Zero2Neuro Roadmap
Our priorities for continued development include:
- The ability for users to provide plug-in code to implement custom functionality inside of Zero2Neuro, including:
- Loading custom file formats
- Building custom networks
- Preprocessing of data after the training, validation, and test data sets have been constructed
- Custom fold-to-data-set translation
-
Splitting data randomly with stratification
-
UNet schema (connection to UNet 3+)
-
Deep neural network clustering schema
-
Support for Scikit-Learn models
- Orthogonalized cross-validation