2.training&testing.md

Requirements

Environment

1. Python 3.6.*
2. CUDA 11.1
3. PyTorch
4. TorchVision

Install

Create a virtual environment and activate it.

conda create -n grew python=3.6.12
conda activate grew

The code has been tested with PyTorch 1.0 and Cuda 11.1.

conda install pytorch=1.0.0 torchvision=0.2.1
conda install matplotlib tqdm
conda install tensorboard tensorboardX
conda install scipy scikit-image opencv

Train

Train a model by

python train.py

• --cache if set as TRUE all the training data will be loaded at once before the training start. This will accelerate the training. Note that if this arg is set as FALSE, samples will NOT be kept in the memory even if they have been used in the former iterations. #Default: TRUE

Test

python build_submission.py

Participants must package the submission.csv for submission using zip xxx.zip $CSV_PATH and then upload it to codebench.(Due to the close of the original Codalab website, we have moved the benchmark to this Codabench site. For the old leaderboard of GREW benchmark, please refer to codalab.)