Skip to content

How to split the meta-test and test datasets? #6

Description

@jm-xiong

What is the rules or pattern for the data partitioning used in the meta-test and evaluation phases? Why the RealSnow dataset contains 65,000 pairs for meta-test while 240 pairs for evaluation, here 65,000>>240?

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions