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
When running mlflow.set_experiment(f"demo_at_{time.time()}") I got
ERRO[0000] SQL error app_file="github.com/mlflow/mlflow-go-backend/pkg/tracking/store/sql/experiments.go:255" app_func="github.com/mlflow/mlflow-go-backend/pkg/tracking/store/sql.TrackingSQLStore.GetExperimentByName()" elapsed=0.631ms error="record not found" rows=0 sql="SELECT * FROM \"experiments\" WHERE name = 'demo_at_1737635915.371468' ORDER BY \"experiments\".\"experiment_id\" LIMIT 1"
2025/01/23 13:38:35 INFO mlflow.tracking.fluent: Experiment with name 'demo_at_1737635915.371468' does not exist. Creating a new experiment.
When running
mlflow.set_experiment(f"demo_at_{time.time()}")
I gotTests to reproduce:
Run postgres docker container:
Install
uv add mlflow-go
Script:
uv run ./script.py
I believe the problem is here that our Go implementation returns an error if didn't find anything.
While Python returns None instead:
https://github.com/mlflow/mlflow/blob/d4bccf76dfb79165731a3643a1397085060e6593/mlflow/store/tracking/sqlalchemy_store.py#L402-L417
While we do:
mlflow-go-backend/pkg/tracking/store/sql/experiments.go
Lines 250 to 272 in ed0105f
The text was updated successfully, but these errors were encountered: