Examples#

Runnable, commented examples for PyAMICA.

Run on your own data#

For EEG data, start from 01_mne_integration.py to fit AMICA on an MNE raw object.

If your data is already a NumPy array with shape (n_channels, n_samples), use 02_pure_jax_fitting.py.

Installation#

From a local clone: pip install -e ".[all]"

JAX uses an available GPU automatically when the compatible GPU package is installed; otherwise it runs on CPU.

AMICA via MNE-Python

AMICA via MNE-Python

Pure-JAX AMICA on a NumPy array

Pure-JAX AMICA on a NumPy array

Gallery generated by Sphinx-Gallery