/CheXpert_Baseline_PyTorch

A simple baseline model for CheXpert - DenseNet121. Contain train and test module. Also a simple and easy-used deep learning framework by PyTorch!

Primary LanguagePython

CheXpert Baseline PyTorch

0. Update remark

  • 2019.08.07

I am trying to repeat the CheXpert baseline model and doing some research on Pneumonia. Welcome anyone interested in finishing this work to join the project and train and improve the model together!.

  • 2019.08.08

The baseline model have been trained. The result is almost same as the paper said.

  • 2020.05.29

Note: The project is my first deep learning project by Torch. Maybe it is naive, but it would be helpful if you are also new to PyTorch! If so, I also recomment my new project Vision Research Toolkit by PyTorch, which will be a powful tool in your way of learning PyTorch

1. How to run the code.

-> run 'main.py'

-> select mission

2. Notation

  • The model is trained through GTX 1060. if you want to run the code with cpu, you should set 'use_gpu = False' in 'config.py' and if you want to load checkpoint to the model stored in cpu, you should change the code in 'main.py' line 167
  • Don't forget to change the 'data_root' in 'config.py' to satisfy the root location where you store CheXpert data.