Requests/voting system for preprocessed datasets #197
Description
Hey, I proposed this idea to @bruAristimunha, I'm putting it here before we forget.
With the EEGDash Storage that will host preprocessed datasets, the possible combinations of preprocessing steps to apply are infinite, but storage will be limited at some point.
- A possibility would be to define a standard preprocessing pipeline and always apply the same, but it's a bit rigid...
- Another possibility would be to implement a requests/voting system, such that users of the library could ask for exactly what they want, in a community-driven manner. It's already possible through various channels, such as GitHub issues, but a centralized system would make it more systematic!
The best solution is probably to have both, a default preprocessing for each dataset, plus the possibility to easily take requests
On top of this, it would be relevant to keep track of the number of downloads for each preprocessed dataset. 1) the statistics would be interesting, and 2) to be able to detect which ones are not used and could be deleted to free storage space.