byom_install_location | Teradata Package for Python - byom_install_location - Teradata Package for Python

The byom_install_location configuration option indicates the name of the database where Bring Your Own Model functions are installed. This option is internally used by BYOM functions like H2OPredict() and PMMLPredict().

Example

• Import all modules.

  >>> from teradataml import *

• Set the configuration option byom_install_location to "mldb".

  >>> # Set the teradataml Configuration option 'byom_install_location' to "mldb"
  >>> configure.byom_install_location = "mldb"

• Load example data.

  >>> load_example_data("byom", "iris_test")

• Load model files.

  >>> # Load model file into Vantage.
  >>> model_file = os.path.join(os.path.dirname(teradataml.__file__), "data", "models", "iris_mojo_glm_h2o_model")

• Save the model.

  >>> save_byom("iris_mojo_glm_h2o_model", model_file, "byom_models")

• Retrieve the saved model and directly pass the output as an input to the H2OPredict function.

  >>> modeldata = retrieve_byom("iris_mojo_glm_h2o_model", table_name="byom_models")

  >>> result = H2OPredict(newdata=iris_test,
                          newdata_partition_column='id',
                          newdata_order_column='id',
                          modeldata=modeldata,
                          modeldata_order_column='model_id',
                          model_output_fields=['label', 'classProbabilities'],
                          accumulate=['id', 'sepal_length', 'petal_length'],
                          overwrite_cached_models='*',
                          enable_options='stageProbabilities',
                          model_type='OpenSource'
                         )

• Print the results of the H2OPredict function.

  >>> # Print the results.
  >>> print(result.result)