 6032, time: 0.064) G_GAN: 0.255 G_GAN_Feat: 0.682 G_ID: 0.589 G_Rec: 0.370 D_GP: 0.036 D_real: 1.037 D_fake: 0.746 +(epoch: 47, iters: 6432, time: 0.064) G_GAN: 0.324 G_GAN_Feat: 0.705 G_ID: 0.288 G_Rec: 0.345 D_GP: 0.046 D_real: 1.023 D_fake: 0.678 +(epoch: 47, iters: 6832, time: 0.064) G_GAN: 0.262 G_GAN_Feat: 1.209 G_ID: 0.544 G_Rec: 0.363 D_GP: 0.480 D_real: 0.486 D_fake: 0.738 +(epoch: 47, iters: 7232, time: 0.064) G_GAN: -0.062 G_GAN_Feat: 0.693 G_ID: 0.270 G_Rec: 0.365 D_GP: 0.041 D_real: 0.670 D_fake: 1.062 +(epoch: 47, iters: 7632, time: 0.064) G_GAN: 0.233 G_GAN_Feat: 1.098 G_ID: 0.542 G_Rec: 0.374 D_GP: 0.092 D_real: 0.430 D_fake: 0.767 +(epoch: 47, iters: 8032, time: 0.064) G_GAN: 0.631 G_GAN_Feat: 0.796 G_ID: 0.259 G_Rec: 0.442 D_GP: 0.053 D_real: 1.214 D_fake: 0.376 +(epoch: 47, iters: 8432, time: 0.064) G_GAN: 0.441 G_GAN_Feat: 0.773 G_ID: 0.508 G_Rec: 0.379 D_GP: 0.039 D_real: 1.155 D_fake: 0.560 +(epoch: 48, iters: 224, time: 0.064) G_GAN: 0.160 G_GAN_Feat: 0.707 G_ID: 0.266 G_Rec: 0.444 D_GP: 0.029 D_real: 0.850 D_fake: 0.840 +(epoch: 48, iters: 624, time: 0.064) G_GAN: 0.493 G_GAN_Feat: 0.829 G_ID: 0.525 G_Rec: 0.380 D_GP: 0.041 D_real: 1.105 D_fake: 0.508 +(epoch: 48, iters: 1024, time: 0.064) G_GAN: 0.172 G_GAN_Feat: 0.648 G_ID: 0.262 G_Rec: 0.400 D_GP: 0.024 D_real: 0.996 D_fake: 0.829 +(epoch: 48, iters: 1424, time: 0.064) G_GAN: 0.153 G_GAN_Feat: 0.745 G_ID: 0.507 G_Rec: 0.369 D_GP: 0.029 D_real: 0.929 D_fake: 0.848 +(epoch: 48, iters: 1824, time: 0.064) G_GAN: 0.174 G_GAN_Feat: 0.698 G_ID: 0.249 G_Rec: 0.370 D_GP: 0.030 D_real: 1.020 D_fake: 0.826 +(epoch: 48, iters: 2224, time: 0.064) G_GAN: 0.197 G_GAN_Feat: 0.859 G_ID: 0.509 G_Rec: 0.384 D_GP: 0.052 D_real: 0.912 D_fake: 0.804 +(epoch: 48, iters: 2624, time: 0.064) G_GAN: 0.193 G_GAN_Feat: 0.919 G_ID: 0.302 G_Rec: 0.354 D_GP: 0.037 D_real: 1.114 D_fake: 0.807 +(epoch: 48, iters: 3024, time: 0.064) G_GAN: 0.390 G_GAN_Feat: 1.001 G_ID: 0.501 G_Rec: 0.375 D_GP: 0.047 D_real: 0.974 D_fake: 0.614 +(epoch: 48, iters: 3424, time: 0.064) G_GAN: 0.509 G_GAN_Feat: 0.772 G_ID: 0.255 G_Rec: 0.339 D_GP: 0.049 D_real: 1.243 D_fake: 0.544 +(epoch: 48, iters: 3824, time: 0.064) G_GAN: 0.208 G_GAN_Feat: 0.798 G_ID: 0.565 G_Rec: 0.374 D_GP: 0.032 D_real: 1.082 D_fake: 0.793 +(epoch: 48, iters: 4224, time: 0.064) G_GAN: 0.290 G_GAN_Feat: 0.794 G_ID: 0.289 G_Rec: 0.339 D_GP: 0.044 D_real: 0.949 D_fake: 0.712 +(epoch: 48, iters: 4624, time: 0.064) G_GAN: -0.050 G_GAN_Feat: 0.874 G_ID: 0.490 G_Rec: 0.388 D_GP: 0.037 D_real: 0.548 D_fake: 1.050 +(epoch: 48, iters: 5024, time: 0.064) G_GAN: 0.296 G_GAN_Feat: 0.708 G_ID: 0.249 G_Rec: 0.346 D_GP: 0.023 D_real: 1.069 D_fake: 0.705 +(epoch: 48, iters: 5424, time: 0.064) G_GAN: 0.125 G_GAN_Feat: 0.704 G_ID: 0.496 G_Rec: 0.333 D_GP: 0.037 D_real: 0.955 D_fake: 0.875 +(epoch: 48, iters: 5824, time: 0.064) G_GAN: -0.213 G_GAN_Feat: 0.858 G_ID: 0.253 G_Rec: 0.376 D_GP: 0.047 D_real: 0.513 D_fake: 1.213 +(epoch: 48, iters: 6224, time: 0.064) G_GAN: 0.419 G_GAN_Feat: 0.918 G_ID: 0.493 G_Rec: 0.348 D_GP: 0.052 D_real: 1.234 D_fake: 0.582 +(epoch: 48, iters: 6624, time: 0.064) G_GAN: 0.211 G_GAN_Feat: 0.590 G_ID: 0.259 G_Rec: 0.328 D_GP: 0.022 D_real: 1.079 D_fake: 0.789 +(epoch: 48, iters: 7024, time: 0.064) G_GAN: 0.259 G_GAN_Feat: 0.682 G_ID: 0.513 G_Rec: 0.458 D_GP: 0.026 D_real: 1.002 D_fake: 0.741 +(epoch: 48, iters: 7424, time: 0.064) G_GAN: 0.188 G_GAN_Feat: 0.721 G_ID: 0.291 G_Rec: 0.370 D_GP: 0.053 D_real: 0.915 D_fake: 0.812 +(epoch: 48, iters: 7824, time: 0.064) G_GAN: 0.322 G_GAN_Feat: 0.796 G_ID: 0.565 G_Rec: 0.380 D_GP: 0.039 D_real: 0.992 D_fake: 0.678 +(epoch: 48, iters: 8224, time: 0.064) G_GAN: 0.666 G_GAN_Feat: 1.164 G_ID: 0.272 G_Rec: 0.438 D_GP: 0.114 D_real: 1.092 D_fake: 0.339 +(epoch: 49, iters: 16, time: 0.064) G_GAN: 0.107 G_GAN_Feat: 0.975 G_ID: 0.546 G_Rec: 0.350 D_GP: 0.107 D_real: 0.411 D_fake: 0.894 +(epoch: 49, iters: 416, time: 0.064) G_GAN: 0.489 G_GAN_Feat: 0.847 G_ID: 0.270 G_Rec: 0.400 D_GP: 0.046 D_real: 1.108 D_fake: 0.511 +(epoch: 49, iters: 816, time: 0.064) G_GAN: 0.659 G_GAN_Feat: 0.967 G_ID: 0.535 G_Rec: 0.347 D_GP: 0.160 D_real: 0.997 D_fake: 0.448 +(epoch: 49, iters: 1216, time: 0.064) G_GAN: 0.771 G_GAN_Feat: 0.986 G_ID: 0.304 G_Rec: 0.406 D_GP: 0.102 D_real: 1.035 D_fake: 0.256 +(epoch: 49, iters: 1616, time: 0.064) G_GAN: 0.767 G_GAN_Feat: 1.042 G_ID: 0.509 G_Rec: 0.479 D_GP: 0.073 D_real: 1.235 D_fake: 0.446 +(epoch: 49, iters: 2016, time: 0.064) G_GAN: 0.362 G_GAN_Feat: 0.850 G_ID: 0.267 G_Rec: 0.348 D_GP: 0.060 D_real: 0.860 D_fake: 0.638 +(epoch: 49, iters: 2416, time: 0.064) G_GAN: 0.317 G_GAN_Feat: 0.890 G_ID: 0.528 G_Rec: 0.399 D_GP: 0.071 D_real: 0.779 D_fake: 0.684 +(epoch: 49, iters: 2816, time: 0.064) G_GAN: 0.460 G_GAN_Feat: 0.717 G_ID: 0.291 G_Rec: 0.369 D_GP: 0.029 D_real: 1.219 D_fake: 0.541 +(epoch: 49, iters: 3216, time: 0.064) G_GAN: 0.340 G_GAN_Feat: 0.637 G_ID: 0.534 G_Rec: 0.400 D_GP: 0.025 D_real: 1.210 D_fake: 0.660 +(epoch: 49, iters: 3616, time: 0.064) G_GAN: 0.406 G_GAN_Feat: 0.547 G_ID: 0.233 G_Rec: 0.489 D_GP: 0.020 D_real: 1.155 D_fake: 0.595 +(epoch: 49, iters: 4016, time: 0.064) G_GAN: 0.349 G_GAN_Feat: 0.598 G_ID: 0.532 G_Rec: 0.349 D_GP: 0.022 D_real: 1.290 D_fake: 0.656 +(epoch: 49, iters: 4416, time: 0.064) G_GAN: 0.278 G_GAN_Feat: 0.672 G_ID: 0.232 G_Rec: 0.346 D_GP: 0.030 D_real: 1.038 D_fake: 0.722 +(epoch: 49, iters: 4816, time: 0.064) G_GAN: 0.272 G_GAN_Feat: 0.753 G_ID: 0.481 G_Rec: 0.338 D_GP: 0.076 D_real: 0.857 D_fake: 0.728 +(epoch: 49, iters: 5216, time: 0.064) G_GAN: 0.389 G_GAN_Feat: 0.821 G_ID: 0.268 G_Rec: 0.384 D_GP: 0.066 D_real: 1.075 D_fake: 0.631 +(epoch: 49, iters: 5616, time: 0.064) G_GAN: 0.721 G_GAN_Feat: 0.890 G_ID: 0.568 G_Rec: 0.440 D_GP: 0.084 D_real: 1.115 D_fake: 0.443 +(epoch: 49, iters: 6016, time: 0.064) G_GAN: 0.246 G_GAN_Feat: 0.725 G_ID: 0.233 G_Rec: 0.356 D_GP: 0.042 D_real: 1.021 D_fake: 0.754 +(epoch: 49, iters: 6416, time: 0.064) G_GAN: 0.088 G_GAN_Feat: 0.695 G_ID: 0.531 G_Rec: 0.359 D_GP: 0.030 D_real: 0.847 D_fake: 0.912 +(epoch: 49, iters: 6816, time: 0.064) G_GAN: 0.488 G_GAN_Feat: 0.672 G_ID: 0.262 G_Rec: 0.363 D_GP: 0.029 D_real: 1.305 D_fake: 0.512 +(epoch: 49, iters: 7216, time: 0.064) G_GAN: 0.635 G_GAN_Feat: 0.903 G_ID: 0.488 G_Rec: 0.408 D_GP: 0.064 D_real: 1.338 D_fake: 0.386 +(epoch: 49, iters: 7616, time: 0.064) G_GAN: 0.470 G_GAN_Feat: 0.694 G_ID: 0.277 G_Rec: 0.373 D_GP: 0.030 D_real: 1.190 D_fake: 0.531 +(epoch: 49, iters: 8016, time: 0.064) G_GAN: 0.317 G_GAN_Feat: 0.715 G_ID: 0.544 G_Rec: 0.337 D_GP: 0.029 D_real: 1.112 D_fake: 0.683 +(epoch: 49, iters: 8416, time: 0.064) G_GAN: 0.325 G_GAN_Feat: 0.987 G_ID: 0.242 G_Rec: 0.364 D_GP: 0.061 D_real: 0.585 D_fake: 0.675 +(epoch: 50, iters: 208, time: 0.064) G_GAN: 0.761 G_GAN_Feat: 1.155 G_ID: 0.451 G_Rec: 0.411 D_GP: 0.059 D_real: 1.154 D_fake: 0.385 +(epoch: 50, iters: 608, time: 0.064) G_GAN: 0.061 G_GAN_Feat: 0.697 G_ID: 0.263 G_Rec: 0.403 D_GP: 0.031 D_real: 0.892 D_fake: 0.940 +(epoch: 50, iters: 1008, time: 0.064) G_GAN: 0.694 G_GAN_Feat: 0.848 G_ID: 0.512 G_Rec: 0.374 D_GP: 0.045 D_real: 1.268 D_fake: 0.335 +(epoch: 50, iters: 1408, time: 0.064) G_GAN: 0.453 G_GAN_Feat: 0.782 G_ID: 0.242 G_Rec: 0.369 D_GP: 0.037 D_real: 1.045 D_fake: 0.548 +(epoch: 50, iters: 1808, time: 0.064) G_GAN: 0.510 G_GAN_Feat: 0.807 G_ID: 0.550 G_Rec: 0.335 D_GP: 0.049 D_real: 1.120 D_fake: 0.490 +(epoch: 50, iters: 2208, time: 0.064) G_GAN: 0.510 G_GAN_Feat: 0.766 G_ID: 0.274 G_Rec: 0.453 D_GP: 0.047 D_real: 1.260 D_fake: 0.493 +(epoch: 50, iters: 2608, time: 0.064) G_GAN: 0.154 G_GAN_Feat: 1.035 G_ID: 0.502 G_Rec: 0.459 D_GP: 0.032 D_real: 1.054 D_fake: 0.881 +(epoch: 50, iters: 3008, time: 0.064) G_GAN: 0.355 G_GAN_Feat: 0.629 G_ID: 0.241 G_Rec: 0.379 D_GP: 0.020 D_real: 1.200 D_fake: 0.645 +(epoch: 50, iters: 3408, time: 0.064) G_GAN: 0.111 G_GAN_Feat: 0.625 G_ID: 0.520 G_Rec: 0.350 D_GP: 0.021 D_real: 0.994 D_fake: 0.889 +(epoch: 50, iters: 3808, time: 0.064) G_GAN: 0.229 G_GAN_Feat: 0.656 G_ID: 0.262 G_Rec: 0.343 D_GP: 0.029 D_real: 1.054 D_fake: 0.771 +(epoch: 50, iters: 4208, time: 0.064) G_GAN: 0.312 G_GAN_Feat: 0.766 G_ID: 0.501 G_Rec: 0.360 D_GP: 0.031 D_real: 1.136 D_fake: 0.689 +(epoch: 50, iters: 4608, time: 0.064) G_GAN: 0.600 G_GAN_Feat: 0.793 G_ID: 0.275 G_Rec: 0.451 D_GP: 0.059 D_real: 1.283 D_fake: 0.408 +(epoch: 50, iters: 5008, time: 0.064) G_GAN: 0.064 G_GAN_Feat: 0.783 G_ID: 0.516 G_Rec: 0.404 D_GP: 0.083 D_real: 0.795 D_fake: 0.937 +(epoch: 50, iters: 5408, time: 0.064) G_GAN: 0.309 G_GAN_Feat: 1.103 G_ID: 0.254 G_Rec: 0.388 D_GP: 0.080 D_real: 0.584 D_fake: 0.691 +(epoch: 50, iters: 5808, time: 0.064) G_GAN: 0.389 G_GAN_Feat: 1.068 G_ID: 0.525 G_Rec: 0.456 D_GP: 0.092 D_real: 0.743 D_fake: 0.614 +(epoch: 50, iters: 6208, time: 0.064) G_GAN: 0.264 G_GAN_Feat: 0.743 G_ID: 0.227 G_Rec: 0.373 D_GP: 0.045 D_real: 0.978 D_fake: 0.738 +(epoch: 50, iters: 6608, time: 0.064) G_GAN: 0.119 G_GAN_Feat: 0.809 G_ID: 0.481 G_Rec: 0.393 D_GP: 0.062 D_real: 0.665 D_fake: 0.881 +(epoch: 50, iters: 7008, time: 0.064) G_GAN: 0.774 G_GAN_Feat: 1.046 G_ID: 0.266 G_Rec: 0.502 D_GP: 0.282 D_real: 1.175 D_fake: 0.273 +(epoch: 50, iters: 7408, time: 0.064) G_GAN: 0.147 G_GAN_Feat: 1.229 G_ID: 0.510 G_Rec: 0.425 D_GP: 0.146 D_real: 0.299 D_fake: 0.854 +(epoch: 50, iters: 7808, time: 0.064) G_GAN: 0.294 G_GAN_Feat: 0.825 G_ID: 0.296 G_Rec: 0.369 D_GP: 0.052 D_real: 0.932 D_fake: 0.706 +(epoch: 50, iters: 8208, time: 0.064) G_GAN: 0.300 G_GAN_Feat: 0.771 G_ID: 0.524 G_Rec: 0.348 D_GP: 0.027 D_real: 1.042 D_fake: 0.700 +(epoch: 50, iters: 8608, time: 0.064) G_GAN: 0.237 G_GAN_Feat: 0.723 G_ID: 0.267 G_Rec: 0.377 D_GP: 0.039 D_real: 0.898 D_fake: 0.763 +(epoch: 51, iters: 400, time: 0.064) G_GAN: 0.118 G_GAN_Feat: 0.875 G_ID: 0.466 G_Rec: 0.363 D_GP: 0.048 D_real: 0.627 D_fake: 0.882 +(epoch: 51, iters: 800, time: 0.064) G_GAN: 0.335 G_GAN_Feat: 0.845 G_ID: 0.285 G_Rec: 0.368 D_GP: 0.060 D_real: 0.774 D_fake: 0.665 +(epoch: 51, iters: 1200, time: 0.064) G_GAN: 0.566 G_GAN_Feat: 0.790 G_ID: 0.540 G_Rec: 0.419 D_GP: 0.038 D_real: 1.254 D_fake: 0.495 +(epoch: 51, iters: 1600, time: 0.064) G_GAN: 0.250 G_GAN_Feat: 1.325 G_ID: 0.213 G_Rec: 0.374 D_GP: 0.357 D_real: 0.511 D_fake: 0.750 +(epoch: 51, iters: 2000, time: 0.064) G_GAN: 0.430 G_GAN_Feat: 0.620 G_ID: 0.482 G_Rec: 0.383 D_GP: 0.024 D_real: 1.238 D_fake: 0.570 +(epoch: 51, iters: 2400, time: 0.064) G_GAN: 0.533 G_GAN_Feat: 0.851 G_ID: 0.265 G_Rec: 0.330 D_GP: 0.033 D_real: 1.379 D_fake: 0.470 +(epoch: 51, iters: 2800, time: 0.064) G_GAN: -0.086 G_GAN_Feat: 0.759 G_ID: 0.498 G_Rec: 0.432 D_GP: 0.047 D_real: 0.652 D_fake: 1.086 +(epoch: 51, iters: 3200, time: 0.064) G_GAN: 0.440 G_GAN_Feat: 0.832 G_ID: 0.222 G_Rec: 0.388 D_GP: 0.041 D_real: 1.068 D_fake: 0.561 +(epoch: 51, iters: 3600, time: 0.064) G_GAN: 0.638 G_GAN_Feat: 0.801 G_ID: 0.501 G_Rec: 0.419 D_GP: 0.059 D_real: 1.270 D_fake: 0.377 +(epoch: 51, iters: 4000, time: 0.064) G_GAN: 0.709 G_GAN_Feat: 0.889 G_ID: 0.224 G_Rec: 0.439 D_GP: 0.080 D_real: 1.144 D_fake: 0.310 +(epoch: 51, iters: 4400, time: 0.064) G_GAN: 0.199 G_GAN_Feat: 0.819 G_ID: 0.439 G_Rec: 0.366 D_GP: 0.064 D_real: 0.936 D_fake: 0.801 +(epoch: 51, iters: 4800, time: 0.064) G_GAN: 0.438 G_GAN_Feat: 0.873 G_ID: 0.238 G_Rec: 0.401 D_GP: 0.037 D_real: 1.138 D_fake: 0.562 +(epoch: 51, iters: 5200, time: 0.064) G_GAN: 0.911 G_GAN_Feat: 0.749 G_ID: 0.470 G_Rec: 0.429 D_GP: 0.026 D_real: 1.645 D_fake: 0.219 +(epoch: 51, iters: 5600, time: 0.064) G_GAN: 0.473 G_GAN_Feat: 0.685 G_ID: 0.221 G_Rec: 0.444 D_GP: 0.024 D_real: 1.258 D_fake: 0.527 +(epoch: 51, iters: 6000, time: 0.064) G_GAN: 0.343 G_GAN_Feat: 0.823 G_ID: 0.550 G_Rec: 0.430 D_GP: 0.040 D_real: 1.012 D_fake: 0.657 +(epoch: 51, iters: 6400, time: 0.064) G_GAN: 0.381 G_GAN_Feat: 0.886 G_ID: 0.248 G_Rec: 0.370 D_GP: 0.107 D_real: 0.782 D_fake: 0.625 +(epoch: 51, iters: 6800, time: 0.064) G_GAN: 0.302 G_GAN_Feat: 1.157 G_ID: 0.510 G_Rec: 0.372 D_GP: 0.621 D_real: 0.273 D_fake: 0.706 +(epoch: 51, iters: 7200, time: 0.064) G_GAN: 0.249 G_GAN_Feat: 1.245 G_ID: 0.248 G_Rec: 0.374 D_GP: 0.592 D_real: 0.356 D_fake: 0.751 +(epoch: 51, iters: 7600, time: 0.064) G_GAN: 0.815 G_GAN_Feat: 0.930 G_ID: 0.514 G_Rec: 0.379 D_GP: 0.039 D_real: 1.413 D_fake: 0.335 +(epoch: 51, iters: 8000, time: 0.064) G_GAN: 0.398 G_GAN_Feat: 0.861 G_ID: 0.319 G_Rec: 0.376 D_GP: 0.032 D_real: 1.088 D_fake: 0.602 +(epoch: 51, iters: 8400, time: 0.064) G_GAN: 0.266 G_GAN_Feat: 1.222 G_ID: 0.542 G_Rec: 0.401 D_GP: 0.142 D_real: 0.348 D_fake: 0.734 +(epoch: 52, iters: 192, time: 0.064) G_GAN: 0.476 G_GAN_Feat: 0.906 G_ID: 0.269 G_Rec: 0.360 D_GP: 0.045 D_real: 1.082 D_fake: 0.525 +(epoch: 52, iters: 592, time: 0.064) G_GAN: 0.391 G_GAN_Feat: 0.888 G_ID: 0.520 G_Rec: 0.377 D_GP: 0.067 D_real: 0.926 D_fake: 0.611 +(epoch: 52, iters: 992, time: 0.064) G_GAN: 0.142 G_GAN_Feat: 1.152 G_ID: 0.283 G_Rec: 0.411 D_GP: 0.137 D_real: 0.351 D_fake: 0.859 +(epoch: 52, iters: 1392, time: 0.064) G_GAN: 0.391 G_GAN_Feat: 1.085 G_ID: 0.502 G_Rec: 0.390 D_GP: 0.085 D_real: 0.651 D_fake: 0.613 +(epoch: 52, iters: 1792, time: 0.064) G_GAN: 0.452 G_GAN_Feat: 0.931 G_ID: 0.246 G_Rec: 0.413 D_GP: 0.028 D_real: 1.224 D_fake: 0.548 +(epoch: 52, iters: 2192, time: 0.064) G_GAN: 0.386 G_GAN_Feat: 0.798 G_ID: 0.542 G_Rec: 0.355 D_GP: 0.025 D_real: 1.119 D_fake: 0.621 +(epoch: 52, iters: 2592, time: 0.064) G_GAN: 0.063 G_GAN_Feat: 0.989 G_ID: 0.267 G_Rec: 0.371 D_GP: 0.116 D_real: 0.407 D_fake: 0.937 +(epoch: 52, iters: 2992, time: 0.064) G_GAN: 0.464 G_GAN_Feat: 0.771 G_ID: 0.517 G_Rec: 0.388 D_GP: 0.026 D_real: 1.271 D_fake: 0.622 +(epoch: 52, iters: 3392, time: 0.064) G_GAN: -0.042 G_GAN_Feat: 0.522 G_ID: 0.267 G_Rec: 0.343 D_GP: 0.022 D_real: 0.939 D_fake: 1.042 +(epoch: 52, iters: 3792, time: 0.064) G_GAN: 0.623 G_GAN_Feat: 0.600 G_ID: 0.545 G_Rec: 0.397 D_GP: 0.018 D_real: 1.488 D_fake: 0.432 +(epoch: 52, iters: 4192, time: 0.064) G_GAN: 0.174 G_GAN_Feat: 0.541 G_ID: 0.216 G_Rec: 0.387 D_GP: 0.020 D_real: 1.044 D_fake: 0.826 +(epoch: 52, iters: 4592, time: 0.064) G_GAN: 0.247 G_GAN_Feat: 0.606 G_ID: 0.476 G_Rec: 0.510 D_GP: 0.026 D_real: 1.064 D_fake: 0.754 +(epoch: 52, iters: 4992, time: 0.064) G_GAN: 0.325 G_GAN_Feat: 0.671 G_ID: 0.267 G_Rec: 0.382 D_GP: 0.032 D_real: 1.036 D_fake: 0.678 +(epoch: 52, iters: 5392, time: 0.064) G_GAN: 0.178 G_GAN_Feat: 0.642 G_ID: 0.547 G_Rec: 0.364 D_GP: 0.044 D_real: 0.896 D_fake: 0.822 +(epoch: 52, iters: 5792, time: 0.064) G_GAN: 0.341 G_GAN_Feat: 0.637 G_ID: 0.230 G_Rec: 0.315 D_GP: 0.045 D_real: 1.062 D_fake: 0.659 +(epoch: 52, iters: 6192, time: 0.064) G_GAN: 0.162 G_GAN_Feat: 0.760 G_ID: 0.512 G_Rec: 0.393 D_GP: 0.051 D_real: 0.845 D_fake: 0.838 +(epoch: 52, iters: 6592, time: 0.064) G_GAN: -0.081 G_GAN_Feat: 0.839 G_ID: 0.228 G_Rec: 0.446 D_GP: 0.080 D_real: 0.308 D_fake: 1.081 +(epoch: 52, iters: 6992, time: 0.064) G_GAN: 0.228 G_GAN_Feat: 0.919 G_ID: 0.532 G_Rec: 0.361 D_GP: 0.069 D_real: 0.669 D_fake: 0.772 +(epoch: 52, iters: 7392, time: 0.064) G_GAN: 0.249 G_GAN_Feat: 0.648 G_ID: 0.265 G_Rec: 0.394 D_GP: 0.025 D_real: 1.012 D_fake: 0.751 +(epoch: 52, iters: 7792, time: 0.064) G_GAN: 0.595 G_GAN_Feat: 0.771 G_ID: 0.502 G_Rec: 0.382 D_GP: 0.055 D_real: 1.312 D_fake: 0.408 +(epoch: 52, iters: 8192, time: 0.064) G_GAN: 0.400 G_GAN_Feat: 0.873 G_ID: 0.254 G_Rec: 0.503 D_GP: 0.039 D_real: 1.159 D_fake: 0.600 +(epoch: 52, iters: 8592, time: 0.064) G_GAN: 0.562 G_GAN_Feat: 0.733 G_ID: 0.488 G_Rec: 0.353 D_GP: 0.044 D_real: 1.305 D_fake: 0.461 +(epoch: 53, iters: 384, time: 0.064) G_GAN: 0.060 G_GAN_Feat: 1.258 G_ID: 0.229 G_Rec: 0.350 D_GP: 0.112 D_real: 0.348 D_fake: 0.940 +(epoch: 53, iters: 784, time: 0.064) G_GAN: 0.480 G_GAN_Feat: 0.709 G_ID: 0.477 G_Rec: 0.430 D_GP: 0.030 D_real: 1.341 D_fake: 0.521 +(epoch: 53, iters: 1184, time: 0.064) G_GAN: 0.406 G_GAN_Feat: 0.882 G_ID: 0.256 G_Rec: 0.395 D_GP: 0.048 D_real: 1.080 D_fake: 0.594 +(epoch: 53, iters: 1584, time: 0.064) G_GAN: 0.336 G_GAN_Feat: 0.823 G_ID: 0.520 G_Rec: 0.407 D_GP: 0.030 D_real: 1.040 D_fake: 0.664 +(epoch: 53, iters: 1984, time: 0.064) G_GAN: 0.271 G_GAN_Feat: 0.684 G_ID: 0.248 G_Rec: 0.419 D_GP: 0.027 D_real: 1.025 D_fake: 0.729 +(epoch: 53, iters: 2384, time: 0.064) G_GAN: 0.290 G_GAN_Feat: 0.832 G_ID: 0.508 G_Rec: 0.368 D_GP: 0.054 D_real: 0.931 D_fake: 0.711 +(epoch: 53, iters: 2784, time: 0.064) G_GAN: 0.417 G_GAN_Feat: 1.146 G_ID: 0.295 G_Rec: 0.393 D_GP: 0.055 D_real: 0.469 D_fake: 0.583 +(epoch: 53, iters: 3184, time: 0.064) G_GAN: 0.725 G_GAN_Feat: 0.985 G_ID: 0.504 G_Rec: 0.372 D_GP: 0.090 D_real: 1.224 D_fake: 0.311 +(epoch: 53, iters: 3584, time: 0.064) G_GAN: 0.573 G_GAN_Feat: 0.745 G_ID: 0.242 G_Rec: 0.395 D_GP: 0.035 D_real: 1.345 D_fake: 0.435 +(epoch: 53, iters: 3984, time: 0.064) G_GAN: 0.735 G_GAN_Feat: 1.027 G_ID: 0.510 G_Rec: 0.418 D_GP: 0.033 D_real: 1.309 D_fake: 0.503 +(epoch: 53, iters: 4384, time: 0.064) G_GAN: 0.383 G_GAN_Feat: 0.953 G_ID: 0.252 G_Rec: 0.383 D_GP: 0.051 D_real: 0.905 D_fake: 0.617 +(epoch: 53, iters: 4784, time: 0.064) G_GAN: 1.063 G_GAN_Feat: 0.949 G_ID: 0.446 G_Rec: 0.391 D_GP: 0.096 D_real: 1.526 D_fake: 0.229 +(epoch: 53, iters: 5184, time: 0.064) G_GAN: 0.582 G_GAN_Feat: 0.802 G_ID: 0.201 G_Rec: 0.426 D_GP: 0.040 D_real: 1.252 D_fake: 0.429 +(epoch: 53, iters: 5584, time: 0.064) G_GAN: 0.428 G_GAN_Feat: 0.726 G_ID: 0.509 G_Rec: 0.349 D_GP: 0.028 D_real: 1.211 D_fake: 0.572 +(epoch: 53, iters: 5984, time: 0.064) G_GAN: 0.382 G_GAN_Feat: 0.874 G_ID: 0.246 G_Rec: 0.405 D_GP: 0.037 D_real: 0.928 D_fake: 0.618 +(epoch: 53, iters: 6384, time: 0.064) G_GAN: 0.406 G_GAN_Feat: 0.832 G_ID: 0.495 G_Rec: 0.351 D_GP: 0.035 D_real: 1.194 D_fake: 0.594 +(epoch: 53, iters: 6784, time: 0.064) G_GAN: 0.811 G_GAN_Feat: 0.852 G_ID: 0.275 G_Rec: 0.361 D_GP: 0.066 D_real: 1.361 D_fake: 0.293 +(epoch: 53, iters: 7184, time: 0.064) G_GAN: 0.206 G_GAN_Feat: 0.768 G_ID: 0.456 G_Rec: 0.363 D_GP: 0.056 D_real: 0.838 D_fake: 0.795 +(epoch: 53, iters: 7584, time: 0.064) G_GAN: 0.155 G_GAN_Feat: 0.725 G_ID: 0.235 G_Rec: 0.460 D_GP: 0.022 D_real: 0.962 D_fake: 0.845 +(epoch: 53, iters: 7984, time: 0.064) G_GAN: 0.044 G_GAN_Feat: 0.971 G_ID: 0.470 G_Rec: 0.385 D_GP: 0.175 D_real: 0.508 D_fake: 0.956 +(epoch: 53, iters: 8384, time: 0.064) G_GAN: 0.371 G_GAN_Feat: 1.015 G_ID: 0.219 G_Rec: 0.455 D_GP: 0.108 D_real: 0.558 D_fake: 0.630 +(epoch: 54, iters: 176, time: 0.064) G_GAN: 0.156 G_GAN_Feat: 0.934 G_ID: 0.483 G_Rec: 0.427 D_GP: 0.032 D_real: 0.833 D_fake: 0.844 +(epoch: 54, iters: 576, time: 0.064) G_GAN: 0.500 G_GAN_Feat: 0.846 G_ID: 0.226 G_Rec: 0.356 D_GP: 0.046 D_real: 1.087 D_fake: 0.501 +(epoch: 54, iters: 976, time: 0.064) G_GAN: 0.187 G_GAN_Feat: 0.828 G_ID: 0.484 G_Rec: 0.412 D_GP: 0.026 D_real: 0.981 D_fake: 0.813 +(epoch: 54, iters: 1376, time: 0.064) G_GAN: 0.522 G_GAN_Feat: 0.666 G_ID: 0.228 G_Rec: 0.345 D_GP: 0.027 D_real: 1.301 D_fake: 0.479 +(epoch: 54, iters: 1776, time: 0.064) G_GAN: 0.497 G_GAN_Feat: 0.856 G_ID: 0.507 G_Rec: 0.419 D_GP: 0.039 D_real: 1.230 D_fake: 0.506 +(epoch: 54, iters: 2176, time: 0.064) G_GAN: 0.377 G_GAN_Feat: 0.843 G_ID: 0.238 G_Rec: 0.408 D_GP: 0.068 D_real: 1.014 D_fake: 0.624 +(epoch: 54, iters: 2576, time: 0.064) G_GAN: 0.407 G_GAN_Feat: 0.688 G_ID: 0.473 G_Rec: 0.426 D_GP: 0.020 D_real: 1.334 D_fake: 0.594 +(epoch: 54, iters: 2976, time: 0.064) G_GAN: 0.271 G_GAN_Feat: 0.656 G_ID: 0.221 G_Rec: 0.390 D_GP: 0.025 D_real: 1.078 D_fake: 0.729 +(epoch: 54, iters: 3376, time: 0.064) G_GAN: 0.282 G_GAN_Feat: 0.745 G_ID: 0.512 G_Rec: 0.398 D_GP: 0.043 D_real: 0.995 D_fake: 0.719 +(epoch: 54, iters: 3776, time: 0.064) G_GAN: 0.283 G_GAN_Feat: 0.791 G_ID: 0.213 G_Rec: 0.384 D_GP: 0.057 D_real: 0.849 D_fake: 0.717 +(epoch: 54, iters: 4176, time: 0.064) G_GAN: 0.342 G_GAN_Feat: 1.036 G_ID: 0.472 G_Rec: 0.432 D_GP: 0.039 D_real: 0.736 D_fake: 0.659 +(epoch: 54, iters: 4576, time: 0.064) G_GAN: 0.423 G_GAN_Feat: 0.872 G_ID: 0.212 G_Rec: 0.389 D_GP: 0.033 D_real: 1.123 D_fake: 0.577 +(epoch: 54, iters: 4976, time: 0.064) G_GAN: 0.284 G_GAN_Feat: 0.910 G_ID: 0.491 G_Rec: 0.376 D_GP: 0.094 D_real: 0.729 D_fake: 0.716 +(epoch: 54, iters: 5376, time: 0.064) G_GAN: 0.148 G_GAN_Feat: 0.558 G_ID: 0.238 G_Rec: 0.344 D_GP: 0.020 D_real: 1.082 D_fake: 0.852 +(epoch: 54, iters: 5776, time: 0.064) G_GAN: 0.261 G_GAN_Feat: 0.680 G_ID: 0.514 G_Rec: 0.369 D_GP: 0.029 D_real: 1.007 D_fake: 0.739 +(epoch: 54, iters: 6176, time: 0.064) G_GAN: -0.017 G_GAN_Feat: 0.749 G_ID: 0.268 G_Rec: 0.467 D_GP: 0.067 D_real: 0.560 D_fake: 1.017 +(epoch: 54, iters: 6576, time: 0.064) G_GAN: -0.172 G_GAN_Feat: 0.793 G_ID: 0.486 G_Rec: 0.363 D_GP: 0.238 D_real: 0.646 D_fake: 1.172 +(epoch: 54, iters: 6976, time: 0.064) G_GAN: -0.296 G_GAN_Feat: 0.828 G_ID: 0.218 G_Rec: 0.425 D_GP: 0.108 D_real: 0.515 D_fake: 1.296 +(epoch: 54, iters: 7376, time: 0.064) G_GAN: 0.755 G_GAN_Feat: 0.721 G_ID: 0.467 G_Rec: 0.364 D_GP: 0.034 D_real: 1.498 D_fake: 0.279 +(epoch: 54, iters: 7776, time: 0.064) G_GAN: 0.317 G_GAN_Feat: 0.761 G_ID: 0.233 G_Rec: 0.391 D_GP: 0.040 D_real: 1.053 D_fake: 0.684 +(epoch: 54, iters: 8176, time: 0.064) G_GAN: 0.462 G_GAN_Feat: 1.009 G_ID: 0.442 G_Rec: 0.413 D_GP: 0.057 D_real: 0.768 D_fake: 0.540 +(epoch: 54, iters: 8576, time: 0.064) G_GAN: 0.334 G_GAN_Feat: 0.666 G_ID: 0.258 G_Rec: 0.378 D_GP: 0.030 D_real: 1.099 D_fake: 0.666 +(epoch: 55, iters: 368, time: 0.064) G_GAN: 0.631 G_GAN_Feat: 0.889 G_ID: 0.476 G_Rec: 0.411 D_GP: 0.060 D_real: 1.222 D_fake: 0.456 +(epoch: 55, iters: 768, time: 0.064) G_GAN: 0.339 G_GAN_Feat: 1.137 G_ID: 0.233 G_Rec: 0.403 D_GP: 0.449 D_real: 0.636 D_fake: 0.669 +(epoch: 55, iters: 1168, time: 0.064) G_GAN: 0.324 G_GAN_Feat: 1.004 G_ID: 0.506 G_Rec: 0.378 D_GP: 0.063 D_real: 0.683 D_fake: 0.676 +(epoch: 55, iters: 1568, time: 0.064) G_GAN: 0.550 G_GAN_Feat: 0.932 G_ID: 0.241 G_Rec: 0.357 D_GP: 0.063 D_real: 0.990 D_fake: 0.468 +(epoch: 55, iters: 1968, time: 0.064) G_GAN: 0.247 G_GAN_Feat: 0.804 G_ID: 0.500 G_Rec: 0.406 D_GP: 0.037 D_real: 0.983 D_fake: 0.754 +(epoch: 55, iters: 2368, time: 0.064) G_GAN: 0.370 G_GAN_Feat: 0.752 G_ID: 0.256 G_Rec: 0.362 D_GP: 0.031 D_real: 1.103 D_fake: 0.630 +(epoch: 55, iters: 2768, time: 0.064) G_GAN: 0.093 G_GAN_Feat: 0.820 G_ID: 0.454 G_Rec: 0.493 D_GP: 0.056 D_real: 0.712 D_fake: 0.907 +(epoch: 55, iters: 3168, time: 0.064) G_GAN: 0.493 G_GAN_Feat: 0.818 G_ID: 0.319 G_Rec: 0.368 D_GP: 0.042 D_real: 1.191 D_fake: 0.508 +(epoch: 55, iters: 3568, time: 0.064) G_GAN: 0.283 G_GAN_Feat: 0.788 G_ID: 0.458 G_Rec: 0.345 D_GP: 0.030 D_real: 0.966 D_fake: 0.717 +(epoch: 55, iters: 3968, time: 0.064) G_GAN: 0.965 G_GAN_Feat: 0.896 G_ID: 0.214 G_Rec: 0.382 D_GP: 0.037 D_real: 1.536 D_fake: 0.297 +(epoch: 55, iters: 4368, time: 0.064) G_GAN: 0.246 G_GAN_Feat: 0.824 G_ID: 0.454 G_Rec: 0.382 D_GP: 0.020 D_real: 1.041 D_fake: 0.755 +(epoch: 55, iters: 4768, time: 0.064) G_GAN: -0.098 G_GAN_Feat: 0.719 G_ID: 0.261 G_Rec: 0.372 D_GP: 0.026 D_real: 0.637 D_fake: 1.098 +(epoch: 55, iters: 5168, time: 0.064) G_GAN: 0.057 G_GAN_Feat: 0.839 G_ID: 0.481 G_Rec: 0.378 D_GP: 0.077 D_real: 0.654 D_fake: 0.943 +(epoch: 55, iters: 5568, time: 0.064) G_GAN: 0.334 G_GAN_Feat: 0.633 G_ID: 0.220 G_Rec: 0.336 D_GP: 0.027 D_real: 1.162 D_fake: 0.666 +(epoch: 55, iters: 5968, time: 0.064) G_GAN: 0.333 G_GAN_Feat: 0.836 G_ID: 0.452 G_Rec: 0.390 D_GP: 0.073 D_real: 0.926 D_fake: 0.672 +(epoch: 55, iters: 6368, time: 0.064) G_GAN: 0.234 G_GAN_Feat: 0.854 G_ID: 0.275 G_Rec: 0.428 D_GP: 0.051 D_real: 0.803 D_fake: 0.766 +(epoch: 55, iters: 6768, time: 0.064) G_GAN: 0.764 G_GAN_Feat: 0.690 G_ID: 0.449 G_Rec: 0.524 D_GP: 0.022 D_real: 1.564 D_fake: 0.344 +(epoch: 55, iters: 7168, time: 0.064) G_GAN: 0.265 G_GAN_Feat: 1.077 G_ID: 0.261 G_Rec: 0.375 D_GP: 0.046 D_real: 0.773 D_fake: 0.735 +(epoch: 55, iters: 7568, time: 0.064) G_GAN: 0.622 G_GAN_Feat: 0.777 G_ID: 0.496 G_Rec: 0.378 D_GP: 0.037 D_real: 1.248 D_fake: 0.382 +(epoch: 55, iters: 7968, time: 0.064) G_GAN: 0.421 G_GAN_Feat: 0.775 G_ID: 0.271 G_Rec: 0.321 D_GP: 0.041 D_real: 1.207 D_fake: 0.581 +(epoch: 55, iters: 8368, time: 0.064) G_GAN: 0.524 G_GAN_Feat: 0.849 G_ID: 0.466 G_Rec: 0.379 D_GP: 0.042 D_real: 1.117 D_fake: 0.479 +(epoch: 56, iters: 160, time: 0.064) G_GAN: 0.246 G_GAN_Feat: 0.951 G_ID: 0.246 G_Rec: 0.399 D_GP: 0.052 D_real: 0.534 D_fake: 0.754 +(epoch: 56, iters: 560, time: 0.064) G_GAN: 0.286 G_GAN_Feat: 0.705 G_ID: 0.496 G_Rec: 0.391 D_GP: 0.029 D_real: 1.185 D_fake: 0.716 +(epoch: 56, iters: 960, time: 0.064) G_GAN: 0.433 G_GAN_Feat: 0.609 G_ID: 0.288 G_Rec: 0.297 D_GP: 0.025 D_real: 1.349 D_fake: 0.569 +(epoch: 56, iters: 1360, time: 0.064) G_GAN: 0.462 G_GAN_Feat: 0.857 G_ID: 0.480 G_Rec: 0.418 D_GP: 0.034 D_real: 1.105 D_fake: 0.538 +(epoch: 56, iters: 1760, time: 0.064) G_GAN: 0.161 G_GAN_Feat: 0.851 G_ID: 0.253 G_Rec: 0.350 D_GP: 0.105 D_real: 0.628 D_fake: 0.839 +(epoch: 56, iters: 2160, time: 0.064) G_GAN: 0.397 G_GAN_Feat: 0.897 G_ID: 0.496 G_Rec: 0.377 D_GP: 0.033 D_real: 1.076 D_fake: 0.603 +(epoch: 56, iters: 2560, time: 0.064) G_GAN: 0.493 G_GAN_Feat: 0.763 G_ID: 0.260 G_Rec: 0.390 D_GP: 0.037 D_real: 1.147 D_fake: 0.513 +(epoch: 56, iters: 2960, time: 0.064) G_GAN: 0.450 G_GAN_Feat: 0.810 G_ID: 0.473 G_Rec: 0.362 D_GP: 0.076 D_real: 1.150 D_fake: 0.551 +(epoch: 56, iters: 3360, time: 0.064) G_GAN: 0.358 G_GAN_Feat: 0.933 G_ID: 0.240 G_Rec: 0.348 D_GP: 0.050 D_real: 0.997 D_fake: 0.642 +(epoch: 56, iters: 3760, time: 0.064) G_GAN: 0.435 G_GAN_Feat: 1.205 G_ID: 0.467 G_Rec: 0.387 D_GP: 0.090 D_real: 0.604 D_fake: 0.565 +(epoch: 56, iters: 4160, time: 0.064) G_GAN: 0.370 G_GAN_Feat: 1.041 G_ID: 0.243 G_Rec: 0.379 D_GP: 0.320 D_real: 0.408 D_fake: 0.662 +(epoch: 56, iters: 4560, time: 0.064) G_GAN: 0.073 G_GAN_Feat: 0.742 G_ID: 0.476 G_Rec: 0.358 D_GP: 0.034 D_real: 0.857 D_fake: 0.927 +(epoch: 56, iters: 4960, time: 0.064) G_GAN: 0.623 G_GAN_Feat: 0.891 G_ID: 0.221 G_Rec: 0.406 D_GP: 0.048 D_real: 1.204 D_fake: 0.392 +(epoch: 56, iters: 5360, time: 0.064) G_GAN: 0.706 G_GAN_Feat: 0.696 G_ID: 0.501 G_Rec: 0.409 D_GP: 0.021 D_real: 1.580 D_fake: 0.307 +(epoch: 56, iters: 5760, time: 0.064) G_GAN: 0.379 G_GAN_Feat: 0.713 G_ID: 0.277 G_Rec: 0.340 D_GP: 0.037 D_real: 1.125 D_fake: 0.627 +(epoch: 56, iters: 6160, time: 0.064) G_GAN: 0.313 G_GAN_Feat: 0.708 G_ID: 0.502 G_Rec: 0.376 D_GP: 0.028 D_real: 1.115 D_fake: 0.687 +(epoch: 56, iters: 6560, time: 0.064) G_GAN: 0.210 G_GAN_Feat: 0.877 G_ID: 0.214 G_Rec: 0.382 D_GP: 0.061 D_real: 0.781 D_fake: 0.790 +(epoch: 56, iters: 6960, time: 0.064) G_GAN: 0.338 G_GAN_Feat: 0.698 G_ID: 0.476 G_Rec: 0.381 D_GP: 0.024 D_real: 1.121 D_fake: 0.662 +(epoch: 56, iters: 7360, time: 0.064) G_GAN: 0.081 G_GAN_Feat: 0.593 G_ID: 0.237 G_Rec: 0.337 D_GP: 0.023 D_real: 0.978 D_fake: 0.919 +(epoch: 56, iters: 7760, time: 0.064) G_GAN: 0.485 G_GAN_Feat: 0.841 G_ID: 0.487 G_Rec: 0.366 D_GP: 0.051 D_real: 1.147 D_fake: 0.517 +(epoch: 56, iters: 8160, time: 0.064) G_GAN: 0.524 G_GAN_Feat: 0.700 G_ID: 0.233 G_Rec: 0.370 D_GP: 0.031 D_real: 1.363 D_fake: 0.485 +(epoch: 56, iters: 8560, time: 0.064) G_GAN: 0.355 G_GAN_Feat: 0.851 G_ID: 0.469 G_Rec: 0.355 D_GP: 0.066 D_real: 0.879 D_fake: 0.649 +(epoch: 57, iters: 352, time: 0.064) G_GAN: 0.377 G_GAN_Feat: 0.775 G_ID: 0.252 G_Rec: 0.406 D_GP: 0.028 D_real: 1.206 D_fake: 0.623 +(epoch: 57, iters: 752, time: 0.064) G_GAN: 0.806 G_GAN_Feat: 0.765 G_ID: 0.488 G_Rec: 0.412 D_GP: 0.027 D_real: 1.577 D_fake: 0.437 +(epoch: 57, iters: 1152, time: 0.064) G_GAN: 0.329 G_GAN_Feat: 0.597 G_ID: 0.238 G_Rec: 0.344 D_GP: 0.027 D_real: 1.162 D_fake: 0.673 +(epoch: 57, iters: 1552, time: 0.064) G_GAN: 0.158 G_GAN_Feat: 0.877 G_ID: 0.454 G_Rec: 0.434 D_GP: 0.092 D_real: 0.692 D_fake: 0.844 +(epoch: 57, iters: 1952, time: 0.064) G_GAN: 0.314 G_GAN_Feat: 0.784 G_ID: 0.266 G_Rec: 0.471 D_GP: 0.037 D_real: 1.076 D_fake: 0.686 +(epoch: 57, iters: 2352, time: 0.064) G_GAN: 0.524 G_GAN_Feat: 0.888 G_ID: 0.455 G_Rec: 0.399 D_GP: 0.054 D_real: 1.139 D_fake: 0.493 +(epoch: 57, iters: 2752, time: 0.064) G_GAN: 0.448 G_GAN_Feat: 0.981 G_ID: 0.279 G_Rec: 0.363 D_GP: 0.070 D_real: 0.671 D_fake: 0.554 +(epoch: 57, iters: 3152, time: 0.064) G_GAN: 0.464 G_GAN_Feat: 0.705 G_ID: 0.461 G_Rec: 0.376 D_GP: 0.021 D_real: 1.272 D_fake: 0.544 +(epoch: 57, iters: 3552, time: 0.064) G_GAN: 0.690 G_GAN_Feat: 0.965 G_ID: 0.244 G_Rec: 0.384 D_GP: 0.045 D_real: 1.383 D_fake: 0.353 +(epoch: 57, iters: 3952, time: 0.064) G_GAN: 0.199 G_GAN_Feat: 1.094 G_ID: 0.490 G_Rec: 0.408 D_GP: 0.142 D_real: 0.583 D_fake: 0.804 +(epoch: 57, iters: 4352, time: 0.064) G_GAN: 0.201 G_GAN_Feat: 0.773 G_ID: 0.258 G_Rec: 0.381 D_GP: 0.028 D_real: 1.019 D_fake: 0.799 +(epoch: 57, iters: 4752, time: 0.064) G_GAN: 0.822 G_GAN_Feat: 0.847 G_ID: 0.456 G_Rec: 0.384 D_GP: 0.033 D_real: 1.484 D_fake: 0.424 +(epoch: 57, iters: 5152, time: 0.064) G_GAN: 0.332 G_GAN_Feat: 0.654 G_ID: 0.266 G_Rec: 0.392 D_GP: 0.027 D_real: 1.097 D_fake: 0.668 +(epoch: 57, iters: 5552, time: 0.064) G_GAN: 0.140 G_GAN_Feat: 0.665 G_ID: 0.523 G_Rec: 0.349 D_GP: 0.026 D_real: 0.929 D_fake: 0.860 +(epoch: 57, iters: 5952, time: 0.064) G_GAN: 0.324 G_GAN_Feat: 0.692 G_ID: 0.214 G_Rec: 0.391 D_GP: 0.033 D_real: 1.165 D_fake: 0.676 +(epoch: 57, iters: 6352, time: 0.064) G_GAN: 0.363 G_GAN_Feat: 0.837 G_ID: 0.498 G_Rec: 0.447 D_GP: 0.057 D_real: 0.957 D_fake: 0.642 +(epoch: 57, iters: 6752, time: 0.064) G_GAN: 0.353 G_GAN_Feat: 0.735 G_ID: 0.268 G_Rec: 0.374 D_GP: 0.027 D_real: 1.131 D_fake: 0.648 +(epoch: 57, iters: 7152, time: 0.064) G_GAN: 0.432 G_GAN_Feat: 0.702 G_ID: 0.467 G_Rec: 0.362 D_GP: 0.030 D_real: 1.305 D_fake: 0.569 +(epoch: 57, iters: 7552, time: 0.064) G_GAN: 0.279 G_GAN_Feat: 0.723 G_ID: 0.228 G_Rec: 0.361 D_GP: 0.039 D_real: 1.097 D_fake: 0.721 +(epoch: 57, iters: 7952, time: 0.064) G_GAN: 0.229 G_GAN_Feat: 0.798 G_ID: 0.509 G_Rec: 0.359 D_GP: 0.037 D_real: 0.925 D_fake: 0.771 +(epoch: 57, iters: 8352, time: 0.064) G_GAN: 0.284 G_GAN_Feat: 0.742 G_ID: 0.237 G_Rec: 0.330 D_GP: 0.053 D_real: 1.033 D_fake: 0.719 +(epoch: 58, iters: 144, time: 0.064) G_GAN: 0.262 G_GAN_Feat: 0.741 G_ID: 0.497 G_Rec: 0.401 D_GP: 0.032 D_real: 1.072 D_fake: 0.738 +(epoch: 58, iters: 544, time: 0.064) G_GAN: 0.463 G_GAN_Feat: 0.781 G_ID: 0.270 G_Rec: 0.371 D_GP: 0.044 D_real: 1.251 D_fake: 0.538 +(epoch: 58, iters: 944, time: 0.064) G_GAN: 0.312 G_GAN_Feat: 0.979 G_ID: 0.483 G_Rec: 0.404 D_GP: 0.277 D_real: 0.531 D_fake: 0.707 +(epoch: 58, iters: 1344, time: 0.064) G_GAN: 0.784 G_GAN_Feat: 0.806 G_ID: 0.226 G_Rec: 0.362 D_GP: 0.042 D_real: 1.398 D_fake: 0.234 +(epoch: 58, iters: 1744, time: 0.064) G_GAN: 0.589 G_GAN_Feat: 0.956 G_ID: 0.467 G_Rec: 0.384 D_GP: 0.096 D_real: 0.961 D_fake: 0.523 +(epoch: 58, iters: 2144, time: 0.064) G_GAN: 0.109 G_GAN_Feat: 0.790 G_ID: 0.231 G_Rec: 0.361 D_GP: 0.034 D_real: 0.724 D_fake: 0.891 +(epoch: 58, iters: 2544, time: 0.064) G_GAN: 0.350 G_GAN_Feat: 1.044 G_ID: 0.491 G_Rec: 0.400 D_GP: 0.072 D_real: 0.454 D_fake: 0.651 +(epoch: 58, iters: 2944, time: 0.064) G_GAN: 0.387 G_GAN_Feat: 0.726 G_ID: 0.256 G_Rec: 0.399 D_GP: 0.029 D_real: 1.203 D_fake: 0.613 +(epoch: 58, iters: 3344, time: 0.064) G_GAN: 0.269 G_GAN_Feat: 1.014 G_ID: 0.441 G_Rec: 0.366 D_GP: 0.098 D_real: 0.556 D_fake: 0.732 +(epoch: 58, iters: 3744, time: 0.064) G_GAN: 0.138 G_GAN_Feat: 0.925 G_ID: 0.244 G_Rec: 0.360 D_GP: 0.049 D_real: 0.547 D_fake: 0.862 +(epoch: 58, iters: 4144, time: 0.064) G_GAN: 0.423 G_GAN_Feat: 1.150 G_ID: 0.494 G_Rec: 0.435 D_GP: 0.149 D_real: 0.474 D_fake: 0.579 +(epoch: 58, iters: 4544, time: 0.064) G_GAN: 0.317 G_GAN_Feat: 0.614 G_ID: 0.222 G_Rec: 0.341 D_GP: 0.025 D_real: 1.124 D_fake: 0.685 +(epoch: 58, iters: 4944, time: 0.064) G_GAN: 0.285 G_GAN_Feat: 0.689 G_ID: 0.467 G_Rec: 0.352 D_GP: 0.032 D_real: 1.091 D_fake: 0.716 +(epoch: 58, iters: 5344, time: 0.064) G_GAN: 0.342 G_GAN_Feat: 0.777 G_ID: 0.258 G_Rec: 0.363 D_GP: 0.036 D_real: 1.025 D_fake: 0.659 +(epoch: 58, iters: 5744, time: 0.064) G_GAN: 0.532 G_GAN_Feat: 1.180 G_ID: 0.483 G_Rec: 0.396 D_GP: 0.150 D_real: 0.681 D_fake: 0.495 +(epoch: 58, iters: 6144, time: 0.064) G_GAN: 0.200 G_GAN_Feat: 0.908 G_ID: 0.221 G_Rec: 0.395 D_GP: 0.100 D_real: 0.607 D_fake: 0.802 +(epoch: 58, iters: 6544, time: 0.064) G_GAN: 0.596 G_GAN_Feat: 0.670 G_ID: 0.469 G_Rec: 0.370 D_GP: 0.021 D_real: 1.468 D_fake: 0.438 +(epoch: 58, iters: 6944, time: 0.064) G_GAN: 0.461 G_GAN_Feat: 0.782 G_ID: 0.210 G_Rec: 0.422 D_GP: 0.034 D_real: 1.301 D_fake: 0.543 +(epoch: 58, iters: 7344, time: 0.064) G_GAN: 0.321 G_GAN_Feat: 0.815 G_ID: 0.478 G_Rec: 0.392 D_GP: 0.090 D_real: 0.893 D_fake: 0.680 +(epoch: 58, iters: 7744, time: 0.064) G_GAN: 0.630 G_GAN_Feat: 0.813 G_ID: 0.248 G_Rec: 0.415 D_GP: 0.047 D_real: 1.299 D_fake: 0.376 +(epoch: 58, iters: 8144, time: 0.064) G_GAN: 0.372 G_GAN_Feat: 0.909 G_ID: 0.511 G_Rec: 0.402 D_GP: 0.166 D_real: 0.651 D_fake: 0.632 +(epoch: 58, iters: 8544, time: 0.064) G_GAN: 0.582 G_GAN_Feat: 1.024 G_ID: 0.230 G_Rec: 0.390 D_GP: 0.095 D_real: 0.733 D_fake: 0.424 +(epoch: 59, iters: 336, time: 0.064) G_GAN: 0.108 G_GAN_Feat: 0.856 G_ID: 0.438 G_Rec: 0.370 D_GP: 0.073 D_real: 0.701 D_fake: 0.892 +(epoch: 59, iters: 736, time: 0.064) G_GAN: 0.830 G_GAN_Feat: 1.335 G_ID: 0.217 G_Rec: 0.391 D_GP: 0.405 D_real: 0.730 D_fake: 0.225 +(epoch: 59, iters: 1136, time: 0.064) G_GAN: 0.309 G_GAN_Feat: 1.002 G_ID: 0.470 G_Rec: 0.388 D_GP: 0.040 D_real: 0.750 D_fake: 0.691 +(epoch: 59, iters: 1536, time: 0.064) G_GAN: 0.451 G_GAN_Feat: 0.772 G_ID: 0.276 G_Rec: 0.356 D_GP: 0.040 D_real: 1.147 D_fake: 0.566 +(epoch: 59, iters: 1936, time: 0.064) G_GAN: 0.661 G_GAN_Feat: 0.642 G_ID: 0.496 G_Rec: 0.400 D_GP: 0.020 D_real: 1.477 D_fake: 0.397 +(epoch: 59, iters: 2336, time: 0.064) G_GAN: 0.256 G_GAN_Feat: 0.614 G_ID: 0.252 G_Rec: 0.420 D_GP: 0.028 D_real: 1.127 D_fake: 0.744 +(epoch: 59, iters: 2736, time: 0.064) G_GAN: 0.449 G_GAN_Feat: 0.860 G_ID: 0.483 G_Rec: 0.432 D_GP: 0.065 D_real: 0.914 D_fake: 0.553 +(epoch: 59, iters: 3136, time: 0.064) G_GAN: 0.362 G_GAN_Feat: 0.835 G_ID: 0.195 G_Rec: 0.399 D_GP: 0.043 D_real: 0.960 D_fake: 0.638 +(epoch: 59, iters: 3536, time: 0.064) G_GAN: 0.077 G_GAN_Feat: 0.813 G_ID: 0.517 G_Rec: 0.365 D_GP: 0.060 D_real: 0.814 D_fake: 0.923 +(epoch: 59, iters: 3936, time: 0.064) G_GAN: 0.144 G_GAN_Feat: 0.841 G_ID: 0.269 G_Rec: 0.458 D_GP: 0.025 D_real: 0.792 D_fake: 0.856 +(epoch: 59, iters: 4336, time: 0.064) G_GAN: 0.270 G_GAN_Feat: 0.871 G_ID: 0.469 G_Rec: 0.341 D_GP: 0.083 D_real: 0.782 D_fake: 0.730 +(epoch: 59, iters: 4736, time: 0.064) G_GAN: 0.271 G_GAN_Feat: 0.923 G_ID: 0.278 G_Rec: 0.406 D_GP: 0.069 D_real: 0.614 D_fake: 0.729 +(epoch: 59, iters: 5136, time: 0.064) G_GAN: 0.414 G_GAN_Feat: 0.956 G_ID: 0.455 G_Rec: 0.352 D_GP: 0.116 D_real: 0.773 D_fake: 0.587 +(epoch: 59, iters: 5536, time: 0.064) G_GAN: 0.496 G_GAN_Feat: 0.544 G_ID: 0.237 G_Rec: 0.378 D_GP: 0.015 D_real: 1.431 D_fake: 0.514 +(epoch: 59, iters: 5936, time: 0.064) G_GAN: 0.439 G_GAN_Feat: 0.585 G_ID: 0.486 G_Rec: 0.371 D_GP: 0.018 D_real: 1.286 D_fake: 0.568 +(epoch: 59, iters: 6336, time: 0.064) G_GAN: 0.292 G_GAN_Feat: 0.584 G_ID: 0.183 G_Rec: 0.385 D_GP: 0.022 D_real: 1.267 D_fake: 0.710 +(epoch: 59, iters: 6736, time: 0.064) G_GAN: 0.164 G_GAN_Feat: 0.687 G_ID: 0.457 G_Rec: 0.370 D_GP: 0.045 D_real: 0.925 D_fake: 0.836 +(epoch: 59, iters: 7136, time: 0.064) G_GAN: 0.283 G_GAN_Feat: 0.697 G_ID: 0.235 G_Rec: 0.372 D_GP: 0.039 D_real: 1.043 D_fake: 0.718 +(epoch: 59, iters: 7536, time: 0.064) G_GAN: 0.361 G_GAN_Feat: 0.737 G_ID: 0.467 G_Rec: 0.408 D_GP: 0.075 D_real: 0.964 D_fake: 0.639 +(epoch: 59, iters: 7936, time: 0.064) G_GAN: 0.184 G_GAN_Feat: 0.751 G_ID: 0.257 G_Rec: 0.384 D_GP: 0.070 D_real: 0.729 D_fake: 0.816 +(epoch: 59, iters: 8336, time: 0.064) G_GAN: 0.525 G_GAN_Feat: 0.934 G_ID: 0.494 G_Rec: 0.382 D_GP: 0.080 D_real: 0.893 D_fake: 0.493 +(epoch: 60, iters: 128, time: 0.064) G_GAN: 0.441 G_GAN_Feat: 1.045 G_ID: 0.268 G_Rec: 0.377 D_GP: 0.060 D_real: 0.693 D_fake: 0.564 +(epoch: 60, iters: 528, time: 0.064) G_GAN: 0.290 G_GAN_Feat: 0.748 G_ID: 0.481 G_Rec: 0.383 D_GP: 0.032 D_real: 1.023 D_fake: 0.710 +(epoch: 60, iters: 928, time: 0.064) G_GAN: 0.525 G_GAN_Feat: 0.688 G_ID: 0.264 G_Rec: 0.348 D_GP: 0.030 D_real: 1.275 D_fake: 0.477 +(epoch: 60, iters: 1328, time: 0.064) G_GAN: 0.729 G_GAN_Feat: 0.968 G_ID: 0.465 G_Rec: 0.371 D_GP: 0.120 D_real: 1.170 D_fake: 0.491 +(epoch: 60, iters: 1728, time: 0.064) G_GAN: 0.258 G_GAN_Feat: 0.698 G_ID: 0.278 G_Rec: 0.368 D_GP: 0.024 D_real: 1.142 D_fake: 0.742 +(epoch: 60, iters: 2128, time: 0.064) G_GAN: 0.008 G_GAN_Feat: 0.700 G_ID: 0.448 G_Rec: 0.385 D_GP: 0.044 D_real: 0.855 D_fake: 0.992 +(epoch: 60, iters: 2528, time: 0.064) G_GAN: 0.156 G_GAN_Feat: 0.766 G_ID: 0.225 G_Rec: 0.386 D_GP: 0.062 D_real: 0.794 D_fake: 0.844 +(epoch: 60, iters: 2928, time: 0.064) G_GAN: 0.171 G_GAN_Feat: 1.343 G_ID: 0.499 G_Rec: 0.401 D_GP: 0.776 D_real: 0.430 D_fake: 0.829 +(epoch: 60, iters: 3328, time: 0.064) G_GAN: 0.309 G_GAN_Feat: 0.649 G_ID: 0.250 G_Rec: 0.387 D_GP: 0.024 D_real: 1.060 D_fake: 0.691 +(epoch: 60, iters: 3728, time: 0.064) G_GAN: 0.167 G_GAN_Feat: 0.725 G_ID: 0.495 G_Rec: 0.378 D_GP: 0.043 D_real: 0.890 D_fake: 0.833 +(epoch: 60, iters: 4128, time: 0.064) G_GAN: 0.266 G_GAN_Feat: 0.725 G_ID: 0.260 G_Rec: 0.395 D_GP: 0.042 D_real: 0.992 D_fake: 0.735 +(epoch: 60, iters: 4528, time: 0.064) G_GAN: 0.078 G_GAN_Feat: 0.778 G_ID: 0.432 G_Rec: 0.372 D_GP: 0.056 D_real: 0.739 D_fake: 0.922 +(epoch: 60, iters: 4928, time: 0.064) G_GAN: 0.257 G_GAN_Feat: 0.912 G_ID: 0.260 G_Rec: 0.385 D_GP: 0.058 D_real: 0.741 D_fake: 0.743 +(epoch: 60, iters: 5328, time: 0.064) G_GAN: 0.342 G_GAN_Feat: 0.748 G_ID: 0.465 G_Rec: 0.349 D_GP: 0.047 D_real: 0.992 D_fake: 0.658 +(epoch: 60, iters: 5728, time: 0.064) G_GAN: 0.660 G_GAN_Feat: 0.874 G_ID: 0.260 G_Rec: 0.373 D_GP: 0.081 D_real: 1.166 D_fake: 0.396 +(epoch: 60, iters: 6128, time: 0.064) G_GAN: 0.352 G_GAN_Feat: 1.213 G_ID: 0.454 G_Rec: 0.430 D_GP: 0.837 D_real: 0.389 D_fake: 0.671 +(epoch: 60, iters: 6528, time: 0.064) G_GAN: 0.193 G_GAN_Feat: 0.688 G_ID: 0.189 G_Rec: 0.352 D_GP: 0.031 D_real: 1.111 D_fake: 0.807 +(epoch: 60, iters: 6928, time: 0.064) G_GAN: 0.364 G_GAN_Feat: 1.018 G_ID: 0.407 G_Rec: 0.411 D_GP: 0.262 D_real: 0.478 D_fake: 0.637 +(epoch: 60, iters: 7328, time: 0.064) G_GAN: 0.084 G_GAN_Feat: 0.968 G_ID: 0.238 G_Rec: 0.375 D_GP: 0.120 D_real: 0.415 D_fake: 0.917 +(epoch: 60, iters: 7728, time: 0.064) G_GAN: 0.240 G_GAN_Feat: 0.730 G_ID: 0.469 G_Rec: 0.421 D_GP: 0.033 D_real: 0.981 D_fake: 0.760 +(epoch: 60, iters: 8128, time: 0.064) G_GAN: 0.386 G_GAN_Feat: 0.890 G_ID: 0.241 G_Rec: 0.375 D_GP: 0.031 D_real: 1.034 D_fake: 0.617 +(epoch: 60, iters: 8528, time: 0.064) G_GAN: 0.115 G_GAN_Feat: 1.101 G_ID: 0.490 G_Rec: 0.412 D_GP: 0.079 D_real: 0.454 D_fake: 0.885 +(epoch: 61, iters: 320, time: 0.064) G_GAN: 0.288 G_GAN_Feat: 0.981 G_ID: 0.244 G_Rec: 0.398 D_GP: 0.156 D_real: 0.734 D_fake: 0.712 +(epoch: 61, iters: 720, time: 0.064) G_GAN: 0.652 G_GAN_Feat: 0.814 G_ID: 0.446 G_Rec: 0.360 D_GP: 0.051 D_real: 1.321 D_fake: 0.358 +(epoch: 61, iters: 1120, time: 0.064) G_GAN: 0.449 G_GAN_Feat: 0.930 G_ID: 0.227 G_Rec: 0.380 D_GP: 0.043 D_real: 0.886 D_fake: 0.552 +(epoch: 61, iters: 1520, time: 0.064) G_GAN: 0.320 G_GAN_Feat: 1.078 G_ID: 0.460 G_Rec: 0.404 D_GP: 0.064 D_real: 0.660 D_fake: 0.680 +(epoch: 61, iters: 1920, time: 0.064) G_GAN: 0.376 G_GAN_Feat: 0.733 G_ID: 0.240 G_Rec: 0.372 D_GP: 0.031 D_real: 1.115 D_fake: 0.625 +(epoch: 61, iters: 2320, time: 0.064) G_GAN: 0.491 G_GAN_Feat: 0.634 G_ID: 0.428 G_Rec: 0.400 D_GP: 0.022 D_real: 1.347 D_fake: 0.512 +(epoch: 61, iters: 2720, time: 0.064) G_GAN: 0.220 G_GAN_Feat: 0.654 G_ID: 0.244 G_Rec: 0.406 D_GP: 0.028 D_real: 0.992 D_fake: 0.780 +(epoch: 61, iters: 3120, time: 0.064) G_GAN: 0.563 G_GAN_Feat: 0.851 G_ID: 0.464 G_Rec: 0.389 D_GP: 0.081 D_real: 1.056 D_fake: 0.445 +(epoch: 61, iters: 3520, time: 0.064) G_GAN: 0.435 G_GAN_Feat: 0.730 G_ID: 0.198 G_Rec: 0.373 D_GP: 0.038 D_real: 1.122 D_fake: 0.566 +(epoch: 61, iters: 3920, time: 0.064) G_GAN: 0.132 G_GAN_Feat: 0.932 G_ID: 0.492 G_Rec: 0.396 D_GP: 0.075 D_real: 0.498 D_fake: 0.868 +(epoch: 61, iters: 4320, time: 0.064) G_GAN: 0.232 G_GAN_Feat: 0.657 G_ID: 0.257 G_Rec: 0.376 D_GP: 0.025 D_real: 1.068 D_fake: 0.768 +(epoch: 61, iters: 4720, time: 0.064) G_GAN: 0.263 G_GAN_Feat: 0.705 G_ID: 0.456 G_Rec: 0.372 D_GP: 0.036 D_real: 0.981 D_fake: 0.737 +(epoch: 61, iters: 5120, time: 0.064) G_GAN: 0.208 G_GAN_Feat: 0.672 G_ID: 0.229 G_Rec: 0.333 D_GP: 0.042 D_real: 0.965 D_fake: 0.792 +(epoch: 61, iters: 5520, time: 0.064) G_GAN: 0.326 G_GAN_Feat: 0.858 G_ID: 0.496 G_Rec: 0.376 D_GP: 0.041 D_real: 0.915 D_fake: 0.678 +(epoch: 61, iters: 5920, time: 0.064) G_GAN: 0.343 G_GAN_Feat: 0.797 G_ID: 0.246 G_Rec: 0.368 D_GP: 0.051 D_real: 1.193 D_fake: 0.659 +(epoch: 61, iters: 6320, time: 0.064) G_GAN: 0.415 G_GAN_Feat: 0.821 G_ID: 0.429 G_Rec: 0.362 D_GP: 0.105 D_real: 0.963 D_fake: 0.591 +(epoch: 61, iters: 6720, time: 0.064) G_GAN: 0.209 G_GAN_Feat: 0.624 G_ID: 0.236 G_Rec: 0.447 D_GP: 0.020 D_real: 1.115 D_fake: 0.791 +(epoch: 61, iters: 7120, time: 0.064) G_GAN: 0.307 G_GAN_Feat: 0.862 G_ID: 0.431 G_Rec: 0.401 D_GP: 0.040 D_real: 1.003 D_fake: 0.693 +(epoch: 61, iters: 7520, time: 0.064) G_GAN: 0.199 G_GAN_Feat: 0.816 G_ID: 0.265 G_Rec: 0.391 D_GP: 0.052 D_real: 0.829 D_fake: 0.801 +(epoch: 61, iters: 7920, time: 0.064) G_GAN: 0.394 G_GAN_Feat: 0.830 G_ID: 0.435 G_Rec: 0.403 D_GP: 0.046 D_real: 1.099 D_fake: 0.607 +(epoch: 61, iters: 8320, time: 0.064) G_GAN: 0.470 G_GAN_Feat: 0.767 G_ID: 0.290 G_Rec: 0.366 D_GP: 0.048 D_real: 1.142 D_fake: 0.547 +(epoch: 62, iters: 112, time: 0.064) G_GAN: 0.370 G_GAN_Feat: 0.840 G_ID: 0.480 G_Rec: 0.424 D_GP: 0.037 D_real: 1.037 D_fake: 0.639 +(epoch: 62, iters: 512, time: 0.064) G_GAN: 0.293 G_GAN_Feat: 0.730 G_ID: 0.221 G_Rec: 0.325 D_GP: 0.042 D_real: 1.127 D_fake: 0.709 +(epoch: 62, iters: 912, time: 0.064) G_GAN: 0.222 G_GAN_Feat: 0.862 G_ID: 0.446 G_Rec: 0.405 D_GP: 0.034 D_real: 0.864 D_fake: 0.781 +(epoch: 62, iters: 1312, time: 0.064) G_GAN: 0.337 G_GAN_Feat: 0.883 G_ID: 0.253 G_Rec: 0.395 D_GP: 0.030 D_real: 1.030 D_fake: 0.663 +(epoch: 62, iters: 1712, time: 0.064) G_GAN: 0.194 G_GAN_Feat: 0.890 G_ID: 0.439 G_Rec: 0.363 D_GP: 0.085 D_real: 0.642 D_fake: 0.808 +(epoch: 62, iters: 2112, time: 0.064) G_GAN: 0.658 G_GAN_Feat: 1.064 G_ID: 0.250 G_Rec: 0.397 D_GP: 0.060 D_real: 1.323 D_fake: 0.558 +(epoch: 62, iters: 2512, time: 0.064) G_GAN: 0.366 G_GAN_Feat: 1.194 G_ID: 0.480 G_Rec: 0.424 D_GP: 0.146 D_real: 0.494 D_fake: 0.634 +(epoch: 62, iters: 2912, time: 0.064) G_GAN: 0.115 G_GAN_Feat: 1.057 G_ID: 0.220 G_Rec: 0.443 D_GP: 0.105 D_real: 0.243 D_fake: 0.885 +(epoch: 62, iters: 3312, time: 0.064) G_GAN: 0.284 G_GAN_Feat: 0.762 G_ID: 0.498 G_Rec: 0.390 D_GP: 0.024 D_real: 1.044 D_fake: 0.716 +(epoch: 62, iters: 3712, time: 0.064) G_GAN: 0.453 G_GAN_Feat: 0.746 G_ID: 0.240 G_Rec: 0.358 D_GP: 0.027 D_real: 1.174 D_fake: 0.555 +(epoch: 62, iters: 4112, time: 0.064) G_GAN: 0.550 G_GAN_Feat: 0.896 G_ID: 0.445 G_Rec: 0.362 D_GP: 0.051 D_real: 1.118 D_fake: 0.451 +(epoch: 62, iters: 4512, time: 0.064) G_GAN: 0.345 G_GAN_Feat: 0.685 G_ID: 0.290 G_Rec: 0.372 D_GP: 0.022 D_real: 1.123 D_fake: 0.655 +(epoch: 62, iters: 4912, time: 0.064) G_GAN: 0.088 G_GAN_Feat: 0.815 G_ID: 0.432 G_Rec: 0.387 D_GP: 0.101 D_real: 0.593 D_fake: 0.912 +(epoch: 62, iters: 5312, time: 0.064) G_GAN: 0.572 G_GAN_Feat: 0.715 G_ID: 0.223 G_Rec: 0.362 D_GP: 0.035 D_real: 1.371 D_fake: 0.430 +(epoch: 62, iters: 5712, time: 0.064) G_GAN: 0.314 G_GAN_Feat: 1.145 G_ID: 0.411 G_Rec: 0.434 D_GP: 0.106 D_real: 0.491 D_fake: 0.686 +(epoch: 62, iters: 6112, time: 0.064) G_GAN: 0.351 G_GAN_Feat: 0.857 G_ID: 0.220 G_Rec: 0.481 D_GP: 0.056 D_real: 0.908 D_fake: 0.649 +(epoch: 62, iters: 6512, time: 0.064) G_GAN: 0.664 G_GAN_Feat: 1.151 G_ID: 0.440 G_Rec: 0.405 D_GP: 0.298 D_real: 0.664 D_fake: 0.386 +(epoch: 62, iters: 6912, time: 0.064) G_GAN: 0.416 G_GAN_Feat: 1.093 G_ID: 0.232 G_Rec: 0.438 D_GP: 0.096 D_real: 0.467 D_fake: 0.586 +(epoch: 62, iters: 7312, time: 0.064) G_GAN: 0.239 G_GAN_Feat: 1.073 G_ID: 0.440 G_Rec: 0.402 D_GP: 0.055 D_real: 0.539 D_fake: 0.762 +(epoch: 62, iters: 7712, time: 0.064) G_GAN: 0.685 G_GAN_Feat: 0.764 G_ID: 0.218 G_Rec: 0.331 D_GP: 0.047 D_real: 1.400 D_fake: 0.353 +(epoch: 62, iters: 8112, time: 0.064) G_GAN: 0.232 G_GAN_Feat: 0.674 G_ID: 0.446 G_Rec: 0.405 D_GP: 0.024 D_real: 1.056 D_fake: 0.768 +(epoch: 62, iters: 8512, time: 0.064) G_GAN: 0.299 G_GAN_Feat: 0.842 G_ID: 0.208 G_Rec: 0.365 D_GP: 0.057 D_real: 0.755 D_fake: 0.702 +(epoch: 63, iters: 304, time: 0.064) G_GAN: 0.417 G_GAN_Feat: 0.654 G_ID: 0.442 G_Rec: 0.382 D_GP: 0.027 D_real: 1.212 D_fake: 0.584 +(epoch: 63, iters: 704, time: 0.064) G_GAN: 0.276 G_GAN_Feat: 0.847 G_ID: 0.248 G_Rec: 0.380 D_GP: 0.080 D_real: 0.780 D_fake: 0.725 +(epoch: 63, iters: 1104, time: 0.064) G_GAN: 0.267 G_GAN_Feat: 0.805 G_ID: 0.460 G_Rec: 0.380 D_GP: 0.035 D_real: 0.929 D_fake: 0.734 +(epoch: 63, iters: 1504, time: 0.064) G_GAN: 0.415 G_GAN_Feat: 0.626 G_ID: 0.308 G_Rec: 0.348 D_GP: 0.024 D_real: 1.162 D_fake: 0.591 +(epoch: 63, iters: 1904, time: 0.064) G_GAN: 0.251 G_GAN_Feat: 0.659 G_ID: 0.427 G_Rec: 0.358 D_GP: 0.031 D_real: 1.043 D_fake: 0.749 +(epoch: 63, iters: 2304, time: 0.064) G_GAN: 0.000 G_GAN_Feat: 0.738 G_ID: 0.224 G_Rec: 0.334 D_GP: 0.050 D_real: 0.850 D_fake: 1.000 +(epoch: 63, iters: 2704, time: 0.064) G_GAN: 0.038 G_GAN_Feat: 0.793 G_ID: 0.426 G_Rec: 0.370 D_GP: 0.060 D_real: 0.714 D_fake: 0.962 +(epoch: 63, iters: 3104, time: 0.064) G_GAN: 0.463 G_GAN_Feat: 0.750 G_ID: 0.228 G_Rec: 0.369 D_GP: 0.039 D_real: 1.154 D_fake: 0.537 +(epoch: 63, iters: 3504, time: 0.064) G_GAN: 0.620 G_GAN_Feat: 0.893 G_ID: 0.414 G_Rec: 0.411 D_GP: 0.077 D_real: 1.113 D_fake: 0.414 +(epoch: 63, iters: 3904, time: 0.064) G_GAN: 0.411 G_GAN_Feat: 0.994 G_ID: 0.195 G_Rec: 0.375 D_GP: 0.042 D_real: 0.785 D_fake: 0.590 +(epoch: 63, iters: 4304, time: 0.064) G_GAN: 0.423 G_GAN_Feat: 0.833 G_ID: 0.513 G_Rec: 0.369 D_GP: 0.044 D_real: 1.149 D_fake: 0.577 +(epoch: 63, iters: 4704, time: 0.064) G_GAN: 0.444 G_GAN_Feat: 0.849 G_ID: 0.200 G_Rec: 0.361 D_GP: 0.089 D_real: 0.910 D_fake: 0.632 +(epoch: 63, iters: 5104, time: 0.064) G_GAN: 0.398 G_GAN_Feat: 1.236 G_ID: 0.482 G_Rec: 0.361 D_GP: 0.070 D_real: 0.429 D_fake: 0.604 +(epoch: 63, iters: 5504, time: 0.064) G_GAN: 0.107 G_GAN_Feat: 0.718 G_ID: 0.212 G_Rec: 0.382 D_GP: 0.022 D_real: 0.924 D_fake: 0.893 +(epoch: 63, iters: 5904, time: 0.064) G_GAN: 0.502 G_GAN_Feat: 0.759 G_ID: 0.487 G_Rec: 0.363 D_GP: 0.036 D_real: 1.212 D_fake: 0.515 +(epoch: 63, iters: 6304, time: 0.064) G_GAN: 0.423 G_GAN_Feat: 0.840 G_ID: 0.233 G_Rec: 0.417 D_GP: 0.030 D_real: 1.122 D_fake: 0.578 +(epoch: 63, iters: 6704, time: 0.064) G_GAN: 0.303 G_GAN_Feat: 0.686 G_ID: 0.428 G_Rec: 0.425 D_GP: 0.022 D_real: 1.167 D_fake: 0.697 +(epoch: 63, iters: 7104, time: 0.064) G_GAN: 0.503 G_GAN_Feat: 0.703 G_ID: 0.249 G_Rec: 0.352 D_GP: 0.027 D_real: 1.262 D_fake: 0.505 +(epoch: 63, iters: 7504, time: 0.064) G_GAN: 0.550 G_GAN_Feat: 0.698 G_ID: 0.432 G_Rec: 0.338 D_GP: 0.032 D_real: 1.329 D_fake: 0.453 +(epoch: 63, iters: 7904, time: 0.064) G_GAN: 0.733 G_GAN_Feat: 0.864 G_ID: 0.256 G_Rec: 0.394 D_GP: 0.085 D_real: 1.191 D_fake: 0.300 +(epoch: 63, iters: 8304, time: 0.064) G_GAN: 0.439 G_GAN_Feat: 0.931 G_ID: 0.447 G_Rec: 0.406 D_GP: 0.041 D_real: 0.968 D_fake: 0.563 +(epoch: 64, iters: 96, time: 0.064) G_GAN: -0.106 G_GAN_Feat: 1.209 G_ID: 0.261 G_Rec: 0.390 D_GP: 0.300 D_real: 0.335 D_fake: 1.106 +(epoch: 64, iters: 496, time: 0.064) G_GAN: 0.727 G_GAN_Feat: 0.886 G_ID: 0.450 G_Rec: 0.402 D_GP: 0.033 D_real: 1.333 D_fake: 0.296 +(epoch: 64, iters: 896, time: 0.064) G_GAN: 0.852 G_GAN_Feat: 0.846 G_ID: 0.240 G_Rec: 0.372 D_GP: 0.074 D_real: 1.486 D_fake: 0.315 +(epoch: 64, iters: 1296, time: 0.064) G_GAN: 0.380 G_GAN_Feat: 1.165 G_ID: 0.482 G_Rec: 0.389 D_GP: 0.086 D_real: 0.376 D_fake: 0.620 +(epoch: 64, iters: 1696, time: 0.064) G_GAN: 0.166 G_GAN_Feat: 0.721 G_ID: 0.239 G_Rec: 0.399 D_GP: 0.021 D_real: 1.002 D_fake: 0.835 +(epoch: 64, iters: 2096, time: 0.064) G_GAN: 0.291 G_GAN_Feat: 0.726 G_ID: 0.468 G_Rec: 0.402 D_GP: 0.027 D_real: 0.989 D_fake: 0.717 +(epoch: 64, iters: 2496, time: 0.064) G_GAN: 0.108 G_GAN_Feat: 0.682 G_ID: 0.241 G_Rec: 0.332 D_GP: 0.033 D_real: 0.925 D_fake: 0.892 +(epoch: 64, iters: 2896, time: 0.064) G_GAN: 0.312 G_GAN_Feat: 0.837 G_ID: 0.462 G_Rec: 0.386 D_GP: 0.071 D_real: 0.926 D_fake: 0.701 +(epoch: 64, iters: 3296, time: 0.064) G_GAN: 0.481 G_GAN_Feat: 0.731 G_ID: 0.219 G_Rec: 0.409 D_GP: 0.037 D_real: 1.196 D_fake: 0.519 +(epoch: 64, iters: 3696, time: 0.064) G_GAN: 0.747 G_GAN_Feat: 0.978 G_ID: 0.462 G_Rec: 0.406 D_GP: 0.098 D_real: 0.972 D_fake: 0.351 +(epoch: 64, iters: 4096, time: 0.064) G_GAN: 0.226 G_GAN_Feat: 0.877 G_ID: 0.241 G_Rec: 0.426 D_GP: 0.040 D_real: 0.923 D_fake: 0.774 +(epoch: 64, iters: 4496, time: 0.064) G_GAN: 0.518 G_GAN_Feat: 0.881 G_ID: 0.475 G_Rec: 0.347 D_GP: 0.092 D_real: 0.973 D_fake: 0.483 +(epoch: 64, iters: 4896, time: 0.064) G_GAN: 0.287 G_GAN_Feat: 0.573 G_ID: 0.221 G_Rec: 0.406 D_GP: 0.023 D_real: 1.194 D_fake: 0.714 +(epoch: 64, iters: 5296, time: 0.064) G_GAN: 0.371 G_GAN_Feat: 0.655 G_ID: 0.435 G_Rec: 0.371 D_GP: 0.021 D_real: 1.159 D_fake: 0.630 +(epoch: 64, iters: 5696, time: 0.064) G_GAN: 0.178 G_GAN_Feat: 0.611 G_ID: 0.225 G_Rec: 0.391 D_GP: 0.027 D_real: 1.039 D_fake: 0.822 +(epoch: 64, iters: 6096, time: 0.064) G_GAN: 0.258 G_GAN_Feat: 0.607 G_ID: 0.436 G_Rec: 0.352 D_GP: 0.029 D_real: 1.069 D_fake: 0.743 +(epoch: 64, iters: 6496, time: 0.064) G_GAN: 0.287 G_GAN_Feat: 0.674 G_ID: 0.215 G_Rec: 0.446 D_GP: 0.043 D_real: 1.021 D_fake: 0.714 +(epoch: 64, iters: 6896, time: 0.064) G_GAN: 0.250 G_GAN_Feat: 0.759 G_ID: 0.426 G_Rec: 0.389 D_GP: 0.044 D_real: 0.893 D_fake: 0.750 +(epoch: 64, iters: 7296, time: 0.064) G_GAN: 0.285 G_GAN_Feat: 0.920 G_ID: 0.255 G_Rec: 0.346 D_GP: 0.126 D_real: 0.575 D_fake: 0.715 +(epoch: 64, iters: 7696, time: 0.064) G_GAN: 0.567 G_GAN_Feat: 0.734 G_ID: 0.446 G_Rec: 0.357 D_GP: 0.053 D_real: 1.260 D_fake: 0.438 +(epoch: 64, iters: 8096, time: 0.064) G_GAN: 0.603 G_GAN_Feat: 0.864 G_ID: 0.236 G_Rec: 0.372 D_GP: 0.068 D_real: 1.199 D_fake: 0.445 +(epoch: 64, iters: 8496, time: 0.064) G_GAN: 0.331 G_GAN_Feat: 0.779 G_ID: 0.483 G_Rec: 0.384 D_GP: 0.039 D_real: 1.068 D_fake: 0.669 +(epoch: 65, iters: 288, time: 0.064) G_GAN: 0.787 G_GAN_Feat: 0.860 G_ID: 0.233 G_Rec: 0.358 D_GP: 0.060 D_real: 1.353 D_fake: 0.409 +(epoch: 65, iters: 688, time: 0.064) G_GAN: 0.621 G_GAN_Feat: 0.974 G_ID: 0.463 G_Rec: 0.374 D_GP: 0.056 D_real: 0.990 D_fake: 0.416 +(epoch: 65, iters: 1088, time: 0.064) G_GAN: 0.330 G_GAN_Feat: 0.739 G_ID: 0.221 G_Rec: 0.425 D_GP: 0.032 D_real: 1.264 D_fake: 0.671 +(epoch: 65, iters: 1488, time: 0.064) G_GAN: 0.273 G_GAN_Feat: 0.718 G_ID: 0.415 G_Rec: 0.369 D_GP: 0.026 D_real: 1.063 D_fake: 0.728 +(epoch: 65, iters: 1888, time: 0.064) G_GAN: 0.433 G_GAN_Feat: 1.117 G_ID: 0.205 G_Rec: 0.384 D_GP: 0.212 D_real: 0.556 D_fake: 0.569 +(epoch: 65, iters: 2288, time: 0.064) G_GAN: 0.066 G_GAN_Feat: 0.964 G_ID: 0.473 G_Rec: 0.425 D_GP: 0.105 D_real: 0.289 D_fake: 0.934 +(epoch: 65, iters: 2688, time: 0.064) G_GAN: 0.206 G_GAN_Feat: 0.835 G_ID: 0.255 G_Rec: 0.396 D_GP: 0.065 D_real: 0.688 D_fake: 0.794 +(epoch: 65, iters: 3088, time: 0.064) G_GAN: 0.726 G_GAN_Feat: 0.769 G_ID: 0.416 G_Rec: 0.414 D_GP: 0.034 D_real: 1.426 D_fake: 0.310 +(epoch: 65, iters: 3488, time: 0.064) G_GAN: 0.419 G_GAN_Feat: 0.661 G_ID: 0.195 G_Rec: 0.358 D_GP: 0.026 D_real: 1.282 D_fake: 0.581 +(epoch: 65, iters: 3888, time: 0.064) G_GAN: 0.933 G_GAN_Feat: 1.078 G_ID: 0.378 G_Rec: 0.403 D_GP: 0.082 D_real: 1.074 D_fake: 0.200 +(epoch: 65, iters: 4288, time: 0.064) G_GAN: 0.209 G_GAN_Feat: 0.991 G_ID: 0.229 G_Rec: 0.375 D_GP: 0.152 D_real: 0.367 D_fake: 0.791 +(epoch: 65, iters: 4688, time: 0.064) G_GAN: 0.314 G_GAN_Feat: 0.759 G_ID: 0.438 G_Rec: 0.458 D_GP: 0.042 D_real: 0.967 D_fake: 0.686 +(epoch: 65, iters: 5088, time: 0.064) G_GAN: 0.432 G_GAN_Feat: 0.768 G_ID: 0.247 G_Rec: 0.337 D_GP: 0.045 D_real: 1.139 D_fake: 0.568 +(epoch: 65, iters: 5488, time: 0.064) G_GAN: 0.154 G_GAN_Feat: 0.929 G_ID: 0.399 G_Rec: 0.375 D_GP: 0.080 D_real: 0.917 D_fake: 0.848 +(epoch: 65, iters: 5888, time: 0.064) G_GAN: 0.193 G_GAN_Feat: 0.956 G_ID: 0.229 G_Rec: 0.374 D_GP: 0.063 D_real: 0.647 D_fake: 0.808 +(epoch: 65, iters: 6288, time: 0.064) G_GAN: 0.334 G_GAN_Feat: 1.133 G_ID: 0.437 G_Rec: 0.457 D_GP: 0.044 D_real: 0.532 D_fake: 0.666 +(epoch: 65, iters: 6688, time: 0.064) G_GAN: 0.145 G_GAN_Feat: 0.718 G_ID: 0.238 G_Rec: 0.353 D_GP: 0.025 D_real: 1.065 D_fake: 0.855 +(epoch: 65, iters: 7088, time: 0.064) G_GAN: 0.281 G_GAN_Feat: 0.808 G_ID: 0.450 G_Rec: 0.361 D_GP: 0.031 D_real: 0.966 D_fake: 0.719 +(epoch: 65, iters: 7488, time: 0.064) G_GAN: 0.156 G_GAN_Feat: 0.670 G_ID: 0.283 G_Rec: 0.345 D_GP: 0.026 D_real: 0.923 D_fake: 0.844 +(epoch: 65, iters: 7888, time: 0.064) G_GAN: 0.476 G_GAN_Feat: 0.760 G_ID: 0.417 G_Rec: 0.344 D_GP: 0.036 D_real: 1.190 D_fake: 0.525 +(epoch: 65, iters: 8288, time: 0.064) G_GAN: 0.233 G_GAN_Feat: 0.711 G_ID: 0.236 G_Rec: 0.361 D_GP: 0.032 D_real: 1.045 D_fake: 0.768 +(epoch: 66, iters: 80, time: 0.064) G_GAN: 0.238 G_GAN_Feat: 0.777 G_ID: 0.412 G_Rec: 0.358 D_GP: 0.066 D_real: 0.837 D_fake: 0.763 +(epoch: 66, iters: 480, time: 0.064) G_GAN: 0.280 G_GAN_Feat: 0.996 G_ID: 0.207 G_Rec: 0.417 D_GP: 0.037 D_real: 0.570 D_fake: 0.732 +(epoch: 66, iters: 880, time: 0.064) G_GAN: 0.321 G_GAN_Feat: 0.670 G_ID: 0.410 G_Rec: 0.380 D_GP: 0.020 D_real: 1.110 D_fake: 0.679 +(epoch: 66, iters: 1280, time: 0.064) G_GAN: 0.260 G_GAN_Feat: 0.899 G_ID: 0.218 G_Rec: 0.351 D_GP: 0.069 D_real: 0.722 D_fake: 0.741 +(epoch: 66, iters: 1680, time: 0.064) G_GAN: 0.820 G_GAN_Feat: 1.213 G_ID: 0.444 G_Rec: 0.438 D_GP: 0.120 D_real: 0.699 D_fake: 0.235 +(epoch: 66, iters: 2080, time: 0.064) G_GAN: 0.184 G_GAN_Feat: 1.223 G_ID: 0.216 G_Rec: 0.405 D_GP: 0.200 D_real: 0.499 D_fake: 0.816 +(epoch: 66, iters: 2480, time: 0.064) G_GAN: 0.427 G_GAN_Feat: 1.461 G_ID: 0.441 G_Rec: 0.387 D_GP: 1.862 D_real: 0.440 D_fake: 0.579 +(epoch: 66, iters: 2880, time: 0.064) G_GAN: 0.324 G_GAN_Feat: 0.597 G_ID: 0.218 G_Rec: 0.381 D_GP: 0.025 D_real: 1.174 D_fake: 0.676 +(epoch: 66, iters: 3280, time: 0.064) G_GAN: 0.152 G_GAN_Feat: 0.614 G_ID: 0.462 G_Rec: 0.426 D_GP: 0.022 D_real: 1.067 D_fake: 0.848 +(epoch: 66, iters: 3680, time: 0.064) G_GAN: 0.358 G_GAN_Feat: 0.574 G_ID: 0.222 G_Rec: 0.396 D_GP: 0.019 D_real: 1.238 D_fake: 0.642 +(epoch: 66, iters: 4080, time: 0.064) G_GAN: 0.456 G_GAN_Feat: 0.725 G_ID: 0.423 G_Rec: 0.409 D_GP: 0.043 D_real: 1.227 D_fake: 0.554 +(epoch: 66, iters: 4480, time: 0.064) G_GAN: 0.415 G_GAN_Feat: 0.776 G_ID: 0.219 G_Rec: 0.396 D_GP: 0.053 D_real: 1.110 D_fake: 0.586 +(epoch: 66, iters: 4880, time: 0.064) G_GAN: 0.486 G_GAN_Feat: 0.762 G_ID: 0.428 G_Rec: 0.367 D_GP: 0.054 D_real: 1.240 D_fake: 0.527 +(epoch: 66, iters: 5280, time: 0.064) G_GAN: 0.121 G_GAN_Feat: 0.644 G_ID: 0.251 G_Rec: 0.373 D_GP: 0.024 D_real: 0.934 D_fake: 0.879 +(epoch: 66, iters: 5680, time: 0.064) G_GAN: 0.406 G_GAN_Feat: 0.968 G_ID: 0.465 G_Rec: 0.365 D_GP: 0.049 D_real: 0.869 D_fake: 0.596 +(epoch: 66, iters: 6080, time: 0.064) G_GAN: 0.144 G_GAN_Feat: 0.725 G_ID: 0.245 G_Rec: 0.369 D_GP: 0.044 D_real: 0.908 D_fake: 0.859 +(epoch: 66, iters: 6480, time: 0.064) G_GAN: 0.373 G_GAN_Feat: 0.828 G_ID: 0.497 G_Rec: 0.371 D_GP: 0.061 D_real: 0.956 D_fake: 0.630 +(epoch: 66, iters: 6880, time: 0.064) G_GAN: 0.227 G_GAN_Feat: 0.667 G_ID: 0.223 G_Rec: 0.339 D_GP: 0.030 D_real: 1.074 D_fake: 0.774 +(epoch: 66, iters: 7280, time: 0.064) G_GAN: 0.425 G_GAN_Feat: 0.863 G_ID: 0.404 G_Rec: 0.383 D_GP: 0.041 D_real: 1.000 D_fake: 0.577 +(epoch: 66, iters: 7680, time: 0.064) G_GAN: 0.402 G_GAN_Feat: 0.796 G_ID: 0.226 G_Rec: 0.347 D_GP: 0.062 D_real: 0.961 D_fake: 0.599 +(epoch: 66, iters: 8080, time: 0.064) G_GAN: 0.427 G_GAN_Feat: 0.883 G_ID: 0.429 G_Rec: 0.430 D_GP: 0.038 D_real: 0.964 D_fake: 0.574 +(epoch: 66, iters: 8480, time: 0.064) G_GAN: 0.379 G_GAN_Feat: 0.748 G_ID: 0.205 G_Rec: 0.431 D_GP: 0.034 D_real: 1.134 D_fake: 0.622 +(epoch: 67, iters: 272, time: 0.064) G_GAN: 0.340 G_GAN_Feat: 0.929 G_ID: 0.434 G_Rec: 0.370 D_GP: 0.065 D_real: 1.028 D_fake: 0.691 +(epoch: 67, iters: 672, time: 0.064) G_GAN: 0.384 G_GAN_Feat: 1.081 G_ID: 0.198 G_Rec: 0.374 D_GP: 0.052 D_real: 0.722 D_fake: 0.620 +(epoch: 67, iters: 1072, time: 0.064) G_GAN: 0.177 G_GAN_Feat: 1.121 G_ID: 0.462 G_Rec: 0.424 D_GP: 0.082 D_real: 0.217 D_fake: 0.824 +(epoch: 67, iters: 1472, time: 0.064) G_GAN: 0.455 G_GAN_Feat: 0.829 G_ID: 0.194 G_Rec: 0.408 D_GP: 0.051 D_real: 0.991 D_fake: 0.546 +(epoch: 67, iters: 1872, time: 0.064) G_GAN: 0.407 G_GAN_Feat: 0.784 G_ID: 0.428 G_Rec: 0.403 D_GP: 0.029 D_real: 1.151 D_fake: 0.602 +(epoch: 67, iters: 2272, time: 0.064) G_GAN: 0.082 G_GAN_Feat: 1.408 G_ID: 0.249 G_Rec: 0.365 D_GP: 0.237 D_real: 0.239 D_fake: 0.918 +(epoch: 67, iters: 2672, time: 0.064) G_GAN: 0.433 G_GAN_Feat: 1.056 G_ID: 0.393 G_Rec: 0.394 D_GP: 0.068 D_real: 0.634 D_fake: 0.576 +(epoch: 67, iters: 3072, time: 0.064) G_GAN: 0.453 G_GAN_Feat: 0.662 G_ID: 0.213 G_Rec: 0.358 D_GP: 0.022 D_real: 1.254 D_fake: 0.548 +(epoch: 67, iters: 3472, time: 0.064) G_GAN: 0.448 G_GAN_Feat: 0.746 G_ID: 0.392 G_Rec: 0.405 D_GP: 0.025 D_real: 1.144 D_fake: 0.553 +(epoch: 67, iters: 3872, time: 0.064) G_GAN: 0.382 G_GAN_Feat: 0.826 G_ID: 0.255 G_Rec: 0.361 D_GP: 0.051 D_real: 0.954 D_fake: 0.618 +(epoch: 67, iters: 4272, time: 0.064) G_GAN: 0.166 G_GAN_Feat: 0.997 G_ID: 0.425 G_Rec: 0.393 D_GP: 0.105 D_real: 0.762 D_fake: 0.834 +(epoch: 67, iters: 4672, time: 0.064) G_GAN: 0.262 G_GAN_Feat: 0.684 G_ID: 0.250 G_Rec: 0.402 D_GP: 0.029 D_real: 1.191 D_fake: 0.739 +(epoch: 67, iters: 5072, time: 0.064) G_GAN: 0.525 G_GAN_Feat: 0.717 G_ID: 0.381 G_Rec: 0.384 D_GP: 0.026 D_real: 1.348 D_fake: 0.477 +(epoch: 67, iters: 5472, time: 0.064) G_GAN: 0.286 G_GAN_Feat: 0.784 G_ID: 0.237 G_Rec: 0.361 D_GP: 0.060 D_real: 1.067 D_fake: 0.714 +(epoch: 67, iters: 5872, time: 0.064) G_GAN: 0.520 G_GAN_Feat: 0.897 G_ID: 0.459 G_Rec: 0.436 D_GP: 0.038 D_real: 1.151 D_fake: 0.484 +(epoch: 67, iters: 6272, time: 0.064) G_GAN: 0.366 G_GAN_Feat: 0.742 G_ID: 0.231 G_Rec: 0.376 D_GP: 0.027 D_real: 1.157 D_fake: 0.634 +(epoch: 67, iters: 6672, time: 0.064) G_GAN: 0.413 G_GAN_Feat: 0.806 G_ID: 0.421 G_Rec: 0.403 D_GP: 0.047 D_real: 1.074 D_fake: 0.588 +(epoch: 67, iters: 7072, time: 0.064) G_GAN: 0.394 G_GAN_Feat: 0.804 G_ID: 0.244 G_Rec: 0.437 D_GP: 0.093 D_real: 1.065 D_fake: 0.608 +(epoch: 67, iters: 7472, time: 0.064) G_GAN: 0.389 G_GAN_Feat: 1.086 G_ID: 0.421 G_Rec: 0.431 D_GP: 0.097 D_real: 0.534 D_fake: 0.612 +(epoch: 67, iters: 7872, time: 0.064) G_GAN: 0.299 G_GAN_Feat: 0.961 G_ID: 0.217 G_Rec: 0.370 D_GP: 0.069 D_real: 0.832 D_fake: 0.702 +(epoch: 67, iters: 8272, time: 0.064) G_GAN: 0.331 G_GAN_Feat: 0.757 G_ID: 0.429 G_Rec: 0.369 D_GP: 0.040 D_real: 1.113 D_fake: 0.671 +(epoch: 68, iters: 64, time: 0.064) G_GAN: 0.283 G_GAN_Feat: 0.975 G_ID: 0.240 G_Rec: 0.358 D_GP: 0.062 D_real: 0.651 D_fake: 0.726 +(epoch: 68, iters: 464, time: 0.064) G_GAN: 0.765 G_GAN_Feat: 0.818 G_ID: 0.415 G_Rec: 0.396 D_GP: 0.031 D_real: 1.509 D_fake: 0.262 +(epoch: 68, iters: 864, time: 0.064) G_GAN: 0.696 G_GAN_Feat: 0.869 G_ID: 0.208 G_Rec: 0.399 D_GP: 0.076 D_real: 1.111 D_fake: 0.362 +(epoch: 68, iters: 1264, time: 0.064) G_GAN: 0.618 G_GAN_Feat: 0.812 G_ID: 0.449 G_Rec: 0.419 D_GP: 0.031 D_real: 1.323 D_fake: 0.399 +(epoch: 68, iters: 1664, time: 0.064) G_GAN: 0.378 G_GAN_Feat: 0.737 G_ID: 0.230 G_Rec: 0.369 D_GP: 0.030 D_real: 1.150 D_fake: 0.622 +(epoch: 68, iters: 2064, time: 0.064) G_GAN: 0.496 G_GAN_Feat: 0.826 G_ID: 0.412 G_Rec: 0.385 D_GP: 0.038 D_real: 1.140 D_fake: 0.509 +(epoch: 68, iters: 2464, time: 0.064) G_GAN: 0.317 G_GAN_Feat: 0.687 G_ID: 0.252 G_Rec: 0.366 D_GP: 0.029 D_real: 1.079 D_fake: 0.683 +(epoch: 68, iters: 2864, time: 0.064) G_GAN: 0.747 G_GAN_Feat: 1.004 G_ID: 0.436 G_Rec: 0.397 D_GP: 0.069 D_real: 1.359 D_fake: 0.305 +(epoch: 68, iters: 3264, time: 0.064) G_GAN: 0.398 G_GAN_Feat: 0.826 G_ID: 0.236 G_Rec: 0.364 D_GP: 0.055 D_real: 1.091 D_fake: 0.614 +(epoch: 68, iters: 3664, time: 0.064) G_GAN: 0.593 G_GAN_Feat: 0.971 G_ID: 0.465 G_Rec: 0.406 D_GP: 0.094 D_real: 1.029 D_fake: 0.456 +(epoch: 68, iters: 4064, time: 0.064) G_GAN: 0.161 G_GAN_Feat: 0.678 G_ID: 0.259 G_Rec: 0.401 D_GP: 0.022 D_real: 1.074 D_fake: 0.839 +(epoch: 68, iters: 4464, time: 0.064) G_GAN: 0.349 G_GAN_Feat: 0.994 G_ID: 0.413 G_Rec: 0.414 D_GP: 0.075 D_real: 0.623 D_fake: 0.652 +(epoch: 68, iters: 4864, time: 0.064) G_GAN: 0.499 G_GAN_Feat: 0.817 G_ID: 0.235 G_Rec: 0.378 D_GP: 0.054 D_real: 1.148 D_fake: 0.506 +(epoch: 68, iters: 5264, time: 0.064) G_GAN: 0.460 G_GAN_Feat: 0.825 G_ID: 0.442 G_Rec: 0.416 D_GP: 0.035 D_real: 1.084 D_fake: 0.541 +(epoch: 68, iters: 5664, time: 0.064) G_GAN: 0.591 G_GAN_Feat: 0.926 G_ID: 0.203 G_Rec: 0.408 D_GP: 0.080 D_real: 0.970 D_fake: 0.519 +(epoch: 68, iters: 6064, time: 0.064) G_GAN: 0.694 G_GAN_Feat: 0.977 G_ID: 0.388 G_Rec: 0.378 D_GP: 0.034 D_real: 1.511 D_fake: 0.316 +(epoch: 68, iters: 6464, time: 0.064) G_GAN: 0.177 G_GAN_Feat: 1.100 G_ID: 0.252 G_Rec: 0.392 D_GP: 0.086 D_real: 0.433 D_fake: 0.823 +(epoch: 68, iters: 6864, time: 0.064) G_GAN: 0.049 G_GAN_Feat: 0.866 G_ID: 0.443 G_Rec: 0.436 D_GP: 0.032 D_real: 0.758 D_fake: 0.951 +(epoch: 68, iters: 7264, time: 0.064) G_GAN: 0.153 G_GAN_Feat: 0.909 G_ID: 0.226 G_Rec: 0.386 D_GP: 0.098 D_real: 0.702 D_fake: 0.847 +(epoch: 68, iters: 7664, time: 0.064) G_GAN: 0.096 G_GAN_Feat: 0.767 G_ID: 0.392 G_Rec: 0.364 D_GP: 0.037 D_real: 0.773 D_fake: 0.904 +(epoch: 68, iters: 8064, time: 0.064) G_GAN: 0.405 G_GAN_Feat: 0.833 G_ID: 0.261 G_Rec: 0.431 D_GP: 0.053 D_real: 0.958 D_fake: 0.598 +(epoch: 68, iters: 8464, time: 0.064) G_GAN: 0.864 G_GAN_Feat: 1.298 G_ID: 0.402 G_Rec: 0.394 D_GP: 0.115 D_real: 1.144 D_fake: 0.249 +(epoch: 69, iters: 256, time: 0.064) G_GAN: 0.212 G_GAN_Feat: 0.666 G_ID: 0.206 G_Rec: 0.365 D_GP: 0.029 D_real: 1.035 D_fake: 0.788 +(epoch: 69, iters: 656, time: 0.064) G_GAN: 0.455 G_GAN_Feat: 0.762 G_ID: 0.387 G_Rec: 0.510 D_GP: 0.035 D_real: 1.274 D_fake: 0.547 +(epoch: 69, iters: 1056, time: 0.064) G_GAN: -0.006 G_GAN_Feat: 1.153 G_ID: 0.293 G_Rec: 0.388 D_GP: 0.183 D_real: 0.258 D_fake: 1.006 +(epoch: 69, iters: 1456, time: 0.064) G_GAN: 0.573 G_GAN_Feat: 0.954 G_ID: 0.443 G_Rec: 0.460 D_GP: 0.032 D_real: 1.243 D_fake: 0.476 +(epoch: 69, iters: 1856, time: 0.064) G_GAN: 0.404 G_GAN_Feat: 0.688 G_ID: 0.254 G_Rec: 0.338 D_GP: 0.027 D_real: 1.227 D_fake: 0.596 +(epoch: 69, iters: 2256, time: 0.064) G_GAN: 0.506 G_GAN_Feat: 0.965 G_ID: 0.410 G_Rec: 0.364 D_GP: 0.089 D_real: 0.951 D_fake: 0.497 +(epoch: 69, iters: 2656, time: 0.064) G_GAN: 0.250 G_GAN_Feat: 0.941 G_ID: 0.224 G_Rec: 0.329 D_GP: 0.040 D_real: 0.758 D_fake: 0.750 +(epoch: 69, iters: 3056, time: 0.064) G_GAN: 0.674 G_GAN_Feat: 0.900 G_ID: 0.428 G_Rec: 0.396 D_GP: 0.062 D_real: 1.276 D_fake: 0.350 +(epoch: 69, iters: 3456, time: 0.064) G_GAN: 0.463 G_GAN_Feat: 0.723 G_ID: 0.184 G_Rec: 0.357 D_GP: 0.030 D_real: 1.182 D_fake: 0.538 +(epoch: 69, iters: 3856, time: 0.064) G_GAN: 0.438 G_GAN_Feat: 0.856 G_ID: 0.414 G_Rec: 0.371 D_GP: 0.039 D_real: 1.112 D_fake: 0.564 +(epoch: 69, iters: 4256, time: 0.064) G_GAN: 0.656 G_GAN_Feat: 0.977 G_ID: 0.194 G_Rec: 0.391 D_GP: 0.109 D_real: 0.874 D_fake: 0.408 +(epoch: 69, iters: 4656, time: 0.064) G_GAN: 0.503 G_GAN_Feat: 0.699 G_ID: 0.414 G_Rec: 0.377 D_GP: 0.024 D_real: 1.355 D_fake: 0.501 +(epoch: 69, iters: 5056, time: 0.064) G_GAN: 0.292 G_GAN_Feat: 1.043 G_ID: 0.211 G_Rec: 0.429 D_GP: 0.183 D_real: 0.415 D_fake: 0.710 +(epoch: 69, iters: 5456, time: 0.064) G_GAN: 0.334 G_GAN_Feat: 0.922 G_ID: 0.417 G_Rec: 0.447 D_GP: 0.035 D_real: 1.123 D_fake: 0.666 +(epoch: 69, iters: 5856, time: 0.064) G_GAN: 0.456 G_GAN_Feat: 0.727 G_ID: 0.237 G_Rec: 0.341 D_GP: 0.041 D_real: 1.188 D_fake: 0.546 +(epoch: 69, iters: 6256, time: 0.064) G_GAN: 0.468 G_GAN_Feat: 0.845 G_ID: 0.397 G_Rec: 0.413 D_GP: 0.036 D_real: 1.168 D_fake: 0.533 +(epoch: 69, iters: 6656, time: 0.064) G_GAN: 0.087 G_GAN_Feat: 0.928 G_ID: 0.201 G_Rec: 0.361 D_GP: 0.069 D_real: 0.507 D_fake: 0.913 +(epoch: 69, iters: 7056, time: 0.064) G_GAN: 0.306 G_GAN_Feat: 0.775 G_ID: 0.420 G_Rec: 0.372 D_GP: 0.029 D_real: 1.042 D_fake: 0.694 +(epoch: 69, iters: 7456, time: 0.064) G_GAN: 0.131 G_GAN_Feat: 0.761 G_ID: 0.262 G_Rec: 0.367 D_GP: 0.035 D_real: 0.827 D_fake: 0.869 +(epoch: 69, iters: 7856, time: 0.064) G_GAN: 0.201 G_GAN_Feat: 0.783 G_ID: 0.419 G_Rec: 0.373 D_GP: 0.055 D_real: 0.809 D_fake: 0.799 +(epoch: 69, iters: 8256, time: 0.064) G_GAN: 0.485 G_GAN_Feat: 0.783 G_ID: 0.221 G_Rec: 0.356 D_GP: 0.042 D_real: 1.201 D_fake: 0.518 +(epoch: 70, iters: 48, time: 0.064) G_GAN: 0.383 G_GAN_Feat: 0.749 G_ID: 0.410 G_Rec: 0.385 D_GP: 0.026 D_real: 1.062 D_fake: 0.618 +(epoch: 70, iters: 448, time: 0.064) G_GAN: 0.483 G_GAN_Feat: 0.785 G_ID: 0.223 G_Rec: 0.366 D_GP: 0.056 D_real: 1.189 D_fake: 0.521 +(epoch: 70, iters: 848, time: 0.064) G_GAN: 0.367 G_GAN_Feat: 1.511 G_ID: 0.400 G_Rec: 0.431 D_GP: 0.226 D_real: 0.450 D_fake: 0.637 +(epoch: 70, iters: 1248, time: 0.064) G_GAN: 0.779 G_GAN_Feat: 0.901 G_ID: 0.190 G_Rec: 0.341 D_GP: 0.049 D_real: 1.324 D_fake: 0.238 +(epoch: 70, iters: 1648, time: 0.064) G_GAN: 1.037 G_GAN_Feat: 1.149 G_ID: 0.411 G_Rec: 0.371 D_GP: 0.054 D_real: 1.550 D_fake: 0.220 +(epoch: 70, iters: 2048, time: 0.064) G_GAN: 0.621 G_GAN_Feat: 0.724 G_ID: 0.213 G_Rec: 0.393 D_GP: 0.018 D_real: 1.536 D_fake: 0.391 +(epoch: 70, iters: 2448, time: 0.064) G_GAN: 0.341 G_GAN_Feat: 0.642 G_ID: 0.441 G_Rec: 0.410 D_GP: 0.018 D_real: 1.140 D_fake: 0.659 +(epoch: 70, iters: 2848, time: 0.064) G_GAN: 0.300 G_GAN_Feat: 0.615 G_ID: 0.205 G_Rec: 0.353 D_GP: 0.025 D_real: 1.155 D_fake: 0.700 +(epoch: 70, iters: 3248, time: 0.064) G_GAN: 0.029 G_GAN_Feat: 0.775 G_ID: 0.442 G_Rec: 0.395 D_GP: 0.042 D_real: 0.740 D_fake: 0.971 +(epoch: 70, iters: 3648, time: 0.064) G_GAN: 0.274 G_GAN_Feat: 0.663 G_ID: 0.193 G_Rec: 0.364 D_GP: 0.032 D_real: 1.098 D_fake: 0.727 +(epoch: 70, iters: 4048, time: 0.064) G_GAN: 0.590 G_GAN_Feat: 0.818 G_ID: 0.374 G_Rec: 0.378 D_GP: 0.053 D_real: 1.209 D_fake: 0.425 +(epoch: 70, iters: 4448, time: 0.064) G_GAN: 0.358 G_GAN_Feat: 0.763 G_ID: 0.189 G_Rec: 0.385 D_GP: 0.030 D_real: 1.174 D_fake: 0.642 +(epoch: 70, iters: 4848, time: 0.064) G_GAN: 0.259 G_GAN_Feat: 0.822 G_ID: 0.377 G_Rec: 0.401 D_GP: 0.075 D_real: 0.827 D_fake: 0.741 +(epoch: 70, iters: 5248, time: 0.064) G_GAN: 0.456 G_GAN_Feat: 0.771 G_ID: 0.217 G_Rec: 0.372 D_GP: 0.052 D_real: 1.094 D_fake: 0.545 +(epoch: 70, iters: 5648, time: 0.064) G_GAN: 0.365 G_GAN_Feat: 0.883 G_ID: 0.375 G_Rec: 0.375 D_GP: 0.082 D_real: 0.878 D_fake: 0.635 +(epoch: 70, iters: 6048, time: 0.064) G_GAN: 0.264 G_GAN_Feat: 0.823 G_ID: 0.254 G_Rec: 0.396 D_GP: 0.067 D_real: 0.911 D_fake: 0.736 +(epoch: 70, iters: 6448, time: 0.064) G_GAN: 0.500 G_GAN_Feat: 0.919 G_ID: 0.379 G_Rec: 0.445 D_GP: 0.052 D_real: 0.972 D_fake: 0.507 +(epoch: 70, iters: 6848, time: 0.064) G_GAN: 0.348 G_GAN_Feat: 0.821 G_ID: 0.210 G_Rec: 0.384 D_GP: 0.065 D_real: 0.895 D_fake: 0.653 +(epoch: 70, iters: 7248, time: 0.064) G_GAN: 0.288 G_GAN_Feat: 0.904 G_ID: 0.385 G_Rec: 0.380 D_GP: 0.056 D_real: 0.832 D_fake: 0.726 +(epoch: 70, iters: 7648, time: 0.064) G_GAN: 0.509 G_GAN_Feat: 0.845 G_ID: 0.217 G_Rec: 0.351 D_GP: 0.051 D_real: 1.210 D_fake: 0.495 +(epoch: 70, iters: 8048, time: 0.064) G_GAN: 0.403 G_GAN_Feat: 0.815 G_ID: 0.428 G_Rec: 0.353 D_GP: 0.029 D_real: 1.200 D_fake: 0.598 +(epoch: 70, iters: 8448, time: 0.064) G_GAN: 0.231 G_GAN_Feat: 0.917 G_ID: 0.197 G_Rec: 0.371 D_GP: 0.088 D_real: 0.608 D_fake: 0.769 +(epoch: 71, iters: 240, time: 0.064) G_GAN: 0.048 G_GAN_Feat: 1.406 G_ID: 0.420 G_Rec: 0.489 D_GP: 2.646 D_real: 0.458 D_fake: 0.953 +(epoch: 71, iters: 640, time: 0.064) G_GAN: 0.135 G_GAN_Feat: 0.899 G_ID: 0.209 G_Rec: 0.423 D_GP: 0.027 D_real: 0.900 D_fake: 0.866 +(epoch: 71, iters: 1040, time: 0.064) G_GAN: 0.191 G_GAN_Feat: 0.847 G_ID: 0.420 G_Rec: 0.363 D_GP: 0.036 D_real: 0.904 D_fake: 0.809 +(epoch: 71, iters: 1440, time: 0.064) G_GAN: -0.206 G_GAN_Feat: 0.897 G_ID: 0.238 G_Rec: 0.389 D_GP: 0.057 D_real: 0.656 D_fake: 1.206 +(epoch: 71, iters: 1840, time: 0.064) G_GAN: 0.163 G_GAN_Feat: 0.988 G_ID: 0.387 G_Rec: 0.399 D_GP: 0.089 D_real: 0.713 D_fake: 0.837 +(epoch: 71, iters: 2240, time: 0.064) G_GAN: 0.354 G_GAN_Feat: 0.834 G_ID: 0.232 G_Rec: 0.375 D_GP: 0.049 D_real: 1.041 D_fake: 0.650 +(epoch: 71, iters: 2640, time: 0.064) G_GAN: 0.437 G_GAN_Feat: 0.802 G_ID: 0.350 G_Rec: 0.434 D_GP: 0.030 D_real: 1.247 D_fake: 0.563 +(epoch: 71, iters: 3040, time: 0.064) G_GAN: 0.291 G_GAN_Feat: 1.129 G_ID: 0.221 G_Rec: 0.371 D_GP: 0.093 D_real: 0.774 D_fake: 0.711 +(epoch: 71, iters: 3440, time: 0.064) G_GAN: 0.467 G_GAN_Feat: 0.848 G_ID: 0.382 G_Rec: 0.396 D_GP: 0.044 D_real: 1.176 D_fake: 0.547 +(epoch: 71, iters: 3840, time: 0.064) G_GAN: 0.114 G_GAN_Feat: 0.987 G_ID: 0.255 G_Rec: 0.370 D_GP: 0.058 D_real: 0.705 D_fake: 0.886 +(epoch: 71, iters: 4240, time: 0.064) G_GAN: 0.664 G_GAN_Feat: 0.885 G_ID: 0.394 G_Rec: 0.365 D_GP: 0.041 D_real: 1.376 D_fake: 0.372 +(epoch: 71, iters: 4640, time: 0.064) G_GAN: 0.171 G_GAN_Feat: 0.768 G_ID: 0.253 G_Rec: 0.369 D_GP: 0.032 D_real: 1.029 D_fake: 0.829 +(epoch: 71, iters: 5040, time: 0.064) G_GAN: 0.220 G_GAN_Feat: 0.796 G_ID: 0.338 G_Rec: 0.391 D_GP: 0.038 D_real: 0.973 D_fake: 0.780 +(epoch: 71, iters: 5440, time: 0.064) G_GAN: 0.226 G_GAN_Feat: 0.816 G_ID: 0.220 G_Rec: 0.368 D_GP: 0.057 D_real: 0.898 D_fake: 0.774 +(epoch: 71, iters: 5840, time: 0.064) G_GAN: 0.115 G_GAN_Feat: 0.766 G_ID: 0.414 G_Rec: 0.370 D_GP: 0.034 D_real: 0.777 D_fake: 0.886 +(epoch: 71, iters: 6240, time: 0.064) G_GAN: 0.256 G_GAN_Feat: 0.867 G_ID: 0.200 G_Rec: 0.414 D_GP: 0.058 D_real: 0.816 D_fake: 0.744 +(epoch: 71, iters: 6640, time: 0.064) G_GAN: -0.123 G_GAN_Feat: 0.943 G_ID: 0.424 G_Rec: 0.376 D_GP: 0.090 D_real: 0.694 D_fake: 1.123 +(epoch: 71, iters: 7040, time: 0.064) G_GAN: 0.238 G_GAN_Feat: 0.922 G_ID: 0.195 G_Rec: 0.361 D_GP: 0.047 D_real: 0.738 D_fake: 0.764 +(epoch: 71, iters: 7440, time: 0.064) G_GAN: 0.624 G_GAN_Feat: 0.845 G_ID: 0.398 G_Rec: 0.392 D_GP: 0.035 D_real: 1.334 D_fake: 0.383 +(epoch: 71, iters: 7840, time: 0.064) G_GAN: 0.565 G_GAN_Feat: 1.004 G_ID: 0.257 G_Rec: 0.371 D_GP: 0.037 D_real: 1.238 D_fake: 0.442 +(epoch: 71, iters: 8240, time: 0.064) G_GAN: 0.191 G_GAN_Feat: 0.774 G_ID: 0.415 G_Rec: 0.466 D_GP: 0.028 D_real: 0.895 D_fake: 0.809 +(epoch: 72, iters: 32, time: 0.064) G_GAN: 0.445 G_GAN_Feat: 0.939 G_ID: 0.209 G_Rec: 0.352 D_GP: 0.031 D_real: 1.029 D_fake: 0.555 +(epoch: 72, iters: 432, time: 0.064) G_GAN: 0.271 G_GAN_Feat: 0.845 G_ID: 0.381 G_Rec: 0.433 D_GP: 0.032 D_real: 0.977 D_fake: 0.735 +(epoch: 72, iters: 832, time: 0.064) G_GAN: 0.804 G_GAN_Feat: 0.854 G_ID: 0.210 G_Rec: 0.388 D_GP: 0.052 D_real: 1.471 D_fake: 0.259 +(epoch: 72, iters: 1232, time: 0.064) G_GAN: 0.312 G_GAN_Feat: 1.358 G_ID: 0.403 G_Rec: 0.410 D_GP: 0.230 D_real: 0.474 D_fake: 0.690 +(epoch: 72, iters: 1632, time: 0.064) G_GAN: 0.176 G_GAN_Feat: 1.152 G_ID: 0.190 G_Rec: 0.368 D_GP: 0.065 D_real: 0.462 D_fake: 0.824 +(epoch: 72, iters: 2032, time: 0.064) G_GAN: 0.127 G_GAN_Feat: 0.710 G_ID: 0.438 G_Rec: 0.371 D_GP: 0.029 D_real: 0.913 D_fake: 0.873 +(epoch: 72, iters: 2432, time: 0.064) G_GAN: 0.093 G_GAN_Feat: 0.838 G_ID: 0.198 G_Rec: 0.395 D_GP: 0.058 D_real: 0.715 D_fake: 0.907 +(epoch: 72, iters: 2832, time: 0.064) G_GAN: -0.050 G_GAN_Feat: 0.804 G_ID: 0.337 G_Rec: 0.438 D_GP: 0.024 D_real: 0.869 D_fake: 1.050 +(epoch: 72, iters: 3232, time: 0.064) G_GAN: 0.170 G_GAN_Feat: 0.701 G_ID: 0.176 G_Rec: 0.355 D_GP: 0.029 D_real: 0.938 D_fake: 0.831 +(epoch: 72, iters: 3632, time: 0.064) G_GAN: 0.199 G_GAN_Feat: 0.800 G_ID: 0.391 G_Rec: 0.427 D_GP: 0.038 D_real: 0.872 D_fake: 0.801 +(epoch: 72, iters: 4032, time: 0.064) G_GAN: 0.409 G_GAN_Feat: 0.805 G_ID: 0.217 G_Rec: 0.359 D_GP: 0.031 D_real: 1.127 D_fake: 0.591 +(epoch: 72, iters: 4432, time: 0.064) G_GAN: 0.700 G_GAN_Feat: 0.991 G_ID: 0.407 G_Rec: 0.468 D_GP: 0.079 D_real: 1.180 D_fake: 0.426 +(epoch: 72, iters: 4832, time: 0.064) G_GAN: 0.296 G_GAN_Feat: 0.815 G_ID: 0.246 G_Rec: 0.384 D_GP: 0.038 D_real: 0.937 D_fake: 0.706 +(epoch: 72, iters: 5232, time: 0.064) G_GAN: 0.202 G_GAN_Feat: 1.166 G_ID: 0.388 G_Rec: 0.406 D_GP: 0.162 D_real: 0.362 D_fake: 0.798 +(epoch: 72, iters: 5632, time: 0.064) G_GAN: 0.163 G_GAN_Feat: 0.727 G_ID: 0.244 G_Rec: 0.345 D_GP: 0.033 D_real: 0.936 D_fake: 0.837 +(epoch: 72, iters: 6032, time: 0.064) G_GAN: 0.613 G_GAN_Feat: 0.716 G_ID: 0.384 G_Rec: 0.460 D_GP: 0.027 D_real: 1.412 D_fake: 0.395 +(epoch: 72, iters: 6432, time: 0.064) G_GAN: -0.041 G_GAN_Feat: 0.761 G_ID: 0.223 G_Rec: 0.400 D_GP: 0.034 D_real: 0.840 D_fake: 1.041 +(epoch: 72, iters: 6832, time: 0.064) G_GAN: 0.420 G_GAN_Feat: 0.731 G_ID: 0.385 G_Rec: 0.382 D_GP: 0.029 D_real: 1.174 D_fake: 0.583 +(epoch: 72, iters: 7232, time: 0.064) G_GAN: 0.210 G_GAN_Feat: 0.722 G_ID: 0.236 G_Rec: 0.353 D_GP: 0.028 D_real: 0.990 D_fake: 0.790 +(epoch: 72, iters: 7632, time: 0.064) G_GAN: -0.273 G_GAN_Feat: 0.991 G_ID: 0.382 G_Rec: 0.435 D_GP: 0.190 D_real: 0.310 D_fake: 1.273 +(epoch: 72, iters: 8032, time: 0.064) G_GAN: 0.077 G_GAN_Feat: 1.227 G_ID: 0.208 G_Rec: 0.409 D_GP: 0.077 D_real: 0.358 D_fake: 0.923 +(epoch: 72, iters: 8432, time: 0.064) G_GAN: 0.312 G_GAN_Feat: 0.984 G_ID: 0.421 G_Rec: 0.380 D_GP: 0.032 D_real: 0.886 D_fake: 0.688 +(epoch: 73, iters: 224, time: 0.064) G_GAN: 0.129 G_GAN_Feat: 0.794 G_ID: 0.245 G_Rec: 0.348 D_GP: 0.049 D_real: 0.770 D_fake: 0.871 +(epoch: 73, iters: 624, time: 0.064) G_GAN: 0.494 G_GAN_Feat: 1.083 G_ID: 0.387 G_Rec: 0.414 D_GP: 0.295 D_real: 0.543 D_fake: 0.511 +(epoch: 73, iters: 1024, time: 0.064) G_GAN: 0.340 G_GAN_Feat: 0.874 G_ID: 0.213 G_Rec: 0.377 D_GP: 0.034 D_real: 0.881 D_fake: 0.663 +(epoch: 73, iters: 1424, time: 0.064) G_GAN: 0.200 G_GAN_Feat: 0.848 G_ID: 0.396 G_Rec: 0.384 D_GP: 0.051 D_real: 0.928 D_fake: 0.800 +(epoch: 73, iters: 1824, time: 0.064) G_GAN: 0.395 G_GAN_Feat: 0.676 G_ID: 0.226 G_Rec: 0.337 D_GP: 0.031 D_real: 1.214 D_fake: 0.606 +(epoch: 73, iters: 2224, time: 0.064) G_GAN: 0.463 G_GAN_Feat: 1.215 G_ID: 0.393 G_Rec: 0.397 D_GP: 0.074 D_real: 0.512 D_fake: 0.539 +(epoch: 73, iters: 2624, time: 0.064) G_GAN: 0.309 G_GAN_Feat: 0.934 G_ID: 0.203 G_Rec: 0.458 D_GP: 0.079 D_real: 0.698 D_fake: 0.691 +(epoch: 73, iters: 3024, time: 0.064) G_GAN: 0.316 G_GAN_Feat: 0.914 G_ID: 0.440 G_Rec: 0.413 D_GP: 0.036 D_real: 0.928 D_fake: 0.684 +(epoch: 73, iters: 3424, time: 0.064) G_GAN: 0.534 G_GAN_Feat: 0.880 G_ID: 0.256 G_Rec: 0.356 D_GP: 0.044 D_real: 1.068 D_fake: 0.470 +(epoch: 73, iters: 3824, time: 0.064) G_GAN: 0.377 G_GAN_Feat: 0.857 G_ID: 0.422 G_Rec: 0.411 D_GP: 0.028 D_real: 1.028 D_fake: 0.623 +(epoch: 73, iters: 4224, time: 0.064) G_GAN: 0.583 G_GAN_Feat: 0.788 G_ID: 0.237 G_Rec: 0.397 D_GP: 0.038 D_real: 1.353 D_fake: 0.421 +(epoch: 73, iters: 4624, time: 0.064) G_GAN: 0.188 G_GAN_Feat: 0.808 G_ID: 0.379 G_Rec: 0.406 D_GP: 0.026 D_real: 1.011 D_fake: 0.813 +(epoch: 73, iters: 5024, time: 0.064) G_GAN: 0.265 G_GAN_Feat: 0.678 G_ID: 0.205 G_Rec: 0.397 D_GP: 0.022 D_real: 1.117 D_fake: 0.735 +(epoch: 73, iters: 5424, time: 0.064) G_GAN: 0.331 G_GAN_Feat: 0.825 G_ID: 0.343 G_Rec: 0.410 D_GP: 0.030 D_real: 1.010 D_fake: 0.670 +(epoch: 73, iters: 5824, time: 0.064) G_GAN: 0.458 G_GAN_Feat: 0.755 G_ID: 0.177 G_Rec: 0.359 D_GP: 0.048 D_real: 1.147 D_fake: 0.542 +(epoch: 73, iters: 6224, time: 0.064) G_GAN: 0.354 G_GAN_Feat: 0.895 G_ID: 0.395 G_Rec: 0.373 D_GP: 0.029 D_real: 0.982 D_fake: 0.647 +(epoch: 73, iters: 6624, time: 0.064) G_GAN: 0.262 G_GAN_Feat: 0.929 G_ID: 0.293 G_Rec: 0.376 D_GP: 0.067 D_real: 0.922 D_fake: 0.744 +(epoch: 73, iters: 7024, time: 0.064) G_GAN: 0.331 G_GAN_Feat: 1.020 G_ID: 0.378 G_Rec: 0.371 D_GP: 0.043 D_real: 0.634 D_fake: 0.669 +(epoch: 73, iters: 7424, time: 0.064) G_GAN: 0.377 G_GAN_Feat: 0.677 G_ID: 0.230 G_Rec: 0.349 D_GP: 0.026 D_real: 1.251 D_fake: 0.623 +(epoch: 73, iters: 7824, time: 0.064) G_GAN: 0.162 G_GAN_Feat: 0.908 G_ID: 0.348 G_Rec: 0.410 D_GP: 0.030 D_real: 0.754 D_fake: 0.838 +(epoch: 73, iters: 8224, time: 0.064) G_GAN: 0.334 G_GAN_Feat: 0.813 G_ID: 0.278 G_Rec: 0.359 D_GP: 0.032 D_real: 1.038 D_fake: 0.666 +(epoch: 74, iters: 16, time: 0.064) G_GAN: 0.427 G_GAN_Feat: 0.755 G_ID: 0.385 G_Rec: 0.441 D_GP: 0.027 D_real: 1.197 D_fake: 0.574 +(epoch: 74, iters: 416, time: 0.064) G_GAN: 0.296 G_GAN_Feat: 0.631 G_ID: 0.238 G_Rec: 0.348 D_GP: 0.024 D_real: 1.187 D_fake: 0.705 +(epoch: 74, iters: 816, time: 0.064) G_GAN: 0.453 G_GAN_Feat: 1.044 G_ID: 0.391 G_Rec: 0.472 D_GP: 0.058 D_real: 1.045 D_fake: 0.548 +(epoch: 74, iters: 1216, time: 0.064) G_GAN: 0.330 G_GAN_Feat: 0.723 G_ID: 0.206 G_Rec: 0.430 D_GP: 0.031 D_real: 1.097 D_fake: 0.671 +(epoch: 74, iters: 1616, time: 0.064) G_GAN: 0.485 G_GAN_Feat: 0.865 G_ID: 0.378 G_Rec: 0.376 D_GP: 0.046 D_real: 1.098 D_fake: 0.516 +(epoch: 74, iters: 2016, time: 0.064) G_GAN: 0.473 G_GAN_Feat: 0.879 G_ID: 0.221 G_Rec: 0.354 D_GP: 0.084 D_real: 1.005 D_fake: 0.530 +(epoch: 74, iters: 2416, time: 0.064) G_GAN: 0.389 G_GAN_Feat: 0.896 G_ID: 0.431 G_Rec: 0.382 D_GP: 0.071 D_real: 0.906 D_fake: 0.613 +(epoch: 74, iters: 2816, time: 0.064) G_GAN: 0.332 G_GAN_Feat: 0.675 G_ID: 0.185 G_Rec: 0.370 D_GP: 0.024 D_real: 1.202 D_fake: 0.669 +(epoch: 74, iters: 3216, time: 0.064) G_GAN: 0.292 G_GAN_Feat: 0.784 G_ID: 0.356 G_Rec: 0.397 D_GP: 0.031 D_real: 0.980 D_fake: 0.708 +(epoch: 74, iters: 3616, time: 0.064) G_GAN: 0.274 G_GAN_Feat: 0.979 G_ID: 0.292 G_Rec: 0.393 D_GP: 0.052 D_real: 1.045 D_fake: 0.735 +(epoch: 74, iters: 4016, time: 0.064) G_GAN: 0.370 G_GAN_Feat: 1.154 G_ID: 0.353 G_Rec: 0.530 D_GP: 0.111 D_real: 0.533 D_fake: 0.634 +(epoch: 74, iters: 4416, time: 0.064) G_GAN: 0.218 G_GAN_Feat: 1.098 G_ID: 0.220 G_Rec: 0.400 D_GP: 0.554 D_real: 0.497 D_fake: 0.788 +(epoch: 74, iters: 4816, time: 0.064) G_GAN: 0.016 G_GAN_Feat: 1.093 G_ID: 0.365 G_Rec: 0.413 D_GP: 0.102 D_real: 0.321 D_fake: 0.986 +(epoch: 74, iters: 5216, time: 0.064) G_GAN: 0.410 G_GAN_Feat: 0.754 G_ID: 0.187 G_Rec: 0.375 D_GP: 0.022 D_real: 1.255 D_fake: 0.592 +(epoch: 74, iters: 5616, time: 0.064) G_GAN: 0.580 G_GAN_Feat: 0.977 G_ID: 0.378 G_Rec: 0.412 D_GP: 0.079 D_real: 1.005 D_fake: 0.436 +(epoch: 74, iters: 6016, time: 0.064) G_GAN: 0.769 G_GAN_Feat: 1.065 G_ID: 0.218 G_Rec: 0.398 D_GP: 0.139 D_real: 0.731 D_fake: 0.330 +(epoch: 74, iters: 6416, time: 0.064) G_GAN: 0.095 G_GAN_Feat: 0.969 G_ID: 0.374 G_Rec: 0.425 D_GP: 0.058 D_real: 0.500 D_fake: 0.905 +(epoch: 74, iters: 6816, time: 0.064) G_GAN: 0.295 G_GAN_Feat: 1.009 G_ID: 0.202 G_Rec: 0.393 D_GP: 0.076 D_real: 0.504 D_fake: 0.705 +(epoch: 74, iters: 7216, time: 0.064) G_GAN: 0.709 G_GAN_Feat: 0.979 G_ID: 0.399 G_Rec: 0.396 D_GP: 0.052 D_real: 1.025 D_fake: 0.370 +(epoch: 74, iters: 7616, time: 0.064) G_GAN: 0.328 G_GAN_Feat: 0.664 G_ID: 0.224 G_Rec: 0.365 D_GP: 0.020 D_real: 1.205 D_fake: 0.672 +(epoch: 74, iters: 8016, time: 0.064) G_GAN: 0.211 G_GAN_Feat: 0.692 G_ID: 0.403 G_Rec: 0.379 D_GP: 0.023 D_real: 1.025 D_fake: 0.789 +(epoch: 74, iters: 8416, time: 0.064) G_GAN: 0.179 G_GAN_Feat: 0.776 G_ID: 0.231 G_Rec: 0.399 D_GP: 0.059 D_real: 0.854 D_fake: 0.821 +(epoch: 75, iters: 208, time: 0.064) G_GAN: 0.382 G_GAN_Feat: 0.762 G_ID: 0.359 G_Rec: 0.375 D_GP: 0.026 D_real: 1.185 D_fake: 0.618 +(epoch: 75, iters: 608, time: 0.064) G_GAN: 0.354 G_GAN_Feat: 1.079 G_ID: 0.181 G_Rec: 0.403 D_GP: 0.467 D_real: 0.530 D_fake: 0.650 +(epoch: 75, iters: 1008, time: 0.064) G_GAN: 0.513 G_GAN_Feat: 0.921 G_ID: 0.320 G_Rec: 0.437 D_GP: 0.091 D_real: 0.915 D_fake: 0.490 +(epoch: 75, iters: 1408, time: 0.064) G_GAN: 0.242 G_GAN_Feat: 0.678 G_ID: 0.191 G_Rec: 0.366 D_GP: 0.028 D_real: 1.130 D_fake: 0.758 +(epoch: 75, iters: 1808, time: 0.064) G_GAN: 0.374 G_GAN_Feat: 0.761 G_ID: 0.369 G_Rec: 0.337 D_GP: 0.034 D_real: 1.097 D_fake: 0.626 +(epoch: 75, iters: 2208, time: 0.064) G_GAN: 0.294 G_GAN_Feat: 1.020 G_ID: 0.181 G_Rec: 0.385 D_GP: 0.091 D_real: 0.498 D_fake: 0.707 +(epoch: 75, iters: 2608, time: 0.064) G_GAN: 0.105 G_GAN_Feat: 0.759 G_ID: 0.334 G_Rec: 0.379 D_GP: 0.025 D_real: 0.825 D_fake: 0.896 +(epoch: 75, iters: 3008, time: 0.064) G_GAN: 0.121 G_GAN_Feat: 1.146 G_ID: 0.223 G_Rec: 0.421 D_GP: 0.149 D_real: 0.437 D_fake: 0.879 +(epoch: 75, iters: 3408, time: 0.064) G_GAN: 0.357 G_GAN_Feat: 0.866 G_ID: 0.432 G_Rec: 0.415 D_GP: 0.039 D_real: 1.090 D_fake: 0.648 +(epoch: 75, iters: 3808, time: 0.064) G_GAN: 0.471 G_GAN_Feat: 0.846 G_ID: 0.202 G_Rec: 0.373 D_GP: 0.036 D_real: 1.143 D_fake: 0.530 +(epoch: 75, iters: 4208, time: 0.064) G_GAN: 0.107 G_GAN_Feat: 0.940 G_ID: 0.426 G_Rec: 0.386 D_GP: 0.093 D_real: 0.429 D_fake: 0.893 +(epoch: 75, iters: 4608, time: 0.064) G_GAN: 0.227 G_GAN_Feat: 0.759 G_ID: 0.222 G_Rec: 0.357 D_GP: 0.033 D_real: 1.019 D_fake: 0.774 +(epoch: 75, iters: 5008, time: 0.064) G_GAN: 0.485 G_GAN_Feat: 0.815 G_ID: 0.347 G_Rec: 0.408 D_GP: 0.034 D_real: 1.263 D_fake: 0.515 +(epoch: 75, iters: 5408, time: 0.064) G_GAN: 0.402 G_GAN_Feat: 0.786 G_ID: 0.218 G_Rec: 0.334 D_GP: 0.044 D_real: 1.048 D_fake: 0.599 +(epoch: 75, iters: 5808, time: 0.064) G_GAN: 0.813 G_GAN_Feat: 0.948 G_ID: 0.318 G_Rec: 0.397 D_GP: 0.056 D_real: 1.335 D_fake: 0.232 +(epoch: 75, iters: 6208, time: 0.064) G_GAN: 0.299 G_GAN_Feat: 0.648 G_ID: 0.205 G_Rec: 0.373 D_GP: 0.021 D_real: 1.229 D_fake: 0.702 +(epoch: 75, iters: 6608, time: 0.064) G_GAN: 0.332 G_GAN_Feat: 1.146 G_ID: 0.400 G_Rec: 0.423 D_GP: 0.051 D_real: 0.837 D_fake: 0.672 +(epoch: 75, iters: 7008, time: 0.064) G_GAN: 0.341 G_GAN_Feat: 0.960 G_ID: 0.200 G_Rec: 0.377 D_GP: 0.069 D_real: 1.132 D_fake: 0.663 +(epoch: 75, iters: 7408, time: 0.064) G_GAN: -0.194 G_GAN_Feat: 0.851 G_ID: 0.413 G_Rec: 0.397 D_GP: 0.060 D_real: 0.563 D_fake: 1.194 +(epoch: 75, iters: 7808, time: 0.064) G_GAN: 0.170 G_GAN_Feat: 0.787 G_ID: 0.242 G_Rec: 0.370 D_GP: 0.070 D_real: 0.795 D_fake: 0.830 +(epoch: 75, iters: 8208, time: 0.064) G_GAN: 0.618 G_GAN_Feat: 0.741 G_ID: 0.357 G_Rec: 0.346 D_GP: 0.030 D_real: 1.364 D_fake: 0.391 +(epoch: 75, iters: 8608, time: 0.064) G_GAN: 0.288 G_GAN_Feat: 1.012 G_ID: 0.193 G_Rec: 0.347 D_GP: 0.056 D_real: 0.624 D_fake: 0.712 +(epoch: 76, iters: 400, time: 0.064) G_GAN: 0.329 G_GAN_Feat: 0.883 G_ID: 0.421 G_Rec: 0.428 D_GP: 0.029 D_real: 0.934 D_fake: 0.673 +(epoch: 76, iters: 800, time: 0.064) G_GAN: 0.131 G_GAN_Feat: 0.688 G_ID: 0.197 G_Rec: 0.349 D_GP: 0.027 D_real: 0.939 D_fake: 0.869 +(epoch: 76, iters: 1200, time: 0.064) G_GAN: 0.266 G_GAN_Feat: 0.946 G_ID: 0.350 G_Rec: 0.395 D_GP: 0.065 D_real: 0.823 D_fake: 0.735 +(epoch: 76, iters: 1600, time: 0.064) G_GAN: 0.140 G_GAN_Feat: 0.936 G_ID: 0.206 G_Rec: 0.381 D_GP: 0.168 D_real: 0.722 D_fake: 0.865 +(epoch: 76, iters: 2000, time: 0.064) G_GAN: 0.651 G_GAN_Feat: 0.780 G_ID: 0.327 G_Rec: 0.351 D_GP: 0.057 D_real: 1.356 D_fake: 0.379 +(epoch: 76, iters: 2400, time: 0.064) G_GAN: 0.223 G_GAN_Feat: 1.162 G_ID: 0.248 G_Rec: 0.368 D_GP: 0.173 D_real: 0.333 D_fake: 0.777 +(epoch: 76, iters: 2800, time: 0.064) G_GAN: 0.433 G_GAN_Feat: 0.882 G_ID: 0.306 G_Rec: 0.398 D_GP: 0.030 D_real: 1.104 D_fake: 0.571 +(epoch: 76, iters: 3200, time: 0.064) G_GAN: 0.317 G_GAN_Feat: 0.942 G_ID: 0.236 G_Rec: 0.377 D_GP: 0.087 D_real: 0.926 D_fake: 0.683 +(epoch: 76, iters: 3600, time: 0.064) G_GAN: 0.357 G_GAN_Feat: 0.860 G_ID: 0.437 G_Rec: 0.421 D_GP: 0.038 D_real: 0.988 D_fake: 0.645 +(epoch: 76, iters: 4000, time: 0.064) G_GAN: 0.085 G_GAN_Feat: 1.060 G_ID: 0.211 G_Rec: 0.392 D_GP: 0.404 D_real: 0.395 D_fake: 0.915 +(epoch: 76, iters: 4400, time: 0.064) G_GAN: 0.465 G_GAN_Feat: 0.931 G_ID: 0.355 G_Rec: 0.388 D_GP: 0.045 D_real: 0.963 D_fake: 0.539 +(epoch: 76, iters: 4800, time: 0.064) G_GAN: 0.263 G_GAN_Feat: 0.696 G_ID: 0.184 G_Rec: 0.346 D_GP: 0.025 D_real: 1.148 D_fake: 0.737 +(epoch: 76, iters: 5200, time: 0.064) G_GAN: 0.521 G_GAN_Feat: 0.725 G_ID: 0.326 G_Rec: 0.397 D_GP: 0.022 D_real: 1.292 D_fake: 0.483 +(epoch: 76, iters: 5600, time: 0.064) G_GAN: 0.291 G_GAN_Feat: 0.632 G_ID: 0.216 G_Rec: 0.350 D_GP: 0.025 D_real: 1.085 D_fake: 0.709 +(epoch: 76, iters: 6000, time: 0.064) G_GAN: 0.545 G_GAN_Feat: 0.819 G_ID: 0.359 G_Rec: 0.450 D_GP: 0.041 D_real: 1.243 D_fake: 0.492 +(epoch: 76, iters: 6400, time: 0.064) G_GAN: 0.528 G_GAN_Feat: 0.746 G_ID: 0.218 G_Rec: 0.355 D_GP: 0.031 D_real: 1.321 D_fake: 0.479 +(epoch: 76, iters: 6800, time: 0.064) G_GAN: 0.401 G_GAN_Feat: 0.906 G_ID: 0.354 G_Rec: 0.401 D_GP: 0.043 D_real: 1.089 D_fake: 0.612 +(epoch: 76, iters: 7200, time: 0.064) G_GAN: -0.272 G_GAN_Feat: 0.835 G_ID: 0.216 G_Rec: 0.396 D_GP: 0.066 D_real: 0.402 D_fake: 1.272 +(epoch: 76, iters: 7600, time: 0.064) G_GAN: 0.292 G_GAN_Feat: 0.956 G_ID: 0.332 G_Rec: 0.384 D_GP: 0.032 D_real: 1.163 D_fake: 0.714 +(epoch: 76, iters: 8000, time: 0.064) G_GAN: 0.269 G_GAN_Feat: 0.881 G_ID: 0.238 G_Rec: 0.437 D_GP: 0.034 D_real: 0.907 D_fake: 0.732 +(epoch: 76, iters: 8400, time: 0.064) G_GAN: 0.732 G_GAN_Feat: 0.868 G_ID: 0.333 G_Rec: 0.380 D_GP: 0.072 D_real: 1.407 D_fake: 0.309 +(epoch: 77, iters: 192, time: 0.064) G_GAN: 0.524 G_GAN_Feat: 0.759 G_ID: 0.209 G_Rec: 0.362 D_GP: 0.034 D_real: 1.315 D_fake: 0.478 +(epoch: 77, iters: 592, time: 0.064) G_GAN: 0.238 G_GAN_Feat: 0.940 G_ID: 0.314 G_Rec: 0.388 D_GP: 0.059 D_real: 0.652 D_fake: 0.762 +(epoch: 77, iters: 992, time: 0.064) G_GAN: 0.294 G_GAN_Feat: 0.810 G_ID: 0.237 G_Rec: 0.367 D_GP: 0.064 D_real: 0.879 D_fake: 0.706 +(epoch: 77, iters: 1392, time: 0.064) G_GAN: 0.359 G_GAN_Feat: 0.856 G_ID: 0.399 G_Rec: 0.384 D_GP: 0.045 D_real: 1.092 D_fake: 0.641 +(epoch: 77, iters: 1792, time: 0.064) G_GAN: 0.328 G_GAN_Feat: 0.752 G_ID: 0.228 G_Rec: 0.343 D_GP: 0.033 D_real: 1.115 D_fake: 0.673 +(epoch: 77, iters: 2192, time: 0.064) G_GAN: 0.289 G_GAN_Feat: 0.755 G_ID: 0.378 G_Rec: 0.383 D_GP: 0.026 D_real: 1.132 D_fake: 0.711 +(epoch: 77, iters: 2592, time: 0.064) G_GAN: 0.346 G_GAN_Feat: 0.798 G_ID: 0.242 G_Rec: 0.406 D_GP: 0.044 D_real: 1.038 D_fake: 0.654 +(epoch: 77, iters: 2992, time: 0.064) G_GAN: 0.704 G_GAN_Feat: 0.891 G_ID: 0.345 G_Rec: 0.396 D_GP: 0.073 D_real: 1.208 D_fake: 0.314 +(epoch: 77, iters: 3392, time: 0.064) G_GAN: 0.242 G_GAN_Feat: 0.771 G_ID: 0.228 G_Rec: 0.359 D_GP: 0.034 D_real: 0.974 D_fake: 0.758 +(epoch: 77, iters: 3792, time: 0.064) G_GAN: 0.079 G_GAN_Feat: 0.996 G_ID: 0.330 G_Rec: 0.426 D_GP: 0.101 D_real: 0.266 D_fake: 0.921 +(epoch: 77, iters: 4192, time: 0.064) G_GAN: 0.287 G_GAN_Feat: 0.766 G_ID: 0.187 G_Rec: 0.369 D_GP: 0.039 D_real: 1.047 D_fake: 0.715 +(epoch: 77, iters: 4592, time: 0.064) G_GAN: 0.383 G_GAN_Feat: 0.770 G_ID: 0.369 G_Rec: 0.393 D_GP: 0.034 D_real: 1.112 D_fake: 0.618 +(epoch: 77, iters: 4992, time: 0.064) G_GAN: 0.257 G_GAN_Feat: 0.887 G_ID: 0.202 G_Rec: 0.364 D_GP: 0.088 D_real: 0.668 D_fake: 0.744 +(epoch: 77, iters: 5392, time: 0.064) G_GAN: 0.635 G_GAN_Feat: 0.906 G_ID: 0.367 G_Rec: 0.380 D_GP: 0.054 D_real: 1.263 D_fake: 0.377 +(epoch: 77, iters: 5792, time: 0.064) G_GAN: 0.519 G_GAN_Feat: 1.132 G_ID: 0.182 G_Rec: 0.420 D_GP: 0.354 D_real: 0.814 D_fake: 0.501 +(epoch: 77, iters: 6192, time: 0.064) G_GAN: 0.158 G_GAN_Feat: 1.277 G_ID: 0.382 G_Rec: 0.421 D_GP: 0.127 D_real: 0.255 D_fake: 0.842 +(epoch: 77, iters: 6592, time: 0.064) G_GAN: -0.065 G_GAN_Feat: 0.975 G_ID: 0.226 G_Rec: 0.379 D_GP: 0.138 D_real: 0.461 D_fake: 1.065 +(epoch: 77, iters: 6992, time: 0.064) G_GAN: 0.410 G_GAN_Feat: 1.517 G_ID: 0.382 G_Rec: 0.438 D_GP: 0.386 D_real: 0.553 D_fake: 0.595 +(epoch: 77, iters: 7392, time: 0.064) G_GAN: 0.334 G_GAN_Feat: 0.857 G_ID: 0.215 G_Rec: 0.354 D_GP: 0.221 D_real: 0.865 D_fake: 0.684 +(epoch: 77, iters: 7792, time: 0.064) G_GAN: 0.731 G_GAN_Feat: 1.057 G_ID: 0.324 G_Rec: 0.432 D_GP: 0.128 D_real: 0.861 D_fake: 0.315 +(epoch: 77, iters: 8192, time: 0.064) G_GAN: 0.121 G_GAN_Feat: 0.707 G_ID: 0.216 G_Rec: 0.359 D_GP: 0.031 D_real: 1.009 D_fake: 0.879 +(epoch: 77, iters: 8592, time: 0.064) G_GAN: 0.323 G_GAN_Feat: 0.755 G_ID: 0.391 G_Rec: 0.375 D_GP: 0.026 D_real: 1.165 D_fake: 0.677 +(epoch: 78, iters: 384, time: 0.064) G_GAN: 0.273 G_GAN_Feat: 0.875 G_ID: 0.198 G_Rec: 0.377 D_GP: 0.058 D_real: 0.864 D_fake: 0.729 +(epoch: 78, iters: 784, time: 0.064) G_GAN: 0.479 G_GAN_Feat: 0.781 G_ID: 0.344 G_Rec: 0.427 D_GP: 0.027 D_real: 1.248 D_fake: 0.559 +(epoch: 78, iters: 1184, time: 0.064) G_GAN: 0.210 G_GAN_Feat: 0.776 G_ID: 0.181 G_Rec: 0.351 D_GP: 0.034 D_real: 0.883 D_fake: 0.790 +(epoch: 78, iters: 1584, time: 0.064) G_GAN: 0.386 G_GAN_Feat: 0.753 G_ID: 0.344 G_Rec: 0.386 D_GP: 0.024 D_real: 1.178 D_fake: 0.614 +(epoch: 78, iters: 1984, time: 0.064) G_GAN: 0.576 G_GAN_Feat: 0.873 G_ID: 0.267 G_Rec: 0.382 D_GP: 0.036 D_real: 1.166 D_fake: 0.474 +(epoch: 78, iters: 2384, time: 0.064) G_GAN: 0.489 G_GAN_Feat: 0.898 G_ID: 0.358 G_Rec: 0.392 D_GP: 0.042 D_real: 1.012 D_fake: 0.513 +(epoch: 78, iters: 2784, time: 0.064) G_GAN: 0.422 G_GAN_Feat: 0.749 G_ID: 0.258 G_Rec: 0.399 D_GP: 0.025 D_real: 1.260 D_fake: 0.578 +(epoch: 78, iters: 3184, time: 0.064) G_GAN: 0.279 G_GAN_Feat: 0.841 G_ID: 0.368 G_Rec: 0.375 D_GP: 0.030 D_real: 1.053 D_fake: 0.721 +(epoch: 78, iters: 3584, time: 0.064) G_GAN: 0.351 G_GAN_Feat: 0.928 G_ID: 0.193 G_Rec: 0.377 D_GP: 0.050 D_real: 0.797 D_fake: 0.649 +(epoch: 78, iters: 3984, time: 0.064) G_GAN: 0.811 G_GAN_Feat: 1.147 G_ID: 0.351 G_Rec: 0.398 D_GP: 0.181 D_real: 0.765 D_fake: 0.357 +(epoch: 78, iters: 4384, time: 0.064) G_GAN: 0.254 G_GAN_Feat: 1.130 G_ID: 0.233 G_Rec: 0.396 D_GP: 0.074 D_real: 0.360 D_fake: 0.746 +(epoch: 78, iters: 4784, time: 0.064) G_GAN: 0.132 G_GAN_Feat: 0.920 G_ID: 0.355 G_Rec: 0.386 D_GP: 0.076 D_real: 0.684 D_fake: 0.868 +(epoch: 78, iters: 5184, time: 0.064) G_GAN: 0.150 G_GAN_Feat: 0.861 G_ID: 0.263 G_Rec: 0.393 D_GP: 0.025 D_real: 0.919 D_fake: 0.850 +(epoch: 78, iters: 5584, time: 0.064) G_GAN: 0.293 G_GAN_Feat: 1.217 G_ID: 0.356 G_Rec: 0.470 D_GP: 0.043 D_real: 0.518 D_fake: 0.707 +(epoch: 78, iters: 5984, time: 0.064) G_GAN: 0.240 G_GAN_Feat: 0.787 G_ID: 0.193 G_Rec: 0.388 D_GP: 0.023 D_real: 1.079 D_fake: 0.761 +(epoch: 78, iters: 6384, time: 0.064) G_GAN: 0.211 G_GAN_Feat: 0.808 G_ID: 0.330 G_Rec: 0.364 D_GP: 0.031 D_real: 0.963 D_fake: 0.790 +(epoch: 78, iters: 6784, time: 0.064) G_GAN: 0.543 G_GAN_Feat: 0.734 G_ID: 0.203 G_Rec: 0.384 D_GP: 0.025 D_real: 1.372 D_fake: 0.458 +(epoch: 78, iters: 7184, time: 0.064) G_GAN: 0.603 G_GAN_Feat: 0.904 G_ID: 0.331 G_Rec: 0.414 D_GP: 0.035 D_real: 1.223 D_fake: 0.406 +(epoch: 78, iters: 7584, time: 0.064) G_GAN: 0.529 G_GAN_Feat: 0.870 G_ID: 0.186 G_Rec: 0.379 D_GP: 0.080 D_real: 0.955 D_fake: 0.473 +(epoch: 78, iters: 7984, time: 0.064) G_GAN: 0.242 G_GAN_Feat: 0.920 G_ID: 0.370 G_Rec: 0.383 D_GP: 0.045 D_real: 0.807 D_fake: 0.759 +(epoch: 78, iters: 8384, time: 0.064) G_GAN: 0.341 G_GAN_Feat: 0.762 G_ID: 0.213 G_Rec: 0.369 D_GP: 0.047 D_real: 1.046 D_fake: 0.662 +(epoch: 79, iters: 176, time: 0.064) G_GAN: 0.165 G_GAN_Feat: 0.807 G_ID: 0.425 G_Rec: 0.398 D_GP: 0.026 D_real: 0.884 D_fake: 0.835 +(epoch: 79, iters: 576, time: 0.064) G_GAN: 0.202 G_GAN_Feat: 0.917 G_ID: 0.203 G_Rec: 0.349 D_GP: 0.092 D_real: 0.559 D_fake: 0.799 +(epoch: 79, iters: 976, time: 0.064) G_GAN: 0.897 G_GAN_Feat: 1.038 G_ID: 0.329 G_Rec: 0.406 D_GP: 0.050 D_real: 1.438 D_fake: 0.187 +(epoch: 79, iters: 1376, time: 0.064) G_GAN: 0.374 G_GAN_Feat: 1.265 G_ID: 0.223 G_Rec: 0.422 D_GP: 0.061 D_real: 0.404 D_fake: 0.626 +(epoch: 79, iters: 1776, time: 0.064) G_GAN: 0.791 G_GAN_Feat: 0.783 G_ID: 0.360 G_Rec: 0.394 D_GP: 0.026 D_real: 1.563 D_fake: 0.282 +(epoch: 79, iters: 2176, time: 0.064) G_GAN: 0.565 G_GAN_Feat: 0.802 G_ID: 0.178 G_Rec: 0.365 D_GP: 0.043 D_real: 1.214 D_fake: 0.440 +(epoch: 79, iters: 2576, time: 0.064) G_GAN: 0.118 G_GAN_Feat: 0.943 G_ID: 0.348 G_Rec: 0.427 D_GP: 0.055 D_real: 0.545 D_fake: 0.882 +(epoch: 79, iters: 2976, time: 0.064) G_GAN: 0.314 G_GAN_Feat: 0.917 G_ID: 0.226 G_Rec: 0.387 D_GP: 0.042 D_real: 1.105 D_fake: 0.687 +(epoch: 79, iters: 3376, time: 0.064) G_GAN: 0.388 G_GAN_Feat: 0.996 G_ID: 0.395 G_Rec: 0.400 D_GP: 0.079 D_real: 0.571 D_fake: 0.614 +(epoch: 79, iters: 3776, time: 0.064) G_GAN: 0.229 G_GAN_Feat: 0.893 G_ID: 0.235 G_Rec: 0.395 D_GP: 0.149 D_real: 0.779 D_fake: 0.773 +(epoch: 79, iters: 4176, time: 0.064) G_GAN: 0.474 G_GAN_Feat: 0.826 G_ID: 0.320 G_Rec: 0.483 D_GP: 0.032 D_real: 1.133 D_fake: 0.532 +(epoch: 79, iters: 4576, time: 0.064) G_GAN: 0.388 G_GAN_Feat: 0.993 G_ID: 0.202 G_Rec: 0.346 D_GP: 0.040 D_real: 0.811 D_fake: 0.614 +(epoch: 79, iters: 4976, time: 0.064) G_GAN: 0.274 G_GAN_Feat: 0.823 G_ID: 0.399 G_Rec: 0.385 D_GP: 0.039 D_real: 1.029 D_fake: 0.726 +(epoch: 79, iters: 5376, time: 0.064) G_GAN: 0.304 G_GAN_Feat: 0.776 G_ID: 0.173 G_Rec: 0.388 D_GP: 0.041 D_real: 0.983 D_fake: 0.696 +(epoch: 79, iters: 5776, time: 0.064) G_GAN: 0.283 G_GAN_Feat: 1.278 G_ID: 0.344 G_Rec: 0.431 D_GP: 0.292 D_real: 0.436 D_fake: 0.718 +(epoch: 79, iters: 6176, time: 0.064) G_GAN: 0.357 G_GAN_Feat: 0.761 G_ID: 0.192 G_Rec: 0.390 D_GP: 0.021 D_real: 1.275 D_fake: 0.643 +(epoch: 79, iters: 6576, time: 0.064) G_GAN: 0.429 G_GAN_Feat: 0.836 G_ID: 0.302 G_Rec: 0.391 D_GP: 0.050 D_real: 1.157 D_fake: 0.571 +(epoch: 79, iters: 6976, time: 0.064) G_GAN: 0.412 G_GAN_Feat: 0.801 G_ID: 0.196 G_Rec: 0.356 D_GP: 0.045 D_real: 1.122 D_fake: 0.588 +(epoch: 79, iters: 7376, time: 0.064) G_GAN: 0.388 G_GAN_Feat: 0.837 G_ID: 0.401 G_Rec: 0.445 D_GP: 0.028 D_real: 1.049 D_fake: 0.612 +(epoch: 79, iters: 7776, time: 0.064) G_GAN: 0.409 G_GAN_Feat: 0.884 G_ID: 0.185 G_Rec: 0.432 D_GP: 0.044 D_real: 1.072 D_fake: 0.592 +(epoch: 79, iters: 8176, time: 0.064) G_GAN: 0.402 G_GAN_Feat: 0.838 G_ID: 0.349 G_Rec: 0.366 D_GP: 0.027 D_real: 1.058 D_fake: 0.598 +(epoch: 79, iters: 8576, time: 0.064) G_GAN: 0.273 G_GAN_Feat: 1.235 G_ID: 0.187 G_Rec: 0.384 D_GP: 0.275 D_real: 0.308 D_fake: 0.730 +(epoch: 80, iters: 368, time: 0.064) G_GAN: 0.413 G_GAN_Feat: 0.801 G_ID: 0.323 G_Rec: 0.381 D_GP: 0.032 D_real: 1.183 D_fake: 0.593 +(epoch: 80, iters: 768, time: 0.064) G_GAN: 0.465 G_GAN_Feat: 0.635 G_ID: 0.193 G_Rec: 0.363 D_GP: 0.020 D_real: 1.346 D_fake: 0.535 +(epoch: 80, iters: 1168, time: 0.064) G_GAN: 0.271 G_GAN_Feat: 0.895 G_ID: 0.369 G_Rec: 0.502 D_GP: 0.051 D_real: 0.938 D_fake: 0.729 +(epoch: 80, iters: 1568, time: 0.064) G_GAN: 0.191 G_GAN_Feat: 0.854 G_ID: 0.209 G_Rec: 0.432 D_GP: 0.090 D_real: 0.724 D_fake: 0.809 +(epoch: 80, iters: 1968, time: 0.064) G_GAN: 0.489 G_GAN_Feat: 0.813 G_ID: 0.299 G_Rec: 0.390 D_GP: 0.034 D_real: 1.304 D_fake: 0.514 +(epoch: 80, iters: 2368, time: 0.064) G_GAN: 0.357 G_GAN_Feat: 0.647 G_ID: 0.197 G_Rec: 0.393 D_GP: 0.022 D_real: 1.167 D_fake: 0.644 +(epoch: 80, iters: 2768, time: 0.064) G_GAN: 0.360 G_GAN_Feat: 0.739 G_ID: 0.340 G_Rec: 0.385 D_GP: 0.028 D_real: 1.156 D_fake: 0.640 +(epoch: 80, iters: 3168, time: 0.064) G_GAN: -0.269 G_GAN_Feat: 0.643 G_ID: 0.207 G_Rec: 0.326 D_GP: 0.024 D_real: 0.858 D_fake: 1.269 +(epoch: 80, iters: 3568, time: 0.064) G_GAN: 0.402 G_GAN_Feat: 1.044 G_ID: 0.329 G_Rec: 0.394 D_GP: 0.052 D_real: 0.696 D_fake: 0.598 +(epoch: 80, iters: 3968, time: 0.064) G_GAN: 0.139 G_GAN_Feat: 0.720 G_ID: 0.186 G_Rec: 0.418 D_GP: 0.040 D_real: 0.937 D_fake: 0.861 +(epoch: 80, iters: 4368, time: 0.064) G_GAN: 0.131 G_GAN_Feat: 0.805 G_ID: 0.377 G_Rec: 0.425 D_GP: 0.048 D_real: 0.775 D_fake: 0.869 +(epoch: 80, iters: 4768, time: 0.064) G_GAN: 0.490 G_GAN_Feat: 0.633 G_ID: 0.220 G_Rec: 0.337 D_GP: 0.031 D_real: 1.347 D_fake: 0.533 +(epoch: 80, iters: 5168, time: 0.064) G_GAN: 0.412 G_GAN_Feat: 0.873 G_ID: 0.371 G_Rec: 0.390 D_GP: 0.036 D_real: 1.042 D_fake: 0.593 +(epoch: 80, iters: 5568, time: 0.064) G_GAN: 0.381 G_GAN_Feat: 0.897 G_ID: 0.210 G_Rec: 0.413 D_GP: 0.057 D_real: 0.844 D_fake: 0.620 +(epoch: 80, iters: 5968, time: 0.064) G_GAN: 0.205 G_GAN_Feat: 0.876 G_ID: 0.361 G_Rec: 0.425 D_GP: 0.062 D_real: 0.790 D_fake: 0.796 +(epoch: 80, iters: 6368, time: 0.064) G_GAN: 0.600 G_GAN_Feat: 1.043 G_ID: 0.184 G_Rec: 0.427 D_GP: 0.223 D_real: 0.755 D_fake: 0.523 +(epoch: 80, iters: 6768, time: 0.064) G_GAN: 0.494 G_GAN_Feat: 0.896 G_ID: 0.340 G_Rec: 0.376 D_GP: 0.041 D_real: 1.125 D_fake: 0.510 +(epoch: 80, iters: 7168, time: 0.064) G_GAN: 0.336 G_GAN_Feat: 0.821 G_ID: 0.187 G_Rec: 0.373 D_GP: 0.040 D_real: 1.035 D_fake: 0.664 +(epoch: 80, iters: 7568, time: 0.064) G_GAN: 0.330 G_GAN_Feat: 0.828 G_ID: 0.336 G_Rec: 0.410 D_GP: 0.040 D_real: 1.091 D_fake: 0.670 +(epoch: 80, iters: 7968, time: 0.064) G_GAN: 0.176 G_GAN_Feat: 0.987 G_ID: 0.234 G_Rec: 0.378 D_GP: 0.121 D_real: 0.379 D_fake: 0.824 +(epoch: 80, iters: 8368, time: 0.064) G_GAN: 0.450 G_GAN_Feat: 0.887 G_ID: 0.372 G_Rec: 0.418 D_GP: 0.049 D_real: 1.072 D_fake: 0.552 +(epoch: 81, iters: 160, time: 0.064) G_GAN: 0.215 G_GAN_Feat: 0.688 G_ID: 0.199 G_Rec: 0.329 D_GP: 0.034 D_real: 1.137 D_fake: 0.785 +(epoch: 81, iters: 560, time: 0.064) G_GAN: 0.554 G_GAN_Feat: 0.718 G_ID: 0.375 G_Rec: 0.404 D_GP: 0.024 D_real: 1.415 D_fake: 0.450 +(epoch: 81, iters: 960, time: 0.064) G_GAN: 0.203 G_GAN_Feat: 0.705 G_ID: 0.227 G_Rec: 0.356 D_GP: 0.022 D_real: 1.109 D_fake: 0.797 +(epoch: 81, iters: 1360, time: 0.064) G_GAN: 0.524 G_GAN_Feat: 0.723 G_ID: 0.305 G_Rec: 0.376 D_GP: 0.027 D_real: 1.293 D_fake: 0.477 +(epoch: 81, iters: 1760, time: 0.064) G_GAN: 0.314 G_GAN_Feat: 0.695 G_ID: 0.240 G_Rec: 0.361 D_GP: 0.028 D_real: 1.142 D_fake: 0.686 +(epoch: 81, iters: 2160, time: 0.064) G_GAN: 0.409 G_GAN_Feat: 0.726 G_ID: 0.351 G_Rec: 0.353 D_GP: 0.032 D_real: 1.195 D_fake: 0.591 +(epoch: 81, iters: 2560, time: 0.064) G_GAN: 0.442 G_GAN_Feat: 0.757 G_ID: 0.219 G_Rec: 0.366 D_GP: 0.032 D_real: 1.234 D_fake: 0.559 +(epoch: 81, iters: 2960, time: 0.064) G_GAN: 0.136 G_GAN_Feat: 0.992 G_ID: 0.365 G_Rec: 0.422 D_GP: 0.239 D_real: 0.549 D_fake: 0.865 +(epoch: 81, iters: 3360, time: 0.064) G_GAN: 0.302 G_GAN_Feat: 0.794 G_ID: 0.250 G_Rec: 0.361 D_GP: 0.036 D_real: 1.077 D_fake: 0.698 +(epoch: 81, iters: 3760, time: 0.064) G_GAN: 0.688 G_GAN_Feat: 0.969 G_ID: 0.324 G_Rec: 0.413 D_GP: 0.129 D_real: 1.074 D_fake: 0.325 +(epoch: 81, iters: 4160, time: 0.064) G_GAN: 0.275 G_GAN_Feat: 0.817 G_ID: 0.190 G_Rec: 0.460 D_GP: 0.046 D_real: 1.080 D_fake: 0.725 +(epoch: 81, iters: 4560, time: 0.064) G_GAN: 0.541 G_GAN_Feat: 0.730 G_ID: 0.335 G_Rec: 0.378 D_GP: 0.023 D_real: 1.372 D_fake: 0.476 +(epoch: 81, iters: 4960, time: 0.064) G_GAN: 0.457 G_GAN_Feat: 0.696 G_ID: 0.170 G_Rec: 0.325 D_GP: 0.034 D_real: 1.218 D_fake: 0.545 +(epoch: 81, iters: 5360, time: 0.064) G_GAN: 0.275 G_GAN_Feat: 0.983 G_ID: 0.385 G_Rec: 0.400 D_GP: 0.057 D_real: 0.769 D_fake: 0.725 +(epoch: 81, iters: 5760, time: 0.064) G_GAN: 0.226 G_GAN_Feat: 1.241 G_ID: 0.215 G_Rec: 0.408 D_GP: 0.472 D_real: 0.242 D_fake: 0.781 +(epoch: 81, iters: 6160, time: 0.064) G_GAN: 0.268 G_GAN_Feat: 1.043 G_ID: 0.320 G_Rec: 0.426 D_GP: 0.116 D_real: 0.358 D_fake: 0.732 +(epoch: 81, iters: 6560, time: 0.064) G_GAN: 0.476 G_GAN_Feat: 0.990 G_ID: 0.191 G_Rec: 0.432 D_GP: 0.116 D_real: 1.011 D_fake: 0.542 +(epoch: 81, iters: 6960, time: 0.064) G_GAN: 0.224 G_GAN_Feat: 0.904 G_ID: 0.355 G_Rec: 0.390 D_GP: 0.036 D_real: 0.835 D_fake: 0.776 +(epoch: 81, iters: 7360, time: 0.064) G_GAN: 0.274 G_GAN_Feat: 0.856 G_ID: 0.178 G_Rec: 0.405 D_GP: 0.071 D_real: 0.865 D_fake: 0.727 +(epoch: 81, iters: 7760, time: 0.064) G_GAN: 0.535 G_GAN_Feat: 0.900 G_ID: 0.357 G_Rec: 0.377 D_GP: 0.053 D_real: 1.033 D_fake: 0.473 +(epoch: 81, iters: 8160, time: 0.064) G_GAN: 0.445 G_GAN_Feat: 0.743 G_ID: 0.185 G_Rec: 0.376 D_GP: 0.035 D_real: 1.251 D_fake: 0.571 +(epoch: 81, iters: 8560, time: 0.064) G_GAN: 0.461 G_GAN_Feat: 0.861 G_ID: 0.339 G_Rec: 0.464 D_GP: 0.035 D_real: 1.311 D_fake: 0.543 +(epoch: 82, iters: 352, time: 0.064) G_GAN: 0.202 G_GAN_Feat: 0.759 G_ID: 0.193 G_Rec: 0.394 D_GP: 0.025 D_real: 0.985 D_fake: 0.798 +(epoch: 82, iters: 752, time: 0.064) G_GAN: 0.606 G_GAN_Feat: 0.797 G_ID: 0.322 G_Rec: 0.413 D_GP: 0.024 D_real: 1.345 D_fake: 0.433 +(epoch: 82, iters: 1152, time: 0.064) G_GAN: 0.276 G_GAN_Feat: 0.779 G_ID: 0.174 G_Rec: 0.346 D_GP: 0.037 D_real: 1.090 D_fake: 0.724 +(epoch: 82, iters: 1552, time: 0.064) G_GAN: 0.427 G_GAN_Feat: 1.059 G_ID: 0.287 G_Rec: 0.421 D_GP: 0.047 D_real: 1.010 D_fake: 0.575 +(epoch: 82, iters: 1952, time: 0.064) G_GAN: 0.333 G_GAN_Feat: 0.990 G_ID: 0.188 G_Rec: 0.387 D_GP: 0.089 D_real: 1.004 D_fake: 0.688 +(epoch: 82, iters: 2352, time: 0.064) G_GAN: 0.196 G_GAN_Feat: 0.908 G_ID: 0.312 G_Rec: 0.406 D_GP: 0.066 D_real: 0.721 D_fake: 0.804 +(epoch: 82, iters: 2752, time: 0.064) G_GAN: 0.500 G_GAN_Feat: 0.988 G_ID: 0.191 G_Rec: 0.385 D_GP: 0.197 D_real: 0.825 D_fake: 0.517 +(epoch: 82, iters: 3152, time: 0.064) G_GAN: 0.527 G_GAN_Feat: 0.823 G_ID: 0.312 G_Rec: 0.373 D_GP: 0.024 D_real: 1.267 D_fake: 0.476 +(epoch: 82, iters: 3552, time: 0.064) G_GAN: 0.214 G_GAN_Feat: 0.718 G_ID: 0.226 G_Rec: 0.383 D_GP: 0.033 D_real: 1.061 D_fake: 0.787 +(epoch: 82, iters: 3952, time: 0.064) G_GAN: 0.406 G_GAN_Feat: 0.750 G_ID: 0.368 G_Rec: 0.369 D_GP: 0.030 D_real: 1.159 D_fake: 0.599 +(epoch: 82, iters: 4352, time: 0.064) G_GAN: 0.342 G_GAN_Feat: 0.767 G_ID: 0.229 G_Rec: 0.340 D_GP: 0.034 D_real: 1.233 D_fake: 0.659 +(epoch: 82, iters: 4752, time: 0.064) G_GAN: 0.544 G_GAN_Feat: 0.801 G_ID: 0.330 G_Rec: 0.340 D_GP: 0.029 D_real: 1.286 D_fake: 0.459 +(epoch: 82, iters: 5152, time: 0.064) G_GAN: 0.571 G_GAN_Feat: 0.768 G_ID: 0.184 G_Rec: 0.351 D_GP: 0.026 D_real: 1.474 D_fake: 0.484 +(epoch: 82, iters: 5552, time: 0.064) G_GAN: 0.576 G_GAN_Feat: 0.833 G_ID: 0.294 G_Rec: 0.393 D_GP: 0.027 D_real: 1.209 D_fake: 0.439 +(epoch: 82, iters: 5952, time: 0.064) G_GAN: 0.357 G_GAN_Feat: 0.806 G_ID: 0.163 G_Rec: 0.397 D_GP: 0.062 D_real: 1.115 D_fake: 0.643 +(epoch: 82, iters: 6352, time: 0.064) G_GAN: 0.161 G_GAN_Feat: 0.923 G_ID: 0.378 G_Rec: 0.401 D_GP: 0.045 D_real: 0.638 D_fake: 0.839 +(epoch: 82, iters: 6752, time: 0.064) G_GAN: 0.382 G_GAN_Feat: 0.877 G_ID: 0.183 G_Rec: 0.356 D_GP: 0.054 D_real: 0.901 D_fake: 0.618 +(epoch: 82, iters: 7152, time: 0.064) G_GAN: 0.894 G_GAN_Feat: 0.715 G_ID: 0.314 G_Rec: 0.393 D_GP: 0.024 D_real: 1.620 D_fake: 0.162 +(epoch: 82, iters: 7552, time: 0.064) G_GAN: 0.331 G_GAN_Feat: 0.643 G_ID: 0.178 G_Rec: 0.346 D_GP: 0.025 D_real: 1.233 D_fake: 0.669 +(epoch: 82, iters: 7952, time: 0.064) G_GAN: 0.289 G_GAN_Feat: 1.045 G_ID: 0.366 G_Rec: 0.469 D_GP: 0.063 D_real: 0.591 D_fake: 0.714 +(epoch: 82, iters: 8352, time: 0.064) G_GAN: 0.073 G_GAN_Feat: 0.813 G_ID: 0.214 G_Rec: 0.392 D_GP: 0.062 D_real: 0.795 D_fake: 0.927 +(epoch: 83, iters: 144, time: 0.064) G_GAN: 0.415 G_GAN_Feat: 1.152 G_ID: 0.375 G_Rec: 0.430 D_GP: 0.269 D_real: 0.503 D_fake: 0.589 +(epoch: 83, iters: 544, time: 0.064) G_GAN: 0.490 G_GAN_Feat: 0.770 G_ID: 0.173 G_Rec: 0.331 D_GP: 0.048 D_real: 1.188 D_fake: 0.510 +(epoch: 83, iters: 944, time: 0.064) G_GAN: -0.063 G_GAN_Feat: 0.809 G_ID: 0.299 G_Rec: 0.441 D_GP: 0.030 D_real: 0.769 D_fake: 1.063 +(epoch: 83, iters: 1344, time: 0.064) G_GAN: 0.331 G_GAN_Feat: 0.812 G_ID: 0.203 G_Rec: 0.385 D_GP: 0.045 D_real: 0.997 D_fake: 0.670 +(epoch: 83, iters: 1744, time: 0.064) G_GAN: 0.027 G_GAN_Feat: 0.847 G_ID: 0.350 G_Rec: 0.387 D_GP: 0.081 D_real: 0.571 D_fake: 0.973 +(epoch: 83, iters: 2144, time: 0.064) G_GAN: 0.320 G_GAN_Feat: 0.913 G_ID: 0.209 G_Rec: 0.337 D_GP: 0.047 D_real: 0.703 D_fake: 0.680 +(epoch: 83, iters: 2544, time: 0.064) G_GAN: 0.693 G_GAN_Feat: 0.719 G_ID: 0.352 G_Rec: 0.370 D_GP: 0.026 D_real: 1.497 D_fake: 0.319 +(epoch: 83, iters: 2944, time: 0.064) G_GAN: 0.712 G_GAN_Feat: 0.878 G_ID: 0.220 G_Rec: 0.367 D_GP: 0.055 D_real: 1.270 D_fake: 0.333 +(epoch: 83, iters: 3344, time: 0.064) G_GAN: -0.064 G_GAN_Feat: 0.887 G_ID: 0.434 G_Rec: 0.406 D_GP: 0.072 D_real: 0.570 D_fake: 1.064 +(epoch: 83, iters: 3744, time: 0.064) G_GAN: 0.395 G_GAN_Feat: 1.000 G_ID: 0.228 G_Rec: 0.405 D_GP: 0.067 D_real: 0.636 D_fake: 0.611 +(epoch: 83, iters: 4144, time: 0.064) G_GAN: 0.385 G_GAN_Feat: 1.060 G_ID: 0.324 G_Rec: 0.411 D_GP: 0.127 D_real: 0.464 D_fake: 0.627 +(epoch: 83, iters: 4544, time: 0.064) G_GAN: 0.276 G_GAN_Feat: 0.780 G_ID: 0.208 G_Rec: 0.350 D_GP: 0.035 D_real: 1.025 D_fake: 0.724 +(epoch: 83, iters: 4944, time: 0.064) G_GAN: 0.533 G_GAN_Feat: 1.089 G_ID: 0.334 G_Rec: 0.413 D_GP: 0.098 D_real: 0.593 D_fake: 0.474 +(epoch: 83, iters: 5344, time: 0.064) G_GAN: 0.094 G_GAN_Feat: 0.754 G_ID: 0.211 G_Rec: 0.429 D_GP: 0.023 D_real: 0.980 D_fake: 0.906 +(epoch: 83, iters: 5744, time: 0.064) G_GAN: 0.466 G_GAN_Feat: 0.739 G_ID: 0.307 G_Rec: 0.453 D_GP: 0.024 D_real: 1.177 D_fake: 0.535 +(epoch: 83, iters: 6144, time: 0.064) G_GAN: 0.312 G_GAN_Feat: 0.697 G_ID: 0.172 G_Rec: 0.356 D_GP: 0.027 D_real: 1.133 D_fake: 0.689 +(epoch: 83, iters: 6544, time: 0.064) G_GAN: 0.164 G_GAN_Feat: 0.845 G_ID: 0.354 G_Rec: 0.404 D_GP: 0.069 D_real: 0.805 D_fake: 0.836 +(epoch: 83, iters: 6944, time: 0.064) G_GAN: 0.327 G_GAN_Feat: 0.826 G_ID: 0.157 G_Rec: 0.357 D_GP: 0.037 D_real: 0.988 D_fake: 0.675 +(epoch: 83, iters: 7344, time: 0.064) G_GAN: 0.985 G_GAN_Feat: 1.213 G_ID: 0.345 G_Rec: 0.466 D_GP: 0.405 D_real: 0.853 D_fake: 0.135 +(epoch: 83, iters: 7744, time: 0.064) G_GAN: 0.201 G_GAN_Feat: 0.885 G_ID: 0.249 G_Rec: 0.424 D_GP: 0.062 D_real: 0.694 D_fake: 0.800 +(epoch: 83, iters: 8144, time: 0.064) G_GAN: 0.456 G_GAN_Feat: 0.817 G_ID: 0.302 G_Rec: 0.378 D_GP: 0.034 D_real: 1.215 D_fake: 0.554 +(epoch: 83, iters: 8544, time: 0.064) G_GAN: 0.446 G_GAN_Feat: 0.731 G_ID: 0.193 G_Rec: 0.375 D_GP: 0.033 D_real: 1.272 D_fake: 0.557 +(epoch: 84, iters: 336, time: 0.064) G_GAN: 0.280 G_GAN_Feat: 1.004 G_ID: 0.354 G_Rec: 0.400 D_GP: 0.063 D_real: 0.651 D_fake: 0.720 +(epoch: 84, iters: 736, time: 0.064) G_GAN: 0.365 G_GAN_Feat: 0.944 G_ID: 0.184 G_Rec: 0.366 D_GP: 0.035 D_real: 1.038 D_fake: 0.635 +(epoch: 84, iters: 1136, time: 0.064) G_GAN: -0.063 G_GAN_Feat: 0.832 G_ID: 0.363 G_Rec: 0.417 D_GP: 0.037 D_real: 0.663 D_fake: 1.065 +(epoch: 84, iters: 1536, time: 0.064) G_GAN: 0.245 G_GAN_Feat: 0.790 G_ID: 0.222 G_Rec: 0.379 D_GP: 0.039 D_real: 1.100 D_fake: 0.757 +(epoch: 84, iters: 1936, time: 0.064) G_GAN: 0.361 G_GAN_Feat: 0.795 G_ID: 0.313 G_Rec: 0.397 D_GP: 0.026 D_real: 1.131 D_fake: 0.639 +(epoch: 84, iters: 2336, time: 0.064) G_GAN: -0.136 G_GAN_Feat: 0.881 G_ID: 0.235 G_Rec: 0.407 D_GP: 0.064 D_real: 0.615 D_fake: 1.136 +(epoch: 84, iters: 2736, time: 0.064) G_GAN: 0.593 G_GAN_Feat: 0.957 G_ID: 0.353 G_Rec: 0.402 D_GP: 0.068 D_real: 1.199 D_fake: 0.420 +(epoch: 84, iters: 3136, time: 0.064) G_GAN: 0.110 G_GAN_Feat: 0.979 G_ID: 0.187 G_Rec: 0.425 D_GP: 0.026 D_real: 0.669 D_fake: 0.890 +(epoch: 84, iters: 3536, time: 0.064) G_GAN: 0.248 G_GAN_Feat: 0.791 G_ID: 0.364 G_Rec: 0.374 D_GP: 0.031 D_real: 1.034 D_fake: 0.753 +(epoch: 84, iters: 3936, time: 0.064) G_GAN: 0.383 G_GAN_Feat: 0.867 G_ID: 0.209 G_Rec: 0.383 D_GP: 0.036 D_real: 1.138 D_fake: 0.626 +(epoch: 84, iters: 4336, time: 0.064) G_GAN: 0.454 G_GAN_Feat: 1.132 G_ID: 0.302 G_Rec: 0.428 D_GP: 0.095 D_real: 0.463 D_fake: 0.547 +(epoch: 84, iters: 4736, time: 0.064) G_GAN: 0.304 G_GAN_Feat: 0.867 G_ID: 0.181 G_Rec: 0.410 D_GP: 0.035 D_real: 0.892 D_fake: 0.697 +(epoch: 84, iters: 5136, time: 0.064) G_GAN: 0.256 G_GAN_Feat: 0.995 G_ID: 0.350 G_Rec: 0.436 D_GP: 0.148 D_real: 0.682 D_fake: 0.746 +(epoch: 84, iters: 5536, time: 0.064) G_GAN: 0.092 G_GAN_Feat: 0.830 G_ID: 0.193 G_Rec: 0.362 D_GP: 0.066 D_real: 0.748 D_fake: 0.908 +(epoch: 84, iters: 5936, time: 0.064) G_GAN: 0.585 G_GAN_Feat: 0.961 G_ID: 0.321 G_Rec: 0.483 D_GP: 0.054 D_real: 0.973 D_fake: 0.430 +(epoch: 84, iters: 6336, time: 0.064) G_GAN: 0.090 G_GAN_Feat: 0.867 G_ID: 0.178 G_Rec: 0.374 D_GP: 0.033 D_real: 0.668 D_fake: 0.910 +(epoch: 84, iters: 6736, time: 0.064) G_GAN: 0.397 G_GAN_Feat: 0.915 G_ID: 0.348 G_Rec: 0.367 D_GP: 0.035 D_real: 1.052 D_fake: 0.605 +(epoch: 84, iters: 7136, time: 0.064) G_GAN: 0.380 G_GAN_Feat: 0.751 G_ID: 0.193 G_Rec: 0.363 D_GP: 0.023 D_real: 1.251 D_fake: 0.620 +(epoch: 84, iters: 7536, time: 0.064) G_GAN: 0.500 G_GAN_Feat: 1.016 G_ID: 0.353 G_Rec: 0.412 D_GP: 0.046 D_real: 1.069 D_fake: 0.528 +(epoch: 84, iters: 7936, time: 0.064) G_GAN: 0.200 G_GAN_Feat: 0.819 G_ID: 0.184 G_Rec: 0.357 D_GP: 0.075 D_real: 0.856 D_fake: 0.800 +(epoch: 84, iters: 8336, time: 0.064) G_GAN: 0.416 G_GAN_Feat: 0.998 G_ID: 0.351 G_Rec: 0.434 D_GP: 0.066 D_real: 0.736 D_fake: 0.584 +(epoch: 85, iters: 128, time: 0.064) G_GAN: 0.205 G_GAN_Feat: 1.168 G_ID: 0.223 G_Rec: 0.360 D_GP: 0.102 D_real: 0.305 D_fake: 0.795 +(epoch: 85, iters: 528, time: 0.064) G_GAN: -0.478 G_GAN_Feat: 1.075 G_ID: 0.339 G_Rec: 0.469 D_GP: 0.258 D_real: 0.179 D_fake: 1.478 +(epoch: 85, iters: 928, time: 0.064) G_GAN: 0.537 G_GAN_Feat: 0.758 G_ID: 0.168 G_Rec: 0.348 D_GP: 0.030 D_real: 1.364 D_fake: 0.465 +(epoch: 85, iters: 1328, time: 0.064) G_GAN: 0.555 G_GAN_Feat: 0.812 G_ID: 0.319 G_Rec: 0.400 D_GP: 0.029 D_real: 1.233 D_fake: 0.448 +(epoch: 85, iters: 1728, time: 0.064) G_GAN: 0.521 G_GAN_Feat: 0.707 G_ID: 0.208 G_Rec: 0.391 D_GP: 0.024 D_real: 1.416 D_fake: 0.485 +(epoch: 85, iters: 2128, time: 0.064) G_GAN: 0.278 G_GAN_Feat: 0.978 G_ID: 0.347 G_Rec: 0.404 D_GP: 0.080 D_real: 0.624 D_fake: 0.726 +(epoch: 85, iters: 2528, time: 0.064) G_GAN: 0.519 G_GAN_Feat: 0.894 G_ID: 0.208 G_Rec: 0.387 D_GP: 0.072 D_real: 1.047 D_fake: 0.509 +(epoch: 85, iters: 2928, time: 0.064) G_GAN: 0.778 G_GAN_Feat: 0.980 G_ID: 0.277 G_Rec: 0.442 D_GP: 0.044 D_real: 1.342 D_fake: 0.300 +(epoch: 85, iters: 3328, time: 0.064) G_GAN: 0.177 G_GAN_Feat: 0.611 G_ID: 0.208 G_Rec: 0.390 D_GP: 0.022 D_real: 0.976 D_fake: 0.823 +(epoch: 85, iters: 3728, time: 0.064) G_GAN: 0.391 G_GAN_Feat: 0.719 G_ID: 0.299 G_Rec: 0.398 D_GP: 0.020 D_real: 1.121 D_fake: 0.610 +(epoch: 85, iters: 4128, time: 0.064) G_GAN: 0.421 G_GAN_Feat: 0.636 G_ID: 0.170 G_Rec: 0.345 D_GP: 0.023 D_real: 1.262 D_fake: 0.579 +(epoch: 85, iters: 4528, time: 0.064) G_GAN: 0.421 G_GAN_Feat: 0.966 G_ID: 0.278 G_Rec: 0.526 D_GP: 0.216 D_real: 0.903 D_fake: 0.584 +(epoch: 85, iters: 4928, time: 0.064) G_GAN: 0.251 G_GAN_Feat: 0.753 G_ID: 0.191 G_Rec: 0.339 D_GP: 0.036 D_real: 1.026 D_fake: 0.749 +(epoch: 85, iters: 5328, time: 0.064) G_GAN: 0.381 G_GAN_Feat: 0.795 G_ID: 0.357 G_Rec: 0.366 D_GP: 0.029 D_real: 1.131 D_fake: 0.619 +(epoch: 85, iters: 5728, time: 0.064) G_GAN: 0.486 G_GAN_Feat: 0.696 G_ID: 0.163 G_Rec: 0.362 D_GP: 0.024 D_real: 1.287 D_fake: 0.515 +(epoch: 85, iters: 6128, time: 0.064) G_GAN: 0.501 G_GAN_Feat: 1.178 G_ID: 0.321 G_Rec: 0.432 D_GP: 0.477 D_real: 0.587 D_fake: 0.513 +(epoch: 85, iters: 6528, time: 0.064) G_GAN: 0.433 G_GAN_Feat: 0.757 G_ID: 0.188 G_Rec: 0.351 D_GP: 0.035 D_real: 1.221 D_fake: 0.568 +(epoch: 85, iters: 6928, time: 0.064) G_GAN: 0.263 G_GAN_Feat: 1.059 G_ID: 0.331 G_Rec: 0.426 D_GP: 0.052 D_real: 0.546 D_fake: 0.737 +(epoch: 85, iters: 7328, time: 0.064) G_GAN: 0.309 G_GAN_Feat: 0.810 G_ID: 0.211 G_Rec: 0.396 D_GP: 0.024 D_real: 1.079 D_fake: 0.691 +(epoch: 85, iters: 7728, time: 0.064) G_GAN: 0.619 G_GAN_Feat: 0.816 G_ID: 0.323 G_Rec: 0.368 D_GP: 0.036 D_real: 1.283 D_fake: 0.391 +(epoch: 85, iters: 8128, time: 0.064) G_GAN: 0.017 G_GAN_Feat: 1.081 G_ID: 0.228 G_Rec: 0.348 D_GP: 0.048 D_real: 0.363 D_fake: 0.983 +(epoch: 85, iters: 8528, time: 0.064) G_GAN: 0.406 G_GAN_Feat: 0.823 G_ID: 0.336 G_Rec: 0.421 D_GP: 0.023 D_real: 1.232 D_fake: 0.605 +(epoch: 86, iters: 320, time: 0.064) G_GAN: 0.008 G_GAN_Feat: 0.733 G_ID: 0.202 G_Rec: 0.343 D_GP: 0.034 D_real: 0.890 D_fake: 0.992 +(epoch: 86, iters: 720, time: 0.064) G_GAN: 0.515 G_GAN_Feat: 0.879 G_ID: 0.371 G_Rec: 0.405 D_GP: 0.026 D_real: 1.201 D_fake: 0.487 +(epoch: 86, iters: 1120, time: 0.064) G_GAN: 0.341 G_GAN_Feat: 0.897 G_ID: 0.212 G_Rec: 0.359 D_GP: 0.065 D_real: 0.800 D_fake: 0.662 +(epoch: 86, iters: 1520, time: 0.064) G_GAN: 0.611 G_GAN_Feat: 0.785 G_ID: 0.304 G_Rec: 0.363 D_GP: 0.033 D_real: 1.313 D_fake: 0.396 +(epoch: 86, iters: 1920, time: 0.064) G_GAN: 0.247 G_GAN_Feat: 0.897 G_ID: 0.209 G_Rec: 0.368 D_GP: 0.053 D_real: 0.786 D_fake: 0.753 +(epoch: 86, iters: 2320, time: 0.064) G_GAN: 0.548 G_GAN_Feat: 0.868 G_ID: 0.341 G_Rec: 0.394 D_GP: 0.038 D_real: 1.209 D_fake: 0.457 +(epoch: 86, iters: 2720, time: 0.064) G_GAN: 0.327 G_GAN_Feat: 0.684 G_ID: 0.205 G_Rec: 0.392 D_GP: 0.023 D_real: 1.226 D_fake: 0.673 +(epoch: 86, iters: 3120, time: 0.064) G_GAN: 0.580 G_GAN_Feat: 1.041 G_ID: 0.313 G_Rec: 0.443 D_GP: 0.074 D_real: 0.910 D_fake: 0.427 +(epoch: 86, iters: 3520, time: 0.064) G_GAN: 0.174 G_GAN_Feat: 0.822 G_ID: 0.187 G_Rec: 0.379 D_GP: 0.035 D_real: 0.922 D_fake: 0.826 +(epoch: 86, iters: 3920, time: 0.064) G_GAN: 0.249 G_GAN_Feat: 1.131 G_ID: 0.356 G_Rec: 0.387 D_GP: 0.064 D_real: 0.395 D_fake: 0.751 +(epoch: 86, iters: 4320, time: 0.064) G_GAN: 0.374 G_GAN_Feat: 0.712 G_ID: 0.193 G_Rec: 0.400 D_GP: 0.024 D_real: 1.262 D_fake: 0.626 +(epoch: 86, iters: 4720, time: 0.064) G_GAN: 0.414 G_GAN_Feat: 0.848 G_ID: 0.285 G_Rec: 0.428 D_GP: 0.034 D_real: 1.042 D_fake: 0.593 +(epoch: 86, iters: 5120, time: 0.064) G_GAN: -0.022 G_GAN_Feat: 1.048 G_ID: 0.223 G_Rec: 0.423 D_GP: 0.131 D_real: 0.169 D_fake: 1.022 +(epoch: 86, iters: 5520, time: 0.064) G_GAN: 0.292 G_GAN_Feat: 0.900 G_ID: 0.370 G_Rec: 0.399 D_GP: 0.035 D_real: 0.896 D_fake: 0.708 +(epoch: 86, iters: 5920, time: 0.064) G_GAN: 0.366 G_GAN_Feat: 0.774 G_ID: 0.195 G_Rec: 0.340 D_GP: 0.039 D_real: 1.017 D_fake: 0.634 +(epoch: 86, iters: 6320, time: 0.064) G_GAN: 0.622 G_GAN_Feat: 0.842 G_ID: 0.325 G_Rec: 0.438 D_GP: 0.032 D_real: 1.288 D_fake: 0.386 +(epoch: 86, iters: 6720, time: 0.064) G_GAN: 0.468 G_GAN_Feat: 0.739 G_ID: 0.231 G_Rec: 0.353 D_GP: 0.037 D_real: 1.285 D_fake: 0.533 +(epoch: 86, iters: 7120, time: 0.064) G_GAN: 0.690 G_GAN_Feat: 0.859 G_ID: 0.332 G_Rec: 0.421 D_GP: 0.036 D_real: 1.312 D_fake: 0.322 +(epoch: 86, iters: 7520, time: 0.064) G_GAN: 0.563 G_GAN_Feat: 0.831 G_ID: 0.200 G_Rec: 0.391 D_GP: 0.027 D_real: 1.204 D_fake: 0.448 +(epoch: 86, iters: 7920, time: 0.064) G_GAN: 0.302 G_GAN_Feat: 1.238 G_ID: 0.312 G_Rec: 0.415 D_GP: 0.060 D_real: 0.450 D_fake: 0.698 +(epoch: 86, iters: 8320, time: 0.064) G_GAN: 0.425 G_GAN_Feat: 1.133 G_ID: 0.172 G_Rec: 0.361 D_GP: 0.097 D_real: 0.426 D_fake: 0.580 +(epoch: 87, iters: 112, time: 0.064) G_GAN: 0.296 G_GAN_Feat: 0.761 G_ID: 0.341 G_Rec: 0.399 D_GP: 0.025 D_real: 1.151 D_fake: 0.704 +(epoch: 87, iters: 512, time: 0.064) G_GAN: 0.161 G_GAN_Feat: 0.679 G_ID: 0.175 G_Rec: 0.357 D_GP: 0.024 D_real: 1.027 D_fake: 0.839 +(epoch: 87, iters: 912, time: 0.064) G_GAN: 0.286 G_GAN_Feat: 1.003 G_ID: 0.309 G_Rec: 0.466 D_GP: 0.065 D_real: 0.597 D_fake: 0.714 +(epoch: 87, iters: 1312, time: 0.064) G_GAN: 0.467 G_GAN_Feat: 0.754 G_ID: 0.182 G_Rec: 0.370 D_GP: 0.025 D_real: 1.379 D_fake: 0.533 +(epoch: 87, iters: 1712, time: 0.064) G_GAN: 0.307 G_GAN_Feat: 0.769 G_ID: 0.305 G_Rec: 0.362 D_GP: 0.034 D_real: 1.095 D_fake: 0.693 +(epoch: 87, iters: 2112, time: 0.064) G_GAN: 0.190 G_GAN_Feat: 0.659 G_ID: 0.200 G_Rec: 0.345 D_GP: 0.027 D_real: 1.029 D_fake: 0.811 +(epoch: 87, iters: 2512, time: 0.064) G_GAN: 0.369 G_GAN_Feat: 0.842 G_ID: 0.304 G_Rec: 0.359 D_GP: 0.067 D_real: 0.935 D_fake: 0.632 +(epoch: 87, iters: 2912, time: 0.064) G_GAN: 0.291 G_GAN_Feat: 0.840 G_ID: 0.180 G_Rec: 0.360 D_GP: 0.038 D_real: 1.094 D_fake: 0.709 +(epoch: 87, iters: 3312, time: 0.064) G_GAN: 0.478 G_GAN_Feat: 0.900 G_ID: 0.311 G_Rec: 0.383 D_GP: 0.084 D_real: 0.983 D_fake: 0.524 +(epoch: 87, iters: 3712, time: 0.064) G_GAN: 0.210 G_GAN_Feat: 0.627 G_ID: 0.260 G_Rec: 0.342 D_GP: 0.023 D_real: 1.091 D_fake: 0.790 +(epoch: 87, iters: 4112, time: 0.064) G_GAN: 0.425 G_GAN_Feat: 0.674 G_ID: 0.344 G_Rec: 0.362 D_GP: 0.025 D_real: 1.299 D_fake: 0.582 +(epoch: 87, iters: 4512, time: 0.064) G_GAN: 0.152 G_GAN_Feat: 0.739 G_ID: 0.182 G_Rec: 0.353 D_GP: 0.065 D_real: 0.909 D_fake: 0.849 +(epoch: 87, iters: 4912, time: 0.064) G_GAN: 0.379 G_GAN_Feat: 0.705 G_ID: 0.331 G_Rec: 0.436 D_GP: 0.023 D_real: 1.163 D_fake: 0.621 +(epoch: 87, iters: 5312, time: 0.064) G_GAN: 0.420 G_GAN_Feat: 0.702 G_ID: 0.186 G_Rec: 0.365 D_GP: 0.035 D_real: 1.254 D_fake: 0.580 +(epoch: 87, iters: 5712, time: 0.064) G_GAN: 0.425 G_GAN_Feat: 0.875 G_ID: 0.298 G_Rec: 0.397 D_GP: 0.043 D_real: 1.123 D_fake: 0.575 +(epoch: 87, iters: 6112, time: 0.064) G_GAN: 0.072 G_GAN_Feat: 0.681 G_ID: 0.226 G_Rec: 0.376 D_GP: 0.026 D_real: 0.954 D_fake: 0.928 +(epoch: 87, iters: 6512, time: 0.064) G_GAN: 0.599 G_GAN_Feat: 1.263 G_ID: 0.306 G_Rec: 0.456 D_GP: 0.148 D_real: 0.517 D_fake: 0.405 +(epoch: 87, iters: 6912, time: 0.064) G_GAN: 0.093 G_GAN_Feat: 1.052 G_ID: 0.212 G_Rec: 0.381 D_GP: 0.183 D_real: 0.255 D_fake: 0.907 +(epoch: 87, iters: 7312, time: 0.064) G_GAN: 0.327 G_GAN_Feat: 0.941 G_ID: 0.294 G_Rec: 0.415 D_GP: 0.071 D_real: 0.667 D_fake: 0.674 +(epoch: 87, iters: 7712, time: 0.064) G_GAN: 0.519 G_GAN_Feat: 0.624 G_ID: 0.181 G_Rec: 0.342 D_GP: 0.022 D_real: 1.476 D_fake: 0.484 +(epoch: 87, iters: 8112, time: 0.064) G_GAN: 0.613 G_GAN_Feat: 0.960 G_ID: 0.290 G_Rec: 0.367 D_GP: 0.060 D_real: 0.984 D_fake: 0.391 +(epoch: 87, iters: 8512, time: 0.064) G_GAN: 0.340 G_GAN_Feat: 0.729 G_ID: 0.202 G_Rec: 0.395 D_GP: 0.026 D_real: 1.178 D_fake: 0.660 +(epoch: 88, iters: 304, time: 0.064) G_GAN: -0.161 G_GAN_Feat: 0.940 G_ID: 0.320 G_Rec: 0.423 D_GP: 0.077 D_real: 0.316 D_fake: 1.161 +(epoch: 88, iters: 704, time: 0.064) G_GAN: 0.240 G_GAN_Feat: 0.772 G_ID: 0.197 G_Rec: 0.360 D_GP: 0.024 D_real: 1.047 D_fake: 0.760 +(epoch: 88, iters: 1104, time: 0.064) G_GAN: 0.400 G_GAN_Feat: 0.804 G_ID: 0.327 G_Rec: 0.464 D_GP: 0.042 D_real: 1.063 D_fake: 0.600 +(epoch: 88, iters: 1504, time: 0.064) G_GAN: 0.414 G_GAN_Feat: 0.828 G_ID: 0.193 G_Rec: 0.367 D_GP: 0.056 D_real: 1.043 D_fake: 0.586 +(epoch: 88, iters: 1904, time: 0.064) G_GAN: 0.283 G_GAN_Feat: 0.729 G_ID: 0.294 G_Rec: 0.396 D_GP: 0.026 D_real: 1.077 D_fake: 0.717 +(epoch: 88, iters: 2304, time: 0.064) G_GAN: 0.285 G_GAN_Feat: 0.780 G_ID: 0.192 G_Rec: 0.443 D_GP: 0.061 D_real: 0.949 D_fake: 0.715 +(epoch: 88, iters: 2704, time: 0.064) G_GAN: 0.242 G_GAN_Feat: 0.918 G_ID: 0.319 G_Rec: 0.404 D_GP: 0.027 D_real: 0.773 D_fake: 0.758 +(epoch: 88, iters: 3104, time: 0.064) G_GAN: 0.173 G_GAN_Feat: 0.725 G_ID: 0.188 G_Rec: 0.380 D_GP: 0.043 D_real: 1.006 D_fake: 0.827 +(epoch: 88, iters: 3504, time: 0.064) G_GAN: 0.269 G_GAN_Feat: 1.234 G_ID: 0.309 G_Rec: 0.389 D_GP: 0.079 D_real: 0.308 D_fake: 0.731 +(epoch: 88, iters: 3904, time: 0.064) G_GAN: 0.130 G_GAN_Feat: 0.992 G_ID: 0.199 G_Rec: 0.400 D_GP: 0.114 D_real: 0.378 D_fake: 0.870 +(epoch: 88, iters: 4304, time: 0.064) G_GAN: 0.407 G_GAN_Feat: 0.868 G_ID: 0.350 G_Rec: 0.416 D_GP: 0.031 D_real: 1.062 D_fake: 0.594 +(epoch: 88, iters: 4704, time: 0.064) G_GAN: 0.687 G_GAN_Feat: 1.040 G_ID: 0.160 G_Rec: 0.412 D_GP: 0.115 D_real: 1.227 D_fake: 0.473 +(epoch: 88, iters: 5104, time: 0.064) G_GAN: 0.338 G_GAN_Feat: 0.894 G_ID: 0.303 G_Rec: 0.440 D_GP: 0.045 D_real: 0.848 D_fake: 0.662 +(epoch: 88, iters: 5504, time: 0.064) G_GAN: 0.036 G_GAN_Feat: 0.691 G_ID: 0.204 G_Rec: 0.347 D_GP: 0.032 D_real: 0.848 D_fake: 0.964 +(epoch: 88, iters: 5904, time: 0.064) G_GAN: 0.449 G_GAN_Feat: 0.818 G_ID: 0.280 G_Rec: 0.395 D_GP: 0.028 D_real: 1.246 D_fake: 0.552 +(epoch: 88, iters: 6304, time: 0.064) G_GAN: 0.534 G_GAN_Feat: 0.877 G_ID: 0.228 G_Rec: 0.407 D_GP: 0.062 D_real: 1.074 D_fake: 0.533 +(epoch: 88, iters: 6704, time: 0.064) G_GAN: 0.280 G_GAN_Feat: 0.873 G_ID: 0.352 G_Rec: 0.371 D_GP: 0.070 D_real: 0.735 D_fake: 0.721 +(epoch: 88, iters: 7104, time: 0.064) G_GAN: 0.301 G_GAN_Feat: 0.928 G_ID: 0.189 G_Rec: 0.359 D_GP: 0.060 D_real: 0.573 D_fake: 0.699 +(epoch: 88, iters: 7504, time: 0.064) G_GAN: 0.519 G_GAN_Feat: 0.953 G_ID: 0.310 G_Rec: 0.388 D_GP: 0.057 D_real: 0.881 D_fake: 0.486 +(epoch: 88, iters: 7904, time: 0.064) G_GAN: 0.650 G_GAN_Feat: 0.891 G_ID: 0.203 G_Rec: 0.359 D_GP: 0.087 D_real: 1.188 D_fake: 0.390 +(epoch: 88, iters: 8304, time: 0.064) G_GAN: 0.632 G_GAN_Feat: 1.012 G_ID: 0.328 G_Rec: 0.411 D_GP: 0.111 D_real: 0.964 D_fake: 0.433 +(epoch: 89, iters: 96, time: 0.064) G_GAN: 0.379 G_GAN_Feat: 0.810 G_ID: 0.187 G_Rec: 0.371 D_GP: 0.045 D_real: 1.204 D_fake: 0.622 +(epoch: 89, iters: 496, time: 0.064) G_GAN: 0.228 G_GAN_Feat: 0.796 G_ID: 0.316 G_Rec: 0.367 D_GP: 0.032 D_real: 1.121 D_fake: 0.773 +(epoch: 89, iters: 896, time: 0.064) G_GAN: 0.074 G_GAN_Feat: 0.744 G_ID: 0.168 G_Rec: 0.353 D_GP: 0.026 D_real: 0.894 D_fake: 0.926 +(epoch: 89, iters: 1296, time: 0.064) G_GAN: 0.168 G_GAN_Feat: 1.238 G_ID: 0.317 G_Rec: 0.483 D_GP: 0.948 D_real: 0.587 D_fake: 0.836 +(epoch: 89, iters: 1696, time: 0.064) G_GAN: 0.137 G_GAN_Feat: 0.724 G_ID: 0.201 G_Rec: 0.407 D_GP: 0.026 D_real: 0.959 D_fake: 0.863 +(epoch: 89, iters: 2096, time: 0.064) G_GAN: 0.428 G_GAN_Feat: 0.950 G_ID: 0.314 G_Rec: 0.405 D_GP: 0.074 D_real: 0.876 D_fake: 0.576 +(epoch: 89, iters: 2496, time: 0.064) G_GAN: 0.292 G_GAN_Feat: 0.696 G_ID: 0.177 G_Rec: 0.328 D_GP: 0.029 D_real: 1.195 D_fake: 0.708 +(epoch: 89, iters: 2896, time: 0.064) G_GAN: 0.314 G_GAN_Feat: 0.887 G_ID: 0.346 G_Rec: 0.472 D_GP: 0.039 D_real: 1.030 D_fake: 0.686 +(epoch: 89, iters: 3296, time: 0.064) G_GAN: 0.080 G_GAN_Feat: 0.977 G_ID: 0.194 G_Rec: 0.366 D_GP: 0.041 D_real: 0.510 D_fake: 0.920 +(epoch: 89, iters: 3696, time: 0.064) G_GAN: 0.407 G_GAN_Feat: 0.816 G_ID: 0.325 G_Rec: 0.375 D_GP: 0.030 D_real: 1.151 D_fake: 0.594 +(epoch: 89, iters: 4096, time: 0.064) G_GAN: 0.318 G_GAN_Feat: 0.768 G_ID: 0.163 G_Rec: 0.394 D_GP: 0.023 D_real: 1.132 D_fake: 0.684 +(epoch: 89, iters: 4496, time: 0.064) G_GAN: 0.340 G_GAN_Feat: 0.802 G_ID: 0.287 G_Rec: 0.403 D_GP: 0.028 D_real: 1.096 D_fake: 0.662 +(epoch: 89, iters: 4896, time: 0.064) G_GAN: 0.349 G_GAN_Feat: 0.723 G_ID: 0.189 G_Rec: 0.381 D_GP: 0.028 D_real: 1.138 D_fake: 0.651 +(epoch: 89, iters: 5296, time: 0.064) G_GAN: 0.565 G_GAN_Feat: 0.879 G_ID: 0.337 G_Rec: 0.371 D_GP: 0.030 D_real: 1.226 D_fake: 0.437 +(epoch: 89, iters: 5696, time: 0.064) G_GAN: -0.010 G_GAN_Feat: 0.669 G_ID: 0.241 G_Rec: 0.376 D_GP: 0.025 D_real: 0.948 D_fake: 1.011 +(epoch: 89, iters: 6096, time: 0.064) G_GAN: 0.417 G_GAN_Feat: 0.741 G_ID: 0.300 G_Rec: 0.402 D_GP: 0.025 D_real: 1.225 D_fake: 0.584 +(epoch: 89, iters: 6496, time: 0.064) G_GAN: 0.288 G_GAN_Feat: 0.836 G_ID: 0.188 G_Rec: 0.357 D_GP: 0.069 D_real: 0.883 D_fake: 0.713 +(epoch: 89, iters: 6896, time: 0.064) G_GAN: 0.498 G_GAN_Feat: 0.954 G_ID: 0.307 G_Rec: 0.420 D_GP: 0.069 D_real: 0.887 D_fake: 0.506 +(epoch: 89, iters: 7296, time: 0.064) G_GAN: 0.147 G_GAN_Feat: 0.787 G_ID: 0.158 G_Rec: 0.341 D_GP: 0.034 D_real: 0.843 D_fake: 0.853 +(epoch: 89, iters: 7696, time: 0.064) G_GAN: 0.514 G_GAN_Feat: 0.952 G_ID: 0.284 G_Rec: 0.429 D_GP: 0.098 D_real: 1.136 D_fake: 0.488 +(epoch: 89, iters: 8096, time: 0.064) G_GAN: 0.174 G_GAN_Feat: 1.019 G_ID: 0.204 G_Rec: 0.386 D_GP: 0.100 D_real: 1.039 D_fake: 0.830 +(epoch: 89, iters: 8496, time: 0.064) G_GAN: 0.167 G_GAN_Feat: 0.919 G_ID: 0.299 G_Rec: 0.436 D_GP: 0.050 D_real: 0.799 D_fake: 0.833 +(epoch: 90, iters: 288, time: 0.064) G_GAN: 0.827 G_GAN_Feat: 1.069 G_ID: 0.218 G_Rec: 0.399 D_GP: 0.236 D_real: 1.116 D_fake: 0.605 +(epoch: 90, iters: 688, time: 0.064) G_GAN: 0.497 G_GAN_Feat: 0.726 G_ID: 0.310 G_Rec: 0.347 D_GP: 0.028 D_real: 1.209 D_fake: 0.504 +(epoch: 90, iters: 1088, time: 0.064) G_GAN: 0.543 G_GAN_Feat: 0.681 G_ID: 0.159 G_Rec: 0.380 D_GP: 0.026 D_real: 1.376 D_fake: 0.459 +(epoch: 90, iters: 1488, time: 0.064) G_GAN: 0.341 G_GAN_Feat: 0.991 G_ID: 0.336 G_Rec: 0.389 D_GP: 0.111 D_real: 0.585 D_fake: 0.660 +(epoch: 90, iters: 1888, time: 0.064) G_GAN: 0.373 G_GAN_Feat: 1.040 G_ID: 0.193 G_Rec: 0.382 D_GP: 0.114 D_real: 0.553 D_fake: 0.627 +(epoch: 90, iters: 2288, time: 0.064) G_GAN: 0.475 G_GAN_Feat: 1.033 G_ID: 0.309 G_Rec: 0.420 D_GP: 0.070 D_real: 0.833 D_fake: 0.533 +(epoch: 90, iters: 2688, time: 0.064) G_GAN: 0.285 G_GAN_Feat: 1.022 G_ID: 0.179 G_Rec: 0.366 D_GP: 0.103 D_real: 0.557 D_fake: 0.718 +(epoch: 90, iters: 3088, time: 0.064) G_GAN: 0.409 G_GAN_Feat: 0.829 G_ID: 0.338 G_Rec: 0.383 D_GP: 0.033 D_real: 1.057 D_fake: 0.592 +(epoch: 90, iters: 3488, time: 0.064) G_GAN: 0.419 G_GAN_Feat: 0.769 G_ID: 0.177 G_Rec: 0.386 D_GP: 0.027 D_real: 1.186 D_fake: 0.582 +(epoch: 90, iters: 3888, time: 0.064) G_GAN: 0.481 G_GAN_Feat: 0.864 G_ID: 0.296 G_Rec: 0.419 D_GP: 0.043 D_real: 1.161 D_fake: 0.521 +(epoch: 90, iters: 4288, time: 0.064) G_GAN: 0.233 G_GAN_Feat: 0.731 G_ID: 0.214 G_Rec: 0.356 D_GP: 0.029 D_real: 1.043 D_fake: 0.767 +(epoch: 90, iters: 4688, time: 0.064) G_GAN: 0.390 G_GAN_Feat: 1.026 G_ID: 0.274 G_Rec: 0.442 D_GP: 0.138 D_real: 0.567 D_fake: 0.611 +(epoch: 90, iters: 5088, time: 0.064) G_GAN: 0.343 G_GAN_Feat: 0.780 G_ID: 0.198 G_Rec: 0.359 D_GP: 0.042 D_real: 1.122 D_fake: 0.658 +(epoch: 90, iters: 5488, time: 0.064) G_GAN: 0.251 G_GAN_Feat: 0.819 G_ID: 0.304 G_Rec: 0.378 D_GP: 0.030 D_real: 0.989 D_fake: 0.749 +(epoch: 90, iters: 5888, time: 0.064) G_GAN: 0.102 G_GAN_Feat: 1.083 G_ID: 0.218 G_Rec: 0.430 D_GP: 0.183 D_real: 0.231 D_fake: 0.898 +(epoch: 90, iters: 6288, time: 0.064) G_GAN: 0.462 G_GAN_Feat: 0.905 G_ID: 0.311 G_Rec: 0.437 D_GP: 0.044 D_real: 1.145 D_fake: 0.638 +(epoch: 90, iters: 6688, time: 0.064) G_GAN: 0.066 G_GAN_Feat: 1.139 G_ID: 0.176 G_Rec: 0.376 D_GP: 0.716 D_real: 0.492 D_fake: 0.936 +(epoch: 90, iters: 7088, time: 0.064) G_GAN: 0.409 G_GAN_Feat: 0.922 G_ID: 0.301 G_Rec: 0.416 D_GP: 0.026 D_real: 1.109 D_fake: 0.591 +(epoch: 90, iters: 7488, time: 0.064) G_GAN: 0.555 G_GAN_Feat: 0.943 G_ID: 0.191 G_Rec: 0.350 D_GP: 0.102 D_real: 1.166 D_fake: 0.462 +(epoch: 90, iters: 7888, time: 0.064) G_GAN: 0.505 G_GAN_Feat: 0.806 G_ID: 0.303 G_Rec: 0.392 D_GP: 0.028 D_real: 1.219 D_fake: 0.496 +(epoch: 90, iters: 8288, time: 0.064) G_GAN: 0.134 G_GAN_Feat: 0.995 G_ID: 0.199 G_Rec: 0.383 D_GP: 0.044 D_real: 0.506 D_fake: 0.866 +(epoch: 91, iters: 80, time: 0.064) G_GAN: 0.276 G_GAN_Feat: 0.869 G_ID: 0.295 G_Rec: 0.421 D_GP: 0.024 D_real: 0.955 D_fake: 0.731 +(epoch: 91, iters: 480, time: 0.064) G_GAN: 0.153 G_GAN_Feat: 0.800 G_ID: 0.175 G_Rec: 0.376 D_GP: 0.038 D_real: 0.877 D_fake: 0.850 +(epoch: 91, iters: 880, time: 0.064) G_GAN: 0.448 G_GAN_Feat: 0.891 G_ID: 0.305 G_Rec: 0.361 D_GP: 0.040 D_real: 1.033 D_fake: 0.554 +(epoch: 91, iters: 1280, time: 0.064) G_GAN: 0.148 G_GAN_Feat: 0.934 G_ID: 0.212 G_Rec: 0.376 D_GP: 0.118 D_real: 0.639 D_fake: 0.852 +(epoch: 91, iters: 1680, time: 0.064) G_GAN: 0.533 G_GAN_Feat: 1.260 G_ID: 0.281 G_Rec: 0.414 D_GP: 0.204 D_real: 0.514 D_fake: 0.472 +(epoch: 91, iters: 2080, time: 0.064) G_GAN: 0.133 G_GAN_Feat: 0.717 G_ID: 0.192 G_Rec: 0.342 D_GP: 0.027 D_real: 1.013 D_fake: 0.867 +(epoch: 91, iters: 2480, time: 0.064) G_GAN: 0.526 G_GAN_Feat: 0.849 G_ID: 0.300 G_Rec: 0.392 D_GP: 0.029 D_real: 1.187 D_fake: 0.476 +(epoch: 91, iters: 2880, time: 0.064) G_GAN: 0.384 G_GAN_Feat: 0.831 G_ID: 0.177 G_Rec: 0.361 D_GP: 0.039 D_real: 1.188 D_fake: 0.631 +(epoch: 91, iters: 3280, time: 0.064) G_GAN: 0.644 G_GAN_Feat: 0.865 G_ID: 0.317 G_Rec: 0.464 D_GP: 0.023 D_real: 1.450 D_fake: 0.363 +(epoch: 91, iters: 3680, time: 0.064) G_GAN: 0.423 G_GAN_Feat: 0.830 G_ID: 0.175 G_Rec: 0.427 D_GP: 0.050 D_real: 1.129 D_fake: 0.579 +(epoch: 91, iters: 4080, time: 0.064) G_GAN: 0.349 G_GAN_Feat: 0.800 G_ID: 0.299 G_Rec: 0.381 D_GP: 0.025 D_real: 1.069 D_fake: 0.653 +(epoch: 91, iters: 4480, time: 0.064) G_GAN: 0.408 G_GAN_Feat: 0.800 G_ID: 0.187 G_Rec: 0.375 D_GP: 0.035 D_real: 1.097 D_fake: 0.593 +(epoch: 91, iters: 4880, time: 0.064) G_GAN: 0.519 G_GAN_Feat: 1.155 G_ID: 0.344 G_Rec: 0.381 D_GP: 0.069 D_real: 0.504 D_fake: 0.485 +(epoch: 91, iters: 5280, time: 0.064) G_GAN: 0.322 G_GAN_Feat: 0.706 G_ID: 0.204 G_Rec: 0.446 D_GP: 0.023 D_real: 1.194 D_fake: 0.678 +(epoch: 91, iters: 5680, time: 0.064) G_GAN: 0.453 G_GAN_Feat: 0.959 G_ID: 0.283 G_Rec: 0.384 D_GP: 0.040 D_real: 0.904 D_fake: 0.550 +(epoch: 91, iters: 6080, time: 0.064) G_GAN: 0.370 G_GAN_Feat: 1.003 G_ID: 0.197 G_Rec: 0.421 D_GP: 0.040 D_real: 1.167 D_fake: 0.656 +(epoch: 91, iters: 6480, time: 0.064) G_GAN: 0.547 G_GAN_Feat: 1.039 G_ID: 0.325 G_Rec: 0.398 D_GP: 0.079 D_real: 0.840 D_fake: 0.478 +(epoch: 91, iters: 6880, time: 0.064) G_GAN: 0.472 G_GAN_Feat: 0.770 G_ID: 0.158 G_Rec: 0.363 D_GP: 0.024 D_real: 1.252 D_fake: 0.530 +(epoch: 91, iters: 7280, time: 0.064) G_GAN: 0.453 G_GAN_Feat: 1.223 G_ID: 0.311 G_Rec: 0.419 D_GP: 0.066 D_real: 0.554 D_fake: 0.548 +(epoch: 91, iters: 7680, time: 0.064) G_GAN: 0.197 G_GAN_Feat: 1.147 G_ID: 0.177 G_Rec: 0.434 D_GP: 0.385 D_real: 0.432 D_fake: 0.805 +(epoch: 91, iters: 8080, time: 0.064) G_GAN: 0.074 G_GAN_Feat: 0.811 G_ID: 0.334 G_Rec: 0.386 D_GP: 0.037 D_real: 0.806 D_fake: 0.926 +(epoch: 91, iters: 8480, time: 0.064) G_GAN: 0.035 G_GAN_Feat: 0.751 G_ID: 0.192 G_Rec: 0.363 D_GP: 0.027 D_real: 0.872 D_fake: 0.965 +(epoch: 92, iters: 272, time: 0.064) G_GAN: 0.664 G_GAN_Feat: 0.931 G_ID: 0.308 G_Rec: 0.387 D_GP: 0.042 D_real: 1.213 D_fake: 0.348 +(epoch: 92, iters: 672, time: 0.064) G_GAN: 0.390 G_GAN_Feat: 0.654 G_ID: 0.199 G_Rec: 0.360 D_GP: 0.023 D_real: 1.361 D_fake: 0.617 +(epoch: 92, iters: 1072, time: 0.064) G_GAN: 0.280 G_GAN_Feat: 0.900 G_ID: 0.333 G_Rec: 0.375 D_GP: 0.037 D_real: 1.035 D_fake: 0.720 +(epoch: 92, iters: 1472, time: 0.064) G_GAN: 0.401 G_GAN_Feat: 0.668 G_ID: 0.245 G_Rec: 0.345 D_GP: 0.024 D_real: 1.284 D_fake: 0.602 +(epoch: 92, iters: 1872, time: 0.064) G_GAN: 0.487 G_GAN_Feat: 0.740 G_ID: 0.311 G_Rec: 0.387 D_GP: 0.025 D_real: 1.233 D_fake: 0.514 +(epoch: 92, iters: 2272, time: 0.064) G_GAN: 0.310 G_GAN_Feat: 0.870 G_ID: 0.194 G_Rec: 0.361 D_GP: 0.060 D_real: 0.840 D_fake: 0.690 +(epoch: 92, iters: 2672, time: 0.064) G_GAN: 0.578 G_GAN_Feat: 1.151 G_ID: 0.278 G_Rec: 0.399 D_GP: 0.281 D_real: 0.467 D_fake: 0.448 +(epoch: 92, iters: 3072, time: 0.064) G_GAN: 0.540 G_GAN_Feat: 0.850 G_ID: 0.230 G_Rec: 0.415 D_GP: 0.046 D_real: 1.235 D_fake: 0.496 +(epoch: 92, iters: 3472, time: 0.064) G_GAN: 0.761 G_GAN_Feat: 1.052 G_ID: 0.303 G_Rec: 0.437 D_GP: 0.169 D_real: 1.014 D_fake: 0.338 +(epoch: 92, iters: 3872, time: 0.064) G_GAN: 0.347 G_GAN_Feat: 0.585 G_ID: 0.165 G_Rec: 0.414 D_GP: 0.016 D_real: 1.260 D_fake: 0.654 +(epoch: 92, iters: 4272, time: 0.064) G_GAN: 0.413 G_GAN_Feat: 0.683 G_ID: 0.288 G_Rec: 0.416 D_GP: 0.018 D_real: 1.316 D_fake: 0.592 +(epoch: 92, iters: 4672, time: 0.064) G_GAN: 0.292 G_GAN_Feat: 0.737 G_ID: 0.173 G_Rec: 0.406 D_GP: 0.023 D_real: 1.151 D_fake: 0.709 +(epoch: 92, iters: 5072, time: 0.064) G_GAN: 0.178 G_GAN_Feat: 0.766 G_ID: 0.292 G_Rec: 0.419 D_GP: 0.027 D_real: 0.906 D_fake: 0.823 +(epoch: 92, iters: 5472, time: 0.064) G_GAN: 0.282 G_GAN_Feat: 0.716 G_ID: 0.193 G_Rec: 0.344 D_GP: 0.040 D_real: 1.026 D_fake: 0.719 +(epoch: 92, iters: 5872, time: 0.064) G_GAN: 0.237 G_GAN_Feat: 0.826 G_ID: 0.302 G_Rec: 0.377 D_GP: 0.054 D_real: 0.852 D_fake: 0.763 +(epoch: 92, iters: 6272, time: 0.064) G_GAN: 0.296 G_GAN_Feat: 0.914 G_ID: 0.170 G_Rec: 0.404 D_GP: 0.075 D_real: 0.739 D_fake: 0.704 +(epoch: 92, iters: 6672, time: 0.064) G_GAN: 0.310 G_GAN_Feat: 0.873 G_ID: 0.313 G_Rec: 0.387 D_GP: 0.043 D_real: 1.039 D_fake: 0.693 +(epoch: 92, iters: 7072, time: 0.064) G_GAN: 0.472 G_GAN_Feat: 0.722 G_ID: 0.202 G_Rec: 0.367 D_GP: 0.029 D_real: 1.293 D_fake: 0.541 +(epoch: 92, iters: 7472, time: 0.064) G_GAN: 0.614 G_GAN_Feat: 0.937 G_ID: 0.309 G_Rec: 0.415 D_GP: 0.086 D_real: 0.985 D_fake: 0.395 +(epoch: 92, iters: 7872, time: 0.064) G_GAN: 0.304 G_GAN_Feat: 0.710 G_ID: 0.172 G_Rec: 0.391 D_GP: 0.025 D_real: 1.147 D_fake: 0.696 +(epoch: 92, iters: 8272, time: 0.064) G_GAN: 0.226 G_GAN_Feat: 0.787 G_ID: 0.303 G_Rec: 0.400 D_GP: 0.033 D_real: 0.955 D_fake: 0.774 +(epoch: 93, iters: 64, time: 0.064) G_GAN: 0.164 G_GAN_Feat: 0.836 G_ID: 0.182 G_Rec: 0.396 D_GP: 0.108 D_real: 0.759 D_fake: 0.837 +(epoch: 93, iters: 464, time: 0.064) G_GAN: 0.309 G_GAN_Feat: 0.854 G_ID: 0.333 G_Rec: 0.405 D_GP: 0.023 D_real: 1.024 D_fake: 0.691 +(epoch: 93, iters: 864, time: 0.064) G_GAN: 0.346 G_GAN_Feat: 0.863 G_ID: 0.220 G_Rec: 0.397 D_GP: 0.049 D_real: 0.974 D_fake: 0.655 +(epoch: 93, iters: 1264, time: 0.064) G_GAN: 0.180 G_GAN_Feat: 0.971 G_ID: 0.319 G_Rec: 0.387 D_GP: 0.207 D_real: 0.612 D_fake: 0.821 +(epoch: 93, iters: 1664, time: 0.064) G_GAN: 0.205 G_GAN_Feat: 0.706 G_ID: 0.205 G_Rec: 0.333 D_GP: 0.033 D_real: 1.022 D_fake: 0.795 +(epoch: 93, iters: 2064, time: 0.064) G_GAN: 0.099 G_GAN_Feat: 0.753 G_ID: 0.338 G_Rec: 0.428 D_GP: 0.025 D_real: 0.879 D_fake: 0.901 +(epoch: 93, iters: 2464, time: 0.064) G_GAN: 0.031 G_GAN_Feat: 0.666 G_ID: 0.154 G_Rec: 0.360 D_GP: 0.025 D_real: 0.872 D_fake: 0.969 +(epoch: 93, iters: 2864, time: 0.064) G_GAN: 0.378 G_GAN_Feat: 0.853 G_ID: 0.309 G_Rec: 0.378 D_GP: 0.047 D_real: 1.020 D_fake: 0.631 +(epoch: 93, iters: 3264, time: 0.064) G_GAN: 0.000 G_GAN_Feat: 0.748 G_ID: 0.155 G_Rec: 0.370 D_GP: 0.053 D_real: 0.699 D_fake: 1.000 +(epoch: 93, iters: 3664, time: 0.064) G_GAN: 0.802 G_GAN_Feat: 0.808 G_ID: 0.305 G_Rec: 0.357 D_GP: 0.032 D_real: 1.417 D_fake: 0.232 +(epoch: 93, iters: 4064, time: 0.064) G_GAN: 0.441 G_GAN_Feat: 0.798 G_ID: 0.188 G_Rec: 0.339 D_GP: 0.032 D_real: 1.269 D_fake: 0.622 +(epoch: 93, iters: 4464, time: 0.064) G_GAN: 0.094 G_GAN_Feat: 0.936 G_ID: 0.248 G_Rec: 0.376 D_GP: 0.032 D_real: 0.734 D_fake: 0.906 +(epoch: 93, iters: 4864, time: 0.064) G_GAN: 0.424 G_GAN_Feat: 0.846 G_ID: 0.199 G_Rec: 0.398 D_GP: 0.058 D_real: 1.208 D_fake: 0.579 +(epoch: 93, iters: 5264, time: 0.064) G_GAN: 0.330 G_GAN_Feat: 1.089 G_ID: 0.268 G_Rec: 0.414 D_GP: 0.083 D_real: 0.365 D_fake: 0.670 +(epoch: 93, iters: 5664, time: 0.064) G_GAN: -0.014 G_GAN_Feat: 0.766 G_ID: 0.194 G_Rec: 0.446 D_GP: 0.025 D_real: 0.779 D_fake: 1.014 +(epoch: 93, iters: 6064, time: 0.064) G_GAN: 0.041 G_GAN_Feat: 0.839 G_ID: 0.294 G_Rec: 0.368 D_GP: 0.036 D_real: 0.707 D_fake: 0.959 +(epoch: 93, iters: 6464, time: 0.064) G_GAN: 0.377 G_GAN_Feat: 1.039 G_ID: 0.174 G_Rec: 0.393 D_GP: 0.205 D_real: 0.538 D_fake: 0.625 +(epoch: 93, iters: 6864, time: 0.064) G_GAN: 0.217 G_GAN_Feat: 0.975 G_ID: 0.297 G_Rec: 0.445 D_GP: 0.049 D_real: 0.690 D_fake: 0.784 +(epoch: 93, iters: 7264, time: 0.064) G_GAN: 0.664 G_GAN_Feat: 0.872 G_ID: 0.183 G_Rec: 0.342 D_GP: 0.063 D_real: 1.233 D_fake: 0.496 +(epoch: 93, iters: 7664, time: 0.064) G_GAN: 0.685 G_GAN_Feat: 1.040 G_ID: 0.312 G_Rec: 0.446 D_GP: 0.112 D_real: 1.056 D_fake: 0.388 +(epoch: 93, iters: 8064, time: 0.064) G_GAN: 0.856 G_GAN_Feat: 0.879 G_ID: 0.210 G_Rec: 0.381 D_GP: 0.064 D_real: 1.354 D_fake: 0.403 +(epoch: 93, iters: 8464, time: 0.064) G_GAN: 0.301 G_GAN_Feat: 1.010 G_ID: 0.278 G_Rec: 0.413 D_GP: 0.035 D_real: 0.783 D_fake: 0.704 +(epoch: 94, iters: 256, time: 0.064) G_GAN: 0.450 G_GAN_Feat: 0.619 G_ID: 0.221 G_Rec: 0.408 D_GP: 0.020 D_real: 1.334 D_fake: 0.551 +(epoch: 94, iters: 656, time: 0.064) G_GAN: 0.488 G_GAN_Feat: 0.762 G_ID: 0.302 G_Rec: 0.398 D_GP: 0.027 D_real: 1.233 D_fake: 0.514 +(epoch: 94, iters: 1056, time: 0.064) G_GAN: 0.330 G_GAN_Feat: 0.969 G_ID: 0.200 G_Rec: 0.378 D_GP: 0.070 D_real: 0.793 D_fake: 0.670 +(epoch: 94, iters: 1456, time: 0.064) G_GAN: 0.674 G_GAN_Feat: 0.747 G_ID: 0.270 G_Rec: 0.411 D_GP: 0.026 D_real: 1.398 D_fake: 0.345 +(epoch: 94, iters: 1856, time: 0.064) G_GAN: 0.526 G_GAN_Feat: 0.712 G_ID: 0.165 G_Rec: 0.364 D_GP: 0.033 D_real: 1.375 D_fake: 0.475 +(epoch: 94, iters: 2256, time: 0.064) G_GAN: 0.431 G_GAN_Feat: 0.728 G_ID: 0.305 G_Rec: 0.347 D_GP: 0.029 D_real: 1.250 D_fake: 0.569 +(epoch: 94, iters: 2656, time: 0.064) G_GAN: 0.297 G_GAN_Feat: 0.665 G_ID: 0.197 G_Rec: 0.361 D_GP: 0.025 D_real: 1.078 D_fake: 0.704 +(epoch: 94, iters: 3056, time: 0.064) G_GAN: 0.588 G_GAN_Feat: 0.961 G_ID: 0.310 G_Rec: 0.389 D_GP: 0.054 D_real: 1.054 D_fake: 0.423 +(epoch: 94, iters: 3456, time: 0.064) G_GAN: 0.219 G_GAN_Feat: 0.745 G_ID: 0.185 G_Rec: 0.354 D_GP: 0.036 D_real: 0.958 D_fake: 0.781 +(epoch: 94, iters: 3856, time: 0.064) G_GAN: 0.568 G_GAN_Feat: 0.804 G_ID: 0.263 G_Rec: 0.368 D_GP: 0.035 D_real: 1.342 D_fake: 0.443 +(epoch: 94, iters: 4256, time: 0.064) G_GAN: 0.273 G_GAN_Feat: 0.630 G_ID: 0.186 G_Rec: 0.366 D_GP: 0.023 D_real: 1.114 D_fake: 0.728 +(epoch: 94, iters: 4656, time: 0.064) G_GAN: 0.153 G_GAN_Feat: 0.819 G_ID: 0.320 G_Rec: 0.387 D_GP: 0.043 D_real: 0.773 D_fake: 0.847 +(epoch: 94, iters: 5056, time: 0.064) G_GAN: 0.188 G_GAN_Feat: 0.642 G_ID: 0.172 G_Rec: 0.378 D_GP: 0.030 D_real: 1.077 D_fake: 0.813 +(epoch: 94, iters: 5456, time: 0.064) G_GAN: 0.539 G_GAN_Feat: 0.892 G_ID: 0.316 G_Rec: 0.393 D_GP: 0.062 D_real: 1.066 D_fake: 0.499 +(epoch: 94, iters: 5856, time: 0.064) G_GAN: 0.296 G_GAN_Feat: 0.680 G_ID: 0.183 G_Rec: 0.374 D_GP: 0.027 D_real: 1.086 D_fake: 0.704 +(epoch: 94, iters: 6256, time: 0.064) G_GAN: 0.595 G_GAN_Feat: 0.836 G_ID: 0.313 G_Rec: 0.375 D_GP: 0.044 D_real: 1.223 D_fake: 0.409 +(epoch: 94, iters: 6656, time: 0.064) G_GAN: 0.306 G_GAN_Feat: 1.022 G_ID: 0.193 G_Rec: 0.379 D_GP: 0.201 D_real: 0.454 D_fake: 0.694 +(epoch: 94, iters: 7056, time: 0.064) G_GAN: 0.556 G_GAN_Feat: 1.168 G_ID: 0.271 G_Rec: 0.419 D_GP: 0.263 D_real: 0.653 D_fake: 0.457 +(epoch: 94, iters: 7456, time: 0.064) G_GAN: 0.339 G_GAN_Feat: 0.640 G_ID: 0.161 G_Rec: 0.346 D_GP: 0.024 D_real: 1.238 D_fake: 0.662 +(epoch: 94, iters: 7856, time: 0.064) G_GAN: 0.542 G_GAN_Feat: 0.787 G_ID: 0.275 G_Rec: 0.381 D_GP: 0.034 D_real: 1.293 D_fake: 0.488 +(epoch: 94, iters: 8256, time: 0.064) G_GAN: 0.477 G_GAN_Feat: 0.796 G_ID: 0.241 G_Rec: 0.357 D_GP: 0.054 D_real: 1.208 D_fake: 0.526 +(epoch: 95, iters: 48, time: 0.064) G_GAN: 0.480 G_GAN_Feat: 1.107 G_ID: 0.267 G_Rec: 0.435 D_GP: 0.122 D_real: 0.533 D_fake: 0.521 +(epoch: 95, iters: 448, time: 0.064) G_GAN: -0.071 G_GAN_Feat: 0.872 G_ID: 0.200 G_Rec: 0.391 D_GP: 0.070 D_real: 0.561 D_fake: 1.071 +(epoch: 95, iters: 848, time: 0.064) G_GAN: 0.634 G_GAN_Feat: 1.181 G_ID: 0.275 G_Rec: 0.387 D_GP: 0.449 D_real: 0.591 D_fake: 0.382 +(epoch: 95, iters: 1248, time: 0.064) G_GAN: 0.024 G_GAN_Feat: 0.871 G_ID: 0.226 G_Rec: 0.383 D_GP: 0.093 D_real: 0.593 D_fake: 0.976 +(epoch: 95, iters: 1648, time: 0.064) G_GAN: 0.436 G_GAN_Feat: 0.670 G_ID: 0.275 G_Rec: 0.416 D_GP: 0.018 D_real: 1.343 D_fake: 0.567 +(epoch: 95, iters: 2048, time: 0.064) G_GAN: 0.430 G_GAN_Feat: 0.624 G_ID: 0.223 G_Rec: 0.356 D_GP: 0.019 D_real: 1.301 D_fake: 0.574 +(epoch: 95, iters: 2448, time: 0.064) G_GAN: 0.123 G_GAN_Feat: 0.729 G_ID: 0.261 G_Rec: 0.403 D_GP: 0.023 D_real: 0.976 D_fake: 0.877 +(epoch: 95, iters: 2848, time: 0.064) G_GAN: 0.291 G_GAN_Feat: 0.634 G_ID: 0.166 G_Rec: 0.359 D_GP: 0.025 D_real: 1.165 D_fake: 0.709 +(epoch: 95, iters: 3248, time: 0.064) G_GAN: 0.170 G_GAN_Feat: 0.744 G_ID: 0.309 G_Rec: 0.383 D_GP: 0.031 D_real: 0.952 D_fake: 0.830 +(epoch: 95, iters: 3648, time: 0.064) G_GAN: -0.005 G_GAN_Feat: 0.720 G_ID: 0.225 G_Rec: 0.390 D_GP: 0.047 D_real: 0.750 D_fake: 1.005 +(epoch: 95, iters: 4048, time: 0.064) G_GAN: 0.316 G_GAN_Feat: 0.752 G_ID: 0.306 G_Rec: 0.403 D_GP: 0.022 D_real: 0.986 D_fake: 0.685 +(epoch: 95, iters: 4448, time: 0.064) G_GAN: -0.037 G_GAN_Feat: 0.797 G_ID: 0.160 G_Rec: 0.362 D_GP: 0.081 D_real: 0.670 D_fake: 1.037 +(epoch: 95, iters: 4848, time: 0.064) G_GAN: 0.319 G_GAN_Feat: 0.805 G_ID: 0.330 G_Rec: 0.418 D_GP: 0.033 D_real: 1.093 D_fake: 0.682 +(epoch: 95, iters: 5248, time: 0.064) G_GAN: 0.273 G_GAN_Feat: 0.822 G_ID: 0.215 G_Rec: 0.353 D_GP: 0.040 D_real: 0.965 D_fake: 0.727 +(epoch: 95, iters: 5648, time: 0.064) G_GAN: 0.328 G_GAN_Feat: 0.815 G_ID: 0.268 G_Rec: 0.384 D_GP: 0.039 D_real: 1.038 D_fake: 0.677 +(epoch: 95, iters: 6048, time: 0.064) G_GAN: 0.325 G_GAN_Feat: 0.784 G_ID: 0.200 G_Rec: 0.335 D_GP: 0.041 D_real: 0.935 D_fake: 0.675 +(epoch: 95, iters: 6448, time: 0.064) G_GAN: -0.013 G_GAN_Feat: 1.069 G_ID: 0.309 G_Rec: 0.388 D_GP: 0.047 D_real: 0.220 D_fake: 1.013 +(epoch: 95, iters: 6848, time: 0.064) G_GAN: 0.319 G_GAN_Feat: 0.676 G_ID: 0.200 G_Rec: 0.324 D_GP: 0.020 D_real: 1.170 D_fake: 0.682 +(epoch: 95, iters: 7248, time: 0.064) G_GAN: 0.546 G_GAN_Feat: 0.870 G_ID: 0.270 G_Rec: 0.405 D_GP: 0.029 D_real: 1.116 D_fake: 0.457 +(epoch: 95, iters: 7648, time: 0.064) G_GAN: 0.459 G_GAN_Feat: 0.790 G_ID: 0.170 G_Rec: 0.350 D_GP: 0.075 D_real: 1.140 D_fake: 0.549 +(epoch: 95, iters: 8048, time: 0.064) G_GAN: 0.327 G_GAN_Feat: 1.308 G_ID: 0.320 G_Rec: 0.402 D_GP: 0.421 D_real: 0.552 D_fake: 0.678 +(epoch: 95, iters: 8448, time: 0.064) G_GAN: 0.347 G_GAN_Feat: 1.082 G_ID: 0.167 G_Rec: 0.391 D_GP: 0.193 D_real: 0.457 D_fake: 0.653 +(epoch: 96, iters: 240, time: 0.064) G_GAN: 0.447 G_GAN_Feat: 1.021 G_ID: 0.273 G_Rec: 0.432 D_GP: 0.111 D_real: 1.027 D_fake: 0.559 +(epoch: 96, iters: 640, time: 0.064) G_GAN: 0.037 G_GAN_Feat: 0.747 G_ID: 0.173 G_Rec: 0.337 D_GP: 0.031 D_real: 0.857 D_fake: 0.963 +(epoch: 96, iters: 1040, time: 0.064) G_GAN: 0.323 G_GAN_Feat: 0.729 G_ID: 0.336 G_Rec: 0.393 D_GP: 0.022 D_real: 1.153 D_fake: 0.677 +(epoch: 96, iters: 1440, time: 0.064) G_GAN: 0.380 G_GAN_Feat: 0.624 G_ID: 0.165 G_Rec: 0.367 D_GP: 0.020 D_real: 1.308 D_fake: 0.620 +(epoch: 96, iters: 1840, time: 0.064) G_GAN: 0.366 G_GAN_Feat: 0.759 G_ID: 0.307 G_Rec: 0.394 D_GP: 0.027 D_real: 1.066 D_fake: 0.636 +(epoch: 96, iters: 2240, time: 0.064) G_GAN: 0.185 G_GAN_Feat: 0.727 G_ID: 0.190 G_Rec: 0.369 D_GP: 0.048 D_real: 1.012 D_fake: 0.815 +(epoch: 96, iters: 2640, time: 0.064) G_GAN: 0.471 G_GAN_Feat: 0.898 G_ID: 0.277 G_Rec: 0.390 D_GP: 0.069 D_real: 1.019 D_fake: 0.531 +(epoch: 96, iters: 3040, time: 0.064) G_GAN: 0.325 G_GAN_Feat: 0.751 G_ID: 0.196 G_Rec: 0.351 D_GP: 0.050 D_real: 1.047 D_fake: 0.675 +(epoch: 96, iters: 3440, time: 0.064) G_GAN: 0.410 G_GAN_Feat: 0.903 G_ID: 0.279 G_Rec: 0.395 D_GP: 0.043 D_real: 0.955 D_fake: 0.593 +(epoch: 96, iters: 3840, time: 0.064) G_GAN: 0.273 G_GAN_Feat: 0.867 G_ID: 0.190 G_Rec: 0.374 D_GP: 0.048 D_real: 0.782 D_fake: 0.727 +(epoch: 96, iters: 4240, time: 0.064) G_GAN: 0.087 G_GAN_Feat: 0.755 G_ID: 0.293 G_Rec: 0.380 D_GP: 0.028 D_real: 0.813 D_fake: 0.913 +(epoch: 96, iters: 4640, time: 0.064) G_GAN: 0.225 G_GAN_Feat: 0.962 G_ID: 0.168 G_Rec: 0.394 D_GP: 0.141 D_real: 0.509 D_fake: 0.776 +(epoch: 96, iters: 5040, time: 0.064) G_GAN: 0.315 G_GAN_Feat: 0.806 G_ID: 0.297 G_Rec: 0.403 D_GP: 0.035 D_real: 1.100 D_fake: 0.686 +(epoch: 96, iters: 5440, time: 0.064) G_GAN: 0.147 G_GAN_Feat: 0.690 G_ID: 0.183 G_Rec: 0.334 D_GP: 0.024 D_real: 1.002 D_fake: 0.853 +(epoch: 96, iters: 5840, time: 0.064) G_GAN: 0.449 G_GAN_Feat: 1.015 G_ID: 0.274 G_Rec: 0.397 D_GP: 0.139 D_real: 0.762 D_fake: 0.553 +(epoch: 96, iters: 6240, time: 0.064) G_GAN: 0.283 G_GAN_Feat: 0.667 G_ID: 0.185 G_Rec: 0.350 D_GP: 0.027 D_real: 1.164 D_fake: 0.719 +(epoch: 96, iters: 6640, time: 0.064) G_GAN: 0.340 G_GAN_Feat: 0.867 G_ID: 0.264 G_Rec: 0.406 D_GP: 0.044 D_real: 0.978 D_fake: 0.661 +(epoch: 96, iters: 7040, time: 0.064) G_GAN: 0.496 G_GAN_Feat: 0.764 G_ID: 0.189 G_Rec: 0.358 D_GP: 0.031 D_real: 1.257 D_fake: 0.513 +(epoch: 96, iters: 7440, time: 0.064) G_GAN: 0.303 G_GAN_Feat: 0.821 G_ID: 0.243 G_Rec: 0.365 D_GP: 0.034 D_real: 1.058 D_fake: 0.698 +(epoch: 96, iters: 7840, time: 0.064) G_GAN: 0.419 G_GAN_Feat: 0.901 G_ID: 0.174 G_Rec: 0.361 D_GP: 0.043 D_real: 1.159 D_fake: 0.581 +(epoch: 96, iters: 8240, time: 0.064) G_GAN: 0.312 G_GAN_Feat: 0.999 G_ID: 0.304 G_Rec: 0.393 D_GP: 0.088 D_real: 0.528 D_fake: 0.688 +(epoch: 97, iters: 32, time: 0.064) G_GAN: 0.337 G_GAN_Feat: 0.786 G_ID: 0.190 G_Rec: 0.364 D_GP: 0.048 D_real: 1.145 D_fake: 0.663 +(epoch: 97, iters: 432, time: 0.064) G_GAN: 0.140 G_GAN_Feat: 0.799 G_ID: 0.333 G_Rec: 0.397 D_GP: 0.041 D_real: 0.840 D_fake: 0.861 +(epoch: 97, iters: 832, time: 0.064) G_GAN: 0.164 G_GAN_Feat: 1.120 G_ID: 0.218 G_Rec: 0.404 D_GP: 0.161 D_real: 0.382 D_fake: 0.836 +(epoch: 97, iters: 1232, time: 0.064) G_GAN: 0.655 G_GAN_Feat: 0.976 G_ID: 0.276 G_Rec: 0.412 D_GP: 0.050 D_real: 1.069 D_fake: 0.361 +(epoch: 97, iters: 1632, time: 0.064) G_GAN: 0.334 G_GAN_Feat: 0.914 G_ID: 0.200 G_Rec: 0.335 D_GP: 0.046 D_real: 0.964 D_fake: 0.671 +(epoch: 97, iters: 2032, time: 0.064) G_GAN: 0.507 G_GAN_Feat: 0.716 G_ID: 0.228 G_Rec: 0.357 D_GP: 0.024 D_real: 1.333 D_fake: 0.498 +(epoch: 97, iters: 2432, time: 0.064) G_GAN: 0.297 G_GAN_Feat: 0.658 G_ID: 0.186 G_Rec: 0.332 D_GP: 0.025 D_real: 1.149 D_fake: 0.703 +(epoch: 97, iters: 2832, time: 0.064) G_GAN: 0.612 G_GAN_Feat: 0.833 G_ID: 0.295 G_Rec: 0.374 D_GP: 0.035 D_real: 1.345 D_fake: 0.393 +(epoch: 97, iters: 3232, time: 0.064) G_GAN: 0.362 G_GAN_Feat: 0.959 G_ID: 0.160 G_Rec: 0.376 D_GP: 0.054 D_real: 0.683 D_fake: 0.642 +(epoch: 97, iters: 3632, time: 0.064) G_GAN: 0.413 G_GAN_Feat: 0.839 G_ID: 0.302 G_Rec: 0.397 D_GP: 0.032 D_real: 1.124 D_fake: 0.588 +(epoch: 97, iters: 4032, time: 0.064) G_GAN: 0.277 G_GAN_Feat: 0.888 G_ID: 0.150 G_Rec: 0.341 D_GP: 0.102 D_real: 0.751 D_fake: 0.724 +(epoch: 97, iters: 4432, time: 0.064) G_GAN: 0.452 G_GAN_Feat: 0.965 G_ID: 0.296 G_Rec: 0.396 D_GP: 0.045 D_real: 1.031 D_fake: 0.548 +(epoch: 97, iters: 4832, time: 0.064) G_GAN: 0.577 G_GAN_Feat: 0.789 G_ID: 0.209 G_Rec: 0.388 D_GP: 0.062 D_real: 1.308 D_fake: 0.443 +(epoch: 97, iters: 5232, time: 0.064) G_GAN: 0.086 G_GAN_Feat: 0.832 G_ID: 0.285 G_Rec: 0.393 D_GP: 0.032 D_real: 0.723 D_fake: 0.914 +(epoch: 97, iters: 5632, time: 0.064) G_GAN: 0.317 G_GAN_Feat: 0.754 G_ID: 0.187 G_Rec: 0.388 D_GP: 0.041 D_real: 1.057 D_fake: 0.683 +(epoch: 97, iters: 6032, time: 0.064) G_GAN: 0.449 G_GAN_Feat: 0.989 G_ID: 0.318 G_Rec: 0.381 D_GP: 0.086 D_real: 0.670 D_fake: 0.555 +(epoch: 97, iters: 6432, time: 0.064) G_GAN: 0.201 G_GAN_Feat: 0.682 G_ID: 0.179 G_Rec: 0.369 D_GP: 0.023 D_real: 1.106 D_fake: 0.799 +(epoch: 97, iters: 6832, time: 0.064) G_GAN: 0.397 G_GAN_Feat: 0.758 G_ID: 0.270 G_Rec: 0.387 D_GP: 0.025 D_real: 1.109 D_fake: 0.604 +(epoch: 97, iters: 7232, time: 0.064) G_GAN: 0.015 G_GAN_Feat: 0.906 G_ID: 0.178 G_Rec: 0.390 D_GP: 0.190 D_real: 0.594 D_fake: 0.985 +(epoch: 97, iters: 7632, time: 0.064) G_GAN: 0.519 G_GAN_Feat: 0.876 G_ID: 0.317 G_Rec: 0.408 D_GP: 0.050 D_real: 1.175 D_fake: 0.502 +(epoch: 97, iters: 8032, time: 0.064) G_GAN: 0.514 G_GAN_Feat: 0.786 G_ID: 0.173 G_Rec: 0.348 D_GP: 0.033 D_real: 1.290 D_fake: 0.488 +(epoch: 97, iters: 8432, time: 0.064) G_GAN: 0.585 G_GAN_Feat: 1.036 G_ID: 0.307 G_Rec: 0.432 D_GP: 0.125 D_real: 0.794 D_fake: 0.428 +(epoch: 98, iters: 224, time: 0.064) G_GAN: 0.375 G_GAN_Feat: 0.692 G_ID: 0.173 G_Rec: 0.345 D_GP: 0.030 D_real: 1.170 D_fake: 0.626 +(epoch: 98, iters: 624, time: 0.064) G_GAN: 0.352 G_GAN_Feat: 0.762 G_ID: 0.321 G_Rec: 0.407 D_GP: 0.027 D_real: 1.108 D_fake: 0.649 +(epoch: 98, iters: 1024, time: 0.064) G_GAN: 0.177 G_GAN_Feat: 0.962 G_ID: 0.182 G_Rec: 0.399 D_GP: 0.076 D_real: 0.500 D_fake: 0.823 +(epoch: 98, iters: 1424, time: 0.064) G_GAN: 0.450 G_GAN_Feat: 0.978 G_ID: 0.261 G_Rec: 0.443 D_GP: 0.045 D_real: 0.920 D_fake: 0.577 +(epoch: 98, iters: 1824, time: 0.064) G_GAN: 0.319 G_GAN_Feat: 0.927 G_ID: 0.174 G_Rec: 0.391 D_GP: 0.028 D_real: 1.051 D_fake: 0.710 +(epoch: 98, iters: 2224, time: 0.064) G_GAN: 0.167 G_GAN_Feat: 0.780 G_ID: 0.288 G_Rec: 0.402 D_GP: 0.022 D_real: 0.942 D_fake: 0.833 +(epoch: 98, iters: 2624, time: 0.064) G_GAN: 0.144 G_GAN_Feat: 0.835 G_ID: 0.176 G_Rec: 0.361 D_GP: 0.055 D_real: 0.792 D_fake: 0.857 +(epoch: 98, iters: 3024, time: 0.064) G_GAN: 0.242 G_GAN_Feat: 1.069 G_ID: 0.318 G_Rec: 0.459 D_GP: 0.140 D_real: 0.429 D_fake: 0.762 +(epoch: 98, iters: 3424, time: 0.064) G_GAN: 0.611 G_GAN_Feat: 0.826 G_ID: 0.164 G_Rec: 0.374 D_GP: 0.037 D_real: 1.342 D_fake: 0.423 +(epoch: 98, iters: 3824, time: 0.064) G_GAN: 0.473 G_GAN_Feat: 0.926 G_ID: 0.330 G_Rec: 0.387 D_GP: 0.049 D_real: 0.994 D_fake: 0.528 +(epoch: 98, iters: 4224, time: 0.064) G_GAN: 0.285 G_GAN_Feat: 0.760 G_ID: 0.191 G_Rec: 0.357 D_GP: 0.031 D_real: 1.024 D_fake: 0.715 +(epoch: 98, iters: 4624, time: 0.064) G_GAN: 0.594 G_GAN_Feat: 1.090 G_ID: 0.263 G_Rec: 0.401 D_GP: 0.103 D_real: 0.863 D_fake: 0.442 +(epoch: 98, iters: 5024, time: 0.064) G_GAN: 0.049 G_GAN_Feat: 0.795 G_ID: 0.163 G_Rec: 0.347 D_GP: 0.045 D_real: 0.762 D_fake: 0.952 +(epoch: 98, iters: 5424, time: 0.064) G_GAN: 0.344 G_GAN_Feat: 0.824 G_ID: 0.287 G_Rec: 0.372 D_GP: 0.030 D_real: 1.030 D_fake: 0.659 +(epoch: 98, iters: 5824, time: 0.064) G_GAN: 0.084 G_GAN_Feat: 0.910 G_ID: 0.168 G_Rec: 0.425 D_GP: 0.061 D_real: 0.591 D_fake: 0.916 +(epoch: 98, iters: 6224, time: 0.064) G_GAN: 0.270 G_GAN_Feat: 1.198 G_ID: 0.255 G_Rec: 0.386 D_GP: 0.368 D_real: 0.549 D_fake: 0.735 +(epoch: 98, iters: 6624, time: 0.064) G_GAN: 0.399 G_GAN_Feat: 0.751 G_ID: 0.188 G_Rec: 0.372 D_GP: 0.030 D_real: 1.237 D_fake: 0.604 +(epoch: 98, iters: 7024, time: 0.064) G_GAN: 0.154 G_GAN_Feat: 0.812 G_ID: 0.260 G_Rec: 0.401 D_GP: 0.032 D_real: 0.875 D_fake: 0.846 +(epoch: 98, iters: 7424, time: 0.064) G_GAN: 0.246 G_GAN_Feat: 0.841 G_ID: 0.172 G_Rec: 0.356 D_GP: 0.062 D_real: 0.803 D_fake: 0.754 +(epoch: 98, iters: 7824, time: 0.064) G_GAN: 0.498 G_GAN_Feat: 0.709 G_ID: 0.310 G_Rec: 0.393 D_GP: 0.020 D_real: 1.324 D_fake: 0.504 +(epoch: 98, iters: 8224, time: 0.064) G_GAN: 0.024 G_GAN_Feat: 0.717 G_ID: 0.170 G_Rec: 0.374 D_GP: 0.023 D_real: 0.944 D_fake: 0.976 +(epoch: 99, iters: 16, time: 0.064) G_GAN: 0.287 G_GAN_Feat: 0.863 G_ID: 0.308 G_Rec: 0.400 D_GP: 0.034 D_real: 0.918 D_fake: 0.713 +(epoch: 99, iters: 416, time: 0.064) G_GAN: 0.574 G_GAN_Feat: 0.781 G_ID: 0.160 G_Rec: 0.361 D_GP: 0.037 D_real: 1.394 D_fake: 0.434 +(epoch: 99, iters: 816, time: 0.064) G_GAN: 0.298 G_GAN_Feat: 0.903 G_ID: 0.290 G_Rec: 0.366 D_GP: 0.055 D_real: 0.860 D_fake: 0.703 +(epoch: 99, iters: 1216, time: 0.064) G_GAN: 0.735 G_GAN_Feat: 0.858 G_ID: 0.168 G_Rec: 0.344 D_GP: 0.078 D_real: 1.299 D_fake: 0.336 +(epoch: 99, iters: 1616, time: 0.064) G_GAN: 0.372 G_GAN_Feat: 1.304 G_ID: 0.291 G_Rec: 0.384 D_GP: 0.108 D_real: 0.440 D_fake: 0.631 +(epoch: 99, iters: 2016, time: 0.064) G_GAN: 0.296 G_GAN_Feat: 0.617 G_ID: 0.162 G_Rec: 0.391 D_GP: 0.024 D_real: 1.259 D_fake: 0.706 +(epoch: 99, iters: 2416, time: 0.064) G_GAN: 0.228 G_GAN_Feat: 0.957 G_ID: 0.313 G_Rec: 0.451 D_GP: 0.059 D_real: 0.790 D_fake: 0.773 +(epoch: 99, iters: 2816, time: 0.064) G_GAN: 0.308 G_GAN_Feat: 0.720 G_ID: 0.185 G_Rec: 0.355 D_GP: 0.026 D_real: 1.118 D_fake: 0.692 +(epoch: 99, iters: 3216, time: 0.064) G_GAN: 0.482 G_GAN_Feat: 0.816 G_ID: 0.270 G_Rec: 0.356 D_GP: 0.053 D_real: 1.133 D_fake: 0.522 +(epoch: 99, iters: 3616, time: 0.064) G_GAN: 0.042 G_GAN_Feat: 0.754 G_ID: 0.216 G_Rec: 0.380 D_GP: 0.031 D_real: 0.903 D_fake: 0.958 +(epoch: 99, iters: 4016, time: 0.064) G_GAN: 0.394 G_GAN_Feat: 0.868 G_ID: 0.240 G_Rec: 0.386 D_GP: 0.033 D_real: 1.036 D_fake: 0.608 +(epoch: 99, iters: 4416, time: 0.064) G_GAN: 0.330 G_GAN_Feat: 0.806 G_ID: 0.169 G_Rec: 0.362 D_GP: 0.038 D_real: 1.211 D_fake: 0.681 +(epoch: 99, iters: 4816, time: 0.064) G_GAN: 0.413 G_GAN_Feat: 1.025 G_ID: 0.298 G_Rec: 0.392 D_GP: 0.075 D_real: 0.773 D_fake: 0.588 +(epoch: 99, iters: 5216, time: 0.064) G_GAN: 0.349 G_GAN_Feat: 0.957 G_ID: 0.183 G_Rec: 0.356 D_GP: 0.087 D_real: 0.868 D_fake: 0.652 +(epoch: 99, iters: 5616, time: 0.064) G_GAN: 0.260 G_GAN_Feat: 0.986 G_ID: 0.304 G_Rec: 0.422 D_GP: 0.112 D_real: 0.756 D_fake: 0.756 +(epoch: 99, iters: 6016, time: 0.064) G_GAN: 0.633 G_GAN_Feat: 1.019 G_ID: 0.175 G_Rec: 0.400 D_GP: 0.139 D_real: 1.024 D_fake: 0.499 +(epoch: 99, iters: 6416, time: 0.064) G_GAN: 0.263 G_GAN_Feat: 1.122 G_ID: 0.290 G_Rec: 0.450 D_GP: 0.189 D_real: 0.560 D_fake: 0.745 +(epoch: 99, iters: 6816, time: 0.064) G_GAN: 0.071 G_GAN_Feat: 0.854 G_ID: 0.179 G_Rec: 0.342 D_GP: 0.043 D_real: 0.693 D_fake: 0.929 +(epoch: 99, iters: 7216, time: 0.064) G_GAN: 0.434 G_GAN_Feat: 1.059 G_ID: 0.274 G_Rec: 0.374 D_GP: 0.062 D_real: 0.752 D_fake: 0.578 +(epoch: 99, iters: 7616, time: 0.064) G_GAN: 0.691 G_GAN_Feat: 1.015 G_ID: 0.178 G_Rec: 0.370 D_GP: 0.115 D_real: 0.837 D_fake: 0.331 +(epoch: 99, iters: 8016, time: 0.064) G_GAN: 0.423 G_GAN_Feat: 0.936 G_ID: 0.328 G_Rec: 0.374 D_GP: 0.093 D_real: 0.807 D_fake: 0.579 +(epoch: 99, iters: 8416, time: 0.064) G_GAN: 0.324 G_GAN_Feat: 0.873 G_ID: 0.160 G_Rec: 0.381 D_GP: 0.031 D_real: 1.032 D_fake: 0.676 +(epoch: 100, iters: 208, time: 0.064) G_GAN: 0.745 G_GAN_Feat: 0.973 G_ID: 0.293 G_Rec: 0.393 D_GP: 0.050 D_real: 1.343 D_fake: 0.344 +(epoch: 100, iters: 608, time: 0.064) G_GAN: 0.366 G_GAN_Feat: 0.784 G_ID: 0.187 G_Rec: 0.360 D_GP: 0.045 D_real: 1.105 D_fake: 0.634 +(epoch: 100, iters: 1008, time: 0.064) G_GAN: 0.634 G_GAN_Feat: 1.135 G_ID: 0.303 G_Rec: 0.407 D_GP: 0.197 D_real: 0.779 D_fake: 0.448 +(epoch: 100, iters: 1408, time: 0.064) G_GAN: 0.312 G_GAN_Feat: 0.724 G_ID: 0.160 G_Rec: 0.402 D_GP: 0.029 D_real: 1.147 D_fake: 0.689 +(epoch: 100, iters: 1808, time: 0.064) G_GAN: 0.320 G_GAN_Feat: 0.993 G_ID: 0.274 G_Rec: 0.387 D_GP: 0.046 D_real: 0.629 D_fake: 0.681 +(epoch: 100, iters: 2208, time: 0.064) G_GAN: 0.394 G_GAN_Feat: 1.004 G_ID: 0.186 G_Rec: 0.400 D_GP: 0.101 D_real: 0.568 D_fake: 0.763 +(epoch: 100, iters: 2608, time: 0.064) G_GAN: 0.178 G_GAN_Feat: 0.894 G_ID: 0.344 G_Rec: 0.399 D_GP: 0.024 D_real: 0.935 D_fake: 0.822 +(epoch: 100, iters: 3008, time: 0.064) G_GAN: 0.154 G_GAN_Feat: 0.778 G_ID: 0.180 G_Rec: 0.366 D_GP: 0.032 D_real: 1.063 D_fake: 0.847 +(epoch: 100, iters: 3408, time: 0.064) G_GAN: 0.287 G_GAN_Feat: 0.838 G_ID: 0.273 G_Rec: 0.444 D_GP: 0.020 D_real: 1.137 D_fake: 0.713 +(epoch: 100, iters: 3808, time: 0.064) G_GAN: 0.286 G_GAN_Feat: 0.832 G_ID: 0.194 G_Rec: 0.389 D_GP: 0.031 D_real: 1.034 D_fake: 0.714 +(epoch: 100, iters: 4208, time: 0.064) G_GAN: 0.291 G_GAN_Feat: 0.820 G_ID: 0.300 G_Rec: 0.448 D_GP: 0.022 D_real: 1.021 D_fake: 0.710 +(epoch: 100, iters: 4608, time: 0.064) G_GAN: 0.303 G_GAN_Feat: 0.716 G_ID: 0.170 G_Rec: 0.317 D_GP: 0.036 D_real: 1.090 D_fake: 0.698 +(epoch: 100, iters: 5008, time: 0.064) G_GAN: 0.338 G_GAN_Feat: 1.058 G_ID: 0.261 G_Rec: 0.405 D_GP: 0.050 D_real: 0.481 D_fake: 0.662 +(epoch: 100, iters: 5408, time: 0.064) G_GAN: 0.528 G_GAN_Feat: 1.171 G_ID: 0.172 G_Rec: 0.401 D_GP: 0.259 D_real: 1.067 D_fake: 0.523 +(epoch: 100, iters: 5808, time: 0.064) G_GAN: 0.593 G_GAN_Feat: 0.989 G_ID: 0.294 G_Rec: 0.399 D_GP: 0.029 D_real: 1.040 D_fake: 0.411 +(epoch: 100, iters: 6208, time: 0.064) G_GAN: 0.492 G_GAN_Feat: 0.846 G_ID: 0.172 G_Rec: 0.403 D_GP: 0.049 D_real: 1.210 D_fake: 0.552 +(epoch: 100, iters: 6608, time: 0.064) G_GAN: 0.238 G_GAN_Feat: 0.954 G_ID: 0.265 G_Rec: 0.454 D_GP: 0.033 D_real: 0.674 D_fake: 0.762 +(epoch: 100, iters: 7008, time: 0.064) G_GAN: 0.214 G_GAN_Feat: 0.745 G_ID: 0.151 G_Rec: 0.356 D_GP: 0.022 D_real: 1.046 D_fake: 0.786 +(epoch: 100, iters: 7408, time: 0.064) G_GAN: 0.106 G_GAN_Feat: 0.738 G_ID: 0.286 G_Rec: 0.356 D_GP: 0.027 D_real: 0.928 D_fake: 0.894 +(epoch: 100, iters: 7808, time: 0.064) G_GAN: 0.134 G_GAN_Feat: 1.120 G_ID: 0.187 G_Rec: 0.370 D_GP: 0.102 D_real: 0.413 D_fake: 0.866 +(epoch: 100, iters: 8208, time: 0.064) G_GAN: 0.051 G_GAN_Feat: 1.090 G_ID: 0.322 G_Rec: 0.414 D_GP: 0.390 D_real: 0.526 D_fake: 0.950 +(epoch: 100, iters: 8608, time: 0.064) G_GAN: 0.190 G_GAN_Feat: 0.715 G_ID: 0.187 G_Rec: 0.339 D_GP: 0.033 D_real: 1.008 D_fake: 0.810 +(epoch: 101, iters: 400, time: 0.064) G_GAN: 0.617 G_GAN_Feat: 0.800 G_ID: 0.295 G_Rec: 0.403 D_GP: 0.030 D_real: 1.377 D_fake: 0.389 +(epoch: 101, iters: 800, time: 0.064) G_GAN: 0.827 G_GAN_Feat: 0.825 G_ID: 0.156 G_Rec: 0.389 D_GP: 0.021 D_real: 1.616 D_fake: 0.234 +(epoch: 101, iters: 1200, time: 0.064) G_GAN: 0.331 G_GAN_Feat: 0.776 G_ID: 0.267 G_Rec: 0.404 D_GP: 0.029 D_real: 1.045 D_fake: 0.670 +(epoch: 101, iters: 1600, time: 0.064) G_GAN: 0.249 G_GAN_Feat: 0.865 G_ID: 0.164 G_Rec: 0.411 D_GP: 0.067 D_real: 0.733 D_fake: 0.752 +(epoch: 101, iters: 2000, time: 0.064) G_GAN: 0.372 G_GAN_Feat: 0.930 G_ID: 0.279 G_Rec: 0.407 D_GP: 0.084 D_real: 0.772 D_fake: 0.629 +(epoch: 101, iters: 2400, time: 0.064) G_GAN: 0.677 G_GAN_Feat: 0.845 G_ID: 0.218 G_Rec: 0.335 D_GP: 0.041 D_real: 1.510 D_fake: 0.357 +(epoch: 101, iters: 2800, time: 0.064) G_GAN: 0.293 G_GAN_Feat: 1.006 G_ID: 0.289 G_Rec: 0.417 D_GP: 0.036 D_real: 0.857 D_fake: 0.707 +(epoch: 101, iters: 3200, time: 0.064) G_GAN: 0.164 G_GAN_Feat: 0.875 G_ID: 0.176 G_Rec: 0.368 D_GP: 0.027 D_real: 1.063 D_fake: 0.836 +(epoch: 101, iters: 3600, time: 0.064) G_GAN: 0.449 G_GAN_Feat: 1.347 G_ID: 0.275 G_Rec: 0.463 D_GP: 0.059 D_real: 0.580 D_fake: 0.560 +(epoch: 101, iters: 4000, time: 0.064) G_GAN: 0.352 G_GAN_Feat: 0.705 G_ID: 0.172 G_Rec: 0.354 D_GP: 0.028 D_real: 1.209 D_fake: 0.648 +(epoch: 101, iters: 4400, time: 0.064) G_GAN: 0.561 G_GAN_Feat: 1.155 G_ID: 0.265 G_Rec: 0.396 D_GP: 0.069 D_real: 0.738 D_fake: 0.443 +(epoch: 101, iters: 4800, time: 0.064) G_GAN: 0.138 G_GAN_Feat: 0.706 G_ID: 0.188 G_Rec: 0.347 D_GP: 0.024 D_real: 0.971 D_fake: 0.862 +(epoch: 101, iters: 5200, time: 0.064) G_GAN: 0.245 G_GAN_Feat: 1.010 G_ID: 0.255 G_Rec: 0.434 D_GP: 0.104 D_real: 0.591 D_fake: 0.755 +(epoch: 101, iters: 5600, time: 0.064) G_GAN: 0.176 G_GAN_Feat: 0.700 G_ID: 0.185 G_Rec: 0.308 D_GP: 0.025 D_real: 1.125 D_fake: 0.824 +(epoch: 101, iters: 6000, time: 0.064) G_GAN: 0.726 G_GAN_Feat: 1.096 G_ID: 0.244 G_Rec: 0.431 D_GP: 0.114 D_real: 0.923 D_fake: 0.311 +(epoch: 101, iters: 6400, time: 0.064) G_GAN: 0.841 G_GAN_Feat: 0.856 G_ID: 0.164 G_Rec: 0.372 D_GP: 0.031 D_real: 1.522 D_fake: 0.360 +(epoch: 101, iters: 6800, time: 0.064) G_GAN: 0.374 G_GAN_Feat: 0.881 G_ID: 0.274 G_Rec: 0.416 D_GP: 0.042 D_real: 1.061 D_fake: 0.633 +(epoch: 101, iters: 7200, time: 0.064) G_GAN: 0.447 G_GAN_Feat: 0.779 G_ID: 0.164 G_Rec: 0.350 D_GP: 0.035 D_real: 1.232 D_fake: 0.557 +(epoch: 101, iters: 7600, time: 0.064) G_GAN: 0.196 G_GAN_Feat: 0.907 G_ID: 0.254 G_Rec: 0.397 D_GP: 0.039 D_real: 0.846 D_fake: 0.804 +(epoch: 101, iters: 8000, time: 0.064) G_GAN: 0.217 G_GAN_Feat: 0.740 G_ID: 0.193 G_Rec: 0.423 D_GP: 0.020 D_real: 1.117 D_fake: 0.783 +(epoch: 101, iters: 8400, time: 0.064) G_GAN: 0.600 G_GAN_Feat: 0.707 G_ID: 0.268 G_Rec: 0.473 D_GP: 0.020 D_real: 1.385 D_fake: 0.405 +(epoch: 102, iters: 192, time: 0.064) G_GAN: 0.387 G_GAN_Feat: 0.614 G_ID: 0.199 G_Rec: 0.328 D_GP: 0.021 D_real: 1.336 D_fake: 0.613 +(epoch: 102, iters: 592, time: 0.064) G_GAN: 0.275 G_GAN_Feat: 0.735 G_ID: 0.252 G_Rec: 0.363 D_GP: 0.026 D_real: 1.138 D_fake: 0.725 +(epoch: 102, iters: 992, time: 0.064) G_GAN: 0.278 G_GAN_Feat: 0.637 G_ID: 0.204 G_Rec: 0.351 D_GP: 0.027 D_real: 1.137 D_fake: 0.722 +(epoch: 102, iters: 1392, time: 0.064) G_GAN: 0.544 G_GAN_Feat: 0.850 G_ID: 0.243 G_Rec: 0.413 D_GP: 0.042 D_real: 1.137 D_fake: 0.465 +(epoch: 102, iters: 1792, time: 0.064) G_GAN: -0.030 G_GAN_Feat: 0.881 G_ID: 0.199 G_Rec: 0.389 D_GP: 0.272 D_real: 0.540 D_fake: 1.030 +(epoch: 102, iters: 2192, time: 0.064) G_GAN: 0.541 G_GAN_Feat: 0.794 G_ID: 0.283 G_Rec: 0.402 D_GP: 0.025 D_real: 1.340 D_fake: 0.479 +(epoch: 102, iters: 2592, time: 0.064) G_GAN: 0.261 G_GAN_Feat: 0.798 G_ID: 0.193 G_Rec: 0.369 D_GP: 0.042 D_real: 0.971 D_fake: 0.739 +(epoch: 102, iters: 2992, time: 0.064) G_GAN: 0.457 G_GAN_Feat: 0.913 G_ID: 0.269 G_Rec: 0.403 D_GP: 0.078 D_real: 0.838 D_fake: 0.544 +(epoch: 102, iters: 3392, time: 0.064) G_GAN: 0.343 G_GAN_Feat: 0.891 G_ID: 0.159 G_Rec: 0.378 D_GP: 0.122 D_real: 0.953 D_fake: 0.658 +(epoch: 102, iters: 3792, time: 0.064) G_GAN: 0.214 G_GAN_Feat: 0.992 G_ID: 0.293 G_Rec: 0.440 D_GP: 0.084 D_real: 0.596 D_fake: 0.787 +(epoch: 102, iters: 4192, time: 0.064) G_GAN: 0.234 G_GAN_Feat: 0.735 G_ID: 0.196 G_Rec: 0.410 D_GP: 0.040 D_real: 1.128 D_fake: 0.777 +(epoch: 102, iters: 4592, time: 0.064) G_GAN: 0.449 G_GAN_Feat: 0.789 G_ID: 0.249 G_Rec: 0.370 D_GP: 0.034 D_real: 1.126 D_fake: 0.553 +(epoch: 102, iters: 4992, time: 0.064) G_GAN: 0.486 G_GAN_Feat: 0.874 G_ID: 0.178 G_Rec: 0.344 D_GP: 0.063 D_real: 1.047 D_fake: 0.516 +(epoch: 102, iters: 5392, time: 0.064) G_GAN: 0.351 G_GAN_Feat: 0.783 G_ID: 0.300 G_Rec: 0.440 D_GP: 0.026 D_real: 1.064 D_fake: 0.649 +(epoch: 102, iters: 5792, time: 0.064) G_GAN: 0.228 G_GAN_Feat: 0.726 G_ID: 0.172 G_Rec: 0.366 D_GP: 0.034 D_real: 1.069 D_fake: 0.772 +(epoch: 102, iters: 6192, time: 0.064) G_GAN: 0.412 G_GAN_Feat: 0.853 G_ID: 0.304 G_Rec: 0.425 D_GP: 0.033 D_real: 1.101 D_fake: 0.594 +(epoch: 102, iters: 6592, time: 0.064) G_GAN: 0.171 G_GAN_Feat: 0.798 G_ID: 0.186 G_Rec: 0.356 D_GP: 0.052 D_real: 0.867 D_fake: 0.829 +(epoch: 102, iters: 6992, time: 0.064) G_GAN: 0.260 G_GAN_Feat: 1.008 G_ID: 0.275 G_Rec: 0.400 D_GP: 0.086 D_real: 0.571 D_fake: 0.740 +(epoch: 102, iters: 7392, time: 0.064) G_GAN: 0.566 G_GAN_Feat: 0.804 G_ID: 0.169 G_Rec: 0.334 D_GP: 0.064 D_real: 1.192 D_fake: 0.440 +(epoch: 102, iters: 7792, time: 0.064) G_GAN: 0.551 G_GAN_Feat: 0.985 G_ID: 0.289 G_Rec: 0.429 D_GP: 0.039 D_real: 0.975 D_fake: 0.466 +(epoch: 102, iters: 8192, time: 0.064) G_GAN: 0.132 G_GAN_Feat: 0.913 G_ID: 0.183 G_Rec: 0.347 D_GP: 0.038 D_real: 0.659 D_fake: 0.868 +(epoch: 102, iters: 8592, time: 0.064) G_GAN: 0.547 G_GAN_Feat: 1.145 G_ID: 0.249 G_Rec: 0.424 D_GP: 0.023 D_real: 1.071 D_fake: 0.454 +(epoch: 103, iters: 384, time: 0.064) G_GAN: 0.245 G_GAN_Feat: 0.950 G_ID: 0.172 G_Rec: 0.396 D_GP: 0.473 D_real: 0.577 D_fake: 0.756 +(epoch: 103, iters: 784, time: 0.064) G_GAN: 0.080 G_GAN_Feat: 0.992 G_ID: 0.283 G_Rec: 0.395 D_GP: 0.121 D_real: 0.435 D_fake: 0.921 +(epoch: 103, iters: 1184, time: 0.064) G_GAN: 0.067 G_GAN_Feat: 0.902 G_ID: 0.231 G_Rec: 0.362 D_GP: 0.087 D_real: 0.705 D_fake: 0.934 +(epoch: 103, iters: 1584, time: 0.064) G_GAN: 0.810 G_GAN_Feat: 0.876 G_ID: 0.301 G_Rec: 0.390 D_GP: 0.033 D_real: 1.448 D_fake: 0.235 +(epoch: 103, iters: 1984, time: 0.064) G_GAN: 0.309 G_GAN_Feat: 0.710 G_ID: 0.223 G_Rec: 0.331 D_GP: 0.023 D_real: 1.098 D_fake: 0.693 +(epoch: 103, iters: 2384, time: 0.064) G_GAN: 0.593 G_GAN_Feat: 0.711 G_ID: 0.249 G_Rec: 0.351 D_GP: 0.023 D_real: 1.355 D_fake: 0.412 +(epoch: 103, iters: 2784, time: 0.064) G_GAN: 0.417 G_GAN_Feat: 0.748 G_ID: 0.163 G_Rec: 0.356 D_GP: 0.034 D_real: 1.219 D_fake: 0.583 +(epoch: 103, iters: 3184, time: 0.064) G_GAN: 0.353 G_GAN_Feat: 1.020 G_ID: 0.293 G_Rec: 0.392 D_GP: 0.062 D_real: 0.639 D_fake: 0.648 +(epoch: 103, iters: 3584, time: 0.064) G_GAN: 0.350 G_GAN_Feat: 0.620 G_ID: 0.187 G_Rec: 0.345 D_GP: 0.020 D_real: 1.312 D_fake: 0.650 +(epoch: 103, iters: 3984, time: 0.064) G_GAN: 0.271 G_GAN_Feat: 0.813 G_ID: 0.269 G_Rec: 0.454 D_GP: 0.028 D_real: 0.983 D_fake: 0.733 +(epoch: 103, iters: 4384, time: 0.064) G_GAN: 0.246 G_GAN_Feat: 0.795 G_ID: 0.173 G_Rec: 0.472 D_GP: 0.052 D_real: 1.046 D_fake: 0.754 +(epoch: 103, iters: 4784, time: 0.064) G_GAN: 0.541 G_GAN_Feat: 0.925 G_ID: 0.264 G_Rec: 0.421 D_GP: 0.045 D_real: 1.124 D_fake: 0.480 +(epoch: 103, iters: 5184, time: 0.064) G_GAN: 0.149 G_GAN_Feat: 0.756 G_ID: 0.178 G_Rec: 0.365 D_GP: 0.024 D_real: 0.986 D_fake: 0.851 +(epoch: 103, iters: 5584, time: 0.064) G_GAN: 0.764 G_GAN_Feat: 0.828 G_ID: 0.283 G_Rec: 0.375 D_GP: 0.032 D_real: 1.475 D_fake: 0.270 +(epoch: 103, iters: 5984, time: 0.064) G_GAN: 0.245 G_GAN_Feat: 1.027 G_ID: 0.161 G_Rec: 0.371 D_GP: 0.055 D_real: 0.583 D_fake: 0.756 +(epoch: 103, iters: 6384, time: 0.064) G_GAN: 0.269 G_GAN_Feat: 0.818 G_ID: 0.280 G_Rec: 0.381 D_GP: 0.039 D_real: 0.959 D_fake: 0.731 +(epoch: 103, iters: 6784, time: 0.064) G_GAN: 0.474 G_GAN_Feat: 0.754 G_ID: 0.199 G_Rec: 0.320 D_GP: 0.042 D_real: 1.219 D_fake: 0.531 +(epoch: 103, iters: 7184, time: 0.064) G_GAN: 0.318 G_GAN_Feat: 0.889 G_ID: 0.266 G_Rec: 0.389 D_GP: 0.029 D_real: 0.958 D_fake: 0.682 +(epoch: 103, iters: 7584, time: 0.064) G_GAN: 0.349 G_GAN_Feat: 0.814 G_ID: 0.190 G_Rec: 0.326 D_GP: 0.052 D_real: 0.972 D_fake: 0.652 +(epoch: 103, iters: 7984, time: 0.064) G_GAN: 0.452 G_GAN_Feat: 0.838 G_ID: 0.258 G_Rec: 0.439 D_GP: 0.027 D_real: 1.166 D_fake: 0.551 +(epoch: 103, iters: 8384, time: 0.064) G_GAN: 0.084 G_GAN_Feat: 0.758 G_ID: 0.169 G_Rec: 0.355 D_GP: 0.044 D_real: 0.950 D_fake: 0.916 +(epoch: 104, iters: 176, time: 0.064) G_GAN: 0.281 G_GAN_Feat: 0.813 G_ID: 0.330 G_Rec: 0.369 D_GP: 0.030 D_real: 1.014 D_fake: 0.720 +(epoch: 104, iters: 576, time: 0.064) G_GAN: 0.481 G_GAN_Feat: 0.658 G_ID: 0.181 G_Rec: 0.326 D_GP: 0.028 D_real: 1.349 D_fake: 0.520 +(epoch: 104, iters: 976, time: 0.064) G_GAN: 0.529 G_GAN_Feat: 0.759 G_ID: 0.259 G_Rec: 0.375 D_GP: 0.035 D_real: 1.252 D_fake: 0.480 +(epoch: 104, iters: 1376, time: 0.064) G_GAN: 0.216 G_GAN_Feat: 0.884 G_ID: 0.152 G_Rec: 0.326 D_GP: 0.236 D_real: 0.509 D_fake: 0.785 +(epoch: 104, iters: 1776, time: 0.064) G_GAN: 0.594 G_GAN_Feat: 0.768 G_ID: 0.271 G_Rec: 0.433 D_GP: 0.023 D_real: 1.310 D_fake: 0.413 +(epoch: 104, iters: 2176, time: 0.064) G_GAN: 0.355 G_GAN_Feat: 0.871 G_ID: 0.170 G_Rec: 0.353 D_GP: 0.033 D_real: 1.054 D_fake: 0.654 +(epoch: 104, iters: 2576, time: 0.064) G_GAN: 0.501 G_GAN_Feat: 0.875 G_ID: 0.317 G_Rec: 0.383 D_GP: 0.026 D_real: 1.253 D_fake: 0.515 +(epoch: 104, iters: 2976, time: 0.064) G_GAN: 0.122 G_GAN_Feat: 1.163 G_ID: 0.174 G_Rec: 0.383 D_GP: 3.192 D_real: 0.560 D_fake: 0.883 +(epoch: 104, iters: 3376, time: 0.064) G_GAN: 0.313 G_GAN_Feat: 0.641 G_ID: 0.301 G_Rec: 0.345 D_GP: 0.020 D_real: 1.168 D_fake: 0.687 +(epoch: 104, iters: 3776, time: 0.064) G_GAN: 0.359 G_GAN_Feat: 0.651 G_ID: 0.160 G_Rec: 0.372 D_GP: 0.022 D_real: 1.216 D_fake: 0.641 +(epoch: 104, iters: 4176, time: 0.064) G_GAN: 0.439 G_GAN_Feat: 0.669 G_ID: 0.273 G_Rec: 0.365 D_GP: 0.023 D_real: 1.266 D_fake: 0.563 +(epoch: 104, iters: 4576, time: 0.064) G_GAN: 0.319 G_GAN_Feat: 0.618 G_ID: 0.188 G_Rec: 0.331 D_GP: 0.023 D_real: 1.231 D_fake: 0.681 +(epoch: 104, iters: 4976, time: 0.064) G_GAN: 0.344 G_GAN_Feat: 0.845 G_ID: 0.243 G_Rec: 0.396 D_GP: 0.038 D_real: 1.059 D_fake: 0.660 +(epoch: 104, iters: 5376, time: 0.064) G_GAN: 0.354 G_GAN_Feat: 0.752 G_ID: 0.172 G_Rec: 0.366 D_GP: 0.055 D_real: 1.005 D_fake: 0.649 +(epoch: 104, iters: 5776, time: 0.064) G_GAN: 0.176 G_GAN_Feat: 0.883 G_ID: 0.242 G_Rec: 0.412 D_GP: 0.057 D_real: 0.856 D_fake: 0.824 +(epoch: 104, iters: 6176, time: 0.064) G_GAN: 0.301 G_GAN_Feat: 0.955 G_ID: 0.187 G_Rec: 0.385 D_GP: 0.080 D_real: 0.628 D_fake: 0.699 +(epoch: 104, iters: 6576, time: 0.064) G_GAN: 0.308 G_GAN_Feat: 1.101 G_ID: 0.229 G_Rec: 0.444 D_GP: 0.232 D_real: 0.509 D_fake: 0.699 +(epoch: 104, iters: 6976, time: 0.064) G_GAN: 0.230 G_GAN_Feat: 0.830 G_ID: 0.160 G_Rec: 0.349 D_GP: 0.056 D_real: 1.042 D_fake: 0.771 +(epoch: 104, iters: 7376, time: 0.064) G_GAN: 0.187 G_GAN_Feat: 0.917 G_ID: 0.287 G_Rec: 0.409 D_GP: 0.076 D_real: 0.739 D_fake: 0.813 +(epoch: 104, iters: 7776, time: 0.064) G_GAN: 0.474 G_GAN_Feat: 0.765 G_ID: 0.143 G_Rec: 0.377 D_GP: 0.034 D_real: 1.172 D_fake: 0.529 +(epoch: 104, iters: 8176, time: 0.064) G_GAN: 0.774 G_GAN_Feat: 0.906 G_ID: 0.255 G_Rec: 0.397 D_GP: 0.041 D_real: 1.385 D_fake: 0.283 +(epoch: 104, iters: 8576, time: 0.064) G_GAN: 0.160 G_GAN_Feat: 0.977 G_ID: 0.200 G_Rec: 0.367 D_GP: 0.112 D_real: 0.408 D_fake: 0.841 +(epoch: 105, iters: 368, time: 0.064) G_GAN: 0.466 G_GAN_Feat: 0.988 G_ID: 0.312 G_Rec: 0.369 D_GP: 0.047 D_real: 0.992 D_fake: 0.534 +(epoch: 105, iters: 768, time: 0.064) G_GAN: 0.310 G_GAN_Feat: 0.855 G_ID: 0.160 G_Rec: 0.341 D_GP: 0.058 D_real: 1.197 D_fake: 0.694 +(epoch: 105, iters: 1168, time: 0.064) G_GAN: 0.398 G_GAN_Feat: 0.937 G_ID: 0.295 G_Rec: 0.389 D_GP: 0.036 D_real: 1.003 D_fake: 0.603 +(epoch: 105, iters: 1568, time: 0.064) G_GAN: 0.094 G_GAN_Feat: 0.936 G_ID: 0.177 G_Rec: 0.362 D_GP: 0.100 D_real: 0.530 D_fake: 0.906 +(epoch: 105, iters: 1968, time: 0.064) G_GAN: 0.638 G_GAN_Feat: 1.135 G_ID: 0.294 G_Rec: 0.422 D_GP: 0.068 D_real: 0.757 D_fake: 0.372 +(epoch: 105, iters: 2368, time: 0.064) G_GAN: -0.263 G_GAN_Feat: 0.875 G_ID: 0.209 G_Rec: 0.430 D_GP: 0.119 D_real: 0.577 D_fake: 1.268 +(epoch: 105, iters: 2768, time: 0.064) G_GAN: 0.553 G_GAN_Feat: 0.841 G_ID: 0.251 G_Rec: 0.379 D_GP: 0.031 D_real: 1.313 D_fake: 0.449 +(epoch: 105, iters: 3168, time: 0.064) G_GAN: 0.213 G_GAN_Feat: 0.823 G_ID: 0.170 G_Rec: 0.338 D_GP: 0.042 D_real: 0.904 D_fake: 0.788 +(epoch: 105, iters: 3568, time: 0.064) G_GAN: 0.209 G_GAN_Feat: 0.936 G_ID: 0.287 G_Rec: 0.418 D_GP: 0.037 D_real: 0.769 D_fake: 0.791 +(epoch: 105, iters: 3968, time: 0.064) G_GAN: -0.465 G_GAN_Feat: 0.915 G_ID: 0.215 G_Rec: 0.360 D_GP: 0.115 D_real: 0.177 D_fake: 1.465 +(epoch: 105, iters: 4368, time: 0.064) G_GAN: 0.118 G_GAN_Feat: 0.964 G_ID: 0.273 G_Rec: 0.424 D_GP: 0.080 D_real: 0.593 D_fake: 0.882 +(epoch: 105, iters: 4768, time: 0.064) G_GAN: 0.383 G_GAN_Feat: 0.767 G_ID: 0.154 G_Rec: 0.326 D_GP: 0.027 D_real: 1.232 D_fake: 0.617 +(epoch: 105, iters: 5168, time: 0.064) G_GAN: 0.327 G_GAN_Feat: 0.841 G_ID: 0.326 G_Rec: 0.405 D_GP: 0.023 D_real: 1.116 D_fake: 0.673 +(epoch: 105, iters: 5568, time: 0.064) G_GAN: 0.210 G_GAN_Feat: 0.789 G_ID: 0.177 G_Rec: 0.366 D_GP: 0.031 D_real: 0.988 D_fake: 0.791 +(epoch: 105, iters: 5968, time: 0.064) G_GAN: 0.332 G_GAN_Feat: 0.892 G_ID: 0.223 G_Rec: 0.417 D_GP: 0.028 D_real: 0.935 D_fake: 0.669 +(epoch: 105, iters: 6368, time: 0.064) G_GAN: 0.220 G_GAN_Feat: 0.750 G_ID: 0.191 G_Rec: 0.349 D_GP: 0.050 D_real: 0.942 D_fake: 0.782 +(epoch: 105, iters: 6768, time: 0.064) G_GAN: 0.369 G_GAN_Feat: 0.811 G_ID: 0.257 G_Rec: 0.398 D_GP: 0.032 D_real: 0.985 D_fake: 0.631 +(epoch: 105, iters: 7168, time: 0.064) G_GAN: 0.244 G_GAN_Feat: 1.161 G_ID: 0.184 G_Rec: 0.379 D_GP: 0.848 D_real: 0.369 D_fake: 0.756 +(epoch: 105, iters: 7568, time: 0.064) G_GAN: 0.496 G_GAN_Feat: 1.034 G_ID: 0.285 G_Rec: 0.404 D_GP: 0.031 D_real: 1.040 D_fake: 0.507 +(epoch: 105, iters: 7968, time: 0.064) G_GAN: 0.415 G_GAN_Feat: 0.724 G_ID: 0.142 G_Rec: 0.324 D_GP: 0.027 D_real: 1.330 D_fake: 0.586 +(epoch: 105, iters: 8368, time: 0.064) G_GAN: 0.238 G_GAN_Feat: 1.019 G_ID: 0.251 G_Rec: 0.373 D_GP: 0.107 D_real: 0.283 D_fake: 0.762 +(epoch: 106, iters: 160, time: 0.064) G_GAN: 0.389 G_GAN_Feat: 0.857 G_ID: 0.177 G_Rec: 0.412 D_GP: 0.038 D_real: 1.185 D_fake: 0.666 +(epoch: 106, iters: 560, time: 0.064) G_GAN: 0.533 G_GAN_Feat: 0.852 G_ID: 0.266 G_Rec: 0.383 D_GP: 0.032 D_real: 1.239 D_fake: 0.488 +(epoch: 106, iters: 960, time: 0.064) G_GAN: 0.221 G_GAN_Feat: 1.048 G_ID: 0.180 G_Rec: 0.395 D_GP: 0.095 D_real: 0.549 D_fake: 0.780 +(epoch: 106, iters: 1360, time: 0.064) G_GAN: 0.531 G_GAN_Feat: 0.871 G_ID: 0.291 G_Rec: 0.406 D_GP: 0.022 D_real: 1.202 D_fake: 0.471 +(epoch: 106, iters: 1760, time: 0.064) G_GAN: 0.286 G_GAN_Feat: 0.747 G_ID: 0.214 G_Rec: 0.326 D_GP: 0.036 D_real: 1.089 D_fake: 0.714 +(epoch: 106, iters: 2160, time: 0.064) G_GAN: 0.307 G_GAN_Feat: 0.871 G_ID: 0.285 G_Rec: 0.416 D_GP: 0.029 D_real: 1.122 D_fake: 0.697 +(epoch: 106, iters: 2560, time: 0.064) G_GAN: 0.483 G_GAN_Feat: 0.910 G_ID: 0.170 G_Rec: 0.365 D_GP: 0.059 D_real: 1.072 D_fake: 0.521 +(epoch: 106, iters: 2960, time: 0.064) G_GAN: 0.278 G_GAN_Feat: 0.879 G_ID: 0.282 G_Rec: 0.397 D_GP: 0.051 D_real: 0.952 D_fake: 0.724 +(epoch: 106, iters: 3360, time: 0.064) G_GAN: 0.134 G_GAN_Feat: 0.712 G_ID: 0.169 G_Rec: 0.321 D_GP: 0.029 D_real: 0.979 D_fake: 0.866 +(epoch: 106, iters: 3760, time: 0.064) G_GAN: 0.432 G_GAN_Feat: 0.968 G_ID: 0.324 G_Rec: 0.408 D_GP: 0.061 D_real: 0.952 D_fake: 0.569 +(epoch: 106, iters: 4160, time: 0.064) G_GAN: -0.348 G_GAN_Feat: 0.800 G_ID: 0.159 G_Rec: 0.384 D_GP: 0.033 D_real: 0.446 D_fake: 1.348 +(epoch: 106, iters: 4560, time: 0.064) G_GAN: 0.044 G_GAN_Feat: 0.852 G_ID: 0.259 G_Rec: 0.435 D_GP: 0.031 D_real: 0.684 D_fake: 0.956 +(epoch: 106, iters: 4960, time: 0.064) G_GAN: 0.505 G_GAN_Feat: 0.654 G_ID: 0.183 G_Rec: 0.347 D_GP: 0.021 D_real: 1.380 D_fake: 0.500 +(epoch: 106, iters: 5360, time: 0.064) G_GAN: 0.603 G_GAN_Feat: 0.828 G_ID: 0.222 G_Rec: 0.407 D_GP: 0.027 D_real: 1.258 D_fake: 0.406 +(epoch: 106, iters: 5760, time: 0.064) G_GAN: 0.260 G_GAN_Feat: 0.636 G_ID: 0.189 G_Rec: 0.346 D_GP: 0.024 D_real: 1.170 D_fake: 0.740 +(epoch: 106, iters: 6160, time: 0.064) G_GAN: 0.412 G_GAN_Feat: 1.017 G_ID: 0.258 G_Rec: 0.396 D_GP: 0.080 D_real: 0.736 D_fake: 0.590 +(epoch: 106, iters: 6560, time: 0.064) G_GAN: 0.508 G_GAN_Feat: 0.911 G_ID: 0.159 G_Rec: 0.346 D_GP: 0.039 D_real: 1.035 D_fake: 0.493 +(epoch: 106, iters: 6960, time: 0.064) G_GAN: 0.619 G_GAN_Feat: 1.052 G_ID: 0.234 G_Rec: 0.415 D_GP: 0.119 D_real: 0.731 D_fake: 0.438 +(epoch: 106, iters: 7360, time: 0.064) G_GAN: -0.261 G_GAN_Feat: 1.081 G_ID: 0.160 G_Rec: 0.384 D_GP: 0.074 D_real: 0.097 D_fake: 1.261 +(epoch: 106, iters: 7760, time: 0.064) G_GAN: 0.507 G_GAN_Feat: 1.044 G_ID: 0.260 G_Rec: 0.399 D_GP: 0.057 D_real: 0.886 D_fake: 0.495 +(epoch: 106, iters: 8160, time: 0.064) G_GAN: -0.028 G_GAN_Feat: 0.749 G_ID: 0.176 G_Rec: 0.369 D_GP: 0.022 D_real: 0.792 D_fake: 1.028 +(epoch: 106, iters: 8560, time: 0.064) G_GAN: 0.315 G_GAN_Feat: 0.913 G_ID: 0.277 G_Rec: 0.442 D_GP: 0.038 D_real: 0.839 D_fake: 0.688 +(epoch: 107, iters: 352, time: 0.064) G_GAN: 0.262 G_GAN_Feat: 1.035 G_ID: 0.147 G_Rec: 0.361 D_GP: 0.133 D_real: 0.387 D_fake: 0.739 +(epoch: 107, iters: 752, time: 0.064) G_GAN: 0.186 G_GAN_Feat: 0.924 G_ID: 0.286 G_Rec: 0.416 D_GP: 0.035 D_real: 0.921 D_fake: 0.814 +(epoch: 107, iters: 1152, time: 0.064) G_GAN: 0.353 G_GAN_Feat: 0.802 G_ID: 0.155 G_Rec: 0.352 D_GP: 0.087 D_real: 1.089 D_fake: 0.647 +(epoch: 107, iters: 1552, time: 0.064) G_GAN: 0.377 G_GAN_Feat: 0.861 G_ID: 0.287 G_Rec: 0.416 D_GP: 0.061 D_real: 0.953 D_fake: 0.623 +(epoch: 107, iters: 1952, time: 0.064) G_GAN: 0.294 G_GAN_Feat: 0.795 G_ID: 0.156 G_Rec: 0.351 D_GP: 0.046 D_real: 0.887 D_fake: 0.706 +(epoch: 107, iters: 2352, time: 0.064) G_GAN: 0.566 G_GAN_Feat: 0.970 G_ID: 0.240 G_Rec: 0.444 D_GP: 0.046 D_real: 0.991 D_fake: 0.449 +(epoch: 107, iters: 2752, time: 0.064) G_GAN: 0.311 G_GAN_Feat: 0.666 G_ID: 0.174 G_Rec: 0.331 D_GP: 0.026 D_real: 1.203 D_fake: 0.689 +(epoch: 107, iters: 3152, time: 0.064) G_GAN: 0.502 G_GAN_Feat: 0.938 G_ID: 0.251 G_Rec: 0.373 D_GP: 0.065 D_real: 1.034 D_fake: 0.509 +(epoch: 107, iters: 3552, time: 0.064) G_GAN: 0.276 G_GAN_Feat: 1.182 G_ID: 0.175 G_Rec: 0.364 D_GP: 0.188 D_real: 1.159 D_fake: 0.750 +(epoch: 107, iters: 3952, time: 0.064) G_GAN: 0.393 G_GAN_Feat: 0.694 G_ID: 0.285 G_Rec: 0.406 D_GP: 0.021 D_real: 1.284 D_fake: 0.609 +(epoch: 107, iters: 4352, time: 0.064) G_GAN: 0.258 G_GAN_Feat: 0.667 G_ID: 0.141 G_Rec: 0.343 D_GP: 0.026 D_real: 1.108 D_fake: 0.742 +(epoch: 107, iters: 4752, time: 0.064) G_GAN: 0.505 G_GAN_Feat: 0.882 G_ID: 0.277 G_Rec: 0.474 D_GP: 0.033 D_real: 0.999 D_fake: 0.502 +(epoch: 107, iters: 5152, time: 0.064) G_GAN: 0.166 G_GAN_Feat: 0.658 G_ID: 0.170 G_Rec: 0.349 D_GP: 0.024 D_real: 1.021 D_fake: 0.839 +(epoch: 107, iters: 5552, time: 0.064) G_GAN: 0.449 G_GAN_Feat: 0.959 G_ID: 0.273 G_Rec: 0.393 D_GP: 0.068 D_real: 0.794 D_fake: 0.552 +(epoch: 107, iters: 5952, time: 0.064) G_GAN: 0.455 G_GAN_Feat: 0.814 G_ID: 0.156 G_Rec: 0.346 D_GP: 0.048 D_real: 1.112 D_fake: 0.550 +(epoch: 107, iters: 6352, time: 0.064) G_GAN: 0.652 G_GAN_Feat: 0.750 G_ID: 0.252 G_Rec: 0.361 D_GP: 0.024 D_real: 1.383 D_fake: 0.354 +(epoch: 107, iters: 6752, time: 0.064) G_GAN: 0.159 G_GAN_Feat: 0.964 G_ID: 0.162 G_Rec: 0.330 D_GP: 0.046 D_real: 0.620 D_fake: 0.841 +(epoch: 107, iters: 7152, time: 0.064) G_GAN: 0.805 G_GAN_Feat: 0.828 G_ID: 0.249 G_Rec: 0.409 D_GP: 0.027 D_real: 1.548 D_fake: 0.238 +(epoch: 107, iters: 7552, time: 0.064) G_GAN: 0.124 G_GAN_Feat: 1.030 G_ID: 0.151 G_Rec: 0.384 D_GP: 0.172 D_real: 0.409 D_fake: 0.876 +(epoch: 107, iters: 7952, time: 0.064) G_GAN: 0.245 G_GAN_Feat: 0.907 G_ID: 0.263 G_Rec: 0.398 D_GP: 0.032 D_real: 0.850 D_fake: 0.755 +(epoch: 107, iters: 8352, time: 0.064) G_GAN: 0.199 G_GAN_Feat: 0.686 G_ID: 0.173 G_Rec: 0.335 D_GP: 0.027 D_real: 1.079 D_fake: 0.801 +(epoch: 108, iters: 144, time: 0.064) G_GAN: 0.269 G_GAN_Feat: 1.008 G_ID: 0.299 G_Rec: 0.391 D_GP: 0.037 D_real: 0.693 D_fake: 0.734 +(epoch: 108, iters: 544, time: 0.064) G_GAN: 0.538 G_GAN_Feat: 0.750 G_ID: 0.186 G_Rec: 0.415 D_GP: 0.033 D_real: 1.378 D_fake: 0.475 +(epoch: 108, iters: 944, time: 0.064) G_GAN: 0.214 G_GAN_Feat: 0.925 G_ID: 0.249 G_Rec: 0.420 D_GP: 0.023 D_real: 0.796 D_fake: 0.786 +(epoch: 108, iters: 1344, time: 0.064) G_GAN: 0.363 G_GAN_Feat: 0.915 G_ID: 0.183 G_Rec: 0.350 D_GP: 0.032 D_real: 0.865 D_fake: 0.638 +(epoch: 108, iters: 1744, time: 0.064) G_GAN: 0.548 G_GAN_Feat: 0.902 G_ID: 0.255 G_Rec: 0.384 D_GP: 0.029 D_real: 1.200 D_fake: 0.454 +(epoch: 108, iters: 2144, time: 0.064) G_GAN: 0.592 G_GAN_Feat: 0.814 G_ID: 0.163 G_Rec: 0.457 D_GP: 0.026 D_real: 1.329 D_fake: 0.426 +(epoch: 108, iters: 2544, time: 0.064) G_GAN: 0.593 G_GAN_Feat: 1.065 G_ID: 0.263 G_Rec: 0.382 D_GP: 0.122 D_real: 0.791 D_fake: 0.424 +(epoch: 108, iters: 2944, time: 0.064) G_GAN: 0.270 G_GAN_Feat: 0.710 G_ID: 0.147 G_Rec: 0.377 D_GP: 0.025 D_real: 1.125 D_fake: 0.731 +(epoch: 108, iters: 3344, time: 0.064) G_GAN: 0.381 G_GAN_Feat: 0.963 G_ID: 0.249 G_Rec: 0.397 D_GP: 0.049 D_real: 1.029 D_fake: 0.624 +(epoch: 108, iters: 3744, time: 0.064) G_GAN: 0.230 G_GAN_Feat: 0.776 G_ID: 0.147 G_Rec: 0.348 D_GP: 0.024 D_real: 1.078 D_fake: 0.770 +(epoch: 108, iters: 4144, time: 0.064) G_GAN: 0.465 G_GAN_Feat: 0.884 G_ID: 0.265 G_Rec: 0.423 D_GP: 0.029 D_real: 1.110 D_fake: 0.542 +(epoch: 108, iters: 4544, time: 0.064) G_GAN: 0.166 G_GAN_Feat: 0.797 G_ID: 0.159 G_Rec: 0.387 D_GP: 0.047 D_real: 0.943 D_fake: 0.852 +(epoch: 108, iters: 4944, time: 0.064) G_GAN: 0.169 G_GAN_Feat: 0.872 G_ID: 0.272 G_Rec: 0.422 D_GP: 0.028 D_real: 0.906 D_fake: 0.831 +(epoch: 108, iters: 5344, time: 0.064) G_GAN: 0.212 G_GAN_Feat: 0.869 G_ID: 0.207 G_Rec: 0.368 D_GP: 0.094 D_real: 0.745 D_fake: 0.788 +(epoch: 108, iters: 5744, time: 0.064) G_GAN: 0.386 G_GAN_Feat: 0.857 G_ID: 0.247 G_Rec: 0.370 D_GP: 0.035 D_real: 0.951 D_fake: 0.615 +(epoch: 108, iters: 6144, time: 0.064) G_GAN: 0.371 G_GAN_Feat: 0.863 G_ID: 0.173 G_Rec: 0.398 D_GP: 0.057 D_real: 1.006 D_fake: 0.632 +(epoch: 108, iters: 6544, time: 0.064) G_GAN: 0.336 G_GAN_Feat: 0.956 G_ID: 0.232 G_Rec: 0.476 D_GP: 0.080 D_real: 0.788 D_fake: 0.666 +(epoch: 108, iters: 6944, time: 0.064) G_GAN: 0.320 G_GAN_Feat: 1.266 G_ID: 0.153 G_Rec: 0.640 D_GP: 0.204 D_real: 0.391 D_fake: 0.680 +(epoch: 108, iters: 7344, time: 0.064) G_GAN: 0.495 G_GAN_Feat: 1.201 G_ID: 0.289 G_Rec: 0.440 D_GP: 0.043 D_real: 0.925 D_fake: 0.507 +(epoch: 108, iters: 7744, time: 0.064) G_GAN: 0.194 G_GAN_Feat: 0.946 G_ID: 0.192 G_Rec: 0.365 D_GP: 0.228 D_real: 0.510 D_fake: 0.808 +(epoch: 108, iters: 8144, time: 0.064) G_GAN: 0.662 G_GAN_Feat: 0.806 G_ID: 0.250 G_Rec: 0.420 D_GP: 0.025 D_real: 1.372 D_fake: 0.350 +(epoch: 108, iters: 8544, time: 0.064) G_GAN: 0.306 G_GAN_Feat: 0.820 G_ID: 0.154 G_Rec: 0.377 D_GP: 0.046 D_real: 0.989 D_fake: 0.699 +(epoch: 109, iters: 336, time: 0.064) G_GAN: 0.393 G_GAN_Feat: 0.825 G_ID: 0.256 G_Rec: 0.415 D_GP: 0.023 D_real: 1.121 D_fake: 0.609 +(epoch: 109, iters: 736, time: 0.064) G_GAN: 0.187 G_GAN_Feat: 0.839 G_ID: 0.150 G_Rec: 0.338 D_GP: 0.055 D_real: 0.784 D_fake: 0.813 +(epoch: 109, iters: 1136, time: 0.064) G_GAN: 0.504 G_GAN_Feat: 1.022 G_ID: 0.238 G_Rec: 0.400 D_GP: 0.065 D_real: 0.811 D_fake: 0.499 +(epoch: 109, iters: 1536, time: 0.064) G_GAN: 0.251 G_GAN_Feat: 0.793 G_ID: 0.147 G_Rec: 0.325 D_GP: 0.033 D_real: 1.018 D_fake: 0.749 +(epoch: 109, iters: 1936, time: 0.064) G_GAN: 0.629 G_GAN_Feat: 0.924 G_ID: 0.239 G_Rec: 0.415 D_GP: 0.035 D_real: 1.086 D_fake: 0.379 +(epoch: 109, iters: 2336, time: 0.064) G_GAN: 0.231 G_GAN_Feat: 0.724 G_ID: 0.175 G_Rec: 0.385 D_GP: 0.028 D_real: 1.127 D_fake: 0.769 +(epoch: 109, iters: 2736, time: 0.064) G_GAN: 0.345 G_GAN_Feat: 0.834 G_ID: 0.306 G_Rec: 0.446 D_GP: 0.025 D_real: 1.130 D_fake: 0.656 +(epoch: 109, iters: 3136, time: 0.064) G_GAN: -0.056 G_GAN_Feat: 0.843 G_ID: 0.141 G_Rec: 0.360 D_GP: 0.145 D_real: 0.639 D_fake: 1.056 +(epoch: 109, iters: 3536, time: 0.064) G_GAN: 0.472 G_GAN_Feat: 1.003 G_ID: 0.302 G_Rec: 0.421 D_GP: 0.048 D_real: 1.201 D_fake: 0.531 +(epoch: 109, iters: 3936, time: 0.064) G_GAN: 0.186 G_GAN_Feat: 0.675 G_ID: 0.195 G_Rec: 0.330 D_GP: 0.024 D_real: 1.019 D_fake: 0.814 +(epoch: 109, iters: 4336, time: 0.064) G_GAN: 0.513 G_GAN_Feat: 1.299 G_ID: 0.241 G_Rec: 0.430 D_GP: 0.692 D_real: 0.609 D_fake: 0.495 +(epoch: 109, iters: 4736, time: 0.064) G_GAN: 0.341 G_GAN_Feat: 0.744 G_ID: 0.154 G_Rec: 0.357 D_GP: 0.027 D_real: 1.154 D_fake: 0.659 +(epoch: 109, iters: 5136, time: 0.064) G_GAN: 0.282 G_GAN_Feat: 0.915 G_ID: 0.280 G_Rec: 0.419 D_GP: 0.028 D_real: 0.936 D_fake: 0.719 +(epoch: 109, iters: 5536, time: 0.064) G_GAN: 0.533 G_GAN_Feat: 0.882 G_ID: 0.165 G_Rec: 0.362 D_GP: 0.078 D_real: 0.968 D_fake: 0.483 +(epoch: 109, iters: 5936, time: 0.064) G_GAN: 0.541 G_GAN_Feat: 0.874 G_ID: 0.245 G_Rec: 0.431 D_GP: 0.023 D_real: 1.193 D_fake: 0.466 +(epoch: 109, iters: 6336, time: 0.064) G_GAN: 0.409 G_GAN_Feat: 0.752 G_ID: 0.166 G_Rec: 0.346 D_GP: 0.030 D_real: 1.219 D_fake: 0.593 +(epoch: 109, iters: 6736, time: 0.064) G_GAN: 0.426 G_GAN_Feat: 0.873 G_ID: 0.254 G_Rec: 0.393 D_GP: 0.036 D_real: 1.069 D_fake: 0.578 +(epoch: 109, iters: 7136, time: 0.064) G_GAN: 0.310 G_GAN_Feat: 0.790 G_ID: 0.151 G_Rec: 0.350 D_GP: 0.057 D_real: 0.993 D_fake: 0.693 +(epoch: 109, iters: 7536, time: 0.064) G_GAN: 0.353 G_GAN_Feat: 0.699 G_ID: 0.281 G_Rec: 0.369 D_GP: 0.025 D_real: 1.148 D_fake: 0.648 +(epoch: 109, iters: 7936, time: 0.064) G_GAN: 0.193 G_GAN_Feat: 0.772 G_ID: 0.175 G_Rec: 0.359 D_GP: 0.054 D_real: 0.972 D_fake: 0.808 +(epoch: 109, iters: 8336, time: 0.064) G_GAN: 0.431 G_GAN_Feat: 0.849 G_ID: 0.262 G_Rec: 0.375 D_GP: 0.029 D_real: 1.177 D_fake: 0.569 +(epoch: 110, iters: 128, time: 0.064) G_GAN: 0.370 G_GAN_Feat: 0.761 G_ID: 0.177 G_Rec: 0.332 D_GP: 0.039 D_real: 1.112 D_fake: 0.630 +(epoch: 110, iters: 528, time: 0.064) G_GAN: 0.220 G_GAN_Feat: 0.847 G_ID: 0.259 G_Rec: 0.420 D_GP: 0.038 D_real: 0.945 D_fake: 0.781 +(epoch: 110, iters: 928, time: 0.064) G_GAN: 0.403 G_GAN_Feat: 0.899 G_ID: 0.145 G_Rec: 0.360 D_GP: 0.128 D_real: 0.826 D_fake: 0.600 +(epoch: 110, iters: 1328, time: 0.064) G_GAN: 0.461 G_GAN_Feat: 0.842 G_ID: 0.272 G_Rec: 0.407 D_GP: 0.028 D_real: 1.231 D_fake: 0.545 +(epoch: 110, iters: 1728, time: 0.064) G_GAN: 0.198 G_GAN_Feat: 0.698 G_ID: 0.189 G_Rec: 0.319 D_GP: 0.033 D_real: 1.023 D_fake: 0.802 +(epoch: 110, iters: 2128, time: 0.064) G_GAN: 0.425 G_GAN_Feat: 0.927 G_ID: 0.210 G_Rec: 0.408 D_GP: 0.037 D_real: 0.940 D_fake: 0.576 +(epoch: 110, iters: 2528, time: 0.064) G_GAN: 0.492 G_GAN_Feat: 0.696 G_ID: 0.165 G_Rec: 0.341 D_GP: 0.024 D_real: 1.342 D_fake: 0.510 +(epoch: 110, iters: 2928, time: 0.064) G_GAN: 0.508 G_GAN_Feat: 0.889 G_ID: 0.254 G_Rec: 0.388 D_GP: 0.036 D_real: 1.164 D_fake: 0.494 +(epoch: 110, iters: 3328, time: 0.064) G_GAN: 0.168 G_GAN_Feat: 0.863 G_ID: 0.175 G_Rec: 0.366 D_GP: 0.053 D_real: 0.717 D_fake: 0.842 +(epoch: 110, iters: 3728, time: 0.064) G_GAN: 0.371 G_GAN_Feat: 0.912 G_ID: 0.302 G_Rec: 0.401 D_GP: 0.056 D_real: 1.032 D_fake: 0.634 +(epoch: 110, iters: 4128, time: 0.064) G_GAN: 0.446 G_GAN_Feat: 0.888 G_ID: 0.172 G_Rec: 0.345 D_GP: 0.051 D_real: 1.055 D_fake: 0.555 +(epoch: 110, iters: 4528, time: 0.064) G_GAN: 0.530 G_GAN_Feat: 0.929 G_ID: 0.247 G_Rec: 0.377 D_GP: 0.029 D_real: 1.207 D_fake: 0.472 +(epoch: 110, iters: 4928, time: 0.064) G_GAN: 0.459 G_GAN_Feat: 0.655 G_ID: 0.196 G_Rec: 0.309 D_GP: 0.026 D_real: 1.348 D_fake: 0.541 +(epoch: 110, iters: 5328, time: 0.064) G_GAN: 0.332 G_GAN_Feat: 0.948 G_ID: 0.250 G_Rec: 0.390 D_GP: 0.046 D_real: 1.030 D_fake: 0.686 +(epoch: 110, iters: 5728, time: 0.064) G_GAN: 0.403 G_GAN_Feat: 0.765 G_ID: 0.152 G_Rec: 0.367 D_GP: 0.024 D_real: 1.237 D_fake: 0.598 +(epoch: 110, iters: 6128, time: 0.064) G_GAN: 0.505 G_GAN_Feat: 0.832 G_ID: 0.229 G_Rec: 0.385 D_GP: 0.029 D_real: 1.241 D_fake: 0.495 +(epoch: 110, iters: 6528, time: 0.064) G_GAN: -0.085 G_GAN_Feat: 0.932 G_ID: 0.177 G_Rec: 0.355 D_GP: 0.073 D_real: 0.992 D_fake: 1.085 +(epoch: 110, iters: 6928, time: 0.064) G_GAN: 0.864 G_GAN_Feat: 0.975 G_ID: 0.265 G_Rec: 0.410 D_GP: 0.046 D_real: 1.439 D_fake: 0.183 +(epoch: 110, iters: 7328, time: 0.064) G_GAN: 0.561 G_GAN_Feat: 0.816 G_ID: 0.177 G_Rec: 0.362 D_GP: 0.042 D_real: 1.295 D_fake: 0.494 +(epoch: 110, iters: 7728, time: 0.064) G_GAN: 0.593 G_GAN_Feat: 0.765 G_ID: 0.264 G_Rec: 0.377 D_GP: 0.023 D_real: 1.377 D_fake: 0.434 +(epoch: 110, iters: 8128, time: 0.064) G_GAN: 0.254 G_GAN_Feat: 0.895 G_ID: 0.147 G_Rec: 0.342 D_GP: 0.056 D_real: 0.750 D_fake: 0.746 +(epoch: 110, iters: 8528, time: 0.064) G_GAN: 0.787 G_GAN_Feat: 1.238 G_ID: 0.258 G_Rec: 0.464 D_GP: 0.154 D_real: 0.614 D_fake: 0.319 +(epoch: 111, iters: 320, time: 0.064) G_GAN: 0.364 G_GAN_Feat: 0.786 G_ID: 0.165 G_Rec: 0.402 D_GP: 0.030 D_real: 1.214 D_fake: 0.636 +(epoch: 111, iters: 720, time: 0.064) G_GAN: 0.410 G_GAN_Feat: 0.870 G_ID: 0.263 G_Rec: 0.371 D_GP: 0.039 D_real: 1.075 D_fake: 0.592 +(epoch: 111, iters: 1120, time: 0.064) G_GAN: -0.811 G_GAN_Feat: 1.135 G_ID: 0.174 G_Rec: 0.456 D_GP: 0.940 D_real: 0.077 D_fake: 1.811 +(epoch: 111, iters: 1520, time: 0.064) G_GAN: 0.361 G_GAN_Feat: 0.738 G_ID: 0.247 G_Rec: 0.368 D_GP: 0.022 D_real: 1.113 D_fake: 0.640 +(epoch: 111, iters: 1920, time: 0.064) G_GAN: 0.486 G_GAN_Feat: 0.620 G_ID: 0.173 G_Rec: 0.310 D_GP: 0.027 D_real: 1.412 D_fake: 0.521 +(epoch: 111, iters: 2320, time: 0.064) G_GAN: 0.467 G_GAN_Feat: 0.814 G_ID: 0.269 G_Rec: 0.396 D_GP: 0.054 D_real: 1.034 D_fake: 0.549 +(epoch: 111, iters: 2720, time: 0.064) G_GAN: 0.119 G_GAN_Feat: 0.889 G_ID: 0.184 G_Rec: 0.424 D_GP: 0.034 D_real: 0.662 D_fake: 0.882 +(epoch: 111, iters: 3120, time: 0.064) G_GAN: 0.483 G_GAN_Feat: 0.789 G_ID: 0.227 G_Rec: 0.395 D_GP: 0.026 D_real: 1.232 D_fake: 0.518 +(epoch: 111, iters: 3520, time: 0.064) G_GAN: 0.233 G_GAN_Feat: 0.956 G_ID: 0.170 G_Rec: 0.391 D_GP: 0.072 D_real: 0.717 D_fake: 0.768 +(epoch: 111, iters: 3920, time: 0.064) G_GAN: 0.633 G_GAN_Feat: 1.234 G_ID: 0.255 G_Rec: 0.466 D_GP: 0.141 D_real: 0.863 D_fake: 0.452 +(epoch: 111, iters: 4320, time: 0.064) G_GAN: 0.501 G_GAN_Feat: 0.871 G_ID: 0.154 G_Rec: 0.337 D_GP: 0.033 D_real: 1.393 D_fake: 0.540 +(epoch: 111, iters: 4720, time: 0.064) G_GAN: 0.013 G_GAN_Feat: 0.745 G_ID: 0.269 G_Rec: 0.385 D_GP: 0.023 D_real: 0.853 D_fake: 0.987 +(epoch: 111, iters: 5120, time: 0.064) G_GAN: 0.154 G_GAN_Feat: 0.697 G_ID: 0.154 G_Rec: 0.357 D_GP: 0.027 D_real: 0.978 D_fake: 0.846 +(epoch: 111, iters: 5520, time: 0.064) G_GAN: 0.353 G_GAN_Feat: 0.829 G_ID: 0.257 G_Rec: 0.397 D_GP: 0.031 D_real: 0.958 D_fake: 0.647 +(epoch: 111, iters: 5920, time: 0.064) G_GAN: 0.259 G_GAN_Feat: 0.714 G_ID: 0.151 G_Rec: 0.309 D_GP: 0.026 D_real: 1.184 D_fake: 0.742 +(epoch: 111, iters: 6320, time: 0.064) G_GAN: 0.375 G_GAN_Feat: 0.881 G_ID: 0.250 G_Rec: 0.430 D_GP: 0.061 D_real: 1.010 D_fake: 0.628 +(epoch: 111, iters: 6720, time: 0.064) G_GAN: 0.486 G_GAN_Feat: 0.844 G_ID: 0.152 G_Rec: 0.372 D_GP: 0.074 D_real: 1.107 D_fake: 0.521 +(epoch: 111, iters: 7120, time: 0.064) G_GAN: 0.262 G_GAN_Feat: 0.852 G_ID: 0.252 G_Rec: 0.386 D_GP: 0.027 D_real: 1.010 D_fake: 0.739 +(epoch: 111, iters: 7520, time: 0.064) G_GAN: 0.735 G_GAN_Feat: 0.796 G_ID: 0.170 G_Rec: 0.341 D_GP: 0.037 D_real: 1.465 D_fake: 0.303 +(epoch: 111, iters: 7920, time: 0.064) G_GAN: 0.409 G_GAN_Feat: 0.834 G_ID: 0.269 G_Rec: 0.418 D_GP: 0.028 D_real: 1.163 D_fake: 0.594 +(epoch: 111, iters: 8320, time: 0.064) G_GAN: 0.206 G_GAN_Feat: 0.849 G_ID: 0.178 G_Rec: 0.396 D_GP: 0.059 D_real: 0.796 D_fake: 0.794 +(epoch: 112, iters: 112, time: 0.064) G_GAN: 0.611 G_GAN_Feat: 0.906 G_ID: 0.222 G_Rec: 0.427 D_GP: 0.043 D_real: 1.210 D_fake: 0.405 +(epoch: 112, iters: 512, time: 0.064) G_GAN: 0.319 G_GAN_Feat: 0.813 G_ID: 0.154 G_Rec: 0.339 D_GP: 0.033 D_real: 0.982 D_fake: 0.681 +(epoch: 112, iters: 912, time: 0.064) G_GAN: 0.232 G_GAN_Feat: 0.713 G_ID: 0.277 G_Rec: 0.390 D_GP: 0.023 D_real: 1.108 D_fake: 0.768 +(epoch: 112, iters: 1312, time: 0.064) G_GAN: 0.105 G_GAN_Feat: 0.815 G_ID: 0.179 G_Rec: 0.381 D_GP: 0.062 D_real: 0.844 D_fake: 0.895 +(epoch: 112, iters: 1712, time: 0.064) G_GAN: 0.675 G_GAN_Feat: 0.824 G_ID: 0.280 G_Rec: 0.387 D_GP: 0.031 D_real: 1.358 D_fake: 0.349 +(epoch: 112, iters: 2112, time: 0.064) G_GAN: 0.300 G_GAN_Feat: 0.795 G_ID: 0.178 G_Rec: 0.349 D_GP: 0.039 D_real: 1.030 D_fake: 0.701 +(epoch: 112, iters: 2512, time: 0.064) G_GAN: 0.328 G_GAN_Feat: 0.949 G_ID: 0.273 G_Rec: 0.404 D_GP: 0.033 D_real: 0.839 D_fake: 0.672 +(epoch: 112, iters: 2912, time: 0.064) G_GAN: 0.122 G_GAN_Feat: 1.091 G_ID: 0.220 G_Rec: 0.356 D_GP: 0.481 D_real: 0.246 D_fake: 0.902 +(epoch: 112, iters: 3312, time: 0.064) G_GAN: 0.381 G_GAN_Feat: 1.155 G_ID: 0.232 G_Rec: 0.394 D_GP: 0.086 D_real: 0.450 D_fake: 0.620 +(epoch: 112, iters: 3712, time: 0.064) G_GAN: 0.455 G_GAN_Feat: 0.762 G_ID: 0.153 G_Rec: 0.342 D_GP: 0.027 D_real: 1.257 D_fake: 0.548 +(epoch: 112, iters: 4112, time: 0.064) G_GAN: 0.341 G_GAN_Feat: 0.740 G_ID: 0.241 G_Rec: 0.406 D_GP: 0.026 D_real: 1.170 D_fake: 0.660 +(epoch: 112, iters: 4512, time: 0.064) G_GAN: 0.406 G_GAN_Feat: 0.751 G_ID: 0.156 G_Rec: 0.348 D_GP: 0.031 D_real: 1.138 D_fake: 0.594 +(epoch: 112, iters: 4912, time: 0.064) G_GAN: 0.347 G_GAN_Feat: 0.885 G_ID: 0.270 G_Rec: 0.388 D_GP: 0.027 D_real: 1.025 D_fake: 0.653 +(epoch: 112, iters: 5312, time: 0.064) G_GAN: 0.194 G_GAN_Feat: 0.674 G_ID: 0.183 G_Rec: 0.381 D_GP: 0.026 D_real: 1.066 D_fake: 0.807 +(epoch: 112, iters: 5712, time: 0.064) G_GAN: 0.265 G_GAN_Feat: 0.844 G_ID: 0.262 G_Rec: 0.386 D_GP: 0.043 D_real: 0.924 D_fake: 0.735 +(epoch: 112, iters: 6112, time: 0.064) G_GAN: 0.460 G_GAN_Feat: 0.777 G_ID: 0.157 G_Rec: 0.346 D_GP: 0.033 D_real: 1.286 D_fake: 0.545 +(epoch: 112, iters: 6512, time: 0.064) G_GAN: 0.457 G_GAN_Feat: 0.976 G_ID: 0.258 G_Rec: 0.449 D_GP: 0.031 D_real: 0.991 D_fake: 0.543 +(epoch: 112, iters: 6912, time: 0.064) G_GAN: 0.222 G_GAN_Feat: 0.886 G_ID: 0.157 G_Rec: 0.367 D_GP: 0.083 D_real: 0.728 D_fake: 0.779 +(epoch: 112, iters: 7312, time: 0.064) G_GAN: 0.324 G_GAN_Feat: 0.834 G_ID: 0.240 G_Rec: 0.377 D_GP: 0.031 D_real: 1.090 D_fake: 0.678 +(epoch: 112, iters: 7712, time: 0.064) G_GAN: 0.204 G_GAN_Feat: 0.710 G_ID: 0.198 G_Rec: 0.329 D_GP: 0.030 D_real: 1.054 D_fake: 0.796 +(epoch: 112, iters: 8112, time: 0.064) G_GAN: 0.959 G_GAN_Feat: 0.929 G_ID: 0.253 G_Rec: 0.389 D_GP: 0.050 D_real: 1.422 D_fake: 0.139 +(epoch: 112, iters: 8512, time: 0.064) G_GAN: 0.419 G_GAN_Feat: 0.719 G_ID: 0.148 G_Rec: 0.341 D_GP: 0.027 D_real: 1.309 D_fake: 0.586 +(epoch: 113, iters: 304, time: 0.064) G_GAN: 0.398 G_GAN_Feat: 0.948 G_ID: 0.243 G_Rec: 0.409 D_GP: 0.046 D_real: 0.874 D_fake: 0.602 +(epoch: 113, iters: 704, time: 0.064) G_GAN: 0.266 G_GAN_Feat: 0.766 G_ID: 0.194 G_Rec: 0.359 D_GP: 0.035 D_real: 1.128 D_fake: 0.734 +(epoch: 113, iters: 1104, time: 0.064) G_GAN: 0.560 G_GAN_Feat: 0.942 G_ID: 0.255 G_Rec: 0.369 D_GP: 0.059 D_real: 1.017 D_fake: 0.498 +(epoch: 113, iters: 1504, time: 0.064) G_GAN: 0.286 G_GAN_Feat: 0.757 G_ID: 0.163 G_Rec: 0.330 D_GP: 0.036 D_real: 1.142 D_fake: 0.715 +(epoch: 113, iters: 1904, time: 0.064) G_GAN: 0.536 G_GAN_Feat: 0.842 G_ID: 0.273 G_Rec: 0.437 D_GP: 0.023 D_real: 1.310 D_fake: 0.467 +(epoch: 113, iters: 2304, time: 0.064) G_GAN: 0.275 G_GAN_Feat: 0.902 G_ID: 0.171 G_Rec: 0.369 D_GP: 0.092 D_real: 0.614 D_fake: 0.725 +(epoch: 113, iters: 2704, time: 0.064) G_GAN: 0.526 G_GAN_Feat: 1.052 G_ID: 0.271 G_Rec: 0.455 D_GP: 0.078 D_real: 0.960 D_fake: 0.520 +(epoch: 113, iters: 3104, time: 0.064) G_GAN: 0.321 G_GAN_Feat: 0.982 G_ID: 0.164 G_Rec: 0.351 D_GP: 0.040 D_real: 0.803 D_fake: 0.679 +(epoch: 113, iters: 3504, time: 0.064) G_GAN: 0.326 G_GAN_Feat: 0.869 G_ID: 0.221 G_Rec: 0.364 D_GP: 0.028 D_real: 0.977 D_fake: 0.674 +(epoch: 113, iters: 3904, time: 0.064) G_GAN: 0.438 G_GAN_Feat: 0.760 G_ID: 0.167 G_Rec: 0.354 D_GP: 0.027 D_real: 1.275 D_fake: 0.603 +(epoch: 113, iters: 4304, time: 0.064) G_GAN: 0.322 G_GAN_Feat: 0.898 G_ID: 0.242 G_Rec: 0.401 D_GP: 0.031 D_real: 0.938 D_fake: 0.678 +(epoch: 113, iters: 4704, time: 0.064) G_GAN: 0.280 G_GAN_Feat: 0.720 G_ID: 0.140 G_Rec: 0.369 D_GP: 0.022 D_real: 1.104 D_fake: 0.720 +(epoch: 113, iters: 5104, time: 0.064) G_GAN: 0.474 G_GAN_Feat: 0.851 G_ID: 0.271 G_Rec: 0.406 D_GP: 0.032 D_real: 1.184 D_fake: 0.529 +(epoch: 113, iters: 5504, time: 0.064) G_GAN: 0.332 G_GAN_Feat: 0.928 G_ID: 0.167 G_Rec: 0.399 D_GP: 0.067 D_real: 0.853 D_fake: 0.691 +(epoch: 113, iters: 5904, time: 0.064) G_GAN: 0.426 G_GAN_Feat: 0.823 G_ID: 0.277 G_Rec: 0.407 D_GP: 0.021 D_real: 1.084 D_fake: 0.576 +(epoch: 113, iters: 6304, time: 0.064) G_GAN: 0.286 G_GAN_Feat: 0.797 G_ID: 0.160 G_Rec: 0.362 D_GP: 0.022 D_real: 1.103 D_fake: 0.714 +(epoch: 113, iters: 6704, time: 0.064) G_GAN: 0.406 G_GAN_Feat: 0.885 G_ID: 0.231 G_Rec: 0.386 D_GP: 0.030 D_real: 1.138 D_fake: 0.598 +(epoch: 113, iters: 7104, time: 0.064) G_GAN: -0.012 G_GAN_Feat: 0.814 G_ID: 0.164 G_Rec: 0.352 D_GP: 0.053 D_real: 0.757 D_fake: 1.012 +(epoch: 113, iters: 7504, time: 0.064) G_GAN: 0.337 G_GAN_Feat: 0.783 G_ID: 0.284 G_Rec: 0.392 D_GP: 0.026 D_real: 1.135 D_fake: 0.664 +(epoch: 113, iters: 7904, time: 0.064) G_GAN: 0.257 G_GAN_Feat: 0.744 G_ID: 0.218 G_Rec: 0.379 D_GP: 0.032 D_real: 1.068 D_fake: 0.743 +(epoch: 113, iters: 8304, time: 0.064) G_GAN: 0.417 G_GAN_Feat: 1.091 G_ID: 0.239 G_Rec: 0.500 D_GP: 0.090 D_real: 0.917 D_fake: 0.618 +(epoch: 114, iters: 96, time: 0.064) G_GAN: 0.280 G_GAN_Feat: 0.811 G_ID: 0.159 G_Rec: 0.401 D_GP: 0.038 D_real: 1.014 D_fake: 0.720 +(epoch: 114, iters: 496, time: 0.064) G_GAN: 0.621 G_GAN_Feat: 1.083 G_ID: 0.267 G_Rec: 0.424 D_GP: 0.134 D_real: 0.898 D_fake: 0.420 +(epoch: 114, iters: 896, time: 0.064) G_GAN: 0.521 G_GAN_Feat: 1.020 G_ID: 0.177 G_Rec: 0.423 D_GP: 0.290 D_real: 0.845 D_fake: 0.529 +(epoch: 114, iters: 1296, time: 0.064) G_GAN: 0.690 G_GAN_Feat: 0.862 G_ID: 0.240 G_Rec: 0.383 D_GP: 0.032 D_real: 1.412 D_fake: 0.333 +(epoch: 114, iters: 1696, time: 0.064) G_GAN: 0.138 G_GAN_Feat: 1.013 G_ID: 0.164 G_Rec: 0.365 D_GP: 0.122 D_real: 0.470 D_fake: 0.862 +(epoch: 114, iters: 2096, time: 0.064) G_GAN: 0.677 G_GAN_Feat: 0.745 G_ID: 0.272 G_Rec: 0.371 D_GP: 0.022 D_real: 1.484 D_fake: 0.346 +(epoch: 114, iters: 2496, time: 0.064) G_GAN: 0.367 G_GAN_Feat: 0.730 G_ID: 0.166 G_Rec: 0.389 D_GP: 0.031 D_real: 1.136 D_fake: 0.633 +(epoch: 114, iters: 2896, time: 0.064) G_GAN: 0.521 G_GAN_Feat: 0.905 G_ID: 0.241 G_Rec: 0.380 D_GP: 0.060 D_real: 1.170 D_fake: 0.527 +(epoch: 114, iters: 3296, time: 0.064) G_GAN: 0.392 G_GAN_Feat: 0.734 G_ID: 0.172 G_Rec: 0.357 D_GP: 0.030 D_real: 1.215 D_fake: 0.610 +(epoch: 114, iters: 3696, time: 0.064) G_GAN: 0.518 G_GAN_Feat: 0.867 G_ID: 0.223 G_Rec: 0.403 D_GP: 0.050 D_real: 1.165 D_fake: 0.493 +(epoch: 114, iters: 4096, time: 0.064) G_GAN: 0.350 G_GAN_Feat: 0.754 G_ID: 0.132 G_Rec: 0.374 D_GP: 0.028 D_real: 1.192 D_fake: 0.650 +(epoch: 114, iters: 4496, time: 0.064) G_GAN: 0.310 G_GAN_Feat: 0.880 G_ID: 0.278 G_Rec: 0.389 D_GP: 0.043 D_real: 0.980 D_fake: 0.691 +(epoch: 114, iters: 4896, time: 0.064) G_GAN: 0.219 G_GAN_Feat: 0.921 G_ID: 0.177 G_Rec: 0.381 D_GP: 0.046 D_real: 0.772 D_fake: 0.781 +(epoch: 114, iters: 5296, time: 0.064) G_GAN: 0.225 G_GAN_Feat: 0.931 G_ID: 0.221 G_Rec: 0.390 D_GP: 0.060 D_real: 0.643 D_fake: 0.776 +(epoch: 114, iters: 5696, time: 0.064) G_GAN: 0.125 G_GAN_Feat: 1.089 G_ID: 0.184 G_Rec: 0.396 D_GP: 0.518 D_real: 0.335 D_fake: 0.876 +(epoch: 114, iters: 6096, time: 0.064) G_GAN: 0.072 G_GAN_Feat: 1.178 G_ID: 0.209 G_Rec: 0.423 D_GP: 0.137 D_real: 0.415 D_fake: 0.928 +(epoch: 114, iters: 6496, time: 0.064) G_GAN: 0.286 G_GAN_Feat: 0.844 G_ID: 0.134 G_Rec: 0.356 D_GP: 0.068 D_real: 0.833 D_fake: 0.715 +(epoch: 114, iters: 6896, time: 0.064) G_GAN: 0.606 G_GAN_Feat: 0.947 G_ID: 0.299 G_Rec: 0.434 D_GP: 0.030 D_real: 1.152 D_fake: 0.405 +(epoch: 114, iters: 7296, time: 0.064) G_GAN: 0.296 G_GAN_Feat: 0.900 G_ID: 0.155 G_Rec: 0.424 D_GP: 0.048 D_real: 0.735 D_fake: 0.705 +(epoch: 114, iters: 7696, time: 0.064) G_GAN: 0.273 G_GAN_Feat: 0.835 G_ID: 0.245 G_Rec: 0.366 D_GP: 0.041 D_real: 1.078 D_fake: 0.729 +(epoch: 114, iters: 8096, time: 0.064) G_GAN: 0.324 G_GAN_Feat: 0.759 G_ID: 0.188 G_Rec: 0.372 D_GP: 0.026 D_real: 1.168 D_fake: 0.677 +(epoch: 114, iters: 8496, time: 0.064) G_GAN: 0.441 G_GAN_Feat: 1.087 G_ID: 0.237 G_Rec: 0.396 D_GP: 0.072 D_real: 0.570 D_fake: 0.564 +(epoch: 115, iters: 288, time: 0.064) G_GAN: 0.221 G_GAN_Feat: 0.814 G_ID: 0.159 G_Rec: 0.342 D_GP: 0.037 D_real: 0.921 D_fake: 0.780 +(epoch: 115, iters: 688, time: 0.064) G_GAN: 0.476 G_GAN_Feat: 0.873 G_ID: 0.272 G_Rec: 0.372 D_GP: 0.037 D_real: 1.181 D_fake: 0.531 +(epoch: 115, iters: 1088, time: 0.064) G_GAN: 0.059 G_GAN_Feat: 0.800 G_ID: 0.153 G_Rec: 0.418 D_GP: 0.028 D_real: 0.870 D_fake: 0.941 +(epoch: 115, iters: 1488, time: 0.064) G_GAN: 0.408 G_GAN_Feat: 0.974 G_ID: 0.264 G_Rec: 0.413 D_GP: 0.077 D_real: 0.861 D_fake: 0.595 +(epoch: 115, iters: 1888, time: 0.064) G_GAN: 0.381 G_GAN_Feat: 1.015 G_ID: 0.148 G_Rec: 0.406 D_GP: 0.029 D_real: 0.855 D_fake: 0.619 +(epoch: 115, iters: 2288, time: 0.064) G_GAN: 0.508 G_GAN_Feat: 0.759 G_ID: 0.252 G_Rec: 0.374 D_GP: 0.025 D_real: 1.268 D_fake: 0.495 +(epoch: 115, iters: 2688, time: 0.064) G_GAN: 0.663 G_GAN_Feat: 0.903 G_ID: 0.162 G_Rec: 0.374 D_GP: 0.121 D_real: 1.110 D_fake: 0.424 +(epoch: 115, iters: 3088, time: 0.064) G_GAN: 0.734 G_GAN_Feat: 0.917 G_ID: 0.252 G_Rec: 0.397 D_GP: 0.028 D_real: 1.366 D_fake: 0.339 +(epoch: 115, iters: 3488, time: 0.064) G_GAN: 0.075 G_GAN_Feat: 0.678 G_ID: 0.144 G_Rec: 0.428 D_GP: 0.026 D_real: 0.940 D_fake: 0.925 +(epoch: 115, iters: 3888, time: 0.064) G_GAN: 0.614 G_GAN_Feat: 0.917 G_ID: 0.241 G_Rec: 0.392 D_GP: 0.029 D_real: 1.264 D_fake: 0.392 +(epoch: 115, iters: 4288, time: 0.064) G_GAN: 0.263 G_GAN_Feat: 0.715 G_ID: 0.188 G_Rec: 0.322 D_GP: 0.027 D_real: 1.046 D_fake: 0.737 +(epoch: 115, iters: 4688, time: 0.064) G_GAN: 0.266 G_GAN_Feat: 0.891 G_ID: 0.245 G_Rec: 0.415 D_GP: 0.030 D_real: 0.920 D_fake: 0.737 +(epoch: 115, iters: 5088, time: 0.064) G_GAN: 0.296 G_GAN_Feat: 0.806 G_ID: 0.171 G_Rec: 0.361 D_GP: 0.042 D_real: 1.054 D_fake: 0.705 +(epoch: 115, iters: 5488, time: 0.064) G_GAN: 0.169 G_GAN_Feat: 1.214 G_ID: 0.241 G_Rec: 0.453 D_GP: 0.114 D_real: 0.346 D_fake: 0.833 +(epoch: 115, iters: 5888, time: 0.064) G_GAN: -0.023 G_GAN_Feat: 0.785 G_ID: 0.189 G_Rec: 0.525 D_GP: 0.033 D_real: 0.822 D_fake: 1.023 +(epoch: 115, iters: 6288, time: 0.064) G_GAN: 0.646 G_GAN_Feat: 1.036 G_ID: 0.249 G_Rec: 0.448 D_GP: 0.061 D_real: 0.941 D_fake: 0.364 +(epoch: 115, iters: 6688, time: 0.064) G_GAN: 0.029 G_GAN_Feat: 1.008 G_ID: 0.169 G_Rec: 0.343 D_GP: 0.049 D_real: 0.378 D_fake: 0.971 +(epoch: 115, iters: 7088, time: 0.064) G_GAN: 0.391 G_GAN_Feat: 0.936 G_ID: 0.234 G_Rec: 0.429 D_GP: 0.048 D_real: 1.056 D_fake: 0.615 +(epoch: 115, iters: 7488, time: 0.064) G_GAN: 0.645 G_GAN_Feat: 0.900 G_ID: 0.170 G_Rec: 0.361 D_GP: 0.056 D_real: 1.142 D_fake: 0.377 +(epoch: 115, iters: 7888, time: 0.064) G_GAN: 0.603 G_GAN_Feat: 0.970 G_ID: 0.244 G_Rec: 0.381 D_GP: 0.052 D_real: 0.996 D_fake: 0.440 +(epoch: 115, iters: 8288, time: 0.064) G_GAN: 0.400 G_GAN_Feat: 0.581 G_ID: 0.151 G_Rec: 0.333 D_GP: 0.017 D_real: 1.387 D_fake: 0.601 +(epoch: 116, iters: 80, time: 0.064) G_GAN: 0.276 G_GAN_Feat: 0.681 G_ID: 0.235 G_Rec: 0.402 D_GP: 0.018 D_real: 1.158 D_fake: 0.724 +(epoch: 116, iters: 480, time: 0.064) G_GAN: 0.376 G_GAN_Feat: 0.619 G_ID: 0.141 G_Rec: 0.369 D_GP: 0.019 D_real: 1.328 D_fake: 0.626 +(epoch: 116, iters: 880, time: 0.064) G_GAN: 0.259 G_GAN_Feat: 0.723 G_ID: 0.271 G_Rec: 0.386 D_GP: 0.025 D_real: 1.036 D_fake: 0.741 +(epoch: 116, iters: 1280, time: 0.064) G_GAN: 0.069 G_GAN_Feat: 0.608 G_ID: 0.215 G_Rec: 0.350 D_GP: 0.024 D_real: 0.912 D_fake: 0.931 +(epoch: 116, iters: 1680, time: 0.064) G_GAN: 0.312 G_GAN_Feat: 0.809 G_ID: 0.241 G_Rec: 0.406 D_GP: 0.044 D_real: 0.893 D_fake: 0.690 +(epoch: 116, iters: 2080, time: 0.064) G_GAN: 0.060 G_GAN_Feat: 0.611 G_ID: 0.165 G_Rec: 0.336 D_GP: 0.033 D_real: 0.983 D_fake: 0.940 +(epoch: 116, iters: 2480, time: 0.064) G_GAN: 0.068 G_GAN_Feat: 0.697 G_ID: 0.264 G_Rec: 0.373 D_GP: 0.034 D_real: 0.810 D_fake: 0.932 +(epoch: 116, iters: 2880, time: 0.064) G_GAN: -0.023 G_GAN_Feat: 0.690 G_ID: 0.175 G_Rec: 0.351 D_GP: 0.045 D_real: 0.776 D_fake: 1.023 +(epoch: 116, iters: 3280, time: 0.064) G_GAN: 0.010 G_GAN_Feat: 0.792 G_ID: 0.239 G_Rec: 0.412 D_GP: 0.039 D_real: 0.787 D_fake: 0.990 +(epoch: 116, iters: 3680, time: 0.064) G_GAN: 0.018 G_GAN_Feat: 0.804 G_ID: 0.207 G_Rec: 0.359 D_GP: 0.102 D_real: 0.575 D_fake: 0.985 +(epoch: 116, iters: 4080, time: 0.064) G_GAN: 0.538 G_GAN_Feat: 0.836 G_ID: 0.281 G_Rec: 0.394 D_GP: 0.032 D_real: 1.249 D_fake: 0.467 +(epoch: 116, iters: 4480, time: 0.064) G_GAN: 0.256 G_GAN_Feat: 0.811 G_ID: 0.179 G_Rec: 0.363 D_GP: 0.057 D_real: 1.028 D_fake: 0.767 +(epoch: 116, iters: 4880, time: 0.064) G_GAN: 0.369 G_GAN_Feat: 1.089 G_ID: 0.231 G_Rec: 0.429 D_GP: 0.092 D_real: 0.643 D_fake: 0.634 +(epoch: 116, iters: 5280, time: 0.064) G_GAN: 0.274 G_GAN_Feat: 0.863 G_ID: 0.152 G_Rec: 0.413 D_GP: 0.098 D_real: 0.988 D_fake: 0.736 +(epoch: 116, iters: 5680, time: 0.064) G_GAN: 0.260 G_GAN_Feat: 1.049 G_ID: 0.266 G_Rec: 0.409 D_GP: 0.074 D_real: 0.634 D_fake: 0.741 +(epoch: 116, iters: 6080, time: 0.064) G_GAN: 0.541 G_GAN_Feat: 0.838 G_ID: 0.180 G_Rec: 0.365 D_GP: 0.094 D_real: 1.119 D_fake: 0.466 +(epoch: 116, iters: 6480, time: 0.064) G_GAN: 0.247 G_GAN_Feat: 0.748 G_ID: 0.241 G_Rec: 0.383 D_GP: 0.023 D_real: 0.959 D_fake: 0.754 +(epoch: 116, iters: 6880, time: 0.064) G_GAN: 0.077 G_GAN_Feat: 0.695 G_ID: 0.153 G_Rec: 0.331 D_GP: 0.035 D_real: 0.897 D_fake: 0.923 +(epoch: 116, iters: 7280, time: 0.064) G_GAN: 0.676 G_GAN_Feat: 0.862 G_ID: 0.252 G_Rec: 0.411 D_GP: 0.029 D_real: 1.423 D_fake: 0.345 +(epoch: 116, iters: 7680, time: 0.064) G_GAN: 0.249 G_GAN_Feat: 0.713 G_ID: 0.168 G_Rec: 0.467 D_GP: 0.028 D_real: 1.149 D_fake: 0.751 +(epoch: 116, iters: 8080, time: 0.064) G_GAN: 0.464 G_GAN_Feat: 1.079 G_ID: 0.261 G_Rec: 0.420 D_GP: 0.152 D_real: 0.575 D_fake: 0.550 +(epoch: 116, iters: 8480, time: 0.064) G_GAN: 0.338 G_GAN_Feat: 0.895 G_ID: 0.132 G_Rec: 0.339 D_GP: 0.100 D_real: 0.743 D_fake: 0.662 +(epoch: 117, iters: 272, time: 0.064) G_GAN: 0.276 G_GAN_Feat: 0.735 G_ID: 0.259 G_Rec: 0.421 D_GP: 0.022 D_real: 1.059 D_fake: 0.725 +(epoch: 117, iters: 672, time: 0.064) G_GAN: 0.128 G_GAN_Feat: 0.644 G_ID: 0.184 G_Rec: 0.330 D_GP: 0.026 D_real: 1.047 D_fake: 0.872 +(epoch: 117, iters: 1072, time: 0.064) G_GAN: 0.446 G_GAN_Feat: 0.805 G_ID: 0.280 G_Rec: 0.400 D_GP: 0.045 D_real: 1.187 D_fake: 0.556 +(epoch: 117, iters: 1472, time: 0.064) G_GAN: 0.442 G_GAN_Feat: 0.721 G_ID: 0.160 G_Rec: 0.366 D_GP: 0.046 D_real: 1.229 D_fake: 0.560 +(epoch: 117, iters: 1872, time: 0.064) G_GAN: 0.778 G_GAN_Feat: 0.825 G_ID: 0.229 G_Rec: 0.398 D_GP: 0.042 D_real: 1.411 D_fake: 0.273 +(epoch: 117, iters: 2272, time: 0.064) G_GAN: 0.271 G_GAN_Feat: 0.708 G_ID: 0.161 G_Rec: 0.345 D_GP: 0.039 D_real: 1.106 D_fake: 0.729 +(epoch: 117, iters: 2672, time: 0.064) G_GAN: 0.560 G_GAN_Feat: 0.835 G_ID: 0.220 G_Rec: 0.376 D_GP: 0.040 D_real: 1.151 D_fake: 0.455 +(epoch: 117, iters: 3072, time: 0.064) G_GAN: 0.226 G_GAN_Feat: 0.641 G_ID: 0.149 G_Rec: 0.316 D_GP: 0.023 D_real: 1.128 D_fake: 0.774 +(epoch: 117, iters: 3472, time: 0.064) G_GAN: 0.322 G_GAN_Feat: 0.985 G_ID: 0.255 G_Rec: 0.381 D_GP: 0.289 D_real: 0.539 D_fake: 0.689 +(epoch: 117, iters: 3872, time: 0.064) G_GAN: 0.140 G_GAN_Feat: 0.874 G_ID: 0.143 G_Rec: 0.360 D_GP: 0.233 D_real: 0.502 D_fake: 0.860 +(epoch: 117, iters: 4272, time: 0.064) G_GAN: 0.419 G_GAN_Feat: 0.952 G_ID: 0.249 G_Rec: 0.442 D_GP: 0.088 D_real: 0.792 D_fake: 0.583 +(epoch: 117, iters: 4672, time: 0.064) G_GAN: 0.261 G_GAN_Feat: 0.672 G_ID: 0.187 G_Rec: 0.336 D_GP: 0.025 D_real: 1.113 D_fake: 0.739 +(epoch: 117, iters: 5072, time: 0.064) G_GAN: 0.812 G_GAN_Feat: 0.880 G_ID: 0.214 G_Rec: 0.409 D_GP: 0.040 D_real: 1.355 D_fake: 0.287 +(epoch: 117, iters: 5472, time: 0.064) G_GAN: 0.285 G_GAN_Feat: 1.045 G_ID: 0.174 G_Rec: 0.388 D_GP: 0.333 D_real: 0.484 D_fake: 0.715 +(epoch: 117, iters: 5872, time: 0.064) G_GAN: 0.215 G_GAN_Feat: 0.908 G_ID: 0.314 G_Rec: 0.369 D_GP: 0.049 D_real: 0.722 D_fake: 0.785 +(epoch: 117, iters: 6272, time: 0.064) G_GAN: 0.152 G_GAN_Feat: 0.843 G_ID: 0.178 G_Rec: 0.369 D_GP: 0.152 D_real: 0.643 D_fake: 0.849 +(epoch: 117, iters: 6672, time: 0.064) G_GAN: 0.497 G_GAN_Feat: 1.211 G_ID: 0.276 G_Rec: 0.471 D_GP: 0.147 D_real: 1.091 D_fake: 0.516 +(epoch: 117, iters: 7072, time: 0.064) G_GAN: 0.354 G_GAN_Feat: 0.740 G_ID: 0.192 G_Rec: 0.402 D_GP: 0.035 D_real: 1.114 D_fake: 0.647 +(epoch: 117, iters: 7472, time: 0.064) G_GAN: 0.643 G_GAN_Feat: 0.933 G_ID: 0.301 G_Rec: 0.390 D_GP: 0.071 D_real: 1.061 D_fake: 0.378 +(epoch: 117, iters: 7872, time: 0.064) G_GAN: 0.434 G_GAN_Feat: 0.817 G_ID: 0.162 G_Rec: 0.356 D_GP: 0.047 D_real: 1.080 D_fake: 0.569 +(epoch: 117, iters: 8272, time: 0.064) G_GAN: 0.540 G_GAN_Feat: 0.773 G_ID: 0.243 G_Rec: 0.393 D_GP: 0.027 D_real: 1.321 D_fake: 0.464 +(epoch: 118, iters: 64, time: 0.064) G_GAN: 0.216 G_GAN_Feat: 0.729 G_ID: 0.189 G_Rec: 0.327 D_GP: 0.042 D_real: 1.058 D_fake: 0.784 +(epoch: 118, iters: 464, time: 0.064) G_GAN: 0.156 G_GAN_Feat: 1.131 G_ID: 0.255 G_Rec: 0.396 D_GP: 0.190 D_real: 0.175 D_fake: 0.846 +(epoch: 118, iters: 864, time: 0.064) G_GAN: 0.377 G_GAN_Feat: 0.923 G_ID: 0.161 G_Rec: 0.370 D_GP: 0.136 D_real: 0.778 D_fake: 0.633 +(epoch: 118, iters: 1264, time: 0.064) G_GAN: 0.551 G_GAN_Feat: 0.919 G_ID: 0.214 G_Rec: 0.405 D_GP: 0.026 D_real: 1.308 D_fake: 0.452 +(epoch: 118, iters: 1664, time: 0.064) G_GAN: 0.526 G_GAN_Feat: 0.796 G_ID: 0.138 G_Rec: 0.328 D_GP: 0.061 D_real: 1.153 D_fake: 0.484 +(epoch: 118, iters: 2064, time: 0.064) G_GAN: 0.545 G_GAN_Feat: 0.857 G_ID: 0.215 G_Rec: 0.394 D_GP: 0.032 D_real: 1.126 D_fake: 0.458 +(epoch: 118, iters: 2464, time: 0.064) G_GAN: 0.393 G_GAN_Feat: 0.761 G_ID: 0.169 G_Rec: 0.396 D_GP: 0.028 D_real: 1.169 D_fake: 0.608 +(epoch: 118, iters: 2864, time: 0.064) G_GAN: 0.662 G_GAN_Feat: 0.762 G_ID: 0.224 G_Rec: 0.396 D_GP: 0.023 D_real: 1.411 D_fake: 0.352 +(epoch: 118, iters: 3264, time: 0.064) G_GAN: 0.220 G_GAN_Feat: 0.584 G_ID: 0.164 G_Rec: 0.350 D_GP: 0.022 D_real: 1.154 D_fake: 0.780 +(epoch: 118, iters: 3664, time: 0.064) G_GAN: 0.413 G_GAN_Feat: 0.753 G_ID: 0.263 G_Rec: 0.370 D_GP: 0.029 D_real: 1.070 D_fake: 0.587 +(epoch: 118, iters: 4064, time: 0.064) G_GAN: 0.202 G_GAN_Feat: 0.746 G_ID: 0.174 G_Rec: 0.337 D_GP: 0.045 D_real: 0.990 D_fake: 0.799 +(epoch: 118, iters: 4464, time: 0.064) G_GAN: 0.598 G_GAN_Feat: 0.941 G_ID: 0.224 G_Rec: 0.386 D_GP: 0.250 D_real: 0.879 D_fake: 0.422 +(epoch: 118, iters: 4864, time: 0.064) G_GAN: 0.283 G_GAN_Feat: 0.779 G_ID: 0.149 G_Rec: 0.373 D_GP: 0.051 D_real: 0.986 D_fake: 0.718 +(epoch: 118, iters: 5264, time: 0.064) G_GAN: 0.339 G_GAN_Feat: 0.894 G_ID: 0.255 G_Rec: 0.371 D_GP: 0.080 D_real: 0.820 D_fake: 0.662 +(epoch: 118, iters: 5664, time: 0.064) G_GAN: 0.122 G_GAN_Feat: 0.820 G_ID: 0.168 G_Rec: 0.392 D_GP: 0.047 D_real: 0.724 D_fake: 0.878 +(epoch: 118, iters: 6064, time: 0.064) G_GAN: 0.924 G_GAN_Feat: 1.193 G_ID: 0.223 G_Rec: 0.467 D_GP: 0.052 D_real: 1.473 D_fake: 0.244 +(epoch: 118, iters: 6464, time: 0.064) G_GAN: 0.113 G_GAN_Feat: 0.945 G_ID: 0.131 G_Rec: 0.339 D_GP: 0.116 D_real: 0.448 D_fake: 0.887 +(epoch: 118, iters: 6864, time: 0.064) G_GAN: 0.332 G_GAN_Feat: 0.689 G_ID: 0.243 G_Rec: 0.415 D_GP: 0.018 D_real: 1.109 D_fake: 0.668 +(epoch: 118, iters: 7264, time: 0.064) G_GAN: 0.047 G_GAN_Feat: 0.692 G_ID: 0.182 G_Rec: 0.363 D_GP: 0.023 D_real: 0.972 D_fake: 0.953 +(epoch: 118, iters: 7664, time: 0.064) G_GAN: 0.384 G_GAN_Feat: 0.777 G_ID: 0.204 G_Rec: 0.408 D_GP: 0.023 D_real: 1.165 D_fake: 0.620 +(epoch: 118, iters: 8064, time: 0.064) G_GAN: 0.106 G_GAN_Feat: 0.653 G_ID: 0.168 G_Rec: 0.365 D_GP: 0.022 D_real: 0.970 D_fake: 0.895 +(epoch: 118, iters: 8464, time: 0.064) G_GAN: 0.174 G_GAN_Feat: 0.796 G_ID: 0.235 G_Rec: 0.415 D_GP: 0.034 D_real: 0.912 D_fake: 0.826 +(epoch: 119, iters: 256, time: 0.064) G_GAN: 0.214 G_GAN_Feat: 0.740 G_ID: 0.154 G_Rec: 0.375 D_GP: 0.078 D_real: 0.915 D_fake: 0.788 +(epoch: 119, iters: 656, time: 0.064) G_GAN: 0.330 G_GAN_Feat: 1.097 G_ID: 0.216 G_Rec: 0.418 D_GP: 0.075 D_real: 0.446 D_fake: 0.670 +(epoch: 119, iters: 1056, time: 0.064) G_GAN: 0.158 G_GAN_Feat: 0.703 G_ID: 0.197 G_Rec: 0.333 D_GP: 0.033 D_real: 1.051 D_fake: 0.842 +(epoch: 119, iters: 1456, time: 0.064) G_GAN: 0.455 G_GAN_Feat: 1.001 G_ID: 0.207 G_Rec: 0.396 D_GP: 0.049 D_real: 0.765 D_fake: 0.547 +(epoch: 119, iters: 1856, time: 0.064) G_GAN: 0.379 G_GAN_Feat: 0.712 G_ID: 0.171 G_Rec: 0.326 D_GP: 0.027 D_real: 1.252 D_fake: 0.621 +(epoch: 119, iters: 2256, time: 0.064) G_GAN: 0.233 G_GAN_Feat: 0.713 G_ID: 0.249 G_Rec: 0.367 D_GP: 0.027 D_real: 1.059 D_fake: 0.768 +(epoch: 119, iters: 2656, time: 0.064) G_GAN: 0.104 G_GAN_Feat: 0.695 G_ID: 0.138 G_Rec: 0.353 D_GP: 0.037 D_real: 0.932 D_fake: 0.896 +(epoch: 119, iters: 3056, time: 0.064) G_GAN: 0.580 G_GAN_Feat: 0.882 G_ID: 0.276 G_Rec: 0.392 D_GP: 0.060 D_real: 1.073 D_fake: 0.459 +(epoch: 119, iters: 3456, time: 0.064) G_GAN: 0.177 G_GAN_Feat: 0.707 G_ID: 0.166 G_Rec: 0.333 D_GP: 0.037 D_real: 0.947 D_fake: 0.823 +(epoch: 119, iters: 3856, time: 0.064) G_GAN: 0.189 G_GAN_Feat: 0.859 G_ID: 0.260 G_Rec: 0.439 D_GP: 0.038 D_real: 0.852 D_fake: 0.811 +(epoch: 119, iters: 4256, time: 0.064) G_GAN: 0.223 G_GAN_Feat: 0.701 G_ID: 0.138 G_Rec: 0.308 D_GP: 0.034 D_real: 1.027 D_fake: 0.777 +(epoch: 119, iters: 4656, time: 0.064) G_GAN: 0.278 G_GAN_Feat: 0.922 G_ID: 0.260 G_Rec: 0.384 D_GP: 0.057 D_real: 0.966 D_fake: 0.726 +(epoch: 119, iters: 5056, time: 0.064) G_GAN: -0.072 G_GAN_Feat: 0.697 G_ID: 0.176 G_Rec: 0.346 D_GP: 0.028 D_real: 0.742 D_fake: 1.072 +(epoch: 119, iters: 5456, time: 0.064) G_GAN: 0.740 G_GAN_Feat: 0.805 G_ID: 0.221 G_Rec: 0.376 D_GP: 0.028 D_real: 1.429 D_fake: 0.280 +(epoch: 119, iters: 5856, time: 0.064) G_GAN: 0.223 G_GAN_Feat: 0.770 G_ID: 0.152 G_Rec: 0.347 D_GP: 0.049 D_real: 1.043 D_fake: 0.777 +(epoch: 119, iters: 6256, time: 0.064) G_GAN: 0.411 G_GAN_Feat: 0.838 G_ID: 0.282 G_Rec: 0.384 D_GP: 0.026 D_real: 1.144 D_fake: 0.591 +(epoch: 119, iters: 6656, time: 0.064) G_GAN: 0.211 G_GAN_Feat: 0.912 G_ID: 0.157 G_Rec: 0.386 D_GP: 0.038 D_real: 0.595 D_fake: 0.789 +(epoch: 119, iters: 7056, time: 0.064) G_GAN: 0.227 G_GAN_Feat: 1.007 G_ID: 0.213 G_Rec: 0.367 D_GP: 0.063 D_real: 0.496 D_fake: 0.774 +(epoch: 119, iters: 7456, time: 0.064) G_GAN: 0.306 G_GAN_Feat: 0.867 G_ID: 0.174 G_Rec: 0.348 D_GP: 0.049 D_real: 0.771 D_fake: 0.694 +(epoch: 119, iters: 7856, time: 0.064) G_GAN: 0.629 G_GAN_Feat: 0.988 G_ID: 0.244 G_Rec: 0.418 D_GP: 0.036 D_real: 1.101 D_fake: 0.379 +(epoch: 119, iters: 8256, time: 0.064) G_GAN: 0.307 G_GAN_Feat: 0.845 G_ID: 0.155 G_Rec: 0.349 D_GP: 0.051 D_real: 0.841 D_fake: 0.693 +(epoch: 120, iters: 48, time: 0.064) G_GAN: 0.504 G_GAN_Feat: 0.841 G_ID: 0.196 G_Rec: 0.402 D_GP: 0.024 D_real: 1.160 D_fake: 0.500 +(epoch: 120, iters: 448, time: 0.064) G_GAN: 0.285 G_GAN_Feat: 1.148 G_ID: 0.161 G_Rec: 0.401 D_GP: 1.558 D_real: 0.469 D_fake: 0.724 +(epoch: 120, iters: 848, time: 0.064) G_GAN: 0.344 G_GAN_Feat: 0.824 G_ID: 0.248 G_Rec: 0.418 D_GP: 0.025 D_real: 1.095 D_fake: 0.656 +(epoch: 120, iters: 1248, time: 0.064) G_GAN: 0.474 G_GAN_Feat: 0.770 G_ID: 0.159 G_Rec: 0.386 D_GP: 0.044 D_real: 1.215 D_fake: 0.537 +(epoch: 120, iters: 1648, time: 0.064) G_GAN: 0.405 G_GAN_Feat: 0.898 G_ID: 0.225 G_Rec: 0.415 D_GP: 0.039 D_real: 0.968 D_fake: 0.596 +(epoch: 120, iters: 2048, time: 0.064) G_GAN: 0.278 G_GAN_Feat: 0.745 G_ID: 0.153 G_Rec: 0.366 D_GP: 0.041 D_real: 1.021 D_fake: 0.722 +(epoch: 120, iters: 2448, time: 0.064) G_GAN: 0.589 G_GAN_Feat: 0.986 G_ID: 0.257 G_Rec: 0.397 D_GP: 0.067 D_real: 0.998 D_fake: 0.443 +(epoch: 120, iters: 2848, time: 0.064) G_GAN: 0.201 G_GAN_Feat: 0.699 G_ID: 0.165 G_Rec: 0.327 D_GP: 0.034 D_real: 1.106 D_fake: 0.800 +(epoch: 120, iters: 3248, time: 0.064) G_GAN: 0.517 G_GAN_Feat: 0.830 G_ID: 0.217 G_Rec: 0.388 D_GP: 0.027 D_real: 1.312 D_fake: 0.485 +(epoch: 120, iters: 3648, time: 0.064) G_GAN: 0.510 G_GAN_Feat: 0.768 G_ID: 0.148 G_Rec: 0.347 D_GP: 0.038 D_real: 1.274 D_fake: 0.520 +(epoch: 120, iters: 4048, time: 0.064) G_GAN: 0.618 G_GAN_Feat: 0.885 G_ID: 0.206 G_Rec: 0.389 D_GP: 0.054 D_real: 1.253 D_fake: 0.427 +(epoch: 120, iters: 4448, time: 0.064) G_GAN: 0.479 G_GAN_Feat: 0.879 G_ID: 0.164 G_Rec: 0.366 D_GP: 0.064 D_real: 0.966 D_fake: 0.529 +(epoch: 120, iters: 4848, time: 0.064) G_GAN: 0.647 G_GAN_Feat: 0.769 G_ID: 0.244 G_Rec: 0.374 D_GP: 0.026 D_real: 1.443 D_fake: 0.356 +(epoch: 120, iters: 5248, time: 0.064) G_GAN: 0.259 G_GAN_Feat: 0.639 G_ID: 0.175 G_Rec: 0.327 D_GP: 0.023 D_real: 1.203 D_fake: 0.741 +(epoch: 120, iters: 5648, time: 0.064) G_GAN: 0.399 G_GAN_Feat: 0.933 G_ID: 0.275 G_Rec: 0.392 D_GP: 0.052 D_real: 0.865 D_fake: 0.625 +(epoch: 120, iters: 6048, time: 0.064) G_GAN: 0.300 G_GAN_Feat: 0.968 G_ID: 0.208 G_Rec: 0.355 D_GP: 0.085 D_real: 0.509 D_fake: 0.700 +(epoch: 120, iters: 6448, time: 0.064) G_GAN: 0.201 G_GAN_Feat: 0.804 G_ID: 0.230 G_Rec: 0.385 D_GP: 0.033 D_real: 1.028 D_fake: 0.799 +(epoch: 120, iters: 6848, time: 0.064) G_GAN: 0.270 G_GAN_Feat: 0.930 G_ID: 0.153 G_Rec: 0.366 D_GP: 0.245 D_real: 0.590 D_fake: 0.733 +(epoch: 120, iters: 7248, time: 0.064) G_GAN: 0.698 G_GAN_Feat: 0.902 G_ID: 0.240 G_Rec: 0.401 D_GP: 0.036 D_real: 1.237 D_fake: 0.315 +(epoch: 120, iters: 7648, time: 0.064) G_GAN: 0.155 G_GAN_Feat: 0.748 G_ID: 0.196 G_Rec: 0.400 D_GP: 0.030 D_real: 0.959 D_fake: 0.845 +(epoch: 120, iters: 8048, time: 0.064) G_GAN: 0.340 G_GAN_Feat: 0.869 G_ID: 0.262 G_Rec: 0.364 D_GP: 0.048 D_real: 0.917 D_fake: 0.660 +(epoch: 120, iters: 8448, time: 0.064) G_GAN: 0.331 G_GAN_Feat: 0.700 G_ID: 0.151 G_Rec: 0.312 D_GP: 0.024 D_real: 1.153 D_fake: 0.669 +(epoch: 121, iters: 240, time: 0.064) G_GAN: 0.146 G_GAN_Feat: 0.771 G_ID: 0.226 G_Rec: 0.383 D_GP: 0.022 D_real: 0.931 D_fake: 0.854 +(epoch: 121, iters: 640, time: 0.064) G_GAN: 0.056 G_GAN_Feat: 0.733 G_ID: 0.182 G_Rec: 0.317 D_GP: 0.040 D_real: 0.951 D_fake: 0.944 +(epoch: 121, iters: 1040, time: 0.064) G_GAN: 0.343 G_GAN_Feat: 0.871 G_ID: 0.251 G_Rec: 0.373 D_GP: 0.023 D_real: 1.073 D_fake: 0.657 +(epoch: 121, iters: 1440, time: 0.064) G_GAN: 0.516 G_GAN_Feat: 0.853 G_ID: 0.156 G_Rec: 0.335 D_GP: 0.059 D_real: 1.016 D_fake: 0.488 +(epoch: 121, iters: 1840, time: 0.064) G_GAN: 0.528 G_GAN_Feat: 0.852 G_ID: 0.214 G_Rec: 0.373 D_GP: 0.028 D_real: 1.241 D_fake: 0.516 +(epoch: 121, iters: 2240, time: 0.064) G_GAN: 0.247 G_GAN_Feat: 0.806 G_ID: 0.176 G_Rec: 0.361 D_GP: 0.028 D_real: 0.956 D_fake: 0.753 +(epoch: 121, iters: 2640, time: 0.064) G_GAN: 0.350 G_GAN_Feat: 0.819 G_ID: 0.248 G_Rec: 0.391 D_GP: 0.022 D_real: 1.080 D_fake: 0.650 +(epoch: 121, iters: 3040, time: 0.064) G_GAN: 0.042 G_GAN_Feat: 0.892 G_ID: 0.158 G_Rec: 0.338 D_GP: 0.052 D_real: 0.393 D_fake: 0.958 +(epoch: 121, iters: 3440, time: 0.064) G_GAN: 0.389 G_GAN_Feat: 0.853 G_ID: 0.231 G_Rec: 0.422 D_GP: 0.026 D_real: 1.124 D_fake: 0.614 +(epoch: 121, iters: 3840, time: 0.064) G_GAN: 0.704 G_GAN_Feat: 0.712 G_ID: 0.159 G_Rec: 0.308 D_GP: 0.025 D_real: 1.548 D_fake: 0.324 +(epoch: 121, iters: 4240, time: 0.064) G_GAN: 0.405 G_GAN_Feat: 0.828 G_ID: 0.241 G_Rec: 0.379 D_GP: 0.036 D_real: 1.112 D_fake: 0.597 +(epoch: 121, iters: 4640, time: 0.064) G_GAN: 0.650 G_GAN_Feat: 0.837 G_ID: 0.140 G_Rec: 0.361 D_GP: 0.045 D_real: 1.312 D_fake: 0.425 +(epoch: 121, iters: 5040, time: 0.064) G_GAN: 0.476 G_GAN_Feat: 0.802 G_ID: 0.218 G_Rec: 0.390 D_GP: 0.028 D_real: 1.203 D_fake: 0.526 +(epoch: 121, iters: 5440, time: 0.064) G_GAN: -0.101 G_GAN_Feat: 0.659 G_ID: 0.219 G_Rec: 0.354 D_GP: 0.024 D_real: 0.854 D_fake: 1.101 +(epoch: 121, iters: 5840, time: 0.064) G_GAN: 0.583 G_GAN_Feat: 0.876 G_ID: 0.235 G_Rec: 0.398 D_GP: 0.039 D_real: 1.189 D_fake: 0.422 +(epoch: 121, iters: 6240, time: 0.064) G_GAN: 0.146 G_GAN_Feat: 0.819 G_ID: 0.162 G_Rec: 0.335 D_GP: 0.067 D_real: 0.791 D_fake: 0.854 +(epoch: 121, iters: 6640, time: 0.064) G_GAN: 0.738 G_GAN_Feat: 0.931 G_ID: 0.228 G_Rec: 0.384 D_GP: 0.030 D_real: 1.234 D_fake: 0.300 +(epoch: 121, iters: 7040, time: 0.064) G_GAN: 0.605 G_GAN_Feat: 0.720 G_ID: 0.156 G_Rec: 0.388 D_GP: 0.025 D_real: 1.408 D_fake: 0.439 +(epoch: 121, iters: 7440, time: 0.064) G_GAN: 0.517 G_GAN_Feat: 0.972 G_ID: 0.285 G_Rec: 0.416 D_GP: 0.051 D_real: 0.950 D_fake: 0.526 +(epoch: 121, iters: 7840, time: 0.064) G_GAN: 0.408 G_GAN_Feat: 0.766 G_ID: 0.151 G_Rec: 0.330 D_GP: 0.034 D_real: 1.207 D_fake: 0.593 +(epoch: 121, iters: 8240, time: 0.064) G_GAN: 0.432 G_GAN_Feat: 0.857 G_ID: 0.221 G_Rec: 0.500 D_GP: 0.024 D_real: 1.224 D_fake: 0.577 +(epoch: 122, iters: 32, time: 0.064) G_GAN: 0.190 G_GAN_Feat: 0.668 G_ID: 0.166 G_Rec: 0.356 D_GP: 0.025 D_real: 1.008 D_fake: 0.810 +(epoch: 122, iters: 432, time: 0.064) G_GAN: 0.389 G_GAN_Feat: 0.999 G_ID: 0.272 G_Rec: 0.401 D_GP: 0.068 D_real: 0.868 D_fake: 0.612 +(epoch: 122, iters: 832, time: 0.064) G_GAN: 0.209 G_GAN_Feat: 0.745 G_ID: 0.134 G_Rec: 0.327 D_GP: 0.041 D_real: 0.995 D_fake: 0.791 +(epoch: 122, iters: 1232, time: 0.064) G_GAN: 0.693 G_GAN_Feat: 0.963 G_ID: 0.248 G_Rec: 0.376 D_GP: 0.031 D_real: 1.188 D_fake: 0.345 +(epoch: 122, iters: 1632, time: 0.064) G_GAN: 0.505 G_GAN_Feat: 0.650 G_ID: 0.152 G_Rec: 0.339 D_GP: 0.022 D_real: 1.449 D_fake: 0.496 +(epoch: 122, iters: 2032, time: 0.064) G_GAN: 0.484 G_GAN_Feat: 0.867 G_ID: 0.240 G_Rec: 0.378 D_GP: 0.055 D_real: 1.062 D_fake: 0.520 +(epoch: 122, iters: 2432, time: 0.064) G_GAN: 0.251 G_GAN_Feat: 0.756 G_ID: 0.169 G_Rec: 0.361 D_GP: 0.030 D_real: 1.087 D_fake: 0.749 +(epoch: 122, iters: 2832, time: 0.064) G_GAN: 0.524 G_GAN_Feat: 1.122 G_ID: 0.237 G_Rec: 0.438 D_GP: 0.254 D_real: 0.546 D_fake: 0.483 +(epoch: 122, iters: 3232, time: 0.064) G_GAN: 0.230 G_GAN_Feat: 0.723 G_ID: 0.145 G_Rec: 0.327 D_GP: 0.030 D_real: 1.059 D_fake: 0.770 +(epoch: 122, iters: 3632, time: 0.064) G_GAN: 0.626 G_GAN_Feat: 0.824 G_ID: 0.268 G_Rec: 0.415 D_GP: 0.026 D_real: 1.386 D_fake: 0.390 +(epoch: 122, iters: 4032, time: 0.064) G_GAN: 0.267 G_GAN_Feat: 0.703 G_ID: 0.158 G_Rec: 0.315 D_GP: 0.028 D_real: 1.148 D_fake: 0.733 +(epoch: 122, iters: 4432, time: 0.064) G_GAN: 0.154 G_GAN_Feat: 0.995 G_ID: 0.266 G_Rec: 0.417 D_GP: 0.087 D_real: 0.426 D_fake: 0.846 +(epoch: 122, iters: 4832, time: 0.064) G_GAN: 0.416 G_GAN_Feat: 0.730 G_ID: 0.155 G_Rec: 0.340 D_GP: 0.025 D_real: 1.278 D_fake: 0.586 +(epoch: 122, iters: 5232, time: 0.064) G_GAN: 0.463 G_GAN_Feat: 0.821 G_ID: 0.223 G_Rec: 0.415 D_GP: 0.020 D_real: 1.282 D_fake: 0.538 +(epoch: 122, iters: 5632, time: 0.064) G_GAN: 0.276 G_GAN_Feat: 0.968 G_ID: 0.157 G_Rec: 0.360 D_GP: 0.326 D_real: 0.536 D_fake: 0.729 +(epoch: 122, iters: 6032, time: 0.064) G_GAN: 0.229 G_GAN_Feat: 0.875 G_ID: 0.218 G_Rec: 0.396 D_GP: 0.041 D_real: 0.949 D_fake: 0.771 +(epoch: 122, iters: 6432, time: 0.064) G_GAN: 0.430 G_GAN_Feat: 0.818 G_ID: 0.156 G_Rec: 0.363 D_GP: 0.037 D_real: 1.086 D_fake: 0.577 +(epoch: 122, iters: 6832, time: 0.064) G_GAN: 0.455 G_GAN_Feat: 0.976 G_ID: 0.223 G_Rec: 0.414 D_GP: 0.086 D_real: 0.771 D_fake: 0.552 +(epoch: 122, iters: 7232, time: 0.064) G_GAN: 0.087 G_GAN_Feat: 0.679 G_ID: 0.157 G_Rec: 0.343 D_GP: 0.034 D_real: 0.932 D_fake: 0.913 +(epoch: 122, iters: 7632, time: 0.064) G_GAN: 0.037 G_GAN_Feat: 0.902 G_ID: 0.243 G_Rec: 0.404 D_GP: 0.084 D_real: 0.483 D_fake: 0.963 +(epoch: 122, iters: 8032, time: 0.064) G_GAN: 0.263 G_GAN_Feat: 0.738 G_ID: 0.164 G_Rec: 0.343 D_GP: 0.035 D_real: 1.134 D_fake: 0.738 +(epoch: 122, iters: 8432, time: 0.064) G_GAN: 0.385 G_GAN_Feat: 0.948 G_ID: 0.236 G_Rec: 0.410 D_GP: 0.061 D_real: 0.767 D_fake: 0.620 +(epoch: 123, iters: 224, time: 0.064) G_GAN: 0.609 G_GAN_Feat: 1.026 G_ID: 0.168 G_Rec: 0.396 D_GP: 0.749 D_real: 0.695 D_fake: 0.466 +(epoch: 123, iters: 624, time: 0.064) G_GAN: 0.062 G_GAN_Feat: 1.014 G_ID: 0.270 G_Rec: 0.414 D_GP: 0.184 D_real: 0.294 D_fake: 0.938 +(epoch: 123, iters: 1024, time: 0.064) G_GAN: 0.370 G_GAN_Feat: 0.857 G_ID: 0.153 G_Rec: 0.341 D_GP: 0.054 D_real: 1.074 D_fake: 0.635 +(epoch: 123, iters: 1424, time: 0.064) G_GAN: 0.497 G_GAN_Feat: 0.783 G_ID: 0.212 G_Rec: 0.411 D_GP: 0.021 D_real: 1.257 D_fake: 0.522 +(epoch: 123, iters: 1824, time: 0.064) G_GAN: -0.062 G_GAN_Feat: 0.740 G_ID: 0.167 G_Rec: 0.323 D_GP: 0.032 D_real: 0.855 D_fake: 1.062 +(epoch: 123, iters: 2224, time: 0.064) G_GAN: 0.885 G_GAN_Feat: 1.047 G_ID: 0.258 G_Rec: 0.424 D_GP: 0.183 D_real: 0.945 D_fake: 0.274 +(epoch: 123, iters: 2624, time: 0.064) G_GAN: 0.471 G_GAN_Feat: 0.823 G_ID: 0.153 G_Rec: 0.353 D_GP: 0.033 D_real: 1.139 D_fake: 0.541 +(epoch: 123, iters: 3024, time: 0.064) G_GAN: 0.508 G_GAN_Feat: 0.904 G_ID: 0.223 G_Rec: 0.403 D_GP: 0.032 D_real: 1.142 D_fake: 0.494 +(epoch: 123, iters: 3424, time: 0.064) G_GAN: 0.521 G_GAN_Feat: 0.931 G_ID: 0.147 G_Rec: 0.354 D_GP: 0.195 D_real: 0.833 D_fake: 0.502 +(epoch: 123, iters: 3824, time: 0.064) G_GAN: 0.683 G_GAN_Feat: 0.973 G_ID: 0.260 G_Rec: 0.461 D_GP: 0.039 D_real: 1.352 D_fake: 0.350 +(epoch: 123, iters: 4224, time: 0.064) G_GAN: 0.315 G_GAN_Feat: 0.679 G_ID: 0.163 G_Rec: 0.333 D_GP: 0.028 D_real: 1.145 D_fake: 0.686 +(epoch: 123, iters: 4624, time: 0.064) G_GAN: 0.497 G_GAN_Feat: 0.918 G_ID: 0.240 G_Rec: 0.383 D_GP: 0.058 D_real: 0.972 D_fake: 0.511 +(epoch: 123, iters: 5024, time: 0.064) G_GAN: 0.425 G_GAN_Feat: 0.627 G_ID: 0.150 G_Rec: 0.363 D_GP: 0.019 D_real: 1.272 D_fake: 0.575 +(epoch: 123, iters: 5424, time: 0.064) G_GAN: 0.522 G_GAN_Feat: 0.821 G_ID: 0.233 G_Rec: 0.385 D_GP: 0.035 D_real: 1.116 D_fake: 0.485 +(epoch: 123, iters: 5824, time: 0.064) G_GAN: 0.119 G_GAN_Feat: 0.703 G_ID: 0.142 G_Rec: 0.324 D_GP: 0.037 D_real: 0.939 D_fake: 0.881 +(epoch: 123, iters: 6224, time: 0.064) G_GAN: 0.566 G_GAN_Feat: 0.826 G_ID: 0.247 G_Rec: 0.404 D_GP: 0.023 D_real: 1.339 D_fake: 0.435 +(epoch: 123, iters: 6624, time: 0.064) G_GAN: 0.297 G_GAN_Feat: 0.805 G_ID: 0.169 G_Rec: 0.358 D_GP: 0.055 D_real: 0.939 D_fake: 0.704 +(epoch: 123, iters: 7024, time: 0.064) G_GAN: 0.756 G_GAN_Feat: 1.036 G_ID: 0.235 G_Rec: 0.421 D_GP: 0.121 D_real: 0.878 D_fake: 0.322 +(epoch: 123, iters: 7424, time: 0.064) G_GAN: 0.244 G_GAN_Feat: 0.700 G_ID: 0.169 G_Rec: 0.352 D_GP: 0.031 D_real: 1.088 D_fake: 0.756 +(epoch: 123, iters: 7824, time: 0.064) G_GAN: 0.488 G_GAN_Feat: 1.001 G_ID: 0.231 G_Rec: 0.398 D_GP: 0.068 D_real: 0.808 D_fake: 0.516 +(epoch: 123, iters: 8224, time: 0.064) G_GAN: 0.500 G_GAN_Feat: 0.790 G_ID: 0.168 G_Rec: 0.337 D_GP: 0.036 D_real: 1.253 D_fake: 0.511 +(epoch: 124, iters: 16, time: 0.064) G_GAN: 0.553 G_GAN_Feat: 0.947 G_ID: 0.224 G_Rec: 0.411 D_GP: 0.046 D_real: 1.129 D_fake: 0.462 +(epoch: 124, iters: 416, time: 0.064) G_GAN: 0.003 G_GAN_Feat: 0.846 G_ID: 0.239 G_Rec: 0.345 D_GP: 0.055 D_real: 0.625 D_fake: 0.999 +(epoch: 124, iters: 816, time: 0.064) G_GAN: 0.741 G_GAN_Feat: 1.034 G_ID: 0.251 G_Rec: 0.387 D_GP: 0.121 D_real: 1.060 D_fake: 0.287 +(epoch: 124, iters: 1216, time: 0.064) G_GAN: 0.238 G_GAN_Feat: 0.682 G_ID: 0.148 G_Rec: 0.379 D_GP: 0.021 D_real: 1.194 D_fake: 0.762 +(epoch: 124, iters: 1616, time: 0.064) G_GAN: 0.532 G_GAN_Feat: 1.000 G_ID: 0.237 G_Rec: 0.438 D_GP: 0.139 D_real: 0.971 D_fake: 0.473 +(epoch: 124, iters: 2016, time: 0.064) G_GAN: 0.170 G_GAN_Feat: 0.727 G_ID: 0.158 G_Rec: 0.318 D_GP: 0.047 D_real: 0.946 D_fake: 0.830 +(epoch: 124, iters: 2416, time: 0.064) G_GAN: 0.430 G_GAN_Feat: 0.979 G_ID: 0.239 G_Rec: 0.419 D_GP: 0.067 D_real: 0.881 D_fake: 0.571 +(epoch: 124, iters: 2816, time: 0.064) G_GAN: 0.357 G_GAN_Feat: 0.820 G_ID: 0.153 G_Rec: 0.328 D_GP: 0.040 D_real: 1.148 D_fake: 0.644 +(epoch: 124, iters: 3216, time: 0.064) G_GAN: 0.608 G_GAN_Feat: 0.981 G_ID: 0.216 G_Rec: 0.381 D_GP: 0.066 D_real: 0.851 D_fake: 0.417 +(epoch: 124, iters: 3616, time: 0.064) G_GAN: 0.330 G_GAN_Feat: 0.709 G_ID: 0.182 G_Rec: 0.360 D_GP: 0.028 D_real: 1.210 D_fake: 0.671 +(epoch: 124, iters: 4016, time: 0.064) G_GAN: 0.633 G_GAN_Feat: 0.839 G_ID: 0.237 G_Rec: 0.399 D_GP: 0.026 D_real: 1.321 D_fake: 0.398 +(epoch: 124, iters: 4416, time: 0.064) G_GAN: 0.316 G_GAN_Feat: 0.790 G_ID: 0.173 G_Rec: 0.359 D_GP: 0.043 D_real: 1.077 D_fake: 0.685 +(epoch: 124, iters: 4816, time: 0.064) G_GAN: 0.296 G_GAN_Feat: 0.934 G_ID: 0.265 G_Rec: 0.398 D_GP: 0.060 D_real: 0.726 D_fake: 0.704 +(epoch: 124, iters: 5216, time: 0.064) G_GAN: -0.025 G_GAN_Feat: 0.862 G_ID: 0.168 G_Rec: 0.373 D_GP: 0.066 D_real: 0.758 D_fake: 1.025 +(epoch: 124, iters: 5616, time: 0.064) G_GAN: 0.280 G_GAN_Feat: 0.802 G_ID: 0.237 G_Rec: 0.348 D_GP: 0.029 D_real: 1.017 D_fake: 0.721 +(epoch: 124, iters: 6016, time: 0.064) G_GAN: 0.377 G_GAN_Feat: 0.806 G_ID: 0.149 G_Rec: 0.359 D_GP: 0.063 D_real: 1.078 D_fake: 0.624 +(epoch: 124, iters: 6416, time: 0.064) G_GAN: 0.588 G_GAN_Feat: 1.019 G_ID: 0.237 G_Rec: 0.441 D_GP: 0.106 D_real: 1.096 D_fake: 0.473 +(epoch: 124, iters: 6816, time: 0.064) G_GAN: -0.010 G_GAN_Feat: 0.873 G_ID: 0.157 G_Rec: 0.369 D_GP: 0.030 D_real: 0.750 D_fake: 1.012 +(epoch: 124, iters: 7216, time: 0.064) G_GAN: 0.559 G_GAN_Feat: 1.126 G_ID: 0.269 G_Rec: 0.380 D_GP: 0.097 D_real: 0.622 D_fake: 0.444 +(epoch: 124, iters: 7616, time: 0.064) G_GAN: 0.144 G_GAN_Feat: 0.726 G_ID: 0.125 G_Rec: 0.330 D_GP: 0.024 D_real: 1.026 D_fake: 0.856 +(epoch: 124, iters: 8016, time: 0.064) G_GAN: 0.604 G_GAN_Feat: 0.912 G_ID: 0.217 G_Rec: 0.393 D_GP: 0.034 D_real: 1.225 D_fake: 0.403 +(epoch: 124, iters: 8416, time: 0.064) G_GAN: 0.154 G_GAN_Feat: 0.945 G_ID: 0.161 G_Rec: 0.373 D_GP: 0.366 D_real: 0.520 D_fake: 0.848 +(epoch: 125, iters: 208, time: 0.064) G_GAN: 0.476 G_GAN_Feat: 0.916 G_ID: 0.239 G_Rec: 0.384 D_GP: 0.035 D_real: 1.059 D_fake: 0.530 +(epoch: 125, iters: 608, time: 0.064) G_GAN: 0.081 G_GAN_Feat: 0.702 G_ID: 0.178 G_Rec: 0.336 D_GP: 0.027 D_real: 0.980 D_fake: 0.919 +(epoch: 125, iters: 1008, time: 0.064) G_GAN: 0.233 G_GAN_Feat: 1.068 G_ID: 0.216 G_Rec: 0.429 D_GP: 0.066 D_real: 0.766 D_fake: 0.771 +(epoch: 125, iters: 1408, time: 0.064) G_GAN: -0.032 G_GAN_Feat: 0.886 G_ID: 0.153 G_Rec: 0.337 D_GP: 0.053 D_real: 0.454 D_fake: 1.032 +(epoch: 125, iters: 1808, time: 0.064) G_GAN: 0.180 G_GAN_Feat: 1.065 G_ID: 0.253 G_Rec: 0.390 D_GP: 0.069 D_real: 0.426 D_fake: 0.820 +(epoch: 125, iters: 2208, time: 0.064) G_GAN: 0.128 G_GAN_Feat: 0.987 G_ID: 0.165 G_Rec: 0.334 D_GP: 0.194 D_real: 0.516 D_fake: 0.872 +(epoch: 125, iters: 2608, time: 0.064) G_GAN: 0.368 G_GAN_Feat: 0.908 G_ID: 0.242 G_Rec: 0.424 D_GP: 0.025 D_real: 1.086 D_fake: 0.632 +(epoch: 125, iters: 3008, time: 0.064) G_GAN: 0.179 G_GAN_Feat: 0.632 G_ID: 0.183 G_Rec: 0.342 D_GP: 0.020 D_real: 1.102 D_fake: 0.821 +(epoch: 125, iters: 3408, time: 0.064) G_GAN: 0.354 G_GAN_Feat: 0.764 G_ID: 0.268 G_Rec: 0.372 D_GP: 0.024 D_real: 1.107 D_fake: 0.652 +(epoch: 125, iters: 3808, time: 0.064) G_GAN: 0.168 G_GAN_Feat: 0.732 G_ID: 0.135 G_Rec: 0.336 D_GP: 0.062 D_real: 0.869 D_fake: 0.833 +(epoch: 125, iters: 4208, time: 0.064) G_GAN: 0.585 G_GAN_Feat: 0.800 G_ID: 0.249 G_Rec: 0.382 D_GP: 0.027 D_real: 1.313 D_fake: 0.424 +(epoch: 125, iters: 4608, time: 0.064) G_GAN: 0.179 G_GAN_Feat: 1.047 G_ID: 0.183 G_Rec: 0.351 D_GP: 0.093 D_real: 0.528 D_fake: 0.821 +(epoch: 125, iters: 5008, time: 0.064) G_GAN: 0.531 G_GAN_Feat: 0.866 G_ID: 0.245 G_Rec: 0.379 D_GP: 0.049 D_real: 1.131 D_fake: 0.476 +(epoch: 125, iters: 5408, time: 0.064) G_GAN: 0.452 G_GAN_Feat: 0.869 G_ID: 0.147 G_Rec: 0.358 D_GP: 0.043 D_real: 1.127 D_fake: 0.549 +(epoch: 125, iters: 5808, time: 0.064) G_GAN: 0.549 G_GAN_Feat: 1.077 G_ID: 0.257 G_Rec: 0.471 D_GP: 0.210 D_real: 0.810 D_fake: 0.460 +(epoch: 125, iters: 6208, time: 0.064) G_GAN: 0.286 G_GAN_Feat: 0.855 G_ID: 0.170 G_Rec: 0.348 D_GP: 0.068 D_real: 0.863 D_fake: 0.717 +(epoch: 125, iters: 6608, time: 0.064) G_GAN: 0.081 G_GAN_Feat: 0.831 G_ID: 0.213 G_Rec: 0.361 D_GP: 0.026 D_real: 0.885 D_fake: 0.919 +(epoch: 125, iters: 7008, time: 0.064) G_GAN: 0.405 G_GAN_Feat: 0.813 G_ID: 0.141 G_Rec: 0.376 D_GP: 0.040 D_real: 1.153 D_fake: 0.605 +(epoch: 125, iters: 7408, time: 0.064) G_GAN: 0.831 G_GAN_Feat: 1.029 G_ID: 0.268 G_Rec: 0.425 D_GP: 0.088 D_real: 1.023 D_fake: 0.254 +(epoch: 125, iters: 7808, time: 0.064) G_GAN: 0.282 G_GAN_Feat: 0.799 G_ID: 0.149 G_Rec: 0.358 D_GP: 0.033 D_real: 0.979 D_fake: 0.718 +(epoch: 125, iters: 8208, time: 0.064) G_GAN: 0.458 G_GAN_Feat: 1.034 G_ID: 0.257 G_Rec: 0.410 D_GP: 0.046 D_real: 0.767 D_fake: 0.551 +(epoch: 125, iters: 8608, time: 0.064) G_GAN: 0.393 G_GAN_Feat: 0.723 G_ID: 0.174 G_Rec: 0.347 D_GP: 0.025 D_real: 1.328 D_fake: 0.608 +(epoch: 126, iters: 400, time: 0.064) G_GAN: 0.246 G_GAN_Feat: 0.938 G_ID: 0.241 G_Rec: 0.383 D_GP: 0.046 D_real: 0.953 D_fake: 0.755 +(epoch: 126, iters: 800, time: 0.064) G_GAN: 0.381 G_GAN_Feat: 0.855 G_ID: 0.150 G_Rec: 0.364 D_GP: 0.046 D_real: 1.070 D_fake: 0.627 +(epoch: 126, iters: 1200, time: 0.064) G_GAN: 0.648 G_GAN_Feat: 1.118 G_ID: 0.217 G_Rec: 0.415 D_GP: 0.097 D_real: 0.801 D_fake: 0.414 +(epoch: 126, iters: 1600, time: 0.064) G_GAN: 0.280 G_GAN_Feat: 0.835 G_ID: 0.147 G_Rec: 0.363 D_GP: 0.029 D_real: 1.040 D_fake: 0.720 +(epoch: 126, iters: 2000, time: 0.064) G_GAN: 0.591 G_GAN_Feat: 0.918 G_ID: 0.245 G_Rec: 0.404 D_GP: 0.027 D_real: 1.183 D_fake: 0.414 +(epoch: 126, iters: 2400, time: 0.064) G_GAN: 0.627 G_GAN_Feat: 0.839 G_ID: 0.136 G_Rec: 0.351 D_GP: 0.058 D_real: 1.328 D_fake: 0.386 +(epoch: 126, iters: 2800, time: 0.064) G_GAN: 0.380 G_GAN_Feat: 0.966 G_ID: 0.227 G_Rec: 0.431 D_GP: 0.039 D_real: 0.905 D_fake: 0.624 +(epoch: 126, iters: 3200, time: 0.064) G_GAN: 0.336 G_GAN_Feat: 0.651 G_ID: 0.165 G_Rec: 0.337 D_GP: 0.021 D_real: 1.219 D_fake: 0.664 +(epoch: 126, iters: 3600, time: 0.064) G_GAN: 0.417 G_GAN_Feat: 0.981 G_ID: 0.233 G_Rec: 0.442 D_GP: 0.056 D_real: 0.843 D_fake: 0.586 +(epoch: 126, iters: 4000, time: 0.064) G_GAN: 0.233 G_GAN_Feat: 0.709 G_ID: 0.189 G_Rec: 0.325 D_GP: 0.027 D_real: 1.094 D_fake: 0.768 +(epoch: 126, iters: 4400, time: 0.064) G_GAN: 0.316 G_GAN_Feat: 0.831 G_ID: 0.257 G_Rec: 0.397 D_GP: 0.035 D_real: 1.044 D_fake: 0.685 +(epoch: 126, iters: 4800, time: 0.064) G_GAN: 0.346 G_GAN_Feat: 0.818 G_ID: 0.192 G_Rec: 0.343 D_GP: 0.038 D_real: 1.091 D_fake: 0.654 +(epoch: 126, iters: 5200, time: 0.064) G_GAN: 0.392 G_GAN_Feat: 0.921 G_ID: 0.222 G_Rec: 0.376 D_GP: 0.040 D_real: 0.954 D_fake: 0.608 +(epoch: 126, iters: 5600, time: 0.064) G_GAN: 0.114 G_GAN_Feat: 0.715 G_ID: 0.190 G_Rec: 0.350 D_GP: 0.026 D_real: 1.051 D_fake: 0.887 +(epoch: 126, iters: 6000, time: 0.064) G_GAN: 0.350 G_GAN_Feat: 0.831 G_ID: 0.193 G_Rec: 0.394 D_GP: 0.036 D_real: 1.003 D_fake: 0.652 +(epoch: 126, iters: 6400, time: 0.064) G_GAN: 0.396 G_GAN_Feat: 0.861 G_ID: 0.144 G_Rec: 0.371 D_GP: 0.052 D_real: 0.966 D_fake: 0.633 +(epoch: 126, iters: 6800, time: 0.064) G_GAN: 0.172 G_GAN_Feat: 1.216 G_ID: 0.290 G_Rec: 0.424 D_GP: 0.241 D_real: 0.306 D_fake: 0.828 +(epoch: 126, iters: 7200, time: 0.064) G_GAN: 0.272 G_GAN_Feat: 0.779 G_ID: 0.144 G_Rec: 0.326 D_GP: 0.030 D_real: 1.037 D_fake: 0.728 +(epoch: 126, iters: 7600, time: 0.064) G_GAN: 0.407 G_GAN_Feat: 0.866 G_ID: 0.236 G_Rec: 0.398 D_GP: 0.030 D_real: 1.095 D_fake: 0.594 +(epoch: 126, iters: 8000, time: 0.064) G_GAN: 0.291 G_GAN_Feat: 0.929 G_ID: 0.155 G_Rec: 0.352 D_GP: 0.051 D_real: 0.645 D_fake: 0.709 +(epoch: 126, iters: 8400, time: 0.064) G_GAN: 0.328 G_GAN_Feat: 0.869 G_ID: 0.207 G_Rec: 0.355 D_GP: 0.033 D_real: 1.058 D_fake: 0.673 +(epoch: 127, iters: 192, time: 0.064) G_GAN: 0.595 G_GAN_Feat: 0.758 G_ID: 0.141 G_Rec: 0.381 D_GP: 0.037 D_real: 1.391 D_fake: 0.413 +(epoch: 127, iters: 592, time: 0.064) G_GAN: 0.268 G_GAN_Feat: 0.877 G_ID: 0.232 G_Rec: 0.430 D_GP: 0.026 D_real: 0.967 D_fake: 0.733 +(epoch: 127, iters: 992, time: 0.064) G_GAN: 0.170 G_GAN_Feat: 0.610 G_ID: 0.140 G_Rec: 0.310 D_GP: 0.025 D_real: 1.056 D_fake: 0.830 +(epoch: 127, iters: 1392, time: 0.064) G_GAN: 0.290 G_GAN_Feat: 0.724 G_ID: 0.195 G_Rec: 0.354 D_GP: 0.026 D_real: 1.007 D_fake: 0.714 +(epoch: 127, iters: 1792, time: 0.064) G_GAN: -0.018 G_GAN_Feat: 0.745 G_ID: 0.160 G_Rec: 0.337 D_GP: 0.059 D_real: 0.753 D_fake: 1.018 +(epoch: 127, iters: 2192, time: 0.064) G_GAN: 0.382 G_GAN_Feat: 0.912 G_ID: 0.271 G_Rec: 0.415 D_GP: 0.058 D_real: 0.855 D_fake: 0.620 +(epoch: 127, iters: 2592, time: 0.064) G_GAN: 0.605 G_GAN_Feat: 0.773 G_ID: 0.131 G_Rec: 0.364 D_GP: 0.048 D_real: 1.401 D_fake: 0.412 +(epoch: 127, iters: 2992, time: 0.064) G_GAN: 0.417 G_GAN_Feat: 0.864 G_ID: 0.248 G_Rec: 0.396 D_GP: 0.034 D_real: 1.134 D_fake: 0.584 +(epoch: 127, iters: 3392, time: 0.064) G_GAN: 0.056 G_GAN_Feat: 0.943 G_ID: 0.147 G_Rec: 0.332 D_GP: 0.116 D_real: 0.360 D_fake: 0.944 +(epoch: 127, iters: 3792, time: 0.064) G_GAN: 0.632 G_GAN_Feat: 0.949 G_ID: 0.204 G_Rec: 0.392 D_GP: 0.064 D_real: 1.086 D_fake: 0.394 +(epoch: 127, iters: 4192, time: 0.064) G_GAN: 0.093 G_GAN_Feat: 1.055 G_ID: 0.168 G_Rec: 0.354 D_GP: 0.246 D_real: 0.424 D_fake: 0.907 +(epoch: 127, iters: 4592, time: 0.064) G_GAN: 0.395 G_GAN_Feat: 0.864 G_ID: 0.233 G_Rec: 0.401 D_GP: 0.027 D_real: 1.162 D_fake: 0.609 +(epoch: 127, iters: 4992, time: 0.064) G_GAN: 0.073 G_GAN_Feat: 0.948 G_ID: 0.174 G_Rec: 0.381 D_GP: 0.210 D_real: 0.481 D_fake: 0.928 +(epoch: 127, iters: 5392, time: 0.064) G_GAN: 0.300 G_GAN_Feat: 1.121 G_ID: 0.247 G_Rec: 0.438 D_GP: 0.332 D_real: 0.563 D_fake: 0.712 +(epoch: 127, iters: 5792, time: 0.064) G_GAN: 0.322 G_GAN_Feat: 0.798 G_ID: 0.178 G_Rec: 0.334 D_GP: 0.040 D_real: 1.050 D_fake: 0.683 +(epoch: 127, iters: 6192, time: 0.064) G_GAN: 0.573 G_GAN_Feat: 0.914 G_ID: 0.201 G_Rec: 0.393 D_GP: 0.040 D_real: 1.037 D_fake: 0.439 +(epoch: 127, iters: 6592, time: 0.064) G_GAN: 0.273 G_GAN_Feat: 0.732 G_ID: 0.173 G_Rec: 0.310 D_GP: 0.032 D_real: 1.135 D_fake: 0.727 +(epoch: 127, iters: 6992, time: 0.064) G_GAN: 0.547 G_GAN_Feat: 1.269 G_ID: 0.209 G_Rec: 0.420 D_GP: 0.058 D_real: 0.808 D_fake: 0.455 +(epoch: 127, iters: 7392, time: 0.064) G_GAN: 0.223 G_GAN_Feat: 0.971 G_ID: 0.133 G_Rec: 0.366 D_GP: 0.055 D_real: 0.626 D_fake: 0.777 +(epoch: 127, iters: 7792, time: 0.064) G_GAN: 0.873 G_GAN_Feat: 1.044 G_ID: 0.224 G_Rec: 0.441 D_GP: 0.157 D_real: 1.217 D_fake: 0.369 +(epoch: 127, iters: 8192, time: 0.064) G_GAN: 0.392 G_GAN_Feat: 0.974 G_ID: 0.149 G_Rec: 0.410 D_GP: 0.067 D_real: 0.627 D_fake: 0.609 +(epoch: 127, iters: 8592, time: 0.064) G_GAN: 0.412 G_GAN_Feat: 1.075 G_ID: 0.288 G_Rec: 0.412 D_GP: 0.088 D_real: 0.534 D_fake: 0.605 +(epoch: 128, iters: 384, time: 0.064) G_GAN: -0.048 G_GAN_Feat: 0.785 G_ID: 0.205 G_Rec: 0.389 D_GP: 0.063 D_real: 0.935 D_fake: 1.051 +(epoch: 128, iters: 784, time: 0.064) G_GAN: 0.307 G_GAN_Feat: 0.803 G_ID: 0.232 G_Rec: 0.388 D_GP: 0.033 D_real: 0.994 D_fake: 0.698 +(epoch: 128, iters: 1184, time: 0.064) G_GAN: 0.324 G_GAN_Feat: 0.819 G_ID: 0.167 G_Rec: 0.368 D_GP: 0.039 D_real: 1.056 D_fake: 0.681 +(epoch: 128, iters: 1584, time: 0.064) G_GAN: 0.094 G_GAN_Feat: 1.005 G_ID: 0.205 G_Rec: 0.403 D_GP: 0.032 D_real: 0.840 D_fake: 0.906 +(epoch: 128, iters: 1984, time: 0.064) G_GAN: 0.418 G_GAN_Feat: 0.969 G_ID: 0.150 G_Rec: 0.374 D_GP: 0.041 D_real: 1.124 D_fake: 0.585 +(epoch: 128, iters: 2384, time: 0.064) G_GAN: 0.339 G_GAN_Feat: 0.877 G_ID: 0.230 G_Rec: 0.367 D_GP: 0.044 D_real: 0.924 D_fake: 0.666 +(epoch: 128, iters: 2784, time: 0.064) G_GAN: 0.053 G_GAN_Feat: 1.163 G_ID: 0.154 G_Rec: 0.368 D_GP: 0.139 D_real: 1.110 D_fake: 0.947 +(epoch: 128, iters: 3184, time: 0.064) G_GAN: 0.378 G_GAN_Feat: 0.990 G_ID: 0.234 G_Rec: 0.421 D_GP: 0.047 D_real: 1.118 D_fake: 0.625 +(epoch: 128, iters: 3584, time: 0.064) G_GAN: 0.248 G_GAN_Feat: 0.734 G_ID: 0.158 G_Rec: 0.357 D_GP: 0.026 D_real: 1.105 D_fake: 0.753 +(epoch: 128, iters: 3984, time: 0.064) G_GAN: 0.216 G_GAN_Feat: 0.848 G_ID: 0.255 G_Rec: 0.404 D_GP: 0.032 D_real: 0.916 D_fake: 0.784 +(epoch: 128, iters: 4384, time: 0.064) G_GAN: -0.008 G_GAN_Feat: 0.767 G_ID: 0.170 G_Rec: 0.371 D_GP: 0.032 D_real: 0.742 D_fake: 1.009 +(epoch: 128, iters: 4784, time: 0.064) G_GAN: 0.573 G_GAN_Feat: 1.194 G_ID: 0.232 G_Rec: 0.369 D_GP: 0.072 D_real: 0.439 D_fake: 0.437 +(epoch: 128, iters: 5184, time: 0.064) G_GAN: 0.301 G_GAN_Feat: 0.677 G_ID: 0.156 G_Rec: 0.374 D_GP: 0.021 D_real: 1.207 D_fake: 0.700 +(epoch: 128, iters: 5584, time: 0.064) G_GAN: 0.496 G_GAN_Feat: 0.792 G_ID: 0.215 G_Rec: 0.390 D_GP: 0.027 D_real: 1.174 D_fake: 0.522 +(epoch: 128, iters: 5984, time: 0.064) G_GAN: 0.277 G_GAN_Feat: 0.696 G_ID: 0.133 G_Rec: 0.336 D_GP: 0.049 D_real: 1.119 D_fake: 0.724 +(epoch: 128, iters: 6384, time: 0.064) G_GAN: 0.337 G_GAN_Feat: 0.843 G_ID: 0.235 G_Rec: 0.411 D_GP: 0.025 D_real: 1.090 D_fake: 0.665 +(epoch: 128, iters: 6784, time: 0.064) G_GAN: -0.064 G_GAN_Feat: 0.831 G_ID: 0.181 G_Rec: 0.348 D_GP: 0.133 D_real: 0.521 D_fake: 1.065 +(epoch: 128, iters: 7184, time: 0.064) G_GAN: 0.331 G_GAN_Feat: 0.751 G_ID: 0.248 G_Rec: 0.378 D_GP: 0.020 D_real: 1.151 D_fake: 0.669 +(epoch: 128, iters: 7584, time: 0.064) G_GAN: 0.252 G_GAN_Feat: 0.675 G_ID: 0.165 G_Rec: 0.307 D_GP: 0.041 D_real: 1.054 D_fake: 0.748 +(epoch: 128, iters: 7984, time: 0.064) G_GAN: -0.039 G_GAN_Feat: 0.986 G_ID: 0.226 G_Rec: 0.382 D_GP: 0.229 D_real: 0.313 D_fake: 1.040 +(epoch: 128, iters: 8384, time: 0.064) G_GAN: 0.026 G_GAN_Feat: 1.019 G_ID: 0.146 G_Rec: 0.364 D_GP: 0.469 D_real: 0.464 D_fake: 0.974 +(epoch: 129, iters: 176, time: 0.064) G_GAN: 0.567 G_GAN_Feat: 0.870 G_ID: 0.249 G_Rec: 0.363 D_GP: 0.061 D_real: 0.943 D_fake: 0.444 +(epoch: 129, iters: 576, time: 0.064) G_GAN: 0.452 G_GAN_Feat: 0.801 G_ID: 0.161 G_Rec: 0.357 D_GP: 0.030 D_real: 1.257 D_fake: 0.553 +(epoch: 129, iters: 976, time: 0.064) G_GAN: 0.537 G_GAN_Feat: 0.980 G_ID: 0.214 G_Rec: 0.424 D_GP: 0.034 D_real: 0.985 D_fake: 0.470 +(epoch: 129, iters: 1376, time: 0.064) G_GAN: 0.345 G_GAN_Feat: 0.840 G_ID: 0.154 G_Rec: 0.360 D_GP: 0.031 D_real: 1.073 D_fake: 0.657 +(epoch: 129, iters: 1776, time: 0.064) G_GAN: 0.469 G_GAN_Feat: 1.039 G_ID: 0.215 G_Rec: 0.442 D_GP: 0.091 D_real: 0.830 D_fake: 0.562 +(epoch: 129, iters: 2176, time: 0.064) G_GAN: 0.307 G_GAN_Feat: 0.863 G_ID: 0.146 G_Rec: 0.330 D_GP: 0.033 D_real: 0.972 D_fake: 0.695 +(epoch: 129, iters: 2576, time: 0.064) G_GAN: 0.210 G_GAN_Feat: 0.838 G_ID: 0.245 G_Rec: 0.393 D_GP: 0.033 D_real: 0.881 D_fake: 0.790 +(epoch: 129, iters: 2976, time: 0.064) G_GAN: 0.271 G_GAN_Feat: 0.788 G_ID: 0.175 G_Rec: 0.313 D_GP: 0.040 D_real: 1.011 D_fake: 0.732 +(epoch: 129, iters: 3376, time: 0.064) G_GAN: 0.511 G_GAN_Feat: 0.835 G_ID: 0.248 G_Rec: 0.373 D_GP: 0.030 D_real: 1.278 D_fake: 0.492 +(epoch: 129, iters: 3776, time: 0.064) G_GAN: 0.146 G_GAN_Feat: 0.751 G_ID: 0.148 G_Rec: 0.359 D_GP: 0.028 D_real: 0.973 D_fake: 0.854 +(epoch: 129, iters: 4176, time: 0.064) G_GAN: 0.714 G_GAN_Feat: 1.004 G_ID: 0.177 G_Rec: 0.404 D_GP: 0.042 D_real: 1.085 D_fake: 0.316 +(epoch: 129, iters: 4576, time: 0.064) G_GAN: 0.105 G_GAN_Feat: 0.836 G_ID: 0.147 G_Rec: 0.348 D_GP: 0.041 D_real: 0.867 D_fake: 0.895 +(epoch: 129, iters: 4976, time: 0.064) G_GAN: 0.390 G_GAN_Feat: 1.134 G_ID: 0.210 G_Rec: 0.418 D_GP: 0.332 D_real: 0.543 D_fake: 0.615 +(epoch: 129, iters: 5376, time: 0.064) G_GAN: 0.217 G_GAN_Feat: 0.689 G_ID: 0.165 G_Rec: 0.366 D_GP: 0.025 D_real: 1.103 D_fake: 0.783 +(epoch: 129, iters: 5776, time: 0.064) G_GAN: 0.368 G_GAN_Feat: 0.938 G_ID: 0.216 G_Rec: 0.399 D_GP: 0.069 D_real: 0.929 D_fake: 0.633 +(epoch: 129, iters: 6176, time: 0.064) G_GAN: 0.295 G_GAN_Feat: 0.860 G_ID: 0.165 G_Rec: 0.314 D_GP: 0.107 D_real: 0.736 D_fake: 0.705 +(epoch: 129, iters: 6576, time: 0.064) G_GAN: 0.493 G_GAN_Feat: 0.952 G_ID: 0.239 G_Rec: 0.431 D_GP: 0.037 D_real: 0.934 D_fake: 0.512 +(epoch: 129, iters: 6976, time: 0.064) G_GAN: 0.210 G_GAN_Feat: 0.737 G_ID: 0.171 G_Rec: 0.347 D_GP: 0.031 D_real: 1.012 D_fake: 0.790 +(epoch: 129, iters: 7376, time: 0.064) G_GAN: 0.605 G_GAN_Feat: 0.866 G_ID: 0.206 G_Rec: 0.381 D_GP: 0.032 D_real: 1.196 D_fake: 0.399 +(epoch: 129, iters: 7776, time: 0.064) G_GAN: 0.331 G_GAN_Feat: 0.641 G_ID: 0.167 G_Rec: 0.336 D_GP: 0.018 D_real: 1.226 D_fake: 0.670 +(epoch: 129, iters: 8176, time: 0.064) G_GAN: 0.502 G_GAN_Feat: 0.858 G_ID: 0.277 G_Rec: 0.418 D_GP: 0.023 D_real: 1.206 D_fake: 0.509 +(epoch: 129, iters: 8576, time: 0.064) G_GAN: 0.318 G_GAN_Feat: 0.721 G_ID: 0.157 G_Rec: 0.328 D_GP: 0.039 D_real: 1.083 D_fake: 0.683 +(epoch: 130, iters: 368, time: 0.064) G_GAN: 0.487 G_GAN_Feat: 0.923 G_ID: 0.225 G_Rec: 0.394 D_GP: 0.042 D_real: 0.917 D_fake: 0.515 +(epoch: 130, iters: 768, time: 0.064) G_GAN: 0.135 G_GAN_Feat: 0.706 G_ID: 0.197 G_Rec: 0.328 D_GP: 0.035 D_real: 0.989 D_fake: 0.866 +(epoch: 130, iters: 1168, time: 0.064) G_GAN: 0.213 G_GAN_Feat: 0.876 G_ID: 0.218 G_Rec: 0.386 D_GP: 0.044 D_real: 0.870 D_fake: 0.793 +(epoch: 130, iters: 1568, time: 0.064) G_GAN: 0.175 G_GAN_Feat: 0.886 G_ID: 0.159 G_Rec: 0.336 D_GP: 0.168 D_real: 0.707 D_fake: 0.826 +(epoch: 130, iters: 1968, time: 0.064) G_GAN: 0.253 G_GAN_Feat: 1.193 G_ID: 0.237 G_Rec: 0.394 D_GP: 0.165 D_real: 0.257 D_fake: 0.748 +(epoch: 130, iters: 2368, time: 0.064) G_GAN: 0.271 G_GAN_Feat: 1.080 G_ID: 0.134 G_Rec: 0.354 D_GP: 0.071 D_real: 0.403 D_fake: 0.730 +(epoch: 130, iters: 2768, time: 0.064) G_GAN: 0.455 G_GAN_Feat: 0.843 G_ID: 0.224 G_Rec: 0.398 D_GP: 0.027 D_real: 1.162 D_fake: 0.547 +(epoch: 130, iters: 3168, time: 0.064) G_GAN: 0.375 G_GAN_Feat: 0.777 G_ID: 0.172 G_Rec: 0.351 D_GP: 0.038 D_real: 1.197 D_fake: 0.626 +(epoch: 130, iters: 3568, time: 0.064) G_GAN: 0.748 G_GAN_Feat: 1.039 G_ID: 0.195 G_Rec: 0.394 D_GP: 0.279 D_real: 0.878 D_fake: 0.387 +(epoch: 130, iters: 3968, time: 0.064) G_GAN: 0.350 G_GAN_Feat: 0.795 G_ID: 0.159 G_Rec: 0.328 D_GP: 0.038 D_real: 1.155 D_fake: 0.650 +(epoch: 130, iters: 4368, time: 0.064) G_GAN: 0.409 G_GAN_Feat: 0.882 G_ID: 0.211 G_Rec: 0.442 D_GP: 0.033 D_real: 1.191 D_fake: 0.594 +(epoch: 130, iters: 4768, time: 0.064) G_GAN: 0.344 G_GAN_Feat: 0.685 G_ID: 0.153 G_Rec: 0.328 D_GP: 0.022 D_real: 1.279 D_fake: 0.656 +(epoch: 130, iters: 5168, time: 0.064) G_GAN: 0.729 G_GAN_Feat: 0.883 G_ID: 0.207 G_Rec: 0.395 D_GP: 0.027 D_real: 1.403 D_fake: 0.288 +(epoch: 130, iters: 5568, time: 0.064) G_GAN: -0.156 G_GAN_Feat: 0.928 G_ID: 0.159 G_Rec: 0.360 D_GP: 0.062 D_real: 0.958 D_fake: 1.156 +(epoch: 130, iters: 5968, time: 0.064) G_GAN: 0.538 G_GAN_Feat: 1.093 G_ID: 0.190 G_Rec: 0.394 D_GP: 0.042 D_real: 0.654 D_fake: 0.468 +(epoch: 130, iters: 6368, time: 0.064) G_GAN: 0.208 G_GAN_Feat: 0.680 G_ID: 0.154 G_Rec: 0.355 D_GP: 0.022 D_real: 1.162 D_fake: 0.794 +(epoch: 130, iters: 6768, time: 0.064) G_GAN: 0.276 G_GAN_Feat: 0.826 G_ID: 0.227 G_Rec: 0.388 D_GP: 0.027 D_real: 1.005 D_fake: 0.724 +(epoch: 130, iters: 7168, time: 0.064) G_GAN: 0.301 G_GAN_Feat: 0.670 G_ID: 0.147 G_Rec: 0.335 D_GP: 0.024 D_real: 1.221 D_fake: 0.699 +(epoch: 130, iters: 7568, time: 0.064) G_GAN: 0.506 G_GAN_Feat: 0.878 G_ID: 0.246 G_Rec: 0.400 D_GP: 0.028 D_real: 1.217 D_fake: 0.514 +(epoch: 130, iters: 7968, time: 0.064) G_GAN: 0.224 G_GAN_Feat: 0.751 G_ID: 0.150 G_Rec: 0.341 D_GP: 0.032 D_real: 1.093 D_fake: 0.777 +(epoch: 130, iters: 8368, time: 0.064) G_GAN: 0.369 G_GAN_Feat: 0.892 G_ID: 0.242 G_Rec: 0.395 D_GP: 0.047 D_real: 0.900 D_fake: 0.631 +(epoch: 131, iters: 160, time: 0.064) G_GAN: 0.329 G_GAN_Feat: 0.778 G_ID: 0.124 G_Rec: 0.339 D_GP: 0.040 D_real: 1.137 D_fake: 0.673 +(epoch: 131, iters: 560, time: 0.064) G_GAN: 0.341 G_GAN_Feat: 0.828 G_ID: 0.228 G_Rec: 0.388 D_GP: 0.034 D_real: 1.080 D_fake: 0.662 +(epoch: 131, iters: 960, time: 0.064) G_GAN: 0.330 G_GAN_Feat: 0.766 G_ID: 0.144 G_Rec: 0.323 D_GP: 0.038 D_real: 1.056 D_fake: 0.672 +(epoch: 131, iters: 1360, time: 0.064) G_GAN: 0.514 G_GAN_Feat: 0.894 G_ID: 0.218 G_Rec: 0.408 D_GP: 0.040 D_real: 1.118 D_fake: 0.491 +(epoch: 131, iters: 1760, time: 0.064) G_GAN: 0.349 G_GAN_Feat: 0.928 G_ID: 0.152 G_Rec: 0.353 D_GP: 0.123 D_real: 0.694 D_fake: 0.657 +(epoch: 131, iters: 2160, time: 0.064) G_GAN: 0.471 G_GAN_Feat: 0.981 G_ID: 0.201 G_Rec: 0.432 D_GP: 0.051 D_real: 0.885 D_fake: 0.551 +(epoch: 131, iters: 2560, time: 0.064) G_GAN: 0.360 G_GAN_Feat: 0.752 G_ID: 0.155 G_Rec: 0.312 D_GP: 0.030 D_real: 1.136 D_fake: 0.641 +(epoch: 131, iters: 2960, time: 0.064) G_GAN: 0.468 G_GAN_Feat: 0.993 G_ID: 0.232 G_Rec: 0.379 D_GP: 0.042 D_real: 0.918 D_fake: 0.532 +(epoch: 131, iters: 3360, time: 0.064) G_GAN: 0.026 G_GAN_Feat: 0.619 G_ID: 0.167 G_Rec: 0.317 D_GP: 0.023 D_real: 0.973 D_fake: 0.974 +(epoch: 131, iters: 3760, time: 0.064) G_GAN: 0.336 G_GAN_Feat: 0.817 G_ID: 0.202 G_Rec: 0.387 D_GP: 0.022 D_real: 1.136 D_fake: 0.670 +(epoch: 131, iters: 4160, time: 0.064) G_GAN: 0.099 G_GAN_Feat: 0.715 G_ID: 0.152 G_Rec: 0.363 D_GP: 0.043 D_real: 0.923 D_fake: 0.902 +(epoch: 131, iters: 4560, time: 0.064) G_GAN: 0.402 G_GAN_Feat: 0.888 G_ID: 0.242 G_Rec: 0.398 D_GP: 0.094 D_real: 0.911 D_fake: 0.620 +(epoch: 131, iters: 4960, time: 0.064) G_GAN: 0.169 G_GAN_Feat: 0.813 G_ID: 0.133 G_Rec: 0.329 D_GP: 0.092 D_real: 0.833 D_fake: 0.831 +(epoch: 131, iters: 5360, time: 0.064) G_GAN: 0.324 G_GAN_Feat: 0.893 G_ID: 0.220 G_Rec: 0.391 D_GP: 0.029 D_real: 0.944 D_fake: 0.676 +(epoch: 131, iters: 5760, time: 0.064) G_GAN: 0.012 G_GAN_Feat: 1.061 G_ID: 0.160 G_Rec: 0.432 D_GP: 1.313 D_real: 0.363 D_fake: 0.989 +(epoch: 131, iters: 6160, time: 0.064) G_GAN: 0.314 G_GAN_Feat: 0.970 G_ID: 0.250 G_Rec: 0.439 D_GP: 0.037 D_real: 0.775 D_fake: 0.692 +(epoch: 131, iters: 6560, time: 0.064) G_GAN: 0.321 G_GAN_Feat: 0.724 G_ID: 0.170 G_Rec: 0.315 D_GP: 0.035 D_real: 1.073 D_fake: 0.681 +(epoch: 131, iters: 6960, time: 0.064) G_GAN: 0.370 G_GAN_Feat: 0.990 G_ID: 0.268 G_Rec: 0.366 D_GP: 0.036 D_real: 0.901 D_fake: 0.631 +(epoch: 131, iters: 7360, time: 0.064) G_GAN: 0.514 G_GAN_Feat: 0.977 G_ID: 0.146 G_Rec: 0.355 D_GP: 0.198 D_real: 0.677 D_fake: 0.548 +(epoch: 131, iters: 7760, time: 0.064) G_GAN: 0.453 G_GAN_Feat: 1.001 G_ID: 0.209 G_Rec: 0.421 D_GP: 0.050 D_real: 0.809 D_fake: 0.552 +(epoch: 131, iters: 8160, time: 0.064) G_GAN: 0.190 G_GAN_Feat: 0.923 G_ID: 0.151 G_Rec: 0.355 D_GP: 0.030 D_real: 0.636 D_fake: 0.810 +(epoch: 131, iters: 8560, time: 0.064) G_GAN: 1.133 G_GAN_Feat: 0.864 G_ID: 0.241 G_Rec: 0.375 D_GP: 0.031 D_real: 1.804 D_fake: 0.135 +(epoch: 132, iters: 352, time: 0.064) G_GAN: 0.409 G_GAN_Feat: 0.928 G_ID: 0.141 G_Rec: 0.357 D_GP: 0.056 D_real: 0.756 D_fake: 0.592 +(epoch: 132, iters: 752, time: 0.064) G_GAN: 0.171 G_GAN_Feat: 0.833 G_ID: 0.259 G_Rec: 0.377 D_GP: 0.026 D_real: 0.975 D_fake: 0.829 +(epoch: 132, iters: 1152, time: 0.064) G_GAN: 0.397 G_GAN_Feat: 0.732 G_ID: 0.166 G_Rec: 0.325 D_GP: 0.024 D_real: 1.311 D_fake: 0.603 +(epoch: 132, iters: 1552, time: 0.064) G_GAN: 0.539 G_GAN_Feat: 0.877 G_ID: 0.189 G_Rec: 0.402 D_GP: 0.028 D_real: 1.208 D_fake: 0.463 +(epoch: 132, iters: 1952, time: 0.064) G_GAN: 0.153 G_GAN_Feat: 0.788 G_ID: 0.180 G_Rec: 0.329 D_GP: 0.043 D_real: 0.911 D_fake: 0.848 +(epoch: 132, iters: 2352, time: 0.064) G_GAN: 0.736 G_GAN_Feat: 1.027 G_ID: 0.257 G_Rec: 0.393 D_GP: 0.052 D_real: 0.881 D_fake: 0.357 +(epoch: 132, iters: 2752, time: 0.064) G_GAN: 0.283 G_GAN_Feat: 0.818 G_ID: 0.189 G_Rec: 0.365 D_GP: 0.025 D_real: 1.148 D_fake: 0.717 +(epoch: 132, iters: 3152, time: 0.064) G_GAN: 0.492 G_GAN_Feat: 0.850 G_ID: 0.210 G_Rec: 0.343 D_GP: 0.029 D_real: 1.214 D_fake: 0.510 +(epoch: 132, iters: 3552, time: 0.064) G_GAN: 0.653 G_GAN_Feat: 1.002 G_ID: 0.137 G_Rec: 0.334 D_GP: 0.104 D_real: 0.783 D_fake: 0.409 +(epoch: 132, iters: 3952, time: 0.064) G_GAN: 0.516 G_GAN_Feat: 0.706 G_ID: 0.207 G_Rec: 0.354 D_GP: 0.021 D_real: 1.307 D_fake: 0.494 +(epoch: 132, iters: 4352, time: 0.064) G_GAN: 0.045 G_GAN_Feat: 0.805 G_ID: 0.152 G_Rec: 0.376 D_GP: 0.069 D_real: 0.786 D_fake: 0.955 +(epoch: 132, iters: 4752, time: 0.064) G_GAN: -0.031 G_GAN_Feat: 1.121 G_ID: 0.195 G_Rec: 0.387 D_GP: 0.049 D_real: 1.077 D_fake: 1.032 +(epoch: 132, iters: 5152, time: 0.064) G_GAN: 0.230 G_GAN_Feat: 0.813 G_ID: 0.130 G_Rec: 0.347 D_GP: 0.029 D_real: 1.005 D_fake: 0.770 +(epoch: 132, iters: 5552, time: 0.064) G_GAN: 0.389 G_GAN_Feat: 1.189 G_ID: 0.228 G_Rec: 0.430 D_GP: 0.205 D_real: 0.479 D_fake: 0.613 +(epoch: 132, iters: 5952, time: 0.064) G_GAN: 0.433 G_GAN_Feat: 0.934 G_ID: 0.157 G_Rec: 0.341 D_GP: 0.059 D_real: 1.041 D_fake: 0.605 +(epoch: 132, iters: 6352, time: 0.064) G_GAN: 0.333 G_GAN_Feat: 0.883 G_ID: 0.214 G_Rec: 0.382 D_GP: 0.026 D_real: 1.092 D_fake: 0.670 +(epoch: 132, iters: 6752, time: 0.064) G_GAN: 0.327 G_GAN_Feat: 0.825 G_ID: 0.141 G_Rec: 0.344 D_GP: 0.033 D_real: 0.915 D_fake: 0.673 +(epoch: 132, iters: 7152, time: 0.064) G_GAN: 0.603 G_GAN_Feat: 0.852 G_ID: 0.212 G_Rec: 0.537 D_GP: 0.024 D_real: 1.328 D_fake: 0.404 +(epoch: 132, iters: 7552, time: 0.064) G_GAN: -0.014 G_GAN_Feat: 0.749 G_ID: 0.184 G_Rec: 0.317 D_GP: 0.036 D_real: 0.905 D_fake: 1.016 +(epoch: 132, iters: 7952, time: 0.064) G_GAN: 0.302 G_GAN_Feat: 0.853 G_ID: 0.214 G_Rec: 0.369 D_GP: 0.027 D_real: 1.024 D_fake: 0.698 +(epoch: 132, iters: 8352, time: 0.064) G_GAN: 0.207 G_GAN_Feat: 0.753 G_ID: 0.167 G_Rec: 0.357 D_GP: 0.025 D_real: 1.061 D_fake: 0.794 +(epoch: 133, iters: 144, time: 0.064) G_GAN: 0.394 G_GAN_Feat: 0.854 G_ID: 0.217 G_Rec: 0.370 D_GP: 0.027 D_real: 1.130 D_fake: 0.607 +(epoch: 133, iters: 544, time: 0.064) G_GAN: 0.440 G_GAN_Feat: 1.079 G_ID: 0.177 G_Rec: 0.374 D_GP: 1.310 D_real: 0.565 D_fake: 0.570 +(epoch: 133, iters: 944, time: 0.064) G_GAN: 0.426 G_GAN_Feat: 0.860 G_ID: 0.224 G_Rec: 0.436 D_GP: 0.031 D_real: 1.062 D_fake: 0.574 +(epoch: 133, iters: 1344, time: 0.064) G_GAN: 0.742 G_GAN_Feat: 0.850 G_ID: 0.159 G_Rec: 0.363 D_GP: 0.059 D_real: 1.310 D_fake: 0.307 +(epoch: 133, iters: 1744, time: 0.064) G_GAN: 0.259 G_GAN_Feat: 0.806 G_ID: 0.242 G_Rec: 0.386 D_GP: 0.021 D_real: 1.046 D_fake: 0.741 +(epoch: 133, iters: 2144, time: 0.064) G_GAN: 0.084 G_GAN_Feat: 0.702 G_ID: 0.154 G_Rec: 0.323 D_GP: 0.026 D_real: 0.943 D_fake: 0.916 +(epoch: 133, iters: 2544, time: 0.064) G_GAN: 0.667 G_GAN_Feat: 1.187 G_ID: 0.259 G_Rec: 0.431 D_GP: 0.405 D_real: 0.625 D_fake: 0.357 +(epoch: 133, iters: 2944, time: 0.064) G_GAN: 0.506 G_GAN_Feat: 0.816 G_ID: 0.145 G_Rec: 0.343 D_GP: 0.035 D_real: 1.183 D_fake: 0.499 +(epoch: 133, iters: 3344, time: 0.064) G_GAN: 0.132 G_GAN_Feat: 0.877 G_ID: 0.218 G_Rec: 0.393 D_GP: 0.078 D_real: 0.786 D_fake: 0.868 +(epoch: 133, iters: 3744, time: 0.064) G_GAN: 0.231 G_GAN_Feat: 0.718 G_ID: 0.186 G_Rec: 0.310 D_GP: 0.029 D_real: 1.051 D_fake: 0.769 +(epoch: 133, iters: 4144, time: 0.064) G_GAN: 0.606 G_GAN_Feat: 0.985 G_ID: 0.215 G_Rec: 0.409 D_GP: 0.046 D_real: 1.004 D_fake: 0.403 +(epoch: 133, iters: 4544, time: 0.064) G_GAN: 0.272 G_GAN_Feat: 0.941 G_ID: 0.171 G_Rec: 0.351 D_GP: 0.193 D_real: 0.682 D_fake: 0.729 +(epoch: 133, iters: 4944, time: 0.064) G_GAN: 0.465 G_GAN_Feat: 0.914 G_ID: 0.190 G_Rec: 0.387 D_GP: 0.034 D_real: 1.115 D_fake: 0.535 +(epoch: 133, iters: 5344, time: 0.064) G_GAN: 0.151 G_GAN_Feat: 0.873 G_ID: 0.153 G_Rec: 0.348 D_GP: 0.053 D_real: 0.723 D_fake: 0.849 +(epoch: 133, iters: 5744, time: 0.064) G_GAN: 0.479 G_GAN_Feat: 1.133 G_ID: 0.216 G_Rec: 0.449 D_GP: 0.402 D_real: 0.515 D_fake: 0.527 +(epoch: 133, iters: 6144, time: 0.064) G_GAN: 0.210 G_GAN_Feat: 0.857 G_ID: 0.157 G_Rec: 0.325 D_GP: 0.028 D_real: 0.906 D_fake: 0.792 +(epoch: 133, iters: 6544, time: 0.064) G_GAN: 0.543 G_GAN_Feat: 1.165 G_ID: 0.231 G_Rec: 0.409 D_GP: 0.070 D_real: 0.826 D_fake: 0.464 +(epoch: 133, iters: 6944, time: 0.064) G_GAN: 0.271 G_GAN_Feat: 0.735 G_ID: 0.158 G_Rec: 0.326 D_GP: 0.030 D_real: 1.225 D_fake: 0.731 +(epoch: 133, iters: 7344, time: 0.064) G_GAN: 0.428 G_GAN_Feat: 0.918 G_ID: 0.203 G_Rec: 0.379 D_GP: 0.034 D_real: 1.107 D_fake: 0.578 +(epoch: 133, iters: 7744, time: 0.064) G_GAN: 0.409 G_GAN_Feat: 0.813 G_ID: 0.159 G_Rec: 0.319 D_GP: 0.044 D_real: 1.082 D_fake: 0.593 +(epoch: 133, iters: 8144, time: 0.064) G_GAN: 0.524 G_GAN_Feat: 0.981 G_ID: 0.221 G_Rec: 0.409 D_GP: 0.048 D_real: 0.962 D_fake: 0.486 +(epoch: 133, iters: 8544, time: 0.064) G_GAN: 0.335 G_GAN_Feat: 0.852 G_ID: 0.158 G_Rec: 0.336 D_GP: 0.028 D_real: 1.046 D_fake: 0.667 +(epoch: 134, iters: 336, time: 0.064) G_GAN: 0.241 G_GAN_Feat: 1.112 G_ID: 0.223 G_Rec: 0.424 D_GP: 0.046 D_real: 0.419 D_fake: 0.761 +(epoch: 134, iters: 736, time: 0.064) G_GAN: 0.251 G_GAN_Feat: 0.931 G_ID: 0.149 G_Rec: 0.354 D_GP: 0.066 D_real: 0.606 D_fake: 0.752 +(epoch: 134, iters: 1136, time: 0.064) G_GAN: -0.017 G_GAN_Feat: 0.859 G_ID: 0.209 G_Rec: 0.396 D_GP: 0.030 D_real: 0.818 D_fake: 1.017 +(epoch: 134, iters: 1536, time: 0.064) G_GAN: 0.191 G_GAN_Feat: 0.707 G_ID: 0.131 G_Rec: 0.335 D_GP: 0.028 D_real: 1.054 D_fake: 0.809 +(epoch: 134, iters: 1936, time: 0.064) G_GAN: 0.728 G_GAN_Feat: 1.032 G_ID: 0.224 G_Rec: 0.433 D_GP: 0.051 D_real: 1.223 D_fake: 0.313 +(epoch: 134, iters: 2336, time: 0.064) G_GAN: 0.303 G_GAN_Feat: 0.901 G_ID: 0.159 G_Rec: 0.407 D_GP: 0.069 D_real: 0.787 D_fake: 0.698 +(epoch: 134, iters: 2736, time: 0.064) G_GAN: 0.644 G_GAN_Feat: 1.087 G_ID: 0.211 G_Rec: 0.416 D_GP: 0.068 D_real: 0.816 D_fake: 0.366 +(epoch: 134, iters: 3136, time: 0.064) G_GAN: 0.130 G_GAN_Feat: 0.763 G_ID: 0.142 G_Rec: 0.382 D_GP: 0.027 D_real: 0.959 D_fake: 0.870 +(epoch: 134, iters: 3536, time: 0.064) G_GAN: 0.358 G_GAN_Feat: 0.826 G_ID: 0.233 G_Rec: 0.351 D_GP: 0.027 D_real: 1.116 D_fake: 0.645 +(epoch: 134, iters: 3936, time: 0.064) G_GAN: 0.216 G_GAN_Feat: 1.020 G_ID: 0.154 G_Rec: 0.380 D_GP: 0.119 D_real: 0.494 D_fake: 0.787 +(epoch: 134, iters: 4336, time: 0.064) G_GAN: 0.373 G_GAN_Feat: 0.916 G_ID: 0.259 G_Rec: 0.378 D_GP: 0.027 D_real: 0.931 D_fake: 0.627 +(epoch: 134, iters: 4736, time: 0.064) G_GAN: 0.170 G_GAN_Feat: 1.045 G_ID: 0.175 G_Rec: 0.321 D_GP: 0.150 D_real: 0.363 D_fake: 0.830 +(epoch: 134, iters: 5136, time: 0.064) G_GAN: 0.252 G_GAN_Feat: 0.928 G_ID: 0.211 G_Rec: 0.426 D_GP: 0.050 D_real: 0.679 D_fake: 0.749 +(epoch: 134, iters: 5536, time: 0.064) G_GAN: 0.153 G_GAN_Feat: 0.818 G_ID: 0.140 G_Rec: 0.359 D_GP: 0.062 D_real: 0.731 D_fake: 0.848 +(epoch: 134, iters: 5936, time: 0.064) G_GAN: 0.308 G_GAN_Feat: 0.978 G_ID: 0.199 G_Rec: 0.430 D_GP: 0.037 D_real: 0.844 D_fake: 0.692 +(epoch: 134, iters: 6336, time: 0.064) G_GAN: 0.505 G_GAN_Feat: 0.779 G_ID: 0.130 G_Rec: 0.328 D_GP: 0.029 D_real: 1.305 D_fake: 0.501 +(epoch: 134, iters: 6736, time: 0.064) G_GAN: 0.438 G_GAN_Feat: 0.757 G_ID: 0.212 G_Rec: 0.390 D_GP: 0.021 D_real: 1.258 D_fake: 0.564 +(epoch: 134, iters: 7136, time: 0.064) G_GAN: 0.344 G_GAN_Feat: 0.932 G_ID: 0.189 G_Rec: 0.348 D_GP: 0.052 D_real: 0.613 D_fake: 0.658 +(epoch: 134, iters: 7536, time: 0.064) G_GAN: 0.429 G_GAN_Feat: 1.105 G_ID: 0.288 G_Rec: 0.411 D_GP: 0.112 D_real: 0.832 D_fake: 0.596 +(epoch: 134, iters: 7936, time: 0.064) G_GAN: 0.434 G_GAN_Feat: 0.812 G_ID: 0.162 G_Rec: 0.320 D_GP: 0.029 D_real: 1.189 D_fake: 0.566 +(epoch: 134, iters: 8336, time: 0.064) G_GAN: 0.518 G_GAN_Feat: 0.981 G_ID: 0.244 G_Rec: 0.431 D_GP: 0.059 D_real: 0.947 D_fake: 0.483 +(epoch: 135, iters: 128, time: 0.064) G_GAN: 0.238 G_GAN_Feat: 0.715 G_ID: 0.154 G_Rec: 0.322 D_GP: 0.026 D_real: 1.109 D_fake: 0.762 +(epoch: 135, iters: 528, time: 0.064) G_GAN: 0.392 G_GAN_Feat: 0.971 G_ID: 0.224 G_Rec: 0.381 D_GP: 0.031 D_real: 1.267 D_fake: 0.623 +(epoch: 135, iters: 928, time: 0.064) G_GAN: 0.255 G_GAN_Feat: 0.730 G_ID: 0.164 G_Rec: 0.307 D_GP: 0.032 D_real: 1.111 D_fake: 0.745 +(epoch: 135, iters: 1328, time: 0.064) G_GAN: 0.174 G_GAN_Feat: 0.830 G_ID: 0.246 G_Rec: 0.381 D_GP: 0.026 D_real: 0.839 D_fake: 0.826 +(epoch: 135, iters: 1728, time: 0.064) G_GAN: 0.515 G_GAN_Feat: 0.883 G_ID: 0.157 G_Rec: 0.347 D_GP: 0.085 D_real: 0.821 D_fake: 0.494 +(epoch: 135, iters: 2128, time: 0.064) G_GAN: 0.364 G_GAN_Feat: 0.853 G_ID: 0.212 G_Rec: 0.392 D_GP: 0.022 D_real: 1.112 D_fake: 0.636 +(epoch: 135, iters: 2528, time: 0.064) G_GAN: 0.234 G_GAN_Feat: 0.850 G_ID: 0.172 G_Rec: 0.320 D_GP: 0.037 D_real: 0.756 D_fake: 0.766 +(epoch: 135, iters: 2928, time: 0.064) G_GAN: 0.560 G_GAN_Feat: 0.930 G_ID: 0.224 G_Rec: 0.381 D_GP: 0.039 D_real: 1.047 D_fake: 0.447 +(epoch: 135, iters: 3328, time: 0.064) G_GAN: 0.127 G_GAN_Feat: 0.708 G_ID: 0.172 G_Rec: 0.328 D_GP: 0.028 D_real: 1.020 D_fake: 0.873 +(epoch: 135, iters: 3728, time: 0.064) G_GAN: 0.403 G_GAN_Feat: 1.139 G_ID: 0.203 G_Rec: 0.434 D_GP: 0.183 D_real: 0.453 D_fake: 0.598 +(epoch: 135, iters: 4128, time: 0.064) G_GAN: 0.126 G_GAN_Feat: 0.758 G_ID: 0.161 G_Rec: 0.303 D_GP: 0.036 D_real: 1.082 D_fake: 0.874 +(epoch: 135, iters: 4528, time: 0.064) G_GAN: 0.468 G_GAN_Feat: 1.092 G_ID: 0.202 G_Rec: 0.416 D_GP: 0.143 D_real: 0.882 D_fake: 0.642 +(epoch: 135, iters: 4928, time: 0.064) G_GAN: 0.203 G_GAN_Feat: 0.708 G_ID: 0.152 G_Rec: 0.342 D_GP: 0.021 D_real: 1.143 D_fake: 0.798 +(epoch: 135, iters: 5328, time: 0.064) G_GAN: 0.453 G_GAN_Feat: 0.792 G_ID: 0.231 G_Rec: 0.351 D_GP: 0.025 D_real: 1.222 D_fake: 0.553 +(epoch: 135, iters: 5728, time: 0.064) G_GAN: 0.289 G_GAN_Feat: 0.686 G_ID: 0.156 G_Rec: 0.321 D_GP: 0.025 D_real: 1.194 D_fake: 0.711 +(epoch: 135, iters: 6128, time: 0.064) G_GAN: 0.565 G_GAN_Feat: 1.040 G_ID: 0.245 G_Rec: 0.430 D_GP: 0.041 D_real: 1.050 D_fake: 0.452 +(epoch: 135, iters: 6528, time: 0.064) G_GAN: 0.215 G_GAN_Feat: 0.934 G_ID: 0.135 G_Rec: 0.301 D_GP: 0.045 D_real: 0.626 D_fake: 0.786 +(epoch: 135, iters: 6928, time: 0.064) G_GAN: 0.616 G_GAN_Feat: 0.891 G_ID: 0.200 G_Rec: 0.385 D_GP: 0.035 D_real: 1.247 D_fake: 0.409 +(epoch: 135, iters: 7328, time: 0.064) G_GAN: 0.322 G_GAN_Feat: 0.910 G_ID: 0.139 G_Rec: 0.352 D_GP: 0.069 D_real: 0.896 D_fake: 0.688 +(epoch: 135, iters: 7728, time: 0.064) G_GAN: 0.216 G_GAN_Feat: 0.891 G_ID: 0.251 G_Rec: 0.383 D_GP: 0.032 D_real: 0.779 D_fake: 0.787 +(epoch: 135, iters: 8128, time: 0.064) G_GAN: 0.224 G_GAN_Feat: 0.853 G_ID: 0.144 G_Rec: 0.326 D_GP: 0.034 D_real: 0.938 D_fake: 0.776 +(epoch: 135, iters: 8528, time: 0.064) G_GAN: 0.357 G_GAN_Feat: 1.028 G_ID: 0.221 G_Rec: 0.442 D_GP: 0.056 D_real: 0.899 D_fake: 0.647 +(epoch: 136, iters: 320, time: 0.064) G_GAN: 0.180 G_GAN_Feat: 0.836 G_ID: 0.141 G_Rec: 0.326 D_GP: 0.029 D_real: 0.817 D_fake: 0.821 +(epoch: 136, iters: 720, time: 0.064) G_GAN: 0.514 G_GAN_Feat: 0.838 G_ID: 0.205 G_Rec: 0.408 D_GP: 0.024 D_real: 1.227 D_fake: 0.490 +(epoch: 136, iters: 1120, time: 0.064) G_GAN: 0.071 G_GAN_Feat: 0.929 G_ID: 0.138 G_Rec: 0.429 D_GP: 0.136 D_real: 0.402 D_fake: 0.929 +(epoch: 136, iters: 1520, time: 0.064) G_GAN: 0.465 G_GAN_Feat: 1.096 G_ID: 0.204 G_Rec: 0.405 D_GP: 0.132 D_real: 0.459 D_fake: 0.541 +(epoch: 136, iters: 1920, time: 0.064) G_GAN: 0.010 G_GAN_Feat: 0.665 G_ID: 0.154 G_Rec: 0.299 D_GP: 0.027 D_real: 0.928 D_fake: 0.990 +(epoch: 136, iters: 2320, time: 0.064) G_GAN: 0.559 G_GAN_Feat: 0.963 G_ID: 0.221 G_Rec: 0.416 D_GP: 0.043 D_real: 1.027 D_fake: 0.447 +(epoch: 136, iters: 2720, time: 0.064) G_GAN: 0.446 G_GAN_Feat: 0.726 G_ID: 0.140 G_Rec: 0.325 D_GP: 0.026 D_real: 1.305 D_fake: 0.554 +(epoch: 136, iters: 3120, time: 0.064) G_GAN: 0.466 G_GAN_Feat: 0.886 G_ID: 0.218 G_Rec: 0.377 D_GP: 0.040 D_real: 1.111 D_fake: 0.544 +(epoch: 136, iters: 3520, time: 0.064) G_GAN: 0.270 G_GAN_Feat: 0.675 G_ID: 0.154 G_Rec: 0.333 D_GP: 0.022 D_real: 1.258 D_fake: 0.731 +(epoch: 136, iters: 3920, time: 0.064) G_GAN: 0.200 G_GAN_Feat: 0.793 G_ID: 0.224 G_Rec: 0.394 D_GP: 0.023 D_real: 0.916 D_fake: 0.801 +(epoch: 136, iters: 4320, time: 0.064) G_GAN: 0.119 G_GAN_Feat: 0.712 G_ID: 0.161 G_Rec: 0.414 D_GP: 0.049 D_real: 0.918 D_fake: 0.881 +(epoch: 136, iters: 4720, time: 0.064) G_GAN: 0.106 G_GAN_Feat: 0.867 G_ID: 0.251 G_Rec: 0.420 D_GP: 0.036 D_real: 0.743 D_fake: 0.894 +(epoch: 136, iters: 5120, time: 0.064) G_GAN: 0.213 G_GAN_Feat: 0.699 G_ID: 0.161 G_Rec: 0.351 D_GP: 0.028 D_real: 1.021 D_fake: 0.787 +(epoch: 136, iters: 5520, time: 0.064) G_GAN: 0.542 G_GAN_Feat: 1.095 G_ID: 0.251 G_Rec: 0.461 D_GP: 0.131 D_real: 1.025 D_fake: 0.478 +(epoch: 136, iters: 5920, time: 0.064) G_GAN: 0.095 G_GAN_Feat: 0.953 G_ID: 0.168 G_Rec: 0.362 D_GP: 0.225 D_real: 0.603 D_fake: 0.906 +(epoch: 136, iters: 6320, time: 0.064) G_GAN: 0.235 G_GAN_Feat: 0.828 G_ID: 0.246 G_Rec: 0.340 D_GP: 0.049 D_real: 0.899 D_fake: 0.765 +(epoch: 136, iters: 6720, time: 0.064) G_GAN: 0.575 G_GAN_Feat: 0.834 G_ID: 0.157 G_Rec: 0.369 D_GP: 0.051 D_real: 1.159 D_fake: 0.452 +(epoch: 136, iters: 7120, time: 0.064) G_GAN: 0.803 G_GAN_Feat: 0.902 G_ID: 0.220 G_Rec: 0.399 D_GP: 0.033 D_real: 1.396 D_fake: 0.316 +(epoch: 136, iters: 7520, time: 0.064) G_GAN: 0.053 G_GAN_Feat: 0.726 G_ID: 0.174 G_Rec: 0.302 D_GP: 0.035 D_real: 0.891 D_fake: 0.947 +(epoch: 136, iters: 7920, time: 0.064) G_GAN: 0.359 G_GAN_Feat: 0.912 G_ID: 0.223 G_Rec: 0.393 D_GP: 0.049 D_real: 0.948 D_fake: 0.643 +(epoch: 136, iters: 8320, time: 0.064) G_GAN: 0.449 G_GAN_Feat: 0.707 G_ID: 0.179 G_Rec: 0.340 D_GP: 0.023 D_real: 1.374 D_fake: 0.555 +(epoch: 137, iters: 112, time: 0.064) G_GAN: 0.393 G_GAN_Feat: 0.946 G_ID: 0.222 G_Rec: 0.402 D_GP: 0.030 D_real: 1.026 D_fake: 0.607 +(epoch: 137, iters: 512, time: 0.064) G_GAN: 0.328 G_GAN_Feat: 1.035 G_ID: 0.178 G_Rec: 0.354 D_GP: 0.191 D_real: 0.446 D_fake: 0.675 +(epoch: 137, iters: 912, time: 0.064) G_GAN: 0.755 G_GAN_Feat: 1.053 G_ID: 0.244 G_Rec: 0.427 D_GP: 0.040 D_real: 1.309 D_fake: 0.348 +(epoch: 137, iters: 1312, time: 0.064) G_GAN: 0.232 G_GAN_Feat: 0.859 G_ID: 0.170 G_Rec: 0.319 D_GP: 0.063 D_real: 0.724 D_fake: 0.768 +(epoch: 137, iters: 1712, time: 0.064) G_GAN: 0.377 G_GAN_Feat: 1.009 G_ID: 0.206 G_Rec: 0.395 D_GP: 0.064 D_real: 0.669 D_fake: 0.624 +(epoch: 137, iters: 2112, time: 0.064) G_GAN: 0.090 G_GAN_Feat: 0.841 G_ID: 0.191 G_Rec: 0.365 D_GP: 0.037 D_real: 0.719 D_fake: 0.910 +(epoch: 137, iters: 2512, time: 0.064) G_GAN: 0.247 G_GAN_Feat: 1.115 G_ID: 0.270 G_Rec: 0.424 D_GP: 0.272 D_real: 0.387 D_fake: 0.771 +(epoch: 137, iters: 2912, time: 0.064) G_GAN: 0.454 G_GAN_Feat: 0.818 G_ID: 0.150 G_Rec: 0.298 D_GP: 0.082 D_real: 1.065 D_fake: 0.562 +(epoch: 137, iters: 3312, time: 0.064) G_GAN: 0.474 G_GAN_Feat: 0.804 G_ID: 0.247 G_Rec: 0.392 D_GP: 0.024 D_real: 1.250 D_fake: 0.527 +(epoch: 137, iters: 3712, time: 0.064) G_GAN: 0.083 G_GAN_Feat: 0.799 G_ID: 0.140 G_Rec: 0.339 D_GP: 0.046 D_real: 0.788 D_fake: 0.917 +(epoch: 137, iters: 4112, time: 0.064) G_GAN: 0.786 G_GAN_Feat: 0.931 G_ID: 0.192 G_Rec: 0.401 D_GP: 0.028 D_real: 1.435 D_fake: 0.244 +(epoch: 137, iters: 4512, time: 0.064) G_GAN: 0.422 G_GAN_Feat: 0.827 G_ID: 0.181 G_Rec: 0.345 D_GP: 0.045 D_real: 1.082 D_fake: 0.580 +(epoch: 137, iters: 4912, time: 0.064) G_GAN: 0.485 G_GAN_Feat: 1.103 G_ID: 0.237 G_Rec: 0.421 D_GP: 0.332 D_real: 0.467 D_fake: 0.543 +(epoch: 137, iters: 5312, time: 0.064) G_GAN: 0.339 G_GAN_Feat: 0.676 G_ID: 0.141 G_Rec: 0.328 D_GP: 0.024 D_real: 1.245 D_fake: 0.661 +(epoch: 137, iters: 5712, time: 0.064) G_GAN: 0.258 G_GAN_Feat: 0.899 G_ID: 0.199 G_Rec: 0.374 D_GP: 0.065 D_real: 0.849 D_fake: 0.752 +(epoch: 137, iters: 6112, time: 0.064) G_GAN: 0.161 G_GAN_Feat: 0.816 G_ID: 0.123 G_Rec: 0.340 D_GP: 0.054 D_real: 0.779 D_fake: 0.840 +(epoch: 137, iters: 6512, time: 0.064) G_GAN: 0.564 G_GAN_Feat: 0.937 G_ID: 0.196 G_Rec: 0.400 D_GP: 0.038 D_real: 1.159 D_fake: 0.448 +(epoch: 137, iters: 6912, time: 0.064) G_GAN: 0.234 G_GAN_Feat: 0.772 G_ID: 0.194 G_Rec: 0.340 D_GP: 0.035 D_real: 1.127 D_fake: 0.766 +(epoch: 137, iters: 7312, time: 0.064) G_GAN: 0.538 G_GAN_Feat: 0.883 G_ID: 0.197 G_Rec: 0.408 D_GP: 0.023 D_real: 1.303 D_fake: 0.471 +(epoch: 137, iters: 7712, time: 0.064) G_GAN: 0.416 G_GAN_Feat: 1.032 G_ID: 0.205 G_Rec: 0.384 D_GP: 0.381 D_real: 0.346 D_fake: 0.641 +(epoch: 137, iters: 8112, time: 0.064) G_GAN: 0.136 G_GAN_Feat: 0.785 G_ID: 0.214 G_Rec: 0.379 D_GP: 0.023 D_real: 0.929 D_fake: 0.864 +(epoch: 137, iters: 8512, time: 0.064) G_GAN: 0.344 G_GAN_Feat: 0.678 G_ID: 0.196 G_Rec: 0.330 D_GP: 0.023 D_real: 1.171 D_fake: 0.657 +(epoch: 138, iters: 304, time: 0.064) G_GAN: 0.302 G_GAN_Feat: 0.909 G_ID: 0.254 G_Rec: 0.418 D_GP: 0.029 D_real: 0.996 D_fake: 0.700 +(epoch: 138, iters: 704, time: 0.064) G_GAN: 0.211 G_GAN_Feat: 0.671 G_ID: 0.160 G_Rec: 0.344 D_GP: 0.023 D_real: 1.128 D_fake: 0.789 +(epoch: 138, iters: 1104, time: 0.064) G_GAN: 0.587 G_GAN_Feat: 1.110 G_ID: 0.201 G_Rec: 0.417 D_GP: 0.155 D_real: 0.622 D_fake: 0.423 +(epoch: 138, iters: 1504, time: 0.064) G_GAN: 0.292 G_GAN_Feat: 0.911 G_ID: 0.171 G_Rec: 0.353 D_GP: 0.090 D_real: 1.020 D_fake: 0.709 +(epoch: 138, iters: 1904, time: 0.064) G_GAN: 0.461 G_GAN_Feat: 0.990 G_ID: 0.205 G_Rec: 0.423 D_GP: 0.034 D_real: 1.069 D_fake: 0.540 +(epoch: 138, iters: 2304, time: 0.064) G_GAN: 0.354 G_GAN_Feat: 0.699 G_ID: 0.142 G_Rec: 0.322 D_GP: 0.027 D_real: 1.318 D_fake: 0.655 +(epoch: 138, iters: 2704, time: 0.064) G_GAN: 0.457 G_GAN_Feat: 1.059 G_ID: 0.214 G_Rec: 0.447 D_GP: 0.037 D_real: 0.804 D_fake: 0.545 +(epoch: 138, iters: 3104, time: 0.064) G_GAN: 0.525 G_GAN_Feat: 0.795 G_ID: 0.141 G_Rec: 0.310 D_GP: 0.046 D_real: 1.201 D_fake: 0.485 +(epoch: 138, iters: 3504, time: 0.064) G_GAN: 0.071 G_GAN_Feat: 0.850 G_ID: 0.248 G_Rec: 0.402 D_GP: 0.025 D_real: 0.851 D_fake: 0.929 +(epoch: 138, iters: 3904, time: 0.064) G_GAN: 0.361 G_GAN_Feat: 0.908 G_ID: 0.151 G_Rec: 0.345 D_GP: 0.074 D_real: 0.864 D_fake: 0.647 +(epoch: 138, iters: 4304, time: 0.064) G_GAN: 0.440 G_GAN_Feat: 0.810 G_ID: 0.254 G_Rec: 0.390 D_GP: 0.027 D_real: 1.202 D_fake: 0.562 +(epoch: 138, iters: 4704, time: 0.064) G_GAN: 0.077 G_GAN_Feat: 0.678 G_ID: 0.143 G_Rec: 0.351 D_GP: 0.029 D_real: 0.925 D_fake: 0.923 +(epoch: 138, iters: 5104, time: 0.064) G_GAN: 0.641 G_GAN_Feat: 0.942 G_ID: 0.204 G_Rec: 0.425 D_GP: 0.031 D_real: 1.185 D_fake: 0.384 +(epoch: 138, iters: 5504, time: 0.064) G_GAN: 0.048 G_GAN_Feat: 1.229 G_ID: 0.176 G_Rec: 0.388 D_GP: 0.662 D_real: 0.309 D_fake: 0.953 +(epoch: 138, iters: 5904, time: 0.064) G_GAN: 0.390 G_GAN_Feat: 0.963 G_ID: 0.194 G_Rec: 0.406 D_GP: 0.036 D_real: 0.918 D_fake: 0.611 +(epoch: 138, iters: 6304, time: 0.064) G_GAN: 0.140 G_GAN_Feat: 0.861 G_ID: 0.184 G_Rec: 0.327 D_GP: 0.078 D_real: 0.826 D_fake: 0.896 +(epoch: 138, iters: 6704, time: 0.064) G_GAN: 0.363 G_GAN_Feat: 0.871 G_ID: 0.234 G_Rec: 0.412 D_GP: 0.029 D_real: 1.032 D_fake: 0.646 +(epoch: 138, iters: 7104, time: 0.064) G_GAN: 0.367 G_GAN_Feat: 1.124 G_ID: 0.164 G_Rec: 0.367 D_GP: 0.288 D_real: 0.476 D_fake: 0.639 +(epoch: 138, iters: 7504, time: 0.064) G_GAN: 0.414 G_GAN_Feat: 0.923 G_ID: 0.216 G_Rec: 0.400 D_GP: 0.038 D_real: 0.923 D_fake: 0.590 +(epoch: 138, iters: 7904, time: 0.064) G_GAN: 0.350 G_GAN_Feat: 1.039 G_ID: 0.154 G_Rec: 0.383 D_GP: 0.876 D_real: 0.382 D_fake: 0.665 +(epoch: 138, iters: 8304, time: 0.064) G_GAN: 0.431 G_GAN_Feat: 0.890 G_ID: 0.223 G_Rec: 0.390 D_GP: 0.034 D_real: 1.006 D_fake: 0.570 +(epoch: 139, iters: 96, time: 0.064) G_GAN: 0.183 G_GAN_Feat: 0.875 G_ID: 0.158 G_Rec: 0.321 D_GP: 0.115 D_real: 0.683 D_fake: 0.817 +(epoch: 139, iters: 496, time: 0.064) G_GAN: 0.382 G_GAN_Feat: 0.801 G_ID: 0.202 G_Rec: 0.378 D_GP: 0.025 D_real: 1.137 D_fake: 0.627 +(epoch: 139, iters: 896, time: 0.064) G_GAN: 0.174 G_GAN_Feat: 0.812 G_ID: 0.137 G_Rec: 0.317 D_GP: 0.050 D_real: 0.827 D_fake: 0.826 +(epoch: 139, iters: 1296, time: 0.064) G_GAN: 0.511 G_GAN_Feat: 0.846 G_ID: 0.244 G_Rec: 0.397 D_GP: 0.026 D_real: 1.175 D_fake: 0.504 +(epoch: 139, iters: 1696, time: 0.064) G_GAN: 0.226 G_GAN_Feat: 0.776 G_ID: 0.160 G_Rec: 0.374 D_GP: 0.023 D_real: 1.122 D_fake: 0.774 +(epoch: 139, iters: 2096, time: 0.064) G_GAN: 0.278 G_GAN_Feat: 0.927 G_ID: 0.276 G_Rec: 0.411 D_GP: 0.029 D_real: 0.874 D_fake: 0.724 +(epoch: 139, iters: 2496, time: 0.064) G_GAN: 0.295 G_GAN_Feat: 0.654 G_ID: 0.148 G_Rec: 0.296 D_GP: 0.027 D_real: 1.147 D_fake: 0.707 +(epoch: 139, iters: 2896, time: 0.064) G_GAN: 0.179 G_GAN_Feat: 0.833 G_ID: 0.224 G_Rec: 0.380 D_GP: 0.036 D_real: 0.917 D_fake: 0.821 +(epoch: 139, iters: 3296, time: 0.064) G_GAN: 0.185 G_GAN_Feat: 0.695 G_ID: 0.182 G_Rec: 0.371 D_GP: 0.026 D_real: 1.048 D_fake: 0.816 +(epoch: 139, iters: 3696, time: 0.064) G_GAN: 0.527 G_GAN_Feat: 0.954 G_ID: 0.203 G_Rec: 0.401 D_GP: 0.048 D_real: 0.986 D_fake: 0.482 +(epoch: 139, iters: 4096, time: 0.064) G_GAN: 0.219 G_GAN_Feat: 0.707 G_ID: 0.151 G_Rec: 0.442 D_GP: 0.027 D_real: 1.083 D_fake: 0.782 +(epoch: 139, iters: 4496, time: 0.064) G_GAN: 0.388 G_GAN_Feat: 1.039 G_ID: 0.210 G_Rec: 0.419 D_GP: 0.031 D_real: 0.968 D_fake: 0.612 +(epoch: 139, iters: 4896, time: 0.064) G_GAN: 0.393 G_GAN_Feat: 0.763 G_ID: 0.153 G_Rec: 0.362 D_GP: 0.032 D_real: 1.219 D_fake: 0.612 +(epoch: 139, iters: 5296, time: 0.064) G_GAN: 0.461 G_GAN_Feat: 0.803 G_ID: 0.209 G_Rec: 0.397 D_GP: 0.022 D_real: 1.285 D_fake: 0.547 +(epoch: 139, iters: 5696, time: 0.064) G_GAN: 0.347 G_GAN_Feat: 0.882 G_ID: 0.135 G_Rec: 0.345 D_GP: 0.048 D_real: 0.964 D_fake: 0.657 +(epoch: 139, iters: 6096, time: 0.064) G_GAN: 0.768 G_GAN_Feat: 0.943 G_ID: 0.188 G_Rec: 0.406 D_GP: 0.034 D_real: 1.235 D_fake: 0.272 +(epoch: 139, iters: 6496, time: 0.064) G_GAN: -0.027 G_GAN_Feat: 0.975 G_ID: 0.159 G_Rec: 0.327 D_GP: 0.079 D_real: 1.021 D_fake: 1.028 +(epoch: 139, iters: 6896, time: 0.064) G_GAN: 0.456 G_GAN_Feat: 1.193 G_ID: 0.221 G_Rec: 0.488 D_GP: 0.063 D_real: 0.684 D_fake: 0.552 +(epoch: 139, iters: 7296, time: 0.064) G_GAN: -0.019 G_GAN_Feat: 0.874 G_ID: 0.163 G_Rec: 0.317 D_GP: 0.079 D_real: 0.423 D_fake: 1.019 +(epoch: 139, iters: 7696, time: 0.064) G_GAN: 0.661 G_GAN_Feat: 0.952 G_ID: 0.217 G_Rec: 0.396 D_GP: 0.034 D_real: 1.306 D_fake: 0.345 +(epoch: 139, iters: 8096, time: 0.064) G_GAN: 0.187 G_GAN_Feat: 0.771 G_ID: 0.175 G_Rec: 0.343 D_GP: 0.044 D_real: 0.952 D_fake: 0.813 +(epoch: 139, iters: 8496, time: 0.064) G_GAN: 0.658 G_GAN_Feat: 0.881 G_ID: 0.275 G_Rec: 0.408 D_GP: 0.024 D_real: 1.409 D_fake: 0.383 +(epoch: 140, iters: 288, time: 0.064) G_GAN: 0.093 G_GAN_Feat: 0.905 G_ID: 0.163 G_Rec: 0.355 D_GP: 0.060 D_real: 0.546 D_fake: 0.910 +(epoch: 140, iters: 688, time: 0.064) G_GAN: 0.399 G_GAN_Feat: 0.789 G_ID: 0.226 G_Rec: 0.404 D_GP: 0.023 D_real: 1.138 D_fake: 0.604 +(epoch: 140, iters: 1088, time: 0.064) G_GAN: 0.209 G_GAN_Feat: 0.565 G_ID: 0.142 G_Rec: 0.289 D_GP: 0.021 D_real: 1.171 D_fake: 0.791 +(epoch: 140, iters: 1488, time: 0.064) G_GAN: 0.365 G_GAN_Feat: 0.792 G_ID: 0.239 G_Rec: 0.417 D_GP: 0.028 D_real: 1.104 D_fake: 0.636 +(epoch: 140, iters: 1888, time: 0.064) G_GAN: 0.239 G_GAN_Feat: 0.640 G_ID: 0.147 G_Rec: 0.346 D_GP: 0.025 D_real: 1.090 D_fake: 0.761 +(epoch: 140, iters: 2288, time: 0.064) G_GAN: 0.141 G_GAN_Feat: 0.895 G_ID: 0.222 G_Rec: 0.457 D_GP: 0.047 D_real: 0.779 D_fake: 0.859 +(epoch: 140, iters: 2688, time: 0.064) G_GAN: -0.026 G_GAN_Feat: 0.741 G_ID: 0.152 G_Rec: 0.328 D_GP: 0.034 D_real: 0.932 D_fake: 1.026 +(epoch: 140, iters: 3088, time: 0.064) G_GAN: 0.386 G_GAN_Feat: 0.940 G_ID: 0.227 G_Rec: 0.424 D_GP: 0.035 D_real: 1.047 D_fake: 0.614 +(epoch: 140, iters: 3488, time: 0.064) G_GAN: 0.466 G_GAN_Feat: 0.879 G_ID: 0.161 G_Rec: 0.345 D_GP: 0.036 D_real: 0.974 D_fake: 0.543 +(epoch: 140, iters: 3888, time: 0.064) G_GAN: 0.355 G_GAN_Feat: 0.993 G_ID: 0.230 G_Rec: 0.404 D_GP: 0.045 D_real: 0.722 D_fake: 0.647 +(epoch: 140, iters: 4288, time: 0.064) G_GAN: 0.200 G_GAN_Feat: 0.803 G_ID: 0.164 G_Rec: 0.326 D_GP: 0.031 D_real: 0.948 D_fake: 0.800 +(epoch: 140, iters: 4688, time: 0.064) G_GAN: 0.538 G_GAN_Feat: 1.266 G_ID: 0.172 G_Rec: 0.461 D_GP: 0.165 D_real: 0.564 D_fake: 0.492 +(epoch: 140, iters: 5088, time: 0.064) G_GAN: 0.347 G_GAN_Feat: 0.786 G_ID: 0.134 G_Rec: 0.310 D_GP: 0.038 D_real: 1.075 D_fake: 0.654 +(epoch: 140, iters: 5488, time: 0.064) G_GAN: 0.157 G_GAN_Feat: 0.762 G_ID: 0.216 G_Rec: 0.391 D_GP: 0.024 D_real: 0.976 D_fake: 0.843 +(epoch: 140, iters: 5888, time: 0.064) G_GAN: 0.198 G_GAN_Feat: 0.735 G_ID: 0.159 G_Rec: 0.351 D_GP: 0.030 D_real: 1.056 D_fake: 0.803 +(epoch: 140, iters: 6288, time: 0.064) G_GAN: 0.088 G_GAN_Feat: 0.911 G_ID: 0.234 G_Rec: 0.397 D_GP: 0.061 D_real: 0.695 D_fake: 0.913 +(epoch: 140, iters: 6688, time: 0.064) G_GAN: 0.159 G_GAN_Feat: 0.786 G_ID: 0.192 G_Rec: 0.318 D_GP: 0.034 D_real: 1.113 D_fake: 0.841 +(epoch: 140, iters: 7088, time: 0.064) G_GAN: 0.529 G_GAN_Feat: 1.075 G_ID: 0.223 G_Rec: 0.431 D_GP: 0.141 D_real: 0.578 D_fake: 0.475 +(epoch: 140, iters: 7488, time: 0.064) G_GAN: 0.192 G_GAN_Feat: 0.701 G_ID: 0.135 G_Rec: 0.303 D_GP: 0.027 D_real: 1.073 D_fake: 0.808 +(epoch: 140, iters: 7888, time: 0.064) G_GAN: 0.498 G_GAN_Feat: 0.999 G_ID: 0.225 G_Rec: 0.381 D_GP: 0.089 D_real: 0.947 D_fake: 0.518 +(epoch: 140, iters: 8288, time: 0.064) G_GAN: 0.249 G_GAN_Feat: 0.698 G_ID: 0.159 G_Rec: 0.324 D_GP: 0.025 D_real: 1.181 D_fake: 0.751 +(epoch: 141, iters: 80, time: 0.064) G_GAN: 0.383 G_GAN_Feat: 1.028 G_ID: 0.212 G_Rec: 0.439 D_GP: 0.031 D_real: 0.857 D_fake: 0.621 +(epoch: 141, iters: 480, time: 0.064) G_GAN: 0.189 G_GAN_Feat: 0.778 G_ID: 0.178 G_Rec: 0.334 D_GP: 0.031 D_real: 0.982 D_fake: 0.811 +(epoch: 141, iters: 880, time: 0.064) G_GAN: 0.482 G_GAN_Feat: 0.822 G_ID: 0.213 G_Rec: 0.376 D_GP: 0.025 D_real: 1.137 D_fake: 0.521 +(epoch: 141, iters: 1280, time: 0.064) G_GAN: 0.189 G_GAN_Feat: 0.684 G_ID: 0.158 G_Rec: 0.301 D_GP: 0.024 D_real: 1.064 D_fake: 0.812 +(epoch: 141, iters: 1680, time: 0.064) G_GAN: 0.383 G_GAN_Feat: 0.832 G_ID: 0.211 G_Rec: 0.356 D_GP: 0.030 D_real: 1.199 D_fake: 0.620 +(epoch: 141, iters: 2080, time: 0.064) G_GAN: 0.225 G_GAN_Feat: 0.844 G_ID: 0.170 G_Rec: 0.379 D_GP: 0.055 D_real: 0.768 D_fake: 0.776 +(epoch: 141, iters: 2480, time: 0.064) G_GAN: 0.645 G_GAN_Feat: 1.002 G_ID: 0.224 G_Rec: 0.408 D_GP: 0.045 D_real: 1.097 D_fake: 0.392 +(epoch: 141, iters: 2880, time: 0.064) G_GAN: 0.075 G_GAN_Feat: 0.689 G_ID: 0.154 G_Rec: 0.352 D_GP: 0.030 D_real: 0.915 D_fake: 0.926 +(epoch: 141, iters: 3280, time: 0.064) G_GAN: 0.447 G_GAN_Feat: 0.895 G_ID: 0.218 G_Rec: 0.434 D_GP: 0.033 D_real: 1.130 D_fake: 0.555 +(epoch: 141, iters: 3680, time: 0.064) G_GAN: 0.196 G_GAN_Feat: 0.620 G_ID: 0.152 G_Rec: 0.336 D_GP: 0.030 D_real: 1.119 D_fake: 0.804 +(epoch: 141, iters: 4080, time: 0.064) G_GAN: 0.185 G_GAN_Feat: 0.853 G_ID: 0.223 G_Rec: 0.402 D_GP: 0.053 D_real: 0.782 D_fake: 0.815 +(epoch: 141, iters: 4480, time: 0.064) G_GAN: 0.145 G_GAN_Feat: 0.866 G_ID: 0.153 G_Rec: 0.416 D_GP: 0.105 D_real: 0.519 D_fake: 0.855 +(epoch: 141, iters: 4880, time: 0.064) G_GAN: 0.574 G_GAN_Feat: 0.836 G_ID: 0.232 G_Rec: 0.371 D_GP: 0.027 D_real: 1.201 D_fake: 0.442 +(epoch: 141, iters: 5280, time: 0.064) G_GAN: 0.223 G_GAN_Feat: 0.759 G_ID: 0.136 G_Rec: 0.314 D_GP: 0.048 D_real: 0.979 D_fake: 0.779 +(epoch: 141, iters: 5680, time: 0.064) G_GAN: 0.441 G_GAN_Feat: 1.018 G_ID: 0.216 G_Rec: 0.442 D_GP: 0.100 D_real: 0.556 D_fake: 0.584 +(epoch: 141, iters: 6080, time: 0.064) G_GAN: 0.169 G_GAN_Feat: 0.855 G_ID: 0.147 G_Rec: 0.406 D_GP: 0.291 D_real: 0.801 D_fake: 0.833 +(epoch: 141, iters: 6480, time: 0.064) G_GAN: 0.393 G_GAN_Feat: 0.987 G_ID: 0.217 G_Rec: 0.431 D_GP: 0.042 D_real: 0.881 D_fake: 0.608 +(epoch: 141, iters: 6880, time: 0.064) G_GAN: 0.003 G_GAN_Feat: 0.928 G_ID: 0.159 G_Rec: 0.357 D_GP: 0.131 D_real: 0.391 D_fake: 0.997 +(epoch: 141, iters: 7280, time: 0.064) G_GAN: 0.311 G_GAN_Feat: 0.985 G_ID: 0.247 G_Rec: 0.404 D_GP: 0.032 D_real: 0.808 D_fake: 0.690 +(epoch: 141, iters: 7680, time: 0.064) G_GAN: 0.421 G_GAN_Feat: 0.936 G_ID: 0.187 G_Rec: 0.345 D_GP: 0.065 D_real: 0.739 D_fake: 0.584 +(epoch: 141, iters: 8080, time: 0.064) G_GAN: 0.555 G_GAN_Feat: 0.943 G_ID: 0.214 G_Rec: 0.386 D_GP: 0.028 D_real: 1.162 D_fake: 0.454 +(epoch: 141, iters: 8480, time: 0.064) G_GAN: 0.501 G_GAN_Feat: 0.703 G_ID: 0.147 G_Rec: 0.361 D_GP: 0.020 D_real: 1.413 D_fake: 0.509 +(epoch: 142, iters: 272, time: 0.064) G_GAN: 0.301 G_GAN_Feat: 1.042 G_ID: 0.189 G_Rec: 0.431 D_GP: 0.100 D_real: 0.706 D_fake: 0.705 +(epoch: 142, iters: 672, time: 0.064) G_GAN: 0.284 G_GAN_Feat: 0.809 G_ID: 0.160 G_Rec: 0.356 D_GP: 0.028 D_real: 0.901 D_fake: 0.717 +(epoch: 142, iters: 1072, time: 0.064) G_GAN: 0.516 G_GAN_Feat: 0.894 G_ID: 0.203 G_Rec: 0.374 D_GP: 0.038 D_real: 1.088 D_fake: 0.494 +(epoch: 142, iters: 1472, time: 0.064) G_GAN: 0.282 G_GAN_Feat: 0.731 G_ID: 0.152 G_Rec: 0.365 D_GP: 0.019 D_real: 1.167 D_fake: 0.718 +(epoch: 142, iters: 1872, time: 0.064) G_GAN: 0.314 G_GAN_Feat: 0.775 G_ID: 0.224 G_Rec: 0.358 D_GP: 0.027 D_real: 1.086 D_fake: 0.687 +(epoch: 142, iters: 2272, time: 0.064) G_GAN: 0.140 G_GAN_Feat: 0.734 G_ID: 0.177 G_Rec: 0.387 D_GP: 0.037 D_real: 0.929 D_fake: 0.860 +(epoch: 142, iters: 2672, time: 0.064) G_GAN: 0.823 G_GAN_Feat: 0.892 G_ID: 0.206 G_Rec: 0.389 D_GP: 0.042 D_real: 1.342 D_fake: 0.334 +(epoch: 142, iters: 3072, time: 0.064) G_GAN: 0.049 G_GAN_Feat: 0.823 G_ID: 0.157 G_Rec: 0.382 D_GP: 0.066 D_real: 1.018 D_fake: 0.953 +(epoch: 142, iters: 3472, time: 0.064) G_GAN: 0.383 G_GAN_Feat: 0.857 G_ID: 0.208 G_Rec: 0.429 D_GP: 0.029 D_real: 1.114 D_fake: 0.617 +(epoch: 142, iters: 3872, time: 0.064) G_GAN: 0.203 G_GAN_Feat: 0.799 G_ID: 0.153 G_Rec: 0.324 D_GP: 0.031 D_real: 1.048 D_fake: 0.797 +(epoch: 142, iters: 4272, time: 0.064) G_GAN: 0.521 G_GAN_Feat: 0.931 G_ID: 0.199 G_Rec: 0.439 D_GP: 0.042 D_real: 1.013 D_fake: 0.493 +(epoch: 142, iters: 4672, time: 0.064) G_GAN: 0.232 G_GAN_Feat: 0.686 G_ID: 0.164 G_Rec: 0.304 D_GP: 0.025 D_real: 1.100 D_fake: 0.768 +(epoch: 142, iters: 5072, time: 0.064) G_GAN: 0.517 G_GAN_Feat: 0.831 G_ID: 0.244 G_Rec: 0.387 D_GP: 0.027 D_real: 1.255 D_fake: 0.488 +(epoch: 142, iters: 5472, time: 0.064) G_GAN: 0.142 G_GAN_Feat: 0.735 G_ID: 0.122 G_Rec: 0.301 D_GP: 0.044 D_real: 0.938 D_fake: 0.858 +(epoch: 142, iters: 5872, time: 0.064) G_GAN: 0.522 G_GAN_Feat: 0.882 G_ID: 0.248 G_Rec: 0.414 D_GP: 0.026 D_real: 1.230 D_fake: 0.481 +(epoch: 142, iters: 6272, time: 0.064) G_GAN: 0.064 G_GAN_Feat: 0.878 G_ID: 0.155 G_Rec: 0.311 D_GP: 0.092 D_real: 0.454 D_fake: 0.937 +(epoch: 142, iters: 6672, time: 0.064) G_GAN: 0.145 G_GAN_Feat: 0.997 G_ID: 0.211 G_Rec: 0.417 D_GP: 0.077 D_real: 0.787 D_fake: 0.856 +(epoch: 142, iters: 7072, time: 0.064) G_GAN: 0.258 G_GAN_Feat: 0.631 G_ID: 0.191 G_Rec: 0.305 D_GP: 0.024 D_real: 1.139 D_fake: 0.742 +(epoch: 142, iters: 7472, time: 0.064) G_GAN: 0.342 G_GAN_Feat: 1.085 G_ID: 0.241 G_Rec: 0.392 D_GP: 0.083 D_real: 0.417 D_fake: 0.660 +(epoch: 142, iters: 7872, time: 0.064) G_GAN: 0.205 G_GAN_Feat: 0.860 G_ID: 0.176 G_Rec: 0.342 D_GP: 0.074 D_real: 0.816 D_fake: 0.806 +(epoch: 142, iters: 8272, time: 0.064) G_GAN: 0.735 G_GAN_Feat: 0.860 G_ID: 0.247 G_Rec: 0.414 D_GP: 0.024 D_real: 1.431 D_fake: 0.355 +(epoch: 143, iters: 64, time: 0.064) G_GAN: 0.159 G_GAN_Feat: 0.747 G_ID: 0.175 G_Rec: 0.349 D_GP: 0.034 D_real: 0.935 D_fake: 0.841 +(epoch: 143, iters: 464, time: 0.064) G_GAN: 0.152 G_GAN_Feat: 1.016 G_ID: 0.222 G_Rec: 0.438 D_GP: 0.078 D_real: 0.885 D_fake: 0.850 +(epoch: 143, iters: 864, time: 0.064) G_GAN: 0.325 G_GAN_Feat: 0.665 G_ID: 0.147 G_Rec: 0.315 D_GP: 0.021 D_real: 1.265 D_fake: 0.675 +(epoch: 143, iters: 1264, time: 0.064) G_GAN: 0.375 G_GAN_Feat: 1.109 G_ID: 0.203 G_Rec: 0.377 D_GP: 0.048 D_real: 0.565 D_fake: 0.630 +(epoch: 143, iters: 1664, time: 0.064) G_GAN: 0.531 G_GAN_Feat: 1.075 G_ID: 0.159 G_Rec: 0.410 D_GP: 0.068 D_real: 1.299 D_fake: 0.546 +(epoch: 143, iters: 2064, time: 0.064) G_GAN: 0.428 G_GAN_Feat: 0.817 G_ID: 0.208 G_Rec: 0.399 D_GP: 0.021 D_real: 1.257 D_fake: 0.574 +(epoch: 143, iters: 2464, time: 0.064) G_GAN: 0.222 G_GAN_Feat: 0.908 G_ID: 0.151 G_Rec: 0.349 D_GP: 0.070 D_real: 0.741 D_fake: 0.780 +(epoch: 143, iters: 2864, time: 0.064) G_GAN: 0.682 G_GAN_Feat: 1.008 G_ID: 0.214 G_Rec: 0.432 D_GP: 0.056 D_real: 1.047 D_fake: 0.365 +(epoch: 143, iters: 3264, time: 0.064) G_GAN: -0.041 G_GAN_Feat: 0.802 G_ID: 0.131 G_Rec: 0.409 D_GP: 0.033 D_real: 0.696 D_fake: 1.041 +(epoch: 143, iters: 3664, time: 0.064) G_GAN: 0.458 G_GAN_Feat: 0.818 G_ID: 0.221 G_Rec: 0.392 D_GP: 0.023 D_real: 1.182 D_fake: 0.545 +(epoch: 143, iters: 4064, time: 0.064) G_GAN: 0.240 G_GAN_Feat: 0.807 G_ID: 0.151 G_Rec: 0.319 D_GP: 0.050 D_real: 0.813 D_fake: 0.764 +(epoch: 143, iters: 4464, time: 0.064) G_GAN: 0.556 G_GAN_Feat: 0.902 G_ID: 0.212 G_Rec: 0.437 D_GP: 0.023 D_real: 1.200 D_fake: 0.448 +(epoch: 143, iters: 4864, time: 0.064) G_GAN: 0.242 G_GAN_Feat: 0.818 G_ID: 0.152 G_Rec: 0.328 D_GP: 0.052 D_real: 0.963 D_fake: 0.759 +(epoch: 143, iters: 5264, time: 0.064) G_GAN: 0.199 G_GAN_Feat: 0.939 G_ID: 0.239 G_Rec: 0.423 D_GP: 0.031 D_real: 0.860 D_fake: 0.803 +(epoch: 143, iters: 5664, time: 0.064) G_GAN: 0.077 G_GAN_Feat: 0.855 G_ID: 0.117 G_Rec: 0.334 D_GP: 0.090 D_real: 0.825 D_fake: 0.923 +(epoch: 143, iters: 6064, time: 0.064) G_GAN: 0.297 G_GAN_Feat: 0.917 G_ID: 0.216 G_Rec: 0.461 D_GP: 0.026 D_real: 1.159 D_fake: 0.703 +(epoch: 143, iters: 6464, time: 0.064) G_GAN: 0.191 G_GAN_Feat: 0.697 G_ID: 0.145 G_Rec: 0.326 D_GP: 0.024 D_real: 1.112 D_fake: 0.810 +(epoch: 143, iters: 6864, time: 0.064) G_GAN: 0.641 G_GAN_Feat: 0.902 G_ID: 0.230 G_Rec: 0.404 D_GP: 0.040 D_real: 1.127 D_fake: 0.394 +(epoch: 143, iters: 7264, time: 0.064) G_GAN: 0.150 G_GAN_Feat: 1.085 G_ID: 0.134 G_Rec: 0.343 D_GP: 0.109 D_real: 0.312 D_fake: 0.850 +(epoch: 143, iters: 7664, time: 0.064) G_GAN: 0.661 G_GAN_Feat: 0.939 G_ID: 0.198 G_Rec: 0.439 D_GP: 0.029 D_real: 1.348 D_fake: 0.345 +(epoch: 143, iters: 8064, time: 0.064) G_GAN: 0.305 G_GAN_Feat: 0.831 G_ID: 0.149 G_Rec: 0.357 D_GP: 0.058 D_real: 0.989 D_fake: 0.696 +(epoch: 143, iters: 8464, time: 0.064) G_GAN: 0.820 G_GAN_Feat: 0.945 G_ID: 0.222 G_Rec: 0.415 D_GP: 0.043 D_real: 1.384 D_fake: 0.221 +(epoch: 144, iters: 256, time: 0.064) G_GAN: 0.070 G_GAN_Feat: 1.022 G_ID: 0.178 G_Rec: 0.328 D_GP: 0.064 D_real: 0.293 D_fake: 0.930 +(epoch: 144, iters: 656, time: 0.064) G_GAN: 0.648 G_GAN_Feat: 0.924 G_ID: 0.204 G_Rec: 0.422 D_GP: 0.025 D_real: 1.330 D_fake: 0.361 +(epoch: 144, iters: 1056, time: 0.064) G_GAN: 0.407 G_GAN_Feat: 0.639 G_ID: 0.136 G_Rec: 0.298 D_GP: 0.026 D_real: 1.335 D_fake: 0.594 +(epoch: 144, iters: 1456, time: 0.064) G_GAN: 0.145 G_GAN_Feat: 0.845 G_ID: 0.213 G_Rec: 0.402 D_GP: 0.037 D_real: 0.853 D_fake: 0.855 +(epoch: 144, iters: 1856, time: 0.064) G_GAN: -0.023 G_GAN_Feat: 0.724 G_ID: 0.172 G_Rec: 0.310 D_GP: 0.045 D_real: 0.870 D_fake: 1.023 +(epoch: 144, iters: 2256, time: 0.064) G_GAN: 0.439 G_GAN_Feat: 0.840 G_ID: 0.192 G_Rec: 0.392 D_GP: 0.030 D_real: 1.155 D_fake: 0.576 +(epoch: 144, iters: 2656, time: 0.064) G_GAN: 0.298 G_GAN_Feat: 0.814 G_ID: 0.173 G_Rec: 0.340 D_GP: 0.038 D_real: 1.035 D_fake: 0.704 +(epoch: 144, iters: 3056, time: 0.064) G_GAN: 0.230 G_GAN_Feat: 0.949 G_ID: 0.232 G_Rec: 0.401 D_GP: 0.074 D_real: 0.746 D_fake: 0.771 +(epoch: 144, iters: 3456, time: 0.064) G_GAN: 0.286 G_GAN_Feat: 0.746 G_ID: 0.164 G_Rec: 0.327 D_GP: 0.037 D_real: 1.046 D_fake: 0.714 +(epoch: 144, iters: 3856, time: 0.064) G_GAN: 0.388 G_GAN_Feat: 1.013 G_ID: 0.229 G_Rec: 0.420 D_GP: 0.135 D_real: 0.831 D_fake: 0.615 +(epoch: 144, iters: 4256, time: 0.064) G_GAN: 0.260 G_GAN_Feat: 0.654 G_ID: 0.153 G_Rec: 0.320 D_GP: 0.026 D_real: 1.165 D_fake: 0.740 +(epoch: 144, iters: 4656, time: 0.064) G_GAN: 0.081 G_GAN_Feat: 1.095 G_ID: 0.213 G_Rec: 0.452 D_GP: 0.203 D_real: 0.293 D_fake: 0.919 +(epoch: 144, iters: 5056, time: 0.064) G_GAN: 0.175 G_GAN_Feat: 0.949 G_ID: 0.154 G_Rec: 0.359 D_GP: 0.097 D_real: 0.541 D_fake: 0.825 +(epoch: 144, iters: 5456, time: 0.064) G_GAN: 0.258 G_GAN_Feat: 1.064 G_ID: 0.219 G_Rec: 0.460 D_GP: 0.212 D_real: 0.523 D_fake: 0.750 +(epoch: 144, iters: 5856, time: 0.064) G_GAN: 0.405 G_GAN_Feat: 0.651 G_ID: 0.151 G_Rec: 0.289 D_GP: 0.026 D_real: 1.266 D_fake: 0.595 +(epoch: 144, iters: 6256, time: 0.064) G_GAN: 0.804 G_GAN_Feat: 1.326 G_ID: 0.239 G_Rec: 0.466 D_GP: 0.497 D_real: 0.731 D_fake: 0.246 +(epoch: 144, iters: 6656, time: 0.064) G_GAN: 0.331 G_GAN_Feat: 0.883 G_ID: 0.156 G_Rec: 0.360 D_GP: 0.042 D_real: 0.739 D_fake: 0.673 +(epoch: 144, iters: 7056, time: 0.064) G_GAN: 0.481 G_GAN_Feat: 1.109 G_ID: 0.219 G_Rec: 0.410 D_GP: 0.501 D_real: 0.427 D_fake: 0.648 +(epoch: 144, iters: 7456, time: 0.064) G_GAN: 0.308 G_GAN_Feat: 0.702 G_ID: 0.151 G_Rec: 0.329 D_GP: 0.021 D_real: 1.173 D_fake: 0.692 +(epoch: 144, iters: 7856, time: 0.064) G_GAN: 0.473 G_GAN_Feat: 1.036 G_ID: 0.189 G_Rec: 0.417 D_GP: 0.144 D_real: 0.781 D_fake: 0.533 +(epoch: 144, iters: 8256, time: 0.064) G_GAN: 0.101 G_GAN_Feat: 0.631 G_ID: 0.177 G_Rec: 0.288 D_GP: 0.024 D_real: 1.019 D_fake: 0.899 +(epoch: 145, iters: 48, time: 0.064) G_GAN: 0.333 G_GAN_Feat: 0.957 G_ID: 0.193 G_Rec: 0.435 D_GP: 0.047 D_real: 1.067 D_fake: 0.669 +(epoch: 145, iters: 448, time: 0.064) G_GAN: 0.265 G_GAN_Feat: 0.614 G_ID: 0.161 G_Rec: 0.418 D_GP: 0.021 D_real: 1.177 D_fake: 0.735 +(epoch: 145, iters: 848, time: 0.064) G_GAN: 0.601 G_GAN_Feat: 0.787 G_ID: 0.231 G_Rec: 0.390 D_GP: 0.022 D_real: 1.272 D_fake: 0.422 +(epoch: 145, iters: 1248, time: 0.064) G_GAN: 0.272 G_GAN_Feat: 0.737 G_ID: 0.170 G_Rec: 0.325 D_GP: 0.080 D_real: 0.975 D_fake: 0.730 +(epoch: 145, iters: 1648, time: 0.064) G_GAN: 0.676 G_GAN_Feat: 1.044 G_ID: 0.213 G_Rec: 0.426 D_GP: 0.070 D_real: 0.969 D_fake: 0.366 +(epoch: 145, iters: 2048, time: 0.064) G_GAN: 0.561 G_GAN_Feat: 0.906 G_ID: 0.137 G_Rec: 0.337 D_GP: 0.059 D_real: 1.327 D_fake: 0.448 +(epoch: 145, iters: 2448, time: 0.064) G_GAN: 0.457 G_GAN_Feat: 0.877 G_ID: 0.192 G_Rec: 0.400 D_GP: 0.031 D_real: 1.080 D_fake: 0.553 +(epoch: 145, iters: 2848, time: 0.064) G_GAN: 0.308 G_GAN_Feat: 0.967 G_ID: 0.176 G_Rec: 0.357 D_GP: 0.105 D_real: 0.629 D_fake: 0.693 +(epoch: 145, iters: 3248, time: 0.064) G_GAN: 0.378 G_GAN_Feat: 0.895 G_ID: 0.197 G_Rec: 0.416 D_GP: 0.032 D_real: 0.929 D_fake: 0.623 +(epoch: 145, iters: 3648, time: 0.064) G_GAN: 0.322 G_GAN_Feat: 0.812 G_ID: 0.139 G_Rec: 0.379 D_GP: 0.033 D_real: 1.088 D_fake: 0.680 +(epoch: 145, iters: 4048, time: 0.064) G_GAN: 0.193 G_GAN_Feat: 0.806 G_ID: 0.202 G_Rec: 0.377 D_GP: 0.025 D_real: 0.963 D_fake: 0.808 +(epoch: 145, iters: 4448, time: 0.064) G_GAN: 0.100 G_GAN_Feat: 0.749 G_ID: 0.146 G_Rec: 0.339 D_GP: 0.044 D_real: 0.882 D_fake: 0.901 +(epoch: 145, iters: 4848, time: 0.064) G_GAN: 0.373 G_GAN_Feat: 0.919 G_ID: 0.218 G_Rec: 0.415 D_GP: 0.030 D_real: 1.050 D_fake: 0.627 +(epoch: 145, iters: 5248, time: 0.064) G_GAN: 0.063 G_GAN_Feat: 0.750 G_ID: 0.144 G_Rec: 0.323 D_GP: 0.032 D_real: 0.874 D_fake: 0.937 +(epoch: 145, iters: 5648, time: 0.064) G_GAN: 0.573 G_GAN_Feat: 0.946 G_ID: 0.206 G_Rec: 0.392 D_GP: 0.055 D_real: 1.223 D_fake: 0.442 +(epoch: 145, iters: 6048, time: 0.064) G_GAN: 0.193 G_GAN_Feat: 0.880 G_ID: 0.143 G_Rec: 0.324 D_GP: 0.039 D_real: 0.770 D_fake: 0.807 +(epoch: 145, iters: 6448, time: 0.064) G_GAN: 0.665 G_GAN_Feat: 1.094 G_ID: 0.215 G_Rec: 0.425 D_GP: 0.113 D_real: 0.958 D_fake: 0.414 +(epoch: 145, iters: 6848, time: 0.064) G_GAN: 0.315 G_GAN_Feat: 0.796 G_ID: 0.145 G_Rec: 0.330 D_GP: 0.027 D_real: 1.155 D_fake: 0.685 +(epoch: 145, iters: 7248, time: 0.064) G_GAN: 0.253 G_GAN_Feat: 0.826 G_ID: 0.203 G_Rec: 0.410 D_GP: 0.022 D_real: 0.997 D_fake: 0.747 +(epoch: 145, iters: 7648, time: 0.064) G_GAN: 0.138 G_GAN_Feat: 0.622 G_ID: 0.152 G_Rec: 0.321 D_GP: 0.027 D_real: 1.032 D_fake: 0.862 +(epoch: 145, iters: 8048, time: 0.064) G_GAN: 0.244 G_GAN_Feat: 0.845 G_ID: 0.191 G_Rec: 0.394 D_GP: 0.036 D_real: 0.898 D_fake: 0.757 +(epoch: 145, iters: 8448, time: 0.064) G_GAN: 0.253 G_GAN_Feat: 0.854 G_ID: 0.181 G_Rec: 0.344 D_GP: 0.037 D_real: 0.869 D_fake: 0.747 +(epoch: 146, iters: 240, time: 0.064) G_GAN: 0.553 G_GAN_Feat: 0.853 G_ID: 0.185 G_Rec: 0.423 D_GP: 0.022 D_real: 1.189 D_fake: 0.454 +(epoch: 146, iters: 640, time: 0.064) G_GAN: 0.331 G_GAN_Feat: 0.758 G_ID: 0.158 G_Rec: 0.333 D_GP: 0.051 D_real: 1.076 D_fake: 0.670 +(epoch: 146, iters: 1040, time: 0.064) G_GAN: 0.666 G_GAN_Feat: 1.046 G_ID: 0.217 G_Rec: 0.430 D_GP: 0.108 D_real: 0.980 D_fake: 0.345 +(epoch: 146, iters: 1440, time: 0.064) G_GAN: 0.382 G_GAN_Feat: 0.873 G_ID: 0.129 G_Rec: 0.323 D_GP: 0.055 D_real: 0.760 D_fake: 0.618 +(epoch: 146, iters: 1840, time: 0.064) G_GAN: 0.519 G_GAN_Feat: 0.832 G_ID: 0.214 G_Rec: 0.390 D_GP: 0.027 D_real: 1.285 D_fake: 0.483 +(epoch: 146, iters: 2240, time: 0.064) G_GAN: 0.038 G_GAN_Feat: 0.911 G_ID: 0.163 G_Rec: 0.337 D_GP: 0.096 D_real: 0.499 D_fake: 0.963 +(epoch: 146, iters: 2640, time: 0.064) G_GAN: 0.669 G_GAN_Feat: 1.052 G_ID: 0.197 G_Rec: 0.401 D_GP: 0.031 D_real: 1.038 D_fake: 0.362 +(epoch: 146, iters: 3040, time: 0.064) G_GAN: 0.125 G_GAN_Feat: 1.187 G_ID: 0.145 G_Rec: 0.422 D_GP: 1.129 D_real: 0.859 D_fake: 0.876 +(epoch: 146, iters: 3440, time: 0.064) G_GAN: 0.290 G_GAN_Feat: 0.862 G_ID: 0.236 G_Rec: 0.419 D_GP: 0.026 D_real: 1.088 D_fake: 0.713 +(epoch: 146, iters: 3840, time: 0.064) G_GAN: 0.014 G_GAN_Feat: 0.752 G_ID: 0.169 G_Rec: 0.354 D_GP: 0.038 D_real: 0.792 D_fake: 0.986 +(epoch: 146, iters: 4240, time: 0.064) G_GAN: 0.397 G_GAN_Feat: 0.966 G_ID: 0.208 G_Rec: 0.410 D_GP: 0.053 D_real: 0.955 D_fake: 0.605 +(epoch: 146, iters: 4640, time: 0.064) G_GAN: 0.263 G_GAN_Feat: 0.715 G_ID: 0.140 G_Rec: 0.294 D_GP: 0.033 D_real: 1.131 D_fake: 0.737 +(epoch: 146, iters: 5040, time: 0.064) G_GAN: 0.293 G_GAN_Feat: 0.930 G_ID: 0.265 G_Rec: 0.415 D_GP: 0.033 D_real: 0.869 D_fake: 0.708 +(epoch: 146, iters: 5440, time: 0.064) G_GAN: 0.304 G_GAN_Feat: 0.828 G_ID: 0.153 G_Rec: 0.315 D_GP: 0.027 D_real: 1.078 D_fake: 0.697 +(epoch: 146, iters: 5840, time: 0.064) G_GAN: 0.711 G_GAN_Feat: 0.846 G_ID: 0.206 G_Rec: 0.396 D_GP: 0.025 D_real: 1.345 D_fake: 0.302 +(epoch: 146, iters: 6240, time: 0.064) G_GAN: 0.319 G_GAN_Feat: 0.671 G_ID: 0.152 G_Rec: 0.322 D_GP: 0.023 D_real: 1.231 D_fake: 0.685 +(epoch: 146, iters: 6640, time: 0.064) G_GAN: 0.516 G_GAN_Feat: 0.993 G_ID: 0.207 G_Rec: 0.418 D_GP: 0.124 D_real: 0.830 D_fake: 0.491 +(epoch: 146, iters: 7040, time: 0.064) G_GAN: 0.160 G_GAN_Feat: 0.786 G_ID: 0.128 G_Rec: 0.349 D_GP: 0.036 D_real: 1.019 D_fake: 0.841 +(epoch: 146, iters: 7440, time: 0.064) G_GAN: 0.010 G_GAN_Feat: 1.240 G_ID: 0.223 G_Rec: 0.411 D_GP: 0.057 D_real: 1.244 D_fake: 0.991 +(epoch: 146, iters: 7840, time: 0.064) G_GAN: -0.070 G_GAN_Feat: 0.650 G_ID: 0.173 G_Rec: 0.338 D_GP: 0.020 D_real: 0.894 D_fake: 1.070 +(epoch: 146, iters: 8240, time: 0.064) G_GAN: 0.229 G_GAN_Feat: 0.957 G_ID: 0.224 G_Rec: 0.425 D_GP: 0.069 D_real: 0.712 D_fake: 0.772 +(epoch: 147, iters: 32, time: 0.064) G_GAN: 0.400 G_GAN_Feat: 0.680 G_ID: 0.128 G_Rec: 0.337 D_GP: 0.024 D_real: 1.338 D_fake: 0.600 +(epoch: 147, iters: 432, time: 0.064) G_GAN: 0.792 G_GAN_Feat: 1.117 G_ID: 0.217 G_Rec: 0.436 D_GP: 0.095 D_real: 0.850 D_fake: 0.263 +(epoch: 147, iters: 832, time: 0.064) G_GAN: 0.282 G_GAN_Feat: 0.743 G_ID: 0.145 G_Rec: 0.352 D_GP: 0.028 D_real: 1.208 D_fake: 0.721 +(epoch: 147, iters: 1232, time: 0.064) G_GAN: 0.250 G_GAN_Feat: 0.824 G_ID: 0.218 G_Rec: 0.376 D_GP: 0.027 D_real: 1.030 D_fake: 0.750 +(epoch: 147, iters: 1632, time: 0.064) G_GAN: 0.155 G_GAN_Feat: 0.869 G_ID: 0.171 G_Rec: 0.320 D_GP: 0.045 D_real: 0.688 D_fake: 0.846 +(epoch: 147, iters: 2032, time: 0.064) G_GAN: 0.204 G_GAN_Feat: 1.225 G_ID: 0.226 G_Rec: 0.467 D_GP: 0.851 D_real: 0.573 D_fake: 0.825 +(epoch: 147, iters: 2432, time: 0.064) G_GAN: 0.089 G_GAN_Feat: 0.727 G_ID: 0.146 G_Rec: 0.302 D_GP: 0.037 D_real: 0.941 D_fake: 0.911 +(epoch: 147, iters: 2832, time: 0.064) G_GAN: 0.477 G_GAN_Feat: 0.760 G_ID: 0.203 G_Rec: 0.399 D_GP: 0.021 D_real: 1.260 D_fake: 0.526 +(epoch: 147, iters: 3232, time: 0.064) G_GAN: 0.113 G_GAN_Feat: 0.642 G_ID: 0.146 G_Rec: 0.301 D_GP: 0.023 D_real: 1.053 D_fake: 0.887 +(epoch: 147, iters: 3632, time: 0.064) G_GAN: 0.273 G_GAN_Feat: 0.795 G_ID: 0.216 G_Rec: 0.407 D_GP: 0.032 D_real: 0.919 D_fake: 0.734 +(epoch: 147, iters: 4032, time: 0.064) G_GAN: 0.082 G_GAN_Feat: 0.742 G_ID: 0.154 G_Rec: 0.328 D_GP: 0.081 D_real: 0.894 D_fake: 0.918 +(epoch: 147, iters: 4432, time: 0.064) G_GAN: 0.643 G_GAN_Feat: 0.847 G_ID: 0.223 G_Rec: 0.386 D_GP: 0.026 D_real: 1.351 D_fake: 0.362 +(epoch: 147, iters: 4832, time: 0.064) G_GAN: 0.300 G_GAN_Feat: 0.824 G_ID: 0.182 G_Rec: 0.343 D_GP: 0.047 D_real: 0.936 D_fake: 0.700 +(epoch: 147, iters: 5232, time: 0.064) G_GAN: 0.595 G_GAN_Feat: 0.861 G_ID: 0.210 G_Rec: 0.391 D_GP: 0.027 D_real: 1.307 D_fake: 0.425 +(epoch: 147, iters: 5632, time: 0.064) G_GAN: 0.169 G_GAN_Feat: 0.793 G_ID: 0.167 G_Rec: 0.331 D_GP: 0.044 D_real: 0.885 D_fake: 0.831 +(epoch: 147, iters: 6032, time: 0.064) G_GAN: 0.617 G_GAN_Feat: 0.851 G_ID: 0.197 G_Rec: 0.415 D_GP: 0.026 D_real: 1.357 D_fake: 0.413 +(epoch: 147, iters: 6432, time: 0.064) G_GAN: 0.066 G_GAN_Feat: 0.830 G_ID: 0.138 G_Rec: 0.310 D_GP: 0.131 D_real: 0.447 D_fake: 0.936 +(epoch: 147, iters: 6832, time: 0.064) G_GAN: 0.557 G_GAN_Feat: 0.882 G_ID: 0.186 G_Rec: 0.456 D_GP: 0.022 D_real: 1.192 D_fake: 0.454 +(epoch: 147, iters: 7232, time: 0.064) G_GAN: 0.299 G_GAN_Feat: 0.602 G_ID: 0.154 G_Rec: 0.299 D_GP: 0.022 D_real: 1.221 D_fake: 0.701 +(epoch: 147, iters: 7632, time: 0.064) G_GAN: 0.707 G_GAN_Feat: 0.810 G_ID: 0.216 G_Rec: 0.358 D_GP: 0.028 D_real: 1.385 D_fake: 0.336 +(epoch: 147, iters: 8032, time: 0.064) G_GAN: 0.252 G_GAN_Feat: 0.807 G_ID: 0.163 G_Rec: 0.332 D_GP: 0.059 D_real: 0.822 D_fake: 0.748 +(epoch: 147, iters: 8432, time: 0.064) G_GAN: 0.203 G_GAN_Feat: 0.801 G_ID: 0.244 G_Rec: 0.426 D_GP: 0.031 D_real: 0.880 D_fake: 0.797 +(epoch: 148, iters: 224, time: 0.064) G_GAN: 0.244 G_GAN_Feat: 0.654 G_ID: 0.132 G_Rec: 0.296 D_GP: 0.027 D_real: 1.160 D_fake: 0.756 +(epoch: 148, iters: 624, time: 0.064) G_GAN: 0.477 G_GAN_Feat: 0.900 G_ID: 0.237 G_Rec: 0.395 D_GP: 0.028 D_real: 1.196 D_fake: 0.532 +(epoch: 148, iters: 1024, time: 0.064) G_GAN: 0.193 G_GAN_Feat: 0.707 G_ID: 0.146 G_Rec: 0.332 D_GP: 0.025 D_real: 1.089 D_fake: 0.808 +(epoch: 148, iters: 1424, time: 0.064) G_GAN: 0.365 G_GAN_Feat: 0.974 G_ID: 0.206 G_Rec: 0.407 D_GP: 0.044 D_real: 0.804 D_fake: 0.645 +(epoch: 148, iters: 1824, time: 0.064) G_GAN: 0.307 G_GAN_Feat: 0.836 G_ID: 0.163 G_Rec: 0.353 D_GP: 0.032 D_real: 1.149 D_fake: 0.696 +(epoch: 148, iters: 2224, time: 0.064) G_GAN: 0.684 G_GAN_Feat: 1.097 G_ID: 0.212 G_Rec: 0.436 D_GP: 0.091 D_real: 0.733 D_fake: 0.359 +(epoch: 148, iters: 2624, time: 0.064) G_GAN: 0.171 G_GAN_Feat: 0.750 G_ID: 0.157 G_Rec: 0.356 D_GP: 0.031 D_real: 0.998 D_fake: 0.829 +(epoch: 148, iters: 3024, time: 0.064) G_GAN: 0.172 G_GAN_Feat: 0.897 G_ID: 0.206 G_Rec: 0.410 D_GP: 0.060 D_real: 0.668 D_fake: 0.828 +(epoch: 148, iters: 3424, time: 0.064) G_GAN: 0.205 G_GAN_Feat: 0.751 G_ID: 0.133 G_Rec: 0.313 D_GP: 0.096 D_real: 0.862 D_fake: 0.796 +(epoch: 148, iters: 3824, time: 0.064) G_GAN: 0.611 G_GAN_Feat: 0.985 G_ID: 0.225 G_Rec: 0.382 D_GP: 0.058 D_real: 1.019 D_fake: 0.408 +(epoch: 148, iters: 4224, time: 0.064) G_GAN: 0.286 G_GAN_Feat: 0.763 G_ID: 0.159 G_Rec: 0.321 D_GP: 0.026 D_real: 1.154 D_fake: 0.718 +(epoch: 148, iters: 4624, time: 0.064) G_GAN: 0.324 G_GAN_Feat: 0.861 G_ID: 0.249 G_Rec: 0.394 D_GP: 0.027 D_real: 1.001 D_fake: 0.676 +(epoch: 148, iters: 5024, time: 0.064) G_GAN: 0.171 G_GAN_Feat: 0.878 G_ID: 0.138 G_Rec: 0.329 D_GP: 0.060 D_real: 0.704 D_fake: 0.829 +(epoch: 148, iters: 5424, time: 0.064) G_GAN: 0.384 G_GAN_Feat: 1.043 G_ID: 0.219 G_Rec: 0.416 D_GP: 0.054 D_real: 0.676 D_fake: 0.659 +(epoch: 148, iters: 5824, time: 0.064) G_GAN: 0.546 G_GAN_Feat: 0.835 G_ID: 0.145 G_Rec: 0.315 D_GP: 0.073 D_real: 1.027 D_fake: 0.493 +(epoch: 148, iters: 6224, time: 0.064) G_GAN: 0.499 G_GAN_Feat: 0.966 G_ID: 0.207 G_Rec: 0.414 D_GP: 0.055 D_real: 0.887 D_fake: 0.507 +(epoch: 148, iters: 6624, time: 0.064) G_GAN: 0.082 G_GAN_Feat: 0.752 G_ID: 0.126 G_Rec: 0.342 D_GP: 0.028 D_real: 0.938 D_fake: 0.918 +(epoch: 148, iters: 7024, time: 0.064) G_GAN: 0.392 G_GAN_Feat: 0.881 G_ID: 0.228 G_Rec: 0.423 D_GP: 0.024 D_real: 1.043 D_fake: 0.618 +(epoch: 148, iters: 7424, time: 0.064) G_GAN: 0.292 G_GAN_Feat: 0.715 G_ID: 0.140 G_Rec: 0.342 D_GP: 0.035 D_real: 1.059 D_fake: 0.709 +(epoch: 148, iters: 7824, time: 0.064) G_GAN: 0.588 G_GAN_Feat: 0.887 G_ID: 0.181 G_Rec: 0.412 D_GP: 0.027 D_real: 1.220 D_fake: 0.427 +(epoch: 148, iters: 8224, time: 0.064) G_GAN: 0.225 G_GAN_Feat: 0.739 G_ID: 0.154 G_Rec: 0.340 D_GP: 0.025 D_real: 1.060 D_fake: 0.782 +(epoch: 149, iters: 16, time: 0.064) G_GAN: 0.703 G_GAN_Feat: 0.783 G_ID: 0.235 G_Rec: 0.402 D_GP: 0.026 D_real: 1.415 D_fake: 0.364 +(epoch: 149, iters: 416, time: 0.064) G_GAN: 0.272 G_GAN_Feat: 0.688 G_ID: 0.154 G_Rec: 0.328 D_GP: 0.020 D_real: 1.184 D_fake: 0.728 +(epoch: 149, iters: 816, time: 0.064) G_GAN: 0.333 G_GAN_Feat: 0.941 G_ID: 0.192 G_Rec: 0.444 D_GP: 0.025 D_real: 0.868 D_fake: 0.668 +(epoch: 149, iters: 1216, time: 0.064) G_GAN: 0.005 G_GAN_Feat: 0.843 G_ID: 0.133 G_Rec: 0.361 D_GP: 0.098 D_real: 0.481 D_fake: 0.995 +(epoch: 149, iters: 1616, time: 0.064) G_GAN: 0.530 G_GAN_Feat: 0.862 G_ID: 0.211 G_Rec: 0.392 D_GP: 0.026 D_real: 1.154 D_fake: 0.479 +(epoch: 149, iters: 2016, time: 0.064) G_GAN: 0.206 G_GAN_Feat: 0.935 G_ID: 0.136 G_Rec: 0.366 D_GP: 0.198 D_real: 0.477 D_fake: 0.794 +(epoch: 149, iters: 2416, time: 0.064) G_GAN: 0.458 G_GAN_Feat: 0.900 G_ID: 0.216 G_Rec: 0.384 D_GP: 0.058 D_real: 0.920 D_fake: 0.547 +(epoch: 149, iters: 2816, time: 0.064) G_GAN: 0.311 G_GAN_Feat: 0.732 G_ID: 0.148 G_Rec: 0.298 D_GP: 0.027 D_real: 1.152 D_fake: 0.689 +(epoch: 149, iters: 3216, time: 0.064) G_GAN: 0.376 G_GAN_Feat: 0.798 G_ID: 0.210 G_Rec: 0.346 D_GP: 0.028 D_real: 1.077 D_fake: 0.624 +(epoch: 149, iters: 3616, time: 0.064) G_GAN: 0.237 G_GAN_Feat: 0.967 G_ID: 0.149 G_Rec: 0.415 D_GP: 0.063 D_real: 0.768 D_fake: 0.764 +(epoch: 149, iters: 4016, time: 0.064) G_GAN: 1.019 G_GAN_Feat: 1.046 G_ID: 0.213 G_Rec: 0.421 D_GP: 0.062 D_real: 1.202 D_fake: 0.127 +(epoch: 149, iters: 4416, time: 0.064) G_GAN: 0.496 G_GAN_Feat: 0.974 G_ID: 0.162 G_Rec: 0.375 D_GP: 0.085 D_real: 0.555 D_fake: 0.545 +(epoch: 149, iters: 4816, time: 0.064) G_GAN: 0.622 G_GAN_Feat: 0.863 G_ID: 0.188 G_Rec: 0.420 D_GP: 0.025 D_real: 1.388 D_fake: 0.402 +(epoch: 149, iters: 5216, time: 0.064) G_GAN: 0.261 G_GAN_Feat: 0.879 G_ID: 0.158 G_Rec: 0.353 D_GP: 0.042 D_real: 1.013 D_fake: 0.740 +(epoch: 149, iters: 5616, time: 0.064) G_GAN: 0.095 G_GAN_Feat: 1.059 G_ID: 0.223 G_Rec: 0.423 D_GP: 0.063 D_real: 0.362 D_fake: 0.905 +(epoch: 149, iters: 6016, time: 0.064) G_GAN: 0.269 G_GAN_Feat: 0.825 G_ID: 0.150 G_Rec: 0.377 D_GP: 0.055 D_real: 0.994 D_fake: 0.732 +(epoch: 149, iters: 6416, time: 0.064) G_GAN: 0.447 G_GAN_Feat: 0.879 G_ID: 0.235 G_Rec: 0.450 D_GP: 0.025 D_real: 1.147 D_fake: 0.562 +(epoch: 149, iters: 6816, time: 0.064) G_GAN: 0.342 G_GAN_Feat: 0.849 G_ID: 0.151 G_Rec: 0.404 D_GP: 0.031 D_real: 1.053 D_fake: 0.658 +(epoch: 149, iters: 7216, time: 0.064) G_GAN: 0.584 G_GAN_Feat: 0.902 G_ID: 0.187 G_Rec: 0.383 D_GP: 0.028 D_real: 1.156 D_fake: 0.420 +(epoch: 149, iters: 7616, time: 0.064) G_GAN: 0.127 G_GAN_Feat: 0.838 G_ID: 0.143 G_Rec: 0.366 D_GP: 0.061 D_real: 0.760 D_fake: 0.873 +(epoch: 149, iters: 8016, time: 0.064) G_GAN: 0.603 G_GAN_Feat: 0.864 G_ID: 0.212 G_Rec: 0.425 D_GP: 0.025 D_real: 1.375 D_fake: 0.414 +(epoch: 149, iters: 8416, time: 0.064) G_GAN: 0.159 G_GAN_Feat: 0.678 G_ID: 0.144 G_Rec: 0.332 D_GP: 0.021 D_real: 1.037 D_fake: 0.841 +(epoch: 150, iters: 208, time: 0.064) G_GAN: 0.288 G_GAN_Feat: 0.764 G_ID: 0.203 G_Rec: 0.378 D_GP: 0.029 D_real: 1.069 D_fake: 0.713 +(epoch: 150, iters: 608, time: 0.064) G_GAN: 0.164 G_GAN_Feat: 0.645 G_ID: 0.169 G_Rec: 0.362 D_GP: 0.021 D_real: 1.127 D_fake: 0.836 +(epoch: 150, iters: 1008, time: 0.064) G_GAN: 0.310 G_GAN_Feat: 0.909 G_ID: 0.216 G_Rec: 0.415 D_GP: 0.065 D_real: 0.896 D_fake: 0.692 +(epoch: 150, iters: 1408, time: 0.064) G_GAN: 0.162 G_GAN_Feat: 0.739 G_ID: 0.130 G_Rec: 0.331 D_GP: 0.033 D_real: 1.045 D_fake: 0.838 +(epoch: 150, iters: 1808, time: 0.064) G_GAN: 0.680 G_GAN_Feat: 1.296 G_ID: 0.187 G_Rec: 0.429 D_GP: 0.589 D_real: 0.471 D_fake: 0.381 +(epoch: 150, iters: 2208, time: 0.064) G_GAN: -0.039 G_GAN_Feat: 0.916 G_ID: 0.144 G_Rec: 0.351 D_GP: 0.029 D_real: 0.641 D_fake: 1.039 +(epoch: 150, iters: 2608, time: 0.064) G_GAN: 0.545 G_GAN_Feat: 0.924 G_ID: 0.214 G_Rec: 0.385 D_GP: 0.032 D_real: 1.149 D_fake: 0.474 +(epoch: 150, iters: 3008, time: 0.064) G_GAN: 0.354 G_GAN_Feat: 0.685 G_ID: 0.143 G_Rec: 0.343 D_GP: 0.023 D_real: 1.261 D_fake: 0.646 +(epoch: 150, iters: 3408, time: 0.064) G_GAN: 0.611 G_GAN_Feat: 0.840 G_ID: 0.227 G_Rec: 0.417 D_GP: 0.031 D_real: 1.456 D_fake: 0.410 +(epoch: 150, iters: 3808, time: 0.064) G_GAN: 0.222 G_GAN_Feat: 0.674 G_ID: 0.136 G_Rec: 0.344 D_GP: 0.028 D_real: 1.061 D_fake: 0.779 +(epoch: 150, iters: 4208, time: 0.064) G_GAN: 0.526 G_GAN_Feat: 0.897 G_ID: 0.198 G_Rec: 0.406 D_GP: 0.044 D_real: 1.115 D_fake: 0.478 +(epoch: 150, iters: 4608, time: 0.064) G_GAN: 0.375 G_GAN_Feat: 0.876 G_ID: 0.138 G_Rec: 0.331 D_GP: 0.221 D_real: 0.881 D_fake: 0.627 +(epoch: 150, iters: 5008, time: 0.064) G_GAN: 0.210 G_GAN_Feat: 0.975 G_ID: 0.200 G_Rec: 0.395 D_GP: 0.056 D_real: 0.731 D_fake: 0.791 +(epoch: 150, iters: 5408, time: 0.064) G_GAN: 0.287 G_GAN_Feat: 0.692 G_ID: 0.154 G_Rec: 0.320 D_GP: 0.035 D_real: 1.189 D_fake: 0.713 +(epoch: 150, iters: 5808, time: 0.064) G_GAN: 0.724 G_GAN_Feat: 1.092 G_ID: 0.192 G_Rec: 0.412 D_GP: 0.048 D_real: 0.846 D_fake: 0.318 +(epoch: 150, iters: 6208, time: 0.064) G_GAN: 0.325 G_GAN_Feat: 0.911 G_ID: 0.137 G_Rec: 0.332 D_GP: 0.033 D_real: 0.928 D_fake: 0.675 +(epoch: 150, iters: 6608, time: 0.064) G_GAN: 0.429 G_GAN_Feat: 1.071 G_ID: 0.209 G_Rec: 0.409 D_GP: 0.049 D_real: 0.630 D_fake: 0.573 +(epoch: 150, iters: 7008, time: 0.064) G_GAN: 0.122 G_GAN_Feat: 0.673 G_ID: 0.132 G_Rec: 0.327 D_GP: 0.024 D_real: 1.087 D_fake: 0.878 +(epoch: 150, iters: 7408, time: 0.064) G_GAN: 0.541 G_GAN_Feat: 1.043 G_ID: 0.199 G_Rec: 0.420 D_GP: 0.185 D_real: 0.853 D_fake: 0.487 +(epoch: 150, iters: 7808, time: 0.064) G_GAN: 0.111 G_GAN_Feat: 1.065 G_ID: 0.161 G_Rec: 0.391 D_GP: 0.201 D_real: 0.364 D_fake: 0.889 +(epoch: 150, iters: 8208, time: 0.064) G_GAN: 0.222 G_GAN_Feat: 1.018 G_ID: 0.172 G_Rec: 0.436 D_GP: 0.024 D_real: 0.848 D_fake: 0.778 +(epoch: 150, iters: 8608, time: 0.064) G_GAN: 0.170 G_GAN_Feat: 0.854 G_ID: 0.148 G_Rec: 0.325 D_GP: 0.038 D_real: 1.112 D_fake: 0.831 +(epoch: 151, iters: 400, time: 0.064) G_GAN: 0.468 G_GAN_Feat: 0.817 G_ID: 0.169 G_Rec: 0.373 D_GP: 0.024 D_real: 1.196 D_fake: 0.539 +(epoch: 151, iters: 800, time: 0.064) G_GAN: 0.115 G_GAN_Feat: 0.762 G_ID: 0.173 G_Rec: 0.372 D_GP: 0.026 D_real: 0.957 D_fake: 0.885 +(epoch: 151, iters: 1200, time: 0.064) G_GAN: 0.379 G_GAN_Feat: 1.035 G_ID: 0.227 G_Rec: 0.401 D_GP: 0.087 D_real: 0.465 D_fake: 0.621 +(epoch: 151, iters: 1600, time: 0.064) G_GAN: 0.318 G_GAN_Feat: 0.812 G_ID: 0.128 G_Rec: 0.343 D_GP: 0.039 D_real: 1.103 D_fake: 0.688 +(epoch: 151, iters: 2000, time: 0.064) G_GAN: 0.398 G_GAN_Feat: 0.806 G_ID: 0.186 G_Rec: 0.389 D_GP: 0.025 D_real: 1.259 D_fake: 0.602 +(epoch: 151, iters: 2400, time: 0.064) G_GAN: -0.120 G_GAN_Feat: 0.876 G_ID: 0.147 G_Rec: 0.313 D_GP: 0.049 D_real: 0.380 D_fake: 1.120 +(epoch: 151, iters: 2800, time: 0.064) G_GAN: 0.611 G_GAN_Feat: 0.898 G_ID: 0.208 G_Rec: 0.406 D_GP: 0.037 D_real: 1.251 D_fake: 0.461 +(epoch: 151, iters: 3200, time: 0.064) G_GAN: 0.270 G_GAN_Feat: 0.759 G_ID: 0.150 G_Rec: 0.332 D_GP: 0.035 D_real: 1.087 D_fake: 0.731 +(epoch: 151, iters: 3600, time: 0.064) G_GAN: 0.663 G_GAN_Feat: 0.910 G_ID: 0.188 G_Rec: 0.397 D_GP: 0.038 D_real: 1.130 D_fake: 0.354 +(epoch: 151, iters: 4000, time: 0.064) G_GAN: 0.015 G_GAN_Feat: 1.077 G_ID: 0.159 G_Rec: 0.353 D_GP: 0.299 D_real: 0.226 D_fake: 0.985 +(epoch: 151, iters: 4400, time: 0.064) G_GAN: 0.637 G_GAN_Feat: 1.154 G_ID: 0.200 G_Rec: 0.507 D_GP: 0.132 D_real: 1.082 D_fake: 0.397 +(epoch: 151, iters: 4800, time: 0.064) G_GAN: -0.154 G_GAN_Feat: 0.822 G_ID: 0.162 G_Rec: 0.366 D_GP: 0.076 D_real: 0.446 D_fake: 1.154 +(epoch: 151, iters: 5200, time: 0.064) G_GAN: 0.197 G_GAN_Feat: 1.012 G_ID: 0.252 G_Rec: 0.442 D_GP: 0.043 D_real: 0.799 D_fake: 0.803 +(epoch: 151, iters: 5600, time: 0.064) G_GAN: -0.118 G_GAN_Feat: 0.742 G_ID: 0.144 G_Rec: 0.336 D_GP: 0.040 D_real: 0.762 D_fake: 1.118 +(epoch: 151, iters: 6000, time: 0.064) G_GAN: 0.471 G_GAN_Feat: 0.819 G_ID: 0.194 G_Rec: 0.419 D_GP: 0.024 D_real: 1.200 D_fake: 0.529 +(epoch: 151, iters: 6400, time: 0.064) G_GAN: 0.239 G_GAN_Feat: 0.703 G_ID: 0.160 G_Rec: 0.327 D_GP: 0.036 D_real: 1.093 D_fake: 0.761 +(epoch: 151, iters: 6800, time: 0.064) G_GAN: 0.469 G_GAN_Feat: 0.867 G_ID: 0.200 G_Rec: 0.395 D_GP: 0.035 D_real: 1.136 D_fake: 0.533 +(epoch: 151, iters: 7200, time: 0.064) G_GAN: 0.129 G_GAN_Feat: 0.708 G_ID: 0.132 G_Rec: 0.324 D_GP: 0.025 D_real: 1.014 D_fake: 0.871 +(epoch: 151, iters: 7600, time: 0.064) G_GAN: 0.464 G_GAN_Feat: 0.837 G_ID: 0.181 G_Rec: 0.391 D_GP: 0.033 D_real: 1.183 D_fake: 0.543 +(epoch: 151, iters: 8000, time: 0.064) G_GAN: 0.144 G_GAN_Feat: 0.727 G_ID: 0.150 G_Rec: 0.318 D_GP: 0.035 D_real: 0.896 D_fake: 0.856 +(epoch: 151, iters: 8400, time: 0.064) G_GAN: 0.535 G_GAN_Feat: 1.021 G_ID: 0.212 G_Rec: 0.420 D_GP: 0.053 D_real: 0.845 D_fake: 0.469 +(epoch: 152, iters: 192, time: 0.064) G_GAN: 0.339 G_GAN_Feat: 1.190 G_ID: 0.145 G_Rec: 0.347 D_GP: 0.189 D_real: 0.698 D_fake: 0.662 +(epoch: 152, iters: 592, time: 0.064) G_GAN: 0.508 G_GAN_Feat: 0.963 G_ID: 0.212 G_Rec: 0.378 D_GP: 0.041 D_real: 1.100 D_fake: 0.496 +(epoch: 152, iters: 992, time: 0.064) G_GAN: -0.044 G_GAN_Feat: 0.722 G_ID: 0.133 G_Rec: 0.342 D_GP: 0.021 D_real: 0.849 D_fake: 1.044 +(epoch: 152, iters: 1392, time: 0.064) G_GAN: 0.129 G_GAN_Feat: 1.248 G_ID: 0.190 G_Rec: 0.459 D_GP: 0.140 D_real: 0.376 D_fake: 0.871 +(epoch: 152, iters: 1792, time: 0.064) G_GAN: 0.387 G_GAN_Feat: 0.719 G_ID: 0.145 G_Rec: 0.330 D_GP: 0.026 D_real: 1.245 D_fake: 0.622 +(epoch: 152, iters: 2192, time: 0.064) G_GAN: 0.730 G_GAN_Feat: 0.953 G_ID: 0.200 G_Rec: 0.408 D_GP: 0.030 D_real: 1.307 D_fake: 0.333 +(epoch: 152, iters: 2592, time: 0.064) G_GAN: 0.397 G_GAN_Feat: 0.838 G_ID: 0.161 G_Rec: 0.367 D_GP: 0.031 D_real: 0.930 D_fake: 0.605 +(epoch: 152, iters: 2992, time: 0.064) G_GAN: 0.383 G_GAN_Feat: 0.749 G_ID: 0.215 G_Rec: 0.359 D_GP: 0.037 D_real: 1.196 D_fake: 0.619 +(epoch: 152, iters: 3392, time: 0.064) G_GAN: 0.121 G_GAN_Feat: 0.732 G_ID: 0.135 G_Rec: 0.314 D_GP: 0.041 D_real: 0.985 D_fake: 0.879 +(epoch: 152, iters: 3792, time: 0.064) G_GAN: 0.446 G_GAN_Feat: 0.849 G_ID: 0.211 G_Rec: 0.391 D_GP: 0.045 D_real: 1.121 D_fake: 0.561 +(epoch: 152, iters: 4192, time: 0.064) G_GAN: 0.174 G_GAN_Feat: 0.896 G_ID: 0.146 G_Rec: 0.354 D_GP: 0.106 D_real: 0.788 D_fake: 0.827 +(epoch: 152, iters: 4592, time: 0.064) G_GAN: 0.497 G_GAN_Feat: 0.900 G_ID: 0.201 G_Rec: 0.420 D_GP: 0.030 D_real: 1.236 D_fake: 0.509 +(epoch: 152, iters: 4992, time: 0.064) G_GAN: 0.206 G_GAN_Feat: 0.758 G_ID: 0.172 G_Rec: 0.324 D_GP: 0.027 D_real: 1.080 D_fake: 0.794 +(epoch: 152, iters: 5392, time: 0.064) G_GAN: 0.499 G_GAN_Feat: 0.954 G_ID: 0.188 G_Rec: 0.390 D_GP: 0.042 D_real: 1.064 D_fake: 0.511 +(epoch: 152, iters: 5792, time: 0.064) G_GAN: 0.326 G_GAN_Feat: 0.691 G_ID: 0.160 G_Rec: 0.308 D_GP: 0.022 D_real: 1.181 D_fake: 0.674 +(epoch: 152, iters: 6192, time: 0.064) G_GAN: 0.540 G_GAN_Feat: 1.061 G_ID: 0.218 G_Rec: 0.403 D_GP: 0.092 D_real: 1.075 D_fake: 0.556 +(epoch: 152, iters: 6592, time: 0.064) G_GAN: 0.390 G_GAN_Feat: 0.648 G_ID: 0.152 G_Rec: 0.334 D_GP: 0.020 D_real: 1.377 D_fake: 0.613 +(epoch: 152, iters: 6992, time: 0.064) G_GAN: 0.466 G_GAN_Feat: 0.838 G_ID: 0.200 G_Rec: 0.407 D_GP: 0.020 D_real: 1.181 D_fake: 0.538 +(epoch: 152, iters: 7392, time: 0.064) G_GAN: 0.119 G_GAN_Feat: 0.749 G_ID: 0.186 G_Rec: 0.381 D_GP: 0.028 D_real: 0.936 D_fake: 0.881 +(epoch: 152, iters: 7792, time: 0.064) G_GAN: 0.219 G_GAN_Feat: 0.834 G_ID: 0.235 G_Rec: 0.402 D_GP: 0.036 D_real: 0.883 D_fake: 0.782 +(epoch: 152, iters: 8192, time: 0.064) G_GAN: 0.063 G_GAN_Feat: 0.634 G_ID: 0.142 G_Rec: 0.317 D_GP: 0.026 D_real: 0.978 D_fake: 0.937 +(epoch: 152, iters: 8592, time: 0.064) G_GAN: 0.558 G_GAN_Feat: 0.985 G_ID: 0.197 G_Rec: 0.393 D_GP: 0.089 D_real: 0.898 D_fake: 0.460 +(epoch: 153, iters: 384, time: 0.064) G_GAN: 0.225 G_GAN_Feat: 0.972 G_ID: 0.139 G_Rec: 0.324 D_GP: 0.064 D_real: 0.507 D_fake: 0.775 +(epoch: 153, iters: 784, time: 0.064) G_GAN: 0.266 G_GAN_Feat: 0.805 G_ID: 0.221 G_Rec: 0.357 D_GP: 0.031 D_real: 1.110 D_fake: 0.734 +(epoch: 153, iters: 1184, time: 0.064) G_GAN: -0.074 G_GAN_Feat: 0.875 G_ID: 0.157 G_Rec: 0.355 D_GP: 0.050 D_real: 0.647 D_fake: 1.074 +(epoch: 153, iters: 1584, time: 0.064) G_GAN: 0.507 G_GAN_Feat: 1.022 G_ID: 0.202 G_Rec: 0.433 D_GP: 0.041 D_real: 0.894 D_fake: 0.499 +(epoch: 153, iters: 1984, time: 0.064) G_GAN: 0.032 G_GAN_Feat: 0.703 G_ID: 0.160 G_Rec: 0.316 D_GP: 0.025 D_real: 0.967 D_fake: 0.968 +(epoch: 153, iters: 2384, time: 0.064) G_GAN: 0.527 G_GAN_Feat: 0.923 G_ID: 0.221 G_Rec: 0.402 D_GP: 0.031 D_real: 1.109 D_fake: 0.500 +(epoch: 153, iters: 2784, time: 0.064) G_GAN: 0.117 G_GAN_Feat: 0.946 G_ID: 0.161 G_Rec: 0.346 D_GP: 0.198 D_real: 0.399 D_fake: 0.884 +(epoch: 153, iters: 3184, time: 0.064) G_GAN: 0.480 G_GAN_Feat: 0.847 G_ID: 0.181 G_Rec: 0.352 D_GP: 0.030 D_real: 1.187 D_fake: 0.533 +(epoch: 153, iters: 3584, time: 0.064) G_GAN: 0.229 G_GAN_Feat: 0.951 G_ID: 0.177 G_Rec: 0.361 D_GP: 0.274 D_real: 0.389 D_fake: 0.771 +(epoch: 153, iters: 3984, time: 0.064) G_GAN: 0.334 G_GAN_Feat: 0.884 G_ID: 0.208 G_Rec: 0.377 D_GP: 0.027 D_real: 1.107 D_fake: 0.666 +(epoch: 153, iters: 4384, time: 0.064) G_GAN: 0.499 G_GAN_Feat: 0.917 G_ID: 0.139 G_Rec: 0.332 D_GP: 0.042 D_real: 0.688 D_fake: 0.506 +(epoch: 153, iters: 4784, time: 0.064) G_GAN: 0.205 G_GAN_Feat: 1.066 G_ID: 0.199 G_Rec: 0.423 D_GP: 0.047 D_real: 0.345 D_fake: 0.798 +(epoch: 153, iters: 5184, time: 0.064) G_GAN: 0.030 G_GAN_Feat: 0.685 G_ID: 0.171 G_Rec: 0.325 D_GP: 0.022 D_real: 1.024 D_fake: 0.970 +(epoch: 153, iters: 5584, time: 0.064) G_GAN: 0.096 G_GAN_Feat: 1.005 G_ID: 0.236 G_Rec: 0.425 D_GP: 0.040 D_real: 0.537 D_fake: 0.904 +(epoch: 153, iters: 5984, time: 0.064) G_GAN: 0.400 G_GAN_Feat: 0.726 G_ID: 0.135 G_Rec: 0.333 D_GP: 0.025 D_real: 1.226 D_fake: 0.600 +(epoch: 153, iters: 6384, time: 0.064) G_GAN: 0.325 G_GAN_Feat: 0.840 G_ID: 0.188 G_Rec: 0.473 D_GP: 0.021 D_real: 1.055 D_fake: 0.675 +(epoch: 153, iters: 6784, time: 0.064) G_GAN: 0.350 G_GAN_Feat: 0.819 G_ID: 0.157 G_Rec: 0.328 D_GP: 0.026 D_real: 1.101 D_fake: 0.657 +(epoch: 153, iters: 7184, time: 0.064) G_GAN: 0.515 G_GAN_Feat: 0.865 G_ID: 0.180 G_Rec: 0.407 D_GP: 0.021 D_real: 1.310 D_fake: 0.496 +(epoch: 153, iters: 7584, time: 0.064) G_GAN: 0.667 G_GAN_Feat: 1.064 G_ID: 0.148 G_Rec: 0.343 D_GP: 0.630 D_real: 0.558 D_fake: 0.568 +(epoch: 153, iters: 7984, time: 0.064) G_GAN: 0.118 G_GAN_Feat: 0.869 G_ID: 0.205 G_Rec: 0.419 D_GP: 0.021 D_real: 0.821 D_fake: 0.885 +(epoch: 153, iters: 8384, time: 0.064) G_GAN: 0.052 G_GAN_Feat: 0.745 G_ID: 0.119 G_Rec: 0.301 D_GP: 0.054 D_real: 0.797 D_fake: 0.948 +(epoch: 154, iters: 176, time: 0.064) G_GAN: 0.103 G_GAN_Feat: 0.966 G_ID: 0.232 G_Rec: 0.418 D_GP: 0.046 D_real: 0.711 D_fake: 0.898 +(epoch: 154, iters: 576, time: 0.064) G_GAN: 0.277 G_GAN_Feat: 0.872 G_ID: 0.136 G_Rec: 0.326 D_GP: 0.038 D_real: 0.861 D_fake: 0.724 +(epoch: 154, iters: 976, time: 0.064) G_GAN: 0.497 G_GAN_Feat: 0.824 G_ID: 0.210 G_Rec: 0.408 D_GP: 0.022 D_real: 1.232 D_fake: 0.511 +(epoch: 154, iters: 1376, time: 0.064) G_GAN: 0.035 G_GAN_Feat: 0.681 G_ID: 0.167 G_Rec: 0.330 D_GP: 0.026 D_real: 0.951 D_fake: 0.965 +(epoch: 154, iters: 1776, time: 0.064) G_GAN: 0.345 G_GAN_Feat: 0.846 G_ID: 0.190 G_Rec: 0.406 D_GP: 0.034 D_real: 0.915 D_fake: 0.656 +(epoch: 154, iters: 2176, time: 0.064) G_GAN: 0.202 G_GAN_Feat: 0.759 G_ID: 0.139 G_Rec: 0.362 D_GP: 0.038 D_real: 1.043 D_fake: 0.799 +(epoch: 154, iters: 2576, time: 0.064) G_GAN: 0.410 G_GAN_Feat: 0.831 G_ID: 0.210 G_Rec: 0.394 D_GP: 0.028 D_real: 1.110 D_fake: 0.593 +(epoch: 154, iters: 2976, time: 0.064) G_GAN: 0.086 G_GAN_Feat: 0.729 G_ID: 0.160 G_Rec: 0.321 D_GP: 0.036 D_real: 0.952 D_fake: 0.914 +(epoch: 154, iters: 3376, time: 0.064) G_GAN: 0.682 G_GAN_Feat: 0.879 G_ID: 0.192 G_Rec: 0.408 D_GP: 0.028 D_real: 1.454 D_fake: 0.353 +(epoch: 154, iters: 3776, time: 0.064) G_GAN: 0.426 G_GAN_Feat: 0.922 G_ID: 0.151 G_Rec: 0.358 D_GP: 0.047 D_real: 0.824 D_fake: 0.575 +(epoch: 154, iters: 4176, time: 0.064) G_GAN: 0.687 G_GAN_Feat: 0.928 G_ID: 0.196 G_Rec: 0.404 D_GP: 0.031 D_real: 1.212 D_fake: 0.346 +(epoch: 154, iters: 4576, time: 0.064) G_GAN: 0.206 G_GAN_Feat: 0.596 G_ID: 0.143 G_Rec: 0.368 D_GP: 0.018 D_real: 1.198 D_fake: 0.794 +(epoch: 154, iters: 4976, time: 0.064) G_GAN: 0.393 G_GAN_Feat: 0.827 G_ID: 0.214 G_Rec: 0.382 D_GP: 0.026 D_real: 1.071 D_fake: 0.609 +(epoch: 154, iters: 5376, time: 0.064) G_GAN: 0.199 G_GAN_Feat: 0.665 G_ID: 0.132 G_Rec: 0.308 D_GP: 0.022 D_real: 1.092 D_fake: 0.802 +(epoch: 154, iters: 5776, time: 0.064) G_GAN: 0.613 G_GAN_Feat: 0.900 G_ID: 0.180 G_Rec: 0.439 D_GP: 0.025 D_real: 1.196 D_fake: 0.397 +(epoch: 154, iters: 6176, time: 0.064) G_GAN: -0.055 G_GAN_Feat: 0.713 G_ID: 0.167 G_Rec: 0.307 D_GP: 0.031 D_real: 0.806 D_fake: 1.055 +(epoch: 154, iters: 6576, time: 0.064) G_GAN: 0.787 G_GAN_Feat: 0.840 G_ID: 0.235 G_Rec: 0.386 D_GP: 0.024 D_real: 1.544 D_fake: 0.248 +(epoch: 154, iters: 6976, time: 0.064) G_GAN: 0.148 G_GAN_Feat: 0.750 G_ID: 0.172 G_Rec: 0.319 D_GP: 0.036 D_real: 0.913 D_fake: 0.852 +(epoch: 154, iters: 7376, time: 0.064) G_GAN: 0.434 G_GAN_Feat: 0.869 G_ID: 0.194 G_Rec: 0.381 D_GP: 0.037 D_real: 1.064 D_fake: 0.567 +(epoch: 154, iters: 7776, time: 0.064) G_GAN: 0.181 G_GAN_Feat: 0.867 G_ID: 0.176 G_Rec: 0.340 D_GP: 0.059 D_real: 0.782 D_fake: 0.819 +(epoch: 154, iters: 8176, time: 0.064) G_GAN: 0.459 G_GAN_Feat: 1.054 G_ID: 0.232 G_Rec: 0.468 D_GP: 0.024 D_real: 1.043 D_fake: 0.544 +(epoch: 154, iters: 8576, time: 0.064) G_GAN: 0.286 G_GAN_Feat: 0.689 G_ID: 0.143 G_Rec: 0.304 D_GP: 0.025 D_real: 1.174 D_fake: 0.714 +(epoch: 155, iters: 368, time: 0.064) G_GAN: 0.408 G_GAN_Feat: 1.207 G_ID: 0.210 G_Rec: 0.407 D_GP: 0.257 D_real: 0.465 D_fake: 0.596 +(epoch: 155, iters: 768, time: 0.064) G_GAN: 0.222 G_GAN_Feat: 0.651 G_ID: 0.132 G_Rec: 0.304 D_GP: 0.023 D_real: 1.111 D_fake: 0.778 +(epoch: 155, iters: 1168, time: 0.064) G_GAN: 0.221 G_GAN_Feat: 1.075 G_ID: 0.184 G_Rec: 0.464 D_GP: 0.126 D_real: 0.487 D_fake: 0.790 +(epoch: 155, iters: 1568, time: 0.064) G_GAN: 0.264 G_GAN_Feat: 0.642 G_ID: 0.126 G_Rec: 0.300 D_GP: 0.023 D_real: 1.223 D_fake: 0.736 +(epoch: 155, iters: 1968, time: 0.064) G_GAN: 0.768 G_GAN_Feat: 0.837 G_ID: 0.191 G_Rec: 0.409 D_GP: 0.026 D_real: 1.509 D_fake: 0.318 +(epoch: 155, iters: 2368, time: 0.064) G_GAN: 0.414 G_GAN_Feat: 0.785 G_ID: 0.132 G_Rec: 0.321 D_GP: 0.035 D_real: 1.241 D_fake: 0.589 +(epoch: 155, iters: 2768, time: 0.064) G_GAN: 0.599 G_GAN_Feat: 0.853 G_ID: 0.200 G_Rec: 0.382 D_GP: 0.024 D_real: 1.394 D_fake: 0.429 +(epoch: 155, iters: 3168, time: 0.064) G_GAN: 0.287 G_GAN_Feat: 0.711 G_ID: 0.146 G_Rec: 0.341 D_GP: 0.024 D_real: 1.124 D_fake: 0.713 +(epoch: 155, iters: 3568, time: 0.064) G_GAN: 0.455 G_GAN_Feat: 0.918 G_ID: 0.190 G_Rec: 0.388 D_GP: 0.029 D_real: 1.188 D_fake: 0.546 +(epoch: 155, iters: 3968, time: 0.064) G_GAN: 0.190 G_GAN_Feat: 0.752 G_ID: 0.146 G_Rec: 0.332 D_GP: 0.031 D_real: 1.030 D_fake: 0.810 +(epoch: 155, iters: 4368, time: 0.064) G_GAN: 0.880 G_GAN_Feat: 0.833 G_ID: 0.180 G_Rec: 0.362 D_GP: 0.035 D_real: 1.615 D_fake: 0.232 +(epoch: 155, iters: 4768, time: 0.064) G_GAN: 0.429 G_GAN_Feat: 0.677 G_ID: 0.155 G_Rec: 0.344 D_GP: 0.025 D_real: 1.360 D_fake: 0.580 +(epoch: 155, iters: 5168, time: 0.064) G_GAN: 0.191 G_GAN_Feat: 0.920 G_ID: 0.221 G_Rec: 0.451 D_GP: 0.037 D_real: 0.766 D_fake: 0.810 +(epoch: 155, iters: 5568, time: 0.064) G_GAN: 0.557 G_GAN_Feat: 0.916 G_ID: 0.137 G_Rec: 0.356 D_GP: 0.440 D_real: 0.656 D_fake: 0.459 +(epoch: 155, iters: 5968, time: 0.064) G_GAN: 0.532 G_GAN_Feat: 0.871 G_ID: 0.191 G_Rec: 0.399 D_GP: 0.022 D_real: 1.177 D_fake: 0.471 +(epoch: 155, iters: 6368, time: 0.064) G_GAN: 0.250 G_GAN_Feat: 1.064 G_ID: 0.146 G_Rec: 0.314 D_GP: 0.148 D_real: 0.409 D_fake: 0.751 +(epoch: 155, iters: 6768, time: 0.064) G_GAN: 0.480 G_GAN_Feat: 0.867 G_ID: 0.229 G_Rec: 0.437 D_GP: 0.021 D_real: 1.212 D_fake: 0.522 +(epoch: 155, iters: 7168, time: 0.064) G_GAN: 0.356 G_GAN_Feat: 0.659 G_ID: 0.141 G_Rec: 0.297 D_GP: 0.025 D_real: 1.222 D_fake: 0.644 +(epoch: 155, iters: 7568, time: 0.064) G_GAN: 0.354 G_GAN_Feat: 0.819 G_ID: 0.172 G_Rec: 0.391 D_GP: 0.025 D_real: 1.096 D_fake: 0.647 +(epoch: 155, iters: 7968, time: 0.064) G_GAN: 0.323 G_GAN_Feat: 0.905 G_ID: 0.129 G_Rec: 0.336 D_GP: 0.123 D_real: 0.897 D_fake: 0.701 +(epoch: 155, iters: 8368, time: 0.064) G_GAN: 0.578 G_GAN_Feat: 0.873 G_ID: 0.193 G_Rec: 0.395 D_GP: 0.027 D_real: 1.272 D_fake: 0.427 +(epoch: 156, iters: 160, time: 0.064) G_GAN: 0.084 G_GAN_Feat: 0.854 G_ID: 0.183 G_Rec: 0.313 D_GP: 0.038 D_real: 0.600 D_fake: 0.916 +(epoch: 156, iters: 560, time: 0.064) G_GAN: 0.533 G_GAN_Feat: 0.794 G_ID: 0.214 G_Rec: 0.398 D_GP: 0.023 D_real: 1.241 D_fake: 0.530 +(epoch: 156, iters: 960, time: 0.064) G_GAN: -0.086 G_GAN_Feat: 0.748 G_ID: 0.151 G_Rec: 0.336 D_GP: 0.051 D_real: 0.646 D_fake: 1.086 +(epoch: 156, iters: 1360, time: 0.064) G_GAN: 0.666 G_GAN_Feat: 0.907 G_ID: 0.197 G_Rec: 0.382 D_GP: 0.043 D_real: 1.275 D_fake: 0.351 +(epoch: 156, iters: 1760, time: 0.064) G_GAN: 0.217 G_GAN_Feat: 0.803 G_ID: 0.140 G_Rec: 0.322 D_GP: 0.043 D_real: 0.886 D_fake: 0.783 +(epoch: 156, iters: 2160, time: 0.064) G_GAN: 0.329 G_GAN_Feat: 1.125 G_ID: 0.216 G_Rec: 0.450 D_GP: 0.100 D_real: 0.434 D_fake: 0.681 +(epoch: 156, iters: 2560, time: 0.064) G_GAN: 0.484 G_GAN_Feat: 0.792 G_ID: 0.130 G_Rec: 0.314 D_GP: 0.041 D_real: 1.025 D_fake: 0.520 +(epoch: 156, iters: 2960, time: 0.064) G_GAN: 0.638 G_GAN_Feat: 0.742 G_ID: 0.230 G_Rec: 0.417 D_GP: 0.020 D_real: 1.377 D_fake: 0.388 +(epoch: 156, iters: 3360, time: 0.064) G_GAN: 0.269 G_GAN_Feat: 0.577 G_ID: 0.166 G_Rec: 0.287 D_GP: 0.020 D_real: 1.224 D_fake: 0.732 +(epoch: 156, iters: 3760, time: 0.064) G_GAN: 0.324 G_GAN_Feat: 0.792 G_ID: 0.189 G_Rec: 0.415 D_GP: 0.027 D_real: 1.045 D_fake: 0.681 +(epoch: 156, iters: 4160, time: 0.064) G_GAN: 0.121 G_GAN_Feat: 0.594 G_ID: 0.166 G_Rec: 0.294 D_GP: 0.022 D_real: 1.065 D_fake: 0.879 +(epoch: 156, iters: 4560, time: 0.064) G_GAN: 0.434 G_GAN_Feat: 0.851 G_ID: 0.170 G_Rec: 0.405 D_GP: 0.052 D_real: 1.070 D_fake: 0.569 +(epoch: 156, iters: 4960, time: 0.064) G_GAN: 0.140 G_GAN_Feat: 0.739 G_ID: 0.160 G_Rec: 0.369 D_GP: 0.030 D_real: 1.059 D_fake: 0.860 +(epoch: 156, iters: 5360, time: 0.064) G_GAN: 0.481 G_GAN_Feat: 0.957 G_ID: 0.212 G_Rec: 0.437 D_GP: 0.047 D_real: 0.972 D_fake: 0.528 +(epoch: 156, iters: 5760, time: 0.064) G_GAN: 0.181 G_GAN_Feat: 0.675 G_ID: 0.142 G_Rec: 0.321 D_GP: 0.020 D_real: 1.079 D_fake: 0.819 +(epoch: 156, iters: 6160, time: 0.064) G_GAN: 0.478 G_GAN_Feat: 0.901 G_ID: 0.193 G_Rec: 0.389 D_GP: 0.045 D_real: 1.098 D_fake: 0.527 +(epoch: 156, iters: 6560, time: 0.064) G_GAN: 0.280 G_GAN_Feat: 0.808 G_ID: 0.142 G_Rec: 0.306 D_GP: 0.039 D_real: 0.983 D_fake: 0.720 +(epoch: 156, iters: 6960, time: 0.064) G_GAN: 0.530 G_GAN_Feat: 0.784 G_ID: 0.174 G_Rec: 0.396 D_GP: 0.024 D_real: 1.288 D_fake: 0.470 +(epoch: 156, iters: 7360, time: 0.064) G_GAN: 0.314 G_GAN_Feat: 0.914 G_ID: 0.183 G_Rec: 0.360 D_GP: 0.052 D_real: 0.926 D_fake: 0.686 +(epoch: 156, iters: 7760, time: 0.064) G_GAN: 0.415 G_GAN_Feat: 0.914 G_ID: 0.189 G_Rec: 0.410 D_GP: 0.030 D_real: 1.085 D_fake: 0.587 +(epoch: 156, iters: 8160, time: 0.064) G_GAN: 0.431 G_GAN_Feat: 0.820 G_ID: 0.148 G_Rec: 0.315 D_GP: 0.050 D_real: 1.025 D_fake: 0.578 +(epoch: 156, iters: 8560, time: 0.064) G_GAN: 0.483 G_GAN_Feat: 0.920 G_ID: 0.177 G_Rec: 0.385 D_GP: 0.040 D_real: 1.086 D_fake: 0.519 +(epoch: 157, iters: 352, time: 0.064) G_GAN: 0.675 G_GAN_Feat: 0.856 G_ID: 0.147 G_Rec: 0.370 D_GP: 0.081 D_real: 1.232 D_fake: 0.350 +(epoch: 157, iters: 752, time: 0.064) G_GAN: 0.430 G_GAN_Feat: 0.881 G_ID: 0.220 G_Rec: 0.393 D_GP: 0.046 D_real: 1.095 D_fake: 0.576 +(epoch: 157, iters: 1152, time: 0.064) G_GAN: 0.488 G_GAN_Feat: 0.805 G_ID: 0.152 G_Rec: 0.317 D_GP: 0.034 D_real: 1.347 D_fake: 0.532 +(epoch: 157, iters: 1552, time: 0.064) G_GAN: 0.632 G_GAN_Feat: 1.200 G_ID: 0.186 G_Rec: 0.454 D_GP: 0.078 D_real: 0.388 D_fake: 0.423 +(epoch: 157, iters: 1952, time: 0.064) G_GAN: 0.361 G_GAN_Feat: 0.819 G_ID: 0.141 G_Rec: 0.324 D_GP: 0.036 D_real: 1.124 D_fake: 0.645 +(epoch: 157, iters: 2352, time: 0.064) G_GAN: 0.528 G_GAN_Feat: 0.824 G_ID: 0.189 G_Rec: 0.395 D_GP: 0.035 D_real: 1.254 D_fake: 0.492 +(epoch: 157, iters: 2752, time: 0.064) G_GAN: 0.074 G_GAN_Feat: 0.677 G_ID: 0.142 G_Rec: 0.305 D_GP: 0.031 D_real: 1.021 D_fake: 0.926 +(epoch: 157, iters: 3152, time: 0.064) G_GAN: 0.544 G_GAN_Feat: 1.067 G_ID: 0.204 G_Rec: 0.426 D_GP: 0.048 D_real: 1.129 D_fake: 0.522 +(epoch: 157, iters: 3552, time: 0.064) G_GAN: 0.387 G_GAN_Feat: 0.782 G_ID: 0.148 G_Rec: 0.325 D_GP: 0.036 D_real: 1.035 D_fake: 0.617 +(epoch: 157, iters: 3952, time: 0.064) G_GAN: 0.143 G_GAN_Feat: 0.992 G_ID: 0.239 G_Rec: 0.408 D_GP: 0.091 D_real: 0.460 D_fake: 0.857 +(epoch: 157, iters: 4352, time: 0.064) G_GAN: 0.525 G_GAN_Feat: 0.799 G_ID: 0.137 G_Rec: 0.331 D_GP: 0.030 D_real: 1.316 D_fake: 0.503 +(epoch: 157, iters: 4752, time: 0.064) G_GAN: 0.676 G_GAN_Feat: 1.242 G_ID: 0.234 G_Rec: 0.414 D_GP: 0.103 D_real: 0.703 D_fake: 0.338 +(epoch: 157, iters: 5152, time: 0.064) G_GAN: 0.384 G_GAN_Feat: 0.835 G_ID: 0.135 G_Rec: 0.318 D_GP: 0.046 D_real: 1.068 D_fake: 0.620 +(epoch: 157, iters: 5552, time: 0.064) G_GAN: 0.421 G_GAN_Feat: 1.192 G_ID: 0.199 G_Rec: 0.483 D_GP: 0.165 D_real: 0.447 D_fake: 0.585 +(epoch: 157, iters: 5952, time: 0.064) G_GAN: -0.126 G_GAN_Feat: 0.763 G_ID: 0.132 G_Rec: 0.333 D_GP: 0.026 D_real: 0.718 D_fake: 1.126 +(epoch: 157, iters: 6352, time: 0.064) G_GAN: 0.127 G_GAN_Feat: 0.856 G_ID: 0.200 G_Rec: 0.397 D_GP: 0.027 D_real: 0.960 D_fake: 0.874 +(epoch: 157, iters: 6752, time: 0.064) G_GAN: 0.036 G_GAN_Feat: 0.664 G_ID: 0.167 G_Rec: 0.324 D_GP: 0.023 D_real: 0.932 D_fake: 0.964 +(epoch: 157, iters: 7152, time: 0.064) G_GAN: 0.467 G_GAN_Feat: 0.982 G_ID: 0.211 G_Rec: 0.438 D_GP: 0.045 D_real: 0.880 D_fake: 0.552 +(epoch: 157, iters: 7552, time: 0.064) G_GAN: 0.307 G_GAN_Feat: 0.724 G_ID: 0.126 G_Rec: 0.314 D_GP: 0.022 D_real: 1.172 D_fake: 0.694 +(epoch: 157, iters: 7952, time: 0.064) G_GAN: 0.479 G_GAN_Feat: 0.927 G_ID: 0.224 G_Rec: 0.401 D_GP: 0.028 D_real: 1.109 D_fake: 0.522 +(epoch: 157, iters: 8352, time: 0.064) G_GAN: 0.383 G_GAN_Feat: 0.777 G_ID: 0.143 G_Rec: 0.317 D_GP: 0.052 D_real: 1.149 D_fake: 0.618 +(epoch: 158, iters: 144, time: 0.064) G_GAN: 0.325 G_GAN_Feat: 0.938 G_ID: 0.215 G_Rec: 0.461 D_GP: 0.043 D_real: 0.928 D_fake: 0.677 +(epoch: 158, iters: 544, time: 0.064) G_GAN: 0.248 G_GAN_Feat: 0.730 G_ID: 0.157 G_Rec: 0.337 D_GP: 0.070 D_real: 1.008 D_fake: 0.758 +(epoch: 158, iters: 944, time: 0.064) G_GAN: 0.568 G_GAN_Feat: 1.000 G_ID: 0.206 G_Rec: 0.409 D_GP: 0.114 D_real: 0.860 D_fake: 0.436 +(epoch: 158, iters: 1344, time: 0.064) G_GAN: 0.435 G_GAN_Feat: 0.890 G_ID: 0.128 G_Rec: 0.351 D_GP: 0.112 D_real: 0.770 D_fake: 0.571 +(epoch: 158, iters: 1744, time: 0.064) G_GAN: 0.228 G_GAN_Feat: 0.941 G_ID: 0.207 G_Rec: 0.411 D_GP: 0.033 D_real: 0.929 D_fake: 0.773 +(epoch: 158, iters: 2144, time: 0.064) G_GAN: 0.193 G_GAN_Feat: 0.781 G_ID: 0.150 G_Rec: 0.315 D_GP: 0.028 D_real: 0.927 D_fake: 0.807 +(epoch: 158, iters: 2544, time: 0.064) G_GAN: 0.706 G_GAN_Feat: 1.024 G_ID: 0.218 G_Rec: 0.411 D_GP: 0.080 D_real: 0.984 D_fake: 0.371 +(epoch: 158, iters: 2944, time: 0.064) G_GAN: 0.013 G_GAN_Feat: 0.664 G_ID: 0.140 G_Rec: 0.323 D_GP: 0.022 D_real: 1.002 D_fake: 0.987 +(epoch: 158, iters: 3344, time: 0.064) G_GAN: 0.579 G_GAN_Feat: 0.784 G_ID: 0.189 G_Rec: 0.380 D_GP: 0.025 D_real: 1.289 D_fake: 0.433 +(epoch: 158, iters: 3744, time: 0.064) G_GAN: 0.090 G_GAN_Feat: 0.882 G_ID: 0.137 G_Rec: 0.389 D_GP: 0.113 D_real: 0.410 D_fake: 0.910 +(epoch: 158, iters: 4144, time: 0.064) G_GAN: 0.638 G_GAN_Feat: 1.228 G_ID: 0.237 G_Rec: 0.451 D_GP: 0.064 D_real: 1.043 D_fake: 0.674 +(epoch: 158, iters: 4544, time: 0.064) G_GAN: 0.483 G_GAN_Feat: 0.921 G_ID: 0.146 G_Rec: 0.350 D_GP: 0.119 D_real: 0.717 D_fake: 0.533 +(epoch: 158, iters: 4944, time: 0.064) G_GAN: 0.264 G_GAN_Feat: 0.852 G_ID: 0.208 G_Rec: 0.411 D_GP: 0.025 D_real: 1.027 D_fake: 0.737 +(epoch: 158, iters: 5344, time: 0.064) G_GAN: 0.088 G_GAN_Feat: 0.737 G_ID: 0.169 G_Rec: 0.359 D_GP: 0.030 D_real: 0.865 D_fake: 0.912 +(epoch: 158, iters: 5744, time: 0.064) G_GAN: 0.282 G_GAN_Feat: 0.795 G_ID: 0.198 G_Rec: 0.374 D_GP: 0.023 D_real: 1.066 D_fake: 0.719 +(epoch: 158, iters: 6144, time: 0.064) G_GAN: 0.142 G_GAN_Feat: 0.678 G_ID: 0.124 G_Rec: 0.321 D_GP: 0.037 D_real: 1.049 D_fake: 0.858 +(epoch: 158, iters: 6544, time: 0.064) G_GAN: 0.337 G_GAN_Feat: 0.873 G_ID: 0.190 G_Rec: 0.390 D_GP: 0.029 D_real: 1.010 D_fake: 0.664 +(epoch: 158, iters: 6944, time: 0.064) G_GAN: 0.362 G_GAN_Feat: 0.749 G_ID: 0.125 G_Rec: 0.335 D_GP: 0.030 D_real: 1.207 D_fake: 0.638 +(epoch: 158, iters: 7344, time: 0.064) G_GAN: 0.445 G_GAN_Feat: 0.970 G_ID: 0.196 G_Rec: 0.376 D_GP: 0.059 D_real: 0.797 D_fake: 0.565 +(epoch: 158, iters: 7744, time: 0.064) G_GAN: 0.011 G_GAN_Feat: 0.842 G_ID: 0.180 G_Rec: 0.326 D_GP: 0.039 D_real: 0.875 D_fake: 1.013 +(epoch: 158, iters: 8144, time: 0.064) G_GAN: 0.909 G_GAN_Feat: 1.059 G_ID: 0.198 G_Rec: 0.438 D_GP: 0.033 D_real: 1.252 D_fake: 0.171 +(epoch: 158, iters: 8544, time: 0.064) G_GAN: 0.342 G_GAN_Feat: 0.716 G_ID: 0.152 G_Rec: 0.329 D_GP: 0.023 D_real: 1.149 D_fake: 0.661 +(epoch: 159, iters: 336, time: 0.064) G_GAN: 0.411 G_GAN_Feat: 0.898 G_ID: 0.168 G_Rec: 0.435 D_GP: 0.031 D_real: 1.041 D_fake: 0.593 +(epoch: 159, iters: 736, time: 0.064) G_GAN: 0.365 G_GAN_Feat: 0.678 G_ID: 0.137 G_Rec: 0.320 D_GP: 0.026 D_real: 1.175 D_fake: 0.637 +(epoch: 159, iters: 1136, time: 0.064) G_GAN: 0.427 G_GAN_Feat: 0.773 G_ID: 0.209 G_Rec: 0.390 D_GP: 0.025 D_real: 1.171 D_fake: 0.576 +(epoch: 159, iters: 1536, time: 0.064) G_GAN: 0.197 G_GAN_Feat: 0.628 G_ID: 0.140 G_Rec: 0.287 D_GP: 0.025 D_real: 1.123 D_fake: 0.803 +(epoch: 159, iters: 1936, time: 0.064) G_GAN: 0.377 G_GAN_Feat: 0.819 G_ID: 0.207 G_Rec: 0.422 D_GP: 0.020 D_real: 1.111 D_fake: 0.624 +(epoch: 159, iters: 2336, time: 0.064) G_GAN: 0.179 G_GAN_Feat: 0.679 G_ID: 0.148 G_Rec: 0.296 D_GP: 0.032 D_real: 0.957 D_fake: 0.822 +(epoch: 159, iters: 2736, time: 0.064) G_GAN: 0.601 G_GAN_Feat: 1.109 G_ID: 0.226 G_Rec: 0.442 D_GP: 0.041 D_real: 1.181 D_fake: 0.448 +(epoch: 159, iters: 3136, time: 0.064) G_GAN: 0.534 G_GAN_Feat: 0.748 G_ID: 0.128 G_Rec: 0.305 D_GP: 0.032 D_real: 1.271 D_fake: 0.469 +(epoch: 159, iters: 3536, time: 0.064) G_GAN: 0.579 G_GAN_Feat: 1.004 G_ID: 0.214 G_Rec: 0.409 D_GP: 0.037 D_real: 1.034 D_fake: 0.434 +(epoch: 159, iters: 3936, time: 0.064) G_GAN: 0.348 G_GAN_Feat: 0.904 G_ID: 0.154 G_Rec: 0.369 D_GP: 0.065 D_real: 0.794 D_fake: 0.656 +(epoch: 159, iters: 4336, time: 0.064) G_GAN: 0.284 G_GAN_Feat: 0.752 G_ID: 0.227 G_Rec: 0.370 D_GP: 0.024 D_real: 1.097 D_fake: 0.720 +(epoch: 159, iters: 4736, time: 0.064) G_GAN: 0.186 G_GAN_Feat: 0.643 G_ID: 0.152 G_Rec: 0.322 D_GP: 0.027 D_real: 1.091 D_fake: 0.814 +(epoch: 159, iters: 5136, time: 0.064) G_GAN: 0.219 G_GAN_Feat: 0.796 G_ID: 0.202 G_Rec: 0.399 D_GP: 0.024 D_real: 1.010 D_fake: 0.781 +(epoch: 159, iters: 5536, time: 0.064) G_GAN: 0.037 G_GAN_Feat: 0.680 G_ID: 0.139 G_Rec: 0.308 D_GP: 0.032 D_real: 0.965 D_fake: 0.963 +(epoch: 159, iters: 5936, time: 0.064) G_GAN: 0.312 G_GAN_Feat: 0.750 G_ID: 0.188 G_Rec: 0.355 D_GP: 0.028 D_real: 1.144 D_fake: 0.689 +(epoch: 159, iters: 6336, time: 0.064) G_GAN: 0.059 G_GAN_Feat: 0.758 G_ID: 0.136 G_Rec: 0.332 D_GP: 0.053 D_real: 0.860 D_fake: 0.941 +(epoch: 159, iters: 6736, time: 0.064) G_GAN: 0.180 G_GAN_Feat: 0.906 G_ID: 0.202 G_Rec: 0.375 D_GP: 0.056 D_real: 0.990 D_fake: 0.820 +(epoch: 159, iters: 7136, time: 0.064) G_GAN: 0.126 G_GAN_Feat: 0.663 G_ID: 0.123 G_Rec: 0.298 D_GP: 0.028 D_real: 1.061 D_fake: 0.874 +(epoch: 159, iters: 7536, time: 0.064) G_GAN: 0.392 G_GAN_Feat: 0.910 G_ID: 0.208 G_Rec: 0.404 D_GP: 0.029 D_real: 1.137 D_fake: 0.609 +(epoch: 159, iters: 7936, time: 0.064) G_GAN: 0.207 G_GAN_Feat: 0.676 G_ID: 0.133 G_Rec: 0.314 D_GP: 0.027 D_real: 1.100 D_fake: 0.793 +(epoch: 159, iters: 8336, time: 0.064) G_GAN: 0.427 G_GAN_Feat: 0.852 G_ID: 0.211 G_Rec: 0.401 D_GP: 0.029 D_real: 1.118 D_fake: 0.580 +(epoch: 160, iters: 128, time: 0.064) G_GAN: 0.049 G_GAN_Feat: 0.768 G_ID: 0.123 G_Rec: 0.340 D_GP: 0.059 D_real: 0.899 D_fake: 0.951 +(epoch: 160, iters: 528, time: 0.064) G_GAN: 0.363 G_GAN_Feat: 0.964 G_ID: 0.203 G_Rec: 0.455 D_GP: 0.046 D_real: 0.871 D_fake: 0.652 +(epoch: 160, iters: 928, time: 0.064) G_GAN: 0.202 G_GAN_Feat: 1.082 G_ID: 0.139 G_Rec: 0.353 D_GP: 0.036 D_real: 0.824 D_fake: 0.798 +(epoch: 160, iters: 1328, time: 0.064) G_GAN: 0.600 G_GAN_Feat: 0.987 G_ID: 0.182 G_Rec: 0.382 D_GP: 0.031 D_real: 1.209 D_fake: 0.410 +(epoch: 160, iters: 1728, time: 0.064) G_GAN: 0.044 G_GAN_Feat: 0.917 G_ID: 0.153 G_Rec: 0.346 D_GP: 0.225 D_real: 0.468 D_fake: 0.956 +(epoch: 160, iters: 2128, time: 0.064) G_GAN: 0.276 G_GAN_Feat: 0.874 G_ID: 0.194 G_Rec: 0.393 D_GP: 0.029 D_real: 0.924 D_fake: 0.724 +(epoch: 160, iters: 2528, time: 0.064) G_GAN: 0.251 G_GAN_Feat: 0.737 G_ID: 0.116 G_Rec: 0.417 D_GP: 0.028 D_real: 0.971 D_fake: 0.749 +(epoch: 160, iters: 2928, time: 0.064) G_GAN: 0.414 G_GAN_Feat: 0.890 G_ID: 0.191 G_Rec: 0.413 D_GP: 0.030 D_real: 1.123 D_fake: 0.591 +(epoch: 160, iters: 3328, time: 0.064) G_GAN: 0.281 G_GAN_Feat: 0.703 G_ID: 0.162 G_Rec: 0.326 D_GP: 0.021 D_real: 1.214 D_fake: 0.719 +(epoch: 160, iters: 3728, time: 0.064) G_GAN: 0.454 G_GAN_Feat: 0.842 G_ID: 0.193 G_Rec: 0.398 D_GP: 0.028 D_real: 1.179 D_fake: 0.560 +(epoch: 160, iters: 4128, time: 0.064) G_GAN: 0.390 G_GAN_Feat: 0.739 G_ID: 0.138 G_Rec: 0.356 D_GP: 0.035 D_real: 1.221 D_fake: 0.611 +(epoch: 160, iters: 4528, time: 0.064) G_GAN: 0.292 G_GAN_Feat: 1.166 G_ID: 0.198 G_Rec: 0.421 D_GP: 0.244 D_real: 0.302 D_fake: 0.709 +(epoch: 160, iters: 4928, time: 0.064) G_GAN: 0.298 G_GAN_Feat: 0.845 G_ID: 0.143 G_Rec: 0.327 D_GP: 0.093 D_real: 0.785 D_fake: 0.706 +(epoch: 160, iters: 5328, time: 0.064) G_GAN: 0.289 G_GAN_Feat: 0.835 G_ID: 0.201 G_Rec: 0.399 D_GP: 0.021 D_real: 1.108 D_fake: 0.713 +(epoch: 160, iters: 5728, time: 0.064) G_GAN: 0.251 G_GAN_Feat: 0.645 G_ID: 0.142 G_Rec: 0.319 D_GP: 0.033 D_real: 1.092 D_fake: 0.749 +(epoch: 160, iters: 6128, time: 0.064) G_GAN: 0.218 G_GAN_Feat: 0.858 G_ID: 0.205 G_Rec: 0.395 D_GP: 0.029 D_real: 0.995 D_fake: 0.782 +(epoch: 160, iters: 6528, time: 0.064) G_GAN: 0.127 G_GAN_Feat: 0.768 G_ID: 0.136 G_Rec: 0.305 D_GP: 0.109 D_real: 0.837 D_fake: 0.873 +(epoch: 160, iters: 6928, time: 0.064) G_GAN: 0.517 G_GAN_Feat: 0.996 G_ID: 0.198 G_Rec: 0.422 D_GP: 0.035 D_real: 1.105 D_fake: 0.492 +(epoch: 160, iters: 7328, time: 0.064) G_GAN: 0.093 G_GAN_Feat: 0.849 G_ID: 0.137 G_Rec: 0.404 D_GP: 0.052 D_real: 0.714 D_fake: 0.908 +(epoch: 160, iters: 7728, time: 0.064) G_GAN: 0.397 G_GAN_Feat: 0.981 G_ID: 0.179 G_Rec: 0.424 D_GP: 0.067 D_real: 0.784 D_fake: 0.616 +(epoch: 160, iters: 8128, time: 0.064) G_GAN: 0.323 G_GAN_Feat: 0.850 G_ID: 0.134 G_Rec: 0.320 D_GP: 0.116 D_real: 0.848 D_fake: 0.679 +(epoch: 160, iters: 8528, time: 0.064) G_GAN: 0.583 G_GAN_Feat: 0.883 G_ID: 0.163 G_Rec: 0.410 D_GP: 0.025 D_real: 1.338 D_fake: 0.422 +(epoch: 161, iters: 320, time: 0.064) G_GAN: 0.213 G_GAN_Feat: 0.741 G_ID: 0.147 G_Rec: 0.307 D_GP: 0.041 D_real: 0.940 D_fake: 0.787 +(epoch: 161, iters: 720, time: 0.064) G_GAN: 0.214 G_GAN_Feat: 0.872 G_ID: 0.190 G_Rec: 0.376 D_GP: 0.039 D_real: 1.028 D_fake: 0.786 +(epoch: 161, iters: 1120, time: 0.064) G_GAN: 0.232 G_GAN_Feat: 0.882 G_ID: 0.173 G_Rec: 0.340 D_GP: 0.029 D_real: 1.091 D_fake: 0.774 +(epoch: 161, iters: 1520, time: 0.064) G_GAN: 0.779 G_GAN_Feat: 1.058 G_ID: 0.190 G_Rec: 0.444 D_GP: 0.032 D_real: 1.155 D_fake: 0.251 +(epoch: 161, iters: 1920, time: 0.064) G_GAN: -0.040 G_GAN_Feat: 0.896 G_ID: 0.152 G_Rec: 0.338 D_GP: 0.048 D_real: 0.418 D_fake: 1.040 +(epoch: 161, iters: 2320, time: 0.064) G_GAN: 0.423 G_GAN_Feat: 0.798 G_ID: 0.208 G_Rec: 0.406 D_GP: 0.023 D_real: 1.194 D_fake: 0.585 +(epoch: 161, iters: 2720, time: 0.064) G_GAN: 0.119 G_GAN_Feat: 0.644 G_ID: 0.145 G_Rec: 0.321 D_GP: 0.026 D_real: 1.062 D_fake: 0.881 +(epoch: 161, iters: 3120, time: 0.064) G_GAN: 0.521 G_GAN_Feat: 0.812 G_ID: 0.168 G_Rec: 0.393 D_GP: 0.031 D_real: 1.304 D_fake: 0.491 +(epoch: 161, iters: 3520, time: 0.064) G_GAN: 0.138 G_GAN_Feat: 0.728 G_ID: 0.141 G_Rec: 0.329 D_GP: 0.046 D_real: 1.002 D_fake: 0.862 +(epoch: 161, iters: 3920, time: 0.064) G_GAN: 0.175 G_GAN_Feat: 0.855 G_ID: 0.211 G_Rec: 0.411 D_GP: 0.027 D_real: 0.893 D_fake: 0.825 +(epoch: 161, iters: 4320, time: 0.064) G_GAN: 0.222 G_GAN_Feat: 0.704 G_ID: 0.149 G_Rec: 0.302 D_GP: 0.025 D_real: 1.133 D_fake: 0.778 +(epoch: 161, iters: 4720, time: 0.064) G_GAN: 0.426 G_GAN_Feat: 0.937 G_ID: 0.185 G_Rec: 0.454 D_GP: 0.055 D_real: 0.986 D_fake: 0.581 +(epoch: 161, iters: 5120, time: 0.064) G_GAN: 0.265 G_GAN_Feat: 0.791 G_ID: 0.162 G_Rec: 0.361 D_GP: 0.081 D_real: 0.930 D_fake: 0.736 +(epoch: 161, iters: 5520, time: 0.064) G_GAN: 0.144 G_GAN_Feat: 0.904 G_ID: 0.211 G_Rec: 0.393 D_GP: 0.036 D_real: 0.819 D_fake: 0.857 +(epoch: 161, iters: 5920, time: 0.064) G_GAN: 0.215 G_GAN_Feat: 0.754 G_ID: 0.159 G_Rec: 0.370 D_GP: 0.039 D_real: 0.911 D_fake: 0.785 +(epoch: 161, iters: 6320, time: 0.064) G_GAN: 0.337 G_GAN_Feat: 0.866 G_ID: 0.193 G_Rec: 0.408 D_GP: 0.052 D_real: 0.962 D_fake: 0.665 +(epoch: 161, iters: 6720, time: 0.064) G_GAN: 0.188 G_GAN_Feat: 0.742 G_ID: 0.142 G_Rec: 0.345 D_GP: 0.025 D_real: 1.092 D_fake: 0.815 +(epoch: 161, iters: 7120, time: 0.064) G_GAN: 0.536 G_GAN_Feat: 0.951 G_ID: 0.247 G_Rec: 0.393 D_GP: 0.050 D_real: 0.966 D_fake: 0.478 +(epoch: 161, iters: 7520, time: 0.064) G_GAN: 0.084 G_GAN_Feat: 0.881 G_ID: 0.155 G_Rec: 0.342 D_GP: 0.033 D_real: 0.731 D_fake: 0.916 +(epoch: 161, iters: 7920, time: 0.064) G_GAN: 0.656 G_GAN_Feat: 0.993 G_ID: 0.218 G_Rec: 0.435 D_GP: 0.039 D_real: 1.069 D_fake: 0.368 +(epoch: 161, iters: 8320, time: 0.064) G_GAN: 0.329 G_GAN_Feat: 0.650 G_ID: 0.133 G_Rec: 0.290 D_GP: 0.020 D_real: 1.297 D_fake: 0.671 +(epoch: 162, iters: 112, time: 0.064) G_GAN: 0.346 G_GAN_Feat: 0.976 G_ID: 0.225 G_Rec: 0.428 D_GP: 0.071 D_real: 0.720 D_fake: 0.655 +(epoch: 162, iters: 512, time: 0.064) G_GAN: 0.249 G_GAN_Feat: 0.670 G_ID: 0.172 G_Rec: 0.299 D_GP: 0.021 D_real: 1.191 D_fake: 0.751 +(epoch: 162, iters: 912, time: 0.064) G_GAN: 0.362 G_GAN_Feat: 0.839 G_ID: 0.240 G_Rec: 0.379 D_GP: 0.046 D_real: 1.001 D_fake: 0.640 +(epoch: 162, iters: 1312, time: 0.064) G_GAN: 0.697 G_GAN_Feat: 0.774 G_ID: 0.138 G_Rec: 0.342 D_GP: 0.025 D_real: 1.562 D_fake: 0.422 +(epoch: 162, iters: 1712, time: 0.064) G_GAN: 0.212 G_GAN_Feat: 1.069 G_ID: 0.231 G_Rec: 0.405 D_GP: 0.065 D_real: 0.415 D_fake: 0.788 +(epoch: 162, iters: 2112, time: 0.064) G_GAN: 0.376 G_GAN_Feat: 0.782 G_ID: 0.165 G_Rec: 0.333 D_GP: 0.057 D_real: 1.025 D_fake: 0.645 +(epoch: 162, iters: 2512, time: 0.064) G_GAN: 0.505 G_GAN_Feat: 0.945 G_ID: 0.187 G_Rec: 0.427 D_GP: 0.049 D_real: 1.061 D_fake: 0.512 +(epoch: 162, iters: 2912, time: 0.064) G_GAN: 0.083 G_GAN_Feat: 0.716 G_ID: 0.154 G_Rec: 0.351 D_GP: 0.021 D_real: 0.989 D_fake: 0.917 +(epoch: 162, iters: 3312, time: 0.064) G_GAN: 0.350 G_GAN_Feat: 1.468 G_ID: 0.177 G_Rec: 0.454 D_GP: 0.124 D_real: 0.483 D_fake: 0.659 +(epoch: 162, iters: 3712, time: 0.064) G_GAN: 0.403 G_GAN_Feat: 0.804 G_ID: 0.156 G_Rec: 0.323 D_GP: 0.085 D_real: 1.005 D_fake: 0.598 +(epoch: 162, iters: 4112, time: 0.064) G_GAN: 0.347 G_GAN_Feat: 0.868 G_ID: 0.209 G_Rec: 0.429 D_GP: 0.035 D_real: 1.015 D_fake: 0.653 +(epoch: 162, iters: 4512, time: 0.064) G_GAN: 0.172 G_GAN_Feat: 0.732 G_ID: 0.145 G_Rec: 0.296 D_GP: 0.067 D_real: 0.864 D_fake: 0.828 +(epoch: 162, iters: 4912, time: 0.064) G_GAN: 0.599 G_GAN_Feat: 0.895 G_ID: 0.212 G_Rec: 0.444 D_GP: 0.032 D_real: 1.281 D_fake: 0.409 +(epoch: 162, iters: 5312, time: 0.064) G_GAN: 0.064 G_GAN_Feat: 0.639 G_ID: 0.159 G_Rec: 0.312 D_GP: 0.023 D_real: 1.002 D_fake: 0.936 +(epoch: 162, iters: 5712, time: 0.064) G_GAN: 0.326 G_GAN_Feat: 0.855 G_ID: 0.167 G_Rec: 0.362 D_GP: 0.046 D_real: 0.956 D_fake: 0.674 +(epoch: 162, iters: 6112, time: 0.064) G_GAN: 0.166 G_GAN_Feat: 0.681 G_ID: 0.139 G_Rec: 0.317 D_GP: 0.033 D_real: 1.058 D_fake: 0.834 +(epoch: 162, iters: 6512, time: 0.064) G_GAN: 0.443 G_GAN_Feat: 0.998 G_ID: 0.225 G_Rec: 0.422 D_GP: 0.135 D_real: 0.664 D_fake: 0.562 +(epoch: 162, iters: 6912, time: 0.064) G_GAN: 0.200 G_GAN_Feat: 0.842 G_ID: 0.166 G_Rec: 0.340 D_GP: 0.057 D_real: 0.740 D_fake: 0.800 +(epoch: 162, iters: 7312, time: 0.064) G_GAN: 0.628 G_GAN_Feat: 0.977 G_ID: 0.198 G_Rec: 0.392 D_GP: 0.037 D_real: 1.004 D_fake: 0.377 +(epoch: 162, iters: 7712, time: 0.064) G_GAN: 0.164 G_GAN_Feat: 0.609 G_ID: 0.138 G_Rec: 0.294 D_GP: 0.021 D_real: 1.123 D_fake: 0.836 +(epoch: 162, iters: 8112, time: 0.064) G_GAN: 0.212 G_GAN_Feat: 0.829 G_ID: 0.199 G_Rec: 0.381 D_GP: 0.030 D_real: 0.842 D_fake: 0.789 +(epoch: 162, iters: 8512, time: 0.064) G_GAN: 0.125 G_GAN_Feat: 0.802 G_ID: 0.144 G_Rec: 0.340 D_GP: 0.122 D_real: 0.763 D_fake: 0.879 +(epoch: 163, iters: 304, time: 0.064) G_GAN: 0.451 G_GAN_Feat: 0.941 G_ID: 0.199 G_Rec: 0.406 D_GP: 0.044 D_real: 1.051 D_fake: 0.550 +(epoch: 163, iters: 704, time: 0.064) G_GAN: 0.068 G_GAN_Feat: 0.823 G_ID: 0.140 G_Rec: 0.360 D_GP: 0.047 D_real: 0.638 D_fake: 0.932 +(epoch: 163, iters: 1104, time: 0.064) G_GAN: 0.231 G_GAN_Feat: 0.988 G_ID: 0.209 G_Rec: 0.444 D_GP: 0.079 D_real: 0.771 D_fake: 0.775 +(epoch: 163, iters: 1504, time: 0.064) G_GAN: 0.275 G_GAN_Feat: 0.732 G_ID: 0.137 G_Rec: 0.319 D_GP: 0.025 D_real: 1.151 D_fake: 0.725 +(epoch: 163, iters: 1904, time: 0.064) G_GAN: 0.344 G_GAN_Feat: 0.916 G_ID: 0.239 G_Rec: 0.438 D_GP: 0.022 D_real: 1.035 D_fake: 0.656 +(epoch: 163, iters: 2304, time: 0.064) G_GAN: 0.374 G_GAN_Feat: 0.989 G_ID: 0.146 G_Rec: 0.342 D_GP: 0.054 D_real: 0.730 D_fake: 0.631 +(epoch: 163, iters: 2704, time: 0.064) G_GAN: 0.429 G_GAN_Feat: 0.877 G_ID: 0.217 G_Rec: 0.433 D_GP: 0.029 D_real: 1.081 D_fake: 0.573 +(epoch: 163, iters: 3104, time: 0.064) G_GAN: 0.246 G_GAN_Feat: 0.906 G_ID: 0.185 G_Rec: 0.353 D_GP: 0.053 D_real: 0.692 D_fake: 0.781 +(epoch: 163, iters: 3504, time: 0.064) G_GAN: 0.920 G_GAN_Feat: 0.931 G_ID: 0.190 G_Rec: 0.425 D_GP: 0.029 D_real: 1.655 D_fake: 0.167 +(epoch: 163, iters: 3904, time: 0.064) G_GAN: 0.143 G_GAN_Feat: 0.771 G_ID: 0.143 G_Rec: 0.325 D_GP: 0.035 D_real: 1.061 D_fake: 0.857 +(epoch: 163, iters: 4304, time: 0.064) G_GAN: 0.518 G_GAN_Feat: 1.005 G_ID: 0.193 G_Rec: 0.420 D_GP: 0.096 D_real: 0.856 D_fake: 0.495 +(epoch: 163, iters: 4704, time: 0.064) G_GAN: 0.459 G_GAN_Feat: 0.895 G_ID: 0.163 G_Rec: 0.323 D_GP: 0.072 D_real: 1.021 D_fake: 0.573 +(epoch: 163, iters: 5104, time: 0.064) G_GAN: 0.304 G_GAN_Feat: 0.840 G_ID: 0.207 G_Rec: 0.374 D_GP: 0.026 D_real: 1.133 D_fake: 0.701 +(epoch: 163, iters: 5504, time: 0.064) G_GAN: 0.107 G_GAN_Feat: 0.765 G_ID: 0.133 G_Rec: 0.322 D_GP: 0.028 D_real: 0.945 D_fake: 0.893 +(epoch: 163, iters: 5904, time: 0.064) G_GAN: 0.450 G_GAN_Feat: 0.924 G_ID: 0.201 G_Rec: 0.408 D_GP: 0.048 D_real: 1.060 D_fake: 0.552 +(epoch: 163, iters: 6304, time: 0.064) G_GAN: 0.224 G_GAN_Feat: 0.694 G_ID: 0.153 G_Rec: 0.322 D_GP: 0.022 D_real: 1.087 D_fake: 0.776 +(epoch: 163, iters: 6704, time: 0.064) G_GAN: 0.382 G_GAN_Feat: 0.931 G_ID: 0.196 G_Rec: 0.482 D_GP: 0.022 D_real: 1.107 D_fake: 0.619 +(epoch: 163, iters: 7104, time: 0.064) G_GAN: -0.076 G_GAN_Feat: 0.861 G_ID: 0.148 G_Rec: 0.348 D_GP: 0.085 D_real: 0.470 D_fake: 1.076 +(epoch: 163, iters: 7504, time: 0.064) G_GAN: 0.615 G_GAN_Feat: 0.941 G_ID: 0.213 G_Rec: 0.428 D_GP: 0.031 D_real: 1.325 D_fake: 0.425 +(epoch: 163, iters: 7904, time: 0.064) G_GAN: 0.107 G_GAN_Feat: 0.805 G_ID: 0.126 G_Rec: 0.383 D_GP: 0.035 D_real: 0.805 D_fake: 0.893 +(epoch: 163, iters: 8304, time: 0.064) G_GAN: 0.413 G_GAN_Feat: 0.980 G_ID: 0.188 G_Rec: 0.399 D_GP: 0.077 D_real: 1.070 D_fake: 0.607 +(epoch: 164, iters: 96, time: 0.064) G_GAN: 0.301 G_GAN_Feat: 0.908 G_ID: 0.147 G_Rec: 0.361 D_GP: 0.088 D_real: 0.568 D_fake: 0.718 +(epoch: 164, iters: 496, time: 0.064) G_GAN: 0.448 G_GAN_Feat: 0.834 G_ID: 0.206 G_Rec: 0.396 D_GP: 0.021 D_real: 1.184 D_fake: 0.554 +(epoch: 164, iters: 896, time: 0.064) G_GAN: 0.268 G_GAN_Feat: 0.706 G_ID: 0.144 G_Rec: 0.302 D_GP: 0.030 D_real: 1.135 D_fake: 0.732 +(epoch: 164, iters: 1296, time: 0.064) G_GAN: 0.343 G_GAN_Feat: 0.877 G_ID: 0.212 G_Rec: 0.385 D_GP: 0.026 D_real: 1.121 D_fake: 0.658 +(epoch: 164, iters: 1696, time: 0.064) G_GAN: 0.359 G_GAN_Feat: 0.747 G_ID: 0.137 G_Rec: 0.292 D_GP: 0.039 D_real: 1.112 D_fake: 0.642 +(epoch: 164, iters: 2096, time: 0.064) G_GAN: 0.173 G_GAN_Feat: 0.763 G_ID: 0.230 G_Rec: 0.386 D_GP: 0.023 D_real: 0.988 D_fake: 0.834 +(epoch: 164, iters: 2496, time: 0.064) G_GAN: 0.116 G_GAN_Feat: 0.680 G_ID: 0.155 G_Rec: 0.324 D_GP: 0.035 D_real: 0.960 D_fake: 0.884 +(epoch: 164, iters: 2896, time: 0.064) G_GAN: 0.301 G_GAN_Feat: 0.855 G_ID: 0.214 G_Rec: 0.412 D_GP: 0.034 D_real: 0.987 D_fake: 0.702 +(epoch: 164, iters: 3296, time: 0.064) G_GAN: 0.374 G_GAN_Feat: 0.641 G_ID: 0.161 G_Rec: 0.285 D_GP: 0.025 D_real: 1.267 D_fake: 0.629 +(epoch: 164, iters: 3696, time: 0.064) G_GAN: 0.370 G_GAN_Feat: 1.061 G_ID: 0.199 G_Rec: 0.520 D_GP: 0.031 D_real: 0.904 D_fake: 0.632 +(epoch: 164, iters: 4096, time: 0.064) G_GAN: 0.192 G_GAN_Feat: 0.746 G_ID: 0.137 G_Rec: 0.311 D_GP: 0.034 D_real: 0.898 D_fake: 0.808 +(epoch: 164, iters: 4496, time: 0.064) G_GAN: 0.566 G_GAN_Feat: 0.953 G_ID: 0.198 G_Rec: 0.392 D_GP: 0.044 D_real: 1.037 D_fake: 0.442 +(epoch: 164, iters: 4896, time: 0.064) G_GAN: 0.368 G_GAN_Feat: 0.812 G_ID: 0.132 G_Rec: 0.337 D_GP: 0.026 D_real: 1.119 D_fake: 0.633 +(epoch: 164, iters: 5296, time: 0.064) G_GAN: 0.511 G_GAN_Feat: 0.982 G_ID: 0.193 G_Rec: 0.476 D_GP: 0.028 D_real: 1.001 D_fake: 0.497 +(epoch: 164, iters: 5696, time: 0.064) G_GAN: 0.185 G_GAN_Feat: 0.657 G_ID: 0.124 G_Rec: 0.283 D_GP: 0.025 D_real: 1.066 D_fake: 0.815 +(epoch: 164, iters: 6096, time: 0.064) G_GAN: 0.669 G_GAN_Feat: 0.846 G_ID: 0.219 G_Rec: 0.402 D_GP: 0.026 D_real: 1.301 D_fake: 0.348 +(epoch: 164, iters: 6496, time: 0.064) G_GAN: 0.170 G_GAN_Feat: 0.894 G_ID: 0.141 G_Rec: 0.343 D_GP: 0.047 D_real: 0.694 D_fake: 0.830 +(epoch: 164, iters: 6896, time: 0.064) G_GAN: 0.273 G_GAN_Feat: 0.905 G_ID: 0.176 G_Rec: 0.423 D_GP: 0.030 D_real: 0.884 D_fake: 0.731 +(epoch: 164, iters: 7296, time: 0.064) G_GAN: 0.420 G_GAN_Feat: 0.921 G_ID: 0.136 G_Rec: 0.373 D_GP: 0.159 D_real: 0.902 D_fake: 0.624 +(epoch: 164, iters: 7696, time: 0.064) G_GAN: 0.274 G_GAN_Feat: 1.067 G_ID: 0.245 G_Rec: 0.389 D_GP: 0.061 D_real: 0.475 D_fake: 0.732 +(epoch: 164, iters: 8096, time: 0.064) G_GAN: 0.145 G_GAN_Feat: 0.614 G_ID: 0.156 G_Rec: 0.301 D_GP: 0.020 D_real: 1.139 D_fake: 0.855 +(epoch: 164, iters: 8496, time: 0.064) G_GAN: 0.455 G_GAN_Feat: 0.825 G_ID: 0.189 G_Rec: 0.401 D_GP: 0.026 D_real: 1.107 D_fake: 0.558 +(epoch: 165, iters: 288, time: 0.064) G_GAN: 0.098 G_GAN_Feat: 0.696 G_ID: 0.144 G_Rec: 0.310 D_GP: 0.026 D_real: 0.993 D_fake: 0.902 +(epoch: 165, iters: 688, time: 0.064) G_GAN: 0.336 G_GAN_Feat: 0.831 G_ID: 0.215 G_Rec: 0.382 D_GP: 0.026 D_real: 1.049 D_fake: 0.667 +(epoch: 165, iters: 1088, time: 0.064) G_GAN: 0.113 G_GAN_Feat: 0.721 G_ID: 0.154 G_Rec: 0.315 D_GP: 0.021 D_real: 0.972 D_fake: 0.887 +(epoch: 165, iters: 1488, time: 0.064) G_GAN: 0.424 G_GAN_Feat: 0.935 G_ID: 0.207 G_Rec: 0.443 D_GP: 0.124 D_real: 0.746 D_fake: 0.582 +(epoch: 165, iters: 1888, time: 0.064) G_GAN: -0.250 G_GAN_Feat: 0.733 G_ID: 0.161 G_Rec: 0.308 D_GP: 0.023 D_real: 0.736 D_fake: 1.250 +(epoch: 165, iters: 2288, time: 0.064) G_GAN: 0.534 G_GAN_Feat: 0.915 G_ID: 0.204 G_Rec: 0.402 D_GP: 0.024 D_real: 1.191 D_fake: 0.470 +(epoch: 165, iters: 2688, time: 0.064) G_GAN: 0.349 G_GAN_Feat: 0.726 G_ID: 0.143 G_Rec: 0.340 D_GP: 0.025 D_real: 1.331 D_fake: 0.655 +(epoch: 165, iters: 3088, time: 0.064) G_GAN: 0.400 G_GAN_Feat: 0.829 G_ID: 0.200 G_Rec: 0.394 D_GP: 0.029 D_real: 1.091 D_fake: 0.603 +(epoch: 165, iters: 3488, time: 0.064) G_GAN: 0.149 G_GAN_Feat: 0.865 G_ID: 0.156 G_Rec: 0.336 D_GP: 0.142 D_real: 0.523 D_fake: 0.851 +(epoch: 165, iters: 3888, time: 0.064) G_GAN: 0.279 G_GAN_Feat: 1.203 G_ID: 0.202 G_Rec: 0.447 D_GP: 0.640 D_real: 0.301 D_fake: 0.725 +(epoch: 165, iters: 4288, time: 0.064) G_GAN: -0.151 G_GAN_Feat: 0.806 G_ID: 0.127 G_Rec: 0.357 D_GP: 0.042 D_real: 0.516 D_fake: 1.151 +(epoch: 165, iters: 4688, time: 0.064) G_GAN: 0.855 G_GAN_Feat: 1.045 G_ID: 0.210 G_Rec: 0.426 D_GP: 0.036 D_real: 1.361 D_fake: 0.221 +(epoch: 165, iters: 5088, time: 0.064) G_GAN: 0.044 G_GAN_Feat: 0.801 G_ID: 0.124 G_Rec: 0.336 D_GP: 0.050 D_real: 0.744 D_fake: 0.956 +(epoch: 165, iters: 5488, time: 0.064) G_GAN: 0.478 G_GAN_Feat: 0.789 G_ID: 0.190 G_Rec: 0.382 D_GP: 0.022 D_real: 1.297 D_fake: 0.528 +(epoch: 165, iters: 5888, time: 0.064) G_GAN: 0.227 G_GAN_Feat: 0.650 G_ID: 0.122 G_Rec: 0.318 D_GP: 0.027 D_real: 1.139 D_fake: 0.773 +(epoch: 165, iters: 6288, time: 0.064) G_GAN: 0.441 G_GAN_Feat: 0.872 G_ID: 0.176 G_Rec: 0.382 D_GP: 0.027 D_real: 1.193 D_fake: 0.564 +(epoch: 165, iters: 6688, time: 0.064) G_GAN: 0.306 G_GAN_Feat: 0.702 G_ID: 0.168 G_Rec: 0.311 D_GP: 0.030 D_real: 1.061 D_fake: 0.694 +(epoch: 165, iters: 7088, time: 0.064) G_GAN: 0.587 G_GAN_Feat: 1.025 G_ID: 0.191 G_Rec: 0.428 D_GP: 0.078 D_real: 0.824 D_fake: 0.431 +(epoch: 165, iters: 7488, time: 0.064) G_GAN: 0.264 G_GAN_Feat: 0.937 G_ID: 0.165 G_Rec: 0.343 D_GP: 0.067 D_real: 0.960 D_fake: 0.741 +(epoch: 165, iters: 7888, time: 0.064) G_GAN: 0.357 G_GAN_Feat: 0.908 G_ID: 0.206 G_Rec: 0.425 D_GP: 0.024 D_real: 1.050 D_fake: 0.649 +(epoch: 165, iters: 8288, time: 0.064) G_GAN: 0.231 G_GAN_Feat: 0.756 G_ID: 0.138 G_Rec: 0.318 D_GP: 0.032 D_real: 1.050 D_fake: 0.769 +(epoch: 166, iters: 80, time: 0.064) G_GAN: 0.504 G_GAN_Feat: 0.909 G_ID: 0.181 G_Rec: 0.429 D_GP: 0.054 D_real: 1.121 D_fake: 0.498 +(epoch: 166, iters: 480, time: 0.064) G_GAN: 0.331 G_GAN_Feat: 0.740 G_ID: 0.158 G_Rec: 0.300 D_GP: 0.089 D_real: 1.035 D_fake: 0.677 +(epoch: 166, iters: 880, time: 0.064) G_GAN: 0.496 G_GAN_Feat: 0.950 G_ID: 0.175 G_Rec: 0.431 D_GP: 0.072 D_real: 0.969 D_fake: 0.508 +(epoch: 166, iters: 1280, time: 0.064) G_GAN: 0.303 G_GAN_Feat: 0.824 G_ID: 0.149 G_Rec: 0.341 D_GP: 0.039 D_real: 1.016 D_fake: 0.700 +(epoch: 166, iters: 1680, time: 0.064) G_GAN: 0.365 G_GAN_Feat: 0.906 G_ID: 0.194 G_Rec: 0.381 D_GP: 0.036 D_real: 0.978 D_fake: 0.636 +(epoch: 166, iters: 2080, time: 0.064) G_GAN: 0.192 G_GAN_Feat: 0.656 G_ID: 0.132 G_Rec: 0.297 D_GP: 0.022 D_real: 1.092 D_fake: 0.808 +(epoch: 166, iters: 2480, time: 0.064) G_GAN: 0.478 G_GAN_Feat: 0.794 G_ID: 0.198 G_Rec: 0.384 D_GP: 0.027 D_real: 1.303 D_fake: 0.528 +(epoch: 166, iters: 2880, time: 0.064) G_GAN: -0.007 G_GAN_Feat: 0.819 G_ID: 0.131 G_Rec: 0.375 D_GP: 0.109 D_real: 0.701 D_fake: 1.007 +(epoch: 166, iters: 3280, time: 0.064) G_GAN: 0.366 G_GAN_Feat: 0.963 G_ID: 0.186 G_Rec: 0.417 D_GP: 0.037 D_real: 0.992 D_fake: 0.639 +(epoch: 166, iters: 3680, time: 0.064) G_GAN: 0.126 G_GAN_Feat: 0.816 G_ID: 0.149 G_Rec: 0.359 D_GP: 0.094 D_real: 0.782 D_fake: 0.874 +(epoch: 166, iters: 4080, time: 0.064) G_GAN: 0.452 G_GAN_Feat: 0.831 G_ID: 0.186 G_Rec: 0.395 D_GP: 0.025 D_real: 1.160 D_fake: 0.551 +(epoch: 166, iters: 4480, time: 0.064) G_GAN: 0.209 G_GAN_Feat: 0.665 G_ID: 0.124 G_Rec: 0.308 D_GP: 0.022 D_real: 1.064 D_fake: 0.792 +(epoch: 166, iters: 4880, time: 0.064) G_GAN: 0.446 G_GAN_Feat: 0.973 G_ID: 0.227 G_Rec: 0.448 D_GP: 0.066 D_real: 0.888 D_fake: 0.565 +(epoch: 166, iters: 5280, time: 0.064) G_GAN: 0.282 G_GAN_Feat: 0.815 G_ID: 0.142 G_Rec: 0.350 D_GP: 0.032 D_real: 1.007 D_fake: 0.718 +(epoch: 166, iters: 5680, time: 0.064) G_GAN: 0.399 G_GAN_Feat: 1.040 G_ID: 0.171 G_Rec: 0.407 D_GP: 0.040 D_real: 0.676 D_fake: 0.607 +(epoch: 166, iters: 6080, time: 0.064) G_GAN: 0.184 G_GAN_Feat: 0.714 G_ID: 0.166 G_Rec: 0.310 D_GP: 0.025 D_real: 1.099 D_fake: 0.816 +(epoch: 166, iters: 6480, time: 0.064) G_GAN: 0.419 G_GAN_Feat: 0.928 G_ID: 0.216 G_Rec: 0.374 D_GP: 0.030 D_real: 1.000 D_fake: 0.585 +(epoch: 166, iters: 6880, time: 0.064) G_GAN: 0.295 G_GAN_Feat: 0.724 G_ID: 0.131 G_Rec: 0.345 D_GP: 0.028 D_real: 1.212 D_fake: 0.705 +(epoch: 166, iters: 7280, time: 0.064) G_GAN: 0.765 G_GAN_Feat: 0.814 G_ID: 0.222 G_Rec: 0.382 D_GP: 0.022 D_real: 1.504 D_fake: 0.332 +(epoch: 166, iters: 7680, time: 0.064) G_GAN: 0.137 G_GAN_Feat: 0.812 G_ID: 0.140 G_Rec: 0.323 D_GP: 0.070 D_real: 0.725 D_fake: 0.863 +(epoch: 166, iters: 8080, time: 0.064) G_GAN: 0.576 G_GAN_Feat: 1.225 G_ID: 0.184 G_Rec: 0.475 D_GP: 0.438 D_real: 0.440 D_fake: 0.445 +(epoch: 166, iters: 8480, time: 0.064) G_GAN: 0.039 G_GAN_Feat: 0.807 G_ID: 0.185 G_Rec: 0.309 D_GP: 0.025 D_real: 0.815 D_fake: 0.961 +(epoch: 167, iters: 272, time: 0.064) G_GAN: 0.522 G_GAN_Feat: 1.132 G_ID: 0.198 G_Rec: 0.465 D_GP: 0.302 D_real: 0.469 D_fake: 0.487 +(epoch: 167, iters: 672, time: 0.064) G_GAN: 0.283 G_GAN_Feat: 0.817 G_ID: 0.129 G_Rec: 0.325 D_GP: 0.022 D_real: 1.055 D_fake: 0.718 +(epoch: 167, iters: 1072, time: 0.064) G_GAN: 0.315 G_GAN_Feat: 0.735 G_ID: 0.205 G_Rec: 0.405 D_GP: 0.019 D_real: 1.221 D_fake: 0.685 +(epoch: 167, iters: 1472, time: 0.064) G_GAN: 0.281 G_GAN_Feat: 0.567 G_ID: 0.163 G_Rec: 0.279 D_GP: 0.021 D_real: 1.221 D_fake: 0.719 +(epoch: 167, iters: 1872, time: 0.064) G_GAN: 0.384 G_GAN_Feat: 0.837 G_ID: 0.180 G_Rec: 0.384 D_GP: 0.023 D_real: 1.130 D_fake: 0.616 +(epoch: 167, iters: 2272, time: 0.064) G_GAN: 0.315 G_GAN_Feat: 0.644 G_ID: 0.147 G_Rec: 0.336 D_GP: 0.025 D_real: 1.290 D_fake: 0.685 +(epoch: 167, iters: 2672, time: 0.064) G_GAN: 0.573 G_GAN_Feat: 0.753 G_ID: 0.185 G_Rec: 0.375 D_GP: 0.026 D_real: 1.356 D_fake: 0.432 +(epoch: 167, iters: 3072, time: 0.064) G_GAN: 0.156 G_GAN_Feat: 0.643 G_ID: 0.154 G_Rec: 0.320 D_GP: 0.025 D_real: 1.031 D_fake: 0.844 +(epoch: 167, iters: 3472, time: 0.064) G_GAN: 0.593 G_GAN_Feat: 0.987 G_ID: 0.177 G_Rec: 0.420 D_GP: 0.051 D_real: 1.297 D_fake: 0.421 +(epoch: 167, iters: 3872, time: 0.064) G_GAN: 0.037 G_GAN_Feat: 0.737 G_ID: 0.177 G_Rec: 0.320 D_GP: 0.047 D_real: 0.787 D_fake: 0.963 +(epoch: 167, iters: 4272, time: 0.064) G_GAN: 0.578 G_GAN_Feat: 0.925 G_ID: 0.203 G_Rec: 0.417 D_GP: 0.044 D_real: 1.175 D_fake: 0.442 +(epoch: 167, iters: 4672, time: 0.064) G_GAN: -0.319 G_GAN_Feat: 1.039 G_ID: 0.137 G_Rec: 0.447 D_GP: 0.318 D_real: 0.343 D_fake: 1.319 +(epoch: 167, iters: 5072, time: 0.064) G_GAN: 0.035 G_GAN_Feat: 0.992 G_ID: 0.225 G_Rec: 0.448 D_GP: 0.041 D_real: 0.503 D_fake: 0.965 +(epoch: 167, iters: 5472, time: 0.064) G_GAN: -0.107 G_GAN_Feat: 0.858 G_ID: 0.132 G_Rec: 0.340 D_GP: 0.133 D_real: 0.507 D_fake: 1.107 +(epoch: 167, iters: 5872, time: 0.064) G_GAN: 0.637 G_GAN_Feat: 0.906 G_ID: 0.178 G_Rec: 0.382 D_GP: 0.083 D_real: 1.198 D_fake: 0.383 +(epoch: 167, iters: 6272, time: 0.064) G_GAN: 0.173 G_GAN_Feat: 0.730 G_ID: 0.152 G_Rec: 0.344 D_GP: 0.030 D_real: 0.961 D_fake: 0.827 +(epoch: 167, iters: 6672, time: 0.064) G_GAN: 0.521 G_GAN_Feat: 1.008 G_ID: 0.194 G_Rec: 0.457 D_GP: 0.099 D_real: 0.940 D_fake: 0.488 +(epoch: 167, iters: 7072, time: 0.064) G_GAN: -0.060 G_GAN_Feat: 0.796 G_ID: 0.135 G_Rec: 0.316 D_GP: 0.089 D_real: 0.590 D_fake: 1.060 +(epoch: 167, iters: 7472, time: 0.064) G_GAN: 0.164 G_GAN_Feat: 0.908 G_ID: 0.206 G_Rec: 0.393 D_GP: 0.098 D_real: 0.854 D_fake: 0.837 +(epoch: 167, iters: 7872, time: 0.064) G_GAN: 0.116 G_GAN_Feat: 0.829 G_ID: 0.151 G_Rec: 0.311 D_GP: 0.034 D_real: 0.749 D_fake: 0.884 +(epoch: 167, iters: 8272, time: 0.064) G_GAN: 0.521 G_GAN_Feat: 1.158 G_ID: 0.196 G_Rec: 0.454 D_GP: 0.221 D_real: 0.555 D_fake: 0.522 +(epoch: 168, iters: 64, time: 0.064) G_GAN: 0.214 G_GAN_Feat: 0.698 G_ID: 0.162 G_Rec: 0.300 D_GP: 0.029 D_real: 1.097 D_fake: 0.786 +(epoch: 168, iters: 464, time: 0.064) G_GAN: 0.227 G_GAN_Feat: 0.964 G_ID: 0.215 G_Rec: 0.436 D_GP: 0.030 D_real: 0.848 D_fake: 0.775 +(epoch: 168, iters: 864, time: 0.064) G_GAN: 0.206 G_GAN_Feat: 0.776 G_ID: 0.150 G_Rec: 0.341 D_GP: 0.075 D_real: 0.958 D_fake: 0.795 +(epoch: 168, iters: 1264, time: 0.064) G_GAN: 0.277 G_GAN_Feat: 0.907 G_ID: 0.183 G_Rec: 0.378 D_GP: 0.029 D_real: 0.963 D_fake: 0.723 +(epoch: 168, iters: 1664, time: 0.064) G_GAN: 0.328 G_GAN_Feat: 0.839 G_ID: 0.139 G_Rec: 0.331 D_GP: 0.045 D_real: 1.087 D_fake: 0.687 +(epoch: 168, iters: 2064, time: 0.064) G_GAN: 0.449 G_GAN_Feat: 0.782 G_ID: 0.208 G_Rec: 0.393 D_GP: 0.021 D_real: 1.218 D_fake: 0.562 +(epoch: 168, iters: 2464, time: 0.064) G_GAN: 0.008 G_GAN_Feat: 0.629 G_ID: 0.135 G_Rec: 0.369 D_GP: 0.025 D_real: 0.934 D_fake: 0.992 +(epoch: 168, iters: 2864, time: 0.064) G_GAN: 0.286 G_GAN_Feat: 0.877 G_ID: 0.196 G_Rec: 0.420 D_GP: 0.034 D_real: 1.027 D_fake: 0.720 +(epoch: 168, iters: 3264, time: 0.064) G_GAN: 0.051 G_GAN_Feat: 0.622 G_ID: 0.150 G_Rec: 0.298 D_GP: 0.022 D_real: 0.958 D_fake: 0.949 +(epoch: 168, iters: 3664, time: 0.064) G_GAN: 0.547 G_GAN_Feat: 0.757 G_ID: 0.195 G_Rec: 0.353 D_GP: 0.022 D_real: 1.337 D_fake: 0.454 +(epoch: 168, iters: 4064, time: 0.064) G_GAN: 0.342 G_GAN_Feat: 0.683 G_ID: 0.154 G_Rec: 0.416 D_GP: 0.025 D_real: 1.252 D_fake: 0.659 +(epoch: 168, iters: 4464, time: 0.064) G_GAN: 0.612 G_GAN_Feat: 1.021 G_ID: 0.178 G_Rec: 0.409 D_GP: 0.174 D_real: 0.779 D_fake: 0.395 +(epoch: 168, iters: 4864, time: 0.064) G_GAN: 0.031 G_GAN_Feat: 0.772 G_ID: 0.137 G_Rec: 0.354 D_GP: 0.023 D_real: 0.858 D_fake: 0.970 +(epoch: 168, iters: 5264, time: 0.064) G_GAN: 0.389 G_GAN_Feat: 0.860 G_ID: 0.193 G_Rec: 0.372 D_GP: 0.032 D_real: 1.087 D_fake: 0.617 +(epoch: 168, iters: 5664, time: 0.064) G_GAN: 0.221 G_GAN_Feat: 0.801 G_ID: 0.177 G_Rec: 0.353 D_GP: 0.046 D_real: 0.852 D_fake: 0.782 +(epoch: 168, iters: 6064, time: 0.064) G_GAN: 0.604 G_GAN_Feat: 0.989 G_ID: 0.212 G_Rec: 0.410 D_GP: 0.043 D_real: 0.955 D_fake: 0.403 +(epoch: 168, iters: 6464, time: 0.064) G_GAN: 0.277 G_GAN_Feat: 0.783 G_ID: 0.163 G_Rec: 0.311 D_GP: 0.029 D_real: 1.070 D_fake: 0.724 +(epoch: 168, iters: 6864, time: 0.064) G_GAN: 0.662 G_GAN_Feat: 0.967 G_ID: 0.195 G_Rec: 0.379 D_GP: 0.061 D_real: 0.926 D_fake: 0.354 +(epoch: 168, iters: 7264, time: 0.064) G_GAN: 0.288 G_GAN_Feat: 0.738 G_ID: 0.149 G_Rec: 0.306 D_GP: 0.029 D_real: 1.103 D_fake: 0.713 +(epoch: 168, iters: 7664, time: 0.064) G_GAN: 0.614 G_GAN_Feat: 0.894 G_ID: 0.186 G_Rec: 0.452 D_GP: 0.024 D_real: 1.313 D_fake: 0.390 +(epoch: 168, iters: 8064, time: 0.064) G_GAN: 0.129 G_GAN_Feat: 0.604 G_ID: 0.150 G_Rec: 0.289 D_GP: 0.018 D_real: 1.058 D_fake: 0.871 +(epoch: 168, iters: 8464, time: 0.064) G_GAN: 0.244 G_GAN_Feat: 0.907 G_ID: 0.189 G_Rec: 0.386 D_GP: 0.036 D_real: 0.964 D_fake: 0.758 +(epoch: 169, iters: 256, time: 0.064) G_GAN: 0.213 G_GAN_Feat: 0.711 G_ID: 0.153 G_Rec: 0.283 D_GP: 0.032 D_real: 1.018 D_fake: 0.787 +(epoch: 169, iters: 656, time: 0.064) G_GAN: 0.763 G_GAN_Feat: 0.899 G_ID: 0.202 G_Rec: 0.426 D_GP: 0.039 D_real: 1.352 D_fake: 0.281 +(epoch: 169, iters: 1056, time: 0.064) G_GAN: 0.065 G_GAN_Feat: 0.823 G_ID: 0.122 G_Rec: 0.308 D_GP: 0.055 D_real: 0.499 D_fake: 0.936 +(epoch: 169, iters: 1456, time: 0.064) G_GAN: 0.369 G_GAN_Feat: 1.072 G_ID: 0.198 G_Rec: 0.405 D_GP: 0.041 D_real: 0.606 D_fake: 0.634 +(epoch: 169, iters: 1856, time: 0.064) G_GAN: 0.580 G_GAN_Feat: 0.945 G_ID: 0.126 G_Rec: 0.328 D_GP: 0.166 D_real: 0.557 D_fake: 0.550 +(epoch: 169, iters: 2256, time: 0.064) G_GAN: 0.507 G_GAN_Feat: 0.802 G_ID: 0.179 G_Rec: 0.406 D_GP: 0.017 D_real: 1.279 D_fake: 0.495 +(epoch: 169, iters: 2656, time: 0.064) G_GAN: 0.126 G_GAN_Feat: 0.679 G_ID: 0.155 G_Rec: 0.310 D_GP: 0.036 D_real: 1.068 D_fake: 0.875 +(epoch: 169, iters: 3056, time: 0.064) G_GAN: 0.642 G_GAN_Feat: 1.058 G_ID: 0.178 G_Rec: 0.417 D_GP: 0.043 D_real: 1.171 D_fake: 0.380 +(epoch: 169, iters: 3456, time: 0.064) G_GAN: 0.165 G_GAN_Feat: 0.747 G_ID: 0.151 G_Rec: 0.324 D_GP: 0.029 D_real: 0.996 D_fake: 0.835 +(epoch: 169, iters: 3856, time: 0.064) G_GAN: 0.477 G_GAN_Feat: 0.946 G_ID: 0.189 G_Rec: 0.428 D_GP: 0.026 D_real: 1.172 D_fake: 0.528 +(epoch: 169, iters: 4256, time: 0.064) G_GAN: 0.090 G_GAN_Feat: 0.856 G_ID: 0.142 G_Rec: 0.348 D_GP: 0.096 D_real: 0.585 D_fake: 0.910 +(epoch: 169, iters: 4656, time: 0.064) G_GAN: 0.479 G_GAN_Feat: 1.026 G_ID: 0.184 G_Rec: 0.416 D_GP: 0.043 D_real: 1.116 D_fake: 0.537 +(epoch: 169, iters: 5056, time: 0.064) G_GAN: 0.236 G_GAN_Feat: 0.709 G_ID: 0.180 G_Rec: 0.316 D_GP: 0.025 D_real: 1.086 D_fake: 0.764 +(epoch: 169, iters: 5456, time: 0.064) G_GAN: 0.268 G_GAN_Feat: 0.805 G_ID: 0.205 G_Rec: 0.416 D_GP: 0.028 D_real: 1.028 D_fake: 0.732 +(epoch: 169, iters: 5856, time: 0.064) G_GAN: 0.339 G_GAN_Feat: 0.733 G_ID: 0.141 G_Rec: 0.303 D_GP: 0.051 D_real: 1.061 D_fake: 0.661 +(epoch: 169, iters: 6256, time: 0.064) G_GAN: 0.387 G_GAN_Feat: 0.947 G_ID: 0.195 G_Rec: 0.365 D_GP: 0.032 D_real: 0.914 D_fake: 0.614 +(epoch: 169, iters: 6656, time: 0.064) G_GAN: 0.547 G_GAN_Feat: 0.871 G_ID: 0.118 G_Rec: 0.320 D_GP: 0.133 D_real: 0.829 D_fake: 0.564 +(epoch: 169, iters: 7056, time: 0.064) G_GAN: 0.372 G_GAN_Feat: 0.841 G_ID: 0.201 G_Rec: 0.407 D_GP: 0.025 D_real: 1.063 D_fake: 0.631 +(epoch: 169, iters: 7456, time: 0.064) G_GAN: 0.079 G_GAN_Feat: 0.658 G_ID: 0.136 G_Rec: 0.316 D_GP: 0.025 D_real: 0.981 D_fake: 0.921 +(epoch: 169, iters: 7856, time: 0.064) G_GAN: 0.433 G_GAN_Feat: 1.058 G_ID: 0.150 G_Rec: 0.451 D_GP: 0.091 D_real: 0.778 D_fake: 0.572 +(epoch: 169, iters: 8256, time: 0.064) G_GAN: -0.009 G_GAN_Feat: 0.785 G_ID: 0.134 G_Rec: 0.332 D_GP: 0.032 D_real: 0.737 D_fake: 1.009 +(epoch: 170, iters: 48, time: 0.064) G_GAN: 0.546 G_GAN_Feat: 1.171 G_ID: 0.188 G_Rec: 0.447 D_GP: 0.108 D_real: 0.427 D_fake: 0.461 +(epoch: 170, iters: 448, time: 0.064) G_GAN: 0.092 G_GAN_Feat: 0.774 G_ID: 0.123 G_Rec: 0.307 D_GP: 0.023 D_real: 0.984 D_fake: 0.908 +(epoch: 170, iters: 848, time: 0.064) G_GAN: 0.144 G_GAN_Feat: 0.898 G_ID: 0.213 G_Rec: 0.393 D_GP: 0.045 D_real: 0.631 D_fake: 0.856 +(epoch: 170, iters: 1248, time: 0.064) G_GAN: 0.164 G_GAN_Feat: 0.742 G_ID: 0.112 G_Rec: 0.320 D_GP: 0.018 D_real: 1.112 D_fake: 0.836 +(epoch: 170, iters: 1648, time: 0.064) G_GAN: 0.254 G_GAN_Feat: 0.871 G_ID: 0.205 G_Rec: 0.426 D_GP: 0.025 D_real: 0.927 D_fake: 0.753 +(epoch: 170, iters: 2048, time: 0.064) G_GAN: 0.130 G_GAN_Feat: 0.720 G_ID: 0.119 G_Rec: 0.301 D_GP: 0.048 D_real: 0.896 D_fake: 0.872 +(epoch: 170, iters: 2448, time: 0.064) G_GAN: 0.548 G_GAN_Feat: 0.929 G_ID: 0.197 G_Rec: 0.390 D_GP: 0.030 D_real: 1.132 D_fake: 0.464 +(epoch: 170, iters: 2848, time: 0.064) G_GAN: 0.313 G_GAN_Feat: 0.947 G_ID: 0.123 G_Rec: 0.314 D_GP: 0.414 D_real: 0.492 D_fake: 0.692 +(epoch: 170, iters: 3248, time: 0.064) G_GAN: 0.651 G_GAN_Feat: 0.863 G_ID: 0.175 G_Rec: 0.388 D_GP: 0.024 D_real: 1.472 D_fake: 0.359 +(epoch: 170, iters: 3648, time: 0.064) G_GAN: 0.177 G_GAN_Feat: 0.644 G_ID: 0.123 G_Rec: 0.321 D_GP: 0.019 D_real: 1.140 D_fake: 0.823 +(epoch: 170, iters: 4048, time: 0.064) G_GAN: 0.689 G_GAN_Feat: 0.834 G_ID: 0.187 G_Rec: 0.385 D_GP: 0.023 D_real: 1.458 D_fake: 0.335 +(epoch: 170, iters: 4448, time: 0.064) G_GAN: 0.229 G_GAN_Feat: 0.644 G_ID: 0.140 G_Rec: 0.313 D_GP: 0.023 D_real: 1.118 D_fake: 0.772 +(epoch: 170, iters: 4848, time: 0.064) G_GAN: 0.580 G_GAN_Feat: 1.055 G_ID: 0.209 G_Rec: 0.387 D_GP: 0.057 D_real: 1.147 D_fake: 0.518 +(epoch: 170, iters: 5248, time: 0.064) G_GAN: 0.461 G_GAN_Feat: 0.769 G_ID: 0.163 G_Rec: 0.315 D_GP: 0.027 D_real: 1.274 D_fake: 0.554 +(epoch: 170, iters: 5648, time: 0.064) G_GAN: 0.259 G_GAN_Feat: 0.802 G_ID: 0.200 G_Rec: 0.362 D_GP: 0.025 D_real: 1.034 D_fake: 0.741 +(epoch: 170, iters: 6048, time: 0.064) G_GAN: 0.149 G_GAN_Feat: 0.830 G_ID: 0.146 G_Rec: 0.353 D_GP: 0.056 D_real: 0.758 D_fake: 0.852 +(epoch: 170, iters: 6448, time: 0.064) G_GAN: 0.709 G_GAN_Feat: 0.966 G_ID: 0.182 G_Rec: 0.410 D_GP: 0.027 D_real: 1.410 D_fake: 0.342 +(epoch: 170, iters: 6848, time: 0.064) G_GAN: 0.364 G_GAN_Feat: 0.734 G_ID: 0.126 G_Rec: 0.317 D_GP: 0.024 D_real: 1.320 D_fake: 0.640 +(epoch: 170, iters: 7248, time: 0.064) G_GAN: 0.174 G_GAN_Feat: 0.838 G_ID: 0.198 G_Rec: 0.389 D_GP: 0.029 D_real: 0.844 D_fake: 0.826 +(epoch: 170, iters: 7648, time: 0.064) G_GAN: 0.190 G_GAN_Feat: 0.746 G_ID: 0.179 G_Rec: 0.331 D_GP: 0.033 D_real: 1.066 D_fake: 0.812 +(epoch: 170, iters: 8048, time: 0.064) G_GAN: 0.464 G_GAN_Feat: 0.805 G_ID: 0.190 G_Rec: 0.362 D_GP: 0.023 D_real: 1.183 D_fake: 0.541 +(epoch: 170, iters: 8448, time: 0.064) G_GAN: 0.243 G_GAN_Feat: 0.729 G_ID: 0.146 G_Rec: 0.371 D_GP: 0.033 D_real: 1.065 D_fake: 0.759 +(epoch: 171, iters: 240, time: 0.064) G_GAN: 0.392 G_GAN_Feat: 0.947 G_ID: 0.176 G_Rec: 0.428 D_GP: 0.042 D_real: 0.984 D_fake: 0.617 +(epoch: 171, iters: 640, time: 0.064) G_GAN: -0.068 G_GAN_Feat: 1.016 G_ID: 0.152 G_Rec: 0.398 D_GP: 0.204 D_real: 0.111 D_fake: 1.070 +(epoch: 171, iters: 1040, time: 0.064) G_GAN: 0.313 G_GAN_Feat: 0.767 G_ID: 0.218 G_Rec: 0.396 D_GP: 0.022 D_real: 1.103 D_fake: 0.687 +(epoch: 171, iters: 1440, time: 0.064) G_GAN: 0.348 G_GAN_Feat: 0.642 G_ID: 0.131 G_Rec: 0.292 D_GP: 0.030 D_real: 1.242 D_fake: 0.653 +(epoch: 171, iters: 1840, time: 0.064) G_GAN: 0.384 G_GAN_Feat: 0.868 G_ID: 0.219 G_Rec: 0.397 D_GP: 0.035 D_real: 1.080 D_fake: 0.618 +(epoch: 171, iters: 2240, time: 0.064) G_GAN: 0.368 G_GAN_Feat: 0.826 G_ID: 0.129 G_Rec: 0.313 D_GP: 0.057 D_real: 0.897 D_fake: 0.697 +(epoch: 171, iters: 2640, time: 0.064) G_GAN: 0.619 G_GAN_Feat: 0.773 G_ID: 0.189 G_Rec: 0.432 D_GP: 0.020 D_real: 1.365 D_fake: 0.389 +(epoch: 171, iters: 3040, time: 0.064) G_GAN: 0.468 G_GAN_Feat: 0.679 G_ID: 0.129 G_Rec: 0.305 D_GP: 0.026 D_real: 1.340 D_fake: 0.534 +(epoch: 171, iters: 3440, time: 0.064) G_GAN: 0.173 G_GAN_Feat: 1.036 G_ID: 0.213 G_Rec: 0.397 D_GP: 0.143 D_real: 0.361 D_fake: 0.828 +(epoch: 171, iters: 3840, time: 0.064) G_GAN: 0.083 G_GAN_Feat: 1.031 G_ID: 0.170 G_Rec: 0.379 D_GP: 0.229 D_real: 0.372 D_fake: 0.917 +(epoch: 171, iters: 4240, time: 0.064) G_GAN: 0.746 G_GAN_Feat: 1.012 G_ID: 0.233 G_Rec: 0.421 D_GP: 0.170 D_real: 0.986 D_fake: 0.514 +(epoch: 171, iters: 4640, time: 0.064) G_GAN: 0.203 G_GAN_Feat: 0.909 G_ID: 0.145 G_Rec: 0.359 D_GP: 0.067 D_real: 0.591 D_fake: 0.798 +(epoch: 171, iters: 5040, time: 0.064) G_GAN: 0.443 G_GAN_Feat: 0.983 G_ID: 0.196 G_Rec: 0.407 D_GP: 0.032 D_real: 0.891 D_fake: 0.559 +(epoch: 171, iters: 5440, time: 0.064) G_GAN: -0.026 G_GAN_Feat: 0.800 G_ID: 0.142 G_Rec: 0.319 D_GP: 0.076 D_real: 0.763 D_fake: 1.026 +(epoch: 171, iters: 5840, time: 0.064) G_GAN: 0.424 G_GAN_Feat: 0.928 G_ID: 0.178 G_Rec: 0.438 D_GP: 0.034 D_real: 0.989 D_fake: 0.588 +(epoch: 171, iters: 6240, time: 0.064) G_GAN: 0.309 G_GAN_Feat: 0.929 G_ID: 0.133 G_Rec: 0.358 D_GP: 0.350 D_real: 0.380 D_fake: 0.714 +(epoch: 171, iters: 6640, time: 0.064) G_GAN: 0.684 G_GAN_Feat: 0.952 G_ID: 0.205 G_Rec: 0.426 D_GP: 0.032 D_real: 1.224 D_fake: 0.376 +(epoch: 171, iters: 7040, time: 0.064) G_GAN: 0.148 G_GAN_Feat: 0.666 G_ID: 0.138 G_Rec: 0.341 D_GP: 0.021 D_real: 1.115 D_fake: 0.852 +(epoch: 171, iters: 7440, time: 0.064) G_GAN: 0.332 G_GAN_Feat: 0.798 G_ID: 0.181 G_Rec: 0.387 D_GP: 0.026 D_real: 1.006 D_fake: 0.670 +(epoch: 171, iters: 7840, time: 0.064) G_GAN: 0.048 G_GAN_Feat: 0.737 G_ID: 0.132 G_Rec: 0.337 D_GP: 0.062 D_real: 0.718 D_fake: 0.952 +(epoch: 171, iters: 8240, time: 0.064) G_GAN: 0.146 G_GAN_Feat: 0.815 G_ID: 0.217 G_Rec: 0.374 D_GP: 0.034 D_real: 0.884 D_fake: 0.854 +(epoch: 172, iters: 32, time: 0.064) G_GAN: 0.070 G_GAN_Feat: 0.712 G_ID: 0.134 G_Rec: 0.323 D_GP: 0.032 D_real: 0.935 D_fake: 0.930 +(epoch: 172, iters: 432, time: 0.064) G_GAN: 0.543 G_GAN_Feat: 0.806 G_ID: 0.203 G_Rec: 0.402 D_GP: 0.024 D_real: 1.262 D_fake: 0.468 +(epoch: 172, iters: 832, time: 0.064) G_GAN: 0.270 G_GAN_Feat: 0.708 G_ID: 0.154 G_Rec: 0.320 D_GP: 0.025 D_real: 1.170 D_fake: 0.730 +(epoch: 172, iters: 1232, time: 0.064) G_GAN: 0.129 G_GAN_Feat: 0.932 G_ID: 0.179 G_Rec: 0.391 D_GP: 0.026 D_real: 0.826 D_fake: 0.871 +(epoch: 172, iters: 1632, time: 0.064) G_GAN: 0.053 G_GAN_Feat: 0.678 G_ID: 0.163 G_Rec: 0.295 D_GP: 0.028 D_real: 0.936 D_fake: 0.947 +(epoch: 172, iters: 2032, time: 0.064) G_GAN: 0.626 G_GAN_Feat: 0.891 G_ID: 0.210 G_Rec: 0.398 D_GP: 0.028 D_real: 1.260 D_fake: 0.387 +(epoch: 172, iters: 2432, time: 0.064) G_GAN: 0.372 G_GAN_Feat: 0.701 G_ID: 0.148 G_Rec: 0.299 D_GP: 0.022 D_real: 1.257 D_fake: 0.628 +(epoch: 172, iters: 2832, time: 0.064) G_GAN: 0.446 G_GAN_Feat: 0.905 G_ID: 0.194 G_Rec: 0.383 D_GP: 0.034 D_real: 1.033 D_fake: 0.556 +(epoch: 172, iters: 3232, time: 0.064) G_GAN: 0.173 G_GAN_Feat: 0.844 G_ID: 0.155 G_Rec: 0.338 D_GP: 0.058 D_real: 0.783 D_fake: 0.827 +(epoch: 172, iters: 3632, time: 0.064) G_GAN: 0.372 G_GAN_Feat: 0.916 G_ID: 0.192 G_Rec: 0.405 D_GP: 0.039 D_real: 0.922 D_fake: 0.628 +(epoch: 172, iters: 4032, time: 0.064) G_GAN: 0.286 G_GAN_Feat: 0.693 G_ID: 0.160 G_Rec: 0.306 D_GP: 0.024 D_real: 1.201 D_fake: 0.715 +(epoch: 172, iters: 4432, time: 0.064) G_GAN: 0.509 G_GAN_Feat: 0.889 G_ID: 0.187 G_Rec: 0.382 D_GP: 0.037 D_real: 1.189 D_fake: 0.525 +(epoch: 172, iters: 4832, time: 0.064) G_GAN: 0.156 G_GAN_Feat: 0.681 G_ID: 0.152 G_Rec: 0.335 D_GP: 0.024 D_real: 1.104 D_fake: 0.844 +(epoch: 172, iters: 5232, time: 0.064) G_GAN: 0.764 G_GAN_Feat: 1.108 G_ID: 0.201 G_Rec: 0.424 D_GP: 0.061 D_real: 0.830 D_fake: 0.298 +(epoch: 172, iters: 5632, time: 0.064) G_GAN: 0.127 G_GAN_Feat: 0.697 G_ID: 0.146 G_Rec: 0.310 D_GP: 0.024 D_real: 0.996 D_fake: 0.873 +(epoch: 172, iters: 6032, time: 0.064) G_GAN: 0.504 G_GAN_Feat: 0.849 G_ID: 0.174 G_Rec: 0.389 D_GP: 0.037 D_real: 1.203 D_fake: 0.509 +(epoch: 172, iters: 6432, time: 0.064) G_GAN: 0.203 G_GAN_Feat: 0.896 G_ID: 0.111 G_Rec: 0.341 D_GP: 0.133 D_real: 0.631 D_fake: 0.797 +(epoch: 172, iters: 6832, time: 0.064) G_GAN: 0.568 G_GAN_Feat: 0.974 G_ID: 0.195 G_Rec: 0.464 D_GP: 0.034 D_real: 1.096 D_fake: 0.436 +(epoch: 172, iters: 7232, time: 0.064) G_GAN: 0.354 G_GAN_Feat: 0.817 G_ID: 0.136 G_Rec: 0.364 D_GP: 0.019 D_real: 1.249 D_fake: 0.664 +(epoch: 172, iters: 7632, time: 0.064) G_GAN: 0.426 G_GAN_Feat: 0.832 G_ID: 0.170 G_Rec: 0.422 D_GP: 0.023 D_real: 1.176 D_fake: 0.594 +(epoch: 172, iters: 8032, time: 0.064) G_GAN: -0.030 G_GAN_Feat: 0.643 G_ID: 0.143 G_Rec: 0.332 D_GP: 0.020 D_real: 0.832 D_fake: 1.030 +(epoch: 172, iters: 8432, time: 0.064) G_GAN: 0.120 G_GAN_Feat: 0.814 G_ID: 0.194 G_Rec: 0.416 D_GP: 0.024 D_real: 0.854 D_fake: 0.881 +(epoch: 173, iters: 224, time: 0.064) G_GAN: 0.059 G_GAN_Feat: 0.653 G_ID: 0.126 G_Rec: 0.322 D_GP: 0.029 D_real: 0.911 D_fake: 0.942 +(epoch: 173, iters: 624, time: 0.064) G_GAN: 0.291 G_GAN_Feat: 0.897 G_ID: 0.177 G_Rec: 0.404 D_GP: 0.043 D_real: 0.821 D_fake: 0.715 +(epoch: 173, iters: 1024, time: 0.064) G_GAN: 0.154 G_GAN_Feat: 0.656 G_ID: 0.142 G_Rec: 0.297 D_GP: 0.030 D_real: 1.071 D_fake: 0.847 +(epoch: 173, iters: 1424, time: 0.064) G_GAN: 0.526 G_GAN_Feat: 0.893 G_ID: 0.177 G_Rec: 0.385 D_GP: 0.039 D_real: 1.088 D_fake: 0.482 +(epoch: 173, iters: 1824, time: 0.064) G_GAN: 0.277 G_GAN_Feat: 0.768 G_ID: 0.146 G_Rec: 0.297 D_GP: 0.025 D_real: 1.131 D_fake: 0.725 +(epoch: 173, iters: 2224, time: 0.064) G_GAN: 0.349 G_GAN_Feat: 0.837 G_ID: 0.190 G_Rec: 0.372 D_GP: 0.026 D_real: 1.048 D_fake: 0.652 +(epoch: 173, iters: 2624, time: 0.064) G_GAN: 0.059 G_GAN_Feat: 0.722 G_ID: 0.137 G_Rec: 0.334 D_GP: 0.025 D_real: 0.946 D_fake: 0.941 +(epoch: 173, iters: 3024, time: 0.064) G_GAN: 0.587 G_GAN_Feat: 0.947 G_ID: 0.219 G_Rec: 0.399 D_GP: 0.031 D_real: 1.160 D_fake: 0.426 +(epoch: 173, iters: 3424, time: 0.064) G_GAN: 0.023 G_GAN_Feat: 0.703 G_ID: 0.144 G_Rec: 0.310 D_GP: 0.028 D_real: 0.928 D_fake: 0.977 +(epoch: 173, iters: 3824, time: 0.064) G_GAN: 0.400 G_GAN_Feat: 0.979 G_ID: 0.233 G_Rec: 0.429 D_GP: 0.068 D_real: 0.842 D_fake: 0.606 +(epoch: 173, iters: 4224, time: 0.064) G_GAN: 0.279 G_GAN_Feat: 0.843 G_ID: 0.141 G_Rec: 0.315 D_GP: 0.032 D_real: 0.910 D_fake: 0.721 +(epoch: 173, iters: 4624, time: 0.064) G_GAN: 0.751 G_GAN_Feat: 0.874 G_ID: 0.177 G_Rec: 0.384 D_GP: 0.024 D_real: 1.467 D_fake: 0.351 +(epoch: 173, iters: 5024, time: 0.064) G_GAN: 0.144 G_GAN_Feat: 0.585 G_ID: 0.120 G_Rec: 0.306 D_GP: 0.020 D_real: 1.068 D_fake: 0.856 +(epoch: 173, iters: 5424, time: 0.064) G_GAN: 0.469 G_GAN_Feat: 0.710 G_ID: 0.187 G_Rec: 0.370 D_GP: 0.020 D_real: 1.285 D_fake: 0.535 +(epoch: 173, iters: 5824, time: 0.064) G_GAN: -0.010 G_GAN_Feat: 0.599 G_ID: 0.162 G_Rec: 0.324 D_GP: 0.024 D_real: 0.919 D_fake: 1.010 +(epoch: 173, iters: 6224, time: 0.064) G_GAN: 0.297 G_GAN_Feat: 0.770 G_ID: 0.194 G_Rec: 0.383 D_GP: 0.031 D_real: 1.034 D_fake: 0.703 +(epoch: 173, iters: 6624, time: 0.064) G_GAN: -0.007 G_GAN_Feat: 0.685 G_ID: 0.139 G_Rec: 0.334 D_GP: 0.035 D_real: 0.813 D_fake: 1.007 +(epoch: 173, iters: 7024, time: 0.064) G_GAN: 0.045 G_GAN_Feat: 0.938 G_ID: 0.189 G_Rec: 0.416 D_GP: 0.184 D_real: 0.517 D_fake: 0.955 +(epoch: 173, iters: 7424, time: 0.064) G_GAN: 0.158 G_GAN_Feat: 0.636 G_ID: 0.130 G_Rec: 0.317 D_GP: 0.066 D_real: 0.979 D_fake: 0.843 +(epoch: 173, iters: 7824, time: 0.064) G_GAN: 0.222 G_GAN_Feat: 0.776 G_ID: 0.200 G_Rec: 0.360 D_GP: 0.041 D_real: 0.861 D_fake: 0.779 +(epoch: 173, iters: 8224, time: 0.064) G_GAN: 0.305 G_GAN_Feat: 0.891 G_ID: 0.131 G_Rec: 0.340 D_GP: 0.083 D_real: 0.933 D_fake: 0.698 +(epoch: 174, iters: 16, time: 0.064) G_GAN: 0.246 G_GAN_Feat: 0.873 G_ID: 0.204 G_Rec: 0.390 D_GP: 0.033 D_real: 0.904 D_fake: 0.754 +(epoch: 174, iters: 416, time: 0.064) G_GAN: 0.325 G_GAN_Feat: 0.766 G_ID: 0.152 G_Rec: 0.320 D_GP: 0.040 D_real: 1.111 D_fake: 0.675 +(epoch: 174, iters: 816, time: 0.064) G_GAN: 0.373 G_GAN_Feat: 0.836 G_ID: 0.177 G_Rec: 0.419 D_GP: 0.023 D_real: 1.195 D_fake: 0.628 +(epoch: 174, iters: 1216, time: 0.064) G_GAN: 0.253 G_GAN_Feat: 0.610 G_ID: 0.158 G_Rec: 0.307 D_GP: 0.025 D_real: 1.149 D_fake: 0.747 +(epoch: 174, iters: 1616, time: 0.064) G_GAN: 0.423 G_GAN_Feat: 0.782 G_ID: 0.197 G_Rec: 0.385 D_GP: 0.031 D_real: 1.158 D_fake: 0.578 +(epoch: 174, iters: 2016, time: 0.064) G_GAN: 0.128 G_GAN_Feat: 0.721 G_ID: 0.116 G_Rec: 0.295 D_GP: 0.044 D_real: 0.926 D_fake: 0.872 +(epoch: 174, iters: 2416, time: 0.064) G_GAN: 0.510 G_GAN_Feat: 0.961 G_ID: 0.172 G_Rec: 0.452 D_GP: 0.037 D_real: 1.031 D_fake: 0.494 +(epoch: 174, iters: 2816, time: 0.064) G_GAN: 0.381 G_GAN_Feat: 0.908 G_ID: 0.109 G_Rec: 0.359 D_GP: 0.100 D_real: 0.757 D_fake: 0.623 +(epoch: 174, iters: 3216, time: 0.064) G_GAN: 0.644 G_GAN_Feat: 0.895 G_ID: 0.260 G_Rec: 0.424 D_GP: 0.031 D_real: 1.301 D_fake: 0.386 +(epoch: 174, iters: 3616, time: 0.064) G_GAN: 0.355 G_GAN_Feat: 0.786 G_ID: 0.148 G_Rec: 0.304 D_GP: 0.066 D_real: 0.981 D_fake: 0.653 +(epoch: 174, iters: 4016, time: 0.064) G_GAN: 0.384 G_GAN_Feat: 0.875 G_ID: 0.197 G_Rec: 0.387 D_GP: 0.041 D_real: 0.916 D_fake: 0.617 +(epoch: 174, iters: 4416, time: 0.064) G_GAN: 0.083 G_GAN_Feat: 0.765 G_ID: 0.124 G_Rec: 0.325 D_GP: 0.066 D_real: 0.783 D_fake: 0.917 +(epoch: 174, iters: 4816, time: 0.064) G_GAN: 0.602 G_GAN_Feat: 0.995 G_ID: 0.190 G_Rec: 0.410 D_GP: 0.076 D_real: 0.934 D_fake: 0.414 +(epoch: 174, iters: 5216, time: 0.064) G_GAN: 0.597 G_GAN_Feat: 1.010 G_ID: 0.131 G_Rec: 0.406 D_GP: 0.870 D_real: 0.777 D_fake: 0.427 +(epoch: 174, iters: 5616, time: 0.064) G_GAN: 0.436 G_GAN_Feat: 0.994 G_ID: 0.189 G_Rec: 0.498 D_GP: 0.061 D_real: 0.923 D_fake: 0.582 +(epoch: 174, iters: 6016, time: 0.064) G_GAN: 0.054 G_GAN_Feat: 0.733 G_ID: 0.122 G_Rec: 0.320 D_GP: 0.025 D_real: 1.036 D_fake: 0.946 +(epoch: 174, iters: 6416, time: 0.064) G_GAN: 0.516 G_GAN_Feat: 0.882 G_ID: 0.190 G_Rec: 0.428 D_GP: 0.024 D_real: 1.183 D_fake: 0.501 +(epoch: 174, iters: 6816, time: 0.064) G_GAN: 0.424 G_GAN_Feat: 0.911 G_ID: 0.155 G_Rec: 0.349 D_GP: 0.060 D_real: 0.715 D_fake: 0.634 +(epoch: 174, iters: 7216, time: 0.064) G_GAN: 0.269 G_GAN_Feat: 0.786 G_ID: 0.186 G_Rec: 0.410 D_GP: 0.019 D_real: 1.014 D_fake: 0.731 +(epoch: 174, iters: 7616, time: 0.064) G_GAN: 0.234 G_GAN_Feat: 0.691 G_ID: 0.145 G_Rec: 0.320 D_GP: 0.023 D_real: 1.227 D_fake: 0.766 +(epoch: 174, iters: 8016, time: 0.064) G_GAN: 0.266 G_GAN_Feat: 0.833 G_ID: 0.204 G_Rec: 0.383 D_GP: 0.025 D_real: 0.948 D_fake: 0.735 +(epoch: 174, iters: 8416, time: 0.064) G_GAN: 0.259 G_GAN_Feat: 0.709 G_ID: 0.146 G_Rec: 0.302 D_GP: 0.036 D_real: 1.084 D_fake: 0.741 +(epoch: 175, iters: 208, time: 0.064) G_GAN: 0.620 G_GAN_Feat: 0.880 G_ID: 0.203 G_Rec: 0.391 D_GP: 0.027 D_real: 1.312 D_fake: 0.403 +(epoch: 175, iters: 608, time: 0.064) G_GAN: 0.202 G_GAN_Feat: 0.790 G_ID: 0.158 G_Rec: 0.339 D_GP: 0.037 D_real: 0.930 D_fake: 0.798 +(epoch: 175, iters: 1008, time: 0.064) G_GAN: 0.636 G_GAN_Feat: 1.072 G_ID: 0.181 G_Rec: 0.468 D_GP: 0.069 D_real: 0.840 D_fake: 0.395 +(epoch: 175, iters: 1408, time: 0.064) G_GAN: 0.296 G_GAN_Feat: 0.751 G_ID: 0.143 G_Rec: 0.337 D_GP: 0.022 D_real: 1.155 D_fake: 0.704 +(epoch: 175, iters: 1808, time: 0.064) G_GAN: 0.615 G_GAN_Feat: 0.932 G_ID: 0.174 G_Rec: 0.406 D_GP: 0.038 D_real: 1.246 D_fake: 0.416 +(epoch: 175, iters: 2208, time: 0.064) G_GAN: 0.486 G_GAN_Feat: 0.696 G_ID: 0.125 G_Rec: 0.315 D_GP: 0.025 D_real: 1.448 D_fake: 0.515 +(epoch: 175, iters: 2608, time: 0.064) G_GAN: 0.592 G_GAN_Feat: 0.977 G_ID: 0.201 G_Rec: 0.422 D_GP: 0.040 D_real: 0.978 D_fake: 0.420 +(epoch: 175, iters: 3008, time: 0.064) G_GAN: 0.284 G_GAN_Feat: 0.686 G_ID: 0.130 G_Rec: 0.347 D_GP: 0.024 D_real: 1.213 D_fake: 0.716 +(epoch: 175, iters: 3408, time: 0.064) G_GAN: 0.185 G_GAN_Feat: 0.886 G_ID: 0.222 G_Rec: 0.367 D_GP: 0.047 D_real: 0.817 D_fake: 0.817 +(epoch: 175, iters: 3808, time: 0.064) G_GAN: 0.375 G_GAN_Feat: 0.680 G_ID: 0.130 G_Rec: 0.299 D_GP: 0.022 D_real: 1.295 D_fake: 0.625 +(epoch: 175, iters: 4208, time: 0.064) G_GAN: 0.562 G_GAN_Feat: 0.902 G_ID: 0.185 G_Rec: 0.378 D_GP: 0.035 D_real: 1.212 D_fake: 0.448 +(epoch: 175, iters: 4608, time: 0.064) G_GAN: 0.117 G_GAN_Feat: 0.706 G_ID: 0.137 G_Rec: 0.295 D_GP: 0.022 D_real: 1.097 D_fake: 0.883 +(epoch: 175, iters: 5008, time: 0.064) G_GAN: 0.104 G_GAN_Feat: 0.938 G_ID: 0.227 G_Rec: 0.409 D_GP: 0.049 D_real: 0.848 D_fake: 0.897 +(epoch: 175, iters: 5408, time: 0.064) G_GAN: 0.346 G_GAN_Feat: 0.662 G_ID: 0.152 G_Rec: 0.284 D_GP: 0.026 D_real: 1.303 D_fake: 0.654 +(epoch: 175, iters: 5808, time: 0.064) G_GAN: 0.463 G_GAN_Feat: 0.961 G_ID: 0.182 G_Rec: 0.428 D_GP: 0.072 D_real: 0.861 D_fake: 0.545 +(epoch: 175, iters: 6208, time: 0.064) G_GAN: 0.502 G_GAN_Feat: 0.766 G_ID: 0.138 G_Rec: 0.305 D_GP: 0.031 D_real: 1.289 D_fake: 0.502 +(epoch: 175, iters: 6608, time: 0.064) G_GAN: 0.598 G_GAN_Feat: 0.806 G_ID: 0.206 G_Rec: 0.350 D_GP: 0.029 D_real: 1.389 D_fake: 0.419 +(epoch: 175, iters: 7008, time: 0.064) G_GAN: 0.525 G_GAN_Feat: 0.707 G_ID: 0.149 G_Rec: 0.296 D_GP: 0.023 D_real: 1.465 D_fake: 0.489 +(epoch: 175, iters: 7408, time: 0.064) G_GAN: 0.392 G_GAN_Feat: 0.780 G_ID: 0.181 G_Rec: 0.392 D_GP: 0.021 D_real: 1.218 D_fake: 0.611 +(epoch: 175, iters: 7808, time: 0.064) G_GAN: -0.001 G_GAN_Feat: 0.688 G_ID: 0.142 G_Rec: 0.340 D_GP: 0.029 D_real: 0.905 D_fake: 1.001 +(epoch: 175, iters: 8208, time: 0.064) G_GAN: 0.330 G_GAN_Feat: 0.872 G_ID: 0.201 G_Rec: 0.403 D_GP: 0.043 D_real: 0.968 D_fake: 0.674 +(epoch: 175, iters: 8608, time: 0.064) G_GAN: 0.352 G_GAN_Feat: 0.790 G_ID: 0.147 G_Rec: 0.286 D_GP: 0.026 D_real: 1.098 D_fake: 0.648 +(epoch: 176, iters: 400, time: 0.064) G_GAN: 0.345 G_GAN_Feat: 0.862 G_ID: 0.182 G_Rec: 0.395 D_GP: 0.025 D_real: 1.133 D_fake: 0.657 +(epoch: 176, iters: 800, time: 0.064) G_GAN: 0.451 G_GAN_Feat: 0.680 G_ID: 0.141 G_Rec: 0.277 D_GP: 0.027 D_real: 1.344 D_fake: 0.550 +(epoch: 176, iters: 1200, time: 0.064) G_GAN: 0.269 G_GAN_Feat: 0.723 G_ID: 0.184 G_Rec: 0.376 D_GP: 0.022 D_real: 1.038 D_fake: 0.731 +(epoch: 176, iters: 1600, time: 0.064) G_GAN: 0.105 G_GAN_Feat: 0.712 G_ID: 0.138 G_Rec: 0.334 D_GP: 0.040 D_real: 0.885 D_fake: 0.896 +(epoch: 176, iters: 2000, time: 0.064) G_GAN: 0.142 G_GAN_Feat: 0.944 G_ID: 0.190 G_Rec: 0.419 D_GP: 0.271 D_real: 0.555 D_fake: 0.861 +(epoch: 176, iters: 2400, time: 0.064) G_GAN: 0.310 G_GAN_Feat: 0.803 G_ID: 0.157 G_Rec: 0.312 D_GP: 0.043 D_real: 0.921 D_fake: 0.692 +(epoch: 176, iters: 2800, time: 0.064) G_GAN: 0.489 G_GAN_Feat: 0.657 G_ID: 0.176 G_Rec: 0.366 D_GP: 0.021 D_real: 1.264 D_fake: 0.515 +(epoch: 176, iters: 3200, time: 0.064) G_GAN: 0.218 G_GAN_Feat: 0.531 G_ID: 0.140 G_Rec: 0.307 D_GP: 0.019 D_real: 1.175 D_fake: 0.782 +(epoch: 176, iters: 3600, time: 0.064) G_GAN: 0.342 G_GAN_Feat: 0.749 G_ID: 0.215 G_Rec: 0.403 D_GP: 0.026 D_real: 1.004 D_fake: 0.658 +(epoch: 176, iters: 4000, time: 0.064) G_GAN: 0.093 G_GAN_Feat: 0.626 G_ID: 0.146 G_Rec: 0.347 D_GP: 0.034 D_real: 1.001 D_fake: 0.907 +(epoch: 176, iters: 4400, time: 0.064) G_GAN: 0.462 G_GAN_Feat: 0.743 G_ID: 0.185 G_Rec: 0.371 D_GP: 0.029 D_real: 1.218 D_fake: 0.541 +(epoch: 176, iters: 4800, time: 0.064) G_GAN: 0.128 G_GAN_Feat: 0.710 G_ID: 0.139 G_Rec: 0.314 D_GP: 0.071 D_real: 0.897 D_fake: 0.872 +(epoch: 176, iters: 5200, time: 0.064) G_GAN: 0.338 G_GAN_Feat: 0.832 G_ID: 0.197 G_Rec: 0.385 D_GP: 0.039 D_real: 1.066 D_fake: 0.662 +(epoch: 176, iters: 5600, time: 0.064) G_GAN: 0.120 G_GAN_Feat: 0.680 G_ID: 0.142 G_Rec: 0.274 D_GP: 0.044 D_real: 0.870 D_fake: 0.880 +(epoch: 176, iters: 6000, time: 0.064) G_GAN: 0.484 G_GAN_Feat: 0.880 G_ID: 0.176 G_Rec: 0.382 D_GP: 0.036 D_real: 1.136 D_fake: 0.521 +(epoch: 176, iters: 6400, time: 0.064) G_GAN: 0.336 G_GAN_Feat: 0.699 G_ID: 0.122 G_Rec: 0.301 D_GP: 0.024 D_real: 1.241 D_fake: 0.665 +(epoch: 176, iters: 6800, time: 0.064) G_GAN: 0.441 G_GAN_Feat: 0.918 G_ID: 0.170 G_Rec: 0.432 D_GP: 0.030 D_real: 1.030 D_fake: 0.560 +(epoch: 176, iters: 7200, time: 0.064) G_GAN: 0.172 G_GAN_Feat: 0.641 G_ID: 0.141 G_Rec: 0.305 D_GP: 0.028 D_real: 1.090 D_fake: 0.828 +(epoch: 176, iters: 7600, time: 0.064) G_GAN: 0.502 G_GAN_Feat: 1.102 G_ID: 0.223 G_Rec: 0.416 D_GP: 0.100 D_real: 0.528 D_fake: 0.586 +(epoch: 176, iters: 8000, time: 0.064) G_GAN: 0.178 G_GAN_Feat: 0.617 G_ID: 0.173 G_Rec: 0.311 D_GP: 0.018 D_real: 1.091 D_fake: 0.822 +(epoch: 176, iters: 8400, time: 0.064) G_GAN: 0.338 G_GAN_Feat: 0.788 G_ID: 0.206 G_Rec: 0.380 D_GP: 0.028 D_real: 1.054 D_fake: 0.666 +(epoch: 177, iters: 192, time: 0.064) G_GAN: 0.412 G_GAN_Feat: 0.785 G_ID: 0.130 G_Rec: 0.339 D_GP: 0.061 D_real: 1.077 D_fake: 0.590 +(epoch: 177, iters: 592, time: 0.064) G_GAN: 0.523 G_GAN_Feat: 1.016 G_ID: 0.173 G_Rec: 0.429 D_GP: 0.132 D_real: 0.780 D_fake: 0.485 +(epoch: 177, iters: 992, time: 0.064) G_GAN: 0.251 G_GAN_Feat: 0.840 G_ID: 0.162 G_Rec: 0.343 D_GP: 0.043 D_real: 0.880 D_fake: 0.750 +(epoch: 177, iters: 1392, time: 0.064) G_GAN: 0.427 G_GAN_Feat: 0.917 G_ID: 0.230 G_Rec: 0.392 D_GP: 0.042 D_real: 0.972 D_fake: 0.585 +(epoch: 177, iters: 1792, time: 0.064) G_GAN: 0.067 G_GAN_Feat: 0.696 G_ID: 0.150 G_Rec: 0.309 D_GP: 0.022 D_real: 1.012 D_fake: 0.935 +(epoch: 177, iters: 2192, time: 0.064) G_GAN: 0.269 G_GAN_Feat: 0.858 G_ID: 0.185 G_Rec: 0.435 D_GP: 0.023 D_real: 0.973 D_fake: 0.732 +(epoch: 177, iters: 2592, time: 0.064) G_GAN: 0.243 G_GAN_Feat: 0.740 G_ID: 0.152 G_Rec: 0.323 D_GP: 0.032 D_real: 1.063 D_fake: 0.758 +(epoch: 177, iters: 2992, time: 0.064) G_GAN: 0.322 G_GAN_Feat: 0.796 G_ID: 0.191 G_Rec: 0.374 D_GP: 0.032 D_real: 1.078 D_fake: 0.680 +(epoch: 177, iters: 3392, time: 0.064) G_GAN: 0.160 G_GAN_Feat: 0.665 G_ID: 0.151 G_Rec: 0.287 D_GP: 0.025 D_real: 1.047 D_fake: 0.840 +(epoch: 177, iters: 3792, time: 0.064) G_GAN: 0.139 G_GAN_Feat: 1.088 G_ID: 0.222 G_Rec: 0.419 D_GP: 0.379 D_real: 0.483 D_fake: 0.862 +(epoch: 177, iters: 4192, time: 0.064) G_GAN: 0.187 G_GAN_Feat: 0.684 G_ID: 0.175 G_Rec: 0.335 D_GP: 0.024 D_real: 1.175 D_fake: 0.813 +(epoch: 177, iters: 4592, time: 0.064) G_GAN: 0.261 G_GAN_Feat: 0.850 G_ID: 0.195 G_Rec: 0.357 D_GP: 0.028 D_real: 1.010 D_fake: 0.739 +(epoch: 177, iters: 4992, time: 0.064) G_GAN: 0.142 G_GAN_Feat: 0.894 G_ID: 0.161 G_Rec: 0.353 D_GP: 0.044 D_real: 0.742 D_fake: 0.858 +(epoch: 177, iters: 5392, time: 0.064) G_GAN: 0.367 G_GAN_Feat: 0.997 G_ID: 0.232 G_Rec: 0.415 D_GP: 0.058 D_real: 0.695 D_fake: 0.637 +(epoch: 177, iters: 5792, time: 0.064) G_GAN: 0.188 G_GAN_Feat: 0.807 G_ID: 0.138 G_Rec: 0.298 D_GP: 0.030 D_real: 0.872 D_fake: 0.812 +(epoch: 177, iters: 6192, time: 0.064) G_GAN: 0.193 G_GAN_Feat: 0.932 G_ID: 0.194 G_Rec: 0.390 D_GP: 0.030 D_real: 0.807 D_fake: 0.807 +(epoch: 177, iters: 6592, time: 0.064) G_GAN: -0.117 G_GAN_Feat: 0.692 G_ID: 0.140 G_Rec: 0.330 D_GP: 0.036 D_real: 0.827 D_fake: 1.117 +(epoch: 177, iters: 6992, time: 0.064) G_GAN: 0.176 G_GAN_Feat: 0.866 G_ID: 0.191 G_Rec: 0.397 D_GP: 0.041 D_real: 0.881 D_fake: 0.825 +(epoch: 177, iters: 7392, time: 0.064) G_GAN: 0.266 G_GAN_Feat: 0.718 G_ID: 0.137 G_Rec: 0.281 D_GP: 0.054 D_real: 1.141 D_fake: 0.735 +(epoch: 177, iters: 7792, time: 0.064) G_GAN: 0.325 G_GAN_Feat: 1.102 G_ID: 0.183 G_Rec: 0.418 D_GP: 0.046 D_real: 1.263 D_fake: 0.679 +(epoch: 177, iters: 8192, time: 0.064) G_GAN: 0.259 G_GAN_Feat: 0.747 G_ID: 0.139 G_Rec: 0.394 D_GP: 0.026 D_real: 1.099 D_fake: 0.742 +(epoch: 177, iters: 8592, time: 0.064) G_GAN: 0.182 G_GAN_Feat: 1.002 G_ID: 0.188 G_Rec: 0.408 D_GP: 0.039 D_real: 0.746 D_fake: 0.818 +(epoch: 178, iters: 384, time: 0.064) G_GAN: 0.184 G_GAN_Feat: 0.780 G_ID: 0.119 G_Rec: 0.312 D_GP: 0.293 D_real: 0.762 D_fake: 0.817 +(epoch: 178, iters: 784, time: 0.064) G_GAN: 0.616 G_GAN_Feat: 0.909 G_ID: 0.187 G_Rec: 0.393 D_GP: 0.046 D_real: 1.136 D_fake: 0.397 +(epoch: 178, iters: 1184, time: 0.064) G_GAN: 0.349 G_GAN_Feat: 0.793 G_ID: 0.147 G_Rec: 0.304 D_GP: 0.033 D_real: 0.997 D_fake: 0.653 +(epoch: 178, iters: 1584, time: 0.064) G_GAN: 0.556 G_GAN_Feat: 1.062 G_ID: 0.175 G_Rec: 0.452 D_GP: 0.053 D_real: 0.782 D_fake: 0.448 +(epoch: 178, iters: 1984, time: 0.064) G_GAN: 0.102 G_GAN_Feat: 0.703 G_ID: 0.164 G_Rec: 0.324 D_GP: 0.046 D_real: 0.920 D_fake: 0.898 +(epoch: 178, iters: 2384, time: 0.064) G_GAN: 0.657 G_GAN_Feat: 0.838 G_ID: 0.196 G_Rec: 0.395 D_GP: 0.046 D_real: 1.257 D_fake: 0.351 +(epoch: 178, iters: 2784, time: 0.064) G_GAN: 0.171 G_GAN_Feat: 0.836 G_ID: 0.140 G_Rec: 0.338 D_GP: 0.173 D_real: 0.537 D_fake: 0.829 +(epoch: 178, iters: 3184, time: 0.064) G_GAN: 0.414 G_GAN_Feat: 0.847 G_ID: 0.186 G_Rec: 0.416 D_GP: 0.022 D_real: 1.167 D_fake: 0.591 +(epoch: 178, iters: 3584, time: 0.064) G_GAN: 0.250 G_GAN_Feat: 0.733 G_ID: 0.170 G_Rec: 0.327 D_GP: 0.024 D_real: 1.108 D_fake: 0.751 +(epoch: 178, iters: 3984, time: 0.064) G_GAN: 0.289 G_GAN_Feat: 0.990 G_ID: 0.187 G_Rec: 0.431 D_GP: 0.055 D_real: 0.674 D_fake: 0.714 +(epoch: 178, iters: 4384, time: 0.064) G_GAN: 0.196 G_GAN_Feat: 0.976 G_ID: 0.161 G_Rec: 0.360 D_GP: 0.075 D_real: 0.722 D_fake: 0.807 +(epoch: 178, iters: 4784, time: 0.064) G_GAN: 0.505 G_GAN_Feat: 0.906 G_ID: 0.178 G_Rec: 0.389 D_GP: 0.032 D_real: 1.213 D_fake: 0.503 +(epoch: 178, iters: 5184, time: 0.064) G_GAN: 0.401 G_GAN_Feat: 0.812 G_ID: 0.142 G_Rec: 0.338 D_GP: 0.048 D_real: 1.214 D_fake: 0.601 +(epoch: 178, iters: 5584, time: 0.064) G_GAN: 0.323 G_GAN_Feat: 1.047 G_ID: 0.229 G_Rec: 0.412 D_GP: 0.132 D_real: 0.430 D_fake: 0.678 +(epoch: 178, iters: 5984, time: 0.064) G_GAN: 0.355 G_GAN_Feat: 1.019 G_ID: 0.123 G_Rec: 0.319 D_GP: 0.385 D_real: 0.299 D_fake: 0.708 +(epoch: 178, iters: 6384, time: 0.064) G_GAN: 0.442 G_GAN_Feat: 0.904 G_ID: 0.202 G_Rec: 0.387 D_GP: 0.034 D_real: 1.099 D_fake: 0.562 +(epoch: 178, iters: 6784, time: 0.064) G_GAN: 0.276 G_GAN_Feat: 0.777 G_ID: 0.143 G_Rec: 0.322 D_GP: 0.030 D_real: 1.026 D_fake: 0.724 +(epoch: 178, iters: 7184, time: 0.064) G_GAN: 0.047 G_GAN_Feat: 0.896 G_ID: 0.202 G_Rec: 0.366 D_GP: 0.024 D_real: 0.848 D_fake: 0.953 +(epoch: 178, iters: 7584, time: 0.064) G_GAN: 0.340 G_GAN_Feat: 0.859 G_ID: 0.134 G_Rec: 0.324 D_GP: 0.036 D_real: 1.000 D_fake: 0.661 +(epoch: 178, iters: 7984, time: 0.064) G_GAN: 0.546 G_GAN_Feat: 1.125 G_ID: 0.192 G_Rec: 0.413 D_GP: 0.061 D_real: 0.676 D_fake: 0.461 +(epoch: 178, iters: 8384, time: 0.064) G_GAN: 0.248 G_GAN_Feat: 0.760 G_ID: 0.199 G_Rec: 0.357 D_GP: 0.031 D_real: 1.028 D_fake: 0.758 +(epoch: 179, iters: 176, time: 0.064) G_GAN: 0.545 G_GAN_Feat: 0.941 G_ID: 0.197 G_Rec: 0.414 D_GP: 0.054 D_real: 1.031 D_fake: 0.459 +(epoch: 179, iters: 576, time: 0.064) G_GAN: 0.419 G_GAN_Feat: 0.706 G_ID: 0.129 G_Rec: 0.301 D_GP: 0.040 D_real: 1.228 D_fake: 0.581 +(epoch: 179, iters: 976, time: 0.064) G_GAN: 0.666 G_GAN_Feat: 0.821 G_ID: 0.192 G_Rec: 0.383 D_GP: 0.023 D_real: 1.358 D_fake: 0.345 +(epoch: 179, iters: 1376, time: 0.064) G_GAN: 0.290 G_GAN_Feat: 0.775 G_ID: 0.141 G_Rec: 0.358 D_GP: 0.038 D_real: 0.977 D_fake: 0.715 +(epoch: 179, iters: 1776, time: 0.064) G_GAN: 0.416 G_GAN_Feat: 1.099 G_ID: 0.238 G_Rec: 0.407 D_GP: 0.179 D_real: 0.642 D_fake: 0.589 +(epoch: 179, iters: 2176, time: 0.064) G_GAN: 0.310 G_GAN_Feat: 0.891 G_ID: 0.132 G_Rec: 0.358 D_GP: 0.084 D_real: 0.737 D_fake: 0.692 +(epoch: 179, iters: 2576, time: 0.064) G_GAN: 0.572 G_GAN_Feat: 0.949 G_ID: 0.206 G_Rec: 0.394 D_GP: 0.045 D_real: 1.100 D_fake: 0.438 +(epoch: 179, iters: 2976, time: 0.064) G_GAN: 0.413 G_GAN_Feat: 0.696 G_ID: 0.168 G_Rec: 0.308 D_GP: 0.023 D_real: 1.290 D_fake: 0.624 +(epoch: 179, iters: 3376, time: 0.064) G_GAN: 0.509 G_GAN_Feat: 0.890 G_ID: 0.217 G_Rec: 0.393 D_GP: 0.024 D_real: 1.210 D_fake: 0.492 +(epoch: 179, iters: 3776, time: 0.064) G_GAN: 0.264 G_GAN_Feat: 0.639 G_ID: 0.154 G_Rec: 0.284 D_GP: 0.033 D_real: 1.173 D_fake: 0.740 +(epoch: 179, iters: 4176, time: 0.064) G_GAN: 0.262 G_GAN_Feat: 0.878 G_ID: 0.204 G_Rec: 0.395 D_GP: 0.035 D_real: 1.008 D_fake: 0.739 +(epoch: 179, iters: 4576, time: 0.064) G_GAN: 0.566 G_GAN_Feat: 0.833 G_ID: 0.122 G_Rec: 0.298 D_GP: 0.053 D_real: 1.117 D_fake: 0.443 +(epoch: 179, iters: 4976, time: 0.064) G_GAN: 0.469 G_GAN_Feat: 0.908 G_ID: 0.173 G_Rec: 0.408 D_GP: 0.035 D_real: 1.105 D_fake: 0.534 +(epoch: 179, iters: 5376, time: 0.064) G_GAN: 0.358 G_GAN_Feat: 0.789 G_ID: 0.138 G_Rec: 0.323 D_GP: 0.037 D_real: 1.164 D_fake: 0.651 +(epoch: 179, iters: 5776, time: 0.064) G_GAN: 0.193 G_GAN_Feat: 0.732 G_ID: 0.193 G_Rec: 0.393 D_GP: 0.018 D_real: 1.068 D_fake: 0.807 +(epoch: 179, iters: 6176, time: 0.064) G_GAN: 0.216 G_GAN_Feat: 0.619 G_ID: 0.140 G_Rec: 0.289 D_GP: 0.019 D_real: 1.191 D_fake: 0.784 +(epoch: 179, iters: 6576, time: 0.064) G_GAN: 0.382 G_GAN_Feat: 0.771 G_ID: 0.193 G_Rec: 0.372 D_GP: 0.021 D_real: 1.173 D_fake: 0.619 +(epoch: 179, iters: 6976, time: 0.064) G_GAN: 0.189 G_GAN_Feat: 0.726 G_ID: 0.161 G_Rec: 0.385 D_GP: 0.036 D_real: 1.074 D_fake: 0.811 +(epoch: 179, iters: 7376, time: 0.064) G_GAN: 0.234 G_GAN_Feat: 0.804 G_ID: 0.200 G_Rec: 0.353 D_GP: 0.040 D_real: 1.026 D_fake: 0.767 +(epoch: 179, iters: 7776, time: 0.064) G_GAN: 0.296 G_GAN_Feat: 0.755 G_ID: 0.181 G_Rec: 0.303 D_GP: 0.048 D_real: 0.981 D_fake: 0.707 +(epoch: 179, iters: 8176, time: 0.064) G_GAN: 0.413 G_GAN_Feat: 0.917 G_ID: 0.174 G_Rec: 0.399 D_GP: 0.031 D_real: 1.099 D_fake: 0.588 +(epoch: 179, iters: 8576, time: 0.064) G_GAN: 0.123 G_GAN_Feat: 0.852 G_ID: 0.134 G_Rec: 0.337 D_GP: 0.080 D_real: 0.974 D_fake: 0.877 +(epoch: 180, iters: 368, time: 0.064) G_GAN: 0.296 G_GAN_Feat: 0.805 G_ID: 0.228 G_Rec: 0.378 D_GP: 0.024 D_real: 1.073 D_fake: 0.704 +(epoch: 180, iters: 768, time: 0.064) G_GAN: 0.157 G_GAN_Feat: 0.731 G_ID: 0.151 G_Rec: 0.316 D_GP: 0.035 D_real: 1.009 D_fake: 0.843 +(epoch: 180, iters: 1168, time: 0.064) G_GAN: 0.491 G_GAN_Feat: 0.914 G_ID: 0.177 G_Rec: 0.413 D_GP: 0.038 D_real: 1.157 D_fake: 0.514 +(epoch: 180, iters: 1568, time: 0.064) G_GAN: 0.175 G_GAN_Feat: 0.736 G_ID: 0.160 G_Rec: 0.332 D_GP: 0.028 D_real: 1.023 D_fake: 0.827 +(epoch: 180, iters: 1968, time: 0.064) G_GAN: 0.957 G_GAN_Feat: 0.945 G_ID: 0.183 G_Rec: 0.377 D_GP: 0.168 D_real: 1.364 D_fake: 0.190 +(epoch: 180, iters: 2368, time: 0.064) G_GAN: 0.298 G_GAN_Feat: 0.774 G_ID: 0.146 G_Rec: 0.375 D_GP: 0.034 D_real: 1.044 D_fake: 0.702 +(epoch: 180, iters: 2768, time: 0.064) G_GAN: 0.531 G_GAN_Feat: 0.930 G_ID: 0.197 G_Rec: 0.371 D_GP: 0.034 D_real: 1.075 D_fake: 0.472 +(epoch: 180, iters: 3168, time: 0.064) G_GAN: 0.656 G_GAN_Feat: 0.855 G_ID: 0.147 G_Rec: 0.328 D_GP: 0.057 D_real: 1.220 D_fake: 0.363 +(epoch: 180, iters: 3568, time: 0.064) G_GAN: 0.815 G_GAN_Feat: 0.903 G_ID: 0.211 G_Rec: 0.378 D_GP: 0.100 D_real: 1.391 D_fake: 0.346 +(epoch: 180, iters: 3968, time: 0.064) G_GAN: -0.063 G_GAN_Feat: 0.669 G_ID: 0.130 G_Rec: 0.318 D_GP: 0.022 D_real: 0.933 D_fake: 1.063 +(epoch: 180, iters: 4368, time: 0.064) G_GAN: 0.230 G_GAN_Feat: 0.822 G_ID: 0.194 G_Rec: 0.433 D_GP: 0.024 D_real: 0.942 D_fake: 0.772 +(epoch: 180, iters: 4768, time: 0.064) G_GAN: -0.006 G_GAN_Feat: 0.705 G_ID: 0.124 G_Rec: 0.322 D_GP: 0.037 D_real: 0.739 D_fake: 1.006 +(epoch: 180, iters: 5168, time: 0.064) G_GAN: 0.290 G_GAN_Feat: 0.871 G_ID: 0.181 G_Rec: 0.409 D_GP: 0.025 D_real: 0.967 D_fake: 0.710 +(epoch: 180, iters: 5568, time: 0.064) G_GAN: 0.222 G_GAN_Feat: 0.848 G_ID: 0.115 G_Rec: 0.309 D_GP: 0.032 D_real: 0.921 D_fake: 0.779 +(epoch: 180, iters: 5968, time: 0.064) G_GAN: 0.506 G_GAN_Feat: 0.837 G_ID: 0.194 G_Rec: 0.390 D_GP: 0.024 D_real: 1.207 D_fake: 0.495 +(epoch: 180, iters: 6368, time: 0.064) G_GAN: 0.138 G_GAN_Feat: 0.823 G_ID: 0.150 G_Rec: 0.314 D_GP: 0.039 D_real: 0.839 D_fake: 0.862 +(epoch: 180, iters: 6768, time: 0.064) G_GAN: 0.485 G_GAN_Feat: 0.932 G_ID: 0.185 G_Rec: 0.395 D_GP: 0.030 D_real: 0.972 D_fake: 0.521 +(epoch: 180, iters: 7168, time: 0.064) G_GAN: 0.197 G_GAN_Feat: 0.937 G_ID: 0.123 G_Rec: 0.329 D_GP: 0.120 D_real: 0.423 D_fake: 0.804 +(epoch: 180, iters: 7568, time: 0.064) G_GAN: 0.283 G_GAN_Feat: 0.762 G_ID: 0.184 G_Rec: 0.411 D_GP: 0.022 D_real: 1.074 D_fake: 0.718 +(epoch: 180, iters: 7968, time: 0.064) G_GAN: -0.015 G_GAN_Feat: 0.647 G_ID: 0.163 G_Rec: 0.342 D_GP: 0.020 D_real: 0.956 D_fake: 1.015 +(epoch: 180, iters: 8368, time: 0.064) G_GAN: 0.166 G_GAN_Feat: 0.794 G_ID: 0.204 G_Rec: 0.388 D_GP: 0.025 D_real: 0.908 D_fake: 0.834 +(epoch: 181, iters: 160, time: 0.064) G_GAN: 0.091 G_GAN_Feat: 0.646 G_ID: 0.145 G_Rec: 0.287 D_GP: 0.025 D_real: 1.086 D_fake: 0.910 +(epoch: 181, iters: 560, time: 0.064) G_GAN: 0.092 G_GAN_Feat: 0.972 G_ID: 0.203 G_Rec: 0.407 D_GP: 0.150 D_real: 0.424 D_fake: 0.913 +(epoch: 181, iters: 960, time: 0.064) G_GAN: 0.162 G_GAN_Feat: 0.717 G_ID: 0.135 G_Rec: 0.311 D_GP: 0.022 D_real: 1.007 D_fake: 0.838 +(epoch: 181, iters: 1360, time: 0.064) G_GAN: 0.417 G_GAN_Feat: 0.893 G_ID: 0.179 G_Rec: 0.428 D_GP: 0.034 D_real: 1.032 D_fake: 0.586 +(epoch: 181, iters: 1760, time: 0.064) G_GAN: 0.243 G_GAN_Feat: 0.897 G_ID: 0.131 G_Rec: 0.331 D_GP: 0.109 D_real: 0.639 D_fake: 0.758 +(epoch: 181, iters: 2160, time: 0.064) G_GAN: 0.429 G_GAN_Feat: 0.860 G_ID: 0.205 G_Rec: 0.396 D_GP: 0.036 D_real: 1.058 D_fake: 0.603 +(epoch: 181, iters: 2560, time: 0.064) G_GAN: 0.158 G_GAN_Feat: 1.108 G_ID: 0.163 G_Rec: 0.373 D_GP: 0.217 D_real: 0.312 D_fake: 0.842 +(epoch: 181, iters: 2960, time: 0.064) G_GAN: 0.249 G_GAN_Feat: 0.938 G_ID: 0.225 G_Rec: 0.421 D_GP: 0.077 D_real: 0.737 D_fake: 0.751 +(epoch: 181, iters: 3360, time: 0.064) G_GAN: -0.018 G_GAN_Feat: 1.023 G_ID: 0.130 G_Rec: 0.433 D_GP: 0.345 D_real: 0.386 D_fake: 1.018 +(epoch: 181, iters: 3760, time: 0.064) G_GAN: 0.689 G_GAN_Feat: 0.946 G_ID: 0.165 G_Rec: 0.407 D_GP: 0.029 D_real: 1.247 D_fake: 0.325 +(epoch: 181, iters: 4160, time: 0.064) G_GAN: 0.093 G_GAN_Feat: 0.876 G_ID: 0.136 G_Rec: 0.333 D_GP: 0.104 D_real: 0.532 D_fake: 0.909 +(epoch: 181, iters: 4560, time: 0.064) G_GAN: 0.336 G_GAN_Feat: 1.035 G_ID: 0.212 G_Rec: 0.394 D_GP: 0.037 D_real: 0.778 D_fake: 0.665 +(epoch: 181, iters: 4960, time: 0.064) G_GAN: 0.300 G_GAN_Feat: 0.680 G_ID: 0.154 G_Rec: 0.300 D_GP: 0.023 D_real: 1.255 D_fake: 0.702 +(epoch: 181, iters: 5360, time: 0.064) G_GAN: 0.868 G_GAN_Feat: 0.791 G_ID: 0.201 G_Rec: 0.410 D_GP: 0.018 D_real: 1.709 D_fake: 0.236 +(epoch: 181, iters: 5760, time: 0.064) G_GAN: 0.101 G_GAN_Feat: 0.653 G_ID: 0.150 G_Rec: 0.338 D_GP: 0.028 D_real: 1.016 D_fake: 0.899 +(epoch: 181, iters: 6160, time: 0.064) G_GAN: 0.539 G_GAN_Feat: 0.799 G_ID: 0.217 G_Rec: 0.365 D_GP: 0.027 D_real: 1.302 D_fake: 0.464 +(epoch: 181, iters: 6560, time: 0.064) G_GAN: 0.227 G_GAN_Feat: 0.754 G_ID: 0.125 G_Rec: 0.343 D_GP: 0.036 D_real: 0.985 D_fake: 0.773 +(epoch: 181, iters: 6960, time: 0.064) G_GAN: 0.502 G_GAN_Feat: 0.846 G_ID: 0.214 G_Rec: 0.387 D_GP: 0.024 D_real: 1.245 D_fake: 0.499 +(epoch: 181, iters: 7360, time: 0.064) G_GAN: 0.010 G_GAN_Feat: 0.619 G_ID: 0.161 G_Rec: 0.279 D_GP: 0.020 D_real: 0.977 D_fake: 0.990 +(epoch: 181, iters: 7760, time: 0.064) G_GAN: 0.253 G_GAN_Feat: 0.900 G_ID: 0.185 G_Rec: 0.374 D_GP: 0.069 D_real: 0.834 D_fake: 0.747 +(epoch: 181, iters: 8160, time: 0.064) G_GAN: 0.625 G_GAN_Feat: 0.666 G_ID: 0.117 G_Rec: 0.292 D_GP: 0.028 D_real: 1.524 D_fake: 0.397 +(epoch: 181, iters: 8560, time: 0.064) G_GAN: 0.317 G_GAN_Feat: 1.080 G_ID: 0.196 G_Rec: 0.494 D_GP: 0.173 D_real: 0.622 D_fake: 0.709 +(epoch: 182, iters: 352, time: 0.064) G_GAN: 0.114 G_GAN_Feat: 0.778 G_ID: 0.160 G_Rec: 0.305 D_GP: 0.050 D_real: 0.714 D_fake: 0.887 +(epoch: 182, iters: 752, time: 0.064) G_GAN: 0.360 G_GAN_Feat: 0.943 G_ID: 0.178 G_Rec: 0.386 D_GP: 0.044 D_real: 0.938 D_fake: 0.641 +(epoch: 182, iters: 1152, time: 0.064) G_GAN: 0.372 G_GAN_Feat: 0.766 G_ID: 0.180 G_Rec: 0.279 D_GP: 0.028 D_real: 1.113 D_fake: 0.647 +(epoch: 182, iters: 1552, time: 0.064) G_GAN: 0.318 G_GAN_Feat: 0.730 G_ID: 0.229 G_Rec: 0.353 D_GP: 0.029 D_real: 1.126 D_fake: 0.684 +(epoch: 182, iters: 1952, time: 0.064) G_GAN: 0.087 G_GAN_Feat: 0.763 G_ID: 0.139 G_Rec: 0.333 D_GP: 0.076 D_real: 0.778 D_fake: 0.913 +(epoch: 182, iters: 2352, time: 0.064) G_GAN: 0.372 G_GAN_Feat: 0.967 G_ID: 0.178 G_Rec: 0.448 D_GP: 0.031 D_real: 0.879 D_fake: 0.634 +(epoch: 182, iters: 2752, time: 0.064) G_GAN: 0.173 G_GAN_Feat: 0.794 G_ID: 0.117 G_Rec: 0.349 D_GP: 0.025 D_real: 1.043 D_fake: 0.827 +(epoch: 182, iters: 3152, time: 0.064) G_GAN: 0.467 G_GAN_Feat: 0.819 G_ID: 0.184 G_Rec: 0.361 D_GP: 0.023 D_real: 1.250 D_fake: 0.534 +(epoch: 182, iters: 3552, time: 0.064) G_GAN: 0.224 G_GAN_Feat: 0.691 G_ID: 0.155 G_Rec: 0.333 D_GP: 0.039 D_real: 1.091 D_fake: 0.776 +(epoch: 182, iters: 3952, time: 0.064) G_GAN: 0.867 G_GAN_Feat: 1.025 G_ID: 0.173 G_Rec: 0.411 D_GP: 0.051 D_real: 1.355 D_fake: 0.237 +(epoch: 182, iters: 4352, time: 0.064) G_GAN: 0.188 G_GAN_Feat: 0.862 G_ID: 0.141 G_Rec: 0.323 D_GP: 0.065 D_real: 0.640 D_fake: 0.812 +(epoch: 182, iters: 4752, time: 0.064) G_GAN: 0.874 G_GAN_Feat: 0.961 G_ID: 0.190 G_Rec: 0.419 D_GP: 0.028 D_real: 1.382 D_fake: 0.238 +(epoch: 182, iters: 5152, time: 0.064) G_GAN: 0.446 G_GAN_Feat: 0.688 G_ID: 0.152 G_Rec: 0.293 D_GP: 0.022 D_real: 1.409 D_fake: 0.558 +(epoch: 182, iters: 5552, time: 0.064) G_GAN: 0.485 G_GAN_Feat: 1.106 G_ID: 0.185 G_Rec: 0.438 D_GP: 0.033 D_real: 0.913 D_fake: 0.517 +(epoch: 182, iters: 5952, time: 0.064) G_GAN: 0.277 G_GAN_Feat: 0.911 G_ID: 0.138 G_Rec: 0.334 D_GP: 0.083 D_real: 0.710 D_fake: 0.744 +(epoch: 182, iters: 6352, time: 0.064) G_GAN: 0.683 G_GAN_Feat: 0.856 G_ID: 0.190 G_Rec: 0.388 D_GP: 0.022 D_real: 1.485 D_fake: 0.339 +(epoch: 182, iters: 6752, time: 0.064) G_GAN: 0.336 G_GAN_Feat: 0.709 G_ID: 0.161 G_Rec: 0.314 D_GP: 0.029 D_real: 1.131 D_fake: 0.668 +(epoch: 182, iters: 7152, time: 0.064) G_GAN: 0.198 G_GAN_Feat: 0.946 G_ID: 0.211 G_Rec: 0.353 D_GP: 0.028 D_real: 1.072 D_fake: 0.803 +(epoch: 182, iters: 7552, time: 0.064) G_GAN: 0.278 G_GAN_Feat: 0.718 G_ID: 0.150 G_Rec: 0.320 D_GP: 0.032 D_real: 1.243 D_fake: 0.723 +(epoch: 182, iters: 7952, time: 0.064) G_GAN: 0.602 G_GAN_Feat: 1.042 G_ID: 0.207 G_Rec: 0.436 D_GP: 0.047 D_real: 1.072 D_fake: 0.417 +(epoch: 182, iters: 8352, time: 0.064) G_GAN: 0.328 G_GAN_Feat: 0.608 G_ID: 0.131 G_Rec: 0.319 D_GP: 0.021 D_real: 1.326 D_fake: 0.674 +(epoch: 183, iters: 144, time: 0.064) G_GAN: 0.438 G_GAN_Feat: 0.818 G_ID: 0.174 G_Rec: 0.418 D_GP: 0.025 D_real: 1.190 D_fake: 0.584 +(epoch: 183, iters: 544, time: 0.064) G_GAN: 0.174 G_GAN_Feat: 0.679 G_ID: 0.147 G_Rec: 0.319 D_GP: 0.031 D_real: 1.037 D_fake: 0.827 +(epoch: 183, iters: 944, time: 0.064) G_GAN: 0.415 G_GAN_Feat: 0.863 G_ID: 0.186 G_Rec: 0.423 D_GP: 0.026 D_real: 1.052 D_fake: 0.593 +(epoch: 183, iters: 1344, time: 0.064) G_GAN: 0.321 G_GAN_Feat: 0.698 G_ID: 0.144 G_Rec: 0.294 D_GP: 0.056 D_real: 1.181 D_fake: 0.679 +(epoch: 183, iters: 1744, time: 0.064) G_GAN: 0.244 G_GAN_Feat: 1.034 G_ID: 0.214 G_Rec: 0.417 D_GP: 0.222 D_real: 0.324 D_fake: 0.760 +(epoch: 183, iters: 2144, time: 0.064) G_GAN: 0.288 G_GAN_Feat: 0.801 G_ID: 0.144 G_Rec: 0.328 D_GP: 0.048 D_real: 0.856 D_fake: 0.713 +(epoch: 183, iters: 2544, time: 0.064) G_GAN: 0.592 G_GAN_Feat: 0.784 G_ID: 0.189 G_Rec: 0.392 D_GP: 0.022 D_real: 1.374 D_fake: 0.414 +(epoch: 183, iters: 2944, time: 0.064) G_GAN: 0.356 G_GAN_Feat: 0.769 G_ID: 0.144 G_Rec: 0.321 D_GP: 0.077 D_real: 0.926 D_fake: 0.644 +(epoch: 183, iters: 3344, time: 0.064) G_GAN: 0.246 G_GAN_Feat: 0.906 G_ID: 0.186 G_Rec: 0.380 D_GP: 0.092 D_real: 0.761 D_fake: 0.758 +(epoch: 183, iters: 3744, time: 0.064) G_GAN: 0.345 G_GAN_Feat: 0.704 G_ID: 0.144 G_Rec: 0.300 D_GP: 0.030 D_real: 1.136 D_fake: 0.657 +(epoch: 183, iters: 4144, time: 0.064) G_GAN: 0.186 G_GAN_Feat: 0.893 G_ID: 0.196 G_Rec: 0.385 D_GP: 0.041 D_real: 0.757 D_fake: 0.814 +(epoch: 183, iters: 4544, time: 0.064) G_GAN: 0.232 G_GAN_Feat: 0.818 G_ID: 0.133 G_Rec: 0.360 D_GP: 0.063 D_real: 0.918 D_fake: 0.771 +(epoch: 183, iters: 4944, time: 0.064) G_GAN: 0.210 G_GAN_Feat: 0.802 G_ID: 0.178 G_Rec: 0.376 D_GP: 0.031 D_real: 0.873 D_fake: 0.790 +(epoch: 183, iters: 5344, time: 0.064) G_GAN: 0.307 G_GAN_Feat: 0.765 G_ID: 0.116 G_Rec: 0.303 D_GP: 0.038 D_real: 1.133 D_fake: 0.693 +(epoch: 183, iters: 5744, time: 0.064) G_GAN: 0.400 G_GAN_Feat: 0.842 G_ID: 0.196 G_Rec: 0.392 D_GP: 0.028 D_real: 1.080 D_fake: 0.602 +(epoch: 183, iters: 6144, time: 0.064) G_GAN: 0.197 G_GAN_Feat: 0.882 G_ID: 0.140 G_Rec: 0.302 D_GP: 0.182 D_real: 0.450 D_fake: 0.803 +(epoch: 183, iters: 6544, time: 0.064) G_GAN: 0.284 G_GAN_Feat: 1.015 G_ID: 0.189 G_Rec: 0.461 D_GP: 0.082 D_real: 0.826 D_fake: 0.717 +(epoch: 183, iters: 6944, time: 0.064) G_GAN: 0.416 G_GAN_Feat: 0.663 G_ID: 0.139 G_Rec: 0.295 D_GP: 0.024 D_real: 1.310 D_fake: 0.585 +(epoch: 183, iters: 7344, time: 0.064) G_GAN: 0.353 G_GAN_Feat: 1.111 G_ID: 0.207 G_Rec: 0.402 D_GP: 0.034 D_real: 0.453 D_fake: 0.649 +(epoch: 183, iters: 7744, time: 0.064) G_GAN: 0.240 G_GAN_Feat: 0.665 G_ID: 0.133 G_Rec: 0.305 D_GP: 0.025 D_real: 1.173 D_fake: 0.762 +(epoch: 183, iters: 8144, time: 0.064) G_GAN: 0.349 G_GAN_Feat: 0.876 G_ID: 0.168 G_Rec: 0.405 D_GP: 0.027 D_real: 1.012 D_fake: 0.651 +(epoch: 183, iters: 8544, time: 0.064) G_GAN: 0.163 G_GAN_Feat: 0.763 G_ID: 0.148 G_Rec: 0.319 D_GP: 0.028 D_real: 1.080 D_fake: 0.838 +(epoch: 184, iters: 336, time: 0.064) G_GAN: 0.280 G_GAN_Feat: 1.025 G_ID: 0.226 G_Rec: 0.448 D_GP: 0.086 D_real: 0.798 D_fake: 0.721 +(epoch: 184, iters: 736, time: 0.064) G_GAN: 0.170 G_GAN_Feat: 0.872 G_ID: 0.147 G_Rec: 0.334 D_GP: 0.175 D_real: 0.732 D_fake: 0.840 +(epoch: 184, iters: 1136, time: 0.064) G_GAN: 0.303 G_GAN_Feat: 0.971 G_ID: 0.195 G_Rec: 0.390 D_GP: 0.084 D_real: 0.661 D_fake: 0.697 +(epoch: 184, iters: 1536, time: 0.064) G_GAN: 0.350 G_GAN_Feat: 0.776 G_ID: 0.133 G_Rec: 0.344 D_GP: 0.026 D_real: 1.175 D_fake: 0.652 +(epoch: 184, iters: 1936, time: 0.064) G_GAN: 0.326 G_GAN_Feat: 0.943 G_ID: 0.211 G_Rec: 0.403 D_GP: 0.027 D_real: 1.019 D_fake: 0.677 +(epoch: 184, iters: 2336, time: 0.064) G_GAN: 0.006 G_GAN_Feat: 0.711 G_ID: 0.176 G_Rec: 0.352 D_GP: 0.037 D_real: 0.796 D_fake: 0.994 +(epoch: 184, iters: 2736, time: 0.064) G_GAN: 0.628 G_GAN_Feat: 0.851 G_ID: 0.172 G_Rec: 0.372 D_GP: 0.028 D_real: 1.333 D_fake: 0.384 +(epoch: 184, iters: 3136, time: 0.064) G_GAN: -0.093 G_GAN_Feat: 0.796 G_ID: 0.130 G_Rec: 0.330 D_GP: 0.025 D_real: 0.883 D_fake: 1.098 +(epoch: 184, iters: 3536, time: 0.064) G_GAN: 0.903 G_GAN_Feat: 0.948 G_ID: 0.152 G_Rec: 0.399 D_GP: 0.026 D_real: 1.482 D_fake: 0.152 +(epoch: 184, iters: 3936, time: 0.064) G_GAN: 0.322 G_GAN_Feat: 0.912 G_ID: 0.151 G_Rec: 0.382 D_GP: 0.078 D_real: 1.024 D_fake: 0.714 +(epoch: 184, iters: 4336, time: 0.064) G_GAN: 0.246 G_GAN_Feat: 0.799 G_ID: 0.214 G_Rec: 0.402 D_GP: 0.021 D_real: 1.016 D_fake: 0.754 +(epoch: 184, iters: 4736, time: 0.064) G_GAN: 0.203 G_GAN_Feat: 0.766 G_ID: 0.138 G_Rec: 0.327 D_GP: 0.044 D_real: 0.879 D_fake: 0.797 +(epoch: 184, iters: 5136, time: 0.064) G_GAN: 0.547 G_GAN_Feat: 0.933 G_ID: 0.186 G_Rec: 0.409 D_GP: 0.038 D_real: 1.036 D_fake: 0.457 +(epoch: 184, iters: 5536, time: 0.064) G_GAN: 0.664 G_GAN_Feat: 0.848 G_ID: 0.107 G_Rec: 0.358 D_GP: 0.167 D_real: 1.265 D_fake: 0.366 +(epoch: 184, iters: 5936, time: 0.064) G_GAN: 0.240 G_GAN_Feat: 1.010 G_ID: 0.180 G_Rec: 0.433 D_GP: 0.104 D_real: 0.472 D_fake: 0.761 +(epoch: 184, iters: 6336, time: 0.064) G_GAN: 0.263 G_GAN_Feat: 0.819 G_ID: 0.136 G_Rec: 0.322 D_GP: 0.038 D_real: 0.944 D_fake: 0.738 +(epoch: 184, iters: 6736, time: 0.064) G_GAN: 0.595 G_GAN_Feat: 1.113 G_ID: 0.198 G_Rec: 0.415 D_GP: 0.091 D_real: 0.572 D_fake: 0.414 +(epoch: 184, iters: 7136, time: 0.064) G_GAN: 0.157 G_GAN_Feat: 0.830 G_ID: 0.168 G_Rec: 0.345 D_GP: 0.112 D_real: 0.739 D_fake: 0.844 +(epoch: 184, iters: 7536, time: 0.064) G_GAN: 0.349 G_GAN_Feat: 1.130 G_ID: 0.165 G_Rec: 0.425 D_GP: 0.138 D_real: 0.556 D_fake: 0.652 +(epoch: 184, iters: 7936, time: 0.064) G_GAN: 0.203 G_GAN_Feat: 0.743 G_ID: 0.129 G_Rec: 0.301 D_GP: 0.022 D_real: 1.015 D_fake: 0.797 +(epoch: 184, iters: 8336, time: 0.064) G_GAN: 0.548 G_GAN_Feat: 1.078 G_ID: 0.203 G_Rec: 0.454 D_GP: 0.031 D_real: 0.863 D_fake: 0.484 +(epoch: 185, iters: 128, time: 0.064) G_GAN: 0.466 G_GAN_Feat: 0.727 G_ID: 0.131 G_Rec: 0.293 D_GP: 0.022 D_real: 1.396 D_fake: 0.538 +(epoch: 185, iters: 528, time: 0.064) G_GAN: 0.372 G_GAN_Feat: 0.994 G_ID: 0.174 G_Rec: 0.414 D_GP: 0.040 D_real: 0.826 D_fake: 0.630 +(epoch: 185, iters: 928, time: 0.064) G_GAN: 0.560 G_GAN_Feat: 0.813 G_ID: 0.143 G_Rec: 0.340 D_GP: 0.026 D_real: 1.346 D_fake: 0.544 +(epoch: 185, iters: 1328, time: 0.064) G_GAN: 0.446 G_GAN_Feat: 1.038 G_ID: 0.189 G_Rec: 0.418 D_GP: 0.024 D_real: 1.230 D_fake: 0.555 +(epoch: 185, iters: 1728, time: 0.064) G_GAN: 0.214 G_GAN_Feat: 0.688 G_ID: 0.135 G_Rec: 0.311 D_GP: 0.027 D_real: 1.035 D_fake: 0.786 +(epoch: 185, iters: 2128, time: 0.064) G_GAN: 0.398 G_GAN_Feat: 1.004 G_ID: 0.192 G_Rec: 0.401 D_GP: 0.133 D_real: 0.731 D_fake: 0.617 +(epoch: 185, iters: 2528, time: 0.064) G_GAN: 0.249 G_GAN_Feat: 0.715 G_ID: 0.130 G_Rec: 0.308 D_GP: 0.021 D_real: 1.111 D_fake: 0.751 +(epoch: 185, iters: 2928, time: 0.064) G_GAN: 0.147 G_GAN_Feat: 1.102 G_ID: 0.186 G_Rec: 0.451 D_GP: 0.035 D_real: 0.520 D_fake: 0.854 +(epoch: 185, iters: 3328, time: 0.064) G_GAN: 0.215 G_GAN_Feat: 0.686 G_ID: 0.148 G_Rec: 0.325 D_GP: 0.022 D_real: 1.199 D_fake: 0.785 +(epoch: 185, iters: 3728, time: 0.064) G_GAN: 0.472 G_GAN_Feat: 1.111 G_ID: 0.187 G_Rec: 0.442 D_GP: 0.086 D_real: 0.743 D_fake: 0.608 +(epoch: 185, iters: 4128, time: 0.064) G_GAN: 0.487 G_GAN_Feat: 0.746 G_ID: 0.146 G_Rec: 0.328 D_GP: 0.021 D_real: 1.467 D_fake: 0.516 +(epoch: 185, iters: 4528, time: 0.064) G_GAN: 0.271 G_GAN_Feat: 0.818 G_ID: 0.182 G_Rec: 0.417 D_GP: 0.022 D_real: 1.043 D_fake: 0.731 +(epoch: 185, iters: 4928, time: 0.064) G_GAN: -0.084 G_GAN_Feat: 0.776 G_ID: 0.157 G_Rec: 0.329 D_GP: 0.059 D_real: 0.769 D_fake: 1.084 +(epoch: 185, iters: 5328, time: 0.064) G_GAN: 0.615 G_GAN_Feat: 0.827 G_ID: 0.176 G_Rec: 0.379 D_GP: 0.022 D_real: 1.353 D_fake: 0.398 +(epoch: 185, iters: 5728, time: 0.064) G_GAN: 0.260 G_GAN_Feat: 0.688 G_ID: 0.142 G_Rec: 0.313 D_GP: 0.027 D_real: 1.156 D_fake: 0.740 +(epoch: 185, iters: 6128, time: 0.064) G_GAN: 0.531 G_GAN_Feat: 1.114 G_ID: 0.200 G_Rec: 0.463 D_GP: 0.140 D_real: 0.476 D_fake: 0.485 +(epoch: 185, iters: 6528, time: 0.064) G_GAN: -0.015 G_GAN_Feat: 0.866 G_ID: 0.142 G_Rec: 0.328 D_GP: 0.131 D_real: 0.393 D_fake: 1.015 +(epoch: 185, iters: 6928, time: 0.064) G_GAN: 0.359 G_GAN_Feat: 0.968 G_ID: 0.180 G_Rec: 0.443 D_GP: 0.027 D_real: 0.994 D_fake: 0.642 +(epoch: 185, iters: 7328, time: 0.064) G_GAN: 0.180 G_GAN_Feat: 0.745 G_ID: 0.120 G_Rec: 0.291 D_GP: 0.025 D_real: 0.983 D_fake: 0.821 +(epoch: 185, iters: 7728, time: 0.064) G_GAN: 0.518 G_GAN_Feat: 0.941 G_ID: 0.186 G_Rec: 0.404 D_GP: 0.027 D_real: 1.115 D_fake: 0.484 +(epoch: 185, iters: 8128, time: 0.064) G_GAN: 0.502 G_GAN_Feat: 0.876 G_ID: 0.137 G_Rec: 0.311 D_GP: 0.073 D_real: 0.979 D_fake: 0.518 +(epoch: 185, iters: 8528, time: 0.064) G_GAN: 0.141 G_GAN_Feat: 0.929 G_ID: 0.209 G_Rec: 0.386 D_GP: 0.037 D_real: 0.877 D_fake: 0.859 +(epoch: 186, iters: 320, time: 0.064) G_GAN: 0.315 G_GAN_Feat: 0.764 G_ID: 0.137 G_Rec: 0.309 D_GP: 0.048 D_real: 1.169 D_fake: 0.689 +(epoch: 186, iters: 720, time: 0.064) G_GAN: 0.201 G_GAN_Feat: 0.949 G_ID: 0.218 G_Rec: 0.438 D_GP: 0.084 D_real: 0.808 D_fake: 0.800 +(epoch: 186, iters: 1120, time: 0.064) G_GAN: 0.144 G_GAN_Feat: 0.757 G_ID: 0.139 G_Rec: 0.317 D_GP: 0.027 D_real: 0.988 D_fake: 0.857 +(epoch: 186, iters: 1520, time: 0.064) G_GAN: 0.559 G_GAN_Feat: 0.898 G_ID: 0.157 G_Rec: 0.395 D_GP: 0.027 D_real: 1.188 D_fake: 0.452 +(epoch: 186, iters: 1920, time: 0.064) G_GAN: -0.094 G_GAN_Feat: 0.783 G_ID: 0.160 G_Rec: 0.302 D_GP: 0.021 D_real: 0.792 D_fake: 1.094 +(epoch: 186, iters: 2320, time: 0.064) G_GAN: 0.446 G_GAN_Feat: 1.102 G_ID: 0.216 G_Rec: 0.406 D_GP: 0.564 D_real: 0.450 D_fake: 0.565 +(epoch: 186, iters: 2720, time: 0.064) G_GAN: 0.296 G_GAN_Feat: 0.638 G_ID: 0.139 G_Rec: 0.322 D_GP: 0.020 D_real: 1.257 D_fake: 0.704 +(epoch: 186, iters: 3120, time: 0.064) G_GAN: 0.258 G_GAN_Feat: 1.056 G_ID: 0.219 G_Rec: 0.432 D_GP: 0.117 D_real: 0.399 D_fake: 0.746 +(epoch: 186, iters: 3520, time: 0.064) G_GAN: 0.252 G_GAN_Feat: 0.787 G_ID: 0.130 G_Rec: 0.305 D_GP: 0.040 D_real: 1.016 D_fake: 0.748 +(epoch: 186, iters: 3920, time: 0.064) G_GAN: 0.414 G_GAN_Feat: 1.015 G_ID: 0.203 G_Rec: 0.417 D_GP: 0.050 D_real: 0.798 D_fake: 0.592 +(epoch: 186, iters: 4320, time: 0.064) G_GAN: 0.223 G_GAN_Feat: 1.039 G_ID: 0.134 G_Rec: 0.372 D_GP: 0.089 D_real: 0.425 D_fake: 0.778 +(epoch: 186, iters: 4720, time: 0.064) G_GAN: 0.570 G_GAN_Feat: 0.802 G_ID: 0.171 G_Rec: 0.406 D_GP: 0.017 D_real: 1.396 D_fake: 0.437 +(epoch: 186, iters: 5120, time: 0.064) G_GAN: 0.096 G_GAN_Feat: 0.720 G_ID: 0.148 G_Rec: 0.356 D_GP: 0.033 D_real: 0.936 D_fake: 0.904 +(epoch: 186, iters: 5520, time: 0.064) G_GAN: 0.466 G_GAN_Feat: 0.990 G_ID: 0.235 G_Rec: 0.413 D_GP: 0.032 D_real: 0.922 D_fake: 0.537 +(epoch: 186, iters: 5920, time: 0.064) G_GAN: 0.168 G_GAN_Feat: 0.840 G_ID: 0.130 G_Rec: 0.333 D_GP: 0.068 D_real: 0.865 D_fake: 0.832 +(epoch: 186, iters: 6320, time: 0.064) G_GAN: 0.303 G_GAN_Feat: 0.922 G_ID: 0.192 G_Rec: 0.416 D_GP: 0.033 D_real: 0.870 D_fake: 0.699 +(epoch: 186, iters: 6720, time: 0.064) G_GAN: -0.011 G_GAN_Feat: 0.724 G_ID: 0.157 G_Rec: 0.315 D_GP: 0.027 D_real: 0.970 D_fake: 1.013 +(epoch: 186, iters: 7120, time: 0.064) G_GAN: 0.327 G_GAN_Feat: 1.017 G_ID: 0.185 G_Rec: 0.403 D_GP: 0.043 D_real: 0.970 D_fake: 0.677 +(epoch: 186, iters: 7520, time: 0.064) G_GAN: 0.064 G_GAN_Feat: 0.748 G_ID: 0.133 G_Rec: 0.303 D_GP: 0.018 D_real: 1.026 D_fake: 0.937 +(epoch: 186, iters: 7920, time: 0.064) G_GAN: 0.400 G_GAN_Feat: 0.905 G_ID: 0.219 G_Rec: 0.416 D_GP: 0.022 D_real: 1.125 D_fake: 0.604 +(epoch: 186, iters: 8320, time: 0.064) G_GAN: 0.219 G_GAN_Feat: 0.768 G_ID: 0.144 G_Rec: 0.316 D_GP: 0.038 D_real: 0.895 D_fake: 0.783 +(epoch: 187, iters: 112, time: 0.064) G_GAN: 0.367 G_GAN_Feat: 0.849 G_ID: 0.163 G_Rec: 0.400 D_GP: 0.024 D_real: 1.166 D_fake: 0.635 +(epoch: 187, iters: 512, time: 0.064) G_GAN: 0.295 G_GAN_Feat: 0.807 G_ID: 0.141 G_Rec: 0.346 D_GP: 0.025 D_real: 1.195 D_fake: 0.705 +(epoch: 187, iters: 912, time: 0.064) G_GAN: 0.294 G_GAN_Feat: 0.885 G_ID: 0.172 G_Rec: 0.407 D_GP: 0.024 D_real: 1.124 D_fake: 0.706 +(epoch: 187, iters: 1312, time: 0.064) G_GAN: 0.175 G_GAN_Feat: 0.705 G_ID: 0.138 G_Rec: 0.319 D_GP: 0.027 D_real: 1.029 D_fake: 0.828 +(epoch: 187, iters: 1712, time: 0.064) G_GAN: 0.368 G_GAN_Feat: 0.995 G_ID: 0.208 G_Rec: 0.413 D_GP: 0.027 D_real: 1.005 D_fake: 0.633 +(epoch: 187, iters: 2112, time: 0.064) G_GAN: 0.153 G_GAN_Feat: 0.718 G_ID: 0.122 G_Rec: 0.302 D_GP: 0.020 D_real: 1.023 D_fake: 0.847 +(epoch: 187, iters: 2512, time: 0.064) G_GAN: 0.365 G_GAN_Feat: 0.953 G_ID: 0.192 G_Rec: 0.424 D_GP: 0.027 D_real: 1.064 D_fake: 0.635 +(epoch: 187, iters: 2912, time: 0.064) G_GAN: 0.194 G_GAN_Feat: 1.027 G_ID: 0.140 G_Rec: 0.363 D_GP: 0.086 D_real: 0.948 D_fake: 0.807 +(epoch: 187, iters: 3312, time: 0.064) G_GAN: 0.351 G_GAN_Feat: 0.845 G_ID: 0.208 G_Rec: 0.352 D_GP: 0.026 D_real: 1.101 D_fake: 0.651 +(epoch: 187, iters: 3712, time: 0.064) G_GAN: 0.305 G_GAN_Feat: 0.709 G_ID: 0.165 G_Rec: 0.314 D_GP: 0.025 D_real: 1.249 D_fake: 0.699 +(epoch: 187, iters: 4112, time: 0.064) G_GAN: 0.419 G_GAN_Feat: 0.828 G_ID: 0.192 G_Rec: 0.383 D_GP: 0.023 D_real: 1.207 D_fake: 0.584 +(epoch: 187, iters: 4512, time: 0.064) G_GAN: -0.122 G_GAN_Feat: 0.871 G_ID: 0.146 G_Rec: 0.326 D_GP: 0.164 D_real: 0.178 D_fake: 1.122 +(epoch: 187, iters: 4912, time: 0.064) G_GAN: 0.117 G_GAN_Feat: 0.847 G_ID: 0.181 G_Rec: 0.469 D_GP: 0.025 D_real: 0.883 D_fake: 0.883 +(epoch: 187, iters: 5312, time: 0.064) G_GAN: 0.289 G_GAN_Feat: 0.734 G_ID: 0.138 G_Rec: 0.300 D_GP: 0.033 D_real: 1.111 D_fake: 0.711 +(epoch: 187, iters: 5712, time: 0.064) G_GAN: 0.923 G_GAN_Feat: 0.989 G_ID: 0.183 G_Rec: 0.423 D_GP: 0.028 D_real: 1.405 D_fake: 0.190 +(epoch: 187, iters: 6112, time: 0.064) G_GAN: -0.082 G_GAN_Feat: 0.983 G_ID: 0.175 G_Rec: 0.340 D_GP: 0.118 D_real: 0.423 D_fake: 1.082 +(epoch: 187, iters: 6512, time: 0.064) G_GAN: 0.435 G_GAN_Feat: 0.768 G_ID: 0.183 G_Rec: 0.384 D_GP: 0.019 D_real: 1.195 D_fake: 0.579 +(epoch: 187, iters: 6912, time: 0.064) G_GAN: 0.267 G_GAN_Feat: 0.949 G_ID: 0.152 G_Rec: 0.333 D_GP: 0.172 D_real: 0.453 D_fake: 0.741 +(epoch: 187, iters: 7312, time: 0.064) G_GAN: 0.559 G_GAN_Feat: 1.122 G_ID: 0.184 G_Rec: 0.456 D_GP: 0.080 D_real: 0.598 D_fake: 0.488 +(epoch: 187, iters: 7712, time: 0.064) G_GAN: 0.122 G_GAN_Feat: 0.734 G_ID: 0.120 G_Rec: 0.285 D_GP: 0.023 D_real: 1.064 D_fake: 0.878 +(epoch: 187, iters: 8112, time: 0.064) G_GAN: 0.522 G_GAN_Feat: 1.076 G_ID: 0.179 G_Rec: 0.396 D_GP: 0.031 D_real: 0.829 D_fake: 0.492 +(epoch: 187, iters: 8512, time: 0.064) G_GAN: 0.211 G_GAN_Feat: 0.923 G_ID: 0.138 G_Rec: 0.333 D_GP: 0.051 D_real: 0.587 D_fake: 0.789 +(epoch: 188, iters: 304, time: 0.064) G_GAN: 0.878 G_GAN_Feat: 1.021 G_ID: 0.163 G_Rec: 0.429 D_GP: 0.027 D_real: 1.388 D_fake: 0.213 +(epoch: 188, iters: 704, time: 0.064) G_GAN: -0.213 G_GAN_Feat: 0.713 G_ID: 0.156 G_Rec: 0.291 D_GP: 0.025 D_real: 0.818 D_fake: 1.213 +(epoch: 188, iters: 1104, time: 0.064) G_GAN: 0.266 G_GAN_Feat: 0.971 G_ID: 0.191 G_Rec: 0.400 D_GP: 0.027 D_real: 0.830 D_fake: 0.734 +(epoch: 188, iters: 1504, time: 0.064) G_GAN: 0.277 G_GAN_Feat: 0.938 G_ID: 0.132 G_Rec: 0.327 D_GP: 0.101 D_real: 0.459 D_fake: 0.724 +(epoch: 188, iters: 1904, time: 0.064) G_GAN: 0.415 G_GAN_Feat: 0.790 G_ID: 0.186 G_Rec: 0.409 D_GP: 0.019 D_real: 1.235 D_fake: 0.586 +(epoch: 188, iters: 2304, time: 0.064) G_GAN: 0.319 G_GAN_Feat: 0.788 G_ID: 0.125 G_Rec: 0.305 D_GP: 0.031 D_real: 1.189 D_fake: 0.682 +(epoch: 188, iters: 2704, time: 0.064) G_GAN: 0.283 G_GAN_Feat: 0.850 G_ID: 0.213 G_Rec: 0.396 D_GP: 0.024 D_real: 1.032 D_fake: 0.717 +(epoch: 188, iters: 3104, time: 0.064) G_GAN: 0.388 G_GAN_Feat: 0.705 G_ID: 0.140 G_Rec: 0.330 D_GP: 0.022 D_real: 1.243 D_fake: 0.614 +(epoch: 188, iters: 3504, time: 0.064) G_GAN: 0.294 G_GAN_Feat: 0.824 G_ID: 0.194 G_Rec: 0.362 D_GP: 0.027 D_real: 1.000 D_fake: 0.707 +(epoch: 188, iters: 3904, time: 0.064) G_GAN: 0.053 G_GAN_Feat: 0.834 G_ID: 0.144 G_Rec: 0.317 D_GP: 0.110 D_real: 0.685 D_fake: 0.947 +(epoch: 188, iters: 4304, time: 0.064) G_GAN: 0.372 G_GAN_Feat: 0.893 G_ID: 0.196 G_Rec: 0.380 D_GP: 0.037 D_real: 0.926 D_fake: 0.631 +(epoch: 188, iters: 4704, time: 0.064) G_GAN: 0.166 G_GAN_Feat: 0.605 G_ID: 0.138 G_Rec: 0.288 D_GP: 0.023 D_real: 1.163 D_fake: 0.834 +(epoch: 188, iters: 5104, time: 0.064) G_GAN: 0.211 G_GAN_Feat: 1.058 G_ID: 0.187 G_Rec: 0.427 D_GP: 0.129 D_real: 0.467 D_fake: 0.789 +(epoch: 188, iters: 5504, time: 0.064) G_GAN: 0.248 G_GAN_Feat: 0.710 G_ID: 0.157 G_Rec: 0.333 D_GP: 0.023 D_real: 1.115 D_fake: 0.752 +(epoch: 188, iters: 5904, time: 0.064) G_GAN: 0.281 G_GAN_Feat: 0.947 G_ID: 0.181 G_Rec: 0.373 D_GP: 0.066 D_real: 0.675 D_fake: 0.722 +(epoch: 188, iters: 6304, time: 0.064) G_GAN: 0.240 G_GAN_Feat: 0.771 G_ID: 0.146 G_Rec: 0.292 D_GP: 0.031 D_real: 1.050 D_fake: 0.763 +(epoch: 188, iters: 6704, time: 0.064) G_GAN: 0.558 G_GAN_Feat: 1.002 G_ID: 0.177 G_Rec: 0.398 D_GP: 0.035 D_real: 1.052 D_fake: 0.449 +(epoch: 188, iters: 7104, time: 0.064) G_GAN: 0.262 G_GAN_Feat: 0.808 G_ID: 0.147 G_Rec: 0.301 D_GP: 0.050 D_real: 0.912 D_fake: 0.740 +(epoch: 188, iters: 7504, time: 0.064) G_GAN: 0.400 G_GAN_Feat: 0.969 G_ID: 0.172 G_Rec: 0.383 D_GP: 0.025 D_real: 1.005 D_fake: 0.602 +(epoch: 188, iters: 7904, time: 0.064) G_GAN: 0.462 G_GAN_Feat: 0.635 G_ID: 0.125 G_Rec: 0.318 D_GP: 0.020 D_real: 1.458 D_fake: 0.543 +(epoch: 188, iters: 8304, time: 0.064) G_GAN: 0.147 G_GAN_Feat: 0.882 G_ID: 0.207 G_Rec: 0.416 D_GP: 0.031 D_real: 0.777 D_fake: 0.854 +(epoch: 189, iters: 96, time: 0.064) G_GAN: 0.191 G_GAN_Feat: 0.719 G_ID: 0.131 G_Rec: 0.298 D_GP: 0.040 D_real: 0.980 D_fake: 0.809 +(epoch: 189, iters: 496, time: 0.064) G_GAN: 0.514 G_GAN_Feat: 0.860 G_ID: 0.199 G_Rec: 0.382 D_GP: 0.029 D_real: 1.188 D_fake: 0.501 +(epoch: 189, iters: 896, time: 0.064) G_GAN: 0.276 G_GAN_Feat: 0.733 G_ID: 0.116 G_Rec: 0.308 D_GP: 0.042 D_real: 0.996 D_fake: 0.725 +(epoch: 189, iters: 1296, time: 0.064) G_GAN: 0.817 G_GAN_Feat: 1.011 G_ID: 0.197 G_Rec: 0.416 D_GP: 0.084 D_real: 1.229 D_fake: 0.455 +(epoch: 189, iters: 1696, time: 0.064) G_GAN: 0.318 G_GAN_Feat: 0.733 G_ID: 0.133 G_Rec: 0.360 D_GP: 0.023 D_real: 1.203 D_fake: 0.683 +(epoch: 189, iters: 2096, time: 0.064) G_GAN: 0.295 G_GAN_Feat: 1.021 G_ID: 0.186 G_Rec: 0.457 D_GP: 0.773 D_real: 0.586 D_fake: 0.707 +(epoch: 189, iters: 2496, time: 0.064) G_GAN: 0.042 G_GAN_Feat: 0.615 G_ID: 0.167 G_Rec: 0.295 D_GP: 0.020 D_real: 0.936 D_fake: 0.958 +(epoch: 189, iters: 2896, time: 0.064) G_GAN: 0.262 G_GAN_Feat: 1.159 G_ID: 0.195 G_Rec: 0.457 D_GP: 0.404 D_real: 0.433 D_fake: 0.742 +(epoch: 189, iters: 3296, time: 0.064) G_GAN: 0.261 G_GAN_Feat: 0.834 G_ID: 0.159 G_Rec: 0.388 D_GP: 0.102 D_real: 0.814 D_fake: 0.751 +(epoch: 189, iters: 3696, time: 0.064) G_GAN: 0.489 G_GAN_Feat: 0.812 G_ID: 0.158 G_Rec: 0.398 D_GP: 0.027 D_real: 1.283 D_fake: 0.516 +(epoch: 189, iters: 4096, time: 0.064) G_GAN: 0.110 G_GAN_Feat: 0.744 G_ID: 0.161 G_Rec: 0.299 D_GP: 0.032 D_real: 0.980 D_fake: 0.890 +(epoch: 189, iters: 4496, time: 0.064) G_GAN: 0.505 G_GAN_Feat: 0.857 G_ID: 0.190 G_Rec: 0.393 D_GP: 0.026 D_real: 1.184 D_fake: 0.499 +(epoch: 189, iters: 4896, time: 0.064) G_GAN: 0.431 G_GAN_Feat: 0.723 G_ID: 0.131 G_Rec: 0.323 D_GP: 0.023 D_real: 1.295 D_fake: 0.569 +(epoch: 189, iters: 5296, time: 0.064) G_GAN: 0.378 G_GAN_Feat: 0.942 G_ID: 0.187 G_Rec: 0.394 D_GP: 0.029 D_real: 1.054 D_fake: 0.623 +(epoch: 189, iters: 5696, time: 0.064) G_GAN: 0.404 G_GAN_Feat: 0.573 G_ID: 0.123 G_Rec: 0.293 D_GP: 0.026 D_real: 1.400 D_fake: 0.596 +(epoch: 189, iters: 6096, time: 0.064) G_GAN: 0.289 G_GAN_Feat: 0.919 G_ID: 0.183 G_Rec: 0.457 D_GP: 0.105 D_real: 0.725 D_fake: 0.712 +(epoch: 189, iters: 6496, time: 0.064) G_GAN: -0.029 G_GAN_Feat: 0.661 G_ID: 0.127 G_Rec: 0.332 D_GP: 0.027 D_real: 0.880 D_fake: 1.029 +(epoch: 189, iters: 6896, time: 0.064) G_GAN: 0.448 G_GAN_Feat: 0.844 G_ID: 0.192 G_Rec: 0.340 D_GP: 0.037 D_real: 1.176 D_fake: 0.554 +(epoch: 189, iters: 7296, time: 0.064) G_GAN: 0.322 G_GAN_Feat: 0.774 G_ID: 0.177 G_Rec: 0.313 D_GP: 0.038 D_real: 1.083 D_fake: 0.690 +(epoch: 189, iters: 7696, time: 0.064) G_GAN: 0.281 G_GAN_Feat: 0.969 G_ID: 0.197 G_Rec: 0.378 D_GP: 0.056 D_real: 0.580 D_fake: 0.719 +(epoch: 189, iters: 8096, time: 0.064) G_GAN: 0.227 G_GAN_Feat: 0.714 G_ID: 0.172 G_Rec: 0.314 D_GP: 0.024 D_real: 1.172 D_fake: 0.773 +(epoch: 189, iters: 8496, time: 0.064) G_GAN: 0.448 G_GAN_Feat: 1.146 G_ID: 0.202 G_Rec: 0.462 D_GP: 0.063 D_real: 0.324 D_fake: 0.554 +(epoch: 190, iters: 288, time: 0.064) G_GAN: 0.098 G_GAN_Feat: 0.563 G_ID: 0.149 G_Rec: 0.298 D_GP: 0.019 D_real: 1.038 D_fake: 0.902 +(epoch: 190, iters: 688, time: 0.064) G_GAN: 0.432 G_GAN_Feat: 0.760 G_ID: 0.181 G_Rec: 0.399 D_GP: 0.021 D_real: 1.152 D_fake: 0.569 +(epoch: 190, iters: 1088, time: 0.064) G_GAN: 0.155 G_GAN_Feat: 0.573 G_ID: 0.132 G_Rec: 0.302 D_GP: 0.020 D_real: 1.110 D_fake: 0.845 +(epoch: 190, iters: 1488, time: 0.064) G_GAN: 0.407 G_GAN_Feat: 0.807 G_ID: 0.200 G_Rec: 0.395 D_GP: 0.039 D_real: 1.115 D_fake: 0.598 +(epoch: 190, iters: 1888, time: 0.064) G_GAN: 0.112 G_GAN_Feat: 0.659 G_ID: 0.125 G_Rec: 0.319 D_GP: 0.034 D_real: 1.003 D_fake: 0.888 +(epoch: 190, iters: 2288, time: 0.064) G_GAN: 0.224 G_GAN_Feat: 0.786 G_ID: 0.182 G_Rec: 0.381 D_GP: 0.027 D_real: 0.986 D_fake: 0.776 +(epoch: 190, iters: 2688, time: 0.064) G_GAN: 0.173 G_GAN_Feat: 0.671 G_ID: 0.117 G_Rec: 0.301 D_GP: 0.053 D_real: 0.988 D_fake: 0.828 +(epoch: 190, iters: 3088, time: 0.064) G_GAN: 0.185 G_GAN_Feat: 0.803 G_ID: 0.208 G_Rec: 0.365 D_GP: 0.062 D_real: 0.823 D_fake: 0.816 +(epoch: 190, iters: 3488, time: 0.064) G_GAN: 0.314 G_GAN_Feat: 0.736 G_ID: 0.139 G_Rec: 0.303 D_GP: 0.077 D_real: 0.947 D_fake: 0.692 +(epoch: 190, iters: 3888, time: 0.064) G_GAN: 0.660 G_GAN_Feat: 0.843 G_ID: 0.174 G_Rec: 0.380 D_GP: 0.025 D_real: 1.362 D_fake: 0.375 +(epoch: 190, iters: 4288, time: 0.064) G_GAN: 0.431 G_GAN_Feat: 0.708 G_ID: 0.125 G_Rec: 0.306 D_GP: 0.034 D_real: 1.266 D_fake: 0.603 +(epoch: 190, iters: 4688, time: 0.064) G_GAN: 0.173 G_GAN_Feat: 0.888 G_ID: 0.208 G_Rec: 0.372 D_GP: 0.032 D_real: 0.838 D_fake: 0.827 +(epoch: 190, iters: 5088, time: 0.064) G_GAN: 0.192 G_GAN_Feat: 0.938 G_ID: 0.151 G_Rec: 0.303 D_GP: 0.060 D_real: 0.536 D_fake: 0.808 +(epoch: 190, iters: 5488, time: 0.064) G_GAN: 0.641 G_GAN_Feat: 0.881 G_ID: 0.190 G_Rec: 0.397 D_GP: 0.026 D_real: 1.291 D_fake: 0.373 +(epoch: 190, iters: 5888, time: 0.064) G_GAN: 0.459 G_GAN_Feat: 0.986 G_ID: 0.172 G_Rec: 0.336 D_GP: 0.079 D_real: 0.994 D_fake: 0.697 +(epoch: 190, iters: 6288, time: 0.064) G_GAN: 0.885 G_GAN_Feat: 0.911 G_ID: 0.196 G_Rec: 0.406 D_GP: 0.029 D_real: 1.648 D_fake: 0.222 +(epoch: 190, iters: 6688, time: 0.064) G_GAN: 0.584 G_GAN_Feat: 1.004 G_ID: 0.140 G_Rec: 0.363 D_GP: 0.085 D_real: 0.770 D_fake: 0.485 +(epoch: 190, iters: 7088, time: 0.064) G_GAN: 0.655 G_GAN_Feat: 0.926 G_ID: 0.176 G_Rec: 0.404 D_GP: 0.030 D_real: 1.265 D_fake: 0.354 +(epoch: 190, iters: 7488, time: 0.064) G_GAN: 0.154 G_GAN_Feat: 0.823 G_ID: 0.144 G_Rec: 0.311 D_GP: 0.064 D_real: 0.724 D_fake: 0.846 +(epoch: 190, iters: 7888, time: 0.064) G_GAN: 0.445 G_GAN_Feat: 0.848 G_ID: 0.206 G_Rec: 0.400 D_GP: 0.025 D_real: 1.144 D_fake: 0.557 +(epoch: 190, iters: 8288, time: 0.064) G_GAN: 0.273 G_GAN_Feat: 0.775 G_ID: 0.124 G_Rec: 0.337 D_GP: 0.040 D_real: 1.012 D_fake: 0.727 +(epoch: 191, iters: 80, time: 0.064) G_GAN: 0.682 G_GAN_Feat: 1.216 G_ID: 0.200 G_Rec: 0.454 D_GP: 0.277 D_real: 0.460 D_fake: 0.374 +(epoch: 191, iters: 480, time: 0.064) G_GAN: 0.289 G_GAN_Feat: 0.695 G_ID: 0.138 G_Rec: 0.369 D_GP: 0.021 D_real: 1.166 D_fake: 0.711 +(epoch: 191, iters: 880, time: 0.064) G_GAN: 0.448 G_GAN_Feat: 0.843 G_ID: 0.185 G_Rec: 0.381 D_GP: 0.028 D_real: 1.138 D_fake: 0.556 +(epoch: 191, iters: 1280, time: 0.064) G_GAN: 0.052 G_GAN_Feat: 0.824 G_ID: 0.142 G_Rec: 0.328 D_GP: 0.106 D_real: 0.468 D_fake: 0.948 +(epoch: 191, iters: 1680, time: 0.064) G_GAN: 0.391 G_GAN_Feat: 0.829 G_ID: 0.169 G_Rec: 0.377 D_GP: 0.022 D_real: 1.153 D_fake: 0.609 +(epoch: 191, iters: 2080, time: 0.064) G_GAN: 0.238 G_GAN_Feat: 0.831 G_ID: 0.136 G_Rec: 0.389 D_GP: 0.064 D_real: 0.936 D_fake: 0.765 +(epoch: 191, iters: 2480, time: 0.064) G_GAN: 0.514 G_GAN_Feat: 0.817 G_ID: 0.202 G_Rec: 0.390 D_GP: 0.029 D_real: 1.281 D_fake: 0.492 +(epoch: 191, iters: 2880, time: 0.064) G_GAN: 0.359 G_GAN_Feat: 0.895 G_ID: 0.144 G_Rec: 0.367 D_GP: 0.140 D_real: 0.548 D_fake: 0.655 +(epoch: 191, iters: 3280, time: 0.064) G_GAN: 0.374 G_GAN_Feat: 1.096 G_ID: 0.157 G_Rec: 0.375 D_GP: 0.047 D_real: 0.571 D_fake: 0.627 +(epoch: 191, iters: 3680, time: 0.064) G_GAN: 0.331 G_GAN_Feat: 0.826 G_ID: 0.131 G_Rec: 0.320 D_GP: 0.072 D_real: 1.161 D_fake: 0.672 +(epoch: 191, iters: 4080, time: 0.064) G_GAN: 0.435 G_GAN_Feat: 0.936 G_ID: 0.193 G_Rec: 0.396 D_GP: 0.031 D_real: 1.106 D_fake: 0.568 +(epoch: 191, iters: 4480, time: 0.064) G_GAN: 0.085 G_GAN_Feat: 0.794 G_ID: 0.124 G_Rec: 0.324 D_GP: 0.045 D_real: 0.640 D_fake: 0.915 +(epoch: 191, iters: 4880, time: 0.064) G_GAN: 0.350 G_GAN_Feat: 1.117 G_ID: 0.195 G_Rec: 0.454 D_GP: 2.094 D_real: 0.499 D_fake: 0.652 +(epoch: 191, iters: 5280, time: 0.064) G_GAN: 0.247 G_GAN_Feat: 0.601 G_ID: 0.137 G_Rec: 0.465 D_GP: 0.018 D_real: 1.230 D_fake: 0.753 +(epoch: 191, iters: 5680, time: 0.064) G_GAN: 0.380 G_GAN_Feat: 0.764 G_ID: 0.198 G_Rec: 0.405 D_GP: 0.019 D_real: 1.091 D_fake: 0.621 +(epoch: 191, iters: 6080, time: 0.064) G_GAN: 0.171 G_GAN_Feat: 0.693 G_ID: 0.140 G_Rec: 0.331 D_GP: 0.037 D_real: 0.940 D_fake: 0.829 +(epoch: 191, iters: 6480, time: 0.064) G_GAN: 0.232 G_GAN_Feat: 1.013 G_ID: 0.186 G_Rec: 0.404 D_GP: 0.093 D_real: 0.552 D_fake: 0.769 +(epoch: 191, iters: 6880, time: 0.064) G_GAN: 0.285 G_GAN_Feat: 0.746 G_ID: 0.143 G_Rec: 0.314 D_GP: 0.059 D_real: 0.927 D_fake: 0.715 +(epoch: 191, iters: 7280, time: 0.064) G_GAN: 0.656 G_GAN_Feat: 0.981 G_ID: 0.205 G_Rec: 0.426 D_GP: 0.059 D_real: 1.085 D_fake: 0.430 +(epoch: 191, iters: 7680, time: 0.064) G_GAN: 0.438 G_GAN_Feat: 0.776 G_ID: 0.151 G_Rec: 0.351 D_GP: 0.028 D_real: 1.386 D_fake: 0.575 +(epoch: 191, iters: 8080, time: 0.064) G_GAN: 0.093 G_GAN_Feat: 0.950 G_ID: 0.203 G_Rec: 0.371 D_GP: 0.076 D_real: 0.729 D_fake: 0.908 +(epoch: 191, iters: 8480, time: 0.064) G_GAN: 0.217 G_GAN_Feat: 0.749 G_ID: 0.132 G_Rec: 0.310 D_GP: 0.040 D_real: 0.964 D_fake: 0.784 +(epoch: 192, iters: 272, time: 0.064) G_GAN: 0.495 G_GAN_Feat: 0.823 G_ID: 0.192 G_Rec: 0.391 D_GP: 0.022 D_real: 1.333 D_fake: 0.508 +(epoch: 192, iters: 672, time: 0.064) G_GAN: 0.411 G_GAN_Feat: 0.803 G_ID: 0.146 G_Rec: 0.323 D_GP: 0.070 D_real: 1.019 D_fake: 0.593 +(epoch: 192, iters: 1072, time: 0.064) G_GAN: 0.557 G_GAN_Feat: 0.959 G_ID: 0.182 G_Rec: 0.418 D_GP: 0.037 D_real: 1.019 D_fake: 0.448 +(epoch: 192, iters: 1472, time: 0.064) G_GAN: 0.339 G_GAN_Feat: 0.732 G_ID: 0.139 G_Rec: 0.336 D_GP: 0.040 D_real: 1.122 D_fake: 0.665 +(epoch: 192, iters: 1872, time: 0.064) G_GAN: 0.483 G_GAN_Feat: 0.811 G_ID: 0.188 G_Rec: 0.396 D_GP: 0.025 D_real: 1.161 D_fake: 0.517 +(epoch: 192, iters: 2272, time: 0.064) G_GAN: 0.230 G_GAN_Feat: 0.646 G_ID: 0.138 G_Rec: 0.285 D_GP: 0.028 D_real: 1.097 D_fake: 0.770 +(epoch: 192, iters: 2672, time: 0.064) G_GAN: 0.347 G_GAN_Feat: 1.044 G_ID: 0.196 G_Rec: 0.443 D_GP: 0.112 D_real: 0.568 D_fake: 0.654 +(epoch: 192, iters: 3072, time: 0.064) G_GAN: 0.233 G_GAN_Feat: 0.663 G_ID: 0.139 G_Rec: 0.330 D_GP: 0.042 D_real: 1.136 D_fake: 0.767 +(epoch: 192, iters: 3472, time: 0.064) G_GAN: 0.700 G_GAN_Feat: 0.928 G_ID: 0.185 G_Rec: 0.415 D_GP: 0.036 D_real: 1.219 D_fake: 0.334 +(epoch: 192, iters: 3872, time: 0.064) G_GAN: 0.285 G_GAN_Feat: 0.757 G_ID: 0.144 G_Rec: 0.321 D_GP: 0.060 D_real: 1.014 D_fake: 0.715 +(epoch: 192, iters: 4272, time: 0.064) G_GAN: 0.363 G_GAN_Feat: 1.098 G_ID: 0.204 G_Rec: 0.450 D_GP: 0.202 D_real: 0.338 D_fake: 0.638 +(epoch: 192, iters: 4672, time: 0.064) G_GAN: 0.256 G_GAN_Feat: 0.683 G_ID: 0.126 G_Rec: 0.309 D_GP: 0.022 D_real: 1.166 D_fake: 0.744 +(epoch: 192, iters: 5072, time: 0.064) G_GAN: 0.387 G_GAN_Feat: 0.862 G_ID: 0.235 G_Rec: 0.392 D_GP: 0.030 D_real: 1.022 D_fake: 0.613 +(epoch: 192, iters: 5472, time: 0.064) G_GAN: 0.454 G_GAN_Feat: 0.721 G_ID: 0.128 G_Rec: 0.315 D_GP: 0.040 D_real: 1.260 D_fake: 0.559 +(epoch: 192, iters: 5872, time: 0.064) G_GAN: 0.331 G_GAN_Feat: 0.873 G_ID: 0.185 G_Rec: 0.384 D_GP: 0.024 D_real: 0.961 D_fake: 0.669 +(epoch: 192, iters: 6272, time: 0.064) G_GAN: 0.267 G_GAN_Feat: 0.777 G_ID: 0.139 G_Rec: 0.312 D_GP: 0.035 D_real: 1.029 D_fake: 0.739 +(epoch: 192, iters: 6672, time: 0.064) G_GAN: 0.489 G_GAN_Feat: 0.987 G_ID: 0.216 G_Rec: 0.429 D_GP: 0.054 D_real: 0.924 D_fake: 0.517 +(epoch: 192, iters: 7072, time: 0.064) G_GAN: -0.049 G_GAN_Feat: 1.017 G_ID: 0.144 G_Rec: 0.431 D_GP: 0.291 D_real: 0.382 D_fake: 1.051 +(epoch: 192, iters: 7472, time: 0.064) G_GAN: 0.268 G_GAN_Feat: 0.869 G_ID: 0.200 G_Rec: 0.402 D_GP: 0.024 D_real: 1.069 D_fake: 0.733 +(epoch: 192, iters: 7872, time: 0.064) G_GAN: 0.381 G_GAN_Feat: 0.656 G_ID: 0.172 G_Rec: 0.281 D_GP: 0.027 D_real: 1.297 D_fake: 0.621 +(epoch: 192, iters: 8272, time: 0.064) G_GAN: 0.540 G_GAN_Feat: 0.970 G_ID: 0.205 G_Rec: 0.434 D_GP: 0.037 D_real: 1.114 D_fake: 0.504 +(epoch: 193, iters: 64, time: 0.064) G_GAN: 0.264 G_GAN_Feat: 0.708 G_ID: 0.136 G_Rec: 0.291 D_GP: 0.025 D_real: 1.148 D_fake: 0.736 +(epoch: 193, iters: 464, time: 0.064) G_GAN: 0.486 G_GAN_Feat: 0.830 G_ID: 0.208 G_Rec: 0.350 D_GP: 0.026 D_real: 1.205 D_fake: 0.514 +(epoch: 193, iters: 864, time: 0.064) G_GAN: 0.207 G_GAN_Feat: 0.630 G_ID: 0.137 G_Rec: 0.307 D_GP: 0.020 D_real: 1.211 D_fake: 0.794 +(epoch: 193, iters: 1264, time: 0.064) G_GAN: 0.312 G_GAN_Feat: 0.794 G_ID: 0.195 G_Rec: 0.391 D_GP: 0.024 D_real: 1.067 D_fake: 0.689 +(epoch: 193, iters: 1664, time: 0.064) G_GAN: 0.145 G_GAN_Feat: 0.664 G_ID: 0.130 G_Rec: 0.312 D_GP: 0.029 D_real: 1.048 D_fake: 0.855 +(epoch: 193, iters: 2064, time: 0.064) G_GAN: 0.368 G_GAN_Feat: 0.869 G_ID: 0.180 G_Rec: 0.404 D_GP: 0.034 D_real: 0.943 D_fake: 0.634 +(epoch: 193, iters: 2464, time: 0.064) G_GAN: -0.166 G_GAN_Feat: 0.827 G_ID: 0.157 G_Rec: 0.367 D_GP: 0.051 D_real: 0.430 D_fake: 1.168 +(epoch: 193, iters: 2864, time: 0.064) G_GAN: 0.278 G_GAN_Feat: 0.951 G_ID: 0.192 G_Rec: 0.383 D_GP: 0.057 D_real: 0.758 D_fake: 0.723 +(epoch: 193, iters: 3264, time: 0.064) G_GAN: 0.323 G_GAN_Feat: 1.028 G_ID: 0.157 G_Rec: 0.322 D_GP: 0.047 D_real: 0.404 D_fake: 0.711 +(epoch: 193, iters: 3664, time: 0.064) G_GAN: 0.665 G_GAN_Feat: 0.891 G_ID: 0.192 G_Rec: 0.399 D_GP: 0.030 D_real: 1.254 D_fake: 0.347 +(epoch: 193, iters: 4064, time: 0.064) G_GAN: 0.231 G_GAN_Feat: 0.757 G_ID: 0.136 G_Rec: 0.333 D_GP: 0.035 D_real: 0.960 D_fake: 0.769 +(epoch: 193, iters: 4464, time: 0.064) G_GAN: 0.729 G_GAN_Feat: 0.905 G_ID: 0.196 G_Rec: 0.387 D_GP: 0.027 D_real: 1.491 D_fake: 0.299 +(epoch: 193, iters: 4864, time: 0.064) G_GAN: -0.100 G_GAN_Feat: 0.749 G_ID: 0.156 G_Rec: 0.331 D_GP: 0.060 D_real: 0.627 D_fake: 1.100 +(epoch: 193, iters: 5264, time: 0.064) G_GAN: 0.562 G_GAN_Feat: 0.970 G_ID: 0.180 G_Rec: 0.382 D_GP: 0.096 D_real: 0.826 D_fake: 0.461 +(epoch: 193, iters: 5664, time: 0.064) G_GAN: 0.022 G_GAN_Feat: 0.742 G_ID: 0.157 G_Rec: 0.329 D_GP: 0.032 D_real: 0.911 D_fake: 0.978 +(epoch: 193, iters: 6064, time: 0.064) G_GAN: 0.579 G_GAN_Feat: 1.055 G_ID: 0.211 G_Rec: 0.417 D_GP: 0.067 D_real: 0.630 D_fake: 0.430 +(epoch: 193, iters: 6464, time: 0.064) G_GAN: 0.143 G_GAN_Feat: 0.966 G_ID: 0.160 G_Rec: 0.373 D_GP: 0.441 D_real: 0.368 D_fake: 0.857 +(epoch: 193, iters: 6864, time: 0.064) G_GAN: 0.573 G_GAN_Feat: 1.010 G_ID: 0.185 G_Rec: 0.420 D_GP: 0.074 D_real: 0.889 D_fake: 0.452 +(epoch: 193, iters: 7264, time: 0.064) G_GAN: 0.351 G_GAN_Feat: 0.702 G_ID: 0.141 G_Rec: 0.341 D_GP: 0.022 D_real: 1.289 D_fake: 0.649 +(epoch: 193, iters: 7664, time: 0.064) G_GAN: 0.598 G_GAN_Feat: 0.860 G_ID: 0.165 G_Rec: 0.390 D_GP: 0.030 D_real: 1.335 D_fake: 0.420 +(epoch: 193, iters: 8064, time: 0.064) G_GAN: 0.152 G_GAN_Feat: 0.715 G_ID: 0.128 G_Rec: 0.298 D_GP: 0.027 D_real: 0.934 D_fake: 0.848 +(epoch: 193, iters: 8464, time: 0.064) G_GAN: 0.663 G_GAN_Feat: 0.867 G_ID: 0.176 G_Rec: 0.401 D_GP: 0.023 D_real: 1.396 D_fake: 0.344 +(epoch: 194, iters: 256, time: 0.064) G_GAN: 0.200 G_GAN_Feat: 0.686 G_ID: 0.155 G_Rec: 0.296 D_GP: 0.025 D_real: 1.140 D_fake: 0.800 +(epoch: 194, iters: 656, time: 0.064) G_GAN: 0.389 G_GAN_Feat: 0.951 G_ID: 0.170 G_Rec: 0.381 D_GP: 0.088 D_real: 0.755 D_fake: 0.612 +(epoch: 194, iters: 1056, time: 0.064) G_GAN: 0.559 G_GAN_Feat: 0.798 G_ID: 0.154 G_Rec: 0.326 D_GP: 0.090 D_real: 1.117 D_fake: 0.455 +(epoch: 194, iters: 1456, time: 0.064) G_GAN: 0.591 G_GAN_Feat: 0.961 G_ID: 0.176 G_Rec: 0.403 D_GP: 0.040 D_real: 1.165 D_fake: 0.424 +(epoch: 194, iters: 1856, time: 0.064) G_GAN: 0.177 G_GAN_Feat: 0.779 G_ID: 0.152 G_Rec: 0.323 D_GP: 0.071 D_real: 0.828 D_fake: 0.824 +(epoch: 194, iters: 2256, time: 0.064) G_GAN: 0.388 G_GAN_Feat: 0.821 G_ID: 0.188 G_Rec: 0.374 D_GP: 0.023 D_real: 1.087 D_fake: 0.620 +(epoch: 194, iters: 2656, time: 0.064) G_GAN: 0.070 G_GAN_Feat: 0.754 G_ID: 0.135 G_Rec: 0.310 D_GP: 0.032 D_real: 0.790 D_fake: 0.930 +(epoch: 194, iters: 3056, time: 0.064) G_GAN: 0.575 G_GAN_Feat: 0.967 G_ID: 0.170 G_Rec: 0.447 D_GP: 0.043 D_real: 1.133 D_fake: 0.438 +(epoch: 194, iters: 3456, time: 0.064) G_GAN: 0.392 G_GAN_Feat: 0.711 G_ID: 0.137 G_Rec: 0.307 D_GP: 0.029 D_real: 1.206 D_fake: 0.649 +(epoch: 194, iters: 3856, time: 0.064) G_GAN: 0.805 G_GAN_Feat: 0.961 G_ID: 0.178 G_Rec: 0.384 D_GP: 0.034 D_real: 1.444 D_fake: 0.251 +(epoch: 194, iters: 4256, time: 0.064) G_GAN: 0.103 G_GAN_Feat: 0.832 G_ID: 0.155 G_Rec: 0.319 D_GP: 0.040 D_real: 0.771 D_fake: 0.897 +(epoch: 194, iters: 4656, time: 0.064) G_GAN: 0.609 G_GAN_Feat: 0.844 G_ID: 0.168 G_Rec: 0.350 D_GP: 0.032 D_real: 1.291 D_fake: 0.407 +(epoch: 194, iters: 5056, time: 0.064) G_GAN: 0.392 G_GAN_Feat: 0.775 G_ID: 0.134 G_Rec: 0.298 D_GP: 0.038 D_real: 1.144 D_fake: 0.608 +(epoch: 194, iters: 5456, time: 0.064) G_GAN: 0.588 G_GAN_Feat: 1.154 G_ID: 0.172 G_Rec: 0.401 D_GP: 0.068 D_real: 0.476 D_fake: 0.425 +(epoch: 194, iters: 5856, time: 0.064) G_GAN: 0.429 G_GAN_Feat: 0.891 G_ID: 0.137 G_Rec: 0.301 D_GP: 0.045 D_real: 0.978 D_fake: 0.572 +(epoch: 194, iters: 6256, time: 0.064) G_GAN: 0.437 G_GAN_Feat: 0.832 G_ID: 0.198 G_Rec: 0.421 D_GP: 0.025 D_real: 1.128 D_fake: 0.573 +(epoch: 194, iters: 6656, time: 0.064) G_GAN: -0.079 G_GAN_Feat: 0.744 G_ID: 0.162 G_Rec: 0.333 D_GP: 0.050 D_real: 0.686 D_fake: 1.079 +(epoch: 194, iters: 7056, time: 0.064) G_GAN: 0.328 G_GAN_Feat: 0.916 G_ID: 0.191 G_Rec: 0.397 D_GP: 0.046 D_real: 0.839 D_fake: 0.678 +(epoch: 194, iters: 7456, time: 0.064) G_GAN: 0.073 G_GAN_Feat: 0.738 G_ID: 0.135 G_Rec: 0.311 D_GP: 0.053 D_real: 0.875 D_fake: 0.927 +(epoch: 194, iters: 7856, time: 0.064) G_GAN: 0.343 G_GAN_Feat: 1.008 G_ID: 0.189 G_Rec: 0.414 D_GP: 0.037 D_real: 0.682 D_fake: 0.659 +(epoch: 194, iters: 8256, time: 0.064) G_GAN: 0.268 G_GAN_Feat: 0.632 G_ID: 0.142 G_Rec: 0.268 D_GP: 0.026 D_real: 1.204 D_fake: 0.732 +(epoch: 195, iters: 48, time: 0.064) G_GAN: 0.270 G_GAN_Feat: 1.022 G_ID: 0.181 G_Rec: 0.404 D_GP: 0.086 D_real: 0.542 D_fake: 0.732 +(epoch: 195, iters: 448, time: 0.064) G_GAN: 0.479 G_GAN_Feat: 0.790 G_ID: 0.159 G_Rec: 0.312 D_GP: 0.034 D_real: 1.196 D_fake: 0.533 +(epoch: 195, iters: 848, time: 0.064) G_GAN: 0.445 G_GAN_Feat: 1.128 G_ID: 0.199 G_Rec: 0.418 D_GP: 0.084 D_real: 0.372 D_fake: 0.558 +(epoch: 195, iters: 1248, time: 0.064) G_GAN: 0.357 G_GAN_Feat: 0.783 G_ID: 0.165 G_Rec: 0.338 D_GP: 0.047 D_real: 1.064 D_fake: 0.645 +(epoch: 195, iters: 1648, time: 0.064) G_GAN: 0.240 G_GAN_Feat: 0.997 G_ID: 0.200 G_Rec: 0.398 D_GP: 0.074 D_real: 0.450 D_fake: 0.762 +(epoch: 195, iters: 2048, time: 0.064) G_GAN: 0.427 G_GAN_Feat: 0.667 G_ID: 0.143 G_Rec: 0.316 D_GP: 0.021 D_real: 1.393 D_fake: 0.574 +(epoch: 195, iters: 2448, time: 0.064) G_GAN: 0.215 G_GAN_Feat: 0.892 G_ID: 0.180 G_Rec: 0.418 D_GP: 0.024 D_real: 1.006 D_fake: 0.785 +(epoch: 195, iters: 2848, time: 0.064) G_GAN: 0.192 G_GAN_Feat: 0.746 G_ID: 0.137 G_Rec: 0.295 D_GP: 0.027 D_real: 1.059 D_fake: 0.808 +(epoch: 195, iters: 3248, time: 0.064) G_GAN: 0.481 G_GAN_Feat: 0.898 G_ID: 0.185 G_Rec: 0.431 D_GP: 0.022 D_real: 1.198 D_fake: 0.537 +(epoch: 195, iters: 3648, time: 0.064) G_GAN: 0.020 G_GAN_Feat: 0.697 G_ID: 0.126 G_Rec: 0.288 D_GP: 0.025 D_real: 0.871 D_fake: 0.980 +(epoch: 195, iters: 4048, time: 0.064) G_GAN: 0.448 G_GAN_Feat: 0.941 G_ID: 0.180 G_Rec: 0.392 D_GP: 0.074 D_real: 0.814 D_fake: 0.559 +(epoch: 195, iters: 4448, time: 0.064) G_GAN: 0.179 G_GAN_Feat: 0.754 G_ID: 0.153 G_Rec: 0.311 D_GP: 0.028 D_real: 1.036 D_fake: 0.823 +(epoch: 195, iters: 4848, time: 0.064) G_GAN: 0.424 G_GAN_Feat: 0.935 G_ID: 0.178 G_Rec: 0.358 D_GP: 0.045 D_real: 0.848 D_fake: 0.577 +(epoch: 195, iters: 5248, time: 0.064) G_GAN: 0.331 G_GAN_Feat: 0.885 G_ID: 0.159 G_Rec: 0.341 D_GP: 0.028 D_real: 0.994 D_fake: 0.670 +(epoch: 195, iters: 5648, time: 0.064) G_GAN: 0.238 G_GAN_Feat: 1.153 G_ID: 0.166 G_Rec: 0.399 D_GP: 0.481 D_real: 0.379 D_fake: 0.768 +(epoch: 195, iters: 6048, time: 0.064) G_GAN: 0.324 G_GAN_Feat: 0.642 G_ID: 0.137 G_Rec: 0.350 D_GP: 0.020 D_real: 1.278 D_fake: 0.676 +(epoch: 195, iters: 6448, time: 0.064) G_GAN: 0.712 G_GAN_Feat: 1.001 G_ID: 0.192 G_Rec: 0.450 D_GP: 0.054 D_real: 1.007 D_fake: 0.311 +(epoch: 195, iters: 6848, time: 0.064) G_GAN: 0.140 G_GAN_Feat: 0.909 G_ID: 0.140 G_Rec: 0.309 D_GP: 0.037 D_real: 0.472 D_fake: 0.861 +(epoch: 195, iters: 7248, time: 0.064) G_GAN: 0.409 G_GAN_Feat: 0.881 G_ID: 0.168 G_Rec: 0.422 D_GP: 0.033 D_real: 1.044 D_fake: 0.593 +(epoch: 195, iters: 7648, time: 0.064) G_GAN: 0.286 G_GAN_Feat: 0.907 G_ID: 0.147 G_Rec: 0.353 D_GP: 0.198 D_real: 0.858 D_fake: 0.716 +(epoch: 195, iters: 8048, time: 0.064) G_GAN: 0.656 G_GAN_Feat: 0.871 G_ID: 0.198 G_Rec: 0.390 D_GP: 0.030 D_real: 1.399 D_fake: 0.380 +(epoch: 195, iters: 8448, time: 0.064) G_GAN: 0.135 G_GAN_Feat: 0.703 G_ID: 0.140 G_Rec: 0.275 D_GP: 0.027 D_real: 0.976 D_fake: 0.865 +(epoch: 196, iters: 240, time: 0.064) G_GAN: 0.351 G_GAN_Feat: 0.827 G_ID: 0.195 G_Rec: 0.395 D_GP: 0.030 D_real: 1.175 D_fake: 0.651 +(epoch: 196, iters: 640, time: 0.064) G_GAN: -0.215 G_GAN_Feat: 0.917 G_ID: 0.164 G_Rec: 0.363 D_GP: 0.057 D_real: 0.903 D_fake: 1.215 +(epoch: 196, iters: 1040, time: 0.064) G_GAN: 0.411 G_GAN_Feat: 0.923 G_ID: 0.182 G_Rec: 0.418 D_GP: 0.031 D_real: 1.123 D_fake: 0.596 +(epoch: 196, iters: 1440, time: 0.064) G_GAN: -0.116 G_GAN_Feat: 0.741 G_ID: 0.131 G_Rec: 0.333 D_GP: 0.110 D_real: 0.654 D_fake: 1.116 +(epoch: 196, iters: 1840, time: 0.064) G_GAN: 0.231 G_GAN_Feat: 0.935 G_ID: 0.194 G_Rec: 0.400 D_GP: 0.031 D_real: 1.093 D_fake: 0.769 +(epoch: 196, iters: 2240, time: 0.064) G_GAN: 0.420 G_GAN_Feat: 0.818 G_ID: 0.118 G_Rec: 0.337 D_GP: 0.054 D_real: 1.179 D_fake: 0.697 +(epoch: 196, iters: 2640, time: 0.064) G_GAN: 0.315 G_GAN_Feat: 0.846 G_ID: 0.180 G_Rec: 0.429 D_GP: 0.026 D_real: 0.956 D_fake: 0.694 +(epoch: 196, iters: 3040, time: 0.064) G_GAN: 0.158 G_GAN_Feat: 0.633 G_ID: 0.137 G_Rec: 0.317 D_GP: 0.024 D_real: 1.079 D_fake: 0.842 +(epoch: 196, iters: 3440, time: 0.064) G_GAN: 0.285 G_GAN_Feat: 1.062 G_ID: 0.192 G_Rec: 0.460 D_GP: 0.167 D_real: 0.597 D_fake: 0.732 +(epoch: 196, iters: 3840, time: 0.064) G_GAN: -0.087 G_GAN_Feat: 0.767 G_ID: 0.139 G_Rec: 0.333 D_GP: 0.044 D_real: 0.730 D_fake: 1.087 +(epoch: 196, iters: 4240, time: 0.064) G_GAN: 0.503 G_GAN_Feat: 0.780 G_ID: 0.154 G_Rec: 0.384 D_GP: 0.020 D_real: 1.366 D_fake: 0.514 +(epoch: 196, iters: 4640, time: 0.064) G_GAN: 0.159 G_GAN_Feat: 0.744 G_ID: 0.159 G_Rec: 0.316 D_GP: 0.027 D_real: 0.972 D_fake: 0.841 +(epoch: 196, iters: 5040, time: 0.064) G_GAN: 0.533 G_GAN_Feat: 0.893 G_ID: 0.179 G_Rec: 0.419 D_GP: 0.026 D_real: 1.285 D_fake: 0.517 +(epoch: 196, iters: 5440, time: 0.064) G_GAN: 0.249 G_GAN_Feat: 0.850 G_ID: 0.139 G_Rec: 0.341 D_GP: 0.057 D_real: 0.727 D_fake: 0.752 +(epoch: 196, iters: 5840, time: 0.064) G_GAN: 0.439 G_GAN_Feat: 1.166 G_ID: 0.192 G_Rec: 0.420 D_GP: 0.074 D_real: 0.659 D_fake: 0.616 +(epoch: 196, iters: 6240, time: 0.064) G_GAN: 0.213 G_GAN_Feat: 0.593 G_ID: 0.158 G_Rec: 0.289 D_GP: 0.018 D_real: 1.159 D_fake: 0.787 +(epoch: 196, iters: 6640, time: 0.064) G_GAN: 0.274 G_GAN_Feat: 0.857 G_ID: 0.173 G_Rec: 0.431 D_GP: 0.023 D_real: 0.989 D_fake: 0.726 +(epoch: 196, iters: 7040, time: 0.064) G_GAN: 0.127 G_GAN_Feat: 1.115 G_ID: 0.130 G_Rec: 0.378 D_GP: 0.054 D_real: 1.326 D_fake: 0.875 +(epoch: 196, iters: 7440, time: 0.064) G_GAN: 0.125 G_GAN_Feat: 0.805 G_ID: 0.184 G_Rec: 0.404 D_GP: 0.022 D_real: 0.936 D_fake: 0.879 +(epoch: 196, iters: 7840, time: 0.064) G_GAN: -0.015 G_GAN_Feat: 0.700 G_ID: 0.139 G_Rec: 0.304 D_GP: 0.037 D_real: 0.829 D_fake: 1.016 +(epoch: 196, iters: 8240, time: 0.064) G_GAN: 0.313 G_GAN_Feat: 0.808 G_ID: 0.190 G_Rec: 0.358 D_GP: 0.035 D_real: 1.111 D_fake: 0.688 +(epoch: 197, iters: 32, time: 0.064) G_GAN: 0.283 G_GAN_Feat: 0.973 G_ID: 0.152 G_Rec: 0.323 D_GP: 0.126 D_real: 1.055 D_fake: 0.755 +(epoch: 197, iters: 432, time: 0.064) G_GAN: 0.473 G_GAN_Feat: 0.844 G_ID: 0.221 G_Rec: 0.374 D_GP: 0.023 D_real: 1.237 D_fake: 0.534 +(epoch: 197, iters: 832, time: 0.064) G_GAN: 0.196 G_GAN_Feat: 0.814 G_ID: 0.151 G_Rec: 0.300 D_GP: 0.071 D_real: 0.671 D_fake: 0.805 +(epoch: 197, iters: 1232, time: 0.064) G_GAN: 0.332 G_GAN_Feat: 0.824 G_ID: 0.220 G_Rec: 0.410 D_GP: 0.024 D_real: 1.064 D_fake: 0.673 +(epoch: 197, iters: 1632, time: 0.064) G_GAN: 0.062 G_GAN_Feat: 0.668 G_ID: 0.116 G_Rec: 0.333 D_GP: 0.020 D_real: 0.973 D_fake: 0.938 +(epoch: 197, iters: 2032, time: 0.064) G_GAN: 0.236 G_GAN_Feat: 0.863 G_ID: 0.190 G_Rec: 0.388 D_GP: 0.046 D_real: 0.882 D_fake: 0.764 +(epoch: 197, iters: 2432, time: 0.064) G_GAN: 0.023 G_GAN_Feat: 0.827 G_ID: 0.146 G_Rec: 0.332 D_GP: 0.145 D_real: 0.507 D_fake: 0.977 +(epoch: 197, iters: 2832, time: 0.064) G_GAN: 0.299 G_GAN_Feat: 0.968 G_ID: 0.185 G_Rec: 0.427 D_GP: 0.033 D_real: 0.836 D_fake: 0.701 +(epoch: 197, iters: 3232, time: 0.064) G_GAN: 0.437 G_GAN_Feat: 0.655 G_ID: 0.129 G_Rec: 0.354 D_GP: 0.028 D_real: 1.331 D_fake: 0.563 +(epoch: 197, iters: 3632, time: 0.064) G_GAN: 0.622 G_GAN_Feat: 1.025 G_ID: 0.206 G_Rec: 0.414 D_GP: 0.058 D_real: 0.803 D_fake: 0.402 +(epoch: 197, iters: 4032, time: 0.064) G_GAN: 0.423 G_GAN_Feat: 0.744 G_ID: 0.146 G_Rec: 0.308 D_GP: 0.025 D_real: 1.326 D_fake: 0.582 +(epoch: 197, iters: 4432, time: 0.064) G_GAN: 0.317 G_GAN_Feat: 0.924 G_ID: 0.184 G_Rec: 0.413 D_GP: 0.027 D_real: 0.996 D_fake: 0.684 +(epoch: 197, iters: 4832, time: 0.064) G_GAN: 0.402 G_GAN_Feat: 0.712 G_ID: 0.129 G_Rec: 0.312 D_GP: 0.027 D_real: 1.289 D_fake: 0.617 +(epoch: 197, iters: 5232, time: 0.064) G_GAN: 0.691 G_GAN_Feat: 0.862 G_ID: 0.203 G_Rec: 0.380 D_GP: 0.027 D_real: 1.338 D_fake: 0.347 +(epoch: 197, iters: 5632, time: 0.064) G_GAN: 0.156 G_GAN_Feat: 0.677 G_ID: 0.168 G_Rec: 0.350 D_GP: 0.030 D_real: 1.035 D_fake: 0.844 +(epoch: 197, iters: 6032, time: 0.064) G_GAN: 0.700 G_GAN_Feat: 0.786 G_ID: 0.179 G_Rec: 0.381 D_GP: 0.025 D_real: 1.423 D_fake: 0.349 +(epoch: 197, iters: 6432, time: 0.064) G_GAN: 0.343 G_GAN_Feat: 0.674 G_ID: 0.129 G_Rec: 0.314 D_GP: 0.020 D_real: 1.254 D_fake: 0.658 +(epoch: 197, iters: 6832, time: 0.064) G_GAN: 0.134 G_GAN_Feat: 0.787 G_ID: 0.211 G_Rec: 0.420 D_GP: 0.033 D_real: 0.888 D_fake: 0.866 +(epoch: 197, iters: 7232, time: 0.064) G_GAN: 0.259 G_GAN_Feat: 0.682 G_ID: 0.136 G_Rec: 0.305 D_GP: 0.043 D_real: 1.130 D_fake: 0.743 +(epoch: 197, iters: 7632, time: 0.064) G_GAN: 0.412 G_GAN_Feat: 0.959 G_ID: 0.187 G_Rec: 0.402 D_GP: 0.159 D_real: 0.802 D_fake: 0.598 +(epoch: 197, iters: 8032, time: 0.064) G_GAN: 0.122 G_GAN_Feat: 0.810 G_ID: 0.152 G_Rec: 0.321 D_GP: 0.043 D_real: 0.793 D_fake: 0.879 +(epoch: 197, iters: 8432, time: 0.064) G_GAN: 0.395 G_GAN_Feat: 0.962 G_ID: 0.238 G_Rec: 0.389 D_GP: 0.043 D_real: 0.778 D_fake: 0.605 +(epoch: 198, iters: 224, time: 0.064) G_GAN: 0.079 G_GAN_Feat: 0.866 G_ID: 0.128 G_Rec: 0.343 D_GP: 0.052 D_real: 0.672 D_fake: 0.921 +(epoch: 198, iters: 624, time: 0.064) G_GAN: 0.036 G_GAN_Feat: 0.873 G_ID: 0.184 G_Rec: 0.393 D_GP: 0.035 D_real: 0.705 D_fake: 0.964 +(epoch: 198, iters: 1024, time: 0.064) G_GAN: 0.116 G_GAN_Feat: 0.953 G_ID: 0.124 G_Rec: 0.365 D_GP: 0.980 D_real: 0.376 D_fake: 0.885 +(epoch: 198, iters: 1424, time: 0.064) G_GAN: 0.545 G_GAN_Feat: 0.777 G_ID: 0.170 G_Rec: 0.363 D_GP: 0.021 D_real: 1.330 D_fake: 0.460 +(epoch: 198, iters: 1824, time: 0.064) G_GAN: 0.401 G_GAN_Feat: 0.633 G_ID: 0.121 G_Rec: 0.278 D_GP: 0.028 D_real: 1.307 D_fake: 0.601 +(epoch: 198, iters: 2224, time: 0.064) G_GAN: 0.444 G_GAN_Feat: 0.824 G_ID: 0.182 G_Rec: 0.369 D_GP: 0.035 D_real: 1.177 D_fake: 0.557 +(epoch: 198, iters: 2624, time: 0.064) G_GAN: 0.416 G_GAN_Feat: 0.765 G_ID: 0.144 G_Rec: 0.304 D_GP: 0.098 D_real: 1.102 D_fake: 0.586 +(epoch: 198, iters: 3024, time: 0.064) G_GAN: 0.377 G_GAN_Feat: 1.047 G_ID: 0.176 G_Rec: 0.426 D_GP: 0.162 D_real: 0.896 D_fake: 0.634 +(epoch: 198, iters: 3424, time: 0.064) G_GAN: -0.074 G_GAN_Feat: 0.803 G_ID: 0.121 G_Rec: 0.377 D_GP: 0.024 D_real: 0.800 D_fake: 1.074 +(epoch: 198, iters: 3824, time: 0.064) G_GAN: 0.436 G_GAN_Feat: 1.107 G_ID: 0.176 G_Rec: 0.391 D_GP: 0.246 D_real: 0.369 D_fake: 0.576 +(epoch: 198, iters: 4224, time: 0.064) G_GAN: 0.133 G_GAN_Feat: 0.746 G_ID: 0.132 G_Rec: 0.314 D_GP: 0.029 D_real: 1.013 D_fake: 0.867 +(epoch: 198, iters: 4624, time: 0.064) G_GAN: 0.563 G_GAN_Feat: 0.974 G_ID: 0.207 G_Rec: 0.408 D_GP: 0.031 D_real: 1.174 D_fake: 0.443 +(epoch: 198, iters: 5024, time: 0.064) G_GAN: 0.239 G_GAN_Feat: 0.671 G_ID: 0.130 G_Rec: 0.298 D_GP: 0.021 D_real: 1.173 D_fake: 0.761 +(epoch: 198, iters: 5424, time: 0.064) G_GAN: 0.253 G_GAN_Feat: 0.895 G_ID: 0.199 G_Rec: 0.437 D_GP: 0.080 D_real: 0.829 D_fake: 0.750 +(epoch: 198, iters: 5824, time: 0.064) G_GAN: 0.060 G_GAN_Feat: 0.742 G_ID: 0.115 G_Rec: 0.305 D_GP: 0.052 D_real: 0.847 D_fake: 0.940 +(epoch: 198, iters: 6224, time: 0.064) G_GAN: 0.514 G_GAN_Feat: 0.807 G_ID: 0.161 G_Rec: 0.375 D_GP: 0.023 D_real: 1.242 D_fake: 0.489 +(epoch: 198, iters: 6624, time: 0.064) G_GAN: 0.218 G_GAN_Feat: 0.683 G_ID: 0.137 G_Rec: 0.305 D_GP: 0.023 D_real: 1.110 D_fake: 0.782 +(epoch: 198, iters: 7024, time: 0.064) G_GAN: 0.268 G_GAN_Feat: 0.869 G_ID: 0.191 G_Rec: 0.388 D_GP: 0.024 D_real: 1.021 D_fake: 0.733 +(epoch: 198, iters: 7424, time: 0.064) G_GAN: 0.319 G_GAN_Feat: 0.847 G_ID: 0.140 G_Rec: 0.297 D_GP: 0.050 D_real: 0.775 D_fake: 0.684 +(epoch: 198, iters: 7824, time: 0.064) G_GAN: 0.265 G_GAN_Feat: 0.768 G_ID: 0.190 G_Rec: 0.401 D_GP: 0.022 D_real: 1.049 D_fake: 0.735 +(epoch: 198, iters: 8224, time: 0.064) G_GAN: 0.245 G_GAN_Feat: 0.586 G_ID: 0.134 G_Rec: 0.313 D_GP: 0.021 D_real: 1.250 D_fake: 0.755 +(epoch: 199, iters: 16, time: 0.064) G_GAN: 0.254 G_GAN_Feat: 0.712 G_ID: 0.163 G_Rec: 0.360 D_GP: 0.022 D_real: 1.104 D_fake: 0.749 +(epoch: 199, iters: 416, time: 0.064) G_GAN: 0.058 G_GAN_Feat: 0.619 G_ID: 0.198 G_Rec: 0.307 D_GP: 0.025 D_real: 1.056 D_fake: 0.942 +(epoch: 199, iters: 816, time: 0.064) G_GAN: 0.608 G_GAN_Feat: 0.807 G_ID: 0.166 G_Rec: 0.416 D_GP: 0.021 D_real: 1.313 D_fake: 0.423 +(epoch: 199, iters: 1216, time: 0.064) G_GAN: 0.223 G_GAN_Feat: 0.747 G_ID: 0.117 G_Rec: 0.288 D_GP: 0.030 D_real: 1.139 D_fake: 0.779 +(epoch: 199, iters: 1616, time: 0.064) G_GAN: 0.253 G_GAN_Feat: 0.884 G_ID: 0.188 G_Rec: 0.433 D_GP: 0.029 D_real: 0.988 D_fake: 0.747 +(epoch: 199, iters: 2016, time: 0.064) G_GAN: 0.146 G_GAN_Feat: 0.788 G_ID: 0.162 G_Rec: 0.307 D_GP: 0.049 D_real: 0.703 D_fake: 0.854 +(epoch: 199, iters: 2416, time: 0.064) G_GAN: 0.462 G_GAN_Feat: 0.714 G_ID: 0.195 G_Rec: 0.364 D_GP: 0.021 D_real: 1.217 D_fake: 0.539 +(epoch: 199, iters: 2816, time: 0.064) G_GAN: 0.053 G_GAN_Feat: 0.628 G_ID: 0.136 G_Rec: 0.297 D_GP: 0.030 D_real: 1.049 D_fake: 0.947 +(epoch: 199, iters: 3216, time: 0.064) G_GAN: 0.282 G_GAN_Feat: 0.793 G_ID: 0.188 G_Rec: 0.400 D_GP: 0.027 D_real: 1.076 D_fake: 0.718 +(epoch: 199, iters: 3616, time: 0.064) G_GAN: 0.433 G_GAN_Feat: 0.767 G_ID: 0.143 G_Rec: 0.349 D_GP: 0.040 D_real: 1.246 D_fake: 0.568 +(epoch: 199, iters: 4016, time: 0.064) G_GAN: 0.168 G_GAN_Feat: 0.855 G_ID: 0.182 G_Rec: 0.385 D_GP: 0.029 D_real: 0.842 D_fake: 0.833 +(epoch: 199, iters: 4416, time: 0.064) G_GAN: 0.280 G_GAN_Feat: 0.717 G_ID: 0.136 G_Rec: 0.311 D_GP: 0.044 D_real: 1.147 D_fake: 0.720 +(epoch: 199, iters: 4816, time: 0.064) G_GAN: 0.476 G_GAN_Feat: 1.011 G_ID: 0.185 G_Rec: 0.422 D_GP: 0.052 D_real: 0.899 D_fake: 0.527 +(epoch: 199, iters: 5216, time: 0.064) G_GAN: 0.249 G_GAN_Feat: 0.730 G_ID: 0.128 G_Rec: 0.307 D_GP: 0.028 D_real: 1.075 D_fake: 0.751 +(epoch: 199, iters: 5616, time: 0.064) G_GAN: 0.398 G_GAN_Feat: 0.830 G_ID: 0.180 G_Rec: 0.400 D_GP: 0.020 D_real: 1.123 D_fake: 0.603 +(epoch: 199, iters: 6016, time: 0.064) G_GAN: 0.312 G_GAN_Feat: 0.602 G_ID: 0.133 G_Rec: 0.278 D_GP: 0.022 D_real: 1.254 D_fake: 0.688 +(epoch: 199, iters: 6416, time: 0.064) G_GAN: 0.330 G_GAN_Feat: 0.809 G_ID: 0.193 G_Rec: 0.396 D_GP: 0.036 D_real: 1.038 D_fake: 0.670 +(epoch: 199, iters: 6816, time: 0.064) G_GAN: 0.133 G_GAN_Feat: 0.749 G_ID: 0.138 G_Rec: 0.305 D_GP: 0.054 D_real: 1.007 D_fake: 0.867 +(epoch: 199, iters: 7216, time: 0.064) G_GAN: 0.553 G_GAN_Feat: 0.837 G_ID: 0.205 G_Rec: 0.388 D_GP: 0.020 D_real: 1.287 D_fake: 0.456 +(epoch: 199, iters: 7616, time: 0.064) G_GAN: 0.118 G_GAN_Feat: 0.701 G_ID: 0.128 G_Rec: 0.308 D_GP: 0.038 D_real: 1.019 D_fake: 0.882 +(epoch: 199, iters: 8016, time: 0.064) G_GAN: 0.615 G_GAN_Feat: 1.115 G_ID: 0.191 G_Rec: 0.451 D_GP: 0.090 D_real: 0.596 D_fake: 0.407 +(epoch: 199, iters: 8416, time: 0.064) G_GAN: 0.293 G_GAN_Feat: 0.685 G_ID: 0.140 G_Rec: 0.313 D_GP: 0.022 D_real: 1.228 D_fake: 0.709 +(epoch: 200, iters: 208, time: 0.064) G_GAN: 0.584 G_GAN_Feat: 1.043 G_ID: 0.197 G_Rec: 0.411 D_GP: 0.054 D_real: 0.665 D_fake: 0.438 +(epoch: 200, iters: 608, time: 0.064) G_GAN: 0.236 G_GAN_Feat: 0.598 G_ID: 0.137 G_Rec: 0.277 D_GP: 0.021 D_real: 1.199 D_fake: 0.764 +(epoch: 200, iters: 1008, time: 0.064) G_GAN: 0.278 G_GAN_Feat: 0.796 G_ID: 0.209 G_Rec: 0.388 D_GP: 0.027 D_real: 1.000 D_fake: 0.723 +(epoch: 200, iters: 1408, time: 0.064) G_GAN: 0.014 G_GAN_Feat: 0.772 G_ID: 0.142 G_Rec: 0.338 D_GP: 0.053 D_real: 0.666 D_fake: 0.987 +(epoch: 200, iters: 1808, time: 0.064) G_GAN: 0.429 G_GAN_Feat: 0.997 G_ID: 0.189 G_Rec: 0.450 D_GP: 0.083 D_real: 0.664 D_fake: 0.575 +(epoch: 200, iters: 2208, time: 0.064) G_GAN: 0.303 G_GAN_Feat: 0.856 G_ID: 0.145 G_Rec: 0.312 D_GP: 0.061 D_real: 0.683 D_fake: 0.697 +(epoch: 200, iters: 2608, time: 0.064) G_GAN: 0.444 G_GAN_Feat: 0.866 G_ID: 0.183 G_Rec: 0.398 D_GP: 0.026 D_real: 1.174 D_fake: 0.567 +(epoch: 200, iters: 3008, time: 0.064) G_GAN: 0.082 G_GAN_Feat: 0.693 G_ID: 0.126 G_Rec: 0.308 D_GP: 0.024 D_real: 0.956 D_fake: 0.918 +(epoch: 200, iters: 3408, time: 0.064) G_GAN: 0.273 G_GAN_Feat: 0.992 G_ID: 0.183 G_Rec: 0.435 D_GP: 0.072 D_real: 0.699 D_fake: 0.727 +(epoch: 200, iters: 3808, time: 0.064) G_GAN: 0.418 G_GAN_Feat: 0.734 G_ID: 0.148 G_Rec: 0.291 D_GP: 0.025 D_real: 1.273 D_fake: 0.595 +(epoch: 200, iters: 4208, time: 0.064) G_GAN: 0.550 G_GAN_Feat: 0.900 G_ID: 0.188 G_Rec: 0.421 D_GP: 0.030 D_real: 1.239 D_fake: 0.465 +(epoch: 200, iters: 4608, time: 0.064) G_GAN: 0.280 G_GAN_Feat: 1.008 G_ID: 0.149 G_Rec: 0.322 D_GP: 0.214 D_real: 0.368 D_fake: 0.732 +(epoch: 200, iters: 5008, time: 0.064) G_GAN: 0.559 G_GAN_Feat: 0.969 G_ID: 0.193 G_Rec: 0.485 D_GP: 0.025 D_real: 1.216 D_fake: 0.453 +(epoch: 200, iters: 5408, time: 0.064) G_GAN: 0.115 G_GAN_Feat: 0.648 G_ID: 0.128 G_Rec: 0.310 D_GP: 0.026 D_real: 1.082 D_fake: 0.888 +(epoch: 200, iters: 5808, time: 0.064) G_GAN: 0.445 G_GAN_Feat: 0.784 G_ID: 0.189 G_Rec: 0.376 D_GP: 0.029 D_real: 1.253 D_fake: 0.567 +(epoch: 200, iters: 6208, time: 0.064) G_GAN: 0.217 G_GAN_Feat: 0.718 G_ID: 0.125 G_Rec: 0.323 D_GP: 0.038 D_real: 0.946 D_fake: 0.783 +(epoch: 200, iters: 6608, time: 0.064) G_GAN: 0.410 G_GAN_Feat: 1.081 G_ID: 0.190 G_Rec: 0.442 D_GP: 0.061 D_real: 0.627 D_fake: 0.594 +(epoch: 200, iters: 7008, time: 0.064) G_GAN: 0.053 G_GAN_Feat: 0.777 G_ID: 0.152 G_Rec: 0.339 D_GP: 0.031 D_real: 0.959 D_fake: 0.947 +(epoch: 200, iters: 7408, time: 0.064) G_GAN: 0.512 G_GAN_Feat: 0.910 G_ID: 0.166 G_Rec: 0.411 D_GP: 0.037 D_real: 1.197 D_fake: 0.493 +(epoch: 200, iters: 7808, time: 0.064) G_GAN: 0.259 G_GAN_Feat: 0.676 G_ID: 0.144 G_Rec: 0.283 D_GP: 0.025 D_real: 1.123 D_fake: 0.741 +(epoch: 200, iters: 8208, time: 0.064) G_GAN: 0.611 G_GAN_Feat: 0.860 G_ID: 0.184 G_Rec: 0.391 D_GP: 0.023 D_real: 1.351 D_fake: 0.394 +(epoch: 200, iters: 8608, time: 0.064) G_GAN: 0.303 G_GAN_Feat: 0.798 G_ID: 0.133 G_Rec: 0.322 D_GP: 0.026 D_real: 1.020 D_fake: 0.697 +(epoch: 201, iters: 400, time: 0.064) G_GAN: 0.426 G_GAN_Feat: 0.898 G_ID: 0.174 G_Rec: 0.368 D_GP: 0.033 D_real: 0.942 D_fake: 0.575 +(epoch: 201, iters: 800, time: 0.064) G_GAN: 0.275 G_GAN_Feat: 0.745 G_ID: 0.141 G_Rec: 0.308 D_GP: 0.029 D_real: 1.020 D_fake: 0.725 +(epoch: 201, iters: 1200, time: 0.064) G_GAN: 0.517 G_GAN_Feat: 1.131 G_ID: 0.170 G_Rec: 0.414 D_GP: 0.209 D_real: 0.425 D_fake: 0.489 +(epoch: 201, iters: 1600, time: 0.064) G_GAN: 0.287 G_GAN_Feat: 0.775 G_ID: 0.132 G_Rec: 0.301 D_GP: 0.027 D_real: 1.152 D_fake: 0.713 +(epoch: 201, iters: 2000, time: 0.064) G_GAN: 0.602 G_GAN_Feat: 0.732 G_ID: 0.166 G_Rec: 0.387 D_GP: 0.020 D_real: 1.479 D_fake: 0.423 +(epoch: 201, iters: 2400, time: 0.064) G_GAN: 0.119 G_GAN_Feat: 0.649 G_ID: 0.147 G_Rec: 0.306 D_GP: 0.020 D_real: 1.026 D_fake: 0.882 +(epoch: 201, iters: 2800, time: 0.064) G_GAN: 0.464 G_GAN_Feat: 0.796 G_ID: 0.194 G_Rec: 0.391 D_GP: 0.027 D_real: 1.161 D_fake: 0.556 +(epoch: 201, iters: 3200, time: 0.064) G_GAN: 0.089 G_GAN_Feat: 0.625 G_ID: 0.138 G_Rec: 0.281 D_GP: 0.025 D_real: 1.016 D_fake: 0.914 +(epoch: 201, iters: 3600, time: 0.064) G_GAN: 0.247 G_GAN_Feat: 0.970 G_ID: 0.165 G_Rec: 0.402 D_GP: 0.085 D_real: 0.571 D_fake: 0.756 +(epoch: 201, iters: 4000, time: 0.064) G_GAN: 0.144 G_GAN_Feat: 0.649 G_ID: 0.142 G_Rec: 0.268 D_GP: 0.028 D_real: 1.034 D_fake: 0.856 +(epoch: 201, iters: 4400, time: 0.064) G_GAN: 0.474 G_GAN_Feat: 0.977 G_ID: 0.199 G_Rec: 0.449 D_GP: 0.033 D_real: 0.995 D_fake: 0.535 +(epoch: 201, iters: 4800, time: 0.064) G_GAN: 0.079 G_GAN_Feat: 0.690 G_ID: 0.136 G_Rec: 0.296 D_GP: 0.026 D_real: 1.061 D_fake: 0.921 +(epoch: 201, iters: 5200, time: 0.064) G_GAN: 0.518 G_GAN_Feat: 0.950 G_ID: 0.179 G_Rec: 0.406 D_GP: 0.027 D_real: 1.081 D_fake: 0.486 +(epoch: 201, iters: 5600, time: 0.064) G_GAN: 0.075 G_GAN_Feat: 0.920 G_ID: 0.141 G_Rec: 0.321 D_GP: 0.104 D_real: 0.559 D_fake: 0.925 +(epoch: 201, iters: 6000, time: 0.064) G_GAN: 0.410 G_GAN_Feat: 0.813 G_ID: 0.166 G_Rec: 0.369 D_GP: 0.020 D_real: 1.220 D_fake: 0.591 +(epoch: 201, iters: 6400, time: 0.064) G_GAN: -0.013 G_GAN_Feat: 0.845 G_ID: 0.136 G_Rec: 0.302 D_GP: 0.116 D_real: 0.501 D_fake: 1.013 +(epoch: 201, iters: 6800, time: 0.064) G_GAN: 0.578 G_GAN_Feat: 0.863 G_ID: 0.172 G_Rec: 0.405 D_GP: 0.023 D_real: 1.341 D_fake: 0.430 +(epoch: 201, iters: 7200, time: 0.064) G_GAN: 0.369 G_GAN_Feat: 0.706 G_ID: 0.133 G_Rec: 0.293 D_GP: 0.025 D_real: 1.230 D_fake: 0.632 +(epoch: 201, iters: 7600, time: 0.064) G_GAN: 0.457 G_GAN_Feat: 0.857 G_ID: 0.193 G_Rec: 0.397 D_GP: 0.028 D_real: 1.075 D_fake: 0.550 +(epoch: 201, iters: 8000, time: 0.064) G_GAN: 0.635 G_GAN_Feat: 0.964 G_ID: 0.126 G_Rec: 0.325 D_GP: 0.037 D_real: 1.354 D_fake: 0.528 +(epoch: 201, iters: 8400, time: 0.064) G_GAN: 0.484 G_GAN_Feat: 0.789 G_ID: 0.195 G_Rec: 0.401 D_GP: 0.026 D_real: 1.257 D_fake: 0.525 +(epoch: 202, iters: 192, time: 0.064) G_GAN: 0.097 G_GAN_Feat: 0.551 G_ID: 0.139 G_Rec: 0.345 D_GP: 0.017 D_real: 1.048 D_fake: 0.903 +(epoch: 202, iters: 592, time: 0.064) G_GAN: 0.329 G_GAN_Feat: 0.679 G_ID: 0.176 G_Rec: 0.337 D_GP: 0.023 D_real: 1.172 D_fake: 0.672 +(epoch: 202, iters: 992, time: 0.064) G_GAN: 0.155 G_GAN_Feat: 0.602 G_ID: 0.154 G_Rec: 0.307 D_GP: 0.024 D_real: 1.023 D_fake: 0.845 +(epoch: 202, iters: 1392, time: 0.064) G_GAN: 0.160 G_GAN_Feat: 0.773 G_ID: 0.181 G_Rec: 0.382 D_GP: 0.026 D_real: 0.926 D_fake: 0.840 +(epoch: 202, iters: 1792, time: 0.064) G_GAN: 0.142 G_GAN_Feat: 0.617 G_ID: 0.126 G_Rec: 0.289 D_GP: 0.024 D_real: 1.005 D_fake: 0.859 +(epoch: 202, iters: 2192, time: 0.064) G_GAN: 0.348 G_GAN_Feat: 0.927 G_ID: 0.178 G_Rec: 0.429 D_GP: 0.116 D_real: 0.768 D_fake: 0.654 +(epoch: 202, iters: 2592, time: 0.064) G_GAN: 0.296 G_GAN_Feat: 0.637 G_ID: 0.133 G_Rec: 0.337 D_GP: 0.020 D_real: 1.242 D_fake: 0.704 +(epoch: 202, iters: 2992, time: 0.064) G_GAN: 0.525 G_GAN_Feat: 0.915 G_ID: 0.183 G_Rec: 0.474 D_GP: 0.026 D_real: 1.348 D_fake: 0.480 +(epoch: 202, iters: 3392, time: 0.064) G_GAN: 0.283 G_GAN_Feat: 0.593 G_ID: 0.137 G_Rec: 0.304 D_GP: 0.021 D_real: 1.242 D_fake: 0.717 +(epoch: 202, iters: 3792, time: 0.064) G_GAN: 0.451 G_GAN_Feat: 0.842 G_ID: 0.189 G_Rec: 0.380 D_GP: 0.044 D_real: 1.040 D_fake: 0.550 +(epoch: 202, iters: 4192, time: 0.064) G_GAN: -0.026 G_GAN_Feat: 0.750 G_ID: 0.150 G_Rec: 0.312 D_GP: 0.069 D_real: 0.667 D_fake: 1.026 +(epoch: 202, iters: 4592, time: 0.064) G_GAN: 0.442 G_GAN_Feat: 1.014 G_ID: 0.183 G_Rec: 0.443 D_GP: 0.517 D_real: 0.659 D_fake: 0.592 +(epoch: 202, iters: 4992, time: 0.064) G_GAN: 0.179 G_GAN_Feat: 0.745 G_ID: 0.139 G_Rec: 0.335 D_GP: 0.036 D_real: 0.963 D_fake: 0.822 +(epoch: 202, iters: 5392, time: 0.064) G_GAN: 0.328 G_GAN_Feat: 0.866 G_ID: 0.177 G_Rec: 0.399 D_GP: 0.023 D_real: 1.024 D_fake: 0.672 +(epoch: 202, iters: 5792, time: 0.064) G_GAN: 0.339 G_GAN_Feat: 0.885 G_ID: 0.118 G_Rec: 0.324 D_GP: 0.453 D_real: 0.587 D_fake: 0.684 +(epoch: 202, iters: 6192, time: 0.064) G_GAN: 0.407 G_GAN_Feat: 0.859 G_ID: 0.192 G_Rec: 0.389 D_GP: 0.034 D_real: 1.080 D_fake: 0.596 +(epoch: 202, iters: 6592, time: 0.064) G_GAN: 0.420 G_GAN_Feat: 0.703 G_ID: 0.127 G_Rec: 0.286 D_GP: 0.040 D_real: 1.199 D_fake: 0.581 +(epoch: 202, iters: 6992, time: 0.064) G_GAN: -0.206 G_GAN_Feat: 1.055 G_ID: 0.224 G_Rec: 0.423 D_GP: 0.113 D_real: 0.079 D_fake: 1.206 +(epoch: 202, iters: 7392, time: 0.064) G_GAN: 0.067 G_GAN_Feat: 0.848 G_ID: 0.170 G_Rec: 0.308 D_GP: 0.030 D_real: 0.723 D_fake: 0.933 +(epoch: 202, iters: 7792, time: 0.064) G_GAN: 0.372 G_GAN_Feat: 0.840 G_ID: 0.172 G_Rec: 0.368 D_GP: 0.024 D_real: 1.134 D_fake: 0.628 +(epoch: 202, iters: 8192, time: 0.064) G_GAN: 0.241 G_GAN_Feat: 0.763 G_ID: 0.137 G_Rec: 0.316 D_GP: 0.045 D_real: 0.908 D_fake: 0.760 +(epoch: 202, iters: 8592, time: 0.064) G_GAN: 0.283 G_GAN_Feat: 1.020 G_ID: 0.174 G_Rec: 0.399 D_GP: 0.052 D_real: 0.968 D_fake: 0.727 +(epoch: 203, iters: 384, time: 0.064) G_GAN: 0.195 G_GAN_Feat: 0.914 G_ID: 0.131 G_Rec: 0.365 D_GP: 0.033 D_real: 0.498 D_fake: 0.810 +(epoch: 203, iters: 784, time: 0.064) G_GAN: 0.516 G_GAN_Feat: 0.923 G_ID: 0.153 G_Rec: 0.367 D_GP: 0.072 D_real: 0.959 D_fake: 0.498 +(epoch: 203, iters: 1184, time: 0.064) G_GAN: 0.268 G_GAN_Feat: 0.759 G_ID: 0.138 G_Rec: 0.328 D_GP: 0.031 D_real: 1.097 D_fake: 0.733 +(epoch: 203, iters: 1584, time: 0.064) G_GAN: 0.693 G_GAN_Feat: 0.821 G_ID: 0.169 G_Rec: 0.370 D_GP: 0.022 D_real: 1.442 D_fake: 0.322 +(epoch: 203, iters: 1984, time: 0.064) G_GAN: 0.344 G_GAN_Feat: 0.647 G_ID: 0.138 G_Rec: 0.280 D_GP: 0.024 D_real: 1.229 D_fake: 0.660 +(epoch: 203, iters: 2384, time: 0.064) G_GAN: 0.437 G_GAN_Feat: 0.861 G_ID: 0.183 G_Rec: 0.387 D_GP: 0.022 D_real: 1.161 D_fake: 0.564 +(epoch: 203, iters: 2784, time: 0.064) G_GAN: 0.353 G_GAN_Feat: 0.850 G_ID: 0.119 G_Rec: 0.327 D_GP: 0.077 D_real: 0.649 D_fake: 0.659 +(epoch: 203, iters: 3184, time: 0.064) G_GAN: 0.477 G_GAN_Feat: 1.064 G_ID: 0.216 G_Rec: 0.419 D_GP: 0.048 D_real: 0.579 D_fake: 0.533 +(epoch: 203, iters: 3584, time: 0.064) G_GAN: 0.452 G_GAN_Feat: 0.699 G_ID: 0.116 G_Rec: 0.321 D_GP: 0.021 D_real: 1.356 D_fake: 0.550 +(epoch: 203, iters: 3984, time: 0.064) G_GAN: 0.231 G_GAN_Feat: 0.914 G_ID: 0.204 G_Rec: 0.436 D_GP: 0.033 D_real: 0.904 D_fake: 0.769 +(epoch: 203, iters: 4384, time: 0.064) G_GAN: 0.293 G_GAN_Feat: 0.640 G_ID: 0.127 G_Rec: 0.361 D_GP: 0.021 D_real: 1.214 D_fake: 0.710 +(epoch: 203, iters: 4784, time: 0.064) G_GAN: 0.355 G_GAN_Feat: 0.924 G_ID: 0.196 G_Rec: 0.445 D_GP: 0.052 D_real: 0.802 D_fake: 0.650 +(epoch: 203, iters: 5184, time: 0.064) G_GAN: 0.045 G_GAN_Feat: 0.778 G_ID: 0.133 G_Rec: 0.291 D_GP: 0.065 D_real: 0.618 D_fake: 0.956 +(epoch: 203, iters: 5584, time: 0.064) G_GAN: 0.490 G_GAN_Feat: 1.057 G_ID: 0.197 G_Rec: 0.475 D_GP: 0.046 D_real: 0.753 D_fake: 0.526 +(epoch: 203, iters: 5984, time: 0.064) G_GAN: 0.490 G_GAN_Feat: 0.576 G_ID: 0.141 G_Rec: 0.314 D_GP: 0.018 D_real: 1.503 D_fake: 0.513 +(epoch: 203, iters: 6384, time: 0.064) G_GAN: 0.366 G_GAN_Feat: 0.814 G_ID: 0.221 G_Rec: 0.447 D_GP: 0.022 D_real: 1.048 D_fake: 0.636 +(epoch: 203, iters: 6784, time: 0.064) G_GAN: 0.223 G_GAN_Feat: 0.623 G_ID: 0.136 G_Rec: 0.312 D_GP: 0.022 D_real: 1.182 D_fake: 0.777 +(epoch: 203, iters: 7184, time: 0.064) G_GAN: 0.450 G_GAN_Feat: 0.697 G_ID: 0.176 G_Rec: 0.403 D_GP: 0.022 D_real: 1.184 D_fake: 0.551 +(epoch: 203, iters: 7584, time: 0.064) G_GAN: 0.013 G_GAN_Feat: 0.624 G_ID: 0.194 G_Rec: 0.319 D_GP: 0.027 D_real: 0.939 D_fake: 0.988 +(epoch: 203, iters: 7984, time: 0.064) G_GAN: 0.273 G_GAN_Feat: 0.788 G_ID: 0.192 G_Rec: 0.384 D_GP: 0.022 D_real: 1.114 D_fake: 0.728 +(epoch: 203, iters: 8384, time: 0.064) G_GAN: 0.272 G_GAN_Feat: 0.552 G_ID: 0.138 G_Rec: 0.285 D_GP: 0.020 D_real: 1.219 D_fake: 0.728 +(epoch: 204, iters: 176, time: 0.064) G_GAN: 0.516 G_GAN_Feat: 0.803 G_ID: 0.183 G_Rec: 0.399 D_GP: 0.026 D_real: 1.202 D_fake: 0.491 +(epoch: 204, iters: 576, time: 0.064) G_GAN: 0.240 G_GAN_Feat: 0.647 G_ID: 0.124 G_Rec: 0.306 D_GP: 0.045 D_real: 1.149 D_fake: 0.760 +(epoch: 204, iters: 976, time: 0.064) G_GAN: 0.234 G_GAN_Feat: 0.828 G_ID: 0.221 G_Rec: 0.402 D_GP: 0.023 D_real: 0.999 D_fake: 0.766 +(epoch: 204, iters: 1376, time: 0.064) G_GAN: 0.124 G_GAN_Feat: 0.661 G_ID: 0.137 G_Rec: 0.313 D_GP: 0.042 D_real: 1.022 D_fake: 0.876 +(epoch: 204, iters: 1776, time: 0.064) G_GAN: 0.331 G_GAN_Feat: 0.783 G_ID: 0.239 G_Rec: 0.404 D_GP: 0.028 D_real: 1.106 D_fake: 0.671 +(epoch: 204, iters: 2176, time: 0.064) G_GAN: 0.110 G_GAN_Feat: 0.705 G_ID: 0.144 G_Rec: 0.309 D_GP: 0.064 D_real: 1.002 D_fake: 0.890 +(epoch: 204, iters: 2576, time: 0.064) G_GAN: 0.343 G_GAN_Feat: 0.865 G_ID: 0.179 G_Rec: 0.405 D_GP: 0.035 D_real: 1.060 D_fake: 0.662 +(epoch: 204, iters: 2976, time: 0.064) G_GAN: -0.006 G_GAN_Feat: 0.838 G_ID: 0.124 G_Rec: 0.364 D_GP: 0.108 D_real: 0.773 D_fake: 1.007 +(epoch: 204, iters: 3376, time: 0.064) G_GAN: 0.656 G_GAN_Feat: 0.855 G_ID: 0.168 G_Rec: 0.418 D_GP: 0.030 D_real: 1.312 D_fake: 0.373 +(epoch: 204, iters: 3776, time: 0.064) G_GAN: 0.263 G_GAN_Feat: 0.668 G_ID: 0.144 G_Rec: 0.296 D_GP: 0.021 D_real: 1.192 D_fake: 0.737 +(epoch: 204, iters: 4176, time: 0.064) G_GAN: 0.345 G_GAN_Feat: 0.743 G_ID: 0.166 G_Rec: 0.358 D_GP: 0.026 D_real: 1.068 D_fake: 0.656 +(epoch: 204, iters: 4576, time: 0.064) G_GAN: 0.365 G_GAN_Feat: 0.689 G_ID: 0.126 G_Rec: 0.293 D_GP: 0.027 D_real: 1.165 D_fake: 0.636 +(epoch: 204, iters: 4976, time: 0.064) G_GAN: 0.388 G_GAN_Feat: 0.727 G_ID: 0.181 G_Rec: 0.371 D_GP: 0.023 D_real: 1.170 D_fake: 0.612 +(epoch: 204, iters: 5376, time: 0.064) G_GAN: 0.372 G_GAN_Feat: 0.639 G_ID: 0.122 G_Rec: 0.287 D_GP: 0.025 D_real: 1.289 D_fake: 0.631 +(epoch: 204, iters: 5776, time: 0.064) G_GAN: 0.344 G_GAN_Feat: 0.799 G_ID: 0.212 G_Rec: 0.412 D_GP: 0.025 D_real: 1.150 D_fake: 0.658 +(epoch: 204, iters: 6176, time: 0.064) G_GAN: 0.073 G_GAN_Feat: 0.734 G_ID: 0.141 G_Rec: 0.325 D_GP: 0.033 D_real: 0.883 D_fake: 0.927 +(epoch: 204, iters: 6576, time: 0.064) G_GAN: 0.473 G_GAN_Feat: 0.809 G_ID: 0.194 G_Rec: 0.395 D_GP: 0.023 D_real: 1.271 D_fake: 0.530 +(epoch: 204, iters: 6976, time: 0.064) G_GAN: 0.493 G_GAN_Feat: 0.612 G_ID: 0.145 G_Rec: 0.269 D_GP: 0.025 D_real: 1.408 D_fake: 0.509 +(epoch: 204, iters: 7376, time: 0.064) G_GAN: 0.694 G_GAN_Feat: 0.911 G_ID: 0.173 G_Rec: 0.465 D_GP: 0.028 D_real: 1.416 D_fake: 0.329 +(epoch: 204, iters: 7776, time: 0.064) G_GAN: 0.437 G_GAN_Feat: 0.802 G_ID: 0.143 G_Rec: 0.316 D_GP: 0.039 D_real: 1.153 D_fake: 0.564 +(epoch: 204, iters: 8176, time: 0.064) G_GAN: 0.584 G_GAN_Feat: 0.817 G_ID: 0.160 G_Rec: 0.395 D_GP: 0.020 D_real: 1.343 D_fake: 0.431 +(epoch: 204, iters: 8576, time: 0.064) G_GAN: 0.202 G_GAN_Feat: 0.657 G_ID: 0.141 G_Rec: 0.308 D_GP: 0.029 D_real: 1.057 D_fake: 0.798 +(epoch: 205, iters: 368, time: 0.064) G_GAN: 0.404 G_GAN_Feat: 0.954 G_ID: 0.154 G_Rec: 0.459 D_GP: 0.041 D_real: 0.876 D_fake: 0.624 +(epoch: 205, iters: 768, time: 0.064) G_GAN: 0.272 G_GAN_Feat: 0.764 G_ID: 0.123 G_Rec: 0.348 D_GP: 0.033 D_real: 1.044 D_fake: 0.730 +(epoch: 205, iters: 1168, time: 0.064) G_GAN: 0.547 G_GAN_Feat: 1.158 G_ID: 0.173 G_Rec: 0.462 D_GP: 0.134 D_real: 1.037 D_fake: 0.471 +(epoch: 205, iters: 1568, time: 0.064) G_GAN: 0.360 G_GAN_Feat: 0.710 G_ID: 0.134 G_Rec: 0.296 D_GP: 0.029 D_real: 1.250 D_fake: 0.653 +(epoch: 205, iters: 1968, time: 0.064) G_GAN: 0.413 G_GAN_Feat: 0.951 G_ID: 0.154 G_Rec: 0.415 D_GP: 0.044 D_real: 0.848 D_fake: 0.591 +(epoch: 205, iters: 2368, time: 0.064) G_GAN: 0.331 G_GAN_Feat: 0.725 G_ID: 0.125 G_Rec: 0.317 D_GP: 0.037 D_real: 1.121 D_fake: 0.681 +(epoch: 205, iters: 2768, time: 0.064) G_GAN: 0.865 G_GAN_Feat: 0.880 G_ID: 0.175 G_Rec: 0.375 D_GP: 0.029 D_real: 1.547 D_fake: 0.206 +(epoch: 205, iters: 3168, time: 0.064) G_GAN: 0.344 G_GAN_Feat: 0.771 G_ID: 0.124 G_Rec: 0.320 D_GP: 0.035 D_real: 1.136 D_fake: 0.662 +(epoch: 205, iters: 3568, time: 0.064) G_GAN: 0.864 G_GAN_Feat: 1.269 G_ID: 0.172 G_Rec: 0.440 D_GP: 0.133 D_real: 0.624 D_fake: 0.254 +(epoch: 205, iters: 3968, time: 0.064) G_GAN: 0.418 G_GAN_Feat: 0.737 G_ID: 0.143 G_Rec: 0.327 D_GP: 0.023 D_real: 1.325 D_fake: 0.590 +(epoch: 205, iters: 4368, time: 0.064) G_GAN: 0.444 G_GAN_Feat: 1.039 G_ID: 0.188 G_Rec: 0.402 D_GP: 0.028 D_real: 0.778 D_fake: 0.561 +(epoch: 205, iters: 4768, time: 0.064) G_GAN: 0.381 G_GAN_Feat: 0.617 G_ID: 0.117 G_Rec: 0.288 D_GP: 0.023 D_real: 1.318 D_fake: 0.619 +(epoch: 205, iters: 5168, time: 0.064) G_GAN: 0.386 G_GAN_Feat: 0.892 G_ID: 0.181 G_Rec: 0.456 D_GP: 0.035 D_real: 1.023 D_fake: 0.617 +(epoch: 205, iters: 5568, time: 0.064) G_GAN: 0.283 G_GAN_Feat: 0.676 G_ID: 0.131 G_Rec: 0.307 D_GP: 0.033 D_real: 1.099 D_fake: 0.718 +(epoch: 205, iters: 5968, time: 0.064) G_GAN: 0.396 G_GAN_Feat: 0.856 G_ID: 0.167 G_Rec: 0.445 D_GP: 0.026 D_real: 1.003 D_fake: 0.608 +(epoch: 205, iters: 6368, time: 0.064) G_GAN: 0.203 G_GAN_Feat: 0.608 G_ID: 0.149 G_Rec: 0.270 D_GP: 0.023 D_real: 1.137 D_fake: 0.797 +(epoch: 205, iters: 6768, time: 0.064) G_GAN: 0.350 G_GAN_Feat: 1.003 G_ID: 0.175 G_Rec: 0.406 D_GP: 0.048 D_real: 0.760 D_fake: 0.652 +(epoch: 205, iters: 7168, time: 0.064) G_GAN: 0.232 G_GAN_Feat: 0.696 G_ID: 0.144 G_Rec: 0.280 D_GP: 0.022 D_real: 1.067 D_fake: 0.768 +(epoch: 205, iters: 7568, time: 0.064) G_GAN: 0.476 G_GAN_Feat: 1.145 G_ID: 0.173 G_Rec: 0.439 D_GP: 0.106 D_real: 0.507 D_fake: 0.526 +(epoch: 205, iters: 7968, time: 0.064) G_GAN: 0.119 G_GAN_Feat: 0.743 G_ID: 0.148 G_Rec: 0.317 D_GP: 0.020 D_real: 0.997 D_fake: 0.881 +(epoch: 205, iters: 8368, time: 0.064) G_GAN: 0.506 G_GAN_Feat: 1.257 G_ID: 0.207 G_Rec: 0.499 D_GP: 0.362 D_real: 0.349 D_fake: 0.502 +(epoch: 206, iters: 160, time: 0.064) G_GAN: 0.098 G_GAN_Feat: 0.752 G_ID: 0.137 G_Rec: 0.374 D_GP: 0.035 D_real: 0.779 D_fake: 0.902 +(epoch: 206, iters: 560, time: 0.064) G_GAN: 0.472 G_GAN_Feat: 0.921 G_ID: 0.209 G_Rec: 0.427 D_GP: 0.024 D_real: 1.107 D_fake: 0.528 +(epoch: 206, iters: 960, time: 0.064) G_GAN: 0.145 G_GAN_Feat: 0.673 G_ID: 0.172 G_Rec: 0.279 D_GP: 0.029 D_real: 1.174 D_fake: 0.856 +(epoch: 206, iters: 1360, time: 0.064) G_GAN: 0.462 G_GAN_Feat: 0.933 G_ID: 0.185 G_Rec: 0.365 D_GP: 0.030 D_real: 0.996 D_fake: 0.540 +(epoch: 206, iters: 1760, time: 0.064) G_GAN: 0.219 G_GAN_Feat: 0.764 G_ID: 0.131 G_Rec: 0.324 D_GP: 0.039 D_real: 0.962 D_fake: 0.781 +(epoch: 206, iters: 2160, time: 0.064) G_GAN: 0.244 G_GAN_Feat: 0.834 G_ID: 0.202 G_Rec: 0.415 D_GP: 0.027 D_real: 0.960 D_fake: 0.756 +(epoch: 206, iters: 2560, time: 0.064) G_GAN: 0.322 G_GAN_Feat: 0.704 G_ID: 0.172 G_Rec: 0.283 D_GP: 0.027 D_real: 1.186 D_fake: 0.679 +(epoch: 206, iters: 2960, time: 0.064) G_GAN: 0.275 G_GAN_Feat: 0.785 G_ID: 0.182 G_Rec: 0.324 D_GP: 0.022 D_real: 1.022 D_fake: 0.726 +(epoch: 206, iters: 3360, time: 0.064) G_GAN: 0.650 G_GAN_Feat: 0.748 G_ID: 0.131 G_Rec: 0.333 D_GP: 0.024 D_real: 1.523 D_fake: 0.367 +(epoch: 206, iters: 3760, time: 0.064) G_GAN: 0.556 G_GAN_Feat: 0.852 G_ID: 0.155 G_Rec: 0.394 D_GP: 0.026 D_real: 1.312 D_fake: 0.448 +(epoch: 206, iters: 4160, time: 0.064) G_GAN: 0.274 G_GAN_Feat: 0.577 G_ID: 0.171 G_Rec: 0.311 D_GP: 0.016 D_real: 1.214 D_fake: 0.729 +(epoch: 206, iters: 4560, time: 0.064) G_GAN: 0.406 G_GAN_Feat: 0.696 G_ID: 0.187 G_Rec: 0.332 D_GP: 0.021 D_real: 1.214 D_fake: 0.594 +(epoch: 206, iters: 4960, time: 0.064) G_GAN: 0.201 G_GAN_Feat: 0.667 G_ID: 0.147 G_Rec: 0.332 D_GP: 0.035 D_real: 1.052 D_fake: 0.799 +(epoch: 206, iters: 5360, time: 0.064) G_GAN: 0.419 G_GAN_Feat: 0.880 G_ID: 0.171 G_Rec: 0.378 D_GP: 0.028 D_real: 1.137 D_fake: 0.582 +(epoch: 206, iters: 5760, time: 0.064) G_GAN: 0.182 G_GAN_Feat: 0.873 G_ID: 0.165 G_Rec: 0.312 D_GP: 0.197 D_real: 0.482 D_fake: 0.818 +(epoch: 206, iters: 6160, time: 0.064) G_GAN: 0.466 G_GAN_Feat: 0.907 G_ID: 0.166 G_Rec: 0.389 D_GP: 0.038 D_real: 1.030 D_fake: 0.537 +(epoch: 206, iters: 6560, time: 0.064) G_GAN: 0.455 G_GAN_Feat: 0.640 G_ID: 0.123 G_Rec: 0.288 D_GP: 0.023 D_real: 1.384 D_fake: 0.547 +(epoch: 206, iters: 6960, time: 0.064) G_GAN: 0.346 G_GAN_Feat: 1.217 G_ID: 0.170 G_Rec: 0.410 D_GP: 0.488 D_real: 0.362 D_fake: 0.664 +(epoch: 206, iters: 7360, time: 0.064) G_GAN: 0.146 G_GAN_Feat: 0.723 G_ID: 0.139 G_Rec: 0.284 D_GP: 0.026 D_real: 1.040 D_fake: 0.854 +(epoch: 206, iters: 7760, time: 0.064) G_GAN: 0.604 G_GAN_Feat: 0.935 G_ID: 0.169 G_Rec: 0.392 D_GP: 0.043 D_real: 1.218 D_fake: 0.481 +(epoch: 206, iters: 8160, time: 0.064) G_GAN: 0.258 G_GAN_Feat: 0.893 G_ID: 0.165 G_Rec: 0.353 D_GP: 0.060 D_real: 0.652 D_fake: 0.756 +(epoch: 206, iters: 8560, time: 0.064) G_GAN: 0.261 G_GAN_Feat: 0.662 G_ID: 0.190 G_Rec: 0.368 D_GP: 0.018 D_real: 1.103 D_fake: 0.740 +(epoch: 207, iters: 352, time: 0.064) G_GAN: 0.386 G_GAN_Feat: 0.533 G_ID: 0.113 G_Rec: 0.289 D_GP: 0.016 D_real: 1.356 D_fake: 0.615 +(epoch: 207, iters: 752, time: 0.064) G_GAN: 0.343 G_GAN_Feat: 0.721 G_ID: 0.177 G_Rec: 0.422 D_GP: 0.017 D_real: 1.181 D_fake: 0.659 +(epoch: 207, iters: 1152, time: 0.064) G_GAN: 0.041 G_GAN_Feat: 0.529 G_ID: 0.175 G_Rec: 0.295 D_GP: 0.018 D_real: 1.017 D_fake: 0.959 +(epoch: 207, iters: 1552, time: 0.064) G_GAN: 0.354 G_GAN_Feat: 0.750 G_ID: 0.162 G_Rec: 0.452 D_GP: 0.022 D_real: 1.067 D_fake: 0.647 +(epoch: 207, iters: 1952, time: 0.064) G_GAN: 0.062 G_GAN_Feat: 0.561 G_ID: 0.149 G_Rec: 0.298 D_GP: 0.020 D_real: 1.011 D_fake: 0.938 +(epoch: 207, iters: 2352, time: 0.064) G_GAN: 0.040 G_GAN_Feat: 0.753 G_ID: 0.163 G_Rec: 0.391 D_GP: 0.024 D_real: 0.751 D_fake: 0.960 +(epoch: 207, iters: 2752, time: 0.064) G_GAN: 0.122 G_GAN_Feat: 0.568 G_ID: 0.148 G_Rec: 0.315 D_GP: 0.020 D_real: 1.072 D_fake: 0.879 +(epoch: 207, iters: 3152, time: 0.064) G_GAN: 0.150 G_GAN_Feat: 0.772 G_ID: 0.191 G_Rec: 0.375 D_GP: 0.028 D_real: 0.918 D_fake: 0.850 +(epoch: 207, iters: 3552, time: 0.064) G_GAN: -0.010 G_GAN_Feat: 0.667 G_ID: 0.128 G_Rec: 0.334 D_GP: 0.055 D_real: 0.905 D_fake: 1.010 +(epoch: 207, iters: 3952, time: 0.064) G_GAN: 0.245 G_GAN_Feat: 0.936 G_ID: 0.186 G_Rec: 0.417 D_GP: 0.050 D_real: 0.720 D_fake: 0.756 +(epoch: 207, iters: 4352, time: 0.064) G_GAN: 0.145 G_GAN_Feat: 0.621 G_ID: 0.145 G_Rec: 0.285 D_GP: 0.026 D_real: 1.088 D_fake: 0.855 +(epoch: 207, iters: 4752, time: 0.064) G_GAN: 0.471 G_GAN_Feat: 0.887 G_ID: 0.165 G_Rec: 0.398 D_GP: 0.049 D_real: 1.079 D_fake: 0.539 +(epoch: 207, iters: 5152, time: 0.064) G_GAN: -0.056 G_GAN_Feat: 0.733 G_ID: 0.121 G_Rec: 0.313 D_GP: 0.083 D_real: 0.651 D_fake: 1.056 +(epoch: 207, iters: 5552, time: 0.064) G_GAN: 0.238 G_GAN_Feat: 0.872 G_ID: 0.198 G_Rec: 0.393 D_GP: 0.030 D_real: 0.907 D_fake: 0.762 +(epoch: 207, iters: 5952, time: 0.064) G_GAN: 0.117 G_GAN_Feat: 0.757 G_ID: 0.158 G_Rec: 0.334 D_GP: 0.030 D_real: 0.990 D_fake: 0.885 +(epoch: 207, iters: 6352, time: 0.064) G_GAN: 0.498 G_GAN_Feat: 0.825 G_ID: 0.162 G_Rec: 0.379 D_GP: 0.024 D_real: 1.218 D_fake: 0.509 +(epoch: 207, iters: 6752, time: 0.064) G_GAN: 0.148 G_GAN_Feat: 0.684 G_ID: 0.136 G_Rec: 0.317 D_GP: 0.030 D_real: 0.965 D_fake: 0.853 +(epoch: 207, iters: 7152, time: 0.064) G_GAN: 0.346 G_GAN_Feat: 1.046 G_ID: 0.165 G_Rec: 0.457 D_GP: 0.107 D_real: 0.657 D_fake: 0.654 +(epoch: 207, iters: 7552, time: 0.064) G_GAN: 0.278 G_GAN_Feat: 0.765 G_ID: 0.144 G_Rec: 0.324 D_GP: 0.041 D_real: 0.941 D_fake: 0.723 +(epoch: 207, iters: 7952, time: 0.064) G_GAN: 0.313 G_GAN_Feat: 0.769 G_ID: 0.205 G_Rec: 0.386 D_GP: 0.023 D_real: 1.103 D_fake: 0.687 +(epoch: 207, iters: 8352, time: 0.064) G_GAN: 0.038 G_GAN_Feat: 0.779 G_ID: 0.152 G_Rec: 0.345 D_GP: 0.058 D_real: 0.823 D_fake: 0.962 +(epoch: 208, iters: 144, time: 0.064) G_GAN: 0.402 G_GAN_Feat: 1.182 G_ID: 0.170 G_Rec: 0.414 D_GP: 0.101 D_real: 0.458 D_fake: 0.602 +(epoch: 208, iters: 544, time: 0.064) G_GAN: 0.214 G_GAN_Feat: 0.742 G_ID: 0.139 G_Rec: 0.311 D_GP: 0.053 D_real: 0.832 D_fake: 0.787 +(epoch: 208, iters: 944, time: 0.064) G_GAN: 0.381 G_GAN_Feat: 0.963 G_ID: 0.193 G_Rec: 0.446 D_GP: 0.060 D_real: 0.924 D_fake: 0.622 +(epoch: 208, iters: 1344, time: 0.064) G_GAN: 0.275 G_GAN_Feat: 0.791 G_ID: 0.131 G_Rec: 0.352 D_GP: 0.034 D_real: 0.956 D_fake: 0.726 +(epoch: 208, iters: 1744, time: 0.064) G_GAN: 0.199 G_GAN_Feat: 0.951 G_ID: 0.193 G_Rec: 0.402 D_GP: 0.044 D_real: 0.589 D_fake: 0.801 +(epoch: 208, iters: 2144, time: 0.064) G_GAN: -0.002 G_GAN_Feat: 0.654 G_ID: 0.154 G_Rec: 0.286 D_GP: 0.024 D_real: 0.962 D_fake: 1.002 +(epoch: 208, iters: 2544, time: 0.064) G_GAN: 0.270 G_GAN_Feat: 0.897 G_ID: 0.188 G_Rec: 0.379 D_GP: 0.023 D_real: 0.961 D_fake: 0.730 +(epoch: 208, iters: 2944, time: 0.064) G_GAN: 0.287 G_GAN_Feat: 0.856 G_ID: 0.119 G_Rec: 0.322 D_GP: 0.108 D_real: 1.013 D_fake: 0.714 +(epoch: 208, iters: 3344, time: 0.064) G_GAN: 0.432 G_GAN_Feat: 0.818 G_ID: 0.189 G_Rec: 0.354 D_GP: 0.028 D_real: 1.155 D_fake: 0.572 +(epoch: 208, iters: 3744, time: 0.064) G_GAN: 0.053 G_GAN_Feat: 0.812 G_ID: 0.163 G_Rec: 0.335 D_GP: 0.073 D_real: 0.726 D_fake: 0.947 +(epoch: 208, iters: 4144, time: 0.064) G_GAN: 0.925 G_GAN_Feat: 1.102 G_ID: 0.203 G_Rec: 0.418 D_GP: 0.047 D_real: 1.338 D_fake: 0.455 +(epoch: 208, iters: 4544, time: 0.064) G_GAN: -0.071 G_GAN_Feat: 0.645 G_ID: 0.152 G_Rec: 0.291 D_GP: 0.025 D_real: 0.862 D_fake: 1.071 +(epoch: 208, iters: 4944, time: 0.064) G_GAN: 0.379 G_GAN_Feat: 0.991 G_ID: 0.180 G_Rec: 0.414 D_GP: 0.057 D_real: 0.725 D_fake: 0.624 +(epoch: 208, iters: 5344, time: 0.064) G_GAN: 0.591 G_GAN_Feat: 0.758 G_ID: 0.141 G_Rec: 0.311 D_GP: 0.035 D_real: 1.423 D_fake: 0.466 +(epoch: 208, iters: 5744, time: 0.064) G_GAN: 0.204 G_GAN_Feat: 0.831 G_ID: 0.172 G_Rec: 0.400 D_GP: 0.025 D_real: 0.920 D_fake: 0.797 +(epoch: 208, iters: 6144, time: 0.064) G_GAN: 0.125 G_GAN_Feat: 0.816 G_ID: 0.189 G_Rec: 0.298 D_GP: 0.043 D_real: 0.740 D_fake: 0.875 +(epoch: 208, iters: 6544, time: 0.064) G_GAN: 0.783 G_GAN_Feat: 1.110 G_ID: 0.203 G_Rec: 0.459 D_GP: 0.407 D_real: 0.805 D_fake: 0.329 +(epoch: 208, iters: 6944, time: 0.064) G_GAN: 0.329 G_GAN_Feat: 0.706 G_ID: 0.143 G_Rec: 0.295 D_GP: 0.034 D_real: 1.167 D_fake: 0.671 +(epoch: 208, iters: 7344, time: 0.064) G_GAN: 0.529 G_GAN_Feat: 0.970 G_ID: 0.167 G_Rec: 0.411 D_GP: 0.080 D_real: 0.939 D_fake: 0.476 +(epoch: 208, iters: 7744, time: 0.064) G_GAN: 0.037 G_GAN_Feat: 0.774 G_ID: 0.138 G_Rec: 0.351 D_GP: 0.067 D_real: 0.679 D_fake: 0.964 +(epoch: 208, iters: 8144, time: 0.064) G_GAN: 0.488 G_GAN_Feat: 0.892 G_ID: 0.154 G_Rec: 0.408 D_GP: 0.025 D_real: 1.096 D_fake: 0.519 +(epoch: 208, iters: 8544, time: 0.064) G_GAN: 0.192 G_GAN_Feat: 0.849 G_ID: 0.133 G_Rec: 0.319 D_GP: 0.029 D_real: 0.849 D_fake: 0.808 +(epoch: 209, iters: 336, time: 0.064) G_GAN: 0.592 G_GAN_Feat: 0.878 G_ID: 0.180 G_Rec: 0.431 D_GP: 0.027 D_real: 1.293 D_fake: 0.429 +(epoch: 209, iters: 736, time: 0.064) G_GAN: 0.068 G_GAN_Feat: 0.612 G_ID: 0.144 G_Rec: 0.312 D_GP: 0.023 D_real: 1.019 D_fake: 0.932 +(epoch: 209, iters: 1136, time: 0.064) G_GAN: 0.196 G_GAN_Feat: 0.853 G_ID: 0.190 G_Rec: 0.423 D_GP: 0.035 D_real: 0.870 D_fake: 0.804 +(epoch: 209, iters: 1536, time: 0.064) G_GAN: -0.025 G_GAN_Feat: 0.722 G_ID: 0.156 G_Rec: 0.300 D_GP: 0.055 D_real: 0.671 D_fake: 1.026 +(epoch: 209, iters: 1936, time: 0.064) G_GAN: 0.354 G_GAN_Feat: 0.919 G_ID: 0.163 G_Rec: 0.387 D_GP: 0.279 D_real: 0.689 D_fake: 0.653 +(epoch: 209, iters: 2336, time: 0.064) G_GAN: 0.187 G_GAN_Feat: 0.696 G_ID: 0.118 G_Rec: 0.316 D_GP: 0.036 D_real: 0.968 D_fake: 0.814 +(epoch: 209, iters: 2736, time: 0.064) G_GAN: 0.475 G_GAN_Feat: 0.778 G_ID: 0.192 G_Rec: 0.329 D_GP: 0.023 D_real: 1.316 D_fake: 0.535 +(epoch: 209, iters: 3136, time: 0.064) G_GAN: 0.299 G_GAN_Feat: 0.661 G_ID: 0.130 G_Rec: 0.279 D_GP: 0.023 D_real: 1.242 D_fake: 0.701 +(epoch: 209, iters: 3536, time: 0.064) G_GAN: 0.113 G_GAN_Feat: 0.780 G_ID: 0.179 G_Rec: 0.354 D_GP: 0.021 D_real: 1.033 D_fake: 0.887 +(epoch: 209, iters: 3936, time: 0.064) G_GAN: 0.244 G_GAN_Feat: 0.688 G_ID: 0.136 G_Rec: 0.344 D_GP: 0.025 D_real: 1.143 D_fake: 0.756 +(epoch: 209, iters: 4336, time: 0.064) G_GAN: 0.281 G_GAN_Feat: 0.891 G_ID: 0.173 G_Rec: 0.370 D_GP: 0.037 D_real: 0.937 D_fake: 0.719 +(epoch: 209, iters: 4736, time: 0.064) G_GAN: 0.305 G_GAN_Feat: 0.755 G_ID: 0.126 G_Rec: 0.287 D_GP: 0.049 D_real: 0.902 D_fake: 0.696 +(epoch: 209, iters: 5136, time: 0.064) G_GAN: 0.554 G_GAN_Feat: 1.060 G_ID: 0.169 G_Rec: 0.438 D_GP: 0.059 D_real: 0.889 D_fake: 0.462 +(epoch: 209, iters: 5536, time: 0.064) G_GAN: 0.361 G_GAN_Feat: 0.713 G_ID: 0.135 G_Rec: 0.282 D_GP: 0.026 D_real: 1.240 D_fake: 0.640 +(epoch: 209, iters: 5936, time: 0.064) G_GAN: 0.562 G_GAN_Feat: 0.965 G_ID: 0.156 G_Rec: 0.430 D_GP: 0.028 D_real: 1.194 D_fake: 0.446 +(epoch: 209, iters: 6336, time: 0.064) G_GAN: 0.193 G_GAN_Feat: 0.783 G_ID: 0.128 G_Rec: 0.317 D_GP: 0.027 D_real: 0.919 D_fake: 0.812 +(epoch: 209, iters: 6736, time: 0.064) G_GAN: 0.470 G_GAN_Feat: 0.834 G_ID: 0.162 G_Rec: 0.368 D_GP: 0.022 D_real: 1.328 D_fake: 0.533 +(epoch: 209, iters: 7136, time: 0.064) G_GAN: 0.012 G_GAN_Feat: 0.749 G_ID: 0.134 G_Rec: 0.297 D_GP: 0.042 D_real: 0.863 D_fake: 0.988 +(epoch: 209, iters: 7536, time: 0.064) G_GAN: 0.634 G_GAN_Feat: 0.911 G_ID: 0.155 G_Rec: 0.449 D_GP: 0.026 D_real: 1.351 D_fake: 0.378 +(epoch: 209, iters: 7936, time: 0.064) G_GAN: 0.289 G_GAN_Feat: 0.772 G_ID: 0.132 G_Rec: 0.385 D_GP: 0.132 D_real: 0.912 D_fake: 0.712 +(epoch: 209, iters: 8336, time: 0.064) G_GAN: 0.415 G_GAN_Feat: 1.038 G_ID: 0.222 G_Rec: 0.410 D_GP: 0.039 D_real: 0.653 D_fake: 0.590 +(epoch: 210, iters: 128, time: 0.064) G_GAN: 0.107 G_GAN_Feat: 0.967 G_ID: 0.128 G_Rec: 0.361 D_GP: 0.224 D_real: 0.536 D_fake: 0.897 +(epoch: 210, iters: 528, time: 0.064) G_GAN: 0.654 G_GAN_Feat: 0.975 G_ID: 0.147 G_Rec: 0.394 D_GP: 0.026 D_real: 1.238 D_fake: 0.371 +(epoch: 210, iters: 928, time: 0.064) G_GAN: 0.361 G_GAN_Feat: 0.774 G_ID: 0.139 G_Rec: 0.314 D_GP: 0.025 D_real: 1.169 D_fake: 0.640 +(epoch: 210, iters: 1328, time: 0.064) G_GAN: 0.464 G_GAN_Feat: 0.933 G_ID: 0.179 G_Rec: 0.420 D_GP: 0.036 D_real: 1.015 D_fake: 0.540 +(epoch: 210, iters: 1728, time: 0.064) G_GAN: 0.058 G_GAN_Feat: 0.723 G_ID: 0.157 G_Rec: 0.347 D_GP: 0.025 D_real: 0.936 D_fake: 0.942 +(epoch: 210, iters: 2128, time: 0.064) G_GAN: 0.374 G_GAN_Feat: 0.879 G_ID: 0.181 G_Rec: 0.395 D_GP: 0.056 D_real: 0.885 D_fake: 0.633 +(epoch: 210, iters: 2528, time: 0.064) G_GAN: 0.512 G_GAN_Feat: 0.698 G_ID: 0.141 G_Rec: 0.316 D_GP: 0.023 D_real: 1.378 D_fake: 0.491 +(epoch: 210, iters: 2928, time: 0.064) G_GAN: 0.501 G_GAN_Feat: 0.840 G_ID: 0.193 G_Rec: 0.426 D_GP: 0.021 D_real: 1.229 D_fake: 0.505 +(epoch: 210, iters: 3328, time: 0.064) G_GAN: 0.406 G_GAN_Feat: 0.683 G_ID: 0.132 G_Rec: 0.307 D_GP: 0.048 D_real: 1.212 D_fake: 0.596 +(epoch: 210, iters: 3728, time: 0.064) G_GAN: 0.627 G_GAN_Feat: 0.919 G_ID: 0.180 G_Rec: 0.409 D_GP: 0.030 D_real: 1.349 D_fake: 0.401 +(epoch: 210, iters: 4128, time: 0.064) G_GAN: 0.201 G_GAN_Feat: 0.740 G_ID: 0.125 G_Rec: 0.285 D_GP: 0.027 D_real: 1.051 D_fake: 0.799 +(epoch: 210, iters: 4528, time: 0.064) G_GAN: 0.177 G_GAN_Feat: 1.167 G_ID: 0.228 G_Rec: 0.434 D_GP: 0.544 D_real: 0.306 D_fake: 0.823 +(epoch: 210, iters: 4928, time: 0.064) G_GAN: 0.317 G_GAN_Feat: 0.940 G_ID: 0.111 G_Rec: 0.318 D_GP: 0.057 D_real: 0.473 D_fake: 0.683 +(epoch: 210, iters: 5328, time: 0.064) G_GAN: 0.498 G_GAN_Feat: 0.759 G_ID: 0.172 G_Rec: 0.371 D_GP: 0.023 D_real: 1.277 D_fake: 0.515 +(epoch: 210, iters: 5728, time: 0.064) G_GAN: 0.077 G_GAN_Feat: 0.654 G_ID: 0.163 G_Rec: 0.330 D_GP: 0.025 D_real: 0.973 D_fake: 0.923 +(epoch: 210, iters: 6128, time: 0.064) G_GAN: 0.167 G_GAN_Feat: 0.801 G_ID: 0.209 G_Rec: 0.357 D_GP: 0.027 D_real: 0.874 D_fake: 0.833 +(epoch: 210, iters: 6528, time: 0.064) G_GAN: 0.241 G_GAN_Feat: 0.648 G_ID: 0.143 G_Rec: 0.339 D_GP: 0.021 D_real: 1.148 D_fake: 0.759 +(epoch: 210, iters: 6928, time: 0.064) G_GAN: 0.008 G_GAN_Feat: 1.020 G_ID: 0.211 G_Rec: 0.417 D_GP: 0.212 D_real: 0.192 D_fake: 0.992 +(epoch: 210, iters: 7328, time: 0.064) G_GAN: 0.308 G_GAN_Feat: 0.642 G_ID: 0.158 G_Rec: 0.279 D_GP: 0.022 D_real: 1.237 D_fake: 0.693 +(epoch: 210, iters: 7728, time: 0.064) G_GAN: 0.390 G_GAN_Feat: 0.842 G_ID: 0.192 G_Rec: 0.398 D_GP: 0.025 D_real: 1.186 D_fake: 0.611 +(epoch: 210, iters: 8128, time: 0.064) G_GAN: 0.327 G_GAN_Feat: 0.820 G_ID: 0.142 G_Rec: 0.310 D_GP: 0.027 D_real: 0.997 D_fake: 0.673 +(epoch: 210, iters: 8528, time: 0.064) G_GAN: 0.688 G_GAN_Feat: 0.952 G_ID: 0.165 G_Rec: 0.440 D_GP: 0.021 D_real: 1.403 D_fake: 0.348 +(epoch: 211, iters: 320, time: 0.064) G_GAN: 0.309 G_GAN_Feat: 0.594 G_ID: 0.142 G_Rec: 0.285 D_GP: 0.019 D_real: 1.238 D_fake: 0.692 +(epoch: 211, iters: 720, time: 0.064) G_GAN: 0.481 G_GAN_Feat: 1.014 G_ID: 0.181 G_Rec: 0.440 D_GP: 0.058 D_real: 0.849 D_fake: 0.538 +(epoch: 211, iters: 1120, time: 0.064) G_GAN: 0.210 G_GAN_Feat: 0.710 G_ID: 0.154 G_Rec: 0.304 D_GP: 0.023 D_real: 1.032 D_fake: 0.790 +(epoch: 211, iters: 1520, time: 0.064) G_GAN: 0.601 G_GAN_Feat: 0.863 G_ID: 0.190 G_Rec: 0.406 D_GP: 0.027 D_real: 1.422 D_fake: 0.426 +(epoch: 211, iters: 1920, time: 0.064) G_GAN: 0.249 G_GAN_Feat: 0.744 G_ID: 0.142 G_Rec: 0.342 D_GP: 0.053 D_real: 0.993 D_fake: 0.751 +(epoch: 211, iters: 2320, time: 0.064) G_GAN: 0.428 G_GAN_Feat: 0.865 G_ID: 0.164 G_Rec: 0.363 D_GP: 0.028 D_real: 1.122 D_fake: 0.573 +(epoch: 211, iters: 2720, time: 0.064) G_GAN: 0.127 G_GAN_Feat: 0.675 G_ID: 0.130 G_Rec: 0.286 D_GP: 0.022 D_real: 1.065 D_fake: 0.873 +(epoch: 211, iters: 3120, time: 0.064) G_GAN: 0.386 G_GAN_Feat: 0.824 G_ID: 0.179 G_Rec: 0.401 D_GP: 0.019 D_real: 1.092 D_fake: 0.616 +(epoch: 211, iters: 3520, time: 0.064) G_GAN: 0.426 G_GAN_Feat: 0.710 G_ID: 0.133 G_Rec: 0.292 D_GP: 0.025 D_real: 1.300 D_fake: 0.576 +(epoch: 211, iters: 3920, time: 0.064) G_GAN: 0.552 G_GAN_Feat: 1.065 G_ID: 0.183 G_Rec: 0.385 D_GP: 0.044 D_real: 0.547 D_fake: 0.451 +(epoch: 211, iters: 4320, time: 0.064) G_GAN: 0.187 G_GAN_Feat: 0.643 G_ID: 0.122 G_Rec: 0.280 D_GP: 0.024 D_real: 1.109 D_fake: 0.813 +(epoch: 211, iters: 4720, time: 0.064) G_GAN: 0.521 G_GAN_Feat: 0.928 G_ID: 0.184 G_Rec: 0.393 D_GP: 0.060 D_real: 1.055 D_fake: 0.484 +(epoch: 211, iters: 5120, time: 0.064) G_GAN: 0.469 G_GAN_Feat: 0.786 G_ID: 0.138 G_Rec: 0.315 D_GP: 0.035 D_real: 1.068 D_fake: 0.534 +(epoch: 211, iters: 5520, time: 0.064) G_GAN: 0.522 G_GAN_Feat: 1.029 G_ID: 0.175 G_Rec: 0.410 D_GP: 0.033 D_real: 0.757 D_fake: 0.512 +(epoch: 211, iters: 5920, time: 0.064) G_GAN: 0.385 G_GAN_Feat: 0.652 G_ID: 0.122 G_Rec: 0.284 D_GP: 0.021 D_real: 1.306 D_fake: 0.615 +(epoch: 211, iters: 6320, time: 0.064) G_GAN: 0.375 G_GAN_Feat: 0.905 G_ID: 0.192 G_Rec: 0.420 D_GP: 0.024 D_real: 1.024 D_fake: 0.634 +(epoch: 211, iters: 6720, time: 0.064) G_GAN: 0.008 G_GAN_Feat: 0.754 G_ID: 0.141 G_Rec: 0.337 D_GP: 0.032 D_real: 0.830 D_fake: 0.992 +(epoch: 211, iters: 7120, time: 0.064) G_GAN: 0.312 G_GAN_Feat: 0.958 G_ID: 0.157 G_Rec: 0.424 D_GP: 0.047 D_real: 0.817 D_fake: 0.690 +(epoch: 211, iters: 7520, time: 0.064) G_GAN: 0.203 G_GAN_Feat: 0.884 G_ID: 0.138 G_Rec: 0.347 D_GP: 0.090 D_real: 0.571 D_fake: 0.797 +(epoch: 211, iters: 7920, time: 0.064) G_GAN: 0.371 G_GAN_Feat: 0.937 G_ID: 0.205 G_Rec: 0.392 D_GP: 0.029 D_real: 0.984 D_fake: 0.633 +(epoch: 211, iters: 8320, time: 0.064) G_GAN: 0.152 G_GAN_Feat: 0.718 G_ID: 0.140 G_Rec: 0.331 D_GP: 0.021 D_real: 1.085 D_fake: 0.863 +(epoch: 212, iters: 112, time: 0.064) G_GAN: 0.367 G_GAN_Feat: 0.814 G_ID: 0.165 G_Rec: 0.410 D_GP: 0.023 D_real: 1.093 D_fake: 0.634 +(epoch: 212, iters: 512, time: 0.064) G_GAN: 0.137 G_GAN_Feat: 0.608 G_ID: 0.139 G_Rec: 0.292 D_GP: 0.021 D_real: 1.160 D_fake: 0.863 +(epoch: 212, iters: 912, time: 0.064) G_GAN: 0.427 G_GAN_Feat: 0.787 G_ID: 0.147 G_Rec: 0.401 D_GP: 0.019 D_real: 1.219 D_fake: 0.576 +(epoch: 212, iters: 1312, time: 0.064) G_GAN: 0.203 G_GAN_Feat: 0.680 G_ID: 0.123 G_Rec: 0.310 D_GP: 0.025 D_real: 1.085 D_fake: 0.797 +(epoch: 212, iters: 1712, time: 0.064) G_GAN: 0.747 G_GAN_Feat: 0.956 G_ID: 0.181 G_Rec: 0.427 D_GP: 0.025 D_real: 1.476 D_fake: 0.276 +(epoch: 212, iters: 2112, time: 0.064) G_GAN: 0.343 G_GAN_Feat: 0.733 G_ID: 0.128 G_Rec: 0.316 D_GP: 0.028 D_real: 1.220 D_fake: 0.657 +(epoch: 212, iters: 2512, time: 0.064) G_GAN: 0.446 G_GAN_Feat: 0.823 G_ID: 0.168 G_Rec: 0.378 D_GP: 0.025 D_real: 1.085 D_fake: 0.562 +(epoch: 212, iters: 2912, time: 0.064) G_GAN: 0.255 G_GAN_Feat: 0.688 G_ID: 0.122 G_Rec: 0.288 D_GP: 0.034 D_real: 1.048 D_fake: 0.745 +(epoch: 212, iters: 3312, time: 0.064) G_GAN: 0.498 G_GAN_Feat: 0.850 G_ID: 0.173 G_Rec: 0.395 D_GP: 0.026 D_real: 1.307 D_fake: 0.504 +(epoch: 212, iters: 3712, time: 0.064) G_GAN: 0.038 G_GAN_Feat: 0.593 G_ID: 0.135 G_Rec: 0.289 D_GP: 0.019 D_real: 1.005 D_fake: 0.962 +(epoch: 212, iters: 4112, time: 0.064) G_GAN: 0.222 G_GAN_Feat: 0.821 G_ID: 0.185 G_Rec: 0.450 D_GP: 0.027 D_real: 0.909 D_fake: 0.779 +(epoch: 212, iters: 4512, time: 0.064) G_GAN: 0.113 G_GAN_Feat: 0.670 G_ID: 0.119 G_Rec: 0.307 D_GP: 0.072 D_real: 0.901 D_fake: 0.887 +(epoch: 212, iters: 4912, time: 0.064) G_GAN: 0.452 G_GAN_Feat: 0.852 G_ID: 0.185 G_Rec: 0.401 D_GP: 0.031 D_real: 1.119 D_fake: 0.573 +(epoch: 212, iters: 5312, time: 0.064) G_GAN: 0.291 G_GAN_Feat: 0.812 G_ID: 0.125 G_Rec: 0.324 D_GP: 0.056 D_real: 0.871 D_fake: 0.709 +(epoch: 212, iters: 5712, time: 0.064) G_GAN: 0.123 G_GAN_Feat: 0.930 G_ID: 0.166 G_Rec: 0.394 D_GP: 0.058 D_real: 0.764 D_fake: 0.885 +(epoch: 212, iters: 6112, time: 0.064) G_GAN: 0.211 G_GAN_Feat: 0.788 G_ID: 0.151 G_Rec: 0.316 D_GP: 0.064 D_real: 0.773 D_fake: 0.789 +(epoch: 212, iters: 6512, time: 0.064) G_GAN: 0.513 G_GAN_Feat: 0.889 G_ID: 0.187 G_Rec: 0.403 D_GP: 0.025 D_real: 1.197 D_fake: 0.493 +(epoch: 212, iters: 6912, time: 0.064) G_GAN: 0.386 G_GAN_Feat: 0.871 G_ID: 0.121 G_Rec: 0.308 D_GP: 0.102 D_real: 0.618 D_fake: 0.617 +(epoch: 212, iters: 7312, time: 0.064) G_GAN: 0.669 G_GAN_Feat: 0.813 G_ID: 0.158 G_Rec: 0.460 D_GP: 0.019 D_real: 1.467 D_fake: 0.366 +(epoch: 212, iters: 7712, time: 0.064) G_GAN: 0.169 G_GAN_Feat: 0.624 G_ID: 0.133 G_Rec: 0.313 D_GP: 0.026 D_real: 1.094 D_fake: 0.831 +(epoch: 212, iters: 8112, time: 0.064) G_GAN: 0.436 G_GAN_Feat: 0.925 G_ID: 0.155 G_Rec: 0.409 D_GP: 0.064 D_real: 0.956 D_fake: 0.569 +(epoch: 212, iters: 8512, time: 0.064) G_GAN: 0.403 G_GAN_Feat: 0.757 G_ID: 0.144 G_Rec: 0.297 D_GP: 0.041 D_real: 1.132 D_fake: 0.597 +(epoch: 213, iters: 304, time: 0.064) G_GAN: 0.464 G_GAN_Feat: 0.830 G_ID: 0.183 G_Rec: 0.387 D_GP: 0.025 D_real: 1.154 D_fake: 0.548 +(epoch: 213, iters: 704, time: 0.064) G_GAN: 0.272 G_GAN_Feat: 0.732 G_ID: 0.122 G_Rec: 0.301 D_GP: 0.042 D_real: 1.067 D_fake: 0.728 +(epoch: 213, iters: 1104, time: 0.064) G_GAN: 0.299 G_GAN_Feat: 0.976 G_ID: 0.162 G_Rec: 0.447 D_GP: 0.025 D_real: 0.893 D_fake: 0.702 +(epoch: 213, iters: 1504, time: 0.064) G_GAN: -0.027 G_GAN_Feat: 0.676 G_ID: 0.132 G_Rec: 0.298 D_GP: 0.022 D_real: 0.873 D_fake: 1.027 +(epoch: 213, iters: 1904, time: 0.064) G_GAN: 0.373 G_GAN_Feat: 1.053 G_ID: 0.194 G_Rec: 0.446 D_GP: 0.159 D_real: 0.492 D_fake: 0.630 +(epoch: 213, iters: 2304, time: 0.064) G_GAN: 0.315 G_GAN_Feat: 0.816 G_ID: 0.139 G_Rec: 0.350 D_GP: 0.048 D_real: 0.892 D_fake: 0.686 +(epoch: 213, iters: 2704, time: 0.064) G_GAN: 0.662 G_GAN_Feat: 0.784 G_ID: 0.179 G_Rec: 0.399 D_GP: 0.019 D_real: 1.392 D_fake: 0.372 +(epoch: 213, iters: 3104, time: 0.064) G_GAN: 0.148 G_GAN_Feat: 0.592 G_ID: 0.134 G_Rec: 0.312 D_GP: 0.018 D_real: 1.097 D_fake: 0.852 +(epoch: 213, iters: 3504, time: 0.064) G_GAN: 0.332 G_GAN_Feat: 0.710 G_ID: 0.166 G_Rec: 0.350 D_GP: 0.026 D_real: 1.213 D_fake: 0.669 +(epoch: 213, iters: 3904, time: 0.064) G_GAN: 0.224 G_GAN_Feat: 0.599 G_ID: 0.106 G_Rec: 0.291 D_GP: 0.028 D_real: 1.143 D_fake: 0.776 +(epoch: 213, iters: 4304, time: 0.064) G_GAN: 0.346 G_GAN_Feat: 0.818 G_ID: 0.174 G_Rec: 0.408 D_GP: 0.046 D_real: 1.057 D_fake: 0.657 +(epoch: 213, iters: 4704, time: 0.064) G_GAN: 0.277 G_GAN_Feat: 0.768 G_ID: 0.135 G_Rec: 0.327 D_GP: 0.027 D_real: 1.111 D_fake: 0.723 +(epoch: 213, iters: 5104, time: 0.064) G_GAN: 0.515 G_GAN_Feat: 0.854 G_ID: 0.180 G_Rec: 0.400 D_GP: 0.039 D_real: 1.175 D_fake: 0.490 +(epoch: 213, iters: 5504, time: 0.064) G_GAN: 0.043 G_GAN_Feat: 0.741 G_ID: 0.124 G_Rec: 0.329 D_GP: 0.073 D_real: 0.856 D_fake: 0.957 +(epoch: 213, iters: 5904, time: 0.064) G_GAN: 0.267 G_GAN_Feat: 0.942 G_ID: 0.179 G_Rec: 0.399 D_GP: 0.139 D_real: 0.592 D_fake: 0.733 +(epoch: 213, iters: 6304, time: 0.064) G_GAN: 0.308 G_GAN_Feat: 0.830 G_ID: 0.158 G_Rec: 0.341 D_GP: 0.054 D_real: 0.912 D_fake: 0.694 +(epoch: 213, iters: 6704, time: 0.064) G_GAN: 0.259 G_GAN_Feat: 0.824 G_ID: 0.207 G_Rec: 0.355 D_GP: 0.026 D_real: 0.957 D_fake: 0.742 +(epoch: 213, iters: 7104, time: 0.064) G_GAN: 0.331 G_GAN_Feat: 0.704 G_ID: 0.125 G_Rec: 0.327 D_GP: 0.025 D_real: 1.228 D_fake: 0.674 +(epoch: 213, iters: 7504, time: 0.064) G_GAN: 0.404 G_GAN_Feat: 0.935 G_ID: 0.186 G_Rec: 0.386 D_GP: 0.043 D_real: 0.859 D_fake: 0.599 +(epoch: 213, iters: 7904, time: 0.064) G_GAN: 0.555 G_GAN_Feat: 0.793 G_ID: 0.121 G_Rec: 0.318 D_GP: 0.077 D_real: 1.223 D_fake: 0.474 +(epoch: 213, iters: 8304, time: 0.064) G_GAN: 0.224 G_GAN_Feat: 0.852 G_ID: 0.156 G_Rec: 0.413 D_GP: 0.026 D_real: 0.943 D_fake: 0.777 +(epoch: 214, iters: 96, time: 0.064) G_GAN: 0.225 G_GAN_Feat: 0.713 G_ID: 0.131 G_Rec: 0.316 D_GP: 0.050 D_real: 1.050 D_fake: 0.775 +(epoch: 214, iters: 496, time: 0.064) G_GAN: 0.328 G_GAN_Feat: 0.974 G_ID: 0.168 G_Rec: 0.386 D_GP: 0.025 D_real: 1.043 D_fake: 0.672 +(epoch: 214, iters: 896, time: 0.064) G_GAN: 0.213 G_GAN_Feat: 0.809 G_ID: 0.136 G_Rec: 0.319 D_GP: 0.038 D_real: 0.840 D_fake: 0.787 +(epoch: 214, iters: 1296, time: 0.064) G_GAN: 0.762 G_GAN_Feat: 0.903 G_ID: 0.190 G_Rec: 0.376 D_GP: 0.038 D_real: 1.428 D_fake: 0.347 +(epoch: 214, iters: 1696, time: 0.064) G_GAN: 0.118 G_GAN_Feat: 0.785 G_ID: 0.120 G_Rec: 0.278 D_GP: 0.070 D_real: 1.051 D_fake: 0.882 +(epoch: 214, iters: 2096, time: 0.064) G_GAN: 0.138 G_GAN_Feat: 0.995 G_ID: 0.171 G_Rec: 0.469 D_GP: 0.071 D_real: 0.617 D_fake: 0.864 +(epoch: 214, iters: 2496, time: 0.064) G_GAN: 0.010 G_GAN_Feat: 0.848 G_ID: 0.161 G_Rec: 0.332 D_GP: 0.141 D_real: 0.417 D_fake: 0.991 +(epoch: 214, iters: 2896, time: 0.064) G_GAN: 0.879 G_GAN_Feat: 0.807 G_ID: 0.166 G_Rec: 0.371 D_GP: 0.025 D_real: 1.612 D_fake: 0.220 +(epoch: 214, iters: 3296, time: 0.064) G_GAN: 0.137 G_GAN_Feat: 0.666 G_ID: 0.189 G_Rec: 0.288 D_GP: 0.025 D_real: 1.022 D_fake: 0.863 +(epoch: 214, iters: 3696, time: 0.064) G_GAN: 0.355 G_GAN_Feat: 0.951 G_ID: 0.198 G_Rec: 0.443 D_GP: 0.059 D_real: 1.032 D_fake: 0.647 +(epoch: 214, iters: 4096, time: 0.064) G_GAN: 0.268 G_GAN_Feat: 0.690 G_ID: 0.134 G_Rec: 0.269 D_GP: 0.023 D_real: 1.154 D_fake: 0.732 +(epoch: 214, iters: 4496, time: 0.064) G_GAN: 0.722 G_GAN_Feat: 0.771 G_ID: 0.148 G_Rec: 0.349 D_GP: 0.026 D_real: 1.522 D_fake: 0.291 +(epoch: 214, iters: 4896, time: 0.064) G_GAN: 0.237 G_GAN_Feat: 0.899 G_ID: 0.128 G_Rec: 0.324 D_GP: 0.622 D_real: 0.450 D_fake: 0.767 +(epoch: 214, iters: 5296, time: 0.064) G_GAN: 0.453 G_GAN_Feat: 0.840 G_ID: 0.180 G_Rec: 0.382 D_GP: 0.028 D_real: 1.162 D_fake: 0.548 +(epoch: 214, iters: 5696, time: 0.064) G_GAN: 0.307 G_GAN_Feat: 0.781 G_ID: 0.144 G_Rec: 0.335 D_GP: 0.054 D_real: 0.885 D_fake: 0.694 +(epoch: 214, iters: 6096, time: 0.064) G_GAN: 0.672 G_GAN_Feat: 0.883 G_ID: 0.171 G_Rec: 0.412 D_GP: 0.036 D_real: 1.309 D_fake: 0.364 +(epoch: 214, iters: 6496, time: 0.064) G_GAN: 0.214 G_GAN_Feat: 0.908 G_ID: 0.143 G_Rec: 0.341 D_GP: 0.092 D_real: 0.485 D_fake: 0.792 +(epoch: 214, iters: 6896, time: 0.064) G_GAN: 0.420 G_GAN_Feat: 0.828 G_ID: 0.184 G_Rec: 0.447 D_GP: 0.020 D_real: 1.188 D_fake: 0.584 +(epoch: 214, iters: 7296, time: 0.064) G_GAN: 0.336 G_GAN_Feat: 0.796 G_ID: 0.128 G_Rec: 0.318 D_GP: 0.043 D_real: 1.055 D_fake: 0.665 +(epoch: 214, iters: 7696, time: 0.064) G_GAN: 0.422 G_GAN_Feat: 0.804 G_ID: 0.161 G_Rec: 0.361 D_GP: 0.022 D_real: 1.236 D_fake: 0.583 +(epoch: 214, iters: 8096, time: 0.064) G_GAN: 0.326 G_GAN_Feat: 0.655 G_ID: 0.131 G_Rec: 0.283 D_GP: 0.024 D_real: 1.215 D_fake: 0.674 +(epoch: 214, iters: 8496, time: 0.064) G_GAN: 0.534 G_GAN_Feat: 0.957 G_ID: 0.171 G_Rec: 0.403 D_GP: 0.024 D_real: 1.106 D_fake: 0.473 +(epoch: 215, iters: 288, time: 0.064) G_GAN: 0.041 G_GAN_Feat: 0.610 G_ID: 0.129 G_Rec: 0.308 D_GP: 0.020 D_real: 0.972 D_fake: 0.959 +(epoch: 215, iters: 688, time: 0.064) G_GAN: 0.246 G_GAN_Feat: 0.835 G_ID: 0.168 G_Rec: 0.374 D_GP: 0.026 D_real: 0.958 D_fake: 0.755 +(epoch: 215, iters: 1088, time: 0.064) G_GAN: 0.106 G_GAN_Feat: 0.687 G_ID: 0.128 G_Rec: 0.300 D_GP: 0.025 D_real: 1.063 D_fake: 0.894 +(epoch: 215, iters: 1488, time: 0.064) G_GAN: 0.515 G_GAN_Feat: 0.950 G_ID: 0.160 G_Rec: 0.380 D_GP: 0.034 D_real: 0.982 D_fake: 0.489 +(epoch: 215, iters: 1888, time: 0.064) G_GAN: 0.395 G_GAN_Feat: 0.708 G_ID: 0.133 G_Rec: 0.281 D_GP: 0.030 D_real: 1.196 D_fake: 0.605 +(epoch: 215, iters: 2288, time: 0.064) G_GAN: 0.477 G_GAN_Feat: 0.986 G_ID: 0.199 G_Rec: 0.430 D_GP: 0.038 D_real: 0.963 D_fake: 0.534 +(epoch: 215, iters: 2688, time: 0.064) G_GAN: -0.024 G_GAN_Feat: 0.772 G_ID: 0.119 G_Rec: 0.344 D_GP: 0.085 D_real: 0.585 D_fake: 1.024 +(epoch: 215, iters: 3088, time: 0.064) G_GAN: 0.099 G_GAN_Feat: 1.141 G_ID: 0.162 G_Rec: 0.438 D_GP: 0.505 D_real: 0.312 D_fake: 0.902 +(epoch: 215, iters: 3488, time: 0.064) G_GAN: 0.340 G_GAN_Feat: 0.636 G_ID: 0.151 G_Rec: 0.287 D_GP: 0.027 D_real: 1.272 D_fake: 0.661 +(epoch: 215, iters: 3888, time: 0.064) G_GAN: 0.599 G_GAN_Feat: 1.068 G_ID: 0.174 G_Rec: 0.476 D_GP: 0.104 D_real: 0.723 D_fake: 0.412 +(epoch: 215, iters: 4288, time: 0.064) G_GAN: 0.129 G_GAN_Feat: 0.654 G_ID: 0.128 G_Rec: 0.286 D_GP: 0.028 D_real: 1.042 D_fake: 0.872 +(epoch: 215, iters: 4688, time: 0.064) G_GAN: 0.695 G_GAN_Feat: 0.927 G_ID: 0.171 G_Rec: 0.406 D_GP: 0.049 D_real: 1.248 D_fake: 0.323 +(epoch: 215, iters: 5088, time: 0.064) G_GAN: 0.267 G_GAN_Feat: 0.803 G_ID: 0.147 G_Rec: 0.312 D_GP: 0.076 D_real: 0.690 D_fake: 0.741 +(epoch: 215, iters: 5488, time: 0.064) G_GAN: 0.198 G_GAN_Feat: 0.843 G_ID: 0.190 G_Rec: 0.370 D_GP: 0.020 D_real: 1.005 D_fake: 0.802 +(epoch: 215, iters: 5888, time: 0.064) G_GAN: 0.183 G_GAN_Feat: 0.626 G_ID: 0.120 G_Rec: 0.282 D_GP: 0.022 D_real: 1.093 D_fake: 0.817 +(epoch: 215, iters: 6288, time: 0.064) G_GAN: 0.307 G_GAN_Feat: 0.923 G_ID: 0.197 G_Rec: 0.431 D_GP: 0.082 D_real: 0.754 D_fake: 0.693 +(epoch: 215, iters: 6688, time: 0.064) G_GAN: 0.080 G_GAN_Feat: 0.835 G_ID: 0.156 G_Rec: 0.353 D_GP: 0.029 D_real: 0.710 D_fake: 0.921 +(epoch: 215, iters: 7088, time: 0.064) G_GAN: 0.538 G_GAN_Feat: 0.981 G_ID: 0.187 G_Rec: 0.418 D_GP: 0.040 D_real: 0.910 D_fake: 0.483 +(epoch: 215, iters: 7488, time: 0.064) G_GAN: 0.157 G_GAN_Feat: 0.857 G_ID: 0.136 G_Rec: 0.290 D_GP: 0.052 D_real: 0.609 D_fake: 0.843 +(epoch: 215, iters: 7888, time: 0.064) G_GAN: 0.145 G_GAN_Feat: 0.869 G_ID: 0.192 G_Rec: 0.388 D_GP: 0.020 D_real: 0.913 D_fake: 0.855 +(epoch: 215, iters: 8288, time: 0.064) G_GAN: 0.254 G_GAN_Feat: 0.738 G_ID: 0.135 G_Rec: 0.298 D_GP: 0.022 D_real: 1.140 D_fake: 0.746 +(epoch: 216, iters: 80, time: 0.064) G_GAN: 0.150 G_GAN_Feat: 0.889 G_ID: 0.169 G_Rec: 0.349 D_GP: 0.021 D_real: 0.768 D_fake: 0.853 +(epoch: 216, iters: 480, time: 0.064) G_GAN: 0.020 G_GAN_Feat: 0.748 G_ID: 0.114 G_Rec: 0.316 D_GP: 0.028 D_real: 0.859 D_fake: 0.980 +(epoch: 216, iters: 880, time: 0.064) G_GAN: 0.768 G_GAN_Feat: 1.086 G_ID: 0.179 G_Rec: 0.445 D_GP: 0.233 D_real: 0.697 D_fake: 0.405 +(epoch: 216, iters: 1280, time: 0.064) G_GAN: 0.214 G_GAN_Feat: 0.684 G_ID: 0.133 G_Rec: 0.344 D_GP: 0.027 D_real: 1.138 D_fake: 0.791 +(epoch: 216, iters: 1680, time: 0.064) G_GAN: 0.231 G_GAN_Feat: 0.852 G_ID: 0.165 G_Rec: 0.403 D_GP: 0.033 D_real: 0.939 D_fake: 0.769 +(epoch: 216, iters: 2080, time: 0.064) G_GAN: 0.292 G_GAN_Feat: 0.608 G_ID: 0.143 G_Rec: 0.278 D_GP: 0.020 D_real: 1.248 D_fake: 0.709 +(epoch: 216, iters: 2480, time: 0.064) G_GAN: 0.387 G_GAN_Feat: 0.885 G_ID: 0.162 G_Rec: 0.389 D_GP: 0.029 D_real: 1.069 D_fake: 0.615 +(epoch: 216, iters: 2880, time: 0.064) G_GAN: -0.017 G_GAN_Feat: 0.648 G_ID: 0.133 G_Rec: 0.273 D_GP: 0.033 D_real: 0.867 D_fake: 1.017 +(epoch: 216, iters: 3280, time: 0.064) G_GAN: 0.439 G_GAN_Feat: 0.771 G_ID: 0.170 G_Rec: 0.391 D_GP: 0.022 D_real: 1.268 D_fake: 0.562 +(epoch: 216, iters: 3680, time: 0.064) G_GAN: 0.048 G_GAN_Feat: 0.580 G_ID: 0.173 G_Rec: 0.281 D_GP: 0.019 D_real: 1.041 D_fake: 0.952 +(epoch: 216, iters: 4080, time: 0.064) G_GAN: 0.339 G_GAN_Feat: 0.795 G_ID: 0.169 G_Rec: 0.388 D_GP: 0.022 D_real: 1.091 D_fake: 0.661 +(epoch: 216, iters: 4480, time: 0.064) G_GAN: 0.288 G_GAN_Feat: 0.668 G_ID: 0.143 G_Rec: 0.320 D_GP: 0.034 D_real: 1.115 D_fake: 0.714 +(epoch: 216, iters: 4880, time: 0.064) G_GAN: 0.281 G_GAN_Feat: 0.781 G_ID: 0.182 G_Rec: 0.365 D_GP: 0.030 D_real: 0.990 D_fake: 0.720 +(epoch: 216, iters: 5280, time: 0.064) G_GAN: 0.169 G_GAN_Feat: 0.645 G_ID: 0.129 G_Rec: 0.320 D_GP: 0.025 D_real: 1.024 D_fake: 0.831 +(epoch: 216, iters: 5680, time: 0.064) G_GAN: 0.610 G_GAN_Feat: 0.873 G_ID: 0.171 G_Rec: 0.403 D_GP: 0.041 D_real: 1.213 D_fake: 0.421 +(epoch: 216, iters: 6080, time: 0.064) G_GAN: 0.236 G_GAN_Feat: 0.753 G_ID: 0.128 G_Rec: 0.303 D_GP: 0.048 D_real: 0.899 D_fake: 0.765 +(epoch: 216, iters: 6480, time: 0.064) G_GAN: 0.237 G_GAN_Feat: 0.894 G_ID: 0.180 G_Rec: 0.427 D_GP: 0.031 D_real: 0.879 D_fake: 0.765 +(epoch: 216, iters: 6880, time: 0.064) G_GAN: 0.143 G_GAN_Feat: 0.608 G_ID: 0.136 G_Rec: 0.257 D_GP: 0.027 D_real: 1.038 D_fake: 0.857 +(epoch: 216, iters: 7280, time: 0.064) G_GAN: 0.303 G_GAN_Feat: 0.941 G_ID: 0.170 G_Rec: 0.439 D_GP: 0.040 D_real: 0.794 D_fake: 0.698 +(epoch: 216, iters: 7680, time: 0.064) G_GAN: 0.301 G_GAN_Feat: 0.840 G_ID: 0.128 G_Rec: 0.319 D_GP: 0.088 D_real: 0.680 D_fake: 0.712 +(epoch: 216, iters: 8080, time: 0.064) G_GAN: 0.413 G_GAN_Feat: 0.807 G_ID: 0.157 G_Rec: 0.384 D_GP: 0.022 D_real: 1.261 D_fake: 0.588 +(epoch: 216, iters: 8480, time: 0.064) G_GAN: 0.059 G_GAN_Feat: 0.707 G_ID: 0.149 G_Rec: 0.294 D_GP: 0.043 D_real: 0.941 D_fake: 0.941 +(epoch: 217, iters: 272, time: 0.064) G_GAN: 0.276 G_GAN_Feat: 0.879 G_ID: 0.176 G_Rec: 0.406 D_GP: 0.076 D_real: 1.063 D_fake: 0.726 +(epoch: 217, iters: 672, time: 0.064) G_GAN: 0.076 G_GAN_Feat: 0.719 G_ID: 0.122 G_Rec: 0.299 D_GP: 0.035 D_real: 0.935 D_fake: 0.924 +(epoch: 217, iters: 1072, time: 0.064) G_GAN: 0.186 G_GAN_Feat: 0.872 G_ID: 0.183 G_Rec: 0.367 D_GP: 0.030 D_real: 0.785 D_fake: 0.816 +(epoch: 217, iters: 1472, time: 0.064) G_GAN: 0.044 G_GAN_Feat: 0.720 G_ID: 0.166 G_Rec: 0.296 D_GP: 0.022 D_real: 0.942 D_fake: 0.956 +(epoch: 217, iters: 1872, time: 0.064) G_GAN: 0.832 G_GAN_Feat: 0.938 G_ID: 0.183 G_Rec: 0.426 D_GP: 0.024 D_real: 1.501 D_fake: 0.229 +(epoch: 217, iters: 2272, time: 0.064) G_GAN: 0.102 G_GAN_Feat: 0.704 G_ID: 0.120 G_Rec: 0.336 D_GP: 0.020 D_real: 0.933 D_fake: 0.898 +(epoch: 217, iters: 2672, time: 0.064) G_GAN: 0.352 G_GAN_Feat: 0.903 G_ID: 0.151 G_Rec: 0.395 D_GP: 0.047 D_real: 0.892 D_fake: 0.651 +(epoch: 217, iters: 3072, time: 0.064) G_GAN: -0.176 G_GAN_Feat: 0.802 G_ID: 0.139 G_Rec: 0.308 D_GP: 0.061 D_real: 0.617 D_fake: 1.176 +(epoch: 217, iters: 3472, time: 0.064) G_GAN: 0.404 G_GAN_Feat: 0.852 G_ID: 0.201 G_Rec: 0.401 D_GP: 0.024 D_real: 1.210 D_fake: 0.596 +(epoch: 217, iters: 3872, time: 0.064) G_GAN: 0.485 G_GAN_Feat: 0.887 G_ID: 0.194 G_Rec: 0.336 D_GP: 0.110 D_real: 0.807 D_fake: 0.556 +(epoch: 217, iters: 4272, time: 0.064) G_GAN: 0.669 G_GAN_Feat: 0.911 G_ID: 0.175 G_Rec: 0.409 D_GP: 0.091 D_real: 1.361 D_fake: 0.381 +(epoch: 217, iters: 4672, time: 0.064) G_GAN: 0.062 G_GAN_Feat: 0.742 G_ID: 0.168 G_Rec: 0.332 D_GP: 0.057 D_real: 0.836 D_fake: 0.938 +(epoch: 217, iters: 5072, time: 0.064) G_GAN: 0.393 G_GAN_Feat: 0.774 G_ID: 0.181 G_Rec: 0.395 D_GP: 0.022 D_real: 1.257 D_fake: 0.611 +(epoch: 217, iters: 5472, time: 0.064) G_GAN: 0.213 G_GAN_Feat: 0.618 G_ID: 0.148 G_Rec: 0.281 D_GP: 0.022 D_real: 1.129 D_fake: 0.788 +(epoch: 217, iters: 5872, time: 0.064) G_GAN: 0.343 G_GAN_Feat: 0.955 G_ID: 0.194 G_Rec: 0.411 D_GP: 0.037 D_real: 0.718 D_fake: 0.658 +(epoch: 217, iters: 6272, time: 0.064) G_GAN: 0.050 G_GAN_Feat: 0.741 G_ID: 0.159 G_Rec: 0.322 D_GP: 0.025 D_real: 0.896 D_fake: 0.950 +(epoch: 217, iters: 6672, time: 0.064) G_GAN: 0.517 G_GAN_Feat: 0.918 G_ID: 0.172 G_Rec: 0.401 D_GP: 0.031 D_real: 1.166 D_fake: 0.492 +(epoch: 217, iters: 7072, time: 0.064) G_GAN: 0.202 G_GAN_Feat: 0.741 G_ID: 0.119 G_Rec: 0.302 D_GP: 0.048 D_real: 0.969 D_fake: 0.798 +(epoch: 217, iters: 7472, time: 0.064) G_GAN: 0.319 G_GAN_Feat: 0.830 G_ID: 0.212 G_Rec: 0.387 D_GP: 0.028 D_real: 1.066 D_fake: 0.682 +(epoch: 217, iters: 7872, time: 0.064) G_GAN: 0.218 G_GAN_Feat: 0.681 G_ID: 0.142 G_Rec: 0.318 D_GP: 0.022 D_real: 1.124 D_fake: 0.782 +(epoch: 217, iters: 8272, time: 0.064) G_GAN: 0.422 G_GAN_Feat: 1.053 G_ID: 0.183 G_Rec: 0.411 D_GP: 0.218 D_real: 0.554 D_fake: 0.583 +(epoch: 218, iters: 64, time: 0.064) G_GAN: 0.207 G_GAN_Feat: 0.600 G_ID: 0.135 G_Rec: 0.316 D_GP: 0.019 D_real: 1.186 D_fake: 0.793 +(epoch: 218, iters: 464, time: 0.064) G_GAN: 0.620 G_GAN_Feat: 0.807 G_ID: 0.142 G_Rec: 0.398 D_GP: 0.019 D_real: 1.382 D_fake: 0.396 +(epoch: 218, iters: 864, time: 0.064) G_GAN: 0.219 G_GAN_Feat: 0.643 G_ID: 0.119 G_Rec: 0.311 D_GP: 0.028 D_real: 1.129 D_fake: 0.781 +(epoch: 218, iters: 1264, time: 0.064) G_GAN: 0.401 G_GAN_Feat: 0.801 G_ID: 0.219 G_Rec: 0.390 D_GP: 0.031 D_real: 1.148 D_fake: 0.608 +(epoch: 218, iters: 1664, time: 0.064) G_GAN: 0.262 G_GAN_Feat: 0.727 G_ID: 0.140 G_Rec: 0.308 D_GP: 0.033 D_real: 1.009 D_fake: 0.738 +(epoch: 218, iters: 2064, time: 0.064) G_GAN: 0.697 G_GAN_Feat: 0.853 G_ID: 0.166 G_Rec: 0.402 D_GP: 0.025 D_real: 1.369 D_fake: 0.330 +(epoch: 218, iters: 2464, time: 0.064) G_GAN: 0.321 G_GAN_Feat: 0.725 G_ID: 0.151 G_Rec: 0.310 D_GP: 0.029 D_real: 1.161 D_fake: 0.682 +(epoch: 218, iters: 2864, time: 0.064) G_GAN: 0.354 G_GAN_Feat: 0.974 G_ID: 0.191 G_Rec: 0.417 D_GP: 0.028 D_real: 0.820 D_fake: 0.647 +(epoch: 218, iters: 3264, time: 0.064) G_GAN: -0.050 G_GAN_Feat: 0.750 G_ID: 0.125 G_Rec: 0.299 D_GP: 0.026 D_real: 0.791 D_fake: 1.050 +(epoch: 218, iters: 3664, time: 0.064) G_GAN: 0.199 G_GAN_Feat: 0.806 G_ID: 0.212 G_Rec: 0.392 D_GP: 0.027 D_real: 1.033 D_fake: 0.801 +(epoch: 218, iters: 4064, time: 0.064) G_GAN: 0.108 G_GAN_Feat: 0.661 G_ID: 0.132 G_Rec: 0.311 D_GP: 0.028 D_real: 1.012 D_fake: 0.893 +(epoch: 218, iters: 4464, time: 0.064) G_GAN: 0.336 G_GAN_Feat: 0.781 G_ID: 0.186 G_Rec: 0.370 D_GP: 0.020 D_real: 1.147 D_fake: 0.664 +(epoch: 218, iters: 4864, time: 0.064) G_GAN: 0.256 G_GAN_Feat: 0.610 G_ID: 0.136 G_Rec: 0.263 D_GP: 0.021 D_real: 1.173 D_fake: 0.745 +(epoch: 218, iters: 5264, time: 0.064) G_GAN: 0.797 G_GAN_Feat: 1.289 G_ID: 0.185 G_Rec: 0.441 D_GP: 0.033 D_real: 1.469 D_fake: 0.291 +(epoch: 218, iters: 5664, time: 0.064) G_GAN: 0.008 G_GAN_Feat: 0.646 G_ID: 0.141 G_Rec: 0.318 D_GP: 0.018 D_real: 0.958 D_fake: 0.992 +(epoch: 218, iters: 6064, time: 0.064) G_GAN: 0.079 G_GAN_Feat: 0.818 G_ID: 0.199 G_Rec: 0.394 D_GP: 0.030 D_real: 0.848 D_fake: 0.922 +(epoch: 218, iters: 6464, time: 0.064) G_GAN: -0.068 G_GAN_Feat: 0.669 G_ID: 0.140 G_Rec: 0.327 D_GP: 0.028 D_real: 0.804 D_fake: 1.069 +(epoch: 218, iters: 6864, time: 0.064) G_GAN: 0.413 G_GAN_Feat: 0.935 G_ID: 0.177 G_Rec: 0.424 D_GP: 0.060 D_real: 0.922 D_fake: 0.589 +(epoch: 218, iters: 7264, time: 0.064) G_GAN: 0.287 G_GAN_Feat: 0.680 G_ID: 0.162 G_Rec: 0.271 D_GP: 0.024 D_real: 1.147 D_fake: 0.714 +(epoch: 218, iters: 7664, time: 0.064) G_GAN: 0.427 G_GAN_Feat: 0.938 G_ID: 0.183 G_Rec: 0.411 D_GP: 0.026 D_real: 1.200 D_fake: 0.575 +(epoch: 218, iters: 8064, time: 0.064) G_GAN: 0.178 G_GAN_Feat: 0.722 G_ID: 0.145 G_Rec: 0.325 D_GP: 0.037 D_real: 1.023 D_fake: 0.825 +(epoch: 218, iters: 8464, time: 0.064) G_GAN: 0.383 G_GAN_Feat: 0.876 G_ID: 0.163 G_Rec: 0.404 D_GP: 0.022 D_real: 1.065 D_fake: 0.617 +(epoch: 219, iters: 256, time: 0.064) G_GAN: 0.448 G_GAN_Feat: 0.846 G_ID: 0.152 G_Rec: 0.319 D_GP: 0.097 D_real: 0.847 D_fake: 0.639 +(epoch: 219, iters: 656, time: 0.064) G_GAN: 0.644 G_GAN_Feat: 1.005 G_ID: 0.158 G_Rec: 0.409 D_GP: 0.037 D_real: 1.431 D_fake: 0.371 +(epoch: 219, iters: 1056, time: 0.064) G_GAN: 0.223 G_GAN_Feat: 0.675 G_ID: 0.152 G_Rec: 0.319 D_GP: 0.018 D_real: 1.131 D_fake: 0.779 +(epoch: 219, iters: 1456, time: 0.064) G_GAN: 0.466 G_GAN_Feat: 0.903 G_ID: 0.149 G_Rec: 0.382 D_GP: 0.046 D_real: 1.016 D_fake: 0.542 +(epoch: 219, iters: 1856, time: 0.064) G_GAN: -0.294 G_GAN_Feat: 0.723 G_ID: 0.126 G_Rec: 0.299 D_GP: 0.042 D_real: 0.592 D_fake: 1.294 +(epoch: 219, iters: 2256, time: 0.064) G_GAN: 0.390 G_GAN_Feat: 1.016 G_ID: 0.183 G_Rec: 0.458 D_GP: 0.067 D_real: 0.621 D_fake: 0.614 +(epoch: 219, iters: 2656, time: 0.064) G_GAN: 0.084 G_GAN_Feat: 0.715 G_ID: 0.142 G_Rec: 0.313 D_GP: 0.029 D_real: 1.025 D_fake: 0.919 +(epoch: 219, iters: 3056, time: 0.064) G_GAN: 0.350 G_GAN_Feat: 0.916 G_ID: 0.159 G_Rec: 0.394 D_GP: 0.027 D_real: 1.070 D_fake: 0.654 +(epoch: 219, iters: 3456, time: 0.064) G_GAN: 0.361 G_GAN_Feat: 0.732 G_ID: 0.130 G_Rec: 0.295 D_GP: 0.038 D_real: 1.120 D_fake: 0.640 +(epoch: 219, iters: 3856, time: 0.064) G_GAN: 0.232 G_GAN_Feat: 0.899 G_ID: 0.163 G_Rec: 0.387 D_GP: 0.042 D_real: 0.922 D_fake: 0.771 +(epoch: 219, iters: 4256, time: 0.064) G_GAN: 0.369 G_GAN_Feat: 0.799 G_ID: 0.143 G_Rec: 0.326 D_GP: 0.040 D_real: 1.075 D_fake: 0.725 +(epoch: 219, iters: 4656, time: 0.064) G_GAN: 0.433 G_GAN_Feat: 1.211 G_ID: 0.200 G_Rec: 0.436 D_GP: 0.584 D_real: 0.359 D_fake: 0.577 +(epoch: 219, iters: 5056, time: 0.064) G_GAN: 0.179 G_GAN_Feat: 0.715 G_ID: 0.134 G_Rec: 0.290 D_GP: 0.028 D_real: 1.048 D_fake: 0.822 +(epoch: 219, iters: 5456, time: 0.064) G_GAN: 0.363 G_GAN_Feat: 0.884 G_ID: 0.196 G_Rec: 0.408 D_GP: 0.021 D_real: 1.036 D_fake: 0.637 +(epoch: 219, iters: 5856, time: 0.064) G_GAN: 0.318 G_GAN_Feat: 0.751 G_ID: 0.126 G_Rec: 0.293 D_GP: 0.025 D_real: 1.111 D_fake: 0.682 +(epoch: 219, iters: 6256, time: 0.064) G_GAN: 0.436 G_GAN_Feat: 0.905 G_ID: 0.167 G_Rec: 0.395 D_GP: 0.025 D_real: 1.149 D_fake: 0.572 +(epoch: 219, iters: 6656, time: 0.064) G_GAN: 0.269 G_GAN_Feat: 0.621 G_ID: 0.148 G_Rec: 0.275 D_GP: 0.023 D_real: 1.264 D_fake: 0.731 +(epoch: 219, iters: 7056, time: 0.064) G_GAN: 0.237 G_GAN_Feat: 0.858 G_ID: 0.194 G_Rec: 0.372 D_GP: 0.029 D_real: 0.947 D_fake: 0.763 +(epoch: 219, iters: 7456, time: 0.064) G_GAN: 0.202 G_GAN_Feat: 0.845 G_ID: 0.146 G_Rec: 0.355 D_GP: 0.077 D_real: 0.733 D_fake: 0.798 +(epoch: 219, iters: 7856, time: 0.064) G_GAN: 0.418 G_GAN_Feat: 1.020 G_ID: 0.204 G_Rec: 0.439 D_GP: 0.258 D_real: 0.634 D_fake: 0.591 +(epoch: 219, iters: 8256, time: 0.064) G_GAN: 0.264 G_GAN_Feat: 0.737 G_ID: 0.114 G_Rec: 0.282 D_GP: 0.035 D_real: 1.148 D_fake: 0.736 +(epoch: 220, iters: 48, time: 0.064) G_GAN: 0.340 G_GAN_Feat: 0.929 G_ID: 0.187 G_Rec: 0.410 D_GP: 0.048 D_real: 0.815 D_fake: 0.665 +(epoch: 220, iters: 448, time: 0.064) G_GAN: 0.085 G_GAN_Feat: 0.884 G_ID: 0.134 G_Rec: 0.330 D_GP: 0.062 D_real: 0.493 D_fake: 0.915 +(epoch: 220, iters: 848, time: 0.064) G_GAN: 0.207 G_GAN_Feat: 1.007 G_ID: 0.208 G_Rec: 0.390 D_GP: 0.043 D_real: 0.417 D_fake: 0.794 +(epoch: 220, iters: 1248, time: 0.064) G_GAN: 0.254 G_GAN_Feat: 0.695 G_ID: 0.171 G_Rec: 0.279 D_GP: 0.026 D_real: 1.197 D_fake: 0.746 +(epoch: 220, iters: 1648, time: 0.064) G_GAN: 0.534 G_GAN_Feat: 0.954 G_ID: 0.160 G_Rec: 0.423 D_GP: 0.041 D_real: 1.161 D_fake: 0.472 +(epoch: 220, iters: 2048, time: 0.064) G_GAN: 0.179 G_GAN_Feat: 0.739 G_ID: 0.134 G_Rec: 0.341 D_GP: 0.023 D_real: 1.026 D_fake: 0.821 +(epoch: 220, iters: 2448, time: 0.064) G_GAN: 0.556 G_GAN_Feat: 0.940 G_ID: 0.180 G_Rec: 0.398 D_GP: 0.025 D_real: 1.193 D_fake: 0.449 +(epoch: 220, iters: 2848, time: 0.064) G_GAN: 0.059 G_GAN_Feat: 0.692 G_ID: 0.126 G_Rec: 0.307 D_GP: 0.050 D_real: 0.904 D_fake: 0.942 +(epoch: 220, iters: 3248, time: 0.064) G_GAN: 0.370 G_GAN_Feat: 0.749 G_ID: 0.175 G_Rec: 0.375 D_GP: 0.019 D_real: 1.188 D_fake: 0.631 +(epoch: 220, iters: 3648, time: 0.064) G_GAN: 0.195 G_GAN_Feat: 0.640 G_ID: 0.145 G_Rec: 0.326 D_GP: 0.028 D_real: 1.084 D_fake: 0.805 +(epoch: 220, iters: 4048, time: 0.064) G_GAN: 0.326 G_GAN_Feat: 0.799 G_ID: 0.194 G_Rec: 0.387 D_GP: 0.031 D_real: 1.038 D_fake: 0.676 +(epoch: 220, iters: 4448, time: 0.064) G_GAN: 0.166 G_GAN_Feat: 0.638 G_ID: 0.125 G_Rec: 0.263 D_GP: 0.039 D_real: 0.998 D_fake: 0.834 +(epoch: 220, iters: 4848, time: 0.064) G_GAN: 0.319 G_GAN_Feat: 0.932 G_ID: 0.177 G_Rec: 0.422 D_GP: 0.063 D_real: 0.866 D_fake: 0.683 +(epoch: 220, iters: 5248, time: 0.064) G_GAN: 0.181 G_GAN_Feat: 0.809 G_ID: 0.133 G_Rec: 0.324 D_GP: 0.082 D_real: 0.717 D_fake: 0.819 +(epoch: 220, iters: 5648, time: 0.064) G_GAN: 0.350 G_GAN_Feat: 0.996 G_ID: 0.190 G_Rec: 0.454 D_GP: 0.126 D_real: 0.699 D_fake: 0.658 +(epoch: 220, iters: 6048, time: 0.064) G_GAN: 0.377 G_GAN_Feat: 0.875 G_ID: 0.139 G_Rec: 0.342 D_GP: 0.028 D_real: 1.088 D_fake: 0.623 +(epoch: 220, iters: 6448, time: 0.064) G_GAN: 0.441 G_GAN_Feat: 0.832 G_ID: 0.205 G_Rec: 0.348 D_GP: 0.036 D_real: 1.212 D_fake: 0.562 +(epoch: 220, iters: 6848, time: 0.064) G_GAN: -0.083 G_GAN_Feat: 0.882 G_ID: 0.195 G_Rec: 0.318 D_GP: 0.225 D_real: 0.294 D_fake: 1.084 +(epoch: 220, iters: 7248, time: 0.064) G_GAN: 0.210 G_GAN_Feat: 1.002 G_ID: 0.189 G_Rec: 0.388 D_GP: 0.031 D_real: 0.737 D_fake: 0.795 +(epoch: 220, iters: 7648, time: 0.064) G_GAN: 0.768 G_GAN_Feat: 1.046 G_ID: 0.151 G_Rec: 0.345 D_GP: 0.057 D_real: 1.100 D_fake: 0.655 +(epoch: 220, iters: 8048, time: 0.064) G_GAN: 0.443 G_GAN_Feat: 0.997 G_ID: 0.165 G_Rec: 0.442 D_GP: 0.022 D_real: 1.171 D_fake: 0.558 +(epoch: 220, iters: 8448, time: 0.064) G_GAN: 0.079 G_GAN_Feat: 0.786 G_ID: 0.122 G_Rec: 0.334 D_GP: 0.084 D_real: 0.678 D_fake: 0.921 +(epoch: 221, iters: 240, time: 0.064) G_GAN: 0.629 G_GAN_Feat: 0.899 G_ID: 0.164 G_Rec: 0.390 D_GP: 0.043 D_real: 1.293 D_fake: 0.383 +(epoch: 221, iters: 640, time: 0.064) G_GAN: 0.165 G_GAN_Feat: 0.627 G_ID: 0.146 G_Rec: 0.309 D_GP: 0.023 D_real: 1.128 D_fake: 0.835 +(epoch: 221, iters: 1040, time: 0.064) G_GAN: 0.541 G_GAN_Feat: 0.825 G_ID: 0.162 G_Rec: 0.362 D_GP: 0.026 D_real: 1.243 D_fake: 0.480 +(epoch: 221, iters: 1440, time: 0.064) G_GAN: 0.420 G_GAN_Feat: 0.643 G_ID: 0.133 G_Rec: 0.277 D_GP: 0.020 D_real: 1.365 D_fake: 0.582 +(epoch: 221, iters: 1840, time: 0.064) G_GAN: 0.425 G_GAN_Feat: 1.092 G_ID: 0.159 G_Rec: 0.468 D_GP: 0.046 D_real: 0.557 D_fake: 0.580 +(epoch: 221, iters: 2240, time: 0.064) G_GAN: -0.043 G_GAN_Feat: 0.687 G_ID: 0.161 G_Rec: 0.314 D_GP: 0.028 D_real: 0.801 D_fake: 1.044 +(epoch: 221, iters: 2640, time: 0.064) G_GAN: 0.381 G_GAN_Feat: 0.937 G_ID: 0.181 G_Rec: 0.439 D_GP: 0.033 D_real: 0.997 D_fake: 0.626 +(epoch: 221, iters: 3040, time: 0.064) G_GAN: 0.316 G_GAN_Feat: 0.750 G_ID: 0.139 G_Rec: 0.339 D_GP: 0.032 D_real: 1.172 D_fake: 0.686 +(epoch: 221, iters: 3440, time: 0.064) G_GAN: 0.490 G_GAN_Feat: 0.882 G_ID: 0.164 G_Rec: 0.396 D_GP: 0.023 D_real: 1.219 D_fake: 0.511 +(epoch: 221, iters: 3840, time: 0.064) G_GAN: 0.284 G_GAN_Feat: 0.642 G_ID: 0.142 G_Rec: 0.292 D_GP: 0.022 D_real: 1.221 D_fake: 0.717 +(epoch: 221, iters: 4240, time: 0.064) G_GAN: 0.539 G_GAN_Feat: 0.863 G_ID: 0.152 G_Rec: 0.368 D_GP: 0.024 D_real: 1.302 D_fake: 0.463 +(epoch: 221, iters: 4640, time: 0.064) G_GAN: 0.218 G_GAN_Feat: 0.537 G_ID: 0.130 G_Rec: 0.299 D_GP: 0.018 D_real: 1.211 D_fake: 0.782 +(epoch: 221, iters: 5040, time: 0.064) G_GAN: 0.276 G_GAN_Feat: 0.713 G_ID: 0.180 G_Rec: 0.393 D_GP: 0.020 D_real: 1.145 D_fake: 0.725 +(epoch: 221, iters: 5440, time: 0.064) G_GAN: 0.171 G_GAN_Feat: 0.569 G_ID: 0.136 G_Rec: 0.315 D_GP: 0.017 D_real: 1.119 D_fake: 0.829 +(epoch: 221, iters: 5840, time: 0.064) G_GAN: 0.394 G_GAN_Feat: 0.775 G_ID: 0.170 G_Rec: 0.455 D_GP: 0.022 D_real: 1.142 D_fake: 0.608 +(epoch: 221, iters: 6240, time: 0.064) G_GAN: 0.115 G_GAN_Feat: 0.589 G_ID: 0.156 G_Rec: 0.317 D_GP: 0.027 D_real: 1.019 D_fake: 0.885 +(epoch: 221, iters: 6640, time: 0.064) G_GAN: 0.191 G_GAN_Feat: 0.717 G_ID: 0.194 G_Rec: 0.381 D_GP: 0.026 D_real: 0.997 D_fake: 0.809 +(epoch: 221, iters: 7040, time: 0.064) G_GAN: -0.034 G_GAN_Feat: 0.538 G_ID: 0.132 G_Rec: 0.306 D_GP: 0.022 D_real: 0.939 D_fake: 1.034 +(epoch: 221, iters: 7440, time: 0.064) G_GAN: 0.111 G_GAN_Feat: 0.725 G_ID: 0.177 G_Rec: 0.401 D_GP: 0.027 D_real: 0.860 D_fake: 0.889 +(epoch: 221, iters: 7840, time: 0.064) G_GAN: 0.016 G_GAN_Feat: 0.627 G_ID: 0.175 G_Rec: 0.306 D_GP: 0.038 D_real: 0.883 D_fake: 0.984 +(epoch: 221, iters: 8240, time: 0.064) G_GAN: 0.134 G_GAN_Feat: 0.757 G_ID: 0.195 G_Rec: 0.365 D_GP: 0.037 D_real: 0.853 D_fake: 0.867 +(epoch: 222, iters: 32, time: 0.064) G_GAN: 0.098 G_GAN_Feat: 0.610 G_ID: 0.119 G_Rec: 0.299 D_GP: 0.027 D_real: 1.126 D_fake: 0.905 +(epoch: 222, iters: 432, time: 0.064) G_GAN: 0.120 G_GAN_Feat: 0.787 G_ID: 0.190 G_Rec: 0.408 D_GP: 0.025 D_real: 0.915 D_fake: 0.881 +(epoch: 222, iters: 832, time: 0.064) G_GAN: 0.057 G_GAN_Feat: 0.581 G_ID: 0.115 G_Rec: 0.268 D_GP: 0.026 D_real: 1.005 D_fake: 0.943 +(epoch: 222, iters: 1232, time: 0.064) G_GAN: 0.426 G_GAN_Feat: 0.740 G_ID: 0.163 G_Rec: 0.367 D_GP: 0.027 D_real: 1.248 D_fake: 0.580 +(epoch: 222, iters: 1632, time: 0.064) G_GAN: 0.073 G_GAN_Feat: 0.611 G_ID: 0.153 G_Rec: 0.308 D_GP: 0.039 D_real: 0.946 D_fake: 0.927 +(epoch: 222, iters: 2032, time: 0.064) G_GAN: 0.284 G_GAN_Feat: 0.801 G_ID: 0.170 G_Rec: 0.402 D_GP: 0.023 D_real: 1.066 D_fake: 0.718 +(epoch: 222, iters: 2432, time: 0.064) G_GAN: 0.022 G_GAN_Feat: 0.689 G_ID: 0.129 G_Rec: 0.293 D_GP: 0.039 D_real: 0.831 D_fake: 0.978 +(epoch: 222, iters: 2832, time: 0.064) G_GAN: 0.228 G_GAN_Feat: 0.814 G_ID: 0.189 G_Rec: 0.372 D_GP: 0.037 D_real: 0.927 D_fake: 0.772 +(epoch: 222, iters: 3232, time: 0.064) G_GAN: -0.181 G_GAN_Feat: 0.711 G_ID: 0.165 G_Rec: 0.336 D_GP: 0.035 D_real: 0.814 D_fake: 1.182 +(epoch: 222, iters: 3632, time: 0.064) G_GAN: 0.345 G_GAN_Feat: 0.790 G_ID: 0.180 G_Rec: 0.400 D_GP: 0.025 D_real: 1.082 D_fake: 0.656 +(epoch: 222, iters: 4032, time: 0.064) G_GAN: 0.325 G_GAN_Feat: 0.601 G_ID: 0.121 G_Rec: 0.288 D_GP: 0.023 D_real: 1.273 D_fake: 0.675 +(epoch: 222, iters: 4432, time: 0.064) G_GAN: 0.463 G_GAN_Feat: 0.831 G_ID: 0.173 G_Rec: 0.461 D_GP: 0.042 D_real: 1.090 D_fake: 0.543 +(epoch: 222, iters: 4832, time: 0.064) G_GAN: 0.196 G_GAN_Feat: 0.553 G_ID: 0.122 G_Rec: 0.261 D_GP: 0.022 D_real: 1.127 D_fake: 0.804 +(epoch: 222, iters: 5232, time: 0.064) G_GAN: 0.445 G_GAN_Feat: 1.118 G_ID: 0.182 G_Rec: 0.512 D_GP: 1.347 D_real: 0.623 D_fake: 0.578 +(epoch: 222, iters: 5632, time: 0.064) G_GAN: 0.241 G_GAN_Feat: 0.577 G_ID: 0.129 G_Rec: 0.279 D_GP: 0.021 D_real: 1.176 D_fake: 0.759 +(epoch: 222, iters: 6032, time: 0.064) G_GAN: 0.504 G_GAN_Feat: 0.838 G_ID: 0.154 G_Rec: 0.382 D_GP: 0.041 D_real: 1.107 D_fake: 0.498 +(epoch: 222, iters: 6432, time: 0.064) G_GAN: 0.205 G_GAN_Feat: 0.650 G_ID: 0.128 G_Rec: 0.291 D_GP: 0.031 D_real: 1.142 D_fake: 0.795 +(epoch: 222, iters: 6832, time: 0.064) G_GAN: 0.260 G_GAN_Feat: 0.895 G_ID: 0.184 G_Rec: 0.432 D_GP: 0.053 D_real: 0.800 D_fake: 0.740 +(epoch: 222, iters: 7232, time: 0.064) G_GAN: 0.197 G_GAN_Feat: 0.618 G_ID: 0.140 G_Rec: 0.290 D_GP: 0.022 D_real: 1.065 D_fake: 0.803 +(epoch: 222, iters: 7632, time: 0.064) G_GAN: 0.442 G_GAN_Feat: 0.773 G_ID: 0.175 G_Rec: 0.380 D_GP: 0.026 D_real: 1.204 D_fake: 0.559 +(epoch: 222, iters: 8032, time: 0.064) G_GAN: 0.367 G_GAN_Feat: 0.591 G_ID: 0.120 G_Rec: 0.277 D_GP: 0.022 D_real: 1.326 D_fake: 0.635 +(epoch: 222, iters: 8432, time: 0.064) G_GAN: 0.508 G_GAN_Feat: 0.895 G_ID: 0.199 G_Rec: 0.370 D_GP: 0.062 D_real: 0.908 D_fake: 0.496 +(epoch: 223, iters: 224, time: 0.064) G_GAN: 0.183 G_GAN_Feat: 0.676 G_ID: 0.141 G_Rec: 0.301 D_GP: 0.032 D_real: 0.958 D_fake: 0.817 +(epoch: 223, iters: 624, time: 0.064) G_GAN: 0.431 G_GAN_Feat: 0.817 G_ID: 0.179 G_Rec: 0.460 D_GP: 0.020 D_real: 1.241 D_fake: 0.570 +(epoch: 223, iters: 1024, time: 0.064) G_GAN: 0.159 G_GAN_Feat: 0.730 G_ID: 0.166 G_Rec: 0.315 D_GP: 0.035 D_real: 0.971 D_fake: 0.841 +(epoch: 223, iters: 1424, time: 0.064) G_GAN: 0.503 G_GAN_Feat: 0.811 G_ID: 0.169 G_Rec: 0.450 D_GP: 0.024 D_real: 1.189 D_fake: 0.506 +(epoch: 223, iters: 1824, time: 0.064) G_GAN: -0.041 G_GAN_Feat: 0.662 G_ID: 0.156 G_Rec: 0.301 D_GP: 0.042 D_real: 0.776 D_fake: 1.041 +(epoch: 223, iters: 2224, time: 0.064) G_GAN: 0.299 G_GAN_Feat: 0.940 G_ID: 0.166 G_Rec: 0.431 D_GP: 0.035 D_real: 0.898 D_fake: 0.702 +(epoch: 223, iters: 2624, time: 0.064) G_GAN: 0.079 G_GAN_Feat: 0.731 G_ID: 0.126 G_Rec: 0.333 D_GP: 0.101 D_real: 0.770 D_fake: 0.921 +(epoch: 223, iters: 3024, time: 0.064) G_GAN: 0.572 G_GAN_Feat: 0.841 G_ID: 0.170 G_Rec: 0.382 D_GP: 0.024 D_real: 1.228 D_fake: 0.452 +(epoch: 223, iters: 3424, time: 0.064) G_GAN: 0.203 G_GAN_Feat: 0.755 G_ID: 0.141 G_Rec: 0.324 D_GP: 0.058 D_real: 0.958 D_fake: 0.798 +(epoch: 223, iters: 3824, time: 0.064) G_GAN: 0.568 G_GAN_Feat: 1.018 G_ID: 0.163 G_Rec: 0.425 D_GP: 0.157 D_real: 0.682 D_fake: 0.440 +(epoch: 223, iters: 4224, time: 0.064) G_GAN: 0.172 G_GAN_Feat: 0.702 G_ID: 0.144 G_Rec: 0.374 D_GP: 0.029 D_real: 1.098 D_fake: 0.829 +(epoch: 223, iters: 4624, time: 0.064) G_GAN: 0.586 G_GAN_Feat: 0.838 G_ID: 0.168 G_Rec: 0.390 D_GP: 0.020 D_real: 1.330 D_fake: 0.419 +(epoch: 223, iters: 5024, time: 0.064) G_GAN: 0.074 G_GAN_Feat: 0.718 G_ID: 0.134 G_Rec: 0.301 D_GP: 0.057 D_real: 0.790 D_fake: 0.926 +(epoch: 223, iters: 5424, time: 0.064) G_GAN: 0.566 G_GAN_Feat: 0.883 G_ID: 0.194 G_Rec: 0.404 D_GP: 0.025 D_real: 1.245 D_fake: 0.440 +(epoch: 223, iters: 5824, time: 0.064) G_GAN: 0.571 G_GAN_Feat: 0.788 G_ID: 0.136 G_Rec: 0.321 D_GP: 0.036 D_real: 1.409 D_fake: 0.539 +(epoch: 223, iters: 6224, time: 0.064) G_GAN: 0.126 G_GAN_Feat: 0.952 G_ID: 0.160 G_Rec: 0.373 D_GP: 0.161 D_real: 0.397 D_fake: 0.874 +(epoch: 223, iters: 6624, time: 0.064) G_GAN: 0.271 G_GAN_Feat: 0.729 G_ID: 0.126 G_Rec: 0.357 D_GP: 0.033 D_real: 1.132 D_fake: 0.729 +(epoch: 223, iters: 7024, time: 0.064) G_GAN: 0.251 G_GAN_Feat: 0.764 G_ID: 0.194 G_Rec: 0.421 D_GP: 0.020 D_real: 1.110 D_fake: 0.750 +(epoch: 223, iters: 7424, time: 0.064) G_GAN: 0.093 G_GAN_Feat: 0.836 G_ID: 0.134 G_Rec: 0.356 D_GP: 0.021 D_real: 0.933 D_fake: 0.907 +(epoch: 223, iters: 7824, time: 0.064) G_GAN: 0.025 G_GAN_Feat: 0.752 G_ID: 0.163 G_Rec: 0.359 D_GP: 0.020 D_real: 0.878 D_fake: 0.978 +(epoch: 223, iters: 8224, time: 0.064) G_GAN: 0.024 G_GAN_Feat: 0.654 G_ID: 0.138 G_Rec: 0.301 D_GP: 0.030 D_real: 0.956 D_fake: 0.976 +(epoch: 224, iters: 16, time: 0.064) G_GAN: 0.408 G_GAN_Feat: 1.031 G_ID: 0.151 G_Rec: 0.429 D_GP: 0.085 D_real: 0.633 D_fake: 0.595 +(epoch: 224, iters: 416, time: 0.064) G_GAN: 0.390 G_GAN_Feat: 0.587 G_ID: 0.124 G_Rec: 0.251 D_GP: 0.025 D_real: 1.350 D_fake: 0.614 +(epoch: 224, iters: 816, time: 0.064) G_GAN: 0.241 G_GAN_Feat: 0.698 G_ID: 0.172 G_Rec: 0.334 D_GP: 0.021 D_real: 1.019 D_fake: 0.759 +(epoch: 224, iters: 1216, time: 0.064) G_GAN: 0.036 G_GAN_Feat: 0.628 G_ID: 0.143 G_Rec: 0.295 D_GP: 0.024 D_real: 0.940 D_fake: 0.965 +(epoch: 224, iters: 1616, time: 0.064) G_GAN: 0.114 G_GAN_Feat: 0.831 G_ID: 0.170 G_Rec: 0.377 D_GP: 0.035 D_real: 0.826 D_fake: 0.886 +(epoch: 224, iters: 2016, time: 0.064) G_GAN: -0.224 G_GAN_Feat: 0.790 G_ID: 0.118 G_Rec: 0.325 D_GP: 0.100 D_real: 0.361 D_fake: 1.224 +(epoch: 224, iters: 2416, time: 0.064) G_GAN: 0.515 G_GAN_Feat: 0.917 G_ID: 0.164 G_Rec: 0.399 D_GP: 0.088 D_real: 0.913 D_fake: 0.489 +(epoch: 224, iters: 2816, time: 0.064) G_GAN: 0.068 G_GAN_Feat: 0.905 G_ID: 0.122 G_Rec: 0.328 D_GP: 0.085 D_real: 0.513 D_fake: 0.933 +(epoch: 224, iters: 3216, time: 0.064) G_GAN: 0.316 G_GAN_Feat: 0.926 G_ID: 0.195 G_Rec: 0.401 D_GP: 0.037 D_real: 1.107 D_fake: 0.684 +(epoch: 224, iters: 3616, time: 0.064) G_GAN: 0.256 G_GAN_Feat: 0.707 G_ID: 0.154 G_Rec: 0.300 D_GP: 0.034 D_real: 1.044 D_fake: 0.744 +(epoch: 224, iters: 4016, time: 0.064) G_GAN: 0.694 G_GAN_Feat: 0.831 G_ID: 0.173 G_Rec: 0.394 D_GP: 0.023 D_real: 1.421 D_fake: 0.330 +(epoch: 224, iters: 4416, time: 0.064) G_GAN: 0.398 G_GAN_Feat: 0.743 G_ID: 0.130 G_Rec: 0.288 D_GP: 0.041 D_real: 1.111 D_fake: 0.605 +(epoch: 224, iters: 4816, time: 0.064) G_GAN: 0.566 G_GAN_Feat: 0.797 G_ID: 0.167 G_Rec: 0.366 D_GP: 0.025 D_real: 1.356 D_fake: 0.441 +(epoch: 224, iters: 5216, time: 0.064) G_GAN: 0.349 G_GAN_Feat: 0.682 G_ID: 0.133 G_Rec: 0.298 D_GP: 0.030 D_real: 1.239 D_fake: 0.652 +(epoch: 224, iters: 5616, time: 0.064) G_GAN: 0.833 G_GAN_Feat: 0.994 G_ID: 0.173 G_Rec: 0.440 D_GP: 0.071 D_real: 1.141 D_fake: 0.389 +(epoch: 224, iters: 6016, time: 0.064) G_GAN: 0.326 G_GAN_Feat: 0.909 G_ID: 0.134 G_Rec: 0.354 D_GP: 0.140 D_real: 0.478 D_fake: 0.684 +(epoch: 224, iters: 6416, time: 0.064) G_GAN: 0.401 G_GAN_Feat: 1.206 G_ID: 0.179 G_Rec: 0.435 D_GP: 0.369 D_real: 0.270 D_fake: 0.611 +(epoch: 224, iters: 6816, time: 0.064) G_GAN: 0.332 G_GAN_Feat: 0.877 G_ID: 0.118 G_Rec: 0.308 D_GP: 0.050 D_real: 1.008 D_fake: 0.726 +(epoch: 224, iters: 7216, time: 0.064) G_GAN: 0.607 G_GAN_Feat: 0.878 G_ID: 0.186 G_Rec: 0.412 D_GP: 0.020 D_real: 1.313 D_fake: 0.403 +(epoch: 224, iters: 7616, time: 0.064) G_GAN: 0.248 G_GAN_Feat: 0.643 G_ID: 0.152 G_Rec: 0.287 D_GP: 0.023 D_real: 1.190 D_fake: 0.752 +(epoch: 224, iters: 8016, time: 0.064) G_GAN: 0.299 G_GAN_Feat: 1.062 G_ID: 0.188 G_Rec: 0.451 D_GP: 0.315 D_real: 0.377 D_fake: 0.707 +(epoch: 224, iters: 8416, time: 0.064) G_GAN: 0.271 G_GAN_Feat: 0.744 G_ID: 0.127 G_Rec: 0.299 D_GP: 0.030 D_real: 1.035 D_fake: 0.730 +(epoch: 225, iters: 208, time: 0.064) G_GAN: 0.346 G_GAN_Feat: 0.946 G_ID: 0.174 G_Rec: 0.413 D_GP: 0.057 D_real: 0.889 D_fake: 0.661 +(epoch: 225, iters: 608, time: 0.064) G_GAN: -0.008 G_GAN_Feat: 0.724 G_ID: 0.149 G_Rec: 0.306 D_GP: 0.021 D_real: 0.926 D_fake: 1.008 +(epoch: 225, iters: 1008, time: 0.064) G_GAN: 0.515 G_GAN_Feat: 0.847 G_ID: 0.162 G_Rec: 0.368 D_GP: 0.034 D_real: 1.212 D_fake: 0.495 +(epoch: 225, iters: 1408, time: 0.064) G_GAN: 0.405 G_GAN_Feat: 0.655 G_ID: 0.138 G_Rec: 0.294 D_GP: 0.020 D_real: 1.290 D_fake: 0.655 +(epoch: 225, iters: 1808, time: 0.064) G_GAN: 0.104 G_GAN_Feat: 0.838 G_ID: 0.173 G_Rec: 0.373 D_GP: 0.021 D_real: 0.885 D_fake: 0.896 +(epoch: 225, iters: 2208, time: 0.064) G_GAN: 0.388 G_GAN_Feat: 0.876 G_ID: 0.156 G_Rec: 0.325 D_GP: 0.098 D_real: 0.539 D_fake: 0.618 +(epoch: 225, iters: 2608, time: 0.064) G_GAN: 0.229 G_GAN_Feat: 0.822 G_ID: 0.165 G_Rec: 0.411 D_GP: 0.024 D_real: 0.964 D_fake: 0.771 +(epoch: 225, iters: 3008, time: 0.064) G_GAN: 0.217 G_GAN_Feat: 0.631 G_ID: 0.121 G_Rec: 0.295 D_GP: 0.023 D_real: 1.177 D_fake: 0.783 +(epoch: 225, iters: 3408, time: 0.064) G_GAN: 0.426 G_GAN_Feat: 0.938 G_ID: 0.169 G_Rec: 0.432 D_GP: 0.036 D_real: 0.976 D_fake: 0.581 +(epoch: 225, iters: 3808, time: 0.064) G_GAN: -0.226 G_GAN_Feat: 0.563 G_ID: 0.122 G_Rec: 0.317 D_GP: 0.016 D_real: 0.738 D_fake: 1.227 +(epoch: 225, iters: 4208, time: 0.064) G_GAN: -0.061 G_GAN_Feat: 0.753 G_ID: 0.200 G_Rec: 0.384 D_GP: 0.021 D_real: 0.740 D_fake: 1.062 +(epoch: 225, iters: 4608, time: 0.064) G_GAN: -0.231 G_GAN_Feat: 0.730 G_ID: 0.123 G_Rec: 0.355 D_GP: 0.041 D_real: 0.633 D_fake: 1.232 +(epoch: 225, iters: 5008, time: 0.064) G_GAN: -0.018 G_GAN_Feat: 0.769 G_ID: 0.220 G_Rec: 0.374 D_GP: 0.026 D_real: 0.779 D_fake: 1.018 +(epoch: 225, iters: 5408, time: 0.064) G_GAN: 0.013 G_GAN_Feat: 0.621 G_ID: 0.127 G_Rec: 0.286 D_GP: 0.029 D_real: 0.937 D_fake: 0.987 +(epoch: 225, iters: 5808, time: 0.064) G_GAN: 0.495 G_GAN_Feat: 0.782 G_ID: 0.159 G_Rec: 0.377 D_GP: 0.025 D_real: 1.241 D_fake: 0.520 +(epoch: 225, iters: 6208, time: 0.064) G_GAN: 0.235 G_GAN_Feat: 0.898 G_ID: 0.120 G_Rec: 0.335 D_GP: 0.159 D_real: 0.549 D_fake: 0.767 +(epoch: 225, iters: 6608, time: 0.064) G_GAN: 0.329 G_GAN_Feat: 0.890 G_ID: 0.190 G_Rec: 0.375 D_GP: 0.045 D_real: 0.858 D_fake: 0.672 +(epoch: 225, iters: 7008, time: 0.064) G_GAN: 0.274 G_GAN_Feat: 0.688 G_ID: 0.152 G_Rec: 0.365 D_GP: 0.020 D_real: 1.254 D_fake: 0.726 +(epoch: 225, iters: 7408, time: 0.064) G_GAN: 0.386 G_GAN_Feat: 0.819 G_ID: 0.168 G_Rec: 0.413 D_GP: 0.023 D_real: 1.159 D_fake: 0.623 +(epoch: 225, iters: 7808, time: 0.064) G_GAN: -0.070 G_GAN_Feat: 0.702 G_ID: 0.143 G_Rec: 0.303 D_GP: 0.034 D_real: 0.831 D_fake: 1.070 +(epoch: 225, iters: 8208, time: 0.064) G_GAN: 0.590 G_GAN_Feat: 0.835 G_ID: 0.165 G_Rec: 0.371 D_GP: 0.022 D_real: 1.268 D_fake: 0.417 +(epoch: 225, iters: 8608, time: 0.064) G_GAN: 0.186 G_GAN_Feat: 0.714 G_ID: 0.129 G_Rec: 0.290 D_GP: 0.032 D_real: 1.042 D_fake: 0.814 +(epoch: 226, iters: 400, time: 0.064) G_GAN: 0.393 G_GAN_Feat: 0.809 G_ID: 0.168 G_Rec: 0.412 D_GP: 0.024 D_real: 1.131 D_fake: 0.608 +(epoch: 226, iters: 800, time: 0.064) G_GAN: 0.169 G_GAN_Feat: 0.773 G_ID: 0.113 G_Rec: 0.314 D_GP: 0.100 D_real: 0.778 D_fake: 0.831 +(epoch: 226, iters: 1200, time: 0.064) G_GAN: 0.427 G_GAN_Feat: 0.873 G_ID: 0.211 G_Rec: 0.387 D_GP: 0.024 D_real: 1.095 D_fake: 0.575 +(epoch: 226, iters: 1600, time: 0.064) G_GAN: 0.185 G_GAN_Feat: 0.857 G_ID: 0.156 G_Rec: 0.319 D_GP: 0.159 D_real: 0.505 D_fake: 0.819 +(epoch: 226, iters: 2000, time: 0.064) G_GAN: 0.436 G_GAN_Feat: 0.812 G_ID: 0.224 G_Rec: 0.338 D_GP: 0.028 D_real: 1.146 D_fake: 0.565 +(epoch: 226, iters: 2400, time: 0.064) G_GAN: 0.294 G_GAN_Feat: 0.906 G_ID: 0.134 G_Rec: 0.317 D_GP: 0.032 D_real: 0.708 D_fake: 0.707 +(epoch: 226, iters: 2800, time: 0.064) G_GAN: 0.321 G_GAN_Feat: 0.717 G_ID: 0.178 G_Rec: 0.385 D_GP: 0.021 D_real: 1.179 D_fake: 0.679 +(epoch: 226, iters: 3200, time: 0.064) G_GAN: 0.325 G_GAN_Feat: 0.568 G_ID: 0.144 G_Rec: 0.310 D_GP: 0.019 D_real: 1.284 D_fake: 0.675 +(epoch: 226, iters: 3600, time: 0.064) G_GAN: 0.338 G_GAN_Feat: 0.701 G_ID: 0.172 G_Rec: 0.353 D_GP: 0.021 D_real: 1.209 D_fake: 0.662 +(epoch: 226, iters: 4000, time: 0.064) G_GAN: 0.166 G_GAN_Feat: 0.652 G_ID: 0.134 G_Rec: 0.322 D_GP: 0.028 D_real: 1.016 D_fake: 0.834 +(epoch: 226, iters: 4400, time: 0.064) G_GAN: 0.072 G_GAN_Feat: 0.791 G_ID: 0.173 G_Rec: 0.391 D_GP: 0.037 D_real: 0.805 D_fake: 0.928 +(epoch: 226, iters: 4800, time: 0.064) G_GAN: 0.093 G_GAN_Feat: 0.617 G_ID: 0.138 G_Rec: 0.313 D_GP: 0.028 D_real: 1.018 D_fake: 0.909 +(epoch: 226, iters: 5200, time: 0.064) G_GAN: 0.350 G_GAN_Feat: 0.827 G_ID: 0.164 G_Rec: 0.396 D_GP: 0.033 D_real: 1.185 D_fake: 0.651 +(epoch: 226, iters: 5600, time: 0.064) G_GAN: 0.239 G_GAN_Feat: 0.611 G_ID: 0.120 G_Rec: 0.299 D_GP: 0.031 D_real: 1.165 D_fake: 0.763 +(epoch: 226, iters: 6000, time: 0.064) G_GAN: 0.472 G_GAN_Feat: 0.953 G_ID: 0.190 G_Rec: 0.438 D_GP: 0.049 D_real: 1.021 D_fake: 0.534 +(epoch: 226, iters: 6400, time: 0.064) G_GAN: 0.188 G_GAN_Feat: 0.703 G_ID: 0.142 G_Rec: 0.329 D_GP: 0.053 D_real: 0.948 D_fake: 0.812 +(epoch: 226, iters: 6800, time: 0.064) G_GAN: 0.431 G_GAN_Feat: 0.767 G_ID: 0.151 G_Rec: 0.381 D_GP: 0.025 D_real: 1.168 D_fake: 0.572 +(epoch: 226, iters: 7200, time: 0.064) G_GAN: 0.318 G_GAN_Feat: 0.644 G_ID: 0.136 G_Rec: 0.310 D_GP: 0.034 D_real: 1.257 D_fake: 0.687 +(epoch: 226, iters: 7600, time: 0.064) G_GAN: 0.442 G_GAN_Feat: 0.898 G_ID: 0.173 G_Rec: 0.402 D_GP: 0.038 D_real: 1.109 D_fake: 0.563 +(epoch: 226, iters: 8000, time: 0.064) G_GAN: -0.020 G_GAN_Feat: 0.722 G_ID: 0.154 G_Rec: 0.305 D_GP: 0.039 D_real: 0.684 D_fake: 1.020 +(epoch: 226, iters: 8400, time: 0.064) G_GAN: 0.203 G_GAN_Feat: 1.019 G_ID: 0.180 G_Rec: 0.413 D_GP: 0.124 D_real: 0.449 D_fake: 0.799 +(epoch: 227, iters: 192, time: 0.064) G_GAN: 0.231 G_GAN_Feat: 0.639 G_ID: 0.130 G_Rec: 0.314 D_GP: 0.017 D_real: 1.214 D_fake: 0.769 +(epoch: 227, iters: 592, time: 0.064) G_GAN: 0.480 G_GAN_Feat: 0.777 G_ID: 0.161 G_Rec: 0.400 D_GP: 0.020 D_real: 1.218 D_fake: 0.522 +(epoch: 227, iters: 992, time: 0.064) G_GAN: 0.183 G_GAN_Feat: 0.706 G_ID: 0.131 G_Rec: 0.350 D_GP: 0.040 D_real: 0.977 D_fake: 0.817 +(epoch: 227, iters: 1392, time: 0.064) G_GAN: 0.491 G_GAN_Feat: 0.759 G_ID: 0.167 G_Rec: 0.384 D_GP: 0.029 D_real: 1.296 D_fake: 0.511 +(epoch: 227, iters: 1792, time: 0.064) G_GAN: 0.252 G_GAN_Feat: 0.729 G_ID: 0.130 G_Rec: 0.362 D_GP: 0.052 D_real: 1.129 D_fake: 0.749 +(epoch: 227, iters: 2192, time: 0.064) G_GAN: 0.405 G_GAN_Feat: 0.861 G_ID: 0.190 G_Rec: 0.446 D_GP: 0.029 D_real: 1.008 D_fake: 0.597 +(epoch: 227, iters: 2592, time: 0.064) G_GAN: 0.278 G_GAN_Feat: 0.632 G_ID: 0.114 G_Rec: 0.295 D_GP: 0.031 D_real: 1.181 D_fake: 0.722 +(epoch: 227, iters: 2992, time: 0.064) G_GAN: 0.390 G_GAN_Feat: 1.011 G_ID: 0.182 G_Rec: 0.420 D_GP: 0.088 D_real: 0.805 D_fake: 0.632 +(epoch: 227, iters: 3392, time: 0.064) G_GAN: 0.085 G_GAN_Feat: 0.804 G_ID: 0.127 G_Rec: 0.306 D_GP: 0.040 D_real: 0.695 D_fake: 0.915 +(epoch: 227, iters: 3792, time: 0.064) G_GAN: 0.123 G_GAN_Feat: 0.897 G_ID: 0.150 G_Rec: 0.416 D_GP: 0.023 D_real: 0.877 D_fake: 0.878 +(epoch: 227, iters: 4192, time: 0.064) G_GAN: 0.272 G_GAN_Feat: 0.695 G_ID: 0.139 G_Rec: 0.287 D_GP: 0.026 D_real: 1.176 D_fake: 0.729 +(epoch: 227, iters: 4592, time: 0.064) G_GAN: 0.567 G_GAN_Feat: 1.122 G_ID: 0.228 G_Rec: 0.439 D_GP: 0.090 D_real: 0.423 D_fake: 0.489 +(epoch: 227, iters: 4992, time: 0.064) G_GAN: 0.311 G_GAN_Feat: 0.695 G_ID: 0.132 G_Rec: 0.331 D_GP: 0.031 D_real: 1.121 D_fake: 0.691 +(epoch: 227, iters: 5392, time: 0.064) G_GAN: 0.354 G_GAN_Feat: 0.869 G_ID: 0.174 G_Rec: 0.379 D_GP: 0.042 D_real: 0.942 D_fake: 0.646 +(epoch: 227, iters: 5792, time: 0.064) G_GAN: 0.313 G_GAN_Feat: 0.881 G_ID: 0.141 G_Rec: 0.322 D_GP: 0.067 D_real: 1.356 D_fake: 0.721 +(epoch: 227, iters: 6192, time: 0.064) G_GAN: 0.347 G_GAN_Feat: 0.854 G_ID: 0.167 G_Rec: 0.409 D_GP: 0.028 D_real: 0.962 D_fake: 0.653 +(epoch: 227, iters: 6592, time: 0.064) G_GAN: 0.132 G_GAN_Feat: 0.630 G_ID: 0.111 G_Rec: 0.302 D_GP: 0.031 D_real: 1.082 D_fake: 0.868 +(epoch: 227, iters: 6992, time: 0.064) G_GAN: 0.366 G_GAN_Feat: 1.008 G_ID: 0.164 G_Rec: 0.454 D_GP: 0.057 D_real: 0.789 D_fake: 0.638 +(epoch: 227, iters: 7392, time: 0.064) G_GAN: 0.076 G_GAN_Feat: 0.681 G_ID: 0.129 G_Rec: 0.327 D_GP: 0.037 D_real: 1.011 D_fake: 0.925 +(epoch: 227, iters: 7792, time: 0.064) G_GAN: 0.529 G_GAN_Feat: 0.942 G_ID: 0.147 G_Rec: 0.395 D_GP: 0.066 D_real: 0.956 D_fake: 0.478 +(epoch: 227, iters: 8192, time: 0.064) G_GAN: 0.113 G_GAN_Feat: 0.596 G_ID: 0.143 G_Rec: 0.252 D_GP: 0.022 D_real: 1.034 D_fake: 0.887 +(epoch: 227, iters: 8592, time: 0.064) G_GAN: 0.391 G_GAN_Feat: 0.824 G_ID: 0.169 G_Rec: 0.403 D_GP: 0.031 D_real: 1.077 D_fake: 0.610 +(epoch: 228, iters: 384, time: 0.064) G_GAN: 0.187 G_GAN_Feat: 0.652 G_ID: 0.153 G_Rec: 0.331 D_GP: 0.028 D_real: 1.130 D_fake: 0.813 +(epoch: 228, iters: 784, time: 0.064) G_GAN: 0.171 G_GAN_Feat: 0.885 G_ID: 0.158 G_Rec: 0.390 D_GP: 0.063 D_real: 0.848 D_fake: 0.832 +(epoch: 228, iters: 1184, time: 0.064) G_GAN: 0.026 G_GAN_Feat: 0.741 G_ID: 0.130 G_Rec: 0.308 D_GP: 0.044 D_real: 0.797 D_fake: 0.974 +(epoch: 228, iters: 1584, time: 0.064) G_GAN: 0.647 G_GAN_Feat: 1.097 G_ID: 0.160 G_Rec: 0.422 D_GP: 0.039 D_real: 1.221 D_fake: 0.408 +(epoch: 228, iters: 1984, time: 0.064) G_GAN: 0.094 G_GAN_Feat: 0.825 G_ID: 0.142 G_Rec: 0.313 D_GP: 0.124 D_real: 0.709 D_fake: 0.907 +(epoch: 228, iters: 2384, time: 0.064) G_GAN: 0.438 G_GAN_Feat: 0.851 G_ID: 0.198 G_Rec: 0.413 D_GP: 0.028 D_real: 1.116 D_fake: 0.570 +(epoch: 228, iters: 2784, time: 0.064) G_GAN: 0.148 G_GAN_Feat: 0.630 G_ID: 0.128 G_Rec: 0.273 D_GP: 0.026 D_real: 1.027 D_fake: 0.852 +(epoch: 228, iters: 3184, time: 0.064) G_GAN: 0.331 G_GAN_Feat: 0.801 G_ID: 0.163 G_Rec: 0.368 D_GP: 0.033 D_real: 1.074 D_fake: 0.671 +(epoch: 228, iters: 3584, time: 0.064) G_GAN: 0.057 G_GAN_Feat: 1.051 G_ID: 0.117 G_Rec: 0.328 D_GP: 0.169 D_real: 0.537 D_fake: 0.945 +(epoch: 228, iters: 3984, time: 0.064) G_GAN: 0.720 G_GAN_Feat: 0.966 G_ID: 0.198 G_Rec: 0.409 D_GP: 0.036 D_real: 1.287 D_fake: 0.359 +(epoch: 228, iters: 4384, time: 0.064) G_GAN: 0.210 G_GAN_Feat: 1.016 G_ID: 0.126 G_Rec: 0.327 D_GP: 0.316 D_real: 0.385 D_fake: 0.792 +(epoch: 228, iters: 4784, time: 0.064) G_GAN: 0.356 G_GAN_Feat: 0.908 G_ID: 0.203 G_Rec: 0.372 D_GP: 0.050 D_real: 0.856 D_fake: 0.648 +(epoch: 228, iters: 5184, time: 0.064) G_GAN: 0.167 G_GAN_Feat: 0.720 G_ID: 0.155 G_Rec: 0.307 D_GP: 0.023 D_real: 1.047 D_fake: 0.838 +(epoch: 228, iters: 5584, time: 0.064) G_GAN: 0.251 G_GAN_Feat: 0.873 G_ID: 0.202 G_Rec: 0.398 D_GP: 0.026 D_real: 0.913 D_fake: 0.750 +(epoch: 228, iters: 5984, time: 0.064) G_GAN: 0.083 G_GAN_Feat: 0.699 G_ID: 0.147 G_Rec: 0.305 D_GP: 0.020 D_real: 0.993 D_fake: 0.917 +(epoch: 228, iters: 6384, time: 0.064) G_GAN: 0.385 G_GAN_Feat: 0.874 G_ID: 0.176 G_Rec: 0.353 D_GP: 0.062 D_real: 0.839 D_fake: 0.618 +(epoch: 228, iters: 6784, time: 0.064) G_GAN: 0.225 G_GAN_Feat: 0.600 G_ID: 0.123 G_Rec: 0.285 D_GP: 0.022 D_real: 1.159 D_fake: 0.775 +(epoch: 228, iters: 7184, time: 0.064) G_GAN: 0.461 G_GAN_Feat: 0.872 G_ID: 0.183 G_Rec: 0.365 D_GP: 0.026 D_real: 1.190 D_fake: 0.544 +(epoch: 228, iters: 7584, time: 0.064) G_GAN: 0.200 G_GAN_Feat: 0.892 G_ID: 0.141 G_Rec: 0.308 D_GP: 0.047 D_real: 0.616 D_fake: 0.801 +(epoch: 228, iters: 7984, time: 0.064) G_GAN: 0.456 G_GAN_Feat: 1.145 G_ID: 0.180 G_Rec: 0.421 D_GP: 0.243 D_real: 0.421 D_fake: 0.552 +(epoch: 228, iters: 8384, time: 0.064) G_GAN: 0.377 G_GAN_Feat: 0.842 G_ID: 0.126 G_Rec: 0.310 D_GP: 0.077 D_real: 0.751 D_fake: 0.695 +(epoch: 229, iters: 176, time: 0.064) G_GAN: 0.331 G_GAN_Feat: 0.838 G_ID: 0.144 G_Rec: 0.366 D_GP: 0.027 D_real: 1.103 D_fake: 0.669 +(epoch: 229, iters: 576, time: 0.064) G_GAN: 0.276 G_GAN_Feat: 0.779 G_ID: 0.116 G_Rec: 0.354 D_GP: 0.023 D_real: 1.099 D_fake: 0.725 +(epoch: 229, iters: 976, time: 0.064) G_GAN: 0.422 G_GAN_Feat: 0.904 G_ID: 0.162 G_Rec: 0.421 D_GP: 0.024 D_real: 1.211 D_fake: 0.581 +(epoch: 229, iters: 1376, time: 0.064) G_GAN: -0.210 G_GAN_Feat: 0.656 G_ID: 0.143 G_Rec: 0.304 D_GP: 0.020 D_real: 0.745 D_fake: 1.210 +(epoch: 229, iters: 1776, time: 0.064) G_GAN: -0.130 G_GAN_Feat: 0.750 G_ID: 0.170 G_Rec: 0.389 D_GP: 0.017 D_real: 0.675 D_fake: 1.131 +(epoch: 229, iters: 2176, time: 0.064) G_GAN: -0.192 G_GAN_Feat: 0.564 G_ID: 0.112 G_Rec: 0.267 D_GP: 0.019 D_real: 0.803 D_fake: 1.192 +(epoch: 229, iters: 2576, time: 0.064) G_GAN: 0.048 G_GAN_Feat: 0.808 G_ID: 0.171 G_Rec: 0.386 D_GP: 0.030 D_real: 0.781 D_fake: 0.953 +(epoch: 229, iters: 2976, time: 0.064) G_GAN: -0.275 G_GAN_Feat: 0.742 G_ID: 0.122 G_Rec: 0.326 D_GP: 0.060 D_real: 0.403 D_fake: 1.275 +(epoch: 229, iters: 3376, time: 0.064) G_GAN: 0.488 G_GAN_Feat: 0.971 G_ID: 0.166 G_Rec: 0.444 D_GP: 0.035 D_real: 1.150 D_fake: 0.550 +(epoch: 229, iters: 3776, time: 0.064) G_GAN: 0.284 G_GAN_Feat: 0.784 G_ID: 0.125 G_Rec: 0.315 D_GP: 0.038 D_real: 0.920 D_fake: 0.721 +(epoch: 229, iters: 4176, time: 0.064) G_GAN: 0.118 G_GAN_Feat: 0.969 G_ID: 0.184 G_Rec: 0.368 D_GP: 0.102 D_real: 0.348 D_fake: 0.882 +(epoch: 229, iters: 4576, time: 0.064) G_GAN: 0.287 G_GAN_Feat: 0.587 G_ID: 0.110 G_Rec: 0.295 D_GP: 0.020 D_real: 1.260 D_fake: 0.714 +(epoch: 229, iters: 4976, time: 0.064) G_GAN: 0.325 G_GAN_Feat: 0.718 G_ID: 0.158 G_Rec: 0.413 D_GP: 0.017 D_real: 1.136 D_fake: 0.676 +(epoch: 229, iters: 5376, time: 0.064) G_GAN: 0.190 G_GAN_Feat: 0.607 G_ID: 0.131 G_Rec: 0.330 D_GP: 0.022 D_real: 1.136 D_fake: 0.810 +(epoch: 229, iters: 5776, time: 0.064) G_GAN: 0.150 G_GAN_Feat: 0.695 G_ID: 0.181 G_Rec: 0.339 D_GP: 0.023 D_real: 0.937 D_fake: 0.850 +(epoch: 229, iters: 6176, time: 0.064) G_GAN: 0.069 G_GAN_Feat: 0.584 G_ID: 0.137 G_Rec: 0.342 D_GP: 0.026 D_real: 0.994 D_fake: 0.932 +(epoch: 229, iters: 6576, time: 0.064) G_GAN: 0.123 G_GAN_Feat: 0.827 G_ID: 0.167 G_Rec: 0.480 D_GP: 0.035 D_real: 0.750 D_fake: 0.878 +(epoch: 229, iters: 6976, time: 0.064) G_GAN: 0.046 G_GAN_Feat: 0.589 G_ID: 0.162 G_Rec: 0.283 D_GP: 0.028 D_real: 0.998 D_fake: 0.954 +(epoch: 229, iters: 7376, time: 0.064) G_GAN: 0.204 G_GAN_Feat: 0.816 G_ID: 0.189 G_Rec: 0.392 D_GP: 0.064 D_real: 0.777 D_fake: 0.799 +(epoch: 229, iters: 7776, time: 0.064) G_GAN: 0.111 G_GAN_Feat: 0.622 G_ID: 0.142 G_Rec: 0.314 D_GP: 0.036 D_real: 1.002 D_fake: 0.889 +(epoch: 229, iters: 8176, time: 0.064) G_GAN: 0.267 G_GAN_Feat: 0.844 G_ID: 0.173 G_Rec: 0.406 D_GP: 0.056 D_real: 0.785 D_fake: 0.736 +(epoch: 229, iters: 8576, time: 0.064) G_GAN: 0.109 G_GAN_Feat: 0.591 G_ID: 0.136 G_Rec: 0.303 D_GP: 0.021 D_real: 1.033 D_fake: 0.891 +(epoch: 230, iters: 368, time: 0.064) G_GAN: 0.171 G_GAN_Feat: 0.805 G_ID: 0.155 G_Rec: 0.379 D_GP: 0.032 D_real: 0.862 D_fake: 0.830 +(epoch: 230, iters: 768, time: 0.064) G_GAN: -0.028 G_GAN_Feat: 0.709 G_ID: 0.122 G_Rec: 0.328 D_GP: 0.029 D_real: 0.930 D_fake: 1.028 +(epoch: 230, iters: 1168, time: 0.064) G_GAN: 0.441 G_GAN_Feat: 0.825 G_ID: 0.166 G_Rec: 0.424 D_GP: 0.023 D_real: 1.162 D_fake: 0.566 +(epoch: 230, iters: 1568, time: 0.064) G_GAN: 0.144 G_GAN_Feat: 0.646 G_ID: 0.130 G_Rec: 0.449 D_GP: 0.025 D_real: 1.048 D_fake: 0.856 +(epoch: 230, iters: 1968, time: 0.064) G_GAN: 0.370 G_GAN_Feat: 0.804 G_ID: 0.171 G_Rec: 0.396 D_GP: 0.028 D_real: 1.142 D_fake: 0.631 +(epoch: 230, iters: 2368, time: 0.064) G_GAN: -0.019 G_GAN_Feat: 0.718 G_ID: 0.145 G_Rec: 0.301 D_GP: 0.074 D_real: 0.715 D_fake: 1.019 +(epoch: 230, iters: 2768, time: 0.064) G_GAN: 0.365 G_GAN_Feat: 0.812 G_ID: 0.193 G_Rec: 0.386 D_GP: 0.029 D_real: 1.040 D_fake: 0.638 +(epoch: 230, iters: 3168, time: 0.064) G_GAN: -0.105 G_GAN_Feat: 0.639 G_ID: 0.160 G_Rec: 0.325 D_GP: 0.022 D_real: 0.921 D_fake: 1.105 +(epoch: 230, iters: 3568, time: 0.064) G_GAN: 0.099 G_GAN_Feat: 0.842 G_ID: 0.169 G_Rec: 0.395 D_GP: 0.025 D_real: 0.842 D_fake: 0.901 +(epoch: 230, iters: 3968, time: 0.064) G_GAN: 0.432 G_GAN_Feat: 0.691 G_ID: 0.112 G_Rec: 0.303 D_GP: 0.035 D_real: 1.258 D_fake: 0.572 +(epoch: 230, iters: 4368, time: 0.064) G_GAN: 0.238 G_GAN_Feat: 0.722 G_ID: 0.166 G_Rec: 0.362 D_GP: 0.019 D_real: 1.082 D_fake: 0.762 +(epoch: 230, iters: 4768, time: 0.064) G_GAN: 0.402 G_GAN_Feat: 0.579 G_ID: 0.116 G_Rec: 0.295 D_GP: 0.020 D_real: 1.372 D_fake: 0.598 +(epoch: 230, iters: 5168, time: 0.064) G_GAN: 0.453 G_GAN_Feat: 0.794 G_ID: 0.184 G_Rec: 0.363 D_GP: 0.030 D_real: 1.169 D_fake: 0.548 +(epoch: 230, iters: 5568, time: 0.064) G_GAN: 0.054 G_GAN_Feat: 0.679 G_ID: 0.137 G_Rec: 0.310 D_GP: 0.030 D_real: 1.036 D_fake: 0.946 +(epoch: 230, iters: 5968, time: 0.064) G_GAN: 0.413 G_GAN_Feat: 0.871 G_ID: 0.175 G_Rec: 0.380 D_GP: 0.036 D_real: 1.011 D_fake: 0.587 +(epoch: 230, iters: 6368, time: 0.064) G_GAN: 0.171 G_GAN_Feat: 0.674 G_ID: 0.113 G_Rec: 0.313 D_GP: 0.041 D_real: 1.029 D_fake: 0.829 +(epoch: 230, iters: 6768, time: 0.064) G_GAN: 0.390 G_GAN_Feat: 0.946 G_ID: 0.180 G_Rec: 0.452 D_GP: 0.041 D_real: 0.887 D_fake: 0.614 +(epoch: 230, iters: 7168, time: 0.064) G_GAN: 0.489 G_GAN_Feat: 0.857 G_ID: 0.138 G_Rec: 0.310 D_GP: 0.132 D_real: 0.931 D_fake: 0.536 +(epoch: 230, iters: 7568, time: 0.064) G_GAN: 0.332 G_GAN_Feat: 0.871 G_ID: 0.176 G_Rec: 0.479 D_GP: 0.019 D_real: 1.257 D_fake: 0.669 +(epoch: 230, iters: 7968, time: 0.064) G_GAN: 0.084 G_GAN_Feat: 0.622 G_ID: 0.129 G_Rec: 0.312 D_GP: 0.020 D_real: 0.984 D_fake: 0.916 +(epoch: 230, iters: 8368, time: 0.064) G_GAN: -0.020 G_GAN_Feat: 0.805 G_ID: 0.172 G_Rec: 0.363 D_GP: 0.031 D_real: 0.738 D_fake: 1.020 +(epoch: 231, iters: 160, time: 0.064) G_GAN: 0.161 G_GAN_Feat: 0.620 G_ID: 0.114 G_Rec: 0.341 D_GP: 0.028 D_real: 1.038 D_fake: 0.839 +(epoch: 231, iters: 560, time: 0.064) G_GAN: 0.392 G_GAN_Feat: 0.808 G_ID: 0.154 G_Rec: 0.360 D_GP: 0.037 D_real: 1.091 D_fake: 0.610 +(epoch: 231, iters: 960, time: 0.064) G_GAN: 0.152 G_GAN_Feat: 0.696 G_ID: 0.138 G_Rec: 0.337 D_GP: 0.033 D_real: 1.008 D_fake: 0.849 +(epoch: 231, iters: 1360, time: 0.064) G_GAN: 0.830 G_GAN_Feat: 0.911 G_ID: 0.166 G_Rec: 0.408 D_GP: 0.057 D_real: 1.468 D_fake: 0.211 +(epoch: 231, iters: 1760, time: 0.064) G_GAN: 0.373 G_GAN_Feat: 0.706 G_ID: 0.135 G_Rec: 0.334 D_GP: 0.025 D_real: 1.193 D_fake: 0.643 +(epoch: 231, iters: 2160, time: 0.064) G_GAN: 0.473 G_GAN_Feat: 0.778 G_ID: 0.178 G_Rec: 0.379 D_GP: 0.021 D_real: 1.210 D_fake: 0.532 +(epoch: 231, iters: 2560, time: 0.064) G_GAN: 0.062 G_GAN_Feat: 0.677 G_ID: 0.138 G_Rec: 0.309 D_GP: 0.024 D_real: 0.919 D_fake: 0.938 +(epoch: 231, iters: 2960, time: 0.064) G_GAN: 0.489 G_GAN_Feat: 0.984 G_ID: 0.184 G_Rec: 0.424 D_GP: 0.039 D_real: 0.939 D_fake: 0.522 +(epoch: 231, iters: 3360, time: 0.064) G_GAN: 0.192 G_GAN_Feat: 0.662 G_ID: 0.130 G_Rec: 0.308 D_GP: 0.025 D_real: 1.110 D_fake: 0.808 +(epoch: 231, iters: 3760, time: 0.064) G_GAN: 0.525 G_GAN_Feat: 0.739 G_ID: 0.173 G_Rec: 0.351 D_GP: 0.032 D_real: 1.271 D_fake: 0.477 +(epoch: 231, iters: 4160, time: 0.064) G_GAN: 0.025 G_GAN_Feat: 0.731 G_ID: 0.135 G_Rec: 0.308 D_GP: 0.049 D_real: 0.808 D_fake: 0.975 +(epoch: 231, iters: 4560, time: 0.064) G_GAN: -0.008 G_GAN_Feat: 1.070 G_ID: 0.212 G_Rec: 0.413 D_GP: 0.069 D_real: 0.145 D_fake: 1.013 +(epoch: 231, iters: 4960, time: 0.064) G_GAN: 0.174 G_GAN_Feat: 0.715 G_ID: 0.141 G_Rec: 0.299 D_GP: 0.039 D_real: 0.924 D_fake: 0.826 +(epoch: 231, iters: 5360, time: 0.064) G_GAN: 0.343 G_GAN_Feat: 0.950 G_ID: 0.189 G_Rec: 0.415 D_GP: 0.057 D_real: 0.715 D_fake: 0.658 +(epoch: 231, iters: 5760, time: 0.064) G_GAN: 0.221 G_GAN_Feat: 0.784 G_ID: 0.108 G_Rec: 0.287 D_GP: 0.072 D_real: 0.905 D_fake: 0.779 +(epoch: 231, iters: 6160, time: 0.064) G_GAN: 0.675 G_GAN_Feat: 0.865 G_ID: 0.193 G_Rec: 0.357 D_GP: 0.029 D_real: 1.363 D_fake: 0.337 +(epoch: 231, iters: 6560, time: 0.064) G_GAN: 0.115 G_GAN_Feat: 0.724 G_ID: 0.138 G_Rec: 0.299 D_GP: 0.042 D_real: 0.877 D_fake: 0.885 +(epoch: 231, iters: 6960, time: 0.064) G_GAN: 0.362 G_GAN_Feat: 0.921 G_ID: 0.179 G_Rec: 0.401 D_GP: 0.033 D_real: 0.998 D_fake: 0.640 +(epoch: 231, iters: 7360, time: 0.064) G_GAN: 0.362 G_GAN_Feat: 0.742 G_ID: 0.133 G_Rec: 0.305 D_GP: 0.026 D_real: 1.179 D_fake: 0.646 +(epoch: 231, iters: 7760, time: 0.064) G_GAN: 0.410 G_GAN_Feat: 0.729 G_ID: 0.187 G_Rec: 0.379 D_GP: 0.021 D_real: 1.303 D_fake: 0.593 +(epoch: 231, iters: 8160, time: 0.064) G_GAN: 0.227 G_GAN_Feat: 0.591 G_ID: 0.124 G_Rec: 0.307 D_GP: 0.018 D_real: 1.212 D_fake: 0.773 +(epoch: 231, iters: 8560, time: 0.064) G_GAN: 0.304 G_GAN_Feat: 0.798 G_ID: 0.193 G_Rec: 0.433 D_GP: 0.018 D_real: 1.063 D_fake: 0.697 +(epoch: 232, iters: 352, time: 0.064) G_GAN: 0.103 G_GAN_Feat: 0.558 G_ID: 0.132 G_Rec: 0.279 D_GP: 0.025 D_real: 1.052 D_fake: 0.898 +(epoch: 232, iters: 752, time: 0.064) G_GAN: 0.320 G_GAN_Feat: 0.818 G_ID: 0.174 G_Rec: 0.407 D_GP: 0.028 D_real: 0.989 D_fake: 0.686 +(epoch: 232, iters: 1152, time: 0.064) G_GAN: 0.064 G_GAN_Feat: 0.686 G_ID: 0.142 G_Rec: 0.324 D_GP: 0.035 D_real: 0.873 D_fake: 0.942 +(epoch: 232, iters: 1552, time: 0.064) G_GAN: 0.273 G_GAN_Feat: 0.871 G_ID: 0.193 G_Rec: 0.432 D_GP: 0.056 D_real: 0.855 D_fake: 0.729 +(epoch: 232, iters: 1952, time: 0.064) G_GAN: 0.195 G_GAN_Feat: 0.669 G_ID: 0.131 G_Rec: 0.291 D_GP: 0.044 D_real: 0.994 D_fake: 0.807 +(epoch: 232, iters: 2352, time: 0.064) G_GAN: 0.293 G_GAN_Feat: 0.896 G_ID: 0.214 G_Rec: 0.430 D_GP: 0.045 D_real: 0.888 D_fake: 0.711 +(epoch: 232, iters: 2752, time: 0.064) G_GAN: 0.058 G_GAN_Feat: 0.667 G_ID: 0.132 G_Rec: 0.310 D_GP: 0.052 D_real: 0.937 D_fake: 0.942 +(epoch: 232, iters: 3152, time: 0.064) G_GAN: 0.451 G_GAN_Feat: 0.864 G_ID: 0.181 G_Rec: 0.436 D_GP: 0.022 D_real: 1.218 D_fake: 0.550 +(epoch: 232, iters: 3552, time: 0.064) G_GAN: 0.264 G_GAN_Feat: 0.656 G_ID: 0.153 G_Rec: 0.272 D_GP: 0.043 D_real: 1.119 D_fake: 0.736 +(epoch: 232, iters: 3952, time: 0.064) G_GAN: 0.571 G_GAN_Feat: 0.900 G_ID: 0.191 G_Rec: 0.434 D_GP: 0.027 D_real: 1.199 D_fake: 0.445 +(epoch: 232, iters: 4352, time: 0.064) G_GAN: 0.238 G_GAN_Feat: 0.669 G_ID: 0.130 G_Rec: 0.300 D_GP: 0.026 D_real: 1.215 D_fake: 0.763 +(epoch: 232, iters: 4752, time: 0.064) G_GAN: 0.416 G_GAN_Feat: 0.914 G_ID: 0.172 G_Rec: 0.410 D_GP: 0.025 D_real: 1.063 D_fake: 0.587 +(epoch: 232, iters: 5152, time: 0.064) G_GAN: 0.053 G_GAN_Feat: 0.662 G_ID: 0.124 G_Rec: 0.275 D_GP: 0.042 D_real: 0.907 D_fake: 0.947 +(epoch: 232, iters: 5552, time: 0.064) G_GAN: 0.530 G_GAN_Feat: 0.754 G_ID: 0.174 G_Rec: 0.381 D_GP: 0.020 D_real: 1.321 D_fake: 0.475 +(epoch: 232, iters: 5952, time: 0.064) G_GAN: 0.243 G_GAN_Feat: 0.666 G_ID: 0.165 G_Rec: 0.296 D_GP: 0.025 D_real: 1.130 D_fake: 0.757 +(epoch: 232, iters: 6352, time: 0.064) G_GAN: 0.585 G_GAN_Feat: 0.785 G_ID: 0.178 G_Rec: 0.383 D_GP: 0.026 D_real: 1.300 D_fake: 0.437 +(epoch: 232, iters: 6752, time: 0.064) G_GAN: 0.057 G_GAN_Feat: 0.788 G_ID: 0.142 G_Rec: 0.333 D_GP: 0.058 D_real: 0.663 D_fake: 0.943 +(epoch: 232, iters: 7152, time: 0.064) G_GAN: 0.574 G_GAN_Feat: 0.913 G_ID: 0.177 G_Rec: 0.408 D_GP: 0.050 D_real: 0.960 D_fake: 0.437 +(epoch: 232, iters: 7552, time: 0.064) G_GAN: 0.322 G_GAN_Feat: 0.672 G_ID: 0.128 G_Rec: 0.315 D_GP: 0.024 D_real: 1.200 D_fake: 0.682 +(epoch: 232, iters: 7952, time: 0.064) G_GAN: 0.619 G_GAN_Feat: 0.982 G_ID: 0.179 G_Rec: 0.449 D_GP: 0.037 D_real: 1.156 D_fake: 0.413 +(epoch: 232, iters: 8352, time: 0.064) G_GAN: 0.198 G_GAN_Feat: 0.954 G_ID: 0.145 G_Rec: 0.296 D_GP: 0.080 D_real: 0.390 D_fake: 0.802 +(epoch: 233, iters: 144, time: 0.064) G_GAN: 0.682 G_GAN_Feat: 1.001 G_ID: 0.170 G_Rec: 0.425 D_GP: 0.021 D_real: 1.414 D_fake: 0.394 +(epoch: 233, iters: 544, time: 0.064) G_GAN: 0.181 G_GAN_Feat: 0.631 G_ID: 0.139 G_Rec: 0.297 D_GP: 0.018 D_real: 1.122 D_fake: 0.819 +(epoch: 233, iters: 944, time: 0.064) G_GAN: 0.267 G_GAN_Feat: 0.848 G_ID: 0.179 G_Rec: 0.424 D_GP: 0.023 D_real: 1.001 D_fake: 0.733 +(epoch: 233, iters: 1344, time: 0.064) G_GAN: 0.030 G_GAN_Feat: 0.598 G_ID: 0.149 G_Rec: 0.287 D_GP: 0.022 D_real: 0.992 D_fake: 0.970 +(epoch: 233, iters: 1744, time: 0.064) G_GAN: 0.341 G_GAN_Feat: 0.874 G_ID: 0.175 G_Rec: 0.402 D_GP: 0.045 D_real: 0.870 D_fake: 0.667 +(epoch: 233, iters: 2144, time: 0.064) G_GAN: 0.307 G_GAN_Feat: 0.721 G_ID: 0.169 G_Rec: 0.311 D_GP: 0.025 D_real: 1.206 D_fake: 0.693 +(epoch: 233, iters: 2544, time: 0.064) G_GAN: 0.442 G_GAN_Feat: 1.023 G_ID: 0.184 G_Rec: 0.421 D_GP: 0.167 D_real: 0.560 D_fake: 0.578 +(epoch: 233, iters: 2944, time: 0.064) G_GAN: 0.081 G_GAN_Feat: 0.673 G_ID: 0.130 G_Rec: 0.321 D_GP: 0.022 D_real: 0.973 D_fake: 0.919 +(epoch: 233, iters: 3344, time: 0.064) G_GAN: 0.750 G_GAN_Feat: 0.787 G_ID: 0.178 G_Rec: 0.367 D_GP: 0.026 D_real: 1.502 D_fake: 0.315 +(epoch: 233, iters: 3744, time: 0.064) G_GAN: 0.357 G_GAN_Feat: 0.780 G_ID: 0.155 G_Rec: 0.324 D_GP: 0.038 D_real: 1.016 D_fake: 0.644 +(epoch: 233, iters: 4144, time: 0.064) G_GAN: 0.350 G_GAN_Feat: 1.061 G_ID: 0.167 G_Rec: 0.428 D_GP: 0.065 D_real: 0.470 D_fake: 0.652 +(epoch: 233, iters: 4544, time: 0.064) G_GAN: -0.002 G_GAN_Feat: 0.683 G_ID: 0.131 G_Rec: 0.303 D_GP: 0.030 D_real: 0.947 D_fake: 1.002 +(epoch: 233, iters: 4944, time: 0.064) G_GAN: 0.507 G_GAN_Feat: 0.923 G_ID: 0.172 G_Rec: 0.402 D_GP: 0.039 D_real: 1.091 D_fake: 0.499 +(epoch: 233, iters: 5344, time: 0.064) G_GAN: 0.369 G_GAN_Feat: 0.646 G_ID: 0.145 G_Rec: 0.306 D_GP: 0.020 D_real: 1.375 D_fake: 0.632 +(epoch: 233, iters: 5744, time: 0.064) G_GAN: 0.291 G_GAN_Feat: 0.869 G_ID: 0.210 G_Rec: 0.430 D_GP: 0.021 D_real: 1.006 D_fake: 0.711 +(epoch: 233, iters: 6144, time: 0.064) G_GAN: 0.062 G_GAN_Feat: 0.648 G_ID: 0.133 G_Rec: 0.335 D_GP: 0.023 D_real: 0.997 D_fake: 0.939 +(epoch: 233, iters: 6544, time: 0.064) G_GAN: 0.469 G_GAN_Feat: 0.771 G_ID: 0.148 G_Rec: 0.370 D_GP: 0.025 D_real: 1.252 D_fake: 0.553 +(epoch: 233, iters: 6944, time: 0.064) G_GAN: 0.222 G_GAN_Feat: 0.667 G_ID: 0.110 G_Rec: 0.289 D_GP: 0.036 D_real: 1.065 D_fake: 0.780 +(epoch: 233, iters: 7344, time: 0.064) G_GAN: 0.426 G_GAN_Feat: 0.904 G_ID: 0.172 G_Rec: 0.393 D_GP: 0.030 D_real: 1.111 D_fake: 0.577 +(epoch: 233, iters: 7744, time: 0.064) G_GAN: -0.033 G_GAN_Feat: 0.935 G_ID: 0.131 G_Rec: 0.340 D_GP: 0.172 D_real: 0.608 D_fake: 1.034 +(epoch: 233, iters: 8144, time: 0.064) G_GAN: 0.343 G_GAN_Feat: 0.906 G_ID: 0.186 G_Rec: 0.375 D_GP: 0.034 D_real: 1.043 D_fake: 0.658 +(epoch: 233, iters: 8544, time: 0.064) G_GAN: 0.263 G_GAN_Feat: 0.656 G_ID: 0.123 G_Rec: 0.297 D_GP: 0.020 D_real: 1.205 D_fake: 0.737 +(epoch: 234, iters: 336, time: 0.064) G_GAN: 0.329 G_GAN_Feat: 0.953 G_ID: 0.164 G_Rec: 0.427 D_GP: 0.051 D_real: 0.704 D_fake: 0.671 +(epoch: 234, iters: 736, time: 0.064) G_GAN: 0.323 G_GAN_Feat: 0.670 G_ID: 0.115 G_Rec: 0.295 D_GP: 0.026 D_real: 1.217 D_fake: 0.683 +(epoch: 234, iters: 1136, time: 0.064) G_GAN: 0.787 G_GAN_Feat: 0.919 G_ID: 0.163 G_Rec: 0.375 D_GP: 0.323 D_real: 1.141 D_fake: 0.281 +(epoch: 234, iters: 1536, time: 0.064) G_GAN: -0.271 G_GAN_Feat: 0.589 G_ID: 0.121 G_Rec: 0.281 D_GP: 0.017 D_real: 0.694 D_fake: 1.271 +(epoch: 234, iters: 1936, time: 0.064) G_GAN: -0.093 G_GAN_Feat: 0.774 G_ID: 0.182 G_Rec: 0.381 D_GP: 0.019 D_real: 0.740 D_fake: 1.094 +(epoch: 234, iters: 2336, time: 0.064) G_GAN: -0.090 G_GAN_Feat: 0.605 G_ID: 0.127 G_Rec: 0.288 D_GP: 0.023 D_real: 0.831 D_fake: 1.090 +(epoch: 234, iters: 2736, time: 0.064) G_GAN: -0.038 G_GAN_Feat: 0.816 G_ID: 0.161 G_Rec: 0.396 D_GP: 0.026 D_real: 0.711 D_fake: 1.038 +(epoch: 234, iters: 3136, time: 0.064) G_GAN: -0.005 G_GAN_Feat: 0.645 G_ID: 0.129 G_Rec: 0.282 D_GP: 0.030 D_real: 0.887 D_fake: 1.007 +(epoch: 234, iters: 3536, time: 0.064) G_GAN: 0.561 G_GAN_Feat: 0.933 G_ID: 0.160 G_Rec: 0.428 D_GP: 0.043 D_real: 1.107 D_fake: 0.445 +(epoch: 234, iters: 3936, time: 0.064) G_GAN: -0.069 G_GAN_Feat: 0.788 G_ID: 0.111 G_Rec: 0.300 D_GP: 0.115 D_real: 0.518 D_fake: 1.069 +(epoch: 234, iters: 4336, time: 0.064) G_GAN: 0.544 G_GAN_Feat: 1.121 G_ID: 0.148 G_Rec: 0.413 D_GP: 0.209 D_real: 0.506 D_fake: 0.537 +(epoch: 234, iters: 4736, time: 0.064) G_GAN: 0.170 G_GAN_Feat: 0.674 G_ID: 0.139 G_Rec: 0.261 D_GP: 0.029 D_real: 1.061 D_fake: 0.830 +(epoch: 234, iters: 5136, time: 0.064) G_GAN: 0.450 G_GAN_Feat: 0.972 G_ID: 0.182 G_Rec: 0.392 D_GP: 0.071 D_real: 0.719 D_fake: 0.556 +(epoch: 234, iters: 5536, time: 0.064) G_GAN: 0.564 G_GAN_Feat: 0.738 G_ID: 0.124 G_Rec: 0.285 D_GP: 0.025 D_real: 1.436 D_fake: 0.481 +(epoch: 234, iters: 5936, time: 0.064) G_GAN: 0.557 G_GAN_Feat: 0.920 G_ID: 0.157 G_Rec: 0.400 D_GP: 0.032 D_real: 1.108 D_fake: 0.460 +(epoch: 234, iters: 6336, time: 0.064) G_GAN: 0.531 G_GAN_Feat: 0.992 G_ID: 0.125 G_Rec: 0.324 D_GP: 0.157 D_real: 0.590 D_fake: 0.488 +(epoch: 234, iters: 6736, time: 0.064) G_GAN: 0.385 G_GAN_Feat: 0.859 G_ID: 0.185 G_Rec: 0.413 D_GP: 0.050 D_real: 1.168 D_fake: 0.662 +(epoch: 234, iters: 7136, time: 0.064) G_GAN: 0.287 G_GAN_Feat: 0.731 G_ID: 0.115 G_Rec: 0.317 D_GP: 0.024 D_real: 1.142 D_fake: 0.713 +(epoch: 234, iters: 7536, time: 0.064) G_GAN: 0.068 G_GAN_Feat: 0.877 G_ID: 0.165 G_Rec: 0.396 D_GP: 0.040 D_real: 0.797 D_fake: 0.933 +(epoch: 234, iters: 7936, time: 0.064) G_GAN: 0.121 G_GAN_Feat: 0.625 G_ID: 0.153 G_Rec: 0.298 D_GP: 0.024 D_real: 1.026 D_fake: 0.879 +(epoch: 234, iters: 8336, time: 0.064) G_GAN: 0.439 G_GAN_Feat: 0.821 G_ID: 0.179 G_Rec: 0.381 D_GP: 0.024 D_real: 1.254 D_fake: 0.567 +(epoch: 235, iters: 128, time: 0.064) G_GAN: 0.114 G_GAN_Feat: 0.670 G_ID: 0.128 G_Rec: 0.291 D_GP: 0.035 D_real: 0.978 D_fake: 0.886 +(epoch: 235, iters: 528, time: 0.064) G_GAN: 0.754 G_GAN_Feat: 1.018 G_ID: 0.176 G_Rec: 0.402 D_GP: 0.059 D_real: 1.222 D_fake: 0.315 +(epoch: 235, iters: 928, time: 0.064) G_GAN: 0.350 G_GAN_Feat: 0.813 G_ID: 0.138 G_Rec: 0.318 D_GP: 0.032 D_real: 1.006 D_fake: 0.657 +(epoch: 235, iters: 1328, time: 0.064) G_GAN: 0.477 G_GAN_Feat: 1.059 G_ID: 0.200 G_Rec: 0.434 D_GP: 0.069 D_real: 0.537 D_fake: 0.528 +(epoch: 235, iters: 1728, time: 0.064) G_GAN: 0.486 G_GAN_Feat: 0.933 G_ID: 0.122 G_Rec: 0.335 D_GP: 0.035 D_real: 0.925 D_fake: 0.516 +(epoch: 235, iters: 2128, time: 0.064) G_GAN: 0.393 G_GAN_Feat: 0.848 G_ID: 0.168 G_Rec: 0.364 D_GP: 0.024 D_real: 1.089 D_fake: 0.608 +(epoch: 235, iters: 2528, time: 0.064) G_GAN: 0.186 G_GAN_Feat: 0.781 G_ID: 0.123 G_Rec: 0.331 D_GP: 0.050 D_real: 0.828 D_fake: 0.817 +(epoch: 235, iters: 2928, time: 0.064) G_GAN: 0.458 G_GAN_Feat: 0.894 G_ID: 0.178 G_Rec: 0.385 D_GP: 0.036 D_real: 1.134 D_fake: 0.548 +(epoch: 235, iters: 3328, time: 0.064) G_GAN: 0.028 G_GAN_Feat: 0.790 G_ID: 0.156 G_Rec: 0.324 D_GP: 0.041 D_real: 0.996 D_fake: 0.973 +(epoch: 235, iters: 3728, time: 0.064) G_GAN: 0.296 G_GAN_Feat: 1.083 G_ID: 0.176 G_Rec: 0.412 D_GP: 0.100 D_real: 0.804 D_fake: 0.707 +(epoch: 235, iters: 4128, time: 0.064) G_GAN: 0.372 G_GAN_Feat: 0.663 G_ID: 0.131 G_Rec: 0.314 D_GP: 0.024 D_real: 1.352 D_fake: 0.628 +(epoch: 235, iters: 4528, time: 0.064) G_GAN: 0.270 G_GAN_Feat: 0.909 G_ID: 0.177 G_Rec: 0.410 D_GP: 0.024 D_real: 1.064 D_fake: 0.731 +(epoch: 235, iters: 4928, time: 0.064) G_GAN: 0.073 G_GAN_Feat: 0.670 G_ID: 0.114 G_Rec: 0.318 D_GP: 0.116 D_real: 0.825 D_fake: 0.927 +(epoch: 235, iters: 5328, time: 0.064) G_GAN: 0.420 G_GAN_Feat: 0.861 G_ID: 0.164 G_Rec: 0.420 D_GP: 0.022 D_real: 1.157 D_fake: 0.582 +(epoch: 235, iters: 5728, time: 0.064) G_GAN: 0.112 G_GAN_Feat: 0.674 G_ID: 0.154 G_Rec: 0.332 D_GP: 0.043 D_real: 0.939 D_fake: 0.888 +(epoch: 235, iters: 6128, time: 0.064) G_GAN: -0.043 G_GAN_Feat: 1.048 G_ID: 0.172 G_Rec: 0.396 D_GP: 0.305 D_real: 0.430 D_fake: 1.043 +(epoch: 235, iters: 6528, time: 0.064) G_GAN: 0.289 G_GAN_Feat: 0.734 G_ID: 0.136 G_Rec: 0.288 D_GP: 0.039 D_real: 1.120 D_fake: 0.712 +(epoch: 235, iters: 6928, time: 0.064) G_GAN: 0.361 G_GAN_Feat: 0.834 G_ID: 0.177 G_Rec: 0.366 D_GP: 0.027 D_real: 1.115 D_fake: 0.649 +(epoch: 235, iters: 7328, time: 0.064) G_GAN: 0.290 G_GAN_Feat: 0.671 G_ID: 0.122 G_Rec: 0.296 D_GP: 0.033 D_real: 1.149 D_fake: 0.711 +(epoch: 235, iters: 7728, time: 0.064) G_GAN: 0.217 G_GAN_Feat: 0.858 G_ID: 0.170 G_Rec: 0.386 D_GP: 0.025 D_real: 0.891 D_fake: 0.783 +(epoch: 235, iters: 8128, time: 0.064) G_GAN: 0.195 G_GAN_Feat: 0.633 G_ID: 0.119 G_Rec: 0.284 D_GP: 0.022 D_real: 1.070 D_fake: 0.805 +(epoch: 235, iters: 8528, time: 0.064) G_GAN: 0.434 G_GAN_Feat: 1.078 G_ID: 0.185 G_Rec: 0.436 D_GP: 0.153 D_real: 0.542 D_fake: 0.569 +(epoch: 236, iters: 320, time: 0.064) G_GAN: 0.364 G_GAN_Feat: 0.788 G_ID: 0.121 G_Rec: 0.287 D_GP: 0.049 D_real: 0.944 D_fake: 0.637 +(epoch: 236, iters: 720, time: 0.064) G_GAN: 0.799 G_GAN_Feat: 0.943 G_ID: 0.186 G_Rec: 0.420 D_GP: 0.027 D_real: 1.425 D_fake: 0.246 +(epoch: 236, iters: 1120, time: 0.064) G_GAN: 0.233 G_GAN_Feat: 0.687 G_ID: 0.136 G_Rec: 0.289 D_GP: 0.024 D_real: 1.157 D_fake: 0.767 +(epoch: 236, iters: 1520, time: 0.064) G_GAN: 0.407 G_GAN_Feat: 1.033 G_ID: 0.174 G_Rec: 0.434 D_GP: 0.080 D_real: 0.516 D_fake: 0.604 +(epoch: 236, iters: 1920, time: 0.064) G_GAN: 0.098 G_GAN_Feat: 0.709 G_ID: 0.154 G_Rec: 0.291 D_GP: 0.022 D_real: 1.042 D_fake: 0.902 +(epoch: 236, iters: 2320, time: 0.064) G_GAN: 0.440 G_GAN_Feat: 1.039 G_ID: 0.168 G_Rec: 0.405 D_GP: 0.026 D_real: 0.862 D_fake: 0.565 +(epoch: 236, iters: 2720, time: 0.064) G_GAN: 0.177 G_GAN_Feat: 0.734 G_ID: 0.139 G_Rec: 0.346 D_GP: 0.028 D_real: 1.057 D_fake: 0.824 +(epoch: 236, iters: 3120, time: 0.064) G_GAN: 0.565 G_GAN_Feat: 1.226 G_ID: 0.186 G_Rec: 0.498 D_GP: 0.044 D_real: 0.292 D_fake: 0.453 +(epoch: 236, iters: 3520, time: 0.064) G_GAN: 0.237 G_GAN_Feat: 1.003 G_ID: 0.138 G_Rec: 0.351 D_GP: 0.041 D_real: 0.410 D_fake: 0.764 +(epoch: 236, iters: 3920, time: 0.064) G_GAN: 0.246 G_GAN_Feat: 0.804 G_ID: 0.161 G_Rec: 0.411 D_GP: 0.020 D_real: 1.031 D_fake: 0.754 +(epoch: 236, iters: 4320, time: 0.064) G_GAN: 0.081 G_GAN_Feat: 0.691 G_ID: 0.144 G_Rec: 0.295 D_GP: 0.034 D_real: 0.877 D_fake: 0.919 +(epoch: 236, iters: 4720, time: 0.064) G_GAN: 0.263 G_GAN_Feat: 0.947 G_ID: 0.161 G_Rec: 0.420 D_GP: 0.033 D_real: 1.164 D_fake: 0.738 +(epoch: 236, iters: 5120, time: 0.064) G_GAN: 0.141 G_GAN_Feat: 0.620 G_ID: 0.113 G_Rec: 0.302 D_GP: 0.019 D_real: 1.143 D_fake: 0.859 +(epoch: 236, iters: 5520, time: 0.064) G_GAN: 0.409 G_GAN_Feat: 0.862 G_ID: 0.154 G_Rec: 0.382 D_GP: 0.024 D_real: 1.158 D_fake: 0.594 +(epoch: 236, iters: 5920, time: 0.064) G_GAN: 0.097 G_GAN_Feat: 0.882 G_ID: 0.123 G_Rec: 0.313 D_GP: 0.066 D_real: 0.480 D_fake: 0.903 +(epoch: 236, iters: 6320, time: 0.064) G_GAN: 0.356 G_GAN_Feat: 0.887 G_ID: 0.200 G_Rec: 0.407 D_GP: 0.023 D_real: 1.097 D_fake: 0.644 +(epoch: 236, iters: 6720, time: 0.064) G_GAN: 0.352 G_GAN_Feat: 0.780 G_ID: 0.126 G_Rec: 0.303 D_GP: 0.027 D_real: 1.135 D_fake: 0.648 +(epoch: 236, iters: 7120, time: 0.064) G_GAN: 0.527 G_GAN_Feat: 0.897 G_ID: 0.168 G_Rec: 0.405 D_GP: 0.029 D_real: 1.173 D_fake: 0.479 +(epoch: 236, iters: 7520, time: 0.064) G_GAN: 0.322 G_GAN_Feat: 0.721 G_ID: 0.118 G_Rec: 0.294 D_GP: 0.040 D_real: 1.154 D_fake: 0.679 +(epoch: 236, iters: 7920, time: 0.064) G_GAN: 0.556 G_GAN_Feat: 0.865 G_ID: 0.192 G_Rec: 0.402 D_GP: 0.027 D_real: 1.231 D_fake: 0.451 +(epoch: 236, iters: 8320, time: 0.064) G_GAN: 0.184 G_GAN_Feat: 0.762 G_ID: 0.131 G_Rec: 0.315 D_GP: 0.023 D_real: 0.977 D_fake: 0.816 +(epoch: 237, iters: 112, time: 0.064) G_GAN: 0.504 G_GAN_Feat: 0.941 G_ID: 0.189 G_Rec: 0.370 D_GP: 0.085 D_real: 0.743 D_fake: 0.500 +(epoch: 237, iters: 512, time: 0.064) G_GAN: 0.295 G_GAN_Feat: 0.961 G_ID: 0.116 G_Rec: 0.296 D_GP: 0.394 D_real: 0.523 D_fake: 0.705 +(epoch: 237, iters: 912, time: 0.064) G_GAN: 0.312 G_GAN_Feat: 1.026 G_ID: 0.158 G_Rec: 0.398 D_GP: 0.069 D_real: 0.564 D_fake: 0.690 +(epoch: 237, iters: 1312, time: 0.064) G_GAN: 0.334 G_GAN_Feat: 0.755 G_ID: 0.151 G_Rec: 0.292 D_GP: 0.031 D_real: 1.116 D_fake: 0.668 +(epoch: 237, iters: 1712, time: 0.064) G_GAN: 0.451 G_GAN_Feat: 0.728 G_ID: 0.157 G_Rec: 0.422 D_GP: 0.019 D_real: 1.298 D_fake: 0.550 +(epoch: 237, iters: 2112, time: 0.064) G_GAN: 0.331 G_GAN_Feat: 0.557 G_ID: 0.117 G_Rec: 0.269 D_GP: 0.019 D_real: 1.303 D_fake: 0.669 +(epoch: 237, iters: 2512, time: 0.064) G_GAN: 0.339 G_GAN_Feat: 0.953 G_ID: 0.166 G_Rec: 0.448 D_GP: 0.036 D_real: 0.949 D_fake: 0.663 +(epoch: 237, iters: 2912, time: 0.064) G_GAN: 0.174 G_GAN_Feat: 0.798 G_ID: 0.130 G_Rec: 0.307 D_GP: 0.026 D_real: 0.835 D_fake: 0.826 +(epoch: 237, iters: 3312, time: 0.064) G_GAN: 0.216 G_GAN_Feat: 0.774 G_ID: 0.188 G_Rec: 0.376 D_GP: 0.021 D_real: 1.032 D_fake: 0.784 +(epoch: 237, iters: 3712, time: 0.064) G_GAN: 0.210 G_GAN_Feat: 0.710 G_ID: 0.128 G_Rec: 0.308 D_GP: 0.064 D_real: 0.951 D_fake: 0.790 +(epoch: 237, iters: 4112, time: 0.064) G_GAN: 0.409 G_GAN_Feat: 0.876 G_ID: 0.150 G_Rec: 0.384 D_GP: 0.035 D_real: 1.041 D_fake: 0.593 +(epoch: 237, iters: 4512, time: 0.064) G_GAN: 0.205 G_GAN_Feat: 0.719 G_ID: 0.122 G_Rec: 0.301 D_GP: 0.024 D_real: 1.123 D_fake: 0.796 +(epoch: 237, iters: 4912, time: 0.064) G_GAN: 0.610 G_GAN_Feat: 0.833 G_ID: 0.149 G_Rec: 0.446 D_GP: 0.020 D_real: 1.357 D_fake: 0.398 +(epoch: 237, iters: 5312, time: 0.064) G_GAN: 0.286 G_GAN_Feat: 0.619 G_ID: 0.106 G_Rec: 0.292 D_GP: 0.021 D_real: 1.336 D_fake: 0.714 +(epoch: 237, iters: 5712, time: 0.064) G_GAN: 0.375 G_GAN_Feat: 0.804 G_ID: 0.189 G_Rec: 0.396 D_GP: 0.024 D_real: 1.128 D_fake: 0.627 +(epoch: 237, iters: 6112, time: 0.064) G_GAN: 0.221 G_GAN_Feat: 0.615 G_ID: 0.137 G_Rec: 0.313 D_GP: 0.028 D_real: 1.141 D_fake: 0.780 +(epoch: 237, iters: 6512, time: 0.064) G_GAN: 0.363 G_GAN_Feat: 0.750 G_ID: 0.165 G_Rec: 0.358 D_GP: 0.030 D_real: 1.137 D_fake: 0.637 +(epoch: 237, iters: 6912, time: 0.064) G_GAN: 0.084 G_GAN_Feat: 0.648 G_ID: 0.143 G_Rec: 0.294 D_GP: 0.035 D_real: 0.930 D_fake: 0.916 +(epoch: 237, iters: 7312, time: 0.064) G_GAN: 0.417 G_GAN_Feat: 0.855 G_ID: 0.200 G_Rec: 0.395 D_GP: 0.104 D_real: 1.030 D_fake: 0.592 +(epoch: 237, iters: 7712, time: 0.064) G_GAN: 0.164 G_GAN_Feat: 0.705 G_ID: 0.135 G_Rec: 0.302 D_GP: 0.039 D_real: 0.921 D_fake: 0.836 +(epoch: 237, iters: 8112, time: 0.064) G_GAN: 0.329 G_GAN_Feat: 0.882 G_ID: 0.173 G_Rec: 0.377 D_GP: 0.098 D_real: 0.844 D_fake: 0.673 +(epoch: 237, iters: 8512, time: 0.064) G_GAN: -0.053 G_GAN_Feat: 0.684 G_ID: 0.156 G_Rec: 0.292 D_GP: 0.033 D_real: 0.855 D_fake: 1.053 +(epoch: 238, iters: 304, time: 0.064) G_GAN: 0.431 G_GAN_Feat: 0.787 G_ID: 0.163 G_Rec: 0.345 D_GP: 0.036 D_real: 1.160 D_fake: 0.574 +(epoch: 238, iters: 704, time: 0.064) G_GAN: 0.122 G_GAN_Feat: 0.680 G_ID: 0.137 G_Rec: 0.289 D_GP: 0.026 D_real: 0.982 D_fake: 0.878 +(epoch: 238, iters: 1104, time: 0.064) G_GAN: 0.280 G_GAN_Feat: 0.810 G_ID: 0.186 G_Rec: 0.366 D_GP: 0.033 D_real: 1.053 D_fake: 0.721 +(epoch: 238, iters: 1504, time: 0.064) G_GAN: 0.186 G_GAN_Feat: 0.706 G_ID: 0.143 G_Rec: 0.324 D_GP: 0.027 D_real: 1.059 D_fake: 0.814 +(epoch: 238, iters: 1904, time: 0.064) G_GAN: 0.553 G_GAN_Feat: 1.064 G_ID: 0.162 G_Rec: 0.422 D_GP: 0.121 D_real: 0.578 D_fake: 0.472 +(epoch: 238, iters: 2304, time: 0.064) G_GAN: 0.375 G_GAN_Feat: 0.613 G_ID: 0.109 G_Rec: 0.284 D_GP: 0.019 D_real: 1.317 D_fake: 0.626 +(epoch: 238, iters: 2704, time: 0.064) G_GAN: 0.451 G_GAN_Feat: 0.811 G_ID: 0.173 G_Rec: 0.377 D_GP: 0.028 D_real: 1.151 D_fake: 0.553 +(epoch: 238, iters: 3104, time: 0.064) G_GAN: 0.134 G_GAN_Feat: 0.697 G_ID: 0.125 G_Rec: 0.296 D_GP: 0.059 D_real: 0.989 D_fake: 0.866 +(epoch: 238, iters: 3504, time: 0.064) G_GAN: 0.360 G_GAN_Feat: 0.784 G_ID: 0.160 G_Rec: 0.363 D_GP: 0.023 D_real: 1.157 D_fake: 0.641 +(epoch: 238, iters: 3904, time: 0.064) G_GAN: 0.171 G_GAN_Feat: 0.676 G_ID: 0.123 G_Rec: 0.296 D_GP: 0.028 D_real: 1.003 D_fake: 0.829 +(epoch: 238, iters: 4304, time: 0.064) G_GAN: 0.341 G_GAN_Feat: 0.819 G_ID: 0.199 G_Rec: 0.358 D_GP: 0.026 D_real: 1.089 D_fake: 0.661 +(epoch: 238, iters: 4704, time: 0.064) G_GAN: 0.414 G_GAN_Feat: 0.721 G_ID: 0.136 G_Rec: 0.298 D_GP: 0.041 D_real: 1.179 D_fake: 0.589 +(epoch: 238, iters: 5104, time: 0.064) G_GAN: 0.461 G_GAN_Feat: 0.765 G_ID: 0.154 G_Rec: 0.326 D_GP: 0.026 D_real: 1.267 D_fake: 0.540 +(epoch: 238, iters: 5504, time: 0.064) G_GAN: 0.128 G_GAN_Feat: 0.725 G_ID: 0.127 G_Rec: 0.302 D_GP: 0.022 D_real: 1.018 D_fake: 0.872 +(epoch: 238, iters: 5904, time: 0.064) G_GAN: 0.567 G_GAN_Feat: 0.927 G_ID: 0.208 G_Rec: 0.402 D_GP: 0.031 D_real: 1.127 D_fake: 0.449 +(epoch: 238, iters: 6304, time: 0.064) G_GAN: 0.398 G_GAN_Feat: 0.842 G_ID: 0.122 G_Rec: 0.299 D_GP: 0.065 D_real: 0.705 D_fake: 0.604 +(epoch: 238, iters: 6704, time: 0.064) G_GAN: 0.348 G_GAN_Feat: 0.747 G_ID: 0.195 G_Rec: 0.399 D_GP: 0.019 D_real: 1.182 D_fake: 0.653 +(epoch: 238, iters: 7104, time: 0.064) G_GAN: 0.125 G_GAN_Feat: 0.657 G_ID: 0.179 G_Rec: 0.357 D_GP: 0.022 D_real: 1.079 D_fake: 0.876 +(epoch: 238, iters: 7504, time: 0.064) G_GAN: 0.339 G_GAN_Feat: 0.766 G_ID: 0.166 G_Rec: 0.375 D_GP: 0.022 D_real: 1.099 D_fake: 0.661 +(epoch: 238, iters: 7904, time: 0.064) G_GAN: 0.069 G_GAN_Feat: 0.575 G_ID: 0.126 G_Rec: 0.276 D_GP: 0.024 D_real: 1.004 D_fake: 0.931 +(epoch: 238, iters: 8304, time: 0.064) G_GAN: 0.331 G_GAN_Feat: 0.940 G_ID: 0.187 G_Rec: 0.441 D_GP: 0.044 D_real: 0.653 D_fake: 0.671 +(epoch: 239, iters: 96, time: 0.064) G_GAN: 0.117 G_GAN_Feat: 0.839 G_ID: 0.126 G_Rec: 0.311 D_GP: 0.216 D_real: 0.646 D_fake: 0.883 +(epoch: 239, iters: 496, time: 0.064) G_GAN: 0.079 G_GAN_Feat: 0.965 G_ID: 0.193 G_Rec: 0.367 D_GP: 0.079 D_real: 0.306 D_fake: 0.921 +(epoch: 239, iters: 896, time: 0.064) G_GAN: 0.075 G_GAN_Feat: 0.640 G_ID: 0.140 G_Rec: 0.285 D_GP: 0.021 D_real: 1.012 D_fake: 0.925 +(epoch: 239, iters: 1296, time: 0.064) G_GAN: 0.357 G_GAN_Feat: 0.821 G_ID: 0.175 G_Rec: 0.405 D_GP: 0.032 D_real: 1.038 D_fake: 0.647 +(epoch: 239, iters: 1696, time: 0.064) G_GAN: 0.023 G_GAN_Feat: 0.657 G_ID: 0.127 G_Rec: 0.292 D_GP: 0.024 D_real: 0.908 D_fake: 0.977 +(epoch: 239, iters: 2096, time: 0.064) G_GAN: 0.610 G_GAN_Feat: 0.848 G_ID: 0.174 G_Rec: 0.387 D_GP: 0.038 D_real: 1.152 D_fake: 0.425 +(epoch: 239, iters: 2496, time: 0.064) G_GAN: 0.285 G_GAN_Feat: 0.776 G_ID: 0.115 G_Rec: 0.320 D_GP: 0.042 D_real: 1.153 D_fake: 0.722 +(epoch: 239, iters: 2896, time: 0.064) G_GAN: 0.401 G_GAN_Feat: 1.128 G_ID: 0.176 G_Rec: 0.440 D_GP: 0.061 D_real: 0.416 D_fake: 0.601 +(epoch: 239, iters: 3296, time: 0.064) G_GAN: 0.380 G_GAN_Feat: 0.738 G_ID: 0.134 G_Rec: 0.300 D_GP: 0.056 D_real: 1.105 D_fake: 0.622 +(epoch: 239, iters: 3696, time: 0.064) G_GAN: 0.198 G_GAN_Feat: 0.826 G_ID: 0.215 G_Rec: 0.369 D_GP: 0.027 D_real: 1.034 D_fake: 0.803 +(epoch: 239, iters: 4096, time: 0.064) G_GAN: 0.294 G_GAN_Feat: 0.833 G_ID: 0.111 G_Rec: 0.311 D_GP: 0.031 D_real: 0.877 D_fake: 0.708 +(epoch: 239, iters: 4496, time: 0.064) G_GAN: 0.053 G_GAN_Feat: 0.832 G_ID: 0.158 G_Rec: 0.420 D_GP: 0.021 D_real: 0.842 D_fake: 0.947 +(epoch: 239, iters: 4896, time: 0.064) G_GAN: 0.202 G_GAN_Feat: 0.691 G_ID: 0.133 G_Rec: 0.307 D_GP: 0.022 D_real: 1.133 D_fake: 0.798 +(epoch: 239, iters: 5296, time: 0.064) G_GAN: 0.540 G_GAN_Feat: 1.122 G_ID: 0.164 G_Rec: 0.465 D_GP: 0.192 D_real: 0.372 D_fake: 0.511 +(epoch: 239, iters: 5696, time: 0.064) G_GAN: 0.242 G_GAN_Feat: 0.896 G_ID: 0.127 G_Rec: 0.342 D_GP: 0.057 D_real: 0.528 D_fake: 0.769 +(epoch: 239, iters: 6096, time: 0.064) G_GAN: 0.441 G_GAN_Feat: 0.885 G_ID: 0.167 G_Rec: 0.409 D_GP: 0.022 D_real: 1.182 D_fake: 0.565 +(epoch: 239, iters: 6496, time: 0.064) G_GAN: 0.177 G_GAN_Feat: 0.656 G_ID: 0.154 G_Rec: 0.307 D_GP: 0.019 D_real: 1.105 D_fake: 0.823 +(epoch: 239, iters: 6896, time: 0.064) G_GAN: 0.369 G_GAN_Feat: 0.769 G_ID: 0.174 G_Rec: 0.415 D_GP: 0.019 D_real: 1.141 D_fake: 0.632 +(epoch: 239, iters: 7296, time: 0.064) G_GAN: 0.240 G_GAN_Feat: 0.537 G_ID: 0.131 G_Rec: 0.285 D_GP: 0.016 D_real: 1.191 D_fake: 0.761 +(epoch: 239, iters: 7696, time: 0.064) G_GAN: 0.468 G_GAN_Feat: 0.781 G_ID: 0.184 G_Rec: 0.410 D_GP: 0.025 D_real: 1.233 D_fake: 0.542 +(epoch: 239, iters: 8096, time: 0.064) G_GAN: 0.201 G_GAN_Feat: 0.597 G_ID: 0.130 G_Rec: 0.288 D_GP: 0.024 D_real: 1.082 D_fake: 0.800 +(epoch: 239, iters: 8496, time: 0.064) G_GAN: 0.359 G_GAN_Feat: 0.775 G_ID: 0.172 G_Rec: 0.362 D_GP: 0.029 D_real: 1.124 D_fake: 0.641 +(epoch: 240, iters: 288, time: 0.064) G_GAN: 0.182 G_GAN_Feat: 0.614 G_ID: 0.123 G_Rec: 0.273 D_GP: 0.025 D_real: 1.116 D_fake: 0.818 +(epoch: 240, iters: 688, time: 0.064) G_GAN: 0.429 G_GAN_Feat: 0.943 G_ID: 0.175 G_Rec: 0.412 D_GP: 0.037 D_real: 0.933 D_fake: 0.574 +(epoch: 240, iters: 1088, time: 0.064) G_GAN: 0.052 G_GAN_Feat: 0.732 G_ID: 0.144 G_Rec: 0.327 D_GP: 0.043 D_real: 0.801 D_fake: 0.948 +(epoch: 240, iters: 1488, time: 0.064) G_GAN: 0.491 G_GAN_Feat: 0.811 G_ID: 0.169 G_Rec: 0.355 D_GP: 0.022 D_real: 1.336 D_fake: 0.513 +(epoch: 240, iters: 1888, time: 0.064) G_GAN: 0.182 G_GAN_Feat: 0.710 G_ID: 0.146 G_Rec: 0.333 D_GP: 0.029 D_real: 0.975 D_fake: 0.819 +(epoch: 240, iters: 2288, time: 0.064) G_GAN: 0.378 G_GAN_Feat: 1.051 G_ID: 0.182 G_Rec: 0.429 D_GP: 0.155 D_real: 0.514 D_fake: 0.640 +(epoch: 240, iters: 2688, time: 0.064) G_GAN: 0.292 G_GAN_Feat: 0.722 G_ID: 0.126 G_Rec: 0.314 D_GP: 0.021 D_real: 1.128 D_fake: 0.710 +(epoch: 240, iters: 3088, time: 0.064) G_GAN: 0.312 G_GAN_Feat: 0.784 G_ID: 0.185 G_Rec: 0.372 D_GP: 0.034 D_real: 1.069 D_fake: 0.690 +(epoch: 240, iters: 3488, time: 0.064) G_GAN: 0.070 G_GAN_Feat: 0.730 G_ID: 0.118 G_Rec: 0.271 D_GP: 0.042 D_real: 0.800 D_fake: 0.930 +(epoch: 240, iters: 3888, time: 0.064) G_GAN: 0.457 G_GAN_Feat: 0.882 G_ID: 0.205 G_Rec: 0.378 D_GP: 0.103 D_real: 0.809 D_fake: 0.548 +(epoch: 240, iters: 4288, time: 0.064) G_GAN: 0.258 G_GAN_Feat: 0.923 G_ID: 0.123 G_Rec: 0.320 D_GP: 0.129 D_real: 0.514 D_fake: 0.744 +(epoch: 240, iters: 4688, time: 0.064) G_GAN: 0.431 G_GAN_Feat: 0.845 G_ID: 0.168 G_Rec: 0.367 D_GP: 0.022 D_real: 1.207 D_fake: 0.573 +(epoch: 240, iters: 5088, time: 0.064) G_GAN: -0.036 G_GAN_Feat: 0.680 G_ID: 0.146 G_Rec: 0.298 D_GP: 0.020 D_real: 0.913 D_fake: 1.036 +(epoch: 240, iters: 5488, time: 0.064) G_GAN: 0.645 G_GAN_Feat: 0.819 G_ID: 0.163 G_Rec: 0.432 D_GP: 0.027 D_real: 1.316 D_fake: 0.366 +(epoch: 240, iters: 5888, time: 0.064) G_GAN: 0.132 G_GAN_Feat: 0.650 G_ID: 0.143 G_Rec: 0.291 D_GP: 0.021 D_real: 1.073 D_fake: 0.869 +(epoch: 240, iters: 6288, time: 0.064) G_GAN: 0.459 G_GAN_Feat: 0.931 G_ID: 0.153 G_Rec: 0.410 D_GP: 0.063 D_real: 0.977 D_fake: 0.544 +(epoch: 240, iters: 6688, time: 0.064) G_GAN: 0.113 G_GAN_Feat: 0.808 G_ID: 0.157 G_Rec: 0.337 D_GP: 0.083 D_real: 0.524 D_fake: 0.888 +(epoch: 240, iters: 7088, time: 0.064) G_GAN: 0.389 G_GAN_Feat: 0.950 G_ID: 0.175 G_Rec: 0.366 D_GP: 0.098 D_real: 0.522 D_fake: 0.612 +(epoch: 240, iters: 7488, time: 0.064) G_GAN: 0.236 G_GAN_Feat: 0.726 G_ID: 0.129 G_Rec: 0.298 D_GP: 0.026 D_real: 1.073 D_fake: 0.765 +(epoch: 240, iters: 7888, time: 0.064) G_GAN: 0.240 G_GAN_Feat: 0.927 G_ID: 0.185 G_Rec: 0.390 D_GP: 0.112 D_real: 0.798 D_fake: 0.762 +(epoch: 240, iters: 8288, time: 0.064) G_GAN: 0.489 G_GAN_Feat: 0.761 G_ID: 0.139 G_Rec: 0.309 D_GP: 0.025 D_real: 1.402 D_fake: 0.524 +(epoch: 241, iters: 80, time: 0.064) G_GAN: 0.117 G_GAN_Feat: 1.083 G_ID: 0.187 G_Rec: 0.426 D_GP: 0.943 D_real: 0.430 D_fake: 0.889 +(epoch: 241, iters: 480, time: 0.064) G_GAN: 0.205 G_GAN_Feat: 0.687 G_ID: 0.110 G_Rec: 0.290 D_GP: 0.021 D_real: 1.105 D_fake: 0.795 +(epoch: 241, iters: 880, time: 0.064) G_GAN: 0.448 G_GAN_Feat: 0.945 G_ID: 0.170 G_Rec: 0.410 D_GP: 0.023 D_real: 1.086 D_fake: 0.553 +(epoch: 241, iters: 1280, time: 0.064) G_GAN: 0.188 G_GAN_Feat: 0.827 G_ID: 0.147 G_Rec: 0.339 D_GP: 0.088 D_real: 0.568 D_fake: 0.814 +(epoch: 241, iters: 1680, time: 0.064) G_GAN: 0.502 G_GAN_Feat: 0.979 G_ID: 0.191 G_Rec: 0.378 D_GP: 0.036 D_real: 0.856 D_fake: 0.508 +(epoch: 241, iters: 2080, time: 0.064) G_GAN: 0.269 G_GAN_Feat: 0.733 G_ID: 0.134 G_Rec: 0.287 D_GP: 0.035 D_real: 1.049 D_fake: 0.733 +(epoch: 241, iters: 2480, time: 0.064) G_GAN: 0.332 G_GAN_Feat: 0.903 G_ID: 0.163 G_Rec: 0.407 D_GP: 0.104 D_real: 0.819 D_fake: 0.691 +(epoch: 241, iters: 2880, time: 0.064) G_GAN: -0.060 G_GAN_Feat: 0.719 G_ID: 0.140 G_Rec: 0.297 D_GP: 0.032 D_real: 0.749 D_fake: 1.060 +(epoch: 241, iters: 3280, time: 0.064) G_GAN: 0.814 G_GAN_Feat: 1.002 G_ID: 0.160 G_Rec: 0.441 D_GP: 0.039 D_real: 1.226 D_fake: 0.256 +(epoch: 241, iters: 3680, time: 0.064) G_GAN: 0.177 G_GAN_Feat: 0.845 G_ID: 0.124 G_Rec: 0.319 D_GP: 0.068 D_real: 0.784 D_fake: 0.824 +(epoch: 241, iters: 4080, time: 0.064) G_GAN: 0.373 G_GAN_Feat: 0.749 G_ID: 0.173 G_Rec: 0.368 D_GP: 0.018 D_real: 1.359 D_fake: 0.634 +(epoch: 241, iters: 4480, time: 0.064) G_GAN: 0.227 G_GAN_Feat: 0.619 G_ID: 0.124 G_Rec: 0.318 D_GP: 0.022 D_real: 1.187 D_fake: 0.774 +(epoch: 241, iters: 4880, time: 0.064) G_GAN: 0.463 G_GAN_Feat: 0.697 G_ID: 0.129 G_Rec: 0.346 D_GP: 0.015 D_real: 1.326 D_fake: 0.539 +(epoch: 241, iters: 5280, time: 0.064) G_GAN: 0.161 G_GAN_Feat: 0.649 G_ID: 0.126 G_Rec: 0.319 D_GP: 0.036 D_real: 1.064 D_fake: 0.839 +(epoch: 241, iters: 5680, time: 0.064) G_GAN: 0.455 G_GAN_Feat: 0.782 G_ID: 0.168 G_Rec: 0.380 D_GP: 0.024 D_real: 1.198 D_fake: 0.554 +(epoch: 241, iters: 6080, time: 0.064) G_GAN: 0.049 G_GAN_Feat: 0.633 G_ID: 0.133 G_Rec: 0.288 D_GP: 0.044 D_real: 0.923 D_fake: 0.951 +(epoch: 241, iters: 6480, time: 0.064) G_GAN: 0.370 G_GAN_Feat: 0.732 G_ID: 0.172 G_Rec: 0.364 D_GP: 0.020 D_real: 1.175 D_fake: 0.631 +(epoch: 241, iters: 6880, time: 0.064) G_GAN: 0.145 G_GAN_Feat: 0.622 G_ID: 0.112 G_Rec: 0.306 D_GP: 0.039 D_real: 1.018 D_fake: 0.855 +(epoch: 241, iters: 7280, time: 0.064) G_GAN: 0.392 G_GAN_Feat: 0.924 G_ID: 0.163 G_Rec: 0.411 D_GP: 0.026 D_real: 1.018 D_fake: 0.610 +(epoch: 241, iters: 7680, time: 0.064) G_GAN: 0.203 G_GAN_Feat: 0.592 G_ID: 0.132 G_Rec: 0.263 D_GP: 0.022 D_real: 1.119 D_fake: 0.798 +(epoch: 241, iters: 8080, time: 0.064) G_GAN: 0.344 G_GAN_Feat: 0.872 G_ID: 0.154 G_Rec: 0.436 D_GP: 0.041 D_real: 1.032 D_fake: 0.658 +(epoch: 241, iters: 8480, time: 0.064) G_GAN: 0.048 G_GAN_Feat: 0.690 G_ID: 0.106 G_Rec: 0.293 D_GP: 0.128 D_real: 0.817 D_fake: 0.952 +(epoch: 242, iters: 272, time: 0.064) G_GAN: 0.213 G_GAN_Feat: 0.777 G_ID: 0.152 G_Rec: 0.378 D_GP: 0.022 D_real: 1.000 D_fake: 0.788 +(epoch: 242, iters: 672, time: 0.064) G_GAN: 0.097 G_GAN_Feat: 0.635 G_ID: 0.107 G_Rec: 0.319 D_GP: 0.028 D_real: 1.095 D_fake: 0.903 +(epoch: 242, iters: 1072, time: 0.064) G_GAN: 0.348 G_GAN_Feat: 0.815 G_ID: 0.155 G_Rec: 0.391 D_GP: 0.027 D_real: 1.101 D_fake: 0.654 +(epoch: 242, iters: 1472, time: 0.064) G_GAN: 0.115 G_GAN_Feat: 0.818 G_ID: 0.118 G_Rec: 0.322 D_GP: 0.457 D_real: 0.715 D_fake: 0.886 +(epoch: 242, iters: 1872, time: 0.064) G_GAN: 0.239 G_GAN_Feat: 0.813 G_ID: 0.166 G_Rec: 0.360 D_GP: 0.023 D_real: 0.931 D_fake: 0.761 +(epoch: 242, iters: 2272, time: 0.064) G_GAN: 0.239 G_GAN_Feat: 0.683 G_ID: 0.125 G_Rec: 0.298 D_GP: 0.034 D_real: 1.034 D_fake: 0.761 +(epoch: 242, iters: 2672, time: 0.064) G_GAN: 0.386 G_GAN_Feat: 0.922 G_ID: 0.187 G_Rec: 0.425 D_GP: 0.028 D_real: 1.065 D_fake: 0.618 +(epoch: 242, iters: 3072, time: 0.064) G_GAN: -0.087 G_GAN_Feat: 0.760 G_ID: 0.133 G_Rec: 0.301 D_GP: 0.102 D_real: 0.521 D_fake: 1.087 +(epoch: 242, iters: 3472, time: 0.064) G_GAN: 0.395 G_GAN_Feat: 0.756 G_ID: 0.174 G_Rec: 0.356 D_GP: 0.022 D_real: 1.235 D_fake: 0.606 +(epoch: 242, iters: 3872, time: 0.064) G_GAN: 0.021 G_GAN_Feat: 0.707 G_ID: 0.130 G_Rec: 0.328 D_GP: 0.022 D_real: 0.913 D_fake: 0.979 +(epoch: 242, iters: 4272, time: 0.064) G_GAN: 0.214 G_GAN_Feat: 0.841 G_ID: 0.168 G_Rec: 0.409 D_GP: 0.026 D_real: 0.993 D_fake: 0.787 +(epoch: 242, iters: 4672, time: 0.064) G_GAN: 0.106 G_GAN_Feat: 0.769 G_ID: 0.139 G_Rec: 0.342 D_GP: 0.083 D_real: 0.739 D_fake: 0.896 +(epoch: 242, iters: 5072, time: 0.064) G_GAN: 0.368 G_GAN_Feat: 0.807 G_ID: 0.194 G_Rec: 0.403 D_GP: 0.024 D_real: 1.163 D_fake: 0.632 +(epoch: 242, iters: 5472, time: 0.064) G_GAN: 0.222 G_GAN_Feat: 0.597 G_ID: 0.130 G_Rec: 0.279 D_GP: 0.019 D_real: 1.165 D_fake: 0.778 +(epoch: 242, iters: 5872, time: 0.064) G_GAN: 0.330 G_GAN_Feat: 0.807 G_ID: 0.180 G_Rec: 0.379 D_GP: 0.026 D_real: 1.073 D_fake: 0.670 +(epoch: 242, iters: 6272, time: 0.064) G_GAN: 0.266 G_GAN_Feat: 0.734 G_ID: 0.132 G_Rec: 0.269 D_GP: 0.035 D_real: 1.028 D_fake: 0.736 +(epoch: 242, iters: 6672, time: 0.064) G_GAN: 0.647 G_GAN_Feat: 1.153 G_ID: 0.170 G_Rec: 0.455 D_GP: 0.033 D_real: 0.958 D_fake: 0.402 +(epoch: 242, iters: 7072, time: 0.064) G_GAN: -0.231 G_GAN_Feat: 0.743 G_ID: 0.126 G_Rec: 0.285 D_GP: 0.036 D_real: 0.517 D_fake: 1.231 +(epoch: 242, iters: 7472, time: 0.064) G_GAN: 0.424 G_GAN_Feat: 0.941 G_ID: 0.196 G_Rec: 0.418 D_GP: 0.059 D_real: 0.823 D_fake: 0.580 +(epoch: 242, iters: 7872, time: 0.064) G_GAN: 0.287 G_GAN_Feat: 0.571 G_ID: 0.116 G_Rec: 0.282 D_GP: 0.018 D_real: 1.270 D_fake: 0.713 +(epoch: 242, iters: 8272, time: 0.064) G_GAN: 0.254 G_GAN_Feat: 0.874 G_ID: 0.169 G_Rec: 0.432 D_GP: 0.034 D_real: 0.937 D_fake: 0.748 +(epoch: 243, iters: 64, time: 0.064) G_GAN: 0.001 G_GAN_Feat: 0.671 G_ID: 0.140 G_Rec: 0.292 D_GP: 0.035 D_real: 0.863 D_fake: 0.999 +(epoch: 243, iters: 464, time: 0.064) G_GAN: 0.401 G_GAN_Feat: 0.936 G_ID: 0.186 G_Rec: 0.406 D_GP: 0.037 D_real: 0.855 D_fake: 0.601 +(epoch: 243, iters: 864, time: 0.064) G_GAN: 0.517 G_GAN_Feat: 0.641 G_ID: 0.129 G_Rec: 0.271 D_GP: 0.025 D_real: 1.405 D_fake: 0.484 +(epoch: 243, iters: 1264, time: 0.064) G_GAN: 0.070 G_GAN_Feat: 0.857 G_ID: 0.177 G_Rec: 0.442 D_GP: 0.020 D_real: 0.765 D_fake: 0.930 +(epoch: 243, iters: 1664, time: 0.064) G_GAN: -0.112 G_GAN_Feat: 0.659 G_ID: 0.137 G_Rec: 0.312 D_GP: 0.026 D_real: 0.765 D_fake: 1.112 +(epoch: 243, iters: 2064, time: 0.064) G_GAN: -0.069 G_GAN_Feat: 0.976 G_ID: 0.185 G_Rec: 0.452 D_GP: 0.098 D_real: 0.404 D_fake: 1.069 +(epoch: 243, iters: 2464, time: 0.064) G_GAN: 0.132 G_GAN_Feat: 0.672 G_ID: 0.121 G_Rec: 0.283 D_GP: 0.021 D_real: 1.061 D_fake: 0.868 +(epoch: 243, iters: 2864, time: 0.064) G_GAN: 0.561 G_GAN_Feat: 0.847 G_ID: 0.189 G_Rec: 0.376 D_GP: 0.033 D_real: 1.224 D_fake: 0.481 +(epoch: 243, iters: 3264, time: 0.064) G_GAN: 0.301 G_GAN_Feat: 0.690 G_ID: 0.107 G_Rec: 0.283 D_GP: 0.024 D_real: 1.189 D_fake: 0.699 +(epoch: 243, iters: 3664, time: 0.064) G_GAN: 0.716 G_GAN_Feat: 0.934 G_ID: 0.177 G_Rec: 0.391 D_GP: 0.022 D_real: 1.452 D_fake: 0.374 +(epoch: 243, iters: 4064, time: 0.064) G_GAN: 0.474 G_GAN_Feat: 0.640 G_ID: 0.132 G_Rec: 0.290 D_GP: 0.023 D_real: 1.431 D_fake: 0.532 +(epoch: 243, iters: 4464, time: 0.064) G_GAN: 0.611 G_GAN_Feat: 0.980 G_ID: 0.164 G_Rec: 0.429 D_GP: 0.033 D_real: 1.093 D_fake: 0.407 +(epoch: 243, iters: 4864, time: 0.064) G_GAN: 0.201 G_GAN_Feat: 0.556 G_ID: 0.140 G_Rec: 0.287 D_GP: 0.015 D_real: 1.163 D_fake: 0.799 +(epoch: 243, iters: 5264, time: 0.064) G_GAN: 0.371 G_GAN_Feat: 0.722 G_ID: 0.187 G_Rec: 0.373 D_GP: 0.018 D_real: 1.174 D_fake: 0.630 +(epoch: 243, iters: 5664, time: 0.064) G_GAN: 0.218 G_GAN_Feat: 0.540 G_ID: 0.138 G_Rec: 0.278 D_GP: 0.016 D_real: 1.190 D_fake: 0.782 +(epoch: 243, iters: 6064, time: 0.064) G_GAN: 0.169 G_GAN_Feat: 0.768 G_ID: 0.163 G_Rec: 0.375 D_GP: 0.029 D_real: 0.988 D_fake: 0.831 +(epoch: 243, iters: 6464, time: 0.064) G_GAN: 0.102 G_GAN_Feat: 0.546 G_ID: 0.127 G_Rec: 0.265 D_GP: 0.022 D_real: 1.032 D_fake: 0.898 +(epoch: 243, iters: 6864, time: 0.064) G_GAN: 0.240 G_GAN_Feat: 0.868 G_ID: 0.150 G_Rec: 0.421 D_GP: 0.048 D_real: 0.824 D_fake: 0.761 +(epoch: 243, iters: 7264, time: 0.064) G_GAN: 0.142 G_GAN_Feat: 0.720 G_ID: 0.124 G_Rec: 0.325 D_GP: 0.043 D_real: 0.885 D_fake: 0.858 +(epoch: 243, iters: 7664, time: 0.064) G_GAN: 0.387 G_GAN_Feat: 0.909 G_ID: 0.173 G_Rec: 0.433 D_GP: 0.052 D_real: 1.008 D_fake: 0.617 +(epoch: 243, iters: 8064, time: 0.064) G_GAN: 0.281 G_GAN_Feat: 0.783 G_ID: 0.124 G_Rec: 0.326 D_GP: 0.073 D_real: 0.922 D_fake: 0.726 +(epoch: 243, iters: 8464, time: 0.064) G_GAN: 0.369 G_GAN_Feat: 0.801 G_ID: 0.175 G_Rec: 0.414 D_GP: 0.021 D_real: 1.079 D_fake: 0.631 +(epoch: 244, iters: 256, time: 0.064) G_GAN: 0.110 G_GAN_Feat: 0.731 G_ID: 0.127 G_Rec: 0.312 D_GP: 0.077 D_real: 0.782 D_fake: 0.890 +(epoch: 244, iters: 656, time: 0.064) G_GAN: 0.427 G_GAN_Feat: 0.886 G_ID: 0.157 G_Rec: 0.392 D_GP: 0.043 D_real: 0.929 D_fake: 0.575 +(epoch: 244, iters: 1056, time: 0.064) G_GAN: 0.391 G_GAN_Feat: 0.691 G_ID: 0.133 G_Rec: 0.312 D_GP: 0.024 D_real: 1.269 D_fake: 0.609 +(epoch: 244, iters: 1456, time: 0.064) G_GAN: 0.494 G_GAN_Feat: 0.868 G_ID: 0.183 G_Rec: 0.400 D_GP: 0.024 D_real: 1.116 D_fake: 0.507 +(epoch: 244, iters: 1856, time: 0.064) G_GAN: 0.099 G_GAN_Feat: 0.620 G_ID: 0.147 G_Rec: 0.310 D_GP: 0.024 D_real: 0.997 D_fake: 0.901 +(epoch: 244, iters: 2256, time: 0.064) G_GAN: 0.482 G_GAN_Feat: 0.755 G_ID: 0.162 G_Rec: 0.358 D_GP: 0.031 D_real: 1.262 D_fake: 0.523 +(epoch: 244, iters: 2656, time: 0.064) G_GAN: 0.348 G_GAN_Feat: 0.689 G_ID: 0.116 G_Rec: 0.309 D_GP: 0.039 D_real: 1.078 D_fake: 0.653 +(epoch: 244, iters: 3056, time: 0.064) G_GAN: 0.671 G_GAN_Feat: 0.905 G_ID: 0.183 G_Rec: 0.411 D_GP: 0.039 D_real: 1.187 D_fake: 0.357 +(epoch: 244, iters: 3456, time: 0.064) G_GAN: 0.260 G_GAN_Feat: 0.651 G_ID: 0.108 G_Rec: 0.260 D_GP: 0.023 D_real: 1.212 D_fake: 0.740 +(epoch: 244, iters: 3856, time: 0.064) G_GAN: 0.169 G_GAN_Feat: 1.079 G_ID: 0.175 G_Rec: 0.422 D_GP: 0.046 D_real: 0.255 D_fake: 0.831 +(epoch: 244, iters: 4256, time: 0.064) G_GAN: 0.087 G_GAN_Feat: 0.689 G_ID: 0.164 G_Rec: 0.319 D_GP: 0.021 D_real: 0.931 D_fake: 0.913 +(epoch: 244, iters: 4656, time: 0.064) G_GAN: 0.364 G_GAN_Feat: 0.852 G_ID: 0.147 G_Rec: 0.396 D_GP: 0.028 D_real: 0.993 D_fake: 0.668 +(epoch: 244, iters: 5056, time: 0.064) G_GAN: 0.016 G_GAN_Feat: 0.606 G_ID: 0.119 G_Rec: 0.266 D_GP: 0.023 D_real: 0.962 D_fake: 0.984 +(epoch: 244, iters: 5456, time: 0.064) G_GAN: 0.447 G_GAN_Feat: 0.919 G_ID: 0.163 G_Rec: 0.378 D_GP: 0.032 D_real: 1.177 D_fake: 0.556 +(epoch: 244, iters: 5856, time: 0.064) G_GAN: 0.239 G_GAN_Feat: 0.737 G_ID: 0.131 G_Rec: 0.329 D_GP: 0.025 D_real: 1.130 D_fake: 0.761 +(epoch: 244, iters: 6256, time: 0.064) G_GAN: 0.417 G_GAN_Feat: 0.891 G_ID: 0.186 G_Rec: 0.377 D_GP: 0.035 D_real: 1.017 D_fake: 0.586 +(epoch: 244, iters: 6656, time: 0.064) G_GAN: 0.131 G_GAN_Feat: 0.687 G_ID: 0.132 G_Rec: 0.288 D_GP: 0.051 D_real: 0.873 D_fake: 0.869 +(epoch: 244, iters: 7056, time: 0.064) G_GAN: 0.462 G_GAN_Feat: 0.934 G_ID: 0.160 G_Rec: 0.430 D_GP: 0.099 D_real: 0.856 D_fake: 0.544 +(epoch: 244, iters: 7456, time: 0.064) G_GAN: 0.363 G_GAN_Feat: 0.621 G_ID: 0.145 G_Rec: 0.287 D_GP: 0.023 D_real: 1.196 D_fake: 0.638 +(epoch: 244, iters: 7856, time: 0.064) G_GAN: 0.767 G_GAN_Feat: 0.826 G_ID: 0.162 G_Rec: 0.388 D_GP: 0.023 D_real: 1.565 D_fake: 0.265 +(epoch: 244, iters: 8256, time: 0.064) G_GAN: -0.168 G_GAN_Feat: 0.932 G_ID: 0.120 G_Rec: 0.308 D_GP: 0.740 D_real: 0.328 D_fake: 1.170 +(epoch: 245, iters: 48, time: 0.064) G_GAN: 0.394 G_GAN_Feat: 0.863 G_ID: 0.157 G_Rec: 0.379 D_GP: 0.024 D_real: 1.127 D_fake: 0.609 +(epoch: 245, iters: 448, time: 0.064) G_GAN: 0.180 G_GAN_Feat: 0.799 G_ID: 0.157 G_Rec: 0.366 D_GP: 0.056 D_real: 0.774 D_fake: 0.823 +(epoch: 245, iters: 848, time: 0.064) G_GAN: 0.492 G_GAN_Feat: 0.968 G_ID: 0.183 G_Rec: 0.405 D_GP: 0.052 D_real: 0.848 D_fake: 0.519 +(epoch: 245, iters: 1248, time: 0.064) G_GAN: -0.267 G_GAN_Feat: 0.670 G_ID: 0.122 G_Rec: 0.370 D_GP: 0.025 D_real: 0.693 D_fake: 1.267 +(epoch: 245, iters: 1648, time: 0.064) G_GAN: 0.404 G_GAN_Feat: 0.903 G_ID: 0.156 G_Rec: 0.389 D_GP: 0.031 D_real: 1.088 D_fake: 0.606 +(epoch: 245, iters: 2048, time: 0.064) G_GAN: 0.493 G_GAN_Feat: 0.661 G_ID: 0.134 G_Rec: 0.276 D_GP: 0.025 D_real: 1.450 D_fake: 0.510 +(epoch: 245, iters: 2448, time: 0.064) G_GAN: 0.326 G_GAN_Feat: 1.027 G_ID: 0.173 G_Rec: 0.450 D_GP: 0.033 D_real: 0.854 D_fake: 0.677 +(epoch: 245, iters: 2848, time: 0.064) G_GAN: 0.223 G_GAN_Feat: 0.679 G_ID: 0.124 G_Rec: 0.314 D_GP: 0.022 D_real: 1.148 D_fake: 0.777 +(epoch: 245, iters: 3248, time: 0.064) G_GAN: 0.405 G_GAN_Feat: 0.822 G_ID: 0.167 G_Rec: 0.379 D_GP: 0.023 D_real: 1.182 D_fake: 0.597 +(epoch: 245, iters: 3648, time: 0.064) G_GAN: 0.140 G_GAN_Feat: 0.720 G_ID: 0.134 G_Rec: 0.311 D_GP: 0.026 D_real: 0.991 D_fake: 0.860 +(epoch: 245, iters: 4048, time: 0.064) G_GAN: 0.264 G_GAN_Feat: 0.812 G_ID: 0.188 G_Rec: 0.372 D_GP: 0.022 D_real: 1.026 D_fake: 0.737 +(epoch: 245, iters: 4448, time: 0.064) G_GAN: -0.289 G_GAN_Feat: 0.867 G_ID: 0.134 G_Rec: 0.331 D_GP: 0.644 D_real: 0.332 D_fake: 1.289 +(epoch: 245, iters: 4848, time: 0.064) G_GAN: 0.358 G_GAN_Feat: 0.957 G_ID: 0.167 G_Rec: 0.411 D_GP: 0.057 D_real: 0.763 D_fake: 0.643 +(epoch: 245, iters: 5248, time: 0.064) G_GAN: 0.328 G_GAN_Feat: 0.691 G_ID: 0.115 G_Rec: 0.283 D_GP: 0.035 D_real: 1.135 D_fake: 0.674 +(epoch: 245, iters: 5648, time: 0.064) G_GAN: 0.587 G_GAN_Feat: 0.883 G_ID: 0.171 G_Rec: 0.394 D_GP: 0.027 D_real: 1.333 D_fake: 0.462 +(epoch: 245, iters: 6048, time: 0.064) G_GAN: 0.313 G_GAN_Feat: 0.759 G_ID: 0.131 G_Rec: 0.300 D_GP: 0.031 D_real: 1.080 D_fake: 0.690 +(epoch: 245, iters: 6448, time: 0.064) G_GAN: 0.846 G_GAN_Feat: 0.907 G_ID: 0.174 G_Rec: 0.404 D_GP: 0.024 D_real: 1.545 D_fake: 0.217 +(epoch: 245, iters: 6848, time: 0.064) G_GAN: 0.235 G_GAN_Feat: 0.599 G_ID: 0.119 G_Rec: 0.280 D_GP: 0.019 D_real: 1.132 D_fake: 0.765 +(epoch: 245, iters: 7248, time: 0.064) G_GAN: 0.192 G_GAN_Feat: 0.839 G_ID: 0.208 G_Rec: 0.400 D_GP: 0.026 D_real: 0.928 D_fake: 0.808 +(epoch: 245, iters: 7648, time: 0.064) G_GAN: 0.181 G_GAN_Feat: 0.809 G_ID: 0.120 G_Rec: 0.336 D_GP: 0.378 D_real: 0.736 D_fake: 0.821 +(epoch: 245, iters: 8048, time: 0.064) G_GAN: 0.455 G_GAN_Feat: 1.014 G_ID: 0.166 G_Rec: 0.450 D_GP: 0.090 D_real: 0.621 D_fake: 0.548 +(epoch: 245, iters: 8448, time: 0.064) G_GAN: 0.215 G_GAN_Feat: 0.872 G_ID: 0.126 G_Rec: 0.290 D_GP: 0.059 D_real: 0.510 D_fake: 0.785 +(epoch: 246, iters: 240, time: 0.064) G_GAN: 0.641 G_GAN_Feat: 0.854 G_ID: 0.181 G_Rec: 0.403 D_GP: 0.021 D_real: 1.361 D_fake: 0.419 +(epoch: 246, iters: 640, time: 0.064) G_GAN: 0.110 G_GAN_Feat: 0.758 G_ID: 0.121 G_Rec: 0.324 D_GP: 0.026 D_real: 0.901 D_fake: 0.890 +(epoch: 246, iters: 1040, time: 0.064) G_GAN: 0.800 G_GAN_Feat: 1.047 G_ID: 0.160 G_Rec: 0.438 D_GP: 0.065 D_real: 0.601 D_fake: 0.321 +(epoch: 246, iters: 1440, time: 0.064) G_GAN: 0.194 G_GAN_Feat: 0.626 G_ID: 0.127 G_Rec: 0.311 D_GP: 0.017 D_real: 1.193 D_fake: 0.806 +(epoch: 246, iters: 1840, time: 0.064) G_GAN: 0.260 G_GAN_Feat: 0.857 G_ID: 0.175 G_Rec: 0.411 D_GP: 0.021 D_real: 1.018 D_fake: 0.741 +(epoch: 246, iters: 2240, time: 0.064) G_GAN: 0.107 G_GAN_Feat: 0.722 G_ID: 0.117 G_Rec: 0.318 D_GP: 0.029 D_real: 1.055 D_fake: 0.894 +(epoch: 246, iters: 2640, time: 0.064) G_GAN: 0.751 G_GAN_Feat: 0.856 G_ID: 0.208 G_Rec: 0.384 D_GP: 0.025 D_real: 1.545 D_fake: 0.295 +(epoch: 246, iters: 3040, time: 0.064) G_GAN: 0.300 G_GAN_Feat: 0.780 G_ID: 0.118 G_Rec: 0.283 D_GP: 0.114 D_real: 0.746 D_fake: 0.702 +(epoch: 246, iters: 3440, time: 0.064) G_GAN: 0.468 G_GAN_Feat: 0.925 G_ID: 0.168 G_Rec: 0.463 D_GP: 0.022 D_real: 1.104 D_fake: 0.537 +(epoch: 246, iters: 3840, time: 0.064) G_GAN: 0.262 G_GAN_Feat: 0.638 G_ID: 0.122 G_Rec: 0.308 D_GP: 0.022 D_real: 1.241 D_fake: 0.740 +(epoch: 246, iters: 4240, time: 0.064) G_GAN: 0.413 G_GAN_Feat: 0.700 G_ID: 0.190 G_Rec: 0.346 D_GP: 0.018 D_real: 1.278 D_fake: 0.587 +(epoch: 246, iters: 4640, time: 0.064) G_GAN: 0.226 G_GAN_Feat: 0.557 G_ID: 0.129 G_Rec: 0.274 D_GP: 0.019 D_real: 1.178 D_fake: 0.774 +(epoch: 246, iters: 5040, time: 0.064) G_GAN: 0.291 G_GAN_Feat: 0.807 G_ID: 0.164 G_Rec: 0.381 D_GP: 0.030 D_real: 1.069 D_fake: 0.709 +(epoch: 246, iters: 5440, time: 0.064) G_GAN: 0.158 G_GAN_Feat: 0.707 G_ID: 0.116 G_Rec: 0.301 D_GP: 0.032 D_real: 1.005 D_fake: 0.844 +(epoch: 246, iters: 5840, time: 0.064) G_GAN: 0.491 G_GAN_Feat: 0.898 G_ID: 0.206 G_Rec: 0.389 D_GP: 0.026 D_real: 1.117 D_fake: 0.519 +(epoch: 246, iters: 6240, time: 0.064) G_GAN: 0.192 G_GAN_Feat: 0.600 G_ID: 0.159 G_Rec: 0.295 D_GP: 0.023 D_real: 1.166 D_fake: 0.808 +(epoch: 246, iters: 6640, time: 0.064) G_GAN: 0.329 G_GAN_Feat: 0.840 G_ID: 0.174 G_Rec: 0.360 D_GP: 0.029 D_real: 1.183 D_fake: 0.672 +(epoch: 246, iters: 7040, time: 0.064) G_GAN: 0.329 G_GAN_Feat: 0.702 G_ID: 0.106 G_Rec: 0.297 D_GP: 0.032 D_real: 1.136 D_fake: 0.672 +(epoch: 246, iters: 7440, time: 0.064) G_GAN: 0.151 G_GAN_Feat: 0.864 G_ID: 0.192 G_Rec: 0.358 D_GP: 0.055 D_real: 0.698 D_fake: 0.849 +(epoch: 246, iters: 7840, time: 0.064) G_GAN: 0.266 G_GAN_Feat: 0.752 G_ID: 0.123 G_Rec: 0.283 D_GP: 0.083 D_real: 0.917 D_fake: 0.735 +(epoch: 246, iters: 8240, time: 0.064) G_GAN: 0.409 G_GAN_Feat: 0.753 G_ID: 0.163 G_Rec: 0.356 D_GP: 0.023 D_real: 1.186 D_fake: 0.593 +(epoch: 247, iters: 32, time: 0.064) G_GAN: 0.141 G_GAN_Feat: 0.875 G_ID: 0.144 G_Rec: 0.340 D_GP: 0.335 D_real: 0.523 D_fake: 0.861 +(epoch: 247, iters: 432, time: 0.064) G_GAN: 0.544 G_GAN_Feat: 1.140 G_ID: 0.184 G_Rec: 0.411 D_GP: 0.155 D_real: 0.488 D_fake: 0.475 +(epoch: 247, iters: 832, time: 0.064) G_GAN: 0.283 G_GAN_Feat: 0.586 G_ID: 0.154 G_Rec: 0.326 D_GP: 0.020 D_real: 1.250 D_fake: 0.717 +(epoch: 247, iters: 1232, time: 0.064) G_GAN: 0.486 G_GAN_Feat: 0.805 G_ID: 0.169 G_Rec: 0.387 D_GP: 0.020 D_real: 1.230 D_fake: 0.529 +(epoch: 247, iters: 1632, time: 0.064) G_GAN: 0.160 G_GAN_Feat: 0.611 G_ID: 0.109 G_Rec: 0.278 D_GP: 0.026 D_real: 1.059 D_fake: 0.840 +(epoch: 247, iters: 2032, time: 0.064) G_GAN: 0.265 G_GAN_Feat: 0.793 G_ID: 0.162 G_Rec: 0.385 D_GP: 0.024 D_real: 0.977 D_fake: 0.736 +(epoch: 247, iters: 2432, time: 0.064) G_GAN: 0.224 G_GAN_Feat: 0.610 G_ID: 0.158 G_Rec: 0.290 D_GP: 0.025 D_real: 1.096 D_fake: 0.776 +(epoch: 247, iters: 2832, time: 0.064) G_GAN: 0.400 G_GAN_Feat: 0.865 G_ID: 0.178 G_Rec: 0.384 D_GP: 0.042 D_real: 1.050 D_fake: 0.600 +(epoch: 247, iters: 3232, time: 0.064) G_GAN: 0.096 G_GAN_Feat: 0.666 G_ID: 0.141 G_Rec: 0.294 D_GP: 0.028 D_real: 0.932 D_fake: 0.904 +(epoch: 247, iters: 3632, time: 0.064) G_GAN: 0.651 G_GAN_Feat: 0.852 G_ID: 0.153 G_Rec: 0.375 D_GP: 0.027 D_real: 1.382 D_fake: 0.365 +(epoch: 247, iters: 4032, time: 0.064) G_GAN: 0.246 G_GAN_Feat: 0.747 G_ID: 0.115 G_Rec: 0.290 D_GP: 0.051 D_real: 1.002 D_fake: 0.755 +(epoch: 247, iters: 4432, time: 0.064) G_GAN: 0.478 G_GAN_Feat: 0.977 G_ID: 0.183 G_Rec: 0.440 D_GP: 0.044 D_real: 0.863 D_fake: 0.527 +(epoch: 247, iters: 4832, time: 0.064) G_GAN: -0.109 G_GAN_Feat: 1.017 G_ID: 0.130 G_Rec: 0.356 D_GP: 3.370 D_real: 0.390 D_fake: 1.110 +(epoch: 247, iters: 5232, time: 0.064) G_GAN: 0.629 G_GAN_Feat: 0.817 G_ID: 0.138 G_Rec: 0.421 D_GP: 0.028 D_real: 1.368 D_fake: 0.391 +(epoch: 247, iters: 5632, time: 0.064) G_GAN: 0.154 G_GAN_Feat: 0.664 G_ID: 0.130 G_Rec: 0.335 D_GP: 0.034 D_real: 1.040 D_fake: 0.847 +(epoch: 247, iters: 6032, time: 0.064) G_GAN: 0.212 G_GAN_Feat: 0.872 G_ID: 0.170 G_Rec: 0.448 D_GP: 0.046 D_real: 0.818 D_fake: 0.788 +(epoch: 247, iters: 6432, time: 0.064) G_GAN: 0.025 G_GAN_Feat: 0.599 G_ID: 0.119 G_Rec: 0.303 D_GP: 0.047 D_real: 0.868 D_fake: 0.975 +(epoch: 247, iters: 6832, time: 0.064) G_GAN: 0.275 G_GAN_Feat: 0.806 G_ID: 0.154 G_Rec: 0.393 D_GP: 0.030 D_real: 1.026 D_fake: 0.725 +(epoch: 247, iters: 7232, time: 0.064) G_GAN: 0.161 G_GAN_Feat: 0.601 G_ID: 0.131 G_Rec: 0.289 D_GP: 0.022 D_real: 1.112 D_fake: 0.839 +(epoch: 247, iters: 7632, time: 0.064) G_GAN: 0.188 G_GAN_Feat: 0.842 G_ID: 0.185 G_Rec: 0.372 D_GP: 0.057 D_real: 0.741 D_fake: 0.813 +(epoch: 247, iters: 8032, time: 0.064) G_GAN: 0.286 G_GAN_Feat: 0.598 G_ID: 0.137 G_Rec: 0.304 D_GP: 0.020 D_real: 1.194 D_fake: 0.715 +(epoch: 247, iters: 8432, time: 0.064) G_GAN: 0.306 G_GAN_Feat: 0.999 G_ID: 0.183 G_Rec: 0.390 D_GP: 0.396 D_real: 0.516 D_fake: 0.698 +(epoch: 248, iters: 224, time: 0.064) G_GAN: 0.206 G_GAN_Feat: 0.637 G_ID: 0.120 G_Rec: 0.283 D_GP: 0.026 D_real: 1.126 D_fake: 0.794 +(epoch: 248, iters: 624, time: 0.064) G_GAN: 0.513 G_GAN_Feat: 0.959 G_ID: 0.159 G_Rec: 0.407 D_GP: 0.089 D_real: 0.940 D_fake: 0.496 +(epoch: 248, iters: 1024, time: 0.064) G_GAN: 0.069 G_GAN_Feat: 0.933 G_ID: 0.137 G_Rec: 0.337 D_GP: 0.128 D_real: 0.502 D_fake: 0.942 +(epoch: 248, iters: 1424, time: 0.064) G_GAN: 0.238 G_GAN_Feat: 0.824 G_ID: 0.189 G_Rec: 0.385 D_GP: 0.029 D_real: 1.000 D_fake: 0.763 +(epoch: 248, iters: 1824, time: 0.064) G_GAN: 0.041 G_GAN_Feat: 0.766 G_ID: 0.134 G_Rec: 0.311 D_GP: 0.146 D_real: 0.979 D_fake: 0.959 +(epoch: 248, iters: 2224, time: 0.064) G_GAN: 0.508 G_GAN_Feat: 0.869 G_ID: 0.170 G_Rec: 0.390 D_GP: 0.037 D_real: 1.117 D_fake: 0.497 +(epoch: 248, iters: 2624, time: 0.064) G_GAN: 0.344 G_GAN_Feat: 0.684 G_ID: 0.149 G_Rec: 0.277 D_GP: 0.025 D_real: 1.279 D_fake: 0.665 +(epoch: 248, iters: 3024, time: 0.064) G_GAN: 0.596 G_GAN_Feat: 1.048 G_ID: 0.158 G_Rec: 0.456 D_GP: 0.101 D_real: 1.093 D_fake: 0.428 +(epoch: 248, iters: 3424, time: 0.064) G_GAN: -0.177 G_GAN_Feat: 0.846 G_ID: 0.153 G_Rec: 0.308 D_GP: 0.086 D_real: 0.816 D_fake: 1.177 +(epoch: 248, iters: 3824, time: 0.064) G_GAN: 0.392 G_GAN_Feat: 0.853 G_ID: 0.177 G_Rec: 0.391 D_GP: 0.021 D_real: 1.107 D_fake: 0.613 +(epoch: 248, iters: 4224, time: 0.064) G_GAN: 0.080 G_GAN_Feat: 0.771 G_ID: 0.131 G_Rec: 0.303 D_GP: 0.037 D_real: 0.821 D_fake: 0.920 +(epoch: 248, iters: 4624, time: 0.064) G_GAN: 0.667 G_GAN_Feat: 0.895 G_ID: 0.159 G_Rec: 0.408 D_GP: 0.023 D_real: 1.340 D_fake: 0.343 +(epoch: 248, iters: 5024, time: 0.064) G_GAN: 0.234 G_GAN_Feat: 0.645 G_ID: 0.124 G_Rec: 0.262 D_GP: 0.022 D_real: 1.147 D_fake: 0.766 +(epoch: 248, iters: 5424, time: 0.064) G_GAN: 0.278 G_GAN_Feat: 0.856 G_ID: 0.173 G_Rec: 0.390 D_GP: 0.034 D_real: 0.885 D_fake: 0.722 +(epoch: 248, iters: 5824, time: 0.064) G_GAN: -0.004 G_GAN_Feat: 0.763 G_ID: 0.129 G_Rec: 0.313 D_GP: 0.048 D_real: 0.760 D_fake: 1.005 +(epoch: 248, iters: 6224, time: 0.064) G_GAN: 0.535 G_GAN_Feat: 1.123 G_ID: 0.169 G_Rec: 0.400 D_GP: 0.134 D_real: 0.492 D_fake: 0.467 +(epoch: 248, iters: 6624, time: 0.064) G_GAN: 0.320 G_GAN_Feat: 0.647 G_ID: 0.142 G_Rec: 0.306 D_GP: 0.024 D_real: 1.228 D_fake: 0.681 +(epoch: 248, iters: 7024, time: 0.064) G_GAN: 0.502 G_GAN_Feat: 0.875 G_ID: 0.172 G_Rec: 0.377 D_GP: 0.025 D_real: 1.109 D_fake: 0.517 +(epoch: 248, iters: 7424, time: 0.064) G_GAN: 0.216 G_GAN_Feat: 0.829 G_ID: 0.133 G_Rec: 0.308 D_GP: 0.034 D_real: 0.841 D_fake: 0.785 +(epoch: 248, iters: 7824, time: 0.064) G_GAN: 0.613 G_GAN_Feat: 0.946 G_ID: 0.162 G_Rec: 0.388 D_GP: 0.034 D_real: 1.143 D_fake: 0.409 +(epoch: 248, iters: 8224, time: 0.064) G_GAN: -0.272 G_GAN_Feat: 0.981 G_ID: 0.135 G_Rec: 0.338 D_GP: 0.176 D_real: 0.135 D_fake: 1.272 +(epoch: 249, iters: 16, time: 0.064) G_GAN: 0.546 G_GAN_Feat: 0.888 G_ID: 0.175 G_Rec: 0.448 D_GP: 0.026 D_real: 1.283 D_fake: 0.463 +(epoch: 249, iters: 416, time: 0.064) G_GAN: 0.200 G_GAN_Feat: 0.651 G_ID: 0.116 G_Rec: 0.374 D_GP: 0.071 D_real: 1.020 D_fake: 0.801 +(epoch: 249, iters: 816, time: 0.064) G_GAN: 0.519 G_GAN_Feat: 0.793 G_ID: 0.143 G_Rec: 0.391 D_GP: 0.022 D_real: 1.195 D_fake: 0.488 +(epoch: 249, iters: 1216, time: 0.064) G_GAN: 0.223 G_GAN_Feat: 0.632 G_ID: 0.104 G_Rec: 0.289 D_GP: 0.023 D_real: 1.113 D_fake: 0.777 +(epoch: 249, iters: 1616, time: 0.064) G_GAN: 0.288 G_GAN_Feat: 0.827 G_ID: 0.161 G_Rec: 0.358 D_GP: 0.044 D_real: 1.003 D_fake: 0.712 +(epoch: 249, iters: 2016, time: 0.064) G_GAN: 0.215 G_GAN_Feat: 0.635 G_ID: 0.149 G_Rec: 0.296 D_GP: 0.022 D_real: 1.173 D_fake: 0.785 +(epoch: 249, iters: 2416, time: 0.064) G_GAN: 0.155 G_GAN_Feat: 0.706 G_ID: 0.171 G_Rec: 0.367 D_GP: 0.018 D_real: 1.122 D_fake: 0.846 +(epoch: 249, iters: 2816, time: 0.064) G_GAN: 0.199 G_GAN_Feat: 0.568 G_ID: 0.126 G_Rec: 0.317 D_GP: 0.021 D_real: 1.181 D_fake: 0.801 +(epoch: 249, iters: 3216, time: 0.064) G_GAN: 0.316 G_GAN_Feat: 0.725 G_ID: 0.162 G_Rec: 0.367 D_GP: 0.019 D_real: 1.098 D_fake: 0.686 +(epoch: 249, iters: 3616, time: 0.064) G_GAN: 0.099 G_GAN_Feat: 0.561 G_ID: 0.148 G_Rec: 0.259 D_GP: 0.029 D_real: 1.006 D_fake: 0.901 +(epoch: 249, iters: 4016, time: 0.064) G_GAN: 0.516 G_GAN_Feat: 0.741 G_ID: 0.164 G_Rec: 0.365 D_GP: 0.021 D_real: 1.301 D_fake: 0.491 +(epoch: 249, iters: 4416, time: 0.064) G_GAN: -0.059 G_GAN_Feat: 0.562 G_ID: 0.143 G_Rec: 0.321 D_GP: 0.022 D_real: 0.850 D_fake: 1.059 +(epoch: 249, iters: 4816, time: 0.064) G_GAN: 0.365 G_GAN_Feat: 0.792 G_ID: 0.143 G_Rec: 0.387 D_GP: 0.024 D_real: 1.122 D_fake: 0.637 +(epoch: 249, iters: 5216, time: 0.064) G_GAN: 0.243 G_GAN_Feat: 0.648 G_ID: 0.156 G_Rec: 0.289 D_GP: 0.024 D_real: 1.206 D_fake: 0.757 +(epoch: 249, iters: 5616, time: 0.064) G_GAN: 0.425 G_GAN_Feat: 0.835 G_ID: 0.159 G_Rec: 0.398 D_GP: 0.024 D_real: 1.162 D_fake: 0.578 +(epoch: 249, iters: 6016, time: 0.064) G_GAN: 0.136 G_GAN_Feat: 0.719 G_ID: 0.136 G_Rec: 0.306 D_GP: 0.052 D_real: 0.858 D_fake: 0.867 +(epoch: 249, iters: 6416, time: 0.064) G_GAN: 0.493 G_GAN_Feat: 0.878 G_ID: 0.196 G_Rec: 0.414 D_GP: 0.025 D_real: 1.266 D_fake: 0.511 +(epoch: 249, iters: 6816, time: 0.064) G_GAN: 0.292 G_GAN_Feat: 0.752 G_ID: 0.134 G_Rec: 0.340 D_GP: 0.054 D_real: 0.975 D_fake: 0.710 +(epoch: 249, iters: 7216, time: 0.064) G_GAN: 0.841 G_GAN_Feat: 0.894 G_ID: 0.236 G_Rec: 0.450 D_GP: 0.035 D_real: 1.402 D_fake: 0.276 +(epoch: 249, iters: 7616, time: 0.064) G_GAN: 0.156 G_GAN_Feat: 0.711 G_ID: 0.120 G_Rec: 0.325 D_GP: 0.039 D_real: 0.899 D_fake: 0.844 +(epoch: 249, iters: 8016, time: 0.064) G_GAN: 0.236 G_GAN_Feat: 0.865 G_ID: 0.189 G_Rec: 0.392 D_GP: 0.028 D_real: 0.954 D_fake: 0.764 +(epoch: 249, iters: 8416, time: 0.064) G_GAN: 0.110 G_GAN_Feat: 0.820 G_ID: 0.140 G_Rec: 0.302 D_GP: 0.084 D_real: 0.597 D_fake: 0.894 +(epoch: 250, iters: 208, time: 0.064) G_GAN: 0.271 G_GAN_Feat: 0.926 G_ID: 0.163 G_Rec: 0.362 D_GP: 0.063 D_real: 0.723 D_fake: 0.730 +(epoch: 250, iters: 608, time: 0.064) G_GAN: 0.092 G_GAN_Feat: 0.612 G_ID: 0.150 G_Rec: 0.288 D_GP: 0.019 D_real: 1.068 D_fake: 0.908 +(epoch: 250, iters: 1008, time: 0.064) G_GAN: 0.257 G_GAN_Feat: 0.808 G_ID: 0.177 G_Rec: 0.382 D_GP: 0.033 D_real: 0.971 D_fake: 0.746 +(epoch: 250, iters: 1408, time: 0.064) G_GAN: 0.090 G_GAN_Feat: 0.691 G_ID: 0.121 G_Rec: 0.304 D_GP: 0.041 D_real: 0.967 D_fake: 0.910 +(epoch: 250, iters: 1808, time: 0.064) G_GAN: 0.461 G_GAN_Feat: 0.949 G_ID: 0.173 G_Rec: 0.476 D_GP: 0.031 D_real: 1.224 D_fake: 0.550 +(epoch: 250, iters: 2208, time: 0.064) G_GAN: 0.125 G_GAN_Feat: 0.710 G_ID: 0.123 G_Rec: 0.296 D_GP: 0.028 D_real: 1.008 D_fake: 0.875 +(epoch: 250, iters: 2608, time: 0.064) G_GAN: 0.543 G_GAN_Feat: 0.763 G_ID: 0.146 G_Rec: 0.340 D_GP: 0.021 D_real: 1.310 D_fake: 0.470 +(epoch: 250, iters: 3008, time: 0.064) G_GAN: 0.302 G_GAN_Feat: 0.603 G_ID: 0.126 G_Rec: 0.260 D_GP: 0.022 D_real: 1.252 D_fake: 0.702 +(epoch: 250, iters: 3408, time: 0.064) G_GAN: 0.122 G_GAN_Feat: 0.939 G_ID: 0.194 G_Rec: 0.394 D_GP: 0.029 D_real: 0.774 D_fake: 0.878 +(epoch: 250, iters: 3808, time: 0.064) G_GAN: 0.192 G_GAN_Feat: 0.711 G_ID: 0.126 G_Rec: 0.310 D_GP: 0.033 D_real: 0.957 D_fake: 0.808 +(epoch: 250, iters: 4208, time: 0.064) G_GAN: 0.342 G_GAN_Feat: 0.788 G_ID: 0.179 G_Rec: 0.341 D_GP: 0.028 D_real: 1.112 D_fake: 0.661 +(epoch: 250, iters: 4608, time: 0.064) G_GAN: 0.282 G_GAN_Feat: 0.785 G_ID: 0.125 G_Rec: 0.313 D_GP: 0.044 D_real: 0.881 D_fake: 0.718 +(epoch: 250, iters: 5008, time: 0.064) G_GAN: 0.474 G_GAN_Feat: 0.888 G_ID: 0.203 G_Rec: 0.422 D_GP: 0.029 D_real: 1.167 D_fake: 0.542 +(epoch: 250, iters: 5408, time: 0.064) G_GAN: 0.538 G_GAN_Feat: 1.047 G_ID: 0.138 G_Rec: 0.352 D_GP: 0.948 D_real: 0.578 D_fake: 0.568 +(epoch: 250, iters: 5808, time: 0.064) G_GAN: 0.447 G_GAN_Feat: 0.826 G_ID: 0.178 G_Rec: 0.369 D_GP: 0.027 D_real: 1.170 D_fake: 0.555 +(epoch: 250, iters: 6208, time: 0.064) G_GAN: 0.314 G_GAN_Feat: 0.656 G_ID: 0.132 G_Rec: 0.291 D_GP: 0.027 D_real: 1.212 D_fake: 0.686 +(epoch: 250, iters: 6608, time: 0.064) G_GAN: 0.701 G_GAN_Feat: 0.881 G_ID: 0.156 G_Rec: 0.441 D_GP: 0.029 D_real: 1.397 D_fake: 0.315 +(epoch: 250, iters: 7008, time: 0.064) G_GAN: 0.083 G_GAN_Feat: 0.569 G_ID: 0.130 G_Rec: 0.287 D_GP: 0.018 D_real: 1.049 D_fake: 0.918 +(epoch: 250, iters: 7408, time: 0.064) G_GAN: 0.156 G_GAN_Feat: 0.877 G_ID: 0.139 G_Rec: 0.462 D_GP: 0.032 D_real: 0.743 D_fake: 0.844 +(epoch: 250, iters: 7808, time: 0.064) G_GAN: -0.004 G_GAN_Feat: 0.680 G_ID: 0.153 G_Rec: 0.326 D_GP: 0.027 D_real: 0.847 D_fake: 1.004 +(epoch: 250, iters: 8208, time: 0.064) G_GAN: 0.293 G_GAN_Feat: 0.867 G_ID: 0.211 G_Rec: 0.397 D_GP: 0.025 D_real: 0.913 D_fake: 0.712 +(epoch: 250, iters: 8608, time: 0.064) G_GAN: 0.237 G_GAN_Feat: 0.728 G_ID: 0.144 G_Rec: 0.294 D_GP: 0.037 D_real: 1.070 D_fake: 0.763 +(epoch: 251, iters: 400, time: 0.064) G_GAN: 0.701 G_GAN_Feat: 0.823 G_ID: 0.168 G_Rec: 0.358 D_GP: 0.024 D_real: 1.507 D_fake: 0.319 +(epoch: 251, iters: 800, time: 0.064) G_GAN: 0.256 G_GAN_Feat: 0.560 G_ID: 0.137 G_Rec: 0.289 D_GP: 0.019 D_real: 1.241 D_fake: 0.745 +(epoch: 251, iters: 1200, time: 0.064) G_GAN: 0.559 G_GAN_Feat: 0.883 G_ID: 0.143 G_Rec: 0.415 D_GP: 0.031 D_real: 1.199 D_fake: 0.452 +(epoch: 251, iters: 1600, time: 0.064) G_GAN: 0.082 G_GAN_Feat: 0.586 G_ID: 0.123 G_Rec: 0.286 D_GP: 0.022 D_real: 1.041 D_fake: 0.918 +(epoch: 251, iters: 2000, time: 0.064) G_GAN: 0.465 G_GAN_Feat: 0.895 G_ID: 0.169 G_Rec: 0.401 D_GP: 0.030 D_real: 1.086 D_fake: 0.543 +(epoch: 251, iters: 2400, time: 0.064) G_GAN: 0.287 G_GAN_Feat: 0.705 G_ID: 0.120 G_Rec: 0.323 D_GP: 0.033 D_real: 1.065 D_fake: 0.714 +(epoch: 251, iters: 2800, time: 0.064) G_GAN: 0.136 G_GAN_Feat: 0.928 G_ID: 0.165 G_Rec: 0.401 D_GP: 0.176 D_real: 0.709 D_fake: 0.864 +(epoch: 251, iters: 3200, time: 0.064) G_GAN: 0.258 G_GAN_Feat: 0.751 G_ID: 0.120 G_Rec: 0.303 D_GP: 0.036 D_real: 0.965 D_fake: 0.742 +(epoch: 251, iters: 3600, time: 0.064) G_GAN: 0.376 G_GAN_Feat: 0.866 G_ID: 0.161 G_Rec: 0.404 D_GP: 0.025 D_real: 1.068 D_fake: 0.625 +(epoch: 251, iters: 4000, time: 0.064) G_GAN: 0.114 G_GAN_Feat: 0.797 G_ID: 0.128 G_Rec: 0.335 D_GP: 0.295 D_real: 0.631 D_fake: 0.887 +(epoch: 251, iters: 4400, time: 0.064) G_GAN: 0.392 G_GAN_Feat: 0.835 G_ID: 0.170 G_Rec: 0.378 D_GP: 0.030 D_real: 1.072 D_fake: 0.609 +(epoch: 251, iters: 4800, time: 0.064) G_GAN: 0.120 G_GAN_Feat: 0.696 G_ID: 0.130 G_Rec: 0.289 D_GP: 0.030 D_real: 0.952 D_fake: 0.880 +(epoch: 251, iters: 5200, time: 0.064) G_GAN: 0.750 G_GAN_Feat: 0.782 G_ID: 0.155 G_Rec: 0.347 D_GP: 0.022 D_real: 1.556 D_fake: 0.348 +(epoch: 251, iters: 5600, time: 0.064) G_GAN: 0.344 G_GAN_Feat: 0.680 G_ID: 0.140 G_Rec: 0.281 D_GP: 0.022 D_real: 1.230 D_fake: 0.656 +(epoch: 251, iters: 6000, time: 0.064) G_GAN: 0.299 G_GAN_Feat: 0.955 G_ID: 0.184 G_Rec: 0.387 D_GP: 0.027 D_real: 0.717 D_fake: 0.702 +(epoch: 251, iters: 6400, time: 0.064) G_GAN: 0.063 G_GAN_Feat: 0.775 G_ID: 0.147 G_Rec: 0.325 D_GP: 0.070 D_real: 0.826 D_fake: 0.937 +(epoch: 251, iters: 6800, time: 0.064) G_GAN: -0.008 G_GAN_Feat: 1.032 G_ID: 0.201 G_Rec: 0.402 D_GP: 0.378 D_real: 0.406 D_fake: 1.008 +(epoch: 251, iters: 7200, time: 0.064) G_GAN: 0.162 G_GAN_Feat: 0.696 G_ID: 0.150 G_Rec: 0.297 D_GP: 0.030 D_real: 1.046 D_fake: 0.838 +(epoch: 251, iters: 7600, time: 0.064) G_GAN: 0.578 G_GAN_Feat: 0.862 G_ID: 0.199 G_Rec: 0.412 D_GP: 0.025 D_real: 1.177 D_fake: 0.430 +(epoch: 251, iters: 8000, time: 0.064) G_GAN: 0.315 G_GAN_Feat: 0.646 G_ID: 0.132 G_Rec: 0.281 D_GP: 0.022 D_real: 1.250 D_fake: 0.685 +(epoch: 251, iters: 8400, time: 0.064) G_GAN: 0.352 G_GAN_Feat: 0.975 G_ID: 0.199 G_Rec: 0.418 D_GP: 0.118 D_real: 0.709 D_fake: 0.650 +(epoch: 252, iters: 192, time: 0.064) G_GAN: 0.246 G_GAN_Feat: 0.687 G_ID: 0.136 G_Rec: 0.284 D_GP: 0.025 D_real: 1.072 D_fake: 0.755 +(epoch: 252, iters: 592, time: 0.064) G_GAN: 0.513 G_GAN_Feat: 0.834 G_ID: 0.157 G_Rec: 0.387 D_GP: 0.022 D_real: 1.297 D_fake: 0.491 +(epoch: 252, iters: 992, time: 0.064) G_GAN: 0.171 G_GAN_Feat: 0.670 G_ID: 0.124 G_Rec: 0.251 D_GP: 0.026 D_real: 1.064 D_fake: 0.830 +(epoch: 252, iters: 1392, time: 0.064) G_GAN: 0.276 G_GAN_Feat: 0.782 G_ID: 0.176 G_Rec: 0.361 D_GP: 0.024 D_real: 1.070 D_fake: 0.724 +(epoch: 252, iters: 1792, time: 0.064) G_GAN: 0.100 G_GAN_Feat: 0.603 G_ID: 0.155 G_Rec: 0.254 D_GP: 0.028 D_real: 1.020 D_fake: 0.900 +(epoch: 252, iters: 2192, time: 0.064) G_GAN: 0.578 G_GAN_Feat: 0.981 G_ID: 0.155 G_Rec: 0.418 D_GP: 0.042 D_real: 0.917 D_fake: 0.430 +(epoch: 252, iters: 2592, time: 0.064) G_GAN: -0.129 G_GAN_Feat: 0.894 G_ID: 0.132 G_Rec: 0.348 D_GP: 1.274 D_real: 0.261 D_fake: 1.129 +(epoch: 252, iters: 2992, time: 0.064) G_GAN: 0.387 G_GAN_Feat: 0.842 G_ID: 0.180 G_Rec: 0.371 D_GP: 0.030 D_real: 1.075 D_fake: 0.618 +(epoch: 252, iters: 3392, time: 0.064) G_GAN: 0.418 G_GAN_Feat: 0.706 G_ID: 0.126 G_Rec: 0.299 D_GP: 0.020 D_real: 1.357 D_fake: 0.583 +(epoch: 252, iters: 3792, time: 0.064) G_GAN: 0.410 G_GAN_Feat: 1.019 G_ID: 0.171 G_Rec: 0.438 D_GP: 0.056 D_real: 0.731 D_fake: 0.592 +(epoch: 252, iters: 4192, time: 0.064) G_GAN: 0.291 G_GAN_Feat: 0.896 G_ID: 0.113 G_Rec: 0.298 D_GP: 0.037 D_real: 0.733 D_fake: 0.711 +(epoch: 252, iters: 4592, time: 0.064) G_GAN: 0.411 G_GAN_Feat: 0.934 G_ID: 0.159 G_Rec: 0.384 D_GP: 0.023 D_real: 1.040 D_fake: 0.589 +(epoch: 252, iters: 4992, time: 0.064) G_GAN: -0.270 G_GAN_Feat: 0.900 G_ID: 0.124 G_Rec: 0.360 D_GP: 0.214 D_real: 0.575 D_fake: 1.270 +(epoch: 252, iters: 5392, time: 0.064) G_GAN: 0.344 G_GAN_Feat: 0.876 G_ID: 0.176 G_Rec: 0.384 D_GP: 0.047 D_real: 0.913 D_fake: 0.660 +(epoch: 252, iters: 5792, time: 0.064) G_GAN: 0.382 G_GAN_Feat: 0.905 G_ID: 0.142 G_Rec: 0.330 D_GP: 0.174 D_real: 0.712 D_fake: 0.627 +(epoch: 252, iters: 6192, time: 0.064) G_GAN: 0.605 G_GAN_Feat: 1.006 G_ID: 0.173 G_Rec: 0.407 D_GP: 0.031 D_real: 0.970 D_fake: 0.405 +(epoch: 252, iters: 6592, time: 0.064) G_GAN: 0.151 G_GAN_Feat: 0.696 G_ID: 0.123 G_Rec: 0.297 D_GP: 0.024 D_real: 1.104 D_fake: 0.849 +(epoch: 252, iters: 6992, time: 0.064) G_GAN: 0.486 G_GAN_Feat: 1.005 G_ID: 0.162 G_Rec: 0.421 D_GP: 0.078 D_real: 1.027 D_fake: 0.533 +(epoch: 252, iters: 7392, time: 0.064) G_GAN: 0.347 G_GAN_Feat: 0.994 G_ID: 0.139 G_Rec: 0.342 D_GP: 1.859 D_real: 0.348 D_fake: 0.718 +(epoch: 252, iters: 7792, time: 0.064) G_GAN: 0.640 G_GAN_Feat: 0.752 G_ID: 0.161 G_Rec: 0.346 D_GP: 0.023 D_real: 1.411 D_fake: 0.425 +(epoch: 252, iters: 8192, time: 0.064) G_GAN: 0.039 G_GAN_Feat: 0.658 G_ID: 0.113 G_Rec: 0.319 D_GP: 0.026 D_real: 0.901 D_fake: 0.961 +(epoch: 252, iters: 8592, time: 0.064) G_GAN: 0.177 G_GAN_Feat: 1.051 G_ID: 0.178 G_Rec: 0.452 D_GP: 0.072 D_real: 0.979 D_fake: 0.826 +(epoch: 253, iters: 384, time: 0.064) G_GAN: 0.177 G_GAN_Feat: 0.665 G_ID: 0.134 G_Rec: 0.326 D_GP: 0.045 D_real: 1.004 D_fake: 0.823 +(epoch: 253, iters: 784, time: 0.064) G_GAN: 0.487 G_GAN_Feat: 0.953 G_ID: 0.149 G_Rec: 0.381 D_GP: 0.038 D_real: 0.968 D_fake: 0.517 +(epoch: 253, iters: 1184, time: 0.064) G_GAN: 0.169 G_GAN_Feat: 0.676 G_ID: 0.118 G_Rec: 0.285 D_GP: 0.019 D_real: 1.128 D_fake: 0.831 +(epoch: 253, iters: 1584, time: 0.064) G_GAN: 0.669 G_GAN_Feat: 0.953 G_ID: 0.148 G_Rec: 0.419 D_GP: 0.029 D_real: 1.156 D_fake: 0.375 +(epoch: 253, iters: 1984, time: 0.064) G_GAN: 0.359 G_GAN_Feat: 0.829 G_ID: 0.120 G_Rec: 0.358 D_GP: 0.038 D_real: 1.042 D_fake: 0.675 +(epoch: 253, iters: 2384, time: 0.064) G_GAN: 0.419 G_GAN_Feat: 1.011 G_ID: 0.144 G_Rec: 0.422 D_GP: 0.022 D_real: 0.963 D_fake: 0.582 +(epoch: 253, iters: 2784, time: 0.064) G_GAN: -0.106 G_GAN_Feat: 0.732 G_ID: 0.145 G_Rec: 0.287 D_GP: 0.027 D_real: 0.689 D_fake: 1.106 +(epoch: 253, iters: 3184, time: 0.064) G_GAN: 0.267 G_GAN_Feat: 0.849 G_ID: 0.180 G_Rec: 0.409 D_GP: 0.022 D_real: 1.034 D_fake: 0.736 +(epoch: 253, iters: 3584, time: 0.064) G_GAN: -0.003 G_GAN_Feat: 0.740 G_ID: 0.130 G_Rec: 0.319 D_GP: 0.032 D_real: 0.780 D_fake: 1.004 +(epoch: 253, iters: 3984, time: 0.064) G_GAN: 0.708 G_GAN_Feat: 0.823 G_ID: 0.164 G_Rec: 0.369 D_GP: 0.021 D_real: 1.509 D_fake: 0.312 +(epoch: 253, iters: 4384, time: 0.064) G_GAN: 0.299 G_GAN_Feat: 0.644 G_ID: 0.130 G_Rec: 0.308 D_GP: 0.022 D_real: 1.219 D_fake: 0.702 +(epoch: 253, iters: 4784, time: 0.064) G_GAN: 0.069 G_GAN_Feat: 0.972 G_ID: 0.155 G_Rec: 0.415 D_GP: 0.084 D_real: 0.697 D_fake: 0.932 +(epoch: 253, iters: 5184, time: 0.064) G_GAN: 0.425 G_GAN_Feat: 0.891 G_ID: 0.141 G_Rec: 0.337 D_GP: 0.105 D_real: 1.200 D_fake: 0.657 +(epoch: 253, iters: 5584, time: 0.064) G_GAN: 0.556 G_GAN_Feat: 0.963 G_ID: 0.184 G_Rec: 0.416 D_GP: 0.046 D_real: 0.880 D_fake: 0.455 +(epoch: 253, iters: 5984, time: 0.064) G_GAN: 0.193 G_GAN_Feat: 0.747 G_ID: 0.132 G_Rec: 0.302 D_GP: 0.021 D_real: 1.131 D_fake: 0.807 +(epoch: 253, iters: 6384, time: 0.064) G_GAN: 0.178 G_GAN_Feat: 0.765 G_ID: 0.145 G_Rec: 0.347 D_GP: 0.022 D_real: 1.061 D_fake: 0.823 +(epoch: 253, iters: 6784, time: 0.064) G_GAN: 0.324 G_GAN_Feat: 0.834 G_ID: 0.127 G_Rec: 0.324 D_GP: 0.045 D_real: 1.059 D_fake: 0.687 +(epoch: 253, iters: 7184, time: 0.064) G_GAN: 0.168 G_GAN_Feat: 0.875 G_ID: 0.161 G_Rec: 0.400 D_GP: 0.046 D_real: 0.741 D_fake: 0.832 +(epoch: 253, iters: 7584, time: 0.064) G_GAN: 0.244 G_GAN_Feat: 0.665 G_ID: 0.130 G_Rec: 0.317 D_GP: 0.023 D_real: 1.164 D_fake: 0.756 +(epoch: 253, iters: 7984, time: 0.064) G_GAN: 0.314 G_GAN_Feat: 0.873 G_ID: 0.165 G_Rec: 0.441 D_GP: 0.024 D_real: 1.042 D_fake: 0.688 +(epoch: 253, iters: 8384, time: 0.064) G_GAN: 0.223 G_GAN_Feat: 0.724 G_ID: 0.116 G_Rec: 0.306 D_GP: 0.039 D_real: 0.925 D_fake: 0.777 +(epoch: 254, iters: 176, time: 0.064) G_GAN: 0.314 G_GAN_Feat: 0.980 G_ID: 0.180 G_Rec: 0.387 D_GP: 0.035 D_real: 0.634 D_fake: 0.688 +(epoch: 254, iters: 576, time: 0.064) G_GAN: 0.316 G_GAN_Feat: 0.756 G_ID: 0.139 G_Rec: 0.311 D_GP: 0.030 D_real: 0.992 D_fake: 0.687 +(epoch: 254, iters: 976, time: 0.064) G_GAN: 0.490 G_GAN_Feat: 0.866 G_ID: 0.180 G_Rec: 0.332 D_GP: 0.027 D_real: 1.280 D_fake: 0.515 +(epoch: 254, iters: 1376, time: 0.064) G_GAN: 0.524 G_GAN_Feat: 0.729 G_ID: 0.117 G_Rec: 0.305 D_GP: 0.036 D_real: 1.297 D_fake: 0.485 +(epoch: 254, iters: 1776, time: 0.064) G_GAN: 0.540 G_GAN_Feat: 0.801 G_ID: 0.195 G_Rec: 0.386 D_GP: 0.019 D_real: 1.445 D_fake: 0.480 +(epoch: 254, iters: 2176, time: 0.064) G_GAN: 0.231 G_GAN_Feat: 0.772 G_ID: 0.160 G_Rec: 0.333 D_GP: 0.039 D_real: 1.009 D_fake: 0.769 +(epoch: 254, iters: 2576, time: 0.064) G_GAN: 0.262 G_GAN_Feat: 0.821 G_ID: 0.150 G_Rec: 0.381 D_GP: 0.021 D_real: 1.024 D_fake: 0.739 +(epoch: 254, iters: 2976, time: 0.064) G_GAN: 0.259 G_GAN_Feat: 0.831 G_ID: 0.136 G_Rec: 0.332 D_GP: 0.074 D_real: 0.586 D_fake: 0.743 +(epoch: 254, iters: 3376, time: 0.064) G_GAN: 0.508 G_GAN_Feat: 0.909 G_ID: 0.169 G_Rec: 0.409 D_GP: 0.032 D_real: 1.176 D_fake: 0.497 +(epoch: 254, iters: 3776, time: 0.064) G_GAN: 0.041 G_GAN_Feat: 0.895 G_ID: 0.126 G_Rec: 0.315 D_GP: 0.125 D_real: 0.434 D_fake: 0.959 +(epoch: 254, iters: 4176, time: 0.064) G_GAN: 0.877 G_GAN_Feat: 1.115 G_ID: 0.180 G_Rec: 0.425 D_GP: 0.098 D_real: 1.177 D_fake: 0.408 +(epoch: 254, iters: 4576, time: 0.064) G_GAN: 0.233 G_GAN_Feat: 0.701 G_ID: 0.128 G_Rec: 0.300 D_GP: 0.031 D_real: 1.119 D_fake: 0.768 +(epoch: 254, iters: 4976, time: 0.064) G_GAN: 0.474 G_GAN_Feat: 0.810 G_ID: 0.174 G_Rec: 0.407 D_GP: 0.020 D_real: 1.239 D_fake: 0.530 +(epoch: 254, iters: 5376, time: 0.064) G_GAN: 0.271 G_GAN_Feat: 0.621 G_ID: 0.119 G_Rec: 0.294 D_GP: 0.024 D_real: 1.268 D_fake: 0.729 +(epoch: 254, iters: 5776, time: 0.064) G_GAN: 0.773 G_GAN_Feat: 1.024 G_ID: 0.177 G_Rec: 0.438 D_GP: 0.046 D_real: 1.067 D_fake: 0.368 +(epoch: 254, iters: 6176, time: 0.064) G_GAN: 0.247 G_GAN_Feat: 0.684 G_ID: 0.138 G_Rec: 0.344 D_GP: 0.018 D_real: 1.153 D_fake: 0.754 +(epoch: 254, iters: 6576, time: 0.064) G_GAN: 0.539 G_GAN_Feat: 0.784 G_ID: 0.146 G_Rec: 0.370 D_GP: 0.023 D_real: 1.307 D_fake: 0.464 +(epoch: 254, iters: 6976, time: 0.064) G_GAN: 0.231 G_GAN_Feat: 0.590 G_ID: 0.141 G_Rec: 0.340 D_GP: 0.020 D_real: 1.228 D_fake: 0.769 +(epoch: 254, iters: 7376, time: 0.064) G_GAN: 0.457 G_GAN_Feat: 0.793 G_ID: 0.168 G_Rec: 0.387 D_GP: 0.027 D_real: 1.169 D_fake: 0.546 +(epoch: 254, iters: 7776, time: 0.064) G_GAN: 0.269 G_GAN_Feat: 0.601 G_ID: 0.113 G_Rec: 0.289 D_GP: 0.024 D_real: 1.148 D_fake: 0.731 +(epoch: 254, iters: 8176, time: 0.064) G_GAN: 0.481 G_GAN_Feat: 0.815 G_ID: 0.162 G_Rec: 0.422 D_GP: 0.029 D_real: 1.164 D_fake: 0.527 +(epoch: 254, iters: 8576, time: 0.064) G_GAN: 0.195 G_GAN_Feat: 0.736 G_ID: 0.127 G_Rec: 0.306 D_GP: 0.026 D_real: 1.053 D_fake: 0.805 +(epoch: 255, iters: 368, time: 0.064) G_GAN: 0.360 G_GAN_Feat: 0.873 G_ID: 0.159 G_Rec: 0.364 D_GP: 0.029 D_real: 0.874 D_fake: 0.640 +(epoch: 255, iters: 768, time: 0.064) G_GAN: 0.027 G_GAN_Feat: 0.662 G_ID: 0.170 G_Rec: 0.283 D_GP: 0.038 D_real: 0.812 D_fake: 0.974 +(epoch: 255, iters: 1168, time: 0.064) G_GAN: 0.348 G_GAN_Feat: 0.924 G_ID: 0.186 G_Rec: 0.393 D_GP: 0.068 D_real: 0.858 D_fake: 0.657 +(epoch: 255, iters: 1568, time: 0.064) G_GAN: 0.060 G_GAN_Feat: 0.751 G_ID: 0.123 G_Rec: 0.306 D_GP: 0.071 D_real: 0.651 D_fake: 0.941 +(epoch: 255, iters: 1968, time: 0.064) G_GAN: 0.300 G_GAN_Feat: 0.905 G_ID: 0.179 G_Rec: 0.366 D_GP: 0.053 D_real: 0.744 D_fake: 0.700 +(epoch: 255, iters: 2368, time: 0.064) G_GAN: 0.454 G_GAN_Feat: 0.714 G_ID: 0.120 G_Rec: 0.300 D_GP: 0.028 D_real: 1.241 D_fake: 0.547 +(epoch: 255, iters: 2768, time: 0.064) G_GAN: 0.742 G_GAN_Feat: 1.026 G_ID: 0.193 G_Rec: 0.520 D_GP: 0.081 D_real: 1.132 D_fake: 0.512 +(epoch: 255, iters: 3168, time: 0.064) G_GAN: 0.076 G_GAN_Feat: 0.638 G_ID: 0.148 G_Rec: 0.276 D_GP: 0.021 D_real: 0.994 D_fake: 0.924 +(epoch: 255, iters: 3568, time: 0.064) G_GAN: 0.437 G_GAN_Feat: 0.865 G_ID: 0.168 G_Rec: 0.394 D_GP: 0.022 D_real: 1.102 D_fake: 0.566 +(epoch: 255, iters: 3968, time: 0.064) G_GAN: 0.085 G_GAN_Feat: 0.642 G_ID: 0.170 G_Rec: 0.272 D_GP: 0.021 D_real: 1.055 D_fake: 0.915 +(epoch: 255, iters: 4368, time: 0.064) G_GAN: 0.256 G_GAN_Feat: 0.998 G_ID: 0.182 G_Rec: 0.439 D_GP: 0.081 D_real: 0.579 D_fake: 0.745 +(epoch: 255, iters: 4768, time: 0.064) G_GAN: 0.075 G_GAN_Feat: 0.612 G_ID: 0.132 G_Rec: 0.271 D_GP: 0.023 D_real: 1.116 D_fake: 0.925 +(epoch: 255, iters: 5168, time: 0.064) G_GAN: 0.198 G_GAN_Feat: 0.938 G_ID: 0.166 G_Rec: 0.427 D_GP: 0.046 D_real: 0.736 D_fake: 0.805 +(epoch: 255, iters: 5568, time: 0.064) G_GAN: 0.102 G_GAN_Feat: 0.692 G_ID: 0.117 G_Rec: 0.282 D_GP: 0.038 D_real: 0.896 D_fake: 0.898 +(epoch: 255, iters: 5968, time: 0.064) G_GAN: 0.619 G_GAN_Feat: 0.904 G_ID: 0.197 G_Rec: 0.378 D_GP: 0.026 D_real: 1.283 D_fake: 0.390 +(epoch: 255, iters: 6368, time: 0.064) G_GAN: 0.249 G_GAN_Feat: 0.748 G_ID: 0.157 G_Rec: 0.294 D_GP: 0.033 D_real: 1.039 D_fake: 0.751 +(epoch: 255, iters: 6768, time: 0.064) G_GAN: 0.616 G_GAN_Feat: 0.956 G_ID: 0.176 G_Rec: 0.386 D_GP: 0.025 D_real: 1.307 D_fake: 0.387 +(epoch: 255, iters: 7168, time: 0.064) G_GAN: 0.101 G_GAN_Feat: 0.959 G_ID: 0.106 G_Rec: 0.350 D_GP: 0.049 D_real: 0.576 D_fake: 0.900 +(epoch: 255, iters: 7568, time: 0.064) G_GAN: 0.455 G_GAN_Feat: 0.973 G_ID: 0.151 G_Rec: 0.413 D_GP: 0.033 D_real: 1.086 D_fake: 0.547 +(epoch: 255, iters: 7968, time: 0.064) G_GAN: 0.232 G_GAN_Feat: 0.940 G_ID: 0.131 G_Rec: 0.304 D_GP: 0.188 D_real: 0.388 D_fake: 0.768 +(epoch: 255, iters: 8368, time: 0.064) G_GAN: 0.638 G_GAN_Feat: 0.841 G_ID: 0.175 G_Rec: 0.402 D_GP: 0.022 D_real: 1.380 D_fake: 0.388 +(epoch: 256, iters: 160, time: 0.064) G_GAN: 0.198 G_GAN_Feat: 0.924 G_ID: 0.137 G_Rec: 0.323 D_GP: 0.034 D_real: 0.539 D_fake: 0.803 +(epoch: 256, iters: 560, time: 0.064) G_GAN: 0.530 G_GAN_Feat: 0.837 G_ID: 0.155 G_Rec: 0.408 D_GP: 0.033 D_real: 1.351 D_fake: 0.499 +(epoch: 256, iters: 960, time: 0.064) G_GAN: 0.012 G_GAN_Feat: 0.561 G_ID: 0.125 G_Rec: 0.288 D_GP: 0.018 D_real: 0.976 D_fake: 0.988 +(epoch: 256, iters: 1360, time: 0.064) G_GAN: 0.215 G_GAN_Feat: 0.717 G_ID: 0.169 G_Rec: 0.383 D_GP: 0.021 D_real: 1.057 D_fake: 0.788 +(epoch: 256, iters: 1760, time: 0.064) G_GAN: 0.070 G_GAN_Feat: 0.603 G_ID: 0.144 G_Rec: 0.347 D_GP: 0.018 D_real: 1.017 D_fake: 0.930 +(epoch: 256, iters: 2160, time: 0.064) G_GAN: 0.178 G_GAN_Feat: 0.853 G_ID: 0.147 G_Rec: 0.435 D_GP: 0.039 D_real: 0.944 D_fake: 0.822 +(epoch: 256, iters: 2560, time: 0.064) G_GAN: 0.027 G_GAN_Feat: 0.653 G_ID: 0.138 G_Rec: 0.315 D_GP: 0.032 D_real: 0.856 D_fake: 0.973 +(epoch: 256, iters: 2960, time: 0.064) G_GAN: 0.241 G_GAN_Feat: 0.743 G_ID: 0.167 G_Rec: 0.346 D_GP: 0.026 D_real: 1.061 D_fake: 0.761 +(epoch: 256, iters: 3360, time: 0.064) G_GAN: -0.107 G_GAN_Feat: 0.659 G_ID: 0.108 G_Rec: 0.315 D_GP: 0.042 D_real: 0.711 D_fake: 1.107 +(epoch: 256, iters: 3760, time: 0.064) G_GAN: 0.154 G_GAN_Feat: 0.899 G_ID: 0.172 G_Rec: 0.434 D_GP: 0.040 D_real: 0.755 D_fake: 0.847 +(epoch: 256, iters: 4160, time: 0.064) G_GAN: 0.002 G_GAN_Feat: 0.687 G_ID: 0.121 G_Rec: 0.298 D_GP: 0.064 D_real: 0.812 D_fake: 0.998 +(epoch: 256, iters: 4560, time: 0.064) G_GAN: 0.092 G_GAN_Feat: 0.851 G_ID: 0.204 G_Rec: 0.395 D_GP: 0.034 D_real: 0.785 D_fake: 0.908 +(epoch: 256, iters: 4960, time: 0.064) G_GAN: 0.429 G_GAN_Feat: 0.605 G_ID: 0.129 G_Rec: 0.264 D_GP: 0.021 D_real: 1.341 D_fake: 0.572 +(epoch: 256, iters: 5360, time: 0.064) G_GAN: 0.429 G_GAN_Feat: 0.871 G_ID: 0.159 G_Rec: 0.359 D_GP: 0.028 D_real: 1.068 D_fake: 0.575 +(epoch: 256, iters: 5760, time: 0.064) G_GAN: 0.027 G_GAN_Feat: 0.726 G_ID: 0.130 G_Rec: 0.313 D_GP: 0.026 D_real: 0.803 D_fake: 0.973 +(epoch: 256, iters: 6160, time: 0.064) G_GAN: 0.464 G_GAN_Feat: 0.903 G_ID: 0.176 G_Rec: 0.402 D_GP: 0.040 D_real: 0.989 D_fake: 0.542 +(epoch: 256, iters: 6560, time: 0.064) G_GAN: 0.335 G_GAN_Feat: 0.615 G_ID: 0.161 G_Rec: 0.303 D_GP: 0.022 D_real: 1.309 D_fake: 0.668 +(epoch: 256, iters: 6960, time: 0.064) G_GAN: 0.444 G_GAN_Feat: 0.858 G_ID: 0.173 G_Rec: 0.397 D_GP: 0.033 D_real: 1.068 D_fake: 0.558 +(epoch: 256, iters: 7360, time: 0.064) G_GAN: 0.382 G_GAN_Feat: 0.574 G_ID: 0.140 G_Rec: 0.255 D_GP: 0.019 D_real: 1.418 D_fake: 0.654 +(epoch: 256, iters: 7760, time: 0.064) G_GAN: 0.432 G_GAN_Feat: 0.873 G_ID: 0.182 G_Rec: 0.415 D_GP: 0.028 D_real: 1.126 D_fake: 0.573 +(epoch: 256, iters: 8160, time: 0.064) G_GAN: 0.276 G_GAN_Feat: 0.758 G_ID: 0.134 G_Rec: 0.318 D_GP: 0.065 D_real: 0.909 D_fake: 0.725 +(epoch: 256, iters: 8560, time: 0.064) G_GAN: 0.389 G_GAN_Feat: 0.834 G_ID: 0.167 G_Rec: 0.367 D_GP: 0.028 D_real: 1.103 D_fake: 0.614 +(epoch: 257, iters: 352, time: 0.064) G_GAN: 0.237 G_GAN_Feat: 0.684 G_ID: 0.131 G_Rec: 0.275 D_GP: 0.021 D_real: 1.125 D_fake: 0.763 +(epoch: 257, iters: 752, time: 0.064) G_GAN: 0.492 G_GAN_Feat: 0.848 G_ID: 0.158 G_Rec: 0.375 D_GP: 0.023 D_real: 1.257 D_fake: 0.511 +(epoch: 257, iters: 1152, time: 0.064) G_GAN: 0.407 G_GAN_Feat: 0.785 G_ID: 0.155 G_Rec: 0.354 D_GP: 0.043 D_real: 1.104 D_fake: 0.595 +(epoch: 257, iters: 1552, time: 0.064) G_GAN: 0.309 G_GAN_Feat: 0.999 G_ID: 0.177 G_Rec: 0.427 D_GP: 0.023 D_real: 0.785 D_fake: 0.692 +(epoch: 257, iters: 1952, time: 0.064) G_GAN: 0.265 G_GAN_Feat: 0.805 G_ID: 0.142 G_Rec: 0.297 D_GP: 0.066 D_real: 0.611 D_fake: 0.736 +(epoch: 257, iters: 2352, time: 0.064) G_GAN: 0.720 G_GAN_Feat: 1.041 G_ID: 0.157 G_Rec: 0.412 D_GP: 0.033 D_real: 1.179 D_fake: 0.351 +(epoch: 257, iters: 2752, time: 0.064) G_GAN: 0.142 G_GAN_Feat: 0.761 G_ID: 0.129 G_Rec: 0.313 D_GP: 0.037 D_real: 0.887 D_fake: 0.860 +(epoch: 257, iters: 3152, time: 0.064) G_GAN: 0.111 G_GAN_Feat: 0.789 G_ID: 0.157 G_Rec: 0.376 D_GP: 0.017 D_real: 0.963 D_fake: 0.892 +(epoch: 257, iters: 3552, time: 0.064) G_GAN: 0.148 G_GAN_Feat: 0.644 G_ID: 0.116 G_Rec: 0.284 D_GP: 0.027 D_real: 1.049 D_fake: 0.852 +(epoch: 257, iters: 3952, time: 0.064) G_GAN: 0.385 G_GAN_Feat: 0.812 G_ID: 0.177 G_Rec: 0.384 D_GP: 0.024 D_real: 1.200 D_fake: 0.615 +(epoch: 257, iters: 4352, time: 0.064) G_GAN: 0.429 G_GAN_Feat: 0.728 G_ID: 0.109 G_Rec: 0.290 D_GP: 0.033 D_real: 1.247 D_fake: 0.572 +(epoch: 257, iters: 4752, time: 0.064) G_GAN: 0.599 G_GAN_Feat: 0.846 G_ID: 0.190 G_Rec: 0.391 D_GP: 0.034 D_real: 1.251 D_fake: 0.461 +(epoch: 257, iters: 5152, time: 0.064) G_GAN: -0.062 G_GAN_Feat: 0.698 G_ID: 0.105 G_Rec: 0.296 D_GP: 0.030 D_real: 0.755 D_fake: 1.062 +(epoch: 257, iters: 5552, time: 0.064) G_GAN: 0.216 G_GAN_Feat: 0.921 G_ID: 0.170 G_Rec: 0.413 D_GP: 0.046 D_real: 0.697 D_fake: 0.785 +(epoch: 257, iters: 5952, time: 0.064) G_GAN: -0.057 G_GAN_Feat: 0.835 G_ID: 0.132 G_Rec: 0.307 D_GP: 0.036 D_real: 1.008 D_fake: 1.060 +(epoch: 257, iters: 6352, time: 0.064) G_GAN: 0.503 G_GAN_Feat: 0.977 G_ID: 0.157 G_Rec: 0.424 D_GP: 0.039 D_real: 0.931 D_fake: 0.500 +(epoch: 257, iters: 6752, time: 0.064) G_GAN: 0.142 G_GAN_Feat: 0.646 G_ID: 0.131 G_Rec: 0.328 D_GP: 0.018 D_real: 1.129 D_fake: 0.858 +(epoch: 257, iters: 7152, time: 0.064) G_GAN: 0.327 G_GAN_Feat: 0.840 G_ID: 0.181 G_Rec: 0.407 D_GP: 0.021 D_real: 1.089 D_fake: 0.675 +(epoch: 257, iters: 7552, time: 0.064) G_GAN: 0.309 G_GAN_Feat: 0.678 G_ID: 0.113 G_Rec: 0.305 D_GP: 0.020 D_real: 1.227 D_fake: 0.691 +(epoch: 257, iters: 7952, time: 0.064) G_GAN: 0.463 G_GAN_Feat: 0.916 G_ID: 0.169 G_Rec: 0.404 D_GP: 0.047 D_real: 0.925 D_fake: 0.547 +(epoch: 257, iters: 8352, time: 0.064) G_GAN: 0.147 G_GAN_Feat: 0.763 G_ID: 0.123 G_Rec: 0.335 D_GP: 0.051 D_real: 0.914 D_fake: 0.853 +(epoch: 258, iters: 144, time: 0.064) G_GAN: 0.988 G_GAN_Feat: 1.237 G_ID: 0.168 G_Rec: 0.407 D_GP: 0.028 D_real: 1.428 D_fake: 0.232 +(epoch: 258, iters: 544, time: 0.064) G_GAN: -0.014 G_GAN_Feat: 0.724 G_ID: 0.148 G_Rec: 0.304 D_GP: 0.031 D_real: 0.816 D_fake: 1.014 +(epoch: 258, iters: 944, time: 0.064) G_GAN: 0.429 G_GAN_Feat: 0.978 G_ID: 0.170 G_Rec: 0.422 D_GP: 0.075 D_real: 0.904 D_fake: 0.577 +(epoch: 258, iters: 1344, time: 0.064) G_GAN: 0.288 G_GAN_Feat: 0.712 G_ID: 0.118 G_Rec: 0.307 D_GP: 0.020 D_real: 1.128 D_fake: 0.712 +(epoch: 258, iters: 1744, time: 0.064) G_GAN: 0.483 G_GAN_Feat: 0.939 G_ID: 0.179 G_Rec: 0.380 D_GP: 0.066 D_real: 0.840 D_fake: 0.522 +(epoch: 258, iters: 2144, time: 0.064) G_GAN: -0.038 G_GAN_Feat: 0.841 G_ID: 0.138 G_Rec: 0.301 D_GP: 0.099 D_real: 0.392 D_fake: 1.038 +(epoch: 258, iters: 2544, time: 0.064) G_GAN: 0.495 G_GAN_Feat: 0.882 G_ID: 0.144 G_Rec: 0.401 D_GP: 0.021 D_real: 1.239 D_fake: 0.507 +(epoch: 258, iters: 2944, time: 0.064) G_GAN: 0.357 G_GAN_Feat: 0.831 G_ID: 0.176 G_Rec: 0.314 D_GP: 0.035 D_real: 0.945 D_fake: 0.652 +(epoch: 258, iters: 3344, time: 0.064) G_GAN: 0.743 G_GAN_Feat: 0.986 G_ID: 0.170 G_Rec: 0.395 D_GP: 0.066 D_real: 1.195 D_fake: 0.299 +(epoch: 258, iters: 3744, time: 0.064) G_GAN: 0.388 G_GAN_Feat: 0.793 G_ID: 0.124 G_Rec: 0.306 D_GP: 0.040 D_real: 1.041 D_fake: 0.615 +(epoch: 258, iters: 4144, time: 0.064) G_GAN: 0.358 G_GAN_Feat: 0.787 G_ID: 0.162 G_Rec: 0.385 D_GP: 0.019 D_real: 1.194 D_fake: 0.644 +(epoch: 258, iters: 4544, time: 0.064) G_GAN: 0.111 G_GAN_Feat: 0.629 G_ID: 0.125 G_Rec: 0.315 D_GP: 0.019 D_real: 1.087 D_fake: 0.889 +(epoch: 258, iters: 4944, time: 0.064) G_GAN: 0.262 G_GAN_Feat: 0.816 G_ID: 0.186 G_Rec: 0.388 D_GP: 0.024 D_real: 0.961 D_fake: 0.739 +(epoch: 258, iters: 5344, time: 0.064) G_GAN: 0.052 G_GAN_Feat: 0.645 G_ID: 0.135 G_Rec: 0.297 D_GP: 0.024 D_real: 1.021 D_fake: 0.948 +(epoch: 258, iters: 5744, time: 0.064) G_GAN: 0.456 G_GAN_Feat: 0.960 G_ID: 0.161 G_Rec: 0.443 D_GP: 0.106 D_real: 0.957 D_fake: 0.558 +(epoch: 258, iters: 6144, time: 0.064) G_GAN: 0.112 G_GAN_Feat: 0.678 G_ID: 0.114 G_Rec: 0.319 D_GP: 0.026 D_real: 0.985 D_fake: 0.888 +(epoch: 258, iters: 6544, time: 0.064) G_GAN: 0.357 G_GAN_Feat: 0.788 G_ID: 0.169 G_Rec: 0.400 D_GP: 0.023 D_real: 1.094 D_fake: 0.645 +(epoch: 258, iters: 6944, time: 0.064) G_GAN: 0.170 G_GAN_Feat: 0.781 G_ID: 0.158 G_Rec: 0.318 D_GP: 0.105 D_real: 0.717 D_fake: 0.834 +(epoch: 258, iters: 7344, time: 0.064) G_GAN: 0.354 G_GAN_Feat: 0.819 G_ID: 0.157 G_Rec: 0.359 D_GP: 0.031 D_real: 1.080 D_fake: 0.647 +(epoch: 258, iters: 7744, time: 0.064) G_GAN: 0.112 G_GAN_Feat: 0.777 G_ID: 0.121 G_Rec: 0.316 D_GP: 0.025 D_real: 1.255 D_fake: 0.890 +(epoch: 258, iters: 8144, time: 0.064) G_GAN: 0.363 G_GAN_Feat: 0.777 G_ID: 0.167 G_Rec: 0.410 D_GP: 0.018 D_real: 1.153 D_fake: 0.638 +(epoch: 258, iters: 8544, time: 0.064) G_GAN: 0.270 G_GAN_Feat: 0.535 G_ID: 0.125 G_Rec: 0.270 D_GP: 0.019 D_real: 1.224 D_fake: 0.730 +(epoch: 259, iters: 336, time: 0.064) G_GAN: 0.431 G_GAN_Feat: 0.771 G_ID: 0.164 G_Rec: 0.410 D_GP: 0.025 D_real: 1.174 D_fake: 0.572 +(epoch: 259, iters: 736, time: 0.064) G_GAN: 0.109 G_GAN_Feat: 0.629 G_ID: 0.131 G_Rec: 0.317 D_GP: 0.030 D_real: 1.038 D_fake: 0.892 +(epoch: 259, iters: 1136, time: 0.064) G_GAN: 0.351 G_GAN_Feat: 0.771 G_ID: 0.186 G_Rec: 0.357 D_GP: 0.035 D_real: 1.058 D_fake: 0.652 +(epoch: 259, iters: 1536, time: 0.064) G_GAN: 0.069 G_GAN_Feat: 0.673 G_ID: 0.126 G_Rec: 0.301 D_GP: 0.049 D_real: 0.893 D_fake: 0.931 +(epoch: 259, iters: 1936, time: 0.064) G_GAN: 0.257 G_GAN_Feat: 0.863 G_ID: 0.168 G_Rec: 0.383 D_GP: 0.024 D_real: 1.025 D_fake: 0.743 +(epoch: 259, iters: 2336, time: 0.064) G_GAN: 0.101 G_GAN_Feat: 0.625 G_ID: 0.130 G_Rec: 0.254 D_GP: 0.033 D_real: 1.013 D_fake: 0.899 +(epoch: 259, iters: 2736, time: 0.064) G_GAN: 0.217 G_GAN_Feat: 0.753 G_ID: 0.172 G_Rec: 0.351 D_GP: 0.025 D_real: 1.052 D_fake: 0.783 +(epoch: 259, iters: 3136, time: 0.064) G_GAN: 0.045 G_GAN_Feat: 0.705 G_ID: 0.135 G_Rec: 0.310 D_GP: 0.024 D_real: 0.908 D_fake: 0.955 +(epoch: 259, iters: 3536, time: 0.064) G_GAN: 0.362 G_GAN_Feat: 0.840 G_ID: 0.175 G_Rec: 0.368 D_GP: 0.030 D_real: 1.090 D_fake: 0.640 +(epoch: 259, iters: 3936, time: 0.064) G_GAN: 0.133 G_GAN_Feat: 0.737 G_ID: 0.109 G_Rec: 0.312 D_GP: 0.059 D_real: 0.897 D_fake: 0.868 +(epoch: 259, iters: 4336, time: 0.064) G_GAN: 0.418 G_GAN_Feat: 0.934 G_ID: 0.151 G_Rec: 0.391 D_GP: 0.025 D_real: 1.088 D_fake: 0.587 +(epoch: 259, iters: 4736, time: 0.064) G_GAN: 0.041 G_GAN_Feat: 0.678 G_ID: 0.125 G_Rec: 0.312 D_GP: 0.026 D_real: 0.923 D_fake: 0.959 +(epoch: 259, iters: 5136, time: 0.064) G_GAN: 0.393 G_GAN_Feat: 0.770 G_ID: 0.160 G_Rec: 0.342 D_GP: 0.024 D_real: 1.168 D_fake: 0.608 +(epoch: 259, iters: 5536, time: 0.064) G_GAN: 0.138 G_GAN_Feat: 0.859 G_ID: 0.136 G_Rec: 0.323 D_GP: 0.055 D_real: 0.537 D_fake: 0.862 +(epoch: 259, iters: 5936, time: 0.064) G_GAN: 0.324 G_GAN_Feat: 0.958 G_ID: 0.161 G_Rec: 0.369 D_GP: 0.039 D_real: 0.743 D_fake: 0.678 +(epoch: 259, iters: 6336, time: 0.064) G_GAN: 0.263 G_GAN_Feat: 0.751 G_ID: 0.116 G_Rec: 0.332 D_GP: 0.036 D_real: 1.049 D_fake: 0.737 +(epoch: 259, iters: 6736, time: 0.064) G_GAN: 0.225 G_GAN_Feat: 1.081 G_ID: 0.173 G_Rec: 0.493 D_GP: 0.209 D_real: 0.368 D_fake: 0.777 +(epoch: 259, iters: 7136, time: 0.064) G_GAN: 0.237 G_GAN_Feat: 0.647 G_ID: 0.125 G_Rec: 0.277 D_GP: 0.018 D_real: 1.166 D_fake: 0.763 +(epoch: 259, iters: 7536, time: 0.064) G_GAN: 0.465 G_GAN_Feat: 0.885 G_ID: 0.161 G_Rec: 0.404 D_GP: 0.038 D_real: 1.173 D_fake: 0.539 +(epoch: 259, iters: 7936, time: 0.064) G_GAN: 0.348 G_GAN_Feat: 0.721 G_ID: 0.130 G_Rec: 0.302 D_GP: 0.029 D_real: 1.130 D_fake: 0.652 +(epoch: 259, iters: 8336, time: 0.064) G_GAN: 0.852 G_GAN_Feat: 1.092 G_ID: 0.154 G_Rec: 0.452 D_GP: 0.084 D_real: 1.379 D_fake: 0.309 +(epoch: 260, iters: 128, time: 0.064) G_GAN: -0.148 G_GAN_Feat: 0.699 G_ID: 0.131 G_Rec: 0.286 D_GP: 0.021 D_real: 0.904 D_fake: 1.149 +(epoch: 260, iters: 528, time: 0.064) G_GAN: 0.498 G_GAN_Feat: 0.842 G_ID: 0.154 G_Rec: 0.488 D_GP: 0.028 D_real: 1.133 D_fake: 0.502 +(epoch: 260, iters: 928, time: 0.064) G_GAN: 0.246 G_GAN_Feat: 0.911 G_ID: 0.146 G_Rec: 0.322 D_GP: 0.298 D_real: 0.357 D_fake: 0.768 +(epoch: 260, iters: 1328, time: 0.064) G_GAN: 0.813 G_GAN_Feat: 1.025 G_ID: 0.159 G_Rec: 0.402 D_GP: 0.039 D_real: 1.064 D_fake: 0.234 +(epoch: 260, iters: 1728, time: 0.064) G_GAN: 0.100 G_GAN_Feat: 0.544 G_ID: 0.130 G_Rec: 0.256 D_GP: 0.019 D_real: 1.085 D_fake: 0.900 +(epoch: 260, iters: 2128, time: 0.064) G_GAN: 0.423 G_GAN_Feat: 0.784 G_ID: 0.194 G_Rec: 0.383 D_GP: 0.019 D_real: 1.239 D_fake: 0.578 +(epoch: 260, iters: 2528, time: 0.064) G_GAN: 0.151 G_GAN_Feat: 0.577 G_ID: 0.137 G_Rec: 0.286 D_GP: 0.020 D_real: 1.046 D_fake: 0.849 +(epoch: 260, iters: 2928, time: 0.064) G_GAN: 0.519 G_GAN_Feat: 0.835 G_ID: 0.130 G_Rec: 0.403 D_GP: 0.030 D_real: 1.279 D_fake: 0.494 +(epoch: 260, iters: 3328, time: 0.064) G_GAN: 0.119 G_GAN_Feat: 0.568 G_ID: 0.125 G_Rec: 0.312 D_GP: 0.028 D_real: 1.052 D_fake: 0.881 +(epoch: 260, iters: 3728, time: 0.064) G_GAN: 0.354 G_GAN_Feat: 0.843 G_ID: 0.156 G_Rec: 0.418 D_GP: 0.031 D_real: 0.989 D_fake: 0.649 +(epoch: 260, iters: 4128, time: 0.064) G_GAN: -0.029 G_GAN_Feat: 0.738 G_ID: 0.135 G_Rec: 0.304 D_GP: 0.066 D_real: 0.706 D_fake: 1.029 +(epoch: 260, iters: 4528, time: 0.064) G_GAN: 0.518 G_GAN_Feat: 0.890 G_ID: 0.144 G_Rec: 0.439 D_GP: 0.030 D_real: 1.088 D_fake: 0.491 +(epoch: 260, iters: 4928, time: 0.064) G_GAN: 0.185 G_GAN_Feat: 0.639 G_ID: 0.135 G_Rec: 0.266 D_GP: 0.026 D_real: 1.089 D_fake: 0.816 +(epoch: 260, iters: 5328, time: 0.064) G_GAN: 0.558 G_GAN_Feat: 0.925 G_ID: 0.151 G_Rec: 0.404 D_GP: 0.031 D_real: 1.156 D_fake: 0.446 +(epoch: 260, iters: 5728, time: 0.064) G_GAN: 0.159 G_GAN_Feat: 0.673 G_ID: 0.113 G_Rec: 0.282 D_GP: 0.040 D_real: 1.025 D_fake: 0.841 +(epoch: 260, iters: 6128, time: 0.064) G_GAN: 0.522 G_GAN_Feat: 0.975 G_ID: 0.162 G_Rec: 0.368 D_GP: 0.101 D_real: 0.872 D_fake: 0.483 +(epoch: 260, iters: 6528, time: 0.064) G_GAN: -0.072 G_GAN_Feat: 0.804 G_ID: 0.151 G_Rec: 0.341 D_GP: 0.188 D_real: 0.485 D_fake: 1.073 +(epoch: 260, iters: 6928, time: 0.064) G_GAN: 0.436 G_GAN_Feat: 0.912 G_ID: 0.165 G_Rec: 0.404 D_GP: 0.037 D_real: 1.088 D_fake: 0.572 +(epoch: 260, iters: 7328, time: 0.064) G_GAN: 0.091 G_GAN_Feat: 0.705 G_ID: 0.154 G_Rec: 0.299 D_GP: 0.029 D_real: 0.948 D_fake: 0.909 +(epoch: 260, iters: 7728, time: 0.064) G_GAN: 0.383 G_GAN_Feat: 0.952 G_ID: 0.173 G_Rec: 0.415 D_GP: 0.068 D_real: 0.863 D_fake: 0.621 +(epoch: 260, iters: 8128, time: 0.064) G_GAN: 0.192 G_GAN_Feat: 0.637 G_ID: 0.119 G_Rec: 0.265 D_GP: 0.022 D_real: 1.161 D_fake: 0.808 +(epoch: 260, iters: 8528, time: 0.064) G_GAN: 0.560 G_GAN_Feat: 0.947 G_ID: 0.168 G_Rec: 0.430 D_GP: 0.032 D_real: 1.043 D_fake: 0.445 +(epoch: 261, iters: 320, time: 0.064) G_GAN: 0.326 G_GAN_Feat: 0.850 G_ID: 0.128 G_Rec: 0.289 D_GP: 0.033 D_real: 0.825 D_fake: 0.676 +(epoch: 261, iters: 720, time: 0.064) G_GAN: 0.398 G_GAN_Feat: 0.864 G_ID: 0.187 G_Rec: 0.379 D_GP: 0.029 D_real: 1.028 D_fake: 0.609 +(epoch: 261, iters: 1120, time: 0.064) G_GAN: 0.221 G_GAN_Feat: 0.677 G_ID: 0.125 G_Rec: 0.298 D_GP: 0.026 D_real: 1.143 D_fake: 0.779 +(epoch: 261, iters: 1520, time: 0.064) G_GAN: 0.257 G_GAN_Feat: 0.848 G_ID: 0.178 G_Rec: 0.403 D_GP: 0.022 D_real: 1.007 D_fake: 0.743 +(epoch: 261, iters: 1920, time: 0.064) G_GAN: 0.280 G_GAN_Feat: 0.864 G_ID: 0.123 G_Rec: 0.339 D_GP: 0.068 D_real: 0.935 D_fake: 0.723 +(epoch: 261, iters: 2320, time: 0.064) G_GAN: 0.529 G_GAN_Feat: 0.919 G_ID: 0.168 G_Rec: 0.359 D_GP: 0.052 D_real: 0.871 D_fake: 0.476 +(epoch: 261, iters: 2720, time: 0.064) G_GAN: 0.093 G_GAN_Feat: 0.636 G_ID: 0.121 G_Rec: 0.311 D_GP: 0.019 D_real: 1.090 D_fake: 0.907 +(epoch: 261, iters: 3120, time: 0.064) G_GAN: 0.336 G_GAN_Feat: 0.816 G_ID: 0.172 G_Rec: 0.397 D_GP: 0.021 D_real: 1.130 D_fake: 0.666 +(epoch: 261, iters: 3520, time: 0.064) G_GAN: 0.272 G_GAN_Feat: 0.586 G_ID: 0.126 G_Rec: 0.289 D_GP: 0.018 D_real: 1.201 D_fake: 0.729 +(epoch: 261, iters: 3920, time: 0.064) G_GAN: 0.180 G_GAN_Feat: 0.728 G_ID: 0.169 G_Rec: 0.350 D_GP: 0.019 D_real: 0.980 D_fake: 0.820 +(epoch: 261, iters: 4320, time: 0.064) G_GAN: 0.173 G_GAN_Feat: 0.624 G_ID: 0.119 G_Rec: 0.301 D_GP: 0.024 D_real: 1.077 D_fake: 0.827 +(epoch: 261, iters: 4720, time: 0.064) G_GAN: 0.507 G_GAN_Feat: 0.848 G_ID: 0.152 G_Rec: 0.396 D_GP: 0.025 D_real: 1.189 D_fake: 0.502 +(epoch: 261, iters: 5120, time: 0.064) G_GAN: 0.267 G_GAN_Feat: 0.665 G_ID: 0.132 G_Rec: 0.288 D_GP: 0.021 D_real: 1.138 D_fake: 0.733 +(epoch: 261, iters: 5520, time: 0.064) G_GAN: 0.432 G_GAN_Feat: 0.948 G_ID: 0.162 G_Rec: 0.415 D_GP: 0.100 D_real: 0.753 D_fake: 0.570 +(epoch: 261, iters: 5920, time: 0.064) G_GAN: 0.004 G_GAN_Feat: 0.837 G_ID: 0.147 G_Rec: 0.330 D_GP: 0.234 D_real: 0.456 D_fake: 0.996 +(epoch: 261, iters: 6320, time: 0.064) G_GAN: 0.348 G_GAN_Feat: 0.913 G_ID: 0.156 G_Rec: 0.425 D_GP: 0.021 D_real: 1.054 D_fake: 0.655 +(epoch: 261, iters: 6720, time: 0.064) G_GAN: 0.144 G_GAN_Feat: 0.709 G_ID: 0.107 G_Rec: 0.275 D_GP: 0.030 D_real: 0.886 D_fake: 0.856 +(epoch: 261, iters: 7120, time: 0.064) G_GAN: 0.542 G_GAN_Feat: 1.020 G_ID: 0.161 G_Rec: 0.442 D_GP: 0.032 D_real: 0.889 D_fake: 0.462 +(epoch: 261, iters: 7520, time: 0.064) G_GAN: 0.317 G_GAN_Feat: 0.657 G_ID: 0.124 G_Rec: 0.272 D_GP: 0.022 D_real: 1.227 D_fake: 0.683 +(epoch: 261, iters: 7920, time: 0.064) G_GAN: 0.208 G_GAN_Feat: 1.138 G_ID: 0.191 G_Rec: 0.450 D_GP: 0.039 D_real: 0.959 D_fake: 0.800 +(epoch: 261, iters: 8320, time: 0.064) G_GAN: 0.200 G_GAN_Feat: 0.613 G_ID: 0.117 G_Rec: 0.261 D_GP: 0.023 D_real: 1.207 D_fake: 0.800 +(epoch: 262, iters: 112, time: 0.064) G_GAN: 0.373 G_GAN_Feat: 0.878 G_ID: 0.169 G_Rec: 0.354 D_GP: 0.022 D_real: 1.063 D_fake: 0.629 +(epoch: 262, iters: 512, time: 0.064) G_GAN: 0.163 G_GAN_Feat: 0.867 G_ID: 0.121 G_Rec: 0.284 D_GP: 0.055 D_real: 0.545 D_fake: 0.837 +(epoch: 262, iters: 912, time: 0.064) G_GAN: 0.329 G_GAN_Feat: 0.897 G_ID: 0.180 G_Rec: 0.423 D_GP: 0.026 D_real: 1.007 D_fake: 0.675 +(epoch: 262, iters: 1312, time: 0.064) G_GAN: 0.077 G_GAN_Feat: 0.677 G_ID: 0.143 G_Rec: 0.308 D_GP: 0.038 D_real: 0.919 D_fake: 0.923 +(epoch: 262, iters: 1712, time: 0.064) G_GAN: 0.370 G_GAN_Feat: 0.886 G_ID: 0.158 G_Rec: 0.383 D_GP: 0.037 D_real: 0.931 D_fake: 0.632 +(epoch: 262, iters: 2112, time: 0.064) G_GAN: 0.304 G_GAN_Feat: 0.737 G_ID: 0.128 G_Rec: 0.291 D_GP: 0.032 D_real: 1.012 D_fake: 0.696 +(epoch: 262, iters: 2512, time: 0.064) G_GAN: 0.599 G_GAN_Feat: 0.918 G_ID: 0.164 G_Rec: 0.388 D_GP: 0.040 D_real: 1.170 D_fake: 0.408 +(epoch: 262, iters: 2912, time: 0.064) G_GAN: 0.124 G_GAN_Feat: 0.789 G_ID: 0.150 G_Rec: 0.369 D_GP: 0.029 D_real: 0.816 D_fake: 0.876 +(epoch: 262, iters: 3312, time: 0.064) G_GAN: 0.541 G_GAN_Feat: 1.124 G_ID: 0.149 G_Rec: 0.413 D_GP: 0.028 D_real: 1.403 D_fake: 0.475 +(epoch: 262, iters: 3712, time: 0.064) G_GAN: 0.036 G_GAN_Feat: 0.677 G_ID: 0.151 G_Rec: 0.312 D_GP: 0.019 D_real: 0.985 D_fake: 0.964 +(epoch: 262, iters: 4112, time: 0.064) G_GAN: 0.176 G_GAN_Feat: 0.769 G_ID: 0.207 G_Rec: 0.331 D_GP: 0.033 D_real: 0.884 D_fake: 0.824 +(epoch: 262, iters: 4512, time: 0.064) G_GAN: 0.236 G_GAN_Feat: 0.668 G_ID: 0.121 G_Rec: 0.262 D_GP: 0.031 D_real: 1.094 D_fake: 0.764 +(epoch: 262, iters: 4912, time: 0.064) G_GAN: 0.613 G_GAN_Feat: 0.970 G_ID: 0.149 G_Rec: 0.382 D_GP: 0.027 D_real: 1.334 D_fake: 0.402 +(epoch: 262, iters: 5312, time: 0.064) G_GAN: 0.195 G_GAN_Feat: 0.749 G_ID: 0.133 G_Rec: 0.274 D_GP: 0.025 D_real: 0.938 D_fake: 0.805 +(epoch: 262, iters: 5712, time: 0.064) G_GAN: 0.271 G_GAN_Feat: 0.915 G_ID: 0.167 G_Rec: 0.372 D_GP: 0.044 D_real: 0.823 D_fake: 0.731 +(epoch: 262, iters: 6112, time: 0.064) G_GAN: 0.304 G_GAN_Feat: 0.603 G_ID: 0.141 G_Rec: 0.280 D_GP: 0.021 D_real: 1.300 D_fake: 0.698 +(epoch: 262, iters: 6512, time: 0.064) G_GAN: 0.208 G_GAN_Feat: 0.894 G_ID: 0.194 G_Rec: 0.396 D_GP: 0.033 D_real: 0.859 D_fake: 0.792 +(epoch: 262, iters: 6912, time: 0.064) G_GAN: 0.211 G_GAN_Feat: 0.833 G_ID: 0.135 G_Rec: 0.324 D_GP: 0.022 D_real: 1.019 D_fake: 0.789 +(epoch: 262, iters: 7312, time: 0.064) G_GAN: 0.307 G_GAN_Feat: 0.839 G_ID: 0.193 G_Rec: 0.390 D_GP: 0.024 D_real: 1.083 D_fake: 0.694 +(epoch: 262, iters: 7712, time: 0.064) G_GAN: 0.248 G_GAN_Feat: 0.823 G_ID: 0.121 G_Rec: 0.296 D_GP: 0.070 D_real: 0.670 D_fake: 0.752 +(epoch: 262, iters: 8112, time: 0.064) G_GAN: 0.596 G_GAN_Feat: 1.346 G_ID: 0.161 G_Rec: 0.447 D_GP: 0.109 D_real: 1.281 D_fake: 0.582 +(epoch: 262, iters: 8512, time: 0.064) G_GAN: -0.011 G_GAN_Feat: 0.657 G_ID: 0.132 G_Rec: 0.287 D_GP: 0.021 D_real: 0.953 D_fake: 1.011 +(epoch: 263, iters: 304, time: 0.064) G_GAN: 0.209 G_GAN_Feat: 0.898 G_ID: 0.151 G_Rec: 0.398 D_GP: 0.032 D_real: 0.829 D_fake: 0.792 +(epoch: 263, iters: 704, time: 0.064) G_GAN: 0.335 G_GAN_Feat: 0.714 G_ID: 0.105 G_Rec: 0.281 D_GP: 0.027 D_real: 1.139 D_fake: 0.665 +(epoch: 263, iters: 1104, time: 0.064) G_GAN: 0.597 G_GAN_Feat: 0.802 G_ID: 0.158 G_Rec: 0.417 D_GP: 0.021 D_real: 1.319 D_fake: 0.413 +(epoch: 263, iters: 1504, time: 0.064) G_GAN: 0.128 G_GAN_Feat: 0.614 G_ID: 0.110 G_Rec: 0.311 D_GP: 0.026 D_real: 1.047 D_fake: 0.872 +(epoch: 263, iters: 1904, time: 0.064) G_GAN: 0.319 G_GAN_Feat: 0.775 G_ID: 0.169 G_Rec: 0.385 D_GP: 0.032 D_real: 1.136 D_fake: 0.682 +(epoch: 263, iters: 2304, time: 0.064) G_GAN: 0.128 G_GAN_Feat: 0.578 G_ID: 0.123 G_Rec: 0.250 D_GP: 0.022 D_real: 1.079 D_fake: 0.872 +(epoch: 263, iters: 2704, time: 0.064) G_GAN: 0.490 G_GAN_Feat: 0.770 G_ID: 0.166 G_Rec: 0.384 D_GP: 0.025 D_real: 1.264 D_fake: 0.530 +(epoch: 263, iters: 3104, time: 0.064) G_GAN: 0.277 G_GAN_Feat: 0.575 G_ID: 0.123 G_Rec: 0.261 D_GP: 0.019 D_real: 1.229 D_fake: 0.723 +(epoch: 263, iters: 3504, time: 0.064) G_GAN: 0.240 G_GAN_Feat: 0.912 G_ID: 0.173 G_Rec: 0.417 D_GP: 0.049 D_real: 0.951 D_fake: 0.762 +(epoch: 263, iters: 3904, time: 0.064) G_GAN: 0.232 G_GAN_Feat: 0.614 G_ID: 0.124 G_Rec: 0.278 D_GP: 0.019 D_real: 1.216 D_fake: 0.769 +(epoch: 263, iters: 4304, time: 0.064) G_GAN: 0.137 G_GAN_Feat: 0.984 G_ID: 0.159 G_Rec: 0.426 D_GP: 0.142 D_real: 0.492 D_fake: 0.865 +(epoch: 263, iters: 4704, time: 0.064) G_GAN: 0.232 G_GAN_Feat: 0.695 G_ID: 0.130 G_Rec: 0.306 D_GP: 0.033 D_real: 1.065 D_fake: 0.768 +(epoch: 263, iters: 5104, time: 0.064) G_GAN: 0.349 G_GAN_Feat: 0.795 G_ID: 0.172 G_Rec: 0.358 D_GP: 0.022 D_real: 1.135 D_fake: 0.652 +(epoch: 263, iters: 5504, time: 0.064) G_GAN: 0.168 G_GAN_Feat: 0.676 G_ID: 0.120 G_Rec: 0.292 D_GP: 0.073 D_real: 0.927 D_fake: 0.832 +(epoch: 263, iters: 5904, time: 0.064) G_GAN: 0.340 G_GAN_Feat: 0.793 G_ID: 0.156 G_Rec: 0.330 D_GP: 0.028 D_real: 1.117 D_fake: 0.661 +(epoch: 263, iters: 6304, time: 0.064) G_GAN: 0.016 G_GAN_Feat: 0.752 G_ID: 0.137 G_Rec: 0.300 D_GP: 0.052 D_real: 0.667 D_fake: 0.984 +(epoch: 263, iters: 6704, time: 0.064) G_GAN: 0.249 G_GAN_Feat: 0.889 G_ID: 0.174 G_Rec: 0.372 D_GP: 0.041 D_real: 0.895 D_fake: 0.752 +(epoch: 263, iters: 7104, time: 0.064) G_GAN: -0.069 G_GAN_Feat: 0.759 G_ID: 0.152 G_Rec: 0.308 D_GP: 0.042 D_real: 0.666 D_fake: 1.070 +(epoch: 263, iters: 7504, time: 0.064) G_GAN: 0.707 G_GAN_Feat: 0.918 G_ID: 0.157 G_Rec: 0.427 D_GP: 0.022 D_real: 1.333 D_fake: 0.333 +(epoch: 263, iters: 7904, time: 0.064) G_GAN: 0.181 G_GAN_Feat: 0.651 G_ID: 0.114 G_Rec: 0.276 D_GP: 0.029 D_real: 1.184 D_fake: 0.819 +(epoch: 263, iters: 8304, time: 0.064) G_GAN: 0.629 G_GAN_Feat: 0.811 G_ID: 0.158 G_Rec: 0.367 D_GP: 0.024 D_real: 1.411 D_fake: 0.385 +(epoch: 264, iters: 96, time: 0.064) G_GAN: 0.139 G_GAN_Feat: 0.649 G_ID: 0.141 G_Rec: 0.272 D_GP: 0.023 D_real: 1.111 D_fake: 0.862 +(epoch: 264, iters: 496, time: 0.064) G_GAN: 0.380 G_GAN_Feat: 0.754 G_ID: 0.164 G_Rec: 0.332 D_GP: 0.026 D_real: 1.235 D_fake: 0.622 +(epoch: 264, iters: 896, time: 0.064) G_GAN: -0.013 G_GAN_Feat: 0.707 G_ID: 0.114 G_Rec: 0.308 D_GP: 0.051 D_real: 0.826 D_fake: 1.013 +(epoch: 264, iters: 1296, time: 0.064) G_GAN: 0.351 G_GAN_Feat: 0.832 G_ID: 0.152 G_Rec: 0.436 D_GP: 0.035 D_real: 1.044 D_fake: 0.650 +(epoch: 264, iters: 1696, time: 0.064) G_GAN: 0.065 G_GAN_Feat: 0.720 G_ID: 0.120 G_Rec: 0.315 D_GP: 0.034 D_real: 0.828 D_fake: 0.935 +(epoch: 264, iters: 2096, time: 0.064) G_GAN: 0.354 G_GAN_Feat: 0.878 G_ID: 0.173 G_Rec: 0.412 D_GP: 0.045 D_real: 0.993 D_fake: 0.649 +(epoch: 264, iters: 2496, time: 0.064) G_GAN: 0.328 G_GAN_Feat: 0.726 G_ID: 0.114 G_Rec: 0.281 D_GP: 0.053 D_real: 1.063 D_fake: 0.674 +(epoch: 264, iters: 2896, time: 0.064) G_GAN: 0.072 G_GAN_Feat: 0.955 G_ID: 0.173 G_Rec: 0.388 D_GP: 0.034 D_real: 0.984 D_fake: 0.936 +(epoch: 264, iters: 3296, time: 0.064) G_GAN: 0.485 G_GAN_Feat: 0.869 G_ID: 0.132 G_Rec: 0.362 D_GP: 0.061 D_real: 0.972 D_fake: 0.575 +(epoch: 264, iters: 3696, time: 0.064) G_GAN: 0.436 G_GAN_Feat: 0.827 G_ID: 0.176 G_Rec: 0.379 D_GP: 0.023 D_real: 1.225 D_fake: 0.567 +(epoch: 264, iters: 4096, time: 0.064) G_GAN: 0.294 G_GAN_Feat: 0.643 G_ID: 0.115 G_Rec: 0.273 D_GP: 0.024 D_real: 1.249 D_fake: 0.706 +(epoch: 264, iters: 4496, time: 0.064) G_GAN: 0.491 G_GAN_Feat: 1.000 G_ID: 0.154 G_Rec: 0.404 D_GP: 0.070 D_real: 0.906 D_fake: 0.515 +(epoch: 264, iters: 4896, time: 0.064) G_GAN: 0.107 G_GAN_Feat: 0.784 G_ID: 0.133 G_Rec: 0.318 D_GP: 0.048 D_real: 0.678 D_fake: 0.893 +(epoch: 264, iters: 5296, time: 0.064) G_GAN: 0.398 G_GAN_Feat: 0.831 G_ID: 0.159 G_Rec: 0.408 D_GP: 0.020 D_real: 1.141 D_fake: 0.603 +(epoch: 264, iters: 5696, time: 0.064) G_GAN: 0.117 G_GAN_Feat: 0.660 G_ID: 0.128 G_Rec: 0.285 D_GP: 0.030 D_real: 0.982 D_fake: 0.883 +(epoch: 264, iters: 6096, time: 0.064) G_GAN: 0.465 G_GAN_Feat: 0.900 G_ID: 0.191 G_Rec: 0.390 D_GP: 0.036 D_real: 1.168 D_fake: 0.540 +(epoch: 264, iters: 6496, time: 0.064) G_GAN: 0.223 G_GAN_Feat: 0.697 G_ID: 0.125 G_Rec: 0.267 D_GP: 0.033 D_real: 0.986 D_fake: 0.777 +(epoch: 264, iters: 6896, time: 0.064) G_GAN: 0.402 G_GAN_Feat: 0.994 G_ID: 0.169 G_Rec: 0.410 D_GP: 0.047 D_real: 0.749 D_fake: 0.599 +(epoch: 264, iters: 7296, time: 0.064) G_GAN: 0.341 G_GAN_Feat: 0.931 G_ID: 0.137 G_Rec: 0.331 D_GP: 0.496 D_real: 0.554 D_fake: 0.668 +(epoch: 264, iters: 7696, time: 0.064) G_GAN: 0.566 G_GAN_Feat: 0.933 G_ID: 0.168 G_Rec: 0.402 D_GP: 0.023 D_real: 1.188 D_fake: 0.440 +(epoch: 264, iters: 8096, time: 0.064) G_GAN: 0.390 G_GAN_Feat: 0.671 G_ID: 0.122 G_Rec: 0.309 D_GP: 0.022 D_real: 1.377 D_fake: 0.614 +(epoch: 264, iters: 8496, time: 0.064) G_GAN: 0.515 G_GAN_Feat: 1.106 G_ID: 0.151 G_Rec: 0.432 D_GP: 0.130 D_real: 0.797 D_fake: 0.492 +(epoch: 265, iters: 288, time: 0.064) G_GAN: 0.244 G_GAN_Feat: 0.716 G_ID: 0.134 G_Rec: 0.318 D_GP: 0.024 D_real: 1.207 D_fake: 0.758 +(epoch: 265, iters: 688, time: 0.064) G_GAN: 0.460 G_GAN_Feat: 0.835 G_ID: 0.156 G_Rec: 0.356 D_GP: 0.055 D_real: 1.123 D_fake: 0.543 +(epoch: 265, iters: 1088, time: 0.064) G_GAN: 0.232 G_GAN_Feat: 0.869 G_ID: 0.113 G_Rec: 0.293 D_GP: 0.147 D_real: 0.409 D_fake: 0.772 +(epoch: 265, iters: 1488, time: 0.064) G_GAN: 0.484 G_GAN_Feat: 0.951 G_ID: 0.157 G_Rec: 0.388 D_GP: 0.029 D_real: 0.985 D_fake: 0.519 +(epoch: 265, iters: 1888, time: 0.064) G_GAN: 0.362 G_GAN_Feat: 0.703 G_ID: 0.108 G_Rec: 0.296 D_GP: 0.026 D_real: 1.352 D_fake: 0.641 +(epoch: 265, iters: 2288, time: 0.064) G_GAN: 0.418 G_GAN_Feat: 1.019 G_ID: 0.183 G_Rec: 0.437 D_GP: 0.082 D_real: 0.808 D_fake: 0.584 +(epoch: 265, iters: 2688, time: 0.064) G_GAN: 0.275 G_GAN_Feat: 0.937 G_ID: 0.132 G_Rec: 0.322 D_GP: 0.063 D_real: 0.468 D_fake: 0.727 +(epoch: 265, iters: 3088, time: 0.064) G_GAN: 0.508 G_GAN_Feat: 0.796 G_ID: 0.165 G_Rec: 0.384 D_GP: 0.019 D_real: 1.231 D_fake: 0.527 +(epoch: 265, iters: 3488, time: 0.064) G_GAN: -0.013 G_GAN_Feat: 0.647 G_ID: 0.122 G_Rec: 0.294 D_GP: 0.018 D_real: 0.928 D_fake: 1.013 +(epoch: 265, iters: 3888, time: 0.064) G_GAN: 0.468 G_GAN_Feat: 0.965 G_ID: 0.165 G_Rec: 0.406 D_GP: 0.039 D_real: 0.924 D_fake: 0.535 +(epoch: 265, iters: 4288, time: 0.064) G_GAN: 0.393 G_GAN_Feat: 0.729 G_ID: 0.111 G_Rec: 0.300 D_GP: 0.028 D_real: 1.190 D_fake: 0.608 +(epoch: 265, iters: 4688, time: 0.064) G_GAN: 0.515 G_GAN_Feat: 0.827 G_ID: 0.158 G_Rec: 0.349 D_GP: 0.022 D_real: 1.249 D_fake: 0.487 +(epoch: 265, iters: 5088, time: 0.064) G_GAN: 0.286 G_GAN_Feat: 0.624 G_ID: 0.140 G_Rec: 0.332 D_GP: 0.019 D_real: 1.245 D_fake: 0.723 +(epoch: 265, iters: 5488, time: 0.064) G_GAN: 0.439 G_GAN_Feat: 0.666 G_ID: 0.165 G_Rec: 0.350 D_GP: 0.020 D_real: 1.261 D_fake: 0.562 +(epoch: 265, iters: 5888, time: 0.064) G_GAN: 0.179 G_GAN_Feat: 0.576 G_ID: 0.108 G_Rec: 0.282 D_GP: 0.023 D_real: 1.131 D_fake: 0.821 +(epoch: 265, iters: 6288, time: 0.064) G_GAN: 0.424 G_GAN_Feat: 0.814 G_ID: 0.159 G_Rec: 0.435 D_GP: 0.034 D_real: 1.143 D_fake: 0.576 +(epoch: 265, iters: 6688, time: 0.064) G_GAN: 0.073 G_GAN_Feat: 0.716 G_ID: 0.119 G_Rec: 0.309 D_GP: 0.052 D_real: 0.839 D_fake: 0.927 +(epoch: 265, iters: 7088, time: 0.064) G_GAN: 0.501 G_GAN_Feat: 0.795 G_ID: 0.150 G_Rec: 0.387 D_GP: 0.022 D_real: 1.231 D_fake: 0.511 +(epoch: 265, iters: 7488, time: 0.064) G_GAN: 0.264 G_GAN_Feat: 0.669 G_ID: 0.144 G_Rec: 0.284 D_GP: 0.020 D_real: 1.141 D_fake: 0.737 +(epoch: 265, iters: 7888, time: 0.064) G_GAN: 0.490 G_GAN_Feat: 0.924 G_ID: 0.160 G_Rec: 0.387 D_GP: 0.046 D_real: 0.985 D_fake: 0.518 +(epoch: 265, iters: 8288, time: 0.064) G_GAN: 0.312 G_GAN_Feat: 0.713 G_ID: 0.107 G_Rec: 0.330 D_GP: 0.035 D_real: 1.045 D_fake: 0.689 +(epoch: 266, iters: 80, time: 0.064) G_GAN: 0.250 G_GAN_Feat: 0.962 G_ID: 0.173 G_Rec: 0.421 D_GP: 0.033 D_real: 0.747 D_fake: 0.752 +(epoch: 266, iters: 480, time: 0.064) G_GAN: 0.100 G_GAN_Feat: 0.847 G_ID: 0.151 G_Rec: 0.327 D_GP: 0.045 D_real: 0.475 D_fake: 0.900 +(epoch: 266, iters: 880, time: 0.064) G_GAN: 0.253 G_GAN_Feat: 0.823 G_ID: 0.163 G_Rec: 0.376 D_GP: 0.024 D_real: 1.130 D_fake: 0.754 +(epoch: 266, iters: 1280, time: 0.064) G_GAN: 0.081 G_GAN_Feat: 0.696 G_ID: 0.129 G_Rec: 0.298 D_GP: 0.031 D_real: 0.932 D_fake: 0.919 +(epoch: 266, iters: 1680, time: 0.064) G_GAN: 0.943 G_GAN_Feat: 1.122 G_ID: 0.147 G_Rec: 0.440 D_GP: 0.070 D_real: 1.356 D_fake: 0.339 +(epoch: 266, iters: 2080, time: 0.064) G_GAN: -0.157 G_GAN_Feat: 0.752 G_ID: 0.151 G_Rec: 0.313 D_GP: 0.052 D_real: 0.572 D_fake: 1.157 +(epoch: 266, iters: 2480, time: 0.064) G_GAN: 0.263 G_GAN_Feat: 0.851 G_ID: 0.171 G_Rec: 0.384 D_GP: 0.032 D_real: 0.933 D_fake: 0.737 +(epoch: 266, iters: 2880, time: 0.064) G_GAN: 0.221 G_GAN_Feat: 0.736 G_ID: 0.134 G_Rec: 0.287 D_GP: 0.051 D_real: 1.097 D_fake: 0.780 +(epoch: 266, iters: 3280, time: 0.064) G_GAN: 0.293 G_GAN_Feat: 0.771 G_ID: 0.182 G_Rec: 0.370 D_GP: 0.021 D_real: 1.093 D_fake: 0.709 +(epoch: 266, iters: 3680, time: 0.064) G_GAN: -0.063 G_GAN_Feat: 0.687 G_ID: 0.148 G_Rec: 0.287 D_GP: 0.050 D_real: 0.798 D_fake: 1.063 +(epoch: 266, iters: 4080, time: 0.064) G_GAN: 0.392 G_GAN_Feat: 0.927 G_ID: 0.163 G_Rec: 0.418 D_GP: 0.034 D_real: 0.993 D_fake: 0.622 +(epoch: 266, iters: 4480, time: 0.064) G_GAN: 0.313 G_GAN_Feat: 1.043 G_ID: 0.116 G_Rec: 0.330 D_GP: 1.340 D_real: 0.541 D_fake: 0.691 +(epoch: 266, iters: 4880, time: 0.064) G_GAN: 0.409 G_GAN_Feat: 0.871 G_ID: 0.179 G_Rec: 0.382 D_GP: 0.023 D_real: 1.067 D_fake: 0.599 +(epoch: 266, iters: 5280, time: 0.064) G_GAN: 0.265 G_GAN_Feat: 0.876 G_ID: 0.147 G_Rec: 0.341 D_GP: 0.057 D_real: 0.762 D_fake: 0.738 +(epoch: 266, iters: 5680, time: 0.064) G_GAN: 0.559 G_GAN_Feat: 0.913 G_ID: 0.165 G_Rec: 0.389 D_GP: 0.027 D_real: 1.172 D_fake: 0.445 +(epoch: 266, iters: 6080, time: 0.064) G_GAN: 0.386 G_GAN_Feat: 0.658 G_ID: 0.103 G_Rec: 0.265 D_GP: 0.022 D_real: 1.311 D_fake: 0.614 +(epoch: 266, iters: 6480, time: 0.064) G_GAN: 0.392 G_GAN_Feat: 0.910 G_ID: 0.169 G_Rec: 0.411 D_GP: 0.028 D_real: 1.036 D_fake: 0.609 +(epoch: 266, iters: 6880, time: 0.064) G_GAN: 0.514 G_GAN_Feat: 0.779 G_ID: 0.119 G_Rec: 0.303 D_GP: 0.052 D_real: 1.251 D_fake: 0.491 +(epoch: 266, iters: 7280, time: 0.064) G_GAN: 0.706 G_GAN_Feat: 0.852 G_ID: 0.161 G_Rec: 0.347 D_GP: 0.023 D_real: 1.561 D_fake: 0.317 +(epoch: 266, iters: 7680, time: 0.064) G_GAN: 0.410 G_GAN_Feat: 0.681 G_ID: 0.126 G_Rec: 0.310 D_GP: 0.027 D_real: 1.349 D_fake: 0.600 +(epoch: 266, iters: 8080, time: 0.064) G_GAN: 0.609 G_GAN_Feat: 1.046 G_ID: 0.167 G_Rec: 0.400 D_GP: 0.080 D_real: 0.564 D_fake: 0.404 +(epoch: 266, iters: 8480, time: 0.064) G_GAN: 0.339 G_GAN_Feat: 0.699 G_ID: 0.131 G_Rec: 0.292 D_GP: 0.026 D_real: 1.143 D_fake: 0.662 +(epoch: 267, iters: 272, time: 0.064) G_GAN: 0.501 G_GAN_Feat: 0.993 G_ID: 0.160 G_Rec: 0.413 D_GP: 0.029 D_real: 1.023 D_fake: 0.594 +(epoch: 267, iters: 672, time: 0.064) G_GAN: 0.251 G_GAN_Feat: 0.835 G_ID: 0.139 G_Rec: 0.296 D_GP: 0.112 D_real: 0.743 D_fake: 0.750 +(epoch: 267, iters: 1072, time: 0.064) G_GAN: 0.339 G_GAN_Feat: 0.797 G_ID: 0.199 G_Rec: 0.398 D_GP: 0.019 D_real: 1.091 D_fake: 0.662 +(epoch: 267, iters: 1472, time: 0.064) G_GAN: 0.291 G_GAN_Feat: 0.583 G_ID: 0.120 G_Rec: 0.316 D_GP: 0.020 D_real: 1.256 D_fake: 0.709 +(epoch: 267, iters: 1872, time: 0.064) G_GAN: 0.331 G_GAN_Feat: 0.951 G_ID: 0.208 G_Rec: 0.452 D_GP: 0.055 D_real: 0.878 D_fake: 0.671 +(epoch: 267, iters: 2272, time: 0.064) G_GAN: 0.155 G_GAN_Feat: 0.687 G_ID: 0.119 G_Rec: 0.294 D_GP: 0.033 D_real: 1.013 D_fake: 0.845 +(epoch: 267, iters: 2672, time: 0.064) G_GAN: 0.382 G_GAN_Feat: 0.797 G_ID: 0.193 G_Rec: 0.413 D_GP: 0.020 D_real: 1.101 D_fake: 0.622 +(epoch: 267, iters: 3072, time: 0.064) G_GAN: 0.182 G_GAN_Feat: 0.621 G_ID: 0.125 G_Rec: 0.298 D_GP: 0.028 D_real: 1.112 D_fake: 0.818 +(epoch: 267, iters: 3472, time: 0.064) G_GAN: 0.402 G_GAN_Feat: 0.845 G_ID: 0.156 G_Rec: 0.405 D_GP: 0.030 D_real: 1.119 D_fake: 0.600 +(epoch: 267, iters: 3872, time: 0.064) G_GAN: 0.151 G_GAN_Feat: 0.612 G_ID: 0.124 G_Rec: 0.286 D_GP: 0.024 D_real: 1.080 D_fake: 0.849 +(epoch: 267, iters: 4272, time: 0.064) G_GAN: 0.228 G_GAN_Feat: 0.865 G_ID: 0.186 G_Rec: 0.388 D_GP: 0.139 D_real: 0.852 D_fake: 0.772 +(epoch: 267, iters: 4672, time: 0.064) G_GAN: 0.072 G_GAN_Feat: 0.713 G_ID: 0.132 G_Rec: 0.293 D_GP: 0.041 D_real: 0.812 D_fake: 0.928 +(epoch: 267, iters: 5072, time: 0.064) G_GAN: 0.348 G_GAN_Feat: 0.756 G_ID: 0.173 G_Rec: 0.335 D_GP: 0.020 D_real: 1.117 D_fake: 0.653 +(epoch: 267, iters: 5472, time: 0.064) G_GAN: 0.356 G_GAN_Feat: 0.814 G_ID: 0.132 G_Rec: 0.314 D_GP: 0.093 D_real: 1.083 D_fake: 0.662 +(epoch: 267, iters: 5872, time: 0.064) G_GAN: 0.316 G_GAN_Feat: 1.252 G_ID: 0.149 G_Rec: 0.427 D_GP: 0.091 D_real: 1.158 D_fake: 0.697 +(epoch: 267, iters: 6272, time: 0.064) G_GAN: 0.043 G_GAN_Feat: 0.691 G_ID: 0.133 G_Rec: 0.296 D_GP: 0.031 D_real: 0.830 D_fake: 0.957 +(epoch: 267, iters: 6672, time: 0.064) G_GAN: 0.285 G_GAN_Feat: 0.951 G_ID: 0.153 G_Rec: 0.433 D_GP: 0.023 D_real: 1.093 D_fake: 0.716 +(epoch: 267, iters: 7072, time: 0.064) G_GAN: 0.115 G_GAN_Feat: 0.732 G_ID: 0.124 G_Rec: 0.296 D_GP: 0.040 D_real: 0.987 D_fake: 0.885 +(epoch: 267, iters: 7472, time: 0.064) G_GAN: 0.572 G_GAN_Feat: 1.029 G_ID: 0.160 G_Rec: 0.432 D_GP: 0.039 D_real: 0.963 D_fake: 0.451 +(epoch: 267, iters: 7872, time: 0.064) G_GAN: 0.319 G_GAN_Feat: 0.747 G_ID: 0.169 G_Rec: 0.296 D_GP: 0.024 D_real: 1.145 D_fake: 0.682 +(epoch: 267, iters: 8272, time: 0.064) G_GAN: 0.414 G_GAN_Feat: 0.818 G_ID: 0.141 G_Rec: 0.403 D_GP: 0.018 D_real: 1.227 D_fake: 0.599 +(epoch: 268, iters: 64, time: 0.064) G_GAN: 0.023 G_GAN_Feat: 0.704 G_ID: 0.140 G_Rec: 0.306 D_GP: 0.027 D_real: 0.863 D_fake: 0.977 +(epoch: 268, iters: 464, time: 0.064) G_GAN: 0.489 G_GAN_Feat: 0.907 G_ID: 0.137 G_Rec: 0.403 D_GP: 0.033 D_real: 0.998 D_fake: 0.518 +(epoch: 268, iters: 864, time: 0.064) G_GAN: 0.472 G_GAN_Feat: 0.725 G_ID: 0.108 G_Rec: 0.295 D_GP: 0.104 D_real: 1.142 D_fake: 0.540 +(epoch: 268, iters: 1264, time: 0.064) G_GAN: 0.514 G_GAN_Feat: 0.990 G_ID: 0.141 G_Rec: 0.427 D_GP: 0.036 D_real: 1.046 D_fake: 0.518 +(epoch: 268, iters: 1664, time: 0.064) G_GAN: 0.501 G_GAN_Feat: 0.725 G_ID: 0.141 G_Rec: 0.306 D_GP: 0.021 D_real: 1.427 D_fake: 0.506 +(epoch: 268, iters: 2064, time: 0.064) G_GAN: -0.010 G_GAN_Feat: 0.968 G_ID: 0.187 G_Rec: 0.393 D_GP: 0.051 D_real: 0.470 D_fake: 1.010 +(epoch: 268, iters: 2464, time: 0.064) G_GAN: 0.366 G_GAN_Feat: 0.728 G_ID: 0.143 G_Rec: 0.329 D_GP: 0.023 D_real: 1.242 D_fake: 0.638 +(epoch: 268, iters: 2864, time: 0.064) G_GAN: 0.362 G_GAN_Feat: 0.891 G_ID: 0.183 G_Rec: 0.363 D_GP: 0.020 D_real: 1.082 D_fake: 0.642 +(epoch: 268, iters: 3264, time: 0.064) G_GAN: 0.283 G_GAN_Feat: 0.854 G_ID: 0.096 G_Rec: 0.330 D_GP: 0.056 D_real: 0.877 D_fake: 0.730 +(epoch: 268, iters: 3664, time: 0.064) G_GAN: 0.518 G_GAN_Feat: 0.866 G_ID: 0.170 G_Rec: 0.366 D_GP: 0.047 D_real: 1.245 D_fake: 0.486 +(epoch: 268, iters: 4064, time: 0.064) G_GAN: 0.147 G_GAN_Feat: 0.571 G_ID: 0.115 G_Rec: 0.293 D_GP: 0.022 D_real: 1.112 D_fake: 0.853 +(epoch: 268, iters: 4464, time: 0.064) G_GAN: 0.282 G_GAN_Feat: 0.858 G_ID: 0.150 G_Rec: 0.416 D_GP: 0.034 D_real: 0.963 D_fake: 0.720 +(epoch: 268, iters: 4864, time: 0.064) G_GAN: 0.256 G_GAN_Feat: 0.628 G_ID: 0.118 G_Rec: 0.278 D_GP: 0.024 D_real: 1.194 D_fake: 0.744 +(epoch: 268, iters: 5264, time: 0.064) G_GAN: 0.337 G_GAN_Feat: 0.962 G_ID: 0.179 G_Rec: 0.433 D_GP: 0.157 D_real: 0.609 D_fake: 0.667 +(epoch: 268, iters: 5664, time: 0.064) G_GAN: 0.032 G_GAN_Feat: 0.689 G_ID: 0.130 G_Rec: 0.290 D_GP: 0.029 D_real: 0.908 D_fake: 0.968 +(epoch: 268, iters: 6064, time: 0.064) G_GAN: 0.105 G_GAN_Feat: 0.981 G_ID: 0.179 G_Rec: 0.410 D_GP: 0.033 D_real: 0.511 D_fake: 0.895 +(epoch: 268, iters: 6464, time: 0.064) G_GAN: 0.091 G_GAN_Feat: 0.711 G_ID: 0.121 G_Rec: 0.281 D_GP: 0.033 D_real: 1.063 D_fake: 0.909 +(epoch: 268, iters: 6864, time: 0.064) G_GAN: 0.666 G_GAN_Feat: 0.945 G_ID: 0.186 G_Rec: 0.407 D_GP: 0.032 D_real: 1.256 D_fake: 0.359 +(epoch: 268, iters: 7264, time: 0.064) G_GAN: 0.265 G_GAN_Feat: 0.596 G_ID: 0.122 G_Rec: 0.299 D_GP: 0.018 D_real: 1.254 D_fake: 0.735 +(epoch: 268, iters: 7664, time: 0.064) G_GAN: 0.303 G_GAN_Feat: 0.714 G_ID: 0.153 G_Rec: 0.324 D_GP: 0.022 D_real: 1.134 D_fake: 0.697 +(epoch: 268, iters: 8064, time: 0.064) G_GAN: 0.007 G_GAN_Feat: 0.641 G_ID: 0.131 G_Rec: 0.290 D_GP: 0.022 D_real: 0.921 D_fake: 0.993 +(epoch: 268, iters: 8464, time: 0.064) G_GAN: 0.363 G_GAN_Feat: 0.808 G_ID: 0.173 G_Rec: 0.388 D_GP: 0.023 D_real: 1.105 D_fake: 0.637 +(epoch: 269, iters: 256, time: 0.064) G_GAN: -0.070 G_GAN_Feat: 0.762 G_ID: 0.188 G_Rec: 0.281 D_GP: 0.200 D_real: 0.384 D_fake: 1.070 +(epoch: 269, iters: 656, time: 0.064) G_GAN: 0.312 G_GAN_Feat: 0.937 G_ID: 0.167 G_Rec: 0.380 D_GP: 0.025 D_real: 0.939 D_fake: 0.688 +(epoch: 269, iters: 1056, time: 0.064) G_GAN: 0.255 G_GAN_Feat: 0.815 G_ID: 0.127 G_Rec: 0.334 D_GP: 0.028 D_real: 0.901 D_fake: 0.746 +(epoch: 269, iters: 1456, time: 0.064) G_GAN: 0.139 G_GAN_Feat: 0.740 G_ID: 0.200 G_Rec: 0.391 D_GP: 0.021 D_real: 0.993 D_fake: 0.861 +(epoch: 269, iters: 1856, time: 0.064) G_GAN: 0.131 G_GAN_Feat: 0.696 G_ID: 0.101 G_Rec: 0.302 D_GP: 0.027 D_real: 1.006 D_fake: 0.869 +(epoch: 269, iters: 2256, time: 0.064) G_GAN: 0.369 G_GAN_Feat: 0.838 G_ID: 0.187 G_Rec: 0.393 D_GP: 0.031 D_real: 1.056 D_fake: 0.631 +(epoch: 269, iters: 2656, time: 0.064) G_GAN: 0.336 G_GAN_Feat: 0.743 G_ID: 0.111 G_Rec: 0.295 D_GP: 0.034 D_real: 0.976 D_fake: 0.665 +(epoch: 269, iters: 3056, time: 0.064) G_GAN: 0.359 G_GAN_Feat: 0.851 G_ID: 0.166 G_Rec: 0.422 D_GP: 0.021 D_real: 1.096 D_fake: 0.643 +(epoch: 269, iters: 3456, time: 0.064) G_GAN: 0.257 G_GAN_Feat: 0.696 G_ID: 0.126 G_Rec: 0.274 D_GP: 0.035 D_real: 1.017 D_fake: 0.744 +(epoch: 269, iters: 3856, time: 0.064) G_GAN: 0.397 G_GAN_Feat: 0.788 G_ID: 0.153 G_Rec: 0.355 D_GP: 0.021 D_real: 1.205 D_fake: 0.616 +(epoch: 269, iters: 4256, time: 0.064) G_GAN: 0.231 G_GAN_Feat: 0.692 G_ID: 0.109 G_Rec: 0.284 D_GP: 0.020 D_real: 1.145 D_fake: 0.770 +(epoch: 269, iters: 4656, time: 0.064) G_GAN: 0.579 G_GAN_Feat: 1.175 G_ID: 0.164 G_Rec: 0.436 D_GP: 1.962 D_real: 0.407 D_fake: 0.490 +(epoch: 269, iters: 5056, time: 0.064) G_GAN: 0.250 G_GAN_Feat: 0.603 G_ID: 0.123 G_Rec: 0.272 D_GP: 0.018 D_real: 1.202 D_fake: 0.750 +(epoch: 269, iters: 5456, time: 0.064) G_GAN: 0.472 G_GAN_Feat: 0.841 G_ID: 0.150 G_Rec: 0.395 D_GP: 0.023 D_real: 1.251 D_fake: 0.531 +(epoch: 269, iters: 5856, time: 0.064) G_GAN: 0.040 G_GAN_Feat: 0.804 G_ID: 0.111 G_Rec: 0.315 D_GP: 0.049 D_real: 1.047 D_fake: 0.961 +(epoch: 269, iters: 6256, time: 0.064) G_GAN: 0.394 G_GAN_Feat: 0.834 G_ID: 0.162 G_Rec: 0.420 D_GP: 0.043 D_real: 1.018 D_fake: 0.609 +(epoch: 269, iters: 6656, time: 0.064) G_GAN: 0.149 G_GAN_Feat: 0.624 G_ID: 0.140 G_Rec: 0.260 D_GP: 0.027 D_real: 1.013 D_fake: 0.851 +(epoch: 269, iters: 7056, time: 0.064) G_GAN: 0.017 G_GAN_Feat: 0.876 G_ID: 0.163 G_Rec: 0.424 D_GP: 0.025 D_real: 0.757 D_fake: 0.983 +(epoch: 269, iters: 7456, time: 0.064) G_GAN: 0.176 G_GAN_Feat: 0.740 G_ID: 0.140 G_Rec: 0.309 D_GP: 0.044 D_real: 0.883 D_fake: 0.825 +(epoch: 269, iters: 7856, time: 0.064) G_GAN: 0.358 G_GAN_Feat: 0.936 G_ID: 0.205 G_Rec: 0.423 D_GP: 0.031 D_real: 0.999 D_fake: 0.646 +(epoch: 269, iters: 8256, time: 0.064) G_GAN: 0.262 G_GAN_Feat: 0.674 G_ID: 0.124 G_Rec: 0.280 D_GP: 0.035 D_real: 1.054 D_fake: 0.738 +(epoch: 270, iters: 48, time: 0.064) G_GAN: 0.362 G_GAN_Feat: 0.825 G_ID: 0.168 G_Rec: 0.400 D_GP: 0.022 D_real: 1.070 D_fake: 0.640 +(epoch: 270, iters: 448, time: 0.064) G_GAN: 0.155 G_GAN_Feat: 0.712 G_ID: 0.109 G_Rec: 0.264 D_GP: 0.033 D_real: 0.888 D_fake: 0.845 +(epoch: 270, iters: 848, time: 0.064) G_GAN: 0.362 G_GAN_Feat: 0.867 G_ID: 0.163 G_Rec: 0.352 D_GP: 0.040 D_real: 1.048 D_fake: 0.640 +(epoch: 270, iters: 1248, time: 0.064) G_GAN: 0.191 G_GAN_Feat: 0.873 G_ID: 0.131 G_Rec: 0.316 D_GP: 0.065 D_real: 0.910 D_fake: 0.819 +(epoch: 270, iters: 1648, time: 0.064) G_GAN: 0.226 G_GAN_Feat: 0.971 G_ID: 0.161 G_Rec: 0.402 D_GP: 0.025 D_real: 0.989 D_fake: 0.774 +(epoch: 270, iters: 2048, time: 0.064) G_GAN: -0.054 G_GAN_Feat: 0.688 G_ID: 0.162 G_Rec: 0.308 D_GP: 0.026 D_real: 0.812 D_fake: 1.054 +(epoch: 270, iters: 2448, time: 0.064) G_GAN: 0.272 G_GAN_Feat: 0.887 G_ID: 0.175 G_Rec: 0.409 D_GP: 0.032 D_real: 0.965 D_fake: 0.732 +(epoch: 270, iters: 2848, time: 0.064) G_GAN: 0.145 G_GAN_Feat: 0.774 G_ID: 0.122 G_Rec: 0.329 D_GP: 0.029 D_real: 0.793 D_fake: 0.855 +(epoch: 270, iters: 3248, time: 0.064) G_GAN: 0.574 G_GAN_Feat: 0.883 G_ID: 0.170 G_Rec: 0.383 D_GP: 0.081 D_real: 1.177 D_fake: 0.440 +(epoch: 270, iters: 3648, time: 0.064) G_GAN: 0.119 G_GAN_Feat: 0.739 G_ID: 0.137 G_Rec: 0.349 D_GP: 0.044 D_real: 0.785 D_fake: 0.882 +(epoch: 270, iters: 4048, time: 0.064) G_GAN: 0.482 G_GAN_Feat: 1.002 G_ID: 0.150 G_Rec: 0.389 D_GP: 0.052 D_real: 0.659 D_fake: 0.522 +(epoch: 270, iters: 4448, time: 0.064) G_GAN: 0.279 G_GAN_Feat: 0.683 G_ID: 0.136 G_Rec: 0.291 D_GP: 0.021 D_real: 1.157 D_fake: 0.721 +(epoch: 270, iters: 4848, time: 0.064) G_GAN: 0.445 G_GAN_Feat: 0.906 G_ID: 0.174 G_Rec: 0.381 D_GP: 0.028 D_real: 1.129 D_fake: 0.566 +(epoch: 270, iters: 5248, time: 0.064) G_GAN: 0.131 G_GAN_Feat: 0.726 G_ID: 0.118 G_Rec: 0.313 D_GP: 0.050 D_real: 0.946 D_fake: 0.871 +(epoch: 270, iters: 5648, time: 0.064) G_GAN: 0.555 G_GAN_Feat: 1.019 G_ID: 0.146 G_Rec: 0.439 D_GP: 0.069 D_real: 0.874 D_fake: 0.452 +(epoch: 270, iters: 6048, time: 0.064) G_GAN: 0.528 G_GAN_Feat: 0.743 G_ID: 0.138 G_Rec: 0.304 D_GP: 0.024 D_real: 1.371 D_fake: 0.476 +(epoch: 270, iters: 6448, time: 0.064) G_GAN: 0.394 G_GAN_Feat: 0.828 G_ID: 0.159 G_Rec: 0.382 D_GP: 0.021 D_real: 1.247 D_fake: 0.608 +(epoch: 270, iters: 6848, time: 0.064) G_GAN: 0.140 G_GAN_Feat: 0.600 G_ID: 0.124 G_Rec: 0.268 D_GP: 0.019 D_real: 1.097 D_fake: 0.860 +(epoch: 270, iters: 7248, time: 0.064) G_GAN: 0.374 G_GAN_Feat: 0.866 G_ID: 0.163 G_Rec: 0.389 D_GP: 0.022 D_real: 1.147 D_fake: 0.627 +(epoch: 270, iters: 7648, time: 0.064) G_GAN: 0.144 G_GAN_Feat: 0.704 G_ID: 0.140 G_Rec: 0.302 D_GP: 0.032 D_real: 0.981 D_fake: 0.856 +(epoch: 270, iters: 8048, time: 0.064) G_GAN: 0.201 G_GAN_Feat: 0.907 G_ID: 0.182 G_Rec: 0.386 D_GP: 0.029 D_real: 0.888 D_fake: 0.800 +(epoch: 270, iters: 8448, time: 0.064) G_GAN: 0.295 G_GAN_Feat: 0.745 G_ID: 0.157 G_Rec: 0.308 D_GP: 0.060 D_real: 0.953 D_fake: 0.705 +(epoch: 271, iters: 240, time: 0.064) G_GAN: 0.677 G_GAN_Feat: 0.908 G_ID: 0.149 G_Rec: 0.390 D_GP: 0.027 D_real: 1.356 D_fake: 0.335 +(epoch: 271, iters: 640, time: 0.064) G_GAN: 0.172 G_GAN_Feat: 0.797 G_ID: 0.121 G_Rec: 0.291 D_GP: 0.027 D_real: 0.761 D_fake: 0.828 +(epoch: 271, iters: 1040, time: 0.064) G_GAN: 0.544 G_GAN_Feat: 0.930 G_ID: 0.180 G_Rec: 0.394 D_GP: 0.027 D_real: 1.103 D_fake: 0.463 +(epoch: 271, iters: 1440, time: 0.064) G_GAN: 0.120 G_GAN_Feat: 0.678 G_ID: 0.140 G_Rec: 0.279 D_GP: 0.021 D_real: 1.022 D_fake: 0.880 +(epoch: 271, iters: 1840, time: 0.064) G_GAN: 0.452 G_GAN_Feat: 0.820 G_ID: 0.156 G_Rec: 0.413 D_GP: 0.020 D_real: 1.126 D_fake: 0.549 +(epoch: 271, iters: 2240, time: 0.064) G_GAN: 0.200 G_GAN_Feat: 0.715 G_ID: 0.129 G_Rec: 0.314 D_GP: 0.044 D_real: 1.077 D_fake: 0.800 +(epoch: 271, iters: 2640, time: 0.064) G_GAN: 0.507 G_GAN_Feat: 0.857 G_ID: 0.165 G_Rec: 0.403 D_GP: 0.021 D_real: 1.187 D_fake: 0.498 +(epoch: 271, iters: 3040, time: 0.064) G_GAN: 0.254 G_GAN_Feat: 0.647 G_ID: 0.176 G_Rec: 0.281 D_GP: 0.022 D_real: 1.201 D_fake: 0.764 +(epoch: 271, iters: 3440, time: 0.064) G_GAN: 0.552 G_GAN_Feat: 0.915 G_ID: 0.154 G_Rec: 0.371 D_GP: 0.023 D_real: 1.229 D_fake: 0.462 +(epoch: 271, iters: 3840, time: 0.064) G_GAN: 0.163 G_GAN_Feat: 0.887 G_ID: 0.129 G_Rec: 0.305 D_GP: 0.034 D_real: 0.439 D_fake: 0.837 +(epoch: 271, iters: 4240, time: 0.064) G_GAN: 0.393 G_GAN_Feat: 0.914 G_ID: 0.193 G_Rec: 0.432 D_GP: 0.024 D_real: 1.213 D_fake: 0.611 +(epoch: 271, iters: 4640, time: 0.064) G_GAN: 0.163 G_GAN_Feat: 0.615 G_ID: 0.144 G_Rec: 0.299 D_GP: 0.018 D_real: 1.124 D_fake: 0.838 +(epoch: 271, iters: 5040, time: 0.064) G_GAN: 0.165 G_GAN_Feat: 0.809 G_ID: 0.190 G_Rec: 0.366 D_GP: 0.030 D_real: 0.969 D_fake: 0.835 +(epoch: 271, iters: 5440, time: 0.064) G_GAN: 0.250 G_GAN_Feat: 0.663 G_ID: 0.127 G_Rec: 0.297 D_GP: 0.032 D_real: 1.127 D_fake: 0.753 +(epoch: 271, iters: 5840, time: 0.064) G_GAN: 0.282 G_GAN_Feat: 0.833 G_ID: 0.174 G_Rec: 0.390 D_GP: 0.027 D_real: 0.959 D_fake: 0.719 +(epoch: 271, iters: 6240, time: 0.064) G_GAN: 0.268 G_GAN_Feat: 0.836 G_ID: 0.105 G_Rec: 0.322 D_GP: 0.058 D_real: 0.816 D_fake: 0.736 +(epoch: 271, iters: 6640, time: 0.064) G_GAN: 0.218 G_GAN_Feat: 0.930 G_ID: 0.151 G_Rec: 0.408 D_GP: 0.024 D_real: 1.076 D_fake: 0.783 +(epoch: 271, iters: 7040, time: 0.064) G_GAN: 0.102 G_GAN_Feat: 0.785 G_ID: 0.119 G_Rec: 0.325 D_GP: 0.043 D_real: 0.872 D_fake: 0.900 +(epoch: 271, iters: 7440, time: 0.064) G_GAN: 0.571 G_GAN_Feat: 0.954 G_ID: 0.170 G_Rec: 0.349 D_GP: 0.035 D_real: 1.418 D_fake: 0.439 +(epoch: 271, iters: 7840, time: 0.064) G_GAN: 0.172 G_GAN_Feat: 0.725 G_ID: 0.128 G_Rec: 0.285 D_GP: 0.034 D_real: 0.977 D_fake: 0.828 +(epoch: 271, iters: 8240, time: 0.064) G_GAN: 0.449 G_GAN_Feat: 0.852 G_ID: 0.155 G_Rec: 0.357 D_GP: 0.023 D_real: 1.311 D_fake: 0.554 +(epoch: 272, iters: 32, time: 0.064) G_GAN: 0.301 G_GAN_Feat: 0.978 G_ID: 0.132 G_Rec: 0.341 D_GP: 0.108 D_real: 0.445 D_fake: 0.706 +(epoch: 272, iters: 432, time: 0.064) G_GAN: 0.330 G_GAN_Feat: 0.915 G_ID: 0.175 G_Rec: 0.397 D_GP: 0.027 D_real: 1.022 D_fake: 0.670 +(epoch: 272, iters: 832, time: 0.064) G_GAN: 0.289 G_GAN_Feat: 0.682 G_ID: 0.124 G_Rec: 0.336 D_GP: 0.028 D_real: 1.284 D_fake: 0.712 +(epoch: 272, iters: 1232, time: 0.064) G_GAN: 0.300 G_GAN_Feat: 0.776 G_ID: 0.174 G_Rec: 0.397 D_GP: 0.020 D_real: 1.082 D_fake: 0.700 +(epoch: 272, iters: 1632, time: 0.064) G_GAN: 0.283 G_GAN_Feat: 0.560 G_ID: 0.099 G_Rec: 0.254 D_GP: 0.021 D_real: 1.225 D_fake: 0.717 +(epoch: 272, iters: 2032, time: 0.064) G_GAN: -0.015 G_GAN_Feat: 0.963 G_ID: 0.158 G_Rec: 0.443 D_GP: 0.180 D_real: 0.612 D_fake: 1.015 +(epoch: 272, iters: 2432, time: 0.064) G_GAN: 0.263 G_GAN_Feat: 0.633 G_ID: 0.141 G_Rec: 0.273 D_GP: 0.037 D_real: 1.110 D_fake: 0.739 +(epoch: 272, iters: 2832, time: 0.064) G_GAN: 0.452 G_GAN_Feat: 0.974 G_ID: 0.169 G_Rec: 0.400 D_GP: 0.028 D_real: 1.195 D_fake: 0.552 +(epoch: 272, iters: 3232, time: 0.064) G_GAN: 0.048 G_GAN_Feat: 0.756 G_ID: 0.115 G_Rec: 0.330 D_GP: 0.044 D_real: 0.871 D_fake: 0.952 +(epoch: 272, iters: 3632, time: 0.064) G_GAN: 0.461 G_GAN_Feat: 0.862 G_ID: 0.164 G_Rec: 0.386 D_GP: 0.020 D_real: 1.258 D_fake: 0.540 +(epoch: 272, iters: 4032, time: 0.064) G_GAN: 0.245 G_GAN_Feat: 0.758 G_ID: 0.121 G_Rec: 0.312 D_GP: 0.040 D_real: 0.977 D_fake: 0.755 +(epoch: 272, iters: 4432, time: 0.064) G_GAN: 0.542 G_GAN_Feat: 0.946 G_ID: 0.160 G_Rec: 0.374 D_GP: 0.064 D_real: 0.897 D_fake: 0.462 +(epoch: 272, iters: 4832, time: 0.064) G_GAN: 0.663 G_GAN_Feat: 0.775 G_ID: 0.126 G_Rec: 0.281 D_GP: 0.112 D_real: 1.288 D_fake: 0.545 +(epoch: 272, iters: 5232, time: 0.064) G_GAN: 0.443 G_GAN_Feat: 1.027 G_ID: 0.160 G_Rec: 0.396 D_GP: 0.186 D_real: 0.459 D_fake: 0.561 +(epoch: 272, iters: 5632, time: 0.064) G_GAN: 0.297 G_GAN_Feat: 0.652 G_ID: 0.150 G_Rec: 0.307 D_GP: 0.019 D_real: 1.324 D_fake: 0.704 +(epoch: 272, iters: 6032, time: 0.064) G_GAN: 0.457 G_GAN_Feat: 0.780 G_ID: 0.180 G_Rec: 0.375 D_GP: 0.020 D_real: 1.271 D_fake: 0.546 +(epoch: 272, iters: 6432, time: 0.064) G_GAN: 0.284 G_GAN_Feat: 0.618 G_ID: 0.119 G_Rec: 0.297 D_GP: 0.022 D_real: 1.211 D_fake: 0.716 +(epoch: 272, iters: 6832, time: 0.064) G_GAN: 0.491 G_GAN_Feat: 0.735 G_ID: 0.173 G_Rec: 0.352 D_GP: 0.023 D_real: 1.333 D_fake: 0.512 +(epoch: 272, iters: 7232, time: 0.064) G_GAN: 0.136 G_GAN_Feat: 0.696 G_ID: 0.131 G_Rec: 0.328 D_GP: 0.034 D_real: 0.962 D_fake: 0.864 +(epoch: 272, iters: 7632, time: 0.064) G_GAN: 0.343 G_GAN_Feat: 0.974 G_ID: 0.169 G_Rec: 0.411 D_GP: 0.069 D_real: 0.892 D_fake: 0.660 +(epoch: 272, iters: 8032, time: 0.064) G_GAN: 0.165 G_GAN_Feat: 0.622 G_ID: 0.111 G_Rec: 0.272 D_GP: 0.026 D_real: 1.077 D_fake: 0.835 +(epoch: 272, iters: 8432, time: 0.064) G_GAN: 0.382 G_GAN_Feat: 0.822 G_ID: 0.166 G_Rec: 0.373 D_GP: 0.037 D_real: 1.076 D_fake: 0.619 +(epoch: 273, iters: 224, time: 0.064) G_GAN: 0.177 G_GAN_Feat: 0.631 G_ID: 0.105 G_Rec: 0.329 D_GP: 0.035 D_real: 1.056 D_fake: 0.823 +(epoch: 273, iters: 624, time: 0.064) G_GAN: 0.352 G_GAN_Feat: 0.886 G_ID: 0.141 G_Rec: 0.392 D_GP: 0.036 D_real: 0.927 D_fake: 0.650 +(epoch: 273, iters: 1024, time: 0.064) G_GAN: 0.109 G_GAN_Feat: 0.755 G_ID: 0.135 G_Rec: 0.358 D_GP: 0.053 D_real: 0.760 D_fake: 0.891 +(epoch: 273, iters: 1424, time: 0.064) G_GAN: 0.337 G_GAN_Feat: 0.811 G_ID: 0.175 G_Rec: 0.365 D_GP: 0.029 D_real: 1.064 D_fake: 0.663 +(epoch: 273, iters: 1824, time: 0.064) G_GAN: 0.383 G_GAN_Feat: 0.762 G_ID: 0.134 G_Rec: 0.341 D_GP: 0.047 D_real: 1.085 D_fake: 0.625 +(epoch: 273, iters: 2224, time: 0.064) G_GAN: 0.253 G_GAN_Feat: 1.014 G_ID: 0.157 G_Rec: 0.453 D_GP: 0.153 D_real: 0.456 D_fake: 0.748 +(epoch: 273, iters: 2624, time: 0.064) G_GAN: 0.244 G_GAN_Feat: 0.815 G_ID: 0.128 G_Rec: 0.296 D_GP: 0.028 D_real: 1.134 D_fake: 0.757 +(epoch: 273, iters: 3024, time: 0.064) G_GAN: 0.553 G_GAN_Feat: 0.934 G_ID: 0.172 G_Rec: 0.396 D_GP: 0.060 D_real: 0.954 D_fake: 0.458 +(epoch: 273, iters: 3424, time: 0.064) G_GAN: 0.190 G_GAN_Feat: 0.706 G_ID: 0.097 G_Rec: 0.277 D_GP: 0.027 D_real: 1.021 D_fake: 0.810 +(epoch: 273, iters: 3824, time: 0.064) G_GAN: 0.320 G_GAN_Feat: 0.825 G_ID: 0.172 G_Rec: 0.392 D_GP: 0.022 D_real: 1.116 D_fake: 0.681 +(epoch: 273, iters: 4224, time: 0.064) G_GAN: 0.073 G_GAN_Feat: 0.622 G_ID: 0.128 G_Rec: 0.289 D_GP: 0.019 D_real: 1.069 D_fake: 0.927 +(epoch: 273, iters: 4624, time: 0.064) G_GAN: 0.381 G_GAN_Feat: 0.879 G_ID: 0.155 G_Rec: 0.445 D_GP: 0.037 D_real: 0.930 D_fake: 0.625 +(epoch: 273, iters: 5024, time: 0.064) G_GAN: -0.040 G_GAN_Feat: 0.789 G_ID: 0.135 G_Rec: 0.318 D_GP: 0.121 D_real: 0.510 D_fake: 1.041 +(epoch: 273, iters: 5424, time: 0.064) G_GAN: 0.446 G_GAN_Feat: 0.822 G_ID: 0.176 G_Rec: 0.374 D_GP: 0.026 D_real: 1.134 D_fake: 0.555 +(epoch: 273, iters: 5824, time: 0.064) G_GAN: 0.212 G_GAN_Feat: 0.636 G_ID: 0.134 G_Rec: 0.285 D_GP: 0.028 D_real: 1.090 D_fake: 0.788 +(epoch: 273, iters: 6224, time: 0.064) G_GAN: 0.331 G_GAN_Feat: 0.798 G_ID: 0.168 G_Rec: 0.385 D_GP: 0.030 D_real: 1.098 D_fake: 0.669 +(epoch: 273, iters: 6624, time: 0.064) G_GAN: 0.252 G_GAN_Feat: 1.061 G_ID: 0.135 G_Rec: 0.371 D_GP: 1.437 D_real: 0.384 D_fake: 0.766 +(epoch: 273, iters: 7024, time: 0.064) G_GAN: 0.344 G_GAN_Feat: 0.859 G_ID: 0.161 G_Rec: 0.429 D_GP: 0.022 D_real: 1.153 D_fake: 0.659 +(epoch: 273, iters: 7424, time: 0.064) G_GAN: -0.042 G_GAN_Feat: 0.580 G_ID: 0.123 G_Rec: 0.302 D_GP: 0.019 D_real: 0.897 D_fake: 1.042 +(epoch: 273, iters: 7824, time: 0.064) G_GAN: 0.268 G_GAN_Feat: 0.852 G_ID: 0.165 G_Rec: 0.409 D_GP: 0.036 D_real: 0.943 D_fake: 0.732 +(epoch: 273, iters: 8224, time: 0.064) G_GAN: 0.130 G_GAN_Feat: 0.636 G_ID: 0.118 G_Rec: 0.272 D_GP: 0.024 D_real: 1.041 D_fake: 0.870 +(epoch: 274, iters: 16, time: 0.064) G_GAN: 0.438 G_GAN_Feat: 0.906 G_ID: 0.155 G_Rec: 0.420 D_GP: 0.060 D_real: 1.038 D_fake: 0.571 +(epoch: 274, iters: 416, time: 0.064) G_GAN: 0.296 G_GAN_Feat: 0.669 G_ID: 0.112 G_Rec: 0.302 D_GP: 0.022 D_real: 1.259 D_fake: 0.704 +(epoch: 274, iters: 816, time: 0.064) G_GAN: 0.280 G_GAN_Feat: 0.978 G_ID: 0.161 G_Rec: 0.397 D_GP: 0.302 D_real: 0.393 D_fake: 0.721 +(epoch: 274, iters: 1216, time: 0.064) G_GAN: 0.048 G_GAN_Feat: 0.773 G_ID: 0.122 G_Rec: 0.296 D_GP: 0.081 D_real: 0.699 D_fake: 0.952 +(epoch: 274, iters: 1616, time: 0.064) G_GAN: 0.353 G_GAN_Feat: 0.709 G_ID: 0.158 G_Rec: 0.381 D_GP: 0.017 D_real: 1.104 D_fake: 0.647 +(epoch: 274, iters: 2016, time: 0.064) G_GAN: 0.220 G_GAN_Feat: 0.557 G_ID: 0.110 G_Rec: 0.313 D_GP: 0.016 D_real: 1.154 D_fake: 0.781 +(epoch: 274, iters: 2416, time: 0.064) G_GAN: 0.251 G_GAN_Feat: 0.661 G_ID: 0.168 G_Rec: 0.351 D_GP: 0.021 D_real: 1.044 D_fake: 0.749 +(epoch: 274, iters: 2816, time: 0.064) G_GAN: 0.179 G_GAN_Feat: 0.527 G_ID: 0.103 G_Rec: 0.266 D_GP: 0.018 D_real: 1.102 D_fake: 0.821 +(epoch: 274, iters: 3216, time: 0.064) G_GAN: 0.353 G_GAN_Feat: 0.706 G_ID: 0.182 G_Rec: 0.349 D_GP: 0.025 D_real: 1.154 D_fake: 0.648 +(epoch: 274, iters: 3616, time: 0.064) G_GAN: -0.057 G_GAN_Feat: 0.721 G_ID: 0.119 G_Rec: 0.333 D_GP: 0.042 D_real: 0.737 D_fake: 1.058 +(epoch: 274, iters: 4016, time: 0.064) G_GAN: 0.442 G_GAN_Feat: 0.794 G_ID: 0.157 G_Rec: 0.404 D_GP: 0.029 D_real: 1.122 D_fake: 0.570 +(epoch: 274, iters: 4416, time: 0.064) G_GAN: 0.150 G_GAN_Feat: 0.641 G_ID: 0.140 G_Rec: 0.291 D_GP: 0.038 D_real: 1.020 D_fake: 0.852 +(epoch: 274, iters: 4816, time: 0.064) G_GAN: 0.055 G_GAN_Feat: 0.832 G_ID: 0.169 G_Rec: 0.399 D_GP: 0.044 D_real: 0.713 D_fake: 0.945 +(epoch: 274, iters: 5216, time: 0.064) G_GAN: 0.087 G_GAN_Feat: 0.780 G_ID: 0.130 G_Rec: 0.360 D_GP: 0.064 D_real: 0.675 D_fake: 0.913 +(epoch: 274, iters: 5616, time: 0.064) G_GAN: 0.288 G_GAN_Feat: 0.779 G_ID: 0.218 G_Rec: 0.399 D_GP: 0.028 D_real: 1.010 D_fake: 0.715 +(epoch: 274, iters: 6016, time: 0.064) G_GAN: 0.050 G_GAN_Feat: 0.715 G_ID: 0.130 G_Rec: 0.289 D_GP: 0.051 D_real: 0.846 D_fake: 0.950 +(epoch: 274, iters: 6416, time: 0.064) G_GAN: 0.353 G_GAN_Feat: 0.913 G_ID: 0.196 G_Rec: 0.396 D_GP: 0.050 D_real: 0.959 D_fake: 0.655 +(epoch: 274, iters: 6816, time: 0.064) G_GAN: 0.176 G_GAN_Feat: 0.599 G_ID: 0.115 G_Rec: 0.267 D_GP: 0.024 D_real: 1.105 D_fake: 0.825 +(epoch: 274, iters: 7216, time: 0.064) G_GAN: 0.214 G_GAN_Feat: 0.844 G_ID: 0.154 G_Rec: 0.400 D_GP: 0.021 D_real: 0.960 D_fake: 0.786 +(epoch: 274, iters: 7616, time: 0.064) G_GAN: 0.203 G_GAN_Feat: 0.609 G_ID: 0.127 G_Rec: 0.278 D_GP: 0.024 D_real: 1.057 D_fake: 0.797 +(epoch: 274, iters: 8016, time: 0.064) G_GAN: 0.301 G_GAN_Feat: 0.773 G_ID: 0.177 G_Rec: 0.361 D_GP: 0.029 D_real: 1.140 D_fake: 0.705 +(epoch: 274, iters: 8416, time: 0.064) G_GAN: 0.269 G_GAN_Feat: 0.673 G_ID: 0.117 G_Rec: 0.306 D_GP: 0.056 D_real: 1.071 D_fake: 0.732 +(epoch: 275, iters: 208, time: 0.064) G_GAN: 0.342 G_GAN_Feat: 0.720 G_ID: 0.172 G_Rec: 0.348 D_GP: 0.023 D_real: 1.172 D_fake: 0.658 +(epoch: 275, iters: 608, time: 0.064) G_GAN: 0.228 G_GAN_Feat: 0.738 G_ID: 0.113 G_Rec: 0.322 D_GP: 0.037 D_real: 1.037 D_fake: 0.774 +(epoch: 275, iters: 1008, time: 0.064) G_GAN: 0.171 G_GAN_Feat: 0.800 G_ID: 0.177 G_Rec: 0.359 D_GP: 0.032 D_real: 0.883 D_fake: 0.829 +(epoch: 275, iters: 1408, time: 0.064) G_GAN: 0.192 G_GAN_Feat: 0.705 G_ID: 0.117 G_Rec: 0.292 D_GP: 0.082 D_real: 0.957 D_fake: 0.808 +(epoch: 275, iters: 1808, time: 0.064) G_GAN: 0.319 G_GAN_Feat: 0.906 G_ID: 0.160 G_Rec: 0.384 D_GP: 0.064 D_real: 0.830 D_fake: 0.682 +(epoch: 275, iters: 2208, time: 0.064) G_GAN: 0.450 G_GAN_Feat: 0.746 G_ID: 0.131 G_Rec: 0.298 D_GP: 0.051 D_real: 1.119 D_fake: 0.552 +(epoch: 275, iters: 2608, time: 0.064) G_GAN: 0.404 G_GAN_Feat: 1.119 G_ID: 0.153 G_Rec: 0.424 D_GP: 0.226 D_real: 1.161 D_fake: 0.615 +(epoch: 275, iters: 3008, time: 0.064) G_GAN: 0.181 G_GAN_Feat: 0.718 G_ID: 0.157 G_Rec: 0.303 D_GP: 0.020 D_real: 1.024 D_fake: 0.821 +(epoch: 275, iters: 3408, time: 0.064) G_GAN: 0.267 G_GAN_Feat: 0.753 G_ID: 0.167 G_Rec: 0.335 D_GP: 0.035 D_real: 1.051 D_fake: 0.734 +(epoch: 275, iters: 3808, time: 0.064) G_GAN: 0.313 G_GAN_Feat: 0.665 G_ID: 0.131 G_Rec: 0.278 D_GP: 0.020 D_real: 1.241 D_fake: 0.687 +(epoch: 275, iters: 4208, time: 0.064) G_GAN: 0.575 G_GAN_Feat: 0.837 G_ID: 0.144 G_Rec: 0.401 D_GP: 0.032 D_real: 1.266 D_fake: 0.439 +(epoch: 275, iters: 4608, time: 0.064) G_GAN: 0.212 G_GAN_Feat: 0.730 G_ID: 0.126 G_Rec: 0.293 D_GP: 0.035 D_real: 0.967 D_fake: 0.790 +(epoch: 275, iters: 5008, time: 0.064) G_GAN: 0.502 G_GAN_Feat: 0.936 G_ID: 0.157 G_Rec: 0.401 D_GP: 0.026 D_real: 1.042 D_fake: 0.499 +(epoch: 275, iters: 5408, time: 0.064) G_GAN: 0.235 G_GAN_Feat: 0.674 G_ID: 0.138 G_Rec: 0.276 D_GP: 0.023 D_real: 1.068 D_fake: 0.765 +(epoch: 275, iters: 5808, time: 0.064) G_GAN: 0.299 G_GAN_Feat: 1.014 G_ID: 0.181 G_Rec: 0.428 D_GP: 0.213 D_real: 0.539 D_fake: 0.702 +(epoch: 275, iters: 6208, time: 0.064) G_GAN: 0.267 G_GAN_Feat: 0.884 G_ID: 0.108 G_Rec: 0.301 D_GP: 0.836 D_real: 0.518 D_fake: 0.742 +(epoch: 275, iters: 6608, time: 0.064) G_GAN: 0.254 G_GAN_Feat: 0.767 G_ID: 0.173 G_Rec: 0.348 D_GP: 0.020 D_real: 1.097 D_fake: 0.747 +(epoch: 275, iters: 7008, time: 0.064) G_GAN: 0.117 G_GAN_Feat: 0.883 G_ID: 0.110 G_Rec: 0.301 D_GP: 0.041 D_real: 0.495 D_fake: 0.883 +(epoch: 275, iters: 7408, time: 0.064) G_GAN: 0.559 G_GAN_Feat: 1.080 G_ID: 0.144 G_Rec: 0.442 D_GP: 0.079 D_real: 0.562 D_fake: 0.458 +(epoch: 275, iters: 7808, time: 0.064) G_GAN: 0.364 G_GAN_Feat: 0.873 G_ID: 0.118 G_Rec: 0.318 D_GP: 0.031 D_real: 0.811 D_fake: 0.645 +(epoch: 275, iters: 8208, time: 0.064) G_GAN: 0.446 G_GAN_Feat: 0.832 G_ID: 0.166 G_Rec: 0.429 D_GP: 0.018 D_real: 1.228 D_fake: 0.558 +(epoch: 275, iters: 8608, time: 0.064) G_GAN: -0.088 G_GAN_Feat: 0.584 G_ID: 0.147 G_Rec: 0.254 D_GP: 0.021 D_real: 0.836 D_fake: 1.088 +(epoch: 276, iters: 400, time: 0.064) G_GAN: 0.599 G_GAN_Feat: 0.962 G_ID: 0.170 G_Rec: 0.417 D_GP: 0.046 D_real: 1.069 D_fake: 0.418 +(epoch: 276, iters: 800, time: 0.064) G_GAN: 0.374 G_GAN_Feat: 0.687 G_ID: 0.120 G_Rec: 0.313 D_GP: 0.036 D_real: 1.207 D_fake: 0.627 +(epoch: 276, iters: 1200, time: 0.064) G_GAN: 0.619 G_GAN_Feat: 0.967 G_ID: 0.162 G_Rec: 0.431 D_GP: 0.025 D_real: 1.114 D_fake: 0.430 +(epoch: 276, iters: 1600, time: 0.064) G_GAN: 0.276 G_GAN_Feat: 0.695 G_ID: 0.158 G_Rec: 0.309 D_GP: 0.020 D_real: 1.165 D_fake: 0.724 +(epoch: 276, iters: 2000, time: 0.064) G_GAN: 0.449 G_GAN_Feat: 0.978 G_ID: 0.161 G_Rec: 0.402 D_GP: 0.062 D_real: 0.715 D_fake: 0.567 +(epoch: 276, iters: 2400, time: 0.064) G_GAN: 0.088 G_GAN_Feat: 0.663 G_ID: 0.115 G_Rec: 0.296 D_GP: 0.022 D_real: 1.178 D_fake: 0.914 +(epoch: 276, iters: 2800, time: 0.064) G_GAN: 0.427 G_GAN_Feat: 0.677 G_ID: 0.159 G_Rec: 0.327 D_GP: 0.020 D_real: 1.334 D_fake: 0.573 +(epoch: 276, iters: 3200, time: 0.064) G_GAN: 0.321 G_GAN_Feat: 0.572 G_ID: 0.126 G_Rec: 0.277 D_GP: 0.021 D_real: 1.285 D_fake: 0.679 +(epoch: 276, iters: 3600, time: 0.064) G_GAN: 0.432 G_GAN_Feat: 0.735 G_ID: 0.144 G_Rec: 0.340 D_GP: 0.022 D_real: 1.276 D_fake: 0.573 +(epoch: 276, iters: 4000, time: 0.064) G_GAN: 0.211 G_GAN_Feat: 0.611 G_ID: 0.132 G_Rec: 0.290 D_GP: 0.025 D_real: 1.097 D_fake: 0.789 +(epoch: 276, iters: 4400, time: 0.064) G_GAN: 0.442 G_GAN_Feat: 0.806 G_ID: 0.160 G_Rec: 0.386 D_GP: 0.023 D_real: 1.188 D_fake: 0.565 +(epoch: 276, iters: 4800, time: 0.064) G_GAN: 0.091 G_GAN_Feat: 0.718 G_ID: 0.132 G_Rec: 0.317 D_GP: 0.060 D_real: 0.803 D_fake: 0.909 +(epoch: 276, iters: 5200, time: 0.064) G_GAN: 0.350 G_GAN_Feat: 0.810 G_ID: 0.166 G_Rec: 0.400 D_GP: 0.024 D_real: 1.126 D_fake: 0.651 +(epoch: 276, iters: 5600, time: 0.064) G_GAN: 0.064 G_GAN_Feat: 0.678 G_ID: 0.117 G_Rec: 0.314 D_GP: 0.030 D_real: 0.922 D_fake: 0.936 +(epoch: 276, iters: 6000, time: 0.064) G_GAN: 0.259 G_GAN_Feat: 0.785 G_ID: 0.166 G_Rec: 0.375 D_GP: 0.028 D_real: 1.015 D_fake: 0.741 +(epoch: 276, iters: 6400, time: 0.064) G_GAN: 0.160 G_GAN_Feat: 0.642 G_ID: 0.113 G_Rec: 0.272 D_GP: 0.022 D_real: 1.048 D_fake: 0.840 +(epoch: 276, iters: 6800, time: 0.064) G_GAN: 0.366 G_GAN_Feat: 0.857 G_ID: 0.150 G_Rec: 0.375 D_GP: 0.028 D_real: 1.078 D_fake: 0.635 +(epoch: 276, iters: 7200, time: 0.064) G_GAN: 0.182 G_GAN_Feat: 0.577 G_ID: 0.119 G_Rec: 0.261 D_GP: 0.024 D_real: 1.133 D_fake: 0.818 +(epoch: 276, iters: 7600, time: 0.064) G_GAN: 0.382 G_GAN_Feat: 0.905 G_ID: 0.140 G_Rec: 0.397 D_GP: 0.036 D_real: 0.950 D_fake: 0.621 +(epoch: 276, iters: 8000, time: 0.064) G_GAN: 0.037 G_GAN_Feat: 0.830 G_ID: 0.144 G_Rec: 0.339 D_GP: 0.083 D_real: 0.769 D_fake: 0.963 +(epoch: 276, iters: 8400, time: 0.064) G_GAN: 0.249 G_GAN_Feat: 0.859 G_ID: 0.176 G_Rec: 0.384 D_GP: 0.021 D_real: 0.939 D_fake: 0.751 +(epoch: 277, iters: 192, time: 0.064) G_GAN: 0.315 G_GAN_Feat: 0.704 G_ID: 0.116 G_Rec: 0.282 D_GP: 0.042 D_real: 1.137 D_fake: 0.686 +(epoch: 277, iters: 592, time: 0.064) G_GAN: 0.380 G_GAN_Feat: 0.853 G_ID: 0.157 G_Rec: 0.376 D_GP: 0.022 D_real: 1.127 D_fake: 0.621 +(epoch: 277, iters: 992, time: 0.064) G_GAN: 0.200 G_GAN_Feat: 0.742 G_ID: 0.147 G_Rec: 0.384 D_GP: 0.047 D_real: 0.957 D_fake: 0.802 +(epoch: 277, iters: 1392, time: 0.064) G_GAN: 0.368 G_GAN_Feat: 0.935 G_ID: 0.161 G_Rec: 0.408 D_GP: 0.036 D_real: 0.943 D_fake: 0.634 +(epoch: 277, iters: 1792, time: 0.064) G_GAN: 0.273 G_GAN_Feat: 0.684 G_ID: 0.125 G_Rec: 0.271 D_GP: 0.029 D_real: 1.092 D_fake: 0.729 +(epoch: 277, iters: 2192, time: 0.064) G_GAN: 0.260 G_GAN_Feat: 0.834 G_ID: 0.185 G_Rec: 0.366 D_GP: 0.021 D_real: 1.043 D_fake: 0.742 +(epoch: 277, iters: 2592, time: 0.064) G_GAN: 0.237 G_GAN_Feat: 0.902 G_ID: 0.118 G_Rec: 0.316 D_GP: 0.095 D_real: 0.582 D_fake: 0.764 +(epoch: 277, iters: 2992, time: 0.064) G_GAN: 0.489 G_GAN_Feat: 0.929 G_ID: 0.188 G_Rec: 0.380 D_GP: 0.047 D_real: 1.153 D_fake: 0.557 +(epoch: 277, iters: 3392, time: 0.064) G_GAN: 0.255 G_GAN_Feat: 0.631 G_ID: 0.114 G_Rec: 0.301 D_GP: 0.019 D_real: 1.244 D_fake: 0.747 +(epoch: 277, iters: 3792, time: 0.064) G_GAN: 0.219 G_GAN_Feat: 0.730 G_ID: 0.148 G_Rec: 0.354 D_GP: 0.022 D_real: 1.025 D_fake: 0.781 +(epoch: 277, iters: 4192, time: 0.064) G_GAN: 0.087 G_GAN_Feat: 0.719 G_ID: 0.135 G_Rec: 0.306 D_GP: 0.051 D_real: 0.914 D_fake: 0.914 +(epoch: 277, iters: 4592, time: 0.064) G_GAN: 0.340 G_GAN_Feat: 0.953 G_ID: 0.167 G_Rec: 0.395 D_GP: 0.055 D_real: 0.806 D_fake: 0.661 +(epoch: 277, iters: 4992, time: 0.064) G_GAN: 0.379 G_GAN_Feat: 0.800 G_ID: 0.122 G_Rec: 0.333 D_GP: 0.039 D_real: 1.110 D_fake: 0.630 +(epoch: 277, iters: 5392, time: 0.064) G_GAN: 0.105 G_GAN_Feat: 0.849 G_ID: 0.180 G_Rec: 0.338 D_GP: 0.053 D_real: 0.761 D_fake: 0.898 +(epoch: 277, iters: 5792, time: 0.064) G_GAN: -0.208 G_GAN_Feat: 0.902 G_ID: 0.144 G_Rec: 0.336 D_GP: 0.130 D_real: 0.122 D_fake: 1.209 +(epoch: 277, iters: 6192, time: 0.064) G_GAN: 0.638 G_GAN_Feat: 0.976 G_ID: 0.173 G_Rec: 0.400 D_GP: 0.038 D_real: 1.188 D_fake: 0.382 +(epoch: 277, iters: 6592, time: 0.064) G_GAN: -0.005 G_GAN_Feat: 0.747 G_ID: 0.127 G_Rec: 0.310 D_GP: 0.028 D_real: 0.881 D_fake: 1.005 +(epoch: 277, iters: 6992, time: 0.064) G_GAN: 0.222 G_GAN_Feat: 0.905 G_ID: 0.166 G_Rec: 0.397 D_GP: 0.022 D_real: 0.907 D_fake: 0.778 +(epoch: 277, iters: 7392, time: 0.064) G_GAN: 0.389 G_GAN_Feat: 0.722 G_ID: 0.117 G_Rec: 0.328 D_GP: 0.036 D_real: 1.295 D_fake: 0.612 +(epoch: 277, iters: 7792, time: 0.064) G_GAN: 0.486 G_GAN_Feat: 1.067 G_ID: 0.151 G_Rec: 0.418 D_GP: 0.046 D_real: 0.470 D_fake: 0.517 +(epoch: 277, iters: 8192, time: 0.064) G_GAN: 0.255 G_GAN_Feat: 0.709 G_ID: 0.116 G_Rec: 0.309 D_GP: 0.018 D_real: 1.195 D_fake: 0.746 +(epoch: 277, iters: 8592, time: 0.064) G_GAN: 0.231 G_GAN_Feat: 0.935 G_ID: 0.149 G_Rec: 0.389 D_GP: 0.039 D_real: 0.759 D_fake: 0.780 +(epoch: 278, iters: 384, time: 0.064) G_GAN: 0.180 G_GAN_Feat: 0.815 G_ID: 0.117 G_Rec: 0.314 D_GP: 0.107 D_real: 0.558 D_fake: 0.821 +(epoch: 278, iters: 784, time: 0.064) G_GAN: 0.221 G_GAN_Feat: 0.914 G_ID: 0.187 G_Rec: 0.405 D_GP: 0.032 D_real: 1.071 D_fake: 0.780 +(epoch: 278, iters: 1184, time: 0.064) G_GAN: 0.091 G_GAN_Feat: 0.626 G_ID: 0.125 G_Rec: 0.277 D_GP: 0.020 D_real: 1.053 D_fake: 0.909 +(epoch: 278, iters: 1584, time: 0.064) G_GAN: 0.460 G_GAN_Feat: 0.803 G_ID: 0.161 G_Rec: 0.361 D_GP: 0.021 D_real: 1.282 D_fake: 0.541 +(epoch: 278, iters: 1984, time: 0.064) G_GAN: 0.248 G_GAN_Feat: 1.031 G_ID: 0.121 G_Rec: 0.314 D_GP: 0.384 D_real: 0.402 D_fake: 0.753 +(epoch: 278, iters: 2384, time: 0.064) G_GAN: -0.027 G_GAN_Feat: 1.030 G_ID: 0.178 G_Rec: 0.381 D_GP: 0.096 D_real: 0.196 D_fake: 1.027 +(epoch: 278, iters: 2784, time: 0.064) G_GAN: 0.105 G_GAN_Feat: 0.763 G_ID: 0.117 G_Rec: 0.291 D_GP: 0.035 D_real: 0.870 D_fake: 0.895 +(epoch: 278, iters: 3184, time: 0.064) G_GAN: 0.418 G_GAN_Feat: 1.005 G_ID: 0.151 G_Rec: 0.419 D_GP: 0.044 D_real: 0.834 D_fake: 0.585 +(epoch: 278, iters: 3584, time: 0.064) G_GAN: 0.406 G_GAN_Feat: 0.721 G_ID: 0.142 G_Rec: 0.293 D_GP: 0.025 D_real: 1.204 D_fake: 0.597 +(epoch: 278, iters: 3984, time: 0.064) G_GAN: 0.563 G_GAN_Feat: 0.875 G_ID: 0.141 G_Rec: 0.369 D_GP: 0.025 D_real: 1.214 D_fake: 0.441 +(epoch: 278, iters: 4384, time: 0.064) G_GAN: 0.390 G_GAN_Feat: 0.696 G_ID: 0.151 G_Rec: 0.308 D_GP: 0.022 D_real: 1.246 D_fake: 0.615 +(epoch: 278, iters: 4784, time: 0.064) G_GAN: 0.345 G_GAN_Feat: 0.742 G_ID: 0.181 G_Rec: 0.391 D_GP: 0.018 D_real: 1.171 D_fake: 0.660 +(epoch: 278, iters: 5184, time: 0.064) G_GAN: 0.111 G_GAN_Feat: 0.606 G_ID: 0.107 G_Rec: 0.301 D_GP: 0.017 D_real: 1.112 D_fake: 0.889 +(epoch: 278, iters: 5584, time: 0.064) G_GAN: 0.093 G_GAN_Feat: 0.708 G_ID: 0.191 G_Rec: 0.358 D_GP: 0.019 D_real: 0.903 D_fake: 0.907 +(epoch: 278, iters: 5984, time: 0.064) G_GAN: 0.128 G_GAN_Feat: 0.518 G_ID: 0.115 G_Rec: 0.252 D_GP: 0.020 D_real: 1.100 D_fake: 0.872 +(epoch: 278, iters: 6384, time: 0.064) G_GAN: 0.134 G_GAN_Feat: 0.723 G_ID: 0.167 G_Rec: 0.344 D_GP: 0.024 D_real: 0.964 D_fake: 0.868 +(epoch: 278, iters: 6784, time: 0.064) G_GAN: -0.101 G_GAN_Feat: 0.694 G_ID: 0.138 G_Rec: 0.324 D_GP: 0.040 D_real: 0.737 D_fake: 1.101 +(epoch: 278, iters: 7184, time: 0.064) G_GAN: 0.480 G_GAN_Feat: 0.901 G_ID: 0.153 G_Rec: 0.384 D_GP: 0.037 D_real: 1.084 D_fake: 0.534 +(epoch: 278, iters: 7584, time: 0.064) G_GAN: 0.298 G_GAN_Feat: 0.622 G_ID: 0.116 G_Rec: 0.300 D_GP: 0.032 D_real: 1.219 D_fake: 0.702 +(epoch: 278, iters: 7984, time: 0.064) G_GAN: 0.311 G_GAN_Feat: 0.784 G_ID: 0.154 G_Rec: 0.362 D_GP: 0.028 D_real: 1.134 D_fake: 0.691 +(epoch: 278, iters: 8384, time: 0.064) G_GAN: -0.182 G_GAN_Feat: 0.758 G_ID: 0.120 G_Rec: 0.314 D_GP: 0.048 D_real: 0.958 D_fake: 1.182 +(epoch: 279, iters: 176, time: 0.064) G_GAN: 0.279 G_GAN_Feat: 0.910 G_ID: 0.177 G_Rec: 0.380 D_GP: 0.117 D_real: 0.782 D_fake: 0.725 +(epoch: 279, iters: 576, time: 0.064) G_GAN: 0.332 G_GAN_Feat: 0.689 G_ID: 0.134 G_Rec: 0.299 D_GP: 0.039 D_real: 1.131 D_fake: 0.669 +(epoch: 279, iters: 976, time: 0.064) G_GAN: 0.278 G_GAN_Feat: 0.907 G_ID: 0.167 G_Rec: 0.437 D_GP: 0.126 D_real: 0.660 D_fake: 0.725 +(epoch: 279, iters: 1376, time: 0.064) G_GAN: 0.261 G_GAN_Feat: 0.852 G_ID: 0.124 G_Rec: 0.315 D_GP: 0.104 D_real: 0.675 D_fake: 0.754 +(epoch: 279, iters: 1776, time: 0.064) G_GAN: 0.070 G_GAN_Feat: 0.781 G_ID: 0.167 G_Rec: 0.381 D_GP: 0.025 D_real: 0.922 D_fake: 0.930 +(epoch: 279, iters: 2176, time: 0.064) G_GAN: -0.009 G_GAN_Feat: 0.930 G_ID: 0.122 G_Rec: 0.358 D_GP: 0.307 D_real: 0.308 D_fake: 1.010 +(epoch: 279, iters: 2576, time: 0.064) G_GAN: 0.715 G_GAN_Feat: 1.205 G_ID: 0.163 G_Rec: 0.436 D_GP: 0.066 D_real: 1.284 D_fake: 0.405 +(epoch: 279, iters: 2976, time: 0.064) G_GAN: 0.040 G_GAN_Feat: 0.623 G_ID: 0.128 G_Rec: 0.284 D_GP: 0.020 D_real: 0.974 D_fake: 0.962 +(epoch: 279, iters: 3376, time: 0.064) G_GAN: 0.032 G_GAN_Feat: 0.929 G_ID: 0.181 G_Rec: 0.392 D_GP: 0.080 D_real: 0.496 D_fake: 0.968 +(epoch: 279, iters: 3776, time: 0.064) G_GAN: 0.130 G_GAN_Feat: 0.813 G_ID: 0.118 G_Rec: 0.317 D_GP: 0.116 D_real: 0.442 D_fake: 0.874 +(epoch: 279, iters: 4176, time: 0.064) G_GAN: 0.361 G_GAN_Feat: 0.849 G_ID: 0.170 G_Rec: 0.405 D_GP: 0.030 D_real: 1.098 D_fake: 0.640 +(epoch: 279, iters: 4576, time: 0.064) G_GAN: 0.283 G_GAN_Feat: 0.656 G_ID: 0.140 G_Rec: 0.285 D_GP: 0.021 D_real: 1.162 D_fake: 0.719 +(epoch: 279, iters: 4976, time: 0.064) G_GAN: 0.605 G_GAN_Feat: 0.806 G_ID: 0.156 G_Rec: 0.386 D_GP: 0.022 D_real: 1.365 D_fake: 0.399 +(epoch: 279, iters: 5376, time: 0.064) G_GAN: 0.244 G_GAN_Feat: 0.626 G_ID: 0.108 G_Rec: 0.279 D_GP: 0.028 D_real: 1.149 D_fake: 0.757 +(epoch: 279, iters: 5776, time: 0.064) G_GAN: 0.228 G_GAN_Feat: 0.940 G_ID: 0.193 G_Rec: 0.419 D_GP: 0.056 D_real: 0.611 D_fake: 0.772 +(epoch: 279, iters: 6176, time: 0.064) G_GAN: 0.153 G_GAN_Feat: 0.792 G_ID: 0.126 G_Rec: 0.314 D_GP: 0.075 D_real: 0.773 D_fake: 0.848 +(epoch: 279, iters: 6576, time: 0.064) G_GAN: 0.553 G_GAN_Feat: 1.053 G_ID: 0.186 G_Rec: 0.406 D_GP: 0.160 D_real: 0.700 D_fake: 0.556 +(epoch: 279, iters: 6976, time: 0.064) G_GAN: 0.399 G_GAN_Feat: 0.705 G_ID: 0.124 G_Rec: 0.306 D_GP: 0.024 D_real: 1.382 D_fake: 0.615 +(epoch: 279, iters: 7376, time: 0.064) G_GAN: 0.316 G_GAN_Feat: 0.935 G_ID: 0.181 G_Rec: 0.362 D_GP: 0.032 D_real: 0.795 D_fake: 0.685 +(epoch: 279, iters: 7776, time: 0.064) G_GAN: 0.371 G_GAN_Feat: 0.707 G_ID: 0.110 G_Rec: 0.289 D_GP: 0.025 D_real: 1.170 D_fake: 0.633 +(epoch: 279, iters: 8176, time: 0.064) G_GAN: 0.578 G_GAN_Feat: 0.877 G_ID: 0.154 G_Rec: 0.389 D_GP: 0.021 D_real: 1.427 D_fake: 0.445 +(epoch: 279, iters: 8576, time: 0.064) G_GAN: 0.230 G_GAN_Feat: 0.717 G_ID: 0.135 G_Rec: 0.296 D_GP: 0.024 D_real: 1.182 D_fake: 0.770 +(epoch: 280, iters: 368, time: 0.064) G_GAN: 0.381 G_GAN_Feat: 0.890 G_ID: 0.132 G_Rec: 0.394 D_GP: 0.026 D_real: 1.073 D_fake: 0.620 +(epoch: 280, iters: 768, time: 0.064) G_GAN: 0.132 G_GAN_Feat: 0.737 G_ID: 0.131 G_Rec: 0.314 D_GP: 0.057 D_real: 0.818 D_fake: 0.869 +(epoch: 280, iters: 1168, time: 0.064) G_GAN: 0.466 G_GAN_Feat: 0.924 G_ID: 0.202 G_Rec: 0.383 D_GP: 0.078 D_real: 1.066 D_fake: 0.540 +(epoch: 280, iters: 1568, time: 0.064) G_GAN: 0.368 G_GAN_Feat: 0.666 G_ID: 0.136 G_Rec: 0.302 D_GP: 0.028 D_real: 1.303 D_fake: 0.633 +(epoch: 280, iters: 1968, time: 0.064) G_GAN: 0.324 G_GAN_Feat: 0.963 G_ID: 0.176 G_Rec: 0.441 D_GP: 0.207 D_real: 0.649 D_fake: 0.677 +(epoch: 280, iters: 2368, time: 0.064) G_GAN: 0.114 G_GAN_Feat: 0.787 G_ID: 0.145 G_Rec: 0.309 D_GP: 0.027 D_real: 0.871 D_fake: 0.886 +(epoch: 280, iters: 2768, time: 0.064) G_GAN: 0.320 G_GAN_Feat: 0.944 G_ID: 0.172 G_Rec: 0.430 D_GP: 0.044 D_real: 0.918 D_fake: 0.682 +(epoch: 280, iters: 3168, time: 0.064) G_GAN: 0.290 G_GAN_Feat: 1.021 G_ID: 0.132 G_Rec: 0.350 D_GP: 0.360 D_real: 0.416 D_fake: 0.722 +(epoch: 280, iters: 3568, time: 0.064) G_GAN: -0.012 G_GAN_Feat: 1.043 G_ID: 0.173 G_Rec: 0.399 D_GP: 0.042 D_real: 0.545 D_fake: 1.012 +(epoch: 280, iters: 3968, time: 0.064) G_GAN: 0.230 G_GAN_Feat: 0.720 G_ID: 0.125 G_Rec: 0.272 D_GP: 0.028 D_real: 0.983 D_fake: 0.770 +(epoch: 280, iters: 4368, time: 0.064) G_GAN: 0.297 G_GAN_Feat: 0.747 G_ID: 0.183 G_Rec: 0.379 D_GP: 0.019 D_real: 1.070 D_fake: 0.703 +(epoch: 280, iters: 4768, time: 0.064) G_GAN: 0.013 G_GAN_Feat: 0.601 G_ID: 0.118 G_Rec: 0.269 D_GP: 0.032 D_real: 0.946 D_fake: 0.987 +(epoch: 280, iters: 5168, time: 0.064) G_GAN: 0.250 G_GAN_Feat: 0.733 G_ID: 0.151 G_Rec: 0.330 D_GP: 0.023 D_real: 1.099 D_fake: 0.750 +(epoch: 280, iters: 5568, time: 0.064) G_GAN: -0.002 G_GAN_Feat: 0.633 G_ID: 0.126 G_Rec: 0.294 D_GP: 0.021 D_real: 0.962 D_fake: 1.002 +(epoch: 280, iters: 5968, time: 0.064) G_GAN: 0.307 G_GAN_Feat: 0.875 G_ID: 0.173 G_Rec: 0.416 D_GP: 0.067 D_real: 0.933 D_fake: 0.695 +(epoch: 280, iters: 6368, time: 0.064) G_GAN: 0.254 G_GAN_Feat: 0.771 G_ID: 0.130 G_Rec: 0.343 D_GP: 0.029 D_real: 1.005 D_fake: 0.747 +(epoch: 280, iters: 6768, time: 0.064) G_GAN: 0.386 G_GAN_Feat: 0.849 G_ID: 0.173 G_Rec: 0.393 D_GP: 0.026 D_real: 1.177 D_fake: 0.617 +(epoch: 280, iters: 7168, time: 0.064) G_GAN: 0.160 G_GAN_Feat: 0.581 G_ID: 0.132 G_Rec: 0.247 D_GP: 0.022 D_real: 1.126 D_fake: 0.840 +(epoch: 280, iters: 7568, time: 0.064) G_GAN: 0.181 G_GAN_Feat: 0.830 G_ID: 0.168 G_Rec: 0.349 D_GP: 0.038 D_real: 0.942 D_fake: 0.819 +(epoch: 280, iters: 7968, time: 0.064) G_GAN: 0.272 G_GAN_Feat: 0.743 G_ID: 0.128 G_Rec: 0.299 D_GP: 0.044 D_real: 0.975 D_fake: 0.731 +(epoch: 280, iters: 8368, time: 0.064) G_GAN: 0.298 G_GAN_Feat: 1.089 G_ID: 0.174 G_Rec: 0.432 D_GP: 0.177 D_real: 0.297 D_fake: 0.703 +(epoch: 281, iters: 160, time: 0.064) G_GAN: 0.283 G_GAN_Feat: 0.629 G_ID: 0.116 G_Rec: 0.275 D_GP: 0.022 D_real: 1.210 D_fake: 0.719 +(epoch: 281, iters: 560, time: 0.064) G_GAN: 0.355 G_GAN_Feat: 0.846 G_ID: 0.175 G_Rec: 0.347 D_GP: 0.029 D_real: 1.018 D_fake: 0.645 +(epoch: 281, iters: 960, time: 0.064) G_GAN: 0.210 G_GAN_Feat: 0.727 G_ID: 0.135 G_Rec: 0.274 D_GP: 0.029 D_real: 0.964 D_fake: 0.790 +(epoch: 281, iters: 1360, time: 0.064) G_GAN: 0.578 G_GAN_Feat: 0.923 G_ID: 0.180 G_Rec: 0.399 D_GP: 0.028 D_real: 1.236 D_fake: 0.429 +(epoch: 281, iters: 1760, time: 0.064) G_GAN: 0.325 G_GAN_Feat: 0.701 G_ID: 0.133 G_Rec: 0.282 D_GP: 0.022 D_real: 1.167 D_fake: 0.676 +(epoch: 281, iters: 2160, time: 0.064) G_GAN: 0.217 G_GAN_Feat: 0.786 G_ID: 0.189 G_Rec: 0.355 D_GP: 0.021 D_real: 1.024 D_fake: 0.783 +(epoch: 281, iters: 2560, time: 0.064) G_GAN: 0.095 G_GAN_Feat: 0.650 G_ID: 0.133 G_Rec: 0.310 D_GP: 0.022 D_real: 1.073 D_fake: 0.905 +(epoch: 281, iters: 2960, time: 0.064) G_GAN: 0.258 G_GAN_Feat: 0.846 G_ID: 0.148 G_Rec: 0.385 D_GP: 0.058 D_real: 0.975 D_fake: 0.743 +(epoch: 281, iters: 3360, time: 0.064) G_GAN: 0.276 G_GAN_Feat: 0.645 G_ID: 0.113 G_Rec: 0.274 D_GP: 0.031 D_real: 1.137 D_fake: 0.724 +(epoch: 281, iters: 3760, time: 0.064) G_GAN: 0.246 G_GAN_Feat: 0.939 G_ID: 0.178 G_Rec: 0.409 D_GP: 0.076 D_real: 0.793 D_fake: 0.754 +(epoch: 281, iters: 4160, time: 0.064) G_GAN: 0.153 G_GAN_Feat: 0.797 G_ID: 0.136 G_Rec: 0.283 D_GP: 0.048 D_real: 0.601 D_fake: 0.847 +(epoch: 281, iters: 4560, time: 0.064) G_GAN: 0.844 G_GAN_Feat: 0.932 G_ID: 0.148 G_Rec: 0.406 D_GP: 0.025 D_real: 1.586 D_fake: 0.203 +(epoch: 281, iters: 4960, time: 0.064) G_GAN: -0.065 G_GAN_Feat: 0.661 G_ID: 0.149 G_Rec: 0.333 D_GP: 0.019 D_real: 0.863 D_fake: 1.065 +(epoch: 281, iters: 5360, time: 0.064) G_GAN: 0.188 G_GAN_Feat: 0.871 G_ID: 0.163 G_Rec: 0.376 D_GP: 0.024 D_real: 0.871 D_fake: 0.812 +(epoch: 281, iters: 5760, time: 0.064) G_GAN: -0.009 G_GAN_Feat: 0.716 G_ID: 0.149 G_Rec: 0.330 D_GP: 0.045 D_real: 0.756 D_fake: 1.009 +(epoch: 281, iters: 6160, time: 0.064) G_GAN: 0.259 G_GAN_Feat: 0.844 G_ID: 0.165 G_Rec: 0.365 D_GP: 0.029 D_real: 1.031 D_fake: 0.744 +(epoch: 281, iters: 6560, time: 0.064) G_GAN: 0.222 G_GAN_Feat: 0.753 G_ID: 0.117 G_Rec: 0.315 D_GP: 0.031 D_real: 0.969 D_fake: 0.782 +(epoch: 281, iters: 6960, time: 0.064) G_GAN: 0.477 G_GAN_Feat: 0.916 G_ID: 0.164 G_Rec: 0.416 D_GP: 0.035 D_real: 1.073 D_fake: 0.527 +(epoch: 281, iters: 7360, time: 0.064) G_GAN: 0.348 G_GAN_Feat: 0.797 G_ID: 0.114 G_Rec: 0.333 D_GP: 0.033 D_real: 0.977 D_fake: 0.662 +(epoch: 281, iters: 7760, time: 0.064) G_GAN: 0.438 G_GAN_Feat: 0.864 G_ID: 0.156 G_Rec: 0.386 D_GP: 0.022 D_real: 1.196 D_fake: 0.564 +(epoch: 281, iters: 8160, time: 0.064) G_GAN: 0.336 G_GAN_Feat: 0.642 G_ID: 0.134 G_Rec: 0.281 D_GP: 0.020 D_real: 1.241 D_fake: 0.664 +(epoch: 281, iters: 8560, time: 0.064) G_GAN: 0.336 G_GAN_Feat: 0.877 G_ID: 0.169 G_Rec: 0.413 D_GP: 0.023 D_real: 1.132 D_fake: 0.665 +(epoch: 282, iters: 352, time: 0.064) G_GAN: 0.050 G_GAN_Feat: 0.645 G_ID: 0.129 G_Rec: 0.282 D_GP: 0.023 D_real: 0.969 D_fake: 0.950 +(epoch: 282, iters: 752, time: 0.064) G_GAN: -0.065 G_GAN_Feat: 0.836 G_ID: 0.194 G_Rec: 0.367 D_GP: 0.026 D_real: 0.768 D_fake: 1.065 +(epoch: 282, iters: 1152, time: 0.064) G_GAN: 0.345 G_GAN_Feat: 0.740 G_ID: 0.122 G_Rec: 0.288 D_GP: 0.028 D_real: 1.044 D_fake: 0.656 +(epoch: 282, iters: 1552, time: 0.064) G_GAN: 0.667 G_GAN_Feat: 1.217 G_ID: 0.175 G_Rec: 0.497 D_GP: 0.052 D_real: 0.630 D_fake: 0.377 +(epoch: 282, iters: 1952, time: 0.064) G_GAN: 0.143 G_GAN_Feat: 0.845 G_ID: 0.132 G_Rec: 0.297 D_GP: 0.111 D_real: 0.491 D_fake: 0.858 +(epoch: 282, iters: 2352, time: 0.064) G_GAN: 0.524 G_GAN_Feat: 0.972 G_ID: 0.149 G_Rec: 0.371 D_GP: 0.021 D_real: 1.132 D_fake: 0.490 +(epoch: 282, iters: 2752, time: 0.064) G_GAN: 0.008 G_GAN_Feat: 0.689 G_ID: 0.125 G_Rec: 0.307 D_GP: 0.028 D_real: 0.902 D_fake: 0.992 +(epoch: 282, iters: 3152, time: 0.064) G_GAN: 0.264 G_GAN_Feat: 0.797 G_ID: 0.171 G_Rec: 0.371 D_GP: 0.021 D_real: 1.099 D_fake: 0.737 +(epoch: 282, iters: 3552, time: 0.064) G_GAN: -0.077 G_GAN_Feat: 0.685 G_ID: 0.135 G_Rec: 0.283 D_GP: 0.035 D_real: 0.853 D_fake: 1.077 +(epoch: 282, iters: 3952, time: 0.064) G_GAN: 0.336 G_GAN_Feat: 1.006 G_ID: 0.165 G_Rec: 0.394 D_GP: 0.050 D_real: 0.737 D_fake: 0.665 +(epoch: 282, iters: 4352, time: 0.064) G_GAN: 0.208 G_GAN_Feat: 0.635 G_ID: 0.148 G_Rec: 0.289 D_GP: 0.022 D_real: 1.152 D_fake: 0.792 +(epoch: 282, iters: 4752, time: 0.064) G_GAN: 0.352 G_GAN_Feat: 0.835 G_ID: 0.163 G_Rec: 0.351 D_GP: 0.030 D_real: 1.164 D_fake: 0.649 +(epoch: 282, iters: 5152, time: 0.064) G_GAN: 0.173 G_GAN_Feat: 0.740 G_ID: 0.116 G_Rec: 0.287 D_GP: 0.033 D_real: 0.920 D_fake: 0.827 +(epoch: 282, iters: 5552, time: 0.064) G_GAN: 0.418 G_GAN_Feat: 0.941 G_ID: 0.194 G_Rec: 0.371 D_GP: 0.068 D_real: 1.108 D_fake: 0.584 +(epoch: 282, iters: 5952, time: 0.064) G_GAN: 0.373 G_GAN_Feat: 0.774 G_ID: 0.131 G_Rec: 0.294 D_GP: 0.045 D_real: 1.189 D_fake: 0.632 +(epoch: 282, iters: 6352, time: 0.064) G_GAN: 0.569 G_GAN_Feat: 1.026 G_ID: 0.165 G_Rec: 0.397 D_GP: 0.036 D_real: 0.847 D_fake: 0.441 +(epoch: 282, iters: 6752, time: 0.064) G_GAN: 0.147 G_GAN_Feat: 0.671 G_ID: 0.131 G_Rec: 0.319 D_GP: 0.021 D_real: 1.026 D_fake: 0.854 +(epoch: 282, iters: 7152, time: 0.064) G_GAN: 0.149 G_GAN_Feat: 0.826 G_ID: 0.179 G_Rec: 0.397 D_GP: 0.029 D_real: 0.849 D_fake: 0.851 +(epoch: 282, iters: 7552, time: 0.064) G_GAN: -0.072 G_GAN_Feat: 0.663 G_ID: 0.114 G_Rec: 0.304 D_GP: 0.054 D_real: 0.810 D_fake: 1.072 +(epoch: 282, iters: 7952, time: 0.064) G_GAN: 0.277 G_GAN_Feat: 0.838 G_ID: 0.172 G_Rec: 0.412 D_GP: 0.030 D_real: 1.019 D_fake: 0.724 +(epoch: 282, iters: 8352, time: 0.064) G_GAN: 0.112 G_GAN_Feat: 0.781 G_ID: 0.120 G_Rec: 0.331 D_GP: 0.053 D_real: 0.766 D_fake: 0.888 +(epoch: 283, iters: 144, time: 0.064) G_GAN: 0.398 G_GAN_Feat: 0.925 G_ID: 0.169 G_Rec: 0.399 D_GP: 0.031 D_real: 1.060 D_fake: 0.602 +(epoch: 283, iters: 544, time: 0.064) G_GAN: 0.097 G_GAN_Feat: 0.815 G_ID: 0.148 G_Rec: 0.312 D_GP: 0.046 D_real: 0.758 D_fake: 0.903 +(epoch: 283, iters: 944, time: 0.064) G_GAN: 0.473 G_GAN_Feat: 0.845 G_ID: 0.136 G_Rec: 0.382 D_GP: 0.029 D_real: 1.152 D_fake: 0.531 +(epoch: 283, iters: 1344, time: 0.064) G_GAN: 0.136 G_GAN_Feat: 0.789 G_ID: 0.133 G_Rec: 0.350 D_GP: 0.063 D_real: 0.911 D_fake: 0.866 +(epoch: 283, iters: 1744, time: 0.064) G_GAN: 0.428 G_GAN_Feat: 1.041 G_ID: 0.158 G_Rec: 0.451 D_GP: 0.040 D_real: 0.818 D_fake: 0.579 +(epoch: 283, iters: 2144, time: 0.064) G_GAN: 0.131 G_GAN_Feat: 0.850 G_ID: 0.122 G_Rec: 0.291 D_GP: 0.031 D_real: 0.599 D_fake: 0.870 +(epoch: 283, iters: 2544, time: 0.064) G_GAN: 0.636 G_GAN_Feat: 0.836 G_ID: 0.187 G_Rec: 0.366 D_GP: 0.022 D_real: 1.366 D_fake: 0.406 +(epoch: 283, iters: 2944, time: 0.064) G_GAN: 0.039 G_GAN_Feat: 0.780 G_ID: 0.152 G_Rec: 0.287 D_GP: 0.026 D_real: 0.782 D_fake: 0.961 +(epoch: 283, iters: 3344, time: 0.064) G_GAN: 0.387 G_GAN_Feat: 0.847 G_ID: 0.161 G_Rec: 0.373 D_GP: 0.029 D_real: 1.100 D_fake: 0.615 +(epoch: 283, iters: 3744, time: 0.064) G_GAN: 0.169 G_GAN_Feat: 0.651 G_ID: 0.142 G_Rec: 0.326 D_GP: 0.019 D_real: 1.086 D_fake: 0.831 +(epoch: 283, iters: 4144, time: 0.064) G_GAN: 0.203 G_GAN_Feat: 0.935 G_ID: 0.154 G_Rec: 0.375 D_GP: 0.054 D_real: 0.614 D_fake: 0.801 +(epoch: 283, iters: 4544, time: 0.064) G_GAN: 0.484 G_GAN_Feat: 0.830 G_ID: 0.150 G_Rec: 0.320 D_GP: 0.039 D_real: 1.029 D_fake: 0.592 +(epoch: 283, iters: 4944, time: 0.064) G_GAN: 0.379 G_GAN_Feat: 0.813 G_ID: 0.178 G_Rec: 0.385 D_GP: 0.021 D_real: 1.201 D_fake: 0.637 +(epoch: 283, iters: 5344, time: 0.064) G_GAN: 0.342 G_GAN_Feat: 0.572 G_ID: 0.111 G_Rec: 0.288 D_GP: 0.018 D_real: 1.320 D_fake: 0.658 +(epoch: 283, iters: 5744, time: 0.064) G_GAN: 0.397 G_GAN_Feat: 0.851 G_ID: 0.166 G_Rec: 0.378 D_GP: 0.021 D_real: 1.217 D_fake: 0.605 +(epoch: 283, iters: 6144, time: 0.064) G_GAN: 0.089 G_GAN_Feat: 0.696 G_ID: 0.110 G_Rec: 0.314 D_GP: 0.027 D_real: 0.909 D_fake: 0.911 +(epoch: 283, iters: 6544, time: 0.064) G_GAN: 0.341 G_GAN_Feat: 0.878 G_ID: 0.184 G_Rec: 0.409 D_GP: 0.024 D_real: 1.097 D_fake: 0.663 +(epoch: 283, iters: 6944, time: 0.064) G_GAN: 0.071 G_GAN_Feat: 0.627 G_ID: 0.165 G_Rec: 0.306 D_GP: 0.019 D_real: 1.034 D_fake: 0.929 +(epoch: 283, iters: 7344, time: 0.064) G_GAN: 0.351 G_GAN_Feat: 0.820 G_ID: 0.160 G_Rec: 0.388 D_GP: 0.024 D_real: 1.075 D_fake: 0.651 +(epoch: 283, iters: 7744, time: 0.064) G_GAN: 0.165 G_GAN_Feat: 0.682 G_ID: 0.134 G_Rec: 0.309 D_GP: 0.030 D_real: 1.040 D_fake: 0.835 +(epoch: 283, iters: 8144, time: 0.064) G_GAN: 0.427 G_GAN_Feat: 0.900 G_ID: 0.160 G_Rec: 0.391 D_GP: 0.028 D_real: 1.111 D_fake: 0.578 +(epoch: 283, iters: 8544, time: 0.064) G_GAN: 0.051 G_GAN_Feat: 0.783 G_ID: 0.125 G_Rec: 0.296 D_GP: 0.036 D_real: 0.845 D_fake: 0.952 +(epoch: 284, iters: 336, time: 0.064) G_GAN: 0.506 G_GAN_Feat: 0.843 G_ID: 0.148 G_Rec: 0.376 D_GP: 0.025 D_real: 1.207 D_fake: 0.500 +(epoch: 284, iters: 736, time: 0.064) G_GAN: 0.229 G_GAN_Feat: 0.659 G_ID: 0.127 G_Rec: 0.272 D_GP: 0.020 D_real: 1.094 D_fake: 0.771 +(epoch: 284, iters: 1136, time: 0.064) G_GAN: 0.372 G_GAN_Feat: 0.895 G_ID: 0.169 G_Rec: 0.352 D_GP: 0.028 D_real: 1.216 D_fake: 0.632 +(epoch: 284, iters: 1536, time: 0.064) G_GAN: 0.232 G_GAN_Feat: 0.681 G_ID: 0.118 G_Rec: 0.281 D_GP: 0.022 D_real: 1.152 D_fake: 0.768 +(epoch: 284, iters: 1936, time: 0.064) G_GAN: 0.528 G_GAN_Feat: 0.818 G_ID: 0.153 G_Rec: 0.369 D_GP: 0.022 D_real: 1.327 D_fake: 0.484 +(epoch: 284, iters: 2336, time: 0.064) G_GAN: 0.154 G_GAN_Feat: 0.585 G_ID: 0.121 G_Rec: 0.279 D_GP: 0.016 D_real: 1.087 D_fake: 0.846 +(epoch: 284, iters: 2736, time: 0.064) G_GAN: 0.290 G_GAN_Feat: 0.804 G_ID: 0.148 G_Rec: 0.374 D_GP: 0.025 D_real: 1.014 D_fake: 0.711 +(epoch: 284, iters: 3136, time: 0.064) G_GAN: 0.050 G_GAN_Feat: 0.765 G_ID: 0.142 G_Rec: 0.305 D_GP: 0.038 D_real: 0.622 D_fake: 0.950 +(epoch: 284, iters: 3536, time: 0.064) G_GAN: 0.313 G_GAN_Feat: 0.886 G_ID: 0.156 G_Rec: 0.389 D_GP: 0.033 D_real: 0.929 D_fake: 0.687 +(epoch: 284, iters: 3936, time: 0.064) G_GAN: 0.217 G_GAN_Feat: 0.692 G_ID: 0.107 G_Rec: 0.266 D_GP: 0.024 D_real: 1.089 D_fake: 0.784 +(epoch: 284, iters: 4336, time: 0.064) G_GAN: 0.215 G_GAN_Feat: 0.856 G_ID: 0.164 G_Rec: 0.367 D_GP: 0.028 D_real: 0.990 D_fake: 0.786 +(epoch: 284, iters: 4736, time: 0.064) G_GAN: 0.094 G_GAN_Feat: 0.872 G_ID: 0.130 G_Rec: 0.353 D_GP: 0.154 D_real: 0.446 D_fake: 0.906 +(epoch: 284, iters: 5136, time: 0.064) G_GAN: 0.548 G_GAN_Feat: 0.975 G_ID: 0.144 G_Rec: 0.396 D_GP: 0.026 D_real: 1.228 D_fake: 0.470 +(epoch: 284, iters: 5536, time: 0.064) G_GAN: 0.202 G_GAN_Feat: 0.905 G_ID: 0.176 G_Rec: 0.378 D_GP: 0.030 D_real: 0.699 D_fake: 0.806 +(epoch: 284, iters: 5936, time: 0.064) G_GAN: 0.249 G_GAN_Feat: 0.921 G_ID: 0.151 G_Rec: 0.397 D_GP: 0.030 D_real: 0.956 D_fake: 0.752 +(epoch: 284, iters: 6336, time: 0.064) G_GAN: 0.226 G_GAN_Feat: 0.715 G_ID: 0.104 G_Rec: 0.270 D_GP: 0.029 D_real: 1.010 D_fake: 0.775 +(epoch: 284, iters: 6736, time: 0.064) G_GAN: 0.705 G_GAN_Feat: 0.962 G_ID: 0.161 G_Rec: 0.454 D_GP: 0.032 D_real: 1.379 D_fake: 0.400 +(epoch: 284, iters: 7136, time: 0.064) G_GAN: 0.361 G_GAN_Feat: 0.643 G_ID: 0.117 G_Rec: 0.254 D_GP: 0.025 D_real: 1.325 D_fake: 0.641 +(epoch: 284, iters: 7536, time: 0.064) G_GAN: 0.411 G_GAN_Feat: 0.940 G_ID: 0.150 G_Rec: 0.406 D_GP: 0.049 D_real: 1.188 D_fake: 0.595 +(epoch: 284, iters: 7936, time: 0.064) G_GAN: 0.167 G_GAN_Feat: 0.657 G_ID: 0.126 G_Rec: 0.287 D_GP: 0.021 D_real: 1.076 D_fake: 0.833 +(epoch: 284, iters: 8336, time: 0.064) G_GAN: 0.486 G_GAN_Feat: 0.891 G_ID: 0.147 G_Rec: 0.402 D_GP: 0.044 D_real: 1.131 D_fake: 0.524 +(epoch: 285, iters: 128, time: 0.064) G_GAN: 0.416 G_GAN_Feat: 0.786 G_ID: 0.138 G_Rec: 0.292 D_GP: 0.034 D_real: 1.082 D_fake: 0.591 +(epoch: 285, iters: 528, time: 0.064) G_GAN: 0.503 G_GAN_Feat: 0.905 G_ID: 0.161 G_Rec: 0.352 D_GP: 0.033 D_real: 1.030 D_fake: 0.501 +(epoch: 285, iters: 928, time: 0.064) G_GAN: 0.360 G_GAN_Feat: 0.856 G_ID: 0.116 G_Rec: 0.322 D_GP: 0.038 D_real: 0.746 D_fake: 0.642 +(epoch: 285, iters: 1328, time: 0.064) G_GAN: 0.421 G_GAN_Feat: 0.742 G_ID: 0.159 G_Rec: 0.474 D_GP: 0.019 D_real: 1.232 D_fake: 0.590 +(epoch: 285, iters: 1728, time: 0.064) G_GAN: -0.022 G_GAN_Feat: 0.610 G_ID: 0.142 G_Rec: 0.282 D_GP: 0.020 D_real: 0.917 D_fake: 1.022 +(epoch: 285, iters: 2128, time: 0.064) G_GAN: 0.447 G_GAN_Feat: 0.806 G_ID: 0.147 G_Rec: 0.361 D_GP: 0.025 D_real: 1.203 D_fake: 0.565 +(epoch: 285, iters: 2528, time: 0.064) G_GAN: -0.081 G_GAN_Feat: 0.669 G_ID: 0.127 G_Rec: 0.298 D_GP: 0.023 D_real: 0.833 D_fake: 1.081 +(epoch: 285, iters: 2928, time: 0.064) G_GAN: 0.397 G_GAN_Feat: 0.834 G_ID: 0.155 G_Rec: 0.379 D_GP: 0.026 D_real: 1.107 D_fake: 0.615 +(epoch: 285, iters: 3328, time: 0.064) G_GAN: 0.028 G_GAN_Feat: 0.714 G_ID: 0.117 G_Rec: 0.297 D_GP: 0.037 D_real: 0.766 D_fake: 0.972 +(epoch: 285, iters: 3728, time: 0.064) G_GAN: 0.446 G_GAN_Feat: 1.061 G_ID: 0.163 G_Rec: 0.439 D_GP: 0.199 D_real: 0.637 D_fake: 0.566 +(epoch: 285, iters: 4128, time: 0.064) G_GAN: 0.089 G_GAN_Feat: 0.717 G_ID: 0.106 G_Rec: 0.307 D_GP: 0.022 D_real: 0.927 D_fake: 0.911 +(epoch: 285, iters: 4528, time: 0.064) G_GAN: 0.586 G_GAN_Feat: 0.910 G_ID: 0.179 G_Rec: 0.407 D_GP: 0.033 D_real: 1.094 D_fake: 0.456 +(epoch: 285, iters: 4928, time: 0.064) G_GAN: 0.306 G_GAN_Feat: 0.640 G_ID: 0.124 G_Rec: 0.311 D_GP: 0.019 D_real: 1.310 D_fake: 0.696 +(epoch: 285, iters: 5328, time: 0.064) G_GAN: 0.395 G_GAN_Feat: 0.814 G_ID: 0.174 G_Rec: 0.363 D_GP: 0.024 D_real: 1.216 D_fake: 0.605 +(epoch: 285, iters: 5728, time: 0.064) G_GAN: 0.213 G_GAN_Feat: 0.734 G_ID: 0.135 G_Rec: 0.296 D_GP: 0.048 D_real: 0.947 D_fake: 0.788 +(epoch: 285, iters: 6128, time: 0.064) G_GAN: 0.389 G_GAN_Feat: 0.919 G_ID: 0.160 G_Rec: 0.411 D_GP: 0.031 D_real: 0.907 D_fake: 0.612 +(epoch: 285, iters: 6528, time: 0.064) G_GAN: -0.023 G_GAN_Feat: 0.705 G_ID: 0.110 G_Rec: 0.288 D_GP: 0.023 D_real: 0.835 D_fake: 1.023 +(epoch: 285, iters: 6928, time: 0.064) G_GAN: 0.697 G_GAN_Feat: 0.866 G_ID: 0.156 G_Rec: 0.380 D_GP: 0.024 D_real: 1.479 D_fake: 0.318 +(epoch: 285, iters: 7328, time: 0.064) G_GAN: -0.066 G_GAN_Feat: 0.614 G_ID: 0.149 G_Rec: 0.294 D_GP: 0.025 D_real: 1.015 D_fake: 1.066 +(epoch: 285, iters: 7728, time: 0.064) G_GAN: 0.179 G_GAN_Feat: 0.753 G_ID: 0.154 G_Rec: 0.397 D_GP: 0.033 D_real: 0.968 D_fake: 0.821 +(epoch: 285, iters: 8128, time: 0.064) G_GAN: 0.103 G_GAN_Feat: 0.604 G_ID: 0.125 G_Rec: 0.289 D_GP: 0.028 D_real: 1.009 D_fake: 0.897 +(epoch: 285, iters: 8528, time: 0.064) G_GAN: 0.411 G_GAN_Feat: 0.806 G_ID: 0.169 G_Rec: 0.382 D_GP: 0.029 D_real: 1.196 D_fake: 0.599 +(epoch: 286, iters: 320, time: 0.064) G_GAN: 0.043 G_GAN_Feat: 0.733 G_ID: 0.114 G_Rec: 0.316 D_GP: 0.288 D_real: 0.617 D_fake: 0.958 +(epoch: 286, iters: 720, time: 0.064) G_GAN: 0.331 G_GAN_Feat: 0.851 G_ID: 0.153 G_Rec: 0.416 D_GP: 0.025 D_real: 0.967 D_fake: 0.670 +(epoch: 286, iters: 1120, time: 0.064) G_GAN: 0.235 G_GAN_Feat: 0.631 G_ID: 0.138 G_Rec: 0.284 D_GP: 0.019 D_real: 1.163 D_fake: 0.765 +(epoch: 286, iters: 1520, time: 0.064) G_GAN: 0.375 G_GAN_Feat: 0.790 G_ID: 0.168 G_Rec: 0.344 D_GP: 0.024 D_real: 1.114 D_fake: 0.628 +(epoch: 286, iters: 1920, time: 0.064) G_GAN: 0.428 G_GAN_Feat: 0.755 G_ID: 0.147 G_Rec: 0.337 D_GP: 0.022 D_real: 1.340 D_fake: 0.587 +(epoch: 286, iters: 2320, time: 0.064) G_GAN: 0.334 G_GAN_Feat: 0.642 G_ID: 0.146 G_Rec: 0.320 D_GP: 0.017 D_real: 1.143 D_fake: 0.666 +(epoch: 286, iters: 2720, time: 0.064) G_GAN: 0.184 G_GAN_Feat: 0.573 G_ID: 0.135 G_Rec: 0.288 D_GP: 0.018 D_real: 1.109 D_fake: 0.816 +(epoch: 286, iters: 3120, time: 0.064) G_GAN: 0.351 G_GAN_Feat: 0.709 G_ID: 0.153 G_Rec: 0.345 D_GP: 0.021 D_real: 1.091 D_fake: 0.650 +(epoch: 286, iters: 3520, time: 0.064) G_GAN: 0.302 G_GAN_Feat: 0.649 G_ID: 0.120 G_Rec: 0.314 D_GP: 0.034 D_real: 1.160 D_fake: 0.700 +(epoch: 286, iters: 3920, time: 0.064) G_GAN: 0.235 G_GAN_Feat: 0.765 G_ID: 0.158 G_Rec: 0.371 D_GP: 0.041 D_real: 1.029 D_fake: 0.765 +(epoch: 286, iters: 4320, time: 0.064) G_GAN: 0.061 G_GAN_Feat: 0.622 G_ID: 0.133 G_Rec: 0.302 D_GP: 0.029 D_real: 0.917 D_fake: 0.939 +(epoch: 286, iters: 4720, time: 0.064) G_GAN: 0.471 G_GAN_Feat: 0.891 G_ID: 0.162 G_Rec: 0.386 D_GP: 0.072 D_real: 1.123 D_fake: 0.532 +(epoch: 286, iters: 5120, time: 0.064) G_GAN: 0.083 G_GAN_Feat: 0.685 G_ID: 0.116 G_Rec: 0.290 D_GP: 0.054 D_real: 0.887 D_fake: 0.917 +(epoch: 286, iters: 5520, time: 0.064) G_GAN: 0.520 G_GAN_Feat: 0.881 G_ID: 0.161 G_Rec: 0.403 D_GP: 0.022 D_real: 1.201 D_fake: 0.487 +(epoch: 286, iters: 5920, time: 0.064) G_GAN: 0.129 G_GAN_Feat: 0.699 G_ID: 0.149 G_Rec: 0.304 D_GP: 0.082 D_real: 0.851 D_fake: 0.871 +(epoch: 286, iters: 6320, time: 0.064) G_GAN: 0.464 G_GAN_Feat: 0.888 G_ID: 0.155 G_Rec: 0.393 D_GP: 0.040 D_real: 1.004 D_fake: 0.544 +(epoch: 286, iters: 6720, time: 0.064) G_GAN: 0.259 G_GAN_Feat: 0.650 G_ID: 0.116 G_Rec: 0.295 D_GP: 0.020 D_real: 1.207 D_fake: 0.741 +(epoch: 286, iters: 7120, time: 0.064) G_GAN: -0.019 G_GAN_Feat: 0.840 G_ID: 0.161 G_Rec: 0.412 D_GP: 0.036 D_real: 0.804 D_fake: 1.019 +(epoch: 286, iters: 7520, time: 0.064) G_GAN: 0.082 G_GAN_Feat: 0.636 G_ID: 0.148 G_Rec: 0.271 D_GP: 0.025 D_real: 0.991 D_fake: 0.918 +(epoch: 286, iters: 7920, time: 0.064) G_GAN: 0.395 G_GAN_Feat: 0.772 G_ID: 0.173 G_Rec: 0.340 D_GP: 0.022 D_real: 1.193 D_fake: 0.607 +(epoch: 286, iters: 8320, time: 0.064) G_GAN: 0.116 G_GAN_Feat: 0.729 G_ID: 0.120 G_Rec: 0.290 D_GP: 0.030 D_real: 1.037 D_fake: 0.885 +(epoch: 287, iters: 112, time: 0.064) G_GAN: 0.706 G_GAN_Feat: 0.952 G_ID: 0.117 G_Rec: 0.402 D_GP: 0.032 D_real: 1.213 D_fake: 0.310 +(epoch: 287, iters: 512, time: 0.064) G_GAN: 0.122 G_GAN_Feat: 0.666 G_ID: 0.115 G_Rec: 0.290 D_GP: 0.018 D_real: 1.076 D_fake: 0.878 +(epoch: 287, iters: 912, time: 0.064) G_GAN: 0.342 G_GAN_Feat: 0.787 G_ID: 0.141 G_Rec: 0.360 D_GP: 0.022 D_real: 1.035 D_fake: 0.664 +(epoch: 287, iters: 1312, time: 0.064) G_GAN: 0.353 G_GAN_Feat: 0.891 G_ID: 0.130 G_Rec: 0.319 D_GP: 0.028 D_real: 1.391 D_fake: 0.676 +(epoch: 287, iters: 1712, time: 0.064) G_GAN: 0.566 G_GAN_Feat: 0.898 G_ID: 0.149 G_Rec: 0.387 D_GP: 0.022 D_real: 1.277 D_fake: 0.458 +(epoch: 287, iters: 2112, time: 0.064) G_GAN: 0.275 G_GAN_Feat: 0.691 G_ID: 0.117 G_Rec: 0.308 D_GP: 0.024 D_real: 1.151 D_fake: 0.726 +(epoch: 287, iters: 2512, time: 0.064) G_GAN: 0.640 G_GAN_Feat: 0.993 G_ID: 0.164 G_Rec: 0.397 D_GP: 0.040 D_real: 1.085 D_fake: 0.451 +(epoch: 287, iters: 2912, time: 0.064) G_GAN: 0.168 G_GAN_Feat: 0.676 G_ID: 0.118 G_Rec: 0.316 D_GP: 0.018 D_real: 1.081 D_fake: 0.833 +(epoch: 287, iters: 3312, time: 0.064) G_GAN: 0.544 G_GAN_Feat: 0.838 G_ID: 0.159 G_Rec: 0.384 D_GP: 0.020 D_real: 1.274 D_fake: 0.464 +(epoch: 287, iters: 3712, time: 0.064) G_GAN: 0.011 G_GAN_Feat: 0.761 G_ID: 0.138 G_Rec: 0.378 D_GP: 0.077 D_real: 0.827 D_fake: 0.989 +(epoch: 287, iters: 4112, time: 0.064) G_GAN: 0.778 G_GAN_Feat: 0.796 G_ID: 0.165 G_Rec: 0.376 D_GP: 0.021 D_real: 1.519 D_fake: 0.261 +(epoch: 287, iters: 4512, time: 0.064) G_GAN: 0.413 G_GAN_Feat: 0.629 G_ID: 0.128 G_Rec: 0.269 D_GP: 0.023 D_real: 1.353 D_fake: 0.589 +(epoch: 287, iters: 4912, time: 0.064) G_GAN: 0.430 G_GAN_Feat: 0.763 G_ID: 0.155 G_Rec: 0.350 D_GP: 0.023 D_real: 1.231 D_fake: 0.573 +(epoch: 287, iters: 5312, time: 0.064) G_GAN: 0.106 G_GAN_Feat: 0.720 G_ID: 0.124 G_Rec: 0.302 D_GP: 0.039 D_real: 0.870 D_fake: 0.894 +(epoch: 287, iters: 5712, time: 0.064) G_GAN: 0.169 G_GAN_Feat: 1.019 G_ID: 0.183 G_Rec: 0.437 D_GP: 0.142 D_real: 0.496 D_fake: 0.835 +(epoch: 287, iters: 6112, time: 0.064) G_GAN: 0.156 G_GAN_Feat: 0.815 G_ID: 0.121 G_Rec: 0.324 D_GP: 0.099 D_real: 0.780 D_fake: 0.846 +(epoch: 287, iters: 6512, time: 0.064) G_GAN: 0.318 G_GAN_Feat: 0.939 G_ID: 0.164 G_Rec: 0.404 D_GP: 0.053 D_real: 0.624 D_fake: 0.684 +(epoch: 287, iters: 6912, time: 0.064) G_GAN: 0.284 G_GAN_Feat: 0.686 G_ID: 0.115 G_Rec: 0.302 D_GP: 0.016 D_real: 1.234 D_fake: 0.716 +(epoch: 287, iters: 7312, time: 0.064) G_GAN: 0.472 G_GAN_Feat: 0.956 G_ID: 0.165 G_Rec: 0.446 D_GP: 0.054 D_real: 1.078 D_fake: 0.537 +(epoch: 287, iters: 7712, time: 0.064) G_GAN: 0.059 G_GAN_Feat: 0.734 G_ID: 0.122 G_Rec: 0.319 D_GP: 0.044 D_real: 0.903 D_fake: 0.941 +(epoch: 287, iters: 8112, time: 0.064) G_GAN: 0.201 G_GAN_Feat: 0.882 G_ID: 0.178 G_Rec: 0.436 D_GP: 0.024 D_real: 1.082 D_fake: 0.799 +(epoch: 287, iters: 8512, time: 0.064) G_GAN: 0.026 G_GAN_Feat: 0.701 G_ID: 0.164 G_Rec: 0.274 D_GP: 0.040 D_real: 0.820 D_fake: 0.974 +(epoch: 288, iters: 304, time: 0.064) G_GAN: 0.520 G_GAN_Feat: 1.012 G_ID: 0.143 G_Rec: 0.482 D_GP: 0.054 D_real: 0.883 D_fake: 0.488 +(epoch: 288, iters: 704, time: 0.064) G_GAN: 0.256 G_GAN_Feat: 0.658 G_ID: 0.114 G_Rec: 0.269 D_GP: 0.027 D_real: 1.147 D_fake: 0.744 +(epoch: 288, iters: 1104, time: 0.064) G_GAN: 0.203 G_GAN_Feat: 0.947 G_ID: 0.142 G_Rec: 0.422 D_GP: 0.038 D_real: 0.610 D_fake: 0.799 +(epoch: 288, iters: 1504, time: 0.064) G_GAN: 0.056 G_GAN_Feat: 0.598 G_ID: 0.151 G_Rec: 0.299 D_GP: 0.022 D_real: 1.030 D_fake: 0.944 +(epoch: 288, iters: 1904, time: 0.064) G_GAN: 0.310 G_GAN_Feat: 0.837 G_ID: 0.158 G_Rec: 0.402 D_GP: 0.027 D_real: 1.042 D_fake: 0.691 +(epoch: 288, iters: 2304, time: 0.064) G_GAN: 0.195 G_GAN_Feat: 0.703 G_ID: 0.146 G_Rec: 0.332 D_GP: 0.051 D_real: 1.002 D_fake: 0.805 +(epoch: 288, iters: 2704, time: 0.064) G_GAN: 0.483 G_GAN_Feat: 0.937 G_ID: 0.149 G_Rec: 0.431 D_GP: 0.050 D_real: 1.027 D_fake: 0.543 +(epoch: 288, iters: 3104, time: 0.064) G_GAN: 0.005 G_GAN_Feat: 0.804 G_ID: 0.119 G_Rec: 0.333 D_GP: 0.198 D_real: 0.625 D_fake: 0.995 +(epoch: 288, iters: 3504, time: 0.064) G_GAN: 0.395 G_GAN_Feat: 0.985 G_ID: 0.156 G_Rec: 0.444 D_GP: 0.061 D_real: 0.870 D_fake: 0.606 +(epoch: 288, iters: 3904, time: 0.064) G_GAN: 0.298 G_GAN_Feat: 0.657 G_ID: 0.124 G_Rec: 0.297 D_GP: 0.031 D_real: 1.121 D_fake: 0.704 +(epoch: 288, iters: 4304, time: 0.064) G_GAN: 0.472 G_GAN_Feat: 0.921 G_ID: 0.155 G_Rec: 0.377 D_GP: 0.032 D_real: 1.022 D_fake: 0.535 +(epoch: 288, iters: 4704, time: 0.064) G_GAN: 0.316 G_GAN_Feat: 0.703 G_ID: 0.136 G_Rec: 0.295 D_GP: 0.039 D_real: 1.000 D_fake: 0.685 +(epoch: 288, iters: 5104, time: 0.064) G_GAN: 0.322 G_GAN_Feat: 0.819 G_ID: 0.165 G_Rec: 0.371 D_GP: 0.034 D_real: 1.008 D_fake: 0.678 +(epoch: 288, iters: 5504, time: 0.064) G_GAN: 0.220 G_GAN_Feat: 0.740 G_ID: 0.148 G_Rec: 0.369 D_GP: 0.035 D_real: 0.856 D_fake: 0.781 +(epoch: 288, iters: 5904, time: 0.064) G_GAN: 0.594 G_GAN_Feat: 0.876 G_ID: 0.195 G_Rec: 0.389 D_GP: 0.023 D_real: 1.315 D_fake: 0.425 +(epoch: 288, iters: 6304, time: 0.064) G_GAN: 0.215 G_GAN_Feat: 0.666 G_ID: 0.130 G_Rec: 0.290 D_GP: 0.031 D_real: 1.092 D_fake: 0.785 +(epoch: 288, iters: 6704, time: 0.064) G_GAN: 0.482 G_GAN_Feat: 0.848 G_ID: 0.158 G_Rec: 0.365 D_GP: 0.030 D_real: 1.071 D_fake: 0.520 +(epoch: 288, iters: 7104, time: 0.064) G_GAN: 0.303 G_GAN_Feat: 0.869 G_ID: 0.134 G_Rec: 0.303 D_GP: 0.631 D_real: 0.472 D_fake: 0.707 +(epoch: 288, iters: 7504, time: 0.064) G_GAN: 0.383 G_GAN_Feat: 1.075 G_ID: 0.181 G_Rec: 0.457 D_GP: 0.277 D_real: 0.454 D_fake: 0.642 +(epoch: 288, iters: 7904, time: 0.064) G_GAN: 0.276 G_GAN_Feat: 0.791 G_ID: 0.142 G_Rec: 0.330 D_GP: 0.027 D_real: 1.010 D_fake: 0.753 +(epoch: 288, iters: 8304, time: 0.064) G_GAN: 0.307 G_GAN_Feat: 0.982 G_ID: 0.157 G_Rec: 0.434 D_GP: 0.024 D_real: 0.927 D_fake: 0.694 +(epoch: 289, iters: 96, time: 0.064) G_GAN: -0.114 G_GAN_Feat: 0.674 G_ID: 0.142 G_Rec: 0.281 D_GP: 0.031 D_real: 0.832 D_fake: 1.114 +(epoch: 289, iters: 496, time: 0.064) G_GAN: 0.437 G_GAN_Feat: 1.056 G_ID: 0.172 G_Rec: 0.433 D_GP: 0.034 D_real: 1.343 D_fake: 0.586 +(epoch: 289, iters: 896, time: 0.064) G_GAN: -0.011 G_GAN_Feat: 0.640 G_ID: 0.131 G_Rec: 0.282 D_GP: 0.032 D_real: 0.942 D_fake: 1.011 +(epoch: 289, iters: 1296, time: 0.064) G_GAN: 0.307 G_GAN_Feat: 0.780 G_ID: 0.173 G_Rec: 0.364 D_GP: 0.021 D_real: 1.103 D_fake: 0.693 +(epoch: 289, iters: 1696, time: 0.064) G_GAN: 0.144 G_GAN_Feat: 0.561 G_ID: 0.103 G_Rec: 0.282 D_GP: 0.021 D_real: 1.079 D_fake: 0.856 +(epoch: 289, iters: 2096, time: 0.064) G_GAN: 0.272 G_GAN_Feat: 0.737 G_ID: 0.173 G_Rec: 0.372 D_GP: 0.022 D_real: 1.067 D_fake: 0.729 +(epoch: 289, iters: 2496, time: 0.064) G_GAN: 0.123 G_GAN_Feat: 0.595 G_ID: 0.128 G_Rec: 0.274 D_GP: 0.021 D_real: 1.055 D_fake: 0.877 +(epoch: 289, iters: 2896, time: 0.064) G_GAN: 0.376 G_GAN_Feat: 0.757 G_ID: 0.170 G_Rec: 0.398 D_GP: 0.025 D_real: 1.126 D_fake: 0.626 +(epoch: 289, iters: 3296, time: 0.064) G_GAN: 0.196 G_GAN_Feat: 0.733 G_ID: 0.126 G_Rec: 0.308 D_GP: 0.035 D_real: 1.083 D_fake: 0.805 +(epoch: 289, iters: 3696, time: 0.064) G_GAN: 0.464 G_GAN_Feat: 0.851 G_ID: 0.150 G_Rec: 0.392 D_GP: 0.022 D_real: 1.197 D_fake: 0.545 +(epoch: 289, iters: 4096, time: 0.064) G_GAN: 0.113 G_GAN_Feat: 0.923 G_ID: 0.127 G_Rec: 0.318 D_GP: 0.084 D_real: 0.898 D_fake: 0.893 +(epoch: 289, iters: 4496, time: 0.064) G_GAN: 0.819 G_GAN_Feat: 0.880 G_ID: 0.158 G_Rec: 0.353 D_GP: 0.049 D_real: 1.361 D_fake: 0.252 +(epoch: 289, iters: 4896, time: 0.064) G_GAN: 0.297 G_GAN_Feat: 0.857 G_ID: 0.128 G_Rec: 0.304 D_GP: 0.092 D_real: 0.628 D_fake: 0.704 +(epoch: 289, iters: 5296, time: 0.064) G_GAN: 0.659 G_GAN_Feat: 0.862 G_ID: 0.137 G_Rec: 0.392 D_GP: 0.022 D_real: 1.377 D_fake: 0.357 +(epoch: 289, iters: 5696, time: 0.064) G_GAN: 0.115 G_GAN_Feat: 0.614 G_ID: 0.115 G_Rec: 0.264 D_GP: 0.020 D_real: 1.076 D_fake: 0.885 +(epoch: 289, iters: 6096, time: 0.064) G_GAN: 0.332 G_GAN_Feat: 0.845 G_ID: 0.158 G_Rec: 0.402 D_GP: 0.032 D_real: 1.067 D_fake: 0.669 +(epoch: 289, iters: 6496, time: 0.064) G_GAN: 0.226 G_GAN_Feat: 0.667 G_ID: 0.122 G_Rec: 0.276 D_GP: 0.026 D_real: 1.102 D_fake: 0.774 +(epoch: 289, iters: 6896, time: 0.064) G_GAN: 0.250 G_GAN_Feat: 0.911 G_ID: 0.177 G_Rec: 0.373 D_GP: 0.036 D_real: 0.836 D_fake: 0.751 +(epoch: 289, iters: 7296, time: 0.064) G_GAN: 0.300 G_GAN_Feat: 0.744 G_ID: 0.134 G_Rec: 0.280 D_GP: 0.029 D_real: 1.198 D_fake: 0.703 +(epoch: 289, iters: 7696, time: 0.064) G_GAN: 0.637 G_GAN_Feat: 0.970 G_ID: 0.198 G_Rec: 0.400 D_GP: 0.038 D_real: 1.201 D_fake: 0.424 +(epoch: 289, iters: 8096, time: 0.064) G_GAN: -0.038 G_GAN_Feat: 0.851 G_ID: 0.119 G_Rec: 0.318 D_GP: 0.038 D_real: 0.604 D_fake: 1.038 +(epoch: 289, iters: 8496, time: 0.064) G_GAN: 0.436 G_GAN_Feat: 0.986 G_ID: 0.175 G_Rec: 0.383 D_GP: 0.080 D_real: 0.623 D_fake: 0.567 +(epoch: 290, iters: 288, time: 0.064) G_GAN: 0.107 G_GAN_Feat: 0.648 G_ID: 0.126 G_Rec: 0.302 D_GP: 0.023 D_real: 1.065 D_fake: 0.893 +(epoch: 290, iters: 688, time: 0.064) G_GAN: 0.356 G_GAN_Feat: 0.999 G_ID: 0.145 G_Rec: 0.450 D_GP: 0.046 D_real: 0.994 D_fake: 0.648 +(epoch: 290, iters: 1088, time: 0.064) G_GAN: 0.101 G_GAN_Feat: 0.634 G_ID: 0.122 G_Rec: 0.237 D_GP: 0.024 D_real: 1.106 D_fake: 0.899 +(epoch: 290, iters: 1488, time: 0.064) G_GAN: 0.084 G_GAN_Feat: 0.879 G_ID: 0.143 G_Rec: 0.425 D_GP: 0.022 D_real: 0.810 D_fake: 0.918 +(epoch: 290, iters: 1888, time: 0.064) G_GAN: 0.172 G_GAN_Feat: 0.684 G_ID: 0.119 G_Rec: 0.272 D_GP: 0.027 D_real: 1.121 D_fake: 0.828 +(epoch: 290, iters: 2288, time: 0.064) G_GAN: 0.456 G_GAN_Feat: 0.897 G_ID: 0.167 G_Rec: 0.424 D_GP: 0.034 D_real: 1.117 D_fake: 0.546 +(epoch: 290, iters: 2688, time: 0.064) G_GAN: 0.322 G_GAN_Feat: 0.636 G_ID: 0.116 G_Rec: 0.292 D_GP: 0.028 D_real: 1.224 D_fake: 0.681 +(epoch: 290, iters: 3088, time: 0.064) G_GAN: 0.515 G_GAN_Feat: 0.780 G_ID: 0.156 G_Rec: 0.357 D_GP: 0.024 D_real: 1.272 D_fake: 0.495 +(epoch: 290, iters: 3488, time: 0.064) G_GAN: 0.485 G_GAN_Feat: 0.777 G_ID: 0.121 G_Rec: 0.289 D_GP: 0.026 D_real: 1.354 D_fake: 0.546 +(epoch: 290, iters: 3888, time: 0.064) G_GAN: 0.316 G_GAN_Feat: 1.113 G_ID: 0.167 G_Rec: 0.448 D_GP: 0.335 D_real: 0.462 D_fake: 0.689 +(epoch: 290, iters: 4288, time: 0.064) G_GAN: 0.341 G_GAN_Feat: 0.547 G_ID: 0.122 G_Rec: 0.272 D_GP: 0.017 D_real: 1.319 D_fake: 0.659 +(epoch: 290, iters: 4688, time: 0.064) G_GAN: 0.412 G_GAN_Feat: 0.836 G_ID: 0.166 G_Rec: 0.448 D_GP: 0.025 D_real: 1.093 D_fake: 0.589 +(epoch: 290, iters: 5088, time: 0.064) G_GAN: 0.136 G_GAN_Feat: 0.583 G_ID: 0.121 G_Rec: 0.276 D_GP: 0.021 D_real: 1.102 D_fake: 0.864 +(epoch: 290, iters: 5488, time: 0.064) G_GAN: 0.238 G_GAN_Feat: 0.795 G_ID: 0.151 G_Rec: 0.397 D_GP: 0.024 D_real: 1.025 D_fake: 0.762 +(epoch: 290, iters: 5888, time: 0.064) G_GAN: 0.057 G_GAN_Feat: 0.685 G_ID: 0.137 G_Rec: 0.302 D_GP: 0.043 D_real: 0.836 D_fake: 0.943 +(epoch: 290, iters: 6288, time: 0.064) G_GAN: 0.352 G_GAN_Feat: 0.829 G_ID: 0.153 G_Rec: 0.391 D_GP: 0.026 D_real: 1.071 D_fake: 0.648 +(epoch: 290, iters: 6688, time: 0.064) G_GAN: 0.237 G_GAN_Feat: 0.673 G_ID: 0.128 G_Rec: 0.309 D_GP: 0.038 D_real: 1.065 D_fake: 0.765 +(epoch: 290, iters: 7088, time: 0.064) G_GAN: 0.288 G_GAN_Feat: 0.781 G_ID: 0.146 G_Rec: 0.353 D_GP: 0.027 D_real: 1.094 D_fake: 0.714 +(epoch: 290, iters: 7488, time: 0.064) G_GAN: 0.204 G_GAN_Feat: 0.682 G_ID: 0.128 G_Rec: 0.318 D_GP: 0.133 D_real: 0.873 D_fake: 0.797 +(epoch: 290, iters: 7888, time: 0.064) G_GAN: 0.334 G_GAN_Feat: 0.850 G_ID: 0.177 G_Rec: 0.363 D_GP: 0.035 D_real: 1.050 D_fake: 0.671 +(epoch: 290, iters: 8288, time: 0.064) G_GAN: 0.253 G_GAN_Feat: 0.637 G_ID: 0.120 G_Rec: 0.318 D_GP: 0.024 D_real: 1.207 D_fake: 0.747 +(epoch: 291, iters: 80, time: 0.064) G_GAN: 0.317 G_GAN_Feat: 1.008 G_ID: 0.138 G_Rec: 0.430 D_GP: 0.116 D_real: 0.556 D_fake: 0.683 +(epoch: 291, iters: 480, time: 0.064) G_GAN: 0.227 G_GAN_Feat: 0.664 G_ID: 0.097 G_Rec: 0.242 D_GP: 0.026 D_real: 1.169 D_fake: 0.773 +(epoch: 291, iters: 880, time: 0.064) G_GAN: 0.461 G_GAN_Feat: 0.756 G_ID: 0.147 G_Rec: 0.381 D_GP: 0.020 D_real: 1.281 D_fake: 0.541 +(epoch: 291, iters: 1280, time: 0.064) G_GAN: 0.121 G_GAN_Feat: 0.665 G_ID: 0.123 G_Rec: 0.301 D_GP: 0.037 D_real: 1.010 D_fake: 0.879 +(epoch: 291, iters: 1680, time: 0.064) G_GAN: 0.182 G_GAN_Feat: 0.728 G_ID: 0.168 G_Rec: 0.341 D_GP: 0.030 D_real: 0.985 D_fake: 0.818 +(epoch: 291, iters: 2080, time: 0.064) G_GAN: 0.148 G_GAN_Feat: 0.857 G_ID: 0.138 G_Rec: 0.342 D_GP: 0.068 D_real: 0.698 D_fake: 0.852 +(epoch: 291, iters: 2480, time: 0.064) G_GAN: 0.503 G_GAN_Feat: 0.846 G_ID: 0.151 G_Rec: 0.419 D_GP: 0.022 D_real: 1.219 D_fake: 0.500 +(epoch: 291, iters: 2880, time: 0.064) G_GAN: 0.108 G_GAN_Feat: 0.854 G_ID: 0.124 G_Rec: 0.327 D_GP: 0.234 D_real: 0.597 D_fake: 0.892 +(epoch: 291, iters: 3280, time: 0.064) G_GAN: 0.422 G_GAN_Feat: 0.878 G_ID: 0.150 G_Rec: 0.368 D_GP: 0.022 D_real: 1.126 D_fake: 0.581 +(epoch: 291, iters: 3680, time: 0.064) G_GAN: 0.265 G_GAN_Feat: 0.756 G_ID: 0.122 G_Rec: 0.299 D_GP: 0.091 D_real: 0.761 D_fake: 0.736 +(epoch: 291, iters: 4080, time: 0.064) G_GAN: 0.211 G_GAN_Feat: 0.867 G_ID: 0.170 G_Rec: 0.383 D_GP: 0.030 D_real: 1.002 D_fake: 0.790 +(epoch: 291, iters: 4480, time: 0.064) G_GAN: 0.215 G_GAN_Feat: 0.807 G_ID: 0.107 G_Rec: 0.312 D_GP: 0.054 D_real: 0.757 D_fake: 0.785 +(epoch: 291, iters: 4880, time: 0.064) G_GAN: 0.345 G_GAN_Feat: 1.092 G_ID: 0.162 G_Rec: 0.402 D_GP: 0.113 D_real: 0.423 D_fake: 0.660 +(epoch: 291, iters: 5280, time: 0.064) G_GAN: 0.303 G_GAN_Feat: 0.596 G_ID: 0.117 G_Rec: 0.245 D_GP: 0.017 D_real: 1.262 D_fake: 0.699 +(epoch: 291, iters: 5680, time: 0.064) G_GAN: 0.509 G_GAN_Feat: 0.814 G_ID: 0.161 G_Rec: 0.402 D_GP: 0.020 D_real: 1.247 D_fake: 0.504 +(epoch: 291, iters: 6080, time: 0.064) G_GAN: 0.154 G_GAN_Feat: 0.631 G_ID: 0.127 G_Rec: 0.284 D_GP: 0.023 D_real: 1.107 D_fake: 0.846 +(epoch: 291, iters: 6480, time: 0.064) G_GAN: 0.092 G_GAN_Feat: 0.815 G_ID: 0.207 G_Rec: 0.343 D_GP: 0.027 D_real: 0.927 D_fake: 0.908 +(epoch: 291, iters: 6880, time: 0.064) G_GAN: 0.387 G_GAN_Feat: 0.726 G_ID: 0.113 G_Rec: 0.269 D_GP: 0.038 D_real: 1.199 D_fake: 0.632 +(epoch: 291, iters: 7280, time: 0.064) G_GAN: 0.285 G_GAN_Feat: 0.898 G_ID: 0.155 G_Rec: 0.404 D_GP: 0.027 D_real: 0.913 D_fake: 0.715 +(epoch: 291, iters: 7680, time: 0.064) G_GAN: 0.224 G_GAN_Feat: 0.724 G_ID: 0.107 G_Rec: 0.288 D_GP: 0.122 D_real: 0.895 D_fake: 0.776 +(epoch: 291, iters: 8080, time: 0.064) G_GAN: 0.599 G_GAN_Feat: 0.838 G_ID: 0.163 G_Rec: 0.400 D_GP: 0.019 D_real: 1.389 D_fake: 0.405 +(epoch: 291, iters: 8480, time: 0.064) G_GAN: 0.188 G_GAN_Feat: 0.739 G_ID: 0.116 G_Rec: 0.304 D_GP: 0.055 D_real: 1.102 D_fake: 0.812 +(epoch: 292, iters: 272, time: 0.064) G_GAN: 0.580 G_GAN_Feat: 1.096 G_ID: 0.182 G_Rec: 0.414 D_GP: 0.101 D_real: 0.396 D_fake: 0.448 +(epoch: 292, iters: 672, time: 0.064) G_GAN: 0.488 G_GAN_Feat: 0.658 G_ID: 0.117 G_Rec: 0.259 D_GP: 0.021 D_real: 1.439 D_fake: 0.597 +(epoch: 292, iters: 1072, time: 0.064) G_GAN: 0.539 G_GAN_Feat: 0.797 G_ID: 0.182 G_Rec: 0.378 D_GP: 0.020 D_real: 1.375 D_fake: 0.475 +(epoch: 292, iters: 1472, time: 0.064) G_GAN: 0.239 G_GAN_Feat: 0.637 G_ID: 0.106 G_Rec: 0.277 D_GP: 0.019 D_real: 1.174 D_fake: 0.761 +(epoch: 292, iters: 1872, time: 0.064) G_GAN: 0.212 G_GAN_Feat: 0.916 G_ID: 0.170 G_Rec: 0.386 D_GP: 0.117 D_real: 0.739 D_fake: 0.791 +(epoch: 292, iters: 2272, time: 0.064) G_GAN: 0.335 G_GAN_Feat: 0.771 G_ID: 0.138 G_Rec: 0.313 D_GP: 0.047 D_real: 0.963 D_fake: 0.668 +(epoch: 292, iters: 2672, time: 0.064) G_GAN: 0.478 G_GAN_Feat: 0.984 G_ID: 0.149 G_Rec: 0.429 D_GP: 0.039 D_real: 1.132 D_fake: 0.529 +(epoch: 292, iters: 3072, time: 0.064) G_GAN: 0.298 G_GAN_Feat: 0.727 G_ID: 0.161 G_Rec: 0.269 D_GP: 0.024 D_real: 1.223 D_fake: 0.705 +(epoch: 292, iters: 3472, time: 0.064) G_GAN: 0.042 G_GAN_Feat: 0.833 G_ID: 0.161 G_Rec: 0.365 D_GP: 0.022 D_real: 0.943 D_fake: 0.958 +(epoch: 292, iters: 3872, time: 0.064) G_GAN: 0.147 G_GAN_Feat: 0.645 G_ID: 0.116 G_Rec: 0.280 D_GP: 0.023 D_real: 1.059 D_fake: 0.854 +(epoch: 292, iters: 4272, time: 0.064) G_GAN: 0.361 G_GAN_Feat: 0.665 G_ID: 0.148 G_Rec: 0.330 D_GP: 0.019 D_real: 1.281 D_fake: 0.640 +(epoch: 292, iters: 4672, time: 0.064) G_GAN: -0.000 G_GAN_Feat: 0.602 G_ID: 0.133 G_Rec: 0.294 D_GP: 0.024 D_real: 0.921 D_fake: 1.000 +(epoch: 292, iters: 5072, time: 0.064) G_GAN: 0.188 G_GAN_Feat: 0.753 G_ID: 0.165 G_Rec: 0.353 D_GP: 0.027 D_real: 0.981 D_fake: 0.812 +(epoch: 292, iters: 5472, time: 0.064) G_GAN: 0.080 G_GAN_Feat: 0.636 G_ID: 0.123 G_Rec: 0.296 D_GP: 0.027 D_real: 1.046 D_fake: 0.920 +(epoch: 292, iters: 5872, time: 0.064) G_GAN: 0.169 G_GAN_Feat: 0.837 G_ID: 0.177 G_Rec: 0.390 D_GP: 0.023 D_real: 0.813 D_fake: 0.832 +(epoch: 292, iters: 6272, time: 0.064) G_GAN: 0.121 G_GAN_Feat: 0.706 G_ID: 0.122 G_Rec: 0.358 D_GP: 0.029 D_real: 0.935 D_fake: 0.880 +(epoch: 292, iters: 6672, time: 0.064) G_GAN: 0.420 G_GAN_Feat: 0.780 G_ID: 0.150 G_Rec: 0.364 D_GP: 0.022 D_real: 1.202 D_fake: 0.580 +(epoch: 292, iters: 7072, time: 0.064) G_GAN: 0.361 G_GAN_Feat: 0.787 G_ID: 0.139 G_Rec: 0.308 D_GP: 0.039 D_real: 1.101 D_fake: 0.641 +(epoch: 292, iters: 7472, time: 0.064) G_GAN: 0.507 G_GAN_Feat: 0.888 G_ID: 0.141 G_Rec: 0.377 D_GP: 0.042 D_real: 1.021 D_fake: 0.499 +(epoch: 292, iters: 7872, time: 0.064) G_GAN: 0.546 G_GAN_Feat: 0.747 G_ID: 0.117 G_Rec: 0.288 D_GP: 0.041 D_real: 1.336 D_fake: 0.488 +(epoch: 292, iters: 8272, time: 0.064) G_GAN: 0.424 G_GAN_Feat: 0.849 G_ID: 0.157 G_Rec: 0.400 D_GP: 0.022 D_real: 1.208 D_fake: 0.577 +(epoch: 293, iters: 64, time: 0.064) G_GAN: 0.308 G_GAN_Feat: 0.659 G_ID: 0.140 G_Rec: 0.270 D_GP: 0.020 D_real: 1.233 D_fake: 0.692 +(epoch: 293, iters: 464, time: 0.064) G_GAN: 0.295 G_GAN_Feat: 0.961 G_ID: 0.157 G_Rec: 0.400 D_GP: 0.051 D_real: 0.732 D_fake: 0.709 +(epoch: 293, iters: 864, time: 0.064) G_GAN: 0.409 G_GAN_Feat: 0.747 G_ID: 0.147 G_Rec: 0.290 D_GP: 0.057 D_real: 1.042 D_fake: 0.603 +(epoch: 293, iters: 1264, time: 0.064) G_GAN: 0.247 G_GAN_Feat: 0.894 G_ID: 0.170 G_Rec: 0.395 D_GP: 0.029 D_real: 0.961 D_fake: 0.753 +(epoch: 293, iters: 1664, time: 0.064) G_GAN: 0.247 G_GAN_Feat: 0.625 G_ID: 0.119 G_Rec: 0.281 D_GP: 0.022 D_real: 1.175 D_fake: 0.753 +(epoch: 293, iters: 2064, time: 0.064) G_GAN: 0.350 G_GAN_Feat: 0.846 G_ID: 0.154 G_Rec: 0.365 D_GP: 0.032 D_real: 1.163 D_fake: 0.659 +(epoch: 293, iters: 2464, time: 0.064) G_GAN: 0.021 G_GAN_Feat: 0.689 G_ID: 0.139 G_Rec: 0.297 D_GP: 0.062 D_real: 0.757 D_fake: 0.980 +(epoch: 293, iters: 2864, time: 0.064) G_GAN: 0.537 G_GAN_Feat: 0.923 G_ID: 0.174 G_Rec: 0.382 D_GP: 0.038 D_real: 1.085 D_fake: 0.484 +(epoch: 293, iters: 3264, time: 0.064) G_GAN: 0.104 G_GAN_Feat: 0.629 G_ID: 0.145 G_Rec: 0.252 D_GP: 0.022 D_real: 0.992 D_fake: 0.896 +(epoch: 293, iters: 3664, time: 0.064) G_GAN: 0.300 G_GAN_Feat: 1.114 G_ID: 0.158 G_Rec: 0.399 D_GP: 0.141 D_real: 0.351 D_fake: 0.701 +(epoch: 293, iters: 4064, time: 0.064) G_GAN: 0.263 G_GAN_Feat: 0.648 G_ID: 0.158 G_Rec: 0.262 D_GP: 0.026 D_real: 1.161 D_fake: 0.738 +(epoch: 293, iters: 4464, time: 0.064) G_GAN: 0.256 G_GAN_Feat: 0.895 G_ID: 0.157 G_Rec: 0.404 D_GP: 0.033 D_real: 0.913 D_fake: 0.746 +(epoch: 293, iters: 4864, time: 0.064) G_GAN: 0.086 G_GAN_Feat: 0.681 G_ID: 0.122 G_Rec: 0.269 D_GP: 0.030 D_real: 0.923 D_fake: 0.914 +(epoch: 293, iters: 5264, time: 0.064) G_GAN: 0.603 G_GAN_Feat: 0.976 G_ID: 0.177 G_Rec: 0.398 D_GP: 0.095 D_real: 0.983 D_fake: 0.432 +(epoch: 293, iters: 5664, time: 0.064) G_GAN: 0.268 G_GAN_Feat: 0.870 G_ID: 0.125 G_Rec: 0.338 D_GP: 0.084 D_real: 0.491 D_fake: 0.735 +(epoch: 293, iters: 6064, time: 0.064) G_GAN: 0.157 G_GAN_Feat: 0.885 G_ID: 0.171 G_Rec: 0.376 D_GP: 0.028 D_real: 0.861 D_fake: 0.844 +(epoch: 293, iters: 6464, time: 0.064) G_GAN: 0.297 G_GAN_Feat: 0.831 G_ID: 0.113 G_Rec: 0.294 D_GP: 0.077 D_real: 0.616 D_fake: 0.730 +(epoch: 293, iters: 6864, time: 0.064) G_GAN: 0.303 G_GAN_Feat: 1.158 G_ID: 0.161 G_Rec: 0.452 D_GP: 0.279 D_real: 0.482 D_fake: 0.699 +(epoch: 293, iters: 7264, time: 0.064) G_GAN: 0.023 G_GAN_Feat: 0.835 G_ID: 0.129 G_Rec: 0.331 D_GP: 0.034 D_real: 0.876 D_fake: 0.999 +(epoch: 293, iters: 7664, time: 0.064) G_GAN: 0.191 G_GAN_Feat: 0.865 G_ID: 0.165 G_Rec: 0.378 D_GP: 0.023 D_real: 1.005 D_fake: 0.809 +(epoch: 293, iters: 8064, time: 0.064) G_GAN: 0.107 G_GAN_Feat: 0.779 G_ID: 0.126 G_Rec: 0.343 D_GP: 0.061 D_real: 0.809 D_fake: 0.895 +(epoch: 293, iters: 8464, time: 0.064) G_GAN: 0.213 G_GAN_Feat: 0.903 G_ID: 0.165 G_Rec: 0.409 D_GP: 0.048 D_real: 0.775 D_fake: 0.788 +(epoch: 294, iters: 256, time: 0.064) G_GAN: -0.003 G_GAN_Feat: 0.734 G_ID: 0.134 G_Rec: 0.292 D_GP: 0.074 D_real: 0.689 D_fake: 1.003 +(epoch: 294, iters: 656, time: 0.064) G_GAN: 0.546 G_GAN_Feat: 0.864 G_ID: 0.173 G_Rec: 0.376 D_GP: 0.026 D_real: 1.269 D_fake: 0.463 +(epoch: 294, iters: 1056, time: 0.064) G_GAN: 0.385 G_GAN_Feat: 0.686 G_ID: 0.133 G_Rec: 0.284 D_GP: 0.025 D_real: 1.308 D_fake: 0.616 +(epoch: 294, iters: 1456, time: 0.064) G_GAN: 0.498 G_GAN_Feat: 0.922 G_ID: 0.144 G_Rec: 0.389 D_GP: 0.029 D_real: 1.086 D_fake: 0.504 +(epoch: 294, iters: 1856, time: 0.064) G_GAN: 0.405 G_GAN_Feat: 0.839 G_ID: 0.114 G_Rec: 0.287 D_GP: 0.026 D_real: 0.869 D_fake: 0.615 +(epoch: 294, iters: 2256, time: 0.064) G_GAN: 0.259 G_GAN_Feat: 0.838 G_ID: 0.156 G_Rec: 0.386 D_GP: 0.023 D_real: 1.044 D_fake: 0.743 +(epoch: 294, iters: 2656, time: 0.064) G_GAN: 0.206 G_GAN_Feat: 0.628 G_ID: 0.128 G_Rec: 0.268 D_GP: 0.022 D_real: 1.142 D_fake: 0.795 +(epoch: 294, iters: 3056, time: 0.064) G_GAN: 0.353 G_GAN_Feat: 0.847 G_ID: 0.176 G_Rec: 0.367 D_GP: 0.022 D_real: 1.216 D_fake: 0.648 +(epoch: 294, iters: 3456, time: 0.064) G_GAN: 0.334 G_GAN_Feat: 0.899 G_ID: 0.101 G_Rec: 0.315 D_GP: 0.065 D_real: 0.614 D_fake: 0.673 +(epoch: 294, iters: 3856, time: 0.064) G_GAN: 0.391 G_GAN_Feat: 0.869 G_ID: 0.159 G_Rec: 0.393 D_GP: 0.022 D_real: 1.132 D_fake: 0.609 +(epoch: 294, iters: 4256, time: 0.064) G_GAN: 0.179 G_GAN_Feat: 0.817 G_ID: 0.146 G_Rec: 0.292 D_GP: 0.039 D_real: 1.119 D_fake: 0.822 +(epoch: 294, iters: 4656, time: 0.064) G_GAN: 0.386 G_GAN_Feat: 0.937 G_ID: 0.153 G_Rec: 0.349 D_GP: 0.027 D_real: 0.826 D_fake: 0.615 +(epoch: 294, iters: 5056, time: 0.064) G_GAN: 0.032 G_GAN_Feat: 0.619 G_ID: 0.141 G_Rec: 0.282 D_GP: 0.020 D_real: 0.991 D_fake: 0.968 +(epoch: 294, iters: 5456, time: 0.064) G_GAN: 0.305 G_GAN_Feat: 0.953 G_ID: 0.156 G_Rec: 0.398 D_GP: 0.101 D_real: 0.696 D_fake: 0.698 +(epoch: 294, iters: 5856, time: 0.064) G_GAN: 0.307 G_GAN_Feat: 0.605 G_ID: 0.137 G_Rec: 0.285 D_GP: 0.018 D_real: 1.282 D_fake: 0.694 +(epoch: 294, iters: 6256, time: 0.064) G_GAN: 0.518 G_GAN_Feat: 0.861 G_ID: 0.144 G_Rec: 0.442 D_GP: 0.029 D_real: 1.215 D_fake: 0.491 +(epoch: 294, iters: 6656, time: 0.064) G_GAN: 0.276 G_GAN_Feat: 0.708 G_ID: 0.122 G_Rec: 0.298 D_GP: 0.021 D_real: 1.172 D_fake: 0.724 +(epoch: 294, iters: 7056, time: 0.064) G_GAN: 0.500 G_GAN_Feat: 0.903 G_ID: 0.158 G_Rec: 0.377 D_GP: 0.051 D_real: 1.135 D_fake: 0.501 +(epoch: 294, iters: 7456, time: 0.064) G_GAN: 0.254 G_GAN_Feat: 0.733 G_ID: 0.123 G_Rec: 0.279 D_GP: 0.040 D_real: 0.867 D_fake: 0.746 +(epoch: 294, iters: 7856, time: 0.064) G_GAN: 0.374 G_GAN_Feat: 0.967 G_ID: 0.166 G_Rec: 0.413 D_GP: 0.054 D_real: 0.682 D_fake: 0.629 +(epoch: 294, iters: 8256, time: 0.064) G_GAN: 0.256 G_GAN_Feat: 0.779 G_ID: 0.109 G_Rec: 0.288 D_GP: 0.050 D_real: 0.701 D_fake: 0.744 +(epoch: 295, iters: 48, time: 0.064) G_GAN: 0.506 G_GAN_Feat: 0.832 G_ID: 0.159 G_Rec: 0.364 D_GP: 0.030 D_real: 1.209 D_fake: 0.500 +(epoch: 295, iters: 448, time: 0.064) G_GAN: 0.329 G_GAN_Feat: 0.830 G_ID: 0.107 G_Rec: 0.308 D_GP: 0.039 D_real: 0.936 D_fake: 0.672 +(epoch: 295, iters: 848, time: 0.064) G_GAN: 0.616 G_GAN_Feat: 1.167 G_ID: 0.160 G_Rec: 0.428 D_GP: 0.039 D_real: 0.372 D_fake: 0.415 +(epoch: 295, iters: 1248, time: 0.064) G_GAN: -0.006 G_GAN_Feat: 0.603 G_ID: 0.121 G_Rec: 0.311 D_GP: 0.017 D_real: 0.978 D_fake: 1.006 +(epoch: 295, iters: 1648, time: 0.064) G_GAN: 0.297 G_GAN_Feat: 0.689 G_ID: 0.145 G_Rec: 0.356 D_GP: 0.019 D_real: 1.196 D_fake: 0.705 +(epoch: 295, iters: 2048, time: 0.064) G_GAN: 0.220 G_GAN_Feat: 0.555 G_ID: 0.106 G_Rec: 0.271 D_GP: 0.026 D_real: 1.185 D_fake: 0.781 +(epoch: 295, iters: 2448, time: 0.064) G_GAN: 0.191 G_GAN_Feat: 0.748 G_ID: 0.149 G_Rec: 0.401 D_GP: 0.027 D_real: 0.947 D_fake: 0.809 +(epoch: 295, iters: 2848, time: 0.064) G_GAN: 0.108 G_GAN_Feat: 0.566 G_ID: 0.122 G_Rec: 0.256 D_GP: 0.027 D_real: 1.021 D_fake: 0.892 +(epoch: 295, iters: 3248, time: 0.064) G_GAN: 0.324 G_GAN_Feat: 0.755 G_ID: 0.165 G_Rec: 0.388 D_GP: 0.019 D_real: 1.156 D_fake: 0.676 +(epoch: 295, iters: 3648, time: 0.064) G_GAN: 0.113 G_GAN_Feat: 0.674 G_ID: 0.165 G_Rec: 0.296 D_GP: 0.023 D_real: 0.948 D_fake: 0.887 +(epoch: 295, iters: 4048, time: 0.064) G_GAN: 0.426 G_GAN_Feat: 0.792 G_ID: 0.166 G_Rec: 0.357 D_GP: 0.039 D_real: 1.166 D_fake: 0.581 +(epoch: 295, iters: 4448, time: 0.064) G_GAN: 0.096 G_GAN_Feat: 0.744 G_ID: 0.105 G_Rec: 0.328 D_GP: 0.063 D_real: 0.826 D_fake: 0.904 +(epoch: 295, iters: 4848, time: 0.064) G_GAN: 0.261 G_GAN_Feat: 0.838 G_ID: 0.160 G_Rec: 0.384 D_GP: 0.045 D_real: 0.969 D_fake: 0.739 +(epoch: 295, iters: 5248, time: 0.064) G_GAN: 0.104 G_GAN_Feat: 0.653 G_ID: 0.113 G_Rec: 0.277 D_GP: 0.025 D_real: 1.033 D_fake: 0.896 +(epoch: 295, iters: 5648, time: 0.064) G_GAN: 0.460 G_GAN_Feat: 0.823 G_ID: 0.150 G_Rec: 0.398 D_GP: 0.024 D_real: 1.215 D_fake: 0.543 +(epoch: 295, iters: 6048, time: 0.064) G_GAN: 0.102 G_GAN_Feat: 0.709 G_ID: 0.179 G_Rec: 0.300 D_GP: 0.048 D_real: 0.786 D_fake: 0.898 +(epoch: 295, iters: 6448, time: 0.064) G_GAN: 0.285 G_GAN_Feat: 0.894 G_ID: 0.155 G_Rec: 0.367 D_GP: 0.059 D_real: 0.724 D_fake: 0.724 +(epoch: 295, iters: 6848, time: 0.064) G_GAN: 0.468 G_GAN_Feat: 0.702 G_ID: 0.122 G_Rec: 0.270 D_GP: 0.024 D_real: 1.284 D_fake: 0.532 +(epoch: 295, iters: 7248, time: 0.064) G_GAN: 0.822 G_GAN_Feat: 1.100 G_ID: 0.174 G_Rec: 0.402 D_GP: 0.064 D_real: 1.073 D_fake: 0.390 +(epoch: 295, iters: 7648, time: 0.064) G_GAN: 0.179 G_GAN_Feat: 0.710 G_ID: 0.135 G_Rec: 0.299 D_GP: 0.025 D_real: 1.120 D_fake: 0.821 +(epoch: 295, iters: 8048, time: 0.064) G_GAN: 0.640 G_GAN_Feat: 0.921 G_ID: 0.163 G_Rec: 0.414 D_GP: 0.037 D_real: 1.147 D_fake: 0.405 +(epoch: 295, iters: 8448, time: 0.064) G_GAN: 0.166 G_GAN_Feat: 0.611 G_ID: 0.117 G_Rec: 0.264 D_GP: 0.019 D_real: 1.106 D_fake: 0.834 +(epoch: 296, iters: 240, time: 0.064) G_GAN: 0.497 G_GAN_Feat: 0.943 G_ID: 0.156 G_Rec: 0.425 D_GP: 0.065 D_real: 0.959 D_fake: 0.510 +(epoch: 296, iters: 640, time: 0.064) G_GAN: 0.127 G_GAN_Feat: 0.893 G_ID: 0.135 G_Rec: 0.334 D_GP: 0.393 D_real: 0.603 D_fake: 0.903 +(epoch: 296, iters: 1040, time: 0.064) G_GAN: 0.694 G_GAN_Feat: 0.800 G_ID: 0.148 G_Rec: 0.356 D_GP: 0.022 D_real: 1.492 D_fake: 0.320 +(epoch: 296, iters: 1440, time: 0.064) G_GAN: 0.179 G_GAN_Feat: 0.702 G_ID: 0.133 G_Rec: 0.322 D_GP: 0.020 D_real: 1.125 D_fake: 0.822 +(epoch: 296, iters: 1840, time: 0.064) G_GAN: 0.481 G_GAN_Feat: 0.829 G_ID: 0.151 G_Rec: 0.378 D_GP: 0.019 D_real: 1.317 D_fake: 0.527 +(epoch: 296, iters: 2240, time: 0.064) G_GAN: 0.374 G_GAN_Feat: 0.853 G_ID: 0.131 G_Rec: 0.337 D_GP: 0.100 D_real: 0.897 D_fake: 0.656 +(epoch: 296, iters: 2640, time: 0.064) G_GAN: 0.273 G_GAN_Feat: 0.882 G_ID: 0.176 G_Rec: 0.356 D_GP: 0.037 D_real: 1.108 D_fake: 0.729 +(epoch: 296, iters: 3040, time: 0.064) G_GAN: 0.447 G_GAN_Feat: 0.631 G_ID: 0.126 G_Rec: 0.282 D_GP: 0.019 D_real: 1.369 D_fake: 0.558 +(epoch: 296, iters: 3440, time: 0.064) G_GAN: 0.330 G_GAN_Feat: 0.891 G_ID: 0.170 G_Rec: 0.362 D_GP: 0.030 D_real: 1.053 D_fake: 0.673 +(epoch: 296, iters: 3840, time: 0.064) G_GAN: 0.318 G_GAN_Feat: 0.792 G_ID: 0.126 G_Rec: 0.293 D_GP: 0.073 D_real: 1.019 D_fake: 0.688 +(epoch: 296, iters: 4240, time: 0.064) G_GAN: 0.412 G_GAN_Feat: 0.954 G_ID: 0.159 G_Rec: 0.383 D_GP: 0.046 D_real: 0.856 D_fake: 0.589 +(epoch: 296, iters: 4640, time: 0.064) G_GAN: 0.272 G_GAN_Feat: 0.863 G_ID: 0.123 G_Rec: 0.312 D_GP: 0.136 D_real: 0.586 D_fake: 0.738 +(epoch: 296, iters: 5040, time: 0.064) G_GAN: 0.735 G_GAN_Feat: 1.073 G_ID: 0.172 G_Rec: 0.411 D_GP: 1.489 D_real: 0.654 D_fake: 0.317 +(epoch: 296, iters: 5440, time: 0.064) G_GAN: 0.106 G_GAN_Feat: 0.540 G_ID: 0.122 G_Rec: 0.257 D_GP: 0.017 D_real: 1.075 D_fake: 0.894 +(epoch: 296, iters: 5840, time: 0.064) G_GAN: 0.354 G_GAN_Feat: 0.824 G_ID: 0.157 G_Rec: 0.409 D_GP: 0.018 D_real: 1.060 D_fake: 0.650 +(epoch: 296, iters: 6240, time: 0.064) G_GAN: 0.266 G_GAN_Feat: 0.606 G_ID: 0.103 G_Rec: 0.280 D_GP: 0.019 D_real: 1.214 D_fake: 0.738 +(epoch: 296, iters: 6640, time: 0.064) G_GAN: 0.197 G_GAN_Feat: 0.765 G_ID: 0.178 G_Rec: 0.368 D_GP: 0.027 D_real: 1.060 D_fake: 0.804 +(epoch: 296, iters: 7040, time: 0.064) G_GAN: 0.034 G_GAN_Feat: 0.583 G_ID: 0.140 G_Rec: 0.276 D_GP: 0.026 D_real: 0.937 D_fake: 0.966 +(epoch: 296, iters: 7440, time: 0.064) G_GAN: 0.063 G_GAN_Feat: 0.780 G_ID: 0.184 G_Rec: 0.377 D_GP: 0.037 D_real: 0.791 D_fake: 0.937 +(epoch: 296, iters: 7840, time: 0.064) G_GAN: 0.091 G_GAN_Feat: 0.653 G_ID: 0.129 G_Rec: 0.293 D_GP: 0.035 D_real: 0.982 D_fake: 0.909 +(epoch: 296, iters: 8240, time: 0.064) G_GAN: 0.139 G_GAN_Feat: 0.756 G_ID: 0.164 G_Rec: 0.370 D_GP: 0.023 D_real: 0.932 D_fake: 0.861 +(epoch: 297, iters: 32, time: 0.064) G_GAN: 0.135 G_GAN_Feat: 0.675 G_ID: 0.128 G_Rec: 0.300 D_GP: 0.030 D_real: 1.053 D_fake: 0.865 +(epoch: 297, iters: 432, time: 0.064) G_GAN: 0.417 G_GAN_Feat: 0.854 G_ID: 0.163 G_Rec: 0.399 D_GP: 0.048 D_real: 1.058 D_fake: 0.585 +(epoch: 297, iters: 832, time: 0.064) G_GAN: 0.052 G_GAN_Feat: 0.940 G_ID: 0.122 G_Rec: 0.344 D_GP: 0.022 D_real: 1.280 D_fake: 0.951 +(epoch: 297, iters: 1232, time: 0.064) G_GAN: 0.179 G_GAN_Feat: 0.787 G_ID: 0.177 G_Rec: 0.402 D_GP: 0.020 D_real: 0.934 D_fake: 0.821 +(epoch: 297, iters: 1632, time: 0.064) G_GAN: 0.212 G_GAN_Feat: 0.598 G_ID: 0.133 G_Rec: 0.291 D_GP: 0.019 D_real: 1.106 D_fake: 0.788 +(epoch: 297, iters: 2032, time: 0.064) G_GAN: 0.394 G_GAN_Feat: 0.796 G_ID: 0.143 G_Rec: 0.398 D_GP: 0.024 D_real: 1.171 D_fake: 0.609 +(epoch: 297, iters: 2432, time: 0.064) G_GAN: 0.098 G_GAN_Feat: 0.631 G_ID: 0.153 G_Rec: 0.328 D_GP: 0.023 D_real: 1.040 D_fake: 0.902 +(epoch: 297, iters: 2832, time: 0.064) G_GAN: 0.422 G_GAN_Feat: 0.722 G_ID: 0.162 G_Rec: 0.315 D_GP: 0.035 D_real: 1.241 D_fake: 0.584 +(epoch: 297, iters: 3232, time: 0.064) G_GAN: -0.030 G_GAN_Feat: 0.706 G_ID: 0.144 G_Rec: 0.287 D_GP: 0.052 D_real: 0.745 D_fake: 1.030 +(epoch: 297, iters: 3632, time: 0.064) G_GAN: 0.319 G_GAN_Feat: 0.868 G_ID: 0.139 G_Rec: 0.389 D_GP: 0.046 D_real: 0.966 D_fake: 0.681 +(epoch: 297, iters: 4032, time: 0.064) G_GAN: 0.328 G_GAN_Feat: 0.621 G_ID: 0.105 G_Rec: 0.262 D_GP: 0.035 D_real: 1.236 D_fake: 0.672 +(epoch: 297, iters: 4432, time: 0.064) G_GAN: -0.221 G_GAN_Feat: 1.105 G_ID: 0.201 G_Rec: 0.466 D_GP: 1.106 D_real: 0.257 D_fake: 1.221 +(epoch: 297, iters: 4832, time: 0.064) G_GAN: 0.263 G_GAN_Feat: 0.719 G_ID: 0.115 G_Rec: 0.311 D_GP: 0.027 D_real: 1.065 D_fake: 0.737 +(epoch: 297, iters: 5232, time: 0.064) G_GAN: 0.544 G_GAN_Feat: 0.813 G_ID: 0.166 G_Rec: 0.347 D_GP: 0.030 D_real: 1.297 D_fake: 0.466 +(epoch: 297, iters: 5632, time: 0.064) G_GAN: 0.225 G_GAN_Feat: 0.657 G_ID: 0.110 G_Rec: 0.293 D_GP: 0.026 D_real: 1.184 D_fake: 0.775 +(epoch: 297, iters: 6032, time: 0.064) G_GAN: 0.644 G_GAN_Feat: 1.034 G_ID: 0.154 G_Rec: 0.439 D_GP: 0.077 D_real: 0.854 D_fake: 0.385 +(epoch: 297, iters: 6432, time: 0.064) G_GAN: 0.132 G_GAN_Feat: 0.687 G_ID: 0.128 G_Rec: 0.286 D_GP: 0.033 D_real: 1.019 D_fake: 0.868 +(epoch: 297, iters: 6832, time: 0.064) G_GAN: 0.266 G_GAN_Feat: 0.946 G_ID: 0.171 G_Rec: 0.377 D_GP: 0.148 D_real: 0.583 D_fake: 0.735 +(epoch: 297, iters: 7232, time: 0.064) G_GAN: 0.266 G_GAN_Feat: 0.736 G_ID: 0.131 G_Rec: 0.274 D_GP: 0.027 D_real: 1.053 D_fake: 0.735 +(epoch: 297, iters: 7632, time: 0.064) G_GAN: 0.365 G_GAN_Feat: 0.895 G_ID: 0.157 G_Rec: 0.419 D_GP: 0.030 D_real: 1.020 D_fake: 0.637 +(epoch: 297, iters: 8032, time: 0.064) G_GAN: 0.289 G_GAN_Feat: 0.621 G_ID: 0.116 G_Rec: 0.273 D_GP: 0.019 D_real: 1.220 D_fake: 0.711 +(epoch: 297, iters: 8432, time: 0.064) G_GAN: 0.638 G_GAN_Feat: 0.876 G_ID: 0.187 G_Rec: 0.407 D_GP: 0.034 D_real: 1.275 D_fake: 0.373 +(epoch: 298, iters: 224, time: 0.064) G_GAN: 0.229 G_GAN_Feat: 0.660 G_ID: 0.113 G_Rec: 0.290 D_GP: 0.029 D_real: 1.118 D_fake: 0.771 +(epoch: 298, iters: 624, time: 0.064) G_GAN: 0.325 G_GAN_Feat: 0.936 G_ID: 0.180 G_Rec: 0.386 D_GP: 0.102 D_real: 0.748 D_fake: 0.676 +(epoch: 298, iters: 1024, time: 0.064) G_GAN: 0.279 G_GAN_Feat: 0.774 G_ID: 0.103 G_Rec: 0.303 D_GP: 0.099 D_real: 0.800 D_fake: 0.723 +(epoch: 298, iters: 1424, time: 0.064) G_GAN: 0.787 G_GAN_Feat: 0.846 G_ID: 0.135 G_Rec: 0.414 D_GP: 0.026 D_real: 1.396 D_fake: 0.270 +(epoch: 298, iters: 1824, time: 0.064) G_GAN: 0.294 G_GAN_Feat: 0.659 G_ID: 0.133 G_Rec: 0.293 D_GP: 0.035 D_real: 1.145 D_fake: 0.707 +(epoch: 298, iters: 2224, time: 0.064) G_GAN: 0.457 G_GAN_Feat: 0.890 G_ID: 0.139 G_Rec: 0.385 D_GP: 0.028 D_real: 1.049 D_fake: 0.545 +(epoch: 298, iters: 2624, time: 0.064) G_GAN: 0.251 G_GAN_Feat: 0.740 G_ID: 0.116 G_Rec: 0.276 D_GP: 0.027 D_real: 0.843 D_fake: 0.749 +(epoch: 298, iters: 3024, time: 0.064) G_GAN: 0.069 G_GAN_Feat: 0.852 G_ID: 0.163 G_Rec: 0.397 D_GP: 0.035 D_real: 0.771 D_fake: 0.931 +(epoch: 298, iters: 3424, time: 0.064) G_GAN: -0.169 G_GAN_Feat: 0.733 G_ID: 0.123 G_Rec: 0.270 D_GP: 0.027 D_real: 0.654 D_fake: 1.169 +(epoch: 298, iters: 3824, time: 0.064) G_GAN: 0.498 G_GAN_Feat: 0.928 G_ID: 0.160 G_Rec: 0.415 D_GP: 0.025 D_real: 1.107 D_fake: 0.507 +(epoch: 298, iters: 4224, time: 0.064) G_GAN: 0.245 G_GAN_Feat: 0.782 G_ID: 0.150 G_Rec: 0.310 D_GP: 0.030 D_real: 0.961 D_fake: 0.755 +(epoch: 298, iters: 4624, time: 0.064) G_GAN: 0.482 G_GAN_Feat: 0.878 G_ID: 0.147 G_Rec: 0.381 D_GP: 0.030 D_real: 1.163 D_fake: 0.521 +(epoch: 298, iters: 5024, time: 0.064) G_GAN: 0.074 G_GAN_Feat: 0.736 G_ID: 0.152 G_Rec: 0.356 D_GP: 0.026 D_real: 0.972 D_fake: 0.926 +(epoch: 298, iters: 5424, time: 0.064) G_GAN: 0.136 G_GAN_Feat: 0.830 G_ID: 0.192 G_Rec: 0.359 D_GP: 0.023 D_real: 0.834 D_fake: 0.864 +(epoch: 298, iters: 5824, time: 0.064) G_GAN: 0.216 G_GAN_Feat: 0.756 G_ID: 0.132 G_Rec: 0.311 D_GP: 0.048 D_real: 0.996 D_fake: 0.785 +(epoch: 298, iters: 6224, time: 0.064) G_GAN: 0.414 G_GAN_Feat: 0.901 G_ID: 0.158 G_Rec: 0.411 D_GP: 0.027 D_real: 1.076 D_fake: 0.590 +(epoch: 298, iters: 6624, time: 0.064) G_GAN: 0.170 G_GAN_Feat: 0.704 G_ID: 0.121 G_Rec: 0.255 D_GP: 0.039 D_real: 0.935 D_fake: 0.830 +(epoch: 298, iters: 7024, time: 0.064) G_GAN: 0.184 G_GAN_Feat: 1.070 G_ID: 0.169 G_Rec: 0.424 D_GP: 0.194 D_real: 0.442 D_fake: 0.817 +(epoch: 298, iters: 7424, time: 0.064) G_GAN: 0.140 G_GAN_Feat: 0.638 G_ID: 0.132 G_Rec: 0.300 D_GP: 0.021 D_real: 1.087 D_fake: 0.860 +(epoch: 298, iters: 7824, time: 0.064) G_GAN: 0.137 G_GAN_Feat: 0.918 G_ID: 0.177 G_Rec: 0.432 D_GP: 0.043 D_real: 0.733 D_fake: 0.864 +(epoch: 298, iters: 8224, time: 0.064) G_GAN: 0.099 G_GAN_Feat: 0.678 G_ID: 0.123 G_Rec: 0.274 D_GP: 0.043 D_real: 0.917 D_fake: 0.901 +(epoch: 299, iters: 16, time: 0.064) G_GAN: 0.359 G_GAN_Feat: 0.879 G_ID: 0.152 G_Rec: 0.349 D_GP: 0.031 D_real: 0.918 D_fake: 0.642 +(epoch: 299, iters: 416, time: 0.064) G_GAN: 0.155 G_GAN_Feat: 0.653 G_ID: 0.133 G_Rec: 0.271 D_GP: 0.029 D_real: 1.036 D_fake: 0.845 +(epoch: 299, iters: 816, time: 0.064) G_GAN: 0.476 G_GAN_Feat: 0.880 G_ID: 0.165 G_Rec: 0.372 D_GP: 0.025 D_real: 1.067 D_fake: 0.531 +(epoch: 299, iters: 1216, time: 0.064) G_GAN: 0.190 G_GAN_Feat: 0.704 G_ID: 0.105 G_Rec: 0.267 D_GP: 0.025 D_real: 0.974 D_fake: 0.810 +(epoch: 299, iters: 1616, time: 0.064) G_GAN: 0.244 G_GAN_Feat: 1.013 G_ID: 0.169 G_Rec: 0.374 D_GP: 0.041 D_real: 0.709 D_fake: 0.757 +(epoch: 299, iters: 2016, time: 0.064) G_GAN: 0.236 G_GAN_Feat: 0.706 G_ID: 0.119 G_Rec: 0.278 D_GP: 0.023 D_real: 1.099 D_fake: 0.764 +(epoch: 299, iters: 2416, time: 0.064) G_GAN: 0.244 G_GAN_Feat: 0.831 G_ID: 0.155 G_Rec: 0.376 D_GP: 0.024 D_real: 1.029 D_fake: 0.756 +(epoch: 299, iters: 2816, time: 0.064) G_GAN: 0.248 G_GAN_Feat: 0.702 G_ID: 0.140 G_Rec: 0.264 D_GP: 0.026 D_real: 1.128 D_fake: 0.752 +(epoch: 299, iters: 3216, time: 0.064) G_GAN: 0.364 G_GAN_Feat: 0.773 G_ID: 0.171 G_Rec: 0.362 D_GP: 0.020 D_real: 1.175 D_fake: 0.636 +(epoch: 299, iters: 3616, time: 0.064) G_GAN: 0.032 G_GAN_Feat: 0.726 G_ID: 0.128 G_Rec: 0.292 D_GP: 0.030 D_real: 0.876 D_fake: 0.968 +(epoch: 299, iters: 4016, time: 0.064) G_GAN: 0.179 G_GAN_Feat: 0.827 G_ID: 0.164 G_Rec: 0.365 D_GP: 0.025 D_real: 0.941 D_fake: 0.821 +(epoch: 299, iters: 4416, time: 0.064) G_GAN: 0.271 G_GAN_Feat: 0.704 G_ID: 0.109 G_Rec: 0.283 D_GP: 0.038 D_real: 1.090 D_fake: 0.730 +(epoch: 299, iters: 4816, time: 0.064) G_GAN: 0.428 G_GAN_Feat: 0.799 G_ID: 0.144 G_Rec: 0.363 D_GP: 0.021 D_real: 1.185 D_fake: 0.573 +(epoch: 299, iters: 5216, time: 0.064) G_GAN: 0.158 G_GAN_Feat: 0.975 G_ID: 0.121 G_Rec: 0.319 D_GP: 0.041 D_real: 1.169 D_fake: 0.855 +(epoch: 299, iters: 5616, time: 0.064) G_GAN: 0.917 G_GAN_Feat: 1.094 G_ID: 0.165 G_Rec: 0.443 D_GP: 0.097 D_real: 1.350 D_fake: 0.368 +(epoch: 299, iters: 6016, time: 0.064) G_GAN: 0.421 G_GAN_Feat: 0.707 G_ID: 0.113 G_Rec: 0.280 D_GP: 0.040 D_real: 1.152 D_fake: 0.581 +(epoch: 299, iters: 6416, time: 0.064) G_GAN: 0.370 G_GAN_Feat: 1.107 G_ID: 0.188 G_Rec: 0.446 D_GP: 0.104 D_real: 0.365 D_fake: 0.632 +(epoch: 299, iters: 6816, time: 0.064) G_GAN: -0.078 G_GAN_Feat: 0.605 G_ID: 0.135 G_Rec: 0.277 D_GP: 0.019 D_real: 0.902 D_fake: 1.078 +(epoch: 299, iters: 7216, time: 0.064) G_GAN: 0.251 G_GAN_Feat: 0.846 G_ID: 0.178 G_Rec: 0.400 D_GP: 0.025 D_real: 0.952 D_fake: 0.752 +(epoch: 299, iters: 7616, time: 0.064) G_GAN: 0.118 G_GAN_Feat: 0.658 G_ID: 0.112 G_Rec: 0.302 D_GP: 0.034 D_real: 0.946 D_fake: 0.883 +(epoch: 299, iters: 8016, time: 0.064) G_GAN: 0.391 G_GAN_Feat: 0.783 G_ID: 0.178 G_Rec: 0.371 D_GP: 0.021 D_real: 1.236 D_fake: 0.609 +(epoch: 299, iters: 8416, time: 0.064) G_GAN: 0.104 G_GAN_Feat: 0.905 G_ID: 0.140 G_Rec: 0.337 D_GP: 0.115 D_real: 0.324 D_fake: 0.896 +(epoch: 300, iters: 208, time: 0.064) G_GAN: 0.457 G_GAN_Feat: 0.866 G_ID: 0.166 G_Rec: 0.389 D_GP: 0.027 D_real: 1.124 D_fake: 0.555 +(epoch: 300, iters: 608, time: 0.064) G_GAN: 0.278 G_GAN_Feat: 0.843 G_ID: 0.114 G_Rec: 0.327 D_GP: 0.066 D_real: 0.517 D_fake: 0.745 +(epoch: 300, iters: 1008, time: 0.064) G_GAN: -0.031 G_GAN_Feat: 0.777 G_ID: 0.203 G_Rec: 0.357 D_GP: 0.019 D_real: 0.806 D_fake: 1.031 +(epoch: 300, iters: 1408, time: 0.064) G_GAN: -0.008 G_GAN_Feat: 0.613 G_ID: 0.124 G_Rec: 0.269 D_GP: 0.018 D_real: 0.945 D_fake: 1.008 +(epoch: 300, iters: 1808, time: 0.064) G_GAN: -0.030 G_GAN_Feat: 0.755 G_ID: 0.167 G_Rec: 0.371 D_GP: 0.019 D_real: 0.801 D_fake: 1.030 +(epoch: 300, iters: 2208, time: 0.064) G_GAN: -0.094 G_GAN_Feat: 0.680 G_ID: 0.116 G_Rec: 0.300 D_GP: 0.027 D_real: 0.727 D_fake: 1.094 +(epoch: 300, iters: 2608, time: 0.064) G_GAN: 0.438 G_GAN_Feat: 0.863 G_ID: 0.152 G_Rec: 0.422 D_GP: 0.036 D_real: 1.144 D_fake: 0.576 +(epoch: 300, iters: 3008, time: 0.064) G_GAN: 0.152 G_GAN_Feat: 0.712 G_ID: 0.151 G_Rec: 0.288 D_GP: 0.027 D_real: 0.906 D_fake: 0.850 +(epoch: 300, iters: 3408, time: 0.064) G_GAN: 0.387 G_GAN_Feat: 0.877 G_ID: 0.174 G_Rec: 0.389 D_GP: 0.038 D_real: 1.000 D_fake: 0.614 +(epoch: 300, iters: 3808, time: 0.064) G_GAN: 0.298 G_GAN_Feat: 0.705 G_ID: 0.112 G_Rec: 0.300 D_GP: 0.037 D_real: 1.135 D_fake: 0.703 +(epoch: 300, iters: 4208, time: 0.064) G_GAN: 0.339 G_GAN_Feat: 0.881 G_ID: 0.165 G_Rec: 0.393 D_GP: 0.031 D_real: 1.135 D_fake: 0.662 +(epoch: 300, iters: 4608, time: 0.064) G_GAN: 0.516 G_GAN_Feat: 0.764 G_ID: 0.123 G_Rec: 0.312 D_GP: 0.037 D_real: 1.222 D_fake: 0.491 +(epoch: 300, iters: 5008, time: 0.064) G_GAN: 0.346 G_GAN_Feat: 0.971 G_ID: 0.166 G_Rec: 0.410 D_GP: 0.069 D_real: 0.706 D_fake: 0.659 +(epoch: 300, iters: 5408, time: 0.064) G_GAN: 0.339 G_GAN_Feat: 0.705 G_ID: 0.124 G_Rec: 0.281 D_GP: 0.029 D_real: 1.175 D_fake: 0.661 +(epoch: 300, iters: 5808, time: 0.064) G_GAN: 0.200 G_GAN_Feat: 0.956 G_ID: 0.179 G_Rec: 0.404 D_GP: 0.042 D_real: 0.744 D_fake: 0.801 +(epoch: 300, iters: 6208, time: 0.064) G_GAN: 0.474 G_GAN_Feat: 0.815 G_ID: 0.131 G_Rec: 0.289 D_GP: 0.038 D_real: 1.044 D_fake: 0.547 +(epoch: 300, iters: 6608, time: 0.064) G_GAN: 0.711 G_GAN_Feat: 0.986 G_ID: 0.155 G_Rec: 0.413 D_GP: 0.039 D_real: 1.024 D_fake: 0.330 +(epoch: 300, iters: 7008, time: 0.064) G_GAN: 0.506 G_GAN_Feat: 0.856 G_ID: 0.125 G_Rec: 0.291 D_GP: 0.047 D_real: 0.700 D_fake: 0.530 +(epoch: 300, iters: 7408, time: 0.064) G_GAN: 0.510 G_GAN_Feat: 0.805 G_ID: 0.174 G_Rec: 0.382 D_GP: 0.022 D_real: 1.228 D_fake: 0.499 +(epoch: 300, iters: 7808, time: 0.064) G_GAN: 0.095 G_GAN_Feat: 0.789 G_ID: 0.120 G_Rec: 0.311 D_GP: 0.116 D_real: 0.751 D_fake: 0.908 +(epoch: 300, iters: 8208, time: 0.064) G_GAN: 0.182 G_GAN_Feat: 0.813 G_ID: 0.138 G_Rec: 0.378 D_GP: 0.023 D_real: 0.958 D_fake: 0.819 +(epoch: 300, iters: 8608, time: 0.064) G_GAN: 0.257 G_GAN_Feat: 0.633 G_ID: 0.116 G_Rec: 0.282 D_GP: 0.023 D_real: 1.149 D_fake: 0.743 +(epoch: 301, iters: 400, time: 0.064) G_GAN: 0.597 G_GAN_Feat: 0.908 G_ID: 0.144 G_Rec: 0.399 D_GP: 0.027 D_real: 1.318 D_fake: 0.410 +(epoch: 301, iters: 800, time: 0.064) G_GAN: 0.132 G_GAN_Feat: 0.757 G_ID: 0.117 G_Rec: 0.295 D_GP: 0.056 D_real: 0.995 D_fake: 0.868 +(epoch: 301, iters: 1200, time: 0.064) G_GAN: 0.637 G_GAN_Feat: 0.881 G_ID: 0.156 G_Rec: 0.341 D_GP: 0.026 D_real: 1.305 D_fake: 0.388 +(epoch: 301, iters: 1600, time: 0.064) G_GAN: 0.565 G_GAN_Feat: 0.796 G_ID: 0.101 G_Rec: 0.305 D_GP: 0.025 D_real: 1.458 D_fake: 0.466 +(epoch: 301, iters: 2000, time: 0.064) G_GAN: 0.345 G_GAN_Feat: 0.902 G_ID: 0.145 G_Rec: 0.378 D_GP: 0.021 D_real: 1.092 D_fake: 0.656 +(epoch: 301, iters: 2400, time: 0.064) G_GAN: 0.165 G_GAN_Feat: 0.648 G_ID: 0.114 G_Rec: 0.270 D_GP: 0.020 D_real: 1.098 D_fake: 0.835 +(epoch: 301, iters: 2800, time: 0.064) G_GAN: 0.625 G_GAN_Feat: 0.888 G_ID: 0.176 G_Rec: 0.408 D_GP: 0.032 D_real: 1.455 D_fake: 0.386 +(epoch: 301, iters: 3200, time: 0.064) G_GAN: 0.011 G_GAN_Feat: 0.604 G_ID: 0.127 G_Rec: 0.280 D_GP: 0.024 D_real: 0.944 D_fake: 0.989 +(epoch: 301, iters: 3600, time: 0.064) G_GAN: 0.287 G_GAN_Feat: 0.833 G_ID: 0.186 G_Rec: 0.417 D_GP: 0.032 D_real: 1.038 D_fake: 0.714 +(epoch: 301, iters: 4000, time: 0.064) G_GAN: 0.062 G_GAN_Feat: 0.686 G_ID: 0.122 G_Rec: 0.294 D_GP: 0.040 D_real: 0.808 D_fake: 0.938 +(epoch: 301, iters: 4400, time: 0.064) G_GAN: 0.704 G_GAN_Feat: 0.960 G_ID: 0.144 G_Rec: 0.380 D_GP: 0.207 D_real: 1.058 D_fake: 0.321 +(epoch: 301, iters: 4800, time: 0.064) G_GAN: 0.007 G_GAN_Feat: 0.675 G_ID: 0.127 G_Rec: 0.292 D_GP: 0.020 D_real: 1.027 D_fake: 0.993 +(epoch: 301, iters: 5200, time: 0.064) G_GAN: 0.233 G_GAN_Feat: 0.846 G_ID: 0.189 G_Rec: 0.370 D_GP: 0.030 D_real: 0.925 D_fake: 0.767 +(epoch: 301, iters: 5600, time: 0.064) G_GAN: 0.394 G_GAN_Feat: 0.626 G_ID: 0.122 G_Rec: 0.276 D_GP: 0.018 D_real: 1.363 D_fake: 0.611 +(epoch: 301, iters: 6000, time: 0.064) G_GAN: 0.381 G_GAN_Feat: 0.888 G_ID: 0.176 G_Rec: 0.372 D_GP: 0.053 D_real: 0.866 D_fake: 0.622 +(epoch: 301, iters: 6400, time: 0.064) G_GAN: 0.017 G_GAN_Feat: 0.602 G_ID: 0.120 G_Rec: 0.283 D_GP: 0.021 D_real: 1.024 D_fake: 0.983 +(epoch: 301, iters: 6800, time: 0.064) G_GAN: 0.282 G_GAN_Feat: 0.841 G_ID: 0.163 G_Rec: 0.375 D_GP: 0.045 D_real: 1.013 D_fake: 0.724 +(epoch: 301, iters: 7200, time: 0.064) G_GAN: 0.091 G_GAN_Feat: 0.717 G_ID: 0.110 G_Rec: 0.310 D_GP: 0.048 D_real: 0.875 D_fake: 0.910 +(epoch: 301, iters: 7600, time: 0.064) G_GAN: 0.370 G_GAN_Feat: 0.986 G_ID: 0.146 G_Rec: 0.419 D_GP: 0.029 D_real: 0.848 D_fake: 0.633 +(epoch: 301, iters: 8000, time: 0.064) G_GAN: 0.166 G_GAN_Feat: 0.801 G_ID: 0.123 G_Rec: 0.292 D_GP: 0.028 D_real: 0.751 D_fake: 0.834 +(epoch: 301, iters: 8400, time: 0.064) G_GAN: 0.509 G_GAN_Feat: 0.844 G_ID: 0.170 G_Rec: 0.347 D_GP: 0.041 D_real: 1.292 D_fake: 0.503 +(epoch: 302, iters: 192, time: 0.064) G_GAN: 0.255 G_GAN_Feat: 0.812 G_ID: 0.124 G_Rec: 0.300 D_GP: 0.029 D_real: 0.897 D_fake: 0.748 +(epoch: 302, iters: 592, time: 0.064) G_GAN: 0.256 G_GAN_Feat: 0.915 G_ID: 0.142 G_Rec: 0.361 D_GP: 0.024 D_real: 0.788 D_fake: 0.744 +(epoch: 302, iters: 992, time: 0.064) G_GAN: 0.363 G_GAN_Feat: 0.716 G_ID: 0.113 G_Rec: 0.316 D_GP: 0.025 D_real: 1.279 D_fake: 0.637 +(epoch: 302, iters: 1392, time: 0.064) G_GAN: 0.394 G_GAN_Feat: 1.158 G_ID: 0.157 G_Rec: 0.414 D_GP: 0.553 D_real: 0.309 D_fake: 0.612 +(epoch: 302, iters: 1792, time: 0.064) G_GAN: 0.331 G_GAN_Feat: 0.810 G_ID: 0.113 G_Rec: 0.304 D_GP: 0.030 D_real: 0.881 D_fake: 0.678 +(epoch: 302, iters: 2192, time: 0.064) G_GAN: 0.203 G_GAN_Feat: 0.836 G_ID: 0.145 G_Rec: 0.326 D_GP: 0.024 D_real: 0.951 D_fake: 0.797 +(epoch: 302, iters: 2592, time: 0.064) G_GAN: -0.098 G_GAN_Feat: 0.872 G_ID: 0.138 G_Rec: 0.315 D_GP: 0.022 D_real: 0.501 D_fake: 1.098 +(epoch: 302, iters: 2992, time: 0.064) G_GAN: 0.267 G_GAN_Feat: 1.042 G_ID: 0.171 G_Rec: 0.412 D_GP: 0.034 D_real: 0.442 D_fake: 0.733 +(epoch: 302, iters: 3392, time: 0.064) G_GAN: 0.398 G_GAN_Feat: 0.567 G_ID: 0.128 G_Rec: 0.261 D_GP: 0.016 D_real: 1.372 D_fake: 0.604 +(epoch: 302, iters: 3792, time: 0.064) G_GAN: 0.513 G_GAN_Feat: 0.802 G_ID: 0.153 G_Rec: 0.355 D_GP: 0.022 D_real: 1.216 D_fake: 0.490 +(epoch: 302, iters: 4192, time: 0.064) G_GAN: 0.135 G_GAN_Feat: 0.615 G_ID: 0.126 G_Rec: 0.285 D_GP: 0.023 D_real: 1.021 D_fake: 0.865 +(epoch: 302, iters: 4592, time: 0.064) G_GAN: 0.515 G_GAN_Feat: 1.033 G_ID: 0.169 G_Rec: 0.415 D_GP: 0.100 D_real: 0.536 D_fake: 0.517 +(epoch: 302, iters: 4992, time: 0.064) G_GAN: 0.280 G_GAN_Feat: 0.697 G_ID: 0.129 G_Rec: 0.294 D_GP: 0.030 D_real: 1.134 D_fake: 0.720 +(epoch: 302, iters: 5392, time: 0.064) G_GAN: 0.234 G_GAN_Feat: 0.946 G_ID: 0.134 G_Rec: 0.409 D_GP: 0.038 D_real: 1.019 D_fake: 0.767 +(epoch: 302, iters: 5792, time: 0.064) G_GAN: 0.559 G_GAN_Feat: 1.010 G_ID: 0.138 G_Rec: 0.318 D_GP: 0.046 D_real: 1.284 D_fake: 0.610 +(epoch: 302, iters: 6192, time: 0.064) G_GAN: 0.242 G_GAN_Feat: 0.822 G_ID: 0.175 G_Rec: 0.357 D_GP: 0.027 D_real: 1.018 D_fake: 0.758 +(epoch: 302, iters: 6592, time: 0.064) G_GAN: 0.232 G_GAN_Feat: 0.806 G_ID: 0.128 G_Rec: 0.321 D_GP: 0.039 D_real: 0.844 D_fake: 0.770 +(epoch: 302, iters: 6992, time: 0.064) G_GAN: 1.149 G_GAN_Feat: 1.001 G_ID: 0.162 G_Rec: 0.452 D_GP: 0.104 D_real: 1.608 D_fake: 0.274 +(epoch: 302, iters: 7392, time: 0.064) G_GAN: 0.085 G_GAN_Feat: 0.625 G_ID: 0.155 G_Rec: 0.274 D_GP: 0.022 D_real: 0.962 D_fake: 0.915 +(epoch: 302, iters: 7792, time: 0.064) G_GAN: 0.277 G_GAN_Feat: 0.885 G_ID: 0.172 G_Rec: 0.461 D_GP: 0.028 D_real: 0.937 D_fake: 0.723 +(epoch: 302, iters: 8192, time: 0.064) G_GAN: 0.158 G_GAN_Feat: 0.660 G_ID: 0.110 G_Rec: 0.285 D_GP: 0.030 D_real: 0.966 D_fake: 0.842 +(epoch: 302, iters: 8592, time: 0.064) G_GAN: 0.542 G_GAN_Feat: 0.985 G_ID: 0.143 G_Rec: 0.421 D_GP: 0.047 D_real: 0.864 D_fake: 0.482 +(epoch: 303, iters: 384, time: 0.064) G_GAN: 0.112 G_GAN_Feat: 0.863 G_ID: 0.135 G_Rec: 0.306 D_GP: 0.048 D_real: 0.477 D_fake: 0.888 +(epoch: 303, iters: 784, time: 0.064) G_GAN: 0.722 G_GAN_Feat: 1.021 G_ID: 0.191 G_Rec: 0.410 D_GP: 0.085 D_real: 0.898 D_fake: 0.535 +(epoch: 303, iters: 1184, time: 0.064) G_GAN: 0.221 G_GAN_Feat: 0.912 G_ID: 0.098 G_Rec: 0.309 D_GP: 0.253 D_real: 0.360 D_fake: 0.792 +(epoch: 303, iters: 1584, time: 0.064) G_GAN: 0.628 G_GAN_Feat: 1.083 G_ID: 0.191 G_Rec: 0.435 D_GP: 0.042 D_real: 0.924 D_fake: 0.431 +(epoch: 303, iters: 1984, time: 0.064) G_GAN: 0.397 G_GAN_Feat: 0.968 G_ID: 0.127 G_Rec: 0.282 D_GP: 0.059 D_real: 0.499 D_fake: 0.606 +(epoch: 303, iters: 2384, time: 0.064) G_GAN: 0.586 G_GAN_Feat: 0.988 G_ID: 0.167 G_Rec: 0.426 D_GP: 0.022 D_real: 1.358 D_fake: 0.428 +(epoch: 303, iters: 2784, time: 0.064) G_GAN: -0.022 G_GAN_Feat: 0.799 G_ID: 0.153 G_Rec: 0.297 D_GP: 0.060 D_real: 0.605 D_fake: 1.022 +(epoch: 303, iters: 3184, time: 0.064) G_GAN: 0.381 G_GAN_Feat: 0.994 G_ID: 0.143 G_Rec: 0.404 D_GP: 0.163 D_real: 0.475 D_fake: 0.620 +(epoch: 303, iters: 3584, time: 0.064) G_GAN: 0.997 G_GAN_Feat: 0.787 G_ID: 0.124 G_Rec: 0.331 D_GP: 0.366 D_real: 1.473 D_fake: 0.482 +(epoch: 303, iters: 3984, time: 0.064) G_GAN: 0.320 G_GAN_Feat: 0.778 G_ID: 0.140 G_Rec: 0.383 D_GP: 0.020 D_real: 1.194 D_fake: 0.680 +(epoch: 303, iters: 4384, time: 0.064) G_GAN: 0.211 G_GAN_Feat: 0.659 G_ID: 0.131 G_Rec: 0.304 D_GP: 0.025 D_real: 1.090 D_fake: 0.789 +(epoch: 303, iters: 4784, time: 0.064) G_GAN: 0.305 G_GAN_Feat: 0.825 G_ID: 0.166 G_Rec: 0.383 D_GP: 0.034 D_real: 1.029 D_fake: 0.698 +(epoch: 303, iters: 5184, time: 0.064) G_GAN: 0.176 G_GAN_Feat: 0.688 G_ID: 0.104 G_Rec: 0.315 D_GP: 0.037 D_real: 1.027 D_fake: 0.824 +(epoch: 303, iters: 5584, time: 0.064) G_GAN: 0.249 G_GAN_Feat: 0.793 G_ID: 0.148 G_Rec: 0.360 D_GP: 0.026 D_real: 1.094 D_fake: 0.752 +(epoch: 303, iters: 5984, time: 0.064) G_GAN: 0.094 G_GAN_Feat: 0.649 G_ID: 0.141 G_Rec: 0.296 D_GP: 0.030 D_real: 1.006 D_fake: 0.907 +(epoch: 303, iters: 6384, time: 0.064) G_GAN: 0.175 G_GAN_Feat: 0.893 G_ID: 0.173 G_Rec: 0.434 D_GP: 0.082 D_real: 0.708 D_fake: 0.826 +(epoch: 303, iters: 6784, time: 0.064) G_GAN: 0.082 G_GAN_Feat: 0.697 G_ID: 0.143 G_Rec: 0.304 D_GP: 0.026 D_real: 1.050 D_fake: 0.918 +(epoch: 303, iters: 7184, time: 0.064) G_GAN: 0.323 G_GAN_Feat: 0.817 G_ID: 0.150 G_Rec: 0.346 D_GP: 0.028 D_real: 1.106 D_fake: 0.678 +(epoch: 303, iters: 7584, time: 0.064) G_GAN: 0.170 G_GAN_Feat: 0.726 G_ID: 0.117 G_Rec: 0.317 D_GP: 0.061 D_real: 1.022 D_fake: 0.830 +(epoch: 303, iters: 7984, time: 0.064) G_GAN: 0.484 G_GAN_Feat: 0.819 G_ID: 0.146 G_Rec: 0.394 D_GP: 0.025 D_real: 1.256 D_fake: 0.522 +(epoch: 303, iters: 8384, time: 0.064) G_GAN: 0.230 G_GAN_Feat: 0.777 G_ID: 0.131 G_Rec: 0.359 D_GP: 0.130 D_real: 0.780 D_fake: 0.771 +(epoch: 304, iters: 176, time: 0.064) G_GAN: 0.668 G_GAN_Feat: 0.874 G_ID: 0.171 G_Rec: 0.375 D_GP: 0.054 D_real: 1.139 D_fake: 0.358 +(epoch: 304, iters: 576, time: 0.064) G_GAN: 0.155 G_GAN_Feat: 0.543 G_ID: 0.130 G_Rec: 0.292 D_GP: 0.018 D_real: 1.161 D_fake: 0.845 +(epoch: 304, iters: 976, time: 0.064) G_GAN: 0.391 G_GAN_Feat: 0.717 G_ID: 0.141 G_Rec: 0.346 D_GP: 0.018 D_real: 1.265 D_fake: 0.609 +(epoch: 304, iters: 1376, time: 0.064) G_GAN: 0.199 G_GAN_Feat: 0.591 G_ID: 0.122 G_Rec: 0.273 D_GP: 0.024 D_real: 1.087 D_fake: 0.801 +(epoch: 304, iters: 1776, time: 0.064) G_GAN: 0.442 G_GAN_Feat: 0.774 G_ID: 0.168 G_Rec: 0.371 D_GP: 0.029 D_real: 1.114 D_fake: 0.559 +(epoch: 304, iters: 2176, time: 0.064) G_GAN: 0.249 G_GAN_Feat: 0.691 G_ID: 0.118 G_Rec: 0.301 D_GP: 0.036 D_real: 1.089 D_fake: 0.751 +(epoch: 304, iters: 2576, time: 0.064) G_GAN: 0.402 G_GAN_Feat: 0.924 G_ID: 0.147 G_Rec: 0.399 D_GP: 0.106 D_real: 0.723 D_fake: 0.607 +(epoch: 304, iters: 2976, time: 0.064) G_GAN: 0.304 G_GAN_Feat: 0.685 G_ID: 0.100 G_Rec: 0.304 D_GP: 0.076 D_real: 1.088 D_fake: 0.697 +(epoch: 304, iters: 3376, time: 0.064) G_GAN: 0.044 G_GAN_Feat: 0.862 G_ID: 0.183 G_Rec: 0.416 D_GP: 0.030 D_real: 0.736 D_fake: 0.956 +(epoch: 304, iters: 3776, time: 0.064) G_GAN: 0.129 G_GAN_Feat: 0.694 G_ID: 0.120 G_Rec: 0.295 D_GP: 0.029 D_real: 0.987 D_fake: 0.871 +(epoch: 304, iters: 4176, time: 0.064) G_GAN: 0.380 G_GAN_Feat: 0.816 G_ID: 0.166 G_Rec: 0.369 D_GP: 0.030 D_real: 1.090 D_fake: 0.624 +(epoch: 304, iters: 4576, time: 0.064) G_GAN: 0.252 G_GAN_Feat: 0.776 G_ID: 0.121 G_Rec: 0.289 D_GP: 0.088 D_real: 0.822 D_fake: 0.749 +(epoch: 304, iters: 4976, time: 0.064) G_GAN: 0.401 G_GAN_Feat: 1.402 G_ID: 0.166 G_Rec: 0.506 D_GP: 1.087 D_real: 0.459 D_fake: 0.607 +(epoch: 304, iters: 5376, time: 0.064) G_GAN: 0.155 G_GAN_Feat: 0.797 G_ID: 0.126 G_Rec: 0.319 D_GP: 0.049 D_real: 0.828 D_fake: 0.846 +(epoch: 304, iters: 5776, time: 0.064) G_GAN: 0.601 G_GAN_Feat: 0.972 G_ID: 0.140 G_Rec: 0.421 D_GP: 0.103 D_real: 0.883 D_fake: 0.416 +(epoch: 304, iters: 6176, time: 0.064) G_GAN: 0.077 G_GAN_Feat: 0.766 G_ID: 0.117 G_Rec: 0.280 D_GP: 0.078 D_real: 0.666 D_fake: 0.923 +(epoch: 304, iters: 6576, time: 0.064) G_GAN: 0.636 G_GAN_Feat: 1.085 G_ID: 0.152 G_Rec: 0.447 D_GP: 0.065 D_real: 0.721 D_fake: 0.383 +(epoch: 304, iters: 6976, time: 0.064) G_GAN: 0.113 G_GAN_Feat: 0.778 G_ID: 0.123 G_Rec: 0.286 D_GP: 0.031 D_real: 0.855 D_fake: 0.887 +(epoch: 304, iters: 7376, time: 0.064) G_GAN: 0.510 G_GAN_Feat: 0.989 G_ID: 0.157 G_Rec: 0.411 D_GP: 0.028 D_real: 1.061 D_fake: 0.494 +(epoch: 304, iters: 7776, time: 0.064) G_GAN: 0.172 G_GAN_Feat: 0.776 G_ID: 0.135 G_Rec: 0.282 D_GP: 0.032 D_real: 0.778 D_fake: 0.829 +(epoch: 304, iters: 8176, time: 0.064) G_GAN: 0.258 G_GAN_Feat: 0.887 G_ID: 0.193 G_Rec: 0.375 D_GP: 0.026 D_real: 0.952 D_fake: 0.742 +(epoch: 304, iters: 8576, time: 0.064) G_GAN: 0.278 G_GAN_Feat: 0.901 G_ID: 0.136 G_Rec: 0.307 D_GP: 0.123 D_real: 0.443 D_fake: 0.726 +(epoch: 305, iters: 368, time: 0.064) G_GAN: 0.387 G_GAN_Feat: 0.782 G_ID: 0.172 G_Rec: 0.410 D_GP: 0.019 D_real: 1.223 D_fake: 0.614 +(epoch: 305, iters: 768, time: 0.064) G_GAN: 0.192 G_GAN_Feat: 0.554 G_ID: 0.104 G_Rec: 0.272 D_GP: 0.021 D_real: 1.220 D_fake: 0.809 +(epoch: 305, iters: 1168, time: 0.064) G_GAN: 0.252 G_GAN_Feat: 0.777 G_ID: 0.155 G_Rec: 0.363 D_GP: 0.029 D_real: 0.999 D_fake: 0.748 +(epoch: 305, iters: 1568, time: 0.064) G_GAN: 0.149 G_GAN_Feat: 0.550 G_ID: 0.103 G_Rec: 0.267 D_GP: 0.018 D_real: 1.164 D_fake: 0.851 +(epoch: 305, iters: 1968, time: 0.064) G_GAN: 0.443 G_GAN_Feat: 0.773 G_ID: 0.148 G_Rec: 0.359 D_GP: 0.020 D_real: 1.265 D_fake: 0.565 +(epoch: 305, iters: 2368, time: 0.064) G_GAN: 0.094 G_GAN_Feat: 0.677 G_ID: 0.146 G_Rec: 0.308 D_GP: 0.033 D_real: 0.906 D_fake: 0.906 +(epoch: 305, iters: 2768, time: 0.064) G_GAN: 0.351 G_GAN_Feat: 0.903 G_ID: 0.140 G_Rec: 0.392 D_GP: 0.037 D_real: 0.952 D_fake: 0.649 +(epoch: 305, iters: 3168, time: 0.064) G_GAN: 0.150 G_GAN_Feat: 0.742 G_ID: 0.094 G_Rec: 0.309 D_GP: 0.125 D_real: 0.799 D_fake: 0.850 +(epoch: 305, iters: 3568, time: 0.064) G_GAN: 0.539 G_GAN_Feat: 0.905 G_ID: 0.179 G_Rec: 0.405 D_GP: 0.024 D_real: 1.246 D_fake: 0.468 +(epoch: 305, iters: 3968, time: 0.064) G_GAN: 0.311 G_GAN_Feat: 0.719 G_ID: 0.145 G_Rec: 0.304 D_GP: 0.052 D_real: 1.034 D_fake: 0.693 +(epoch: 305, iters: 4368, time: 0.064) G_GAN: 0.584 G_GAN_Feat: 0.810 G_ID: 0.160 G_Rec: 0.383 D_GP: 0.025 D_real: 1.429 D_fake: 0.426 +(epoch: 305, iters: 4768, time: 0.064) G_GAN: 0.032 G_GAN_Feat: 0.641 G_ID: 0.139 G_Rec: 0.285 D_GP: 0.046 D_real: 0.862 D_fake: 0.968 +(epoch: 305, iters: 5168, time: 0.064) G_GAN: 0.429 G_GAN_Feat: 0.796 G_ID: 0.152 G_Rec: 0.382 D_GP: 0.023 D_real: 1.152 D_fake: 0.573 +(epoch: 305, iters: 5568, time: 0.064) G_GAN: 0.338 G_GAN_Feat: 0.908 G_ID: 0.153 G_Rec: 0.346 D_GP: 0.398 D_real: 0.576 D_fake: 0.665 +(epoch: 305, iters: 5968, time: 0.064) G_GAN: 0.197 G_GAN_Feat: 0.897 G_ID: 0.179 G_Rec: 0.414 D_GP: 0.025 D_real: 0.875 D_fake: 0.803 +(epoch: 305, iters: 6368, time: 0.064) G_GAN: 0.248 G_GAN_Feat: 0.727 G_ID: 0.119 G_Rec: 0.310 D_GP: 0.022 D_real: 1.144 D_fake: 0.752 +(epoch: 305, iters: 6768, time: 0.064) G_GAN: 0.489 G_GAN_Feat: 0.792 G_ID: 0.166 G_Rec: 0.351 D_GP: 0.021 D_real: 1.260 D_fake: 0.520 +(epoch: 305, iters: 7168, time: 0.064) G_GAN: 0.088 G_GAN_Feat: 0.787 G_ID: 0.118 G_Rec: 0.288 D_GP: 0.060 D_real: 0.486 D_fake: 0.915 +(epoch: 305, iters: 7568, time: 0.064) G_GAN: 0.494 G_GAN_Feat: 0.839 G_ID: 0.168 G_Rec: 0.375 D_GP: 0.023 D_real: 1.232 D_fake: 0.520 +(epoch: 305, iters: 7968, time: 0.064) G_GAN: 0.403 G_GAN_Feat: 0.849 G_ID: 0.135 G_Rec: 0.325 D_GP: 0.070 D_real: 0.853 D_fake: 0.604 +(epoch: 305, iters: 8368, time: 0.064) G_GAN: 0.433 G_GAN_Feat: 0.919 G_ID: 0.172 G_Rec: 0.357 D_GP: 0.027 D_real: 1.199 D_fake: 0.570 +(epoch: 306, iters: 160, time: 0.064) G_GAN: 0.160 G_GAN_Feat: 0.733 G_ID: 0.109 G_Rec: 0.286 D_GP: 0.024 D_real: 0.988 D_fake: 0.840 +(epoch: 306, iters: 560, time: 0.064) G_GAN: 0.990 G_GAN_Feat: 1.099 G_ID: 0.143 G_Rec: 0.454 D_GP: 0.024 D_real: 1.433 D_fake: 0.251 +(epoch: 306, iters: 960, time: 0.064) G_GAN: 0.366 G_GAN_Feat: 1.023 G_ID: 0.132 G_Rec: 0.323 D_GP: 0.629 D_real: 0.455 D_fake: 0.637 +(epoch: 306, iters: 1360, time: 0.064) G_GAN: 0.354 G_GAN_Feat: 0.856 G_ID: 0.150 G_Rec: 0.348 D_GP: 0.021 D_real: 1.042 D_fake: 0.651 +(epoch: 306, iters: 1760, time: 0.064) G_GAN: 0.145 G_GAN_Feat: 0.727 G_ID: 0.111 G_Rec: 0.285 D_GP: 0.027 D_real: 0.979 D_fake: 0.855 +(epoch: 306, iters: 2160, time: 0.064) G_GAN: 0.247 G_GAN_Feat: 0.756 G_ID: 0.188 G_Rec: 0.320 D_GP: 0.025 D_real: 1.103 D_fake: 0.754 +(epoch: 306, iters: 2560, time: 0.064) G_GAN: 0.215 G_GAN_Feat: 0.680 G_ID: 0.125 G_Rec: 0.288 D_GP: 0.023 D_real: 1.173 D_fake: 0.786 +(epoch: 306, iters: 2960, time: 0.064) G_GAN: 0.578 G_GAN_Feat: 1.235 G_ID: 0.171 G_Rec: 0.429 D_GP: 0.068 D_real: 0.527 D_fake: 0.432 +(epoch: 306, iters: 3360, time: 0.064) G_GAN: 0.192 G_GAN_Feat: 0.535 G_ID: 0.129 G_Rec: 0.306 D_GP: 0.018 D_real: 1.198 D_fake: 0.808 +(epoch: 306, iters: 3760, time: 0.064) G_GAN: 0.425 G_GAN_Feat: 0.739 G_ID: 0.159 G_Rec: 0.388 D_GP: 0.019 D_real: 1.257 D_fake: 0.577 +(epoch: 306, iters: 4160, time: 0.064) G_GAN: 0.186 G_GAN_Feat: 0.558 G_ID: 0.125 G_Rec: 0.287 D_GP: 0.032 D_real: 1.108 D_fake: 0.814 +(epoch: 306, iters: 4560, time: 0.064) G_GAN: 0.418 G_GAN_Feat: 0.715 G_ID: 0.155 G_Rec: 0.378 D_GP: 0.022 D_real: 1.173 D_fake: 0.583 +(epoch: 306, iters: 4960, time: 0.064) G_GAN: 0.290 G_GAN_Feat: 0.557 G_ID: 0.120 G_Rec: 0.285 D_GP: 0.024 D_real: 1.260 D_fake: 0.711 +(epoch: 306, iters: 5360, time: 0.064) G_GAN: 0.304 G_GAN_Feat: 0.737 G_ID: 0.157 G_Rec: 0.373 D_GP: 0.024 D_real: 1.098 D_fake: 0.697 +(epoch: 306, iters: 5760, time: 0.064) G_GAN: 0.060 G_GAN_Feat: 0.553 G_ID: 0.134 G_Rec: 0.266 D_GP: 0.024 D_real: 0.988 D_fake: 0.940 +(epoch: 306, iters: 6160, time: 0.064) G_GAN: 0.370 G_GAN_Feat: 0.808 G_ID: 0.199 G_Rec: 0.387 D_GP: 0.035 D_real: 0.996 D_fake: 0.636 +(epoch: 306, iters: 6560, time: 0.064) G_GAN: 0.033 G_GAN_Feat: 0.587 G_ID: 0.114 G_Rec: 0.260 D_GP: 0.038 D_real: 0.913 D_fake: 0.967 +(epoch: 306, iters: 6960, time: 0.064) G_GAN: 0.267 G_GAN_Feat: 0.790 G_ID: 0.145 G_Rec: 0.389 D_GP: 0.046 D_real: 1.007 D_fake: 0.736 +(epoch: 306, iters: 7360, time: 0.064) G_GAN: 0.121 G_GAN_Feat: 0.676 G_ID: 0.137 G_Rec: 0.304 D_GP: 0.034 D_real: 1.017 D_fake: 0.879 +(epoch: 306, iters: 7760, time: 0.064) G_GAN: 0.262 G_GAN_Feat: 0.815 G_ID: 0.150 G_Rec: 0.375 D_GP: 0.030 D_real: 0.907 D_fake: 0.739 +(epoch: 306, iters: 8160, time: 0.064) G_GAN: 0.310 G_GAN_Feat: 0.658 G_ID: 0.119 G_Rec: 0.311 D_GP: 0.038 D_real: 1.142 D_fake: 0.691 +(epoch: 306, iters: 8560, time: 0.064) G_GAN: 0.518 G_GAN_Feat: 0.891 G_ID: 0.168 G_Rec: 0.378 D_GP: 0.037 D_real: 1.102 D_fake: 0.489 +(epoch: 307, iters: 352, time: 0.064) G_GAN: 0.174 G_GAN_Feat: 0.644 G_ID: 0.121 G_Rec: 0.317 D_GP: 0.024 D_real: 1.000 D_fake: 0.827 +(epoch: 307, iters: 752, time: 0.064) G_GAN: 0.546 G_GAN_Feat: 0.814 G_ID: 0.158 G_Rec: 0.382 D_GP: 0.024 D_real: 1.273 D_fake: 0.457 +(epoch: 307, iters: 1152, time: 0.064) G_GAN: -0.025 G_GAN_Feat: 0.784 G_ID: 0.158 G_Rec: 0.334 D_GP: 0.182 D_real: 0.506 D_fake: 1.025 +(epoch: 307, iters: 1552, time: 0.064) G_GAN: 0.225 G_GAN_Feat: 1.010 G_ID: 0.153 G_Rec: 0.397 D_GP: 0.219 D_real: 0.335 D_fake: 0.780 +(epoch: 307, iters: 1952, time: 0.064) G_GAN: 0.279 G_GAN_Feat: 0.718 G_ID: 0.115 G_Rec: 0.260 D_GP: 0.070 D_real: 0.815 D_fake: 0.722 +(epoch: 307, iters: 2352, time: 0.064) G_GAN: 0.221 G_GAN_Feat: 0.888 G_ID: 0.160 G_Rec: 0.332 D_GP: 0.031 D_real: 0.985 D_fake: 0.781 +(epoch: 307, iters: 2752, time: 0.064) G_GAN: 0.167 G_GAN_Feat: 0.841 G_ID: 0.123 G_Rec: 0.307 D_GP: 0.024 D_real: 0.909 D_fake: 0.833 +(epoch: 307, iters: 3152, time: 0.064) G_GAN: -0.090 G_GAN_Feat: 1.281 G_ID: 0.148 G_Rec: 0.450 D_GP: 0.051 D_real: 1.212 D_fake: 1.093 +(epoch: 307, iters: 3552, time: 0.064) G_GAN: -0.280 G_GAN_Feat: 0.760 G_ID: 0.123 G_Rec: 0.318 D_GP: 0.020 D_real: 0.644 D_fake: 1.280 +(epoch: 307, iters: 3952, time: 0.064) G_GAN: 0.097 G_GAN_Feat: 0.789 G_ID: 0.156 G_Rec: 0.359 D_GP: 0.021 D_real: 0.939 D_fake: 0.904 +(epoch: 307, iters: 4352, time: 0.064) G_GAN: -0.034 G_GAN_Feat: 0.663 G_ID: 0.112 G_Rec: 0.308 D_GP: 0.023 D_real: 0.855 D_fake: 1.034 +(epoch: 307, iters: 4752, time: 0.064) G_GAN: 0.052 G_GAN_Feat: 0.840 G_ID: 0.184 G_Rec: 0.342 D_GP: 0.038 D_real: 0.784 D_fake: 0.949 +(epoch: 307, iters: 5152, time: 0.064) G_GAN: 0.100 G_GAN_Feat: 0.875 G_ID: 0.135 G_Rec: 0.312 D_GP: 0.201 D_real: 0.469 D_fake: 0.901 +(epoch: 307, iters: 5552, time: 0.064) G_GAN: 0.574 G_GAN_Feat: 0.907 G_ID: 0.151 G_Rec: 0.430 D_GP: 0.043 D_real: 1.206 D_fake: 0.439 +(epoch: 307, iters: 5952, time: 0.064) G_GAN: 0.339 G_GAN_Feat: 0.614 G_ID: 0.119 G_Rec: 0.269 D_GP: 0.021 D_real: 1.308 D_fake: 0.661 +(epoch: 307, iters: 6352, time: 0.064) G_GAN: 0.520 G_GAN_Feat: 1.062 G_ID: 0.150 G_Rec: 0.440 D_GP: 0.173 D_real: 0.592 D_fake: 0.494 +(epoch: 307, iters: 6752, time: 0.064) G_GAN: 0.392 G_GAN_Feat: 0.897 G_ID: 0.118 G_Rec: 0.318 D_GP: 0.024 D_real: 0.789 D_fake: 0.609 +(epoch: 307, iters: 7152, time: 0.064) G_GAN: 0.252 G_GAN_Feat: 0.965 G_ID: 0.160 G_Rec: 0.396 D_GP: 0.109 D_real: 0.674 D_fake: 0.754 +(epoch: 307, iters: 7552, time: 0.064) G_GAN: 0.380 G_GAN_Feat: 0.739 G_ID: 0.121 G_Rec: 0.311 D_GP: 0.023 D_real: 1.253 D_fake: 0.636 +(epoch: 307, iters: 7952, time: 0.064) G_GAN: 0.428 G_GAN_Feat: 0.959 G_ID: 0.161 G_Rec: 0.382 D_GP: 0.032 D_real: 0.937 D_fake: 0.572 +(epoch: 307, iters: 8352, time: 0.064) G_GAN: 0.326 G_GAN_Feat: 0.787 G_ID: 0.134 G_Rec: 0.268 D_GP: 0.032 D_real: 0.846 D_fake: 0.675 +(epoch: 308, iters: 144, time: 0.064) G_GAN: 0.476 G_GAN_Feat: 0.824 G_ID: 0.145 G_Rec: 0.374 D_GP: 0.021 D_real: 1.245 D_fake: 0.531 +(epoch: 308, iters: 544, time: 0.064) G_GAN: 0.203 G_GAN_Feat: 0.644 G_ID: 0.156 G_Rec: 0.303 D_GP: 0.018 D_real: 1.139 D_fake: 0.797 +(epoch: 308, iters: 944, time: 0.064) G_GAN: 0.385 G_GAN_Feat: 0.883 G_ID: 0.151 G_Rec: 0.382 D_GP: 0.026 D_real: 1.085 D_fake: 0.615 +(epoch: 308, iters: 1344, time: 0.064) G_GAN: 0.358 G_GAN_Feat: 0.620 G_ID: 0.115 G_Rec: 0.290 D_GP: 0.024 D_real: 1.269 D_fake: 0.642 +(epoch: 308, iters: 1744, time: 0.064) G_GAN: 0.333 G_GAN_Feat: 0.788 G_ID: 0.151 G_Rec: 0.351 D_GP: 0.031 D_real: 1.105 D_fake: 0.667 +(epoch: 308, iters: 2144, time: 0.064) G_GAN: -0.406 G_GAN_Feat: 0.635 G_ID: 0.140 G_Rec: 0.316 D_GP: 0.020 D_real: 0.609 D_fake: 1.406 +(epoch: 308, iters: 2544, time: 0.064) G_GAN: 0.065 G_GAN_Feat: 0.767 G_ID: 0.149 G_Rec: 0.355 D_GP: 0.021 D_real: 0.813 D_fake: 0.937 +(epoch: 308, iters: 2944, time: 0.064) G_GAN: -0.096 G_GAN_Feat: 0.602 G_ID: 0.122 G_Rec: 0.267 D_GP: 0.022 D_real: 0.834 D_fake: 1.099 +(epoch: 308, iters: 3344, time: 0.064) G_GAN: 0.010 G_GAN_Feat: 0.909 G_ID: 0.169 G_Rec: 0.399 D_GP: 0.049 D_real: 0.559 D_fake: 0.990 +(epoch: 308, iters: 3744, time: 0.064) G_GAN: 0.058 G_GAN_Feat: 0.669 G_ID: 0.130 G_Rec: 0.279 D_GP: 0.037 D_real: 0.834 D_fake: 0.942 +(epoch: 308, iters: 4144, time: 0.064) G_GAN: 0.517 G_GAN_Feat: 0.871 G_ID: 0.135 G_Rec: 0.379 D_GP: 0.042 D_real: 0.999 D_fake: 0.489 +(epoch: 308, iters: 4544, time: 0.064) G_GAN: 0.466 G_GAN_Feat: 0.715 G_ID: 0.121 G_Rec: 0.283 D_GP: 0.037 D_real: 1.242 D_fake: 0.543 +(epoch: 308, iters: 4944, time: 0.064) G_GAN: 0.289 G_GAN_Feat: 0.850 G_ID: 0.169 G_Rec: 0.431 D_GP: 0.025 D_real: 1.101 D_fake: 0.711 +(epoch: 308, iters: 5344, time: 0.064) G_GAN: 0.218 G_GAN_Feat: 0.618 G_ID: 0.114 G_Rec: 0.275 D_GP: 0.021 D_real: 1.133 D_fake: 0.782 +(epoch: 308, iters: 5744, time: 0.064) G_GAN: 0.294 G_GAN_Feat: 0.873 G_ID: 0.157 G_Rec: 0.387 D_GP: 0.036 D_real: 0.952 D_fake: 0.710 +(epoch: 308, iters: 6144, time: 0.064) G_GAN: 0.169 G_GAN_Feat: 0.611 G_ID: 0.131 G_Rec: 0.264 D_GP: 0.025 D_real: 1.046 D_fake: 0.832 +(epoch: 308, iters: 6544, time: 0.064) G_GAN: 0.506 G_GAN_Feat: 0.873 G_ID: 0.163 G_Rec: 0.419 D_GP: 0.032 D_real: 1.177 D_fake: 0.511 +(epoch: 308, iters: 6944, time: 0.064) G_GAN: 0.124 G_GAN_Feat: 0.627 G_ID: 0.114 G_Rec: 0.291 D_GP: 0.023 D_real: 1.027 D_fake: 0.876 +(epoch: 308, iters: 7344, time: 0.064) G_GAN: 0.428 G_GAN_Feat: 0.769 G_ID: 0.139 G_Rec: 0.358 D_GP: 0.024 D_real: 1.137 D_fake: 0.574 +(epoch: 308, iters: 7744, time: 0.064) G_GAN: 0.107 G_GAN_Feat: 0.578 G_ID: 0.121 G_Rec: 0.275 D_GP: 0.023 D_real: 1.028 D_fake: 0.893 +(epoch: 308, iters: 8144, time: 0.064) G_GAN: 0.491 G_GAN_Feat: 0.842 G_ID: 0.170 G_Rec: 0.369 D_GP: 0.033 D_real: 1.141 D_fake: 0.516 +(epoch: 308, iters: 8544, time: 0.064) G_GAN: 0.137 G_GAN_Feat: 0.652 G_ID: 0.119 G_Rec: 0.278 D_GP: 0.052 D_real: 0.941 D_fake: 0.863 +(epoch: 309, iters: 336, time: 0.064) G_GAN: 0.484 G_GAN_Feat: 0.971 G_ID: 0.161 G_Rec: 0.450 D_GP: 0.153 D_real: 0.790 D_fake: 0.522 +(epoch: 309, iters: 736, time: 0.064) G_GAN: 0.111 G_GAN_Feat: 0.703 G_ID: 0.124 G_Rec: 0.283 D_GP: 0.066 D_real: 0.799 D_fake: 0.889 +(epoch: 309, iters: 1136, time: 0.064) G_GAN: 0.565 G_GAN_Feat: 0.887 G_ID: 0.147 G_Rec: 0.392 D_GP: 0.040 D_real: 1.222 D_fake: 0.441 +(epoch: 309, iters: 1536, time: 0.064) G_GAN: 0.377 G_GAN_Feat: 0.714 G_ID: 0.119 G_Rec: 0.260 D_GP: 0.032 D_real: 1.204 D_fake: 0.624 +(epoch: 309, iters: 1936, time: 0.064) G_GAN: 0.363 G_GAN_Feat: 0.876 G_ID: 0.160 G_Rec: 0.384 D_GP: 0.025 D_real: 1.040 D_fake: 0.637 +(epoch: 309, iters: 2336, time: 0.064) G_GAN: 0.187 G_GAN_Feat: 0.861 G_ID: 0.128 G_Rec: 0.323 D_GP: 0.169 D_real: 0.598 D_fake: 0.817 +(epoch: 309, iters: 2736, time: 0.064) G_GAN: 0.459 G_GAN_Feat: 0.934 G_ID: 0.170 G_Rec: 0.404 D_GP: 0.039 D_real: 1.229 D_fake: 0.544 +(epoch: 309, iters: 3136, time: 0.064) G_GAN: 0.633 G_GAN_Feat: 0.861 G_ID: 0.126 G_Rec: 0.310 D_GP: 0.215 D_real: 0.899 D_fake: 0.557 +(epoch: 309, iters: 3536, time: 0.064) G_GAN: 0.413 G_GAN_Feat: 0.935 G_ID: 0.148 G_Rec: 0.388 D_GP: 0.033 D_real: 1.126 D_fake: 0.603 +(epoch: 309, iters: 3936, time: 0.064) G_GAN: 0.349 G_GAN_Feat: 0.670 G_ID: 0.124 G_Rec: 0.292 D_GP: 0.025 D_real: 1.164 D_fake: 0.652 +(epoch: 309, iters: 4336, time: 0.064) G_GAN: 0.454 G_GAN_Feat: 1.086 G_ID: 0.158 G_Rec: 0.436 D_GP: 0.168 D_real: 0.509 D_fake: 0.552 +(epoch: 309, iters: 4736, time: 0.064) G_GAN: 0.455 G_GAN_Feat: 0.767 G_ID: 0.109 G_Rec: 0.297 D_GP: 0.039 D_real: 1.044 D_fake: 0.550 +(epoch: 309, iters: 5136, time: 0.064) G_GAN: 0.366 G_GAN_Feat: 0.850 G_ID: 0.146 G_Rec: 0.447 D_GP: 0.023 D_real: 1.177 D_fake: 0.643 +(epoch: 309, iters: 5536, time: 0.064) G_GAN: -0.072 G_GAN_Feat: 0.664 G_ID: 0.107 G_Rec: 0.292 D_GP: 0.019 D_real: 0.816 D_fake: 1.072 +(epoch: 309, iters: 5936, time: 0.064) G_GAN: 0.217 G_GAN_Feat: 0.828 G_ID: 0.136 G_Rec: 0.364 D_GP: 0.025 D_real: 0.933 D_fake: 0.785 +(epoch: 309, iters: 6336, time: 0.064) G_GAN: 0.081 G_GAN_Feat: 0.654 G_ID: 0.126 G_Rec: 0.287 D_GP: 0.025 D_real: 0.965 D_fake: 0.919 +(epoch: 309, iters: 6736, time: 0.064) G_GAN: 0.250 G_GAN_Feat: 0.937 G_ID: 0.162 G_Rec: 0.386 D_GP: 0.028 D_real: 0.806 D_fake: 0.750 +(epoch: 309, iters: 7136, time: 0.064) G_GAN: 0.314 G_GAN_Feat: 0.771 G_ID: 0.121 G_Rec: 0.308 D_GP: 0.031 D_real: 0.920 D_fake: 0.687 +(epoch: 309, iters: 7536, time: 0.064) G_GAN: 0.442 G_GAN_Feat: 0.872 G_ID: 0.166 G_Rec: 0.358 D_GP: 0.023 D_real: 1.220 D_fake: 0.560 +(epoch: 309, iters: 7936, time: 0.064) G_GAN: 0.192 G_GAN_Feat: 0.616 G_ID: 0.156 G_Rec: 0.262 D_GP: 0.023 D_real: 1.053 D_fake: 0.808 +(epoch: 309, iters: 8336, time: 0.064) G_GAN: 0.421 G_GAN_Feat: 0.946 G_ID: 0.162 G_Rec: 0.359 D_GP: 0.053 D_real: 0.886 D_fake: 0.584 +(epoch: 310, iters: 128, time: 0.064) G_GAN: 0.272 G_GAN_Feat: 0.757 G_ID: 0.138 G_Rec: 0.303 D_GP: 0.053 D_real: 1.131 D_fake: 0.729 +(epoch: 310, iters: 528, time: 0.064) G_GAN: 0.245 G_GAN_Feat: 1.012 G_ID: 0.144 G_Rec: 0.418 D_GP: 0.135 D_real: 0.522 D_fake: 0.757 +(epoch: 310, iters: 928, time: 0.064) G_GAN: 0.379 G_GAN_Feat: 0.716 G_ID: 0.129 G_Rec: 0.290 D_GP: 0.030 D_real: 1.193 D_fake: 0.623 +(epoch: 310, iters: 1328, time: 0.064) G_GAN: 0.797 G_GAN_Feat: 1.103 G_ID: 0.167 G_Rec: 0.428 D_GP: 0.343 D_real: 0.508 D_fake: 0.352 +(epoch: 310, iters: 1728, time: 0.064) G_GAN: 0.340 G_GAN_Feat: 0.703 G_ID: 0.139 G_Rec: 0.294 D_GP: 0.028 D_real: 1.247 D_fake: 0.661 +(epoch: 310, iters: 2128, time: 0.064) G_GAN: 0.324 G_GAN_Feat: 0.751 G_ID: 0.145 G_Rec: 0.313 D_GP: 0.022 D_real: 1.231 D_fake: 0.677 +(epoch: 310, iters: 2528, time: 0.064) G_GAN: 0.320 G_GAN_Feat: 0.806 G_ID: 0.115 G_Rec: 0.302 D_GP: 0.047 D_real: 1.154 D_fake: 0.686 +(epoch: 310, iters: 2928, time: 0.064) G_GAN: 0.690 G_GAN_Feat: 0.983 G_ID: 0.153 G_Rec: 0.377 D_GP: 0.038 D_real: 1.078 D_fake: 0.367 +(epoch: 310, iters: 3328, time: 0.064) G_GAN: 0.338 G_GAN_Feat: 0.701 G_ID: 0.102 G_Rec: 0.265 D_GP: 0.032 D_real: 1.091 D_fake: 0.662 +(epoch: 310, iters: 3728, time: 0.064) G_GAN: 0.851 G_GAN_Feat: 0.918 G_ID: 0.152 G_Rec: 0.408 D_GP: 0.068 D_real: 1.521 D_fake: 0.220 +(epoch: 310, iters: 4128, time: 0.064) G_GAN: 0.110 G_GAN_Feat: 0.621 G_ID: 0.126 G_Rec: 0.267 D_GP: 0.022 D_real: 1.008 D_fake: 0.890 +(epoch: 310, iters: 4528, time: 0.064) G_GAN: 0.361 G_GAN_Feat: 0.883 G_ID: 0.177 G_Rec: 0.410 D_GP: 0.025 D_real: 1.013 D_fake: 0.646 +(epoch: 310, iters: 4928, time: 0.064) G_GAN: 0.185 G_GAN_Feat: 0.792 G_ID: 0.138 G_Rec: 0.292 D_GP: 0.067 D_real: 0.836 D_fake: 0.819 +(epoch: 310, iters: 5328, time: 0.064) G_GAN: 0.727 G_GAN_Feat: 0.954 G_ID: 0.158 G_Rec: 0.402 D_GP: 0.021 D_real: 1.398 D_fake: 0.289 +(epoch: 310, iters: 5728, time: 0.064) G_GAN: 0.216 G_GAN_Feat: 0.703 G_ID: 0.123 G_Rec: 0.281 D_GP: 0.019 D_real: 1.095 D_fake: 0.784 +(epoch: 310, iters: 6128, time: 0.064) G_GAN: 0.579 G_GAN_Feat: 0.977 G_ID: 0.178 G_Rec: 0.406 D_GP: 0.026 D_real: 1.051 D_fake: 0.425 +(epoch: 310, iters: 6528, time: 0.064) G_GAN: 0.255 G_GAN_Feat: 0.595 G_ID: 0.120 G_Rec: 0.289 D_GP: 0.020 D_real: 1.301 D_fake: 0.746 +(epoch: 310, iters: 6928, time: 0.064) G_GAN: 0.420 G_GAN_Feat: 0.847 G_ID: 0.165 G_Rec: 0.491 D_GP: 0.031 D_real: 1.105 D_fake: 0.583 +(epoch: 310, iters: 7328, time: 0.064) G_GAN: 0.102 G_GAN_Feat: 0.661 G_ID: 0.147 G_Rec: 0.297 D_GP: 0.029 D_real: 0.989 D_fake: 0.898 +(epoch: 310, iters: 7728, time: 0.064) G_GAN: 0.310 G_GAN_Feat: 0.824 G_ID: 0.144 G_Rec: 0.391 D_GP: 0.032 D_real: 0.961 D_fake: 0.692 +(epoch: 310, iters: 8128, time: 0.064) G_GAN: 0.183 G_GAN_Feat: 0.655 G_ID: 0.114 G_Rec: 0.298 D_GP: 0.023 D_real: 1.121 D_fake: 0.817 +(epoch: 310, iters: 8528, time: 0.064) G_GAN: 0.567 G_GAN_Feat: 0.879 G_ID: 0.152 G_Rec: 0.399 D_GP: 0.039 D_real: 1.284 D_fake: 0.449 +(epoch: 311, iters: 320, time: 0.064) G_GAN: 0.246 G_GAN_Feat: 0.622 G_ID: 0.112 G_Rec: 0.289 D_GP: 0.019 D_real: 1.181 D_fake: 0.754 +(epoch: 311, iters: 720, time: 0.064) G_GAN: 0.510 G_GAN_Feat: 0.934 G_ID: 0.145 G_Rec: 0.375 D_GP: 0.085 D_real: 0.875 D_fake: 0.510 +(epoch: 311, iters: 1120, time: 0.064) G_GAN: 0.112 G_GAN_Feat: 0.826 G_ID: 0.129 G_Rec: 0.296 D_GP: 0.061 D_real: 0.463 D_fake: 0.889 +(epoch: 311, iters: 1520, time: 0.064) G_GAN: 0.402 G_GAN_Feat: 0.794 G_ID: 0.170 G_Rec: 0.386 D_GP: 0.017 D_real: 1.261 D_fake: 0.604 +(epoch: 311, iters: 1920, time: 0.064) G_GAN: 0.187 G_GAN_Feat: 0.662 G_ID: 0.110 G_Rec: 0.336 D_GP: 0.021 D_real: 1.123 D_fake: 0.814 +(epoch: 311, iters: 2320, time: 0.064) G_GAN: 0.420 G_GAN_Feat: 0.745 G_ID: 0.162 G_Rec: 0.362 D_GP: 0.019 D_real: 1.273 D_fake: 0.580 +(epoch: 311, iters: 2720, time: 0.064) G_GAN: 0.211 G_GAN_Feat: 0.577 G_ID: 0.116 G_Rec: 0.299 D_GP: 0.024 D_real: 1.156 D_fake: 0.789 +(epoch: 311, iters: 3120, time: 0.064) G_GAN: 0.360 G_GAN_Feat: 0.729 G_ID: 0.150 G_Rec: 0.337 D_GP: 0.023 D_real: 1.204 D_fake: 0.640 +(epoch: 311, iters: 3520, time: 0.064) G_GAN: 0.011 G_GAN_Feat: 0.700 G_ID: 0.136 G_Rec: 0.327 D_GP: 0.104 D_real: 0.881 D_fake: 0.989 +(epoch: 311, iters: 3920, time: 0.064) G_GAN: 0.353 G_GAN_Feat: 0.747 G_ID: 0.165 G_Rec: 0.346 D_GP: 0.024 D_real: 1.159 D_fake: 0.650 +(epoch: 311, iters: 4320, time: 0.064) G_GAN: 0.174 G_GAN_Feat: 0.597 G_ID: 0.130 G_Rec: 0.268 D_GP: 0.023 D_real: 1.122 D_fake: 0.826 +(epoch: 311, iters: 4720, time: 0.064) G_GAN: 0.345 G_GAN_Feat: 0.799 G_ID: 0.161 G_Rec: 0.370 D_GP: 0.029 D_real: 1.108 D_fake: 0.655 +(epoch: 311, iters: 5120, time: 0.064) G_GAN: 0.064 G_GAN_Feat: 0.755 G_ID: 0.119 G_Rec: 0.306 D_GP: 0.242 D_real: 0.525 D_fake: 0.936 +(epoch: 311, iters: 5520, time: 0.064) G_GAN: 0.641 G_GAN_Feat: 0.854 G_ID: 0.161 G_Rec: 0.396 D_GP: 0.023 D_real: 1.384 D_fake: 0.369 +(epoch: 311, iters: 5920, time: 0.064) G_GAN: 0.305 G_GAN_Feat: 0.744 G_ID: 0.097 G_Rec: 0.315 D_GP: 0.100 D_real: 0.963 D_fake: 0.695 +(epoch: 311, iters: 6320, time: 0.064) G_GAN: 0.470 G_GAN_Feat: 0.833 G_ID: 0.168 G_Rec: 0.360 D_GP: 0.025 D_real: 1.207 D_fake: 0.538 +(epoch: 311, iters: 6720, time: 0.064) G_GAN: 0.270 G_GAN_Feat: 0.764 G_ID: 0.109 G_Rec: 0.332 D_GP: 0.038 D_real: 0.977 D_fake: 0.737 +(epoch: 311, iters: 7120, time: 0.064) G_GAN: 0.395 G_GAN_Feat: 0.873 G_ID: 0.169 G_Rec: 0.425 D_GP: 0.051 D_real: 0.975 D_fake: 0.605 +(epoch: 311, iters: 7520, time: 0.064) G_GAN: 0.252 G_GAN_Feat: 0.832 G_ID: 0.113 G_Rec: 0.298 D_GP: 0.051 D_real: 0.835 D_fake: 0.797 +(epoch: 311, iters: 7920, time: 0.064) G_GAN: 0.530 G_GAN_Feat: 0.946 G_ID: 0.156 G_Rec: 0.388 D_GP: 0.030 D_real: 1.207 D_fake: 0.475 +(epoch: 311, iters: 8320, time: 0.064) G_GAN: 0.200 G_GAN_Feat: 0.568 G_ID: 0.111 G_Rec: 0.241 D_GP: 0.018 D_real: 1.157 D_fake: 0.800 +(epoch: 312, iters: 112, time: 0.064) G_GAN: 0.325 G_GAN_Feat: 0.843 G_ID: 0.154 G_Rec: 0.371 D_GP: 0.037 D_real: 1.051 D_fake: 0.680 +(epoch: 312, iters: 512, time: 0.064) G_GAN: 0.220 G_GAN_Feat: 0.664 G_ID: 0.103 G_Rec: 0.282 D_GP: 0.059 D_real: 1.140 D_fake: 0.781 +(epoch: 312, iters: 912, time: 0.064) G_GAN: 0.593 G_GAN_Feat: 0.893 G_ID: 0.159 G_Rec: 0.391 D_GP: 0.022 D_real: 1.352 D_fake: 0.410 +(epoch: 312, iters: 1312, time: 0.064) G_GAN: 0.455 G_GAN_Feat: 0.630 G_ID: 0.113 G_Rec: 0.272 D_GP: 0.017 D_real: 1.369 D_fake: 0.545 +(epoch: 312, iters: 1712, time: 0.064) G_GAN: 0.424 G_GAN_Feat: 1.030 G_ID: 0.148 G_Rec: 0.419 D_GP: 0.151 D_real: 0.559 D_fake: 0.583 +(epoch: 312, iters: 2112, time: 0.064) G_GAN: 0.130 G_GAN_Feat: 0.733 G_ID: 0.111 G_Rec: 0.307 D_GP: 0.023 D_real: 0.941 D_fake: 0.870 +(epoch: 312, iters: 2512, time: 0.064) G_GAN: 0.078 G_GAN_Feat: 0.752 G_ID: 0.170 G_Rec: 0.340 D_GP: 0.021 D_real: 0.984 D_fake: 0.922 +(epoch: 312, iters: 2912, time: 0.064) G_GAN: 0.219 G_GAN_Feat: 0.685 G_ID: 0.117 G_Rec: 0.282 D_GP: 0.046 D_real: 1.098 D_fake: 0.782 +(epoch: 312, iters: 3312, time: 0.064) G_GAN: 0.483 G_GAN_Feat: 1.071 G_ID: 0.159 G_Rec: 0.416 D_GP: 0.555 D_real: 0.662 D_fake: 0.538 +(epoch: 312, iters: 3712, time: 0.064) G_GAN: 0.217 G_GAN_Feat: 0.672 G_ID: 0.100 G_Rec: 0.274 D_GP: 0.023 D_real: 1.089 D_fake: 0.783 +(epoch: 312, iters: 4112, time: 0.064) G_GAN: 0.675 G_GAN_Feat: 1.069 G_ID: 0.162 G_Rec: 0.439 D_GP: 0.181 D_real: 1.007 D_fake: 0.471 +(epoch: 312, iters: 4512, time: 0.064) G_GAN: 0.217 G_GAN_Feat: 0.846 G_ID: 0.135 G_Rec: 0.336 D_GP: 0.082 D_real: 0.558 D_fake: 0.787 +(epoch: 312, iters: 4912, time: 0.064) G_GAN: 0.308 G_GAN_Feat: 0.969 G_ID: 0.169 G_Rec: 0.381 D_GP: 0.063 D_real: 0.748 D_fake: 0.694 +(epoch: 312, iters: 5312, time: 0.064) G_GAN: 0.309 G_GAN_Feat: 0.709 G_ID: 0.151 G_Rec: 0.292 D_GP: 0.029 D_real: 1.101 D_fake: 0.691 +(epoch: 312, iters: 5712, time: 0.064) G_GAN: 0.281 G_GAN_Feat: 0.763 G_ID: 0.162 G_Rec: 0.427 D_GP: 0.020 D_real: 1.104 D_fake: 0.723 +(epoch: 312, iters: 6112, time: 0.064) G_GAN: 0.113 G_GAN_Feat: 0.503 G_ID: 0.109 G_Rec: 0.299 D_GP: 0.020 D_real: 1.057 D_fake: 0.887 +(epoch: 312, iters: 6512, time: 0.064) G_GAN: -0.018 G_GAN_Feat: 0.677 G_ID: 0.152 G_Rec: 0.354 D_GP: 0.021 D_real: 0.821 D_fake: 1.018 +(epoch: 312, iters: 6912, time: 0.064) G_GAN: 0.183 G_GAN_Feat: 0.505 G_ID: 0.128 G_Rec: 0.285 D_GP: 0.021 D_real: 1.135 D_fake: 0.817 +(epoch: 312, iters: 7312, time: 0.064) G_GAN: 0.296 G_GAN_Feat: 0.715 G_ID: 0.148 G_Rec: 0.373 D_GP: 0.021 D_real: 1.032 D_fake: 0.711 +(epoch: 312, iters: 7712, time: 0.064) G_GAN: 0.091 G_GAN_Feat: 0.545 G_ID: 0.113 G_Rec: 0.273 D_GP: 0.022 D_real: 1.037 D_fake: 0.909 +(epoch: 312, iters: 8112, time: 0.064) G_GAN: 0.242 G_GAN_Feat: 0.745 G_ID: 0.163 G_Rec: 0.370 D_GP: 0.026 D_real: 0.966 D_fake: 0.760 +(epoch: 312, iters: 8512, time: 0.064) G_GAN: 0.065 G_GAN_Feat: 0.562 G_ID: 0.133 G_Rec: 0.267 D_GP: 0.028 D_real: 1.018 D_fake: 0.935 +(epoch: 313, iters: 304, time: 0.064) G_GAN: -0.055 G_GAN_Feat: 0.804 G_ID: 0.166 G_Rec: 0.400 D_GP: 0.034 D_real: 0.668 D_fake: 1.055 +(epoch: 313, iters: 704, time: 0.064) G_GAN: 0.098 G_GAN_Feat: 0.571 G_ID: 0.136 G_Rec: 0.261 D_GP: 0.022 D_real: 1.079 D_fake: 0.903 +(epoch: 313, iters: 1104, time: 0.064) G_GAN: 0.157 G_GAN_Feat: 0.863 G_ID: 0.182 G_Rec: 0.408 D_GP: 0.045 D_real: 0.797 D_fake: 0.848 +(epoch: 313, iters: 1504, time: 0.064) G_GAN: 0.076 G_GAN_Feat: 0.663 G_ID: 0.125 G_Rec: 0.307 D_GP: 0.034 D_real: 0.860 D_fake: 0.924 +(epoch: 313, iters: 1904, time: 0.064) G_GAN: 0.418 G_GAN_Feat: 0.828 G_ID: 0.142 G_Rec: 0.398 D_GP: 0.022 D_real: 1.177 D_fake: 0.590 +(epoch: 313, iters: 2304, time: 0.064) G_GAN: 0.024 G_GAN_Feat: 0.614 G_ID: 0.134 G_Rec: 0.276 D_GP: 0.036 D_real: 0.917 D_fake: 0.976 +(epoch: 313, iters: 2704, time: 0.064) G_GAN: 0.254 G_GAN_Feat: 0.737 G_ID: 0.149 G_Rec: 0.319 D_GP: 0.027 D_real: 1.089 D_fake: 0.747 +(epoch: 313, iters: 3104, time: 0.064) G_GAN: 0.299 G_GAN_Feat: 0.628 G_ID: 0.118 G_Rec: 0.267 D_GP: 0.021 D_real: 1.290 D_fake: 0.702 +(epoch: 313, iters: 3504, time: 0.064) G_GAN: 0.449 G_GAN_Feat: 0.852 G_ID: 0.157 G_Rec: 0.415 D_GP: 0.021 D_real: 1.215 D_fake: 0.552 +(epoch: 313, iters: 3904, time: 0.064) G_GAN: 0.220 G_GAN_Feat: 0.699 G_ID: 0.111 G_Rec: 0.280 D_GP: 0.057 D_real: 0.896 D_fake: 0.780 +(epoch: 313, iters: 4304, time: 0.064) G_GAN: 0.507 G_GAN_Feat: 0.934 G_ID: 0.155 G_Rec: 0.423 D_GP: 0.082 D_real: 0.957 D_fake: 0.504 +(epoch: 313, iters: 4704, time: 0.064) G_GAN: -0.016 G_GAN_Feat: 0.755 G_ID: 0.123 G_Rec: 0.312 D_GP: 0.032 D_real: 1.108 D_fake: 1.017 +(epoch: 313, iters: 5104, time: 0.064) G_GAN: 0.249 G_GAN_Feat: 0.736 G_ID: 0.183 G_Rec: 0.355 D_GP: 0.018 D_real: 1.144 D_fake: 0.751 +(epoch: 313, iters: 5504, time: 0.064) G_GAN: 0.269 G_GAN_Feat: 0.668 G_ID: 0.118 G_Rec: 0.303 D_GP: 0.028 D_real: 1.147 D_fake: 0.732 +(epoch: 313, iters: 5904, time: 0.064) G_GAN: 0.248 G_GAN_Feat: 0.824 G_ID: 0.148 G_Rec: 0.387 D_GP: 0.028 D_real: 0.955 D_fake: 0.752 +(epoch: 313, iters: 6304, time: 0.064) G_GAN: -0.050 G_GAN_Feat: 0.706 G_ID: 0.125 G_Rec: 0.312 D_GP: 0.043 D_real: 0.718 D_fake: 1.050 +(epoch: 313, iters: 6704, time: 0.064) G_GAN: 0.306 G_GAN_Feat: 0.758 G_ID: 0.147 G_Rec: 0.359 D_GP: 0.020 D_real: 1.141 D_fake: 0.694 +(epoch: 313, iters: 7104, time: 0.064) G_GAN: 0.193 G_GAN_Feat: 0.616 G_ID: 0.114 G_Rec: 0.264 D_GP: 0.027 D_real: 1.091 D_fake: 0.808 +(epoch: 313, iters: 7504, time: 0.064) G_GAN: 0.299 G_GAN_Feat: 0.866 G_ID: 0.168 G_Rec: 0.387 D_GP: 0.049 D_real: 0.821 D_fake: 0.702 +(epoch: 313, iters: 7904, time: 0.064) G_GAN: 0.104 G_GAN_Feat: 0.696 G_ID: 0.133 G_Rec: 0.295 D_GP: 0.027 D_real: 0.957 D_fake: 0.896 +(epoch: 313, iters: 8304, time: 0.064) G_GAN: 0.418 G_GAN_Feat: 0.852 G_ID: 0.160 G_Rec: 0.374 D_GP: 0.030 D_real: 1.049 D_fake: 0.585 +(epoch: 314, iters: 96, time: 0.064) G_GAN: 0.367 G_GAN_Feat: 0.805 G_ID: 0.131 G_Rec: 0.331 D_GP: 0.049 D_real: 1.055 D_fake: 0.770 +(epoch: 314, iters: 496, time: 0.064) G_GAN: 0.304 G_GAN_Feat: 1.013 G_ID: 0.156 G_Rec: 0.415 D_GP: 0.091 D_real: 0.623 D_fake: 0.697 +(epoch: 314, iters: 896, time: 0.064) G_GAN: 0.190 G_GAN_Feat: 0.639 G_ID: 0.106 G_Rec: 0.259 D_GP: 0.029 D_real: 1.131 D_fake: 0.810 +(epoch: 314, iters: 1296, time: 0.064) G_GAN: 0.302 G_GAN_Feat: 0.999 G_ID: 0.152 G_Rec: 0.423 D_GP: 0.039 D_real: 0.806 D_fake: 0.699 +(epoch: 314, iters: 1696, time: 0.064) G_GAN: 0.157 G_GAN_Feat: 0.643 G_ID: 0.141 G_Rec: 0.296 D_GP: 0.021 D_real: 1.064 D_fake: 0.843 +(epoch: 314, iters: 2096, time: 0.064) G_GAN: 0.335 G_GAN_Feat: 0.931 G_ID: 0.153 G_Rec: 0.398 D_GP: 0.081 D_real: 0.893 D_fake: 0.665 +(epoch: 314, iters: 2496, time: 0.064) G_GAN: 0.415 G_GAN_Feat: 0.670 G_ID: 0.124 G_Rec: 0.289 D_GP: 0.025 D_real: 1.270 D_fake: 0.591 +(epoch: 314, iters: 2896, time: 0.064) G_GAN: 0.296 G_GAN_Feat: 0.866 G_ID: 0.140 G_Rec: 0.397 D_GP: 0.028 D_real: 1.030 D_fake: 0.706 +(epoch: 314, iters: 3296, time: 0.064) G_GAN: 0.113 G_GAN_Feat: 0.707 G_ID: 0.126 G_Rec: 0.303 D_GP: 0.041 D_real: 0.927 D_fake: 0.887 +(epoch: 314, iters: 3696, time: 0.064) G_GAN: 0.133 G_GAN_Feat: 0.859 G_ID: 0.177 G_Rec: 0.435 D_GP: 0.026 D_real: 0.985 D_fake: 0.868 +(epoch: 314, iters: 4096, time: 0.064) G_GAN: -0.030 G_GAN_Feat: 0.631 G_ID: 0.127 G_Rec: 0.280 D_GP: 0.037 D_real: 0.885 D_fake: 1.030 +(epoch: 314, iters: 4496, time: 0.064) G_GAN: 0.240 G_GAN_Feat: 0.785 G_ID: 0.168 G_Rec: 0.359 D_GP: 0.028 D_real: 0.959 D_fake: 0.765 +(epoch: 314, iters: 4896, time: 0.064) G_GAN: 0.116 G_GAN_Feat: 0.688 G_ID: 0.148 G_Rec: 0.285 D_GP: 0.090 D_real: 0.813 D_fake: 0.885 +(epoch: 314, iters: 5296, time: 0.064) G_GAN: 0.108 G_GAN_Feat: 1.000 G_ID: 0.184 G_Rec: 0.419 D_GP: 0.046 D_real: 0.509 D_fake: 0.894 +(epoch: 314, iters: 5696, time: 0.064) G_GAN: 0.187 G_GAN_Feat: 0.610 G_ID: 0.110 G_Rec: 0.289 D_GP: 0.023 D_real: 1.169 D_fake: 0.813 +(epoch: 314, iters: 6096, time: 0.064) G_GAN: 0.379 G_GAN_Feat: 0.901 G_ID: 0.171 G_Rec: 0.394 D_GP: 0.045 D_real: 0.865 D_fake: 0.622 +(epoch: 314, iters: 6496, time: 0.064) G_GAN: 0.337 G_GAN_Feat: 0.659 G_ID: 0.124 G_Rec: 0.307 D_GP: 0.026 D_real: 1.134 D_fake: 0.663 +(epoch: 314, iters: 6896, time: 0.064) G_GAN: 0.376 G_GAN_Feat: 1.048 G_ID: 0.163 G_Rec: 0.413 D_GP: 0.082 D_real: 0.329 D_fake: 0.637 +(epoch: 314, iters: 7296, time: 0.064) G_GAN: 0.390 G_GAN_Feat: 0.822 G_ID: 0.118 G_Rec: 0.311 D_GP: 0.025 D_real: 0.949 D_fake: 0.610 +(epoch: 314, iters: 7696, time: 0.064) G_GAN: 0.399 G_GAN_Feat: 0.809 G_ID: 0.159 G_Rec: 0.374 D_GP: 0.023 D_real: 1.229 D_fake: 0.604 +(epoch: 314, iters: 8096, time: 0.064) G_GAN: 0.092 G_GAN_Feat: 0.823 G_ID: 0.128 G_Rec: 0.335 D_GP: 0.082 D_real: 0.668 D_fake: 0.910 +(epoch: 314, iters: 8496, time: 0.064) G_GAN: 0.469 G_GAN_Feat: 0.969 G_ID: 0.151 G_Rec: 0.403 D_GP: 0.088 D_real: 0.736 D_fake: 0.534 +(epoch: 315, iters: 288, time: 0.064) G_GAN: 0.369 G_GAN_Feat: 0.862 G_ID: 0.135 G_Rec: 0.315 D_GP: 0.063 D_real: 0.690 D_fake: 0.639 +(epoch: 315, iters: 688, time: 0.064) G_GAN: 0.214 G_GAN_Feat: 0.926 G_ID: 0.158 G_Rec: 0.422 D_GP: 0.020 D_real: 0.980 D_fake: 0.788 +(epoch: 315, iters: 1088, time: 0.064) G_GAN: -0.014 G_GAN_Feat: 0.715 G_ID: 0.133 G_Rec: 0.386 D_GP: 0.021 D_real: 0.888 D_fake: 1.014 +(epoch: 315, iters: 1488, time: 0.064) G_GAN: 0.107 G_GAN_Feat: 0.779 G_ID: 0.159 G_Rec: 0.374 D_GP: 0.023 D_real: 0.946 D_fake: 0.893 +(epoch: 315, iters: 1888, time: 0.064) G_GAN: -0.040 G_GAN_Feat: 0.623 G_ID: 0.121 G_Rec: 0.285 D_GP: 0.024 D_real: 0.879 D_fake: 1.040 +(epoch: 315, iters: 2288, time: 0.064) G_GAN: 0.352 G_GAN_Feat: 0.816 G_ID: 0.137 G_Rec: 0.391 D_GP: 0.023 D_real: 1.120 D_fake: 0.661 +(epoch: 315, iters: 2688, time: 0.064) G_GAN: -0.143 G_GAN_Feat: 0.667 G_ID: 0.121 G_Rec: 0.285 D_GP: 0.038 D_real: 0.736 D_fake: 1.143 +(epoch: 315, iters: 3088, time: 0.064) G_GAN: 0.367 G_GAN_Feat: 0.900 G_ID: 0.119 G_Rec: 0.396 D_GP: 0.034 D_real: 1.046 D_fake: 0.636 +(epoch: 315, iters: 3488, time: 0.064) G_GAN: 0.704 G_GAN_Feat: 0.861 G_ID: 0.116 G_Rec: 0.343 D_GP: 0.047 D_real: 1.123 D_fake: 0.508 +(epoch: 315, iters: 3888, time: 0.064) G_GAN: 0.201 G_GAN_Feat: 0.865 G_ID: 0.173 G_Rec: 0.370 D_GP: 0.051 D_real: 0.907 D_fake: 0.800 +(epoch: 315, iters: 4288, time: 0.064) G_GAN: 0.588 G_GAN_Feat: 0.824 G_ID: 0.115 G_Rec: 0.306 D_GP: 0.045 D_real: 1.416 D_fake: 0.468 +(epoch: 315, iters: 4688, time: 0.064) G_GAN: 0.257 G_GAN_Feat: 1.123 G_ID: 0.158 G_Rec: 0.468 D_GP: 0.215 D_real: 0.298 D_fake: 0.745 +(epoch: 315, iters: 5088, time: 0.064) G_GAN: 0.146 G_GAN_Feat: 0.633 G_ID: 0.131 G_Rec: 0.271 D_GP: 0.023 D_real: 1.075 D_fake: 0.854 +(epoch: 315, iters: 5488, time: 0.064) G_GAN: 0.637 G_GAN_Feat: 1.161 G_ID: 0.162 G_Rec: 0.480 D_GP: 0.208 D_real: 0.444 D_fake: 0.434 +(epoch: 315, iters: 5888, time: 0.064) G_GAN: 0.116 G_GAN_Feat: 0.712 G_ID: 0.131 G_Rec: 0.305 D_GP: 0.033 D_real: 0.880 D_fake: 0.885 +(epoch: 315, iters: 6288, time: 0.064) G_GAN: 0.542 G_GAN_Feat: 1.027 G_ID: 0.169 G_Rec: 0.394 D_GP: 0.276 D_real: 0.638 D_fake: 0.491 +(epoch: 315, iters: 6688, time: 0.064) G_GAN: 0.594 G_GAN_Feat: 0.982 G_ID: 0.114 G_Rec: 0.289 D_GP: 0.037 D_real: 1.397 D_fake: 0.561 +(epoch: 315, iters: 7088, time: 0.064) G_GAN: 0.631 G_GAN_Feat: 0.859 G_ID: 0.144 G_Rec: 0.354 D_GP: 0.030 D_real: 1.267 D_fake: 0.383 +(epoch: 315, iters: 7488, time: 0.064) G_GAN: 0.389 G_GAN_Feat: 0.900 G_ID: 0.113 G_Rec: 0.299 D_GP: 0.067 D_real: 0.379 D_fake: 0.659 +(epoch: 315, iters: 7888, time: 0.064) G_GAN: 0.304 G_GAN_Feat: 1.046 G_ID: 0.163 G_Rec: 0.419 D_GP: 0.043 D_real: 0.788 D_fake: 0.698 +(epoch: 315, iters: 8288, time: 0.064) G_GAN: 0.221 G_GAN_Feat: 0.747 G_ID: 0.118 G_Rec: 0.280 D_GP: 0.037 D_real: 0.957 D_fake: 0.782 +(epoch: 316, iters: 80, time: 0.064) G_GAN: 0.395 G_GAN_Feat: 1.271 G_ID: 0.167 G_Rec: 0.452 D_GP: 0.528 D_real: 0.312 D_fake: 0.612 +(epoch: 316, iters: 480, time: 0.064) G_GAN: -0.151 G_GAN_Feat: 0.633 G_ID: 0.101 G_Rec: 0.319 D_GP: 0.021 D_real: 0.872 D_fake: 1.151 +(epoch: 316, iters: 880, time: 0.064) G_GAN: 0.174 G_GAN_Feat: 0.858 G_ID: 0.145 G_Rec: 0.408 D_GP: 0.018 D_real: 0.895 D_fake: 0.832 +(epoch: 316, iters: 1280, time: 0.064) G_GAN: -0.111 G_GAN_Feat: 0.700 G_ID: 0.122 G_Rec: 0.327 D_GP: 0.026 D_real: 0.785 D_fake: 1.112 +(epoch: 316, iters: 1680, time: 0.064) G_GAN: 0.168 G_GAN_Feat: 0.760 G_ID: 0.147 G_Rec: 0.352 D_GP: 0.024 D_real: 0.893 D_fake: 0.832 +(epoch: 316, iters: 2080, time: 0.064) G_GAN: 0.012 G_GAN_Feat: 0.765 G_ID: 0.125 G_Rec: 0.306 D_GP: 0.082 D_real: 0.707 D_fake: 0.988 +(epoch: 316, iters: 2480, time: 0.064) G_GAN: 0.409 G_GAN_Feat: 0.985 G_ID: 0.137 G_Rec: 0.415 D_GP: 0.062 D_real: 0.720 D_fake: 0.598 +(epoch: 316, iters: 2880, time: 0.064) G_GAN: 0.117 G_GAN_Feat: 0.665 G_ID: 0.120 G_Rec: 0.306 D_GP: 0.029 D_real: 1.157 D_fake: 0.889 +(epoch: 316, iters: 3280, time: 0.064) G_GAN: 0.387 G_GAN_Feat: 0.816 G_ID: 0.149 G_Rec: 0.381 D_GP: 0.031 D_real: 1.118 D_fake: 0.613 +(epoch: 316, iters: 3680, time: 0.064) G_GAN: 0.191 G_GAN_Feat: 0.685 G_ID: 0.121 G_Rec: 0.319 D_GP: 0.041 D_real: 1.057 D_fake: 0.809 +(epoch: 316, iters: 4080, time: 0.064) G_GAN: 0.581 G_GAN_Feat: 0.785 G_ID: 0.162 G_Rec: 0.344 D_GP: 0.041 D_real: 1.327 D_fake: 0.432 +(epoch: 316, iters: 4480, time: 0.064) G_GAN: -0.057 G_GAN_Feat: 0.621 G_ID: 0.118 G_Rec: 0.271 D_GP: 0.023 D_real: 0.844 D_fake: 1.057 +(epoch: 316, iters: 4880, time: 0.064) G_GAN: 0.459 G_GAN_Feat: 0.953 G_ID: 0.168 G_Rec: 0.412 D_GP: 0.030 D_real: 1.118 D_fake: 0.547 +(epoch: 316, iters: 5280, time: 0.064) G_GAN: 0.307 G_GAN_Feat: 0.742 G_ID: 0.126 G_Rec: 0.290 D_GP: 0.023 D_real: 1.154 D_fake: 0.693 +(epoch: 316, iters: 5680, time: 0.064) G_GAN: 0.433 G_GAN_Feat: 1.048 G_ID: 0.154 G_Rec: 0.446 D_GP: 0.038 D_real: 1.003 D_fake: 0.574 +(epoch: 316, iters: 6080, time: 0.064) G_GAN: 0.477 G_GAN_Feat: 0.898 G_ID: 0.155 G_Rec: 0.315 D_GP: 0.122 D_real: 0.582 D_fake: 0.563 +================ Training Loss (Tue Apr 21 12:46:57 2020) ================ +(epoch: 315, iters: 7488, time: 0.075) G_GAN: 0.436 G_GAN_Feat: 0.885 G_ID: 0.143 G_Rec: 0.375 D_GP: 0.027 D_real: 1.123 D_fake: 0.566 +(epoch: 315, iters: 7888, time: 0.063) G_GAN: 0.172 G_GAN_Feat: 0.831 G_ID: 0.129 G_Rec: 0.312 D_GP: 0.032 D_real: 0.934 D_fake: 0.857 +(epoch: 315, iters: 8288, time: 0.063) G_GAN: 0.629 G_GAN_Feat: 1.244 G_ID: 0.151 G_Rec: 0.425 D_GP: 0.072 D_real: 0.481 D_fake: 0.384 +(epoch: 315, iters: 8688, time: 0.064) G_GAN: 0.307 G_GAN_Feat: 0.816 G_ID: 0.118 G_Rec: 0.341 D_GP: 0.073 D_real: 0.723 D_fake: 0.699 +(epoch: 315, iters: 9088, time: 0.063) G_GAN: 0.632 G_GAN_Feat: 0.942 G_ID: 0.132 G_Rec: 0.419 D_GP: 0.027 D_real: 1.228 D_fake: 0.406 +(epoch: 315, iters: 9488, time: 0.063) G_GAN: 0.339 G_GAN_Feat: 0.769 G_ID: 0.094 G_Rec: 0.305 D_GP: 0.038 D_real: 1.035 D_fake: 0.667 +(epoch: 315, iters: 9888, time: 0.063) G_GAN: 0.933 G_GAN_Feat: 1.035 G_ID: 0.140 G_Rec: 0.405 D_GP: 0.032 D_real: 1.385 D_fake: 0.148 +(epoch: 315, iters: 10288, time: 0.064) G_GAN: 0.382 G_GAN_Feat: 0.686 G_ID: 0.124 G_Rec: 0.289 D_GP: 0.023 D_real: 1.286 D_fake: 0.618 +(epoch: 315, iters: 10688, time: 0.063) G_GAN: 0.751 G_GAN_Feat: 1.101 G_ID: 0.159 G_Rec: 0.440 D_GP: 0.062 D_real: 0.495 D_fake: 0.286 +(epoch: 315, iters: 11088, time: 0.063) G_GAN: 0.415 G_GAN_Feat: 0.733 G_ID: 0.112 G_Rec: 0.283 D_GP: 0.029 D_real: 1.292 D_fake: 0.592 +(epoch: 315, iters: 11488, time: 0.063) G_GAN: 0.606 G_GAN_Feat: 1.195 G_ID: 0.163 G_Rec: 0.421 D_GP: 0.142 D_real: 0.290 D_fake: 0.508 +(epoch: 315, iters: 11888, time: 0.064) G_GAN: 0.422 G_GAN_Feat: 0.732 G_ID: 0.102 G_Rec: 0.286 D_GP: 0.027 D_real: 1.314 D_fake: 0.580 +(epoch: 315, iters: 12288, time: 0.064) G_GAN: 0.795 G_GAN_Feat: 1.130 G_ID: 0.150 G_Rec: 0.454 D_GP: 0.036 D_real: 1.068 D_fake: 0.261 +(epoch: 315, iters: 12688, time: 0.064) G_GAN: 0.339 G_GAN_Feat: 0.978 G_ID: 0.119 G_Rec: 0.361 D_GP: 0.087 D_real: 0.949 D_fake: 0.661 +(epoch: 315, iters: 13088, time: 0.064) G_GAN: 0.483 G_GAN_Feat: 1.163 G_ID: 0.145 G_Rec: 0.473 D_GP: 0.084 D_real: 0.491 D_fake: 0.524 +(epoch: 315, iters: 13488, time: 0.064) G_GAN: 0.385 G_GAN_Feat: 0.744 G_ID: 0.114 G_Rec: 0.312 D_GP: 0.033 D_real: 1.316 D_fake: 0.624 +(epoch: 315, iters: 13888, time: 0.064) G_GAN: 0.449 G_GAN_Feat: 0.893 G_ID: 0.142 G_Rec: 0.406 D_GP: 0.045 D_real: 0.997 D_fake: 0.561 +(epoch: 315, iters: 14288, time: 0.064) G_GAN: 0.393 G_GAN_Feat: 0.828 G_ID: 0.125 G_Rec: 0.318 D_GP: 0.166 D_real: 0.875 D_fake: 0.619 +(epoch: 315, iters: 14688, time: 0.064) G_GAN: 0.534 G_GAN_Feat: 0.946 G_ID: 0.135 G_Rec: 0.411 D_GP: 0.028 D_real: 1.180 D_fake: 0.469 +(epoch: 315, iters: 15088, time: 0.064) G_GAN: 0.344 G_GAN_Feat: 0.644 G_ID: 0.109 G_Rec: 0.272 D_GP: 0.026 D_real: 1.276 D_fake: 0.658 +(epoch: 315, iters: 15488, time: 0.063) G_GAN: 0.653 G_GAN_Feat: 0.875 G_ID: 0.142 G_Rec: 0.405 D_GP: 0.030 D_real: 1.307 D_fake: 0.364 +(epoch: 316, iters: 286, time: 0.063) G_GAN: 0.178 G_GAN_Feat: 0.858 G_ID: 0.099 G_Rec: 0.324 D_GP: 0.109 D_real: 0.509 D_fake: 0.824 +(epoch: 316, iters: 686, time: 0.063) G_GAN: 0.728 G_GAN_Feat: 1.022 G_ID: 0.135 G_Rec: 0.447 D_GP: 0.034 D_real: 1.279 D_fake: 0.301 +(epoch: 316, iters: 1086, time: 0.064) G_GAN: 0.131 G_GAN_Feat: 0.753 G_ID: 0.120 G_Rec: 0.326 D_GP: 0.030 D_real: 0.947 D_fake: 0.869 +(epoch: 316, iters: 1486, time: 0.063) G_GAN: 0.516 G_GAN_Feat: 0.917 G_ID: 0.154 G_Rec: 0.416 D_GP: 0.042 D_real: 1.112 D_fake: 0.488 +(epoch: 316, iters: 1886, time: 0.063) G_GAN: 0.396 G_GAN_Feat: 0.821 G_ID: 0.095 G_Rec: 0.286 D_GP: 0.057 D_real: 0.844 D_fake: 0.605 +(epoch: 316, iters: 2286, time: 0.063) G_GAN: 0.320 G_GAN_Feat: 0.992 G_ID: 0.138 G_Rec: 0.433 D_GP: 0.033 D_real: 0.959 D_fake: 0.681 +(epoch: 316, iters: 2686, time: 0.064) G_GAN: 0.424 G_GAN_Feat: 0.833 G_ID: 0.110 G_Rec: 0.334 D_GP: 0.029 D_real: 1.181 D_fake: 0.578 +(epoch: 316, iters: 3086, time: 0.063) G_GAN: 0.819 G_GAN_Feat: 1.221 G_ID: 0.150 G_Rec: 0.472 D_GP: 0.570 D_real: 0.536 D_fake: 0.269 +(epoch: 316, iters: 3486, time: 0.063) G_GAN: 0.366 G_GAN_Feat: 0.850 G_ID: 0.106 G_Rec: 0.313 D_GP: 0.028 D_real: 0.922 D_fake: 0.635 +(epoch: 316, iters: 3886, time: 0.063) G_GAN: 0.928 G_GAN_Feat: 1.102 G_ID: 0.154 G_Rec: 0.452 D_GP: 0.040 D_real: 1.486 D_fake: 0.242 +(epoch: 316, iters: 4286, time: 0.064) G_GAN: 0.201 G_GAN_Feat: 0.660 G_ID: 0.120 G_Rec: 0.299 D_GP: 0.025 D_real: 1.111 D_fake: 0.799 +(epoch: 316, iters: 4686, time: 0.063) G_GAN: 0.411 G_GAN_Feat: 0.899 G_ID: 0.157 G_Rec: 0.403 D_GP: 0.034 D_real: 0.993 D_fake: 0.592 +(epoch: 316, iters: 5086, time: 0.063) G_GAN: 0.332 G_GAN_Feat: 0.691 G_ID: 0.098 G_Rec: 0.289 D_GP: 0.027 D_real: 1.151 D_fake: 0.669 +(epoch: 316, iters: 5486, time: 0.063) G_GAN: 0.598 G_GAN_Feat: 0.986 G_ID: 0.155 G_Rec: 0.445 D_GP: 0.056 D_real: 1.043 D_fake: 0.461 +(epoch: 316, iters: 5886, time: 0.064) G_GAN: 0.316 G_GAN_Feat: 0.755 G_ID: 0.102 G_Rec: 0.284 D_GP: 0.044 D_real: 0.985 D_fake: 0.686 +(epoch: 316, iters: 6286, time: 0.063) G_GAN: 0.524 G_GAN_Feat: 0.996 G_ID: 0.134 G_Rec: 0.420 D_GP: 0.064 D_real: 0.750 D_fake: 0.479 +(epoch: 316, iters: 6686, time: 0.063) G_GAN: 0.219 G_GAN_Feat: 0.752 G_ID: 0.105 G_Rec: 0.302 D_GP: 0.028 D_real: 0.955 D_fake: 0.782 +(epoch: 316, iters: 7086, time: 0.063) G_GAN: 0.843 G_GAN_Feat: 0.939 G_ID: 0.124 G_Rec: 0.404 D_GP: 0.032 D_real: 1.306 D_fake: 0.231 +(epoch: 316, iters: 7486, time: 0.064) G_GAN: 0.179 G_GAN_Feat: 0.809 G_ID: 0.128 G_Rec: 0.305 D_GP: 0.063 D_real: 0.570 D_fake: 0.823 +(epoch: 316, iters: 7886, time: 0.063) G_GAN: 0.548 G_GAN_Feat: 1.000 G_ID: 0.152 G_Rec: 0.410 D_GP: 0.178 D_real: 0.841 D_fake: 0.497 +(epoch: 316, iters: 8286, time: 0.063) G_GAN: 0.303 G_GAN_Feat: 0.812 G_ID: 0.126 G_Rec: 0.293 D_GP: 0.045 D_real: 0.767 D_fake: 0.698 +(epoch: 316, iters: 8686, time: 0.063) G_GAN: 0.602 G_GAN_Feat: 0.872 G_ID: 0.122 G_Rec: 0.364 D_GP: 0.030 D_real: 1.261 D_fake: 0.408 +(epoch: 317, iters: 478, time: 0.064) G_GAN: 0.330 G_GAN_Feat: 0.821 G_ID: 0.115 G_Rec: 0.311 D_GP: 0.033 D_real: 0.967 D_fake: 0.674 +(epoch: 317, iters: 878, time: 0.063) G_GAN: 0.710 G_GAN_Feat: 1.150 G_ID: 0.122 G_Rec: 0.448 D_GP: 0.270 D_real: 0.414 D_fake: 0.327 +(epoch: 317, iters: 1278, time: 0.063) G_GAN: 0.463 G_GAN_Feat: 0.963 G_ID: 0.116 G_Rec: 0.306 D_GP: 0.074 D_real: 0.394 D_fake: 0.559 +(epoch: 317, iters: 1678, time: 0.063) G_GAN: 0.306 G_GAN_Feat: 0.960 G_ID: 0.137 G_Rec: 0.456 D_GP: 0.032 D_real: 1.053 D_fake: 0.695 +(epoch: 317, iters: 2078, time: 0.064) G_GAN: 0.110 G_GAN_Feat: 0.768 G_ID: 0.145 G_Rec: 0.318 D_GP: 0.031 D_real: 0.970 D_fake: 0.891 +(epoch: 317, iters: 2478, time: 0.063) G_GAN: 0.473 G_GAN_Feat: 1.213 G_ID: 0.145 G_Rec: 0.467 D_GP: 0.138 D_real: 0.291 D_fake: 0.534 +(epoch: 317, iters: 2878, time: 0.063) G_GAN: 0.492 G_GAN_Feat: 0.886 G_ID: 0.112 G_Rec: 0.337 D_GP: 0.043 D_real: 1.042 D_fake: 0.563 +(epoch: 317, iters: 3278, time: 0.063) G_GAN: 0.559 G_GAN_Feat: 1.035 G_ID: 0.111 G_Rec: 0.435 D_GP: 0.028 D_real: 1.002 D_fake: 0.468 +(epoch: 317, iters: 3678, time: 0.064) G_GAN: -0.027 G_GAN_Feat: 0.831 G_ID: 0.131 G_Rec: 0.335 D_GP: 0.067 D_real: 0.549 D_fake: 1.028 +(epoch: 317, iters: 4078, time: 0.063) G_GAN: 0.615 G_GAN_Feat: 0.968 G_ID: 0.124 G_Rec: 0.427 D_GP: 0.028 D_real: 1.118 D_fake: 0.400 +(epoch: 317, iters: 4478, time: 0.063) G_GAN: 0.224 G_GAN_Feat: 0.738 G_ID: 0.106 G_Rec: 0.294 D_GP: 0.040 D_real: 0.949 D_fake: 0.776 +(epoch: 317, iters: 4878, time: 0.063) G_GAN: 0.715 G_GAN_Feat: 0.935 G_ID: 0.119 G_Rec: 0.435 D_GP: 0.026 D_real: 1.450 D_fake: 0.307 +(epoch: 317, iters: 5278, time: 0.064) G_GAN: 0.151 G_GAN_Feat: 0.743 G_ID: 0.119 G_Rec: 0.302 D_GP: 0.032 D_real: 0.858 D_fake: 0.849 +(epoch: 317, iters: 5678, time: 0.064) G_GAN: 0.432 G_GAN_Feat: 1.023 G_ID: 0.124 G_Rec: 0.407 D_GP: 0.068 D_real: 0.721 D_fake: 0.573 +(epoch: 317, iters: 6078, time: 0.064) G_GAN: 0.118 G_GAN_Feat: 0.875 G_ID: 0.108 G_Rec: 0.316 D_GP: 0.064 D_real: 0.325 D_fake: 0.885 +(epoch: 317, iters: 6478, time: 0.064) G_GAN: 0.503 G_GAN_Feat: 1.193 G_ID: 0.138 G_Rec: 0.462 D_GP: 0.093 D_real: 0.214 D_fake: 0.532 +(epoch: 317, iters: 6878, time: 0.064) G_GAN: 0.386 G_GAN_Feat: 0.676 G_ID: 0.106 G_Rec: 0.297 D_GP: 0.023 D_real: 1.273 D_fake: 0.618 +(epoch: 317, iters: 7278, time: 0.063) G_GAN: 0.526 G_GAN_Feat: 0.972 G_ID: 0.129 G_Rec: 0.431 D_GP: 0.037 D_real: 0.985 D_fake: 0.479 +(epoch: 317, iters: 7678, time: 0.063) G_GAN: 0.338 G_GAN_Feat: 0.747 G_ID: 0.133 G_Rec: 0.289 D_GP: 0.037 D_real: 1.165 D_fake: 0.663 +(epoch: 317, iters: 8078, time: 0.063) G_GAN: 0.738 G_GAN_Feat: 0.902 G_ID: 0.119 G_Rec: 0.400 D_GP: 0.030 D_real: 1.337 D_fake: 0.286 +(epoch: 317, iters: 8478, time: 0.064) G_GAN: 0.083 G_GAN_Feat: 0.769 G_ID: 0.132 G_Rec: 0.327 D_GP: 0.054 D_real: 0.908 D_fake: 0.918 +(epoch: 318, iters: 270, time: 0.063) G_GAN: 0.506 G_GAN_Feat: 0.874 G_ID: 0.138 G_Rec: 0.409 D_GP: 0.034 D_real: 1.092 D_fake: 0.506 +(epoch: 318, iters: 670, time: 0.063) G_GAN: 0.173 G_GAN_Feat: 0.904 G_ID: 0.121 G_Rec: 0.333 D_GP: 0.063 D_real: 0.853 D_fake: 0.845 +(epoch: 318, iters: 1070, time: 0.063) G_GAN: 0.588 G_GAN_Feat: 0.811 G_ID: 0.128 G_Rec: 0.389 D_GP: 0.031 D_real: 1.325 D_fake: 0.423 +(epoch: 318, iters: 1470, time: 0.064) G_GAN: 0.011 G_GAN_Feat: 0.656 G_ID: 0.127 G_Rec: 0.297 D_GP: 0.027 D_real: 0.933 D_fake: 0.989 +(epoch: 318, iters: 1870, time: 0.063) G_GAN: 0.528 G_GAN_Feat: 0.946 G_ID: 0.121 G_Rec: 0.468 D_GP: 0.033 D_real: 1.161 D_fake: 0.485 +(epoch: 318, iters: 2270, time: 0.063) G_GAN: 0.187 G_GAN_Feat: 0.641 G_ID: 0.096 G_Rec: 0.292 D_GP: 0.028 D_real: 1.127 D_fake: 0.814 +(epoch: 318, iters: 2670, time: 0.063) G_GAN: 0.557 G_GAN_Feat: 0.844 G_ID: 0.128 G_Rec: 0.396 D_GP: 0.033 D_real: 1.299 D_fake: 0.470 +(epoch: 318, iters: 3070, time: 0.064) G_GAN: 0.032 G_GAN_Feat: 0.662 G_ID: 0.121 G_Rec: 0.302 D_GP: 0.036 D_real: 0.849 D_fake: 0.968 +(epoch: 318, iters: 3470, time: 0.063) G_GAN: 0.214 G_GAN_Feat: 0.871 G_ID: 0.124 G_Rec: 0.410 D_GP: 0.066 D_real: 0.811 D_fake: 0.790 +(epoch: 318, iters: 3870, time: 0.063) G_GAN: -0.087 G_GAN_Feat: 0.765 G_ID: 0.102 G_Rec: 0.324 D_GP: 0.055 D_real: 0.675 D_fake: 1.087 +(epoch: 318, iters: 4270, time: 0.063) G_GAN: 0.190 G_GAN_Feat: 1.108 G_ID: 0.132 G_Rec: 0.491 D_GP: 0.049 D_real: 0.874 D_fake: 0.829 +(epoch: 318, iters: 4670, time: 0.064) G_GAN: 0.224 G_GAN_Feat: 0.731 G_ID: 0.097 G_Rec: 0.331 D_GP: 0.043 D_real: 1.075 D_fake: 0.776 +(epoch: 318, iters: 5070, time: 0.063) G_GAN: 0.409 G_GAN_Feat: 0.916 G_ID: 0.131 G_Rec: 0.414 D_GP: 0.046 D_real: 1.045 D_fake: 0.596 +(epoch: 318, iters: 5470, time: 0.063) G_GAN: 0.054 G_GAN_Feat: 0.740 G_ID: 0.121 G_Rec: 0.338 D_GP: 0.037 D_real: 0.907 D_fake: 0.946 +(epoch: 318, iters: 5870, time: 0.063) G_GAN: 0.628 G_GAN_Feat: 1.041 G_ID: 0.129 G_Rec: 0.479 D_GP: 0.062 D_real: 0.883 D_fake: 0.391 +(epoch: 318, iters: 6270, time: 0.064) G_GAN: 0.240 G_GAN_Feat: 0.896 G_ID: 0.117 G_Rec: 0.332 D_GP: 0.160 D_real: 0.557 D_fake: 0.761 +(epoch: 318, iters: 6670, time: 0.063) G_GAN: 0.372 G_GAN_Feat: 0.878 G_ID: 0.147 G_Rec: 0.369 D_GP: 0.040 D_real: 0.925 D_fake: 0.628 +(epoch: 318, iters: 7070, time: 0.063) G_GAN: 0.367 G_GAN_Feat: 0.770 G_ID: 0.107 G_Rec: 0.320 D_GP: 0.056 D_real: 1.046 D_fake: 0.640 +(epoch: 318, iters: 7470, time: 0.063) G_GAN: 0.956 G_GAN_Feat: 0.974 G_ID: 0.126 G_Rec: 0.417 D_GP: 0.032 D_real: 1.510 D_fake: 0.164 +(epoch: 318, iters: 7870, time: 0.064) G_GAN: 0.211 G_GAN_Feat: 0.789 G_ID: 0.110 G_Rec: 0.308 D_GP: 0.057 D_real: 0.782 D_fake: 0.790 +(epoch: 318, iters: 8270, time: 0.063) G_GAN: 0.369 G_GAN_Feat: 0.964 G_ID: 0.139 G_Rec: 0.426 D_GP: 0.032 D_real: 0.954 D_fake: 0.632 +(epoch: 318, iters: 8670, time: 0.063) G_GAN: -0.080 G_GAN_Feat: 0.844 G_ID: 0.119 G_Rec: 0.316 D_GP: 0.130 D_real: 0.593 D_fake: 1.082 +(epoch: 319, iters: 462, time: 0.063) G_GAN: 0.703 G_GAN_Feat: 0.989 G_ID: 0.140 G_Rec: 0.413 D_GP: 0.053 D_real: 1.029 D_fake: 0.329 +(epoch: 319, iters: 862, time: 0.064) G_GAN: 0.263 G_GAN_Feat: 0.944 G_ID: 0.102 G_Rec: 0.351 D_GP: 0.480 D_real: 0.456 D_fake: 0.739 +(epoch: 319, iters: 1262, time: 0.063) G_GAN: 0.399 G_GAN_Feat: 1.009 G_ID: 0.143 G_Rec: 0.429 D_GP: 0.039 D_real: 0.882 D_fake: 0.618 +(epoch: 319, iters: 1662, time: 0.063) G_GAN: -0.030 G_GAN_Feat: 0.719 G_ID: 0.119 G_Rec: 0.276 D_GP: 0.028 D_real: 0.882 D_fake: 1.030 +(epoch: 319, iters: 2062, time: 0.063) G_GAN: 0.440 G_GAN_Feat: 1.050 G_ID: 0.152 G_Rec: 0.400 D_GP: 0.033 D_real: 0.495 D_fake: 0.561 +(epoch: 319, iters: 2462, time: 0.064) G_GAN: 0.305 G_GAN_Feat: 0.965 G_ID: 0.107 G_Rec: 0.307 D_GP: 0.102 D_real: 0.273 D_fake: 0.714 +(epoch: 319, iters: 2862, time: 0.063) G_GAN: 0.485 G_GAN_Feat: 0.882 G_ID: 0.151 G_Rec: 0.475 D_GP: 0.028 D_real: 1.196 D_fake: 0.543 +(epoch: 319, iters: 3262, time: 0.063) G_GAN: -0.078 G_GAN_Feat: 0.677 G_ID: 0.108 G_Rec: 0.305 D_GP: 0.027 D_real: 0.830 D_fake: 1.078 +(epoch: 319, iters: 3662, time: 0.063) G_GAN: 0.284 G_GAN_Feat: 0.851 G_ID: 0.117 G_Rec: 0.418 D_GP: 0.028 D_real: 0.893 D_fake: 0.718 +(epoch: 319, iters: 4062, time: 0.064) G_GAN: -0.106 G_GAN_Feat: 0.658 G_ID: 0.113 G_Rec: 0.315 D_GP: 0.028 D_real: 0.733 D_fake: 1.106 +(epoch: 319, iters: 4462, time: 0.063) G_GAN: 0.190 G_GAN_Feat: 0.861 G_ID: 0.123 G_Rec: 0.410 D_GP: 0.041 D_real: 0.828 D_fake: 0.811 +(epoch: 319, iters: 4862, time: 0.063) G_GAN: -0.178 G_GAN_Feat: 0.735 G_ID: 0.132 G_Rec: 0.315 D_GP: 0.038 D_real: 0.624 D_fake: 1.178 +(epoch: 319, iters: 5262, time: 0.063) G_GAN: 0.233 G_GAN_Feat: 0.828 G_ID: 0.147 G_Rec: 0.376 D_GP: 0.033 D_real: 0.913 D_fake: 0.768 +(epoch: 319, iters: 5662, time: 0.063) G_GAN: 0.223 G_GAN_Feat: 0.757 G_ID: 0.107 G_Rec: 0.308 D_GP: 0.038 D_real: 0.955 D_fake: 0.778 +(epoch: 319, iters: 6062, time: 0.063) G_GAN: 0.430 G_GAN_Feat: 0.954 G_ID: 0.142 G_Rec: 0.418 D_GP: 0.047 D_real: 0.920 D_fake: 0.571 +(epoch: 319, iters: 6462, time: 0.063) G_GAN: 0.162 G_GAN_Feat: 0.722 G_ID: 0.120 G_Rec: 0.319 D_GP: 0.040 D_real: 1.021 D_fake: 0.838 +(epoch: 319, iters: 6862, time: 0.064) G_GAN: 0.627 G_GAN_Feat: 0.936 G_ID: 0.130 G_Rec: 0.403 D_GP: 0.039 D_real: 1.063 D_fake: 0.381 +(epoch: 319, iters: 7262, time: 0.063) G_GAN: 0.014 G_GAN_Feat: 0.815 G_ID: 0.103 G_Rec: 0.350 D_GP: 0.068 D_real: 0.587 D_fake: 0.987 +(epoch: 319, iters: 7662, time: 0.063) G_GAN: 0.418 G_GAN_Feat: 0.989 G_ID: 0.138 G_Rec: 0.419 D_GP: 0.062 D_real: 0.914 D_fake: 0.585 +(epoch: 319, iters: 8062, time: 0.063) G_GAN: 0.225 G_GAN_Feat: 0.781 G_ID: 0.103 G_Rec: 0.304 D_GP: 0.043 D_real: 0.857 D_fake: 0.776 +(epoch: 319, iters: 8462, time: 0.064) G_GAN: 0.383 G_GAN_Feat: 1.014 G_ID: 0.141 G_Rec: 0.466 D_GP: 0.062 D_real: 0.823 D_fake: 0.622 +(epoch: 320, iters: 254, time: 0.063) G_GAN: 0.443 G_GAN_Feat: 0.874 G_ID: 0.102 G_Rec: 0.319 D_GP: 0.176 D_real: 0.753 D_fake: 0.593 +(epoch: 320, iters: 654, time: 0.063) G_GAN: 0.678 G_GAN_Feat: 0.975 G_ID: 0.129 G_Rec: 0.460 D_GP: 0.034 D_real: 1.260 D_fake: 0.335 +(epoch: 320, iters: 1054, time: 0.063) G_GAN: -0.069 G_GAN_Feat: 0.931 G_ID: 0.130 G_Rec: 0.331 D_GP: 0.040 D_real: 0.648 D_fake: 1.069 +(epoch: 320, iters: 1454, time: 0.064) G_GAN: 0.265 G_GAN_Feat: 1.001 G_ID: 0.111 G_Rec: 0.455 D_GP: 0.035 D_real: 0.944 D_fake: 0.736 +(epoch: 320, iters: 1854, time: 0.063) G_GAN: 0.997 G_GAN_Feat: 0.909 G_ID: 0.113 G_Rec: 0.360 D_GP: 0.174 D_real: 1.446 D_fake: 0.361 +(epoch: 320, iters: 2254, time: 0.063) G_GAN: 0.581 G_GAN_Feat: 1.215 G_ID: 0.147 G_Rec: 0.484 D_GP: 0.247 D_real: 0.512 D_fake: 0.443 +(epoch: 320, iters: 2654, time: 0.063) G_GAN: 0.380 G_GAN_Feat: 0.813 G_ID: 0.108 G_Rec: 0.324 D_GP: 0.045 D_real: 0.884 D_fake: 0.623 +(epoch: 320, iters: 3054, time: 0.064) G_GAN: 0.453 G_GAN_Feat: 0.918 G_ID: 0.175 G_Rec: 0.438 D_GP: 0.031 D_real: 1.162 D_fake: 0.555 +(epoch: 320, iters: 3454, time: 0.063) G_GAN: 0.216 G_GAN_Feat: 0.740 G_ID: 0.140 G_Rec: 0.343 D_GP: 0.031 D_real: 1.041 D_fake: 0.785 +(epoch: 320, iters: 3854, time: 0.063) G_GAN: 0.404 G_GAN_Feat: 0.960 G_ID: 0.152 G_Rec: 0.415 D_GP: 0.045 D_real: 0.854 D_fake: 0.601 +(epoch: 320, iters: 4254, time: 0.063) G_GAN: 0.185 G_GAN_Feat: 0.700 G_ID: 0.106 G_Rec: 0.305 D_GP: 0.028 D_real: 1.057 D_fake: 0.815 +(epoch: 320, iters: 4654, time: 0.064) G_GAN: 0.511 G_GAN_Feat: 0.986 G_ID: 0.124 G_Rec: 0.426 D_GP: 0.032 D_real: 1.062 D_fake: 0.491 +(epoch: 320, iters: 5054, time: 0.063) G_GAN: 0.322 G_GAN_Feat: 0.987 G_ID: 0.087 G_Rec: 0.325 D_GP: 1.654 D_real: 0.347 D_fake: 0.692 +(epoch: 320, iters: 5454, time: 0.063) G_GAN: 0.445 G_GAN_Feat: 0.975 G_ID: 0.111 G_Rec: 0.417 D_GP: 0.034 D_real: 0.997 D_fake: 0.557 +(epoch: 320, iters: 5854, time: 0.063) G_GAN: 0.008 G_GAN_Feat: 1.035 G_ID: 0.107 G_Rec: 0.354 D_GP: 0.436 D_real: 0.323 D_fake: 0.992 +(epoch: 320, iters: 6254, time: 0.064) G_GAN: 0.661 G_GAN_Feat: 1.234 G_ID: 0.142 G_Rec: 0.504 D_GP: 0.178 D_real: 0.386 D_fake: 0.415 +(epoch: 320, iters: 6654, time: 0.063) G_GAN: 0.478 G_GAN_Feat: 0.854 G_ID: 0.121 G_Rec: 0.340 D_GP: 0.038 D_real: 1.063 D_fake: 0.552 +(epoch: 320, iters: 7054, time: 0.063) G_GAN: 0.490 G_GAN_Feat: 0.951 G_ID: 0.134 G_Rec: 0.484 D_GP: 0.027 D_real: 1.212 D_fake: 0.523 +(epoch: 320, iters: 7454, time: 0.063) G_GAN: 0.059 G_GAN_Feat: 0.732 G_ID: 0.101 G_Rec: 0.313 D_GP: 0.031 D_real: 0.853 D_fake: 0.941 +(epoch: 320, iters: 7854, time: 0.064) G_GAN: 0.608 G_GAN_Feat: 1.072 G_ID: 0.117 G_Rec: 0.454 D_GP: 0.143 D_real: 0.673 D_fake: 0.414 +(epoch: 320, iters: 8254, time: 0.063) G_GAN: 0.337 G_GAN_Feat: 0.790 G_ID: 0.112 G_Rec: 0.335 D_GP: 0.050 D_real: 1.066 D_fake: 0.694 +(epoch: 320, iters: 8654, time: 0.063) G_GAN: 0.557 G_GAN_Feat: 1.223 G_ID: 0.135 G_Rec: 0.496 D_GP: 0.137 D_real: 0.499 D_fake: 0.469 +(epoch: 321, iters: 446, time: 0.063) G_GAN: 0.476 G_GAN_Feat: 0.743 G_ID: 0.106 G_Rec: 0.315 D_GP: 0.027 D_real: 1.314 D_fake: 0.530 +(epoch: 321, iters: 846, time: 0.064) G_GAN: -0.190 G_GAN_Feat: 1.140 G_ID: 0.146 G_Rec: 0.483 D_GP: 0.029 D_real: 0.513 D_fake: 1.191 +(epoch: 321, iters: 1246, time: 0.063) G_GAN: -0.088 G_GAN_Feat: 0.686 G_ID: 0.106 G_Rec: 0.342 D_GP: 0.024 D_real: 0.830 D_fake: 1.088 +(epoch: 321, iters: 1646, time: 0.063) G_GAN: -0.018 G_GAN_Feat: 0.813 G_ID: 0.123 G_Rec: 0.393 D_GP: 0.030 D_real: 0.764 D_fake: 1.018 +(epoch: 321, iters: 2046, time: 0.063) G_GAN: 0.077 G_GAN_Feat: 0.631 G_ID: 0.106 G_Rec: 0.291 D_GP: 0.034 D_real: 0.953 D_fake: 0.925 +(epoch: 321, iters: 2446, time: 0.064) G_GAN: 0.128 G_GAN_Feat: 0.803 G_ID: 0.133 G_Rec: 0.389 D_GP: 0.032 D_real: 0.807 D_fake: 0.873 +(epoch: 321, iters: 2846, time: 0.063) G_GAN: -0.111 G_GAN_Feat: 0.738 G_ID: 0.114 G_Rec: 0.366 D_GP: 0.059 D_real: 0.682 D_fake: 1.112 +(epoch: 321, iters: 3246, time: 0.063) G_GAN: 0.573 G_GAN_Feat: 0.888 G_ID: 0.115 G_Rec: 0.441 D_GP: 0.039 D_real: 1.091 D_fake: 0.503 +(epoch: 321, iters: 3646, time: 0.063) G_GAN: 0.023 G_GAN_Feat: 0.653 G_ID: 0.112 G_Rec: 0.304 D_GP: 0.040 D_real: 0.856 D_fake: 0.977 +(epoch: 321, iters: 4046, time: 0.064) G_GAN: 0.452 G_GAN_Feat: 0.944 G_ID: 0.135 G_Rec: 0.450 D_GP: 0.057 D_real: 0.986 D_fake: 0.575 +(epoch: 321, iters: 4446, time: 0.063) G_GAN: -0.041 G_GAN_Feat: 0.759 G_ID: 0.116 G_Rec: 0.321 D_GP: 0.051 D_real: 0.626 D_fake: 1.045 +(epoch: 321, iters: 4846, time: 0.063) G_GAN: 0.276 G_GAN_Feat: 0.901 G_ID: 0.121 G_Rec: 0.416 D_GP: 0.049 D_real: 0.908 D_fake: 0.725 +(epoch: 321, iters: 5246, time: 0.063) G_GAN: 0.098 G_GAN_Feat: 0.669 G_ID: 0.094 G_Rec: 0.319 D_GP: 0.037 D_real: 0.971 D_fake: 0.903 +(epoch: 321, iters: 5646, time: 0.064) G_GAN: 0.572 G_GAN_Feat: 0.960 G_ID: 0.122 G_Rec: 0.444 D_GP: 0.053 D_real: 1.109 D_fake: 0.480 +(epoch: 321, iters: 6046, time: 0.063) G_GAN: 0.234 G_GAN_Feat: 0.670 G_ID: 0.112 G_Rec: 0.296 D_GP: 0.041 D_real: 1.148 D_fake: 0.768 +(epoch: 321, iters: 6446, time: 0.063) G_GAN: 0.365 G_GAN_Feat: 0.892 G_ID: 0.131 G_Rec: 0.385 D_GP: 0.047 D_real: 0.975 D_fake: 0.639 +(epoch: 321, iters: 6846, time: 0.063) G_GAN: 0.217 G_GAN_Feat: 0.688 G_ID: 0.109 G_Rec: 0.312 D_GP: 0.035 D_real: 1.165 D_fake: 0.783 +(epoch: 321, iters: 7246, time: 0.064) G_GAN: 0.288 G_GAN_Feat: 0.920 G_ID: 0.154 G_Rec: 0.449 D_GP: 0.037 D_real: 0.798 D_fake: 0.714 +(epoch: 321, iters: 7646, time: 0.063) G_GAN: 0.144 G_GAN_Feat: 0.766 G_ID: 0.112 G_Rec: 0.304 D_GP: 0.051 D_real: 0.930 D_fake: 0.856 +(epoch: 321, iters: 8046, time: 0.063) G_GAN: 0.322 G_GAN_Feat: 1.001 G_ID: 0.129 G_Rec: 0.429 D_GP: 0.055 D_real: 0.854 D_fake: 0.680 +(epoch: 321, iters: 8446, time: 0.063) G_GAN: 0.262 G_GAN_Feat: 0.855 G_ID: 0.106 G_Rec: 0.314 D_GP: 0.082 D_real: 0.673 D_fake: 0.739 +(epoch: 322, iters: 238, time: 0.064) G_GAN: 0.394 G_GAN_Feat: 1.217 G_ID: 0.152 G_Rec: 0.482 D_GP: 0.238 D_real: 0.429 D_fake: 0.616 +(epoch: 322, iters: 638, time: 0.063) G_GAN: 0.190 G_GAN_Feat: 0.740 G_ID: 0.112 G_Rec: 0.315 D_GP: 0.030 D_real: 1.056 D_fake: 0.810 +(epoch: 322, iters: 1038, time: 0.063) G_GAN: 0.236 G_GAN_Feat: 0.947 G_ID: 0.128 G_Rec: 0.443 D_GP: 0.037 D_real: 0.878 D_fake: 0.765 +(epoch: 322, iters: 1438, time: 0.063) G_GAN: 0.050 G_GAN_Feat: 0.723 G_ID: 0.105 G_Rec: 0.309 D_GP: 0.048 D_real: 0.793 D_fake: 0.950 +(epoch: 322, iters: 1838, time: 0.064) G_GAN: 0.153 G_GAN_Feat: 1.027 G_ID: 0.137 G_Rec: 0.406 D_GP: 0.059 D_real: 0.408 D_fake: 0.847 +(epoch: 322, iters: 2238, time: 0.063) G_GAN: 0.455 G_GAN_Feat: 0.783 G_ID: 0.092 G_Rec: 0.315 D_GP: 0.039 D_real: 1.130 D_fake: 0.559 +(epoch: 322, iters: 2638, time: 0.063) G_GAN: 0.643 G_GAN_Feat: 0.969 G_ID: 0.131 G_Rec: 0.452 D_GP: 0.031 D_real: 1.250 D_fake: 0.377 +(epoch: 322, iters: 3038, time: 0.063) G_GAN: 0.232 G_GAN_Feat: 0.889 G_ID: 0.114 G_Rec: 0.354 D_GP: 0.041 D_real: 0.827 D_fake: 0.769 +(epoch: 322, iters: 3438, time: 0.064) G_GAN: 0.029 G_GAN_Feat: 1.008 G_ID: 0.151 G_Rec: 0.465 D_GP: 0.057 D_real: 0.581 D_fake: 0.973 +(epoch: 322, iters: 3838, time: 0.063) G_GAN: 0.091 G_GAN_Feat: 0.699 G_ID: 0.111 G_Rec: 0.286 D_GP: 0.030 D_real: 0.979 D_fake: 0.909 +(epoch: 322, iters: 4238, time: 0.063) G_GAN: 0.682 G_GAN_Feat: 0.900 G_ID: 0.115 G_Rec: 0.471 D_GP: 0.029 D_real: 1.326 D_fake: 0.371 +(epoch: 322, iters: 4638, time: 0.063) G_GAN: 0.193 G_GAN_Feat: 0.725 G_ID: 0.103 G_Rec: 0.305 D_GP: 0.033 D_real: 1.142 D_fake: 0.808 +(epoch: 322, iters: 5038, time: 0.064) G_GAN: 0.606 G_GAN_Feat: 0.961 G_ID: 0.140 G_Rec: 0.462 D_GP: 0.033 D_real: 1.108 D_fake: 0.441 +(epoch: 322, iters: 5438, time: 0.063) G_GAN: 0.423 G_GAN_Feat: 0.702 G_ID: 0.096 G_Rec: 0.295 D_GP: 0.029 D_real: 1.355 D_fake: 0.581 +(epoch: 322, iters: 5838, time: 0.063) G_GAN: 0.427 G_GAN_Feat: 0.937 G_ID: 0.128 G_Rec: 0.393 D_GP: 0.039 D_real: 0.889 D_fake: 0.574 +(epoch: 322, iters: 6238, time: 0.063) G_GAN: -0.181 G_GAN_Feat: 0.811 G_ID: 0.115 G_Rec: 0.365 D_GP: 0.128 D_real: 0.477 D_fake: 1.181 +(epoch: 322, iters: 6638, time: 0.064) G_GAN: 0.106 G_GAN_Feat: 0.995 G_ID: 0.160 G_Rec: 0.453 D_GP: 0.060 D_real: 0.668 D_fake: 0.895 +(epoch: 322, iters: 7038, time: 0.063) G_GAN: 0.422 G_GAN_Feat: 0.812 G_ID: 0.105 G_Rec: 0.342 D_GP: 0.043 D_real: 1.037 D_fake: 0.579 +(epoch: 322, iters: 7438, time: 0.063) G_GAN: 0.825 G_GAN_Feat: 0.981 G_ID: 0.130 G_Rec: 0.441 D_GP: 0.028 D_real: 1.389 D_fake: 0.236 +(epoch: 322, iters: 7838, time: 0.063) G_GAN: 0.274 G_GAN_Feat: 0.795 G_ID: 0.098 G_Rec: 0.305 D_GP: 0.045 D_real: 1.194 D_fake: 0.727 +(epoch: 322, iters: 8238, time: 0.064) G_GAN: 0.320 G_GAN_Feat: 1.082 G_ID: 0.144 G_Rec: 0.439 D_GP: 0.062 D_real: 0.510 D_fake: 0.682 +(epoch: 322, iters: 8638, time: 0.063) G_GAN: 0.242 G_GAN_Feat: 0.947 G_ID: 0.116 G_Rec: 0.318 D_GP: 0.112 D_real: 0.362 D_fake: 0.758 +(epoch: 323, iters: 430, time: 0.063) G_GAN: 0.023 G_GAN_Feat: 1.024 G_ID: 0.152 G_Rec: 0.431 D_GP: 0.036 D_real: 0.841 D_fake: 0.980 +(epoch: 323, iters: 830, time: 0.063) G_GAN: 0.309 G_GAN_Feat: 0.715 G_ID: 0.103 G_Rec: 0.321 D_GP: 0.032 D_real: 1.107 D_fake: 0.697 +(epoch: 323, iters: 1230, time: 0.064) G_GAN: 0.447 G_GAN_Feat: 0.916 G_ID: 0.118 G_Rec: 0.440 D_GP: 0.029 D_real: 1.076 D_fake: 0.554 +(epoch: 323, iters: 1630, time: 0.063) G_GAN: 0.420 G_GAN_Feat: 0.898 G_ID: 0.112 G_Rec: 0.316 D_GP: 0.034 D_real: 0.737 D_fake: 0.591 +(epoch: 323, iters: 2030, time: 0.063) G_GAN: 0.144 G_GAN_Feat: 1.214 G_ID: 0.142 G_Rec: 0.484 D_GP: 0.337 D_real: 0.288 D_fake: 0.856 +(epoch: 323, iters: 2430, time: 0.063) G_GAN: 0.273 G_GAN_Feat: 0.684 G_ID: 0.113 G_Rec: 0.324 D_GP: 0.025 D_real: 1.173 D_fake: 0.727 +(epoch: 323, iters: 2830, time: 0.064) G_GAN: 0.554 G_GAN_Feat: 0.946 G_ID: 0.123 G_Rec: 0.475 D_GP: 0.032 D_real: 1.147 D_fake: 0.463 +(epoch: 323, iters: 3230, time: 0.063) G_GAN: 0.159 G_GAN_Feat: 0.670 G_ID: 0.095 G_Rec: 0.302 D_GP: 0.030 D_real: 0.986 D_fake: 0.841 +(epoch: 323, iters: 3630, time: 0.063) G_GAN: 0.416 G_GAN_Feat: 1.023 G_ID: 0.121 G_Rec: 0.491 D_GP: 0.079 D_real: 0.794 D_fake: 0.594 +(epoch: 323, iters: 4030, time: 0.063) G_GAN: 0.133 G_GAN_Feat: 0.771 G_ID: 0.108 G_Rec: 0.314 D_GP: 0.060 D_real: 0.713 D_fake: 0.867 +(epoch: 323, iters: 4430, time: 0.064) G_GAN: 0.812 G_GAN_Feat: 0.952 G_ID: 0.131 G_Rec: 0.413 D_GP: 0.050 D_real: 1.408 D_fake: 0.267 +(epoch: 323, iters: 4830, time: 0.063) G_GAN: 0.289 G_GAN_Feat: 0.717 G_ID: 0.094 G_Rec: 0.273 D_GP: 0.034 D_real: 1.131 D_fake: 0.711 +(epoch: 323, iters: 5230, time: 0.063) G_GAN: 0.396 G_GAN_Feat: 1.105 G_ID: 0.120 G_Rec: 0.432 D_GP: 0.057 D_real: 0.553 D_fake: 0.606 +(epoch: 323, iters: 5630, time: 0.063) G_GAN: -0.037 G_GAN_Feat: 0.738 G_ID: 0.115 G_Rec: 0.343 D_GP: 0.026 D_real: 0.888 D_fake: 1.039 +(epoch: 323, iters: 6030, time: 0.064) G_GAN: 0.601 G_GAN_Feat: 0.833 G_ID: 0.121 G_Rec: 0.384 D_GP: 0.031 D_real: 1.268 D_fake: 0.415 +(epoch: 323, iters: 6430, time: 0.063) G_GAN: 0.084 G_GAN_Feat: 0.708 G_ID: 0.110 G_Rec: 0.318 D_GP: 0.034 D_real: 0.986 D_fake: 0.916 +(epoch: 323, iters: 6830, time: 0.063) G_GAN: 0.412 G_GAN_Feat: 0.950 G_ID: 0.138 G_Rec: 0.417 D_GP: 0.054 D_real: 0.910 D_fake: 0.595 +(epoch: 323, iters: 7230, time: 0.063) G_GAN: 0.086 G_GAN_Feat: 0.686 G_ID: 0.104 G_Rec: 0.295 D_GP: 0.050 D_real: 0.891 D_fake: 0.914 +(epoch: 323, iters: 7630, time: 0.064) G_GAN: 0.532 G_GAN_Feat: 0.988 G_ID: 0.146 G_Rec: 0.419 D_GP: 0.057 D_real: 0.904 D_fake: 0.477 +(epoch: 323, iters: 8030, time: 0.063) G_GAN: 0.359 G_GAN_Feat: 0.760 G_ID: 0.099 G_Rec: 0.311 D_GP: 0.032 D_real: 1.181 D_fake: 0.642 +(epoch: 323, iters: 8430, time: 0.063) G_GAN: 0.333 G_GAN_Feat: 0.950 G_ID: 0.130 G_Rec: 0.421 D_GP: 0.039 D_real: 0.905 D_fake: 0.668 +(epoch: 324, iters: 222, time: 0.063) G_GAN: 0.507 G_GAN_Feat: 0.759 G_ID: 0.104 G_Rec: 0.308 D_GP: 0.032 D_real: 1.287 D_fake: 0.496 +(epoch: 324, iters: 622, time: 0.064) G_GAN: 0.512 G_GAN_Feat: 0.912 G_ID: 0.153 G_Rec: 0.417 D_GP: 0.029 D_real: 1.219 D_fake: 0.494 +(epoch: 324, iters: 1022, time: 0.063) G_GAN: -0.085 G_GAN_Feat: 0.767 G_ID: 0.106 G_Rec: 0.330 D_GP: 0.061 D_real: 0.564 D_fake: 1.085 +(epoch: 324, iters: 1422, time: 0.063) G_GAN: 0.293 G_GAN_Feat: 1.016 G_ID: 0.129 G_Rec: 0.450 D_GP: 0.056 D_real: 1.147 D_fake: 0.713 +(epoch: 324, iters: 1822, time: 0.063) G_GAN: 0.129 G_GAN_Feat: 0.705 G_ID: 0.128 G_Rec: 0.299 D_GP: 0.035 D_real: 0.938 D_fake: 0.871 +(epoch: 324, iters: 2222, time: 0.063) G_GAN: 0.484 G_GAN_Feat: 0.904 G_ID: 0.138 G_Rec: 0.419 D_GP: 0.034 D_real: 1.059 D_fake: 0.519 +(epoch: 324, iters: 2622, time: 0.063) G_GAN: 0.438 G_GAN_Feat: 0.840 G_ID: 0.119 G_Rec: 0.332 D_GP: 0.077 D_real: 0.938 D_fake: 0.566 +(epoch: 324, iters: 3022, time: 0.063) G_GAN: 0.367 G_GAN_Feat: 1.145 G_ID: 0.123 G_Rec: 0.464 D_GP: 0.910 D_real: 0.545 D_fake: 0.687 +(epoch: 324, iters: 3422, time: 0.064) G_GAN: 0.087 G_GAN_Feat: 0.736 G_ID: 0.082 G_Rec: 0.315 D_GP: 0.023 D_real: 1.002 D_fake: 0.913 +(epoch: 324, iters: 3822, time: 0.063) G_GAN: 0.280 G_GAN_Feat: 0.874 G_ID: 0.121 G_Rec: 0.399 D_GP: 0.029 D_real: 0.971 D_fake: 0.721 +(epoch: 324, iters: 4222, time: 0.063) G_GAN: -0.046 G_GAN_Feat: 0.760 G_ID: 0.109 G_Rec: 0.317 D_GP: 0.062 D_real: 0.731 D_fake: 1.046 +(epoch: 324, iters: 4622, time: 0.063) G_GAN: 0.562 G_GAN_Feat: 0.928 G_ID: 0.117 G_Rec: 0.384 D_GP: 0.040 D_real: 1.144 D_fake: 0.459 +(epoch: 324, iters: 5022, time: 0.064) G_GAN: 0.140 G_GAN_Feat: 0.762 G_ID: 0.128 G_Rec: 0.310 D_GP: 0.117 D_real: 0.812 D_fake: 0.861 +(epoch: 324, iters: 5422, time: 0.063) G_GAN: 0.559 G_GAN_Feat: 0.894 G_ID: 0.112 G_Rec: 0.392 D_GP: 0.035 D_real: 1.188 D_fake: 0.444 +(epoch: 324, iters: 5822, time: 0.063) G_GAN: 0.090 G_GAN_Feat: 0.760 G_ID: 0.119 G_Rec: 0.334 D_GP: 0.028 D_real: 0.989 D_fake: 0.910 +(epoch: 324, iters: 6222, time: 0.063) G_GAN: 0.475 G_GAN_Feat: 0.913 G_ID: 0.130 G_Rec: 0.448 D_GP: 0.028 D_real: 1.114 D_fake: 0.539 +(epoch: 324, iters: 6622, time: 0.064) G_GAN: 0.138 G_GAN_Feat: 0.671 G_ID: 0.105 G_Rec: 0.288 D_GP: 0.031 D_real: 1.041 D_fake: 0.862 +(epoch: 324, iters: 7022, time: 0.063) G_GAN: 0.542 G_GAN_Feat: 0.921 G_ID: 0.138 G_Rec: 0.403 D_GP: 0.047 D_real: 1.120 D_fake: 0.463 +(epoch: 324, iters: 7422, time: 0.063) G_GAN: 0.528 G_GAN_Feat: 0.802 G_ID: 0.100 G_Rec: 0.334 D_GP: 0.055 D_real: 1.200 D_fake: 0.494 +(epoch: 324, iters: 7822, time: 0.063) G_GAN: 0.855 G_GAN_Feat: 1.003 G_ID: 0.145 G_Rec: 0.424 D_GP: 0.121 D_real: 1.269 D_fake: 0.303 +(epoch: 324, iters: 8222, time: 0.064) G_GAN: 0.380 G_GAN_Feat: 0.750 G_ID: 0.117 G_Rec: 0.311 D_GP: 0.031 D_real: 1.209 D_fake: 0.627 +(epoch: 324, iters: 8622, time: 0.063) G_GAN: 0.396 G_GAN_Feat: 0.998 G_ID: 0.139 G_Rec: 0.440 D_GP: 0.030 D_real: 0.977 D_fake: 0.608 +(epoch: 325, iters: 414, time: 0.063) G_GAN: -0.025 G_GAN_Feat: 0.742 G_ID: 0.092 G_Rec: 0.297 D_GP: 0.033 D_real: 0.780 D_fake: 1.025 +(epoch: 325, iters: 814, time: 0.063) G_GAN: 0.696 G_GAN_Feat: 0.949 G_ID: 0.122 G_Rec: 0.435 D_GP: 0.041 D_real: 1.202 D_fake: 0.323 +(epoch: 325, iters: 1214, time: 0.064) G_GAN: 0.517 G_GAN_Feat: 0.756 G_ID: 0.100 G_Rec: 0.289 D_GP: 0.032 D_real: 1.253 D_fake: 0.486 +(epoch: 325, iters: 1614, time: 0.063) G_GAN: 0.449 G_GAN_Feat: 1.050 G_ID: 0.128 G_Rec: 0.444 D_GP: 0.043 D_real: 0.862 D_fake: 0.555 +(epoch: 325, iters: 2014, time: 0.063) G_GAN: 0.178 G_GAN_Feat: 1.134 G_ID: 0.113 G_Rec: 0.350 D_GP: 0.029 D_real: 1.547 D_fake: 0.835 +(epoch: 325, iters: 2414, time: 0.063) G_GAN: 0.224 G_GAN_Feat: 0.788 G_ID: 0.141 G_Rec: 0.376 D_GP: 0.025 D_real: 1.054 D_fake: 0.779 +(epoch: 325, iters: 2814, time: 0.064) G_GAN: -0.105 G_GAN_Feat: 0.661 G_ID: 0.106 G_Rec: 0.283 D_GP: 0.023 D_real: 0.806 D_fake: 1.106 +(epoch: 325, iters: 3214, time: 0.063) G_GAN: 0.405 G_GAN_Feat: 0.924 G_ID: 0.130 G_Rec: 0.444 D_GP: 0.036 D_real: 1.056 D_fake: 0.624 +(epoch: 325, iters: 3614, time: 0.063) G_GAN: -0.100 G_GAN_Feat: 0.715 G_ID: 0.129 G_Rec: 0.337 D_GP: 0.030 D_real: 0.651 D_fake: 1.100 +(epoch: 325, iters: 4014, time: 0.063) G_GAN: 0.413 G_GAN_Feat: 0.923 G_ID: 0.123 G_Rec: 0.416 D_GP: 0.032 D_real: 1.038 D_fake: 0.589 +(epoch: 325, iters: 4414, time: 0.064) G_GAN: 0.096 G_GAN_Feat: 0.697 G_ID: 0.115 G_Rec: 0.298 D_GP: 0.035 D_real: 0.861 D_fake: 0.904 +(epoch: 325, iters: 4814, time: 0.063) G_GAN: 0.328 G_GAN_Feat: 0.896 G_ID: 0.143 G_Rec: 0.388 D_GP: 0.050 D_real: 0.910 D_fake: 0.673 +(epoch: 325, iters: 5214, time: 0.063) G_GAN: 0.227 G_GAN_Feat: 0.854 G_ID: 0.104 G_Rec: 0.337 D_GP: 0.063 D_real: 0.694 D_fake: 0.773 +(epoch: 325, iters: 5614, time: 0.063) G_GAN: 0.540 G_GAN_Feat: 0.969 G_ID: 0.130 G_Rec: 0.413 D_GP: 0.060 D_real: 0.936 D_fake: 0.466 +(epoch: 325, iters: 6014, time: 0.064) G_GAN: 0.191 G_GAN_Feat: 0.684 G_ID: 0.110 G_Rec: 0.296 D_GP: 0.028 D_real: 1.085 D_fake: 0.809 +(epoch: 325, iters: 6414, time: 0.063) G_GAN: 0.452 G_GAN_Feat: 1.129 G_ID: 0.172 G_Rec: 0.501 D_GP: 0.130 D_real: 0.462 D_fake: 0.569 +(epoch: 325, iters: 6814, time: 0.063) G_GAN: 0.252 G_GAN_Feat: 0.823 G_ID: 0.098 G_Rec: 0.307 D_GP: 0.048 D_real: 0.784 D_fake: 0.749 +(epoch: 325, iters: 7214, time: 0.063) G_GAN: 0.278 G_GAN_Feat: 0.965 G_ID: 0.136 G_Rec: 0.431 D_GP: 0.035 D_real: 0.846 D_fake: 0.724 +(epoch: 325, iters: 7614, time: 0.064) G_GAN: 0.255 G_GAN_Feat: 0.807 G_ID: 0.130 G_Rec: 0.326 D_GP: 0.034 D_real: 0.883 D_fake: 0.745 +(epoch: 325, iters: 8014, time: 0.063) G_GAN: 0.987 G_GAN_Feat: 1.171 G_ID: 0.135 G_Rec: 0.466 D_GP: 0.117 D_real: 0.844 D_fake: 0.141 +(epoch: 325, iters: 8414, time: 0.063) G_GAN: 0.191 G_GAN_Feat: 0.797 G_ID: 0.114 G_Rec: 0.303 D_GP: 0.037 D_real: 1.001 D_fake: 0.818 +(epoch: 326, iters: 206, time: 0.063) G_GAN: 0.378 G_GAN_Feat: 1.044 G_ID: 0.140 G_Rec: 0.476 D_GP: 0.038 D_real: 0.851 D_fake: 0.623 +(epoch: 326, iters: 606, time: 0.064) G_GAN: 0.216 G_GAN_Feat: 0.673 G_ID: 0.094 G_Rec: 0.273 D_GP: 0.031 D_real: 1.107 D_fake: 0.784 +(epoch: 326, iters: 1006, time: 0.063) G_GAN: 0.723 G_GAN_Feat: 0.964 G_ID: 0.126 G_Rec: 0.429 D_GP: 0.047 D_real: 1.243 D_fake: 0.312 +(epoch: 326, iters: 1406, time: 0.063) G_GAN: 0.349 G_GAN_Feat: 0.796 G_ID: 0.097 G_Rec: 0.300 D_GP: 0.046 D_real: 1.023 D_fake: 0.652 +(epoch: 326, iters: 1806, time: 0.063) G_GAN: 0.673 G_GAN_Feat: 1.133 G_ID: 0.132 G_Rec: 0.436 D_GP: 0.106 D_real: 0.628 D_fake: 0.344 +(epoch: 326, iters: 2206, time: 0.064) G_GAN: 0.176 G_GAN_Feat: 1.108 G_ID: 0.108 G_Rec: 0.367 D_GP: 0.129 D_real: 1.109 D_fake: 0.858 +(epoch: 326, iters: 2606, time: 0.063) G_GAN: 0.499 G_GAN_Feat: 1.036 G_ID: 0.127 G_Rec: 0.450 D_GP: 0.039 D_real: 0.952 D_fake: 0.522 +(epoch: 326, iters: 3006, time: 0.063) G_GAN: 0.157 G_GAN_Feat: 0.775 G_ID: 0.116 G_Rec: 0.310 D_GP: 0.045 D_real: 0.948 D_fake: 0.844 +(epoch: 326, iters: 3406, time: 0.063) G_GAN: 0.756 G_GAN_Feat: 1.238 G_ID: 0.129 G_Rec: 0.464 D_GP: 0.052 D_real: 0.470 D_fake: 0.291 +(epoch: 326, iters: 3806, time: 0.064) G_GAN: 0.068 G_GAN_Feat: 0.796 G_ID: 0.122 G_Rec: 0.346 D_GP: 0.039 D_real: 0.771 D_fake: 0.932 +(epoch: 326, iters: 4206, time: 0.063) G_GAN: 0.642 G_GAN_Feat: 1.234 G_ID: 0.139 G_Rec: 0.483 D_GP: 0.076 D_real: 0.460 D_fake: 0.379 +(epoch: 326, iters: 4606, time: 0.063) G_GAN: -0.075 G_GAN_Feat: 0.911 G_ID: 0.114 G_Rec: 0.346 D_GP: 0.071 D_real: 0.270 D_fake: 1.077 +(epoch: 326, iters: 5006, time: 0.063) G_GAN: 0.778 G_GAN_Feat: 1.176 G_ID: 0.114 G_Rec: 0.455 D_GP: 0.044 D_real: 0.778 D_fake: 0.274 +(epoch: 326, iters: 5406, time: 0.064) G_GAN: 0.216 G_GAN_Feat: 0.736 G_ID: 0.127 G_Rec: 0.303 D_GP: 0.045 D_real: 1.082 D_fake: 0.795 +(epoch: 326, iters: 5806, time: 0.063) G_GAN: 0.502 G_GAN_Feat: 1.104 G_ID: 0.119 G_Rec: 0.455 D_GP: 0.038 D_real: 0.982 D_fake: 0.503 +(epoch: 326, iters: 6206, time: 0.064) G_GAN: 0.505 G_GAN_Feat: 0.827 G_ID: 0.111 G_Rec: 0.349 D_GP: 0.038 D_real: 1.262 D_fake: 0.604 +(epoch: 326, iters: 6606, time: 0.063) G_GAN: 0.201 G_GAN_Feat: 0.938 G_ID: 0.133 G_Rec: 0.385 D_GP: 0.036 D_real: 0.853 D_fake: 0.799 +(epoch: 326, iters: 7006, time: 0.064) G_GAN: 0.244 G_GAN_Feat: 0.828 G_ID: 0.083 G_Rec: 0.314 D_GP: 0.046 D_real: 0.894 D_fake: 0.756 +(epoch: 326, iters: 7406, time: 0.063) G_GAN: 0.424 G_GAN_Feat: 1.009 G_ID: 0.142 G_Rec: 0.453 D_GP: 0.032 D_real: 0.929 D_fake: 0.578 +(epoch: 326, iters: 7806, time: 0.063) G_GAN: 0.236 G_GAN_Feat: 0.729 G_ID: 0.107 G_Rec: 0.335 D_GP: 0.026 D_real: 1.219 D_fake: 0.764 +(epoch: 326, iters: 8206, time: 0.063) G_GAN: 0.654 G_GAN_Feat: 0.914 G_ID: 0.132 G_Rec: 0.446 D_GP: 0.030 D_real: 1.304 D_fake: 0.376 +(epoch: 326, iters: 8606, time: 0.064) G_GAN: 0.283 G_GAN_Feat: 0.721 G_ID: 0.089 G_Rec: 0.309 D_GP: 0.041 D_real: 1.120 D_fake: 0.717 +(epoch: 327, iters: 398, time: 0.063) G_GAN: 0.639 G_GAN_Feat: 0.908 G_ID: 0.115 G_Rec: 0.421 D_GP: 0.046 D_real: 1.194 D_fake: 0.403 +(epoch: 327, iters: 798, time: 0.063) G_GAN: 0.059 G_GAN_Feat: 0.650 G_ID: 0.107 G_Rec: 0.299 D_GP: 0.038 D_real: 0.928 D_fake: 0.941 +(epoch: 327, iters: 1198, time: 0.063) G_GAN: 0.042 G_GAN_Feat: 0.985 G_ID: 0.155 G_Rec: 0.463 D_GP: 0.071 D_real: 0.489 D_fake: 0.958 +(epoch: 327, iters: 1598, time: 0.064) G_GAN: 0.204 G_GAN_Feat: 0.726 G_ID: 0.127 G_Rec: 0.328 D_GP: 0.038 D_real: 0.961 D_fake: 0.796 +(epoch: 327, iters: 1998, time: 0.063) G_GAN: 0.863 G_GAN_Feat: 0.893 G_ID: 0.115 G_Rec: 0.422 D_GP: 0.027 D_real: 1.579 D_fake: 0.191 +(epoch: 327, iters: 2398, time: 0.063) G_GAN: 0.163 G_GAN_Feat: 0.693 G_ID: 0.102 G_Rec: 0.300 D_GP: 0.040 D_real: 1.025 D_fake: 0.837 +(epoch: 327, iters: 2798, time: 0.063) G_GAN: 0.493 G_GAN_Feat: 1.035 G_ID: 0.127 G_Rec: 0.469 D_GP: 0.060 D_real: 0.759 D_fake: 0.520 +(epoch: 327, iters: 3198, time: 0.064) G_GAN: 0.090 G_GAN_Feat: 0.983 G_ID: 0.115 G_Rec: 0.355 D_GP: 0.368 D_real: 0.338 D_fake: 0.911 +(epoch: 327, iters: 3598, time: 0.063) G_GAN: 0.377 G_GAN_Feat: 0.980 G_ID: 0.116 G_Rec: 0.421 D_GP: 0.045 D_real: 0.873 D_fake: 0.626 +(epoch: 327, iters: 3998, time: 0.063) G_GAN: 0.214 G_GAN_Feat: 0.684 G_ID: 0.099 G_Rec: 0.283 D_GP: 0.033 D_real: 1.106 D_fake: 0.786 +(epoch: 327, iters: 4398, time: 0.063) G_GAN: 0.627 G_GAN_Feat: 1.123 G_ID: 0.126 G_Rec: 0.447 D_GP: 0.079 D_real: 0.604 D_fake: 0.418 +(epoch: 327, iters: 4798, time: 0.064) G_GAN: 0.413 G_GAN_Feat: 0.894 G_ID: 0.103 G_Rec: 0.340 D_GP: 0.042 D_real: 0.792 D_fake: 0.596 +(epoch: 327, iters: 5198, time: 0.063) G_GAN: 0.746 G_GAN_Feat: 1.153 G_ID: 0.151 G_Rec: 0.467 D_GP: 0.086 D_real: 0.612 D_fake: 0.314 +(epoch: 327, iters: 5598, time: 0.063) G_GAN: 0.098 G_GAN_Feat: 0.657 G_ID: 0.105 G_Rec: 0.305 D_GP: 0.032 D_real: 1.047 D_fake: 0.902 +(epoch: 327, iters: 5998, time: 0.063) G_GAN: 0.351 G_GAN_Feat: 0.857 G_ID: 0.139 G_Rec: 0.399 D_GP: 0.031 D_real: 1.009 D_fake: 0.651 +(epoch: 327, iters: 6398, time: 0.064) G_GAN: 0.213 G_GAN_Feat: 0.650 G_ID: 0.112 G_Rec: 0.316 D_GP: 0.029 D_real: 1.167 D_fake: 0.788 +(epoch: 327, iters: 6798, time: 0.063) G_GAN: 0.394 G_GAN_Feat: 0.912 G_ID: 0.145 G_Rec: 0.443 D_GP: 0.059 D_real: 0.978 D_fake: 0.620 +(epoch: 327, iters: 7198, time: 0.063) G_GAN: 0.090 G_GAN_Feat: 0.700 G_ID: 0.115 G_Rec: 0.309 D_GP: 0.039 D_real: 0.892 D_fake: 0.910 +(epoch: 327, iters: 7598, time: 0.063) G_GAN: 0.386 G_GAN_Feat: 0.783 G_ID: 0.139 G_Rec: 0.356 D_GP: 0.030 D_real: 1.133 D_fake: 0.619 +(epoch: 327, iters: 7998, time: 0.064) G_GAN: 0.279 G_GAN_Feat: 0.763 G_ID: 0.114 G_Rec: 0.312 D_GP: 0.085 D_real: 0.989 D_fake: 0.730 +(epoch: 327, iters: 8398, time: 0.063) G_GAN: 0.482 G_GAN_Feat: 0.838 G_ID: 0.119 G_Rec: 0.397 D_GP: 0.030 D_real: 1.196 D_fake: 0.540 +(epoch: 328, iters: 190, time: 0.063) G_GAN: -0.165 G_GAN_Feat: 0.827 G_ID: 0.114 G_Rec: 0.330 D_GP: 0.081 D_real: 0.548 D_fake: 1.165 +(epoch: 328, iters: 590, time: 0.063) G_GAN: 0.394 G_GAN_Feat: 0.889 G_ID: 0.127 G_Rec: 0.409 D_GP: 0.030 D_real: 1.072 D_fake: 0.609 +(epoch: 328, iters: 990, time: 0.064) G_GAN: 0.074 G_GAN_Feat: 0.754 G_ID: 0.100 G_Rec: 0.331 D_GP: 0.031 D_real: 0.896 D_fake: 0.926 +(epoch: 328, iters: 1390, time: 0.063) G_GAN: 0.666 G_GAN_Feat: 1.010 G_ID: 0.117 G_Rec: 0.454 D_GP: 0.037 D_real: 1.062 D_fake: 0.376 +(epoch: 328, iters: 1790, time: 0.063) G_GAN: 0.025 G_GAN_Feat: 0.840 G_ID: 0.112 G_Rec: 0.333 D_GP: 0.068 D_real: 0.535 D_fake: 0.975 +(epoch: 328, iters: 2190, time: 0.063) G_GAN: 0.491 G_GAN_Feat: 0.917 G_ID: 0.123 G_Rec: 0.406 D_GP: 0.038 D_real: 1.099 D_fake: 0.519 +(epoch: 328, iters: 2590, time: 0.064) G_GAN: 0.184 G_GAN_Feat: 0.890 G_ID: 0.101 G_Rec: 0.342 D_GP: 1.264 D_real: 0.495 D_fake: 0.821 +(epoch: 328, iters: 2990, time: 0.063) G_GAN: 0.127 G_GAN_Feat: 0.905 G_ID: 0.149 G_Rec: 0.406 D_GP: 0.054 D_real: 0.720 D_fake: 0.874 +(epoch: 328, iters: 3390, time: 0.064) G_GAN: 0.186 G_GAN_Feat: 0.767 G_ID: 0.115 G_Rec: 0.317 D_GP: 0.043 D_real: 0.865 D_fake: 0.815 +(epoch: 328, iters: 3790, time: 0.064) G_GAN: 0.378 G_GAN_Feat: 0.838 G_ID: 0.129 G_Rec: 0.398 D_GP: 0.028 D_real: 1.185 D_fake: 0.623 +(epoch: 328, iters: 4190, time: 0.064) G_GAN: 0.187 G_GAN_Feat: 0.695 G_ID: 0.104 G_Rec: 0.302 D_GP: 0.033 D_real: 1.005 D_fake: 0.813 +(epoch: 328, iters: 4590, time: 0.063) G_GAN: 0.368 G_GAN_Feat: 0.980 G_ID: 0.166 G_Rec: 0.445 D_GP: 0.046 D_real: 0.950 D_fake: 0.637 +(epoch: 328, iters: 4990, time: 0.063) G_GAN: -0.170 G_GAN_Feat: 0.887 G_ID: 0.100 G_Rec: 0.330 D_GP: 0.129 D_real: 0.251 D_fake: 1.170 +(epoch: 328, iters: 5390, time: 0.063) G_GAN: 0.502 G_GAN_Feat: 0.976 G_ID: 0.129 G_Rec: 0.454 D_GP: 0.038 D_real: 1.013 D_fake: 0.501 +(epoch: 328, iters: 5790, time: 0.064) G_GAN: 0.211 G_GAN_Feat: 0.836 G_ID: 0.120 G_Rec: 0.333 D_GP: 0.043 D_real: 0.770 D_fake: 0.796 +(epoch: 328, iters: 6190, time: 0.063) G_GAN: 0.801 G_GAN_Feat: 1.225 G_ID: 0.137 G_Rec: 0.487 D_GP: 0.095 D_real: 0.612 D_fake: 0.257 +(epoch: 328, iters: 6590, time: 0.063) G_GAN: 0.402 G_GAN_Feat: 0.676 G_ID: 0.100 G_Rec: 0.334 D_GP: 0.028 D_real: 1.305 D_fake: 0.600 +(epoch: 328, iters: 6990, time: 0.063) G_GAN: 0.414 G_GAN_Feat: 0.905 G_ID: 0.142 G_Rec: 0.443 D_GP: 0.034 D_real: 1.000 D_fake: 0.596 +(epoch: 328, iters: 7390, time: 0.063) G_GAN: 0.010 G_GAN_Feat: 0.796 G_ID: 0.123 G_Rec: 0.349 D_GP: 0.056 D_real: 0.668 D_fake: 0.990 +(epoch: 328, iters: 7790, time: 0.064) G_GAN: 0.389 G_GAN_Feat: 0.873 G_ID: 0.130 G_Rec: 0.376 D_GP: 0.095 D_real: 1.025 D_fake: 0.617 +(epoch: 328, iters: 8190, time: 0.064) G_GAN: 0.442 G_GAN_Feat: 0.678 G_ID: 0.104 G_Rec: 0.292 D_GP: 0.034 D_real: 1.240 D_fake: 0.560 +(epoch: 328, iters: 8590, time: 0.064) G_GAN: 0.535 G_GAN_Feat: 1.042 G_ID: 0.144 G_Rec: 0.460 D_GP: 0.045 D_real: 0.907 D_fake: 0.486 +(epoch: 329, iters: 382, time: 0.064) G_GAN: 0.105 G_GAN_Feat: 0.741 G_ID: 0.114 G_Rec: 0.307 D_GP: 0.039 D_real: 0.762 D_fake: 0.895 +(epoch: 329, iters: 782, time: 0.064) G_GAN: 0.658 G_GAN_Feat: 0.981 G_ID: 0.128 G_Rec: 0.440 D_GP: 0.085 D_real: 1.008 D_fake: 0.360 +(epoch: 329, iters: 1182, time: 0.064) G_GAN: 0.134 G_GAN_Feat: 0.944 G_ID: 0.117 G_Rec: 0.349 D_GP: 0.240 D_real: 0.373 D_fake: 0.867 +(epoch: 329, iters: 1582, time: 0.064) G_GAN: 0.635 G_GAN_Feat: 1.044 G_ID: 0.140 G_Rec: 0.448 D_GP: 0.038 D_real: 1.023 D_fake: 0.383 +(epoch: 329, iters: 1982, time: 0.064) G_GAN: 0.258 G_GAN_Feat: 0.871 G_ID: 0.109 G_Rec: 0.327 D_GP: 0.057 D_real: 0.607 D_fake: 0.764 +(epoch: 329, iters: 2382, time: 0.064) G_GAN: 0.495 G_GAN_Feat: 1.211 G_ID: 0.121 G_Rec: 0.465 D_GP: 1.065 D_real: 0.505 D_fake: 0.552 +(epoch: 329, iters: 2782, time: 0.064) G_GAN: 0.114 G_GAN_Feat: 0.657 G_ID: 0.095 G_Rec: 0.291 D_GP: 0.030 D_real: 0.971 D_fake: 0.887 +(epoch: 329, iters: 3182, time: 0.064) G_GAN: 0.382 G_GAN_Feat: 1.068 G_ID: 0.123 G_Rec: 0.499 D_GP: 0.043 D_real: 0.825 D_fake: 0.621 +(epoch: 329, iters: 3582, time: 0.064) G_GAN: 0.264 G_GAN_Feat: 0.771 G_ID: 0.110 G_Rec: 0.316 D_GP: 0.087 D_real: 0.868 D_fake: 0.740 +(epoch: 329, iters: 3982, time: 0.064) G_GAN: 0.402 G_GAN_Feat: 1.036 G_ID: 0.147 G_Rec: 0.455 D_GP: 0.092 D_real: 0.547 D_fake: 0.602 +(epoch: 329, iters: 4382, time: 0.064) G_GAN: 0.111 G_GAN_Feat: 0.850 G_ID: 0.099 G_Rec: 0.307 D_GP: 0.084 D_real: 0.489 D_fake: 0.908 +(epoch: 329, iters: 4782, time: 0.064) G_GAN: 0.028 G_GAN_Feat: 1.079 G_ID: 0.141 G_Rec: 0.481 D_GP: 0.034 D_real: 0.498 D_fake: 0.972 +(epoch: 329, iters: 5182, time: 0.064) G_GAN: 0.268 G_GAN_Feat: 0.783 G_ID: 0.087 G_Rec: 0.304 D_GP: 0.045 D_real: 0.945 D_fake: 0.733 +(epoch: 329, iters: 5582, time: 0.064) G_GAN: 0.684 G_GAN_Feat: 1.000 G_ID: 0.150 G_Rec: 0.477 D_GP: 0.033 D_real: 1.229 D_fake: 0.335 +(epoch: 329, iters: 5982, time: 0.064) G_GAN: -0.092 G_GAN_Feat: 0.791 G_ID: 0.117 G_Rec: 0.326 D_GP: 0.063 D_real: 0.745 D_fake: 1.093 +(epoch: 329, iters: 6382, time: 0.064) G_GAN: 0.860 G_GAN_Feat: 1.086 G_ID: 0.116 G_Rec: 0.457 D_GP: 0.101 D_real: 0.891 D_fake: 0.231 +(epoch: 329, iters: 6782, time: 0.063) G_GAN: 0.421 G_GAN_Feat: 0.812 G_ID: 0.107 G_Rec: 0.331 D_GP: 0.143 D_real: 0.862 D_fake: 0.588 +(epoch: 329, iters: 7182, time: 0.063) G_GAN: 0.442 G_GAN_Feat: 0.873 G_ID: 0.132 G_Rec: 0.433 D_GP: 0.029 D_real: 1.149 D_fake: 0.559 +(epoch: 329, iters: 7582, time: 0.063) G_GAN: 0.065 G_GAN_Feat: 0.733 G_ID: 0.141 G_Rec: 0.320 D_GP: 0.034 D_real: 0.925 D_fake: 0.935 +(epoch: 329, iters: 7982, time: 0.064) G_GAN: 0.478 G_GAN_Feat: 1.000 G_ID: 0.139 G_Rec: 0.441 D_GP: 0.036 D_real: 1.183 D_fake: 0.549 +(epoch: 329, iters: 8382, time: 0.063) G_GAN: 0.311 G_GAN_Feat: 0.716 G_ID: 0.132 G_Rec: 0.289 D_GP: 0.042 D_real: 1.194 D_fake: 0.690 +(epoch: 330, iters: 174, time: 0.063) G_GAN: 0.700 G_GAN_Feat: 0.909 G_ID: 0.119 G_Rec: 0.445 D_GP: 0.029 D_real: 1.335 D_fake: 0.348 +(epoch: 330, iters: 574, time: 0.063) G_GAN: 0.171 G_GAN_Feat: 0.773 G_ID: 0.108 G_Rec: 0.319 D_GP: 0.047 D_real: 0.782 D_fake: 0.830 +(epoch: 330, iters: 974, time: 0.064) G_GAN: 0.406 G_GAN_Feat: 1.135 G_ID: 0.137 G_Rec: 0.447 D_GP: 0.069 D_real: 0.445 D_fake: 0.600 +(epoch: 330, iters: 1374, time: 0.063) G_GAN: 0.333 G_GAN_Feat: 0.925 G_ID: 0.094 G_Rec: 0.334 D_GP: 0.491 D_real: 0.397 D_fake: 0.703 +(epoch: 330, iters: 1774, time: 0.063) G_GAN: 0.460 G_GAN_Feat: 0.912 G_ID: 0.129 G_Rec: 0.409 D_GP: 0.032 D_real: 1.188 D_fake: 0.555 +(epoch: 330, iters: 2174, time: 0.063) G_GAN: 0.149 G_GAN_Feat: 0.795 G_ID: 0.118 G_Rec: 0.308 D_GP: 0.038 D_real: 0.809 D_fake: 0.853 +(epoch: 330, iters: 2574, time: 0.064) G_GAN: 0.569 G_GAN_Feat: 1.046 G_ID: 0.153 G_Rec: 0.429 D_GP: 0.033 D_real: 0.729 D_fake: 0.434 +(epoch: 330, iters: 2974, time: 0.063) G_GAN: 0.352 G_GAN_Feat: 0.786 G_ID: 0.119 G_Rec: 0.330 D_GP: 0.034 D_real: 1.166 D_fake: 0.656 +(epoch: 330, iters: 3374, time: 0.063) G_GAN: 0.757 G_GAN_Feat: 1.016 G_ID: 0.168 G_Rec: 0.424 D_GP: 0.072 D_real: 1.066 D_fake: 0.299 +(epoch: 330, iters: 3774, time: 0.063) G_GAN: 0.312 G_GAN_Feat: 0.777 G_ID: 0.119 G_Rec: 0.295 D_GP: 0.041 D_real: 1.080 D_fake: 0.699 +(epoch: 330, iters: 4174, time: 0.064) G_GAN: 0.251 G_GAN_Feat: 0.950 G_ID: 0.144 G_Rec: 0.410 D_GP: 0.031 D_real: 0.927 D_fake: 0.750 +(epoch: 330, iters: 4574, time: 0.064) G_GAN: 0.265 G_GAN_Feat: 0.850 G_ID: 0.111 G_Rec: 0.315 D_GP: 0.031 D_real: 0.882 D_fake: 0.752 +(epoch: 330, iters: 4974, time: 0.064) G_GAN: 0.662 G_GAN_Feat: 1.028 G_ID: 0.144 G_Rec: 0.447 D_GP: 0.036 D_real: 1.105 D_fake: 0.353 +(epoch: 330, iters: 5374, time: 0.064) G_GAN: 0.159 G_GAN_Feat: 0.760 G_ID: 0.096 G_Rec: 0.315 D_GP: 0.033 D_real: 0.911 D_fake: 0.844 +(epoch: 330, iters: 5774, time: 0.064) G_GAN: 0.531 G_GAN_Feat: 1.161 G_ID: 0.121 G_Rec: 0.484 D_GP: 0.115 D_real: 0.626 D_fake: 0.503 +(epoch: 330, iters: 6174, time: 0.064) G_GAN: -0.057 G_GAN_Feat: 0.792 G_ID: 0.132 G_Rec: 0.325 D_GP: 0.047 D_real: 0.626 D_fake: 1.057 +(epoch: 330, iters: 6574, time: 0.064) G_GAN: 0.770 G_GAN_Feat: 0.955 G_ID: 0.147 G_Rec: 0.414 D_GP: 0.033 D_real: 1.111 D_fake: 0.307 +(epoch: 330, iters: 6974, time: 0.064) G_GAN: 0.188 G_GAN_Feat: 0.703 G_ID: 0.111 G_Rec: 0.282 D_GP: 0.051 D_real: 1.009 D_fake: 0.812 +(epoch: 330, iters: 7374, time: 0.064) G_GAN: 0.563 G_GAN_Feat: 1.023 G_ID: 0.140 G_Rec: 0.445 D_GP: 0.055 D_real: 0.960 D_fake: 0.445 +(epoch: 330, iters: 7774, time: 0.064) G_GAN: 0.279 G_GAN_Feat: 0.833 G_ID: 0.098 G_Rec: 0.322 D_GP: 0.043 D_real: 0.951 D_fake: 0.721 +(epoch: 330, iters: 8174, time: 0.064) G_GAN: 0.011 G_GAN_Feat: 1.101 G_ID: 0.131 G_Rec: 0.469 D_GP: 0.110 D_real: 0.911 D_fake: 1.004 +(epoch: 330, iters: 8574, time: 0.064) G_GAN: -0.030 G_GAN_Feat: 0.650 G_ID: 0.118 G_Rec: 0.307 D_GP: 0.026 D_real: 0.956 D_fake: 1.031 +(epoch: 331, iters: 366, time: 0.064) G_GAN: 0.392 G_GAN_Feat: 0.840 G_ID: 0.134 G_Rec: 0.439 D_GP: 0.029 D_real: 1.108 D_fake: 0.619 +(epoch: 331, iters: 766, time: 0.064) G_GAN: -0.016 G_GAN_Feat: 0.652 G_ID: 0.111 G_Rec: 0.292 D_GP: 0.035 D_real: 0.826 D_fake: 1.016 +(epoch: 331, iters: 1166, time: 0.064) G_GAN: 0.187 G_GAN_Feat: 0.911 G_ID: 0.114 G_Rec: 0.440 D_GP: 0.034 D_real: 0.856 D_fake: 0.813 +(epoch: 331, iters: 1566, time: 0.064) G_GAN: -0.075 G_GAN_Feat: 0.660 G_ID: 0.115 G_Rec: 0.323 D_GP: 0.031 D_real: 0.776 D_fake: 1.075 +(epoch: 331, iters: 1966, time: 0.064) G_GAN: 0.338 G_GAN_Feat: 0.845 G_ID: 0.129 G_Rec: 0.389 D_GP: 0.035 D_real: 0.949 D_fake: 0.662 +(epoch: 331, iters: 2366, time: 0.064) G_GAN: 0.130 G_GAN_Feat: 0.677 G_ID: 0.098 G_Rec: 0.288 D_GP: 0.036 D_real: 0.978 D_fake: 0.870 +(epoch: 331, iters: 2766, time: 0.064) G_GAN: 0.501 G_GAN_Feat: 0.978 G_ID: 0.114 G_Rec: 0.471 D_GP: 0.050 D_real: 0.968 D_fake: 0.522 +(epoch: 331, iters: 3166, time: 0.064) G_GAN: 0.094 G_GAN_Feat: 0.696 G_ID: 0.108 G_Rec: 0.301 D_GP: 0.035 D_real: 0.977 D_fake: 0.906 +(epoch: 331, iters: 3566, time: 0.064) G_GAN: 0.168 G_GAN_Feat: 1.085 G_ID: 0.103 G_Rec: 0.477 D_GP: 0.085 D_real: 0.524 D_fake: 0.839 +(epoch: 331, iters: 3966, time: 0.063) G_GAN: 0.072 G_GAN_Feat: 0.782 G_ID: 0.123 G_Rec: 0.333 D_GP: 0.073 D_real: 0.745 D_fake: 0.929 +(epoch: 331, iters: 4366, time: 0.063) G_GAN: 0.541 G_GAN_Feat: 0.975 G_ID: 0.135 G_Rec: 0.419 D_GP: 0.066 D_real: 1.157 D_fake: 0.507 +(epoch: 331, iters: 4766, time: 0.063) G_GAN: 0.233 G_GAN_Feat: 0.704 G_ID: 0.126 G_Rec: 0.314 D_GP: 0.029 D_real: 1.090 D_fake: 0.768 +(epoch: 331, iters: 5166, time: 0.064) G_GAN: 0.594 G_GAN_Feat: 0.916 G_ID: 0.122 G_Rec: 0.431 D_GP: 0.050 D_real: 1.023 D_fake: 0.439 +(epoch: 331, iters: 5566, time: 0.063) G_GAN: 0.147 G_GAN_Feat: 0.787 G_ID: 0.104 G_Rec: 0.346 D_GP: 0.084 D_real: 0.817 D_fake: 0.856 +(epoch: 331, iters: 5966, time: 0.063) G_GAN: 0.659 G_GAN_Feat: 0.956 G_ID: 0.126 G_Rec: 0.410 D_GP: 0.054 D_real: 1.133 D_fake: 0.412 +(epoch: 331, iters: 6366, time: 0.063) G_GAN: 0.058 G_GAN_Feat: 0.786 G_ID: 0.114 G_Rec: 0.315 D_GP: 0.043 D_real: 0.741 D_fake: 0.942 +(epoch: 331, iters: 6766, time: 0.064) G_GAN: 0.125 G_GAN_Feat: 0.929 G_ID: 0.119 G_Rec: 0.446 D_GP: 0.028 D_real: 0.755 D_fake: 0.875 +(epoch: 331, iters: 7166, time: 0.063) G_GAN: 0.507 G_GAN_Feat: 0.860 G_ID: 0.112 G_Rec: 0.351 D_GP: 0.094 D_real: 0.923 D_fake: 0.509 +(epoch: 331, iters: 7566, time: 0.063) G_GAN: 0.473 G_GAN_Feat: 0.863 G_ID: 0.113 G_Rec: 0.383 D_GP: 0.029 D_real: 1.120 D_fake: 0.533 +(epoch: 331, iters: 7966, time: 0.063) G_GAN: 0.234 G_GAN_Feat: 0.865 G_ID: 0.105 G_Rec: 0.376 D_GP: 0.326 D_real: 0.721 D_fake: 0.804 +(epoch: 331, iters: 8366, time: 0.064) G_GAN: 0.477 G_GAN_Feat: 0.854 G_ID: 0.148 G_Rec: 0.370 D_GP: 0.032 D_real: 1.146 D_fake: 0.528 +(epoch: 332, iters: 158, time: 0.063) G_GAN: 0.222 G_GAN_Feat: 0.725 G_ID: 0.107 G_Rec: 0.298 D_GP: 0.033 D_real: 1.077 D_fake: 0.778 +(epoch: 332, iters: 558, time: 0.063) G_GAN: 0.270 G_GAN_Feat: 0.933 G_ID: 0.168 G_Rec: 0.405 D_GP: 0.033 D_real: 0.860 D_fake: 0.730 +(epoch: 332, iters: 958, time: 0.063) G_GAN: 0.124 G_GAN_Feat: 0.822 G_ID: 0.119 G_Rec: 0.342 D_GP: 0.044 D_real: 0.781 D_fake: 0.876 +(epoch: 332, iters: 1358, time: 0.064) G_GAN: 0.322 G_GAN_Feat: 0.929 G_ID: 0.145 G_Rec: 0.413 D_GP: 0.050 D_real: 0.781 D_fake: 0.679 +(epoch: 332, iters: 1758, time: 0.063) G_GAN: 0.183 G_GAN_Feat: 0.841 G_ID: 0.090 G_Rec: 0.351 D_GP: 0.081 D_real: 0.849 D_fake: 0.818 +(epoch: 332, iters: 2158, time: 0.063) G_GAN: 0.257 G_GAN_Feat: 1.195 G_ID: 0.167 G_Rec: 0.477 D_GP: 0.169 D_real: 0.174 D_fake: 0.755 +(epoch: 332, iters: 2558, time: 0.063) G_GAN: 0.333 G_GAN_Feat: 0.909 G_ID: 0.108 G_Rec: 0.418 D_GP: 0.063 D_real: 0.640 D_fake: 0.668 +(epoch: 332, iters: 2958, time: 0.064) G_GAN: 0.855 G_GAN_Feat: 0.971 G_ID: 0.128 G_Rec: 0.436 D_GP: 0.039 D_real: 1.453 D_fake: 0.215 +(epoch: 332, iters: 3358, time: 0.063) G_GAN: 0.356 G_GAN_Feat: 0.715 G_ID: 0.123 G_Rec: 0.327 D_GP: 0.030 D_real: 1.160 D_fake: 0.646 +(epoch: 332, iters: 3758, time: 0.063) G_GAN: 0.424 G_GAN_Feat: 0.949 G_ID: 0.131 G_Rec: 0.445 D_GP: 0.040 D_real: 1.032 D_fake: 0.595 +(epoch: 332, iters: 4158, time: 0.063) G_GAN: 0.087 G_GAN_Feat: 0.676 G_ID: 0.088 G_Rec: 0.301 D_GP: 0.038 D_real: 0.947 D_fake: 0.913 +(epoch: 332, iters: 4558, time: 0.064) G_GAN: 0.386 G_GAN_Feat: 1.157 G_ID: 0.147 G_Rec: 0.504 D_GP: 0.084 D_real: 0.497 D_fake: 0.624 +(epoch: 332, iters: 4958, time: 0.063) G_GAN: 0.254 G_GAN_Feat: 0.779 G_ID: 0.097 G_Rec: 0.321 D_GP: 0.033 D_real: 0.966 D_fake: 0.746 +(epoch: 332, iters: 5358, time: 0.063) G_GAN: 0.838 G_GAN_Feat: 1.120 G_ID: 0.134 G_Rec: 0.477 D_GP: 0.066 D_real: 0.966 D_fake: 0.236 +(epoch: 332, iters: 5758, time: 0.063) G_GAN: 0.205 G_GAN_Feat: 0.791 G_ID: 0.115 G_Rec: 0.300 D_GP: 0.069 D_real: 0.807 D_fake: 0.796 +(epoch: 332, iters: 6158, time: 0.064) G_GAN: 0.489 G_GAN_Feat: 1.106 G_ID: 0.120 G_Rec: 0.477 D_GP: 0.135 D_real: 0.529 D_fake: 0.526 +(epoch: 332, iters: 6558, time: 0.063) G_GAN: 0.223 G_GAN_Feat: 0.963 G_ID: 0.106 G_Rec: 0.330 D_GP: 0.469 D_real: 0.253 D_fake: 0.781 +(epoch: 332, iters: 6958, time: 0.063) G_GAN: 0.268 G_GAN_Feat: 0.932 G_ID: 0.139 G_Rec: 0.442 D_GP: 0.031 D_real: 1.043 D_fake: 0.733 +(epoch: 332, iters: 7358, time: 0.063) G_GAN: 0.187 G_GAN_Feat: 0.856 G_ID: 0.122 G_Rec: 0.321 D_GP: 0.035 D_real: 0.946 D_fake: 0.815 +(epoch: 332, iters: 7758, time: 0.064) G_GAN: 0.654 G_GAN_Feat: 0.997 G_ID: 0.125 G_Rec: 0.453 D_GP: 0.029 D_real: 1.156 D_fake: 0.387 +(epoch: 332, iters: 8158, time: 0.063) G_GAN: 0.462 G_GAN_Feat: 0.751 G_ID: 0.112 G_Rec: 0.297 D_GP: 0.034 D_real: 1.270 D_fake: 0.539 +(epoch: 332, iters: 8558, time: 0.063) G_GAN: 0.849 G_GAN_Feat: 1.023 G_ID: 0.119 G_Rec: 0.490 D_GP: 0.031 D_real: 1.287 D_fake: 0.259 +(epoch: 333, iters: 350, time: 0.063) G_GAN: 0.116 G_GAN_Feat: 0.766 G_ID: 0.114 G_Rec: 0.330 D_GP: 0.030 D_real: 0.936 D_fake: 0.884 +(epoch: 333, iters: 750, time: 0.064) G_GAN: 0.696 G_GAN_Feat: 0.960 G_ID: 0.129 G_Rec: 0.415 D_GP: 0.030 D_real: 1.200 D_fake: 0.332 +(epoch: 333, iters: 1150, time: 0.063) G_GAN: 0.248 G_GAN_Feat: 0.848 G_ID: 0.115 G_Rec: 0.297 D_GP: 0.051 D_real: 0.717 D_fake: 0.752 +(epoch: 333, iters: 1550, time: 0.063) G_GAN: 0.403 G_GAN_Feat: 0.943 G_ID: 0.134 G_Rec: 0.414 D_GP: 0.043 D_real: 1.020 D_fake: 0.601 +(epoch: 333, iters: 1950, time: 0.063) G_GAN: 0.665 G_GAN_Feat: 0.716 G_ID: 0.112 G_Rec: 0.352 D_GP: 0.046 D_real: 1.502 D_fake: 0.479 +(epoch: 333, iters: 2350, time: 0.064) G_GAN: 0.606 G_GAN_Feat: 0.914 G_ID: 0.120 G_Rec: 0.447 D_GP: 0.030 D_real: 1.258 D_fake: 0.425 +(epoch: 333, iters: 2750, time: 0.063) G_GAN: 0.105 G_GAN_Feat: 0.658 G_ID: 0.108 G_Rec: 0.297 D_GP: 0.031 D_real: 0.947 D_fake: 0.895 +(epoch: 333, iters: 3150, time: 0.063) G_GAN: 0.485 G_GAN_Feat: 0.866 G_ID: 0.137 G_Rec: 0.411 D_GP: 0.041 D_real: 1.044 D_fake: 0.538 +(epoch: 333, iters: 3550, time: 0.063) G_GAN: 0.587 G_GAN_Feat: 0.690 G_ID: 0.094 G_Rec: 0.299 D_GP: 0.029 D_real: 1.515 D_fake: 0.448 +(epoch: 333, iters: 3950, time: 0.063) G_GAN: 0.266 G_GAN_Feat: 0.963 G_ID: 0.157 G_Rec: 0.436 D_GP: 0.065 D_real: 0.727 D_fake: 0.738 +(epoch: 333, iters: 4350, time: 0.063) G_GAN: 0.247 G_GAN_Feat: 0.754 G_ID: 0.104 G_Rec: 0.332 D_GP: 0.036 D_real: 0.985 D_fake: 0.755 +(epoch: 333, iters: 4750, time: 0.063) G_GAN: 0.599 G_GAN_Feat: 1.076 G_ID: 0.131 G_Rec: 0.451 D_GP: 0.239 D_real: 0.539 D_fake: 0.445 +(epoch: 333, iters: 5150, time: 0.064) G_GAN: 0.244 G_GAN_Feat: 0.718 G_ID: 0.095 G_Rec: 0.311 D_GP: 0.033 D_real: 1.185 D_fake: 0.756 +(epoch: 333, iters: 5550, time: 0.063) G_GAN: 0.057 G_GAN_Feat: 1.135 G_ID: 0.122 G_Rec: 0.488 D_GP: 0.065 D_real: 0.804 D_fake: 0.946 +(epoch: 333, iters: 5950, time: 0.063) G_GAN: 0.214 G_GAN_Feat: 0.763 G_ID: 0.134 G_Rec: 0.324 D_GP: 0.061 D_real: 0.818 D_fake: 0.789 +(epoch: 333, iters: 6350, time: 0.063) G_GAN: 0.624 G_GAN_Feat: 0.972 G_ID: 0.137 G_Rec: 0.431 D_GP: 0.037 D_real: 1.092 D_fake: 0.416 +(epoch: 333, iters: 6750, time: 0.064) G_GAN: 0.037 G_GAN_Feat: 0.747 G_ID: 0.099 G_Rec: 0.323 D_GP: 0.035 D_real: 0.965 D_fake: 0.964 +(epoch: 333, iters: 7150, time: 0.063) G_GAN: 0.364 G_GAN_Feat: 0.953 G_ID: 0.133 G_Rec: 0.432 D_GP: 0.037 D_real: 1.023 D_fake: 0.636 +(epoch: 333, iters: 7550, time: 0.063) G_GAN: 0.171 G_GAN_Feat: 0.736 G_ID: 0.132 G_Rec: 0.292 D_GP: 0.033 D_real: 1.048 D_fake: 0.829 +(epoch: 333, iters: 7950, time: 0.063) G_GAN: 0.505 G_GAN_Feat: 1.155 G_ID: 0.133 G_Rec: 0.486 D_GP: 0.037 D_real: 0.825 D_fake: 0.504 +(epoch: 333, iters: 8350, time: 0.064) G_GAN: 0.277 G_GAN_Feat: 0.760 G_ID: 0.125 G_Rec: 0.324 D_GP: 0.031 D_real: 1.148 D_fake: 0.730 +(epoch: 334, iters: 142, time: 0.063) G_GAN: 0.576 G_GAN_Feat: 1.002 G_ID: 0.120 G_Rec: 0.426 D_GP: 0.044 D_real: 1.150 D_fake: 0.461 +(epoch: 334, iters: 542, time: 0.063) G_GAN: 0.370 G_GAN_Feat: 0.891 G_ID: 0.115 G_Rec: 0.368 D_GP: 0.089 D_real: 0.927 D_fake: 0.656 +(epoch: 334, iters: 942, time: 0.063) G_GAN: 0.347 G_GAN_Feat: 1.217 G_ID: 0.129 G_Rec: 0.505 D_GP: 0.486 D_real: 0.311 D_fake: 0.657 +(epoch: 334, iters: 1342, time: 0.064) G_GAN: 0.446 G_GAN_Feat: 0.880 G_ID: 0.096 G_Rec: 0.299 D_GP: 0.034 D_real: 1.101 D_fake: 0.565 +(epoch: 334, iters: 1742, time: 0.063) G_GAN: 0.868 G_GAN_Feat: 1.241 G_ID: 0.127 G_Rec: 0.442 D_GP: 0.054 D_real: 0.620 D_fake: 0.188 +(epoch: 334, iters: 2142, time: 0.063) G_GAN: 0.209 G_GAN_Feat: 1.024 G_ID: 0.112 G_Rec: 0.388 D_GP: 0.360 D_real: 0.089 D_fake: 0.895 +(epoch: 334, iters: 2542, time: 0.063) G_GAN: 0.586 G_GAN_Feat: 1.002 G_ID: 0.133 G_Rec: 0.409 D_GP: 0.031 D_real: 1.127 D_fake: 0.422 +(epoch: 334, iters: 2942, time: 0.064) G_GAN: 0.754 G_GAN_Feat: 0.897 G_ID: 0.105 G_Rec: 0.308 D_GP: 0.044 D_real: 1.163 D_fake: 0.366 +(epoch: 334, iters: 3342, time: 0.063) G_GAN: 0.759 G_GAN_Feat: 1.186 G_ID: 0.116 G_Rec: 0.503 D_GP: 0.040 D_real: 1.337 D_fake: 0.338 +(epoch: 334, iters: 3742, time: 0.063) G_GAN: 0.111 G_GAN_Feat: 0.654 G_ID: 0.116 G_Rec: 0.308 D_GP: 0.024 D_real: 1.099 D_fake: 0.890 +(epoch: 334, iters: 4142, time: 0.063) G_GAN: 0.369 G_GAN_Feat: 0.835 G_ID: 0.113 G_Rec: 0.434 D_GP: 0.026 D_real: 1.162 D_fake: 0.636 +(epoch: 334, iters: 4542, time: 0.064) G_GAN: 0.162 G_GAN_Feat: 0.626 G_ID: 0.094 G_Rec: 0.299 D_GP: 0.027 D_real: 1.122 D_fake: 0.839 +(epoch: 334, iters: 4942, time: 0.063) G_GAN: 0.325 G_GAN_Feat: 0.800 G_ID: 0.130 G_Rec: 0.376 D_GP: 0.029 D_real: 1.115 D_fake: 0.679 +(epoch: 334, iters: 5342, time: 0.063) G_GAN: 0.041 G_GAN_Feat: 0.697 G_ID: 0.131 G_Rec: 0.314 D_GP: 0.031 D_real: 0.934 D_fake: 0.960 +(epoch: 334, iters: 5742, time: 0.063) G_GAN: 0.338 G_GAN_Feat: 0.813 G_ID: 0.123 G_Rec: 0.391 D_GP: 0.027 D_real: 1.098 D_fake: 0.668 +(epoch: 334, iters: 6142, time: 0.064) G_GAN: -0.011 G_GAN_Feat: 0.639 G_ID: 0.123 G_Rec: 0.275 D_GP: 0.034 D_real: 0.844 D_fake: 1.011 +(epoch: 334, iters: 6542, time: 0.063) G_GAN: 0.311 G_GAN_Feat: 0.860 G_ID: 0.131 G_Rec: 0.395 D_GP: 0.042 D_real: 0.920 D_fake: 0.695 +(epoch: 334, iters: 6942, time: 0.063) G_GAN: -0.163 G_GAN_Feat: 0.716 G_ID: 0.116 G_Rec: 0.307 D_GP: 0.038 D_real: 0.620 D_fake: 1.163 +(epoch: 334, iters: 7342, time: 0.063) G_GAN: 0.479 G_GAN_Feat: 0.866 G_ID: 0.115 G_Rec: 0.405 D_GP: 0.034 D_real: 1.123 D_fake: 0.569 +(epoch: 334, iters: 7742, time: 0.064) G_GAN: -0.168 G_GAN_Feat: 0.731 G_ID: 0.105 G_Rec: 0.326 D_GP: 0.079 D_real: 0.681 D_fake: 1.168 +(epoch: 334, iters: 8142, time: 0.063) G_GAN: 0.485 G_GAN_Feat: 0.965 G_ID: 0.146 G_Rec: 0.435 D_GP: 0.042 D_real: 0.906 D_fake: 0.536 +(epoch: 334, iters: 8542, time: 0.063) G_GAN: 0.101 G_GAN_Feat: 0.813 G_ID: 0.118 G_Rec: 0.341 D_GP: 0.044 D_real: 0.799 D_fake: 0.900 +(epoch: 335, iters: 334, time: 0.063) G_GAN: 0.159 G_GAN_Feat: 0.916 G_ID: 0.139 G_Rec: 0.424 D_GP: 0.037 D_real: 0.708 D_fake: 0.842 +(epoch: 335, iters: 734, time: 0.064) G_GAN: -0.005 G_GAN_Feat: 0.740 G_ID: 0.096 G_Rec: 0.315 D_GP: 0.076 D_real: 0.742 D_fake: 1.006 +(epoch: 335, iters: 1134, time: 0.063) G_GAN: 0.693 G_GAN_Feat: 0.946 G_ID: 0.126 G_Rec: 0.432 D_GP: 0.033 D_real: 1.308 D_fake: 0.357 +(epoch: 335, iters: 1534, time: 0.063) G_GAN: 0.051 G_GAN_Feat: 0.682 G_ID: 0.093 G_Rec: 0.292 D_GP: 0.028 D_real: 0.918 D_fake: 0.949 +(epoch: 335, iters: 1934, time: 0.063) G_GAN: 0.721 G_GAN_Feat: 0.993 G_ID: 0.127 G_Rec: 0.433 D_GP: 0.061 D_real: 1.185 D_fake: 0.337 +(epoch: 335, iters: 2334, time: 0.064) G_GAN: 0.093 G_GAN_Feat: 0.693 G_ID: 0.109 G_Rec: 0.304 D_GP: 0.030 D_real: 0.992 D_fake: 0.907 +(epoch: 335, iters: 2734, time: 0.063) G_GAN: 0.353 G_GAN_Feat: 1.129 G_ID: 0.134 G_Rec: 0.509 D_GP: 0.360 D_real: 0.502 D_fake: 0.661 +(epoch: 335, iters: 3134, time: 0.063) G_GAN: 0.390 G_GAN_Feat: 0.706 G_ID: 0.096 G_Rec: 0.315 D_GP: 0.026 D_real: 1.209 D_fake: 0.616 +(epoch: 335, iters: 3534, time: 0.063) G_GAN: 0.576 G_GAN_Feat: 0.990 G_ID: 0.126 G_Rec: 0.450 D_GP: 0.035 D_real: 1.141 D_fake: 0.441 +(epoch: 335, iters: 3934, time: 0.064) G_GAN: 0.287 G_GAN_Feat: 0.714 G_ID: 0.107 G_Rec: 0.335 D_GP: 0.030 D_real: 1.204 D_fake: 0.713 +(epoch: 335, iters: 4334, time: 0.063) G_GAN: 0.506 G_GAN_Feat: 0.934 G_ID: 0.131 G_Rec: 0.437 D_GP: 0.037 D_real: 1.121 D_fake: 0.519 +(epoch: 335, iters: 4734, time: 0.063) G_GAN: -0.340 G_GAN_Feat: 0.774 G_ID: 0.120 G_Rec: 0.320 D_GP: 0.072 D_real: 0.497 D_fake: 1.340 +(epoch: 335, iters: 5134, time: 0.063) G_GAN: 0.518 G_GAN_Feat: 1.083 G_ID: 0.126 G_Rec: 0.465 D_GP: 0.161 D_real: 0.739 D_fake: 0.515 +(epoch: 335, iters: 5534, time: 0.064) G_GAN: 0.385 G_GAN_Feat: 0.793 G_ID: 0.106 G_Rec: 0.305 D_GP: 0.059 D_real: 1.025 D_fake: 0.617 +(epoch: 335, iters: 5934, time: 0.063) G_GAN: 0.493 G_GAN_Feat: 1.079 G_ID: 0.130 G_Rec: 0.473 D_GP: 0.099 D_real: 0.607 D_fake: 0.526 +(epoch: 335, iters: 6334, time: 0.063) G_GAN: -0.003 G_GAN_Feat: 0.784 G_ID: 0.116 G_Rec: 0.334 D_GP: 0.032 D_real: 0.879 D_fake: 1.003 +(epoch: 335, iters: 6734, time: 0.063) G_GAN: 0.446 G_GAN_Feat: 0.946 G_ID: 0.119 G_Rec: 0.463 D_GP: 0.032 D_real: 1.041 D_fake: 0.559 +(epoch: 335, iters: 7134, time: 0.064) G_GAN: 0.250 G_GAN_Feat: 0.805 G_ID: 0.111 G_Rec: 0.313 D_GP: 0.064 D_real: 0.814 D_fake: 0.750 +(epoch: 335, iters: 7534, time: 0.063) G_GAN: 0.867 G_GAN_Feat: 0.998 G_ID: 0.120 G_Rec: 0.419 D_GP: 0.044 D_real: 1.353 D_fake: 0.238 +(epoch: 335, iters: 7934, time: 0.063) G_GAN: 0.296 G_GAN_Feat: 0.790 G_ID: 0.113 G_Rec: 0.339 D_GP: 0.032 D_real: 1.165 D_fake: 0.705 +(epoch: 335, iters: 8334, time: 0.063) G_GAN: 0.574 G_GAN_Feat: 0.822 G_ID: 0.108 G_Rec: 0.419 D_GP: 0.028 D_real: 1.264 D_fake: 0.438 +(epoch: 336, iters: 126, time: 0.064) G_GAN: 0.314 G_GAN_Feat: 0.744 G_ID: 0.107 G_Rec: 0.286 D_GP: 0.045 D_real: 1.081 D_fake: 0.697 +(epoch: 336, iters: 526, time: 0.063) G_GAN: 0.342 G_GAN_Feat: 0.997 G_ID: 0.153 G_Rec: 0.441 D_GP: 0.034 D_real: 0.866 D_fake: 0.665 +(epoch: 336, iters: 926, time: 0.063) G_GAN: 0.184 G_GAN_Feat: 0.738 G_ID: 0.110 G_Rec: 0.297 D_GP: 0.038 D_real: 0.954 D_fake: 0.818 +(epoch: 336, iters: 1326, time: 0.064) G_GAN: 0.804 G_GAN_Feat: 1.140 G_ID: 0.125 G_Rec: 0.536 D_GP: 0.050 D_real: 0.986 D_fake: 0.300 +(epoch: 336, iters: 1726, time: 0.064) G_GAN: 0.262 G_GAN_Feat: 0.884 G_ID: 0.106 G_Rec: 0.330 D_GP: 0.076 D_real: 0.739 D_fake: 0.765 +(epoch: 336, iters: 2126, time: 0.064) G_GAN: 0.331 G_GAN_Feat: 1.021 G_ID: 0.127 G_Rec: 0.437 D_GP: 0.057 D_real: 0.524 D_fake: 0.677 +(epoch: 336, iters: 2526, time: 0.064) G_GAN: 0.093 G_GAN_Feat: 0.770 G_ID: 0.112 G_Rec: 0.354 D_GP: 0.033 D_real: 0.896 D_fake: 0.908 +(epoch: 336, iters: 2926, time: 0.064) G_GAN: 0.336 G_GAN_Feat: 1.097 G_ID: 0.138 G_Rec: 0.434 D_GP: 0.048 D_real: 0.464 D_fake: 0.667 +(epoch: 336, iters: 3326, time: 0.064) G_GAN: 0.193 G_GAN_Feat: 0.747 G_ID: 0.100 G_Rec: 0.306 D_GP: 0.058 D_real: 0.911 D_fake: 0.809 +(epoch: 336, iters: 3726, time: 0.064) G_GAN: 0.391 G_GAN_Feat: 0.977 G_ID: 0.147 G_Rec: 0.463 D_GP: 0.031 D_real: 0.967 D_fake: 0.614 +(epoch: 336, iters: 4126, time: 0.064) G_GAN: 0.478 G_GAN_Feat: 0.743 G_ID: 0.097 G_Rec: 0.296 D_GP: 0.034 D_real: 1.263 D_fake: 0.525 +(epoch: 336, iters: 4526, time: 0.064) G_GAN: 0.371 G_GAN_Feat: 0.978 G_ID: 0.142 G_Rec: 0.443 D_GP: 0.034 D_real: 1.044 D_fake: 0.631 +(epoch: 336, iters: 4926, time: 0.064) G_GAN: 0.161 G_GAN_Feat: 0.807 G_ID: 0.114 G_Rec: 0.343 D_GP: 0.033 D_real: 0.876 D_fake: 0.840 +(epoch: 336, iters: 5326, time: 0.064) G_GAN: 0.559 G_GAN_Feat: 1.205 G_ID: 0.112 G_Rec: 0.465 D_GP: 0.413 D_real: 0.271 D_fake: 0.490 +(epoch: 336, iters: 5726, time: 0.064) G_GAN: 0.041 G_GAN_Feat: 0.688 G_ID: 0.099 G_Rec: 0.292 D_GP: 0.028 D_real: 0.967 D_fake: 0.959 +(epoch: 336, iters: 6126, time: 0.064) G_GAN: 0.522 G_GAN_Feat: 0.955 G_ID: 0.138 G_Rec: 0.432 D_GP: 0.052 D_real: 0.951 D_fake: 0.498 +(epoch: 336, iters: 6526, time: 0.064) G_GAN: 0.333 G_GAN_Feat: 0.908 G_ID: 0.113 G_Rec: 0.330 D_GP: 0.068 D_real: 0.474 D_fake: 0.674 +(epoch: 336, iters: 6926, time: 0.064) G_GAN: 0.542 G_GAN_Feat: 0.961 G_ID: 0.120 G_Rec: 0.509 D_GP: 0.036 D_real: 1.132 D_fake: 0.495 +(epoch: 336, iters: 7326, time: 0.064) G_GAN: 0.169 G_GAN_Feat: 0.609 G_ID: 0.110 G_Rec: 0.280 D_GP: 0.031 D_real: 1.100 D_fake: 0.831 +(epoch: 336, iters: 7726, time: 0.064) G_GAN: 0.252 G_GAN_Feat: 0.910 G_ID: 0.136 G_Rec: 0.445 D_GP: 0.040 D_real: 0.941 D_fake: 0.749 +(epoch: 336, iters: 8126, time: 0.064) G_GAN: 0.233 G_GAN_Feat: 0.766 G_ID: 0.098 G_Rec: 0.326 D_GP: 0.058 D_real: 0.950 D_fake: 0.768 +(epoch: 336, iters: 8526, time: 0.064) G_GAN: 0.113 G_GAN_Feat: 1.005 G_ID: 0.133 G_Rec: 0.478 D_GP: 0.104 D_real: 0.418 D_fake: 0.888 +(epoch: 337, iters: 318, time: 0.064) G_GAN: 0.358 G_GAN_Feat: 0.701 G_ID: 0.110 G_Rec: 0.296 D_GP: 0.030 D_real: 1.228 D_fake: 0.643 +(epoch: 337, iters: 718, time: 0.064) G_GAN: 0.393 G_GAN_Feat: 0.961 G_ID: 0.128 G_Rec: 0.408 D_GP: 0.061 D_real: 0.740 D_fake: 0.619 +(epoch: 337, iters: 1118, time: 0.064) G_GAN: 0.225 G_GAN_Feat: 0.847 G_ID: 0.105 G_Rec: 0.315 D_GP: 0.055 D_real: 0.623 D_fake: 0.776 +(epoch: 337, iters: 1518, time: 0.063) G_GAN: 0.728 G_GAN_Feat: 1.060 G_ID: 0.140 G_Rec: 0.440 D_GP: 0.042 D_real: 1.081 D_fake: 0.315 +(epoch: 337, iters: 1918, time: 0.063) G_GAN: 0.040 G_GAN_Feat: 0.782 G_ID: 0.118 G_Rec: 0.323 D_GP: 0.067 D_real: 0.630 D_fake: 0.960 +(epoch: 337, iters: 2318, time: 0.063) G_GAN: 0.492 G_GAN_Feat: 0.952 G_ID: 0.138 G_Rec: 0.436 D_GP: 0.034 D_real: 1.094 D_fake: 0.512 +(epoch: 337, iters: 2718, time: 0.064) G_GAN: 0.248 G_GAN_Feat: 0.666 G_ID: 0.091 G_Rec: 0.290 D_GP: 0.035 D_real: 1.191 D_fake: 0.752 +(epoch: 337, iters: 3118, time: 0.063) G_GAN: 0.583 G_GAN_Feat: 1.027 G_ID: 0.127 G_Rec: 0.464 D_GP: 0.038 D_real: 1.026 D_fake: 0.438 +(epoch: 337, iters: 3518, time: 0.063) G_GAN: 0.098 G_GAN_Feat: 0.826 G_ID: 0.108 G_Rec: 0.333 D_GP: 0.089 D_real: 0.524 D_fake: 0.903 +(epoch: 337, iters: 3918, time: 0.063) G_GAN: 0.389 G_GAN_Feat: 0.921 G_ID: 0.130 G_Rec: 0.429 D_GP: 0.030 D_real: 1.079 D_fake: 0.615 +(epoch: 337, iters: 4318, time: 0.064) G_GAN: -0.207 G_GAN_Feat: 0.656 G_ID: 0.103 G_Rec: 0.320 D_GP: 0.032 D_real: 0.799 D_fake: 1.207 +(epoch: 337, iters: 4718, time: 0.063) G_GAN: 0.535 G_GAN_Feat: 0.866 G_ID: 0.129 G_Rec: 0.400 D_GP: 0.033 D_real: 1.269 D_fake: 0.483 +(epoch: 337, iters: 5118, time: 0.063) G_GAN: 0.227 G_GAN_Feat: 0.665 G_ID: 0.102 G_Rec: 0.310 D_GP: 0.028 D_real: 1.105 D_fake: 0.773 +(epoch: 337, iters: 5518, time: 0.063) G_GAN: 0.439 G_GAN_Feat: 0.963 G_ID: 0.117 G_Rec: 0.473 D_GP: 0.028 D_real: 1.169 D_fake: 0.565 +(epoch: 337, iters: 5918, time: 0.064) G_GAN: -0.001 G_GAN_Feat: 0.714 G_ID: 0.130 G_Rec: 0.287 D_GP: 0.042 D_real: 0.845 D_fake: 1.001 +(epoch: 337, iters: 6318, time: 0.063) G_GAN: 0.547 G_GAN_Feat: 1.089 G_ID: 0.142 G_Rec: 0.485 D_GP: 0.475 D_real: 0.858 D_fake: 0.543 +(epoch: 337, iters: 6718, time: 0.063) G_GAN: 0.143 G_GAN_Feat: 0.700 G_ID: 0.107 G_Rec: 0.305 D_GP: 0.034 D_real: 1.085 D_fake: 0.857 +(epoch: 337, iters: 7118, time: 0.063) G_GAN: 0.247 G_GAN_Feat: 0.899 G_ID: 0.133 G_Rec: 0.404 D_GP: 0.041 D_real: 0.856 D_fake: 0.754 +(epoch: 337, iters: 7518, time: 0.064) G_GAN: 0.134 G_GAN_Feat: 0.697 G_ID: 0.110 G_Rec: 0.285 D_GP: 0.031 D_real: 1.053 D_fake: 0.866 +(epoch: 337, iters: 7918, time: 0.063) G_GAN: 0.427 G_GAN_Feat: 0.929 G_ID: 0.119 G_Rec: 0.410 D_GP: 0.035 D_real: 1.128 D_fake: 0.579 +(epoch: 337, iters: 8318, time: 0.063) G_GAN: 0.341 G_GAN_Feat: 0.755 G_ID: 0.110 G_Rec: 0.319 D_GP: 0.048 D_real: 1.018 D_fake: 0.660 +(epoch: 338, iters: 110, time: 0.063) G_GAN: 0.574 G_GAN_Feat: 1.122 G_ID: 0.140 G_Rec: 0.468 D_GP: 0.360 D_real: 0.549 D_fake: 0.451 +(epoch: 338, iters: 510, time: 0.063) G_GAN: 0.213 G_GAN_Feat: 0.934 G_ID: 0.107 G_Rec: 0.365 D_GP: 0.377 D_real: 0.391 D_fake: 0.794 +(epoch: 338, iters: 910, time: 0.063) G_GAN: 0.662 G_GAN_Feat: 1.012 G_ID: 0.153 G_Rec: 0.424 D_GP: 0.052 D_real: 0.964 D_fake: 0.364 +(epoch: 338, iters: 1310, time: 0.064) G_GAN: 0.146 G_GAN_Feat: 0.988 G_ID: 0.120 G_Rec: 0.352 D_GP: 0.318 D_real: 0.380 D_fake: 0.857 +(epoch: 338, iters: 1710, time: 0.064) G_GAN: 0.869 G_GAN_Feat: 1.009 G_ID: 0.128 G_Rec: 0.476 D_GP: 0.033 D_real: 1.549 D_fake: 0.188 +(epoch: 338, iters: 2110, time: 0.063) G_GAN: 0.569 G_GAN_Feat: 0.825 G_ID: 0.102 G_Rec: 0.353 D_GP: 0.038 D_real: 1.388 D_fake: 0.444 +(epoch: 338, iters: 2510, time: 0.064) G_GAN: 0.350 G_GAN_Feat: 1.111 G_ID: 0.128 G_Rec: 0.483 D_GP: 0.139 D_real: 0.715 D_fake: 0.662 +(epoch: 338, iters: 2910, time: 0.063) G_GAN: 0.158 G_GAN_Feat: 0.805 G_ID: 0.108 G_Rec: 0.336 D_GP: 0.063 D_real: 0.804 D_fake: 0.842 +(epoch: 338, iters: 3310, time: 0.064) G_GAN: 0.348 G_GAN_Feat: 1.006 G_ID: 0.147 G_Rec: 0.419 D_GP: 0.051 D_real: 0.725 D_fake: 0.660 +(epoch: 338, iters: 3710, time: 0.063) G_GAN: 0.151 G_GAN_Feat: 0.841 G_ID: 0.104 G_Rec: 0.354 D_GP: 0.031 D_real: 0.950 D_fake: 0.849 +(epoch: 338, iters: 4110, time: 0.063) G_GAN: 0.441 G_GAN_Feat: 1.120 G_ID: 0.127 G_Rec: 0.505 D_GP: 0.125 D_real: 0.294 D_fake: 0.586 +(epoch: 338, iters: 4510, time: 0.063) G_GAN: -0.428 G_GAN_Feat: 0.842 G_ID: 0.113 G_Rec: 0.326 D_GP: 0.043 D_real: 0.431 D_fake: 1.428 +(epoch: 338, iters: 4910, time: 0.064) G_GAN: 0.325 G_GAN_Feat: 0.902 G_ID: 0.145 G_Rec: 0.416 D_GP: 0.034 D_real: 0.997 D_fake: 0.675 +(epoch: 338, iters: 5310, time: 0.063) G_GAN: 0.200 G_GAN_Feat: 0.865 G_ID: 0.116 G_Rec: 0.339 D_GP: 0.078 D_real: 0.542 D_fake: 0.801 +(epoch: 338, iters: 5710, time: 0.063) G_GAN: 0.303 G_GAN_Feat: 1.165 G_ID: 0.142 G_Rec: 0.487 D_GP: 0.244 D_real: 0.513 D_fake: 0.715 +(epoch: 338, iters: 6110, time: 0.063) G_GAN: 0.200 G_GAN_Feat: 1.136 G_ID: 0.095 G_Rec: 0.384 D_GP: 1.795 D_real: 0.811 D_fake: 0.805 +(epoch: 338, iters: 6510, time: 0.064) G_GAN: 0.158 G_GAN_Feat: 0.893 G_ID: 0.118 G_Rec: 0.437 D_GP: 0.028 D_real: 0.854 D_fake: 0.845 +(epoch: 338, iters: 6910, time: 0.063) G_GAN: -0.024 G_GAN_Feat: 0.812 G_ID: 0.102 G_Rec: 0.350 D_GP: 0.091 D_real: 0.736 D_fake: 1.024 +(epoch: 338, iters: 7310, time: 0.063) G_GAN: 0.208 G_GAN_Feat: 0.863 G_ID: 0.128 G_Rec: 0.411 D_GP: 0.030 D_real: 0.891 D_fake: 0.795 +(epoch: 338, iters: 7710, time: 0.063) G_GAN: -0.261 G_GAN_Feat: 0.752 G_ID: 0.101 G_Rec: 0.342 D_GP: 0.057 D_real: 0.498 D_fake: 1.261 +(epoch: 338, iters: 8110, time: 0.064) G_GAN: 0.142 G_GAN_Feat: 0.941 G_ID: 0.139 G_Rec: 0.394 D_GP: 0.081 D_real: 0.613 D_fake: 0.861 +(epoch: 338, iters: 8510, time: 0.063) G_GAN: 0.009 G_GAN_Feat: 0.855 G_ID: 0.125 G_Rec: 0.340 D_GP: 0.108 D_real: 0.590 D_fake: 0.991 +(epoch: 339, iters: 302, time: 0.063) G_GAN: 0.434 G_GAN_Feat: 1.076 G_ID: 0.132 G_Rec: 0.487 D_GP: 0.164 D_real: 0.507 D_fake: 0.578 +(epoch: 339, iters: 702, time: 0.063) G_GAN: 0.326 G_GAN_Feat: 0.679 G_ID: 0.094 G_Rec: 0.323 D_GP: 0.023 D_real: 1.241 D_fake: 0.675 +(epoch: 339, iters: 1102, time: 0.064) G_GAN: 0.639 G_GAN_Feat: 1.058 G_ID: 0.137 G_Rec: 0.482 D_GP: 0.099 D_real: 0.946 D_fake: 0.370 +(epoch: 339, iters: 1502, time: 0.064) G_GAN: 0.243 G_GAN_Feat: 0.805 G_ID: 0.122 G_Rec: 0.341 D_GP: 0.031 D_real: 1.074 D_fake: 0.758 +(epoch: 339, iters: 1902, time: 0.063) G_GAN: 0.421 G_GAN_Feat: 1.063 G_ID: 0.128 G_Rec: 0.469 D_GP: 0.040 D_real: 0.611 D_fake: 0.580 +(epoch: 339, iters: 2302, time: 0.063) G_GAN: 0.096 G_GAN_Feat: 0.728 G_ID: 0.117 G_Rec: 0.320 D_GP: 0.024 D_real: 0.984 D_fake: 0.905 +(epoch: 339, iters: 2702, time: 0.064) G_GAN: 0.340 G_GAN_Feat: 1.002 G_ID: 0.120 G_Rec: 0.449 D_GP: 0.050 D_real: 0.874 D_fake: 0.665 +(epoch: 339, iters: 3102, time: 0.063) G_GAN: 0.035 G_GAN_Feat: 0.688 G_ID: 0.110 G_Rec: 0.284 D_GP: 0.031 D_real: 0.833 D_fake: 0.966 +(epoch: 339, iters: 3502, time: 0.064) G_GAN: 0.206 G_GAN_Feat: 0.946 G_ID: 0.157 G_Rec: 0.407 D_GP: 0.042 D_real: 0.709 D_fake: 0.798 +(epoch: 339, iters: 3902, time: 0.063) G_GAN: 0.160 G_GAN_Feat: 0.742 G_ID: 0.097 G_Rec: 0.326 D_GP: 0.040 D_real: 1.037 D_fake: 0.840 +(epoch: 339, iters: 4302, time: 0.064) G_GAN: 0.421 G_GAN_Feat: 1.027 G_ID: 0.130 G_Rec: 0.476 D_GP: 0.045 D_real: 0.762 D_fake: 0.595 +(epoch: 339, iters: 4702, time: 0.063) G_GAN: 0.116 G_GAN_Feat: 0.839 G_ID: 0.112 G_Rec: 0.342 D_GP: 0.102 D_real: 0.642 D_fake: 0.885 +(epoch: 339, iters: 5102, time: 0.064) G_GAN: 0.576 G_GAN_Feat: 0.967 G_ID: 0.130 G_Rec: 0.426 D_GP: 0.041 D_real: 1.199 D_fake: 0.445 +(epoch: 339, iters: 5502, time: 0.064) G_GAN: 0.222 G_GAN_Feat: 0.847 G_ID: 0.101 G_Rec: 0.354 D_GP: 0.032 D_real: 0.719 D_fake: 0.785 +(epoch: 339, iters: 5902, time: 0.064) G_GAN: 0.538 G_GAN_Feat: 1.017 G_ID: 0.122 G_Rec: 0.423 D_GP: 0.049 D_real: 0.785 D_fake: 0.465 +(epoch: 339, iters: 6302, time: 0.064) G_GAN: 0.356 G_GAN_Feat: 0.709 G_ID: 0.110 G_Rec: 0.315 D_GP: 0.027 D_real: 1.284 D_fake: 0.645 +(epoch: 339, iters: 6702, time: 0.064) G_GAN: 0.253 G_GAN_Feat: 0.884 G_ID: 0.134 G_Rec: 0.422 D_GP: 0.032 D_real: 0.907 D_fake: 0.748 +(epoch: 339, iters: 7102, time: 0.064) G_GAN: -0.023 G_GAN_Feat: 0.748 G_ID: 0.117 G_Rec: 0.327 D_GP: 0.035 D_real: 0.835 D_fake: 1.023 +(epoch: 339, iters: 7502, time: 0.064) G_GAN: 0.182 G_GAN_Feat: 0.970 G_ID: 0.160 G_Rec: 0.433 D_GP: 0.043 D_real: 0.586 D_fake: 0.819 +(epoch: 339, iters: 7902, time: 0.064) G_GAN: 0.320 G_GAN_Feat: 0.745 G_ID: 0.117 G_Rec: 0.308 D_GP: 0.082 D_real: 0.995 D_fake: 0.683 +(epoch: 339, iters: 8302, time: 0.063) G_GAN: 0.724 G_GAN_Feat: 1.103 G_ID: 0.137 G_Rec: 0.486 D_GP: 0.108 D_real: 0.704 D_fake: 0.344 +(epoch: 339, iters: 8702, time: 0.063) G_GAN: 0.311 G_GAN_Feat: 0.774 G_ID: 0.115 G_Rec: 0.303 D_GP: 0.034 D_real: 1.134 D_fake: 0.689 +(epoch: 340, iters: 494, time: 0.064) G_GAN: 0.306 G_GAN_Feat: 1.015 G_ID: 0.127 G_Rec: 0.425 D_GP: 0.080 D_real: 0.651 D_fake: 0.703 +(epoch: 340, iters: 894, time: 0.063) G_GAN: 0.278 G_GAN_Feat: 0.783 G_ID: 0.105 G_Rec: 0.361 D_GP: 0.026 D_real: 1.100 D_fake: 0.726 +(epoch: 340, iters: 1294, time: 0.064) G_GAN: 0.620 G_GAN_Feat: 1.071 G_ID: 0.138 G_Rec: 0.456 D_GP: 0.037 D_real: 1.223 D_fake: 0.420 +(epoch: 340, iters: 1694, time: 0.064) G_GAN: 0.221 G_GAN_Feat: 0.816 G_ID: 0.143 G_Rec: 0.325 D_GP: 0.062 D_real: 0.904 D_fake: 0.779 +(epoch: 340, iters: 2094, time: 0.064) G_GAN: 0.800 G_GAN_Feat: 1.232 G_ID: 0.152 G_Rec: 0.484 D_GP: 0.238 D_real: 0.404 D_fake: 0.252 +(epoch: 340, iters: 2494, time: 0.064) G_GAN: -0.021 G_GAN_Feat: 0.752 G_ID: 0.109 G_Rec: 0.335 D_GP: 0.038 D_real: 0.832 D_fake: 1.021 +(epoch: 340, iters: 2894, time: 0.063) G_GAN: 0.449 G_GAN_Feat: 0.977 G_ID: 0.139 G_Rec: 0.463 D_GP: 0.046 D_real: 0.863 D_fake: 0.562 +(epoch: 340, iters: 3294, time: 0.063) G_GAN: 0.220 G_GAN_Feat: 0.729 G_ID: 0.098 G_Rec: 0.306 D_GP: 0.033 D_real: 1.077 D_fake: 0.781 +(epoch: 340, iters: 3694, time: 0.064) G_GAN: 0.517 G_GAN_Feat: 1.164 G_ID: 0.142 G_Rec: 0.460 D_GP: 0.864 D_real: 0.274 D_fake: 0.506 +(epoch: 340, iters: 4094, time: 0.063) G_GAN: 0.254 G_GAN_Feat: 0.700 G_ID: 0.092 G_Rec: 0.342 D_GP: 0.032 D_real: 1.202 D_fake: 0.747 +(epoch: 340, iters: 4494, time: 0.063) G_GAN: 0.605 G_GAN_Feat: 1.060 G_ID: 0.122 G_Rec: 0.429 D_GP: 0.063 D_real: 0.794 D_fake: 0.440 +(epoch: 340, iters: 4894, time: 0.063) G_GAN: 0.264 G_GAN_Feat: 0.703 G_ID: 0.116 G_Rec: 0.315 D_GP: 0.027 D_real: 1.192 D_fake: 0.827 +(epoch: 340, iters: 5294, time: 0.064) G_GAN: 0.356 G_GAN_Feat: 0.851 G_ID: 0.121 G_Rec: 0.426 D_GP: 0.026 D_real: 1.058 D_fake: 0.648 +(epoch: 340, iters: 5694, time: 0.063) G_GAN: 0.220 G_GAN_Feat: 0.613 G_ID: 0.101 G_Rec: 0.309 D_GP: 0.026 D_real: 1.194 D_fake: 0.780 +(epoch: 340, iters: 6094, time: 0.063) G_GAN: 0.456 G_GAN_Feat: 0.936 G_ID: 0.116 G_Rec: 0.472 D_GP: 0.040 D_real: 1.071 D_fake: 0.554 +(epoch: 340, iters: 6494, time: 0.063) G_GAN: 0.134 G_GAN_Feat: 0.694 G_ID: 0.103 G_Rec: 0.331 D_GP: 0.046 D_real: 0.992 D_fake: 0.867 +(epoch: 340, iters: 6894, time: 0.064) G_GAN: 0.269 G_GAN_Feat: 0.791 G_ID: 0.114 G_Rec: 0.375 D_GP: 0.038 D_real: 0.977 D_fake: 0.735 +(epoch: 340, iters: 7294, time: 0.064) G_GAN: 0.095 G_GAN_Feat: 0.737 G_ID: 0.097 G_Rec: 0.370 D_GP: 0.061 D_real: 0.834 D_fake: 0.912 +(epoch: 340, iters: 7694, time: 0.063) G_GAN: 0.310 G_GAN_Feat: 0.911 G_ID: 0.138 G_Rec: 0.429 D_GP: 0.047 D_real: 0.880 D_fake: 0.698 +(epoch: 340, iters: 8094, time: 0.063) G_GAN: 0.001 G_GAN_Feat: 0.720 G_ID: 0.101 G_Rec: 0.327 D_GP: 0.068 D_real: 0.808 D_fake: 0.999 +(epoch: 340, iters: 8494, time: 0.064) G_GAN: 0.414 G_GAN_Feat: 0.880 G_ID: 0.111 G_Rec: 0.417 D_GP: 0.037 D_real: 0.995 D_fake: 0.591 +(epoch: 341, iters: 286, time: 0.064) G_GAN: 0.116 G_GAN_Feat: 0.687 G_ID: 0.099 G_Rec: 0.306 D_GP: 0.038 D_real: 1.007 D_fake: 0.884 +(epoch: 341, iters: 686, time: 0.063) G_GAN: 0.112 G_GAN_Feat: 0.931 G_ID: 0.128 G_Rec: 0.450 D_GP: 0.049 D_real: 0.752 D_fake: 0.893 +(epoch: 341, iters: 1086, time: 0.063) G_GAN: 0.079 G_GAN_Feat: 0.660 G_ID: 0.125 G_Rec: 0.280 D_GP: 0.031 D_real: 0.994 D_fake: 0.921 +(epoch: 341, iters: 1486, time: 0.064) G_GAN: 0.594 G_GAN_Feat: 0.969 G_ID: 0.132 G_Rec: 0.425 D_GP: 0.066 D_real: 1.092 D_fake: 0.446 +(epoch: 341, iters: 1886, time: 0.064) G_GAN: 0.063 G_GAN_Feat: 0.754 G_ID: 0.103 G_Rec: 0.321 D_GP: 0.042 D_real: 0.883 D_fake: 0.937 +(epoch: 341, iters: 2286, time: 0.063) G_GAN: 0.118 G_GAN_Feat: 0.906 G_ID: 0.159 G_Rec: 0.385 D_GP: 0.048 D_real: 0.694 D_fake: 0.882 +(epoch: 341, iters: 2686, time: 0.064) G_GAN: 0.036 G_GAN_Feat: 0.773 G_ID: 0.099 G_Rec: 0.317 D_GP: 0.030 D_real: 0.859 D_fake: 0.964 +(epoch: 341, iters: 3086, time: 0.064) G_GAN: 0.333 G_GAN_Feat: 1.017 G_ID: 0.115 G_Rec: 0.425 D_GP: 0.059 D_real: 0.746 D_fake: 0.669 +(epoch: 341, iters: 3486, time: 0.063) G_GAN: 0.059 G_GAN_Feat: 0.768 G_ID: 0.107 G_Rec: 0.333 D_GP: 0.038 D_real: 0.877 D_fake: 0.942 +(epoch: 341, iters: 3886, time: 0.063) G_GAN: 0.293 G_GAN_Feat: 0.893 G_ID: 0.121 G_Rec: 0.378 D_GP: 0.041 D_real: 0.878 D_fake: 0.708 +(epoch: 341, iters: 4286, time: 0.064) G_GAN: 0.012 G_GAN_Feat: 0.795 G_ID: 0.093 G_Rec: 0.322 D_GP: 0.042 D_real: 0.851 D_fake: 0.989 +(epoch: 341, iters: 4686, time: 0.064) G_GAN: 0.483 G_GAN_Feat: 1.001 G_ID: 0.132 G_Rec: 0.492 D_GP: 0.058 D_real: 0.987 D_fake: 0.522 +(epoch: 341, iters: 5086, time: 0.063) G_GAN: 0.151 G_GAN_Feat: 0.746 G_ID: 0.121 G_Rec: 0.310 D_GP: 0.030 D_real: 1.021 D_fake: 0.849 +(epoch: 341, iters: 5486, time: 0.063) G_GAN: 0.286 G_GAN_Feat: 0.955 G_ID: 0.126 G_Rec: 0.435 D_GP: 0.033 D_real: 0.899 D_fake: 0.716 +(epoch: 341, iters: 5886, time: 0.064) G_GAN: 0.048 G_GAN_Feat: 0.905 G_ID: 0.112 G_Rec: 0.337 D_GP: 0.051 D_real: 0.483 D_fake: 0.952 +(epoch: 341, iters: 6286, time: 0.064) G_GAN: 0.507 G_GAN_Feat: 0.961 G_ID: 0.132 G_Rec: 0.402 D_GP: 0.058 D_real: 0.917 D_fake: 0.506 +(epoch: 341, iters: 6686, time: 0.064) G_GAN: 0.004 G_GAN_Feat: 0.749 G_ID: 0.125 G_Rec: 0.321 D_GP: 0.026 D_real: 0.885 D_fake: 0.996 +(epoch: 341, iters: 7086, time: 0.064) G_GAN: 0.376 G_GAN_Feat: 1.061 G_ID: 0.128 G_Rec: 0.417 D_GP: 0.033 D_real: 0.874 D_fake: 0.625 +(epoch: 341, iters: 7486, time: 0.063) G_GAN: 0.201 G_GAN_Feat: 0.861 G_ID: 0.111 G_Rec: 0.319 D_GP: 0.104 D_real: 0.550 D_fake: 0.799 +(epoch: 341, iters: 7886, time: 0.064) G_GAN: 0.340 G_GAN_Feat: 0.902 G_ID: 0.146 G_Rec: 0.447 D_GP: 0.029 D_real: 1.069 D_fake: 0.665 +(epoch: 341, iters: 8286, time: 0.064) G_GAN: 0.239 G_GAN_Feat: 0.730 G_ID: 0.089 G_Rec: 0.314 D_GP: 0.028 D_real: 1.144 D_fake: 0.761 +(epoch: 341, iters: 8686, time: 0.064) G_GAN: 0.320 G_GAN_Feat: 0.993 G_ID: 0.146 G_Rec: 0.453 D_GP: 0.033 D_real: 0.844 D_fake: 0.681 +(epoch: 342, iters: 478, time: 0.064) G_GAN: 0.384 G_GAN_Feat: 0.777 G_ID: 0.119 G_Rec: 0.310 D_GP: 0.038 D_real: 1.159 D_fake: 0.616 +(epoch: 342, iters: 878, time: 0.064) G_GAN: 0.452 G_GAN_Feat: 0.988 G_ID: 0.132 G_Rec: 0.440 D_GP: 0.063 D_real: 0.970 D_fake: 0.554 +(epoch: 342, iters: 1278, time: 0.064) G_GAN: 0.121 G_GAN_Feat: 0.845 G_ID: 0.112 G_Rec: 0.320 D_GP: 0.060 D_real: 0.614 D_fake: 0.880 +(epoch: 342, iters: 1678, time: 0.064) G_GAN: 0.255 G_GAN_Feat: 0.965 G_ID: 0.157 G_Rec: 0.431 D_GP: 0.029 D_real: 0.774 D_fake: 0.745 +(epoch: 342, iters: 2078, time: 0.063) G_GAN: 0.370 G_GAN_Feat: 0.789 G_ID: 0.100 G_Rec: 0.313 D_GP: 0.043 D_real: 1.070 D_fake: 0.633 +(epoch: 342, iters: 2478, time: 0.064) G_GAN: 0.239 G_GAN_Feat: 1.074 G_ID: 0.142 G_Rec: 0.462 D_GP: 0.079 D_real: 0.513 D_fake: 0.763 +(epoch: 342, iters: 2878, time: 0.064) G_GAN: 0.208 G_GAN_Feat: 0.788 G_ID: 0.110 G_Rec: 0.358 D_GP: 0.030 D_real: 1.062 D_fake: 0.792 +(epoch: 342, iters: 3278, time: 0.064) G_GAN: 0.705 G_GAN_Feat: 0.880 G_ID: 0.129 G_Rec: 0.447 D_GP: 0.031 D_real: 1.268 D_fake: 0.397 +(epoch: 342, iters: 3678, time: 0.063) G_GAN: 0.106 G_GAN_Feat: 0.689 G_ID: 0.099 G_Rec: 0.314 D_GP: 0.030 D_real: 0.979 D_fake: 0.894 +(epoch: 342, iters: 4078, time: 0.064) G_GAN: 0.472 G_GAN_Feat: 0.988 G_ID: 0.130 G_Rec: 0.478 D_GP: 0.037 D_real: 1.095 D_fake: 0.534 +(epoch: 342, iters: 4478, time: 0.063) G_GAN: 0.314 G_GAN_Feat: 0.789 G_ID: 0.117 G_Rec: 0.356 D_GP: 0.053 D_real: 0.972 D_fake: 0.688 +(epoch: 342, iters: 4878, time: 0.064) G_GAN: 0.257 G_GAN_Feat: 0.960 G_ID: 0.146 G_Rec: 0.449 D_GP: 0.218 D_real: 0.874 D_fake: 0.746 +(epoch: 342, iters: 5278, time: 0.064) G_GAN: 0.002 G_GAN_Feat: 0.728 G_ID: 0.096 G_Rec: 0.289 D_GP: 0.048 D_real: 0.849 D_fake: 0.998 +(epoch: 342, iters: 5678, time: 0.063) G_GAN: 0.413 G_GAN_Feat: 1.042 G_ID: 0.134 G_Rec: 0.446 D_GP: 0.057 D_real: 0.943 D_fake: 0.604 +(epoch: 342, iters: 6078, time: 0.063) G_GAN: 0.014 G_GAN_Feat: 0.830 G_ID: 0.096 G_Rec: 0.323 D_GP: 0.033 D_real: 0.770 D_fake: 0.987 +(epoch: 342, iters: 6478, time: 0.063) G_GAN: 0.291 G_GAN_Feat: 1.018 G_ID: 0.124 G_Rec: 0.438 D_GP: 0.053 D_real: 0.724 D_fake: 0.710 +(epoch: 342, iters: 6878, time: 0.064) G_GAN: 0.336 G_GAN_Feat: 0.726 G_ID: 0.112 G_Rec: 0.295 D_GP: 0.030 D_real: 1.149 D_fake: 0.664 +(epoch: 342, iters: 7278, time: 0.063) G_GAN: 0.410 G_GAN_Feat: 0.947 G_ID: 0.152 G_Rec: 0.396 D_GP: 0.035 D_real: 1.167 D_fake: 0.593 +(epoch: 342, iters: 7678, time: 0.063) G_GAN: 0.137 G_GAN_Feat: 1.002 G_ID: 0.103 G_Rec: 0.405 D_GP: 0.245 D_real: 0.440 D_fake: 0.865 +(epoch: 342, iters: 8078, time: 0.063) G_GAN: 0.448 G_GAN_Feat: 1.180 G_ID: 0.145 G_Rec: 0.448 D_GP: 0.067 D_real: 0.460 D_fake: 0.554 +(epoch: 342, iters: 8478, time: 0.064) G_GAN: 0.290 G_GAN_Feat: 0.967 G_ID: 0.125 G_Rec: 0.334 D_GP: 0.080 D_real: 0.643 D_fake: 0.714 +(epoch: 343, iters: 270, time: 0.063) G_GAN: 0.520 G_GAN_Feat: 1.167 G_ID: 0.150 G_Rec: 0.447 D_GP: 0.032 D_real: 0.905 D_fake: 0.485 +(epoch: 343, iters: 670, time: 0.063) G_GAN: 0.062 G_GAN_Feat: 0.700 G_ID: 0.110 G_Rec: 0.347 D_GP: 0.027 D_real: 0.964 D_fake: 0.938 +(epoch: 343, iters: 1070, time: 0.064) G_GAN: 0.430 G_GAN_Feat: 0.931 G_ID: 0.157 G_Rec: 0.435 D_GP: 0.036 D_real: 0.926 D_fake: 0.590 +(epoch: 343, iters: 1470, time: 0.064) G_GAN: -0.107 G_GAN_Feat: 0.818 G_ID: 0.118 G_Rec: 0.361 D_GP: 0.121 D_real: 0.617 D_fake: 1.107 +(epoch: 343, iters: 1870, time: 0.063) G_GAN: 0.242 G_GAN_Feat: 0.864 G_ID: 0.132 G_Rec: 0.372 D_GP: 0.043 D_real: 0.989 D_fake: 0.760 +(epoch: 343, iters: 2270, time: 0.063) G_GAN: 0.386 G_GAN_Feat: 0.692 G_ID: 0.106 G_Rec: 0.302 D_GP: 0.036 D_real: 1.285 D_fake: 0.621 +(epoch: 343, iters: 2670, time: 0.063) G_GAN: 0.560 G_GAN_Feat: 0.997 G_ID: 0.151 G_Rec: 0.486 D_GP: 0.070 D_real: 1.069 D_fake: 0.453 +(epoch: 343, iters: 3070, time: 0.064) G_GAN: 0.214 G_GAN_Feat: 0.721 G_ID: 0.111 G_Rec: 0.287 D_GP: 0.042 D_real: 0.967 D_fake: 0.786 +(epoch: 343, iters: 3470, time: 0.064) G_GAN: 0.796 G_GAN_Feat: 1.043 G_ID: 0.141 G_Rec: 0.429 D_GP: 0.094 D_real: 0.997 D_fake: 0.293 +(epoch: 343, iters: 3870, time: 0.063) G_GAN: -0.083 G_GAN_Feat: 0.800 G_ID: 0.110 G_Rec: 0.312 D_GP: 0.044 D_real: 0.786 D_fake: 1.084 +(epoch: 343, iters: 4270, time: 0.064) G_GAN: 0.692 G_GAN_Feat: 1.064 G_ID: 0.125 G_Rec: 0.458 D_GP: 0.039 D_real: 0.959 D_fake: 0.343 +(epoch: 343, iters: 4670, time: 0.064) G_GAN: 0.266 G_GAN_Feat: 0.943 G_ID: 0.114 G_Rec: 0.354 D_GP: 0.981 D_real: 0.401 D_fake: 0.739 +(epoch: 343, iters: 5070, time: 0.064) G_GAN: 0.221 G_GAN_Feat: 1.181 G_ID: 0.133 G_Rec: 0.500 D_GP: 0.155 D_real: 0.439 D_fake: 0.786 +(epoch: 343, iters: 5470, time: 0.064) G_GAN: 0.263 G_GAN_Feat: 0.724 G_ID: 0.115 G_Rec: 0.307 D_GP: 0.033 D_real: 1.126 D_fake: 0.738 +(epoch: 343, iters: 5870, time: 0.064) G_GAN: 0.315 G_GAN_Feat: 1.066 G_ID: 0.127 G_Rec: 0.394 D_GP: 0.057 D_real: 0.386 D_fake: 0.723 +(epoch: 343, iters: 6270, time: 0.064) G_GAN: 0.517 G_GAN_Feat: 0.857 G_ID: 0.103 G_Rec: 0.349 D_GP: 0.035 D_real: 1.306 D_fake: 0.493 +(epoch: 343, iters: 6670, time: 0.063) G_GAN: 0.105 G_GAN_Feat: 1.221 G_ID: 0.147 G_Rec: 0.490 D_GP: 0.172 D_real: 0.089 D_fake: 0.901 +(epoch: 343, iters: 7070, time: 0.063) G_GAN: 0.022 G_GAN_Feat: 0.750 G_ID: 0.122 G_Rec: 0.289 D_GP: 0.043 D_real: 0.745 D_fake: 0.978 +(epoch: 343, iters: 7470, time: 0.063) G_GAN: 0.286 G_GAN_Feat: 1.123 G_ID: 0.122 G_Rec: 0.434 D_GP: 0.093 D_real: 0.751 D_fake: 0.722 +(epoch: 343, iters: 7870, time: 0.064) G_GAN: 0.586 G_GAN_Feat: 0.821 G_ID: 0.099 G_Rec: 0.334 D_GP: 0.046 D_real: 1.389 D_fake: 0.468 +(epoch: 343, iters: 8270, time: 0.063) G_GAN: 0.401 G_GAN_Feat: 0.990 G_ID: 0.122 G_Rec: 0.428 D_GP: 0.029 D_real: 0.990 D_fake: 0.599 +(epoch: 343, iters: 8670, time: 0.063) G_GAN: 0.786 G_GAN_Feat: 0.828 G_ID: 0.113 G_Rec: 0.330 D_GP: 0.037 D_real: 1.386 D_fake: 0.275 +(epoch: 344, iters: 462, time: 0.063) G_GAN: 0.416 G_GAN_Feat: 0.961 G_ID: 0.136 G_Rec: 0.436 D_GP: 0.032 D_real: 1.005 D_fake: 0.587 +(epoch: 344, iters: 862, time: 0.064) G_GAN: 0.416 G_GAN_Feat: 0.836 G_ID: 0.097 G_Rec: 0.329 D_GP: 0.034 D_real: 1.131 D_fake: 0.585 +(epoch: 344, iters: 1262, time: 0.063) G_GAN: 0.545 G_GAN_Feat: 0.873 G_ID: 0.132 G_Rec: 0.378 D_GP: 0.032 D_real: 1.331 D_fake: 0.459 +(epoch: 344, iters: 1662, time: 0.063) G_GAN: 0.348 G_GAN_Feat: 0.764 G_ID: 0.123 G_Rec: 0.315 D_GP: 0.045 D_real: 1.153 D_fake: 0.658 +(epoch: 344, iters: 2062, time: 0.064) G_GAN: 0.637 G_GAN_Feat: 0.885 G_ID: 0.146 G_Rec: 0.453 D_GP: 0.033 D_real: 1.288 D_fake: 0.374 +(epoch: 344, iters: 2462, time: 0.064) G_GAN: 0.168 G_GAN_Feat: 0.732 G_ID: 0.113 G_Rec: 0.318 D_GP: 0.044 D_real: 0.979 D_fake: 0.832 +(epoch: 344, iters: 2862, time: 0.064) G_GAN: 0.366 G_GAN_Feat: 0.980 G_ID: 0.149 G_Rec: 0.445 D_GP: 0.033 D_real: 1.053 D_fake: 0.641 +(epoch: 344, iters: 3262, time: 0.063) G_GAN: 0.201 G_GAN_Feat: 0.681 G_ID: 0.102 G_Rec: 0.305 D_GP: 0.038 D_real: 1.129 D_fake: 0.799 +(epoch: 344, iters: 3662, time: 0.063) G_GAN: 0.500 G_GAN_Feat: 0.838 G_ID: 0.130 G_Rec: 0.379 D_GP: 0.039 D_real: 1.179 D_fake: 0.510 +(epoch: 344, iters: 4062, time: 0.064) G_GAN: 0.102 G_GAN_Feat: 0.687 G_ID: 0.108 G_Rec: 0.302 D_GP: 0.038 D_real: 0.934 D_fake: 0.898 +(epoch: 344, iters: 4462, time: 0.063) G_GAN: 0.623 G_GAN_Feat: 0.937 G_ID: 0.118 G_Rec: 0.413 D_GP: 0.032 D_real: 1.313 D_fake: 0.390 +(epoch: 344, iters: 4862, time: 0.063) G_GAN: 0.238 G_GAN_Feat: 0.706 G_ID: 0.111 G_Rec: 0.308 D_GP: 0.028 D_real: 1.040 D_fake: 0.762 +(epoch: 344, iters: 5262, time: 0.063) G_GAN: 0.667 G_GAN_Feat: 0.998 G_ID: 0.130 G_Rec: 0.440 D_GP: 0.034 D_real: 1.393 D_fake: 0.354 +(epoch: 344, iters: 5662, time: 0.064) G_GAN: -0.011 G_GAN_Feat: 0.892 G_ID: 0.100 G_Rec: 0.326 D_GP: 0.235 D_real: 0.322 D_fake: 1.012 +(epoch: 344, iters: 6062, time: 0.063) G_GAN: 0.635 G_GAN_Feat: 1.193 G_ID: 0.127 G_Rec: 0.447 D_GP: 0.050 D_real: 0.682 D_fake: 0.446 +(epoch: 344, iters: 6462, time: 0.063) G_GAN: 0.184 G_GAN_Feat: 0.736 G_ID: 0.093 G_Rec: 0.269 D_GP: 0.036 D_real: 0.927 D_fake: 0.816 +(epoch: 344, iters: 6862, time: 0.063) G_GAN: 0.786 G_GAN_Feat: 1.003 G_ID: 0.128 G_Rec: 0.411 D_GP: 0.033 D_real: 1.091 D_fake: 0.247 +(epoch: 344, iters: 7262, time: 0.064) G_GAN: 0.099 G_GAN_Feat: 0.854 G_ID: 0.126 G_Rec: 0.316 D_GP: 0.035 D_real: 0.753 D_fake: 0.901 +(epoch: 344, iters: 7662, time: 0.063) G_GAN: 0.648 G_GAN_Feat: 1.212 G_ID: 0.148 G_Rec: 0.488 D_GP: 0.044 D_real: 0.626 D_fake: 0.366 +(epoch: 344, iters: 8062, time: 0.063) G_GAN: 0.194 G_GAN_Feat: 0.735 G_ID: 0.105 G_Rec: 0.327 D_GP: 0.026 D_real: 1.125 D_fake: 0.809 +(epoch: 344, iters: 8462, time: 0.063) G_GAN: 0.723 G_GAN_Feat: 0.922 G_ID: 0.122 G_Rec: 0.434 D_GP: 0.031 D_real: 1.366 D_fake: 0.382 +(epoch: 345, iters: 254, time: 0.064) G_GAN: 0.213 G_GAN_Feat: 0.998 G_ID: 0.103 G_Rec: 0.341 D_GP: 0.153 D_real: 0.435 D_fake: 0.794 +(epoch: 345, iters: 654, time: 0.063) G_GAN: 0.741 G_GAN_Feat: 0.936 G_ID: 0.115 G_Rec: 0.384 D_GP: 0.033 D_real: 1.320 D_fake: 0.281 +(epoch: 345, iters: 1054, time: 0.063) G_GAN: 0.170 G_GAN_Feat: 0.817 G_ID: 0.105 G_Rec: 0.342 D_GP: 0.036 D_real: 0.928 D_fake: 0.832 +(epoch: 345, iters: 1454, time: 0.063) G_GAN: 0.591 G_GAN_Feat: 1.013 G_ID: 0.114 G_Rec: 0.446 D_GP: 0.036 D_real: 1.055 D_fake: 0.418 +(epoch: 345, iters: 1854, time: 0.064) G_GAN: 0.467 G_GAN_Feat: 1.051 G_ID: 0.101 G_Rec: 0.385 D_GP: 0.389 D_real: 0.475 D_fake: 0.625 +(epoch: 345, iters: 2254, time: 0.063) G_GAN: 0.382 G_GAN_Feat: 0.923 G_ID: 0.119 G_Rec: 0.429 D_GP: 0.028 D_real: 1.078 D_fake: 0.621 +(epoch: 345, iters: 2654, time: 0.063) G_GAN: 0.362 G_GAN_Feat: 0.843 G_ID: 0.097 G_Rec: 0.332 D_GP: 0.034 D_real: 1.169 D_fake: 0.655 +(epoch: 345, iters: 3054, time: 0.064) G_GAN: 0.506 G_GAN_Feat: 0.946 G_ID: 0.133 G_Rec: 0.417 D_GP: 0.035 D_real: 1.050 D_fake: 0.500 +(epoch: 345, iters: 3454, time: 0.064) G_GAN: 0.571 G_GAN_Feat: 0.910 G_ID: 0.106 G_Rec: 0.351 D_GP: 0.065 D_real: 0.935 D_fake: 0.529 +(epoch: 345, iters: 3854, time: 0.064) G_GAN: 0.547 G_GAN_Feat: 1.176 G_ID: 0.152 G_Rec: 0.470 D_GP: 0.153 D_real: 0.404 D_fake: 0.468 +(epoch: 345, iters: 4254, time: 0.064) G_GAN: 0.205 G_GAN_Feat: 0.858 G_ID: 0.103 G_Rec: 0.374 D_GP: 0.031 D_real: 0.671 D_fake: 0.796 +(epoch: 345, iters: 4654, time: 0.064) G_GAN: 0.426 G_GAN_Feat: 1.148 G_ID: 0.140 G_Rec: 0.509 D_GP: 0.205 D_real: 0.419 D_fake: 0.597 +(epoch: 345, iters: 5054, time: 0.064) G_GAN: 0.184 G_GAN_Feat: 0.824 G_ID: 0.094 G_Rec: 0.318 D_GP: 0.028 D_real: 1.014 D_fake: 0.816 +(epoch: 345, iters: 5454, time: 0.064) G_GAN: 0.826 G_GAN_Feat: 0.964 G_ID: 0.118 G_Rec: 0.469 D_GP: 0.036 D_real: 1.351 D_fake: 0.336 +(epoch: 345, iters: 5854, time: 0.064) G_GAN: 0.150 G_GAN_Feat: 0.661 G_ID: 0.086 G_Rec: 0.288 D_GP: 0.029 D_real: 1.027 D_fake: 0.850 +(epoch: 345, iters: 6254, time: 0.064) G_GAN: 0.276 G_GAN_Feat: 0.896 G_ID: 0.125 G_Rec: 0.400 D_GP: 0.031 D_real: 0.997 D_fake: 0.725 +(epoch: 345, iters: 6654, time: 0.064) G_GAN: 0.545 G_GAN_Feat: 0.924 G_ID: 0.097 G_Rec: 0.361 D_GP: 0.490 D_real: 0.963 D_fake: 0.472 +(epoch: 345, iters: 7054, time: 0.064) G_GAN: 0.314 G_GAN_Feat: 1.033 G_ID: 0.124 G_Rec: 0.469 D_GP: 0.093 D_real: 0.682 D_fake: 0.689 +(epoch: 345, iters: 7454, time: 0.064) G_GAN: 0.178 G_GAN_Feat: 0.764 G_ID: 0.100 G_Rec: 0.302 D_GP: 0.036 D_real: 0.936 D_fake: 0.822 +(epoch: 345, iters: 7854, time: 0.064) G_GAN: 0.275 G_GAN_Feat: 1.005 G_ID: 0.135 G_Rec: 0.443 D_GP: 0.041 D_real: 0.768 D_fake: 0.726 +(epoch: 345, iters: 8254, time: 0.064) G_GAN: 0.084 G_GAN_Feat: 0.835 G_ID: 0.101 G_Rec: 0.319 D_GP: 0.046 D_real: 0.783 D_fake: 0.917 +(epoch: 345, iters: 8654, time: 0.064) G_GAN: 0.497 G_GAN_Feat: 0.982 G_ID: 0.135 G_Rec: 0.444 D_GP: 0.033 D_real: 1.192 D_fake: 0.505 +(epoch: 346, iters: 446, time: 0.064) G_GAN: 0.531 G_GAN_Feat: 1.044 G_ID: 0.104 G_Rec: 0.356 D_GP: 0.286 D_real: 0.468 D_fake: 0.536 +(epoch: 346, iters: 846, time: 0.063) G_GAN: 0.501 G_GAN_Feat: 1.169 G_ID: 0.117 G_Rec: 0.465 D_GP: 0.037 D_real: 0.421 D_fake: 0.506 +(epoch: 346, iters: 1246, time: 0.064) G_GAN: 0.122 G_GAN_Feat: 0.772 G_ID: 0.117 G_Rec: 0.342 D_GP: 0.030 D_real: 0.969 D_fake: 0.878 +(epoch: 346, iters: 1646, time: 0.063) G_GAN: 0.395 G_GAN_Feat: 0.933 G_ID: 0.122 G_Rec: 0.452 D_GP: 0.032 D_real: 1.008 D_fake: 0.614 +(epoch: 346, iters: 2046, time: 0.063) G_GAN: -0.136 G_GAN_Feat: 0.649 G_ID: 0.115 G_Rec: 0.290 D_GP: 0.030 D_real: 0.815 D_fake: 1.136 +(epoch: 346, iters: 2446, time: 0.063) G_GAN: 0.569 G_GAN_Feat: 0.990 G_ID: 0.116 G_Rec: 0.483 D_GP: 0.035 D_real: 0.963 D_fake: 0.477 +(epoch: 346, iters: 2846, time: 0.064) G_GAN: 0.221 G_GAN_Feat: 0.758 G_ID: 0.104 G_Rec: 0.375 D_GP: 0.027 D_real: 1.022 D_fake: 0.779 +(epoch: 346, iters: 3246, time: 0.064) G_GAN: 0.438 G_GAN_Feat: 0.909 G_ID: 0.120 G_Rec: 0.430 D_GP: 0.031 D_real: 1.138 D_fake: 0.570 +(epoch: 346, iters: 3646, time: 0.063) G_GAN: -0.001 G_GAN_Feat: 0.734 G_ID: 0.147 G_Rec: 0.338 D_GP: 0.048 D_real: 0.836 D_fake: 1.001 +(epoch: 346, iters: 4046, time: 0.064) G_GAN: 0.431 G_GAN_Feat: 1.103 G_ID: 0.131 G_Rec: 0.480 D_GP: 0.231 D_real: 0.482 D_fake: 0.582 +(epoch: 346, iters: 4446, time: 0.064) G_GAN: 0.139 G_GAN_Feat: 0.744 G_ID: 0.120 G_Rec: 0.296 D_GP: 0.035 D_real: 1.025 D_fake: 0.870 +(epoch: 346, iters: 4846, time: 0.064) G_GAN: 0.424 G_GAN_Feat: 1.016 G_ID: 0.107 G_Rec: 0.426 D_GP: 0.042 D_real: 0.911 D_fake: 0.588 +(epoch: 346, iters: 5246, time: 0.064) G_GAN: 0.059 G_GAN_Feat: 0.886 G_ID: 0.119 G_Rec: 0.320 D_GP: 0.043 D_real: 0.555 D_fake: 0.941 +(epoch: 346, iters: 5646, time: 0.063) G_GAN: 0.287 G_GAN_Feat: 1.013 G_ID: 0.110 G_Rec: 0.406 D_GP: 0.044 D_real: 0.614 D_fake: 0.713 +(epoch: 346, iters: 6046, time: 0.064) G_GAN: 0.364 G_GAN_Feat: 0.822 G_ID: 0.101 G_Rec: 0.317 D_GP: 0.049 D_real: 1.076 D_fake: 0.645 +(epoch: 346, iters: 6446, time: 0.063) G_GAN: 0.664 G_GAN_Feat: 0.996 G_ID: 0.127 G_Rec: 0.460 D_GP: 0.031 D_real: 1.309 D_fake: 0.394 +(epoch: 346, iters: 6846, time: 0.063) G_GAN: 0.270 G_GAN_Feat: 0.732 G_ID: 0.106 G_Rec: 0.292 D_GP: 0.028 D_real: 1.103 D_fake: 0.730 +(epoch: 346, iters: 7246, time: 0.063) G_GAN: 0.538 G_GAN_Feat: 1.005 G_ID: 0.136 G_Rec: 0.453 D_GP: 0.037 D_real: 1.035 D_fake: 0.483 +(epoch: 346, iters: 7646, time: 0.064) G_GAN: 0.095 G_GAN_Feat: 0.793 G_ID: 0.122 G_Rec: 0.317 D_GP: 0.038 D_real: 0.967 D_fake: 0.906 +(epoch: 346, iters: 8046, time: 0.063) G_GAN: 0.738 G_GAN_Feat: 0.954 G_ID: 0.121 G_Rec: 0.445 D_GP: 0.032 D_real: 1.399 D_fake: 0.350 +(epoch: 346, iters: 8446, time: 0.063) G_GAN: 0.206 G_GAN_Feat: 0.782 G_ID: 0.116 G_Rec: 0.319 D_GP: 0.027 D_real: 1.058 D_fake: 0.794 +(epoch: 347, iters: 238, time: 0.063) G_GAN: 0.578 G_GAN_Feat: 1.049 G_ID: 0.111 G_Rec: 0.488 D_GP: 0.032 D_real: 1.070 D_fake: 0.427 +(epoch: 347, iters: 638, time: 0.064) G_GAN: 0.276 G_GAN_Feat: 0.787 G_ID: 0.113 G_Rec: 0.316 D_GP: 0.031 D_real: 1.086 D_fake: 0.724 +(epoch: 347, iters: 1038, time: 0.063) G_GAN: 0.453 G_GAN_Feat: 1.089 G_ID: 0.130 G_Rec: 0.420 D_GP: 0.052 D_real: 0.690 D_fake: 0.551 +(epoch: 347, iters: 1438, time: 0.063) G_GAN: 0.682 G_GAN_Feat: 0.799 G_ID: 0.091 G_Rec: 0.345 D_GP: 0.035 D_real: 1.443 D_fake: 0.486 +(epoch: 347, iters: 1838, time: 0.063) G_GAN: 0.334 G_GAN_Feat: 0.865 G_ID: 0.136 G_Rec: 0.420 D_GP: 0.031 D_real: 1.081 D_fake: 0.667 +(epoch: 347, iters: 2238, time: 0.064) G_GAN: 0.190 G_GAN_Feat: 0.696 G_ID: 0.108 G_Rec: 0.331 D_GP: 0.028 D_real: 1.086 D_fake: 0.810 +(epoch: 347, iters: 2638, time: 0.064) G_GAN: 0.411 G_GAN_Feat: 0.883 G_ID: 0.126 G_Rec: 0.400 D_GP: 0.035 D_real: 1.017 D_fake: 0.596 +(epoch: 347, iters: 3038, time: 0.063) G_GAN: -0.197 G_GAN_Feat: 0.726 G_ID: 0.122 G_Rec: 0.347 D_GP: 0.039 D_real: 0.632 D_fake: 1.197 +(epoch: 347, iters: 3438, time: 0.064) G_GAN: 0.425 G_GAN_Feat: 0.885 G_ID: 0.122 G_Rec: 0.418 D_GP: 0.034 D_real: 1.079 D_fake: 0.592 +(epoch: 347, iters: 3838, time: 0.064) G_GAN: -0.071 G_GAN_Feat: 0.782 G_ID: 0.111 G_Rec: 0.347 D_GP: 0.053 D_real: 0.609 D_fake: 1.071 +(epoch: 347, iters: 4238, time: 0.063) G_GAN: 0.473 G_GAN_Feat: 0.984 G_ID: 0.126 G_Rec: 0.491 D_GP: 0.040 D_real: 1.144 D_fake: 0.546 +(epoch: 347, iters: 4638, time: 0.063) G_GAN: 0.082 G_GAN_Feat: 0.667 G_ID: 0.107 G_Rec: 0.295 D_GP: 0.037 D_real: 0.958 D_fake: 0.918 +(epoch: 347, iters: 5038, time: 0.064) G_GAN: 0.474 G_GAN_Feat: 0.948 G_ID: 0.130 G_Rec: 0.403 D_GP: 0.056 D_real: 1.092 D_fake: 0.568 +(epoch: 347, iters: 5438, time: 0.063) G_GAN: 0.107 G_GAN_Feat: 0.698 G_ID: 0.115 G_Rec: 0.325 D_GP: 0.035 D_real: 0.960 D_fake: 0.893 +(epoch: 347, iters: 5838, time: 0.063) G_GAN: 0.189 G_GAN_Feat: 0.977 G_ID: 0.138 G_Rec: 0.466 D_GP: 0.082 D_real: 0.604 D_fake: 0.811 +(epoch: 347, iters: 6238, time: 0.063) G_GAN: 0.014 G_GAN_Feat: 0.834 G_ID: 0.137 G_Rec: 0.343 D_GP: 0.042 D_real: 1.090 D_fake: 0.986 +(epoch: 347, iters: 6638, time: 0.064) G_GAN: 0.623 G_GAN_Feat: 0.944 G_ID: 0.111 G_Rec: 0.435 D_GP: 0.072 D_real: 1.297 D_fake: 0.395 +(epoch: 347, iters: 7038, time: 0.063) G_GAN: 0.351 G_GAN_Feat: 0.804 G_ID: 0.101 G_Rec: 0.369 D_GP: 0.075 D_real: 1.011 D_fake: 0.657 +(epoch: 347, iters: 7438, time: 0.063) G_GAN: 0.546 G_GAN_Feat: 0.994 G_ID: 0.127 G_Rec: 0.454 D_GP: 0.036 D_real: 1.015 D_fake: 0.468 +(epoch: 347, iters: 7838, time: 0.063) G_GAN: 0.221 G_GAN_Feat: 0.813 G_ID: 0.101 G_Rec: 0.328 D_GP: 0.037 D_real: 0.941 D_fake: 0.781 +(epoch: 347, iters: 8238, time: 0.064) G_GAN: 0.628 G_GAN_Feat: 0.997 G_ID: 0.126 G_Rec: 0.450 D_GP: 0.033 D_real: 1.245 D_fake: 0.389 +(epoch: 347, iters: 8638, time: 0.063) G_GAN: 0.077 G_GAN_Feat: 0.878 G_ID: 0.114 G_Rec: 0.350 D_GP: 0.067 D_real: 0.480 D_fake: 0.923 +(epoch: 348, iters: 430, time: 0.063) G_GAN: 0.409 G_GAN_Feat: 0.967 G_ID: 0.113 G_Rec: 0.444 D_GP: 0.039 D_real: 0.905 D_fake: 0.591 +(epoch: 348, iters: 830, time: 0.063) G_GAN: 0.387 G_GAN_Feat: 1.042 G_ID: 0.114 G_Rec: 0.362 D_GP: 0.087 D_real: 0.246 D_fake: 0.659 +(epoch: 348, iters: 1230, time: 0.064) G_GAN: 0.622 G_GAN_Feat: 1.152 G_ID: 0.124 G_Rec: 0.474 D_GP: 0.089 D_real: 0.969 D_fake: 0.422 +(epoch: 348, iters: 1630, time: 0.063) G_GAN: 0.435 G_GAN_Feat: 0.747 G_ID: 0.101 G_Rec: 0.305 D_GP: 0.044 D_real: 1.238 D_fake: 0.575 +(epoch: 348, iters: 2030, time: 0.063) G_GAN: 0.451 G_GAN_Feat: 1.107 G_ID: 0.114 G_Rec: 0.471 D_GP: 0.046 D_real: 0.663 D_fake: 0.557 +(epoch: 348, iters: 2430, time: 0.063) G_GAN: 0.165 G_GAN_Feat: 0.773 G_ID: 0.109 G_Rec: 0.308 D_GP: 0.033 D_real: 0.949 D_fake: 0.836 +(epoch: 348, iters: 2830, time: 0.064) G_GAN: 0.652 G_GAN_Feat: 1.000 G_ID: 0.144 G_Rec: 0.423 D_GP: 0.034 D_real: 1.255 D_fake: 0.371 +(epoch: 348, iters: 3230, time: 0.063) G_GAN: 0.078 G_GAN_Feat: 0.665 G_ID: 0.120 G_Rec: 0.287 D_GP: 0.032 D_real: 1.023 D_fake: 0.922 +(epoch: 348, iters: 3630, time: 0.063) G_GAN: 0.294 G_GAN_Feat: 1.051 G_ID: 0.130 G_Rec: 0.474 D_GP: 0.053 D_real: 0.872 D_fake: 0.706 +(epoch: 348, iters: 4030, time: 0.063) G_GAN: 0.200 G_GAN_Feat: 0.742 G_ID: 0.117 G_Rec: 0.311 D_GP: 0.029 D_real: 0.897 D_fake: 0.800 +(epoch: 348, iters: 4430, time: 0.064) G_GAN: 0.682 G_GAN_Feat: 1.080 G_ID: 0.114 G_Rec: 0.477 D_GP: 0.052 D_real: 0.955 D_fake: 0.352 +(epoch: 348, iters: 4830, time: 0.064) G_GAN: 0.147 G_GAN_Feat: 0.796 G_ID: 0.103 G_Rec: 0.322 D_GP: 0.051 D_real: 1.133 D_fake: 0.853 +(epoch: 348, iters: 5230, time: 0.064) G_GAN: 0.386 G_GAN_Feat: 1.031 G_ID: 0.122 G_Rec: 0.434 D_GP: 0.034 D_real: 0.794 D_fake: 0.618 +(epoch: 348, iters: 5630, time: 0.064) G_GAN: 1.125 G_GAN_Feat: 1.159 G_ID: 0.119 G_Rec: 0.408 D_GP: 0.116 D_real: 1.515 D_fake: 0.335 +(epoch: 348, iters: 6030, time: 0.064) G_GAN: 0.015 G_GAN_Feat: 0.883 G_ID: 0.137 G_Rec: 0.438 D_GP: 0.028 D_real: 0.751 D_fake: 0.986 +(epoch: 348, iters: 6430, time: 0.064) G_GAN: -0.104 G_GAN_Feat: 0.712 G_ID: 0.109 G_Rec: 0.313 D_GP: 0.032 D_real: 0.757 D_fake: 1.104 +(epoch: 348, iters: 6830, time: 0.064) G_GAN: 0.318 G_GAN_Feat: 0.948 G_ID: 0.125 G_Rec: 0.457 D_GP: 0.041 D_real: 0.940 D_fake: 0.697 +(epoch: 348, iters: 7230, time: 0.064) G_GAN: 0.121 G_GAN_Feat: 0.723 G_ID: 0.112 G_Rec: 0.298 D_GP: 0.041 D_real: 0.983 D_fake: 0.879 +(epoch: 348, iters: 7630, time: 0.064) G_GAN: 1.004 G_GAN_Feat: 1.187 G_ID: 0.123 G_Rec: 0.544 D_GP: 0.085 D_real: 1.040 D_fake: 0.166 +(epoch: 348, iters: 8030, time: 0.063) G_GAN: 0.400 G_GAN_Feat: 0.778 G_ID: 0.118 G_Rec: 0.339 D_GP: 0.032 D_real: 1.206 D_fake: 0.612 +(epoch: 348, iters: 8430, time: 0.063) G_GAN: 0.656 G_GAN_Feat: 1.044 G_ID: 0.123 G_Rec: 0.434 D_GP: 0.054 D_real: 1.004 D_fake: 0.359 +(epoch: 349, iters: 222, time: 0.063) G_GAN: 0.096 G_GAN_Feat: 0.956 G_ID: 0.112 G_Rec: 0.367 D_GP: 0.050 D_real: 0.365 D_fake: 0.905 +(epoch: 349, iters: 622, time: 0.064) G_GAN: 0.354 G_GAN_Feat: 1.153 G_ID: 0.148 G_Rec: 0.462 D_GP: 0.079 D_real: 0.372 D_fake: 0.650 +(epoch: 349, iters: 1022, time: 0.063) G_GAN: 0.572 G_GAN_Feat: 0.951 G_ID: 0.095 G_Rec: 0.328 D_GP: 0.780 D_real: 0.589 D_fake: 0.461 +(epoch: 349, iters: 1422, time: 0.063) G_GAN: 0.590 G_GAN_Feat: 1.076 G_ID: 0.127 G_Rec: 0.441 D_GP: 0.071 D_real: 0.633 D_fake: 0.432 +(epoch: 349, iters: 1822, time: 0.063) G_GAN: 0.112 G_GAN_Feat: 0.844 G_ID: 0.100 G_Rec: 0.326 D_GP: 0.060 D_real: 0.744 D_fake: 0.889 +(epoch: 349, iters: 2222, time: 0.064) G_GAN: 0.800 G_GAN_Feat: 1.086 G_ID: 0.120 G_Rec: 0.399 D_GP: 0.042 D_real: 1.006 D_fake: 0.230 +(epoch: 349, iters: 2622, time: 0.063) G_GAN: 0.568 G_GAN_Feat: 1.078 G_ID: 0.115 G_Rec: 0.385 D_GP: 0.063 D_real: 1.012 D_fake: 0.615 +(epoch: 349, iters: 3022, time: 0.063) G_GAN: 0.435 G_GAN_Feat: 1.076 G_ID: 0.136 G_Rec: 0.431 D_GP: 0.040 D_real: 0.577 D_fake: 0.568 +(epoch: 349, iters: 3422, time: 0.064) G_GAN: 0.391 G_GAN_Feat: 0.975 G_ID: 0.103 G_Rec: 0.337 D_GP: 0.030 D_real: 0.785 D_fake: 0.609 +(epoch: 349, iters: 3822, time: 0.064) G_GAN: 0.916 G_GAN_Feat: 1.068 G_ID: 0.130 G_Rec: 0.415 D_GP: 0.037 D_real: 1.180 D_fake: 0.166 +(epoch: 349, iters: 4222, time: 0.063) G_GAN: 0.097 G_GAN_Feat: 1.066 G_ID: 0.117 G_Rec: 0.364 D_GP: 0.325 D_real: 0.259 D_fake: 0.908 +(epoch: 349, iters: 4622, time: 0.063) G_GAN: 0.694 G_GAN_Feat: 1.063 G_ID: 0.130 G_Rec: 0.599 D_GP: 0.037 D_real: 1.199 D_fake: 0.326 +(epoch: 349, iters: 5022, time: 0.064) G_GAN: -0.028 G_GAN_Feat: 0.792 G_ID: 0.095 G_Rec: 0.303 D_GP: 0.061 D_real: 0.716 D_fake: 1.029 +(epoch: 349, iters: 5422, time: 0.064) G_GAN: 0.589 G_GAN_Feat: 0.971 G_ID: 0.125 G_Rec: 0.477 D_GP: 0.041 D_real: 1.082 D_fake: 0.424 +(epoch: 349, iters: 5822, time: 0.063) G_GAN: 0.536 G_GAN_Feat: 0.745 G_ID: 0.102 G_Rec: 0.367 D_GP: 0.040 D_real: 1.371 D_fake: 0.470 +(epoch: 349, iters: 6222, time: 0.063) G_GAN: 0.552 G_GAN_Feat: 1.104 G_ID: 0.129 G_Rec: 0.438 D_GP: 0.042 D_real: 0.831 D_fake: 0.457 +(epoch: 349, iters: 6622, time: 0.064) G_GAN: 0.250 G_GAN_Feat: 0.677 G_ID: 0.121 G_Rec: 0.327 D_GP: 0.027 D_real: 1.175 D_fake: 0.751 +(epoch: 349, iters: 7022, time: 0.064) G_GAN: 0.367 G_GAN_Feat: 0.862 G_ID: 0.115 G_Rec: 0.384 D_GP: 0.044 D_real: 1.094 D_fake: 0.633 +(epoch: 349, iters: 7422, time: 0.063) G_GAN: 0.295 G_GAN_Feat: 0.788 G_ID: 0.127 G_Rec: 0.334 D_GP: 0.080 D_real: 0.967 D_fake: 0.708 +(epoch: 349, iters: 7822, time: 0.063) G_GAN: 0.467 G_GAN_Feat: 1.029 G_ID: 0.113 G_Rec: 0.450 D_GP: 0.040 D_real: 1.003 D_fake: 0.546 +(epoch: 349, iters: 8222, time: 0.063) G_GAN: 0.126 G_GAN_Feat: 0.795 G_ID: 0.090 G_Rec: 0.313 D_GP: 0.027 D_real: 1.063 D_fake: 0.874 +(epoch: 349, iters: 8622, time: 0.064) G_GAN: 0.384 G_GAN_Feat: 1.074 G_ID: 0.146 G_Rec: 0.471 D_GP: 0.094 D_real: 0.657 D_fake: 0.624 +(epoch: 350, iters: 414, time: 0.063) G_GAN: 0.203 G_GAN_Feat: 0.933 G_ID: 0.111 G_Rec: 0.323 D_GP: 0.059 D_real: 1.159 D_fake: 0.800 +(epoch: 350, iters: 814, time: 0.063) G_GAN: 0.876 G_GAN_Feat: 1.041 G_ID: 0.139 G_Rec: 0.436 D_GP: 0.037 D_real: 1.475 D_fake: 0.194 +(epoch: 350, iters: 1214, time: 0.063) G_GAN: 0.002 G_GAN_Feat: 0.795 G_ID: 0.107 G_Rec: 0.310 D_GP: 0.025 D_real: 0.758 D_fake: 0.998 +(epoch: 350, iters: 1614, time: 0.064) G_GAN: 0.620 G_GAN_Feat: 1.113 G_ID: 0.110 G_Rec: 0.408 D_GP: 0.068 D_real: 0.417 D_fake: 0.391 +(epoch: 350, iters: 2014, time: 0.063) G_GAN: 0.178 G_GAN_Feat: 0.874 G_ID: 0.097 G_Rec: 0.332 D_GP: 0.028 D_real: 0.941 D_fake: 0.822 +(epoch: 350, iters: 2414, time: 0.063) G_GAN: 1.007 G_GAN_Feat: 1.268 G_ID: 0.122 G_Rec: 0.507 D_GP: 0.182 D_real: 0.532 D_fake: 0.257 +(epoch: 350, iters: 2814, time: 0.063) G_GAN: 0.044 G_GAN_Feat: 0.686 G_ID: 0.103 G_Rec: 0.346 D_GP: 0.025 D_real: 0.956 D_fake: 0.956 +(epoch: 350, iters: 3214, time: 0.064) G_GAN: 0.401 G_GAN_Feat: 0.794 G_ID: 0.133 G_Rec: 0.382 D_GP: 0.026 D_real: 1.222 D_fake: 0.610 +(epoch: 350, iters: 3614, time: 0.063) G_GAN: 0.173 G_GAN_Feat: 0.650 G_ID: 0.107 G_Rec: 0.297 D_GP: 0.029 D_real: 1.071 D_fake: 0.828 +(epoch: 350, iters: 4014, time: 0.063) G_GAN: -0.070 G_GAN_Feat: 0.874 G_ID: 0.177 G_Rec: 0.411 D_GP: 0.031 D_real: 0.646 D_fake: 1.071 +(epoch: 350, iters: 4414, time: 0.063) G_GAN: 0.109 G_GAN_Feat: 0.678 G_ID: 0.096 G_Rec: 0.310 D_GP: 0.030 D_real: 0.964 D_fake: 0.892 +(epoch: 350, iters: 4814, time: 0.064) G_GAN: 0.078 G_GAN_Feat: 0.936 G_ID: 0.127 G_Rec: 0.463 D_GP: 0.039 D_real: 0.691 D_fake: 0.923 +(epoch: 350, iters: 5214, time: 0.063) G_GAN: 0.205 G_GAN_Feat: 0.703 G_ID: 0.101 G_Rec: 0.301 D_GP: 0.036 D_real: 1.060 D_fake: 0.802 +(epoch: 350, iters: 5614, time: 0.063) G_GAN: 0.383 G_GAN_Feat: 0.966 G_ID: 0.128 G_Rec: 0.461 D_GP: 0.049 D_real: 0.814 D_fake: 0.634 +(epoch: 350, iters: 6014, time: 0.063) G_GAN: 0.132 G_GAN_Feat: 0.753 G_ID: 0.097 G_Rec: 0.350 D_GP: 0.057 D_real: 0.796 D_fake: 0.868 +(epoch: 350, iters: 6414, time: 0.064) G_GAN: 0.325 G_GAN_Feat: 0.908 G_ID: 0.125 G_Rec: 0.412 D_GP: 0.052 D_real: 0.787 D_fake: 0.679 +(epoch: 350, iters: 6814, time: 0.063) G_GAN: -0.100 G_GAN_Feat: 0.685 G_ID: 0.120 G_Rec: 0.301 D_GP: 0.034 D_real: 0.742 D_fake: 1.100 +(epoch: 350, iters: 7214, time: 0.063) G_GAN: 0.579 G_GAN_Feat: 1.001 G_ID: 0.138 G_Rec: 0.458 D_GP: 0.082 D_real: 0.844 D_fake: 0.441 +(epoch: 350, iters: 7614, time: 0.063) G_GAN: 0.053 G_GAN_Feat: 0.734 G_ID: 0.093 G_Rec: 0.343 D_GP: 0.030 D_real: 1.024 D_fake: 0.950 +(epoch: 350, iters: 8014, time: 0.064) G_GAN: 0.267 G_GAN_Feat: 0.997 G_ID: 0.126 G_Rec: 0.441 D_GP: 0.055 D_real: 0.590 D_fake: 0.734 +(epoch: 350, iters: 8414, time: 0.064) G_GAN: 0.279 G_GAN_Feat: 0.793 G_ID: 0.110 G_Rec: 0.356 D_GP: 0.064 D_real: 0.953 D_fake: 0.726 +(epoch: 351, iters: 206, time: 0.064) G_GAN: 0.674 G_GAN_Feat: 1.022 G_ID: 0.133 G_Rec: 0.437 D_GP: 0.090 D_real: 0.877 D_fake: 0.376 +(epoch: 351, iters: 606, time: 0.064) G_GAN: 0.187 G_GAN_Feat: 0.811 G_ID: 0.101 G_Rec: 0.313 D_GP: 0.123 D_real: 0.690 D_fake: 0.814 +(epoch: 351, iters: 1006, time: 0.064) G_GAN: 0.513 G_GAN_Feat: 1.072 G_ID: 0.134 G_Rec: 0.449 D_GP: 0.062 D_real: 0.933 D_fake: 0.493 +(epoch: 351, iters: 1406, time: 0.064) G_GAN: 0.524 G_GAN_Feat: 0.703 G_ID: 0.093 G_Rec: 0.272 D_GP: 0.034 D_real: 1.402 D_fake: 0.488 +(epoch: 351, iters: 1806, time: 0.064) G_GAN: 0.604 G_GAN_Feat: 1.152 G_ID: 0.131 G_Rec: 0.464 D_GP: 0.319 D_real: 0.411 D_fake: 0.439 +(epoch: 351, iters: 2206, time: 0.064) G_GAN: 0.368 G_GAN_Feat: 0.853 G_ID: 0.110 G_Rec: 0.322 D_GP: 0.338 D_real: 0.614 D_fake: 0.644 +(epoch: 351, iters: 2606, time: 0.064) G_GAN: 0.497 G_GAN_Feat: 1.051 G_ID: 0.120 G_Rec: 0.431 D_GP: 0.066 D_real: 0.847 D_fake: 0.509 +(epoch: 351, iters: 3006, time: 0.064) G_GAN: 0.415 G_GAN_Feat: 0.923 G_ID: 0.116 G_Rec: 0.352 D_GP: 0.103 D_real: 0.514 D_fake: 0.609 +(epoch: 351, iters: 3406, time: 0.063) G_GAN: 0.568 G_GAN_Feat: 0.983 G_ID: 0.129 G_Rec: 0.456 D_GP: 0.042 D_real: 0.979 D_fake: 0.443 +(epoch: 351, iters: 3806, time: 0.063) G_GAN: 0.353 G_GAN_Feat: 0.889 G_ID: 0.106 G_Rec: 0.305 D_GP: 0.049 D_real: 0.624 D_fake: 0.649 +(epoch: 351, iters: 4206, time: 0.064) G_GAN: 0.238 G_GAN_Feat: 1.053 G_ID: 0.137 G_Rec: 0.404 D_GP: 0.044 D_real: 0.564 D_fake: 0.763 +(epoch: 351, iters: 4606, time: 0.064) G_GAN: 0.484 G_GAN_Feat: 0.749 G_ID: 0.102 G_Rec: 0.319 D_GP: 0.041 D_real: 1.449 D_fake: 0.523 +(epoch: 351, iters: 5006, time: 0.064) G_GAN: 0.445 G_GAN_Feat: 1.006 G_ID: 0.143 G_Rec: 0.434 D_GP: 0.080 D_real: 0.634 D_fake: 0.561 +(epoch: 351, iters: 5406, time: 0.064) G_GAN: 0.350 G_GAN_Feat: 0.987 G_ID: 0.114 G_Rec: 0.358 D_GP: 0.356 D_real: 0.402 D_fake: 0.651 +(epoch: 351, iters: 5806, time: 0.064) G_GAN: 0.194 G_GAN_Feat: 0.974 G_ID: 0.115 G_Rec: 0.461 D_GP: 0.032 D_real: 0.848 D_fake: 0.807 +(epoch: 351, iters: 6206, time: 0.064) G_GAN: 0.046 G_GAN_Feat: 0.652 G_ID: 0.106 G_Rec: 0.313 D_GP: 0.032 D_real: 0.879 D_fake: 0.954 +(epoch: 351, iters: 6606, time: 0.064) G_GAN: 0.426 G_GAN_Feat: 0.983 G_ID: 0.126 G_Rec: 0.466 D_GP: 0.048 D_real: 1.015 D_fake: 0.598 +(epoch: 351, iters: 7006, time: 0.063) G_GAN: 0.213 G_GAN_Feat: 0.791 G_ID: 0.100 G_Rec: 0.348 D_GP: 0.080 D_real: 0.799 D_fake: 0.787 +(epoch: 351, iters: 7406, time: 0.063) G_GAN: 0.357 G_GAN_Feat: 0.950 G_ID: 0.131 G_Rec: 0.436 D_GP: 0.048 D_real: 0.914 D_fake: 0.660 +(epoch: 351, iters: 7806, time: 0.064) G_GAN: 0.168 G_GAN_Feat: 0.872 G_ID: 0.108 G_Rec: 0.363 D_GP: 0.063 D_real: 0.709 D_fake: 0.833 +(epoch: 351, iters: 8206, time: 0.063) G_GAN: 0.594 G_GAN_Feat: 1.010 G_ID: 0.120 G_Rec: 0.490 D_GP: 0.048 D_real: 1.162 D_fake: 0.436 +(epoch: 351, iters: 8606, time: 0.064) G_GAN: 0.130 G_GAN_Feat: 0.739 G_ID: 0.104 G_Rec: 0.331 D_GP: 0.043 D_real: 0.863 D_fake: 0.871 +(epoch: 352, iters: 398, time: 0.063) G_GAN: 0.590 G_GAN_Feat: 0.852 G_ID: 0.113 G_Rec: 0.378 D_GP: 0.031 D_real: 1.353 D_fake: 0.442 +(epoch: 352, iters: 798, time: 0.063) G_GAN: 0.011 G_GAN_Feat: 0.805 G_ID: 0.097 G_Rec: 0.338 D_GP: 0.037 D_real: 0.797 D_fake: 0.989 +(epoch: 352, iters: 1198, time: 0.063) G_GAN: 0.370 G_GAN_Feat: 0.992 G_ID: 0.132 G_Rec: 0.453 D_GP: 0.045 D_real: 0.720 D_fake: 0.641 +(epoch: 352, iters: 1598, time: 0.064) G_GAN: 0.217 G_GAN_Feat: 0.850 G_ID: 0.105 G_Rec: 0.319 D_GP: 0.161 D_real: 0.482 D_fake: 0.783 +(epoch: 352, iters: 1998, time: 0.063) G_GAN: 0.700 G_GAN_Feat: 1.200 G_ID: 0.112 G_Rec: 0.445 D_GP: 0.089 D_real: 0.561 D_fake: 0.329 +(epoch: 352, iters: 2398, time: 0.063) G_GAN: 0.284 G_GAN_Feat: 0.721 G_ID: 0.092 G_Rec: 0.283 D_GP: 0.031 D_real: 1.123 D_fake: 0.716 +(epoch: 352, iters: 2798, time: 0.063) G_GAN: 0.577 G_GAN_Feat: 0.984 G_ID: 0.121 G_Rec: 0.423 D_GP: 0.036 D_real: 1.187 D_fake: 0.430 +(epoch: 352, iters: 3198, time: 0.064) G_GAN: 0.512 G_GAN_Feat: 0.928 G_ID: 0.105 G_Rec: 0.331 D_GP: 0.044 D_real: 1.014 D_fake: 0.489 +(epoch: 352, iters: 3598, time: 0.064) G_GAN: 0.583 G_GAN_Feat: 1.057 G_ID: 0.112 G_Rec: 0.426 D_GP: 0.040 D_real: 1.040 D_fake: 0.435 +(epoch: 352, iters: 3998, time: 0.064) G_GAN: -0.089 G_GAN_Feat: 0.886 G_ID: 0.102 G_Rec: 0.373 D_GP: 0.156 D_real: 0.558 D_fake: 1.090 +(epoch: 352, iters: 4398, time: 0.064) G_GAN: 0.268 G_GAN_Feat: 0.944 G_ID: 0.120 G_Rec: 0.424 D_GP: 0.032 D_real: 0.911 D_fake: 0.733 +(epoch: 352, iters: 4798, time: 0.064) G_GAN: -0.094 G_GAN_Feat: 0.710 G_ID: 0.115 G_Rec: 0.299 D_GP: 0.032 D_real: 0.875 D_fake: 1.094 +(epoch: 352, iters: 5198, time: 0.064) G_GAN: 0.549 G_GAN_Feat: 1.031 G_ID: 0.126 G_Rec: 0.420 D_GP: 0.044 D_real: 0.980 D_fake: 0.461 +(epoch: 352, iters: 5598, time: 0.064) G_GAN: 0.384 G_GAN_Feat: 0.902 G_ID: 0.119 G_Rec: 0.319 D_GP: 0.037 D_real: 0.874 D_fake: 0.629 +(epoch: 352, iters: 5998, time: 0.064) G_GAN: 0.673 G_GAN_Feat: 1.119 G_ID: 0.113 G_Rec: 0.429 D_GP: 0.064 D_real: 0.576 D_fake: 0.352 +(epoch: 352, iters: 6398, time: 0.064) G_GAN: 0.040 G_GAN_Feat: 0.681 G_ID: 0.103 G_Rec: 0.299 D_GP: 0.027 D_real: 0.967 D_fake: 0.960 +(epoch: 352, iters: 6798, time: 0.064) G_GAN: 0.661 G_GAN_Feat: 1.068 G_ID: 0.112 G_Rec: 0.454 D_GP: 0.039 D_real: 1.037 D_fake: 0.368 +(epoch: 352, iters: 7198, time: 0.064) G_GAN: 0.116 G_GAN_Feat: 0.828 G_ID: 0.122 G_Rec: 0.320 D_GP: 0.036 D_real: 0.806 D_fake: 0.897 +(epoch: 352, iters: 7598, time: 0.064) G_GAN: 0.588 G_GAN_Feat: 1.122 G_ID: 0.128 G_Rec: 0.467 D_GP: 0.053 D_real: 0.980 D_fake: 0.449 +(epoch: 352, iters: 7998, time: 0.064) G_GAN: 0.170 G_GAN_Feat: 0.899 G_ID: 0.129 G_Rec: 0.327 D_GP: 0.322 D_real: 0.703 D_fake: 0.838 +(epoch: 352, iters: 8398, time: 0.064) G_GAN: 0.516 G_GAN_Feat: 0.954 G_ID: 0.125 G_Rec: 0.393 D_GP: 0.053 D_real: 1.097 D_fake: 0.498 +(epoch: 353, iters: 190, time: 0.064) G_GAN: 0.053 G_GAN_Feat: 0.761 G_ID: 0.123 G_Rec: 0.327 D_GP: 0.028 D_real: 0.987 D_fake: 0.947 +(epoch: 353, iters: 590, time: 0.064) G_GAN: 0.445 G_GAN_Feat: 0.948 G_ID: 0.121 G_Rec: 0.437 D_GP: 0.035 D_real: 1.053 D_fake: 0.565 +(epoch: 353, iters: 990, time: 0.064) G_GAN: 0.380 G_GAN_Feat: 0.761 G_ID: 0.097 G_Rec: 0.307 D_GP: 0.032 D_real: 1.239 D_fake: 0.621 +(epoch: 353, iters: 1390, time: 0.064) G_GAN: 0.372 G_GAN_Feat: 0.877 G_ID: 0.130 G_Rec: 0.440 D_GP: 0.032 D_real: 1.200 D_fake: 0.628 +(epoch: 353, iters: 1790, time: 0.064) G_GAN: 0.235 G_GAN_Feat: 0.716 G_ID: 0.104 G_Rec: 0.295 D_GP: 0.031 D_real: 1.149 D_fake: 0.766 +(epoch: 353, iters: 2190, time: 0.064) G_GAN: 0.453 G_GAN_Feat: 0.894 G_ID: 0.138 G_Rec: 0.408 D_GP: 0.035 D_real: 0.967 D_fake: 0.552 +(epoch: 353, iters: 2590, time: 0.064) G_GAN: 0.081 G_GAN_Feat: 0.754 G_ID: 0.114 G_Rec: 0.372 D_GP: 0.030 D_real: 0.968 D_fake: 0.919 +(epoch: 353, iters: 2990, time: 0.064) G_GAN: 0.367 G_GAN_Feat: 1.039 G_ID: 0.130 G_Rec: 0.439 D_GP: 0.044 D_real: 0.722 D_fake: 0.636 +(epoch: 353, iters: 3390, time: 0.064) G_GAN: 0.344 G_GAN_Feat: 0.970 G_ID: 0.100 G_Rec: 0.345 D_GP: 0.185 D_real: 0.327 D_fake: 0.692 +(epoch: 353, iters: 3790, time: 0.064) G_GAN: 0.312 G_GAN_Feat: 1.251 G_ID: 0.146 G_Rec: 0.461 D_GP: 0.065 D_real: 0.202 D_fake: 0.691 +(epoch: 353, iters: 4190, time: 0.064) G_GAN: 0.266 G_GAN_Feat: 0.741 G_ID: 0.084 G_Rec: 0.341 D_GP: 0.030 D_real: 1.238 D_fake: 0.736 +(epoch: 353, iters: 4590, time: 0.064) G_GAN: 0.337 G_GAN_Feat: 0.884 G_ID: 0.134 G_Rec: 0.440 D_GP: 0.031 D_real: 1.074 D_fake: 0.664 +(epoch: 353, iters: 4990, time: 0.064) G_GAN: 0.014 G_GAN_Feat: 0.656 G_ID: 0.109 G_Rec: 0.303 D_GP: 0.029 D_real: 0.903 D_fake: 0.987 +(epoch: 353, iters: 5390, time: 0.064) G_GAN: 0.348 G_GAN_Feat: 0.840 G_ID: 0.119 G_Rec: 0.388 D_GP: 0.034 D_real: 0.958 D_fake: 0.663 +(epoch: 353, iters: 5790, time: 0.064) G_GAN: 0.093 G_GAN_Feat: 0.565 G_ID: 0.081 G_Rec: 0.269 D_GP: 0.028 D_real: 1.081 D_fake: 0.908 +(epoch: 353, iters: 6190, time: 0.064) G_GAN: 0.436 G_GAN_Feat: 0.801 G_ID: 0.132 G_Rec: 0.380 D_GP: 0.035 D_real: 1.142 D_fake: 0.596 +(epoch: 353, iters: 6590, time: 0.064) G_GAN: -0.069 G_GAN_Feat: 0.719 G_ID: 0.099 G_Rec: 0.366 D_GP: 0.045 D_real: 0.805 D_fake: 1.069 +(epoch: 353, iters: 6990, time: 0.064) G_GAN: -0.044 G_GAN_Feat: 0.840 G_ID: 0.129 G_Rec: 0.405 D_GP: 0.039 D_real: 0.591 D_fake: 1.044 +(epoch: 353, iters: 7390, time: 0.064) G_GAN: -0.122 G_GAN_Feat: 0.687 G_ID: 0.110 G_Rec: 0.321 D_GP: 0.039 D_real: 0.709 D_fake: 1.122 +(epoch: 353, iters: 7790, time: 0.064) G_GAN: 0.538 G_GAN_Feat: 0.943 G_ID: 0.121 G_Rec: 0.493 D_GP: 0.042 D_real: 1.152 D_fake: 0.546 +(epoch: 353, iters: 8190, time: 0.064) G_GAN: 0.093 G_GAN_Feat: 0.677 G_ID: 0.106 G_Rec: 0.314 D_GP: 0.048 D_real: 0.854 D_fake: 0.907 +(epoch: 353, iters: 8590, time: 0.064) G_GAN: 0.155 G_GAN_Feat: 0.896 G_ID: 0.130 G_Rec: 0.415 D_GP: 0.060 D_real: 0.749 D_fake: 0.845 +(epoch: 354, iters: 382, time: 0.064) G_GAN: -0.078 G_GAN_Feat: 0.752 G_ID: 0.123 G_Rec: 0.339 D_GP: 0.052 D_real: 0.759 D_fake: 1.079 +(epoch: 354, iters: 782, time: 0.064) G_GAN: -0.084 G_GAN_Feat: 1.019 G_ID: 0.131 G_Rec: 0.483 D_GP: 0.052 D_real: 0.392 D_fake: 1.084 +(epoch: 354, iters: 1182, time: 0.064) G_GAN: 0.093 G_GAN_Feat: 0.687 G_ID: 0.086 G_Rec: 0.314 D_GP: 0.074 D_real: 0.805 D_fake: 0.907 +(epoch: 354, iters: 1582, time: 0.064) G_GAN: 0.425 G_GAN_Feat: 0.878 G_ID: 0.131 G_Rec: 0.403 D_GP: 0.040 D_real: 1.055 D_fake: 0.581 +(epoch: 354, iters: 1982, time: 0.064) G_GAN: 0.209 G_GAN_Feat: 0.679 G_ID: 0.099 G_Rec: 0.286 D_GP: 0.045 D_real: 1.021 D_fake: 0.794 +(epoch: 354, iters: 2382, time: 0.064) G_GAN: 0.201 G_GAN_Feat: 0.916 G_ID: 0.131 G_Rec: 0.425 D_GP: 0.064 D_real: 0.724 D_fake: 0.799 +(epoch: 354, iters: 2782, time: 0.064) G_GAN: 0.077 G_GAN_Feat: 0.785 G_ID: 0.102 G_Rec: 0.343 D_GP: 0.050 D_real: 0.763 D_fake: 0.923 +(epoch: 354, iters: 3182, time: 0.064) G_GAN: 0.285 G_GAN_Feat: 0.901 G_ID: 0.118 G_Rec: 0.440 D_GP: 0.042 D_real: 0.914 D_fake: 0.715 +(epoch: 354, iters: 3582, time: 0.064) G_GAN: 0.159 G_GAN_Feat: 0.812 G_ID: 0.109 G_Rec: 0.333 D_GP: 0.061 D_real: 0.823 D_fake: 0.841 +(epoch: 354, iters: 3982, time: 0.064) G_GAN: 0.057 G_GAN_Feat: 0.955 G_ID: 0.129 G_Rec: 0.426 D_GP: 0.047 D_real: 0.730 D_fake: 0.943 +(epoch: 354, iters: 4382, time: 0.064) G_GAN: 0.287 G_GAN_Feat: 0.869 G_ID: 0.102 G_Rec: 0.341 D_GP: 0.089 D_real: 1.064 D_fake: 0.720 +(epoch: 354, iters: 4782, time: 0.064) G_GAN: 0.499 G_GAN_Feat: 1.049 G_ID: 0.127 G_Rec: 0.462 D_GP: 0.427 D_real: 0.629 D_fake: 0.511 +(epoch: 354, iters: 5182, time: 0.064) G_GAN: 0.121 G_GAN_Feat: 0.888 G_ID: 0.101 G_Rec: 0.351 D_GP: 0.087 D_real: 0.514 D_fake: 0.879 +(epoch: 354, iters: 5582, time: 0.064) G_GAN: 0.627 G_GAN_Feat: 0.963 G_ID: 0.115 G_Rec: 0.432 D_GP: 0.066 D_real: 1.143 D_fake: 0.395 +(epoch: 354, iters: 5982, time: 0.064) G_GAN: 0.083 G_GAN_Feat: 0.824 G_ID: 0.113 G_Rec: 0.326 D_GP: 0.095 D_real: 0.520 D_fake: 0.919 +(epoch: 354, iters: 6382, time: 0.064) G_GAN: 0.401 G_GAN_Feat: 0.904 G_ID: 0.113 G_Rec: 0.425 D_GP: 0.041 D_real: 1.068 D_fake: 0.601 +(epoch: 354, iters: 6782, time: 0.064) G_GAN: 0.107 G_GAN_Feat: 0.758 G_ID: 0.104 G_Rec: 0.319 D_GP: 0.045 D_real: 0.910 D_fake: 0.893 +(epoch: 354, iters: 7182, time: 0.064) G_GAN: 0.733 G_GAN_Feat: 0.912 G_ID: 0.118 G_Rec: 0.433 D_GP: 0.032 D_real: 1.412 D_fake: 0.318 +(epoch: 354, iters: 7582, time: 0.064) G_GAN: 0.357 G_GAN_Feat: 0.801 G_ID: 0.109 G_Rec: 0.402 D_GP: 0.041 D_real: 0.934 D_fake: 0.644 +(epoch: 354, iters: 7982, time: 0.063) G_GAN: 0.196 G_GAN_Feat: 1.052 G_ID: 0.140 G_Rec: 0.454 D_GP: 0.157 D_real: 0.269 D_fake: 0.814 +(epoch: 354, iters: 8382, time: 0.064) G_GAN: 0.138 G_GAN_Feat: 0.855 G_ID: 0.107 G_Rec: 0.318 D_GP: 0.064 D_real: 0.526 D_fake: 0.862 +(epoch: 355, iters: 174, time: 0.064) G_GAN: 0.465 G_GAN_Feat: 1.020 G_ID: 0.115 G_Rec: 0.431 D_GP: 0.040 D_real: 0.749 D_fake: 0.547 +(epoch: 355, iters: 574, time: 0.064) G_GAN: 0.123 G_GAN_Feat: 0.894 G_ID: 0.115 G_Rec: 0.364 D_GP: 0.083 D_real: 0.536 D_fake: 0.882 +(epoch: 355, iters: 974, time: 0.064) G_GAN: 0.604 G_GAN_Feat: 0.922 G_ID: 0.114 G_Rec: 0.399 D_GP: 0.035 D_real: 1.320 D_fake: 0.464 +(epoch: 355, iters: 1374, time: 0.064) G_GAN: 0.236 G_GAN_Feat: 0.777 G_ID: 0.104 G_Rec: 0.349 D_GP: 0.033 D_real: 0.984 D_fake: 0.765 +(epoch: 355, iters: 1774, time: 0.064) G_GAN: 0.476 G_GAN_Feat: 0.936 G_ID: 0.115 G_Rec: 0.420 D_GP: 0.037 D_real: 1.120 D_fake: 0.525 +(epoch: 355, iters: 2174, time: 0.063) G_GAN: 0.548 G_GAN_Feat: 0.835 G_ID: 0.092 G_Rec: 0.326 D_GP: 0.044 D_real: 1.094 D_fake: 0.481 +(epoch: 355, iters: 2574, time: 0.064) G_GAN: 0.627 G_GAN_Feat: 0.888 G_ID: 0.118 G_Rec: 0.417 D_GP: 0.027 D_real: 1.321 D_fake: 0.387 +(epoch: 355, iters: 2974, time: 0.064) G_GAN: 0.134 G_GAN_Feat: 0.729 G_ID: 0.093 G_Rec: 0.287 D_GP: 0.038 D_real: 0.968 D_fake: 0.866 +(epoch: 355, iters: 3374, time: 0.064) G_GAN: 0.782 G_GAN_Feat: 0.875 G_ID: 0.111 G_Rec: 0.372 D_GP: 0.046 D_real: 1.342 D_fake: 0.286 +(epoch: 355, iters: 3774, time: 0.064) G_GAN: 0.702 G_GAN_Feat: 0.899 G_ID: 0.090 G_Rec: 0.304 D_GP: 0.039 D_real: 1.479 D_fake: 0.356 +(epoch: 355, iters: 4174, time: 0.064) G_GAN: 0.300 G_GAN_Feat: 1.080 G_ID: 0.117 G_Rec: 0.487 D_GP: 0.047 D_real: 0.728 D_fake: 0.705 +(epoch: 355, iters: 4574, time: 0.064) G_GAN: 0.299 G_GAN_Feat: 0.823 G_ID: 0.096 G_Rec: 0.340 D_GP: 0.049 D_real: 0.924 D_fake: 0.704 +(epoch: 355, iters: 4974, time: 0.063) G_GAN: 0.400 G_GAN_Feat: 0.919 G_ID: 0.142 G_Rec: 0.444 D_GP: 0.034 D_real: 1.013 D_fake: 0.600 +(epoch: 355, iters: 5374, time: 0.063) G_GAN: 0.327 G_GAN_Feat: 0.751 G_ID: 0.084 G_Rec: 0.359 D_GP: 0.029 D_real: 1.102 D_fake: 0.673 +(epoch: 355, iters: 5774, time: 0.064) G_GAN: 1.076 G_GAN_Feat: 1.116 G_ID: 0.107 G_Rec: 0.439 D_GP: 0.064 D_real: 1.194 D_fake: 0.163 +(epoch: 355, iters: 6174, time: 0.064) G_GAN: 0.573 G_GAN_Feat: 0.803 G_ID: 0.120 G_Rec: 0.341 D_GP: 0.032 D_real: 1.384 D_fake: 0.446 +(epoch: 355, iters: 6574, time: 0.064) G_GAN: 0.482 G_GAN_Feat: 1.060 G_ID: 0.130 G_Rec: 0.497 D_GP: 0.042 D_real: 0.945 D_fake: 0.525 +(epoch: 355, iters: 6974, time: 0.064) G_GAN: 0.520 G_GAN_Feat: 0.895 G_ID: 0.108 G_Rec: 0.354 D_GP: 0.061 D_real: 1.020 D_fake: 0.544 +(epoch: 355, iters: 7374, time: 0.064) G_GAN: 0.686 G_GAN_Feat: 1.369 G_ID: 0.125 G_Rec: 0.501 D_GP: 0.126 D_real: 1.111 D_fake: 0.403 +(epoch: 355, iters: 7774, time: 0.064) G_GAN: -0.080 G_GAN_Feat: 0.691 G_ID: 0.128 G_Rec: 0.331 D_GP: 0.027 D_real: 0.827 D_fake: 1.080 +(epoch: 355, iters: 8174, time: 0.064) G_GAN: 0.227 G_GAN_Feat: 1.034 G_ID: 0.145 G_Rec: 0.486 D_GP: 0.037 D_real: 0.767 D_fake: 0.778 +(epoch: 355, iters: 8574, time: 0.064) G_GAN: 0.195 G_GAN_Feat: 0.724 G_ID: 0.101 G_Rec: 0.349 D_GP: 0.064 D_real: 0.955 D_fake: 0.808 +(epoch: 356, iters: 366, time: 0.064) G_GAN: 0.332 G_GAN_Feat: 0.930 G_ID: 0.141 G_Rec: 0.465 D_GP: 0.043 D_real: 0.840 D_fake: 0.670 +(epoch: 356, iters: 766, time: 0.064) G_GAN: 0.245 G_GAN_Feat: 0.775 G_ID: 0.127 G_Rec: 0.317 D_GP: 0.033 D_real: 1.151 D_fake: 0.756 +(epoch: 356, iters: 1166, time: 0.064) G_GAN: 0.595 G_GAN_Feat: 1.021 G_ID: 0.124 G_Rec: 0.406 D_GP: 0.073 D_real: 0.691 D_fake: 0.421 +(epoch: 356, iters: 1566, time: 0.064) G_GAN: 0.451 G_GAN_Feat: 0.910 G_ID: 0.117 G_Rec: 0.318 D_GP: 0.057 D_real: 0.500 D_fake: 0.564 +(epoch: 356, iters: 1966, time: 0.064) G_GAN: 0.751 G_GAN_Feat: 0.939 G_ID: 0.124 G_Rec: 0.432 D_GP: 0.033 D_real: 1.369 D_fake: 0.316 +(epoch: 356, iters: 2366, time: 0.064) G_GAN: 0.006 G_GAN_Feat: 0.723 G_ID: 0.099 G_Rec: 0.316 D_GP: 0.029 D_real: 0.997 D_fake: 0.994 +(epoch: 356, iters: 2766, time: 0.064) G_GAN: 0.496 G_GAN_Feat: 0.977 G_ID: 0.130 G_Rec: 0.449 D_GP: 0.037 D_real: 0.990 D_fake: 0.546 +(epoch: 356, iters: 3166, time: 0.064) G_GAN: 0.027 G_GAN_Feat: 0.853 G_ID: 0.099 G_Rec: 0.360 D_GP: 0.077 D_real: 0.673 D_fake: 0.975 +(epoch: 356, iters: 3566, time: 0.064) G_GAN: 0.247 G_GAN_Feat: 0.975 G_ID: 0.119 G_Rec: 0.433 D_GP: 0.056 D_real: 0.682 D_fake: 0.754 +(epoch: 356, iters: 3966, time: 0.064) G_GAN: 0.147 G_GAN_Feat: 0.693 G_ID: 0.112 G_Rec: 0.293 D_GP: 0.039 D_real: 1.032 D_fake: 0.853 +(epoch: 356, iters: 4366, time: 0.064) G_GAN: 0.636 G_GAN_Feat: 1.074 G_ID: 0.135 G_Rec: 0.500 D_GP: 0.071 D_real: 0.858 D_fake: 0.389 +(epoch: 356, iters: 4766, time: 0.064) G_GAN: 0.567 G_GAN_Feat: 0.973 G_ID: 0.125 G_Rec: 0.340 D_GP: 0.628 D_real: 0.482 D_fake: 0.486 +(epoch: 356, iters: 5166, time: 0.064) G_GAN: 0.874 G_GAN_Feat: 1.138 G_ID: 0.118 G_Rec: 0.448 D_GP: 0.137 D_real: 0.637 D_fake: 0.242 +(epoch: 356, iters: 5566, time: 0.064) G_GAN: -0.077 G_GAN_Feat: 0.819 G_ID: 0.117 G_Rec: 0.314 D_GP: 0.039 D_real: 0.682 D_fake: 1.077 +(epoch: 356, iters: 5966, time: 0.064) G_GAN: 0.430 G_GAN_Feat: 0.986 G_ID: 0.151 G_Rec: 0.440 D_GP: 0.034 D_real: 1.162 D_fake: 0.575 +(epoch: 356, iters: 6366, time: 0.064) G_GAN: -0.030 G_GAN_Feat: 0.676 G_ID: 0.135 G_Rec: 0.299 D_GP: 0.030 D_real: 0.837 D_fake: 1.030 +(epoch: 356, iters: 6766, time: 0.064) G_GAN: 0.290 G_GAN_Feat: 0.931 G_ID: 0.133 G_Rec: 0.426 D_GP: 0.033 D_real: 0.930 D_fake: 0.711 +(epoch: 356, iters: 7166, time: 0.064) G_GAN: 0.261 G_GAN_Feat: 0.771 G_ID: 0.091 G_Rec: 0.328 D_GP: 0.129 D_real: 0.941 D_fake: 0.742 +(epoch: 356, iters: 7566, time: 0.064) G_GAN: 0.242 G_GAN_Feat: 0.997 G_ID: 0.143 G_Rec: 0.449 D_GP: 0.267 D_real: 0.515 D_fake: 0.760 +(epoch: 356, iters: 7966, time: 0.064) G_GAN: 0.208 G_GAN_Feat: 0.782 G_ID: 0.095 G_Rec: 0.346 D_GP: 0.030 D_real: 0.940 D_fake: 0.793 +(epoch: 356, iters: 8366, time: 0.064) G_GAN: 0.654 G_GAN_Feat: 0.956 G_ID: 0.124 G_Rec: 0.412 D_GP: 0.039 D_real: 1.110 D_fake: 0.365 +(epoch: 357, iters: 158, time: 0.064) G_GAN: 0.127 G_GAN_Feat: 0.669 G_ID: 0.104 G_Rec: 0.303 D_GP: 0.026 D_real: 1.118 D_fake: 0.874 +(epoch: 357, iters: 558, time: 0.064) G_GAN: 0.243 G_GAN_Feat: 0.891 G_ID: 0.138 G_Rec: 0.440 D_GP: 0.036 D_real: 0.957 D_fake: 0.760 +(epoch: 357, iters: 958, time: 0.064) G_GAN: -0.155 G_GAN_Feat: 0.710 G_ID: 0.111 G_Rec: 0.310 D_GP: 0.049 D_real: 0.617 D_fake: 1.155 +(epoch: 357, iters: 1358, time: 0.064) G_GAN: 0.080 G_GAN_Feat: 0.967 G_ID: 0.131 G_Rec: 0.449 D_GP: 0.053 D_real: 0.527 D_fake: 0.922 +(epoch: 357, iters: 1758, time: 0.064) G_GAN: -0.001 G_GAN_Feat: 0.799 G_ID: 0.107 G_Rec: 0.331 D_GP: 0.046 D_real: 0.721 D_fake: 1.001 +(epoch: 357, iters: 2158, time: 0.064) G_GAN: 0.797 G_GAN_Feat: 0.924 G_ID: 0.111 G_Rec: 0.426 D_GP: 0.035 D_real: 1.349 D_fake: 0.349 +(epoch: 357, iters: 2558, time: 0.064) G_GAN: 0.261 G_GAN_Feat: 0.819 G_ID: 0.098 G_Rec: 0.332 D_GP: 0.067 D_real: 1.027 D_fake: 0.741 +(epoch: 357, iters: 2958, time: 0.064) G_GAN: 0.325 G_GAN_Feat: 1.086 G_ID: 0.113 G_Rec: 0.450 D_GP: 0.178 D_real: 0.455 D_fake: 0.681 +(epoch: 357, iters: 3358, time: 0.064) G_GAN: -0.191 G_GAN_Feat: 0.898 G_ID: 0.112 G_Rec: 0.335 D_GP: 0.292 D_real: 0.262 D_fake: 1.191 +(epoch: 357, iters: 3758, time: 0.064) G_GAN: 0.275 G_GAN_Feat: 0.995 G_ID: 0.142 G_Rec: 0.471 D_GP: 0.033 D_real: 0.798 D_fake: 0.725 +(epoch: 357, iters: 4158, time: 0.064) G_GAN: 0.085 G_GAN_Feat: 0.804 G_ID: 0.095 G_Rec: 0.372 D_GP: 0.083 D_real: 0.661 D_fake: 0.915 +(epoch: 357, iters: 4558, time: 0.064) G_GAN: 0.111 G_GAN_Feat: 0.984 G_ID: 0.134 G_Rec: 0.388 D_GP: 0.044 D_real: 0.735 D_fake: 0.889 +(epoch: 357, iters: 4958, time: 0.064) G_GAN: 0.446 G_GAN_Feat: 0.973 G_ID: 0.100 G_Rec: 0.359 D_GP: 0.098 D_real: 0.627 D_fake: 0.560 +(epoch: 357, iters: 5358, time: 0.064) G_GAN: 0.657 G_GAN_Feat: 0.933 G_ID: 0.121 G_Rec: 0.380 D_GP: 0.036 D_real: 1.303 D_fake: 0.371 +(epoch: 357, iters: 5758, time: 0.064) G_GAN: 0.377 G_GAN_Feat: 0.819 G_ID: 0.131 G_Rec: 0.328 D_GP: 0.051 D_real: 1.174 D_fake: 0.632 +(epoch: 357, iters: 6158, time: 0.064) G_GAN: 0.446 G_GAN_Feat: 0.950 G_ID: 0.114 G_Rec: 0.456 D_GP: 0.034 D_real: 1.010 D_fake: 0.555 +(epoch: 357, iters: 6558, time: 0.063) G_GAN: 0.419 G_GAN_Feat: 0.753 G_ID: 0.093 G_Rec: 0.311 D_GP: 0.033 D_real: 1.235 D_fake: 0.589 +(epoch: 357, iters: 6958, time: 0.063) G_GAN: 0.365 G_GAN_Feat: 0.952 G_ID: 0.103 G_Rec: 0.430 D_GP: 0.035 D_real: 1.049 D_fake: 0.636 +(epoch: 357, iters: 7358, time: 0.063) G_GAN: 0.137 G_GAN_Feat: 0.842 G_ID: 0.089 G_Rec: 0.303 D_GP: 0.028 D_real: 0.846 D_fake: 0.863 +(epoch: 357, iters: 7758, time: 0.064) G_GAN: 0.694 G_GAN_Feat: 0.974 G_ID: 0.114 G_Rec: 0.474 D_GP: 0.030 D_real: 1.328 D_fake: 0.344 +(epoch: 357, iters: 8158, time: 0.063) G_GAN: 0.130 G_GAN_Feat: 0.815 G_ID: 0.093 G_Rec: 0.347 D_GP: 0.049 D_real: 0.842 D_fake: 0.872 +(epoch: 357, iters: 8558, time: 0.063) G_GAN: 0.442 G_GAN_Feat: 1.233 G_ID: 0.125 G_Rec: 0.512 D_GP: 0.220 D_real: 0.406 D_fake: 0.570 +(epoch: 358, iters: 350, time: 0.064) G_GAN: 0.363 G_GAN_Feat: 0.826 G_ID: 0.094 G_Rec: 0.304 D_GP: 0.045 D_real: 0.951 D_fake: 0.638 +(epoch: 358, iters: 750, time: 0.064) G_GAN: 0.651 G_GAN_Feat: 1.036 G_ID: 0.137 G_Rec: 0.399 D_GP: 0.045 D_real: 0.954 D_fake: 0.377 +(epoch: 358, iters: 1150, time: 0.064) G_GAN: 0.099 G_GAN_Feat: 0.837 G_ID: 0.091 G_Rec: 0.332 D_GP: 0.040 D_real: 0.849 D_fake: 0.901 +(epoch: 358, iters: 1550, time: 0.063) G_GAN: 0.453 G_GAN_Feat: 0.894 G_ID: 0.148 G_Rec: 0.437 D_GP: 0.030 D_real: 1.066 D_fake: 0.551 +(epoch: 358, iters: 1950, time: 0.063) G_GAN: 0.186 G_GAN_Feat: 0.902 G_ID: 0.109 G_Rec: 0.321 D_GP: 0.099 D_real: 0.385 D_fake: 0.830 +(epoch: 358, iters: 2350, time: 0.064) G_GAN: -0.043 G_GAN_Feat: 1.263 G_ID: 0.139 G_Rec: 0.490 D_GP: 0.124 D_real: 0.712 D_fake: 1.044 +(epoch: 358, iters: 2750, time: 0.063) G_GAN: 0.125 G_GAN_Feat: 0.720 G_ID: 0.140 G_Rec: 0.307 D_GP: 0.030 D_real: 0.956 D_fake: 0.875 +(epoch: 358, iters: 3150, time: 0.064) G_GAN: 0.382 G_GAN_Feat: 0.933 G_ID: 0.149 G_Rec: 0.440 D_GP: 0.030 D_real: 1.101 D_fake: 0.620 +(epoch: 358, iters: 3550, time: 0.063) G_GAN: 0.285 G_GAN_Feat: 0.835 G_ID: 0.103 G_Rec: 0.340 D_GP: 0.052 D_real: 0.777 D_fake: 0.723 +(epoch: 358, iters: 3950, time: 0.064) G_GAN: 0.466 G_GAN_Feat: 0.890 G_ID: 0.124 G_Rec: 0.384 D_GP: 0.039 D_real: 1.159 D_fake: 0.542 +(epoch: 358, iters: 4350, time: 0.063) G_GAN: 0.437 G_GAN_Feat: 0.636 G_ID: 0.091 G_Rec: 0.270 D_GP: 0.029 D_real: 1.364 D_fake: 0.564 +(epoch: 358, iters: 4750, time: 0.063) G_GAN: 0.479 G_GAN_Feat: 0.946 G_ID: 0.131 G_Rec: 0.423 D_GP: 0.035 D_real: 0.938 D_fake: 0.533 +(epoch: 358, iters: 5150, time: 0.063) G_GAN: 0.074 G_GAN_Feat: 0.768 G_ID: 0.092 G_Rec: 0.319 D_GP: 0.050 D_real: 0.695 D_fake: 0.926 +(epoch: 358, iters: 5550, time: 0.064) G_GAN: 0.600 G_GAN_Feat: 1.044 G_ID: 0.124 G_Rec: 0.473 D_GP: 0.064 D_real: 1.018 D_fake: 0.460 +(epoch: 358, iters: 5950, time: 0.063) G_GAN: 0.438 G_GAN_Feat: 0.775 G_ID: 0.102 G_Rec: 0.340 D_GP: 0.055 D_real: 1.218 D_fake: 0.568 +(epoch: 358, iters: 6350, time: 0.063) G_GAN: 0.585 G_GAN_Feat: 1.227 G_ID: 0.122 G_Rec: 0.446 D_GP: 0.178 D_real: 0.519 D_fake: 0.437 +(epoch: 358, iters: 6750, time: 0.063) G_GAN: 0.333 G_GAN_Feat: 0.869 G_ID: 0.114 G_Rec: 0.330 D_GP: 0.043 D_real: 0.894 D_fake: 0.668 +(epoch: 358, iters: 7150, time: 0.064) G_GAN: 0.841 G_GAN_Feat: 1.128 G_ID: 0.130 G_Rec: 0.405 D_GP: 0.412 D_real: 0.743 D_fake: 0.202 +(epoch: 358, iters: 7550, time: 0.063) G_GAN: 0.307 G_GAN_Feat: 0.800 G_ID: 0.102 G_Rec: 0.332 D_GP: 0.034 D_real: 1.208 D_fake: 0.700 +(epoch: 358, iters: 7950, time: 0.063) G_GAN: 0.435 G_GAN_Feat: 1.150 G_ID: 0.131 G_Rec: 0.455 D_GP: 0.165 D_real: 0.302 D_fake: 0.575 +(epoch: 358, iters: 8350, time: 0.063) G_GAN: 0.039 G_GAN_Feat: 0.663 G_ID: 0.102 G_Rec: 0.321 D_GP: 0.028 D_real: 0.952 D_fake: 0.961 +(epoch: 359, iters: 142, time: 0.064) G_GAN: 0.370 G_GAN_Feat: 0.947 G_ID: 0.116 G_Rec: 0.436 D_GP: 0.035 D_real: 1.080 D_fake: 0.654 +(epoch: 359, iters: 542, time: 0.063) G_GAN: 0.088 G_GAN_Feat: 0.643 G_ID: 0.092 G_Rec: 0.314 D_GP: 0.028 D_real: 0.993 D_fake: 0.912 +(epoch: 359, iters: 942, time: 0.064) G_GAN: 0.323 G_GAN_Feat: 0.850 G_ID: 0.123 G_Rec: 0.386 D_GP: 0.032 D_real: 1.042 D_fake: 0.686 +(epoch: 359, iters: 1342, time: 0.063) G_GAN: 0.090 G_GAN_Feat: 0.645 G_ID: 0.101 G_Rec: 0.298 D_GP: 0.029 D_real: 1.045 D_fake: 0.910 +(epoch: 359, iters: 1742, time: 0.064) G_GAN: 0.329 G_GAN_Feat: 0.874 G_ID: 0.138 G_Rec: 0.402 D_GP: 0.044 D_real: 0.930 D_fake: 0.676 +(epoch: 359, iters: 2142, time: 0.063) G_GAN: 0.308 G_GAN_Feat: 0.705 G_ID: 0.098 G_Rec: 0.286 D_GP: 0.040 D_real: 1.116 D_fake: 0.694 +(epoch: 359, iters: 2542, time: 0.063) G_GAN: 0.518 G_GAN_Feat: 1.013 G_ID: 0.107 G_Rec: 0.474 D_GP: 0.064 D_real: 1.022 D_fake: 0.530 +(epoch: 359, iters: 2942, time: 0.063) G_GAN: 0.290 G_GAN_Feat: 0.721 G_ID: 0.122 G_Rec: 0.325 D_GP: 0.030 D_real: 1.142 D_fake: 0.710 +(epoch: 359, iters: 3342, time: 0.064) G_GAN: 0.514 G_GAN_Feat: 1.005 G_ID: 0.109 G_Rec: 0.478 D_GP: 0.042 D_real: 1.121 D_fake: 0.492 +(epoch: 359, iters: 3742, time: 0.063) G_GAN: 0.160 G_GAN_Feat: 0.654 G_ID: 0.097 G_Rec: 0.294 D_GP: 0.029 D_real: 1.006 D_fake: 0.840 +(epoch: 359, iters: 4142, time: 0.063) G_GAN: 0.237 G_GAN_Feat: 0.920 G_ID: 0.136 G_Rec: 0.434 D_GP: 0.044 D_real: 0.860 D_fake: 0.764 +(epoch: 359, iters: 4542, time: 0.063) G_GAN: 0.307 G_GAN_Feat: 0.728 G_ID: 0.104 G_Rec: 0.299 D_GP: 0.034 D_real: 1.081 D_fake: 0.694 +(epoch: 359, iters: 4942, time: 0.064) G_GAN: 0.604 G_GAN_Feat: 1.026 G_ID: 0.128 G_Rec: 0.464 D_GP: 0.060 D_real: 1.031 D_fake: 0.407 +(epoch: 359, iters: 5342, time: 0.063) G_GAN: 0.188 G_GAN_Feat: 0.752 G_ID: 0.101 G_Rec: 0.308 D_GP: 0.037 D_real: 0.986 D_fake: 0.812 +(epoch: 359, iters: 5742, time: 0.063) G_GAN: 0.011 G_GAN_Feat: 1.137 G_ID: 0.123 G_Rec: 0.469 D_GP: 0.041 D_real: 0.471 D_fake: 0.989 +(epoch: 359, iters: 6142, time: 0.063) G_GAN: 0.192 G_GAN_Feat: 0.876 G_ID: 0.108 G_Rec: 0.340 D_GP: 0.151 D_real: 0.493 D_fake: 0.808 +(epoch: 359, iters: 6542, time: 0.064) G_GAN: 0.712 G_GAN_Feat: 0.903 G_ID: 0.133 G_Rec: 0.448 D_GP: 0.032 D_real: 1.365 D_fake: 0.303 +(epoch: 359, iters: 6942, time: 0.063) G_GAN: 0.232 G_GAN_Feat: 0.751 G_ID: 0.121 G_Rec: 0.300 D_GP: 0.038 D_real: 1.101 D_fake: 0.769 +(epoch: 359, iters: 7342, time: 0.063) G_GAN: 0.180 G_GAN_Feat: 0.987 G_ID: 0.140 G_Rec: 0.454 D_GP: 0.029 D_real: 0.816 D_fake: 0.821 +(epoch: 359, iters: 7742, time: 0.063) G_GAN: -0.106 G_GAN_Feat: 0.741 G_ID: 0.107 G_Rec: 0.326 D_GP: 0.029 D_real: 0.783 D_fake: 1.106 +(epoch: 359, iters: 8142, time: 0.064) G_GAN: 0.427 G_GAN_Feat: 0.936 G_ID: 0.104 G_Rec: 0.432 D_GP: 0.030 D_real: 0.931 D_fake: 0.591 +(epoch: 359, iters: 8542, time: 0.063) G_GAN: 0.102 G_GAN_Feat: 0.791 G_ID: 0.090 G_Rec: 0.335 D_GP: 0.058 D_real: 0.792 D_fake: 0.899 +(epoch: 360, iters: 334, time: 0.063) G_GAN: 0.321 G_GAN_Feat: 0.936 G_ID: 0.134 G_Rec: 0.420 D_GP: 0.036 D_real: 0.981 D_fake: 0.679 +(epoch: 360, iters: 734, time: 0.063) G_GAN: 0.270 G_GAN_Feat: 0.736 G_ID: 0.091 G_Rec: 0.314 D_GP: 0.035 D_real: 1.148 D_fake: 0.731 +(epoch: 360, iters: 1134, time: 0.064) G_GAN: 0.738 G_GAN_Feat: 0.923 G_ID: 0.122 G_Rec: 0.511 D_GP: 0.029 D_real: 1.408 D_fake: 0.299 +(epoch: 360, iters: 1534, time: 0.063) G_GAN: 0.306 G_GAN_Feat: 0.800 G_ID: 0.104 G_Rec: 0.296 D_GP: 0.060 D_real: 0.792 D_fake: 0.739 +(epoch: 360, iters: 1934, time: 0.063) G_GAN: 0.393 G_GAN_Feat: 1.121 G_ID: 0.137 G_Rec: 0.440 D_GP: 0.094 D_real: 0.257 D_fake: 0.616 +(epoch: 360, iters: 2334, time: 0.063) G_GAN: 0.285 G_GAN_Feat: 0.762 G_ID: 0.109 G_Rec: 0.304 D_GP: 0.032 D_real: 1.089 D_fake: 0.715 +(epoch: 360, iters: 2734, time: 0.064) G_GAN: 0.566 G_GAN_Feat: 1.007 G_ID: 0.121 G_Rec: 0.464 D_GP: 0.034 D_real: 1.033 D_fake: 0.442 +(epoch: 360, iters: 3134, time: 0.063) G_GAN: 0.043 G_GAN_Feat: 0.675 G_ID: 0.108 G_Rec: 0.253 D_GP: 0.035 D_real: 1.017 D_fake: 0.957 +(epoch: 360, iters: 3534, time: 0.063) G_GAN: 0.385 G_GAN_Feat: 0.889 G_ID: 0.133 G_Rec: 0.417 D_GP: 0.029 D_real: 1.045 D_fake: 0.626 +(epoch: 360, iters: 3934, time: 0.063) G_GAN: 0.408 G_GAN_Feat: 0.967 G_ID: 0.118 G_Rec: 0.356 D_GP: 0.101 D_real: 0.394 D_fake: 0.647 +(epoch: 360, iters: 4334, time: 0.064) G_GAN: 0.549 G_GAN_Feat: 1.021 G_ID: 0.129 G_Rec: 0.441 D_GP: 0.041 D_real: 0.993 D_fake: 0.466 +(epoch: 360, iters: 4734, time: 0.063) G_GAN: 0.238 G_GAN_Feat: 0.722 G_ID: 0.101 G_Rec: 0.310 D_GP: 0.029 D_real: 1.012 D_fake: 0.763 +(epoch: 360, iters: 5134, time: 0.063) G_GAN: 0.571 G_GAN_Feat: 0.920 G_ID: 0.119 G_Rec: 0.441 D_GP: 0.029 D_real: 1.321 D_fake: 0.456 +(epoch: 360, iters: 5534, time: 0.063) G_GAN: -0.041 G_GAN_Feat: 0.743 G_ID: 0.121 G_Rec: 0.333 D_GP: 0.036 D_real: 0.824 D_fake: 1.041 +(epoch: 360, iters: 5934, time: 0.064) G_GAN: 0.418 G_GAN_Feat: 0.876 G_ID: 0.114 G_Rec: 0.413 D_GP: 0.033 D_real: 1.052 D_fake: 0.596 +(epoch: 360, iters: 6334, time: 0.063) G_GAN: 0.144 G_GAN_Feat: 0.718 G_ID: 0.114 G_Rec: 0.330 D_GP: 0.042 D_real: 0.910 D_fake: 0.856 +(epoch: 360, iters: 6734, time: 0.063) G_GAN: 0.576 G_GAN_Feat: 0.943 G_ID: 0.134 G_Rec: 0.422 D_GP: 0.033 D_real: 1.098 D_fake: 0.462 +(epoch: 360, iters: 7134, time: 0.063) G_GAN: 0.161 G_GAN_Feat: 0.725 G_ID: 0.096 G_Rec: 0.328 D_GP: 0.029 D_real: 1.029 D_fake: 0.839 +(epoch: 360, iters: 7534, time: 0.064) G_GAN: 0.809 G_GAN_Feat: 0.981 G_ID: 0.120 G_Rec: 0.455 D_GP: 0.035 D_real: 1.350 D_fake: 0.247 +(epoch: 360, iters: 7934, time: 0.063) G_GAN: 0.299 G_GAN_Feat: 0.752 G_ID: 0.104 G_Rec: 0.314 D_GP: 0.028 D_real: 1.078 D_fake: 0.701 +(epoch: 360, iters: 8334, time: 0.063) G_GAN: 0.601 G_GAN_Feat: 1.238 G_ID: 0.151 G_Rec: 0.509 D_GP: 0.079 D_real: 0.232 D_fake: 0.438 +(epoch: 361, iters: 126, time: 0.063) G_GAN: 0.532 G_GAN_Feat: 0.971 G_ID: 0.099 G_Rec: 0.339 D_GP: 0.913 D_real: 0.581 D_fake: 0.588 +(epoch: 361, iters: 526, time: 0.063) G_GAN: 0.307 G_GAN_Feat: 0.998 G_ID: 0.152 G_Rec: 0.470 D_GP: 0.035 D_real: 0.896 D_fake: 0.695 +(epoch: 361, iters: 926, time: 0.064) G_GAN: 0.330 G_GAN_Feat: 0.950 G_ID: 0.102 G_Rec: 0.332 D_GP: 0.169 D_real: 0.470 D_fake: 0.674 +(epoch: 361, iters: 1326, time: 0.063) G_GAN: 0.647 G_GAN_Feat: 1.090 G_ID: 0.123 G_Rec: 0.420 D_GP: 0.092 D_real: 0.496 D_fake: 0.372 +(epoch: 361, iters: 1726, time: 0.064) G_GAN: 0.256 G_GAN_Feat: 0.656 G_ID: 0.098 G_Rec: 0.272 D_GP: 0.035 D_real: 1.198 D_fake: 0.744 +(epoch: 361, iters: 2126, time: 0.063) G_GAN: 0.335 G_GAN_Feat: 0.855 G_ID: 0.128 G_Rec: 0.409 D_GP: 0.033 D_real: 0.990 D_fake: 0.669 +(epoch: 361, iters: 2526, time: 0.063) G_GAN: 0.199 G_GAN_Feat: 0.760 G_ID: 0.107 G_Rec: 0.311 D_GP: 0.033 D_real: 0.930 D_fake: 0.801 +(epoch: 361, iters: 2926, time: 0.064) G_GAN: 0.437 G_GAN_Feat: 1.049 G_ID: 0.128 G_Rec: 0.452 D_GP: 0.073 D_real: 0.831 D_fake: 0.575 +(epoch: 361, iters: 3326, time: 0.064) G_GAN: 0.150 G_GAN_Feat: 0.724 G_ID: 0.101 G_Rec: 0.298 D_GP: 0.032 D_real: 0.891 D_fake: 0.850 +(epoch: 361, iters: 3726, time: 0.063) G_GAN: 0.141 G_GAN_Feat: 0.917 G_ID: 0.137 G_Rec: 0.445 D_GP: 0.036 D_real: 0.765 D_fake: 0.861 +(epoch: 361, iters: 4126, time: 0.064) G_GAN: 0.011 G_GAN_Feat: 0.791 G_ID: 0.111 G_Rec: 0.295 D_GP: 0.037 D_real: 0.745 D_fake: 0.989 +(epoch: 361, iters: 4526, time: 0.064) G_GAN: 0.799 G_GAN_Feat: 1.035 G_ID: 0.132 G_Rec: 0.482 D_GP: 0.038 D_real: 1.266 D_fake: 0.268 +(epoch: 361, iters: 4926, time: 0.064) G_GAN: 0.289 G_GAN_Feat: 0.731 G_ID: 0.102 G_Rec: 0.312 D_GP: 0.035 D_real: 1.114 D_fake: 0.711 +(epoch: 361, iters: 5326, time: 0.063) G_GAN: 0.563 G_GAN_Feat: 0.977 G_ID: 0.124 G_Rec: 0.425 D_GP: 0.073 D_real: 0.789 D_fake: 0.459 +(epoch: 361, iters: 5726, time: 0.063) G_GAN: 0.338 G_GAN_Feat: 0.805 G_ID: 0.103 G_Rec: 0.326 D_GP: 0.038 D_real: 1.253 D_fake: 0.667 +(epoch: 361, iters: 6126, time: 0.063) G_GAN: 0.703 G_GAN_Feat: 0.860 G_ID: 0.115 G_Rec: 0.417 D_GP: 0.027 D_real: 1.415 D_fake: 0.352 +(epoch: 361, iters: 6526, time: 0.064) G_GAN: 0.082 G_GAN_Feat: 0.717 G_ID: 0.109 G_Rec: 0.317 D_GP: 0.036 D_real: 0.918 D_fake: 0.918 +(epoch: 361, iters: 6926, time: 0.063) G_GAN: 0.579 G_GAN_Feat: 0.863 G_ID: 0.124 G_Rec: 0.403 D_GP: 0.027 D_real: 1.267 D_fake: 0.454 +(epoch: 361, iters: 7326, time: 0.063) G_GAN: 0.263 G_GAN_Feat: 0.743 G_ID: 0.088 G_Rec: 0.301 D_GP: 0.034 D_real: 1.117 D_fake: 0.738 +(epoch: 361, iters: 7726, time: 0.063) G_GAN: 0.402 G_GAN_Feat: 1.013 G_ID: 0.127 G_Rec: 0.442 D_GP: 0.045 D_real: 0.817 D_fake: 0.599 +(epoch: 361, iters: 8126, time: 0.064) G_GAN: 0.395 G_GAN_Feat: 0.785 G_ID: 0.094 G_Rec: 0.336 D_GP: 0.054 D_real: 0.984 D_fake: 0.606 +(epoch: 361, iters: 8526, time: 0.063) G_GAN: 0.437 G_GAN_Feat: 0.877 G_ID: 0.121 G_Rec: 0.413 D_GP: 0.030 D_real: 1.180 D_fake: 0.567 +(epoch: 362, iters: 318, time: 0.063) G_GAN: 0.106 G_GAN_Feat: 0.818 G_ID: 0.112 G_Rec: 0.341 D_GP: 0.087 D_real: 0.752 D_fake: 0.895 +(epoch: 362, iters: 718, time: 0.063) G_GAN: 0.281 G_GAN_Feat: 0.934 G_ID: 0.120 G_Rec: 0.398 D_GP: 0.037 D_real: 0.765 D_fake: 0.721 +(epoch: 362, iters: 1118, time: 0.064) G_GAN: 0.119 G_GAN_Feat: 0.768 G_ID: 0.101 G_Rec: 0.438 D_GP: 0.029 D_real: 0.843 D_fake: 0.882 +(epoch: 362, iters: 1518, time: 0.063) G_GAN: 0.700 G_GAN_Feat: 1.002 G_ID: 0.119 G_Rec: 0.428 D_GP: 0.036 D_real: 1.297 D_fake: 0.313 +(epoch: 362, iters: 1918, time: 0.063) G_GAN: 0.436 G_GAN_Feat: 0.817 G_ID: 0.096 G_Rec: 0.309 D_GP: 0.037 D_real: 1.201 D_fake: 0.568 +(epoch: 362, iters: 2318, time: 0.063) G_GAN: 0.418 G_GAN_Feat: 1.088 G_ID: 0.120 G_Rec: 0.465 D_GP: 0.037 D_real: 0.942 D_fake: 0.585 +(epoch: 362, iters: 2718, time: 0.064) G_GAN: -0.110 G_GAN_Feat: 0.872 G_ID: 0.116 G_Rec: 0.375 D_GP: 0.036 D_real: 0.666 D_fake: 1.110 +(epoch: 362, iters: 3118, time: 0.064) G_GAN: 0.539 G_GAN_Feat: 1.029 G_ID: 0.138 G_Rec: 0.438 D_GP: 0.088 D_real: 0.735 D_fake: 0.481 +(epoch: 362, iters: 3518, time: 0.064) G_GAN: 0.137 G_GAN_Feat: 0.923 G_ID: 0.123 G_Rec: 0.320 D_GP: 0.067 D_real: 0.220 D_fake: 0.870 +(epoch: 362, iters: 3918, time: 0.063) G_GAN: 0.240 G_GAN_Feat: 0.837 G_ID: 0.132 G_Rec: 0.469 D_GP: 0.031 D_real: 1.006 D_fake: 0.761 +(epoch: 362, iters: 4318, time: 0.064) G_GAN: 0.085 G_GAN_Feat: 0.602 G_ID: 0.099 G_Rec: 0.313 D_GP: 0.029 D_real: 0.983 D_fake: 0.915 +(epoch: 362, iters: 4718, time: 0.064) G_GAN: 0.429 G_GAN_Feat: 0.792 G_ID: 0.128 G_Rec: 0.405 D_GP: 0.031 D_real: 1.200 D_fake: 0.574 +(epoch: 362, iters: 5118, time: 0.064) G_GAN: 0.264 G_GAN_Feat: 0.655 G_ID: 0.103 G_Rec: 0.315 D_GP: 0.038 D_real: 1.137 D_fake: 0.740 +(epoch: 362, iters: 5518, time: 0.063) G_GAN: 0.388 G_GAN_Feat: 0.875 G_ID: 0.120 G_Rec: 0.460 D_GP: 0.042 D_real: 1.037 D_fake: 0.616 +(epoch: 362, iters: 5918, time: 0.064) G_GAN: 0.216 G_GAN_Feat: 0.599 G_ID: 0.105 G_Rec: 0.285 D_GP: 0.033 D_real: 1.144 D_fake: 0.784 +(epoch: 362, iters: 6318, time: 0.064) G_GAN: 0.382 G_GAN_Feat: 0.930 G_ID: 0.124 G_Rec: 0.497 D_GP: 0.062 D_real: 0.930 D_fake: 0.628 +(epoch: 362, iters: 6718, time: 0.064) G_GAN: 0.036 G_GAN_Feat: 0.693 G_ID: 0.104 G_Rec: 0.325 D_GP: 0.040 D_real: 0.873 D_fake: 0.964 +(epoch: 362, iters: 7118, time: 0.064) G_GAN: 0.233 G_GAN_Feat: 0.792 G_ID: 0.125 G_Rec: 0.408 D_GP: 0.039 D_real: 0.936 D_fake: 0.771 +(epoch: 362, iters: 7518, time: 0.064) G_GAN: 0.097 G_GAN_Feat: 0.611 G_ID: 0.093 G_Rec: 0.288 D_GP: 0.029 D_real: 1.002 D_fake: 0.903 +(epoch: 362, iters: 7918, time: 0.064) G_GAN: 0.206 G_GAN_Feat: 0.936 G_ID: 0.126 G_Rec: 0.468 D_GP: 0.068 D_real: 0.694 D_fake: 0.795 +(epoch: 362, iters: 8318, time: 0.063) G_GAN: 0.037 G_GAN_Feat: 0.656 G_ID: 0.132 G_Rec: 0.285 D_GP: 0.042 D_real: 0.886 D_fake: 0.963 +(epoch: 363, iters: 110, time: 0.064) G_GAN: 0.584 G_GAN_Feat: 0.924 G_ID: 0.141 G_Rec: 0.419 D_GP: 0.043 D_real: 1.128 D_fake: 0.436 +(epoch: 363, iters: 510, time: 0.064) G_GAN: 0.305 G_GAN_Feat: 0.691 G_ID: 0.099 G_Rec: 0.304 D_GP: 0.042 D_real: 1.124 D_fake: 0.698 +(epoch: 363, iters: 910, time: 0.064) G_GAN: 0.550 G_GAN_Feat: 0.924 G_ID: 0.115 G_Rec: 0.434 D_GP: 0.036 D_real: 1.082 D_fake: 0.475 +(epoch: 363, iters: 1310, time: 0.063) G_GAN: 0.210 G_GAN_Feat: 0.736 G_ID: 0.094 G_Rec: 0.302 D_GP: 0.052 D_real: 0.890 D_fake: 0.791 +(epoch: 363, iters: 1710, time: 0.064) G_GAN: 0.539 G_GAN_Feat: 0.905 G_ID: 0.126 G_Rec: 0.407 D_GP: 0.073 D_real: 1.059 D_fake: 0.467 +(epoch: 363, iters: 2110, time: 0.064) G_GAN: 0.008 G_GAN_Feat: 0.782 G_ID: 0.130 G_Rec: 0.321 D_GP: 0.064 D_real: 0.658 D_fake: 0.992 +(epoch: 363, iters: 2510, time: 0.063) G_GAN: 0.545 G_GAN_Feat: 0.967 G_ID: 0.136 G_Rec: 0.466 D_GP: 0.064 D_real: 0.945 D_fake: 0.480 +(epoch: 363, iters: 2910, time: 0.064) G_GAN: -0.059 G_GAN_Feat: 0.778 G_ID: 0.118 G_Rec: 0.353 D_GP: 0.037 D_real: 0.941 D_fake: 1.060 +(epoch: 363, iters: 3310, time: 0.064) G_GAN: 0.368 G_GAN_Feat: 0.916 G_ID: 0.132 G_Rec: 0.446 D_GP: 0.086 D_real: 0.875 D_fake: 0.632 +(epoch: 363, iters: 3710, time: 0.064) G_GAN: 0.288 G_GAN_Feat: 0.801 G_ID: 0.101 G_Rec: 0.316 D_GP: 0.096 D_real: 0.797 D_fake: 0.719 +(epoch: 363, iters: 4110, time: 0.063) G_GAN: 0.582 G_GAN_Feat: 0.959 G_ID: 0.137 G_Rec: 0.391 D_GP: 0.046 D_real: 1.212 D_fake: 0.434 +(epoch: 363, iters: 4510, time: 0.063) G_GAN: 0.351 G_GAN_Feat: 0.841 G_ID: 0.099 G_Rec: 0.362 D_GP: 0.211 D_real: 0.564 D_fake: 0.657 +(epoch: 363, iters: 4910, time: 0.064) G_GAN: 0.564 G_GAN_Feat: 0.929 G_ID: 0.113 G_Rec: 0.449 D_GP: 0.034 D_real: 1.226 D_fake: 0.450 +(epoch: 363, iters: 5310, time: 0.064) G_GAN: 0.460 G_GAN_Feat: 0.786 G_ID: 0.104 G_Rec: 0.315 D_GP: 0.047 D_real: 1.017 D_fake: 0.545 +(epoch: 363, iters: 5710, time: 0.063) G_GAN: 0.403 G_GAN_Feat: 0.882 G_ID: 0.119 G_Rec: 0.391 D_GP: 0.032 D_real: 1.205 D_fake: 0.598 +(epoch: 363, iters: 6110, time: 0.063) G_GAN: 0.239 G_GAN_Feat: 0.797 G_ID: 0.127 G_Rec: 0.318 D_GP: 0.049 D_real: 0.994 D_fake: 0.762 +(epoch: 363, iters: 6510, time: 0.064) G_GAN: 0.653 G_GAN_Feat: 1.099 G_ID: 0.147 G_Rec: 0.451 D_GP: 0.059 D_real: 0.671 D_fake: 0.362 +(epoch: 363, iters: 6910, time: 0.064) G_GAN: 0.207 G_GAN_Feat: 0.858 G_ID: 0.109 G_Rec: 0.320 D_GP: 0.128 D_real: 0.603 D_fake: 0.811 +(epoch: 363, iters: 7310, time: 0.064) G_GAN: 0.122 G_GAN_Feat: 1.085 G_ID: 0.140 G_Rec: 0.469 D_GP: 0.260 D_real: 0.494 D_fake: 0.890 +(epoch: 363, iters: 7710, time: 0.064) G_GAN: 0.260 G_GAN_Feat: 0.747 G_ID: 0.103 G_Rec: 0.322 D_GP: 0.046 D_real: 1.019 D_fake: 0.740 +(epoch: 363, iters: 8110, time: 0.064) G_GAN: 0.006 G_GAN_Feat: 0.997 G_ID: 0.147 G_Rec: 0.397 D_GP: 0.067 D_real: 0.446 D_fake: 0.994 +(epoch: 363, iters: 8510, time: 0.064) G_GAN: -0.213 G_GAN_Feat: 0.849 G_ID: 0.096 G_Rec: 0.335 D_GP: 0.181 D_real: 0.239 D_fake: 1.214 +(epoch: 364, iters: 302, time: 0.064) G_GAN: 0.516 G_GAN_Feat: 0.884 G_ID: 0.119 G_Rec: 0.390 D_GP: 0.034 D_real: 1.105 D_fake: 0.489 +(epoch: 364, iters: 702, time: 0.063) G_GAN: 0.446 G_GAN_Feat: 1.005 G_ID: 0.113 G_Rec: 0.364 D_GP: 0.467 D_real: 0.555 D_fake: 0.563 +(epoch: 364, iters: 1102, time: 0.064) G_GAN: 0.769 G_GAN_Feat: 1.082 G_ID: 0.150 G_Rec: 0.503 D_GP: 0.055 D_real: 1.026 D_fake: 0.267 +(epoch: 364, iters: 1502, time: 0.064) G_GAN: -0.060 G_GAN_Feat: 0.964 G_ID: 0.112 G_Rec: 0.360 D_GP: 0.330 D_real: 0.188 D_fake: 1.063 +(epoch: 364, iters: 1902, time: 0.063) G_GAN: 0.926 G_GAN_Feat: 1.001 G_ID: 0.134 G_Rec: 0.458 D_GP: 0.032 D_real: 1.391 D_fake: 0.169 +(epoch: 364, iters: 2302, time: 0.064) G_GAN: 0.164 G_GAN_Feat: 0.925 G_ID: 0.104 G_Rec: 0.358 D_GP: 0.056 D_real: 0.491 D_fake: 0.836 +(epoch: 364, iters: 2702, time: 0.064) G_GAN: 0.751 G_GAN_Feat: 0.992 G_ID: 0.121 G_Rec: 0.414 D_GP: 0.031 D_real: 1.239 D_fake: 0.316 +(epoch: 364, iters: 3102, time: 0.064) G_GAN: 0.102 G_GAN_Feat: 0.643 G_ID: 0.121 G_Rec: 0.303 D_GP: 0.027 D_real: 0.997 D_fake: 0.898 +(epoch: 364, iters: 3502, time: 0.063) G_GAN: 0.166 G_GAN_Feat: 0.851 G_ID: 0.143 G_Rec: 0.439 D_GP: 0.031 D_real: 0.891 D_fake: 0.834 +(epoch: 364, iters: 3902, time: 0.063) G_GAN: -0.066 G_GAN_Feat: 0.688 G_ID: 0.115 G_Rec: 0.311 D_GP: 0.043 D_real: 0.748 D_fake: 1.066 +(epoch: 364, iters: 4302, time: 0.063) G_GAN: 0.118 G_GAN_Feat: 0.900 G_ID: 0.132 G_Rec: 0.407 D_GP: 0.045 D_real: 0.693 D_fake: 0.882 +(epoch: 364, iters: 4702, time: 0.064) G_GAN: 0.086 G_GAN_Feat: 0.740 G_ID: 0.109 G_Rec: 0.339 D_GP: 0.139 D_real: 0.736 D_fake: 0.914 +(epoch: 364, iters: 5102, time: 0.063) G_GAN: 0.531 G_GAN_Feat: 0.994 G_ID: 0.122 G_Rec: 0.470 D_GP: 0.196 D_real: 0.872 D_fake: 0.506 +(epoch: 364, iters: 5502, time: 0.063) G_GAN: 0.160 G_GAN_Feat: 0.619 G_ID: 0.101 G_Rec: 0.295 D_GP: 0.029 D_real: 1.091 D_fake: 0.840 +(epoch: 364, iters: 5902, time: 0.063) G_GAN: 0.755 G_GAN_Feat: 0.999 G_ID: 0.138 G_Rec: 0.445 D_GP: 0.054 D_real: 1.217 D_fake: 0.363 +(epoch: 364, iters: 6302, time: 0.064) G_GAN: -0.082 G_GAN_Feat: 0.737 G_ID: 0.099 G_Rec: 0.355 D_GP: 0.070 D_real: 0.669 D_fake: 1.082 +(epoch: 364, iters: 6702, time: 0.063) G_GAN: 0.343 G_GAN_Feat: 1.039 G_ID: 0.111 G_Rec: 0.485 D_GP: 0.052 D_real: 0.756 D_fake: 0.662 +(epoch: 364, iters: 7102, time: 0.063) G_GAN: -0.126 G_GAN_Feat: 0.705 G_ID: 0.100 G_Rec: 0.292 D_GP: 0.049 D_real: 0.673 D_fake: 1.126 +(epoch: 364, iters: 7502, time: 0.063) G_GAN: 0.284 G_GAN_Feat: 1.019 G_ID: 0.115 G_Rec: 0.474 D_GP: 0.054 D_real: 0.766 D_fake: 0.719 +(epoch: 364, iters: 7902, time: 0.064) G_GAN: 0.161 G_GAN_Feat: 0.714 G_ID: 0.105 G_Rec: 0.284 D_GP: 0.032 D_real: 1.019 D_fake: 0.840 +(epoch: 364, iters: 8302, time: 0.063) G_GAN: 0.438 G_GAN_Feat: 0.848 G_ID: 0.118 G_Rec: 0.367 D_GP: 0.037 D_real: 1.118 D_fake: 0.566 +(epoch: 364, iters: 8702, time: 0.064) G_GAN: 0.067 G_GAN_Feat: 0.773 G_ID: 0.128 G_Rec: 0.359 D_GP: 0.029 D_real: 0.966 D_fake: 0.934 +(epoch: 365, iters: 494, time: 0.063) G_GAN: 0.509 G_GAN_Feat: 1.055 G_ID: 0.112 G_Rec: 0.458 D_GP: 0.054 D_real: 0.800 D_fake: 0.497 +(epoch: 365, iters: 894, time: 0.064) G_GAN: 0.265 G_GAN_Feat: 0.757 G_ID: 0.105 G_Rec: 0.299 D_GP: 0.032 D_real: 1.001 D_fake: 0.737 +(epoch: 365, iters: 1294, time: 0.063) G_GAN: 0.588 G_GAN_Feat: 0.956 G_ID: 0.105 G_Rec: 0.400 D_GP: 0.031 D_real: 1.112 D_fake: 0.445 +(epoch: 365, iters: 1694, time: 0.063) G_GAN: 0.259 G_GAN_Feat: 0.840 G_ID: 0.090 G_Rec: 0.322 D_GP: 0.112 D_real: 0.752 D_fake: 0.747 +(epoch: 365, iters: 2094, time: 0.063) G_GAN: 0.500 G_GAN_Feat: 1.062 G_ID: 0.115 G_Rec: 0.421 D_GP: 0.041 D_real: 0.798 D_fake: 0.505 +(epoch: 365, iters: 2494, time: 0.064) G_GAN: 0.885 G_GAN_Feat: 1.002 G_ID: 0.095 G_Rec: 0.333 D_GP: 0.066 D_real: 0.980 D_fake: 0.205 +(epoch: 365, iters: 2894, time: 0.064) G_GAN: 0.603 G_GAN_Feat: 1.064 G_ID: 0.120 G_Rec: 0.419 D_GP: 0.030 D_real: 0.811 D_fake: 0.409 +(epoch: 365, iters: 3294, time: 0.063) G_GAN: 0.386 G_GAN_Feat: 0.930 G_ID: 0.101 G_Rec: 0.313 D_GP: 0.042 D_real: 0.556 D_fake: 0.666 +(epoch: 365, iters: 3694, time: 0.063) G_GAN: 0.775 G_GAN_Feat: 1.091 G_ID: 0.124 G_Rec: 0.497 D_GP: 0.112 D_real: 0.631 D_fake: 0.295 +(epoch: 365, iters: 4094, time: 0.064) G_GAN: 0.022 G_GAN_Feat: 0.940 G_ID: 0.100 G_Rec: 0.327 D_GP: 0.050 D_real: 0.232 D_fake: 0.978 +(epoch: 365, iters: 4494, time: 0.063) G_GAN: 0.545 G_GAN_Feat: 1.259 G_ID: 0.124 G_Rec: 0.486 D_GP: 0.040 D_real: 1.136 D_fake: 0.535 +(epoch: 365, iters: 4894, time: 0.063) G_GAN: 0.322 G_GAN_Feat: 0.714 G_ID: 0.098 G_Rec: 0.317 D_GP: 0.026 D_real: 1.278 D_fake: 0.683 +(epoch: 365, iters: 5294, time: 0.063) G_GAN: 0.424 G_GAN_Feat: 0.942 G_ID: 0.124 G_Rec: 0.426 D_GP: 0.034 D_real: 1.034 D_fake: 0.578 +(epoch: 365, iters: 5694, time: 0.063) G_GAN: 0.676 G_GAN_Feat: 0.782 G_ID: 0.101 G_Rec: 0.324 D_GP: 0.090 D_real: 1.270 D_fake: 0.343 +(epoch: 365, iters: 6094, time: 0.063) G_GAN: 0.489 G_GAN_Feat: 0.826 G_ID: 0.131 G_Rec: 0.392 D_GP: 0.030 D_real: 1.214 D_fake: 0.518 +(epoch: 365, iters: 6494, time: 0.063) G_GAN: 0.225 G_GAN_Feat: 0.771 G_ID: 0.109 G_Rec: 0.316 D_GP: 0.037 D_real: 0.979 D_fake: 0.776 +(epoch: 365, iters: 6894, time: 0.064) G_GAN: 0.497 G_GAN_Feat: 1.249 G_ID: 0.118 G_Rec: 0.518 D_GP: 0.050 D_real: 1.182 D_fake: 0.544 +(epoch: 365, iters: 7294, time: 0.063) G_GAN: 0.288 G_GAN_Feat: 0.749 G_ID: 0.126 G_Rec: 0.325 D_GP: 0.030 D_real: 1.152 D_fake: 0.718 +(epoch: 365, iters: 7694, time: 0.063) G_GAN: 0.605 G_GAN_Feat: 0.875 G_ID: 0.131 G_Rec: 0.408 D_GP: 0.041 D_real: 1.180 D_fake: 0.469 +(epoch: 365, iters: 8094, time: 0.064) G_GAN: 0.218 G_GAN_Feat: 0.737 G_ID: 0.095 G_Rec: 0.327 D_GP: 0.036 D_real: 0.904 D_fake: 0.783 +(epoch: 365, iters: 8494, time: 0.064) G_GAN: 0.453 G_GAN_Feat: 0.894 G_ID: 0.135 G_Rec: 0.434 D_GP: 0.030 D_real: 1.183 D_fake: 0.550 +(epoch: 366, iters: 286, time: 0.064) G_GAN: 0.236 G_GAN_Feat: 0.859 G_ID: 0.111 G_Rec: 0.346 D_GP: 0.034 D_real: 0.859 D_fake: 0.765 +(epoch: 366, iters: 686, time: 0.064) G_GAN: 0.733 G_GAN_Feat: 1.119 G_ID: 0.145 G_Rec: 0.493 D_GP: 0.047 D_real: 0.870 D_fake: 0.319 +(epoch: 366, iters: 1086, time: 0.064) G_GAN: 0.036 G_GAN_Feat: 1.082 G_ID: 0.126 G_Rec: 0.347 D_GP: 0.073 D_real: 0.088 D_fake: 0.964 +(epoch: 366, iters: 1486, time: 0.064) G_GAN: 0.448 G_GAN_Feat: 0.867 G_ID: 0.115 G_Rec: 0.419 D_GP: 0.027 D_real: 1.125 D_fake: 0.555 +(epoch: 366, iters: 1886, time: 0.063) G_GAN: 0.022 G_GAN_Feat: 0.734 G_ID: 0.087 G_Rec: 0.303 D_GP: 0.030 D_real: 0.909 D_fake: 0.978 +(epoch: 366, iters: 2286, time: 0.063) G_GAN: 0.221 G_GAN_Feat: 0.883 G_ID: 0.130 G_Rec: 0.402 D_GP: 0.037 D_real: 0.930 D_fake: 0.780 +(epoch: 366, iters: 2686, time: 0.063) G_GAN: 0.140 G_GAN_Feat: 0.735 G_ID: 0.095 G_Rec: 0.284 D_GP: 0.047 D_real: 0.787 D_fake: 0.860 +(epoch: 366, iters: 3086, time: 0.064) G_GAN: 0.591 G_GAN_Feat: 1.020 G_ID: 0.123 G_Rec: 0.430 D_GP: 0.036 D_real: 0.925 D_fake: 0.425 +(epoch: 366, iters: 3486, time: 0.064) G_GAN: 0.272 G_GAN_Feat: 0.908 G_ID: 0.098 G_Rec: 0.334 D_GP: 0.086 D_real: 0.517 D_fake: 0.728 +(epoch: 366, iters: 3886, time: 0.063) G_GAN: 0.567 G_GAN_Feat: 1.193 G_ID: 0.137 G_Rec: 0.446 D_GP: 0.131 D_real: 0.259 D_fake: 0.488 +(epoch: 366, iters: 4286, time: 0.063) G_GAN: 0.363 G_GAN_Feat: 0.856 G_ID: 0.116 G_Rec: 0.312 D_GP: 0.052 D_real: 0.966 D_fake: 0.643 +(epoch: 366, iters: 4686, time: 0.064) G_GAN: 0.403 G_GAN_Feat: 0.979 G_ID: 0.135 G_Rec: 0.476 D_GP: 0.037 D_real: 0.964 D_fake: 0.603 +(epoch: 366, iters: 5086, time: 0.063) G_GAN: 0.189 G_GAN_Feat: 0.731 G_ID: 0.096 G_Rec: 0.327 D_GP: 0.030 D_real: 1.018 D_fake: 0.816 +(epoch: 366, iters: 5486, time: 0.063) G_GAN: 0.574 G_GAN_Feat: 1.018 G_ID: 0.128 G_Rec: 0.485 D_GP: 0.096 D_real: 0.896 D_fake: 0.474 +(epoch: 366, iters: 5886, time: 0.063) G_GAN: 0.194 G_GAN_Feat: 0.736 G_ID: 0.101 G_Rec: 0.301 D_GP: 0.036 D_real: 1.037 D_fake: 0.807 +(epoch: 366, iters: 6286, time: 0.064) G_GAN: 0.282 G_GAN_Feat: 0.866 G_ID: 0.137 G_Rec: 0.380 D_GP: 0.036 D_real: 1.001 D_fake: 0.719 +(epoch: 366, iters: 6686, time: 0.063) G_GAN: 0.028 G_GAN_Feat: 0.957 G_ID: 0.115 G_Rec: 0.360 D_GP: 0.432 D_real: 0.184 D_fake: 0.979 +(epoch: 366, iters: 7086, time: 0.063) G_GAN: 0.662 G_GAN_Feat: 1.098 G_ID: 0.126 G_Rec: 0.477 D_GP: 0.049 D_real: 0.907 D_fake: 0.377 +(epoch: 366, iters: 7486, time: 0.063) G_GAN: 0.226 G_GAN_Feat: 0.915 G_ID: 0.097 G_Rec: 0.342 D_GP: 0.072 D_real: 0.502 D_fake: 0.775 +(epoch: 366, iters: 7886, time: 0.064) G_GAN: 0.358 G_GAN_Feat: 1.164 G_ID: 0.122 G_Rec: 0.437 D_GP: 0.089 D_real: 0.491 D_fake: 0.647 +(epoch: 366, iters: 8286, time: 0.063) G_GAN: 0.336 G_GAN_Feat: 0.846 G_ID: 0.106 G_Rec: 0.307 D_GP: 0.062 D_real: 0.725 D_fake: 0.664 +(epoch: 366, iters: 8686, time: 0.063) G_GAN: 0.511 G_GAN_Feat: 1.022 G_ID: 0.141 G_Rec: 0.462 D_GP: 0.038 D_real: 1.033 D_fake: 0.502 +(epoch: 367, iters: 478, time: 0.063) G_GAN: 0.251 G_GAN_Feat: 0.778 G_ID: 0.098 G_Rec: 0.320 D_GP: 0.043 D_real: 1.115 D_fake: 0.749 +(epoch: 367, iters: 878, time: 0.064) G_GAN: 0.324 G_GAN_Feat: 0.926 G_ID: 0.136 G_Rec: 0.453 D_GP: 0.037 D_real: 0.964 D_fake: 0.683 +(epoch: 367, iters: 1278, time: 0.063) G_GAN: 0.274 G_GAN_Feat: 0.811 G_ID: 0.105 G_Rec: 0.313 D_GP: 0.039 D_real: 0.893 D_fake: 0.727 +(epoch: 367, iters: 1678, time: 0.063) G_GAN: 0.975 G_GAN_Feat: 1.169 G_ID: 0.105 G_Rec: 0.464 D_GP: 0.324 D_real: 0.414 D_fake: 0.193 +(epoch: 367, iters: 2078, time: 0.063) G_GAN: 0.457 G_GAN_Feat: 0.946 G_ID: 0.104 G_Rec: 0.342 D_GP: 0.063 D_real: 0.967 D_fake: 0.615 +(epoch: 367, iters: 2478, time: 0.064) G_GAN: 0.701 G_GAN_Feat: 1.117 G_ID: 0.125 G_Rec: 0.460 D_GP: 0.101 D_real: 0.711 D_fake: 0.335 +(epoch: 367, iters: 2878, time: 0.063) G_GAN: -0.062 G_GAN_Feat: 0.864 G_ID: 0.115 G_Rec: 0.342 D_GP: 0.029 D_real: 0.781 D_fake: 1.062 +(epoch: 367, iters: 3278, time: 0.063) G_GAN: 0.897 G_GAN_Feat: 1.111 G_ID: 0.105 G_Rec: 0.464 D_GP: 0.033 D_real: 1.112 D_fake: 0.171 +(epoch: 367, iters: 3678, time: 0.063) G_GAN: -0.066 G_GAN_Feat: 0.731 G_ID: 0.120 G_Rec: 0.311 D_GP: 0.028 D_real: 0.872 D_fake: 1.066 +(epoch: 367, iters: 4078, time: 0.064) G_GAN: 0.408 G_GAN_Feat: 1.128 G_ID: 0.121 G_Rec: 0.470 D_GP: 0.069 D_real: 0.731 D_fake: 0.603 +(epoch: 367, iters: 4478, time: 0.064) G_GAN: -0.045 G_GAN_Feat: 0.846 G_ID: 0.103 G_Rec: 0.332 D_GP: 0.118 D_real: 0.501 D_fake: 1.045 +(epoch: 367, iters: 4878, time: 0.064) G_GAN: 0.936 G_GAN_Feat: 1.109 G_ID: 0.125 G_Rec: 0.480 D_GP: 0.105 D_real: 1.054 D_fake: 0.214 +(epoch: 367, iters: 5278, time: 0.063) G_GAN: 0.287 G_GAN_Feat: 0.756 G_ID: 0.102 G_Rec: 0.315 D_GP: 0.036 D_real: 1.115 D_fake: 0.714 +(epoch: 367, iters: 5678, time: 0.064) G_GAN: 0.584 G_GAN_Feat: 0.937 G_ID: 0.120 G_Rec: 0.397 D_GP: 0.042 D_real: 1.211 D_fake: 0.431 +(epoch: 367, iters: 6078, time: 0.063) G_GAN: 0.159 G_GAN_Feat: 0.890 G_ID: 0.107 G_Rec: 0.304 D_GP: 0.033 D_real: 0.558 D_fake: 0.841 +(epoch: 367, iters: 6478, time: 0.063) G_GAN: 0.785 G_GAN_Feat: 1.078 G_ID: 0.144 G_Rec: 0.484 D_GP: 0.032 D_real: 0.998 D_fake: 0.346 +(epoch: 367, iters: 6878, time: 0.063) G_GAN: 0.381 G_GAN_Feat: 1.062 G_ID: 0.094 G_Rec: 0.369 D_GP: 0.032 D_real: 1.135 D_fake: 0.687 +(epoch: 367, iters: 7278, time: 0.064) G_GAN: 0.325 G_GAN_Feat: 0.871 G_ID: 0.136 G_Rec: 0.442 D_GP: 0.027 D_real: 1.085 D_fake: 0.681 +(epoch: 367, iters: 7678, time: 0.063) G_GAN: 0.111 G_GAN_Feat: 0.620 G_ID: 0.081 G_Rec: 0.308 D_GP: 0.026 D_real: 1.092 D_fake: 0.890 +(epoch: 367, iters: 8078, time: 0.063) G_GAN: 0.398 G_GAN_Feat: 0.808 G_ID: 0.113 G_Rec: 0.416 D_GP: 0.028 D_real: 1.203 D_fake: 0.612 +(epoch: 367, iters: 8478, time: 0.063) G_GAN: 0.049 G_GAN_Feat: 0.614 G_ID: 0.090 G_Rec: 0.293 D_GP: 0.033 D_real: 0.979 D_fake: 0.951 +(epoch: 368, iters: 270, time: 0.064) G_GAN: 0.308 G_GAN_Feat: 0.826 G_ID: 0.121 G_Rec: 0.467 D_GP: 0.038 D_real: 0.992 D_fake: 0.719 +(epoch: 368, iters: 670, time: 0.064) G_GAN: 0.100 G_GAN_Feat: 0.623 G_ID: 0.103 G_Rec: 0.291 D_GP: 0.037 D_real: 0.927 D_fake: 0.901 +(epoch: 368, iters: 1070, time: 0.063) G_GAN: 0.089 G_GAN_Feat: 0.759 G_ID: 0.110 G_Rec: 0.366 D_GP: 0.038 D_real: 0.797 D_fake: 0.911 +(epoch: 368, iters: 1470, time: 0.063) G_GAN: -0.104 G_GAN_Feat: 0.648 G_ID: 0.105 G_Rec: 0.331 D_GP: 0.035 D_real: 0.796 D_fake: 1.104 +(epoch: 368, iters: 1870, time: 0.064) G_GAN: 0.029 G_GAN_Feat: 0.853 G_ID: 0.127 G_Rec: 0.412 D_GP: 0.046 D_real: 0.694 D_fake: 0.971 +(epoch: 368, iters: 2270, time: 0.063) G_GAN: -0.177 G_GAN_Feat: 0.678 G_ID: 0.111 G_Rec: 0.320 D_GP: 0.050 D_real: 0.673 D_fake: 1.177 +(epoch: 368, iters: 2670, time: 0.064) G_GAN: 0.370 G_GAN_Feat: 0.817 G_ID: 0.111 G_Rec: 0.405 D_GP: 0.041 D_real: 1.053 D_fake: 0.646 +(epoch: 368, iters: 3070, time: 0.063) G_GAN: 0.005 G_GAN_Feat: 0.697 G_ID: 0.096 G_Rec: 0.319 D_GP: 0.039 D_real: 0.903 D_fake: 0.995 +(epoch: 368, iters: 3470, time: 0.064) G_GAN: 0.309 G_GAN_Feat: 0.857 G_ID: 0.134 G_Rec: 0.381 D_GP: 0.044 D_real: 1.030 D_fake: 0.697 +(epoch: 368, iters: 3870, time: 0.063) G_GAN: 0.322 G_GAN_Feat: 0.704 G_ID: 0.105 G_Rec: 0.309 D_GP: 0.036 D_real: 1.160 D_fake: 0.679 +(epoch: 368, iters: 4270, time: 0.063) G_GAN: 0.112 G_GAN_Feat: 0.833 G_ID: 0.124 G_Rec: 0.373 D_GP: 0.043 D_real: 0.771 D_fake: 0.889 +(epoch: 368, iters: 4670, time: 0.063) G_GAN: 0.196 G_GAN_Feat: 0.783 G_ID: 0.101 G_Rec: 0.301 D_GP: 0.107 D_real: 0.769 D_fake: 0.804 +(epoch: 368, iters: 5070, time: 0.064) G_GAN: 0.622 G_GAN_Feat: 0.901 G_ID: 0.108 G_Rec: 0.378 D_GP: 0.047 D_real: 1.204 D_fake: 0.388 +(epoch: 368, iters: 5470, time: 0.064) G_GAN: 0.088 G_GAN_Feat: 0.659 G_ID: 0.094 G_Rec: 0.303 D_GP: 0.031 D_real: 1.007 D_fake: 0.913 +(epoch: 368, iters: 5870, time: 0.064) G_GAN: 0.344 G_GAN_Feat: 0.890 G_ID: 0.119 G_Rec: 0.432 D_GP: 0.034 D_real: 0.996 D_fake: 0.662 +(epoch: 368, iters: 6270, time: 0.064) G_GAN: -0.100 G_GAN_Feat: 0.721 G_ID: 0.145 G_Rec: 0.321 D_GP: 0.047 D_real: 0.722 D_fake: 1.100 +(epoch: 368, iters: 6670, time: 0.064) G_GAN: 0.118 G_GAN_Feat: 1.023 G_ID: 0.145 G_Rec: 0.473 D_GP: 0.103 D_real: 0.430 D_fake: 0.886 +(epoch: 368, iters: 7070, time: 0.064) G_GAN: 0.129 G_GAN_Feat: 0.702 G_ID: 0.104 G_Rec: 0.302 D_GP: 0.038 D_real: 0.869 D_fake: 0.871 +(epoch: 368, iters: 7470, time: 0.064) G_GAN: 0.678 G_GAN_Feat: 0.968 G_ID: 0.126 G_Rec: 0.474 D_GP: 0.035 D_real: 1.217 D_fake: 0.351 +(epoch: 368, iters: 7870, time: 0.064) G_GAN: 0.112 G_GAN_Feat: 0.785 G_ID: 0.121 G_Rec: 0.354 D_GP: 0.050 D_real: 0.817 D_fake: 0.888 +(epoch: 368, iters: 8270, time: 0.064) G_GAN: 0.377 G_GAN_Feat: 0.883 G_ID: 0.129 G_Rec: 0.379 D_GP: 0.041 D_real: 1.061 D_fake: 0.624 +(epoch: 368, iters: 8670, time: 0.064) G_GAN: 0.305 G_GAN_Feat: 0.771 G_ID: 0.104 G_Rec: 0.354 D_GP: 0.033 D_real: 1.103 D_fake: 0.695 +(epoch: 369, iters: 462, time: 0.064) G_GAN: 0.568 G_GAN_Feat: 0.923 G_ID: 0.131 G_Rec: 0.388 D_GP: 0.048 D_real: 1.243 D_fake: 0.439 +(epoch: 369, iters: 862, time: 0.064) G_GAN: 0.308 G_GAN_Feat: 0.668 G_ID: 0.103 G_Rec: 0.292 D_GP: 0.029 D_real: 1.245 D_fake: 0.692 +(epoch: 369, iters: 1262, time: 0.064) G_GAN: 0.413 G_GAN_Feat: 0.958 G_ID: 0.122 G_Rec: 0.434 D_GP: 0.042 D_real: 1.006 D_fake: 0.595 +(epoch: 369, iters: 1662, time: 0.064) G_GAN: 0.090 G_GAN_Feat: 0.788 G_ID: 0.117 G_Rec: 0.311 D_GP: 0.063 D_real: 0.659 D_fake: 0.910 +(epoch: 369, iters: 2062, time: 0.064) G_GAN: 0.456 G_GAN_Feat: 1.080 G_ID: 0.138 G_Rec: 0.490 D_GP: 0.051 D_real: 0.743 D_fake: 0.551 +(epoch: 369, iters: 2462, time: 0.064) G_GAN: 0.471 G_GAN_Feat: 0.839 G_ID: 0.111 G_Rec: 0.332 D_GP: 0.042 D_real: 1.314 D_fake: 0.531 +(epoch: 369, iters: 2862, time: 0.064) G_GAN: 0.874 G_GAN_Feat: 0.868 G_ID: 0.104 G_Rec: 0.433 D_GP: 0.028 D_real: 1.570 D_fake: 0.324 +(epoch: 369, iters: 3262, time: 0.064) G_GAN: -0.026 G_GAN_Feat: 0.637 G_ID: 0.119 G_Rec: 0.292 D_GP: 0.029 D_real: 0.899 D_fake: 1.026 +(epoch: 369, iters: 3662, time: 0.064) G_GAN: 0.378 G_GAN_Feat: 0.845 G_ID: 0.120 G_Rec: 0.406 D_GP: 0.036 D_real: 0.959 D_fake: 0.634 +(epoch: 369, iters: 4062, time: 0.064) G_GAN: 0.130 G_GAN_Feat: 0.626 G_ID: 0.113 G_Rec: 0.278 D_GP: 0.035 D_real: 1.015 D_fake: 0.871 +(epoch: 369, iters: 4462, time: 0.064) G_GAN: 0.188 G_GAN_Feat: 0.919 G_ID: 0.156 G_Rec: 0.440 D_GP: 0.037 D_real: 0.745 D_fake: 0.814 +(epoch: 369, iters: 4862, time: 0.064) G_GAN: 0.119 G_GAN_Feat: 0.724 G_ID: 0.099 G_Rec: 0.300 D_GP: 0.047 D_real: 0.871 D_fake: 0.881 +(epoch: 369, iters: 5262, time: 0.064) G_GAN: 0.564 G_GAN_Feat: 0.938 G_ID: 0.131 G_Rec: 0.427 D_GP: 0.074 D_real: 1.047 D_fake: 0.463 +(epoch: 369, iters: 5662, time: 0.064) G_GAN: 0.259 G_GAN_Feat: 0.805 G_ID: 0.092 G_Rec: 0.331 D_GP: 0.045 D_real: 0.951 D_fake: 0.742 +(epoch: 369, iters: 6062, time: 0.064) G_GAN: 0.617 G_GAN_Feat: 0.992 G_ID: 0.128 G_Rec: 0.442 D_GP: 0.041 D_real: 1.215 D_fake: 0.391 +(epoch: 369, iters: 6462, time: 0.064) G_GAN: 0.258 G_GAN_Feat: 0.725 G_ID: 0.092 G_Rec: 0.287 D_GP: 0.032 D_real: 1.147 D_fake: 0.742 +(epoch: 369, iters: 6862, time: 0.064) G_GAN: 0.649 G_GAN_Feat: 1.054 G_ID: 0.111 G_Rec: 0.451 D_GP: 0.053 D_real: 0.942 D_fake: 0.371 +(epoch: 369, iters: 7262, time: 0.064) G_GAN: 0.307 G_GAN_Feat: 0.871 G_ID: 0.100 G_Rec: 0.337 D_GP: 0.070 D_real: 0.704 D_fake: 0.694 +(epoch: 369, iters: 7662, time: 0.064) G_GAN: 0.485 G_GAN_Feat: 0.960 G_ID: 0.131 G_Rec: 0.452 D_GP: 0.052 D_real: 1.012 D_fake: 0.555 +(epoch: 369, iters: 8062, time: 0.064) G_GAN: 0.228 G_GAN_Feat: 0.717 G_ID: 0.116 G_Rec: 0.317 D_GP: 0.036 D_real: 1.043 D_fake: 0.773 +(epoch: 369, iters: 8462, time: 0.064) G_GAN: 0.344 G_GAN_Feat: 1.100 G_ID: 0.150 G_Rec: 0.489 D_GP: 0.061 D_real: 0.652 D_fake: 0.657 +(epoch: 370, iters: 254, time: 0.064) G_GAN: 0.385 G_GAN_Feat: 0.808 G_ID: 0.095 G_Rec: 0.303 D_GP: 0.034 D_real: 1.031 D_fake: 0.615 +(epoch: 370, iters: 654, time: 0.064) G_GAN: 1.037 G_GAN_Feat: 1.029 G_ID: 0.112 G_Rec: 0.478 D_GP: 0.102 D_real: 1.488 D_fake: 0.246 +(epoch: 370, iters: 1054, time: 0.064) G_GAN: 0.005 G_GAN_Feat: 0.754 G_ID: 0.112 G_Rec: 0.322 D_GP: 0.034 D_real: 0.766 D_fake: 0.995 +(epoch: 370, iters: 1454, time: 0.064) G_GAN: 0.057 G_GAN_Feat: 1.027 G_ID: 0.142 G_Rec: 0.487 D_GP: 0.068 D_real: 0.460 D_fake: 0.945 +(epoch: 370, iters: 1854, time: 0.064) G_GAN: 0.105 G_GAN_Feat: 0.681 G_ID: 0.088 G_Rec: 0.290 D_GP: 0.033 D_real: 1.019 D_fake: 0.895 +(epoch: 370, iters: 2254, time: 0.064) G_GAN: 0.332 G_GAN_Feat: 0.941 G_ID: 0.124 G_Rec: 0.424 D_GP: 0.031 D_real: 1.013 D_fake: 0.668 +(epoch: 370, iters: 2654, time: 0.064) G_GAN: 0.249 G_GAN_Feat: 0.769 G_ID: 0.091 G_Rec: 0.311 D_GP: 0.037 D_real: 0.960 D_fake: 0.751 +(epoch: 370, iters: 3054, time: 0.064) G_GAN: 0.478 G_GAN_Feat: 1.105 G_ID: 0.135 G_Rec: 0.496 D_GP: 0.095 D_real: 0.658 D_fake: 0.536 +(epoch: 370, iters: 3454, time: 0.064) G_GAN: 0.136 G_GAN_Feat: 0.817 G_ID: 0.105 G_Rec: 0.313 D_GP: 0.050 D_real: 0.792 D_fake: 0.866 +(epoch: 370, iters: 3854, time: 0.064) G_GAN: 0.806 G_GAN_Feat: 1.058 G_ID: 0.119 G_Rec: 0.459 D_GP: 0.067 D_real: 1.067 D_fake: 0.290 +(epoch: 370, iters: 4254, time: 0.064) G_GAN: 0.334 G_GAN_Feat: 0.838 G_ID: 0.098 G_Rec: 0.311 D_GP: 0.041 D_real: 0.946 D_fake: 0.671 +(epoch: 370, iters: 4654, time: 0.064) G_GAN: 0.456 G_GAN_Feat: 1.115 G_ID: 0.161 G_Rec: 0.454 D_GP: 0.138 D_real: 0.322 D_fake: 0.552 +(epoch: 370, iters: 5054, time: 0.064) G_GAN: 0.228 G_GAN_Feat: 0.832 G_ID: 0.101 G_Rec: 0.321 D_GP: 0.053 D_real: 0.942 D_fake: 0.772 +(epoch: 370, iters: 5454, time: 0.064) G_GAN: 1.058 G_GAN_Feat: 1.194 G_ID: 0.128 G_Rec: 0.543 D_GP: 0.156 D_real: 0.601 D_fake: 0.117 +(epoch: 370, iters: 5854, time: 0.064) G_GAN: 0.161 G_GAN_Feat: 0.802 G_ID: 0.087 G_Rec: 0.315 D_GP: 0.069 D_real: 0.930 D_fake: 0.839 +(epoch: 370, iters: 6254, time: 0.064) G_GAN: 0.376 G_GAN_Feat: 1.024 G_ID: 0.124 G_Rec: 0.462 D_GP: 0.034 D_real: 0.878 D_fake: 0.624 +(epoch: 370, iters: 6654, time: 0.064) G_GAN: 0.278 G_GAN_Feat: 0.836 G_ID: 0.100 G_Rec: 0.335 D_GP: 0.037 D_real: 0.878 D_fake: 0.722 +(epoch: 370, iters: 7054, time: 0.064) G_GAN: 0.428 G_GAN_Feat: 1.035 G_ID: 0.119 G_Rec: 0.474 D_GP: 0.032 D_real: 0.988 D_fake: 0.575 +(epoch: 370, iters: 7454, time: 0.064) G_GAN: 0.364 G_GAN_Feat: 0.772 G_ID: 0.123 G_Rec: 0.291 D_GP: 0.032 D_real: 1.138 D_fake: 0.647 +(epoch: 370, iters: 7854, time: 0.064) G_GAN: 0.029 G_GAN_Feat: 1.009 G_ID: 0.118 G_Rec: 0.458 D_GP: 0.047 D_real: 0.539 D_fake: 0.971 +(epoch: 370, iters: 8254, time: 0.064) G_GAN: 0.336 G_GAN_Feat: 0.895 G_ID: 0.117 G_Rec: 0.348 D_GP: 0.136 D_real: 0.632 D_fake: 0.666 +(epoch: 370, iters: 8654, time: 0.064) G_GAN: 0.714 G_GAN_Feat: 1.005 G_ID: 0.126 G_Rec: 0.453 D_GP: 0.036 D_real: 1.389 D_fake: 0.315 +(epoch: 371, iters: 446, time: 0.064) G_GAN: 0.219 G_GAN_Feat: 0.814 G_ID: 0.102 G_Rec: 0.316 D_GP: 0.040 D_real: 0.931 D_fake: 0.786 +(epoch: 371, iters: 846, time: 0.064) G_GAN: 0.809 G_GAN_Feat: 1.174 G_ID: 0.122 G_Rec: 0.483 D_GP: 0.061 D_real: 0.634 D_fake: 0.237 +(epoch: 371, iters: 1246, time: 0.064) G_GAN: 0.239 G_GAN_Feat: 0.960 G_ID: 0.098 G_Rec: 0.377 D_GP: 0.103 D_real: 0.679 D_fake: 0.770 +(epoch: 371, iters: 1646, time: 0.064) G_GAN: 0.855 G_GAN_Feat: 1.034 G_ID: 0.133 G_Rec: 0.466 D_GP: 0.042 D_real: 1.432 D_fake: 0.207 +(epoch: 371, iters: 2046, time: 0.064) G_GAN: -0.022 G_GAN_Feat: 0.752 G_ID: 0.109 G_Rec: 0.363 D_GP: 0.031 D_real: 0.790 D_fake: 1.022 +(epoch: 371, iters: 2446, time: 0.064) G_GAN: 0.374 G_GAN_Feat: 1.056 G_ID: 0.114 G_Rec: 0.432 D_GP: 0.054 D_real: 0.614 D_fake: 0.633 +(epoch: 371, iters: 2846, time: 0.064) G_GAN: 0.084 G_GAN_Feat: 0.725 G_ID: 0.102 G_Rec: 0.315 D_GP: 0.030 D_real: 0.954 D_fake: 0.916 +(epoch: 371, iters: 3246, time: 0.064) G_GAN: 0.567 G_GAN_Feat: 0.952 G_ID: 0.127 G_Rec: 0.417 D_GP: 0.046 D_real: 1.104 D_fake: 0.476 +(epoch: 371, iters: 3646, time: 0.064) G_GAN: -0.083 G_GAN_Feat: 0.861 G_ID: 0.111 G_Rec: 0.324 D_GP: 0.134 D_real: 0.353 D_fake: 1.091 +(epoch: 371, iters: 4046, time: 0.064) G_GAN: 0.446 G_GAN_Feat: 0.953 G_ID: 0.126 G_Rec: 0.449 D_GP: 0.055 D_real: 1.046 D_fake: 0.566 +(epoch: 371, iters: 4446, time: 0.064) G_GAN: 0.029 G_GAN_Feat: 0.836 G_ID: 0.106 G_Rec: 0.347 D_GP: 0.102 D_real: 0.622 D_fake: 0.972 +(epoch: 371, iters: 4846, time: 0.064) G_GAN: -0.088 G_GAN_Feat: 1.010 G_ID: 0.129 G_Rec: 0.433 D_GP: 1.688 D_real: 0.349 D_fake: 1.090 +(epoch: 371, iters: 5246, time: 0.064) G_GAN: 0.228 G_GAN_Feat: 0.629 G_ID: 0.088 G_Rec: 0.290 D_GP: 0.026 D_real: 1.169 D_fake: 0.773 +(epoch: 371, iters: 5646, time: 0.064) G_GAN: 0.459 G_GAN_Feat: 0.897 G_ID: 0.111 G_Rec: 0.454 D_GP: 0.036 D_real: 1.101 D_fake: 0.556 +(epoch: 371, iters: 6046, time: 0.064) G_GAN: 0.088 G_GAN_Feat: 0.681 G_ID: 0.098 G_Rec: 0.279 D_GP: 0.037 D_real: 0.896 D_fake: 0.912 +(epoch: 371, iters: 6446, time: 0.064) G_GAN: 0.361 G_GAN_Feat: 0.937 G_ID: 0.144 G_Rec: 0.439 D_GP: 0.050 D_real: 0.975 D_fake: 0.644 +(epoch: 371, iters: 6846, time: 0.064) G_GAN: -0.067 G_GAN_Feat: 0.694 G_ID: 0.096 G_Rec: 0.305 D_GP: 0.038 D_real: 0.757 D_fake: 1.067 +(epoch: 371, iters: 7246, time: 0.064) G_GAN: 0.551 G_GAN_Feat: 0.986 G_ID: 0.124 G_Rec: 0.453 D_GP: 0.052 D_real: 0.939 D_fake: 0.471 +(epoch: 371, iters: 7646, time: 0.064) G_GAN: 0.336 G_GAN_Feat: 0.776 G_ID: 0.153 G_Rec: 0.360 D_GP: 0.070 D_real: 1.038 D_fake: 0.665 +(epoch: 371, iters: 8046, time: 0.064) G_GAN: 0.683 G_GAN_Feat: 1.005 G_ID: 0.134 G_Rec: 0.481 D_GP: 0.045 D_real: 1.072 D_fake: 0.351 +(epoch: 371, iters: 8446, time: 0.064) G_GAN: 0.527 G_GAN_Feat: 1.036 G_ID: 0.116 G_Rec: 0.381 D_GP: 0.080 D_real: 0.819 D_fake: 0.520 +(epoch: 372, iters: 238, time: 0.064) G_GAN: 0.614 G_GAN_Feat: 0.998 G_ID: 0.118 G_Rec: 0.431 D_GP: 0.047 D_real: 0.910 D_fake: 0.407 +(epoch: 372, iters: 638, time: 0.064) G_GAN: 0.404 G_GAN_Feat: 0.866 G_ID: 0.110 G_Rec: 0.336 D_GP: 0.072 D_real: 0.753 D_fake: 0.612 +(epoch: 372, iters: 1038, time: 0.064) G_GAN: 0.318 G_GAN_Feat: 1.147 G_ID: 0.113 G_Rec: 0.451 D_GP: 0.100 D_real: 0.310 D_fake: 0.693 +(epoch: 372, iters: 1438, time: 0.064) G_GAN: 0.652 G_GAN_Feat: 0.761 G_ID: 0.114 G_Rec: 0.310 D_GP: 0.077 D_real: 1.337 D_fake: 0.546 +(epoch: 372, iters: 1838, time: 0.064) G_GAN: 0.346 G_GAN_Feat: 0.891 G_ID: 0.118 G_Rec: 0.398 D_GP: 0.030 D_real: 1.038 D_fake: 0.666 +(epoch: 372, iters: 2238, time: 0.064) G_GAN: 0.198 G_GAN_Feat: 0.708 G_ID: 0.097 G_Rec: 0.309 D_GP: 0.033 D_real: 1.121 D_fake: 0.802 +(epoch: 372, iters: 2638, time: 0.064) G_GAN: 0.101 G_GAN_Feat: 1.085 G_ID: 0.143 G_Rec: 0.461 D_GP: 0.109 D_real: 0.353 D_fake: 0.907 +(epoch: 372, iters: 3038, time: 0.064) G_GAN: 0.257 G_GAN_Feat: 0.983 G_ID: 0.104 G_Rec: 0.342 D_GP: 0.039 D_real: 1.480 D_fake: 0.773 +(epoch: 372, iters: 3438, time: 0.064) G_GAN: 0.662 G_GAN_Feat: 0.929 G_ID: 0.132 G_Rec: 0.432 D_GP: 0.032 D_real: 1.298 D_fake: 0.389 +(epoch: 372, iters: 3838, time: 0.064) G_GAN: 0.508 G_GAN_Feat: 0.772 G_ID: 0.110 G_Rec: 0.295 D_GP: 0.036 D_real: 1.337 D_fake: 0.513 +(epoch: 372, iters: 4238, time: 0.064) G_GAN: 0.709 G_GAN_Feat: 1.208 G_ID: 0.119 G_Rec: 0.502 D_GP: 0.648 D_real: 0.494 D_fake: 0.361 +(epoch: 372, iters: 4638, time: 0.064) G_GAN: 0.143 G_GAN_Feat: 0.722 G_ID: 0.102 G_Rec: 0.309 D_GP: 0.028 D_real: 1.052 D_fake: 0.858 +(epoch: 372, iters: 5038, time: 0.064) G_GAN: 0.636 G_GAN_Feat: 0.847 G_ID: 0.129 G_Rec: 0.404 D_GP: 0.031 D_real: 1.377 D_fake: 0.371 +(epoch: 372, iters: 5438, time: 0.064) G_GAN: 0.085 G_GAN_Feat: 0.629 G_ID: 0.092 G_Rec: 0.285 D_GP: 0.028 D_real: 1.058 D_fake: 0.915 +(epoch: 372, iters: 5838, time: 0.064) G_GAN: 0.274 G_GAN_Feat: 0.929 G_ID: 0.140 G_Rec: 0.426 D_GP: 0.038 D_real: 0.878 D_fake: 0.727 +(epoch: 372, iters: 6238, time: 0.064) G_GAN: 0.213 G_GAN_Feat: 0.710 G_ID: 0.088 G_Rec: 0.295 D_GP: 0.083 D_real: 0.892 D_fake: 0.787 +(epoch: 372, iters: 6638, time: 0.064) G_GAN: 0.336 G_GAN_Feat: 0.876 G_ID: 0.140 G_Rec: 0.440 D_GP: 0.033 D_real: 1.105 D_fake: 0.665 +(epoch: 372, iters: 7038, time: 0.064) G_GAN: 0.184 G_GAN_Feat: 0.760 G_ID: 0.125 G_Rec: 0.328 D_GP: 0.038 D_real: 0.951 D_fake: 0.817 +(epoch: 372, iters: 7438, time: 0.064) G_GAN: 0.622 G_GAN_Feat: 0.845 G_ID: 0.119 G_Rec: 0.372 D_GP: 0.032 D_real: 1.287 D_fake: 0.414 +(epoch: 372, iters: 7838, time: 0.064) G_GAN: 0.360 G_GAN_Feat: 0.805 G_ID: 0.102 G_Rec: 0.357 D_GP: 0.051 D_real: 1.086 D_fake: 0.642 +(epoch: 372, iters: 8238, time: 0.064) G_GAN: 0.234 G_GAN_Feat: 1.218 G_ID: 0.121 G_Rec: 0.444 D_GP: 0.053 D_real: 1.501 D_fake: 0.799 +(epoch: 372, iters: 8638, time: 0.064) G_GAN: 0.161 G_GAN_Feat: 0.721 G_ID: 0.100 G_Rec: 0.309 D_GP: 0.035 D_real: 1.013 D_fake: 0.840 +(epoch: 373, iters: 430, time: 0.064) G_GAN: 0.292 G_GAN_Feat: 1.126 G_ID: 0.128 G_Rec: 0.490 D_GP: 0.144 D_real: 0.415 D_fake: 0.713 +(epoch: 373, iters: 830, time: 0.064) G_GAN: 0.520 G_GAN_Feat: 0.723 G_ID: 0.101 G_Rec: 0.305 D_GP: 0.035 D_real: 1.402 D_fake: 0.488 +(epoch: 373, iters: 1230, time: 0.064) G_GAN: 0.732 G_GAN_Feat: 1.019 G_ID: 0.113 G_Rec: 0.409 D_GP: 0.061 D_real: 1.046 D_fake: 0.286 +(epoch: 373, iters: 1630, time: 0.064) G_GAN: 0.223 G_GAN_Feat: 0.915 G_ID: 0.110 G_Rec: 0.336 D_GP: 0.117 D_real: 0.319 D_fake: 0.805 +(epoch: 373, iters: 2030, time: 0.064) G_GAN: 0.339 G_GAN_Feat: 1.053 G_ID: 0.138 G_Rec: 0.477 D_GP: 0.062 D_real: 0.581 D_fake: 0.667 +(epoch: 373, iters: 2430, time: 0.064) G_GAN: 0.199 G_GAN_Feat: 0.752 G_ID: 0.103 G_Rec: 0.307 D_GP: 0.029 D_real: 1.061 D_fake: 0.801 +(epoch: 373, iters: 2830, time: 0.064) G_GAN: 0.166 G_GAN_Feat: 1.005 G_ID: 0.123 G_Rec: 0.503 D_GP: 0.039 D_real: 0.821 D_fake: 0.835 +(epoch: 373, iters: 3230, time: 0.064) G_GAN: 0.204 G_GAN_Feat: 0.686 G_ID: 0.099 G_Rec: 0.300 D_GP: 0.032 D_real: 1.114 D_fake: 0.796 +(epoch: 373, iters: 3630, time: 0.064) G_GAN: 0.298 G_GAN_Feat: 0.902 G_ID: 0.113 G_Rec: 0.382 D_GP: 0.035 D_real: 0.956 D_fake: 0.703 +(epoch: 373, iters: 4030, time: 0.064) G_GAN: 0.126 G_GAN_Feat: 0.814 G_ID: 0.105 G_Rec: 0.323 D_GP: 0.062 D_real: 0.898 D_fake: 0.874 +(epoch: 373, iters: 4430, time: 0.064) G_GAN: 0.444 G_GAN_Feat: 1.233 G_ID: 0.123 G_Rec: 0.471 D_GP: 0.054 D_real: 0.320 D_fake: 0.560 +(epoch: 373, iters: 4830, time: 0.064) G_GAN: 0.150 G_GAN_Feat: 0.649 G_ID: 0.107 G_Rec: 0.312 D_GP: 0.029 D_real: 1.108 D_fake: 0.850 +(epoch: 373, iters: 5230, time: 0.064) G_GAN: 0.263 G_GAN_Feat: 0.874 G_ID: 0.137 G_Rec: 0.404 D_GP: 0.032 D_real: 0.981 D_fake: 0.738 +(epoch: 373, iters: 5630, time: 0.064) G_GAN: 0.299 G_GAN_Feat: 0.609 G_ID: 0.100 G_Rec: 0.276 D_GP: 0.029 D_real: 1.197 D_fake: 0.704 +(epoch: 373, iters: 6030, time: 0.064) G_GAN: 0.455 G_GAN_Feat: 0.918 G_ID: 0.120 G_Rec: 0.446 D_GP: 0.035 D_real: 1.105 D_fake: 0.571 +(epoch: 373, iters: 6430, time: 0.064) G_GAN: -0.168 G_GAN_Feat: 0.629 G_ID: 0.102 G_Rec: 0.272 D_GP: 0.033 D_real: 0.731 D_fake: 1.168 +(epoch: 373, iters: 6830, time: 0.064) G_GAN: 0.446 G_GAN_Feat: 0.814 G_ID: 0.106 G_Rec: 0.379 D_GP: 0.033 D_real: 1.166 D_fake: 0.576 +(epoch: 373, iters: 7230, time: 0.064) G_GAN: -0.168 G_GAN_Feat: 0.755 G_ID: 0.131 G_Rec: 0.329 D_GP: 0.065 D_real: 0.546 D_fake: 1.168 +(epoch: 373, iters: 7630, time: 0.064) G_GAN: 0.303 G_GAN_Feat: 0.901 G_ID: 0.118 G_Rec: 0.406 D_GP: 0.048 D_real: 0.821 D_fake: 0.701 +(epoch: 373, iters: 8030, time: 0.064) G_GAN: 0.133 G_GAN_Feat: 0.771 G_ID: 0.107 G_Rec: 0.306 D_GP: 0.032 D_real: 0.885 D_fake: 0.867 +(epoch: 373, iters: 8430, time: 0.064) G_GAN: 0.518 G_GAN_Feat: 0.984 G_ID: 0.123 G_Rec: 0.439 D_GP: 0.062 D_real: 1.066 D_fake: 0.491 +(epoch: 374, iters: 222, time: 0.064) G_GAN: 0.286 G_GAN_Feat: 0.740 G_ID: 0.111 G_Rec: 0.320 D_GP: 0.043 D_real: 1.077 D_fake: 0.714 +(epoch: 374, iters: 622, time: 0.064) G_GAN: 0.744 G_GAN_Feat: 1.133 G_ID: 0.117 G_Rec: 0.498 D_GP: 0.737 D_real: 0.619 D_fake: 0.330 +(epoch: 374, iters: 1022, time: 0.064) G_GAN: 0.229 G_GAN_Feat: 0.687 G_ID: 0.108 G_Rec: 0.291 D_GP: 0.030 D_real: 1.099 D_fake: 0.771 +(epoch: 374, iters: 1422, time: 0.064) G_GAN: 0.219 G_GAN_Feat: 1.022 G_ID: 0.131 G_Rec: 0.402 D_GP: 0.084 D_real: 0.360 D_fake: 0.782 +(epoch: 374, iters: 1822, time: 0.064) G_GAN: 0.222 G_GAN_Feat: 0.789 G_ID: 0.101 G_Rec: 0.302 D_GP: 0.044 D_real: 0.885 D_fake: 0.786 +(epoch: 374, iters: 2222, time: 0.064) G_GAN: 0.620 G_GAN_Feat: 0.970 G_ID: 0.124 G_Rec: 0.402 D_GP: 0.053 D_real: 1.145 D_fake: 0.412 +(epoch: 374, iters: 2622, time: 0.064) G_GAN: 0.174 G_GAN_Feat: 0.860 G_ID: 0.103 G_Rec: 0.351 D_GP: 0.040 D_real: 0.906 D_fake: 0.826 +(epoch: 374, iters: 3022, time: 0.063) G_GAN: 0.945 G_GAN_Feat: 1.029 G_ID: 0.117 G_Rec: 0.449 D_GP: 0.048 D_real: 1.238 D_fake: 0.163 +(epoch: 374, iters: 3422, time: 0.063) G_GAN: -0.120 G_GAN_Feat: 0.982 G_ID: 0.096 G_Rec: 0.351 D_GP: 0.244 D_real: 0.289 D_fake: 1.120 +(epoch: 374, iters: 3822, time: 0.063) G_GAN: 0.394 G_GAN_Feat: 0.964 G_ID: 0.136 G_Rec: 0.417 D_GP: 0.034 D_real: 0.888 D_fake: 0.620 +(epoch: 374, iters: 4222, time: 0.064) G_GAN: 0.366 G_GAN_Feat: 0.796 G_ID: 0.107 G_Rec: 0.331 D_GP: 0.030 D_real: 1.087 D_fake: 0.643 +(epoch: 374, iters: 4622, time: 0.063) G_GAN: 0.762 G_GAN_Feat: 1.119 G_ID: 0.118 G_Rec: 0.426 D_GP: 0.078 D_real: 0.706 D_fake: 0.384 +(epoch: 374, iters: 5022, time: 0.063) G_GAN: -0.127 G_GAN_Feat: 0.835 G_ID: 0.096 G_Rec: 0.288 D_GP: 0.037 D_real: 0.579 D_fake: 1.127 +(epoch: 374, iters: 5422, time: 0.063) G_GAN: 0.432 G_GAN_Feat: 1.009 G_ID: 0.124 G_Rec: 0.458 D_GP: 0.037 D_real: 1.002 D_fake: 0.571 +(epoch: 374, iters: 5822, time: 0.064) G_GAN: -0.164 G_GAN_Feat: 0.827 G_ID: 0.105 G_Rec: 0.333 D_GP: 0.068 D_real: 0.406 D_fake: 1.164 +(epoch: 374, iters: 6222, time: 0.063) G_GAN: 0.672 G_GAN_Feat: 1.031 G_ID: 0.137 G_Rec: 0.454 D_GP: 0.051 D_real: 0.930 D_fake: 0.346 +(epoch: 374, iters: 6622, time: 0.063) G_GAN: 0.926 G_GAN_Feat: 1.004 G_ID: 0.110 G_Rec: 0.358 D_GP: 0.071 D_real: 1.320 D_fake: 0.332 +(epoch: 374, iters: 7022, time: 0.063) G_GAN: 0.550 G_GAN_Feat: 0.982 G_ID: 0.123 G_Rec: 0.463 D_GP: 0.030 D_real: 1.274 D_fake: 0.472 +(epoch: 374, iters: 7422, time: 0.063) G_GAN: -0.049 G_GAN_Feat: 0.708 G_ID: 0.110 G_Rec: 0.305 D_GP: 0.028 D_real: 0.876 D_fake: 1.049 +(epoch: 374, iters: 7822, time: 0.063) G_GAN: 0.423 G_GAN_Feat: 0.901 G_ID: 0.125 G_Rec: 0.455 D_GP: 0.040 D_real: 1.187 D_fake: 0.593 +(epoch: 374, iters: 8222, time: 0.063) G_GAN: 0.021 G_GAN_Feat: 0.693 G_ID: 0.096 G_Rec: 0.305 D_GP: 0.035 D_real: 0.837 D_fake: 0.979 +(epoch: 374, iters: 8622, time: 0.064) G_GAN: -0.011 G_GAN_Feat: 0.821 G_ID: 0.132 G_Rec: 0.401 D_GP: 0.027 D_real: 0.896 D_fake: 1.011 +(epoch: 375, iters: 414, time: 0.063) G_GAN: 0.324 G_GAN_Feat: 0.692 G_ID: 0.093 G_Rec: 0.307 D_GP: 0.026 D_real: 1.257 D_fake: 0.678 +(epoch: 375, iters: 814, time: 0.063) G_GAN: 0.409 G_GAN_Feat: 0.871 G_ID: 0.122 G_Rec: 0.392 D_GP: 0.036 D_real: 1.105 D_fake: 0.596 +(epoch: 375, iters: 1214, time: 0.063) G_GAN: 0.249 G_GAN_Feat: 0.723 G_ID: 0.094 G_Rec: 0.304 D_GP: 0.041 D_real: 1.083 D_fake: 0.751 +(epoch: 375, iters: 1614, time: 0.064) G_GAN: 0.457 G_GAN_Feat: 0.942 G_ID: 0.112 G_Rec: 0.459 D_GP: 0.051 D_real: 1.066 D_fake: 0.567 +(epoch: 375, iters: 2014, time: 0.063) G_GAN: -0.091 G_GAN_Feat: 0.792 G_ID: 0.109 G_Rec: 0.331 D_GP: 0.046 D_real: 0.641 D_fake: 1.091 +(epoch: 375, iters: 2414, time: 0.063) G_GAN: 0.191 G_GAN_Feat: 1.021 G_ID: 0.124 G_Rec: 0.498 D_GP: 0.050 D_real: 0.795 D_fake: 0.810 +(epoch: 375, iters: 2814, time: 0.063) G_GAN: 0.139 G_GAN_Feat: 0.816 G_ID: 0.107 G_Rec: 0.336 D_GP: 0.076 D_real: 0.667 D_fake: 0.861 +(epoch: 375, iters: 3214, time: 0.064) G_GAN: 0.488 G_GAN_Feat: 0.995 G_ID: 0.130 G_Rec: 0.424 D_GP: 0.056 D_real: 0.948 D_fake: 0.518 +(epoch: 375, iters: 3614, time: 0.063) G_GAN: 0.475 G_GAN_Feat: 0.909 G_ID: 0.109 G_Rec: 0.340 D_GP: 0.180 D_real: 0.776 D_fake: 0.570 +(epoch: 375, iters: 4014, time: 0.063) G_GAN: 0.971 G_GAN_Feat: 1.253 G_ID: 0.115 G_Rec: 0.500 D_GP: 0.036 D_real: 1.680 D_fake: 0.171 +(epoch: 375, iters: 4414, time: 0.063) G_GAN: 0.145 G_GAN_Feat: 0.807 G_ID: 0.111 G_Rec: 0.348 D_GP: 0.056 D_real: 0.945 D_fake: 0.856 +(epoch: 375, iters: 4814, time: 0.064) G_GAN: 0.697 G_GAN_Feat: 0.999 G_ID: 0.105 G_Rec: 0.434 D_GP: 0.080 D_real: 1.050 D_fake: 0.343 +(epoch: 375, iters: 5214, time: 0.063) G_GAN: 0.282 G_GAN_Feat: 0.814 G_ID: 0.102 G_Rec: 0.308 D_GP: 0.057 D_real: 0.991 D_fake: 0.752 +(epoch: 375, iters: 5614, time: 0.063) G_GAN: 0.684 G_GAN_Feat: 1.123 G_ID: 0.131 G_Rec: 0.425 D_GP: 0.052 D_real: 0.537 D_fake: 0.348 +(epoch: 375, iters: 6014, time: 0.063) G_GAN: 0.226 G_GAN_Feat: 0.936 G_ID: 0.100 G_Rec: 0.326 D_GP: 0.167 D_real: 0.544 D_fake: 0.774 +(epoch: 375, iters: 6414, time: 0.064) G_GAN: 0.788 G_GAN_Feat: 1.075 G_ID: 0.128 G_Rec: 0.433 D_GP: 0.056 D_real: 0.720 D_fake: 0.253 +(epoch: 375, iters: 6814, time: 0.063) G_GAN: 0.492 G_GAN_Feat: 0.934 G_ID: 0.102 G_Rec: 0.360 D_GP: 0.055 D_real: 0.893 D_fake: 0.527 +(epoch: 375, iters: 7214, time: 0.063) G_GAN: 0.422 G_GAN_Feat: 0.842 G_ID: 0.124 G_Rec: 0.404 D_GP: 0.034 D_real: 1.109 D_fake: 0.582 +(epoch: 375, iters: 7614, time: 0.063) G_GAN: 0.142 G_GAN_Feat: 0.662 G_ID: 0.101 G_Rec: 0.285 D_GP: 0.032 D_real: 1.007 D_fake: 0.858 +(epoch: 375, iters: 8014, time: 0.064) G_GAN: 0.328 G_GAN_Feat: 1.007 G_ID: 0.127 G_Rec: 0.471 D_GP: 0.048 D_real: 0.739 D_fake: 0.679 +(epoch: 375, iters: 8414, time: 0.063) G_GAN: 0.096 G_GAN_Feat: 0.854 G_ID: 0.111 G_Rec: 0.384 D_GP: 0.086 D_real: 0.783 D_fake: 0.905 +(epoch: 376, iters: 206, time: 0.063) G_GAN: 0.524 G_GAN_Feat: 0.903 G_ID: 0.120 G_Rec: 0.429 D_GP: 0.031 D_real: 1.184 D_fake: 0.482 +(epoch: 376, iters: 606, time: 0.063) G_GAN: 0.147 G_GAN_Feat: 0.882 G_ID: 0.107 G_Rec: 0.360 D_GP: 0.347 D_real: 0.499 D_fake: 0.868 +(epoch: 376, iters: 1006, time: 0.064) G_GAN: 0.422 G_GAN_Feat: 0.964 G_ID: 0.125 G_Rec: 0.435 D_GP: 0.058 D_real: 0.949 D_fake: 0.581 +(epoch: 376, iters: 1406, time: 0.063) G_GAN: 0.299 G_GAN_Feat: 0.852 G_ID: 0.102 G_Rec: 0.361 D_GP: 0.097 D_real: 0.882 D_fake: 0.707 +(epoch: 376, iters: 1806, time: 0.063) G_GAN: 0.505 G_GAN_Feat: 1.016 G_ID: 0.176 G_Rec: 0.493 D_GP: 0.051 D_real: 1.059 D_fake: 0.498 +(epoch: 376, iters: 2206, time: 0.063) G_GAN: 0.355 G_GAN_Feat: 0.734 G_ID: 0.095 G_Rec: 0.314 D_GP: 0.033 D_real: 1.184 D_fake: 0.649 +(epoch: 376, iters: 2606, time: 0.064) G_GAN: 0.640 G_GAN_Feat: 1.011 G_ID: 0.121 G_Rec: 0.437 D_GP: 0.039 D_real: 1.019 D_fake: 0.393 +(epoch: 376, iters: 3006, time: 0.064) G_GAN: 0.644 G_GAN_Feat: 0.768 G_ID: 0.097 G_Rec: 0.332 D_GP: 0.036 D_real: 1.551 D_fake: 0.370 +(epoch: 376, iters: 3406, time: 0.064) G_GAN: 0.758 G_GAN_Feat: 0.934 G_ID: 0.108 G_Rec: 0.391 D_GP: 0.039 D_real: 1.409 D_fake: 0.277 +(epoch: 376, iters: 3806, time: 0.064) G_GAN: 0.426 G_GAN_Feat: 0.741 G_ID: 0.116 G_Rec: 0.272 D_GP: 0.038 D_real: 1.223 D_fake: 0.607 +(epoch: 376, iters: 4206, time: 0.064) G_GAN: 0.298 G_GAN_Feat: 1.187 G_ID: 0.134 G_Rec: 0.424 D_GP: 0.035 D_real: 1.157 D_fake: 0.710 +(epoch: 376, iters: 4606, time: 0.064) G_GAN: 0.087 G_GAN_Feat: 0.617 G_ID: 0.108 G_Rec: 0.271 D_GP: 0.028 D_real: 1.108 D_fake: 0.913 +(epoch: 376, iters: 5006, time: 0.064) G_GAN: 0.362 G_GAN_Feat: 0.827 G_ID: 0.114 G_Rec: 0.462 D_GP: 0.030 D_real: 1.082 D_fake: 0.651 +(epoch: 376, iters: 5406, time: 0.064) G_GAN: 0.180 G_GAN_Feat: 0.678 G_ID: 0.097 G_Rec: 0.314 D_GP: 0.038 D_real: 1.061 D_fake: 0.820 +(epoch: 376, iters: 5806, time: 0.064) G_GAN: 0.289 G_GAN_Feat: 0.969 G_ID: 0.121 G_Rec: 0.482 D_GP: 0.055 D_real: 0.796 D_fake: 0.715 +(epoch: 376, iters: 6206, time: 0.064) G_GAN: 0.116 G_GAN_Feat: 0.816 G_ID: 0.108 G_Rec: 0.360 D_GP: 0.114 D_real: 0.726 D_fake: 0.885 +(epoch: 376, iters: 6606, time: 0.064) G_GAN: 0.562 G_GAN_Feat: 0.985 G_ID: 0.120 G_Rec: 0.416 D_GP: 0.034 D_real: 1.028 D_fake: 0.453 +(epoch: 376, iters: 7006, time: 0.064) G_GAN: 0.326 G_GAN_Feat: 0.686 G_ID: 0.086 G_Rec: 0.296 D_GP: 0.035 D_real: 1.220 D_fake: 0.674 +(epoch: 376, iters: 7406, time: 0.064) G_GAN: 0.472 G_GAN_Feat: 0.991 G_ID: 0.120 G_Rec: 0.453 D_GP: 0.063 D_real: 0.844 D_fake: 0.534 +(epoch: 376, iters: 7806, time: 0.064) G_GAN: 0.219 G_GAN_Feat: 0.782 G_ID: 0.103 G_Rec: 0.313 D_GP: 0.043 D_real: 0.961 D_fake: 0.781 +(epoch: 376, iters: 8206, time: 0.064) G_GAN: 0.507 G_GAN_Feat: 0.937 G_ID: 0.128 G_Rec: 0.410 D_GP: 0.035 D_real: 1.150 D_fake: 0.498 +(epoch: 376, iters: 8606, time: 0.064) G_GAN: 0.167 G_GAN_Feat: 0.761 G_ID: 0.087 G_Rec: 0.315 D_GP: 0.033 D_real: 1.050 D_fake: 0.833 +(epoch: 377, iters: 398, time: 0.064) G_GAN: 0.580 G_GAN_Feat: 1.013 G_ID: 0.119 G_Rec: 0.452 D_GP: 0.036 D_real: 1.035 D_fake: 0.431 +(epoch: 377, iters: 798, time: 0.063) G_GAN: -0.068 G_GAN_Feat: 0.941 G_ID: 0.113 G_Rec: 0.337 D_GP: 0.578 D_real: 0.206 D_fake: 1.079 +(epoch: 377, iters: 1198, time: 0.063) G_GAN: 0.450 G_GAN_Feat: 1.001 G_ID: 0.109 G_Rec: 0.428 D_GP: 0.049 D_real: 0.853 D_fake: 0.561 +(epoch: 377, iters: 1598, time: 0.063) G_GAN: 0.109 G_GAN_Feat: 0.707 G_ID: 0.119 G_Rec: 0.361 D_GP: 0.029 D_real: 0.924 D_fake: 0.891 +(epoch: 377, iters: 1998, time: 0.064) G_GAN: 0.525 G_GAN_Feat: 1.086 G_ID: 0.120 G_Rec: 0.450 D_GP: 0.042 D_real: 0.861 D_fake: 0.493 +(epoch: 377, iters: 2398, time: 0.063) G_GAN: 0.332 G_GAN_Feat: 1.021 G_ID: 0.093 G_Rec: 0.349 D_GP: 0.107 D_real: 0.479 D_fake: 0.677 +(epoch: 377, iters: 2798, time: 0.063) G_GAN: 0.529 G_GAN_Feat: 0.967 G_ID: 0.133 G_Rec: 0.441 D_GP: 0.034 D_real: 1.093 D_fake: 0.488 +(epoch: 377, iters: 3198, time: 0.063) G_GAN: 0.109 G_GAN_Feat: 0.697 G_ID: 0.122 G_Rec: 0.313 D_GP: 0.038 D_real: 0.996 D_fake: 0.891 +(epoch: 377, iters: 3598, time: 0.064) G_GAN: 0.346 G_GAN_Feat: 0.931 G_ID: 0.111 G_Rec: 0.429 D_GP: 0.033 D_real: 0.976 D_fake: 0.665 +(epoch: 377, iters: 3998, time: 0.063) G_GAN: 0.143 G_GAN_Feat: 0.666 G_ID: 0.102 G_Rec: 0.288 D_GP: 0.042 D_real: 0.971 D_fake: 0.857 +(epoch: 377, iters: 4398, time: 0.063) G_GAN: 0.231 G_GAN_Feat: 0.977 G_ID: 0.142 G_Rec: 0.457 D_GP: 0.040 D_real: 1.041 D_fake: 0.776 +(epoch: 377, iters: 4798, time: 0.063) G_GAN: 0.159 G_GAN_Feat: 0.723 G_ID: 0.090 G_Rec: 0.305 D_GP: 0.031 D_real: 1.039 D_fake: 0.841 +(epoch: 377, iters: 5198, time: 0.064) G_GAN: 0.423 G_GAN_Feat: 0.953 G_ID: 0.130 G_Rec: 0.425 D_GP: 0.063 D_real: 0.880 D_fake: 0.592 +(epoch: 377, iters: 5598, time: 0.063) G_GAN: 0.118 G_GAN_Feat: 0.852 G_ID: 0.107 G_Rec: 0.325 D_GP: 0.059 D_real: 0.560 D_fake: 0.882 +(epoch: 377, iters: 5998, time: 0.063) G_GAN: 0.316 G_GAN_Feat: 1.014 G_ID: 0.109 G_Rec: 0.480 D_GP: 0.047 D_real: 0.888 D_fake: 0.685 +(epoch: 377, iters: 6398, time: 0.063) G_GAN: 0.209 G_GAN_Feat: 0.857 G_ID: 0.103 G_Rec: 0.330 D_GP: 0.037 D_real: 1.006 D_fake: 0.791 +(epoch: 377, iters: 6798, time: 0.064) G_GAN: 0.996 G_GAN_Feat: 1.008 G_ID: 0.120 G_Rec: 0.434 D_GP: 0.035 D_real: 1.554 D_fake: 0.115 +(epoch: 377, iters: 7198, time: 0.063) G_GAN: 0.458 G_GAN_Feat: 0.798 G_ID: 0.096 G_Rec: 0.319 D_GP: 0.055 D_real: 1.445 D_fake: 0.544 +(epoch: 377, iters: 7598, time: 0.063) G_GAN: 0.283 G_GAN_Feat: 0.995 G_ID: 0.148 G_Rec: 0.442 D_GP: 0.033 D_real: 0.901 D_fake: 0.719 +(epoch: 377, iters: 7998, time: 0.063) G_GAN: -0.124 G_GAN_Feat: 0.867 G_ID: 0.107 G_Rec: 0.321 D_GP: 0.048 D_real: 0.538 D_fake: 1.124 +(epoch: 377, iters: 8398, time: 0.064) G_GAN: 0.679 G_GAN_Feat: 1.026 G_ID: 0.119 G_Rec: 0.440 D_GP: 0.052 D_real: 1.110 D_fake: 0.354 +(epoch: 378, iters: 190, time: 0.063) G_GAN: 0.271 G_GAN_Feat: 0.804 G_ID: 0.114 G_Rec: 0.310 D_GP: 0.041 D_real: 1.022 D_fake: 0.731 +(epoch: 378, iters: 590, time: 0.063) G_GAN: 0.509 G_GAN_Feat: 0.922 G_ID: 0.115 G_Rec: 0.456 D_GP: 0.026 D_real: 1.230 D_fake: 0.497 +(epoch: 378, iters: 990, time: 0.063) G_GAN: 0.104 G_GAN_Feat: 0.793 G_ID: 0.107 G_Rec: 0.335 D_GP: 0.042 D_real: 0.755 D_fake: 0.896 +(epoch: 378, iters: 1390, time: 0.064) G_GAN: 0.678 G_GAN_Feat: 1.002 G_ID: 0.126 G_Rec: 0.440 D_GP: 0.036 D_real: 0.979 D_fake: 0.346 +(epoch: 378, iters: 1790, time: 0.063) G_GAN: 0.418 G_GAN_Feat: 1.007 G_ID: 0.114 G_Rec: 0.356 D_GP: 0.416 D_real: 0.314 D_fake: 0.712 +(epoch: 378, iters: 2190, time: 0.063) G_GAN: 0.416 G_GAN_Feat: 0.858 G_ID: 0.130 G_Rec: 0.392 D_GP: 0.036 D_real: 1.179 D_fake: 0.589 +(epoch: 378, iters: 2590, time: 0.063) G_GAN: 0.108 G_GAN_Feat: 0.692 G_ID: 0.102 G_Rec: 0.293 D_GP: 0.028 D_real: 1.064 D_fake: 0.893 +(epoch: 378, iters: 2990, time: 0.064) G_GAN: 0.241 G_GAN_Feat: 1.003 G_ID: 0.135 G_Rec: 0.453 D_GP: 0.040 D_real: 0.892 D_fake: 0.764 +(epoch: 378, iters: 3390, time: 0.063) G_GAN: 0.133 G_GAN_Feat: 0.701 G_ID: 0.105 G_Rec: 0.296 D_GP: 0.040 D_real: 1.023 D_fake: 0.870 +(epoch: 378, iters: 3790, time: 0.063) G_GAN: 0.125 G_GAN_Feat: 1.040 G_ID: 0.135 G_Rec: 0.462 D_GP: 0.071 D_real: 0.438 D_fake: 0.875 +(epoch: 378, iters: 4190, time: 0.064) G_GAN: 0.154 G_GAN_Feat: 0.756 G_ID: 0.110 G_Rec: 0.333 D_GP: 0.034 D_real: 0.988 D_fake: 0.847 +(epoch: 378, iters: 4590, time: 0.064) G_GAN: 0.538 G_GAN_Feat: 0.985 G_ID: 0.122 G_Rec: 0.484 D_GP: 0.042 D_real: 1.262 D_fake: 0.509 +(epoch: 378, iters: 4990, time: 0.063) G_GAN: 0.042 G_GAN_Feat: 0.886 G_ID: 0.110 G_Rec: 0.335 D_GP: 0.052 D_real: 0.405 D_fake: 0.958 +(epoch: 378, iters: 5390, time: 0.063) G_GAN: 0.713 G_GAN_Feat: 0.999 G_ID: 0.124 G_Rec: 0.424 D_GP: 0.037 D_real: 1.180 D_fake: 0.309 +(epoch: 378, iters: 5790, time: 0.063) G_GAN: 0.237 G_GAN_Feat: 0.836 G_ID: 0.100 G_Rec: 0.341 D_GP: 0.034 D_real: 1.046 D_fake: 0.769 +(epoch: 378, iters: 6190, time: 0.064) G_GAN: 0.334 G_GAN_Feat: 0.900 G_ID: 0.129 G_Rec: 0.423 D_GP: 0.028 D_real: 1.110 D_fake: 0.669 +(epoch: 378, iters: 6590, time: 0.063) G_GAN: 0.119 G_GAN_Feat: 0.726 G_ID: 0.100 G_Rec: 0.293 D_GP: 0.031 D_real: 0.944 D_fake: 0.881 +(epoch: 378, iters: 6990, time: 0.063) G_GAN: 0.388 G_GAN_Feat: 1.031 G_ID: 0.140 G_Rec: 0.496 D_GP: 0.095 D_real: 0.838 D_fake: 0.624 +(epoch: 378, iters: 7390, time: 0.063) G_GAN: 0.069 G_GAN_Feat: 0.747 G_ID: 0.098 G_Rec: 0.308 D_GP: 0.040 D_real: 0.933 D_fake: 0.932 +(epoch: 378, iters: 7790, time: 0.064) G_GAN: 0.494 G_GAN_Feat: 0.926 G_ID: 0.121 G_Rec: 0.415 D_GP: 0.038 D_real: 1.042 D_fake: 0.527 +(epoch: 378, iters: 8190, time: 0.063) G_GAN: -0.035 G_GAN_Feat: 0.767 G_ID: 0.095 G_Rec: 0.326 D_GP: 0.047 D_real: 0.728 D_fake: 1.035 +(epoch: 378, iters: 8590, time: 0.063) G_GAN: 0.530 G_GAN_Feat: 1.029 G_ID: 0.121 G_Rec: 0.456 D_GP: 0.045 D_real: 0.833 D_fake: 0.489 +(epoch: 379, iters: 382, time: 0.063) G_GAN: 0.220 G_GAN_Feat: 0.673 G_ID: 0.105 G_Rec: 0.269 D_GP: 0.034 D_real: 1.092 D_fake: 0.780 +(epoch: 379, iters: 782, time: 0.064) G_GAN: 0.665 G_GAN_Feat: 0.956 G_ID: 0.112 G_Rec: 0.494 D_GP: 0.037 D_real: 1.285 D_fake: 0.352 +(epoch: 379, iters: 1182, time: 0.063) G_GAN: 0.479 G_GAN_Feat: 0.869 G_ID: 0.079 G_Rec: 0.308 D_GP: 0.124 D_real: 0.730 D_fake: 0.526 +(epoch: 379, iters: 1582, time: 0.063) G_GAN: 0.491 G_GAN_Feat: 1.218 G_ID: 0.138 G_Rec: 0.557 D_GP: 0.043 D_real: 0.987 D_fake: 0.541 +(epoch: 379, iters: 1982, time: 0.063) G_GAN: -0.050 G_GAN_Feat: 0.709 G_ID: 0.110 G_Rec: 0.305 D_GP: 0.030 D_real: 0.824 D_fake: 1.050 +(epoch: 379, iters: 2382, time: 0.064) G_GAN: 0.463 G_GAN_Feat: 1.035 G_ID: 0.140 G_Rec: 0.467 D_GP: 0.120 D_real: 0.530 D_fake: 0.548 +(epoch: 379, iters: 2782, time: 0.063) G_GAN: 0.432 G_GAN_Feat: 0.827 G_ID: 0.089 G_Rec: 0.315 D_GP: 0.059 D_real: 0.933 D_fake: 0.573 +(epoch: 379, iters: 3182, time: 0.063) G_GAN: 0.682 G_GAN_Feat: 0.995 G_ID: 0.132 G_Rec: 0.414 D_GP: 0.047 D_real: 1.255 D_fake: 0.343 +(epoch: 379, iters: 3582, time: 0.063) G_GAN: 0.102 G_GAN_Feat: 0.805 G_ID: 0.109 G_Rec: 0.329 D_GP: 0.062 D_real: 0.748 D_fake: 0.898 +(epoch: 379, iters: 3982, time: 0.063) G_GAN: 1.118 G_GAN_Feat: 0.951 G_ID: 0.158 G_Rec: 0.421 D_GP: 0.035 D_real: 1.666 D_fake: 0.106 +(epoch: 379, iters: 4382, time: 0.063) G_GAN: 0.336 G_GAN_Feat: 0.734 G_ID: 0.100 G_Rec: 0.292 D_GP: 0.035 D_real: 1.114 D_fake: 0.667 +(epoch: 379, iters: 4782, time: 0.063) G_GAN: 0.678 G_GAN_Feat: 1.050 G_ID: 0.124 G_Rec: 0.495 D_GP: 0.036 D_real: 1.211 D_fake: 0.367 +(epoch: 379, iters: 5182, time: 0.064) G_GAN: 0.071 G_GAN_Feat: 0.879 G_ID: 0.106 G_Rec: 0.342 D_GP: 0.116 D_real: 0.667 D_fake: 0.929 +(epoch: 379, iters: 5582, time: 0.063) G_GAN: 0.606 G_GAN_Feat: 1.127 G_ID: 0.114 G_Rec: 0.473 D_GP: 0.079 D_real: 0.714 D_fake: 0.409 +(epoch: 379, iters: 5982, time: 0.063) G_GAN: -0.109 G_GAN_Feat: 0.785 G_ID: 0.128 G_Rec: 0.347 D_GP: 0.035 D_real: 0.797 D_fake: 1.109 +(epoch: 379, iters: 6382, time: 0.063) G_GAN: 0.673 G_GAN_Feat: 0.990 G_ID: 0.106 G_Rec: 0.440 D_GP: 0.033 D_real: 1.273 D_fake: 0.359 +(epoch: 379, iters: 6782, time: 0.064) G_GAN: 0.234 G_GAN_Feat: 0.791 G_ID: 0.109 G_Rec: 0.340 D_GP: 0.037 D_real: 0.899 D_fake: 0.767 +(epoch: 379, iters: 7182, time: 0.063) G_GAN: 0.559 G_GAN_Feat: 0.873 G_ID: 0.128 G_Rec: 0.395 D_GP: 0.029 D_real: 1.240 D_fake: 0.456 +(epoch: 379, iters: 7582, time: 0.063) G_GAN: 0.158 G_GAN_Feat: 0.806 G_ID: 0.125 G_Rec: 0.340 D_GP: 0.036 D_real: 0.762 D_fake: 0.843 +(epoch: 379, iters: 7982, time: 0.063) G_GAN: 0.330 G_GAN_Feat: 0.906 G_ID: 0.123 G_Rec: 0.414 D_GP: 0.034 D_real: 1.036 D_fake: 0.673 +(epoch: 379, iters: 8382, time: 0.064) G_GAN: -0.235 G_GAN_Feat: 0.697 G_ID: 0.100 G_Rec: 0.290 D_GP: 0.033 D_real: 0.648 D_fake: 1.235 +(epoch: 380, iters: 174, time: 0.063) G_GAN: 0.348 G_GAN_Feat: 1.080 G_ID: 0.118 G_Rec: 0.491 D_GP: 0.050 D_real: 0.647 D_fake: 0.668 +(epoch: 380, iters: 574, time: 0.063) G_GAN: 0.077 G_GAN_Feat: 0.714 G_ID: 0.098 G_Rec: 0.299 D_GP: 0.043 D_real: 0.873 D_fake: 0.923 +(epoch: 380, iters: 974, time: 0.063) G_GAN: 0.488 G_GAN_Feat: 1.011 G_ID: 0.109 G_Rec: 0.450 D_GP: 0.037 D_real: 1.028 D_fake: 0.517 +(epoch: 380, iters: 1374, time: 0.064) G_GAN: 0.339 G_GAN_Feat: 0.790 G_ID: 0.098 G_Rec: 0.307 D_GP: 0.034 D_real: 1.179 D_fake: 0.662 +(epoch: 380, iters: 1774, time: 0.063) G_GAN: 0.900 G_GAN_Feat: 1.237 G_ID: 0.123 G_Rec: 0.463 D_GP: 0.072 D_real: 1.352 D_fake: 0.254 +(epoch: 380, iters: 2174, time: 0.063) G_GAN: 0.166 G_GAN_Feat: 0.740 G_ID: 0.113 G_Rec: 0.299 D_GP: 0.029 D_real: 1.000 D_fake: 0.835 +(epoch: 380, iters: 2574, time: 0.063) G_GAN: 0.189 G_GAN_Feat: 1.083 G_ID: 0.128 G_Rec: 0.431 D_GP: 0.049 D_real: 0.467 D_fake: 0.811 +(epoch: 380, iters: 2974, time: 0.064) G_GAN: 0.107 G_GAN_Feat: 0.804 G_ID: 0.098 G_Rec: 0.312 D_GP: 0.034 D_real: 0.809 D_fake: 0.893 +(epoch: 380, iters: 3374, time: 0.063) G_GAN: 0.732 G_GAN_Feat: 1.098 G_ID: 0.116 G_Rec: 0.452 D_GP: 0.054 D_real: 1.093 D_fake: 0.308 +(epoch: 380, iters: 3774, time: 0.063) G_GAN: 0.307 G_GAN_Feat: 0.907 G_ID: 0.094 G_Rec: 0.311 D_GP: 0.057 D_real: 0.625 D_fake: 0.695 +(epoch: 380, iters: 4174, time: 0.063) G_GAN: 0.562 G_GAN_Feat: 0.845 G_ID: 0.112 G_Rec: 0.401 D_GP: 0.032 D_real: 1.270 D_fake: 0.459 +(epoch: 380, iters: 4574, time: 0.064) G_GAN: 0.339 G_GAN_Feat: 0.729 G_ID: 0.092 G_Rec: 0.325 D_GP: 0.038 D_real: 1.182 D_fake: 0.669 +(epoch: 380, iters: 4974, time: 0.063) G_GAN: 0.533 G_GAN_Feat: 0.948 G_ID: 0.127 G_Rec: 0.440 D_GP: 0.039 D_real: 1.158 D_fake: 0.497 +(epoch: 380, iters: 5374, time: 0.063) G_GAN: 0.262 G_GAN_Feat: 0.797 G_ID: 0.105 G_Rec: 0.336 D_GP: 0.082 D_real: 0.832 D_fake: 0.743 +(epoch: 380, iters: 5774, time: 0.063) G_GAN: 0.558 G_GAN_Feat: 0.912 G_ID: 0.116 G_Rec: 0.417 D_GP: 0.033 D_real: 1.073 D_fake: 0.450 +(epoch: 380, iters: 6174, time: 0.064) G_GAN: 0.143 G_GAN_Feat: 0.848 G_ID: 0.117 G_Rec: 0.323 D_GP: 0.046 D_real: 0.883 D_fake: 0.858 +(epoch: 380, iters: 6574, time: 0.063) G_GAN: 0.431 G_GAN_Feat: 1.079 G_ID: 0.109 G_Rec: 0.453 D_GP: 0.044 D_real: 0.637 D_fake: 0.572 +(epoch: 380, iters: 6974, time: 0.063) G_GAN: -0.222 G_GAN_Feat: 0.888 G_ID: 0.114 G_Rec: 0.352 D_GP: 0.051 D_real: 0.412 D_fake: 1.222 +(epoch: 380, iters: 7374, time: 0.063) G_GAN: 0.697 G_GAN_Feat: 1.016 G_ID: 0.120 G_Rec: 0.476 D_GP: 0.035 D_real: 1.323 D_fake: 0.325 +(epoch: 380, iters: 7774, time: 0.064) G_GAN: 0.181 G_GAN_Feat: 0.734 G_ID: 0.099 G_Rec: 0.318 D_GP: 0.041 D_real: 1.026 D_fake: 0.819 +(epoch: 380, iters: 8174, time: 0.063) G_GAN: 0.499 G_GAN_Feat: 1.039 G_ID: 0.123 G_Rec: 0.477 D_GP: 0.062 D_real: 0.778 D_fake: 0.511 +(epoch: 380, iters: 8574, time: 0.063) G_GAN: -0.057 G_GAN_Feat: 0.987 G_ID: 0.123 G_Rec: 0.349 D_GP: 0.219 D_real: 0.392 D_fake: 1.060 +(epoch: 381, iters: 366, time: 0.063) G_GAN: 0.834 G_GAN_Feat: 1.283 G_ID: 0.130 G_Rec: 0.520 D_GP: 0.636 D_real: 0.404 D_fake: 0.315 +(epoch: 381, iters: 766, time: 0.064) G_GAN: 0.244 G_GAN_Feat: 0.849 G_ID: 0.127 G_Rec: 0.300 D_GP: 0.037 D_real: 0.891 D_fake: 0.764 +(epoch: 381, iters: 1166, time: 0.063) G_GAN: 0.305 G_GAN_Feat: 0.927 G_ID: 0.120 G_Rec: 0.481 D_GP: 0.037 D_real: 1.025 D_fake: 0.699 +(epoch: 381, iters: 1566, time: 0.063) G_GAN: 0.275 G_GAN_Feat: 0.756 G_ID: 0.103 G_Rec: 0.331 D_GP: 0.036 D_real: 1.124 D_fake: 0.725 +(epoch: 381, iters: 1966, time: 0.063) G_GAN: 0.739 G_GAN_Feat: 0.943 G_ID: 0.122 G_Rec: 0.440 D_GP: 0.044 D_real: 1.320 D_fake: 0.386 +(epoch: 381, iters: 2366, time: 0.064) G_GAN: -0.055 G_GAN_Feat: 0.799 G_ID: 0.109 G_Rec: 0.354 D_GP: 0.097 D_real: 0.610 D_fake: 1.056 +(epoch: 381, iters: 2766, time: 0.063) G_GAN: 0.211 G_GAN_Feat: 0.932 G_ID: 0.121 G_Rec: 0.438 D_GP: 0.056 D_real: 0.735 D_fake: 0.790 +(epoch: 381, iters: 3166, time: 0.063) G_GAN: 0.306 G_GAN_Feat: 0.739 G_ID: 0.106 G_Rec: 0.318 D_GP: 0.037 D_real: 1.220 D_fake: 0.696 +(epoch: 381, iters: 3566, time: 0.063) G_GAN: 0.494 G_GAN_Feat: 1.105 G_ID: 0.117 G_Rec: 0.486 D_GP: 0.093 D_real: 0.469 D_fake: 0.507 +(epoch: 381, iters: 3966, time: 0.064) G_GAN: 0.175 G_GAN_Feat: 0.909 G_ID: 0.107 G_Rec: 0.328 D_GP: 0.293 D_real: 0.483 D_fake: 0.838 +(epoch: 381, iters: 4366, time: 0.063) G_GAN: 0.532 G_GAN_Feat: 1.052 G_ID: 0.130 G_Rec: 0.412 D_GP: 0.042 D_real: 0.846 D_fake: 0.477 +(epoch: 381, iters: 4766, time: 0.063) G_GAN: 0.082 G_GAN_Feat: 0.791 G_ID: 0.121 G_Rec: 0.329 D_GP: 0.034 D_real: 0.911 D_fake: 0.919 +(epoch: 381, iters: 5166, time: 0.063) G_GAN: 0.472 G_GAN_Feat: 1.171 G_ID: 0.138 G_Rec: 0.478 D_GP: 0.153 D_real: 0.294 D_fake: 0.536 +(epoch: 381, iters: 5566, time: 0.064) G_GAN: 0.244 G_GAN_Feat: 0.737 G_ID: 0.094 G_Rec: 0.312 D_GP: 0.036 D_real: 1.039 D_fake: 0.757 +(epoch: 381, iters: 5966, time: 0.063) G_GAN: 0.540 G_GAN_Feat: 1.054 G_ID: 0.129 G_Rec: 0.432 D_GP: 0.132 D_real: 0.610 D_fake: 0.471 +(epoch: 381, iters: 6366, time: 0.063) G_GAN: -0.027 G_GAN_Feat: 0.871 G_ID: 0.103 G_Rec: 0.319 D_GP: 0.036 D_real: 0.352 D_fake: 1.027 +(epoch: 381, iters: 6766, time: 0.063) G_GAN: 0.611 G_GAN_Feat: 1.153 G_ID: 0.126 G_Rec: 0.501 D_GP: 0.375 D_real: 0.829 D_fake: 0.518 +(epoch: 381, iters: 7166, time: 0.064) G_GAN: 0.363 G_GAN_Feat: 0.836 G_ID: 0.119 G_Rec: 0.318 D_GP: 0.035 D_real: 0.982 D_fake: 0.637 +(epoch: 381, iters: 7566, time: 0.063) G_GAN: 0.196 G_GAN_Feat: 0.985 G_ID: 0.128 G_Rec: 0.457 D_GP: 0.033 D_real: 0.862 D_fake: 0.804 +(epoch: 381, iters: 7966, time: 0.063) G_GAN: 0.314 G_GAN_Feat: 0.958 G_ID: 0.095 G_Rec: 0.354 D_GP: 0.125 D_real: 0.595 D_fake: 0.691 +(epoch: 381, iters: 8366, time: 0.063) G_GAN: 0.580 G_GAN_Feat: 0.867 G_ID: 0.112 G_Rec: 0.436 D_GP: 0.029 D_real: 1.406 D_fake: 0.429 +(epoch: 382, iters: 158, time: 0.064) G_GAN: 0.238 G_GAN_Feat: 0.707 G_ID: 0.093 G_Rec: 0.321 D_GP: 0.033 D_real: 1.192 D_fake: 0.763 +(epoch: 382, iters: 558, time: 0.063) G_GAN: 0.439 G_GAN_Feat: 0.937 G_ID: 0.122 G_Rec: 0.447 D_GP: 0.031 D_real: 1.038 D_fake: 0.563 +(epoch: 382, iters: 958, time: 0.063) G_GAN: -0.036 G_GAN_Feat: 0.679 G_ID: 0.090 G_Rec: 0.303 D_GP: 0.031 D_real: 0.880 D_fake: 1.036 +(epoch: 382, iters: 1358, time: 0.063) G_GAN: 0.281 G_GAN_Feat: 0.899 G_ID: 0.144 G_Rec: 0.402 D_GP: 0.035 D_real: 0.967 D_fake: 0.719 +(epoch: 382, iters: 1758, time: 0.064) G_GAN: 0.142 G_GAN_Feat: 0.821 G_ID: 0.112 G_Rec: 0.349 D_GP: 0.029 D_real: 1.019 D_fake: 0.858 +(epoch: 382, iters: 2158, time: 0.063) G_GAN: 0.208 G_GAN_Feat: 0.903 G_ID: 0.129 G_Rec: 0.423 D_GP: 0.047 D_real: 0.767 D_fake: 0.794 +(epoch: 382, iters: 2558, time: 0.063) G_GAN: 0.063 G_GAN_Feat: 0.738 G_ID: 0.110 G_Rec: 0.306 D_GP: 0.033 D_real: 0.894 D_fake: 0.937 +(epoch: 382, iters: 2958, time: 0.063) G_GAN: 0.350 G_GAN_Feat: 1.111 G_ID: 0.133 G_Rec: 0.444 D_GP: 0.104 D_real: 0.349 D_fake: 0.655 +(epoch: 382, iters: 3358, time: 0.064) G_GAN: 0.090 G_GAN_Feat: 0.693 G_ID: 0.130 G_Rec: 0.299 D_GP: 0.032 D_real: 0.960 D_fake: 0.910 +(epoch: 382, iters: 3758, time: 0.063) G_GAN: 0.471 G_GAN_Feat: 1.006 G_ID: 0.112 G_Rec: 0.425 D_GP: 0.033 D_real: 1.011 D_fake: 0.530 +(epoch: 382, iters: 4158, time: 0.063) G_GAN: 0.465 G_GAN_Feat: 0.686 G_ID: 0.094 G_Rec: 0.282 D_GP: 0.035 D_real: 1.435 D_fake: 0.543 +(epoch: 382, iters: 4558, time: 0.063) G_GAN: 0.479 G_GAN_Feat: 0.945 G_ID: 0.108 G_Rec: 0.411 D_GP: 0.030 D_real: 1.102 D_fake: 0.536 +(epoch: 382, iters: 4958, time: 0.064) G_GAN: 0.256 G_GAN_Feat: 0.886 G_ID: 0.126 G_Rec: 0.360 D_GP: 0.211 D_real: 0.679 D_fake: 0.747 +(epoch: 382, iters: 5358, time: 0.063) G_GAN: 0.276 G_GAN_Feat: 0.998 G_ID: 0.123 G_Rec: 0.412 D_GP: 0.103 D_real: 0.624 D_fake: 0.724 +(epoch: 382, iters: 5758, time: 0.063) G_GAN: 0.279 G_GAN_Feat: 0.759 G_ID: 0.102 G_Rec: 0.328 D_GP: 0.033 D_real: 1.151 D_fake: 0.722 +(epoch: 382, iters: 6158, time: 0.063) G_GAN: 0.677 G_GAN_Feat: 1.040 G_ID: 0.114 G_Rec: 0.444 D_GP: 0.108 D_real: 1.105 D_fake: 0.374 +(epoch: 382, iters: 6558, time: 0.064) G_GAN: 0.671 G_GAN_Feat: 0.819 G_ID: 0.101 G_Rec: 0.323 D_GP: 0.037 D_real: 1.556 D_fake: 0.409 +(epoch: 382, iters: 6958, time: 0.063) G_GAN: 0.099 G_GAN_Feat: 1.060 G_ID: 0.121 G_Rec: 0.435 D_GP: 0.114 D_real: 0.348 D_fake: 0.901 +(epoch: 382, iters: 7358, time: 0.063) G_GAN: 0.510 G_GAN_Feat: 0.928 G_ID: 0.095 G_Rec: 0.346 D_GP: 0.086 D_real: 0.633 D_fake: 0.497 +(epoch: 382, iters: 7758, time: 0.063) G_GAN: 0.271 G_GAN_Feat: 0.993 G_ID: 0.131 G_Rec: 0.450 D_GP: 0.036 D_real: 0.900 D_fake: 0.730 +(epoch: 382, iters: 8158, time: 0.064) G_GAN: 0.457 G_GAN_Feat: 0.729 G_ID: 0.106 G_Rec: 0.281 D_GP: 0.034 D_real: 1.220 D_fake: 0.544 +(epoch: 382, iters: 8558, time: 0.063) G_GAN: 0.811 G_GAN_Feat: 1.118 G_ID: 0.151 G_Rec: 0.459 D_GP: 0.160 D_real: 0.918 D_fake: 0.397 +(epoch: 383, iters: 350, time: 0.063) G_GAN: -0.408 G_GAN_Feat: 0.774 G_ID: 0.118 G_Rec: 0.354 D_GP: 0.031 D_real: 0.676 D_fake: 1.408 +(epoch: 383, iters: 750, time: 0.063) G_GAN: 0.278 G_GAN_Feat: 0.903 G_ID: 0.151 G_Rec: 0.418 D_GP: 0.034 D_real: 0.895 D_fake: 0.724 +(epoch: 383, iters: 1150, time: 0.064) G_GAN: 0.079 G_GAN_Feat: 0.771 G_ID: 0.093 G_Rec: 0.339 D_GP: 0.035 D_real: 0.832 D_fake: 0.922 +(epoch: 383, iters: 1550, time: 0.063) G_GAN: 0.210 G_GAN_Feat: 0.997 G_ID: 0.127 G_Rec: 0.409 D_GP: 0.051 D_real: 0.752 D_fake: 0.794 +(epoch: 383, iters: 1950, time: 0.063) G_GAN: 0.321 G_GAN_Feat: 0.944 G_ID: 0.104 G_Rec: 0.317 D_GP: 0.396 D_real: 0.503 D_fake: 0.679 +(epoch: 383, iters: 2350, time: 0.063) G_GAN: 0.672 G_GAN_Feat: 0.886 G_ID: 0.105 G_Rec: 0.372 D_GP: 0.029 D_real: 1.361 D_fake: 0.352 +(epoch: 383, iters: 2750, time: 0.064) G_GAN: 0.279 G_GAN_Feat: 0.858 G_ID: 0.091 G_Rec: 0.355 D_GP: 0.032 D_real: 0.881 D_fake: 0.721 +(epoch: 383, iters: 3150, time: 0.063) G_GAN: 0.469 G_GAN_Feat: 0.974 G_ID: 0.140 G_Rec: 0.455 D_GP: 0.041 D_real: 1.018 D_fake: 0.539 +(epoch: 383, iters: 3550, time: 0.063) G_GAN: 0.077 G_GAN_Feat: 0.757 G_ID: 0.106 G_Rec: 0.301 D_GP: 0.042 D_real: 0.791 D_fake: 0.923 +(epoch: 383, iters: 3950, time: 0.063) G_GAN: 0.409 G_GAN_Feat: 1.034 G_ID: 0.133 G_Rec: 0.466 D_GP: 0.052 D_real: 0.830 D_fake: 0.598 +(epoch: 383, iters: 4350, time: 0.064) G_GAN: 0.318 G_GAN_Feat: 0.896 G_ID: 0.109 G_Rec: 0.298 D_GP: 0.067 D_real: 0.588 D_fake: 0.683 +(epoch: 383, iters: 4750, time: 0.063) G_GAN: 0.298 G_GAN_Feat: 1.149 G_ID: 0.129 G_Rec: 0.422 D_GP: 0.042 D_real: 0.473 D_fake: 0.704 +(epoch: 383, iters: 5150, time: 0.063) G_GAN: 0.314 G_GAN_Feat: 0.794 G_ID: 0.109 G_Rec: 0.315 D_GP: 0.033 D_real: 1.096 D_fake: 0.687 +(epoch: 383, iters: 5550, time: 0.063) G_GAN: 0.314 G_GAN_Feat: 0.903 G_ID: 0.138 G_Rec: 0.423 D_GP: 0.030 D_real: 1.050 D_fake: 0.695 +(epoch: 383, iters: 5950, time: 0.064) G_GAN: 0.126 G_GAN_Feat: 0.833 G_ID: 0.088 G_Rec: 0.355 D_GP: 0.040 D_real: 0.871 D_fake: 0.875 +(epoch: 383, iters: 6350, time: 0.063) G_GAN: 0.212 G_GAN_Feat: 1.000 G_ID: 0.133 G_Rec: 0.471 D_GP: 0.037 D_real: 0.728 D_fake: 0.788 +(epoch: 383, iters: 6750, time: 0.063) G_GAN: 0.143 G_GAN_Feat: 0.722 G_ID: 0.096 G_Rec: 0.297 D_GP: 0.031 D_real: 1.047 D_fake: 0.857 +(epoch: 383, iters: 7150, time: 0.063) G_GAN: 0.351 G_GAN_Feat: 1.014 G_ID: 0.120 G_Rec: 0.435 D_GP: 0.040 D_real: 0.851 D_fake: 0.652 +(epoch: 383, iters: 7550, time: 0.064) G_GAN: -0.022 G_GAN_Feat: 0.791 G_ID: 0.103 G_Rec: 0.312 D_GP: 0.035 D_real: 0.904 D_fake: 1.022 +(epoch: 383, iters: 7950, time: 0.063) G_GAN: 0.423 G_GAN_Feat: 1.052 G_ID: 0.126 G_Rec: 0.426 D_GP: 0.152 D_real: 0.500 D_fake: 0.582 +(epoch: 383, iters: 8350, time: 0.064) G_GAN: 0.555 G_GAN_Feat: 0.932 G_ID: 0.116 G_Rec: 0.359 D_GP: 0.179 D_real: 0.663 D_fake: 0.507 +(epoch: 384, iters: 142, time: 0.064) G_GAN: 0.610 G_GAN_Feat: 0.938 G_ID: 0.131 G_Rec: 0.421 D_GP: 0.030 D_real: 1.207 D_fake: 0.443 +(epoch: 384, iters: 542, time: 0.064) G_GAN: 0.594 G_GAN_Feat: 1.078 G_ID: 0.100 G_Rec: 0.374 D_GP: 0.038 D_real: 1.363 D_fake: 0.510 +(epoch: 384, iters: 942, time: 0.064) G_GAN: 0.763 G_GAN_Feat: 1.143 G_ID: 0.122 G_Rec: 0.455 D_GP: 0.082 D_real: 0.677 D_fake: 0.275 +(epoch: 384, iters: 1342, time: 0.064) G_GAN: -0.047 G_GAN_Feat: 0.845 G_ID: 0.119 G_Rec: 0.349 D_GP: 0.036 D_real: 0.728 D_fake: 1.047 +(epoch: 384, iters: 1742, time: 0.064) G_GAN: 0.412 G_GAN_Feat: 0.950 G_ID: 0.121 G_Rec: 0.420 D_GP: 0.044 D_real: 0.939 D_fake: 0.590 +(epoch: 384, iters: 2142, time: 0.063) G_GAN: 0.195 G_GAN_Feat: 0.758 G_ID: 0.101 G_Rec: 0.291 D_GP: 0.045 D_real: 0.941 D_fake: 0.807 +(epoch: 384, iters: 2542, time: 0.064) G_GAN: 0.369 G_GAN_Feat: 1.161 G_ID: 0.130 G_Rec: 0.464 D_GP: 0.325 D_real: 0.152 D_fake: 0.642 +(epoch: 384, iters: 2942, time: 0.063) G_GAN: 0.583 G_GAN_Feat: 1.005 G_ID: 0.113 G_Rec: 0.365 D_GP: 0.286 D_real: 0.439 D_fake: 0.560 +(epoch: 384, iters: 3342, time: 0.064) G_GAN: 0.296 G_GAN_Feat: 1.238 G_ID: 0.119 G_Rec: 0.465 D_GP: 0.045 D_real: 0.323 D_fake: 0.705 +(epoch: 384, iters: 3742, time: 0.063) G_GAN: 0.094 G_GAN_Feat: 0.891 G_ID: 0.109 G_Rec: 0.349 D_GP: 0.039 D_real: 0.695 D_fake: 0.907 +(epoch: 384, iters: 4142, time: 0.063) G_GAN: 0.508 G_GAN_Feat: 1.031 G_ID: 0.129 G_Rec: 0.491 D_GP: 0.031 D_real: 1.070 D_fake: 0.494 +(epoch: 384, iters: 4542, time: 0.063) G_GAN: 0.715 G_GAN_Feat: 1.099 G_ID: 0.114 G_Rec: 0.406 D_GP: 0.136 D_real: 0.492 D_fake: 0.441 +(epoch: 384, iters: 4942, time: 0.064) G_GAN: 0.842 G_GAN_Feat: 1.189 G_ID: 0.120 G_Rec: 0.469 D_GP: 0.047 D_real: 1.175 D_fake: 0.232 +(epoch: 384, iters: 5342, time: 0.063) G_GAN: 0.300 G_GAN_Feat: 1.004 G_ID: 0.111 G_Rec: 0.365 D_GP: 0.094 D_real: 0.881 D_fake: 0.707 +(epoch: 384, iters: 5742, time: 0.063) G_GAN: 0.393 G_GAN_Feat: 1.075 G_ID: 0.137 G_Rec: 0.479 D_GP: 0.050 D_real: 0.727 D_fake: 0.608 +(epoch: 384, iters: 6142, time: 0.064) G_GAN: 0.218 G_GAN_Feat: 1.001 G_ID: 0.096 G_Rec: 0.408 D_GP: 0.270 D_real: 0.387 D_fake: 0.786 +(epoch: 384, iters: 6542, time: 0.064) G_GAN: 0.865 G_GAN_Feat: 1.090 G_ID: 0.131 G_Rec: 0.464 D_GP: 0.038 D_real: 0.927 D_fake: 0.181 +(epoch: 384, iters: 6942, time: 0.063) G_GAN: 0.149 G_GAN_Feat: 0.701 G_ID: 0.106 G_Rec: 0.314 D_GP: 0.024 D_real: 1.086 D_fake: 0.851 +(epoch: 384, iters: 7342, time: 0.063) G_GAN: 0.177 G_GAN_Feat: 1.059 G_ID: 0.126 G_Rec: 0.467 D_GP: 0.040 D_real: 0.658 D_fake: 0.823 +(epoch: 384, iters: 7742, time: 0.063) G_GAN: 0.203 G_GAN_Feat: 0.704 G_ID: 0.124 G_Rec: 0.309 D_GP: 0.031 D_real: 1.053 D_fake: 0.798 +(epoch: 384, iters: 8142, time: 0.064) G_GAN: 0.786 G_GAN_Feat: 1.094 G_ID: 0.114 G_Rec: 0.488 D_GP: 0.078 D_real: 1.361 D_fake: 0.268 +(epoch: 384, iters: 8542, time: 0.063) G_GAN: 0.058 G_GAN_Feat: 0.718 G_ID: 0.121 G_Rec: 0.338 D_GP: 0.033 D_real: 0.832 D_fake: 0.942 +(epoch: 385, iters: 334, time: 0.063) G_GAN: 0.068 G_GAN_Feat: 0.882 G_ID: 0.135 G_Rec: 0.401 D_GP: 0.038 D_real: 0.681 D_fake: 0.932 +(epoch: 385, iters: 734, time: 0.063) G_GAN: 0.022 G_GAN_Feat: 0.742 G_ID: 0.101 G_Rec: 0.323 D_GP: 0.035 D_real: 0.848 D_fake: 0.979 +(epoch: 385, iters: 1134, time: 0.064) G_GAN: 0.479 G_GAN_Feat: 0.975 G_ID: 0.129 G_Rec: 0.463 D_GP: 0.030 D_real: 1.176 D_fake: 0.542 +(epoch: 385, iters: 1534, time: 0.064) G_GAN: 0.194 G_GAN_Feat: 0.764 G_ID: 0.099 G_Rec: 0.348 D_GP: 0.031 D_real: 0.992 D_fake: 0.808 +(epoch: 385, iters: 1934, time: 0.063) G_GAN: 0.512 G_GAN_Feat: 0.860 G_ID: 0.122 G_Rec: 0.404 D_GP: 0.037 D_real: 1.220 D_fake: 0.521 +(epoch: 385, iters: 2334, time: 0.063) G_GAN: 0.125 G_GAN_Feat: 0.713 G_ID: 0.121 G_Rec: 0.316 D_GP: 0.039 D_real: 1.010 D_fake: 0.875 +(epoch: 385, iters: 2734, time: 0.064) G_GAN: 0.056 G_GAN_Feat: 1.003 G_ID: 0.144 G_Rec: 0.458 D_GP: 0.060 D_real: 0.421 D_fake: 0.944 +(epoch: 385, iters: 3134, time: 0.063) G_GAN: -0.093 G_GAN_Feat: 0.903 G_ID: 0.128 G_Rec: 0.342 D_GP: 0.294 D_real: 0.302 D_fake: 1.093 +(epoch: 385, iters: 3534, time: 0.063) G_GAN: 0.422 G_GAN_Feat: 1.075 G_ID: 0.118 G_Rec: 0.473 D_GP: 0.111 D_real: 0.505 D_fake: 0.591 +(epoch: 385, iters: 3934, time: 0.063) G_GAN: 0.761 G_GAN_Feat: 0.754 G_ID: 0.092 G_Rec: 0.313 D_GP: 0.034 D_real: 1.682 D_fake: 0.274 +(epoch: 385, iters: 4334, time: 0.064) G_GAN: 0.827 G_GAN_Feat: 0.991 G_ID: 0.124 G_Rec: 0.407 D_GP: 0.045 D_real: 1.404 D_fake: 0.219 +(epoch: 385, iters: 4734, time: 0.063) G_GAN: 0.421 G_GAN_Feat: 0.935 G_ID: 0.098 G_Rec: 0.312 D_GP: 0.044 D_real: 1.214 D_fake: 0.629 +(epoch: 385, iters: 5134, time: 0.063) G_GAN: 0.460 G_GAN_Feat: 0.983 G_ID: 0.157 G_Rec: 0.413 D_GP: 0.043 D_real: 1.117 D_fake: 0.541 +(epoch: 385, iters: 5534, time: 0.063) G_GAN: 0.448 G_GAN_Feat: 0.798 G_ID: 0.101 G_Rec: 0.312 D_GP: 0.032 D_real: 1.372 D_fake: 0.557 +(epoch: 385, iters: 5934, time: 0.064) G_GAN: 0.711 G_GAN_Feat: 1.036 G_ID: 0.124 G_Rec: 0.459 D_GP: 0.055 D_real: 1.056 D_fake: 0.319 +(epoch: 385, iters: 6334, time: 0.063) G_GAN: 0.328 G_GAN_Feat: 0.771 G_ID: 0.099 G_Rec: 0.352 D_GP: 0.059 D_real: 1.122 D_fake: 0.697 +(epoch: 385, iters: 6734, time: 0.063) G_GAN: 0.552 G_GAN_Feat: 0.911 G_ID: 0.132 G_Rec: 0.441 D_GP: 0.040 D_real: 1.096 D_fake: 0.480 +(epoch: 385, iters: 7134, time: 0.063) G_GAN: 0.053 G_GAN_Feat: 0.687 G_ID: 0.104 G_Rec: 0.328 D_GP: 0.032 D_real: 0.915 D_fake: 0.947 +(epoch: 385, iters: 7534, time: 0.064) G_GAN: 0.335 G_GAN_Feat: 0.882 G_ID: 0.123 G_Rec: 0.425 D_GP: 0.031 D_real: 1.022 D_fake: 0.672 +(epoch: 385, iters: 7934, time: 0.063) G_GAN: 0.195 G_GAN_Feat: 0.723 G_ID: 0.124 G_Rec: 0.301 D_GP: 0.042 D_real: 0.877 D_fake: 0.805 +(epoch: 385, iters: 8334, time: 0.063) G_GAN: 0.519 G_GAN_Feat: 0.927 G_ID: 0.124 G_Rec: 0.435 D_GP: 0.029 D_real: 1.099 D_fake: 0.500 +(epoch: 386, iters: 126, time: 0.063) G_GAN: 0.039 G_GAN_Feat: 0.844 G_ID: 0.106 G_Rec: 0.337 D_GP: 0.043 D_real: 0.698 D_fake: 0.961 +(epoch: 386, iters: 526, time: 0.064) G_GAN: 0.810 G_GAN_Feat: 0.900 G_ID: 0.106 G_Rec: 0.410 D_GP: 0.033 D_real: 1.514 D_fake: 0.233 +(epoch: 386, iters: 926, time: 0.063) G_GAN: 0.454 G_GAN_Feat: 0.740 G_ID: 0.095 G_Rec: 0.294 D_GP: 0.030 D_real: 1.297 D_fake: 0.550 +(epoch: 386, iters: 1326, time: 0.063) G_GAN: 0.686 G_GAN_Feat: 1.033 G_ID: 0.120 G_Rec: 0.412 D_GP: 0.076 D_real: 0.909 D_fake: 0.345 +(epoch: 386, iters: 1726, time: 0.064) G_GAN: 0.142 G_GAN_Feat: 0.744 G_ID: 0.107 G_Rec: 0.340 D_GP: 0.037 D_real: 1.011 D_fake: 0.860 +(epoch: 386, iters: 2126, time: 0.064) G_GAN: 0.125 G_GAN_Feat: 0.888 G_ID: 0.140 G_Rec: 0.432 D_GP: 0.033 D_real: 0.847 D_fake: 0.875 +(epoch: 386, iters: 2526, time: 0.064) G_GAN: 0.077 G_GAN_Feat: 0.642 G_ID: 0.095 G_Rec: 0.259 D_GP: 0.027 D_real: 1.010 D_fake: 0.923 +(epoch: 386, iters: 2926, time: 0.064) G_GAN: 0.535 G_GAN_Feat: 0.994 G_ID: 0.127 G_Rec: 0.468 D_GP: 0.032 D_real: 1.127 D_fake: 0.475 +(epoch: 386, iters: 3326, time: 0.064) G_GAN: 0.041 G_GAN_Feat: 0.852 G_ID: 0.118 G_Rec: 0.332 D_GP: 0.061 D_real: 0.565 D_fake: 0.959 +(epoch: 386, iters: 3726, time: 0.064) G_GAN: 0.618 G_GAN_Feat: 0.924 G_ID: 0.131 G_Rec: 0.392 D_GP: 0.041 D_real: 1.098 D_fake: 0.391 +(epoch: 386, iters: 4126, time: 0.064) G_GAN: 0.335 G_GAN_Feat: 0.794 G_ID: 0.096 G_Rec: 0.294 D_GP: 0.035 D_real: 1.002 D_fake: 0.666 +(epoch: 386, iters: 4526, time: 0.064) G_GAN: 0.527 G_GAN_Feat: 0.934 G_ID: 0.133 G_Rec: 0.454 D_GP: 0.036 D_real: 1.240 D_fake: 0.492 +(epoch: 386, iters: 4926, time: 0.064) G_GAN: 0.212 G_GAN_Feat: 0.805 G_ID: 0.101 G_Rec: 0.366 D_GP: 0.039 D_real: 1.032 D_fake: 0.792 +(epoch: 386, iters: 5326, time: 0.064) G_GAN: 0.373 G_GAN_Feat: 0.916 G_ID: 0.113 G_Rec: 0.441 D_GP: 0.034 D_real: 0.960 D_fake: 0.628 +(epoch: 386, iters: 5726, time: 0.064) G_GAN: 0.028 G_GAN_Feat: 0.723 G_ID: 0.107 G_Rec: 0.319 D_GP: 0.035 D_real: 0.877 D_fake: 0.972 +(epoch: 386, iters: 6126, time: 0.064) G_GAN: 0.305 G_GAN_Feat: 0.919 G_ID: 0.120 G_Rec: 0.418 D_GP: 0.053 D_real: 0.913 D_fake: 0.696 +(epoch: 386, iters: 6526, time: 0.064) G_GAN: 0.056 G_GAN_Feat: 0.915 G_ID: 0.115 G_Rec: 0.365 D_GP: 0.045 D_real: 0.589 D_fake: 0.944 +(epoch: 386, iters: 6926, time: 0.064) G_GAN: 0.348 G_GAN_Feat: 0.943 G_ID: 0.126 G_Rec: 0.402 D_GP: 0.036 D_real: 0.903 D_fake: 0.657 +(epoch: 386, iters: 7326, time: 0.064) G_GAN: 0.445 G_GAN_Feat: 0.724 G_ID: 0.105 G_Rec: 0.287 D_GP: 0.033 D_real: 1.294 D_fake: 0.560 +(epoch: 386, iters: 7726, time: 0.064) G_GAN: 0.574 G_GAN_Feat: 1.071 G_ID: 0.118 G_Rec: 0.434 D_GP: 0.092 D_real: 0.590 D_fake: 0.442 +(epoch: 386, iters: 8126, time: 0.064) G_GAN: 0.281 G_GAN_Feat: 0.935 G_ID: 0.097 G_Rec: 0.346 D_GP: 0.078 D_real: 0.771 D_fake: 0.722 +(epoch: 386, iters: 8526, time: 0.064) G_GAN: 0.659 G_GAN_Feat: 1.151 G_ID: 0.135 G_Rec: 0.517 D_GP: 0.093 D_real: 1.027 D_fake: 0.544 +(epoch: 387, iters: 318, time: 0.064) G_GAN: 0.386 G_GAN_Feat: 0.900 G_ID: 0.106 G_Rec: 0.339 D_GP: 0.142 D_real: 0.651 D_fake: 0.626 +(epoch: 387, iters: 718, time: 0.064) G_GAN: 0.257 G_GAN_Feat: 1.173 G_ID: 0.163 G_Rec: 0.487 D_GP: 0.046 D_real: 0.545 D_fake: 0.745 +(epoch: 387, iters: 1118, time: 0.064) G_GAN: 0.278 G_GAN_Feat: 0.771 G_ID: 0.107 G_Rec: 0.315 D_GP: 0.037 D_real: 1.160 D_fake: 0.723 +(epoch: 387, iters: 1518, time: 0.064) G_GAN: 0.717 G_GAN_Feat: 1.098 G_ID: 0.142 G_Rec: 0.439 D_GP: 0.043 D_real: 0.950 D_fake: 0.331 +(epoch: 387, iters: 1918, time: 0.064) G_GAN: -0.143 G_GAN_Feat: 0.997 G_ID: 0.110 G_Rec: 0.322 D_GP: 0.040 D_real: 0.462 D_fake: 1.143 +(epoch: 387, iters: 2318, time: 0.064) G_GAN: 0.827 G_GAN_Feat: 1.036 G_ID: 0.116 G_Rec: 0.462 D_GP: 0.054 D_real: 1.362 D_fake: 0.298 +(epoch: 387, iters: 2718, time: 0.064) G_GAN: 0.143 G_GAN_Feat: 0.855 G_ID: 0.109 G_Rec: 0.371 D_GP: 0.043 D_real: 0.939 D_fake: 0.857 +(epoch: 387, iters: 3118, time: 0.064) G_GAN: 0.714 G_GAN_Feat: 0.975 G_ID: 0.122 G_Rec: 0.404 D_GP: 0.048 D_real: 1.273 D_fake: 0.330 +(epoch: 387, iters: 3518, time: 0.063) G_GAN: 0.742 G_GAN_Feat: 0.980 G_ID: 0.115 G_Rec: 0.335 D_GP: 0.100 D_real: 0.525 D_fake: 0.506 +(epoch: 387, iters: 3918, time: 0.063) G_GAN: 1.195 G_GAN_Feat: 1.289 G_ID: 0.116 G_Rec: 0.489 D_GP: 0.251 D_real: 0.618 D_fake: 0.130 +(epoch: 387, iters: 4318, time: 0.063) G_GAN: 0.363 G_GAN_Feat: 0.888 G_ID: 0.107 G_Rec: 0.310 D_GP: 0.040 D_real: 0.801 D_fake: 0.680 +(epoch: 387, iters: 4718, time: 0.064) G_GAN: 0.445 G_GAN_Feat: 0.952 G_ID: 0.125 G_Rec: 0.467 D_GP: 0.028 D_real: 1.109 D_fake: 0.557 +(epoch: 387, iters: 5118, time: 0.063) G_GAN: 0.095 G_GAN_Feat: 0.660 G_ID: 0.131 G_Rec: 0.306 D_GP: 0.029 D_real: 1.019 D_fake: 0.905 +(epoch: 387, iters: 5518, time: 0.063) G_GAN: 0.424 G_GAN_Feat: 0.837 G_ID: 0.132 G_Rec: 0.381 D_GP: 0.033 D_real: 1.077 D_fake: 0.581 +(epoch: 387, iters: 5918, time: 0.063) G_GAN: -0.076 G_GAN_Feat: 0.703 G_ID: 0.085 G_Rec: 0.308 D_GP: 0.031 D_real: 0.800 D_fake: 1.076 +(epoch: 387, iters: 6318, time: 0.064) G_GAN: 0.455 G_GAN_Feat: 0.969 G_ID: 0.118 G_Rec: 0.489 D_GP: 0.069 D_real: 1.032 D_fake: 0.557 +(epoch: 387, iters: 6718, time: 0.063) G_GAN: -0.000 G_GAN_Feat: 0.713 G_ID: 0.169 G_Rec: 0.314 D_GP: 0.036 D_real: 0.768 D_fake: 1.000 +(epoch: 387, iters: 7118, time: 0.063) G_GAN: 0.388 G_GAN_Feat: 1.100 G_ID: 0.119 G_Rec: 0.498 D_GP: 0.127 D_real: 0.493 D_fake: 0.635 +(epoch: 387, iters: 7518, time: 0.063) G_GAN: 0.078 G_GAN_Feat: 0.717 G_ID: 0.126 G_Rec: 0.291 D_GP: 0.042 D_real: 0.743 D_fake: 0.922 +(epoch: 387, iters: 7918, time: 0.064) G_GAN: 0.397 G_GAN_Feat: 0.975 G_ID: 0.127 G_Rec: 0.425 D_GP: 0.052 D_real: 0.771 D_fake: 0.606 +(epoch: 387, iters: 8318, time: 0.063) G_GAN: -0.091 G_GAN_Feat: 0.861 G_ID: 0.101 G_Rec: 0.354 D_GP: 0.036 D_real: 1.012 D_fake: 1.108 +(epoch: 388, iters: 110, time: 0.063) G_GAN: 0.626 G_GAN_Feat: 0.885 G_ID: 0.114 G_Rec: 0.426 D_GP: 0.030 D_real: 1.340 D_fake: 0.402 +(epoch: 388, iters: 510, time: 0.063) G_GAN: 0.215 G_GAN_Feat: 0.813 G_ID: 0.106 G_Rec: 0.301 D_GP: 0.096 D_real: 0.787 D_fake: 0.786 +(epoch: 388, iters: 910, time: 0.064) G_GAN: 0.829 G_GAN_Feat: 1.056 G_ID: 0.110 G_Rec: 0.476 D_GP: 0.040 D_real: 1.196 D_fake: 0.223 +(epoch: 388, iters: 1310, time: 0.064) G_GAN: 0.400 G_GAN_Feat: 0.802 G_ID: 0.113 G_Rec: 0.337 D_GP: 0.098 D_real: 1.095 D_fake: 0.603 +(epoch: 388, iters: 1710, time: 0.064) G_GAN: 0.591 G_GAN_Feat: 1.053 G_ID: 0.125 G_Rec: 0.417 D_GP: 0.047 D_real: 0.802 D_fake: 0.415 +(epoch: 388, iters: 2110, time: 0.063) G_GAN: 0.577 G_GAN_Feat: 0.833 G_ID: 0.153 G_Rec: 0.307 D_GP: 0.045 D_real: 1.149 D_fake: 0.429 +(epoch: 388, iters: 2510, time: 0.064) G_GAN: -0.090 G_GAN_Feat: 0.868 G_ID: 0.121 G_Rec: 0.389 D_GP: 0.032 D_real: 0.843 D_fake: 1.090 +(epoch: 388, iters: 2910, time: 0.063) G_GAN: 0.432 G_GAN_Feat: 0.851 G_ID: 0.104 G_Rec: 0.291 D_GP: 0.091 D_real: 0.836 D_fake: 0.569 +(epoch: 388, iters: 3310, time: 0.063) G_GAN: 0.609 G_GAN_Feat: 1.048 G_ID: 0.143 G_Rec: 0.455 D_GP: 0.040 D_real: 1.187 D_fake: 0.401 +(epoch: 388, iters: 3710, time: 0.063) G_GAN: 0.225 G_GAN_Feat: 0.757 G_ID: 0.111 G_Rec: 0.317 D_GP: 0.034 D_real: 1.087 D_fake: 0.776 +(epoch: 388, iters: 4110, time: 0.064) G_GAN: 0.170 G_GAN_Feat: 0.984 G_ID: 0.142 G_Rec: 0.460 D_GP: 0.038 D_real: 0.786 D_fake: 0.830 +(epoch: 388, iters: 4510, time: 0.063) G_GAN: 0.171 G_GAN_Feat: 0.755 G_ID: 0.122 G_Rec: 0.308 D_GP: 0.031 D_real: 1.091 D_fake: 0.829 +(epoch: 388, iters: 4910, time: 0.063) G_GAN: 0.327 G_GAN_Feat: 1.299 G_ID: 0.138 G_Rec: 0.499 D_GP: 0.331 D_real: 0.862 D_fake: 0.688 +(epoch: 388, iters: 5310, time: 0.063) G_GAN: 0.132 G_GAN_Feat: 0.752 G_ID: 0.112 G_Rec: 0.317 D_GP: 0.030 D_real: 0.940 D_fake: 0.868 +(epoch: 388, iters: 5710, time: 0.064) G_GAN: 0.545 G_GAN_Feat: 0.916 G_ID: 0.110 G_Rec: 0.421 D_GP: 0.038 D_real: 1.076 D_fake: 0.476 +(epoch: 388, iters: 6110, time: 0.063) G_GAN: 0.375 G_GAN_Feat: 0.797 G_ID: 0.107 G_Rec: 0.321 D_GP: 0.046 D_real: 0.986 D_fake: 0.633 +(epoch: 388, iters: 6510, time: 0.064) G_GAN: 0.823 G_GAN_Feat: 0.982 G_ID: 0.126 G_Rec: 0.390 D_GP: 0.063 D_real: 1.372 D_fake: 0.264 +(epoch: 388, iters: 6910, time: 0.064) G_GAN: -0.004 G_GAN_Feat: 0.936 G_ID: 0.101 G_Rec: 0.365 D_GP: 0.053 D_real: 0.953 D_fake: 1.004 +(epoch: 388, iters: 7310, time: 0.063) G_GAN: 0.623 G_GAN_Feat: 0.958 G_ID: 0.143 G_Rec: 0.417 D_GP: 0.047 D_real: 1.262 D_fake: 0.415 +(epoch: 388, iters: 7710, time: 0.063) G_GAN: 0.322 G_GAN_Feat: 0.749 G_ID: 0.110 G_Rec: 0.281 D_GP: 0.034 D_real: 0.967 D_fake: 0.680 +(epoch: 388, iters: 8110, time: 0.063) G_GAN: 0.520 G_GAN_Feat: 1.062 G_ID: 0.119 G_Rec: 0.418 D_GP: 0.089 D_real: 0.859 D_fake: 0.526 +(epoch: 388, iters: 8510, time: 0.064) G_GAN: 0.474 G_GAN_Feat: 0.842 G_ID: 0.096 G_Rec: 0.310 D_GP: 0.062 D_real: 1.039 D_fake: 0.545 +(epoch: 389, iters: 302, time: 0.063) G_GAN: 0.570 G_GAN_Feat: 1.065 G_ID: 0.118 G_Rec: 0.475 D_GP: 0.035 D_real: 1.066 D_fake: 0.439 +(epoch: 389, iters: 702, time: 0.063) G_GAN: 0.260 G_GAN_Feat: 0.894 G_ID: 0.080 G_Rec: 0.324 D_GP: 1.041 D_real: 0.477 D_fake: 0.744 +(epoch: 389, iters: 1102, time: 0.063) G_GAN: 0.577 G_GAN_Feat: 1.008 G_ID: 0.107 G_Rec: 0.467 D_GP: 0.031 D_real: 1.118 D_fake: 0.471 +(epoch: 389, iters: 1502, time: 0.064) G_GAN: 0.574 G_GAN_Feat: 0.739 G_ID: 0.092 G_Rec: 0.297 D_GP: 0.031 D_real: 1.361 D_fake: 0.444 +(epoch: 389, iters: 1902, time: 0.063) G_GAN: 0.276 G_GAN_Feat: 1.119 G_ID: 0.136 G_Rec: 0.478 D_GP: 0.031 D_real: 0.711 D_fake: 0.725 +(epoch: 389, iters: 2302, time: 0.063) G_GAN: 0.747 G_GAN_Feat: 1.047 G_ID: 0.112 G_Rec: 0.401 D_GP: 0.152 D_real: 0.524 D_fake: 0.551 +(epoch: 389, iters: 2702, time: 0.063) G_GAN: 0.541 G_GAN_Feat: 0.960 G_ID: 0.122 G_Rec: 0.434 D_GP: 0.030 D_real: 1.275 D_fake: 0.463 +(epoch: 389, iters: 3102, time: 0.064) G_GAN: 0.137 G_GAN_Feat: 0.979 G_ID: 0.096 G_Rec: 0.349 D_GP: 0.044 D_real: 0.545 D_fake: 0.863 +(epoch: 389, iters: 3502, time: 0.063) G_GAN: 0.895 G_GAN_Feat: 1.142 G_ID: 0.133 G_Rec: 0.458 D_GP: 0.210 D_real: 0.843 D_fake: 0.207 +(epoch: 389, iters: 3902, time: 0.063) G_GAN: 0.102 G_GAN_Feat: 0.793 G_ID: 0.108 G_Rec: 0.339 D_GP: 0.033 D_real: 0.947 D_fake: 0.898 +(epoch: 389, iters: 4302, time: 0.063) G_GAN: 0.779 G_GAN_Feat: 0.995 G_ID: 0.115 G_Rec: 0.442 D_GP: 0.036 D_real: 1.339 D_fake: 0.250 +(epoch: 389, iters: 4702, time: 0.064) G_GAN: 0.586 G_GAN_Feat: 0.887 G_ID: 0.088 G_Rec: 0.323 D_GP: 0.046 D_real: 0.820 D_fake: 0.540 +(epoch: 389, iters: 5102, time: 0.063) G_GAN: 0.443 G_GAN_Feat: 1.160 G_ID: 0.132 G_Rec: 0.465 D_GP: 0.043 D_real: 1.019 D_fake: 0.563 +(epoch: 389, iters: 5502, time: 0.063) G_GAN: -0.071 G_GAN_Feat: 0.880 G_ID: 0.115 G_Rec: 0.328 D_GP: 0.039 D_real: 0.465 D_fake: 1.072 +(epoch: 389, iters: 5902, time: 0.063) G_GAN: 0.477 G_GAN_Feat: 1.022 G_ID: 0.114 G_Rec: 0.461 D_GP: 0.047 D_real: 0.988 D_fake: 0.531 +(epoch: 389, iters: 6302, time: 0.064) G_GAN: 0.207 G_GAN_Feat: 0.806 G_ID: 0.095 G_Rec: 0.318 D_GP: 0.044 D_real: 0.992 D_fake: 0.793 +(epoch: 389, iters: 6702, time: 0.063) G_GAN: 0.635 G_GAN_Feat: 0.942 G_ID: 0.121 G_Rec: 0.434 D_GP: 0.030 D_real: 1.401 D_fake: 0.387 +(epoch: 389, iters: 7102, time: 0.063) G_GAN: 0.207 G_GAN_Feat: 0.857 G_ID: 0.088 G_Rec: 0.298 D_GP: 0.030 D_real: 0.741 D_fake: 0.794 +(epoch: 389, iters: 7502, time: 0.063) G_GAN: 0.394 G_GAN_Feat: 1.119 G_ID: 0.136 G_Rec: 0.466 D_GP: 0.047 D_real: 0.597 D_fake: 0.610 +(epoch: 389, iters: 7902, time: 0.064) G_GAN: -0.389 G_GAN_Feat: 0.762 G_ID: 0.097 G_Rec: 0.343 D_GP: 0.036 D_real: 0.634 D_fake: 1.389 +(epoch: 389, iters: 8302, time: 0.063) G_GAN: 0.487 G_GAN_Feat: 0.962 G_ID: 0.132 G_Rec: 0.420 D_GP: 0.033 D_real: 1.037 D_fake: 0.525 +(epoch: 389, iters: 8702, time: 0.063) G_GAN: 0.264 G_GAN_Feat: 0.757 G_ID: 0.122 G_Rec: 0.322 D_GP: 0.033 D_real: 1.114 D_fake: 0.736 +(epoch: 390, iters: 494, time: 0.063) G_GAN: 0.454 G_GAN_Feat: 0.989 G_ID: 0.124 G_Rec: 0.422 D_GP: 0.042 D_real: 0.942 D_fake: 0.562 +(epoch: 390, iters: 894, time: 0.064) G_GAN: 0.325 G_GAN_Feat: 0.776 G_ID: 0.105 G_Rec: 0.313 D_GP: 0.041 D_real: 1.364 D_fake: 0.715 +(epoch: 390, iters: 1294, time: 0.063) G_GAN: 0.560 G_GAN_Feat: 0.886 G_ID: 0.113 G_Rec: 0.407 D_GP: 0.037 D_real: 1.276 D_fake: 0.443 +(epoch: 390, iters: 1694, time: 0.063) G_GAN: 0.313 G_GAN_Feat: 0.729 G_ID: 0.097 G_Rec: 0.304 D_GP: 0.032 D_real: 1.216 D_fake: 0.688 +(epoch: 390, iters: 2094, time: 0.063) G_GAN: 0.529 G_GAN_Feat: 0.991 G_ID: 0.116 G_Rec: 0.423 D_GP: 0.038 D_real: 1.180 D_fake: 0.492 +(epoch: 390, iters: 2494, time: 0.064) G_GAN: -0.001 G_GAN_Feat: 0.920 G_ID: 0.119 G_Rec: 0.332 D_GP: 0.050 D_real: 0.349 D_fake: 1.001 +(epoch: 390, iters: 2894, time: 0.063) G_GAN: 0.561 G_GAN_Feat: 0.954 G_ID: 0.115 G_Rec: 0.440 D_GP: 0.030 D_real: 1.256 D_fake: 0.455 +(epoch: 390, iters: 3294, time: 0.063) G_GAN: -0.004 G_GAN_Feat: 0.741 G_ID: 0.104 G_Rec: 0.347 D_GP: 0.051 D_real: 0.781 D_fake: 1.004 +(epoch: 390, iters: 3694, time: 0.063) G_GAN: 0.614 G_GAN_Feat: 1.079 G_ID: 0.143 G_Rec: 0.464 D_GP: 0.156 D_real: 0.806 D_fake: 0.398 +(epoch: 390, iters: 4094, time: 0.064) G_GAN: -0.120 G_GAN_Feat: 0.943 G_ID: 0.135 G_Rec: 0.332 D_GP: 0.063 D_real: 0.188 D_fake: 1.120 +(epoch: 390, iters: 4494, time: 0.063) G_GAN: 0.461 G_GAN_Feat: 1.048 G_ID: 0.135 G_Rec: 0.387 D_GP: 0.046 D_real: 0.813 D_fake: 0.540 +(epoch: 390, iters: 4894, time: 0.064) G_GAN: 0.274 G_GAN_Feat: 0.999 G_ID: 0.113 G_Rec: 0.362 D_GP: 0.065 D_real: 0.383 D_fake: 0.730 +(epoch: 390, iters: 5294, time: 0.063) G_GAN: 0.293 G_GAN_Feat: 1.224 G_ID: 0.136 G_Rec: 0.469 D_GP: 0.049 D_real: 0.461 D_fake: 0.709 +(epoch: 390, iters: 5694, time: 0.064) G_GAN: 0.126 G_GAN_Feat: 0.872 G_ID: 0.114 G_Rec: 0.346 D_GP: 0.151 D_real: 0.535 D_fake: 0.889 +(epoch: 390, iters: 6094, time: 0.063) G_GAN: 0.630 G_GAN_Feat: 1.109 G_ID: 0.138 G_Rec: 0.401 D_GP: 0.047 D_real: 0.760 D_fake: 0.414 +(epoch: 390, iters: 6494, time: 0.063) G_GAN: 0.428 G_GAN_Feat: 0.863 G_ID: 0.099 G_Rec: 0.300 D_GP: 0.042 D_real: 0.790 D_fake: 0.592 +(epoch: 390, iters: 6894, time: 0.063) G_GAN: 0.386 G_GAN_Feat: 0.965 G_ID: 0.130 G_Rec: 0.427 D_GP: 0.033 D_real: 1.018 D_fake: 0.615 +(epoch: 390, iters: 7294, time: 0.064) G_GAN: 0.412 G_GAN_Feat: 0.800 G_ID: 0.094 G_Rec: 0.305 D_GP: 0.039 D_real: 1.158 D_fake: 0.589 +(epoch: 390, iters: 7694, time: 0.063) G_GAN: 0.726 G_GAN_Feat: 0.972 G_ID: 0.143 G_Rec: 0.434 D_GP: 0.038 D_real: 1.293 D_fake: 0.296 +(epoch: 390, iters: 8094, time: 0.063) G_GAN: 0.488 G_GAN_Feat: 1.095 G_ID: 0.104 G_Rec: 0.381 D_GP: 0.058 D_real: 1.474 D_fake: 0.548 +(epoch: 390, iters: 8494, time: 0.063) G_GAN: 0.383 G_GAN_Feat: 0.988 G_ID: 0.127 G_Rec: 0.471 D_GP: 0.031 D_real: 1.039 D_fake: 0.652 +(epoch: 391, iters: 286, time: 0.064) G_GAN: -0.046 G_GAN_Feat: 0.711 G_ID: 0.103 G_Rec: 0.281 D_GP: 0.034 D_real: 0.930 D_fake: 1.046 +(epoch: 391, iters: 686, time: 0.063) G_GAN: 0.424 G_GAN_Feat: 1.056 G_ID: 0.121 G_Rec: 0.448 D_GP: 0.038 D_real: 0.784 D_fake: 0.578 +(epoch: 391, iters: 1086, time: 0.063) G_GAN: 0.347 G_GAN_Feat: 0.907 G_ID: 0.104 G_Rec: 0.313 D_GP: 0.037 D_real: 0.582 D_fake: 0.657 +(epoch: 391, iters: 1486, time: 0.064) G_GAN: 0.820 G_GAN_Feat: 1.011 G_ID: 0.124 G_Rec: 0.425 D_GP: 0.070 D_real: 1.651 D_fake: 0.238 +(epoch: 391, iters: 1886, time: 0.064) G_GAN: 0.180 G_GAN_Feat: 0.692 G_ID: 0.092 G_Rec: 0.309 D_GP: 0.028 D_real: 1.100 D_fake: 0.820 +(epoch: 391, iters: 2286, time: 0.064) G_GAN: 0.453 G_GAN_Feat: 0.875 G_ID: 0.122 G_Rec: 0.410 D_GP: 0.033 D_real: 0.998 D_fake: 0.552 +(epoch: 391, iters: 2686, time: 0.064) G_GAN: -0.302 G_GAN_Feat: 0.735 G_ID: 0.092 G_Rec: 0.306 D_GP: 0.035 D_real: 0.602 D_fake: 1.302 +(epoch: 391, iters: 3086, time: 0.064) G_GAN: 0.774 G_GAN_Feat: 1.079 G_ID: 0.116 G_Rec: 0.479 D_GP: 0.112 D_real: 1.048 D_fake: 0.369 +(epoch: 391, iters: 3486, time: 0.064) G_GAN: 0.095 G_GAN_Feat: 0.808 G_ID: 0.099 G_Rec: 0.326 D_GP: 0.047 D_real: 0.754 D_fake: 0.906 +(epoch: 391, iters: 3886, time: 0.063) G_GAN: 0.433 G_GAN_Feat: 1.054 G_ID: 0.118 G_Rec: 0.451 D_GP: 0.053 D_real: 0.892 D_fake: 0.575 +(epoch: 391, iters: 4286, time: 0.063) G_GAN: 0.156 G_GAN_Feat: 0.895 G_ID: 0.129 G_Rec: 0.372 D_GP: 0.051 D_real: 0.766 D_fake: 0.845 +(epoch: 391, iters: 4686, time: 0.063) G_GAN: 0.663 G_GAN_Feat: 1.049 G_ID: 0.125 G_Rec: 0.469 D_GP: 0.052 D_real: 0.930 D_fake: 0.363 +(epoch: 391, iters: 5086, time: 0.064) G_GAN: 0.542 G_GAN_Feat: 0.876 G_ID: 0.083 G_Rec: 0.332 D_GP: 0.050 D_real: 1.201 D_fake: 0.553 +(epoch: 391, iters: 5486, time: 0.063) G_GAN: 0.746 G_GAN_Feat: 1.114 G_ID: 0.121 G_Rec: 0.475 D_GP: 0.053 D_real: 0.878 D_fake: 0.340 +(epoch: 391, iters: 5886, time: 0.063) G_GAN: 0.160 G_GAN_Feat: 0.972 G_ID: 0.092 G_Rec: 0.315 D_GP: 0.112 D_real: 0.285 D_fake: 0.840 +(epoch: 391, iters: 6286, time: 0.063) G_GAN: 0.653 G_GAN_Feat: 1.321 G_ID: 0.115 G_Rec: 0.475 D_GP: 0.172 D_real: 0.254 D_fake: 0.382 +(epoch: 391, iters: 6686, time: 0.064) G_GAN: 0.391 G_GAN_Feat: 0.935 G_ID: 0.099 G_Rec: 0.322 D_GP: 0.045 D_real: 0.490 D_fake: 0.626 +(epoch: 391, iters: 7086, time: 0.063) G_GAN: 0.364 G_GAN_Feat: 1.025 G_ID: 0.133 G_Rec: 0.432 D_GP: 0.045 D_real: 0.711 D_fake: 0.637 +(epoch: 391, iters: 7486, time: 0.063) G_GAN: 0.224 G_GAN_Feat: 0.940 G_ID: 0.103 G_Rec: 0.301 D_GP: 0.079 D_real: 0.324 D_fake: 0.781 +(epoch: 391, iters: 7886, time: 0.063) G_GAN: 0.474 G_GAN_Feat: 1.043 G_ID: 0.147 G_Rec: 0.490 D_GP: 0.033 D_real: 1.014 D_fake: 0.530 +(epoch: 391, iters: 8286, time: 0.064) G_GAN: -0.029 G_GAN_Feat: 0.753 G_ID: 0.109 G_Rec: 0.324 D_GP: 0.030 D_real: 0.851 D_fake: 1.029 +(epoch: 391, iters: 8686, time: 0.063) G_GAN: 0.147 G_GAN_Feat: 1.043 G_ID: 0.109 G_Rec: 0.448 D_GP: 0.084 D_real: 0.460 D_fake: 0.855 +(epoch: 392, iters: 478, time: 0.063) G_GAN: 0.234 G_GAN_Feat: 0.951 G_ID: 0.123 G_Rec: 0.324 D_GP: 0.211 D_real: 0.372 D_fake: 0.774 +(epoch: 392, iters: 878, time: 0.063) G_GAN: 0.851 G_GAN_Feat: 1.320 G_ID: 0.121 G_Rec: 0.544 D_GP: 0.060 D_real: 0.427 D_fake: 0.212 +(epoch: 392, iters: 1278, time: 0.064) G_GAN: 0.422 G_GAN_Feat: 0.879 G_ID: 0.091 G_Rec: 0.288 D_GP: 0.077 D_real: 0.680 D_fake: 0.580 +(epoch: 392, iters: 1678, time: 0.063) G_GAN: 0.485 G_GAN_Feat: 1.001 G_ID: 0.124 G_Rec: 0.506 D_GP: 0.035 D_real: 1.173 D_fake: 0.523 +(epoch: 392, iters: 2078, time: 0.063) G_GAN: 0.093 G_GAN_Feat: 0.717 G_ID: 0.097 G_Rec: 0.302 D_GP: 0.040 D_real: 0.995 D_fake: 0.908 +(epoch: 392, iters: 2478, time: 0.063) G_GAN: 0.424 G_GAN_Feat: 1.116 G_ID: 0.129 G_Rec: 0.436 D_GP: 0.227 D_real: 0.388 D_fake: 0.584 +(epoch: 392, iters: 2878, time: 0.064) G_GAN: 0.426 G_GAN_Feat: 0.852 G_ID: 0.108 G_Rec: 0.399 D_GP: 0.069 D_real: 0.851 D_fake: 0.619 +(epoch: 392, iters: 3278, time: 0.063) G_GAN: 0.342 G_GAN_Feat: 0.932 G_ID: 0.106 G_Rec: 0.441 D_GP: 0.032 D_real: 1.056 D_fake: 0.664 +(epoch: 392, iters: 3678, time: 0.063) G_GAN: 0.218 G_GAN_Feat: 0.825 G_ID: 0.106 G_Rec: 0.323 D_GP: 0.056 D_real: 0.845 D_fake: 0.782 +(epoch: 392, iters: 4078, time: 0.063) G_GAN: 0.466 G_GAN_Feat: 1.183 G_ID: 0.134 G_Rec: 0.473 D_GP: 0.042 D_real: 1.184 D_fake: 0.591 +(epoch: 392, iters: 4478, time: 0.064) G_GAN: 0.207 G_GAN_Feat: 0.676 G_ID: 0.099 G_Rec: 0.304 D_GP: 0.025 D_real: 1.125 D_fake: 0.794 +(epoch: 392, iters: 4878, time: 0.063) G_GAN: 0.271 G_GAN_Feat: 0.916 G_ID: 0.110 G_Rec: 0.475 D_GP: 0.036 D_real: 0.841 D_fake: 0.730 +(epoch: 392, iters: 5278, time: 0.063) G_GAN: -0.082 G_GAN_Feat: 0.677 G_ID: 0.093 G_Rec: 0.309 D_GP: 0.033 D_real: 0.793 D_fake: 1.082 +(epoch: 392, iters: 5678, time: 0.063) G_GAN: 0.108 G_GAN_Feat: 0.953 G_ID: 0.118 G_Rec: 0.461 D_GP: 0.090 D_real: 0.672 D_fake: 0.894 +(epoch: 392, iters: 6078, time: 0.064) G_GAN: 0.061 G_GAN_Feat: 0.658 G_ID: 0.121 G_Rec: 0.298 D_GP: 0.031 D_real: 0.975 D_fake: 0.939 +(epoch: 392, iters: 6478, time: 0.063) G_GAN: 0.324 G_GAN_Feat: 0.862 G_ID: 0.104 G_Rec: 0.399 D_GP: 0.033 D_real: 0.991 D_fake: 0.677 +(epoch: 392, iters: 6878, time: 0.063) G_GAN: -0.030 G_GAN_Feat: 0.850 G_ID: 0.126 G_Rec: 0.356 D_GP: 0.143 D_real: 0.549 D_fake: 1.030 +(epoch: 392, iters: 7278, time: 0.063) G_GAN: 0.469 G_GAN_Feat: 1.013 G_ID: 0.113 G_Rec: 0.424 D_GP: 0.081 D_real: 0.721 D_fake: 0.551 +(epoch: 392, iters: 7678, time: 0.064) G_GAN: 0.224 G_GAN_Feat: 0.717 G_ID: 0.093 G_Rec: 0.307 D_GP: 0.042 D_real: 1.076 D_fake: 0.776 +(epoch: 392, iters: 8078, time: 0.063) G_GAN: 0.283 G_GAN_Feat: 0.921 G_ID: 0.136 G_Rec: 0.408 D_GP: 0.039 D_real: 0.964 D_fake: 0.725 +(epoch: 392, iters: 8478, time: 0.063) G_GAN: 0.333 G_GAN_Feat: 0.859 G_ID: 0.117 G_Rec: 0.326 D_GP: 0.082 D_real: 0.849 D_fake: 0.692 +(epoch: 393, iters: 270, time: 0.063) G_GAN: 0.475 G_GAN_Feat: 0.919 G_ID: 0.122 G_Rec: 0.448 D_GP: 0.032 D_real: 1.162 D_fake: 0.562 +(epoch: 393, iters: 670, time: 0.064) G_GAN: 0.093 G_GAN_Feat: 0.652 G_ID: 0.111 G_Rec: 0.292 D_GP: 0.029 D_real: 1.013 D_fake: 0.908 +(epoch: 393, iters: 1070, time: 0.063) G_GAN: 0.227 G_GAN_Feat: 0.857 G_ID: 0.103 G_Rec: 0.411 D_GP: 0.039 D_real: 0.877 D_fake: 0.774 +(epoch: 393, iters: 1470, time: 0.063) G_GAN: 0.014 G_GAN_Feat: 0.768 G_ID: 0.096 G_Rec: 0.336 D_GP: 0.044 D_real: 0.810 D_fake: 0.986 +(epoch: 393, iters: 1870, time: 0.063) G_GAN: 0.801 G_GAN_Feat: 1.007 G_ID: 0.107 G_Rec: 0.416 D_GP: 0.045 D_real: 1.190 D_fake: 0.318 +(epoch: 393, iters: 2270, time: 0.063) G_GAN: 0.265 G_GAN_Feat: 0.850 G_ID: 0.115 G_Rec: 0.342 D_GP: 0.095 D_real: 0.947 D_fake: 0.757 +(epoch: 393, iters: 2670, time: 0.063) G_GAN: 0.606 G_GAN_Feat: 1.033 G_ID: 0.114 G_Rec: 0.435 D_GP: 0.042 D_real: 1.239 D_fake: 0.401 +(epoch: 393, iters: 3070, time: 0.063) G_GAN: 0.368 G_GAN_Feat: 0.896 G_ID: 0.098 G_Rec: 0.321 D_GP: 0.205 D_real: 0.605 D_fake: 0.642 +(epoch: 393, iters: 3470, time: 0.064) G_GAN: 0.311 G_GAN_Feat: 1.272 G_ID: 0.137 G_Rec: 0.470 D_GP: 0.116 D_real: 0.304 D_fake: 0.690 +(epoch: 393, iters: 3870, time: 0.063) G_GAN: 0.436 G_GAN_Feat: 0.881 G_ID: 0.102 G_Rec: 0.322 D_GP: 0.049 D_real: 0.909 D_fake: 0.568 +(epoch: 393, iters: 4270, time: 0.063) G_GAN: 0.791 G_GAN_Feat: 1.346 G_ID: 0.139 G_Rec: 0.506 D_GP: 0.073 D_real: 0.442 D_fake: 0.261 +(epoch: 393, iters: 4670, time: 0.064) G_GAN: 0.298 G_GAN_Feat: 0.683 G_ID: 0.090 G_Rec: 0.365 D_GP: 0.026 D_real: 1.157 D_fake: 0.702 +(epoch: 393, iters: 5070, time: 0.064) G_GAN: 0.433 G_GAN_Feat: 0.828 G_ID: 0.111 G_Rec: 0.395 D_GP: 0.033 D_real: 1.198 D_fake: 0.582 +(epoch: 393, iters: 5470, time: 0.064) G_GAN: 0.185 G_GAN_Feat: 0.619 G_ID: 0.109 G_Rec: 0.284 D_GP: 0.029 D_real: 1.104 D_fake: 0.817 +(epoch: 393, iters: 5870, time: 0.064) G_GAN: 0.424 G_GAN_Feat: 0.868 G_ID: 0.121 G_Rec: 0.444 D_GP: 0.030 D_real: 1.101 D_fake: 0.589 +(epoch: 393, iters: 6270, time: 0.064) G_GAN: 0.134 G_GAN_Feat: 0.703 G_ID: 0.084 G_Rec: 0.325 D_GP: 0.046 D_real: 1.000 D_fake: 0.866 +(epoch: 393, iters: 6670, time: 0.064) G_GAN: 0.534 G_GAN_Feat: 0.859 G_ID: 0.110 G_Rec: 0.406 D_GP: 0.034 D_real: 1.208 D_fake: 0.514 +(epoch: 393, iters: 7070, time: 0.064) G_GAN: 0.313 G_GAN_Feat: 0.652 G_ID: 0.099 G_Rec: 0.310 D_GP: 0.032 D_real: 1.298 D_fake: 0.688 +(epoch: 393, iters: 7470, time: 0.064) G_GAN: 0.467 G_GAN_Feat: 0.867 G_ID: 0.126 G_Rec: 0.429 D_GP: 0.041 D_real: 1.012 D_fake: 0.554 +(epoch: 393, iters: 7870, time: 0.064) G_GAN: 0.244 G_GAN_Feat: 0.638 G_ID: 0.098 G_Rec: 0.289 D_GP: 0.043 D_real: 1.122 D_fake: 0.756 +(epoch: 393, iters: 8270, time: 0.064) G_GAN: 0.482 G_GAN_Feat: 0.888 G_ID: 0.135 G_Rec: 0.448 D_GP: 0.037 D_real: 1.133 D_fake: 0.536 +(epoch: 393, iters: 8670, time: 0.064) G_GAN: -0.030 G_GAN_Feat: 0.780 G_ID: 0.134 G_Rec: 0.366 D_GP: 0.072 D_real: 0.669 D_fake: 1.030 +(epoch: 394, iters: 462, time: 0.064) G_GAN: 0.425 G_GAN_Feat: 0.892 G_ID: 0.120 G_Rec: 0.431 D_GP: 0.039 D_real: 1.097 D_fake: 0.586 +(epoch: 394, iters: 862, time: 0.064) G_GAN: -0.153 G_GAN_Feat: 0.675 G_ID: 0.103 G_Rec: 0.309 D_GP: 0.040 D_real: 0.729 D_fake: 1.153 +(epoch: 394, iters: 1262, time: 0.064) G_GAN: 0.474 G_GAN_Feat: 0.934 G_ID: 0.112 G_Rec: 0.446 D_GP: 0.049 D_real: 0.943 D_fake: 0.553 +(epoch: 394, iters: 1662, time: 0.064) G_GAN: 0.015 G_GAN_Feat: 0.685 G_ID: 0.096 G_Rec: 0.321 D_GP: 0.037 D_real: 0.879 D_fake: 0.987 +(epoch: 394, iters: 2062, time: 0.064) G_GAN: 0.337 G_GAN_Feat: 0.899 G_ID: 0.119 G_Rec: 0.422 D_GP: 0.060 D_real: 0.895 D_fake: 0.666 +(epoch: 394, iters: 2462, time: 0.064) G_GAN: 0.157 G_GAN_Feat: 0.729 G_ID: 0.106 G_Rec: 0.296 D_GP: 0.082 D_real: 0.750 D_fake: 0.843 +(epoch: 394, iters: 2862, time: 0.064) G_GAN: 0.605 G_GAN_Feat: 0.880 G_ID: 0.103 G_Rec: 0.409 D_GP: 0.038 D_real: 1.254 D_fake: 0.429 +(epoch: 394, iters: 3262, time: 0.064) G_GAN: 0.203 G_GAN_Feat: 0.740 G_ID: 0.117 G_Rec: 0.325 D_GP: 0.043 D_real: 0.845 D_fake: 0.798 +(epoch: 394, iters: 3662, time: 0.064) G_GAN: 0.500 G_GAN_Feat: 0.940 G_ID: 0.126 G_Rec: 0.409 D_GP: 0.128 D_real: 0.796 D_fake: 0.509 +(epoch: 394, iters: 4062, time: 0.064) G_GAN: 0.196 G_GAN_Feat: 0.727 G_ID: 0.120 G_Rec: 0.333 D_GP: 0.045 D_real: 0.953 D_fake: 0.805 +(epoch: 394, iters: 4462, time: 0.064) G_GAN: 0.570 G_GAN_Feat: 0.970 G_ID: 0.109 G_Rec: 0.499 D_GP: 0.069 D_real: 1.076 D_fake: 0.444 +(epoch: 394, iters: 4862, time: 0.063) G_GAN: 0.380 G_GAN_Feat: 0.707 G_ID: 0.112 G_Rec: 0.285 D_GP: 0.045 D_real: 1.170 D_fake: 0.623 +(epoch: 394, iters: 5262, time: 0.063) G_GAN: 0.275 G_GAN_Feat: 0.930 G_ID: 0.130 G_Rec: 0.439 D_GP: 0.054 D_real: 0.841 D_fake: 0.725 +(epoch: 394, iters: 5662, time: 0.063) G_GAN: 0.246 G_GAN_Feat: 0.741 G_ID: 0.103 G_Rec: 0.303 D_GP: 0.051 D_real: 1.085 D_fake: 0.754 +(epoch: 394, iters: 6062, time: 0.064) G_GAN: 0.755 G_GAN_Feat: 1.090 G_ID: 0.119 G_Rec: 0.488 D_GP: 0.137 D_real: 0.802 D_fake: 0.334 +(epoch: 394, iters: 6462, time: 0.063) G_GAN: -0.105 G_GAN_Feat: 0.731 G_ID: 0.107 G_Rec: 0.348 D_GP: 0.041 D_real: 0.852 D_fake: 1.105 +(epoch: 394, iters: 6862, time: 0.063) G_GAN: 0.452 G_GAN_Feat: 0.871 G_ID: 0.116 G_Rec: 0.456 D_GP: 0.027 D_real: 1.262 D_fake: 0.562 +(epoch: 394, iters: 7262, time: 0.063) G_GAN: 0.064 G_GAN_Feat: 0.635 G_ID: 0.083 G_Rec: 0.278 D_GP: 0.030 D_real: 1.014 D_fake: 0.936 +(epoch: 394, iters: 7662, time: 0.064) G_GAN: 0.280 G_GAN_Feat: 0.861 G_ID: 0.107 G_Rec: 0.423 D_GP: 0.033 D_real: 1.003 D_fake: 0.720 +(epoch: 394, iters: 8062, time: 0.063) G_GAN: 0.094 G_GAN_Feat: 0.648 G_ID: 0.108 G_Rec: 0.283 D_GP: 0.038 D_real: 0.983 D_fake: 0.907 +(epoch: 394, iters: 8462, time: 0.063) G_GAN: 0.214 G_GAN_Feat: 0.925 G_ID: 0.130 G_Rec: 0.438 D_GP: 0.052 D_real: 0.854 D_fake: 0.789 +(epoch: 395, iters: 254, time: 0.063) G_GAN: 0.112 G_GAN_Feat: 0.635 G_ID: 0.101 G_Rec: 0.291 D_GP: 0.044 D_real: 0.979 D_fake: 0.888 +(epoch: 395, iters: 654, time: 0.064) G_GAN: 0.043 G_GAN_Feat: 0.887 G_ID: 0.142 G_Rec: 0.418 D_GP: 0.046 D_real: 0.668 D_fake: 0.957 +(epoch: 395, iters: 1054, time: 0.063) G_GAN: 0.186 G_GAN_Feat: 0.717 G_ID: 0.083 G_Rec: 0.309 D_GP: 0.114 D_real: 0.865 D_fake: 0.815 +(epoch: 395, iters: 1454, time: 0.063) G_GAN: 0.800 G_GAN_Feat: 0.999 G_ID: 0.117 G_Rec: 0.470 D_GP: 0.060 D_real: 1.170 D_fake: 0.273 +(epoch: 395, iters: 1854, time: 0.063) G_GAN: 0.346 G_GAN_Feat: 0.699 G_ID: 0.117 G_Rec: 0.281 D_GP: 0.042 D_real: 1.103 D_fake: 0.654 +(epoch: 395, iters: 2254, time: 0.064) G_GAN: 0.445 G_GAN_Feat: 1.001 G_ID: 0.128 G_Rec: 0.476 D_GP: 0.040 D_real: 0.866 D_fake: 0.559 +(epoch: 395, iters: 2654, time: 0.063) G_GAN: 0.204 G_GAN_Feat: 0.723 G_ID: 0.100 G_Rec: 0.325 D_GP: 0.033 D_real: 0.986 D_fake: 0.796 +(epoch: 395, iters: 3054, time: 0.063) G_GAN: 0.551 G_GAN_Feat: 0.946 G_ID: 0.115 G_Rec: 0.446 D_GP: 0.042 D_real: 1.329 D_fake: 0.456 +(epoch: 395, iters: 3454, time: 0.063) G_GAN: 0.204 G_GAN_Feat: 0.669 G_ID: 0.103 G_Rec: 0.290 D_GP: 0.031 D_real: 1.090 D_fake: 0.796 +(epoch: 395, iters: 3854, time: 0.064) G_GAN: 0.375 G_GAN_Feat: 0.896 G_ID: 0.135 G_Rec: 0.400 D_GP: 0.035 D_real: 0.898 D_fake: 0.625 +(epoch: 395, iters: 4254, time: 0.063) G_GAN: 0.453 G_GAN_Feat: 0.869 G_ID: 0.109 G_Rec: 0.329 D_GP: 0.070 D_real: 0.853 D_fake: 0.552 +(epoch: 395, iters: 4654, time: 0.063) G_GAN: 0.340 G_GAN_Feat: 0.993 G_ID: 0.121 G_Rec: 0.427 D_GP: 0.035 D_real: 0.857 D_fake: 0.664 +(epoch: 395, iters: 5054, time: 0.063) G_GAN: 0.559 G_GAN_Feat: 0.747 G_ID: 0.098 G_Rec: 0.290 D_GP: 0.034 D_real: 1.281 D_fake: 0.445 +(epoch: 395, iters: 5454, time: 0.064) G_GAN: 0.145 G_GAN_Feat: 1.023 G_ID: 0.131 G_Rec: 0.481 D_GP: 0.041 D_real: 0.650 D_fake: 0.855 +(epoch: 395, iters: 5854, time: 0.063) G_GAN: 0.361 G_GAN_Feat: 0.714 G_ID: 0.105 G_Rec: 0.307 D_GP: 0.046 D_real: 1.127 D_fake: 0.643 +(epoch: 395, iters: 6254, time: 0.063) G_GAN: 0.288 G_GAN_Feat: 1.067 G_ID: 0.126 G_Rec: 0.494 D_GP: 0.115 D_real: 0.441 D_fake: 0.714 +(epoch: 395, iters: 6654, time: 0.063) G_GAN: 0.189 G_GAN_Feat: 0.754 G_ID: 0.088 G_Rec: 0.292 D_GP: 0.049 D_real: 0.857 D_fake: 0.811 +(epoch: 395, iters: 7054, time: 0.064) G_GAN: 0.911 G_GAN_Feat: 0.999 G_ID: 0.134 G_Rec: 0.448 D_GP: 0.151 D_real: 1.397 D_fake: 0.206 +(epoch: 395, iters: 7454, time: 0.063) G_GAN: 0.183 G_GAN_Feat: 0.927 G_ID: 0.106 G_Rec: 0.338 D_GP: 0.184 D_real: 0.399 D_fake: 0.817 +(epoch: 395, iters: 7854, time: 0.063) G_GAN: 0.492 G_GAN_Feat: 1.107 G_ID: 0.128 G_Rec: 0.441 D_GP: 0.163 D_real: 0.533 D_fake: 0.514 +(epoch: 395, iters: 8254, time: 0.063) G_GAN: 0.190 G_GAN_Feat: 0.936 G_ID: 0.113 G_Rec: 0.336 D_GP: 0.123 D_real: 0.477 D_fake: 0.814 +(epoch: 395, iters: 8654, time: 0.064) G_GAN: 0.779 G_GAN_Feat: 1.140 G_ID: 0.129 G_Rec: 0.466 D_GP: 0.459 D_real: 0.863 D_fake: 0.400 +(epoch: 396, iters: 446, time: 0.063) G_GAN: 0.327 G_GAN_Feat: 0.922 G_ID: 0.113 G_Rec: 0.348 D_GP: 0.072 D_real: 0.465 D_fake: 0.749 +(epoch: 396, iters: 846, time: 0.063) G_GAN: 0.266 G_GAN_Feat: 1.186 G_ID: 0.126 G_Rec: 0.457 D_GP: 0.364 D_real: 0.209 D_fake: 0.746 +(epoch: 396, iters: 1246, time: 0.063) G_GAN: 0.160 G_GAN_Feat: 0.859 G_ID: 0.106 G_Rec: 0.382 D_GP: 0.050 D_real: 0.964 D_fake: 0.851 +(epoch: 396, iters: 1646, time: 0.064) G_GAN: 0.465 G_GAN_Feat: 0.991 G_ID: 0.112 G_Rec: 0.444 D_GP: 0.035 D_real: 0.924 D_fake: 0.539 +(epoch: 396, iters: 2046, time: 0.063) G_GAN: 0.291 G_GAN_Feat: 0.714 G_ID: 0.086 G_Rec: 0.269 D_GP: 0.036 D_real: 1.198 D_fake: 0.711 +(epoch: 396, iters: 2446, time: 0.063) G_GAN: 0.686 G_GAN_Feat: 0.929 G_ID: 0.129 G_Rec: 0.418 D_GP: 0.035 D_real: 1.287 D_fake: 0.368 +(epoch: 396, iters: 2846, time: 0.063) G_GAN: 0.249 G_GAN_Feat: 0.716 G_ID: 0.096 G_Rec: 0.288 D_GP: 0.029 D_real: 1.135 D_fake: 0.752 +(epoch: 396, iters: 3246, time: 0.064) G_GAN: 0.381 G_GAN_Feat: 0.999 G_ID: 0.122 G_Rec: 0.386 D_GP: 0.055 D_real: 0.776 D_fake: 0.620 +(epoch: 396, iters: 3646, time: 0.063) G_GAN: 0.288 G_GAN_Feat: 0.744 G_ID: 0.110 G_Rec: 0.290 D_GP: 0.040 D_real: 1.132 D_fake: 0.712 +(epoch: 396, iters: 4046, time: 0.063) G_GAN: 0.768 G_GAN_Feat: 1.230 G_ID: 0.120 G_Rec: 0.440 D_GP: 0.116 D_real: 0.589 D_fake: 0.298 +(epoch: 396, iters: 4446, time: 0.063) G_GAN: 0.517 G_GAN_Feat: 0.772 G_ID: 0.095 G_Rec: 0.311 D_GP: 0.033 D_real: 1.348 D_fake: 0.488 +(epoch: 396, iters: 4846, time: 0.064) G_GAN: 0.599 G_GAN_Feat: 0.958 G_ID: 0.106 G_Rec: 0.463 D_GP: 0.040 D_real: 1.338 D_fake: 0.405 +(epoch: 396, iters: 5246, time: 0.063) G_GAN: 0.499 G_GAN_Feat: 0.777 G_ID: 0.102 G_Rec: 0.302 D_GP: 0.044 D_real: 1.269 D_fake: 0.522 +(epoch: 396, iters: 5646, time: 0.063) G_GAN: 0.647 G_GAN_Feat: 1.050 G_ID: 0.122 G_Rec: 0.467 D_GP: 0.044 D_real: 1.031 D_fake: 0.368 +(epoch: 396, iters: 6046, time: 0.063) G_GAN: 0.481 G_GAN_Feat: 1.134 G_ID: 0.096 G_Rec: 0.347 D_GP: 0.625 D_real: 1.176 D_fake: 0.585 +(epoch: 396, iters: 6446, time: 0.064) G_GAN: 0.086 G_GAN_Feat: 0.908 G_ID: 0.114 G_Rec: 0.447 D_GP: 0.027 D_real: 0.849 D_fake: 0.918 +(epoch: 396, iters: 6846, time: 0.063) G_GAN: -0.146 G_GAN_Feat: 0.684 G_ID: 0.090 G_Rec: 0.309 D_GP: 0.028 D_real: 0.775 D_fake: 1.146 +(epoch: 396, iters: 7246, time: 0.063) G_GAN: 0.062 G_GAN_Feat: 0.980 G_ID: 0.116 G_Rec: 0.474 D_GP: 0.037 D_real: 0.670 D_fake: 0.939 +(epoch: 396, iters: 7646, time: 0.063) G_GAN: -0.055 G_GAN_Feat: 0.762 G_ID: 0.092 G_Rec: 0.312 D_GP: 0.039 D_real: 0.791 D_fake: 1.055 +(epoch: 396, iters: 8046, time: 0.064) G_GAN: -0.025 G_GAN_Feat: 1.147 G_ID: 0.115 G_Rec: 0.465 D_GP: 0.401 D_real: 0.233 D_fake: 1.026 +(epoch: 396, iters: 8446, time: 0.063) G_GAN: 0.034 G_GAN_Feat: 0.757 G_ID: 0.101 G_Rec: 0.300 D_GP: 0.056 D_real: 0.822 D_fake: 0.966 +(epoch: 397, iters: 238, time: 0.063) G_GAN: 0.258 G_GAN_Feat: 1.038 G_ID: 0.128 G_Rec: 0.427 D_GP: 0.096 D_real: 0.515 D_fake: 0.742 +(epoch: 397, iters: 638, time: 0.063) G_GAN: 0.152 G_GAN_Feat: 0.839 G_ID: 0.100 G_Rec: 0.347 D_GP: 0.035 D_real: 0.865 D_fake: 0.848 +(epoch: 397, iters: 1038, time: 0.064) G_GAN: 0.664 G_GAN_Feat: 1.157 G_ID: 0.117 G_Rec: 0.441 D_GP: 0.051 D_real: 0.793 D_fake: 0.393 +(epoch: 397, iters: 1438, time: 0.063) G_GAN: 0.247 G_GAN_Feat: 0.812 G_ID: 0.094 G_Rec: 0.334 D_GP: 0.063 D_real: 0.957 D_fake: 0.754 +(epoch: 397, iters: 1838, time: 0.063) G_GAN: 0.679 G_GAN_Feat: 0.887 G_ID: 0.130 G_Rec: 0.447 D_GP: 0.039 D_real: 1.282 D_fake: 0.357 +(epoch: 397, iters: 2238, time: 0.063) G_GAN: 0.049 G_GAN_Feat: 0.795 G_ID: 0.096 G_Rec: 0.327 D_GP: 0.054 D_real: 0.840 D_fake: 0.951 +(epoch: 397, iters: 2638, time: 0.064) G_GAN: 0.467 G_GAN_Feat: 1.017 G_ID: 0.134 G_Rec: 0.455 D_GP: 0.057 D_real: 0.869 D_fake: 0.564 +(epoch: 397, iters: 3038, time: 0.063) G_GAN: 0.238 G_GAN_Feat: 0.933 G_ID: 0.117 G_Rec: 0.355 D_GP: 0.101 D_real: 0.678 D_fake: 0.764 +(epoch: 397, iters: 3438, time: 0.063) G_GAN: 0.580 G_GAN_Feat: 1.003 G_ID: 0.139 G_Rec: 0.458 D_GP: 0.047 D_real: 1.116 D_fake: 0.434 +(epoch: 397, iters: 3838, time: 0.063) G_GAN: 0.399 G_GAN_Feat: 0.749 G_ID: 0.093 G_Rec: 0.282 D_GP: 0.037 D_real: 1.222 D_fake: 0.604 +(epoch: 397, iters: 4238, time: 0.064) G_GAN: 0.920 G_GAN_Feat: 0.983 G_ID: 0.130 G_Rec: 0.446 D_GP: 0.053 D_real: 1.527 D_fake: 0.224 +(epoch: 397, iters: 4638, time: 0.063) G_GAN: 0.371 G_GAN_Feat: 0.661 G_ID: 0.084 G_Rec: 0.279 D_GP: 0.029 D_real: 1.244 D_fake: 0.633 +(epoch: 397, iters: 5038, time: 0.063) G_GAN: 0.137 G_GAN_Feat: 0.994 G_ID: 0.121 G_Rec: 0.446 D_GP: 0.043 D_real: 0.507 D_fake: 0.863 +(epoch: 397, iters: 5438, time: 0.063) G_GAN: 0.168 G_GAN_Feat: 0.895 G_ID: 0.095 G_Rec: 0.328 D_GP: 0.095 D_real: 0.396 D_fake: 0.837 +(epoch: 397, iters: 5838, time: 0.064) G_GAN: 0.662 G_GAN_Feat: 0.951 G_ID: 0.115 G_Rec: 0.418 D_GP: 0.031 D_real: 1.277 D_fake: 0.357 +(epoch: 397, iters: 6238, time: 0.063) G_GAN: 0.360 G_GAN_Feat: 0.877 G_ID: 0.093 G_Rec: 0.331 D_GP: 0.074 D_real: 0.613 D_fake: 0.691 +(epoch: 397, iters: 6638, time: 0.063) G_GAN: 0.375 G_GAN_Feat: 0.921 G_ID: 0.118 G_Rec: 0.427 D_GP: 0.034 D_real: 1.017 D_fake: 0.627 +(epoch: 397, iters: 7038, time: 0.063) G_GAN: 0.519 G_GAN_Feat: 0.729 G_ID: 0.101 G_Rec: 0.355 D_GP: 0.030 D_real: 1.350 D_fake: 0.496 +(epoch: 397, iters: 7438, time: 0.063) G_GAN: 0.733 G_GAN_Feat: 1.051 G_ID: 0.118 G_Rec: 0.448 D_GP: 0.068 D_real: 1.002 D_fake: 0.305 +(epoch: 397, iters: 7838, time: 0.063) G_GAN: 0.233 G_GAN_Feat: 0.801 G_ID: 0.115 G_Rec: 0.321 D_GP: 0.041 D_real: 0.803 D_fake: 0.771 +(epoch: 397, iters: 8238, time: 0.063) G_GAN: 0.234 G_GAN_Feat: 0.973 G_ID: 0.139 G_Rec: 0.462 D_GP: 0.065 D_real: 0.908 D_fake: 0.767 +(epoch: 397, iters: 8638, time: 0.064) G_GAN: 0.150 G_GAN_Feat: 0.740 G_ID: 0.093 G_Rec: 0.316 D_GP: 0.057 D_real: 1.092 D_fake: 0.850 +(epoch: 398, iters: 430, time: 0.063) G_GAN: 0.269 G_GAN_Feat: 0.981 G_ID: 0.108 G_Rec: 0.437 D_GP: 0.063 D_real: 0.891 D_fake: 0.732 +(epoch: 398, iters: 830, time: 0.063) G_GAN: 0.181 G_GAN_Feat: 0.699 G_ID: 0.093 G_Rec: 0.295 D_GP: 0.042 D_real: 1.148 D_fake: 0.821 +(epoch: 398, iters: 1230, time: 0.063) G_GAN: 0.347 G_GAN_Feat: 0.984 G_ID: 0.121 G_Rec: 0.459 D_GP: 0.067 D_real: 0.886 D_fake: 0.660 +(epoch: 398, iters: 1630, time: 0.064) G_GAN: 0.013 G_GAN_Feat: 0.816 G_ID: 0.104 G_Rec: 0.349 D_GP: 0.069 D_real: 0.700 D_fake: 0.987 +(epoch: 398, iters: 2030, time: 0.063) G_GAN: 0.285 G_GAN_Feat: 0.923 G_ID: 0.125 G_Rec: 0.427 D_GP: 0.054 D_real: 0.858 D_fake: 0.725 +(epoch: 398, iters: 2430, time: 0.063) G_GAN: 0.157 G_GAN_Feat: 0.736 G_ID: 0.092 G_Rec: 0.330 D_GP: 0.048 D_real: 0.984 D_fake: 0.844 +(epoch: 398, iters: 2830, time: 0.063) G_GAN: 0.218 G_GAN_Feat: 0.933 G_ID: 0.123 G_Rec: 0.417 D_GP: 0.102 D_real: 0.787 D_fake: 0.783 +(epoch: 398, iters: 3230, time: 0.064) G_GAN: 0.071 G_GAN_Feat: 0.703 G_ID: 0.106 G_Rec: 0.304 D_GP: 0.039 D_real: 0.881 D_fake: 0.929 +(epoch: 398, iters: 3630, time: 0.063) G_GAN: 0.533 G_GAN_Feat: 0.993 G_ID: 0.124 G_Rec: 0.466 D_GP: 0.050 D_real: 1.028 D_fake: 0.486 +(epoch: 398, iters: 4030, time: 0.063) G_GAN: 0.147 G_GAN_Feat: 0.774 G_ID: 0.119 G_Rec: 0.337 D_GP: 0.043 D_real: 0.986 D_fake: 0.854 +(epoch: 398, iters: 4430, time: 0.063) G_GAN: 0.409 G_GAN_Feat: 1.007 G_ID: 0.102 G_Rec: 0.428 D_GP: 0.048 D_real: 1.181 D_fake: 0.608 +(epoch: 398, iters: 4830, time: 0.064) G_GAN: 0.121 G_GAN_Feat: 0.772 G_ID: 0.106 G_Rec: 0.305 D_GP: 0.052 D_real: 0.863 D_fake: 0.879 +(epoch: 398, iters: 5230, time: 0.063) G_GAN: 0.209 G_GAN_Feat: 0.949 G_ID: 0.125 G_Rec: 0.420 D_GP: 0.051 D_real: 0.804 D_fake: 0.794 +(epoch: 398, iters: 5630, time: 0.063) G_GAN: 0.212 G_GAN_Feat: 0.736 G_ID: 0.117 G_Rec: 0.296 D_GP: 0.036 D_real: 1.050 D_fake: 0.790 +(epoch: 398, iters: 6030, time: 0.063) G_GAN: 0.381 G_GAN_Feat: 0.915 G_ID: 0.101 G_Rec: 0.438 D_GP: 0.038 D_real: 0.969 D_fake: 0.620 +(epoch: 398, iters: 6430, time: 0.064) G_GAN: 0.014 G_GAN_Feat: 0.805 G_ID: 0.110 G_Rec: 0.336 D_GP: 0.075 D_real: 0.789 D_fake: 0.986 +(epoch: 398, iters: 6830, time: 0.063) G_GAN: 0.463 G_GAN_Feat: 0.970 G_ID: 0.108 G_Rec: 0.443 D_GP: 0.041 D_real: 0.900 D_fake: 0.548 +(epoch: 398, iters: 7230, time: 0.063) G_GAN: 0.070 G_GAN_Feat: 0.793 G_ID: 0.094 G_Rec: 0.321 D_GP: 0.052 D_real: 0.658 D_fake: 0.930 +(epoch: 398, iters: 7630, time: 0.063) G_GAN: 0.663 G_GAN_Feat: 1.020 G_ID: 0.105 G_Rec: 0.431 D_GP: 0.036 D_real: 1.196 D_fake: 0.399 +(epoch: 398, iters: 8030, time: 0.064) G_GAN: 0.155 G_GAN_Feat: 0.803 G_ID: 0.099 G_Rec: 0.310 D_GP: 0.049 D_real: 0.801 D_fake: 0.845 +(epoch: 398, iters: 8430, time: 0.063) G_GAN: 0.538 G_GAN_Feat: 1.030 G_ID: 0.115 G_Rec: 0.437 D_GP: 0.330 D_real: 0.752 D_fake: 0.475 +(epoch: 399, iters: 222, time: 0.063) G_GAN: 0.085 G_GAN_Feat: 0.759 G_ID: 0.097 G_Rec: 0.306 D_GP: 0.050 D_real: 0.793 D_fake: 0.915 +(epoch: 399, iters: 622, time: 0.063) G_GAN: 0.490 G_GAN_Feat: 0.933 G_ID: 0.129 G_Rec: 0.412 D_GP: 0.037 D_real: 1.099 D_fake: 0.513 +(epoch: 399, iters: 1022, time: 0.064) G_GAN: 0.369 G_GAN_Feat: 0.867 G_ID: 0.105 G_Rec: 0.334 D_GP: 0.121 D_real: 0.831 D_fake: 0.632 +(epoch: 399, iters: 1422, time: 0.063) G_GAN: 0.406 G_GAN_Feat: 1.030 G_ID: 0.119 G_Rec: 0.432 D_GP: 0.043 D_real: 0.869 D_fake: 0.597 +(epoch: 399, iters: 1822, time: 0.063) G_GAN: -0.107 G_GAN_Feat: 0.755 G_ID: 0.101 G_Rec: 0.294 D_GP: 0.035 D_real: 0.694 D_fake: 1.107 +(epoch: 399, iters: 2222, time: 0.063) G_GAN: 0.379 G_GAN_Feat: 0.967 G_ID: 0.129 G_Rec: 0.457 D_GP: 0.035 D_real: 1.036 D_fake: 0.622 +(epoch: 399, iters: 2622, time: 0.064) G_GAN: 0.397 G_GAN_Feat: 0.793 G_ID: 0.095 G_Rec: 0.302 D_GP: 0.034 D_real: 1.299 D_fake: 0.653 +(epoch: 399, iters: 3022, time: 0.063) G_GAN: 0.218 G_GAN_Feat: 0.920 G_ID: 0.123 G_Rec: 0.399 D_GP: 0.042 D_real: 0.828 D_fake: 0.784 +(epoch: 399, iters: 3422, time: 0.063) G_GAN: 0.296 G_GAN_Feat: 0.803 G_ID: 0.091 G_Rec: 0.307 D_GP: 0.051 D_real: 0.973 D_fake: 0.704 +(epoch: 399, iters: 3822, time: 0.063) G_GAN: 0.898 G_GAN_Feat: 1.011 G_ID: 0.123 G_Rec: 0.436 D_GP: 0.045 D_real: 1.359 D_fake: 0.213 +(epoch: 399, iters: 4222, time: 0.064) G_GAN: 0.323 G_GAN_Feat: 0.802 G_ID: 0.102 G_Rec: 0.318 D_GP: 0.039 D_real: 1.048 D_fake: 0.683 +(epoch: 399, iters: 4622, time: 0.063) G_GAN: 0.558 G_GAN_Feat: 1.073 G_ID: 0.123 G_Rec: 0.452 D_GP: 0.117 D_real: 0.694 D_fake: 0.451 +(epoch: 399, iters: 5022, time: 0.063) G_GAN: 0.602 G_GAN_Feat: 0.828 G_ID: 0.092 G_Rec: 0.272 D_GP: 0.038 D_real: 1.167 D_fake: 0.412 +(epoch: 399, iters: 5422, time: 0.063) G_GAN: 0.619 G_GAN_Feat: 0.975 G_ID: 0.138 G_Rec: 0.433 D_GP: 0.033 D_real: 1.271 D_fake: 0.390 +(epoch: 399, iters: 5822, time: 0.064) G_GAN: 0.387 G_GAN_Feat: 0.896 G_ID: 0.103 G_Rec: 0.335 D_GP: 0.063 D_real: 0.795 D_fake: 0.748 +(epoch: 399, iters: 6222, time: 0.063) G_GAN: 0.607 G_GAN_Feat: 1.048 G_ID: 0.114 G_Rec: 0.435 D_GP: 0.040 D_real: 1.089 D_fake: 0.408 +(epoch: 399, iters: 6622, time: 0.063) G_GAN: 0.261 G_GAN_Feat: 1.030 G_ID: 0.108 G_Rec: 0.365 D_GP: 0.031 D_real: 1.354 D_fake: 0.775 +(epoch: 399, iters: 7022, time: 0.063) G_GAN: 0.156 G_GAN_Feat: 0.807 G_ID: 0.136 G_Rec: 0.420 D_GP: 0.026 D_real: 0.921 D_fake: 0.844 +(epoch: 399, iters: 7422, time: 0.064) G_GAN: 0.083 G_GAN_Feat: 0.672 G_ID: 0.100 G_Rec: 0.362 D_GP: 0.025 D_real: 0.986 D_fake: 0.917 +(epoch: 399, iters: 7822, time: 0.063) G_GAN: 0.238 G_GAN_Feat: 0.843 G_ID: 0.122 G_Rec: 0.430 D_GP: 0.030 D_real: 0.895 D_fake: 0.763 +(epoch: 399, iters: 8222, time: 0.063) G_GAN: -0.026 G_GAN_Feat: 0.635 G_ID: 0.102 G_Rec: 0.319 D_GP: 0.035 D_real: 0.844 D_fake: 1.026 +(epoch: 399, iters: 8622, time: 0.063) G_GAN: -0.030 G_GAN_Feat: 0.879 G_ID: 0.156 G_Rec: 0.442 D_GP: 0.055 D_real: 0.561 D_fake: 1.031 +(epoch: 400, iters: 414, time: 0.064) G_GAN: -0.273 G_GAN_Feat: 0.859 G_ID: 0.118 G_Rec: 0.355 D_GP: 0.091 D_real: 0.374 D_fake: 1.273 +(epoch: 400, iters: 814, time: 0.063) G_GAN: 0.541 G_GAN_Feat: 0.848 G_ID: 0.106 G_Rec: 0.404 D_GP: 0.036 D_real: 1.260 D_fake: 0.514 +(epoch: 400, iters: 1214, time: 0.063) G_GAN: 0.054 G_GAN_Feat: 0.699 G_ID: 0.104 G_Rec: 0.296 D_GP: 0.037 D_real: 0.966 D_fake: 0.946 +(epoch: 400, iters: 1614, time: 0.063) G_GAN: 0.619 G_GAN_Feat: 1.016 G_ID: 0.120 G_Rec: 0.440 D_GP: 0.042 D_real: 1.027 D_fake: 0.417 +(epoch: 400, iters: 2014, time: 0.064) G_GAN: 0.132 G_GAN_Feat: 0.592 G_ID: 0.092 G_Rec: 0.273 D_GP: 0.027 D_real: 1.076 D_fake: 0.869 +(epoch: 400, iters: 2414, time: 0.063) G_GAN: 0.347 G_GAN_Feat: 0.873 G_ID: 0.105 G_Rec: 0.445 D_GP: 0.036 D_real: 1.002 D_fake: 0.668 +(epoch: 400, iters: 2814, time: 0.063) G_GAN: -0.021 G_GAN_Feat: 0.617 G_ID: 0.091 G_Rec: 0.305 D_GP: 0.032 D_real: 0.796 D_fake: 1.021 +(epoch: 400, iters: 3214, time: 0.063) G_GAN: 0.339 G_GAN_Feat: 0.872 G_ID: 0.127 G_Rec: 0.411 D_GP: 0.077 D_real: 0.914 D_fake: 0.687 +(epoch: 400, iters: 3614, time: 0.064) G_GAN: 0.100 G_GAN_Feat: 0.662 G_ID: 0.093 G_Rec: 0.351 D_GP: 0.035 D_real: 0.986 D_fake: 0.901 +(epoch: 400, iters: 4014, time: 0.063) G_GAN: 0.268 G_GAN_Feat: 1.007 G_ID: 0.118 G_Rec: 0.459 D_GP: 0.107 D_real: 0.775 D_fake: 0.742 +(epoch: 400, iters: 4414, time: 0.063) G_GAN: -0.157 G_GAN_Feat: 0.657 G_ID: 0.104 G_Rec: 0.320 D_GP: 0.029 D_real: 0.694 D_fake: 1.157 +(epoch: 400, iters: 4814, time: 0.063) G_GAN: 0.149 G_GAN_Feat: 1.035 G_ID: 0.137 G_Rec: 0.466 D_GP: 0.056 D_real: 0.547 D_fake: 0.854 +(epoch: 400, iters: 5214, time: 0.064) G_GAN: 0.367 G_GAN_Feat: 0.641 G_ID: 0.093 G_Rec: 0.262 D_GP: 0.033 D_real: 1.224 D_fake: 0.641 +(epoch: 400, iters: 5614, time: 0.063) G_GAN: 0.336 G_GAN_Feat: 0.999 G_ID: 0.132 G_Rec: 0.429 D_GP: 0.155 D_real: 0.761 D_fake: 0.672 +(epoch: 400, iters: 6014, time: 0.063) G_GAN: 0.041 G_GAN_Feat: 0.836 G_ID: 0.117 G_Rec: 0.343 D_GP: 0.096 D_real: 0.682 D_fake: 0.959 +(epoch: 400, iters: 6414, time: 0.063) G_GAN: 0.282 G_GAN_Feat: 1.012 G_ID: 0.121 G_Rec: 0.426 D_GP: 0.075 D_real: 0.553 D_fake: 0.719 +(epoch: 400, iters: 6814, time: 0.064) G_GAN: 0.058 G_GAN_Feat: 0.775 G_ID: 0.086 G_Rec: 0.291 D_GP: 0.040 D_real: 0.743 D_fake: 0.942 +(epoch: 400, iters: 7214, time: 0.063) G_GAN: 0.519 G_GAN_Feat: 0.868 G_ID: 0.134 G_Rec: 0.438 D_GP: 0.029 D_real: 1.245 D_fake: 0.509 +(epoch: 400, iters: 7614, time: 0.063) G_GAN: 0.317 G_GAN_Feat: 0.679 G_ID: 0.097 G_Rec: 0.310 D_GP: 0.030 D_real: 1.207 D_fake: 0.683 +(epoch: 400, iters: 8014, time: 0.063) G_GAN: 0.424 G_GAN_Feat: 0.842 G_ID: 0.121 G_Rec: 0.381 D_GP: 0.035 D_real: 1.034 D_fake: 0.581 +(epoch: 400, iters: 8414, time: 0.064) G_GAN: 0.245 G_GAN_Feat: 0.687 G_ID: 0.122 G_Rec: 0.285 D_GP: 0.037 D_real: 1.101 D_fake: 0.755 +(epoch: 401, iters: 206, time: 0.064) G_GAN: 0.659 G_GAN_Feat: 0.965 G_ID: 0.101 G_Rec: 0.435 D_GP: 0.032 D_real: 1.337 D_fake: 0.349 +(epoch: 401, iters: 606, time: 0.063) G_GAN: 0.376 G_GAN_Feat: 0.890 G_ID: 0.108 G_Rec: 0.363 D_GP: 0.115 D_real: 0.576 D_fake: 0.650 +(epoch: 401, iters: 1006, time: 0.063) G_GAN: 0.886 G_GAN_Feat: 1.085 G_ID: 0.130 G_Rec: 0.426 D_GP: 0.146 D_real: 0.897 D_fake: 0.349 +(epoch: 401, iters: 1406, time: 0.064) G_GAN: 0.307 G_GAN_Feat: 0.744 G_ID: 0.114 G_Rec: 0.308 D_GP: 0.028 D_real: 1.186 D_fake: 0.694 +(epoch: 401, iters: 1806, time: 0.063) G_GAN: 0.896 G_GAN_Feat: 1.130 G_ID: 0.112 G_Rec: 0.478 D_GP: 0.129 D_real: 0.765 D_fake: 0.211 +(epoch: 401, iters: 2206, time: 0.063) G_GAN: 0.226 G_GAN_Feat: 1.001 G_ID: 0.097 G_Rec: 0.338 D_GP: 0.809 D_real: 0.295 D_fake: 0.779 +(epoch: 401, iters: 2606, time: 0.063) G_GAN: 0.119 G_GAN_Feat: 0.951 G_ID: 0.134 G_Rec: 0.439 D_GP: 0.060 D_real: 0.643 D_fake: 0.881 +(epoch: 401, iters: 3006, time: 0.064) G_GAN: 0.239 G_GAN_Feat: 0.765 G_ID: 0.102 G_Rec: 0.324 D_GP: 0.032 D_real: 1.131 D_fake: 0.761 +(epoch: 401, iters: 3406, time: 0.063) G_GAN: 0.476 G_GAN_Feat: 0.955 G_ID: 0.117 G_Rec: 0.412 D_GP: 0.040 D_real: 1.047 D_fake: 0.531 +(epoch: 401, iters: 3806, time: 0.063) G_GAN: 0.271 G_GAN_Feat: 0.725 G_ID: 0.101 G_Rec: 0.297 D_GP: 0.036 D_real: 1.106 D_fake: 0.730 +(epoch: 401, iters: 4206, time: 0.063) G_GAN: 0.541 G_GAN_Feat: 0.917 G_ID: 0.127 G_Rec: 0.450 D_GP: 0.033 D_real: 1.237 D_fake: 0.470 +(epoch: 401, iters: 4606, time: 0.064) G_GAN: 0.198 G_GAN_Feat: 0.811 G_ID: 0.098 G_Rec: 0.318 D_GP: 0.037 D_real: 0.871 D_fake: 0.802 +(epoch: 401, iters: 5006, time: 0.063) G_GAN: 0.710 G_GAN_Feat: 0.976 G_ID: 0.123 G_Rec: 0.385 D_GP: 0.046 D_real: 1.122 D_fake: 0.308 +(epoch: 401, iters: 5406, time: 0.063) G_GAN: 0.168 G_GAN_Feat: 0.722 G_ID: 0.116 G_Rec: 0.301 D_GP: 0.038 D_real: 1.045 D_fake: 0.836 +(epoch: 401, iters: 5806, time: 0.063) G_GAN: 0.307 G_GAN_Feat: 1.036 G_ID: 0.149 G_Rec: 0.472 D_GP: 0.285 D_real: 0.616 D_fake: 0.702 +(epoch: 401, iters: 6206, time: 0.064) G_GAN: 0.405 G_GAN_Feat: 0.738 G_ID: 0.099 G_Rec: 0.327 D_GP: 0.038 D_real: 1.207 D_fake: 0.596 +(epoch: 401, iters: 6606, time: 0.063) G_GAN: 0.672 G_GAN_Feat: 1.126 G_ID: 0.122 G_Rec: 0.483 D_GP: 0.154 D_real: 0.500 D_fake: 0.373 +(epoch: 401, iters: 7006, time: 0.063) G_GAN: 0.284 G_GAN_Feat: 0.795 G_ID: 0.108 G_Rec: 0.344 D_GP: 0.038 D_real: 1.101 D_fake: 0.716 +(epoch: 401, iters: 7406, time: 0.063) G_GAN: 0.673 G_GAN_Feat: 0.899 G_ID: 0.116 G_Rec: 0.390 D_GP: 0.041 D_real: 1.323 D_fake: 0.365 +(epoch: 401, iters: 7806, time: 0.064) G_GAN: 0.362 G_GAN_Feat: 0.879 G_ID: 0.088 G_Rec: 0.352 D_GP: 0.054 D_real: 0.818 D_fake: 0.638 +(epoch: 401, iters: 8206, time: 0.063) G_GAN: 0.626 G_GAN_Feat: 1.027 G_ID: 0.121 G_Rec: 0.434 D_GP: 0.032 D_real: 1.125 D_fake: 0.425 +(epoch: 401, iters: 8606, time: 0.063) G_GAN: -0.029 G_GAN_Feat: 0.964 G_ID: 0.110 G_Rec: 0.351 D_GP: 0.214 D_real: 0.426 D_fake: 1.029 +(epoch: 402, iters: 398, time: 0.063) G_GAN: 0.343 G_GAN_Feat: 0.909 G_ID: 0.130 G_Rec: 0.429 D_GP: 0.030 D_real: 1.018 D_fake: 0.660 +(epoch: 402, iters: 798, time: 0.064) G_GAN: 0.090 G_GAN_Feat: 0.684 G_ID: 0.105 G_Rec: 0.311 D_GP: 0.027 D_real: 1.031 D_fake: 0.910 +(epoch: 402, iters: 1198, time: 0.063) G_GAN: 0.425 G_GAN_Feat: 0.966 G_ID: 0.119 G_Rec: 0.457 D_GP: 0.045 D_real: 0.892 D_fake: 0.610 +(epoch: 402, iters: 1598, time: 0.063) G_GAN: 0.100 G_GAN_Feat: 0.681 G_ID: 0.091 G_Rec: 0.303 D_GP: 0.037 D_real: 1.020 D_fake: 0.900 +(epoch: 402, iters: 1998, time: 0.063) G_GAN: 0.323 G_GAN_Feat: 0.868 G_ID: 0.110 G_Rec: 0.435 D_GP: 0.041 D_real: 1.002 D_fake: 0.692 +(epoch: 402, iters: 2398, time: 0.064) G_GAN: 0.047 G_GAN_Feat: 0.731 G_ID: 0.098 G_Rec: 0.306 D_GP: 0.041 D_real: 0.830 D_fake: 0.953 +(epoch: 402, iters: 2798, time: 0.063) G_GAN: 0.556 G_GAN_Feat: 0.961 G_ID: 0.108 G_Rec: 0.476 D_GP: 0.036 D_real: 1.260 D_fake: 0.484 +(epoch: 402, iters: 3198, time: 0.063) G_GAN: 0.004 G_GAN_Feat: 0.928 G_ID: 0.118 G_Rec: 0.387 D_GP: 0.556 D_real: 0.441 D_fake: 0.997 +(epoch: 402, iters: 3598, time: 0.063) G_GAN: 0.412 G_GAN_Feat: 1.053 G_ID: 0.104 G_Rec: 0.500 D_GP: 0.040 D_real: 0.857 D_fake: 0.592 +(epoch: 402, iters: 3998, time: 0.063) G_GAN: 0.405 G_GAN_Feat: 0.700 G_ID: 0.091 G_Rec: 0.316 D_GP: 0.049 D_real: 1.268 D_fake: 0.604 +(epoch: 402, iters: 4398, time: 0.063) G_GAN: 0.469 G_GAN_Feat: 0.978 G_ID: 0.114 G_Rec: 0.417 D_GP: 0.113 D_real: 0.904 D_fake: 0.538 +(epoch: 402, iters: 4798, time: 0.063) G_GAN: 0.224 G_GAN_Feat: 0.775 G_ID: 0.103 G_Rec: 0.317 D_GP: 0.088 D_real: 0.867 D_fake: 0.777 +(epoch: 402, iters: 5198, time: 0.064) G_GAN: 0.730 G_GAN_Feat: 0.993 G_ID: 0.135 G_Rec: 0.469 D_GP: 0.041 D_real: 1.173 D_fake: 0.303 +(epoch: 402, iters: 5598, time: 0.063) G_GAN: 0.145 G_GAN_Feat: 0.745 G_ID: 0.122 G_Rec: 0.313 D_GP: 0.036 D_real: 0.932 D_fake: 0.855 +(epoch: 402, iters: 5998, time: 0.063) G_GAN: 0.696 G_GAN_Feat: 1.016 G_ID: 0.118 G_Rec: 0.440 D_GP: 0.545 D_real: 0.854 D_fake: 0.346 +(epoch: 402, iters: 6398, time: 0.063) G_GAN: 0.220 G_GAN_Feat: 0.739 G_ID: 0.111 G_Rec: 0.299 D_GP: 0.038 D_real: 1.044 D_fake: 0.780 +(epoch: 402, iters: 6798, time: 0.064) G_GAN: 0.557 G_GAN_Feat: 1.016 G_ID: 0.127 G_Rec: 0.466 D_GP: 0.047 D_real: 1.023 D_fake: 0.450 +(epoch: 402, iters: 7198, time: 0.063) G_GAN: 0.416 G_GAN_Feat: 0.855 G_ID: 0.097 G_Rec: 0.314 D_GP: 0.080 D_real: 0.793 D_fake: 0.591 +(epoch: 402, iters: 7598, time: 0.063) G_GAN: 0.257 G_GAN_Feat: 1.197 G_ID: 0.124 G_Rec: 0.510 D_GP: 0.255 D_real: 0.240 D_fake: 0.749 +(epoch: 402, iters: 7998, time: 0.063) G_GAN: 0.298 G_GAN_Feat: 0.858 G_ID: 0.119 G_Rec: 0.356 D_GP: 0.049 D_real: 0.969 D_fake: 0.712 +(epoch: 402, iters: 8398, time: 0.064) G_GAN: 0.423 G_GAN_Feat: 1.039 G_ID: 0.113 G_Rec: 0.420 D_GP: 0.050 D_real: 0.581 D_fake: 0.580 +(epoch: 403, iters: 190, time: 0.063) G_GAN: 0.673 G_GAN_Feat: 1.016 G_ID: 0.110 G_Rec: 0.318 D_GP: 0.484 D_real: 0.566 D_fake: 0.386 +(epoch: 403, iters: 590, time: 0.063) G_GAN: 0.966 G_GAN_Feat: 1.084 G_ID: 0.120 G_Rec: 0.471 D_GP: 0.039 D_real: 1.289 D_fake: 0.181 +(epoch: 403, iters: 990, time: 0.063) G_GAN: 0.238 G_GAN_Feat: 0.978 G_ID: 0.114 G_Rec: 0.355 D_GP: 0.394 D_real: 0.284 D_fake: 0.833 +(epoch: 403, iters: 1390, time: 0.064) G_GAN: 0.329 G_GAN_Feat: 0.890 G_ID: 0.124 G_Rec: 0.413 D_GP: 0.029 D_real: 1.097 D_fake: 0.677 +(epoch: 403, iters: 1790, time: 0.063) G_GAN: -0.012 G_GAN_Feat: 0.878 G_ID: 0.110 G_Rec: 0.331 D_GP: 0.129 D_real: 0.382 D_fake: 1.012 +(epoch: 403, iters: 2190, time: 0.063) G_GAN: 0.717 G_GAN_Feat: 0.987 G_ID: 0.110 G_Rec: 0.395 D_GP: 0.032 D_real: 1.191 D_fake: 0.305 +(epoch: 403, iters: 2590, time: 0.063) G_GAN: 0.192 G_GAN_Feat: 0.807 G_ID: 0.120 G_Rec: 0.332 D_GP: 0.066 D_real: 0.864 D_fake: 0.808 +(epoch: 403, iters: 2990, time: 0.064) G_GAN: 0.605 G_GAN_Feat: 0.936 G_ID: 0.139 G_Rec: 0.412 D_GP: 0.034 D_real: 1.316 D_fake: 0.423 +(epoch: 403, iters: 3390, time: 0.063) G_GAN: 0.147 G_GAN_Feat: 0.854 G_ID: 0.090 G_Rec: 0.319 D_GP: 0.032 D_real: 0.913 D_fake: 0.853 +(epoch: 403, iters: 3790, time: 0.063) G_GAN: 0.417 G_GAN_Feat: 0.914 G_ID: 0.127 G_Rec: 0.453 D_GP: 0.036 D_real: 1.209 D_fake: 0.586 +(epoch: 403, iters: 4190, time: 0.064) G_GAN: 0.066 G_GAN_Feat: 0.693 G_ID: 0.104 G_Rec: 0.295 D_GP: 0.043 D_real: 0.814 D_fake: 0.934 +(epoch: 403, iters: 4590, time: 0.064) G_GAN: 0.502 G_GAN_Feat: 0.997 G_ID: 0.115 G_Rec: 0.443 D_GP: 0.057 D_real: 0.892 D_fake: 0.511 +(epoch: 403, iters: 4990, time: 0.063) G_GAN: 0.289 G_GAN_Feat: 0.838 G_ID: 0.104 G_Rec: 0.323 D_GP: 0.034 D_real: 0.873 D_fake: 0.712 +(epoch: 403, iters: 5390, time: 0.063) G_GAN: 0.451 G_GAN_Feat: 0.986 G_ID: 0.106 G_Rec: 0.464 D_GP: 0.038 D_real: 1.075 D_fake: 0.552 +(epoch: 403, iters: 5790, time: 0.063) G_GAN: 0.285 G_GAN_Feat: 0.781 G_ID: 0.096 G_Rec: 0.324 D_GP: 0.110 D_real: 0.864 D_fake: 0.715 +(epoch: 403, iters: 6190, time: 0.064) G_GAN: 0.362 G_GAN_Feat: 1.015 G_ID: 0.139 G_Rec: 0.448 D_GP: 0.067 D_real: 0.864 D_fake: 0.639 +(epoch: 403, iters: 6590, time: 0.063) G_GAN: 0.309 G_GAN_Feat: 0.798 G_ID: 0.095 G_Rec: 0.299 D_GP: 0.083 D_real: 0.856 D_fake: 0.693 +(epoch: 403, iters: 6990, time: 0.063) G_GAN: 0.700 G_GAN_Feat: 1.163 G_ID: 0.126 G_Rec: 0.425 D_GP: 0.740 D_real: 0.589 D_fake: 0.325 +(epoch: 403, iters: 7390, time: 0.063) G_GAN: 0.180 G_GAN_Feat: 0.836 G_ID: 0.096 G_Rec: 0.301 D_GP: 0.049 D_real: 0.689 D_fake: 0.821 +(epoch: 403, iters: 7790, time: 0.064) G_GAN: 0.538 G_GAN_Feat: 0.981 G_ID: 0.118 G_Rec: 0.431 D_GP: 0.039 D_real: 1.048 D_fake: 0.472 +(epoch: 403, iters: 8190, time: 0.063) G_GAN: 0.230 G_GAN_Feat: 0.822 G_ID: 0.096 G_Rec: 0.320 D_GP: 0.039 D_real: 0.981 D_fake: 0.770 +(epoch: 403, iters: 8590, time: 0.063) G_GAN: 0.654 G_GAN_Feat: 1.029 G_ID: 0.108 G_Rec: 0.440 D_GP: 0.035 D_real: 1.161 D_fake: 0.383 +(epoch: 404, iters: 382, time: 0.063) G_GAN: 0.260 G_GAN_Feat: 0.879 G_ID: 0.111 G_Rec: 0.336 D_GP: 0.096 D_real: 0.367 D_fake: 0.744 +(epoch: 404, iters: 782, time: 0.064) G_GAN: 0.584 G_GAN_Feat: 1.254 G_ID: 0.126 G_Rec: 0.533 D_GP: 1.268 D_real: 0.302 D_fake: 0.489 +(epoch: 404, iters: 1182, time: 0.063) G_GAN: -0.237 G_GAN_Feat: 0.670 G_ID: 0.115 G_Rec: 0.296 D_GP: 0.025 D_real: 0.681 D_fake: 1.237 +(epoch: 404, iters: 1582, time: 0.063) G_GAN: 0.061 G_GAN_Feat: 0.958 G_ID: 0.115 G_Rec: 0.479 D_GP: 0.024 D_real: 0.840 D_fake: 0.940 +(epoch: 404, iters: 1982, time: 0.063) G_GAN: -0.116 G_GAN_Feat: 0.698 G_ID: 0.094 G_Rec: 0.327 D_GP: 0.031 D_real: 0.756 D_fake: 1.117 +(epoch: 404, iters: 2382, time: 0.064) G_GAN: 0.057 G_GAN_Feat: 0.819 G_ID: 0.111 G_Rec: 0.382 D_GP: 0.035 D_real: 0.706 D_fake: 0.946 +(epoch: 404, iters: 2782, time: 0.063) G_GAN: -0.041 G_GAN_Feat: 0.669 G_ID: 0.089 G_Rec: 0.308 D_GP: 0.034 D_real: 0.838 D_fake: 1.041 +(epoch: 404, iters: 3182, time: 0.063) G_GAN: -0.019 G_GAN_Feat: 0.943 G_ID: 0.118 G_Rec: 0.418 D_GP: 0.055 D_real: 0.532 D_fake: 1.019 +(epoch: 404, iters: 3582, time: 0.063) G_GAN: 0.201 G_GAN_Feat: 0.742 G_ID: 0.096 G_Rec: 0.314 D_GP: 0.057 D_real: 0.930 D_fake: 0.804 +(epoch: 404, iters: 3982, time: 0.064) G_GAN: 0.533 G_GAN_Feat: 0.965 G_ID: 0.114 G_Rec: 0.365 D_GP: 0.043 D_real: 0.979 D_fake: 0.475 +(epoch: 404, iters: 4382, time: 0.063) G_GAN: 0.168 G_GAN_Feat: 0.722 G_ID: 0.091 G_Rec: 0.276 D_GP: 0.041 D_real: 0.907 D_fake: 0.832 +(epoch: 404, iters: 4782, time: 0.063) G_GAN: 0.668 G_GAN_Feat: 0.991 G_ID: 0.103 G_Rec: 0.421 D_GP: 0.036 D_real: 1.075 D_fake: 0.356 +(epoch: 404, iters: 5182, time: 0.063) G_GAN: 0.660 G_GAN_Feat: 0.853 G_ID: 0.095 G_Rec: 0.341 D_GP: 0.039 D_real: 1.451 D_fake: 0.355 +(epoch: 404, iters: 5582, time: 0.064) G_GAN: 0.160 G_GAN_Feat: 1.081 G_ID: 0.119 G_Rec: 0.443 D_GP: 0.045 D_real: 0.760 D_fake: 0.842 +(epoch: 404, iters: 5982, time: 0.063) G_GAN: 0.229 G_GAN_Feat: 0.853 G_ID: 0.107 G_Rec: 0.303 D_GP: 0.044 D_real: 0.528 D_fake: 0.771 +(epoch: 404, iters: 6382, time: 0.063) G_GAN: 0.335 G_GAN_Feat: 0.928 G_ID: 0.120 G_Rec: 0.445 D_GP: 0.031 D_real: 1.058 D_fake: 0.671 +(epoch: 404, iters: 6782, time: 0.063) G_GAN: 0.216 G_GAN_Feat: 0.712 G_ID: 0.116 G_Rec: 0.301 D_GP: 0.034 D_real: 1.054 D_fake: 0.786 +(epoch: 404, iters: 7182, time: 0.064) G_GAN: 0.474 G_GAN_Feat: 1.107 G_ID: 0.114 G_Rec: 0.518 D_GP: 0.127 D_real: 0.702 D_fake: 0.555 +(epoch: 404, iters: 7582, time: 0.063) G_GAN: 0.299 G_GAN_Feat: 0.768 G_ID: 0.103 G_Rec: 0.335 D_GP: 0.037 D_real: 0.998 D_fake: 0.704 +(epoch: 404, iters: 7982, time: 0.063) G_GAN: 0.429 G_GAN_Feat: 1.163 G_ID: 0.111 G_Rec: 0.505 D_GP: 0.433 D_real: 1.237 D_fake: 0.809 +(epoch: 404, iters: 8382, time: 0.063) G_GAN: -0.106 G_GAN_Feat: 0.821 G_ID: 0.109 G_Rec: 0.347 D_GP: 0.032 D_real: 0.647 D_fake: 1.106 +(epoch: 405, iters: 174, time: 0.064) G_GAN: 0.275 G_GAN_Feat: 1.016 G_ID: 0.133 G_Rec: 0.422 D_GP: 0.037 D_real: 0.772 D_fake: 0.727 +(epoch: 405, iters: 574, time: 0.063) G_GAN: 0.284 G_GAN_Feat: 0.791 G_ID: 0.115 G_Rec: 0.298 D_GP: 0.065 D_real: 0.915 D_fake: 0.717 +(epoch: 405, iters: 974, time: 0.063) G_GAN: 0.379 G_GAN_Feat: 0.911 G_ID: 0.121 G_Rec: 0.432 D_GP: 0.032 D_real: 1.046 D_fake: 0.623 +(epoch: 405, iters: 1374, time: 0.063) G_GAN: 0.063 G_GAN_Feat: 0.824 G_ID: 0.112 G_Rec: 0.338 D_GP: 0.077 D_real: 0.713 D_fake: 0.937 +(epoch: 405, iters: 1774, time: 0.064) G_GAN: 0.522 G_GAN_Feat: 1.061 G_ID: 0.110 G_Rec: 0.424 D_GP: 0.050 D_real: 0.735 D_fake: 0.479 +(epoch: 405, iters: 2174, time: 0.063) G_GAN: 0.209 G_GAN_Feat: 0.966 G_ID: 0.095 G_Rec: 0.367 D_GP: 0.642 D_real: 0.271 D_fake: 0.804 +(epoch: 405, iters: 2574, time: 0.063) G_GAN: 0.399 G_GAN_Feat: 0.969 G_ID: 0.112 G_Rec: 0.423 D_GP: 0.041 D_real: 1.064 D_fake: 0.601 +(epoch: 405, iters: 2974, time: 0.063) G_GAN: 0.169 G_GAN_Feat: 1.022 G_ID: 0.114 G_Rec: 0.359 D_GP: 3.856 D_real: 0.381 D_fake: 0.835 +(epoch: 405, iters: 3374, time: 0.064) G_GAN: 0.422 G_GAN_Feat: 0.859 G_ID: 0.110 G_Rec: 0.434 D_GP: 0.030 D_real: 1.175 D_fake: 0.581 +(epoch: 405, iters: 3774, time: 0.063) G_GAN: 0.059 G_GAN_Feat: 0.730 G_ID: 0.091 G_Rec: 0.354 D_GP: 0.052 D_real: 0.824 D_fake: 0.941 +(epoch: 405, iters: 4174, time: 0.063) G_GAN: 0.282 G_GAN_Feat: 0.919 G_ID: 0.113 G_Rec: 0.447 D_GP: 0.034 D_real: 0.945 D_fake: 0.719 +(epoch: 405, iters: 4574, time: 0.063) G_GAN: 0.042 G_GAN_Feat: 0.723 G_ID: 0.092 G_Rec: 0.311 D_GP: 0.051 D_real: 0.813 D_fake: 0.958 +(epoch: 405, iters: 4974, time: 0.064) G_GAN: 0.429 G_GAN_Feat: 0.895 G_ID: 0.118 G_Rec: 0.391 D_GP: 0.063 D_real: 0.983 D_fake: 0.589 +(epoch: 405, iters: 5374, time: 0.063) G_GAN: -0.120 G_GAN_Feat: 0.748 G_ID: 0.122 G_Rec: 0.331 D_GP: 0.063 D_real: 0.551 D_fake: 1.120 +(epoch: 405, iters: 5774, time: 0.063) G_GAN: 0.506 G_GAN_Feat: 0.837 G_ID: 0.138 G_Rec: 0.375 D_GP: 0.028 D_real: 1.275 D_fake: 0.505 +(epoch: 405, iters: 6174, time: 0.063) G_GAN: -0.044 G_GAN_Feat: 0.754 G_ID: 0.111 G_Rec: 0.334 D_GP: 0.049 D_real: 0.625 D_fake: 1.044 +(epoch: 405, iters: 6574, time: 0.064) G_GAN: 0.424 G_GAN_Feat: 0.956 G_ID: 0.113 G_Rec: 0.460 D_GP: 0.029 D_real: 1.049 D_fake: 0.578 +(epoch: 405, iters: 6974, time: 0.063) G_GAN: -0.126 G_GAN_Feat: 0.842 G_ID: 0.098 G_Rec: 0.359 D_GP: 0.255 D_real: 0.419 D_fake: 1.127 +(epoch: 405, iters: 7374, time: 0.063) G_GAN: 0.559 G_GAN_Feat: 0.956 G_ID: 0.122 G_Rec: 0.520 D_GP: 0.029 D_real: 1.104 D_fake: 0.445 +(epoch: 405, iters: 7774, time: 0.063) G_GAN: 0.437 G_GAN_Feat: 0.713 G_ID: 0.093 G_Rec: 0.292 D_GP: 0.040 D_real: 1.203 D_fake: 0.571 +(epoch: 405, iters: 8174, time: 0.064) G_GAN: 0.488 G_GAN_Feat: 1.115 G_ID: 0.132 G_Rec: 0.491 D_GP: 0.115 D_real: 0.488 D_fake: 0.519 +(epoch: 405, iters: 8574, time: 0.063) G_GAN: 0.278 G_GAN_Feat: 0.712 G_ID: 0.102 G_Rec: 0.289 D_GP: 0.035 D_real: 1.048 D_fake: 0.722 +(epoch: 406, iters: 366, time: 0.063) G_GAN: 0.518 G_GAN_Feat: 1.114 G_ID: 0.122 G_Rec: 0.471 D_GP: 0.219 D_real: 0.439 D_fake: 0.510 +(epoch: 406, iters: 766, time: 0.063) G_GAN: 0.320 G_GAN_Feat: 0.854 G_ID: 0.112 G_Rec: 0.361 D_GP: 0.205 D_real: 0.985 D_fake: 0.817 +(epoch: 406, iters: 1166, time: 0.064) G_GAN: 0.009 G_GAN_Feat: 0.925 G_ID: 0.142 G_Rec: 0.480 D_GP: 0.033 D_real: 0.681 D_fake: 0.991 +(epoch: 406, iters: 1566, time: 0.063) G_GAN: 0.070 G_GAN_Feat: 0.784 G_ID: 0.099 G_Rec: 0.311 D_GP: 0.054 D_real: 0.882 D_fake: 0.930 +(epoch: 406, iters: 1966, time: 0.063) G_GAN: 0.452 G_GAN_Feat: 1.114 G_ID: 0.127 G_Rec: 0.507 D_GP: 0.080 D_real: 0.363 D_fake: 0.553 +(epoch: 406, iters: 2366, time: 0.063) G_GAN: 0.236 G_GAN_Feat: 1.011 G_ID: 0.097 G_Rec: 0.365 D_GP: 0.888 D_real: 0.374 D_fake: 0.767 +(epoch: 406, iters: 2766, time: 0.064) G_GAN: 0.458 G_GAN_Feat: 0.961 G_ID: 0.151 G_Rec: 0.417 D_GP: 0.038 D_real: 1.036 D_fake: 0.544 +(epoch: 406, iters: 3166, time: 0.063) G_GAN: 0.314 G_GAN_Feat: 0.816 G_ID: 0.094 G_Rec: 0.325 D_GP: 0.046 D_real: 1.074 D_fake: 0.688 +(epoch: 406, iters: 3566, time: 0.063) G_GAN: 0.793 G_GAN_Feat: 1.001 G_ID: 0.113 G_Rec: 0.469 D_GP: 0.041 D_real: 1.345 D_fake: 0.276 +(epoch: 406, iters: 3966, time: 0.063) G_GAN: 0.233 G_GAN_Feat: 0.862 G_ID: 0.101 G_Rec: 0.310 D_GP: 0.065 D_real: 0.650 D_fake: 0.772 +(epoch: 406, iters: 4366, time: 0.064) G_GAN: 0.700 G_GAN_Feat: 1.133 G_ID: 0.112 G_Rec: 0.445 D_GP: 0.035 D_real: 1.184 D_fake: 0.349 +(epoch: 406, iters: 4766, time: 0.063) G_GAN: 0.157 G_GAN_Feat: 0.999 G_ID: 0.092 G_Rec: 0.337 D_GP: 0.189 D_real: 0.511 D_fake: 0.857 +(epoch: 406, iters: 5166, time: 0.063) G_GAN: 0.414 G_GAN_Feat: 1.284 G_ID: 0.150 G_Rec: 0.451 D_GP: 0.153 D_real: 0.139 D_fake: 0.616 +(epoch: 406, iters: 5566, time: 0.063) G_GAN: 0.049 G_GAN_Feat: 0.975 G_ID: 0.141 G_Rec: 0.357 D_GP: 0.552 D_real: 0.383 D_fake: 0.954 +(epoch: 406, iters: 5966, time: 0.064) G_GAN: 0.418 G_GAN_Feat: 0.919 G_ID: 0.131 G_Rec: 0.426 D_GP: 0.031 D_real: 1.048 D_fake: 0.592 +(epoch: 406, iters: 6366, time: 0.063) G_GAN: 0.220 G_GAN_Feat: 0.841 G_ID: 0.112 G_Rec: 0.351 D_GP: 0.033 D_real: 1.014 D_fake: 0.781 +(epoch: 406, iters: 6766, time: 0.063) G_GAN: 0.220 G_GAN_Feat: 1.107 G_ID: 0.123 G_Rec: 0.464 D_GP: 0.100 D_real: 0.260 D_fake: 0.785 +(epoch: 406, iters: 7166, time: 0.063) G_GAN: 0.162 G_GAN_Feat: 0.720 G_ID: 0.123 G_Rec: 0.309 D_GP: 0.031 D_real: 1.111 D_fake: 0.867 +(epoch: 406, iters: 7566, time: 0.064) G_GAN: 0.202 G_GAN_Feat: 0.841 G_ID: 0.140 G_Rec: 0.415 D_GP: 0.026 D_real: 0.957 D_fake: 0.798 +(epoch: 406, iters: 7966, time: 0.063) G_GAN: 0.167 G_GAN_Feat: 0.778 G_ID: 0.118 G_Rec: 0.340 D_GP: 0.031 D_real: 0.987 D_fake: 0.834 +(epoch: 406, iters: 8366, time: 0.063) G_GAN: 0.391 G_GAN_Feat: 0.937 G_ID: 0.127 G_Rec: 0.455 D_GP: 0.031 D_real: 1.052 D_fake: 0.623 +(epoch: 407, iters: 158, time: 0.063) G_GAN: 0.022 G_GAN_Feat: 0.704 G_ID: 0.091 G_Rec: 0.314 D_GP: 0.032 D_real: 0.925 D_fake: 0.978 +(epoch: 407, iters: 558, time: 0.063) G_GAN: 0.108 G_GAN_Feat: 0.899 G_ID: 0.134 G_Rec: 0.416 D_GP: 0.055 D_real: 0.709 D_fake: 0.894 +(epoch: 407, iters: 958, time: 0.064) G_GAN: -0.131 G_GAN_Feat: 0.720 G_ID: 0.098 G_Rec: 0.379 D_GP: 0.040 D_real: 0.647 D_fake: 1.131 +(epoch: 407, iters: 1358, time: 0.064) G_GAN: 0.478 G_GAN_Feat: 1.076 G_ID: 0.110 G_Rec: 0.486 D_GP: 0.049 D_real: 0.895 D_fake: 0.529 +(epoch: 407, iters: 1758, time: 0.064) G_GAN: 0.036 G_GAN_Feat: 0.812 G_ID: 0.101 G_Rec: 0.316 D_GP: 0.121 D_real: 0.638 D_fake: 0.964 +(epoch: 407, iters: 2158, time: 0.063) G_GAN: 0.347 G_GAN_Feat: 1.045 G_ID: 0.118 G_Rec: 0.449 D_GP: 0.046 D_real: 0.864 D_fake: 0.655 +(epoch: 407, iters: 2558, time: 0.063) G_GAN: 0.401 G_GAN_Feat: 0.832 G_ID: 0.095 G_Rec: 0.305 D_GP: 0.039 D_real: 1.033 D_fake: 0.600 +(epoch: 407, iters: 2958, time: 0.063) G_GAN: 0.661 G_GAN_Feat: 1.216 G_ID: 0.115 G_Rec: 0.454 D_GP: 0.067 D_real: 0.678 D_fake: 0.373 +(epoch: 407, iters: 3358, time: 0.064) G_GAN: 0.662 G_GAN_Feat: 0.723 G_ID: 0.082 G_Rec: 0.269 D_GP: 0.034 D_real: 1.668 D_fake: 0.495 +(epoch: 407, iters: 3758, time: 0.063) G_GAN: 0.827 G_GAN_Feat: 1.122 G_ID: 0.115 G_Rec: 0.435 D_GP: 0.046 D_real: 0.592 D_fake: 0.226 +(epoch: 407, iters: 4158, time: 0.063) G_GAN: -0.015 G_GAN_Feat: 0.690 G_ID: 0.105 G_Rec: 0.339 D_GP: 0.045 D_real: 0.925 D_fake: 1.015 +(epoch: 407, iters: 4558, time: 0.063) G_GAN: 0.074 G_GAN_Feat: 0.769 G_ID: 0.111 G_Rec: 0.403 D_GP: 0.035 D_real: 0.894 D_fake: 0.927 +(epoch: 407, iters: 4958, time: 0.064) G_GAN: 0.161 G_GAN_Feat: 0.591 G_ID: 0.097 G_Rec: 0.292 D_GP: 0.035 D_real: 1.077 D_fake: 0.839 +(epoch: 407, iters: 5358, time: 0.063) G_GAN: 0.189 G_GAN_Feat: 0.801 G_ID: 0.140 G_Rec: 0.407 D_GP: 0.038 D_real: 0.828 D_fake: 0.813 +(epoch: 407, iters: 5758, time: 0.063) G_GAN: 0.036 G_GAN_Feat: 0.625 G_ID: 0.093 G_Rec: 0.317 D_GP: 0.038 D_real: 0.998 D_fake: 0.964 +(epoch: 407, iters: 6158, time: 0.063) G_GAN: 0.340 G_GAN_Feat: 0.804 G_ID: 0.118 G_Rec: 0.397 D_GP: 0.038 D_real: 0.976 D_fake: 0.669 +(epoch: 407, iters: 6558, time: 0.064) G_GAN: -0.125 G_GAN_Feat: 0.682 G_ID: 0.097 G_Rec: 0.347 D_GP: 0.049 D_real: 0.750 D_fake: 1.125 +(epoch: 407, iters: 6958, time: 0.063) G_GAN: 0.210 G_GAN_Feat: 0.882 G_ID: 0.128 G_Rec: 0.466 D_GP: 0.047 D_real: 0.742 D_fake: 0.795 +(epoch: 407, iters: 7358, time: 0.063) G_GAN: -0.012 G_GAN_Feat: 0.711 G_ID: 0.097 G_Rec: 0.344 D_GP: 0.050 D_real: 0.809 D_fake: 1.012 +(epoch: 407, iters: 7758, time: 0.063) G_GAN: 0.079 G_GAN_Feat: 0.787 G_ID: 0.118 G_Rec: 0.377 D_GP: 0.037 D_real: 0.736 D_fake: 0.921 +(epoch: 407, iters: 8158, time: 0.064) G_GAN: -0.139 G_GAN_Feat: 0.577 G_ID: 0.105 G_Rec: 0.272 D_GP: 0.037 D_real: 0.848 D_fake: 1.139 +(epoch: 407, iters: 8558, time: 0.063) G_GAN: 0.155 G_GAN_Feat: 0.850 G_ID: 0.116 G_Rec: 0.410 D_GP: 0.059 D_real: 0.850 D_fake: 0.847 +(epoch: 408, iters: 350, time: 0.063) G_GAN: -0.262 G_GAN_Feat: 0.665 G_ID: 0.099 G_Rec: 0.313 D_GP: 0.043 D_real: 0.597 D_fake: 1.262 +(epoch: 408, iters: 750, time: 0.063) G_GAN: -0.075 G_GAN_Feat: 0.848 G_ID: 0.137 G_Rec: 0.415 D_GP: 0.044 D_real: 0.579 D_fake: 1.075 +(epoch: 408, iters: 1150, time: 0.064) G_GAN: -0.060 G_GAN_Feat: 0.675 G_ID: 0.107 G_Rec: 0.331 D_GP: 0.058 D_real: 0.714 D_fake: 1.063 +(epoch: 408, iters: 1550, time: 0.063) G_GAN: 0.243 G_GAN_Feat: 0.934 G_ID: 0.113 G_Rec: 0.472 D_GP: 0.058 D_real: 0.837 D_fake: 0.760 +(epoch: 408, iters: 1950, time: 0.063) G_GAN: 0.062 G_GAN_Feat: 0.600 G_ID: 0.094 G_Rec: 0.307 D_GP: 0.037 D_real: 0.901 D_fake: 0.939 +(epoch: 408, iters: 2350, time: 0.063) G_GAN: 0.369 G_GAN_Feat: 0.792 G_ID: 0.115 G_Rec: 0.385 D_GP: 0.036 D_real: 0.982 D_fake: 0.645 +(epoch: 408, iters: 2750, time: 0.064) G_GAN: 0.190 G_GAN_Feat: 0.621 G_ID: 0.106 G_Rec: 0.281 D_GP: 0.034 D_real: 1.064 D_fake: 0.815 +(epoch: 408, iters: 3150, time: 0.063) G_GAN: 0.315 G_GAN_Feat: 0.888 G_ID: 0.119 G_Rec: 0.442 D_GP: 0.040 D_real: 0.941 D_fake: 0.696 +(epoch: 408, iters: 3550, time: 0.063) G_GAN: 0.052 G_GAN_Feat: 0.659 G_ID: 0.098 G_Rec: 0.302 D_GP: 0.033 D_real: 0.897 D_fake: 0.949 +(epoch: 408, iters: 3950, time: 0.063) G_GAN: 0.267 G_GAN_Feat: 0.819 G_ID: 0.105 G_Rec: 0.376 D_GP: 0.039 D_real: 1.016 D_fake: 0.734 +(epoch: 408, iters: 4350, time: 0.064) G_GAN: -0.021 G_GAN_Feat: 0.747 G_ID: 0.107 G_Rec: 0.326 D_GP: 0.053 D_real: 0.673 D_fake: 1.021 +(epoch: 408, iters: 4750, time: 0.063) G_GAN: 0.383 G_GAN_Feat: 0.955 G_ID: 0.125 G_Rec: 0.420 D_GP: 0.079 D_real: 0.867 D_fake: 0.623 +(epoch: 408, iters: 5150, time: 0.063) G_GAN: 0.126 G_GAN_Feat: 0.723 G_ID: 0.111 G_Rec: 0.328 D_GP: 0.039 D_real: 0.959 D_fake: 0.874 +(epoch: 408, iters: 5550, time: 0.063) G_GAN: 0.444 G_GAN_Feat: 0.932 G_ID: 0.123 G_Rec: 0.418 D_GP: 0.061 D_real: 0.959 D_fake: 0.568 +(epoch: 408, iters: 5950, time: 0.064) G_GAN: -0.063 G_GAN_Feat: 0.692 G_ID: 0.111 G_Rec: 0.337 D_GP: 0.036 D_real: 0.814 D_fake: 1.063 +(epoch: 408, iters: 6350, time: 0.063) G_GAN: 0.505 G_GAN_Feat: 1.121 G_ID: 0.108 G_Rec: 0.507 D_GP: 0.142 D_real: 0.570 D_fake: 0.523 +(epoch: 408, iters: 6750, time: 0.063) G_GAN: 0.335 G_GAN_Feat: 0.704 G_ID: 0.109 G_Rec: 0.300 D_GP: 0.043 D_real: 1.149 D_fake: 0.666 +(epoch: 408, iters: 7150, time: 0.063) G_GAN: 0.112 G_GAN_Feat: 1.008 G_ID: 0.127 G_Rec: 0.488 D_GP: 0.052 D_real: 0.536 D_fake: 0.888 +(epoch: 408, iters: 7550, time: 0.064) G_GAN: 0.055 G_GAN_Feat: 0.691 G_ID: 0.102 G_Rec: 0.301 D_GP: 0.035 D_real: 0.944 D_fake: 0.946 +(epoch: 408, iters: 7950, time: 0.063) G_GAN: 0.384 G_GAN_Feat: 1.007 G_ID: 0.113 G_Rec: 0.436 D_GP: 0.057 D_real: 0.757 D_fake: 0.622 +(epoch: 408, iters: 8350, time: 0.063) G_GAN: 0.227 G_GAN_Feat: 0.747 G_ID: 0.104 G_Rec: 0.328 D_GP: 0.039 D_real: 1.041 D_fake: 0.774 +(epoch: 409, iters: 142, time: 0.063) G_GAN: 0.317 G_GAN_Feat: 0.911 G_ID: 0.142 G_Rec: 0.403 D_GP: 0.045 D_real: 0.796 D_fake: 0.683 +(epoch: 409, iters: 542, time: 0.064) G_GAN: 0.260 G_GAN_Feat: 0.748 G_ID: 0.114 G_Rec: 0.330 D_GP: 0.072 D_real: 0.869 D_fake: 0.741 +(epoch: 409, iters: 942, time: 0.063) G_GAN: 0.229 G_GAN_Feat: 0.981 G_ID: 0.120 G_Rec: 0.449 D_GP: 0.041 D_real: 0.845 D_fake: 0.773 +(epoch: 409, iters: 1342, time: 0.063) G_GAN: -0.051 G_GAN_Feat: 0.819 G_ID: 0.122 G_Rec: 0.320 D_GP: 0.082 D_real: 0.321 D_fake: 1.051 +(epoch: 409, iters: 1742, time: 0.063) G_GAN: 0.478 G_GAN_Feat: 1.079 G_ID: 0.111 G_Rec: 0.491 D_GP: 0.083 D_real: 0.568 D_fake: 0.531 +(epoch: 409, iters: 2142, time: 0.064) G_GAN: 0.174 G_GAN_Feat: 0.638 G_ID: 0.100 G_Rec: 0.291 D_GP: 0.027 D_real: 1.080 D_fake: 0.826 +(epoch: 409, iters: 2542, time: 0.063) G_GAN: 0.597 G_GAN_Feat: 0.953 G_ID: 0.114 G_Rec: 0.447 D_GP: 0.030 D_real: 1.158 D_fake: 0.421 +(epoch: 409, iters: 2942, time: 0.063) G_GAN: 0.233 G_GAN_Feat: 0.711 G_ID: 0.092 G_Rec: 0.332 D_GP: 0.036 D_real: 1.113 D_fake: 0.768 +(epoch: 409, iters: 3342, time: 0.063) G_GAN: 0.394 G_GAN_Feat: 1.051 G_ID: 0.123 G_Rec: 0.481 D_GP: 0.113 D_real: 0.662 D_fake: 0.615 +(epoch: 409, iters: 3742, time: 0.064) G_GAN: 0.364 G_GAN_Feat: 0.668 G_ID: 0.097 G_Rec: 0.336 D_GP: 0.031 D_real: 1.306 D_fake: 0.638 +(epoch: 409, iters: 4142, time: 0.063) G_GAN: 0.344 G_GAN_Feat: 0.995 G_ID: 0.119 G_Rec: 0.449 D_GP: 0.054 D_real: 0.705 D_fake: 0.657 +(epoch: 409, iters: 4542, time: 0.063) G_GAN: 0.080 G_GAN_Feat: 0.760 G_ID: 0.108 G_Rec: 0.334 D_GP: 0.039 D_real: 0.830 D_fake: 0.920 +(epoch: 409, iters: 4942, time: 0.063) G_GAN: 0.752 G_GAN_Feat: 0.959 G_ID: 0.122 G_Rec: 0.452 D_GP: 0.038 D_real: 1.221 D_fake: 0.288 +(epoch: 409, iters: 5342, time: 0.064) G_GAN: 0.340 G_GAN_Feat: 0.730 G_ID: 0.111 G_Rec: 0.352 D_GP: 0.028 D_real: 1.187 D_fake: 0.661 +(epoch: 409, iters: 5742, time: 0.063) G_GAN: 0.426 G_GAN_Feat: 0.880 G_ID: 0.122 G_Rec: 0.405 D_GP: 0.037 D_real: 1.111 D_fake: 0.577 +(epoch: 409, iters: 6142, time: 0.063) G_GAN: 0.528 G_GAN_Feat: 0.724 G_ID: 0.091 G_Rec: 0.297 D_GP: 0.031 D_real: 1.371 D_fake: 0.473 +(epoch: 409, iters: 6542, time: 0.063) G_GAN: 0.531 G_GAN_Feat: 0.948 G_ID: 0.125 G_Rec: 0.446 D_GP: 0.033 D_real: 1.142 D_fake: 0.473 +(epoch: 409, iters: 6942, time: 0.064) G_GAN: -0.000 G_GAN_Feat: 0.672 G_ID: 0.099 G_Rec: 0.267 D_GP: 0.031 D_real: 0.951 D_fake: 1.000 +(epoch: 409, iters: 7342, time: 0.063) G_GAN: 0.388 G_GAN_Feat: 0.916 G_ID: 0.127 G_Rec: 0.461 D_GP: 0.034 D_real: 1.159 D_fake: 0.614 +(epoch: 409, iters: 7742, time: 0.063) G_GAN: 0.312 G_GAN_Feat: 0.743 G_ID: 0.116 G_Rec: 0.323 D_GP: 0.036 D_real: 1.089 D_fake: 0.689 +(epoch: 409, iters: 8142, time: 0.063) G_GAN: 0.443 G_GAN_Feat: 0.913 G_ID: 0.113 G_Rec: 0.412 D_GP: 0.038 D_real: 0.973 D_fake: 0.567 +(epoch: 409, iters: 8542, time: 0.064) G_GAN: 0.248 G_GAN_Feat: 0.826 G_ID: 0.102 G_Rec: 0.327 D_GP: 0.094 D_real: 0.729 D_fake: 0.752 +(epoch: 410, iters: 334, time: 0.063) G_GAN: 0.660 G_GAN_Feat: 1.062 G_ID: 0.119 G_Rec: 0.460 D_GP: 0.050 D_real: 0.799 D_fake: 0.363 +(epoch: 410, iters: 734, time: 0.063) G_GAN: 0.089 G_GAN_Feat: 0.859 G_ID: 0.089 G_Rec: 0.339 D_GP: 0.089 D_real: 0.314 D_fake: 0.911 +(epoch: 410, iters: 1134, time: 0.063) G_GAN: 0.655 G_GAN_Feat: 1.004 G_ID: 0.121 G_Rec: 0.456 D_GP: 0.032 D_real: 1.113 D_fake: 0.363 +(epoch: 410, iters: 1534, time: 0.064) G_GAN: 0.002 G_GAN_Feat: 0.756 G_ID: 0.103 G_Rec: 0.306 D_GP: 0.028 D_real: 0.938 D_fake: 0.998 +(epoch: 410, iters: 1934, time: 0.063) G_GAN: 0.223 G_GAN_Feat: 1.045 G_ID: 0.109 G_Rec: 0.471 D_GP: 0.078 D_real: 0.697 D_fake: 0.779 +(epoch: 410, iters: 2334, time: 0.063) G_GAN: 0.208 G_GAN_Feat: 0.702 G_ID: 0.096 G_Rec: 0.280 D_GP: 0.038 D_real: 1.034 D_fake: 0.793 +(epoch: 410, iters: 2734, time: 0.063) G_GAN: 0.343 G_GAN_Feat: 1.007 G_ID: 0.138 G_Rec: 0.433 D_GP: 0.044 D_real: 0.740 D_fake: 0.657 +(epoch: 410, iters: 3134, time: 0.064) G_GAN: 0.278 G_GAN_Feat: 0.774 G_ID: 0.093 G_Rec: 0.321 D_GP: 0.033 D_real: 1.050 D_fake: 0.725 +(epoch: 410, iters: 3534, time: 0.063) G_GAN: 0.458 G_GAN_Feat: 0.873 G_ID: 0.121 G_Rec: 0.374 D_GP: 0.031 D_real: 1.148 D_fake: 0.551 +(epoch: 410, iters: 3934, time: 0.063) G_GAN: 0.045 G_GAN_Feat: 0.878 G_ID: 0.116 G_Rec: 0.361 D_GP: 0.123 D_real: 0.600 D_fake: 0.955 +(epoch: 410, iters: 4334, time: 0.063) G_GAN: 0.319 G_GAN_Feat: 0.919 G_ID: 0.119 G_Rec: 0.472 D_GP: 0.030 D_real: 0.974 D_fake: 0.684 +(epoch: 410, iters: 4734, time: 0.064) G_GAN: 0.054 G_GAN_Feat: 0.819 G_ID: 0.111 G_Rec: 0.311 D_GP: 0.031 D_real: 0.719 D_fake: 0.946 +(epoch: 410, iters: 5134, time: 0.063) G_GAN: -0.186 G_GAN_Feat: 0.991 G_ID: 0.133 G_Rec: 0.428 D_GP: 0.166 D_real: 0.168 D_fake: 1.187 +(epoch: 410, iters: 5534, time: 0.063) G_GAN: 0.602 G_GAN_Feat: 0.739 G_ID: 0.089 G_Rec: 0.305 D_GP: 0.035 D_real: 1.351 D_fake: 0.405 +(epoch: 410, iters: 5934, time: 0.063) G_GAN: 0.377 G_GAN_Feat: 0.939 G_ID: 0.117 G_Rec: 0.430 D_GP: 0.032 D_real: 1.083 D_fake: 0.624 +(epoch: 410, iters: 6334, time: 0.064) G_GAN: 0.138 G_GAN_Feat: 0.766 G_ID: 0.107 G_Rec: 0.335 D_GP: 0.032 D_real: 0.950 D_fake: 0.863 +(epoch: 410, iters: 6734, time: 0.063) G_GAN: 0.564 G_GAN_Feat: 0.966 G_ID: 0.123 G_Rec: 0.479 D_GP: 0.030 D_real: 1.167 D_fake: 0.440 +(epoch: 410, iters: 7134, time: 0.063) G_GAN: 0.490 G_GAN_Feat: 0.841 G_ID: 0.100 G_Rec: 0.347 D_GP: 0.053 D_real: 1.020 D_fake: 0.530 +(epoch: 410, iters: 7534, time: 0.063) G_GAN: 0.884 G_GAN_Feat: 0.980 G_ID: 0.110 G_Rec: 0.425 D_GP: 0.050 D_real: 1.330 D_fake: 0.165 +(epoch: 410, iters: 7934, time: 0.064) G_GAN: 0.106 G_GAN_Feat: 0.786 G_ID: 0.125 G_Rec: 0.307 D_GP: 0.034 D_real: 0.893 D_fake: 0.895 +(epoch: 410, iters: 8334, time: 0.063) G_GAN: 0.883 G_GAN_Feat: 1.061 G_ID: 0.122 G_Rec: 0.444 D_GP: 0.062 D_real: 1.184 D_fake: 0.285 +(epoch: 411, iters: 126, time: 0.063) G_GAN: 0.219 G_GAN_Feat: 0.720 G_ID: 0.102 G_Rec: 0.335 D_GP: 0.029 D_real: 1.052 D_fake: 0.781 +(epoch: 411, iters: 526, time: 0.063) G_GAN: 0.244 G_GAN_Feat: 0.960 G_ID: 0.124 G_Rec: 0.409 D_GP: 0.050 D_real: 0.746 D_fake: 0.756 +(epoch: 411, iters: 926, time: 0.064) G_GAN: 0.077 G_GAN_Feat: 0.950 G_ID: 0.119 G_Rec: 0.335 D_GP: 0.549 D_real: 0.266 D_fake: 0.923 +(epoch: 411, iters: 1326, time: 0.063) G_GAN: 0.676 G_GAN_Feat: 1.017 G_ID: 0.126 G_Rec: 0.431 D_GP: 0.035 D_real: 1.182 D_fake: 0.343 +(epoch: 411, iters: 1726, time: 0.063) G_GAN: 0.516 G_GAN_Feat: 0.868 G_ID: 0.100 G_Rec: 0.337 D_GP: 0.052 D_real: 0.934 D_fake: 0.492 +(epoch: 411, iters: 2126, time: 0.063) G_GAN: 0.577 G_GAN_Feat: 0.977 G_ID: 0.140 G_Rec: 0.400 D_GP: 0.029 D_real: 1.085 D_fake: 0.453 +(epoch: 411, iters: 2526, time: 0.064) G_GAN: 0.284 G_GAN_Feat: 0.882 G_ID: 0.099 G_Rec: 0.357 D_GP: 0.141 D_real: 0.775 D_fake: 0.719 +(epoch: 411, iters: 2926, time: 0.063) G_GAN: 0.953 G_GAN_Feat: 0.979 G_ID: 0.122 G_Rec: 0.442 D_GP: 0.034 D_real: 1.602 D_fake: 0.213 +(epoch: 411, iters: 3326, time: 0.063) G_GAN: 0.126 G_GAN_Feat: 0.808 G_ID: 0.101 G_Rec: 0.284 D_GP: 0.047 D_real: 0.862 D_fake: 0.874 +(epoch: 411, iters: 3726, time: 0.063) G_GAN: 0.314 G_GAN_Feat: 1.035 G_ID: 0.122 G_Rec: 0.456 D_GP: 0.073 D_real: 0.788 D_fake: 0.687 +(epoch: 411, iters: 4126, time: 0.064) G_GAN: 0.258 G_GAN_Feat: 0.726 G_ID: 0.103 G_Rec: 0.303 D_GP: 0.034 D_real: 1.001 D_fake: 0.742 +(epoch: 411, iters: 4526, time: 0.063) G_GAN: 0.704 G_GAN_Feat: 1.181 G_ID: 0.120 G_Rec: 0.476 D_GP: 0.988 D_real: 0.790 D_fake: 0.342 +(epoch: 411, iters: 4926, time: 0.063) G_GAN: 0.309 G_GAN_Feat: 0.798 G_ID: 0.096 G_Rec: 0.309 D_GP: 0.059 D_real: 1.044 D_fake: 0.691 +(epoch: 411, iters: 5326, time: 0.063) G_GAN: 0.730 G_GAN_Feat: 0.916 G_ID: 0.119 G_Rec: 0.411 D_GP: 0.060 D_real: 1.339 D_fake: 0.389 +(epoch: 411, iters: 5726, time: 0.063) G_GAN: 0.072 G_GAN_Feat: 0.833 G_ID: 0.112 G_Rec: 0.348 D_GP: 0.045 D_real: 0.792 D_fake: 0.929 +(epoch: 411, iters: 6126, time: 0.063) G_GAN: 0.423 G_GAN_Feat: 0.929 G_ID: 0.113 G_Rec: 0.425 D_GP: 0.030 D_real: 1.054 D_fake: 0.579 +(epoch: 411, iters: 6526, time: 0.063) G_GAN: 0.583 G_GAN_Feat: 0.791 G_ID: 0.086 G_Rec: 0.315 D_GP: 0.032 D_real: 1.330 D_fake: 0.436 +(epoch: 411, iters: 6926, time: 0.064) G_GAN: 0.628 G_GAN_Feat: 1.150 G_ID: 0.115 G_Rec: 0.454 D_GP: 0.049 D_real: 0.920 D_fake: 0.391 +(epoch: 411, iters: 7326, time: 0.063) G_GAN: 0.583 G_GAN_Feat: 1.013 G_ID: 0.104 G_Rec: 0.344 D_GP: 0.059 D_real: 0.516 D_fake: 0.433 +(epoch: 411, iters: 7726, time: 0.063) G_GAN: 0.871 G_GAN_Feat: 1.070 G_ID: 0.115 G_Rec: 0.492 D_GP: 0.040 D_real: 1.262 D_fake: 0.186 +(epoch: 411, iters: 8126, time: 0.063) G_GAN: 0.713 G_GAN_Feat: 0.851 G_ID: 0.099 G_Rec: 0.329 D_GP: 0.060 D_real: 1.275 D_fake: 0.475 +(epoch: 411, iters: 8526, time: 0.064) G_GAN: 0.016 G_GAN_Feat: 0.992 G_ID: 0.134 G_Rec: 0.439 D_GP: 0.044 D_real: 0.649 D_fake: 0.984 +(epoch: 412, iters: 318, time: 0.063) G_GAN: 0.521 G_GAN_Feat: 0.756 G_ID: 0.094 G_Rec: 0.285 D_GP: 0.038 D_real: 1.254 D_fake: 0.486 +(epoch: 412, iters: 718, time: 0.063) G_GAN: 0.288 G_GAN_Feat: 1.164 G_ID: 0.129 G_Rec: 0.439 D_GP: 0.048 D_real: 0.369 D_fake: 0.713 +(epoch: 412, iters: 1118, time: 0.063) G_GAN: 0.376 G_GAN_Feat: 0.930 G_ID: 0.106 G_Rec: 0.329 D_GP: 0.270 D_real: 0.511 D_fake: 0.627 +(epoch: 412, iters: 1518, time: 0.064) G_GAN: 0.801 G_GAN_Feat: 0.990 G_ID: 0.107 G_Rec: 0.451 D_GP: 0.032 D_real: 1.392 D_fake: 0.233 +(epoch: 412, iters: 1918, time: 0.063) G_GAN: 0.402 G_GAN_Feat: 0.721 G_ID: 0.089 G_Rec: 0.287 D_GP: 0.028 D_real: 1.310 D_fake: 0.603 +(epoch: 412, iters: 2318, time: 0.063) G_GAN: 0.314 G_GAN_Feat: 0.835 G_ID: 0.119 G_Rec: 0.376 D_GP: 0.026 D_real: 1.118 D_fake: 0.687 +(epoch: 412, iters: 2718, time: 0.063) G_GAN: 0.215 G_GAN_Feat: 0.779 G_ID: 0.097 G_Rec: 0.319 D_GP: 0.028 D_real: 1.114 D_fake: 0.787 +(epoch: 412, iters: 3118, time: 0.064) G_GAN: 0.793 G_GAN_Feat: 1.228 G_ID: 0.121 G_Rec: 0.473 D_GP: 0.138 D_real: 0.674 D_fake: 0.387 +(epoch: 412, iters: 3518, time: 0.063) G_GAN: 0.186 G_GAN_Feat: 0.765 G_ID: 0.090 G_Rec: 0.292 D_GP: 0.035 D_real: 1.077 D_fake: 0.814 +(epoch: 412, iters: 3918, time: 0.063) G_GAN: 0.136 G_GAN_Feat: 0.943 G_ID: 0.123 G_Rec: 0.394 D_GP: 0.037 D_real: 0.800 D_fake: 0.865 +(epoch: 412, iters: 4318, time: 0.063) G_GAN: 0.082 G_GAN_Feat: 0.785 G_ID: 0.111 G_Rec: 0.305 D_GP: 0.038 D_real: 0.867 D_fake: 0.918 +(epoch: 412, iters: 4718, time: 0.064) G_GAN: 0.459 G_GAN_Feat: 1.002 G_ID: 0.112 G_Rec: 0.470 D_GP: 0.067 D_real: 0.909 D_fake: 0.546 +(epoch: 412, iters: 5118, time: 0.063) G_GAN: 0.291 G_GAN_Feat: 0.800 G_ID: 0.077 G_Rec: 0.296 D_GP: 0.080 D_real: 0.936 D_fake: 0.711 +(epoch: 412, iters: 5518, time: 0.063) G_GAN: 0.509 G_GAN_Feat: 1.028 G_ID: 0.118 G_Rec: 0.422 D_GP: 0.057 D_real: 0.934 D_fake: 0.499 +(epoch: 412, iters: 5918, time: 0.063) G_GAN: 0.263 G_GAN_Feat: 0.835 G_ID: 0.093 G_Rec: 0.310 D_GP: 0.074 D_real: 0.691 D_fake: 0.737 +(epoch: 412, iters: 6318, time: 0.064) G_GAN: 0.492 G_GAN_Feat: 1.043 G_ID: 0.136 G_Rec: 0.504 D_GP: 0.031 D_real: 0.984 D_fake: 0.518 +(epoch: 412, iters: 6718, time: 0.063) G_GAN: 0.358 G_GAN_Feat: 0.841 G_ID: 0.097 G_Rec: 0.323 D_GP: 0.071 D_real: 0.847 D_fake: 0.645 +(epoch: 412, iters: 7118, time: 0.063) G_GAN: 0.593 G_GAN_Feat: 1.006 G_ID: 0.116 G_Rec: 0.422 D_GP: 0.044 D_real: 0.990 D_fake: 0.417 +(epoch: 412, iters: 7518, time: 0.063) G_GAN: 0.260 G_GAN_Feat: 0.673 G_ID: 0.091 G_Rec: 0.295 D_GP: 0.028 D_real: 1.173 D_fake: 0.742 +(epoch: 412, iters: 7918, time: 0.064) G_GAN: 0.312 G_GAN_Feat: 0.930 G_ID: 0.128 G_Rec: 0.423 D_GP: 0.039 D_real: 0.956 D_fake: 0.689 +(epoch: 412, iters: 8318, time: 0.063) G_GAN: -0.061 G_GAN_Feat: 0.720 G_ID: 0.107 G_Rec: 0.313 D_GP: 0.036 D_real: 0.770 D_fake: 1.061 +(epoch: 413, iters: 110, time: 0.063) G_GAN: 0.760 G_GAN_Feat: 0.921 G_ID: 0.100 G_Rec: 0.441 D_GP: 0.033 D_real: 1.294 D_fake: 0.353 +(epoch: 413, iters: 510, time: 0.063) G_GAN: 0.127 G_GAN_Feat: 0.741 G_ID: 0.089 G_Rec: 0.298 D_GP: 0.035 D_real: 1.099 D_fake: 0.873 +(epoch: 413, iters: 910, time: 0.064) G_GAN: 0.426 G_GAN_Feat: 0.988 G_ID: 0.121 G_Rec: 0.439 D_GP: 0.113 D_real: 0.776 D_fake: 0.577 +(epoch: 413, iters: 1310, time: 0.063) G_GAN: 0.216 G_GAN_Feat: 0.763 G_ID: 0.105 G_Rec: 0.303 D_GP: 0.035 D_real: 1.060 D_fake: 0.786 +(epoch: 413, iters: 1710, time: 0.063) G_GAN: 0.591 G_GAN_Feat: 0.984 G_ID: 0.117 G_Rec: 0.460 D_GP: 0.030 D_real: 1.197 D_fake: 0.419 +(epoch: 413, iters: 2110, time: 0.063) G_GAN: 0.389 G_GAN_Feat: 0.701 G_ID: 0.099 G_Rec: 0.306 D_GP: 0.026 D_real: 1.297 D_fake: 0.612 +(epoch: 413, iters: 2510, time: 0.064) G_GAN: 0.286 G_GAN_Feat: 0.917 G_ID: 0.112 G_Rec: 0.421 D_GP: 0.033 D_real: 0.888 D_fake: 0.723 +(epoch: 413, iters: 2910, time: 0.063) G_GAN: 0.459 G_GAN_Feat: 0.815 G_ID: 0.105 G_Rec: 0.319 D_GP: 0.055 D_real: 0.987 D_fake: 0.561 +(epoch: 413, iters: 3310, time: 0.063) G_GAN: 0.905 G_GAN_Feat: 1.078 G_ID: 0.103 G_Rec: 0.456 D_GP: 0.052 D_real: 0.999 D_fake: 0.170 +(epoch: 413, iters: 3710, time: 0.063) G_GAN: 0.545 G_GAN_Feat: 0.879 G_ID: 0.104 G_Rec: 0.339 D_GP: 0.059 D_real: 1.197 D_fake: 0.574 +(epoch: 413, iters: 4110, time: 0.064) G_GAN: 0.454 G_GAN_Feat: 1.059 G_ID: 0.124 G_Rec: 0.438 D_GP: 0.040 D_real: 0.625 D_fake: 0.554 +(epoch: 413, iters: 4510, time: 0.063) G_GAN: 0.027 G_GAN_Feat: 0.771 G_ID: 0.121 G_Rec: 0.313 D_GP: 0.034 D_real: 0.843 D_fake: 0.973 +(epoch: 413, iters: 4910, time: 0.063) G_GAN: 0.698 G_GAN_Feat: 1.031 G_ID: 0.111 G_Rec: 0.436 D_GP: 0.068 D_real: 0.776 D_fake: 0.335 +(epoch: 413, iters: 5310, time: 0.063) G_GAN: -0.034 G_GAN_Feat: 0.890 G_ID: 0.103 G_Rec: 0.343 D_GP: 0.035 D_real: 0.696 D_fake: 1.034 +(epoch: 413, iters: 5710, time: 0.064) G_GAN: 0.285 G_GAN_Feat: 0.998 G_ID: 0.132 G_Rec: 0.448 D_GP: 0.038 D_real: 0.729 D_fake: 0.716 +(epoch: 413, iters: 6110, time: 0.063) G_GAN: 0.369 G_GAN_Feat: 0.935 G_ID: 0.107 G_Rec: 0.314 D_GP: 0.089 D_real: 0.591 D_fake: 0.632 +(epoch: 413, iters: 6510, time: 0.063) G_GAN: 0.397 G_GAN_Feat: 0.837 G_ID: 0.102 G_Rec: 0.418 D_GP: 0.033 D_real: 1.171 D_fake: 0.604 +(epoch: 413, iters: 6910, time: 0.063) G_GAN: 0.168 G_GAN_Feat: 0.693 G_ID: 0.103 G_Rec: 0.313 D_GP: 0.029 D_real: 1.082 D_fake: 0.832 +(epoch: 413, iters: 7310, time: 0.064) G_GAN: 0.452 G_GAN_Feat: 1.013 G_ID: 0.115 G_Rec: 0.486 D_GP: 0.032 D_real: 0.890 D_fake: 0.571 +(epoch: 413, iters: 7710, time: 0.063) G_GAN: -0.017 G_GAN_Feat: 0.677 G_ID: 0.102 G_Rec: 0.317 D_GP: 0.033 D_real: 0.868 D_fake: 1.017 +(epoch: 413, iters: 8110, time: 0.063) G_GAN: 0.217 G_GAN_Feat: 0.888 G_ID: 0.126 G_Rec: 0.416 D_GP: 0.037 D_real: 0.894 D_fake: 0.789 +(epoch: 413, iters: 8510, time: 0.063) G_GAN: 0.073 G_GAN_Feat: 0.745 G_ID: 0.090 G_Rec: 0.309 D_GP: 0.039 D_real: 0.858 D_fake: 0.927 +(epoch: 414, iters: 302, time: 0.064) G_GAN: 0.479 G_GAN_Feat: 0.982 G_ID: 0.145 G_Rec: 0.430 D_GP: 0.065 D_real: 0.835 D_fake: 0.564 +(epoch: 414, iters: 702, time: 0.063) G_GAN: 0.006 G_GAN_Feat: 0.746 G_ID: 0.106 G_Rec: 0.333 D_GP: 0.056 D_real: 0.618 D_fake: 0.994 +(epoch: 414, iters: 1102, time: 0.063) G_GAN: 0.445 G_GAN_Feat: 0.948 G_ID: 0.124 G_Rec: 0.455 D_GP: 0.040 D_real: 1.024 D_fake: 0.557 +(epoch: 414, iters: 1502, time: 0.063) G_GAN: -0.012 G_GAN_Feat: 0.836 G_ID: 0.124 G_Rec: 0.346 D_GP: 0.040 D_real: 0.551 D_fake: 1.012 +(epoch: 414, iters: 1902, time: 0.064) G_GAN: 0.576 G_GAN_Feat: 1.049 G_ID: 0.106 G_Rec: 0.421 D_GP: 0.066 D_real: 0.728 D_fake: 0.431 +(epoch: 414, iters: 2302, time: 0.063) G_GAN: 0.505 G_GAN_Feat: 0.803 G_ID: 0.139 G_Rec: 0.336 D_GP: 0.034 D_real: 1.241 D_fake: 0.513 +(epoch: 414, iters: 2702, time: 0.063) G_GAN: 0.412 G_GAN_Feat: 0.950 G_ID: 0.113 G_Rec: 0.440 D_GP: 0.034 D_real: 0.955 D_fake: 0.596 +(epoch: 414, iters: 3102, time: 0.063) G_GAN: 0.339 G_GAN_Feat: 0.835 G_ID: 0.105 G_Rec: 0.326 D_GP: 0.032 D_real: 1.328 D_fake: 0.667 +(epoch: 414, iters: 3502, time: 0.064) G_GAN: 0.435 G_GAN_Feat: 0.859 G_ID: 0.117 G_Rec: 0.434 D_GP: 0.028 D_real: 1.172 D_fake: 0.566 +(epoch: 414, iters: 3902, time: 0.063) G_GAN: 0.215 G_GAN_Feat: 0.716 G_ID: 0.092 G_Rec: 0.332 D_GP: 0.040 D_real: 1.043 D_fake: 0.785 +(epoch: 414, iters: 4302, time: 0.063) G_GAN: 0.554 G_GAN_Feat: 0.913 G_ID: 0.132 G_Rec: 0.418 D_GP: 0.047 D_real: 1.152 D_fake: 0.456 +(epoch: 414, iters: 4702, time: 0.063) G_GAN: 0.156 G_GAN_Feat: 0.737 G_ID: 0.090 G_Rec: 0.309 D_GP: 0.044 D_real: 0.932 D_fake: 0.844 +(epoch: 414, iters: 5102, time: 0.064) G_GAN: 0.085 G_GAN_Feat: 0.924 G_ID: 0.151 G_Rec: 0.409 D_GP: 0.036 D_real: 0.767 D_fake: 0.915 +(epoch: 414, iters: 5502, time: 0.063) G_GAN: 0.084 G_GAN_Feat: 0.805 G_ID: 0.113 G_Rec: 0.352 D_GP: 0.065 D_real: 0.771 D_fake: 0.917 +(epoch: 414, iters: 5902, time: 0.063) G_GAN: 0.699 G_GAN_Feat: 1.030 G_ID: 0.123 G_Rec: 0.416 D_GP: 0.162 D_real: 0.967 D_fake: 0.354 +(epoch: 414, iters: 6302, time: 0.063) G_GAN: 0.168 G_GAN_Feat: 0.778 G_ID: 0.099 G_Rec: 0.316 D_GP: 0.057 D_real: 0.788 D_fake: 0.832 +(epoch: 414, iters: 6702, time: 0.064) G_GAN: 0.534 G_GAN_Feat: 0.966 G_ID: 0.114 G_Rec: 0.427 D_GP: 0.032 D_real: 1.099 D_fake: 0.471 +(epoch: 414, iters: 7102, time: 0.063) G_GAN: 0.617 G_GAN_Feat: 0.818 G_ID: 0.095 G_Rec: 0.323 D_GP: 0.047 D_real: 1.413 D_fake: 0.505 +(epoch: 414, iters: 7502, time: 0.063) G_GAN: 0.629 G_GAN_Feat: 0.989 G_ID: 0.117 G_Rec: 0.432 D_GP: 0.056 D_real: 1.287 D_fake: 0.387 +(epoch: 414, iters: 7902, time: 0.063) G_GAN: 0.103 G_GAN_Feat: 0.841 G_ID: 0.106 G_Rec: 0.335 D_GP: 0.052 D_real: 0.604 D_fake: 0.897 +(epoch: 414, iters: 8302, time: 0.064) G_GAN: 0.685 G_GAN_Feat: 1.036 G_ID: 0.127 G_Rec: 0.447 D_GP: 0.036 D_real: 1.007 D_fake: 0.337 +(epoch: 414, iters: 8702, time: 0.063) G_GAN: 0.148 G_GAN_Feat: 0.859 G_ID: 0.092 G_Rec: 0.340 D_GP: 0.038 D_real: 0.745 D_fake: 0.857 +(epoch: 415, iters: 494, time: 0.063) G_GAN: 0.652 G_GAN_Feat: 1.149 G_ID: 0.149 G_Rec: 0.477 D_GP: 0.045 D_real: 1.011 D_fake: 0.390 +(epoch: 415, iters: 894, time: 0.063) G_GAN: 0.214 G_GAN_Feat: 0.651 G_ID: 0.100 G_Rec: 0.316 D_GP: 0.030 D_real: 1.127 D_fake: 0.787 +(epoch: 415, iters: 1294, time: 0.064) G_GAN: 0.458 G_GAN_Feat: 0.895 G_ID: 0.120 G_Rec: 0.479 D_GP: 0.031 D_real: 1.231 D_fake: 0.562 +(epoch: 415, iters: 1694, time: 0.063) G_GAN: 0.077 G_GAN_Feat: 0.692 G_ID: 0.092 G_Rec: 0.316 D_GP: 0.028 D_real: 1.041 D_fake: 0.924 +(epoch: 415, iters: 2094, time: 0.063) G_GAN: 0.485 G_GAN_Feat: 0.872 G_ID: 0.101 G_Rec: 0.437 D_GP: 0.033 D_real: 1.140 D_fake: 0.528 +(epoch: 415, iters: 2494, time: 0.063) G_GAN: -0.188 G_GAN_Feat: 0.651 G_ID: 0.089 G_Rec: 0.300 D_GP: 0.037 D_real: 0.769 D_fake: 1.188 +(epoch: 415, iters: 2894, time: 0.064) G_GAN: 0.229 G_GAN_Feat: 0.896 G_ID: 0.106 G_Rec: 0.450 D_GP: 0.037 D_real: 0.839 D_fake: 0.773 +(epoch: 415, iters: 3294, time: 0.063) G_GAN: -0.046 G_GAN_Feat: 0.643 G_ID: 0.100 G_Rec: 0.292 D_GP: 0.044 D_real: 0.793 D_fake: 1.046 +(epoch: 415, iters: 3694, time: 0.063) G_GAN: 0.464 G_GAN_Feat: 0.906 G_ID: 0.114 G_Rec: 0.452 D_GP: 0.041 D_real: 1.050 D_fake: 0.588 +(epoch: 415, iters: 4094, time: 0.063) G_GAN: -0.055 G_GAN_Feat: 0.742 G_ID: 0.101 G_Rec: 0.357 D_GP: 0.064 D_real: 0.668 D_fake: 1.055 +(epoch: 415, iters: 4494, time: 0.064) G_GAN: 0.917 G_GAN_Feat: 0.910 G_ID: 0.120 G_Rec: 0.448 D_GP: 0.041 D_real: 1.518 D_fake: 0.410 +(epoch: 415, iters: 4894, time: 0.063) G_GAN: -0.131 G_GAN_Feat: 0.710 G_ID: 0.093 G_Rec: 0.324 D_GP: 0.048 D_real: 0.621 D_fake: 1.131 +(epoch: 415, iters: 5294, time: 0.063) G_GAN: 0.199 G_GAN_Feat: 0.910 G_ID: 0.110 G_Rec: 0.469 D_GP: 0.036 D_real: 0.802 D_fake: 0.801 +(epoch: 415, iters: 5694, time: 0.063) G_GAN: -0.059 G_GAN_Feat: 0.736 G_ID: 0.101 G_Rec: 0.320 D_GP: 0.054 D_real: 0.745 D_fake: 1.060 +(epoch: 415, iters: 6094, time: 0.064) G_GAN: 0.226 G_GAN_Feat: 0.887 G_ID: 0.117 G_Rec: 0.407 D_GP: 0.048 D_real: 0.731 D_fake: 0.775 +(epoch: 415, iters: 6494, time: 0.063) G_GAN: 0.241 G_GAN_Feat: 0.711 G_ID: 0.101 G_Rec: 0.298 D_GP: 0.037 D_real: 1.056 D_fake: 0.760 +(epoch: 415, iters: 6894, time: 0.063) G_GAN: 0.417 G_GAN_Feat: 0.965 G_ID: 0.119 G_Rec: 0.431 D_GP: 0.056 D_real: 0.869 D_fake: 0.596 +(epoch: 415, iters: 7294, time: 0.063) G_GAN: -0.007 G_GAN_Feat: 0.838 G_ID: 0.100 G_Rec: 0.337 D_GP: 0.797 D_real: 0.414 D_fake: 1.008 +(epoch: 415, iters: 7694, time: 0.064) G_GAN: 0.512 G_GAN_Feat: 1.013 G_ID: 0.117 G_Rec: 0.481 D_GP: 0.081 D_real: 0.942 D_fake: 0.500 +(epoch: 415, iters: 8094, time: 0.063) G_GAN: 0.041 G_GAN_Feat: 0.730 G_ID: 0.105 G_Rec: 0.303 D_GP: 0.048 D_real: 0.781 D_fake: 0.959 +(epoch: 415, iters: 8494, time: 0.063) G_GAN: 0.305 G_GAN_Feat: 0.946 G_ID: 0.130 G_Rec: 0.461 D_GP: 0.050 D_real: 0.885 D_fake: 0.695 +(epoch: 416, iters: 286, time: 0.063) G_GAN: 0.517 G_GAN_Feat: 0.833 G_ID: 0.101 G_Rec: 0.352 D_GP: 0.056 D_real: 1.210 D_fake: 0.496 +(epoch: 416, iters: 686, time: 0.064) G_GAN: 0.725 G_GAN_Feat: 0.912 G_ID: 0.111 G_Rec: 0.449 D_GP: 0.033 D_real: 1.306 D_fake: 0.294 +(epoch: 416, iters: 1086, time: 0.063) G_GAN: 0.153 G_GAN_Feat: 0.864 G_ID: 0.097 G_Rec: 0.358 D_GP: 0.140 D_real: 0.351 D_fake: 0.852 +(epoch: 416, iters: 1486, time: 0.063) G_GAN: 0.852 G_GAN_Feat: 0.936 G_ID: 0.099 G_Rec: 0.401 D_GP: 0.036 D_real: 1.466 D_fake: 0.202 +(epoch: 416, iters: 1886, time: 0.063) G_GAN: 0.008 G_GAN_Feat: 0.603 G_ID: 0.100 G_Rec: 0.306 D_GP: 0.028 D_real: 0.990 D_fake: 0.992 +(epoch: 416, iters: 2286, time: 0.063) G_GAN: 0.275 G_GAN_Feat: 0.792 G_ID: 0.130 G_Rec: 0.406 D_GP: 0.026 D_real: 1.134 D_fake: 0.727 +(epoch: 416, iters: 2686, time: 0.063) G_GAN: 0.062 G_GAN_Feat: 0.656 G_ID: 0.089 G_Rec: 0.300 D_GP: 0.029 D_real: 0.924 D_fake: 0.938 +(epoch: 416, iters: 3086, time: 0.063) G_GAN: 0.228 G_GAN_Feat: 0.917 G_ID: 0.121 G_Rec: 0.432 D_GP: 0.057 D_real: 0.926 D_fake: 0.779 +(epoch: 416, iters: 3486, time: 0.064) G_GAN: -0.010 G_GAN_Feat: 0.791 G_ID: 0.093 G_Rec: 0.327 D_GP: 0.092 D_real: 0.562 D_fake: 1.010 +(epoch: 416, iters: 3886, time: 0.063) G_GAN: 0.653 G_GAN_Feat: 0.989 G_ID: 0.115 G_Rec: 0.481 D_GP: 0.038 D_real: 1.109 D_fake: 0.381 +(epoch: 416, iters: 4286, time: 0.063) G_GAN: 0.080 G_GAN_Feat: 0.725 G_ID: 0.094 G_Rec: 0.302 D_GP: 0.051 D_real: 0.879 D_fake: 0.920 +(epoch: 416, iters: 4686, time: 0.063) G_GAN: 0.567 G_GAN_Feat: 0.840 G_ID: 0.111 G_Rec: 0.421 D_GP: 0.036 D_real: 1.199 D_fake: 0.438 +(epoch: 416, iters: 5086, time: 0.064) G_GAN: 0.503 G_GAN_Feat: 0.719 G_ID: 0.094 G_Rec: 0.306 D_GP: 0.044 D_real: 1.262 D_fake: 0.501 +(epoch: 416, iters: 5486, time: 0.063) G_GAN: 0.429 G_GAN_Feat: 0.967 G_ID: 0.117 G_Rec: 0.464 D_GP: 0.048 D_real: 0.954 D_fake: 0.575 +(epoch: 416, iters: 5886, time: 0.063) G_GAN: -0.007 G_GAN_Feat: 0.799 G_ID: 0.102 G_Rec: 0.358 D_GP: 0.036 D_real: 0.751 D_fake: 1.007 +(epoch: 416, iters: 6286, time: 0.063) G_GAN: 0.484 G_GAN_Feat: 0.900 G_ID: 0.137 G_Rec: 0.411 D_GP: 0.043 D_real: 1.019 D_fake: 0.521 +(epoch: 416, iters: 6686, time: 0.064) G_GAN: 0.308 G_GAN_Feat: 0.788 G_ID: 0.089 G_Rec: 0.321 D_GP: 0.045 D_real: 0.953 D_fake: 0.692 +(epoch: 416, iters: 7086, time: 0.063) G_GAN: 0.537 G_GAN_Feat: 1.008 G_ID: 0.128 G_Rec: 0.418 D_GP: 0.071 D_real: 0.856 D_fake: 0.474 +(epoch: 416, iters: 7486, time: 0.063) G_GAN: 0.425 G_GAN_Feat: 0.851 G_ID: 0.091 G_Rec: 0.313 D_GP: 0.050 D_real: 0.874 D_fake: 0.576 +(epoch: 416, iters: 7886, time: 0.063) G_GAN: 0.410 G_GAN_Feat: 1.245 G_ID: 0.130 G_Rec: 0.468 D_GP: 0.065 D_real: 0.885 D_fake: 0.613 +(epoch: 416, iters: 8286, time: 0.064) G_GAN: -0.103 G_GAN_Feat: 0.689 G_ID: 0.097 G_Rec: 0.318 D_GP: 0.028 D_real: 0.861 D_fake: 1.103 +(epoch: 416, iters: 8686, time: 0.063) G_GAN: 0.139 G_GAN_Feat: 0.872 G_ID: 0.112 G_Rec: 0.425 D_GP: 0.029 D_real: 0.788 D_fake: 0.863 +(epoch: 417, iters: 478, time: 0.063) G_GAN: -0.050 G_GAN_Feat: 0.666 G_ID: 0.122 G_Rec: 0.289 D_GP: 0.036 D_real: 0.826 D_fake: 1.050 +(epoch: 417, iters: 878, time: 0.063) G_GAN: 0.395 G_GAN_Feat: 0.916 G_ID: 0.110 G_Rec: 0.451 D_GP: 0.043 D_real: 0.902 D_fake: 0.628 +(epoch: 417, iters: 1278, time: 0.064) G_GAN: -0.024 G_GAN_Feat: 0.668 G_ID: 0.097 G_Rec: 0.308 D_GP: 0.039 D_real: 0.854 D_fake: 1.024 +(epoch: 417, iters: 1678, time: 0.063) G_GAN: 0.534 G_GAN_Feat: 0.942 G_ID: 0.114 G_Rec: 0.478 D_GP: 0.035 D_real: 1.074 D_fake: 0.507 +(epoch: 417, iters: 2078, time: 0.063) G_GAN: -0.031 G_GAN_Feat: 0.671 G_ID: 0.109 G_Rec: 0.325 D_GP: 0.029 D_real: 0.864 D_fake: 1.031 +(epoch: 417, iters: 2478, time: 0.063) G_GAN: 0.150 G_GAN_Feat: 0.942 G_ID: 0.127 G_Rec: 0.477 D_GP: 0.042 D_real: 0.674 D_fake: 0.852 +(epoch: 417, iters: 2878, time: 0.064) G_GAN: 0.002 G_GAN_Feat: 0.649 G_ID: 0.112 G_Rec: 0.282 D_GP: 0.031 D_real: 0.964 D_fake: 0.999 +(epoch: 417, iters: 3278, time: 0.063) G_GAN: 0.501 G_GAN_Feat: 0.949 G_ID: 0.154 G_Rec: 0.452 D_GP: 0.046 D_real: 0.993 D_fake: 0.523 +(epoch: 417, iters: 3678, time: 0.063) G_GAN: 0.271 G_GAN_Feat: 0.776 G_ID: 0.095 G_Rec: 0.334 D_GP: 0.051 D_real: 1.045 D_fake: 0.732 +(epoch: 417, iters: 4078, time: 0.063) G_GAN: 0.543 G_GAN_Feat: 0.860 G_ID: 0.113 G_Rec: 0.385 D_GP: 0.032 D_real: 1.276 D_fake: 0.470 +(epoch: 417, iters: 4478, time: 0.064) G_GAN: 0.227 G_GAN_Feat: 0.762 G_ID: 0.087 G_Rec: 0.305 D_GP: 0.045 D_real: 0.932 D_fake: 0.775 +(epoch: 417, iters: 4878, time: 0.063) G_GAN: 0.496 G_GAN_Feat: 0.930 G_ID: 0.116 G_Rec: 0.440 D_GP: 0.035 D_real: 1.145 D_fake: 0.512 +(epoch: 417, iters: 5278, time: 0.063) G_GAN: 0.216 G_GAN_Feat: 0.733 G_ID: 0.093 G_Rec: 0.304 D_GP: 0.040 D_real: 1.056 D_fake: 0.786 +(epoch: 417, iters: 5678, time: 0.063) G_GAN: 0.486 G_GAN_Feat: 0.925 G_ID: 0.113 G_Rec: 0.448 D_GP: 0.063 D_real: 0.930 D_fake: 0.530 +(epoch: 417, iters: 6078, time: 0.064) G_GAN: 0.364 G_GAN_Feat: 0.789 G_ID: 0.088 G_Rec: 0.307 D_GP: 0.064 D_real: 0.879 D_fake: 0.637 +(epoch: 417, iters: 6478, time: 0.063) G_GAN: 0.549 G_GAN_Feat: 0.875 G_ID: 0.125 G_Rec: 0.391 D_GP: 0.031 D_real: 1.153 D_fake: 0.467 +(epoch: 417, iters: 6878, time: 0.063) G_GAN: -0.040 G_GAN_Feat: 0.815 G_ID: 0.113 G_Rec: 0.340 D_GP: 0.098 D_real: 0.464 D_fake: 1.040 +(epoch: 417, iters: 7278, time: 0.063) G_GAN: 0.528 G_GAN_Feat: 1.117 G_ID: 0.108 G_Rec: 0.492 D_GP: 0.110 D_real: 0.744 D_fake: 0.487 +(epoch: 417, iters: 7678, time: 0.064) G_GAN: 0.241 G_GAN_Feat: 0.960 G_ID: 0.115 G_Rec: 0.356 D_GP: 0.054 D_real: 0.811 D_fake: 0.870 +(epoch: 417, iters: 8078, time: 0.063) G_GAN: 0.703 G_GAN_Feat: 0.977 G_ID: 0.110 G_Rec: 0.430 D_GP: 0.033 D_real: 1.298 D_fake: 0.333 +(epoch: 417, iters: 8478, time: 0.063) G_GAN: -0.024 G_GAN_Feat: 0.938 G_ID: 0.102 G_Rec: 0.322 D_GP: 3.707 D_real: 0.330 D_fake: 1.025 +(epoch: 418, iters: 270, time: 0.063) G_GAN: 0.360 G_GAN_Feat: 0.882 G_ID: 0.122 G_Rec: 0.441 D_GP: 0.029 D_real: 1.077 D_fake: 0.640 +(epoch: 418, iters: 670, time: 0.064) G_GAN: -0.062 G_GAN_Feat: 0.643 G_ID: 0.140 G_Rec: 0.328 D_GP: 0.028 D_real: 0.808 D_fake: 1.062 +(epoch: 418, iters: 1070, time: 0.063) G_GAN: 0.241 G_GAN_Feat: 0.837 G_ID: 0.116 G_Rec: 0.409 D_GP: 0.035 D_real: 0.984 D_fake: 0.760 +(epoch: 418, iters: 1470, time: 0.063) G_GAN: -0.007 G_GAN_Feat: 0.603 G_ID: 0.104 G_Rec: 0.285 D_GP: 0.034 D_real: 0.902 D_fake: 1.007 +(epoch: 418, iters: 1870, time: 0.063) G_GAN: 0.316 G_GAN_Feat: 0.810 G_ID: 0.126 G_Rec: 0.416 D_GP: 0.038 D_real: 1.006 D_fake: 0.698 +(epoch: 418, iters: 2270, time: 0.064) G_GAN: -0.061 G_GAN_Feat: 0.698 G_ID: 0.101 G_Rec: 0.323 D_GP: 0.054 D_real: 0.768 D_fake: 1.061 +(epoch: 418, iters: 2670, time: 0.064) G_GAN: 0.247 G_GAN_Feat: 0.876 G_ID: 0.121 G_Rec: 0.433 D_GP: 0.046 D_real: 0.971 D_fake: 0.763 +(epoch: 418, iters: 3070, time: 0.063) G_GAN: -0.143 G_GAN_Feat: 0.669 G_ID: 0.078 G_Rec: 0.312 D_GP: 0.049 D_real: 0.664 D_fake: 1.143 +(epoch: 418, iters: 3470, time: 0.063) G_GAN: 0.296 G_GAN_Feat: 0.928 G_ID: 0.120 G_Rec: 0.462 D_GP: 0.044 D_real: 0.926 D_fake: 0.713 +(epoch: 418, iters: 3870, time: 0.064) G_GAN: 0.251 G_GAN_Feat: 0.676 G_ID: 0.110 G_Rec: 0.290 D_GP: 0.037 D_real: 1.156 D_fake: 0.762 +(epoch: 418, iters: 4270, time: 0.063) G_GAN: 0.485 G_GAN_Feat: 0.961 G_ID: 0.107 G_Rec: 0.457 D_GP: 0.042 D_real: 1.220 D_fake: 0.540 +(epoch: 418, iters: 4670, time: 0.063) G_GAN: 0.526 G_GAN_Feat: 0.783 G_ID: 0.119 G_Rec: 0.343 D_GP: 0.033 D_real: 1.416 D_fake: 0.481 +(epoch: 418, iters: 5070, time: 0.063) G_GAN: 0.316 G_GAN_Feat: 0.952 G_ID: 0.120 G_Rec: 0.439 D_GP: 0.082 D_real: 0.763 D_fake: 0.691 +(epoch: 418, iters: 5470, time: 0.064) G_GAN: 0.239 G_GAN_Feat: 0.669 G_ID: 0.096 G_Rec: 0.298 D_GP: 0.040 D_real: 1.127 D_fake: 0.761 +(epoch: 418, iters: 5870, time: 0.063) G_GAN: 0.342 G_GAN_Feat: 1.010 G_ID: 0.126 G_Rec: 0.472 D_GP: 0.052 D_real: 0.840 D_fake: 0.659 +(epoch: 418, iters: 6270, time: 0.063) G_GAN: 0.094 G_GAN_Feat: 0.656 G_ID: 0.099 G_Rec: 0.302 D_GP: 0.028 D_real: 1.014 D_fake: 0.909 +(epoch: 418, iters: 6670, time: 0.063) G_GAN: 0.448 G_GAN_Feat: 1.090 G_ID: 0.124 G_Rec: 0.464 D_GP: 0.085 D_real: 0.463 D_fake: 0.557 +(epoch: 418, iters: 7070, time: 0.064) G_GAN: 0.449 G_GAN_Feat: 0.743 G_ID: 0.091 G_Rec: 0.311 D_GP: 0.065 D_real: 1.264 D_fake: 0.587 +(epoch: 418, iters: 7470, time: 0.063) G_GAN: 0.487 G_GAN_Feat: 0.970 G_ID: 0.118 G_Rec: 0.455 D_GP: 0.091 D_real: 1.075 D_fake: 0.523 +(epoch: 418, iters: 7870, time: 0.063) G_GAN: 0.056 G_GAN_Feat: 0.660 G_ID: 0.110 G_Rec: 0.374 D_GP: 0.035 D_real: 0.955 D_fake: 0.944 +(epoch: 418, iters: 8270, time: 0.063) G_GAN: 0.142 G_GAN_Feat: 0.943 G_ID: 0.111 G_Rec: 0.450 D_GP: 0.050 D_real: 0.777 D_fake: 0.858 +(epoch: 418, iters: 8670, time: 0.064) G_GAN: 0.262 G_GAN_Feat: 0.760 G_ID: 0.107 G_Rec: 0.318 D_GP: 0.034 D_real: 1.026 D_fake: 0.738 +(epoch: 419, iters: 462, time: 0.063) G_GAN: 0.383 G_GAN_Feat: 0.891 G_ID: 0.124 G_Rec: 0.419 D_GP: 0.031 D_real: 0.951 D_fake: 0.618 +(epoch: 419, iters: 862, time: 0.063) G_GAN: 0.431 G_GAN_Feat: 0.656 G_ID: 0.100 G_Rec: 0.270 D_GP: 0.025 D_real: 1.312 D_fake: 0.571 +(epoch: 419, iters: 1262, time: 0.063) G_GAN: 0.541 G_GAN_Feat: 0.906 G_ID: 0.104 G_Rec: 0.408 D_GP: 0.034 D_real: 1.124 D_fake: 0.467 +(epoch: 419, iters: 1662, time: 0.064) G_GAN: 0.272 G_GAN_Feat: 0.739 G_ID: 0.105 G_Rec: 0.281 D_GP: 0.051 D_real: 1.160 D_fake: 0.728 +(epoch: 419, iters: 2062, time: 0.063) G_GAN: 0.835 G_GAN_Feat: 0.984 G_ID: 0.114 G_Rec: 0.470 D_GP: 0.037 D_real: 1.401 D_fake: 0.250 +(epoch: 419, iters: 2462, time: 0.063) G_GAN: 0.089 G_GAN_Feat: 0.718 G_ID: 0.096 G_Rec: 0.328 D_GP: 0.036 D_real: 0.977 D_fake: 0.911 +(epoch: 419, iters: 2862, time: 0.063) G_GAN: 0.512 G_GAN_Feat: 0.891 G_ID: 0.101 G_Rec: 0.391 D_GP: 0.038 D_real: 1.121 D_fake: 0.507 +(epoch: 419, iters: 3262, time: 0.064) G_GAN: 0.195 G_GAN_Feat: 0.746 G_ID: 0.096 G_Rec: 0.320 D_GP: 0.042 D_real: 1.021 D_fake: 0.805 +(epoch: 419, iters: 3662, time: 0.063) G_GAN: 0.500 G_GAN_Feat: 0.911 G_ID: 0.112 G_Rec: 0.422 D_GP: 0.031 D_real: 1.218 D_fake: 0.512 +(epoch: 419, iters: 4062, time: 0.064) G_GAN: 0.217 G_GAN_Feat: 0.710 G_ID: 0.122 G_Rec: 0.290 D_GP: 0.032 D_real: 1.060 D_fake: 0.784 +(epoch: 419, iters: 4462, time: 0.064) G_GAN: 0.474 G_GAN_Feat: 0.914 G_ID: 0.126 G_Rec: 0.392 D_GP: 0.037 D_real: 1.051 D_fake: 0.529 +(epoch: 419, iters: 4862, time: 0.064) G_GAN: -0.112 G_GAN_Feat: 0.785 G_ID: 0.120 G_Rec: 0.325 D_GP: 0.042 D_real: 0.600 D_fake: 1.112 +(epoch: 419, iters: 5262, time: 0.064) G_GAN: 0.134 G_GAN_Feat: 0.892 G_ID: 0.119 G_Rec: 0.376 D_GP: 0.030 D_real: 0.707 D_fake: 0.866 +(epoch: 419, iters: 5662, time: 0.064) G_GAN: 0.325 G_GAN_Feat: 0.732 G_ID: 0.103 G_Rec: 0.294 D_GP: 0.053 D_real: 1.033 D_fake: 0.675 +(epoch: 419, iters: 6062, time: 0.064) G_GAN: 0.436 G_GAN_Feat: 1.059 G_ID: 0.131 G_Rec: 0.459 D_GP: 0.062 D_real: 0.647 D_fake: 0.569 +(epoch: 419, iters: 6462, time: 0.064) G_GAN: 0.246 G_GAN_Feat: 0.812 G_ID: 0.105 G_Rec: 0.322 D_GP: 0.046 D_real: 0.995 D_fake: 0.756 +(epoch: 419, iters: 6862, time: 0.064) G_GAN: 0.800 G_GAN_Feat: 1.035 G_ID: 0.132 G_Rec: 0.437 D_GP: 0.064 D_real: 0.818 D_fake: 0.255 +(epoch: 419, iters: 7262, time: 0.064) G_GAN: -0.239 G_GAN_Feat: 0.832 G_ID: 0.100 G_Rec: 0.321 D_GP: 0.048 D_real: 0.546 D_fake: 1.240 +(epoch: 419, iters: 7662, time: 0.064) G_GAN: 0.429 G_GAN_Feat: 1.029 G_ID: 0.127 G_Rec: 0.433 D_GP: 0.034 D_real: 0.831 D_fake: 0.572 +(epoch: 419, iters: 8062, time: 0.064) G_GAN: 0.229 G_GAN_Feat: 0.716 G_ID: 0.099 G_Rec: 0.316 D_GP: 0.039 D_real: 1.180 D_fake: 0.775 +(epoch: 419, iters: 8462, time: 0.064) G_GAN: 0.292 G_GAN_Feat: 0.951 G_ID: 0.114 G_Rec: 0.431 D_GP: 0.029 D_real: 0.974 D_fake: 0.708 +(epoch: 420, iters: 254, time: 0.064) G_GAN: 0.346 G_GAN_Feat: 0.723 G_ID: 0.106 G_Rec: 0.274 D_GP: 0.036 D_real: 1.177 D_fake: 0.654 +(epoch: 420, iters: 654, time: 0.064) G_GAN: 0.510 G_GAN_Feat: 0.924 G_ID: 0.124 G_Rec: 0.430 D_GP: 0.034 D_real: 1.078 D_fake: 0.509 +(epoch: 420, iters: 1054, time: 0.064) G_GAN: 0.353 G_GAN_Feat: 0.793 G_ID: 0.096 G_Rec: 0.313 D_GP: 0.057 D_real: 0.953 D_fake: 0.651 +(epoch: 420, iters: 1454, time: 0.063) G_GAN: 0.533 G_GAN_Feat: 0.929 G_ID: 0.117 G_Rec: 0.486 D_GP: 0.027 D_real: 1.179 D_fake: 0.473 +(epoch: 420, iters: 1854, time: 0.063) G_GAN: 0.153 G_GAN_Feat: 0.796 G_ID: 0.105 G_Rec: 0.314 D_GP: 0.097 D_real: 0.840 D_fake: 0.848 +(epoch: 420, iters: 2254, time: 0.063) G_GAN: 0.617 G_GAN_Feat: 0.916 G_ID: 0.112 G_Rec: 0.423 D_GP: 0.034 D_real: 1.198 D_fake: 0.395 +(epoch: 420, iters: 2654, time: 0.064) G_GAN: 0.436 G_GAN_Feat: 0.801 G_ID: 0.087 G_Rec: 0.310 D_GP: 0.051 D_real: 1.149 D_fake: 0.572 +(epoch: 420, iters: 3054, time: 0.063) G_GAN: 0.931 G_GAN_Feat: 1.168 G_ID: 0.113 G_Rec: 0.467 D_GP: 0.057 D_real: 0.877 D_fake: 0.237 +(epoch: 420, iters: 3454, time: 0.063) G_GAN: 0.242 G_GAN_Feat: 0.901 G_ID: 0.110 G_Rec: 0.322 D_GP: 0.077 D_real: 0.493 D_fake: 0.761 +(epoch: 420, iters: 3854, time: 0.063) G_GAN: 0.454 G_GAN_Feat: 1.039 G_ID: 0.125 G_Rec: 0.499 D_GP: 0.030 D_real: 1.089 D_fake: 0.558 +(epoch: 420, iters: 4254, time: 0.064) G_GAN: 0.365 G_GAN_Feat: 0.752 G_ID: 0.095 G_Rec: 0.283 D_GP: 0.037 D_real: 1.136 D_fake: 0.637 +(epoch: 420, iters: 4654, time: 0.063) G_GAN: 1.416 G_GAN_Feat: 1.097 G_ID: 0.132 G_Rec: 0.460 D_GP: 0.121 D_real: 1.590 D_fake: 0.148 +(epoch: 420, iters: 5054, time: 0.063) G_GAN: 0.045 G_GAN_Feat: 0.731 G_ID: 0.104 G_Rec: 0.308 D_GP: 0.030 D_real: 0.996 D_fake: 0.955 +(epoch: 420, iters: 5454, time: 0.063) G_GAN: 0.241 G_GAN_Feat: 0.929 G_ID: 0.116 G_Rec: 0.418 D_GP: 0.046 D_real: 0.877 D_fake: 0.762 +(epoch: 420, iters: 5854, time: 0.064) G_GAN: 0.280 G_GAN_Feat: 0.882 G_ID: 0.102 G_Rec: 0.336 D_GP: 0.083 D_real: 0.611 D_fake: 0.728 +(epoch: 420, iters: 6254, time: 0.063) G_GAN: 0.442 G_GAN_Feat: 0.936 G_ID: 0.138 G_Rec: 0.457 D_GP: 0.035 D_real: 1.051 D_fake: 0.561 +(epoch: 420, iters: 6654, time: 0.063) G_GAN: 0.237 G_GAN_Feat: 0.701 G_ID: 0.086 G_Rec: 0.307 D_GP: 0.047 D_real: 1.140 D_fake: 0.763 +(epoch: 420, iters: 7054, time: 0.063) G_GAN: 0.643 G_GAN_Feat: 0.952 G_ID: 0.137 G_Rec: 0.463 D_GP: 0.038 D_real: 1.136 D_fake: 0.383 +(epoch: 420, iters: 7454, time: 0.063) G_GAN: 0.407 G_GAN_Feat: 0.643 G_ID: 0.090 G_Rec: 0.271 D_GP: 0.043 D_real: 1.285 D_fake: 0.594 +(epoch: 420, iters: 7854, time: 0.063) G_GAN: 0.480 G_GAN_Feat: 0.925 G_ID: 0.126 G_Rec: 0.391 D_GP: 0.043 D_real: 1.072 D_fake: 0.529 +(epoch: 420, iters: 8254, time: 0.063) G_GAN: 0.503 G_GAN_Feat: 0.933 G_ID: 0.098 G_Rec: 0.353 D_GP: 0.282 D_real: 0.645 D_fake: 0.526 +(epoch: 420, iters: 8654, time: 0.064) G_GAN: 0.643 G_GAN_Feat: 1.010 G_ID: 0.116 G_Rec: 0.439 D_GP: 0.037 D_real: 1.175 D_fake: 0.371 +(epoch: 421, iters: 446, time: 0.063) G_GAN: 0.376 G_GAN_Feat: 0.771 G_ID: 0.122 G_Rec: 0.327 D_GP: 0.037 D_real: 1.211 D_fake: 0.640 +(epoch: 421, iters: 846, time: 0.063) G_GAN: 0.577 G_GAN_Feat: 1.018 G_ID: 0.120 G_Rec: 0.499 D_GP: 0.135 D_real: 0.932 D_fake: 0.431 +(epoch: 421, iters: 1246, time: 0.063) G_GAN: 0.283 G_GAN_Feat: 0.789 G_ID: 0.098 G_Rec: 0.318 D_GP: 0.051 D_real: 0.818 D_fake: 0.718 +(epoch: 421, iters: 1646, time: 0.064) G_GAN: 0.443 G_GAN_Feat: 1.042 G_ID: 0.131 G_Rec: 0.459 D_GP: 0.065 D_real: 0.620 D_fake: 0.563 +(epoch: 421, iters: 2046, time: 0.063) G_GAN: 0.112 G_GAN_Feat: 0.854 G_ID: 0.130 G_Rec: 0.334 D_GP: 0.044 D_real: 0.822 D_fake: 0.889 +(epoch: 421, iters: 2446, time: 0.063) G_GAN: 0.676 G_GAN_Feat: 1.279 G_ID: 0.128 G_Rec: 0.467 D_GP: 0.047 D_real: 0.531 D_fake: 0.345 +(epoch: 421, iters: 2846, time: 0.063) G_GAN: 0.250 G_GAN_Feat: 0.842 G_ID: 0.119 G_Rec: 0.339 D_GP: 0.032 D_real: 1.092 D_fake: 0.751 +(epoch: 421, iters: 3246, time: 0.064) G_GAN: 0.535 G_GAN_Feat: 0.991 G_ID: 0.122 G_Rec: 0.458 D_GP: 0.034 D_real: 1.111 D_fake: 0.491 +(epoch: 421, iters: 3646, time: 0.063) G_GAN: -0.159 G_GAN_Feat: 0.712 G_ID: 0.095 G_Rec: 0.301 D_GP: 0.040 D_real: 0.669 D_fake: 1.159 +(epoch: 421, iters: 4046, time: 0.063) G_GAN: 0.416 G_GAN_Feat: 0.971 G_ID: 0.136 G_Rec: 0.484 D_GP: 0.033 D_real: 1.027 D_fake: 0.590 +(epoch: 421, iters: 4446, time: 0.063) G_GAN: 0.431 G_GAN_Feat: 0.990 G_ID: 0.098 G_Rec: 0.338 D_GP: 0.057 D_real: 0.657 D_fake: 0.574 +(epoch: 421, iters: 4846, time: 0.064) G_GAN: 1.212 G_GAN_Feat: 1.140 G_ID: 0.105 G_Rec: 0.446 D_GP: 0.179 D_real: 1.433 D_fake: 0.236 +(epoch: 421, iters: 5246, time: 0.063) G_GAN: 0.358 G_GAN_Feat: 1.009 G_ID: 0.097 G_Rec: 0.334 D_GP: 0.644 D_real: 0.382 D_fake: 0.708 +(epoch: 421, iters: 5646, time: 0.063) G_GAN: 0.298 G_GAN_Feat: 1.038 G_ID: 0.133 G_Rec: 0.468 D_GP: 0.072 D_real: 0.572 D_fake: 0.706 +(epoch: 421, iters: 6046, time: 0.063) G_GAN: 0.347 G_GAN_Feat: 0.963 G_ID: 0.097 G_Rec: 0.349 D_GP: 0.078 D_real: 0.379 D_fake: 0.665 +(epoch: 421, iters: 6446, time: 0.064) G_GAN: 0.369 G_GAN_Feat: 1.223 G_ID: 0.165 G_Rec: 0.464 D_GP: 0.069 D_real: 0.773 D_fake: 0.711 +(epoch: 421, iters: 6846, time: 0.063) G_GAN: 0.190 G_GAN_Feat: 0.975 G_ID: 0.099 G_Rec: 0.364 D_GP: 0.072 D_real: 0.331 D_fake: 0.812 +(epoch: 421, iters: 7246, time: 0.063) G_GAN: 0.490 G_GAN_Feat: 1.242 G_ID: 0.153 G_Rec: 0.485 D_GP: 0.060 D_real: 0.315 D_fake: 0.515 +(epoch: 421, iters: 7646, time: 0.063) G_GAN: 0.534 G_GAN_Feat: 0.817 G_ID: 0.087 G_Rec: 0.348 D_GP: 0.037 D_real: 1.109 D_fake: 0.480 +(epoch: 421, iters: 8046, time: 0.064) G_GAN: 0.241 G_GAN_Feat: 0.802 G_ID: 0.120 G_Rec: 0.422 D_GP: 0.026 D_real: 1.051 D_fake: 0.759 +(epoch: 421, iters: 8446, time: 0.063) G_GAN: 0.155 G_GAN_Feat: 0.579 G_ID: 0.093 G_Rec: 0.284 D_GP: 0.026 D_real: 1.107 D_fake: 0.845 +(epoch: 422, iters: 238, time: 0.063) G_GAN: 0.459 G_GAN_Feat: 0.847 G_ID: 0.120 G_Rec: 0.413 D_GP: 0.030 D_real: 1.177 D_fake: 0.547 +(epoch: 422, iters: 638, time: 0.063) G_GAN: 0.253 G_GAN_Feat: 0.612 G_ID: 0.095 G_Rec: 0.344 D_GP: 0.030 D_real: 1.178 D_fake: 0.748 +(epoch: 422, iters: 1038, time: 0.064) G_GAN: 0.447 G_GAN_Feat: 0.880 G_ID: 0.113 G_Rec: 0.422 D_GP: 0.047 D_real: 1.069 D_fake: 0.566 +(epoch: 422, iters: 1438, time: 0.063) G_GAN: 0.181 G_GAN_Feat: 0.617 G_ID: 0.096 G_Rec: 0.287 D_GP: 0.035 D_real: 1.117 D_fake: 0.820 +(epoch: 422, iters: 1838, time: 0.063) G_GAN: 0.516 G_GAN_Feat: 0.887 G_ID: 0.101 G_Rec: 0.411 D_GP: 0.039 D_real: 1.112 D_fake: 0.517 +(epoch: 422, iters: 2238, time: 0.063) G_GAN: -0.082 G_GAN_Feat: 0.618 G_ID: 0.095 G_Rec: 0.293 D_GP: 0.030 D_real: 0.766 D_fake: 1.082 +(epoch: 422, iters: 2638, time: 0.064) G_GAN: 0.416 G_GAN_Feat: 0.882 G_ID: 0.112 G_Rec: 0.432 D_GP: 0.037 D_real: 1.122 D_fake: 0.599 +(epoch: 422, iters: 3038, time: 0.063) G_GAN: -0.025 G_GAN_Feat: 0.657 G_ID: 0.114 G_Rec: 0.302 D_GP: 0.035 D_real: 0.796 D_fake: 1.025 +(epoch: 422, iters: 3438, time: 0.063) G_GAN: 0.311 G_GAN_Feat: 0.906 G_ID: 0.107 G_Rec: 0.426 D_GP: 0.036 D_real: 0.897 D_fake: 0.696 +(epoch: 422, iters: 3838, time: 0.063) G_GAN: 0.108 G_GAN_Feat: 0.708 G_ID: 0.101 G_Rec: 0.341 D_GP: 0.038 D_real: 0.949 D_fake: 0.894 +(epoch: 422, iters: 4238, time: 0.064) G_GAN: 0.216 G_GAN_Feat: 0.882 G_ID: 0.127 G_Rec: 0.446 D_GP: 0.034 D_real: 0.903 D_fake: 0.784 +(epoch: 422, iters: 4638, time: 0.063) G_GAN: -0.196 G_GAN_Feat: 0.767 G_ID: 0.119 G_Rec: 0.364 D_GP: 0.067 D_real: 0.471 D_fake: 1.196 +(epoch: 422, iters: 5038, time: 0.063) G_GAN: 0.376 G_GAN_Feat: 0.966 G_ID: 0.126 G_Rec: 0.490 D_GP: 0.039 D_real: 0.916 D_fake: 0.633 +(epoch: 422, iters: 5438, time: 0.063) G_GAN: 0.125 G_GAN_Feat: 0.697 G_ID: 0.116 G_Rec: 0.307 D_GP: 0.041 D_real: 0.892 D_fake: 0.875 +(epoch: 422, iters: 5838, time: 0.064) G_GAN: 0.228 G_GAN_Feat: 0.907 G_ID: 0.115 G_Rec: 0.456 D_GP: 0.036 D_real: 0.861 D_fake: 0.773 +(epoch: 422, iters: 6238, time: 0.063) G_GAN: 0.227 G_GAN_Feat: 0.691 G_ID: 0.100 G_Rec: 0.302 D_GP: 0.031 D_real: 1.094 D_fake: 0.774 +(epoch: 422, iters: 6638, time: 0.063) G_GAN: 0.458 G_GAN_Feat: 0.966 G_ID: 0.112 G_Rec: 0.439 D_GP: 0.055 D_real: 1.055 D_fake: 0.544 +(epoch: 422, iters: 7038, time: 0.063) G_GAN: 0.202 G_GAN_Feat: 0.831 G_ID: 0.102 G_Rec: 0.333 D_GP: 0.126 D_real: 0.808 D_fake: 0.800 +(epoch: 422, iters: 7438, time: 0.064) G_GAN: 0.351 G_GAN_Feat: 0.916 G_ID: 0.128 G_Rec: 0.422 D_GP: 0.031 D_real: 1.007 D_fake: 0.650 +(epoch: 422, iters: 7838, time: 0.063) G_GAN: 0.234 G_GAN_Feat: 0.604 G_ID: 0.096 G_Rec: 0.291 D_GP: 0.027 D_real: 1.219 D_fake: 0.766 +(epoch: 422, iters: 8238, time: 0.063) G_GAN: 0.485 G_GAN_Feat: 0.915 G_ID: 0.118 G_Rec: 0.425 D_GP: 0.032 D_real: 1.266 D_fake: 0.520 +(epoch: 422, iters: 8638, time: 0.063) G_GAN: 0.208 G_GAN_Feat: 0.704 G_ID: 0.102 G_Rec: 0.320 D_GP: 0.035 D_real: 1.066 D_fake: 0.792 +(epoch: 423, iters: 430, time: 0.064) G_GAN: 0.622 G_GAN_Feat: 0.925 G_ID: 0.109 G_Rec: 0.424 D_GP: 0.050 D_real: 1.197 D_fake: 0.395 +(epoch: 423, iters: 830, time: 0.063) G_GAN: -0.040 G_GAN_Feat: 0.766 G_ID: 0.108 G_Rec: 0.358 D_GP: 0.062 D_real: 0.821 D_fake: 1.040 +(epoch: 423, iters: 1230, time: 0.063) G_GAN: 0.518 G_GAN_Feat: 0.987 G_ID: 0.112 G_Rec: 0.484 D_GP: 0.068 D_real: 0.872 D_fake: 0.509 +(epoch: 423, iters: 1630, time: 0.063) G_GAN: 0.139 G_GAN_Feat: 0.756 G_ID: 0.099 G_Rec: 0.295 D_GP: 0.051 D_real: 0.779 D_fake: 0.861 +(epoch: 423, iters: 2030, time: 0.064) G_GAN: 0.418 G_GAN_Feat: 0.910 G_ID: 0.136 G_Rec: 0.413 D_GP: 0.058 D_real: 0.891 D_fake: 0.585 +(epoch: 423, iters: 2430, time: 0.063) G_GAN: 0.436 G_GAN_Feat: 0.783 G_ID: 0.108 G_Rec: 0.310 D_GP: 0.067 D_real: 0.959 D_fake: 0.575 +(epoch: 423, iters: 2830, time: 0.063) G_GAN: 0.418 G_GAN_Feat: 0.910 G_ID: 0.105 G_Rec: 0.394 D_GP: 0.036 D_real: 1.052 D_fake: 0.583 +(epoch: 423, iters: 3230, time: 0.063) G_GAN: 0.246 G_GAN_Feat: 0.730 G_ID: 0.083 G_Rec: 0.304 D_GP: 0.038 D_real: 1.088 D_fake: 0.754 +(epoch: 423, iters: 3630, time: 0.064) G_GAN: 0.671 G_GAN_Feat: 0.981 G_ID: 0.108 G_Rec: 0.429 D_GP: 0.126 D_real: 0.951 D_fake: 0.370 +(epoch: 423, iters: 4030, time: 0.063) G_GAN: 0.148 G_GAN_Feat: 0.776 G_ID: 0.108 G_Rec: 0.307 D_GP: 0.054 D_real: 0.771 D_fake: 0.852 +(epoch: 423, iters: 4430, time: 0.063) G_GAN: 0.315 G_GAN_Feat: 1.010 G_ID: 0.113 G_Rec: 0.494 D_GP: 0.043 D_real: 0.852 D_fake: 0.700 +(epoch: 423, iters: 4830, time: 0.063) G_GAN: 0.199 G_GAN_Feat: 0.780 G_ID: 0.086 G_Rec: 0.315 D_GP: 0.067 D_real: 0.911 D_fake: 0.803 +(epoch: 423, iters: 5230, time: 0.064) G_GAN: 0.171 G_GAN_Feat: 0.978 G_ID: 0.123 G_Rec: 0.443 D_GP: 0.046 D_real: 0.670 D_fake: 0.829 +(epoch: 423, iters: 5630, time: 0.064) G_GAN: -0.109 G_GAN_Feat: 0.899 G_ID: 0.092 G_Rec: 0.373 D_GP: 0.155 D_real: 0.205 D_fake: 1.111 +(epoch: 423, iters: 6030, time: 0.063) G_GAN: 0.274 G_GAN_Feat: 0.974 G_ID: 0.135 G_Rec: 0.424 D_GP: 0.040 D_real: 0.683 D_fake: 0.731 +(epoch: 423, iters: 6430, time: 0.063) G_GAN: 0.108 G_GAN_Feat: 0.861 G_ID: 0.101 G_Rec: 0.357 D_GP: 0.094 D_real: 0.569 D_fake: 0.892 +(epoch: 423, iters: 6830, time: 0.064) G_GAN: 0.629 G_GAN_Feat: 0.995 G_ID: 0.116 G_Rec: 0.490 D_GP: 0.052 D_real: 0.995 D_fake: 0.402 +(epoch: 423, iters: 7230, time: 0.063) G_GAN: 0.221 G_GAN_Feat: 0.752 G_ID: 0.097 G_Rec: 0.340 D_GP: 0.034 D_real: 1.074 D_fake: 0.779 +(epoch: 423, iters: 7630, time: 0.063) G_GAN: 0.546 G_GAN_Feat: 0.962 G_ID: 0.114 G_Rec: 0.430 D_GP: 0.043 D_real: 1.177 D_fake: 0.500 +(epoch: 423, iters: 8030, time: 0.063) G_GAN: 0.442 G_GAN_Feat: 0.835 G_ID: 0.091 G_Rec: 0.315 D_GP: 0.054 D_real: 1.045 D_fake: 0.562 +(epoch: 423, iters: 8430, time: 0.064) G_GAN: 0.382 G_GAN_Feat: 0.997 G_ID: 0.104 G_Rec: 0.529 D_GP: 0.036 D_real: 0.774 D_fake: 0.622 +(epoch: 424, iters: 222, time: 0.063) G_GAN: 0.245 G_GAN_Feat: 0.826 G_ID: 0.111 G_Rec: 0.297 D_GP: 0.050 D_real: 0.698 D_fake: 0.761 +(epoch: 424, iters: 622, time: 0.063) G_GAN: 0.446 G_GAN_Feat: 0.869 G_ID: 0.114 G_Rec: 0.415 D_GP: 0.026 D_real: 1.185 D_fake: 0.557 +(epoch: 424, iters: 1022, time: 0.063) G_GAN: 0.259 G_GAN_Feat: 0.692 G_ID: 0.103 G_Rec: 0.305 D_GP: 0.032 D_real: 1.113 D_fake: 0.742 +(epoch: 424, iters: 1422, time: 0.064) G_GAN: 0.317 G_GAN_Feat: 1.008 G_ID: 0.111 G_Rec: 0.450 D_GP: 0.050 D_real: 0.725 D_fake: 0.684 +(epoch: 424, iters: 1822, time: 0.063) G_GAN: 0.479 G_GAN_Feat: 0.781 G_ID: 0.093 G_Rec: 0.301 D_GP: 0.045 D_real: 0.999 D_fake: 0.532 +(epoch: 424, iters: 2222, time: 0.063) G_GAN: 0.962 G_GAN_Feat: 1.097 G_ID: 0.120 G_Rec: 0.462 D_GP: 0.139 D_real: 0.944 D_fake: 0.201 +(epoch: 424, iters: 2622, time: 0.063) G_GAN: 0.223 G_GAN_Feat: 0.715 G_ID: 0.093 G_Rec: 0.304 D_GP: 0.036 D_real: 1.119 D_fake: 0.780 +(epoch: 424, iters: 3022, time: 0.064) G_GAN: 0.512 G_GAN_Feat: 0.880 G_ID: 0.133 G_Rec: 0.382 D_GP: 0.029 D_real: 1.202 D_fake: 0.524 +(epoch: 424, iters: 3422, time: 0.063) G_GAN: 0.479 G_GAN_Feat: 0.907 G_ID: 0.105 G_Rec: 0.342 D_GP: 0.136 D_real: 0.499 D_fake: 0.588 +(epoch: 424, iters: 3822, time: 0.063) G_GAN: 0.869 G_GAN_Feat: 1.011 G_ID: 0.131 G_Rec: 0.454 D_GP: 0.045 D_real: 1.296 D_fake: 0.209 +(epoch: 424, iters: 4222, time: 0.063) G_GAN: 0.123 G_GAN_Feat: 0.700 G_ID: 0.089 G_Rec: 0.309 D_GP: 0.025 D_real: 1.073 D_fake: 0.877 +(epoch: 424, iters: 4622, time: 0.064) G_GAN: 0.241 G_GAN_Feat: 0.960 G_ID: 0.103 G_Rec: 0.437 D_GP: 0.031 D_real: 0.849 D_fake: 0.765 +(epoch: 424, iters: 5022, time: 0.063) G_GAN: -0.190 G_GAN_Feat: 0.747 G_ID: 0.108 G_Rec: 0.297 D_GP: 0.033 D_real: 0.817 D_fake: 1.191 +(epoch: 424, iters: 5422, time: 0.063) G_GAN: 0.808 G_GAN_Feat: 1.071 G_ID: 0.105 G_Rec: 0.485 D_GP: 0.034 D_real: 0.946 D_fake: 0.274 +(epoch: 424, iters: 5822, time: 0.063) G_GAN: 0.190 G_GAN_Feat: 0.788 G_ID: 0.101 G_Rec: 0.313 D_GP: 0.040 D_real: 0.954 D_fake: 0.811 +(epoch: 424, iters: 6222, time: 0.064) G_GAN: 0.243 G_GAN_Feat: 1.124 G_ID: 0.131 G_Rec: 0.444 D_GP: 0.042 D_real: 0.984 D_fake: 0.761 +(epoch: 424, iters: 6622, time: 0.063) G_GAN: -0.164 G_GAN_Feat: 0.752 G_ID: 0.121 G_Rec: 0.325 D_GP: 0.031 D_real: 0.752 D_fake: 1.164 +(epoch: 424, iters: 7022, time: 0.063) G_GAN: 0.260 G_GAN_Feat: 0.974 G_ID: 0.113 G_Rec: 0.457 D_GP: 0.035 D_real: 0.852 D_fake: 0.742 +(epoch: 424, iters: 7422, time: 0.063) G_GAN: 0.350 G_GAN_Feat: 0.750 G_ID: 0.090 G_Rec: 0.313 D_GP: 0.035 D_real: 1.176 D_fake: 0.652 +(epoch: 424, iters: 7822, time: 0.064) G_GAN: 0.216 G_GAN_Feat: 0.935 G_ID: 0.110 G_Rec: 0.419 D_GP: 0.027 D_real: 0.901 D_fake: 0.784 +(epoch: 424, iters: 8222, time: 0.063) G_GAN: -0.044 G_GAN_Feat: 0.764 G_ID: 0.100 G_Rec: 0.322 D_GP: 0.034 D_real: 0.742 D_fake: 1.045 +(epoch: 424, iters: 8622, time: 0.063) G_GAN: 0.254 G_GAN_Feat: 1.003 G_ID: 0.111 G_Rec: 0.425 D_GP: 0.147 D_real: 0.652 D_fake: 0.747 +(epoch: 425, iters: 414, time: 0.063) G_GAN: 0.214 G_GAN_Feat: 0.683 G_ID: 0.123 G_Rec: 0.278 D_GP: 0.031 D_real: 1.165 D_fake: 0.786 +(epoch: 425, iters: 814, time: 0.064) G_GAN: 0.849 G_GAN_Feat: 1.068 G_ID: 0.107 G_Rec: 0.485 D_GP: 0.111 D_real: 1.138 D_fake: 0.230 +(epoch: 425, iters: 1214, time: 0.063) G_GAN: 0.504 G_GAN_Feat: 0.728 G_ID: 0.096 G_Rec: 0.310 D_GP: 0.031 D_real: 1.430 D_fake: 0.504 +(epoch: 425, iters: 1614, time: 0.063) G_GAN: 0.214 G_GAN_Feat: 0.852 G_ID: 0.134 G_Rec: 0.354 D_GP: 0.038 D_real: 0.920 D_fake: 0.786 +(epoch: 425, iters: 2014, time: 0.063) G_GAN: 0.030 G_GAN_Feat: 0.739 G_ID: 0.128 G_Rec: 0.299 D_GP: 0.043 D_real: 0.943 D_fake: 0.970 +(epoch: 425, iters: 2414, time: 0.064) G_GAN: 0.696 G_GAN_Feat: 0.861 G_ID: 0.132 G_Rec: 0.419 D_GP: 0.034 D_real: 1.407 D_fake: 0.330 +(epoch: 425, iters: 2814, time: 0.063) G_GAN: 0.143 G_GAN_Feat: 0.644 G_ID: 0.105 G_Rec: 0.287 D_GP: 0.030 D_real: 1.033 D_fake: 0.857 +(epoch: 425, iters: 3214, time: 0.063) G_GAN: 0.397 G_GAN_Feat: 0.875 G_ID: 0.120 G_Rec: 0.434 D_GP: 0.037 D_real: 1.033 D_fake: 0.615 +(epoch: 425, iters: 3614, time: 0.063) G_GAN: 0.035 G_GAN_Feat: 0.696 G_ID: 0.107 G_Rec: 0.318 D_GP: 0.044 D_real: 0.885 D_fake: 0.965 +(epoch: 425, iters: 4014, time: 0.063) G_GAN: 0.194 G_GAN_Feat: 0.910 G_ID: 0.111 G_Rec: 0.406 D_GP: 0.057 D_real: 0.761 D_fake: 0.807 +(epoch: 425, iters: 4414, time: 0.063) G_GAN: 0.018 G_GAN_Feat: 0.732 G_ID: 0.093 G_Rec: 0.321 D_GP: 0.056 D_real: 0.805 D_fake: 0.982 +(epoch: 425, iters: 4814, time: 0.063) G_GAN: 0.223 G_GAN_Feat: 0.843 G_ID: 0.129 G_Rec: 0.387 D_GP: 0.032 D_real: 0.990 D_fake: 0.777 +(epoch: 425, iters: 5214, time: 0.064) G_GAN: 0.167 G_GAN_Feat: 0.819 G_ID: 0.092 G_Rec: 0.360 D_GP: 0.084 D_real: 0.753 D_fake: 0.834 +(epoch: 425, iters: 5614, time: 0.063) G_GAN: 0.247 G_GAN_Feat: 0.939 G_ID: 0.113 G_Rec: 0.449 D_GP: 0.038 D_real: 0.906 D_fake: 0.754 +(epoch: 425, iters: 6014, time: 0.063) G_GAN: 0.247 G_GAN_Feat: 0.686 G_ID: 0.099 G_Rec: 0.283 D_GP: 0.052 D_real: 1.038 D_fake: 0.753 +(epoch: 425, iters: 6414, time: 0.063) G_GAN: 0.917 G_GAN_Feat: 0.970 G_ID: 0.102 G_Rec: 0.435 D_GP: 0.038 D_real: 1.505 D_fake: 0.173 +(epoch: 425, iters: 6814, time: 0.064) G_GAN: 0.136 G_GAN_Feat: 0.766 G_ID: 0.110 G_Rec: 0.304 D_GP: 0.033 D_real: 0.848 D_fake: 0.865 +(epoch: 425, iters: 7214, time: 0.063) G_GAN: 0.529 G_GAN_Feat: 0.988 G_ID: 0.115 G_Rec: 0.441 D_GP: 0.033 D_real: 1.092 D_fake: 0.474 +(epoch: 425, iters: 7614, time: 0.063) G_GAN: 0.290 G_GAN_Feat: 0.806 G_ID: 0.112 G_Rec: 0.312 D_GP: 0.046 D_real: 0.861 D_fake: 0.731 +(epoch: 425, iters: 8014, time: 0.063) G_GAN: 0.491 G_GAN_Feat: 0.907 G_ID: 0.145 G_Rec: 0.474 D_GP: 0.034 D_real: 1.101 D_fake: 0.515 +(epoch: 425, iters: 8414, time: 0.064) G_GAN: 0.271 G_GAN_Feat: 0.690 G_ID: 0.113 G_Rec: 0.333 D_GP: 0.031 D_real: 1.146 D_fake: 0.731 +(epoch: 426, iters: 206, time: 0.063) G_GAN: 0.746 G_GAN_Feat: 0.903 G_ID: 0.114 G_Rec: 0.446 D_GP: 0.035 D_real: 1.331 D_fake: 0.345 +(epoch: 426, iters: 606, time: 0.063) G_GAN: 0.028 G_GAN_Feat: 0.817 G_ID: 0.095 G_Rec: 0.351 D_GP: 0.061 D_real: 0.648 D_fake: 0.972 +(epoch: 426, iters: 1006, time: 0.063) G_GAN: 0.474 G_GAN_Feat: 0.867 G_ID: 0.122 G_Rec: 0.392 D_GP: 0.036 D_real: 1.146 D_fake: 0.545 +(epoch: 426, iters: 1406, time: 0.064) G_GAN: 0.184 G_GAN_Feat: 0.803 G_ID: 0.104 G_Rec: 0.373 D_GP: 0.042 D_real: 0.956 D_fake: 0.818 +(epoch: 426, iters: 1806, time: 0.063) G_GAN: 0.716 G_GAN_Feat: 1.045 G_ID: 0.106 G_Rec: 0.486 D_GP: 0.033 D_real: 1.416 D_fake: 0.305 +(epoch: 426, iters: 2206, time: 0.063) G_GAN: 0.216 G_GAN_Feat: 0.706 G_ID: 0.088 G_Rec: 0.267 D_GP: 0.039 D_real: 1.048 D_fake: 0.784 +(epoch: 426, iters: 2606, time: 0.063) G_GAN: 0.434 G_GAN_Feat: 0.916 G_ID: 0.100 G_Rec: 0.440 D_GP: 0.027 D_real: 1.137 D_fake: 0.572 +(epoch: 426, iters: 3006, time: 0.064) G_GAN: 0.248 G_GAN_Feat: 0.774 G_ID: 0.090 G_Rec: 0.298 D_GP: 0.032 D_real: 1.020 D_fake: 0.752 +(epoch: 426, iters: 3406, time: 0.063) G_GAN: 0.445 G_GAN_Feat: 0.951 G_ID: 0.137 G_Rec: 0.397 D_GP: 0.039 D_real: 1.055 D_fake: 0.574 +(epoch: 426, iters: 3806, time: 0.063) G_GAN: -0.328 G_GAN_Feat: 1.137 G_ID: 0.119 G_Rec: 0.375 D_GP: 0.220 D_real: 0.927 D_fake: 1.328 +(epoch: 426, iters: 4206, time: 0.063) G_GAN: 0.348 G_GAN_Feat: 0.880 G_ID: 0.126 G_Rec: 0.462 D_GP: 0.024 D_real: 1.076 D_fake: 0.658 +(epoch: 426, iters: 4606, time: 0.064) G_GAN: -0.098 G_GAN_Feat: 0.696 G_ID: 0.092 G_Rec: 0.311 D_GP: 0.037 D_real: 0.810 D_fake: 1.098 +(epoch: 426, iters: 5006, time: 0.063) G_GAN: 0.276 G_GAN_Feat: 0.955 G_ID: 0.129 G_Rec: 0.436 D_GP: 0.063 D_real: 0.751 D_fake: 0.726 +(epoch: 426, iters: 5406, time: 0.063) G_GAN: 0.359 G_GAN_Feat: 0.823 G_ID: 0.103 G_Rec: 0.343 D_GP: 0.044 D_real: 0.845 D_fake: 0.645 +(epoch: 426, iters: 5806, time: 0.063) G_GAN: 0.337 G_GAN_Feat: 0.897 G_ID: 0.114 G_Rec: 0.414 D_GP: 0.035 D_real: 0.999 D_fake: 0.668 +(epoch: 426, iters: 6206, time: 0.064) G_GAN: 0.126 G_GAN_Feat: 0.717 G_ID: 0.092 G_Rec: 0.314 D_GP: 0.033 D_real: 0.986 D_fake: 0.874 +(epoch: 426, iters: 6606, time: 0.063) G_GAN: 0.324 G_GAN_Feat: 1.064 G_ID: 0.125 G_Rec: 0.438 D_GP: 0.063 D_real: 0.718 D_fake: 0.677 +(epoch: 426, iters: 7006, time: 0.063) G_GAN: 0.179 G_GAN_Feat: 0.767 G_ID: 0.090 G_Rec: 0.319 D_GP: 0.060 D_real: 0.827 D_fake: 0.821 +(epoch: 426, iters: 7406, time: 0.063) G_GAN: 0.669 G_GAN_Feat: 1.028 G_ID: 0.111 G_Rec: 0.441 D_GP: 0.049 D_real: 0.932 D_fake: 0.347 +(epoch: 426, iters: 7806, time: 0.064) G_GAN: -0.125 G_GAN_Feat: 0.771 G_ID: 0.090 G_Rec: 0.330 D_GP: 0.068 D_real: 0.626 D_fake: 1.125 +(epoch: 426, iters: 8206, time: 0.063) G_GAN: 0.573 G_GAN_Feat: 1.013 G_ID: 0.106 G_Rec: 0.491 D_GP: 0.030 D_real: 0.967 D_fake: 0.461 +(epoch: 426, iters: 8606, time: 0.063) G_GAN: 0.355 G_GAN_Feat: 0.839 G_ID: 0.108 G_Rec: 0.310 D_GP: 0.067 D_real: 0.961 D_fake: 0.647 +(epoch: 427, iters: 398, time: 0.063) G_GAN: 0.832 G_GAN_Feat: 1.136 G_ID: 0.117 G_Rec: 0.452 D_GP: 0.113 D_real: 0.678 D_fake: 0.328 +(epoch: 427, iters: 798, time: 0.064) G_GAN: -0.039 G_GAN_Feat: 0.724 G_ID: 0.119 G_Rec: 0.315 D_GP: 0.034 D_real: 0.835 D_fake: 1.040 +(epoch: 427, iters: 1198, time: 0.064) G_GAN: 0.630 G_GAN_Feat: 0.965 G_ID: 0.117 G_Rec: 0.454 D_GP: 0.036 D_real: 1.197 D_fake: 0.387 +(epoch: 427, iters: 1598, time: 0.063) G_GAN: -0.024 G_GAN_Feat: 0.826 G_ID: 0.089 G_Rec: 0.344 D_GP: 0.076 D_real: 0.561 D_fake: 1.025 +(epoch: 427, iters: 1998, time: 0.063) G_GAN: 1.528 G_GAN_Feat: 1.058 G_ID: 0.116 G_Rec: 0.505 D_GP: 0.165 D_real: 1.850 D_fake: 0.025 +(epoch: 427, iters: 2398, time: 0.064) G_GAN: 0.119 G_GAN_Feat: 0.718 G_ID: 0.099 G_Rec: 0.309 D_GP: 0.040 D_real: 0.904 D_fake: 0.881 +(epoch: 427, iters: 2798, time: 0.063) G_GAN: 0.768 G_GAN_Feat: 0.973 G_ID: 0.124 G_Rec: 0.433 D_GP: 0.134 D_real: 1.051 D_fake: 0.351 +(epoch: 427, iters: 3198, time: 0.063) G_GAN: 0.173 G_GAN_Feat: 0.837 G_ID: 0.100 G_Rec: 0.338 D_GP: 0.047 D_real: 0.698 D_fake: 0.828 +(epoch: 427, iters: 3598, time: 0.063) G_GAN: 0.965 G_GAN_Feat: 1.248 G_ID: 0.135 G_Rec: 0.495 D_GP: 0.084 D_real: 0.494 D_fake: 0.150 +(epoch: 427, iters: 3998, time: 0.064) G_GAN: 0.425 G_GAN_Feat: 0.954 G_ID: 0.099 G_Rec: 0.346 D_GP: 0.055 D_real: 0.729 D_fake: 0.605 +(epoch: 427, iters: 4398, time: 0.063) G_GAN: 0.846 G_GAN_Feat: 0.996 G_ID: 0.110 G_Rec: 0.381 D_GP: 0.038 D_real: 1.394 D_fake: 0.211 +(epoch: 427, iters: 4798, time: 0.063) G_GAN: 0.285 G_GAN_Feat: 0.765 G_ID: 0.116 G_Rec: 0.339 D_GP: 0.033 D_real: 1.064 D_fake: 0.716 +(epoch: 427, iters: 5198, time: 0.063) G_GAN: 0.209 G_GAN_Feat: 0.924 G_ID: 0.136 G_Rec: 0.463 D_GP: 0.031 D_real: 0.901 D_fake: 0.791 +(epoch: 427, iters: 5598, time: 0.064) G_GAN: 0.103 G_GAN_Feat: 0.859 G_ID: 0.091 G_Rec: 0.344 D_GP: 0.352 D_real: 0.583 D_fake: 0.897 +(epoch: 427, iters: 5998, time: 0.063) G_GAN: 0.187 G_GAN_Feat: 0.964 G_ID: 0.124 G_Rec: 0.421 D_GP: 0.038 D_real: 0.852 D_fake: 0.815 +(epoch: 427, iters: 6398, time: 0.063) G_GAN: 0.252 G_GAN_Feat: 0.693 G_ID: 0.116 G_Rec: 0.295 D_GP: 0.027 D_real: 1.105 D_fake: 0.748 +(epoch: 427, iters: 6798, time: 0.063) G_GAN: 0.672 G_GAN_Feat: 0.931 G_ID: 0.119 G_Rec: 0.434 D_GP: 0.032 D_real: 1.382 D_fake: 0.343 +(epoch: 427, iters: 7198, time: 0.064) G_GAN: 0.078 G_GAN_Feat: 0.805 G_ID: 0.114 G_Rec: 0.356 D_GP: 0.107 D_real: 0.756 D_fake: 0.924 +(epoch: 427, iters: 7598, time: 0.063) G_GAN: 0.397 G_GAN_Feat: 1.204 G_ID: 0.115 G_Rec: 0.467 D_GP: 0.090 D_real: 0.229 D_fake: 0.606 +(epoch: 427, iters: 7998, time: 0.063) G_GAN: 0.124 G_GAN_Feat: 0.740 G_ID: 0.091 G_Rec: 0.306 D_GP: 0.029 D_real: 1.053 D_fake: 0.877 +(epoch: 427, iters: 8398, time: 0.063) G_GAN: 0.544 G_GAN_Feat: 0.955 G_ID: 0.122 G_Rec: 0.417 D_GP: 0.042 D_real: 1.097 D_fake: 0.490 +(epoch: 428, iters: 190, time: 0.064) G_GAN: 0.424 G_GAN_Feat: 0.761 G_ID: 0.097 G_Rec: 0.299 D_GP: 0.052 D_real: 1.226 D_fake: 0.656 +(epoch: 428, iters: 590, time: 0.063) G_GAN: 0.763 G_GAN_Feat: 1.187 G_ID: 0.134 G_Rec: 0.524 D_GP: 0.151 D_real: 0.419 D_fake: 0.306 +(epoch: 428, iters: 990, time: 0.063) G_GAN: 0.535 G_GAN_Feat: 0.816 G_ID: 0.109 G_Rec: 0.315 D_GP: 0.060 D_real: 1.268 D_fake: 0.509 +(epoch: 428, iters: 1390, time: 0.063) G_GAN: 0.498 G_GAN_Feat: 1.129 G_ID: 0.146 G_Rec: 0.450 D_GP: 0.091 D_real: 0.368 D_fake: 0.504 +(epoch: 428, iters: 1790, time: 0.064) G_GAN: 0.865 G_GAN_Feat: 0.956 G_ID: 0.109 G_Rec: 0.338 D_GP: 0.050 D_real: 0.883 D_fake: 0.204 +(epoch: 428, iters: 2190, time: 0.063) G_GAN: 0.955 G_GAN_Feat: 1.040 G_ID: 0.103 G_Rec: 0.433 D_GP: 0.078 D_real: 1.454 D_fake: 0.224 +(epoch: 428, iters: 2590, time: 0.063) G_GAN: 0.107 G_GAN_Feat: 0.649 G_ID: 0.094 G_Rec: 0.263 D_GP: 0.026 D_real: 1.068 D_fake: 0.893 +(epoch: 428, iters: 2990, time: 0.063) G_GAN: 0.514 G_GAN_Feat: 1.101 G_ID: 0.125 G_Rec: 0.488 D_GP: 0.090 D_real: 0.563 D_fake: 0.500 +(epoch: 428, iters: 3390, time: 0.064) G_GAN: 0.442 G_GAN_Feat: 0.897 G_ID: 0.093 G_Rec: 0.300 D_GP: 0.143 D_real: 0.531 D_fake: 0.561 +(epoch: 428, iters: 3790, time: 0.064) G_GAN: 0.662 G_GAN_Feat: 1.053 G_ID: 0.118 G_Rec: 0.433 D_GP: 0.052 D_real: 0.966 D_fake: 0.374 +(epoch: 428, iters: 4190, time: 0.064) G_GAN: 0.328 G_GAN_Feat: 0.817 G_ID: 0.090 G_Rec: 0.318 D_GP: 0.049 D_real: 0.826 D_fake: 0.674 +(epoch: 428, iters: 4590, time: 0.064) G_GAN: 0.169 G_GAN_Feat: 1.190 G_ID: 0.144 G_Rec: 0.439 D_GP: 0.044 D_real: 1.077 D_fake: 0.837 +(epoch: 428, iters: 4990, time: 0.064) G_GAN: 0.240 G_GAN_Feat: 0.610 G_ID: 0.097 G_Rec: 0.297 D_GP: 0.022 D_real: 1.201 D_fake: 0.760 +(epoch: 428, iters: 5390, time: 0.063) G_GAN: 0.533 G_GAN_Feat: 0.913 G_ID: 0.105 G_Rec: 0.429 D_GP: 0.031 D_real: 1.190 D_fake: 0.491 +(epoch: 428, iters: 5790, time: 0.063) G_GAN: 0.191 G_GAN_Feat: 0.693 G_ID: 0.092 G_Rec: 0.304 D_GP: 0.037 D_real: 1.011 D_fake: 0.809 +(epoch: 428, iters: 6190, time: 0.063) G_GAN: 0.236 G_GAN_Feat: 0.887 G_ID: 0.123 G_Rec: 0.450 D_GP: 0.039 D_real: 0.760 D_fake: 0.765 +(epoch: 428, iters: 6590, time: 0.064) G_GAN: 0.035 G_GAN_Feat: 0.763 G_ID: 0.106 G_Rec: 0.327 D_GP: 0.061 D_real: 0.926 D_fake: 0.966 +(epoch: 428, iters: 6990, time: 0.063) G_GAN: 0.642 G_GAN_Feat: 0.908 G_ID: 0.112 G_Rec: 0.438 D_GP: 0.040 D_real: 1.195 D_fake: 0.402 +(epoch: 428, iters: 7390, time: 0.063) G_GAN: 0.058 G_GAN_Feat: 0.812 G_ID: 0.086 G_Rec: 0.346 D_GP: 0.146 D_real: 0.801 D_fake: 0.942 +(epoch: 428, iters: 7790, time: 0.063) G_GAN: 0.409 G_GAN_Feat: 0.963 G_ID: 0.114 G_Rec: 0.406 D_GP: 0.044 D_real: 0.891 D_fake: 0.595 +(epoch: 428, iters: 8190, time: 0.064) G_GAN: 0.340 G_GAN_Feat: 0.786 G_ID: 0.094 G_Rec: 0.320 D_GP: 0.068 D_real: 0.978 D_fake: 0.666 +(epoch: 428, iters: 8590, time: 0.063) G_GAN: 0.313 G_GAN_Feat: 1.035 G_ID: 0.144 G_Rec: 0.417 D_GP: 0.060 D_real: 0.576 D_fake: 0.688 +(epoch: 429, iters: 382, time: 0.063) G_GAN: 0.439 G_GAN_Feat: 0.815 G_ID: 0.099 G_Rec: 0.334 D_GP: 0.061 D_real: 0.999 D_fake: 0.563 +(epoch: 429, iters: 782, time: 0.063) G_GAN: 0.384 G_GAN_Feat: 0.975 G_ID: 0.124 G_Rec: 0.411 D_GP: 0.043 D_real: 0.904 D_fake: 0.618 +(epoch: 429, iters: 1182, time: 0.064) G_GAN: -0.109 G_GAN_Feat: 0.890 G_ID: 0.100 G_Rec: 0.349 D_GP: 0.141 D_real: 0.510 D_fake: 1.109 +(epoch: 429, iters: 1582, time: 0.063) G_GAN: 0.679 G_GAN_Feat: 1.033 G_ID: 0.098 G_Rec: 0.417 D_GP: 0.050 D_real: 1.031 D_fake: 0.338 +(epoch: 429, iters: 1982, time: 0.063) G_GAN: 0.089 G_GAN_Feat: 0.731 G_ID: 0.096 G_Rec: 0.300 D_GP: 0.041 D_real: 0.952 D_fake: 0.911 +(epoch: 429, iters: 2382, time: 0.064) G_GAN: 0.362 G_GAN_Feat: 1.041 G_ID: 0.116 G_Rec: 0.454 D_GP: 0.092 D_real: 0.644 D_fake: 0.640 +(epoch: 429, iters: 2782, time: 0.064) G_GAN: 0.401 G_GAN_Feat: 0.755 G_ID: 0.093 G_Rec: 0.297 D_GP: 0.046 D_real: 1.242 D_fake: 0.691 +(epoch: 429, iters: 3182, time: 0.064) G_GAN: 1.030 G_GAN_Feat: 1.183 G_ID: 0.110 G_Rec: 0.479 D_GP: 0.110 D_real: 0.924 D_fake: 0.184 +(epoch: 429, iters: 3582, time: 0.064) G_GAN: 0.166 G_GAN_Feat: 0.811 G_ID: 0.100 G_Rec: 0.332 D_GP: 0.032 D_real: 1.019 D_fake: 0.834 +(epoch: 429, iters: 3982, time: 0.064) G_GAN: 0.368 G_GAN_Feat: 1.111 G_ID: 0.125 G_Rec: 0.464 D_GP: 0.074 D_real: 0.684 D_fake: 0.644 +(epoch: 429, iters: 4382, time: 0.064) G_GAN: 0.483 G_GAN_Feat: 0.730 G_ID: 0.092 G_Rec: 0.302 D_GP: 0.032 D_real: 1.438 D_fake: 0.520 +(epoch: 429, iters: 4782, time: 0.064) G_GAN: 0.617 G_GAN_Feat: 1.035 G_ID: 0.108 G_Rec: 0.475 D_GP: 0.057 D_real: 1.071 D_fake: 0.395 +(epoch: 429, iters: 5182, time: 0.064) G_GAN: 0.374 G_GAN_Feat: 0.754 G_ID: 0.091 G_Rec: 0.310 D_GP: 0.051 D_real: 0.995 D_fake: 0.629 +(epoch: 429, iters: 5582, time: 0.064) G_GAN: 0.701 G_GAN_Feat: 1.055 G_ID: 0.111 G_Rec: 0.410 D_GP: 0.108 D_real: 0.908 D_fake: 0.418 +(epoch: 429, iters: 5982, time: 0.064) G_GAN: 0.164 G_GAN_Feat: 0.887 G_ID: 0.111 G_Rec: 0.314 D_GP: 0.039 D_real: 0.859 D_fake: 0.836 +(epoch: 429, iters: 6382, time: 0.064) G_GAN: 0.688 G_GAN_Feat: 1.203 G_ID: 0.116 G_Rec: 0.487 D_GP: 0.063 D_real: 0.342 D_fake: 0.336 +(epoch: 429, iters: 6782, time: 0.064) G_GAN: 0.097 G_GAN_Feat: 0.901 G_ID: 0.108 G_Rec: 0.309 D_GP: 0.037 D_real: 0.597 D_fake: 0.904 +(epoch: 429, iters: 7182, time: 0.064) G_GAN: 0.390 G_GAN_Feat: 0.797 G_ID: 0.112 G_Rec: 0.398 D_GP: 0.027 D_real: 1.157 D_fake: 0.611 +(epoch: 429, iters: 7582, time: 0.064) G_GAN: 0.207 G_GAN_Feat: 0.664 G_ID: 0.115 G_Rec: 0.347 D_GP: 0.026 D_real: 1.120 D_fake: 0.796 +(epoch: 429, iters: 7982, time: 0.064) G_GAN: 0.342 G_GAN_Feat: 0.899 G_ID: 0.121 G_Rec: 0.429 D_GP: 0.030 D_real: 1.035 D_fake: 0.663 +(epoch: 429, iters: 8382, time: 0.064) G_GAN: 0.137 G_GAN_Feat: 0.641 G_ID: 0.094 G_Rec: 0.296 D_GP: 0.028 D_real: 1.071 D_fake: 0.863 +(epoch: 430, iters: 174, time: 0.064) G_GAN: 0.308 G_GAN_Feat: 0.855 G_ID: 0.128 G_Rec: 0.408 D_GP: 0.032 D_real: 0.930 D_fake: 0.692 +(epoch: 430, iters: 574, time: 0.064) G_GAN: 0.106 G_GAN_Feat: 0.709 G_ID: 0.120 G_Rec: 0.343 D_GP: 0.035 D_real: 0.857 D_fake: 0.897 +(epoch: 430, iters: 974, time: 0.064) G_GAN: 0.333 G_GAN_Feat: 0.863 G_ID: 0.124 G_Rec: 0.426 D_GP: 0.038 D_real: 0.944 D_fake: 0.673 +(epoch: 430, iters: 1374, time: 0.064) G_GAN: -0.065 G_GAN_Feat: 0.762 G_ID: 0.116 G_Rec: 0.357 D_GP: 0.072 D_real: 0.683 D_fake: 1.065 +(epoch: 430, iters: 1774, time: 0.064) G_GAN: 0.233 G_GAN_Feat: 0.918 G_ID: 0.110 G_Rec: 0.416 D_GP: 0.054 D_real: 0.786 D_fake: 0.769 +(epoch: 430, iters: 2174, time: 0.063) G_GAN: 0.160 G_GAN_Feat: 0.729 G_ID: 0.095 G_Rec: 0.314 D_GP: 0.042 D_real: 0.932 D_fake: 0.841 +(epoch: 430, iters: 2574, time: 0.063) G_GAN: 0.215 G_GAN_Feat: 0.874 G_ID: 0.124 G_Rec: 0.403 D_GP: 0.040 D_real: 0.808 D_fake: 0.785 +(epoch: 430, iters: 2974, time: 0.063) G_GAN: 0.228 G_GAN_Feat: 0.825 G_ID: 0.123 G_Rec: 0.374 D_GP: 0.064 D_real: 0.933 D_fake: 0.780 +(epoch: 430, iters: 3374, time: 0.064) G_GAN: 0.582 G_GAN_Feat: 0.940 G_ID: 0.125 G_Rec: 0.442 D_GP: 0.033 D_real: 1.160 D_fake: 0.425 +(epoch: 430, iters: 3774, time: 0.064) G_GAN: 0.300 G_GAN_Feat: 0.734 G_ID: 0.094 G_Rec: 0.300 D_GP: 0.048 D_real: 1.053 D_fake: 0.701 +(epoch: 430, iters: 4174, time: 0.064) G_GAN: 0.701 G_GAN_Feat: 1.006 G_ID: 0.116 G_Rec: 0.446 D_GP: 0.049 D_real: 1.091 D_fake: 0.328 +(epoch: 430, iters: 4574, time: 0.064) G_GAN: 0.103 G_GAN_Feat: 0.745 G_ID: 0.092 G_Rec: 0.304 D_GP: 0.046 D_real: 0.928 D_fake: 0.897 +(epoch: 430, iters: 4974, time: 0.064) G_GAN: 0.591 G_GAN_Feat: 0.989 G_ID: 0.104 G_Rec: 0.468 D_GP: 0.032 D_real: 1.195 D_fake: 0.427 +(epoch: 430, iters: 5374, time: 0.064) G_GAN: 0.318 G_GAN_Feat: 0.769 G_ID: 0.106 G_Rec: 0.311 D_GP: 0.035 D_real: 1.108 D_fake: 0.683 +(epoch: 430, iters: 5774, time: 0.064) G_GAN: 0.663 G_GAN_Feat: 1.013 G_ID: 0.118 G_Rec: 0.450 D_GP: 0.039 D_real: 1.114 D_fake: 0.349 +(epoch: 430, iters: 6174, time: 0.064) G_GAN: 0.283 G_GAN_Feat: 0.654 G_ID: 0.102 G_Rec: 0.302 D_GP: 0.033 D_real: 1.209 D_fake: 0.718 +(epoch: 430, iters: 6574, time: 0.064) G_GAN: 0.325 G_GAN_Feat: 0.906 G_ID: 0.129 G_Rec: 0.443 D_GP: 0.048 D_real: 0.925 D_fake: 0.676 +(epoch: 430, iters: 6974, time: 0.064) G_GAN: 0.333 G_GAN_Feat: 0.715 G_ID: 0.088 G_Rec: 0.301 D_GP: 0.037 D_real: 1.183 D_fake: 0.669 +(epoch: 430, iters: 7374, time: 0.064) G_GAN: 0.487 G_GAN_Feat: 0.978 G_ID: 0.149 G_Rec: 0.449 D_GP: 0.063 D_real: 0.984 D_fake: 0.525 +(epoch: 430, iters: 7774, time: 0.064) G_GAN: -0.058 G_GAN_Feat: 0.732 G_ID: 0.119 G_Rec: 0.291 D_GP: 0.046 D_real: 0.743 D_fake: 1.058 +(epoch: 430, iters: 8174, time: 0.064) G_GAN: 0.579 G_GAN_Feat: 0.944 G_ID: 0.110 G_Rec: 0.432 D_GP: 0.029 D_real: 1.274 D_fake: 0.427 +(epoch: 430, iters: 8574, time: 0.064) G_GAN: 0.269 G_GAN_Feat: 0.729 G_ID: 0.106 G_Rec: 0.319 D_GP: 0.034 D_real: 1.220 D_fake: 0.737 +(epoch: 431, iters: 366, time: 0.064) G_GAN: 0.463 G_GAN_Feat: 0.983 G_ID: 0.112 G_Rec: 0.459 D_GP: 0.043 D_real: 0.979 D_fake: 0.542 +(epoch: 431, iters: 766, time: 0.064) G_GAN: 0.118 G_GAN_Feat: 0.723 G_ID: 0.106 G_Rec: 0.299 D_GP: 0.044 D_real: 0.905 D_fake: 0.886 +(epoch: 431, iters: 1166, time: 0.064) G_GAN: 0.435 G_GAN_Feat: 0.938 G_ID: 0.115 G_Rec: 0.427 D_GP: 0.042 D_real: 0.951 D_fake: 0.577 +(epoch: 431, iters: 1566, time: 0.064) G_GAN: 0.244 G_GAN_Feat: 0.709 G_ID: 0.093 G_Rec: 0.282 D_GP: 0.035 D_real: 1.155 D_fake: 0.756 +(epoch: 431, iters: 1966, time: 0.063) G_GAN: 0.336 G_GAN_Feat: 1.187 G_ID: 0.115 G_Rec: 0.508 D_GP: 0.118 D_real: 0.372 D_fake: 0.678 +(epoch: 431, iters: 2366, time: 0.063) G_GAN: 0.290 G_GAN_Feat: 0.789 G_ID: 0.094 G_Rec: 0.301 D_GP: 0.040 D_real: 0.850 D_fake: 0.711 +(epoch: 431, iters: 2766, time: 0.064) G_GAN: 0.704 G_GAN_Feat: 1.024 G_ID: 0.122 G_Rec: 0.447 D_GP: 0.035 D_real: 1.315 D_fake: 0.312 +(epoch: 431, iters: 3166, time: 0.063) G_GAN: 0.444 G_GAN_Feat: 1.087 G_ID: 0.112 G_Rec: 0.364 D_GP: 0.141 D_real: 0.416 D_fake: 0.563 +(epoch: 431, iters: 3566, time: 0.063) G_GAN: 0.630 G_GAN_Feat: 0.856 G_ID: 0.113 G_Rec: 0.399 D_GP: 0.037 D_real: 1.315 D_fake: 0.427 +(epoch: 431, iters: 3966, time: 0.063) G_GAN: 0.029 G_GAN_Feat: 0.631 G_ID: 0.100 G_Rec: 0.301 D_GP: 0.026 D_real: 0.988 D_fake: 0.971 +(epoch: 431, iters: 4366, time: 0.064) G_GAN: 0.222 G_GAN_Feat: 0.886 G_ID: 0.120 G_Rec: 0.426 D_GP: 0.030 D_real: 0.829 D_fake: 0.782 +(epoch: 431, iters: 4766, time: 0.063) G_GAN: 0.024 G_GAN_Feat: 0.708 G_ID: 0.079 G_Rec: 0.326 D_GP: 0.031 D_real: 0.934 D_fake: 0.976 +(epoch: 431, iters: 5166, time: 0.063) G_GAN: 0.489 G_GAN_Feat: 1.008 G_ID: 0.112 G_Rec: 0.487 D_GP: 0.056 D_real: 1.053 D_fake: 0.559 +(epoch: 431, iters: 5566, time: 0.063) G_GAN: -0.026 G_GAN_Feat: 0.664 G_ID: 0.104 G_Rec: 0.310 D_GP: 0.029 D_real: 0.949 D_fake: 1.027 +(epoch: 431, iters: 5966, time: 0.064) G_GAN: 0.123 G_GAN_Feat: 1.018 G_ID: 0.118 G_Rec: 0.466 D_GP: 0.049 D_real: 0.548 D_fake: 0.880 +(epoch: 431, iters: 6366, time: 0.063) G_GAN: -0.160 G_GAN_Feat: 0.732 G_ID: 0.109 G_Rec: 0.323 D_GP: 0.065 D_real: 0.672 D_fake: 1.161 +(epoch: 431, iters: 6766, time: 0.063) G_GAN: 0.249 G_GAN_Feat: 1.001 G_ID: 0.110 G_Rec: 0.463 D_GP: 0.043 D_real: 0.666 D_fake: 0.754 +(epoch: 431, iters: 7166, time: 0.063) G_GAN: 0.063 G_GAN_Feat: 0.794 G_ID: 0.115 G_Rec: 0.321 D_GP: 0.039 D_real: 1.049 D_fake: 0.937 +(epoch: 431, iters: 7566, time: 0.064) G_GAN: 0.637 G_GAN_Feat: 0.936 G_ID: 0.123 G_Rec: 0.442 D_GP: 0.045 D_real: 1.095 D_fake: 0.407 +(epoch: 431, iters: 7966, time: 0.063) G_GAN: 0.097 G_GAN_Feat: 0.675 G_ID: 0.098 G_Rec: 0.311 D_GP: 0.044 D_real: 0.978 D_fake: 0.904 +(epoch: 431, iters: 8366, time: 0.063) G_GAN: 0.365 G_GAN_Feat: 0.995 G_ID: 0.111 G_Rec: 0.459 D_GP: 0.049 D_real: 0.908 D_fake: 0.639 +(epoch: 432, iters: 158, time: 0.063) G_GAN: 0.126 G_GAN_Feat: 0.829 G_ID: 0.106 G_Rec: 0.332 D_GP: 0.044 D_real: 1.008 D_fake: 0.877 +(epoch: 432, iters: 558, time: 0.064) G_GAN: 0.360 G_GAN_Feat: 0.953 G_ID: 0.117 G_Rec: 0.435 D_GP: 0.038 D_real: 0.876 D_fake: 0.644 +(epoch: 432, iters: 958, time: 0.063) G_GAN: -0.059 G_GAN_Feat: 0.771 G_ID: 0.095 G_Rec: 0.322 D_GP: 0.048 D_real: 0.695 D_fake: 1.059 +(epoch: 432, iters: 1358, time: 0.063) G_GAN: 0.301 G_GAN_Feat: 0.988 G_ID: 0.153 G_Rec: 0.458 D_GP: 0.036 D_real: 0.833 D_fake: 0.700 +(epoch: 432, iters: 1758, time: 0.063) G_GAN: 0.457 G_GAN_Feat: 0.892 G_ID: 0.093 G_Rec: 0.330 D_GP: 0.086 D_real: 1.296 D_fake: 0.554 +(epoch: 432, iters: 2158, time: 0.064) G_GAN: 0.348 G_GAN_Feat: 1.000 G_ID: 0.124 G_Rec: 0.438 D_GP: 0.049 D_real: 0.816 D_fake: 0.654 +(epoch: 432, iters: 2558, time: 0.063) G_GAN: 0.175 G_GAN_Feat: 0.983 G_ID: 0.107 G_Rec: 0.344 D_GP: 0.423 D_real: 0.265 D_fake: 0.835 +(epoch: 432, iters: 2958, time: 0.063) G_GAN: 0.573 G_GAN_Feat: 1.078 G_ID: 0.125 G_Rec: 0.458 D_GP: 0.052 D_real: 0.807 D_fake: 0.438 +(epoch: 432, iters: 3358, time: 0.063) G_GAN: 0.141 G_GAN_Feat: 0.870 G_ID: 0.094 G_Rec: 0.327 D_GP: 0.166 D_real: 0.335 D_fake: 0.860 +(epoch: 432, iters: 3758, time: 0.064) G_GAN: 0.460 G_GAN_Feat: 0.974 G_ID: 0.104 G_Rec: 0.439 D_GP: 0.037 D_real: 1.158 D_fake: 0.554 +(epoch: 432, iters: 4158, time: 0.063) G_GAN: -0.037 G_GAN_Feat: 0.736 G_ID: 0.095 G_Rec: 0.333 D_GP: 0.064 D_real: 0.693 D_fake: 1.038 +(epoch: 432, iters: 4558, time: 0.063) G_GAN: 0.379 G_GAN_Feat: 1.041 G_ID: 0.116 G_Rec: 0.478 D_GP: 0.043 D_real: 1.084 D_fake: 0.627 +(epoch: 432, iters: 4958, time: 0.063) G_GAN: 0.154 G_GAN_Feat: 0.792 G_ID: 0.107 G_Rec: 0.325 D_GP: 0.034 D_real: 0.869 D_fake: 0.846 +(epoch: 432, iters: 5358, time: 0.064) G_GAN: 0.700 G_GAN_Feat: 1.027 G_ID: 0.132 G_Rec: 0.445 D_GP: 0.059 D_real: 1.139 D_fake: 0.439 +(epoch: 432, iters: 5758, time: 0.063) G_GAN: 0.214 G_GAN_Feat: 0.873 G_ID: 0.122 G_Rec: 0.347 D_GP: 0.045 D_real: 0.901 D_fake: 0.787 +(epoch: 432, iters: 6158, time: 0.063) G_GAN: 0.515 G_GAN_Feat: 1.031 G_ID: 0.115 G_Rec: 0.437 D_GP: 0.055 D_real: 0.803 D_fake: 0.496 +(epoch: 432, iters: 6558, time: 0.063) G_GAN: 0.425 G_GAN_Feat: 0.889 G_ID: 0.116 G_Rec: 0.300 D_GP: 0.048 D_real: 0.663 D_fake: 0.584 +(epoch: 432, iters: 6958, time: 0.064) G_GAN: 0.558 G_GAN_Feat: 1.172 G_ID: 0.117 G_Rec: 0.455 D_GP: 0.051 D_real: 0.372 D_fake: 0.449 +(epoch: 432, iters: 7358, time: 0.063) G_GAN: 0.149 G_GAN_Feat: 0.861 G_ID: 0.125 G_Rec: 0.335 D_GP: 0.043 D_real: 0.925 D_fake: 0.851 +(epoch: 432, iters: 7758, time: 0.063) G_GAN: 0.714 G_GAN_Feat: 1.009 G_ID: 0.105 G_Rec: 0.433 D_GP: 0.038 D_real: 1.143 D_fake: 0.343 +(epoch: 432, iters: 8158, time: 0.063) G_GAN: -0.154 G_GAN_Feat: 0.777 G_ID: 0.099 G_Rec: 0.282 D_GP: 0.030 D_real: 0.689 D_fake: 1.154 +(epoch: 432, iters: 8558, time: 0.064) G_GAN: 0.532 G_GAN_Feat: 1.028 G_ID: 0.119 G_Rec: 0.398 D_GP: 0.038 D_real: 0.971 D_fake: 0.470 +(epoch: 433, iters: 350, time: 0.063) G_GAN: 0.295 G_GAN_Feat: 0.753 G_ID: 0.123 G_Rec: 0.320 D_GP: 0.031 D_real: 1.233 D_fake: 0.706 +(epoch: 433, iters: 750, time: 0.063) G_GAN: 0.369 G_GAN_Feat: 0.935 G_ID: 0.118 G_Rec: 0.399 D_GP: 0.028 D_real: 1.035 D_fake: 0.636 +(epoch: 433, iters: 1150, time: 0.063) G_GAN: 0.124 G_GAN_Feat: 0.759 G_ID: 0.094 G_Rec: 0.314 D_GP: 0.029 D_real: 1.058 D_fake: 0.876 +(epoch: 433, iters: 1550, time: 0.064) G_GAN: 0.279 G_GAN_Feat: 0.930 G_ID: 0.107 G_Rec: 0.413 D_GP: 0.037 D_real: 0.951 D_fake: 0.724 +(epoch: 433, iters: 1950, time: 0.063) G_GAN: 0.062 G_GAN_Feat: 0.788 G_ID: 0.090 G_Rec: 0.295 D_GP: 0.031 D_real: 0.918 D_fake: 0.938 +(epoch: 433, iters: 2350, time: 0.063) G_GAN: 0.850 G_GAN_Feat: 1.063 G_ID: 0.124 G_Rec: 0.412 D_GP: 0.072 D_real: 0.975 D_fake: 0.243 +(epoch: 433, iters: 2750, time: 0.064) G_GAN: 0.142 G_GAN_Feat: 0.904 G_ID: 0.105 G_Rec: 0.324 D_GP: 0.046 D_real: 0.540 D_fake: 0.869 +(epoch: 433, iters: 3150, time: 0.064) G_GAN: 0.814 G_GAN_Feat: 1.386 G_ID: 0.128 G_Rec: 0.517 D_GP: 0.047 D_real: 1.185 D_fake: 0.283 +(epoch: 433, iters: 3550, time: 0.063) G_GAN: 0.189 G_GAN_Feat: 0.754 G_ID: 0.107 G_Rec: 0.307 D_GP: 0.034 D_real: 1.044 D_fake: 0.813 +(epoch: 433, iters: 3950, time: 0.063) G_GAN: 0.921 G_GAN_Feat: 0.891 G_ID: 0.115 G_Rec: 0.446 D_GP: 0.034 D_real: 1.563 D_fake: 0.161 +(epoch: 433, iters: 4350, time: 0.064) G_GAN: 0.150 G_GAN_Feat: 0.617 G_ID: 0.101 G_Rec: 0.273 D_GP: 0.027 D_real: 1.115 D_fake: 0.850 +(epoch: 433, iters: 4750, time: 0.064) G_GAN: 0.262 G_GAN_Feat: 0.897 G_ID: 0.122 G_Rec: 0.492 D_GP: 0.041 D_real: 0.855 D_fake: 0.739 +(epoch: 433, iters: 5150, time: 0.063) G_GAN: 0.089 G_GAN_Feat: 0.628 G_ID: 0.092 G_Rec: 0.296 D_GP: 0.032 D_real: 0.993 D_fake: 0.911 +(epoch: 433, iters: 5550, time: 0.063) G_GAN: 0.424 G_GAN_Feat: 0.849 G_ID: 0.111 G_Rec: 0.404 D_GP: 0.038 D_real: 1.104 D_fake: 0.585 +(epoch: 433, iters: 5950, time: 0.064) G_GAN: -0.160 G_GAN_Feat: 0.635 G_ID: 0.106 G_Rec: 0.272 D_GP: 0.037 D_real: 0.740 D_fake: 1.160 +(epoch: 433, iters: 6350, time: 0.064) G_GAN: 0.479 G_GAN_Feat: 0.934 G_ID: 0.118 G_Rec: 0.450 D_GP: 0.041 D_real: 1.089 D_fake: 0.529 +(epoch: 433, iters: 6750, time: 0.063) G_GAN: 0.060 G_GAN_Feat: 0.655 G_ID: 0.101 G_Rec: 0.331 D_GP: 0.034 D_real: 0.924 D_fake: 0.940 +(epoch: 433, iters: 7150, time: 0.063) G_GAN: 0.561 G_GAN_Feat: 0.967 G_ID: 0.114 G_Rec: 0.440 D_GP: 0.063 D_real: 0.890 D_fake: 0.459 +(epoch: 433, iters: 7550, time: 0.063) G_GAN: 0.252 G_GAN_Feat: 0.770 G_ID: 0.107 G_Rec: 0.313 D_GP: 0.162 D_real: 0.708 D_fake: 0.757 +(epoch: 433, iters: 7950, time: 0.064) G_GAN: 0.229 G_GAN_Feat: 0.975 G_ID: 0.123 G_Rec: 0.416 D_GP: 0.046 D_real: 0.748 D_fake: 0.772 +(epoch: 433, iters: 8350, time: 0.063) G_GAN: 0.149 G_GAN_Feat: 0.794 G_ID: 0.102 G_Rec: 0.333 D_GP: 0.044 D_real: 1.012 D_fake: 0.854 +(epoch: 434, iters: 142, time: 0.063) G_GAN: 0.396 G_GAN_Feat: 0.971 G_ID: 0.131 G_Rec: 0.436 D_GP: 0.030 D_real: 1.037 D_fake: 0.610 +(epoch: 434, iters: 542, time: 0.063) G_GAN: 0.264 G_GAN_Feat: 0.782 G_ID: 0.099 G_Rec: 0.309 D_GP: 0.040 D_real: 1.000 D_fake: 0.737 +(epoch: 434, iters: 942, time: 0.064) G_GAN: 0.856 G_GAN_Feat: 0.977 G_ID: 0.106 G_Rec: 0.401 D_GP: 0.050 D_real: 1.326 D_fake: 0.184 +(epoch: 434, iters: 1342, time: 0.063) G_GAN: 0.493 G_GAN_Feat: 0.798 G_ID: 0.085 G_Rec: 0.269 D_GP: 0.039 D_real: 0.919 D_fake: 0.513 +(epoch: 434, iters: 1742, time: 0.063) G_GAN: 0.482 G_GAN_Feat: 1.127 G_ID: 0.124 G_Rec: 0.443 D_GP: 0.042 D_real: 0.246 D_fake: 0.526 +(epoch: 434, iters: 2142, time: 0.063) G_GAN: 0.546 G_GAN_Feat: 0.823 G_ID: 0.110 G_Rec: 0.315 D_GP: 0.043 D_real: 1.387 D_fake: 0.493 +(epoch: 434, iters: 2542, time: 0.064) G_GAN: 0.710 G_GAN_Feat: 1.067 G_ID: 0.120 G_Rec: 0.448 D_GP: 0.048 D_real: 1.075 D_fake: 0.317 +(epoch: 434, iters: 2942, time: 0.064) G_GAN: 0.373 G_GAN_Feat: 0.797 G_ID: 0.107 G_Rec: 0.316 D_GP: 0.031 D_real: 1.072 D_fake: 0.632 +(epoch: 434, iters: 3342, time: 0.063) G_GAN: 0.560 G_GAN_Feat: 0.952 G_ID: 0.125 G_Rec: 0.418 D_GP: 0.039 D_real: 1.220 D_fake: 0.490 +(epoch: 434, iters: 3742, time: 0.063) G_GAN: -0.041 G_GAN_Feat: 0.772 G_ID: 0.093 G_Rec: 0.311 D_GP: 0.038 D_real: 0.881 D_fake: 1.041 +(epoch: 434, iters: 4142, time: 0.064) G_GAN: 0.706 G_GAN_Feat: 1.127 G_ID: 0.104 G_Rec: 0.459 D_GP: 0.031 D_real: 1.344 D_fake: 0.347 +(epoch: 434, iters: 4542, time: 0.063) G_GAN: 0.023 G_GAN_Feat: 0.774 G_ID: 0.110 G_Rec: 0.324 D_GP: 0.031 D_real: 0.925 D_fake: 0.978 +(epoch: 434, iters: 4942, time: 0.063) G_GAN: 0.318 G_GAN_Feat: 0.935 G_ID: 0.122 G_Rec: 0.413 D_GP: 0.038 D_real: 0.914 D_fake: 0.684 +(epoch: 434, iters: 5342, time: 0.063) G_GAN: 0.271 G_GAN_Feat: 0.786 G_ID: 0.108 G_Rec: 0.321 D_GP: 0.036 D_real: 1.025 D_fake: 0.729 +(epoch: 434, iters: 5742, time: 0.063) G_GAN: 0.715 G_GAN_Feat: 1.093 G_ID: 0.117 G_Rec: 0.467 D_GP: 0.044 D_real: 1.114 D_fake: 0.339 +(epoch: 434, iters: 6142, time: 0.063) G_GAN: 0.120 G_GAN_Feat: 0.740 G_ID: 0.087 G_Rec: 0.313 D_GP: 0.036 D_real: 0.924 D_fake: 0.880 +(epoch: 434, iters: 6542, time: 0.063) G_GAN: 0.589 G_GAN_Feat: 1.213 G_ID: 0.126 G_Rec: 0.507 D_GP: 0.281 D_real: 0.396 D_fake: 0.436 +(epoch: 434, iters: 6942, time: 0.064) G_GAN: 0.334 G_GAN_Feat: 0.827 G_ID: 0.107 G_Rec: 0.321 D_GP: 0.040 D_real: 1.162 D_fake: 0.669 +(epoch: 434, iters: 7342, time: 0.063) G_GAN: 0.857 G_GAN_Feat: 1.037 G_ID: 0.115 G_Rec: 0.425 D_GP: 0.054 D_real: 1.160 D_fake: 0.197 +(epoch: 434, iters: 7742, time: 0.063) G_GAN: 0.656 G_GAN_Feat: 0.959 G_ID: 0.094 G_Rec: 0.338 D_GP: 0.093 D_real: 0.511 D_fake: 0.558 +(epoch: 434, iters: 8142, time: 0.063) G_GAN: 0.565 G_GAN_Feat: 1.003 G_ID: 0.114 G_Rec: 0.440 D_GP: 0.034 D_real: 1.090 D_fake: 0.462 +(epoch: 434, iters: 8542, time: 0.064) G_GAN: 0.223 G_GAN_Feat: 0.862 G_ID: 0.108 G_Rec: 0.300 D_GP: 0.042 D_real: 0.603 D_fake: 0.778 +(epoch: 435, iters: 334, time: 0.063) G_GAN: 0.535 G_GAN_Feat: 1.015 G_ID: 0.114 G_Rec: 0.425 D_GP: 0.069 D_real: 0.719 D_fake: 0.485 +(epoch: 435, iters: 734, time: 0.063) G_GAN: 0.480 G_GAN_Feat: 0.957 G_ID: 0.108 G_Rec: 0.331 D_GP: 0.037 D_real: 0.981 D_fake: 0.527 +(epoch: 435, iters: 1134, time: 0.063) G_GAN: 0.606 G_GAN_Feat: 0.961 G_ID: 0.127 G_Rec: 0.416 D_GP: 0.040 D_real: 1.243 D_fake: 0.427 +(epoch: 435, iters: 1534, time: 0.064) G_GAN: 0.026 G_GAN_Feat: 0.761 G_ID: 0.112 G_Rec: 0.349 D_GP: 0.035 D_real: 0.878 D_fake: 0.974 +(epoch: 435, iters: 1934, time: 0.063) G_GAN: 0.188 G_GAN_Feat: 0.964 G_ID: 0.111 G_Rec: 0.445 D_GP: 0.040 D_real: 0.841 D_fake: 0.812 +(epoch: 435, iters: 2334, time: 0.063) G_GAN: 0.173 G_GAN_Feat: 0.755 G_ID: 0.092 G_Rec: 0.320 D_GP: 0.034 D_real: 0.984 D_fake: 0.827 +(epoch: 435, iters: 2734, time: 0.063) G_GAN: 0.573 G_GAN_Feat: 0.968 G_ID: 0.110 G_Rec: 0.505 D_GP: 0.033 D_real: 1.073 D_fake: 0.457 +(epoch: 435, iters: 3134, time: 0.064) G_GAN: -0.100 G_GAN_Feat: 0.697 G_ID: 0.108 G_Rec: 0.310 D_GP: 0.027 D_real: 0.868 D_fake: 1.100 +(epoch: 435, iters: 3534, time: 0.063) G_GAN: 0.448 G_GAN_Feat: 1.001 G_ID: 0.116 G_Rec: 0.509 D_GP: 0.035 D_real: 1.011 D_fake: 0.554 +(epoch: 435, iters: 3934, time: 0.063) G_GAN: 0.410 G_GAN_Feat: 0.896 G_ID: 0.092 G_Rec: 0.351 D_GP: 0.197 D_real: 0.739 D_fake: 0.592 +(epoch: 435, iters: 4334, time: 0.063) G_GAN: 0.658 G_GAN_Feat: 0.910 G_ID: 0.107 G_Rec: 0.385 D_GP: 0.043 D_real: 1.278 D_fake: 0.360 +(epoch: 435, iters: 4734, time: 0.064) G_GAN: 0.449 G_GAN_Feat: 0.826 G_ID: 0.096 G_Rec: 0.333 D_GP: 0.039 D_real: 1.444 D_fake: 0.560 +(epoch: 435, iters: 5134, time: 0.063) G_GAN: 0.548 G_GAN_Feat: 1.077 G_ID: 0.106 G_Rec: 0.449 D_GP: 0.040 D_real: 0.765 D_fake: 0.465 +(epoch: 435, iters: 5534, time: 0.063) G_GAN: 0.241 G_GAN_Feat: 0.723 G_ID: 0.083 G_Rec: 0.302 D_GP: 0.034 D_real: 1.140 D_fake: 0.760 +(epoch: 435, iters: 5934, time: 0.063) G_GAN: 0.504 G_GAN_Feat: 0.899 G_ID: 0.121 G_Rec: 0.405 D_GP: 0.037 D_real: 1.064 D_fake: 0.512 +(epoch: 435, iters: 6334, time: 0.064) G_GAN: 0.043 G_GAN_Feat: 0.746 G_ID: 0.097 G_Rec: 0.317 D_GP: 0.040 D_real: 0.832 D_fake: 0.957 +(epoch: 435, iters: 6734, time: 0.063) G_GAN: 0.581 G_GAN_Feat: 1.014 G_ID: 0.115 G_Rec: 0.442 D_GP: 0.124 D_real: 0.903 D_fake: 0.460 +(epoch: 435, iters: 7134, time: 0.063) G_GAN: 0.436 G_GAN_Feat: 0.849 G_ID: 0.097 G_Rec: 0.339 D_GP: 0.042 D_real: 0.996 D_fake: 0.576 +(epoch: 435, iters: 7534, time: 0.063) G_GAN: 0.320 G_GAN_Feat: 0.941 G_ID: 0.124 G_Rec: 0.427 D_GP: 0.036 D_real: 0.958 D_fake: 0.680 +(epoch: 435, iters: 7934, time: 0.064) G_GAN: 0.248 G_GAN_Feat: 0.825 G_ID: 0.101 G_Rec: 0.303 D_GP: 0.066 D_real: 0.634 D_fake: 0.753 +(epoch: 435, iters: 8334, time: 0.063) G_GAN: 0.159 G_GAN_Feat: 0.903 G_ID: 0.135 G_Rec: 0.507 D_GP: 0.032 D_real: 0.994 D_fake: 0.842 +(epoch: 436, iters: 126, time: 0.063) G_GAN: 0.085 G_GAN_Feat: 0.645 G_ID: 0.096 G_Rec: 0.307 D_GP: 0.025 D_real: 1.012 D_fake: 0.917 +(epoch: 436, iters: 526, time: 0.063) G_GAN: 0.388 G_GAN_Feat: 0.830 G_ID: 0.118 G_Rec: 0.423 D_GP: 0.031 D_real: 1.165 D_fake: 0.619 +(epoch: 436, iters: 926, time: 0.064) G_GAN: 0.121 G_GAN_Feat: 0.718 G_ID: 0.100 G_Rec: 0.320 D_GP: 0.039 D_real: 0.881 D_fake: 0.879 +(epoch: 436, iters: 1326, time: 0.063) G_GAN: 0.225 G_GAN_Feat: 0.986 G_ID: 0.120 G_Rec: 0.454 D_GP: 0.073 D_real: 0.667 D_fake: 0.784 +(epoch: 436, iters: 1726, time: 0.063) G_GAN: -0.241 G_GAN_Feat: 0.707 G_ID: 0.101 G_Rec: 0.307 D_GP: 0.038 D_real: 0.706 D_fake: 1.241 +(epoch: 436, iters: 2126, time: 0.063) G_GAN: 0.487 G_GAN_Feat: 0.929 G_ID: 0.106 G_Rec: 0.422 D_GP: 0.049 D_real: 1.062 D_fake: 0.535 +(epoch: 436, iters: 2526, time: 0.064) G_GAN: 0.220 G_GAN_Feat: 0.647 G_ID: 0.090 G_Rec: 0.289 D_GP: 0.037 D_real: 1.150 D_fake: 0.781 +(epoch: 436, iters: 2926, time: 0.063) G_GAN: 0.249 G_GAN_Feat: 0.926 G_ID: 0.118 G_Rec: 0.453 D_GP: 0.033 D_real: 0.925 D_fake: 0.755 +(epoch: 436, iters: 3326, time: 0.063) G_GAN: -0.070 G_GAN_Feat: 0.826 G_ID: 0.111 G_Rec: 0.355 D_GP: 0.110 D_real: 0.385 D_fake: 1.070 +(epoch: 436, iters: 3726, time: 0.063) G_GAN: 0.460 G_GAN_Feat: 0.925 G_ID: 0.111 G_Rec: 0.443 D_GP: 0.034 D_real: 1.107 D_fake: 0.545 +(epoch: 436, iters: 4126, time: 0.064) G_GAN: 0.138 G_GAN_Feat: 0.690 G_ID: 0.091 G_Rec: 0.322 D_GP: 0.029 D_real: 1.058 D_fake: 0.862 +(epoch: 436, iters: 4526, time: 0.063) G_GAN: 0.352 G_GAN_Feat: 0.974 G_ID: 0.117 G_Rec: 0.476 D_GP: 0.038 D_real: 1.013 D_fake: 0.648 +(epoch: 436, iters: 4926, time: 0.063) G_GAN: 0.307 G_GAN_Feat: 0.744 G_ID: 0.102 G_Rec: 0.378 D_GP: 0.039 D_real: 1.125 D_fake: 0.696 +(epoch: 436, iters: 5326, time: 0.063) G_GAN: 0.712 G_GAN_Feat: 0.929 G_ID: 0.102 G_Rec: 0.428 D_GP: 0.038 D_real: 1.351 D_fake: 0.316 +(epoch: 436, iters: 5726, time: 0.064) G_GAN: 0.411 G_GAN_Feat: 0.788 G_ID: 0.115 G_Rec: 0.310 D_GP: 0.061 D_real: 1.031 D_fake: 0.594 +(epoch: 436, iters: 6126, time: 0.063) G_GAN: 0.570 G_GAN_Feat: 0.983 G_ID: 0.113 G_Rec: 0.421 D_GP: 0.043 D_real: 1.022 D_fake: 0.441 +(epoch: 436, iters: 6526, time: 0.063) G_GAN: 0.299 G_GAN_Feat: 0.709 G_ID: 0.089 G_Rec: 0.279 D_GP: 0.035 D_real: 1.136 D_fake: 0.701 +(epoch: 436, iters: 6926, time: 0.063) G_GAN: 0.890 G_GAN_Feat: 0.996 G_ID: 0.126 G_Rec: 0.410 D_GP: 0.061 D_real: 1.257 D_fake: 0.184 +(epoch: 436, iters: 7326, time: 0.064) G_GAN: 0.019 G_GAN_Feat: 0.862 G_ID: 0.093 G_Rec: 0.343 D_GP: 0.038 D_real: 0.656 D_fake: 0.981 +(epoch: 436, iters: 7726, time: 0.063) G_GAN: 0.533 G_GAN_Feat: 1.205 G_ID: 0.117 G_Rec: 0.537 D_GP: 0.438 D_real: 0.445 D_fake: 0.482 +(epoch: 436, iters: 8126, time: 0.063) G_GAN: 0.456 G_GAN_Feat: 0.919 G_ID: 0.092 G_Rec: 0.369 D_GP: 0.126 D_real: 0.656 D_fake: 0.555 +(epoch: 436, iters: 8526, time: 0.063) G_GAN: 1.024 G_GAN_Feat: 0.951 G_ID: 0.114 G_Rec: 0.440 D_GP: 0.059 D_real: 1.508 D_fake: 0.107 +(epoch: 437, iters: 318, time: 0.064) G_GAN: 0.227 G_GAN_Feat: 0.960 G_ID: 0.097 G_Rec: 0.347 D_GP: 0.331 D_real: 0.378 D_fake: 0.775 +(epoch: 437, iters: 718, time: 0.063) G_GAN: 0.675 G_GAN_Feat: 1.061 G_ID: 0.117 G_Rec: 0.402 D_GP: 0.037 D_real: 0.872 D_fake: 0.335 +(epoch: 437, iters: 1118, time: 0.063) G_GAN: 0.124 G_GAN_Feat: 0.918 G_ID: 0.111 G_Rec: 0.334 D_GP: 0.058 D_real: 0.281 D_fake: 0.879 +(epoch: 437, iters: 1518, time: 0.063) G_GAN: 0.913 G_GAN_Feat: 1.362 G_ID: 0.117 G_Rec: 0.511 D_GP: 0.077 D_real: 1.543 D_fake: 0.255 +(epoch: 437, iters: 1918, time: 0.064) G_GAN: -0.207 G_GAN_Feat: 0.609 G_ID: 0.095 G_Rec: 0.312 D_GP: 0.026 D_real: 0.823 D_fake: 1.207 +(epoch: 437, iters: 2318, time: 0.063) G_GAN: 0.515 G_GAN_Feat: 0.893 G_ID: 0.111 G_Rec: 0.451 D_GP: 0.026 D_real: 1.140 D_fake: 0.526 +(epoch: 437, iters: 2718, time: 0.063) G_GAN: 0.043 G_GAN_Feat: 0.664 G_ID: 0.102 G_Rec: 0.328 D_GP: 0.033 D_real: 0.954 D_fake: 0.957 +(epoch: 437, iters: 3118, time: 0.063) G_GAN: 0.078 G_GAN_Feat: 0.890 G_ID: 0.106 G_Rec: 0.439 D_GP: 0.048 D_real: 0.757 D_fake: 0.923 +(epoch: 437, iters: 3518, time: 0.064) G_GAN: 0.011 G_GAN_Feat: 0.619 G_ID: 0.092 G_Rec: 0.276 D_GP: 0.031 D_real: 0.989 D_fake: 0.990 +(epoch: 437, iters: 3918, time: 0.063) G_GAN: 0.320 G_GAN_Feat: 0.874 G_ID: 0.113 G_Rec: 0.418 D_GP: 0.035 D_real: 1.012 D_fake: 0.691 +(epoch: 437, iters: 4318, time: 0.063) G_GAN: 0.185 G_GAN_Feat: 0.697 G_ID: 0.093 G_Rec: 0.302 D_GP: 0.033 D_real: 1.074 D_fake: 0.817 +(epoch: 437, iters: 4718, time: 0.063) G_GAN: 0.370 G_GAN_Feat: 0.882 G_ID: 0.124 G_Rec: 0.376 D_GP: 0.041 D_real: 1.007 D_fake: 0.635 +(epoch: 437, iters: 5118, time: 0.064) G_GAN: 0.149 G_GAN_Feat: 0.805 G_ID: 0.118 G_Rec: 0.334 D_GP: 0.037 D_real: 0.942 D_fake: 0.851 +(epoch: 437, iters: 5518, time: 0.063) G_GAN: 0.488 G_GAN_Feat: 0.849 G_ID: 0.118 G_Rec: 0.406 D_GP: 0.030 D_real: 1.282 D_fake: 0.515 +(epoch: 437, iters: 5918, time: 0.063) G_GAN: 0.309 G_GAN_Feat: 0.647 G_ID: 0.087 G_Rec: 0.298 D_GP: 0.029 D_real: 1.234 D_fake: 0.692 +(epoch: 437, iters: 6318, time: 0.063) G_GAN: 0.384 G_GAN_Feat: 0.879 G_ID: 0.117 G_Rec: 0.405 D_GP: 0.040 D_real: 1.017 D_fake: 0.622 +(epoch: 437, iters: 6718, time: 0.064) G_GAN: 0.104 G_GAN_Feat: 0.620 G_ID: 0.093 G_Rec: 0.264 D_GP: 0.032 D_real: 1.029 D_fake: 0.896 +(epoch: 437, iters: 7118, time: 0.063) G_GAN: 0.067 G_GAN_Feat: 0.987 G_ID: 0.116 G_Rec: 0.438 D_GP: 0.083 D_real: 0.571 D_fake: 0.934 +(epoch: 437, iters: 7518, time: 0.063) G_GAN: 0.183 G_GAN_Feat: 0.741 G_ID: 0.115 G_Rec: 0.314 D_GP: 0.045 D_real: 0.955 D_fake: 0.818 +(epoch: 437, iters: 7918, time: 0.063) G_GAN: 0.639 G_GAN_Feat: 0.950 G_ID: 0.125 G_Rec: 0.445 D_GP: 0.032 D_real: 1.199 D_fake: 0.394 +(epoch: 437, iters: 8318, time: 0.064) G_GAN: 0.213 G_GAN_Feat: 0.797 G_ID: 0.144 G_Rec: 0.314 D_GP: 0.037 D_real: 0.850 D_fake: 0.791 +(epoch: 438, iters: 110, time: 0.063) G_GAN: 0.306 G_GAN_Feat: 0.929 G_ID: 0.119 G_Rec: 0.420 D_GP: 0.037 D_real: 1.014 D_fake: 0.694 +(epoch: 438, iters: 510, time: 0.063) G_GAN: 0.132 G_GAN_Feat: 0.660 G_ID: 0.097 G_Rec: 0.287 D_GP: 0.029 D_real: 1.031 D_fake: 0.868 +(epoch: 438, iters: 910, time: 0.063) G_GAN: 0.457 G_GAN_Feat: 0.853 G_ID: 0.135 G_Rec: 0.394 D_GP: 0.042 D_real: 1.140 D_fake: 0.557 +(epoch: 438, iters: 1310, time: 0.064) G_GAN: 0.244 G_GAN_Feat: 0.778 G_ID: 0.083 G_Rec: 0.332 D_GP: 0.059 D_real: 1.072 D_fake: 0.756 +(epoch: 438, iters: 1710, time: 0.063) G_GAN: 0.200 G_GAN_Feat: 1.002 G_ID: 0.100 G_Rec: 0.424 D_GP: 0.099 D_real: 0.523 D_fake: 0.803 +(epoch: 438, iters: 2110, time: 0.063) G_GAN: 0.284 G_GAN_Feat: 0.764 G_ID: 0.101 G_Rec: 0.316 D_GP: 0.031 D_real: 1.075 D_fake: 0.716 +(epoch: 438, iters: 2510, time: 0.063) G_GAN: 0.508 G_GAN_Feat: 0.976 G_ID: 0.112 G_Rec: 0.415 D_GP: 0.034 D_real: 1.106 D_fake: 0.497 +(epoch: 438, iters: 2910, time: 0.064) G_GAN: 0.206 G_GAN_Feat: 0.846 G_ID: 0.102 G_Rec: 0.334 D_GP: 0.044 D_real: 0.776 D_fake: 0.795 +(epoch: 438, iters: 3310, time: 0.063) G_GAN: 0.524 G_GAN_Feat: 1.100 G_ID: 0.107 G_Rec: 0.420 D_GP: 0.127 D_real: 0.373 D_fake: 0.487 +(epoch: 438, iters: 3710, time: 0.063) G_GAN: 0.332 G_GAN_Feat: 0.827 G_ID: 0.094 G_Rec: 0.301 D_GP: 0.032 D_real: 0.844 D_fake: 0.676 +(epoch: 438, iters: 4110, time: 0.063) G_GAN: 0.495 G_GAN_Feat: 1.074 G_ID: 0.107 G_Rec: 0.399 D_GP: 0.061 D_real: 0.577 D_fake: 0.516 +(epoch: 438, iters: 4510, time: 0.064) G_GAN: 0.100 G_GAN_Feat: 0.746 G_ID: 0.089 G_Rec: 0.324 D_GP: 0.031 D_real: 0.968 D_fake: 0.900 +(epoch: 438, iters: 4910, time: 0.063) G_GAN: 0.494 G_GAN_Feat: 1.052 G_ID: 0.115 G_Rec: 0.486 D_GP: 0.044 D_real: 0.939 D_fake: 0.546 +(epoch: 438, iters: 5310, time: 0.063) G_GAN: 0.140 G_GAN_Feat: 0.855 G_ID: 0.098 G_Rec: 0.343 D_GP: 0.163 D_real: 0.651 D_fake: 0.860 +(epoch: 438, iters: 5710, time: 0.063) G_GAN: 0.595 G_GAN_Feat: 1.098 G_ID: 0.126 G_Rec: 0.490 D_GP: 0.085 D_real: 0.756 D_fake: 0.423 +(epoch: 438, iters: 6110, time: 0.064) G_GAN: 0.069 G_GAN_Feat: 0.933 G_ID: 0.098 G_Rec: 0.338 D_GP: 0.191 D_real: 0.423 D_fake: 0.932 +(epoch: 438, iters: 6510, time: 0.063) G_GAN: 0.268 G_GAN_Feat: 0.861 G_ID: 0.107 G_Rec: 0.419 D_GP: 0.031 D_real: 1.021 D_fake: 0.733 +(epoch: 438, iters: 6910, time: 0.063) G_GAN: -0.022 G_GAN_Feat: 0.658 G_ID: 0.110 G_Rec: 0.300 D_GP: 0.029 D_real: 0.877 D_fake: 1.022 +(epoch: 438, iters: 7310, time: 0.063) G_GAN: 0.106 G_GAN_Feat: 0.856 G_ID: 0.124 G_Rec: 0.429 D_GP: 0.035 D_real: 0.925 D_fake: 0.897 +(epoch: 438, iters: 7710, time: 0.064) G_GAN: 0.052 G_GAN_Feat: 0.709 G_ID: 0.098 G_Rec: 0.359 D_GP: 0.053 D_real: 0.930 D_fake: 0.950 +(epoch: 438, iters: 8110, time: 0.063) G_GAN: 0.183 G_GAN_Feat: 0.920 G_ID: 0.130 G_Rec: 0.474 D_GP: 0.043 D_real: 0.728 D_fake: 0.818 +(epoch: 438, iters: 8510, time: 0.063) G_GAN: -0.089 G_GAN_Feat: 0.627 G_ID: 0.120 G_Rec: 0.309 D_GP: 0.030 D_real: 0.789 D_fake: 1.089 +(epoch: 439, iters: 302, time: 0.063) G_GAN: 0.365 G_GAN_Feat: 0.870 G_ID: 0.118 G_Rec: 0.423 D_GP: 0.042 D_real: 1.014 D_fake: 0.648 +(epoch: 439, iters: 702, time: 0.064) G_GAN: -0.253 G_GAN_Feat: 0.731 G_ID: 0.096 G_Rec: 0.336 D_GP: 0.046 D_real: 0.570 D_fake: 1.253 +(epoch: 439, iters: 1102, time: 0.063) G_GAN: 0.374 G_GAN_Feat: 0.919 G_ID: 0.132 G_Rec: 0.454 D_GP: 0.052 D_real: 0.924 D_fake: 0.640 +(epoch: 439, iters: 1502, time: 0.063) G_GAN: 0.011 G_GAN_Feat: 0.680 G_ID: 0.083 G_Rec: 0.308 D_GP: 0.048 D_real: 0.865 D_fake: 0.989 +(epoch: 439, iters: 1902, time: 0.063) G_GAN: 0.419 G_GAN_Feat: 0.873 G_ID: 0.126 G_Rec: 0.421 D_GP: 0.032 D_real: 0.981 D_fake: 0.593 +(epoch: 439, iters: 2302, time: 0.063) G_GAN: 0.080 G_GAN_Feat: 0.687 G_ID: 0.114 G_Rec: 0.298 D_GP: 0.038 D_real: 0.903 D_fake: 0.920 +(epoch: 439, iters: 2702, time: 0.063) G_GAN: 0.376 G_GAN_Feat: 0.948 G_ID: 0.137 G_Rec: 0.437 D_GP: 0.050 D_real: 0.969 D_fake: 0.633 +(epoch: 439, iters: 3102, time: 0.063) G_GAN: 0.314 G_GAN_Feat: 0.647 G_ID: 0.089 G_Rec: 0.275 D_GP: 0.035 D_real: 1.233 D_fake: 0.687 +(epoch: 439, iters: 3502, time: 0.064) G_GAN: 0.347 G_GAN_Feat: 0.959 G_ID: 0.125 G_Rec: 0.449 D_GP: 0.059 D_real: 0.833 D_fake: 0.661 +(epoch: 439, iters: 3902, time: 0.063) G_GAN: 0.370 G_GAN_Feat: 0.737 G_ID: 0.090 G_Rec: 0.301 D_GP: 0.063 D_real: 0.994 D_fake: 0.640 +(epoch: 439, iters: 4302, time: 0.063) G_GAN: 0.400 G_GAN_Feat: 0.911 G_ID: 0.123 G_Rec: 0.410 D_GP: 0.037 D_real: 1.001 D_fake: 0.605 +(epoch: 439, iters: 4702, time: 0.063) G_GAN: 0.074 G_GAN_Feat: 0.691 G_ID: 0.094 G_Rec: 0.314 D_GP: 0.025 D_real: 0.999 D_fake: 0.926 +(epoch: 439, iters: 5102, time: 0.064) G_GAN: 0.344 G_GAN_Feat: 0.976 G_ID: 0.108 G_Rec: 0.446 D_GP: 0.052 D_real: 0.905 D_fake: 0.675 +(epoch: 439, iters: 5502, time: 0.063) G_GAN: 0.121 G_GAN_Feat: 0.667 G_ID: 0.099 G_Rec: 0.281 D_GP: 0.036 D_real: 1.013 D_fake: 0.879 +(epoch: 439, iters: 5902, time: 0.063) G_GAN: 0.575 G_GAN_Feat: 0.925 G_ID: 0.113 G_Rec: 0.402 D_GP: 0.061 D_real: 1.159 D_fake: 0.465 +(epoch: 439, iters: 6302, time: 0.063) G_GAN: 0.133 G_GAN_Feat: 0.766 G_ID: 0.094 G_Rec: 0.297 D_GP: 0.044 D_real: 0.897 D_fake: 0.867 +(epoch: 439, iters: 6702, time: 0.064) G_GAN: 0.602 G_GAN_Feat: 1.025 G_ID: 0.134 G_Rec: 0.469 D_GP: 0.038 D_real: 1.087 D_fake: 0.422 +(epoch: 439, iters: 7102, time: 0.063) G_GAN: -0.103 G_GAN_Feat: 0.795 G_ID: 0.097 G_Rec: 0.321 D_GP: 0.049 D_real: 0.567 D_fake: 1.103 +(epoch: 439, iters: 7502, time: 0.063) G_GAN: 0.327 G_GAN_Feat: 1.029 G_ID: 0.111 G_Rec: 0.463 D_GP: 0.100 D_real: 0.797 D_fake: 0.674 +(epoch: 439, iters: 7902, time: 0.063) G_GAN: 0.163 G_GAN_Feat: 0.696 G_ID: 0.092 G_Rec: 0.267 D_GP: 0.051 D_real: 0.950 D_fake: 0.837 +(epoch: 439, iters: 8302, time: 0.064) G_GAN: 0.565 G_GAN_Feat: 1.052 G_ID: 0.125 G_Rec: 0.460 D_GP: 0.051 D_real: 0.908 D_fake: 0.441 +(epoch: 439, iters: 8702, time: 0.063) G_GAN: -0.022 G_GAN_Feat: 0.828 G_ID: 0.103 G_Rec: 0.342 D_GP: 0.039 D_real: 0.675 D_fake: 1.022 +(epoch: 440, iters: 494, time: 0.063) G_GAN: 0.201 G_GAN_Feat: 1.013 G_ID: 0.138 G_Rec: 0.460 D_GP: 0.086 D_real: 0.597 D_fake: 0.801 +(epoch: 440, iters: 894, time: 0.063) G_GAN: 0.149 G_GAN_Feat: 0.757 G_ID: 0.093 G_Rec: 0.328 D_GP: 0.041 D_real: 0.920 D_fake: 0.851 +(epoch: 440, iters: 1294, time: 0.064) G_GAN: 0.396 G_GAN_Feat: 1.011 G_ID: 0.114 G_Rec: 0.448 D_GP: 0.041 D_real: 0.734 D_fake: 0.604 +(epoch: 440, iters: 1694, time: 0.063) G_GAN: 0.189 G_GAN_Feat: 0.779 G_ID: 0.108 G_Rec: 0.332 D_GP: 0.032 D_real: 1.048 D_fake: 0.811 +(epoch: 440, iters: 2094, time: 0.063) G_GAN: 0.289 G_GAN_Feat: 1.100 G_ID: 0.120 G_Rec: 0.480 D_GP: 0.347 D_real: 0.258 D_fake: 0.716 +(epoch: 440, iters: 2494, time: 0.063) G_GAN: 0.099 G_GAN_Feat: 0.882 G_ID: 0.103 G_Rec: 0.347 D_GP: 0.073 D_real: 0.295 D_fake: 0.901 +(epoch: 440, iters: 2894, time: 0.064) G_GAN: 0.504 G_GAN_Feat: 0.984 G_ID: 0.104 G_Rec: 0.447 D_GP: 0.038 D_real: 1.062 D_fake: 0.506 +(epoch: 440, iters: 3294, time: 0.063) G_GAN: 0.367 G_GAN_Feat: 0.860 G_ID: 0.093 G_Rec: 0.364 D_GP: 0.036 D_real: 1.078 D_fake: 0.635 +(epoch: 440, iters: 3694, time: 0.063) G_GAN: 0.767 G_GAN_Feat: 1.168 G_ID: 0.123 G_Rec: 0.461 D_GP: 0.132 D_real: 0.501 D_fake: 0.284 +(epoch: 440, iters: 4094, time: 0.063) G_GAN: 0.289 G_GAN_Feat: 0.749 G_ID: 0.097 G_Rec: 0.311 D_GP: 0.033 D_real: 1.287 D_fake: 0.716 +(epoch: 440, iters: 4494, time: 0.064) G_GAN: 0.443 G_GAN_Feat: 0.874 G_ID: 0.135 G_Rec: 0.417 D_GP: 0.028 D_real: 1.158 D_fake: 0.562 +(epoch: 440, iters: 4894, time: 0.063) G_GAN: 0.190 G_GAN_Feat: 0.726 G_ID: 0.091 G_Rec: 0.317 D_GP: 0.048 D_real: 0.992 D_fake: 0.810 +(epoch: 440, iters: 5294, time: 0.063) G_GAN: 0.326 G_GAN_Feat: 0.939 G_ID: 0.122 G_Rec: 0.452 D_GP: 0.054 D_real: 0.902 D_fake: 0.677 +(epoch: 440, iters: 5694, time: 0.063) G_GAN: 0.146 G_GAN_Feat: 0.703 G_ID: 0.091 G_Rec: 0.323 D_GP: 0.031 D_real: 0.960 D_fake: 0.855 +(epoch: 440, iters: 6094, time: 0.064) G_GAN: 0.526 G_GAN_Feat: 0.981 G_ID: 0.109 G_Rec: 0.473 D_GP: 0.028 D_real: 1.081 D_fake: 0.484 +(epoch: 440, iters: 6494, time: 0.063) G_GAN: -0.093 G_GAN_Feat: 0.730 G_ID: 0.104 G_Rec: 0.330 D_GP: 0.064 D_real: 0.667 D_fake: 1.093 +(epoch: 440, iters: 6894, time: 0.063) G_GAN: 0.470 G_GAN_Feat: 1.038 G_ID: 0.131 G_Rec: 0.448 D_GP: 0.096 D_real: 0.605 D_fake: 0.545 +(epoch: 440, iters: 7294, time: 0.063) G_GAN: 0.203 G_GAN_Feat: 0.779 G_ID: 0.095 G_Rec: 0.299 D_GP: 0.040 D_real: 0.914 D_fake: 0.797 +(epoch: 440, iters: 7694, time: 0.064) G_GAN: 0.409 G_GAN_Feat: 0.907 G_ID: 0.112 G_Rec: 0.413 D_GP: 0.034 D_real: 1.054 D_fake: 0.592 +(epoch: 440, iters: 8094, time: 0.063) G_GAN: 0.125 G_GAN_Feat: 0.714 G_ID: 0.090 G_Rec: 0.273 D_GP: 0.032 D_real: 1.004 D_fake: 0.875 +(epoch: 440, iters: 8494, time: 0.063) G_GAN: 0.477 G_GAN_Feat: 0.932 G_ID: 0.164 G_Rec: 0.418 D_GP: 0.035 D_real: 1.011 D_fake: 0.524 +(epoch: 441, iters: 286, time: 0.063) G_GAN: 0.486 G_GAN_Feat: 0.802 G_ID: 0.102 G_Rec: 0.335 D_GP: 0.046 D_real: 1.235 D_fake: 0.523 +(epoch: 441, iters: 686, time: 0.064) G_GAN: 0.856 G_GAN_Feat: 1.025 G_ID: 0.133 G_Rec: 0.467 D_GP: 0.034 D_real: 1.165 D_fake: 0.227 +(epoch: 441, iters: 1086, time: 0.063) G_GAN: 0.232 G_GAN_Feat: 0.836 G_ID: 0.123 G_Rec: 0.326 D_GP: 0.063 D_real: 0.820 D_fake: 0.779 +(epoch: 441, iters: 1486, time: 0.063) G_GAN: 0.412 G_GAN_Feat: 0.947 G_ID: 0.124 G_Rec: 0.424 D_GP: 0.030 D_real: 1.010 D_fake: 0.626 +(epoch: 441, iters: 1886, time: 0.063) G_GAN: 0.036 G_GAN_Feat: 0.775 G_ID: 0.087 G_Rec: 0.314 D_GP: 0.034 D_real: 0.811 D_fake: 0.968 +(epoch: 441, iters: 2286, time: 0.064) G_GAN: 0.392 G_GAN_Feat: 0.996 G_ID: 0.125 G_Rec: 0.439 D_GP: 0.078 D_real: 0.839 D_fake: 0.614 +(epoch: 441, iters: 2686, time: 0.063) G_GAN: -0.070 G_GAN_Feat: 0.710 G_ID: 0.104 G_Rec: 0.280 D_GP: 0.028 D_real: 0.831 D_fake: 1.070 +(epoch: 441, iters: 3086, time: 0.063) G_GAN: 0.618 G_GAN_Feat: 1.185 G_ID: 0.128 G_Rec: 0.455 D_GP: 0.060 D_real: 0.390 D_fake: 0.427 +(epoch: 441, iters: 3486, time: 0.063) G_GAN: 0.542 G_GAN_Feat: 0.791 G_ID: 0.101 G_Rec: 0.304 D_GP: 0.063 D_real: 1.164 D_fake: 0.470 +(epoch: 441, iters: 3886, time: 0.064) G_GAN: 0.351 G_GAN_Feat: 1.193 G_ID: 0.121 G_Rec: 0.445 D_GP: 0.147 D_real: 0.259 D_fake: 0.651 +(epoch: 441, iters: 4286, time: 0.063) G_GAN: 0.513 G_GAN_Feat: 0.783 G_ID: 0.106 G_Rec: 0.300 D_GP: 0.041 D_real: 1.226 D_fake: 0.494 +(epoch: 441, iters: 4686, time: 0.063) G_GAN: 0.529 G_GAN_Feat: 0.914 G_ID: 0.105 G_Rec: 0.442 D_GP: 0.029 D_real: 1.225 D_fake: 0.520 +(epoch: 441, iters: 5086, time: 0.063) G_GAN: -0.083 G_GAN_Feat: 0.699 G_ID: 0.102 G_Rec: 0.295 D_GP: 0.030 D_real: 0.819 D_fake: 1.083 +(epoch: 441, iters: 5486, time: 0.064) G_GAN: 0.056 G_GAN_Feat: 0.905 G_ID: 0.119 G_Rec: 0.443 D_GP: 0.038 D_real: 0.719 D_fake: 0.945 +(epoch: 441, iters: 5886, time: 0.063) G_GAN: -0.166 G_GAN_Feat: 0.712 G_ID: 0.108 G_Rec: 0.314 D_GP: 0.036 D_real: 0.707 D_fake: 1.166 +(epoch: 441, iters: 6286, time: 0.063) G_GAN: -0.010 G_GAN_Feat: 0.992 G_ID: 0.134 G_Rec: 0.464 D_GP: 0.109 D_real: 0.509 D_fake: 1.010 +(epoch: 441, iters: 6686, time: 0.063) G_GAN: -0.204 G_GAN_Feat: 0.697 G_ID: 0.096 G_Rec: 0.315 D_GP: 0.043 D_real: 0.666 D_fake: 1.204 +(epoch: 441, iters: 7086, time: 0.064) G_GAN: 0.324 G_GAN_Feat: 1.031 G_ID: 0.115 G_Rec: 0.475 D_GP: 0.069 D_real: 0.796 D_fake: 0.688 +(epoch: 441, iters: 7486, time: 0.063) G_GAN: -0.298 G_GAN_Feat: 0.911 G_ID: 0.114 G_Rec: 0.371 D_GP: 0.127 D_real: 0.204 D_fake: 1.298 +(epoch: 441, iters: 7886, time: 0.063) G_GAN: 0.466 G_GAN_Feat: 0.986 G_ID: 0.121 G_Rec: 0.480 D_GP: 0.073 D_real: 0.839 D_fake: 0.574 +(epoch: 441, iters: 8286, time: 0.063) G_GAN: 0.230 G_GAN_Feat: 0.698 G_ID: 0.100 G_Rec: 0.276 D_GP: 0.042 D_real: 1.094 D_fake: 0.772 +(epoch: 441, iters: 8686, time: 0.064) G_GAN: 0.423 G_GAN_Feat: 0.994 G_ID: 0.118 G_Rec: 0.492 D_GP: 0.038 D_real: 1.025 D_fake: 0.579 +(epoch: 442, iters: 478, time: 0.063) G_GAN: 0.271 G_GAN_Feat: 0.813 G_ID: 0.110 G_Rec: 0.320 D_GP: 0.111 D_real: 0.778 D_fake: 0.735 +(epoch: 442, iters: 878, time: 0.063) G_GAN: 0.926 G_GAN_Feat: 0.994 G_ID: 0.114 G_Rec: 0.452 D_GP: 0.047 D_real: 1.296 D_fake: 0.203 +(epoch: 442, iters: 1278, time: 0.063) G_GAN: 0.022 G_GAN_Feat: 0.718 G_ID: 0.116 G_Rec: 0.366 D_GP: 0.028 D_real: 0.975 D_fake: 0.978 +(epoch: 442, iters: 1678, time: 0.064) G_GAN: 0.342 G_GAN_Feat: 0.888 G_ID: 0.118 G_Rec: 0.430 D_GP: 0.026 D_real: 1.124 D_fake: 0.660 +(epoch: 442, iters: 2078, time: 0.063) G_GAN: 0.111 G_GAN_Feat: 0.676 G_ID: 0.106 G_Rec: 0.324 D_GP: 0.030 D_real: 1.039 D_fake: 0.890 +(epoch: 442, iters: 2478, time: 0.063) G_GAN: 0.161 G_GAN_Feat: 0.838 G_ID: 0.106 G_Rec: 0.384 D_GP: 0.036 D_real: 0.958 D_fake: 0.839 +(epoch: 442, iters: 2878, time: 0.063) G_GAN: 0.236 G_GAN_Feat: 0.637 G_ID: 0.095 G_Rec: 0.260 D_GP: 0.032 D_real: 1.133 D_fake: 0.764 +(epoch: 442, iters: 3278, time: 0.064) G_GAN: 0.161 G_GAN_Feat: 0.880 G_ID: 0.143 G_Rec: 0.392 D_GP: 0.032 D_real: 0.883 D_fake: 0.840 +(epoch: 442, iters: 3678, time: 0.063) G_GAN: -0.039 G_GAN_Feat: 0.803 G_ID: 0.097 G_Rec: 0.347 D_GP: 0.073 D_real: 0.627 D_fake: 1.039 +(epoch: 442, iters: 4078, time: 0.063) G_GAN: 0.537 G_GAN_Feat: 0.981 G_ID: 0.114 G_Rec: 0.430 D_GP: 0.044 D_real: 1.037 D_fake: 0.470 +(epoch: 442, iters: 4478, time: 0.063) G_GAN: 0.335 G_GAN_Feat: 0.735 G_ID: 0.079 G_Rec: 0.298 D_GP: 0.033 D_real: 1.169 D_fake: 0.665 +(epoch: 442, iters: 4878, time: 0.064) G_GAN: 0.628 G_GAN_Feat: 1.020 G_ID: 0.101 G_Rec: 0.420 D_GP: 0.094 D_real: 0.998 D_fake: 0.396 +(epoch: 442, iters: 5278, time: 0.063) G_GAN: 0.062 G_GAN_Feat: 0.736 G_ID: 0.109 G_Rec: 0.294 D_GP: 0.031 D_real: 0.874 D_fake: 0.938 +(epoch: 442, iters: 5678, time: 0.063) G_GAN: 0.503 G_GAN_Feat: 1.020 G_ID: 0.128 G_Rec: 0.431 D_GP: 0.053 D_real: 0.955 D_fake: 0.512 +(epoch: 442, iters: 6078, time: 0.063) G_GAN: 0.208 G_GAN_Feat: 0.759 G_ID: 0.086 G_Rec: 0.303 D_GP: 0.047 D_real: 0.900 D_fake: 0.792 +(epoch: 442, iters: 6478, time: 0.064) G_GAN: 0.646 G_GAN_Feat: 0.893 G_ID: 0.122 G_Rec: 0.366 D_GP: 0.032 D_real: 1.351 D_fake: 0.370 +(epoch: 442, iters: 6878, time: 0.063) G_GAN: 0.409 G_GAN_Feat: 0.970 G_ID: 0.121 G_Rec: 0.353 D_GP: 0.050 D_real: 1.071 D_fake: 0.657 +(epoch: 442, iters: 7278, time: 0.063) G_GAN: 0.676 G_GAN_Feat: 1.024 G_ID: 0.104 G_Rec: 0.455 D_GP: 0.039 D_real: 1.304 D_fake: 0.455 +(epoch: 442, iters: 7678, time: 0.063) G_GAN: 0.238 G_GAN_Feat: 0.804 G_ID: 0.094 G_Rec: 0.317 D_GP: 0.043 D_real: 1.062 D_fake: 0.762 +(epoch: 442, iters: 8078, time: 0.064) G_GAN: 0.735 G_GAN_Feat: 1.119 G_ID: 0.130 G_Rec: 0.508 D_GP: 0.046 D_real: 0.930 D_fake: 0.403 +(epoch: 442, iters: 8478, time: 0.063) G_GAN: 0.104 G_GAN_Feat: 0.774 G_ID: 0.091 G_Rec: 0.296 D_GP: 0.028 D_real: 0.984 D_fake: 0.896 +(epoch: 443, iters: 270, time: 0.063) G_GAN: 0.572 G_GAN_Feat: 1.013 G_ID: 0.120 G_Rec: 0.452 D_GP: 0.036 D_real: 1.103 D_fake: 0.433 +(epoch: 443, iters: 670, time: 0.063) G_GAN: 0.204 G_GAN_Feat: 1.287 G_ID: 0.130 G_Rec: 0.430 D_GP: 0.212 D_real: 0.909 D_fake: 0.833 +(epoch: 443, iters: 1070, time: 0.064) G_GAN: 0.179 G_GAN_Feat: 0.883 G_ID: 0.140 G_Rec: 0.457 D_GP: 0.028 D_real: 0.964 D_fake: 0.843 +(epoch: 443, iters: 1470, time: 0.063) G_GAN: -0.003 G_GAN_Feat: 0.691 G_ID: 0.107 G_Rec: 0.304 D_GP: 0.029 D_real: 0.837 D_fake: 1.003 +(epoch: 443, iters: 1870, time: 0.063) G_GAN: 0.197 G_GAN_Feat: 0.978 G_ID: 0.115 G_Rec: 0.488 D_GP: 0.038 D_real: 0.740 D_fake: 0.810 +(epoch: 443, iters: 2270, time: 0.063) G_GAN: 0.021 G_GAN_Feat: 0.721 G_ID: 0.103 G_Rec: 0.287 D_GP: 0.037 D_real: 0.836 D_fake: 0.979 +(epoch: 443, iters: 2670, time: 0.064) G_GAN: 0.509 G_GAN_Feat: 0.969 G_ID: 0.124 G_Rec: 0.402 D_GP: 0.041 D_real: 1.108 D_fake: 0.533 +(epoch: 443, iters: 3070, time: 0.063) G_GAN: 0.183 G_GAN_Feat: 0.850 G_ID: 0.091 G_Rec: 0.299 D_GP: 0.037 D_real: 0.791 D_fake: 0.817 +(epoch: 443, iters: 3470, time: 0.063) G_GAN: 0.689 G_GAN_Feat: 0.967 G_ID: 0.099 G_Rec: 0.413 D_GP: 0.038 D_real: 1.382 D_fake: 0.331 +(epoch: 443, iters: 3870, time: 0.063) G_GAN: 0.182 G_GAN_Feat: 0.845 G_ID: 0.111 G_Rec: 0.305 D_GP: 0.039 D_real: 0.832 D_fake: 0.819 +(epoch: 443, iters: 4270, time: 0.064) G_GAN: 0.892 G_GAN_Feat: 1.015 G_ID: 0.114 G_Rec: 0.432 D_GP: 0.043 D_real: 1.315 D_fake: 0.179 +(epoch: 443, iters: 4670, time: 0.063) G_GAN: 0.237 G_GAN_Feat: 0.616 G_ID: 0.094 G_Rec: 0.306 D_GP: 0.026 D_real: 1.130 D_fake: 0.763 +(epoch: 443, iters: 5070, time: 0.063) G_GAN: -0.049 G_GAN_Feat: 0.902 G_ID: 0.121 G_Rec: 0.447 D_GP: 0.050 D_real: 0.605 D_fake: 1.049 +(epoch: 443, iters: 5470, time: 0.063) G_GAN: -0.037 G_GAN_Feat: 0.687 G_ID: 0.113 G_Rec: 0.326 D_GP: 0.045 D_real: 0.834 D_fake: 1.037 +(epoch: 443, iters: 5870, time: 0.064) G_GAN: 0.297 G_GAN_Feat: 0.924 G_ID: 0.139 G_Rec: 0.484 D_GP: 0.043 D_real: 0.834 D_fake: 0.706 +(epoch: 443, iters: 6270, time: 0.063) G_GAN: 0.159 G_GAN_Feat: 0.698 G_ID: 0.096 G_Rec: 0.304 D_GP: 0.068 D_real: 0.953 D_fake: 0.843 +(epoch: 443, iters: 6670, time: 0.063) G_GAN: 0.435 G_GAN_Feat: 1.017 G_ID: 0.133 G_Rec: 0.477 D_GP: 0.153 D_real: 0.558 D_fake: 0.580 +(epoch: 443, iters: 7070, time: 0.063) G_GAN: 0.335 G_GAN_Feat: 0.759 G_ID: 0.089 G_Rec: 0.321 D_GP: 0.046 D_real: 1.280 D_fake: 0.669 +(epoch: 443, iters: 7470, time: 0.063) G_GAN: 0.476 G_GAN_Feat: 0.892 G_ID: 0.129 G_Rec: 0.380 D_GP: 0.041 D_real: 1.070 D_fake: 0.536 +(epoch: 443, iters: 7870, time: 0.063) G_GAN: 0.444 G_GAN_Feat: 0.751 G_ID: 0.100 G_Rec: 0.320 D_GP: 0.048 D_real: 1.253 D_fake: 0.571 +(epoch: 443, iters: 8270, time: 0.063) G_GAN: 0.129 G_GAN_Feat: 0.941 G_ID: 0.123 G_Rec: 0.433 D_GP: 0.040 D_real: 0.771 D_fake: 0.871 +(epoch: 443, iters: 8670, time: 0.064) G_GAN: -0.131 G_GAN_Feat: 0.714 G_ID: 0.113 G_Rec: 0.297 D_GP: 0.045 D_real: 0.809 D_fake: 1.131 +(epoch: 444, iters: 462, time: 0.063) G_GAN: 0.182 G_GAN_Feat: 0.929 G_ID: 0.131 G_Rec: 0.434 D_GP: 0.030 D_real: 0.882 D_fake: 0.818 +(epoch: 444, iters: 862, time: 0.063) G_GAN: 0.296 G_GAN_Feat: 0.913 G_ID: 0.105 G_Rec: 0.351 D_GP: 0.211 D_real: 0.403 D_fake: 0.731 +(epoch: 444, iters: 1262, time: 0.063) G_GAN: 0.215 G_GAN_Feat: 1.019 G_ID: 0.146 G_Rec: 0.442 D_GP: 0.044 D_real: 0.497 D_fake: 0.788 +(epoch: 444, iters: 1662, time: 0.064) G_GAN: 0.280 G_GAN_Feat: 1.010 G_ID: 0.112 G_Rec: 0.361 D_GP: 0.205 D_real: 0.282 D_fake: 0.771 +(epoch: 444, iters: 2062, time: 0.063) G_GAN: 0.571 G_GAN_Feat: 0.932 G_ID: 0.119 G_Rec: 0.378 D_GP: 0.033 D_real: 1.087 D_fake: 0.469 +(epoch: 444, iters: 2462, time: 0.063) G_GAN: 0.047 G_GAN_Feat: 0.899 G_ID: 0.105 G_Rec: 0.328 D_GP: 0.107 D_real: 0.378 D_fake: 0.960 +(epoch: 444, iters: 2862, time: 0.063) G_GAN: 0.388 G_GAN_Feat: 1.069 G_ID: 0.137 G_Rec: 0.389 D_GP: 0.044 D_real: 0.844 D_fake: 0.614 +(epoch: 444, iters: 3262, time: 0.064) G_GAN: 0.135 G_GAN_Feat: 0.780 G_ID: 0.096 G_Rec: 0.318 D_GP: 0.036 D_real: 0.867 D_fake: 0.865 +(epoch: 444, iters: 3662, time: 0.064) G_GAN: 0.589 G_GAN_Feat: 1.128 G_ID: 0.117 G_Rec: 0.497 D_GP: 0.094 D_real: 0.765 D_fake: 0.423 +(epoch: 444, iters: 4062, time: 0.064) G_GAN: -0.105 G_GAN_Feat: 0.879 G_ID: 0.095 G_Rec: 0.339 D_GP: 0.422 D_real: 0.351 D_fake: 1.105 +(epoch: 444, iters: 4462, time: 0.064) G_GAN: 0.247 G_GAN_Feat: 1.028 G_ID: 0.133 G_Rec: 0.412 D_GP: 0.036 D_real: 1.095 D_fake: 0.755 +(epoch: 444, iters: 4862, time: 0.064) G_GAN: 0.460 G_GAN_Feat: 0.772 G_ID: 0.095 G_Rec: 0.291 D_GP: 0.029 D_real: 1.238 D_fake: 0.542 +(epoch: 444, iters: 5262, time: 0.064) G_GAN: 0.366 G_GAN_Feat: 0.903 G_ID: 0.104 G_Rec: 0.381 D_GP: 0.035 D_real: 0.980 D_fake: 0.635 +(epoch: 444, iters: 5662, time: 0.063) G_GAN: 0.263 G_GAN_Feat: 0.828 G_ID: 0.095 G_Rec: 0.293 D_GP: 0.033 D_real: 0.815 D_fake: 0.738 +(epoch: 444, iters: 6062, time: 0.064) G_GAN: 0.323 G_GAN_Feat: 0.961 G_ID: 0.114 G_Rec: 0.459 D_GP: 0.034 D_real: 0.949 D_fake: 0.678 +(epoch: 444, iters: 6462, time: 0.064) G_GAN: 0.347 G_GAN_Feat: 0.758 G_ID: 0.095 G_Rec: 0.337 D_GP: 0.032 D_real: 1.246 D_fake: 0.665 +(epoch: 444, iters: 6862, time: 0.064) G_GAN: 0.075 G_GAN_Feat: 0.891 G_ID: 0.117 G_Rec: 0.420 D_GP: 0.033 D_real: 0.702 D_fake: 0.925 +(epoch: 444, iters: 7262, time: 0.064) G_GAN: -0.182 G_GAN_Feat: 0.697 G_ID: 0.123 G_Rec: 0.315 D_GP: 0.033 D_real: 0.702 D_fake: 1.182 +(epoch: 444, iters: 7662, time: 0.064) G_GAN: 0.164 G_GAN_Feat: 0.857 G_ID: 0.115 G_Rec: 0.397 D_GP: 0.034 D_real: 0.929 D_fake: 0.839 +(epoch: 444, iters: 8062, time: 0.064) G_GAN: 0.097 G_GAN_Feat: 0.758 G_ID: 0.100 G_Rec: 0.346 D_GP: 0.050 D_real: 0.865 D_fake: 0.903 +(epoch: 444, iters: 8462, time: 0.064) G_GAN: 0.405 G_GAN_Feat: 0.816 G_ID: 0.125 G_Rec: 0.374 D_GP: 0.034 D_real: 1.152 D_fake: 0.602 +(epoch: 445, iters: 254, time: 0.064) G_GAN: -0.104 G_GAN_Feat: 0.699 G_ID: 0.100 G_Rec: 0.314 D_GP: 0.031 D_real: 0.790 D_fake: 1.104 +(epoch: 445, iters: 654, time: 0.064) G_GAN: 0.202 G_GAN_Feat: 1.007 G_ID: 0.119 G_Rec: 0.458 D_GP: 0.054 D_real: 0.712 D_fake: 0.799 +(epoch: 445, iters: 1054, time: 0.064) G_GAN: 0.397 G_GAN_Feat: 0.752 G_ID: 0.089 G_Rec: 0.300 D_GP: 0.042 D_real: 1.175 D_fake: 0.606 +(epoch: 445, iters: 1454, time: 0.064) G_GAN: 0.638 G_GAN_Feat: 0.935 G_ID: 0.114 G_Rec: 0.424 D_GP: 0.039 D_real: 1.195 D_fake: 0.379 +(epoch: 445, iters: 1854, time: 0.064) G_GAN: 0.492 G_GAN_Feat: 0.744 G_ID: 0.104 G_Rec: 0.338 D_GP: 0.034 D_real: 1.378 D_fake: 0.517 +(epoch: 445, iters: 2254, time: 0.063) G_GAN: 0.552 G_GAN_Feat: 0.845 G_ID: 0.103 G_Rec: 0.460 D_GP: 0.028 D_real: 1.274 D_fake: 0.466 +(epoch: 445, iters: 2654, time: 0.064) G_GAN: 0.254 G_GAN_Feat: 0.618 G_ID: 0.087 G_Rec: 0.285 D_GP: 0.030 D_real: 1.124 D_fake: 0.747 +(epoch: 445, iters: 3054, time: 0.064) G_GAN: 0.027 G_GAN_Feat: 0.930 G_ID: 0.121 G_Rec: 0.449 D_GP: 0.043 D_real: 0.703 D_fake: 0.973 +(epoch: 445, iters: 3454, time: 0.063) G_GAN: -0.039 G_GAN_Feat: 0.744 G_ID: 0.106 G_Rec: 0.309 D_GP: 0.043 D_real: 0.784 D_fake: 1.039 +(epoch: 445, iters: 3854, time: 0.063) G_GAN: 0.545 G_GAN_Feat: 0.957 G_ID: 0.110 G_Rec: 0.433 D_GP: 0.052 D_real: 1.070 D_fake: 0.473 +(epoch: 445, iters: 4254, time: 0.064) G_GAN: -0.029 G_GAN_Feat: 0.816 G_ID: 0.110 G_Rec: 0.338 D_GP: 0.047 D_real: 0.793 D_fake: 1.031 +(epoch: 445, iters: 4654, time: 0.063) G_GAN: 0.570 G_GAN_Feat: 1.041 G_ID: 0.115 G_Rec: 0.410 D_GP: 0.172 D_real: 0.685 D_fake: 0.440 +(epoch: 445, iters: 5054, time: 0.063) G_GAN: 0.110 G_GAN_Feat: 0.760 G_ID: 0.120 G_Rec: 0.300 D_GP: 0.037 D_real: 0.854 D_fake: 0.890 +(epoch: 445, iters: 5454, time: 0.063) G_GAN: 0.251 G_GAN_Feat: 1.000 G_ID: 0.107 G_Rec: 0.413 D_GP: 0.038 D_real: 0.799 D_fake: 0.749 +(epoch: 445, iters: 5854, time: 0.064) G_GAN: 0.370 G_GAN_Feat: 0.858 G_ID: 0.094 G_Rec: 0.327 D_GP: 0.038 D_real: 1.058 D_fake: 0.630 +(epoch: 445, iters: 6254, time: 0.063) G_GAN: 0.483 G_GAN_Feat: 0.967 G_ID: 0.128 G_Rec: 0.416 D_GP: 0.036 D_real: 1.037 D_fake: 0.523 +(epoch: 445, iters: 6654, time: 0.063) G_GAN: 0.350 G_GAN_Feat: 1.045 G_ID: 0.126 G_Rec: 0.354 D_GP: 0.211 D_real: 0.277 D_fake: 0.725 +(epoch: 445, iters: 7054, time: 0.063) G_GAN: 0.526 G_GAN_Feat: 1.251 G_ID: 0.124 G_Rec: 0.502 D_GP: 0.359 D_real: 0.528 D_fake: 0.499 +(epoch: 445, iters: 7454, time: 0.064) G_GAN: 0.327 G_GAN_Feat: 0.947 G_ID: 0.101 G_Rec: 0.401 D_GP: 0.045 D_real: 1.064 D_fake: 0.699 +(epoch: 445, iters: 7854, time: 0.063) G_GAN: 0.266 G_GAN_Feat: 0.860 G_ID: 0.104 G_Rec: 0.439 D_GP: 0.027 D_real: 0.987 D_fake: 0.738 +(epoch: 445, iters: 8254, time: 0.063) G_GAN: 0.121 G_GAN_Feat: 0.667 G_ID: 0.082 G_Rec: 0.298 D_GP: 0.029 D_real: 1.063 D_fake: 0.879 +(epoch: 445, iters: 8654, time: 0.063) G_GAN: 0.166 G_GAN_Feat: 0.874 G_ID: 0.126 G_Rec: 0.411 D_GP: 0.038 D_real: 0.785 D_fake: 0.834 +(epoch: 446, iters: 446, time: 0.064) G_GAN: -0.054 G_GAN_Feat: 0.684 G_ID: 0.084 G_Rec: 0.349 D_GP: 0.055 D_real: 0.846 D_fake: 1.054 +(epoch: 446, iters: 846, time: 0.063) G_GAN: 0.390 G_GAN_Feat: 0.909 G_ID: 0.114 G_Rec: 0.439 D_GP: 0.033 D_real: 0.940 D_fake: 0.621 +(epoch: 446, iters: 1246, time: 0.063) G_GAN: 0.207 G_GAN_Feat: 0.735 G_ID: 0.100 G_Rec: 0.320 D_GP: 0.039 D_real: 1.041 D_fake: 0.794 +(epoch: 446, iters: 1646, time: 0.063) G_GAN: 0.307 G_GAN_Feat: 0.994 G_ID: 0.112 G_Rec: 0.427 D_GP: 0.037 D_real: 0.852 D_fake: 0.693 +(epoch: 446, iters: 2046, time: 0.064) G_GAN: 0.309 G_GAN_Feat: 0.705 G_ID: 0.102 G_Rec: 0.299 D_GP: 0.040 D_real: 1.180 D_fake: 0.691 +(epoch: 446, iters: 2446, time: 0.063) G_GAN: 0.486 G_GAN_Feat: 0.985 G_ID: 0.130 G_Rec: 0.439 D_GP: 0.037 D_real: 1.112 D_fake: 0.553 +(epoch: 446, iters: 2846, time: 0.063) G_GAN: 0.337 G_GAN_Feat: 0.749 G_ID: 0.098 G_Rec: 0.336 D_GP: 0.033 D_real: 1.214 D_fake: 0.666 +(epoch: 446, iters: 3246, time: 0.063) G_GAN: 0.335 G_GAN_Feat: 0.843 G_ID: 0.109 G_Rec: 0.430 D_GP: 0.029 D_real: 1.067 D_fake: 0.666 +(epoch: 446, iters: 3646, time: 0.064) G_GAN: 0.368 G_GAN_Feat: 0.735 G_ID: 0.099 G_Rec: 0.324 D_GP: 0.033 D_real: 1.176 D_fake: 0.635 +(epoch: 446, iters: 4046, time: 0.063) G_GAN: 0.318 G_GAN_Feat: 0.924 G_ID: 0.129 G_Rec: 0.395 D_GP: 0.061 D_real: 0.957 D_fake: 0.685 +(epoch: 446, iters: 4446, time: 0.063) G_GAN: 0.371 G_GAN_Feat: 0.838 G_ID: 0.105 G_Rec: 0.319 D_GP: 0.084 D_real: 0.656 D_fake: 0.643 +(epoch: 446, iters: 4846, time: 0.063) G_GAN: 0.593 G_GAN_Feat: 1.064 G_ID: 0.106 G_Rec: 0.439 D_GP: 0.061 D_real: 0.848 D_fake: 0.419 +(epoch: 446, iters: 5246, time: 0.064) G_GAN: 0.283 G_GAN_Feat: 0.718 G_ID: 0.095 G_Rec: 0.283 D_GP: 0.028 D_real: 1.153 D_fake: 0.717 +(epoch: 446, iters: 5646, time: 0.064) G_GAN: 0.459 G_GAN_Feat: 1.025 G_ID: 0.123 G_Rec: 0.412 D_GP: 0.044 D_real: 0.914 D_fake: 0.547 +(epoch: 446, iters: 6046, time: 0.064) G_GAN: 0.643 G_GAN_Feat: 0.789 G_ID: 0.107 G_Rec: 0.321 D_GP: 0.040 D_real: 1.457 D_fake: 0.385 +(epoch: 446, iters: 6446, time: 0.063) G_GAN: 0.482 G_GAN_Feat: 1.030 G_ID: 0.130 G_Rec: 0.408 D_GP: 0.056 D_real: 0.687 D_fake: 0.537 +(epoch: 446, iters: 6846, time: 0.064) G_GAN: 0.259 G_GAN_Feat: 0.872 G_ID: 0.110 G_Rec: 0.302 D_GP: 0.039 D_real: 0.525 D_fake: 0.747 +(epoch: 446, iters: 7246, time: 0.063) G_GAN: 0.258 G_GAN_Feat: 0.917 G_ID: 0.118 G_Rec: 0.443 D_GP: 0.034 D_real: 1.034 D_fake: 0.753 +(epoch: 446, iters: 7646, time: 0.063) G_GAN: 0.103 G_GAN_Feat: 0.674 G_ID: 0.087 G_Rec: 0.301 D_GP: 0.040 D_real: 1.018 D_fake: 0.897 +(epoch: 446, iters: 8046, time: 0.063) G_GAN: 0.222 G_GAN_Feat: 1.004 G_ID: 0.117 G_Rec: 0.502 D_GP: 0.076 D_real: 0.801 D_fake: 0.783 +(epoch: 446, iters: 8446, time: 0.064) G_GAN: 0.011 G_GAN_Feat: 0.660 G_ID: 0.093 G_Rec: 0.296 D_GP: 0.039 D_real: 0.869 D_fake: 0.989 +(epoch: 447, iters: 238, time: 0.064) G_GAN: 0.347 G_GAN_Feat: 0.907 G_ID: 0.106 G_Rec: 0.454 D_GP: 0.034 D_real: 0.918 D_fake: 0.661 +(epoch: 447, iters: 638, time: 0.063) G_GAN: 0.145 G_GAN_Feat: 0.700 G_ID: 0.091 G_Rec: 0.315 D_GP: 0.033 D_real: 1.045 D_fake: 0.858 +(epoch: 447, iters: 1038, time: 0.063) G_GAN: 0.569 G_GAN_Feat: 0.886 G_ID: 0.119 G_Rec: 0.400 D_GP: 0.032 D_real: 1.147 D_fake: 0.480 +(epoch: 447, iters: 1438, time: 0.064) G_GAN: 0.095 G_GAN_Feat: 0.723 G_ID: 0.091 G_Rec: 0.310 D_GP: 0.054 D_real: 0.870 D_fake: 0.905 +(epoch: 447, iters: 1838, time: 0.063) G_GAN: 0.288 G_GAN_Feat: 0.897 G_ID: 0.128 G_Rec: 0.401 D_GP: 0.036 D_real: 0.882 D_fake: 0.713 +(epoch: 447, iters: 2238, time: 0.063) G_GAN: 0.316 G_GAN_Feat: 0.772 G_ID: 0.105 G_Rec: 0.336 D_GP: 0.039 D_real: 1.020 D_fake: 0.690 +(epoch: 447, iters: 2638, time: 0.063) G_GAN: 0.382 G_GAN_Feat: 1.035 G_ID: 0.110 G_Rec: 0.478 D_GP: 0.067 D_real: 0.851 D_fake: 0.626 +(epoch: 447, iters: 3038, time: 0.064) G_GAN: 0.217 G_GAN_Feat: 0.768 G_ID: 0.101 G_Rec: 0.333 D_GP: 0.043 D_real: 1.119 D_fake: 0.783 +(epoch: 447, iters: 3438, time: 0.063) G_GAN: 0.507 G_GAN_Feat: 1.064 G_ID: 0.116 G_Rec: 0.455 D_GP: 0.131 D_real: 0.866 D_fake: 0.503 +(epoch: 447, iters: 3838, time: 0.064) G_GAN: 0.243 G_GAN_Feat: 0.753 G_ID: 0.100 G_Rec: 0.329 D_GP: 0.054 D_real: 0.989 D_fake: 0.757 +(epoch: 447, iters: 4238, time: 0.063) G_GAN: 0.411 G_GAN_Feat: 0.953 G_ID: 0.138 G_Rec: 0.429 D_GP: 0.039 D_real: 0.934 D_fake: 0.593 +(epoch: 447, iters: 4638, time: 0.064) G_GAN: 0.242 G_GAN_Feat: 0.686 G_ID: 0.107 G_Rec: 0.292 D_GP: 0.032 D_real: 1.129 D_fake: 0.759 +(epoch: 447, iters: 5038, time: 0.064) G_GAN: 0.337 G_GAN_Feat: 0.948 G_ID: 0.113 G_Rec: 0.402 D_GP: 0.051 D_real: 0.770 D_fake: 0.664 +(epoch: 447, iters: 5438, time: 0.063) G_GAN: 0.244 G_GAN_Feat: 0.685 G_ID: 0.102 G_Rec: 0.326 D_GP: 0.032 D_real: 1.147 D_fake: 0.756 +(epoch: 447, iters: 5838, time: 0.063) G_GAN: 0.238 G_GAN_Feat: 0.983 G_ID: 0.116 G_Rec: 0.477 D_GP: 0.037 D_real: 0.687 D_fake: 0.762 +(epoch: 447, iters: 6238, time: 0.064) G_GAN: -0.070 G_GAN_Feat: 0.797 G_ID: 0.115 G_Rec: 0.314 D_GP: 0.044 D_real: 0.622 D_fake: 1.070 +(epoch: 447, iters: 6638, time: 0.064) G_GAN: 0.688 G_GAN_Feat: 0.983 G_ID: 0.121 G_Rec: 0.443 D_GP: 0.040 D_real: 1.216 D_fake: 0.381 +(epoch: 447, iters: 7038, time: 0.063) G_GAN: 0.177 G_GAN_Feat: 0.805 G_ID: 0.106 G_Rec: 0.344 D_GP: 0.041 D_real: 0.834 D_fake: 0.824 +(epoch: 447, iters: 7438, time: 0.063) G_GAN: 0.363 G_GAN_Feat: 1.012 G_ID: 0.137 G_Rec: 0.423 D_GP: 0.050 D_real: 0.842 D_fake: 0.638 +(epoch: 447, iters: 7838, time: 0.064) G_GAN: 0.251 G_GAN_Feat: 0.695 G_ID: 0.097 G_Rec: 0.309 D_GP: 0.036 D_real: 1.084 D_fake: 0.749 +(epoch: 447, iters: 8238, time: 0.063) G_GAN: 0.520 G_GAN_Feat: 1.038 G_ID: 0.112 G_Rec: 0.443 D_GP: 0.040 D_real: 0.914 D_fake: 0.485 +(epoch: 447, iters: 8638, time: 0.063) G_GAN: 0.306 G_GAN_Feat: 0.766 G_ID: 0.107 G_Rec: 0.313 D_GP: 0.039 D_real: 1.013 D_fake: 0.695 +(epoch: 448, iters: 430, time: 0.063) G_GAN: 0.564 G_GAN_Feat: 0.913 G_ID: 0.130 G_Rec: 0.411 D_GP: 0.037 D_real: 1.244 D_fake: 0.453 +(epoch: 448, iters: 830, time: 0.064) G_GAN: 0.080 G_GAN_Feat: 0.882 G_ID: 0.098 G_Rec: 0.346 D_GP: 0.137 D_real: 0.365 D_fake: 0.921 +(epoch: 448, iters: 1230, time: 0.064) G_GAN: 0.483 G_GAN_Feat: 0.994 G_ID: 0.114 G_Rec: 0.433 D_GP: 0.035 D_real: 0.927 D_fake: 0.521 +(epoch: 448, iters: 1630, time: 0.064) G_GAN: 0.123 G_GAN_Feat: 0.836 G_ID: 0.121 G_Rec: 0.351 D_GP: 0.054 D_real: 0.541 D_fake: 0.878 +(epoch: 448, iters: 2030, time: 0.064) G_GAN: 0.360 G_GAN_Feat: 1.002 G_ID: 0.116 G_Rec: 0.420 D_GP: 0.117 D_real: 0.525 D_fake: 0.649 +(epoch: 448, iters: 2430, time: 0.064) G_GAN: -0.086 G_GAN_Feat: 0.836 G_ID: 0.083 G_Rec: 0.319 D_GP: 0.247 D_real: 0.370 D_fake: 1.091 +(epoch: 448, iters: 2830, time: 0.064) G_GAN: 0.619 G_GAN_Feat: 1.047 G_ID: 0.097 G_Rec: 0.456 D_GP: 0.040 D_real: 1.098 D_fake: 0.394 +(epoch: 448, iters: 3230, time: 0.064) G_GAN: 0.135 G_GAN_Feat: 0.865 G_ID: 0.090 G_Rec: 0.332 D_GP: 0.187 D_real: 0.507 D_fake: 0.866 +(epoch: 448, iters: 3630, time: 0.064) G_GAN: 0.457 G_GAN_Feat: 1.024 G_ID: 0.128 G_Rec: 0.434 D_GP: 0.052 D_real: 0.790 D_fake: 0.552 +(epoch: 448, iters: 4030, time: 0.064) G_GAN: 0.011 G_GAN_Feat: 0.848 G_ID: 0.112 G_Rec: 0.305 D_GP: 0.079 D_real: 0.350 D_fake: 0.994 +(epoch: 448, iters: 4430, time: 0.064) G_GAN: 0.359 G_GAN_Feat: 0.994 G_ID: 0.121 G_Rec: 0.455 D_GP: 0.036 D_real: 1.001 D_fake: 0.643 +(epoch: 448, iters: 4830, time: 0.064) G_GAN: -0.046 G_GAN_Feat: 0.799 G_ID: 0.092 G_Rec: 0.333 D_GP: 0.061 D_real: 0.792 D_fake: 1.046 +(epoch: 448, iters: 5230, time: 0.064) G_GAN: 0.203 G_GAN_Feat: 1.029 G_ID: 0.133 G_Rec: 0.434 D_GP: 0.180 D_real: 0.273 D_fake: 0.800 +(epoch: 448, iters: 5630, time: 0.063) G_GAN: 0.113 G_GAN_Feat: 0.747 G_ID: 0.090 G_Rec: 0.296 D_GP: 0.033 D_real: 0.792 D_fake: 0.888 +(epoch: 448, iters: 6030, time: 0.063) G_GAN: 0.386 G_GAN_Feat: 1.100 G_ID: 0.131 G_Rec: 0.485 D_GP: 0.043 D_real: 0.940 D_fake: 0.619 +(epoch: 448, iters: 6430, time: 0.063) G_GAN: 0.388 G_GAN_Feat: 0.757 G_ID: 0.096 G_Rec: 0.315 D_GP: 0.030 D_real: 1.162 D_fake: 0.614 +(epoch: 448, iters: 6830, time: 0.064) G_GAN: 0.956 G_GAN_Feat: 1.229 G_ID: 0.121 G_Rec: 0.461 D_GP: 0.071 D_real: 0.469 D_fake: 0.193 +(epoch: 448, iters: 7230, time: 0.063) G_GAN: 0.158 G_GAN_Feat: 0.747 G_ID: 0.090 G_Rec: 0.329 D_GP: 0.029 D_real: 1.080 D_fake: 0.842 +(epoch: 448, iters: 7630, time: 0.063) G_GAN: 0.050 G_GAN_Feat: 0.968 G_ID: 0.115 G_Rec: 0.429 D_GP: 0.036 D_real: 0.755 D_fake: 0.950 +(epoch: 448, iters: 8030, time: 0.063) G_GAN: -0.038 G_GAN_Feat: 0.858 G_ID: 0.110 G_Rec: 0.316 D_GP: 0.045 D_real: 0.657 D_fake: 1.038 +(epoch: 448, iters: 8430, time: 0.064) G_GAN: 0.674 G_GAN_Feat: 0.941 G_ID: 0.135 G_Rec: 0.354 D_GP: 0.041 D_real: 1.109 D_fake: 0.343 +(epoch: 449, iters: 222, time: 0.064) G_GAN: 0.652 G_GAN_Feat: 0.769 G_ID: 0.097 G_Rec: 0.328 D_GP: 0.045 D_real: 1.480 D_fake: 0.379 +(epoch: 449, iters: 622, time: 0.064) G_GAN: 0.516 G_GAN_Feat: 0.947 G_ID: 0.118 G_Rec: 0.439 D_GP: 0.033 D_real: 1.008 D_fake: 0.499 +(epoch: 449, iters: 1022, time: 0.064) G_GAN: 0.301 G_GAN_Feat: 0.730 G_ID: 0.116 G_Rec: 0.295 D_GP: 0.033 D_real: 1.142 D_fake: 0.699 +(epoch: 449, iters: 1422, time: 0.064) G_GAN: 0.771 G_GAN_Feat: 1.042 G_ID: 0.119 G_Rec: 0.461 D_GP: 0.048 D_real: 1.159 D_fake: 0.274 +(epoch: 449, iters: 1822, time: 0.063) G_GAN: 0.289 G_GAN_Feat: 0.806 G_ID: 0.105 G_Rec: 0.322 D_GP: 0.067 D_real: 0.844 D_fake: 0.716 +(epoch: 449, iters: 2222, time: 0.063) G_GAN: 0.686 G_GAN_Feat: 1.162 G_ID: 0.123 G_Rec: 0.435 D_GP: 0.748 D_real: 0.494 D_fake: 0.343 +(epoch: 449, iters: 2622, time: 0.063) G_GAN: 0.168 G_GAN_Feat: 0.943 G_ID: 0.096 G_Rec: 0.339 D_GP: 0.766 D_real: 0.371 D_fake: 0.838 +(epoch: 449, iters: 3022, time: 0.064) G_GAN: 0.605 G_GAN_Feat: 1.038 G_ID: 0.145 G_Rec: 0.445 D_GP: 0.035 D_real: 0.957 D_fake: 0.427 +(epoch: 449, iters: 3422, time: 0.063) G_GAN: 0.429 G_GAN_Feat: 0.733 G_ID: 0.106 G_Rec: 0.268 D_GP: 0.035 D_real: 1.241 D_fake: 0.571 +(epoch: 449, iters: 3822, time: 0.063) G_GAN: 0.192 G_GAN_Feat: 0.930 G_ID: 0.131 G_Rec: 0.483 D_GP: 0.031 D_real: 0.835 D_fake: 0.810 +(epoch: 449, iters: 4222, time: 0.063) G_GAN: -0.063 G_GAN_Feat: 0.752 G_ID: 0.090 G_Rec: 0.297 D_GP: 0.055 D_real: 0.632 D_fake: 1.063 +(epoch: 449, iters: 4622, time: 0.064) G_GAN: 0.351 G_GAN_Feat: 0.955 G_ID: 0.138 G_Rec: 0.417 D_GP: 0.038 D_real: 0.948 D_fake: 0.651 +(epoch: 449, iters: 5022, time: 0.063) G_GAN: 0.054 G_GAN_Feat: 0.879 G_ID: 0.107 G_Rec: 0.336 D_GP: 0.130 D_real: 0.340 D_fake: 0.947 +(epoch: 449, iters: 5422, time: 0.063) G_GAN: 0.565 G_GAN_Feat: 1.059 G_ID: 0.110 G_Rec: 0.487 D_GP: 0.044 D_real: 0.784 D_fake: 0.445 +(epoch: 449, iters: 5822, time: 0.063) G_GAN: -0.012 G_GAN_Feat: 0.763 G_ID: 0.111 G_Rec: 0.287 D_GP: 0.034 D_real: 0.789 D_fake: 1.014 +(epoch: 449, iters: 6222, time: 0.064) G_GAN: 0.638 G_GAN_Feat: 0.947 G_ID: 0.123 G_Rec: 0.449 D_GP: 0.042 D_real: 1.353 D_fake: 0.392 +(epoch: 449, iters: 6622, time: 0.063) G_GAN: 0.179 G_GAN_Feat: 0.682 G_ID: 0.099 G_Rec: 0.275 D_GP: 0.028 D_real: 1.141 D_fake: 0.821 +(epoch: 449, iters: 7022, time: 0.063) G_GAN: 0.325 G_GAN_Feat: 0.852 G_ID: 0.117 G_Rec: 0.359 D_GP: 0.033 D_real: 1.079 D_fake: 0.676 +(epoch: 449, iters: 7422, time: 0.063) G_GAN: 0.218 G_GAN_Feat: 0.738 G_ID: 0.093 G_Rec: 0.304 D_GP: 0.031 D_real: 1.124 D_fake: 0.783 +(epoch: 449, iters: 7822, time: 0.064) G_GAN: 0.348 G_GAN_Feat: 1.022 G_ID: 0.131 G_Rec: 0.465 D_GP: 0.067 D_real: 0.752 D_fake: 0.658 +(epoch: 449, iters: 8222, time: 0.063) G_GAN: 0.242 G_GAN_Feat: 0.866 G_ID: 0.113 G_Rec: 0.377 D_GP: 0.223 D_real: 0.752 D_fake: 0.765 +(epoch: 449, iters: 8622, time: 0.063) G_GAN: 0.498 G_GAN_Feat: 0.936 G_ID: 0.130 G_Rec: 0.425 D_GP: 0.033 D_real: 1.152 D_fake: 0.507 +(epoch: 450, iters: 414, time: 0.063) G_GAN: -0.158 G_GAN_Feat: 0.748 G_ID: 0.116 G_Rec: 0.315 D_GP: 0.038 D_real: 0.656 D_fake: 1.158 +(epoch: 450, iters: 814, time: 0.064) G_GAN: 0.616 G_GAN_Feat: 0.930 G_ID: 0.129 G_Rec: 0.445 D_GP: 0.033 D_real: 1.230 D_fake: 0.417 +(epoch: 450, iters: 1214, time: 0.063) G_GAN: 0.158 G_GAN_Feat: 0.851 G_ID: 0.134 G_Rec: 0.330 D_GP: 0.053 D_real: 0.673 D_fake: 0.842 +(epoch: 450, iters: 1614, time: 0.063) G_GAN: 0.694 G_GAN_Feat: 1.133 G_ID: 0.103 G_Rec: 0.454 D_GP: 0.079 D_real: 0.648 D_fake: 0.329 +(epoch: 450, iters: 2014, time: 0.063) G_GAN: 0.423 G_GAN_Feat: 0.873 G_ID: 0.098 G_Rec: 0.328 D_GP: 0.036 D_real: 0.937 D_fake: 0.581 +(epoch: 450, iters: 2414, time: 0.064) G_GAN: 0.716 G_GAN_Feat: 0.886 G_ID: 0.110 G_Rec: 0.394 D_GP: 0.080 D_real: 1.279 D_fake: 0.397 +(epoch: 450, iters: 2814, time: 0.063) G_GAN: 0.068 G_GAN_Feat: 0.742 G_ID: 0.091 G_Rec: 0.306 D_GP: 0.032 D_real: 0.938 D_fake: 0.932 +(epoch: 450, iters: 3214, time: 0.063) G_GAN: 0.392 G_GAN_Feat: 0.967 G_ID: 0.156 G_Rec: 0.421 D_GP: 0.037 D_real: 0.964 D_fake: 0.610 +(epoch: 450, iters: 3614, time: 0.063) G_GAN: 0.256 G_GAN_Feat: 0.720 G_ID: 0.103 G_Rec: 0.288 D_GP: 0.033 D_real: 1.177 D_fake: 0.745 +(epoch: 450, iters: 4014, time: 0.064) G_GAN: 0.544 G_GAN_Feat: 0.978 G_ID: 0.111 G_Rec: 0.449 D_GP: 0.036 D_real: 1.233 D_fake: 0.460 +(epoch: 450, iters: 4414, time: 0.063) G_GAN: 0.430 G_GAN_Feat: 0.716 G_ID: 0.079 G_Rec: 0.272 D_GP: 0.025 D_real: 1.304 D_fake: 0.571 +(epoch: 450, iters: 4814, time: 0.063) G_GAN: 0.833 G_GAN_Feat: 1.063 G_ID: 0.128 G_Rec: 0.403 D_GP: 0.048 D_real: 1.121 D_fake: 0.203 +(epoch: 450, iters: 5214, time: 0.063) G_GAN: 0.312 G_GAN_Feat: 0.900 G_ID: 0.091 G_Rec: 0.327 D_GP: 0.040 D_real: 0.600 D_fake: 0.688 +(epoch: 450, iters: 5614, time: 0.064) G_GAN: 0.812 G_GAN_Feat: 0.911 G_ID: 0.109 G_Rec: 0.419 D_GP: 0.027 D_real: 1.513 D_fake: 0.279 +(epoch: 450, iters: 6014, time: 0.064) G_GAN: 0.351 G_GAN_Feat: 0.788 G_ID: 0.087 G_Rec: 0.324 D_GP: 0.039 D_real: 1.079 D_fake: 0.651 +(epoch: 450, iters: 6414, time: 0.063) G_GAN: 0.649 G_GAN_Feat: 1.042 G_ID: 0.109 G_Rec: 0.460 D_GP: 0.044 D_real: 1.218 D_fake: 0.378 +(epoch: 450, iters: 6814, time: 0.063) G_GAN: 0.420 G_GAN_Feat: 0.689 G_ID: 0.092 G_Rec: 0.281 D_GP: 0.027 D_real: 1.272 D_fake: 0.581 +(epoch: 450, iters: 7214, time: 0.064) G_GAN: 0.712 G_GAN_Feat: 0.958 G_ID: 0.127 G_Rec: 0.388 D_GP: 0.069 D_real: 1.002 D_fake: 0.306 +(epoch: 450, iters: 7614, time: 0.063) G_GAN: 0.281 G_GAN_Feat: 0.785 G_ID: 0.096 G_Rec: 0.300 D_GP: 0.030 D_real: 0.946 D_fake: 0.719 +(epoch: 450, iters: 8014, time: 0.063) G_GAN: 0.675 G_GAN_Feat: 1.119 G_ID: 0.123 G_Rec: 0.420 D_GP: 0.059 D_real: 0.604 D_fake: 0.356 +(epoch: 450, iters: 8414, time: 0.063) G_GAN: 0.547 G_GAN_Feat: 0.956 G_ID: 0.101 G_Rec: 0.337 D_GP: 0.091 D_real: 0.629 D_fake: 0.522 +(epoch: 451, iters: 206, time: 0.064) G_GAN: 0.356 G_GAN_Feat: 0.867 G_ID: 0.131 G_Rec: 0.411 D_GP: 0.031 D_real: 1.075 D_fake: 0.647 +(epoch: 451, iters: 606, time: 0.063) G_GAN: 0.222 G_GAN_Feat: 0.659 G_ID: 0.096 G_Rec: 0.280 D_GP: 0.026 D_real: 1.208 D_fake: 0.778 +(epoch: 451, iters: 1006, time: 0.063) G_GAN: 0.358 G_GAN_Feat: 0.981 G_ID: 0.113 G_Rec: 0.443 D_GP: 0.037 D_real: 0.979 D_fake: 0.649 +(epoch: 451, iters: 1406, time: 0.063) G_GAN: -0.209 G_GAN_Feat: 0.712 G_ID: 0.118 G_Rec: 0.359 D_GP: 0.032 D_real: 0.731 D_fake: 1.209 +(epoch: 451, iters: 1806, time: 0.064) G_GAN: 0.130 G_GAN_Feat: 0.884 G_ID: 0.122 G_Rec: 0.396 D_GP: 0.043 D_real: 0.769 D_fake: 0.871 +(epoch: 451, iters: 2206, time: 0.063) G_GAN: -0.086 G_GAN_Feat: 0.755 G_ID: 0.121 G_Rec: 0.321 D_GP: 0.042 D_real: 0.565 D_fake: 1.086 +(epoch: 451, iters: 2606, time: 0.063) G_GAN: 0.088 G_GAN_Feat: 0.970 G_ID: 0.117 G_Rec: 0.430 D_GP: 0.053 D_real: 0.581 D_fake: 0.912 +(epoch: 451, iters: 3006, time: 0.063) G_GAN: 0.248 G_GAN_Feat: 0.791 G_ID: 0.082 G_Rec: 0.300 D_GP: 0.080 D_real: 0.794 D_fake: 0.757 +(epoch: 451, iters: 3406, time: 0.064) G_GAN: 0.364 G_GAN_Feat: 0.997 G_ID: 0.113 G_Rec: 0.452 D_GP: 0.035 D_real: 0.984 D_fake: 0.637 +(epoch: 451, iters: 3806, time: 0.063) G_GAN: 0.264 G_GAN_Feat: 0.723 G_ID: 0.098 G_Rec: 0.317 D_GP: 0.038 D_real: 1.151 D_fake: 0.736 +(epoch: 451, iters: 4206, time: 0.063) G_GAN: 0.463 G_GAN_Feat: 0.956 G_ID: 0.117 G_Rec: 0.417 D_GP: 0.054 D_real: 0.938 D_fake: 0.546 +(epoch: 451, iters: 4606, time: 0.064) G_GAN: 0.255 G_GAN_Feat: 0.763 G_ID: 0.089 G_Rec: 0.308 D_GP: 0.064 D_real: 0.858 D_fake: 0.747 +(epoch: 451, iters: 5006, time: 0.064) G_GAN: 0.500 G_GAN_Feat: 1.044 G_ID: 0.112 G_Rec: 0.455 D_GP: 0.097 D_real: 0.599 D_fake: 0.508 +(epoch: 451, iters: 5406, time: 0.063) G_GAN: 0.298 G_GAN_Feat: 0.988 G_ID: 0.113 G_Rec: 0.388 D_GP: 0.149 D_real: 0.320 D_fake: 0.723 +(epoch: 451, iters: 5806, time: 0.063) G_GAN: 0.432 G_GAN_Feat: 1.177 G_ID: 0.115 G_Rec: 0.471 D_GP: 0.080 D_real: 0.422 D_fake: 0.568 +(epoch: 451, iters: 6206, time: 0.063) G_GAN: 0.072 G_GAN_Feat: 0.942 G_ID: 0.126 G_Rec: 0.347 D_GP: 0.141 D_real: 0.234 D_fake: 0.932 +(epoch: 451, iters: 6606, time: 0.064) G_GAN: 0.281 G_GAN_Feat: 1.001 G_ID: 0.147 G_Rec: 0.453 D_GP: 0.052 D_real: 0.673 D_fake: 0.724 +(epoch: 451, iters: 7006, time: 0.063) G_GAN: 0.078 G_GAN_Feat: 0.776 G_ID: 0.107 G_Rec: 0.287 D_GP: 0.065 D_real: 0.748 D_fake: 0.924 +(epoch: 451, iters: 7406, time: 0.063) G_GAN: 0.609 G_GAN_Feat: 1.211 G_ID: 0.109 G_Rec: 0.473 D_GP: 0.049 D_real: 0.603 D_fake: 0.407 +(epoch: 451, iters: 7806, time: 0.063) G_GAN: 0.249 G_GAN_Feat: 0.723 G_ID: 0.090 G_Rec: 0.291 D_GP: 0.029 D_real: 1.132 D_fake: 0.751 +(epoch: 451, iters: 8206, time: 0.064) G_GAN: 0.371 G_GAN_Feat: 1.003 G_ID: 0.129 G_Rec: 0.440 D_GP: 0.047 D_real: 0.866 D_fake: 0.631 +(epoch: 451, iters: 8606, time: 0.063) G_GAN: 0.166 G_GAN_Feat: 0.902 G_ID: 0.093 G_Rec: 0.358 D_GP: 0.186 D_real: 0.555 D_fake: 0.841 +(epoch: 452, iters: 398, time: 0.063) G_GAN: 0.602 G_GAN_Feat: 1.172 G_ID: 0.126 G_Rec: 0.459 D_GP: 0.073 D_real: 0.292 D_fake: 0.405 +(epoch: 452, iters: 798, time: 0.063) G_GAN: 0.479 G_GAN_Feat: 0.907 G_ID: 0.092 G_Rec: 0.324 D_GP: 0.067 D_real: 0.709 D_fake: 0.525 +(epoch: 452, iters: 1198, time: 0.064) G_GAN: 0.450 G_GAN_Feat: 1.089 G_ID: 0.129 G_Rec: 0.465 D_GP: 0.040 D_real: 0.608 D_fake: 0.552 +(epoch: 452, iters: 1598, time: 0.063) G_GAN: 0.190 G_GAN_Feat: 0.735 G_ID: 0.090 G_Rec: 0.290 D_GP: 0.027 D_real: 1.056 D_fake: 0.810 +(epoch: 452, iters: 1998, time: 0.064) G_GAN: 0.679 G_GAN_Feat: 1.083 G_ID: 0.120 G_Rec: 0.455 D_GP: 0.048 D_real: 0.846 D_fake: 0.354 +(epoch: 452, iters: 2398, time: 0.064) G_GAN: 0.364 G_GAN_Feat: 1.104 G_ID: 0.095 G_Rec: 0.352 D_GP: 0.224 D_real: 0.415 D_fake: 0.645 +(epoch: 452, iters: 2798, time: 0.064) G_GAN: 1.045 G_GAN_Feat: 1.166 G_ID: 0.110 G_Rec: 0.442 D_GP: 0.091 D_real: 0.887 D_fake: 0.111 +(epoch: 452, iters: 3198, time: 0.064) G_GAN: 0.092 G_GAN_Feat: 0.839 G_ID: 0.087 G_Rec: 0.293 D_GP: 0.044 D_real: 0.645 D_fake: 0.908 +(epoch: 452, iters: 3598, time: 0.064) G_GAN: 0.565 G_GAN_Feat: 1.020 G_ID: 0.119 G_Rec: 0.408 D_GP: 0.049 D_real: 0.658 D_fake: 0.446 +(epoch: 452, iters: 3998, time: 0.063) G_GAN: 0.492 G_GAN_Feat: 0.776 G_ID: 0.092 G_Rec: 0.283 D_GP: 0.041 D_real: 1.099 D_fake: 0.524 +(epoch: 452, iters: 4398, time: 0.064) G_GAN: 0.237 G_GAN_Feat: 0.926 G_ID: 0.113 G_Rec: 0.431 D_GP: 0.029 D_real: 0.896 D_fake: 0.766 +(epoch: 452, iters: 4798, time: 0.063) G_GAN: 0.169 G_GAN_Feat: 0.741 G_ID: 0.088 G_Rec: 0.302 D_GP: 0.099 D_real: 0.970 D_fake: 0.832 +(epoch: 452, iters: 5198, time: 0.063) G_GAN: 0.505 G_GAN_Feat: 1.003 G_ID: 0.129 G_Rec: 0.408 D_GP: 0.039 D_real: 0.927 D_fake: 0.503 +(epoch: 452, iters: 5598, time: 0.064) G_GAN: 0.357 G_GAN_Feat: 0.738 G_ID: 0.089 G_Rec: 0.310 D_GP: 0.033 D_real: 1.339 D_fake: 0.643 +(epoch: 452, iters: 5998, time: 0.064) G_GAN: 0.640 G_GAN_Feat: 1.038 G_ID: 0.116 G_Rec: 0.415 D_GP: 0.064 D_real: 0.529 D_fake: 0.375 +(epoch: 452, iters: 6398, time: 0.063) G_GAN: 0.081 G_GAN_Feat: 0.708 G_ID: 0.096 G_Rec: 0.326 D_GP: 0.034 D_real: 0.868 D_fake: 0.919 +(epoch: 452, iters: 6798, time: 0.063) G_GAN: 0.439 G_GAN_Feat: 1.017 G_ID: 0.125 G_Rec: 0.432 D_GP: 0.085 D_real: 0.741 D_fake: 0.584 +(epoch: 452, iters: 7198, time: 0.063) G_GAN: -0.148 G_GAN_Feat: 0.798 G_ID: 0.095 G_Rec: 0.326 D_GP: 0.037 D_real: 0.724 D_fake: 1.148 +(epoch: 452, iters: 7598, time: 0.064) G_GAN: 0.196 G_GAN_Feat: 0.943 G_ID: 0.135 G_Rec: 0.414 D_GP: 0.046 D_real: 0.796 D_fake: 0.809 +(epoch: 452, iters: 7998, time: 0.063) G_GAN: -0.047 G_GAN_Feat: 0.732 G_ID: 0.081 G_Rec: 0.291 D_GP: 0.055 D_real: 0.840 D_fake: 1.047 +(epoch: 452, iters: 8398, time: 0.063) G_GAN: 0.517 G_GAN_Feat: 1.036 G_ID: 0.116 G_Rec: 0.458 D_GP: 0.045 D_real: 0.936 D_fake: 0.498 +(epoch: 453, iters: 190, time: 0.063) G_GAN: 0.296 G_GAN_Feat: 0.772 G_ID: 0.096 G_Rec: 0.301 D_GP: 0.038 D_real: 1.136 D_fake: 0.705 +(epoch: 453, iters: 590, time: 0.063) G_GAN: 0.379 G_GAN_Feat: 0.931 G_ID: 0.116 G_Rec: 0.424 D_GP: 0.033 D_real: 1.096 D_fake: 0.621 +(epoch: 453, iters: 990, time: 0.063) G_GAN: 0.298 G_GAN_Feat: 0.858 G_ID: 0.090 G_Rec: 0.335 D_GP: 0.059 D_real: 0.833 D_fake: 0.702 +(epoch: 453, iters: 1390, time: 0.063) G_GAN: 0.452 G_GAN_Feat: 1.078 G_ID: 0.098 G_Rec: 0.470 D_GP: 0.051 D_real: 0.769 D_fake: 0.551 +(epoch: 453, iters: 1790, time: 0.064) G_GAN: 0.283 G_GAN_Feat: 0.894 G_ID: 0.107 G_Rec: 0.334 D_GP: 0.037 D_real: 0.750 D_fake: 0.717 +(epoch: 453, iters: 2190, time: 0.063) G_GAN: 0.861 G_GAN_Feat: 1.103 G_ID: 0.122 G_Rec: 0.478 D_GP: 0.076 D_real: 0.714 D_fake: 0.233 +(epoch: 453, iters: 2590, time: 0.063) G_GAN: 0.131 G_GAN_Feat: 0.720 G_ID: 0.084 G_Rec: 0.306 D_GP: 0.032 D_real: 1.057 D_fake: 0.869 +(epoch: 453, iters: 2990, time: 0.063) G_GAN: 0.432 G_GAN_Feat: 0.929 G_ID: 0.124 G_Rec: 0.449 D_GP: 0.041 D_real: 1.066 D_fake: 0.587 +(epoch: 453, iters: 3390, time: 0.064) G_GAN: -0.082 G_GAN_Feat: 0.625 G_ID: 0.100 G_Rec: 0.276 D_GP: 0.032 D_real: 0.764 D_fake: 1.082 +(epoch: 453, iters: 3790, time: 0.063) G_GAN: 0.197 G_GAN_Feat: 0.922 G_ID: 0.110 G_Rec: 0.412 D_GP: 0.034 D_real: 0.823 D_fake: 0.804 +(epoch: 453, iters: 4190, time: 0.063) G_GAN: 0.038 G_GAN_Feat: 0.755 G_ID: 0.103 G_Rec: 0.318 D_GP: 0.052 D_real: 0.818 D_fake: 0.962 +(epoch: 453, iters: 4590, time: 0.063) G_GAN: 0.228 G_GAN_Feat: 1.014 G_ID: 0.115 G_Rec: 0.478 D_GP: 0.043 D_real: 0.722 D_fake: 0.774 +(epoch: 453, iters: 4990, time: 0.064) G_GAN: 0.277 G_GAN_Feat: 0.714 G_ID: 0.080 G_Rec: 0.292 D_GP: 0.037 D_real: 1.091 D_fake: 0.724 +(epoch: 453, iters: 5390, time: 0.063) G_GAN: 0.237 G_GAN_Feat: 0.907 G_ID: 0.121 G_Rec: 0.407 D_GP: 0.028 D_real: 1.087 D_fake: 0.763 +(epoch: 453, iters: 5790, time: 0.063) G_GAN: 0.182 G_GAN_Feat: 0.728 G_ID: 0.107 G_Rec: 0.281 D_GP: 0.033 D_real: 1.101 D_fake: 0.818 +(epoch: 453, iters: 6190, time: 0.063) G_GAN: 0.377 G_GAN_Feat: 0.995 G_ID: 0.125 G_Rec: 0.427 D_GP: 0.072 D_real: 0.793 D_fake: 0.625 +(epoch: 453, iters: 6590, time: 0.064) G_GAN: 0.266 G_GAN_Feat: 0.859 G_ID: 0.104 G_Rec: 0.330 D_GP: 0.064 D_real: 0.850 D_fake: 0.735 +(epoch: 453, iters: 6990, time: 0.063) G_GAN: 0.563 G_GAN_Feat: 0.970 G_ID: 0.120 G_Rec: 0.445 D_GP: 0.032 D_real: 1.151 D_fake: 0.448 +(epoch: 453, iters: 7390, time: 0.063) G_GAN: 0.157 G_GAN_Feat: 0.774 G_ID: 0.106 G_Rec: 0.317 D_GP: 0.033 D_real: 0.907 D_fake: 0.843 +(epoch: 453, iters: 7790, time: 0.063) G_GAN: 0.657 G_GAN_Feat: 1.036 G_ID: 0.121 G_Rec: 0.465 D_GP: 0.037 D_real: 1.157 D_fake: 0.355 +(epoch: 453, iters: 8190, time: 0.064) G_GAN: 0.255 G_GAN_Feat: 0.875 G_ID: 0.116 G_Rec: 0.350 D_GP: 0.075 D_real: 0.639 D_fake: 0.752 +(epoch: 453, iters: 8590, time: 0.063) G_GAN: 0.750 G_GAN_Feat: 1.182 G_ID: 0.122 G_Rec: 0.476 D_GP: 0.073 D_real: 0.651 D_fake: 0.274 +(epoch: 454, iters: 382, time: 0.063) G_GAN: -0.017 G_GAN_Feat: 0.883 G_ID: 0.110 G_Rec: 0.339 D_GP: 0.039 D_real: 0.375 D_fake: 1.017 +(epoch: 454, iters: 782, time: 0.063) G_GAN: 0.324 G_GAN_Feat: 0.948 G_ID: 0.124 G_Rec: 0.389 D_GP: 0.037 D_real: 0.900 D_fake: 0.677 +(epoch: 454, iters: 1182, time: 0.064) G_GAN: 0.349 G_GAN_Feat: 0.737 G_ID: 0.078 G_Rec: 0.326 D_GP: 0.029 D_real: 1.211 D_fake: 0.651 +(epoch: 454, iters: 1582, time: 0.063) G_GAN: 0.629 G_GAN_Feat: 0.850 G_ID: 0.110 G_Rec: 0.383 D_GP: 0.042 D_real: 1.296 D_fake: 0.403 +(epoch: 454, iters: 1982, time: 0.063) G_GAN: 0.090 G_GAN_Feat: 0.751 G_ID: 0.090 G_Rec: 0.330 D_GP: 0.036 D_real: 0.881 D_fake: 0.910 +(epoch: 454, iters: 2382, time: 0.063) G_GAN: 0.545 G_GAN_Feat: 1.022 G_ID: 0.110 G_Rec: 0.410 D_GP: 0.035 D_real: 1.092 D_fake: 0.464 +(epoch: 454, iters: 2782, time: 0.064) G_GAN: 0.106 G_GAN_Feat: 0.697 G_ID: 0.079 G_Rec: 0.293 D_GP: 0.032 D_real: 0.953 D_fake: 0.894 +(epoch: 454, iters: 3182, time: 0.063) G_GAN: 0.353 G_GAN_Feat: 1.042 G_ID: 0.111 G_Rec: 0.464 D_GP: 0.055 D_real: 0.930 D_fake: 0.649 +(epoch: 454, iters: 3582, time: 0.063) G_GAN: 0.527 G_GAN_Feat: 0.751 G_ID: 0.095 G_Rec: 0.330 D_GP: 0.037 D_real: 1.406 D_fake: 0.477 +(epoch: 454, iters: 3982, time: 0.063) G_GAN: 0.523 G_GAN_Feat: 0.894 G_ID: 0.132 G_Rec: 0.396 D_GP: 0.034 D_real: 1.143 D_fake: 0.487 +(epoch: 454, iters: 4382, time: 0.064) G_GAN: 0.360 G_GAN_Feat: 0.690 G_ID: 0.112 G_Rec: 0.341 D_GP: 0.031 D_real: 1.279 D_fake: 0.642 +(epoch: 454, iters: 4782, time: 0.063) G_GAN: 0.362 G_GAN_Feat: 1.059 G_ID: 0.115 G_Rec: 0.472 D_GP: 0.090 D_real: 0.698 D_fake: 0.647 +(epoch: 454, iters: 5182, time: 0.063) G_GAN: 0.344 G_GAN_Feat: 0.886 G_ID: 0.102 G_Rec: 0.299 D_GP: 0.072 D_real: 0.590 D_fake: 0.658 +(epoch: 454, iters: 5582, time: 0.063) G_GAN: 0.421 G_GAN_Feat: 1.151 G_ID: 0.125 G_Rec: 0.471 D_GP: 0.672 D_real: 0.344 D_fake: 0.590 +(epoch: 454, iters: 5982, time: 0.064) G_GAN: 0.369 G_GAN_Feat: 0.925 G_ID: 0.085 G_Rec: 0.321 D_GP: 0.132 D_real: 0.324 D_fake: 0.679 +(epoch: 454, iters: 6382, time: 0.063) G_GAN: 0.697 G_GAN_Feat: 1.006 G_ID: 0.119 G_Rec: 0.424 D_GP: 0.035 D_real: 1.197 D_fake: 0.316 +(epoch: 454, iters: 6782, time: 0.063) G_GAN: 0.118 G_GAN_Feat: 0.817 G_ID: 0.094 G_Rec: 0.330 D_GP: 0.044 D_real: 0.796 D_fake: 0.882 +(epoch: 454, iters: 7182, time: 0.063) G_GAN: 0.426 G_GAN_Feat: 0.936 G_ID: 0.126 G_Rec: 0.407 D_GP: 0.036 D_real: 1.076 D_fake: 0.582 +(epoch: 454, iters: 7582, time: 0.064) G_GAN: 0.154 G_GAN_Feat: 0.821 G_ID: 0.100 G_Rec: 0.333 D_GP: 0.075 D_real: 0.720 D_fake: 0.847 +(epoch: 454, iters: 7982, time: 0.063) G_GAN: 0.548 G_GAN_Feat: 1.156 G_ID: 0.119 G_Rec: 0.480 D_GP: 0.067 D_real: 0.442 D_fake: 0.466 +(epoch: 454, iters: 8382, time: 0.063) G_GAN: 0.097 G_GAN_Feat: 0.708 G_ID: 0.095 G_Rec: 0.294 D_GP: 0.027 D_real: 0.977 D_fake: 0.903 +(epoch: 455, iters: 174, time: 0.063) G_GAN: 0.863 G_GAN_Feat: 1.181 G_ID: 0.123 G_Rec: 0.457 D_GP: 0.091 D_real: 0.645 D_fake: 0.376 +(epoch: 455, iters: 574, time: 0.064) G_GAN: 0.145 G_GAN_Feat: 0.784 G_ID: 0.110 G_Rec: 0.319 D_GP: 0.035 D_real: 0.965 D_fake: 0.855 +(epoch: 455, iters: 974, time: 0.063) G_GAN: 0.609 G_GAN_Feat: 1.105 G_ID: 0.120 G_Rec: 0.447 D_GP: 0.051 D_real: 0.950 D_fake: 0.409 +(epoch: 455, iters: 1374, time: 0.063) G_GAN: 0.118 G_GAN_Feat: 0.754 G_ID: 0.093 G_Rec: 0.320 D_GP: 0.036 D_real: 0.996 D_fake: 0.882 +(epoch: 455, iters: 1774, time: 0.063) G_GAN: 0.689 G_GAN_Feat: 1.051 G_ID: 0.103 G_Rec: 0.489 D_GP: 0.036 D_real: 1.127 D_fake: 0.371 +(epoch: 455, iters: 2174, time: 0.064) G_GAN: 0.312 G_GAN_Feat: 0.833 G_ID: 0.088 G_Rec: 0.353 D_GP: 0.096 D_real: 0.808 D_fake: 0.690 +(epoch: 455, iters: 2574, time: 0.063) G_GAN: 0.431 G_GAN_Feat: 1.051 G_ID: 0.136 G_Rec: 0.475 D_GP: 0.050 D_real: 0.759 D_fake: 0.582 +(epoch: 455, iters: 2974, time: 0.063) G_GAN: 0.246 G_GAN_Feat: 1.074 G_ID: 0.095 G_Rec: 0.340 D_GP: 0.213 D_real: 0.301 D_fake: 0.756 +(epoch: 455, iters: 3374, time: 0.063) G_GAN: 1.050 G_GAN_Feat: 1.150 G_ID: 0.128 G_Rec: 0.406 D_GP: 0.095 D_real: 0.853 D_fake: 0.234 +(epoch: 455, iters: 3774, time: 0.064) G_GAN: 0.408 G_GAN_Feat: 0.869 G_ID: 0.101 G_Rec: 0.327 D_GP: 0.050 D_real: 0.801 D_fake: 0.621 +(epoch: 455, iters: 4174, time: 0.063) G_GAN: 0.664 G_GAN_Feat: 1.026 G_ID: 0.123 G_Rec: 0.479 D_GP: 0.035 D_real: 1.279 D_fake: 0.400 +(epoch: 455, iters: 4574, time: 0.063) G_GAN: -0.041 G_GAN_Feat: 0.749 G_ID: 0.106 G_Rec: 0.321 D_GP: 0.039 D_real: 0.749 D_fake: 1.041 +(epoch: 455, iters: 4974, time: 0.063) G_GAN: 0.718 G_GAN_Feat: 0.996 G_ID: 0.106 G_Rec: 0.422 D_GP: 0.054 D_real: 1.200 D_fake: 0.310 +(epoch: 455, iters: 5374, time: 0.064) G_GAN: 0.572 G_GAN_Feat: 0.809 G_ID: 0.099 G_Rec: 0.303 D_GP: 0.087 D_real: 1.164 D_fake: 0.437 +(epoch: 455, iters: 5774, time: 0.063) G_GAN: 0.816 G_GAN_Feat: 1.038 G_ID: 0.125 G_Rec: 0.427 D_GP: 0.043 D_real: 1.259 D_fake: 0.313 +(epoch: 455, iters: 6174, time: 0.063) G_GAN: 0.465 G_GAN_Feat: 0.772 G_ID: 0.102 G_Rec: 0.297 D_GP: 0.036 D_real: 1.273 D_fake: 0.544 +(epoch: 455, iters: 6574, time: 0.063) G_GAN: 0.454 G_GAN_Feat: 1.150 G_ID: 0.127 G_Rec: 0.468 D_GP: 0.066 D_real: 0.543 D_fake: 0.559 +(epoch: 455, iters: 6974, time: 0.064) G_GAN: 0.236 G_GAN_Feat: 0.778 G_ID: 0.100 G_Rec: 0.345 D_GP: 0.034 D_real: 1.052 D_fake: 0.764 +(epoch: 455, iters: 7374, time: 0.063) G_GAN: 0.141 G_GAN_Feat: 0.897 G_ID: 0.134 G_Rec: 0.440 D_GP: 0.041 D_real: 0.834 D_fake: 0.860 +(epoch: 455, iters: 7774, time: 0.063) G_GAN: 0.167 G_GAN_Feat: 0.723 G_ID: 0.092 G_Rec: 0.317 D_GP: 0.037 D_real: 1.072 D_fake: 0.834 +(epoch: 455, iters: 8174, time: 0.063) G_GAN: 0.369 G_GAN_Feat: 0.922 G_ID: 0.104 G_Rec: 0.433 D_GP: 0.069 D_real: 1.028 D_fake: 0.636 +(epoch: 455, iters: 8574, time: 0.064) G_GAN: -0.033 G_GAN_Feat: 0.754 G_ID: 0.118 G_Rec: 0.329 D_GP: 0.042 D_real: 0.747 D_fake: 1.033 +(epoch: 456, iters: 366, time: 0.063) G_GAN: 0.385 G_GAN_Feat: 0.931 G_ID: 0.116 G_Rec: 0.423 D_GP: 0.045 D_real: 0.900 D_fake: 0.621 +(epoch: 456, iters: 766, time: 0.063) G_GAN: 0.242 G_GAN_Feat: 0.645 G_ID: 0.107 G_Rec: 0.299 D_GP: 0.028 D_real: 1.126 D_fake: 0.759 +(epoch: 456, iters: 1166, time: 0.063) G_GAN: 0.479 G_GAN_Feat: 0.893 G_ID: 0.122 G_Rec: 0.402 D_GP: 0.048 D_real: 1.078 D_fake: 0.528 +(epoch: 456, iters: 1566, time: 0.064) G_GAN: 0.239 G_GAN_Feat: 0.769 G_ID: 0.091 G_Rec: 0.316 D_GP: 0.060 D_real: 1.007 D_fake: 0.764 +(epoch: 456, iters: 1966, time: 0.063) G_GAN: 0.512 G_GAN_Feat: 0.925 G_ID: 0.115 G_Rec: 0.402 D_GP: 0.028 D_real: 1.152 D_fake: 0.492 +(epoch: 456, iters: 2366, time: 0.064) G_GAN: 0.354 G_GAN_Feat: 0.930 G_ID: 0.087 G_Rec: 0.347 D_GP: 0.162 D_real: 0.510 D_fake: 0.660 +(epoch: 456, iters: 2766, time: 0.064) G_GAN: 0.141 G_GAN_Feat: 1.048 G_ID: 0.135 G_Rec: 0.461 D_GP: 0.036 D_real: 0.868 D_fake: 0.861 +(epoch: 456, iters: 3166, time: 0.064) G_GAN: -0.067 G_GAN_Feat: 0.737 G_ID: 0.124 G_Rec: 0.342 D_GP: 0.029 D_real: 0.840 D_fake: 1.067 +(epoch: 456, iters: 3566, time: 0.064) G_GAN: 0.170 G_GAN_Feat: 0.890 G_ID: 0.122 G_Rec: 0.419 D_GP: 0.033 D_real: 0.838 D_fake: 0.835 +(epoch: 456, iters: 3966, time: 0.063) G_GAN: -0.153 G_GAN_Feat: 0.728 G_ID: 0.102 G_Rec: 0.327 D_GP: 0.035 D_real: 0.704 D_fake: 1.153 +(epoch: 456, iters: 4366, time: 0.064) G_GAN: 0.296 G_GAN_Feat: 0.937 G_ID: 0.123 G_Rec: 0.426 D_GP: 0.040 D_real: 0.889 D_fake: 0.708 +(epoch: 456, iters: 4766, time: 0.064) G_GAN: 0.130 G_GAN_Feat: 0.641 G_ID: 0.097 G_Rec: 0.269 D_GP: 0.035 D_real: 1.053 D_fake: 0.872 +(epoch: 456, iters: 5166, time: 0.064) G_GAN: 0.198 G_GAN_Feat: 0.991 G_ID: 0.113 G_Rec: 0.459 D_GP: 0.060 D_real: 0.566 D_fake: 0.808 +(epoch: 456, iters: 5566, time: 0.064) G_GAN: -0.027 G_GAN_Feat: 0.670 G_ID: 0.107 G_Rec: 0.287 D_GP: 0.041 D_real: 0.813 D_fake: 1.030 +(epoch: 456, iters: 5966, time: 0.064) G_GAN: 0.361 G_GAN_Feat: 1.002 G_ID: 0.118 G_Rec: 0.459 D_GP: 0.052 D_real: 0.750 D_fake: 0.648 +(epoch: 456, iters: 6366, time: 0.064) G_GAN: -0.178 G_GAN_Feat: 0.789 G_ID: 0.105 G_Rec: 0.347 D_GP: 0.070 D_real: 0.464 D_fake: 1.178 +(epoch: 456, iters: 6766, time: 0.064) G_GAN: 0.456 G_GAN_Feat: 1.039 G_ID: 0.120 G_Rec: 0.415 D_GP: 0.063 D_real: 0.743 D_fake: 0.549 +(epoch: 456, iters: 7166, time: 0.064) G_GAN: 0.246 G_GAN_Feat: 0.827 G_ID: 0.085 G_Rec: 0.331 D_GP: 0.047 D_real: 0.866 D_fake: 0.758 +(epoch: 456, iters: 7566, time: 0.064) G_GAN: 0.609 G_GAN_Feat: 1.169 G_ID: 0.138 G_Rec: 0.479 D_GP: 0.073 D_real: 0.324 D_fake: 0.442 +(epoch: 456, iters: 7966, time: 0.064) G_GAN: 0.280 G_GAN_Feat: 0.762 G_ID: 0.088 G_Rec: 0.304 D_GP: 0.036 D_real: 1.181 D_fake: 0.725 +(epoch: 456, iters: 8366, time: 0.064) G_GAN: 0.854 G_GAN_Feat: 1.106 G_ID: 0.115 G_Rec: 0.503 D_GP: 0.055 D_real: 1.121 D_fake: 0.221 +(epoch: 457, iters: 158, time: 0.064) G_GAN: 0.499 G_GAN_Feat: 0.947 G_ID: 0.099 G_Rec: 0.333 D_GP: 0.046 D_real: 0.573 D_fake: 0.520 +(epoch: 457, iters: 558, time: 0.064) G_GAN: 0.251 G_GAN_Feat: 0.904 G_ID: 0.104 G_Rec: 0.449 D_GP: 0.030 D_real: 1.006 D_fake: 0.751 +(epoch: 457, iters: 958, time: 0.064) G_GAN: 0.077 G_GAN_Feat: 0.680 G_ID: 0.098 G_Rec: 0.300 D_GP: 0.029 D_real: 0.943 D_fake: 0.923 +(epoch: 457, iters: 1358, time: 0.063) G_GAN: 0.477 G_GAN_Feat: 0.965 G_ID: 0.122 G_Rec: 0.417 D_GP: 0.039 D_real: 1.033 D_fake: 0.526 +(epoch: 457, iters: 1758, time: 0.063) G_GAN: 0.095 G_GAN_Feat: 0.656 G_ID: 0.093 G_Rec: 0.267 D_GP: 0.038 D_real: 0.967 D_fake: 0.905 +(epoch: 457, iters: 2158, time: 0.063) G_GAN: 0.476 G_GAN_Feat: 0.987 G_ID: 0.112 G_Rec: 0.444 D_GP: 0.037 D_real: 1.024 D_fake: 0.526 +(epoch: 457, iters: 2558, time: 0.064) G_GAN: 0.111 G_GAN_Feat: 0.775 G_ID: 0.099 G_Rec: 0.279 D_GP: 0.066 D_real: 0.730 D_fake: 0.889 +(epoch: 457, iters: 2958, time: 0.063) G_GAN: 0.647 G_GAN_Feat: 1.095 G_ID: 0.113 G_Rec: 0.470 D_GP: 0.048 D_real: 0.950 D_fake: 0.482 +(epoch: 457, iters: 3358, time: 0.063) G_GAN: 0.339 G_GAN_Feat: 0.671 G_ID: 0.124 G_Rec: 0.335 D_GP: 0.044 D_real: 1.275 D_fake: 0.663 +(epoch: 457, iters: 3758, time: 0.063) G_GAN: 0.479 G_GAN_Feat: 0.809 G_ID: 0.112 G_Rec: 0.391 D_GP: 0.029 D_real: 1.123 D_fake: 0.531 +(epoch: 457, iters: 4158, time: 0.064) G_GAN: 0.220 G_GAN_Feat: 0.714 G_ID: 0.107 G_Rec: 0.319 D_GP: 0.040 D_real: 1.097 D_fake: 0.781 +(epoch: 457, iters: 4558, time: 0.063) G_GAN: 0.359 G_GAN_Feat: 0.822 G_ID: 0.117 G_Rec: 0.398 D_GP: 0.029 D_real: 1.077 D_fake: 0.653 +(epoch: 457, iters: 4958, time: 0.063) G_GAN: -0.047 G_GAN_Feat: 0.658 G_ID: 0.096 G_Rec: 0.318 D_GP: 0.034 D_real: 0.871 D_fake: 1.047 +(epoch: 457, iters: 5358, time: 0.063) G_GAN: 0.081 G_GAN_Feat: 0.856 G_ID: 0.114 G_Rec: 0.426 D_GP: 0.033 D_real: 0.738 D_fake: 0.919 +(epoch: 457, iters: 5758, time: 0.063) G_GAN: 0.073 G_GAN_Feat: 0.686 G_ID: 0.104 G_Rec: 0.331 D_GP: 0.040 D_real: 0.937 D_fake: 0.927 +(epoch: 457, iters: 6158, time: 0.063) G_GAN: 0.311 G_GAN_Feat: 0.846 G_ID: 0.100 G_Rec: 0.413 D_GP: 0.037 D_real: 1.017 D_fake: 0.704 +(epoch: 457, iters: 6558, time: 0.063) G_GAN: 0.022 G_GAN_Feat: 0.650 G_ID: 0.090 G_Rec: 0.294 D_GP: 0.046 D_real: 0.926 D_fake: 0.979 +(epoch: 457, iters: 6958, time: 0.064) G_GAN: 0.158 G_GAN_Feat: 0.869 G_ID: 0.110 G_Rec: 0.409 D_GP: 0.037 D_real: 0.820 D_fake: 0.844 +(epoch: 457, iters: 7358, time: 0.063) G_GAN: 0.157 G_GAN_Feat: 0.674 G_ID: 0.091 G_Rec: 0.340 D_GP: 0.035 D_real: 0.927 D_fake: 0.843 +(epoch: 457, iters: 7758, time: 0.063) G_GAN: 0.348 G_GAN_Feat: 0.911 G_ID: 0.110 G_Rec: 0.476 D_GP: 0.036 D_real: 0.974 D_fake: 0.658 +(epoch: 457, iters: 8158, time: 0.063) G_GAN: 0.068 G_GAN_Feat: 0.639 G_ID: 0.102 G_Rec: 0.275 D_GP: 0.036 D_real: 0.946 D_fake: 0.932 +(epoch: 457, iters: 8558, time: 0.064) G_GAN: 0.455 G_GAN_Feat: 1.028 G_ID: 0.113 G_Rec: 0.471 D_GP: 0.098 D_real: 0.804 D_fake: 0.568 +(epoch: 458, iters: 350, time: 0.063) G_GAN: 0.230 G_GAN_Feat: 0.670 G_ID: 0.084 G_Rec: 0.287 D_GP: 0.042 D_real: 1.092 D_fake: 0.774 +(epoch: 458, iters: 750, time: 0.063) G_GAN: 0.168 G_GAN_Feat: 0.983 G_ID: 0.120 G_Rec: 0.467 D_GP: 0.060 D_real: 0.685 D_fake: 0.835 +(epoch: 458, iters: 1150, time: 0.063) G_GAN: 0.274 G_GAN_Feat: 0.748 G_ID: 0.107 G_Rec: 0.307 D_GP: 0.084 D_real: 0.882 D_fake: 0.729 +(epoch: 458, iters: 1550, time: 0.064) G_GAN: 0.294 G_GAN_Feat: 0.882 G_ID: 0.145 G_Rec: 0.418 D_GP: 0.044 D_real: 0.884 D_fake: 0.706 +(epoch: 458, iters: 1950, time: 0.063) G_GAN: 0.148 G_GAN_Feat: 0.734 G_ID: 0.084 G_Rec: 0.309 D_GP: 0.040 D_real: 0.940 D_fake: 0.852 +(epoch: 458, iters: 2350, time: 0.063) G_GAN: 0.276 G_GAN_Feat: 0.915 G_ID: 0.127 G_Rec: 0.417 D_GP: 0.087 D_real: 0.818 D_fake: 0.724 +(epoch: 458, iters: 2750, time: 0.063) G_GAN: 0.389 G_GAN_Feat: 0.765 G_ID: 0.099 G_Rec: 0.322 D_GP: 0.039 D_real: 1.165 D_fake: 0.626 +(epoch: 458, iters: 3150, time: 0.064) G_GAN: 0.452 G_GAN_Feat: 0.945 G_ID: 0.139 G_Rec: 0.403 D_GP: 0.042 D_real: 0.988 D_fake: 0.551 +(epoch: 458, iters: 3550, time: 0.063) G_GAN: 0.121 G_GAN_Feat: 0.781 G_ID: 0.104 G_Rec: 0.332 D_GP: 0.051 D_real: 0.799 D_fake: 0.879 +(epoch: 458, iters: 3950, time: 0.063) G_GAN: 0.793 G_GAN_Feat: 0.967 G_ID: 0.111 G_Rec: 0.405 D_GP: 0.047 D_real: 1.485 D_fake: 0.252 +(epoch: 458, iters: 4350, time: 0.063) G_GAN: 0.019 G_GAN_Feat: 0.848 G_ID: 0.080 G_Rec: 0.346 D_GP: 0.092 D_real: 0.393 D_fake: 0.981 +(epoch: 458, iters: 4750, time: 0.064) G_GAN: 0.330 G_GAN_Feat: 0.970 G_ID: 0.116 G_Rec: 0.432 D_GP: 0.044 D_real: 0.788 D_fake: 0.671 +(epoch: 458, iters: 5150, time: 0.063) G_GAN: 0.421 G_GAN_Feat: 0.765 G_ID: 0.103 G_Rec: 0.300 D_GP: 0.032 D_real: 1.137 D_fake: 0.583 +(epoch: 458, iters: 5550, time: 0.063) G_GAN: 0.478 G_GAN_Feat: 0.969 G_ID: 0.128 G_Rec: 0.423 D_GP: 0.037 D_real: 1.105 D_fake: 0.527 +(epoch: 458, iters: 5950, time: 0.063) G_GAN: 0.031 G_GAN_Feat: 0.880 G_ID: 0.109 G_Rec: 0.335 D_GP: 0.084 D_real: 0.419 D_fake: 0.969 +(epoch: 458, iters: 6350, time: 0.064) G_GAN: 0.452 G_GAN_Feat: 0.937 G_ID: 0.096 G_Rec: 0.407 D_GP: 0.052 D_real: 0.967 D_fake: 0.550 +(epoch: 458, iters: 6750, time: 0.063) G_GAN: -0.005 G_GAN_Feat: 0.782 G_ID: 0.104 G_Rec: 0.324 D_GP: 0.042 D_real: 0.570 D_fake: 1.005 +(epoch: 458, iters: 7150, time: 0.063) G_GAN: 0.398 G_GAN_Feat: 0.889 G_ID: 0.113 G_Rec: 0.429 D_GP: 0.035 D_real: 1.120 D_fake: 0.604 +(epoch: 458, iters: 7550, time: 0.063) G_GAN: 0.258 G_GAN_Feat: 0.688 G_ID: 0.090 G_Rec: 0.325 D_GP: 0.055 D_real: 1.068 D_fake: 0.743 +(epoch: 458, iters: 7950, time: 0.064) G_GAN: 0.634 G_GAN_Feat: 1.037 G_ID: 0.114 G_Rec: 0.429 D_GP: 0.158 D_real: 0.790 D_fake: 0.406 +(epoch: 458, iters: 8350, time: 0.063) G_GAN: 0.177 G_GAN_Feat: 0.756 G_ID: 0.102 G_Rec: 0.308 D_GP: 0.053 D_real: 0.938 D_fake: 0.827 +(epoch: 459, iters: 142, time: 0.063) G_GAN: 0.084 G_GAN_Feat: 0.985 G_ID: 0.112 G_Rec: 0.461 D_GP: 0.036 D_real: 0.612 D_fake: 0.916 +(epoch: 459, iters: 542, time: 0.063) G_GAN: -0.042 G_GAN_Feat: 0.692 G_ID: 0.099 G_Rec: 0.289 D_GP: 0.054 D_real: 0.763 D_fake: 1.044 +(epoch: 459, iters: 942, time: 0.064) G_GAN: 0.446 G_GAN_Feat: 1.067 G_ID: 0.145 G_Rec: 0.457 D_GP: 0.054 D_real: 0.789 D_fake: 0.562 +(epoch: 459, iters: 1342, time: 0.063) G_GAN: -0.025 G_GAN_Feat: 0.892 G_ID: 0.121 G_Rec: 0.332 D_GP: 0.685 D_real: 0.179 D_fake: 1.031 +(epoch: 459, iters: 1742, time: 0.063) G_GAN: 0.356 G_GAN_Feat: 1.064 G_ID: 0.116 G_Rec: 0.447 D_GP: 0.082 D_real: 0.566 D_fake: 0.645 +(epoch: 459, iters: 2142, time: 0.063) G_GAN: 0.234 G_GAN_Feat: 0.766 G_ID: 0.083 G_Rec: 0.297 D_GP: 0.033 D_real: 1.021 D_fake: 0.766 +(epoch: 459, iters: 2542, time: 0.064) G_GAN: 0.611 G_GAN_Feat: 0.899 G_ID: 0.121 G_Rec: 0.394 D_GP: 0.045 D_real: 1.255 D_fake: 0.533 +(epoch: 459, iters: 2942, time: 0.063) G_GAN: 0.272 G_GAN_Feat: 0.719 G_ID: 0.104 G_Rec: 0.307 D_GP: 0.032 D_real: 1.065 D_fake: 0.731 +(epoch: 459, iters: 3342, time: 0.063) G_GAN: 0.585 G_GAN_Feat: 0.982 G_ID: 0.108 G_Rec: 0.395 D_GP: 0.042 D_real: 1.080 D_fake: 0.424 +(epoch: 459, iters: 3742, time: 0.063) G_GAN: 0.128 G_GAN_Feat: 0.821 G_ID: 0.125 G_Rec: 0.278 D_GP: 0.082 D_real: 0.591 D_fake: 0.874 +(epoch: 459, iters: 4142, time: 0.064) G_GAN: 0.196 G_GAN_Feat: 1.077 G_ID: 0.108 G_Rec: 0.452 D_GP: 0.131 D_real: 0.584 D_fake: 0.808 +(epoch: 459, iters: 4542, time: 0.063) G_GAN: 0.244 G_GAN_Feat: 0.753 G_ID: 0.111 G_Rec: 0.300 D_GP: 0.035 D_real: 0.976 D_fake: 0.757 +(epoch: 459, iters: 4942, time: 0.063) G_GAN: 0.505 G_GAN_Feat: 0.990 G_ID: 0.115 G_Rec: 0.421 D_GP: 0.097 D_real: 0.747 D_fake: 0.511 +(epoch: 459, iters: 5342, time: 0.063) G_GAN: -0.010 G_GAN_Feat: 0.857 G_ID: 0.114 G_Rec: 0.308 D_GP: 0.051 D_real: 0.528 D_fake: 1.010 +(epoch: 459, iters: 5742, time: 0.064) G_GAN: 0.601 G_GAN_Feat: 1.082 G_ID: 0.128 G_Rec: 0.402 D_GP: 0.041 D_real: 0.629 D_fake: 0.408 +(epoch: 459, iters: 6142, time: 0.063) G_GAN: 0.597 G_GAN_Feat: 0.867 G_ID: 0.102 G_Rec: 0.331 D_GP: 0.041 D_real: 1.389 D_fake: 0.427 +(epoch: 459, iters: 6542, time: 0.063) G_GAN: 0.226 G_GAN_Feat: 0.995 G_ID: 0.119 G_Rec: 0.430 D_GP: 0.182 D_real: 0.505 D_fake: 0.775 +(epoch: 459, iters: 6942, time: 0.063) G_GAN: 0.099 G_GAN_Feat: 0.759 G_ID: 0.091 G_Rec: 0.299 D_GP: 0.033 D_real: 0.924 D_fake: 0.901 +(epoch: 459, iters: 7342, time: 0.064) G_GAN: 0.471 G_GAN_Feat: 1.064 G_ID: 0.122 G_Rec: 0.480 D_GP: 0.039 D_real: 0.941 D_fake: 0.536 +(epoch: 459, iters: 7742, time: 0.063) G_GAN: 0.620 G_GAN_Feat: 0.914 G_ID: 0.094 G_Rec: 0.312 D_GP: 0.057 D_real: 0.685 D_fake: 0.402 +(epoch: 459, iters: 8142, time: 0.063) G_GAN: 0.346 G_GAN_Feat: 0.931 G_ID: 0.126 G_Rec: 0.443 D_GP: 0.033 D_real: 0.985 D_fake: 0.654 +(epoch: 459, iters: 8542, time: 0.063) G_GAN: 0.134 G_GAN_Feat: 0.795 G_ID: 0.098 G_Rec: 0.336 D_GP: 0.078 D_real: 0.803 D_fake: 0.867 +(epoch: 460, iters: 334, time: 0.064) G_GAN: 0.411 G_GAN_Feat: 0.947 G_ID: 0.108 G_Rec: 0.434 D_GP: 0.033 D_real: 1.018 D_fake: 0.592 +(epoch: 460, iters: 734, time: 0.063) G_GAN: 0.200 G_GAN_Feat: 0.787 G_ID: 0.112 G_Rec: 0.303 D_GP: 0.048 D_real: 0.959 D_fake: 0.802 +(epoch: 460, iters: 1134, time: 0.063) G_GAN: 0.459 G_GAN_Feat: 1.056 G_ID: 0.122 G_Rec: 0.449 D_GP: 0.033 D_real: 0.979 D_fake: 0.544 +(epoch: 460, iters: 1534, time: 0.063) G_GAN: 0.069 G_GAN_Feat: 0.795 G_ID: 0.095 G_Rec: 0.333 D_GP: 0.045 D_real: 0.849 D_fake: 0.931 +(epoch: 460, iters: 1934, time: 0.064) G_GAN: 0.503 G_GAN_Feat: 0.998 G_ID: 0.122 G_Rec: 0.424 D_GP: 0.037 D_real: 0.957 D_fake: 0.505 +(epoch: 460, iters: 2334, time: 0.063) G_GAN: 0.367 G_GAN_Feat: 0.874 G_ID: 0.099 G_Rec: 0.376 D_GP: 0.034 D_real: 0.953 D_fake: 0.633 +(epoch: 460, iters: 2734, time: 0.063) G_GAN: 0.706 G_GAN_Feat: 1.097 G_ID: 0.132 G_Rec: 0.437 D_GP: 0.128 D_real: 0.548 D_fake: 0.329 +(epoch: 460, iters: 3134, time: 0.063) G_GAN: 0.053 G_GAN_Feat: 0.748 G_ID: 0.109 G_Rec: 0.358 D_GP: 0.032 D_real: 0.958 D_fake: 0.947 +(epoch: 460, iters: 3534, time: 0.064) G_GAN: 0.248 G_GAN_Feat: 0.884 G_ID: 0.122 G_Rec: 0.411 D_GP: 0.033 D_real: 1.043 D_fake: 0.754 +(epoch: 460, iters: 3934, time: 0.063) G_GAN: 0.022 G_GAN_Feat: 0.683 G_ID: 0.115 G_Rec: 0.279 D_GP: 0.035 D_real: 0.883 D_fake: 0.978 +(epoch: 460, iters: 4334, time: 0.063) G_GAN: 0.286 G_GAN_Feat: 0.889 G_ID: 0.135 G_Rec: 0.397 D_GP: 0.036 D_real: 0.932 D_fake: 0.717 +(epoch: 460, iters: 4734, time: 0.063) G_GAN: -0.009 G_GAN_Feat: 0.784 G_ID: 0.100 G_Rec: 0.338 D_GP: 0.057 D_real: 0.760 D_fake: 1.009 +(epoch: 460, iters: 5134, time: 0.064) G_GAN: 0.423 G_GAN_Feat: 0.950 G_ID: 0.117 G_Rec: 0.423 D_GP: 0.034 D_real: 1.086 D_fake: 0.579 +(epoch: 460, iters: 5534, time: 0.064) G_GAN: 0.298 G_GAN_Feat: 0.739 G_ID: 0.095 G_Rec: 0.286 D_GP: 0.036 D_real: 1.142 D_fake: 0.702 +(epoch: 460, iters: 5934, time: 0.064) G_GAN: 0.309 G_GAN_Feat: 0.988 G_ID: 0.112 G_Rec: 0.454 D_GP: 0.033 D_real: 0.945 D_fake: 0.693 +(epoch: 460, iters: 6334, time: 0.064) G_GAN: 0.280 G_GAN_Feat: 0.849 G_ID: 0.098 G_Rec: 0.307 D_GP: 0.050 D_real: 0.632 D_fake: 0.721 +(epoch: 460, iters: 6734, time: 0.064) G_GAN: 0.191 G_GAN_Feat: 1.172 G_ID: 0.123 G_Rec: 0.479 D_GP: 0.072 D_real: 0.183 D_fake: 0.810 +(epoch: 460, iters: 7134, time: 0.064) G_GAN: 0.248 G_GAN_Feat: 0.741 G_ID: 0.091 G_Rec: 0.310 D_GP: 0.033 D_real: 1.206 D_fake: 0.754 +(epoch: 460, iters: 7534, time: 0.064) G_GAN: 0.292 G_GAN_Feat: 0.993 G_ID: 0.128 G_Rec: 0.430 D_GP: 0.039 D_real: 0.755 D_fake: 0.708 +(epoch: 460, iters: 7934, time: 0.064) G_GAN: 0.423 G_GAN_Feat: 0.868 G_ID: 0.097 G_Rec: 0.303 D_GP: 0.068 D_real: 0.775 D_fake: 0.583 +(epoch: 460, iters: 8334, time: 0.064) G_GAN: 0.745 G_GAN_Feat: 1.277 G_ID: 0.109 G_Rec: 0.506 D_GP: 0.106 D_real: 0.942 D_fake: 0.375 +(epoch: 461, iters: 126, time: 0.064) G_GAN: 0.372 G_GAN_Feat: 0.967 G_ID: 0.101 G_Rec: 0.366 D_GP: 0.227 D_real: 0.323 D_fake: 0.696 +(epoch: 461, iters: 526, time: 0.063) G_GAN: 0.256 G_GAN_Feat: 1.001 G_ID: 0.149 G_Rec: 0.449 D_GP: 0.038 D_real: 0.911 D_fake: 0.745 +(epoch: 461, iters: 926, time: 0.064) G_GAN: 0.208 G_GAN_Feat: 0.836 G_ID: 0.099 G_Rec: 0.326 D_GP: 0.042 D_real: 0.797 D_fake: 0.792 +(epoch: 461, iters: 1326, time: 0.064) G_GAN: 0.455 G_GAN_Feat: 0.948 G_ID: 0.115 G_Rec: 0.388 D_GP: 0.035 D_real: 1.173 D_fake: 0.546 +(epoch: 461, iters: 1726, time: 0.064) G_GAN: 0.173 G_GAN_Feat: 0.812 G_ID: 0.127 G_Rec: 0.357 D_GP: 0.039 D_real: 1.075 D_fake: 0.827 +(epoch: 461, iters: 2126, time: 0.064) G_GAN: 0.356 G_GAN_Feat: 0.871 G_ID: 0.104 G_Rec: 0.419 D_GP: 0.030 D_real: 1.061 D_fake: 0.645 +(epoch: 461, iters: 2526, time: 0.063) G_GAN: 0.067 G_GAN_Feat: 0.656 G_ID: 0.110 G_Rec: 0.294 D_GP: 0.033 D_real: 1.012 D_fake: 0.934 +(epoch: 461, iters: 2926, time: 0.064) G_GAN: 0.335 G_GAN_Feat: 0.882 G_ID: 0.115 G_Rec: 0.410 D_GP: 0.045 D_real: 0.996 D_fake: 0.672 +(epoch: 461, iters: 3326, time: 0.064) G_GAN: 0.109 G_GAN_Feat: 0.728 G_ID: 0.081 G_Rec: 0.321 D_GP: 0.035 D_real: 0.928 D_fake: 0.892 +(epoch: 461, iters: 3726, time: 0.063) G_GAN: 0.317 G_GAN_Feat: 0.959 G_ID: 0.106 G_Rec: 0.434 D_GP: 0.038 D_real: 0.980 D_fake: 0.689 +(epoch: 461, iters: 4126, time: 0.064) G_GAN: -0.106 G_GAN_Feat: 0.697 G_ID: 0.109 G_Rec: 0.297 D_GP: 0.041 D_real: 0.785 D_fake: 1.106 +(epoch: 461, iters: 4526, time: 0.064) G_GAN: 0.210 G_GAN_Feat: 0.873 G_ID: 0.121 G_Rec: 0.404 D_GP: 0.040 D_real: 0.831 D_fake: 0.790 +(epoch: 461, iters: 4926, time: 0.063) G_GAN: 0.216 G_GAN_Feat: 0.761 G_ID: 0.111 G_Rec: 0.331 D_GP: 0.036 D_real: 0.980 D_fake: 0.785 +(epoch: 461, iters: 5326, time: 0.064) G_GAN: 0.633 G_GAN_Feat: 0.916 G_ID: 0.102 G_Rec: 0.403 D_GP: 0.043 D_real: 1.341 D_fake: 0.405 +(epoch: 461, iters: 5726, time: 0.064) G_GAN: 0.565 G_GAN_Feat: 0.679 G_ID: 0.101 G_Rec: 0.293 D_GP: 0.030 D_real: 1.464 D_fake: 0.462 +(epoch: 461, iters: 6126, time: 0.064) G_GAN: 0.348 G_GAN_Feat: 1.004 G_ID: 0.118 G_Rec: 0.455 D_GP: 0.052 D_real: 0.813 D_fake: 0.662 +(epoch: 461, iters: 6526, time: 0.064) G_GAN: 0.048 G_GAN_Feat: 0.791 G_ID: 0.082 G_Rec: 0.378 D_GP: 0.044 D_real: 0.954 D_fake: 0.953 +(epoch: 461, iters: 6926, time: 0.064) G_GAN: 0.350 G_GAN_Feat: 1.033 G_ID: 0.113 G_Rec: 0.453 D_GP: 0.045 D_real: 0.798 D_fake: 0.652 +(epoch: 461, iters: 7326, time: 0.064) G_GAN: 0.149 G_GAN_Feat: 0.743 G_ID: 0.094 G_Rec: 0.296 D_GP: 0.044 D_real: 0.931 D_fake: 0.851 +(epoch: 461, iters: 7726, time: 0.064) G_GAN: 0.437 G_GAN_Feat: 1.151 G_ID: 0.112 G_Rec: 0.496 D_GP: 0.209 D_real: 0.413 D_fake: 0.565 +(epoch: 461, iters: 8126, time: 0.064) G_GAN: 0.466 G_GAN_Feat: 0.858 G_ID: 0.097 G_Rec: 0.305 D_GP: 0.040 D_real: 1.133 D_fake: 0.541 +(epoch: 461, iters: 8526, time: 0.064) G_GAN: 0.543 G_GAN_Feat: 1.046 G_ID: 0.114 G_Rec: 0.458 D_GP: 0.065 D_real: 0.949 D_fake: 0.523 +(epoch: 462, iters: 318, time: 0.064) G_GAN: 0.018 G_GAN_Feat: 0.733 G_ID: 0.102 G_Rec: 0.330 D_GP: 0.037 D_real: 0.960 D_fake: 0.982 +(epoch: 462, iters: 718, time: 0.064) G_GAN: 0.684 G_GAN_Feat: 0.891 G_ID: 0.104 G_Rec: 0.420 D_GP: 0.030 D_real: 1.405 D_fake: 0.349 +(epoch: 462, iters: 1118, time: 0.064) G_GAN: 0.077 G_GAN_Feat: 0.798 G_ID: 0.109 G_Rec: 0.311 D_GP: 0.055 D_real: 0.697 D_fake: 0.923 +(epoch: 462, iters: 1518, time: 0.064) G_GAN: 0.859 G_GAN_Feat: 0.964 G_ID: 0.106 G_Rec: 0.447 D_GP: 0.039 D_real: 1.615 D_fake: 0.239 +(epoch: 462, iters: 1918, time: 0.064) G_GAN: -0.037 G_GAN_Feat: 0.780 G_ID: 0.107 G_Rec: 0.330 D_GP: 0.031 D_real: 0.821 D_fake: 1.037 +(epoch: 462, iters: 2318, time: 0.064) G_GAN: 0.086 G_GAN_Feat: 0.871 G_ID: 0.116 G_Rec: 0.387 D_GP: 0.029 D_real: 0.749 D_fake: 0.914 +(epoch: 462, iters: 2718, time: 0.064) G_GAN: 0.123 G_GAN_Feat: 0.641 G_ID: 0.104 G_Rec: 0.280 D_GP: 0.033 D_real: 1.030 D_fake: 0.877 +(epoch: 462, iters: 3118, time: 0.064) G_GAN: 0.133 G_GAN_Feat: 0.920 G_ID: 0.126 G_Rec: 0.399 D_GP: 0.050 D_real: 0.685 D_fake: 0.867 +(epoch: 462, iters: 3518, time: 0.064) G_GAN: 0.256 G_GAN_Feat: 0.650 G_ID: 0.096 G_Rec: 0.293 D_GP: 0.032 D_real: 1.146 D_fake: 0.745 +(epoch: 462, iters: 3918, time: 0.064) G_GAN: 0.415 G_GAN_Feat: 1.026 G_ID: 0.124 G_Rec: 0.467 D_GP: 0.072 D_real: 0.726 D_fake: 0.594 +(epoch: 462, iters: 4318, time: 0.064) G_GAN: 0.184 G_GAN_Feat: 0.897 G_ID: 0.108 G_Rec: 0.368 D_GP: 0.071 D_real: 0.727 D_fake: 0.817 +(epoch: 462, iters: 4718, time: 0.064) G_GAN: 0.607 G_GAN_Feat: 0.920 G_ID: 0.114 G_Rec: 0.390 D_GP: 0.037 D_real: 1.238 D_fake: 0.415 +(epoch: 462, iters: 5118, time: 0.064) G_GAN: 0.005 G_GAN_Feat: 0.821 G_ID: 0.099 G_Rec: 0.323 D_GP: 0.043 D_real: 0.719 D_fake: 0.995 +(epoch: 462, iters: 5518, time: 0.064) G_GAN: 0.523 G_GAN_Feat: 1.201 G_ID: 0.128 G_Rec: 0.477 D_GP: 0.188 D_real: 0.300 D_fake: 0.491 +(epoch: 462, iters: 5918, time: 0.064) G_GAN: 0.026 G_GAN_Feat: 1.074 G_ID: 0.096 G_Rec: 0.339 D_GP: 1.703 D_real: 0.660 D_fake: 0.996 +(epoch: 462, iters: 6318, time: 0.064) G_GAN: 0.573 G_GAN_Feat: 1.032 G_ID: 0.109 G_Rec: 0.476 D_GP: 0.034 D_real: 1.135 D_fake: 0.451 +(epoch: 462, iters: 6718, time: 0.064) G_GAN: 0.206 G_GAN_Feat: 0.700 G_ID: 0.101 G_Rec: 0.305 D_GP: 0.040 D_real: 1.139 D_fake: 0.794 +(epoch: 462, iters: 7118, time: 0.064) G_GAN: 0.491 G_GAN_Feat: 1.024 G_ID: 0.127 G_Rec: 0.389 D_GP: 0.050 D_real: 1.141 D_fake: 0.559 +(epoch: 462, iters: 7518, time: 0.064) G_GAN: 0.570 G_GAN_Feat: 0.687 G_ID: 0.101 G_Rec: 0.281 D_GP: 0.031 D_real: 1.424 D_fake: 0.440 +(epoch: 462, iters: 7918, time: 0.064) G_GAN: 0.354 G_GAN_Feat: 1.115 G_ID: 0.127 G_Rec: 0.415 D_GP: 0.047 D_real: 0.976 D_fake: 0.647 +(epoch: 462, iters: 8318, time: 0.064) G_GAN: 0.369 G_GAN_Feat: 0.910 G_ID: 0.103 G_Rec: 0.326 D_GP: 0.059 D_real: 0.395 D_fake: 0.695 +(epoch: 463, iters: 110, time: 0.064) G_GAN: 0.492 G_GAN_Feat: 1.282 G_ID: 0.116 G_Rec: 0.497 D_GP: 0.174 D_real: 0.351 D_fake: 0.513 +(epoch: 463, iters: 510, time: 0.064) G_GAN: 0.363 G_GAN_Feat: 0.946 G_ID: 0.098 G_Rec: 0.336 D_GP: 0.092 D_real: 0.584 D_fake: 0.638 +(epoch: 463, iters: 910, time: 0.064) G_GAN: 0.435 G_GAN_Feat: 0.964 G_ID: 0.112 G_Rec: 0.459 D_GP: 0.030 D_real: 1.104 D_fake: 0.572 +(epoch: 463, iters: 1310, time: 0.064) G_GAN: 0.242 G_GAN_Feat: 0.671 G_ID: 0.095 G_Rec: 0.291 D_GP: 0.029 D_real: 1.191 D_fake: 0.758 +(epoch: 463, iters: 1710, time: 0.064) G_GAN: 0.091 G_GAN_Feat: 0.855 G_ID: 0.123 G_Rec: 0.418 D_GP: 0.037 D_real: 0.796 D_fake: 0.909 +(epoch: 463, iters: 2110, time: 0.064) G_GAN: -0.052 G_GAN_Feat: 0.719 G_ID: 0.095 G_Rec: 0.344 D_GP: 0.036 D_real: 0.802 D_fake: 1.052 +(epoch: 463, iters: 2510, time: 0.063) G_GAN: 0.283 G_GAN_Feat: 0.883 G_ID: 0.115 G_Rec: 0.405 D_GP: 0.034 D_real: 1.023 D_fake: 0.730 +(epoch: 463, iters: 2910, time: 0.064) G_GAN: 0.236 G_GAN_Feat: 0.670 G_ID: 0.093 G_Rec: 0.308 D_GP: 0.035 D_real: 1.105 D_fake: 0.768 +(epoch: 463, iters: 3310, time: 0.064) G_GAN: 0.114 G_GAN_Feat: 0.918 G_ID: 0.113 G_Rec: 0.428 D_GP: 0.033 D_real: 0.677 D_fake: 0.886 +(epoch: 463, iters: 3710, time: 0.064) G_GAN: 0.111 G_GAN_Feat: 0.699 G_ID: 0.088 G_Rec: 0.311 D_GP: 0.050 D_real: 0.926 D_fake: 0.890 +(epoch: 463, iters: 4110, time: 0.064) G_GAN: 0.329 G_GAN_Feat: 1.008 G_ID: 0.112 G_Rec: 0.465 D_GP: 0.057 D_real: 0.596 D_fake: 0.693 +(epoch: 463, iters: 4510, time: 0.064) G_GAN: 0.235 G_GAN_Feat: 0.763 G_ID: 0.100 G_Rec: 0.285 D_GP: 0.044 D_real: 1.197 D_fake: 0.765 +(epoch: 463, iters: 4910, time: 0.064) G_GAN: 0.677 G_GAN_Feat: 0.970 G_ID: 0.128 G_Rec: 0.452 D_GP: 0.046 D_real: 1.199 D_fake: 0.342 +(epoch: 463, iters: 5310, time: 0.064) G_GAN: 0.331 G_GAN_Feat: 0.753 G_ID: 0.089 G_Rec: 0.305 D_GP: 0.045 D_real: 1.012 D_fake: 0.670 +(epoch: 463, iters: 5710, time: 0.064) G_GAN: 0.440 G_GAN_Feat: 0.911 G_ID: 0.118 G_Rec: 0.423 D_GP: 0.031 D_real: 1.213 D_fake: 0.564 +(epoch: 463, iters: 6110, time: 0.064) G_GAN: 0.287 G_GAN_Feat: 0.722 G_ID: 0.092 G_Rec: 0.295 D_GP: 0.032 D_real: 1.158 D_fake: 0.714 +(epoch: 463, iters: 6510, time: 0.064) G_GAN: 0.426 G_GAN_Feat: 0.894 G_ID: 0.107 G_Rec: 0.400 D_GP: 0.032 D_real: 1.054 D_fake: 0.575 +(epoch: 463, iters: 6910, time: 0.064) G_GAN: 0.253 G_GAN_Feat: 0.860 G_ID: 0.111 G_Rec: 0.317 D_GP: 0.152 D_real: 0.574 D_fake: 0.750 +(epoch: 463, iters: 7310, time: 0.064) G_GAN: 0.312 G_GAN_Feat: 1.070 G_ID: 0.130 G_Rec: 0.530 D_GP: 0.704 D_real: 0.307 D_fake: 0.700 +(epoch: 463, iters: 7710, time: 0.064) G_GAN: 0.427 G_GAN_Feat: 0.818 G_ID: 0.096 G_Rec: 0.321 D_GP: 0.049 D_real: 1.063 D_fake: 0.578 +(epoch: 463, iters: 8110, time: 0.063) G_GAN: 0.463 G_GAN_Feat: 0.946 G_ID: 0.110 G_Rec: 0.469 D_GP: 0.035 D_real: 1.056 D_fake: 0.554 +(epoch: 463, iters: 8510, time: 0.064) G_GAN: 0.045 G_GAN_Feat: 0.585 G_ID: 0.098 G_Rec: 0.279 D_GP: 0.027 D_real: 0.997 D_fake: 0.955 +(epoch: 464, iters: 302, time: 0.064) G_GAN: 0.216 G_GAN_Feat: 0.968 G_ID: 0.113 G_Rec: 0.465 D_GP: 0.036 D_real: 0.712 D_fake: 0.786 +(epoch: 464, iters: 702, time: 0.064) G_GAN: -0.066 G_GAN_Feat: 0.719 G_ID: 0.104 G_Rec: 0.325 D_GP: 0.050 D_real: 0.769 D_fake: 1.067 +(epoch: 464, iters: 1102, time: 0.064) G_GAN: 0.455 G_GAN_Feat: 0.910 G_ID: 0.111 G_Rec: 0.439 D_GP: 0.065 D_real: 1.015 D_fake: 0.569 +(epoch: 464, iters: 1502, time: 0.064) G_GAN: -0.025 G_GAN_Feat: 0.697 G_ID: 0.117 G_Rec: 0.294 D_GP: 0.037 D_real: 0.847 D_fake: 1.025 +(epoch: 464, iters: 1902, time: 0.064) G_GAN: 0.385 G_GAN_Feat: 0.911 G_ID: 0.113 G_Rec: 0.428 D_GP: 0.036 D_real: 0.980 D_fake: 0.629 +(epoch: 464, iters: 2302, time: 0.064) G_GAN: 0.125 G_GAN_Feat: 0.656 G_ID: 0.091 G_Rec: 0.277 D_GP: 0.030 D_real: 0.997 D_fake: 0.875 +(epoch: 464, iters: 2702, time: 0.064) G_GAN: 0.344 G_GAN_Feat: 1.050 G_ID: 0.117 G_Rec: 0.495 D_GP: 0.079 D_real: 0.628 D_fake: 0.670 +(epoch: 464, iters: 3102, time: 0.064) G_GAN: 0.171 G_GAN_Feat: 0.699 G_ID: 0.090 G_Rec: 0.307 D_GP: 0.038 D_real: 1.065 D_fake: 0.829 +(epoch: 464, iters: 3502, time: 0.064) G_GAN: 0.675 G_GAN_Feat: 1.026 G_ID: 0.120 G_Rec: 0.399 D_GP: 0.042 D_real: 1.078 D_fake: 0.380 +(epoch: 464, iters: 3902, time: 0.064) G_GAN: 0.246 G_GAN_Feat: 0.763 G_ID: 0.097 G_Rec: 0.360 D_GP: 0.036 D_real: 1.061 D_fake: 0.754 +(epoch: 464, iters: 4302, time: 0.064) G_GAN: 0.559 G_GAN_Feat: 0.990 G_ID: 0.115 G_Rec: 0.442 D_GP: 0.056 D_real: 0.813 D_fake: 0.452 +(epoch: 464, iters: 4702, time: 0.064) G_GAN: 0.408 G_GAN_Feat: 0.795 G_ID: 0.103 G_Rec: 0.329 D_GP: 0.046 D_real: 1.120 D_fake: 0.597 +(epoch: 464, iters: 5102, time: 0.064) G_GAN: 0.819 G_GAN_Feat: 1.049 G_ID: 0.104 G_Rec: 0.471 D_GP: 0.061 D_real: 1.081 D_fake: 0.274 +(epoch: 464, iters: 5502, time: 0.064) G_GAN: 0.086 G_GAN_Feat: 0.798 G_ID: 0.099 G_Rec: 0.332 D_GP: 0.032 D_real: 0.825 D_fake: 0.914 +(epoch: 464, iters: 5902, time: 0.064) G_GAN: -0.233 G_GAN_Feat: 1.176 G_ID: 0.115 G_Rec: 0.527 D_GP: 0.066 D_real: 0.815 D_fake: 1.234 +(epoch: 464, iters: 6302, time: 0.064) G_GAN: 0.091 G_GAN_Feat: 0.687 G_ID: 0.096 G_Rec: 0.303 D_GP: 0.024 D_real: 1.057 D_fake: 0.909 +(epoch: 464, iters: 6702, time: 0.064) G_GAN: 0.167 G_GAN_Feat: 0.953 G_ID: 0.113 G_Rec: 0.423 D_GP: 0.030 D_real: 0.796 D_fake: 0.838 +(epoch: 464, iters: 7102, time: 0.064) G_GAN: -0.184 G_GAN_Feat: 0.746 G_ID: 0.094 G_Rec: 0.340 D_GP: 0.044 D_real: 0.651 D_fake: 1.184 +(epoch: 464, iters: 7502, time: 0.064) G_GAN: 0.216 G_GAN_Feat: 1.011 G_ID: 0.119 G_Rec: 0.475 D_GP: 0.048 D_real: 0.742 D_fake: 0.788 +(epoch: 464, iters: 7902, time: 0.064) G_GAN: -0.217 G_GAN_Feat: 0.773 G_ID: 0.096 G_Rec: 0.323 D_GP: 0.062 D_real: 0.553 D_fake: 1.217 +(epoch: 464, iters: 8302, time: 0.064) G_GAN: 0.404 G_GAN_Feat: 0.930 G_ID: 0.116 G_Rec: 0.380 D_GP: 0.054 D_real: 1.021 D_fake: 0.598 +(epoch: 464, iters: 8702, time: 0.064) G_GAN: 0.153 G_GAN_Feat: 0.879 G_ID: 0.091 G_Rec: 0.321 D_GP: 0.533 D_real: 0.395 D_fake: 0.847 +(epoch: 465, iters: 494, time: 0.064) G_GAN: 0.619 G_GAN_Feat: 1.027 G_ID: 0.124 G_Rec: 0.475 D_GP: 0.036 D_real: 1.316 D_fake: 0.486 +(epoch: 465, iters: 894, time: 0.064) G_GAN: 0.249 G_GAN_Feat: 0.679 G_ID: 0.081 G_Rec: 0.272 D_GP: 0.028 D_real: 1.153 D_fake: 0.752 +(epoch: 465, iters: 1294, time: 0.064) G_GAN: 0.590 G_GAN_Feat: 0.950 G_ID: 0.119 G_Rec: 0.441 D_GP: 0.040 D_real: 1.252 D_fake: 0.417 +(epoch: 465, iters: 1694, time: 0.064) G_GAN: 0.489 G_GAN_Feat: 0.958 G_ID: 0.088 G_Rec: 0.318 D_GP: 0.120 D_real: 0.620 D_fake: 0.518 +(epoch: 465, iters: 2094, time: 0.064) G_GAN: 0.254 G_GAN_Feat: 0.933 G_ID: 0.126 G_Rec: 0.451 D_GP: 0.041 D_real: 1.009 D_fake: 0.747 +(epoch: 465, iters: 2494, time: 0.064) G_GAN: 0.027 G_GAN_Feat: 0.731 G_ID: 0.106 G_Rec: 0.317 D_GP: 0.035 D_real: 0.906 D_fake: 0.973 +(epoch: 465, iters: 2894, time: 0.064) G_GAN: 0.399 G_GAN_Feat: 0.848 G_ID: 0.098 G_Rec: 0.424 D_GP: 0.027 D_real: 1.119 D_fake: 0.607 +(epoch: 465, iters: 3294, time: 0.064) G_GAN: 0.129 G_GAN_Feat: 0.657 G_ID: 0.104 G_Rec: 0.303 D_GP: 0.035 D_real: 1.013 D_fake: 0.872 +(epoch: 465, iters: 3694, time: 0.064) G_GAN: 0.339 G_GAN_Feat: 0.893 G_ID: 0.107 G_Rec: 0.451 D_GP: 0.032 D_real: 1.003 D_fake: 0.675 +(epoch: 465, iters: 4094, time: 0.064) G_GAN: -0.059 G_GAN_Feat: 0.668 G_ID: 0.095 G_Rec: 0.324 D_GP: 0.030 D_real: 0.847 D_fake: 1.059 +(epoch: 465, iters: 4494, time: 0.064) G_GAN: 0.420 G_GAN_Feat: 0.913 G_ID: 0.108 G_Rec: 0.446 D_GP: 0.039 D_real: 1.077 D_fake: 0.602 +(epoch: 465, iters: 4894, time: 0.064) G_GAN: 0.113 G_GAN_Feat: 0.728 G_ID: 0.090 G_Rec: 0.325 D_GP: 0.044 D_real: 0.948 D_fake: 0.887 +(epoch: 465, iters: 5294, time: 0.064) G_GAN: 0.325 G_GAN_Feat: 0.939 G_ID: 0.108 G_Rec: 0.455 D_GP: 0.035 D_real: 0.926 D_fake: 0.679 +(epoch: 465, iters: 5694, time: 0.064) G_GAN: 0.105 G_GAN_Feat: 0.707 G_ID: 0.095 G_Rec: 0.301 D_GP: 0.041 D_real: 0.949 D_fake: 0.896 +(epoch: 465, iters: 6094, time: 0.064) G_GAN: 0.210 G_GAN_Feat: 1.010 G_ID: 0.119 G_Rec: 0.448 D_GP: 0.057 D_real: 0.789 D_fake: 0.791 +(epoch: 465, iters: 6494, time: 0.064) G_GAN: -0.049 G_GAN_Feat: 0.721 G_ID: 0.103 G_Rec: 0.277 D_GP: 0.043 D_real: 0.714 D_fake: 1.049 +(epoch: 465, iters: 6894, time: 0.063) G_GAN: 0.194 G_GAN_Feat: 0.879 G_ID: 0.129 G_Rec: 0.364 D_GP: 0.040 D_real: 0.973 D_fake: 0.809 +(epoch: 465, iters: 7294, time: 0.064) G_GAN: 0.054 G_GAN_Feat: 0.791 G_ID: 0.104 G_Rec: 0.323 D_GP: 0.062 D_real: 0.615 D_fake: 0.946 +(epoch: 465, iters: 7694, time: 0.064) G_GAN: 0.762 G_GAN_Feat: 1.002 G_ID: 0.117 G_Rec: 0.428 D_GP: 0.043 D_real: 1.193 D_fake: 0.301 +(epoch: 465, iters: 8094, time: 0.064) G_GAN: 0.402 G_GAN_Feat: 0.845 G_ID: 0.103 G_Rec: 0.340 D_GP: 0.063 D_real: 0.945 D_fake: 0.603 +(epoch: 465, iters: 8494, time: 0.064) G_GAN: 0.658 G_GAN_Feat: 1.031 G_ID: 0.110 G_Rec: 0.467 D_GP: 0.049 D_real: 1.049 D_fake: 0.373 +(epoch: 466, iters: 286, time: 0.064) G_GAN: 0.288 G_GAN_Feat: 0.846 G_ID: 0.103 G_Rec: 0.332 D_GP: 0.033 D_real: 0.916 D_fake: 0.719 +(epoch: 466, iters: 686, time: 0.064) G_GAN: 0.449 G_GAN_Feat: 1.186 G_ID: 0.097 G_Rec: 0.467 D_GP: 0.136 D_real: 0.450 D_fake: 0.561 +(epoch: 466, iters: 1086, time: 0.064) G_GAN: 0.355 G_GAN_Feat: 0.972 G_ID: 0.105 G_Rec: 0.356 D_GP: 0.075 D_real: 0.494 D_fake: 0.648 +(epoch: 466, iters: 1486, time: 0.064) G_GAN: 0.954 G_GAN_Feat: 1.159 G_ID: 0.130 G_Rec: 0.465 D_GP: 0.124 D_real: 0.724 D_fake: 0.204 +(epoch: 466, iters: 1886, time: 0.063) G_GAN: -0.078 G_GAN_Feat: 0.836 G_ID: 0.106 G_Rec: 0.359 D_GP: 0.037 D_real: 0.605 D_fake: 1.078 +(epoch: 466, iters: 2286, time: 0.064) G_GAN: 0.768 G_GAN_Feat: 1.033 G_ID: 0.120 G_Rec: 0.451 D_GP: 0.042 D_real: 1.474 D_fake: 0.288 +(epoch: 466, iters: 2686, time: 0.064) G_GAN: 0.166 G_GAN_Feat: 0.748 G_ID: 0.099 G_Rec: 0.276 D_GP: 0.029 D_real: 0.998 D_fake: 0.834 +(epoch: 466, iters: 3086, time: 0.063) G_GAN: 0.986 G_GAN_Feat: 1.179 G_ID: 0.125 G_Rec: 0.449 D_GP: 0.056 D_real: 1.087 D_fake: 0.195 +(epoch: 466, iters: 3486, time: 0.063) G_GAN: 0.270 G_GAN_Feat: 0.735 G_ID: 0.083 G_Rec: 0.329 D_GP: 0.031 D_real: 1.142 D_fake: 0.731 +(epoch: 466, iters: 3886, time: 0.064) G_GAN: 0.393 G_GAN_Feat: 0.933 G_ID: 0.113 G_Rec: 0.475 D_GP: 0.028 D_real: 1.063 D_fake: 0.611 +(epoch: 466, iters: 4286, time: 0.064) G_GAN: 0.140 G_GAN_Feat: 0.648 G_ID: 0.107 G_Rec: 0.289 D_GP: 0.028 D_real: 1.025 D_fake: 0.860 +(epoch: 466, iters: 4686, time: 0.063) G_GAN: 0.333 G_GAN_Feat: 0.922 G_ID: 0.104 G_Rec: 0.441 D_GP: 0.042 D_real: 0.845 D_fake: 0.680 +(epoch: 466, iters: 5086, time: 0.063) G_GAN: 0.124 G_GAN_Feat: 0.686 G_ID: 0.081 G_Rec: 0.315 D_GP: 0.059 D_real: 0.903 D_fake: 0.878 +(epoch: 466, iters: 5486, time: 0.063) G_GAN: 0.549 G_GAN_Feat: 0.920 G_ID: 0.117 G_Rec: 0.464 D_GP: 0.038 D_real: 1.165 D_fake: 0.487 +(epoch: 466, iters: 5886, time: 0.064) G_GAN: 0.159 G_GAN_Feat: 0.697 G_ID: 0.088 G_Rec: 0.294 D_GP: 0.039 D_real: 0.978 D_fake: 0.841 +(epoch: 466, iters: 6286, time: 0.063) G_GAN: 0.422 G_GAN_Feat: 0.953 G_ID: 0.110 G_Rec: 0.414 D_GP: 0.036 D_real: 1.113 D_fake: 0.581 +(epoch: 466, iters: 6686, time: 0.063) G_GAN: 0.218 G_GAN_Feat: 0.770 G_ID: 0.084 G_Rec: 0.318 D_GP: 0.034 D_real: 1.007 D_fake: 0.783 +(epoch: 466, iters: 7086, time: 0.063) G_GAN: 0.472 G_GAN_Feat: 0.926 G_ID: 0.110 G_Rec: 0.414 D_GP: 0.036 D_real: 1.159 D_fake: 0.529 +(epoch: 466, iters: 7486, time: 0.064) G_GAN: 0.138 G_GAN_Feat: 0.813 G_ID: 0.095 G_Rec: 0.309 D_GP: 0.039 D_real: 1.155 D_fake: 0.863 +(epoch: 466, iters: 7886, time: 0.063) G_GAN: 0.597 G_GAN_Feat: 1.018 G_ID: 0.110 G_Rec: 0.433 D_GP: 0.051 D_real: 1.090 D_fake: 0.423 +(epoch: 466, iters: 8286, time: 0.063) G_GAN: 0.369 G_GAN_Feat: 0.756 G_ID: 0.091 G_Rec: 0.295 D_GP: 0.030 D_real: 1.263 D_fake: 0.656 +(epoch: 466, iters: 8686, time: 0.064) G_GAN: 0.636 G_GAN_Feat: 1.090 G_ID: 0.127 G_Rec: 0.446 D_GP: 0.047 D_real: 0.926 D_fake: 0.373 +(epoch: 467, iters: 478, time: 0.063) G_GAN: 0.696 G_GAN_Feat: 0.719 G_ID: 0.084 G_Rec: 0.258 D_GP: 0.031 D_real: 1.548 D_fake: 0.316 +(epoch: 467, iters: 878, time: 0.063) G_GAN: 0.449 G_GAN_Feat: 0.965 G_ID: 0.112 G_Rec: 0.465 D_GP: 0.030 D_real: 1.055 D_fake: 0.557 +(epoch: 467, iters: 1278, time: 0.063) G_GAN: 0.267 G_GAN_Feat: 0.808 G_ID: 0.113 G_Rec: 0.320 D_GP: 0.034 D_real: 0.969 D_fake: 0.733 +(epoch: 467, iters: 1678, time: 0.064) G_GAN: 0.545 G_GAN_Feat: 1.066 G_ID: 0.110 G_Rec: 0.472 D_GP: 0.032 D_real: 1.127 D_fake: 0.460 +(epoch: 467, iters: 2078, time: 0.063) G_GAN: 0.309 G_GAN_Feat: 0.740 G_ID: 0.087 G_Rec: 0.287 D_GP: 0.035 D_real: 1.276 D_fake: 0.691 +(epoch: 467, iters: 2478, time: 0.064) G_GAN: 0.608 G_GAN_Feat: 0.952 G_ID: 0.127 G_Rec: 0.412 D_GP: 0.034 D_real: 1.346 D_fake: 0.415 +(epoch: 467, iters: 2878, time: 0.064) G_GAN: 0.137 G_GAN_Feat: 0.965 G_ID: 0.119 G_Rec: 0.340 D_GP: 0.149 D_real: 0.378 D_fake: 0.864 +(epoch: 467, iters: 3278, time: 0.064) G_GAN: 0.820 G_GAN_Feat: 0.962 G_ID: 0.122 G_Rec: 0.427 D_GP: 0.034 D_real: 1.334 D_fake: 0.272 +(epoch: 467, iters: 3678, time: 0.064) G_GAN: 0.282 G_GAN_Feat: 1.008 G_ID: 0.108 G_Rec: 0.349 D_GP: 0.056 D_real: 0.330 D_fake: 0.719 +(epoch: 467, iters: 4078, time: 0.064) G_GAN: 0.589 G_GAN_Feat: 1.235 G_ID: 0.112 G_Rec: 0.426 D_GP: 0.035 D_real: 0.378 D_fake: 0.425 +(epoch: 467, iters: 4478, time: 0.064) G_GAN: 0.377 G_GAN_Feat: 0.914 G_ID: 0.105 G_Rec: 0.361 D_GP: 0.076 D_real: 0.767 D_fake: 0.790 +(epoch: 467, iters: 4878, time: 0.064) G_GAN: 0.561 G_GAN_Feat: 0.953 G_ID: 0.106 G_Rec: 0.437 D_GP: 0.030 D_real: 1.169 D_fake: 0.463 +(epoch: 467, iters: 5278, time: 0.064) G_GAN: 0.378 G_GAN_Feat: 0.589 G_ID: 0.093 G_Rec: 0.265 D_GP: 0.024 D_real: 1.313 D_fake: 0.624 +(epoch: 467, iters: 5678, time: 0.063) G_GAN: 0.441 G_GAN_Feat: 0.872 G_ID: 0.114 G_Rec: 0.396 D_GP: 0.030 D_real: 1.129 D_fake: 0.566 +(epoch: 467, iters: 6078, time: 0.063) G_GAN: 0.272 G_GAN_Feat: 0.639 G_ID: 0.100 G_Rec: 0.290 D_GP: 0.027 D_real: 1.209 D_fake: 0.731 +(epoch: 467, iters: 6478, time: 0.064) G_GAN: 0.226 G_GAN_Feat: 0.926 G_ID: 0.107 G_Rec: 0.442 D_GP: 0.040 D_real: 0.769 D_fake: 0.778 +(epoch: 467, iters: 6878, time: 0.064) G_GAN: 0.018 G_GAN_Feat: 0.687 G_ID: 0.086 G_Rec: 0.275 D_GP: 0.037 D_real: 0.900 D_fake: 0.982 +(epoch: 467, iters: 7278, time: 0.064) G_GAN: 0.678 G_GAN_Feat: 0.975 G_ID: 0.112 G_Rec: 0.453 D_GP: 0.035 D_real: 1.317 D_fake: 0.360 +(epoch: 467, iters: 7678, time: 0.064) G_GAN: -0.001 G_GAN_Feat: 0.796 G_ID: 0.157 G_Rec: 0.318 D_GP: 0.034 D_real: 0.829 D_fake: 1.001 +(epoch: 467, iters: 8078, time: 0.064) G_GAN: 0.356 G_GAN_Feat: 0.930 G_ID: 0.115 G_Rec: 0.421 D_GP: 0.061 D_real: 0.845 D_fake: 0.649 +(epoch: 467, iters: 8478, time: 0.064) G_GAN: 0.437 G_GAN_Feat: 0.781 G_ID: 0.082 G_Rec: 0.294 D_GP: 0.072 D_real: 0.888 D_fake: 0.565 +(epoch: 468, iters: 270, time: 0.064) G_GAN: 0.454 G_GAN_Feat: 0.993 G_ID: 0.107 G_Rec: 0.485 D_GP: 0.032 D_real: 1.116 D_fake: 0.551 +(epoch: 468, iters: 670, time: 0.064) G_GAN: 0.193 G_GAN_Feat: 0.918 G_ID: 0.103 G_Rec: 0.334 D_GP: 0.095 D_real: 0.377 D_fake: 0.808 +(epoch: 468, iters: 1070, time: 0.064) G_GAN: 0.980 G_GAN_Feat: 1.116 G_ID: 0.111 G_Rec: 0.498 D_GP: 0.096 D_real: 0.955 D_fake: 0.154 +(epoch: 468, iters: 1470, time: 0.064) G_GAN: -0.035 G_GAN_Feat: 0.903 G_ID: 0.114 G_Rec: 0.369 D_GP: 0.137 D_real: 0.414 D_fake: 1.035 +(epoch: 468, iters: 1870, time: 0.064) G_GAN: 0.687 G_GAN_Feat: 1.186 G_ID: 0.108 G_Rec: 0.451 D_GP: 0.103 D_real: 1.061 D_fake: 0.382 +(epoch: 468, iters: 2270, time: 0.064) G_GAN: 0.284 G_GAN_Feat: 0.732 G_ID: 0.100 G_Rec: 0.283 D_GP: 0.034 D_real: 1.210 D_fake: 0.717 +(epoch: 468, iters: 2670, time: 0.064) G_GAN: 0.873 G_GAN_Feat: 1.053 G_ID: 0.111 G_Rec: 0.494 D_GP: 0.035 D_real: 1.337 D_fake: 0.215 +(epoch: 468, iters: 3070, time: 0.064) G_GAN: 0.280 G_GAN_Feat: 0.840 G_ID: 0.102 G_Rec: 0.315 D_GP: 0.111 D_real: 0.995 D_fake: 0.723 +(epoch: 468, iters: 3470, time: 0.063) G_GAN: 0.398 G_GAN_Feat: 1.071 G_ID: 0.124 G_Rec: 0.456 D_GP: 0.039 D_real: 0.818 D_fake: 0.603 +(epoch: 468, iters: 3870, time: 0.064) G_GAN: -0.048 G_GAN_Feat: 0.727 G_ID: 0.098 G_Rec: 0.315 D_GP: 0.034 D_real: 0.784 D_fake: 1.048 +(epoch: 468, iters: 4270, time: 0.064) G_GAN: 0.281 G_GAN_Feat: 0.937 G_ID: 0.118 G_Rec: 0.434 D_GP: 0.036 D_real: 0.971 D_fake: 0.719 +(epoch: 468, iters: 4670, time: 0.063) G_GAN: 0.175 G_GAN_Feat: 0.715 G_ID: 0.090 G_Rec: 0.294 D_GP: 0.031 D_real: 1.027 D_fake: 0.825 +(epoch: 468, iters: 5070, time: 0.063) G_GAN: 0.499 G_GAN_Feat: 1.058 G_ID: 0.116 G_Rec: 0.469 D_GP: 0.049 D_real: 0.843 D_fake: 0.508 +(epoch: 468, iters: 5470, time: 0.064) G_GAN: 0.173 G_GAN_Feat: 0.740 G_ID: 0.107 G_Rec: 0.287 D_GP: 0.057 D_real: 0.840 D_fake: 0.827 +(epoch: 468, iters: 5870, time: 0.064) G_GAN: 0.490 G_GAN_Feat: 0.838 G_ID: 0.115 G_Rec: 0.434 D_GP: 0.026 D_real: 1.266 D_fake: 0.514 +(epoch: 468, iters: 6270, time: 0.063) G_GAN: 0.289 G_GAN_Feat: 0.651 G_ID: 0.093 G_Rec: 0.299 D_GP: 0.027 D_real: 1.181 D_fake: 0.715 +(epoch: 468, iters: 6670, time: 0.064) G_GAN: 0.646 G_GAN_Feat: 0.922 G_ID: 0.110 G_Rec: 0.476 D_GP: 0.032 D_real: 1.282 D_fake: 0.405 +(epoch: 468, iters: 7070, time: 0.063) G_GAN: 0.191 G_GAN_Feat: 0.720 G_ID: 0.101 G_Rec: 0.359 D_GP: 0.045 D_real: 1.023 D_fake: 0.811 +(epoch: 468, iters: 7470, time: 0.064) G_GAN: 0.393 G_GAN_Feat: 0.949 G_ID: 0.101 G_Rec: 0.446 D_GP: 0.039 D_real: 1.074 D_fake: 0.609 +(epoch: 468, iters: 7870, time: 0.064) G_GAN: 0.136 G_GAN_Feat: 0.745 G_ID: 0.080 G_Rec: 0.354 D_GP: 0.039 D_real: 0.917 D_fake: 0.864 +(epoch: 468, iters: 8270, time: 0.064) G_GAN: 0.435 G_GAN_Feat: 0.941 G_ID: 0.137 G_Rec: 0.417 D_GP: 0.036 D_real: 1.034 D_fake: 0.566 +(epoch: 468, iters: 8670, time: 0.064) G_GAN: 0.224 G_GAN_Feat: 0.829 G_ID: 0.097 G_Rec: 0.324 D_GP: 0.085 D_real: 0.845 D_fake: 0.777 +(epoch: 469, iters: 462, time: 0.064) G_GAN: 0.561 G_GAN_Feat: 0.969 G_ID: 0.116 G_Rec: 0.484 D_GP: 0.031 D_real: 1.136 D_fake: 0.447 +(epoch: 469, iters: 862, time: 0.064) G_GAN: 0.157 G_GAN_Feat: 0.785 G_ID: 0.132 G_Rec: 0.308 D_GP: 0.049 D_real: 0.839 D_fake: 0.844 +(epoch: 469, iters: 1262, time: 0.064) G_GAN: 0.756 G_GAN_Feat: 1.061 G_ID: 0.116 G_Rec: 0.423 D_GP: 0.048 D_real: 1.180 D_fake: 0.360 +(epoch: 469, iters: 1662, time: 0.064) G_GAN: 0.066 G_GAN_Feat: 0.656 G_ID: 0.125 G_Rec: 0.338 D_GP: 0.028 D_real: 0.969 D_fake: 0.934 +(epoch: 469, iters: 2062, time: 0.064) G_GAN: 0.278 G_GAN_Feat: 0.786 G_ID: 0.104 G_Rec: 0.373 D_GP: 0.031 D_real: 1.027 D_fake: 0.726 +(epoch: 469, iters: 2462, time: 0.064) G_GAN: -0.207 G_GAN_Feat: 0.616 G_ID: 0.093 G_Rec: 0.294 D_GP: 0.030 D_real: 0.755 D_fake: 1.207 +(epoch: 469, iters: 2862, time: 0.064) G_GAN: 0.316 G_GAN_Feat: 0.830 G_ID: 0.118 G_Rec: 0.443 D_GP: 0.029 D_real: 0.944 D_fake: 0.705 +(epoch: 469, iters: 3262, time: 0.064) G_GAN: -0.113 G_GAN_Feat: 0.640 G_ID: 0.097 G_Rec: 0.302 D_GP: 0.029 D_real: 0.750 D_fake: 1.113 +(epoch: 469, iters: 3662, time: 0.064) G_GAN: 0.321 G_GAN_Feat: 0.868 G_ID: 0.112 G_Rec: 0.431 D_GP: 0.039 D_real: 0.962 D_fake: 0.698 +(epoch: 469, iters: 4062, time: 0.064) G_GAN: 0.051 G_GAN_Feat: 0.630 G_ID: 0.097 G_Rec: 0.285 D_GP: 0.040 D_real: 0.864 D_fake: 0.949 +(epoch: 469, iters: 4462, time: 0.063) G_GAN: 0.351 G_GAN_Feat: 0.838 G_ID: 0.113 G_Rec: 0.387 D_GP: 0.036 D_real: 1.030 D_fake: 0.659 +(epoch: 469, iters: 4862, time: 0.064) G_GAN: -0.032 G_GAN_Feat: 0.689 G_ID: 0.118 G_Rec: 0.309 D_GP: 0.047 D_real: 0.757 D_fake: 1.032 +(epoch: 469, iters: 5262, time: 0.064) G_GAN: 0.161 G_GAN_Feat: 0.974 G_ID: 0.118 G_Rec: 0.455 D_GP: 0.050 D_real: 0.687 D_fake: 0.840 +(epoch: 469, iters: 5662, time: 0.064) G_GAN: 0.174 G_GAN_Feat: 0.684 G_ID: 0.101 G_Rec: 0.322 D_GP: 0.037 D_real: 0.970 D_fake: 0.826 +(epoch: 469, iters: 6062, time: 0.064) G_GAN: 0.160 G_GAN_Feat: 0.954 G_ID: 0.125 G_Rec: 0.466 D_GP: 0.064 D_real: 0.788 D_fake: 0.841 +(epoch: 469, iters: 6462, time: 0.064) G_GAN: 0.224 G_GAN_Feat: 0.799 G_ID: 0.099 G_Rec: 0.349 D_GP: 0.056 D_real: 0.848 D_fake: 0.778 +(epoch: 469, iters: 6862, time: 0.064) G_GAN: 0.331 G_GAN_Feat: 0.888 G_ID: 0.130 G_Rec: 0.423 D_GP: 0.035 D_real: 0.931 D_fake: 0.673 +(epoch: 469, iters: 7262, time: 0.064) G_GAN: 0.220 G_GAN_Feat: 0.709 G_ID: 0.096 G_Rec: 0.317 D_GP: 0.034 D_real: 1.105 D_fake: 0.780 +(epoch: 469, iters: 7662, time: 0.064) G_GAN: 0.562 G_GAN_Feat: 0.946 G_ID: 0.105 G_Rec: 0.426 D_GP: 0.037 D_real: 1.113 D_fake: 0.448 +(epoch: 469, iters: 8062, time: 0.064) G_GAN: -0.186 G_GAN_Feat: 0.749 G_ID: 0.096 G_Rec: 0.334 D_GP: 0.056 D_real: 0.795 D_fake: 1.186 +(epoch: 469, iters: 8462, time: 0.064) G_GAN: 0.293 G_GAN_Feat: 0.953 G_ID: 0.112 G_Rec: 0.443 D_GP: 0.077 D_real: 0.840 D_fake: 0.710 +(epoch: 470, iters: 254, time: 0.064) G_GAN: 0.179 G_GAN_Feat: 0.706 G_ID: 0.084 G_Rec: 0.290 D_GP: 0.034 D_real: 1.144 D_fake: 0.821 +(epoch: 470, iters: 654, time: 0.064) G_GAN: 0.335 G_GAN_Feat: 1.077 G_ID: 0.135 G_Rec: 0.487 D_GP: 0.185 D_real: 0.306 D_fake: 0.666 +(epoch: 470, iters: 1054, time: 0.064) G_GAN: 0.168 G_GAN_Feat: 0.731 G_ID: 0.108 G_Rec: 0.359 D_GP: 0.028 D_real: 1.080 D_fake: 0.832 +(epoch: 470, iters: 1454, time: 0.064) G_GAN: 0.235 G_GAN_Feat: 0.827 G_ID: 0.127 G_Rec: 0.389 D_GP: 0.033 D_real: 0.892 D_fake: 0.765 +(epoch: 470, iters: 1854, time: 0.064) G_GAN: 0.074 G_GAN_Feat: 0.706 G_ID: 0.106 G_Rec: 0.298 D_GP: 0.036 D_real: 0.861 D_fake: 0.926 +(epoch: 470, iters: 2254, time: 0.064) G_GAN: 0.107 G_GAN_Feat: 0.962 G_ID: 0.117 G_Rec: 0.445 D_GP: 0.041 D_real: 0.798 D_fake: 0.893 +(epoch: 470, iters: 2654, time: 0.064) G_GAN: 0.108 G_GAN_Feat: 0.750 G_ID: 0.093 G_Rec: 0.330 D_GP: 0.109 D_real: 0.850 D_fake: 0.892 +(epoch: 470, iters: 3054, time: 0.064) G_GAN: 0.302 G_GAN_Feat: 0.905 G_ID: 0.142 G_Rec: 0.399 D_GP: 0.051 D_real: 0.837 D_fake: 0.698 +(epoch: 470, iters: 3454, time: 0.064) G_GAN: 0.169 G_GAN_Feat: 0.679 G_ID: 0.090 G_Rec: 0.281 D_GP: 0.030 D_real: 1.079 D_fake: 0.832 +(epoch: 470, iters: 3854, time: 0.064) G_GAN: 0.486 G_GAN_Feat: 0.883 G_ID: 0.119 G_Rec: 0.405 D_GP: 0.038 D_real: 1.088 D_fake: 0.526 +(epoch: 470, iters: 4254, time: 0.064) G_GAN: 0.013 G_GAN_Feat: 0.951 G_ID: 0.121 G_Rec: 0.337 D_GP: 0.483 D_real: 0.217 D_fake: 0.990 +(epoch: 470, iters: 4654, time: 0.064) G_GAN: 0.684 G_GAN_Feat: 1.056 G_ID: 0.127 G_Rec: 0.463 D_GP: 0.093 D_real: 0.454 D_fake: 0.416 +(epoch: 470, iters: 5054, time: 0.064) G_GAN: 0.238 G_GAN_Feat: 0.628 G_ID: 0.106 G_Rec: 0.323 D_GP: 0.025 D_real: 1.168 D_fake: 0.768 +(epoch: 470, iters: 5454, time: 0.064) G_GAN: 0.461 G_GAN_Feat: 0.825 G_ID: 0.128 G_Rec: 0.418 D_GP: 0.028 D_real: 1.213 D_fake: 0.546 +(epoch: 470, iters: 5854, time: 0.064) G_GAN: 0.155 G_GAN_Feat: 0.607 G_ID: 0.093 G_Rec: 0.291 D_GP: 0.024 D_real: 1.102 D_fake: 0.845 +(epoch: 470, iters: 6254, time: 0.064) G_GAN: 0.264 G_GAN_Feat: 0.797 G_ID: 0.125 G_Rec: 0.390 D_GP: 0.025 D_real: 1.041 D_fake: 0.736 +(epoch: 470, iters: 6654, time: 0.064) G_GAN: 0.086 G_GAN_Feat: 0.633 G_ID: 0.101 G_Rec: 0.293 D_GP: 0.029 D_real: 1.050 D_fake: 0.914 +(epoch: 470, iters: 7054, time: 0.064) G_GAN: 0.525 G_GAN_Feat: 0.882 G_ID: 0.098 G_Rec: 0.431 D_GP: 0.035 D_real: 1.153 D_fake: 0.491 +(epoch: 470, iters: 7454, time: 0.064) G_GAN: 0.134 G_GAN_Feat: 0.639 G_ID: 0.087 G_Rec: 0.292 D_GP: 0.039 D_real: 1.008 D_fake: 0.866 +(epoch: 470, iters: 7854, time: 0.064) G_GAN: 0.185 G_GAN_Feat: 0.850 G_ID: 0.127 G_Rec: 0.406 D_GP: 0.043 D_real: 0.858 D_fake: 0.815 +(epoch: 470, iters: 8254, time: 0.064) G_GAN: 0.091 G_GAN_Feat: 0.641 G_ID: 0.108 G_Rec: 0.290 D_GP: 0.035 D_real: 0.954 D_fake: 0.909 +(epoch: 470, iters: 8654, time: 0.064) G_GAN: 0.211 G_GAN_Feat: 0.888 G_ID: 0.118 G_Rec: 0.439 D_GP: 0.039 D_real: 0.792 D_fake: 0.791 +(epoch: 471, iters: 446, time: 0.064) G_GAN: 0.193 G_GAN_Feat: 0.674 G_ID: 0.097 G_Rec: 0.293 D_GP: 0.036 D_real: 1.029 D_fake: 0.809 +(epoch: 471, iters: 846, time: 0.064) G_GAN: 0.368 G_GAN_Feat: 0.890 G_ID: 0.117 G_Rec: 0.425 D_GP: 0.038 D_real: 0.973 D_fake: 0.639 +(epoch: 471, iters: 1246, time: 0.064) G_GAN: 0.047 G_GAN_Feat: 0.695 G_ID: 0.082 G_Rec: 0.296 D_GP: 0.082 D_real: 0.851 D_fake: 0.953 +(epoch: 471, iters: 1646, time: 0.064) G_GAN: 0.194 G_GAN_Feat: 0.880 G_ID: 0.125 G_Rec: 0.387 D_GP: 0.043 D_real: 0.857 D_fake: 0.807 +(epoch: 471, iters: 2046, time: 0.064) G_GAN: -0.042 G_GAN_Feat: 0.850 G_ID: 0.101 G_Rec: 0.337 D_GP: 0.070 D_real: 0.844 D_fake: 1.043 +(epoch: 471, iters: 2446, time: 0.064) G_GAN: 0.136 G_GAN_Feat: 0.881 G_ID: 0.138 G_Rec: 0.415 D_GP: 0.032 D_real: 0.828 D_fake: 0.864 +(epoch: 471, iters: 2846, time: 0.064) G_GAN: 0.002 G_GAN_Feat: 0.679 G_ID: 0.101 G_Rec: 0.284 D_GP: 0.044 D_real: 0.818 D_fake: 0.998 +(epoch: 471, iters: 3246, time: 0.063) G_GAN: 0.486 G_GAN_Feat: 0.881 G_ID: 0.109 G_Rec: 0.389 D_GP: 0.037 D_real: 1.185 D_fake: 0.533 +(epoch: 471, iters: 3646, time: 0.064) G_GAN: 0.203 G_GAN_Feat: 0.654 G_ID: 0.090 G_Rec: 0.274 D_GP: 0.031 D_real: 1.106 D_fake: 0.797 +(epoch: 471, iters: 4046, time: 0.064) G_GAN: 0.608 G_GAN_Feat: 0.870 G_ID: 0.103 G_Rec: 0.389 D_GP: 0.032 D_real: 1.370 D_fake: 0.404 +(epoch: 471, iters: 4446, time: 0.063) G_GAN: -0.100 G_GAN_Feat: 0.846 G_ID: 0.096 G_Rec: 0.362 D_GP: 0.193 D_real: 0.666 D_fake: 1.102 +(epoch: 471, iters: 4846, time: 0.064) G_GAN: 0.390 G_GAN_Feat: 0.910 G_ID: 0.120 G_Rec: 0.451 D_GP: 0.031 D_real: 1.006 D_fake: 0.621 +(epoch: 471, iters: 5246, time: 0.064) G_GAN: 0.193 G_GAN_Feat: 0.641 G_ID: 0.078 G_Rec: 0.274 D_GP: 0.030 D_real: 1.097 D_fake: 0.807 +(epoch: 471, iters: 5646, time: 0.064) G_GAN: 0.691 G_GAN_Feat: 0.931 G_ID: 0.110 G_Rec: 0.442 D_GP: 0.032 D_real: 1.248 D_fake: 0.345 +(epoch: 471, iters: 6046, time: 0.063) G_GAN: 0.224 G_GAN_Feat: 0.678 G_ID: 0.094 G_Rec: 0.290 D_GP: 0.028 D_real: 1.074 D_fake: 0.776 +(epoch: 471, iters: 6446, time: 0.063) G_GAN: 0.530 G_GAN_Feat: 0.897 G_ID: 0.106 G_Rec: 0.408 D_GP: 0.031 D_real: 1.215 D_fake: 0.476 +(epoch: 471, iters: 6846, time: 0.064) G_GAN: 0.282 G_GAN_Feat: 0.781 G_ID: 0.090 G_Rec: 0.315 D_GP: 0.046 D_real: 1.059 D_fake: 0.719 +(epoch: 471, iters: 7246, time: 0.064) G_GAN: 0.102 G_GAN_Feat: 0.924 G_ID: 0.154 G_Rec: 0.430 D_GP: 0.033 D_real: 0.782 D_fake: 0.898 +(epoch: 471, iters: 7646, time: 0.063) G_GAN: 0.281 G_GAN_Feat: 0.702 G_ID: 0.115 G_Rec: 0.301 D_GP: 0.034 D_real: 1.153 D_fake: 0.725 +(epoch: 471, iters: 8046, time: 0.064) G_GAN: 0.514 G_GAN_Feat: 0.921 G_ID: 0.116 G_Rec: 0.441 D_GP: 0.037 D_real: 1.176 D_fake: 0.499 +(epoch: 471, iters: 8446, time: 0.064) G_GAN: 0.162 G_GAN_Feat: 0.664 G_ID: 0.095 G_Rec: 0.294 D_GP: 0.035 D_real: 1.006 D_fake: 0.839 +(epoch: 472, iters: 238, time: 0.064) G_GAN: 0.277 G_GAN_Feat: 1.018 G_ID: 0.125 G_Rec: 0.448 D_GP: 0.070 D_real: 0.821 D_fake: 0.724 +(epoch: 472, iters: 638, time: 0.063) G_GAN: 0.085 G_GAN_Feat: 0.650 G_ID: 0.102 G_Rec: 0.268 D_GP: 0.037 D_real: 1.019 D_fake: 0.915 +(epoch: 472, iters: 1038, time: 0.063) G_GAN: 0.523 G_GAN_Feat: 0.970 G_ID: 0.138 G_Rec: 0.409 D_GP: 0.037 D_real: 1.050 D_fake: 0.487 +(epoch: 472, iters: 1438, time: 0.064) G_GAN: 0.259 G_GAN_Feat: 0.788 G_ID: 0.109 G_Rec: 0.334 D_GP: 0.044 D_real: 0.972 D_fake: 0.742 +(epoch: 472, iters: 1838, time: 0.063) G_GAN: 0.354 G_GAN_Feat: 1.044 G_ID: 0.136 G_Rec: 0.422 D_GP: 0.151 D_real: 0.395 D_fake: 0.649 +(epoch: 472, iters: 2238, time: 0.063) G_GAN: 0.050 G_GAN_Feat: 0.755 G_ID: 0.109 G_Rec: 0.330 D_GP: 0.032 D_real: 0.973 D_fake: 0.951 +(epoch: 472, iters: 2638, time: 0.064) G_GAN: 0.275 G_GAN_Feat: 0.996 G_ID: 0.134 G_Rec: 0.494 D_GP: 0.048 D_real: 0.701 D_fake: 0.737 +(epoch: 472, iters: 3038, time: 0.064) G_GAN: 0.041 G_GAN_Feat: 0.779 G_ID: 0.098 G_Rec: 0.329 D_GP: 0.058 D_real: 0.786 D_fake: 0.960 +(epoch: 472, iters: 3438, time: 0.063) G_GAN: 0.734 G_GAN_Feat: 1.099 G_ID: 0.112 G_Rec: 0.464 D_GP: 0.039 D_real: 1.296 D_fake: 0.306 +(epoch: 472, iters: 3838, time: 0.063) G_GAN: -0.092 G_GAN_Feat: 0.850 G_ID: 0.091 G_Rec: 0.347 D_GP: 0.063 D_real: 0.405 D_fake: 1.092 +(epoch: 472, iters: 4238, time: 0.063) G_GAN: 0.788 G_GAN_Feat: 0.992 G_ID: 0.115 G_Rec: 0.433 D_GP: 0.057 D_real: 1.053 D_fake: 0.268 +(epoch: 472, iters: 4638, time: 0.064) G_GAN: 0.082 G_GAN_Feat: 0.783 G_ID: 0.097 G_Rec: 0.337 D_GP: 0.044 D_real: 0.774 D_fake: 0.918 +(epoch: 472, iters: 5038, time: 0.064) G_GAN: 0.362 G_GAN_Feat: 0.969 G_ID: 0.135 G_Rec: 0.450 D_GP: 0.041 D_real: 0.893 D_fake: 0.640 +(epoch: 472, iters: 5438, time: 0.063) G_GAN: 0.405 G_GAN_Feat: 0.804 G_ID: 0.114 G_Rec: 0.314 D_GP: 0.036 D_real: 1.112 D_fake: 0.596 +(epoch: 472, iters: 5838, time: 0.064) G_GAN: 0.604 G_GAN_Feat: 1.121 G_ID: 0.110 G_Rec: 0.424 D_GP: 0.041 D_real: 0.642 D_fake: 0.406 +(epoch: 472, iters: 6238, time: 0.064) G_GAN: 0.018 G_GAN_Feat: 0.699 G_ID: 0.106 G_Rec: 0.334 D_GP: 0.030 D_real: 0.934 D_fake: 0.984 +(epoch: 472, iters: 6638, time: 0.064) G_GAN: 0.126 G_GAN_Feat: 0.937 G_ID: 0.119 G_Rec: 0.417 D_GP: 0.046 D_real: 0.677 D_fake: 0.878 +(epoch: 472, iters: 7038, time: 0.063) G_GAN: -0.155 G_GAN_Feat: 0.699 G_ID: 0.100 G_Rec: 0.289 D_GP: 0.033 D_real: 0.838 D_fake: 1.155 +(epoch: 472, iters: 7438, time: 0.064) G_GAN: 0.309 G_GAN_Feat: 0.914 G_ID: 0.109 G_Rec: 0.387 D_GP: 0.034 D_real: 1.107 D_fake: 0.692 +(epoch: 472, iters: 7838, time: 0.064) G_GAN: 0.161 G_GAN_Feat: 0.756 G_ID: 0.107 G_Rec: 0.294 D_GP: 0.044 D_real: 1.012 D_fake: 0.839 +(epoch: 472, iters: 8238, time: 0.063) G_GAN: 0.682 G_GAN_Feat: 1.063 G_ID: 0.115 G_Rec: 0.476 D_GP: 0.062 D_real: 0.994 D_fake: 0.348 +(epoch: 472, iters: 8638, time: 0.063) G_GAN: 0.484 G_GAN_Feat: 0.855 G_ID: 0.110 G_Rec: 0.310 D_GP: 0.122 D_real: 0.763 D_fake: 0.530 +(epoch: 473, iters: 430, time: 0.063) G_GAN: 0.739 G_GAN_Feat: 1.186 G_ID: 0.104 G_Rec: 0.479 D_GP: 0.040 D_real: 0.778 D_fake: 0.317 +(epoch: 473, iters: 830, time: 0.064) G_GAN: 0.353 G_GAN_Feat: 0.868 G_ID: 0.103 G_Rec: 0.309 D_GP: 0.092 D_real: 0.787 D_fake: 0.647 +(epoch: 473, iters: 1230, time: 0.063) G_GAN: 0.447 G_GAN_Feat: 0.934 G_ID: 0.111 G_Rec: 0.472 D_GP: 0.031 D_real: 1.143 D_fake: 0.567 +(epoch: 473, iters: 1630, time: 0.064) G_GAN: 0.193 G_GAN_Feat: 0.662 G_ID: 0.100 G_Rec: 0.302 D_GP: 0.026 D_real: 1.151 D_fake: 0.808 +(epoch: 473, iters: 2030, time: 0.064) G_GAN: 0.376 G_GAN_Feat: 0.880 G_ID: 0.104 G_Rec: 0.419 D_GP: 0.033 D_real: 1.145 D_fake: 0.628 +(epoch: 473, iters: 2430, time: 0.064) G_GAN: 0.211 G_GAN_Feat: 0.703 G_ID: 0.105 G_Rec: 0.318 D_GP: 0.040 D_real: 0.983 D_fake: 0.790 +(epoch: 473, iters: 2830, time: 0.064) G_GAN: 0.280 G_GAN_Feat: 0.922 G_ID: 0.120 G_Rec: 0.463 D_GP: 0.049 D_real: 0.808 D_fake: 0.726 +(epoch: 473, iters: 3230, time: 0.064) G_GAN: 0.215 G_GAN_Feat: 0.671 G_ID: 0.085 G_Rec: 0.322 D_GP: 0.040 D_real: 1.131 D_fake: 0.786 +(epoch: 473, iters: 3630, time: 0.063) G_GAN: 0.372 G_GAN_Feat: 0.973 G_ID: 0.106 G_Rec: 0.433 D_GP: 0.052 D_real: 0.906 D_fake: 0.639 +(epoch: 473, iters: 4030, time: 0.064) G_GAN: 0.047 G_GAN_Feat: 0.662 G_ID: 0.099 G_Rec: 0.292 D_GP: 0.039 D_real: 0.934 D_fake: 0.953 +(epoch: 473, iters: 4430, time: 0.064) G_GAN: 0.350 G_GAN_Feat: 1.006 G_ID: 0.119 G_Rec: 0.473 D_GP: 0.074 D_real: 0.939 D_fake: 0.661 +(epoch: 473, iters: 4830, time: 0.064) G_GAN: 0.100 G_GAN_Feat: 0.702 G_ID: 0.103 G_Rec: 0.290 D_GP: 0.040 D_real: 0.895 D_fake: 0.900 +(epoch: 473, iters: 5230, time: 0.064) G_GAN: 0.320 G_GAN_Feat: 0.951 G_ID: 0.122 G_Rec: 0.410 D_GP: 0.054 D_real: 0.901 D_fake: 0.689 +(epoch: 473, iters: 5630, time: 0.064) G_GAN: 0.239 G_GAN_Feat: 0.716 G_ID: 0.085 G_Rec: 0.285 D_GP: 0.050 D_real: 0.993 D_fake: 0.762 +(epoch: 473, iters: 6030, time: 0.064) G_GAN: 0.415 G_GAN_Feat: 1.015 G_ID: 0.108 G_Rec: 0.421 D_GP: 0.064 D_real: 0.855 D_fake: 0.592 +(epoch: 473, iters: 6430, time: 0.064) G_GAN: 0.170 G_GAN_Feat: 0.768 G_ID: 0.109 G_Rec: 0.302 D_GP: 0.048 D_real: 0.963 D_fake: 0.830 +(epoch: 473, iters: 6830, time: 0.063) G_GAN: 0.101 G_GAN_Feat: 1.208 G_ID: 0.109 G_Rec: 0.494 D_GP: 4.292 D_real: 0.437 D_fake: 0.930 +(epoch: 473, iters: 7230, time: 0.064) G_GAN: 0.197 G_GAN_Feat: 0.626 G_ID: 0.098 G_Rec: 0.286 D_GP: 0.028 D_real: 1.154 D_fake: 0.804 +(epoch: 473, iters: 7630, time: 0.064) G_GAN: 0.115 G_GAN_Feat: 0.793 G_ID: 0.118 G_Rec: 0.391 D_GP: 0.036 D_real: 0.812 D_fake: 0.886 +(epoch: 473, iters: 8030, time: 0.064) G_GAN: 0.056 G_GAN_Feat: 0.636 G_ID: 0.086 G_Rec: 0.300 D_GP: 0.027 D_real: 1.040 D_fake: 0.944 +(epoch: 473, iters: 8430, time: 0.064) G_GAN: 0.062 G_GAN_Feat: 0.873 G_ID: 0.101 G_Rec: 0.427 D_GP: 0.054 D_real: 0.569 D_fake: 0.941 +(epoch: 474, iters: 222, time: 0.064) G_GAN: -0.104 G_GAN_Feat: 0.715 G_ID: 0.094 G_Rec: 0.315 D_GP: 0.072 D_real: 0.684 D_fake: 1.105 +(epoch: 474, iters: 622, time: 0.064) G_GAN: 0.132 G_GAN_Feat: 0.896 G_ID: 0.113 G_Rec: 0.417 D_GP: 0.036 D_real: 0.687 D_fake: 0.868 +(epoch: 474, iters: 1022, time: 0.064) G_GAN: 0.096 G_GAN_Feat: 0.706 G_ID: 0.097 G_Rec: 0.321 D_GP: 0.041 D_real: 0.850 D_fake: 0.904 +(epoch: 474, iters: 1422, time: 0.064) G_GAN: 0.266 G_GAN_Feat: 0.901 G_ID: 0.104 G_Rec: 0.418 D_GP: 0.048 D_real: 0.759 D_fake: 0.738 +(epoch: 474, iters: 1822, time: 0.064) G_GAN: -0.061 G_GAN_Feat: 0.790 G_ID: 0.121 G_Rec: 0.322 D_GP: 0.067 D_real: 0.579 D_fake: 1.061 +(epoch: 474, iters: 2222, time: 0.064) G_GAN: 0.613 G_GAN_Feat: 0.973 G_ID: 0.122 G_Rec: 0.467 D_GP: 0.039 D_real: 1.159 D_fake: 0.412 +(epoch: 474, iters: 2622, time: 0.064) G_GAN: 0.261 G_GAN_Feat: 0.689 G_ID: 0.123 G_Rec: 0.272 D_GP: 0.043 D_real: 1.072 D_fake: 0.739 +(epoch: 474, iters: 3022, time: 0.063) G_GAN: 0.549 G_GAN_Feat: 0.944 G_ID: 0.139 G_Rec: 0.442 D_GP: 0.036 D_real: 1.018 D_fake: 0.459 +(epoch: 474, iters: 3422, time: 0.064) G_GAN: 0.413 G_GAN_Feat: 0.739 G_ID: 0.090 G_Rec: 0.323 D_GP: 0.042 D_real: 1.121 D_fake: 0.591 +(epoch: 474, iters: 3822, time: 0.063) G_GAN: 0.345 G_GAN_Feat: 1.057 G_ID: 0.120 G_Rec: 0.506 D_GP: 0.042 D_real: 1.081 D_fake: 0.679 +(epoch: 474, iters: 4222, time: 0.064) G_GAN: -0.026 G_GAN_Feat: 0.686 G_ID: 0.103 G_Rec: 0.324 D_GP: 0.031 D_real: 0.919 D_fake: 1.026 +(epoch: 474, iters: 4622, time: 0.063) G_GAN: -0.059 G_GAN_Feat: 0.939 G_ID: 0.138 G_Rec: 0.451 D_GP: 0.050 D_real: 0.665 D_fake: 1.059 +(epoch: 474, iters: 5022, time: 0.064) G_GAN: -0.027 G_GAN_Feat: 0.731 G_ID: 0.097 G_Rec: 0.325 D_GP: 0.065 D_real: 0.716 D_fake: 1.028 +(epoch: 474, iters: 5422, time: 0.063) G_GAN: 0.069 G_GAN_Feat: 0.822 G_ID: 0.162 G_Rec: 0.370 D_GP: 0.040 D_real: 0.668 D_fake: 0.931 +(epoch: 474, iters: 5822, time: 0.063) G_GAN: 0.012 G_GAN_Feat: 0.730 G_ID: 0.091 G_Rec: 0.303 D_GP: 0.042 D_real: 0.923 D_fake: 0.988 +(epoch: 474, iters: 6222, time: 0.064) G_GAN: 0.529 G_GAN_Feat: 0.952 G_ID: 0.107 G_Rec: 0.407 D_GP: 0.046 D_real: 1.174 D_fake: 0.517 +(epoch: 474, iters: 6622, time: 0.064) G_GAN: 0.245 G_GAN_Feat: 0.792 G_ID: 0.107 G_Rec: 0.346 D_GP: 0.050 D_real: 0.996 D_fake: 0.762 +(epoch: 474, iters: 7022, time: 0.064) G_GAN: 0.503 G_GAN_Feat: 0.892 G_ID: 0.103 G_Rec: 0.417 D_GP: 0.033 D_real: 1.050 D_fake: 0.510 +(epoch: 474, iters: 7422, time: 0.064) G_GAN: -0.088 G_GAN_Feat: 0.853 G_ID: 0.105 G_Rec: 0.355 D_GP: 0.105 D_real: 0.372 D_fake: 1.088 +(epoch: 474, iters: 7822, time: 0.064) G_GAN: 0.464 G_GAN_Feat: 0.983 G_ID: 0.111 G_Rec: 0.403 D_GP: 0.037 D_real: 0.882 D_fake: 0.549 +(epoch: 474, iters: 8222, time: 0.064) G_GAN: 0.049 G_GAN_Feat: 0.976 G_ID: 0.097 G_Rec: 0.365 D_GP: 1.100 D_real: 0.371 D_fake: 0.955 +(epoch: 474, iters: 8622, time: 0.064) G_GAN: 0.490 G_GAN_Feat: 0.956 G_ID: 0.108 G_Rec: 0.498 D_GP: 0.028 D_real: 1.058 D_fake: 0.549 +(epoch: 475, iters: 414, time: 0.064) G_GAN: -0.010 G_GAN_Feat: 0.752 G_ID: 0.092 G_Rec: 0.296 D_GP: 0.042 D_real: 0.700 D_fake: 1.010 +(epoch: 475, iters: 814, time: 0.064) G_GAN: -0.160 G_GAN_Feat: 0.994 G_ID: 0.125 G_Rec: 0.433 D_GP: 0.146 D_real: 0.351 D_fake: 1.161 +(epoch: 475, iters: 1214, time: 0.064) G_GAN: 0.160 G_GAN_Feat: 0.847 G_ID: 0.088 G_Rec: 0.363 D_GP: 0.132 D_real: 0.580 D_fake: 0.840 +(epoch: 475, iters: 1614, time: 0.063) G_GAN: 0.579 G_GAN_Feat: 0.999 G_ID: 0.120 G_Rec: 0.427 D_GP: 0.039 D_real: 0.931 D_fake: 0.443 +(epoch: 475, iters: 2014, time: 0.063) G_GAN: 0.226 G_GAN_Feat: 0.771 G_ID: 0.095 G_Rec: 0.284 D_GP: 0.033 D_real: 0.853 D_fake: 0.774 +(epoch: 475, iters: 2414, time: 0.064) G_GAN: 0.774 G_GAN_Feat: 1.023 G_ID: 0.136 G_Rec: 0.419 D_GP: 0.054 D_real: 1.159 D_fake: 0.287 +(epoch: 475, iters: 2814, time: 0.064) G_GAN: 0.142 G_GAN_Feat: 0.998 G_ID: 0.117 G_Rec: 0.367 D_GP: 0.685 D_real: 0.275 D_fake: 0.942 +(epoch: 475, iters: 3214, time: 0.063) G_GAN: 1.014 G_GAN_Feat: 1.042 G_ID: 0.127 G_Rec: 0.445 D_GP: 0.074 D_real: 1.303 D_fake: 0.113 +(epoch: 475, iters: 3614, time: 0.064) G_GAN: 0.037 G_GAN_Feat: 0.692 G_ID: 0.104 G_Rec: 0.303 D_GP: 0.032 D_real: 0.972 D_fake: 0.963 +(epoch: 475, iters: 4014, time: 0.063) G_GAN: -0.029 G_GAN_Feat: 0.915 G_ID: 0.143 G_Rec: 0.426 D_GP: 0.036 D_real: 0.667 D_fake: 1.031 +(epoch: 475, iters: 4414, time: 0.064) G_GAN: -0.022 G_GAN_Feat: 0.717 G_ID: 0.088 G_Rec: 0.319 D_GP: 0.069 D_real: 0.819 D_fake: 1.022 +(epoch: 475, iters: 4814, time: 0.063) G_GAN: -0.041 G_GAN_Feat: 0.963 G_ID: 0.117 G_Rec: 0.445 D_GP: 0.047 D_real: 0.593 D_fake: 1.041 +(epoch: 475, iters: 5214, time: 0.063) G_GAN: 0.008 G_GAN_Feat: 0.894 G_ID: 0.104 G_Rec: 0.342 D_GP: 0.059 D_real: 0.496 D_fake: 0.992 +(epoch: 475, iters: 5614, time: 0.064) G_GAN: 0.333 G_GAN_Feat: 0.928 G_ID: 0.157 G_Rec: 0.416 D_GP: 0.039 D_real: 0.965 D_fake: 0.671 +(epoch: 475, iters: 6014, time: 0.064) G_GAN: 0.370 G_GAN_Feat: 0.691 G_ID: 0.091 G_Rec: 0.276 D_GP: 0.037 D_real: 1.232 D_fake: 0.634 +(epoch: 475, iters: 6414, time: 0.064) G_GAN: 0.748 G_GAN_Feat: 0.953 G_ID: 0.117 G_Rec: 0.425 D_GP: 0.031 D_real: 1.278 D_fake: 0.301 +(epoch: 475, iters: 6814, time: 0.064) G_GAN: 0.280 G_GAN_Feat: 0.710 G_ID: 0.088 G_Rec: 0.291 D_GP: 0.029 D_real: 1.176 D_fake: 0.720 +(epoch: 475, iters: 7214, time: 0.064) G_GAN: 0.528 G_GAN_Feat: 1.020 G_ID: 0.112 G_Rec: 0.437 D_GP: 0.060 D_real: 1.020 D_fake: 0.478 +(epoch: 475, iters: 7614, time: 0.064) G_GAN: 0.302 G_GAN_Feat: 0.804 G_ID: 0.103 G_Rec: 0.343 D_GP: 0.041 D_real: 1.021 D_fake: 0.713 +(epoch: 475, iters: 8014, time: 0.064) G_GAN: 0.588 G_GAN_Feat: 0.969 G_ID: 0.097 G_Rec: 0.462 D_GP: 0.032 D_real: 1.213 D_fake: 0.441 +(epoch: 475, iters: 8414, time: 0.064) G_GAN: 0.059 G_GAN_Feat: 0.712 G_ID: 0.112 G_Rec: 0.286 D_GP: 0.030 D_real: 0.978 D_fake: 0.941 +(epoch: 476, iters: 206, time: 0.064) G_GAN: 0.474 G_GAN_Feat: 0.994 G_ID: 0.132 G_Rec: 0.419 D_GP: 0.039 D_real: 0.910 D_fake: 0.546 +(epoch: 476, iters: 606, time: 0.064) G_GAN: 0.227 G_GAN_Feat: 0.657 G_ID: 0.093 G_Rec: 0.277 D_GP: 0.027 D_real: 1.176 D_fake: 0.773 +(epoch: 476, iters: 1006, time: 0.064) G_GAN: 0.493 G_GAN_Feat: 1.029 G_ID: 0.110 G_Rec: 0.496 D_GP: 0.037 D_real: 1.066 D_fake: 0.526 +(epoch: 476, iters: 1406, time: 0.064) G_GAN: 0.164 G_GAN_Feat: 0.651 G_ID: 0.098 G_Rec: 0.273 D_GP: 0.029 D_real: 1.062 D_fake: 0.837 +(epoch: 476, iters: 1806, time: 0.064) G_GAN: 0.334 G_GAN_Feat: 1.023 G_ID: 0.110 G_Rec: 0.479 D_GP: 0.068 D_real: 0.710 D_fake: 0.667 +(epoch: 476, iters: 2206, time: 0.064) G_GAN: -0.134 G_GAN_Feat: 0.749 G_ID: 0.080 G_Rec: 0.309 D_GP: 0.194 D_real: 0.504 D_fake: 1.134 +(epoch: 476, iters: 2606, time: 0.064) G_GAN: 0.274 G_GAN_Feat: 1.013 G_ID: 0.110 G_Rec: 0.445 D_GP: 0.033 D_real: 0.822 D_fake: 0.726 +(epoch: 476, iters: 3006, time: 0.064) G_GAN: 0.088 G_GAN_Feat: 0.667 G_ID: 0.095 G_Rec: 0.263 D_GP: 0.033 D_real: 0.995 D_fake: 0.912 +(epoch: 476, iters: 3406, time: 0.064) G_GAN: 0.894 G_GAN_Feat: 0.981 G_ID: 0.109 G_Rec: 0.435 D_GP: 0.041 D_real: 1.557 D_fake: 0.272 +(epoch: 476, iters: 3806, time: 0.064) G_GAN: 0.059 G_GAN_Feat: 0.736 G_ID: 0.097 G_Rec: 0.299 D_GP: 0.040 D_real: 0.865 D_fake: 0.941 +(epoch: 476, iters: 4206, time: 0.064) G_GAN: 0.434 G_GAN_Feat: 0.950 G_ID: 0.129 G_Rec: 0.424 D_GP: 0.037 D_real: 1.120 D_fake: 0.569 +(epoch: 476, iters: 4606, time: 0.064) G_GAN: 0.259 G_GAN_Feat: 0.748 G_ID: 0.095 G_Rec: 0.280 D_GP: 0.039 D_real: 1.056 D_fake: 0.742 +(epoch: 476, iters: 5006, time: 0.064) G_GAN: 0.785 G_GAN_Feat: 1.064 G_ID: 0.101 G_Rec: 0.460 D_GP: 0.061 D_real: 0.905 D_fake: 0.254 +(epoch: 476, iters: 5406, time: 0.064) G_GAN: 0.396 G_GAN_Feat: 0.910 G_ID: 0.108 G_Rec: 0.359 D_GP: 0.168 D_real: 0.714 D_fake: 0.718 +(epoch: 476, iters: 5806, time: 0.064) G_GAN: 0.058 G_GAN_Feat: 0.866 G_ID: 0.127 G_Rec: 0.396 D_GP: 0.030 D_real: 0.844 D_fake: 0.943 +(epoch: 476, iters: 6206, time: 0.064) G_GAN: -0.141 G_GAN_Feat: 0.754 G_ID: 0.095 G_Rec: 0.321 D_GP: 0.047 D_real: 0.594 D_fake: 1.141 +(epoch: 476, iters: 6606, time: 0.064) G_GAN: 0.303 G_GAN_Feat: 1.067 G_ID: 0.117 G_Rec: 0.478 D_GP: 0.101 D_real: 0.482 D_fake: 0.701 +(epoch: 476, iters: 7006, time: 0.064) G_GAN: 0.624 G_GAN_Feat: 0.870 G_ID: 0.088 G_Rec: 0.349 D_GP: 0.045 D_real: 1.102 D_fake: 0.549 +(epoch: 476, iters: 7406, time: 0.064) G_GAN: 0.455 G_GAN_Feat: 0.998 G_ID: 0.116 G_Rec: 0.427 D_GP: 0.052 D_real: 1.289 D_fake: 0.560 +(epoch: 476, iters: 7806, time: 0.064) G_GAN: 0.291 G_GAN_Feat: 0.734 G_ID: 0.079 G_Rec: 0.328 D_GP: 0.036 D_real: 1.199 D_fake: 0.709 +(epoch: 476, iters: 8206, time: 0.064) G_GAN: 0.451 G_GAN_Feat: 0.987 G_ID: 0.113 G_Rec: 0.424 D_GP: 0.041 D_real: 0.956 D_fake: 0.556 +(epoch: 476, iters: 8606, time: 0.064) G_GAN: 0.348 G_GAN_Feat: 0.824 G_ID: 0.088 G_Rec: 0.317 D_GP: 0.034 D_real: 1.082 D_fake: 0.653 +(epoch: 477, iters: 398, time: 0.064) G_GAN: 0.628 G_GAN_Feat: 0.951 G_ID: 0.125 G_Rec: 0.446 D_GP: 0.043 D_real: 1.211 D_fake: 0.386 +(epoch: 477, iters: 798, time: 0.064) G_GAN: 0.067 G_GAN_Feat: 0.814 G_ID: 0.089 G_Rec: 0.335 D_GP: 0.085 D_real: 0.826 D_fake: 0.933 +(epoch: 477, iters: 1198, time: 0.064) G_GAN: 0.619 G_GAN_Feat: 0.953 G_ID: 0.101 G_Rec: 0.401 D_GP: 0.035 D_real: 1.220 D_fake: 0.425 +(epoch: 477, iters: 1598, time: 0.064) G_GAN: 0.279 G_GAN_Feat: 0.849 G_ID: 0.083 G_Rec: 0.333 D_GP: 0.035 D_real: 0.679 D_fake: 0.721 +(epoch: 477, iters: 1998, time: 0.064) G_GAN: 0.635 G_GAN_Feat: 1.021 G_ID: 0.106 G_Rec: 0.462 D_GP: 0.035 D_real: 1.171 D_fake: 0.433 +(epoch: 477, iters: 2398, time: 0.064) G_GAN: 0.477 G_GAN_Feat: 0.871 G_ID: 0.096 G_Rec: 0.301 D_GP: 0.057 D_real: 0.953 D_fake: 0.533 +(epoch: 477, iters: 2798, time: 0.064) G_GAN: 0.494 G_GAN_Feat: 1.052 G_ID: 0.104 G_Rec: 0.452 D_GP: 0.056 D_real: 0.860 D_fake: 0.525 +(epoch: 477, iters: 3198, time: 0.064) G_GAN: 0.495 G_GAN_Feat: 1.013 G_ID: 0.091 G_Rec: 0.359 D_GP: 0.897 D_real: 0.578 D_fake: 0.563 +(epoch: 477, iters: 3598, time: 0.064) G_GAN: 0.357 G_GAN_Feat: 0.988 G_ID: 0.105 G_Rec: 0.461 D_GP: 0.038 D_real: 0.829 D_fake: 0.648 +(epoch: 477, iters: 3998, time: 0.064) G_GAN: 0.071 G_GAN_Feat: 0.901 G_ID: 0.103 G_Rec: 0.334 D_GP: 0.050 D_real: 0.768 D_fake: 0.934 +(epoch: 477, iters: 4398, time: 0.064) G_GAN: 0.256 G_GAN_Feat: 0.945 G_ID: 0.136 G_Rec: 0.403 D_GP: 0.035 D_real: 0.959 D_fake: 0.746 +(epoch: 477, iters: 4798, time: 0.064) G_GAN: 0.041 G_GAN_Feat: 0.685 G_ID: 0.098 G_Rec: 0.306 D_GP: 0.038 D_real: 0.868 D_fake: 0.959 +(epoch: 477, iters: 5198, time: 0.064) G_GAN: 0.324 G_GAN_Feat: 1.078 G_ID: 0.120 G_Rec: 0.441 D_GP: 0.041 D_real: 0.604 D_fake: 0.677 +(epoch: 477, iters: 5598, time: 0.064) G_GAN: 0.303 G_GAN_Feat: 0.908 G_ID: 0.092 G_Rec: 0.343 D_GP: 0.074 D_real: 1.086 D_fake: 0.700 +(epoch: 477, iters: 5998, time: 0.064) G_GAN: 1.046 G_GAN_Feat: 1.147 G_ID: 0.108 G_Rec: 0.477 D_GP: 0.071 D_real: 0.990 D_fake: 0.167 +(epoch: 477, iters: 6398, time: 0.064) G_GAN: 0.151 G_GAN_Feat: 0.694 G_ID: 0.091 G_Rec: 0.294 D_GP: 0.032 D_real: 1.116 D_fake: 0.849 +(epoch: 477, iters: 6798, time: 0.064) G_GAN: 0.159 G_GAN_Feat: 0.921 G_ID: 0.131 G_Rec: 0.424 D_GP: 0.042 D_real: 0.641 D_fake: 0.842 +(epoch: 477, iters: 7198, time: 0.064) G_GAN: -0.195 G_GAN_Feat: 0.824 G_ID: 0.105 G_Rec: 0.328 D_GP: 0.088 D_real: 0.569 D_fake: 1.195 +(epoch: 477, iters: 7598, time: 0.064) G_GAN: 0.832 G_GAN_Feat: 0.941 G_ID: 0.127 G_Rec: 0.392 D_GP: 0.036 D_real: 1.493 D_fake: 0.251 +(epoch: 477, iters: 7998, time: 0.063) G_GAN: 0.318 G_GAN_Feat: 0.835 G_ID: 0.100 G_Rec: 0.379 D_GP: 0.053 D_real: 1.157 D_fake: 0.683 +(epoch: 477, iters: 8398, time: 0.064) G_GAN: 0.460 G_GAN_Feat: 1.017 G_ID: 0.110 G_Rec: 0.403 D_GP: 0.099 D_real: 0.593 D_fake: 0.542 +(epoch: 478, iters: 190, time: 0.063) G_GAN: 0.175 G_GAN_Feat: 0.900 G_ID: 0.091 G_Rec: 0.313 D_GP: 0.034 D_real: 0.844 D_fake: 0.825 +(epoch: 478, iters: 590, time: 0.064) G_GAN: 0.417 G_GAN_Feat: 0.862 G_ID: 0.121 G_Rec: 0.451 D_GP: 0.027 D_real: 1.156 D_fake: 0.587 +(epoch: 478, iters: 990, time: 0.063) G_GAN: 0.221 G_GAN_Feat: 0.646 G_ID: 0.077 G_Rec: 0.311 D_GP: 0.023 D_real: 1.148 D_fake: 0.779 +(epoch: 478, iters: 1390, time: 0.064) G_GAN: 0.280 G_GAN_Feat: 0.891 G_ID: 0.110 G_Rec: 0.462 D_GP: 0.035 D_real: 0.836 D_fake: 0.724 +(epoch: 478, iters: 1790, time: 0.063) G_GAN: 0.144 G_GAN_Feat: 0.603 G_ID: 0.105 G_Rec: 0.288 D_GP: 0.029 D_real: 1.013 D_fake: 0.856 +(epoch: 478, iters: 2190, time: 0.064) G_GAN: 0.170 G_GAN_Feat: 0.908 G_ID: 0.132 G_Rec: 0.453 D_GP: 0.047 D_real: 0.744 D_fake: 0.833 +(epoch: 478, iters: 2590, time: 0.063) G_GAN: 0.027 G_GAN_Feat: 0.690 G_ID: 0.104 G_Rec: 0.328 D_GP: 0.043 D_real: 0.839 D_fake: 0.973 +(epoch: 478, iters: 2990, time: 0.063) G_GAN: 0.339 G_GAN_Feat: 0.927 G_ID: 0.115 G_Rec: 0.470 D_GP: 0.039 D_real: 0.924 D_fake: 0.678 +(epoch: 478, iters: 3390, time: 0.064) G_GAN: 0.006 G_GAN_Feat: 0.732 G_ID: 0.107 G_Rec: 0.366 D_GP: 0.042 D_real: 0.788 D_fake: 0.995 +(epoch: 478, iters: 3790, time: 0.064) G_GAN: 0.113 G_GAN_Feat: 0.905 G_ID: 0.134 G_Rec: 0.465 D_GP: 0.034 D_real: 0.654 D_fake: 0.888 +(epoch: 478, iters: 4190, time: 0.064) G_GAN: 0.084 G_GAN_Feat: 0.678 G_ID: 0.095 G_Rec: 0.313 D_GP: 0.043 D_real: 0.887 D_fake: 0.916 +(epoch: 478, iters: 4590, time: 0.064) G_GAN: 0.140 G_GAN_Feat: 0.877 G_ID: 0.112 G_Rec: 0.402 D_GP: 0.040 D_real: 0.863 D_fake: 0.867 +(epoch: 478, iters: 4990, time: 0.063) G_GAN: -0.141 G_GAN_Feat: 0.734 G_ID: 0.097 G_Rec: 0.334 D_GP: 0.058 D_real: 0.583 D_fake: 1.141 +(epoch: 478, iters: 5390, time: 0.064) G_GAN: 0.159 G_GAN_Feat: 0.907 G_ID: 0.118 G_Rec: 0.439 D_GP: 0.065 D_real: 0.777 D_fake: 0.849 +(epoch: 478, iters: 5790, time: 0.064) G_GAN: 0.064 G_GAN_Feat: 0.753 G_ID: 0.093 G_Rec: 0.312 D_GP: 0.085 D_real: 0.747 D_fake: 0.937 +(epoch: 478, iters: 6190, time: 0.064) G_GAN: 0.420 G_GAN_Feat: 0.909 G_ID: 0.104 G_Rec: 0.415 D_GP: 0.048 D_real: 1.077 D_fake: 0.603 +(epoch: 478, iters: 6590, time: 0.064) G_GAN: 0.170 G_GAN_Feat: 0.654 G_ID: 0.095 G_Rec: 0.289 D_GP: 0.033 D_real: 0.999 D_fake: 0.830 +(epoch: 478, iters: 6990, time: 0.064) G_GAN: 0.375 G_GAN_Feat: 0.900 G_ID: 0.105 G_Rec: 0.383 D_GP: 0.039 D_real: 1.194 D_fake: 0.632 +(epoch: 478, iters: 7390, time: 0.064) G_GAN: 0.091 G_GAN_Feat: 0.694 G_ID: 0.120 G_Rec: 0.326 D_GP: 0.030 D_real: 0.955 D_fake: 0.909 +(epoch: 478, iters: 7790, time: 0.063) G_GAN: 0.382 G_GAN_Feat: 0.933 G_ID: 0.116 G_Rec: 0.455 D_GP: 0.043 D_real: 0.927 D_fake: 0.626 +(epoch: 478, iters: 8190, time: 0.064) G_GAN: 0.216 G_GAN_Feat: 0.648 G_ID: 0.104 G_Rec: 0.279 D_GP: 0.037 D_real: 1.049 D_fake: 0.784 +(epoch: 478, iters: 8590, time: 0.064) G_GAN: 0.072 G_GAN_Feat: 0.884 G_ID: 0.123 G_Rec: 0.402 D_GP: 0.049 D_real: 0.704 D_fake: 0.928 +(epoch: 479, iters: 382, time: 0.064) G_GAN: 0.144 G_GAN_Feat: 0.725 G_ID: 0.093 G_Rec: 0.307 D_GP: 0.044 D_real: 1.020 D_fake: 0.857 +(epoch: 479, iters: 782, time: 0.064) G_GAN: 0.636 G_GAN_Feat: 0.957 G_ID: 0.108 G_Rec: 0.434 D_GP: 0.043 D_real: 1.126 D_fake: 0.402 +(epoch: 479, iters: 1182, time: 0.064) G_GAN: 0.111 G_GAN_Feat: 0.762 G_ID: 0.094 G_Rec: 0.321 D_GP: 0.053 D_real: 0.803 D_fake: 0.889 +(epoch: 479, iters: 1582, time: 0.064) G_GAN: 0.304 G_GAN_Feat: 1.079 G_ID: 0.119 G_Rec: 0.484 D_GP: 0.640 D_real: 0.501 D_fake: 0.702 +(epoch: 479, iters: 1982, time: 0.064) G_GAN: 0.206 G_GAN_Feat: 0.743 G_ID: 0.105 G_Rec: 0.294 D_GP: 0.035 D_real: 1.122 D_fake: 0.794 +(epoch: 479, iters: 2382, time: 0.064) G_GAN: 0.274 G_GAN_Feat: 0.822 G_ID: 0.126 G_Rec: 0.373 D_GP: 0.032 D_real: 1.088 D_fake: 0.728 +(epoch: 479, iters: 2782, time: 0.064) G_GAN: 0.212 G_GAN_Feat: 0.740 G_ID: 0.089 G_Rec: 0.324 D_GP: 0.036 D_real: 1.008 D_fake: 0.788 +(epoch: 479, iters: 3182, time: 0.064) G_GAN: 0.834 G_GAN_Feat: 1.053 G_ID: 0.135 G_Rec: 0.481 D_GP: 0.050 D_real: 1.405 D_fake: 0.280 +(epoch: 479, iters: 3582, time: 0.063) G_GAN: 0.076 G_GAN_Feat: 0.869 G_ID: 0.091 G_Rec: 0.323 D_GP: 0.128 D_real: 0.548 D_fake: 0.924 +(epoch: 479, iters: 3982, time: 0.064) G_GAN: 0.443 G_GAN_Feat: 1.002 G_ID: 0.115 G_Rec: 0.486 D_GP: 0.064 D_real: 0.902 D_fake: 0.562 +(epoch: 479, iters: 4382, time: 0.064) G_GAN: -0.057 G_GAN_Feat: 0.723 G_ID: 0.109 G_Rec: 0.309 D_GP: 0.034 D_real: 0.904 D_fake: 1.057 +(epoch: 479, iters: 4782, time: 0.064) G_GAN: 0.593 G_GAN_Feat: 0.903 G_ID: 0.122 G_Rec: 0.391 D_GP: 0.037 D_real: 1.132 D_fake: 0.425 +(epoch: 479, iters: 5182, time: 0.064) G_GAN: 0.186 G_GAN_Feat: 0.709 G_ID: 0.101 G_Rec: 0.288 D_GP: 0.030 D_real: 1.079 D_fake: 0.814 +(epoch: 479, iters: 5582, time: 0.064) G_GAN: 0.633 G_GAN_Feat: 1.028 G_ID: 0.099 G_Rec: 0.447 D_GP: 0.088 D_real: 0.876 D_fake: 0.389 +(epoch: 479, iters: 5982, time: 0.064) G_GAN: 0.117 G_GAN_Feat: 0.713 G_ID: 0.102 G_Rec: 0.296 D_GP: 0.032 D_real: 0.973 D_fake: 0.883 +(epoch: 479, iters: 6382, time: 0.064) G_GAN: 0.567 G_GAN_Feat: 0.957 G_ID: 0.128 G_Rec: 0.442 D_GP: 0.035 D_real: 1.079 D_fake: 0.458 +(epoch: 479, iters: 6782, time: 0.064) G_GAN: 0.314 G_GAN_Feat: 0.727 G_ID: 0.104 G_Rec: 0.301 D_GP: 0.034 D_real: 1.237 D_fake: 0.687 +(epoch: 479, iters: 7182, time: 0.064) G_GAN: 0.492 G_GAN_Feat: 0.980 G_ID: 0.131 G_Rec: 0.400 D_GP: 0.036 D_real: 1.112 D_fake: 0.530 +(epoch: 479, iters: 7582, time: 0.064) G_GAN: 0.286 G_GAN_Feat: 0.734 G_ID: 0.089 G_Rec: 0.288 D_GP: 0.034 D_real: 1.112 D_fake: 0.714 +(epoch: 479, iters: 7982, time: 0.064) G_GAN: 0.978 G_GAN_Feat: 1.002 G_ID: 0.114 G_Rec: 0.432 D_GP: 0.040 D_real: 1.573 D_fake: 0.190 +(epoch: 479, iters: 8382, time: 0.064) G_GAN: 0.275 G_GAN_Feat: 0.698 G_ID: 0.092 G_Rec: 0.277 D_GP: 0.032 D_real: 1.152 D_fake: 0.725 +(epoch: 480, iters: 174, time: 0.064) G_GAN: 1.056 G_GAN_Feat: 1.016 G_ID: 0.125 G_Rec: 0.426 D_GP: 0.062 D_real: 1.402 D_fake: 0.373 +(epoch: 480, iters: 574, time: 0.063) G_GAN: 0.076 G_GAN_Feat: 0.655 G_ID: 0.100 G_Rec: 0.270 D_GP: 0.032 D_real: 0.996 D_fake: 0.925 +(epoch: 480, iters: 974, time: 0.064) G_GAN: 0.629 G_GAN_Feat: 1.076 G_ID: 0.108 G_Rec: 0.462 D_GP: 0.094 D_real: 0.732 D_fake: 0.416 +(epoch: 480, iters: 1374, time: 0.064) G_GAN: 0.063 G_GAN_Feat: 0.693 G_ID: 0.114 G_Rec: 0.264 D_GP: 0.033 D_real: 0.940 D_fake: 0.937 +(epoch: 480, iters: 1774, time: 0.064) G_GAN: -0.048 G_GAN_Feat: 1.182 G_ID: 0.140 G_Rec: 0.483 D_GP: 0.070 D_real: 0.102 D_fake: 1.048 +(epoch: 480, iters: 2174, time: 0.064) G_GAN: 0.012 G_GAN_Feat: 0.779 G_ID: 0.093 G_Rec: 0.314 D_GP: 0.041 D_real: 0.700 D_fake: 0.988 +(epoch: 480, iters: 2574, time: 0.064) G_GAN: 0.289 G_GAN_Feat: 0.966 G_ID: 0.106 G_Rec: 0.432 D_GP: 0.032 D_real: 0.893 D_fake: 0.712 +(epoch: 480, iters: 2974, time: 0.064) G_GAN: 0.270 G_GAN_Feat: 0.799 G_ID: 0.102 G_Rec: 0.320 D_GP: 0.044 D_real: 1.123 D_fake: 0.734 +(epoch: 480, iters: 3374, time: 0.063) G_GAN: 0.727 G_GAN_Feat: 1.080 G_ID: 0.113 G_Rec: 0.434 D_GP: 0.041 D_real: 0.927 D_fake: 0.374 +(epoch: 480, iters: 3774, time: 0.064) G_GAN: 0.291 G_GAN_Feat: 1.055 G_ID: 0.086 G_Rec: 0.333 D_GP: 0.116 D_real: 0.496 D_fake: 0.711 +(epoch: 480, iters: 4174, time: 0.064) G_GAN: 0.640 G_GAN_Feat: 1.003 G_ID: 0.118 G_Rec: 0.459 D_GP: 0.036 D_real: 1.256 D_fake: 0.440 +(epoch: 480, iters: 4574, time: 0.064) G_GAN: 0.094 G_GAN_Feat: 0.647 G_ID: 0.104 G_Rec: 0.285 D_GP: 0.028 D_real: 1.025 D_fake: 0.906 +(epoch: 480, iters: 4974, time: 0.064) G_GAN: 0.439 G_GAN_Feat: 0.891 G_ID: 0.097 G_Rec: 0.430 D_GP: 0.032 D_real: 1.090 D_fake: 0.573 +(epoch: 480, iters: 5374, time: 0.064) G_GAN: 0.013 G_GAN_Feat: 0.696 G_ID: 0.098 G_Rec: 0.293 D_GP: 0.034 D_real: 0.864 D_fake: 0.987 +(epoch: 480, iters: 5774, time: 0.063) G_GAN: 0.187 G_GAN_Feat: 0.865 G_ID: 0.110 G_Rec: 0.406 D_GP: 0.047 D_real: 0.789 D_fake: 0.819 +(epoch: 480, iters: 6174, time: 0.063) G_GAN: 0.116 G_GAN_Feat: 0.700 G_ID: 0.094 G_Rec: 0.305 D_GP: 0.051 D_real: 0.942 D_fake: 0.887 +(epoch: 480, iters: 6574, time: 0.064) G_GAN: 0.075 G_GAN_Feat: 0.969 G_ID: 0.112 G_Rec: 0.444 D_GP: 0.048 D_real: 0.639 D_fake: 0.925 +(epoch: 480, iters: 6974, time: 0.064) G_GAN: -0.141 G_GAN_Feat: 0.622 G_ID: 0.075 G_Rec: 0.249 D_GP: 0.032 D_real: 0.868 D_fake: 1.141 +(epoch: 480, iters: 7374, time: 0.064) G_GAN: 0.174 G_GAN_Feat: 0.864 G_ID: 0.132 G_Rec: 0.373 D_GP: 0.039 D_real: 0.795 D_fake: 0.826 +(epoch: 480, iters: 7774, time: 0.063) G_GAN: -0.057 G_GAN_Feat: 0.659 G_ID: 0.085 G_Rec: 0.265 D_GP: 0.044 D_real: 0.933 D_fake: 1.057 +(epoch: 480, iters: 8174, time: 0.064) G_GAN: 0.188 G_GAN_Feat: 0.956 G_ID: 0.123 G_Rec: 0.415 D_GP: 0.038 D_real: 0.850 D_fake: 0.812 +(epoch: 480, iters: 8574, time: 0.064) G_GAN: 0.155 G_GAN_Feat: 0.717 G_ID: 0.099 G_Rec: 0.315 D_GP: 0.036 D_real: 0.968 D_fake: 0.845 +(epoch: 481, iters: 366, time: 0.063) G_GAN: 0.430 G_GAN_Feat: 1.047 G_ID: 0.122 G_Rec: 0.476 D_GP: 0.077 D_real: 0.760 D_fake: 0.580 +(epoch: 481, iters: 766, time: 0.063) G_GAN: -0.508 G_GAN_Feat: 1.083 G_ID: 0.102 G_Rec: 0.384 D_GP: 0.050 D_real: 1.081 D_fake: 1.508 +(epoch: 481, iters: 1166, time: 0.063) G_GAN: 0.226 G_GAN_Feat: 0.832 G_ID: 0.116 G_Rec: 0.417 D_GP: 0.023 D_real: 1.010 D_fake: 0.787 +(epoch: 481, iters: 1566, time: 0.064) G_GAN: -0.196 G_GAN_Feat: 0.636 G_ID: 0.099 G_Rec: 0.285 D_GP: 0.023 D_real: 0.769 D_fake: 1.200 +(epoch: 481, iters: 1966, time: 0.064) G_GAN: 0.187 G_GAN_Feat: 0.886 G_ID: 0.119 G_Rec: 0.438 D_GP: 0.029 D_real: 0.850 D_fake: 0.822 +(epoch: 481, iters: 2366, time: 0.064) G_GAN: -0.035 G_GAN_Feat: 0.664 G_ID: 0.087 G_Rec: 0.295 D_GP: 0.031 D_real: 0.847 D_fake: 1.036 +(epoch: 481, iters: 2766, time: 0.064) G_GAN: -0.042 G_GAN_Feat: 0.894 G_ID: 0.129 G_Rec: 0.436 D_GP: 0.055 D_real: 0.587 D_fake: 1.043 +(epoch: 481, iters: 3166, time: 0.064) G_GAN: 0.171 G_GAN_Feat: 0.697 G_ID: 0.113 G_Rec: 0.301 D_GP: 0.036 D_real: 0.995 D_fake: 0.831 +(epoch: 481, iters: 3566, time: 0.063) G_GAN: 0.561 G_GAN_Feat: 0.896 G_ID: 0.099 G_Rec: 0.396 D_GP: 0.035 D_real: 1.235 D_fake: 0.476 +(epoch: 481, iters: 3966, time: 0.063) G_GAN: 0.014 G_GAN_Feat: 0.802 G_ID: 0.094 G_Rec: 0.315 D_GP: 0.099 D_real: 0.626 D_fake: 0.986 +(epoch: 481, iters: 4366, time: 0.064) G_GAN: 0.290 G_GAN_Feat: 0.961 G_ID: 0.118 G_Rec: 0.443 D_GP: 0.036 D_real: 0.932 D_fake: 0.712 +(epoch: 481, iters: 4766, time: 0.064) G_GAN: 0.216 G_GAN_Feat: 0.858 G_ID: 0.107 G_Rec: 0.327 D_GP: 0.039 D_real: 0.758 D_fake: 0.784 +(epoch: 481, iters: 5166, time: 0.063) G_GAN: 0.370 G_GAN_Feat: 0.982 G_ID: 0.103 G_Rec: 0.424 D_GP: 0.042 D_real: 0.808 D_fake: 0.631 +(epoch: 481, iters: 5566, time: 0.064) G_GAN: 0.192 G_GAN_Feat: 0.813 G_ID: 0.099 G_Rec: 0.312 D_GP: 0.045 D_real: 0.877 D_fake: 0.809 +(epoch: 481, iters: 5966, time: 0.063) G_GAN: 0.324 G_GAN_Feat: 0.924 G_ID: 0.109 G_Rec: 0.392 D_GP: 0.030 D_real: 1.027 D_fake: 0.683 +(epoch: 481, iters: 6366, time: 0.064) G_GAN: -0.083 G_GAN_Feat: 0.799 G_ID: 0.109 G_Rec: 0.327 D_GP: 0.040 D_real: 0.577 D_fake: 1.083 +(epoch: 481, iters: 6766, time: 0.064) G_GAN: 0.632 G_GAN_Feat: 0.999 G_ID: 0.106 G_Rec: 0.458 D_GP: 0.037 D_real: 1.214 D_fake: 0.378 +(epoch: 481, iters: 7166, time: 0.064) G_GAN: 0.443 G_GAN_Feat: 0.801 G_ID: 0.091 G_Rec: 0.340 D_GP: 0.033 D_real: 1.134 D_fake: 0.561 +(epoch: 481, iters: 7566, time: 0.063) G_GAN: 0.720 G_GAN_Feat: 0.985 G_ID: 0.101 G_Rec: 0.431 D_GP: 0.077 D_real: 1.300 D_fake: 0.341 +(epoch: 481, iters: 7966, time: 0.064) G_GAN: 0.174 G_GAN_Feat: 0.844 G_ID: 0.087 G_Rec: 0.350 D_GP: 0.046 D_real: 0.826 D_fake: 0.827 +(epoch: 481, iters: 8366, time: 0.064) G_GAN: 0.542 G_GAN_Feat: 0.983 G_ID: 0.106 G_Rec: 0.407 D_GP: 0.056 D_real: 1.018 D_fake: 0.490 +(epoch: 482, iters: 158, time: 0.063) G_GAN: 0.245 G_GAN_Feat: 0.928 G_ID: 0.099 G_Rec: 0.333 D_GP: 0.040 D_real: 0.558 D_fake: 0.755 +(epoch: 482, iters: 558, time: 0.064) G_GAN: 0.159 G_GAN_Feat: 0.929 G_ID: 0.131 G_Rec: 0.444 D_GP: 0.037 D_real: 0.777 D_fake: 0.841 +(epoch: 482, iters: 958, time: 0.064) G_GAN: 0.277 G_GAN_Feat: 0.917 G_ID: 0.081 G_Rec: 0.324 D_GP: 0.064 D_real: 0.845 D_fake: 0.723 +(epoch: 482, iters: 1358, time: 0.064) G_GAN: 0.713 G_GAN_Feat: 0.996 G_ID: 0.108 G_Rec: 0.377 D_GP: 0.030 D_real: 1.085 D_fake: 0.403 +(epoch: 482, iters: 1758, time: 0.063) G_GAN: -0.066 G_GAN_Feat: 0.778 G_ID: 0.091 G_Rec: 0.322 D_GP: 0.029 D_real: 0.826 D_fake: 1.067 +(epoch: 482, iters: 2158, time: 0.063) G_GAN: 0.162 G_GAN_Feat: 0.896 G_ID: 0.135 G_Rec: 0.408 D_GP: 0.028 D_real: 0.741 D_fake: 0.843 +(epoch: 482, iters: 2558, time: 0.064) G_GAN: 0.136 G_GAN_Feat: 0.784 G_ID: 0.110 G_Rec: 0.327 D_GP: 0.075 D_real: 0.768 D_fake: 0.865 +(epoch: 482, iters: 2958, time: 0.063) G_GAN: 0.565 G_GAN_Feat: 1.084 G_ID: 0.117 G_Rec: 0.426 D_GP: 0.065 D_real: 0.499 D_fake: 0.446 +(epoch: 482, iters: 3358, time: 0.063) G_GAN: 0.127 G_GAN_Feat: 0.733 G_ID: 0.099 G_Rec: 0.334 D_GP: 0.039 D_real: 0.983 D_fake: 0.874 +(epoch: 482, iters: 3758, time: 0.064) G_GAN: 0.574 G_GAN_Feat: 0.939 G_ID: 0.112 G_Rec: 0.443 D_GP: 0.033 D_real: 1.190 D_fake: 0.435 +(epoch: 482, iters: 4158, time: 0.064) G_GAN: 0.350 G_GAN_Feat: 0.982 G_ID: 0.090 G_Rec: 0.349 D_GP: 0.574 D_real: 0.486 D_fake: 0.658 +(epoch: 482, iters: 4558, time: 0.064) G_GAN: 0.575 G_GAN_Feat: 1.140 G_ID: 0.120 G_Rec: 0.436 D_GP: 0.055 D_real: 0.383 D_fake: 0.443 +(epoch: 482, iters: 4958, time: 0.064) G_GAN: 0.502 G_GAN_Feat: 0.981 G_ID: 0.086 G_Rec: 0.338 D_GP: 0.100 D_real: 0.524 D_fake: 0.517 +(epoch: 482, iters: 5358, time: 0.064) G_GAN: 0.388 G_GAN_Feat: 0.965 G_ID: 0.109 G_Rec: 0.436 D_GP: 0.031 D_real: 1.010 D_fake: 0.647 +(epoch: 482, iters: 5758, time: 0.064) G_GAN: 0.295 G_GAN_Feat: 0.871 G_ID: 0.097 G_Rec: 0.313 D_GP: 0.072 D_real: 0.597 D_fake: 0.710 +(epoch: 482, iters: 6158, time: 0.063) G_GAN: 0.556 G_GAN_Feat: 1.191 G_ID: 0.094 G_Rec: 0.452 D_GP: 0.321 D_real: 0.371 D_fake: 0.565 +(epoch: 482, iters: 6558, time: 0.063) G_GAN: -0.132 G_GAN_Feat: 0.774 G_ID: 0.083 G_Rec: 0.326 D_GP: 0.030 D_real: 0.790 D_fake: 1.132 +(epoch: 482, iters: 6958, time: 0.063) G_GAN: 0.481 G_GAN_Feat: 1.156 G_ID: 0.129 G_Rec: 0.440 D_GP: 0.098 D_real: 0.508 D_fake: 0.530 +(epoch: 482, iters: 7358, time: 0.064) G_GAN: 0.286 G_GAN_Feat: 0.898 G_ID: 0.104 G_Rec: 0.321 D_GP: 0.063 D_real: 0.309 D_fake: 0.717 +(epoch: 482, iters: 7758, time: 0.064) G_GAN: 0.579 G_GAN_Feat: 0.982 G_ID: 0.129 G_Rec: 0.449 D_GP: 0.034 D_real: 1.322 D_fake: 0.458 +(epoch: 482, iters: 8158, time: 0.063) G_GAN: 0.054 G_GAN_Feat: 0.773 G_ID: 0.092 G_Rec: 0.301 D_GP: 0.058 D_real: 0.795 D_fake: 0.947 +(epoch: 482, iters: 8558, time: 0.063) G_GAN: 0.516 G_GAN_Feat: 1.008 G_ID: 0.107 G_Rec: 0.475 D_GP: 0.038 D_real: 0.851 D_fake: 0.494 +(epoch: 483, iters: 350, time: 0.064) G_GAN: 0.518 G_GAN_Feat: 0.911 G_ID: 0.099 G_Rec: 0.342 D_GP: 0.336 D_real: 0.545 D_fake: 0.610 +(epoch: 483, iters: 750, time: 0.063) G_GAN: 0.633 G_GAN_Feat: 1.062 G_ID: 0.107 G_Rec: 0.447 D_GP: 0.068 D_real: 0.763 D_fake: 0.378 +(epoch: 483, iters: 1150, time: 0.063) G_GAN: 0.256 G_GAN_Feat: 0.809 G_ID: 0.099 G_Rec: 0.318 D_GP: 0.033 D_real: 0.848 D_fake: 0.755 +(epoch: 483, iters: 1550, time: 0.063) G_GAN: 0.841 G_GAN_Feat: 1.119 G_ID: 0.126 G_Rec: 0.473 D_GP: 0.085 D_real: 1.125 D_fake: 0.287 +(epoch: 483, iters: 1950, time: 0.064) G_GAN: 0.313 G_GAN_Feat: 0.758 G_ID: 0.098 G_Rec: 0.334 D_GP: 0.037 D_real: 1.112 D_fake: 0.690 +(epoch: 483, iters: 2350, time: 0.063) G_GAN: 0.268 G_GAN_Feat: 0.991 G_ID: 0.115 G_Rec: 0.441 D_GP: 0.033 D_real: 0.780 D_fake: 0.732 +(epoch: 483, iters: 2750, time: 0.063) G_GAN: 0.279 G_GAN_Feat: 0.995 G_ID: 0.099 G_Rec: 0.339 D_GP: 0.167 D_real: 0.318 D_fake: 0.733 +(epoch: 483, iters: 3150, time: 0.063) G_GAN: 0.422 G_GAN_Feat: 0.890 G_ID: 0.112 G_Rec: 0.435 D_GP: 0.027 D_real: 1.118 D_fake: 0.585 +(epoch: 483, iters: 3550, time: 0.064) G_GAN: -0.070 G_GAN_Feat: 0.683 G_ID: 0.123 G_Rec: 0.294 D_GP: 0.033 D_real: 0.814 D_fake: 1.070 +(epoch: 483, iters: 3950, time: 0.063) G_GAN: 0.283 G_GAN_Feat: 0.896 G_ID: 0.111 G_Rec: 0.387 D_GP: 0.034 D_real: 0.960 D_fake: 0.718 +(epoch: 483, iters: 4350, time: 0.063) G_GAN: 0.296 G_GAN_Feat: 0.810 G_ID: 0.123 G_Rec: 0.343 D_GP: 0.036 D_real: 1.063 D_fake: 0.712 +(epoch: 483, iters: 4750, time: 0.063) G_GAN: 0.685 G_GAN_Feat: 1.125 G_ID: 0.121 G_Rec: 0.509 D_GP: 0.039 D_real: 0.996 D_fake: 0.385 +(epoch: 483, iters: 5150, time: 0.064) G_GAN: 0.216 G_GAN_Feat: 0.796 G_ID: 0.094 G_Rec: 0.293 D_GP: 0.034 D_real: 0.968 D_fake: 0.785 +(epoch: 483, iters: 5550, time: 0.063) G_GAN: 0.567 G_GAN_Feat: 1.157 G_ID: 0.126 G_Rec: 0.483 D_GP: 0.047 D_real: 1.049 D_fake: 0.456 +(epoch: 483, iters: 5950, time: 0.063) G_GAN: 0.286 G_GAN_Feat: 0.835 G_ID: 0.102 G_Rec: 0.312 D_GP: 0.033 D_real: 0.790 D_fake: 0.715 +(epoch: 483, iters: 6350, time: 0.063) G_GAN: 0.584 G_GAN_Feat: 1.101 G_ID: 0.114 G_Rec: 0.442 D_GP: 0.965 D_real: 0.698 D_fake: 0.461 +(epoch: 483, iters: 6750, time: 0.064) G_GAN: 0.024 G_GAN_Feat: 0.772 G_ID: 0.114 G_Rec: 0.313 D_GP: 0.044 D_real: 0.789 D_fake: 0.976 +(epoch: 483, iters: 7150, time: 0.063) G_GAN: 0.390 G_GAN_Feat: 1.128 G_ID: 0.109 G_Rec: 0.486 D_GP: 0.034 D_real: 0.680 D_fake: 0.611 +(epoch: 483, iters: 7550, time: 0.063) G_GAN: 0.277 G_GAN_Feat: 0.770 G_ID: 0.115 G_Rec: 0.299 D_GP: 0.031 D_real: 1.072 D_fake: 0.724 +(epoch: 483, iters: 7950, time: 0.063) G_GAN: 0.304 G_GAN_Feat: 1.008 G_ID: 0.116 G_Rec: 0.433 D_GP: 0.032 D_real: 0.924 D_fake: 0.697 +(epoch: 483, iters: 8350, time: 0.064) G_GAN: 0.370 G_GAN_Feat: 0.742 G_ID: 0.094 G_Rec: 0.276 D_GP: 0.031 D_real: 1.173 D_fake: 0.632 +(epoch: 484, iters: 142, time: 0.063) G_GAN: 0.475 G_GAN_Feat: 1.036 G_ID: 0.119 G_Rec: 0.441 D_GP: 0.036 D_real: 1.073 D_fake: 0.530 +(epoch: 484, iters: 542, time: 0.063) G_GAN: 0.262 G_GAN_Feat: 0.734 G_ID: 0.093 G_Rec: 0.327 D_GP: 0.029 D_real: 1.174 D_fake: 0.738 +(epoch: 484, iters: 942, time: 0.063) G_GAN: 0.164 G_GAN_Feat: 0.857 G_ID: 0.136 G_Rec: 0.380 D_GP: 0.033 D_real: 0.829 D_fake: 0.837 +(epoch: 484, iters: 1342, time: 0.064) G_GAN: 0.062 G_GAN_Feat: 0.804 G_ID: 0.101 G_Rec: 0.351 D_GP: 0.037 D_real: 0.829 D_fake: 0.938 +(epoch: 484, iters: 1742, time: 0.063) G_GAN: 0.489 G_GAN_Feat: 0.941 G_ID: 0.105 G_Rec: 0.541 D_GP: 0.030 D_real: 1.072 D_fake: 0.531 +(epoch: 484, iters: 2142, time: 0.063) G_GAN: -0.288 G_GAN_Feat: 0.771 G_ID: 0.109 G_Rec: 0.336 D_GP: 0.046 D_real: 0.487 D_fake: 1.288 +(epoch: 484, iters: 2542, time: 0.063) G_GAN: 0.380 G_GAN_Feat: 1.020 G_ID: 0.105 G_Rec: 0.454 D_GP: 0.048 D_real: 0.879 D_fake: 0.624 +(epoch: 484, iters: 2942, time: 0.064) G_GAN: 0.298 G_GAN_Feat: 0.732 G_ID: 0.097 G_Rec: 0.291 D_GP: 0.039 D_real: 1.136 D_fake: 0.704 +(epoch: 484, iters: 3342, time: 0.063) G_GAN: 0.833 G_GAN_Feat: 0.974 G_ID: 0.105 G_Rec: 0.387 D_GP: 0.039 D_real: 1.487 D_fake: 0.212 +(epoch: 484, iters: 3742, time: 0.063) G_GAN: 0.204 G_GAN_Feat: 0.832 G_ID: 0.105 G_Rec: 0.337 D_GP: 0.149 D_real: 0.566 D_fake: 0.801 +(epoch: 484, iters: 4142, time: 0.063) G_GAN: 0.526 G_GAN_Feat: 1.038 G_ID: 0.123 G_Rec: 0.467 D_GP: 0.044 D_real: 0.914 D_fake: 0.477 +(epoch: 484, iters: 4542, time: 0.064) G_GAN: 0.438 G_GAN_Feat: 0.851 G_ID: 0.099 G_Rec: 0.318 D_GP: 0.042 D_real: 1.235 D_fake: 0.564 +(epoch: 484, iters: 4942, time: 0.063) G_GAN: 0.526 G_GAN_Feat: 1.047 G_ID: 0.106 G_Rec: 0.476 D_GP: 0.037 D_real: 1.114 D_fake: 0.493 +(epoch: 484, iters: 5342, time: 0.063) G_GAN: 0.021 G_GAN_Feat: 0.700 G_ID: 0.097 G_Rec: 0.285 D_GP: 0.034 D_real: 0.937 D_fake: 0.979 +(epoch: 484, iters: 5742, time: 0.063) G_GAN: 0.415 G_GAN_Feat: 1.058 G_ID: 0.126 G_Rec: 0.443 D_GP: 0.044 D_real: 0.701 D_fake: 0.591 +(epoch: 484, iters: 6142, time: 0.064) G_GAN: 0.254 G_GAN_Feat: 0.965 G_ID: 0.095 G_Rec: 0.375 D_GP: 0.113 D_real: 0.468 D_fake: 0.746 +(epoch: 484, iters: 6542, time: 0.063) G_GAN: 0.287 G_GAN_Feat: 1.123 G_ID: 0.114 G_Rec: 0.436 D_GP: 0.038 D_real: 0.670 D_fake: 0.714 +(epoch: 484, iters: 6942, time: 0.063) G_GAN: 0.349 G_GAN_Feat: 0.910 G_ID: 0.096 G_Rec: 0.336 D_GP: 0.058 D_real: 1.329 D_fake: 0.655 +(epoch: 484, iters: 7342, time: 0.063) G_GAN: 0.200 G_GAN_Feat: 1.109 G_ID: 0.126 G_Rec: 0.436 D_GP: 0.033 D_real: 0.498 D_fake: 0.801 +(epoch: 484, iters: 7742, time: 0.064) G_GAN: 0.241 G_GAN_Feat: 0.711 G_ID: 0.086 G_Rec: 0.276 D_GP: 0.033 D_real: 1.178 D_fake: 0.759 +(epoch: 484, iters: 8142, time: 0.063) G_GAN: 0.655 G_GAN_Feat: 1.234 G_ID: 0.114 G_Rec: 0.494 D_GP: 0.249 D_real: 0.552 D_fake: 0.373 +(epoch: 484, iters: 8542, time: 0.063) G_GAN: 0.357 G_GAN_Feat: 1.039 G_ID: 0.112 G_Rec: 0.355 D_GP: 0.123 D_real: 0.236 D_fake: 0.687 +(epoch: 485, iters: 334, time: 0.063) G_GAN: 0.423 G_GAN_Feat: 0.947 G_ID: 0.115 G_Rec: 0.434 D_GP: 0.029 D_real: 1.091 D_fake: 0.583 +(epoch: 485, iters: 734, time: 0.064) G_GAN: 0.184 G_GAN_Feat: 0.643 G_ID: 0.096 G_Rec: 0.307 D_GP: 0.027 D_real: 1.116 D_fake: 0.817 +(epoch: 485, iters: 1134, time: 0.063) G_GAN: -0.054 G_GAN_Feat: 0.883 G_ID: 0.146 G_Rec: 0.441 D_GP: 0.033 D_real: 0.768 D_fake: 1.054 +(epoch: 485, iters: 1534, time: 0.063) G_GAN: 0.278 G_GAN_Feat: 0.699 G_ID: 0.092 G_Rec: 0.302 D_GP: 0.034 D_real: 1.211 D_fake: 0.724 +(epoch: 485, iters: 1934, time: 0.063) G_GAN: 0.292 G_GAN_Feat: 0.978 G_ID: 0.122 G_Rec: 0.443 D_GP: 0.049 D_real: 0.854 D_fake: 0.713 +(epoch: 485, iters: 2334, time: 0.063) G_GAN: 0.252 G_GAN_Feat: 0.627 G_ID: 0.093 G_Rec: 0.275 D_GP: 0.027 D_real: 1.178 D_fake: 0.749 +(epoch: 485, iters: 2734, time: 0.063) G_GAN: 0.301 G_GAN_Feat: 0.907 G_ID: 0.117 G_Rec: 0.432 D_GP: 0.033 D_real: 0.895 D_fake: 0.707 +(epoch: 485, iters: 3134, time: 0.063) G_GAN: 0.103 G_GAN_Feat: 0.724 G_ID: 0.104 G_Rec: 0.313 D_GP: 0.043 D_real: 0.925 D_fake: 0.897 +(epoch: 485, iters: 3534, time: 0.064) G_GAN: 0.324 G_GAN_Feat: 0.956 G_ID: 0.131 G_Rec: 0.457 D_GP: 0.042 D_real: 0.861 D_fake: 0.688 +(epoch: 485, iters: 3934, time: 0.063) G_GAN: 0.065 G_GAN_Feat: 0.697 G_ID: 0.101 G_Rec: 0.305 D_GP: 0.029 D_real: 1.052 D_fake: 0.935 +(epoch: 485, iters: 4334, time: 0.063) G_GAN: 0.389 G_GAN_Feat: 1.010 G_ID: 0.105 G_Rec: 0.460 D_GP: 0.051 D_real: 0.819 D_fake: 0.616 +(epoch: 485, iters: 4734, time: 0.063) G_GAN: 0.133 G_GAN_Feat: 0.737 G_ID: 0.092 G_Rec: 0.298 D_GP: 0.032 D_real: 1.025 D_fake: 0.867 +(epoch: 485, iters: 5134, time: 0.064) G_GAN: 0.378 G_GAN_Feat: 0.991 G_ID: 0.131 G_Rec: 0.423 D_GP: 0.065 D_real: 0.784 D_fake: 0.631 +(epoch: 485, iters: 5534, time: 0.063) G_GAN: 0.113 G_GAN_Feat: 0.711 G_ID: 0.102 G_Rec: 0.295 D_GP: 0.029 D_real: 1.008 D_fake: 0.887 +(epoch: 485, iters: 5934, time: 0.063) G_GAN: 0.548 G_GAN_Feat: 1.038 G_ID: 0.106 G_Rec: 0.463 D_GP: 0.049 D_real: 0.918 D_fake: 0.464 +(epoch: 485, iters: 6334, time: 0.063) G_GAN: 0.138 G_GAN_Feat: 0.741 G_ID: 0.108 G_Rec: 0.312 D_GP: 0.033 D_real: 0.887 D_fake: 0.862 +(epoch: 485, iters: 6734, time: 0.064) G_GAN: 0.403 G_GAN_Feat: 0.839 G_ID: 0.110 G_Rec: 0.382 D_GP: 0.031 D_real: 1.213 D_fake: 0.604 +(epoch: 485, iters: 7134, time: 0.063) G_GAN: 0.323 G_GAN_Feat: 0.660 G_ID: 0.088 G_Rec: 0.301 D_GP: 0.034 D_real: 1.221 D_fake: 0.678 +(epoch: 485, iters: 7534, time: 0.063) G_GAN: 0.427 G_GAN_Feat: 0.996 G_ID: 0.115 G_Rec: 0.434 D_GP: 0.085 D_real: 0.778 D_fake: 0.575 +(epoch: 485, iters: 7934, time: 0.063) G_GAN: 0.180 G_GAN_Feat: 0.841 G_ID: 0.116 G_Rec: 0.413 D_GP: 0.056 D_real: 0.939 D_fake: 0.822 +(epoch: 485, iters: 8334, time: 0.064) G_GAN: 0.450 G_GAN_Feat: 0.975 G_ID: 0.136 G_Rec: 0.427 D_GP: 0.037 D_real: 1.199 D_fake: 0.555 +(epoch: 486, iters: 126, time: 0.063) G_GAN: 0.425 G_GAN_Feat: 0.649 G_ID: 0.087 G_Rec: 0.323 D_GP: 0.029 D_real: 1.351 D_fake: 0.575 +(epoch: 486, iters: 526, time: 0.063) G_GAN: 0.652 G_GAN_Feat: 0.893 G_ID: 0.104 G_Rec: 0.437 D_GP: 0.033 D_real: 1.373 D_fake: 0.393 +(epoch: 486, iters: 926, time: 0.063) G_GAN: 0.224 G_GAN_Feat: 0.724 G_ID: 0.105 G_Rec: 0.323 D_GP: 0.036 D_real: 0.998 D_fake: 0.777 +(epoch: 486, iters: 1326, time: 0.064) G_GAN: 0.312 G_GAN_Feat: 0.897 G_ID: 0.121 G_Rec: 0.389 D_GP: 0.043 D_real: 0.941 D_fake: 0.689 +(epoch: 486, iters: 1726, time: 0.063) G_GAN: -0.064 G_GAN_Feat: 0.784 G_ID: 0.108 G_Rec: 0.300 D_GP: 0.070 D_real: 0.560 D_fake: 1.064 +(epoch: 486, iters: 2126, time: 0.063) G_GAN: 0.496 G_GAN_Feat: 1.028 G_ID: 0.107 G_Rec: 0.415 D_GP: 0.040 D_real: 0.883 D_fake: 0.507 +(epoch: 486, iters: 2526, time: 0.063) G_GAN: 0.200 G_GAN_Feat: 0.855 G_ID: 0.117 G_Rec: 0.343 D_GP: 0.057 D_real: 1.034 D_fake: 0.811 +(epoch: 486, iters: 2926, time: 0.064) G_GAN: 0.697 G_GAN_Feat: 0.937 G_ID: 0.100 G_Rec: 0.383 D_GP: 0.034 D_real: 1.359 D_fake: 0.329 +(epoch: 486, iters: 3326, time: 0.063) G_GAN: 0.181 G_GAN_Feat: 0.942 G_ID: 0.099 G_Rec: 0.338 D_GP: 0.226 D_real: 0.142 D_fake: 1.005 +(epoch: 486, iters: 3726, time: 0.063) G_GAN: 0.814 G_GAN_Feat: 1.116 G_ID: 0.102 G_Rec: 0.440 D_GP: 0.133 D_real: 0.816 D_fake: 0.253 +(epoch: 486, iters: 4126, time: 0.063) G_GAN: 0.036 G_GAN_Feat: 0.999 G_ID: 0.108 G_Rec: 0.383 D_GP: 0.539 D_real: 0.214 D_fake: 0.971 +(epoch: 486, iters: 4526, time: 0.064) G_GAN: 0.450 G_GAN_Feat: 1.030 G_ID: 0.121 G_Rec: 0.471 D_GP: 0.036 D_real: 1.188 D_fake: 0.567 +(epoch: 486, iters: 4926, time: 0.063) G_GAN: 0.145 G_GAN_Feat: 0.708 G_ID: 0.099 G_Rec: 0.320 D_GP: 0.031 D_real: 0.925 D_fake: 0.859 +(epoch: 486, iters: 5326, time: 0.063) G_GAN: 0.442 G_GAN_Feat: 0.922 G_ID: 0.125 G_Rec: 0.438 D_GP: 0.034 D_real: 1.109 D_fake: 0.584 +(epoch: 486, iters: 5726, time: 0.063) G_GAN: 0.213 G_GAN_Feat: 0.843 G_ID: 0.101 G_Rec: 0.364 D_GP: 0.038 D_real: 0.732 D_fake: 0.787 +(epoch: 486, iters: 6126, time: 0.064) G_GAN: 0.500 G_GAN_Feat: 1.057 G_ID: 0.110 G_Rec: 0.403 D_GP: 0.041 D_real: 0.848 D_fake: 0.503 +(epoch: 486, iters: 6526, time: 0.063) G_GAN: 0.641 G_GAN_Feat: 0.813 G_ID: 0.100 G_Rec: 0.301 D_GP: 0.036 D_real: 1.368 D_fake: 0.372 +(epoch: 486, iters: 6926, time: 0.063) G_GAN: 0.341 G_GAN_Feat: 1.255 G_ID: 0.114 G_Rec: 0.485 D_GP: 0.885 D_real: 0.329 D_fake: 0.661 +(epoch: 486, iters: 7326, time: 0.063) G_GAN: 0.110 G_GAN_Feat: 0.766 G_ID: 0.078 G_Rec: 0.309 D_GP: 0.032 D_real: 0.798 D_fake: 0.890 +(epoch: 486, iters: 7726, time: 0.064) G_GAN: 0.737 G_GAN_Feat: 1.005 G_ID: 0.114 G_Rec: 0.427 D_GP: 0.059 D_real: 1.191 D_fake: 0.290 +(epoch: 486, iters: 8126, time: 0.063) G_GAN: 0.492 G_GAN_Feat: 0.938 G_ID: 0.110 G_Rec: 0.341 D_GP: 0.041 D_real: 0.883 D_fake: 0.516 +(epoch: 486, iters: 8526, time: 0.063) G_GAN: 0.545 G_GAN_Feat: 0.945 G_ID: 0.147 G_Rec: 0.455 D_GP: 0.034 D_real: 1.256 D_fake: 0.469 +(epoch: 487, iters: 318, time: 0.063) G_GAN: 0.374 G_GAN_Feat: 0.731 G_ID: 0.106 G_Rec: 0.297 D_GP: 0.040 D_real: 1.270 D_fake: 0.627 +(epoch: 487, iters: 718, time: 0.064) G_GAN: 0.353 G_GAN_Feat: 1.037 G_ID: 0.100 G_Rec: 0.424 D_GP: 0.040 D_real: 0.618 D_fake: 0.649 +(epoch: 487, iters: 1118, time: 0.063) G_GAN: 0.226 G_GAN_Feat: 0.649 G_ID: 0.082 G_Rec: 0.285 D_GP: 0.031 D_real: 1.199 D_fake: 0.776 +(epoch: 487, iters: 1518, time: 0.063) G_GAN: 0.459 G_GAN_Feat: 0.875 G_ID: 0.113 G_Rec: 0.407 D_GP: 0.032 D_real: 1.110 D_fake: 0.568 +(epoch: 487, iters: 1918, time: 0.063) G_GAN: -0.023 G_GAN_Feat: 0.732 G_ID: 0.086 G_Rec: 0.321 D_GP: 0.042 D_real: 0.785 D_fake: 1.023 +(epoch: 487, iters: 2318, time: 0.064) G_GAN: 0.183 G_GAN_Feat: 0.917 G_ID: 0.150 G_Rec: 0.374 D_GP: 0.043 D_real: 0.740 D_fake: 0.821 +(epoch: 487, iters: 2718, time: 0.063) G_GAN: 0.270 G_GAN_Feat: 0.739 G_ID: 0.106 G_Rec: 0.312 D_GP: 0.040 D_real: 1.162 D_fake: 0.734 +(epoch: 487, iters: 3118, time: 0.063) G_GAN: 0.478 G_GAN_Feat: 1.032 G_ID: 0.123 G_Rec: 0.467 D_GP: 0.041 D_real: 1.008 D_fake: 0.534 +(epoch: 487, iters: 3518, time: 0.063) G_GAN: 0.430 G_GAN_Feat: 0.746 G_ID: 0.093 G_Rec: 0.316 D_GP: 0.030 D_real: 1.426 D_fake: 0.573 +(epoch: 487, iters: 3918, time: 0.064) G_GAN: 0.267 G_GAN_Feat: 1.059 G_ID: 0.109 G_Rec: 0.444 D_GP: 0.042 D_real: 0.619 D_fake: 0.735 +(epoch: 487, iters: 4318, time: 0.063) G_GAN: 0.688 G_GAN_Feat: 0.883 G_ID: 0.092 G_Rec: 0.325 D_GP: 0.056 D_real: 1.289 D_fake: 0.570 +(epoch: 487, iters: 4718, time: 0.063) G_GAN: 0.674 G_GAN_Feat: 1.163 G_ID: 0.124 G_Rec: 0.468 D_GP: 0.093 D_real: 0.458 D_fake: 0.363 +(epoch: 487, iters: 5118, time: 0.063) G_GAN: -0.061 G_GAN_Feat: 0.788 G_ID: 0.100 G_Rec: 0.279 D_GP: 0.032 D_real: 0.618 D_fake: 1.061 +(epoch: 487, iters: 5518, time: 0.064) G_GAN: 0.270 G_GAN_Feat: 0.904 G_ID: 0.122 G_Rec: 0.422 D_GP: 0.029 D_real: 0.963 D_fake: 0.730 +(epoch: 487, iters: 5918, time: 0.063) G_GAN: 0.190 G_GAN_Feat: 0.729 G_ID: 0.083 G_Rec: 0.285 D_GP: 0.039 D_real: 0.964 D_fake: 0.810 +(epoch: 487, iters: 6318, time: 0.063) G_GAN: 0.443 G_GAN_Feat: 0.989 G_ID: 0.106 G_Rec: 0.381 D_GP: 0.040 D_real: 1.167 D_fake: 0.562 +(epoch: 487, iters: 6718, time: 0.063) G_GAN: 0.198 G_GAN_Feat: 0.735 G_ID: 0.102 G_Rec: 0.304 D_GP: 0.030 D_real: 1.115 D_fake: 0.802 +(epoch: 487, iters: 7118, time: 0.064) G_GAN: 0.418 G_GAN_Feat: 0.972 G_ID: 0.108 G_Rec: 0.432 D_GP: 0.044 D_real: 0.973 D_fake: 0.591 +(epoch: 487, iters: 7518, time: 0.063) G_GAN: 0.321 G_GAN_Feat: 0.786 G_ID: 0.092 G_Rec: 0.345 D_GP: 0.038 D_real: 0.938 D_fake: 0.681 +(epoch: 487, iters: 7918, time: 0.063) G_GAN: 0.629 G_GAN_Feat: 0.940 G_ID: 0.109 G_Rec: 0.421 D_GP: 0.031 D_real: 1.276 D_fake: 0.398 +(epoch: 487, iters: 8318, time: 0.063) G_GAN: 0.423 G_GAN_Feat: 0.709 G_ID: 0.082 G_Rec: 0.295 D_GP: 0.033 D_real: 1.379 D_fake: 0.580 +(epoch: 488, iters: 110, time: 0.064) G_GAN: 0.408 G_GAN_Feat: 0.991 G_ID: 0.094 G_Rec: 0.455 D_GP: 0.036 D_real: 1.045 D_fake: 0.593 +(epoch: 488, iters: 510, time: 0.063) G_GAN: 0.042 G_GAN_Feat: 0.728 G_ID: 0.108 G_Rec: 0.315 D_GP: 0.034 D_real: 0.865 D_fake: 0.959 +(epoch: 488, iters: 910, time: 0.063) G_GAN: 0.275 G_GAN_Feat: 0.932 G_ID: 0.117 G_Rec: 0.424 D_GP: 0.041 D_real: 0.884 D_fake: 0.728 +(epoch: 488, iters: 1310, time: 0.063) G_GAN: 0.150 G_GAN_Feat: 0.690 G_ID: 0.122 G_Rec: 0.293 D_GP: 0.031 D_real: 0.965 D_fake: 0.850 +(epoch: 488, iters: 1710, time: 0.064) G_GAN: 0.336 G_GAN_Feat: 1.097 G_ID: 0.128 G_Rec: 0.476 D_GP: 0.379 D_real: 0.408 D_fake: 0.666 +(epoch: 488, iters: 2110, time: 0.063) G_GAN: 0.067 G_GAN_Feat: 0.755 G_ID: 0.107 G_Rec: 0.315 D_GP: 0.041 D_real: 0.810 D_fake: 0.933 +(epoch: 488, iters: 2510, time: 0.063) G_GAN: 0.666 G_GAN_Feat: 1.091 G_ID: 0.117 G_Rec: 0.466 D_GP: 0.058 D_real: 0.756 D_fake: 0.358 +(epoch: 488, iters: 2910, time: 0.063) G_GAN: 0.176 G_GAN_Feat: 0.744 G_ID: 0.108 G_Rec: 0.271 D_GP: 0.037 D_real: 0.864 D_fake: 0.824 +(epoch: 488, iters: 3310, time: 0.064) G_GAN: 0.938 G_GAN_Feat: 1.050 G_ID: 0.123 G_Rec: 0.509 D_GP: 0.031 D_real: 1.410 D_fake: 0.277 +(epoch: 488, iters: 3710, time: 0.063) G_GAN: 0.187 G_GAN_Feat: 0.741 G_ID: 0.086 G_Rec: 0.277 D_GP: 0.037 D_real: 1.124 D_fake: 0.814 +(epoch: 488, iters: 4110, time: 0.063) G_GAN: 0.662 G_GAN_Feat: 1.027 G_ID: 0.114 G_Rec: 0.433 D_GP: 0.030 D_real: 1.183 D_fake: 0.361 +(epoch: 488, iters: 4510, time: 0.063) G_GAN: 0.435 G_GAN_Feat: 0.827 G_ID: 0.096 G_Rec: 0.310 D_GP: 0.041 D_real: 1.158 D_fake: 0.608 +(epoch: 488, iters: 4910, time: 0.064) G_GAN: 0.598 G_GAN_Feat: 1.177 G_ID: 0.129 G_Rec: 0.469 D_GP: 0.111 D_real: 0.482 D_fake: 0.508 +(epoch: 488, iters: 5310, time: 0.063) G_GAN: 0.160 G_GAN_Feat: 0.905 G_ID: 0.110 G_Rec: 0.364 D_GP: 0.100 D_real: 0.511 D_fake: 0.843 +(epoch: 488, iters: 5710, time: 0.063) G_GAN: 0.596 G_GAN_Feat: 1.242 G_ID: 0.127 G_Rec: 0.465 D_GP: 1.181 D_real: 0.366 D_fake: 0.467 +(epoch: 488, iters: 6110, time: 0.063) G_GAN: 0.145 G_GAN_Feat: 0.829 G_ID: 0.096 G_Rec: 0.320 D_GP: 0.038 D_real: 0.802 D_fake: 0.856 +(epoch: 488, iters: 6510, time: 0.064) G_GAN: 0.781 G_GAN_Feat: 0.982 G_ID: 0.112 G_Rec: 0.422 D_GP: 0.039 D_real: 1.312 D_fake: 0.270 +(epoch: 488, iters: 6910, time: 0.063) G_GAN: 0.060 G_GAN_Feat: 1.035 G_ID: 0.108 G_Rec: 0.389 D_GP: 0.047 D_real: 0.897 D_fake: 0.959 +(epoch: 488, iters: 7310, time: 0.063) G_GAN: 0.301 G_GAN_Feat: 1.058 G_ID: 0.124 G_Rec: 0.455 D_GP: 0.030 D_real: 0.950 D_fake: 0.705 +(epoch: 488, iters: 7710, time: 0.063) G_GAN: -0.017 G_GAN_Feat: 0.760 G_ID: 0.109 G_Rec: 0.312 D_GP: 0.042 D_real: 0.741 D_fake: 1.017 +(epoch: 488, iters: 8110, time: 0.064) G_GAN: 0.609 G_GAN_Feat: 0.978 G_ID: 0.119 G_Rec: 0.410 D_GP: 0.037 D_real: 1.072 D_fake: 0.407 +(epoch: 488, iters: 8510, time: 0.063) G_GAN: 0.562 G_GAN_Feat: 0.834 G_ID: 0.105 G_Rec: 0.321 D_GP: 0.072 D_real: 1.383 D_fake: 0.523 +(epoch: 489, iters: 302, time: 0.063) G_GAN: 0.295 G_GAN_Feat: 1.036 G_ID: 0.116 G_Rec: 0.433 D_GP: 0.052 D_real: 0.309 D_fake: 0.736 +(epoch: 489, iters: 702, time: 0.063) G_GAN: 0.340 G_GAN_Feat: 0.765 G_ID: 0.110 G_Rec: 0.294 D_GP: 0.031 D_real: 1.113 D_fake: 0.661 +(epoch: 489, iters: 1102, time: 0.064) G_GAN: 0.546 G_GAN_Feat: 1.124 G_ID: 0.102 G_Rec: 0.441 D_GP: 0.036 D_real: 0.540 D_fake: 0.464 +(epoch: 489, iters: 1502, time: 0.063) G_GAN: -0.085 G_GAN_Feat: 0.768 G_ID: 0.116 G_Rec: 0.349 D_GP: 0.033 D_real: 0.800 D_fake: 1.086 +(epoch: 489, iters: 1902, time: 0.063) G_GAN: 0.549 G_GAN_Feat: 0.912 G_ID: 0.106 G_Rec: 0.418 D_GP: 0.028 D_real: 1.235 D_fake: 0.470 +(epoch: 489, iters: 2302, time: 0.063) G_GAN: 0.070 G_GAN_Feat: 0.735 G_ID: 0.115 G_Rec: 0.321 D_GP: 0.039 D_real: 0.981 D_fake: 0.931 +(epoch: 489, iters: 2702, time: 0.064) G_GAN: 0.377 G_GAN_Feat: 0.889 G_ID: 0.097 G_Rec: 0.423 D_GP: 0.031 D_real: 1.095 D_fake: 0.624 +(epoch: 489, iters: 3102, time: 0.063) G_GAN: 0.262 G_GAN_Feat: 0.710 G_ID: 0.090 G_Rec: 0.299 D_GP: 0.033 D_real: 1.117 D_fake: 0.739 +(epoch: 489, iters: 3502, time: 0.063) G_GAN: 0.189 G_GAN_Feat: 1.008 G_ID: 0.119 G_Rec: 0.473 D_GP: 0.067 D_real: 0.711 D_fake: 0.813 +(epoch: 489, iters: 3902, time: 0.063) G_GAN: -0.144 G_GAN_Feat: 0.781 G_ID: 0.114 G_Rec: 0.346 D_GP: 0.034 D_real: 0.651 D_fake: 1.144 +(epoch: 489, iters: 4302, time: 0.064) G_GAN: 0.414 G_GAN_Feat: 0.923 G_ID: 0.122 G_Rec: 0.415 D_GP: 0.034 D_real: 1.032 D_fake: 0.593 +(epoch: 489, iters: 4702, time: 0.063) G_GAN: 0.230 G_GAN_Feat: 0.796 G_ID: 0.093 G_Rec: 0.327 D_GP: 0.041 D_real: 0.959 D_fake: 0.771 +(epoch: 489, iters: 5102, time: 0.063) G_GAN: 0.271 G_GAN_Feat: 1.069 G_ID: 0.115 G_Rec: 0.477 D_GP: 0.084 D_real: 0.604 D_fake: 0.730 +(epoch: 489, iters: 5502, time: 0.063) G_GAN: 0.196 G_GAN_Feat: 0.762 G_ID: 0.090 G_Rec: 0.289 D_GP: 0.040 D_real: 0.770 D_fake: 0.804 +(epoch: 489, iters: 5902, time: 0.064) G_GAN: 0.740 G_GAN_Feat: 1.083 G_ID: 0.120 G_Rec: 0.481 D_GP: 0.051 D_real: 0.814 D_fake: 0.331 +(epoch: 489, iters: 6302, time: 0.063) G_GAN: 0.219 G_GAN_Feat: 0.814 G_ID: 0.089 G_Rec: 0.298 D_GP: 0.035 D_real: 0.939 D_fake: 0.781 +(epoch: 489, iters: 6702, time: 0.063) G_GAN: 0.518 G_GAN_Feat: 1.055 G_ID: 0.108 G_Rec: 0.408 D_GP: 0.070 D_real: 0.610 D_fake: 0.492 +(epoch: 489, iters: 7102, time: 0.063) G_GAN: 0.169 G_GAN_Feat: 0.868 G_ID: 0.093 G_Rec: 0.289 D_GP: 0.035 D_real: 0.645 D_fake: 0.831 +(epoch: 489, iters: 7502, time: 0.063) G_GAN: 0.237 G_GAN_Feat: 0.786 G_ID: 0.122 G_Rec: 0.392 D_GP: 0.028 D_real: 1.042 D_fake: 0.763 +(epoch: 489, iters: 7902, time: 0.063) G_GAN: 0.020 G_GAN_Feat: 0.646 G_ID: 0.101 G_Rec: 0.307 D_GP: 0.031 D_real: 0.896 D_fake: 0.981 +(epoch: 489, iters: 8302, time: 0.063) G_GAN: 0.362 G_GAN_Feat: 0.808 G_ID: 0.115 G_Rec: 0.396 D_GP: 0.032 D_real: 1.128 D_fake: 0.640 +(epoch: 489, iters: 8702, time: 0.064) G_GAN: -0.050 G_GAN_Feat: 0.632 G_ID: 0.086 G_Rec: 0.272 D_GP: 0.032 D_real: 0.869 D_fake: 1.050 +(epoch: 490, iters: 494, time: 0.063) G_GAN: 0.242 G_GAN_Feat: 0.869 G_ID: 0.130 G_Rec: 0.412 D_GP: 0.044 D_real: 0.739 D_fake: 0.760 +(epoch: 490, iters: 894, time: 0.063) G_GAN: 0.038 G_GAN_Feat: 0.692 G_ID: 0.092 G_Rec: 0.313 D_GP: 0.038 D_real: 0.988 D_fake: 0.962 +(epoch: 490, iters: 1294, time: 0.063) G_GAN: 0.514 G_GAN_Feat: 0.975 G_ID: 0.114 G_Rec: 0.446 D_GP: 0.040 D_real: 1.014 D_fake: 0.513 +(epoch: 490, iters: 1694, time: 0.064) G_GAN: 0.224 G_GAN_Feat: 0.706 G_ID: 0.082 G_Rec: 0.323 D_GP: 0.049 D_real: 1.063 D_fake: 0.778 +(epoch: 490, iters: 2094, time: 0.063) G_GAN: 0.347 G_GAN_Feat: 0.981 G_ID: 0.129 G_Rec: 0.453 D_GP: 0.053 D_real: 0.739 D_fake: 0.657 +(epoch: 490, iters: 2494, time: 0.063) G_GAN: 0.309 G_GAN_Feat: 0.745 G_ID: 0.109 G_Rec: 0.313 D_GP: 0.058 D_real: 0.988 D_fake: 0.692 +(epoch: 490, iters: 2894, time: 0.063) G_GAN: 0.637 G_GAN_Feat: 0.925 G_ID: 0.117 G_Rec: 0.414 D_GP: 0.037 D_real: 1.333 D_fake: 0.379 +(epoch: 490, iters: 3294, time: 0.064) G_GAN: 0.080 G_GAN_Feat: 0.861 G_ID: 0.109 G_Rec: 0.336 D_GP: 0.126 D_real: 0.293 D_fake: 0.922 +(epoch: 490, iters: 3694, time: 0.063) G_GAN: 0.398 G_GAN_Feat: 0.984 G_ID: 0.113 G_Rec: 0.389 D_GP: 0.037 D_real: 1.004 D_fake: 0.614 +(epoch: 490, iters: 4094, time: 0.063) G_GAN: 0.390 G_GAN_Feat: 0.753 G_ID: 0.086 G_Rec: 0.304 D_GP: 0.039 D_real: 1.216 D_fake: 0.611 +(epoch: 490, iters: 4494, time: 0.063) G_GAN: 0.730 G_GAN_Feat: 1.030 G_ID: 0.120 G_Rec: 0.444 D_GP: 0.043 D_real: 1.261 D_fake: 0.332 +(epoch: 490, iters: 4894, time: 0.064) G_GAN: 0.371 G_GAN_Feat: 0.763 G_ID: 0.099 G_Rec: 0.301 D_GP: 0.043 D_real: 1.116 D_fake: 0.630 +(epoch: 490, iters: 5294, time: 0.063) G_GAN: 0.238 G_GAN_Feat: 0.941 G_ID: 0.120 G_Rec: 0.461 D_GP: 0.025 D_real: 1.019 D_fake: 0.762 +(epoch: 490, iters: 5694, time: 0.063) G_GAN: 0.148 G_GAN_Feat: 0.716 G_ID: 0.114 G_Rec: 0.300 D_GP: 0.035 D_real: 0.957 D_fake: 0.852 +(epoch: 490, iters: 6094, time: 0.063) G_GAN: 0.377 G_GAN_Feat: 0.915 G_ID: 0.104 G_Rec: 0.408 D_GP: 0.037 D_real: 1.091 D_fake: 0.624 +(epoch: 490, iters: 6494, time: 0.064) G_GAN: 0.357 G_GAN_Feat: 0.814 G_ID: 0.096 G_Rec: 0.316 D_GP: 0.092 D_real: 1.053 D_fake: 0.649 +(epoch: 490, iters: 6894, time: 0.063) G_GAN: 0.503 G_GAN_Feat: 1.103 G_ID: 0.123 G_Rec: 0.451 D_GP: 0.084 D_real: 0.506 D_fake: 0.501 +(epoch: 490, iters: 7294, time: 0.063) G_GAN: 0.414 G_GAN_Feat: 0.848 G_ID: 0.098 G_Rec: 0.320 D_GP: 0.076 D_real: 0.754 D_fake: 0.617 +(epoch: 490, iters: 7694, time: 0.063) G_GAN: 0.463 G_GAN_Feat: 1.001 G_ID: 0.142 G_Rec: 0.428 D_GP: 0.062 D_real: 0.831 D_fake: 0.549 +(epoch: 490, iters: 8094, time: 0.064) G_GAN: 0.317 G_GAN_Feat: 0.827 G_ID: 0.107 G_Rec: 0.313 D_GP: 0.052 D_real: 0.745 D_fake: 0.689 +(epoch: 490, iters: 8494, time: 0.063) G_GAN: 0.461 G_GAN_Feat: 0.945 G_ID: 0.118 G_Rec: 0.433 D_GP: 0.028 D_real: 1.237 D_fake: 0.542 +(epoch: 491, iters: 286, time: 0.063) G_GAN: 0.125 G_GAN_Feat: 0.699 G_ID: 0.081 G_Rec: 0.294 D_GP: 0.034 D_real: 0.984 D_fake: 0.875 +(epoch: 491, iters: 686, time: 0.063) G_GAN: 0.302 G_GAN_Feat: 0.950 G_ID: 0.115 G_Rec: 0.416 D_GP: 0.040 D_real: 0.913 D_fake: 0.707 +(epoch: 491, iters: 1086, time: 0.064) G_GAN: 0.231 G_GAN_Feat: 0.839 G_ID: 0.088 G_Rec: 0.357 D_GP: 0.059 D_real: 0.849 D_fake: 0.781 +(epoch: 491, iters: 1486, time: 0.063) G_GAN: 0.794 G_GAN_Feat: 1.043 G_ID: 0.125 G_Rec: 0.426 D_GP: 0.053 D_real: 1.051 D_fake: 0.279 +(epoch: 491, iters: 1886, time: 0.063) G_GAN: 0.106 G_GAN_Feat: 0.948 G_ID: 0.083 G_Rec: 0.347 D_GP: 0.360 D_real: 0.236 D_fake: 0.910 +(epoch: 491, iters: 2286, time: 0.063) G_GAN: 0.696 G_GAN_Feat: 1.123 G_ID: 0.116 G_Rec: 0.448 D_GP: 0.053 D_real: 0.997 D_fake: 0.363 +(epoch: 491, iters: 2686, time: 0.064) G_GAN: 0.299 G_GAN_Feat: 0.962 G_ID: 0.092 G_Rec: 0.351 D_GP: 0.208 D_real: 0.488 D_fake: 0.704 +(epoch: 491, iters: 3086, time: 0.063) G_GAN: 0.466 G_GAN_Feat: 0.963 G_ID: 0.104 G_Rec: 0.453 D_GP: 0.031 D_real: 1.006 D_fake: 0.564 +(epoch: 491, iters: 3486, time: 0.063) G_GAN: 0.252 G_GAN_Feat: 0.738 G_ID: 0.118 G_Rec: 0.295 D_GP: 0.039 D_real: 1.039 D_fake: 0.748 +(epoch: 491, iters: 3886, time: 0.063) G_GAN: 0.375 G_GAN_Feat: 1.025 G_ID: 0.153 G_Rec: 0.421 D_GP: 0.058 D_real: 0.633 D_fake: 0.638 +(epoch: 491, iters: 4286, time: 0.064) G_GAN: 0.322 G_GAN_Feat: 0.903 G_ID: 0.107 G_Rec: 0.336 D_GP: 0.272 D_real: 0.302 D_fake: 0.881 +(epoch: 491, iters: 4686, time: 0.063) G_GAN: 0.488 G_GAN_Feat: 0.977 G_ID: 0.128 G_Rec: 0.462 D_GP: 0.042 D_real: 1.142 D_fake: 0.516 +(epoch: 491, iters: 5086, time: 0.063) G_GAN: 0.217 G_GAN_Feat: 0.686 G_ID: 0.087 G_Rec: 0.312 D_GP: 0.034 D_real: 1.060 D_fake: 0.784 +(epoch: 491, iters: 5486, time: 0.063) G_GAN: 0.280 G_GAN_Feat: 1.025 G_ID: 0.122 G_Rec: 0.472 D_GP: 0.068 D_real: 0.730 D_fake: 0.722 +(epoch: 491, iters: 5886, time: 0.064) G_GAN: 0.010 G_GAN_Feat: 0.797 G_ID: 0.090 G_Rec: 0.339 D_GP: 0.040 D_real: 0.761 D_fake: 0.990 +(epoch: 491, iters: 6286, time: 0.063) G_GAN: 0.430 G_GAN_Feat: 0.891 G_ID: 0.155 G_Rec: 0.410 D_GP: 0.036 D_real: 1.041 D_fake: 0.580 +(epoch: 491, iters: 6686, time: 0.063) G_GAN: 0.008 G_GAN_Feat: 0.814 G_ID: 0.101 G_Rec: 0.345 D_GP: 0.044 D_real: 0.611 D_fake: 0.992 +(epoch: 491, iters: 7086, time: 0.063) G_GAN: 0.459 G_GAN_Feat: 1.068 G_ID: 0.122 G_Rec: 0.468 D_GP: 0.215 D_real: 0.515 D_fake: 0.548 +(epoch: 491, iters: 7486, time: 0.064) G_GAN: 0.161 G_GAN_Feat: 0.840 G_ID: 0.099 G_Rec: 0.372 D_GP: 0.060 D_real: 0.654 D_fake: 0.840 +(epoch: 491, iters: 7886, time: 0.063) G_GAN: 0.373 G_GAN_Feat: 1.348 G_ID: 0.117 G_Rec: 0.484 D_GP: 0.061 D_real: 1.202 D_fake: 0.647 +(epoch: 491, iters: 8286, time: 0.063) G_GAN: 0.002 G_GAN_Feat: 0.649 G_ID: 0.088 G_Rec: 0.305 D_GP: 0.024 D_real: 0.994 D_fake: 0.998 +(epoch: 491, iters: 8686, time: 0.063) G_GAN: 0.321 G_GAN_Feat: 0.925 G_ID: 0.121 G_Rec: 0.458 D_GP: 0.032 D_real: 0.933 D_fake: 0.688 +(epoch: 492, iters: 478, time: 0.064) G_GAN: 0.168 G_GAN_Feat: 0.741 G_ID: 0.081 G_Rec: 0.337 D_GP: 0.042 D_real: 1.025 D_fake: 0.854 +(epoch: 492, iters: 878, time: 0.063) G_GAN: 0.076 G_GAN_Feat: 1.001 G_ID: 0.126 G_Rec: 0.489 D_GP: 0.072 D_real: 0.599 D_fake: 0.925 +(epoch: 492, iters: 1278, time: 0.063) G_GAN: 0.011 G_GAN_Feat: 0.725 G_ID: 0.086 G_Rec: 0.306 D_GP: 0.060 D_real: 0.835 D_fake: 0.989 +(epoch: 492, iters: 1678, time: 0.063) G_GAN: 0.361 G_GAN_Feat: 0.902 G_ID: 0.106 G_Rec: 0.404 D_GP: 0.040 D_real: 0.947 D_fake: 0.647 +(epoch: 492, iters: 2078, time: 0.064) G_GAN: 0.173 G_GAN_Feat: 0.791 G_ID: 0.094 G_Rec: 0.306 D_GP: 0.050 D_real: 0.865 D_fake: 0.827 +(epoch: 492, iters: 2478, time: 0.063) G_GAN: 0.652 G_GAN_Feat: 0.963 G_ID: 0.105 G_Rec: 0.426 D_GP: 0.035 D_real: 1.286 D_fake: 0.380 +(epoch: 492, iters: 2878, time: 0.063) G_GAN: 0.323 G_GAN_Feat: 0.844 G_ID: 0.099 G_Rec: 0.333 D_GP: 0.078 D_real: 0.729 D_fake: 0.677 +(epoch: 492, iters: 3278, time: 0.063) G_GAN: 0.064 G_GAN_Feat: 0.949 G_ID: 0.129 G_Rec: 0.453 D_GP: 0.037 D_real: 0.893 D_fake: 0.937 +(epoch: 492, iters: 3678, time: 0.064) G_GAN: -0.023 G_GAN_Feat: 0.700 G_ID: 0.111 G_Rec: 0.287 D_GP: 0.029 D_real: 0.859 D_fake: 1.023 +(epoch: 492, iters: 4078, time: 0.063) G_GAN: 0.362 G_GAN_Feat: 1.028 G_ID: 0.121 G_Rec: 0.431 D_GP: 0.084 D_real: 0.717 D_fake: 0.646 +(epoch: 492, iters: 4478, time: 0.063) G_GAN: 0.342 G_GAN_Feat: 0.648 G_ID: 0.103 G_Rec: 0.269 D_GP: 0.033 D_real: 1.262 D_fake: 0.659 +(epoch: 492, iters: 4878, time: 0.063) G_GAN: 0.398 G_GAN_Feat: 1.145 G_ID: 0.128 G_Rec: 0.435 D_GP: 0.770 D_real: 0.119 D_fake: 0.711 +(epoch: 492, iters: 5278, time: 0.064) G_GAN: 0.013 G_GAN_Feat: 0.884 G_ID: 0.093 G_Rec: 0.342 D_GP: 0.239 D_real: 0.311 D_fake: 0.988 +(epoch: 492, iters: 5678, time: 0.063) G_GAN: 0.536 G_GAN_Feat: 1.070 G_ID: 0.115 G_Rec: 0.489 D_GP: 0.034 D_real: 1.039 D_fake: 0.471 +(epoch: 492, iters: 6078, time: 0.063) G_GAN: 0.028 G_GAN_Feat: 0.897 G_ID: 0.129 G_Rec: 0.304 D_GP: 0.033 D_real: 0.447 D_fake: 0.972 +(epoch: 492, iters: 6478, time: 0.063) G_GAN: 0.418 G_GAN_Feat: 1.356 G_ID: 0.125 G_Rec: 0.462 D_GP: 0.105 D_real: 0.071 D_fake: 0.602 +(epoch: 492, iters: 6878, time: 0.064) G_GAN: 0.720 G_GAN_Feat: 0.858 G_ID: 0.095 G_Rec: 0.332 D_GP: 0.065 D_real: 1.349 D_fake: 0.654 +(epoch: 492, iters: 7278, time: 0.063) G_GAN: 0.238 G_GAN_Feat: 1.224 G_ID: 0.109 G_Rec: 0.466 D_GP: 0.076 D_real: 0.135 D_fake: 0.766 +(epoch: 492, iters: 7678, time: 0.063) G_GAN: 0.496 G_GAN_Feat: 0.838 G_ID: 0.091 G_Rec: 0.318 D_GP: 0.057 D_real: 1.113 D_fake: 0.510 +(epoch: 492, iters: 8078, time: 0.063) G_GAN: 0.370 G_GAN_Feat: 1.084 G_ID: 0.123 G_Rec: 0.559 D_GP: 0.035 D_real: 0.979 D_fake: 0.632 +(epoch: 492, iters: 8478, time: 0.064) G_GAN: 0.261 G_GAN_Feat: 0.794 G_ID: 0.115 G_Rec: 0.327 D_GP: 0.030 D_real: 1.123 D_fake: 0.740 +(epoch: 493, iters: 270, time: 0.063) G_GAN: 0.690 G_GAN_Feat: 1.012 G_ID: 0.114 G_Rec: 0.438 D_GP: 0.034 D_real: 1.082 D_fake: 0.378 +(epoch: 493, iters: 670, time: 0.063) G_GAN: 0.064 G_GAN_Feat: 0.807 G_ID: 0.084 G_Rec: 0.331 D_GP: 0.078 D_real: 0.733 D_fake: 0.936 +(epoch: 493, iters: 1070, time: 0.063) G_GAN: 0.579 G_GAN_Feat: 1.063 G_ID: 0.138 G_Rec: 0.465 D_GP: 0.050 D_real: 0.843 D_fake: 0.428 +(epoch: 493, iters: 1470, time: 0.064) G_GAN: 0.360 G_GAN_Feat: 0.828 G_ID: 0.104 G_Rec: 0.323 D_GP: 0.081 D_real: 0.831 D_fake: 0.640 +(epoch: 493, iters: 1870, time: 0.063) G_GAN: 0.697 G_GAN_Feat: 1.058 G_ID: 0.114 G_Rec: 0.442 D_GP: 0.065 D_real: 1.052 D_fake: 0.345 +(epoch: 493, iters: 2270, time: 0.063) G_GAN: 0.498 G_GAN_Feat: 0.922 G_ID: 0.124 G_Rec: 0.307 D_GP: 0.059 D_real: 0.951 D_fake: 0.681 +(epoch: 493, iters: 2670, time: 0.063) G_GAN: 0.858 G_GAN_Feat: 0.988 G_ID: 0.110 G_Rec: 0.440 D_GP: 0.031 D_real: 1.446 D_fake: 0.206 +(epoch: 493, iters: 3070, time: 0.064) G_GAN: 0.483 G_GAN_Feat: 0.887 G_ID: 0.105 G_Rec: 0.309 D_GP: 0.050 D_real: 1.212 D_fake: 0.524 +(epoch: 493, iters: 3470, time: 0.063) G_GAN: 0.985 G_GAN_Feat: 1.131 G_ID: 0.115 G_Rec: 0.472 D_GP: 0.039 D_real: 1.577 D_fake: 0.206 +(epoch: 493, iters: 3870, time: 0.063) G_GAN: 0.329 G_GAN_Feat: 0.840 G_ID: 0.099 G_Rec: 0.323 D_GP: 0.030 D_real: 1.109 D_fake: 0.671 +(epoch: 493, iters: 4270, time: 0.063) G_GAN: 0.387 G_GAN_Feat: 1.070 G_ID: 0.126 G_Rec: 0.509 D_GP: 0.032 D_real: 1.034 D_fake: 0.622 +(epoch: 493, iters: 4670, time: 0.064) G_GAN: 0.263 G_GAN_Feat: 0.709 G_ID: 0.100 G_Rec: 0.287 D_GP: 0.035 D_real: 1.169 D_fake: 0.742 +(epoch: 493, iters: 5070, time: 0.063) G_GAN: 0.360 G_GAN_Feat: 0.968 G_ID: 0.109 G_Rec: 0.419 D_GP: 0.079 D_real: 0.768 D_fake: 0.642 +(epoch: 493, iters: 5470, time: 0.063) G_GAN: 0.345 G_GAN_Feat: 0.813 G_ID: 0.096 G_Rec: 0.292 D_GP: 0.127 D_real: 0.784 D_fake: 0.656 +(epoch: 493, iters: 5870, time: 0.063) G_GAN: 0.804 G_GAN_Feat: 0.976 G_ID: 0.120 G_Rec: 0.395 D_GP: 0.041 D_real: 1.347 D_fake: 0.254 +(epoch: 493, iters: 6270, time: 0.064) G_GAN: 0.426 G_GAN_Feat: 0.729 G_ID: 0.094 G_Rec: 0.311 D_GP: 0.032 D_real: 1.359 D_fake: 0.577 +(epoch: 493, iters: 6670, time: 0.063) G_GAN: 0.724 G_GAN_Feat: 1.152 G_ID: 0.149 G_Rec: 0.497 D_GP: 0.037 D_real: 1.048 D_fake: 0.376 +(epoch: 493, iters: 7070, time: 0.063) G_GAN: 0.531 G_GAN_Feat: 0.787 G_ID: 0.109 G_Rec: 0.283 D_GP: 0.038 D_real: 1.233 D_fake: 0.553 +(epoch: 493, iters: 7470, time: 0.063) G_GAN: 0.397 G_GAN_Feat: 0.843 G_ID: 0.123 G_Rec: 0.405 D_GP: 0.031 D_real: 1.164 D_fake: 0.611 +(epoch: 493, iters: 7870, time: 0.064) G_GAN: 0.151 G_GAN_Feat: 0.740 G_ID: 0.095 G_Rec: 0.330 D_GP: 0.033 D_real: 1.025 D_fake: 0.849 +(epoch: 493, iters: 8270, time: 0.063) G_GAN: 0.207 G_GAN_Feat: 0.884 G_ID: 0.121 G_Rec: 0.413 D_GP: 0.032 D_real: 0.985 D_fake: 0.794 +(epoch: 493, iters: 8670, time: 0.063) G_GAN: -0.003 G_GAN_Feat: 0.803 G_ID: 0.099 G_Rec: 0.325 D_GP: 0.081 D_real: 0.617 D_fake: 1.003 +(epoch: 494, iters: 462, time: 0.063) G_GAN: 0.633 G_GAN_Feat: 1.021 G_ID: 0.107 G_Rec: 0.481 D_GP: 0.035 D_real: 1.196 D_fake: 0.404 +(epoch: 494, iters: 862, time: 0.064) G_GAN: 0.189 G_GAN_Feat: 0.656 G_ID: 0.102 G_Rec: 0.284 D_GP: 0.030 D_real: 1.124 D_fake: 0.811 +(epoch: 494, iters: 1262, time: 0.063) G_GAN: 0.209 G_GAN_Feat: 0.950 G_ID: 0.121 G_Rec: 0.437 D_GP: 0.034 D_real: 0.859 D_fake: 0.792 +(epoch: 494, iters: 1662, time: 0.063) G_GAN: 0.082 G_GAN_Feat: 0.779 G_ID: 0.107 G_Rec: 0.318 D_GP: 0.063 D_real: 0.683 D_fake: 0.921 +(epoch: 494, iters: 2062, time: 0.063) G_GAN: 0.532 G_GAN_Feat: 1.008 G_ID: 0.144 G_Rec: 0.460 D_GP: 0.053 D_real: 1.003 D_fake: 0.486 +(epoch: 494, iters: 2462, time: 0.064) G_GAN: 0.270 G_GAN_Feat: 0.771 G_ID: 0.094 G_Rec: 0.305 D_GP: 0.029 D_real: 1.060 D_fake: 0.731 +(epoch: 494, iters: 2862, time: 0.063) G_GAN: 0.478 G_GAN_Feat: 1.028 G_ID: 0.117 G_Rec: 0.445 D_GP: 0.050 D_real: 0.687 D_fake: 0.538 +(epoch: 494, iters: 3262, time: 0.063) G_GAN: -0.159 G_GAN_Feat: 0.765 G_ID: 0.096 G_Rec: 0.293 D_GP: 0.103 D_real: 0.571 D_fake: 1.159 +(epoch: 494, iters: 3662, time: 0.063) G_GAN: 0.641 G_GAN_Feat: 1.048 G_ID: 0.127 G_Rec: 0.481 D_GP: 0.053 D_real: 1.055 D_fake: 0.384 +(epoch: 494, iters: 4062, time: 0.063) G_GAN: 0.313 G_GAN_Feat: 0.643 G_ID: 0.086 G_Rec: 0.326 D_GP: 0.025 D_real: 1.242 D_fake: 0.688 +(epoch: 494, iters: 4462, time: 0.063) G_GAN: 0.579 G_GAN_Feat: 0.904 G_ID: 0.117 G_Rec: 0.448 D_GP: 0.028 D_real: 1.324 D_fake: 0.429 +(epoch: 494, iters: 4862, time: 0.063) G_GAN: 0.245 G_GAN_Feat: 0.660 G_ID: 0.102 G_Rec: 0.276 D_GP: 0.028 D_real: 1.190 D_fake: 0.755 +(epoch: 494, iters: 5262, time: 0.064) G_GAN: 0.404 G_GAN_Feat: 0.951 G_ID: 0.112 G_Rec: 0.419 D_GP: 0.054 D_real: 0.988 D_fake: 0.600 +(epoch: 494, iters: 5662, time: 0.063) G_GAN: 0.296 G_GAN_Feat: 0.733 G_ID: 0.089 G_Rec: 0.290 D_GP: 0.045 D_real: 1.120 D_fake: 0.705 +(epoch: 494, iters: 6062, time: 0.063) G_GAN: 0.505 G_GAN_Feat: 0.954 G_ID: 0.125 G_Rec: 0.435 D_GP: 0.036 D_real: 1.129 D_fake: 0.503 +(epoch: 494, iters: 6462, time: 0.063) G_GAN: 0.437 G_GAN_Feat: 0.744 G_ID: 0.085 G_Rec: 0.309 D_GP: 0.038 D_real: 1.212 D_fake: 0.565 +(epoch: 494, iters: 6862, time: 0.064) G_GAN: 0.461 G_GAN_Feat: 0.941 G_ID: 0.108 G_Rec: 0.429 D_GP: 0.035 D_real: 1.113 D_fake: 0.541 +(epoch: 494, iters: 7262, time: 0.063) G_GAN: 0.290 G_GAN_Feat: 0.720 G_ID: 0.092 G_Rec: 0.310 D_GP: 0.040 D_real: 1.078 D_fake: 0.711 +(epoch: 494, iters: 7662, time: 0.063) G_GAN: 0.308 G_GAN_Feat: 0.973 G_ID: 0.119 G_Rec: 0.400 D_GP: 0.056 D_real: 0.795 D_fake: 0.693 +(epoch: 494, iters: 8062, time: 0.063) G_GAN: 0.304 G_GAN_Feat: 0.771 G_ID: 0.092 G_Rec: 0.304 D_GP: 0.057 D_real: 0.987 D_fake: 0.703 +(epoch: 494, iters: 8462, time: 0.064) G_GAN: 0.364 G_GAN_Feat: 1.012 G_ID: 0.119 G_Rec: 0.475 D_GP: 0.031 D_real: 0.870 D_fake: 0.638 +(epoch: 495, iters: 254, time: 0.063) G_GAN: 0.114 G_GAN_Feat: 0.757 G_ID: 0.097 G_Rec: 0.282 D_GP: 0.028 D_real: 0.919 D_fake: 0.886 +(epoch: 495, iters: 654, time: 0.063) G_GAN: 0.613 G_GAN_Feat: 1.002 G_ID: 0.105 G_Rec: 0.414 D_GP: 0.034 D_real: 1.305 D_fake: 0.410 +(epoch: 495, iters: 1054, time: 0.063) G_GAN: 0.263 G_GAN_Feat: 0.864 G_ID: 0.087 G_Rec: 0.315 D_GP: 0.052 D_real: 0.577 D_fake: 0.739 +(epoch: 495, iters: 1454, time: 0.064) G_GAN: 0.439 G_GAN_Feat: 1.079 G_ID: 0.114 G_Rec: 0.471 D_GP: 0.044 D_real: 0.980 D_fake: 0.569 +(epoch: 495, iters: 1854, time: 0.063) G_GAN: 0.358 G_GAN_Feat: 0.839 G_ID: 0.100 G_Rec: 0.314 D_GP: 0.096 D_real: 0.701 D_fake: 0.646 +(epoch: 495, iters: 2254, time: 0.063) G_GAN: 0.430 G_GAN_Feat: 0.928 G_ID: 0.106 G_Rec: 0.414 D_GP: 0.033 D_real: 1.122 D_fake: 0.580 +(epoch: 495, iters: 2654, time: 0.063) G_GAN: 0.339 G_GAN_Feat: 0.761 G_ID: 0.111 G_Rec: 0.323 D_GP: 0.036 D_real: 1.138 D_fake: 0.662 +(epoch: 495, iters: 3054, time: 0.064) G_GAN: 0.557 G_GAN_Feat: 1.147 G_ID: 0.122 G_Rec: 0.458 D_GP: 0.262 D_real: 0.395 D_fake: 0.465 +(epoch: 495, iters: 3454, time: 0.063) G_GAN: 0.141 G_GAN_Feat: 0.946 G_ID: 0.090 G_Rec: 0.308 D_GP: 0.055 D_real: 0.222 D_fake: 0.862 +(epoch: 495, iters: 3854, time: 0.063) G_GAN: 0.448 G_GAN_Feat: 0.881 G_ID: 0.118 G_Rec: 0.441 D_GP: 0.029 D_real: 1.198 D_fake: 0.559 +(epoch: 495, iters: 4254, time: 0.063) G_GAN: 0.143 G_GAN_Feat: 0.644 G_ID: 0.101 G_Rec: 0.324 D_GP: 0.028 D_real: 1.057 D_fake: 0.857 +(epoch: 495, iters: 4654, time: 0.064) G_GAN: 0.505 G_GAN_Feat: 0.861 G_ID: 0.112 G_Rec: 0.436 D_GP: 0.034 D_real: 1.042 D_fake: 0.512 +(epoch: 495, iters: 5054, time: 0.063) G_GAN: 0.197 G_GAN_Feat: 0.666 G_ID: 0.104 G_Rec: 0.304 D_GP: 0.033 D_real: 1.058 D_fake: 0.803 +(epoch: 495, iters: 5454, time: 0.064) G_GAN: 0.357 G_GAN_Feat: 0.893 G_ID: 0.117 G_Rec: 0.424 D_GP: 0.042 D_real: 1.062 D_fake: 0.644 +(epoch: 495, iters: 5854, time: 0.064) G_GAN: 0.127 G_GAN_Feat: 0.686 G_ID: 0.104 G_Rec: 0.314 D_GP: 0.034 D_real: 0.967 D_fake: 0.873 +(epoch: 495, iters: 6254, time: 0.064) G_GAN: 0.124 G_GAN_Feat: 0.861 G_ID: 0.118 G_Rec: 0.403 D_GP: 0.038 D_real: 0.899 D_fake: 0.876 +(epoch: 495, iters: 6654, time: 0.064) G_GAN: -0.079 G_GAN_Feat: 0.752 G_ID: 0.104 G_Rec: 0.341 D_GP: 0.046 D_real: 0.662 D_fake: 1.079 +(epoch: 495, iters: 7054, time: 0.064) G_GAN: 0.345 G_GAN_Feat: 0.914 G_ID: 0.124 G_Rec: 0.420 D_GP: 0.048 D_real: 0.907 D_fake: 0.670 +(epoch: 495, iters: 7454, time: 0.064) G_GAN: -0.182 G_GAN_Feat: 0.740 G_ID: 0.108 G_Rec: 0.319 D_GP: 0.039 D_real: 0.605 D_fake: 1.182 +(epoch: 495, iters: 7854, time: 0.064) G_GAN: 0.418 G_GAN_Feat: 0.975 G_ID: 0.115 G_Rec: 0.454 D_GP: 0.063 D_real: 0.789 D_fake: 0.615 +(epoch: 495, iters: 8254, time: 0.064) G_GAN: -0.126 G_GAN_Feat: 0.756 G_ID: 0.099 G_Rec: 0.355 D_GP: 0.108 D_real: 0.550 D_fake: 1.126 +(epoch: 495, iters: 8654, time: 0.064) G_GAN: 0.318 G_GAN_Feat: 1.027 G_ID: 0.113 G_Rec: 0.448 D_GP: 0.089 D_real: 0.641 D_fake: 0.691 +(epoch: 496, iters: 446, time: 0.064) G_GAN: 0.122 G_GAN_Feat: 0.726 G_ID: 0.112 G_Rec: 0.306 D_GP: 0.049 D_real: 0.906 D_fake: 0.878 +(epoch: 496, iters: 846, time: 0.064) G_GAN: 0.500 G_GAN_Feat: 0.955 G_ID: 0.113 G_Rec: 0.398 D_GP: 0.050 D_real: 1.036 D_fake: 0.508 +(epoch: 496, iters: 1246, time: 0.064) G_GAN: 0.035 G_GAN_Feat: 0.739 G_ID: 0.095 G_Rec: 0.308 D_GP: 0.053 D_real: 0.950 D_fake: 0.965 +(epoch: 496, iters: 1646, time: 0.064) G_GAN: 0.481 G_GAN_Feat: 0.865 G_ID: 0.107 G_Rec: 0.391 D_GP: 0.036 D_real: 1.251 D_fake: 0.524 +(epoch: 496, iters: 2046, time: 0.064) G_GAN: 0.107 G_GAN_Feat: 0.845 G_ID: 0.100 G_Rec: 0.357 D_GP: 0.077 D_real: 0.487 D_fake: 0.894 +(epoch: 496, iters: 2446, time: 0.064) G_GAN: 0.322 G_GAN_Feat: 1.054 G_ID: 0.140 G_Rec: 0.488 D_GP: 0.110 D_real: 0.452 D_fake: 0.681 +(epoch: 496, iters: 2846, time: 0.064) G_GAN: 0.208 G_GAN_Feat: 0.773 G_ID: 0.099 G_Rec: 0.313 D_GP: 0.074 D_real: 1.015 D_fake: 0.792 +(epoch: 496, iters: 3246, time: 0.063) G_GAN: 0.624 G_GAN_Feat: 0.936 G_ID: 0.116 G_Rec: 0.441 D_GP: 0.070 D_real: 1.160 D_fake: 0.401 +(epoch: 496, iters: 3646, time: 0.063) G_GAN: 0.282 G_GAN_Feat: 0.796 G_ID: 0.092 G_Rec: 0.327 D_GP: 0.050 D_real: 0.868 D_fake: 0.720 +(epoch: 496, iters: 4046, time: 0.064) G_GAN: 0.402 G_GAN_Feat: 0.964 G_ID: 0.132 G_Rec: 0.407 D_GP: 0.050 D_real: 0.837 D_fake: 0.599 +(epoch: 496, iters: 4446, time: 0.064) G_GAN: 0.040 G_GAN_Feat: 0.899 G_ID: 0.093 G_Rec: 0.333 D_GP: 0.117 D_real: 0.413 D_fake: 0.961 +(epoch: 496, iters: 4846, time: 0.064) G_GAN: 0.369 G_GAN_Feat: 0.930 G_ID: 0.109 G_Rec: 0.450 D_GP: 0.026 D_real: 1.146 D_fake: 0.635 +(epoch: 496, iters: 5246, time: 0.064) G_GAN: 0.059 G_GAN_Feat: 0.708 G_ID: 0.095 G_Rec: 0.314 D_GP: 0.031 D_real: 0.904 D_fake: 0.941 +(epoch: 496, iters: 5646, time: 0.064) G_GAN: 0.607 G_GAN_Feat: 0.944 G_ID: 0.109 G_Rec: 0.506 D_GP: 0.046 D_real: 1.106 D_fake: 0.431 +(epoch: 496, iters: 6046, time: 0.063) G_GAN: -0.131 G_GAN_Feat: 0.826 G_ID: 0.105 G_Rec: 0.325 D_GP: 0.115 D_real: 0.415 D_fake: 1.132 +(epoch: 496, iters: 6446, time: 0.063) G_GAN: 0.561 G_GAN_Feat: 0.956 G_ID: 0.103 G_Rec: 0.416 D_GP: 0.041 D_real: 1.115 D_fake: 0.448 +(epoch: 496, iters: 6846, time: 0.063) G_GAN: 0.166 G_GAN_Feat: 0.739 G_ID: 0.106 G_Rec: 0.309 D_GP: 0.036 D_real: 1.014 D_fake: 0.835 +(epoch: 496, iters: 7246, time: 0.064) G_GAN: 0.489 G_GAN_Feat: 0.932 G_ID: 0.117 G_Rec: 0.396 D_GP: 0.039 D_real: 1.177 D_fake: 0.517 +(epoch: 496, iters: 7646, time: 0.063) G_GAN: 0.297 G_GAN_Feat: 0.802 G_ID: 0.097 G_Rec: 0.291 D_GP: 0.039 D_real: 1.088 D_fake: 0.703 +(epoch: 496, iters: 8046, time: 0.063) G_GAN: 0.501 G_GAN_Feat: 1.032 G_ID: 0.129 G_Rec: 0.479 D_GP: 0.078 D_real: 0.903 D_fake: 0.520 +(epoch: 496, iters: 8446, time: 0.063) G_GAN: 0.098 G_GAN_Feat: 0.862 G_ID: 0.109 G_Rec: 0.325 D_GP: 0.057 D_real: 0.582 D_fake: 0.902 +(epoch: 497, iters: 238, time: 0.064) G_GAN: 0.420 G_GAN_Feat: 0.937 G_ID: 0.123 G_Rec: 0.379 D_GP: 0.068 D_real: 1.017 D_fake: 0.583 +(epoch: 497, iters: 638, time: 0.063) G_GAN: 0.251 G_GAN_Feat: 0.721 G_ID: 0.093 G_Rec: 0.314 D_GP: 0.035 D_real: 1.019 D_fake: 0.749 +(epoch: 497, iters: 1038, time: 0.063) G_GAN: 0.638 G_GAN_Feat: 0.963 G_ID: 0.112 G_Rec: 0.428 D_GP: 0.038 D_real: 1.276 D_fake: 0.384 +(epoch: 497, iters: 1438, time: 0.063) G_GAN: 0.373 G_GAN_Feat: 0.717 G_ID: 0.101 G_Rec: 0.289 D_GP: 0.032 D_real: 1.278 D_fake: 0.628 +(epoch: 497, iters: 1838, time: 0.064) G_GAN: 0.575 G_GAN_Feat: 1.022 G_ID: 0.119 G_Rec: 0.459 D_GP: 0.096 D_real: 1.108 D_fake: 0.442 +(epoch: 497, iters: 2238, time: 0.063) G_GAN: 0.196 G_GAN_Feat: 0.802 G_ID: 0.095 G_Rec: 0.295 D_GP: 0.035 D_real: 0.827 D_fake: 0.807 +(epoch: 497, iters: 2638, time: 0.063) G_GAN: 0.370 G_GAN_Feat: 1.019 G_ID: 0.120 G_Rec: 0.403 D_GP: 0.039 D_real: 0.622 D_fake: 0.639 +(epoch: 497, iters: 3038, time: 0.063) G_GAN: 0.088 G_GAN_Feat: 0.783 G_ID: 0.100 G_Rec: 0.289 D_GP: 0.032 D_real: 0.900 D_fake: 0.912 +(epoch: 497, iters: 3438, time: 0.064) G_GAN: 0.858 G_GAN_Feat: 1.025 G_ID: 0.113 G_Rec: 0.430 D_GP: 0.056 D_real: 1.212 D_fake: 0.209 +(epoch: 497, iters: 3838, time: 0.063) G_GAN: -0.051 G_GAN_Feat: 0.969 G_ID: 0.098 G_Rec: 0.371 D_GP: 0.205 D_real: 0.223 D_fake: 1.052 +(epoch: 497, iters: 4238, time: 0.063) G_GAN: 0.387 G_GAN_Feat: 0.927 G_ID: 0.126 G_Rec: 0.408 D_GP: 0.036 D_real: 1.054 D_fake: 0.646 +(epoch: 497, iters: 4638, time: 0.063) G_GAN: 0.384 G_GAN_Feat: 0.772 G_ID: 0.088 G_Rec: 0.287 D_GP: 0.037 D_real: 1.199 D_fake: 0.618 +(epoch: 497, iters: 5038, time: 0.064) G_GAN: 0.278 G_GAN_Feat: 1.063 G_ID: 0.119 G_Rec: 0.461 D_GP: 0.036 D_real: 0.829 D_fake: 0.723 +(epoch: 497, iters: 5438, time: 0.063) G_GAN: 0.103 G_GAN_Feat: 0.841 G_ID: 0.098 G_Rec: 0.305 D_GP: 0.058 D_real: 0.611 D_fake: 0.898 +(epoch: 497, iters: 5838, time: 0.063) G_GAN: 0.773 G_GAN_Feat: 1.073 G_ID: 0.114 G_Rec: 0.428 D_GP: 0.037 D_real: 0.833 D_fake: 0.252 +(epoch: 497, iters: 6238, time: 0.063) G_GAN: 0.376 G_GAN_Feat: 0.630 G_ID: 0.097 G_Rec: 0.290 D_GP: 0.025 D_real: 1.344 D_fake: 0.625 +(epoch: 497, iters: 6638, time: 0.064) G_GAN: 0.424 G_GAN_Feat: 0.845 G_ID: 0.109 G_Rec: 0.383 D_GP: 0.031 D_real: 1.157 D_fake: 0.578 +(epoch: 497, iters: 7038, time: 0.063) G_GAN: 0.139 G_GAN_Feat: 0.785 G_ID: 0.092 G_Rec: 0.348 D_GP: 0.050 D_real: 0.946 D_fake: 0.864 +(epoch: 497, iters: 7438, time: 0.063) G_GAN: 0.635 G_GAN_Feat: 1.005 G_ID: 0.108 G_Rec: 0.429 D_GP: 0.043 D_real: 1.141 D_fake: 0.383 +(epoch: 497, iters: 7838, time: 0.063) G_GAN: 0.278 G_GAN_Feat: 0.764 G_ID: 0.082 G_Rec: 0.316 D_GP: 0.045 D_real: 0.981 D_fake: 0.723 +(epoch: 497, iters: 8238, time: 0.064) G_GAN: 0.656 G_GAN_Feat: 1.051 G_ID: 0.108 G_Rec: 0.462 D_GP: 0.048 D_real: 1.188 D_fake: 0.386 +(epoch: 497, iters: 8638, time: 0.063) G_GAN: 0.452 G_GAN_Feat: 0.668 G_ID: 0.092 G_Rec: 0.281 D_GP: 0.027 D_real: 1.371 D_fake: 0.548 +(epoch: 498, iters: 430, time: 0.063) G_GAN: 0.531 G_GAN_Feat: 1.017 G_ID: 0.110 G_Rec: 0.456 D_GP: 0.062 D_real: 1.048 D_fake: 0.478 +(epoch: 498, iters: 830, time: 0.063) G_GAN: 0.148 G_GAN_Feat: 0.787 G_ID: 0.091 G_Rec: 0.325 D_GP: 0.096 D_real: 0.922 D_fake: 0.853 +(epoch: 498, iters: 1230, time: 0.064) G_GAN: 0.929 G_GAN_Feat: 0.970 G_ID: 0.110 G_Rec: 0.405 D_GP: 0.038 D_real: 1.544 D_fake: 0.147 +(epoch: 498, iters: 1630, time: 0.063) G_GAN: 0.108 G_GAN_Feat: 0.795 G_ID: 0.104 G_Rec: 0.323 D_GP: 0.032 D_real: 0.693 D_fake: 0.892 +(epoch: 498, iters: 2030, time: 0.063) G_GAN: 0.594 G_GAN_Feat: 0.988 G_ID: 0.124 G_Rec: 0.416 D_GP: 0.040 D_real: 1.075 D_fake: 0.431 +(epoch: 498, iters: 2430, time: 0.063) G_GAN: 0.154 G_GAN_Feat: 0.781 G_ID: 0.093 G_Rec: 0.296 D_GP: 0.034 D_real: 0.928 D_fake: 0.846 +(epoch: 498, iters: 2830, time: 0.064) G_GAN: 0.353 G_GAN_Feat: 1.035 G_ID: 0.107 G_Rec: 0.412 D_GP: 0.035 D_real: 0.649 D_fake: 0.665 +(epoch: 498, iters: 3230, time: 0.063) G_GAN: 0.020 G_GAN_Feat: 0.748 G_ID: 0.097 G_Rec: 0.286 D_GP: 0.032 D_real: 0.796 D_fake: 0.980 +(epoch: 498, iters: 3630, time: 0.063) G_GAN: 0.804 G_GAN_Feat: 0.951 G_ID: 0.134 G_Rec: 0.441 D_GP: 0.042 D_real: 1.472 D_fake: 0.289 +(epoch: 498, iters: 4030, time: 0.063) G_GAN: 0.121 G_GAN_Feat: 0.777 G_ID: 0.090 G_Rec: 0.340 D_GP: 0.029 D_real: 1.153 D_fake: 0.880 +(epoch: 498, iters: 4430, time: 0.064) G_GAN: 0.305 G_GAN_Feat: 0.821 G_ID: 0.112 G_Rec: 0.392 D_GP: 0.027 D_real: 1.129 D_fake: 0.701 +(epoch: 498, iters: 4830, time: 0.063) G_GAN: 0.007 G_GAN_Feat: 0.619 G_ID: 0.085 G_Rec: 0.288 D_GP: 0.025 D_real: 0.934 D_fake: 0.994 +(epoch: 498, iters: 5230, time: 0.063) G_GAN: 0.253 G_GAN_Feat: 0.852 G_ID: 0.106 G_Rec: 0.404 D_GP: 0.037 D_real: 0.918 D_fake: 0.752 +(epoch: 498, iters: 5630, time: 0.063) G_GAN: -0.196 G_GAN_Feat: 0.791 G_ID: 0.096 G_Rec: 0.358 D_GP: 0.033 D_real: 0.628 D_fake: 1.196 +(epoch: 498, iters: 6030, time: 0.064) G_GAN: 0.025 G_GAN_Feat: 0.829 G_ID: 0.127 G_Rec: 0.387 D_GP: 0.041 D_real: 0.575 D_fake: 0.976 +(epoch: 498, iters: 6430, time: 0.063) G_GAN: -0.163 G_GAN_Feat: 0.611 G_ID: 0.090 G_Rec: 0.290 D_GP: 0.029 D_real: 0.699 D_fake: 1.163 +(epoch: 498, iters: 6830, time: 0.063) G_GAN: 0.274 G_GAN_Feat: 0.871 G_ID: 0.113 G_Rec: 0.425 D_GP: 0.035 D_real: 0.926 D_fake: 0.748 +(epoch: 498, iters: 7230, time: 0.063) G_GAN: -0.364 G_GAN_Feat: 0.713 G_ID: 0.109 G_Rec: 0.316 D_GP: 0.040 D_real: 0.488 D_fake: 1.364 +(epoch: 498, iters: 7630, time: 0.064) G_GAN: 0.386 G_GAN_Feat: 0.945 G_ID: 0.113 G_Rec: 0.440 D_GP: 0.039 D_real: 0.986 D_fake: 0.632 +(epoch: 498, iters: 8030, time: 0.063) G_GAN: -0.089 G_GAN_Feat: 0.759 G_ID: 0.101 G_Rec: 0.311 D_GP: 0.097 D_real: 0.585 D_fake: 1.089 +(epoch: 498, iters: 8430, time: 0.063) G_GAN: 0.519 G_GAN_Feat: 1.044 G_ID: 0.116 G_Rec: 0.465 D_GP: 0.040 D_real: 1.035 D_fake: 0.497 +(epoch: 499, iters: 222, time: 0.063) G_GAN: 0.299 G_GAN_Feat: 0.680 G_ID: 0.090 G_Rec: 0.295 D_GP: 0.042 D_real: 1.174 D_fake: 0.701 +(epoch: 499, iters: 622, time: 0.063) G_GAN: 0.548 G_GAN_Feat: 0.915 G_ID: 0.117 G_Rec: 0.393 D_GP: 0.032 D_real: 1.210 D_fake: 0.454 +(epoch: 499, iters: 1022, time: 0.063) G_GAN: 0.248 G_GAN_Feat: 0.741 G_ID: 0.111 G_Rec: 0.291 D_GP: 0.030 D_real: 1.092 D_fake: 0.752 +(epoch: 499, iters: 1422, time: 0.063) G_GAN: 0.504 G_GAN_Feat: 1.030 G_ID: 0.114 G_Rec: 0.441 D_GP: 0.036 D_real: 0.938 D_fake: 0.500 +(epoch: 499, iters: 1822, time: 0.064) G_GAN: -0.153 G_GAN_Feat: 0.760 G_ID: 0.097 G_Rec: 0.311 D_GP: 0.033 D_real: 0.845 D_fake: 1.153 +(epoch: 499, iters: 2222, time: 0.063) G_GAN: 0.336 G_GAN_Feat: 1.036 G_ID: 0.107 G_Rec: 0.467 D_GP: 0.057 D_real: 0.848 D_fake: 0.670 +(epoch: 499, iters: 2622, time: 0.063) G_GAN: 0.070 G_GAN_Feat: 0.780 G_ID: 0.111 G_Rec: 0.341 D_GP: 0.047 D_real: 0.834 D_fake: 0.933 +(epoch: 499, iters: 3022, time: 0.063) G_GAN: 0.282 G_GAN_Feat: 0.934 G_ID: 0.126 G_Rec: 0.427 D_GP: 0.037 D_real: 0.857 D_fake: 0.722 +(epoch: 499, iters: 3422, time: 0.064) G_GAN: 0.117 G_GAN_Feat: 0.787 G_ID: 0.098 G_Rec: 0.313 D_GP: 0.067 D_real: 0.788 D_fake: 0.884 +(epoch: 499, iters: 3822, time: 0.063) G_GAN: 0.532 G_GAN_Feat: 1.041 G_ID: 0.123 G_Rec: 0.456 D_GP: 0.060 D_real: 0.887 D_fake: 0.497 +(epoch: 499, iters: 4222, time: 0.063) G_GAN: -0.197 G_GAN_Feat: 0.811 G_ID: 0.115 G_Rec: 0.306 D_GP: 0.077 D_real: 0.553 D_fake: 1.198 +(epoch: 499, iters: 4622, time: 0.063) G_GAN: 0.461 G_GAN_Feat: 1.094 G_ID: 0.108 G_Rec: 0.457 D_GP: 0.063 D_real: 0.883 D_fake: 0.547 +(epoch: 499, iters: 5022, time: 0.064) G_GAN: 0.297 G_GAN_Feat: 0.827 G_ID: 0.087 G_Rec: 0.304 D_GP: 0.060 D_real: 0.839 D_fake: 0.753 +(epoch: 499, iters: 5422, time: 0.063) G_GAN: 0.360 G_GAN_Feat: 0.838 G_ID: 0.108 G_Rec: 0.385 D_GP: 0.033 D_real: 1.073 D_fake: 0.641 +(epoch: 499, iters: 5822, time: 0.063) G_GAN: 0.037 G_GAN_Feat: 0.665 G_ID: 0.097 G_Rec: 0.294 D_GP: 0.032 D_real: 0.985 D_fake: 0.963 +(epoch: 499, iters: 6222, time: 0.063) G_GAN: 0.364 G_GAN_Feat: 0.912 G_ID: 0.113 G_Rec: 0.421 D_GP: 0.049 D_real: 1.049 D_fake: 0.645 +(epoch: 499, iters: 6622, time: 0.064) G_GAN: 0.208 G_GAN_Feat: 0.676 G_ID: 0.080 G_Rec: 0.297 D_GP: 0.032 D_real: 1.083 D_fake: 0.795 +(epoch: 499, iters: 7022, time: 0.063) G_GAN: 0.386 G_GAN_Feat: 0.902 G_ID: 0.103 G_Rec: 0.435 D_GP: 0.029 D_real: 1.020 D_fake: 0.628 +(epoch: 499, iters: 7422, time: 0.063) G_GAN: 0.111 G_GAN_Feat: 0.708 G_ID: 0.103 G_Rec: 0.309 D_GP: 0.033 D_real: 0.997 D_fake: 0.889 +(epoch: 499, iters: 7822, time: 0.063) G_GAN: 0.301 G_GAN_Feat: 0.860 G_ID: 0.115 G_Rec: 0.415 D_GP: 0.031 D_real: 1.046 D_fake: 0.703 +(epoch: 499, iters: 8222, time: 0.064) G_GAN: 0.222 G_GAN_Feat: 0.732 G_ID: 0.085 G_Rec: 0.313 D_GP: 0.035 D_real: 1.033 D_fake: 0.780 +(epoch: 499, iters: 8622, time: 0.063) G_GAN: 0.985 G_GAN_Feat: 1.019 G_ID: 0.104 G_Rec: 0.466 D_GP: 0.045 D_real: 1.403 D_fake: 0.288 +(epoch: 500, iters: 414, time: 0.063) G_GAN: 0.171 G_GAN_Feat: 0.739 G_ID: 0.095 G_Rec: 0.315 D_GP: 0.038 D_real: 1.006 D_fake: 0.829 +(epoch: 500, iters: 814, time: 0.063) G_GAN: 0.383 G_GAN_Feat: 1.007 G_ID: 0.121 G_Rec: 0.411 D_GP: 0.043 D_real: 0.834 D_fake: 0.620 +(epoch: 500, iters: 1214, time: 0.064) G_GAN: 0.332 G_GAN_Feat: 0.775 G_ID: 0.091 G_Rec: 0.286 D_GP: 0.039 D_real: 1.113 D_fake: 0.683 +(epoch: 500, iters: 1614, time: 0.063) G_GAN: 0.360 G_GAN_Feat: 0.888 G_ID: 0.114 G_Rec: 0.383 D_GP: 0.031 D_real: 1.095 D_fake: 0.641 +(epoch: 500, iters: 2014, time: 0.063) G_GAN: 0.261 G_GAN_Feat: 0.677 G_ID: 0.100 G_Rec: 0.289 D_GP: 0.030 D_real: 1.114 D_fake: 0.740 +(epoch: 500, iters: 2414, time: 0.063) G_GAN: 0.539 G_GAN_Feat: 0.982 G_ID: 0.122 G_Rec: 0.390 D_GP: 0.039 D_real: 0.984 D_fake: 0.468 +(epoch: 500, iters: 2814, time: 0.064) G_GAN: 0.168 G_GAN_Feat: 0.898 G_ID: 0.123 G_Rec: 0.306 D_GP: 0.064 D_real: 0.607 D_fake: 0.832 +(epoch: 500, iters: 3214, time: 0.063) G_GAN: 0.628 G_GAN_Feat: 1.152 G_ID: 0.115 G_Rec: 0.439 D_GP: 1.355 D_real: 0.337 D_fake: 0.444 +(epoch: 500, iters: 3614, time: 0.063) G_GAN: 0.118 G_GAN_Feat: 0.673 G_ID: 0.082 G_Rec: 0.286 D_GP: 0.026 D_real: 1.173 D_fake: 0.882 +(epoch: 500, iters: 4014, time: 0.063) G_GAN: 0.391 G_GAN_Feat: 0.989 G_ID: 0.113 G_Rec: 0.427 D_GP: 0.035 D_real: 0.859 D_fake: 0.609 +(epoch: 500, iters: 4414, time: 0.064) G_GAN: 0.345 G_GAN_Feat: 0.765 G_ID: 0.081 G_Rec: 0.291 D_GP: 0.038 D_real: 1.280 D_fake: 0.669 +(epoch: 500, iters: 4814, time: 0.063) G_GAN: 0.602 G_GAN_Feat: 0.998 G_ID: 0.116 G_Rec: 0.490 D_GP: 0.039 D_real: 1.107 D_fake: 0.426 +(epoch: 500, iters: 5214, time: 0.063) G_GAN: 0.256 G_GAN_Feat: 0.785 G_ID: 0.092 G_Rec: 0.288 D_GP: 0.032 D_real: 1.012 D_fake: 0.745 +(epoch: 500, iters: 5614, time: 0.063) G_GAN: 0.795 G_GAN_Feat: 1.131 G_ID: 0.114 G_Rec: 0.525 D_GP: 0.038 D_real: 1.323 D_fake: 0.262 +(epoch: 500, iters: 6014, time: 0.064) G_GAN: 0.114 G_GAN_Feat: 0.826 G_ID: 0.098 G_Rec: 0.294 D_GP: 0.045 D_real: 0.476 D_fake: 0.886 +(epoch: 500, iters: 6414, time: 0.063) G_GAN: 0.318 G_GAN_Feat: 1.043 G_ID: 0.098 G_Rec: 0.414 D_GP: 0.044 D_real: 0.891 D_fake: 0.683 +(epoch: 500, iters: 6814, time: 0.063) G_GAN: 0.361 G_GAN_Feat: 0.713 G_ID: 0.090 G_Rec: 0.266 D_GP: 0.029 D_real: 1.273 D_fake: 0.708 +(epoch: 500, iters: 7214, time: 0.063) G_GAN: 0.339 G_GAN_Feat: 0.934 G_ID: 0.119 G_Rec: 0.413 D_GP: 0.033 D_real: 0.988 D_fake: 0.675 +(epoch: 500, iters: 7614, time: 0.064) G_GAN: 0.074 G_GAN_Feat: 0.669 G_ID: 0.093 G_Rec: 0.281 D_GP: 0.029 D_real: 0.989 D_fake: 0.926 +(epoch: 500, iters: 8014, time: 0.063) G_GAN: 0.255 G_GAN_Feat: 1.055 G_ID: 0.108 G_Rec: 0.509 D_GP: 0.033 D_real: 0.635 D_fake: 0.759 +(epoch: 500, iters: 8414, time: 0.063) G_GAN: 0.030 G_GAN_Feat: 0.729 G_ID: 0.099 G_Rec: 0.300 D_GP: 0.045 D_real: 0.925 D_fake: 0.970 +(epoch: 501, iters: 206, time: 0.063) G_GAN: 0.113 G_GAN_Feat: 0.980 G_ID: 0.114 G_Rec: 0.404 D_GP: 0.092 D_real: 0.449 D_fake: 0.888 +(epoch: 501, iters: 606, time: 0.064) G_GAN: 0.374 G_GAN_Feat: 0.883 G_ID: 0.097 G_Rec: 0.323 D_GP: 0.220 D_real: 0.596 D_fake: 0.634 +(epoch: 501, iters: 1006, time: 0.063) G_GAN: 0.712 G_GAN_Feat: 1.246 G_ID: 0.111 G_Rec: 0.519 D_GP: 0.156 D_real: 0.793 D_fake: 0.348 +(epoch: 501, iters: 1406, time: 0.063) G_GAN: 0.450 G_GAN_Feat: 0.717 G_ID: 0.092 G_Rec: 0.317 D_GP: 0.033 D_real: 1.344 D_fake: 0.553 +(epoch: 501, iters: 1806, time: 0.063) G_GAN: 0.337 G_GAN_Feat: 1.019 G_ID: 0.126 G_Rec: 0.448 D_GP: 0.090 D_real: 0.788 D_fake: 0.664 +(epoch: 501, iters: 2206, time: 0.064) G_GAN: 0.432 G_GAN_Feat: 0.708 G_ID: 0.109 G_Rec: 0.281 D_GP: 0.028 D_real: 1.368 D_fake: 0.568 +(epoch: 501, iters: 2606, time: 0.063) G_GAN: 0.466 G_GAN_Feat: 0.974 G_ID: 0.101 G_Rec: 0.430 D_GP: 0.032 D_real: 1.161 D_fake: 0.545 +(epoch: 501, iters: 3006, time: 0.063) G_GAN: 0.157 G_GAN_Feat: 0.870 G_ID: 0.092 G_Rec: 0.286 D_GP: 0.384 D_real: 0.369 D_fake: 0.844 +(epoch: 501, iters: 3406, time: 0.063) G_GAN: 0.775 G_GAN_Feat: 1.242 G_ID: 0.117 G_Rec: 0.517 D_GP: 0.137 D_real: 0.297 D_fake: 0.293 +(epoch: 501, iters: 3806, time: 0.064) G_GAN: 0.541 G_GAN_Feat: 0.984 G_ID: 0.096 G_Rec: 0.355 D_GP: 0.417 D_real: 0.294 D_fake: 0.619 +(epoch: 501, iters: 4206, time: 0.063) G_GAN: 0.789 G_GAN_Feat: 1.115 G_ID: 0.106 G_Rec: 0.471 D_GP: 0.054 D_real: 0.917 D_fake: 0.291 +(epoch: 501, iters: 4606, time: 0.063) G_GAN: 0.022 G_GAN_Feat: 0.695 G_ID: 0.085 G_Rec: 0.306 D_GP: 0.028 D_real: 1.006 D_fake: 0.978 +(epoch: 501, iters: 5006, time: 0.063) G_GAN: 0.368 G_GAN_Feat: 0.945 G_ID: 0.118 G_Rec: 0.431 D_GP: 0.029 D_real: 1.019 D_fake: 0.643 +(epoch: 501, iters: 5406, time: 0.064) G_GAN: 0.087 G_GAN_Feat: 0.666 G_ID: 0.086 G_Rec: 0.283 D_GP: 0.029 D_real: 0.998 D_fake: 0.913 +(epoch: 501, iters: 5806, time: 0.063) G_GAN: 0.217 G_GAN_Feat: 0.904 G_ID: 0.114 G_Rec: 0.410 D_GP: 0.035 D_real: 0.974 D_fake: 0.787 +(epoch: 501, iters: 6206, time: 0.063) G_GAN: -0.161 G_GAN_Feat: 0.765 G_ID: 0.095 G_Rec: 0.301 D_GP: 0.054 D_real: 0.613 D_fake: 1.161 +(epoch: 501, iters: 6606, time: 0.063) G_GAN: 0.103 G_GAN_Feat: 0.941 G_ID: 0.116 G_Rec: 0.459 D_GP: 0.049 D_real: 0.656 D_fake: 0.897 +(epoch: 501, iters: 7006, time: 0.064) G_GAN: 0.080 G_GAN_Feat: 0.715 G_ID: 0.087 G_Rec: 0.298 D_GP: 0.037 D_real: 0.919 D_fake: 0.920 +(epoch: 501, iters: 7406, time: 0.063) G_GAN: 0.542 G_GAN_Feat: 1.144 G_ID: 0.112 G_Rec: 0.481 D_GP: 0.170 D_real: 0.589 D_fake: 0.483 +(epoch: 501, iters: 7806, time: 0.063) G_GAN: 0.464 G_GAN_Feat: 0.791 G_ID: 0.091 G_Rec: 0.350 D_GP: 0.041 D_real: 1.248 D_fake: 0.537 +(epoch: 501, iters: 8206, time: 0.063) G_GAN: 0.478 G_GAN_Feat: 1.042 G_ID: 0.123 G_Rec: 0.441 D_GP: 0.158 D_real: 0.644 D_fake: 0.529 +(epoch: 501, iters: 8606, time: 0.064) G_GAN: 0.304 G_GAN_Feat: 0.824 G_ID: 0.094 G_Rec: 0.305 D_GP: 0.132 D_real: 0.726 D_fake: 0.701 +(epoch: 502, iters: 398, time: 0.063) G_GAN: 0.542 G_GAN_Feat: 0.906 G_ID: 0.110 G_Rec: 0.379 D_GP: 0.036 D_real: 1.171 D_fake: 0.468 +(epoch: 502, iters: 798, time: 0.063) G_GAN: 0.527 G_GAN_Feat: 0.782 G_ID: 0.084 G_Rec: 0.312 D_GP: 0.046 D_real: 1.259 D_fake: 0.480 +(epoch: 502, iters: 1198, time: 0.063) G_GAN: 0.584 G_GAN_Feat: 1.168 G_ID: 0.130 G_Rec: 0.484 D_GP: 0.095 D_real: 0.449 D_fake: 0.440 +(epoch: 502, iters: 1598, time: 0.064) G_GAN: 0.266 G_GAN_Feat: 0.828 G_ID: 0.095 G_Rec: 0.302 D_GP: 0.136 D_real: 0.642 D_fake: 0.737 +(epoch: 502, iters: 1998, time: 0.063) G_GAN: 0.749 G_GAN_Feat: 1.213 G_ID: 0.103 G_Rec: 0.490 D_GP: 0.068 D_real: 0.525 D_fake: 0.380 +(epoch: 502, iters: 2398, time: 0.063) G_GAN: 0.004 G_GAN_Feat: 1.121 G_ID: 0.113 G_Rec: 0.406 D_GP: 2.753 D_real: 0.450 D_fake: 0.999 +(epoch: 502, iters: 2798, time: 0.063) G_GAN: 0.236 G_GAN_Feat: 0.953 G_ID: 0.105 G_Rec: 0.497 D_GP: 0.030 D_real: 1.049 D_fake: 0.765 +(epoch: 502, iters: 3198, time: 0.064) G_GAN: 0.231 G_GAN_Feat: 0.668 G_ID: 0.086 G_Rec: 0.307 D_GP: 0.027 D_real: 1.150 D_fake: 0.769 +(epoch: 502, iters: 3598, time: 0.063) G_GAN: 0.209 G_GAN_Feat: 0.731 G_ID: 0.110 G_Rec: 0.349 D_GP: 0.028 D_real: 1.011 D_fake: 0.792 +(epoch: 502, iters: 3998, time: 0.063) G_GAN: 0.236 G_GAN_Feat: 0.677 G_ID: 0.096 G_Rec: 0.302 D_GP: 0.053 D_real: 1.105 D_fake: 0.765 +(epoch: 502, iters: 4398, time: 0.063) G_GAN: 0.370 G_GAN_Feat: 0.890 G_ID: 0.115 G_Rec: 0.420 D_GP: 0.035 D_real: 1.054 D_fake: 0.631 +(epoch: 502, iters: 4798, time: 0.064) G_GAN: 0.256 G_GAN_Feat: 0.724 G_ID: 0.093 G_Rec: 0.331 D_GP: 0.040 D_real: 1.130 D_fake: 0.745 +(epoch: 502, iters: 5198, time: 0.063) G_GAN: 0.225 G_GAN_Feat: 0.963 G_ID: 0.120 G_Rec: 0.461 D_GP: 0.045 D_real: 0.759 D_fake: 0.776 +(epoch: 502, iters: 5598, time: 0.063) G_GAN: 0.055 G_GAN_Feat: 0.794 G_ID: 0.094 G_Rec: 0.391 D_GP: 0.122 D_real: 0.620 D_fake: 0.946 +(epoch: 502, iters: 5998, time: 0.063) G_GAN: 0.283 G_GAN_Feat: 0.928 G_ID: 0.128 G_Rec: 0.422 D_GP: 0.044 D_real: 0.833 D_fake: 0.717 +(epoch: 502, iters: 6398, time: 0.064) G_GAN: 0.674 G_GAN_Feat: 0.693 G_ID: 0.097 G_Rec: 0.404 D_GP: 0.027 D_real: 1.561 D_fake: 0.359 +(epoch: 502, iters: 6798, time: 0.063) G_GAN: 0.447 G_GAN_Feat: 0.836 G_ID: 0.113 G_Rec: 0.393 D_GP: 0.030 D_real: 1.152 D_fake: 0.556 +(epoch: 502, iters: 7198, time: 0.063) G_GAN: 0.124 G_GAN_Feat: 0.747 G_ID: 0.089 G_Rec: 0.330 D_GP: 0.039 D_real: 0.922 D_fake: 0.876 +(epoch: 502, iters: 7598, time: 0.063) G_GAN: 0.384 G_GAN_Feat: 0.932 G_ID: 0.115 G_Rec: 0.459 D_GP: 0.037 D_real: 1.066 D_fake: 0.638 +(epoch: 502, iters: 7998, time: 0.064) G_GAN: 0.011 G_GAN_Feat: 0.738 G_ID: 0.084 G_Rec: 0.314 D_GP: 0.062 D_real: 0.871 D_fake: 0.989 +(epoch: 502, iters: 8398, time: 0.063) G_GAN: 0.293 G_GAN_Feat: 0.989 G_ID: 0.114 G_Rec: 0.441 D_GP: 0.041 D_real: 0.886 D_fake: 0.707 +(epoch: 503, iters: 190, time: 0.063) G_GAN: 0.209 G_GAN_Feat: 0.743 G_ID: 0.102 G_Rec: 0.309 D_GP: 0.078 D_real: 0.910 D_fake: 0.792 +(epoch: 503, iters: 590, time: 0.063) G_GAN: 0.504 G_GAN_Feat: 0.956 G_ID: 0.115 G_Rec: 0.394 D_GP: 0.069 D_real: 0.995 D_fake: 0.516 +(epoch: 503, iters: 990, time: 0.064) G_GAN: 0.121 G_GAN_Feat: 0.865 G_ID: 0.109 G_Rec: 0.341 D_GP: 0.043 D_real: 0.775 D_fake: 0.880 +(epoch: 503, iters: 1390, time: 0.063) G_GAN: 0.396 G_GAN_Feat: 0.929 G_ID: 0.102 G_Rec: 0.459 D_GP: 0.028 D_real: 1.196 D_fake: 0.605 +(epoch: 503, iters: 1790, time: 0.063) G_GAN: 0.161 G_GAN_Feat: 0.796 G_ID: 0.100 G_Rec: 0.321 D_GP: 0.043 D_real: 0.967 D_fake: 0.839 +(epoch: 503, iters: 2190, time: 0.063) G_GAN: 0.511 G_GAN_Feat: 0.980 G_ID: 0.111 G_Rec: 0.457 D_GP: 0.041 D_real: 1.062 D_fake: 0.510 +(epoch: 503, iters: 2590, time: 0.064) G_GAN: 0.191 G_GAN_Feat: 0.726 G_ID: 0.095 G_Rec: 0.306 D_GP: 0.035 D_real: 0.982 D_fake: 0.809 +(epoch: 503, iters: 2990, time: 0.063) G_GAN: 0.081 G_GAN_Feat: 0.936 G_ID: 0.114 G_Rec: 0.428 D_GP: 0.047 D_real: 0.648 D_fake: 0.919 +(epoch: 503, iters: 3390, time: 0.063) G_GAN: 0.044 G_GAN_Feat: 0.737 G_ID: 0.100 G_Rec: 0.290 D_GP: 0.046 D_real: 0.909 D_fake: 0.956 +(epoch: 503, iters: 3790, time: 0.063) G_GAN: 0.628 G_GAN_Feat: 0.912 G_ID: 0.106 G_Rec: 0.376 D_GP: 0.034 D_real: 1.443 D_fake: 0.382 +(epoch: 503, iters: 4190, time: 0.064) G_GAN: 0.077 G_GAN_Feat: 0.830 G_ID: 0.125 G_Rec: 0.399 D_GP: 0.042 D_real: 1.127 D_fake: 0.936 +(epoch: 503, iters: 4590, time: 0.063) G_GAN: 0.337 G_GAN_Feat: 0.909 G_ID: 0.118 G_Rec: 0.466 D_GP: 0.024 D_real: 1.117 D_fake: 0.667 +(epoch: 503, iters: 4990, time: 0.063) G_GAN: -0.047 G_GAN_Feat: 0.708 G_ID: 0.105 G_Rec: 0.345 D_GP: 0.027 D_real: 0.900 D_fake: 1.047 +(epoch: 503, iters: 5390, time: 0.063) G_GAN: 0.444 G_GAN_Feat: 0.813 G_ID: 0.111 G_Rec: 0.393 D_GP: 0.030 D_real: 1.195 D_fake: 0.560 +(epoch: 503, iters: 5790, time: 0.063) G_GAN: 0.141 G_GAN_Feat: 0.662 G_ID: 0.090 G_Rec: 0.353 D_GP: 0.028 D_real: 1.028 D_fake: 0.860 +(epoch: 503, iters: 6190, time: 0.063) G_GAN: 0.284 G_GAN_Feat: 0.890 G_ID: 0.127 G_Rec: 0.458 D_GP: 0.040 D_real: 0.920 D_fake: 0.718 +(epoch: 503, iters: 6590, time: 0.063) G_GAN: 0.055 G_GAN_Feat: 0.642 G_ID: 0.094 G_Rec: 0.310 D_GP: 0.045 D_real: 0.896 D_fake: 0.945 +(epoch: 503, iters: 6990, time: 0.064) G_GAN: 0.177 G_GAN_Feat: 0.914 G_ID: 0.131 G_Rec: 0.441 D_GP: 0.054 D_real: 0.769 D_fake: 0.829 +(epoch: 503, iters: 7390, time: 0.063) G_GAN: 0.215 G_GAN_Feat: 0.658 G_ID: 0.096 G_Rec: 0.312 D_GP: 0.039 D_real: 1.161 D_fake: 0.786 +(epoch: 503, iters: 7790, time: 0.063) G_GAN: 0.304 G_GAN_Feat: 0.885 G_ID: 0.124 G_Rec: 0.424 D_GP: 0.044 D_real: 1.026 D_fake: 0.698 +(epoch: 503, iters: 8190, time: 0.063) G_GAN: 0.052 G_GAN_Feat: 0.668 G_ID: 0.094 G_Rec: 0.300 D_GP: 0.040 D_real: 0.933 D_fake: 0.948 +(epoch: 503, iters: 8590, time: 0.064) G_GAN: 0.351 G_GAN_Feat: 0.867 G_ID: 0.100 G_Rec: 0.416 D_GP: 0.037 D_real: 1.056 D_fake: 0.656 +(epoch: 504, iters: 382, time: 0.063) G_GAN: 0.186 G_GAN_Feat: 0.670 G_ID: 0.082 G_Rec: 0.302 D_GP: 0.064 D_real: 1.061 D_fake: 0.815 +(epoch: 504, iters: 782, time: 0.063) G_GAN: 0.170 G_GAN_Feat: 0.904 G_ID: 0.113 G_Rec: 0.443 D_GP: 0.040 D_real: 0.699 D_fake: 0.830 +(epoch: 504, iters: 1182, time: 0.063) G_GAN: 0.007 G_GAN_Feat: 0.719 G_ID: 0.094 G_Rec: 0.310 D_GP: 0.046 D_real: 0.737 D_fake: 0.993 +(epoch: 504, iters: 1582, time: 0.064) G_GAN: 0.322 G_GAN_Feat: 0.888 G_ID: 0.128 G_Rec: 0.417 D_GP: 0.047 D_real: 0.774 D_fake: 0.688 +(epoch: 504, iters: 1982, time: 0.063) G_GAN: 0.040 G_GAN_Feat: 0.696 G_ID: 0.083 G_Rec: 0.296 D_GP: 0.042 D_real: 0.817 D_fake: 0.960 +(epoch: 504, iters: 2382, time: 0.063) G_GAN: 0.320 G_GAN_Feat: 0.961 G_ID: 0.101 G_Rec: 0.462 D_GP: 0.046 D_real: 0.896 D_fake: 0.684 +(epoch: 504, iters: 2782, time: 0.063) G_GAN: 0.155 G_GAN_Feat: 0.760 G_ID: 0.095 G_Rec: 0.339 D_GP: 0.069 D_real: 0.873 D_fake: 0.846 +(epoch: 504, iters: 3182, time: 0.064) G_GAN: 0.340 G_GAN_Feat: 0.819 G_ID: 0.113 G_Rec: 0.367 D_GP: 0.035 D_real: 1.043 D_fake: 0.661 +(epoch: 504, iters: 3582, time: 0.063) G_GAN: 0.255 G_GAN_Feat: 0.715 G_ID: 0.106 G_Rec: 0.309 D_GP: 0.045 D_real: 1.079 D_fake: 0.747 +(epoch: 504, iters: 3982, time: 0.063) G_GAN: 0.646 G_GAN_Feat: 1.013 G_ID: 0.102 G_Rec: 0.434 D_GP: 0.059 D_real: 1.049 D_fake: 0.418 +(epoch: 504, iters: 4382, time: 0.063) G_GAN: 0.145 G_GAN_Feat: 0.818 G_ID: 0.104 G_Rec: 0.366 D_GP: 0.092 D_real: 0.712 D_fake: 0.855 +(epoch: 504, iters: 4782, time: 0.064) G_GAN: 0.368 G_GAN_Feat: 0.986 G_ID: 0.131 G_Rec: 0.465 D_GP: 0.042 D_real: 0.797 D_fake: 0.636 +(epoch: 504, iters: 5182, time: 0.063) G_GAN: 0.072 G_GAN_Feat: 0.792 G_ID: 0.119 G_Rec: 0.318 D_GP: 0.211 D_real: 0.656 D_fake: 0.930 +(epoch: 504, iters: 5582, time: 0.063) G_GAN: 0.442 G_GAN_Feat: 1.085 G_ID: 0.116 G_Rec: 0.502 D_GP: 0.375 D_real: 0.667 D_fake: 0.583 +(epoch: 504, iters: 5982, time: 0.063) G_GAN: -0.199 G_GAN_Feat: 0.921 G_ID: 0.118 G_Rec: 0.393 D_GP: 0.352 D_real: 0.425 D_fake: 1.200 +(epoch: 504, iters: 6382, time: 0.064) G_GAN: 0.500 G_GAN_Feat: 0.966 G_ID: 0.115 G_Rec: 0.422 D_GP: 0.041 D_real: 1.195 D_fake: 0.513 +(epoch: 504, iters: 6782, time: 0.063) G_GAN: -0.129 G_GAN_Feat: 0.770 G_ID: 0.095 G_Rec: 0.314 D_GP: 0.044 D_real: 0.610 D_fake: 1.129 +(epoch: 504, iters: 7182, time: 0.063) G_GAN: 0.541 G_GAN_Feat: 0.944 G_ID: 0.105 G_Rec: 0.392 D_GP: 0.038 D_real: 1.195 D_fake: 0.464 +(epoch: 504, iters: 7582, time: 0.063) G_GAN: 0.340 G_GAN_Feat: 0.798 G_ID: 0.088 G_Rec: 0.287 D_GP: 0.055 D_real: 0.928 D_fake: 0.661 +(epoch: 504, iters: 7982, time: 0.064) G_GAN: 0.295 G_GAN_Feat: 0.977 G_ID: 0.121 G_Rec: 0.445 D_GP: 0.032 D_real: 0.904 D_fake: 0.711 +(epoch: 504, iters: 8382, time: 0.063) G_GAN: 0.167 G_GAN_Feat: 0.743 G_ID: 0.092 G_Rec: 0.301 D_GP: 0.032 D_real: 0.951 D_fake: 0.834 +(epoch: 505, iters: 174, time: 0.063) G_GAN: 1.143 G_GAN_Feat: 1.145 G_ID: 0.114 G_Rec: 0.448 D_GP: 0.061 D_real: 1.485 D_fake: 0.122 +(epoch: 505, iters: 574, time: 0.063) G_GAN: 0.635 G_GAN_Feat: 0.892 G_ID: 0.086 G_Rec: 0.324 D_GP: 0.054 D_real: 0.752 D_fake: 0.416 +(epoch: 505, iters: 974, time: 0.064) G_GAN: 0.343 G_GAN_Feat: 0.953 G_ID: 0.118 G_Rec: 0.447 D_GP: 0.033 D_real: 0.986 D_fake: 0.663 +(epoch: 505, iters: 1374, time: 0.063) G_GAN: 0.195 G_GAN_Feat: 0.657 G_ID: 0.099 G_Rec: 0.288 D_GP: 0.030 D_real: 1.102 D_fake: 0.807 +(epoch: 505, iters: 1774, time: 0.063) G_GAN: 0.148 G_GAN_Feat: 0.861 G_ID: 0.115 G_Rec: 0.425 D_GP: 0.031 D_real: 0.859 D_fake: 0.854 +(epoch: 505, iters: 2174, time: 0.063) G_GAN: 0.073 G_GAN_Feat: 0.665 G_ID: 0.079 G_Rec: 0.304 D_GP: 0.030 D_real: 0.933 D_fake: 0.927 +(epoch: 505, iters: 2574, time: 0.064) G_GAN: 0.312 G_GAN_Feat: 0.879 G_ID: 0.119 G_Rec: 0.423 D_GP: 0.044 D_real: 0.950 D_fake: 0.692 +(epoch: 505, iters: 2974, time: 0.063) G_GAN: -0.209 G_GAN_Feat: 0.681 G_ID: 0.113 G_Rec: 0.296 D_GP: 0.036 D_real: 0.665 D_fake: 1.209 +(epoch: 505, iters: 3374, time: 0.063) G_GAN: 0.400 G_GAN_Feat: 0.934 G_ID: 0.109 G_Rec: 0.510 D_GP: 0.050 D_real: 0.884 D_fake: 0.614 +(epoch: 505, iters: 3774, time: 0.063) G_GAN: 0.057 G_GAN_Feat: 0.751 G_ID: 0.101 G_Rec: 0.314 D_GP: 0.070 D_real: 0.801 D_fake: 0.943 +(epoch: 505, iters: 4174, time: 0.064) G_GAN: 0.221 G_GAN_Feat: 0.994 G_ID: 0.117 G_Rec: 0.450 D_GP: 0.092 D_real: 0.639 D_fake: 0.783 +(epoch: 505, iters: 4574, time: 0.063) G_GAN: 0.064 G_GAN_Feat: 0.789 G_ID: 0.095 G_Rec: 0.338 D_GP: 0.048 D_real: 0.967 D_fake: 0.937 +(epoch: 505, iters: 4974, time: 0.063) G_GAN: 0.454 G_GAN_Feat: 1.077 G_ID: 0.142 G_Rec: 0.472 D_GP: 0.301 D_real: 0.618 D_fake: 0.560 +(epoch: 505, iters: 5374, time: 0.063) G_GAN: 0.252 G_GAN_Feat: 0.741 G_ID: 0.129 G_Rec: 0.297 D_GP: 0.093 D_real: 0.849 D_fake: 0.748 +(epoch: 505, iters: 5774, time: 0.064) G_GAN: 0.378 G_GAN_Feat: 1.036 G_ID: 0.102 G_Rec: 0.470 D_GP: 0.045 D_real: 0.780 D_fake: 0.623 +(epoch: 505, iters: 6174, time: 0.063) G_GAN: 0.615 G_GAN_Feat: 0.721 G_ID: 0.104 G_Rec: 0.295 D_GP: 0.039 D_real: 1.398 D_fake: 0.402 +(epoch: 505, iters: 6574, time: 0.063) G_GAN: 0.728 G_GAN_Feat: 1.079 G_ID: 0.122 G_Rec: 0.526 D_GP: 0.059 D_real: 0.996 D_fake: 0.300 +(epoch: 505, iters: 6974, time: 0.063) G_GAN: 0.244 G_GAN_Feat: 0.881 G_ID: 0.112 G_Rec: 0.342 D_GP: 0.032 D_real: 0.967 D_fake: 0.757 +(epoch: 505, iters: 7374, time: 0.064) G_GAN: 0.163 G_GAN_Feat: 1.016 G_ID: 0.125 G_Rec: 0.465 D_GP: 0.038 D_real: 0.864 D_fake: 0.837 +(epoch: 505, iters: 7774, time: 0.063) G_GAN: 0.051 G_GAN_Feat: 0.673 G_ID: 0.088 G_Rec: 0.307 D_GP: 0.028 D_real: 0.976 D_fake: 0.949 +(epoch: 505, iters: 8174, time: 0.063) G_GAN: 0.155 G_GAN_Feat: 0.972 G_ID: 0.120 G_Rec: 0.447 D_GP: 0.042 D_real: 0.808 D_fake: 0.847 +(epoch: 505, iters: 8574, time: 0.063) G_GAN: 0.005 G_GAN_Feat: 0.781 G_ID: 0.115 G_Rec: 0.313 D_GP: 0.050 D_real: 0.696 D_fake: 0.996 +(epoch: 506, iters: 366, time: 0.064) G_GAN: 0.391 G_GAN_Feat: 0.960 G_ID: 0.132 G_Rec: 0.413 D_GP: 0.045 D_real: 0.940 D_fake: 0.613 +(epoch: 506, iters: 766, time: 0.063) G_GAN: 0.102 G_GAN_Feat: 0.786 G_ID: 0.125 G_Rec: 0.326 D_GP: 0.042 D_real: 0.806 D_fake: 0.898 +(epoch: 506, iters: 1166, time: 0.063) G_GAN: 0.674 G_GAN_Feat: 1.018 G_ID: 0.110 G_Rec: 0.461 D_GP: 0.062 D_real: 1.021 D_fake: 0.344 +(epoch: 506, iters: 1566, time: 0.063) G_GAN: 0.207 G_GAN_Feat: 0.811 G_ID: 0.103 G_Rec: 0.298 D_GP: 0.038 D_real: 0.811 D_fake: 0.794 +(epoch: 506, iters: 1966, time: 0.064) G_GAN: 0.805 G_GAN_Feat: 1.186 G_ID: 0.106 G_Rec: 0.465 D_GP: 0.052 D_real: 0.656 D_fake: 0.268 +(epoch: 506, iters: 2366, time: 0.063) G_GAN: 0.424 G_GAN_Feat: 0.927 G_ID: 0.109 G_Rec: 0.315 D_GP: 0.048 D_real: 0.563 D_fake: 0.602 +(epoch: 506, iters: 2766, time: 0.063) G_GAN: 0.317 G_GAN_Feat: 0.904 G_ID: 0.110 G_Rec: 0.464 D_GP: 0.031 D_real: 1.081 D_fake: 0.688 +(epoch: 506, iters: 3166, time: 0.063) G_GAN: 0.145 G_GAN_Feat: 0.663 G_ID: 0.098 G_Rec: 0.310 D_GP: 0.028 D_real: 1.092 D_fake: 0.855 +(epoch: 506, iters: 3566, time: 0.064) G_GAN: 0.221 G_GAN_Feat: 0.802 G_ID: 0.117 G_Rec: 0.384 D_GP: 0.028 D_real: 0.946 D_fake: 0.781 +(epoch: 506, iters: 3966, time: 0.063) G_GAN: 0.183 G_GAN_Feat: 0.682 G_ID: 0.092 G_Rec: 0.313 D_GP: 0.039 D_real: 1.059 D_fake: 0.818 +(epoch: 506, iters: 4366, time: 0.063) G_GAN: 0.153 G_GAN_Feat: 0.863 G_ID: 0.118 G_Rec: 0.405 D_GP: 0.043 D_real: 0.887 D_fake: 0.847 +(epoch: 506, iters: 4766, time: 0.063) G_GAN: -0.088 G_GAN_Feat: 0.762 G_ID: 0.125 G_Rec: 0.361 D_GP: 0.040 D_real: 0.755 D_fake: 1.088 +(epoch: 506, iters: 5166, time: 0.064) G_GAN: 0.347 G_GAN_Feat: 0.874 G_ID: 0.112 G_Rec: 0.425 D_GP: 0.036 D_real: 1.007 D_fake: 0.662 +(epoch: 506, iters: 5566, time: 0.063) G_GAN: -0.049 G_GAN_Feat: 0.676 G_ID: 0.100 G_Rec: 0.336 D_GP: 0.030 D_real: 0.821 D_fake: 1.049 +(epoch: 506, iters: 5966, time: 0.063) G_GAN: 0.121 G_GAN_Feat: 0.856 G_ID: 0.122 G_Rec: 0.400 D_GP: 0.034 D_real: 0.761 D_fake: 0.880 +(epoch: 506, iters: 6366, time: 0.063) G_GAN: 0.063 G_GAN_Feat: 0.645 G_ID: 0.087 G_Rec: 0.302 D_GP: 0.042 D_real: 0.925 D_fake: 0.937 +(epoch: 506, iters: 6766, time: 0.064) G_GAN: -0.033 G_GAN_Feat: 0.941 G_ID: 0.133 G_Rec: 0.445 D_GP: 0.042 D_real: 0.580 D_fake: 1.033 +(epoch: 506, iters: 7166, time: 0.063) G_GAN: -0.125 G_GAN_Feat: 0.662 G_ID: 0.117 G_Rec: 0.287 D_GP: 0.037 D_real: 0.767 D_fake: 1.125 +(epoch: 506, iters: 7566, time: 0.063) G_GAN: 0.372 G_GAN_Feat: 0.875 G_ID: 0.113 G_Rec: 0.422 D_GP: 0.036 D_real: 1.148 D_fake: 0.636 +(epoch: 506, iters: 7966, time: 0.063) G_GAN: -0.113 G_GAN_Feat: 0.703 G_ID: 0.103 G_Rec: 0.302 D_GP: 0.040 D_real: 0.720 D_fake: 1.113 +(epoch: 506, iters: 8366, time: 0.064) G_GAN: 0.374 G_GAN_Feat: 0.877 G_ID: 0.101 G_Rec: 0.393 D_GP: 0.034 D_real: 1.064 D_fake: 0.633 +(epoch: 507, iters: 158, time: 0.063) G_GAN: 0.378 G_GAN_Feat: 0.712 G_ID: 0.111 G_Rec: 0.323 D_GP: 0.046 D_real: 1.184 D_fake: 0.636 +(epoch: 507, iters: 558, time: 0.063) G_GAN: 0.140 G_GAN_Feat: 0.935 G_ID: 0.123 G_Rec: 0.453 D_GP: 0.070 D_real: 0.605 D_fake: 0.862 +(epoch: 507, iters: 958, time: 0.063) G_GAN: -0.099 G_GAN_Feat: 0.710 G_ID: 0.082 G_Rec: 0.298 D_GP: 0.050 D_real: 0.710 D_fake: 1.099 +(epoch: 507, iters: 1358, time: 0.064) G_GAN: 0.376 G_GAN_Feat: 0.929 G_ID: 0.113 G_Rec: 0.452 D_GP: 0.040 D_real: 1.001 D_fake: 0.639 +(epoch: 507, iters: 1758, time: 0.063) G_GAN: 0.137 G_GAN_Feat: 0.771 G_ID: 0.100 G_Rec: 0.344 D_GP: 0.057 D_real: 0.822 D_fake: 0.864 +(epoch: 507, iters: 2158, time: 0.063) G_GAN: 0.268 G_GAN_Feat: 0.876 G_ID: 0.094 G_Rec: 0.394 D_GP: 0.056 D_real: 0.984 D_fake: 0.735 +(epoch: 507, iters: 2558, time: 0.063) G_GAN: 0.082 G_GAN_Feat: 0.672 G_ID: 0.088 G_Rec: 0.287 D_GP: 0.039 D_real: 1.087 D_fake: 0.919 +(epoch: 507, iters: 2958, time: 0.064) G_GAN: 0.461 G_GAN_Feat: 0.905 G_ID: 0.106 G_Rec: 0.442 D_GP: 0.032 D_real: 1.172 D_fake: 0.545 +(epoch: 507, iters: 3358, time: 0.063) G_GAN: 0.315 G_GAN_Feat: 0.673 G_ID: 0.092 G_Rec: 0.294 D_GP: 0.031 D_real: 1.172 D_fake: 0.687 +(epoch: 507, iters: 3758, time: 0.063) G_GAN: 0.255 G_GAN_Feat: 1.055 G_ID: 0.108 G_Rec: 0.476 D_GP: 0.108 D_real: 0.580 D_fake: 0.747 +(epoch: 507, iters: 4158, time: 0.063) G_GAN: 0.087 G_GAN_Feat: 0.690 G_ID: 0.098 G_Rec: 0.320 D_GP: 0.044 D_real: 0.865 D_fake: 0.913 +(epoch: 507, iters: 4558, time: 0.064) G_GAN: 0.770 G_GAN_Feat: 0.973 G_ID: 0.112 G_Rec: 0.478 D_GP: 0.049 D_real: 1.291 D_fake: 0.313 +(epoch: 507, iters: 4958, time: 0.063) G_GAN: 0.078 G_GAN_Feat: 0.727 G_ID: 0.104 G_Rec: 0.328 D_GP: 0.050 D_real: 0.848 D_fake: 0.922 +(epoch: 507, iters: 5358, time: 0.063) G_GAN: 0.333 G_GAN_Feat: 0.970 G_ID: 0.114 G_Rec: 0.482 D_GP: 0.105 D_real: 0.653 D_fake: 0.672 +(epoch: 507, iters: 5758, time: 0.063) G_GAN: 0.196 G_GAN_Feat: 0.864 G_ID: 0.106 G_Rec: 0.360 D_GP: 0.070 D_real: 0.605 D_fake: 0.806 +(epoch: 507, iters: 6158, time: 0.064) G_GAN: 0.414 G_GAN_Feat: 1.021 G_ID: 0.118 G_Rec: 0.464 D_GP: 0.114 D_real: 0.697 D_fake: 0.594 +(epoch: 507, iters: 6558, time: 0.063) G_GAN: 0.234 G_GAN_Feat: 0.842 G_ID: 0.109 G_Rec: 0.344 D_GP: 0.119 D_real: 0.647 D_fake: 0.776 +(epoch: 507, iters: 6958, time: 0.063) G_GAN: 0.025 G_GAN_Feat: 1.013 G_ID: 0.106 G_Rec: 0.484 D_GP: 0.069 D_real: 0.541 D_fake: 0.976 +(epoch: 507, iters: 7358, time: 0.063) G_GAN: 0.103 G_GAN_Feat: 0.677 G_ID: 0.089 G_Rec: 0.315 D_GP: 0.031 D_real: 1.017 D_fake: 0.897 +(epoch: 507, iters: 7758, time: 0.064) G_GAN: 0.428 G_GAN_Feat: 0.946 G_ID: 0.111 G_Rec: 0.462 D_GP: 0.038 D_real: 1.011 D_fake: 0.584 +(epoch: 507, iters: 8158, time: 0.063) G_GAN: 0.013 G_GAN_Feat: 0.815 G_ID: 0.093 G_Rec: 0.348 D_GP: 0.129 D_real: 0.421 D_fake: 0.989 +(epoch: 507, iters: 8558, time: 0.063) G_GAN: 0.356 G_GAN_Feat: 1.043 G_ID: 0.120 G_Rec: 0.456 D_GP: 0.211 D_real: 0.767 D_fake: 0.656 +(epoch: 508, iters: 350, time: 0.063) G_GAN: 0.061 G_GAN_Feat: 0.674 G_ID: 0.088 G_Rec: 0.275 D_GP: 0.036 D_real: 0.906 D_fake: 0.939 +(epoch: 508, iters: 750, time: 0.064) G_GAN: 0.408 G_GAN_Feat: 1.028 G_ID: 0.122 G_Rec: 0.464 D_GP: 0.046 D_real: 0.826 D_fake: 0.596 +(epoch: 508, iters: 1150, time: 0.063) G_GAN: -0.104 G_GAN_Feat: 0.895 G_ID: 0.105 G_Rec: 0.354 D_GP: 0.749 D_real: 0.223 D_fake: 1.105 +(epoch: 508, iters: 1550, time: 0.063) G_GAN: 0.324 G_GAN_Feat: 0.918 G_ID: 0.117 G_Rec: 0.421 D_GP: 0.043 D_real: 0.893 D_fake: 0.680 +(epoch: 508, iters: 1950, time: 0.063) G_GAN: 0.297 G_GAN_Feat: 0.770 G_ID: 0.088 G_Rec: 0.313 D_GP: 0.105 D_real: 0.818 D_fake: 0.707 +(epoch: 508, iters: 2350, time: 0.063) G_GAN: 0.470 G_GAN_Feat: 1.096 G_ID: 0.091 G_Rec: 0.518 D_GP: 0.027 D_real: 1.159 D_fake: 0.544 +(epoch: 508, iters: 2750, time: 0.063) G_GAN: 0.098 G_GAN_Feat: 0.716 G_ID: 0.100 G_Rec: 0.321 D_GP: 0.027 D_real: 0.924 D_fake: 0.902 +(epoch: 508, iters: 3150, time: 0.063) G_GAN: 0.265 G_GAN_Feat: 0.884 G_ID: 0.110 G_Rec: 0.419 D_GP: 0.034 D_real: 0.975 D_fake: 0.737 +(epoch: 508, iters: 3550, time: 0.064) G_GAN: 0.011 G_GAN_Feat: 0.737 G_ID: 0.083 G_Rec: 0.355 D_GP: 0.057 D_real: 0.825 D_fake: 0.990 +(epoch: 508, iters: 3950, time: 0.063) G_GAN: 0.238 G_GAN_Feat: 0.832 G_ID: 0.116 G_Rec: 0.382 D_GP: 0.036 D_real: 0.974 D_fake: 0.771 +(epoch: 508, iters: 4350, time: 0.063) G_GAN: 0.097 G_GAN_Feat: 0.687 G_ID: 0.097 G_Rec: 0.297 D_GP: 0.049 D_real: 0.948 D_fake: 0.904 +(epoch: 508, iters: 4750, time: 0.063) G_GAN: 0.125 G_GAN_Feat: 0.922 G_ID: 0.109 G_Rec: 0.423 D_GP: 0.056 D_real: 0.718 D_fake: 0.876 +(epoch: 508, iters: 5150, time: 0.064) G_GAN: -0.177 G_GAN_Feat: 0.728 G_ID: 0.080 G_Rec: 0.307 D_GP: 0.052 D_real: 0.608 D_fake: 1.177 +(epoch: 508, iters: 5550, time: 0.063) G_GAN: 0.449 G_GAN_Feat: 1.037 G_ID: 0.105 G_Rec: 0.475 D_GP: 0.174 D_real: 0.660 D_fake: 0.559 +(epoch: 508, iters: 5950, time: 0.063) G_GAN: -0.175 G_GAN_Feat: 0.804 G_ID: 0.115 G_Rec: 0.355 D_GP: 0.115 D_real: 0.477 D_fake: 1.177 +(epoch: 508, iters: 6350, time: 0.063) G_GAN: 0.719 G_GAN_Feat: 1.063 G_ID: 0.121 G_Rec: 0.463 D_GP: 0.045 D_real: 1.108 D_fake: 0.315 +(epoch: 508, iters: 6750, time: 0.064) G_GAN: 0.450 G_GAN_Feat: 0.783 G_ID: 0.082 G_Rec: 0.295 D_GP: 0.049 D_real: 1.122 D_fake: 0.555 +(epoch: 508, iters: 7150, time: 0.063) G_GAN: 0.197 G_GAN_Feat: 0.959 G_ID: 0.129 G_Rec: 0.408 D_GP: 0.044 D_real: 0.859 D_fake: 0.804 +(epoch: 508, iters: 7550, time: 0.063) G_GAN: 0.042 G_GAN_Feat: 0.784 G_ID: 0.115 G_Rec: 0.320 D_GP: 0.032 D_real: 0.951 D_fake: 0.958 +(epoch: 508, iters: 7950, time: 0.063) G_GAN: 0.124 G_GAN_Feat: 0.930 G_ID: 0.116 G_Rec: 0.453 D_GP: 0.044 D_real: 0.667 D_fake: 0.878 +(epoch: 508, iters: 8350, time: 0.064) G_GAN: 0.088 G_GAN_Feat: 0.710 G_ID: 0.078 G_Rec: 0.284 D_GP: 0.068 D_real: 0.736 D_fake: 0.912 +(epoch: 509, iters: 142, time: 0.063) G_GAN: 0.557 G_GAN_Feat: 0.879 G_ID: 0.105 G_Rec: 0.373 D_GP: 0.035 D_real: 1.236 D_fake: 0.477 +(epoch: 509, iters: 542, time: 0.063) G_GAN: 0.122 G_GAN_Feat: 0.737 G_ID: 0.081 G_Rec: 0.287 D_GP: 0.034 D_real: 0.931 D_fake: 0.878 +(epoch: 509, iters: 942, time: 0.063) G_GAN: 0.535 G_GAN_Feat: 1.165 G_ID: 0.124 G_Rec: 0.517 D_GP: 0.080 D_real: 0.495 D_fake: 0.479 +(epoch: 509, iters: 1342, time: 0.064) G_GAN: 0.272 G_GAN_Feat: 0.874 G_ID: 0.092 G_Rec: 0.337 D_GP: 0.056 D_real: 0.586 D_fake: 0.755 +(epoch: 509, iters: 1742, time: 0.063) G_GAN: 0.374 G_GAN_Feat: 1.080 G_ID: 0.122 G_Rec: 0.452 D_GP: 0.054 D_real: 0.792 D_fake: 0.633 +(epoch: 509, iters: 2142, time: 0.063) G_GAN: -0.009 G_GAN_Feat: 0.881 G_ID: 0.124 G_Rec: 0.360 D_GP: 0.048 D_real: 0.659 D_fake: 1.010 +(epoch: 509, iters: 2542, time: 0.063) G_GAN: 0.440 G_GAN_Feat: 0.965 G_ID: 0.133 G_Rec: 0.410 D_GP: 0.043 D_real: 0.843 D_fake: 0.564 +(epoch: 509, iters: 2942, time: 0.064) G_GAN: 0.054 G_GAN_Feat: 0.763 G_ID: 0.117 G_Rec: 0.333 D_GP: 0.037 D_real: 0.991 D_fake: 0.946 +(epoch: 509, iters: 3342, time: 0.063) G_GAN: 0.180 G_GAN_Feat: 1.005 G_ID: 0.109 G_Rec: 0.493 D_GP: 0.033 D_real: 0.894 D_fake: 0.820 +(epoch: 509, iters: 3742, time: 0.063) G_GAN: 0.043 G_GAN_Feat: 0.609 G_ID: 0.088 G_Rec: 0.262 D_GP: 0.027 D_real: 1.061 D_fake: 0.957 +(epoch: 509, iters: 4142, time: 0.063) G_GAN: 0.455 G_GAN_Feat: 0.919 G_ID: 0.111 G_Rec: 0.424 D_GP: 0.035 D_real: 1.048 D_fake: 0.582 +(epoch: 509, iters: 4542, time: 0.064) G_GAN: -0.082 G_GAN_Feat: 0.739 G_ID: 0.117 G_Rec: 0.303 D_GP: 0.045 D_real: 0.670 D_fake: 1.082 +(epoch: 509, iters: 4942, time: 0.063) G_GAN: 0.525 G_GAN_Feat: 0.927 G_ID: 0.108 G_Rec: 0.388 D_GP: 0.055 D_real: 1.084 D_fake: 0.486 +(epoch: 509, iters: 5342, time: 0.063) G_GAN: 0.185 G_GAN_Feat: 0.719 G_ID: 0.099 G_Rec: 0.302 D_GP: 0.036 D_real: 1.106 D_fake: 0.815 +(epoch: 509, iters: 5742, time: 0.063) G_GAN: 0.427 G_GAN_Feat: 0.861 G_ID: 0.118 G_Rec: 0.388 D_GP: 0.043 D_real: 1.092 D_fake: 0.588 +(epoch: 509, iters: 6142, time: 0.064) G_GAN: 0.306 G_GAN_Feat: 0.789 G_ID: 0.093 G_Rec: 0.333 D_GP: 0.037 D_real: 0.968 D_fake: 0.696 +(epoch: 509, iters: 6542, time: 0.063) G_GAN: 0.535 G_GAN_Feat: 0.920 G_ID: 0.106 G_Rec: 0.382 D_GP: 0.037 D_real: 1.199 D_fake: 0.488 +(epoch: 509, iters: 6942, time: 0.063) G_GAN: -0.160 G_GAN_Feat: 0.746 G_ID: 0.106 G_Rec: 0.339 D_GP: 0.032 D_real: 0.792 D_fake: 1.160 +(epoch: 509, iters: 7342, time: 0.063) G_GAN: 0.237 G_GAN_Feat: 0.917 G_ID: 0.106 G_Rec: 0.392 D_GP: 0.031 D_real: 0.897 D_fake: 0.766 +(epoch: 509, iters: 7742, time: 0.064) G_GAN: 0.058 G_GAN_Feat: 0.802 G_ID: 0.085 G_Rec: 0.331 D_GP: 0.039 D_real: 0.813 D_fake: 0.942 +(epoch: 509, iters: 8142, time: 0.063) G_GAN: 0.292 G_GAN_Feat: 0.990 G_ID: 0.110 G_Rec: 0.408 D_GP: 0.036 D_real: 0.917 D_fake: 0.709 +(epoch: 509, iters: 8542, time: 0.063) G_GAN: 0.275 G_GAN_Feat: 0.859 G_ID: 0.087 G_Rec: 0.311 D_GP: 0.086 D_real: 0.740 D_fake: 0.728 +(epoch: 510, iters: 334, time: 0.063) G_GAN: 0.535 G_GAN_Feat: 1.059 G_ID: 0.116 G_Rec: 0.440 D_GP: 0.118 D_real: 0.552 D_fake: 0.487 +(epoch: 510, iters: 734, time: 0.064) G_GAN: 0.377 G_GAN_Feat: 0.979 G_ID: 0.102 G_Rec: 0.325 D_GP: 0.042 D_real: 0.845 D_fake: 0.626 +(epoch: 510, iters: 1134, time: 0.063) G_GAN: 0.563 G_GAN_Feat: 1.039 G_ID: 0.108 G_Rec: 0.443 D_GP: 0.057 D_real: 1.132 D_fake: 0.529 +(epoch: 510, iters: 1534, time: 0.063) G_GAN: 0.246 G_GAN_Feat: 0.763 G_ID: 0.133 G_Rec: 0.298 D_GP: 0.048 D_real: 1.056 D_fake: 0.755 +(epoch: 510, iters: 1934, time: 0.063) G_GAN: 0.268 G_GAN_Feat: 1.093 G_ID: 0.115 G_Rec: 0.430 D_GP: 0.469 D_real: 0.389 D_fake: 0.737 +(epoch: 510, iters: 2334, time: 0.064) G_GAN: 0.660 G_GAN_Feat: 0.713 G_ID: 0.080 G_Rec: 0.283 D_GP: 0.055 D_real: 1.518 D_fake: 0.474 +(epoch: 510, iters: 2734, time: 0.063) G_GAN: 0.428 G_GAN_Feat: 1.022 G_ID: 0.124 G_Rec: 0.455 D_GP: 0.060 D_real: 0.942 D_fake: 0.578 +(epoch: 510, iters: 3134, time: 0.063) G_GAN: 0.060 G_GAN_Feat: 0.752 G_ID: 0.096 G_Rec: 0.318 D_GP: 0.053 D_real: 0.801 D_fake: 0.940 +(epoch: 510, iters: 3534, time: 0.063) G_GAN: 0.319 G_GAN_Feat: 0.899 G_ID: 0.108 G_Rec: 0.408 D_GP: 0.040 D_real: 1.024 D_fake: 0.688 +(epoch: 510, iters: 3934, time: 0.064) G_GAN: -0.123 G_GAN_Feat: 0.822 G_ID: 0.092 G_Rec: 0.331 D_GP: 0.077 D_real: 0.883 D_fake: 1.123 +(epoch: 510, iters: 4334, time: 0.063) G_GAN: 0.568 G_GAN_Feat: 1.035 G_ID: 0.107 G_Rec: 0.481 D_GP: 0.051 D_real: 1.038 D_fake: 0.443 +(epoch: 510, iters: 4734, time: 0.063) G_GAN: 0.424 G_GAN_Feat: 0.824 G_ID: 0.132 G_Rec: 0.335 D_GP: 0.071 D_real: 0.961 D_fake: 0.616 +(epoch: 510, iters: 5134, time: 0.063) G_GAN: 0.454 G_GAN_Feat: 1.032 G_ID: 0.113 G_Rec: 0.473 D_GP: 0.046 D_real: 1.147 D_fake: 0.562 +(epoch: 510, iters: 5534, time: 0.064) G_GAN: 0.265 G_GAN_Feat: 0.789 G_ID: 0.095 G_Rec: 0.290 D_GP: 0.048 D_real: 0.893 D_fake: 0.735 +(epoch: 510, iters: 5934, time: 0.063) G_GAN: 0.409 G_GAN_Feat: 0.970 G_ID: 0.113 G_Rec: 0.431 D_GP: 0.052 D_real: 0.892 D_fake: 0.596 +(epoch: 510, iters: 6334, time: 0.063) G_GAN: 0.296 G_GAN_Feat: 0.881 G_ID: 0.100 G_Rec: 0.342 D_GP: 0.043 D_real: 0.756 D_fake: 0.705 +(epoch: 510, iters: 6734, time: 0.063) G_GAN: 0.581 G_GAN_Feat: 1.077 G_ID: 0.109 G_Rec: 0.466 D_GP: 0.036 D_real: 0.832 D_fake: 0.478 +(epoch: 510, iters: 7134, time: 0.064) G_GAN: 0.296 G_GAN_Feat: 0.856 G_ID: 0.091 G_Rec: 0.311 D_GP: 0.037 D_real: 1.121 D_fake: 0.707 +(epoch: 510, iters: 7534, time: 0.063) G_GAN: 0.575 G_GAN_Feat: 1.111 G_ID: 0.131 G_Rec: 0.456 D_GP: 0.047 D_real: 0.686 D_fake: 0.433 +(epoch: 510, iters: 7934, time: 0.063) G_GAN: -0.049 G_GAN_Feat: 0.859 G_ID: 0.092 G_Rec: 0.328 D_GP: 0.102 D_real: 0.526 D_fake: 1.049 +(epoch: 510, iters: 8334, time: 0.063) G_GAN: 0.311 G_GAN_Feat: 1.026 G_ID: 0.139 G_Rec: 0.476 D_GP: 0.033 D_real: 0.868 D_fake: 0.689 +(epoch: 511, iters: 126, time: 0.064) G_GAN: 0.433 G_GAN_Feat: 0.768 G_ID: 0.087 G_Rec: 0.293 D_GP: 0.032 D_real: 1.156 D_fake: 0.569 +(epoch: 511, iters: 526, time: 0.063) G_GAN: 0.920 G_GAN_Feat: 1.157 G_ID: 0.109 G_Rec: 0.474 D_GP: 0.135 D_real: 0.765 D_fake: 0.183 +(epoch: 511, iters: 926, time: 0.063) G_GAN: 0.307 G_GAN_Feat: 0.825 G_ID: 0.081 G_Rec: 0.327 D_GP: 0.032 D_real: 1.176 D_fake: 0.697 +(epoch: 511, iters: 1326, time: 0.063) G_GAN: 0.636 G_GAN_Feat: 1.069 G_ID: 0.096 G_Rec: 0.474 D_GP: 0.041 D_real: 0.835 D_fake: 0.399 +(epoch: 511, iters: 1726, time: 0.064) G_GAN: 0.007 G_GAN_Feat: 0.858 G_ID: 0.116 G_Rec: 0.318 D_GP: 0.105 D_real: 0.379 D_fake: 0.993 +(epoch: 511, iters: 2126, time: 0.063) G_GAN: 0.835 G_GAN_Feat: 1.212 G_ID: 0.094 G_Rec: 0.474 D_GP: 0.368 D_real: 0.539 D_fake: 0.241 +(epoch: 511, iters: 2526, time: 0.063) G_GAN: 0.326 G_GAN_Feat: 0.900 G_ID: 0.096 G_Rec: 0.332 D_GP: 0.058 D_real: 1.151 D_fake: 0.723 +(epoch: 511, iters: 2926, time: 0.063) G_GAN: 0.419 G_GAN_Feat: 0.994 G_ID: 0.105 G_Rec: 0.435 D_GP: 0.035 D_real: 1.095 D_fake: 0.586 +(epoch: 511, iters: 3326, time: 0.064) G_GAN: 0.386 G_GAN_Feat: 0.716 G_ID: 0.095 G_Rec: 0.317 D_GP: 0.031 D_real: 1.255 D_fake: 0.631 +(epoch: 511, iters: 3726, time: 0.063) G_GAN: 0.256 G_GAN_Feat: 0.916 G_ID: 0.119 G_Rec: 0.451 D_GP: 0.031 D_real: 0.934 D_fake: 0.750 +(epoch: 511, iters: 4126, time: 0.063) G_GAN: 0.079 G_GAN_Feat: 0.778 G_ID: 0.091 G_Rec: 0.335 D_GP: 0.053 D_real: 0.905 D_fake: 0.922 +(epoch: 511, iters: 4526, time: 0.063) G_GAN: 0.268 G_GAN_Feat: 0.891 G_ID: 0.122 G_Rec: 0.396 D_GP: 0.037 D_real: 0.814 D_fake: 0.735 +(epoch: 511, iters: 4926, time: 0.064) G_GAN: 0.383 G_GAN_Feat: 0.708 G_ID: 0.089 G_Rec: 0.283 D_GP: 0.033 D_real: 1.299 D_fake: 0.619 +(epoch: 511, iters: 5326, time: 0.063) G_GAN: 0.567 G_GAN_Feat: 0.962 G_ID: 0.108 G_Rec: 0.446 D_GP: 0.042 D_real: 1.231 D_fake: 0.446 +(epoch: 511, iters: 5726, time: 0.063) G_GAN: 0.540 G_GAN_Feat: 0.845 G_ID: 0.087 G_Rec: 0.311 D_GP: 0.054 D_real: 1.103 D_fake: 0.471 +(epoch: 511, iters: 6126, time: 0.063) G_GAN: 0.910 G_GAN_Feat: 1.030 G_ID: 0.105 G_Rec: 0.460 D_GP: 0.040 D_real: 1.410 D_fake: 0.219 +(epoch: 511, iters: 6526, time: 0.064) G_GAN: 0.340 G_GAN_Feat: 0.992 G_ID: 0.129 G_Rec: 0.368 D_GP: 0.041 D_real: 1.246 D_fake: 0.721 +(epoch: 511, iters: 6926, time: 0.063) G_GAN: 0.425 G_GAN_Feat: 1.031 G_ID: 0.113 G_Rec: 0.482 D_GP: 0.038 D_real: 1.062 D_fake: 0.578 +(epoch: 511, iters: 7326, time: 0.063) G_GAN: 0.376 G_GAN_Feat: 0.962 G_ID: 0.113 G_Rec: 0.342 D_GP: 0.079 D_real: 0.381 D_fake: 0.639 +(epoch: 511, iters: 7726, time: 0.063) G_GAN: 0.692 G_GAN_Feat: 1.029 G_ID: 0.112 G_Rec: 0.431 D_GP: 0.048 D_real: 1.165 D_fake: 0.464 +(epoch: 511, iters: 8126, time: 0.064) G_GAN: -0.064 G_GAN_Feat: 0.736 G_ID: 0.101 G_Rec: 0.323 D_GP: 0.037 D_real: 0.752 D_fake: 1.064 +(epoch: 511, iters: 8526, time: 0.063) G_GAN: 0.653 G_GAN_Feat: 1.027 G_ID: 0.119 G_Rec: 0.451 D_GP: 0.081 D_real: 0.903 D_fake: 0.383 +(epoch: 512, iters: 318, time: 0.063) G_GAN: 0.213 G_GAN_Feat: 0.815 G_ID: 0.104 G_Rec: 0.321 D_GP: 0.077 D_real: 0.640 D_fake: 0.791 +(epoch: 512, iters: 718, time: 0.063) G_GAN: 0.289 G_GAN_Feat: 1.149 G_ID: 0.115 G_Rec: 0.487 D_GP: 0.250 D_real: 0.380 D_fake: 0.723 +(epoch: 512, iters: 1118, time: 0.064) G_GAN: 0.586 G_GAN_Feat: 0.848 G_ID: 0.114 G_Rec: 0.323 D_GP: 0.051 D_real: 1.523 D_fake: 0.500 +(epoch: 512, iters: 1518, time: 0.063) G_GAN: 0.237 G_GAN_Feat: 0.987 G_ID: 0.125 G_Rec: 0.465 D_GP: 0.031 D_real: 0.878 D_fake: 0.763 +(epoch: 512, iters: 1918, time: 0.063) G_GAN: 0.068 G_GAN_Feat: 0.728 G_ID: 0.107 G_Rec: 0.316 D_GP: 0.048 D_real: 0.876 D_fake: 0.932 +(epoch: 512, iters: 2318, time: 0.063) G_GAN: 0.479 G_GAN_Feat: 1.017 G_ID: 0.112 G_Rec: 0.459 D_GP: 0.049 D_real: 0.926 D_fake: 0.536 +(epoch: 512, iters: 2718, time: 0.064) G_GAN: 0.450 G_GAN_Feat: 0.670 G_ID: 0.089 G_Rec: 0.315 D_GP: 0.028 D_real: 1.411 D_fake: 0.553 +(epoch: 512, iters: 3118, time: 0.063) G_GAN: 0.404 G_GAN_Feat: 0.966 G_ID: 0.100 G_Rec: 0.480 D_GP: 0.031 D_real: 0.973 D_fake: 0.606 +(epoch: 512, iters: 3518, time: 0.063) G_GAN: 0.053 G_GAN_Feat: 0.719 G_ID: 0.129 G_Rec: 0.320 D_GP: 0.035 D_real: 0.904 D_fake: 0.947 +(epoch: 512, iters: 3918, time: 0.063) G_GAN: 0.363 G_GAN_Feat: 1.017 G_ID: 0.118 G_Rec: 0.467 D_GP: 0.050 D_real: 0.792 D_fake: 0.647 +(epoch: 512, iters: 4318, time: 0.064) G_GAN: 0.137 G_GAN_Feat: 0.728 G_ID: 0.092 G_Rec: 0.310 D_GP: 0.041 D_real: 0.919 D_fake: 0.863 +(epoch: 512, iters: 4718, time: 0.063) G_GAN: 0.239 G_GAN_Feat: 1.038 G_ID: 0.108 G_Rec: 0.457 D_GP: 0.047 D_real: 0.610 D_fake: 0.762 +(epoch: 512, iters: 5118, time: 0.063) G_GAN: 0.277 G_GAN_Feat: 0.831 G_ID: 0.101 G_Rec: 0.324 D_GP: 0.044 D_real: 0.863 D_fake: 0.723 +(epoch: 512, iters: 5518, time: 0.063) G_GAN: 0.503 G_GAN_Feat: 1.035 G_ID: 0.106 G_Rec: 0.450 D_GP: 0.047 D_real: 0.729 D_fake: 0.507 +(epoch: 512, iters: 5918, time: 0.064) G_GAN: 0.357 G_GAN_Feat: 0.807 G_ID: 0.085 G_Rec: 0.332 D_GP: 0.031 D_real: 1.457 D_fake: 0.646 +(epoch: 512, iters: 6318, time: 0.063) G_GAN: 0.538 G_GAN_Feat: 0.932 G_ID: 0.112 G_Rec: 0.425 D_GP: 0.038 D_real: 1.234 D_fake: 0.474 +(epoch: 512, iters: 6718, time: 0.063) G_GAN: 0.216 G_GAN_Feat: 0.687 G_ID: 0.093 G_Rec: 0.279 D_GP: 0.030 D_real: 1.152 D_fake: 0.784 +(epoch: 512, iters: 7118, time: 0.063) G_GAN: 0.322 G_GAN_Feat: 1.023 G_ID: 0.117 G_Rec: 0.420 D_GP: 0.036 D_real: 0.695 D_fake: 0.679 +(epoch: 512, iters: 7518, time: 0.063) G_GAN: 0.417 G_GAN_Feat: 0.753 G_ID: 0.096 G_Rec: 0.317 D_GP: 0.045 D_real: 1.223 D_fake: 0.589 +(epoch: 512, iters: 7918, time: 0.063) G_GAN: 0.315 G_GAN_Feat: 0.985 G_ID: 0.111 G_Rec: 0.424 D_GP: 0.056 D_real: 0.878 D_fake: 0.686 +(epoch: 512, iters: 8318, time: 0.063) G_GAN: 0.206 G_GAN_Feat: 0.862 G_ID: 0.099 G_Rec: 0.320 D_GP: 0.072 D_real: 0.566 D_fake: 0.795 +(epoch: 513, iters: 110, time: 0.064) G_GAN: 0.312 G_GAN_Feat: 0.854 G_ID: 0.120 G_Rec: 0.376 D_GP: 0.036 D_real: 1.096 D_fake: 0.691 +(epoch: 513, iters: 510, time: 0.063) G_GAN: 0.084 G_GAN_Feat: 0.671 G_ID: 0.108 G_Rec: 0.297 D_GP: 0.033 D_real: 0.933 D_fake: 0.916 +(epoch: 513, iters: 910, time: 0.063) G_GAN: 0.185 G_GAN_Feat: 0.945 G_ID: 0.109 G_Rec: 0.418 D_GP: 0.037 D_real: 0.801 D_fake: 0.817 +(epoch: 513, iters: 1310, time: 0.063) G_GAN: 0.006 G_GAN_Feat: 0.772 G_ID: 0.095 G_Rec: 0.292 D_GP: 0.065 D_real: 0.779 D_fake: 0.995 +(epoch: 513, iters: 1710, time: 0.064) G_GAN: 0.188 G_GAN_Feat: 1.056 G_ID: 0.112 G_Rec: 0.461 D_GP: 0.040 D_real: 0.625 D_fake: 0.813 +(epoch: 513, iters: 2110, time: 0.063) G_GAN: 0.435 G_GAN_Feat: 0.723 G_ID: 0.096 G_Rec: 0.318 D_GP: 0.030 D_real: 1.341 D_fake: 0.565 +(epoch: 513, iters: 2510, time: 0.063) G_GAN: 0.829 G_GAN_Feat: 0.984 G_ID: 0.109 G_Rec: 0.408 D_GP: 0.044 D_real: 1.353 D_fake: 0.254 +(epoch: 513, iters: 2910, time: 0.063) G_GAN: 0.498 G_GAN_Feat: 0.728 G_ID: 0.098 G_Rec: 0.309 D_GP: 0.029 D_real: 1.454 D_fake: 0.513 +(epoch: 513, iters: 3310, time: 0.064) G_GAN: 0.051 G_GAN_Feat: 0.915 G_ID: 0.125 G_Rec: 0.412 D_GP: 0.030 D_real: 0.816 D_fake: 0.949 +(epoch: 513, iters: 3710, time: 0.063) G_GAN: 0.071 G_GAN_Feat: 0.663 G_ID: 0.093 G_Rec: 0.293 D_GP: 0.028 D_real: 0.992 D_fake: 0.931 +(epoch: 513, iters: 4110, time: 0.063) G_GAN: 0.520 G_GAN_Feat: 0.853 G_ID: 0.110 G_Rec: 0.395 D_GP: 0.030 D_real: 1.205 D_fake: 0.503 +(epoch: 513, iters: 4510, time: 0.063) G_GAN: 0.234 G_GAN_Feat: 0.748 G_ID: 0.093 G_Rec: 0.310 D_GP: 0.031 D_real: 1.145 D_fake: 0.767 +(epoch: 513, iters: 4910, time: 0.064) G_GAN: 0.083 G_GAN_Feat: 0.950 G_ID: 0.135 G_Rec: 0.411 D_GP: 0.048 D_real: 0.689 D_fake: 0.917 +(epoch: 513, iters: 5310, time: 0.063) G_GAN: 0.154 G_GAN_Feat: 0.734 G_ID: 0.091 G_Rec: 0.291 D_GP: 0.031 D_real: 0.981 D_fake: 0.846 +(epoch: 513, iters: 5710, time: 0.063) G_GAN: 0.067 G_GAN_Feat: 1.060 G_ID: 0.117 G_Rec: 0.467 D_GP: 0.037 D_real: 0.798 D_fake: 0.933 +(epoch: 513, iters: 6110, time: 0.063) G_GAN: 0.331 G_GAN_Feat: 0.942 G_ID: 0.079 G_Rec: 0.314 D_GP: 0.049 D_real: 0.739 D_fake: 0.670 +(epoch: 513, iters: 6510, time: 0.064) G_GAN: 0.106 G_GAN_Feat: 1.089 G_ID: 0.105 G_Rec: 0.452 D_GP: 0.070 D_real: 0.484 D_fake: 0.899 +(epoch: 513, iters: 6910, time: 0.063) G_GAN: 0.070 G_GAN_Feat: 0.783 G_ID: 0.083 G_Rec: 0.307 D_GP: 0.037 D_real: 0.883 D_fake: 0.930 +(epoch: 513, iters: 7310, time: 0.063) G_GAN: 0.732 G_GAN_Feat: 1.016 G_ID: 0.106 G_Rec: 0.442 D_GP: 0.033 D_real: 1.248 D_fake: 0.303 +(epoch: 513, iters: 7710, time: 0.063) G_GAN: 0.360 G_GAN_Feat: 0.865 G_ID: 0.086 G_Rec: 0.299 D_GP: 0.039 D_real: 0.920 D_fake: 0.641 +(epoch: 513, iters: 8110, time: 0.064) G_GAN: 0.406 G_GAN_Feat: 0.908 G_ID: 0.099 G_Rec: 0.403 D_GP: 0.027 D_real: 1.179 D_fake: 0.600 +(epoch: 513, iters: 8510, time: 0.063) G_GAN: 0.236 G_GAN_Feat: 0.691 G_ID: 0.090 G_Rec: 0.307 D_GP: 0.031 D_real: 1.161 D_fake: 0.768 +(epoch: 514, iters: 302, time: 0.063) G_GAN: 0.212 G_GAN_Feat: 1.016 G_ID: 0.123 G_Rec: 0.456 D_GP: 0.042 D_real: 0.677 D_fake: 0.788 +(epoch: 514, iters: 702, time: 0.063) G_GAN: -0.032 G_GAN_Feat: 0.778 G_ID: 0.106 G_Rec: 0.309 D_GP: 0.049 D_real: 0.684 D_fake: 1.032 +(epoch: 514, iters: 1102, time: 0.064) G_GAN: 0.322 G_GAN_Feat: 0.960 G_ID: 0.111 G_Rec: 0.421 D_GP: 0.055 D_real: 0.924 D_fake: 0.681 +(epoch: 514, iters: 1502, time: 0.063) G_GAN: 0.113 G_GAN_Feat: 0.793 G_ID: 0.099 G_Rec: 0.330 D_GP: 0.054 D_real: 0.896 D_fake: 0.887 +(epoch: 514, iters: 1902, time: 0.063) G_GAN: 0.371 G_GAN_Feat: 0.982 G_ID: 0.116 G_Rec: 0.448 D_GP: 0.050 D_real: 0.918 D_fake: 0.638 +(epoch: 514, iters: 2302, time: 0.063) G_GAN: -0.108 G_GAN_Feat: 0.734 G_ID: 0.103 G_Rec: 0.299 D_GP: 0.044 D_real: 0.650 D_fake: 1.108 +(epoch: 514, iters: 2702, time: 0.064) G_GAN: 0.442 G_GAN_Feat: 1.023 G_ID: 0.114 G_Rec: 0.450 D_GP: 0.048 D_real: 0.895 D_fake: 0.562 +(epoch: 514, iters: 3102, time: 0.063) G_GAN: 0.083 G_GAN_Feat: 0.919 G_ID: 0.119 G_Rec: 0.324 D_GP: 0.255 D_real: 0.203 D_fake: 0.917 +(epoch: 514, iters: 3502, time: 0.063) G_GAN: 0.659 G_GAN_Feat: 1.133 G_ID: 0.138 G_Rec: 0.501 D_GP: 0.117 D_real: 0.868 D_fake: 0.398 +(epoch: 514, iters: 3902, time: 0.063) G_GAN: 0.400 G_GAN_Feat: 0.819 G_ID: 0.097 G_Rec: 0.304 D_GP: 0.057 D_real: 0.742 D_fake: 0.631 +(epoch: 514, iters: 4302, time: 0.064) G_GAN: 0.454 G_GAN_Feat: 1.027 G_ID: 0.125 G_Rec: 0.436 D_GP: 0.039 D_real: 0.861 D_fake: 0.550 +(epoch: 514, iters: 4702, time: 0.063) G_GAN: 0.356 G_GAN_Feat: 0.921 G_ID: 0.117 G_Rec: 0.313 D_GP: 0.253 D_real: 0.452 D_fake: 0.647 +(epoch: 514, iters: 5102, time: 0.063) G_GAN: 0.310 G_GAN_Feat: 1.100 G_ID: 0.120 G_Rec: 0.433 D_GP: 0.039 D_real: 0.575 D_fake: 0.691 +(epoch: 514, iters: 5502, time: 0.063) G_GAN: 0.225 G_GAN_Feat: 0.931 G_ID: 0.104 G_Rec: 0.319 D_GP: 0.070 D_real: 0.403 D_fake: 0.796 +(epoch: 514, iters: 5902, time: 0.064) G_GAN: 0.364 G_GAN_Feat: 0.854 G_ID: 0.098 G_Rec: 0.456 D_GP: 0.027 D_real: 1.116 D_fake: 0.638 +(epoch: 514, iters: 6302, time: 0.063) G_GAN: 0.126 G_GAN_Feat: 0.589 G_ID: 0.104 G_Rec: 0.277 D_GP: 0.029 D_real: 0.998 D_fake: 0.874 +(epoch: 514, iters: 6702, time: 0.063) G_GAN: 0.275 G_GAN_Feat: 0.863 G_ID: 0.121 G_Rec: 0.433 D_GP: 0.041 D_real: 0.932 D_fake: 0.725 +(epoch: 514, iters: 7102, time: 0.063) G_GAN: 0.137 G_GAN_Feat: 0.575 G_ID: 0.091 G_Rec: 0.285 D_GP: 0.029 D_real: 1.071 D_fake: 0.863 +(epoch: 514, iters: 7502, time: 0.064) G_GAN: 0.124 G_GAN_Feat: 0.812 G_ID: 0.110 G_Rec: 0.403 D_GP: 0.033 D_real: 0.924 D_fake: 0.876 +(epoch: 514, iters: 7902, time: 0.063) G_GAN: -0.118 G_GAN_Feat: 0.670 G_ID: 0.098 G_Rec: 0.342 D_GP: 0.037 D_real: 0.823 D_fake: 1.118 +(epoch: 514, iters: 8302, time: 0.063) G_GAN: 0.202 G_GAN_Feat: 0.835 G_ID: 0.123 G_Rec: 0.411 D_GP: 0.038 D_real: 0.882 D_fake: 0.803 +(epoch: 514, iters: 8702, time: 0.063) G_GAN: 0.071 G_GAN_Feat: 0.649 G_ID: 0.088 G_Rec: 0.341 D_GP: 0.032 D_real: 0.941 D_fake: 0.930 +(epoch: 515, iters: 494, time: 0.064) G_GAN: 0.028 G_GAN_Feat: 0.901 G_ID: 0.116 G_Rec: 0.435 D_GP: 0.066 D_real: 0.686 D_fake: 0.972 +(epoch: 515, iters: 894, time: 0.063) G_GAN: 0.085 G_GAN_Feat: 0.591 G_ID: 0.082 G_Rec: 0.287 D_GP: 0.028 D_real: 1.019 D_fake: 0.915 +(epoch: 515, iters: 1294, time: 0.063) G_GAN: 0.096 G_GAN_Feat: 0.902 G_ID: 0.102 G_Rec: 0.444 D_GP: 0.042 D_real: 0.708 D_fake: 0.909 +(epoch: 515, iters: 1694, time: 0.063) G_GAN: -0.418 G_GAN_Feat: 0.742 G_ID: 0.112 G_Rec: 0.362 D_GP: 0.062 D_real: 0.463 D_fake: 1.418 +(epoch: 515, iters: 2094, time: 0.064) G_GAN: 0.276 G_GAN_Feat: 0.802 G_ID: 0.114 G_Rec: 0.390 D_GP: 0.035 D_real: 0.890 D_fake: 0.735 +(epoch: 515, iters: 2494, time: 0.063) G_GAN: 0.012 G_GAN_Feat: 0.657 G_ID: 0.089 G_Rec: 0.298 D_GP: 0.039 D_real: 0.872 D_fake: 0.988 +(epoch: 515, iters: 2894, time: 0.063) G_GAN: 0.002 G_GAN_Feat: 0.922 G_ID: 0.104 G_Rec: 0.444 D_GP: 0.037 D_real: 0.588 D_fake: 0.999 +(epoch: 515, iters: 3294, time: 0.063) G_GAN: -0.062 G_GAN_Feat: 0.730 G_ID: 0.106 G_Rec: 0.353 D_GP: 0.057 D_real: 0.669 D_fake: 1.064 +(epoch: 515, iters: 3694, time: 0.064) G_GAN: 0.198 G_GAN_Feat: 0.898 G_ID: 0.123 G_Rec: 0.430 D_GP: 0.039 D_real: 0.903 D_fake: 0.805 +(epoch: 515, iters: 4094, time: 0.063) G_GAN: -0.121 G_GAN_Feat: 0.683 G_ID: 0.084 G_Rec: 0.302 D_GP: 0.047 D_real: 0.725 D_fake: 1.121 +(epoch: 515, iters: 4494, time: 0.063) G_GAN: 0.244 G_GAN_Feat: 0.858 G_ID: 0.118 G_Rec: 0.414 D_GP: 0.034 D_real: 0.884 D_fake: 0.757 +(epoch: 515, iters: 4894, time: 0.063) G_GAN: 0.029 G_GAN_Feat: 0.759 G_ID: 0.099 G_Rec: 0.317 D_GP: 0.066 D_real: 0.742 D_fake: 0.971 +(epoch: 515, iters: 5294, time: 0.064) G_GAN: 0.350 G_GAN_Feat: 0.912 G_ID: 0.107 G_Rec: 0.411 D_GP: 0.042 D_real: 0.922 D_fake: 0.657 +(epoch: 515, iters: 5694, time: 0.063) G_GAN: 0.333 G_GAN_Feat: 0.636 G_ID: 0.092 G_Rec: 0.279 D_GP: 0.050 D_real: 1.208 D_fake: 0.668 +(epoch: 515, iters: 6094, time: 0.063) G_GAN: 0.287 G_GAN_Feat: 0.916 G_ID: 0.112 G_Rec: 0.407 D_GP: 0.064 D_real: 0.843 D_fake: 0.715 +(epoch: 515, iters: 6494, time: 0.063) G_GAN: 0.338 G_GAN_Feat: 0.671 G_ID: 0.103 G_Rec: 0.282 D_GP: 0.036 D_real: 1.256 D_fake: 0.665 +(epoch: 515, iters: 6894, time: 0.064) G_GAN: 0.406 G_GAN_Feat: 0.909 G_ID: 0.117 G_Rec: 0.419 D_GP: 0.042 D_real: 1.068 D_fake: 0.595 +(epoch: 515, iters: 7294, time: 0.063) G_GAN: 0.125 G_GAN_Feat: 0.742 G_ID: 0.093 G_Rec: 0.405 D_GP: 0.064 D_real: 0.872 D_fake: 0.875 +(epoch: 515, iters: 7694, time: 0.063) G_GAN: 0.302 G_GAN_Feat: 0.981 G_ID: 0.136 G_Rec: 0.440 D_GP: 0.047 D_real: 0.739 D_fake: 0.702 +(epoch: 515, iters: 8094, time: 0.063) G_GAN: 0.118 G_GAN_Feat: 0.744 G_ID: 0.110 G_Rec: 0.313 D_GP: 0.075 D_real: 0.759 D_fake: 0.884 +(epoch: 515, iters: 8494, time: 0.064) G_GAN: 0.362 G_GAN_Feat: 0.991 G_ID: 0.124 G_Rec: 0.434 D_GP: 0.085 D_real: 0.613 D_fake: 0.642 +(epoch: 516, iters: 286, time: 0.063) G_GAN: 0.356 G_GAN_Feat: 0.693 G_ID: 0.092 G_Rec: 0.281 D_GP: 0.040 D_real: 1.304 D_fake: 0.645 +(epoch: 516, iters: 686, time: 0.063) G_GAN: 0.720 G_GAN_Feat: 0.910 G_ID: 0.109 G_Rec: 0.382 D_GP: 0.038 D_real: 1.330 D_fake: 0.337 +(epoch: 516, iters: 1086, time: 0.063) G_GAN: 0.129 G_GAN_Feat: 0.752 G_ID: 0.088 G_Rec: 0.290 D_GP: 0.067 D_real: 0.815 D_fake: 0.872 +(epoch: 516, iters: 1486, time: 0.064) G_GAN: 0.611 G_GAN_Feat: 0.915 G_ID: 0.105 G_Rec: 0.413 D_GP: 0.037 D_real: 1.208 D_fake: 0.457 +(epoch: 516, iters: 1886, time: 0.063) G_GAN: 0.100 G_GAN_Feat: 0.781 G_ID: 0.113 G_Rec: 0.315 D_GP: 0.045 D_real: 0.765 D_fake: 0.900 +(epoch: 516, iters: 2286, time: 0.063) G_GAN: 0.301 G_GAN_Feat: 0.842 G_ID: 0.122 G_Rec: 0.403 D_GP: 0.025 D_real: 1.161 D_fake: 0.707 +(epoch: 516, iters: 2686, time: 0.063) G_GAN: -0.025 G_GAN_Feat: 0.611 G_ID: 0.092 G_Rec: 0.298 D_GP: 0.025 D_real: 0.957 D_fake: 1.025 +(epoch: 516, iters: 3086, time: 0.064) G_GAN: 0.221 G_GAN_Feat: 0.900 G_ID: 0.123 G_Rec: 0.437 D_GP: 0.037 D_real: 0.809 D_fake: 0.782 +(epoch: 516, iters: 3486, time: 0.063) G_GAN: -0.204 G_GAN_Feat: 0.693 G_ID: 0.085 G_Rec: 0.321 D_GP: 0.041 D_real: 0.647 D_fake: 1.204 +(epoch: 516, iters: 3886, time: 0.063) G_GAN: 0.284 G_GAN_Feat: 0.892 G_ID: 0.105 G_Rec: 0.419 D_GP: 0.036 D_real: 1.018 D_fake: 0.721 +(epoch: 516, iters: 4286, time: 0.063) G_GAN: -0.084 G_GAN_Feat: 0.729 G_ID: 0.108 G_Rec: 0.345 D_GP: 0.050 D_real: 0.687 D_fake: 1.084 +(epoch: 516, iters: 4686, time: 0.064) G_GAN: 0.397 G_GAN_Feat: 0.870 G_ID: 0.108 G_Rec: 0.410 D_GP: 0.034 D_real: 1.061 D_fake: 0.621 +(epoch: 516, iters: 5086, time: 0.063) G_GAN: -0.043 G_GAN_Feat: 0.757 G_ID: 0.096 G_Rec: 0.326 D_GP: 0.038 D_real: 0.823 D_fake: 1.043 +(epoch: 516, iters: 5486, time: 0.063) G_GAN: 0.451 G_GAN_Feat: 0.986 G_ID: 0.113 G_Rec: 0.460 D_GP: 0.062 D_real: 0.789 D_fake: 0.562 +(epoch: 516, iters: 5886, time: 0.063) G_GAN: 0.139 G_GAN_Feat: 0.687 G_ID: 0.085 G_Rec: 0.300 D_GP: 0.034 D_real: 1.086 D_fake: 0.862 +(epoch: 516, iters: 6286, time: 0.064) G_GAN: 0.263 G_GAN_Feat: 0.959 G_ID: 0.108 G_Rec: 0.436 D_GP: 0.048 D_real: 0.801 D_fake: 0.737 +(epoch: 516, iters: 6686, time: 0.063) G_GAN: -0.054 G_GAN_Feat: 0.849 G_ID: 0.089 G_Rec: 0.370 D_GP: 0.276 D_real: 0.285 D_fake: 1.054 +(epoch: 516, iters: 7086, time: 0.063) G_GAN: 0.435 G_GAN_Feat: 0.927 G_ID: 0.118 G_Rec: 0.390 D_GP: 0.036 D_real: 1.121 D_fake: 0.565 +(epoch: 516, iters: 7486, time: 0.063) G_GAN: 0.119 G_GAN_Feat: 0.837 G_ID: 0.089 G_Rec: 0.332 D_GP: 0.354 D_real: 0.484 D_fake: 0.881 +(epoch: 516, iters: 7886, time: 0.064) G_GAN: 0.580 G_GAN_Feat: 1.018 G_ID: 0.139 G_Rec: 0.456 D_GP: 0.071 D_real: 0.866 D_fake: 0.430 +(epoch: 516, iters: 8286, time: 0.063) G_GAN: 0.163 G_GAN_Feat: 0.959 G_ID: 0.103 G_Rec: 0.381 D_GP: 0.566 D_real: 0.369 D_fake: 0.841 +(epoch: 516, iters: 8686, time: 0.063) G_GAN: 0.593 G_GAN_Feat: 1.053 G_ID: 0.117 G_Rec: 0.512 D_GP: 0.052 D_real: 1.083 D_fake: 0.511 +(epoch: 517, iters: 478, time: 0.063) G_GAN: 0.446 G_GAN_Feat: 0.821 G_ID: 0.096 G_Rec: 0.316 D_GP: 0.060 D_real: 0.968 D_fake: 0.600 +(epoch: 517, iters: 878, time: 0.064) G_GAN: 0.660 G_GAN_Feat: 0.916 G_ID: 0.092 G_Rec: 0.450 D_GP: 0.029 D_real: 1.410 D_fake: 0.357 +(epoch: 517, iters: 1278, time: 0.063) G_GAN: 0.183 G_GAN_Feat: 0.680 G_ID: 0.078 G_Rec: 0.309 D_GP: 0.028 D_real: 1.094 D_fake: 0.817 +(epoch: 517, iters: 1678, time: 0.063) G_GAN: 0.359 G_GAN_Feat: 0.890 G_ID: 0.108 G_Rec: 0.400 D_GP: 0.032 D_real: 1.141 D_fake: 0.643 +(epoch: 517, iters: 2078, time: 0.063) G_GAN: 0.249 G_GAN_Feat: 0.651 G_ID: 0.090 G_Rec: 0.294 D_GP: 0.028 D_real: 1.132 D_fake: 0.751 +(epoch: 517, iters: 2478, time: 0.064) G_GAN: 0.346 G_GAN_Feat: 0.826 G_ID: 0.099 G_Rec: 0.387 D_GP: 0.033 D_real: 1.102 D_fake: 0.654 +(epoch: 517, iters: 2878, time: 0.063) G_GAN: 0.124 G_GAN_Feat: 0.695 G_ID: 0.086 G_Rec: 0.308 D_GP: 0.044 D_real: 0.914 D_fake: 0.878 +(epoch: 517, iters: 3278, time: 0.063) G_GAN: 0.420 G_GAN_Feat: 0.883 G_ID: 0.109 G_Rec: 0.403 D_GP: 0.039 D_real: 1.117 D_fake: 0.582 +(epoch: 517, iters: 3678, time: 0.063) G_GAN: 0.147 G_GAN_Feat: 0.649 G_ID: 0.087 G_Rec: 0.284 D_GP: 0.035 D_real: 1.010 D_fake: 0.853 +(epoch: 517, iters: 4078, time: 0.063) G_GAN: 0.373 G_GAN_Feat: 0.916 G_ID: 0.100 G_Rec: 0.446 D_GP: 0.034 D_real: 1.105 D_fake: 0.633 +(epoch: 517, iters: 4478, time: 0.063) G_GAN: 0.132 G_GAN_Feat: 0.702 G_ID: 0.091 G_Rec: 0.304 D_GP: 0.041 D_real: 0.948 D_fake: 0.870 +(epoch: 517, iters: 4878, time: 0.063) G_GAN: 0.534 G_GAN_Feat: 0.939 G_ID: 0.118 G_Rec: 0.418 D_GP: 0.050 D_real: 1.174 D_fake: 0.496 +(epoch: 517, iters: 5278, time: 0.064) G_GAN: 0.178 G_GAN_Feat: 0.763 G_ID: 0.085 G_Rec: 0.286 D_GP: 0.038 D_real: 0.952 D_fake: 0.823 +(epoch: 517, iters: 5678, time: 0.063) G_GAN: 0.335 G_GAN_Feat: 1.057 G_ID: 0.112 G_Rec: 0.454 D_GP: 0.064 D_real: 0.698 D_fake: 0.669 +(epoch: 517, iters: 6078, time: 0.063) G_GAN: 0.127 G_GAN_Feat: 0.824 G_ID: 0.105 G_Rec: 0.363 D_GP: 0.244 D_real: 0.693 D_fake: 0.885 +(epoch: 517, iters: 6478, time: 0.063) G_GAN: 0.260 G_GAN_Feat: 1.073 G_ID: 0.116 G_Rec: 0.489 D_GP: 0.157 D_real: 0.398 D_fake: 0.741 +(epoch: 517, iters: 6878, time: 0.064) G_GAN: 0.069 G_GAN_Feat: 0.689 G_ID: 0.088 G_Rec: 0.286 D_GP: 0.035 D_real: 0.903 D_fake: 0.931 +(epoch: 517, iters: 7278, time: 0.063) G_GAN: 0.569 G_GAN_Feat: 0.872 G_ID: 0.118 G_Rec: 0.417 D_GP: 0.032 D_real: 1.194 D_fake: 0.450 +(epoch: 517, iters: 7678, time: 0.063) G_GAN: 0.166 G_GAN_Feat: 0.793 G_ID: 0.095 G_Rec: 0.294 D_GP: 0.049 D_real: 0.779 D_fake: 0.834 +(epoch: 517, iters: 8078, time: 0.063) G_GAN: 0.362 G_GAN_Feat: 0.931 G_ID: 0.107 G_Rec: 0.444 D_GP: 0.029 D_real: 1.042 D_fake: 0.639 +(epoch: 517, iters: 8478, time: 0.064) G_GAN: 0.302 G_GAN_Feat: 0.798 G_ID: 0.104 G_Rec: 0.337 D_GP: 0.062 D_real: 0.988 D_fake: 0.699 +(epoch: 518, iters: 270, time: 0.063) G_GAN: 0.765 G_GAN_Feat: 0.913 G_ID: 0.120 G_Rec: 0.416 D_GP: 0.036 D_real: 1.386 D_fake: 0.312 +(epoch: 518, iters: 670, time: 0.063) G_GAN: 0.358 G_GAN_Feat: 0.852 G_ID: 0.129 G_Rec: 0.335 D_GP: 0.116 D_real: 0.780 D_fake: 0.715 +(epoch: 518, iters: 1070, time: 0.063) G_GAN: 0.431 G_GAN_Feat: 1.048 G_ID: 0.116 G_Rec: 0.422 D_GP: 0.041 D_real: 0.726 D_fake: 0.573 +(epoch: 518, iters: 1470, time: 0.064) G_GAN: 0.200 G_GAN_Feat: 0.829 G_ID: 0.106 G_Rec: 0.293 D_GP: 0.073 D_real: 0.574 D_fake: 0.802 +(epoch: 518, iters: 1870, time: 0.063) G_GAN: 0.074 G_GAN_Feat: 0.915 G_ID: 0.130 G_Rec: 0.419 D_GP: 0.032 D_real: 0.800 D_fake: 0.927 +(epoch: 518, iters: 2270, time: 0.063) G_GAN: 0.422 G_GAN_Feat: 0.835 G_ID: 0.086 G_Rec: 0.334 D_GP: 0.073 D_real: 0.676 D_fake: 0.597 +(epoch: 518, iters: 2670, time: 0.063) G_GAN: 0.484 G_GAN_Feat: 1.032 G_ID: 0.103 G_Rec: 0.447 D_GP: 0.035 D_real: 0.969 D_fake: 0.525 +(epoch: 518, iters: 3070, time: 0.064) G_GAN: 0.403 G_GAN_Feat: 0.802 G_ID: 0.093 G_Rec: 0.307 D_GP: 0.036 D_real: 1.041 D_fake: 0.603 +(epoch: 518, iters: 3470, time: 0.063) G_GAN: 0.611 G_GAN_Feat: 1.158 G_ID: 0.107 G_Rec: 0.455 D_GP: 0.122 D_real: 0.511 D_fake: 0.407 +(epoch: 518, iters: 3870, time: 0.063) G_GAN: 0.049 G_GAN_Feat: 0.874 G_ID: 0.098 G_Rec: 0.322 D_GP: 0.047 D_real: 0.553 D_fake: 0.951 +(epoch: 518, iters: 4270, time: 0.063) G_GAN: 0.559 G_GAN_Feat: 0.893 G_ID: 0.108 G_Rec: 0.415 D_GP: 0.037 D_real: 1.297 D_fake: 0.469 +(epoch: 518, iters: 4670, time: 0.064) G_GAN: -0.035 G_GAN_Feat: 0.726 G_ID: 0.106 G_Rec: 0.330 D_GP: 0.046 D_real: 0.846 D_fake: 1.035 +(epoch: 518, iters: 5070, time: 0.063) G_GAN: 0.163 G_GAN_Feat: 0.944 G_ID: 0.107 G_Rec: 0.457 D_GP: 0.048 D_real: 0.771 D_fake: 0.838 +(epoch: 518, iters: 5470, time: 0.063) G_GAN: -0.036 G_GAN_Feat: 0.723 G_ID: 0.091 G_Rec: 0.318 D_GP: 0.040 D_real: 0.904 D_fake: 1.036 +(epoch: 518, iters: 5870, time: 0.063) G_GAN: 0.400 G_GAN_Feat: 0.936 G_ID: 0.107 G_Rec: 0.413 D_GP: 0.040 D_real: 0.958 D_fake: 0.604 +(epoch: 518, iters: 6270, time: 0.064) G_GAN: 0.190 G_GAN_Feat: 0.654 G_ID: 0.095 G_Rec: 0.266 D_GP: 0.029 D_real: 1.066 D_fake: 0.810 +(epoch: 518, iters: 6670, time: 0.063) G_GAN: 0.356 G_GAN_Feat: 0.933 G_ID: 0.118 G_Rec: 0.431 D_GP: 0.032 D_real: 0.905 D_fake: 0.644 +(epoch: 518, iters: 7070, time: 0.063) G_GAN: 0.416 G_GAN_Feat: 0.782 G_ID: 0.095 G_Rec: 0.300 D_GP: 0.044 D_real: 1.109 D_fake: 0.588 +(epoch: 518, iters: 7470, time: 0.063) G_GAN: 0.812 G_GAN_Feat: 1.093 G_ID: 0.111 G_Rec: 0.467 D_GP: 0.101 D_real: 1.049 D_fake: 0.316 +(epoch: 518, iters: 7870, time: 0.064) G_GAN: 0.517 G_GAN_Feat: 0.924 G_ID: 0.097 G_Rec: 0.370 D_GP: 1.121 D_real: 0.810 D_fake: 0.511 +(epoch: 518, iters: 8270, time: 0.063) G_GAN: 0.311 G_GAN_Feat: 0.954 G_ID: 0.120 G_Rec: 0.464 D_GP: 0.029 D_real: 0.942 D_fake: 0.690 +(epoch: 518, iters: 8670, time: 0.063) G_GAN: 0.223 G_GAN_Feat: 0.665 G_ID: 0.097 G_Rec: 0.293 D_GP: 0.026 D_real: 1.128 D_fake: 0.778 +(epoch: 519, iters: 462, time: 0.063) G_GAN: 0.269 G_GAN_Feat: 0.942 G_ID: 0.100 G_Rec: 0.403 D_GP: 0.045 D_real: 0.932 D_fake: 0.731 +(epoch: 519, iters: 862, time: 0.064) G_GAN: 0.229 G_GAN_Feat: 0.719 G_ID: 0.096 G_Rec: 0.274 D_GP: 0.032 D_real: 1.080 D_fake: 0.772 +(epoch: 519, iters: 1262, time: 0.063) G_GAN: 0.521 G_GAN_Feat: 0.953 G_ID: 0.099 G_Rec: 0.454 D_GP: 0.046 D_real: 1.063 D_fake: 0.486 +(epoch: 519, iters: 1662, time: 0.063) G_GAN: 0.493 G_GAN_Feat: 0.717 G_ID: 0.094 G_Rec: 0.293 D_GP: 0.029 D_real: 1.397 D_fake: 0.512 +(epoch: 519, iters: 2062, time: 0.063) G_GAN: 0.554 G_GAN_Feat: 0.929 G_ID: 0.139 G_Rec: 0.394 D_GP: 0.035 D_real: 1.066 D_fake: 0.453 +(epoch: 519, iters: 2462, time: 0.064) G_GAN: 0.229 G_GAN_Feat: 0.803 G_ID: 0.090 G_Rec: 0.316 D_GP: 0.040 D_real: 0.930 D_fake: 0.773 +(epoch: 519, iters: 2862, time: 0.063) G_GAN: 0.438 G_GAN_Feat: 0.982 G_ID: 0.124 G_Rec: 0.421 D_GP: 0.033 D_real: 1.037 D_fake: 0.601 +(epoch: 519, iters: 3262, time: 0.063) G_GAN: 0.046 G_GAN_Feat: 0.880 G_ID: 0.100 G_Rec: 0.368 D_GP: 0.046 D_real: 0.724 D_fake: 0.955 +(epoch: 519, iters: 3662, time: 0.063) G_GAN: 0.197 G_GAN_Feat: 0.973 G_ID: 0.123 G_Rec: 0.445 D_GP: 0.060 D_real: 0.703 D_fake: 0.803 +(epoch: 519, iters: 4062, time: 0.064) G_GAN: 0.162 G_GAN_Feat: 0.694 G_ID: 0.101 G_Rec: 0.291 D_GP: 0.037 D_real: 1.088 D_fake: 0.838 +(epoch: 519, iters: 4462, time: 0.063) G_GAN: 0.743 G_GAN_Feat: 1.089 G_ID: 0.100 G_Rec: 0.470 D_GP: 0.155 D_real: 0.759 D_fake: 0.312 +(epoch: 519, iters: 4862, time: 0.063) G_GAN: 0.018 G_GAN_Feat: 0.770 G_ID: 0.113 G_Rec: 0.329 D_GP: 0.031 D_real: 0.829 D_fake: 0.982 +(epoch: 519, iters: 5262, time: 0.063) G_GAN: 0.506 G_GAN_Feat: 0.995 G_ID: 0.106 G_Rec: 0.445 D_GP: 0.030 D_real: 1.144 D_fake: 0.495 +(epoch: 519, iters: 5662, time: 0.064) G_GAN: 0.420 G_GAN_Feat: 0.729 G_ID: 0.080 G_Rec: 0.296 D_GP: 0.037 D_real: 1.238 D_fake: 0.586 +(epoch: 519, iters: 6062, time: 0.063) G_GAN: 0.781 G_GAN_Feat: 1.083 G_ID: 0.102 G_Rec: 0.471 D_GP: 0.042 D_real: 1.125 D_fake: 0.250 +(epoch: 519, iters: 6462, time: 0.063) G_GAN: 0.183 G_GAN_Feat: 0.864 G_ID: 0.108 G_Rec: 0.334 D_GP: 0.052 D_real: 0.838 D_fake: 0.827 +(epoch: 519, iters: 6862, time: 0.063) G_GAN: 0.782 G_GAN_Feat: 1.019 G_ID: 0.120 G_Rec: 0.433 D_GP: 0.047 D_real: 1.197 D_fake: 0.267 +(epoch: 519, iters: 7262, time: 0.064) G_GAN: 0.290 G_GAN_Feat: 0.803 G_ID: 0.088 G_Rec: 0.334 D_GP: 0.038 D_real: 1.170 D_fake: 0.717 +(epoch: 519, iters: 7662, time: 0.063) G_GAN: 0.183 G_GAN_Feat: 0.998 G_ID: 0.141 G_Rec: 0.473 D_GP: 0.044 D_real: 0.620 D_fake: 0.822 +(epoch: 519, iters: 8062, time: 0.063) G_GAN: 0.102 G_GAN_Feat: 0.680 G_ID: 0.098 G_Rec: 0.283 D_GP: 0.032 D_real: 1.063 D_fake: 0.898 +(epoch: 519, iters: 8462, time: 0.063) G_GAN: 0.655 G_GAN_Feat: 0.900 G_ID: 0.109 G_Rec: 0.429 D_GP: 0.041 D_real: 1.281 D_fake: 0.431 +(epoch: 520, iters: 254, time: 0.064) G_GAN: -0.002 G_GAN_Feat: 0.753 G_ID: 0.097 G_Rec: 0.321 D_GP: 0.034 D_real: 0.779 D_fake: 1.003 diff --git a/mlflow/registry/Swimswap/checkpoints/people/opt.txt b/mlflow/registry/Swimswap/checkpoints/people/opt.txt new file mode 100644 index 0000000..3cc4ed4 --- /dev/null +++ b/mlflow/registry/Swimswap/checkpoints/people/opt.txt @@ -0,0 +1,72 @@ +------------ Options ------------- +batchSize: 8 +beta1: 0.5 +checkpoints_dir: ./checkpoints +continue_train: False +data_type: 32 +dataroot: ./datasets/cityscapes/ +debug: False +display_freq: 99 +display_winsize: 512 +feat_num: 3 +fineSize: 512 +fp16: False +gan_mode: hinge +gpu_ids: [0] +image_size: 224 +input_nc: 3 +instance_feat: False +isTrain: True +label_feat: False +label_nc: 0 +lambda_GP: 10.0 +lambda_feat: 10.0 +lambda_id: 20.0 +lambda_rec: 10.0 +latent_size: 512 +loadSize: 1024 +load_features: False +load_pretrain: +local_rank: 0 +lr: 0.0002 +max_dataset_size: inf +model: pix2pixHD +nThreads: 2 +n_blocks_global: 6 +n_blocks_local: 3 +n_clusters: 10 +n_downsample_E: 4 +n_downsample_global: 3 +n_layers_D: 4 +n_local_enhancers: 1 +name: people +ndf: 64 +nef: 16 +netG: global +ngf: 64 +niter: 10000 +niter_decay: 10000 +niter_fix_global: 0 +no_flip: False +no_ganFeat_loss: False +no_html: False +no_instance: False +no_vgg_loss: False +norm: batch +norm_G: spectralspadesyncbatch3x3 +num_D: 2 +output_nc: 3 +phase: train +pool_size: 0 +print_freq: 100 +resize_or_crop: scale_width +save_epoch_freq: 10000 +save_latest_freq: 10000 +semantic_nc: 3 +serial_batches: False +tf_log: False +times_G: 1 +use_dropout: False +verbose: False +which_epoch: latest +-------------- End ---------------- diff --git a/train/mlflow/sf2f/options/__init__.py b/mlflow/registry/Swimswap/insightface_func/__init__.py similarity index 100% rename from train/mlflow/sf2f/options/__init__.py rename to mlflow/registry/Swimswap/insightface_func/__init__.py diff --git a/mlflow/registry/Swimswap/insightface_func/face_detect_crop_multi.py b/mlflow/registry/Swimswap/insightface_func/face_detect_crop_multi.py new file mode 100644 index 0000000..d75c36b --- /dev/null +++ b/mlflow/registry/Swimswap/insightface_func/face_detect_crop_multi.py @@ -0,0 +1,100 @@ +''' +Author: Naiyuan liu +Github: https://github.com/NNNNAI +Date: 2021-11-23 17:03:58 +LastEditors: Naiyuan liu +LastEditTime: 2021-11-24 16:45:41 +Description: +''' +from __future__ import division +import collections +import numpy as np +import glob +import os +import os.path as osp +import cv2 +from insightface.model_zoo import model_zoo +from insightface_func.utils import face_align_ffhqandnewarc as face_align + +__all__ = ['Face_detect_crop', 'Face'] + +Face = collections.namedtuple('Face', [ + 'bbox', 'kps', 'det_score', 'embedding', 'gender', 'age', + 'embedding_norm', 'normed_embedding', + 'landmark' +]) + +Face.__new__.__defaults__ = (None, ) * len(Face._fields) + + +class Face_detect_crop: + def __init__(self, name, root='~/.insightface_func/models'): + self.models = {} + root = os.path.expanduser(root) + onnx_files = glob.glob(osp.join(root, name, '*.onnx')) + onnx_files = sorted(onnx_files) + for onnx_file in onnx_files: + if onnx_file.find('_selfgen_')>0: + #print('ignore:', onnx_file) + continue + model = model_zoo.get_model(onnx_file) + if model.taskname not in self.models: + print('find model:', onnx_file, model.taskname) + self.models[model.taskname] = model + else: + print('duplicated model task type, ignore:', onnx_file, model.taskname) + del model + assert 'detection' in self.models + self.det_model = self.models['detection'] + + + def prepare(self, ctx_id, det_thresh=0.5, det_size=(640, 640), mode ='None'): + self.det_thresh = det_thresh + self.mode = mode + assert det_size is not None + print('set det-size:', det_size) + self.det_size = det_size + for taskname, model in self.models.items(): + if taskname=='detection': + model.prepare(ctx_id, input_size=det_size) + else: + model.prepare(ctx_id) + + def get(self, img, crop_size, max_num=0): + bboxes, kpss = self.det_model.detect(img, + threshold=self.det_thresh, + max_num=max_num, + metric='default') + if bboxes.shape[0] == 0: + return None + ret = [] + # for i in range(bboxes.shape[0]): + # bbox = bboxes[i, 0:4] + # det_score = bboxes[i, 4] + # kps = None + # if kpss is not None: + # kps = kpss[i] + # M, _ = face_align.estimate_norm(kps, crop_size, mode ='None') + # align_img = cv2.warpAffine(img, M, (crop_size, crop_size), borderValue=0.0) + align_img_list = [] + M_list = [] + for i in range(bboxes.shape[0]): + kps = None + if kpss is not None: + kps = kpss[i] + M, _ = face_align.estimate_norm(kps, crop_size, mode = self.mode) + align_img = cv2.warpAffine(img, M, (crop_size, crop_size), borderValue=0.0) + align_img_list.append(align_img) + M_list.append(M) + + # det_score = bboxes[..., 4] + + # best_index = np.argmax(det_score) + + # kps = None + # if kpss is not None: + # kps = kpss[best_index] + # M, _ = face_align.estimate_norm(kps, crop_size, mode ='None') + # align_img = cv2.warpAffine(img, M, (crop_size, crop_size), borderValue=0.0) + + return align_img_list, M_list diff --git a/mlflow/registry/Swimswap/insightface_func/face_detect_crop_single.py b/mlflow/registry/Swimswap/insightface_func/face_detect_crop_single.py new file mode 100644 index 0000000..f3f648f --- /dev/null +++ b/mlflow/registry/Swimswap/insightface_func/face_detect_crop_single.py @@ -0,0 +1,97 @@ +''' +Author: Naiyuan liu +Github: https://github.com/NNNNAI +Date: 2021-11-23 17:03:58 +LastEditors: Naiyuan liu +LastEditTime: 2021-11-24 16:46:04 +Description: +''' +from __future__ import division +import collections +import numpy as np +import glob +import os +import os.path as osp +import cv2 +from insightface.model_zoo import model_zoo +from insightface_func.utils import face_align_ffhqandnewarc as face_align + +__all__ = ['Face_detect_crop', 'Face'] + +Face = collections.namedtuple('Face', [ + 'bbox', 'kps', 'det_score', 'embedding', 'gender', 'age', + 'embedding_norm', 'normed_embedding', + 'landmark' +]) + +Face.__new__.__defaults__ = (None, ) * len(Face._fields) + + +class Face_detect_crop: + def __init__(self, name, root='~/.insightface_func/models'): + self.models = {} + root = os.path.expanduser(root) + onnx_files = glob.glob(osp.join(root, name, '*.onnx')) + onnx_files = sorted(onnx_files) + for onnx_file in onnx_files: + if onnx_file.find('_selfgen_')>0: + #print('ignore:', onnx_file) + continue + model = model_zoo.get_model(onnx_file) + if model.taskname not in self.models: + print('find model:', onnx_file, model.taskname) + self.models[model.taskname] = model + else: + print('duplicated model task type, ignore:', onnx_file, model.taskname) + del model + assert 'detection' in self.models + self.det_model = self.models['detection'] + + + def prepare(self, ctx_id, det_thresh=0.5, det_size=(640, 640), mode ='None'): + self.det_thresh = det_thresh + self.mode = mode + assert det_size is not None + print('set det-size:', det_size) + self.det_size = det_size + for taskname, model in self.models.items(): + if taskname=='detection': + model.prepare(ctx_id, input_size=det_size) + else: + model.prepare(ctx_id) + + def get(self, img, crop_size, max_num=0): + bboxes, kpss = self.det_model.detect(img, + threshold=self.det_thresh, + max_num=max_num, + metric='default') + if bboxes.shape[0] == 0: + return None + # ret = [] + # for i in range(bboxes.shape[0]): + # bbox = bboxes[i, 0:4] + # det_score = bboxes[i, 4] + # kps = None + # if kpss is not None: + # kps = kpss[i] + # M, _ = face_align.estimate_norm(kps, crop_size, mode ='None') + # align_img = cv2.warpAffine(img, M, (crop_size, crop_size), borderValue=0.0) + # for i in range(bboxes.shape[0]): + # kps = None + # if kpss is not None: + # kps = kpss[i] + # M, _ = face_align.estimate_norm(kps, crop_size, mode ='None') + # align_img = cv2.warpAffine(img, M, (crop_size, crop_size), borderValue=0.0) + + det_score = bboxes[..., 4] + + # select the face with the hightest detection score + best_index = np.argmax(det_score) + + kps = None + if kpss is not None: + kps = kpss[best_index] + M, _ = face_align.estimate_norm(kps, crop_size, mode = self.mode) + align_img = cv2.warpAffine(img, M, (crop_size, crop_size), borderValue=0.0) + + return [align_img], [M] diff --git a/mlflow/registry/Swimswap/insightface_func/utils/face_align_ffhqandnewarc.py b/mlflow/registry/Swimswap/insightface_func/utils/face_align_ffhqandnewarc.py new file mode 100644 index 0000000..2f5aeb7 --- /dev/null +++ b/mlflow/registry/Swimswap/insightface_func/utils/face_align_ffhqandnewarc.py @@ -0,0 +1,159 @@ +''' +Author: Naiyuan liu +Github: https://github.com/NNNNAI +Date: 2021-11-15 19:42:42 +LastEditors: Naiyuan liu +LastEditTime: 2021-11-15 20:01:47 +Description: +''' + +import cv2 +import numpy as np +from skimage import transform as trans + +src1 = np.array([[51.642, 50.115], [57.617, 49.990], [35.740, 69.007], + [51.157, 89.050], [57.025, 89.702]], + dtype=np.float32) +#<--left +src2 = np.array([[45.031, 50.118], [65.568, 50.872], [39.677, 68.111], + [45.177, 86.190], [64.246, 86.758]], + dtype=np.float32) + +#---frontal +src3 = np.array([[39.730, 51.138], [72.270, 51.138], [56.000, 68.493], + [42.463, 87.010], [69.537, 87.010]], + dtype=np.float32) + +#-->right +src4 = np.array([[46.845, 50.872], [67.382, 50.118], [72.737, 68.111], + [48.167, 86.758], [67.236, 86.190]], + dtype=np.float32) + +#-->right profile +src5 = np.array([[54.796, 49.990], [60.771, 50.115], [76.673, 69.007], + [55.388, 89.702], [61.257, 89.050]], + dtype=np.float32) + +src = np.array([src1, src2, src3, src4, src5]) +src_map = src + +ffhq_src = np.array([[192.98138, 239.94708], [318.90277, 240.1936], [256.63416, 314.01935], + [201.26117, 371.41043], [313.08905, 371.15118]]) +ffhq_src = np.expand_dims(ffhq_src, axis=0) + +# arcface_src = np.array( +# [[38.2946, 51.6963], [73.5318, 51.5014], [56.0252, 71.7366], +# [41.5493, 92.3655], [70.7299, 92.2041]], +# dtype=np.float32) + +# arcface_src = np.expand_dims(arcface_src, axis=0) + +# In[66]: + + +# lmk is prediction; src is template +def estimate_norm(lmk, image_size=112, mode='ffhq'): + assert lmk.shape == (5, 2) + tform = trans.SimilarityTransform() + lmk_tran = np.insert(lmk, 2, values=np.ones(5), axis=1) + min_M = [] + min_index = [] + min_error = float('inf') + if mode == 'ffhq': + # assert image_size == 112 + src = ffhq_src * image_size / 512 + else: + src = src_map * image_size / 112 + for i in np.arange(src.shape[0]): + tform.estimate(lmk, src[i]) + M = tform.params[0:2, :] + results = np.dot(M, lmk_tran.T) + results = results.T + error = np.sum(np.sqrt(np.sum((results - src[i])**2, axis=1))) + # print(error) + if error < min_error: + min_error = error + min_M = M + min_index = i + return min_M, min_index + + +def norm_crop(img, landmark, image_size=112, mode='ffhq'): + if mode == 'Both': + M_None, _ = estimate_norm(landmark, image_size, mode = 'newarc') + M_ffhq, _ = estimate_norm(landmark, image_size, mode='ffhq') + warped_None = cv2.warpAffine(img, M_None, (image_size, image_size), borderValue=0.0) + warped_ffhq = cv2.warpAffine(img, M_ffhq, (image_size, image_size), borderValue=0.0) + return warped_ffhq, warped_None + else: + M, pose_index = estimate_norm(landmark, image_size, mode) + warped = cv2.warpAffine(img, M, (image_size, image_size), borderValue=0.0) + return warped + +def square_crop(im, S): + if im.shape[0] > im.shape[1]: + height = S + width = int(float(im.shape[1]) / im.shape[0] * S) + scale = float(S) / im.shape[0] + else: + width = S + height = int(float(im.shape[0]) / im.shape[1] * S) + scale = float(S) / im.shape[1] + resized_im = cv2.resize(im, (width, height)) + det_im = np.zeros((S, S, 3), dtype=np.uint8) + det_im[:resized_im.shape[0], :resized_im.shape[1], :] = resized_im + return det_im, scale + + +def transform(data, center, output_size, scale, rotation): + scale_ratio = scale + rot = float(rotation) * np.pi / 180.0 + #translation = (output_size/2-center[0]*scale_ratio, output_size/2-center[1]*scale_ratio) + t1 = trans.SimilarityTransform(scale=scale_ratio) + cx = center[0] * scale_ratio + cy = center[1] * scale_ratio + t2 = trans.SimilarityTransform(translation=(-1 * cx, -1 * cy)) + t3 = trans.SimilarityTransform(rotation=rot) + t4 = trans.SimilarityTransform(translation=(output_size / 2, + output_size / 2)) + t = t1 + t2 + t3 + t4 + M = t.params[0:2] + cropped = cv2.warpAffine(data, + M, (output_size, output_size), + borderValue=0.0) + return cropped, M + + +def trans_points2d(pts, M): + new_pts = np.zeros(shape=pts.shape, dtype=np.float32) + for i in range(pts.shape[0]): + pt = pts[i] + new_pt = np.array([pt[0], pt[1], 1.], dtype=np.float32) + new_pt = np.dot(M, new_pt) + #print('new_pt', new_pt.shape, new_pt) + new_pts[i] = new_pt[0:2] + + return new_pts + + +def trans_points3d(pts, M): + scale = np.sqrt(M[0][0] * M[0][0] + M[0][1] * M[0][1]) + #print(scale) + new_pts = np.zeros(shape=pts.shape, dtype=np.float32) + for i in range(pts.shape[0]): + pt = pts[i] + new_pt = np.array([pt[0], pt[1], 1.], dtype=np.float32) + new_pt = np.dot(M, new_pt) + #print('new_pt', new_pt.shape, new_pt) + new_pts[i][0:2] = new_pt[0:2] + new_pts[i][2] = pts[i][2] * scale + + return new_pts + + +def trans_points(pts, M): + if pts.shape[1] == 2: + return trans_points2d(pts, M) + else: + return trans_points3d(pts, M) + diff --git a/mlflow/registry/Swimswap/model_registry.py b/mlflow/registry/Swimswap/model_registry.py new file mode 100644 index 0000000..aa9f9b5 --- /dev/null +++ b/mlflow/registry/Swimswap/model_registry.py @@ -0,0 +1,66 @@ +import cv2 +import torch +import fractions +import numpy as np +from PIL import Image +import torch.nn.functional as F +from torchvision import transforms +from models.models import create_model +from options.test_options import TestOptions +# from insightface_func.face_detect_crop_single import Face_detect_crop +# from util.gifswap import gif_swap +import os + +import mlflow + +os.environ["MLFLOW_S3_ENDPOINT_URL"] = "http://223.130.133.236:9000" +os.environ["MLFLOW_TRACKING_URI"] = "http://223.130.133.236:5001" +os.environ["AWS_ACCESS_KEY_ID"] = "minio" +os.environ["AWS_SECRET_ACCESS_KEY"] = "miniostorage" +mlflow.set_experiment("Swimswap") + + +def lcm(a, b): return abs(a * b) / fractions.gcd(a, b) if a and b else 0 + +transformer = transforms.Compose([ + transforms.ToTensor(), + #transforms.Normalize([0.485, 0.456, 0.406], [0.229, 0.224, 0.225]) + ]) + +transformer_Arcface = transforms.Compose([ + transforms.ToTensor(), + transforms.Normalize([0.485, 0.456, 0.406], [0.229, 0.224, 0.225]) + ]) + +# detransformer = transforms.Compose([ +# transforms.Normalize([0, 0, 0], [1/0.229, 1/0.224, 1/0.225]), +# transforms.Normalize([-0.485, -0.456, -0.406], [1, 1, 1]) +# ]) + +def main(): + opt = TestOptions().parse() + + start_epoch, epoch_iter = 1, 0 + crop_size = opt.crop_size + + torch.nn.Module.dump_patches = True + if crop_size == 512: + opt.which_epoch = 550000 + opt.name = '512' + mode = 'ffhq' + else: + mode = 'None' + model = create_model(opt) + model.eval() + + + with mlflow.start_run(): + mlflow.pytorch.log_model( + pytorch_model=model, + artifact_path = "swimswap_pytorch", + # signature= signature, + # input_example = input_sample, + # pip_requirements = "rec.txt" + ) +if __name__ == "__main__": + main() \ No newline at end of file diff --git a/mlflow/registry/Swimswap/models/__init__.py b/mlflow/registry/Swimswap/models/__init__.py new file mode 100644 index 0000000..289de91 --- /dev/null +++ b/mlflow/registry/Swimswap/models/__init__.py @@ -0,0 +1,4 @@ +from .arcface_models import ArcMarginModel +from .arcface_models import ResNet +from .arcface_models import IRBlock +from .arcface_models import SEBlock \ No newline at end of file diff --git a/mlflow/registry/Swimswap/models/arcface_models.py b/mlflow/registry/Swimswap/models/arcface_models.py new file mode 100644 index 0000000..39a6ac5 --- /dev/null +++ b/mlflow/registry/Swimswap/models/arcface_models.py @@ -0,0 +1,163 @@ +import math +import torch +import torch.nn.functional as F +from torch import nn +from torch.nn import Parameter +from .config import device, num_classes + + + +class SEBlock(nn.Module): + def __init__(self, channel, reduction=16): + super(SEBlock, self).__init__() + self.avg_pool = nn.AdaptiveAvgPool2d(1) + self.fc = nn.Sequential( + nn.Linear(channel, channel // reduction), + nn.PReLU(), + nn.Linear(channel // reduction, channel), + nn.Sigmoid() + ) + + def forward(self, x): + b, c, _, _ = x.size() + y = self.avg_pool(x).view(b, c) + y = self.fc(y).view(b, c, 1, 1) + return x * y + + +class IRBlock(nn.Module): + expansion = 1 + + def __init__(self, inplanes, planes, stride=1, downsample=None, use_se=True): + super(IRBlock, self).__init__() + self.bn0 = nn.BatchNorm2d(inplanes) + self.conv1 = conv3x3(inplanes, inplanes) + self.bn1 = nn.BatchNorm2d(inplanes) + self.prelu = nn.PReLU() + self.conv2 = conv3x3(inplanes, planes, stride) + self.bn2 = nn.BatchNorm2d(planes) + self.downsample = downsample + self.stride = stride + self.use_se = use_se + if self.use_se: + self.se = SEBlock(planes) + + def forward(self, x): + residual = x + out = self.bn0(x) + out = self.conv1(out) + out = self.bn1(out) + out = self.prelu(out) + + out = self.conv2(out) + out = self.bn2(out) + if self.use_se: + out = self.se(out) + + if self.downsample is not None: + residual = self.downsample(x) + + out += residual + out = self.prelu(out) + + return out + + +class ResNet(nn.Module): + + def __init__(self, block, layers, use_se=True): + self.inplanes = 64 + self.use_se = use_se + super(ResNet, self).__init__() + self.conv1 = nn.Conv2d(3, 64, kernel_size=3, stride=1, bias=False) + self.bn1 = nn.BatchNorm2d(64) + self.prelu = nn.PReLU() + self.maxpool = nn.MaxPool2d(kernel_size=2, stride=2) + self.layer1 = self._make_layer(block, 64, layers[0]) + self.layer2 = self._make_layer(block, 128, layers[1], stride=2) + self.layer3 = self._make_layer(block, 256, layers[2], stride=2) + self.layer4 = self._make_layer(block, 512, layers[3], stride=2) + self.bn2 = nn.BatchNorm2d(512) + self.dropout = nn.Dropout() + self.fc = nn.Linear(512 * 7 * 7, 512) + self.bn3 = nn.BatchNorm1d(512) + + for m in self.modules(): + if isinstance(m, nn.Conv2d): + nn.init.xavier_normal_(m.weight) + elif isinstance(m, nn.BatchNorm2d) or isinstance(m, nn.BatchNorm1d): + nn.init.constant_(m.weight, 1) + nn.init.constant_(m.bias, 0) + elif isinstance(m, nn.Linear): + nn.init.xavier_normal_(m.weight) + nn.init.constant_(m.bias, 0) + + def _make_layer(self, block, planes, blocks, stride=1): + downsample = None + if stride != 1 or self.inplanes != planes * block.expansion: + downsample = nn.Sequential( + nn.Conv2d(self.inplanes, planes * block.expansion, + kernel_size=1, stride=stride, bias=False), + nn.BatchNorm2d(planes * block.expansion), + ) + + layers = [] + layers.append(block(self.inplanes, planes, stride, downsample, use_se=self.use_se)) + self.inplanes = planes + for i in range(1, blocks): + layers.append(block(self.inplanes, planes, use_se=self.use_se)) + + return nn.Sequential(*layers) + + def forward(self, x): + x = self.conv1(x) + x = self.bn1(x) + x = self.prelu(x) + x = self.maxpool(x) + + x = self.layer1(x) + x = self.layer2(x) + x = self.layer3(x) + x = self.layer4(x) + + x = self.bn2(x) + x = self.dropout(x) + # feature = x + x = x.view(x.size(0), -1) + x = self.fc(x) + x = self.bn3(x) + + return x + + +class ArcMarginModel(nn.Module): + def __init__(self, args): + super(ArcMarginModel, self).__init__() + + self.weight = Parameter(torch.FloatTensor(num_classes, args.emb_size)) + nn.init.xavier_uniform_(self.weight) + + self.easy_margin = args.easy_margin + self.m = args.margin_m + self.s = args.margin_s + + self.cos_m = math.cos(self.m) + self.sin_m = math.sin(self.m) + self.th = math.cos(math.pi - self.m) + self.mm = math.sin(math.pi - self.m) * self.m + + def forward(self, input, label): + x = F.normalize(input) + W = F.normalize(self.weight) + cosine = F.linear(x, W) + sine = torch.sqrt(1.0 - torch.pow(cosine, 2)) + phi = cosine * self.cos_m - sine * self.sin_m # cos(theta + m) + if self.easy_margin: + phi = torch.where(cosine > 0, phi, cosine) + else: + phi = torch.where(cosine > self.th, phi, cosine - self.mm) + one_hot = torch.zeros(cosine.size(), device=device) + one_hot.scatter_(1, label.view(-1, 1).long(), 1) + output = (one_hot * phi) + ((1.0 - one_hot) * cosine) + output *= self.s + return output \ No newline at end of file diff --git a/mlflow/registry/Swimswap/models/base_model.py b/mlflow/registry/Swimswap/models/base_model.py new file mode 100644 index 0000000..0a6474a --- /dev/null +++ b/mlflow/registry/Swimswap/models/base_model.py @@ -0,0 +1,138 @@ +import os +import torch +import sys + +class BaseModel(torch.nn.Module): + def name(self): + return 'BaseModel' + + def initialize(self, opt): + self.opt = opt + self.gpu_ids = opt.gpu_ids + self.isTrain = opt.isTrain + self.Tensor = torch.cuda.FloatTensor if self.gpu_ids else torch.Tensor + self.save_dir = os.path.join(opt.checkpoints_dir, opt.name) + + def set_input(self, input): + self.input = input + + def forward(self): + pass + + # used in test time, no backprop + def test(self): + pass + + def get_image_paths(self): + pass + + def optimize_parameters(self): + pass + + def get_current_visuals(self): + return self.input + + def get_current_errors(self): + return {} + + def save(self, label): + pass + + # helper saving function that can be used by subclasses + def save_network(self, network, network_label, epoch_label, gpu_ids=None): + save_filename = '{}_net_{}.pth'.format(epoch_label, network_label) + save_path = os.path.join(self.save_dir, save_filename) + torch.save(network.cpu().state_dict(), save_path) + if torch.cuda.is_available(): + network.cuda() + + def save_optim(self, network, network_label, epoch_label, gpu_ids=None): + save_filename = '{}_optim_{}.pth'.format(epoch_label, network_label) + save_path = os.path.join(self.save_dir, save_filename) + torch.save(network.state_dict(), save_path) + + + # helper loading function that can be used by subclasses + def load_network(self, network, network_label, epoch_label, save_dir=''): + save_filename = '%s_net_%s.pth' % (epoch_label, network_label) + if not save_dir: + save_dir = self.save_dir + save_path = os.path.join(save_dir, save_filename) + if not os.path.isfile(save_path): + print('%s not exists yet!' % save_path) + if network_label == 'G': + raise('Generator must exist!') + else: + #network.load_state_dict(torch.load(save_path)) + try: + network.load_state_dict(torch.load(save_path)) + except: + pretrained_dict = torch.load(save_path) + model_dict = network.state_dict() + try: + pretrained_dict = {k: v for k, v in pretrained_dict.items() if k in model_dict} + network.load_state_dict(pretrained_dict) + if self.opt.verbose: + print('Pretrained network %s has excessive layers; Only loading layers that are used' % network_label) + except: + print('Pretrained network %s has fewer layers; The following are not initialized:' % network_label) + for k, v in pretrained_dict.items(): + if v.size() == model_dict[k].size(): + model_dict[k] = v + + if sys.version_info >= (3,0): + not_initialized = set() + else: + from sets import Set + not_initialized = Set() + + for k, v in model_dict.items(): + if k not in pretrained_dict or v.size() != pretrained_dict[k].size(): + not_initialized.add(k.split('.')[0]) + + print(sorted(not_initialized)) + network.load_state_dict(model_dict) + + # helper loading function that can be used by subclasses + def load_optim(self, network, network_label, epoch_label, save_dir=''): + save_filename = '%s_optim_%s.pth' % (epoch_label, network_label) + if not save_dir: + save_dir = self.save_dir + save_path = os.path.join(save_dir, save_filename) + if not os.path.isfile(save_path): + print('%s not exists yet!' % save_path) + if network_label == 'G': + raise('Generator must exist!') + else: + #network.load_state_dict(torch.load(save_path)) + try: + network.load_state_dict(torch.load(save_path, map_location=torch.device("cpu"))) + except: + pretrained_dict = torch.load(save_path, map_location=torch.device("cpu")) + model_dict = network.state_dict() + try: + pretrained_dict = {k: v for k, v in pretrained_dict.items() if k in model_dict} + network.load_state_dict(pretrained_dict) + if self.opt.verbose: + print('Pretrained network %s has excessive layers; Only loading layers that are used' % network_label) + except: + print('Pretrained network %s has fewer layers; The following are not initialized:' % network_label) + for k, v in pretrained_dict.items(): + if v.size() == model_dict[k].size(): + model_dict[k] = v + + if sys.version_info >= (3,0): + not_initialized = set() + else: + from sets import Set + not_initialized = Set() + + for k, v in model_dict.items(): + if k not in pretrained_dict or v.size() != pretrained_dict[k].size(): + not_initialized.add(k.split('.')[0]) + + print(sorted(not_initialized)) + network.load_state_dict(model_dict) + + def update_learning_rate(): + pass diff --git a/mlflow/registry/Swimswap/models/config.py b/mlflow/registry/Swimswap/models/config.py new file mode 100644 index 0000000..eb83edb --- /dev/null +++ b/mlflow/registry/Swimswap/models/config.py @@ -0,0 +1,28 @@ +import os + +import torch + +device = torch.device('cuda' if torch.cuda.is_available() else 'cpu') # sets device for model and PyTorch tensors + +# Model parameters +image_w = 112 +image_h = 112 +channel = 3 +emb_size = 512 + +# Training parameters +num_workers = 1 # for data-loading; right now, only 1 works with h5py +grad_clip = 5. # clip gradients at an absolute value of +print_freq = 100 # print training/validation stats every __ batches +checkpoint = None # path to checkpoint, None if none + +# Data parameters +num_classes = 93431 +num_samples = 5179510 +DATA_DIR = 'data' +# faces_ms1m_folder = 'data/faces_ms1m_112x112' +faces_ms1m_folder = 'data/ms1m-retinaface-t1' +path_imgidx = os.path.join(faces_ms1m_folder, 'train.idx') +path_imgrec = os.path.join(faces_ms1m_folder, 'train.rec') +IMG_DIR = 'data/images' +pickle_file = 'data/faces_ms1m_112x112.pickle' diff --git a/mlflow/registry/Swimswap/models/fs_model.py b/mlflow/registry/Swimswap/models/fs_model.py new file mode 100644 index 0000000..95ac877 --- /dev/null +++ b/mlflow/registry/Swimswap/models/fs_model.py @@ -0,0 +1,242 @@ +import numpy as np +import torch +import torch.nn as nn +import torch.nn.functional as F +import os +from torch.autograd import Variable +from .base_model import BaseModel +from . import networks + +class SpecificNorm(nn.Module): + def __init__(self, epsilon=1e-8): + """ + @notice: avoid in-place ops. + https://discuss.pytorch.org/t/encounter-the-runtimeerror-one-of-the-variables-needed-for-gradient-computation-has-been-modified-by-an-inplace-operation/836/3 + """ + super(SpecificNorm, self).__init__() + self.mean = np.array([0.485, 0.456, 0.406]) + self.mean = torch.from_numpy(self.mean).float().cuda() + self.mean = self.mean.view([1, 3, 1, 1]) + + self.std = np.array([0.229, 0.224, 0.225]) + self.std = torch.from_numpy(self.std).float().cuda() + self.std = self.std.view([1, 3, 1, 1]) + + def forward(self, x): + mean = self.mean.expand([1, 3, x.shape[2], x.shape[3]]) + std = self.std.expand([1, 3, x.shape[2], x.shape[3]]) + + x = (x - mean) / std + + return x + +class fsModel(BaseModel): + def name(self): + return 'fsModel' + + def init_loss_filter(self, use_gan_feat_loss, use_vgg_loss): + flags = (True, use_gan_feat_loss, use_vgg_loss, True, True, True, True, True) + + def loss_filter(g_gan, g_gan_feat, g_vgg, g_id, g_rec, g_mask, d_real, d_fake): + return [l for (l, f) in zip((g_gan, g_gan_feat, g_vgg, g_id, g_rec, g_mask, d_real, d_fake), flags) if f] + + return loss_filter + + def initialize(self, opt): + BaseModel.initialize(self, opt) + if opt.resize_or_crop != 'none' or not opt.isTrain: # when training at full res this causes OOM + torch.backends.cudnn.benchmark = True + self.isTrain = opt.isTrain + + device = torch.device("cuda:0") + + if opt.crop_size == 224: + from .fs_networks import Generator_Adain_Upsample, Discriminator + elif opt.crop_size == 512: + from .fs_networks_512 import Generator_Adain_Upsample, Discriminator + + # Generator network + self.netG = Generator_Adain_Upsample(input_nc=3, output_nc=3, latent_size=512, n_blocks=9, deep=False) + self.netG.to(device) + + # Id network + netArc_checkpoint = opt.Arc_path + netArc_checkpoint = torch.load(netArc_checkpoint, map_location=torch.device("cpu")) + self.netArc = netArc_checkpoint + self.netArc = self.netArc.to(device) + self.netArc.eval() + + if not self.isTrain: + pretrained_path = '' if not self.isTrain else opt.load_pretrain + self.load_network(self.netG, 'G', opt.which_epoch, pretrained_path) + return + + # Discriminator network + if opt.gan_mode == 'original': + use_sigmoid = True + else: + use_sigmoid = False + self.netD1 = Discriminator(input_nc=3, use_sigmoid=use_sigmoid) + self.netD2 = Discriminator(input_nc=3, use_sigmoid=use_sigmoid) + self.netD1.to(device) + self.netD2.to(device) + + # + self.spNorm =SpecificNorm() + self.downsample = nn.AvgPool2d(3, stride=2, padding=[1, 1], count_include_pad=False) + + # load networks + if opt.continue_train or opt.load_pretrain: + pretrained_path = '' if not self.isTrain else opt.load_pretrain + # print (pretrained_path) + self.load_network(self.netG, 'G', opt.which_epoch, pretrained_path) + self.load_network(self.netD1, 'D1', opt.which_epoch, pretrained_path) + self.load_network(self.netD2, 'D2', opt.which_epoch, pretrained_path) + + + + if self.isTrain: + # define loss functions + self.loss_filter = self.init_loss_filter(not opt.no_ganFeat_loss, not opt.no_vgg_loss) + + self.criterionGAN = networks.GANLoss(opt.gan_mode, tensor=self.Tensor, opt=self.opt) + self.criterionFeat = nn.L1Loss() + self.criterionRec = nn.L1Loss() + + # Names so we can breakout loss + self.loss_names = self.loss_filter('G_GAN', 'G_GAN_Feat', 'G_VGG', 'G_ID', 'G_Rec', 'D_GP', + 'D_real', 'D_fake') + + # initialize optimizers + + # optimizer G + params = list(self.netG.parameters()) + self.optimizer_G = torch.optim.Adam(params, lr=opt.lr, betas=(opt.beta1, 0.999)) + + # optimizer D + params = list(self.netD1.parameters()) + list(self.netD2.parameters()) + self.optimizer_D = torch.optim.Adam(params, lr=opt.lr, betas=(opt.beta1, 0.999)) + + def _gradinet_penalty_D(self, netD, img_att, img_fake): + # interpolate sample + bs = img_fake.shape[0] + alpha = torch.rand(bs, 1, 1, 1).expand_as(img_fake).cuda() + interpolated = Variable(alpha * img_att + (1 - alpha) * img_fake, requires_grad=True) + pred_interpolated = netD.forward(interpolated) + pred_interpolated = pred_interpolated[-1] + + # compute gradients + grad = torch.autograd.grad(outputs=pred_interpolated, + inputs=interpolated, + grad_outputs=torch.ones(pred_interpolated.size()).cuda(), + retain_graph=True, + create_graph=True, + only_inputs=True)[0] + + # penalize gradients + grad = grad.view(grad.size(0), -1) + grad_l2norm = torch.sqrt(torch.sum(grad ** 2, dim=1)) + loss_d_gp = torch.mean((grad_l2norm - 1) ** 2) + + return loss_d_gp + + def cosin_metric(self, x1, x2): + #return np.dot(x1, x2) / (np.linalg.norm(x1) * np.linalg.norm(x2)) + return torch.sum(x1 * x2, dim=1) / (torch.norm(x1, dim=1) * torch.norm(x2, dim=1)) + + def forward(self, img_id, img_att, latent_id, latent_att, for_G=False): + loss_D_fake, loss_D_real, loss_D_GP = 0, 0, 0 + loss_G_GAN, loss_G_GAN_Feat, loss_G_VGG, loss_G_ID, loss_G_Rec = 0,0,0,0,0 + + img_fake = self.netG.forward(img_att, latent_id) + if not self.isTrain: + return img_fake + img_fake_downsample = self.downsample(img_fake) + img_att_downsample = self.downsample(img_att) + + + + # D_Fake + fea1_fake = self.netD1.forward(img_fake.detach()) + fea2_fake = self.netD2.forward(img_fake_downsample.detach()) + pred_fake = [fea1_fake, fea2_fake] + loss_D_fake = self.criterionGAN(pred_fake, False, for_discriminator=True) + + + # D_Feal + fea1_real = self.netD1.forward(img_att) + fea2_real = self.netD2.forward(img_att_downsample) + pred_real = [fea1_real, fea2_real] + fea_real = [fea1_real, fea2_real] + loss_D_real = self.criterionGAN(pred_real, True, for_discriminator=True) + #print('=====================D_Real========================') + + # D_GP + + loss_D_GP = 0 + + # G_GAN + fea1_fake = self.netD1.forward(img_fake) + fea2_fake = self.netD2.forward(img_fake_downsample) + #pred_fake = [fea1_fake[-1], fea2_fake[-1]] + pred_fake = [fea1_fake, fea2_fake] + fea_fake = [fea1_fake, fea2_fake] + loss_G_GAN = self.criterionGAN(pred_fake, True, for_discriminator=False) + + # GAN feature matching loss + n_layers_D = 4 + num_D = 2 + if not self.opt.no_ganFeat_loss: + feat_weights = 4.0 / (n_layers_D + 1) + D_weights = 1.0 / num_D + for i in range(num_D): + for j in range(0, len(fea_fake[i]) - 1): + loss_G_GAN_Feat += D_weights * feat_weights * \ + self.criterionFeat(fea_fake[i][j], + fea_real[i][j].detach()) * self.opt.lambda_feat + + + #G_ID + img_fake_down = F.interpolate(img_fake, size=(112,112)) + img_fake_down = self.spNorm(img_fake_down) + latent_fake = self.netArc(img_fake_down) + loss_G_ID = (1 - self.cosin_metric(latent_fake, latent_id)) + #print('=====================G_ID========================') + #print(loss_G_ID) + + #G_Rec + loss_G_Rec = self.criterionRec(img_fake, img_att) * self.opt.lambda_rec + + # Only return the fake_B image if necessary to save BW + return [self.loss_filter(loss_G_GAN, loss_G_GAN_Feat, loss_G_VGG, loss_G_ID, loss_G_Rec, loss_D_GP, loss_D_real, loss_D_fake), + img_fake] + + + def save(self, which_epoch): + self.save_network(self.netG, 'G', which_epoch, self.gpu_ids) + self.save_network(self.netD1, 'D1', which_epoch, self.gpu_ids) + self.save_network(self.netD2, 'D2', which_epoch, self.gpu_ids) + '''if self.gen_features: + self.save_network(self.netE, 'E', which_epoch, self.gpu_ids)''' + + def update_fixed_params(self): + # after fixing the global generator for a number of iterations, also start finetuning it + params = list(self.netG.parameters()) + if self.gen_features: + params += list(self.netE.parameters()) + self.optimizer_G = torch.optim.Adam(params, lr=self.opt.lr, betas=(self.opt.beta1, 0.999)) + if self.opt.verbose: + print('------------ Now also finetuning global generator -----------') + + def update_learning_rate(self): + lrd = self.opt.lr / self.opt.niter_decay + lr = self.old_lr - lrd + for param_group in self.optimizer_D.param_groups: + param_group['lr'] = lr + for param_group in self.optimizer_G.param_groups: + param_group['lr'] = lr + if self.opt.verbose: + print('update learning rate: %f -> %f' % (self.old_lr, lr)) + self.old_lr = lr + + diff --git a/mlflow/registry/Swimswap/models/fs_networks.py b/mlflow/registry/Swimswap/models/fs_networks.py new file mode 100644 index 0000000..227f6fa --- /dev/null +++ b/mlflow/registry/Swimswap/models/fs_networks.py @@ -0,0 +1,215 @@ +""" +Copyright (C) 2019 NVIDIA Corporation. All rights reserved. +Licensed under the CC BY-NC-SA 4.0 license (https://creativecommons.org/licenses/by-nc-sa/4.0/legalcode). +""" + +import torch +import torch.nn as nn + + +class InstanceNorm(nn.Module): + def __init__(self, epsilon=1e-8): + """ + @notice: avoid in-place ops. + https://discuss.pytorch.org/t/encounter-the-runtimeerror-one-of-the-variables-needed-for-gradient-computation-has-been-modified-by-an-inplace-operation/836/3 + """ + super(InstanceNorm, self).__init__() + self.epsilon = epsilon + + def forward(self, x): + x = x - torch.mean(x, (2, 3), True) + tmp = torch.mul(x, x) # or x ** 2 + tmp = torch.rsqrt(torch.mean(tmp, (2, 3), True) + self.epsilon) + return x * tmp + +class ApplyStyle(nn.Module): + """ + @ref: https://github.com/lernapparat/lernapparat/blob/master/style_gan/pytorch_style_gan.ipynb + """ + def __init__(self, latent_size, channels): + super(ApplyStyle, self).__init__() + self.linear = nn.Linear(latent_size, channels * 2) + + def forward(self, x, latent): + style = self.linear(latent) # style => [batch_size, n_channels*2] + shape = [-1, 2, x.size(1), 1, 1] + style = style.view(shape) # [batch_size, 2, n_channels, ...] + #x = x * (style[:, 0] + 1.) + style[:, 1] + x = x * (style[:, 0] * 1 + 1.) + style[:, 1] * 1 + return x + +class ResnetBlock_Adain(nn.Module): + def __init__(self, dim, latent_size, padding_type, activation=nn.ReLU(True)): + super(ResnetBlock_Adain, self).__init__() + + p = 0 + conv1 = [] + if padding_type == 'reflect': + conv1 += [nn.ReflectionPad2d(1)] + elif padding_type == 'replicate': + conv1 += [nn.ReplicationPad2d(1)] + elif padding_type == 'zero': + p = 1 + else: + raise NotImplementedError('padding [%s] is not implemented' % padding_type) + conv1 += [nn.Conv2d(dim, dim, kernel_size=3, padding = p), InstanceNorm()] + self.conv1 = nn.Sequential(*conv1) + self.style1 = ApplyStyle(latent_size, dim) + self.act1 = activation + + p = 0 + conv2 = [] + if padding_type == 'reflect': + conv2 += [nn.ReflectionPad2d(1)] + elif padding_type == 'replicate': + conv2 += [nn.ReplicationPad2d(1)] + elif padding_type == 'zero': + p = 1 + else: + raise NotImplementedError('padding [%s] is not implemented' % padding_type) + conv2 += [nn.Conv2d(dim, dim, kernel_size=3, padding=p), InstanceNorm()] + self.conv2 = nn.Sequential(*conv2) + self.style2 = ApplyStyle(latent_size, dim) + + + def forward(self, x, dlatents_in_slice): + y = self.conv1(x) + y = self.style1(y, dlatents_in_slice) + y = self.act1(y) + y = self.conv2(y) + y = self.style2(y, dlatents_in_slice) + out = x + y + return out + + + +class Generator_Adain_Upsample(nn.Module): + def __init__(self, input_nc, output_nc, latent_size, n_blocks=6, deep=False, + norm_layer=nn.BatchNorm2d, + padding_type='reflect'): + assert (n_blocks >= 0) + super(Generator_Adain_Upsample, self).__init__() + activation = nn.ReLU(True) + self.deep = deep + + self.first_layer = nn.Sequential(nn.ReflectionPad2d(3), nn.Conv2d(input_nc, 64, kernel_size=7, padding=0), + norm_layer(64), activation) + ### downsample + self.down1 = nn.Sequential(nn.Conv2d(64, 128, kernel_size=3, stride=2, padding=1), + norm_layer(128), activation) + self.down2 = nn.Sequential(nn.Conv2d(128, 256, kernel_size=3, stride=2, padding=1), + norm_layer(256), activation) + self.down3 = nn.Sequential(nn.Conv2d(256, 512, kernel_size=3, stride=2, padding=1), + norm_layer(512), activation) + if self.deep: + self.down4 = nn.Sequential(nn.Conv2d(512, 512, kernel_size=3, stride=2, padding=1), + norm_layer(512), activation) + + ### resnet blocks + BN = [] + for i in range(n_blocks): + BN += [ + ResnetBlock_Adain(512, latent_size=latent_size, padding_type=padding_type, activation=activation)] + self.BottleNeck = nn.Sequential(*BN) + + if self.deep: + self.up4 = nn.Sequential( + nn.Upsample(scale_factor=2, mode='bilinear'), + nn.Conv2d(512, 512, kernel_size=3, stride=1, padding=1), + nn.BatchNorm2d(512), activation + ) + self.up3 = nn.Sequential( + nn.Upsample(scale_factor=2, mode='bilinear'), + nn.Conv2d(512, 256, kernel_size=3, stride=1, padding=1), + nn.BatchNorm2d(256), activation + ) + self.up2 = nn.Sequential( + nn.Upsample(scale_factor=2, mode='bilinear'), + nn.Conv2d(256, 128, kernel_size=3, stride=1, padding=1), + nn.BatchNorm2d(128), activation + ) + self.up1 = nn.Sequential( + nn.Upsample(scale_factor=2, mode='bilinear'), + nn.Conv2d(128, 64, kernel_size=3, stride=1, padding=1), + nn.BatchNorm2d(64), activation + ) + self.last_layer = nn.Sequential(nn.ReflectionPad2d(3), nn.Conv2d(64, output_nc, kernel_size=7, padding=0), + nn.Tanh()) + + def forward(self, input, dlatents): + x = input # 3*224*224 + + skip1 = self.first_layer(x) + skip2 = self.down1(skip1) + skip3 = self.down2(skip2) + if self.deep: + skip4 = self.down3(skip3) + x = self.down4(skip4) + else: + x = self.down3(skip3) + + for i in range(len(self.BottleNeck)): + x = self.BottleNeck[i](x, dlatents) + + if self.deep: + x = self.up4(x) + x = self.up3(x) + x = self.up2(x) + x = self.up1(x) + x = self.last_layer(x) + x = (x + 1) / 2 + + return x + +class Discriminator(nn.Module): + def __init__(self, input_nc, norm_layer=nn.BatchNorm2d, use_sigmoid=False): + super(Discriminator, self).__init__() + + kw = 4 + padw = 1 + self.down1 = nn.Sequential( + nn.Conv2d(input_nc, 64, kernel_size=kw, stride=2, padding=padw), nn.LeakyReLU(0.2, True) + ) + self.down2 = nn.Sequential( + nn.Conv2d(64, 128, kernel_size=kw, stride=2, padding=padw), + norm_layer(128), nn.LeakyReLU(0.2, True) + ) + self.down3 = nn.Sequential( + nn.Conv2d(128, 256, kernel_size=kw, stride=2, padding=padw), + norm_layer(256), nn.LeakyReLU(0.2, True) + ) + self.down4 = nn.Sequential( + nn.Conv2d(256, 512, kernel_size=kw, stride=2, padding=padw), + norm_layer(512), nn.LeakyReLU(0.2, True) + ) + self.conv1 = nn.Sequential( + nn.Conv2d(512, 512, kernel_size=kw, stride=1, padding=padw), + norm_layer(512), + nn.LeakyReLU(0.2, True) + ) + + if use_sigmoid: + self.conv2 = nn.Sequential( + nn.Conv2d(512, 1, kernel_size=kw, stride=1, padding=padw), nn.Sigmoid() + ) + else: + self.conv2 = nn.Sequential( + nn.Conv2d(512, 1, kernel_size=kw, stride=1, padding=padw) + ) + + def forward(self, input): + out = [] + x = self.down1(input) + out.append(x) + x = self.down2(x) + out.append(x) + x = self.down3(x) + out.append(x) + x = self.down4(x) + out.append(x) + x = self.conv1(x) + out.append(x) + x = self.conv2(x) + out.append(x) + + return out \ No newline at end of file diff --git a/mlflow/registry/Swimswap/models/fs_networks_512.py b/mlflow/registry/Swimswap/models/fs_networks_512.py new file mode 100644 index 0000000..1b9f9c8 --- /dev/null +++ b/mlflow/registry/Swimswap/models/fs_networks_512.py @@ -0,0 +1,232 @@ +''' +Author: Naiyuan liu +Github: https://github.com/NNNNAI +Date: 2021-11-23 16:55:48 +LastEditors: Naiyuan liu +LastEditTime: 2021-11-24 16:58:06 +Description: +''' +""" +Copyright (C) 2019 NVIDIA Corporation. All rights reserved. +Licensed under the CC BY-NC-SA 4.0 license (https://creativecommons.org/licenses/by-nc-sa/4.0/legalcode). +""" + +import torch +import torch.nn as nn + + +class InstanceNorm(nn.Module): + def __init__(self, epsilon=1e-8): + """ + @notice: avoid in-place ops. + https://discuss.pytorch.org/t/encounter-the-runtimeerror-one-of-the-variables-needed-for-gradient-computation-has-been-modified-by-an-inplace-operation/836/3 + """ + super(InstanceNorm, self).__init__() + self.epsilon = epsilon + + def forward(self, x): + x = x - torch.mean(x, (2, 3), True) + tmp = torch.mul(x, x) # or x ** 2 + tmp = torch.rsqrt(torch.mean(tmp, (2, 3), True) + self.epsilon) + return x * tmp + +class ApplyStyle(nn.Module): + """ + @ref: https://github.com/lernapparat/lernapparat/blob/master/style_gan/pytorch_style_gan.ipynb + """ + def __init__(self, latent_size, channels): + super(ApplyStyle, self).__init__() + self.linear = nn.Linear(latent_size, channels * 2) + + def forward(self, x, latent): + style = self.linear(latent) # style => [batch_size, n_channels*2] + shape = [-1, 2, x.size(1), 1, 1] + style = style.view(shape) # [batch_size, 2, n_channels, ...] + #x = x * (style[:, 0] + 1.) + style[:, 1] + x = x * (style[:, 0] * 1 + 1.) + style[:, 1] * 1 + return x + +class ResnetBlock_Adain(nn.Module): + def __init__(self, dim, latent_size, padding_type, activation=nn.ReLU(True)): + super(ResnetBlock_Adain, self).__init__() + + p = 0 + conv1 = [] + if padding_type == 'reflect': + conv1 += [nn.ReflectionPad2d(1)] + elif padding_type == 'replicate': + conv1 += [nn.ReplicationPad2d(1)] + elif padding_type == 'zero': + p = 1 + else: + raise NotImplementedError('padding [%s] is not implemented' % padding_type) + conv1 += [nn.Conv2d(dim, dim, kernel_size=3, padding = p), InstanceNorm()] + self.conv1 = nn.Sequential(*conv1) + self.style1 = ApplyStyle(latent_size, dim) + self.act1 = activation + + p = 0 + conv2 = [] + if padding_type == 'reflect': + conv2 += [nn.ReflectionPad2d(1)] + elif padding_type == 'replicate': + conv2 += [nn.ReplicationPad2d(1)] + elif padding_type == 'zero': + p = 1 + else: + raise NotImplementedError('padding [%s] is not implemented' % padding_type) + conv2 += [nn.Conv2d(dim, dim, kernel_size=3, padding=p), InstanceNorm()] + self.conv2 = nn.Sequential(*conv2) + self.style2 = ApplyStyle(latent_size, dim) + + + def forward(self, x, dlatents_in_slice): + y = self.conv1(x) + y = self.style1(y, dlatents_in_slice) + y = self.act1(y) + y = self.conv2(y) + y = self.style2(y, dlatents_in_slice) + out = x + y + return out + + + +class Generator_Adain_Upsample(nn.Module): + def __init__(self, input_nc, output_nc, latent_size, n_blocks=6, deep=False, + norm_layer=nn.BatchNorm2d, + padding_type='reflect'): + assert (n_blocks >= 0) + super(Generator_Adain_Upsample, self).__init__() + activation = nn.ReLU(True) + self.deep = deep + + self.first_layer = nn.Sequential(nn.ReflectionPad2d(3), nn.Conv2d(input_nc, 32, kernel_size=7, padding=0), + norm_layer(32), activation) + ### downsample + self.down0 = nn.Sequential(nn.Conv2d(32, 64, kernel_size=3, stride=2, padding=1), + norm_layer(64), activation) + self.down1 = nn.Sequential(nn.Conv2d(64, 128, kernel_size=3, stride=2, padding=1), + norm_layer(128), activation) + self.down2 = nn.Sequential(nn.Conv2d(128, 256, kernel_size=3, stride=2, padding=1), + norm_layer(256), activation) + self.down3 = nn.Sequential(nn.Conv2d(256, 512, kernel_size=3, stride=2, padding=1), + norm_layer(512), activation) + if self.deep: + self.down4 = nn.Sequential(nn.Conv2d(512, 512, kernel_size=3, stride=2, padding=1), + norm_layer(512), activation) + + ### resnet blocks + BN = [] + for i in range(n_blocks): + BN += [ + ResnetBlock_Adain(512, latent_size=latent_size, padding_type=padding_type, activation=activation)] + self.BottleNeck = nn.Sequential(*BN) + + if self.deep: + self.up4 = nn.Sequential( + nn.Upsample(scale_factor=2, mode='bilinear'), + nn.Conv2d(512, 512, kernel_size=3, stride=1, padding=1), + nn.BatchNorm2d(512), activation + ) + self.up3 = nn.Sequential( + nn.Upsample(scale_factor=2, mode='bilinear'), + nn.Conv2d(512, 256, kernel_size=3, stride=1, padding=1), + nn.BatchNorm2d(256), activation + ) + self.up2 = nn.Sequential( + nn.Upsample(scale_factor=2, mode='bilinear'), + nn.Conv2d(256, 128, kernel_size=3, stride=1, padding=1), + nn.BatchNorm2d(128), activation + ) + self.up1 = nn.Sequential( + nn.Upsample(scale_factor=2, mode='bilinear'), + nn.Conv2d(128, 64, kernel_size=3, stride=1, padding=1), + nn.BatchNorm2d(64), activation + ) + self.up0 = nn.Sequential( + nn.Upsample(scale_factor=2, mode='bilinear'), + nn.Conv2d(64, 32, kernel_size=3, stride=1, padding=1), + nn.BatchNorm2d(32), activation + ) + self.last_layer = nn.Sequential(nn.ReflectionPad2d(3), nn.Conv2d(32, output_nc, kernel_size=7, padding=0), + nn.Tanh()) + + def forward(self, input, dlatents): + x = input # 3*224*224 + + skip0 = self.first_layer(x) + skip1 = self.down0(skip0) + skip2 = self.down1(skip1) + skip3 = self.down2(skip2) + if self.deep: + skip4 = self.down3(skip3) + x = self.down4(skip4) + else: + x = self.down3(skip3) + + for i in range(len(self.BottleNeck)): + x = self.BottleNeck[i](x, dlatents) + + if self.deep: + x = self.up4(x) + x = self.up3(x) + x = self.up2(x) + x = self.up1(x) + x = self.up0(x) + x = self.last_layer(x) + x = (x + 1) / 2 + + return x + +class Discriminator(nn.Module): + def __init__(self, input_nc, norm_layer=nn.BatchNorm2d, use_sigmoid=False): + super(Discriminator, self).__init__() + + kw = 4 + padw = 1 + self.down1 = nn.Sequential( + nn.Conv2d(input_nc, 64, kernel_size=kw, stride=2, padding=padw), nn.LeakyReLU(0.2, True) + ) + self.down2 = nn.Sequential( + nn.Conv2d(64, 128, kernel_size=kw, stride=2, padding=padw), + norm_layer(128), nn.LeakyReLU(0.2, True) + ) + self.down3 = nn.Sequential( + nn.Conv2d(128, 256, kernel_size=kw, stride=2, padding=padw), + norm_layer(256), nn.LeakyReLU(0.2, True) + ) + self.down4 = nn.Sequential( + nn.Conv2d(256, 512, kernel_size=kw, stride=2, padding=padw), + norm_layer(512), nn.LeakyReLU(0.2, True) + ) + self.conv1 = nn.Sequential( + nn.Conv2d(512, 512, kernel_size=kw, stride=1, padding=padw), + norm_layer(512), + nn.LeakyReLU(0.2, True) + ) + + if use_sigmoid: + self.conv2 = nn.Sequential( + nn.Conv2d(512, 1, kernel_size=kw, stride=1, padding=padw), nn.Sigmoid() + ) + else: + self.conv2 = nn.Sequential( + nn.Conv2d(512, 1, kernel_size=kw, stride=1, padding=padw) + ) + + def forward(self, input): + out = [] + x = self.down1(input) + out.append(x) + x = self.down2(x) + out.append(x) + x = self.down3(x) + out.append(x) + x = self.down4(x) + out.append(x) + x = self.conv1(x) + out.append(x) + x = self.conv2(x) + out.append(x) + + return out diff --git a/mlflow/registry/Swimswap/models/fs_networks_fix.py b/mlflow/registry/Swimswap/models/fs_networks_fix.py new file mode 100644 index 0000000..c7b0525 --- /dev/null +++ b/mlflow/registry/Swimswap/models/fs_networks_fix.py @@ -0,0 +1,172 @@ +""" +Copyright (C) 2019 NVIDIA Corporation. All rights reserved. +Licensed under the CC BY-NC-SA 4.0 license (https://creativecommons.org/licenses/by-nc-sa/4.0/legalcode). +""" + +import torch +import torch.nn as nn + + +class InstanceNorm(nn.Module): + def __init__(self, epsilon=1e-8): + """ + @notice: avoid in-place ops. + https://discuss.pytorch.org/t/encounter-the-runtimeerror-one-of-the-variables-needed-for-gradient-computation-has-been-modified-by-an-inplace-operation/836/3 + """ + super(InstanceNorm, self).__init__() + self.epsilon = epsilon + + def forward(self, x): + x = x - torch.mean(x, (2, 3), True) + tmp = torch.mul(x, x) # or x ** 2 + tmp = torch.rsqrt(torch.mean(tmp, (2, 3), True) + self.epsilon) + return x * tmp + +class ApplyStyle(nn.Module): + """ + @ref: https://github.com/lernapparat/lernapparat/blob/master/style_gan/pytorch_style_gan.ipynb + """ + def __init__(self, latent_size, channels): + super(ApplyStyle, self).__init__() + self.linear = nn.Linear(latent_size, channels * 2) + + def forward(self, x, latent): + style = self.linear(latent) # style => [batch_size, n_channels*2] + shape = [-1, 2, x.size(1), 1, 1] + style = style.view(shape) # [batch_size, 2, n_channels, ...] + #x = x * (style[:, 0] + 1.) + style[:, 1] + x = x * (style[:, 0] * 1 + 1.) + style[:, 1] * 1 + return x + +class ResnetBlock_Adain(nn.Module): + def __init__(self, dim, latent_size, padding_type, activation=nn.ReLU(True)): + super(ResnetBlock_Adain, self).__init__() + + p = 0 + conv1 = [] + if padding_type == 'reflect': + conv1 += [nn.ReflectionPad2d(1)] + elif padding_type == 'replicate': + conv1 += [nn.ReplicationPad2d(1)] + elif padding_type == 'zero': + p = 1 + else: + raise NotImplementedError('padding [%s] is not implemented' % padding_type) + conv1 += [nn.Conv2d(dim, dim, kernel_size=3, padding = p), InstanceNorm()] + self.conv1 = nn.Sequential(*conv1) + self.style1 = ApplyStyle(latent_size, dim) + self.act1 = activation + + p = 0 + conv2 = [] + if padding_type == 'reflect': + conv2 += [nn.ReflectionPad2d(1)] + elif padding_type == 'replicate': + conv2 += [nn.ReplicationPad2d(1)] + elif padding_type == 'zero': + p = 1 + else: + raise NotImplementedError('padding [%s] is not implemented' % padding_type) + conv2 += [nn.Conv2d(dim, dim, kernel_size=3, padding=p), InstanceNorm()] + self.conv2 = nn.Sequential(*conv2) + self.style2 = ApplyStyle(latent_size, dim) + + + def forward(self, x, dlatents_in_slice): + y = self.conv1(x) + y = self.style1(y, dlatents_in_slice) + y = self.act1(y) + y = self.conv2(y) + y = self.style2(y, dlatents_in_slice) + out = x + y + return out + + + +class Generator_Adain_Upsample(nn.Module): + def __init__(self, input_nc, output_nc, latent_size, n_blocks=6, deep=False, + norm_layer=nn.BatchNorm2d, + padding_type='reflect'): + assert (n_blocks >= 0) + super(Generator_Adain_Upsample, self).__init__() + + activation = nn.ReLU(True) + + self.deep = deep + + self.first_layer = nn.Sequential(nn.ReflectionPad2d(3), nn.Conv2d(input_nc, 64, kernel_size=7, padding=0), + norm_layer(64), activation) + ### downsample + self.down1 = nn.Sequential(nn.Conv2d(64, 128, kernel_size=3, stride=2, padding=1), + norm_layer(128), activation) + self.down2 = nn.Sequential(nn.Conv2d(128, 256, kernel_size=3, stride=2, padding=1), + norm_layer(256), activation) + self.down3 = nn.Sequential(nn.Conv2d(256, 512, kernel_size=3, stride=2, padding=1), + norm_layer(512), activation) + + if self.deep: + self.down4 = nn.Sequential(nn.Conv2d(512, 512, kernel_size=3, stride=2, padding=1), + norm_layer(512), activation) + + ### resnet blocks + BN = [] + for i in range(n_blocks): + BN += [ + ResnetBlock_Adain(512, latent_size=latent_size, padding_type=padding_type, activation=activation)] + self.BottleNeck = nn.Sequential(*BN) + + if self.deep: + self.up4 = nn.Sequential( + nn.Upsample(scale_factor=2, mode='bilinear',align_corners=False), + nn.Conv2d(512, 512, kernel_size=3, stride=1, padding=1), + nn.BatchNorm2d(512), activation + ) + self.up3 = nn.Sequential( + nn.Upsample(scale_factor=2, mode='bilinear',align_corners=False), + nn.Conv2d(512, 256, kernel_size=3, stride=1, padding=1), + nn.BatchNorm2d(256), activation + ) + self.up2 = nn.Sequential( + nn.Upsample(scale_factor=2, mode='bilinear',align_corners=False), + nn.Conv2d(256, 128, kernel_size=3, stride=1, padding=1), + nn.BatchNorm2d(128), activation + ) + self.up1 = nn.Sequential( + nn.Upsample(scale_factor=2, mode='bilinear',align_corners=False), + nn.Conv2d(128, 64, kernel_size=3, stride=1, padding=1), + nn.BatchNorm2d(64), activation + ) + self.last_layer = nn.Sequential(nn.ReflectionPad2d(3), nn.Conv2d(64, output_nc, kernel_size=7, padding=0)) + + def forward(self, input, dlatents): + x = input # 3*224*224 + + skip1 = self.first_layer(x) + skip2 = self.down1(skip1) + skip3 = self.down2(skip2) + if self.deep: + skip4 = self.down3(skip3) + x = self.down4(skip4) + else: + x = self.down3(skip3) + bot = [] + bot.append(x) + features = [] + for i in range(len(self.BottleNeck)): + x = self.BottleNeck[i](x, dlatents) + bot.append(x) + + if self.deep: + x = self.up4(x) + features.append(x) + x = self.up3(x) + features.append(x) + x = self.up2(x) + features.append(x) + x = self.up1(x) + features.append(x) + x = self.last_layer(x) + # x = (x + 1) / 2 + + # return x, bot, features, dlatents + return x \ No newline at end of file diff --git a/mlflow/registry/Swimswap/models/models.py b/mlflow/registry/Swimswap/models/models.py new file mode 100644 index 0000000..bf9e84e --- /dev/null +++ b/mlflow/registry/Swimswap/models/models.py @@ -0,0 +1,177 @@ +import math +import torch +import torch.nn.functional as F +from torch import nn +from torch.nn import Parameter +from .config import device, num_classes + + +def create_model(opt): + #from .pix2pixHD_model import Pix2PixHDModel, InferenceModel + from .fs_model import fsModel + model = fsModel() + + model.initialize(opt) + if opt.verbose: + print("model [%s] was created" % (model.name())) + + if opt.isTrain and len(opt.gpu_ids) and not opt.fp16: + model = torch.nn.DataParallel(model, device_ids=opt.gpu_ids) + + return model + + + +class SEBlock(nn.Module): + def __init__(self, channel, reduction=16): + super(SEBlock, self).__init__() + self.avg_pool = nn.AdaptiveAvgPool2d(1) + self.fc = nn.Sequential( + nn.Linear(channel, channel // reduction), + nn.PReLU(), + nn.Linear(channel // reduction, channel), + nn.Sigmoid() + ) + + def forward(self, x): + b, c, _, _ = x.size() + y = self.avg_pool(x).view(b, c) + y = self.fc(y).view(b, c, 1, 1) + return x * y + + +class IRBlock(nn.Module): + expansion = 1 + + def __init__(self, inplanes, planes, stride=1, downsample=None, use_se=True): + super(IRBlock, self).__init__() + self.bn0 = nn.BatchNorm2d(inplanes) + self.conv1 = conv3x3(inplanes, inplanes) + self.bn1 = nn.BatchNorm2d(inplanes) + self.prelu = nn.PReLU() + self.conv2 = conv3x3(inplanes, planes, stride) + self.bn2 = nn.BatchNorm2d(planes) + self.downsample = downsample + self.stride = stride + self.use_se = use_se + if self.use_se: + self.se = SEBlock(planes) + + def forward(self, x): + residual = x + out = self.bn0(x) + out = self.conv1(out) + out = self.bn1(out) + out = self.prelu(out) + + out = self.conv2(out) + out = self.bn2(out) + if self.use_se: + out = self.se(out) + + if self.downsample is not None: + residual = self.downsample(x) + + out += residual + out = self.prelu(out) + + return out + + +class ResNet(nn.Module): + + def __init__(self, block, layers, use_se=True): + self.inplanes = 64 + self.use_se = use_se + super(ResNet, self).__init__() + self.conv1 = nn.Conv2d(3, 64, kernel_size=3, stride=1, bias=False) + self.bn1 = nn.BatchNorm2d(64) + self.prelu = nn.PReLU() + self.maxpool = nn.MaxPool2d(kernel_size=2, stride=2) + self.layer1 = self._make_layer(block, 64, layers[0]) + self.layer2 = self._make_layer(block, 128, layers[1], stride=2) + self.layer3 = self._make_layer(block, 256, layers[2], stride=2) + self.layer4 = self._make_layer(block, 512, layers[3], stride=2) + self.bn2 = nn.BatchNorm2d(512) + self.dropout = nn.Dropout() + self.fc = nn.Linear(512 * 7 * 7, 512) + self.bn3 = nn.BatchNorm1d(512) + + for m in self.modules(): + if isinstance(m, nn.Conv2d): + nn.init.xavier_normal_(m.weight) + elif isinstance(m, nn.BatchNorm2d) or isinstance(m, nn.BatchNorm1d): + nn.init.constant_(m.weight, 1) + nn.init.constant_(m.bias, 0) + elif isinstance(m, nn.Linear): + nn.init.xavier_normal_(m.weight) + nn.init.constant_(m.bias, 0) + + def _make_layer(self, block, planes, blocks, stride=1): + downsample = None + if stride != 1 or self.inplanes != planes * block.expansion: + downsample = nn.Sequential( + nn.Conv2d(self.inplanes, planes * block.expansion, + kernel_size=1, stride=stride, bias=False), + nn.BatchNorm2d(planes * block.expansion), + ) + + layers = [] + layers.append(block(self.inplanes, planes, stride, downsample, use_se=self.use_se)) + self.inplanes = planes + for i in range(1, blocks): + layers.append(block(self.inplanes, planes, use_se=self.use_se)) + + return nn.Sequential(*layers) + + def forward(self, x): + x = self.conv1(x) + x = self.bn1(x) + x = self.prelu(x) + x = self.maxpool(x) + + x = self.layer1(x) + x = self.layer2(x) + x = self.layer3(x) + x = self.layer4(x) + + x = self.bn2(x) + x = self.dropout(x) + x = x.view(x.size(0), -1) + x = self.fc(x) + x = self.bn3(x) + + return x + + +class ArcMarginModel(nn.Module): + def __init__(self, args): + super(ArcMarginModel, self).__init__() + + self.weight = Parameter(torch.FloatTensor(num_classes, args.emb_size)) + nn.init.xavier_uniform_(self.weight) + + self.easy_margin = args.easy_margin + self.m = args.margin_m + self.s = args.margin_s + + self.cos_m = math.cos(self.m) + self.sin_m = math.sin(self.m) + self.th = math.cos(math.pi - self.m) + self.mm = math.sin(math.pi - self.m) * self.m + + def forward(self, input, label): + x = F.normalize(input) + W = F.normalize(self.weight) + cosine = F.linear(x, W) + sine = torch.sqrt(1.0 - torch.pow(cosine, 2)) + phi = cosine * self.cos_m - sine * self.sin_m # cos(theta + m) + if self.easy_margin: + phi = torch.where(cosine > 0, phi, cosine) + else: + phi = torch.where(cosine > self.th, phi, cosine - self.mm) + one_hot = torch.zeros(cosine.size(), device=device) + one_hot.scatter_(1, label.view(-1, 1).long(), 1) + output = (one_hot * phi) + ((1.0 - one_hot) * cosine) + output *= self.s + return output diff --git a/mlflow/registry/Swimswap/models/networks.py b/mlflow/registry/Swimswap/models/networks.py new file mode 100644 index 0000000..3c7eb4d --- /dev/null +++ b/mlflow/registry/Swimswap/models/networks.py @@ -0,0 +1,845 @@ +import torch +import torch.nn as nn +import functools +from torch.autograd import Variable +import numpy as np +from torchvision import transforms +import torch.nn.functional as F + +############################################################################### +# Functions +############################################################################### +def weights_init(m): + classname = m.__class__.__name__ + if classname.find('Conv') != -1: + m.weight.data.normal_(0.0, 0.02) + elif classname.find('BatchNorm2d') != -1: + m.weight.data.normal_(1.0, 0.02) + m.bias.data.fill_(0) + +def get_norm_layer(norm_type='instance'): + if norm_type == 'batch': + norm_layer = functools.partial(nn.BatchNorm2d, affine=True) + elif norm_type == 'instance': + norm_layer = functools.partial(nn.InstanceNorm2d, affine=False) + else: + raise NotImplementedError('normalization layer [%s] is not found' % norm_type) + return norm_layer + +def define_G(input_nc, output_nc, ngf, netG, n_downsample_global=3, n_blocks_global=9, n_local_enhancers=1, + n_blocks_local=3, norm='instance', gpu_ids=[]): + norm_layer = get_norm_layer(norm_type=norm) + if netG == 'global': + netG = GlobalGenerator(input_nc, output_nc, ngf, n_downsample_global, n_blocks_global, norm_layer) + elif netG == 'local': + netG = LocalEnhancer(input_nc, output_nc, ngf, n_downsample_global, n_blocks_global, + n_local_enhancers, n_blocks_local, norm_layer) + elif netG == 'encoder': + netG = Encoder(input_nc, output_nc, ngf, n_downsample_global, norm_layer) + else: + raise('generator not implemented!') + print(netG) + if len(gpu_ids) > 0: + assert(torch.cuda.is_available()) + netG.cuda(gpu_ids[0]) + netG.apply(weights_init) + return netG + +def define_G_Adain(input_nc, output_nc, latent_size, ngf, netG, n_downsample_global=2, n_blocks_global=4, norm='instance', gpu_ids=[]): + norm_layer = get_norm_layer(norm_type=norm) + netG = Generator_Adain(input_nc, output_nc, latent_size, ngf, n_downsample_global, n_blocks_global, norm_layer) + print(netG) + if len(gpu_ids) > 0: + assert(torch.cuda.is_available()) + netG.cuda(gpu_ids[0]) + netG.apply(weights_init) + return netG + +def define_G_Adain_Mask(input_nc, output_nc, latent_size, ngf, netG, n_downsample_global=2, n_blocks_global=4, norm='instance', gpu_ids=[]): + norm_layer = get_norm_layer(norm_type=norm) + netG = Generator_Adain_Mask(input_nc, output_nc, latent_size, ngf, n_downsample_global, n_blocks_global, norm_layer) + print(netG) + if len(gpu_ids) > 0: + assert(torch.cuda.is_available()) + netG.cuda(gpu_ids[0]) + netG.apply(weights_init) + return netG + +def define_G_Adain_Upsample(input_nc, output_nc, latent_size, ngf, netG, n_downsample_global=2, n_blocks_global=4, norm='instance', gpu_ids=[]): + norm_layer = get_norm_layer(norm_type=norm) + netG = Generator_Adain_Upsample(input_nc, output_nc, latent_size, ngf, n_downsample_global, n_blocks_global, norm_layer) + print(netG) + if len(gpu_ids) > 0: + assert(torch.cuda.is_available()) + netG.cuda(gpu_ids[0]) + netG.apply(weights_init) + return netG + +def define_G_Adain_2(input_nc, output_nc, latent_size, ngf, netG, n_downsample_global=2, n_blocks_global=4, norm='instance', gpu_ids=[]): + norm_layer = get_norm_layer(norm_type=norm) + netG = Generator_Adain_2(input_nc, output_nc, latent_size, ngf, n_downsample_global, n_blocks_global, norm_layer) + print(netG) + if len(gpu_ids) > 0: + assert(torch.cuda.is_available()) + netG.cuda(gpu_ids[0]) + netG.apply(weights_init) + return netG + +def define_D(input_nc, ndf, n_layers_D, norm='instance', use_sigmoid=False, num_D=1, getIntermFeat=False, gpu_ids=[]): + norm_layer = get_norm_layer(norm_type=norm) + netD = MultiscaleDiscriminator(input_nc, ndf, n_layers_D, norm_layer, use_sigmoid, num_D, getIntermFeat) + print(netD) + if len(gpu_ids) > 0: + assert(torch.cuda.is_available()) + netD.cuda(gpu_ids[0]) + netD.apply(weights_init) + return netD + +def print_network(net): + if isinstance(net, list): + net = net[0] + num_params = 0 + for param in net.parameters(): + num_params += param.numel() + print(net) + print('Total number of parameters: %d' % num_params) + +############################################################################## +# Losses +############################################################################## +class GANLoss(nn.Module): + def __init__(self, gan_mode, target_real_label=1.0, target_fake_label=0.0, + tensor=torch.FloatTensor, opt=None): + super(GANLoss, self).__init__() + self.real_label = target_real_label + self.fake_label = target_fake_label + self.real_label_tensor = None + self.fake_label_tensor = None + self.zero_tensor = None + self.Tensor = tensor + self.gan_mode = gan_mode + self.opt = opt + if gan_mode == 'ls': + pass + elif gan_mode == 'original': + pass + elif gan_mode == 'w': + pass + elif gan_mode == 'hinge': + pass + else: + raise ValueError('Unexpected gan_mode {}'.format(gan_mode)) + + def get_target_tensor(self, input, target_is_real): + if target_is_real: + if self.real_label_tensor is None: + self.real_label_tensor = self.Tensor(1).fill_(self.real_label) + self.real_label_tensor.requires_grad_(False) + return self.real_label_tensor.expand_as(input) + else: + if self.fake_label_tensor is None: + self.fake_label_tensor = self.Tensor(1).fill_(self.fake_label) + self.fake_label_tensor.requires_grad_(False) + return self.fake_label_tensor.expand_as(input) + + def get_zero_tensor(self, input): + if self.zero_tensor is None: + self.zero_tensor = self.Tensor(1).fill_(0) + self.zero_tensor.requires_grad_(False) + return self.zero_tensor.expand_as(input) + + def loss(self, input, target_is_real, for_discriminator=True): + if self.gan_mode == 'original': # cross entropy loss + target_tensor = self.get_target_tensor(input, target_is_real) + loss = F.binary_cross_entropy_with_logits(input, target_tensor) + return loss + elif self.gan_mode == 'ls': + target_tensor = self.get_target_tensor(input, target_is_real) + return F.mse_loss(input, target_tensor) + elif self.gan_mode == 'hinge': + if for_discriminator: + if target_is_real: + minval = torch.min(input - 1, self.get_zero_tensor(input)) + loss = -torch.mean(minval) + else: + minval = torch.min(-input - 1, self.get_zero_tensor(input)) + loss = -torch.mean(minval) + else: + assert target_is_real, "The generator's hinge loss must be aiming for real" + loss = -torch.mean(input) + return loss + else: + # wgan + if target_is_real: + return -input.mean() + else: + return input.mean() + + def __call__(self, input, target_is_real, for_discriminator=True): + # computing loss is a bit complicated because |input| may not be + # a tensor, but list of tensors in case of multiscale discriminator + if isinstance(input, list): + loss = 0 + for pred_i in input: + if isinstance(pred_i, list): + pred_i = pred_i[-1] + loss_tensor = self.loss(pred_i, target_is_real, for_discriminator) + bs = 1 if len(loss_tensor.size()) == 0 else loss_tensor.size(0) + new_loss = torch.mean(loss_tensor.view(bs, -1), dim=1) + loss += new_loss + return loss / len(input) + else: + return self.loss(input, target_is_real, for_discriminator) + +class VGGLoss(nn.Module): + def __init__(self, gpu_ids): + super(VGGLoss, self).__init__() + self.vgg = Vgg19().cuda() + self.criterion = nn.L1Loss() + self.weights = [1.0/32, 1.0/16, 1.0/8, 1.0/4, 1.0] + + def forward(self, x, y): + x_vgg, y_vgg = self.vgg(x), self.vgg(y) + loss = 0 + for i in range(len(x_vgg)): + loss += self.weights[i] * self.criterion(x_vgg[i], y_vgg[i].detach()) + return loss + +############################################################################## +# Generator +############################################################################## +class LocalEnhancer(nn.Module): + def __init__(self, input_nc, output_nc, ngf=32, n_downsample_global=3, n_blocks_global=9, + n_local_enhancers=1, n_blocks_local=3, norm_layer=nn.BatchNorm2d, padding_type='reflect'): + super(LocalEnhancer, self).__init__() + self.n_local_enhancers = n_local_enhancers + + ###### global generator model ##### + ngf_global = ngf * (2**n_local_enhancers) + model_global = GlobalGenerator(input_nc, output_nc, ngf_global, n_downsample_global, n_blocks_global, norm_layer).model + model_global = [model_global[i] for i in range(len(model_global)-3)] # get rid of final convolution layers + self.model = nn.Sequential(*model_global) + + ###### local enhancer layers ##### + for n in range(1, n_local_enhancers+1): + ### downsample + ngf_global = ngf * (2**(n_local_enhancers-n)) + model_downsample = [nn.ReflectionPad2d(3), nn.Conv2d(input_nc, ngf_global, kernel_size=7, padding=0), + norm_layer(ngf_global), nn.ReLU(True), + nn.Conv2d(ngf_global, ngf_global * 2, kernel_size=3, stride=2, padding=1), + norm_layer(ngf_global * 2), nn.ReLU(True)] + ### residual blocks + model_upsample = [] + for i in range(n_blocks_local): + model_upsample += [ResnetBlock(ngf_global * 2, padding_type=padding_type, norm_layer=norm_layer)] + + ### upsample + model_upsample += [nn.ConvTranspose2d(ngf_global * 2, ngf_global, kernel_size=3, stride=2, padding=1, output_padding=1), + norm_layer(ngf_global), nn.ReLU(True)] + + ### final convolution + if n == n_local_enhancers: + model_upsample += [nn.ReflectionPad2d(3), nn.Conv2d(ngf, output_nc, kernel_size=7, padding=0), nn.Tanh()] + + setattr(self, 'model'+str(n)+'_1', nn.Sequential(*model_downsample)) + setattr(self, 'model'+str(n)+'_2', nn.Sequential(*model_upsample)) + + self.downsample = nn.AvgPool2d(3, stride=2, padding=[1, 1], count_include_pad=False) + + def forward(self, input): + ### create input pyramid + input_downsampled = [input] + for i in range(self.n_local_enhancers): + input_downsampled.append(self.downsample(input_downsampled[-1])) + + ### output at coarest level + output_prev = self.model(input_downsampled[-1]) + ### build up one layer at a time + for n_local_enhancers in range(1, self.n_local_enhancers+1): + model_downsample = getattr(self, 'model'+str(n_local_enhancers)+'_1') + model_upsample = getattr(self, 'model'+str(n_local_enhancers)+'_2') + input_i = input_downsampled[self.n_local_enhancers-n_local_enhancers] + output_prev = model_upsample(model_downsample(input_i) + output_prev) + return output_prev + +class GlobalGenerator(nn.Module): + def __init__(self, input_nc, output_nc, ngf=64, n_downsampling=3, n_blocks=9, norm_layer=nn.BatchNorm2d, + padding_type='reflect'): + assert(n_blocks >= 0) + super(GlobalGenerator, self).__init__() + activation = nn.ReLU(True) + + model = [nn.ReflectionPad2d(3), nn.Conv2d(input_nc, ngf, kernel_size=7, padding=0), norm_layer(ngf), activation] + ### downsample + for i in range(n_downsampling): + mult = 2**i + model += [nn.Conv2d(ngf * mult, ngf * mult * 2, kernel_size=3, stride=2, padding=1), + norm_layer(ngf * mult * 2), activation] + + ### resnet blocks + mult = 2**n_downsampling + for i in range(n_blocks): + model += [ResnetBlock(ngf * mult, padding_type=padding_type, activation=activation, norm_layer=norm_layer)] + + ### upsample + for i in range(n_downsampling): + mult = 2**(n_downsampling - i) + model += [nn.ConvTranspose2d(ngf * mult, int(ngf * mult / 2), kernel_size=3, stride=2, padding=1, output_padding=1), + norm_layer(int(ngf * mult / 2)), activation] + model += [nn.ReflectionPad2d(3), nn.Conv2d(ngf, output_nc, kernel_size=7, padding=0), nn.Tanh()] + self.model = nn.Sequential(*model) + + def forward(self, input): + return self.model(input) + +# Define a resnet block +class ResnetBlock(nn.Module): + def __init__(self, dim, padding_type, norm_layer, activation=nn.ReLU(True), use_dropout=False): + super(ResnetBlock, self).__init__() + self.conv_block = self.build_conv_block(dim, padding_type, norm_layer, activation, use_dropout) + + def build_conv_block(self, dim, padding_type, norm_layer, activation, use_dropout): + conv_block = [] + p = 0 + if padding_type == 'reflect': + conv_block += [nn.ReflectionPad2d(1)] + elif padding_type == 'replicate': + conv_block += [nn.ReplicationPad2d(1)] + elif padding_type == 'zero': + p = 1 + else: + raise NotImplementedError('padding [%s] is not implemented' % padding_type) + + conv_block += [nn.Conv2d(dim, dim, kernel_size=3, padding=p), + norm_layer(dim), + activation] + if use_dropout: + conv_block += [nn.Dropout(0.5)] + + p = 0 + if padding_type == 'reflect': + conv_block += [nn.ReflectionPad2d(1)] + elif padding_type == 'replicate': + conv_block += [nn.ReplicationPad2d(1)] + elif padding_type == 'zero': + p = 1 + else: + raise NotImplementedError('padding [%s] is not implemented' % padding_type) + conv_block += [nn.Conv2d(dim, dim, kernel_size=3, padding=p), + norm_layer(dim)] + + return nn.Sequential(*conv_block) + + def forward(self, x): + out = x + self.conv_block(x) + return out + +class InstanceNorm(nn.Module): + def __init__(self, epsilon=1e-8): + """ + @notice: avoid in-place ops. + https://discuss.pytorch.org/t/encounter-the-runtimeerror-one-of-the-variables-needed-for-gradient-computation-has-been-modified-by-an-inplace-operation/836/3 + """ + super(InstanceNorm, self).__init__() + self.epsilon = epsilon + + def forward(self, x): + x = x - torch.mean(x, (2, 3), True) + tmp = torch.mul(x, x) # or x ** 2 + tmp = torch.rsqrt(torch.mean(tmp, (2, 3), True) + self.epsilon) + return x * tmp + +class SpecificNorm(nn.Module): + def __init__(self, epsilon=1e-8): + """ + @notice: avoid in-place ops. + https://discuss.pytorch.org/t/encounter-the-runtimeerror-one-of-the-variables-needed-for-gradient-computation-has-been-modified-by-an-inplace-operation/836/3 + """ + super(SpecificNorm, self).__init__() + self.mean = np.array([0.485, 0.456, 0.406]) + self.mean = torch.from_numpy(self.mean).float().cuda() + self.mean = self.mean.view([1, 3, 1, 1]) + + self.std = np.array([0.229, 0.224, 0.225]) + self.std = torch.from_numpy(self.std).float().cuda() + self.std = self.std.view([1, 3, 1, 1]) + + def forward(self, x): + mean = self.mean.expand([1, 3, x.shape[2], x.shape[3]]) + std = self.std.expand([1, 3, x.shape[2], x.shape[3]]) + + x = (x - mean) / std + + return x + +class ApplyStyle(nn.Module): + """ + @ref: https://github.com/lernapparat/lernapparat/blob/master/style_gan/pytorch_style_gan.ipynb + """ + def __init__(self, latent_size, channels): + super(ApplyStyle, self).__init__() + self.linear = nn.Linear(latent_size, channels * 2) + + def forward(self, x, latent): + style = self.linear(latent) # style => [batch_size, n_channels*2] + shape = [-1, 2, x.size(1), 1, 1] + style = style.view(shape) # [batch_size, 2, n_channels, ...] + x = x * (style[:, 0] + 1.) + style[:, 1] + return x + +class ResnetBlock_Adain(nn.Module): + def __init__(self, dim, latent_size, padding_type, activation=nn.ReLU(True)): + super(ResnetBlock_Adain, self).__init__() + + p = 0 + conv1 = [] + if padding_type == 'reflect': + conv1 += [nn.ReflectionPad2d(1)] + elif padding_type == 'replicate': + conv1 += [nn.ReplicationPad2d(1)] + elif padding_type == 'zero': + p = 1 + else: + raise NotImplementedError('padding [%s] is not implemented' % padding_type) + conv1 += [nn.Conv2d(dim, dim, kernel_size=3, padding = p), InstanceNorm()] + self.conv1 = nn.Sequential(*conv1) + self.style1 = ApplyStyle(latent_size, dim) + self.act1 = activation + + p = 0 + conv2 = [] + if padding_type == 'reflect': + conv2 += [nn.ReflectionPad2d(1)] + elif padding_type == 'replicate': + conv2 += [nn.ReplicationPad2d(1)] + elif padding_type == 'zero': + p = 1 + else: + raise NotImplementedError('padding [%s] is not implemented' % padding_type) + conv2 += [nn.Conv2d(dim, dim, kernel_size=3, padding=p), InstanceNorm()] + self.conv2 = nn.Sequential(*conv2) + self.style2 = ApplyStyle(latent_size, dim) + + + def forward(self, x, dlatents_in_slice): + y = self.conv1(x) + y = self.style1(y, dlatents_in_slice) + y = self.act1(y) + y = self.conv2(y) + y = self.style2(y, dlatents_in_slice) + out = x + y + return out + +class UpBlock_Adain(nn.Module): + def __init__(self, dim_in, dim_out, latent_size, padding_type, activation=nn.ReLU(True)): + super(UpBlock_Adain, self).__init__() + + p = 0 + conv1 = [nn.Upsample(scale_factor=2, mode='bilinear')] + if padding_type == 'reflect': + conv1 += [nn.ReflectionPad2d(1)] + elif padding_type == 'replicate': + conv1 += [nn.ReplicationPad2d(1)] + elif padding_type == 'zero': + p = 1 + else: + raise NotImplementedError('padding [%s] is not implemented' % padding_type) + conv1 += [nn.Conv2d(dim_in, dim_out, kernel_size=3, padding = p), InstanceNorm()] + self.conv1 = nn.Sequential(*conv1) + self.style1 = ApplyStyle(latent_size, dim_out) + self.act1 = activation + + + def forward(self, x, dlatents_in_slice): + y = self.conv1(x) + y = self.style1(y, dlatents_in_slice) + y = self.act1(y) + return y + +class Encoder(nn.Module): + def __init__(self, input_nc, output_nc, ngf=32, n_downsampling=4, norm_layer=nn.BatchNorm2d): + super(Encoder, self).__init__() + self.output_nc = output_nc + + model = [nn.ReflectionPad2d(3), nn.Conv2d(input_nc, ngf, kernel_size=7, padding=0), + norm_layer(ngf), nn.ReLU(True)] + ### downsample + for i in range(n_downsampling): + mult = 2**i + model += [nn.Conv2d(ngf * mult, ngf * mult * 2, kernel_size=3, stride=2, padding=1), + norm_layer(ngf * mult * 2), nn.ReLU(True)] + + ### upsample + for i in range(n_downsampling): + mult = 2**(n_downsampling - i) + model += [nn.ConvTranspose2d(ngf * mult, int(ngf * mult / 2), kernel_size=3, stride=2, padding=1, output_padding=1), + norm_layer(int(ngf * mult / 2)), nn.ReLU(True)] + + model += [nn.ReflectionPad2d(3), nn.Conv2d(ngf, output_nc, kernel_size=7, padding=0), nn.Tanh()] + self.model = nn.Sequential(*model) + + def forward(self, input, inst): + outputs = self.model(input) + + # instance-wise average pooling + outputs_mean = outputs.clone() + inst_list = np.unique(inst.cpu().numpy().astype(int)) + for i in inst_list: + for b in range(input.size()[0]): + indices = (inst[b:b+1] == int(i)).nonzero() # n x 4 + for j in range(self.output_nc): + output_ins = outputs[indices[:,0] + b, indices[:,1] + j, indices[:,2], indices[:,3]] + mean_feat = torch.mean(output_ins).expand_as(output_ins) + outputs_mean[indices[:,0] + b, indices[:,1] + j, indices[:,2], indices[:,3]] = mean_feat + return outputs_mean + + +class Generator_Adain(nn.Module): + def __init__(self, input_nc, output_nc, latent_size, ngf=64, n_downsampling=2, n_blocks=4, norm_layer=nn.BatchNorm2d, + padding_type='reflect'): + assert (n_blocks >= 0) + super(Generator_Adain, self).__init__() + activation = nn.ReLU(True) + + Enc = [nn.ReflectionPad2d(3), nn.Conv2d(input_nc, ngf, kernel_size=7, padding=0), norm_layer(ngf), activation] + ### downsample + for i in range(n_downsampling): + mult = 2 ** i + Enc += [nn.Conv2d(ngf * mult, ngf * mult * 2, kernel_size=3, stride=2, padding=1), + norm_layer(ngf * mult * 2), activation] + self.Encoder = nn.Sequential(*Enc) + + ### resnet blocks + BN = [] + mult = 2 ** n_downsampling + for i in range(n_blocks): + BN += [ResnetBlock_Adain(ngf*mult, latent_size=latent_size, padding_type=padding_type, activation=activation)] + self.BottleNeck = nn.Sequential(*BN) + '''self.ResBlockAdain1 = ResnetBlock_Adain(ngf * mult, latent_size=latent_size, padding_type=padding_type, + activation=activation) + self.ResBlockAdain2 = ResnetBlock_Adain(ngf * mult, latent_size=latent_size, padding_type=padding_type, + activation=activation) + self.ResBlockAdain3 = ResnetBlock_Adain(ngf * mult, latent_size=latent_size, padding_type=padding_type, + activation=activation) + self.ResBlockAdain4 = ResnetBlock_Adain(ngf * mult, latent_size=latent_size, padding_type=padding_type, + activation=activation)''' + + ### upsample + Dec = [] + for i in range(n_downsampling): + mult = 2 ** (n_downsampling - i) + Dec += [nn.ConvTranspose2d(ngf * mult, int(ngf * mult / 2), kernel_size=3, stride=2, padding=1, + output_padding=1), + norm_layer(int(ngf * mult / 2)), activation] + Dec += [nn.ReflectionPad2d(3), nn.Conv2d(ngf, output_nc, kernel_size=7, padding=0), nn.Tanh()] + + self.Decoder = nn.Sequential(*Dec) + #self.model = nn.Sequential(*model) + self.spNorm = SpecificNorm() + + def forward(self, input, dlatents): + x = input + x = self.Encoder(x) + + + for i in range(len(self.BottleNeck)): + x = self.BottleNeck[i](x, dlatents) + '''x = self.ResBlockAdain1(x, dlatents) + x = self.ResBlockAdain2(x, dlatents) + x = self.ResBlockAdain3(x, dlatents) + x = self.ResBlockAdain4(x, dlatents)''' + + x = self.Decoder(x) + + x = (x + 1) / 2 + x = self.spNorm(x) + + return x + +class Generator_Adain_Mask(nn.Module): + def __init__(self, input_nc, output_nc, latent_size, ngf=64, n_downsampling=2, n_blocks=4, norm_layer=nn.BatchNorm2d, + padding_type='reflect'): + assert (n_blocks >= 0) + super(Generator_Adain_Mask, self).__init__() + activation = nn.ReLU(True) + + Enc = [nn.ReflectionPad2d(3), nn.Conv2d(input_nc, ngf, kernel_size=7, padding=0), norm_layer(ngf), activation] + ### downsample + for i in range(n_downsampling): + mult = 2 ** i + Enc += [nn.Conv2d(ngf * mult, ngf * mult * 2, kernel_size=3, stride=2, padding=1), + norm_layer(ngf * mult * 2), activation] + self.Encoder = nn.Sequential(*Enc) + + ### resnet blocks + BN = [] + mult = 2 ** n_downsampling + for i in range(n_blocks): + BN += [ResnetBlock_Adain(ngf*mult, latent_size=latent_size, padding_type=padding_type, activation=activation)] + self.BottleNeck = nn.Sequential(*BN) + + ### upsample + Dec = [] + for i in range(n_downsampling): + mult = 2 ** (n_downsampling - i) + Dec += [nn.ConvTranspose2d(ngf * mult, int(ngf * mult / 2), kernel_size=3, stride=2, padding=1, + output_padding=1), + norm_layer(int(ngf * mult / 2)), activation] + Fake_out = [nn.ReflectionPad2d(3), nn.Conv2d(ngf, output_nc, kernel_size=7, padding=0), nn.Tanh()] + Mast_out = [nn.ReflectionPad2d(3), nn.Conv2d(ngf, 1, kernel_size=7, padding=0), nn.Sigmoid()] + + self.Decoder = nn.Sequential(*Dec) + #self.model = nn.Sequential(*model) + self.spNorm = SpecificNorm() + self.Fake_out = nn.Sequential(*Fake_out) + self.Mask_out = nn.Sequential(*Mast_out) + + def forward(self, input, dlatents): + x = input + x = self.Encoder(x) + + + for i in range(len(self.BottleNeck)): + x = self.BottleNeck[i](x, dlatents) + + x = self.Decoder(x) + + fake_out = self.Fake_out(x) + mask_out = self.Mask_out(x) + + fake_out = (fake_out + 1) / 2 + fake_out = self.spNorm(fake_out) + + generated = fake_out * mask_out + input * (1-mask_out) + + return generated, mask_out + +class Generator_Adain_Upsample(nn.Module): + def __init__(self, input_nc, output_nc, latent_size, ngf=64, n_downsampling=2, n_blocks=4, norm_layer=nn.BatchNorm2d, + padding_type='reflect'): + assert (n_blocks >= 0) + super(Generator_Adain_Upsample, self).__init__() + activation = nn.ReLU(True) + + Enc = [nn.ReflectionPad2d(3), nn.Conv2d(input_nc, ngf, kernel_size=7, padding=0), norm_layer(ngf), activation] + ### downsample + for i in range(n_downsampling): + mult = 2 ** i + Enc += [nn.Conv2d(ngf * mult, ngf * mult * 2, kernel_size=3, stride=2, padding=1), + norm_layer(ngf * mult * 2), activation] + self.Encoder = nn.Sequential(*Enc) + + ### resnet blocks + BN = [] + mult = 2 ** n_downsampling + for i in range(n_blocks): + BN += [ResnetBlock_Adain(ngf*mult, latent_size=latent_size, padding_type=padding_type, activation=activation)] + self.BottleNeck = nn.Sequential(*BN) + + ### upsample + Dec = [] + for i in range(n_downsampling): + mult = 2 ** (n_downsampling - i) + '''Dec += [nn.ConvTranspose2d(ngf * mult, int(ngf * mult / 2), kernel_size=3, stride=2, padding=1, + output_padding=1), + norm_layer(int(ngf * mult / 2)), activation]''' + Dec += [nn.Upsample(scale_factor=2, mode='bilinear'), + nn.Conv2d(ngf * mult, int(ngf * mult / 2), kernel_size=3, stride=1, padding=1), + norm_layer(int(ngf * mult / 2)), activation] + Dec += [nn.ReflectionPad2d(3), nn.Conv2d(ngf, output_nc, kernel_size=7, padding=0), nn.Tanh()] + + self.Decoder = nn.Sequential(*Dec) + self.spNorm = SpecificNorm() + + def forward(self, input, dlatents): + x = input + x = self.Encoder(x) + + + for i in range(len(self.BottleNeck)): + x = self.BottleNeck[i](x, dlatents) + + x = self.Decoder(x) + + x = (x + 1) / 2 + x = self.spNorm(x) + + return x + +class Generator_Adain_2(nn.Module): + def __init__(self, input_nc, output_nc, latent_size, ngf=64, n_downsampling=2, n_blocks=4, norm_layer=nn.BatchNorm2d, + padding_type='reflect'): + assert (n_blocks >= 0) + super(Generator_Adain_2, self).__init__() + activation = nn.ReLU(True) + + Enc = [nn.ReflectionPad2d(3), nn.Conv2d(input_nc, ngf, kernel_size=7, padding=0), norm_layer(ngf), activation] + ### downsample + for i in range(n_downsampling): + mult = 2 ** i + Enc += [nn.Conv2d(ngf * mult, ngf * mult * 2, kernel_size=3, stride=2, padding=1), + norm_layer(ngf * mult * 2), activation] + self.Encoder = nn.Sequential(*Enc) + + ### resnet blocks + BN = [] + mult = 2 ** n_downsampling + for i in range(n_blocks): + BN += [ResnetBlock_Adain(ngf*mult, latent_size=latent_size, padding_type=padding_type, activation=activation)] + self.BottleNeck = nn.Sequential(*BN) + + ### upsample + Dec = [] + for i in range(n_downsampling): + mult = 2 ** (n_downsampling - i) + Dec += [UpBlock_Adain(dim_in=ngf * mult, dim_out=int(ngf * mult / 2), latent_size=latent_size, padding_type=padding_type)] + layer_out = [nn.ReflectionPad2d(3), nn.Conv2d(ngf, output_nc, kernel_size=7, padding=0), nn.Tanh()] + + self.Decoder = nn.Sequential(*Dec) + #self.model = nn.Sequential(*model) + self.spNorm = SpecificNorm() + self.layer_out = nn.Sequential(*layer_out) + + def forward(self, input, dlatents): + x = input + x = self.Encoder(x) + + + for i in range(len(self.BottleNeck)): + x = self.BottleNeck[i](x, dlatents) + + for i in range(len(self.Decoder)): + x = self.Decoder[i](x, dlatents) + + x = self.layer_out(x) + + x = (x + 1) / 2 + x = self.spNorm(x) + + return x + +class MultiscaleDiscriminator(nn.Module): + def __init__(self, input_nc, ndf=64, n_layers=3, norm_layer=nn.BatchNorm2d, + use_sigmoid=False, num_D=3, getIntermFeat=False): + super(MultiscaleDiscriminator, self).__init__() + self.num_D = num_D + self.n_layers = n_layers + self.getIntermFeat = getIntermFeat + + for i in range(num_D): + netD = NLayerDiscriminator(input_nc, ndf, n_layers, norm_layer, use_sigmoid, getIntermFeat) + if getIntermFeat: + for j in range(n_layers+2): + setattr(self, 'scale'+str(i)+'_layer'+str(j), getattr(netD, 'model'+str(j))) + else: + setattr(self, 'layer'+str(i), netD.model) + + self.downsample = nn.AvgPool2d(3, stride=2, padding=[1, 1], count_include_pad=False) + + def singleD_forward(self, model, input): + if self.getIntermFeat: + result = [input] + for i in range(len(model)): + result.append(model[i](result[-1])) + return result[1:] + else: + return [model(input)] + + def forward(self, input): + num_D = self.num_D + result = [] + input_downsampled = input + for i in range(num_D): + if self.getIntermFeat: + model = [getattr(self, 'scale'+str(num_D-1-i)+'_layer'+str(j)) for j in range(self.n_layers+2)] + else: + model = getattr(self, 'layer'+str(num_D-1-i)) + result.append(self.singleD_forward(model, input_downsampled)) + if i != (num_D-1): + input_downsampled = self.downsample(input_downsampled) + return result + +# Defines the PatchGAN discriminator with the specified arguments. +class NLayerDiscriminator(nn.Module): + def __init__(self, input_nc, ndf=64, n_layers=3, norm_layer=nn.BatchNorm2d, use_sigmoid=False, getIntermFeat=False): + super(NLayerDiscriminator, self).__init__() + self.getIntermFeat = getIntermFeat + self.n_layers = n_layers + + kw = 4 + padw = 1 + sequence = [[nn.Conv2d(input_nc, ndf, kernel_size=kw, stride=2, padding=padw), nn.LeakyReLU(0.2, True)]] + + nf = ndf + for n in range(1, n_layers): + nf_prev = nf + nf = min(nf * 2, 512) + sequence += [[ + nn.Conv2d(nf_prev, nf, kernel_size=kw, stride=2, padding=padw), + norm_layer(nf), nn.LeakyReLU(0.2, True) + ]] + + nf_prev = nf + nf = min(nf * 2, 512) + sequence += [[ + nn.Conv2d(nf_prev, nf, kernel_size=kw, stride=1, padding=padw), + norm_layer(nf), + nn.LeakyReLU(0.2, True) + ]] + + if use_sigmoid: + sequence += [[nn.Conv2d(nf, 1, kernel_size=kw, stride=1, padding=padw), nn.Sigmoid()]] + else: + sequence += [[nn.Conv2d(nf, 1, kernel_size=kw, stride=1, padding=padw)]] + + if getIntermFeat: + for n in range(len(sequence)): + setattr(self, 'model'+str(n), nn.Sequential(*sequence[n])) + else: + sequence_stream = [] + for n in range(len(sequence)): + sequence_stream += sequence[n] + self.model = nn.Sequential(*sequence_stream) + + def forward(self, input): + if self.getIntermFeat: + res = [input] + for n in range(self.n_layers+2): + model = getattr(self, 'model'+str(n)) + res.append(model(res[-1])) + return res[1:] + else: + return self.model(input) + +from torchvision import models +class Vgg19(torch.nn.Module): + def __init__(self, requires_grad=False): + super(Vgg19, self).__init__() + vgg_pretrained_features = models.vgg19(pretrained=True).features + self.slice1 = torch.nn.Sequential() + self.slice2 = torch.nn.Sequential() + self.slice3 = torch.nn.Sequential() + self.slice4 = torch.nn.Sequential() + self.slice5 = torch.nn.Sequential() + for x in range(2): + self.slice1.add_module(str(x), vgg_pretrained_features[x]) + for x in range(2, 7): + self.slice2.add_module(str(x), vgg_pretrained_features[x]) + for x in range(7, 12): + self.slice3.add_module(str(x), vgg_pretrained_features[x]) + for x in range(12, 21): + self.slice4.add_module(str(x), vgg_pretrained_features[x]) + for x in range(21, 30): + self.slice5.add_module(str(x), vgg_pretrained_features[x]) + if not requires_grad: + for param in self.parameters(): + param.requires_grad = False + + def forward(self, X): + h_relu1 = self.slice1(X) + h_relu2 = self.slice2(h_relu1) + h_relu3 = self.slice3(h_relu2) + h_relu4 = self.slice4(h_relu3) + h_relu5 = self.slice5(h_relu4) + out = [h_relu1, h_relu2, h_relu3, h_relu4, h_relu5] + return out diff --git a/mlflow/registry/Swimswap/models/pix2pixHD_model.py b/mlflow/registry/Swimswap/models/pix2pixHD_model.py new file mode 100644 index 0000000..fafdec0 --- /dev/null +++ b/mlflow/registry/Swimswap/models/pix2pixHD_model.py @@ -0,0 +1,304 @@ +import numpy as np +import torch +import os +from torch.autograd import Variable +from util.image_pool import ImagePool +from .base_model import BaseModel +from . import networks + +class Pix2PixHDModel(BaseModel): + def name(self): + return 'Pix2PixHDModel' + + def init_loss_filter(self, use_gan_feat_loss, use_vgg_loss): + flags = (True, use_gan_feat_loss, use_vgg_loss, True, True) + def loss_filter(g_gan, g_gan_feat, g_vgg, d_real, d_fake): + return [l for (l,f) in zip((g_gan,g_gan_feat,g_vgg,d_real,d_fake),flags) if f] + return loss_filter + + def initialize(self, opt): + BaseModel.initialize(self, opt) + if opt.resize_or_crop != 'none' or not opt.isTrain: # when training at full res this causes OOM + torch.backends.cudnn.benchmark = True + self.isTrain = opt.isTrain + self.use_features = opt.instance_feat or opt.label_feat + self.gen_features = self.use_features and not self.opt.load_features + input_nc = opt.label_nc if opt.label_nc != 0 else opt.input_nc + + ##### define networks + # Generator network + netG_input_nc = input_nc + if not opt.no_instance: + netG_input_nc += 1 + if self.use_features: + netG_input_nc += opt.feat_num + self.netG = networks.define_G(netG_input_nc, opt.output_nc, opt.ngf, opt.netG, + opt.n_downsample_global, opt.n_blocks_global, opt.n_local_enhancers, + opt.n_blocks_local, opt.norm, gpu_ids=self.gpu_ids) + + # Discriminator network + if self.isTrain: + use_sigmoid = opt.no_lsgan + netD_input_nc = input_nc + opt.output_nc + if not opt.no_instance: + netD_input_nc += 1 + self.netD = networks.define_D(netD_input_nc, opt.ndf, opt.n_layers_D, opt.norm, use_sigmoid, + opt.num_D, not opt.no_ganFeat_loss, gpu_ids=self.gpu_ids) + + ### Encoder network + if self.gen_features: + self.netE = networks.define_G(opt.output_nc, opt.feat_num, opt.nef, 'encoder', + opt.n_downsample_E, norm=opt.norm, gpu_ids=self.gpu_ids) + if self.opt.verbose: + print('---------- Networks initialized -------------') + + # load networks + if not self.isTrain or opt.continue_train or opt.load_pretrain: + pretrained_path = '' if not self.isTrain else opt.load_pretrain + self.load_network(self.netG, 'G', opt.which_epoch, pretrained_path) + if self.isTrain: + self.load_network(self.netD, 'D', opt.which_epoch, pretrained_path) + if self.gen_features: + self.load_network(self.netE, 'E', opt.which_epoch, pretrained_path) + + # set loss functions and optimizers + if self.isTrain: + if opt.pool_size > 0 and (len(self.gpu_ids)) > 1: + raise NotImplementedError("Fake Pool Not Implemented for MultiGPU") + self.fake_pool = ImagePool(opt.pool_size) + self.old_lr = opt.lr + + # define loss functions + self.loss_filter = self.init_loss_filter(not opt.no_ganFeat_loss, not opt.no_vgg_loss) + + self.criterionGAN = networks.GANLoss(use_lsgan=not opt.no_lsgan, tensor=self.Tensor) + self.criterionFeat = torch.nn.L1Loss() + if not opt.no_vgg_loss: + self.criterionVGG = networks.VGGLoss(self.gpu_ids) + + + # Names so we can breakout loss + self.loss_names = self.loss_filter('G_GAN','G_GAN_Feat','G_VGG','D_real', 'D_fake') + + # initialize optimizers + # optimizer G + if opt.niter_fix_global > 0: + import sys + if sys.version_info >= (3,0): + finetune_list = set() + else: + from sets import Set + finetune_list = Set() + + params_dict = dict(self.netG.named_parameters()) + params = [] + for key, value in params_dict.items(): + if key.startswith('model' + str(opt.n_local_enhancers)): + params += [value] + finetune_list.add(key.split('.')[0]) + print('------------- Only training the local enhancer network (for %d epochs) ------------' % opt.niter_fix_global) + print('The layers that are finetuned are ', sorted(finetune_list)) + else: + params = list(self.netG.parameters()) + if self.gen_features: + params += list(self.netE.parameters()) + self.optimizer_G = torch.optim.Adam(params, lr=opt.lr, betas=(opt.beta1, 0.999)) + + # optimizer D + params = list(self.netD.parameters()) + self.optimizer_D = torch.optim.Adam(params, lr=opt.lr, betas=(opt.beta1, 0.999)) + + def encode_input(self, label_map, inst_map=None, real_image=None, feat_map=None, infer=False): + if self.opt.label_nc == 0: + input_label = label_map.data.cuda() + else: + # create one-hot vector for label map + size = label_map.size() + oneHot_size = (size[0], self.opt.label_nc, size[2], size[3]) + input_label = torch.cuda.FloatTensor(torch.Size(oneHot_size)).zero_() + input_label = input_label.scatter_(1, label_map.data.long().cuda(), 1.0) + if self.opt.data_type == 16: + input_label = input_label.half() + + # get edges from instance map + if not self.opt.no_instance: + inst_map = inst_map.data.cuda() + edge_map = self.get_edges(inst_map) + input_label = torch.cat((input_label, edge_map), dim=1) + input_label = Variable(input_label, volatile=infer) + + # real images for training + if real_image is not None: + real_image = Variable(real_image.data.cuda()) + + # instance map for feature encoding + if self.use_features: + # get precomputed feature maps + if self.opt.load_features: + feat_map = Variable(feat_map.data.cuda()) + if self.opt.label_feat: + inst_map = label_map.cuda() + + return input_label, inst_map, real_image, feat_map + + def discriminate(self, input_label, test_image, use_pool=False): + input_concat = torch.cat((input_label, test_image.detach()), dim=1) + if use_pool: + fake_query = self.fake_pool.query(input_concat) + return self.netD.forward(fake_query) + else: + return self.netD.forward(input_concat) + + def forward(self, label, inst, image, feat, infer=False): + # Encode Inputs + input_label, inst_map, real_image, feat_map = self.encode_input(label, inst, image, feat) + + # Fake Generation + if self.use_features: + if not self.opt.load_features: + feat_map = self.netE.forward(real_image, inst_map) + input_concat = torch.cat((input_label, feat_map), dim=1) + else: + input_concat = input_label + fake_image = self.netG.forward(input_concat) + + # Fake Detection and Loss + pred_fake_pool = self.discriminate(input_label, fake_image, use_pool=True) + loss_D_fake = self.criterionGAN(pred_fake_pool, False) + + # Real Detection and Loss + pred_real = self.discriminate(input_label, real_image) + loss_D_real = self.criterionGAN(pred_real, True) + + # GAN loss (Fake Passability Loss) + pred_fake = self.netD.forward(torch.cat((input_label, fake_image), dim=1)) + loss_G_GAN = self.criterionGAN(pred_fake, True) + + # GAN feature matching loss + loss_G_GAN_Feat = 0 + if not self.opt.no_ganFeat_loss: + feat_weights = 4.0 / (self.opt.n_layers_D + 1) + D_weights = 1.0 / self.opt.num_D + for i in range(self.opt.num_D): + for j in range(len(pred_fake[i])-1): + loss_G_GAN_Feat += D_weights * feat_weights * \ + self.criterionFeat(pred_fake[i][j], pred_real[i][j].detach()) * self.opt.lambda_feat + + # VGG feature matching loss + loss_G_VGG = 0 + if not self.opt.no_vgg_loss: + loss_G_VGG = self.criterionVGG(fake_image, real_image) * self.opt.lambda_feat + + # Only return the fake_B image if necessary to save BW + return [ self.loss_filter( loss_G_GAN, loss_G_GAN_Feat, loss_G_VGG, loss_D_real, loss_D_fake ), None if not infer else fake_image ] + + def inference(self, label, inst, image=None): + # Encode Inputs + image = Variable(image) if image is not None else None + input_label, inst_map, real_image, _ = self.encode_input(Variable(label), Variable(inst), image, infer=True) + + # Fake Generation + if self.use_features: + if self.opt.use_encoded_image: + # encode the real image to get feature map + feat_map = self.netE.forward(real_image, inst_map) + else: + # sample clusters from precomputed features + feat_map = self.sample_features(inst_map) + input_concat = torch.cat((input_label, feat_map), dim=1) + else: + input_concat = input_label + + if torch.__version__.startswith('0.4'): + with torch.no_grad(): + fake_image = self.netG.forward(input_concat) + else: + fake_image = self.netG.forward(input_concat) + return fake_image + + def sample_features(self, inst): + # read precomputed feature clusters + cluster_path = os.path.join(self.opt.checkpoints_dir, self.opt.name, self.opt.cluster_path) + features_clustered = np.load(cluster_path, encoding='latin1').item() + + # randomly sample from the feature clusters + inst_np = inst.cpu().numpy().astype(int) + feat_map = self.Tensor(inst.size()[0], self.opt.feat_num, inst.size()[2], inst.size()[3]) + for i in np.unique(inst_np): + label = i if i < 1000 else i//1000 + if label in features_clustered: + feat = features_clustered[label] + cluster_idx = np.random.randint(0, feat.shape[0]) + + idx = (inst == int(i)).nonzero() + for k in range(self.opt.feat_num): + feat_map[idx[:,0], idx[:,1] + k, idx[:,2], idx[:,3]] = feat[cluster_idx, k] + if self.opt.data_type==16: + feat_map = feat_map.half() + return feat_map + + def encode_features(self, image, inst): + image = Variable(image.cuda(), volatile=True) + feat_num = self.opt.feat_num + h, w = inst.size()[2], inst.size()[3] + block_num = 32 + feat_map = self.netE.forward(image, inst.cuda()) + inst_np = inst.cpu().numpy().astype(int) + feature = {} + for i in range(self.opt.label_nc): + feature[i] = np.zeros((0, feat_num+1)) + for i in np.unique(inst_np): + label = i if i < 1000 else i//1000 + idx = (inst == int(i)).nonzero() + num = idx.size()[0] + idx = idx[num//2,:] + val = np.zeros((1, feat_num+1)) + for k in range(feat_num): + val[0, k] = feat_map[idx[0], idx[1] + k, idx[2], idx[3]].data[0] + val[0, feat_num] = float(num) / (h * w // block_num) + feature[label] = np.append(feature[label], val, axis=0) + return feature + + def get_edges(self, t): + edge = torch.cuda.ByteTensor(t.size()).zero_() + edge[:,:,:,1:] = edge[:,:,:,1:] | (t[:,:,:,1:] != t[:,:,:,:-1]) + edge[:,:,:,:-1] = edge[:,:,:,:-1] | (t[:,:,:,1:] != t[:,:,:,:-1]) + edge[:,:,1:,:] = edge[:,:,1:,:] | (t[:,:,1:,:] != t[:,:,:-1,:]) + edge[:,:,:-1,:] = edge[:,:,:-1,:] | (t[:,:,1:,:] != t[:,:,:-1,:]) + if self.opt.data_type==16: + return edge.half() + else: + return edge.float() + + def save(self, which_epoch): + self.save_network(self.netG, 'G', which_epoch, self.gpu_ids) + self.save_network(self.netD, 'D', which_epoch, self.gpu_ids) + if self.gen_features: + self.save_network(self.netE, 'E', which_epoch, self.gpu_ids) + + def update_fixed_params(self): + # after fixing the global generator for a number of iterations, also start finetuning it + params = list(self.netG.parameters()) + if self.gen_features: + params += list(self.netE.parameters()) + self.optimizer_G = torch.optim.Adam(params, lr=self.opt.lr, betas=(self.opt.beta1, 0.999)) + if self.opt.verbose: + print('------------ Now also finetuning global generator -----------') + + def update_learning_rate(self): + lrd = self.opt.lr / self.opt.niter_decay + lr = self.old_lr - lrd + for param_group in self.optimizer_D.param_groups: + param_group['lr'] = lr + for param_group in self.optimizer_G.param_groups: + param_group['lr'] = lr + if self.opt.verbose: + print('update learning rate: %f -> %f' % (self.old_lr, lr)) + self.old_lr = lr + +class InferenceModel(Pix2PixHDModel): + def forward(self, inp): + label, inst = inp + return self.inference(label, inst) + + diff --git a/mlflow/registry/Swimswap/models/projected_model.py b/mlflow/registry/Swimswap/models/projected_model.py new file mode 100644 index 0000000..15d70f9 --- /dev/null +++ b/mlflow/registry/Swimswap/models/projected_model.py @@ -0,0 +1,122 @@ +#!/usr/bin/env python3 +# -*- coding:utf-8 -*- +############################################################# +# File: fs_model_fix_idnorm_donggp_saveoptim copy.py +# Created Date: Wednesday January 12th 2022 +# Author: Chen Xuanhong +# Email: chenxuanhongzju@outlook.com +# Last Modified: Saturday, 13th May 2023 9:56:35 am +# Modified By: Chen Xuanhong +# Copyright (c) 2022 Shanghai Jiao Tong University +############################################################# + + +import torch +import torch.nn as nn + +from .base_model import BaseModel +from .fs_networks_fix import Generator_Adain_Upsample + +from pg_modules.projected_discriminator import ProjectedDiscriminator + +def compute_grad2(d_out, x_in): + batch_size = x_in.size(0) + grad_dout = torch.autograd.grad( + outputs=d_out.sum(), inputs=x_in, + create_graph=True, retain_graph=True, only_inputs=True + )[0] + grad_dout2 = grad_dout.pow(2) + assert(grad_dout2.size() == x_in.size()) + reg = grad_dout2.view(batch_size, -1).sum(1) + return reg + +class fsModel(BaseModel): + def name(self): + return 'fsModel' + + def initialize(self, opt): + BaseModel.initialize(self, opt) + # if opt.resize_or_crop != 'none' or not opt.isTrain: # when training at full res this causes OOM + self.isTrain = opt.isTrain + + # Generator network + self.netG = Generator_Adain_Upsample(input_nc=3, output_nc=3, latent_size=512, n_blocks=9, deep=opt.Gdeep) + self.netG.cuda() + + # Id network + netArc_checkpoint = opt.Arc_path + netArc_checkpoint = torch.load(netArc_checkpoint, map_location=torch.device("cpu")) + self.netArc = netArc_checkpoint + self.netArc = self.netArc.cuda() + self.netArc.eval() + self.netArc.requires_grad_(False) + if not self.isTrain: + pretrained_path = opt.checkpoints_dir + self.load_network(self.netG, 'G', opt.which_epoch, pretrained_path) + return + self.netD = ProjectedDiscriminator(diffaug=False, interp224=False, **{}) + # self.netD.feature_network.requires_grad_(False) + self.netD.cuda() + + + if self.isTrain: + # define loss functions + self.criterionFeat = nn.L1Loss() + self.criterionRec = nn.L1Loss() + + + # initialize optimizers + + # optimizer G + params = list(self.netG.parameters()) + self.optimizer_G = torch.optim.Adam(params, lr=opt.lr, betas=(opt.beta1, 0.99),eps=1e-8) + + # optimizer D + params = list(self.netD.parameters()) + self.optimizer_D = torch.optim.Adam(params, lr=opt.lr, betas=(opt.beta1, 0.99),eps=1e-8) + + # load networks + if opt.continue_train: + pretrained_path = '' if not self.isTrain else opt.load_pretrain + # print (pretrained_path) + self.load_network(self.netG, 'G', opt.which_epoch, pretrained_path) + self.load_network(self.netD, 'D', opt.which_epoch, pretrained_path) + self.load_optim(self.optimizer_G, 'G', opt.which_epoch, pretrained_path) + self.load_optim(self.optimizer_D, 'D', opt.which_epoch, pretrained_path) + torch.cuda.empty_cache() + + def cosin_metric(self, x1, x2): + #return np.dot(x1, x2) / (np.linalg.norm(x1) * np.linalg.norm(x2)) + return torch.sum(x1 * x2, dim=1) / (torch.norm(x1, dim=1) * torch.norm(x2, dim=1)) + + + + def save(self, which_epoch): + self.save_network(self.netG, 'G', which_epoch) + self.save_network(self.netD, 'D', which_epoch) + self.save_optim(self.optimizer_G, 'G', which_epoch) + self.save_optim(self.optimizer_D, 'D', which_epoch) + '''if self.gen_features: + self.save_network(self.netE, 'E', which_epoch, self.gpu_ids)''' + + def update_fixed_params(self): + # after fixing the global generator for a number of iterations, also start finetuning it + params = list(self.netG.parameters()) + if self.gen_features: + params += list(self.netE.parameters()) + self.optimizer_G = torch.optim.Adam(params, lr=self.opt.lr, betas=(self.opt.beta1, 0.999)) + if self.opt.verbose: + print('------------ Now also finetuning global generator -----------') + + def update_learning_rate(self): + lrd = self.opt.lr / self.opt.niter_decay + lr = self.old_lr - lrd + for param_group in self.optimizer_D.param_groups: + param_group['lr'] = lr + for param_group in self.optimizer_G.param_groups: + param_group['lr'] = lr + if self.opt.verbose: + print('update learning rate: %f -> %f' % (self.old_lr, lr)) + self.old_lr = lr + + diff --git a/mlflow/registry/Swimswap/models/projectionhead.py b/mlflow/registry/Swimswap/models/projectionhead.py new file mode 100644 index 0000000..a0d18f1 --- /dev/null +++ b/mlflow/registry/Swimswap/models/projectionhead.py @@ -0,0 +1,14 @@ +import torch.nn as nn + +class ProjectionHead(nn.Module): + def __init__(self, proj_dim=256): + super(ProjectionHead, self).__init__() + + self.proj = nn.Sequential( + nn.Linear(proj_dim, proj_dim), + nn.ReLU(), + nn.Linear(proj_dim, proj_dim), + ) + + def forward(self, x): + return self.proj(x) \ No newline at end of file diff --git a/mlflow/registry/Swimswap/models/ui_model.py b/mlflow/registry/Swimswap/models/ui_model.py new file mode 100644 index 0000000..c5b3433 --- /dev/null +++ b/mlflow/registry/Swimswap/models/ui_model.py @@ -0,0 +1,347 @@ +import torch +from torch.autograd import Variable +from collections import OrderedDict +import numpy as np +import os +from PIL import Image +import util.util as util +from .base_model import BaseModel +from . import networks + +class UIModel(BaseModel): + def name(self): + return 'UIModel' + + def initialize(self, opt): + assert(not opt.isTrain) + BaseModel.initialize(self, opt) + self.use_features = opt.instance_feat or opt.label_feat + + netG_input_nc = opt.label_nc + if not opt.no_instance: + netG_input_nc += 1 + if self.use_features: + netG_input_nc += opt.feat_num + + self.netG = networks.define_G(netG_input_nc, opt.output_nc, opt.ngf, opt.netG, + opt.n_downsample_global, opt.n_blocks_global, opt.n_local_enhancers, + opt.n_blocks_local, opt.norm, gpu_ids=self.gpu_ids) + self.load_network(self.netG, 'G', opt.which_epoch) + + print('---------- Networks initialized -------------') + + def toTensor(self, img, normalize=False): + tensor = torch.from_numpy(np.array(img, np.int32, copy=False)) + tensor = tensor.view(1, img.size[1], img.size[0], len(img.mode)) + tensor = tensor.transpose(1, 2).transpose(1, 3).contiguous() + if normalize: + return (tensor.float()/255.0 - 0.5) / 0.5 + return tensor.float() + + def load_image(self, label_path, inst_path, feat_path): + opt = self.opt + # read label map + label_img = Image.open(label_path) + if label_path.find('face') != -1: + label_img = label_img.convert('L') + ow, oh = label_img.size + w = opt.loadSize + h = int(w * oh / ow) + label_img = label_img.resize((w, h), Image.NEAREST) + label_map = self.toTensor(label_img) + + # onehot vector input for label map + self.label_map = label_map.cuda() + oneHot_size = (1, opt.label_nc, h, w) + input_label = self.Tensor(torch.Size(oneHot_size)).zero_() + self.input_label = input_label.scatter_(1, label_map.long().cuda(), 1.0) + + # read instance map + if not opt.no_instance: + inst_img = Image.open(inst_path) + inst_img = inst_img.resize((w, h), Image.NEAREST) + self.inst_map = self.toTensor(inst_img).cuda() + self.edge_map = self.get_edges(self.inst_map) + self.net_input = Variable(torch.cat((self.input_label, self.edge_map), dim=1), volatile=True) + else: + self.net_input = Variable(self.input_label, volatile=True) + + self.features_clustered = np.load(feat_path).item() + self.object_map = self.inst_map if opt.instance_feat else self.label_map + + object_np = self.object_map.cpu().numpy().astype(int) + self.feat_map = self.Tensor(1, opt.feat_num, h, w).zero_() + self.cluster_indices = np.zeros(self.opt.label_nc, np.uint8) + for i in np.unique(object_np): + label = i if i < 1000 else i//1000 + if label in self.features_clustered: + feat = self.features_clustered[label] + np.random.seed(i+1) + cluster_idx = np.random.randint(0, feat.shape[0]) + self.cluster_indices[label] = cluster_idx + idx = (self.object_map == i).nonzero() + self.set_features(idx, feat, cluster_idx) + + self.net_input_original = self.net_input.clone() + self.label_map_original = self.label_map.clone() + self.feat_map_original = self.feat_map.clone() + if not opt.no_instance: + self.inst_map_original = self.inst_map.clone() + + def reset(self): + self.net_input = self.net_input_prev = self.net_input_original.clone() + self.label_map = self.label_map_prev = self.label_map_original.clone() + self.feat_map = self.feat_map_prev = self.feat_map_original.clone() + if not self.opt.no_instance: + self.inst_map = self.inst_map_prev = self.inst_map_original.clone() + self.object_map = self.inst_map if self.opt.instance_feat else self.label_map + + def undo(self): + self.net_input = self.net_input_prev + self.label_map = self.label_map_prev + self.feat_map = self.feat_map_prev + if not self.opt.no_instance: + self.inst_map = self.inst_map_prev + self.object_map = self.inst_map if self.opt.instance_feat else self.label_map + + # get boundary map from instance map + def get_edges(self, t): + edge = torch.cuda.ByteTensor(t.size()).zero_() + edge[:,:,:,1:] = edge[:,:,:,1:] | (t[:,:,:,1:] != t[:,:,:,:-1]) + edge[:,:,:,:-1] = edge[:,:,:,:-1] | (t[:,:,:,1:] != t[:,:,:,:-1]) + edge[:,:,1:,:] = edge[:,:,1:,:] | (t[:,:,1:,:] != t[:,:,:-1,:]) + edge[:,:,:-1,:] = edge[:,:,:-1,:] | (t[:,:,1:,:] != t[:,:,:-1,:]) + return edge.float() + + # change the label at the source position to the label at the target position + def change_labels(self, click_src, click_tgt): + y_src, x_src = click_src[0], click_src[1] + y_tgt, x_tgt = click_tgt[0], click_tgt[1] + label_src = int(self.label_map[0, 0, y_src, x_src]) + inst_src = self.inst_map[0, 0, y_src, x_src] + label_tgt = int(self.label_map[0, 0, y_tgt, x_tgt]) + inst_tgt = self.inst_map[0, 0, y_tgt, x_tgt] + + idx_src = (self.inst_map == inst_src).nonzero() + # need to change 3 things: label map, instance map, and feature map + if idx_src.shape: + # backup current maps + self.backup_current_state() + + # change both the label map and the network input + self.label_map[idx_src[:,0], idx_src[:,1], idx_src[:,2], idx_src[:,3]] = label_tgt + self.net_input[idx_src[:,0], idx_src[:,1] + label_src, idx_src[:,2], idx_src[:,3]] = 0 + self.net_input[idx_src[:,0], idx_src[:,1] + label_tgt, idx_src[:,2], idx_src[:,3]] = 1 + + # update the instance map (and the network input) + if inst_tgt > 1000: + # if different instances have different ids, give the new object a new id + tgt_indices = (self.inst_map > label_tgt * 1000) & (self.inst_map < (label_tgt+1) * 1000) + inst_tgt = self.inst_map[tgt_indices].max() + 1 + self.inst_map[idx_src[:,0], idx_src[:,1], idx_src[:,2], idx_src[:,3]] = inst_tgt + self.net_input[:,-1,:,:] = self.get_edges(self.inst_map) + + # also copy the source features to the target position + idx_tgt = (self.inst_map == inst_tgt).nonzero() + if idx_tgt.shape: + self.copy_features(idx_src, idx_tgt[0,:]) + + self.fake_image = util.tensor2im(self.single_forward(self.net_input, self.feat_map)) + + # add strokes of target label in the image + def add_strokes(self, click_src, label_tgt, bw, save): + # get the region of the new strokes (bw is the brush width) + size = self.net_input.size() + h, w = size[2], size[3] + idx_src = torch.LongTensor(bw**2, 4).fill_(0) + for i in range(bw): + idx_src[i*bw:(i+1)*bw, 2] = min(h-1, max(0, click_src[0]-bw//2 + i)) + for j in range(bw): + idx_src[i*bw+j, 3] = min(w-1, max(0, click_src[1]-bw//2 + j)) + idx_src = idx_src.cuda() + + # again, need to update 3 things + if idx_src.shape: + # backup current maps + if save: + self.backup_current_state() + + # update the label map (and the network input) in the stroke region + self.label_map[idx_src[:,0], idx_src[:,1], idx_src[:,2], idx_src[:,3]] = label_tgt + for k in range(self.opt.label_nc): + self.net_input[idx_src[:,0], idx_src[:,1] + k, idx_src[:,2], idx_src[:,3]] = 0 + self.net_input[idx_src[:,0], idx_src[:,1] + label_tgt, idx_src[:,2], idx_src[:,3]] = 1 + + # update the instance map (and the network input) + self.inst_map[idx_src[:,0], idx_src[:,1], idx_src[:,2], idx_src[:,3]] = label_tgt + self.net_input[:,-1,:,:] = self.get_edges(self.inst_map) + + # also update the features if available + if self.opt.instance_feat: + feat = self.features_clustered[label_tgt] + #np.random.seed(label_tgt+1) + #cluster_idx = np.random.randint(0, feat.shape[0]) + cluster_idx = self.cluster_indices[label_tgt] + self.set_features(idx_src, feat, cluster_idx) + + self.fake_image = util.tensor2im(self.single_forward(self.net_input, self.feat_map)) + + # add an object to the clicked position with selected style + def add_objects(self, click_src, label_tgt, mask, style_id=0): + y, x = click_src[0], click_src[1] + mask = np.transpose(mask, (2, 0, 1))[np.newaxis,...] + idx_src = torch.from_numpy(mask).cuda().nonzero() + idx_src[:,2] += y + idx_src[:,3] += x + + # backup current maps + self.backup_current_state() + + # update label map + self.label_map[idx_src[:,0], idx_src[:,1], idx_src[:,2], idx_src[:,3]] = label_tgt + for k in range(self.opt.label_nc): + self.net_input[idx_src[:,0], idx_src[:,1] + k, idx_src[:,2], idx_src[:,3]] = 0 + self.net_input[idx_src[:,0], idx_src[:,1] + label_tgt, idx_src[:,2], idx_src[:,3]] = 1 + + # update instance map + self.inst_map[idx_src[:,0], idx_src[:,1], idx_src[:,2], idx_src[:,3]] = label_tgt + self.net_input[:,-1,:,:] = self.get_edges(self.inst_map) + + # update feature map + self.set_features(idx_src, self.feat, style_id) + + self.fake_image = util.tensor2im(self.single_forward(self.net_input, self.feat_map)) + + def single_forward(self, net_input, feat_map): + net_input = torch.cat((net_input, feat_map), dim=1) + fake_image = self.netG.forward(net_input) + + if fake_image.size()[0] == 1: + return fake_image.data[0] + return fake_image.data + + + # generate all outputs for different styles + def style_forward(self, click_pt, style_id=-1): + if click_pt is None: + self.fake_image = util.tensor2im(self.single_forward(self.net_input, self.feat_map)) + self.crop = None + self.mask = None + else: + instToChange = int(self.object_map[0, 0, click_pt[0], click_pt[1]]) + self.instToChange = instToChange + label = instToChange if instToChange < 1000 else instToChange//1000 + self.feat = self.features_clustered[label] + self.fake_image = [] + self.mask = self.object_map == instToChange + idx = self.mask.nonzero() + self.get_crop_region(idx) + if idx.size(): + if style_id == -1: + (min_y, min_x, max_y, max_x) = self.crop + ### original + for cluster_idx in range(self.opt.multiple_output): + self.set_features(idx, self.feat, cluster_idx) + fake_image = self.single_forward(self.net_input, self.feat_map) + fake_image = util.tensor2im(fake_image[:,min_y:max_y,min_x:max_x]) + self.fake_image.append(fake_image) + """### To speed up previewing different style results, either crop or downsample the label maps + if instToChange > 1000: + (min_y, min_x, max_y, max_x) = self.crop + ### crop + _, _, h, w = self.net_input.size() + offset = 512 + y_start, x_start = max(0, min_y-offset), max(0, min_x-offset) + y_end, x_end = min(h, (max_y + offset)), min(w, (max_x + offset)) + y_region = slice(y_start, y_start+(y_end-y_start)//16*16) + x_region = slice(x_start, x_start+(x_end-x_start)//16*16) + net_input = self.net_input[:,:,y_region,x_region] + for cluster_idx in range(self.opt.multiple_output): + self.set_features(idx, self.feat, cluster_idx) + fake_image = self.single_forward(net_input, self.feat_map[:,:,y_region,x_region]) + fake_image = util.tensor2im(fake_image[:,min_y-y_start:max_y-y_start,min_x-x_start:max_x-x_start]) + self.fake_image.append(fake_image) + else: + ### downsample + (min_y, min_x, max_y, max_x) = [crop//2 for crop in self.crop] + net_input = self.net_input[:,:,::2,::2] + size = net_input.size() + net_input_batch = net_input.expand(self.opt.multiple_output, size[1], size[2], size[3]) + for cluster_idx in range(self.opt.multiple_output): + self.set_features(idx, self.feat, cluster_idx) + feat_map = self.feat_map[:,:,::2,::2] + if cluster_idx == 0: + feat_map_batch = feat_map + else: + feat_map_batch = torch.cat((feat_map_batch, feat_map), dim=0) + fake_image_batch = self.single_forward(net_input_batch, feat_map_batch) + for i in range(self.opt.multiple_output): + self.fake_image.append(util.tensor2im(fake_image_batch[i,:,min_y:max_y,min_x:max_x]))""" + + else: + self.set_features(idx, self.feat, style_id) + self.cluster_indices[label] = style_id + self.fake_image = util.tensor2im(self.single_forward(self.net_input, self.feat_map)) + + def backup_current_state(self): + self.net_input_prev = self.net_input.clone() + self.label_map_prev = self.label_map.clone() + self.inst_map_prev = self.inst_map.clone() + self.feat_map_prev = self.feat_map.clone() + + # crop the ROI and get the mask of the object + def get_crop_region(self, idx): + size = self.net_input.size() + h, w = size[2], size[3] + min_y, min_x = idx[:,2].min(), idx[:,3].min() + max_y, max_x = idx[:,2].max(), idx[:,3].max() + crop_min = 128 + if max_y - min_y < crop_min: + min_y = max(0, (max_y + min_y) // 2 - crop_min // 2) + max_y = min(h-1, min_y + crop_min) + if max_x - min_x < crop_min: + min_x = max(0, (max_x + min_x) // 2 - crop_min // 2) + max_x = min(w-1, min_x + crop_min) + self.crop = (min_y, min_x, max_y, max_x) + self.mask = self.mask[:,:, min_y:max_y, min_x:max_x] + + # update the feature map once a new object is added or the label is changed + def update_features(self, cluster_idx, mask=None, click_pt=None): + self.feat_map_prev = self.feat_map.clone() + # adding a new object + if mask is not None: + y, x = click_pt[0], click_pt[1] + mask = np.transpose(mask, (2,0,1))[np.newaxis,...] + idx = torch.from_numpy(mask).cuda().nonzero() + idx[:,2] += y + idx[:,3] += x + # changing the label of an existing object + else: + idx = (self.object_map == self.instToChange).nonzero() + + # update feature map + self.set_features(idx, self.feat, cluster_idx) + + # set the class features to the target feature + def set_features(self, idx, feat, cluster_idx): + for k in range(self.opt.feat_num): + self.feat_map[idx[:,0], idx[:,1] + k, idx[:,2], idx[:,3]] = feat[cluster_idx, k] + + # copy the features at the target position to the source position + def copy_features(self, idx_src, idx_tgt): + for k in range(self.opt.feat_num): + val = self.feat_map[idx_tgt[0], idx_tgt[1] + k, idx_tgt[2], idx_tgt[3]] + self.feat_map[idx_src[:,0], idx_src[:,1] + k, idx_src[:,2], idx_src[:,3]] = val + + def get_current_visuals(self, getLabel=False): + mask = self.mask + if self.mask is not None: + mask = np.transpose(self.mask[0].cpu().float().numpy(), (1,2,0)).astype(np.uint8) + + dict_list = [('fake_image', self.fake_image), ('mask', mask)] + + if getLabel: # only output label map if needed to save bandwidth + label = util.tensor2label(self.net_input.data[0], self.opt.label_nc) + dict_list += [('label', label)] + + return OrderedDict(dict_list) \ No newline at end of file diff --git a/mlflow/registry/Swimswap/options/base_options.py b/mlflow/registry/Swimswap/options/base_options.py new file mode 100644 index 0000000..f5e5e8c --- /dev/null +++ b/mlflow/registry/Swimswap/options/base_options.py @@ -0,0 +1,104 @@ +import argparse +import os +from util import util +import torch + +class BaseOptions(): + def __init__(self): + self.parser = argparse.ArgumentParser() + self.initialized = False + + def initialize(self): + # experiment specifics + self.parser.add_argument('--name', type=str, default='people', help='name of the experiment. It decides where to store samples and models') + self.parser.add_argument('--gpu_ids', type=str, default='0', help='gpu ids: e.g. 0 0,1,2, 0,2. use -1 for CPU') + self.parser.add_argument('--checkpoints_dir', type=str, default='./checkpoints', help='models are saved here') + # self.parser.add_argument('--model', type=str, default='pix2pixHD', help='which model to use') + self.parser.add_argument('--norm', type=str, default='batch', help='instance normalization or batch normalization') + self.parser.add_argument('--use_dropout', action='store_true', help='use dropout for the generator') + self.parser.add_argument('--data_type', default=32, type=int, choices=[8, 16, 32], help="Supported data type i.e. 8, 16, 32 bit") + self.parser.add_argument('--verbose', action='store_true', default=False, help='toggles verbose') + self.parser.add_argument('--fp16', action='store_true', default=False, help='train with AMP') + self.parser.add_argument('--local_rank', type=int, default=0, help='local rank for distributed training') + self.parser.add_argument('--isTrain', type=bool, default=True, help='local rank for distributed training') + + # input/output sizes + self.parser.add_argument('--batchSize', type=int, default=8, help='input batch size') + self.parser.add_argument('--loadSize', type=int, default=1024, help='scale images to this size') + self.parser.add_argument('--fineSize', type=int, default=512, help='then crop to this size') + self.parser.add_argument('--label_nc', type=int, default=0, help='# of input label channels') + self.parser.add_argument('--input_nc', type=int, default=3, help='# of input image channels') + self.parser.add_argument('--output_nc', type=int, default=3, help='# of output image channels') + + # for setting inputs + self.parser.add_argument('--dataroot', type=str, default='./SimSwap/datasets/cityscapes/') + self.parser.add_argument('--resize_or_crop', type=str, default='scale_width', help='scaling and cropping of images at load time [resize_and_crop|crop|scale_width|scale_width_and_crop]') + self.parser.add_argument('--serial_batches', action='store_true', help='if true, takes images in order to make batches, otherwise takes them randomly') + self.parser.add_argument('--no_flip', action='store_true', help='if specified, do not flip the images for data argumentation') + self.parser.add_argument('--nThreads', default=2, type=int, help='# threads for loading data') + self.parser.add_argument('--max_dataset_size', type=int, default=float("inf"), help='Maximum number of samples allowed per dataset. If the dataset directory contains more than max_dataset_size, only a subset is loaded.') + + # for displays + self.parser.add_argument('--display_winsize', type=int, default=512, help='display window size') + self.parser.add_argument('--tf_log', action='store_true', help='if specified, use tensorboard logging. Requires tensorflow installed') + + # for generator + self.parser.add_argument('--netG', type=str, default='global', help='selects model to use for netG') + self.parser.add_argument('--latent_size', type=int, default=512, help='latent size of Adain layer') + self.parser.add_argument('--ngf', type=int, default=64, help='# of gen filters in first conv layer') + self.parser.add_argument('--n_downsample_global', type=int, default=3, help='number of downsampling layers in netG') + self.parser.add_argument('--n_blocks_global', type=int, default=6, help='number of residual blocks in the global generator network') + self.parser.add_argument('--n_blocks_local', type=int, default=3, help='number of residual blocks in the local enhancer network') + self.parser.add_argument('--n_local_enhancers', type=int, default=1, help='number of local enhancers to use') + self.parser.add_argument('--niter_fix_global', type=int, default=0, help='number of epochs that we only train the outmost local enhancer') + + # for instance-wise features + self.parser.add_argument('--no_instance', action='store_true', help='if specified, do *not* add instance map as input') + self.parser.add_argument('--instance_feat', action='store_true', help='if specified, add encoded instance features as input') + self.parser.add_argument('--label_feat', action='store_true', help='if specified, add encoded label features as input') + self.parser.add_argument('--feat_num', type=int, default=3, help='vector length for encoded features') + self.parser.add_argument('--load_features', action='store_true', help='if specified, load precomputed feature maps') + self.parser.add_argument('--n_downsample_E', type=int, default=4, help='# of downsampling layers in encoder') + self.parser.add_argument('--nef', type=int, default=16, help='# of encoder filters in the first conv layer') + self.parser.add_argument('--n_clusters', type=int, default=10, help='number of clusters for features') + self.parser.add_argument('--image_size', type=int, default=224, help='number of clusters for features') + self.parser.add_argument('--norm_G', type=str, default='spectralspadesyncbatch3x3', help='instance normalization or batch normalization') + self.parser.add_argument('--semantic_nc', type=int, default=3, help='number of clusters for features') + self.initialized = True + + def parse(self, save=True): + if not self.initialized: + self.initialize() + self.opt = self.parser.parse_args() + self.opt.isTrain = self.isTrain # train or test + + str_ids = self.opt.gpu_ids.split(',') + self.opt.gpu_ids = [] + for str_id in str_ids: + id = int(str_id) + if id >= 0: + self.opt.gpu_ids.append(id) + + # set gpu ids + if len(self.opt.gpu_ids) > 0: + torch.cuda.set_device(self.opt.gpu_ids[0]) + + args = vars(self.opt) + + print('------------ Options -------------') + for k, v in sorted(args.items()): + print('%s: %s' % (str(k), str(v))) + print('-------------- End ----------------') + + # save to the disk + if self.opt.isTrain: + expr_dir = os.path.join(self.opt.checkpoints_dir, self.opt.name) + util.mkdirs(expr_dir) + if save and not self.opt.continue_train: + file_name = os.path.join(expr_dir, 'opt.txt') + with open(file_name, 'wt') as opt_file: + opt_file.write('------------ Options -------------\n') + for k, v in sorted(args.items()): + opt_file.write('%s: %s\n' % (str(k), str(v))) + opt_file.write('-------------- End ----------------\n') + return self.opt diff --git a/mlflow/registry/Swimswap/options/test_options.py b/mlflow/registry/Swimswap/options/test_options.py new file mode 100644 index 0000000..e46e435 --- /dev/null +++ b/mlflow/registry/Swimswap/options/test_options.py @@ -0,0 +1,38 @@ +''' +Author: Naiyuan liu +Github: https://github.com/NNNNAI +Date: 2021-11-23 17:03:58 +LastEditors: Naiyuan liu +LastEditTime: 2021-11-23 17:08:08 +Description: +''' +from .base_options import BaseOptions + +class TestOptions(BaseOptions): + def initialize(self): + BaseOptions.initialize(self) + self.parser.add_argument('--ntest', type=int, default=float("inf"), help='# of test examples.') + self.parser.add_argument('--results_dir', type=str, default='./results/', help='saves results here.') + self.parser.add_argument('--aspect_ratio', type=float, default=1.0, help='aspect ratio of result images') + self.parser.add_argument('--phase', type=str, default='test', help='train, val, test, etc') + self.parser.add_argument('--which_epoch', type=str, default='latest', help='which epoch to load? set to latest to use latest cached model') + self.parser.add_argument('--how_many', type=int, default=50, help='how many test images to run') + self.parser.add_argument('--cluster_path', type=str, default='features_clustered_010.npy', help='the path for clustered results of encoded features') + self.parser.add_argument('--use_encoded_image', action='store_true', help='if specified, encode the real image to get the feature map') + self.parser.add_argument("--export_onnx", type=str, help="export ONNX model to a given file") + self.parser.add_argument("--engine", type=str, help="run serialized TRT engine") + self.parser.add_argument("--onnx", type=str, help="run ONNX model via TRT") + self.parser.add_argument("--Arc_path", type=str, default='SimSwap/arcface_model/arcface_checkpoint.tar', help="run ONNX model via TRT") + self.parser.add_argument("--pic_a_path", type=str, default='G:/swap_data/ID/elon-musk-hero-image.jpeg', help="Person who provides identity information") + self.parser.add_argument("--pic_b_path", type=str, default='./demo_file/multi_people.jpg', help="Person who provides information other than their identity") + self.parser.add_argument("--pic_specific_path", type=str, default='./crop_224/zrf.jpg', help="The specific person to be swapped") + self.parser.add_argument("--multisepcific_dir", type=str, default='./demo_file/multispecific', help="Dir for multi specific") + self.parser.add_argument("--video_path", type=str, default='G:/swap_data/video/HSB_Demo_Trim.mp4', help="path for the video to swap") + self.parser.add_argument("--temp_path", type=str, default='./temp_results', help="path to save temporarily images") + self.parser.add_argument("--output_path", type=str, default='./output/', help="results path") + self.parser.add_argument('--id_thres', type=float, default=0.03, help='how many test images to run') + self.parser.add_argument('--no_simswaplogo', action='store_true', help='Remove the watermark') + self.parser.add_argument('--use_mask', action='store_true', help='Use mask for better result') + self.parser.add_argument('--crop_size', type=int, default=224, help='Crop of size of input image') + + self.isTrain = False \ No newline at end of file diff --git a/mlflow/registry/Swimswap/parsing_model/model.py b/mlflow/registry/Swimswap/parsing_model/model.py new file mode 100644 index 0000000..d434836 --- /dev/null +++ b/mlflow/registry/Swimswap/parsing_model/model.py @@ -0,0 +1,283 @@ +#!/usr/bin/python +# -*- encoding: utf-8 -*- + + +import torch +import torch.nn as nn +import torch.nn.functional as F +import torchvision + +from parsing_model.resnet import Resnet18 +# from modules.bn import InPlaceABNSync as BatchNorm2d + + +class ConvBNReLU(nn.Module): + def __init__(self, in_chan, out_chan, ks=3, stride=1, padding=1, *args, **kwargs): + super(ConvBNReLU, self).__init__() + self.conv = nn.Conv2d(in_chan, + out_chan, + kernel_size = ks, + stride = stride, + padding = padding, + bias = False) + self.bn = nn.BatchNorm2d(out_chan) + self.init_weight() + + def forward(self, x): + x = self.conv(x) + x = F.relu(self.bn(x)) + return x + + def init_weight(self): + for ly in self.children(): + if isinstance(ly, nn.Conv2d): + nn.init.kaiming_normal_(ly.weight, a=1) + if not ly.bias is None: nn.init.constant_(ly.bias, 0) + +class BiSeNetOutput(nn.Module): + def __init__(self, in_chan, mid_chan, n_classes, *args, **kwargs): + super(BiSeNetOutput, self).__init__() + self.conv = ConvBNReLU(in_chan, mid_chan, ks=3, stride=1, padding=1) + self.conv_out = nn.Conv2d(mid_chan, n_classes, kernel_size=1, bias=False) + self.init_weight() + + def forward(self, x): + x = self.conv(x) + x = self.conv_out(x) + return x + + def init_weight(self): + for ly in self.children(): + if isinstance(ly, nn.Conv2d): + nn.init.kaiming_normal_(ly.weight, a=1) + if not ly.bias is None: nn.init.constant_(ly.bias, 0) + + def get_params(self): + wd_params, nowd_params = [], [] + for name, module in self.named_modules(): + if isinstance(module, nn.Linear) or isinstance(module, nn.Conv2d): + wd_params.append(module.weight) + if not module.bias is None: + nowd_params.append(module.bias) + elif isinstance(module, nn.BatchNorm2d): + nowd_params += list(module.parameters()) + return wd_params, nowd_params + + +class AttentionRefinementModule(nn.Module): + def __init__(self, in_chan, out_chan, *args, **kwargs): + super(AttentionRefinementModule, self).__init__() + self.conv = ConvBNReLU(in_chan, out_chan, ks=3, stride=1, padding=1) + self.conv_atten = nn.Conv2d(out_chan, out_chan, kernel_size= 1, bias=False) + self.bn_atten = nn.BatchNorm2d(out_chan) + self.sigmoid_atten = nn.Sigmoid() + self.init_weight() + + def forward(self, x): + feat = self.conv(x) + atten = F.avg_pool2d(feat, feat.size()[2:]) + atten = self.conv_atten(atten) + atten = self.bn_atten(atten) + atten = self.sigmoid_atten(atten) + out = torch.mul(feat, atten) + return out + + def init_weight(self): + for ly in self.children(): + if isinstance(ly, nn.Conv2d): + nn.init.kaiming_normal_(ly.weight, a=1) + if not ly.bias is None: nn.init.constant_(ly.bias, 0) + + +class ContextPath(nn.Module): + def __init__(self, *args, **kwargs): + super(ContextPath, self).__init__() + self.resnet = Resnet18() + self.arm16 = AttentionRefinementModule(256, 128) + self.arm32 = AttentionRefinementModule(512, 128) + self.conv_head32 = ConvBNReLU(128, 128, ks=3, stride=1, padding=1) + self.conv_head16 = ConvBNReLU(128, 128, ks=3, stride=1, padding=1) + self.conv_avg = ConvBNReLU(512, 128, ks=1, stride=1, padding=0) + + self.init_weight() + + def forward(self, x): + H0, W0 = x.size()[2:] + feat8, feat16, feat32 = self.resnet(x) + H8, W8 = feat8.size()[2:] + H16, W16 = feat16.size()[2:] + H32, W32 = feat32.size()[2:] + + avg = F.avg_pool2d(feat32, feat32.size()[2:]) + avg = self.conv_avg(avg) + avg_up = F.interpolate(avg, (H32, W32), mode='nearest') + + feat32_arm = self.arm32(feat32) + feat32_sum = feat32_arm + avg_up + feat32_up = F.interpolate(feat32_sum, (H16, W16), mode='nearest') + feat32_up = self.conv_head32(feat32_up) + + feat16_arm = self.arm16(feat16) + feat16_sum = feat16_arm + feat32_up + feat16_up = F.interpolate(feat16_sum, (H8, W8), mode='nearest') + feat16_up = self.conv_head16(feat16_up) + + return feat8, feat16_up, feat32_up # x8, x8, x16 + + def init_weight(self): + for ly in self.children(): + if isinstance(ly, nn.Conv2d): + nn.init.kaiming_normal_(ly.weight, a=1) + if not ly.bias is None: nn.init.constant_(ly.bias, 0) + + def get_params(self): + wd_params, nowd_params = [], [] + for name, module in self.named_modules(): + if isinstance(module, (nn.Linear, nn.Conv2d)): + wd_params.append(module.weight) + if not module.bias is None: + nowd_params.append(module.bias) + elif isinstance(module, nn.BatchNorm2d): + nowd_params += list(module.parameters()) + return wd_params, nowd_params + + +### This is not used, since I replace this with the resnet feature with the same size +class SpatialPath(nn.Module): + def __init__(self, *args, **kwargs): + super(SpatialPath, self).__init__() + self.conv1 = ConvBNReLU(3, 64, ks=7, stride=2, padding=3) + self.conv2 = ConvBNReLU(64, 64, ks=3, stride=2, padding=1) + self.conv3 = ConvBNReLU(64, 64, ks=3, stride=2, padding=1) + self.conv_out = ConvBNReLU(64, 128, ks=1, stride=1, padding=0) + self.init_weight() + + def forward(self, x): + feat = self.conv1(x) + feat = self.conv2(feat) + feat = self.conv3(feat) + feat = self.conv_out(feat) + return feat + + def init_weight(self): + for ly in self.children(): + if isinstance(ly, nn.Conv2d): + nn.init.kaiming_normal_(ly.weight, a=1) + if not ly.bias is None: nn.init.constant_(ly.bias, 0) + + def get_params(self): + wd_params, nowd_params = [], [] + for name, module in self.named_modules(): + if isinstance(module, nn.Linear) or isinstance(module, nn.Conv2d): + wd_params.append(module.weight) + if not module.bias is None: + nowd_params.append(module.bias) + elif isinstance(module, nn.BatchNorm2d): + nowd_params += list(module.parameters()) + return wd_params, nowd_params + + +class FeatureFusionModule(nn.Module): + def __init__(self, in_chan, out_chan, *args, **kwargs): + super(FeatureFusionModule, self).__init__() + self.convblk = ConvBNReLU(in_chan, out_chan, ks=1, stride=1, padding=0) + self.conv1 = nn.Conv2d(out_chan, + out_chan//4, + kernel_size = 1, + stride = 1, + padding = 0, + bias = False) + self.conv2 = nn.Conv2d(out_chan//4, + out_chan, + kernel_size = 1, + stride = 1, + padding = 0, + bias = False) + self.relu = nn.ReLU(inplace=True) + self.sigmoid = nn.Sigmoid() + self.init_weight() + + def forward(self, fsp, fcp): + fcat = torch.cat([fsp, fcp], dim=1) + feat = self.convblk(fcat) + atten = F.avg_pool2d(feat, feat.size()[2:]) + atten = self.conv1(atten) + atten = self.relu(atten) + atten = self.conv2(atten) + atten = self.sigmoid(atten) + feat_atten = torch.mul(feat, atten) + feat_out = feat_atten + feat + return feat_out + + def init_weight(self): + for ly in self.children(): + if isinstance(ly, nn.Conv2d): + nn.init.kaiming_normal_(ly.weight, a=1) + if not ly.bias is None: nn.init.constant_(ly.bias, 0) + + def get_params(self): + wd_params, nowd_params = [], [] + for name, module in self.named_modules(): + if isinstance(module, nn.Linear) or isinstance(module, nn.Conv2d): + wd_params.append(module.weight) + if not module.bias is None: + nowd_params.append(module.bias) + elif isinstance(module, nn.BatchNorm2d): + nowd_params += list(module.parameters()) + return wd_params, nowd_params + + +class BiSeNet(nn.Module): + def __init__(self, n_classes, *args, **kwargs): + super(BiSeNet, self).__init__() + self.cp = ContextPath() + ## here self.sp is deleted + self.ffm = FeatureFusionModule(256, 256) + self.conv_out = BiSeNetOutput(256, 256, n_classes) + self.conv_out16 = BiSeNetOutput(128, 64, n_classes) + self.conv_out32 = BiSeNetOutput(128, 64, n_classes) + self.init_weight() + + def forward(self, x): + H, W = x.size()[2:] + feat_res8, feat_cp8, feat_cp16 = self.cp(x) # here return res3b1 feature + feat_sp = feat_res8 # use res3b1 feature to replace spatial path feature + feat_fuse = self.ffm(feat_sp, feat_cp8) + + feat_out = self.conv_out(feat_fuse) + feat_out16 = self.conv_out16(feat_cp8) + feat_out32 = self.conv_out32(feat_cp16) + + feat_out = F.interpolate(feat_out, (H, W), mode='bilinear', align_corners=True) + feat_out16 = F.interpolate(feat_out16, (H, W), mode='bilinear', align_corners=True) + feat_out32 = F.interpolate(feat_out32, (H, W), mode='bilinear', align_corners=True) + return feat_out, feat_out16, feat_out32 + + def init_weight(self): + for ly in self.children(): + if isinstance(ly, nn.Conv2d): + nn.init.kaiming_normal_(ly.weight, a=1) + if not ly.bias is None: nn.init.constant_(ly.bias, 0) + + def get_params(self): + wd_params, nowd_params, lr_mul_wd_params, lr_mul_nowd_params = [], [], [], [] + for name, child in self.named_children(): + child_wd_params, child_nowd_params = child.get_params() + if isinstance(child, FeatureFusionModule) or isinstance(child, BiSeNetOutput): + lr_mul_wd_params += child_wd_params + lr_mul_nowd_params += child_nowd_params + else: + wd_params += child_wd_params + nowd_params += child_nowd_params + return wd_params, nowd_params, lr_mul_wd_params, lr_mul_nowd_params + + +if __name__ == "__main__": + net = BiSeNet(19) + net.cuda() + net.eval() + in_ten = torch.randn(16, 3, 640, 480).cuda() + out, out16, out32 = net(in_ten) + print(out.shape) + + net.get_params() diff --git a/mlflow/registry/Swimswap/parsing_model/resnet.py b/mlflow/registry/Swimswap/parsing_model/resnet.py new file mode 100644 index 0000000..aa2bf95 --- /dev/null +++ b/mlflow/registry/Swimswap/parsing_model/resnet.py @@ -0,0 +1,109 @@ +#!/usr/bin/python +# -*- encoding: utf-8 -*- + +import torch +import torch.nn as nn +import torch.nn.functional as F +import torch.utils.model_zoo as modelzoo + +# from modules.bn import InPlaceABNSync as BatchNorm2d + +resnet18_url = 'https://download.pytorch.org/models/resnet18-5c106cde.pth' + + +def conv3x3(in_planes, out_planes, stride=1): + """3x3 convolution with padding""" + return nn.Conv2d(in_planes, out_planes, kernel_size=3, stride=stride, + padding=1, bias=False) + + +class BasicBlock(nn.Module): + def __init__(self, in_chan, out_chan, stride=1): + super(BasicBlock, self).__init__() + self.conv1 = conv3x3(in_chan, out_chan, stride) + self.bn1 = nn.BatchNorm2d(out_chan) + self.conv2 = conv3x3(out_chan, out_chan) + self.bn2 = nn.BatchNorm2d(out_chan) + self.relu = nn.ReLU(inplace=True) + self.downsample = None + if in_chan != out_chan or stride != 1: + self.downsample = nn.Sequential( + nn.Conv2d(in_chan, out_chan, + kernel_size=1, stride=stride, bias=False), + nn.BatchNorm2d(out_chan), + ) + + def forward(self, x): + residual = self.conv1(x) + residual = F.relu(self.bn1(residual)) + residual = self.conv2(residual) + residual = self.bn2(residual) + + shortcut = x + if self.downsample is not None: + shortcut = self.downsample(x) + + out = shortcut + residual + out = self.relu(out) + return out + + +def create_layer_basic(in_chan, out_chan, bnum, stride=1): + layers = [BasicBlock(in_chan, out_chan, stride=stride)] + for i in range(bnum-1): + layers.append(BasicBlock(out_chan, out_chan, stride=1)) + return nn.Sequential(*layers) + + +class Resnet18(nn.Module): + def __init__(self): + super(Resnet18, self).__init__() + self.conv1 = nn.Conv2d(3, 64, kernel_size=7, stride=2, padding=3, + bias=False) + self.bn1 = nn.BatchNorm2d(64) + self.maxpool = nn.MaxPool2d(kernel_size=3, stride=2, padding=1) + self.layer1 = create_layer_basic(64, 64, bnum=2, stride=1) + self.layer2 = create_layer_basic(64, 128, bnum=2, stride=2) + self.layer3 = create_layer_basic(128, 256, bnum=2, stride=2) + self.layer4 = create_layer_basic(256, 512, bnum=2, stride=2) + self.init_weight() + + def forward(self, x): + x = self.conv1(x) + x = F.relu(self.bn1(x)) + x = self.maxpool(x) + + x = self.layer1(x) + feat8 = self.layer2(x) # 1/8 + feat16 = self.layer3(feat8) # 1/16 + feat32 = self.layer4(feat16) # 1/32 + return feat8, feat16, feat32 + + def init_weight(self): + state_dict = modelzoo.load_url(resnet18_url) + self_state_dict = self.state_dict() + for k, v in state_dict.items(): + if 'fc' in k: continue + self_state_dict.update({k: v}) + self.load_state_dict(self_state_dict) + + def get_params(self): + wd_params, nowd_params = [], [] + for name, module in self.named_modules(): + if isinstance(module, (nn.Linear, nn.Conv2d)): + wd_params.append(module.weight) + if not module.bias is None: + nowd_params.append(module.bias) + elif isinstance(module, nn.BatchNorm2d): + nowd_params += list(module.parameters()) + return wd_params, nowd_params + + +if __name__ == "__main__": + net = Resnet18() + x = torch.randn(16, 3, 224, 224) + out = net(x) + print(out[0].size()) + print(out[1].size()) + print(out[2].size()) + net.get_params() diff --git a/mlflow/registry/Swimswap/pg_modules/blocks.py b/mlflow/registry/Swimswap/pg_modules/blocks.py new file mode 100644 index 0000000..78bd113 --- /dev/null +++ b/mlflow/registry/Swimswap/pg_modules/blocks.py @@ -0,0 +1,325 @@ +import functools +import torch +import torch.nn as nn +import torch.nn.functional as F +from torch.nn.utils import spectral_norm + + +### single layers + + +def conv2d(*args, **kwargs): + return spectral_norm(nn.Conv2d(*args, **kwargs)) + + +def convTranspose2d(*args, **kwargs): + return spectral_norm(nn.ConvTranspose2d(*args, **kwargs)) + + +def embedding(*args, **kwargs): + return spectral_norm(nn.Embedding(*args, **kwargs)) + + +def linear(*args, **kwargs): + return spectral_norm(nn.Linear(*args, **kwargs)) + + +def NormLayer(c, mode='batch'): + if mode == 'group': + return nn.GroupNorm(c//2, c) + elif mode == 'batch': + return nn.BatchNorm2d(c) + + +### Activations + + +class GLU(nn.Module): + def forward(self, x): + nc = x.size(1) + assert nc % 2 == 0, 'channels dont divide 2!' + nc = int(nc/2) + return x[:, :nc] * torch.sigmoid(x[:, nc:]) + + +class Swish(nn.Module): + def forward(self, feat): + return feat * torch.sigmoid(feat) + + +### Upblocks + + +class InitLayer(nn.Module): + def __init__(self, nz, channel, sz=4): + super().__init__() + + self.init = nn.Sequential( + convTranspose2d(nz, channel*2, sz, 1, 0, bias=False), + NormLayer(channel*2), + GLU(), + ) + + def forward(self, noise): + noise = noise.view(noise.shape[0], -1, 1, 1) + return self.init(noise) + + +def UpBlockSmall(in_planes, out_planes): + block = nn.Sequential( + nn.Upsample(scale_factor=2, mode='nearest'), + conv2d(in_planes, out_planes*2, 3, 1, 1, bias=False), + NormLayer(out_planes*2), GLU()) + return block + + +class UpBlockSmallCond(nn.Module): + def __init__(self, in_planes, out_planes, z_dim): + super().__init__() + self.in_planes = in_planes + self.out_planes = out_planes + self.up = nn.Upsample(scale_factor=2, mode='nearest') + self.conv = conv2d(in_planes, out_planes*2, 3, 1, 1, bias=False) + + which_bn = functools.partial(CCBN, which_linear=linear, input_size=z_dim) + self.bn = which_bn(2*out_planes) + self.act = GLU() + + def forward(self, x, c): + x = self.up(x) + x = self.conv(x) + x = self.bn(x, c) + x = self.act(x) + return x + + +def UpBlockBig(in_planes, out_planes): + block = nn.Sequential( + nn.Upsample(scale_factor=2, mode='nearest'), + conv2d(in_planes, out_planes*2, 3, 1, 1, bias=False), + NoiseInjection(), + NormLayer(out_planes*2), GLU(), + conv2d(out_planes, out_planes*2, 3, 1, 1, bias=False), + NoiseInjection(), + NormLayer(out_planes*2), GLU() + ) + return block + + +class UpBlockBigCond(nn.Module): + def __init__(self, in_planes, out_planes, z_dim): + super().__init__() + self.in_planes = in_planes + self.out_planes = out_planes + self.up = nn.Upsample(scale_factor=2, mode='nearest') + self.conv1 = conv2d(in_planes, out_planes*2, 3, 1, 1, bias=False) + self.conv2 = conv2d(out_planes, out_planes*2, 3, 1, 1, bias=False) + + which_bn = functools.partial(CCBN, which_linear=linear, input_size=z_dim) + self.bn1 = which_bn(2*out_planes) + self.bn2 = which_bn(2*out_planes) + self.act = GLU() + self.noise = NoiseInjection() + + def forward(self, x, c): + # block 1 + x = self.up(x) + x = self.conv1(x) + x = self.noise(x) + x = self.bn1(x, c) + x = self.act(x) + + # block 2 + x = self.conv2(x) + x = self.noise(x) + x = self.bn2(x, c) + x = self.act(x) + + return x + + +class SEBlock(nn.Module): + def __init__(self, ch_in, ch_out): + super().__init__() + self.main = nn.Sequential( + nn.AdaptiveAvgPool2d(4), + conv2d(ch_in, ch_out, 4, 1, 0, bias=False), + Swish(), + conv2d(ch_out, ch_out, 1, 1, 0, bias=False), + nn.Sigmoid(), + ) + + def forward(self, feat_small, feat_big): + return feat_big * self.main(feat_small) + + +### Downblocks + + +class SeparableConv2d(nn.Module): + def __init__(self, in_channels, out_channels, kernel_size, bias=False): + super(SeparableConv2d, self).__init__() + self.depthwise = conv2d(in_channels, in_channels, kernel_size=kernel_size, + groups=in_channels, bias=bias, padding=1) + self.pointwise = conv2d(in_channels, out_channels, + kernel_size=1, bias=bias) + + def forward(self, x): + out = self.depthwise(x) + out = self.pointwise(out) + return out + + +class DownBlock(nn.Module): + def __init__(self, in_planes, out_planes, separable=False): + super().__init__() + if not separable: + self.main = nn.Sequential( + conv2d(in_planes, out_planes, 4, 2, 1), + NormLayer(out_planes), + nn.LeakyReLU(0.2, inplace=True), + ) + else: + self.main = nn.Sequential( + SeparableConv2d(in_planes, out_planes, 3), + NormLayer(out_planes), + nn.LeakyReLU(0.2, inplace=True), + nn.AvgPool2d(2, 2), + ) + + def forward(self, feat): + return self.main(feat) + + +class DownBlockPatch(nn.Module): + def __init__(self, in_planes, out_planes, separable=False): + super().__init__() + self.main = nn.Sequential( + DownBlock(in_planes, out_planes, separable), + conv2d(out_planes, out_planes, 1, 1, 0, bias=False), + NormLayer(out_planes), + nn.LeakyReLU(0.2, inplace=True), + ) + + def forward(self, feat): + return self.main(feat) + + +### CSM + + +class ResidualConvUnit(nn.Module): + def __init__(self, cin, activation, bn): + super().__init__() + self.conv = nn.Conv2d(cin, cin, kernel_size=3, stride=1, padding=1, bias=True) + self.skip_add = nn.quantized.FloatFunctional() + + def forward(self, x): + return self.skip_add.add(self.conv(x), x) + + +class FeatureFusionBlock(nn.Module): + def __init__(self, features, activation, deconv=False, bn=False, expand=False, align_corners=True, lowest=False): + super().__init__() + + self.deconv = deconv + self.align_corners = align_corners + + self.expand = expand + out_features = features + if self.expand==True: + out_features = features//2 + + self.out_conv = nn.Conv2d(features, out_features, kernel_size=1, stride=1, padding=0, bias=True, groups=1) + self.skip_add = nn.quantized.FloatFunctional() + + def forward(self, *xs): + output = xs[0] + + if len(xs) == 2: + output = self.skip_add.add(output, xs[1]) + + output = nn.functional.interpolate( + output, scale_factor=2, mode="bilinear", align_corners=self.align_corners + ) + + output = self.out_conv(output) + + return output + + +### Misc + + +class NoiseInjection(nn.Module): + def __init__(self): + super().__init__() + self.weight = nn.Parameter(torch.zeros(1), requires_grad=True) + + def forward(self, feat, noise=None): + if noise is None: + batch, _, height, width = feat.shape + noise = torch.randn(batch, 1, height, width).to(feat.device) + + return feat + self.weight * noise + + +class CCBN(nn.Module): + ''' conditional batchnorm ''' + def __init__(self, output_size, input_size, which_linear, eps=1e-5, momentum=0.1): + super().__init__() + self.output_size, self.input_size = output_size, input_size + + # Prepare gain and bias layers + self.gain = which_linear(input_size, output_size) + self.bias = which_linear(input_size, output_size) + + # epsilon to avoid dividing by 0 + self.eps = eps + # Momentum + self.momentum = momentum + + self.register_buffer('stored_mean', torch.zeros(output_size)) + self.register_buffer('stored_var', torch.ones(output_size)) + + def forward(self, x, y): + # Calculate class-conditional gains and biases + gain = (1 + self.gain(y)).view(y.size(0), -1, 1, 1) + bias = self.bias(y).view(y.size(0), -1, 1, 1) + out = F.batch_norm(x, self.stored_mean, self.stored_var, None, None, + self.training, 0.1, self.eps) + return out * gain + bias + + +class Interpolate(nn.Module): + """Interpolation module.""" + + def __init__(self, size, mode='bilinear', align_corners=False): + """Init. + Args: + scale_factor (float): scaling + mode (str): interpolation mode + """ + super(Interpolate, self).__init__() + + self.interp = nn.functional.interpolate + self.size = size + self.mode = mode + self.align_corners = align_corners + + def forward(self, x): + """Forward pass. + Args: + x (tensor): input + Returns: + tensor: interpolated data + """ + + x = self.interp( + x, + size=self.size, + mode=self.mode, + align_corners=self.align_corners, + ) + + return x diff --git a/mlflow/registry/Swimswap/pg_modules/diffaug.py b/mlflow/registry/Swimswap/pg_modules/diffaug.py new file mode 100644 index 0000000..54020be --- /dev/null +++ b/mlflow/registry/Swimswap/pg_modules/diffaug.py @@ -0,0 +1,76 @@ +# Differentiable Augmentation for Data-Efficient GAN Training +# Shengyu Zhao, Zhijian Liu, Ji Lin, Jun-Yan Zhu, and Song Han +# https://arxiv.org/pdf/2006.10738 + +import torch +import torch.nn.functional as F + + +def DiffAugment(x, policy='', channels_first=True): + if policy: + if not channels_first: + x = x.permute(0, 3, 1, 2) + for p in policy.split(','): + for f in AUGMENT_FNS[p]: + x = f(x) + if not channels_first: + x = x.permute(0, 2, 3, 1) + x = x.contiguous() + return x + + +def rand_brightness(x): + x = x + (torch.rand(x.size(0), 1, 1, 1, dtype=x.dtype, device=x.device) - 0.5) + return x + + +def rand_saturation(x): + x_mean = x.mean(dim=1, keepdim=True) + x = (x - x_mean) * (torch.rand(x.size(0), 1, 1, 1, dtype=x.dtype, device=x.device) * 2) + x_mean + return x + + +def rand_contrast(x): + x_mean = x.mean(dim=[1, 2, 3], keepdim=True) + x = (x - x_mean) * (torch.rand(x.size(0), 1, 1, 1, dtype=x.dtype, device=x.device) + 0.5) + x_mean + return x + + +def rand_translation(x, ratio=0.125): + shift_x, shift_y = int(x.size(2) * ratio + 0.5), int(x.size(3) * ratio + 0.5) + translation_x = torch.randint(-shift_x, shift_x + 1, size=[x.size(0), 1, 1], device=x.device) + translation_y = torch.randint(-shift_y, shift_y + 1, size=[x.size(0), 1, 1], device=x.device) + grid_batch, grid_x, grid_y = torch.meshgrid( + torch.arange(x.size(0), dtype=torch.long, device=x.device), + torch.arange(x.size(2), dtype=torch.long, device=x.device), + torch.arange(x.size(3), dtype=torch.long, device=x.device), + ) + grid_x = torch.clamp(grid_x + translation_x + 1, 0, x.size(2) + 1) + grid_y = torch.clamp(grid_y + translation_y + 1, 0, x.size(3) + 1) + x_pad = F.pad(x, [1, 1, 1, 1, 0, 0, 0, 0]) + x = x_pad.permute(0, 2, 3, 1).contiguous()[grid_batch, grid_x, grid_y].permute(0, 3, 1, 2) + return x + + +def rand_cutout(x, ratio=0.2): + cutout_size = int(x.size(2) * ratio + 0.5), int(x.size(3) * ratio + 0.5) + offset_x = torch.randint(0, x.size(2) + (1 - cutout_size[0] % 2), size=[x.size(0), 1, 1], device=x.device) + offset_y = torch.randint(0, x.size(3) + (1 - cutout_size[1] % 2), size=[x.size(0), 1, 1], device=x.device) + grid_batch, grid_x, grid_y = torch.meshgrid( + torch.arange(x.size(0), dtype=torch.long, device=x.device), + torch.arange(cutout_size[0], dtype=torch.long, device=x.device), + torch.arange(cutout_size[1], dtype=torch.long, device=x.device), + ) + grid_x = torch.clamp(grid_x + offset_x - cutout_size[0] // 2, min=0, max=x.size(2) - 1) + grid_y = torch.clamp(grid_y + offset_y - cutout_size[1] // 2, min=0, max=x.size(3) - 1) + mask = torch.ones(x.size(0), x.size(2), x.size(3), dtype=x.dtype, device=x.device) + mask[grid_batch, grid_x, grid_y] = 0 + x = x * mask.unsqueeze(1) + return x + + +AUGMENT_FNS = { + 'color': [rand_brightness, rand_saturation, rand_contrast], + 'translation': [rand_translation], + 'cutout': [rand_cutout], +} diff --git a/mlflow/registry/Swimswap/pg_modules/projected_discriminator.py b/mlflow/registry/Swimswap/pg_modules/projected_discriminator.py new file mode 100644 index 0000000..d0c879f --- /dev/null +++ b/mlflow/registry/Swimswap/pg_modules/projected_discriminator.py @@ -0,0 +1,191 @@ +from functools import partial +import numpy as np +import torch +import torch.nn as nn + +from pg_modules.blocks import DownBlock, DownBlockPatch, conv2d +from pg_modules.projector import F_RandomProj +from pg_modules.diffaug import DiffAugment + + +class SingleDisc(nn.Module): + def __init__(self, nc=None, ndf=None, start_sz=256, end_sz=8, head=None, separable=False, patch=False): + super().__init__() + channel_dict = {4: 512, 8: 512, 16: 256, 32: 128, 64: 64, 128: 64, + 256: 32, 512: 16, 1024: 8} + + # interpolate for start sz that are not powers of two + if start_sz not in channel_dict.keys(): + sizes = np.array(list(channel_dict.keys())) + start_sz = sizes[np.argmin(abs(sizes - start_sz))] + self.start_sz = start_sz + + # if given ndf, allocate all layers with the same ndf + if ndf is None: + nfc = channel_dict + else: + nfc = {k: ndf for k, v in channel_dict.items()} + + # for feature map discriminators with nfc not in channel_dict + # this is the case for the pretrained backbone (midas.pretrained) + if nc is not None and head is None: + nfc[start_sz] = nc + + layers = [] + + # Head if the initial input is the full modality + if head: + layers += [conv2d(nc, nfc[256], 3, 1, 1, bias=False), + nn.LeakyReLU(0.2, inplace=True)] + + # Down Blocks + DB = partial(DownBlockPatch, separable=separable) if patch else partial(DownBlock, separable=separable) + while start_sz > end_sz: + layers.append(DB(nfc[start_sz], nfc[start_sz//2])) + start_sz = start_sz // 2 + + layers.append(conv2d(nfc[end_sz], 1, 4, 1, 0, bias=False)) + self.main = nn.Sequential(*layers) + + def forward(self, x, c): + return self.main(x) + + +class SingleDiscCond(nn.Module): + def __init__(self, nc=None, ndf=None, start_sz=256, end_sz=8, head=None, separable=False, patch=False, c_dim=1000, cmap_dim=64, embedding_dim=128): + super().__init__() + self.cmap_dim = cmap_dim + + # midas channels + channel_dict = {4: 512, 8: 512, 16: 256, 32: 128, 64: 64, 128: 64, + 256: 32, 512: 16, 1024: 8} + + # interpolate for start sz that are not powers of two + if start_sz not in channel_dict.keys(): + sizes = np.array(list(channel_dict.keys())) + start_sz = sizes[np.argmin(abs(sizes - start_sz))] + self.start_sz = start_sz + + # if given ndf, allocate all layers with the same ndf + if ndf is None: + nfc = channel_dict + else: + nfc = {k: ndf for k, v in channel_dict.items()} + + # for feature map discriminators with nfc not in channel_dict + # this is the case for the pretrained backbone (midas.pretrained) + if nc is not None and head is None: + nfc[start_sz] = nc + + layers = [] + + # Head if the initial input is the full modality + if head: + layers += [conv2d(nc, nfc[256], 3, 1, 1, bias=False), + nn.LeakyReLU(0.2, inplace=True)] + + # Down Blocks + DB = partial(DownBlockPatch, separable=separable) if patch else partial(DownBlock, separable=separable) + while start_sz > end_sz: + layers.append(DB(nfc[start_sz], nfc[start_sz//2])) + start_sz = start_sz // 2 + self.main = nn.Sequential(*layers) + + # additions for conditioning on class information + self.cls = conv2d(nfc[end_sz], self.cmap_dim, 4, 1, 0, bias=False) + self.embed = nn.Embedding(num_embeddings=c_dim, embedding_dim=embedding_dim) + self.embed_proj = nn.Sequential( + nn.Linear(self.embed.embedding_dim, self.cmap_dim), + nn.LeakyReLU(0.2, inplace=True), + ) + + def forward(self, x, c): + h = self.main(x) + out = self.cls(h) + + # conditioning via projection + cmap = self.embed_proj(self.embed(c.argmax(1))).unsqueeze(-1).unsqueeze(-1) + out = (out * cmap).sum(dim=1, keepdim=True) * (1 / np.sqrt(self.cmap_dim)) + + return out + + +class MultiScaleD(nn.Module): + def __init__( + self, + channels, + resolutions, + num_discs=4, + proj_type=2, # 0 = no projection, 1 = cross channel mixing, 2 = cross scale mixing + cond=0, + separable=False, + patch=False, + **kwargs, + ): + super().__init__() + + assert num_discs in [1, 2, 3, 4] + + # the first disc is on the lowest level of the backbone + self.disc_in_channels = channels[:num_discs] + self.disc_in_res = resolutions[:num_discs] + Disc = SingleDiscCond if cond else SingleDisc + + mini_discs = [] + for i, (cin, res) in enumerate(zip(self.disc_in_channels, self.disc_in_res)): + start_sz = res if not patch else 16 + mini_discs += [str(i), Disc(nc=cin, start_sz=start_sz, end_sz=8, separable=separable, patch=patch)], + self.mini_discs = nn.ModuleDict(mini_discs) + + def forward(self, features, c): + all_logits = [] + for k, disc in self.mini_discs.items(): + res = disc(features[k], c).view(features[k].size(0), -1) + all_logits.append(res) + + all_logits = torch.cat(all_logits, dim=1) + return all_logits + + +class ProjectedDiscriminator(torch.nn.Module): + def __init__( + self, + diffaug=True, + interp224=True, + backbone_kwargs={}, + **kwargs + ): + super().__init__() + self.diffaug = diffaug + self.interp224 = interp224 + self.feature_network = F_RandomProj(**backbone_kwargs) + self.discriminator = MultiScaleD( + channels=self.feature_network.CHANNELS, + resolutions=self.feature_network.RESOLUTIONS, + **backbone_kwargs, + ) + + def train(self, mode=True): + self.feature_network = self.feature_network.train(False) + self.discriminator = self.discriminator.train(mode) + return self + + def eval(self): + return self.train(False) + + def get_feature(self, x): + features = self.feature_network(x, get_features=True) + return features + + def forward(self, x, c): + # if self.diffaug: + # x = DiffAugment(x, policy='color,translation,cutout') + + # if self.interp224: + # x = F.interpolate(x, 224, mode='bilinear', align_corners=False) + + features,backbone_features = self.feature_network(x) + logits = self.discriminator(features, c) + + return logits,backbone_features + diff --git a/mlflow/registry/Swimswap/pg_modules/projector.py b/mlflow/registry/Swimswap/pg_modules/projector.py new file mode 100644 index 0000000..610a482 --- /dev/null +++ b/mlflow/registry/Swimswap/pg_modules/projector.py @@ -0,0 +1,158 @@ +import torch +import torch.nn as nn +import timm +from pg_modules.blocks import FeatureFusionBlock + + +def _make_scratch_ccm(scratch, in_channels, cout, expand=False): + # shapes + out_channels = [cout, cout*2, cout*4, cout*8] if expand else [cout]*4 + + scratch.layer0_ccm = nn.Conv2d(in_channels[0], out_channels[0], kernel_size=1, stride=1, padding=0, bias=True) + scratch.layer1_ccm = nn.Conv2d(in_channels[1], out_channels[1], kernel_size=1, stride=1, padding=0, bias=True) + scratch.layer2_ccm = nn.Conv2d(in_channels[2], out_channels[2], kernel_size=1, stride=1, padding=0, bias=True) + scratch.layer3_ccm = nn.Conv2d(in_channels[3], out_channels[3], kernel_size=1, stride=1, padding=0, bias=True) + + scratch.CHANNELS = out_channels + + return scratch + + +def _make_scratch_csm(scratch, in_channels, cout, expand): + scratch.layer3_csm = FeatureFusionBlock(in_channels[3], nn.ReLU(False), expand=expand, lowest=True) + scratch.layer2_csm = FeatureFusionBlock(in_channels[2], nn.ReLU(False), expand=expand) + scratch.layer1_csm = FeatureFusionBlock(in_channels[1], nn.ReLU(False), expand=expand) + scratch.layer0_csm = FeatureFusionBlock(in_channels[0], nn.ReLU(False)) + + # last refinenet does not expand to save channels in higher dimensions + scratch.CHANNELS = [cout, cout, cout*2, cout*4] if expand else [cout]*4 + + return scratch + + +def _make_efficientnet(model): + pretrained = nn.Module() + pretrained.layer0 = nn.Sequential(model.conv_stem, model.bn1, model.act1, *model.blocks[0:2]) + pretrained.layer1 = nn.Sequential(*model.blocks[2:3]) + pretrained.layer2 = nn.Sequential(*model.blocks[3:5]) + pretrained.layer3 = nn.Sequential(*model.blocks[5:9]) + return pretrained + + +def calc_channels(pretrained, inp_res=224): + channels = [] + tmp = torch.zeros(1, 3, inp_res, inp_res) + + # forward pass + tmp = pretrained.layer0(tmp) + channels.append(tmp.shape[1]) + tmp = pretrained.layer1(tmp) + channels.append(tmp.shape[1]) + tmp = pretrained.layer2(tmp) + channels.append(tmp.shape[1]) + tmp = pretrained.layer3(tmp) + channels.append(tmp.shape[1]) + + return channels + + +def _make_projector(im_res, cout, proj_type, expand=False): + assert proj_type in [0, 1, 2], "Invalid projection type" + + ### Build pretrained feature network + model = timm.create_model('tf_efficientnet_lite0', pretrained=True) + pretrained = _make_efficientnet(model) + + # determine resolution of feature maps, this is later used to calculate the number + # of down blocks in the discriminators. Interestingly, the best results are achieved + # by fixing this to 256, ie., we use the same number of down blocks per discriminator + # independent of the dataset resolution + im_res = 256 + pretrained.RESOLUTIONS = [im_res//4, im_res//8, im_res//16, im_res//32] + pretrained.CHANNELS = calc_channels(pretrained) + + if proj_type == 0: return pretrained, None + + ### Build CCM + scratch = nn.Module() + scratch = _make_scratch_ccm(scratch, in_channels=pretrained.CHANNELS, cout=cout, expand=expand) + pretrained.CHANNELS = scratch.CHANNELS + + if proj_type == 1: return pretrained, scratch + + ### build CSM + scratch = _make_scratch_csm(scratch, in_channels=scratch.CHANNELS, cout=cout, expand=expand) + + # CSM upsamples x2 so the feature map resolution doubles + pretrained.RESOLUTIONS = [res*2 for res in pretrained.RESOLUTIONS] + pretrained.CHANNELS = scratch.CHANNELS + + return pretrained, scratch + + +class F_RandomProj(nn.Module): + def __init__( + self, + im_res=256, + cout=64, + expand=True, + proj_type=2, # 0 = no projection, 1 = cross channel mixing, 2 = cross scale mixing + **kwargs, + ): + super().__init__() + self.proj_type = proj_type + self.cout = cout + self.expand = expand + + # build pretrained feature network and random decoder (scratch) + self.pretrained, self.scratch = _make_projector(im_res=im_res, cout=self.cout, proj_type=self.proj_type, expand=self.expand) + self.CHANNELS = self.pretrained.CHANNELS + self.RESOLUTIONS = self.pretrained.RESOLUTIONS + + def forward(self, x, get_features=False): + # predict feature maps + out0 = self.pretrained.layer0(x) + out1 = self.pretrained.layer1(out0) + out2 = self.pretrained.layer2(out1) + out3 = self.pretrained.layer3(out2) + + # start enumerating at the lowest layer (this is where we put the first discriminator) + backbone_features = { + '0': out0, + '1': out1, + '2': out2, + '3': out3, + } + if get_features: + return backbone_features + + if self.proj_type == 0: return backbone_features + + out0_channel_mixed = self.scratch.layer0_ccm(backbone_features['0']) + out1_channel_mixed = self.scratch.layer1_ccm(backbone_features['1']) + out2_channel_mixed = self.scratch.layer2_ccm(backbone_features['2']) + out3_channel_mixed = self.scratch.layer3_ccm(backbone_features['3']) + + out = { + '0': out0_channel_mixed, + '1': out1_channel_mixed, + '2': out2_channel_mixed, + '3': out3_channel_mixed, + } + + if self.proj_type == 1: return out + + # from bottom to top + out3_scale_mixed = self.scratch.layer3_csm(out3_channel_mixed) + out2_scale_mixed = self.scratch.layer2_csm(out3_scale_mixed, out2_channel_mixed) + out1_scale_mixed = self.scratch.layer1_csm(out2_scale_mixed, out1_channel_mixed) + out0_scale_mixed = self.scratch.layer0_csm(out1_scale_mixed, out0_channel_mixed) + + out = { + '0': out0_scale_mixed, + '1': out1_scale_mixed, + '2': out2_scale_mixed, + '3': out3_scale_mixed, + } + + return out, backbone_features diff --git a/mlflow/registry/Swimswap/util/add_watermark.py b/mlflow/registry/Swimswap/util/add_watermark.py new file mode 100644 index 0000000..a4c5d3d --- /dev/null +++ b/mlflow/registry/Swimswap/util/add_watermark.py @@ -0,0 +1,118 @@ + +import cv2 +import numpy as np +from PIL import Image +import math +import numpy as np +# import torch +# from torchvision import transforms + +def rotate_image(image, angle, center = None, scale = 1.0): + (h, w) = image.shape[:2] + + if center is None: + center = (w / 2, h / 2) + + # Perform the rotation + M = cv2.getRotationMatrix2D(center, angle, scale) + rotated = cv2.warpAffine(image, M, (w, h)) + + return rotated + +class watermark_image: + def __init__(self, logo_path, size=0.3, oritation="DR", margin=(5,20,20,20), angle=15, rgb_weight=(0,1,1.5), input_frame_shape=None) -> None: + + logo_image = cv2.imread(logo_path, cv2.IMREAD_UNCHANGED) + h,w,c = logo_image.shape + if angle%360 != 0: + new_h = w*math.sin(angle/180*math.pi) + h*math.cos(angle/180*math.pi) + pad_h = int((new_h-h)//2) + + padding = np.zeros((pad_h, w, c), dtype=np.uint8) + logo_image = cv2.vconcat([logo_image, padding]) + logo_image = cv2.vconcat([padding, logo_image]) + + logo_image = rotate_image(logo_image, angle) + print(logo_image.shape) + self.logo_image = logo_image + + if self.logo_image.shape[2] < 4: + print("No alpha channel found!") + self.logo_image = self.__addAlpha__(self.logo_image) #add alpha channel + self.size = size + self.oritation = oritation + self.margin = margin + self.ori_shape = self.logo_image.shape + self.resized = False + self.rgb_weight = rgb_weight + + self.logo_image[:, :, 2] = self.logo_image[:, :, 2]*self.rgb_weight[0] + self.logo_image[:, :, 1] = self.logo_image[:, :, 1]*self.rgb_weight[1] + self.logo_image[:, :, 0] = self.logo_image[:, :, 0]*self.rgb_weight[2] + + if input_frame_shape is not None: + + logo_w = input_frame_shape[1] * self.size + ratio = logo_w / self.ori_shape[1] + logo_h = int(ratio * self.ori_shape[0]) + logo_w = int(logo_w) + + size = (logo_w, logo_h) + self.logo_image = cv2.resize(self.logo_image, size, interpolation = cv2.INTER_CUBIC) + self.resized = True + if oritation == "UL": + self.coor_h = self.margin[1] + self.coor_w = self.margin[0] + elif oritation == "UR": + self.coor_h = self.margin[1] + self.coor_w = input_frame_shape[1] - (logo_w + self.margin[2]) + elif oritation == "DL": + self.coor_h = input_frame_shape[0] - (logo_h + self.margin[1]) + self.coor_w = self.margin[0] + else: + self.coor_h = input_frame_shape[0] - (logo_h + self.margin[3]) + self.coor_w = input_frame_shape[1] - (logo_w + self.margin[2]) + self.logo_w = logo_w + self.logo_h = logo_h + self.mask = self.logo_image[:,:,3] + self.mask = cv2.bitwise_not(self.mask//255) + + def apply_frames(self, frame): + + if not self.resized: + shape = frame.shape + logo_w = shape[1] * self.size + ratio = logo_w / self.ori_shape[1] + logo_h = int(ratio * self.ori_shape[0]) + logo_w = int(logo_w) + + size = (logo_w, logo_h) + self.logo_image = cv2.resize(self.logo_image, size, interpolation = cv2.INTER_CUBIC) + self.resized = True + if self.oritation == "UL": + self.coor_h = self.margin[1] + self.coor_w = self.margin[0] + elif self.oritation == "UR": + self.coor_h = self.margin[1] + self.coor_w = shape[1] - (logo_w + self.margin[2]) + elif self.oritation == "DL": + self.coor_h = shape[0] - (logo_h + self.margin[1]) + self.coor_w = self.margin[0] + else: + self.coor_h = shape[0] - (logo_h + self.margin[3]) + self.coor_w = shape[1] - (logo_w + self.margin[2]) + self.logo_w = logo_w + self.logo_h = logo_h + self.mask = self.logo_image[:,:,3] + self.mask = cv2.bitwise_not(self.mask//255) + + original_frame = frame[self.coor_h:(self.coor_h+self.logo_h), self.coor_w:(self.coor_w+self.logo_w),:] + blending_logo = cv2.add(self.logo_image[:,:,0:3],original_frame,mask = self.mask) + frame[self.coor_h:(self.coor_h+self.logo_h), self.coor_w:(self.coor_w+self.logo_w),:] = blending_logo + return frame + + def __addAlpha__(self, image): + shape = image.shape + alpha_channel = np.ones((shape[0],shape[1],1),np.uint8)*255 + return np.concatenate((image,alpha_channel),2) + diff --git a/mlflow/registry/Swimswap/util/gifswap.py b/mlflow/registry/Swimswap/util/gifswap.py new file mode 100644 index 0000000..e0eaca9 --- /dev/null +++ b/mlflow/registry/Swimswap/util/gifswap.py @@ -0,0 +1,129 @@ +''' +Author: Naiyuan liu +Github: https://github.com/NNNNAI +Date: 2021-11-23 17:03:58 +LastEditors: Naiyuan liu +LastEditTime: 2021-11-24 19:19:52 +Description: +''' +import os +import cv2 +import glob +import torch +import shutil +import numpy as np +from tqdm import tqdm +from util.reverse2original import reverse2wholeimage +import moviepy.editor as mp +from moviepy.editor import AudioFileClip, VideoFileClip +from moviepy.video.io.ImageSequenceClip import ImageSequenceClip +import time +from util.add_watermark import watermark_image +from util.norm import SpecificNorm +from parsing_model.model import BiSeNet +from array2gif import write_gif + + +def _totensor(array): + tensor = torch.from_numpy(array) + img = tensor.transpose(0, 1).transpose(0, 2).contiguous() + return img.float().div(255) + +def gif_swap(video_path, id_vetor, swap_model, detect_model, save_path, temp_results_dir='./temp_results', crop_size=224, no_simswaplogo=False, use_mask=False): + video_forcheck = VideoFileClip(video_path) + if video_forcheck.audio is None: + no_audio = True + else: + no_audio = False + + del video_forcheck + + if not no_audio: + video_audio_clip = AudioFileClip(video_path) + + video = cv2.VideoCapture(video_path) + logoclass = watermark_image('./SimSwap/simswaplogo/simswaplogo.png') + ret = True + frame_index = 0 + + frame_count = int(video.get(cv2.CAP_PROP_FRAME_COUNT)) + + # video_WIDTH = int(video.get(cv2.CAP_PROP_FRAME_WIDTH)) + + # video_HEIGHT = int(video.get(cv2.CAP_PROP_FRAME_HEIGHT)) + + fps = video.get(cv2.CAP_PROP_FPS) + if os.path.exists(temp_results_dir): + shutil.rmtree(temp_results_dir) + + spNorm =SpecificNorm() + if use_mask: + n_classes = 19 + net = BiSeNet(n_classes=n_classes) + net.cuda() + save_pth = os.path.join('./SimSwap/parsing_model/checkpoint', '79999_iter.pth') + net.load_state_dict(torch.load(save_pth)) + net.eval() + else: + net =None + + frames = [] + # while ret: + for frame_index in tqdm(range(frame_count)): + ret, frame = video.read() + if ret: + detect_results = detect_model.get(frame,crop_size) + + if detect_results is not None: + # print(frame_index) + # if not os.path.exists(temp_results_dir): + # os.mkdir(temp_results_dir) + frame_align_crop_list = detect_results[0] + frame_mat_list = detect_results[1] + swap_result_list = [] + frame_align_crop_tenor_list = [] + for frame_align_crop in frame_align_crop_list: + + # BGR TO RGB + # frame_align_crop_RGB = frame_align_crop[...,::-1] + + frame_align_crop_tenor = _totensor(cv2.cvtColor(frame_align_crop,cv2.COLOR_BGR2RGB))[None,...].cuda() + + swap_result = swap_model(None, frame_align_crop_tenor, id_vetor, None, True)[0] + # cv2.imwrite(os.path.join(temp_results_dir, 'frame_{:0>7d}.jpg'.format(frame_index)), frame) # 원본 프레임 img 저장 + swap_result_list.append(swap_result) + frame_align_crop_tenor_list.append(frame_align_crop_tenor) + + + img = reverse2wholeimage(frame_align_crop_tenor_list,swap_result_list, frame_mat_list, crop_size, frame, logoclass,\ + os.path.join(temp_results_dir, 'frame_{:0>7d}.jpg'.format(frame_index)), no_simswaplogo, pasring_model=net, use_mask=use_mask, norm=spNorm) + frames.append(img.tobytes()) + + else: + if not os.path.exists(temp_results_dir): + os.mkdir(temp_results_dir) + frame = frame.astype(np.uint8) + # if not no_simswaplogo: + # frame = logoclass.apply_frames(frame) + cv2.imwrite(os.path.join(temp_results_dir, 'frame_{:0>7d}.jpg'.format(frame_index)), frame) + else: + break + + video.release() + + # image_filename_list = [] + # path = os.path.join(temp_results_dir, '*.jpg') + # image_filenames = sorted(glob.glob(path)) + + # clips = ImageSequenceClip(image_filenames, fps = fps) + + # if not no_audio: + # clips = clips.set_audio(video_audio_clip) + + # # if save_path.split('.')[-1] == 'gif': + # clips.write_gif(save_path) + + # write_gif(frames, save_path, fps=fps) + return frames, fps + + diff --git a/mlflow/registry/Swimswap/util/html.py b/mlflow/registry/Swimswap/util/html.py new file mode 100644 index 0000000..71c48ad --- /dev/null +++ b/mlflow/registry/Swimswap/util/html.py @@ -0,0 +1,63 @@ +import dominate +from dominate.tags import * +import os + + +class HTML: + def __init__(self, web_dir, title, refresh=0): + self.title = title + self.web_dir = web_dir + self.img_dir = os.path.join(self.web_dir, 'images') + if not os.path.exists(self.web_dir): + os.makedirs(self.web_dir) + if not os.path.exists(self.img_dir): + os.makedirs(self.img_dir) + + self.doc = dominate.document(title=title) + if refresh > 0: + with self.doc.head: + meta(http_equiv="refresh", content=str(refresh)) + + def get_image_dir(self): + return self.img_dir + + def add_header(self, str): + with self.doc: + h3(str) + + def add_table(self, border=1): + self.t = table(border=border, style="table-layout: fixed;") + self.doc.add(self.t) + + def add_images(self, ims, txts, links, width=512): + self.add_table() + with self.t: + with tr(): + for im, txt, link in zip(ims, txts, links): + with td(style="word-wrap: break-word;", halign="center", valign="top"): + with p(): + with a(href=os.path.join('images', link)): + img(style="width:%dpx" % (width), src=os.path.join('images', im)) + br() + p(txt) + + def save(self): + html_file = '%s/index.html' % self.web_dir + f = open(html_file, 'wt') + f.write(self.doc.render()) + f.close() + + +if __name__ == '__main__': + html = HTML('web/', 'test_html') + html.add_header('hello world') + + ims = [] + txts = [] + links = [] + for n in range(4): + ims.append('image_%d.jpg' % n) + txts.append('text_%d' % n) + links.append('image_%d.jpg' % n) + html.add_images(ims, txts, links) + html.save() diff --git a/mlflow/registry/Swimswap/util/image_pool.py b/mlflow/registry/Swimswap/util/image_pool.py new file mode 100644 index 0000000..63e1877 --- /dev/null +++ b/mlflow/registry/Swimswap/util/image_pool.py @@ -0,0 +1,31 @@ +import random +import torch +from torch.autograd import Variable +class ImagePool(): + def __init__(self, pool_size): + self.pool_size = pool_size + if self.pool_size > 0: + self.num_imgs = 0 + self.images = [] + + def query(self, images): + if self.pool_size == 0: + return images + return_images = [] + for image in images.data: + image = torch.unsqueeze(image, 0) + if self.num_imgs < self.pool_size: + self.num_imgs = self.num_imgs + 1 + self.images.append(image) + return_images.append(image) + else: + p = random.uniform(0, 1) + if p > 0.5: + random_id = random.randint(0, self.pool_size-1) + tmp = self.images[random_id].clone() + self.images[random_id] = image + return_images.append(tmp) + else: + return_images.append(image) + return_images = Variable(torch.cat(return_images, 0)) + return return_images diff --git a/mlflow/registry/Swimswap/util/json_config.py b/mlflow/registry/Swimswap/util/json_config.py new file mode 100644 index 0000000..c68fbff --- /dev/null +++ b/mlflow/registry/Swimswap/util/json_config.py @@ -0,0 +1,15 @@ +import json + + +def readConfig(path): + with open(path,'r') as cf: + nodelocaltionstr = cf.read() + nodelocaltioninf = json.loads(nodelocaltionstr) + if isinstance(nodelocaltioninf,str): + nodelocaltioninf = json.loads(nodelocaltioninf) + return nodelocaltioninf + +def writeConfig(path, info): + with open(path, 'w') as cf: + configjson = json.dumps(info, indent=4) + cf.writelines(configjson) \ No newline at end of file diff --git a/mlflow/registry/Swimswap/util/logo_class.py b/mlflow/registry/Swimswap/util/logo_class.py new file mode 100644 index 0000000..044dce3 --- /dev/null +++ b/mlflow/registry/Swimswap/util/logo_class.py @@ -0,0 +1,44 @@ +#!/usr/bin/env python3 +# -*- coding:utf-8 -*- +############################################################# +# File: logo_class.py +# Created Date: Tuesday June 29th 2021 +# Author: Chen Xuanhong +# Email: chenxuanhongzju@outlook.com +# Last Modified: Monday, 11th October 2021 12:39:55 am +# Modified By: Chen Xuanhong +# Copyright (c) 2021 Shanghai Jiao Tong University +############################################################# + +class logo_class: + + @staticmethod + def print_group_logo(): + logo_str = """ + +███╗ ██╗██████╗ ███████╗██╗ ██████╗ ███████╗ ██╗████████╗██╗ ██╗ +████╗ ██║██╔══██╗██╔════╝██║██╔════╝ ██╔════╝ ██║╚══██╔══╝██║ ██║ +██╔██╗ ██║██████╔╝███████╗██║██║ ███╗ ███████╗ ██║ ██║ ██║ ██║ +██║╚██╗██║██╔══██╗╚════██║██║██║ ██║ ╚════██║██ ██║ ██║ ██║ ██║ +██║ ╚████║██║ ██║███████║██║╚██████╔╝ ███████║╚█████╔╝ ██║ ╚██████╔╝ +╚═╝ ╚═══╝╚═╝ ╚═╝╚══════╝╚═╝ ╚═════╝ ╚══════╝ ╚════╝ ╚═╝ ╚═════╝ +Neural Rendering Special Interesting Group of SJTU + + """ + print(logo_str) + + @staticmethod + def print_start_training(): + logo_str = """ + _____ __ __ ______ _ _ + / ___/ / /_ ____ _ _____ / /_ /_ __/_____ ____ _ (_)____ (_)____ ____ _ + \__ \ / __// __ `// ___// __/ / / / ___// __ `// // __ \ / // __ \ / __ `/ + ___/ // /_ / /_/ // / / /_ / / / / / /_/ // // / / // // / / // /_/ / +/____/ \__/ \__,_//_/ \__/ /_/ /_/ \__,_//_//_/ /_//_//_/ /_/ \__, / + /____/ + """ + print(logo_str) + +if __name__=="__main__": + # logo_class.print_group_logo() + logo_class.print_start_training() \ No newline at end of file diff --git a/mlflow/registry/Swimswap/util/norm.py b/mlflow/registry/Swimswap/util/norm.py new file mode 100644 index 0000000..981adcf --- /dev/null +++ b/mlflow/registry/Swimswap/util/norm.py @@ -0,0 +1,25 @@ +import torch.nn as nn +import numpy as np +import torch +class SpecificNorm(nn.Module): + def __init__(self, epsilon=1e-8): + """ + @notice: avoid in-place ops. + https://discuss.pytorch.org/t/encounter-the-runtimeerror-one-of-the-variables-needed-for-gradient-computation-has-been-modified-by-an-inplace-operation/836/3 + """ + super(SpecificNorm, self).__init__() + self.mean = np.array([0.485, 0.456, 0.406]) + self.mean = torch.from_numpy(self.mean).float().cuda() + self.mean = self.mean.view([1, 3, 1, 1]) + + self.std = np.array([0.229, 0.224, 0.225]) + self.std = torch.from_numpy(self.std).float().cuda() + self.std = self.std.view([1, 3, 1, 1]) + + def forward(self, x): + mean = self.mean.expand([1, 3, x.shape[2], x.shape[3]]) + std = self.std.expand([1, 3, x.shape[2], x.shape[3]]) + + x = (x - mean) / std + + return x \ No newline at end of file diff --git a/mlflow/registry/Swimswap/util/plot.py b/mlflow/registry/Swimswap/util/plot.py new file mode 100644 index 0000000..0da1c75 --- /dev/null +++ b/mlflow/registry/Swimswap/util/plot.py @@ -0,0 +1,37 @@ +import numpy as np +import math +import PIL + +def postprocess(x): + """[0,1] to uint8.""" + + x = np.clip(255 * x, 0, 255) + x = np.cast[np.uint8](x) + return x + +def tile(X, rows, cols): + """Tile images for display.""" + tiling = np.zeros((rows * X.shape[1], cols * X.shape[2], X.shape[3]), dtype = X.dtype) + for i in range(rows): + for j in range(cols): + idx = i * cols + j + if idx < X.shape[0]: + img = X[idx,...] + tiling[ + i*X.shape[1]:(i+1)*X.shape[1], + j*X.shape[2]:(j+1)*X.shape[2], + :] = img + return tiling + + +def plot_batch(X, out_path): + """Save batch of images tiled.""" + n_channels = X.shape[3] + if n_channels > 3: + X = X[:,:,:,np.random.choice(n_channels, size = 3)] + X = postprocess(X) + rc = math.sqrt(X.shape[0]) + rows = cols = math.ceil(rc) + canvas = tile(X, rows, cols) + canvas = np.squeeze(canvas) + PIL.Image.fromarray(canvas).save(out_path) \ No newline at end of file diff --git a/mlflow/registry/Swimswap/util/reverse2original.py b/mlflow/registry/Swimswap/util/reverse2original.py new file mode 100644 index 0000000..81305b8 --- /dev/null +++ b/mlflow/registry/Swimswap/util/reverse2original.py @@ -0,0 +1,176 @@ +import cv2 +import numpy as np +# import time +import torch +from torch.nn import functional as F +import torch.nn as nn + + +def encode_segmentation_rgb(segmentation, no_neck=True): + parse = segmentation + + face_part_ids = [1, 2, 3, 4, 5, 6, 10, 12, 13] if no_neck else [1, 2, 3, 4, 5, 6, 7, 8, 10, 12, 13, 14] + mouth_id = 11 + # hair_id = 17 + face_map = np.zeros([parse.shape[0], parse.shape[1]]) + mouth_map = np.zeros([parse.shape[0], parse.shape[1]]) + # hair_map = np.zeros([parse.shape[0], parse.shape[1]]) + + for valid_id in face_part_ids: + valid_index = np.where(parse==valid_id) + face_map[valid_index] = 255 + valid_index = np.where(parse==mouth_id) + mouth_map[valid_index] = 255 + # valid_index = np.where(parse==hair_id) + # hair_map[valid_index] = 255 + #return np.stack([face_map, mouth_map,hair_map], axis=2) + return np.stack([face_map, mouth_map], axis=2) + + +class SoftErosion(nn.Module): + def __init__(self, kernel_size=15, threshold=0.6, iterations=1): + super(SoftErosion, self).__init__() + r = kernel_size // 2 + self.padding = r + self.iterations = iterations + self.threshold = threshold + + # Create kernel + y_indices, x_indices = torch.meshgrid(torch.arange(0., kernel_size), torch.arange(0., kernel_size)) + dist = torch.sqrt((x_indices - r) ** 2 + (y_indices - r) ** 2) + kernel = dist.max() - dist + kernel /= kernel.sum() + kernel = kernel.view(1, 1, *kernel.shape) + self.register_buffer('weight', kernel) + + def forward(self, x): + x = x.float() + for i in range(self.iterations - 1): + x = torch.min(x, F.conv2d(x, weight=self.weight, groups=x.shape[1], padding=self.padding)) + x = F.conv2d(x, weight=self.weight, groups=x.shape[1], padding=self.padding) + + mask = x >= self.threshold + x[mask] = 1.0 + x[~mask] /= x[~mask].max() + + return x, mask + + +def postprocess(swapped_face, target, target_mask,smooth_mask): + # target_mask = cv2.resize(target_mask, (self.size, self.size)) + + mask_tensor = torch.from_numpy(target_mask.copy().transpose((2, 0, 1))).float().mul_(1/255.0).cuda() + face_mask_tensor = mask_tensor[0] + mask_tensor[1] + + soft_face_mask_tensor, _ = smooth_mask(face_mask_tensor.unsqueeze_(0).unsqueeze_(0)) + soft_face_mask_tensor.squeeze_() + + soft_face_mask = soft_face_mask_tensor.cpu().numpy() + soft_face_mask = soft_face_mask[:, :, np.newaxis] + + result = swapped_face * soft_face_mask + target * (1 - soft_face_mask) + result = result[:,:,::-1]# .astype(np.uint8) + return result + +def reverse2wholeimage(b_align_crop_tenor_list, swaped_imgs, mats, crop_size, oriimg, logoclass, save_path='', \ + no_simswaplogo=False, pasring_model=None, norm=None, use_mask=False): + + target_image_list = [] + img_mask_list = [] + if use_mask: + smooth_mask = SoftErosion(kernel_size=17, threshold=0.9, iterations=7).cuda() + else: + pass + + # print(len(swaped_imgs)) + # print(mats) + # print(len(b_align_crop_tenor_list)) + for swaped_img, mat ,source_img in zip(swaped_imgs, mats, b_align_crop_tenor_list): + swaped_img = swaped_img.cpu().detach().numpy().transpose((1, 2, 0)) + img_white = np.full((crop_size,crop_size), 255, dtype=float) + + # inverse the Affine transformation matrix + mat_rev = np.zeros([2,3]) + div1 = mat[0][0]*mat[1][1]-mat[0][1]*mat[1][0] + mat_rev[0][0] = mat[1][1]/div1 + mat_rev[0][1] = -mat[0][1]/div1 + mat_rev[0][2] = -(mat[0][2]*mat[1][1]-mat[0][1]*mat[1][2])/div1 + div2 = mat[0][1]*mat[1][0]-mat[0][0]*mat[1][1] + mat_rev[1][0] = mat[1][0]/div2 + mat_rev[1][1] = -mat[0][0]/div2 + mat_rev[1][2] = -(mat[0][2]*mat[1][0]-mat[0][0]*mat[1][2])/div2 + + orisize = (oriimg.shape[1], oriimg.shape[0]) + if use_mask: + source_img_norm = norm(source_img) + source_img_512 = F.interpolate(source_img_norm,size=(512,512)) + out = pasring_model(source_img_512)[0] + parsing = out.squeeze(0).detach().cpu().numpy().argmax(0) + vis_parsing_anno = parsing.copy().astype(np.uint8) + tgt_mask = encode_segmentation_rgb(vis_parsing_anno) + if tgt_mask.sum() >= 5000: + # face_mask_tensor = tgt_mask[...,0] + tgt_mask[...,1] + target_mask = cv2.resize(tgt_mask, (crop_size, crop_size)) + # print(source_img) + target_image_parsing = postprocess(swaped_img, source_img[0].cpu().detach().numpy().transpose((1, 2, 0)), target_mask,smooth_mask) + + + target_image = cv2.warpAffine(target_image_parsing, mat_rev, orisize) + # target_image_parsing = cv2.warpAffine(swaped_img, mat_rev, orisize) + else: + target_image = cv2.warpAffine(swaped_img, mat_rev, orisize)[..., ::-1] + else: + target_image = cv2.warpAffine(swaped_img, mat_rev, orisize) + # source_image = cv2.warpAffine(source_img, mat_rev, orisize) + + img_white = cv2.warpAffine(img_white, mat_rev, orisize) + + + img_white[img_white>20] =255 + + img_mask = img_white + + # if use_mask: + # kernel = np.ones((40,40),np.uint8) + # img_mask = cv2.erode(img_mask,kernel,iterations = 1) + # else: + kernel = np.ones((40,40),np.uint8) + img_mask = cv2.erode(img_mask,kernel,iterations = 1) + kernel_size = (20, 20) + blur_size = tuple(2*i+1 for i in kernel_size) + img_mask = cv2.GaussianBlur(img_mask, blur_size, 0) + + # kernel = np.ones((10,10),np.uint8) + # img_mask = cv2.erode(img_mask,kernel,iterations = 1) + + + + img_mask /= 255 + + img_mask = np.reshape(img_mask, [img_mask.shape[0],img_mask.shape[1],1]) + + # pasing mask + + # target_image_parsing = postprocess(target_image, source_image, tgt_mask) + + if use_mask: + target_image = np.array(target_image, dtype=np.float) * 255 + else: + target_image = np.array(target_image, dtype=np.float)[..., ::-1] * 255 + + + img_mask_list.append(img_mask) + target_image_list.append(target_image) + + + # target_image /= 255 + # target_image = 0 + img = np.array(oriimg, dtype=np.float) + for img_mask, target_image in zip(img_mask_list, target_image_list): + img = img_mask * target_image + (1-img_mask) * img + + final_img = img.astype(np.uint8) + # if not no_simswaplogo: + # final_img = logoclass.apply_frames(final_img) + # cv2.imwrite(save_path, final_img) # swap img 저장 + return final_img diff --git a/mlflow/registry/Swimswap/util/save_heatmap.py b/mlflow/registry/Swimswap/util/save_heatmap.py new file mode 100644 index 0000000..71ce4c9 --- /dev/null +++ b/mlflow/registry/Swimswap/util/save_heatmap.py @@ -0,0 +1,57 @@ +#!/usr/bin/env python3 +# -*- coding:utf-8 -*- +############################################################# +# File: save_heatmap.py +# Created Date: Friday January 15th 2021 +# Author: Chen Xuanhong +# Email: chenxuanhongzju@outlook.com +# Last Modified: Wednesday, 19th January 2022 1:22:47 am +# Modified By: Chen Xuanhong +# Copyright (c) 2021 Shanghai Jiao Tong University +############################################################# + +import os +import shutil +import seaborn as sns +import matplotlib.pyplot as plt +import cv2 +import numpy as np + +def SaveHeatmap(heatmaps, path, row=-1, dpi=72): + """ + The input tensor must be B X 1 X H X W + """ + batch_size = heatmaps.shape[0] + temp_path = ".temp/" + if not os.path.exists(temp_path): + os.makedirs(temp_path) + final_img = None + if row < 1: + col = batch_size + row = 1 + else: + col = batch_size // row + if row * col = col: + col_i = 0 + row_i += 1 + cv2.imwrite(path,final_img) + +if __name__ == "__main__": + random_map = np.random.randn(16,1,10,10) + SaveHeatmap(random_map,"./wocao.png",1) diff --git a/mlflow/registry/Swimswap/util/util.py b/mlflow/registry/Swimswap/util/util.py new file mode 100644 index 0000000..f4f79ec --- /dev/null +++ b/mlflow/registry/Swimswap/util/util.py @@ -0,0 +1,100 @@ +from __future__ import print_function +import torch +import numpy as np +from PIL import Image +import numpy as np +import os + +# Converts a Tensor into a Numpy array +# |imtype|: the desired type of the converted numpy array +def tensor2im(image_tensor, imtype=np.uint8, normalize=True): + if isinstance(image_tensor, list): + image_numpy = [] + for i in range(len(image_tensor)): + image_numpy.append(tensor2im(image_tensor[i], imtype, normalize)) + return image_numpy + image_numpy = image_tensor.cpu().float().numpy() + if normalize: + image_numpy = (np.transpose(image_numpy, (1, 2, 0)) + 1) / 2.0 * 255.0 + else: + image_numpy = np.transpose(image_numpy, (1, 2, 0)) * 255.0 + image_numpy = np.clip(image_numpy, 0, 255) + if image_numpy.shape[2] == 1 or image_numpy.shape[2] > 3: + image_numpy = image_numpy[:,:,0] + return image_numpy.astype(imtype) + +# Converts a one-hot tensor into a colorful label map +def tensor2label(label_tensor, n_label, imtype=np.uint8): + if n_label == 0: + return tensor2im(label_tensor, imtype) + label_tensor = label_tensor.cpu().float() + if label_tensor.size()[0] > 1: + label_tensor = label_tensor.max(0, keepdim=True)[1] + label_tensor = Colorize(n_label)(label_tensor) + label_numpy = np.transpose(label_tensor.numpy(), (1, 2, 0)) + return label_numpy.astype(imtype) + +def save_image(image_numpy, image_path): + image_pil = Image.fromarray(image_numpy) + image_pil.save(image_path) + +def mkdirs(paths): + if isinstance(paths, list) and not isinstance(paths, str): + for path in paths: + mkdir(path) + else: + mkdir(paths) + +def mkdir(path): + if not os.path.exists(path): + os.makedirs(path) + +############################################################################### +# Code from +# https://github.com/ycszen/pytorch-seg/blob/master/transform.py +# Modified so it complies with the Citscape label map colors +############################################################################### +def uint82bin(n, count=8): + """returns the binary of integer n, count refers to amount of bits""" + return ''.join([str((n >> y) & 1) for y in range(count-1, -1, -1)]) + +def labelcolormap(N): + if N == 35: # cityscape + cmap = np.array([( 0, 0, 0), ( 0, 0, 0), ( 0, 0, 0), ( 0, 0, 0), ( 0, 0, 0), (111, 74, 0), ( 81, 0, 81), + (128, 64,128), (244, 35,232), (250,170,160), (230,150,140), ( 70, 70, 70), (102,102,156), (190,153,153), + (180,165,180), (150,100,100), (150,120, 90), (153,153,153), (153,153,153), (250,170, 30), (220,220, 0), + (107,142, 35), (152,251,152), ( 70,130,180), (220, 20, 60), (255, 0, 0), ( 0, 0,142), ( 0, 0, 70), + ( 0, 60,100), ( 0, 0, 90), ( 0, 0,110), ( 0, 80,100), ( 0, 0,230), (119, 11, 32), ( 0, 0,142)], + dtype=np.uint8) + else: + cmap = np.zeros((N, 3), dtype=np.uint8) + for i in range(N): + r, g, b = 0, 0, 0 + id = i + for j in range(7): + str_id = uint82bin(id) + r = r ^ (np.uint8(str_id[-1]) << (7-j)) + g = g ^ (np.uint8(str_id[-2]) << (7-j)) + b = b ^ (np.uint8(str_id[-3]) << (7-j)) + id = id >> 3 + cmap[i, 0] = r + cmap[i, 1] = g + cmap[i, 2] = b + return cmap + +class Colorize(object): + def __init__(self, n=35): + self.cmap = labelcolormap(n) + self.cmap = torch.from_numpy(self.cmap[:n]) + + def __call__(self, gray_image): + size = gray_image.size() + color_image = torch.ByteTensor(3, size[1], size[2]).fill_(0) + + for label in range(0, len(self.cmap)): + mask = (label == gray_image[0]).cpu() + color_image[0][mask] = self.cmap[label][0] + color_image[1][mask] = self.cmap[label][1] + color_image[2][mask] = self.cmap[label][2] + + return color_image diff --git a/mlflow/registry/Swimswap/util/videoswap.py b/mlflow/registry/Swimswap/util/videoswap.py new file mode 100644 index 0000000..99294d5 --- /dev/null +++ b/mlflow/registry/Swimswap/util/videoswap.py @@ -0,0 +1,121 @@ +''' +Author: Naiyuan liu +Github: https://github.com/NNNNAI +Date: 2021-11-23 17:03:58 +LastEditors: Naiyuan liu +LastEditTime: 2021-11-24 19:19:52 +Description: +''' +import os +import cv2 +import glob +import torch +import shutil +import numpy as np +from tqdm import tqdm +from util.reverse2original import reverse2wholeimage +import moviepy.editor as mp +from moviepy.editor import AudioFileClip, VideoFileClip +from moviepy.video.io.ImageSequenceClip import ImageSequenceClip +import time +from util.add_watermark import watermark_image +from util.norm import SpecificNorm +from parsing_model.model import BiSeNet + +def _totensor(array): + tensor = torch.from_numpy(array) + img = tensor.transpose(0, 1).transpose(0, 2).contiguous() + return img.float().div(255) + +def video_swap(video_path, id_vetor, swap_model, detect_model, save_path, temp_results_dir='./temp_results', crop_size=224, no_simswaplogo = False,use_mask =False): + video_forcheck = VideoFileClip(video_path) + if video_forcheck.audio is None: + no_audio = True + else: + no_audio = False + + del video_forcheck + + if not no_audio: + video_audio_clip = AudioFileClip(video_path) + + video = cv2.VideoCapture(video_path) + logoclass = watermark_image('./simswaplogo/simswaplogo.png') + ret = True + frame_index = 0 + + frame_count = int(video.get(cv2.CAP_PROP_FRAME_COUNT)) + + # video_WIDTH = int(video.get(cv2.CAP_PROP_FRAME_WIDTH)) + + # video_HEIGHT = int(video.get(cv2.CAP_PROP_FRAME_HEIGHT)) + + fps = video.get(cv2.CAP_PROP_FPS) + if os.path.exists(temp_results_dir): + shutil.rmtree(temp_results_dir) + + spNorm =SpecificNorm() + if use_mask: + n_classes = 19 + net = BiSeNet(n_classes=n_classes) + net.cuda() + save_pth = os.path.join('./parsing_model/checkpoint', '79999_iter.pth') + net.load_state_dict(torch.load(save_pth)) + net.eval() + else: + net =None + + # while ret: + for frame_index in tqdm(range(frame_count)): + ret, frame = video.read() + if ret: + detect_results = detect_model.get(frame,crop_size) + + if detect_results is not None: + # print(frame_index) + if not os.path.exists(temp_results_dir): + os.mkdir(temp_results_dir) + frame_align_crop_list = detect_results[0] + frame_mat_list = detect_results[1] + swap_result_list = [] + frame_align_crop_tenor_list = [] + for frame_align_crop in frame_align_crop_list: + + # BGR TO RGB + # frame_align_crop_RGB = frame_align_crop[...,::-1] + + frame_align_crop_tenor = _totensor(cv2.cvtColor(frame_align_crop,cv2.COLOR_BGR2RGB))[None,...].cuda() + + swap_result = swap_model(None, frame_align_crop_tenor, id_vetor, None, True)[0] + cv2.imwrite(os.path.join(temp_results_dir, 'frame_{:0>7d}.jpg'.format(frame_index)), frame) + swap_result_list.append(swap_result) + frame_align_crop_tenor_list.append(frame_align_crop_tenor) + + + + reverse2wholeimage(frame_align_crop_tenor_list,swap_result_list, frame_mat_list, crop_size, frame, logoclass,\ + os.path.join(temp_results_dir, 'frame_{:0>7d}.jpg'.format(frame_index)),no_simswaplogo,pasring_model =net,use_mask=use_mask, norm = spNorm) + + else: + if not os.path.exists(temp_results_dir): + os.mkdir(temp_results_dir) + frame = frame.astype(np.uint8) + if not no_simswaplogo: + frame = logoclass.apply_frames(frame) + cv2.imwrite(os.path.join(temp_results_dir, 'frame_{:0>7d}.jpg'.format(frame_index)), frame) + else: + break + + video.release() + + # image_filename_list = [] + path = os.path.join(temp_results_dir,'*.jpg') + image_filenames = sorted(glob.glob(path)) + + clips = ImageSequenceClip(image_filenames,fps = fps) + + if not no_audio: + clips = clips.set_audio(video_audio_clip) + + + clips.write_videofile(save_path,audio_codec='aac') diff --git a/mlflow/registry/Swimswap/util/videoswap_multispecific.py b/mlflow/registry/Swimswap/util/videoswap_multispecific.py new file mode 100644 index 0000000..40e9488 --- /dev/null +++ b/mlflow/registry/Swimswap/util/videoswap_multispecific.py @@ -0,0 +1,146 @@ +import os +import cv2 +import glob +import torch +import shutil +import numpy as np +from tqdm import tqdm +from util.reverse2original import reverse2wholeimage +import moviepy.editor as mp +from moviepy.editor import AudioFileClip, VideoFileClip +from moviepy.video.io.ImageSequenceClip import ImageSequenceClip +import time +from util.add_watermark import watermark_image +from util.norm import SpecificNorm +import torch.nn.functional as F +from parsing_model.model import BiSeNet + +def _totensor(array): + tensor = torch.from_numpy(array) + img = tensor.transpose(0, 1).transpose(0, 2).contiguous() + return img.float().div(255) + +def video_swap(video_path, target_id_norm_list,source_specific_id_nonorm_list,id_thres, swap_model, detect_model, save_path, temp_results_dir='./temp_results', crop_size=224, no_simswaplogo = False,use_mask =False): + video_forcheck = VideoFileClip(video_path) + if video_forcheck.audio is None: + no_audio = True + else: + no_audio = False + + del video_forcheck + + if not no_audio: + video_audio_clip = AudioFileClip(video_path) + + video = cv2.VideoCapture(video_path) + logoclass = watermark_image('./simswaplogo/simswaplogo.png') + ret = True + frame_index = 0 + + frame_count = int(video.get(cv2.CAP_PROP_FRAME_COUNT)) + + # video_WIDTH = int(video.get(cv2.CAP_PROP_FRAME_WIDTH)) + + # video_HEIGHT = int(video.get(cv2.CAP_PROP_FRAME_HEIGHT)) + + fps = video.get(cv2.CAP_PROP_FPS) + if os.path.exists(temp_results_dir): + shutil.rmtree(temp_results_dir) + + spNorm =SpecificNorm() + mse = torch.nn.MSELoss().cuda() + + if use_mask: + n_classes = 19 + net = BiSeNet(n_classes=n_classes) + net.cuda() + save_pth = os.path.join('./parsing_model/checkpoint', '79999_iter.pth') + net.load_state_dict(torch.load(save_pth)) + net.eval() + else: + net =None + + # while ret: + for frame_index in tqdm(range(frame_count)): + ret, frame = video.read() + if ret: + detect_results = detect_model.get(frame,crop_size) + + if detect_results is not None: + # print(frame_index) + if not os.path.exists(temp_results_dir): + os.mkdir(temp_results_dir) + frame_align_crop_list = detect_results[0] + frame_mat_list = detect_results[1] + + id_compare_values = [] + frame_align_crop_tenor_list = [] + for frame_align_crop in frame_align_crop_list: + + # BGR TO RGB + # frame_align_crop_RGB = frame_align_crop[...,::-1] + + frame_align_crop_tenor = _totensor(cv2.cvtColor(frame_align_crop,cv2.COLOR_BGR2RGB))[None,...].cuda() + + frame_align_crop_tenor_arcnorm = spNorm(frame_align_crop_tenor) + frame_align_crop_tenor_arcnorm_downsample = F.interpolate(frame_align_crop_tenor_arcnorm, size=(112,112)) + frame_align_crop_crop_id_nonorm = swap_model.netArc(frame_align_crop_tenor_arcnorm_downsample) + id_compare_values.append([]) + for source_specific_id_nonorm_tmp in source_specific_id_nonorm_list: + id_compare_values[-1].append(mse(frame_align_crop_crop_id_nonorm,source_specific_id_nonorm_tmp).detach().cpu().numpy()) + frame_align_crop_tenor_list.append(frame_align_crop_tenor) + + id_compare_values_array = np.array(id_compare_values).transpose(1,0) + min_indexs = np.argmin(id_compare_values_array,axis=0) + min_value = np.min(id_compare_values_array,axis=0) + + swap_result_list = [] + swap_result_matrix_list = [] + swap_result_ori_pic_list = [] + for tmp_index, min_index in enumerate(min_indexs): + if min_value[tmp_index] < id_thres: + swap_result = swap_model(None, frame_align_crop_tenor_list[tmp_index], target_id_norm_list[min_index], None, True)[0] + swap_result_list.append(swap_result) + swap_result_matrix_list.append(frame_mat_list[tmp_index]) + swap_result_ori_pic_list.append(frame_align_crop_tenor_list[tmp_index]) + else: + pass + + + + if len(swap_result_list) !=0: + + reverse2wholeimage(swap_result_ori_pic_list,swap_result_list, swap_result_matrix_list, crop_size, frame, logoclass,\ + os.path.join(temp_results_dir, 'frame_{:0>7d}.jpg'.format(frame_index)),no_simswaplogo,pasring_model =net,use_mask=use_mask, norm = spNorm) + else: + if not os.path.exists(temp_results_dir): + os.mkdir(temp_results_dir) + frame = frame.astype(np.uint8) + if not no_simswaplogo: + frame = logoclass.apply_frames(frame) + cv2.imwrite(os.path.join(temp_results_dir, 'frame_{:0>7d}.jpg'.format(frame_index)), frame) + + else: + if not os.path.exists(temp_results_dir): + os.mkdir(temp_results_dir) + frame = frame.astype(np.uint8) + if not no_simswaplogo: + frame = logoclass.apply_frames(frame) + cv2.imwrite(os.path.join(temp_results_dir, 'frame_{:0>7d}.jpg'.format(frame_index)), frame) + else: + break + + video.release() + + # image_filename_list = [] + path = os.path.join(temp_results_dir,'*.jpg') + image_filenames = sorted(glob.glob(path)) + + clips = ImageSequenceClip(image_filenames,fps = fps) + + if not no_audio: + clips = clips.set_audio(video_audio_clip) + + + clips.write_videofile(save_path,audio_codec='aac') + diff --git a/mlflow/registry/Swimswap/util/videoswap_specific.py b/mlflow/registry/Swimswap/util/videoswap_specific.py new file mode 100644 index 0000000..8177efb --- /dev/null +++ b/mlflow/registry/Swimswap/util/videoswap_specific.py @@ -0,0 +1,130 @@ +import os +import cv2 +import glob +import torch +import shutil +import numpy as np +from tqdm import tqdm +from util.reverse2original import reverse2wholeimage +import moviepy.editor as mp +from moviepy.editor import AudioFileClip, VideoFileClip +from moviepy.video.io.ImageSequenceClip import ImageSequenceClip +import time +from util.add_watermark import watermark_image +from util.norm import SpecificNorm +import torch.nn.functional as F +from parsing_model.model import BiSeNet + +def _totensor(array): + tensor = torch.from_numpy(array) + img = tensor.transpose(0, 1).transpose(0, 2).contiguous() + return img.float().div(255) + +def video_swap(video_path, id_vetor,specific_person_id_nonorm,id_thres, swap_model, detect_model, save_path, temp_results_dir='./temp_results', crop_size=224, no_simswaplogo = False,use_mask =False): + video_forcheck = VideoFileClip(video_path) + if video_forcheck.audio is None: + no_audio = True + else: + no_audio = False + + del video_forcheck + + if not no_audio: + video_audio_clip = AudioFileClip(video_path) + + video = cv2.VideoCapture(video_path) + logoclass = watermark_image('./simswaplogo/simswaplogo.png') + ret = True + frame_index = 0 + + frame_count = int(video.get(cv2.CAP_PROP_FRAME_COUNT)) + + # video_WIDTH = int(video.get(cv2.CAP_PROP_FRAME_WIDTH)) + + # video_HEIGHT = int(video.get(cv2.CAP_PROP_FRAME_HEIGHT)) + + fps = video.get(cv2.CAP_PROP_FPS) + if os.path.exists(temp_results_dir): + shutil.rmtree(temp_results_dir) + + spNorm =SpecificNorm() + mse = torch.nn.MSELoss().cuda() + + if use_mask: + n_classes = 19 + net = BiSeNet(n_classes=n_classes) + net.cuda() + save_pth = os.path.join('./parsing_model/checkpoint', '79999_iter.pth') + net.load_state_dict(torch.load(save_pth)) + net.eval() + else: + net =None + + # while ret: + for frame_index in tqdm(range(frame_count)): + ret, frame = video.read() + if ret: + detect_results = detect_model.get(frame,crop_size) + + if detect_results is not None: + # print(frame_index) + if not os.path.exists(temp_results_dir): + os.mkdir(temp_results_dir) + frame_align_crop_list = detect_results[0] + frame_mat_list = detect_results[1] + + id_compare_values = [] + frame_align_crop_tenor_list = [] + for frame_align_crop in frame_align_crop_list: + + # BGR TO RGB + # frame_align_crop_RGB = frame_align_crop[...,::-1] + + frame_align_crop_tenor = _totensor(cv2.cvtColor(frame_align_crop,cv2.COLOR_BGR2RGB))[None,...].cuda() + + frame_align_crop_tenor_arcnorm = spNorm(frame_align_crop_tenor) + frame_align_crop_tenor_arcnorm_downsample = F.interpolate(frame_align_crop_tenor_arcnorm, size=(112,112)) + frame_align_crop_crop_id_nonorm = swap_model.netArc(frame_align_crop_tenor_arcnorm_downsample) + + id_compare_values.append(mse(frame_align_crop_crop_id_nonorm,specific_person_id_nonorm).detach().cpu().numpy()) + frame_align_crop_tenor_list.append(frame_align_crop_tenor) + id_compare_values_array = np.array(id_compare_values) + min_index = np.argmin(id_compare_values_array) + min_value = id_compare_values_array[min_index] + if min_value < id_thres: + swap_result = swap_model(None, frame_align_crop_tenor_list[min_index], id_vetor, None, True)[0] + + reverse2wholeimage([frame_align_crop_tenor_list[min_index]], [swap_result], [frame_mat_list[min_index]], crop_size, frame, logoclass,\ + os.path.join(temp_results_dir, 'frame_{:0>7d}.jpg'.format(frame_index)),no_simswaplogo,pasring_model =net,use_mask= use_mask, norm = spNorm) + else: + if not os.path.exists(temp_results_dir): + os.mkdir(temp_results_dir) + frame = frame.astype(np.uint8) + if not no_simswaplogo: + frame = logoclass.apply_frames(frame) + cv2.imwrite(os.path.join(temp_results_dir, 'frame_{:0>7d}.jpg'.format(frame_index)), frame) + + else: + if not os.path.exists(temp_results_dir): + os.mkdir(temp_results_dir) + frame = frame.astype(np.uint8) + if not no_simswaplogo: + frame = logoclass.apply_frames(frame) + cv2.imwrite(os.path.join(temp_results_dir, 'frame_{:0>7d}.jpg'.format(frame_index)), frame) + else: + break + + video.release() + + # image_filename_list = [] + path = os.path.join(temp_results_dir,'*.jpg') + image_filenames = sorted(glob.glob(path)) + + clips = ImageSequenceClip(image_filenames,fps = fps) + + if not no_audio: + clips = clips.set_audio(video_audio_clip) + + + clips.write_videofile(save_path,audio_codec='aac') + diff --git a/mlflow/registry/Swimswap/util/visualizer.py b/mlflow/registry/Swimswap/util/visualizer.py new file mode 100644 index 0000000..584ac45 --- /dev/null +++ b/mlflow/registry/Swimswap/util/visualizer.py @@ -0,0 +1,131 @@ +import numpy as np +import os +import ntpath +import time +from . import util +from . import html +import scipy.misc +try: + from StringIO import StringIO # Python 2.7 +except ImportError: + from io import BytesIO # Python 3.x + +class Visualizer(): + def __init__(self, opt): + # self.opt = opt + self.tf_log = opt.tf_log + self.use_html = opt.isTrain and not opt.no_html + self.win_size = opt.display_winsize + self.name = opt.name + if self.tf_log: + import tensorflow as tf + self.tf = tf + self.log_dir = os.path.join(opt.checkpoints_dir, opt.name, 'logs') + self.writer = tf.summary.FileWriter(self.log_dir) + + if self.use_html: + self.web_dir = os.path.join(opt.checkpoints_dir, opt.name, 'web') + self.img_dir = os.path.join(self.web_dir, 'images') + print('create web directory %s...' % self.web_dir) + util.mkdirs([self.web_dir, self.img_dir]) + self.log_name = os.path.join(opt.checkpoints_dir, opt.name, 'loss_log.txt') + with open(self.log_name, "a") as log_file: + now = time.strftime("%c") + log_file.write('================ Training Loss (%s) ================\n' % now) + + # |visuals|: dictionary of images to display or save + def display_current_results(self, visuals, epoch, step): + if self.tf_log: # show images in tensorboard output + img_summaries = [] + for label, image_numpy in visuals.items(): + # Write the image to a string + try: + s = StringIO() + except: + s = BytesIO() + scipy.misc.toimage(image_numpy).save(s, format="jpeg") + # Create an Image object + img_sum = self.tf.Summary.Image(encoded_image_string=s.getvalue(), height=image_numpy.shape[0], width=image_numpy.shape[1]) + # Create a Summary value + img_summaries.append(self.tf.Summary.Value(tag=label, image=img_sum)) + + # Create and write Summary + summary = self.tf.Summary(value=img_summaries) + self.writer.add_summary(summary, step) + + if self.use_html: # save images to a html file + for label, image_numpy in visuals.items(): + if isinstance(image_numpy, list): + for i in range(len(image_numpy)): + img_path = os.path.join(self.img_dir, 'epoch%.3d_%s_%d.jpg' % (epoch, label, i)) + util.save_image(image_numpy[i], img_path) + else: + img_path = os.path.join(self.img_dir, 'epoch%.3d_%s.jpg' % (epoch, label)) + util.save_image(image_numpy, img_path) + + # update website + webpage = html.HTML(self.web_dir, 'Experiment name = %s' % self.name, refresh=30) + for n in range(epoch, 0, -1): + webpage.add_header('epoch [%d]' % n) + ims = [] + txts = [] + links = [] + + for label, image_numpy in visuals.items(): + if isinstance(image_numpy, list): + for i in range(len(image_numpy)): + img_path = 'epoch%.3d_%s_%d.jpg' % (n, label, i) + ims.append(img_path) + txts.append(label+str(i)) + links.append(img_path) + else: + img_path = 'epoch%.3d_%s.jpg' % (n, label) + ims.append(img_path) + txts.append(label) + links.append(img_path) + if len(ims) < 10: + webpage.add_images(ims, txts, links, width=self.win_size) + else: + num = int(round(len(ims)/2.0)) + webpage.add_images(ims[:num], txts[:num], links[:num], width=self.win_size) + webpage.add_images(ims[num:], txts[num:], links[num:], width=self.win_size) + webpage.save() + + # errors: dictionary of error labels and values + def plot_current_errors(self, errors, step): + if self.tf_log: + for tag, value in errors.items(): + summary = self.tf.Summary(value=[self.tf.Summary.Value(tag=tag, simple_value=value)]) + self.writer.add_summary(summary, step) + + # errors: same format as |errors| of plotCurrentErrors + def print_current_errors(self, epoch, i, errors, t): + message = '(epoch: %d, iters: %d, time: %.3f) ' % (epoch, i, t) + for k, v in errors.items(): + if v != 0: + message += '%s: %.3f ' % (k, v) + + print(message) + with open(self.log_name, "a") as log_file: + log_file.write('%s\n' % message) + + # save image to the disk + def save_images(self, webpage, visuals, image_path): + image_dir = webpage.get_image_dir() + short_path = ntpath.basename(image_path[0]) + name = os.path.splitext(short_path)[0] + + webpage.add_header(name) + ims = [] + txts = [] + links = [] + + for label, image_numpy in visuals.items(): + image_name = '%s_%s.jpg' % (name, label) + save_path = os.path.join(image_dir, image_name) + util.save_image(image_numpy, save_path) + + ims.append(image_name) + txts.append(label) + links.append(image_name) + webpage.add_images(ims, txts, links, width=self.win_size) diff --git a/train/mlflow/sf2f/GETTING_STARTED.md b/mlflow/train/sf2f/GETTING_STARTED.md similarity index 100% rename from train/mlflow/sf2f/GETTING_STARTED.md rename to mlflow/train/sf2f/GETTING_STARTED.md diff --git "a/train/mlflow/sf2f/data/\353\205\271\354\235\214.wav" "b/mlflow/train/sf2f/data/\353\205\271\354\235\214.wav" similarity index 100% rename from "train/mlflow/sf2f/data/\353\205\271\354\235\214.wav" rename to "mlflow/train/sf2f/data/\353\205\271\354\235\214.wav" diff --git a/train/mlflow/sf2f/datasets/__init__.py b/mlflow/train/sf2f/datasets/__init__.py similarity index 100% rename from train/mlflow/sf2f/datasets/__init__.py rename to mlflow/train/sf2f/datasets/__init__.py diff --git a/train/mlflow/sf2f/datasets/build_dataset.py b/mlflow/train/sf2f/datasets/build_dataset.py similarity index 100% rename from train/mlflow/sf2f/datasets/build_dataset.py rename to mlflow/train/sf2f/datasets/build_dataset.py diff --git a/train/mlflow/sf2f/datasets/utils.py b/mlflow/train/sf2f/datasets/utils.py similarity index 100% rename from train/mlflow/sf2f/datasets/utils.py rename to mlflow/train/sf2f/datasets/utils.py diff --git a/train/mlflow/sf2f/datasets/vox_dataset.py b/mlflow/train/sf2f/datasets/vox_dataset.py similarity index 100% rename from train/mlflow/sf2f/datasets/vox_dataset.py rename to mlflow/train/sf2f/datasets/vox_dataset.py diff --git a/train/mlflow/sf2f/infer.py b/mlflow/train/sf2f/infer.py similarity index 100% rename from train/mlflow/sf2f/infer.py rename to mlflow/train/sf2f/infer.py diff --git a/train/mlflow/sf2f/inference_fuser.py b/mlflow/train/sf2f/inference_fuser.py similarity index 100% rename from train/mlflow/sf2f/inference_fuser.py rename to mlflow/train/sf2f/inference_fuser.py diff --git a/train/mlflow/sf2f/model_registry.py b/mlflow/train/sf2f/model_registry.py similarity index 86% rename from train/mlflow/sf2f/model_registry.py rename to mlflow/train/sf2f/model_registry.py index cb82b57..0140dfe 100644 --- a/train/mlflow/sf2f/model_registry.py +++ b/mlflow/train/sf2f/model_registry.py @@ -7,18 +7,18 @@ from datasets import imagenet_deprocess_batch, set_mel_transform, \ deprocess_and_save, window_segment -os.environ["MLFLOW_S3_ENDPOINT_URL"] = "http://localhost:9000" -os.environ["MLFLOW_TRACKING_URI"] = "http://localhost:5001" +os.environ["MLFLOW_S3_ENDPOINT_URL"] = "http://223.130.133.236:9000" +os.environ["MLFLOW_TRACKING_URI"] = "http://223.130.133.236:5001" os.environ["AWS_ACCESS_KEY_ID"] = "minio" os.environ["AWS_SECRET_ACCESS_KEY"] = "miniostorage" mlflow.set_experiment("new-exp") model, _ = models.build_model( options["generator"], image_size=[128,128], - checkpoint_start_from="/home/hojun/Documents/project/boostcamp/final_project/mlops/voice2face-modeling/sf2f/data/best_IS_with_model.pt") + checkpoint_start_from="/home/hojun/Documents/project/boostcamp/final_project/mlops/temp/voice2face-mlops/best_IS_with_model.pt") model.cuda().eval() -voice_path = os.path.join("/home/hojun/Documents/project/boostcamp/final_project/mlops/voice2face-modeling/sf2f/data", '*.wav') +voice_path = os.path.join("/home/hojun/Documents/project/boostcamp/final_project/mlops/mlflow/train/mlflow/sf2f/data/", '*.wav') voice_list = glob.glob(voice_path) filename = voice_list[0] print(filename) @@ -61,5 +61,5 @@ artifact_path = "sf2f_pytorch", signature= signature, input_example = input_sample, - pip_requirements = "rec.txt" + # pip_requirements = "rec.txt" ) \ No newline at end of file diff --git a/train/mlflow/sf2f/models/__init__.py b/mlflow/train/sf2f/models/__init__.py similarity index 100% rename from train/mlflow/sf2f/models/__init__.py rename to mlflow/train/sf2f/models/__init__.py diff --git a/train/mlflow/sf2f/models/attention.py b/mlflow/train/sf2f/models/attention.py similarity index 100% rename from train/mlflow/sf2f/models/attention.py rename to mlflow/train/sf2f/models/attention.py diff --git a/train/mlflow/sf2f/models/crn.py b/mlflow/train/sf2f/models/crn.py similarity index 100% rename from train/mlflow/sf2f/models/crn.py rename to mlflow/train/sf2f/models/crn.py diff --git a/train/mlflow/sf2f/models/discriminators.py b/mlflow/train/sf2f/models/discriminators.py similarity index 100% rename from train/mlflow/sf2f/models/discriminators.py rename to mlflow/train/sf2f/models/discriminators.py diff --git a/train/mlflow/sf2f/models/encoder_decoder.py b/mlflow/train/sf2f/models/encoder_decoder.py similarity index 100% rename from train/mlflow/sf2f/models/encoder_decoder.py rename to mlflow/train/sf2f/models/encoder_decoder.py diff --git a/train/mlflow/sf2f/models/face_decoders.py b/mlflow/train/sf2f/models/face_decoders.py similarity index 100% rename from train/mlflow/sf2f/models/face_decoders.py rename to mlflow/train/sf2f/models/face_decoders.py diff --git a/train/mlflow/sf2f/models/fusers.py b/mlflow/train/sf2f/models/fusers.py similarity index 100% rename from train/mlflow/sf2f/models/fusers.py rename to mlflow/train/sf2f/models/fusers.py diff --git a/train/mlflow/sf2f/models/inception_resnet_v1.py b/mlflow/train/sf2f/models/inception_resnet_v1.py similarity index 100% rename from train/mlflow/sf2f/models/inception_resnet_v1.py rename to mlflow/train/sf2f/models/inception_resnet_v1.py diff --git a/train/mlflow/sf2f/models/layers.py b/mlflow/train/sf2f/models/layers.py similarity index 100% rename from train/mlflow/sf2f/models/layers.py rename to mlflow/train/sf2f/models/layers.py diff --git a/train/mlflow/sf2f/models/model_collection.py b/mlflow/train/sf2f/models/model_collection.py similarity index 100% rename from train/mlflow/sf2f/models/model_collection.py rename to mlflow/train/sf2f/models/model_collection.py diff --git a/train/mlflow/sf2f/models/model_setup.py b/mlflow/train/sf2f/models/model_setup.py similarity index 100% rename from train/mlflow/sf2f/models/model_setup.py rename to mlflow/train/sf2f/models/model_setup.py diff --git a/train/mlflow/sf2f/models/networks.py b/mlflow/train/sf2f/models/networks.py similarity index 100% rename from train/mlflow/sf2f/models/networks.py rename to mlflow/train/sf2f/models/networks.py diff --git a/train/mlflow/sf2f/models/perceptual.py b/mlflow/train/sf2f/models/perceptual.py similarity index 100% rename from train/mlflow/sf2f/models/perceptual.py rename to mlflow/train/sf2f/models/perceptual.py diff --git a/train/mlflow/sf2f/models/voice_encoders.py b/mlflow/train/sf2f/models/voice_encoders.py similarity index 100% rename from train/mlflow/sf2f/models/voice_encoders.py rename to mlflow/train/sf2f/models/voice_encoders.py diff --git a/train/mlflow/sf2f/scripts/DScore/__init__.py b/mlflow/train/sf2f/options/__init__.py similarity index 100% rename from train/mlflow/sf2f/scripts/DScore/__init__.py rename to mlflow/train/sf2f/options/__init__.py diff --git a/train/mlflow/sf2f/options/data_opts/vox.yaml b/mlflow/train/sf2f/options/data_opts/vox.yaml similarity index 100% rename from train/mlflow/sf2f/options/data_opts/vox.yaml rename to mlflow/train/sf2f/options/data_opts/vox.yaml diff --git a/train/mlflow/sf2f/options/opts.py b/mlflow/train/sf2f/options/opts.py similarity index 100% rename from train/mlflow/sf2f/options/opts.py rename to mlflow/train/sf2f/options/opts.py diff --git a/train/mlflow/sf2f/options/vox/baseline/v2f.yaml b/mlflow/train/sf2f/options/vox/baseline/v2f.yaml similarity index 100% rename from train/mlflow/sf2f/options/vox/baseline/v2f.yaml rename to mlflow/train/sf2f/options/vox/baseline/v2f.yaml diff --git a/train/mlflow/sf2f/options/vox/sf2f/sf2f_1st_stage.yaml b/mlflow/train/sf2f/options/vox/sf2f/sf2f_1st_stage.yaml similarity index 100% rename from train/mlflow/sf2f/options/vox/sf2f/sf2f_1st_stage.yaml rename to mlflow/train/sf2f/options/vox/sf2f/sf2f_1st_stage.yaml diff --git a/train/mlflow/sf2f/options/vox/sf2f/sf2f_fuser.yaml b/mlflow/train/sf2f/options/vox/sf2f/sf2f_fuser.yaml similarity index 96% rename from train/mlflow/sf2f/options/vox/sf2f/sf2f_fuser.yaml rename to mlflow/train/sf2f/options/vox/sf2f/sf2f_fuser.yaml index a414da5..c354661 100644 --- a/train/mlflow/sf2f/options/vox/sf2f/sf2f_fuser.yaml +++ b/mlflow/train/sf2f/options/vox/sf2f/sf2f_fuser.yaml @@ -8,7 +8,7 @@ logs: # data-related settings data: dataset: vox - data_opts_path: /home/hojun/Documents/project/boostcamp/final_project/mlops/voice2face-modeling/sf2f/options/data_opts/vox.yaml + data_opts_path: /home/hojun/Documents/project/boostcamp/final_project/mlops/mlflow/train/mlflow/sf2f/options/data_opts/vox.yaml image_size: [64, 64] # model related settings generator: diff --git a/train/mlflow/sf2f/options/vox/sf2f/sf2f_mid_1st_stage.yaml b/mlflow/train/sf2f/options/vox/sf2f/sf2f_mid_1st_stage.yaml similarity index 100% rename from train/mlflow/sf2f/options/vox/sf2f/sf2f_mid_1st_stage.yaml rename to mlflow/train/sf2f/options/vox/sf2f/sf2f_mid_1st_stage.yaml diff --git a/train/mlflow/sf2f/options/vox/sf2f/sf2f_mid_fuser.yaml b/mlflow/train/sf2f/options/vox/sf2f/sf2f_mid_fuser.yaml similarity index 100% rename from train/mlflow/sf2f/options/vox/sf2f/sf2f_mid_fuser.yaml rename to mlflow/train/sf2f/options/vox/sf2f/sf2f_mid_fuser.yaml diff --git a/train/mlflow/sf2f/output/sf2f_1st_stage-20240303-215640/sf2f_1st_stage.yaml b/mlflow/train/sf2f/output/sf2f_1st_stage-20240303-215640/sf2f_1st_stage.yaml similarity index 100% rename from train/mlflow/sf2f/output/sf2f_1st_stage-20240303-215640/sf2f_1st_stage.yaml rename to mlflow/train/sf2f/output/sf2f_1st_stage-20240303-215640/sf2f_1st_stage.yaml diff --git a/train/mlflow/sf2f/output/sf2f_1st_stage-20240303-220349/sf2f_1st_stage.yaml b/mlflow/train/sf2f/output/sf2f_1st_stage-20240303-220349/sf2f_1st_stage.yaml similarity index 100% rename from train/mlflow/sf2f/output/sf2f_1st_stage-20240303-220349/sf2f_1st_stage.yaml rename to mlflow/train/sf2f/output/sf2f_1st_stage-20240303-220349/sf2f_1st_stage.yaml diff --git a/train/mlflow/sf2f/output/sf2f_1st_stage-20240303-221106/sf2f_1st_stage.yaml b/mlflow/train/sf2f/output/sf2f_1st_stage-20240303-221106/sf2f_1st_stage.yaml similarity index 100% rename from train/mlflow/sf2f/output/sf2f_1st_stage-20240303-221106/sf2f_1st_stage.yaml rename to mlflow/train/sf2f/output/sf2f_1st_stage-20240303-221106/sf2f_1st_stage.yaml diff --git a/train/mlflow/sf2f/output/sf2f_1st_stage-20240303-225403/sf2f_1st_stage.yaml b/mlflow/train/sf2f/output/sf2f_1st_stage-20240303-225403/sf2f_1st_stage.yaml similarity index 100% rename from train/mlflow/sf2f/output/sf2f_1st_stage-20240303-225403/sf2f_1st_stage.yaml rename to mlflow/train/sf2f/output/sf2f_1st_stage-20240303-225403/sf2f_1st_stage.yaml diff --git a/train/mlflow/sf2f/output/sf2f_1st_stage-20240303-225533/sf2f_1st_stage.yaml b/mlflow/train/sf2f/output/sf2f_1st_stage-20240303-225533/sf2f_1st_stage.yaml similarity index 100% rename from train/mlflow/sf2f/output/sf2f_1st_stage-20240303-225533/sf2f_1st_stage.yaml rename to mlflow/train/sf2f/output/sf2f_1st_stage-20240303-225533/sf2f_1st_stage.yaml diff --git a/train/mlflow/sf2f/output/sf2f_1st_stage-20240303-225656/sf2f_1st_stage.yaml b/mlflow/train/sf2f/output/sf2f_1st_stage-20240303-225656/sf2f_1st_stage.yaml similarity index 100% rename from train/mlflow/sf2f/output/sf2f_1st_stage-20240303-225656/sf2f_1st_stage.yaml rename to mlflow/train/sf2f/output/sf2f_1st_stage-20240303-225656/sf2f_1st_stage.yaml diff --git a/train/mlflow/sf2f/output/sf2f_1st_stage-20240303-225950/sf2f_1st_stage.yaml b/mlflow/train/sf2f/output/sf2f_1st_stage-20240303-225950/sf2f_1st_stage.yaml similarity index 100% rename from train/mlflow/sf2f/output/sf2f_1st_stage-20240303-225950/sf2f_1st_stage.yaml rename to mlflow/train/sf2f/output/sf2f_1st_stage-20240303-225950/sf2f_1st_stage.yaml diff --git a/train/mlflow/sf2f/output/sf2f_1st_stage-20240303-230032/sf2f_1st_stage.yaml b/mlflow/train/sf2f/output/sf2f_1st_stage-20240303-230032/sf2f_1st_stage.yaml similarity index 100% rename from train/mlflow/sf2f/output/sf2f_1st_stage-20240303-230032/sf2f_1st_stage.yaml rename to mlflow/train/sf2f/output/sf2f_1st_stage-20240303-230032/sf2f_1st_stage.yaml diff --git a/train/mlflow/sf2f/output/sf2f_1st_stage-20240303-230142/sf2f_1st_stage.yaml b/mlflow/train/sf2f/output/sf2f_1st_stage-20240303-230142/sf2f_1st_stage.yaml similarity index 100% rename from train/mlflow/sf2f/output/sf2f_1st_stage-20240303-230142/sf2f_1st_stage.yaml rename to mlflow/train/sf2f/output/sf2f_1st_stage-20240303-230142/sf2f_1st_stage.yaml diff --git a/train/mlflow/sf2f/output/sf2f_1st_stage-20240303-230232/sf2f_1st_stage.yaml b/mlflow/train/sf2f/output/sf2f_1st_stage-20240303-230232/sf2f_1st_stage.yaml similarity index 100% rename from train/mlflow/sf2f/output/sf2f_1st_stage-20240303-230232/sf2f_1st_stage.yaml rename to mlflow/train/sf2f/output/sf2f_1st_stage-20240303-230232/sf2f_1st_stage.yaml diff --git a/train/mlflow/sf2f/output/sf2f_1st_stage-20240303-230355/sf2f_1st_stage.yaml b/mlflow/train/sf2f/output/sf2f_1st_stage-20240303-230355/sf2f_1st_stage.yaml similarity index 100% rename from train/mlflow/sf2f/output/sf2f_1st_stage-20240303-230355/sf2f_1st_stage.yaml rename to mlflow/train/sf2f/output/sf2f_1st_stage-20240303-230355/sf2f_1st_stage.yaml diff --git a/train/mlflow/sf2f/output/sf2f_1st_stage-20240303-230514/sf2f_1st_stage.yaml b/mlflow/train/sf2f/output/sf2f_1st_stage-20240303-230514/sf2f_1st_stage.yaml similarity index 100% rename from train/mlflow/sf2f/output/sf2f_1st_stage-20240303-230514/sf2f_1st_stage.yaml rename to mlflow/train/sf2f/output/sf2f_1st_stage-20240303-230514/sf2f_1st_stage.yaml diff --git a/train/mlflow/sf2f/output/sf2f_1st_stage-20240303-230631/sf2f_1st_stage.yaml b/mlflow/train/sf2f/output/sf2f_1st_stage-20240303-230631/sf2f_1st_stage.yaml similarity index 100% rename from train/mlflow/sf2f/output/sf2f_1st_stage-20240303-230631/sf2f_1st_stage.yaml rename to mlflow/train/sf2f/output/sf2f_1st_stage-20240303-230631/sf2f_1st_stage.yaml diff --git a/train/mlflow/sf2f/output/sf2f_1st_stage-20240303-231044/sf2f_1st_stage.yaml b/mlflow/train/sf2f/output/sf2f_1st_stage-20240303-231044/sf2f_1st_stage.yaml similarity index 100% rename from train/mlflow/sf2f/output/sf2f_1st_stage-20240303-231044/sf2f_1st_stage.yaml rename to mlflow/train/sf2f/output/sf2f_1st_stage-20240303-231044/sf2f_1st_stage.yaml diff --git a/train/mlflow/sf2f/output/sf2f_1st_stage-20240303-231201/sf2f_1st_stage.yaml b/mlflow/train/sf2f/output/sf2f_1st_stage-20240303-231201/sf2f_1st_stage.yaml similarity index 100% rename from train/mlflow/sf2f/output/sf2f_1st_stage-20240303-231201/sf2f_1st_stage.yaml rename to mlflow/train/sf2f/output/sf2f_1st_stage-20240303-231201/sf2f_1st_stage.yaml diff --git a/train/mlflow/sf2f/output/sf2f_1st_stage-20240303-231308/sf2f_1st_stage.yaml b/mlflow/train/sf2f/output/sf2f_1st_stage-20240303-231308/sf2f_1st_stage.yaml similarity index 100% rename from train/mlflow/sf2f/output/sf2f_1st_stage-20240303-231308/sf2f_1st_stage.yaml rename to mlflow/train/sf2f/output/sf2f_1st_stage-20240303-231308/sf2f_1st_stage.yaml diff --git a/train/mlflow/sf2f/output/sf2f_1st_stage-20240303-231542/sf2f_1st_stage.yaml b/mlflow/train/sf2f/output/sf2f_1st_stage-20240303-231542/sf2f_1st_stage.yaml similarity index 100% rename from train/mlflow/sf2f/output/sf2f_1st_stage-20240303-231542/sf2f_1st_stage.yaml rename to mlflow/train/sf2f/output/sf2f_1st_stage-20240303-231542/sf2f_1st_stage.yaml diff --git a/train/mlflow/sf2f/output/sf2f_1st_stage-20240303-231831/sf2f_1st_stage.yaml b/mlflow/train/sf2f/output/sf2f_1st_stage-20240303-231831/sf2f_1st_stage.yaml similarity index 100% rename from train/mlflow/sf2f/output/sf2f_1st_stage-20240303-231831/sf2f_1st_stage.yaml rename to mlflow/train/sf2f/output/sf2f_1st_stage-20240303-231831/sf2f_1st_stage.yaml diff --git a/train/mlflow/sf2f/output/sf2f_1st_stage-20240303-232325/sf2f_1st_stage.yaml b/mlflow/train/sf2f/output/sf2f_1st_stage-20240303-232325/sf2f_1st_stage.yaml similarity index 100% rename from train/mlflow/sf2f/output/sf2f_1st_stage-20240303-232325/sf2f_1st_stage.yaml rename to mlflow/train/sf2f/output/sf2f_1st_stage-20240303-232325/sf2f_1st_stage.yaml diff --git a/train/mlflow/sf2f/output/sf2f_1st_stage-20240303-233127/sf2f_1st_stage.yaml b/mlflow/train/sf2f/output/sf2f_1st_stage-20240303-233127/sf2f_1st_stage.yaml similarity index 100% rename from train/mlflow/sf2f/output/sf2f_1st_stage-20240303-233127/sf2f_1st_stage.yaml rename to mlflow/train/sf2f/output/sf2f_1st_stage-20240303-233127/sf2f_1st_stage.yaml diff --git a/train/mlflow/sf2f/output/sf2f_1st_stage-20240303-233311/sf2f_1st_stage.yaml b/mlflow/train/sf2f/output/sf2f_1st_stage-20240303-233311/sf2f_1st_stage.yaml similarity index 100% rename from train/mlflow/sf2f/output/sf2f_1st_stage-20240303-233311/sf2f_1st_stage.yaml rename to mlflow/train/sf2f/output/sf2f_1st_stage-20240303-233311/sf2f_1st_stage.yaml diff --git a/train/mlflow/sf2f/output/sf2f_1st_stage-20240303-233619/sf2f_1st_stage.yaml b/mlflow/train/sf2f/output/sf2f_1st_stage-20240303-233619/sf2f_1st_stage.yaml similarity index 100% rename from train/mlflow/sf2f/output/sf2f_1st_stage-20240303-233619/sf2f_1st_stage.yaml rename to mlflow/train/sf2f/output/sf2f_1st_stage-20240303-233619/sf2f_1st_stage.yaml diff --git a/train/mlflow/sf2f/scripts/DScore/models/__init__.py b/mlflow/train/sf2f/scripts/DScore/__init__.py similarity index 100% rename from train/mlflow/sf2f/scripts/DScore/models/__init__.py rename to mlflow/train/sf2f/scripts/DScore/__init__.py diff --git a/train/mlflow/sf2f/scripts/DScore/util/__init__.py b/mlflow/train/sf2f/scripts/DScore/models/__init__.py similarity index 100% rename from train/mlflow/sf2f/scripts/DScore/util/__init__.py rename to mlflow/train/sf2f/scripts/DScore/models/__init__.py diff --git a/train/mlflow/sf2f/scripts/DScore/models/base_model.py b/mlflow/train/sf2f/scripts/DScore/models/base_model.py similarity index 100% rename from train/mlflow/sf2f/scripts/DScore/models/base_model.py rename to mlflow/train/sf2f/scripts/DScore/models/base_model.py diff --git a/train/mlflow/sf2f/scripts/DScore/models/dist_model.py b/mlflow/train/sf2f/scripts/DScore/models/dist_model.py similarity index 100% rename from train/mlflow/sf2f/scripts/DScore/models/dist_model.py rename to mlflow/train/sf2f/scripts/DScore/models/dist_model.py diff --git a/train/mlflow/sf2f/scripts/DScore/models/models.py b/mlflow/train/sf2f/scripts/DScore/models/models.py similarity index 100% rename from train/mlflow/sf2f/scripts/DScore/models/models.py rename to mlflow/train/sf2f/scripts/DScore/models/models.py diff --git a/train/mlflow/sf2f/scripts/DScore/models/networks_basic.py b/mlflow/train/sf2f/scripts/DScore/models/networks_basic.py similarity index 100% rename from train/mlflow/sf2f/scripts/DScore/models/networks_basic.py rename to mlflow/train/sf2f/scripts/DScore/models/networks_basic.py diff --git a/train/mlflow/sf2f/scripts/DScore/models/pretrained_networks.py b/mlflow/train/sf2f/scripts/DScore/models/pretrained_networks.py similarity index 100% rename from train/mlflow/sf2f/scripts/DScore/models/pretrained_networks.py rename to mlflow/train/sf2f/scripts/DScore/models/pretrained_networks.py diff --git a/mlflow/train/sf2f/scripts/DScore/util/__init__.py b/mlflow/train/sf2f/scripts/DScore/util/__init__.py new file mode 100644 index 0000000..e69de29 diff --git a/train/mlflow/sf2f/scripts/DScore/util/html.py b/mlflow/train/sf2f/scripts/DScore/util/html.py similarity index 100% rename from train/mlflow/sf2f/scripts/DScore/util/html.py rename to mlflow/train/sf2f/scripts/DScore/util/html.py diff --git a/train/mlflow/sf2f/scripts/DScore/util/util.py b/mlflow/train/sf2f/scripts/DScore/util/util.py similarity index 100% rename from train/mlflow/sf2f/scripts/DScore/util/util.py rename to mlflow/train/sf2f/scripts/DScore/util/util.py diff --git a/train/mlflow/sf2f/scripts/DScore/util/visualizer.py b/mlflow/train/sf2f/scripts/DScore/util/visualizer.py similarity index 100% rename from train/mlflow/sf2f/scripts/DScore/util/visualizer.py rename to mlflow/train/sf2f/scripts/DScore/util/visualizer.py diff --git a/train/mlflow/sf2f/scripts/build_demo_set.py b/mlflow/train/sf2f/scripts/build_demo_set.py similarity index 100% rename from train/mlflow/sf2f/scripts/build_demo_set.py rename to mlflow/train/sf2f/scripts/build_demo_set.py diff --git a/train/mlflow/sf2f/scripts/compute_diversity_score.py b/mlflow/train/sf2f/scripts/compute_diversity_score.py similarity index 100% rename from train/mlflow/sf2f/scripts/compute_diversity_score.py rename to mlflow/train/sf2f/scripts/compute_diversity_score.py diff --git a/train/mlflow/sf2f/scripts/compute_fid_score.py b/mlflow/train/sf2f/scripts/compute_fid_score.py similarity index 100% rename from train/mlflow/sf2f/scripts/compute_fid_score.py rename to mlflow/train/sf2f/scripts/compute_fid_score.py diff --git a/train/mlflow/sf2f/scripts/compute_inception_score.py b/mlflow/train/sf2f/scripts/compute_inception_score.py similarity index 100% rename from train/mlflow/sf2f/scripts/compute_inception_score.py rename to mlflow/train/sf2f/scripts/compute_inception_score.py diff --git a/train/mlflow/sf2f/scripts/compute_mel_mean_var.py b/mlflow/train/sf2f/scripts/compute_mel_mean_var.py similarity index 100% rename from train/mlflow/sf2f/scripts/compute_mel_mean_var.py rename to mlflow/train/sf2f/scripts/compute_mel_mean_var.py diff --git a/train/mlflow/sf2f/scripts/compute_vggface_score.py b/mlflow/train/sf2f/scripts/compute_vggface_score.py similarity index 100% rename from train/mlflow/sf2f/scripts/compute_vggface_score.py rename to mlflow/train/sf2f/scripts/compute_vggface_score.py diff --git a/train/mlflow/sf2f/scripts/convert_wav_to_mel.py b/mlflow/train/sf2f/scripts/convert_wav_to_mel.py similarity index 100% rename from train/mlflow/sf2f/scripts/convert_wav_to_mel.py rename to mlflow/train/sf2f/scripts/convert_wav_to_mel.py diff --git a/train/mlflow/sf2f/scripts/create_split_json.py b/mlflow/train/sf2f/scripts/create_split_json.py similarity index 100% rename from train/mlflow/sf2f/scripts/create_split_json.py rename to mlflow/train/sf2f/scripts/create_split_json.py diff --git a/train/mlflow/sf2f/scripts/download_vggface_weights.sh b/mlflow/train/sf2f/scripts/download_vggface_weights.sh similarity index 100% rename from train/mlflow/sf2f/scripts/download_vggface_weights.sh rename to mlflow/train/sf2f/scripts/download_vggface_weights.sh diff --git a/train/mlflow/sf2f/scripts/install_requirements.py b/mlflow/train/sf2f/scripts/install_requirements.py similarity index 100% rename from train/mlflow/sf2f/scripts/install_requirements.py rename to mlflow/train/sf2f/scripts/install_requirements.py diff --git a/train/mlflow/sf2f/scripts/print_args.py b/mlflow/train/sf2f/scripts/print_args.py similarity index 100% rename from train/mlflow/sf2f/scripts/print_args.py rename to mlflow/train/sf2f/scripts/print_args.py diff --git a/train/mlflow/sf2f/scripts/sample_mel_spectrograms.py b/mlflow/train/sf2f/scripts/sample_mel_spectrograms.py similarity index 100% rename from train/mlflow/sf2f/scripts/sample_mel_spectrograms.py rename to mlflow/train/sf2f/scripts/sample_mel_spectrograms.py diff --git a/train/mlflow/sf2f/scripts/strip_checkpoint.py b/mlflow/train/sf2f/scripts/strip_checkpoint.py similarity index 100% rename from train/mlflow/sf2f/scripts/strip_checkpoint.py rename to mlflow/train/sf2f/scripts/strip_checkpoint.py diff --git a/train/mlflow/sf2f/scripts/strip_old_args.py b/mlflow/train/sf2f/scripts/strip_old_args.py similarity index 100% rename from train/mlflow/sf2f/scripts/strip_old_args.py rename to mlflow/train/sf2f/scripts/strip_old_args.py diff --git a/train/mlflow/sf2f/scripts/watch_data.py b/mlflow/train/sf2f/scripts/watch_data.py similarity index 100% rename from train/mlflow/sf2f/scripts/watch_data.py rename to mlflow/train/sf2f/scripts/watch_data.py diff --git a/train/mlflow/sf2f/train.py b/mlflow/train/sf2f/train.py similarity index 100% rename from train/mlflow/sf2f/train.py rename to mlflow/train/sf2f/train.py diff --git a/train/mlflow/sf2f/train_registry.py b/mlflow/train/sf2f/train_registry.py similarity index 100% rename from train/mlflow/sf2f/train_registry.py rename to mlflow/train/sf2f/train_registry.py diff --git a/train/mlflow/sf2f/utils/__init__.py b/mlflow/train/sf2f/utils/__init__.py similarity index 100% rename from train/mlflow/sf2f/utils/__init__.py rename to mlflow/train/sf2f/utils/__init__.py diff --git a/train/mlflow/sf2f/utils/bilinear.py b/mlflow/train/sf2f/utils/bilinear.py similarity index 100% rename from train/mlflow/sf2f/utils/bilinear.py rename to mlflow/train/sf2f/utils/bilinear.py diff --git a/train/mlflow/sf2f/utils/box_utils.py b/mlflow/train/sf2f/utils/box_utils.py similarity index 100% rename from train/mlflow/sf2f/utils/box_utils.py rename to mlflow/train/sf2f/utils/box_utils.py diff --git a/train/mlflow/sf2f/utils/common.py b/mlflow/train/sf2f/utils/common.py similarity index 100% rename from train/mlflow/sf2f/utils/common.py rename to mlflow/train/sf2f/utils/common.py diff --git a/train/mlflow/sf2f/utils/evaluate.py b/mlflow/train/sf2f/utils/evaluate.py similarity index 100% rename from train/mlflow/sf2f/utils/evaluate.py rename to mlflow/train/sf2f/utils/evaluate.py diff --git a/train/mlflow/sf2f/utils/evaluate_fid.py b/mlflow/train/sf2f/utils/evaluate_fid.py similarity index 100% rename from train/mlflow/sf2f/utils/evaluate_fid.py rename to mlflow/train/sf2f/utils/evaluate_fid.py diff --git a/train/mlflow/sf2f/utils/filter_pickle.py b/mlflow/train/sf2f/utils/filter_pickle.py similarity index 100% rename from train/mlflow/sf2f/utils/filter_pickle.py rename to mlflow/train/sf2f/utils/filter_pickle.py diff --git a/train/mlflow/sf2f/utils/logger.py b/mlflow/train/sf2f/utils/logger.py similarity index 100% rename from train/mlflow/sf2f/utils/logger.py rename to mlflow/train/sf2f/utils/logger.py diff --git a/train/mlflow/sf2f/utils/losses.py b/mlflow/train/sf2f/utils/losses.py similarity index 100% rename from train/mlflow/sf2f/utils/losses.py rename to mlflow/train/sf2f/utils/losses.py diff --git a/train/mlflow/sf2f/utils/metrics.py b/mlflow/train/sf2f/utils/metrics.py similarity index 100% rename from train/mlflow/sf2f/utils/metrics.py rename to mlflow/train/sf2f/utils/metrics.py diff --git a/train/mlflow/sf2f/utils/mlflow_wandb.py b/mlflow/train/sf2f/utils/mlflow_wandb.py similarity index 100% rename from train/mlflow/sf2f/utils/mlflow_wandb.py rename to mlflow/train/sf2f/utils/mlflow_wandb.py diff --git a/train/mlflow/sf2f/utils/pgan_utils.py b/mlflow/train/sf2f/utils/pgan_utils.py similarity index 100% rename from train/mlflow/sf2f/utils/pgan_utils.py rename to mlflow/train/sf2f/utils/pgan_utils.py diff --git a/train/mlflow/sf2f/utils/s2f_evaluator.py b/mlflow/train/sf2f/utils/s2f_evaluator.py similarity index 100% rename from train/mlflow/sf2f/utils/s2f_evaluator.py rename to mlflow/train/sf2f/utils/s2f_evaluator.py diff --git a/train/mlflow/sf2f/utils/training_utils.py b/mlflow/train/sf2f/utils/training_utils.py similarity index 100% rename from train/mlflow/sf2f/utils/training_utils.py rename to mlflow/train/sf2f/utils/training_utils.py diff --git a/train/mlflow/sf2f/utils/utils.py b/mlflow/train/sf2f/utils/utils.py similarity index 100% rename from train/mlflow/sf2f/utils/utils.py rename to mlflow/train/sf2f/utils/utils.py diff --git a/train/mlflow/sf2f/utils/vad_ex.py b/mlflow/train/sf2f/utils/vad_ex.py similarity index 100% rename from train/mlflow/sf2f/utils/vad_ex.py rename to mlflow/train/sf2f/utils/vad_ex.py diff --git a/train/mlflow/sf2f/utils/visualization/__init__.py b/mlflow/train/sf2f/utils/visualization/__init__.py similarity index 100% rename from train/mlflow/sf2f/utils/visualization/__init__.py rename to mlflow/train/sf2f/utils/visualization/__init__.py diff --git a/train/mlflow/sf2f/utils/visualization/html.py b/mlflow/train/sf2f/utils/visualization/html.py similarity index 100% rename from train/mlflow/sf2f/utils/visualization/html.py rename to mlflow/train/sf2f/utils/visualization/html.py diff --git a/train/mlflow/sf2f/utils/visualization/plot.py b/mlflow/train/sf2f/utils/visualization/plot.py similarity index 100% rename from train/mlflow/sf2f/utils/visualization/plot.py rename to mlflow/train/sf2f/utils/visualization/plot.py diff --git a/train/mlflow/sf2f/utils/visualization/tsne.py b/mlflow/train/sf2f/utils/visualization/tsne.py similarity index 100% rename from train/mlflow/sf2f/utils/visualization/tsne.py rename to mlflow/train/sf2f/utils/visualization/tsne.py diff --git a/train/mlflow/sf2f/utils/visualization/vis.py b/mlflow/train/sf2f/utils/visualization/vis.py similarity index 100% rename from train/mlflow/sf2f/utils/visualization/vis.py rename to mlflow/train/sf2f/utils/visualization/vis.py diff --git a/train/mlflow/sf2f/utils/wav2mel.py b/mlflow/train/sf2f/utils/wav2mel.py similarity index 100% rename from train/mlflow/sf2f/utils/wav2mel.py rename to mlflow/train/sf2f/utils/wav2mel.py From 96671046738c71a7a0f480ca784667d9163e2056 Mon Sep 17 00:00:00 2001 From: internationalwe Date: Wed, 13 Mar 2024 07:26:10 +0000 Subject: [PATCH 6/8] Rename : mlflow Docker file - - #3 --- docker/mlflow/{DockerFile_mlflow => DockerFile.mlflow} | 0 1 file changed, 0 insertions(+), 0 deletions(-) rename docker/mlflow/{DockerFile_mlflow => DockerFile.mlflow} (100%) diff --git a/docker/mlflow/DockerFile_mlflow b/docker/mlflow/DockerFile.mlflow similarity index 100% rename from docker/mlflow/DockerFile_mlflow rename to docker/mlflow/DockerFile.mlflow From 3f80a63ca81d1fd88dd8afb261eb79831ae5d2d0 Mon Sep 17 00:00:00 2001 From: internationalwe Date: Thu, 14 Mar 2024 01:20:04 +0000 Subject: [PATCH 7/8] Remove : output - - #3 --- .../sf2f_1st_stage.yaml | 52 ------------------- .../sf2f_1st_stage.yaml | 52 ------------------- .../sf2f_1st_stage.yaml | 52 ------------------- .../sf2f_1st_stage.yaml | 52 ------------------- .../sf2f_1st_stage.yaml | 52 ------------------- .../sf2f_1st_stage.yaml | 52 ------------------- .../sf2f_1st_stage.yaml | 52 ------------------- .../sf2f_1st_stage.yaml | 52 ------------------- .../sf2f_1st_stage.yaml | 52 ------------------- .../sf2f_1st_stage.yaml | 52 ------------------- .../sf2f_1st_stage.yaml | 52 ------------------- .../sf2f_1st_stage.yaml | 52 ------------------- .../sf2f_1st_stage.yaml | 52 ------------------- .../sf2f_1st_stage.yaml | 52 ------------------- .../sf2f_1st_stage.yaml | 52 ------------------- .../sf2f_1st_stage.yaml | 52 ------------------- .../sf2f_1st_stage.yaml | 52 ------------------- .../sf2f_1st_stage.yaml | 52 ------------------- .../sf2f_1st_stage.yaml | 52 ------------------- .../sf2f_1st_stage.yaml | 52 ------------------- .../sf2f_1st_stage.yaml | 52 ------------------- .../sf2f_1st_stage.yaml | 52 ------------------- 22 files changed, 1144 deletions(-) delete mode 100644 mlflow/train/sf2f/output/sf2f_1st_stage-20240303-215640/sf2f_1st_stage.yaml delete mode 100644 mlflow/train/sf2f/output/sf2f_1st_stage-20240303-220349/sf2f_1st_stage.yaml delete mode 100644 mlflow/train/sf2f/output/sf2f_1st_stage-20240303-221106/sf2f_1st_stage.yaml delete mode 100644 mlflow/train/sf2f/output/sf2f_1st_stage-20240303-225403/sf2f_1st_stage.yaml delete mode 100644 mlflow/train/sf2f/output/sf2f_1st_stage-20240303-225533/sf2f_1st_stage.yaml delete mode 100644 mlflow/train/sf2f/output/sf2f_1st_stage-20240303-225656/sf2f_1st_stage.yaml delete mode 100644 mlflow/train/sf2f/output/sf2f_1st_stage-20240303-225950/sf2f_1st_stage.yaml delete mode 100644 mlflow/train/sf2f/output/sf2f_1st_stage-20240303-230032/sf2f_1st_stage.yaml delete mode 100644 mlflow/train/sf2f/output/sf2f_1st_stage-20240303-230142/sf2f_1st_stage.yaml delete mode 100644 mlflow/train/sf2f/output/sf2f_1st_stage-20240303-230232/sf2f_1st_stage.yaml delete mode 100644 mlflow/train/sf2f/output/sf2f_1st_stage-20240303-230355/sf2f_1st_stage.yaml delete mode 100644 mlflow/train/sf2f/output/sf2f_1st_stage-20240303-230514/sf2f_1st_stage.yaml delete mode 100644 mlflow/train/sf2f/output/sf2f_1st_stage-20240303-230631/sf2f_1st_stage.yaml delete mode 100644 mlflow/train/sf2f/output/sf2f_1st_stage-20240303-231044/sf2f_1st_stage.yaml delete mode 100644 mlflow/train/sf2f/output/sf2f_1st_stage-20240303-231201/sf2f_1st_stage.yaml delete mode 100644 mlflow/train/sf2f/output/sf2f_1st_stage-20240303-231308/sf2f_1st_stage.yaml delete mode 100644 mlflow/train/sf2f/output/sf2f_1st_stage-20240303-231542/sf2f_1st_stage.yaml delete mode 100644 mlflow/train/sf2f/output/sf2f_1st_stage-20240303-231831/sf2f_1st_stage.yaml delete mode 100644 mlflow/train/sf2f/output/sf2f_1st_stage-20240303-232325/sf2f_1st_stage.yaml delete mode 100644 mlflow/train/sf2f/output/sf2f_1st_stage-20240303-233127/sf2f_1st_stage.yaml delete mode 100644 mlflow/train/sf2f/output/sf2f_1st_stage-20240303-233311/sf2f_1st_stage.yaml delete mode 100644 mlflow/train/sf2f/output/sf2f_1st_stage-20240303-233619/sf2f_1st_stage.yaml diff --git a/mlflow/train/sf2f/output/sf2f_1st_stage-20240303-215640/sf2f_1st_stage.yaml b/mlflow/train/sf2f/output/sf2f_1st_stage-20240303-215640/sf2f_1st_stage.yaml deleted file mode 100644 index c627b1d..0000000 --- a/mlflow/train/sf2f/output/sf2f_1st_stage-20240303-215640/sf2f_1st_stage.yaml +++ /dev/null @@ -1,52 +0,0 @@ -# Optimal final version of encoder-decoder and Loss Function -# Encoder: Inceptional 1D CNN -# Decoder: Upsampling + CNN -logs: - name: sf2f_1st_stage - output_dir: output/ -# data-related settings -data: - dataset: vox - data_opts_path: options/data_opts/vox.yaml - image_size: [64, 64] -# model related settings -generator: - arch: EncoderDecoder - options: - encoder_arch: V2F1DCNN - encoder_kwargs: - input_channel: 40 - channels: [256, 384, 576, 864] - output_channel: 512 - normalize_embedding: True - inception_mode: True - decoder_arch: FaceGanDecoder - decoder_kwargs: - noise_dim: 512 - mlp_normalization: none - normalization: batch - activation: leakyrelu-0.1 -discriminator: - generic: - normalization: batch - padding: valid - activation: leakyrelu-0.1 - image: - arch: 'C4-64-2,C4-128-2,C4-256-2' - identity: - arch: 'C4-64-2,C4-128-2,C4-256-2' - num_id: 0 # will be updated in train.py -optim: - # Discriminator Loss Weights - d_loss_weight: 1.0 - d_img_weight: 1.0 #0.5 - ac_loss_weight: 0.05 - # Generator Loss Weights - gan_loss_type: 'gan' - l1_pixel_loss_weight: 10.0 - # Perceptual Loss - perceptual_loss_weight: 100.0 -eval: - facenet: - deprocess_and_preprocess: True - crop_faces: True diff --git a/mlflow/train/sf2f/output/sf2f_1st_stage-20240303-220349/sf2f_1st_stage.yaml b/mlflow/train/sf2f/output/sf2f_1st_stage-20240303-220349/sf2f_1st_stage.yaml deleted file mode 100644 index c627b1d..0000000 --- a/mlflow/train/sf2f/output/sf2f_1st_stage-20240303-220349/sf2f_1st_stage.yaml +++ /dev/null @@ -1,52 +0,0 @@ -# Optimal final version of encoder-decoder and Loss Function -# Encoder: Inceptional 1D CNN -# Decoder: Upsampling + CNN -logs: - name: sf2f_1st_stage - output_dir: output/ -# data-related settings -data: - dataset: vox - data_opts_path: options/data_opts/vox.yaml - image_size: [64, 64] -# model related settings -generator: - arch: EncoderDecoder - options: - encoder_arch: V2F1DCNN - encoder_kwargs: - input_channel: 40 - channels: [256, 384, 576, 864] - output_channel: 512 - normalize_embedding: True - inception_mode: True - decoder_arch: FaceGanDecoder - decoder_kwargs: - noise_dim: 512 - mlp_normalization: none - normalization: batch - activation: leakyrelu-0.1 -discriminator: - generic: - normalization: batch - padding: valid - activation: leakyrelu-0.1 - image: - arch: 'C4-64-2,C4-128-2,C4-256-2' - identity: - arch: 'C4-64-2,C4-128-2,C4-256-2' - num_id: 0 # will be updated in train.py -optim: - # Discriminator Loss Weights - d_loss_weight: 1.0 - d_img_weight: 1.0 #0.5 - ac_loss_weight: 0.05 - # Generator Loss Weights - gan_loss_type: 'gan' - l1_pixel_loss_weight: 10.0 - # Perceptual Loss - perceptual_loss_weight: 100.0 -eval: - facenet: - deprocess_and_preprocess: True - crop_faces: True diff --git a/mlflow/train/sf2f/output/sf2f_1st_stage-20240303-221106/sf2f_1st_stage.yaml b/mlflow/train/sf2f/output/sf2f_1st_stage-20240303-221106/sf2f_1st_stage.yaml deleted file mode 100644 index c627b1d..0000000 --- a/mlflow/train/sf2f/output/sf2f_1st_stage-20240303-221106/sf2f_1st_stage.yaml +++ /dev/null @@ -1,52 +0,0 @@ -# Optimal final version of encoder-decoder and Loss Function -# Encoder: Inceptional 1D CNN -# Decoder: Upsampling + CNN -logs: - name: sf2f_1st_stage - output_dir: output/ -# data-related settings -data: - dataset: vox - data_opts_path: options/data_opts/vox.yaml - image_size: [64, 64] -# model related settings -generator: - arch: EncoderDecoder - options: - encoder_arch: V2F1DCNN - encoder_kwargs: - input_channel: 40 - channels: [256, 384, 576, 864] - output_channel: 512 - normalize_embedding: True - inception_mode: True - decoder_arch: FaceGanDecoder - decoder_kwargs: - noise_dim: 512 - mlp_normalization: none - normalization: batch - activation: leakyrelu-0.1 -discriminator: - generic: - normalization: batch - padding: valid - activation: leakyrelu-0.1 - image: - arch: 'C4-64-2,C4-128-2,C4-256-2' - identity: - arch: 'C4-64-2,C4-128-2,C4-256-2' - num_id: 0 # will be updated in train.py -optim: - # Discriminator Loss Weights - d_loss_weight: 1.0 - d_img_weight: 1.0 #0.5 - ac_loss_weight: 0.05 - # Generator Loss Weights - gan_loss_type: 'gan' - l1_pixel_loss_weight: 10.0 - # Perceptual Loss - perceptual_loss_weight: 100.0 -eval: - facenet: - deprocess_and_preprocess: True - crop_faces: True diff --git a/mlflow/train/sf2f/output/sf2f_1st_stage-20240303-225403/sf2f_1st_stage.yaml b/mlflow/train/sf2f/output/sf2f_1st_stage-20240303-225403/sf2f_1st_stage.yaml deleted file mode 100644 index c627b1d..0000000 --- a/mlflow/train/sf2f/output/sf2f_1st_stage-20240303-225403/sf2f_1st_stage.yaml +++ /dev/null @@ -1,52 +0,0 @@ -# Optimal final version of encoder-decoder and Loss Function -# Encoder: Inceptional 1D CNN -# Decoder: Upsampling + CNN -logs: - name: sf2f_1st_stage - output_dir: output/ -# data-related settings -data: - dataset: vox - data_opts_path: options/data_opts/vox.yaml - image_size: [64, 64] -# model related settings -generator: - arch: EncoderDecoder - options: - encoder_arch: V2F1DCNN - encoder_kwargs: - input_channel: 40 - channels: [256, 384, 576, 864] - output_channel: 512 - normalize_embedding: True - inception_mode: True - decoder_arch: FaceGanDecoder - decoder_kwargs: - noise_dim: 512 - mlp_normalization: none - normalization: batch - activation: leakyrelu-0.1 -discriminator: - generic: - normalization: batch - padding: valid - activation: leakyrelu-0.1 - image: - arch: 'C4-64-2,C4-128-2,C4-256-2' - identity: - arch: 'C4-64-2,C4-128-2,C4-256-2' - num_id: 0 # will be updated in train.py -optim: - # Discriminator Loss Weights - d_loss_weight: 1.0 - d_img_weight: 1.0 #0.5 - ac_loss_weight: 0.05 - # Generator Loss Weights - gan_loss_type: 'gan' - l1_pixel_loss_weight: 10.0 - # Perceptual Loss - perceptual_loss_weight: 100.0 -eval: - facenet: - deprocess_and_preprocess: True - crop_faces: True diff --git a/mlflow/train/sf2f/output/sf2f_1st_stage-20240303-225533/sf2f_1st_stage.yaml b/mlflow/train/sf2f/output/sf2f_1st_stage-20240303-225533/sf2f_1st_stage.yaml deleted file mode 100644 index c627b1d..0000000 --- a/mlflow/train/sf2f/output/sf2f_1st_stage-20240303-225533/sf2f_1st_stage.yaml +++ /dev/null @@ -1,52 +0,0 @@ -# Optimal final version of encoder-decoder and Loss Function -# Encoder: Inceptional 1D CNN -# Decoder: Upsampling + CNN -logs: - name: sf2f_1st_stage - output_dir: output/ -# data-related settings -data: - dataset: vox - data_opts_path: options/data_opts/vox.yaml - image_size: [64, 64] -# model related settings -generator: - arch: EncoderDecoder - options: - encoder_arch: V2F1DCNN - encoder_kwargs: - input_channel: 40 - channels: [256, 384, 576, 864] - output_channel: 512 - normalize_embedding: True - inception_mode: True - decoder_arch: FaceGanDecoder - decoder_kwargs: - noise_dim: 512 - mlp_normalization: none - normalization: batch - activation: leakyrelu-0.1 -discriminator: - generic: - normalization: batch - padding: valid - activation: leakyrelu-0.1 - image: - arch: 'C4-64-2,C4-128-2,C4-256-2' - identity: - arch: 'C4-64-2,C4-128-2,C4-256-2' - num_id: 0 # will be updated in train.py -optim: - # Discriminator Loss Weights - d_loss_weight: 1.0 - d_img_weight: 1.0 #0.5 - ac_loss_weight: 0.05 - # Generator Loss Weights - gan_loss_type: 'gan' - l1_pixel_loss_weight: 10.0 - # Perceptual Loss - perceptual_loss_weight: 100.0 -eval: - facenet: - deprocess_and_preprocess: True - crop_faces: True diff --git a/mlflow/train/sf2f/output/sf2f_1st_stage-20240303-225656/sf2f_1st_stage.yaml b/mlflow/train/sf2f/output/sf2f_1st_stage-20240303-225656/sf2f_1st_stage.yaml deleted file mode 100644 index c627b1d..0000000 --- a/mlflow/train/sf2f/output/sf2f_1st_stage-20240303-225656/sf2f_1st_stage.yaml +++ /dev/null @@ -1,52 +0,0 @@ -# Optimal final version of encoder-decoder and Loss Function -# Encoder: Inceptional 1D CNN -# Decoder: Upsampling + CNN -logs: - name: sf2f_1st_stage - output_dir: output/ -# data-related settings -data: - dataset: vox - data_opts_path: options/data_opts/vox.yaml - image_size: [64, 64] -# model related settings -generator: - arch: EncoderDecoder - options: - encoder_arch: V2F1DCNN - encoder_kwargs: - input_channel: 40 - channels: [256, 384, 576, 864] - output_channel: 512 - normalize_embedding: True - inception_mode: True - decoder_arch: FaceGanDecoder - decoder_kwargs: - noise_dim: 512 - mlp_normalization: none - normalization: batch - activation: leakyrelu-0.1 -discriminator: - generic: - normalization: batch - padding: valid - activation: leakyrelu-0.1 - image: - arch: 'C4-64-2,C4-128-2,C4-256-2' - identity: - arch: 'C4-64-2,C4-128-2,C4-256-2' - num_id: 0 # will be updated in train.py -optim: - # Discriminator Loss Weights - d_loss_weight: 1.0 - d_img_weight: 1.0 #0.5 - ac_loss_weight: 0.05 - # Generator Loss Weights - gan_loss_type: 'gan' - l1_pixel_loss_weight: 10.0 - # Perceptual Loss - perceptual_loss_weight: 100.0 -eval: - facenet: - deprocess_and_preprocess: True - crop_faces: True diff --git a/mlflow/train/sf2f/output/sf2f_1st_stage-20240303-225950/sf2f_1st_stage.yaml b/mlflow/train/sf2f/output/sf2f_1st_stage-20240303-225950/sf2f_1st_stage.yaml deleted file mode 100644 index c627b1d..0000000 --- a/mlflow/train/sf2f/output/sf2f_1st_stage-20240303-225950/sf2f_1st_stage.yaml +++ /dev/null @@ -1,52 +0,0 @@ -# Optimal final version of encoder-decoder and Loss Function -# Encoder: Inceptional 1D CNN -# Decoder: Upsampling + CNN -logs: - name: sf2f_1st_stage - output_dir: output/ -# data-related settings -data: - dataset: vox - data_opts_path: options/data_opts/vox.yaml - image_size: [64, 64] -# model related settings -generator: - arch: EncoderDecoder - options: - encoder_arch: V2F1DCNN - encoder_kwargs: - input_channel: 40 - channels: [256, 384, 576, 864] - output_channel: 512 - normalize_embedding: True - inception_mode: True - decoder_arch: FaceGanDecoder - decoder_kwargs: - noise_dim: 512 - mlp_normalization: none - normalization: batch - activation: leakyrelu-0.1 -discriminator: - generic: - normalization: batch - padding: valid - activation: leakyrelu-0.1 - image: - arch: 'C4-64-2,C4-128-2,C4-256-2' - identity: - arch: 'C4-64-2,C4-128-2,C4-256-2' - num_id: 0 # will be updated in train.py -optim: - # Discriminator Loss Weights - d_loss_weight: 1.0 - d_img_weight: 1.0 #0.5 - ac_loss_weight: 0.05 - # Generator Loss Weights - gan_loss_type: 'gan' - l1_pixel_loss_weight: 10.0 - # Perceptual Loss - perceptual_loss_weight: 100.0 -eval: - facenet: - deprocess_and_preprocess: True - crop_faces: True diff --git a/mlflow/train/sf2f/output/sf2f_1st_stage-20240303-230032/sf2f_1st_stage.yaml b/mlflow/train/sf2f/output/sf2f_1st_stage-20240303-230032/sf2f_1st_stage.yaml deleted file mode 100644 index c627b1d..0000000 --- a/mlflow/train/sf2f/output/sf2f_1st_stage-20240303-230032/sf2f_1st_stage.yaml +++ /dev/null @@ -1,52 +0,0 @@ -# Optimal final version of encoder-decoder and Loss Function -# Encoder: Inceptional 1D CNN -# Decoder: Upsampling + CNN -logs: - name: sf2f_1st_stage - output_dir: output/ -# data-related settings -data: - dataset: vox - data_opts_path: options/data_opts/vox.yaml - image_size: [64, 64] -# model related settings -generator: - arch: EncoderDecoder - options: - encoder_arch: V2F1DCNN - encoder_kwargs: - input_channel: 40 - channels: [256, 384, 576, 864] - output_channel: 512 - normalize_embedding: True - inception_mode: True - decoder_arch: FaceGanDecoder - decoder_kwargs: - noise_dim: 512 - mlp_normalization: none - normalization: batch - activation: leakyrelu-0.1 -discriminator: - generic: - normalization: batch - padding: valid - activation: leakyrelu-0.1 - image: - arch: 'C4-64-2,C4-128-2,C4-256-2' - identity: - arch: 'C4-64-2,C4-128-2,C4-256-2' - num_id: 0 # will be updated in train.py -optim: - # Discriminator Loss Weights - d_loss_weight: 1.0 - d_img_weight: 1.0 #0.5 - ac_loss_weight: 0.05 - # Generator Loss Weights - gan_loss_type: 'gan' - l1_pixel_loss_weight: 10.0 - # Perceptual Loss - perceptual_loss_weight: 100.0 -eval: - facenet: - deprocess_and_preprocess: True - crop_faces: True diff --git a/mlflow/train/sf2f/output/sf2f_1st_stage-20240303-230142/sf2f_1st_stage.yaml b/mlflow/train/sf2f/output/sf2f_1st_stage-20240303-230142/sf2f_1st_stage.yaml deleted file mode 100644 index c627b1d..0000000 --- a/mlflow/train/sf2f/output/sf2f_1st_stage-20240303-230142/sf2f_1st_stage.yaml +++ /dev/null @@ -1,52 +0,0 @@ -# Optimal final version of encoder-decoder and Loss Function -# Encoder: Inceptional 1D CNN -# Decoder: Upsampling + CNN -logs: - name: sf2f_1st_stage - output_dir: output/ -# data-related settings -data: - dataset: vox - data_opts_path: options/data_opts/vox.yaml - image_size: [64, 64] -# model related settings -generator: - arch: EncoderDecoder - options: - encoder_arch: V2F1DCNN - encoder_kwargs: - input_channel: 40 - channels: [256, 384, 576, 864] - output_channel: 512 - normalize_embedding: True - inception_mode: True - decoder_arch: FaceGanDecoder - decoder_kwargs: - noise_dim: 512 - mlp_normalization: none - normalization: batch - activation: leakyrelu-0.1 -discriminator: - generic: - normalization: batch - padding: valid - activation: leakyrelu-0.1 - image: - arch: 'C4-64-2,C4-128-2,C4-256-2' - identity: - arch: 'C4-64-2,C4-128-2,C4-256-2' - num_id: 0 # will be updated in train.py -optim: - # Discriminator Loss Weights - d_loss_weight: 1.0 - d_img_weight: 1.0 #0.5 - ac_loss_weight: 0.05 - # Generator Loss Weights - gan_loss_type: 'gan' - l1_pixel_loss_weight: 10.0 - # Perceptual Loss - perceptual_loss_weight: 100.0 -eval: - facenet: - deprocess_and_preprocess: True - crop_faces: True diff --git a/mlflow/train/sf2f/output/sf2f_1st_stage-20240303-230232/sf2f_1st_stage.yaml b/mlflow/train/sf2f/output/sf2f_1st_stage-20240303-230232/sf2f_1st_stage.yaml deleted file mode 100644 index c627b1d..0000000 --- a/mlflow/train/sf2f/output/sf2f_1st_stage-20240303-230232/sf2f_1st_stage.yaml +++ /dev/null @@ -1,52 +0,0 @@ -# Optimal final version of encoder-decoder and Loss Function -# Encoder: Inceptional 1D CNN -# Decoder: Upsampling + CNN -logs: - name: sf2f_1st_stage - output_dir: output/ -# data-related settings -data: - dataset: vox - data_opts_path: options/data_opts/vox.yaml - image_size: [64, 64] -# model related settings -generator: - arch: EncoderDecoder - options: - encoder_arch: V2F1DCNN - encoder_kwargs: - input_channel: 40 - channels: [256, 384, 576, 864] - output_channel: 512 - normalize_embedding: True - inception_mode: True - decoder_arch: FaceGanDecoder - decoder_kwargs: - noise_dim: 512 - mlp_normalization: none - normalization: batch - activation: leakyrelu-0.1 -discriminator: - generic: - normalization: batch - padding: valid - activation: leakyrelu-0.1 - image: - arch: 'C4-64-2,C4-128-2,C4-256-2' - identity: - arch: 'C4-64-2,C4-128-2,C4-256-2' - num_id: 0 # will be updated in train.py -optim: - # Discriminator Loss Weights - d_loss_weight: 1.0 - d_img_weight: 1.0 #0.5 - ac_loss_weight: 0.05 - # Generator Loss Weights - gan_loss_type: 'gan' - l1_pixel_loss_weight: 10.0 - # Perceptual Loss - perceptual_loss_weight: 100.0 -eval: - facenet: - deprocess_and_preprocess: True - crop_faces: True diff --git a/mlflow/train/sf2f/output/sf2f_1st_stage-20240303-230355/sf2f_1st_stage.yaml b/mlflow/train/sf2f/output/sf2f_1st_stage-20240303-230355/sf2f_1st_stage.yaml deleted file mode 100644 index c627b1d..0000000 --- a/mlflow/train/sf2f/output/sf2f_1st_stage-20240303-230355/sf2f_1st_stage.yaml +++ /dev/null @@ -1,52 +0,0 @@ -# Optimal final version of encoder-decoder and Loss Function -# Encoder: Inceptional 1D CNN -# Decoder: Upsampling + CNN -logs: - name: sf2f_1st_stage - output_dir: output/ -# data-related settings -data: - dataset: vox - data_opts_path: options/data_opts/vox.yaml - image_size: [64, 64] -# model related settings -generator: - arch: EncoderDecoder - options: - encoder_arch: V2F1DCNN - encoder_kwargs: - input_channel: 40 - channels: [256, 384, 576, 864] - output_channel: 512 - normalize_embedding: True - inception_mode: True - decoder_arch: FaceGanDecoder - decoder_kwargs: - noise_dim: 512 - mlp_normalization: none - normalization: batch - activation: leakyrelu-0.1 -discriminator: - generic: - normalization: batch - padding: valid - activation: leakyrelu-0.1 - image: - arch: 'C4-64-2,C4-128-2,C4-256-2' - identity: - arch: 'C4-64-2,C4-128-2,C4-256-2' - num_id: 0 # will be updated in train.py -optim: - # Discriminator Loss Weights - d_loss_weight: 1.0 - d_img_weight: 1.0 #0.5 - ac_loss_weight: 0.05 - # Generator Loss Weights - gan_loss_type: 'gan' - l1_pixel_loss_weight: 10.0 - # Perceptual Loss - perceptual_loss_weight: 100.0 -eval: - facenet: - deprocess_and_preprocess: True - crop_faces: True diff --git a/mlflow/train/sf2f/output/sf2f_1st_stage-20240303-230514/sf2f_1st_stage.yaml b/mlflow/train/sf2f/output/sf2f_1st_stage-20240303-230514/sf2f_1st_stage.yaml deleted file mode 100644 index c627b1d..0000000 --- a/mlflow/train/sf2f/output/sf2f_1st_stage-20240303-230514/sf2f_1st_stage.yaml +++ /dev/null @@ -1,52 +0,0 @@ -# Optimal final version of encoder-decoder and Loss Function -# Encoder: Inceptional 1D CNN -# Decoder: Upsampling + CNN -logs: - name: sf2f_1st_stage - output_dir: output/ -# data-related settings -data: - dataset: vox - data_opts_path: options/data_opts/vox.yaml - image_size: [64, 64] -# model related settings -generator: - arch: EncoderDecoder - options: - encoder_arch: V2F1DCNN - encoder_kwargs: - input_channel: 40 - channels: [256, 384, 576, 864] - output_channel: 512 - normalize_embedding: True - inception_mode: True - decoder_arch: FaceGanDecoder - decoder_kwargs: - noise_dim: 512 - mlp_normalization: none - normalization: batch - activation: leakyrelu-0.1 -discriminator: - generic: - normalization: batch - padding: valid - activation: leakyrelu-0.1 - image: - arch: 'C4-64-2,C4-128-2,C4-256-2' - identity: - arch: 'C4-64-2,C4-128-2,C4-256-2' - num_id: 0 # will be updated in train.py -optim: - # Discriminator Loss Weights - d_loss_weight: 1.0 - d_img_weight: 1.0 #0.5 - ac_loss_weight: 0.05 - # Generator Loss Weights - gan_loss_type: 'gan' - l1_pixel_loss_weight: 10.0 - # Perceptual Loss - perceptual_loss_weight: 100.0 -eval: - facenet: - deprocess_and_preprocess: True - crop_faces: True diff --git a/mlflow/train/sf2f/output/sf2f_1st_stage-20240303-230631/sf2f_1st_stage.yaml b/mlflow/train/sf2f/output/sf2f_1st_stage-20240303-230631/sf2f_1st_stage.yaml deleted file mode 100644 index c627b1d..0000000 --- a/mlflow/train/sf2f/output/sf2f_1st_stage-20240303-230631/sf2f_1st_stage.yaml +++ /dev/null @@ -1,52 +0,0 @@ -# Optimal final version of encoder-decoder and Loss Function -# Encoder: Inceptional 1D CNN -# Decoder: Upsampling + CNN -logs: - name: sf2f_1st_stage - output_dir: output/ -# data-related settings -data: - dataset: vox - data_opts_path: options/data_opts/vox.yaml - image_size: [64, 64] -# model related settings -generator: - arch: EncoderDecoder - options: - encoder_arch: V2F1DCNN - encoder_kwargs: - input_channel: 40 - channels: [256, 384, 576, 864] - output_channel: 512 - normalize_embedding: True - inception_mode: True - decoder_arch: FaceGanDecoder - decoder_kwargs: - noise_dim: 512 - mlp_normalization: none - normalization: batch - activation: leakyrelu-0.1 -discriminator: - generic: - normalization: batch - padding: valid - activation: leakyrelu-0.1 - image: - arch: 'C4-64-2,C4-128-2,C4-256-2' - identity: - arch: 'C4-64-2,C4-128-2,C4-256-2' - num_id: 0 # will be updated in train.py -optim: - # Discriminator Loss Weights - d_loss_weight: 1.0 - d_img_weight: 1.0 #0.5 - ac_loss_weight: 0.05 - # Generator Loss Weights - gan_loss_type: 'gan' - l1_pixel_loss_weight: 10.0 - # Perceptual Loss - perceptual_loss_weight: 100.0 -eval: - facenet: - deprocess_and_preprocess: True - crop_faces: True diff --git a/mlflow/train/sf2f/output/sf2f_1st_stage-20240303-231044/sf2f_1st_stage.yaml b/mlflow/train/sf2f/output/sf2f_1st_stage-20240303-231044/sf2f_1st_stage.yaml deleted file mode 100644 index c627b1d..0000000 --- a/mlflow/train/sf2f/output/sf2f_1st_stage-20240303-231044/sf2f_1st_stage.yaml +++ /dev/null @@ -1,52 +0,0 @@ -# Optimal final version of encoder-decoder and Loss Function -# Encoder: Inceptional 1D CNN -# Decoder: Upsampling + CNN -logs: - name: sf2f_1st_stage - output_dir: output/ -# data-related settings -data: - dataset: vox - data_opts_path: options/data_opts/vox.yaml - image_size: [64, 64] -# model related settings -generator: - arch: EncoderDecoder - options: - encoder_arch: V2F1DCNN - encoder_kwargs: - input_channel: 40 - channels: [256, 384, 576, 864] - output_channel: 512 - normalize_embedding: True - inception_mode: True - decoder_arch: FaceGanDecoder - decoder_kwargs: - noise_dim: 512 - mlp_normalization: none - normalization: batch - activation: leakyrelu-0.1 -discriminator: - generic: - normalization: batch - padding: valid - activation: leakyrelu-0.1 - image: - arch: 'C4-64-2,C4-128-2,C4-256-2' - identity: - arch: 'C4-64-2,C4-128-2,C4-256-2' - num_id: 0 # will be updated in train.py -optim: - # Discriminator Loss Weights - d_loss_weight: 1.0 - d_img_weight: 1.0 #0.5 - ac_loss_weight: 0.05 - # Generator Loss Weights - gan_loss_type: 'gan' - l1_pixel_loss_weight: 10.0 - # Perceptual Loss - perceptual_loss_weight: 100.0 -eval: - facenet: - deprocess_and_preprocess: True - crop_faces: True diff --git a/mlflow/train/sf2f/output/sf2f_1st_stage-20240303-231201/sf2f_1st_stage.yaml b/mlflow/train/sf2f/output/sf2f_1st_stage-20240303-231201/sf2f_1st_stage.yaml deleted file mode 100644 index c627b1d..0000000 --- a/mlflow/train/sf2f/output/sf2f_1st_stage-20240303-231201/sf2f_1st_stage.yaml +++ /dev/null @@ -1,52 +0,0 @@ -# Optimal final version of encoder-decoder and Loss Function -# Encoder: Inceptional 1D CNN -# Decoder: Upsampling + CNN -logs: - name: sf2f_1st_stage - output_dir: output/ -# data-related settings -data: - dataset: vox - data_opts_path: options/data_opts/vox.yaml - image_size: [64, 64] -# model related settings -generator: - arch: EncoderDecoder - options: - encoder_arch: V2F1DCNN - encoder_kwargs: - input_channel: 40 - channels: [256, 384, 576, 864] - output_channel: 512 - normalize_embedding: True - inception_mode: True - decoder_arch: FaceGanDecoder - decoder_kwargs: - noise_dim: 512 - mlp_normalization: none - normalization: batch - activation: leakyrelu-0.1 -discriminator: - generic: - normalization: batch - padding: valid - activation: leakyrelu-0.1 - image: - arch: 'C4-64-2,C4-128-2,C4-256-2' - identity: - arch: 'C4-64-2,C4-128-2,C4-256-2' - num_id: 0 # will be updated in train.py -optim: - # Discriminator Loss Weights - d_loss_weight: 1.0 - d_img_weight: 1.0 #0.5 - ac_loss_weight: 0.05 - # Generator Loss Weights - gan_loss_type: 'gan' - l1_pixel_loss_weight: 10.0 - # Perceptual Loss - perceptual_loss_weight: 100.0 -eval: - facenet: - deprocess_and_preprocess: True - crop_faces: True diff --git a/mlflow/train/sf2f/output/sf2f_1st_stage-20240303-231308/sf2f_1st_stage.yaml b/mlflow/train/sf2f/output/sf2f_1st_stage-20240303-231308/sf2f_1st_stage.yaml deleted file mode 100644 index c627b1d..0000000 --- a/mlflow/train/sf2f/output/sf2f_1st_stage-20240303-231308/sf2f_1st_stage.yaml +++ /dev/null @@ -1,52 +0,0 @@ -# Optimal final version of encoder-decoder and Loss Function -# Encoder: Inceptional 1D CNN -# Decoder: Upsampling + CNN -logs: - name: sf2f_1st_stage - output_dir: output/ -# data-related settings -data: - dataset: vox - data_opts_path: options/data_opts/vox.yaml - image_size: [64, 64] -# model related settings -generator: - arch: EncoderDecoder - options: - encoder_arch: V2F1DCNN - encoder_kwargs: - input_channel: 40 - channels: [256, 384, 576, 864] - output_channel: 512 - normalize_embedding: True - inception_mode: True - decoder_arch: FaceGanDecoder - decoder_kwargs: - noise_dim: 512 - mlp_normalization: none - normalization: batch - activation: leakyrelu-0.1 -discriminator: - generic: - normalization: batch - padding: valid - activation: leakyrelu-0.1 - image: - arch: 'C4-64-2,C4-128-2,C4-256-2' - identity: - arch: 'C4-64-2,C4-128-2,C4-256-2' - num_id: 0 # will be updated in train.py -optim: - # Discriminator Loss Weights - d_loss_weight: 1.0 - d_img_weight: 1.0 #0.5 - ac_loss_weight: 0.05 - # Generator Loss Weights - gan_loss_type: 'gan' - l1_pixel_loss_weight: 10.0 - # Perceptual Loss - perceptual_loss_weight: 100.0 -eval: - facenet: - deprocess_and_preprocess: True - crop_faces: True diff --git a/mlflow/train/sf2f/output/sf2f_1st_stage-20240303-231542/sf2f_1st_stage.yaml b/mlflow/train/sf2f/output/sf2f_1st_stage-20240303-231542/sf2f_1st_stage.yaml deleted file mode 100644 index c627b1d..0000000 --- a/mlflow/train/sf2f/output/sf2f_1st_stage-20240303-231542/sf2f_1st_stage.yaml +++ /dev/null @@ -1,52 +0,0 @@ -# Optimal final version of encoder-decoder and Loss Function -# Encoder: Inceptional 1D CNN -# Decoder: Upsampling + CNN -logs: - name: sf2f_1st_stage - output_dir: output/ -# data-related settings -data: - dataset: vox - data_opts_path: options/data_opts/vox.yaml - image_size: [64, 64] -# model related settings -generator: - arch: EncoderDecoder - options: - encoder_arch: V2F1DCNN - encoder_kwargs: - input_channel: 40 - channels: [256, 384, 576, 864] - output_channel: 512 - normalize_embedding: True - inception_mode: True - decoder_arch: FaceGanDecoder - decoder_kwargs: - noise_dim: 512 - mlp_normalization: none - normalization: batch - activation: leakyrelu-0.1 -discriminator: - generic: - normalization: batch - padding: valid - activation: leakyrelu-0.1 - image: - arch: 'C4-64-2,C4-128-2,C4-256-2' - identity: - arch: 'C4-64-2,C4-128-2,C4-256-2' - num_id: 0 # will be updated in train.py -optim: - # Discriminator Loss Weights - d_loss_weight: 1.0 - d_img_weight: 1.0 #0.5 - ac_loss_weight: 0.05 - # Generator Loss Weights - gan_loss_type: 'gan' - l1_pixel_loss_weight: 10.0 - # Perceptual Loss - perceptual_loss_weight: 100.0 -eval: - facenet: - deprocess_and_preprocess: True - crop_faces: True diff --git a/mlflow/train/sf2f/output/sf2f_1st_stage-20240303-231831/sf2f_1st_stage.yaml b/mlflow/train/sf2f/output/sf2f_1st_stage-20240303-231831/sf2f_1st_stage.yaml deleted file mode 100644 index c627b1d..0000000 --- a/mlflow/train/sf2f/output/sf2f_1st_stage-20240303-231831/sf2f_1st_stage.yaml +++ /dev/null @@ -1,52 +0,0 @@ -# Optimal final version of encoder-decoder and Loss Function -# Encoder: Inceptional 1D CNN -# Decoder: Upsampling + CNN -logs: - name: sf2f_1st_stage - output_dir: output/ -# data-related settings -data: - dataset: vox - data_opts_path: options/data_opts/vox.yaml - image_size: [64, 64] -# model related settings -generator: - arch: EncoderDecoder - options: - encoder_arch: V2F1DCNN - encoder_kwargs: - input_channel: 40 - channels: [256, 384, 576, 864] - output_channel: 512 - normalize_embedding: True - inception_mode: True - decoder_arch: FaceGanDecoder - decoder_kwargs: - noise_dim: 512 - mlp_normalization: none - normalization: batch - activation: leakyrelu-0.1 -discriminator: - generic: - normalization: batch - padding: valid - activation: leakyrelu-0.1 - image: - arch: 'C4-64-2,C4-128-2,C4-256-2' - identity: - arch: 'C4-64-2,C4-128-2,C4-256-2' - num_id: 0 # will be updated in train.py -optim: - # Discriminator Loss Weights - d_loss_weight: 1.0 - d_img_weight: 1.0 #0.5 - ac_loss_weight: 0.05 - # Generator Loss Weights - gan_loss_type: 'gan' - l1_pixel_loss_weight: 10.0 - # Perceptual Loss - perceptual_loss_weight: 100.0 -eval: - facenet: - deprocess_and_preprocess: True - crop_faces: True diff --git a/mlflow/train/sf2f/output/sf2f_1st_stage-20240303-232325/sf2f_1st_stage.yaml b/mlflow/train/sf2f/output/sf2f_1st_stage-20240303-232325/sf2f_1st_stage.yaml deleted file mode 100644 index c627b1d..0000000 --- a/mlflow/train/sf2f/output/sf2f_1st_stage-20240303-232325/sf2f_1st_stage.yaml +++ /dev/null @@ -1,52 +0,0 @@ -# Optimal final version of encoder-decoder and Loss Function -# Encoder: Inceptional 1D CNN -# Decoder: Upsampling + CNN -logs: - name: sf2f_1st_stage - output_dir: output/ -# data-related settings -data: - dataset: vox - data_opts_path: options/data_opts/vox.yaml - image_size: [64, 64] -# model related settings -generator: - arch: EncoderDecoder - options: - encoder_arch: V2F1DCNN - encoder_kwargs: - input_channel: 40 - channels: [256, 384, 576, 864] - output_channel: 512 - normalize_embedding: True - inception_mode: True - decoder_arch: FaceGanDecoder - decoder_kwargs: - noise_dim: 512 - mlp_normalization: none - normalization: batch - activation: leakyrelu-0.1 -discriminator: - generic: - normalization: batch - padding: valid - activation: leakyrelu-0.1 - image: - arch: 'C4-64-2,C4-128-2,C4-256-2' - identity: - arch: 'C4-64-2,C4-128-2,C4-256-2' - num_id: 0 # will be updated in train.py -optim: - # Discriminator Loss Weights - d_loss_weight: 1.0 - d_img_weight: 1.0 #0.5 - ac_loss_weight: 0.05 - # Generator Loss Weights - gan_loss_type: 'gan' - l1_pixel_loss_weight: 10.0 - # Perceptual Loss - perceptual_loss_weight: 100.0 -eval: - facenet: - deprocess_and_preprocess: True - crop_faces: True diff --git a/mlflow/train/sf2f/output/sf2f_1st_stage-20240303-233127/sf2f_1st_stage.yaml b/mlflow/train/sf2f/output/sf2f_1st_stage-20240303-233127/sf2f_1st_stage.yaml deleted file mode 100644 index c627b1d..0000000 --- a/mlflow/train/sf2f/output/sf2f_1st_stage-20240303-233127/sf2f_1st_stage.yaml +++ /dev/null @@ -1,52 +0,0 @@ -# Optimal final version of encoder-decoder and Loss Function -# Encoder: Inceptional 1D CNN -# Decoder: Upsampling + CNN -logs: - name: sf2f_1st_stage - output_dir: output/ -# data-related settings -data: - dataset: vox - data_opts_path: options/data_opts/vox.yaml - image_size: [64, 64] -# model related settings -generator: - arch: EncoderDecoder - options: - encoder_arch: V2F1DCNN - encoder_kwargs: - input_channel: 40 - channels: [256, 384, 576, 864] - output_channel: 512 - normalize_embedding: True - inception_mode: True - decoder_arch: FaceGanDecoder - decoder_kwargs: - noise_dim: 512 - mlp_normalization: none - normalization: batch - activation: leakyrelu-0.1 -discriminator: - generic: - normalization: batch - padding: valid - activation: leakyrelu-0.1 - image: - arch: 'C4-64-2,C4-128-2,C4-256-2' - identity: - arch: 'C4-64-2,C4-128-2,C4-256-2' - num_id: 0 # will be updated in train.py -optim: - # Discriminator Loss Weights - d_loss_weight: 1.0 - d_img_weight: 1.0 #0.5 - ac_loss_weight: 0.05 - # Generator Loss Weights - gan_loss_type: 'gan' - l1_pixel_loss_weight: 10.0 - # Perceptual Loss - perceptual_loss_weight: 100.0 -eval: - facenet: - deprocess_and_preprocess: True - crop_faces: True diff --git a/mlflow/train/sf2f/output/sf2f_1st_stage-20240303-233311/sf2f_1st_stage.yaml b/mlflow/train/sf2f/output/sf2f_1st_stage-20240303-233311/sf2f_1st_stage.yaml deleted file mode 100644 index c627b1d..0000000 --- a/mlflow/train/sf2f/output/sf2f_1st_stage-20240303-233311/sf2f_1st_stage.yaml +++ /dev/null @@ -1,52 +0,0 @@ -# Optimal final version of encoder-decoder and Loss Function -# Encoder: Inceptional 1D CNN -# Decoder: Upsampling + CNN -logs: - name: sf2f_1st_stage - output_dir: output/ -# data-related settings -data: - dataset: vox - data_opts_path: options/data_opts/vox.yaml - image_size: [64, 64] -# model related settings -generator: - arch: EncoderDecoder - options: - encoder_arch: V2F1DCNN - encoder_kwargs: - input_channel: 40 - channels: [256, 384, 576, 864] - output_channel: 512 - normalize_embedding: True - inception_mode: True - decoder_arch: FaceGanDecoder - decoder_kwargs: - noise_dim: 512 - mlp_normalization: none - normalization: batch - activation: leakyrelu-0.1 -discriminator: - generic: - normalization: batch - padding: valid - activation: leakyrelu-0.1 - image: - arch: 'C4-64-2,C4-128-2,C4-256-2' - identity: - arch: 'C4-64-2,C4-128-2,C4-256-2' - num_id: 0 # will be updated in train.py -optim: - # Discriminator Loss Weights - d_loss_weight: 1.0 - d_img_weight: 1.0 #0.5 - ac_loss_weight: 0.05 - # Generator Loss Weights - gan_loss_type: 'gan' - l1_pixel_loss_weight: 10.0 - # Perceptual Loss - perceptual_loss_weight: 100.0 -eval: - facenet: - deprocess_and_preprocess: True - crop_faces: True diff --git a/mlflow/train/sf2f/output/sf2f_1st_stage-20240303-233619/sf2f_1st_stage.yaml b/mlflow/train/sf2f/output/sf2f_1st_stage-20240303-233619/sf2f_1st_stage.yaml deleted file mode 100644 index c627b1d..0000000 --- a/mlflow/train/sf2f/output/sf2f_1st_stage-20240303-233619/sf2f_1st_stage.yaml +++ /dev/null @@ -1,52 +0,0 @@ -# Optimal final version of encoder-decoder and Loss Function -# Encoder: Inceptional 1D CNN -# Decoder: Upsampling + CNN -logs: - name: sf2f_1st_stage - output_dir: output/ -# data-related settings -data: - dataset: vox - data_opts_path: options/data_opts/vox.yaml - image_size: [64, 64] -# model related settings -generator: - arch: EncoderDecoder - options: - encoder_arch: V2F1DCNN - encoder_kwargs: - input_channel: 40 - channels: [256, 384, 576, 864] - output_channel: 512 - normalize_embedding: True - inception_mode: True - decoder_arch: FaceGanDecoder - decoder_kwargs: - noise_dim: 512 - mlp_normalization: none - normalization: batch - activation: leakyrelu-0.1 -discriminator: - generic: - normalization: batch - padding: valid - activation: leakyrelu-0.1 - image: - arch: 'C4-64-2,C4-128-2,C4-256-2' - identity: - arch: 'C4-64-2,C4-128-2,C4-256-2' - num_id: 0 # will be updated in train.py -optim: - # Discriminator Loss Weights - d_loss_weight: 1.0 - d_img_weight: 1.0 #0.5 - ac_loss_weight: 0.05 - # Generator Loss Weights - gan_loss_type: 'gan' - l1_pixel_loss_weight: 10.0 - # Perceptual Loss - perceptual_loss_weight: 100.0 -eval: - facenet: - deprocess_and_preprocess: True - crop_faces: True From e9ae801d9f0d3d6fde7870e9deb75bc2cbaba115 Mon Sep 17 00:00:00 2001 From: internationalwe Date: Thu, 14 Mar 2024 01:39:44 +0000 Subject: [PATCH 8/8] Remove : wav file - - #3 --- .../sf2f/data/\353\205\271\354\235\214.wav" | Bin 1138800 -> 0 bytes 1 file changed, 0 insertions(+), 0 deletions(-) delete mode 100644 "mlflow/train/sf2f/data/\353\205\271\354\235\214.wav" diff --git "a/mlflow/train/sf2f/data/\353\205\271\354\235\214.wav" "b/mlflow/train/sf2f/data/\353\205\271\354\235\214.wav" deleted file mode 100644 index dab0b2994b02d24dd9c38c485f72b7e45874879d..0000000000000000000000000000000000000000 GIT binary patch literal 0 HcmV?d00001 literal 1138800 zcmeF)1(+qrku~g|8OzLT21#aSW|^58CEGGHGs(=%C^IvIWSN;6B$=5RH2Tlm>evnE zR^7fm8hQ8sw(ohKiG)b3tcZ-rte$IK@p6~D?Ow-Qx&9@uciCIt>7KhEcV%T|%gWZ3 zhrTa?UvFJG(#mmHu5!g|U+1EUUh!&|yZRL`eXYyvlJL3DdG7N{x#U%kk=QRD`JP|C z@Cm!0@!V%h+O;otont5MDmTCPtDyHL&P-cI zf{g7gDuZ|YO{I-I|I_h9^Beon(|Wl0r1f8=PtnddhPNO3uaN(*Dci>)tMQR%N8*PZ z`tO7OZKSn6p0Qlzjp;{3{GW#|kE!+-mHx+}|M+jUR~^0F6kpGu3s+r5@l|+^_PXYH zb|ii3`7e5omDlOA?e;Po4NtyoYl+_NMfsIiGN9s%=v1Bh{6F!V>dVgP#}QR;^{wXn zh@jQRvUIKwz5EDc+I7vSp_D=K8+OeBi0_Q*~GWM!L(&ayatJe@B)4?>R^QZ?zUrm6@Ys|E9*$ z(?WUHgs0kFRPTDSsAT+ld>hHHI*X2F z=hBO>@^^%m-X3Y+lTPhv`yPBMJeHZ_S(Mhx_rA=9i+@p_mB&^oRN5S^dgtg|xw-Ue zcZ&Y~(0Tmr(0|1|=i;2Hc9%(;8&6N4o1?k(UUnnlx%zs#b<*P%o?Aat>E4N+UUlqj z+S@py!u|ao+vs^y;q~N7m6_tJyq!UJcKuWJ{clOz8S~ZasB!hQ^}*-@&=c~W%R(}^zyuw9`N4W}D9$LR%?=s=Bk2@9rA*Hvj~FIQn>8Q0W~#;5)cZm+B6t!U4yusZ5*`qp0q7MET; zQ?%-wqBYh|i?7o9JXN^z@K&3`{U^QpCQsQ~>z)JNu`OxJ8P{KfzXYhil3Zi0dFXR6 zRloMsx9Epw(N9@*k7JSE%Zd9{+T1#;{KL5aYy342=z71ZqiAnSxPFyhb3RgkL%sh^ zcj5kb)yF*X{+bbdebp~xtLN+adA1#%_n(75HHd?}%I|4?eOMz^FP>gc zg?n9NeN#NWf3*&cq1K$S8Gntt=0%!cJx}4zSMyo_U%*jT<}vm^|b0& z-plXxu`r;n!2jqqGNm24`pgD$F}5+KmGDt#%J!i7kXX~SN;_3 zd=)M}?e(yD%@1nLbc}fw@8c8ptGqdS7}vl4i+B8u^E4N)=Tpz4p5a)Z_*|bV9ZzpZ z+^_QI=wV#{%2#PqbgCYFz3rVHHZL8${+_-g;a*pbx2ILO>K@-G@yhG8%B@GI%2Yqh zb*<$-?>$ZZy-(F2>Ai00g^~j$n>ww$&MSYBJnVU+eC@KfhfCI$T&;FH)rUR?VQC{z z=j-{xe&)00Pk1%4{nOiM$^Y55dH|oFI{bvIPsLMIAN#xs_kNGzUVrsRTIZ{}dw!*L zTKPqLo^YjiTH~^|$k#=3xAI164>0b}@RYpn^;X%QN0yg7uQgt@o?m|_iXV^k{xhEO zXS~K=&!(^Y{;XSi)$FX1Hz3;v5v0Q&Q`lIzENp@%GdLfRy^Zc@BG!K^40%(9u+TsJX1e))HCYON4TEb z996E$7SC8m#VfzcjO8l7=(2qHJMTuqwMJ_mm!VZ=j?UH5=YHhrdBPjb$FDzqPHW!# zbF8`PeIMJZc-2>BP-*=)#Z!DePq@&_3mb2(_j(`13*&oG>$>RpeoUnoPvv!5?Tl34 zs?Q@e##*O+{o$*;n)jLy_0?~zua~R(q}9(HP5tEey2SC-So%0CTzRNp)uEs3^%bvj zRaedHs1te|Pj5>d^{a7@?TVX+TEl&=D-G@Cs;;T|lQ-6Z9iKZCI!qb{%h)N|b^9mi4SE8W~2QNMBS zHaZsN{Z=33+CSCT$`h}#^sz~=a>D&r=|$yL9i4A3j8|K=CW_W`^$gh#M_mVatc_YH zXswmX8}mC)@k-Nn&2RO)#@A_oM=E`azo(7wsQ0nI_x=5y${Tsdc)jbrZ_9+M+(>)b z8e>lr_8ZfBdX?J|T60%x3iYeHs@;ljlve#agHE+w{VO`fvm%d)*riR~@LbJw6p4`>-fJ zRers+vH$C>6Zw9{S8IA%x{MA{SJglEPkg;GEt@}8cTdxYs%z1_%bQw@wT9O#JGN1A zeHrIL-g@=*{Hm+dKE|nVwK>w_?R8bS@`_eET4_~Z#XGIMIa<8E-U?Tqvb7H9)>-A_ z^{YBY+UIZ#*K-{yKELWu(O!3@P0<>=KI(t(>o~SW`!(fieA?;tj&;h{zj2Jxh~0_u@&vloo=jn<&Sk#Ta{jX%GG{l zv=2txK$)pu)x9Hhsy=Djs=n3us?SBM?AUIlFQQckT0iv&ORM?pw94SAHiiAFo=WfZ zFpk6it-5PJGmfo~+5AtTY`3CQXHD`deY7Pjy~jMXn`=A)inpS#Y#C_I+0x}+B>t<%aIskW+r<5((Q`Eyhq zRd2Of@lNsi^|azgd-{6e!)iM0-;uZ4Q0Mqgjq!T_dizxlZ%@~b_?Ta5BgH@d%wf$d zTE9x`>#VO$>0`X}R(Oi1>hA50{Sw!gq7^Q_agN8aPQ|;t>fw@YHTOMlQFyA2o>qOV zKBJX}ufo!6-p6rPp1B(5ywB}ic+vh&nR-q|tDcHiyK`+9U#I=~_GesUt8w?a>S6g) zzbaR>^Q>3KbFAmwpJ{Km!d3rBtG!Cc-`lNch_85rE4`>RzlwJ%y~-AqR=-M{qs7;0 z=k4hgmS1C^qrHx0_^X~N-m$K!c$FWYTcuBxIlO6ebys?))mF{nGPKI5r?0^tt~J_a zQEzt)tFzLoo?gdPxcI6ZT79c=Sd-&>+~4sEm%Jz{eJodTbyodFrPZ(aidK53%1!mP z(rZk`Q}Lpej`lR=wNq_(T4SxW-k(ZaL_2?ltG~TJ^3+r9jkMaRIx1eY(yN|I@3iw& zSpL}7RNOPBPx@JG>6hZKw$yEmTzt+^>Gy?(-#K1HYIr0Oj?mK)2=)cA_Pq7N%aqZdW$USD(CZz?Gga5f)8jQyQ?$>0cdE@k5g-? z`ZkskpZayaF<)M_Rpo13)z?n@SjKSY8^a@S#fz`nL@TX$&`K*Rt*>qAJMtUr>uoL* zuJYRMb6b6>{GydU(#oHrczfN#y?oEB^r?EL(iV-e^H1^heBqI&o=@MeRa()WCtT?x zt-8lH#dq}A>#92XT-JNz{V3VqX}ve~j#k;Kt7va$3>VL`R2$W=Ia>Xy_gR_s>c_L* zuj-iMtGqdC-1Q9mv#sYpbtY1IMSD5nN*`&RnMkiP)pqYkrQxaY9IfYOd=>7zz20hb znR<)2>Z-b{zRI7XeSA~-Q)$)CdbG+GpZ?7KdY`AlRj$@e#dkKX`ulq!ZEUCb=4gL! zrg(b&W4QB=?e=)_?Pz&*RUN(Fsc=0*&s2Eo8TGWOI;8dXHo~vm)SQlEuXxF?QPz(6 z#aFcWYi}-IdCK;%c>neIIR4Hz=85;RcstJ+H*fuU)O*%x%U&E`cqV%aP@bN))*_l>eNn!7tv}D_3QB-uKG7htKP8>H4pWiE5El{X(O$&a}@ug zyvnb((J4&TQDuwb8OzROn$K6yQ@^G=0puH;p?<{PbNRe4p1Hs->;{W09j$Q#Gq z<0ThLeiW5ne8p3AQTo`IF|Kc8+FZQa8ENNNU(H+1skHtp{$7u?9v3#g8viIiD}O9A z#?{-~u0HiXRu~`t`t@?lhSgQ;2Q4|%Y0s~4&+p;NM|(f3Kb1B`i)W-G@0efXN5}cB zd9Srs&!^%eohn~xopD4#X z*y<--^HFlD$47ZpJVo(Vxae3$Twc}N%k;FKUUiN6BhMILRBq%Q`#F}acv0nRT?qI1 zIH$$eY0Y`jdNxHXec3Xe{oJqC#Uh?6TkRH=Ug1Tw`niZ_uCAWm!&O(+Jwv?xNu9Y#4x7X3b`cgk*^{e=L{Vcx2N=^UUpLKBCZcI~6XvsE>V| zeeR~_XH4tyKG)-1_V%kiX%#No>!>ue%J*=k8DoVP%}>qYdUUL#=5UIRyyEzZ&Xwuu z`abuoadn;^u5wfT=xK9&eJ;9AqPu#$>#0gtb_~zeSA3mT8>J_UPU*|yRlfLoT|HcV z=;N=jJip>Wr|Rtaou{``{V3Y=gvUCj;=Nzh&REYF$LCkiw6BpKmfzz&T<=y-+h};I z{i=8BJyn1G^uOj~42zGy;vH$9n|cmYwDVORV;kbtzeTi4h|iUi-k*gy{`!g6uk(%J;u-14 z*Vp^pn(t+*zUogOSD&BC?{wtt@#3pGiuUx*ztM2jvomPT!A9t!cBjTN@>g5^Sxkk; z=h@@MUuk`>(!+gEG~OqeiWh&cqlYWMXysKuI;}o-x?X2uEe9y16 zdLColop&nS``+6s-l?=|v$wTgJyqWv?RC~zicYP+F|ByfMR`+vU3bmN_%Xr)h4byPc3bnbcebmhih)ith*Mdf;#;;Xc(zs68wE83sU z)V^a;S}!YHynX*WhKpyUBVWZwzOl`zade)_A8E};&Cf_{{x(9#I+lr7KNitm&&XGK zQ~j*9P;}&}^pTc)E_uBpwCX^2q^^E-#AICJtYkn5du}sC4sqh@#k$TswtMWH8 z-jTQFtO+I_LT|rjKLm;~vAs zJJzu?;>*_E=WA-5V}6fM%}=#EM{5q|@)xD|dFlO_3wQoS^;TZzUoTu`IvvZ5{TlPf z_{e{F<754E@jiwgu4lCo+RN{(aJ64DsngO6hllq1gw<7cO`U;Pd?W3Q^^Wlx*D`c0 zTm9_)DIWEWb3gU$7x9dF;){M8;h$R9%hp|d>Zow1W8IZEmtOf(blEW#Up>c)7p?T7 zl|DzuzRdAedew_o+K$p**BG|G#j`?G}_~!av`BQYPtH-DKtBv)>xhTKtC_2Ym&#UhPD-Esv+#D_5x%w-; z_?D@&`ch?!_B!Um#k(GzDqmyld2{tF;^}$9Q@@d?+Fs8_FTO>z>R7}hulS12@mKyx z=lFa2IL?aWsqobN%%#kE?+VX7 z8|j{D|4sF&#?joo;+vx*zxWjI9N$#> zn6~KoEt}W*>OCLV2A;|*{_0D`cNVR7itb3gi~3*nPVw}%YJ5efo}IK>Cum>4WwVq` z)oJC;(W!doczb#;SK;d4*#AnOqU-UF`D5Ii^|Yxr=F+O)Xz}#AgvVdcSEk384Og9g z4b)ocdU6bpy0*thUh$s4Y`E%~>QALj^}F&$s=l7q%g%*2%CDYYW(+UWjxzPD=To%O z$9gMX`Dm3{mM&T|%hrK!qkWiax7wS^lQ#ETukFeo%U62MW6?@4{<;2Cdhu1cDOz=} zN5`^L{3Fj8?`7(JT0}el&I;?(SXb}onAhWDUwhuN;ojC5o~vVyuhQqr9iH?W>s%d` zkIwPTrB}bFXsy{=!;9Xp;;nirj#l}e*3&D0s{WoghR6DT8eX9C)l=gP!!ejg5HRtQmsj^dP>gmr^xc?UAb-qR6YPWcLot3tTR(Z7eDs67= zq}Q_-BxTc^LDjo>irdyggn#Bkepryk5U6zn7T`7av;l)zg*^@saj&HKsYb4DS?APuoa%t{>8C-ivnLO4|sn_i?0EM^Bel@shcn&gE75 zSZ6OQTzqKd^}W7urH^!~>>SV7k6v~RtE1AC>+>zV=(+Yaw}@|1UazCVwYH0PUu=ad zZ;n=7Q?&P|=U4hT2Nkb*oukEv&hgBpSJ~=A(cVAdN}r-5?^Im5!}F_l$2w|UwU(@_ zTHCcwD{qcg9cc0NvwLnUcdl+c zW4ZO>bNYKxdhG$wMSOF4W1sri#&C_b=$K#e%3qe&bMAE6vWxhtKT~v3xkY(XbxG^L z8b?nPu5@(F>+!iZ#`Iohqv7hyM#@#0MYLquBA&|YZC6^SJ%5>SFI(YWuI8_&jo~Uc z(#}_Tl~(nlQ@_fe$~(MiRrexV@Ax7fdBulL$%`&)b|gGi=Xz(QzU zoob_}9ns-&-gZWNW1Um+YNyY|&Inh%)n>)3?usue}9xtBi<4F75jCqUVRjzodosm}lB3gCrOj>K8(_Vj1@8RO ze>qwn?WWj7M8@Kjz0NC@ny@_-vrty-ALQLFMXUf4?S(Y zaFs!ex2FmBXE)|e#f!h^E&I&p`q1mzk?^wZSKk-$cb+3EY>XSJyX4BUbeX!Btqb3j zJY1#?JYze{mYd@n>-*mnpX+l^pIYBNUw9GEm^YT)+3~6Qt@SWRccjc*U57V)sxRxM z^)@?i&23NHnc=Fx(~Z>M%k}WMW~Sm({D((gtFPLrdeHIL+uf0|_A>psEfel#=fb^z z6)xV2FGH&gI`ymksXS?8e#O_LV||^!&r$8Kd)j*8DpR!c_jblUR2)zLjccRF`x;s{ z+{-NzY?snFqyk+dSUYGD#rt%lHgJ;aI_fLA^ z%kbB;U61a}y4Gu}^1FN$K0G?9mmTY!idP>-IyVpfdG&R(OnT?(;iSU9R+SFDJdn$M962YCc9HgnN67>gn&>TzD)$#W(Wwc<;+pxOgkCo=3$;TKPMJ_O&ooU(c`b6yJ`dP1RRv zMHkgO&dXGMQTfW#R)u$j?nt{+W#-aqeaz8uJ-?jPWXi_MbM#Iqvg^ll{eL1uc!Fd zqhme2j*W)r)|>S1XYYUObVtIyt{n+iKYQ6qo1*xp(&y5~c^mo0eDRTIE6e)Y$R)B|$>K&iOnC=;l&v$N};G4?p`V~mgO|2yNoFFU*4sporm z)8^KAuXjhru_NWH%%b-C{8o5w?Nt60ohwu6z09KiR9@#7uJlE8EHf1!%TMtvlUC!} z5xPkBPt~`Kewiw_v(su``W#joD!s>rs}Ey(#jCDbt8=v8tGWD2ulg3z;zbwzs*gn% z@lMsbsNABwMeWF|vS{b+>B2Sl=-e9WWjcQkcb*>ZZI9v3r>>3u#`@;sQ}yB<)BAXq z4Ug~XRJoDo@Wy)^;~1vmbMsQ^wJzu0pGq%U{GC=_(Vo}C)#ei<;QdUMe8cUE|+ zAC)%MM&%Wq;+aY-{$=P?d1=e6vAMF9UhP-BQ+Yiutj#fPj8~nDXswODZiL6*m_HY< zHHpsgtd~BvJ;v87KbK#9o64`eWoX%1MaMFgj#l1Oo8!2ac}`=Q9`EJHwb0|8Zz{Zq zS9@cb&M$m;e|=09t~x6|NBcZ?o=UIhQMA%K?_9We(f;e{Jv_EidD8nD>FcD*bXw)m z{_E+>gnQXV;hO7>&_(qvo2R|v9miVfMaS`0I@2>ZP-peftSJ~p}{p?|DYwp)&!^l&4qh9!*j8~tgXzzdJFQdnL zxt$fR_UGt&b2wLSNBTF)pGE6os{XNjUt4qGUS^J`r!V5~c{>vB?W`AGRDaFQA|82D zzxB$D`C~ul^2hY4xc*F~OY6Upr{@We`D6K!zv6Ric}y>UwCDBosc`3AHeB!D$Ujy8 zdTGnf%M{;KnrA93xmxR0XO*pZ<@bKc>*-xK^>E3qMYQt^7w_D+ zij_Y{$8tU1%Z%+F-guvn9l4RcB9IHr(s%;VNHsPtoG(bc$~(t7_Q~aG)*`8M6v2T^Wh*o`_;_vY>+}G$b;q`P& z>9)={Rqxo3G2Y9Jc{>uHnzvrhRQ^<2FH_;-8{?IZRvy~ZJ?HV)c`MxKV=i30ozC(0 z^mNd}#yfwFZHi8<)hWJSUk{J%ZY198Sri`Y ztT`+?%89A@r~47W2-i+Y}H%wqLq$TTBnuYpH&Z6{#d@+ulE=&pKzz; zTSkxEaFmt9aMYEft{iO*M_)Pm$}t*_xpEkewQ{VLV+Y4+IPS`S;dm>@TRDF4AHfNO z69*>=PPT@Vubh146u~Liugi~Tz2KME0+r{A6#J#6-EnqinsG$apf>n*-qtLsX<(R)j3l2kEnj4`fUV8 zWsJsp@xZvj9GDmLW1h^{`2%xj4$Z0gHP7b!EP-d>S$HPk8F^lwo#*Jedd?>etbr2+ zCkU(y>%%&+hO8%R&Dygrtz~kLJ@wb%&%qyq-v_@9eiQsM_<8Wt;77p^f`fyDg6{_3 zS;Myu{nnx12)-74Irw7m`QS6br-Dxe9}7Mbd?+{|*gyC{@V?+Z!MlRD2X77D6udro zZSd;gRRdmm=wW!pq5p-KANuk`Ul!~eyf}DKL&aZs=nD^hUhwST8Nt(nr#6T`eGT%T zxdvhBQ~c7#a?d*SS*vYm3tkZD&r5=r2Ks6Q#$|kk*Bp8njNAC%7`!=&5N4+rMhTz@(+_n!;C5O^k@jpyXKeJ}7le-`{I_+9X)z&fxV=uWcF8X?z6 zE%J(-vW~1L>&hCl)~vabJ=P*QV{MWx)~j_)epuhuy0z~;D4F0L@veApyhGl*D+gB( zt`l58xM6Uk;KspCgFSw0=uR5g3Q(oEI zHmIW@{dU3agF7@x!=r8e&>#J}Nq`o_wXOd*2(BAkJ21AZ1mq2QMShh0EBUcUlOg74 z3SF+4SMr1`A*0CI#0(2dr_--9Vx@#{l>y(%ky zap-@6>?l}gKMW2GtV8Saqk*++9ls;6zOD1O1lGK@?;Y?)cvrmJ-XHIh_t!i3#NbiE zLxTGTcMEPG+%mXXaKqrb!L@^HuHhPo9)_zQdi6uE8eB2BY_Lag;oyA1*@H6%rw&dL zoG3V6aIE0y!I6V)!Afw*_CvP+J@|9*hu}BCFN2>4KMj5y{4n@laA5H5;2Xi$g0BQ$ z4$!X#(!UjaH#jIbI8g5U0a_6MLGYvCC&AAG?SZf~dD;bikp8RScY(hCGtmECf};h; z4o(o9EI3tghQK_WH@HY}so)BMdAnY4)4)8N@4E)~2_6(YJa}C2lz<$3e&D&0dE}gD z>iLpo?+x}3tcA}8Uk&IM>xC}zvku7_Yu28{dZr((bMJ$9=W2oX%DZ-pz&m@V;O@bF zg9itX44xP~GkAXRvf#DBn}fFp?+V@%yg%4K_;B!v;B&!Og6{-B2!0m)Cir9U*Wi#~ z+m>xxjv5>%I8ktl;MBnxg0lwa2+kdxH#mQAq2QvyC4x%^mkBNxTp_qpaMj@I!8L78y@uOvISjYka=R^e2<{l%DL_Hm zh}&Od)Rs$axp;8l z;C#V3gR=!^3CTzq8l2FQ+@{Q*zdi{c0a(|3#9P%(S2tFL_ z7rZ-oYw)_@Wx)%AX9Q0Q9veI=czE!z;Gt`H(DnyyzkhJ=;O@blgWCtU4)zXi9_$(1 zB)DO4{op#mwSsE|R}ZebhO2Db#+%+(sdjs|JF7Z|o*@5A=Kp^C@3;Rc_($OR9wj(V z;JJJLXAS5*>w+G-O1$Phb9Xud-Sn!zO$pP7KZPMG; z=^Fz2{oMh*ZynpGeku57z^-6tus^&9-U<6mHVqrcUXm?D_S@fhm+T+ufIWgs2KG>W z?^rrv@9fd;9NZ^(Xz-Z8`}^#``|Q2;etXXk2tFQ=0p!6~12TfVAUntrGG%*kq%B9< za*W`30l9y&V7K6O!I=YcpS&mQ$vm=&<0Z1AYy;lV?L z2M6Z+zQH|%I|sK3=uUd=dI5b#Z_&>e3_MG^hhE<`pmRk>3f>pkhrB*`MexGFUgZhFBLe#6o&h_AUb$vKr`QMC=hG={ zE_#Iz$X?#Q$vb2Z#fGcBy!XpK{ttn@yZsh>&^!3;;46W5ldbN3_3pB(*}(R9_Y7_m z+$6BSF8xAgTr#*&aIU}}diS8~m#v3xUEMR=f6_C53+zYf8v9lNMC6~!?vK($4pf?EUjjILZ{w0Hp2ImXT7MwoVJve!A(%{6w z34`MY#|n-Tc;Wwc$lnh6L-4EMC&3Sb?*-ovei%smS@82U6wkqj9DK+@!GQrPuADMI z4t^T^GWc!q`#>AN3w|9)|HT@FwOxFFI^<7>{3ZB%5HsYEZQHkPKU#3yz!-K5&Jdg< zIDc@lfDWV+=|}THhoW@h<$^te3kBy6&K{g4I8(!!x1V|YSp#+AQSU{9O9b?)dgy0! zPe;;+_X!>nJUV!MU{CPm!1H@r@a(`deRc4rfX;hg@WJ3i!AAo6Pu!ZY-w>Am$>4JV zUC5>?8_HgVtz|v2qpU4@kj-PSV_lvzxL9D_UM(m+XI=JjrQ&*(R{G(VjEAAc;p#*a#;(Pw-he391(_zl=J zcMRyShXrH;+3=!(tawB4&Vb!Qca_aT)*KYr@7nwRDX@@_w&iH6`(?81?7>9>`*XI2 zz5QJRGLrnHhv=Yp26PQQLQelM_(?!tkmKYz`TqA{+tzJcj~*O1I6-jY;3UCG*Kmrh zr`WoCaK_*~!5+a?gX;!23HA*33X0xh>n*n4D!6@cm*DQfeS!xD4+|a{JT`btApFok z-u(mgA;H6gM+J`#9ueSqKyc6Cj=|o+&4L@RLHL%zZG$@n+J!p@cL;7B=$HOpKe$$K z_26oOv0N#*Y;dvQT)}CA69wkQZ~K<*TmBLJE%;0Dr+_Y{FU_a9wo=gWU`rz zCacf2hVyJc4Cmc`-c=n)58}CCK=xlaK&6)+peN~rlKFV(&}xfar61P|^o=ca&wzc$ z&LjUH7VHzS|BUgzfwAE$TL~{ddY!IdA3ZNHXY_yRhxcxO@2W1K1K31tQuFUQ06oCB zCfzeZ#ijGZ@rnI5un+N!`S(r~oOliPEV~BQ!a0Hq1{V+PZ!WzCc2WmlBJU#oafjgk zfwjjz=Lcg8S)c46=N|OK#{)il^do`wynnz(LO&Qt!^>u}=AEzDr~M@GjyQ*;OZeKo zQ~Y7}i1vzi4eT5572G@Eqo70hF6a>Z%zcBG25cek?6ZQW1y2p09y~AbKEEU27h@Nd zZu#EI_g440rDN=WkG$o`tNY?pZ8_Dd@8$wqF0ks;;oITsLCF}l5&sTbiOqD{fKLb5 zQS2$UmHPg<^3PR24n3wVdhMG5`;7laoPEaEQ$9I+a&qf9!SMrftZX!Gm5p|;E$3R* zgM3Qzj~V!}^GeSBWaTHTtg%fc^zxY4Sc3(8r$*J{{1{ z_WS&d_WQ33%sroyIcJxWnNJGXsSgS67Lc3Q3;2@QxNKcEul+)u0kDU`K7#-M)PZvv z^3=Y=SxhpZ58o5JoZkt)9ymK9d--U^*$CDsUoLl&={ z4tpYY1p2CA-@rTZoPe+3xdDH{OM(}#feh#GcZNomJFDYcw(nwxp!Qx*3hEq>jpE!6 zzq-*nD^M1k8{n5tKS;Bmv&XwjaI4@(f&HL;p>u+31wee|H&-^^=OMP#XgfSg>A*30>XUj{CW8H3dk1q0)LD?NY_r)ss0Y_ zR9yRa44jpy-`@CmU^1@)>fFF&o=taQ`Ccws#hwl}^%L8%sKRU2~zFVLl&QZ!w#E*Bq z0KIW=)1drE&OhimXFcZ$=s9C^mf`$j_rUn+B|7f70Y4I5=iI{H`rv?`&*TfoLg9P&i(B@=n!|TtV?@#>(@PlI=i4lgy|6XMyzM|72X!u7m)!ar^y9= zWU_*-?R_Lu$Y!#%@86wi*u!@lo{S=^yz};I{87$3_@q$#eAK?~%z-@~KlpCJNdsr0 z{8aY*&T{zagEslklPOsX3eG<9ANoZI@qcKMeTv^$!6m?rc~t`^Nab z7x1C*sW@Z!X29mbYtQV=*m?0OgVO|O3eFbrshm47zo>twt2LMPD$^d<xphu zmcFo#*n-xEw)OA)0o|?~9}u189*?~Y`ENhZ{v_+|&+U)wvAiSRo4T*%EWI7914(A;@v~ zwj&3-1U8j_+xoYyzYBgI{3!Tg@Y4YA);Nof8=Q0v!pbV=cbwog>>cM4#3kK&3&K{g&4QJeT#%(7KywKkdJ|4U&ctP-l;6VYO4u8}2f@=g<4^Y0G zD+Ct_&Jyez@MW2glLe;?__U549J1w*Rlkk%l>-7k8Ty0nxMkp6oKB3@#V&EASojEAu7uEuStpW$+)tF@s|S#|cgxNave}T?27AMZj0jN6+US zciMJ-@IME9=f4Q}=f4?zCiqbB-r$|V+k>|TZwcNIye4?%fL9#yibM7dUKl(tcz*EW z;H7Jj{*2)9!NY?G2KNf?65KktNpRiZS^;|FV9zxOUoW^$aQ)zx!Civ~1&<1z7^qvj z+Iw~I+CV>FwT9On^14Ib7Q8PwAoytT;Xru5-~$csKjbh-lmD^cv%!~xuLs`=z85&V zaP~l19d-Lr|MMKdygNtWKjKfiRB)N#l56KEo+;G%!j8z_WLzeusch>)`?4 zzw_=F2m1!@3-F7*Hn5iHJ?rd%fZpT#rzgJ@xEp0X{yLyrtyO18be=P9_m}7oz8j!N z=pXw4djj&G>?ilxJ@yrBN$&_-@wovCb$7|W-rgOxcW1kj|MvXS?fJ_tVGFyT@yUSS zpB?SJB)`3@_DlBYbi!EzcOu=}v(NHQ1AE>c>;{27n!UO6KYkngH~txZ1M-63mH)#& z(fuTQaC_~G1@@zN4;~Tl4ZlD5ad5n?$6NJDUuNrNR^{BD@wDG7czW>m;IqMj0U1qh zlbt8rcEW9E4#?Px1{Vs*JG2Y(BWa-^di>FB|cgKfbs!7+p52geDr1h)M&_*(F}K-upG z2M1pdwD-0^KVG?peYfqq?b*TOf=2`o4;~dfGPr+mtKcfZ1%fjLy9e~iQ3Lwxo52CW zn}e4IPYWIr+$NwuE)bk9pf~vb2`cB|2M3=G-W~AW^V{>;KPb3o;FcEOJR6A5jDBWI zoIPMqxpT%w@Z9;L>fRFH6MM}Xv0khl`vm%9@4z14ojL2w*)dpyrw{Do>5{sGWdGqF z()WYfuUp$}Tej~z19q)RA4`@$$ z!oXhe_S^-#xfrz`u<0qmxgU z56CmL?gBZ#ba!ZvfV=~;jlcW+f%DJP1f`2mXUEQr=`QE6hvY|&(o0(d;Xejstn*?J z{!LJ4y+6tM@ekL|k){1J*k$`J+m9Tm=O2gsWA(h(IkfuZ9X)VHecV9%Cksv$oHlT- zO>U#~EdM>9f64GGZNJj0UUok2jul;9pu78BC}$3Iy8BRf4(=a3I&gmLEdKey3j=<7 za-0q0Ik0E=?a6p~uL`6&<8?mb`IT(<#l&vF= z-6Nl!qOL!#q4?A%U0fT=93!AV*nVsqX>^LZ#NGQvtw+4h39MUoj5W>o59Mp(H!D9G z%C;!~8Gl*%p!nbJ7d$xNhk95bT)rsp&VvH!^i-WWd#}7_^c8(nz9%*b{}aCw9~GP8 z<7;OO_J!;XHU@tao7_3t*8?^Szmt6=pD;ajnt<)kr_ArXS3s7~Mb7k`Lz6Sk58PYf z&wh9C-rxfP+rwGG_kuqJ?o06ju|4((_)VNgI-}&j;j`g;amUBKA2QNiqg}S{vU-k8 zj$S-qgFHBRUhu}?0|A|7fA3{G*|w9d?*8d0x`}O2`sps)?y@S^AGGa3+a4G^EO>nI z{NPo=>w;GV&k5+R#|HAB6g(loPsiOjV2fNnQ2!pmxr37jY!|xo>%r%PPX-?h*d5Og zoKrtw?dfpY;rzQpeapAYyUouyho_BQrF{Iu)_@|(Yx zKlO5fGgCIWcfh;g9=UhQzKq@AJr+jEZtpgKf-^dB4u|qzuq&M5xlhWk@qxgZDY;>f z==_h2v1eps*grlhAj|D3!C4}iBuqY$UF18vje7;kt5f=A17+xh3k3GYY??C$UDnFe4qJy@X0xzs&W+~M8ss{e ze!PHOXA_b0!t{*#)v4dwbvA1pZ?p?7}}c!r*t`&l4-aKLB8@8ynZ`MStW_9VN~w*-7cd^Oe$zue8%d^OHI z$iO=LAPXN7kawuKJ-V|QaqH0@3ePr$bM=26yc69y=g2j~?*Ht5aR}QsT<( zzIM-tPu^W0?_z!Hhl-zX4f4+sun*h~s_*~o(Y(84v+s2HW3{baLWf_e_G{$1@f))< z1>=_H+?fBBpMVd_T%Z+Zs~i%T6LVn!x(8c6rag0)=Mus0f%#s_~pDjyR6_*Vk< zyS?Bm1AD}h=k1DYdF#N6RrA-$W!w6NP)dI*=pa;=V701e{HY+`{1|1uY=mRAGCIV^GApL zXw}DL|L#uaPXl}UgWKIs{-^H-UkE-Nyer^)dUvp2@R5Ko=HtO91AGSt+Oxms*E)7U zKLGiEf}r#VDlWbF=nQGb%q}q}_WktG#REQ@^54)E^xMq?{uzF|y9MU@A;D7v`}O?- z_h;wuxoD!2j;f8kv3G-~wykyqE8$>}qnJ{4U?WyBI(Z&?n9Uv`L51C)xpdbkD8< zO4jR(zLLkz54@|=^ivvLL;gF1Q=hWhDd;o(a*2Rn=}HauhSv{#JICK7k1vW2qL0XW zcTJo*umg-)-h~2bS6=&0*Y~XIa!2K6fp25^9Qdy32zOZAJ$iV+ZXk#1`wM559}f63 z>|xy@a^J@Xmd-<+kJ1hF$=QN)2Imbf7F;>-EycZqCk4)L`BXj@IKLn(-LZ5R^$Y=j z%q0T8ud4(%4(=Mb7y6vw6#-w)e!*9R-v-Cnc8qPi2B%*Go#K4qV!;)ID+iYe_{P`( z=w*Xz2GBDS?Snf9cMNV5+#eK60EU&;Nl8wYfV_rd)G`viL{`wj1>yKAoryw~)IJ22jV_VJ|x=X(DU z{4MsoGe@?7b58n#zkq*XpTIpx`+EKd`*8buvdn$NlK*r7x$kU%ytJ1l!+*YpU*`UN z$?G5I-aT3V&EQ)BIn4)7et$doO7QuBpB%_$ac2pi2>9EbDeND-CwO!4s^BHTO9TFT za-Q5L-|-(1d@vAJr#8MCd@fLzdYnnTDR^V>`k?A`{s8KBX5lPD-TLyWKpkHX=pf_d zKcYvBlYjpkYY;9SMb{X&dEk5cQ^59crb2Hzb1|QE2>XSdLEo}H_>q7OQoesSk?#rk z{P~OM6TUY3+dWRtlbu1Id@S${Av=OC=xmw2XdUp~@UgHT=#x_g><2zWYle@79^prE z?};zpy0S-c=F66#Ti6c#D8PPTXSh%0-V!s5turADhJ4$L9mK%DaPC2QLVo z9y~eV&)O%@=H7umTp`#aIDc@4;CR7bw*F=H4Cqr^KehGU!7GC&1$Pdv6!23WC-7kY zly4V)68JXcodLb_(134`ZuZ^dIf7#a?&!J$%U-k(;7_z?*dwr3t-+%N&K~UNJ`}t? z@LswngLz(%#bD5_dID5s({yiC@|r3>io!))_Usfqpq8KYsecUk~}~s_bPi*#DEYWGtOR*L*GbQZVWk z=_Q-VaOJvuRu@_Bd;lNaLFUs-bV!v|jyzWfy+@||G3uQkh|`JAd4>P72IuB%p0a!B z54H~~Oq;PdG`3^s(!N_@K$|viMH7*>O-p!=7Jt%i_jtZgoj>av&dt^6wIS}Hs^d*bizRa zn~m+}e1g6u3+M#;f&TetKrfXIr+)1iC+MGk@@*KOdEvhTb5>{X_N~G|x6;$@+LKFs zB=-p(9JrIuf8vgnvyJx!&N@){t$Y{Y4lLV+@5SEvNbzYoi+8`u8LqQje=EaX(Ypjs z2wofPA2JdgiIYzQJ1pb^x2;7Xe+v zzjT~{ujzEb*#mc@*c7J;oKv`S#jZJRz^1`>%Ao8Gdf)_sv;X4XK>t0*wicbg5d0rYkJ$QIvPOcxYLFnb<2T`@F zccjcOy}?#=-|d0Ht%9ou?oW88-W|S9=Ud)scdYq{`4;R2-F?4)z<)(=usz5U-~YG= z$-nXW;G=;(DnIyZgBJ#TAx~JtQ}W%aIQjK}z*#b1JDpKysp8HGoFn^IihQI)PQ8ZR z@@=lOTKa_l$-xyFOAeE{>=gTV5H49vZqqURL(ZAWW4^5NXGyDc_3{hx3E9(YtIib4 zUnGsaWPi^`bcuj}h_A@LUfKoMz6%p3-;KpthcgG~;m#e3m&})Tv4HG1e&B23v%6Ms z!?m-Rn?~0P^FcYIF@J2kI%lRc%q=^D&UjcrU(?&q4EP6~CDR$sDn7o3Pv%6~bENyf zz4i^TXX}ovvtYV}uZoY#`IY^HJ8|v}TO(`)w#2Oh_QmZ2`s2v~d*KCv^9T2o`K+9i zu^-;FhS%mk5S#Zcf&I&u*8au?pOkY~>ytdQHzv#cYL87(+f>m3i zp1hMrW|Q$`9yzBRDxDr6zsYl9JY=Q%$xiRgY6tRnd8 z^o2P#*L)DpgPei$A@U`%=bVGE-{=}T29M{%J}DhvdP%+7@H0+h0&`=G^bH+DKaxdk zv$98=k??^zGk;XzO!ldPZ|YtikZ1IWdprIXg!}#O>i!^bzsEOd$H+HqrwuL~*h|wP z&dzTYxL?Xnd42HiV87s_fqSU_&W5uS=jIm-E*J0%-6?R+@S?z(^m~H?f=>ot3)m~< ze6Cn+JNg>fF2ctS=mdIY_khkgbHJtncci49I^erHML>VJFGVl$r~NXZr|LfNx3+$3 z)tB|boRRZmy)Mu6 zM#z-Y1$^Nr2#y`lGkhEFvivr1KJv|gZ~apNp9tM!|4XL5BXF;Y58l}c>RjZ-!83x# z1`i3GeY0Qq#`s>yTJoD5=C2{I-2>v!aV7xtQ1_S719b*MZ#aLZBZQrkuy5E3Yz8{Y z83}z=b`Sf4-f@PaEPdlXk~0_7nS?a@2P)q=1^u&Y!*0=yYzlrob_Bb^SV|8$bEl`+ zW|t01&#>3%6Z*q>_g#Z~1f@HkoO=W2j!u7L@P+_=cR***6?6of1KeG3p3SD*H8@Lf z&cM9~&)7F|bOc}1LxOz*cXS>dJS})h;G0KxamXI$YHSSWul6aN`Og1i-P+5NLl+NFekAq?l#G(czt2XIhDZF8f%Q*@ z38Va0mkp%jVHaqRE@1D_L;6eyklDuNogvfdG%}TpCbR2Inp`EnQM!=q<`=T>H%H=? zMvnU%f#kUJA9@vp-ErcpL)mff3+(^Q`+I^91p5c5djf9^yubW?b#K>mlEyBiKj?kULA*J+QC8X~6IG;NX#gy+0dZ|G?c?a{iA2 zee%!1Hyy_cP7=@`rwP~}bcXX4x`d5Fm#|qnX+%+uQ~FM8GC_ zZJ-p*Kj`KOwGGx&*VL3^IG@z9qz1HdoK?7wtV+yugRb7+g|4+ zzHz%)K=!!L&IdxDxT8ouxo`E|;EMq{Mt}JG2yYJTk)Ia0tGQ3`@PI#Fm~Y;>uYI_E zxqD4Oj)Hyr9s$`)4!d_mkC4m4K!?yL^ap(bWILHI4?nwyOlR-Zxq`C31A(f?{iu@r zbe1*&4?U(IY>kTtWWVo8^q1Xm%Ydyw?i-W40>aL+`I>6X4_y1(8pdueo*1y@*mmp- zcck!`uV)43?`3OuS5dlyUHIYP)4}J~Q1-_6*UmBAg`(%(LG{~~yQK7py#Sljb1eUr z`=jpFu|IsD<9jsUu{o#UyW+#5V}zZTTO&^ieA@&3SOOIJo*8L$tx=TI25dKW zoHV+c+;>iCe%(o6-zhImn11(e@+Ei&GpVNT3b%_03_uK3T)8$HWkNw=`C*i&>2`^El(9mn@) zKVYu_WE5Rb&y)}BzWFAcuf>_Kzn|;xgScC2@BG1lPlmp6cK_qRoh~ww@9S8>NdxD} z=L_}-$WG_Xd|%NY~< z?d5^H3=a(0G3M%W!8wAH1V;(@7yWIL?+50e?qO3tBDh~+r(kXy$J*f0$B)e9QAq(R30CZL1JM79fm{_BF5 z2hR`4STdDtbp}8-^Y6f&0%`U4faJNqYcJ1-hSE`dL+l&ABKLF2f8TAecjzm30^IF& zhnKGj@X=#ayGuCguKJkYHj|!d?(5=prP~ewxrm=6p zW_$G-*lxn+gw5f81N)Bcv0uP{Rp;Sf%6BbnLOMm5jrfDWH`;#;oQL?)|HllRi@1N~ zTTbU-WuIIje$Fekze(a7guQ~>2Yg)Q6?(UT{CZ?S9=FM5%_Fa|Q${@*yj9^V-0R^vAhV1Aq-IZGnP$#$~ddn1i}H>YGh z9YOxnL%@GXC(u8>+0YI>tWV^8`3=2`e2L^go5Q?ommboe7X{MkdJxuMWzgtS(4a)vx@0E>NenqyUyJhYHT{UnI46YU2 zF!0?Odxek1y)Sa=C4u`yZw{QN`2N}7xUiRY2kcKlOybqQ-LzZo5}!R_w_G}Kj>5Nd z`+)!J@qs=1zJc=;e-rP3;PZjMbL&jycR`LXw;XfpF;~CaKIhhRuHKKjM!tY#!>|qR z6Yz_Xy7wPi11uqI77u+LYZ?Zj|lX?Dpw`_X;3-7J<)a`2T zkv)@rl53OTevt?G*eov&l&61`VZXdRC?0+=X?&0L4L{hyYhcf~ zOF{SiDWLcHG2P9gb9M{(IX%y_2mGG&k8}Bp2cEn0dAf)l3v^Q1J2%T+Kzofl1*MDV zBdGnd{kFSFF*qoa|&LLCJFRm^=r2eD#pC z zdXw%cy{U}0=$g_gbUYi+ISc*52gc{+$KQFJfZvOL;rAkA>MVsl!biDBP(Dca-LAIw z|9v=Hp<8_8Mpiu_aK_?}DnE?xpx+&^Q$8B-(fC%{`HH{i$5wH^QufO6bKdNnM=zcPiK`?uEN6?%ud(>zVUs^OaGQ>=m{OU1J^F zW7;d(i+XqJ{DmB`zjY6YZ{Bx`bPjpMemQ+`ih%6$J?}07zYLj1-~2TAR$%YVj^Rsl zCiB67e0+Vto*_?P7`z~OZs072{&`%$#^G0DBk(ctKLMG}cA%Te&MBMc!SPQi1LQtk z0OIO=%G%$_etPVTZeNi1dEH+X-re?VulGst<=*!Zu$9vl?V zvj+y`KZyS#_+wyxejm^S(%5^_{;>u&3H<={INQ+t(iL#D;Ml z|Dd2d><=yyTq2-5=nMCKovZowh>wh2N^jgE*n17mVLX55_UzY^v(`rG4KkRYg)Ajg z#mQnIugAPfqa)~#@*9!)Ri`p^5V@>g?Xdy)d%B#rPU#Ok%9l*1!%Fwy(N^)QyYvgb z(ocBAN1YWVf-v36jF~1jma$xYi zfDg0;BMH#!Dz#f z)og*A1Af2+$qEoUR8Tj&kGF}BSQ0_XGmVEkeHUwmV1 z6%gh(bEd;~@eLz?ns4vvBQ_FS$(=CwpM6W??l9d$&(U{x58MsAc5vB%EyEw?`$oPn z_ZIlV*fDPnUJ~$!vGwQ}_a819uwzag&@+du9I|@W z&)mHY`h_iHZCRK0GuE*8g@1%Tp-cSzAhO9Gm+!vL_4&idH+m#jvR1#1`$@ic|3*NE z91y%OAdBA-ydl^(Aj|0r@}B&sFMxg+{bT%L&a$2H^Odn_ZWMGIhAq=k{xEh;`Ot2h zd(Agl!%gGkVwc<|P-orErK{K=beHpx#|HEd9rv7o&U@L~ImnyW{>BI018)zE)0jUI ze75~RMRW{(^2>mZ_-lavHu!abvi-cU{8~VV@Ga6OV1C(&@_b+J9v8jBemrjAY~@4& zo06X47n4rk`^HF`{6e>1=pg6&1!pt#k-z=t?*W!AV^2XJ`Gy^Y*|+Q)ap!e62$ACQrcL`8>kv^a& z*cJ2>+e$iHtNek^9@*~ntMw1s(x!gsbB!6T`C!u-hyLFzpp)5k&auiaqL)EGO24ve z=+Lf{#MwDzvz2b4o21$My)@`PFk!Zg_n1zh8~D8V!<>IfL0o%YGEcVL11O6y- z+g69V8mKO(f4*P|@@%n(iai2?^E_zKMon8Q8>B=bk!a#lMWv4hxxmVz<=Z?Xx zgFS=m1y@`9`;6?C3k7EjP8l3M`1_W>ul}DT2W>fM)eq<{1AWR~ahC6IY|$mx3oaF$ zEjU?lw7^{nJ~3x2?8UNKd{5+gIoo%qkNxSJE_N!vsJ#WBsdG0zXgY;GY|mq_WZ&i8 zvp4iUddF@Yc&F(X-_o;F&JgStkW=&vSx4^i`P*;XcaxWVV?PR<4e-<0!?R!918|nW zenFih0DWZN&lcD(K-o4Z8{ypny8-AbcExMgd`K_JH~B9O_?cc4ye>fblwK7G;{`nU z=?Z5*D6o~-FLh4BuZ6#K2m1x|!}&Vf#NVE!JKW_3di6U2UGl@g`1z;kT=v_~0`tN~ zVXK&@(l>O7`71p_f6yJr3CuHHB8_cGXZTx!e3am91?B6a8%i&DX6(yzul+4relK^+ z*)VwcI`Mn1^({Qx2H2)}QO}{8Id1d{6vV-Z!=dnNQvWJA*FpKJp_R7?Ahy{Wb7Sf%o>H zHOS+ya!-q1pbuY@LM?#;UlAG+%cn9-WE8c*+2MT z;LM-h!N#Fq-Wbp^^v}M5Z~WOI&UyI1=(9Tq^xF-Bs|2M_&XvECdB))60iE*aeEa$R zz`4=~19y&J8I(?OKkVj#d7_6e9b7P=r%x2n;m)<(>*f!m=f4p6o6Tqg28}=c;39M6Ro_wv=Jssm6;XAOu@*RWyAX!2VkQ4T^An%31B)ebXN{1O38>#~xub0Q%9=Mm2OiEYD9Q7^lPy`mi`n~9ymPv)$~SrG7n(Rsj5 zd3r!sJvX5LUKFrTUKMyZouSZEeAn(GuuK%V;O}x)%-_XEu}`*NK5DF!{|KBb^KqRcaOaEs19plte0iq}&K&TA*?Y5B z_6(d)J7=L&?i)C7*(dOw1UZ-F3czj^MkZ*Z8lT#UCrs zha(05%$;fetb+sJ$GcPPZUTG6xzQT}zCJd|9RlYkmk7=o(7z`S__ye0KE)pdY{d5m z&QaJTzN5c?U?1TB1@KJWy>JKPN`XCt^>Nam?33?huk(fAfWVq`hG(y4uVpQ>$@#Lq z4|@jQDesbfq5UKILe6;Ko&A=*LXPo=k&|R29rA;~-LM0Lg9CeaVKSKALY6A8er@W5 zeIA>CpTN5=k4z#<$PfEndW}r;UXvl@6x)d(hzurA$xZTB{bVqkiaaG(^;HVb(l(UT5;oYV4)>#pna~ z=AgoKk8x?M>l!|={;qgmpjrNg18*^m;TZ@ z^bDO+=O}cEvlRZWuLXQwzy|ti;C>uAz`prMz@`CVf5)ZHP@JF8IleU_M|>}I#lYR5 zdjxzUzN2GXI1`~4*cIJxbY2d{1NUEF7x?zVJ*M{u z{8T>-j+}eIXAQ0yklE#ndPu&9e?V~Gfb70!Q0F1O|8NGuPWV*deB9YMTY>$+N5xJM zW;-}D_x*^oblUe9wujNiMTZ68lgPgCQ zByhIwTN<_z8;QN-T>S;XV*>NSrr0}RQ?Mu4VstB;Vz7* zqw?qyHh}U~k9Oz)Jmm{h#$KarkFqtiNiXT2b2+*imG8Vp9o7%o3&M1T`aoD4WOeb_ zN7NpN+=l8$r7LSp#;f1fo_h>*zcmQ_h5d}sd_&oUrE9HI&w~TT^YmP;56{@X$-1%M zuvX}5JoLCS_ARx>Ju|T0Jx}?bKVEgSUCZAoU;CiEK7hTFcf;PL{G2tH#>{qNFVR)~ zEXjNAedf#i|M+}16A0EH||9(9l_qEUKy3Xr7kK=v3kK;VQXNQ=fC!)sCKj%aF z8&vm&{)PCjCRi!t0oPZE_Xhp%I1v2_I{EjZ4}3v5*4wd&^u9jv0zrH8Y~%s)_Re@( z$EhzFFOAp4it*mqIJSyyV(a)+Y%*c<8Jo}eWPBnvh_%COJ{+HpJ>rn~Mf^Mth;PJK zV!QZKd_Fdcw}oRE7mTNenP0~!>aEKQdf|>oQ0jdkI? zmgacB9p;_&!<{49x>6tM@#tm94SE=NHAAglv)=hj{6F*7a)X(x?Sf~xVek^whReot zLrz#A_?6}(9JlDPMdF!3BhUtXRQ{7$Lh)N)gEpdP216s!M0{rZJTD63xqaQfvoE*Vv|@W^zc6vYscv4Z_s(&KN(*LpJn&hBY0?*p>J<~*sL0#&86G! zAG2y^5WMEA;q`Ez@9MMg;COF7>k%Q}sYiHyyg%#9I#ruk@8XpFsV0(}{1b!ZCV6(m z8JFrW@et}SzwdZlY!W}^ATeEDuq}wew7j}u)E@NwfR|&Nb7>nK{bxFU=P*;n%h7Ar zkJG!LJ7|5{dZ!7y)Ti@Yk9m&gd!1P4evj*&^N75cnnry=vrMl$vD@c zICPCXCWqP9#c12NZF_jJd|9X|iF=9Qv<39V3y4-JMIbJ6( zjapCKx6Yj#$94H%z3sD&9+%uI7pR}~;(r}#4jNHUgeu`>>fgW}*gDt;MfguhL7S^PP!j%(t&xG`>z zyTkHk%sFGu`Qmx8a4Zoa%DD|%O$ zRp1}#ZErF8+hh9OJBMR<_=NKMav{g-b3OjV<5TDU{7+t`8bbb8Z>mMr0`h|G-u7cV z69?!~+p29%E>J_Oh3&8QC;Jf1L1WP%wn4pAUW7~ABkkq>;rZxg@HX_n`62&vT+aGG zJpnpejN*~d(fk|tiDhP&j-Sjk=|7(q?jJo4Ivji${c15$to&?Ep`pDSMNj2hU*^NHZ}`$UMt1RWAS)iJSP?i*B6S#Vu^6MaJc`u z;rg=is(4eZ7;lL;#%p7_cyV~G*E}Pf!)30RGiHlf{d+VRN*-GYxz*-h~*Bq#U{V$sBOe4+m4u~KHRWwhxB2|$U-_hZ z-j>Oap^5boBawr{QolLI8pg<10EmEudie-Vft*ezJj>}eFoQ84t=OBh$(N8A0+16 zm&9^@C+$Pu({J>@x?R1j_EeklTxc9x$8C8*4Xxg{4~-rj-_){RA8J(llh@Ee_8r^u z^f*4QAC=Y?e{B1-v3-UPrpe7M*_XtAy4JoWZ;10_KNIWe;hkcepv$+M9KSBkZ#cnq z_lx^q4*TPmLR_>TVmxMl6(4sD&#}x;gxv9tSU#4D`C|4Ee;=Bhf9u}P2lIuZx&K{%qvWg!=k>Qr zJm!5ZcmB@v@`89nye&Q&p9yisIuk3ziRp1oTv2_szsj zzqgz~$I`U6BN~Ao75l|;abE1+e{wtme^Cz5TjDJay{?9Uyuoi&yN=u;M~LxyQMTs; z!}91|_Ya1T$%B_O;JWuybErXR9~z&wm$%eA!|$QP$MG~H2l{>k&*SCzuIA6|7vimY z$#cYOdPd%mQ^jnW#d?=l^4c)V^;2(W|tt*tW%Kz6wo4 zXNsv}?{*W!VETIKowoL-KiwDNwD?L#)8C5*ZM|f;Ts$m?b{CUrYcX5gJwHy4qvMeHVSFpT8au>Ru}OR= z)`}J5O|e3(8EePJ;dx()Z^V9aa2y#&#Sw8(d_TS$-wBudhWotdkHTwy634^|@$2|a z$OUr4^>KHo51t*%ggoFFlr6&g<1_hP%x^(XI6UZ1xs`uJ-$L9Hv(5;Q$p!N9i9v6w z#XN7sGyWgHoA)ku@eXZk@&JE`cj!K~f$h&F4WcKYt{mRK=h7(Rzxq>dumnqh{@Au&(*{k$-D!v-$1nn&*i@}S8n9rl(&+iawf>*}y z;hm$y=W-6)XNBu>jq`7Ke_C9>)-eb!-M%6m>*l;3uM6J(2SRT^j?iE5efdt#Bjx*y zdaN@Z>{UPC;xBB+G#kS#h9kqaCR^AeN#LI+U0HaF2 zzay@SbK)0aTe5%fUwJk>P5QyMEZ*BU`8zJ%Ru9`g;W_ex$N9T_9sDFP23)i0dR~IJ-NgjOH3VG?0ggYs(Pn-pfojo?UIka zPw?Kq9J>YmduaSDjtt%kADn)FQ_#ZBqhz_xa{;>FvBt}E{-k%u7h>Pgd!@yBFIU7R zaaQnfejN1n*Mo1PFZ;!yxjza?n5X|J9bHcG_=7St(pf+*LnLY!}eO#Owym>KQ z{^8m4?q46PhUKmqD}|o#{GoRFW2k$+9-oYNhhBzrYWT46*+oVSB_~@wd1wZU}kd;xHHV+n^=zn+f_vGzX1jKj4+p zU-E)Fls3}y`d87 zbwl$udrJR1Jpa{`bJoAKzeysGy0q@}EWGBu zF?uXKH90~LTff?GdQ9+?E)L$H9H3@YFDyNQ?&hiJC+IKeJ@6rIduC+4#y@oi9iq4D zw#W2dZJRU%&t>>qH|3%0salrTSiYGcx`sY~Z>VYcnY=3b#(p4I@RRHp^pxBnzsMi@ z(((qquLkfMb%31W`81fkaCO)}?+$t7p|Bt8&&ebDvGRi%PTsC#%j5|@{hl#;wBm>U ztT`0kgB{Wg3{T^?aZ#KfC&$5I2FCmgZ(PjZJH&gv*IxyHMIQKIED?8gZV54!#ujVE z+y5qG#)px3g!6{Y%9^`fC%X#J55p z!MT{;7atDC6z>?$!E{16r>lJAv-!N*cN(D`cx zkDecPbI7A+I~^}4e!nADjg`Y}C;!ZSYAct{(eCj#gyWMPH?c%46)z6QRIC)@JY4rY zUjKhOZv2XH{2!hF>F|3y9>jdl!%sXs)vLdBa*Q%P!RMr7=o@vSZEv^Wsq!*uB-wnsIo?Uc^hF=#4UN?onKR@bV%X?^n?W5&To%KtR`sK&SN9u)h9_%Fus zXLucQfu5+?zf|afig9AoYl60a{REo$6(KK(iHn9lEzf*-NVnx5(dBeH?JW-TWB5I^ zJl)UB*(P?5y@R&*JTd(6@VXO%Z)7IkTHRfXE5*mYUv3x#`I@h+YhOmFw7xbaUbe@rT&1(^}>HaI@x}c$k zwi3VjsZeu^gM@tH7RsRyfHO3#buVmUwOAMt3+Gv_>WE)a{y3n%dU^;|ti zUie;Y9xH}E!J~6LIyJw~gOPVm4YNZ0s0+i4KTnEJb#wT9Uh|C3=E}>W^TquS#m6Ve zWvhici0`9sp_Xvo3Z4~z$MIn1b@aM#4Sm^5u;9B&EneSw%e{Ci@y zCuZa4xIKTU$Cix6Vc)Vpy)o<);yw@5ypBHDGGV_F zr)hWlulP@wi_u5L5n=y7G!BVzDL?#Ra?CJqpN}xSAf5qD$Zyb3j_49+~e@fWV#5b{DU-Y5yPw^-A^Lae>*Z6bL*u0YCLmZ{$U5ce*vByr0>9JYd zroH*@d&M^=xc=>+zr}f)oc12j>h7n_JrDB5(CW1Nc-^714xQ5aJ_p_Y%a9l3gqdMk zL)Y_zctc{nJZBw<=i;&)MtAdf=y&rzABc6s`nA5TfBozo;|n3Dis7Csp1(QdLpt92 zRxd3bZdou=;Sk^TL~T=iMBXA#(f0P5u$_wa zw#&!<``DCrAN^LE19St-AJFy>2K_!O{PQ|>B_C3K;Boh<2W;oIWAC+c@IL8#`w(43 z4~hM>)R;x0>*WE*0;`iRj*G%|HM3gzjQGQ3o|2}&FZer-{arA`e*5qWq5tYqF5vyp z^}JiNe2(Qc^Zigb9}rL3ajW8#nDxT`-i}ztC)q206g(U~%wyuDI5#ef>*L1wQ=A#* zCC$)_vrh@1u<*tubt!{-(0jk?@96vq2IuZ zax9n6PuHIp9zQtLE$fH=$2=j%pA)}`LxXp-PwX7dm+{e9C6t-naa{Lu?V##TD^FJQ7#L54g^IasOMxbNB@O zXs`| zOm?4{bot!&K%?6>{GKDP+g|*(^82XgemR*fr1jNydc)pJjP_5@lz;!ypd-Y2K0Z%n z=oDJryowmC=H#=8#d@i>b^TO6zPW?-!t=#!+TQHTn?o%39^(4q;XTdC*oJMt^p9Bo zuCVV|zGaK`{F?nkFPPVPUhrl98g#tlEbWi>PhQN@;e0^LhJ9C`jaTEGLH1qGbDw?M zthM9!7mX!?k7J*=?>o1Hb2{+x9qTFPUmx`1CxhN}P6A%Xz9GKd6#B1?u1{(7h5LJ+ z^zL%8a;zOZ{_TSoBS-K&-X5OA2j_S2>cx1+&YIJ4d3K1I^uF14`rmW)TD>2ShhH+@ zPfwSZ;yv~YdFQIQD;^8+p8j|H>X3irD0zanv~8#%HVb_PeP20-XT$H;_tnEygQ#=V z0>6thV&nw9-V?*;mmhflW{&g$_+Vx<#eDNRpNO?%r7(y2{BRtve~&b~$=km=_&tZk z9w7(N`Er1HoJC^3cxucDTK|FgYdBB#8F6&z$9yN`0)1NN=rDUp$E*F`6J~Mb19{)J zFy=RHTedNp!FKmx$7QH9X%5@xLBU6~f2eisQ}TfAU0kMvXf6BU@Za@M?VCJLzWW^^ zPk1h`R4qUksfR5`PH`#L(_-?=vw|nzSAK%75=(9m^}F|?_q>NX|0iJ{hgY*n=rO(_ zcr`o({=$fbyoV77c?)zVz3I8U8a@qwMy_~ExOAK6>{9-aJLp!Pklf+YObhM1Qizeu zh8Qach_T{5A6&g4){3!zit~cD=Cjb-v^fq4{tZuEy)gV4zP;C696a``~h)kKCS#G59*uK-g|}h&2tl@#pUyZ4xo|6 zdA^!Uv0r?a4|w0Uhs(q9b#wslo&Uy*5x;E{wgwtLDcM2Z5!1BIGc+ zMIKl@%qI!9`pxo^_vQQO)%vXT|IF~&z8D{m)h08RPiqE;mve16Ci#@G4(a`4Cj6}P z6Ir+4jU8jN(35>#EEfElhpNTyiob;OWQg~!-yD~Q<8*m8y9LkQIpE~>SBJb$6FfEi zd+dqFrgX*M`J9IkDPOp#q*Pfcz`G1d%xNd$M(y6EYvdE~H_L(Up8-8$iQ`QBCL z+IAuSzAD69+W1W|JiHMDU5bBn@rcWOLb_Tk6${PK*#6uWAI)E`8Dh6NFxPFH9$$Yl zW3Ja^`xUpv?0w^aI4sN%h@pH-+dIs~oD^c?xf5T-Ji_%MuKq3V37#kJira%er-!`e ztWY2G=Iu-L`@ozHPli|H`54#5M;>|(B z@on_!R*TmL{brw+2jqaig!j5)avm1f)%NQBH-xxPFPh)}M?A0dPU<0kBAio(9{o*R z7I%fZ;I*+%d^ywyhlPHuI4c3vy&{$l zy?_NmKSHiiho}kEFLIRcwq?9Ec=u0O6B-&VJ>TcYkUBUmk8-e}4$;@`%_w zc=;a>eOz-lj(uA)77Bhn{2tJwHIx5n{=GcFxBpfAICwo@oE&RQ->($55!;IT&$jcl zuuZ8gFAG}2EHW)%pP_ryrDo~XyJl_70PEY18rZ(6Cu`qT-##zQL-007JQojWdR{&4 zrcb60=I3~g`{*~?j&7sx`94nxbvti>2IL9Rh_wBgaeS!%XNK6%^QAlKPxF_%g!ugB z5VL92jY6E~?eHnYJzhRfpI^ftk~h4LmNolnd3wE;DHigpw+PFUL%epxdHP;_7uUsG zUJD;atTv;`Ke;L{jI)EU&s#q(c@ zn`;sC`Q?1_>6%%Lw~f=wV)7+1J-&$%DSg_$|%bNvBxl#an#EBHcYIKLF)yJM-;24er7;aC8DgB^ko zBsZ8vayhQ^hSU{u%Zoz|zz33x)GWTU*~|sQ0R6q4uk*IJIxdd0Co`I~y>+LK+AzfV zR|jqHJZy{r9?pGnPy93f7PrQAae15{yq#YMe@Bf*@Bbj2vyIokYkWS|kGBPnpZ}*O zq+{fM+l%dp&KK)xd@~F7L%QC+%46Zlz&sjX-@Xj{wKz|Bs^k9>H-tVL@7AT7pZ`P4 ziv{|V%fB%Wm&Eog7*Jui2GvEu0fNF-88xw?K7|HA53&{aycAgpH zv!1Y-cV5h$ljCl0oOR<=Jil_*l~ew{IM27y7uHL?A#Mt@k~|?EKYvJH!1CyQJ?CeH z`eEV8EGd79XXP`9`$O~7`;H?J+v$Cul~*Osk3NUIM*GW!;{C{nav@Jg&K38^{FHor zZSZRRPWq$rwH(0X@y~B>Tap9hc-xoXTP-=XyI9VP=OfblV!CZ|Xpb!?9=sU?@$}-W@DBNbMB5`r<)wdLQC^TT_66a z=kVV7qU`_zrclGwQk|+1)DHTG`^Bjt=J1wjP#y&T zd9N^cqEC5U{44yum&HO2@U~C~Y#yJDFT{31>+cX>3Gq@6PzR_L^g{Jd>3ce#hfedu zby{C-aAj`#B3KJ$Lz`)v~MiMPay;yK}%+B-uIFr%-hyK^`u!hAiA|H7c_ zt?S2Tdu-~wr~Dmp|Ms{c%we7rj=wUOxKQ19f-id!Dk(FJG8!BATJEurr$lsbH#HSpVlAR zU%ju^xBsgjyPOK>Sosi2JV$Jp;W~ zahW;F!x6j1bTOW;|83Cz{ChtB;qjAc^OmPH_jB6hm?^cwITO5=$FDB9EG*Nqc|~(~ zZU$bD_%G&*_q6=mL#%&asDF51G(C^YrSITU?|RcvD~bDZqngL}wa$EZHI{X|e^}Sn zxpglmoD#e~GdKD>SBBiczvK6*;p7JY^3L_YVVL(kH)v2kgBayFA@Tofad4a# z*M#`*d}3xm7mg(+EYjaQc~;C7kInwr)ZYbj+^K`n^f~oAmyK72c)5D$1HkMf&wS+x zuFESQjEzHG;rJ?Ek=Yu@%kpE)BCQhQJH4*w%**Hd)A}_3cfyRwep&E(<_h}%+W37q*6OgJ@A>-rxMs5)bL;pE@txM!zjuyi^F4h3ab6zB z?$IW6jyVF`O`)c|Og@-&s85f$6os_mT%z3%;(rGdy0sGyOU~vRvZ*M*P2^7^A=Y^ALm70^+?` zFHZ4nMyxmEDc0{CJH?K{#~3w%IstBb%yYd)%%giP&vNw_z7>}1IbK5}i<#o5_{vj~ z2gF|*eDPQy`0+`>DL>ya0b({ye@oE#YJr=>eY~B!!t?$W^2a0L^=b)mp06ao%pYR@ zVj=Gg-OuaMYv%K>5G#j11h2%_a3v)$v66J>Jz!b=Wu)Y{Z@#7mO-CA6kbQ?dkzgR zUzza-?e0FG&u!n&=aqweKgfe}k=*9C_p@wukUB<;Fc+d{`1deV;#g2|%JHJd#d$&N zn@J&NoHO}lA>PsYbpDP(&wnR)IuN6IJmMt1&->B0^*qN=d5-?=;lcl*rN#8)L(f1> zK=-@O8{z};T%7k|o*DB@`7rz){=V7IeZ%p=zYa5}mxkGYGp9Gj4dIvsb3qr!IdO6v z9f!vELvG*&(f#YhsvLel)*xM402;AslzR zQHcM%H1VGYv_|N$tPsn`OTz3G-_5`ObS#!0ulf87;fw#K z3VPl1hDU2Z=HH6%?jOiAuG`mnPIRE0qc)IpuJpr#%-&nR<=PSY2vfsHJxUHuFa~S$@`W*BcO>aI`znfpC zezKqP&s@6C^Ssvjvaic|^EAh|Qfw3a<6FY@XYZ{5~_o{HI!g&(9yC{Vm^d0CR+M zHt>1$v3=Ihg;}N72d`gkb3>dNhX;Rb+c3BDkr3Oh*H^`hVyWQuFC1zs>;36L?+YFe z#l0cc^L@;5>g)3S)nsP{?_d0vvyTn_KfmXz;T#>l*Kxrf4DZ76NPL)$g0Icr7XS2I&Ez{a z>bAHg#ATky*}=P~jSmj~$o_F)I8TEfFJ^{gyZHX%zI-7+cpo){9H4G+OqKZ_@5gr$ zi_QDLEX?GX&7sfbfGtD6mR~O)e0jollXFUVoUgxe@OktF)D6qW65-ghyE;$9W#K#c zzI>hc$1CHRp`UwsoESX(oq~tY!&xm}8{)gTK3`}MKRVl^oh|uaQ$5_jHlyi$YPU|1 zb6w}}j~PzSf!pJ`H}y9%^?mvO&Z#N(J1+OAY3J1B2Yn%UK>QxLotL*@=mGIM)r7Vk z+sW}^+a7*Dzx7k`WIeHN$G_MQ`9fkeT_#5JYWXx{pXJjJ?|$^`%okWDT<1A%8azmP zg}-mP;=X0^e`tUCMn2Jd<_++Y9{um5Q*)wp<2AucI4i_{K0P0r_T*EGYo7=`OnP;_ z5DWP;;-H)$Zj1Z7gqSxj#dJA9FF=fcGUB4gq2he@vIha z3$=|LG5r5^s%5Mn-(ka$FZ46y3^kKnVSQR>+r(IJ*3q8v-LUTE0Xe|$amM6a3Uc(# zPx5;7h5W8`13f`k$no-l-<@`l7t8|C0rG)u#P(v_u`SWQTL&#n8`J7^`gHGm*J*m1 z2kyDHaaxD(?e(s^&tn+Jb*Xp9HtSODG2V|Rmw&vM+#`R;9W=d5xn$&(ebe_ex96!9 zT#j6ISha~~ zGvncSdB+SnXT#1hGt8e~8g~XAt6%x@;I;EL=yvC&Wn-Gb9sK4!x1mdWa`Nt zI>B{wF?zK89De!e;mQq;g`%Ztes#jm;nHn+!RyU_n*X^q9t_7{InSFpOtU>=_X@$M zUo*_*KwN)otR5a)IXv%;LD!q*TqZ2bKVH&R;r)EBZG)H3$G1NIJ^R1^&v95sc3x_8 z`C`9b$ok=!2;L3v{uQxI$N_ZyJTXV;>C*K)ocqJ&Bf->=C!P{|zjKEB{8sXY{_l-p z=7;vD_r-l4&o^WH*gV91$89YWW;$toJ+Lv$Z(A9$-u5p>)8XPcf1Oqrd+By{IPGs= z9NJ&+R!oNd*8b|46@HDHK^_ptEz|x=zsU(BUx@GaXI_ted-yz~#t;K6n>M8N>BQQD zQ!_*Vn4FvZ(q>G=mSf}K5PN7<8lH|7uf#2}UkuwK#4mA6Uhv$#VrW^qR;}PZxbD88 zcgJ;)$pi9)nt>-SZqnPl5^;Btm^T===6mi8@%^f}Jj8WA(1pRj5#z=DtK)`{6TIf{ zaZm7%hVJJnnIT;=%<`x$#B;IyL&1}h-^B01$a|x&JDwv4T1Wr4dO&S7P%Bt})}x$& z&0?%m>sy~h9_&BYZYg@J5dM~*~ zo|)cf81Y`LS0{+KG?aZyjiKHU%f(vX+*ApXk<=7zkEk2`s~z2K33DTbdf zuk!TQjrC%K34RCn@q*L}?+*Fm&A}6TLGXi~AI}MXKR-w=P(R!qasbcYxtI7ndcV%i z>R1hVUe8EhM=#5^VH>mk9Ubad^)W4?Zl+<>()J&|kNr!{{m!u8h|%_qvG3St>_2eb z{v=;`j(v>3BewJR;X2$ua*KVNH^eKl-|GvD@e&|y|L-s}&MPo0Vs7NDI4<-Tc``Jp zxW;3lX=z-sm-ZDK^;li%e>%pA7Uqw5-ZKX98b?lJIKXnC<+fAuM$ z|4NIC?U%;+aaIg(-z?9u@$=x_yM+0kai9C$=IcAI_tdbA^Fp4W`FTSR#2o*fW2&b> z=ldK>OpZnNIemBE<>fJEJFOQwpT^hYkPpOdIZH0I4#jRBnA`He$d_XMrZM8ZOZUkI z{4#k~eKb6Ny8rO-8=V?@ycfo0LHmpS)8|3&&%1jt9*Nod-OUF)JM?$tfBl{j|5wiw zqD|xgH2}nQh|@0JXPc!hY`3;Kue05azLM<^Zre6J*K7X&mg_O^&6DyP%XZy;UhBQ( zC3Ue}q&~1u=}YkQ>3;r_93d~z`FtaqUhJmH_&WC0o5Ku|zOeqXelPwLVmsWgpKNCD zb-~;HcJPCF#%spz@!R-E%-1>Ioln3_4INL*ZW3lN`RRJA^e!)bk!B^ubGq3a5aPWe!e~*=+sb8+#Jp$ z#D8(#H1Fy2ne9JwGV?EAIG2vvV;oOx-qX1!9EU4^$Y0mQxuKWLL(-pJFTB5bOzXxB zl6vQ=_Dw`fvsTC5^2YTtSC4mu*vKOnA3qwdgD3KVkYA?H@5m)1SI9Ao1|Oa#f6jyj@_WQ+ zm(LEhz}zuc%m{H@PximF{yXdOpzYy0PiNphHH7&dvqC(7@3BzGQ$CA)HfB0~AM-}O zi+KH+*fw?y^@?@EuM)F)`D!0}T}`t`i03rC`o?4W90!J+J2TW-ydAjxqtFAejPZPr znZ5OU=*i0gemi}ddxGyT-amDs`xok*fAl_mA-0O=yhSmU4sly-r^#(2V!HS}`bf4N zI(_=KB(8hh_B85C+m#wsUV*xneo=eMGjxpG7&U>~gZ`0Ae72E;d@i+xZP~tHd*=!5 z5cV6#WgQguDSChCe8&OPW1~O7v*-1T{pK%deA?g44$o$eW{MV%HNvr*dVBW-zy2$6 zb~Ii$J8xox&?n{3n3H#IF>!RR8FNkPbFrK^zFQm+`klvzS(hKj*J6v%N9Btw84HAS zR>Co^^T)HpY$Z)E&hq}nV!GLR8u;})Oy>IOXnsFWh)3jnZ8m;h%S+-dA=kVwc<^Q| zq0ZptZxow{&!q-akQ06w2QCWkn$n$INe%XRwGYlOPTT&kHIn*P?f zC{7AJ?jOdt!+Z{0*B5cW`ib5@Y{JYuvmeBs;oJ;cg?zDL(D?Gkd&4~E8$v&Tu9v6z zIr_Q3i(`V0-zD^GHwkgxG246`G2U_Aj^Xf+CNTd+V-9M^M8+z@8(dAolJkJ-P?gX-1l z5f3ki&Nt&B2AJL9<7n?F5hZP4rP$OZD6+$H|g>0-KAZ~c8g#B^HypwO$8Gx;>)wRQc=;He!QJRJE{4iM+% z4f#XgL(JC;mlMXZ5nf|?C&q6>4WKvUceSnD9sWJ5WBK@owlQ@hT}?|5&v1J0XlRV- zzWcU0F?5MbUcJ~$f2d*2zj(c6iPvJcZCfmt3*;G?FLU3h349LQzt3dulm>xqe|Vdg zMX&fwVm%L4JuSY|@%pv51np%Gloq4Y7muZ5si4#J;T8ywxxHLC7H6w4x2MOqbmw%` zlXXljeQLIZAF*MGXL_Y#J^e2J-W0qSTHTyFf9BJn|LJ^d=Gs3T{yxe=G0tI3SJUuj zD)~P@i6evOE_R#O^nU9H{cg_l`Qh9(=KOg{=KA?0Zwj-IpAEWRKCmoxf|_CYJUpKP zeLiXjdY;yI&VHEf7FU*Nl&Gm`Th8}N!)AoMJFGsua+_W z_4JZ`hqkic%Oj8f_wlKCz!T8JbzBgihsGB-X!_&BJg3;B$NK4@ZOtHx|0_)xdY-<2 zd(goy&0~5@-;*c8BjM%nNq9iRAEBA$lOsuTSND#cjB5Zq5DTxaZLKa)IsFb}d)fel6GXeLnB4pUbZ~E$Aq*oi~#NdO}}K z4Cl9s;Y)^nRW0rqVaHY4hwaZa+$y2}x?$`X&Zp^E9mug(>1Vf^~%go2)*!^3eNhM8f;YMaUb#{?2&NH4ZZH1Yie`*uKPSj!}DLnYWuRdt!}nFHTdX}*>}WuxNe_R zf74ZJiKT9usRxV%*G|HW&uTuhfAydOGWo>sTnhMwLmkNF+CTD-UYtvA7S+lOst=<$_`*>r?U+lp<9$KpA*A=rM! zcYe?G=MPWDcIJ6BjQk>g(=arMeE_tG-mN-8u2K8)XZWAIPc`xJ@tdHP%wW<~j>$47 zH|K0~PU&>}=Fuob)vi0rw$L0jh4f3DqYJNE# zZnj*Hb(P?|e=PVNy994We0PrGyC;7;<>46*&-i;>6ym>GN3Rt-ori$;aa@R*W-iSy zemfjzC9jy%UpvI~W#SpZM{=HFmwfRT2ED&&n9)}k(BSm>?ZIz~L8`9zn5{<8TXpUZN;5On-I!kp>i@zij<)b;WEaGr#NgCDa~Y#p10 zcucR;^p2f+Uwkk=8ax8O=MIx*Xm- z{mzsBdGO=6kM%-4m-8J*#h0PIX>+}Ln!)U@SzcZ|eeQTYeH`1CzK+=(UXJ^{R$S)+ z()2tWueA*II2ImVid1~2A=M1@Hb8feul^R7-HIH zLHDa0d>%D|_gyr69zEWNXFWWXSN_~_RkS;QMy&o}d?)sfed1fOZ^#9Q#WC^gI5YS@ zSH_JYNAZN@Cpm8M5QlxI5r6qNpz(b+b&B~-yqn0CT9@*hzPJ3K#xWP<()t~^ zZGH00)Gp$^9O$>O{M`g4Lm1dn@+_V6CIGx>w(KlHj-FTQ^<91p;kQCEyh zwS&51&rky&6{m)MQ;g@+IQG>(JmUMiXM6XQUejCMC^in4wEG6(`X__7+dGa9$6=b) z5R1er$4I?vj(1II`2RE!%b-iWwf1=koOB4%a>IxoU%J zT>!w>p$(DU-esxfBu`8Ny1(}MWCFT{Tyy?X6y!JA(#=mF=4@Yko;_>k5cD2EEVOjj) zLqflAm!J5U=rC+Js)ATWpK!L^(yx zNQ1~Jd>XZZnu6!U-`^rWALeX!o*b9UFZ$X9J=nd1&Oa#B$7244L5J~hW}j{Lsk&O+ z=eN@Cw7mGfMd-tc?_Z9ugkvf93q9CtV*dVi-p4{r5rftZ{YJ5Ox&D@k*vxkqKWD}< z@!QGYx)t;9iN~VyV$R2BJU(@s#+MuR3&$ww*)A4pgzMsr;JN5&nyWFJ{^@XB7GLIl zK}Yj!-X44;UdfiBH|#ho{tln#)-Wr>qtVkf=ecyeDD+Z&2l1OPq7Icz3K8Z;983-n3(;%plSB&kXVVZ(-&}T-JL%VS?*?oD1Ue_;WaK z!qstEoEN-(=f^)N^k;XAZDWftclnOs%`Y8hC+Ydy;<7j^P7FEwh&Uv8@_ySrW2YEB zSe}i#Zg@84t@wK4`tbMo_N&Kx!g+S&2j|%TOz`+^PwGX_S0jk~=LQ|mW3(-+U&VRe zr1-now7JL`Pt17Yq0V>gn918h9H!rCcl)M&$z%2*dWlY=+vy_OM~`Y|m_^`ay7b<1 z&EAvqz}lD8%Ik)C3-$JLLC4Dz-qV~7zn5PuZkU@9Yw1?noTkKdeJ&r+v_oUk{@Vv1 z!Mr{#et7&m=yUzYQ$qaVDZBK(^t+zq5#clJ9r^+MYM+JA!PB9UX=5=}+}F1qKK%8~ zXo{!2_RA*d^@^!-l$a_$^Y6uCvD9+p5V6$tdE*(OHV{{7abAk=Ev{N0^g7fgqetO! zx?2n$^Qm6HQ1F=KI$GXia-8oW$Bn**!f=7>W-F(AO;#XnT#!Q6z z+HsG(dRlD7uwR>*rt#@}dVZ@A>-Pz<;q2h)@^{}EW&*`5p2-)3H_z`d2fsr&*Ub5G zOWYsrx-*_K=TrI%ga2dB{yjS5(HZ}YKgSv2HG9VwV}r>|y8i7lAujXM`6uGPxa}BN zez=($eOsS_e!n?re)^yPLd)~p&GN`2H2EfBK7Pm8Bk1<;PR@bmGq}%V&V69kX1y?* z!&{OE77gFmjE&hM=h@Vs)xR~r$#bIT4~XvuziH>NuFQJsCz$(OD^`eC#j>$b@SN!O z;X|pP?u)xZuDCt?T^ae{!jJ>Z>zKXyRJ=W28sa_OetldN;yn-U)CsQtCggx$#Nn}j zd@JO5zRu@j(^x0o7kcw+#42Ii5%cwR)Qirk@UqbR(a%>8@N%vQa{+2jy(`->uV%pz ztLY!J(e^bljvvlbrN8WpW-83m**DE!fbLe0>y6qs=`#D4x>`MLmY#oO-*V}_+~+ZM zxB9`kw&V}>!7su*2>qw8M?;G>;<>msybv*vzN9z#@w6vsc5#;e6dU7c=dXnybbk$<-l6nCb!5V3xqr%XNmo4963S`P)7~#OHLi@QLUri+A8E?xzsFR~?jI{dj zZx*S$q-GSy^;Y@$eEWA!=Hq#Wv^$S}%m|3*+s1COPy8Tgb=u_c35Vs+9}@J-fkCr~ z^K|`pW8e5ucpeX*e=pwiRp;+Gs8_^WgN_?9e#_2h{Q3A&>>4y5KiA*H`&oz$dU*c~ ze$Jc2v7J8+UIH!q`4H>Orqj99Pxd={$?RR zzcR#lG5wS9SPYhObi1`bJyz$a_#RNJY`oE5m)oayz<*^`E?z5Ts{7i`P>JC1Rd8Jjt<;o%EFCQn$JA6JCqBjf-v z|IjdRxmWBG+XlTZ<~ui;1f_?|p~;T5=T&XZ>)K6yROO3Ug$@+ibTpG}Mu|HWD{_wLY#6^m(ldj8@Nui?7f zV75m5=Am30;fo{g}C=&3$J2oG<0P znIq?^QOxO>yRp8;?BvXHyY(g49~M6j@mnm1>*Bb_#Bp_!bxSMI0K7gr{i3j*txsD0 zrm%fEcEa{&JFpGeR%|Cc`6c3|Aztg_@YL4}ahZS4OZjH-QfO|qBwuo1TNl6i?qiTMk&)cVYI5hoku}6sa-;E!KUfxM@X7G5=3+Gh%W0=ix?v)2(&e`Xj zn(2E*$Me4@^bO?$@ro!}taIIqYdjZo=H_F@?EA5u=|G$|%W`ixZ-E)j#e+uo@1M=z zn4!_P<*AsBH1~2*$O$(Ez0E&=FwEWG9(u9z1^sQ_-<**CujM)Z>9$ZeShiU1_^R=I zuW>#FGnV=U`ni_teH?@9v*-={G!6-xUf$t7IaZjLhbb`j*5L^yZB_R8~pjz<27NvQa)x#-y1i_B|*<06LkDx z;d}_kg&CZa!|#hz;eD)A z`^2zq+s5e_`-=Tx(=hj7zoLcY1of@H7=LG>;IZ?2)V_LKD^7TGGZg08;C}V6*ALwG z9Q!VxhIjvr;1kIyVgc_&Eg>fB<$WX6{p$b!O#uGyf1bcL!Aqej#VMD3VfT5=Gj|Jd zOFSB0!RejDQ5_-v$XnvOyv2{wGoU5i7LVk(^W&5d%X}uei2hd>h?%Q~I7G{1?YQ z(B|qpnwbnTolc*dpRwKY z0zVm_4%@OBOmUnRKPG+|zYg2J>nFr34JSe0R?|EzXM5!)^D|`Sd<*@AAW(X6((!@pkN=aQ(@+fAO#%^KZ@I(RgMu&GGPo z%5v}EL_s$boeJ@-Ov-@$@Bl6 zQ_akg{;&BP9ujTOPf}}`(HFzTa`(#>W;!2_c{|?Av0&yn`9C}+^?^KbL3qB;?6qnZ zIl=6z9Ki$98}M1okuDq02z_1k)Xg!x{Xg`#!_Erpj<^5SFz55ucuDBF%n@|I9PszJ zJ#Gs5`l^r<R>mc8|^t^We5!=`U!PB`h#C+T6 zk3+AP2Kq#NJnTbea`^dTzx`?Buup9m>x9S6&**3I+CLngXP+}0O@G<%`1kyHGnM*Z za)F-MQXzM!y~TXqj~YR|6@SGJF_x~}H|R|su-OhV-7!I8KF^1TE+@F8y?F^@5?@|C zIyz1W%hPMNY%xVFQBz>V8$N>jU3!igPq9o)<6Di-qGsU#i1~bMx>?+(tHoh)R^M`- zAYS=RV)$R<*2&E0%^jb0XPC$2NqMf<=oj;zEJq%p&-o_&c)DEAf^WWF@SGeQ%!5*M z490EqMDF`ksCCp#UdN|eKg`IB@0LT~yWBkFGq0PTN5^{iJ6O+Rzj|lHd|I9_L%&<6 za=0F=IWD#0@X<&8$X}=H>3Kd3|K){23-B0+p69oihqO)EPW4J{vwRb=T)h2l(BzI4 z5npc&@l~J7HZRWpJIpMof6YVMH_SemO%`Vtn;^a}5oVcR6ymHJTC8=vy8TNpO|Mnl zHADH4P;-mF>j(Y*iQvWW5Hve&MzcH5fq9#+hZv1-1n>5Hp@-+veyy$-%lSRLAGLw^ zQzv-u_lCG{hR^%y`O$tmg<63ZY}wvhE;u;E5_#Z)5O3~|+55LZEE~?>qVKp;tQBT~ zHjK~4_OVm!7_NUlwhTQ>deqLordO%|1lib zYS#4Optbod^9KFBOuQ!aX&tY%dB`byg_tcD&75{#PMV&dFCWP>JbrOryqB}&4!v(N zfBz8k&GYbsEPK<~EL?6Fa)Cbeo8v|C+?X@?K!1(l^E=m$xqir{=fnwdK=$q)7|Gd;8oox>lZq4o|n@=h`81o>jbcs<~khrSq3)xK>X zpC@>+dIvlmx{$6rGsIYZR9c^h;Bw~w1@}*Xj{d?=LyT81i19od@D|*L_wYF1;i#|- z@r%#w{k)e;o`v`(-id`HK8l~>qu9w85tI2RG&Ak|sxYTWlk2V0=|ij2-Zc3#L4S+y zJRu&H*NXEr__!_Ri}!L1pK9>Y&g~#(>pN^0a^CJC=j|CjyS!(eSeGv4E!XA2lj8KS zF2rko`Y*$!{*Cn`PaYq98a0yj%6F3s>2O>eZp)?O{g}a(yQh2bVtl3Ul*h;G^ZWB| zczg2yvY~EVG4%WC2%2G=*dzGlVmz&Fd$e7eA*QEovu2T2#bIo2m7s`)9&8|1=e29$|PVL`jwblD_fgT>u$93}^<8}Ogv0mTTXOL^? zKmNbxS+@8uAFUGdmFx13m@nS{E1au^w&WY|k;A|eFt0@ocy##30|xZ63DoZRT8`@&~bF@Tq+VpVR$5lXc_y*3XD(zsTc0CvJ&Z;oOtY zjc0`Se{8nLrq-L!D^`l12F{Fz88fMx6u4 z55GTnT<@OLG~*Z_{uF;foZ@roMMFO0e_`P;@6C_UpBC?kOnFXSL*MePXy%ID(6nOEnp@j%uH#}n~F)MhT-cK@oe$^?1Z z{bKHT++#dB+M6e5hQ+d98)7tFE~mq7&k?uBa`}40=d*lup<2;8S3lD5>O(r-EP>zb z+o1;a+mHB8f7nj=>om&Uaep}8gnv%Y-xtF-(bob!Kej#FpB>kI9uwc`e3#62Itb`kc0=v*|5botBpa=rD0r9f&3sYh`l}^edx36 z8-6EzPIZBNMN1A}!Q)<|rtltej!XRr@8!C>!+YsZ^O@ff)|9B`5i|6<38UJvC?<*+VT61 z`@L37l`rK~`2D*-OWWI8Vg0#n z&eFD~ru2I2Zal|hKm9{ICL`e;8;X{>ZxIQXHUaz8g9gVwT0KIEzCfO z`RWTj2>mxRa$^2&p;yOS5=+DsI?*{(^!Yq4?z=6{h&y~I&k_5*)||DtBWJ*UVvzSc zFzC!t3)*?bmy z*E-Q-w4Qj#JZb9`uDkztAx2&nJP%q}A78$OzNGu)T{WGY>vtZE-}{V?yS6;P6t3h|%+e?N=;*MTo<% z2^z(Gn(h3xp+_oat7C2V{22R%IA{NHege<)I+}NAb^Db57cYm$L;qWjxbOXzkCnr| zMN^AkUTgV2hn(R4FN8QPUWiThW&5c(C=T#y^C!^F2xivO5C*!vBxs(-}IPT z!~2L8@_;zvyZM~*7fmL1iiegZR{GB4dx{l4j~Zp@x6xA;KYS;MOLD}>DQ3I9M(hwH z{5E2S^(2PqW6QsCu9-k{B-XFb>v!-Rxl`=2j)vD@-A@0Fd>7x7zVthI%yYc2_~d;p zU(E9R$kRh((xJ8k@l($5J$<)XPt2M+$DHlmcI7rrAkNv=d~UaWAIq?v(FeA>@g02@ z?>GH>+9rKZ+l+PT^U<)jY5SFWa%`hMC#?bNY{YM`zcmkX?7Q{}a{_dWWn2>H$K~NT zZO6vi@5EiY`1bHQtrP3szT|$-_q}{3kMVIVTYMIi?Z4uIbt2aDLhY~Oha ze#J5^o8}ku#2hij<37X4Lt=_`N6V>K{9b-n&-MAtcZk#Sl=V1#E&1%=7+N%5{5Y0S?C{TLvz{!={zJ>EXT>?ON?i87Vzih?V~Wq>K3&Bpq0PL;ZOga6 zKRv`KpL1v{aY!5$|7i3>CVmOsDDH~oVxrhA2D?pXirvQt|APmsUgy861H>hHz@=Ev ze;@jij-;<0cS&3GXuU=*v3%dvb3Koy@#3f#^-&^cdyZcSVIysdl_nGw={l0!r`=xzd z{IQK$*FL9yiJWE5$#y0#h(oq3`?lAKNA?@>LL9K2TjtOLwhg|Veb8sKozfYes|FMU z=wRE2^&pqnuI=OWAnj<`_J>j9js00n7IVZN`e=`ng( z4Nd>jkz$k2<$Wx}YrG%L%0IOpXmjgA%=Z0gc3PQ6<`3{c=w|&i+Fl(_Ytp(yd%JYo z{gx#Mh_&LJ@9niT?JmKyu^i8X$LV_dn@;Ajm?h%5@MqKy;=TA|IeZwt3XjG0{logV z4aqC=70-b;!KdK^SubOz+WXV$VzQj*`?~J6KDTYswl3zG+wfZP-S*|5*$Qz?EhPub zJ=XDvleP=j#Yc~e!D66g+NQ>DO;6b;Hw>D^GCf~By*d20_BHWOoV1^cb@oZ|l{Ouo zhW(J{am=>3=XtKvRQ6AJFM3i;5j*YYp67emx9rbiv{-7N5m)pv#T6c&&u=^PeZ_O1 zN1XLs@97v*`?*{uH`zzUTk%Yc^cmzT%d$U;6|}BcYG1I9e1D(EZQqB+lT+=}mPf04 zops>5(NUHqe!HJ;R;P=dJWl;q>z5DWw%lajqeU%)-jz3eNBh6HWVxPiJ^5Yik7A(p zColXo#9uMl?;xL8hV`j;%Fm|r`P^c?^=N;#uD!o_%o8{({vA*6Se{2_dt_>!!~Oge zaoy|X5HVkDrNezT{(;BLdz%B{KluJ)BfVuk`R#o_9+B@a56FM=i+tmI@O^E|zMJ1h zZgrpBVf~Baa;a^`=XaZyfc)(?oo>DPjcl*BQ;&I#@9VRORWyxQDE`>Syr1nw9qfbYD>^oY9eP2%WoKZ{J zAH-xa*7;_{Tk+5~KlGBkZr>9t#Y4~c9mOb`QOq-QBW8+s2Z!&b7O);H%Wdy3&YH#H zZM)<@_?~_PaoqRjgZN&wv-m0>T2FolpI@%=+xTAIOD++sE!$`JTZsYUf!}z{TKJ7@ z6Sg_mtt0D@H@$M0G2-p3Q!K;#Ssp*d@~tD6e0=ZgdA^srL|*dR;TiKKtRrz-PM16M zrR6)nkKf93_%U+D5@Du9J>c=7ziltpp}1swvK{&7_11yUV4JmV+b+dYaarxkxA5=g z;`#7J#83MSJuZIHwYFowmHJ5z_FUV!fAWW#)BZJbx&24Hp-aU$`;_NgZKtW%#wzVn%_Q+dn!@t*Xde?EttrZ-8S zi}hlub>Q>p<-Mcs;%4!}P&0TR`?owIF6+7a+ZdL?m*yu~ru8b9`h0SL&#GRb1Jr!J zhxqTew|>kALVoaFtvhvqW$~bX71oVq*arPBVzS>>UX;^q8}ci!Lrzo|c&*>odiPwq zUJSP0y|;Dm{anf^?vpF5E8kU)k;~j?dsBzey>z^Pwlx~W>*Z>%lXI+p`=H0gkY@(X z;5++$>~nrk&l}zo?QGi<>(%aZi5MuZxNU#2KhoCXtJp5Ch%hl%9G*pi^X<=da@=692A;P#(m5E{uB+>-pA3$5J@pwnSYI>7@ucx%EYd^f z1lw{Awd>HuzGrUnw|Pc?&`0z|b0t18o{UL))b{joV?rBK_#4Et_DfTVd-OHO zf)BKdF|H3fKV#0hYDZ%h=lVXLpfv$U;G5y~<6QBBoG2}f$N4Tp|C<-g8+;P{Xx{UE z^CnKP9dnhr5_kf*2bVafF~^wiY|ASj1TW$s9A_JzpQdeH8|#=eajNrl&0G)Lbo}Nh z98Cjj1N`L`@6n=Q`*i!*j&{Jew&{D~DscYh5Bt&f`kD4}T-J3gw(l6}0-zhMi@WXD zKD4dx;A7i^!L#OZoKNTDXSrDM4-ZT|xwuB$fb+GzHZ~8Lo6Jk{7TVUlxKprPjA=fz zzSzsT*blGB{GbgSlYO}+);Dk2j(Ej6(pts=zSob;k@^ zcZeUySReWdPwW5q!>hiD%WF!?WVG;4K`E6U95mzE^x@ zKEZ9q8-6f%@c?iM{-Ixu!{Nyqhxoz1=|bb49-(uLci+P!#;7*YXY?<9!FaW<_pZ0= ztRLt@L-)~aw3T)>2DGRCq#uYIT^rk=-L;7^i-%|x^D16+T#nhRVZ`?Kq2xpURN z+QBx=50?jRsJ*nIcDL=ZzIayO*RRdJAkL*dO-Z_NSvmHspOwDZzuTxZ`mHqA}?x@)3shW@k6wX&_jwbsFXxR?faEzBXf zMjyblxQ?cEO^gv-A{Mtj+Y!s)EZpgO==b2d)2!xR@$K-!=;h(J=mWUjwFKObj~%nQ z!Ygj`nK9saY+Ju{p5{t@A17!xeVAt}o;9bL`*68)(LeP=F{E~MeXXb6@Q-#ijvW&} z${dN?^(!6hm z^ow(HZ2GJ7rys4{k9Q@s}}z*IXB4(OC8!{gr0d zfAxQ{t7F57t}&j%Nr1ncr}-W1U;o4HK676Bz?iG(4g18^w3TtDKbVWfSmrj@+qG~F z9S0uMzRumTI5*eF*bz_gZLI6qaQT+O+>fv6Pd*2&2#$+BbA9w({Xw4}no$3Ap5}$Y zcfNyH>3r9H#7nmCSk0~GRsG4i=!346Yww?Ns;$R+*A55a2=fPib3CqrYinKer_U^R&i?6NmOE~3Wm~R`?K)@6 z@t<|*Q~P$V`kgVXjq##!WIXD_Vs`q)7s{l>n0 z*IcWO^pW9ByUxRh#sSXTyfpfh@7srdiLZ@S@rxMEJb=gbUHxhBratTYuDO1$?Oboi z;QX8?u61plx9j8@()Zd?``b4D_Ud@dAA_IuVdGOjb`6Y2pWCnP8H=v9>ugT7Ki9$W z*oOIE|8cz5w+-j#{Kwo%=Xme@@iDH`r)|&qYA4@!?OaD(DpLV^pqif^*j4$nMeaE2vv?X3QHuOF1?O1((#PP;D-qY6hX&d&h@8ej@ z9j|`lpU=lQat@9W$GFzGf5eBD=`TYQX+Q00yZV^6bnc@)#UsXnHt-$Wc3gPVHto~c z!adqvAGL4opienhTx?$So_?k+XbHMPJK-8*M7~rXa4e2fo0)I5jq7H)_S7ew7e3Ip zX&r5DIj+Kmw&$AJo;J1*<3nF_oP+OZGv_)qfj%_Nk3>Hm(&^*5gm zXiwWRrt#9~+cXAl*MIQ1KC7?kJ2=An+SGN_p4xfDAv&S*01p!-|m1(Tx(z2 z*8aeGYg^m!IOD#=lGrHzkAO0Gw13vT2nm7m!&1>e}0Mmo2Tu^+-%OV zf7{2`mbw1=!|3DAS^K*#uCH<9y1L%ZU4L{O+FARLKJMCTYwc_d_*`GoHojxt8GXw# z=Z_Ed0pD}I0T1YB+FD;Re(l3QpXpEfpMGu((gViq+~E4*4dd5x>kOaCoaJxV7gyQ0 zV{omsr(<&5BmU7wqrHtg?K1kD^|hmGq5nBH?PuTir_YK9=|;I794gMBB^{T(oSjSc5$|NeF?xLzCK1+gBUGCoH?cWsSD zoQ&J_2m8`4{>H72Pn%m$e4%Y|jhGexd9Phv6R=P7m-9Dfy*e&)fc@Fl*rsbd&fnOv zPp?Clxla0med9TEiSxA$eNmgyLyjLWI4{S5ck~_Wjr|$Z`k8&;X5ZC+^;d0boLk3t z?FaAcD~?Y)*oWgC?LFGjGS|a+alP~(*y@z{s9aDKjrx2&tb>1+NC zzO;XR#rO0v*W7!32G=r=+1eLGft&@!J}SD$eF zKF3k|{rKE?(Ei#Kk2p5{!}WK~wV~s8oVH`E*tcUbX1op!ZGS!kZDoCYr9JdV%N@UM z`JT44Pv-}&9q2F4+kQuzY45=exB{2y3yud*fpfBLZHVu+ul?BX^szY>V^N#oRew7d z`?4QnMBj4^`i$+sxCYwPIoXE(GPvBf=zSW7k7Le|7c^JladW|#Cvby)hJPHlc5*!C z40$>6(!7Af##q+Q*7aT6@tOCotG>r?<;~G`beTTsIJLR!t8H9w+r#C)>wEgS?P){Y zF|M0+Y{T}gJN7l=cm2rr^zFI9^>JOaziVT;W%`kL5vOTi>sTL`;TOPp`n)mhJcpJP ztKeV|tJ{WgH?FU1VqaceBmHMwNBzY5w2ZOtZ`aoTaSa`W>+tUA?~cQ|xK7`9{MPe5 z$1VmL+=9>aQT=%IX~(Tk<3PONy6U6mD#z!z^(p+|Jhk5#6Z)wBq<#4=bb;~gn)_#8 zj=}z&(};QSv*UNH#*S_3llrG?txb)uF&}%?PX6i3j?uN&7rpx1HXN_E)`qrkf7(j> z0F5u67Vo%*<`Hw1c}rV6&e7(^k~YEB`k3=HUhvY`FV1jGq_%Pa)CNoZqHJ*x`kMTy|`P}$%4R{vb`~H{%2H&}^ z&cnF1k1ZH{GH?AV&e)2jQM7aU;HsNAMVn3yc)}{>lmZ9J$($?(KR=I z^*!Hl9S67TKjt6fOdAH z@LkuztNmDq*3g&uq|<%Mcr(uQ6>X-i?8mFW9kb(i?Tz1Y-A8{g9_@ehy`i7Bj?J;!hU@Rzx!$-` z|973V&6o>)XKc@_cK1DP=vvq>p3v^LWxU#seyFco?)roG*0+7v&GES=-n+hD?az5x zKH6L#vt6&Y=^EIF{>q~;7W6gScHEB9wj8H+^UwCQnR9o&M!(a~TqoCLXl%S8R(I_k zkK-N3Y#WyAH@4v%9H)Kj7x=_^diA$!=v{t^{0q;Rx6jx2s=kRA zEEi>6{N?)FmvQBsVYK%chsKz5*ADo}b(@B3@7VE`@#WQd={vq_43BH)Z*8a#x#r_q zTL)a%ajxUsEpv@*!#di*cO8?qfbq}ua&265$LhU*_Un7v5^U49hBk2={y8sw+VMMI z$Lcfv*fxwI%fS9^-&k|b+Q2zGR(%N1(mS|PyK2vI4r5$cf6S+ji{{V=X%@igL&M-q zfBSsscbqye&{Jvz)MI&1kIBiYw;H)QUf;-3*yflk9HVU-w>aG~7{AWJa+=!lk7L(Y zeC~VZ6UX3q^c&YqI~mK)aa_A`{fuMVH`Z*|KFt@}W?cWV?zq;rrQh4GzG}N*xv}Cp z>4UC;*h;^i1`TBV(PJaNoSQy#-SmHB(yKA$y5o9%$Fc$IxK^$?xMucg4B}z@J78ZU zhk*ZS3%bH`W7RPjuVXo`7&>lfN#A!4<}&A}U*jOhqi@^4zNo!HJ2)TwZy)rNc>+H< zC&#K^(-UKU)ZfRr(60Klu`__}C<`Ul>wUxeSyV{p;Z8=y+ zJ&#&A^|Ii-;}M%^YjcR}0NUR^!E(pqSRJ2j`Hpio4>^DR14f%_8-2lba@|}H(8fN~ zZ%03N%?788>*#ZwV%?z=^i!YX4E@-68@}z(+{Qc3ru6`)i?N3u$HUX{w>iN0GjA9( zVu@)DXB&J_v4iod9?000UoftD1j8q_AKSMt+qG}8GJPk<@187qF?A%qZ+o=p`hnjt z>S5F#$T6uSlj~C}<{3h2B-JFi_dpG%x=Hn9{swiE@`CCuw+ud0gQn(3ZKeAj)O4z) z^?VxlD0rTa`^h|O$lsnR=GklG*+iamqn1Zar}|F!52#sF59+ya>F@*sr=c=dXUtcAXc<$2mXnUb~Lk8||xh%Q2|m z(2n5w5TKpL_Z*+^T3=0uzkLr}uaRTrH|PWUmg_v`m47RkHD>yhYp*}*KVU8Z^Qn2> zyg24TbB#F%EH|g%S!2&w#8WhxF%OpedyG-b@Ubz7+vq;y)40bCG$&6`E=kS2+Ee!# zfP9ZyRrf>rTYZ`ODmDM^`*Q!A8WZ;j_ziaOUX8t4O!pRBM?JlJ&(tn?){W;yc#fL; zxW;pD+>7j9V)wPVw^v=J=Yn|_m}iN3hMH%Ctrt9F%)Ptr!-DbbFrRt9Q5esZcJHtI z<=i{wxzV0CW?RFJp0Y_*xb*yYVb@rzk}l0n&a7X z;~boy7PPFO3f8(tJ% z7G4p&TK3fN&~WFhJ7--J{5FfYvSNYDDqz;9Pl6xkB?N9dB+Hhu~3vo3Cly5j){ff8$>JHm{mr&Aa9V_Zxui zj@lL4-1g1Yv^z~qQ`0N79^FOz(K>3X-Rq>^t9KsH!B+$Ay=N|{IX5=cKMyX&k+{*= z@!N>_3s-sWvF8`yAmi+xJ^kF6J|+Ahd_C+Ib_|}U{*mCB>z*&Xb$EC1{9&K@eUVQG z&%5{B`SE*sU%Brq|M{JzUHe^*j|I;tertHs1aIg$@tcKBf@jCSCTtNt5N!KXVeI?u zJ^$S?*v@Ce*TT2MeiQ89^N~GI_<-=uut)HH&zg6UY`Mngs5$QK_{eJty!uZ`>zp>?aTzVYte||sm=6-|gui?+(f^cED zG~5{c^X%v4f@eK@w)6Vo)nUuv+4_!shp8`&Cta;C(=T+B zd!_t7klzFH`$67&eyp*4X~3Zy1)S)a<9@$m`F^{4{qW|=?`eO$-{Zhvp9(vLuLr-~ z;kT5Q?RTAC8a@`j9sKV1Md9W!YtdPYE*TyX9v&VVmI_{%36Bns3Cjk{{JVF~y>tE* zZVXok+d4O#9QPr!&HV!ExhnJc?@{6Fe*4 zZo)5MVc zy%Bqv2Zt9ZMx*iQJ@JkBM_l8({?Ulyb@QQX%1cl)WJHI17f5H5j-yFQ8->W?< z92349z8d~FY#BBR&kn1F=Y}dBO4hBRGE7&~-KMnWLP)ImaCQ`{26J*Yvt;L%(bP zJHp+8MiwtD8$9pWy-@Bc<`szp-4p5eRdMXs!*2tQU3B)Mtg+dCx8g71Hvtb_9_GxR zGot}NIQxS$zd!NYe)HOI+}#we4!;h5ukPTmf1o`t4v+4)_TC|)tW?_r) zrtp$r+2g`PLWi(u-E4-Z7h80(MHdhF_sVcw_)OS1;P@N+UGCq9Urumgzq4oAgXcVW z=J#MRn zA}F^q%Kn-gSKZ41S8|O?XcI zm4VjAr)z~x0zTb3ygP_jJ{mq9b`5?5<$&4eNp51>% z@XUS3h6^S1DoFgO}_<7u4ZcQEkseGBfJ`7P`} zPJS2f{C?~7;y_#dEBK95Jnr{0UL3Xx^sr;04{6Asg)_n*1FpR_i0N^m@jPPl5s%~M zGs6j!-@x#CZV+?(4*umQ@E1I9fWN`};E#Z3+R$b0>+&pl&m|as1+Mm+r1+cXARedX z=>M~WW5vnl5ba7IYfst_N10zf9?W-iJUy$umJfa->fQzS&irQ7b^VqUzB@M@9ljB^ z58H%ShPA@e!sEkZg4b2Tx&ag8VC&-GcZc_fj|F_bOTfi@gs%jgd}#Pdz{&XblyGV| zEu0a~p8VeUIWxaH{_o!%|9!t>cxAXf(A9n)<%QwRfrb}D(E8eh2cnI5FWSsprVV*J z+LQMort%viuJ2`mufmJ*ok!jOs2Lr>8^Z_2D2?;^ApU>a?5E9sb$Cnt1Q9NBBgbHMb1g2Fp$eeoO8pb6zrM-#{O~V9^&W`mwN2I52qK zF>Dn!3{MRHT|4L6Ip>6*g#E*p!}j6b0ne@xaQ|ZA?%8+GzCGL(=JXGDKO<}swhg<6 zuZAy%j|AF%>F|esd)4nAy(&B`_?@~R1RR8uwhTOkmj`@E#|UXmP&`#$9!_(pOd_*#CR zem0@{ea^?|ytu6gd zigq*q@x*bpc!Rd)lktt^*Z9C<3i%xE$7dBA(vyD*#|IoNp2QnC7x!)(EW?#z_Q5k> z?l(~BYCMXst^e|Xn}I&WJ@hWTG0@!43F28A8zEVyiSoE;p1@b!@p8e=^>pd&*kom@V*m?UcL45m$uw}r-s|9fZe}F!|C0rBCSNIP< z?iD^1ULKwlW``^0Uororuz&bW_-Ob*_*mFEd@1Y`_6_@mgTnX155f<_QQ=47*f1~r zCfMdp;a}ml@Q-kP_-nW}{5@FC*nY@@hb&k=tPyDPcZHq8-r4h-Tn9?pq@myM7A60Q#BTXQchEk5C0n={>$@`u3B)9)9Xwb-nk z!yN%X^ILxt{uF){{C+c^;>F=n;qF;?_bY4vnzh*M#b)&8=Ce1S`Q1=l{O9n`P#86v zSFl1@Gi)Bd6!0yt;&lNppBw%W?h6aTP2msW$Kmtgt%0_FaJX*Hb#s0hcocgFocpQp ziQsj|@Y%3O*fUsXzwrIQ+o0cn8h#us<8k~?!1eh2cm2NgXToy>zxwBa=lia}(!Wuzr&(rWbN#g&D10JGFaSSbOKGfDc4tfxW%A@gf)fdP=sb`Wq+BO(t8-|g) zG4|wqc%Q}>u2z5UI$jcRygV?EL+!!bzze|F{1f}*t0_EqY^;yqSKN%3@vS`J@XWOVcoqUa z#@RmStJBeM3w#a!InB*Kr{{PjIK+HH=ktJf3eJh=jf=iI0WE2{cHn`F73fXAtyqF@ z!AJOZplv@8coWYGONYDX-#x>hm(9Oy{)ORJ;q-vFPY5T5Q^T3z-0<7*r*K2a9+|() zf@KywCGbGj3_K8gYx}E*=Z3Yzi-P^~Oz3SI`=Ic%Ks)~_+z{w|982jc`*WGpD z(S0ji6!5h^?>Elg*Kd5vPsyWwD9DfbT}?T%cLZK6ZOsEQF3$xk^l!o)fzQEP=4HQ_m7r9eCLDHabjx@}z= z%qjAF{Nv9B$G=T@d-zEBZ15YPhlZoVPbTo&(tH_l!5hPh;rdzE&-zi=KKT7Px{|Ld z(JWrSdh*-Omn?G0A{Pao0`1ImRwJQaLGFrwiYrGAnY^3)8=qslpr*nz(D>3m;~Nic z&4d14kgs(;J*Q`n@XheUp#Gn4g_~&V+rrS*YVc@H``$9B#o)#8KzJ{X+jA|Q%Z(SxqVZ*RR*fAUsXy*%qwla4=X!e6B`w+?hPuUbrx zpNwZ(+p%pI56;C zP7izz$1Nx5JmeYV6Q2;C7Q_u}gf)Zs%5|o#^$Gm#y6QXTb$Wn)R=YvGt6vc#xmH^Hg9`%9j)UU$_bcT_0 z0^j73{jTxL!~X{SjFZm}XN7se-{*xJ!g6z#oAbuN%iwLSyy(h{ZW%rsz7h5gJBO{q zx?#C+_nf=u{5hNy4hTDkcZ8RRX9wK>nDB`3@IWWe4bKc4hBpM>^d|#7&i|mjY5iRS zZ-h7V$RM6DPnmmg%oVe)m~~9x#r|*L&jO9OL*N07 z&l7_8%LJaS-~ANh^Ci{~4+?4zv+NAoRs8oUH`bi8TY%zHj0 zc$P5##dG4-bMS<5_VIxiyk8jK;W1dBmwip(RXa}SV%~aZ;5$3#F9f~^jeXDrv@Wimh&c4s|+kl(v5@J0R=m{Jd!^N=}D4lfGt47-KH!};O1@bKz7o)T6I zUj2RhoZIL8A@Hle5acA+4vz}+XV0I>W7>!9FF9w)Igbr127K*34yMN+7R`Xz)`2LoD>-sMN2Z`;+ zGlq97KgNT=ue@aZx>r!QMsvR?s7br){<~&kpOdO@^sK?b+57gan4`iu;fAo-B8$z` ztc#KO9_s`(9iH`}CXR0ZV8DU&{P2bGGj66$%@dy6&R_UlxITy@%q9HgXN4Dpjl!nk z&0))+J%CT7&+w!V4s`P+;ihoc1p0@0otKTb_X|6QZ31n*Zr~~LN9g%VeP;c=Jog2` zTq6hTSmiz)`xk>Y0;aShF@eq&oKBQHTuj64G$%Ut@V33yZdPcfl6m&f%#!T84y#<(#n=KNj| zyW;6@2mJekuzQdhk-u-LilKSOHfmPz+Ra5xl@Q83tQT*{?&+wig*T~PsXU`7uj$WS>mJ0t` z8lVs3+bc(B7X9`%dz9>On3Ie#?p1{^Vf(G1t)jkDEYDK=+!Tc=oRg z;)b^cnixOt621~>>YoN4wS2~7XFqnPW?X-GbJ#Sj7gh;AduZUPSyvwN?7;t~x8*N4 z2s{zBH9rdGHC{h%_8w>Z-4Pt_^`IcO;Q5%NHVE3}p8Nm(%-=JsZ#*W5Pvp|v_u;+^ ze8oecb@_QAne+ab`uYLWSq?+W669sr)kwbsLBCj{D<&&`wGFW_-Cz1F9zaU9U_ zG#K8)@wB~t(qrZb-^JaunQe|3S}m7FUuz&b^ z*g3o>@Q>FHtA-~Ayt-WALGvy67>tNJ!#~5dfme;kPY5{t$RMYP)8_@B@xA2_eb07q zGu@7x`P$D3YlaO2jlFH4@g3g@fp<+?Kdjnm-p49|2fsoPGtk^P+~;!0;xBWT*i5|S zdg=@2Pu>QO=6#5Z=mcCXkNoVw&lc<8;b@s=tlh${fgeh1yGM#oDsQz& z|M(-~=m&M5gxL4b0Uv)X@TCtAUU?t%_X>fI;v3LaH-yD!FFxa68nf;J;az_}91#u; z2ZV2h?*#RHYW`^7=gxWVOl|Waa}JsF`*3ThBxF(bdjG0U`hsw1_(Z_Zj|*x)E)TyB zr-oky{>$lsem*Ch5qR5b^3;8Z0n~Z(xM91%ci%1S70gk*?k@(Pi!GK3m(99trcPwL zS=-Ie^W$346VJ_J6-HF9NYmhsO4fd;2_`4M;u52=yh z%dQ@D9Pa)jbCR4Z9;c6a{QCx;2S1yxp4~rlIp2E4uu{-Z%&z%$YR-KS!#@CIm3z9A0dXX8+s8NcFOni*%~VR<#a01YZQ1 z!1uu6rw2OQvY&=S!k2?w-$eYb zdV>=WuKrpriP~#y)r} zaIect1AQtthPwrr^eY~ew;R0u_u_lGyeq=UQJ%NRc{906wZ?LN*1t21@8c=nHeD>o zKwrK;yf*Mo)(bTCW`Q23L%$qwn0evCp!R~lpiUF-%H83?R|I-lzK>SkE6~b3XK}=7 zf%cWpI4S%j92$5YJR+J}zL`e8C*bFm!@5Boq#C2Q1bKjs!v^66L2UNC@a&+U;By?# zfBIXX(~k{b4ceF9b}!9s{m!d7?`MHGzhih$n9l3c)_i>)pPcI*-G_KVps(cswhM0# zFAVPMSTQ_0JUq~>c+A-1JJOSC_ju59hjNMHUNw3=>dgZl$6vHEZHnW02SbbU5acX< zjtB8S-d4v=pZm^X0q@i4#`l1~2XE8-C56;}>&aOyv9ul`t_9uMTy z;QkSL#czgv0)N^h1;%)_pcwl)X4Hu@AEmh)9}vu=1%*f^;Dl26z=yeAB8?U@4K4u=H1thRs$ zq7R=M{uJ&CbF2M&Wl-yNQjiBTuI1H^3V#ZB2R;ce%IXe9umAB80hVTg1QYk%Im{J=R9=A-;kU9P*A5Q&!}GHJz>+}o)$UAJ7(W8 zqp|rev^0%r7>A-uQEKa5sN6ygP zDGz5EZ$qw)UrRglY=NeR$`&U}CHIv_7C_WE?d^?L+-|D7P0sK4M`oPJFBQ8+yK?7P9duwQWBzntZl!|uVh z^f$hU{3cJvahdB*3)S2%PNV@jmVj$?apldt`jfWP?{d=Z+MHoh_NL#_%}1YCSwxFy^l=;jpyZ-v(( z_sHvbd!XUfqI@gxXa*PG*gfC!q3XgP6P6F^T5vP}!+nS}GvCMD%+Jye<}`gm-}p^1 z=eahn=cU1X>6-JY)al3-tF`1!i|6G(;k3U_a9#KN;W@Fb7+0MM z9$zicU9SlETa6uU%ah=j@K$&fv^VeM2Vq{|JLBS|yN6}vux@xoz|~&;{p9fA;QmRz zGp_c1p2nAgIzHN-hJJtG$@5g#3F^&ow)lacN;^9j=XBKs&T)7>7tFd~MswnA+=I7w z3?B?zgiQnf#xqX~%Llr4so;JZ9`*1}E}qn9Y z)gZe@JeI-7=l2{BJWR*CXPc+>xZoaeKE|3sZjWa5Tot_hvOsIUCdeh?=#PY*gZt&b z5pe4Pfwp$uKb+jJcxt)D^8=rV5299&*P*V)+&%mcbE06^JgAT8JLW%Ho(FC|a2)ji zB7sk%EqFxg)p`Z5?Ddd&VyL2{vZ5=+vK0%yqe(=gm9-8jtJQVp-+~K^;54hKP@=e7L zY7cNbel|DZ6!XQ%v)a$dE%I^10^$oEzFeu?A}zgG;Cs-XHwXDeyhn%fcdfH*z}Gs& zZ3}Lj@eux!pFn&6D)1Oi3cL{<${(R`zZhsr)N zJ9w{+4hlaE=4qZ1U2QI<0r(tpit>rHv^w`shrPn#;nYA+TovTT_1{}27|+uCbI$_f zd;BI?_nN@7esU0Fe=fN1?v|i7;MIYSGH&+|M}(h+GXn0C^PeB^GEM!QKvRo{X<6JX zCiXhw<1KoQks6Hm1)cu(!vJ?`AvMylfEci5#$K33)cpI$DabPsjaUH+c^*lY81NaHJ9-s4TOjPEQ#|ECY8V!7o$8q>8C*W0Y z7TyxngYfS^8-}K)nZ+F-SB>9&rY+59>K(PIIY1qWxqr8y?&cFg9puJAoFYcKw4A86 z=$rDXI9eRIOn6}M@6PVIzbW8ou_BH>InXb33XSr)@X4@kcw2Z)kariK%KNxC zghr(^`N24kMppBOSNRaQ)Y$auZ_7^$bm(t`u}b^mLHvv()ywh5#lK=-9wbkO)|W#9 z`6ab0?+;>QbBcN{@w4wQ8|YwK8V{)Fz_)ym+k*Y68KLFn5_!_*3Gnwn0zIoilvR%jN0HIbIyj4`&CQJbde4 zR>No6xq-g+-ZJapQ@o8^_X~T4-2xBUz2~0{H1Y?7+#oK0U*KnK7x24v-Sfh0k>5Nq z(DHu`_XfE^bM5m3zg>Ncd6{OGkKosvYjHJx)&}ygt~Fnu2gOgq$>!RTcc7^+3;K)p z7n|G@%#-rjw6*>vro-8;?}x(nVYjenpq=*)^t7CmetSx|Fwh@rd-;3354BMHg)@V^ z82?qCX}!R^7MB`l_zYi*bLR#A0Iud+(o;N;r-W6*8i7ZND_1d z;b?q;Bb*yQL)@*Vj1T?d@WOzz)Uc`f5MR)yqh^l=TrRku183i}$UQUqm$$(Gz}4Cl zPx2%1^+AEJ;d!NEm%W4co`c~t>l_u14(^4dwQ(%{Oh=mot_*mV#=bq!>NNYq0)JOt zkfv9!Lf_NUYlU$R>VBMq&z+NN{IcNuw+u8ce}lKNZ!kCWvFYI7hKs{r!li*ew(PXP z?=b)IHq3v%kB4oGZzAr{zS`5Ape_aJh#UC!;)`R$_k($c=cBIWt$~hyMo{A@mx>70 z$y^rB4dO{2AN~C0uygoG*e0mm!;>!#G|mfxu_K=|yaF7}zXs!r7R71C94^Fl^pYH( zSKLT1<4)`2c{R=(26_Hh1$9_&4etozU-uH?V?6EH@G{<351_`@+_6y@$M?+cZQ*zF zG#uBMLk72+8$5^8T!Le1aeAIU=E>09Jd8a9A6m>n`}1#rm!s|sXVI7H(WZG=9oyC6 z!pZ$D^U4<h9qy!7>_|zd_sX8R%v3-a6C3%htoOzH6Ob1OEMZ z_(a$-d?DB-4Ni+68));hg1O7_n@D}KDhe_!F|ZdOkqCq;wtGU+|q>gHg);G4O@IL1-pVdM5g!FUq~8?(O%bg9}|^&jF{euo+m zoX7`$TENphhL;EY{GQ<1A-jY!Z!l10=!qR!MBhHq>aZqd=bldCGwn`h2dey zZSm68fs4n@A^cJvDQ}cd;T&k>tpZM>8F9v_pH^RNe)13ZEevf$f6{yECtzr5e2F*B zaWu6YB2LAv{9zsi9vZ;C13vprus;5!&F>97WcMuLGj-hRG2{l#75EkCU)RfWT2$>I zzr%Oj6UhIt4VoE`yT%U*&J%oQx!kCIn49r8?som@T-tf(KtJyqb_nJ-c}X=J`iS=2 zD3}w}j5rS4)27;q7sI!q=k+@}|KKpZ4e`O|;nhJchkm+2SUWsB&=ZdjO9y@jUz&#I z+x;>8I-D2I5B|pA`aF$&b+|l;C217voE3P4M+O?ky}dLI?M>hC2<3Tb9yL?O(k_7x z=A+S{`s!+NvoVX?%_DqG_Z*r(=4)O$%$xP(|KkDAelf`9@-O85%u!+k{7W;75#)q@N36ixQRics zISgN0Z`ZJU;B&|k%SYgA`CwZ6kYFyv)yDg{ye_yu@ht&AZxQ%E{GAU4I+~vNNZ@;@J>&_= zq1c|h;<|yxULiaz$S?k@=k4KVe2j~a4?hV<1moj-ffpi2GtJGsLoq0BR@(~RcGobH+W7D{VnH5Yw@pXbh#-W3T?($ zIU)>Aj4yD97+oD5URIN^PIyM3gL%Vv700S2101_BMx8KDiI96WwV*{VU{9>M2H*6T_ zesd8oMlMr3;9B`k?ZR(1cd04FyLcIgiyv^Y^>8%cXZ0MTcAig->jys%{{ttx7PPb2 zg|{!3lCO0wHwzyN=J=n5i^JXFG2OBD{Gh+88;}$C>>Fd|?qKZFG-CI+2KO%DFYzc& z6Q9Znieqt}IQH@&kM@s18@qt&UzZIlhLwZbs%M2~hv$ap1-{0b3jr_VV_FydZQa$w zssaDva``4Yc+r4|N1o|dvwk&GW2-)l?`hs}{N@qQFTp9e2KV5Q6$37#b$P|O*L;L` z={H=9OT6M=JdAhc^JrKceOd4wT9|J^`_Z9*qJeSh8v`!oS*vRu-ox-6=5`N*8dzSj@hLB- zrh*SC7&UHjEw56{EA~A;sNbTy#LQynQO8H?@k421{IYpaKS3+vB)Llbq?W>~zt;&H z1YWUM>(R{W$Of!Oclyq=C*Wtm>S5eNj|1I*w@ko|2K?}w@h8s3rL?bnARf044Q_pN z0NsmIZG--`JvmqR=-apB;|0sN%GuGcxDl7)Q0Kl=F#l)^{74T^Ygnfko%;(m(E&Dh**tcWAv$)bde7F;5!eIfQ;$GZ}fAK6Y9Ov?Y zdE;LFjbnYb{{+?@HGgp{9kz*&uqfKYN%g3xnGx_b^}whmhavG>~oP7!~LvFSkT zcyF0?w+a3p@%eVuTG-|;friuvz1kJkis6!yu7(2wa|-h}n|#M8$xeH`AuFz}C! zfoF!5!xIA^Ld>To$LKqtw!FCXwQzn6X*brb4c<#~+ze})@_oc*YoxVE|p zIXn3vbMo0vDoJb_2e6$10jJ9MemDF){3@tZ(?=fJZ>Z@{ye;Ez z=RiY<-S8@HBOb*axPp$8|G+i8M=>XVP#wOwQ+&xw6l?PD#GX7~o*>Thj0oN!4&#yX zA!tXuyH$995X;lYbS5sOIS&freYtZ`=Y)rUFahtvGV592_jt3@`E}Y`Z7&`E_P{H6 zb9ikqC#ZokKhW#^12t!km8Rt-jAuj8x8iB%=R9%n@Ilo&@K#68iZ@FTiM4UFxE=p$ zPo9i;pNtG_ppAzPrf_2(tFE%NBl1DMK8*IJvrbm zvAkD*KQ)N&SDnCn%d9gu$d^4fJTBn8Ck5wWyUx$?4bJrHZ+z+8o)v7*epU?hs%_y- zx`K9HHn?A8aOj*x=FI#q*zG-Yk(WTn@_3CK;3-@i@ax~gjT5N-w64#rCwE8t%Hv%S z@bK>fEvybyUB;P#C&9=2Vc-?Z-{Ix|4YaWOh_{BJgYoZrVel^vOiQU##k=avMqS3u z)erL@@aA-F27P>Mz{&DE-s5I<8@L2~Z@}MbJO(Vs-?-YnC+>++qorO8SJ8ZYO!`ke z&fBE3=|)--|IwMeQ~Zfr<-zD+wb;B@euf+%50+-#Wb({DehB?K;Cs9T_YU%CaW!tn z<#a7C7RTGCeLDu))N%0;)GQns@GDN`i)%Oc!iX)z8}4Q1weyJNRq?60bY75$b-nKj z;voH8KQumRAY)e?CZ03K#hYTwprQ}|{4GC?Pw^ArDxCV# zKr8YbaNOI&J0{pFj-@4OU7V-R6z_ReZ)`bE^u3X%@f~>^TG@ASr2PzT#Ow4mejM=J zHatmtn-k0rG^(7G{1q-Vm*B(UpUQ9f3}0Fg%rRaqqv`pbzNf8*e#NIS?G5k*#M|O^ zZ7oi~sbXYxt>RlcQ5`Ssxn;nkbQ(Q|yTrIS@(BS~xULTf^p%O${0Z}eza8YFVX@%e zcHmKZ^}PoL{6eeYAKS%CI1HcaFZMn1RgMLx(~suog*76M1^n!IaUMR#x6cWVU3`u| z9sl6BC3}{?`Y`a`F*!Hz-m(V=$Lw63@BLH%#(9o&8r$&QrNa2Gw(w8;$W0g<>V>_+ z^fDgqZGn%AOYtg>y&~XH8rCZ>0S`}e@8DeZ5tmPDBk-&`ieHA)0)AB^fn#|Oc$Nk} zAjq|SG3*>@;P(Z!qG}rGUYhr1;l+XXAfL8YP$NwHs##h&$c;Zf;NYc$XC#PmaR|tt zgL+~d?R9*vHe$hj3ubDC@h?4mS-3dxE^w<_w%>%IXV2`p8m9ztwcH*4!eca}+&*rj zA-@&8f@S;;unu@Xy{>y9a5Z?pFh{D}ph4+R$3R;S$=iDCDPRl#WKbot| zTjntBOQVWAb_{&@L&EXF9H}4ZZ_5Q*g|@+|d?{m?Hu`QbzUe39UH(Q4D?cJHvQ)r9 zG}~$cR}MZ^8#J)N;z~UD&OkT5cLFgvPQ{IH3)aJ%)4;#H1iXn?<)zg0 z<4^m<-S|>{k^H&boMWY_Xa1E{%|Kc@V^zeXd={x<0hMXH#47itu#E)WW^R?p` z@HhQwos|O3h(m49@!&Z78*>4V3Wtfg<)COnJcW0{af5c1`!oTYfcxuKLeW{jG{I3>x8l&DoPK?G8 zH`1Xr4=t)*j6bVRSB{J}rHSOq+!JLi^9~bH zAIu}p#~h<(n)Vk5XnS>ha{sQS>#rY~@63_rQ1Od?uRqfSyr!*!oGY)4|9@Z*=hH{V zIX|?+V*afJ5;qe|5_+ z_!EZ$?TJ@?Z)i_kj63a*ckt}6YIthEpDTqGg5&ipF!8r~CFddbrZIV(Vs3nh<8YaC zS4SiVLzl}54fy;22l;UxsPD@$$>*FBifs@_gdyTnueG z{L@c&znXcD|LNKdeJaMM1;yzjHXpig+8@Mm;^k$-(t$P`n$LU7@GDNkwO(;1?zG=h1-bUB%FJow)hD za83|ApAv8=&O9pMxxsOAw)CB`qTXtN?}^*w{Ea2rgHMLjXb~E4+kn^9#;UI}_T(t| zqHhU&(NXh?C5bR8EgyOkWNTy(9H(`}SqOJZR@J z^rrTt>0M9PTE0=gP&57GpdayO%-doU{hU5nFYxWv7Sc4vuAB#+FxKfTyn;K$wQ7yz zzVQ)#Ctjy1=|G&PPRV_}o)IIDt{$D=IquEmdww;jZBm!z{vwcjQXBB~z}r+epoYMA zcbUL_0b@I(rog&t3;gXf`;^n-wbHO+d;CiC@&Yyr8wAHqqvC!%>wI{;;&{SKygu|O zy?sRxcgw?#9K75)Uc{C1+3w%sY05+6J-~zGc{b)GZF^{t_ogf9d)jh*j~A-e!o5$Q z4d%0*!uCN-&u^hQ-xb95YCgr(;%jmC;7D_$>x+wM!@)yzrV|va;-FDKq`njX{2|bD z7lpqBzRz_*9BO?K!}^;J zI28PikFDdobTAFTm(Z6-e0@vzq&o)3=$Hn-(x~Dsd`gQN58~tf11*R*@#P-j^FjS7 zO^7!?9cV)L3($gibDA&3(jb=)_;d3x;%m8ITqv#xNBEUE zV0e)DL##aF*5aHrJNESM*rav+Z%e!_USR2(S2@s4CJ)@jc4xbpY?pl zHf#@{P3N$@w?E766KG9;kDQkrmiLz9MBXD!IQWczz)zrOwS`ylG;NFPe6DR^cn#W1 zJ%sgf;lTLr@E~l%wtbGnwLNWazM(nkdVJ~HixtEa`iNM@ysWSDd1wfpym2PKEA}&1 z`EGQO7*cF5_rbrT%a$lVLaVJ9Rtq=>2Z@{cmN*Jm$_Wjw92_UNgC}v~LBYKPo{w-! zI48)*yKmr+6UepyE?EBSpqBK!;QjgG_rbbqUcr0oTj%s}e2}9b_t@iMoch@?IQ0YF z!|Avkzw>ZD_>(@J=1w^@J|+*7F6IB4L)0&c(Z%K3Oid6z#C^EZZ!?UXx4bv_o4-sq zdi6J6G*|s^kT0SmaiTd)J&)Xx{JQIeA9;We$A8xudTsiUnctZF~ ztW784Y4JBN7N`1$R~;*E#f6TYzFe4Nr+o$SJpJjr&e3-8r*pM!`@p64XTQ^4rSmrz zI9})R%W!INp3c{KYHP&U;$vFq5dmK~F&Yf_T^{7i>8+E4+zl^LZKQhv_76NlKH@hf*k>kt zvd@1!LZ9s$_=)lD{yLKO_?Hx6N_i{sW@b+Onc_QB#@VDA5&#O)!54Si`m8SbR=ELN1fJ?@>zTUe2xq0SK63YAA7}7+VZ3#y&jIifWPMd4|RmQ>$ylopd9f{C{zzTBU7+>w+WS9`NMW z6Ntn4&+fM$yvfg$hmmiV%aOwt2hR<(+VC<*433L%4UWNAwBX1~B)A#Wv&ZZZfmvc~;k27(;nBI6Xckr|Lp~leQO}^&$rtYQsX1|4sALVYu z@H7di?pb`I^I;{Z!C+^<#)u>gEwg-yh-oLmD8MflitKB zyg}ZdIvw1E1La%fyVXakiB#L8_Gi?Vstr;jq*h4nky=pq22H~<`Ej4CD;n^(<<^zo zepiq~-zKO>+9GToMxBz}`hewI2J5~vyf=Ix*!Ff|)ZVL8b=+!^@cXDm`bqbR;#Rzh zQ-2-IAAbs$1-h0NHowr%^r)I7UZ#7Oco@UW^jx4X1u?kmZSK^EjSKxsZ4zCgW^boJ z2Z<%c-0BnMM&y=xfX@v)NwuB$krwk@4KXtA+drriqTz58zN6{9`uphMGj;B?B<@ms zZ@F4eHSlyGZHWtMOZD-W1kd@nGF%hXcHTIt`&5^ISGYIK51yOny=AvfJx3OY;!j!> z_uAL7Velso7sreB)iH_NohQ$gZeBjP-#}fKyqB6yb)WRMybr$%IvAOw8T&@Q5{Q z7<=S@#P+xs$Khd|jhFE=uHzZdw$r(w;a}3qxDL-5540eE%J`uD2EXwrJre_uiKlsy zBj19*#M1mn+Qhv|G@x-uv+%-b88t2RAWzWPqzgxUNdL&4sN?jxm~g<~JihTAvE_6e zI$iI{j}uSRm||P8w-_8(h|lrn@c7P}sk-^U|39<*$r<6aK%2^S%d^mUxR$o0BZ2+{ z%fNg36Ccx)2Tk(gc$toMFZ{wa`d_Vf|0_WaraB>*-I&Gjike zy2pU#bo@B-oZx)V4;O?#2fhHFNwp;LwJa}aprS^oQk?0`RXl$_?0Fb`EGd|o+Up0NbtN2 z+KquG`S~DDR?|N6)j0FRVcMS*Gs{=wPqlq&QRJ*~sQ8&q z5krq03mz5o;nIb{F9XZvYdx1rz8c@skjC}$0e>3j{5D>lLTY(mno=CfKjZ~o5%49R zpsOwlv?lJ91HmJ_O}v6NaSc5tFHM`dZyZO8hjHUH2Ody8rZ|1N?&q7;`H1Op8oi49 zaNW+qeL$ZKbSjNVchZZrr~6Z49gY-1jfpQ$2=pHwn-@+EbSGUn;_-94-$+c(!yNE8ZE8K=aZi!$+SefgM^Af}J9dAO z9F$xTF9FBP%Wn~!r<%|W0$u&IFgNfp)hLP0{TsDOazS!m?t>rDo|cK>c^T8sELptz zrh&y~Er!Ntv@72Aosn0%GGElZHFEay>f-(3i_)$%D?RGEy2nS%jyGQs)(G0#wbz&48Iyj8B!(-q*UK5}D9PiSA;5+{f z<8$l!4!y`vv_1Y2eiSG3L!WemQvGd4*O%Q$^_Y#Q} z<&X-T{&w?t3s_k$NlozMZ*f71=>j`s+k33ySyQ$Cx2DE}!A!*Q<(cx+l9ju@N% z8_y#hH8`{i9vgA9xEjRJFk)<D?|fUjQLdODNjDn%_>e{v zZ;Cx};k+l%u*1^9tCcVgmq#fO}1KtCE9(vOEX;b`-|HSLGtK*?>#p8TR`5zjY@4$yJ zhiFsRhd;vi6%&XV%#reS`j!6A4|4w)ZscF7-IH(NtI;#!MVv?*ZWZ`)yhgbdT92+1 z`_gRkBJ#<@C&WjC=kOGs!(HNFoVI8X-->e|5grqk4^If5+41!7%&ht(;dK*uziHScyfnNptQA%dw)^Zey02nKa<8EwUnRiUU1rw{^ya%y@{#iY}N?8MsWgacdW)o9PJr6#)^CjJ)x!tFVcXx(OANd zG#*`Ne5vQ*I~s5Dsq(74L!2hpDmS2(hWS%!<-U8uodK`i5@g4DRJ+(zAHZ@jGtE>DUK{kMpw*;PL@}0Dsi^ zX_Hd}PuBdQefbjRANr9GLgTysUk?0feU66o&$v;4%+vP_20Y2z61N%io;UfFpcY1r zA+G#Tpc$tu{^aLwK!O94~;i;<<&hy`G#;)xIM_d zO!MnQCcfg*J!=lnE*o$y9-JHS?UN>0v1i>qE-Vvxl0EWjMqlzW<;5KfO)1v@MxZC1 zBj40H^Bv`<)(UEZvoRJjjCr?Z0!tlUoIPQC?fEf}G6Lg4*u6|0S=jjz)fkAm{&? zKWVwan{?mcOY-f$nX= zW&XAtza1WMq3_y;ZQ)CEK6rmzz@NOvQL|$oyhhs@UgNkoX7}pHX$P^@&H-1BddN|m zgC}{6{6*fPc$cR*otII!z3_c4?s2iqXMjh6#sqbe<9qm3UPrx(*m)YfN41o6C|!wr z<+AC>d+)nEodeJs;Dm{od@g*&Yr{#X|9sb6Zj>oINaVRbo zKg$XEp5w#6pAS4xpfUFh_!o!L&gK>Q?h$LdR^~m|d#9l9>9=AZ-ndwM{jg=&DX3`{ z$I(CPA&qyMa^-*{NG-Db4Y#ki}R zvbE)zU%PPagdgdb^*J^!)0m97jF_%1+Z%8GB5)plxNtzPbzbA(c!Z6PFU61Par8C3 zbxJUD6K=ytxQ_1+|2;q83*@r|IX%A13&NLTF>GgcIL^gobV__jm!!{@AC-#}m5Z>I2I z=#z3*=qqBe^ccF0`OhC`E_>$%{puOPet|fK7`ClA4Sul|1Aaju?@;UmU*bu*HF6{y zvym6c*8OB4KZtI}hUY`%n`E%yIGl+0?it)CV3Thhu%{moJUFOb&ejGty8T-YzekTN z`2qMqai#bU9ansaKCEtHOD_)&44N;==DvS`zi=fElPl^o{u8&cg}Voif%AHfYvmB( zK|F~QL2TT;aj0u?EBPe;L#L#d(nH0_`HIE2`6u-id;Ele{cT+iopQgRaT)!~4@s|N zdv6+)6Xg)H!ReFYGuK^z-6R%GpS&!nKM?ov1@br2DcR!eVq8Xd#Bp>ydK|rv9yi+H z_TjFAUG6x2vbou3L~rz&?&x^0v90YazJmMskLa9@MYEOYhjc^wfZX85TiDUYKRbG# z;F;6)Fi%PBq%mlDJor#9NOO5`+{kZ@N8`5gnyv46ul(sepL-te6PKnh(hu3@w)Dok z29AxoCp{BaIxcUBk5cZBcr0C#Us8+)XPO)o9^6@bo{n_o0C607FZ547x&FCEJlwgL z2RN5Lic{&Bbj#ZEe3oonx+T5r?LoPbK1eSd^+`O56UBGHcRFObksXe|=zsEp=#OlC z{EH_~4s3n5-|^!D=PLUxf%4r)`N~#~cMvU`=3_SZ5=Y}^98PD&v$*%HfNb#Wd?VUh z-?JacPcb|)ob2bLvPMJx@e2Yze>r{h6l?J0O`D_UD;>!^r~m8}>=wv{!H@Wm4knI6 z$1KD_@Xg@?zb1}kqvInwq}Ve5Ci@#V@)Oea=zRQ_w+r|b`4{;j_X?gh{q50%<`0@b zICxR;g5Y^mJU_onYM*0=1kQVIuy^pdfS;2+PA?Vrk!OjMb2(4+%%xNGGam(cf(O~? zX>(5VdgzO|u{k^C(9y2Og^Mdd?l$2-yof)YBgY4PX0PMUu3H>+Nj^>-S8kIR$ZqJBsbl9q}Q3j{b$0sxRUs@fUVBeGm`gpnZbI zxOZRJeG=z>%H-FVZul?u?U*>YbN37In%Io^4Lh06cYGlBf{*F0>cO_goA?q}vaiKz zYX7Sfo)&k1Za{zCH&`C18}3w=I@4eAD<6V;jms-`r@Mku^K#y{|*mcAG|GiXK-wA%na;*d+_Gq$l#^H zfx*)Qy!+5VyohaH-{o8qZp54ZGJO~EMTx)DJNW|Wy6klI=7;4sV6Tg7(L?zY#LNB0 zFdJUXpM2nVAS>wAazx1!ZNdL3PDIu$540N&rn}NP>7v@2|C0}-xtG?{ixGJ*iT;s0 z**tb}AbRE}1M7a?ADkM9{m?tv=Wh#M8;IRm`zZGFj9~Y`7?4-VPTwV9tM3?W7swUE zvHYC34b}}fe2R3@ZwK_$3j+C%?0EMApT)V^ z_05gM*T5#aC=g4>+c=v)R5^H#-rnxXoxAS0BT71^ovS9C7iMfbylIA`ww$I0!H)ANWx?#=^)`v&(3 z?it9-k)PANolWze_Kky$f(-*(-!}`KCx=JQ&o;q!fjpsI0y#WS4CEBy*K+J0d3PnA z-E)9_*jVn7>*O6hCQwiJl8fY+`qM8*e@x7tap=e$7s(l6=k@)O8Fxmv~y&g8!_ ze%VsSB|THF$kKp+%R3XTJEpgYxr)Jh2M!*Uw?g;)O(51w{}h9u7qQp*^5{_t1b&x?k{&04F~_*eTdPV865T+5MXZ8w7MxG5feM=J`r_P^34K1N%@j|AM9;82<&b7McSPo86Wab zl4)|vPYF&8aG~6C{>mc)e8^{MougbyaXY^kCEnw`LVT9n1pJkJG1eE*E$Nl~*z&^J z;&=-`$;rV@c#y0;Gr&*c%ydb598m&s7aEJ?xGfI0 zpZ(uByz=qm`GI=yZ_+*SBt4TJdvw4jdHD2w=W-p{;B?8S2KdrDcj$?9$mV`mN2DLp z_2_@}J-T609np8T@~9{JY@cIvMCbbK9`5J9>~%g%Hh%q@w!TAso1Zs z@inf+v21d>=V+I+$^RDoA)r(8NwTlmoudy^?m1l(eCF@u%aeD0et?_lo;U?p(?{LI z`TU;XI=)e{BG)+o>|k|Z+w0{mg70i#IqdX1oF%qRujLo@U7peiar!al?Ofl*p5%Y> ztBP-bEl^MO7pL+p>Z^`?v~+CxINyXnZAYgh3&p~0xpIT3Os*em#x zon$=YP`ZNof*95h%qe;(d(M0__sm0c^2k6Oc2&SW#H)O2bRY3_ylM>(9Z8%Qr;3xX zXT?v%iB?Q?Py8a*B5w!p{4fyX!AtTd>4KXFe3jb;xQ)Ju=i~@IIoKn3X0U&N58361 z1uqF~>8R}Y0|MtP57_Fi!JBOU-2?aAC3s}OhTkesc6k%O<4rl7`2J6Uy3$F-jrd^Y zRM9`_n*0);Ra~7u>D_QU1pFIpa{8q9!I!@e*x&rea*J@KHBs^=@g)DUHh*0poC5yl_pJZb*1qb}&ci!m_p|3a z7hF&Ob{_q_d85h)y7#R9t&Zda`0O~HT{)hUe8X|tiVV_^(9hb7ABCLMuiC$TpzWQ< z-z?5X_5nGiJ^8P-x$E^=;a>WWuf}*V29^b6w7J2);g2$AjcGatU4yJ=U(p5VB>xKN zE9^IO)O_VX!?kR_rv`MM0|I^koa;SBIG653k78e*8^}lWz6tgxoeP)n>(Oa(3?Cq$ zp&U^0Cvh2mM)ocn`1OHYAa*ppj}L&Klul^PBAyd-zdT@zivjTq@NFg}6joek{$_zZ zCGmK9nevY0i@J`#mF@nE09VSX=ZnRe=LC3D{=Ik>``vm?xdZYD)Q^pRNWez7-c#(1 z4$42w-@)%87DUGIgVxU}UrCJI-&J{EC||SRV&mh~W;j-Edf&-`W96mOL)i}2>fl*1 zA^ovaKlt)<##$|3S5t0?Z?f9@AOifZR@+ST|Lr&L$r|97pg+@zo#R>5zipttqd&H>EOk&e;g-I8a@W+g-D>|_U7gp2S8_)a$9l2JyC{$Fwc$KiUQ=Z~t0nB=3zW{ca2yKjx4zV*D60a!Bwf+v?)LoD)Bwzy3WCQxK08U!bt@4Vb5N zAMZurFB8Y56WuLfL&~p~Lr8bpIl#gE2;KuB-mx;k3v5+>JU%^o?>PZ~pWGX|@5KS0 zvA$3~d0d#2JV`tx#&w$jFIl7L9l-dRe%u@(Ti@*$$8OH0c#U|rKRG}0Dc%2?8PFvQ z$LQeB6d_|Y-`1o3#|nV%Iu;zm4;mz7su zR7cOihw8aJAOC^pr7vg$?V)Y>c(s$UuFZh_k%tA^l769G`9k{VyZz2_?G=Ic<*%4+ zQ+{J@*|sz93T?_yY!0#;FA7|*uI8fq=`VE#WfY#Fu+Ou(&a;x6?N9et_9!37Ir30n zkYUiuhhTJCl7I%`xpn6VaA3r(sAN>9CHp?>N?}l7%@Keg|->_&K%K)Bj=e(8 z)fS*)smR%^XrND)o2W)@y)cM}m>RSFa@Y%kN0{3z4mVxO0XO9oWL#&g=Cv;qK5^-nn**?dei*Llm#7^ALefbON z$=)kMXU0E`qwpC%K4=Vu9tU{-(ZlZ};;IqiF2%TCtec+hdiFMc6O-X9kuRhzwhqV@ z?W3Kv&5l$0-9} z8Hb>3FpdjcW=@Q-ug9G~UT=T(!!_x&F36`L1<8_oH!Tvk`-f_)eSptJU9CcU&>V`FSUz5`XRp5Zli992ivFm zQ9tW*eLwbp`+n2p2jWFzkA8_SjYZp$GhJ&8=u7=f=fp?2M$A`E4Zj|Lt2`TdQ*sme zUV%-DQ|%WY!FPBT=eh3b0o@A+(_`^oP~Y7CmO`8ghwz5~rrKn9KUdI2hy9`r25-6L{l)lAYvqb*;+L%2oU6X6)A5 zg>(hBVe^H^{6gHtesB&yC0lZ|FWH>#2kt$q`}t1)Yh7?FPE~);;dwk)Z5m^@JV%!@ z56l;w2j)hhjlkS7Z|O^PF8Wq^kxpn#vwO5R`L8eWf^lhlj=a?Ij7P|-ksI+I7%OB!ZxFzq%J@9@8`yA`|8acDvrySge_X>Pg4s?HG)V;-W`NYf(avCagADP@-$DOC# za<;PF%e{?pbxP56BeuiuPnf8t3Df#3BIbD>K#_z@lzQS9coh<2CsrcXj>U& zu9H*b9QaOlkz@28I$z~&eR}jiwg6pk@ z_+V`9TNyZ~*L6F4z4Hp^J6@Y=(ZX8%@A{V8vXu|^JMK*O89S|Z7(bBQO12y>#=p3i zjfcxeE~Q)2JK56Ohn-chtqWYM9qA}^hw3Ng-0CLf)R9l=3v>l?3a6USwXMukbCkZ& zxoHlXYvz}6R9g$DRCk!Qw>sXdKXlytUUUxNug(M9R?ak*abES z2G}%goYCH~kM01@={4iF*N@!XIcNS?KKKkHN6d1CVE zU!n7{KMM1%wum{&7O3B;eia;wC!yTR7N9G#Lux1ENj5Bdmd%Sd##k~PjUIIL5RVQ^* z@A9YT@mzG>>aNvS>zf>7H_dn*wY&6LK^D?y%dhnhvODSgwZHfx z@hV-2?quB1rPvAdSL4d~!m%U2)(2%Q;@Rpf#wWc6@NRWo96HOrbU~ok81JKgY93U_ zF+a?can6k2?O&W(i+W8zcRyan(Ky>2t3Ij>pY2mtIlX+2(_y@}buy>vGSzvupR#5- zww&ATnt0bWj?eOi`c~I3caSNC=Tf%oT#v)?eK{Iu&GKJ)u-lOj*{B6>#D9*Li>vP| zV+7d%l_CA!t#ff>`I3BU-ISrMmTNomvg0k^b$GY;a$jX=)7m`kYwgwk)>rNC+Cj#e zvA8BXW8_o11^WV*`ZG`X^!Y64Bjyj?L@a<#YED%ztskA;X#RouSbutTPJCKlH66)3 zr%$cPUyVcQo%n<9iCb_CT@ByJ?EyRr(v{WAcjq?#SpIYj*E`p<@B`HTx8)N6&#n#VrP@T>(96jk z?WZm453BzSZ{kq)zxKto^?^E8{ggZd{W5Diuov{Be$~ffEc#t6$ha^*_*K{|{*0S) z>8P(7XZRCWlIyq+7a6M~%1JoLxHjL-HFJTUV=j;Sj5*``Qvz{QITQR;){9v;CU=d$ zpWnWDzK%JEpNr4SJd|hQJ%DoC__x4k``p{S#0lo0`B%QhMf5Z0@PF0!k4tcXbMczF zSU-Sg_ssTtM$hWFIjVg|x#By!xcU!y#3rW$6^oYo4YGFFEC^!*!luO3gV)IdO4BfDLBpw2sR9o0Nc*=fUh~sc3PIGS@_VuaHeE#%N+{u?h z#_3yJNoL~6`f}L!y~h3MuFl1+#swawgR=4Kdo$1Jq;%E#uf?~GC$=Djg$ ze7cWvoy*SQZ(s)*qwZzgx;I@HXN%Q1?itB%x^(@+o{=8ye(qcUgKL%VdUL>M_r+7@ z3!XF2a3ubw&)`lz1#=76sW06bN8y@sD$b?TnX`BZXYn7pKV4mU^cC~boW-}|4>-wI zndWu5RT;{&rQhN)d}6(^cQja&b4g$wvK*6@f%n$nJomwso&~3oA!5Vwq`hC(@3e{o zTR(?eaEJcWZ~BH@!mMl`<@<4ldR`p0)_us=%0B(4AFIdWRNu7`S*(1} zciK#!>0h!)+3uyj>PBX|PWkR(Yi#Hr^aOAx~Vzgvo<8d zt?7{ifd7?GHX1kPK$Q>DpR7oXxs`Q38&}xOAkRN;>bWU=St)ZQa5pzagc^|WWW+R$2_^`YvCP6+MH~LggM%O3M zC2$7r)F1R?ei%L+bum87CAxq)N*|z~;UIj)59HbC%5-V{Q4Ys>_}h8v%vXR<>C2vr z{H%P^Z`wv*lc{8Kb$32y_5+!wKjcZurR+K>`Q&mj$u-wj*C)&6LCW`Td+Qsy;AATK ztS{;F>dZ%_zvu~UB({os$|aRMsUJM6zA^W;i@9OG8vEMaJYu)GA0G<7rGIN@?~Xk= zkPGw1z&`DXJIsH4fPW4Q^piXr{Umo!4j<0*{u|fgR&~G;VD7k&cj73+HMm)xG~Hb} z>VdoP3A+|ws5_p+hpxq^&7-DQ8XNe;d?%l^m9gPD$rI(OHx9&edbnA>M7QyRW{YquBBhI^WzgKg>__(r4H4PvQt%NN>}Z#({pIOW-Gb z?0kABZly!&6EQb`#vb{_?qF-^Pr8?S)3wyO-0WIqnw#W=cJiBWek)FnuXZvf)PXHc z9(e}Ws~?$W{5l73=^y<;H(>wZMDv3Fqz|=QWA)^kYxOG`Ypxp?)f?Q8Zl!E}LB=h5 zZ+2G^4T z#)Z1;2W5F?y3oo%pV9ryM{~@arz5MQKEo+=KXcRh@(s;dJc*0wFZ!zOhg;1jdaUtp zj?tg+=V4Qjjr1^n5YLL6^s71Vx$zJVqD$#lW#I?o5~tDk=;_Yaeq^$C*4Ns-x{J1P z4H@Y^>S_M!Q*CIT;ZNgO|9Vz((e=hp^F2I2em3s(H(t{xd}nkZwyXAa%=4Rf#x@SY z1;&Ol__uIp?Kkt^yv5)AQR;5|d4Bz592#TnHeAgnB>%`kd|*x*&-GtvJN<_PU2FU~ zhiqqm>r*ffjbCkNezTeIJ%5V!@tiop^>V?CQ~aT>z`k@p7jxWpgEv^G{(_fRk8 zdluIiN9sl=)DQaH*t2yHGS>b1R>^C=0D7Q4(QbTU><0WxS7BQi+seZi){6lB%NQaT z%JcZvoH8fPZ90NJaVn!JiHAk1=cR7^lXOvT?IHXe`r7=oq)p`jWc_>jv}=Iz5h6CcTJmY+UPm_8q>l z#$%nVR@o|eP_RMZUhH1HMi$Up$ztn$jd5*{+q}~uQ)+VU)_IrhjRJZjxkkPi(;b7J z*BHaSbSb*1a;%&2UJiOD9a8)H!!bDB_;D|7fQQ-N*6>-+XT6I0d)J5lr4z6#^aXyz zakvGy(^H+}J#?!B>#gaxa?kY_ey9K7W#e3%`OMdWYw$a6_KscaF1`Q9+IPAdJyMy* z6kE+aG6r!5n^)iSt@1z6$F1c&C7>shZQ4XRWCM;crnIv*Ge`ZAJ?cnb(Y|Dr=kjOV zvZKsR+~hvSq`81oaW38@Gs$i8*Yg-V+Le4Vmeia5rF`?(yfVLVi)SPI$QL@hv2NU& zXXH11&=_GWSr12V!%OB0*`m&5zxwktk-Ov>8<#C*9h`M@?t$m@KYIvI;|u+#JkOz? z+K2Dax@$JHwQlrNb3>Ws8#WfspmWo6>6iF=MX)STUw%_-+|&i1(qqXs9HqTn>wdDk>H&b8Zm2Rrt&2xRkk8B?Dv$FAV9-9OIsJHXvA?g?PC8M=H zdsiLV{@w>dR^fHy6n~rN`cqx`iEyN6@GNABy73+3bas_7ZjRt^&q+VkPWah4qf@lL ze9vqtx;#Gvn+$KNuW`Vp5$CZk1CQZt$MBW$qprq;HHqv8vI76;S98ibXZ9+6UOlvd zamc^T2gBahzc|d;HU{vx`CfmcI`ezzU;2i=wDwTF)xlh#>zYT#RNKi|^sMwAW$Fvh zr!3CgNDb|5~+0r<-NH$QNJdphPk{cKDb7dRGA8Z+9OUZf6Wt}@uZ?gQkG{dmUw z_blq{xzq!1wC-e*J_qAjJ?KlGS-Wd%V-z1~L)W`j8-jARn{lZ=(* zkA9_G<3@X_GoHdX?!oTV?)psK$P;z0Z_M~M){GtF5#M-LbIbMgAmfIf<{Wj@r`lIP z*^mGAD}HB}v0dTx~w%A-rjxYIpNa+Zd<%fZWqI z>^ZWVe~QnFKc8=moTqb{wERkVorgOgWt`6p#HdS}`pf}(+ z>hAvPPUoaYXg9K*%yLilRd0RjdgbdcI*EIdr|Q6mLC@vS&~ECaEP6qG@8%}%(HD-< z<&{alrC)=-AoH|eWtTqFcdpZx>Z1PYr)@l|`C^>oVdKrY=DhODnZ7%&-rB&k(eJgF zHgb)5Yi=7~I7`{)z47At>9hPa>S3;^7cMcLjZJd_4;Z87x%-*VxXL|@2V3<}FLMrG z;2?9mc^uVS@r$wFdf)-qsT`^uw8k4v_)kK?lq6_I9X);;TPi(KO0}}L1r0?u4S7VANXGT;%@tlEn~wNRhIE! z40?8RR9Wg~Yri@fJ7l*pk5BO*`@uZ4#WU3f$aTIMbKP~FyN;Vf>g^up0J-8^ZLd6i zVZ3<;_3=E$J{jzp$wK9sx45}F7%s*|^iXYLJ_BCBgXTJUfLod)#CG+$x;>t!|KW-1 zDauuE_rx>uhj1hACa>HRkJ|6Jdni+T;28XbANc{u2JtbmVth+Z(LWu>g=83+S9mu2 zajEibor90nhdjbfbw2;Q`Yuh`v^r=z{5$4~kyoJ4@t8-L;#(v~7PXm(0;0$}^_S1!I1e zUpRWShB%d11dlV^|%WuTJz? zW0x+7Bk%>T!E?Bb?PN~k3-SVws~c38tSrEZxB}0RFKjXVtPW%Y_-tRV#YL{8*WoX| zI{qTuWJ^Yja2(G%7e@fzP;d2guIrpJ>A1W&26avk?hJzi)jq0!7$($7px$L`3ndqkH@>82fzS4-lltKzby4ri5uAjZ+yn2G`|NXnbx?0|M%}fAcGbSx z+Oy(q`Xu>M-E@>Il@+cT~Z=QZm?z3;yfJ1=CrjB zrJY~6;t$lPz{Do)vS?Hx~cJP@kj~Dqh z@N0EXd`wsM-PU(p$PV|6?oV#uejJ5!p)#Ruqb};|8NuiBpl7L!pl`aTXCViar{2oO z)7r{$=a1ZJjOagG{e>TmfsTjrpZ;_X{jKAnx}1J>9?sRb+8fA{%E69Df$91M?pb;2J3Bl~Zw=dC$JXgZRTaIHd3$Z{Y^lIe)yy zdH4zk;xT+jXT`hpRD8%UISc!tx-8vxM0H@iT{+_ZbWR+Jm#a_ujyp&0pVhP7n;cNL z)dAgE9o2yiz#b4Mw{0G~ch&1Q76{s+Eib(?QFHB{rZO7AvZ_Z zr(N}l^T|1StnJwL?ol{bKRTx}qg>B-1Z&b&JD%vP)rpPY&L8@UV;yUi)s;iWCn(Fl zmSap7)sF|xYPfzy=j-t`UO zd3FQ6nQZXgeO#;j^1JiZgB+q~lc7Mi(B)mz`l(N0pSpnS?RRhRnS8G;pnP@deMgyB zdEq|JuN))m$YZij-IdodM;2*I&tKieGqr#9zxL24=7#?1JTg9w7h{y}qQBdIhEN#80<=p1~UZX7??;LTSdn%KRpvU2W3j;hsmo^WKWpz>?b+J`< zbEmieeT-G+N$zj-CF04 zH`mzTb~2vmclCE_E9I)Qay`GYjREZqK5G;G13v4o){p$harmY(oNU3J_L+~`P+hdW zXRvKuNBzU~Fka_8*R=25r)82O_^3QqPOI&OcYC~Y2?yfg$|jtL+wde_b-p^_S5Ss$ zaWC*0oKtzKzTmraaQ7(FaBgLrdZ=f82EDJXvR&&K^f{DOIaC>^4GMgx9UTLo9W!pU zpZ3;%#)7f0|9sZ2cnY67f4q--_-?;>;Lp9lUpZLcI>**N_fuBqw6SMg*&4T<$IhGF zbb4R&({<|CacAqh{mS(`EwA-joEIukjhzbibC} za(j-ttEW28*4sIS^R%mHZ@U;*pl!61KmDN%wU4^g~ zZkyY7&d-j2eWD(HZqMHN>KgTMZ{t{bn+zzwsna5hExy0ecgLJd&&9Xu>mJ6I=QReD zg`3*W+68>aDf$I`wr`YGJ;(L>s(t5K%-fDJ-#w4x?Nj}(&x|o+gI-a+$7j%2w#Jw7 zr$5Cd=o{oJkm+P98E-tg7dh?T#*1+S%Imhz=eg9U?OpC^JOzQ(ZgaF~0+c>Or$eICy(pBuZan>nT~%2&4b zR;IQq*3_2vX>Xk6uk+oO?oeLwEaUt!Ki35H=zLedvHjcb`a++m`&e&n(t6*})Bo=$ z&(g6|*{QDnX6viXw23lWm&!a_bt`<=*XoFaj160Lbx!N)p88Kewl3s;YuY}LM<9+Q<}day4z@gyk17r( zKWl{CUb$j&a^<*MXCzlu?y5Cf)^`>5Ip!Q|ujIvA4{GbXb-!|ctw)pdYVCqG?}gv> zwr^$7yx5*&{i-!UeJ_@^Lw*ATe3rZHn0#jIujCS2Bjn!h3qIS|{nj!oM}5@G8VdDP zXSr&9?(rGq&N*j1uldj7hjJ?9Vi(P`(9VweF4sf*izn(2{i1)G_oWZXmC7`IOn#9W z#)`3HOc+n(r13}psm^3u{ml3T-^p0A*4QKZuY!oa+xKQv~~_anzi zyr%h5{6PFajRlGI$k&&fE7l{2c2yvM-}(@F^m0+<@)y?Efi*MMPswTZ?g{y?a!ReC zvu;Ulsr5Adtu61F0QtGrKzZMicP)8ulN?#^t*}n1zcVR!7OWvEeDCj1S`%eWm3M%jV(!hUDDcYhPpd!!2G+HY-?^W4j|$^)O> zv%js_a_m<(zu)LL8vEO4-qYdt8o}rBch9^xrN5Esx;0^G)@N-WxPO0p&6*5@}(NfW2}>&Vyu%<)lbM`^8^2r`((O5^U2(zyRnVg z6Zo6HX)g1li`UDy!VNeSKYGW6^?XRtJ82^4uxDGWV6q_p|5Do*Z-BXXg~o>G|i) zEyDL+bJpBhb6*PlhNj=N{P#eap9wx4sKdvC4+ZZFP73^1u-}tc{}sVYf)@tQ4)zZA z2%Zq^6g)h*e{ip0v*3=wt%4f|e%tNu!JmWQ1=`rVw5%VtHrD!7>wv9w^)31Ko>o?BtGR0=sZ#H@7()hdQ{T+1gVsf5y3g3IK-`xf0`^|L6HV*tQuHV%4 z+uVMm`^LeN;9m>>n*4tE6$@8Pe#`sPg-aKH6I>YlBKT?Wqu>X@_owjLKF6H%tKg!* zb^jgwF1RfCV{m!!*Wk+Fs^FRdS1(+>@b};^!5;$Gy4R0`?*`upz80Jpd?i>NoEdyE z_&!Q>HHMQ)LyI3L$r>q=hxO-yTEaJ#WgAHe7I2zev8y!kBV z2|Z+WV7-p`lQ~3R!Q=E8Iy`HKoj_N@*P8@%XS$bsE;<{%j;?4;oc9P>Uymnn?bgAr z!E*xK^X>pw;yJ(5`-fm|es2Eu!99be!EV7m!Ha^I2P=cu1g{U?5WF#XQ}EW{sNkK! z@xiL#jNs$J=YlT;Uk=Wk;;i{aoIQW`{JFvT!MB4S2EPb?9b6LpF1R%KZE!*Gli<6- z*8=zZbnv0z)IhoK2-NFUfjX=ElY&PC_YF1ROllf=JYPJ;HiayO=<9{O8>CyGaRBtBx>CJ3I z_OG$Uj^*p1JIF~Amo*NJOSxFqK3FG4C-FWzW15aim+`JVx|2CT*Yd8k`v-no8+ZC$ zalbciZp{bREnGKYCp?9(06+a>fMe$8p936+FU{MFgI@=Du<*V6%cpqrI{~i2D_;)& zBft^xxhZhhrvp5O*G>yg2;LFAAvipEX>f4xyx=*(e!=qK8Nt(nCkML*j|#R6wh6Wl z?j77CxO-sh`+b841P=}#9&8^xGT0&5af+Q5c3OCJ@Yvw-!EV7G!Ct{LgT1Fv{+_{8 zgC_>+;5zql&qoBdzPsmRf?Wdlb?)PWCj{=ld+_vNpWwN{A;GHyZSc>v*ucP^QyFZFrl9NA?o6uuiDoWFAZ>cGUl@me=t>lVRng7t!12R94UV_|+_{_nvb zgYx*#=6^Q-qu_hNHv)3tGr{`;GUV04a{}D`(BN*t`oWC?zeVqN>Msbs7tqIX_Nw5h z;8nr%f@cKPrI+S6;~yGq6VR!*4DJ!IJ?<3HtJx)fKl$>&`)IwR=BGjRY44}4KJ8sS z>>cl)!P(x!!|zian*W7AYRt**nv>=~^K9_;rN5jNUsEZ?mL|h z@CB}+>*0{@{dZot^TK8US#Y0#e0X5sUJnVj4IUOO4e&Lt#?5#i=RYfW zen1W#7Q8GVmkQrs5#a3C1aAt+Fwcd%J>%)Y`vN+1w`VMzvGCzQTabz5B;8pXYx~+D zKUw(6gp+^1@bifdP3B$_;9s)Z-`dx&J<*xhyMDci4Rpur?>OmeTVNM?FYfw*t?$Od zor4VnAefKuD+}C~Fn+(AD_X=>mF-upW zv#=YDbv)0m=D0l^c47X)t!J{Wv8_*HO4Ft^s+gp=_seumoy z<=tDYb*l;g`o4bPm~(u0?4|)e$J_r3u9@Px`RnHA*P5Rkcg;ot`LK1cZSaU-r(l=h zvB9oW`26T#$6#sT-1`OO%O-)c*A0~O`+yFAZt$_--2t28z~ITjcELRY?XY%0*3r?~ zCtnZvay}899(<%ebn9FANvvb!3%53iO=gXY^+dM5+izXD zbLsKcck`u-W6#OW4A_eD7ro<;eva4ixv}is zT67t{$ATXj=*ImXcnr5W&$Vw4 zUK_kTctNmlfY0f@^kUqL!|^wsUN=B0e+w=PE)ISf{51GU@Z%|b|9SAM;Mc)#gG&R~ z{3*C1xGMNp5VdFFrU98iM%*R1S3rK?eSA-@upQ{=Yr@V8J5S_6?E&(L&i<4cutUIp zHi2hlBY1A|jO@@BWZdfl^6seMU4gbgHDCjgjdXG{;D-S_i>}R1!^3me&rSSVBj@7x za_y*Z`;31ba}M4o^Kmub$}NI!eZF~s^Nk@s8D)_zxZWRM8ad+}at6mMgRQxBU@Y?a z^Yt5_#$QKlCL2N_x9_#V^CpUfunY#QDuZF1oYLGE6B8rjEB7uSwmiH3%z$r}FJ`^D_2%vyJT~Cler3QXd}=_C<%|ASz%Ox0 z@Yf(=ulZXB{}yZ!@MSzWczCdVz`p@}HV+Hv%(jm080->|DNhad44x6}9q^Cr8|)t( z5b&A2G~namOF1R@M8I!DPp^OG!ubp5FA8kQrC$a=59sx55W4;MgYN{cJ1=mL&jue3 zP7dh#{7C#t&kvRdWWdgW`fn9%9&8ZYI#@R_DCqx}2bTuF3Vs-TBVY&czw@{A-5edf zHXsw8AMme>DeM?LD4_rI(cu4d=iHS`u3U0y;QgE53-CYRjCWnuZ+_yE6DPkpLEk6W zUlkk{@VoJwFAMnL#7xKrv1;qg`RC{Z)?6AV=| zCY{5WJUjSOK)>gwp_7Ocd9ToiXJA~5V|w3^an29PAJ4DDUVV2!w>dg^d%%``OJJU` zZTtIM-oYb&X)e)^*v90Hc}6bqf%1il|FIp!{meffRF)A7jx@&T8T8}wiP3q1Fh;Ee(9J2Jq7uL@on;KdR2 zY2WG8;2c}~s&l_&;U$yLJQ{2{^qmDYY-ok=fEc~pM`x7TLr#@W590T(p$y+#Qs(fFa(|dz=1#b>k1TPNALUNMaykD?UaP#1rxoaje=f`tD zp8Hzx`QU@WiGdgaIm1_{?Y0l@8Hl@Fx8%A>jQ0CWzCY0eyzf{aiMt*e>>rR9{6hME zBUw#-qnEK((S98}qHn+tk`Q7>2 z#Sn2qxmK*~X8|3sF+7~YuU?G)-THa$a~}V1{WQMY?>_Y@@bTg$u}!=VKHH~UJdeBh zfj~XL_26FWp&ULS{^QXHgd_Q>$pqZVKA_*?Qg#Af#jkWB0sjj> zJ0JP1&Rt!bzU?#p`$+-c4Iewd(PILBd7#VF@#*{gJW$==ap$@QTwnjZdsm-dwyqA^l&8RLX(rf>7# z@ul#akMU0&FYZp~CjZ%#dj`t_HY1*Yalqe5k1$?wJiX(60bZxWoE6At8RMRpCad;O z0e=2_aCN}<`Okn49n2Gdc$@wt?$sP@K2*78czTlo=9pZ0JnkHMX!2lOH=B#*ejuji zm~)&gO6c+ zKL~#U-}8&6@6MO2yWP zIA8f}B=!eC)HVSdguS(KAm(z5fbh9`?&?Xroc-~g;LPA70e>Ew<3)is+98nFr2TFZ zT)pJ#iR=}Zm!HGF@Y_JA1V;ztxE!l}1O6cWN=Gm@|KhBzD1znKOm5xYOh`G@l`FP|~{647M z5VynSxMbvSd|o}2U&uA|PkMgizsjJ0+R{bsSDvzzBNm98ltEUIA(c;+1!4k#+sFkr zg7|=#<5__?C7qe?h&t*?JSgxdpA>!Ed1MCgNx7yr2CjX5urk2GxcEf@ ze#X=InmieKn_sHD&E_C`*bnTJ+754uFQ4t;{=k-DC#aY5`MYpCImafz?P4X`L!0As zIzQRowyyu2!d8C@8;pO3PsJGEQ=!MR+xYqD@cb$CcRGeK#5dmjMf&@Z0b9{;&9EQY zlE(%o2CD*G&n~6E)7#mus{=Ya-k;Uk={)rIA5Zfc>FsnO>n(m4kPCkc$O7Ci*8!j6 zUG{W&ANS*XTX_=pJC39Avukj2b0%i_-Z9+mJQ!sJJ3(#*`9l_&pXAh)fx3auYzxPn z~KSF&ScTJz!^LbYG2pQr%1)~gMSCARrnc$rkd=dN-{Ly^T z`vmgj<;LS8`RH=g#W|bbdG?aCmz*DnQ{%A<0`b!3c&01nZW`P^kOyqTu%dAK10QI`iwyuNUCsdk1`A^zm0o@w-4wDJdzHsEQpK^U&qhoW${KiI(!#=7;*{esJM(z zhaNAEOP8m=*56;9e$?l2AztSH$IIn%d?_Ak3%K1ee2?4lD1Xt&@9rf&0sKj80vm$- zD8yA8TV-SPyYpP{o@9q|l&9P=<}t=T93ykUxt^=_Yu#EW_itTVzIGwsv=iAt-f2s+ zj|?OW*-K<2ImuQdGuTyRYHb0sn%vdDWHI^7AH+w`XHRC-|0B*XuZ9nZpHB|JY_0%Z zgT25$ZO-KA7h*4vE%M9whQwa^>?>dBF?^Co1^ku#L}W{Ih!cb;op|MS2--J9*CEOwOj2G%o>lb;OKNuDZyoOp+L z_&Wo(*lPlDr9%RCn)tWa$D@L612V;$0y3Iia_y39CwWg7FS&Tqu3w+AP5jO_xx&{- zKUfy9PmGZV1#}GKnmueDu*uD*%L4O`@6}usuRb&2yZvY&4q*<{75%o-%0PU9>^Wc> z!zwO!aR z3 ziOVY|D=WnZ3Vd(BbDZls_i#T@ma@rSbyH7r*)x#Up4IcSwX_i%PusD_$ow_g5oCfG zFS#HWN56;-uqng`#D4fB$p-zneL%M$3#`RzPR?j!kO6#0bduKw>u8$NG`*ia)HQAVNbe2!)9^d{KwKN=%YU-Ij6AaDjQv0^$RlGvkrS{wu(i)I z=Q!WB;x``-&IsJ!`Z??9`0(BlyfsjtBZF55FAEM1o*V2FJS}))uv74`;C_K;-Y}5g zux_Azt_awMY!&%$?8Y+#a$S3$6wm?4mRAPWJ3K3(U$9B(LdFC=$rv&Q%{B9ojA2`w zyB7!MH{UV6&;0*6H%$DnZy=V9L{NA^DJI&ouM)!Fp&*m8`1K0;* z{p5gr9WexQ+gd8}K>Ly9TLs!wo~HH{W6;*}dbE4j{OSvOfVISOIrURx4{TL^xL2S* zmk07Y#hc{-$yt!QPrnc^tN%(~1Nm}fK*ks&^bxTNV}^brWQKgb}u z3?G;A#+H_cVf=k+noIP>#Ho$Ta|3#i_ac5Ppcfgp^7+^*bR}bXj8*e{ky*w$xdp!s z*fV69I0-q%wxM&)%Co=5zA;y7*ZgDp9tu7w_6_i(k#*Mwop)63Y^3{pWrH<-K2CFxD%uZvkkconby7F=^+yNr%xpV;Kr)}4BuASMfTzegIx+di-$sU!TV$3vK3huu zC>x5;&3XZL62A=| zPw;?1KHRcE`A-eV7sNt!Y+jq= z{A+BU#%S;Z-r&E$FRuycw_>FH9^?_u!e8`ZdNKW&-(^yPD}1-#Ib>G3xc9Jc_CC(Fuji9@$`vQT-|B(G$-|Myak%=5Bd9xlpF9;)=(`}u zSUfI<>^Z$(!87vRPz%k$(rKN7Cjz90sM_sM^`NBm3L zUwjPc1}g^mZa*2}T-UHs-NU_lUt8rUPr2lbdZ?4S$)_fV)D_0M`|g-?oJ&@X&(i0q z-0Cx^!z?_P`W5O6-QRk*uI{PM>hGBAw1?+rj~kzriEqf-eEv5!n7CqPqp?gT%B7W; zLKYh5>@vA3;y&zsHk$YkJ53CTttJj+ew-DMk07^^56|2oFU5#PnfdLkQN^ zm{H?J<{_Dh_sjQ{6O|Q}ndAxEkIcdOm4W!49IJe*3}pYA|Mly!do~X6!iPMa)5V^_E#!gxP``y_Jph?d{~B9nS-_`8KFBd;&yW#(X|E1&mpr1{H2nXK^H)}! zJ&o7+?Q*}JF2*lr!*7Ns*)x1+xO1%}T-_|dsdozK^W+8}+Fb+k1NYuLxNks?JS4!) zd`ofz#4YePf75OOnZoXY-2-uPvSr_ZZwmJx6p#gE!b=0Pgq+}uV}BeTtO(diY!Wf{ z*G<9KWxwOjcb$8De*tHS*DOdT#QCAMuvKI)Oj( zm5kwY)P{VJKMKeh_RTqgxIQ~l{+)aavir1vy(9OFT-RshJlj+rAAhO1|6T!mmY%?O z%U{a|GKNMO#3tu^Fed3$^d>sYD1-Pu=$LFG{t)v@zJR$$FEuaetmF~sI1F$Y&cZF_D*S>I9T!KID}?jOf~7%a16e?yXNQpi;In;X zg>$>kT$};Ks@V!;47tK4V8iuYJ-(Blt|gD0PhM4Sky&J_dsJ4jG03Xgf6ZZLqqvsb ztL7w36yVE<~9KA-mGm(`Dj@2&@*M>)@}PLWsOd+j*qbWHOd{{LVO z7~f|Pd@~Sx;SVD}aAD&vw@xlGAK2XjJ|VG|M+f8q&c*xuN4OvGu~@hL;5*+K znQ&Nu_wl@#hBY!`GW=6V1$T4 z>pFP|d~D!5-xyz&d^#D za+l2}md{5f_Ts%D)*h18Y!>fs`g}l+Sc7tEU=0f2rQ8hf38Vwi2gs1;1@r~F16_h| zR6LLj;Zvo9@TKv=vQv!-a)cc#mxOQI*zsp9HJ^n&Oh2M0(V6%kE)4i6_}1xP-w2FZ zW0#yUj_GyA^nV2GALE@K$UkCU&=u*7d?~=6U!RKk1HL;}xkE0|LCtUe8air$CxHHr zH}D6Y7O&8!@edwiTRnXWvdH(n19pbv;5#_(-0?ie`3&jsiNoHuehIkkJr|g zaXnkcx!_**wM;p*IJs*yaVeg~s|5}wXK;DtD6S^sthXTp$R?<)Bm>AVGGX@pRb8vi zR|#Vbhuz432R`%Xb2==`(KHKMSHm^-wu+QT8$Wn1BvKRO!!LxWK^4K$yyX121H383Qh7?}(Ql(nC4Z4zgeMDb0uh~BQappg{L1wUL$O$q6kMPOk75su{#8}uP;`;Pi@f3c2 z@p>FaMu-8>o5e^f8`utfIs8xJB5VTwA9_Dd7mE<{5bF@{ZhS*dG5!|wuoc_*THrk( zzYN6E{}TK?uuf2}mCy3W{xrp(=l(qRx8PsFjplDOSyy?7tT*2{*gUvfU@K>9gFxQa zt%F+xHxF#B#k8;cZ<2gwf7Xa7M=W3dnj8W3J3A1AQE$&9Kl)<<*|RDjyF51=hTLPb zkTYasZ56VHeZ|h&KG-(6e;~I(%!!;Kf5lz|zyA@CG2{!`!cHN#^^2VEu5Hv`CkJE- ze;AvD?B^3ZG+=YGJ>_(WtvxMZuU4+GX~`71hGLE4eQaT|KfW@tK|a#aR-v=-=d#O< zLHW$aCLcJPh2FznAw$G*YO5IAeC^^n#=0#TL5_^Fg1tgtqo?t)(HrT4)e&v`*_`6< zpmWkkaXIeB`D_ooi*Io)8>D(YJHz}P<3fE$iY*Qnx1dAQo6U2mO=0W1`vCg{Z~?oa zHiYuP_hQ^fSI7UZb)S}@O!X^#FYIHhs5{#PAK@kSZfvV^V&pkoSpOh?^SSVS1UmZyghQa&yL}E@hCDye2h#X2W)*OXB-pDY8=aFu`{u> z#@WVmo#&onT8(F^OJ$?@7TBl0>f#wIL)IiOJ)iozu500I)2W~OfX^*o*~)OO_T;A} zzsZyOSB*1$sbBTK@xYhLZYE3U7REL|ApMBF#2zw7YOk0}qwkKM#x4=JqSNtZvqi+F zE(+*@<}e$C?ZH>ZFJ`Wr|7;98Cm$Dox;(K11Ds;rG@fB!i2w3g@ynAN-wMbLJ}I$v zKB)Ty{7jDw=-PZt{7YgLIFf&dzlR@)eIOsoIuv@oI62OR#>>eOv2iv5j>gp&2IL3+ zCPS_WYzyDX4RS>;GQM9oAP;U6tRLKN3NmB8;8wv+19E0zZeen5*RIR;lIJDI>%2gH z$Op249U|{c4uHI1vSDSwul0g}FN+_mHVC=E{-|7NJbt6B>lBaYyJ~EN-|EtU?@HVV zw6_?&HHzNdD^|jP$lj3qC13B=0a-6bPww+cT4$tR*^P7qGC`gOyOQ6NTrdts@Q>0f z$OmyV@qF<_@innFek{5O_}myJzpSw)7AR)N7H+J?xa8+DMy(~F^Y9t-A=813VRFQH zW`od=jBoz0f^F_I+oW=ajN$j9TbV2LEcQoz4*U%K1Pul_4Cg#5ta z<|nSk$K(SJ2J^VU$L2R4rhDR6oPsyYleWHBXLTG0;TZga<8T|jn0|asKvxEz3*Y5* zS=)1YAWshD{mS>UM%+2_>HJkA_u0O29jUcCeNUgQ@4c6M7xu{~lQ-TxybF_qcTphU zOn#m`J=eQN-PEIy(`TRURRQ|}-+N}yjx%xcqNG`rKV(Sx7q3F?1INnW+7T+dv@?#! z{m$>Tc-;LQ-zacR<7(^zx5v=`_d0 zbj&w?X8N0XXl{zj(Dld+b_Tz-_z52bKEM&;tYiib**U1(z&B*ZTLb%_HxaobfIh&qm(-Aoy+|ub9jsBjtbb zr-_}2^OG%NH}!3?Irz577jl^W#pUTc!&O6 z7CbqSU%oV`e`?FT4MCiPpOU{)Y>f|7E)z8A;KbGGi|e${{J#?YKHpPK_#ebo2r z)qF5GpFF_3@F@PpxqK+#^C-{iH&UP0d#r2sk2-yG z<{YC}dvNKHzMM7LIrg!MM&A&=FZ8ox&T%jLyZhRb6TQ~?{BxBD z__?45kPpTb89~n{CyX;Tq%lXIF#c*E+WHQoZCIHy+KA+fdlh7Y`?s92EZ4J7-Pf3> z)37o4DcQ^91>J{Fl+0jru&eoD`D)EMah3Y1=wRk2|1f#MZ%kf@SBmZPEz|MX8gxH- zWa9383S~{w`J?ix_`~J@vNh!VvNQP6#b(b5*cx(#_(%Q{@SE_V z$So2J_AZqArub#(;BN}Z5$i2h1>y+wb}QNvB7qMI;aa7$H%o+Kn7kFuzS24{x^ZODST~wT6}VRSf>SK>gc}`zY)h- z7Vuw*|FGNm`)2)CcbL1wB!7#KuBcQjes5UiJ5n!%Xg?PB2L9G#RlP*+%!x$3VU|WzQYzeu7aw{(geir;N_-??SAW!Hs#wlHg|A`Ed6KXv3 zE3v5?ZxFX+3y5h}p3tH83Cs`kM7)(>pFANy%vbZud^X=kJr5@o)#1kYFxi2-=%Tm{ zFB*Kae)3W0FLz&&RpIpOO9FcLF9K^mz8`!iu&(9n!PkQGf^&ki0&7S9Bl!1W4a#S- z-sIE4CxVX#9}M0foElia`0n87z`Itw7wffw_hcP0VC7^5{mNm_4M(kuUlW{Rzk9hS zSU(Rf!|@ftD}$E?FAfe44w}Me`wk1N-+y)Bn%4(!4&D~1yEUWMY@Rq>mw8fFWv&XW zzkFYCM&P-vz0?NULYru(F9xfFbA$7PZwB8Deh~a5(Dv+~i--4gv7gvU>>qtq8>qGz z!qVU5Vr3c`OP;ZN$g;{CIt3ggUrKxyHzALk z2J)li2JyRF_aGjUvBj^!WXGZK67?kxH| zz5Xcyp586+{c%$~Depz6|35omYrG*iJ~$)zMDUp@`rZ@v#c2WCf}LSa`&Xw9J*2ufhse$*ic()fhvuCh#@Q{E^xleHKpxe98-+f{? z@t5(3S^xO`KpaQRojqpFrI_l>(FH4^v96URqBNz{bxYknwHVM5#JVV@qUrMY(?Ckpi*#Ke{Kv$u&&|&zN z_)?AM!vl76W6yM>#+KO?;<d!`FE2~kCszH#~*|ct!Fb#I=qHnj6tAJ-qq6@+jm` z@#nCK=KYS8=d z7(6=IKHwkoE-m*5pC3Gh@5)vWv6CAI1xi3fOw$;p{(tD)t~dLk#MSU{z3ml{mODKzCRg&?hzv*r5DU z^bNUFVi?AoIEArjj2gejxp_dJq1VWFGQZ?BiM@3lh(_?)&io598&NgX+xoePsHM5*+>EnfJJp3Eqth?+D0+Q3t1o zlV9Y3bDeWkKz7*rPPRGjT(SbZQ-w@(%sI|~>lDfw%Wga!xv(WJ=&Q;>nA`fCBOJs z*~R>=^rPmBkFu=mGDlfPwy_Chp5%&{iJ9U&PcH+Pgl?2`2YYXs}J5+fDo z*dZWu#@nG+;M4k56ic!13BaVN9cVL^}0^5U%|eG zIfFmn3wSzMcYiS46Uv3-eG~j$bkX*~dqezO*3uR8n*glTt##3SsSv%KX*;4jc)wjiV)7SNRwjDp0@$kc7FO@ZKW6PMchotY2 zInr6+=j}dQ^BjHv_A8mdwxzG?v#zaM-NygG?@XVum93eW53!&5$w3y?2jA;t7T;Ib z6|9GV{Y@Xv>^hU}9D*=N6vbJyJCTd`)~`K}#w%|dwj=JSX1 zKHt1IjP)P7TX|p0@6SI2b3^rF&Y+HFmbv7vdQ_+G)sNr0wmu*X-wnCkcQeT6;#zTy z?P0ykx)=YBc?Y!VFumS;-REr2Gx$2);~VN4n)!@=fu`?Vnx*&nM!V-uKT-}j?|OC1 z(`Ur#IqMT_CE?jm>PJ5t#x?Z5bNy635ATiZe9d+D)pMU!zn<3y(yEWF`*`U(I!l7X^N>s{%dLH4^?XdtZ!ay2w0;?y(2|EVzb??B}b4c(G2>s;yR(xuX z-ubW=ME&FoK%jnIS z*(L9;crV0zj@~Jx2lxf)U%Fs{pf2VO{8aWss~>emXM2xEU8>W+gzJO3$rVAJ+wb5V z9P<>uENklD58eY|BhV4_1HYCw<;${H*Bbl#2Ya*D&3hpBSeW}R8s-gjgn5fKk(Yz{ z3!kcf>>+8t?|aEzFJT9= z33m$Q4tZdFlL0WUE2sSKy1VgbOd6lYYS&AL*G%05x#-A#ZCLWPbpMs3Yh*C_+;=Ig z)7rx`V@J>bzVA+scOOpIX&hZEeh=@zA!FpFGDR6sU0GS~H@Q_=Rhi=m*WK^)m2nM6 zHj$Yi6QS;g>$Gye{wDQL)|27ndZ}#JMqtnLX_9lb>HAI$xmWsLyY&FFur@@MbxTHiCV2Jd}B^O&=OwH9+3GL3w@ELeA86Z50>JC9_d z`HgiL{xr6lLy3YplC^$Dyum&UyJy!f4zAhf|iB}rZw4a$Cp%dv6 zI?{e;`P?|H7v%pP!MtvnV4gR;{-G}H_vLeZCYX0j4)!{zQ+s@`9ePjQUSDlN&szth z|JecdO_+atK4{b3gLbyoPN!@d_{8)JdoirHEf)A%XCLg>qEqxm{nGxwyMr}6eVV`5 zxG?`Pj)vzP#xA|VS8aYz9++p47i0(-L++4IY)keZc}B+7o+A&*P;!<$C(FrT>z`ye z-+cGz)K+JocYeX<=L<2<06V{XgZv)$l?aQO=PR}MrDI+*y#G^PhHZcPrn-x6ntAWd zu#Ks#WHxzSJ6&Cn&+4VVmHJisPBq;-EL-_g>R+Lw&GUw3tUV57to1B7*Dho*U(2v8 zCWHH}EOK^c`8q6P`9R57b^<@lk3zp0;^;TI>mJBq2#@~mHL_ZKWOse5mF@r6QU0uX z$SdfF;bw%x@VZFh&r#HGr!p^scwz`7jT-QpfBkYBtCyZ5og4fa&_Fd5x<`WAA+nqsa%bm22Jw%X zv+$Aa9}Ws%4(2b`XZTONkH|LmehFW`xeWgbyV+cZyyJ7Bf6fnNAU)(=Yj!q$WL@k<{IF9_COz+Uzn5837#wB6d&CwE$PYw zz3RPmd8Kb39=s=_49eo$G5lA`=zB8CO;@WE^}{E5d6;?5qMp09qAj!wzvj__p5f~< zhoNh%x7pXVYw$fTa~OSOqrmz~Ye!qPp zZfOqJk$Qu?$s;{2@5;gkqLb-eeGmVa4%SEY)z{Kp)mg*y8hT2fH-Djn=pW;O-r;L9 zM(nHan!K^oHTljvs#koQ#CWqW%ic;pgw8?e6$??FfACA2gBY*oA^g|YQ|AxnBgV76 z|9sms>l$_@?C&gO0rNY z^v)iE{~hKzWE5S*HYT&odAuJ-FFE>M{N6Y4i4z z9-U{+>Hot0#^ITUr!Rfe#L=_DH2$XHntQ^T1JbpSN4iCR=@;b~)-lRRzbLbLrFu{| z^uDxi1A`>L`OXk=qWZFd(Jpx-+8x%y;pr@&csGER_QHk zl;%);a&%bNabC;bG|q)_9A?-K6<2*1^;-vy( zz&TuZe|TRVn-=cD`EZ!tYkYj*>WxgA8TaV^rfL3(K0AC}yz)WLDT^|xdomFGGTK4e z`E0a7eWBzio1aWY{Y9MvxbNIO^BeOTbB5u$gS}SfJz#^7+2$X$LCAJ<6LyJveby1K z^}6S!(KtJ&@*BhRD(Ayt*+89-)#qi7uFoj1KG%7Qy0Cvs-L#H+PH(73J}Y&u9kh@6 zN!v-g!Ca+l>c)gMb>qX)Z|m&lFYuebsjji_%ed-X##l4v*nxD9BaB6Qi1BNDcCC~y z0)Kev8mjSZTpQzj8^${w}5y3JUVe0YXatMa4w9ne1DF@+OJ?fM|W0l zb{^+9eQKTu=WKQFL+rI?v`H5qfE}F?-}xq+!Ic` zu2-IkyK_`Js6Kw@(jhDB%W)(xU9TRSp-Jx;{oUuQ!>W`09>#U+3v^u98KkRB;;zp0 zyE>6xr_=p*edcsscV9Y=uci0p*|YLlT$3hUt&Wrt)f39A4qA8GNk7sr^a)40!u7VD zwh%^tF&;Wbj04xSk7pe9CHIXx^4^|KN8{2M?AS9F`4>8W5oQ?knaY3T-MFU*PM_8~y9O(o7kG%HhT>o>u_x*OSvb^%19?)Nf zrEi3%Q(QM@$R_f~kqjm0J4creJ>wdgOuh zKd76As}3PYT&s>%m*SI#H07=G#&6|@xZP84)i2c#%I2At_qFs?bxZAPd(XuqU#(N; zu92se?ZPXc_~l2vir@3(apgD}E^PBuouz)nDLna1UpP)4>_rlvIuV}SmM)x=(`~D^ zi}urJ+@}YmDSmlVm-?~3uDv`jJ$ZAKhIZD+jU^aEWV$@ayRxe*58_ScU-{K)V$O=cljYs$4PaNb7c||@L`{Fhhg^`v%MLx(AnM2l- zgJc!mq2DQ^XXKZDA+IX$CPx0yA9RPgh%nNYN3w@r5>MwZ(t+RdtiIa6$qV&TJw@gz zgD~QCUtI1>m&_EuKY26;w3jfR6~P4B!B>7a{%71mP z4BAFmd2^IM*Q6zGamhc#;huDbl^4JDnT8QwI5;J})Jb9KZQ4VRVBXuA@Q2Eqqy9YN9`?6?b*I){OL>T2g;|M;;pTtPbq_TB}4Ud&xuc& z$v*9-?BdlohWmm3#+GDPic>q1|NKgPNW$tX=Jfmo{0!<-{hGh_obY?~L!x8ozPD&b78+eVYu^{?=QB<4^O9c3=Z5gF4Y3`tGE_?`Gbnoz<)Su+_D< zalrp3Z^FwHyOn*dt+bo?JHJz>$|5~=Bs>`<9{N{1dPYB@kJOj`s2t+)yfI;ZTp-6L2Yo{SQFnB#aMD$;l^gOT@A9P%wV!bMo_NS9d6zF^S|4{`IQ>|^ z7Qg=d2u8=ik!=xZ?bu8b~ly`kuU+0S!Cq1Ve z(ojcaqCTUK$v@wPGO&Bqzq+!|jZe~8C12g+_g7!)!gx@AVdYEu%H&V})UkeO%t%vv zYA5yYc{ZuKBD0ON&V`K+bw*A})0j6-^fB#YUagNg_gw8N{i^y`IpvL9()Z<2yNOqy zGH&d#u?I)r)=$VX>FGE6z458;)jQc{E!1=Jp!b3f%-7! zr2e&K<$!T6U1NhT;OkYd%0jMLJFx~Kf65|^y+himeMp1Wn>$xQXc z_e6K7V|_qA$uW5$L&#;i!Wx%xrcH%W7yK*wgmI-$kpq=c+FTjP5aIP}WhF~G-tq(?8%}NwYl=z=dAC@8=b06JTGlyO8&_=_vwGv zwWoPz`@TM`F7-cQ zWaAsw18fy_V|OffaLe8ioH#>%a~>NCcGqkf>A>W`k5J~=MGopZ~-G31YqA?wIi_3WPUtRKj$amsE~ z_xiUq$YX1dWVgIoH?jX=m!K^AhrVsiP#>1AIOWHE@`mlG|B=P|v@*++>tvxe)A!Yr zJW5wsD6_m6PwJd}R!`D2hRHnnR_A1=cwIN9>?M{bVYHRJlWF>-{hs!8k?EdM&)P@Y z;#NQEolLXugH1`6NncwCtIf4j$E$n#hH*kp&|hdv=MP{+!q&h$lNR-B$QHjHcSFj0`Xa$Z2bJ>;-)S>W40&$BZX&XlMP-en9&@?DNpi z$OU_g?TccY)AeM&JskGw&^6?~HX%>chdLzlm0cStzj4BzP$oJ-{oC(jPmaAh+DpH* z)@_a5c+{`#+wtC&J@U%0eYGRqLf81yKJ{N4gZhJV>9_ifwy@X49!Gmw>0|k@$HHC> z{gqr-clHL^gD+h=P7~@cO0rmD4!(N8T!nu<}aQYIl9Z*ff^3yMApS zt+v%pEx$dE<}S*ppO963KV4X5Q4V^*bH)^#i=L=VQU~&=?d4OO=_BGYPK|x>lN)Ra zbwEETi#=8LM%gQ6zm<8F{;5w%M|nI;&TCusWY4I18u?AeE2p~axDk#XrTdi49#7+3 z-3q54>D%&Ly+;PHX~{s(=@Z(341>BQyU0=bQtxCK`DET_|AxKsbS~X1&-8(Dr%u$3 zK0+Vb*JppAy@2Z9-tytS7}BRt$wcL^eJgMFqT2uO9SD0n?SEvyR7bFp{kDI^yH)ld z_WxnC=TLb2EbYUz@3Zf)8gpb9eIae*!v0Y2CVQX3o>TGkeE{L@cNB-~_H^3o>9@V7 z_Mq+@q~m=9>Dp^)ucSPY*UCvIsz2!o({H1ApV+%;_GNmPLAu^uP;cG?@xGiqNyENe zbqf0^TShWp-5Td?7jtjAOTO+WG-DmChQpO)9d?T!dBnVBV@WU<5@BrWRZCc*&;sv6?(z9 zLhX|jmp%XX1^Nb@J?CT!`C{*;{hG>WZ>2JNPtE(#@}e!&4LK|>dkyVvx8Kn3zI)q! zot}d|hGdsK*n8-mgr0lOI|ue)+K>6;Agp}T)sE`QI|=qidS}c2P5Dw*`xDvU(o~=J zdwRD){N9UD*UDz!qjwyvk4$x1@QqP+gvsL`j?`#tNoLpxU(|w%}39GL4d%BR#!>?tG=(qHlamt<|D~xyY zkG&>M`oVZK_Ki&#t7I}6&z@te@F%)XugbHzoqW^%#*cANmb013Xkpn@@XYK6L5Rr}8C#@=KmcPyWe6aVaBPksKux=?MCO45u5_u{nf1sXsb^uJFFK z_qgeX%3`*H{qXkHdq}4{ob|o8?{!(uCW^G0F zXl{a#QEMmv{9w~c^jq9Bswk3OzjAG-6M|~S}`hfV%4?uPrAFk6;{10TJ z@t}T;BjbyVG~ZALc3#WLo?}n5o%s;RI&o-skm1?^+FD(TpKMmQWQuFjS603sa+ZB7 zEi#+DsqesVcDZ^|7i29vQ$472`Rq8M2kBG!^}KPX&rcroef>baibwtL68u&+NLQJ) zt2q!m(QmjXpL|#P1Q}}_sVn2(xiTx4H1$38qOZ~!wVBPYJWFn?e|6)Uz3`jj0>`ojvZT{L)ib#+3Rt$MJrd zK4zYz?#+wTfjnv_z984Nn|T;r&hN(dR#tUp&f$6em9Eupd|}?fYTq^&)28~8c9Fii zpm+3FzC|_#%**(llvNqkkvdU->KM++^tOe1R^R#=IiX(2dGgNQfYSpRN@f^0^%ojD z=(ntBfK0#Nm1R)%!MVSvrv{6OVjKlRlwy<*Txs zzU52Tr`d|ylFl-I%w6;kN6*tOkS}Es7ddW@LFQ|Fb1(Br?Z~I5pOTYgyz$9K(w4@w zwDm9gjE@VfAvsXo4#esvNidBx~4!k&@XhizRZ^QZZiAd zw=;axgHO!7LcQ_d@G4=c-?I}hdx?JF^AfN3fPEK_FN}{$yzCrD?-ctch41J2wubL+c!!uR#OK9F z#vbxr5APTIMu)hr3H*4%^u206Lh;IT-$Awp#g8TqVc}d@deJwnJP+?2dmmeQ)QR}z z!}G%P$(e)FZ^~#6u8iJOrWe_Y`T@JqeQm@pVSi{3`c}VX`_*otGlli6c_-AN_SVkA zYP0HC*F5V;_p&Fotv;hK(f#xR{ix3w3v>k7QS2UL#CX!zJj3pC?i$_VzN2e&Et{8a zV+YYQwTb%Me5&%F{-MX{E(lAvb#CYATJ;>e(K&rYR+x{_vvd#L>fD@y{^f%rFUVW+ znk?b7B9DY~-+aKFjGm_J$aZ>9eEd}OF`Ar z=EHnEp7ji04xL9H3qwA8j=dpW`aoL3iIbfI`S+WSp&QLlpn$wqw8J zOOKJq(xQXJ!=EJI>RkTZ(>}_jjE?$&{-dq^roTF`H)pQjux+dF)MjE2 z2t)7b*YqSiiharUbf52mU%~Iro%L<@o@b0HW6WIBnBf!XGwc!LtiA>FCg;MliP<7e zQ=V%ZvWtZ=COgkF?=ZhH=Q6*bm*^yVmwo6S8=dW49ZJ5CHLlg}rL*a8b}gODr&K$( zzA^X3Yo5Z!HwU91>1q0so@T$w3)#~-k8{@>U+rac0ycuVfON?!d3KZz`NjrjE127p zf8?BaCkAC;I|x_#NA{7O>=|v@u(B!RS(L^u7P%Q%{}>34*H61!p<<)Q%<^18;FyBk)F1wZ$KW~KGITN zdC>;qu6~k+a+<5se{?bb0e^|#Y#eJ{^2`Pl-VxeGoZ5?y)Zh3qY9H#a=0yCr`ZeF1 z_EazOt)Ao;>Q~&_s`+dlwW0dcr#gm=PkmIKvGdfK_|1FSzQ&i|;!rQrqSN)s_CtA5 z-}MRd709!G595Fzf{rBv=wiR=SmVanH|OO;sr>MaF|AFFIdPIB#ubch&(|g=i^vbM zhuomI>*FWu#7$1QE+3AN7te`Ln(!x%%2IPp`rAC0oFl`!rba%}{qDQwp68_3{p?MX z+*Ces2y5<329k%;U<0rz*aPAfXYZ5IYz>gr}$hqnt@}Jx#U*Vi=BvaW7j>3@XmGO>rN5gxjI=%Xa zKC6$Id?l+JH$6i)i_<-Fo-P5MLnf2W(xJmV>*yNYHe;4GZcU;5HD zb>*DTLL2e5XkYDC9jYHeo6>9ANxqa>S)ttOlb)Bh-|C$`B~E#k*Xla`NgvmDJD;SF z^mqMRzeR05ebqCyM{1{uvuWs`!s)|~@~cnM%f^tjI$xEh@y4e0XFM5uwQt2GEn|rt zVf@U{YBTnT7;Eapbz{#QSKP)mIY1s&r;rJ3alb1Ej9dEB81DLx^#XMuF89a_a>z4e z4JtR-`(zPWNUoB3u8}9wbL=_!LU&fCHIJ1)yvptw@ycWE3Tcan z?xWl263{X7E>8KZtRA){-1EHr3ghUvG}H&0SLgDo4D^%o(8tnn&vWXi@q3PbgnY}h z_>~V$OStN4@eGGkZokE$j#_upQ7`Rx`h)bv1_aTx2>y?$3O*8f)7;z(AImE;q-LoQTa zbq?j8Jd#7wB5UYf@`=19AI($6BTh(*?5u1jdn=pCCih(Ptn-OMI^>o7st=G=ev4BW z&xmto$Wc0=G+n>RZMuV=pu_1Jc~zg{cdUG_yqAviqI|U-+^Y7F(dxAxOFl|!E#PDfhCf-zP;PZraGL? zl`r+ta=7mq@o9_N8S(_>ly_mo<9X$y2PcQu3NL@vch$SfB`$eV&t2Pa?po_mIC-Hb z#V=j)$*=tAC)&&N`oQq|NZZFz7=1zC>{xNG4~s_~G=6oU9ppvd5vTfVo|LhDNL{F3 z`S6S*gqNN^I^5UA<-TW{{~4|R>}a`r-lMvZcX-Y-o^@Z|^?&hKcN%xq-^PyoD~J3z z8W+Y(|D4MwoRbOuoC~W?#o;$OL9URuWDvdWTE}5+i zEMx@vR#~Z>{wf!gwPA*3o^#K@^YSUa+WXF3BX{9k-YT2)k;+x~)U!WFc@S^sJ7l`B z?zK$XM&4W#FT_*1PEI%e=3!WNt6O<+&vmk%exX|&)xYblYx#z{RnM(!_uJpaC4J%C zSI6?GE?ie<$^z-hU!Q6Hr~`cg&ZQ@fmO(rXD-Ch?PhCh`erC4$8qRSTad<`=XxqI`$IEVXW$Gm>2zx1#6ZDoyc+F4)J?(VBkf6a^Et*_yGz2C4c zPs<^!_{3Yp)4#^sIMspj4Y&XBZ}ECoobuZ|yVkUuC+Rd^*BW=@(+28pxL>Jf;XDU% zw$H0$_dVDA3F}W<@}jQQrTn?>zBt6y`g6T)C~k2z9%0=Z&YL{8zj$^yO!MM-h)*BZ z=iBe)+ja3aotCe8QLc_t;e~BIYd3jsUCiilGx}Wf-~4v`w4NF-`^{K{@!2si9e=GK z=hA6;9EFjNxZvD%;h_GckCwA#Z8&9724&GU&iyqGN59?Ij@nRL8XKriHfH=5MjZ0% zIdKY4H&xF$7rxi3_vl~epwsDTwg4SmTS9vB0O^T?uB11GsjVgs=~cf;PdH)3Nyj_- z?K=Hg9qbzWqc+HJJmQlkq_2GHQeC&r{qA4uRvoAtf5Pc2`h;`W+-u#5OP|v>YTs*L zZPEVe8Exr$`}J^meN{ZdXn*p!^HuW{{Y@IeIr>fSm@|PM>b#J?@QgW5=T*HgPS4Vp z(vX+-H+gG0^ljs=`I0y1uCXhZGd z8F`kz@Say+?ep40JL^mOrat1DKhKL#-(bVJ*S_X@&)cS+Q8sZm|9-pgS!w8po|RYm z=$Q44KXK@9!bnq}hddY$^5og}Ibr<`m&Z7eo;W(@qj2h67-NHPK>rt?_D}|Y`jr0XPdWW{ zyfuDtsT*mE!_n`~0X*9=pg*WL_32L^@#nXExhMYiXYJm05>9`2Up&$`emYiMcg>%3 zK8*Tkj9qo_zN2yFnsnr?W3A<#F&h6@E*WF)NmriaTNrf%>4{ffTGozt&%pEK40!`} zHJrBm$TK;`uS7mMR)%?&P3_srGxwbLn%})9{>l$&D?^`mtjv;s`Idj>(k{k=aw?-d zimNhGJ}W2P*9M*&F2DONkNl9!^68qo5+2f)7v|{6xcR9bM0Hzqn={ zL^%E=>q6E&tWjCpw|~O1Ya!B*7Qd2bT(?$I`kZwd=kANM*R2~typGm)>{)Q`8DZr^ zc-Pz$rg>=o#4COIS7!NY`Q@$kFudN=@GX1GEbr2je%G72Z^QlzdnEiu_fgopVQ-`T z8r?r)U&raeo=8V~O6}#am&Kk)=fc<wZ(^fIT*@iObO*6?<`tvUU%Sdg>k=>vEHV zqc&5Q>Q+1GtMxrrmpjsJZ0zbdwg}x~ycl2g+1I~sJa;XDEU3(|hGCuE9ysgq*5>UM zw`XscV85RCr|b`~X78T8-`+X!4uJh{-FI*Qx%U-ZbI)Epd*r+e=9>}TXRr_6{`WJ2 zcMt93_bx#H?=9buu;<-A_`VBZFa0&4|9_^V?_l_*hW~%letGW&z`1w2y*J?h0P_xj z?`7EYZ%@7V48+~A(&)D-?BDnOC~aXT&L-|6ff0Jm;Gm^5B^n zU$pqaXnB;?I|)rs9P%Sh?*;UI0QF=qzV}1D2V#G}eg57V^{zno>8odTTfDDfPolbq z-`d6dp}v`6zo+)I?{daY!uO2s{phq+lOv%xV_u^);5!wmdA(!>Qrq@YQg5IAp+=#?0WLF$az2ZKwP{^P7j&!ZKl@@WwD-ctdcUHz?oh zgTEJNd2#gpD(_mUkN!W6WVrv!(Ysl2?i!g*UX!C_B)LhB_WxrhSIJGXpNtodELP`q z2|0awAX6c{d&1KN{=Z0gPm8?wO(Jr<@sa<+$*2E^vj0!Byvw8i_tJB|eIy_BNOelz zN%Jn3en589ePngtS*BCTVSC`|TDp}^b&dWdd+qJBN6kI}kiYf?u;s~N>+sg&+2!UP z<`(9t?A*zL&A_&0!`X|&*05LAdqL#m4}*74$T@P&|5xj~I`Dtd`c2Lm6RuUZ`TmUm z514#2hKwgi?|F7CdH2)1rM@NS-B0g%l6T}A*#__4zLtzLwvF>eg0W8?kQdVikS}Bn zne(qe#*j&5%){Z{aC^8V+#GHVcZGYx-Qlm{FX2z&kKv|pL-<{|I$S>Bs_9ovzc&1S zz%|panSN!sH2gYT7X0?iq#*2dL7cY*X-ezaxs_8^}vj2avcdxv^>^(K_MbXRlbeo@4wwkY*+nMi~N1Bt`uW77!@6dMu z*h$76`Fe98UmpzK;UsG&jF~WIfv{Y7XIL+65w;I|g#!W^cT6}goDjYhP7CLTOT%x& zAH(h8-tbU(JUkH|3w}Q^;NdY3k6!=Rn179VHarvl8HDldo#D^nhCud;pUnPgAfJB_ z$aS*!ut5HQD(oIU61EDP2IW{itP++FONS-G!hsB)J^1r~virZ`{Xg;Q#Q(|uK=6OB zlVxNZd)j-g>{Q=@Bd^%zzMIFs^j`6yfeq;!7yZ9(cg~!^J2icm$GL0d7t9s>E`Il3 zkKU6Z>&QO)jpo?7Rgn<1M_3@ls#adsQIk<96OXuH9zz&A-j=W)#ocmeRsfk zAWw}AGSXP_{=G58ZZw8m>;JD#mKsa!N@Ghn|4)1Wul7FUIq{Iq#-y>y)@1*%Dal(n zC$l?t`~SO(7sJn!+3N;J_LH>aLw@AXcMODQUkNAv+N+M_y!539VO%HU*{_R*1p?c2 z&cHTxB=^CNWfKwzY|7^YJCdw_B#`&y8XZ8ky%>a}AK03%HN3F*2lDp5@KEraIE7`$ zJ`w&A+^0u`_xrIx*NB(BDqn1Bd0!|j8rY|moxU5fa_~I^<<>5~uSyQ~TXMcR+W#Zo zb^XLQNzFZcv%%bj?de?;{mWdZ-w<)F@4T2N@e|PZ{O|nz=2o3w@e}Zkk*nrjd}m~< zzHbg^tkmyke_`ix{EF_eJJ`2;1IDO#>Adr8+>*)l39wuff*8~Fm~ z=KgFnvd7%nyY+t!WE5FNW_c(7(eUs9*V)c5hw)>^kJ^}Q&&9%`;Z0$lKnAiqF%Dov>lpI5_(4IvaS6;J1744O@o~h7W~p z!-oer-zIDsgnQ3SY%*q((YV(hv-X%(!%E@JVW}Wb$|!#ehWP_qTe*~Z?x38n3+iTM z06o}xaL;e|F>7c&x$c?%jT)Q$bZ})2ho89*k=$_j4zPG^7;kyRJNRg-=64up7aL$ zna!+!*PpG=(*^uvAp8C9HU4-0v#t+Vhp3;~n6l1AAMm^JGqbhX+58v$s{Eh)cXR`P z9^ZxWYV7h2_MN=?Ech?^A^9GR{l3rMcg_3WJ-f*}`QDc&E69)P74|x+TYOulx`j+< ztCK_31D!9MJDW$Z6xe)Z*ur7{FzhoRlZPSy_zxz8&ezFBb|pKV46ZC@y9?);J}2yO zIQ|9pJXu_MOx9MGyGAyX*_GksI2lb=KM>p_r-g%aGN1f*kG`Nc$b7P#%;($i#}CoG zyVkVLHRvAsZ5h-HpNo1T|M^+?DcI6<6&)rFpM`M3v_6I9dr@cN9gf>|_4E2rKaTV2 zMD_2yzvq2*-qCoQjw9$Jb0<2fNA0?7;5*?P;U}S+=wf|>{-(!$qlbQHbNEiT{>2}{ zmf;IKH28J`{MJX=@x%UbbIba;`Tf~P-aRoU*iC;6lLJ4BwQ&Bw=YlcDUuf)Eck4P> z*TRg`B?FtzSf-bZaqoB0Pw2bL-dST)(ou9(ZOgvTMplp+5Dvo8XYjkeP&TRO#l>zl zcW3+Z^?1LWu45zfu{!dlH8Fqh$1;FsVFAusyO<-rJ^ZN4nhrkZuv#H;PE~0;Y6Nlc>mi#u**1nCSt=rz#H_VOnkNRz_r_ep- zEBrX-ulk;TC!GF8PYGk*!dFhum~&Wfc8%WhEZ_Tz!5XQ#i1~*#B7Sh=i~Yy;WA~{{ zeo1TE^wiyf?~+fPe&N?NH=%FXm)-}vCHyteU3}`sGhM>w=L2BN^4qXqjelcZeAdl; z$AmsIAE0CEzxFJDkof_>D*IMAvV`t3|1p;l-duoOB9qKX=q)z#mjmCFJd-#4?tD?c zpDaGQuyVk*?{P1+t{jXbmVt*zJ+w+FJ4{Pdh}!&sY;hU?l` z`)gk^($R17R6k%(oEDVBI*K(xHjFu~d9SqWbKoOX=^(<#(-F)4A-2C93ck!J)5X`Hd8&EyMhBpr* zb3M-&er<3hYsp`B{hfhdoUC=9-`-j%{}O+?H2KLL>u2}7-*RLBTMxG8$>#(%nEN$i)ZvdebV^S|Hutvh`-*L zg7HK)8TsA);r^##I*CM~k{`&f>3&?vuBkKt4OTYPs=ohvsTY0Hq zK2m=ZA2S~kJC(iKxeDLW;(>k3XYITC*88j{nA4cguzA_xY;E%kYXs&34}?d<^T9pq zqvByVv$3aT{$kCNFP<-*Ki?XHxrRA~HBj@52LoUJZ^JLcPr?twDdF_+^T1y(jJXG2 zlsS&)?+?}=tT9|1E(nfS1b+NG!sH+y@@>xID1F!e8ucvg^2VV4wSzSR?Pcyze?MD6`|1bg z6>I>yQ-9Nk=}KdR>?9+NLoyHSZ}QSuG=7a~vd_98`od1%7|O?+JYQw*}#>58M9}JHM>jc&Y?hEcqV^X*}I9?N+UluM7=LKuG)@#oS-woCT4i5VbunybbA0PAa zQ6JN0V>TOI8+g~4ca7G`o8vd)&zd8!v)S0zoZ0thhv_TyDnBK8%qAvV3*F5oBLCP6>;`sjZ9BdgHVj+GHMUl5S!R9*_~vpudp7t-_%P}x;>&Q2UxX}i&vU-*OJ18>IPy!7`{XrQUR}Vy z;Tg}9$^1#r26AfHuR|u1kNo^(t8?Bm&-^d$ zk$3bHJwfitr=#_Y`dD037P`k=##+d_LH+Tsm>;QIb*^6d<*eCQBd@- zBl?xjW(Tv4>1jG0)&#ni&+Zk5K4(wckILRtG8RgXY1;DcjCiwr7JZB`*&T zv72Ff%%R!t61o237`|#edL0BWK9F`7m2paD3KsNt6kfGlUhlYKE^3elUM9^vTEmGEtBqISAx7^KlcFt#x$c`nO$vLu=Ol5mmKebkAoEdM% zUH7Z-cd{kfE!OC|myG?`{bTOA?mjtH-#8h-UL^a-IWmqsVJERM$sBeVTak^%eq?7| z8GaQm3~W8N;ZMU^;neV*@XheGa8x)nd?D-`J{|T7yN2z5S9uHhPlJ6 zAybiQFNJ@HC&DA)fpCAgFWejM33m_h+dX0a5rls{{4+Q{8eI3xU&Afo#&CVOF5D1Y zyDNx8n$L#+gcrlhAtHR*`03-P&lTiVo>vUYux{8iygzIcwhKFjJ;L7M^Wl(ibofS4 zU*8XG?Q+`m(?;u4z5XKnCaC|~w)6m7mM)+x%mdi<+J_x(KhGsWJF({>o{NHZJ1?9Y z#Q(G4x9h@O8pI(k_O<;s^py1jy2HGMoTrQR-}{1b$*;`UX#ULh+90s=?A>Eyj?LfO z!!F_L;j-{Rm@qP77j4Uy-V3;SoE^v&zH1?&jPlfx#E#ZprlW9Sp`9|n;Ha33< zJ!Q?Iz7%`N^)c)4!t!_Xf%hBuYGzbTN(C1H#^6x3E*#F*xoJwhiwK8-&%u3SpVBL|7y&7~T{d7Y-A{vSIn~ z=CD*)D9jh;2(t~CecJ5P<_zhs}=@L>3RxINq&9PbMEheyJb;om{{7lUU<>(uzxrun7fCEi;c`1+<#*16Ju`=zX=zFABVHT8R3*bRF^OqlWglcn|UZ(wz916 zgmhkO9&7$>9&V1$uh%s>&$6BQ1LO}|g<<*?G3ql`7QO@ig6<)aZ}ujC1)oGQKadae z1Mv_0VgH;x+xCD_VjO0 zKRz5AjtP#(g|7$Cvz>)y^IjdU3DRQw{v=ednU8!jY#TNVZ1zINm<~}C}`|Q;N-@G}U^-%LjYc1x% z=HTSB^?zf1t?;REYWPF=cUW-jf@9YXWEfupc}Mo06Ugfyg&zjj4-cOVTL-df*+4eE zDJ&Eg3w$APeSt7gbhc!}bitJ4GF#Gq2 zU_H>0T|8sO@N1*z{{KS0Xlt?d7qU6+-SHk5e-B@^XV}&D@$h4#_V)1q-9)E&-^kh} z8<*YPHA(iZJcxg{z=pNIWBFh%NV!{QwkG)W5PW!k+auupc58)qg&PB33tPq7H$Mnp zGM}OS6MTtu30ce+X>T&WC)s#sFwV&g^5iRl4RmlgDC{5h3A+SilE1;Y=W|#mtQFP> z?+kAXl>_GV3kPF+wm=r}Lp&WOhd+l&;iB-PaBBEo_-^=iI60gUjt*ZA2ZsFv8~30A z`%T+#+9$)VfsMO;_-NQ=z|PZVu*bALrtK3v`^6xhPlS(ztp|AS)8WYQ{cvvhbzqDC zCdljW!ym)#fq&zU;CF3ycDmnx4`f~aCCnfkM@5l&{2Hi z^cKI?YGJ#;-q4p$4%QgJw`^UQfA6-SPg>6=gLet`Oz|a~|9&Bu+dmh`OSUn)^7L>| zVAJj#P6@Y!>0y;|tBl({d_UYAUI=rJpL6`YVO)56+|%Q33s;AW!g=As@T+h|m=vxL ze-3{O4}?d<T97Kgaktg~h@mVV*F3-1Kpegg=BIhap|EIJIxHWW4yX0wBae^V7Je5l3+IK?!|{Pn=#zneXs@u(0Kc~jTZDIq&BE4U z`|#ngY2X9m<9TfCV`Kjmt_&9jI`jMCD}nESi?B+d%jiG4?8(5t^NT<~d?tK2kn{BO zI|4rn-}`*wwbmYc{gqi>8TCQ%OV~f6pOMRa0c>tV0_n4M zm@hmw^|7f}h7-dl!zN*=@amLTr`#7Ng|oxg!hvDW@WHTtcuSZgJo(C#uiOxR5{?U> z4cmsb!^AKyJpJ<1FW(dX6s`{Ehwp|%!(QP7VZE?Qm>A{`gupXla`Bg1#XdEvS+Iq+HVzpNfU82Bpaz@G5I5@w%u_E{GRD}*(}#$ofYMc62;8Ws(V#78GQI^mXZ zRk%1@7%mCdhTDU9{~N~5I&Rjv!h&I`@YeA5uzGk`STDRMY#Tln4hTnv%pUw%41YP$V^4(p!=J-d;hgZz@P)8j*eYxgRu3zMrNaEdWQY&yFX5Nr z2f^44$pq0IyMwvP#W z1wLYHrV9sau34rV-9yWk_RYJ$g+GOBf;C3|8{>u@&8Om>M|LwkVGkFd0bA|Na6(`& zv*)%9Y+-he#{!wSRUjW%4axNAvv6LxI9w5gyETa8-(i+1vrL&QNNb6(Y@jPw39E&*!^YwLL0NVQ%A)+r zU6fto8@dbl858LkVrgge4r;odMgJQAJ`FNU#G$4;Fm zEE(1aTZhjFdf@87M$qWsC6Xq-CH#T99gil3&<@NnRKI{SEn>jm>AHNUc zd^GTbj~^L7GGE~Pep^^SY#DY4djz&GoV&)>-6UA=v>v*2SU8A-%{y)Ew6XsV4~M(L zP2s9=abW+R7QP+65!8|Ld^*?@LvGoZqHdLiFNi;z{F)Sg64=D%NcCBKRr3Iam|4*2Y)intPrR z-+96Dtnlr?Uvg0J&cFRFJA|!+{plNocLnyuTf*YO-jcDwUJ~zf@sHc*as9wA?)whj z6|nE0Z~R++II+**<)r=F;{^ zY!%iDZwZTpc>}wNU$VS@`s+>YM)yd(GVPU7KP8#{WOy|2Q~WJhpR@jE&FzA4c3|g| z>Fl%P!ZE=b+@XPg;Li4W3e}$U+vPsK6oUI&%P;r-ukoud)<2u#;WmqMBqc^^Ib1k3nVMZ#1{kkvR3#= zFfad3_*LNlcr>uFUO(>jxrQn=>*K2$i?+P1)%>p}+-{C#MHTRbcbB0$&UKx2TNcYwt&uqZk z!{5V0;prgUtTC$Pdx`Ltz`wCp;N#dlY!yBfwhtc-JBFQtqiY`s^1%Ng?|de6g_vKX zWmH$n%;#}KV2551P77ZRp9_2&>xSjRJb@FEJbxtI7w9N^Rp^mRgZ(Ne2m4j_4D=4Y zL?6x*#ss>Q9Dgi4815PD1Gk6Vo^k7xmxT+0y*u9zCxoMey)0h}hs?wm^L;PAeg2V; z1$$fQaNpM2c)+IFtN6a~!SIo=WB6#;HVDW5wU2TB;LkfdlY>1RY~iZ{y+O9~S05hs z3EPE@!YX0Wz&3v6)fxYH?*rMF$G?1QxH;Su_*3lLpEP(^i##_*$Zz|Ot!H}Q!@8mU zwD!YU`!fFQgEL+)3GC*thR=o%1+seSKt_-i{|fx3Y#=g(zqOFbe5{V$KXGIBpiK(D z2tN+w5_!x=yk{VX-y7Bs?+nWZ`xK($r#(CM*{S!0KZIX}ABFFRuZN?;m&1X9t@i1# zclcP?HJI1$5VjBI`PIcIIhS+B6Hlx>s^_)0%344T(2l0O@u@&Otc7h<$_deKmY5oDujWE)Bm8<{Nhe_48n0 zqdpT}2>v((?AfsQgYCLVSUfBd77Ggna~XR(W)12)uC-BL2Y&~}z;HFk5oUp~WH1-{ zpRjSD*ZAsp5A@pp!CdOAf!?8`=w@@Rhk|tm{b<>+PM~M_*}oU)2mX;2v#MGLk8%0{cINlOx5A0xo59g<*G~;+26nMEOvmo& zsC|0e*yBcRRCelD1DnU9Kkz}5r+f!|(qyV<%l6qP zNVdAaRS@>Q!M>m3ntQ^0C~Oyo<8dx-`PppnUXJz56~fZNp2B$pTi*s9^SA#7{qNy$ zXShCG63z_P55E{b9(on_cK9`cU3z99f3 zX4*E>whHeLY;Lx;BR$~S)`6dO+d#hX%UDZfx9=6mnooz%hXcaF;fTQAKRKKh>?5Od z*t=aHl+W9SxOLjCqwAfwWM4EH1?LTSN7fPPG5YMDa9?;Zn1k3i@qBnWSVNsHyfG{m zmJe)cGR~f9z9+J7$FN)AkJ>v}zqNOquH83~g$IQ%g~P&;fsdgaG5v_?hX-wDy+c_0 zFu+!~KCDgc9Wm#k%g+z`gncab)YunkuZ=a7;k`1}UhJn|H`q6{bMTJR5ka3kJMh0* z(`J863Eo+^ztjFt>pu3uP8c&`R6ZY+ci=7x4+s12)(!TpkuT)pFT!=f{QdD@uiR`S zvyCho_$Jp0_C>SD`30<#u^smf`-cOW83nRXB!x_OG;e>E(IC_9Qd?S22oDzgPJN$TnbL)TK z5Ayo8AnykUW!pRO)v}S_A2teW1nYf^hk1iPvi;G($8%$#D}El-?eT%FY`>qqe(w(N z2#W`M_}I$-2)6|LOU@1K;ja$fWBgS1gV+N?M)H5KW64Rfn|yZe8o5i(vVDamo5@yo zGMtmWmDj?1p8j4Vye-({y;QK)XdlU&1~{kdUANDh4;jvdnIqV1LMGoEZVcoxzY070 zsBmELj;ilaZ5cKUtA=I6{K3TSxqR1VaxjPFzhgI_7ETOb35N{!=~=I_SMQ@?n?N3~ z6O0+_`quKthX3UKVSCZ;3AYaRk4?(AM=uTDMK~v%8oH;$*tKu#qk){Thr{~OieZ_s zcvvXBA+Vju8GEw~-WS3@1KGsR<{L22|3wG+Q-2+< z3cnAxguBC|!QNoDCOdQCuv~asU}y5L@T0IdHyOYm!#{7%uy(MAn4jMK;q76SuyR-_ zIEw2X!T#!Z1@+;1dglXS*oHMPIV5}~93SWt{uMqHz8U)Mtiac#uFYxansWo6&BZ}G zuzSsWeiyC}+Q%Lg*i$~UuAyh{4eVroE^Yg2;1ioaEE)8P{|R*SdxQDeEffU1~@-6 z{3x)^&kMr-DqK3mCE3G$aX3E+rwxA)wC8bycYo+~z7Xxr9`~(8Hut*&na+l1%fB=Eo!;rrestQGjb`BT{9^M==lS;9-xUmCTq>}`HHu({ah{Nn7etHMQreMS$k_1J=J z#0|n*16wlcZMuDVbzsx}ERaKdzvkKOZT@3&|Leg%7q+k?$T9x#ZwB(rGvwJ> z!5p1m_>yo{Ag}m`*~aW%ex64HUoqd$e*&AAY~rKmtDh$<5Ecm&1HJX;pd4h@%3+n@ zxKeQ4oI;r80)GIU!*w>aa9ANU9`g~lH66(}rM~R>**feH%wf!B*v3Z$@=zUrH=Gj4 z#WTWL;p||ZWFPg{3|a9vYdTyjgl|L+9g@ecd~t8^OJM@?PNJQ__biYkbES|$x6Nve%|wgHN?xq zHNmQGcxL#&z7h@({MX&jO-5@=`k9QTtJe(V za&2z*^?ZS!gOagU;@f_F)c*_*h5G`3hknBb_}$>$LALh!fxUf7I4O|RhlT^fK7swf zA7Wi{?ZB74RNx2EUkQhJva{dCo1pWv1ApZzkksWC-Gr3JBlg;EdIqgV}+h<@+ z?+4+uK;E-Atqqd@=LK?`J|MfhPDqBcJAWJ4on*N@urKKm{u=W9@xaccU!DnMyCc7b zybHtE!6u~#?5(ypTzr2B%5-gzmNfZUE)SOt-j8Gh%h#`hbxipj#&vo2th{>1Qa+_6 zUHXQ9g(I4tVF%m0^QN#^SSrvH^a7or4fw0p4eVn&p?;U0hTc=>YdIkBpU@BX=kbTp z5#I{c6MScMg!!5^Y<_~ik;1!w#rxIVxa#@uZKe#1Qi zAdvg?JeTKbe;BI$8(6pN6Dn<2MBI$-0q z3H!c0xGiRDT=x=oCrNNg2Hu}VXoi+~G>ca!}`uzdjT6|}$&`S0Q_zXJ- z&kc;Fe+&L4kc*`C8Dkp<;zRY_8wP4k9vVC-Fb>)HHG{hcp3l5(Kx?@nzPX$Z-J5UD ze>p#pyA*qVDfnz~OmIXXCN<{Seg2T#|Goj+@A-<_eshFQZoWJvVDH7YY`td@#k%~m zRReiX^N=66Ecij7ZsGL6yf)9xdo>U54cG>nDj!`fA+21jy=TCdjCjRgUoofSvM>Bk zHpYH9gWmp*pmv5n%hnWPcITtfJ14tjKl@kfb#C0_{Pbei;GEz(_`URGdVu;3I`gi< zP62(1w*Tw^Z#^@x6mwY5ccxF%6Lt&m+lzwg)Q;_Z&hI+t&AmT5^=L=klV@f=9vl~p zel;yZd*fGr^(Da-0nO}J!8L*Y@$5?^uJqQ5cYELbZ!7-qv0FvE5NE9x;BNL)-{Nv{ z7(1#S?}33j5-}Q^`lx^}E`P!g@P2h;L>$LAkXv~|K-Z&x8%xI2F0=Q5)lU}>)<@@y z95Q=nDnI>0!N&r=`fQKH|FrcAcLo);}d+qdZF}XG*JA zL&0BV-`O_x6?BMoX3s{l@sAGJMtX(k#MyZo24A0@-y`7f(>nGKJd48SzAd0l^QqZk z`BF=;?i~|!yZ&Zh$B>J4Og7v(i!XA(P32z$e8i47-@9(=M%eGw13vnV zd8xtgf}aIq8`{b_vz(@O(q6QXcLi*=+CW^nYe4s=Yi|+IwDrS%gL?;_Z`Viu#9RF8 zD+50D4}$Z8GXu8zvw@nyCBX*+&m|rl7zbiFKJ_aCel;7tTd?Eo-Fxbt*y*hTv7fQj z+~cCz|M$p8Fy7LwbAGp&@prR#OI?+BM_nFxcHa9{4(!ld&@<4oZ zzQ~fmxTlxDEjS=}L*Ti1HCu9+I|k1Qo))O-dVIjgF(=m#)(iMRs|Vs}zW8qgzBnI> zcBwu^9*YkqW*5t|rR zPiqJMCeSxVs9Cgr+|KXb?bKREx`=m*@Lf&T8}oO4@W<@B!F;wq9=2 z*b@_q4dvCJ8+hNNXK{^Jeg*$R4xZ2PuHXZ~VZqUXasS2OjNsdWxb@0_UlVC~?m1yT zkU992V5h*mm5-ujn&V##)PC{V_~gG0#NKkJ>JHcrHDhA+O#=BgTI-GhKT%B@J47Q` z63}efayILN;PQ#quEv~?e*ZuoP;Cp_{j{JOjc4lV?Ry8KMssMMC8W>rr}^3ZZvOXa zf%*|S+;ajN%6EeA1$1~dCKm=j3R;&=xA~ty-jW7W_zr$s9{f14Ki%i7;FN$a!=I-Y z;e!JLcHeU;I|OXKJm+HrS`5CT!+7SAf9^d4_>FH)cj1@Q!H*8)KE*uy1@ai`RkStj zw-Ol7YFP( zP5apaU7J?I*XCz$6tLrJTGtEIka!OLcEK$J&e3(juLEO{Z(%(0sm}{&DSVGpf|CRO zi1Ew6J~E)I7~96Uv2M&8`{KM^+3(i{`v!72{BynuABC^NXZcLPhL3*vWqD`AZvua-=RKYLCix|?FF)&*!P^4Q zWqdC9dT?PN{{C}tn>;1W7Z<0?jXo}r7gHZWTk(DkwuOzUzH)rt+40rj+rhHnx53Tw zM^m9Q@Ux#BJU{UK%Dw^H_pv~&2fh02;0FP_NeicWtEax@Dz}`g30z~9HRkF@?vwWf zJ}~e+%?5#I03REudEa#Qe!|)FEm$>> zFI9uMERZL~9dsJpLR+V+)70_N;^5)I+5!KZ?ba?hjb{FhfF>m;K_~Z|tXzeh#ZH0t zrggK|4-D1{?h@1w_kZiC8To0TZbTjS=!btMZ|xIv(MZ_rQPY0u?El!@A@-WC&0dSy zo)}ac!TU6kM+7vH^#k#pniIOnT7i7!8o`}|+XuG}yl{|~cTFItV?46ud~mVjnE~DJ zq}j7Nba6h(2Ltx|pkVL7_*XwGZsl{c-QHit2cj>s+iL`TD07vMCEhm2&2@UEm|Scw zMpskFHpq*qD-_e$_dY1^`lJD{J-Y||NIoUI?VW9G7{8O9WAoU)hX$kX&8D)KY$(54 zjSJs-m4E|SpSk`rKe-z9RX47{m2(w&SI%kH|C8GED+XLX_g2))=QQPId2WcFeetYr z{erwFRgE8QTfUS2{M{LRx1Mfp`%8l5f#d%ya1H!+*TtXbr_=rU>aO`O0iWGH@c&Ip zv4EI>mhC>p5boRkYX^K&O^8qT?(y*Ut)Y$Ycb}HeYX750pCV!YM#jT)5ZS~+-jv; z&E-TNlK+cri{LqdIm{O~zs210S^RMJK<F?2kiVKg9itAEEa24T7H!P6#|#b7;Vp3tp~asY#$g$+s^*KgD2(x6Ee2=>+C+8&(`zR*>~EvhO74d z%h_}7be8`P=qqE+^Mw4rNVIQb{Otifh305Zh*SCFG!%N}qXRyZcL|Ao#lmV3%teNW zzxJCzyv?SY+vYlZ&gWy#<+lzC_=fyMKI7g2JHjvLU+xm{$vwMSJ%ue|)A`kWXg)C= zf?ZP|9@_pZJIFrLG}ufzMfR8-y)=-IWLrHmWXU#LFaO7ev$bryedPn$-0CCMORi4- z()#(}H`}=e_L)s)i`iiuL$6_f`ReY0kB*;4AAPjB^~bdVeydMTYhr_G?BW+ZNOz~9 z<3)P>O9L7@uB27TCF~!(CE%NXFrcZZML98GqvaIk6WL+?M?3j>fU8@7!XFoRsY3z2 zxxV+uM=$6oVlHxwG-z4`O@u#7|4^So!%%0Ue$ewew+r~eYXxj6o65HGySE6~-KPdm3!WJ~JJ>FG z?hHP|a{|9jVV~j|t2|?sZ3D+rJI~fT$72G%`8vU!gPR5HI9=u&v*-RloM(<`>a=@W zzI&yY(^2S0{^U5X3Vs~OZ=MmzZ5|WIE2vk&iLVHD4*25QnvS(@pw>hoa+Q3B$gFc?PM%V2%1;7*1zknn-PmNeX)NaiX9s5m;>oWD zV$074#|DQ7azpeL^WkNIxub?$Y|Dm=fB8)2q&cdtj4rx77&VkHrRIPSC#QB~z_(MA z!frQ5DKAI6rKi+x(@^;0>g8x5Y#qIVhD?{HHS?tTxaSAw1g8b+Gt@JyYgXTUY@mMS z6T#7erFDK++wp}!{fAl*`=1f06Hzas{#oq^_zceDw{HYr57Zu>8CcrSdfTa$=8NMA z+~Hc;b$)lc>P9|wZ8;vIx$vXK8MWthAbdp^q7P{cnv3`Z=tr~{zO`p|s=0_|o)NI| z@}xTi+I2pye4peY-W2E$zBilx(SZMrXX#&TyFBxkgRcba{8xk10=&#GrNGX_dh1MZ1Hb&20sU?C$?4@Zx@s2$-r1HL-l<*mWH12wfr1)mAb4fAGMz(1#B(l*US^VIw{@98UQ z&1kA@#VuC8#he`%%h!&J?>7s!3aYWt#$OkxtvDj^Jm~p>oTgmny1{0FniTm-dB>vz zdN!^3lHlrq4}PmvZZ-F;4c~X&!1Fi!Z~ie~o1NW0*fsFpzE=nP1)l%$T+fGs!-AuN zPX@;aCj_4gK0Slah2Ly*Z19QT$iQ<R(DnG;KMy?LAm0f-TQ|bGD+BpZc3&R!n*qP%q=3)DhcUj5 zd2>MgDxOsrzDY3FotSs*Iz5zL%2%VM%6-vVl?pHsdozR~)4cK@*$iDO8pBnJtwJB}A^)Pax+L{l~ z&hy{-?do3GcwGCA-~++oGoKwV>$_6}{m53=PSaE9X|xo&x}5f@U4CRRYU6x!+PHWS*l0E8#v<*n^(%6l#;Ke<8+}kv+q@*2_$LE-q0i0! z-!gXEc$edp_ZOd<1D>G}x0)Y+48*qbQkw^2W%JiuX9wsh?DjeV{e*VnnZSP!)J~{N z;U`+YYzCicgL-KWe@mu9x-46)Vbi6q5T#Z_78j74DpO$ZX$6(EX4$a@D zO|#wn;7tPB$yNa$`B}kEf#)t?6&w(}EBI(|OmI^0)!@wF?BLwsn}Ovwg0BUq2d4xl z2cHetaW;HO@ZsQnf#*E&!(PD)0=~NIz$<)m@dEoz&sPIUQ<3AW{oW>ZiFBk5gY^Sk zMqA;VYnK~VzF|&xTAuntHe6nTMoyn%zxm@&3LYNNQskQHUHVBLL*7zO1=j{w2cGXS4*BP_ay|unO*7}K)6J``7|(K_M+YAaYP;#==7?GV z`lR`#UeKIVQ$|Y>FN>vVDf}xrG4}fGfbaIHfUSO4ux}t2$9C|!X|?>$`vvS2AC+BZ zkJ(~=EL+I_(wo^_ww3+-O7QvMGr=*zQNc%p4+Lt)-x};6ygqnU@bciL!Ha|cn8D|l z1b%yE@Y>)F!9Kx$!GXb>f;R^TO>l7Do8=sb1n&&~Yp6$Ga^sRY|6fd?rk&sKdR;d^ zp1l`GTokxJ8V=r}!?<_6grnGXHk}<`GpOG#PbyB)cI-EPH*l~3${BJo=xpICtfBQkf!-LkP@EMH1G2dBR&gW*!=95BMtb{Omm6h1Q}bQM^ijG>7;}v`RWT|4Iz~=HRG+J)o&v8_-kk9;k0q z=f($T+tnS)cdA$3B2cfgZLmwgp7SBuqO$|~!K$ene`xTG;MKtg19c^GgX%!|vbV{* zOuW}*?cn|aJO1#1b|UAvX+Zy`Z_`Bh;jakT@An3LZqN9CCs-EzI`~U)^Za>d*X@GU zf?EeG2Y*`SPjlN`k#}WX6xff=|8j6*z%S>^bbdLVg$;jIz~`oYt4G;7Fdozuvfm(9 zD;}RZ6*hgd;Bmo5L%qtPS>53s^X|b_gKFUX1~y#H3Vmhjlk>}sNxI(^!SX>${*UevkxE$EhK#YD=piYH8 zORJ@Y^95^p)EBZ_WN+vFaZDWxn$W z_SAa5HciB~d~rTCu5b+gx_6fF-TByfgfB0zDR;>R^YLlxY;kpU+~YZ)>gu#|+B#mk zCip`@zo)Obmp*@k$N1XS%GqeS$?7R=G(TJ0KxgH@WeFga6g<2Qw zFCRfmXV=x()6m}*yfQ2xYgVe`+D}`QGxmEnQ3)r#{}j*d!SxL{Ju`Wj)dMzut$^K>cciWC5bPNo9I&_S@7IG1gDZn;gX;s%R48_p8w2+Hn&7J7$H8|4 zzWJ8|dJBL2@Y#D;-k*1~&{W?Hka` zcMi4RbDR@cn%m8PqQ@L9pUQ(Z;o{;^P*TlVS zH?3TbQ@x5j=h=ZgC%>FO{-HoW)5r0*9OaV&c3CaT`oaAJwW;)O-yat6z1eNg^FVF5 zZH=Ql1Z+3Gj%^nIvBP4*(}J%A#-^q3#wHt0Bd3?M+ZP6MIX?5t#geB7#`Q4)8~(O{ zo?=e$b;Pl14r!2LUw#w6T)v8)$*+>UQK&Y45n-=~6I{gI~&|TzDcM8~e_FaxtZBg}fTG*jOEejj3CW&@VZ~12M{eWF(yZPl;2Y(3s ziQ_z9VEn8#t2^}U!3II=Q|R1u6m^C)lxe<`&8B-Bqc0Em8^-K@!NGy&_1SN}_Hlun zCVMSMWG?Vy_%yW0YXWoXPr>yA_~L+gSIVE**Bmq##mIavdZ_vZ8mjq!exQc@wt1dk z{ULkKhCe6ZgTE;F&w$QCW0CJvYvH*dwn!a0yG}3ve!%xsv+O;2I(=`Gb* zmge1CeDTi(d~w?!8nEMU4cPG42YBN@0~*Wo0=68du;(-&JVP78Jq+8mc{jl2ff`D+ z?BbA51ohQlo%rO1vv(nGndgW+|FmWxhM}8Z5zt%YjQQ#B4dg)g2zCs%3APM23i=!; zzg?}Ue5g8dcAu8cho`6GXm%c7(_cOj(9GF)&jQg|_~~+-&2R3U94FmH?o%H0#R2Yz zZfo4gf2wJv!QcgaaL>RPll%0iR;6RoSoA%v&)@vb1sf}^Jg3K4e_cRd=d&B%-wD`u ze#_y3Ik8WmE`iR`6Z=CMPb^AmGf&QHu~FxIoV z?7!{c(-Zj3cJ@0epp_gx!1p5qzuVS1PYL+@t@xRqg@hf%>MjOumrlYXq{P2APvE+dPo2~|nmcmD2*X8u( zf!KBWBE9j-KupVKv)5{R?;4m_>cH4;OLf22Sz5O!;Aio})Q|Dk<-5#r{v7?4-^X9S zN1oSbL!KV6=dTam8hDnRPJUP*=c%5KE%#iDx(v^qtI=2*uv6-&{~X*o?`)9^q(8hg zczfWv`LhCbA>Nm=a*8!>AE-m5!>k*y-2xx$0Kv>Te9*6|mXt z_Zh*rg9`$_`Xzzo4+DNX+fTc}2TKBai&%ht=cBXja<5Mh)HKq8HV!rjT!U*`Be-pV zlW^4Xfd4L6;Qcb*3-I26#;$I0r{Jl9y2bkiboJ|3zJAWW%RhWB(9ZZ0k9xj#tKgBr zdcj=+n!389-vn0$%L3ZE+~_$${q@@OgHzW+b78~vJx$zrFm`Sp{64u$aowfC4`-S- z`^vYrl#h|0{a9cuza?Okje9yGe@6WlZE=KoV!fJN=r-&) z-JC5K<8B(T<+K%ACma6zya&--7ITZi`DvdEjtLGA)JD8JU@zVr>=)2k*p}8Eiv8)i za&z)^n+Nn@c5JJssuLfGsQ0fo*KdUfYkajWfXK(dVX#;|2EJeg(ep8(rV_v>w;unp_({oLv`p&|uhd zdbzqqa9{GI*8e`R4Y2!`>Ui9@ed@!nnj9+sTx^2}qoT!}1d4W1}_Ws8Kjootxv~&4Sj)q)0U!85g zcVJxgnSMU|Lj#(Nd=1@Y%!wM0v=((L{B?eZx+}Ka*f$408K_4&IXEStCz>bVbH6W* zjc+__K8a!Fo7j4O(&_;_PB*84dInqGix2kFK%E*ty)f6+9Q%!b#xLWay*v=N)7(8H zdR%a3K(}R2Xan8@qR!>T!T!Mq0{-TS!IuL5Cwwtr@8wNDJ%i6D2Xq^CjeJ(N?M``r z$@+m>zg>g91Agxb!P0SH0h@o1;0}S>9eR^| zFb#@tkJmmQ)VDu4^_#B?;0fX_fj z=X3uy;Bzbs=rQa)9gwy{n=w}TCUh1YU<@1Q{FdW_BLcZ0nhPG-GhpA@bK7bAXd>~KCU;ldW?EtSY5BTad7IwSZIE`h^VBO&U0h>+3tKH_G8%K{0 z#DT^a&D@xyyYMaeA;zWg`Idm~{%CM?a6)imz^`G$;R`ePEWb&Uq?e0DX_e-J`N2oJ zJorVxzjEO;bF~KMn0l561=U8?D9}(h3$_U8FY;mXV(h(mdb>b-dmY1lI%vPdPs|9G;8WAlNie z+kqdR6YLP|8oV&rJ*f6W_o3szF*q=gQ~Y4?@c@sU9*7Z^&7MUigx>yUx1Y0x54_m} z=W0(D=lM)}KfU|80l(QhZ2u#8S+Hjyzxl?1Z>>g$R(@!3WPn>}NciQP0LT0=P``6& za9Kc)VYe;$?dmcw4lW98%bt73fPK|%s=ZhDcy>U;;p3kgd@)d;&)0YUcLe(duF*9= zE5Ln^4e%gcr#An(ydzyrSA@{_<^=d*n)VAI(En(NYlU;pJm zO|!fyyTOjo-46@c8?`ZP5FcMHMtyubFFl^!r~3k120pWGY#dw1=7|O54B12R0KXM4 z(3sVEtA(ULv%&Oeey`X7NAQCSy0siFn|+5_UE!VcHWwNZ8*YhP{AN4*vGKS=jDb_w zef%-PdcWJw_Riz{<2iaB=ji#I)AiD7?h%X}f+J{H?vI`&U*Vq15B&AfSEqsD&K-t2 z_4@1bANvOSLhN;DKu^P{SBnR5%^3k+DCDU9 zX4`Uva~=|igK*gnfw*AP;2{Biz=w*5N?dSOAP)0R2W|7tVBbJ|uzf%) zQhTZGJp+XsX#Dz+Kk&}U{~@HtMlO~fbdNx-t+6g|B!|T3Hz)Wya;Wl7VqDMNc^;46 z$$u9gKOo?zi>c|TYS7JLaku#U;^4mn_ML6#=kfR0^Wy_G6}9D4&4!Q3uJGl@_Y%_O zfvxu&e{>=E-FCJY4-{%Vtn)j)*}2#+dIKBgIUC&ao%oN!;X`w8)%wt!O=g$rC?5$v7Koi_ zD*SbQfnV8d{ljkKUp&lyt9dyqptZ<}>Qnx@8kq9~x(nNm*ZJ>aHC#`Bp|=x`j1fA! zJbQil_42Hcam0tGyU<+t75ob|={}td1bK3JQ_(L2p!aBe4 z2W{uL_Hk_c7ryscwzZ!;pzCp6ILbBhDfgKNe%n7#-*-^pSsTZ2PUmp_IE!zJr^F?n z3g{B@5BNnKf}`lz_~p{Tdnn~9e;3fG=p8ub4uKj`+~QfShX&&_xBTRtg8vBciCE;I z;N1bAS}gLZ;N(Eg;+#NEnw`Zja?^bCJW8{&XS!%u_X_SCJT%xSV6UGLJSBK~@a*7u z0edc9*(Ja^yAAN&dbK<|1~j4V1N%NR*d}nit%5Cr#|4iJ#0|Ku&r=b8)cpJ+xHNF> zxPdNrZ19ml+_g_27cWlJCip=+;bHA}tKhomN9yy{ZR!Kh#nFk3gEs`mtg$Sg%iqBj zYXtJR;)S0D^dVXhP4hE>c_)_UdwEXVJXI@+D>e-77nt+n0l7DJfsT(O)K~BiPY>7} z^))mC+5y|dPO)2T9-GOIva#}meE+=zx({29Gw=bsKXQcieitX$md&@{2i93EZz`k6}CePw(ILxGvX< z^W;6~H}tT=cktWDi}T%|V`)om(R*}W*HXUJFE|LVh|%ah@-q5Uzv7MNV{rukUkzmW z;fBB7FjsWv9RhMaVo2OT{(f_;>J|7!>jrAs)UnB-^7o$-Y#W#xe0+L5T}I5!x0hdK7uiELneDAD zQt6SL`TJKK+|*aY{7zwns6LURpjLbL~Hkt2{PX9MHei+B`ehG3aw^WrQ2;3nYr!uAFK+A)f*@Aodd-qJezIlt=w_Sbb4 z^cZlxu3PNJKD!t9ixbs9A^)oHY=_GQ1dIlbl(R zFV#!ZbUd@f&h!1Z37#G7F!=p69%Do9RXq*gUoDui!`?3mJ`vFJKNpAr==JRWIe{7n zHIs6#mjpC>gKnvmri3`>a?h`yPpbOy%z7nnwAK(lA z6V6bBK_g=K*hRLEon$X}3D{2cIqayIfo3$;*_mS?j@V-cHn`|IzuS(@#7oe8uw$~- zjt|aN(5al?b(qU+FkAfG!1c1z=C{0|dy$`TUr!F)`$Y*E+I+;A9nCY>B z_Mtt9qrkHSG->sM4-dpx+EU(IZjv3DLJnnw-)(Cju)pItp5r>c^EjXLvrp<_T-Vr7 zbn@zW?py!p8*QZxwXruFVWoSZ;f}nC-U-u zt;J9BFnsgB1Sy4FS^eg1XB?#tsQMavOIVo|WHsh;^bMZU>gU_MejazetuVDNcYx+;W>U+60bJF}=El}fhV>ssz z!PUV}1JAOkX)^zD&uM|2p4jD+0lOjR@!kLj@gw<@Yz>=3a}d+)7|=fLC27`|g_Ac=4QjX#FHD5Er{oIUqLQ zGfi~DwF2$I?&EH4M9Y`=pXP40r#eeqAcw1+@jw4x{%PykJA0pzeAKgo=LGx&Tp=z{ zk4djoTc!@}Wr4h}+WKi-rkM530WQ(^p8KOG;gk=}o^zE$I4Uq!j-Ne0B~L7mAO=1o zFs{xG#20*=*3gNSan168#%erX6^Ntdi*b)}ihuY?>VRlc#xE}Vb3mWMNjC(>wvO`` zGqDok*qEnX;VPOg4ND!7c;=45on|mc_+@GV?-t;*y9Z*Pd(LV|~FX$P*)b-G@E?!o=IFZW1WxGr$tcvw4V4|OS^ePI0D zb-&JYpZ(nJ?c4Trytb)xXjgE~v3d7psRTnjD42BQCqrdAcu43V2$8j!J>eD$1!3AF%b?Zx0F-1%rgj29G>s+fF4!&?l;@n=gEOs%dv129>P733UH3N?0$jka(!YpnxJbJ%iWmg9Nd@O zlzY7_aR1st-b>rabM^h58>p${SLp}+!B-Qz>MQ;Y|4v`>H9X_6*6jU2 z#=G%t%otb3qB*`SxG-Q3*amij&ET7|Bl41J2jyAb9ndat&%S}$ar{H;_MR15hj@s6 z+c9`vfQ#6_Z38h9pP6Qko7BDVquEV596QRM;;Y5M<}+-Tm}=92&7FeoqLzj(5AywZ zPX1>h_}zB4w=XyjIKK0NYj8cZe%HswyWY{pyC?TXOB4&xecZSEubtN>Y`q$V))wPy zc3-=RBiVmC5iZxBfDdSuYW1}h{Kwv z;x~)o_|R;c7*D*{_y*@RMtge^3!n;{ZX9rnV@+cHdfsL4Nl#G z+Elq6S`wSwScjedeV~p27m0V+aq+`y0oyL7kh7r8tsSfz;3WAUytICRn?ODYPk?-Q z;k(~#^WeZf_H_(=;&_hh_|7Ai!GU)T?iAofF_^rGxQ3%FKEpM1GMtKUa4fEMpL9p} z?f%D_T->8xOMRa96EA90+^(HR{t7QyM`z6nB~@fJ4ipxU_t_H1!5J z75I8Z*ZF+}4qga7uH|^H^0s5+ZgW_T1MV(wo8NdFXa6X`)%aQ615RE@?-ZYjZ*CWq zdm7i^9QFq1usyUkc4&-kaE>}=`XIfKc8G82H*`kNrr?@a2J#1No_d4k%h*G?s>1`h zEOpFmX8C2T8<3mue=LmLa`j4A&)Mb1IOP?&4f+(DK8|;>L25 z7|wmFm%ypo0MAZiAbA{hVeG{27vjSa_-*9J^5nQI584kG7LNPRpxbxpoP}eL=O4M59ydbl2J}+! z+3$_#rq}8IwUIW{?)pX_=_@{j{#z)fD}P!R`oDb2CmBEcZj6kvpLh?C;z}G@{;NJ{ ztd-l0zv_c{Obmy+#BO-Wm^F6!RXBqkH>TO`ug%_n+L((k*0_s)*k^G)b5)odo;8`q zWAueFCZjbrE@SgSTn6Tv_zVy53C+L8XyUV*0zYyD4uNr<{e3sL8;^;@8h_E?#8u)c zKB{;~9K<&8U1^u>1)Y-ppfyf;M}9|~Lr;@qqNk0V!%p#=#W!>uIkCnzt@1x8j;V1Y!jGns^59u-9~} zbpm`NF2OUn08h}r=wW@%<^}n`fYh>i-^VM1*9ZFq2Ly8Z2M2Ep4hhtKjj-M`{I;>J zXL9xn-Wa?#a6HHT&p`n>w`*>j40Uj6js!qfnY8|iRrV&u@oKr~PKo48aTidofJ;u<_7--lns!N!R3 zV(ie7#XI7ETEoaG<_y2cTr=;?LGy8pclen6O&W~3%vYt=&~423#yb38c0ruOFK186 zFKiQDVXxROc8slMgW1l;GVEV%B7LiNylLy%Qg6_wh^bXi!?5_ud##c z#D}iAc@8lfznw-kpYM)S3-`{iXZLK6XX#vH%;)&*COgWe($>T@Vw{_b``pXeKIIhm ziyz7(ZI`xFm(@DugQ|7X5b2zKzLefbhoniW8u>m(>ds;^wesa^vvd~Xt6X|*Xv%~Tl>@fm-F;%IZte9ES2xtui95TYj5o? zmcvo{qxma+X55OqjAQ=Nn7b0+i+$*Y{424A_(U9Gj+iTrAI2QksP~mi$|JaBM0o|b z6y=ur_<>H*T+OJ>&G)%nGX1SOo@F(>F~?Kg4(IseG_xIQJa7*mwfrOYnJ=f)dJlC4 zY8l3y4xgBBOsi4rp^kvgwp%d9J@VSs*QPZDY$3l~eg_Bf)!9n9a{jygPUD`1@;fv& zHaI22|C`?#V;?aD9uil`tKcHGosDPf@e$iEFM|tk0&Y+{wo`yFUKpq+^WFf@qTrKP z1bfZ{>(wON4(HepzwHvJsq-9*`a0*rgJP-812I@R$F;5%h<)xD%$N7^OgtS<9l7V? z<$gT(M2p1Vo^6^>gHz8Y{t<`j7r8e+gZda8q%ZX=KSR7D<~0r$1!CjX1LMb^F=kxy zi_8ap7~hP~#-}R>iQ(x9<{*87c7cOv|2RlJHJwD9gNNuS)!nAH7$>m-Ywx6bd@k9tqBI|BgAC`s`h<@ZA6Jzx{1R z24}^6Ifiq&hB3AiJCujTF>Ef|%Lc0-V3*~anipfcn+s#t<*?+q9u=rTX7@p@0v;G) zU11yBgMH`Qe>?_W@%#&vYw%6iS&rP}c(@0zHJ^s>+=u&dUpUge@;mU0cuITdL+zx^ zv@9;+BZzkv;t}I-As!Lm(8op&5y#*U?~V}D)11wNvDS{C#!ip10Bu07eQJ;8Lc|1j z7-|{#eq&xlJpzn1405w#2lEux3HXjQEcyBQ;5p9`wy{0fw{R@SbbRNbcZ_ur?7|4K z%zUs-ubYnVSdQyFuA^ETZAIL}_K1Ddvf-Xh0x=NXO@3OAfWItXjfWP>`HXy29ZpO{ zhignk8)~fqF6uL;V~&S?^`6L^iiPM>@+LQ*lyX6G54c>T;8V|o4PhN=I+1m zVNu(r9z$IQ{ysjSWy+hWH&A;o)}giG8uglVwsK8t4dn3n7<5+Sz?d_>jr&nkhkobZPfibVp zb<-Sz{p=5p1&%ufzjnm<`*goe68AWsV>`F&E$8yT=|%kU zDdRGzKoD0buT(*SYP7$36RV|<9K#5nA_+6ZxlI7i(CyD#p*Ta9(_miVN+Q!Xlh zmBVmQxyrV&nUNuW$=smOib(A-9vdsOj@47%eWfXXdhfNjajs_c(e6( zBiD$@X=!|<#w&C*8d~e^TBm?-_*y>>@D1H-tVL+tfqQ6DVh+CAs7Xz8_Be@OhnM6= zZaK@*H&4H6-kpS>09OEbgJ%Px+PI!B6pY!~(5%Td2lB9Kp`Z2hp+0LHCKa zb^oEp;K9kmJT!P%fHUN1rm()iTeh{2eb*0g%mae^2J6joIpP_bpqPe+=6b|wVjG-u zvp^i@`pY@qztCE0_e@ixNvdy?8>3g^Y55&(C!TDLn`gZC3-JA3ff$F*rq=zYY7F>5 z&F9=PuS+v;?4#y@W`TpmKkB7gd(b!t zC$Sl9ID0+X>e^}Cz*d)^%29ZzT!p8^O~5We^W>J|se+wnzkt1Oofx||1;2Veh{>9V z5X&?+D_1mT;-lk?#xv|GO^9u6T*IfQG2)QwXkr}tSv5v>8h^AV5Pz`aI7GZq4jE$~ zwqI;9jeQ!kh(*SjWsFnCSY^sD)yy1w+^cv_8;C>4I7a*H zpOIT|HU6$1CXO+l#MQ=_T#cNf*qa~3AEJkqU+7`YsZZmXv7XzQr)|l3;FRi5d@_0f zZMu3Bn_N3Q>P`5<+^XH>?h$voUtEe)$2z=z)+T+yUkJIP0Hz4emWGd=ux&-Add)#JP$=O4vD zeWquOfo_*~t+Ds)zIcNsC!Zn)x!+8edq`pvobsq(qoF2YjDy5DVj=O)2dRGY(3>bCi+Yzls2Uuf=PCOk!h2f7`U zv)H4Nx5`~?75Ls5WQ)0|K_ZH$FlBT+t^FE@k-Ml@g4x(8eV;wSvqW%EeQr}=I>`#auPBQfSpBYT5e+H{!o`^Qr_v`d|)7Ht2H49^m)MtX^ zjM(ZQ1@a@_`^k>evHIL8old;L0ie~1LBt}sKz<)z(d=**h+6=EiJ!zWBXHUjez&dt zXn*1_$8~(?aX!3ri$MH?Gu4OF-PH7{Z*#qPNQ|fMLe59r=f32`)CAJv@UncFIPvs= zZi&0ak>W|Ug<>Reruq+juYJ7-Nt??BwH8j_h+SI?Cnr>{qPtFO;nXScON_3Gf&WYkZ_Fn&;78GHzl_T%>Kab7Lg&upYwU`jg**>+wI|01xqP_&I7~ z_&)p~9E5*bL%?Sm`KR^BYT8sOzn4{%VEDk z3<3RYX?`_+;U7+8nAQ_mI{t|2gnr{!PO;Fw3&lj!IH)m@Yi#_(m#yB&PVjNXZQ>s> zA`asJii6mdYIEfxelpEQ&OluBkeNm&Z}ga1Y{b9z?gloD-_8HtCBQ@S5%js%DKsYX zoQgUvdQ#tYAg@vG5gW;`v$5)o`%LP5F%o-CryFbL%0KKr`+u9jvq0q@aS7lcz#Vu9 zkAT=^iH}BC=OD;NUswwwI*Aid3M zH__+W0y(Ft_9t&V);(1JQ{TW6=ZIFTOOMV%KmFjjweP;t{zB zHj!<_D~(6^_iQH{T1|~@ZQWku32_9JPs$vOeZ4rk2S zs5=<-H?;yqwYX|@maS80uE+6PJAiXsn`@R|axc>wHL=)Ev)Z)gcC=Yzp5}M(cHe0is*~~63hVsVeXH5A$+gRk4cKH_1>etS@dA5YP0sm7?Sk#c zQTV_;xEC>t`?B=C_h%b%i#bb!H-}xLW4fm5eb%{W@eF=(T(J>-o)&_S%~$i2{vjuf z%f*zSKDK(O`D@LPYE%j)%OL$*n+1$%|7)YN(A zM4Y6Sn;m2m`^?bDN3GwcJK-dH9X}s0(WDv|ot{{LEfy206O&I72Z=9m4?B-n@PK^H zg9G^|5eX#K!%4-Fm>(Bx=yj;&U~@p??_ zoY(p7CwC-<5;KXD#78(tJmi@l+#|-rJL=RM|EPbH9})wpgTp^~8oYx?`)DWaC3mC^ z)gj6ssgIx5EZ`zMgoE@C-dP-oclj9)3&cQlJG{dm;g>85_$gv%zKi*w-_1MoizaT& znylI7&mMz7)Sh~h2ZxQ_7UG3V~$yUx^+9WK=Hk6GT!P(xn{nU>vtdSr}@;z z`qh3L=U3;&(?0X38Vi=M`DBiXv)d+?&NZH=nnE>Z=K|k>F6}zT7}K#G$319UZR(!2 zpE)RoG%sl)I1K-Y3&m?V(R}wuyJJW2G~0rA*&FspJoLyweuy5YK0%E(eTSbd)?w?s z=U_>IM?9y}|2v2O-WqH*U!M7AcRlw*qgrL z>w^6P&+7R954^YYqrs8ECxa6L?-4yIIC%!|`z-w506trOIrvI&O5p#LIXyTd@Sf1K z0?RW4>y`#z4eWd3?A@Us&wEQh7JMLZEpG|-58e>GI#9>`l3@2>w?OUoa|7@6QM>&w zfxMIalzbHq#926WQ6QcYw~6V*ck)W*E;=6lPE7;dj&3Q2RLel8!|mcuF{nC4x~Mt^ z{(yK@9m6gGKBCRFh7KQjHdQQ)cZ>!8QGGAmf>ZcYa*{ZK&s7c1I1>XIb7CORL*NBt zkw2ua!5lTu>qFrHV^@q*9gbd79jrn#8LKeV;RKQy)0q|wjBJZz?TKwKj}5YMz;fZb((M;;Pa;2$=-+%vxa z&b#2e6VCtZ#_oI1vvwEU>i;EPc{w)IR z^v^oK_5F92<2&%|?|YA7+iqWe^X}8z2ky~*<2F2u+r0k{zxn@?v=g3F_pRMA{} z;W}-M)6{I>F}W)~0v;2q(nIwjKSIBX!RV-TKKUkNz<4k&#KOi3t~0L0&1~?PYbxIv zqwFwFxhc-WNBEBZ$5t1WMhd+`^Ug1Tf#G+-aS;*gO}F z!dx|H`E&d_^BGq(hT;$63v=HeUsJA!4~jGRrs6k#D?j!T0WC!?hrS|@!|xs6-Npvt z5!xBsq^?cQ#`7igGFln`8kf+3-=VW`=1*)CSEG%_J!92Q{7GO2d_-u z0jDm67TSCdjmL%ub6~?V(b}%VvVsb;o_OT zX57-%#2m&tzlRTL9`P;tn&ynTNh8A_v@w2N>k#U9;S~N>d8L||d^0=19^eQzqPdsm zU&#&*ZCwuDu4)D%VwwDcNi`irxz;?6w;uSGUV<7R*)xj@v%FRO=YeDYDsZ0P1kNeW5@Wd@ zf36SL^}VXD9Y^9#_v3x%jhEbGdBrL81P-2+J2KY#A0lUOLu~VKq zWw-DVokp#%`Zjzdx9$0!4+d*FX|7aJ`Opm9Ampp%IaaF%!oZ?WsR>v@5gVwd0r0Uqn~rTA=*8SsMG3BSE6 z5S!o&v6S~Dc+bJ!0UojRT@2%Q+u6r4JdY|4^DOJG0q(>zo^^FiPYYb*R)P497DwB2 zPhuLmCHWz_A+-|Pqw&nB#o>1~+G58(6ExOmtBvz63b7|`(=$QpK=9Abf&SF5-n+g~ ze#e+FHtrj!IWS(vI=R*2*We?%+BD`d_UO3gj%QZoU(E&M)|fXZ`Bug`e+(zJ)&MWn zuft9JMe}{s1K0z8CV#Ur4BNpE#Yb|`?27s^%R>HOpMXYFp6YWpcxcSw;FB?rBbH&y zcA9{VYkfd-JI(L#t;c$4{x*2_W_PXIg{Z8f(Afzvhm{=Xk1>RU;F_l?$5hX&nLX#lzx3Z8~55Kw}F1#h0LGibMGr z`WGMQZ?Ubum*?TDw3f}-pidbm&^SYUTun_ZU|fkC#M7-Wz!~%~zL7Y z=1#c+5AeT?_2yNK|K|6nb;-2j5n={5dOnCHKn|p!S^G2R#(ao)gN-*IIF&$zN<{Aux)ytP~cA6x7tA4JdLcW*L#?v)?DMSuhO9uhItTF@JF4En|2gD&OFT(8+AYAB^6NNL495tiT*ee1Y0M=V-{zBX ztH#c}G>`dM&9AZnd_z7YpNwzFUhvhL|KZ2+>DU?erZJhkDDI>20eeN;z=74nr*#a| zT01t9-%Jn3jWoW3J_C*0s`K%^3w9S*ww?}OLcfpPi9g5Zn98G#@y6vCZw`)u8Gz3=0`j^TUPHCDtujV~=*TkZOQ-c|T6M(ut0JgyVlx@WO& zp*^&b_R@~Hj4z^HwXvE7?XMrIzv?G_hTFv6%@N@-z7H*H)Um3K@tN>e;~lZW7$@Ma zzH6tkg0a|lx$(&wH{dXSnz6eOm*FvE-I&Kk=D?JP*mAbNHoW#5SKtn@2OHjJlGt(c zj2&0kz{j-gXY+9FfPDv_t+QQWTl=>r!ZB&`=C1QNXL)Nr&T@_An{rOM(wqnUX5Wzy z9diVJaxU7=* zcZJHsh+A-{n6vyT_EgKRo_fCkk9yv>JSyL&jzb)TJH)1VL_Hi%RfB+AaO;u)zp72Z zI|aT4@hqrc0Q^+=?ze7hU&nGx$94|qa!%)V4O8HGpc%SWHRJSMzJS~my|qDSG)VawMLz>b$t9IdNiMj?}Y!14dcX^X^pw@YJ3`Z zW4u=#lJ84-NT)J@!(r7w6$S zF#=mkp9J<*Oem*yQ?=!6^Av10+g%%8`(2*wQmlzb#huPmu<_3A8eEI(a(&HHH%E=X z3vmtfyQO=if7#x0jDNI6fj87sYbX3Mg3mDKU>YM?@+-78AEP-FJ_lX6@e*G|%+!2K z{Sw<1Vyt4UUueA-&Z5uh>((%cyFgB+Xe>q(z-#g_v^?X5UQi8;W=2NmQoA_Sc3e(mgmr8@Z44*i2^@nn@eyunOjz#f z5{H$?a2fdScYKFa@gFY4k2td^Uyj^p-!bl?>&=H1?7f0BN8ZGpt{cjqzjg}Sd29bU1!<#I1uOYf5v(Qnm}_t z#==5qoL7w!*Nyt5oRIhq$I&9hci=Nv58}UK+;-$exuo$KBROZ^0ayARZyN7~@79+~ z@uqo!C(RM@V>xo5#W_v0+-OemHAfCK@5Er`KXVh$;UN4~9xIobzvaO$t?Raph0OQH zL**IEat*#&2<57g@5E8%9lo`AY9ZA3wh#Xsd>+yDBaim_TsPo7z{4Z{5x?SP$L%@B z^SXyZd{D5J;Mzx6XB+!C&h&am-gaHCv;N1}clu``*slF%|4}<||6@Puld)ggw&OkZ zJ{IbqalLz;-lO|;T=!aN)57`NR~=97uin3VHHJGzjq54=R-er0@%Qf6_4vvhDPNVd zs%PM?V$>Yy2Y{mix$h)v2)eCs(Cq9;n?#0WBz{{7q~Xp39flW z*IT+L_t$%M@2wH54b-+Hn&%nY+df5)F&?vRI&RZ-U3Pod)a&Z8^bNGHdTotgYLmxU z#plK=mer;#ji322ZZm%VNBei%j>~?x>@~O^f8~>oTlX?;OKqVI#_?vH@!P<71D_l7 z*;ZeUZPs>j&bEhZa-C!RMwhFOIMw~k!|IRbV&|iI2#(XX?EQ}YFn81c;Gee1LYvKZ zyzXPU(0=o+@3F={?e)8t-cRqvcHO5-$Ngu|-+j95HTC{#itzfJe)`8$R>e)N6E zPUpGt0Ua~-i`c9BNgYeI8{=3q?~EgVuBYQi-?PW#*kQ+^I&{BV8dt`c?K_6Fb=%oJ zjrYS=^nOMgT>H@Z(K*ul?b7;gYk#PXaoiELOSLhbH?9Z#xjy|e&Yf!E)wkVO=SunO zrj~7s&ZBYcX_xl@xO5%k-}sx)9XstO$82AX*W1s2>-pVV`@^x^_t+;ZtnlCeZM${Z zUyR4G4cmT>Ip1}yuquE5x7XI?c#Y$jaXf$Hbvkc(qVuBHXW4!?7W#So{iZ(KzQ6Zg z``yyI@!s*~bPTnRw8Pjp9XlN_Q+wPw)Ol39Y(MQY?YpsE+9s~gb-Itvh0ZPWrRcmF zm*e~y?{&Q2&XxA}T<6UGao@f7j!{eP&@oXvJe_0aiTka8SdJQP&1iG_d)v}DTBwa& zp6mWcZ8YBheB+?wz4zXG8}G$6bWH2Fe(v>kJdE2-xAEH)w)y+)`MV98w%c?}wVk?b z`?lTe-}AdR_tL)8m;O5U+J57F=;v`hww=a#(KhS&=$M!;+m>x7%YGi)tZmV2w``kr zoo(D7Ot-hK<2wI%zw_PmIHx+Vj1}m78ab)!x-`c+w(LKSt+re3h<0)foiF3P_x`$E z(K0LUVQk0w?nghi?K^I}Pq*)PeNc3rW!Lq*mR;9l_E^pZ{yKi#W5-bM`=*w*9nl!2 zHpsDiUe_^STXihVx9k|O?|7g6?AX1B-iLb`|2oE}*LG9ejQ7<0FV}Qwz3tj&<8r)z z*EhC-dz#+sKbi+)JG3pvd!OE8uYY>&wFT2<$9$K5AMbO#*PhFAyyv!W*Y(`vz6<@< zb{vo4vvcUz@o#L;j;GoVc4V}h)@$pwU+07I)ApN=ALn#Ez5kn9YA0Tj1SM8-A*mAbMF_(CXo!1w(b?xrU{koQp8N6%k zxTf*^J+FK0IdDb&vXM8&zVmtPyYagJnf0!-`}7{WU)#I)Jw2D-{noMB@!9P=E-meA zdw=?D?04VCzv*xG=~(Fg`mE^ombKYkwr{)5xQ}!8xGQ>mg)Qhg97EsfD}Bay;(Ly@ zR=!)`ey!itc-Vf%U~NIq?;OtAw$kSQT&KVO);83BJ{u3lnXzX~SnnPjx7XTn?!3mB zbN9H;VS9byoZwvk9HZCS_VT@B*ztS3-iP18_Kx9ry;e)#ozL&~b&SIGxz@IsW!p*{ z*k`;Z$M1D_EOi^(`P}cWc|LT`jlcCCx}4sdZM&as{An}y*8a2f9bDT8>$C;fuI)W{ zpxHmY?%v0IZQ1=T9bdcl7>?QTVQd&Djr)ujcB*<%H6Y7s*>q*!?dRC8LqE+o_Kj`h zTHpEI(zf<-O|GSV-ZpiA?#cC9r=Pp8vWX?=fAQKJ2x$ zKRXUBozK40<2gp}&3&13v{>y1KDWKKo!|AtI1bwXoy+5Ijya;+I!=$Lzq^g`qd)a! z^P1I$N9}K1XFK@?xd}OlWr5s+b7(txCG`nvHq=6>*YFI1S`zQ7lFO3A(T6>^rFQX~ zniTbzjw5%Z@8w41K0Fub-0q#-l$TQLpgzU344!FFyCR<@r{o;CRNK0L=a5(PyE+)T zEZ3qPzvE+!1=Y2 zes-^YzO>^@e#>v#$Fa04un~@H>}fOO)cDl?`rJCd+sByE2kHbmmMop4-<{j|vYqk5 z<~gTh>5JCXIfrdPKj{<4_g9-|2pS^69N&QGwd z{;U7sv%c0}^0xB1=1_CH`p9_qyS%JzE%l+CHTbOGI)+@EbGq)fh5KZ8U86p9J?51* z)@HR2+Op3Dt50m(`dyo8JN=|TjDIzaeTKx`FprD_<6pbmPd~IT%uBT)YK_3>juYF0 zF$3*8=d$0jz;W@Ev15+95B;EhjA?U4TWDwZshzc%cIq6meQT&JwU>S{?j29t>8l>M z{i2q@IUL{iwz0qdaO^R+?mBD>^6ciOZMDDg@6VXi2OS%()pa|rYp`8yz42Tw=$z3X zK6h?8Ci}0Q^n>=*Kl&5cFYElKU$wEhr%$!DerSH$Qvd27`>7SPjxB3HwSV-jYcp3| zPi?aOj0NAF!+t)StHzLRyN_#WAM0=5dt7a1UU=`3+C6pj>eAKNtHW1&sP4}1xJ3P= zcGm|!tHHx(>ipE|sXb@+^ojOn$Jix(!A9aTwR?S5PQ9b&w(u3+!e7Rgwsy^~*F4jI zYAB31b5DD!7gj5$_DjvQdNOrpYNyqHEe-IMxu&hvTdC(#4{e)nuii|}9L~aXI8Pty zBlAx!mpU@fo;p|GGp4Rg-Ltwb=V3Elmwg?>SkSk=w@+|j+k)*;li1o)%RZy1c2PZ} z_Eq1A%dO)xs5#XaY8*X}G4A8}H~ZMv_G%v;*Jt}^XEl-5`E1+m1J4iWhs8mC5_4By zxL)&xO#pL(EjMr6mws};>KyqBfWyrXW3;w~%`)!lN9kK_r2bodoc?hf$J36TBaUOd z>kECVmald0`rp2-r}M0V<2jaN>vv0Ig)gEu&u7$g>x_Hzz-QZeZb{pFk56lu)mf|8R&T7YwX<4c zbA?S)EAD)J3fE-buzPsOHM3cbS$!@y;YHhm`(Pv4xbm5Oj7MYDTw}YzXV7oPr@qr? z#~_^!>hA=?Jp z-FIzVU(9+wMQg~lv3c0|NI$lg*nISxZS`M$iq2nsqp#TDv0wCq>vsKY#e8GXJ>qcw zQs)=ntnJJ18|yO7Nx#|8cE)i1L-r8tW1Zs~1KQBI2cNsH`NxpPN<`)o{T zb2dSHg3pd^edmZ{@ZB88(lxub+LDel>%o5Rn_aI>@Yy=o=ziU!d943kkL&Zh{d;cT z-J9*}qjXGYL+95%?3OX3eYKT8ZD1RHqu)UP8Uy;xxF|37z8wQk8V}~Kv0*IfYuow^ z;;xQC=K%XUj^(&+ye{qPuh-UbWgG43zTJQO*ggC1{+865ZLYyN9i#rC`Cxm0J-*)^OFO$q;~TVP=Z~@E8jK&;?wai5 zcktP9T$}A|Q~0h=dmiVuy<^&@cE}iVf3|1uoWq~~&`0KucDK%cY?$_QPS;?&&THG+ z$2@C4xnFSq`q4ZqXR_(@r^5y!c= zk@wuQd&PZ)@76C1e73K4cmIyBy&V^vQ+peS#-Mv`oAe$nwUK+NeRBQn-?o=+#y0GD zF%}$i$~*4eIm;QwoU!M6jCFlfXb;z|@5Dc~(~X7Pi{IU&`EP8o*Y3SzO&jQc+d7{4 z=RB6$q5B(qj^lf`>m1To+RAlyKijy@+Cjhd&s^~5-U`Ps_T0bm)wZ$JRvmMe=9=}L zgB@4))o#wCy8t1y}ac#ykPI2wMKYi|=YNNFYdo;q>VN>}?#!&5SeQEo)P4ta+aV`F| zO~Q8D^0+E=@~fA?ujwEx_1$B1>V*LC-Pdynqh z*f1w-*E!H}W?kor<2jFeW@kre1J|J)^u>IC+NR^g{4P4?+i&)FoWgNO=vU`+9j?c` zfa$f_23)75{d!DG$Lu=S>)MQ8eP*2LH+I`{KKKpvgL`4e9LI5;(=}&s7&D-AAbxgPy+k*S3{8FTj;SncSm2y_Ks&8?dKfYx^u~RaUHIyyod+7U&opAbvwuBS1t;~ zfBett1MEfZo_nl)a6HGwtA+C!&(5JeYY!~lm-D$N_s+ID2Mx!vT+QybjV#$-*X5dM zKQtm*5kJ{JuFo-@(_hDur9bV_Ytg6r)7&v0jF*CaHCFVowsfpszhfDf=3mED$Cz>9 znCyW4j1Svd-*M+0qi^J&Z_Nh(MXSVgbL;WM_-x`ollm&I|i@`=CC<7&KL93*l(QJ`O~>XC#9Rx!R038 zBIG#aI-2k3dpNvH1$?KYH<#6XmYhiUSv;rIFP_`Zx^dfXZ@uFo%;^az6-6n7U$}Hx*oY1`2qO@@8a^V7fave807Kf z7v%E1qhxvyuFEw!pL4T2+FX0HN0!=}E$a2TZu+=;0Q=Io$M@E_PHoh7avt|$`_`v8 zCR^7&9_=HW$lmfH*xK3!k6cC2TgZNnym&&HE+WxV-o9nskvxQc7FQa^EjS;orfK1PT#VcIy*&Em)E4Z%cIcX zmj?1z(|3Q!HG(`0`0V{5^3UEaB5ygqf5bb{yl15E4zaYY9Gg6|T$vnN--GNuFLK=8 zV=}(C%V+Dn59a^zbtmxJRdfHxJLkSD6)I_v=FmV&GDNcm$&f^7o+V9^QtECnlX(_R zglI5^aF-!v$e0YNR8kr!Bz2$D|NFkq^?mH;{PUd8>vi_N_OSNaYp?a2*4k_9BRzfG z7iZ~wec!X^0MnnUe_ML46ze2yBGPBf2 z3*`rWiOO-N#Ag0?elYu>43)Doavk}RFZa?sy6@(3SeDM4jq7zDX#HdQIQ65nKdbu5 zGuK}R>-r?s3H5BKThuf4afoen4%uHDaj5IH3GJutsk(op5XS$snc5RQ*3veIv<=ZC z)e+j5Kk2o7YJ+GyCFA*L+BW-#e9`tRTgZh6gNy|EQ~Y*p<>gQwStq0Hcf4c$Ca1_c z@|axjoLOWj8Cmq18!uwKq<(b$;+->6pR75s%!_3nEn^reb1jW)aE`v*&Yfl1IfM15 zJ5Jg?)_&tzjMFq<)^MJz&TnN~=gcw(mbv%M?QG19`5P_GQ)x~tb77f(tMhPmF0RhQ z)%mxYw(YL#yv}`T>Gy--SZNM1ZpS=Y=Fn0;4>o^R`@Hpgcb)Fi2ku#TCPRHek5*TD zrl1?EH|Z+6Y1j`-XV3%mLSqE#p!(kWtIpQWC7bJGYA@urHq_X$cBfrxr)049OU`KX zWPv%d$Sq?F$UgE8#`ch#=87T*jYsI*i=9`>c!kPG>2y9Vve0(tI@VYrb3K|H%RE=s zJ7>Ky5XK*wFU!2Z=1n%2aKCwDX^vrY2itExE%R!bL(6Y-3!9U;b0v2!W53O#+;n_f z#XQKq|JE_=K=vSc+$Fip zJ|!Q>Nix%zJ#vxk^Nsm_3&z~H=EL>9OY_zH?xOG0Tb>`xW8ZmnJEwh*F&DgVG@4hp zY1rpmCzphNOXjMDt{Q!7+4p1a4^eF+d%j>?>gITMS7#luoz*|2=Dd}wgY z-Qm{YTPwE(zkS2mw`nYWkJE8WjKAl+;>3!hZ`3TEZzn%&%rf~N<;vkPVcoDncv{#z zY&~X^e9v>xwP zrpgwwr}G+i-MonOL%*X|*>~djdo|{oCm+%I=*fNGY9w#_-5K8_hVMQ5mJAucb+D|= z-#lN1@f}Ftb@5G?O~Q=y-(vLLvwj2GG5rqo`VXu>`nKe153KgU%42+o(>Kw4f6_O@ zd=t(0(0o_ZcgB2Utlt;&J!aqMgl)bz*6)lpowXlWd-RQI->mcD#=I%`-m>r4`|b#t=ldkrg}(=SV{&3Ll75MaM~C&o)5nnMPY6#5&kcKr zgTsMgkMZ}$j?Q<%J`{Wl`Rn2QaC-1PX5Y)S&vz<65k3=qE7P}>eN)*tm!0Pu&6fn< z$@GoP?+4o*e?hR%H#ME(8<>utAABqGn&3N`zW01;I63$Rvvd2sOxJ&7czv++%}n2l zmOsZI9$pb%6|5f;b`Lv*t-?0p`C-REx6x_Jiaw-cE$KC7NeA5$l)LXTdj|ck%;TWwH*!uK&n?WKGC#iO%0Hri*d=t1_?uWBUqn3;;}^d- zhLjEhAD7?3&uGj?T~lXaUA^Sfivd}-4e&YCVetwv1%AfRC*T*b8Tk`*0Qv8|0(Lzc zm))y9vypvQ%=b#z@3#Bai8cq{1fCmZE~JabKX zpa;k>a+EBu{*ZR}IcJqXp4m>n_+44-`^UYe?Y>n;PWr}~`?gK`PY!gRZ+cqyU1Z;~ zeNpY|&+`y4|j`VOdXfcE>-#Fa2kZ2K{d5g`k=>}zQ2SbJH0W`C1g@J%e=&tiv@W4_r~y}=K7e0Xx$Jdm9) z3FPR$fzCKIye_;cyf=``9}gc1)fXR}_~67x$B@r&4JU?qq31c?G17QTAj3}z_WKqx z-WJ{*obNa~>C*{e`QlZERuh#AGz^TaVK#o@hJUX47bLfTKDwJ@f-9V&;$I0_ASy4^Z|dJjUz5i zC-Cd(1no#nK|3V>eN)6YqQrsN_3U}Jytoe8-&ji9ymn3wh@0#htdjxG>3MX7{p1FH zLMAxR@xDJv4#7Hk#5W+XhO&)a=$nsZC)?Zifop$0{(;AjY)Ez`xk_GJvQ1%~JO(*W z=JzvA>-)>dV;S&XIQdleNVZ1#&Pl>FG~d9Fs{8!AJ8w|Pp{A~?D>AP*uA)> z`<qug+=S+4N+8zjI0c`#t>LF}Y5Eko$D4vh24!Eq(i#jHl0)xiaQ&(akRiws{`# zd_5~m&&hKKo7!`<&3DPz3FIyMy;62H`?~Y;lka|uk&*voH2KV~SJ(O-8^?4me&;k! zNmq!Ku}{czdVv3`?vmsDUVSVOo8k|PwboDOBa`n?|KS*ZfOjC<_d_3O7kmzWb7N;> zV0>wPVD`12i_guzx;F6Bt1m3YgT(-}TlO3OLOjeLe}WI8eRoWm-;UL<0_(PkozWj+ z(PC(Hh<*<7Y%y-{@9B5Y*Py>vAA|S@yHiYrPvIL=zS(qo5U(PqK!1=`Ah*P>z8vfm z^XfOE#Ky$L$T@z9_!fC*$rqtl#B{`R*w5r9IVz4NZnQ?ATiD%C3QrAN1UC6D;bq~) z!Ebtpo&NH$UwCDpd-e)14P?0GZeyIgQ+Qr*zH}UWP~hwE3-~?s7_3Vdj&mJ(u(Xe! zfPFiM?E{_h%pl+VtJMOZ!Tr%0b3$GdogfW1yyIZK-ypWM-*MzNokEXP2a)N8j;j2Z7jX^x%XRn~miF5&AN|I$ zYsy>WUcIJ!D*OZIN>7}|`HtrU@CDqj^rS;yC?kC^^lkf7ls_F#*Q5R&y-h!m|8#+U zw$%<`^Lxg0f8#IqJKsJ!q~EmEU*y}AbOJws?LpRq?Jn*}=CkeD?)t^p@3Z^J*#D@H zKu4&Pd|LI>dHOwu{Tap$h>?=zWPbbk`2wALv2=`M`4N6s z*R*YESGHNUjp?_aeQxcwZfDc+_vseCN&SCgvc!(HYxbbna{ZC|BHDg+3O&Rpp;zc0 z@rI7g;hUIWM4t=&vPIZ7RLAIFC2#aqn=j$eKvvN=Vp{r)$*_JS*pePHuQMOcO5-jzNvXO`i;uY zx1otgji#~T_X!$bgEx*NhllF6HdwqA9EvG$bcd*Y`CiWMb zj4m>cN1N7nS-Xv2(>`@_fPX{I&`U=L_A7ZrE|E{QdD*@6PU9MM6J15uh-s0NbW!ye z-DI6kqq`axk%oQr6dB7d7q79dpIdB442B%8T~EGSrxWNMdZhjUd0u9wlMs<*1Ue0SON@AsAIBmEHlezH2k zzu^OeuCj0D%g{4p8rU?B13wv#Rmb=_{G0kR;#PDUou)pjpZXv8F&(4EAK}kj66hsM zI*Omu?^OEj7`}-5Z+mIqpY+<_C*Cz%*RkpJ;nR0)y!uDIQWpYVGKS}(k->)Z&^&{|`s>_DH6Q4qV`*K15zy5lC_YVpD z4)8hnAN}*&F^;F}_$jM~#<=JN>$Z#ctsV3aJvr#>-!{~bqHpLxx`xlge=E*&zI0s2 zHDH{9{uKRD^s=~_ekzDJ!a5yp9Dw^bM!;`lBZLw@f9k3_ zYpj|v@LDAV?nhOZOCs+?MJ&6bFkEowJZCyVQrZm%bum@ z=sWn$&h1z+>-I~>zV`K#J@jDVud%zyC{#wZ&tFWP&K$;Q$TV@9#&0U)I_^o@(v_Bd zjq8Z#xR&D@!|(SNYxCPLUGYaTCBA|4<-z&#XuoZZA=&3T;!ET;IbNAw`7D0f?fS3^ z`A-kf4fK8e7(R?NoFk2W#Xk&atiN zDRq$SrxW-|WV`yV@5GoCwi!9!aqeV2KZ)E&x3lxC^OcNgah~|JG}wD$57O6WhB}JR z#4dDA$Mu@haV_x-?Q%GFnVw-|>aU?^7Rq0BP5q&c>)|K)t4^|nSf2NW>SJ}xvNUby ze|3Dm*ox!&u2OXySt1@ows;S#eZ6#+zFzSgV~oiq{+B)hvWmQRor+gC3z<#NGH6q{-*eN9H>G%WntwA_l5|t$jK44u43zPFc|Nu-@Nvk+S0RvJs&C*a*t9 z@TJ%h_Sr^vu{AscV?y{)dxe*c$EfkS`z$?g^?>f;OQ|pFjry}{Fm8Y@Q?J-Uo%^Eo z%dw8L-S~NWZU`Go{nRG318qdzSC@zSu4B)%L+zjKsGSb=oprhn!#SVm#l}0d`?h_0 z6XXD2&yq~h-krlntt_YxWp~n#)rVwX#9qud!>o$SL@^` z+2|Z;y9SwAd22h}N57HH^cB6t52zlf{;-W6BgcCT-9fj|BcM0v0lJ6ZK^I7;dVv0; zgB&~58P2J`>)30*tD}1S(3hd-98>+|ntsb~b+CN8=Fk?Pi`*|g#AfKcl#Smh53ynU zm5KA+yELl5rR%rrxPSNPzIu$~+_U4@2cDa9_e|&{HWNFF4K+WgZ*+$G!v9bgE!89S zmEKUd)I)KE)-QDoj-}uDIeeI=DLr-FIrW$5De0-Zt(q)BQ8|o>2S>Ef=$3ai&``9aVE=qkM z;*enD`(2xVxU}7VdXHXoonF)R=tS3{5BY+w<9!n{Sf7h!-#@W$?_e7_t$)TiKUmjK zLzn7*Gv3d&-3Ptwo}r(gEa#8Nk9FJROPRpBW983r@<$iAws8ZlZJdbw4fE!j%F40s zN7LTourg7cs z{OV)c28P@0qdTErOUuh9%i0jq38IRXIr4l80np)7fBD=)`8*u%4_?sW<0h3^m)0~@EX02 zbk(V0T~YqE_msD1)aTS^WdHD(+0W3v;kix2Ggm(v_f&6Mchs+9Z=FoB4`hqq{NVbi{8c`zzuFT^491(VBiIXU2tEUONe+|YWOx13 z`lkF6&>^3}UzZFmYRbkbE~|?8Ol~c3XdB-~ZyP@<;iqe8Nit|Iqtf{84>p;<(0z z8~ehi{B55Ce9;_sH%G`ubqKzlZOH+2&m>$Gdh+%Lz;4-ff?KPT%J;uE%>>?#F$( zKljQf=R@<=J1(ek0P%s2FFIp9MxGC?-_W>LD1Sb+XW^N6HlC4wM$gTA#m3Bg=4>lA zn7YF5V~40$>=rtoj;5R0#p<;Bt*=~r(!R(%?NvJ^UoHt`16@QIV*sb7XWsp}bNbHdo5OYCH{n;|=i$om z!|=WE&G41*#c<)6^QO<6{&e_wcyD-XI3=7AjtPf_{lYF`hp<)HBs?uVIXp3}6;=08>qMp!dg z_87b7UX;Du+1#Co8E5vIpKNXh2h2FWns^7U^pbaIvf!W56Z*(QQ`GL zSt+x3h4+V#hO@%C;Y;D`;oIRm!87jO6aMSbDd@_p5(GlEzHpSAsC^fH}F2h;EBEg3HMLN+&MNEY+O*vf2p^^sSIl#sO%@-n}$#Lf;V?W1OD;o0Wrc5bJ~ugYgE>3f>=j zQ84aq|6shstHW!ge}6e!q#K9ncQY_yRbueL3rU9>(39} z_TtGGPwo<48Z37XJBH_l=Y(em=~((Lz3s!cVXLrZu=Lw8mYavCgeQc@h2GD4lk1J{ z!S=Pnnqk%O=fWh4!k@=jzdQUR{4LxTZVopE%Ui-(L_;qSpYmQf8;OHMC2lAmPa zV*@#jCxp!dnYVk`H@qT{bteb1>{HbFeHcDYeF**o8rO%Uw;@wCu_yc z*c9@lZnG`K-`LIUX!piOXREtkb)K!y#^7TpE0FX2T(M*^V{uY(Onsn@5sUq56JoOZ z0nM9X3?)0mxFlm(j3qL*#rPNFT{@1(SQvIl$G!A@v5uFqolf%H?Q^`bGSZa3dY~_| zeWBWXW74wnH4(`C__sBs)u4=sR_t z_6sX3@p7?XG4s|_dFZ?AeSge#*mV3s&w;(im*b~Z*Yb5+N5x&}Tk#roQ{B^#qHd@w zbZui}Vs>I*WT2QIe}kOVpQG=`*bla?_d(4UVJw63^c~yCpU~&Te=r7w9j*UL9~D1= zkHGF`$Fs|g19@nmpH~f!3r`5-x$)@F2%80C1HI?^obbZ%vanz9?$c4>m|*jPcw zeJNuTy@y3#pA*gt^a;IkNuax}e?43rz7wttzX^2FZ^KW6?R35Kr6awM1-5{9dA*PMde&^xyH{ z^x@S9@h%p>qdtb`qz|tBZ^rxR|Kv;ZHN{o=rDB-;SFuJhQgwkmAus4W@|Cs*ANLP3_mYy(H-FrzeW_{K1dr0hC znzJ55ej_!bar~AcZ#75|e#=x{gx^1FoUE)=X5=HeNTwOnNUs^+ZhWNt8fVXj=vexWtM}V}$88dfn`F0JH{Q|xk&pDF zzBBz_{60(lXZoe&NB4=XTz;fgj!_9}j;FHwHEo zTT?so&Y$+iCRP7E1G<*(^lUm_i0)KR^o45Yp1b&>`XKfo2En&u8|x3?+tD>-hH+>5 z7xb}_FKiXEM&Ab6qd$huFzRR{#fN%X*f$&!jtOrH9}1rh z7lm(zZ-g&}^TP$iQ81-{$;6Zij*?{>q)4HLf)zYISL^y-(w+2P&cgm73mFz~BA5APi882H|+ z2hWYJyCH}-ib0A^8lz%tBwOPXK|GCaW0UBI6l>+9tsV3SFA~@}Vq$zUGKg(2_H|BR zi_uT)DE;^J6aSz-8oSe&>zPsH+mGh!AuqKHejC|JcZh*$Gv0mmj;y!|pXBZ_f6vS} z*8djBEnmjxLneHQ^7z zx^`~6rMCauAT9glNgnSF{|?e@p5@>D_@h(nW6?MC43tC5z@8#K9n((cD0k!Wz0cM6!jwOo!G1WGF5E11>^}WQ&sEoLk&*m#rF7&~KFwXx z_nYYzc7%IWHp)!7Sb8SP?2aHVCZ-AN%3j%!osO~H@yb&C({oUkY>|YWAMhTvI#Rns zyxztbQ`_TE?2YXJN0lhAMtR2zy?N8FS*nsSu>jJ+}|6KcnY~u^_DcO$Xy*_Mx z@%mLZ58AHr4(w5OX~!Ap0}Ixk2P=ZCEf z{T6IzeG>Ws^h4-R&<|i9Ieh~97?uib@P~&*$9Vr$oY2^Da_VV;U$SbTW0nsq1@WdQ zhK&M$W&N;PV9(P@uH_o5gpGrk)y{zr)vlhB`a z@IUk8*~SN;H_3tTh98ajUi<*|AKynTg4{4RkdGwR!Dk>dDksDwE(!bq>4}TDPxmVg z{snnse`82|A!GFT3GSclGLBq-1HXm;qD9PSJ5`LRKnz9bwGP7R+6SB8HCvF`1{5rMBbKd2jG2;MW~->BnsmDno1 zBaR}*B3>m1qpygJJ|&17-4s@sTw&CIa>nEtlYb7Y%voj5reV9VLwIR8I-D7P5dIKu z2sehmgvq&+b5{?Wh3AHC!-iqGFg0gt&aL6v;Jll{9pP`m`VYf7;SJ#xVVCgS@T{Z`Ec}!3k+4tLAUq`8HutuN#HxXNUKMR|n5v`Ecjtos+)~ zKMI!y{@B~Y-a$Xw%0XY;!-M`Pzt;|YKK)sij|t0%Wx^_fFSJwGH|!rQ4-WjSi^E@o zGF>Sw8y*%O8P*P4h8KpN!n1?=^0e^Ouu*V~bEUyg>hN%I*e#G>PYLS>GVFQbm4Tf2RQSpmvhmH~z_3$ze%K?BmG2544Ic`w zcTylrjtbHuQ?&0@16g<1g1Z*{DO?-K#h(ZA;d9}g;mEK@*g3o?SiUf99iAAL4RgYR z`xo3lkNezQ^B$C3MYh9!kOVm zL7EI)_dGW&71jyQ4*Le>_4@GU@Ud`FP^Q;~>%;HD?*siq_pB105Y`K8gcZW00-Zyr z?HxSBPlT_AABNwBn*)7$f1pp74r_*Igck+6SBy@KO}uTNpbyhnj~&Av;qV|PC$9IN zV62ESBw}NHFTN}PhkwR5Vh5OmihaP|VNZ!qva$F#><;l-cFeWm{-A$XJW5+K23oxQ z_)s6@w7i$ewxb{T75bI=5XLFd6KrPv(3=LaC;b%SPkRP&4}BH81>0T_b_~WFm?ME5 zu0O-?=Z0;A{toGEANUL6CP#;NhEIoY2YnNN2!9U$2vIE~nMIau6vVHL71%i(7Q~6f zl*lV_BylG4>*VmxAP#k9xHa4p{uTZ{=HHY5o}8RBIp-l^Zt(lIaAWvWAakAbk6_ys z;k@wvaAG(j>=}$JAlJxOa&@&}jDS3o!wZKSGalfwa8YodSBJj`_vi2KiMtbZx?wbC zz;-g)n2D!{4T5xSGtNS~#xA@lh~bHE8AIWC*P{nqlOs;nN?(j?sWFR8E;72F7}pO1 zUrX%pli@?*+K{_Q{VM~rdT&-S9gKQey(S%H7V zS9oOLBg_r@kJVBAxcX#3-Wzw$w=iBoI^GZ1DOfk&&^2BkJ{Y{SMsM(M_&WSIb=kN- z@2OoB^xx^b6N~$IFiwv@OYgWhezAJbKQb2Zy5PNT{9|LT9^>z)<#%V`(J<%mCiH+W|M z3dY=s>8>BN3-v@D*gmKa+Q;*PZBGvC1;=e0#6*t@rvx$I-l;APb}3tfxirw1oG>%L97I425}Vqf$E=nv{g_KAx^mayTs*A4peh5;hDS3uqi4U31UWJyU-Tw}(H4Uj;S}ABUY|`NhER z`AB#}V4EBl-W1p^*8e#dcWQdu*fGu!c_+|KM_(ixXSpG1~4_^xxhbscx;-NMTQ z`_QvGGkh(4Kd?8h2tNpxzYI5qe+EALBEj;$sr#n>6|6rptQ*)O>|tyc)F*5jwhrnT z8-(p~UifOz52;U2pW_FE`Y9eP&MYpodSHW?UsKF?gCGv0k5z2P7)dsWb|D5L?%}r> ztC*HpnD)1QFg}Ca*T+FGc>joPVr;i|Os?x6^PaD`Cz&j+WgN8e1;)jRUt+^CWb9r+ zpP0Dd8NpZyVi=1F%<7Y?kH#u)I=Sh{A3l8Y@R5up`zrtRJ-s`e z6}}OE7OoHcX0`*HfsMhQ;Q#PjR}L!$%f|(N((}R|;efDj*fs1J_vUTb4^A8Hg1^)O6;Z=c8 zF3l~%=0X0p56+SP{$Zc+;=nKDgRd7XpAj|&db;7!VU;pGWj$Jb-qvgUf;o)J? z;BT?;sNkGUg0vj3jFq|aXRqBA!~v$ly@Bt^Zo4XcFR%~K4W9~UhmQw+i1grV!>fZO z+m4<)F6g*AgQH^v#&b}_d9vOq7g&(03|67~*!Y3~6H#LSqd)9}ge@eg9R>AlMV_Ep!`t$VbHx7CB^x31hHDWTWa_i2iJEyJ>R|T;ye)}bX&qb$+>5-d!Z@xGg`N8n^ zAZB-BAb(E|;*+lnbl$u`&b~8{%i?`|hh2lXAK%<|dhzH$Zj50Yd8(tj5 zJ+}+u1!4qZb-M+=y=@x=ekDIr%+l{o0)LXO-6*UR)(rIS`oVVTz97&Qj^TsS9pVMz z4B{5__0GX@FAA=^Ul5NKj}XtO9{Rx42Szbl_e{5lZz_w^gEAEdRhG)O_Y3*~wOM`(yOytVMtFBHE>Yi(J{QN?W_kK}PGIlq=$Fwi^|?TA zvV|Q#FX)FnB)mE3AG$V>75aY6^Yr6D$DSXYYx#xo93lFC_)&DOK1B~$|Cl_|$>fVP zFAnVT8v@^uZ?$4rF3bt8C`dIZ%@fAIT`ilBjrEyju7ugx+`#3r{#u)h%gR*x|`oH8&KIOe- zUedauq| zs)No~*VTXbDGg;W9ov<;vTvDtX6lA?)f4^d_KOq9=P^M$&^DBrG8ezlPV7H2D0gLJ zo3deDO(y`>h@aG79>Wjc=c`(9HK;$u0ZIIAa)KJj2a_ZwoSzugkYJ z)}Kssy?+Ji*-lon+g;Cgvd=lvGEU<9;I}+DR^0KLa8rNx$lWKg%IYu!2Oyx5Am zE^q3eXQMon+c$zT?ekFX>VjwQx&AI(9+Z)1z;^X+!N-I1l)JRl6VKW6_Y9@ON7c7) zsm#^08-qHdE=s@k#5QH9-YQdN+`94Oj1hI7GI4M8zjAUvo{N0AhA{%#i1JgO=aM8+kE4K{>IB%$+Ar zWgNy^!|CJk4eGu;>syen@$ScmSA{);F$ikZ_J#uaN{bO{^N`&8nVj|tvuctS7+jb36` zinlvYd(@`YQT0vxXUn!tYOC64WrKFBo~qBVUtQH+jc3x<{ea4qM0Hl7i^&+e+8X9GxQ(;!Xbyye$*8gG_I&x5U? z--0d1zF;2>?FjeFj@zkPKZZTWW)rjESFq#Q*K9tvI$P0LJT^Xmgm1v+H=c!ktv`UhEe+?ei`j>K zr?-aMrP`hQt}X5UpC8;O8(2Ab-s%Q_&2gTUGV$!Y-g5KYhwW6mP`|YWeOz<_IUu&l zF4T7DLG43Zah$fJu4wD(vV5p}uE{oc4Q*u2pq;yhdZ*oK1Jc&c=wIzYzecaMYTk3Q zU!2uFx>xt8E=k+{sq@AVx}J8S&Z*mNht{R(zTAsExRx~Ap51F@2suKA_)X4Ol96l- z_9pqsUZL;kc>0*0BYW95^jO~$q=VQ)bTqk4KN{O){2+b6&o|D`d1N@9PG^ztbWwE# zJ)|#&PV&yOvBJj08%JzRwWa;^TEBxqH#*;vPLm$}VM%AwHDcb)hb6s457Ciyld)Lz zSkt6)?U!b^Nt#x}p76zppZM%lV&Ww(29 zebd&a%OwambE4YSs zz(L&!$=RvFK(Qci5L+AnlJU9H^UTxBIKc7QTc9?nrF(sC{tFMV>^((z;~ z*-h?SvL)opz0pN%0904eSAM$(X}GTZSa*-)J9+Oo_f8l3?cUilbdS71Ugg30?m?L- z1NTNBuxr(Eb-wjhT~;5J3)_^=a*u4)-p9%rAx`J2n&++o$T5M{|=E;7~+r6^`V7)q4nyxKUSvt$kvol7@$t#t z$|o|JjI!Oa;x^=!-~45=ioGC?;XO<~Jl{)QuoTa+pFeC}dg4st8jy~>zIp+r7+0 zs;>DTA747+JYr&S9dUr6Z%^ORf8GaVN70R96m%ooNo+`;hP=pwG^8cnB3>bHuGeMP z9pjq47x%1Ol$GsGtM}GtrEEPL>GXWp_nQv#9K~B2bLsO`7Rsn))AZDb)*W%3J~wd~ zd6O??-MrZ@@2)2nB+b?p^~ZI|6Lq!qL(EbgRA9c|`Y<8``?INS4C7_RapF%gJT3 zOn%*mwn|o#g&-Gd%k=)7M}Ck^!}woq8FIru$FPUU7w1~~&3=&2UPE3SFW+R4yjTA8 z9_R#q5qVSDOvYA5+fGKgN5@){yTxy5Nx!iQ&&)m7?rGdZdG;P13wc&9c(8jXCzUIi z=Xl4_6_v5d*|j|HmbWsuUVYShAz!r{ok!nz&W^K`PW7qd;IFz(x@fxgNfWLi4f|_r zs(WOex~9%mhpLOiy088@$GP$#UC&inxnFg~y?d^nX`gS)&$05T{%TLH>z0-8t>2Du zJ$b9HXxp=|bzge@8VrJEUwL!>usOR=O-o1IQ%BRoHHTx`wOWV6oKAhvveanYyyB~Gi zb=eBNuBG4d()_T2hV}q|hOOc_?NfQVx7G`FuJum+aQ)f!KpTX5=J~XK%XjmvY@I)P zsQKp}Y?B@w=e{jHbH@(l_Ewrr_fv9HhCy(urxpmMSIIlMRdcOU98TIXzU zy|8SZbWC-Gb@g+2?C=@(+1l3j>e$Ma;dzd2`Lv9bPigy9rpiPehQHS32YXh-c1vfI zscaj18`djhE7#hdJriZ74And3tZvmdRhQVDwL5DEtBa0vyz`vz8M+5~RK~7Fp15}N zBrnYy+X(ds9Oqt?J=q`~W!7?02F12v8P0wVeHMMbEqnDs+0t{pXLU_``M=@;>~C#L zTT=#(RYuCd`Rb&0CT;oh+}(%!nX$5F;;fzzS|&3ZaOS`5%B#;v9daMibHBqjGpZE% z*S6ysx+gN^zsa+qF4vCe&FXRYVBPlrCQsEB`A}b#r@Xn(>dxwr-fx$Erj_&3@Ygy% z%)55%xl{(Xjoa@2hV4NfJm2b(nb!UP|5q0s#1ovO26|5NO_nvk%0!-r&qQ5UPvoUKO8&G7xSz@g`^$`lxa_FSXg9FU zerZY9b9X)0^-TO#mdn?Ft8bMrm3hM@d1HUu7u}|9swtkAG~3Qx%X4a7uv94Nown%SoUhI*M|pIe+RxRo zo`ZT=JHvXPjbl7(&qe;U*|q`YrVQKehj}C4Ag}Tx?~Zkj`|?-4wT;H=U%BX*pP2K0WJbNzMoHmWx0+!u2 zd+C0ir*27KTAs7={BO^r&%%9aPn8kFHrDNqL7$uURC!(*uY7HzuZR1{dgY?dsvl4; z=xlZG4oSIeBFm*iEsD6b)o zN6XH-d-EK%YtO9Lt!$8=%0y)W$MzY^)3DxncG8s>Wke6t$3vamIJqSq-F|`UaK~OZ z?$>uO^2pYxtdXZ)*R|nKp4I8Li`I49l%MnI4P_&Ly}oO)D~jLUCU5#|hWU4{G^C{* zT-P<#H}$7=X;|-UBfDxR)Gi<&)K__PpPr>MalMv@CA+4yZ>fDK?e~yp=S#;kk}vIm zY><~;zw&`xC3CBj=p|I=INrJHs6T0X4({7?ZaW>eRrO8TsAH{%>OX%a$%we%h)sl%A#EJ=cC|_&q#MeJIU`XIPm%`y6H2S=HyuBeg)M^~Yo9vhK8Me*b*T2`@VU~h)u*$4Ic29Tm4$1#kKv#DQx@nx z?Mxk4KP~(0+>>;qC*9gN)#0v<)-5&-8^>4yJzzdDUb3ulm){wmtRPy&%rfX#^6pqzmzSpBW9*X;bwXXJ{bb!UYMp62a@_FQwaz%+K54lh*Xy;0 zb=o%hDDolCo{_)N!^l7Dwzer3Ng6XZUu+dlMt=>mF$4wh!^l%b6>W5vwuuPsyi#`)EI^3Z&^hP?PqM>%Gw zH;3!>zlU~7Z4c#Gsw?XI)rRS@wo60a-HUwp-g-|lRRnL>ZbZPrP#0%JF zo@r(Cf0KJ;UTp;H?#=aD7M6aiGt!c#wCQ0wdiK~bzm17A2F{p2V;qg^Ge&)gF>M{! zX3TrXt2^F$$INxiC>y}_+?)G!zx0**DsS#j+4MPjF3Pa7yLDnHzlSnh{V1MWpAG*> z-LxD&Pv;hCdajiZu4TRWtBdJH z*Hk)=j0~>)?XvMsGE#bNgLIF39oEO*!-Fl|bL+hFa;|(!r)`|PA{WTuq3=cq(MR%Y zOxA3SA2SYZI3~?8?yF+}XEfY@zME@%k#@(7^_(8xYfEo`=$pEOgFLW|0-z zIk~S5uNJf?V>-20_O&)^td%jO#;R(E+NO5ealjq#ZA`Oqz%W+XSY+dkhwqDfN633g z#%X)M%2@6bgZIaaca)6j_HK^%-HrYBzLIy?Vcj)@}2C53GCl-h1fY z?>2^9`Y?WcNVj*NyvUb4dOxc8J*4kW8H4T}aqn^+8r-k(<;GPT(+*>@jjc8wd#|8; z`mUd`^u|*gS8d$2G1jo&-yMIx#X?(*PtMPH3TYV(YvT~AsCmYFdvZg)* zdE~FUNZXyQ$Jt8it2T@4f^B4vdh0iPPFvD0v#mJA*koysP6q9`AK}&%=9~-uv*bjrVN43*%iG@4>)(G2VUgUZ{6xyh{S> zaJ+ry)OH-K+wM5;+IYvt(r?>)oa=Z$XNY~Xuhnzp#d|J&Uvlv8}IYa~;J(%h=010?<`(mg(KCZGQD)|N@jQ+R-iPvT zulK*b8|EGUzV}V0s^i*$*aCT`-Fff9J0RqSc}UDnVQw+s0hl{6cl7Opjq7T-1!kxkQKh1M!*>!V!n$y#Krsf@kb;sQvZVuOn z-v!?a{aN^N_+c<7W5VI#z_3TyIqVRg z6`m0`3{MDahsTA-gw?|;VZ~q`%O%6&Vc`&yW#ayY?jOw|W3J0PgZW{UpE(-M^0)Ax+b>*D=P?{<10^parS7iDk$5ATP1AM}hs{+mmLTs98}J<@l5yg%)IMY6W< zPt!r3t9Mhpvr2|qlCR!Z`ek$Ubt9%<~?c_X-aS4Nm<-NVlgSOLm`@E}X&Ls21YiExQ=JeC{%&Fx2 z3(Ev;UK=GJrt^1yxHsGr{t^Bb?g+PrKZWbVHR0;;i*RN5QMf$(AY3-)d-)F7cf)tW zrL&knJ%98))URhQO!E+$2kHyq{9rC9^Lw8e%#r+&@c!`5@YZljI3XMx%(?uU@ak|_ z@EwF#hW*1{;pM@+hc5^_glC6sf;kem2=^L`s{3DaSP%n7=2aBuGj-wd9Q=j%Pp(*wP&PP0?IOKtwAtumL~JA%3D z9+mkIUmiXjeir7=nLC=leUHoycuF`cd_7zqt_(j3-wzjsbAoy5j}Hfi7lqBjI$`Nx z9>m`U-w`__yfGXR_6@H2(lNVD?l!sCm_0H#t$i;F&kx&=zt{Zi%rm<~*fG3#{N3lB zGsmrMn}&76vLREFjbyR!U6Rq?3S>6fO-{P+W5fPo$MN^N)}2^)G~e~InOAbD@UXB% zAomsxlfm3pWFA>(t}AnCnSaXi`fy_~f42Q(;vK<&*n=I`1}p=95RnB$j{+JdDq6f+jK7-%&sHXysOK;GY5xv zvdv-RU2n&G56HRvGkz-h$=BhBi_P#=)Kzj>T!9=WgZcFARd#4)E!)$3KkQHMH?cX{ zoAN{MlDYC>4ioPVdr#c^QS4K)+I>h*Ops4JM4tGy?w!Buk3a0Uee7e`XLEQ@olIxD zdXJfX!Y*X*vGKh7%!XSxkhAPGGM0^GN#>Hhs|7NdjptZ$+LAqFzB}K>Ts1s8^t+y8F?*n;!MIdj#5zMW3{#d4dcKWj;xq5cyZ>((naOQsXJsGm(P2r?) zd@zqa`TF|cw`JumS^KJRKp=;93oi>V2`?S9>-4UpIq&zJ-g9(}rQ@)Ba2^?bXduJM zbn;z(bzmLE1_OHxSsVwde%Hx)BWB5b(P59OLn{;%P zviVy0d^jVxf7gF)I4H>9PGS47Wgu_OVYqfM$K{IQk%3NFats}@T%b1|7tG^Kzw8+H z33Sw3!Y9H-!F~KYuXLW(#DVyCd`fNIycbJnK5X`>?_IG&?+NQ=ZtwlVN5U21 z?_q^GD~!Iy@q*0zx=%PP92?9b`{wYrF>jgkmXVx4X3jCAxnEzLIb6x`$A*W7e`N0a zYr+r0m&2LCeBj501H(>Xvq1i?5Ecvh6h$oFyGL^P#>{(vO)%%~PXask%Yi)pV0cS- zeK;b#GVB}JsxJ@h)#rt6!ZX5?!aCzQ+u5mW1a>R=y`ZLH`=Q>Ir6bHKMMuyN^8z`^FYi1N z{9^X)aLx(-AK9l(@N>yXz9PT3^MCq=B^#Hm%kC{?F8Gbolr}qhufV@=JiyX#wzfGt zI#0m_gmpB;le6Rk;?C#${F!1A>~b>3_bb?l>_W1P+#?6YAx;S&3ZDzUpYV%tQxHd(GdXAUWsj9I z5B1}M?|?idJTq(&whCK^?E-t6&CIrbVPH$YEYt?4N67V8goA>m-}?vq*zWs`**o8h z*ekHvoiCkT0=t}EkiKQFv1h)OazHpF92t%Y^vLVS9GmY=yl#wr)@>_(+y2_|T-~qA zH-6=5_u!h(3)=+#fcsYtYlh{+LxS&SC^PXDaTGe}y+N$x(6Cz&A9-eYe0X%=2eAFw z_2!#0*RA-4XJ&pX^I);R$$he3{F^*xbBjNiN5q^CV(aE^pwIPbtvHS^=))4f_s0$v z>t~Dd+sIqCYU2vt?KjVjd2rM(@zaxn`ey!|&LK1t6+LtmQxNMEOC@u~RUON|5OXDu z*~aFx?3{?^6CktM?B>4{*A?Gozq7a5z3h1J*TXtl&klF2xUcw!xp+8RH?~q}1lvpL1#^U3eNyJ-ds4Qu+fNR@<4)(iEwHEA zwjT`iQDykK)8~$K7Tv{1Aj{>0JpWk`*C8+EmrP|-iwC(M@nZ5_yvTjhjcDAMOy;BT zdB}gVk}g+9Vo&^?#(aGzMSP0B755dlGS}4X*p}zvxp+RFndj#@^0z!!Hbr&3dgb}6 zAKHnz!_-4_w`xOY2eyg&s?MwL>bW+@u3;1Dmt%+P4`ws*MVAQdDE({NF}tdL@oSF9 zvaz=cU$${ZKCR>Vx{nC_?Dkue0s3h43G+4e_lY_3Mfu?R1N8~>#ld#-+jenDvO_+} z5$Q-z?2^6b8uU;5_GkAC>QiDTij|8Ah^4Y4$uaVbTw}NDXETSgd7Rn5V!G^P@d+~0 zHzwKB?CoX3N`a03q_9bNc3{`<8QA+r1@c$i+BZ6Ux8oDxli}PT&Mn^kjlka#&!PV; z$?&fS`+Xltoag-T`9M!PUyS_I;Zs4Z{Nuqj$bGS(H-(b|A4I(THGwZe=N=N|QT%+r zuvgeM>@xmV1K-5=8nzCb1wP7C!Ulo<5Knji%3-mfT$GR45MA$^RpvA_pE>`ceucRV z=@)TvI_9)M&*&4SYs`JkC*TXvIeZ7t$NZw6mH0UOLd?e;j$%EH@#sq&#=q6qRfF*r z#(U^LR7X0_MeIX;QV+EQ^-dgU*cV6-n3HCH(2m4ev@P?jig{~y?0#*L9M+e&doYI` zdCLwYTlI61rDOvcs(*-$)o+^*Whhzn=fI{TN7-v+DXjDL*?RUn?zX^=W0(Fjhz~SQ zVBI!0-VnzQe-F=dE;-01yf288vpwYz{o^Yz>W2Mp^Z~iah9o!H$8?2Q5IIRkDg$;U znMl`=hjb8Kv|UhcWS=<=_XzYF{ib}`0iExbo>ca9uKou4m+tpW^g*y8KsSrsiI<6I ziRZBmW=%0z|6^D1`^4VZ0%rtE@j3F>KBy<`1ajH++D{_w_LH~= z_fmb&4k+C=v;pXTWiZ4x4DAEI)k}5=J45-gMU*A`Lw}62)z8a+RMz@sl)HWzI*6}A zFKrRjWAVPBeZWsquRTxt4q|(rEgy69z~*3+uy@?6a@2=po4equOIYVvM1_8kk{;y_EFORYz20KaYpvBR_#^O5rN z`SNT!bza@q9>_UyaWQRnA~{GNl8fw0GSavKSnn7EdV#!jp7Yt>Y)CqUtYkN8kFKe0 zf}Zi49ObWTqw5Fx5ND!uq{)VLJu;gtC+Fz_`h(v2OQ2K46zCZGMl3<>;EJH1^V>nZ znqB?HATD%45Pt{zUR?XL;nP7M=O=@B_gTTxHv7*D`a|`LN>3lDxbYC%?f=p&9&B5W zX?oIO`|CgCf6Nc!>|*2kYxx6W?&9U*B<@qJhi^d74&x+rsWQ`l>{;;z_yGJN_P;Xr zy!2(%5Af`Uet>7<*^r~2)6fT?|HSq@OYtGTfVxF~(uqYZ2IMpPyW7N>#GBO3LZ6c9 z>=N-nF++ZZI?T7A1Ly;>OL~G|%`c@x=pu5Pj-|8s8Sf8tojxgYod2xfiLGv2D4WaN zj@tU3fgQ)bCcoL%!|&>dVUyXL2ln$p;rKvikzZsPx%S;a){%K+A9-!;$$i0qu!Y0o zVaf1_uw2jwOopxz)(-j;HVkAsdAnI4d$D=gBy1e)cU)!hWAgpmRfF`F2@em81|6EN z-#&pqPW*9{Klzkj`Ii6F!~23h0r$ohKRD1$%le{v|(?-^PDaXZW9FJ!c}O<;Yj$Ck|!) z^%-lbo{{%te8;g`LLaO#t&QFAdC7Qj9*8%otL%Emvpwh#HoraszA+u)y!N}Af08aZ zGtdX@x(@~R9Q#naV0Rhkq7AX7*gM)7Tgp1ys{T70%&}})_L+0pfBbklhRsUfuoY{Q zYIF59v`g(&n%Xy6!8Q}K5qBVm=o&hPjZSXe5yXqdJM_tkfBZ0z=VT)pONNulVl$_Q z)53>>*ti&r7>#i$;xu9~^Md%yaRbEHPYCuo#<5~F?+)TK9}9GV>(CqeC+QP1Se!__ zNIzy}uRPO9jZ@QCWH9~3*BJU4VlwHpVxz`;31hZx+t@xz)p_fh-gEcz0l9>l4*K>YL|ebv{EoF?VwA zs4w-Q`PL*k%C;r1jd3AgR|;e{tlP#OC)bw<`eE6^meO#({p@7NLR{Ro%4z!H-idoh z;}Yo&;~K9Yk4rRGh`cr)@sjYhAU5-va7Oq@AiL?KjzJVxIUyKtb98XeM+UaFc>Ak@ z_zGPH>$dYlUK`lqCx3Z)9r-ru$w!3~3*L+=&j&%13bcb<|d>-*A zx`K|--y$Z%em90u->-2!?0CLY$1obdNI!JkBLB+RMc=B>2P{_O`xE-91wq z(-r37-ad%0k@?~-#@UFqkm+J9#+Il{WV2X`xFnxhThSJ@G4hb@LAJ54`KDy4y3bZo z|3RK=6OJQ4$xQZ$IKOn+BaQD{vg>`fmmD`wfkpZq4bwJ)|bn_FAb zezYaa%9O@)wI%J33`S!a(O%h)Y)Z0QY(*@__>ZCNzB1zw z$xC)DJDknSj{Zy_TPwH4*2Tri-cy6IFXXrJ8RYUCgLwN%L2SMFZM*%BCBL0Z2S`hr zbOPNljJLD<}X{X?B|clreQM|3FN197nji;WSK!+6+me7cyK z7@Am`n3{CN+@$aM(s8aUkMibTl#Q}b27D7J2ix4Y3$F?KTE#oWI@sH6@Rb93u78*PY}|Oqmj89)KZtXX*Yt<}O)}fK^xp<@yW`2P zn7CpT10lb~)5So<$IlGpcVpt@ExAknlEdUExy%+93ulk-9vTOK{>1Yqo*Rry65ro$ zjCK8~_VZa@8l0yb#a5&>Bl!=ScttoM82=;At}oW{I|XTrQ|Xr#TYqe*{5EdKm>pxz zJ*)Y__alw(7jMzOE?!O+v)>O5VrAaTVXNz-6(iT*@ti4WK*8n!Z7{@u`V;cNymP_mUJM82|-;kP(S5fgPBA5uJ> zY!-WAqf7U?z?Yl~#!oI4Drwss@Cxv^<>5}7QXL%xf<@YgHf8-I}|pPe1HS#T^{lb(>CHpx~cLyXxVXUHw` zO+1B+6o(Lx5StLAV2@uM$WivOn8=q0@ek{^+3#5AIUh}H_?wMiSxlC=HdzYt#5TTw z{8*ARjv;I1l`N9)%2F~Ce+;&fo5k;98yQ8$N`su19;`QA=_ngM%HIRoSQ}qof>?{V zndc;qCf=ei!P0Nr8&~5G@C%0T0FjO8pR_BB$z%BK&v9gU@!Ro^kuQ3{vy@N$-Ns7s zO~`Qh*B9en^eMZqYlHiA&&spqt&EksXF#9PUygO2wCH{HM!e40FmZPAJ$5=jQG4OD z>Sxuy$pPc{$a8&>;tS*-So_pHFOc6ig+=BpGG~qOw6Jy9K5QE{4NnX!g+;?% z`S$$p!)1Y-C*w~I#Q+L#7o3TT+cX3vRj;ntTv|Sr(?WZB!)wd-xzKQw+3?l zFTuKPWWJ?s-X)?Rj0s}Li_^opaiZccuIG4qLLBD0Ag*ueTE7oh2l0OIAc>h*U+8Bw zc2i%grMQj$SADUL5zFbjN{zE$Fmb`CFW381d;|8mepY>}^vFwsn2zTmwywXFT`s2X zeWG84?}RS}ea`wqy&wAapbxY$bbX)3x;KVSrjzC5IG>CxFI(oDGx|-Q9>ie8UifbM zR4O~!&+A&9AnkFcAKIaklfbuv}_!&2Xk@n7QPwY9&K7)#?S zV(#>Vv6puQeg1Tb_uKgf`e?=djUAF#zwP!rUmnF?8qaxG#@x6L+rIvQF*PTT;R8UP#CYWIm|%RT{JW># zqvfl^(ZTV23%&yXk>ym|HBF31A zONdE$AJbBdqGMddx{Yx$HiZs?82EX?I2Q4h`9ZveEcR|FJwWc0;odd1B=>I)L%W|0 zr#r-y=n>=YFAK)ckln7Se^*|_+D{DPCE_z|{lml1-e<3iS+V1{3Sua01^vimwSHyq zIvLXxK2NgxkRWEx?)D7HH*&b&*V3omI5<1IGI0`N$DSmY$SZQobzIMNU7tK+S2ji<)-vp$YRbp2}Lhi=U*gduY@1%cMb9 zOOGuqZTdi&DjQ|3Y|wJAUXU01epr_B=vk7_o~Lp|b%TA?4^7W?T+j3Ky!aLJB){&1 zJm&+F=jt7Qlf9x&8&hy~xHo7g>^1Xih^2amTp!-kgZO|k8kYt#Z|TXUN8><@4|#QX zM>sopM@hWpitxk09{*-IFMKq-As9FA{ZixO9~-2l@I+L^PRMa~EE#MpA)O-TOrDeF zVk7KWvQzxod!^(z`5#_-aU{$3p{_W&;{W68 zKEURxs_uc4Mi4{wVr&$Ey$6Pg-9r z!2R+E+8y7^AH)pDNq;1-iqFHV*BdFmh%@%L?R^oOjtb~<9u3ch-=e-slUtjLK;O)o zR6cq6EfcezpWotqxps>j3(tQy*fXHXc`aK9n+N5$@LF&n{#!Gk$$2dWk3Eko+ih0P zRZ(gUzy8_<0(DZc0B`a;@H#I3TX07(v0I09S*}IGN1-KnEIb$WDR5uCvb+}C;@t^> z&%p0bck5N(iO<9H5qo^Maf`bb=|I1$;fhE6ef8S6l0}F$4+M0Ko&df+KF72A5xyGW zSA4*C;fu)Yr+2^nZ~sg63(D*92Y$f&Vt_gxZxioV4>jKA%T(*pK26@Dy!^HzUCT<5)Rvc+qii4CGKa84u!U$J#k{&fM3(PMy#4)6^e8zTtOR$L94q*YZX0^;zPo^DLLJPxe#3 zXRPw_czbH3#hW7W zu)b%v0M}a&bX0Iea6s^_!0W38`d4`7IG3-(zo4gaIK6A#EUmk5z<(YU@R|A38wDE% zZZ{9`pl#GF)iTu|+g=+Sds~2W>1@6U9|srnR`9>PfM-+JhT*~RXA1b-Z!JBK*URJa z-r{#2jhN#4FusqMuWn4!<9I%cS~5=O=j(H?n(WDgPVVhLFJF$dT|mpP7C83mq4Nd$ zX2jbJ-Cn=Ut24aXt6i&!;(K1qZNY@#Pl4K}^LRnPYghN=&G6h?=%I1#E;)esz_&Rn z(C;K45F^Bg4Fi17Z|8p)FY^a zPnnyh`2I}95uBYi;7vXckl7ZMH z-&ioHo*j9B-{&{W3uX-*1Dc<&B#!xh-jCSgw~IMq5AK(v@Z#xvxe9Lw|Cj&d`0#b` zzF5a^$Ng&u_@D3Y*zj~5t6H}t9?TKU6il08+6-?6u4}{NvDWLc;Gy9D;I817pyI-1 zGhEh-31Y#qfx4?)fHyDxuMzOGtvREc>EzuzhV z`y71WJ^tYdoIz{iXS%fDvHkaY@rm#AU4Dn}!Qpg2ZFqgaH=%>+Yx@>5f7KM}W5?a^q`C30_{(>ZhgIyu-};r@;&1h9anCmR+ZZUXMm+WzY9|$k z5rp&LSUNvd-{wi#52!k_eZtwsf^mfVz1R8RlSAcK;(+ssx5ajPhsY=5l%(?!w^qfG?m2Th9mo zm|udUc`5WPj-<=+9$w_l^Uzlaz8owQ;78oaBcYG!cRXjEKJNvO?HJ%~9FEVI4{(>Y zIUef`w-3ebyQH6lmwiN_FA*>DKGZE~Xq-x8)8P1A-2nfp&&ogSGk6U4O)LU^miW51 z!{1onyt7(&xAe}=>-qEQsdPNfzBOEU z9_-YtE!`ko&ciVt)GB%S#>awzvBDo=8qvxx1(O3jZfw!!G_x8iFI``X`^KKFP_K;d`|Il z={@_>YBu_`KF|Aordv9`@`1`V%1go+yuHx5n!QzJlkyPVdv^bho&nH>U+1@A*P$dj5*{`Yc+VKVe_&6JKL;AU270;-&NB zcZp|?vtx^&Y1pp@&h4gwnnv5hZmk;Bbtt^HLqOZp`T7KDU>Y9xiwpRleis9(_qAex zWAFQ0pGThOJXB5HIj67XSU%JK@yP9?@3b~aPO)`hUyM(^ylOMnxgH(h?b8C>{o~-g z!1$Jfy8ec~1bnXlz`VLOxG#7#FlXei^b(y!AIW{?xoS~7*Byex1NA%Fj`o{5>qqtA z@kb5`^kMTJ_~m*H@fv>~pVF?@VS@Xf6ASp&yb{`9A0o}F_mKa*N8q#RQ+$rw!Pntq%~6RUaRa-K>gUypz3Yo7X2!HU5efqu=60-n+4!6w1_-L+#@&e|qD zI(#QOT|Dv`a|Lq;?t8AhdwQO4U%LL&>Dkm{PtS`1_}@RdjrihNhy#vA#ezxcz2V(E z9%2B$N^RKne)#hGF!v314jh|xf)xYa{Jep4^8O6(_q-aq{Eh&(^X2(7deZSaPo_K= z>re5!7|xG7C~(b4Jik}4N3dHUkK3`kjyaE}dM5Q#tr4i*(((KfwM#Wmdg0~37^B~5 z>bnCzIu17u0jJ}39DhTAr^Nx|RbP#|D<7U$ZcL90tSh3&e-Iqo^~4u{t8LOfdcbIN zz63ABe8iLZnC9h;;6U7s!|^%(C|ykx@>*%@0^J6lFL;b6X=8lEnJ?Aysc@KUC99Zxer?#It5a-tW=v_Qo-Ux2Rx3n-{goanYyfDxc z&*wfP&rL4PoMMO`SQNo zKJa;XHsXN&_PZPxI+?dUDRAuZJzXqrIgj`r*NTJUC@%`{UAKRw- zm&Dx~Be>dUikBeXdMqCI$8s^&d&OMwrme$>J7R`j7XFF#8F<|K4YhXK+Za|0Jw4zX z;&S8roZv@I|F5xc4wwtBuVH@3QOz6kNz4){fmKILLVm{UnOv>&g*%v z^I~u{p2p|&IxRnAFf{N_>?jaPxZC&ge_YdD-PLcEJ2)Qdt@cx$RV?#ct6r>5%+h*Y8fdi}S_<4ZmujzFPe_c%An>UqBo1PxvLTbp8x(zc;|; z^g8a==gt$Pw~f&&13XTL7`HUI-pC7s^8&iXc;_e5GUfuW#d;+^qcxv=7wZgZv|9Tl zr?tL`mZQ`9y;pjTjyLb=G#X8wPIsBBxPnGA=lkJ#Ft-b~ulwe^d>hy3i}blZ-}m4b z-@9s{=iS_=AHjXG$!mC_>apTf_3z*i`&W5@eUm$gU6l{W0owRQyW>N>9AX%+iN92F zj4y;cdEDX@A4ja>Iq)86YB5M}j2I+F;a)L|hHp`Er`6*=N1c$bAzu1!-i5e|qj@X# zxj5AC@w*&PvAMXDe%3GISQd9XMvfX_<(qLBy`D!^}`p6j{dcAf>^9~-Dg@?pdk zHGmre@q!Lk2f(#6#e?CCj|_O|G&!$ml|Y@Bjt8}2>!oP><+|&_@Rw+J@qr)Dx54i? zod%bS_#QmYS66=!vv@JqxblVMGkQ7s>v&yVnr}pxt0h-`l|Q4dT-Orw`|!Pbf*9xb zt9dxCYXm$R$9ut`pY#721wScxB)BKQ`{mK7wYDD3>G``yCk9nZJv7%17SEk;HB}m( zUqi$5Y$^`u!PzkQMzBUuzK#Aw8h*}zHy&G{=Zo|IXbN76F^K1JyjrJm$}^L7vrb$H43<~MLlmj@1rsKwLS}cE`J2i;x*p|?t9K_{C3~#Gvo&R34Sha#F2Ow zzv51uTH3rcx!R-}rCMk88q&r3T`mmtvGB7?qw7 z^_wIj@Ni1@RNS58SW!^Jzd_rUj}e3ipe{VhZ0yERj3%hvbpvx$w-5 zV{874&y96so*t)X%mq0s|B@c3hj^U#2M-6l$jJfD=7YW#;B8t8DVcBj#PGLXG8|5O z;c>N8T&{0!yTF{L^Tq^ndESim=ky_Ne=@L^o$)$*fM@tMs|0GEn+CK!Uc%>m1wG%m zncnBCw7_LRm-A;v2V(+tMcTId+Vw%=LcEA0X@0)U#DKqDK0EIN598!d1+@4V0-Ak| z;2XgP0Z)cMqfgWQ4TJT9uLo-e;(+VVEFIAQ#rJ#{zWb0s9N`7=obbH9_sGE-Nt~sNZ5~r)7?i%3p?SpMPUB6kb8n#hD*W-1Xezia! zjWJ{_=}AE4+xO100Cs`OPO4c5Kt4w%5z8s6HUZI_~(|xxmqQ7pL;A zoj>Q7CdQ-k96B3tYCk;R=0jq?_W@lFKCA9~-ZoI5S(=|uF7GS-ZCqN%OoQ{ut!1d| z(%^1mnQwSzK#%`8puGq3%;m4HiD6FAJU0jC2wuN0c(AKKsgcr0yiRkeI9`tYX25?5 z758#yJq~kq{leC`=y5Yg%~!RwEdn}?W;5Rp4)DBM9KH9WK)uzP8hter0=3r%173mp z9`3guYHFbHsGsknf$N9O6{riU)8Zy;c+?B^$MAe`7Y^GdpzCRNJf~-}`rI$c+9sNP zLU3omx8?=uQPCGOEPXJ%nGpfq{-uCk|7L*8t)Chd>>BJI>>2D8xaOn&&Ygnof-M8Q z&#V7hcTIPE&*!1<iFq75^r6DY*_H_4Da@?i7Y1L`l}_%Rm; z`2FI(62k|7YnfFcC>M8$ax28w0 zNUO8;w(zR;l+eX`OwP~z@LS~=>HuPaSfy?uhVhBYV;2h@b9IvH734$l+RGnvuFAtv zdv^Y+rY&B$U+|b;<~;Ly+!j2R1Mq)Z@7`E3=8QkFgx|7d*VlsU`GNcvJWr<>&pgD_ z`s$+mM&63LD9z4`q}zFwe017f{YmcuzXfpqZ2{glf9?$Mzy5>!gTDp*n10V)4;#+c zXGU-7C)Ve-WI&JcSl0^ZHFKPn@Obcd&1;f1pPn^zQE$c+ECD2e^KVVBaYt3{Cc%l zv8DW+{LKIF=kVplptfh8pQMMVydB&x7vb-yP2hgKuePCo65sQ7)LqqG4acV6US*9XoS-v-zJIM4&nui@428#w3Uy|^#_^W*vOG(lbO zS3iw>u+|gJ5g*=KSoQj;|C!q3c{wHh@kn578n<){|6LCNkA~lPRzUyILFKQv*VLo$ z#Q=UyU0aODJ}M9sX#Rq0bh+<&uUk{vqMmcRLtxvj19Ml-uxX&jL5!i} z9Krk-BX9z)kPF}*Ubs8}H}Sv41o5C^0q=$1jnn8`@j$OPj~m~W2ZR6U@bX~(n9dY2 z0!lZ#?|FP#@VHRh+Med7zo)D5|NURB73Wsmsn{d_xE2A3LoGaKJMZy+pYQwat9_O? z@ZgV1@y+Ob{H`u8ziR88xE|yHYM%vzzy5!FiP%()P}G`7XwX@uIe49J}-okVBJYIh&FrV-{|7KD^-|Ly7?e!STq5U%# z`S0`>|FqVR(D<$yMYEaLdM3>EodZ5C&Bw!`19`jZdw&UT2<{B%`p1LGfxiKY=hX)F z^SOqN-fSGNX1i#>=g~8R-_;m*2yhp@%BOL?M7`nq3#|>+Gh;mwZ$v)|{ft}jyPlb) z1HL-Hg$~EjxEjyz5YXy$Jl$UK7_aXb*v7WBJnhfd5i3>?_&sesc7a?IXWl@cjhd|e z&}&(BSzJGD(6qSs@AvAk_7~r~MkSuFnzP?T&*OA`Njx2X4o(-}^ilG$`16j#>E`R-DK+CH&oe&haSM8LxS38v# z<97XRa%J3Y&e8hC$GDg;Uo|uHQ{HW^%FE>d{2FtazUKv)=d>GbXa3`8x{yzgt8qlj zA#EQHk9b=C;UeA)Po4(FUAzh0hUaK^Y>E&0;p2kx!fEu@3;$mF#>WJFZTgs2Zm-3% zPxqJd_D)~A`+8XR4QxYW+rHKCKEq>r+iSGAZ3p^HpH-;u8R)kEJAFR=epJ`{F3*r} z=)=*E!;9zpTf6Fa|16NF$X8s4K%GF%K9J986MjU)9s2HHW;ds9u@U@`o8*(h? zgZAgKAqM@YVuN!@=hNx(JRT7L#~7i_^(@i`bORo5uMG(yziUlP~zad136ulGls(jST2)fMg0)FF-ncV1jp4@wx`Td~bIL#Eq_hlA-ux)Or+@IhJeRLYAE~#>gR6&`es?V} z+FRclj;Fgf4mRjGo{!3l<+Yl_@@yUry`~4Fzc-Hd!vQ>DS`jDEjr2h&i z9%h_W9kjhBKflEo#N{; zbHJ_Vy{?`r7TnYa&v~u1eR(os1nu6!b2Pmepm3(-5bH#v%tcHlQ=u^Co z->mKARj8lpt5?&UIsJ;I(fQ%H9M2Y))7!K+PR7rCmqUXigE7I;!S{k=g0X>XDI69Y z9JJgn4)AGwPHB36%{GDWP@ml-pzHO|uN&BhuLY|H`gQCdugL!Lj%fJ0&W@fRe9mw8 zTY2pCxmxqSUBA2DnGFKmEw@=V_-r69^0sCRW(;^R)i)_;;=wo{a;gi1T8pYrhR3dk zTKb&tB6j0#aooA)tuGX)nHmE;m*Q{!o-sl*=#Qb(jUnUB7&LZ^!zXk-?fSU*j6lPi z1Gt^usXj^i#~i`+<_o?zcYYSARh<)@8JrfJ6qsk`8(n1{nwxmP?Ta*z&1d?Ihx%b0 z&%>qHORM7#K7d|me1cbSJ$}dQZwCJgo)2&nzQT2U2potz@g%;**F0x@Otp9!f`{2IkeHM=!e70Mf*=v;>9Gfngu|eepbUYpY?O^v{*I?&h z#|}H?Z*T9?{GDw&-h24@qnloKpI_hKes}#oub(*+<8Mj;G-rC;DAH5liGg zJRN?HSfH2a>7e*uT*d#T)y3TEwRGO(UZvG>ymK$!^W10xIiNK~_?&i7P>=>x0x@P?9 z&3aa~YPk@-PFy<1;;JN(F^wNnf3FGWZyWI7i zTpEbB5kJfaalza&&rS?R2RPsS#rNj(CIRn;XG6#FSoPbO_k35qZy;x{`)*q=M_xZ7 zz#aOv@jZUw8S@T!$!&eLy!46Td^)<-&0+vlkBir;uZ8w*ds|9hm)9BbJ-z$fIZ8&C9vd=hXv@Z0$`^ahSUJCJ+wZ0HjGk5x}KK7SWn z5#ar+0(@RPj>~Bv!0BQF{x`m%VgsJ1^J)8Ax;5X$_vS=9UmneRCwa5^HaXz!@N>*b zbCXUpU#&x@<7hhl&g!oGS^X9#1>*wV4DGJ>n%AzL>TgR788oElv*R9~aDhLgr)JK8 zmw==6($J{u1v>?5ht@pl8`o>X$FTnMW4U4g|3hz#dM7U3D!{Eg6g~`i%ulgK74O@w z;IZfV>$JHzfVb)U6N9sYp9B{MzYH$vaPi=ad)LWyefl2-X9On%-Y16a6NoE5mybiw ztHTF@I%B+cd6qS+2M78fd2D*wt8S`?iiTf3ST$HBaIRMf)ZCW}%6Fe9-W@$Z zD8S>qc4NrcGUklUKL(cvw74-&kLw?BOW#rH=NO5H*9W{Ju|W>_b=UH5=1%X&62Yp0e#wIa z{qMYKUiBLRo~Mh|P-$ge37+SlxW(VDLqON_%GFud2*d<+Qe2P!@$rEHzvj%~=fM@h z_`o%e9}FG~o(!G}+&&&W9NZV!=B8jma7}P^;5x}Z!*!8=6NnMxf^&l(1pFcKL_NTM z+Bf|;`f$`{`B1oDOuIkeQCS1aud;4eAH6kR)3PCrR(KyD+Xe{b5FDL@{ASwoMxb}<&E;g z$`R>uI-CZlDfl_Y9=Pv08r^oh9eRb9G1iS;v7&|fVBCAnt;fc>@z2+xp?Ep8m%Li9 zvgCIE~Rb8!xd2BwfADG|jvevkW0rWfXmq$M>|L{Ehj#KzGJmiAMe0i_E z6o>~e2k!qP@cPpMUgAabvdg!pYsH9G)0f`&yclw8(E8wf|ANoLdpr(a2R;;2@T53E zU(@9LkwS56)jV;lZ3`aL?>v7TZ2R`TVoKEm#3OtxR{0Hb3Vdu|!2Z{7^t;-a*2X|F zQZB;xSF@(^d18*M<4(`xZ~DD@Y;bq$ukqM`wpU*P_tnbryZDXYYppRqr|rMT^ZY%x zw1V*fwdTTouc`IO4{F`29spW{cW4Y6i^cKf-P1e9u5pahOV|IU=VVUzGCe0xEzWOq z0NafZXnTDO@?P9db3Gf-U*;!IwXWNz7s9;71^D7;0gcZu;2V@@^L%>dc?sgcV}ZJ?{z|Cr^xcRLv_6f`tKU7a7Km@Hmxlf= z|Asc!m#*H5FKOlxf&23X`que7z!z5o6(`h3R}aJqUdtiDDZvH7Rl%(RPJcS^_euU0 zyc)a~xb^tOz-!M3&jz-g983x-E<7~&p~3eAw{+LW|9$?($wh&D;G}>Tv~QrlhsQ4# z>GQOIJSqEKK999ku4AUBQtZ=Tg!2y#tRdJq5F=@PF;hOpr+3_WXO6SH%UTBglCDL; zv)4Dt(`ox`)LU@A{+cZVn%{YzFA(dUe`5(>)AGgB*G11)?&-drpJ^uCt4F4M7W2!z(<6hAdG2y>oUO-()&loEXa0G=`N*H4*KoJq zIa*G?jQLOR;f{B@Ix1eq$&&+|taiv7#yR*@{~x}^skaCC7nk8<{DoWbBVMIhPYdW@ z`uK=|WL`8*)~IAW_<%+q&n#S3-a z*8;b7U%kZ`;0vk0R6d|4BW|eMh!N_#azDLUxF6T+?V|6ET^a+=^Yi7K#`X^aTBP-N z>b{P@&9lh?F2>{Vbb#0CC0a_Ywfs%mt2iHD<8(Z)uEt-TD=?QA z2>7pycGuokhto%a*X8c;$-raVE*9vwSS@gk6a1jJl4n3qULWx4`SWV9H2ts+_{d}R zLmL0Bz_l!^Rx2M^AoxBKZ$$@!bKFD!#JT26& zbEUt{A)MX%>-1J>Ew>(v2mBh_;eYSL_dcU^yw5Q&y+^+dFUPzs-lzTWKG8wHo9k&s z>-FJz-kj$F(2@nu+5Y`_#(Mf_cOJ9YKx2yo^et|pd2trdv3#4->#Y}oqv%+5NBt>c z4E~~VeI9-6eLmCY_}muw5HI>3{X7*LN~5>_xp)L(MsX~jZBcq&Ou^N-`|JRZL;3mQ z1Mc>D;*i*6zuMnfd5Lua@)5tSfS;`y@VuA`9(&#K6muOHy!~R}c;jUHUd)zzIhSHN zF6PbSWBz>AUU+%6?%8w3foHN5Qu9`H`VF2Sw= zt#3^Y{^tp(;qv(PVbYFzwbcmKa>oa>r#yh)jQjiJ0NIwl8rpMJ} zj}G|zG`qau#^Bzd^?=;JKaex#)X4Qi@!TA9 zy|B*9T><}}AH@IJComrL>Cgu1vg$j=sv2yo`zJ)-Umei$#=JSfw=`GGnd+CMhj2aL zlQuWU^hL^*`6)bfb6LKww*tS)=RsZcgTTDU*)+Rt<>~kwtbNoAZ@mnz#?dq!JxA-& ze86*gCBP+KtM8!sd{3+U!~*+*^ZCGhgzEL<71$5^D@NGXXZUP>MD<_ta&WpFf(EUg zlG=wivOdVZ(4ROC*NF#skdCEic_sLt|3UZSdK{16%fF|ai>q-oO^>JXHlC)p@i~;< zw!PPBbh*js;E+J?&3+yB?|Sul|8zcmZ(F&5oW}Rm_nqEz>!<%;-!FF1{Ps^AlCy{r zVi_N5d>~%>jp7_{$8Q!N{eCgku@DO!Kj+VJc3$Xl=Rs}O7!U*Mx-@b;*Zt+=h!4hx z?OfLgMg;mZy-pX11@c2NKu&3V(h=eSO<$0c8o$-&OLy=HJ@)*lKpoh4R|BSb=zj4+ zY$&bIr@S(to33h5JvrAVPrg3To1ow3fk51#>uG;|5OQ!a1l%__UkKcKj)z0D@pE1a z%wM^Ct*0W$_2q~?G@kr`k2_1CmOEQ8doX7pwunb!kURk|@P6C4qAn<===reTwql8R z0^-XiftX`EwF0+li+TjbmIDIbv+FL>r@RVYl3vfh1$-m-)hzW0i7&400uzGU0<~o^ zg!bisuMqG@_6Uv$aHVxhJQQnDU31L3RB=RIb$sBu1$u=hcWZL_@E^%KDY?ahf&O~= zL|t=1PN9yhe}|`|zenE>zCSKFK5*;t*a3P$)g9ym`ii_RKByO;9Pon6nv9DobNjshOQ5k3oV@OUt=(X3Jx!9K?RwXK~Itu&U*Wh4LS@=Tic{@6mzZYdgI~>VV$MC&oFrWSgM) zr|N}nago<>4xX~kz_F1Jl~3jv*?+&owIlGN8a3~tVr2E0;!|-*3=)@sFXFk1k7A|g z#8AH2Q-RnfPq`z&+jtzO7jXN;pm^OH1o?pN@jAUP2M{0Io=fpsErr$>$8mjIkEQ*c zU+3CyagLpTyeD%K4>E$%n7Ys%Q@(uY(`A6c0E4L*yo2 zhyEPpbGd$&{t*yYmJH-BVvX;xA95qH$?vIS;Wyc5-{Ujw-*N#y5{Kj~`v?BUhS@XqyuxiO-p4i7<`gP#@U6ak)2-jxdHQ8VNR({{Yf!Lr< zuOHXnE9CvDDToW_1w28=&v6tZ_)E@#oN-j(ysRC3IWU%t72{BSil=#Vzz;RAc&G9R za}h_HuQ(Ui$=m5RIer1}$@OVEF@i3{eYjsNpc`=ttxtQ?lYlv@!0dWQ%ewM@Gl;&n1Pqo z7u?Pqcuh=cW0Cry*YI`a7d&QphQ~A`KKIYQ=o_>TctI_Ljs*8T=e2g zpGI$#nyVTtJ$-Y)Cv$x{^%HB9hYlLryY7S>h9<|o);3t1ykoFuV0|+Et`01Qz~Rkx zM8~Fv^8J9$*C)y=!tvsZJjZn$#0Yu8GrQ506KhT6nX3h~f!vj5p+}6r(iCEl^DkGHH=8HM zo1D7locUOD%=}WPQx7qx^6f5;o$_e;fK+n77-+P{S^IRat${}ccc}972p67i{3LX#s7RV9!K=KE*^!w@w_KB^s0;S&UjL49~TA}1mlA91AbNOm3izn+p6pP zEcF(CobQs`=n1gj_TTT2C;6>31`Q$Ks_XMR=9RZp&k>DMbqwcG4kve^6-pOUu&c=ns z*Z3Zn`59&UTiB6^Q@iE>}4%Tep zey_u+w`=GcaDeOEx!^YmJ2=?$VcR0auP99y<07bXZ3Wz4|E=H44hYguj`UPy~DY0 zkaVNt-%8UuExc@ARY{!2R@L;^_<*E9KdPnBkD3@g&4z=7jN8xo*B118XjEhy~_>C zw{Jar-kiAQ^L%!juT{*DpW#=%w4QV8vHi!r;)&x4?ia)%s64N7KCk0#aSMmLogCQC zXFCq|!|$|jye4@JFA`U)9f-xeIJrpWdDWjFC&c4oyL?0cgRvn8le@_+#;i9xq&fe`_dLZ!lxgpNbOO=02&tRtKF!GVgW88XN=zpFDqsPQ4xrw7u_;L-|g5&m&z9T2$R#J%kqm>Lc(pLeO(S=O;$mCgTsQu1G&YRfOjSC z9UtH|xrTTr*5SH}b>bb+$v+8PV@>a_x;aNGwTHG3!ew< zljrP@nzHwZ#nu{#!?i#9kJ^24-*e)!?QDN?;4}O#{vEhqupb56`R%~tgL?1Sczf~z zy$at6^gGGj9P^C>yiZ5a6LbZQL2JP3f%7LW=%M&hV7-g@Ade_-QZ6BnEnP%688`As z`Lcf>B+oXkj4xwazHe-rGq(oj6tBP>(-+8Fum)62!2#yJ9u)qwIXWg#YvAkfk#NhX zfX^XU;F0jg_$KlK{5w#+g!ZTP`Rh2>eL7W~Dh@9`r(JO~xbJnGUmP#*z~MBl@5Jr$ z619l7w!%}wm;4!d4sPVbr~~n%{5HQ4UllLv6~>FWP>x0C%DH$ed_#VUd5sJ3gJUgj zJ125@=Zoh^XY!i$G&-N=sB^t;pbyR*G)9aS^G>g$aco?gFRokV?@7>$yic`C^+-BH zeNvrLy^?mM8)*@GQ7%iL(1`Tkp!}m-tkd~Oz*C+r(5s=Bo3_!@#|IcO06gz?H9Oa8 z@t#iv9}i{?eCCXS?@$wko@t!_{fo@+^@CJzroK#CR==SB5c|)Y@SEh>dPVr*uBCZl zP+Wm83dJXNOIx>Gf=A3_yoJZ`8-JKz;&{v1dBuE{odV}V4o{!q$E^eN)_F2_)s)R$ zbGGK_^103g?ln)HYq^D7gO)Q#akzQu08C2*hj~fc8818~xsW+(43+mJ&a^&&xhr;< zziRYi-<<(pk0$`Pb#UMF;-tCG^YcE}Fc7oENVh)Eb1gjQb)Y$YZ~21uN&Z%`(LR@N z$H(BI`AzaVzpp$pzqv<9`Oh&hM{raNaZikEeLG%><5zhg&Qq6>8w19J9yL5}ob>monZIBx@#Ksv z*Cpc3fw2Z+tg&aEo3ZG+Q|7!_0Um?bz;vV$NEV{wCry<-HJeCJLR(P!9nfSFz;JDVjaJ-!pT)1FUU#q#{+{Xg# z`Ju*;9ui~Bc*Cv6r!j2I8{_&_%KyU2;uVcraY~JW-sNG4QECff6n)IYmm7&w`oZXG ze!n;(wuniU1ISHic3S=G7|?iP!2N;z$UMU7 zd^mbgyb*i&&^Vcf#Ib7I<|>ZFje4hW0S?41as?blf780OFa1wTw|a&DQT`Twg(1>1V;p;gQElW zkb=LhR`=anYiNB1{|WeIY7{&(^>FohzL_=G{3`X73xkV-i-Su7H4e27-WH$i@}T<5 z_+EbwcwpBA{Il@^U+ntehCrQOjRo}h7Wi!HHEJ;YG<6$r?Wel~&+CDxT_)5V0)Z=X|8B=ON#uvB`iv$aFz0B%Evj?;?@9(35@oL=a zZ{`E?T=hBgXX#mG3z^gJJ}Ka9aLVzXjh;?B+M>^%b|(X?T%hoSF>3(ppGs3C9GZgTn*ybML_Q1Gfn_ z4K@hA8LSF+SF-N-w)mm-VWXjUJL#e{3Cci zc(%hcQ=geSIe0R7Ja{yCWB@!g^`WT`2KNPb1-At^1vdl}g6o361b+&y41OQ{uEFI~ zFQ59`;Ia9QvS=bNtJbMuS! z(ij)U2QPvi(Ep*QMqdXVuxhi;OOK@*m>wG*fw5P;HsZgyFXr<;^lbqDK;Mh8E4K49 zjcGoDn9a{L=F3AcFZlN6j=4m~ir40vyv_U*ujOlUG;>rQ*ydj5v)oEfDM#W3n*Ws> zsUygR#BV&Ya)2{vbNnHm)9y6B92dvP2gG$b3+|C~$d`F*@@KvmjuP+X()=xa#h3HC z=Xr6J!<07QkMY%jPl)UI8oa;@0)I=VAkVkH^LGL5aAklm^)1}c{S6I0VAk=gwWyQP z6F3+D;$hwbO%Vy%yY4!!E`Nijm?xksXbRlFY#NV5s__njr9p)Cg&^=Yq+BxW$j*BRRfqJqEA0&%58k zHqMLl<2>n~p!+-r?u&_Tc~!ReexK{UYsYwA@1ytfgY-hGks1TQ_qsl~HsBE%C*VH4 zsejV#RYARG9C-|NO#k>&0I@qz=R5lw05`z9;U&V{I>*_}i|TZF3&=MR6G)sx!);@TOc! z4}rW&-=5lI<<)XF94x=)Kfr-SX<%P$SoSWgb< z7P&ruLH;jxZW(+nST0yJm^YX`m?an*&?9sR{UKh{7!L&Z2W~-JzdjKAe-r#NxG1;MB^I)3}?r+jzi>X`mUfXo)rc*Z#Jil3B zn~j3CgH?i+0=H`go?9*WQm}NeOt5mWZs2`e1@#?ncMc8+jtNfcaNN}6dXJBudi2y| zgVTZ^2R{$|UOI?oqJ!MhK6eJs1+*M(Hzc5OrUWxio3W<@SDd!uwDp3G0(xgZu#3 z&;V)1{Vk8gNK8E1aAaW0*0__A$&YoAXp|?E7&C1y2F-3 zwj8oeut#uoaDMQo;D!#@4Y_W}^}&O|Yr*V8XCM06VEN$7f!nVHUk^qFhX-c{7X`l! zeji*JToc?Ayd2CjY?fh*1WO0Y1a6lP)(`dx&Io=RTpe5yT-xE0OEONhaKD7r*(&|JM6Q;+`+7Y-!MFwGIYw&=YpF8zxkA4bZ|)U-C(a^t6-&I?%=&4 z?+y7!@b}>5;MKtWyMkW@-wXB+_6-i|;M}YqEFL(Y&hNj1mxC9A2ZKKZrw88+b_})& zwhlH9jGcvo|3t*=^D)+V71sxR48Di8bvzt7JSv-N!7=$isOeRE(A|1L16#|Ai= zKf8M%r*SK%l<+}sCb8+ya;0FP&IwJUXfIC+Y@ZaLWLczSj zM}xOMc`!~HmF}N=9+Ree;!GpoWfzSC{;IZ3B1K;7hUJ70d-U{9grUd^9Or%+Zd4dta z!olLflEJdUSAw;IZ*=(jl<8P+%6e0_2u20F2YUo=VejCe;OOAw;LPBM!HK1kVKj4E`Ow8N3?^F2g?<{=poh=1V z~3BQIMGUJX{wuVSB^S3QqDSEEyRqvdIP@lZZ+ zO~9+WCy=K+AIL4`QR?4vMLC?<$d5cA;OFvb`L8@w*M!j5E+&eH^9P>|z7nh->=5i9 z921C#7X-fz{v6y8+z~twJQh3?ycE10Fp`Gk-|%3TK%AT__*5`oFmK@gCjtREQI-e=liFq#s&jh~n zzTnn?&bTVLG`KMMVQ^9~Iyf-+cHsAJ8~FW>!y18O^o782bNrnL$I*Fm>}L)9L!9pf zZw9XiuLj~Lz2sVe&vkIWh4^Z_R|4-7d+Ytik=LIMjH{=Er-DarQwhyqKi}v7Ki#OW+^>f^Wg|F*kV?yax3p^H$z3kJksHkL-j%o^Gy}4%i|! zZaMjy0d1^4t=@$%^la!iPz$3E@X5;oe$gB8a6l{Y>u3c&9#8MQfM($RjSlqA%b{uY z-2*u@Ex}XdFY+1b3cfZ?Pha3YIU^2~Q_7p^fBjK%Po|a}R{d0di#z`(cqrg;-xu&g z@#-A`J#uY;Yt^dxroRa;4%Bt)Oztho?T=%bUarzVtQ0Re`)pPI+D+?}BrKvx3tDT1QUi{u#lU z9nPM5c25hPKlS|H`^6yn8K3$O0iT-p&6~a_@cmB)Ja+pe@1?QiZ!}kV(o0NRV%jpn z3c*){wE}tE7Qv{1ZlmSo+3NX@mzZ{Ja6)imcTI7%fBBt00LS<2u4mdUc&^R;&d9Y* zY%4Yv>alZkeBe9kdwf24{+M8N2m8V&w;$dI?iakrbEh^v06cXykaIeC3?A3>wa@m= z@2bzWKlH77$-#l+BF?Lu>>BX#Hw)BFz8$ zTc~l|9;k1O57Z!xA?IA(;@5$(bV2a5z&J8ajHQCd#*ndOTp3@d2B&mz-*aB~9v+7= z>GQ?~d=LHyE$;h_J7c)&BkC-)y76uNsgK+c*k@xAZVl9F{1#pUj|GfVzp?5l>ON{I z#w|<^oxK>{x7>(~i(*|XP51TeqV4Tkud?GMkKOHO_d?r{r z_)_qdfL>GIGVax4whr`|$XlFWV_q+babR4WALtoTkKs}255Yr^2fR4_9y~Xm9(~9A z;*F`j@C|7^Yh!7=!vp+zZg6q%$KZzG{y-m)x(98iM`6K$zoFMiKdl}n{C-5B*1)s9 zEZ`AayT?}+Gt>z9;bO{S0pFkBzgZv-sW%@JoEfOMtEZ0-?g}0ctW|tF7@U8zb$^3( z@%;VMF9mA^8wJ}3dj{g==-`CF^}WsueigVr?_Yx(194aU6^q4Z@&1Eg=#Zg3jW_#{ z*?W46W~0-V43-V(h}D9%1KMJvVDo@CvO};_uuHI8uzNs{>=zsm&|QZIM+8R&V}h{( zKLy+`cub$k^S?g;Jnwbe(>^})u;9?Z_wWRKC%sob!QMmm9f| zXK-^cKA?ZD3@!^U3Vs$icP9tm502`tp|*Ffr@3q3TyGihExs8T11kn!2xy`OgHHt? z4`vMLqBjHAiF-bHB6uLMX6d@XIw=0exZv!7=XPXZTJf#7pxW-zn=|wcKT-Y z_8Q}^Z>M%sYm#nA-Q?=v(m>yVoJWtb`kuZ*{)k$IK0Y-HnrPvG9#WHt72oR>(lc4OM}V7gHO-(#j?Z!X zFM{&|aYQ^28%_*(FZ|-80v^m^!NI|IgYN`986G%4eAj^QyhE^kAb)mS_j%!bY90=3 z*};AOH}I})vu)sgK5JC4Q?N_0Tj0BVr`+6qw|fTmWuIW*V83AhfCp}W4-M=)93J>R ze%JQ`zmYe7Lcl9;W1<);?{{1(Hj24ouj3{bi-%&fSa@wf^NEAvpSZ`)X;+3J1|EqsMjm;Yj`kLfp{F#kAY?^*UeaLdPtpc@FF`jp` zW3Wr0|5@Et+}9VqU)NWy2jP(5uz()GadN<80(}f}LVXCQ1!pvCmaGZVmtg&X{88;! zZ^EwvyvZN>L-41r&P-cO2yme%Yq&j$L!tjW;(=vrrtR!Qct zK<$3E0EhFkM|6FUdI0nV=vDh-@Z~@cn;cmmfqYs2qrOM|ZhWb2g6*3=$KCU{4fYK5 zEb67(H;{Mpt_}(g4i0Uu>!9~e%o!bw36AdYy{?C``WE#o0x$mL;FLgKua6H->%-~! z+iCS&ZR0)O?{maDpZ)#7clvJoU|;Oh;eq|MzxI8??g^loe% z=mq34tQ&lzTh~nctr>`=Ukz3ctohX=$b)da#aOEcw`OLJu4iM`X3dPA4P!y?iukRDVSS9TVvP*#YW+9QLtLL6@ILfZ z7-PIjH4n8sV~-zlXW*I@dM73Z`VZT_2;-SX{&k@D=I6orfjOWrQJ;>!-jf1#6!XVi zIyA60L7$Md2<}kzNU24#YKm71ldE->qq$oHfZ$1!5nc>Cr$ed@#5_D6h$S0NU;5 zfahebz{G%-6GyKLs(0b)^k0gvG~OS&YmD=z#9jW>Wx=HZpNa+)lhwrw9=FfAZ#&z2 zug_6?yCU%UzRP-c>lk=p_Q(FI*WDV}@4EsytKaoV;J4DG)@eHy{|Fo($4cyfC*Z^R zHz-9)I>2#uyd8J?cJ9FW_;euWb*^e1g8oD65cIO@KU|~v8zXK1;dbe%Pzp)@+_1&<=8F<6D1#@lHehDX^aUy1<;UPC(uB_P|-wSNV#c5A@QO>@*-l`CYam(;5_k7rIZAKP8MFo7(c`#oKv$^wZqluPrYE-U)^nAX z(EC_g0-wo;^gq%OxQ>P>{eTDYpB?0Yld7#hi%HWSd zX$bWuygWXjCGa#oabrMJ@U(F_ZDCy`eSz13uR&|5RpEQRa=8D2zGe24Gzy}sYR*eCmFU+uU3_dE3J`E7ou-zx^mMXMLjF>!3fBgaYI zUR-kAc=_U!;|R1^3vtVFhKoAXehX5&cr?EkM`pofpZGhrMK(Ft#fhS z;j9i~B`pc=gV$_hdt=0B_$;64^L?N1ZO4q?V88uVzuhr#JRBRxsQO~*P%xIDdK>H3 zYmO^;kN396xQ=rj>;B{I7~7Bb7}oK1EZgHrC)Y8of4E0*;7x@%)hB?fykTb|1)(*^nTtIKE6yOWF3NHhf$YbO))=bE6 zXb;>oZ=hdF4)lotN6{fT3wO~XxQsxPLs`E&rR&esYt`x!yx6Zx@TmL@ui{s_q+MS@ zm+0N<*CqJ-7l9tmalv^3pG7|>o$|wgR#A7NSHu7w%&CD~QjZw@azb!i@cn>4L&uEi zAdZx#5m)FMc`0op*2q)E9saB}8U6ZZr_|^9Jamm-dAf$@L(}LD{bqOFF!lU31G+}c zqie)JzR(JRcu3=@fh`w^kp=lQ-2=R1P$%mLUBqJ&XXztYyu)JYzZ8QP4rr#40o}xp zqMhao#A`J-eVO9-oB=Im?V+B}f*22aH)joa0Ajye@SNA6w)I}Rz~}kgxdPweyL>1A zz&_9m_RId!4ffSOs|nhFsNYn-(=Fej;21a-&>kbl%`tR+2FKQMcDx;X=fL?`CE!~) zSA0A5NxGCibxxgIUWPg*JxiacZPF-oFP&nn)Vc^`$+$Acj5*`ax<+Ht_%v3H*)s!U znTB!w7dnO>H|A+`dWKFnCwL-s&2Iu4-r7gku%K_Q3TPbjNqu&FpoYLNkqel6asxTS zLxG&Zy2#Qv&t+W{eWT_}+qAl7YSvWY0R9P2)Ossxm`l&dVQ@+5nUU%1#xpdHdRX4-OTk&Krom@;jh2y5;W|2oFTxYyhl~pPy%73Eex~Mqcpzu9J_EnvS-vms zmD^Rk5~IrhC>?{>#YbF^`^8o9Ky0XZEH+o1605`!F-3eCC}vfxqJL-rTHvo8N(<2n zGz0BGN0hHZW6&Ek4t+!0l%^>yL%+z0p9s`^K%U&9p0~~ag<7vb|J3);Kfaf?vJXHD z4TO5V|JJtkKEI`Ol-s&b*ZIBlp5IRc0!>xAidLi>`B1HKI470gJCCJz#A#Zj z;<9ldE*lg5ahYdUYaWav%%mDX>sG( zxTe+VcKW?yIGxXDWb6l9lVv56zYdOQF)6v{m-Cc zwRMm%+SvwrxokL{esvc{_{56dR*JJx8e`Y06r6V>HYDD ze^UDF zv%a6^lA}An{3JT9jdipfO($ok@haBQeZLGU&dK48D|vk7^2QrK%Njyskd8DS<@Ls< z++N%>R>eN`0^`>jLu0ve{fdLey;%5UU@n*wmGhe?=8Jj5e-j&BJAnRuH!#oW;>!Jb zcS8f(xoQpKC0#9M(%7?ipt;Kr5JSytag;A;o`c@30ofG%HtsA89T<7aR$a5KZ-9U7c?dd#;7<`W7ho| zw{BZ_kN4MS^uOkH&(&wSt!;d6<)7wC3-d>e-r~_`e$y_z~*_uV~`Ky0C3;yp7)*>KF4SJ?D~9<{qxxOir;Prg6(QQ`aj=y z`tJ7or|;eL{qWts!{^iR?QeD*94p7gF|K3lmT!r(inrW4FP?Yne6`Qjd%QONUDLO( zb7}i_{L~mTUW`2)Sg6N1(bzOb3+>}}{MxqO?|z}ib&bt|Zu`Hl|2tYP#=phIc(*vY z&jO#@zYldT%*S@V_n%|u zzEE?potwr{&As-vopZ)u!JHdt+)tk+|2+?DOxGN2Z)@z={OjM=>*lCygMir0w-?8O zPL}hE?ewyG8;JJ>8kz>6qrv@x*9x`0+j_45*8Akz>Us1<3wik8135VzQgBTK^$7Z8 zQb4QFEI_{l9RsE3X&XKqxbOMWIo@CJc|N!Q^Xv7x^ z(rxiX@rGOQ-jSV#q$8mom!_mEpu8QLgSMe5>6`-XQ++cu(XxT(ylz|XfmQ>xx~J7X zzN;UlZE9a=p6LqefAvynqxN%KJ>@gK-)H%K^?MyV$HTF5j2&-rqvc02&iScxg$E1u zoZB{L*8Sqpf!<#niZjLPmLp3$lol}7T0ZONG~*Vh6_@q%kjKV&@m+gcJZ78XIrGB& zkPj7inLGH(JgQss${Z`MG9!zViWl&K9IViC1fHnJ?QP33{oD8)zQuz9?}FDU*OT+f z{p5grQ28M6GYaxV@Yrj%tL^K4n?JUnTh9;l9raoDy*}UfsDsE)`yp@bhiz={eLl~A zRUYd%`n`U8aVs9h(T;h`fzCytIIwsS7rGB#E7bONTkk90taIMx$sfj__|kaAi~iBM z#e1bcTfUPY&?2Q7TmCD}*xt6fuK2Mul{ric755d_72lcrH~=5u2YfM*U#+%Qd7>P! z@ZOZhuA{z;DUyhrV{{8z`LJXpt7E!8ovTB^LCKTtlb z7$?3MC&rAiWNaCeG_rAP42$8FM~d;rIj*i8%^c|0n>ZK8;#Qp6>Ob?5X2G2_l6hXf zEWW`9_@VV=d1E}XD+4v6>i25vs{A(ITh&(ivphL9DB#tqNmZSdcL#jDw&u#)7 zJ|#c9zb0oLB+t7aJWS7dou}Df(^luJWA3~DFE}T*t@@ySa_+_j_NjbR=eYLA_w(bO z{{lbNeb0khV(q8*+E2c{&+|Kt6YxDyk9}ty3)^~+_ttm&T)wTkfcm~MS#?S4C&W1M zOe_;i`{P*U0>-|WBRJ7#NL@@VPmS*B;P1hULH*4mJ+XRYtpoP=j`Yh~bZT*IteqYP<=@~yPi;SjS{NP- z>o7d)Wo8Oy4rU2H7R(mR-r*DJ$DS*gyTd%am1Fb#|HodljqSbfqk+$xG4M~1xBb<} z+pbym8`b{wOzF>c92^flI_jzFt$KshWc3oM$*J8Q9XQ77ZhB)Je=)>4!S7-kKF8(a zra0@IiqT@WJOqb}@5X`gKqJr?#)>iHAmV8nnr6ny#u{y7{Lw%(m9a@{RgKK}EiNuy zrv_o1PYBc})GG8J7B{Qen;ZDqJTWKRxo95Me5^TXjWTB zNq-+OPQ;P;Q%y%rPOV2RPM?EXklqJ%HZ>zPrK+V>?M$6YZ>K(r^#Xk{RXbDLs{R;t zvQ@kKm>!wxkI|>0XG8CXo(?@8Z4a=1k?Lj9OQM&h`dIXu=s9UoeJE`&O52CxvwUXt zp6H+PeR@rNw|!{)N%ZIp)Kg;rtFOfGQsb-Ip5O0SI3|vdV^zH+j^U_4&9drmYHwS;**`VbstH!# zOVtC7l`#YKz0e1AgdQ$2&6pEcj6>sAe4<;_mDTjb33JPQ69>#k`pI0Sxy)mHPOF*s z^c`MMqt{=DfANZ50lfv*q_|!&j>26F1ubFgi&Y4FWp^_(E*P)mYSkrq!aC+ePSw}J| z*f3Z*_-ydWKx`0ithdmYfukKmoQ#WIrpVN(Gz^VtJhWV+zZ>5eKWfUxlC=i~k3Fx4&o;L89@-9H(VTRg{y>___tf~pwe$!+ z#?SI`{LN#KE6Zo?J6$1HHO^>yK8{!+$E5w`01F0k#`2x`Rjp^`HQQ9a;Jq{nU!3nJ z*6?TOe%~v$(D#m)aW8I&6S&*{(|(Q>ZLFV6&D}bTx@I))&;wxZ;#ECu_!DR8pTli> z@AT)n-lQJ>_ouzzyLRxubM3_!f@gv!gU5ou1@{Jb2Db(`c5580CAh8+6SMB&hTxXq z_TcW|f#8wA=RF&|7`zs|8xSt8?V(Q)U+ej+HH5fRJP<2zkDdkn2zU+8wCe|*bH_tJ z2~H_K5hsl^8n@N~h_|=}+y~=P47QDJy+=%Syz!FHYsaYkLT`g{XFTdpskJb;5GUeB z9Eof3CJx2<&WTuHU5g$>aYSs<>!=sganLKNkJ7pde+yz-{?$2pCF?L8!{>v^!4tuw z!9(43P@JoKg1Z9etl+Wp;r08wzg^*dbzc6N^)Jqg@9~|!-#8Hm#Xa#&EEB(+6a6N7 zNW^_H3Fn)S&INuq=Nvma!+evA(*EWfeMIxq`?$fFr>*3@^gQm6H=A?3QrevtYmU+E zYVCRp=sey9uE$68p6;+z?o|?r#EE!~RHMJ^Rza--DNeR|D(a-wmb&(*m(#Xn>z*31$oC z2<8sv4MqeD2a5+w1~`4`4(@wyiD1!SWZ*scUi=vrOdT|}_nl(XJmf-5ZZ}@R= zLU3rXN3d0}POyC7n9Lc-5nN};x=qJVKcbw)ah20J&hj4K5D$fCB7frR)7IiKo^<~B z9(eSn;KO2pF~W!7GnmKb7G9?_^%v1NG>smlT2r<|I+gl+xjZi2WDepAnn<6LIVxs| z6UM0cfJf!ia#)%e7l==M6rKug&vW5(iU;xmV-sf^pH)luJF9*!M$jZU-Z3cel0RfD z%QI{%*YJMFhj;I@`Rj69bHSW2H_Q?HB6gZTVgZh-SRlTMn>b1QbKJxP{8-1%{A_v2 zvBRhICT=SINB5cA`poEZy=>+?-HNls5*&cL+MadsrJ#?!da7+_ds@DF-ugALI7R>J zyTd1Ph<2TLoi}~^au8mP-^+i(_0FHz&XeFl$b;~=cLom zj>_I4B&4Efun{UlG*KBg4WyJLqQN#dsuX32Z7dF_QJG>JLzzn%-_G^@?ceo3>;Hc0 zy*{6Jt=}4+;ePJty6@+Hp0$=Sfw2OcyIf#QV!XlD`R*UPhiBP9{QhL{MDTckgC7mp z$bSTX5B?T-{_ddwPZxeW=Ni|$S2|IvNw!@c-Z#P4d#A3o`fNWNV{D{c><*r{&Ymse zE8u%=sr|H>^@qj+))KO9{0(ix2f_FJ5;jvGqNjWp^Amik^1JzMI^KT7w^C4jiI1f} zK-a!F?!5ly+Tq;2J_C+9?>c|(*SE}@>5KZGxIf>N?XW(K-4J6IL*UCAA6UD6VIT%` zd*Ho?c+m68(}R}=uMA!nc%Irb=I;qSOU_<=GWcx3uRbO)X5eR^8hED(zuP=jRlrsnABY2qAw87$ z=ieL{Q(1#;ZM?Pg*4ncVWNfZtM!+^`1OBfVxj$n#elLDCSB2~N27F#V8GX_>=@&ht zZ@7#Nz+d_v&cajp1>fKy&kBnb44;{?Pp9=wd{jTCz9m1#7@Ez&qw+L1#k2fEc7aZt zXA=XW$8?gO8S_jJ#Foqzm;#v1RtUHB*=ro<`CuFb_8r4R}Vs;==GhvWFW0)GDWfpHW+-}^Y%Q#aK77|*!5#rS1{an zjvqzm`Ex@!#y&ako|}UDdbo)#(Wmg1__6+s)5QC|XKha4nI6xmcwWXBgCT27$@pSs zAhzPY9D4+8#UTMZ;a#Ao1m6pA?)icE(v<<*;GJv66L$yf!ehZi{@8^73ThK@`D(#> z!A8NB!FItefjH9v!C}GC!JNSRO}&ThtiU_J;HSZlg6{?=1z!n17l=*m8L%N@O=|~Y zOzeiS&5ePv&}jktw`agUt{E6-8FLv&c$UYQ%(K6qr!hzT_Q2QyXOgjb8nGj>cmB3{ zPwmMrX#e`a#*)UA9XIH6=JND29L_)C*YaW61pbWiHLl@*;UEqJA4@+mrvA(L9T55( z7-Nfv@U8e{`j6)rJ*Q|qOuvog>9jtrf3q9LFyao%NN?$uIV*KCPQcUZ+Hvg6tdJ)I zc4F$Be(geSgY#^OYx~?U`(U1$ezLLRplkzs)cI+7vb*dQ*f0F1$pdfWH1$F+KVT2=Ho0qWe!=XZ;{&$4;{$%_@c}!|?>EltxWLwLykjijTJ5Jkc*z*R zSW&vRemmau%lEeObDwj@Ct#l~Fh=7`496*S#2Ccb({UWgX0R8=Za9)1W2?xMUt=DE zJjI2`ke-U+h|B#n_-1fia71uOaBy%yu-_Q=9s6`}Sn&Da$l&O}eZGZyQt<8I2Z1!_ z2GSQZx-JkK)Hl4_QoCtKV-D@A&5UFGY3q(Z^fzT@_lo zF~L5;CxR~oUkgqP@cwzhrNLFf?}BTB>w+7C>w{|p$DO|N-8A z*Yw%-^cBysF~$~R)_hDp9$!*lrmy-p{-LY*lV2j%!!F=ldS*Q?{jwg`^H}DsX9VW4 z&1>O)wt#%-o%wFC-+6E21~!3yT5BsUYuV_y@dCR*kLATa(0BcUJ)rYqSH?8fFZr`h zk1bG6{mD9LYeJ3N^f9q2ZDKrOEMXjhYw;mY7Q5r~kvn_IC&c4ysCYkns^8-QvS%mA zchqsyd9?+97+1K?SV4@!7(^S2SF!cm2illDV2iaoF6E=+VDWoCCLN(GY!^S1o%YN; zm=jWdz6IOBzrfS#!#>~}wgjX_*USwVql$NHW4ed$@sq9J@^($ z1>X9^zhlIz?j9dN&PH%@H3 z(-~t6K7c-K3`Gw7J+?tV_q?zeuQ(z3Yom@Uw6)k2J@sBR{uvn8`)zK*Iq!!uH;7y4 zh5P6RRA2OC*IY23pTnW?1eD`JF#l8dhn`%UDzTJ zH)0peM}I6hBrrxeIXELYKe#;jb@2OuP548=4tRI-FMV~Hbe8}O6GJ=g;_-CPmhTKj1W{gp4J zpXqz}-iN9g}JhAy%N#-QeY%>(0Ib>yeuZQ}%SANBz3nit>}aej#3G>@J3c zE7cQk;Vir?216$LFBy}S7zE#s{BVFiPrk+l#$@6de+ux8`A)wbXP5B|4$)51lP-I} zR}_~QBP93EOVR^&SR3%o=?Gn+Gx(Pe(U=1c0RF|Wj;>Kxb#~l&ey4G%EiRBoQ{lL4*a2l^OUaTwRUh?Y1B?&Um7WcK8uBnsz@KDAZu0{EySUMB z1M|SHnG@f&bV^|e1CC$ z@fk1<0Q=%XGXwFPt%F&CIf4%d;ya%VJ`)@s91(mu;MW@){5Zhz{C?ws+k^XpzXtyZ z9tj=}9t`ddZVG-AxW{*$jtveD-WR+nurA8jWx?RiX?IS$BCxjV(BR#{YXa+!$mQz5 zyK=lw?V#X&ftZM~0e=>!;VizizNcNaHNW3DQ5+ZF^AY%;{3ZQ_{@{JFA%3J-rgbOc zp!80h$QnWXLXWGrbeKNVUA{m4buAy34mXBvTw%^ko?;Zr!XD|*{7c+fe}RseQ#NOX z_xb0>mwbP5A@#Mk)fm&*$TJon4y>&Fg6&D3ye+3nr&h`tkGZ-{uul&5Svpc{h{+V{8v1Sv-srf;BXAU z-jccguAkQykZ*m#jse(NzM?PcJYSCJx)pH-_VC7FUSLj0pBE$JmypY^25W+GLw!tXk`+6_rj%Rg zFZ)p5=f8Fwi`UHOcbymR=f~hdaZcj^HjpgHlZ=d|_~&fqox#lk|NRevc+Qo9wOY7; zI7awx-a%`OV0>_5FlP+=#tEkd#u3I7X9ed47YEjSu`9m|t`4pV*c^GXKeq*U1?DKs zTeD4ksrvi+nLbvq6Z)Vzk>NPo{E&7w&SnG5$7^Fgp1x-6$KN_IU<^sI58_x~La%?r5&EC_0`on(E1N~dS(XYg5^*_%FcrT>) zREiUsZ}#46>+h{a6(89&ct>DNaA3e5d?PTI+kTHSy(F%ci1979(0=9#l;IF?f5I15g#Qk>;j1fK<>>C&>d^8aA**g$da*b=< zW1iWXRB2kr^sK;K_I>##+8+XQ;3ov14t5L7Ew2+SA1o1&h4+24kH$;Z9(f+XcmS{A zF>|D1-qu!%UE?KV0ZPlFNhbId$YDdjM2DK3<&q*^5Ob%9FObi zA?{%hV5)B7erfSR^+~ZNehL4V?$IlKTYu*(;sJSOGk|}jJ1@fV+T5H zorN*O{Xt_t_|E(cZo{4Wg1VXu5r^UXv%ff5pCxnlk*utJ6n``C2<$=M+bBlFHskb} zfw^a62%L@s`FebM{Zjij)`HJLAJw1rZTzlW`X~EqOpo9BZeYLT80Xmmu#cmMex6tX z{o?0|dExlNZ^!$bFP{<*kc~FyC(2J=(l+ieR#H~H#8%>Xu?+Dnb1CI_dS#qSX66BK zJR4~2#4ZlUO@B^&-na?h;(Tj7#e?uZ-(Osa|IUxUAmG>kB>3_8o#bca{oy|fa67v& zH()2+XZ&zez@D%xI9%S1*C>xM9eb`X8B4QC%B|h4z0jUwfqXITDu!p=27ENUu8-n& z{+oV8m+(2iz8r20hpV-}W94ht6zqIC+}3aYAwFPJ>5(-KbjmXp{5AH(mhNf`{tMgU zTCoJSMO)EvdD1=Yi2JoSU%;RKvq8Z3GUqAA?HL4XjP4FR!}YX0H}!&G-2e|-lQ}yO zFUR%g1^8SX-Tc73z&Jq6$Gq_Uf%T>M)jCqK9(=w*z~8nmi0}QO0B?JyK>YtxL4E$C z@{K2R0jCDOe<`+qad1Vz?+0-p$DDJGv4Jsyd(9K`@!1Bpz_SUSdr%JCZ{AwG>Vn|J z!2CD6Cg!zKApXPdEfp*nSXU%Qc4gqXFmgLA@Sa7yhp)sta1>wFJPi&N(_jN|5xAad$c*PNOc?4*?tCxA_k~d7mE;%PGW%+-tm0I47nrU1JG$Q7m7c zV)({N{1LGxI?7gxJ>f50rQFKTm%*L-kbbAWe0ThfgF0uzCo@hmpU2n417ZN?&G~ud zAHLr$ff$ZBfw=KwfqRUl#47Lv{x>c#M^KKy4Qz)Phcajr?SmiL^7d16Eroup@ACa^ z9Tzjv_WHbjtIgPn&I$0n*$4WfACp7lJb1gjPgdr@*>78%&$sXRhcD00o9{KgVEge! z<2K4e@6-dgcTU9k#rTu&!FT5i^Nab%e0X!>xE4BZPL9_4@TbLi$X6_ey#E;RyN6%{ z_~T+WYym%=Junu4`t0n5w4^IPu{OSswq|$uN_-}quD!ZWRD2JgYbX6lJK$D6kbXiI zLwjBOtPL8MA$#_gE|R&irM|?b+Q+YC-M9=NkG|qx{hZvjYi&dQc)HAADCgsIWy9@Y zpMI3%afR#I3Nb@%$xh3MZ@`}GKm0fIX5u^6XL!fKp#jdaCek~5t$*>&BhSV@H?Wp; zX7H9^m%zNJ=M9bwtQ$RLye3He-}46N2Dp=t?|B1r0F4ECo`6rkTwvan&uwie{x=sZ z<|Ec`o#_@eEa+B zf#(W76wC@X57rA-4VDSSxE2X0wK)I%fp0>2)|ox@e&4eL?^ZT8VlVl$o-r_o!awDk zifx;>$8|UkmztxrP6em&MUB<Xtt@x}P%px@W$V?XGiSdCZ* z+rq}MHReg56(|RO#?xX_fX_iW%j@Q0L0o?2z+5cbr4H6>N>e@92zl80J(0O@9FEKB zne{sEVe_olai6h|czwBDtX^D#eqJ+hI$iI%3;>zuij@sym%i;$z}3{9zo)rr>ewkknBO#@NQV#aNIWj1k#6 zHV+SrxAV98*5WJQ2#yY{!#N~4X#73lkK{Xjdk4D(?+JDuVAs*tsdpWH3v&0#-ACW6 z-fMEN$qxn}3O*dTf8SugV0Q4S;B&!I!PkRRgC7K9{TBts0>2FKKl{Lb;D2+x);+N6 zVlu|T_@6)Dd02eUS0@WTk@ySyN?!O_U(=s12-rz|iO<1D6E71x1N-_|b(ZhM*5ZAf zuPu$~rKN4OwYFh9hMX^7oGzoJ)nX!I)p)c%vMo-9A;?Sq8^!hsVU*mC?Q8E8Z=%MPRMI_&who zUt4>Yq2NFH-+0}+VlfoH_=5qvRo-U5^b5LA7x_K>@~LrS_KNnd z=2!C_^MxX%lV|kXS>9Zm6I9rQX`RZWbF@6=)7E6atOH=!4D{~;n2ey7|Lp-hh`DpH8uM2*S^x1j+#@viG z{;n81ZXCk87Vm|? zo8mHJGS+OG>-7BYjG!@fKKOP4UmF*T(TJb#9*EQ6bMYDLD)G6s`(ie@ewRRu9d}Dt z`u`R1%{?>raKJymH85vve!z3f>M%RlHP|u`Z(lKZW`L*bmwQ)_cMjbW`0mG5f#>4b zJ{-+Ax0cCTO&o|bhu^FcZ#Gvc)+}Z{8fq{O9N6Y78s(6P#_U8Zx;~st+UdHRE2j=XK5577EfA7TLTLGUBM_m}0tHb5| z(%Z-D`ReQ8g~ny9#W2?A&+~8lEDwGV7vukmi;2nL3%tn}#G%IjotF~>=$M*+Xj~0= z7}$r#2FQR<&j#rKVBax*KU?8?Fj%m_9P-Wa?sm=)lBKK@68g9E-i`(g|sZhu)Io+BnNUSodB zm`&ZypWt&ox^bTN=$a_<=MS@owY}tvDbE@oHpH9j_8GHxpHvlxl< zt}_Pc^>`j<#7&gV&LEQXv!M=f*(OZKz1Z))ir&TN8_~4-bwD*cR`96EnLg5D)TPvG-4~L*hj0ad#kg#Ab^lJvxSc?<;yBxF--> za=m#3&ls3XmX`PIc@~Wxv1`7+#SV~%nB6e}Ik0JL*$#m+E4w;~^Qjlz^sKG1qPUQF zFb>yW@SHX8Vglm2#?1J~z|y)0{DZssW!BvB+4$qf1cwLshi}IZFZhw42%O{d9TD*R z@I1cfbMpVJ-H_fH0d6=i!1vZ~vlZs=a6bEi_i={!1b-g?i{*&R^K-@b#WJSO-QiaA z=jQ8ht+^4rkAKAgtWg(F1^Z&}c*i;ZwHUZL1u^`_!Ns1;4|eUM*om^3OSm-gK)Y!x zbFB2j+=;jv8I;G_GIBS@GcF->vV+=SZ7`ghqKAhB>}7fVn9=BJ{ySa8-M~4=alSDn z9TvL*`;Iv$UcxpMeml=D7;Ev##cITGaJbm4xF`DB z_|SU@Z764i`K!5omk6`r}^p9z|$)R<~7Ajti}9mU_Mq{UR>wwz<5DS z{7V75WE^pDa9|)tPrdI-l|<66(l;p|(-@Y`|M;B{k@^1E|?3~)LbTpaX! zfWD=O*R37p=lkZd@{8A)i#IN@c8HB)zr@(h<9NPIyj;wMFGH{SMPk1EOZ^Mq;l3Gx zc}o7dSSbHk?AID)eiz>4JM%roN5s#KwZWPq@nLpBtj2l{>tEap_E!$XRGgPiufxUG zzu^L9;qR7r&0`s>cOFZ*t$h>k)W6l|^g!K=$2$f$c6N*}$7gHYWEgwl(}8P@!%qru z0^iYiMZU^mETCL`PBOyj zHG`K1D+c%#uk)#w2zc7_z3%_}zwqS3Bc6Za$tRxtdvJencYx<_3vB-!JQVnI=Q9Gn z`O3j70{p#Uz$f26*gIf{zZU#B5I<*Ajmyl<8mEbu8$)S(@fCJWe5~u4_y*>-@ws_m z<0(EmE@u~qZECKP&n~uZ{fTzP?Z!XGQXN0dk8~e48XGyrzwdc|K3&G${P|M@aWuZT z_>EY*7`*YRmA2jU8R($@#L3^ofk2{sM* ztm4PwRKRccoDrfF z$My`ZScrG)cveb0#9A72snV8@bvA4bPX2JPM<7OG9saDqdw$*?Y#VGHSfisp;@__h z#Q*DeKR0u`<~bJ&6v8@2oW&Q#RoRP=-^~XWPZ4M46XPkiQf!%@jK{@P z_|4+sxVhk;dJm`b;?U-w-DB<)cS}p0MjXSt+r&cfAsdVzaW`LET;3Q#j9$AcGv2ih z-+P?Z$DD^)5!km@M^M5$9@z21_24(ZoD8hzWV6im;%mM)+re)6W-h(zctja}ifp-1wB9>jPrrd`>*6pYbzsHjeFczxY^uMSr!< z(>x}wGL9At;SYm-KACIrw|KCAjkEP_cA&O^J#frcT$TUU_o2Kt^P<+$;BZ@Iat!R( zzs3!?RGbB`vOn(%@WT+|Eyng@F2xQRf4J6tcttw=b-W`Vu^eS(XUy@clQn$!*_;86 z#%ab2{BGlixdA^_tfcb?N9O$j#w>i<&jhx`p_7MBelj>P*f+qNVk$cZZx8sx;@?{a zn+KZ){B|*r^@6nmagmpgvBu;YBQ6$;c|+iS{{H&{F?RWUHkj&Di;es#uwDi?Yu`Ro zs{Y!DJgmhLb0cs1LWX#l{tV+-<_+*Ezgdi$on$}7z3HH_ARkza&wQfwU&e@@nZ~c? ze#`-R?&I;mph0|)A0i&gZ#K3TFEtOzhc<_`ZeSe*e-0<$3jCq(c!r+;#dp%*^vlKb zK6bHy7X@nvuM6H1>=5i0>>Yd{u-!cnS8xtDRlf7Qz&*bWaI|?+ezdhEVmf?jT+om0OF#?{=-N->z4FzW0X$YY?rsH0I!c<9N1-&BE`m3RVm1 zgD*L4$r112JAODhn zYyG=P|?Sk=|>N7KEDh+(e z&pkE3n{xtjdF2!Hk+-e6R_9zNj$@t3*@5)LsiiMpd}e(gj<)Vk?1qf2y#z5P94^ju zV?bVFO`g#i%IZ6LhDTl18E4{h{K>Z!f5NAYKiTJ>_n7#FxnePC+zw(ej=LA{<9~kr zP{#66PrkE!%}q3RMdoDH*wy)Y)|AZ6ud+kxqE6;7#j5Crbw_NncvWpOPG@^@I(x_V znV+C@#z^KOjElrYjA7`evCTz+d3SU0I2%vXZ+cGO#W=0?(kIpmUK^PIF`m|!^iBO( zj6+N>F;?*jOC@Xf$G#B7J)qZ=i+O8PJeJTZsyDT zHWuHWkB_6pWvnsXG{Do=L#{J^<{w{+vx`sTZR=t1HqJJd<4f=>{8@v-XU5n36l*5U z!{BN$JmdT01M4#m4vatb9b;2Iw6Q(E)%f1p2=NF$6rbAjoLv_#=4~9zch*m>BNhX< zZp8WnoF(>*$M8BX#Z$OW-#1r@1I=ri>%tXy1E=F|TqX94ufRUv6UXBme1_Xy4>-0w zD{hX1*$G>aHf|cC*A2M}XNuVZUdN+&3)hPW@R83MKP%5}unA%W?1nmu@#1gm?9^Qx zzVMr!12KBIIdE;^caK@y1J1b?KZ+ldE1qN<3cuM=F@OGWWB(82U0&>lc)Rw|Rvq&g z6BO(*-?`6Yic5)IiSIOiB|aq%W^B<|2Ytc8#v$`Gh!Dw@s(4cjK&Je zTTUEuCf>x6>bpW2Do7jb5Owivj!>b#z|!@puw))#s|r+G#3j>Z^H$U2a(2j(5cw0vVqY-B1L zm*JQ5$#JuocsW@N#`8#Ej=HgQar8F?;uSLj9F3Rx4dxDw=lBuJ1?Ca?6-x!yls`2v z4m2h-KIDTKH(E1N|JwRle#)i6Ie|Eyc&r%j;enW+7@_`T4$i!r^&~q4VhHAr%pn=; zTHhw#FpOv8a`TprMeE;uD!#dxiFm>A9SQs|e1SV23;1R@hi{6{@DtACC-dd_v3z%& z2sr=B06&TgyADtCb@|oKxdzmoWn0Q<{@|72x1RnKaR7l z#zo=~+6&lTaU5~tVXT;6E6$^BwJ}cZJS&^+IN)68=!5YWADE7??_wE81m*(SeCu$; z&Kf)BC)2Gtf%W*-DT%S+Wyd;C$Y*dKSMwt-3Ouv(t3XWT>M`)P=M(tY#*Y5zJ?<9Q z~*7+;G|j_3L1IKh~l zUxxE94m@v&PjCyK!9_R+KbgNa1Q+}5ymZV5T4N3NrC|)guf@&NgJ%TFAw~=QSaDmiX*_Lx2+jue zPSe1=Y@XxNbgy%s=OGKeur(Rx`OHlgev?_}?8uNjt>+L2Hg{)B=3>vb zV&q+)p-nCt!#>On8c!*w<8|vdw5wy<7{9aaY&@NqJ%+f*!DHB`Gan21-t?&NOQBo* zZeu&Kk<9}-$d@qg!|UFi(r4xIxVc0=wiq5?WWj)LKNiq)JWl6t4sf_<8!j2Imz|qu zQ!Wf_ar&8oF{v$-M#?E*Pm*F%qZZ-hy^YL*WJ`_7Q_GSa{CSTE3?7T5_wt>$twv21}{pQK=t#jpC z;{`q@|C22-UXYeC#PB$dc8z1;8tK^fm~u!%9`g2EnHt-#y>iZVEr&Qbt`|!Habk0~ zowHSM^(Pziv-sb2&a;W^qvvDUP(CnO@`>3~Ho)<|*FYSe{A-7`0iU^X3+<-;0ROYu zpzXx_@jE|Se3}o9*W2#e**&&6AD>%SiO2DI`JKPO$F@d~j+x)JMh|~q5YSJ42E7Dx z6mXoMZj8!5(y#QdjRSmN&d2x0 zz51cJHy8_xTZ)0}v;27w=hT<^Yao^?R;rJ)1^j$GQGf8X1x_2SDLgS_@%n_;!nqEo z@FCd*9AzyU{xOz!zgVuVF@teEuqDR)bH~6_&WrgrUpB|J@!s`{FhFrvAzoGyl{zxVDb*%dE`;{+T!@-;Ukj^I1Otd_Xn>*bi}0YsPvV zKM{YJ|FVwHK90dP;zHK<@hi!uB9jR#ayG_9%ZZFLadpR??vbbK zZ2jSFT;4Uk*7rUmpufhb^j122cXLDZ-ZjRvbYEGVufCV}^&xYb#>CP#-^pik4_}I3 z#Rkj_#5P^uGU#8%2Ysf`b^0&+!Il_T*T(3RuMhZgIKef_$WPPn*dSvJHbvZ(P0{CZ zg7~U_EkDm4iuL0Y;|rW5mZW?y46I4u7h0=e-VNuNual?yW(4>`nrxbM*d#n?y|MWN zX_1HeTJOMb#mDRbf0mD`p6X3L*6PWpb^$-CgL2>@^2KS6HP&T|BOQ0%b?UF&%7J%{ znZ%zRCo8g-renAkFZ;&1`jfr-(-G}X{%mz^v2<~#_O0HzR(j;TW1xM+Md%1$nay^NoIb~Amxgwf zzdmJc6zISD9sZWLIU)Ur55$kxH-8pf7MT0wO(0B_)MoPziDcl}?vmBnx4?sAi{xjM2N%EE>yr+dwJ@%Q+8>Sx@6!|)7T z9Pj~MhpU{EM(eEnWQP}VqWi6#)OL77ozxdU;z{$qKM2?@T%=Cqsm{*fWAO)^NM`a; zce2Ev)~n%F@n31+JiI8MtH$s7As_W|J$b8#I!ljyl}CM~qfUHUww2#14d=;E8tUU7 zGL?>N$&Nm`Ub^ygWmrUdpPyaH~3yhw{(~=N#uxtC#fW1-19` zs_!jreg$5_htiZc9j@%~C@z;Cd9wpJ7`OJEc_7@6%W<-DYeTxGt@S_gOZ^u&;v4rm z$Cj`e`iuTzq8gwoutVx6IZpxo9-3Q z$6Nd|{yyK|H~^pGPWg+oy52GIT0F(?#IHC9N61SV_<#6E{c!X~f%I{&Jj4RPy|@bB zIuG{c&6mXGuEF)LvyWrRjvY}h<;Uyd2#&jttr0ioU&@;=sr=+-U)^!HF^uc^rPgM! zedLW7>#NGQYvjmSJ>@AK>5EOZ4rpCP(~3 z$BYBr<67q$Gtp*z2;7M;>7w7_eCz-l+;L>j^%-uOWzu(W0-p~DSi9o9zCq91pUcnqLH|@ezmeGw>nB+n<=-t%SRdH$quNib2w8P@>gH7(EfCftZxa}2{AEa z0QQhf{Ly89_!9rBFI(mw^DFYEcFm!^LEePwAm^IMelX0H^ZL=)FEh z2gwP)^OJBB{#GCQB2Rk8FU5y^ndT)Qeyp@`hrXrH@NaRh>%l&L)nD{0*YnFje|Jn- zZS`aK;78|iIR0eI?VSIJ!2q`@x~FU8~T*XClOINrVdR=mfi zkh8i=1J~fa#&y^Y{;%?|Rj$W{u5pfBonzaaS9Y9+!eNCI-2j}rRJK{VUk^}$3 zHRR!0a=ht0E>v#i*SB3uew7z_kiGn?+vC3plIb4R@Z26D4gZ&^2J{=x1hF8b>cJi-(i9`6_{7Ss%H=9TP zbelf1A^2R316R{CZN%o*N0W#7V7_PVnKIxG@>W-}F;}O&Yy%FX3-rUa?i1(1>FO*G z`l>#9@t`iP95|S`bb~M-55r^hP?Ip`l$L3dsLgQ9O|mv^c>ge zYj~56i1&c@(~t0_m?W98J#2W#9sC{KqfcpDWnn|n;s(B>{wyClL$>5a26Rgq z@C!c`NAThF!N%qF1)Quu%a3i~6XH~|P+#L+{e}G9%NFr%`R4AWqvW7m^vs%NF=L#^ zAJv8(-_iyAz%Q-cFkee=^i#a4uIh}p#0r$(zCZm;e^n-&MUGl*jr zcQ(j1>@a=Op4yqtxDHQgTXuo%!J%D?fnT&O8Iuj!%gg=B zL+08OpNkKflcxXlO24JYc$Yn)8{#|0Aj*O7wX1TFAKQuR)yH{#h!3gF`S#k5?Q@NC z&{yZ(Pgm4e+T=pU5QnLsd+{w@Q)c?%9{z~*ocE``%0~|J)COdXhn>Sg zohx;X^58N0fwQ#>4#H>TtKZ@qvg3D z`JCr}&@E-6+w8CW)Q|0FOZ0KFQXleA2Wiqpd~S?kzJkrsZ}lmCfxqBh<0vvvW_7|P z`ZhgbAI08XCtrF(#-M)s1UsVdYg^nWuB7bhj6bDEx0F{IfUal{vcz{dmu}V{!S8%u zdgi+60iTK-?30!9ursbxFSd$*if{BAWyB@8LOaO^7vObi;$rpTlaLuX%L6Ydul}ZP zbVmKOmG*Qm9nxmXPmc6Y+4V)~;~V3DKDv99#d+5%hkhm>I!{jR=lDrklvn>zH|?Mw zIH&x!%0yoHfGnh;tnzlAoY-*3hx|?#wXtiILwU3t8R?I7RNC%WH}`92urE!#M=$Xh zK69KbaG|Yo>DSUF5BJa~{aiX^?p`uhUgyYGUFoy>Iz~@xqxC5^fjz|OI0494{k5h0 z@T$J6y>PH&WJ$hwzI9^5=`{UTPWRw(dCFTq!yPyhuei7NP}|dG_iH0*8&lAA?TKsj zb7j>(w4J)sgW3*hXfL|wS{$pNNmm->1ox4#d}?=9goc6$X z{BUt}e{3FmtIw-DE@Xqqxm?NC_^n^5D|ygc^1>Zt?_T;!he6u(i$1X5%0{nnk-U|S z4C%HpvNpgo{A2o0F3QbM!zc4%qnxKh+CjT07hO|ldcrR0%j&HCaH)LAU;gDboT2Tt z6CNi=b;WOVQGd}_^ojNp9Pb=?E3>l6lT64!{oSX1)SJ9))dQEQhkflkH(*x^xPE(+|J(L3S0F(L?=LndPlb=negsmh|b8bIPD>@~)5Joa1;v({*H@FH1YBx}JGSr52McetU z47h?UlvBT^2d<|d`V5)7k1q1>>5)AAk(;p(8_qYv@8Z97ogDRHw$FLx*QRU)9if-n z2;Z>7?p1!*&=WouZj=U{_vaqGNLKWXf2Z%Ozy3hi+^en06*p^ZJgYwLQCIzo9Mp}D z;681^N8sns8C)VJsh?^$Tp||kJQ+z#JChA=kgohIQ{~Ygv>jP%JL3R#&=0k%vdfDO z;Y0QvN0Q$T0h>jK_%!@h*RiY8bDy--L%X?_p6P4aTpWVzYQxmoJ>=KpY(JTk6Q0rk z@SptHf{x#vQ!dw$wff_F*DEvmk%2ZRA8kmV<7}@Z%=!AZTOY}8=_=K*@pKX-Caws#I>!Z@v&+&?W1@h9a`hvRXlgdLM`NrDT zHOeb5f&}BM9*KmiJr}-K6Q5NabXMIaq>4oxWU$W7cm4mLkULLsI zADxoFdfIAp_m)?+t-jzs^>IDD1ZATqpS4V_V1A zIB8Ti%BVl!0A*A+`ba)}4Ej!PxXnB@&QKPdWqgls)rror!P-H3>f<;W`O~kBy~*64 zzN>xdiI{?Mv+*q6HhwVncd^Ov6XtU^Z0^ZsXG~vo%`q^eWgG80~x6w zzlnb&9-yuCtF|xwGKM7sa;87zChc;ve6=MxlLfuNgWxya!-;sw{bZmof%}w=-!J|v zADk~u{Ggra2M)58HvQul;2=7x9O~3^&^25^-p=D1Tp&GmPg`m$e4&2wkcP6jPyYTK z!_WGPeC5wxN`ua*kG`YL^;O5|AKjt<>W!D!8L}mN^3X?=MVi`|PSYv!&?e;oZNmom zO{dk-ADsp5K*r9gv$BZEn3Ki-@@Fg9EcVZ}^w6LD=?;$IM=LWuaZb6kKOLsu^bU8C zuRh23!%w)$zCNMu+E`nvCs~mv|BF2NGv{-^|OzDv}cMo|vN6xqdv?Cj+JjMmK_(eV3 z=MQh|56Y>JOP7tb&)*_f{6lZtD{s0XFJ-Q;!HxlWx>tI}Z@5Pu%F3SUoAP&!b95h1 z(Mk7^pLWp)Dr54+1Rd4yTN7hy7SH^#S%ZGo5 z)9H@BhC8*ZI^r^VQT@VQWTvfLgMV<5_E&zoqn*?X=%7E>tDAm-1GF(+^jrJk)TY6n z(+xV{I`x^C@fjV~*U6dgsTbXsAO6G<^g>?xihQ+$eoUA3Cw5F)y;d8zR-W3*Z)H&@ zeNX-A51p?5HU_HA$J=9rR9rKcPWl9e8h^&=ady|1YG>nJnZ(_R=5={Xl!+EWfpnW3I&) zxC4*SQP(*~-s}Xs&AzeoY#13kFArP&Okc$_^oLyNrDNAkwr5$!0S)~}qS|FQ2tN7R7~JGP*M(xRu1lLw9_8(f0N$sJ$eDQ)dK zeX2U;9^9dixz~O4LLIcHzTzHs*tP0aKaqU>;XCza8?~$UkdF3sTs_p2zN@3_q^ZrM zNhj$q*|LeQBPSf8F4CdH+RAzD#QxB4{fgeWhhHTh`MKUX`8n^HcG3p!(|*dP-|9E6 zb51#2-#$a%a5eo0vLkOYQg%E`@5qI`s&CrYwd?{}t26y0EB#BF&Pj`o;4pfxe>qM* zK*oM+4|3H;@|734_)|vd%Txa3+5FU{`c8k@K>bO5)nA_S*58y54>?aK`ImT%+~~PK zWp^Fj1!>?4X$&DfZRi;L>Hg}8d)!OL(*s2D z>PfEpJ00^|`)E^tWKDKJ7VH6i*$qPkN`1)>qZ={jGOUf?IX(SD%4=p&mtEryf9_uAyJ8mp)17 z9VZ9JwUx4I3;LnoYfJ4){*C9b;h+s11O1E~hc>s@xgU>#V{EWCr2n;}j*~ywmzK7* z)y{OC{wi~MgzZy*_0mpY-!W~EV{kIrYK!WDt@G-sP4s#9IL_vfJ=;M~wFUWWcjsNF zAGn9CTrYj+rPaRToMW}wj=Rs+ar&XZf&KOY<-#-WbDiJX0@T-UY5Ofb^1!uZLwCu{ zK0awV>6qVi#I@SHx~Gll2R)G%y=>X(mwM6>eXLwkUqKo`?!)sqivDSfwhf&G*Xj?- zFIyUtsl|T%G3RhtDisjtAlb&Publ&fAnrt&IN^^85F&-4s`;42)4tBe(Desxi{ z%0`*mrsS?Ju2)C0a7-Qj2K$bYTV?1tdHAiK%CBxvnUjUO((T$jy5YQY$^^8IR0jQtoaqF)ldWU;sN7ebp;Pi~Us9%)OM2?79OMI)MP*c- zt*l&AS+%b;4aeQ*w{tx%O{nh6*Ll}AKXp|`^tO45`=_>-CjC?f zvK;D2%Y~=#B0mCm(y4M;^_*To`-w73SG&kheLMNOsT|eC z%2GaLUHI*s>uu?vKXoS``UdU)?PGL;J);kfDX;s;NIB>OopDZ?q^-VWraj2fJ@T;~ z>aY9hk~%5F&<|oS0e8_&eW4twzsT2p(o-JoO&)Ya8RX9{P7kz?cB>s9dl3nfVw(PM%7_uQbzjFws1`T;Cl6tj&>Z%Px{qC=lyoQ``Ul( zw=Kv`ndRNK)xP>v`K@j582GJ@%0LFn2<0SsmW!%mbe)dl0QdRR9_1(HpeuBk9^o8% z)iTT1IebLFoOiwZMv0BnBX8{{-}Zm~xVj{N$LV>?r0mX^QY#rZ-4$<${$<}!?EXO;qEb2!F%H~hnFYQUM z>69{RS2C;|%U|wQ=R$hqNfzo{+o~MOBY$NlL#PgrwR@CHy<3*%O{R{?R~hIWeRG~n z{AnBMOQ&rr56897uq^Zn90O%_oq9-5UO->LF|r{S{flkUhJGuHa*!3fQy-B`m7{f= zTBdT7ytS8dw=dW#Pk*b2vbCMaqcTxG{fpe%4))uw&R3@#qf^o)tLls2@*qds_Cwbw zzu*3vhjvn4=atiOI->4mqMcmdKBVsQRt{zLyEWcQ$revrLm5blXN(Szgj`}Do_>&L*o<0U&gx{iEpv?NBYrwwqbj%=?gb>*i__LaFln5}EvN9N>P9g%+9q%yAVD~J8+nXR_1-25glZKclg1^dcD zj`Gsh+EAY0kG{E2eH@c6`zo$MCgn^rEa;>%YIAA&bFVT;Pu}Wkj3q84{y z#HEbko#%g+qm-3w%B6lQC)u>lLw{478LKo7ql{#Quk;V+rU!IQ8tN%+Z2pVLa*qyvR0lQS1FTy^6i-2b?%b}{gS4#w*TQ) z`L(=wlx`_cV{OW4-?_?7yR@H@tMnWrCv7W#u&=yeUztI<^i4X3kF>S)$_@0ZI;U(c zU)x%{m+RaoZ-4GrMsQr(?F07xRbR_l`eDz}GxCuRNW-6d?Ay{u_mHXjs)KfthBWDe z!`6&w?A_w}VFVqH+6IqZ+xqvQe6J-YX_qVdji!LjZ zeR+cNwIARz<#t@%K>4(#Hq@T(Z5eG{TbXDR`I0BOlZ*Rp$+6`wcWHC-S6=s&ck2iB z{HQ_3Z6Y0i@@g5oZ5yc*kayd<*ZZwJ2iX)%pPaAp>Q@W947h z(-nWx)b8|B9+ipX%BfuV9B?^(Z+~cc+}HB8Uup|-QbuK&s-x;gCcRHu?geF0rpiU$ zcpLa#wULhF&~jz_jx@Ow}!rX1Ru zyoUZ@{d+c>9N0l^*EV+_U6G!0fqc3p8-L`Zt)M!tT=rd~EG>hraw)5{rluzi@*RTi zw}0yQ{LGGb%2j+$`yDs(9om=4s(ss5pOwBcXans*B2mvej4h z1NtW&z#C)=_N755q5JYtcJ&}vb&;p*=n`d@oZ@8^$r{%_3Voc_^R^S`q5>?TOhb^4gK2iC|~S7MIVysi0L>jbPbv+l%t zAnW0*_naA6^JLw+wV&3pduGO3DC?wb9dnKAtwFSQ)YiPO`Co9ZYpktn>6~%PV11Bv zE$(wIxL;abqvV{lltrE`gSGU^qFgOY%iZf-qz!w0NR2~tYI>5uHCvGP1}KXw7v;{PHlhs zq1}PcXFJrH@@&u3J9<`BZ`|t~&ZVdNj((u8lpE zWPRCkX7j*Z^73aZ4SlZM-E|}S|B#=of3z+UM_D^(eVnaz(2nCTzpa(SVc;6;BEi~5 zT;sTX>CFwSxAUxM;W-h{_<26Xy8*qk%f9E6!1GOxIp=wK*IFy+nO|!xyMD6kvz0-< zplp7(jCc>a?$R1^>&&GoFKau!Tg5TY%vdXEjiCI!k84ig9h0`+FQ%OC@r+BKcafHI z_S-tWe*T-R=KpuhbIZQx;9ZE`b?AEz;N9)sCF8wb-meJV_kADFM0y@q9(|5W+2yOe z%C9`usFF1~S%a_5$f)h$yn8(F>3(%457)MB$qDFy``zoDV|0b?uu1GC)Rxxnu|@nl zKBzbw9jm`wo5)VFMeGwhWglv@9GBO;KwfM`{Rn+R->AKIymo@EpzG`n*yoq%5Bis5 zwLSK0Ak#xWbOEbk2Y#(sfk zrUy6-=i@l(f_=b;czwv3cp67{y(3Pv-`}`VJ@BA z)*ySH+i~lX!Eamd^mUKt%Dm64?~IhTXDw@M+J4GTUbVMf7fU9}s@%@mZyU2G+E`oD z7x&D!NNbI*_nW-uTc8(@u5q3A1^e|WhUwBjy27t1=mxMs9pAGv1s|F&vp2ZEc7X3% zU(GT7M1Rmf_+9!J+n}Eg`z=4QHlm!4{|lUt8}NjEJZr3}@3v3--SsheL7$cf4k)^o zwfFiBewVw;W4Iq`t7-%AJt&(pu@lyo&ka16=lMG5F&tS~ zJjemxvk{;UpxkT$o1mVc&g?LoQ`<~N?k7j-u^XUV%8b9u`Pvwaap)G#W-DqZ9HV!5 z7>DC?TyKm+C+Vy|dg?r#tQ8zybVE?;`M> z2j5ljT^rvP^Bo%R?Do!A->31t9N(Dix9EJ!!S{8(75H|8?+y8Wj^Do91KzLbdpO|R zJia63yE?uF;hPYy2m8K-<6AwpzNOG>)mVXx9dS({SKbuLH{9+7rfT zS(>JDN(0>QxMTJ~y1s4Xdquv>*>4E>9+7L^Cm-dKwr>Ub{*rHERA#RhGyFk)h93ytpMp#Hj>B)z@a$#y4$z z-^O=yymNe3F#L{Wzs0luKi409_s#dpe51_w%KH5`-!oe-@SQW?z4XmD-#PQmGy9G^ z@4INejXD3TY0nz%_uH6``=*+2skv7=;Fxd6`Tm>lvia7g`&}<>`>yxfw>F3Q`>kxg zN9Wt5w!X)vEY7)C+P?YbJ93lx%kaf!^qu3sk!8yTM!4HG)1m6t4 z6`UUU7WPkqvjgA8w(olPOZTeaw}JZH5vZf@-uYg$_8}wRqw~#fZBHNQig5v79>0qp z(Pcg)ABA6|fBBw^?|k8hD+1r{T`1p$Tp?ILcw4Yr@X_Ft!54!QgC7UJ{eFGmJMxbO zi%l#x@vLC^V5Q(if$fT8tUR&u#HztdgVlmHgS7(t&OI+!B1kbO9-j8_=)3g3TYpn< zZQwiizY2Ww{*vJA;5)(BgD(Wiw_mVN@ZsQt!Jfe`!8-%rO5Qx!IG7o{GI&|=!r=MA zvx8>_O9qPu93tNZ^^IuX!Mr!{4a_?N-!1lyHs5RWy-VMC^PPBG-z|0w>|YgJ9^gyg zB=xO19EvA#B%4v-dEbN^ekaZQWW689H_rYQdd_#-*eE`)cjkC63e?wS-}t-z{uy4y zqc~HXhVRZ#=i~BE`Lujbek-2jV~Sm{tHwZLYqpIW+ZS^acVLT+qu6Tk!p1Mllk7ep zgkQk7z=__A;{E^LC+*$M#=G9H>fL|FxA@Vx*!xy}FU+^fz`pMaxW{?d;nQ-f(Y1GK z8h0DVyB>_Ajl;dG#k&AI_U<@a+PIq!#3$m{+46&Y<3M@k=bhN)Si{}Et)}n#KJX?1?)E*&eyiCxhw-6rv*Xt#0{rURn)ncWs}mYcW2~Xjx(}I(NuLs8kUkm)kXWtLL z7r5@^!0}T8_naEo2j^Xj3vrx0aN@TE=WL}TUt7PMe(yPV^4!TYgC7La!k>=eUEB>g z8E@ikHsJRG4*gYt&(-^e;I`oI;Ll^+GkMSCeZhmlU&p|=j|RRaigVc_ZQPzSQ!fOOu1n&;^4Gs(N z+z*3GgX@C(f{FYs9pKM(gH402gLehH1bYS_2@VVn3yutq4UP{^4o)58q=}Ozz7ZS~ zd?7d_aL+!0GJz9H|8(;Rzf$z)vMnu0!gs%(VaVtOXzrg;Zcb@yk z5iaKAid%?h@K?pV`LRR(6^{`E;G>Ekn+M~Y8rOjNj{2kwr+w)7Ufj#r$T-UQ$=K@l zz?j`w(3s4)(pZguYMhFTaU2fAk#zY}0lxBm0^f`DZAIT)^v#%lE7CV&fKQ6oz;|rH zH;Y}%ca_e^gM9+&NypZA32dd;^zplEebdr=q22E~X}Q*K=L^^Q?ijxr&ni3q_6~pV zl(Ti-zHcB%Q+Zk@ezk8)`tD`F&+Z$OzFDb{`<}c0?pwZV2mCJIPv*1MKlROEJmH(e z_?fTDU*n_V=D!E{<=z1A;GZD}-8FgF=)3l~>E-}WT^C#v{2{RQ+wpmUYxujie%~1I z6?^@)lh^*gYjHmtz>mB$kT1J{_mt=U;GqEDu^ntmznxs)9QXMaa=(A<+sEq!>j%C8 zN={n^z60tzp1%FbCnZzg3$^u~WAK}P(3RbTevh>6PdDt-AJ_RdG9AOSuBVIig)Y!1 z?KnH&4;0#*4$ues%s*h4*>Jkvc#Sws;~io&#%$svVoc&v#(Cym3^>eP>W};)oNi3X zU&8JDDRD&ae&dJv&Vg?#;0iGTu>qXJ_rp=X+w0rH#?pL6KI0z(+=l~i4ScJf&*?ks zwvPGk`rW}_0{-b=1ODpc!IMEgr@g?m{5>V$&;DodjNm!JvH>4|rQpQ@Kll~FY5`tm zf7TAx4b}^6`OA)hea8ydygJ}lZy0P8ye{BtZyLNIV9#*>Rsmn!*n&OSK6qDPf7{@# z!8XC0gEt172hxz1blHyA1j?b@D+bF1e0OzRBw$M(UEt9L{ucOF{awLr!3_bwoIg(H zzU|Czuo3KoI0bIU>0&?PKH85@jkC3>IJ5TUI}dqV+l#%6D~aXOi^db!{?3WxX>$&Z ztD2t|*D|l)_yONo4A9uE{4A!1qs77UoAYjqA$Thvr zdB<(L?|0$)!ZBsW)#`+=)lpq>HuX+0C(efe6KvXn=N1$*avYXwt~%ACs-?Bb8z{pfjY5GI9*I;sbGl!ulsv!fyYL+ zuH%Fo7r1dGGq#f)&kgwhV&UY?pVtPy7b-@=|JF|8AYx$uj?0Jhd}7CJJWl6}(+BYo z@w8`;*K>$}m}ldo1KqKXj=%n|z#afSo9fG(*XaD5SPDO$z2M8!)6Q@6NyJpl%iR>1 zFQealdvjg%erP}R6~0a9bo8&TX~5n3-Vp3a=X}cf-LeDh0{c*#p?`~ah=J();w(6U zFIN8@fAHsUFCG{F;49)){I)@W3tt(Cfrwp;gER)R(mz)k#f16X;vvT4(0wsczA7{( zJ=``9Efy`7!e1Ao7DEwN5wqsA8}B!M^2Fp5Bj250{Et99MO*;vJN8K6JRjfn97TTr zB7t!LKYFP^47$13+q>tXYCojW-pD! z*v_dS|DmtV&T1FvSj<=m_|=8q#$W#m?lUecq{Vla5C6Vp81A=xw#r;vtX^!7dP40w zU16VW3%|(-*sy}VhT1mg-6JhBBOh{5cl9!MFgB#K;*j*1Uej|v4BNv8Gq0n6naAX_ z8B<$xVjhxjr%&Snb^}K&AK(sgcVqhv0_zyeWtnfZ{y;29?Dn9*_v3LEpH!TfpNiYA z5BU6iT0F?t6}Mg`;0xndyerm?lRpsP>rV&x{Hwvqfta~C(piBR`nkc+0&$xQgNw$n zKR0mv{4rd2PT<}%g42WV2Hy@&3r-5Y5qu?JOO6P{<_k9GbAfo)p<}Qw$|z36KCx5g z8^rP1EoeNyHfPsqyN+zmj?;D=*`4jCZ8vS}fX!iRW(2Pa*q1c}Hge@)`QW+1RGad| z0#A&_P7f^bz-X*s+;vUx%fPry8?e>J3dV273Fd&=6=OT?#^#$x-Zk*uX5WIarm^#o z=A8Kt)*R9awu8;U|F|Bv^F{DDKZTy*YaC4n=_8+tf2^ie zJTqf>N~}SgKs?~rfw`-T1M^ejsba3;wI>8(yI%>u6dWFWIyf-+Xz=0S1Ht=)-GX-q z?+msH-V|zS#TIk|bT*%)t_oDau(e3QvdCN~N;2wod(7;GGD8oWNR1!=hZje)%0 zJOyu^eCy=fgIR%c?Har{*ga5QW#2D2I5;FYEcjgT`QXUlm_Xf+555tIb2|q1=LBT* z-9YSF4El^fY+4L@ZqWHZay6$1_Q~3OBH4@Mh|!3_i}Q%ZXdf{c@n>wu?Pu`&JCt$i}7D&p3iT#ev31IFHWk5#VN=XPm`vH}=AL;v#H1yHD@B zmctqou|o6O)^J$Ab6p^|O<(yRxR5VGhr144+)>QL8V=yY06*K>BEF4319ovs@l5>5 zcM@OWYw0uAe~GJ@voa1BOZB}O~Jjvm25SW?1xpA19+-Gw;*Y@< z!P&ug1LYRS|7@^-ut)HYV6$Mo;Kjl719PoUPJ43NJ%Ksa-v*Zl_}=_3K0h`v1~?%2 zP{6L>T7EtMebvA?L|kd{V8Ps)lVTiC1r3Hap3IO1GC4EW{82jVQ^>Fk|Yi@6YS z^Zf$fRNgz-b1;vo-PuYyfLF!K*;3;$9BXThrS*60Eso`<<5{{yr})|a=pA(Zn=PH> zSJPj5Y|fH@ZER>uzuA1Y-&!jEbz?MhGY;iHi)$F~8Uyor%}w$z_%(c7<8OW~f0qBs z7c`HT(D10Q>-&wxHa;v?izmc$&1C`p@Y}qWIW4?m-V4WE5?mJeZBFmUfw?vFcE<)s zk8#xGQ6t{_h4y z3f2zR3T$EZ;Kjl6!Lx%UgT;fzf<=O-1RCbC;NjrG;NIY_;Ev$7;MTy_@7sf0f*XS0 z1y=@_1s4V92j>Pq34Rp(AUHkvesD%`W^h*Eysh7k_d3_wcl^BI(%`D#>fridULb#E zyC=9mcpy;bzXr2T(g4Kdo2b%*9FR|%+}4YO*s0wV?1Z#Iis~T&zyMXXwLN+6VJ$z=Zvm1 z_xrNIz3_^_*4mnNgVzM>2h#U@?ZDcb)dSbA8kozK-jczh!9sz-?<2tjfps`H1Xl;v z;IN}t1eXNH1V0MC71aMfZrX9vjtRt)*pE*IYzV)eFTZynb|h{jX0%lxzAr8$7Q|O) zSH$$ieym9{pCR_sn7*}2=0ePs7$5P|FAgpY`0E#p!9TEX?)b7m>|d;{etqWv_!8!Y z&C!U}9T#-HB42~A-#G!kzI8P_1niL*p!ooExond)i1e4w;&~_b$=ohK#ypW{kJu(- zL}SIpgW#`MYgSax#Ru;?}Nus!Ps#u@mk;|k*pTnqLc$7RMQ_}A9&@*KNnj50H@b)RvGx@;TZX|~dM z0|$}|IpJD-+_40io)X|jvhMhS4QIb=!;L4L8;;RVoIG*F<6oWp>S$bo-=_lax4rPN za(q0%@gEF~;n@7bZ#G5yxmLS=F_4FKICR3A8oFaFV%?z_m2m{W$9Pqol5ShyV(!cu zH|yl|-_-(he8&1>1~}v|0bhKX0B5~6*dlmq@UCE|VAo)`;Qaya9AdAD`EU;Q4EVb^ z-aYWXp!>G(4Wub;dAu!PAO4rK`vCf^surrtPC)tE->(If>iVL((uH%K<12qHqmTfgu7ui<##*ELh` zTUGr}H=KL!IeYEB_CB|r_ld#sVCUe$f!akj_zr=;Yf_hZd2n997P?fC;tH-Y=i|Nl);4Mk4ljG*?JEvC`aQ|R=@h?wBGfZb(tTYn=ykc(Ak&kvxB zvCAJ1d}f!`-q2sv=2w4Vm*r&C*ECNnK9IK&AE>Dj7trP9V%5g157=9AZtG#bo}3eX zT&>^n!6yT`N-?t>7!8r_rCrO}sR>bIy-h%0rYV~%{2jFx>Ne;u=8`#OZm|peEAx{7 zO}F4%u_x>io5+5#f%gpV81Rw#$5#bc1bpH1gP#X{;O__D4%P-=5BR?SJNQEI+2Hu# zxWMb@0^k30@Gmp??0eg+367cJGaKfydc)1=@1NfA=?$L@{xSGq@ZR8E!8?M(gTsPV zf%iuQzCSW}XJEVHs15Ttdc)Ck$9VsS_iy-c@QJ`#wT=JIU+0s%#*+g+|91lZ{||y6 z1wRXZ5%AN+3cn8c=vN2)a(|0%ToljzmN(uu*dn+`fM;;W&H=s<)8MWH0{6-N!h?Kh zc3gZ#3sTR>&&8$R4{+-kuZh{jZoZ=fv8ToZThH#F7}#IjhKEP^9zTogP783h<2&a1 z;Icr?qCYX^T?6BgFJNraf%ppi24h&xRlfAXfbX$=z;3HCmY-y|ZN>4Z_8i4Ah7@LCfF^YUF3`kI|sK9 zZWRy|ZdiQ7;%kFH1iudW+kEd|1V0V<+TROK5BS{t_iqGW3FtS+2WlSQ7rZ^tW*qU- zfG)xZXaAoO;DDzFw2~(UVh6tY;{v?EkLFK5Cg4wd?|Z-F4*S@*`}=N1U^|?`=iV`3 z%a;UfJfEAbzHZTVbJ~>JPMq@PU`@a#)4k(pOJrDHml*_ORH_OhK8<6Ulp^` zTjdn3eWAlPXSmr-o85HFK;AFm%ABq<*8I_SE(y$6zPEUo4KSbSHs-%M&)@rMz~`ec zvn_maP#5i${_Ho~usbwnc4M>=*0?x6jfkyK_e?WlGw9Ff1#$v3XgZU&ff(KQez%={ z?5}p3-C?uXxN1($$v#={qV3wFU240v71yzm{N3Yc_2FKNN@r)zj_SVSR_|^_hF1uh@96v^s6}n|7ZwtK)PJ^nLRHt;1Y=)8aY3 zU)%W>>WKK=az6(KG)80FIR8w*SNUc@o0OY0uO1xi70_7uUFxglz?KJWJAIs=$`|G5 zvQKOmKlz=(u>mdPwBSdNx;WcUJ5f_kJ3lnwTdPBN40e*<@W6ntx>Zm=m0x|U zAZgP%U-jBW*Us^aSK0&rHT1lQk?WtE;CsF}-Ic+X!pw%B+6 z172X0@c{o>%@yB*Z8jF!XMXe-gMSYA(#ER#KjT(yF3tSN;D}&Vz*ej0RyQI~WnD#U z0uM+%fqWLsrzpt-9So7TuSo4@8R8^KpHU)c^1wx?y=MIe(bQEAU{xS|ABxuazs#Fm1f$QyD>H&k}l8~o6TeA*)_J5y zkKJX%y#n9T_h1_~1&6RNBah$&>uhj|bux}mZ(xhr(%N6|edjmVV_VmzE$q4Du8gp{Gwd>+UK83N%@1WzUy`=3K|L+Tq3Cth!iZ8}bGM~k`)~B!; zv{7~2;`U<$+5($KS9w9ODp(Vo5YV!J6Kn`J-(>SSf1A(kJstjPiytt@7wjY3&PKAEG#wf?TMs__&33d4x-(yx&H^-Sxk{(d~yf32^`1y!2M?5#c9d)1t;(9eEe0drNf1a;>`+&~O_m&H#JL3Vr zd-d5i#+E<6H$M1vV9c!#tN!QS)9)tamI1-9R3 z-{S)Hgyn=!r+!fW^j!ga&%fC_kULecERIz>&0jNLH`+ot}52F}K_bKt%2{3fTidR8C8hO(V(iDR-= z{5$+&?GGJZTiPZzix260;7fY-o!@Lb!f)&to5z+puIs>C1s-D;U8C)aa-QGIajw6< zKAxjbh!toQxTgGqlk`t*BpV0pqwm;ee8irztB%R1beXpz zdju~G)HbSR#Ph9Jd0lFr@W1%Y>%jpVsiwvDjpH24G5Ph?;+&726|Wf|t@Y#=7%TLW z!e`&N#>Vz?D8^QwSs?zCzc=pGqte1%kMpU)GB(AHd=mBj;>xE6YPHnX(_zG*Y7Y56 z<^*5K920w+$NW0>!J2Zl&vIjGxoNfPG}yg6F23WO4drLcCsM=N>GuX74L%WkGLYl^ zUxE4-8u)VpwIJ*<{hc3uZa^Pq=j8Ng)?R7WY69uTbZgp#^IM1i-N2ep_L0BM*S<91 z5A(12?tJ7;12&wEzACT{-&`)yah-=&{YL4o>~mjx>Wemon`ho{e| z8^!nA1Zr96);hE{`Zv)=E}PY@{3>-SeDVtxfbac&Nnm@n{dd8Y3;6e!r{?9NfXzQE z_+G%LR};pzTl=F{Y)!zI?;8D&CXdgCe;}Z{u;s0VF^A3ReFO7=U2oktAKqLscOMmu zHeN1stHAtYKg>}Y3qOx-Ve{B9c9!j`KgW;br_&?&XS4`@E`L=W1l#@LfX~X;0zN$5`c1(*g7*X;3_cS4Q}FS?D_j48;9bER0vgV<0`{Jcfiu*};14#G zHUexZ8_MskA59D46Vp4udp(2<7~pP!eQ7LwZX6^Q zP?thSr@_nr;xqYK8okf*4S3FR)oqGv9M`_}<;4wR4!-~G0)xgitqX7)o3B<~E`zSV zD$w6-{(b?w&!2xv@YtX_`yI39ilA#P7M)#g$XD|3 z`Sxl@*n9l`;^4Uf9iCRR%j|iG#s~kMU%pwu&a?4sI{&*qH^2MLfbR11;QWBiX1i%L zwd;SFTQ&68|BhM?9-!qo{x1W^{B~g6iaYu0Y&_d;OdlP*BVgBQHPvY3me_bc+kSyK z_IZJrmS4ucS5v36^8dixp}F$s%{B9``BQqjxygTJ=gr^q12%>o@CxR3zoWhKKh1ku zE1S&jo)pwy#SQ#b8j9`sy^g_8=Ns29@Vhqz?2opv4}3#5jbA9w#y14-eODh7PtlE= z4{#hdkze2V!@1Za{H6WcLEq4B?bA+N!M2J^@D1BWU&k9_4K@$2(R0}1#w+3z{J}M&a32n2FKa6u-}%HVY-ROy*Dr_ddil|GaXLAlmOtwr^TE}*^2^yhv752LS2rH` z>GCb*ePe>16w^H}5Yrh4o-4qoW_#sTaEG{oeHO>K|MC>>5zuh-U1J@w5U}gj^4NLs z*>Cz?OeII)`0k~+fWB_r;UZ(ubsCSxC0}Hhz&Z;)i|cr8VC=s?SRJUDH($;O#K!zA z@w43Oy#i~)&2=_k?5`fex_z2~nu>jb!vbq({x#qO^M8L7@Rj+!Y;^SwzAf;l`M+$q zXJql;zZj_9R(E$uz_zKq*)w=S;9PVf{;st+?4fl*?4ulS>r(2=;{;r!u7v%jnXtX= zu-s zK%Bw%*FXA7Tq1takDi-jEh}69x`1ZG&U*&IlLB#-+9Y)>cMEPE(9-z%>w~ib&z@o9 z)v>efe0+6-oU(I22cOi6cvxmSAEd;;ORB=FU zCR@%2XUo_~wvCNr7uh^MHLZpXLXG~_qgHTn#@%lGEj@|ouq^?S|b zB7IjoD3?j^;MdbiXfkKbKtG{ntGT4Z(7D-BcAPEVJ=iNyBd?B99#@V05rJ4jydka- zZ+tAEE2;fi9mp$q^_k`*2Ki7xb2=(`dq9_ZbMVFiVj1wAx*X?xb?}N{??4V&4UW2X zxdGQL?<)W6I;vTnkTpxzqsq1ZW1vrQ8;*NXP%Vc)zcjd0Fs)T&=lS#ecXr)9TN9`` zRD;NluL|(@Apvb({`KX-zQK!v)`EL>xN%@T(!&F_HSGF*0{YFJ0&$>Pm)dv!_!R-a zVf4X`J@)!b!M_B?C4a*hl|SW^e=y*e8^dfk|BFAy2UjP+PvdKuo7Gp$34RSfyr?fu zzZ~mb#N*~J|JK}P7slsWHMhpro2P7jeOf*>8%>)5?|~1`*5d^8oS(@jV|S)=nm=e> zYlFA|*VNCx!%cUX^T}!bd_}sr^Wz7+z>ntl(}37-cACDzMpye`E9pOMHOSx6enu_G z`+`2?ef1%0WNAq3EqGrnq}SNDHkzh`D{v7zE#FYToVGzD;{UVPw2U7FG%kJU8ucIF zxOQ8tAx5Eb>mxSNGke)zHjxbl zGUkVQ@|D@z)l>59adulgdv?GVqJ4^``Aq5_ZV2SN_*!Zr(d`KyDs z2K@4m1-OAfPZy_&)7{Su*mX9Yzl$^2dpd_WKs^k-n?EczQ160ES0@?u&QiW#1mqbY2mNb)FV14|WRp z_;(IA4Lob|il9FJ8L3gH-|*|zq|@bSH*|Zo=I$l?Uq7DRmy^Zyv>W4to!>gRXTZi= z+s~*OYxEW4k52QWz_|QB!M_J=y0L6*tAR1@&4c>jV&&E+nydUT&kYfy@~vo)=C^sy zHqyw=|AK!6KC_qJ(q?tdEVm^zMtQ;0bj6ta+B=3bI^F%Nxm(=oyO5;7SO{F z4GswS&~$8T7Om&8=16V#vf%H7M+aVa39Lo(+qfo(9%}=qL9H2k@c! z@BCw&NV{=Q7t*b815TyuRLjN@#Wc4m?gRYM=higGIr0blUmGtk%f{1**m-qZ>^x0~ zUL*&~z8@Xb{}z*)6Ld#D_`e0_$$tcVq5lkMEhh%22B!z|Q|$at19R-W;KG2tm+KM_ z10Sn-FFu#Kd3B&hna;`|lUq{nd>;*Xb)XjH&jd|<_xjtxE=d@lHW z(AO_4{=(d6zlj6H24V!U!iNKSiuVWd(&7PJfkO@tb`NmEBZ3`+?PkxA#0}!J z8?(;%>fqM_z5kqm-tV5#d(`h(lO&Ie4`}&#KpoCQgXU?~)2L-PzKp|j0=B<83_nHv z3;*R~f$`7B;p?ml)IC;1Rqs%UTWK_OS8=R4MUynY%r$-)e++MoFfaLd=IXA2m{>d@ zR_F8M6Z$J2;b-FpJW)-DJ)wi+2eCVwf-}Sd;Qdq^cMO~~#^$Y?U|(uu=tcP(YCYiv=g}* zv7$IZ{K%hY+l{fk1G>IgQ_Ly$l<$}K|3Dx%{rK#;k84sVz^1DiFy`5GxuvfM{GgM9 zZw36K+H!dHtdK@4}Z2hKnoc!Q&fY^Y4EFQQpxFondpqI2xM$UjQ zO=FTnpzqUELVJs=H;EJIE$ZLJ3T!Yfp4QJ#w`PhTFGq_%=tgWeU*D_GFA3Okem_6I zYptsB?-p%Gu0RYz<)clgzpO9^UH0&G04TS z*^XNsm%m#0&hPfWJdhhubI;Fb2l>7H{IvlMhvv?P^7U!<)!At_?D1g%8$Gr2d~_N* zyUyOvE9*D=@qT0L`Rw??I!}DT&eM6Sr;9I~w|N5fIy5D1;=9WgfSdrZ^cYd>t>-}+ni(HF#iY4%X^NSIzwUG~aYOqhB){p{w$ZundvpNk*967Y>r z4EV|FVC0nL2Ixk#BKZcKATPky(%9Kv^)zAzS`0faX7Id(eS!l58V}u`<|A&P5qYi7 zPY0qQ(SUFQKA`L4jcPuQ`J{mEvun_rA^Bo;`bz^j=i3GT@D)DOcGvN(K<;F!1K|Mu zrVsZGo*g_s5MR-Otbbb^c;@4};0FPXP99f`#=rOcW;q;MzkCjzPfg~Ez`dpSEDi1) z7-zeQmwuC-W=?8hYzp@B)l=FRjt1v#{6p z7c;1%VVBj`IQ|L2NrCg7640;MX+An=%ZPq&8yK~(>Hgn|mUBkH*Qe#vbnu*O#d)q9 z|4r>b+h2``&#w>lr+&uy?i<~PHq&|)d3)o5FTt;<_QLNFSJGbI9*8+<>c%7+Z#_nJ zbiTPU%+9ls)mQ3&v*GOMXt(*}muv)=&g}WT8Z)oQ^u1ShzM#|4>iPBNmD&b14s7_f zvu8;1)!ALB-;Nt>Yd^=-HhiPaI7$ru{nP@9^|AH%tneLs|J}g$=CfnCF8N>Q z7hB*OKA_y5=ZfqUsN<6(cwS)bj=GOG2S){bN*o}^cv3)DI4{5jH0j88bFqN6{pwNf z6^ILHL+lo<;(oycf*pc~1Z*1a*fqc(YFqXWUKzYTz!`W#?C{xut&~r|3H)d=L+khD z4Vp8cvxo=i@AA%K0erwux32f`fmq-f!E*xseZju_PVH&^{n~k4vOM7TI}T0i5rG(^ zbw1msR`ou?U4q(vaf-HyYpkK0ec={!G69j+X(^;&p<{y%E}hvxrFJTP!i#RBp_PYRX?_`tJDX+Ly9^*iE9 zK8kT@j)-B6J7bG}DLxPfcl{Gykk2v?<+sGo;s$dQPmJqiSIzVx8ftmM{N~G97c0M} z)x~1L!p1J@~%A^?hHvy&4@FJ>OqV%?NSD>jSX?h%tb^@A+A;3Ft&L|5r`mJHOk` zK91wq+NaHO!i_QLMYzhfiY?R_wQdJ5eJ8*h`dj`)j36!$58zjvjKAqSG(8#){RSWK z8RT782I5XSyjYaaaabV#gb$22{)w?DezkVbI5l3?#PDg>2K?gs#Ek*!|JT>A?=L1` zgGczz?`&&%1V`Wiwx8XvPCsAacJsgNd^H%p*M`6!U!Tv<7Z(H2W5Bw{jiLHJUyi;{ z*XP^aH!!cQ$zs>#`e-?PLHR#EIQJZ}u-ebsq;Rv>@hnjbktHlIyVKX_`u?w=pf znB{}jqp;Bf(sh>rcN+5h)J+oo~TJj4l}wa(s)1?b}J z^n_|w!^1iIfM7j^Wp+|-KPY& zg1u+!A2xdq&V93PfPe3N&dbi80SOqC-_t5PuEYGSGD`rP<<>IZ9g5Ao+>_|9mxr>`Ca=s z+WhK9>^|)X%=dDD=Yg{Ua>%qKF+l6{mP8L`Gt~Fc@oD@Uq7D5y81u3GP@vls{jT5h z<#EQS3GwZLtt$LpeP}!u&go~mhizz07X{UZ@C=UOL!S}wyZP%lhPI-{kksCD~@gF~buYi^#Zj>)#*ZKHjOm$+$S8Haf$J2D=S=9vd`^Bop z=+^>v{)B*ke0sotp?&aWjBEZs|C-IN|ITk`zkT*SpIJ=M7-9O_c)_;j%Xpl|0JxyB zK;wb(f_cWz5(D6X+J7QMNh>x0W@|Nj$y zkdtJ)ZWnA3hz-;ai3fHJ9ue#ksA=c_vxT&eHwFA*+Bj{Xk(`iQd&TqD%DXG=>XZ;yp-f`t|theAK-2A8vRH7Hv0Te&a-`21djt`xlo#kIKMnZD;t%}527hk`?|m=N$}ZcMZN?||1@FQ4 zVj0`0IkLUug7@9uHue((iHr1y>(w^gJor_cD|htY|Q`LiauTe$%`idPUPhz08ZUyxh?ZomQT zKHs0Wf9*``Vc%~EkflF%UW{$Njrm>novr7GvETF=Hk@y!Hjh@bB-kntudDOiZU*nC zuuZpP`yUX{e&qnw`_%Tc2V#U}0sG*Yvp7J^z-Gt=(2wZ`v>`R)bY<~_9OL0ZIYGR@ z7L^m!r_-seA)r~yS#Ah6S+vPq4)eCDC#7f5wdFN&1G|SC=tH>S$pLP7aqxwXku_rSuK45B0Ec)+gPbh8?K$ysv~0OrO}PbrJT4J$ zu=ifUcYb64ZAaI~17Z_Apq|#X zijUNT?h!m=c%HGC3U}(KZ36L^e%v&;?iSb0#R(VY|J#)t#?3gw^BuZAz&bC_U{VVt zUcl`K1mcA~0{7jR;3M2Gz!755#Q`0&@xhs?T{DOHTWa0ZzO4@A-ONS)o7miZ<+nLT6I@C3Wf z*0Jljp}G)Xmp;LUvYqTLTg(pgi)k2qWPbZXdikl*+gA*L(5%`TLH?*KZub=XV}{|Clq-7VV;Yxn3NB zOXy^HL4WBx{DsS|531#lT+kXFc^Mo~-=CeI)?CW(9TV~;LbzyG9-)aCpvIy(DJEt@;S_kOn>zuZ2+R`c1+N4DE5jNhr@ z| z-~)9t;srW7{fJ$~1I;C~|7(NVe$Vh3pU-nqp3TFat3UN@Oj=Ikg*(iiC(&5p?sHF& zxcgkpaM$FUeXs6F%-|VrYJbEHVh#G#CIQ1a#t(P|ci;-#gNMWo_(5J+yKxy_6E|2J zc32>^Vxb&ejdBe@1yVV z1?dF*#O8NtJnRZzb6m$IUf`3`d)TAv0=B8PexbOaHlP20e$YHXa{%?9>qGOAp}w}? z3*U7+zBZf5$Hp1;;p@lq-@OOF731%0Z}|m>&|`4J2?1M87pgYIrqh4e_-a0MAHW|V zN34bi@B)62JC>8c8MvkVf@2Eb`Q5hD8X)|k4e}v4i*BWjc&yxq_gn|QET8BXeI(w{ z=Q!8BpeNDY#B=T)&UQcD*VgOecew%gej0D!8F9u~FC^}iH^ME|miA73YTWTTXiQzN zWn8KO#V_g@-W9wr_;B!vfG=ZAi+>w8tWExauJrBTjDWUe9*7-&8sHf5f*1kMY%D(b zbF?3Efxj^(5Em5m9kBufVw>^LcYd>t?QvG~2I2>G|6&MpR~%8O@dWvT#uj)Chv6^W zAXX43;1^tkqu76bA>04lV6T8S#BR`<`HwUQ93baNgIE<Adl>9$? z-wXJu>x0G#v?CXc3uqsA4rt+92Xe@seSiz(6y!+v4AdUcRo)OBF{wSp2cHi-TR~jV zH3%o?xf7lZ$M({0noq_Dv?Mwb4M|R!9lv$JzO(b}J-g5T7v+I1@~k2HQMtf=lp55#@5O`ZmCq8&f(D(q)yYIN4MkrpS9r8Vl8L=rJ$@nuK zjqPg#n&;(#Tp8XlN8}d17l^;rFUT>_Zt1vmUGq-t{>eaY;X{F3@=?J%0=Z<|V*Z-P z>i^7Z&uWt+h<9i|Bg8*qp#n$RzIvHsR3oDkIfrvOr}&B;ul6PO!pjTUbJs^d za^3ns?bfIv=|BBRGol@eADTO$6NwXPjI^TK`EgyYdv6@5F_0&&jhBm}8Hpo|B{6FC zBI8Ug)QuXK-wbN|s~decbvEoj?Fa`L(}mCAxAAv*h;jwK0Gd(O z+}~VrzV8I)%xM7*(iT1#PAYu%z2CvM_Ob6mz+)qvYrHOPF0Yxt>u32BbNb5Q>VVFL zqs;r;1>z~adAqNN6oXJBscA#0XP3j!wJbw_#HHaTB z4*b7O;s@IJW{WnPvvcg-cEN*##td|JJVAFqFgP@zx6obaNOBGQZ8lb}cH|0c6;29p zg`6&~uwH@gudat?gfHks?72L!I704$od@rIhfi!{Tl?6T-5>QL`3*I%&L<|22NxI6 zjqrfA!((jVI$jshjPQW``V)i41i144f!Kh?Mkm6n;)81f{U=7SrUMtB7^w65Lhz{o zcdHFh3#^U^2dE?92Rtzlv+@(fn_^6{v3qYEh!>0vx+vSvFESU5P4kJZH`myC{uaMW z++dEf@1tg9?HIehZ@`{6-%vX)hEOZcezO&9H+#*7u+MCBZL+!{cDeAmnA+>{`3-Uw zwbfmpz-EuNrfeI`SDDWL>oc81O(}o2b*1dOIHqgV=876P2Q%=4N6@y|Qy4X_F~=bO z`9tvg;L1Q8G;#vFeM!KUUl5cJs(m#UYHTycKm2#U!<0Ap^&@ad`Gd}Oae$9TzR=eC z{jR5ezkV6l=IP(A{Vrc@nc7kOif`Rtyi7BcZyNInI9@%$7z<4Kp>;(33jT#L_<`A4 zu*Ly>R)bn1oFL{sAz*9S)E@+FHamWHKxbkr8v_{sG^dLKIb*qIxncHGOnuF)K9n8J zZJw*sWKZko^Yshg6~5#DPjd-k0bq+;N7R@=EHK6Z)wM<~Ys@pykmy8gI6tp?5uHh# zFy<53_A!tAv{~+;F@yZU%Aj$BXHU={4hgI?Yc0~d^MB-hDELI6rtx0`&pDuFpC0@; zkOO7A)S=3Yj`@Yn^Z%+>M`HJu1osWp*6@w_$b98z2c9$cvOw+`FUT*5A?Pl+fvzOa zV7&r;iS6}FYg|F6q2@}@PTtWM<)`TzC1vrDv2k(97H#I}e7qI{8T;-C* zDJKSwi;LtCoJUUFIh|Xcc}+0J2XD+;0x=QJYpz+$^r!$g>W?jhy9RhcpZOCb=uf%) zbwT3<9PGJq@;qt))QX7_#Cq1*Uwq&CB0j%)FDK2<=R2w?|XtP8VD_m|I9XC6I>mz zl?6W=YUf8Af5ngoXLIQ+>@TqQ)xOwg_L|+VEoax+_}YFskZM}Asxc2DUqI^t?`b~O zduU+v9$J}mIIp&_{dmdsxJEoS+I)S58}%K&6bnstA3W_IxOeP5t{1miAKsjTd)|0} z_G3KIe#8W7(2Ny%1@U8J0ydxjp>~w+Bex(1;G=ZCFaO%uzk z+v(33r=7|t$f2>pY%iNz&4qtlO}@6ac9yN3f*od$?PFgWt$F2`j_n-IWsb=!;|3go zGyLI~8z$hZf#3YuZq%H*4}LOd#r1y*_;>j04*@Rw-2n62Z?+rtF72ROO*uK*A6YF}`M+EqFe&fv@bXTWyJKYS~YgTNoZ2;@omz^!MM zci1}MH{%F#1Af>)puh9A;z|E(dV8MEtbK7ED_=i4{0%ntEE*(C6u|IVZ5i&nd^Ux5RB&t|b- zo?+u=(?QsA_F5j54_@2M4`-XbvemxlpWB8lw>|&8Hom^R-xkt$*!Ok8`N7o2v+r#3 z)V|9{$Whe3H`iSoFCTGour`p3kUy3Skb{tOkSCB|Xl`N5Ny|gXQ`p|VCk2k@9L|el zv`4#e&j{bmXJF>*nY{NMzH)tdMPK0+eJc(UN9glDKgfM>&%_Y!i~B=!bFXM|G&%Rb zu>?&Iuh0O+lX3~-2il>y6PL(i;StY?#v@}*lK4Rmfv;kWb`1h8iH@{3pvja&=qh8Z zur6x~3b6zDY#zuZh#}zOKpa7X5nr1tAf^yo;3M1EhhHW3Hjl(8VsPh&#w~ucjq{07 z9G5LGpRxNRXZXynv+uTZPVI3{`+McnX@@r90^2)AIZ<3fE8q_nau&X$Ay$h*tGizCEQIALiZhEUJt4=3PMoS>i8AjvV{V*hXY z>Oek74H8~ZGl0WAlSq7s8{`+32kIZh4oikSv@v8nnFo9@+NF7=21KrohHLK0LyH&a zz3Pzodbq-T^{jv~ZlEoxJ+bCgjsS18j&+wj8yIi!HR%iD2d_S}6JiS6$YBG2lm=m+ z#v1GjU7M|8d&&{GMI3Nb94tJ+|1W$V!DnXUaEVxe&BF`q-{k>Yh(B=22)r`FHqh_z z56%z^u&;b~K0Hp~*N^o|?010^GhTp|)%_X=W?u&aQS1c}U&8a#h z_kZtzA5lBswPrL){)SkA?_qp(t=X~pKb%{SG-^b~<+lQwIvaaVAYNcE*-n0rKmKsz z1h$Ut;|q=OUB71^*-2=vDBoXBm%m*2?EBhP$L3%2z1d-Z8G7;jZ-(~xp3&fo@yXd? z=WcG9U+&!6*JC!uu)p|+&rjEKo#KPv&)~i9aEsrs4fKP3%-J>LJApbSCyme!6kPiP6K=&Ug;$xxu9Y z-CPapor0}{odWhy{#71YPFB5+=O6K_KNHZ2Q`w`YLmnZ9|^<@a?o_Ml>seH-a*dgLBTcw4z>P6Fey)?|IP`17~o|6jHlg$ zu3f<6FAm(d6~VL?$^CZ!jURpr|H#~;!_tjBqsv^R-OBT=3d~{inGdJ#h;3gMu;cOs zG-S4d{?B$a9^f}t&zU}Fls>~o@x|FvHjF(R?I)YaHnNf53fRk203VR`L$z&@7gZ=Iga>%ZD-ekO=r(*%SYe6@AXH)ZwEYcZg5_( zE)cUc2Eto70GEkx#6f@)^fQQw@SeV>gW(6fh#zQ5{QvSobtJq{pTD_gT9SKCPZA5# zl*EPn|E^6iMmAPcL{k!T(v$cTe2eNyG)iMn9zpGbcmP-MOO6S|1!^273_%^Ex!~0H2#*&gRz6vd3&Jdv8KkgTXu1UA+1Xa(HYy z8;cA4nRkxC|GzG9O!k`{rv*6|`+j909=T$O3$$4rBgSwoxbT;OYZH5T1>cKVE(pXD zm(5}b$8>ygg!5_xPAC_+Chc@x+O7S#4mUJzpqJ6p~_N#el>#@{Y(#~ZoHyIpMj{^g6GfHJkuiZ z0Xw56$ul+S73z|jUr}*66k+1TWgiXb>#+{;4=c;@OQzeBN<0G z#<>15Fh_8LoSwPV=Utd<;(C5tIe~wtUWE2L<{RW2&3jrxpI6cK3}dc=JyBCMu776# z<%QY$seP};!=BN3*f#z!J6FAj-DCIp&`|q7wc~94h}w4Bv7z=WdaUua$F=QvY{x7= zu*(Z^1iQ}eiy6+GK@34B0&#`-f_@~{Xsj}gPsB_`oxJeI2LqisLU7uarYv zi|Z73=v)0PzG!SgdvaggANR@qlA{(EmN!PtsksFF@80tV_ysgUIR&+9%~h*KmRqnU zW@Rwt3gb*{Ay>UBC`XJnO2+4>Cu^N)P~zZe?drFadpIMY{ht}|gBFS}_}zS^%LDNQ zU!MP6?FN^?H3RycIR&=GJNC1`m_$6_n2vq%ELT7e6Bpnt^YGjNhlvUBnKm|Wf#Up% zoQ>*EY|XOZNdZpbqrNmaD0qFqMsD6jxXqy?-Dtu_~ZD`9HMTSGpwP&RT9bK8tuVnV~!f< z(agk1VkL3K+XL&Q4+zw|%17)PJUqB>VBH|iDH8Eq9n(d5Ql3~%+}7vh9h?6T0hiO9 zmez67aLdM|iR8BWN`^a=Jc(9w&B^(`A2w#?#u`*4N6{7wq*E z{mrXf|5$hAn5{jc8#Q0d#y4l&*ueF;F4w6K*nB!tV+6Ke&cVHN59vubp#S#_9#Eg= zo_8%YUqT*29fNqYxd`!tytB0~c){~9tZl#z#ueYB{4mB4%~RkAaYUb?#J}|*s9wwzrzC)j(o|4$Q*26;w4nwvlLq-r#ye?Rs8ZDU(8ih0N8 zIu6Z;1|_Di{^Wf4hkY*Z{4V;}ly~@gbhC1gIn2HT9s<{ej|$g1!uPhZZF$mh^#P8< zckKhQmNvTvd?DtMHxaMk3^iQzD7kFBL6dqyuzNt4;=|$&n)vGiwo1*VSVRtWb+9r1 zptXxZZwb1VT5XfMCYryzu>7#PRCx;P6)p{|U9;YS z&o57L+hB`8u0re~KG-IZe-MA{7|@$QECP5zjzUb)JcWEV9STp#Q{W~#6s{2GTous7 zE(qk*o4=-^$+O5&;K1gupy(6wqM%rhu@RtebEd1q#K%)nG49))P zp{|v$E!Lsw^SxXTRHSj5+KtR%x8Eu^5AHwr&<75zlQ1{5h6B#gop{yoR=Dn|5lmH5uw=TxYq2 zof3!a8ypZE9I#`!L><5UE1$T}>Nq)h4z*HhWUaGf7p<|Q(bMW__OyGsZ?^PF!!!J> zuX{tlPgf&lO#&T@U8euYamaa#Mfm!13A7^hP;5IJFTWu#T`sw2{^w}*QFuXYLVJSk zX23B^1Dt_(#{4!;!cF2FafY>bp4qf6prhfmYEx>sv{jqm8yp^ptE{biLGZMovBv%K zToSRyZ32BW<}K8kjj_f@Zt;=1web3!uDCL=PQv}6MV33n8T3$N26ynS&J7x4;12WA z^UU~i@{00|{64t~ejz{6bJNUuI<)#l_Mp$?>9z~K_I%ETr?dsM3#auO zaG~o1?_n-jHjo0>-}M~&MD9i(fp}vYTZmC`us$7YQoFXT@dX?2ez5cIjo1mdv-fzO z&3FIAfURd@_r(-qMjB+-yw&!rZJWjsIH1p0;&-TR!UJmNtcCcept_RvZ`Nw!gXXK{ zA8-O4MJ|n|gcp7moHNr-_{jVpnh#%`Z_d_D{bFMtPtfz#@vyOM?WpJTmGK4}$3D|@ z&6}=4lhZA%IbzcRzX1DgTkFXG|->sKJXt4HN1uI{BB#Im)X~Sa6HGy z(_$7}&8E=w+zV~he%Isr8k;Oh4yAe)E?E}f)n^8K1+Q>@MW8HW*hm2b>w(;qHyZ6SN@$ub&4r(6Lip_7FV&2Plh)1lm!z+zPs#nn_ zJhP2nReq^fr5={vU7d;sP7|p{MI)(B#gC<*uya69sYb*GZaIVZ^rD5pP8NRee(WfS zg#f=*W9s|yH~w{XC)!gnYEV->$UeY!LqF4rihk$Sd7X3Gwzi|~w6EiBtUcO+H{>6R zsWygdHil|k_UZnP*Xwg`ZPl0V2M%)2+()@@yyHHPyhF#LVbL-B>{jbBX;#J$txCLM zOm&?#uJH_xa!u=K`8U<3`fNX3GuB42d&V)_$-c3vbRDtCs7HxO*tz=BY&Kt-e{OE@ zr>A*uIt@Nz57~e9vtFmD?{9k?R^Q)oM~>jX^W(JvPn0A0c+HP!v$nei*FtA9SNVS< zN3iSV27Q1BXmsW?;9`*5D5l@h_KI%P&tkm3w;v7Aam9T&-ucQm`!y53n&DQy`-PV1p;q}~&g`O)&7qiaKdhb44 zkS68%uj;pGWjKjeHNtm(vyE-#$!TkN%=4e*Yw*e10MFq&=U)?OleOt$EbF{q9cVu- z@96<;PYzwK2*H^$v_zMgR~?Yg4DTwE2au3*_53r_i{l{(pV{ahrbEJcF33aO~P+ zwwMhTgMqf;A@*3iNB$UXvpBBt!C1$tZ<>2xtDAd}gKn%a<{#9w;dF7M*g?O`MaWNs zm=g95#1!(??vdDnZ@}I+PvIU;^AxLw9Cn|=CvC( zeX8Bm-ZxhhEw0Hm^6~M9>t~P0I77aYtrlb8 z5&fqB#0?{t;1jkU^z#(Hw~hWECJhzZ8M)UtF@pH{)Z zmUIv3lXOaPrW}}Dn4B0N0sqi!#ql%&8UYY7N`i>qC~zMxI%8!@ANmUsc*=s$cTro=P)wj3h|KJtup7uNEWYt-Rw zEH934+)K}P70-4}hI`%rC&Dv})oL^^-sk_}AX=B$7!RrK@Z2ZxECO0APFfp`bwy&7 z0$1^;`Pjza=tr}kQ`=el$u3vFVXtdH`Pj9g?C1#JH&&>fpXy2cb>{#!e<59|JW$)u z|JU|0PH;`Gjeek>%Ju69{h?p<56;$C_*>tJtB0h0y+2&*qqp4bJ7f zVnglF@7g7PY3zc#%2m_2MgQpEauXh6Gx!KNhd%=_twatS&2IZWk$@{B~!Z{BO)E~3qG`HP?@{a$DP7J~3;~p`E+y_3OUE!dE13V!w zf{XBmd=E|nTmtxMgzx-rd;8)td3nctRY31^zI_7RhjXkclylKWc^B(O@XmvR`v+nh zYi(R7j>Wa|_|~9b8R#4Q^YcK>uo^J^=>N|9Qh?Lf1UO!fMqPz_q0U0Rg&1_?Aoo$u zhR#+FqLZrSpj-1lz?&H9t+y7j{*|9=L1$Nu3V zJ~)4z-7E*Ot!!+$2k-Fv`Tt#)KGu=3ZrW}K>^%pd@XJ=M`0OqJzA={=NSq};YEGP%ZGOpH_^X~r-1zVQ=60C;<2X||J~$UN zj-%Zz6wCSUe}Xac8@G4d!nt6)&R%0ROR)}rm%X7u%2Ux{_`*vAelafMBg=uQbD*)Q zsbIgH^VuSm3y9wDUM0`%U*`*V%nspzra6 z_Tmrsz;(DL*H^xG{f#$qp?=!fJ;ALw7T1o?x5vq&4}ibR*?paUXZsP3QBD^Z7Gr$q z9=8tKm~mgLL%Nslq3d#e?i+4(-}Hg|B@Wj=CDTl436jF#_F^`0BhyWKH z9mu^{qiGF4uCP`Uca(c@3U0CfMopQu@oEA_{70VM>$BfHhtan93K!uq>s@f09F%j? z?&PJ!P+})_c4`UKr9CH*%d#fuv4I$A`@l8h-7NxjYW}RF6fdglv(5+)(J(z5Q@?&W z!1MSX?|(SJMa>7{qJ09s11@q;-Cs31TLt1_eu;b#KZaIoE|@doY+OVK5wBN^qqT^E zXmNZ{zG-tpWBgp@1gU_zA}V z+gh&aZ~e~p<($S9)vvUvIT5ytpWmDaJI{s{_3M4cK|bT8@__wufoo?AUAN=AMz8w7 zckPeH9zHkj@anVQy08Ay7p}pz;a2>CcX2U}VE^fb>NM0>s9jLYATK3fHs;D&XWo37 z99Qe!)FOC4zuIR0Iluj_+mHFR=FjBST5HHr_qQRqVOEEJU26B>FM;pX>#56=mveq~ zjM}Q*a)GYJHMurAC9V`JyT`@={?RAKjeczngSgCn8J`y~mZBNs8*!8IOH;-%e5}?k zoDf|MztGU=QFN#=XSFV|5Zy@}WUjE=V~&H(t-T%n{?^K|xoWn|KXt$$2H+bOKKp*_ z8Eh+7u&=fA<}DqarsX+b^ez6R`UvM1d#HJ1`?W{R)Hum|ey87UXM6J>=po>}?``AH zG3<|9@C+Wpdm~@+Ut2HN{E*M>Cp>3Ad?lXp=REpItpT4{T;v*EyL=E%8tcxj-H=mQ z9>@>jC-vmC9vro2@S@ps>tB{vKmGPRD+W)|?Py4zWu-1bj-&ry$=Uh;tu6>~ z6J07M_`hNz@xgt9Z3DFl;-Uu!VhA>zPDk@nCx)-YN^HD(Yq}V_F9z8=z+K`JIvwsR zZ^`$_19@JTwV~oBoC4naPCSKUY=h(MgKKDZ)$7Dz;xXrNF6YFVYT49Ks5_w3Ef2I= z+aDIt9JdM7N|%RjnYs*hZhs1X8{ls1`o)p7&E|?47vZDN1^fW|sOP{O74Q@A5}(1^ zTJbA>qS^5)+_x75xXS&MKjMdoiPbNd6MPn9-ME%hh<2!n;nCH$rjM7*#=r5pVYchzN(mv7N|agZK2cPM7GBJ`nPP3?OLbH z_H(9Nay>3mq&-en5sHtc@MQbV? z3wJp#l*h(;y4Jlr#=R!?~_VhezHCm^kHbdWU6E4nTJtU8{U z2ye;v8}sEZV_x1#OjNrphGCZ*Z#1@O%;7ili*2``d7)0gah-!cjw`i8+s%zJe?)UM zj>lY3wZ}1rYu-m3=h&UQ+HSu3&h_v`%|)@F{q>W+(Ra=b%{BQ=pU_fpxcQ8iT#xIc z@8QC7khrKEO`F5f_!lSB=WtQ$*ZI)6NNxcS$s5Ue)9K`n_6pcETAlpSYXUZpM#L8K z(>^cmiLj>)vMu@4@%-O>{q8ChucEoPysRhsL%L#BRZBA}YTq!P6<6%9N{uCRj!=}^W zBlQb7Nt}e2)MmR6_=$Gvz8oIlC$a5;0Utt5gBldRM(gQlwPI#|89mRqsV^p;5@++r z#N6{9)nbvCI@fLf9UuZRSGCU)`W7k?|fiuJ@v@SYO^+vH+b6V;N#3|5P zHou8qif(HgS`&VuebK1cW%YsO7`9s+BMuRF)b5vO#1t?$xlQmH5{mrSpYOWe_&UE@ zXnwtrI0k&T&^EsJdwC4U;32V7pGlxD5YLEb@XT1B?Yd}keeQr5Zdp*Bt-K@NtHy?R z^f!*iIjyB|Kg6Q^0GcH|Q%ueeaZlajat=R(Um@4eW*b|4o95x!*oA7w*kQIG2e703 zdi+3#5*PRrGgwE;N2|`nr#DB~RJm7RTbItz*v5B$vz-{nehaPXaSr~U+e+sWOVH58 zC9T72Ena!d{3@!w;XYj8I?OvU9o=JH;2P^omV0m%Zq-kC1^?jP!e`&(CwM^VXW%wlPxc%^e%pfODy*r=?AweWShdK;@t2dt9d&4_AwSa5O%r zxix>+IzT$AI8wifi^Pi~KUGI{U)&=cC66YbCN63&Xsq?)i%er=W6wAgJFu~AF}rNM zo*%HQZ0e}Rv6D1ywh~wPqh+z1Q2mPrRvpZ|7%~3V_kFehh#uEDoXdI5G2AHD;OpTh zcAP!eR`z}5q3UPmqF8OL4Ro!zq_yB)ea2I^!!dxn^oe8OX}kkI`yLz{Y}?nq@4mJ# z?BDti`7mv6U$Y_QB-gJ0_`Q4pJ~1B0>+~CO6g`i95drU-9zbpnxCF z2GWh_N9dMrY;U)2u0liQXmPMYzg& z#aP;a7jY!cS{BHUsZGbHYSp(6=#+9vjiJPi{^Z*5vYK~Vo|3+R4?4 zM{&LSJ29&IM(d97RpTfzl=z9R%75TT@F(yU{{mN8i}aLWMNmDj`6rxYjNl}`j5%c9 z7*FPkd19jT6?>@PdKEEsu* zy=J%BZ#<*+n|_5`*nF`DF2E=FLaZbv5ub=v#4u+E;uhZH#lbopW%YQHWvq%1eXQ&|4s0_;P zQsTJgfEqWs4%dUz!281Ik>7Bve37_{E{U6Il=xaaDXtV>>f6RvG(Gpi{c+F4Q(`H3 zBllX|JAHm9Uxn^#{Lp2MDdURwFUn z6|R@o=C@w6>-N5JlfL0s>L*<8)n{|ozKxl18@|JBUO^kYvH{wzo?5ay*YFp3FK6X@@E)h( zLA;2g)}_|3f}kg(!^KzxAk}OO=2qir0>L*X9nsQagx{*$J6;5Pl=_} z;M4VRk$d#2fG=T9;Mcc4P@Ka~s_COg$%PcH3G|s=Wnbx6cu1_n{)#n5JMFzVXUyHu z#%OnPKe&kf$46d^>uA&zTr#n*ktkNP;{tYeKv+_7f7`D`0JHP!v(QVMps>!G~*zIL767jxk* zW1T$)@1eOGzkAGrjqW12T?zt_TN=PYmuj?fiknz^3u9tw-{#1@*`LadD8G2YpFx7~h?Z zR4c}IvZ3QM;%IZ~y{pTK8J5f-ht$vP^bUa>%`O4Gj20&s)HO$RHu@V*qO&zN63>Wp z@WiU1yrWjma}m_z;hOgZxa#o08dvq>UhSin&VFMpgql9}4}MqoSN>_-Bz9^JOMBWT z{3MpbrMOm3%(ZP5(D-nW{!jx?=aXy0M|8gapJ@8f|4WT~)PakM=y%O4c|IQ*s?DuVo=EPfpv|yDbQ(IHToE6cK8L&HyvI7}g=(bBQ}XG! zNbI9NPyB^cLkQuDyJsp%3gd1dcj9sEOZSn#eu%)~CzjnqG=Tb7@vE%D*SQEJ9%PM+I< z$HZ0iJp85pSv(-;FTPR((Dk(JGks5tC3a|SIo(SR%DNtWNAshLf%m>wTQTOT#AWI# z@PHZ%ynr77KiF1`LQA8m5w85<39+45yy1MdwZ=*;C?7;$>vhm4#WeD2^h>pfI9rV4 zJM{~=MVzU=UTmW;#iII{)!d3&z2+aezvoJb5DG5+jKiz z=UmS1xVX`|*g-zJ^_ljyjbrtAj!9o+BW-U#?UM(S6Qe`=-Z7lhd2Fi>`0Q+^>u*kt zU2ScZ{`0+l(N}mK*Xw8d+Rv-LcW>BLnu~3BA#5uKV8hr|{&Dji#*6RSPv>G^Yk%3h+BmjtlUr;uXQPZcHm+mL@#w?s7<*;`c8(9;x`O^D&bN*I`0KV&ufYC+d^(?f^xu7UeA>0|ZDZan4Sdg* znkSBFUdeUv_4szypjY!28}wXkE_-aw`QEYUR1Xd0P2^CVx3G=AX^zP`#TWBR*EVz4 z=~({czi8|23)kH1wjcQ1ZM0kec=d;8_?PZu^)~z95%H3XB*JOX^)pq@F%;>xJz3VDEzwBop*R9{R*ErCIjz#y%eU>kE&m7PGj%7RJ z%2;gek$d4d#)IFCBR&wj!LKv6j6vIV%sQt(pY@mVx-O`n##T9ZZG!RQxco!Ev0vVc zx2kbCmp}XTo*UbISGLl2Z4ckF+6mjJ9rkfQU8n1J9^0^&&cg@9wZ^|W=34cU?;P8< z+N7PX)t~EfPWF`_tsNsrYn!&{FXz-=$8$dYp>3U??H_R@AD-Rkqw)vX2(Qksowfh= zb^cz9HUOTe{)1n%-@fjZc7XTx)jr1r{jMEtpYQAkwrhL!jj>`5yC?3yYjjMu%xB}w zIMSE;(LTnYvFRQfb3S)Y8RKmB(!jl9E8HuuzBg~#VDHTn@ZLOR%h>~S${goc^S^wD z8)yJt`P%FQPH{Z?w7JV)qZQDv=|bv&_};Xl?q^@``FH$0$HYIa0i%Cg3rK(9yZijK zfF|J;PuZ5u;4|94{=U!Jt8MHDU(!C>VgKcU@7W^uf{n7hV>zaFv0dzm`Oog?JN;~1 z{i6N$am|k9vsZoX9vEN7zH!Y?xesiQ`|CcsSI%!geeQF|vfrJTZF66Z8yX0IO5A2l zxbNnlV{0G##5Op$dF1@Yo$<@>q4o2(>~D@5v(9N>Hj9lh&YYVqVV@ksc;}<74fs{& zrg^EpiodBY%9z|Y&sU}GS+iuE@a39Y=7-Wn`BVHfF}8WFZ(P57r!Sn>9CYs5Xmdyo zv)Z$L&0qa&ANR#R=9%`ILwpeT0)M!_j?1r98%)EIbKq0DcJZLNgoddf%uD_2Ub$!b zlkLz3v4FX2F51pD^4;}|b6L}RY#=x1e)ygLt^dqt{mtI#KgZH0_rcsTo{e#1)f_U$ zj5Wt$fB3)JtS#=dIcl7CjNnsT*LjTt8zY+s#+bg+PtL2&?zwxVrbYbay?!;eI%f5s zz9?V%t#vY9@vMGyKJ&pGG|yd!wrGR4)$X`wjn`{0%(Y&xG3~gntMki!V4v`sfFYkdxaZtbNSW&+Ql}wKG&nY zY=isDHrvPeV84wO-)p;efn&QD{14-m?KjT&2=-?;oYVg1vfpf@4ccV<+s^$nW*y6z z@SXFz$J$^!bJKWsJbYsuI*$F^7ss_9?qYvzXB*$y*ZB0=9CN(3$2Ebr*-o3yKijh- z+Uc{l;8VZ5R`)*tw{`oC6Qq-nHmo=h8mE+uu2AEA)py+t`*J zVgK~Ed84iFx4+J5+qv(?n)kIQI9%W9gWi9i-5>T;o3+b*&>o-XPCh%oKF}UxP#avI zYjyqYH+^L6+Rpy^N&9@S9qf}i*WX>QapwLRzqZ%D_NQaGc7NL9IE|CE-9E_jj@7-VHSKruHdt8s>Xot4d4jN1Prt`ve+Q%5uUhS}tKCusY z@0#2v_rv-0wf5*w+qm}jwcmYTd!vo^w@vr8zdy&;9;glRy<_=o9@xfr#)h$@z52~I z&aF?hAM~@|Y@<)iLHCMX(wDZ?F0Zz8AKb@sv3sNKj??41Z@%;Doca!IV}I?^7d~sV zesZn$@w;Q%*YDoDm)c`L-@88WyYI|3zjqF}pN{AEw$10-O4s2U>}wzW;CTAn9I%~s zw~yKfzPFuM-)WnEE0@+zIG62QSC6R={qBBfuW_Vp&gWj*);;sR-`N6vqz%repTM=b zhxT<1^)Xz3$Gr1so3Z0s+Xv(QcP`i7esCSmXFvU>z0T{k=W=}a$8maXezzanV6J=B zuRV@_ay)I)Pp-|pGuE`-zWyB3vAp{1dws#~^xn6BT|YRk^JtT6wXgovHorTD{lR!O zcJ!xxI;LE=KliNH>O1GQjXrYy;QYPc?d#56zqhZQ!?7GwJN((s{*I@;+UC_h-Cq0s zIhI%Znm4|;uQut6w%IwrpX0j5w%0MC?RGA|wY|>Uz8in@yYFqQjcup%`>ns*r(?&f z-?c+qdmTN#V|nkK_HRFU)jsX4?Qk#JkKX&Uk9HVW`n~6~uQ33&>oqy9^R*wftH&9C zv!8SOvu(H6e*M}p=6ip&1Aeow>(U<3AHH+H%$wf#?%Qj$O|Qp&bSc!#-`F z@7w3yr+unzz4qQqzxnHVI}d&5__hbf1>5(1U;XZ{$MEVqkVFTttQlH#E4o=7x(hHgQvu3>DFElQ9jNRE9#C;wps-4Wf)u z(bZKwp&~_dRL^<(f8OVu&$a&N-~I2;>vi_tYp?bFe%E(eYp=b}Z<9wKYCEzj$4`4a z$>n%#i!b}x@Y5dX1Wmlqf{{O&(XyedzPeo)n(SdR6dE4lV zcGa1EX|)`0H0)<3o%JcQlBsfNM_xKa*LmqxXRbT$xw^{KLy!IGot*d^r;XR<0qWw9 zC!6pob#3L%kD+!{zjC6fp6i`|;Gw)eg%(*Vqhp&owT(K}pY~|lYM=T``IT1nq3yUi zM~BR0Q(u|(w(7x{Om63NuK&$_^hUnQj#lZYTfO4zT6Kz+cJ(WD)u}x6kFWaLkq6(7 zA7o^|=4$7S8MO4d^RCL@`H()=mUQM=8G7_kzaWotwDd{2JsE4C+87NoxP}J#)wj7e z`W87{*Dv(R>pERkU+B74o#}&i__ls^rcB-HQ$B0n!&^Doplz$G`iycLdDSOJ^KtWI z>DbSwqYgAS)B*ic-toq(5-4V^KDmC-T% zns?UU=o2mcZRoHk8;Aw=Ys{cs|3C}W)i-psM+7#>Y zb%Phat%HUQPy5P(e%mNlPan`z$6kN%YrNvA9If_4UG4C6T^n`ug^uM2{(zUd)zK)I z`fWezj9lI719+>$R$WI_hfkSXn>*lD`O9Zk^Z)RZ+KbH971_wBZf&l8W$Sf)W24bl z)x&-DpB~Vxp3rn&8tCamWx|h4?9I;{SEuEvjSXL$f@Lcyw(< zpUDN}#1kCjtK3Eo$LP7QxBjEAWM~`p^_zdy$CSBVJ>fU2z1hFjO?j)QKAwC+-@q|m zP+4m$ZSl2UWgT5EQYLR&OVD3@$zpRn%Hy0J)bBkvd#GE##H+R;ADOIujcZ`!D(&_` zJ-SerTx2i~TQBe&#WP>)0r;{B9%QM`oV)LJjm+rjC%)`}C)x02U)QYnf&b&97rJPi zcb(NzH_1Y_<^sA?*Ir-LM++RI(YQgMPRWO_zEltT z0QgiVj+LR`96&b5zA$iv3ISxhiuyD1Gug|`Se5o_!mB8u%Xw!w2ZEs z%k&3b{lj11@PqcX*U`hj{c3-zNAlo>CwgRWee#hNUH%4LFQ`2Hy1rN)(=-0g?QQs= zN00c~(AGbCp$DLAb?Z0Uk_k<4+%kQy53)0mmH&0FtWRj+V_$ud3;*h+HdYUR<>azg zrY@jCetWXuKdbp!-8)Cyo-N2hM~w|OBm-LXj21fDkY79fbiI7&N}pS1U*6Sm^`gJz zF(+t#&|V+#Q;%F^k}KFwU2VwVdijyFwA9fTGAb`Wa;ZAQ!o{ zXXTPr$W%L{S3koal+~V%ZMIbpEp5qzPIb`!jMvm5gF5y- zuRgRX(4}*Hn@8|tQ+3#aZuu|erY@ZuU(o5iqwSUB;ThxjKX;zB zL!Z@=r%-16BsV(bsy^E{eCU_%_zGFnsejNjH23H?ImxR{n;^9k~gKBP?F%(e92`#*jEg$*kMI|Dn5N&Pa`q3`*+_cy#R0NywA zj({@nC3x4!duQHT@(!h#B$K{~cRJBG{!Ko)MeMO9*<4pwye|$)8z21q|Azky@V{H$ zN%Q{&_RhUC=Y2DEymRF}3h=)L+LE#Friep)=oK%qiXV9$dw$cG0f$WxC@_Y$`T@uNphrXFjLA{=jFH@ohR$pI@*MA7(%HX-q0Nx97*+ zZxKh0vA!4QeYL*l>HRC{JB;4>1n~fU2UPs@ohP)#4*t;hwv;(;S>p~J{Jh(S5B|#d zotPAN1wP)nL1&CZI?{i7qQmBCegyo8j_5~RY1BVgui}e;$HnGs<@N8%M{cy5SJcst?}H6pTY1d(ZkE2>(lL&&@m2mN@6_k` ztvn!Byz9d^?Y*}wN4O4rUp+8>8JE!Lm?5t@$A|xlcRnp9)b}1Kx{X8cn4S~ZiwSY& zI`sW%Wz7-ZH9|++h!eE&X`kc?eFc5QA8+m2mdxt6F4;%9t0ULNw){uW&4Y3?&?~#q zxxKpDK-VtVWVD~SENV~w=U91jxv`YZwFN!}d;`CZ>G&#ZZs4!tS}chxaIV}K;#}-_ zcT#MLCGoC4Kj!OV2(*(=-j7{6lWlg#FQ9wUY|GSIJOrn#twVq2%3#KyzvqP zqu=tsj{ElH0zOLCjyKwM%n?^(jED(+(MDVTP6zbCuj!5c zH=%)6b&dznE1#xgJjq)9x~46j<`lq|;sjlBKqi|ye3e|y&GKexv#WFU(Z!>=iayzu z4#l!-&MRN@Y4z=#4E0m7!Ot4gbitSTrak{GTqB?Qor5|@t910G{T6%l-1zDkDYpFo zl=tGj-wghb+5bX#PaC`|u1?3({vSl+wqvMxax72Si)V1&_npxY_xO6J8ejjvfvI4CKXCUF;Qu!KzcKW8eEcp?{}0*q*75%z z>N)rS&iyTuUT?YoPjk&VI{wE6Mw;lLSDIugO?x!_pNZe6(4USmjSo8M+>egUn>vp1 zRbx=mTv-3*XZ5x2I}X+F#Egx9iC^Q0oGEXEoap}1_#sb80Y2# z@>u5v+Q_|thC0o2<`8llIF{?$*14P~-^puorDO9axfdKOb8W1nojDMCym1$QbK+AdbA6;=+O@;~&sOwM{^gC=$d4Yzc`V=3cVn$_w&N|m z6w1g*w%VpXr>}UCy}E5b@KuJk@tRNbZR1z{8-Ldd=e?#(+tMWiS(J%~+N}GwH3#Y3 zR!rH%mwXI1F()@04~#9H*EJ^P>t1u+v9gY*_RZzZ=ks&DI>pGVJhbcek(M%>F{|T_ zF~!`j`lG|<33_5vx~X5WB|RyptLjT#IYryf7tuAI8prS(d3IijpLXb=i`NJ?s@?1R z?4>Td;8#8!w>x%sY_5)~XSO3FzN3GjSD!6kzJ{NEyM`b7Hu|h>=-&8mZXsuzZ9&f>Xsg5K*JPmG<$7h|XYm+JUwjjyog z=9)29edDfsAmgm~_v(#fd+)8rv7KznyWUY=<%bUXU4I|v zA)Ski^OkXLV*aR|v6EiFG0@dm?wT@f%yZ~hY_gx&WzWV1`_3OP>t|#9t1q7T0pH>S zXz~MN@i-5}$0i2Q7Z?0iPHLRVi~MXpl#4h1$#40&O>B?z&dv!sAF$U?Aj|x@?ihpQ zK!1z}@gY9RNoIPgpQASdFaBEl)OO0$qZc|tgWjqGGN7+tbc`Q3A9YP{^rz2cpx==n zn~I~x7e6w~RZo5TXD9rj8(Bb3px5dIe{v~vT^qdhVa(rXH;3b)PI=+MPPO?+zd4U> z<(fk7tG&k@D?d7y_grrsIf>m0_EXlnd}+*|#@WL7Iv;YZ>&D+jbMrv^&OMCP=z-ig z&MS=FoimwxneUlb-51Qcig6C6o%xpe2wG^C|0q-EFX|M^)HhctI-hWDypEQB>VNwr z&$9y?j4@b!(>*)TF@3Y4xT(+b-$Gf(7xw5lIks!v`TP1_$4>s&93zICU--7;5z30j zXY+z{bFm0YGS|p~Z+SMi+beSoEqQ?K%|p&-x&7n*R^HH0{T^dhY|}$^AaAN`Pbc)}+NkeQ z$FrGswyrv*FLyK#$OY^vH#DEjT445XbBSxsC9a@{e*0BjDQjQIuRjG^Q2M3qI)5rE zhrZYLbXk4(o^;f2?KH|$nb7Fis-OBSzcz>C@49;El!pC$y7O&cTPtT9bjVMq;zcZV zp5NFU+tZ{-q`?;=OvQ;l+Y(JD~k3QOTI>uV#t8p{>vg6T*wY60?G}&dGGvFca#+VkX_=p+4 zGx{EX;@|R2{oHkJ@UD;Xf%3yso-c3wN3`6W!hDC!=!ipo=ezdK+b4dkZs!W+#WrkD zr}Z=AR>$v-Vf80Zq%ov5JlPS;zX3pE5pd-0axNceTNvZtzA| z|L)7)au+?+zTKZ-&%%jzjee;TX=&^0b4YmaK9mfJ?#!O*Jjy@$ zP?sJUCVBOZa ze!5;h)%iFk)CcL4-T1&5^J2Ju#V`2^pC0pNv`UF=I@lrN5&;HBOY14;=Sgo5nI9MGswer2q2cU&=Zz z$~)>p?bdU$>j#~UeqLFsXLag-`pFMVw>G7h>ZUO?-!6FQ*O&C9$9}BCPwS)l zRo~HfwCRjp3b{fVpQ;S@&a2~I*B+hrNuL`Z_SJ#D(b?Fq`Lsv+WFU{c2#)bAKRQKQ z-^6BPUi=o#1&x1on;+-PMV9J;kJp#Syj(lzhdx7nliul7pR4cYP5d2$YolKDgYN8Y z?1m2c!0`xW>|Gt#2gZJ(QJ*8@7$2iv>2$uW$GkDx6n*m3^Oz%hUm;g2cU?X8$*NCm zPY$%3r||yY^sL@}=~ln#k3QM9bItq8)9TWLK9o1#)n7K9pQq%i-W!mMFOY-m=;#l= zW8aMt^T!Ew=J%by@GrdaHRtR6(mjS_d9w36yw$HB$Y7&S{ikDnrhDU`@pv2u$)Nw* z=zD#;{cdh12fpT+m8WYqu920@+M`34bWuCVDYITZJ3qjeY=!e)CwKKCKh;0=&9O3m zUcW|Ldv>OCww`Zq<*uuze&dRaw(^iS$q$X!k!SmZ4w`J*xo77box^tiVg6&zNAGN* z&t$0H@nQGoSGw=qtA3`8PqK5zlKFDc8QqP#cC4Jf)YmUI5>t4vDSPs*u1D6-dS9#_ z{`EDovIUxG)2qJer+Kz{E;y!>_MfiQB@_Q7kG{w`Vx3I=WKa)(ZR_K; zJ)7e}7W(czfMYfi1M1K}Jyizu%D;0KeFpLb4j*85 zay7pBTk~h_-o0|y^`SWcU%J7UAK=Nx`0;CcurL4mLa)0vUe}gh$yoc)w{7%U$2R4) z+ql=i`X!mr{QvnQy%g%*7hJDwXxqrZf5E1V9OYj>K~F#VM)N`MPsY47<|g&{6CKhQ z{`ykgkOA<+n;qCv?66PaoR1Xh0Nah=3uxH%!{%7u>L+Z^ALK+ia9n4Ro9E~1mf7=j z;}<{X`)G)T`c(bOeuRGL1DVM|99lg#JJ+s$-L`BF zwOextKjK^RhIv(=oi1813v;f{x8xJ`z?=v8zV`gN^N-QTYB%ygxB8xaW3TdQU-|eh zAFkg!7ax4jroZBwjp$cCFsG9r%<=mC&-&)#z`kRke4~xFXwV+0i;4?hwI_1o^1xu#C}>mQx-1$su4 zj@fdwFT3&&V}^R_<7s0{ybJxRjw%bij`K=-(GCsvLz6A_2M=(p%r(%@`55z!zL24D zPO8vzi)a$s(hJ3iTlsg(fWZ{G4W>dQD zm{OgIq3W$;#hAlguMWE(W36_rFVYpB;+p*Y&E7d3*|Z@CzQznWj6aR<%=iD1`aZn( z=)6a-4qwwoZquI4$)tZaJnZNDmvedm$K>FDBkG4^e&aWz@6`vZLwbSHzsNurVoLwn zn5{dW)aGJR9d*fpW_8l%WaGPq`Y-16-<(jtwX?Bvbx-enu<>0!asj_^PAHSpi zs?q~rbgxZ6p#StIPp~~1=$(J@JwCH2@D7M~E%+4Pu2pWlZET}GAC~7jZxb`_O)pkK_oxrmk2M2YApa`Mf9M|0(?csQ)kR|ErQ4UwWcH?fIlW zi$TDL&6SV#B%gNT$9pHe{a*^f=};xFXX zNBU%w#u&Tu!TNji=IAT@h~DX&T;4q#|L=p`eRqc~$csNY$c$%mfqcjwpfCK3oyi93 z>jN2UKmJ$yv7H#>cl>YsUn)M?ShVMtivweV_RZP#33Ta!Y_{fF`e=;u`Nl9EDd*qp zi>5aEDVCtQRQ{y<#wXwB)A+P+?N@UHzOxx;HV5HRIUMT;blzod#n+U{F>J@4=m9^W zLwaTZ%4AK?dr$n^|3mxVA-?VXD0Xy>--$BPp|NFUV;Qv7VKdAo=)jxj9XYmwsd{YeS z7rN}lx5=lj{`VbQ$7uLpQo2Ed9AwmgHlcrZkoWjs|6hRLu?5}fSH~dj#ijcVJV7oa z16{CFb)xP#W;JgsQwR7JIMy$8pksz>`lzoqzRqXaMXd2j^4X)W9zVvLzmUhqmi$RR z(R;tcR`T&#?+yAtJ93g2Z*c6om{DK6*S@YdpOd>b zW-B@ZvXIYMDfY!tb1*+6v+Ha?Ryx(U&KvR5Pa7J1a-1XbN&m~@yV?Fv)Nj8yH@BAC z@Y9A5gB-_}=+!6KRnBa_W>Y>ze|*PzeM#TQ#8&!m?_59mfj-H(^(Qirw{s^tzz^;E zw?6A5`=G~n*xhmG&W$IwV~5Ts=vZ86&&K3HPdOc`%YT5a*-_k&_GVxFT-PVZ_%xT0 zwRs=y>V^LB>zt>uk-7HM2fWFl9y*OJzKb3>ZhYzo8hnqO^0$1~|Nk^+Q6_ikk8xXn z8w=IHvB9RCoIrMuH9H2E#v(wDNvN^@52qb+^d=t^FoKe<2~ebRSj>?haMcJ}xx zQ;!{$0si*Er)9b^=19FhbcQ^f>gtoqYh~`$os{-H2^bJ4rbU9bQ$O_~k zpN;M4jy~yKU3M2I&DZ$#TcG`xd&j2!<^=ip{(a>JbjC3RA2C4p&1qtRY?V{FSfE=u zwSG)izGc2^-9tWao;L3@|B^e+b>&jN#@}a0z}a;lb|-^56aDC$c@jOdIeEnb8Q2mH zwl%*)*Z(N`-${GCI{!8IaZWB{ivM9;9FW)ijXY#khL+g2=^s8~T#SHz>N7owXK<_^ z^?mwLk8H*a|EuXY1pMxR-!pLRcbxnmCwbi)80R{s(2F<_XXq~qn#b`Lb7I+esNZP# zoi^Wr_bpA|jQ4v7__|IH`X%1=LF|dG+MnMk)8~%i_HqI~=8*Os@5rVtUwc9D+F++S zyl%nk{^vJT{EkY$yD852Om&T(es=tGO+MxKKMEvl9#`mU~J8u)bc@E{Lp6}r9 z5WGEjU+|&esNlrll;B&zH-pmx$KMIg418PlpgDoTQ z!NL0jb;$AW!3Tmb2Hy_8790_%`<~!~!Lh;Tf=>m92FgAX937BToBs?>4!#_GDL5(k zWN=7;-@d{A!G8qr4)B#<=$ic71#b)J;Z1?O<2$5uLzW|hV}g$cdj`I#`qscTdeASl z(a@e=9KSQDKG=c%mSA;XpdJ1KMg>cF>ceJ7S|bV%3g>yN(R zBOk~^;!qyvE8 z7j5~IylDAuGWz}PlvD0KW#)UjC*6C}%x^_1R~HYm<3}&x|Jgg=rweV#=r<_XUV3@(z2J=C=fU;C-+~*1%YqAnp9EJ1%KjAmGWbUDnc(Q)$l$QxgTelRy=!L& zX#6%n8;?_h?*_jP(87-lc>FB5I=CmOE|$G_7OUO6+Px1C$i5)Z_PRiS9}_$$SU*^r z?We(i26&O-2f??3PX&hrWW!HC@zI8iKMcr4N6!o1Jcljs-SXbef|mum2Zsl~k9|b& z-eBjzH?emO_6j}}d^+&kK77wu%NIWt92#hUT)?N^6TB)Q|FeRZ1lt6!49LKT;9r9s zgZ~K7K~HRoMX|=Woty8N$H;YZv)m@9nfu9~#t7fD-ZhZlMn^Tf>)zYEY^b#m3o=LT;H-WTi{ z>>9i?I5M~(xH*`Znwa|gdv}|O7vF3bSI-EZ9&8vqE)ak6fPC@!;H2R6;N0Ll z!54y$1@8{t6}&f)a}EwZ9vl^XGB`0fIgktN(K%tjaZ8U|dQgDIYl9aA&kWWN)(Tb& z76i9W-#UF`a6|Ct;19un1;3cX<UmyG}u-zSKqwM#=mBFur9|dOwUkW}M95shy zrjMC9SLXQW;Mm}V;N;-+;LJc<$MC%X|LX$$7A{@5bShXfy=3~XAgO(NrKKw^T_adG zcuerPV1wX!!Rvx|1p5W=59Dn)FrX89JT&-Fpbj`bBA~Zpg2U%<{Lo*BrU^0~3b zSoMhD1%VuAOg=l13q6y*Dlo@4XEHA`f9pN1`&{?Dau;9ZW9Ep?&410o3v)JjL!hqQ zCO4S_{~-8fa8+&?Ngb8x&v@cQ5l!H&V(g8hQSf)fJq zcwTTx@Q2{0;LbpvlcTN=eji*B{8u1H{4DrUAm5!IXfMyn72ga_4Za?HCHO*cd~jTF zbZ~ee_Kyla8GJU7C)N90AUBZf}ws zq3*}c2hAam3HayhgBJ%I1?CgxVK*(kY3Y9lmj>oB=0N5N7X|Xr*Mctx_U1!h59Akd zvTN|F;Kjl71G(jef%BIIFArWC{A;jTuyOF*z_nKg+XiBE%Ru{Q2O9=Y3jQfrGk9dM zb|7CpJ$P2|oB%ESUJ<-0cxSLrAinMQ2zCj!3$_ln33dqN!4C)KbjJqbTVMVokY7I$ zd^IqCGp{-|kY_&->=|hPo?zeLeSz~g2ipfb2D=5y(UiYG639IV1nA2-=4kTtR|34~ z^qT>FpBJ1Tn5SJBnCJZ}Fjo9EFwZqlTqk%+z?K^ao6Nx+-JI%0!REoMgMSb93_ci` zm+=Sld~>@)gM$Nm^Zw(5uLbUr+*g3#kTe(LOJ@b{v*bDFy@zl=q0IAE>pq@&nWq^i z!MSzo`m8lX>uv58+>%-$Sryk>GCVI( zP6o1i>#xM$*?iCcp|2iF92XD)krK$r6Ww}Y<+Ck`-Q?i}y^6Yrmy z-|arJ`^>1sM*kkL*9(IUf<=K`y<%XlCtu$h$kYE7TohaoSZlJb^Xb6+&OC1Kz`TC9 zz_@CS=6Qi}nBVo8$hfv>P14$;XK%)DjLp7 zHlGx%7d$#xGk8eg-gd>nT^}dDYw2A}Zw|~Wt_l7S{4Tg6FyD{^+~3JZX9e!_%=hGw zPY32Ia@1$$AitQq$Z?$`%HMLCeDk@Wa~t`{y)ASuB)9Dyn0ta;`u0G+ms8&o$aAn$ z@R~rLH|P47z!)IME0d!)4dix^yL&8$%ln%L^7HEh$I6TuAm^iRT)}Ir zK>mGeuzNs0V}iUaKg+@NB;TGEn8(Sn#yDf1d-LA}j_r+qivwe!-0MDCefV+ki@^E- zyU5eZe-vo{$KdwB5q`iY_y<3EV&MM%$-%ROmj~MgZw=lS>=Nt{ z@GoQewt@TQodVa~x4$cp>-PwHFXF!Bu)uwa`xW;s-cyjP`wpct`P!PlwK!{j){m`O zTNAuJ=r!wo^0#$Jab&Jyty{Ht=EcO_vPkE z;uowjyKem!+|yatbw6ibrOaAuVZLQ8*SU7uyO)y(@N-WmKcM5f_1FXFpj>-= z$%K}@wPI@r*0-${7>nTR!CAq%!P&w0g6{|Bhvt>$jz*_H2fqn^5nLKv8OUQ-2hO#z zevY;^b26a+`M}z>b?>hRr_4b;b<`yj{yz|Gj1nQWdqU&A(Z}$tXyGQ;;aC$)I z3j^)p$8&IO4aDBsh<5gPToRz=T)*9q;7N|31ah;zetkE<LZ%^_3PmB059#7lN~+u@wc}QL`T}&tHajrz4#p; ztS_4LTW7G2Z{D(P!2iwhox67v6XrM8o5YCuj(LtbKNz3f=e;h_77h0V>X;u{%MxFn z*ErrR=<_?VxF|5c^6b@`I#{!}W?{Q0Png88y|sRO_oto}nX_5v{%zpi;lBgt)}(Kk zdmqIbb>DBYMlFBH8}{CJvNk8bfHAYr3a#CB95tUbhxGi+zTaEX)_t?<-wj*?>x-Vd z8FQ@(n!8%BJ|R%{#lYUY^%H@0>kkChuJ;SxAAC6YNT7V*xiMRgloy?=Yya-xoxwW- z`P22%Fn9iF@bSRh8RS%S4-Bk_y7rmCoLH_kkGAISnWTQ0FDv^(KyLjqx2A)013IzZ ze{!HbpoOM+`^N%vb@O(5I5r@MbL*_m>EW}1y>s>G4NtVc5u6&3i#*?(dv1caI_d-7 z?7_ZlZO%^?djE2uo%>66W+yVU10MRq=6pzd$DXh72jdNT)^t6?=5L;P@IQX$nX59- zY}K__SDU`$i6-Av*RvG&7uw)=W$=f8Up?}f1M>yXm&jK+=!yQ(Z!Gf- z{q1uo&!T#a2A{)EY={B$wAV(N82Nl4X5<9rWJO1N_nYLY?$G9Y>e`c+ujo77ILAjD zGS-h=XAganf5`2cd{_I?sW$Av7wLrGsl(3JgzpH2 z8~gA@IoWehb+mE+K_B?BvG&$}pYc4m&r#)m3kUY@Z;T7xK|!nU z0~s^4QQqgd?vLD~8(YlpJcoAg)OAeHRE$H$G|yg)TV2oUGiGC&+cB6+dS9zOvm_%wz1S0`>w#=^KmEf3c>R4-Y4`|M6 z-OO`z>uttZeX#dlgfZG)TYK*-8EcKX-dDoYbA#6h#`3;rYz|BxWW`fm>l^rZ7mvQs z^z0oE^w3s@H(K=T`8mCl6%V|~#}4ezPM)uO$Jsj~`rw)}_H&&**_Dm0A?maDy1Xxk zMt#xVx#t_MTdVebjgRqFer@AdU5~`WeS+(JkH4Bv@L%r=jNo(H%e(6GTk{Zg+-LAp zGAO$%5C`2xzvbJ2o^l%*>dR>1Z=DHE{!LDDw_F|X*tvi4ZUTDf;7?z!S!4BXLS^&L z0y-@dM`R^~dnDJ<(SPx7Y?3$JGm!;vdJ^yQs9Yldid*@)V~4hKfxIGL$@_AS93V%@ z0md-b<)O}#O@@-b?TpcA~oJVQT#{>fE)uvcxyhwIC1#xCqD*7%Ruam@C> zuNDRJjhxDtns51(_i6Z}I?kKF>(}0$sV~ai{JMVXSR6G@T=!0nI2F@kMNE;;_`z4T zL680Q1^6ys5Kk@VYhqtJbon|y`i`Eyp`$O_IagPievpwY`Xe5}vHpA4)SiswQLlZX zKYd0QToX>%^b-%V7|oP_zU~(wlyeU)}0SUzM>d8QB(W=+$qur5n7^RiEy~ z4?U8H{`8w}@KmQZp?hPFz8e43>sag9b#(EdH`mnfd>SqKX9u#;P3Qgi;?L$bGA<5` z8^#GdjfG$wagAQVb>oox7I2?cp7cxa=oq(*QSL{KUHG#lo1tec?fs7X7UQ!q%v=D_ zFup1`wt{;g_Y>s6hpp%yUpmpwb>k(SxyDxFhJWBo=20$stKG$?xdA@zr$BwOsK-ax zhhNc+dgd1TiXWY`Kb^BDIrY+_W9Rqoc5gm`*(-@zO6MdIKKR{h$j&aO=kue5cx*?nT#yR&$?wxvFzwkv@nL5TO zW1O<`G=>_Fi{1+v?~I%F+87gEGuFAcGIn;2Ra@g5I99LYGCA4axIkaVKYF$C4}1Dl z-fMI$ro;sM>kGZnp+3uh^v!qaULAVJ%QZeO-sA=L#INH8zZ3^z91Zl)#RDC=x%rE2 zT?hHqxjOuY>~bAn=ksJ?e{if^PV9Ig$MegM7xJoDYOFX%lf2sNJ17%tVon|ceNdl2 z@>O!^k9;aG;M;MG4qT@ba%#UgkTcYEovzsdZ+yrsKe`8yljR?B;^CUSW35Q8lf&dh z{ODCa)=qA|i4!+0aLq4CtvZf1_zF%QblqePe(eFQ2M!yr5V4 zOdI)A8)FIidJSDP+|w9y$RkJMtv)Dk8~%jGIe~RubX}t}If$L)Z9KK3CwW!hI;Qgh zy3mgP!Lc&7WN&(+XY$~O4ju71zQe!pVoN^2U&Rmq2eH;1%iqR1nmo>z#iBZ7L#J~c z*E=U|4s9GdcMT7FF-0EnATAmk`idtx#hjQIUykXD9{CD8sMq+?&e+3lu89Hil#jS} zjt=?dMzT1TyX79a%wFDWKh%?Jhp)u2<(o*b}v4(Ms8Zs*T(tX$w) z>B$SA50z0HIo4P}Hhrak_Mj)%@seZ2ns}7oK{K^=LK{Nz)QuIXHUWCQ&1z~6QHpmRPi#`!9JC?}6T z(RcGAA7EQD;!D2jj10yhGNA)(Ne1H(*~nF$t5=`0=R16f|Kh1kALS^q3UUUYqlv3{j4J6Cq}>4`q*O+ET{9ep4d z-P4`x^u$N#*g1OY={I`%E|jw2)a^aq?*Z?!WzXo~}SXCF34n;iO~4ttQ_ zhCiLxxAXzXZ?9i$jt=ODjh*!YEq$T~wAe+im;3cWUA)jEKRxq5_2nLNl8*=A9c8YKxkFj|Om@Iq9d$t;=tqC_!5GlKtE-Lr=%`aZ{Je3e zo<5QT*j9aXl^ch|9R1LPzR+!R+^p7Te;biL^d0c7jP~sty^s^gmt*Kr-{?Th=oek-2j9{^^z~WZPzMeArX%IX24(p3Bfckh z$ZhH^%6Qv*I{H9U9ea9ny>lJy^>d76^5`cW@L9)vg{1^#{Dkh!)!Vs1E-ZOYHA@^~x9UagA-X zkr$w21^SK`1@0ZhD;wZNR{GHvZ91W6V-LDwvvEa6{D3a`w|k_n<)GV~&pz6bpTDDz zHlGn^`Y$KdSLuQ*WZ<{X*@YZzPgXXjcRtU?WS|@R(8t=B%%ydH+&0BP# zZ){zl!T$K_FCCGOT>3#K`XR5jc#zBZz)w1dBAfQ?#~0ZN4d55;vpV>ZTR+&BTu@u% zOb+6emlg_oF1NJ5>Jyajcj0I4=lGpX}rkC2^IK=PFt>_zHI^qlL!7hA<4xq8E zTwn13^ywXK{b4IK$f{51lHHyk>#OVXk3NWJxzC=QY{5T495~mmW0r9Xy~af2%AQ^M z9^WS?U9hECVMF>-M|*w?YzdC#B%lv<=t`gTk8Hq~^-Vt<;~^he&ythG8vRsH+R%-@ zuqz)YqdMBMn>u9XM?g2)kd?nEQ;)vc)b+-gJ)P@69rH;#l4H%a*tz2Zn%eS7eb7I0 z(WQR6R=v>|-H8Ky`2;@l5MKJP4O`k=LlcyfyK#>n-eRZilrF3KVS!b zNH%(=6ExH<@WcxrvXZBHNI4%D_r^GM*+c(9xtwFqclCpu;ui2nzx^Q-zocVvM9=8Z z1>2#sC=h3Ss(lz^5?$wfLBD$c%?_P=px0c6K00(sK3jE3~1t4J@YO4 zp(}lE9L*5UA2jhRUwTz$uZ+GLua5Z?+4uxq;mJ1m@N@pFZv|P%M7Lz-&*W*J8#j$p zyu=Y*8-tWLA83Q8KI3WM{_+)LCH=}Z=2`NBc`9GxS7aeC`;vuB)!*n>Y(dYiYXj=h z89maCJ}Bd#`rh{ZvHGKL0)^~2us2tCXDuG>0BjPZzvK9hq^ z=!cE*r5k+e%le5nx*e;GTl|}^xvuYY#b5c2yl4(5*E`1>FOU<&Eg5QO{nHOI%6|Ya zJkTdgeFZN%ufFl6C%Qw=o}bcJeT@z20bl)AAC2aGdPGNieGsGClm_1A!+wodw9wQi zu?gj&4q4D)Gxm^w0dG2XtudvaKo@L)HhYj89jHH7pLp2n58C5HH)JjkZS{@p`mQh4 zXZuDL&<2ghGdqZD`_9!C&l|V2tA6Q#FUcW%i*DGO4C?8xy>tC^PJX^r|K)Gh5qa5z zF7#VIGX5?K2qmRZJ zK6Fn1`hh1tWY@m>Xl^BQ`*B}$BJCRIXjf)^BCBi4@Wc!7rbionk-d3b46+3|?C~Dw zXZkwY2oJi!n=jBEoA3#FfStr9{%pxt(QY0wrqCrh8Y|=YR{M~b?DR;V=0ZTn%J8i} zt4l66MxV@l3m-Dj4ZVm5=X7YRuC&)5dmuZ1aI9^|0`$a+a~qzv9`l3RTU*!phknu_ zT|oOw-uim?Xrrk;pVIe^nZ`@`j}EG5x~B^=lNBBG@gXn%HrK#L2l&tfeb=UA-}Qrj z)FHnZ1GK9HZK3q=DqZcsxqgyGd-2HsI`?%=dvcMZayGx}8^2=v#dEpEIosQsm(k}_ z)v0|!2DIr&3~E2-tLl_ri52mqKDp2*vwYIE0`EIo8yBBqhOVmo#=67rFRY^Oo_O9>~c5=$|jx z^pPy%m`3hGdF@JGz6f}e1@FqoHex~>I+P#eqUJ$416{h|b7<(RxFnZ+L)MNz`rufZ zYvgBZ_4$%M^G|%qL=N=yjh*O1Uz;!Owa06|9<(QuzE&5;pRphEC^YX?H*Bll%?J8d z8m)siIrW*GHu|r;^o#$-ss}TH(e8h>d^x}+M+{;^oA#$+4D=WH0yD*_n+!2 zhvu#7b4ICifAqh(r(+wviCghnAG3E(w~Z0?$cz;E@Y45=r|jP`wd47sx%o-QUVYF` z``V*Sb<8}c80>zLg>IoB_64EhR=wP{?TgC6^$Y;yn+0&!>7n&^a6Z(MRn`C;C7q{G$D69vE#T4>T9JXOIuT z@n}b7`q>=8|LF)1yc!=J)0EXm)UnlvW=`^_^6?=uilO>6+Gwy9UBMXRweS2r-D)HE zww-;?`8eNhPNBDsC*l#wAhURE46uK7SUooP^j_ZmJdEd#XXWh}>J9x_aY{Z(Xxf;JjgAd^^FY1T&D)R^LMfl$f z&nLY@<-Jqq-fJ%r^og8QKxLh z2MWBJH)b=!Yz`EI$^mcZ%?;>}wrK9w7y54A=Fj@YH*MPKA3E|5{db(BSAOOlO>pcQ zU8}F1{;)sW<0G%odEb{af1_V|_CK27eYEOZzN&9%&vw;4+T@c9=#zcT8=6nYJSo4h zjacD7WCw9U_P#gCf9ga0$g%f4?fH^Ec8;o_=;FynbSC!suyKZe_4xuH8}n^*Zr{nF z2iMi>dbRfY-}q?mXnwJmW5i+eGo6UTG4HWOeS&WIj66_Xh$ZiP+MBy5rz`r?KhO>h zzALu%wfU1AWMCVpOl;0pWCJ-*4y{i$xAAK}Ek5WH##o_S_SG*NKSm4f=Ggji$9j9< zGx7-eMtQX7mmO2|hwcDus|uBKzY)^};W@9Y06ygv+%2 z|J0*@_d|3}*W{sJxl-;G7ve*n^8UO3QTBhveuu>WDBJs9ga2Rl-o5uaJv%esk@MI` zjO!nL(mmUXYwXeWpjU zk&&!y$M?m7coHMxM~?P>s{bpn*I&8Yo^Je)fNSchPY-gBxT1^Zadu{ZzJ;%Gmkr59 zzRGUS!lux2eb6uGu8SLaLEFYmeVJeDr}B;s`iq}F!|3N^n)Srl`!&9xjL-64e%Sb` zFRQPfzH}@uP37WG`>~#KI^bt?Vr-{tK4U(}=j17K#O5iv%CR!=?sK2-h&i;-wCST< z!aw*9pW^p@pWgfP{-@agMcX^~zrfzDS4Um`C->5^JS5K8iazw6kKij#j2HH7+&PB6 zu)W;sSbpw&OuLSMVw}&>tv<*dy{~cJ99BK*4$u9ee|=|bK2zVf?=hW>^E7>w zhv`Wi>UZZb{EQwuw$p>$KI(!j^uaI0x*W)sivqC&oy)lg5KG2lwsXyRV~n=%*wa{{ zo8~?K#c$>9`gHx;Imlu9P432t^D#Ea-uUGe7CrS0CadjYV>ItNdipKk@+CIu_bj z=E`l~^7f%Ll*=!Cqu?LSJ#+-6@0f1v(M6A3Waj_;r{kS{$9J*7r|3bP@h$!Wj+MEt z9({O}_epb|c7a8|}zPL2T9ruD4$EDw~6PAhz{ge><Ywb1KnjL@+*nwYbiw56puBp8C%3RYIoBHKZpF$5^ zdZbVB(XmNv@)JJ82goz>El+x+ljcIUZH{ex(1HGo0eZVHdy}nw*GD$353>8{r>?87 zU*uq$(VpYjMZe@O$^(6UZywTbd7$d`n&;=X@U+eV+~Wv-!}N$<{EAJ?c~s zWFNQ2u06Yqu{hdMoBCjNULQhR8Jnns)_q~V-RVR>#hgADbfnBRI#X}7Lw%2KY;=Nu zbAID&j5XKzNYT1InyM*W~!KO6l|SwZ*UI_Ptu zys?2F9_<5}oR79~tX!SyW_})??-$PNYj_lmyY5@Ae{|9QRt{zApx5|d&yGFj6?|SS zjN`3wcG28i&O8pB8+*l!Jj_q{N8`A<6Qh8ijs4h@425&o>6D)t(;AQHk7GMNqc4)b zz?(hrpac3f2F;H{F%b6H$Sc)cCTKcOj{c|Y|J;W$I*o-pJ85r+U^9tqWQ&80Y6?A&36rJNjdNM}BGiS2xv< zTzFsiK)uIwUE9*;^Ym?RbIkA9o6m@Kxd$98yRUdDjoOW_*oiF0g6deA{@M4sYaK7_ z$;!5!%UWlI-qUoArS@h6x@3>8O?J()YqQ#*fqv~=oz+L_Mj8E9@74MIn60n6PM6iE za&gjFqN~Pa%jwW~Bo4)2W6a()ZXx#;V>Mqoo~M;8>aQOIu^hqQIJ)wKr?i)>y6Ad+z8t zn04#^c8)cD>-&C>6#SN{b8zhUO#54|e%ICSwW{Ygd$jd?rS_f`fO2j7TdJj}&Za@v z)UBawtDfKYfztJxK>Zya>+K_y7w9>*Mz3%BX-(PMqjh6z^ZKi=_ySt$L(kFiTmxV5 z?1Q}e0_0Se4!a)BRX$F}`>vB-eX>1lF)Y?)Qvd-uLXQzkAcXFHgBQr!#sWtA5K51sin? zc1$<=p%2aRWF?n8Xf55?W6bd^!EXk6zTo#w{r)NVZPNZOsoyd68}@#$-fy@2-FWD? znEcMW-<9{f^?rBW@6-Fedf)iiHrOV3Mc~_lTLoJLzBl9BPrlFUyHNd>i0_Fw2HzEN z&9^pvi?H90^1UeE1w+gCqVVWuK9P5PlhOAP{NA(QkoH^9 z{q}{T_q*e%~2A-){E%)8LzJevi7}y*?q|lk`mx z-`DV+4c`XwjfnFC-^}o>kKYG>3H}}|%(vw3A3P{{Nbt~Ljo?wiy21LvhQYIg&4T9z zzF}p*aq#p&oqr0}2^Iy~uNF)NzJvLvpx>tREezkT@EsH1JMj$`-$BtAzlBW~@)@1W zk?ujnwb<&?&eaZo$?!Z>AboHeF4@1Z&71u+ zs==(Qz z*d=&H@SI@%V4dKR!NY?!0>{c8A3QmDQt;Sd%|IRd#{^Fao*6tlcwVr1uw(G<;Gp2B z;JD!U;N;-^;M!p2r7JIeO0Y%Xn^3-Q_o;w>uL!0C-@5W`BRNTq>G$Y-zo_5-m7jda znXPUNwEbz&Zx5c3Zw>k`mG4o}yYC44&gTOHAI`mb`sV2?f(wFg1YZc~?90Jv!FPi5 zf}aGJ1=k1KJv4Y?;QLg*+jmHCVc=WTa;a|+`<-dIogd2?pA8J$+XucI=UZso2B!rp zPpv$)b8ud;>cUkQz9Kj@xG-2I+vdSnf(I}A;AKA){55#Sa?e<9>);i^#=&EQHRrI# za%(KNFt~ErE0;Yz_;m2O;H==M!5P6`!GnU+mN{*i_XN8K2L)daP7Xd2;Jb2g@xqH2 z9v|!#yfWA@SS7f6>guU41iJ@Y1+NL-ADj}L8=M&I5b|>W8nIp0oh+4Y!a*y+%kE~@9&h>YF3@KIB7#Yw~TmF9yEl=G$z(rTL6N?vkf`OZUcn1K4+b z#nX3#uLeg3dj~rN{}#M3*eH-+RtoYt-X*r>rk9(3P$2iL9jq5TE6`@Q;J^T_6N58@ zUj;V>zUhniior6$WS|TrlG_5`7``<4VW91ofXDP!M=fU<@*8qvO7QY%|G9}_HAopwph6}xZl0^o0-2FKg^ANKgpcieBONBdnV?E zzVqgLZNBT}yK3fNE95(&j}A5qUK#8ZyeBv$_+sE2puY`n4OW<3VdguU&!2q$%(vE$ zm^@aPb^|IW>z*rY@PfEck73 zRj_yt^6>A2YlG{8YXaBR{YCKO;QZjs0FUE>g97<+uRsno{%jj;9mtWd3Y?Q~-{7#| zv%xol?+55!9{eu&bMUvop!1Mm-Qb^sHG|ax+ux`DKK0jtY`+P978vi&2#gn>4-N_5 z6_DYd17n5!enoJ4@Zn&a;PJuq#B{zKf9cFOLO(L`k(qC_zI5WH6Hg8v7?{ugHt;<* z^J?F|^DVr7lh616e1pxmm3_z0cLRNc*>?uJcWmCfU*NlgzALj!uwCF=H109n>w7QQ zJ8152-4D74^*)r}*XC#Lb3PUL-j;hH^8?>kbg$$)jhh4;1oE}{j2zEr{}}u}_+4NR ze+cAa*UWit32qPW4DJfdouaatZ^Rn|9uPb@FrPwOneRucdvoxo;M!nu@Lz$kuP z$+w9g9*~*r@W8-#i_OdR&luNv`T6;NrtcqrJ~%k=on>=9`N#P5*ub3GJXwyoE8l)M zCj2!pUV&p}8H%QtTe{rLOxyhX@qv8i8>BA@UKyZm&g$EyzSI1Pz_&s5^JjrM=$`}j z@O@TuLf^_>C6I%hvyTN1eW8oHgF6DcVUOzqd*^=-YER$%WXDB;xv#O^_v78eZyoR_ zjzkeMa!Y;FW>- z=MKS6fqZLTx_=-?9~FEi_+oHs@SWg?!B6J!gQ*|Pj2ZIs*MhGG_UfJ8a_NI_lmO{3W;{_)}m! zGBz0}j9|u~(}LrJj|T4uUK{*NuwLLE;m*lBCod1q3XTc(2(}1(mwLtE`ibi&ej0o| zI6QcFzy{9^9viF_7_)u9Q_h#)-yX=>qqE_;B!%;G@C8!QsI%!Lh;7!6Ctc!M?%!2JF3L z?=WSgj-c1}TC&#+{{xmBFtgWo=U}_w)xp-mtAba~!MVBkj=@fI z*nY|OOI{O<^|o5F)y&*>lO>xhc}if;zRn!hU9#@X(V9!vT=IzE;eq2v2kQlo4;()# zcxdpDz#Mw5!1WD+jf0m4WYNd>2ksS43{DTu4=xF=4*nASEm%0caQXqkTEWu-_d%}- z+=H;+5rOrGZwA(Geh^qMam?PmPQ5h)Yf$FU#!Peh^@Aq`)@nAG!&8?&b>^PWxac0xT>r&^Z*0FY zcxSMGz?VN67#qJA@OQp%3>O#T#aLyn`ozFkVLtA;pD{>$%IBV+Y#;0&_%^fe`uYy9 zZ}46h$jP_OgKxYmGY6EfSD9F4V!hxc!7hP(eN^zN;P}A)nBe1qx#atU0|Lj#1*Zl- z3*?^VCzqdmZ19p`x8R`Qgy5^e*8}SY^8IM84lA_*@_#UJ=Nh z^68@BnZf44w!tpJ+k?G>4+cjECk5XKz8Cy7xH7mcSZSG+mRS@$DR^e^>^W?-%tp&> z5Ud@n7~HV%hM9J!EIegqO$W`Rf@6YD1V;vk1o+{*Yp`{&QDAO=%hWA1W5$=JzBKj0 zz*@>?ficb6%srF$O#VGs5@dN}26s)~HTma2UcWGKKXh_%d~jH>fAF?o``}f<=D~}C z=LQ=E8_r?l$&F|9{@TgcPQEGFDcCkJ)@~X+C0I9jNU&1i9>G1?4Z(H6wZZR#-vrid z&IrB`938Or`-0tr9fDT`*3O?6tQjm5T)W`fnQuN@w?8uQ-DdX)?hkxZlApRK@PAL{ z8;b(>1bo?C#+=D~$9S_iFpuhUHuIyMbjS-oKPWCj_n^5LgR9%RRnh2c*>2|66C;jmaa3iF17R0oo8~;?n`%{ z$w&HSotEs20y)n*k^2Gb60Z#o4b0y!nZvnD&z-rSIdbWdOFt6m>p{U0!Dj;ZLe}5@ z8n~zOOw79PR)M*|OM>SIn+$US&(EyidFHTNV1DkIp685L2K>NWch!K;cus1r+vlpz ztp|G6Y`*Gwlka|;xB8YTKRhb%z4=1|`Erlooq^|0`^HcraGd<7W+S=c?i$#g3SXt2&`@VTOcp} zYp_A!`JJ^XN!Ig0Fy5K}UKY^HH-l3Hduv+X3BDD0?r1IKW5Ig^&-M6&_cN@6+4BkC zj`i%4FWGxm>iH_Wm`C$h&tTnuDdRt$hk6&o9xZth4>|L5!KVWD<9FVn@ca& zHm`!s7a{2XoH113Md8wJ*6_>*|S!#f@B`}mso7_JVy+hyKi zjYbZ&_pIIg-}4)O>)H9{f%r9-w>BZq+Pfb&r+3dUXZ&aI<>2eVnZenCy|s};f_DU4 z2hRWCf<5Q(o;)kuH`qUTckssG`GGlXZu_Qh zS#nG6C~jJESFrr_^3$sa*3ms1w0`fop}B6K5gJz@>a8%(DZse=s;YI4SsgV4i(uU~c{6 zfUa&2j1R^M&k4R82iC)^d9nKif%TwY2iDfQCNy>L z)XZ9zbuMGO@!-wDfr0x>ekVuve@z`*JAwM8TzXh=Kwz!M-n!!F0{5-v5q)>@!aM^< z!+q>Q!LEVvss8+^dmlCPeuw$^+k#^P`Pq7mca1%V+BGoOIU)FZAcuHA;L6~hz&i!6 z4&<+21*=W2HZy;7?|yjj`M`Y7-1zL^;^6vV`Kje+?)zUh^|GnA1qTNw1>X*=Ygmu4 zMq&Nns=(Uz9lpbT&0+x>&703Y*eW8C?{ z*80Do*S$mct_XmaokP&k5vo z$MX8Ifwh@kgI5P~z`q3Uf!!NFE^z;BKI~l$^UD{+yx{4<o)lvKMefF0Ff&Avcv&HonuD@{S;6s6RfRlnB1@{C`UgpWmY#qEU*e5tVI3u_* zSZmp}mVIHcO|W&aMex$#MZspl;{wU#vSlt?=Kt_@C-9ff_5c5knL$dlsf0nBw5U{+ zokS@uq(!z=qO{Onw$di;5@jbXN_$#lDMg~blG4JAB@{*WWhTGJ{nL4W|K~rq`p)Ba zoAWv6T<1F1_I@qT*Xw_#PxHopL2E`8;8V4 zg6;og&?#sp4~rXT-#DW+)4cVcs42cX_Kvs3dxQ4;{@`=i>nFs6VouO5u8a9WuVBAv z(t1}`4|W}E1r2}e&|ktYyg97X-}?j&|C2$xKO^|D8$xpe?VMksr{6ao5FYQEeb?-Z zgT8W5*yjD?r;sn*7T3l6P!Iokd?(aM#N9iEe00N5OL=n8?c|T@>h}tETTHH> zQa%AWfzRr>6ko_;^-OLPVs-1FpLkTudFmvdcRhP9Uyw778qRZy`(GAnvmWJ)@@2UR ze<7D8MyM^xJH`D+hS=$?K|j$8vs;Mw=_BHN^}5^tb^A=cRV>FRi08$7;=VV>E8_)W zJkwz5vh0EQl1~4nI40PVeS)U)v{)nbSv)kvv|?v{G2?apF?TGyW8pP%Y5XbX$AVz9 zE(o@tEo0xLU39OI!^mB3i_+$q z+(RDn_;^(CyNebsn(?>iE(ra0&xk+UHGJ2XYiIXYs~${uqNUSS(7MH|G~4Hd zSd#WZ8=o5<$KRaUZE<5}rU=U<3V1dVp@cyrKD z*fBcF0YTTN-O%*E6=w#0=4YWtM~r$wTo(TfF|3$)m3U078(Rk7xLer2cg6eSgYo{@ zH`I0C9Pfw^gdQL9{So1BJ%eKXcg7yEZ8+|1u=v||(x-j@9kcG3b#44EY~Q!TzVg%d z#Xi~3*M>UhOXKCSQ@k>E4Y9Uf!xx5r!!2U{7_m3aOI>v7STfl4yW-9$w3_KdTxQlX zDco~sd45A&6_>{qLHG8Y571L{Y5Xl}Z&*SM|LAyfJUuoGzGS!HTj*Vf1f7fD@qE8{ zOY9N0ozAs!JT~qhw=Z$~%y+2a$}`WKK+mG-9TlI9PlQK$5WQ%xpaH#P!gkFu+A8=A zV}kahmw;c=?;wAmXVNR>*YwK6Le4<ye+N&B#|4f3 zVez22U(kx~6ZCie%KAEJM-HObSB#Zo)!=tLiskghsF&d}@%Zqawq?6eC-7+d#`f5T zN5@LxdzTM=HrDmnpgr6_W`%9IEA*n=9k;~agI?j$bIUA@Tn2W7pV4P4HV`vU_XF6M?}YuL zuY5iZjgQ6$Lob(OU{1li;(%a>T}Sdy*OmQVqJQ*AF_UY0m07FIi~)5hxjUP}20vi- z17`0Udg7Lxv*euZS*pfd7yYj~gv;ejT-oH5hCw>!_ z^;x>pva^<*(G?!i{1^6TlThQOk+X+vrAN=%q^)A35a;^Liox#DAAAq{LZ6l|i6b`% z@uS*@IV1R~^Q z@5EQ*#Q0p$bKf8D33=xmgFd@U>=60}{Jm$0C+W2G1Ap%obkAMGvfIaY@sgn5THlA` z@HjSn=NIFY5UrO4!GIAGnhSOs7pU7*}TtWA6 zz5Ft4i|z7n`F}0{uZM-}Mo#3qQD>!{(aT*cdMWL*hlXl7{<+;;tgT^d?eIRz7>w8c?{|& z#;o~Ou3fs0o)`P@`S^42%lilawQ2RQQ(}$!13w$$;5WoML1Wl4z7TYP_3HI_dz=(M ziC@J9ac!@QcMLjT3zx`U zI@rbyWBu4Do*7=JGg(Id?~y;A#`}DVxA8^U+xiai38%p@v)$7 z9Tvxi<8x-%C&%Kda6IS^_SZ2t55ar{vCt*q*!?`jMUJCeJT9zK)f6 z=I!D9==_3H zz53HZQ~zk}7q5xULQkDu>F>t-;zjZBu-@Os_u`m1INl!g+^yp&LAOX&Uy_z_Y5Y3a zEznf{95k0jL2ppIn-`Cmu=0{C&(!knoOS1{E5m0_k5hvVLyw`GoF8g@zC*pwI&X-- z$6vzh-oJ1H`*~T&1=I+M4)+c9!&PJTkTdvxHBJ38ovN92&Yv}Z*8c>1N&`|;JT~;* z92jp1`Uj1JzCkazWrKpo6jtI3wdb0EO!q_}$0CK+v#ZsYOXdVO`B|oE^sTrE9 zaLT0b``CMqozaMosFot1)3bF*&ilcAkG=>s_y357#dLdR!b6M@{%#_e}Uq=(l`-n5StQUKg(k{>ZlRhkTXm zl8<}g1l#Vlmrw4g@apDByfOBRPlR#vk;2Xr|3x_dt?4E3iU#A?mO*55~y7XSV@ zR-3)rOr7m_&3}3HoJY@jLwr51jg{uEGGJ>DCiio@fu_-MQv03;P43D1wBrNll@VSHIow0j}&Fv$-X{A^?Zf%bL<#9p$AWjT%`u-s&k-t1Y zHVZL3U-sajKkPA`kLrQ zPYZoc;&1g)`qTA6H~Lka6{iM$S`Ul9RPpl(6a4LUwNw46=Y-GEwycB3{`a6o(dMn6 zKJPg_p59K2r;+_S&Wx|e=fiwe{!XvVCZX>6fDoJC5Ld+SgOBr@1*gTggWi2qd@}en z8m1cPt}){84e!}-M&Fj>&Wja;UM=nx6Pv5Z9`I-14*pC0_`cXDjA^;PespmxA0x-7 zb%^=IWa2*gKA*Ev(2i&&dTQt;PmZU?&gXc}^TIX^o}7od-^x9m9Ey=80|{Lm|b1cdRGbG=ezkvK5eIaj$_}`(` zTu`r({iI?1Cin%9G#r|Z`9+rJr|CU>Bfqw9>>V^m-^W*N9sH_ev2Hvi)Fb&^J)G*L z`a*3pf2?-OKf9m8ZPi!*7WVg#@%#8)c)TS39FEWR;rKb84-3b8i*PRNFCQn~U!gfn ze1dpg9%7!v(cz!I0%K9nnflW&LX7$H_*`5ZY#}}1Rq??%A=s!hgMDVtXby+R+vE8$ zVsbUmQ{vO0{yoZHasCb(#691}qE{H#cm|TpjXPjM94`rin3xkSYBhLC{b1r^`#(>;vM>p$_+= z*gWVYY7-BOA2!`3G&{_&7YoP#+3;0ozR{cwVrzG<~*}rYnbDIoRy` z1p6w_UMl=evz5c&z3AWgTmE2qww;D*UF)nx{i6_9!2rVTf5b3p#;(oeyB| z#j|{a?GfL~QzYv*#Qb36#PaeFF*|MdfRJ;%B}U9H-hOGYaU)hAf3vsGo8Uchy4Za8 zP-n3oc1d1wWXKWp-=7)h1zSaz5QjS^G>Ca&yX}iQkz=fWB8Jt==NzzWG^+2!>62b^ zF}OTX4F1Ph5LZr^pXMY#q+MCYvU&o=@N&yzL%!!;W&E-_6&;=iLw{1YVNdyJ=Yg-) zm$-WH&(53LjD5Is(Va6oi~NE%LzAb&oF6|4d6c|K{zN~g*;w|bC@h=t@#{2)RD6F> z$Te*LeY-D(>-W9kI#-)j&o@43ox6rujeTQB*q48X_;uG1m(ozg*kWz-Db+R~5o=9& z4(I}LTt*$v7ZOV zdnUZEo)OC(7Ig2UgKhm{d?S7oKM7j+nL&^Cy8WV&d;D7PHG}b7AH6x0*1_KTtnIVz zqr!LUyZ=PkHrr`?KN55#f7@=$eKrma?}@S1q4liL@YMnMP5L+=_w-=5^|7oG^c%iT z&ksL$dC;KF2zLDVpndNjVrli*X9c~R)-4aCSJCO1FM^xUR&;_nX`|mwkmg?{abPxNoc&^J2|OkD<>G zo?MUPda+?_6y`0h9eNe5-@5cY(hmO~^rf@n%W-J1+pmw8#xvu|v0CU8q3O_y=r*(> zYJ2RvIyk+AjfZ%7;JNqdD3)it4-4xOL$h5!3%VAK%Qna% z)TuWJ`=H;GX3kz88EhDP>s-<3_*?oMP41!b=x~ndgPVkNzEwOg`0edOzO`+bUw556IP~?<{f54%2Jz3J5AqBA1HGPK;P>e_^gnve8S$eyHTW3YPqz^Z zoE-EU8lgx2!*wTCI5YSb8jyPi`Of8vG4vH}9CXlk#OLDl&|CH3dU%Y-eM2lHKBJ@4 zalan4Yx7Uk;ppjC#)D@+c=nb-*Ell7!n4!BWMw{6`BX#gdRdW(N`_*rA34LH0PAA{=wKU^clV}#N4k8I=h}i z^G!F5Cx?3B!$J+xzni+(!sT&ch;iv3$HvFQIXACVKeo7;ouo~uAF{Fj-Q4_N^};{I z1@V&@^W4}=8vNlQ-qmOJ!T3PXD-H^_RR5Lh?kyo6enY%2Xn(te9yw!6O@~guMQjvL zi^subv)bS8dA}qI1&VjT4%tI7-}N{Ii>E8+%5JSS!S$;ym%Ld$zD2w}jdh zJEoUJd@62YW8~*z#k1qI5I=q+*rJnyHbFC&=d)969q1M24(S!515ER=dW5|W%Z%kb z`m8+Px(*Na)B5#J&=Gv6I9A+D_qHuKC8oD&*j*a5xSDSA`Y<;}JS}dfSY^0seyK;rQBb=RiGULC}ljHFPQVU+%G0@Q3QA;%G57 zU7K#sp3=cwGi)y#={mYO)MNN{k2eN8J?b)IcApckv%{{l&rf{3da;~L+=oQO;Ii%*0+^SfaV=^x|z z@K4X=V?&H3=8`v?+bvdO2gGu6^Jm1)aZr3Q&Wr1VR!fu6JO1hrce>Y;IG2W?51qC^ z3$8_a_Hw}|49%Kt7Wad#WT)2%S~qQb`Jg2)5qB+l*UWlu>Rz)qhMd8A?+){Q=;5?y z>k#`tEBFF)qTU_y3i-h|gO>fq|oK=IL*(6>VyT%)1zc@J5e2_$tOeJ;`qNk-Vm<}K4qKOG=|1bLlfuIplDea1YhF39v05=KCxSD z8|t%94_eLY;d;2wM(@Mn1bQsIiO~lRo4lQBhpq21NVr<`|b|$wMi|-WE@xI(&51YqPd$SDNrq1T` z)@8lsE7I(36a9icR1>E6&|KVqMVzjmlHNfVxFl%JH0X1J*8JlTU(?)a&g?CJqb|(% z;DpedD83&)NWAZJZ1b7HSN6_r|{Q zj?f3=`k2OR-cuW8@6~46cAx)j*uIk{eM@3;+a|83{kR_0W7v0nNot@=hZz0-@sLk-MRaJ(Cc@b!gH_h6nu+*fAOQw&|?Q)_r7Ic6?*;I0JeZG@R2wy z)Zf^VpT_S(9gSYh9_g(U+wKzYj8DbMq4(snxGiY$YsSW*R!qk>w?{339&>uARhl0) z=1vLx^kDKW4-EcAefg2WU-4T{47mfJBj%Swc=Y^9p(e_fo7c(jS%>~nzD-`hcljRk zwDjp;A6Lg^!GGGOvw{{+7vf*p`ger;X1zGnP@fPh$Gt+mRIWh7G51QZ<7dO1owo+t zE+3%7%i-pQJfO>L#-8iv*Q2jj>W3j0_(BZb=0n|McfT+kyO4(P=%_VST>GKYeIo zG#Y|AJNg;u`+O1&;W=Ty=?8Le<94UmIsCmtxGyY?fiF`h8^?zoubS`pJ$lYQ(rg^}XUCQymvr5+q1HnO@#t84k6pFS7liZVoa`FA zhimt3v0sQiJ|1#iHuj|WT6{b7r2RP9+n>iT;)3{}xG2Oje~twqwh^ZrJ9^x5&;R?o z(B?TATeMG_=R?DITWvDW+W2!1C)%by0^{S^bN`X7B_ndbPc9BkI+@#1)GyerI4Iyz_~>W_L4=Z8E+U!<7Vy;szA>E5)e4MQDu z+t?}S<#h9X;{!o+r>)cCPY8Oxn(BArhjB*yJk(5o6XJ7ro6e)Ys&9YfDjq%OTht@e zVtvM(Eb+WrfOwui^1bg3HQm<+?f$u8d)fI_Lru9d%&dQe{j|UK`&)5hd?x5Yj?G>{ z2YN|7H_Yv%-SeAz$>~7%3-e0#tK1MYeEsA6t$VN?7v|-e_a;7nQP9ktKRW!vdlt_0 zio2fWnn#DRK*xDS7(Z;Dm__diTh2DJS!)0Kk=2j%e6e4&PBv8z&JOsSUdhI?336_> zj-92KvhQ~V8{q;IGr@Q0^%m>O1k( z_(Hg5UBC7j#<+bK%Y7s0y*~}P1f5CUK%Q>D_z>UywBTpt0`3=bKzusB816@=r$YRx zzEIoEj9pr-`XZaZeu(p56g$T5LFayZ>>nQpzvV&4;0tKu`p(%z8plsUAJK0@51|-Y z&x*N1cg370>awd33H^v_q~`l?8GOXf!AI;J?}`J0pHPccBWAa0F>1DYa1IT1=7Hzl zx2zamZJDomThMIW=Viz6U1|Yxm8S)b&Gwilp!Z)cqb{H?iEsO7TpRP_;;@fD4z=6k zLT$kD;RnztLaw&2Zl?c3$o#9&r6RhtB6O?1p%h9XW2&S1dNAw|LEa>Q$$NdX{*QCL^u7`28H^d=j;?4Wm)z}E+cNGKbcdY&!2A^KG3$xc=IW=3CEQW zb{wbuuVXfRBwfwEP3shn8C%TWvd`@FfcftSX>Kli&OhuI;&}6KKOA~Qk+ILFurF*48z5#B_p#H% zezV=|0&Qkm0}vB}Uc+X<-(DM$AFLUJ-*NPvOnw^n`RqryZw{1(6soT zX?==sSR6SCF7B4gSl+tjN&c~)jw4OQb~rDN9X-jhW5d~9 zIF7d2v2-pSPx~>A*Su#L%Ug%_S+Dct+_KBVX4_xeGWOZ_*-rZ-|AK8-d!^CYXMSe* zX4`BVeJ7v6zsLpT2+oNZUQQ6SINa+-yi9w4 ze8{!=ig_U~;6H2!TVBif|7tUImdoQ0L0hszA>VpX*Qy0pG+{nzS@=GDk4=-BjZe&)K+W6oc!7>^3RL7kJH^qSB= zLpwPlXy<%_xki7EzlVODdp9p&rFe9(y&DIA@Z#7x=r(-8JLAA$zuA5^pQc3Tp)2XF zKX$^=^_HsV>gBW?yRSy!J`HjXdOu&o=5G<}1}*D>;ol8)3E4*R|99e;Fb9gwmoIG< z>%^+TSBm@D{tM%b_*&5T4~}<*qb^UAXUZe>3b@YYjq-sz@40g(-l6SX65^qsP39$y zyz#_-6NbjYSExN1d-4V~DL#Y7v0czs*;96xJvNTTiu?z=&9<;B({Ui1zqtPxK4bCN zkgZ`m=tRSIvMKB*n==i2Gi(qW%9gUL^dd33*mr)&AqFFtunarQc3ID~&1K`+dUk8X zq-+-(Y9HA!{^8$h#lQa_dWHCTK)aZRUm1Aaaz1Tc`V3bIdB-?@>^r+`JJ^25j9rG~ zMw?-y+1X*A9W%$s@o)^}0CFt3om_0xOKJEggnFrb;?wc*p!KWGs-5bC@#wjlsG5ZQ z;}cRFSg0J{pu(fQX7|D2`^RSb2d^$Pg2Y;RX z-X0r0@EW{tIiK-)Hhg|quk{;azU!j!y%$bt?Ov0Mh|l@1%ffb8A6xy;5X13f{!Qm1 zY`0o39d~hm#^3Ne{LkY4jGq$cv+d^83_SO`_xU}(OkTsEu_JO3e$k`n@E$*DIm`R( zw7=vh@D z@LWwn&#dLlO*k@WMq+=t1C2;sg7zflrw`GjXj6}md2zpxr>J$Cr)G|t-k#G#O@t1l zF0Q|aUbbhvB3=+`DeHz_{?+3_!9O(JXGVvUoBbg!2)&`|GG7YwuJnh#H~3-aL!QdV zJ9p~_pCMimD=iapVf7=|x@+4tE4H8;j($;nJV%H4Oq`}i)VO?WsI~DMJBM1SJ_qAj zJx?u<4-xzEBjP|YI6axhz$UT_;#je%xKs|oCJ%czVp4X`-)tbe;(fN!vOdFhvZ2<) z-dHCa%hs~FzKi`~m-#VvXy`?@OP*mHZL4jz?e;-^QOwFG+Bdbvf#=?{41cMn!lsQJ zWaJ_87O;7~Lry}MqEpFtY|A(vwwJagZ{i2!1>$jjkso7U)nn9P)WOep35-5;+7-J!@Eo)y{fwi2K-c3#tef^lW72oYm-3-(`@-O#*!OAQY5Vy(eou}i zZlK9IzK(~wjN|IK$xnBR-GZL)T2iNB_ucc6w!dGf!MNr=6k?W-h5C%X8G4Yu1u+dR zTn&dl#7D@9)dBblHDC1r@%40U&4-AMJWk^^y*S2zxcl@Fvx~iHu57)rGVN#h2;+!- zpROmkVCO6`=YMr3ak(62UdZR!bK_EuqAnm#XWQX#ud(-P4)PgwW%-8n$T#R$;y-?b zKe;_9W#5HvJ72LB}flwZ-x z_zQMM{v@Yhd*l*t2=+*AK`cvmk`L$~V!!k#ULH3EU5UMWWYEOfLcK9N$LoWwd|!M# zGQN0w)rvPK0oTWv>G*M_1x!8*t+`w(fh^pdU-Yu z%d+q0YOObg=g$a_<2`zmdiJOjsB!aEwxKranfd@fcFE-T@FyFvTPWe$(FLOY%cr69?%8ENc} zvHteD&sm3cStt9aW}u!S)|5BMBiJOdGF`*Av7yttlWkr+z8B9g?i2W^Q9~H|6Ppds z=}802v2W@M!;V`gpToC_@o6Hqg{|bT9J3L}J9c7udCACqMy_Mu#P=iDvA>S*K;GlE zt0&sh;<}RjYT9QyHojku!PoMY!%xb0Y&XqDeZ>BV)9tVGuJ1-KJ0B-zXT$aF@b$wt z>WP9HEyM$Cyw`@Wv^;<5v#zOu=U(^zK)rXsU%+>P54P^npQY|44pFCnQ;1FYj{`!@ z-}S5yNDjegjC$$t6Er72f^VR=u={MjIEZazC&l{43tOtLU<}cq=ufmCcA1TS(xlf% z-28vV#MvGPULP^K*qkp=FL-qD8*IPt^^fK+cM-q+HrVsC!}i+`*?fV}m@FCvk6UMQidC6V)HruJrZ6D)#B zLd``Gt%f64I6LS`eDD52&wEku3y%uhTyd>9d&Ky|4i5dbstgIGUaEeeCx5 zj2M2}Ua=`7-c(n7RnVYm4Qz%-_QQK}8T;e!@!V(GDEnvK({a9S@%^@$?c@iOmX4|O%r7|)d=VeQAE_^3wzVX5&Ro8KbUD$FZC^YhI}DdK9lu_l2rWc+I%u zL)dV61AEH0vd8?+h{f6H5v$X>=(^rFR_BKpUpz;nvi;t(jAeaJ3{S7}IO2Z$YMC=a zeSzPjr}>O!X;pFvF+XTlv;w&eJ7F6gGoRs4Z0q=5>x3M_vAiyP*71a6Ck`Do_fbDF z#~*TwQA=_BLz5051#Or#H=7pR=p1@YWBh-+O2wKva;hq-r<1b;(0Q^0;vN_lb;*a6p zK9+}j0sTJy6yDQ^dUm*Pg}Maa!B5b&_$4{XBZCHZdH04mD{QxY*f-=1eB+j3t{9)7 zuJVvjqc>OMSE26D=kmWaJ3c|)CI;aX9BV#Zoa5X(hiY)HjS*weUd2e_2-g$*&371A zbX$FT#-TWnj}X(*SJ-c3nN6n+u&?4ewtCoE_)-R@IYuI}0R#Rn{yf40`W3V-_ zt$d1pboEZ(=ezg?Ih60)IPAAaz3ukdzKLb+FW=*{_HWcn9ZNZxb$c{N+cBnxm`_AU zu@2ksQ9Z{x#MZvY_wz@NiS6fe=s0Zcs5|?ceWvyNHAdZ6E+XfUf9d7=P0;-1dLBov zr!PsK_J;}H_ul9Q^LbhlKQeq3d(XdkKK#UZZ}>s)%a?r4GPYff#yJ{3*!f>IcAUOiXOroPOfN3pf+rJUljpd(qI&$*AFbr@q_H|C5zwiv_Cv#a8Au|NMq2jWlY zx}F=e{0cj(4lFlOKd=m+<#X0+8TQp{)<0?^KJRtT*n8~0<;DB{wybUAPvjf6d-#Fj zuN)`)A_s8{jt+!*}Y-lUUeXs$a)ngA;y0JUrHk%|c8~cX?^>6|bGlZD42Bq1jzE3Hm;r zpRv48nD<2k(u?%gP-j+0K4!uRli%$%Z{Y{wQLRCZK_9BVA~^!PKd_Et!}7lhKEqrG zbLI3X`HVg%ISU^}nZNAlkp&oivd^PCk z*2(6G)om9WPcLB;*(CK^wuQgoulOf6n|4iK(9_|%Gfs^W<4n9{ta_cMDo!#s=&;6= zfAV=_liw8=$)m-QV$R`%UCVR;{Tlkh9DA{l<87Ss*R+8#R;{mI$jIY=A2h{ z-@4dv>lrzUNBS7s&7Th2?AVGA$QL>f{5oGS@-G^lT$I1%>*OTY z2ftxi8WG#i_tMcm8T-Vig3pq(^3&o2{$76M`{k)#=hOLEaf`e|d?P1j@5JYPgTJ^X%n_sQtIKU1Vz2jv`JLwQvqxgLb;I1Cd9iXBlbePZZ|BfUvRkMnvPt4X ze2EK zIiYjJZmDC_O!O%IIj#-I$T?BFrn9qwKMo1D+N?e(+s)o=5XOjmVv5z-YW~1HFEt;1 zZu}LE%sehO^`zj7%oS6=p3_`p^N-zs=<(g};qBcIM9-GqU3GM^yu3zl5#8xI-K*l; z@$pzK#$#Uo$Ji}M)eaHK5^mR@$L9~JaN$z zXYP~p)9!>jpgG}aZH#eYs`N*#00Ks^+k4uZF2q5!rAk~gFRMnvp;HBZ0tdC zPCUB%G|3m`h>rVvgU?(mo)Ppo{k{CIc;JfWw8$&vV*Xu|mbvt-rDyKtaZ>ljrb$}Y ztHU*~28k&D=eCmF_eAsyHO}jc3GF-Ann{I3&Ivx5bv-zxjjl<@io~ zH_nI!@v!bSw`Zs$`^Qhv#pVS)haGji`B%P249~Xn;rfg8DX}l?41LUb5+{fs^sTY4 zv;;AseoXqf`l&H2CKNC7lQbIVlKw&47gvh^cM3LJe#fp~Q(gV!ct`9SZx6?hEn@F{ zH$NnoV1w8-KEZy`vYZ3=2>ENjf%f%q&3z+&9am3&5ACw<-Em%=7OvIHx$~{Hi z9rjs|l^(3O#B1aAVSMPrIxO^Ty)AYL*EDvHH-tFGJ@cGzu@k>5x0JKWN9Cb%9X``> z6n{IG>K>tG7vxZ_e5NAup9Dd*pwOCAp|$O7rEbX@;ZsOB_J! zbv!Q%y=mqb*=OhbJK;6Qao;e<{VDO}*f?CT8^v1T+InO>FZ8YPC(eO5MD90gSFRaj zMUEkNqjT90F_5@RUZqx{Z;4-gTyqkPWxq*vukP6(hBhbQytpB5jVt1;&=X`nfSyo2 zqI8;bg1$4me~%8f_;vA-_)7dN{t#?2u8M!gefw?lr^SxpH_P^qcgH(ozxa5t-G{~- zW1YCE`>_5bPKi_FcX3y^x6uaivUp2;Fg_Xw#h$TcSmu`5x6HmM&Ws<2-xF96x5n(8 z&az><9uzCb8sTxNm_K{|%=)16*J-zYBVq4&PwXG>iZ{f|V*R*( z(8j(PZ;CBq^vXKkch9P*?nVvbK|}eau2m7`37y( zn3tQ%IcQI=ao4<9()blKiXYe;PW$TWrZft%o=5R5ok9$2uI@+T)$#PWZ~Sf1-)8PB zcY60fJ~t!RxSam$;K%)jA z*SS2{aQ25?G$w89_5ErfU;Fd;QLuH#$0x(Fcy(+OPl|iTWqFow#K&X*I3V61?~OOc zw(*QuD;^Q6#tL!U-M8I+ar_{@6vxG}aZDT)pNr3j=O@I8;eD@tHVzNZzZyS`U&LkM zhIXs%kH-1M<%*M$( z(Hht?toq{pZav-4^W^d7mUnuT7_=ZBbC|BhO%xcaC# zEA;BnO7x_NrFRKAi1Uwg-kh3RvZH=I{>*>^6 zqb{qT(!KgG4?2c^P~Rbj|5DH}^aP3L)q*|J!fls)XW=#T)p86mp_qgIByLwf74yF{-W`t70daVUgWbDf?dD0F-=?QkekK+i z$L6NyIoqdOLcHjaEnGWx3EIE9zyrJ2(N@8pejpByljH1QkB8k}e$Mj!LfXS;?EKs2 zylu`2aYl&Ye;D72)8dkl3#`y@>fAfz1HX-P;-dIREHihRxets7$I9`T*f?GgFN#g$ zQL$89J?H8=G}JZR0tyUaS;1cF%CW=o7JfY!Q!$S+Ssc z3E%p?_)^fpc8G@tO-n8O(BLcdOo^j64tcUT+c{A0a6aTsv=UmT_=c`4UpEHj?&3S~ z9(~^!ql3^Y*g%>9ZGasVV~anY8S=bm#Kz${Td-k>UBoNwI=i|4geP>rG%?J*V{R-H z_lp%n9<*%Sx#-TBxbb)0fy6Guq7!A=dW3SWGXdeZDwoHsUL~`(NUJV#I3VH!-Q$)3MnpY?J+c zOspC9_laSfw@@uEZ05lY~wn?t~;;K3dfYD zVIRc<2Z#KM#$nC`dw+O%^tv3vv3OT}Amm*>=ecXgHS&mX?)i~FhhuVGIPax^GxxT= zwcj?lEee?zE%mRZW`4I>93iJTJH!@O#4T}0h%Ii2g~2!ROU}ou!Zos8xL1^G!FHb* zVu{n^?2u2oHf{_)i$3$fpmou1)K}EL_$T8<+|QS&KN&-Gd-{;vdek+<*lGiE3o)6x z^dCbUu4XLP_IWuOU#IS3UE&rwh39hi&xPDW9Dh`ZZ+zF6L!D7>!jByr;vYWKXZ5Ij zEDnk{g`7o=ZIf7kQrnX=tQ7pSTyxc+HSw?BlUL9wjSq1b-)_E(`j_!TR}f2CpBkHE zqmDuAu)lOyI`Czo_D=s*<2UysiL~m)oinv{wSRxhU-=jMt(bP{u$=^bL)(MI%Uo&b3PUCiv#0B;rTJ~%{VXq9{kq&v0c0*HVZzCUpqIx z7DvZn;q#vek4MCD@r5`s4v+W891z**rGr-I{!wZ@uZ#8Ko`h{N zk38lHaYgrW5>uEn;~cvV=o7|`>)Q1%_7l&!uIM7{uv*4`L1&m1Y=B(j)A6|we;*fz z#s}lQad5D6N5$vEvL6eNmSz8b5I+oYJU>IP5aaVNPYwQS)!?&+&v`|^L$F)y67DO< z-m!n*4nBk?3}ZwNp{7IM*IT5v?C%j5)7XtQF)TaqsW7gMAMvR%tzXlzVF!+!^q`p! zAhz`xu`XZE1`HiTzEA6;ZKzE`P0(^|gJVpy_&}T-^zFX|jqRo|MvY%u$Tvc~Oh0;C zh{b6g8;2T+JWB2)R**lk4?hYT5g+ozs1e5?2^mQ)SUIO>Sq-fv*U6k`m;K)deA2EqIhkv;m5|w z;ZgngW$}bqD(2TCJGM`~TfZs2n&!=({U+ou`l8IU-YK3JtA={Gd%d0+azESqpzd>R z8`h7Pgg)4x$9=n(>vo}LLDQmb&Fi-%wuo26&XZi?NpqewV}JiO`(LyF5m(2*m9#5o z|1EA0y49m&>yZD*M}8K6hy^i!!j*HbocaCHzs~vVoU7u}@ZODazq$9D`;6Ez_KCN| ztK!+QX51?*cX?P3-^AZ+5WB=X=E0<ySn45Tf930M}^SNSN*YDNobv`)gWNK9MSMzH=8}j!3!yFv* z_0^8mk<=Y&=JJ01C}J1!o*3V?&9-k8j|%z3^6`LJEsTYAL%h6t$Xmpsf1Pks@u(Pg zUKr=}Z0l6RH=j(LNo!gqo6AI7vfW&=Sw`Y`+Ns|e8XVF*#O!fUD>@>#nJXd%*|Kwn{Nwt=2;=PcucGk z%f%fF@0j^5%AYO#*}{{9rnGw3FAdQMXIrv)l1piK)7U#r8!OzPrXNV}I*+EB_SoH#yz) zaeLeo-oJmW9&5*X6XbM{jkRJNgMIqVA-?AKA#QQK?iqY8&6KVGahxAw5j#DvI+R>r zzlNORk}#i5Z5c2pHx$%S?ZgS`+qg~KmHB+V~&ZVL*D6g#ZY^OoK2i3&f+t_KjF0c$!L>uMse~3o3m_g z4b6Y|7+Ra$#{CM_(RT{xQ9dPhV;iiSefUTi-N%^(0 zF)wJc`p>qEX9xY-`Pw6#Grw!0|Ie7!`~0)`T^PIiJM=N!7`KGqhxT|=m|sa3p4(i@ z2gbugU(N=huD@;U9AdP+;`Q;GkW;@TUKnCKi2bm2i20rwo5sfBm~9@<4m~?Bj+e&{ z;nDNw#zr9~)X%La<&Gu({ktD>ep)B=k?1v`af>+}?+e2|yC#hn*RSjNeX)OtPmL+} z4>&ON)bAboAN1L85$f&wn)H~_B=n`-5Z48rLqbEk}F(c6j_j$X|XQbSK*jI-PZm&(iQ{1HRYy+Ya{HHmlp&FYyd5 z5_HS4pK>92sTwD(kk%;%pl#DWY1->eeoIB1Cr%I(c%8mTixAuKx%5(xp3^wxwjb!4h_fui1>W)F(<@Vf(||WiDT#7I?wm5hh)`|Z|TF+=b-kb-Y@@p zWvF$jhtdedf37z^(lvNiToC*|-|qV56U^&bF7#aTzkI~T!Po1}+d0gUP!H0lXWXzU z&>u9$jj;pIKM?xrXg*&JHjvFRKGi17X;iCtOgtslkHzhP`VsrWuF#RxI*t#$yZU(D zJJ)^H2}AtH>XUn-u*=U6wWNJQ&c%kG6mki@#I(D!CY;gz(|;7-k5ea%W#nqVjz7c| z@wdQ{A-^-F^O<+m}83wa9NaL3pxo*EC2`-gfTjqxubN3lau>FL*EIW}^$U`J`)yN39RR{!oW4jczL&~b5c z=*#0z`B%F9rEzW0z{H;7;uS)k$nNXsQyVs~cdHPuZaewy8NOERMBCIiCl)hic8T4B zUl;4XA&eohFP~%Vd^n5~_bqVGIQUz9#z%<@j3?ucFW1Yj_G+Eh@4NXZKHqkU1H@}$ zf}xd*;lvd7$9~;?&)qX)pADePI|gFUFNI?yM^u+!PxPzP`)HMNR>xf~aeeSFt`~X6 zDdD`ay$6JIEsv)C$&c7m`4mm{fgx`;aBnYHQir-a=z)I>In8e;TwK4!C81t)MW`L! z9HoVemaeWw2cw74*w|^dna1e5>9gW=8ZFz(zx*uZ;p%|;6!dD@FFtq6*eT=^a_OVu z+u``QKG;kD*vgfHj_nwa>q6``eOqq1TubH^ zyda#@O~QAu%WH<1Pfnm`L5<|rxPCH+azXd~Qgiw91ThJl0P!S#A8InoUFvo5{ktAt1Az;*3?-(`89`**f@@0q`?*L&sz_&(d=ys??a^yons zJCD9sV^uyRPo*QNAJLTbP@EV1nD|nRuZLopeh)-Ux>jr$FN)oR&!Qoo6h97m4INX@ z*Ye#voloE!wv3ksZA*QLMmc)y=#+dCU!^DSH=#FzK6ZKVKYxjB_(bM^Vk-9lcrNjxFU>A6?jKI`_Go)$eQ`Z&HA?&HN@iTTCtPmk5( zULk)Fd;cW#tbH=rJvPg+Hg22~*RNcL4Q7w!t!f^l&he^xIK?;ecXnP5t#)GE-u%xxcq~dQEbZ=iBtJ6K1D3c4v1~}G4UT?_NJ+Omg_m?qx6)*d+dj}jlbGI#E{ly zonkiMK`;1juz_rgI91(>ZKNyE0BC-6e0Fp0J#%Ml=%_L4OXG*tQ^fZ)d2t;dO()>5 z*(Aq7zZ5?%wsY@FHDtd1fH*MtImeBzpk_leG(VE==a?QIpPST@*f#OOF%$f)#^ilI zz-Q=<)?eJ`Om09 zuU;L5J}E{~x43`k+tZ&yyHR^`O!aAq3;q^j1oz01>&Q1~x^fP=gBqP!V&ooj4S9sz zgI?ykaXumTkjr@8`b1jN-4&Lry;Q4n1n(AAM=^m6wFLNG!&m^1*y7KTH$hi}_ssnE&O6 z#bs{|HtO9WK9kFc(ZpZ}hnS2VJ2dE+;&5|e`qP z`s{uh=L9Y2H*rbOhiGTC8QSXfz37){f>rHdGdTe}yWHeH^Yj@$NKPeZpw+Ay^pdqh z5AoAty;whJDKr;44sAvKWV@+3pER6ZW48(Fa?{9B241(E<>hPoBI!2rHyRP&@SNBp z&}eRCwdSaPrV*P?;U^BfaU3O6XSXy2KAaczR%EJSVoJ zAFu^#1&%4$1N}Sy4ewjdXMEoJd@q|Mo@1+QpLs2|S&n8uoCmQkJwt3NR;6(`K91Md z<14{d(!%vX@NIk|zb4P&JLQP_8~85w-yMEy_xgwVUHK#*tjAb>DIVdw*lPVMViWnN zx&!}Bn;G+EXf^!%=snkq&abik@V6L1tzRx{S)btp%@gDs#69K)@)2?|-^EwRSA0JW z$hL?V^kPk8I|t?gs5NdE_JI%PYxraPMnl>q_-gx2OL7eOaL0sC=a1A6`Fj3eFPxab zF%*x85vCEJVE-6#idaIOa>Oe7)#O`HuY~h+NXXm8G5-ni$#~sz@R`AQZM<%s*6sTs zu7T^ow!n6QzqOqM@zwZyI?l4q*6TZLk2jFy5ol!Sm_&T&qWfxJ;Zz zyHK}tUs@UlP5GPQT4y_4_iO??U@Wjh@-;c1abmo%U+mcQ_^+%sGxm)&^bm1BKdkn|4-Y+r@16F$bgiMq(-((6qJAM4<7{&ZTL6^r{*Jpp_NpDJ%#9ItzSe(=3~jNFAUmh1gC_-6M^;HTw)YMg31 z(|FB$mboDKD4!X4KJdQfrq^e?N4-c*O%2t4xzDlu)4tMD?YG*Wy6Wg{R6}xH)L6wL z>Tr&g$7#eXI3gTRag2C|zob3UqWDe6orcSQii5;M?+U#X&e^nPv1@ZB#98z&ao0;j z3^wrG>)s!li0eQsW<6psx(H2#pLRXK-@a?Fu&mGc9_thDx&CNf{I}SXKjORS4{Fw~ zb3Ttw!AJ0i{38D-PUb5Io_k&FY~094X(aro@k8$bU+U3ov@^?Dr}4%I^Y8pMzr*Jk z!)yba!bb8j>?u3Tm$1!S2OlEu<3rea^=p5tS+n(WB)JkB&Th-2_$BsPUdT4H%WN`V zCP!nNJ$lYw^KEodcAbsq)7bvu&#d3J@F~7;FnkRE!hhKAQE!l=_-^^1{bMKDR5q9` zWYgrGY?!>oF;NF&$M|Kn36BH4EwIdByf&7nQ?h?wTkO7-BL{L$M-DV>ADxQMAZ~D7 zh#}ZOInmKyl*-CI353qbFI_5>54SY zv5kC!?T{OKgxADp#v3dPpA$n`hkaEmwO;?knT~~+({Zug(|FDPSjMvcO>dWYd}xEy zd8_Ri$I!832S&~+rk9^Co}21-0pB)q)RCvMC)0nA=U&73%vgs!7oIQvxcIu=*Lpht zmFKcye5#ro{r;Ij_f>!6Tj{`ZTROkogkoU5i(|FB$mb3gW z6WG>)b@JU{w+G`j?@xcPbyye1_o`9QwfXwR^X*~h)imVX^j7{wt|kUhmlp>NJO@7n z%UIUu_$2G&Kll>g#b1as?5l0Et&SDn;TSrW>Nk!n-IqVnU*mW??hxBJ4{B`AiG0!d z*<^zArC-Cj(>EiRbUwvN&avypH6ibj+qe$JRWudXsqtiN8#Bhs7}Lgw{sQBaJu#N` z80g88ck)5}8UMup=;7vr7oX#`V3gt@TNj8#5($jd-`T%7Tq<)$@D>#|Nh!goyj6We85hmWvd z_ED@Xr{QXL75udrg&p7%`9Z#!{~Lap&tpf}6WR+OM@wOI*d8_D>3oeHW7pIu**$iU zO=R=gZ1$2iI&w0$m8~899P$Erg1kXaL2ILNj$F(=(x!8