robintibor

Created an issue in skorch repository at https://github.com/skorch-dev/skorch/issues/785

We still use MOABB in quite some tests (dataset/windower tests) but do we consider this a problem? to me seems ok or at least not high priority to remove, what...

We could still see if we could remove moabb dataloading from tests and replace it with fake datasets just to make things more decoupled

Somehow it seems on a fresh install moabb will fail if you do not create ~/mne_data ? this made our CI fail temporarily... is this intended, wasn't it different before?...

great

Uh sorry we missed this so far, will have a look

So is this still open or have we implemented it? @gemeinl

Until it is ported, you can use the old braindecode to run this example https://github.com/TNTLFreiburg/braindecode/

We have high-performing decoding code at https://gist.github.com/robintibor/6a95a85088e651392c1bb4f912d1528e Would be worth it to transform to proper braindecode example. Also @TianhaoFu linked code may be interesting for you

Here are potential sensor positions to use: https://gist.github.com/robintibor/051ca311d12894733a1ea27b0f0babfa Last Positions were used to visualize in https://gist.github.com/robintibor/25a8c8d2cb9a7083ac387f1b1ac2bb04 but I guess waveguard is more precise?