Stacking in machine learning π½ 1/4 
- Divide the training data into folds.
- Train a bunch of models: M1, M2.....Mn.
- Create full training predictions (using out of fold training) and test predictions using all these models. 2/4
- Train a bunch of models: M1, M2.....Mn.
- Create full training predictions (using out of fold training) and test predictions using all these models. 2/4
- Till here it is Level β 1 (L1).
- Use the fold predictions from these models as features to another model.
This is now a Level β 2 (L2) model.
- Use the same folds as before to train this L2 model. 3/4
- Use the fold predictions from these models as features to another model.
This is now a Level β 2 (L2) model.
- Use the same folds as before to train this L2 model. 3/4
- Now create OOF (out of fold) predictions on the training set and the test set.
- Now you have L2 predictions for training data and also the final test set predictions. 4/4
- Now you have L2 predictions for training data and also the final test set predictions. 4/4
β’ β’ β’
Missing some Tweet in this thread? You can try to
force a refresh
γ
