Imageomics / Imageomics/SST

Package for PyPI publication

Open
#22 0 comments 1 reaction 0 assignees View on GitHub

Nobody has claimed this yet.

Dominant language
Jupyter Notebook
Stars
14
Forks
3
PR merge metrics
No merged PRs in 30d

Description

Packaging SST for publication and installation via PyPI has been expressed as a desirable update to the project in a few contexts.

The first decision that would be needed for this though is a package name because bare [sst](https://pypi.org/project/sst/) is claimed.

And changes that seem appropriate:
- Drop GroundingDINO (or excise it into a branch for further exploration of text-based segmentation, perhaps for evaluation alongside SAM3)
- Migrate local SAM2 copy use to [sam2](https://pypi.org/project/sam2/) imports and default configs
- Specify `from sst import` to avoid ambiguous imports
- Retrieve model weights from HF automatically rather than as a manual step
- Set entry points and a simple `cli.py` module, with launching the GUI as its own subcommand
- Pin down dependencies a bit
- PyPI Actions workflow
- README and demo updates for `data/`
- Housekeeping DOI, Zenodo, etc.

This should be most of what's needed for packaging. Scope here should probably be limited to just what's needed for PyPI publication.

This would obviously be a breaking change for anyone currently using the scripts from the cloned repo. We should probably want to tag a release for the current 'scripts era' at v1.1.0 and release the updated PyPI as v2.0.0.

Any/all of the above is available for discussion!

Contributor guide

No contributing guide indexed for this repository

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

Start by deciding the package name and narrowing the listed changes to what PyPI publication requires. Review the proposed cli.py entry point, README and data/ updates, dependency and model-weight handling, and the PyPI Actions workflow. Done means the project can be installed from PyPI with documented entry points and a tagged release.

Written by the indexing model from the issue text.

Assessment

Tech stack
huggingface, python
Domain
build-system, cli, documentation, release
Issue type
Feature
Difficulty
5/5
Estimated time
Over a week
Activity status
Quiet
Clarity
Mostly clear
Newbie friendliness
35/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.