├── .gitignore ├── LICENSE ├── README.md ├── __init__.py ├── evaluation.py ├── experiments ├── __init__.py ├── models │ ├── __init__.py │ ├── han16.py │ ├── multilayer.py │ └── singlelayer.py └── settings.py ├── means ├── han16_mean.npy ├── multilayer_mean.npy └── singlelayer_mean.npy ├── metadata ├── classes.pkl ├── irmas_test_meta.csv └── irmas_train_meta.csv ├── preprocessing.py ├── requirements.txt ├── training.py └── weights ├── multilayer └── epoch.128-val_loss.0.866-fbeta.0.782.hdf5 └── singlelayer └── epoch.115-val_loss.0.902-fbeta.0.673.hdf5 /.gitignore: -------------------------------------------------------------------------------- 1 | *.pyc 2 | -------------------------------------------------------------------------------- /LICENSE: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Veleslavia/EUSIPCO2017/HEAD/LICENSE -------------------------------------------------------------------------------- /README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Veleslavia/EUSIPCO2017/HEAD/README.md -------------------------------------------------------------------------------- /__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /evaluation.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Veleslavia/EUSIPCO2017/HEAD/evaluation.py -------------------------------------------------------------------------------- /experiments/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /experiments/models/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /experiments/models/han16.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Veleslavia/EUSIPCO2017/HEAD/experiments/models/han16.py -------------------------------------------------------------------------------- /experiments/models/multilayer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Veleslavia/EUSIPCO2017/HEAD/experiments/models/multilayer.py -------------------------------------------------------------------------------- /experiments/models/singlelayer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Veleslavia/EUSIPCO2017/HEAD/experiments/models/singlelayer.py -------------------------------------------------------------------------------- /experiments/settings.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Veleslavia/EUSIPCO2017/HEAD/experiments/settings.py -------------------------------------------------------------------------------- /means/han16_mean.npy: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Veleslavia/EUSIPCO2017/HEAD/means/han16_mean.npy -------------------------------------------------------------------------------- /means/multilayer_mean.npy: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Veleslavia/EUSIPCO2017/HEAD/means/multilayer_mean.npy -------------------------------------------------------------------------------- /means/singlelayer_mean.npy: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Veleslavia/EUSIPCO2017/HEAD/means/singlelayer_mean.npy -------------------------------------------------------------------------------- /metadata/classes.pkl: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Veleslavia/EUSIPCO2017/HEAD/metadata/classes.pkl -------------------------------------------------------------------------------- /metadata/irmas_test_meta.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Veleslavia/EUSIPCO2017/HEAD/metadata/irmas_test_meta.csv -------------------------------------------------------------------------------- /metadata/irmas_train_meta.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Veleslavia/EUSIPCO2017/HEAD/metadata/irmas_train_meta.csv -------------------------------------------------------------------------------- /preprocessing.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Veleslavia/EUSIPCO2017/HEAD/preprocessing.py -------------------------------------------------------------------------------- /requirements.txt: -------------------------------------------------------------------------------- 1 | keras 2 | librosa 3 | numpy 4 | pandas 5 | sklearn -------------------------------------------------------------------------------- /training.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Veleslavia/EUSIPCO2017/HEAD/training.py -------------------------------------------------------------------------------- /weights/multilayer/epoch.128-val_loss.0.866-fbeta.0.782.hdf5: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Veleslavia/EUSIPCO2017/HEAD/weights/multilayer/epoch.128-val_loss.0.866-fbeta.0.782.hdf5 -------------------------------------------------------------------------------- /weights/singlelayer/epoch.115-val_loss.0.902-fbeta.0.673.hdf5: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Veleslavia/EUSIPCO2017/HEAD/weights/singlelayer/epoch.115-val_loss.0.902-fbeta.0.673.hdf5 --------------------------------------------------------------------------------