Théo Gnassounou

[MRG] Add SPDNet

12

Hi, I started implementing SPDNet in Braindecode, as discussed in the issue [#453](https://github.com/braindecode/braindecode/issues/453). It is just a draft, and the model is not working. I added an example if someone...

[WIP] Add len_last_layer to USleep model as other sleep stager model

1

I added the attribut len_last_layer to USleep model like the other sleep stager models. @agramfort

Problem to access ASSISTments dataset

1

[Dataset] ASSISTments dataset [Question/Issue] When using `get_dataset("assistments")`, I get an error (see screenshot). One dataset is downloaded, but not the proper one, I presume. Its name is skillbuilder-data-2009-2010.zip. I replicated...

Make deep method work with `make_da_pipeline`

1

Deep methods can not be put in a `make_da_pipeline` since the pipeline detects it like an estimator, which should not get access to target data.

In the DeepCoral method, the class `DeepCoralLoss` computes first the covariance matrices and then calls the `deep_coral_loss` function with cov as arguments. I would like to have to covariance estimation...

Put a parameters criterion to deep models function

1

The function to create deep DA models should have a parameter `criterion` or `base_criterion` or `source_criterion`. For now, it is hardcoded as a cross-entropy loss but if the model has...

 Hi, I played a bit with the logo. Let me know what you think about it. Maybe I will still change a bit the colors

We need to provide way better example of the deep methods. For each methods we should provide a training example (which is already done) and then an explanation of what...

Deep methods do not currently support `device='gpu'`. It should break at several locations when we define a new tensor or a `domain_classifier` for the adversarial methods. This issue should not...

During the validation step of the training loop, we should think about the strategy for the dataloading. Should we use the same as the training set? Should we provide a...