The mlflow.tracking module provides a Python CRUD interface to MLflow experiments and runs. This is a lower level API that directly translates to MLflow REST API calls. For a higher level API for managing an “active run”, use the mlflow module.

class mlflow.tracking.MlflowClient(tracking_uri=None)

Bases: object

Client of an MLflow Tracking Server that creates and manages experiments and runs.

create_experiment(name, artifact_location=None)

Create an experiment.

  • name – The experiment name. Must be unique.
  • artifact_location – The location to store run artifacts. If not provided, the server picks an appropriate default.

Integer ID of the created experiment.

create_run(experiment_id, user_id=None, run_name=None, start_time=None, parent_run_id=None, tags=None)

Create a mlflow.entities.Run object that can be associated with metrics, parameters, artifacts, etc. Unlike, creates objects but does not run code. Unlike mlflow.start_run(), does not change the “active run” used by mlflow.log_param().

  • user_id – If not provided, use the current user as a default.
  • start_time – If not provided, use the current timestamp.
:param parent_run_id Optional parent run ID - takes precedence over parent run ID included
in the tags argument.
Parameters:tags – A dictionary of key-value pairs that are converted into mlflow.entities.RunTag objects.
Returns:mlflow.entities.Run that was created.

Delete an experiment from the backend store.

Parameters:experiment_id – The experiment ID returned from create_experiment.

Deletes a run with the given ID.

download_artifacts(run_id, path)

Download an artifact file or directory from a run to a local directory if applicable, and return a local path for it.

  • run_id – The run to download artifacts from.
  • path – Relative source path to the desired artifact.

Local path of desired artifact.

Parameters:experiment_id – The experiment ID returned from create_experiment.
Parameters:name – The experiment name.

Fetch the run from backend store. The resulting Run contains a collection of run metadata - RunInfo, as well as a collection of run parameters, tags, and metrics - :py:class`RunData <mlflow.entities.RunData>`. In the case where multiple metrics with the same key are logged for the run, the RunData contains the value at the latest timestamp for each metric. If there are multiple values with the latest timestamp for a given metric, the maximum of these values is returned.

Parameters:run_uuid – Unique identifier for the run.
Returns:A single mlflow.entities.Run object, if the run exists. Otherwise, raises an exception.
list_artifacts(run_id, path=None)

List the artifacts for a run.

  • run_id – The run to list artifacts from.
  • path – The run’s relative artifact path to list from. By default it is set to None or the root artifact path.

List of mlflow.entities.FileInfo

Returns:List of mlflow.entities.Experiment
list_run_infos(experiment_id, run_view_type=1)
Returns:List of mlflow.entities.RunInfo
log_artifact(run_id, local_path, artifact_path=None)

Write a local file to the remote artifact_uri.

  • local_path – Path to the file to write.
  • artifact_path – If provided, the directory in artifact_uri to write to.
log_artifacts(run_id, local_dir, artifact_path=None)

Write a directory of files to the remote artifact_uri.

  • local_dir – Path to the directory of files to write.
  • artifact_path – If provided, the directory in artifact_uri to write to.
log_batch(run_id, metrics, params, tags)

Log multiple metrics, params, and/or tags.

  • metrics – List of Metric(key, value, timestamp) instances.
  • params – List of Param(key, value) instances.
  • tags – List of RunTag(key, value) instances.

Raises an MlflowException if any errors occur. :returns: None

log_metric(run_id, key, value, timestamp=None)

Log a metric against the run ID. If timestamp is not provided, uses the current timestamp.

log_param(run_id, key, value)

Log a parameter against the run ID. Value is converted to a string.

rename_experiment(experiment_id, new_name)

Update an experiment’s name. The new name must be unique.

Parameters:experiment_id – The experiment ID returned from create_experiment.

Restore a deleted experiment unless permanently deleted.

Parameters:experiment_id – The experiment ID returned from create_experiment.

Restores a deleted run with the given ID.

search_runs(experiment_ids, filter_string, run_view_type=1)

Search experiments that fit the search criteria.

  • experiment_ids – List of experiment IDs
  • filter_string – Filter query string.
  • run_view_type – one of enum values ACTIVE_ONLY, DELETED_ONLY, or ALL runs defined in mlflow.entities.ViewType.

set_tag(run_id, key, value)

Set a tag on the run ID. Value is converted to a string.

set_terminated(run_id, status=None, end_time=None)

Set a run’s status to terminated.

  • status – A string value of mlflow.entities.RunStatus. Defaults to “FINISHED”.
  • end_time – If not provided, defaults to the current time.

Get the current tracking URI. This may not correspond to the tracking URI of the currently active run, since the tracking URI can be updated via set_tracking_uri.

Returns:The tracking URI.

Set the tracking server URI. This does not affect the currently active run (if one exists), but takes effect for successive runs.

  • An empty string, or a local file path, prefixed with file:/. Data is stored locally at the provided file (or ./mlruns if empty).
  • An HTTP URI like https://my-tracking-server:5000.
  • A Databricks workspace, provided as the string “databricks” or, to use a Databricks CLI profile, “databricks://<profileName>”.