DeepSleepNet_Implementation_in_Keras

original paper : https://arxiv.org/abs/1703.04046

original implementation by the authors : https://github.com/akaraspt/deepsleepnet

I followed implementation of the deepsleepnet in tensorflow into keras (actually keras in tensorflow).

Requirements

conda create -n deepsleepnet_keras environment --file requirements.txt

Data Preparation

Download and getting NPZ file is totally same with original implementation from https://github.com/akaraspt/deepsleepnet .

To get 20 fold cross validation data:

python data_preparation.py

Training

python trainer.py

The trained model (both featurenet and deepsleepnet) will be stored at the ./weights as weights.

Performance Evaluation

cat performance.txt

To Do

• Revise 'data preparation' to split patients. For example, in test set, patients 0 - 10, valid set patients 11 - 12, test set 13 - 14.