A repository for stock implementations of popular machine learning models.
Every model should at mininum
- Have separate scripts for training and analyzing models before production
- Use a commandline interface.
Use pipreqs to generate requirements.txt Almost always use the latest packages (no rollbacks).
I'm terrible at Flask, so help me. Issues or PRs welcome.
Add pandas_profiling to the analysis. Learn Flask or FastAPI -- any deployable method. Is a package format really necessary to use these? Maybe it should be a cookie-cutter