Skip to main content

MLflow 1.15.0

· One min read
MLflow maintainers
MLflow maintainers
MLflow maintainers

We are happy to announce the availability of MLflow 1.15.0!

In addition to bug and documentation fixes, MLflow 1.15.0 includes the following features and improvements:

  • Add silent=False option to all autologging APIs, to allow suppressing MLflow warnings and logging statements during autologging setup and training (#4173, @dbczumar)
  • Add disable_for_unsupported_versions=False option to all autologging APIs, to disable autologging for versions of ML frameworks that have not been explicitly tested against the current version of the MLflow client (#4119, @WeichenXu123)

For a comprehensive list of changes, see the release change log, and check out the latest documentation on mlflow.org.

MLflow 1.14.0

· 2 min read
MLflow maintainers
MLflow maintainers
MLflow maintainers

We are happy to announce the availability of MLflow 1.14.0!

In addition to bug and documentation fixes, MLflow 1.14.0 includes the following features and improvements:

  • MLflow's model inference APIs (mlflow.pyfunc.predict), built-in model serving tools (mlflow models serve), and model signatures now support tensor inputs. In particular, MLflow now provides built-in support for scoring PyTorch, TensorFlow, Keras, ONNX, and Gluon models with tensor inputs. For more information, see https://mlflow.org/docs/latest/models.html#deploy-mlflow-models (#3808, #3894, #4084, #4068 @wentinghu; #4041 @tomasatdatabricks, #4099, @arjundc-db)
  • Add new mlflow.shap.log_explainer, mlflow.shap.load_explainer APIs for logging and loading shap.Explainer instances (#3989, @vivekchettiar)
  • The MLflow Python client is now available with a reduced dependency set via the mlflow-skinny PyPI package (#4049, @eedeleon)
  • Add new RequestHeaderProvider plugin interface for passing custom request headers with REST API requests made by the MLflow Python client (#4042, @jimmyxu-db)
  • mlflow.keras.log_model now saves models in the TensorFlow SavedModel format by default instead of the older Keras H5 format (#4043, @harupy)
  • mlflow_log_model now supports logging MLeap models in R (#3819, @yitao-li)
  • Add mlflow.pytorch.log_state_dict, mlflow.pytorch.load_state_dict for logging and loading PyTorch state dicts (#3705, @shrinath-suresh)
  • mlflow gc can now garbage-collect artifacts stored in S3 (#3958, @sklingel)

For a comprehensive list of changes, see the release change log, and check out the latest documentation on mlflow.org.

MLflow 1.13.1

· One min read
MLflow maintainers
MLflow maintainers
MLflow maintainers

We are happy to announce the availability of MLflow 1.13.1!

MLflow 1.13.1 is a patch release containing bug fixes and small changes:

  • Fix bug causing Spark autologging to ignore configuration options specified by mlflow.autolog() (#3917, @dbczumar)
  • Fix bugs causing metrics to be dropped during TensorFlow autologging (#3913, #3914, @dbczumar)
  • Fix incorrect value of optimizer name parameter in autologging PyTorch Lightning (#3901, @harupy)
  • Fix model registry database allow_null_for_run_id migration failure affecting MySQL databases (#3836, @t-henri)
  • Fix failure in transition_model_version_stage when uncanonical stage name is passed (#3929, @harupy)
  • Fix an undefined variable error causing AzureML model deployment to fail (#3922, @eedeleon)
  • Reclassify scikit-learn as a pip dependency in MLflow Model conda environments (#3896, @harupy)
  • Fix experiment view crash and artifact view inconsistency caused by artifact URIs with redundant slashes (#3928, @dbczumar)

MLflow 1.13.0

· 2 min read
MLflow maintainers
MLflow maintainers
MLflow maintainers

We are happy to announce the availability of MLflow 1.13.0!

In addition to bug and documentation fixes, MLflow 1.13.0 includes the following features and improvements:

New fluent APIs for logging in-memory objects as artifacts:​

  • Add mlflow.log_text which logs text as an artifact (#3678, @harupy)
  • Add mlflow.log_dict which logs a dictionary as an artifact (#3685, @harupy)
  • Add mlflow.log_figure which logs a figure object as an artifact (#3707, @harupy)
  • Add mlflow.log_image which logs an image object as an artifact (#3728, @harupy)

UI updates / fixes:​

  • Add model version link in compact experiment table view
  • Add logged/registered model links in experiment runs page view
  • Enhance artifact viewer for MLflow models
  • Model registry UI settings are now persisted across browser sessions
  • Add model version description field to model version table

(#3867, @smurching)

Autologging enhancements:​

More features and improvements:​

For a comprehensive list of changes, see the release change log, and check out the latest documentation on mlflow.org.

MLflow 1.12.1

· One min read
MLflow maintainers
MLflow maintainers
MLflow maintainers

MLflow 1.12.1 is a patch release containing bug fixes and small changes:

MLflow 1.12.1 is a patch release containing bug fixes and small changes:

  • Fix run_link for cross-workspace model versions (#3681, @sueann)
  • Remove hard dependency on matplotlib for sklearn autologging (#3703, @dbczumar)
  • Do not disable existing loggers when initializing alembic (#3653, @arthury1n)

MLflow 1.12.0

· One min read
MLflow maintainers
MLflow maintainers
MLflow maintainers

We are happy to announce the availability of MLflow 1.12.0!

In addition to bug and documentation fixes, MLflow 1.12.0 includes several major features and improvements, in particular a number of improvements to MLflow's Pytorch integrations and autologging:

PyTorch

  • mlflow.pytorch.log_model, mlflow.pytorch.load_model now support logging/loading TorchScript models (#3557, @shrinath-suresh)
  • mlflow.pytorch.log_model supports passing requirements_file & extra_files arguments to log additional artifacts along with a model (#3436, @shrinath-suresh)

Autologging

  • Add universal mlflow.autolog which enables autologging for all supported integrations (#3561, #3590, @andrewnitu)
  • Add mlflow.pytorch.autolog API for automatic logging of metrics, params, and models from Pytorch Lightning training (#3601, @shrinath-suresh, #3636, @karthik-77). This API is also enabled by mlflow.autolog.
  • Scikit-learn, XGBoost, and LightGBM autologging now support logging model signatures and input examples (#3386, #3403, #3449, @andrewnitu)
  • mlflow.sklearn.autolog now supports logging metrics (e.g. accuracy) and plots (e.g. confusion matrix heat map) (#3423, #3327, @willzhan-db, @harupy)

For a comprehensive list of changes, see the release change log, and check out the latest documentation on mlflow.org.

MLflow 1.11.0

· 2 min read
MLflow maintainers
MLflow maintainers
MLflow maintainers

We are happy to announce the availability of MLflow 1.11.0!

In addition to bug and documentation fixes, MLflow 1.11.0 includes the following features and improvements:

  • New mlflow.sklearn.autolog() API for automatic logging of metrics, params, and models from scikit-learn model training (#3287, @harupy; #3323, #3358 @dbczumar)
  • Registered model & model version creation APIs now support specifying an initial description (#3271, @sueann)
  • The R mlflow_log_model and mlflow_load_model APIs now support XGBoost models (#3085, @lorenzwalthert)
  • New mlflow.list_run_infos fluent API for listing run metadata (#3183, @trangevi)
  • Added section for visualizing and comparing model schemas to model version and model-version-comparison UIs (#3209, @zhidongqu-db)
  • Enhanced support for using the model registry across Databricks workspaces: support for registering models to a Databricks workspace from outside the workspace (#3119, @sueann), tracking run-lineage of these models (#3128, #3164, @ankitmathur-db; #3187, @harupy), and calling mlflow.<flavor>.load_model against remote Databricks model registries (#3330, @sueann)
  • UI support for setting/deleting registered model and model version tags (#3187, @harupy)
  • UI support for archiving existing staging/production versions of a model when transitioning a new model version to staging/production (#3134, @harupy)

For a comprehensive list of changes, see the release change log, and check out the latest documentation on mlflow.org.

MLflow 1.10.0

· One min read
MLflow maintainers
MLflow maintainers
MLflow maintainers

We are happy to announce the availability of MLflow 1.10.0!

In addition to bug and documentation fixes, MLflow 1.10.0 includes the following features and improvements:

  • MlflowClient.transition_model_version_stage now supports an archive_existing_versions argument for archiving existing staging or production model versions when transitioning a new model version to staging or production (#3095, @harupy)
  • Added set_registry_uri, get_registry_uri APIs. Setting the model registry URI causes fluent APIs like mlflow.register_model to communicate with the model registry at the specified URI (#3072, @sueann)
  • Added paginated MlflowClient.search_registered_models API (#2939, #3023, #3027 @ankitmathur-db; #2966, @mparkhe)
  • Added syntax highlighting when viewing text files (YAML etc) in the MLflow runs UI (#3041, @harupy)

For a comprehensive list of changes, see the release change log, and check out the latest documentation on mlflow.org.

MLflow 1.9.1

· One min read
MLflow maintainers
MLflow maintainers
MLflow maintainers

MLflow 1.9.1 is a patch release containing a number of bug-fixes and improvements:

  • Fixes AttributeError when pickling an instance of the Python MlflowClient class (#2955, @Polyphenolx)
  • Fixes bug that prevented updating model-version descriptions in the model registry UI (#2969, @AnastasiaKol)
  • Fixes bug where credentials were not properly propagated to artifact CLI commands when logging artifacts from Java to the DatabricksArtifactRepository (#3001, @dbczumar)
  • Removes use of new Pandas API in new MLflow model-schema functionality, so that it can be used with older Pandas versions (#2988, @aarondav)

For a comprehensive list of changes, see the release change log, and check out the latest documentation on mlflow.org.