3 Comments

Great insight, as always ! Thanks Avi Chawla !

Expand full comment

Good content, I remember working on a project where cross-validation and retraining played an important role to finalise the model for an imbalanced class dataset classification prioblem.

Expand full comment

Refitting the model on the full dataset is the default behavior in scikit-learn as well (see GridSearchCV).

Expand full comment