MLflow module

class MLflow.MLApplication(model, preprocessing)

Bases: mlflow.pyfunc.model.PythonModel

Define the MLFlow “PythonModel”. This clas is used to deploy the model as a REST API.

Parameters

  • model – fitted model

  • preprocessing – fitted pipeline

predict(context, x)

Evaluates a pyfunc-compatible input and produces a pyfunc-compatible output. For more information about the pyfunc input/output API, see the pyfunc-inference-api.

Parameters

  • context – A PythonModelContext instance containing artifacts that the model can use to perform inference.

  • model_input – A pyfunc-compatible input for the model to evaluate.

MLflow.track(metrics=None, params=None, artifacts=None, model=None, feature_engineering=None, preprocessing=None, mlflow_dir='./mlruns', artifacts_path='outputs/plots/mlflow_artifacts')

Ensure MLFlow Tracking.

Parameters

  • metrics (dict) – Model’s metrics.

  • params (dict) – Best hyperparameters found.

  • artifacts – Plots.

  • model – fitted model.

  • feature_engineering – Not used for now.

  • preprocessing – Fitted pipeline.

  • mlflow_dir (str) – path describing where to store MLFlows runs. Defaults to ‘./mlruns’.

  • artifacts_path (str) – path of the artifacts to add to MLFlow. Defaults to “outputs/plots/mlflow_artifacts”.