locuslab/TCN

Any preprocessing required for the model?

deancorleone00 opened this issue · 3 comments

I used zero padding to make records in my batch share the same length. Is there any other preprocessing required in order to use the model? (scaling, etc.)

There shouldn't be any on the model side. Depending on the data, you may want to normalize and scale the data before passing them into the TCN model.

Thanks! Do you mean linearly scale (min-max normalize) or standardize (z-score normalize)? Or I should try both?

No harm to try both, but I would go with the latter first.