search

h2oai / h2o-3

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.
http://h2o.ai
Apache License 2.0
6.9k stars 2k forks source link

Clarify default options for GLM #14845

Open exalate-issue-sync[bot] opened 1 year ago

exalate-issue-sync[bot] commented 1 year ago

The GLM interface in Flow has a default alpha of 0.5 and lambda as a data-guided value. This is not clear to the user, so it appears as if a non-regularized model is being run unless some values are specified (unless the user looks into the model summary after running one).

Special values in deep learning might be one option. Added description might be another option, though it becomes difficult since default lambda is slightly different with lambda search than not--though ultimately H2O will decide in both cases.

It might also be useful to make it clear lambda 0 is non-regularized.

exalate-issue-sync[bot] commented 1 year ago

Mark Landry commented: Solver has an impact in the default behavior as well, which can seem confusing (default alpha for LBFGS is 0). Again, merely being transparent will help the users understand what they are doing.

DinukaH2O commented 1 year ago

JIRA Issue Migration Info

Jira Issue: PUBDEV-1982 Assignee: Tomas Nykodym Reporter: Mark Landry State: Open Fix Version: N/A Attachments: N/A Development PRs: N/A