Is there Deep Learning Methods in MNE-Python

✉️ Mailing List Archive (read-only)
1

External Email - Use Caution

I want to try the deep learning methods on my EEG-Data.
I have a question regarding Deep Learning methods (convolutional,
recurrent, and LSTM) to classify EEG-Data. Is there libraries in
MNE-Python? If not, then which libraries would you suggest? Also when
classifying using CSP (Common Spatial Patterns) the data should be epoched
in equal windows. Is it the same case in Deep Learning? Or is there another
way preprocess the data?
Lastly is it possible to classify CSP components using deep learning
methods?
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.nmr.mgh.harvard.edu/pipermail/mne_analysis/attachments/20190223/76b7d8e7/attachment.html

2

External Email - Use Caution

hi,

you don't have any deep model in mne-python directly but it's used
by many recent papers using deep learning with EEG.

one of these packages is: https://robintibor.github.io/braindecode/
for BCI mostly
but you can find more on github:
https://github.com/search?q=mne+tensorflow&type=Code
https://github.com/search?l=Python&q=mne+torch&type=Code

See our recent review paper on EEG and deep learning :
https://arxiv.org/abs/1901.05498v1

HTH
Alex