Pipe fold_column and weights_column through the algos in AutoML

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

6.92k stars 2k forks source link

Pipe fold_column and weights_column through the algos in AutoML #11309

Closed exalate-issue-sync[bot] closed 1 year ago

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

In AutoML, we need to pipe through the fold_column and weights_column arguments and expose them at the user/client level.

The fold_column argument is useful for pooled-repeated measures data (to avoid leakage across CV folds)
Weights column is also useful

This is not part of the AutoML REST API, so it will have to be added and piped through on the backend before we can add it to R/Py. Once this is done on the backend, we can create a new ticket to add this to the R and Python APIs.

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

Navdeep commented: Current blocker is referenced above. Essentially, fold assignment is required for stacked ensembles, but cannot be used when fold_column is set. This constraint needs to be relaxed.

h2o-ops commented 1 year ago

JIRA Issue Migration Info

Jira Issue: PUBDEV-4423 Assignee: Raymond Peck Reporter: Erin LeDell State: Closed Fix Version: 3.10.5.1 Attachments: N/A Development PRs: Available

Linked PRs from JIRA

https://github.com/h2oai/h2o-3/pull/1155 https://github.com/h2oai/h2o-3/pull/1159