A Machine Learning engine for MID music files, based on PyTorch framework.
Project description
Ennio-Learning
Ennio is a machine learning program. It takes musics in MID format to learn a theme. Themes are collections of datasets (collection of MID files) having common features, for example :
- Heroic fantasy
- Thriller
- Horror
- Suspense
- Classic
- Manga
- etc.
Once learning (training) for one theme is done, the model (network weights) is serialized and can be reused for later music generation. Such models are used for instance in Ennio app project.
Training features
Ennio Learning handles Jordan features during training and model evaluation. As these are long processes, Jordan app gives remote access to logs (and ETA), and some control over loops. For example, one may skip a training task if it is not going to be efficient (loss is not decreasing).
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Hashes for enniolearning-0.1.2-py3-none-any.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | ca9bf97ac45872de168d1173df2f5706c9fc2f20862e7f1251d1ce7de045f24f |
|
MD5 | 709de9b7d8410db6333a19267e4f7e27 |
|
BLAKE2b-256 | 2829203a3148e90bf073f703751fc5a866b46de911abd25978f1e8f0caba9ae0 |