Skip to content

About homophily of datasets #37

Open

Description

@huangJC0429

why i compute homophily on cora just 0.537, but in this paper cora achieve 0.82?

Activity

  1. dongkwan-kim commented on Apr 12, 2021

    @dongkwan-kim
    Owner

    In our research community, we have several variants of the Cora dataset and homophily metric. I used the one in the Geom-GCN paper (See Table 4's \beta in https://openreview.net/forum?id=S1e2agrFvS).

  2. UnBuen commented on Jul 19, 2021

    @UnBuen

    In our research community, we have several variants of the Cora dataset and homophily metric. I used the one in the Geom-GCN paper (See Table 4's \beta in https://openreview.net/forum?id=S1e2agrFvS).

    Dear dongkwan-kim,
    Table 4 doesn't exist in https://openreview.net/forum?id=S1e2agrFvS, maybe is table 2?

  3. dongkwan-kim commented on Jul 19, 2021

    @dongkwan-kim
    Owner

    Hi, @UnBuen. Thank you for your interest.
    Table 4 (Mean Classification Accuracy (Percent)) is on page 9. Could you check again? Thank you.

  4. UnBuen commented on Jul 19, 2021

    @UnBuen

    Hi, @UnBuen. Thank you for your interest.
    Table 4 (Mean Classification Accuracy (Percent)) is on page 9. Could you check again? Thank you.

    Oh, thanks, I open the original pdf(the original pdf doesn't exit table 4), now I have found, it's my fault, thank you for your persistence

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

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