Skip to content

lopster return negative predictions for invalid inputs #36

Open

Description

@sfluegel05

Finding based on claude in ChEB-AI/chebifier-web#39 (comment)

Invalid SMILES come back as [] instead of the documented None. Lopster returns an all-negative prediction for unparseable input instead of failing, so the ensemble never marks the input as a complete failure. The fix belongs in python-chebifier. The test is marked xfail(strict=True), so CI will flag the marker once Lopster is fixed.

Activity

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

    Type

    No type

    Projects

    No projects

      Milestone

      No milestone

      Relationships

      None yet

      Development

      No branches or pull requests

      Issue actions