Throw an error if certain parameters are no supported in selected algo

Description

When defining models it would be good to throw an error if a user sets parameters that are not available for that algorithm. We had several (a lot) of those instances, e.g. setting "balance classes" in glm or people (me) not reading the manual on weight columns.

Atm H2O takes silently everything that you pass.

Assignee

New H2O Bugs

Fix versions

None

Reporter

Avkash Chauhan

Support ticket URL

Labels

None

Affected Spark version

None

Customer Request Type

None

Task progress

None

ReleaseNotesHidden

None

CustomerVisible

No

AffectedCustomers

Priority

Major
Configure