Welcome to the Deep Neural Networks for Neurophysiology (DN3) Toolbox documentation!

This Python package is an effort to bridge the gap between the neuroscience library MNE-Python and the deep learning library PyTorch. This package’s main focus is on minimizing boilerplate code, rapid deployment of known solutions, and increasing the reproducibility of new deep-learning solutions for the analysis of M/EEG data (and may be compatible with other similar data… use at your own risk).

Access to the code can be found at https://github.com/SPOClab-ca/dn3

The image above sketches out the structure of how the different modules of DN3 work together, but if you are new, we recommend starting with the configuration guide.

Guides

• The Configuratron
• Datasets
• Metrics
• Trainables
• Transformations and Preprocessors

Documentation

• Configuratron
• Datasets
• Neural Network Building Blocks
  • Models
  • Layers
• Processes
• Transforms
  • Instance Transforms
  • Batch Transforms
  • Preprocessors

Indices and tables

• Index

• Module Index

• Search Page