mlflow

The mlflow module provides an API for starting and managing MLflow runs. For example:

import mlflow
mlflow.start_run()
mlflow.log_param("my", "param")
mlflow.log_metric("score", 100)
mlflow.end_run()

You can also use syntax like this:

with mlflow.start_run() as run:
    ...

which automatically terminates the run at the end of the block.

For a lower level API, see the mlflow.tracking module.

class mlflow.ActiveRun(run)

Bases: mlflow.entities.run.Run

Wrapper around mlflow.entities.Run to enable using Python with syntax.

mlflow.log_param(key, value)

Log a parameter under the current run, creating a run if necessary.

Parameters:	key – Parameter name (string) value – Parameter value (string, but will be string-ified if not)

mlflow.log_metric(key, value)

Log a metric under the current run, creating a run if necessary.

Parameters:	key – Metric name (string). value – Metric value (float).

mlflow.set_tag(key, value)

Set a tag under the current run, creating a run if necessary.

Parameters:	key – Tag name (string) value – Tag value (string, but will be string-ified if not)

mlflow.log_artifacts(local_dir, artifact_path=None)

Log all the contents of a local directory as artifacts of the run.

Parameters:	local_dir – Path to the directory of files to write. artifact_path – If provided, the directory in `artifact_uri` to write to.

mlflow.log_artifact(local_path, artifact_path=None)

Log a local file or directory as an artifact of the currently active run.

Parameters:	local_path – Path to the file to write. artifact_path – If provided, the directory in `artifact_uri` to write to.

mlflow.active_run(): Get the currently active Run, or None if no such run exists.

mlflow.start_run(run_uuid=None, experiment_id=None, source_name=None, source_version=None, entry_point_name=None, source_type=None, run_name=None)

Start a new MLflow run, setting it as the active run under which metrics and parameters will be logged. The return value can be used as a context manager within a with block; otherwise, you must call end_run() to terminate the current run.

If you pass a run_uuid or the MLFLOW_RUN_ID environment variable is set, start_run attempts to resume a run with the specified run ID and other parameters are ignored. run_uuid takes precedence over MLFLOW_RUN_ID.

Parameters:

run_uuid – If specified, get the run with the specified UUID and log parameters and metrics under that run. The run’s end time is unset and its status is set to running, but the run’s other attributes (source_version, source_type, etc.) are not changed.
experiment_id – ID of the experiment under which to create the current run. Used only when run_uuid is unspecified. If unspecified, the run is created under the default experiment.
source_name – Name of the source file or URI of the project to be associated with the run. If none provided defaults to the current file.
source_version – Optional Git commit hash to associate with the run.
entry_point_name – Optional name of the entry point for the current run.
source_type – Integer mlflow.entities.SourceType describing the type of the run (“local”, “project”, etc.). Defaults to mlflow.entities.SourceType.LOCAL (“local”).
run_name – Name of new run. Used only when run_uuid is unspecified.

Returns:

mlflow.ActiveRun object that acts as a context manager wrapping the run’s state.

mlflow.end_run(status='FINISHED'): End an active MLflow run (if there is one).

mlflow.get_artifact_uri(): Get the artifact URI of the currently active run. Calls to log_artifact and log_artifacts write artifact(s) to subdirectories of the returned URI.

mlflow.set_tracking_uri(uri)

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

Parameters:	uri – 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>”.

mlflow.create_experiment(name, artifact_location=None)

Create an experiment.

Parameters:	name – The experiment name. Must be unique. artifact_location – The location to store run artifacts. If not provided, the server picks an appropriate default.
Returns:	Integer ID of the created experiment.

mlflow.run(uri, entry_point='main', version=None, parameters=None, experiment_id=None, mode=None, cluster_spec=None, git_username=None, git_password=None, use_conda=True, storage_dir=None, block=True, run_id=None)

Run an MLflow project. The project can be local or stored at a Git URI.

You can run the project locally or remotely on a Databricks.

For information on using this method in chained workflows, see Building Multistep Workflows.

Raises:	ExecutionException – If a run launched in blocking mode is unsuccessful.
Parameters:	uri – URI of project to run. A local filesystem path or a Git repository URI (e.g. https://github.com/mlflow/mlflow-example) pointing to a project directory containing an MLproject file. entry_point – Entry point to run within the project. If no entry point with the specified name is found, runs the project file `entry_point` as a script, using “python” to run `.py` files and the default shell (specified by environment variable `$SHELL`) to run `.sh` files. version – For Git-based projects, a commit hash. experiment_id – ID of experiment under which to launch the run. mode – Execution mode of the run: “local” or “databricks”. cluster_spec – When `mode` is “databricks”, path to a JSON file containing a Databricks cluster specification to use when launching a run. git_username – Username for HTTP(S) authentication with Git. git_password – Password for HTTP(S) authentication with Git. use_conda – If True (the default), create a new Conda environment for the run and install project dependencies within that environment. Otherwise, run the project in the current environment without installing any project dependencies. storage_dir – Used only if `mode` is “local”. MLflow downloads artifacts from distributed URIs passed to parameters of type `path` to subdirectories of `storage_dir`. block – Whether to block while waiting for a run to complete. Defaults to True. Note that if `block` is False and mode is “local”, this method will return, but the current process will block when exiting until the local run completes. If the current process is interrupted, any asynchronous runs launched via this method will be terminated. run_id – Note: this argument is used internally by the MLflow project APIs and should not be specified. If specified, the run ID will be used instead of creating a new run.
Returns:	`mlflow.projects.SubmittedRun` exposing information (e.g. run ID) about the launched run.