# Copyright (C) Microsoft Corporation. All rights reserved.
# Licensed under the MIT License. See LICENSE in project root for information.
import sys
if sys.version >= '3':
basestring = str
from pyspark import SparkContext, SQLContext
from pyspark.sql import DataFrame
from pyspark.ml.param.shared import *
from pyspark import keyword_only
from pyspark.ml.util import JavaMLReadable, JavaMLWritable
from synapse.ml.core.serialize.java_params_patch import *
from pyspark.ml.wrapper import JavaTransformer, JavaEstimator, JavaModel
from pyspark.ml.evaluation import JavaEvaluator
from pyspark.ml.common import inherit_doc
from synapse.ml.core.schema.Utils import *
from pyspark.ml.param import TypeConverters
from synapse.ml.core.schema.TypeConversionUtils import generateTypeConverter, complexTypeConverter
from synapse.ml.automl.TuneHyperparametersModel import TuneHyperparametersModel
[docs]@inherit_doc
class TuneHyperparameters(ComplexParamsMixin, JavaMLReadable, JavaMLWritable, JavaEstimator):
"""
Args:
evaluationMetric (object): Metric to evaluate models with
models (object): Estimators to run
numFolds (int): Number of folds
numRuns (int): Termination criteria for randomized search
parallelism (int): The number of models to run in parallel
paramSpace (object): Parameter space for generating hyperparameters
seed (long): Random number generator seed
"""
evaluationMetric = Param(Params._dummy(), "evaluationMetric", "Metric to evaluate models with")
models = Param(Params._dummy(), "models", "Estimators to run")
numFolds = Param(Params._dummy(), "numFolds", "Number of folds", typeConverter=TypeConverters.toInt)
numRuns = Param(Params._dummy(), "numRuns", "Termination criteria for randomized search", typeConverter=TypeConverters.toInt)
parallelism = Param(Params._dummy(), "parallelism", "The number of models to run in parallel", typeConverter=TypeConverters.toInt)
paramSpace = Param(Params._dummy(), "paramSpace", "Parameter space for generating hyperparameters")
seed = Param(Params._dummy(), "seed", "Random number generator seed")
@keyword_only
def __init__(
self,
java_obj=None,
evaluationMetric=None,
models=None,
numFolds=None,
numRuns=None,
parallelism=None,
paramSpace=None,
seed=0
):
super(TuneHyperparameters, self).__init__()
if java_obj is None:
self._java_obj = self._new_java_obj("com.microsoft.azure.synapse.ml.automl.TuneHyperparameters", self.uid)
else:
self._java_obj = java_obj
self._setDefault(seed=0)
if hasattr(self, "_input_kwargs"):
kwargs = self._input_kwargs
else:
kwargs = self.__init__._input_kwargs
if java_obj is None:
for k,v in kwargs.items():
if v is not None:
getattr(self, "set" + k[0].upper() + k[1:])(v)
[docs] @keyword_only
def setParams(
self,
evaluationMetric=None,
models=None,
numFolds=None,
numRuns=None,
parallelism=None,
paramSpace=None,
seed=0
):
"""
Set the (keyword only) parameters
"""
if hasattr(self, "_input_kwargs"):
kwargs = self._input_kwargs
else:
kwargs = self.__init__._input_kwargs
return self._set(**kwargs)
[docs] @classmethod
def read(cls):
""" Returns an MLReader instance for this class. """
return JavaMMLReader(cls)
[docs] @staticmethod
def getJavaPackage():
""" Returns package name String. """
return "com.microsoft.azure.synapse.ml.automl.TuneHyperparameters"
@staticmethod
def _from_java(java_stage):
module_name=TuneHyperparameters.__module__
module_name=module_name.rsplit(".", 1)[0] + ".TuneHyperparameters"
return from_java(java_stage, module_name)
[docs] def setEvaluationMetric(self, value):
"""
Args:
evaluationMetric: Metric to evaluate models with
"""
self._set(evaluationMetric=value)
return self
[docs] def setModels(self, value):
"""
Args:
models: Estimators to run
"""
self._set(models=value)
return self
[docs] def setNumFolds(self, value):
"""
Args:
numFolds: Number of folds
"""
self._set(numFolds=value)
return self
[docs] def setNumRuns(self, value):
"""
Args:
numRuns: Termination criteria for randomized search
"""
self._set(numRuns=value)
return self
[docs] def setParallelism(self, value):
"""
Args:
parallelism: The number of models to run in parallel
"""
self._set(parallelism=value)
return self
[docs] def setParamSpace(self, value):
"""
Args:
paramSpace: Parameter space for generating hyperparameters
"""
self._set(paramSpace=value)
return self
[docs] def setSeed(self, value):
"""
Args:
seed: Random number generator seed
"""
self._set(seed=value)
return self
[docs] def getEvaluationMetric(self):
"""
Returns:
evaluationMetric: Metric to evaluate models with
"""
return self.getOrDefault(self.evaluationMetric)
[docs] def getModels(self):
"""
Returns:
models: Estimators to run
"""
return self.getOrDefault(self.models)
[docs] def getNumFolds(self):
"""
Returns:
numFolds: Number of folds
"""
return self.getOrDefault(self.numFolds)
[docs] def getNumRuns(self):
"""
Returns:
numRuns: Termination criteria for randomized search
"""
return self.getOrDefault(self.numRuns)
[docs] def getParallelism(self):
"""
Returns:
parallelism: The number of models to run in parallel
"""
return self.getOrDefault(self.parallelism)
[docs] def getParamSpace(self):
"""
Returns:
paramSpace: Parameter space for generating hyperparameters
"""
return self.getOrDefault(self.paramSpace)
[docs] def getSeed(self):
"""
Returns:
seed: Random number generator seed
"""
return self.getOrDefault(self.seed)
def _create_model(self, java_model):
try:
model = TuneHyperparametersModel(java_obj=java_model)
model._transfer_params_from_java()
except TypeError:
model = TuneHyperparametersModel._from_java(java_model)
return model
def _fit(self, dataset):
java_model = self._fit_java(dataset)
return self._create_model(java_model)