This project provides an initial framework to start for other deep learning projects.

What is needed to use it?

*) add necessary configuration parameters as you like.

*) add dataset loader in nn/data_loader.py, and imports them in utils/configs.py

*) inherit nn.Predictor and writer a custom trainer/tester function in nn/XXX_predictor.py

*) add network structure in nn/model.py

*) add your result logger in utils/train_callbacks.py

Notices:

This is a very raw version to brew the deep learning based networks, and it cannot run as some key functions are left as blank.