search

NeuroTechX / moabb

Mother of All BCI Benchmarks
https://neurotechx.github.io/moabb/
BSD 3-Clause "New" or "Revised" License
672 stars 175 forks source link

Parametrisation of n_component in CCA based SSVEP classification #360

Open emmanuelkalunga opened 1 year ago

emmanuelkalunga commented 1 year ago

Hi @sylvchev, In the the current implementation of the SSVEP CCA classification, CCA classifiers are instantiated with the number of CCA components fixed to one i.e. n_components = 1.
The number of components does affect the performance of CCA-based classifiers significantly. I would like to change that in all CCA classifiers (present and futures) so that n_components can be passed as a parameter. What do you think?

P.S. I am referring to the implementations in moabb/moabb/pipelines/classification.py

sylvchev commented 1 year ago

Ok, definitely a must. We could include it in the PR #359 We need to add an admonition in the docstring to indicate this change, like the example below:

    Notes
    -----
    .. versionadded:: 0.4.7
emmanuelkalunga commented 1 year ago

Ok. I will add it to the PR