Rename Dataset Classes

[robintibor]

I would suggest to rename the dataset classes as follows:

BaseDataset -> EEGRawDataset
BaseConcatDataset -> EEGConcatDataset

Think this is more straightforward, Base has been confusing for me, and then structure is:
EEGConcatDataset is a concatenation of either EEGRawDataset or EEGWindowsDataset

What do you think @bruAristimunha ? Assuming #515 is merged.

[bruAristimunha]

Okay, I'm thinking here @robintibor. I'm not confident about this nomenclature. I would remove EEG from the beginning, we also have support for ecog and any mne format (ieeg, meg...). It would be shorter, which would be great.

[robintibor]

Hm, atm though this is also in other parts of Braindecode, namely EEGClassifier, EEGRegressor.
And if we strip away entirely, it might become a bit generic for ConcatDataset.

Like ConcatDatasetalso exists in pytorch https://pytorch.org/docs/stable/data.html#torch.utils.data.ConcatDataset (and we are indeed subclassing that one).
So at least we need to have another name for that I think? BDConcatDataset might be a bit ugly, BraindecodeConcatDataset a bit long? BraindecodeDatasetmight be an alternative, but seems a bit like mysterious to me. Kind of like the fact that currently, one may be able to guess from the class name that it is a concatenation of one of the other datasets (raw or windows).

[bruAristimunha]

How about just Brain in the prefix?

[bruAristimunha]

• BaseDataset -> BrainRawDataset
• BaseConcatDataset -> BrainConcatDataset

Looks nice for me, what do you think @robintibor?

[robintibor]

So for me a BrainDataset implies a dataset of the Brain, e.g. an anatomical atlas/coordinates etc.
For me what we have is nto a dataset of the Brain but a dataset of BrainSignals, yet BrainSignalRawDataset, BrainSignalConcatDataset seems a bit long? Not sure, let's discuss a bit more? What do you think @agramfort ?

[agramfort]

you have now EEGClassifier and EEGRegressor. I would try to be consistent so EEGDataset so an EEGClassifier will be fed by an EEGDataset.

[bruAristimunha]

I'm ok with the first proposal @robintibor and @agramfort

[bruAristimunha]

Should we focus on this? @robintibor