- flytekitplugins.mlflow.mlflow_autolog(fn=None, *, framework=<module 'mlflow.sklearn' from '/home/docs/checkouts/readthedocs.org/user_builds/flytekit/envs/latest/lib/python3.9/site-packages/mlflow/sklearn/__init__.py'>, experiment_name=None)[source]#
MLFlow decorator to enable autologging of training metrics.
This decorator can be used as a nested decorator for a
@taskand it will automatically enable mlflow autologging, for the given
framework. By default autologging is enabled for
@task @mlflow_autolog(framework=mlflow.tensorflow) def my_tensorflow_trainer(): ...
One benefit of doing so is that the mlflow metrics are then rendered inline using FlyteDecks and can be viewed in jupyter notebook, as well as in hosted Flyte environment:
# jupyter notebook cell with flytekit.new_context() as ctx: my_tensorflow_trainer() ctx.get_deck() # IPython.display
When the task is called in a Flyte backend, the decorator starts a new MLFlow run using the Flyte execution name by default, or a user-provided
experiment_namein the decorator.