Add Generalized Additive Models (GAMs) to H2O

Activity

Show:
Wendy
March 19, 2020, 11:08 PM

The following subtasks will make it to the major release in end of march:

  1. PUBDEV-7178: Generate basic GAM Schema and Algorithm

  2. PUBDEV-7180: added modelmetrics

  3. PUBDEV-7181: Python API

  4. PUBDEV-7182: R API

 

Wendy
January 6, 2020, 6:42 PM
Edited

After digging around the literature, I have come up with the following plan and subtasks:

  1. Generate basic GAM Schema and Algorithm supporting regression (PUBDEV-7178);

  2. Generate GAM dataset and implement Java unit test (PUBDEV-7178);

  3. Generate ModelMetric and implement model metrics calculation (PUBDEV-7180);

  4. Generate Python API o support GAM and implement Python unit test (PUBDEV-7181);

  5. Generate R API to support GAM and implement R unit test (PUBDEV-7182);

  6. Generate Python API to support GAM model metrics and implement Python unit test (PUBDEV-7183);

  7. Generate R API to support GAM model metrics and implement R unit test (PUBDEV-7184);

  8. Generate GAM Mojo (PUBDEV-7185);

Note that task 4 and 5 can start as soon as 1 is completed. Tasks 6, 7 can start as soon as task 3 is completed. Task 8 can start as soon as task 1 is completed.

In addition, we will want to do the following as well:

  1. Add support for more smoothers;

  2. Add methods for choosing knots;

  3. Add support for other families and links: binomial, multinomial, gamma, ….. Check with R and see what families they support in mgcv toolbox.

I will add the appropriate JIRAs as we go along.

 

Fixed

Assignee

Wendy

Fix versions

Reporter

Nidhi Mehta

Support ticket URL