forecastflowml.ForecastFlowML.train#

ForecastFlowML.train(df, local_result=False)[source]#

Train models

Parameters:

  • df (pyspark.sql.dataframe.DataFrame) – Dataset to fit.

  • local_result (bool) – Whether to store trained models as attribute. Only provide True in case of the trained models are not expected to overload the driver node.

Return type:

None if local_result=True. Otherwise, pyspark DataFrame that includes the trained models.