GitHub - ThuryW/ResNet20-for-Cifar10: ResNet20 implement for Cifar10 dataset

ResNet20 for Cifar10

Just run

python train.py # --lr 0.001 ...
python test.py  # --path ./checkpoint/yourfilename

You can add args if you wanna change hyperparameters, see train.py.

Use -r to resume from checkpoint.

Save the model weights in csv files

python export_weights.py

Name		Name	Last commit message	Last commit date
Latest commit History 19 Commits
autorun		autorun
hook/conv_analysis		hook/conv_analysis
model		model
scripts		scripts
.gitignore		.gitignore
README.md		README.md
prune_and_finetune.py		prune_and_finetune.py
prune_and_finetune_auto.py		prune_and_finetune_auto.py
test.py		test.py
test_unmerged.py		test_unmerged.py
train.py		train.py
usp_and_finetune.py		usp_and_finetune.py
utils.py		utils.py