Closed HannahKirk closed 2 years ago
Hi,
which sklearn model are you using exactly? Until now, I thought that scikit-learn classifiers either don't have a validation set or create their own from the training data tha tis provided. Therefore, the current intended use would be to omit the validation set when using with scikit learn classifiers.
But maybe this assumption was wrong?
I am using the ConfidenceEnhancedSVM.
That makes sense. I guess this validation set is only useful for early stopping in the train scheduler of the transformers model. Might be worthwhile adding to documentation so its clear fit() of a non-transformers model doesnt take a validation set. Thank you!
Hi,
I'm initializing an active learner for an Sklearn model with specific validation indices. Minimal code example is:
The error I am getting is
fit() got an unexpected keyword argument 'validation_set'
. Digging into the code, it seems like if you passx_indices_validation
as notNone
this shouldn't happen.Do you have any suggestions?