H2O is an Open Source, Distributed, Fast & Scalable Machine Learning Platform: Deep Learning, Gradient Boosting (GBM) & XGBoost, Random Forest, Generalized Linear Modeling (GLM with Elastic Net), K-Means, PCA, Generalized Additive Models (GAM), RuleFit, Support Vector Machine (SVM), Stacked Ensembles, Automatic Machine Learning (AutoML), etc.
Currently, in AutoML, we exploit the fact that you can pass a list of alphas to the alpha parameter in GLM. This results in a GLM where you can't tell which alpha is used and returns a single GLM with the "best" alpha. Let's move alpha to the GLM grid so that it returns six GLMs, one associated with each alpha that we try (0.0, 0.2, 0.4, 0.6, 0.8, 1.0).
We should definitely benchmark this to see that it doesn't hurt performance too much to have the six GLMs in the ensemble vs one GLM. If not, then we should include all six.
Currently, in AutoML, we exploit the fact that you can pass a list of alphas to the alpha parameter in GLM. This results in a GLM where you can't tell which alpha is used and returns a single GLM with the "best" alpha. Let's move alpha to the GLM grid so that it returns six GLMs, one associated with each alpha that we try (0.0, 0.2, 0.4, 0.6, 0.8, 1.0).
We should definitely benchmark this to see that it doesn't hurt performance too much to have the six GLMs in the ensemble vs one GLM. If not, then we should include all six.