Skip to content

Latest commit

 

History

History
141 lines (123 loc) · 4.62 KB

Advanced.md

File metadata and controls

141 lines (123 loc) · 4.62 KB
Raw

ResNet50 V1 FP32 inference - Advanced Instructions

This document has advanced instructions for running ResNet50 V1 FP32 inference, which provides more control over the individual parameters that are used. For more information on using /benchmarks/launch_benchmark.py, see the launch benchmark documentation.

Prior to using these instructions, please follow the setup instructions from the model's README and/or the AI Kit documentation to get your environment setup (if running on bare metal) and download the dataset, pretrained model, etc. If you are using AI Kit, please exclude the --docker-image flag from the commands below, since you will be running the the TensorFlow conda environment instead of docker.

Any of the launch_benchmark.py commands below can be run on bare metal by removing the --docker-image arg. Ensure that you have all of the required prerequisites installed in your environment before running without the docker container.

If you are new to docker and are running into issues with the container, see this document for troubleshooting tips.

Once your environment is setup, navigate to the benchmarks directory of the model zoo and set environment variables pointing to the directory for the dataset, pretrained model frozen graph, and an output directory where log files will be written.

# cd to the benchmarks directory in the model zoo
cd benchmarks

export DATASET_DIR=<path to the dataset>
export OUTPUT_DIR=<directory where log files will be written>
export PRETRAINED_MODEL=<path to the frozen graph that you downloaded>

ResNet50 V1 FP32 inference can be run to test accuracy, batch inference, or online inference. Use one of the following examples below, depending on your use case.

  • For accuracy run the following command that uses the DATASET_DIR, a batch size of 100, and the --accuracy-only flag:
python launch_benchmark.py \
  --model-name resnet50 \
  --precision fp32 \
  --mode inference \
  --framework tensorflow \
  --in-graph ${PRETRAINED_MODEL} \
  --data-location ${DATASET_DIR} \
  --output-dir ${OUTPUT_DIR} \
  --accuracy-only \
  --batch-size 100 \
  --socket-id 0 \
  --docker-image intel/intel-optimized-tensorflow:latest
  • For batch inference, use the command below that uses the DATASET_DIR and a batch size of 128.
python launch_benchmark.py \
  --model-name resnet50 \
  --precision fp32 \
  --mode inference \
  --framework tensorflow \
  --in-graph ${PRETRAINED_MODEL} \
  --data-location ${DATASET_DIR} \
  --output-dir ${OUTPUT_DIR} \
  --batch-size 128 \
  --socket-id 0 \
  --docker-image intel/intel-optimized-tensorflow:latest
  • For online inference, use the command below that uses the DATASET_DIR and a batch size of 1.
python launch_benchmark.py \
  --model-name resnet50 \
  --precision fp32 \
  --mode inference \
  --framework tensorflow \
  --in-graph ${PRETRAINED_MODEL} \
  --data-location ${DATASET_DIR} \
  --output-dir ${OUTPUT_DIR} \
  --batch-size 1 \
  --socket-id 0 \
  --docker-image intel/intel-optimized-tensorflow:latest

Example log file snippet when testing accuracy:

...
Iteration time: ... ms
Processed 50000 images. (Top1 accuracy, Top5 accuracy) = (0.7430, 0.9188)
Ran inference with batch size 100
Log location outside container: {--output-dir value}/benchmark_resnet50_inference_fp32_20190104_213452.log

Example log file snippet when testing batch inference:

Inference with dummy data.
Iteration 1: ... sec
Iteration 2: ... sec
Iteration 3: ... sec
...
Iteration 38: ... sec
Iteration 39: ... sec
Iteration 40: ... sec
Average time: ... sec
Batch size = 128
Throughput: ... images/sec
Ran inference with batch size 128
Log location outside container: {--output-dir value}/benchmark_resnet50_inference_fp32_20190104_215655.log

Example log file snippet when testing online inference:

Inference with dummy data.
Iteration 1: ... sec
Iteration 2: ... sec
...
Iteration 38: ... sec
Iteration 39: ... sec
Iteration 40: ... sec
Average time: ... sec
Batch size = 1
Latency: ... ms
Throughput: ... images/sec
Ran inference with batch size 1
Log location outside container: {--output-dir value}/benchmark_resnet50_inference_fp32_20190104_215326.log