You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I'm sure internally Google has some great experiment tracking infrastruture, but externally, keeping track of experiments is hard, especially with tree based models that have a lot of hyper parameters, and MLFlow is the defacto standard.
It would be awesome if we could add official support for MLFlow integration.
I've submitted a request to the MLFlow repo as well here.
Plus, if Spark support is added for distributed training, this integration would go well in the databricks environments where MLFlow is typically available.
The text was updated successfully, but these errors were encountered:
I'm sure internally Google has some great experiment tracking infrastruture, but externally, keeping track of experiments is hard, especially with tree based models that have a lot of hyper parameters, and MLFlow is the defacto standard.
It would be awesome if we could add official support for MLFlow integration.
I've submitted a request to the MLFlow repo as well here.
Plus, if Spark support is added for distributed training, this integration would go well in the databricks environments where MLFlow is typically available.
The text was updated successfully, but these errors were encountered: