Quick Example of ML pipeline with dbt-python and Github Actions
This is a quick overview of taking a basic #ML pipeline using #dbt-python and #Snowflake and adding a quick #CI test with #Github-Actions.
Note:
This is a toy example (as always)
I've done nothing around model performance, fairness, drift, etc, e.g., the actual hard stuff in ML
But if you're a data scientist who thinks "hey, if I adopted more software engineering practices I might get more models deployed", this should help build some intuition around basic concepts (and we're very thankful that these types of people reach out to us Continual)
Also, given that dbt-python uses #snowpark to run all of the compute in Snowflake, the default Github Actions compute instance works really well (because I don't need lots of storage and compute on the CI/CD server as everything is just happening in Snowflake). This is at least as feasible as spinning up huge K8s cluster to handle all your ML training/inferencing (and much easier).
Смотрите видео Quick Example of ML pipeline with dbt-python and Github Actions онлайн без регистрации, длительностью часов минут секунд в хорошем качестве. Это видео добавил пользователь Gus Cavanaugh 02 Ноябрь 2022, не забудьте поделиться им ссылкой с друзьями и знакомыми, на нашем сайте его посмотрели 698 раз и оно понравилось 11 людям.