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 AutoML doesn’t expose {{offset_column}} because DRF doesn’t support it (see PUBDEV-5341).
Once supported by Stacked Ensemble (PUBDEV-4915), we can expose {{offset_column}} to client API and decide how it should behave with algos like DRF and XRT:
we could skip those algos.
we could emit a warning (using {{eventLog}}) when the param is set, before training of DRF/XRT models: those models would still be trained ignoring the offset, but user can decide to ignore those models and/or restart the AutoML training without those algos.
Currently AutoML doesn’t expose {{offset_column}} because DRF doesn’t support it (see PUBDEV-5341).
Once supported by Stacked Ensemble (PUBDEV-4915), we can expose {{offset_column}} to client API and decide how it should behave with algos like DRF and XRT:
we could skip those algos.
we could emit a warning (using {{eventLog}}) when the param is set, before training of DRF/XRT models: those models would still be trained ignoring the offset, but user can decide to ignore those models and/or restart the AutoML training without those algos.