Command-Line Interface

The MLflow command-line interface (CLI) provides a simple interface to various functionality in MLflow. You can use the CLI to run projects, start the tracking UI, create and list experiments, download run artifacts, serve MLflow Python Function and scikit-learn models, and serve models on Microsoft Azure Machine Learning and Amazon SageMaker.

$ mlflow --help
Usage: mlflow [OPTIONS] COMMAND [ARGS]...

Options:
  --version  Show the version and exit.
  --help     Show this message and exit.

Commands:
  azureml      Serve models on Azure Machine Learning.
  download     Download the artifact at the specified DBFS or S3 URI.
  experiments  Manage experiments.
  pyfunc       Serve Python models locally.
  run          Run an MLflow project from the given URI.
  sagemaker    Serve models on Amazon SageMaker.
  sklearn      Serve scikit-learn models.
  ui           Run the MLflow tracking UI.

Each individual command has a detailed help screen accessible via mlflow command_name --help.

Table of Contents

Azure Machine Learning Models
Download
Experiments
- Create
- List
- Delete
- Restore
Python Function Models
Run
SageMaker Models
scikit-learn Models
UI

Azure Machine Learning Models

Subcommands to serve models on Azure Machine Learning.

Download

Download the artifact at the specified DBFS or S3 URI into the specified local output path, or the current directory if no output path is specified.

Experiments

Subcommands to manage experiments.

In this section:

Create
List
Delete
Restore

Create

Create an experiment. The command has required argument for experiment name. Additionally, you can provide an artifact location using -l or --artifact-location option. If not provided, backend store will pick default location. Backend store will generate a unique ID for each experiment.

All artifacts generated by runs related to this experiment will be stored under artifact location, organized under specific run_uuid sub-directories.

Implementation of experiment and metadata store is dependent on backend storage. FileStore creates a folder for each experiment ID and stores metadata in meta.yaml. Runs are stored as subfolders.

List

Lists all experiments managed by backend store. Command takes an optional --view or -v argument. Valid arguments are active_only (default), deleted_only, or all.

Delete

Mark an active experiment for deletion. This also applies to experiment’s metadata, runs and associated data, and artifacts if they are store in default location. Use list command to view artifact location. Command takes a required argument for experiment ID. Command will thrown an error if experiment is not found or already marked for deletion.

Experiments marked for deletion can be restored using restore command, unless they are permanently deleted.

Specific implementation of deletion is dependent on backend stores. FileStore moves experiments marked for deletion under a .trash folder under the main folder used to instantiate FileStore. Experiments marked for deletion can be permanently deleted by clearing the .trash folder. It is recommended to use a cron job or an alternate workflow mechanism to clear .trash folder.

Restore

Restore a deleted experiment. This also applies to experiment’s metadata, runs and associated data. The command has a required argument for experiment ID. The command throws an error if the experiment is already active, cannot be found, or permanently deleted.

Python Function Models

Subcommands to serve Python models and apply them for inference.

Run

Run an MLflow project from the given URI.

If running locally (the default), the URI can be either a Git repository URI or a local path. If running on Databricks, the URI must be a Git repository.

By default, Git projects will run in a new working directory with the given parameters, while local projects will run from the project’s root directory.