You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Currently, whenever switching to algo3, we have to modify the FACTORS and CODEVECTORS configs to account for the extra ID factor. We should be able to let the dataset class do it.
Note we should not change VSA class to account for algo since algo is not an attribute of VSA. The only place we can do this is the dataset class.
A better solution may be extending the VSA class for algo3, like how we do it in the multi-concept-MNIST project. Let's see if we can get around it without doing so first.
The text was updated successfully, but these errors were encountered:
Currently, whenever switching to algo3, we have to modify the
FACTORS
andCODEVECTORS
configs to account for the extra ID factor. We should be able to let the dataset class do it.Note we should not change VSA class to account for algo since algo is not an attribute of VSA. The only place we can do this is the dataset class.
A better solution may be extending the VSA class for algo3, like how we do it in the multi-concept-MNIST project. Let's see if we can get around it without doing so first.
The text was updated successfully, but these errors were encountered: