Metadata-Version: 2.1
Name: back
Version: 0.0.1
Summary: Backbone for PyTorch training loop
Home-page: https://github.com/EvgenyKashin/backbone
Author: Evgeny Kashin
Author-email: kashinevge@gmail.com
License: MIT
Keywords: pytorch,deep-learning,training-loop,backbone
Platform: UNKNOWN
Requires-Python: >=3.6.0
Description-Content-Type: text/markdown
Requires-Dist: torch (==1.1.0)
Requires-Dist: torchvision (==0.2.2)
Requires-Dist: tb-nightly (==1.14.0a20190522)
Requires-Dist: pandas (==0.24.2)
Requires-Dist: pillow (==5.4.1)
Requires-Dist: tqdm (==4.31.1)
Requires-Dist: jpeg4py (==0.1.4)
Requires-Dist: scikit-learn (==0.21.1)
Requires-Dist: albumentations (==0.2.3)
Requires-Dist: future (==0.17.1)

## Backbone for PyTorch training loop
Work in progress

### SRM 
[Implementation](cifar10_train.py) of "SRM : A Style-based Recalibration Module for Convolutional Neural Networks"

TODO:
- [ ] Cifar10 benchmark
- [ ] ImageNet benchmark


