Storing MLflow Experiment Artifacts in BucketFS

MLflow allows creating experiments via UI, CLI, and API.

Via UI

Open the UI of your MLflow server, click “Experiments” in the left-hand menu, and click the “Create” button in the upper right:

Via CLI

mlflow experiments create \
  --experiment-name "My Experiment" \
  --artifact-location "exa+bfs://localhost:2580/bfsdefault/default/"

For details, see MLflow CLI Documentation and URI Format.

Via API

import mlflow

uri = "exa+bfs://localhost:2580/bfsdefault/default/"
mlflow.create_experiment("My Experiment", artifact_location=uri)
mlflow.set_experiment("My Experiment")

For details, see the MLflow API Documentation and URI Format.