Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

FLUX.1 fine-tuning, dataset_1024_bs1.toml not found #1832

Open
huangjun12 opened this issue Dec 12, 2024 · 1 comment
Open

FLUX.1 fine-tuning, dataset_1024_bs1.toml not found #1832

huangjun12 opened this issue Dec 12, 2024 · 1 comment

Comments

@huangjun12
Copy link

FLUX.1 fine-tuning

accelerate launch  --mixed_precision bf16 --num_cpu_threads_per_process 1 flux_train.py   
--pretrained_model_name_or_path flux1-dev.safetensors  --clip_l clip_l.safetensors --t5xxl t5xxl_fp16.safetensors --ae ae_dev.safetensors 
--save_model_as safetensors --sdpa --persistent_data_loader_workers --max_data_loader_n_workers 2 
--seed 42 --gradient_checkpointing --mixed_precision bf16 --save_precision bf16 
--dataset_config dataset_1024_bs1.toml  --output_dir path/to/output/dir --output_name output-name 
--learning_rate 5e-5 --max_train_epochs 4  --sdpa --highvram --cache_text_encoder_outputs_to_disk --cache_latents_to_disk --save_every_n_epochs 1 
--optimizer_type adafactor --optimizer_args "relative_step=False" "scale_parameter=False" "warmup_init=False" 
--lr_scheduler constant_with_warmup --max_grad_norm 0.0 
--timestep_sampling shift --discrete_flow_shift 3.1582 --model_prediction_type raw --guidance_scale 1.0 
--fused_backward_pass  --blocks_to_swap 8 --full_bf16 

error log

  File "/picassox/ma-sfs-bj-training/hj-data/Debug-Train/sd-scripts/sd-scripts/library/config_util.py", line 651, in load_user_config
    raise ValueError(f"file not found / ファイルが見つかりません: {file}")
ValueError: file not found / ファイルが見つかりません: dataset_1024_bs1.toml

Is there any example of dataset_1024_bs1.toml

@phyllispeng123
Copy link

you have to generate yourself , the .toml file should be like this:

[general]
flip_aug = true
color_aug = false

[[datasets]]
enable_bucket = true
resolution = [1024,1024]
bucket_reso_steps = 64
max_bucket_reso = 2048
min_bucket_reso = 128
bucket_no_upscale = false
batch_size = 1
random_crop = false
shuffle_caption = false

[[datasets.subsets]]
image_dir = "your image folder containing image + caption (img and txt file sharing the same name)"
num_repeats = 1
caption_extension = ".txt"

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants