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
Our current time-based splitter only returns one split of the data. If users want multiple time-based splits, they would need to make multiple calls. We could definitely write a short function that wrap ours, but I'll just also share this as another option. It would require the data to already be sorted by time https://scikit-learn.org/stable/modules/generated/sklearn.model_selection.TimeSeriesSplit.html
The text was updated successfully, but these errors were encountered:
Yes this specific splitter would ignore the split size arguments. Not sure if that would be confusing to the user but it would be consistent with the function call for TimeSeriesSplit()
Following up on this after some time - now that astartes is used in a few different places, we see that it is easier to require the end users to make repeat calls rather than for us to implement it here. Going to go ahead and close this, but leave it as not completed since this may change in the future.
Our current time-based splitter only returns one split of the data. If users want multiple time-based splits, they would need to make multiple calls. We could definitely write a short function that wrap ours, but I'll just also share this as another option. It would require the data to already be sorted by time
https://scikit-learn.org/stable/modules/generated/sklearn.model_selection.TimeSeriesSplit.html
The text was updated successfully, but these errors were encountered: