Andre Mesarovic
Andre Mesarovic
Sounds interesting, let me check it out. Only issue might be that Databricks notebooks are saved as ".py" files, so we'd have to exclude them somehow.
We might also want to check out how the core mlflow repo handles code formatting so as to be compatible with them.
Good catch. I'm finishing up on another PR now. Will merge yours after that.
Tests passed and I approved the change.
Sounds great. I've thought of doing this too, similar to the mlflow CLI, but time sort of ran out. Do you have a link to your fork?
I am on PTO until August 14, so responses might be delayed. :)
end_time is a strictly read-only value set by the system. start_time is also set by the system with an option in create_run() to pass in a time though w/o being...
MLFLOW_TRACKING_URI has to be set. The tool does not handle file-based store where not MLFLOW_TRACKING_URI is needed. Not sure what the issue is.
Hey there, sorry for the delay. > where the "same" experiment exists already (name=NAME, ID=YYY). Is that an intended use case? Yes, this should be OK. If experiment exists import...
When you create a PAT you specify its lifetime in days. Not aware of a problem for long-running requests. Best to ask in the https://community.databricks.com/s/