Machine Learning utilities#

MLflow#

Arbitrary MLflow models can be used with pandas-on-Spark Dataframes, provided they implement the ‘pyfunc’ flavor. This is the case for most frameworks supported by MLflow (scikit-learn, pytorch, tensorflow, …). See comprehensive examples in load_model() for more information.

Note

The MLflow package must be installed in order to use this module.

PythonModelWrapper(model_uri, return_type_hint)

A wrapper around MLflow's Python object model.

load_model(model_uri[, predict_type])

Loads an MLflow model into a wrapper that can be used both for pandas and pandas-on-Spark DataFrame.