Skip to content

Add the specification of the behavior of n_startup_trials for constant_liar option #3908

@not522

Description

@not522

The n_startup_trials' document says, "The random sampling is used instead of the TPE algorithm until the given number of trials finish in the same study.", but it is inconsistent with group=True or constant_liar=True. We need to specify the behavior for these options.

Metadata

Metadata

Assignees

No one assigned

    Labels

    needs-discussionIssue/PR which needs discussion.

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions