Tutorials

The notebooks below are self-contained and use simulated signals, so they can be run without an EEG recording or private stimulus files. Figures generated by the notebook cells are rendered directly in the HTML documentation.

TRF tutorials

Two complementary tutorials cover Temporal Response Function (TRF) estimation. Both are self-contained and use simulated signals.

A gentle introduction to TRFs (TRF_simulation_tutorial) — a minimal walkthrough: simulate two stimulus features, fit the simplest TRF, and visualise the result. Best starting point.
TRF estimation: inference, regularisation, and solvers (TRF_advanced_tutorial) — an in-depth tutorial covering statistical inference (permutation tests, bootstrap CIs), banded regularisation, cross-validation, robust fitting, solver comparison, and multi-channel TRFs.

Loading Word-level features

from pyeeg.io import AlignedSpeech
speech = AlignedSpeech(onset=0., srate=125, path_audio='path/to/audio/file')

Loading Word vectors

from pyeeg.io import WordLevelFeatures
wlf = wfeats = WordLevelFeatures(path_praat_env=dur_path, path_wordonsets=surp_path, path_surprisal=surp_path)

Loading processed EEG (processed by EEGLAB)

To load EEG data from .set files, the code would be:

from pyeeg.io import eeglab2mne
eeg = eeglab2mne(filepath)

Older, data-dependent examples

These examples are retained for now. They require local EEG/audio files and may need adaptation to current APIs.