Skip to content

Release of training/evaluation dataset (HPA image splits used for ProtiCelli) #1

Description

@auroua

Thanks for releasing the ProtiCelli code and checkpoints — great work in the bioRxiv preprint.

The README documents Model.download_checkpoints() for pre-trained weights, but I don't see a link or instructions for the underlying training/evaluation data (the 1.23M HPA images used to train ProtiCelli, or the held-out evaluation split used for the benchmarks in the paper).

Would it be possible to release:

The list of image/protein/cell-line identifiers (or accession numbers) used for the train/val/test splits, and

Any preprocessing metadata needed to reproduce the exact splits (e.g., HPA image IDs, antibody IDs)?

This would help with reproducing the reported metrics and benchmarking follow-up work.

Thanks!

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