This is a thread to document the use of python packaging in Guild and for ML projects generally.
TODOs
- [ ] How to create packages for a bare model (not trained, empty) and several trained versions
- naming conflicts?
- automated scripts to re-package from bare?
- [ ] proper
- [ ] package untrained model
- [ ] package trained model
- [ ] test Guild packaging
- [ ] test Guild remote execution
Refs
- Main Docs
- Python Packaging Authority — PyPA documentation (via Guild meet [2020-10-15 Thu])
- Python Packaging User Guide — Python Packaging User Guide
- Guides
- Minimal Structure — Python Packaging Tutorial
- activescott/python-package-example: A simple example of creating and consuming a distributable Python package.
- Knowledge Bits — Common Python Packaging Mistakes [2020-08-31 Mon] (via C. Martin)
- Publishing (Perfect) Python Packages on PyPi - YouTube [2020-10-14 Wed 21:55]
- package server
- Relevant packages (also can use as guides for project setup) (via Guild meet [2020-10-15 Thu])
- Fora, Q&A:
- Tools (incomplete)
- mgedmin/check-manifest: Tool to check the completeness of MANIFEST.in for Python packages [2020-10-14 Wed 22:12]
- Coverage.py — Coverage.py 5.3 documentation [2020-10-14 Wed 22:12]
- unittest.mock — getting started — Python 3.9.0 documentation [2020-10-14 Wed 22:12]
- Readme file
- Licenses