Code accompanying "A recurrent neural network for classification of unevenly sampled variable stars".
- Code for scores/figures is found in
figures.ipynb
- Autoencoder network architecture is defined in
autoencoder.py
- Experiments for simulated data are found in
autoencoder.main
- Experiments for light curve data are found in
survey_autoencoder.main
- Light curve data is in
./data
- Light curve data is in
- Model weights are saved in
./keras_logs
See mirror at https://gitlab.com/cesium-ml/IrregularTimeSeriesAutoencoderPaper to download the full set of input data.