Skip to content

The evaluation test set size. #13

Description

@hussien

Thanks for the well-written paper.
Regarding the evaluation results in Table 1, does the test set contain all the questions in the dataset test split, i.e., 1,628 in WebQSP, or just 100 random questions as mentioned in the code test[:100]?

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