# Quick Publish Checklist

## Before Publishing
- [ ] Run `python3 -m venv .venv && source .venv/bin/activate`
- [ ] Run `pip install .`
- [ ] Run `python smoke_test.py`
- [ ] Run `python predict_one.py`
- [ ] Run `python predict_batch.py`
- [ ] Verify `ARTIFACTS_SHA256.txt`
- [ ] Review `CHANGELOG.md`
- [ ] Review `EVALUATION_SUMMARY.md`
- [ ] Review `README.md` for the final release version

## Repository Hygiene
- [ ] Confirm there is no internal branding in the repository
- [ ] Confirm there are no service-layer or conversational features in the repository
- [ ] Confirm only the champion model is present
- [ ] Confirm no large internal historical CSVs are included
- [ ] Confirm only synthetic sample CSVs are included

## Release Metadata
- [ ] Package version is correct, for example `2026.04.1`
- [ ] Model version is correct: `la_liga_score_predictor`
- [ ] Git tag is prepared, for example `release-2026.04.1`

## Publish Targets
- [ ] Push the clean public bundle to the public GitHub repository
- [ ] Publish the same release to Hugging Face
- [ ] Publish or mirror the notebook on Kaggle

## Recommended Source of Truth
The public repository should be created from this public-release branch or from a clean export of this bundle from this branch.
That avoids mixing in internal codepaths that are not part of the public package.