search

nkartik94 / Multi-Label-Text-Classification

Kaggle Toxic Comments Challenge
107 stars 55 forks source link

One Vs Rest classifier #1

Closed NGBoost closed 5 years ago

NGBoost commented 6 years ago

Is it correct to fit an OneVsRestClassifier using''for category in categories:''? I think that this is not correct in multi label case. Instead, OneVsRestClassifier should be fitted in the whole train and test set. Am I correct?

nkartik94 commented 5 years ago

The basic concept of oneVsRest is to fit it that way, its mentioned in its documentation itself and not fitting on the whole data.