Quick example of adding Github Action to MLflow Projects
This is a quick video about MLflow projects. In my attempt to understand this feature, I took a basic ML example and tried to make it reproducible with python and Github Actions.
Even with my toy example, I struggled to get all my dependencies installed, get my script to run, and enable an end user to pass input. My suckyness aside, once my GH-action was actually running, I could see how annoying it would be for anyone else running this project to make changes to it. For one, they'd have to edit my terrible code, leading to a hot mess of diffs in Github that make reviewing what changed quite complicated.
Enter mlflow projects: I put my dependencies, script, and user input in one configuration file. Now if someone makes a change, they don't have to look at my sad code, they can just see a clean diff between config files.
I'm sure there's a lot more to mlflow projects, but at a high level, this is now making more sense to me. With Continual, it's great to work with ML engineers. I'm looking forward to learning more about the tools that are commonly part of their stack
Feel free to connect with me on LI: / gustafrcavanaugh
Смотрите видео Quick example of adding Github Action to MLflow Projects онлайн без регистрации, длительностью часов минут секунд в хорошем качестве. Это видео добавил пользователь Gus Cavanaugh 08 Декабрь 2022, не забудьте поделиться им ссылкой с друзьями и знакомыми, на нашем сайте его посмотрели 944 раз и оно понравилось 10 людям.