Order of parameters for TargetEncoder in Flow UI is not optimal

[exalate-issue-sync]

First user is being asked to select ignored_columns ( from empty set) and only then comes training_frame parameter

[exalate-issue-sync]

Andrey Spiridonov commented: This could be seen as part of a bigger thing ( not sure if it is a design or something we want to improve/fix). For auto generation of estimators' parameters names on client sides we need to specify {{fields}} on backend (e.g. {{GBMParametersV3}}) but those parameters are not shared. Probably it is done to have flexibility of specifying order on per algo basis just by hardcoding it in the list of fields.

[h2o-ops]

JIRA Issue Migration Info

Jira Issue: PUBDEV-7358
Assignee: Andrey Spiridonov
Reporter: Andrey Spiridonov
State: Resolved
Fix Version: 3.28.1.3
Attachments: N/A
Development PRs: Available

Linked PRs from JIRA

#4387