/basemodel-pytorch

mainly from https://github.com/kuangliu/pytorch-cifar and some code files from orig paper added

Primary LanguagePythonMIT LicenseMIT

basemodel-pytorch

Some classification mathods.

Request

python3.8.13, PyTorch1.7.1

How to use

  1. prepare data in which a folder with images acts as a class.

  2. set configs (dataset path, classifier, etc.) in data/config.py.

  3. set input size and modify the corresponding input shape of fc layer.

  4. start training

python train.py

TODO:

move step 3 above into config.py

move preprocess part(data augmentation) from train.py into config.py

add valid.py

finish README.md

ref

https://github.com/kuangliu/pytorch-cifar