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
It is common to specify the mlflow tracking uri through the MLFLOW_TRACKING_URI environment variable. We should have a way to use this variable as the default one in kedro-mlflow
Context
When using kedro-mlflow in a managed cluster where the tracking uri is defined dynamically, I can't use a fixed uri in the mlflow.yml config file.
Possible Implementation
When server.mlflow_tracking_uri is null in mlflow.yml, use MLFLOW_TRACKING_URI instead of mlruns as a default.
Description
It is common to specify the mlflow tracking uri through the
MLFLOW_TRACKING_URI
environment variable. We should have a way to use this variable as the default one inkedro-mlflow
Context
When using kedro-mlflow in a managed cluster where the tracking uri is defined dynamically, I can't use a fixed uri in the
mlflow.yml
config file.Possible Implementation
When
server.mlflow_tracking_uri
isnull
in mlflow.yml, useMLFLOW_TRACKING_URI
instead ofmlruns
as a default.kedro-mlflow/kedro_mlflow/config/kedro_mlflow_config.py
Line 18 in fa13573
The text was updated successfully, but these errors were encountered: