Skip to content

Request for pretrained checkpoint for evaluation #2

Description

@fengyu8

Hi, thank you for releasing the VSCD code and dataset.

I would like to reproduce the evaluation results using the provided eval.py. The README requires a trained model checkpoint (for example, runs/vscd_default/ckpt_best.pt), but I could not find a pretrained VSCD checkpoint in the repository, GitHub Releases, or the author's Hugging Face account.

Are there plans to release the official pretrained checkpoint and, if applicable, the exact configuration used to obtain the reported results? Having these files would make it possible to run evaluation directly and verify the benchmark results without retraining the full model.

Thank you for your help.

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Fields

    No fields configured for issues without a type.

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions