Model Stacking

Model stacking is a way of ensembling models, when we want to give different weights to different models in the ensemble.

Model stacking adds a new learning phase that calculates the optimal way of combining the models in the ensemble.

References

Model Ensembling