Installation
The main dependencies are tensorflow/tensorflow-gpu, numpy, scipy, scikit-learn, matplotlib, pyedflib, tqdm.
An easy way to install all dependencies is to work from a conda env. Create the environment from the provided environment file.
conda env create -f environment.yml
Obtaining data
Please follow instructions from the SHHS website.
Train
-
Modify CONFIG settings in train.py
-
Run train.py and specify your options. To reproduce the model described in the paper, use:
python train.py -training_batches 300000 -learning_rates 0.00003 -featuremap_sizes 128 128 128 128 128 128 256 256 256 256 256 256 -filter_sizes 7 7 7 7 7 7 7 5 5 5 3 3 -strides 2 2 2 2 2 2 2 2 2 2 2 2 --eps_before=2 --balance_classes=False --conv_type=std --batch_norm=False --filter=False --hiddenlayer_size=256
Visualize
To visualize synthetic inputs as described in the paper:
-
Modify CONFIG settings in visualize.py
-
Run:
python visualize.py
Comparison
jupyter notebook comparison.ipynb