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
The handling of 1d input should be consistent in all estimators. I think currently 1d X is assumed to be of shape (1, n_features) - and even that might fail sometimes. That needs to be checked and needs to be consistent across all estimators.
Also, all estimators that don't support nd-input (I'm not sure any do or should do?) need to raise an appropriate error.
The text was updated successfully, but these errors were encountered:
Also compare #1549. I think that asks to interpret 1d input the other way around. I guess we should find out what estimators do atm before changing any behavior.
The handling of 1d input should be consistent in all estimators. I think currently 1d X is assumed to be of shape
(1, n_features)
- and even that might fail sometimes. That needs to be checked and needs to be consistent across all estimators.Also, all estimators that don't support nd-input (I'm not sure any do or should do?) need to raise an appropriate error.
The text was updated successfully, but these errors were encountered: