The mlflow.h2o module provides an API for logging and loading H2O models. This module exports H2O models with the following flavors:

H20 (native) format
This is the main flavor that can be loaded back into H2O.
Produced for use by generic pyfunc-based deployment tools and batch inference.
mlflow.h2o.load_model(path, run_id=None)

Load an H2O model from a local file (if run_id is None) or a run. This function expects there is an H2O instance initialised with h2o.init.

  • path – Local filesystem path or run-relative artifact path to the model saved by mlflow.h2o.save_model().
  • run_id – Run ID. If provided, combined with path to identify the model.

Load a persisted H2O model as a python_function model. This method calls h2o.init, so the right version of h2o(-py) must be in the environment. The arguments given to h2o.init can be customized in path/h2o.yaml under the key init.

Parameters:path – Local filesystem path to the model saved by mlflow.h2o.save_model().
Return type:Pyfunc format model with function model.predict(pandas DataFrame) -> pandas DataFrame.
mlflow.h2o.log_model(h2o_model, artifact_path, **kwargs)

Log an H2O model as an MLflow artifact for the current run.

  • h2o_model – H2O model to be saved.
  • artifact_path – Run-relative artifact path.
  • kwargs – kwargs to pass to h2o.save_model method.
mlflow.h2o.save_model(h2o_model, path, conda_env=None, mlflow_model=<mlflow.models.Model object>, settings=None)

Save an H2O model to a path on the local file system.

  • h2o_model – H2O model to be saved.
  • path – Local path where the model is to be saved.
  • mlflow_modelmlflow.models.Model this flavor is being added to.