# 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.platform import running_on_synapse_internal
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.causal.OrthoForestDMLModel import OrthoForestDMLModel
[docs]@inherit_doc
class OrthoForestDMLEstimator(ComplexParamsMixin, JavaMLReadable, JavaMLWritable, JavaEstimator):
"""
Args:
confidenceLevel (float): confidence level, default value is 0.975
confounderVecCol (str): Confounders to control for
featuresCol (str): The name of the features column
heterogeneityVecCol (str): Vector to divide the treatment by
maxDepth (int): Max Depth of Tree
maxIter (int): maximum number of iterations (>= 0)
minSamplesLeaf (int): Max Depth of Tree
numTrees (int): Number of trees
outcomeCol (str): outcome column
outcomeModel (object): outcome model to run
outcomeResidualCol (str): Outcome Residual Column
outputCol (str): The name of the output column
outputHighCol (str): Output Confidence Interval Low
outputLowCol (str): Output Confidence Interval Low
parallelism (int): the number of threads to use when running parallel algorithms
sampleSplitRatio (list): Sample split ratio for cross-fitting. Default: [0.5, 0.5].
treatmentCol (str): treatment column
treatmentModel (object): treatment model to run
treatmentResidualCol (str): Treatment Residual Column
weightCol (str): The name of the weight column
"""
confidenceLevel = Param(Params._dummy(), "confidenceLevel", "confidence level, default value is 0.975", typeConverter=TypeConverters.toFloat)
confounderVecCol = Param(Params._dummy(), "confounderVecCol", "Confounders to control for", typeConverter=TypeConverters.toString)
featuresCol = Param(Params._dummy(), "featuresCol", "The name of the features column", typeConverter=TypeConverters.toString)
heterogeneityVecCol = Param(Params._dummy(), "heterogeneityVecCol", "Vector to divide the treatment by", typeConverter=TypeConverters.toString)
maxDepth = Param(Params._dummy(), "maxDepth", "Max Depth of Tree", typeConverter=TypeConverters.toInt)
maxIter = Param(Params._dummy(), "maxIter", "maximum number of iterations (>= 0)", typeConverter=TypeConverters.toInt)
minSamplesLeaf = Param(Params._dummy(), "minSamplesLeaf", "Max Depth of Tree", typeConverter=TypeConverters.toInt)
numTrees = Param(Params._dummy(), "numTrees", "Number of trees", typeConverter=TypeConverters.toInt)
outcomeCol = Param(Params._dummy(), "outcomeCol", "outcome column", typeConverter=TypeConverters.toString)
outcomeModel = Param(Params._dummy(), "outcomeModel", "outcome model to run")
outcomeResidualCol = Param(Params._dummy(), "outcomeResidualCol", "Outcome Residual Column", typeConverter=TypeConverters.toString)
outputCol = Param(Params._dummy(), "outputCol", "The name of the output column", typeConverter=TypeConverters.toString)
outputHighCol = Param(Params._dummy(), "outputHighCol", "Output Confidence Interval Low", typeConverter=TypeConverters.toString)
outputLowCol = Param(Params._dummy(), "outputLowCol", "Output Confidence Interval Low", typeConverter=TypeConverters.toString)
parallelism = Param(Params._dummy(), "parallelism", "the number of threads to use when running parallel algorithms", typeConverter=TypeConverters.toInt)
sampleSplitRatio = Param(Params._dummy(), "sampleSplitRatio", "Sample split ratio for cross-fitting. Default: [0.5, 0.5].", typeConverter=TypeConverters.toListFloat)
treatmentCol = Param(Params._dummy(), "treatmentCol", "treatment column", typeConverter=TypeConverters.toString)
treatmentModel = Param(Params._dummy(), "treatmentModel", "treatment model to run")
treatmentResidualCol = Param(Params._dummy(), "treatmentResidualCol", "Treatment Residual Column", typeConverter=TypeConverters.toString)
weightCol = Param(Params._dummy(), "weightCol", "The name of the weight column", typeConverter=TypeConverters.toString)
@keyword_only
def __init__(
self,
java_obj=None,
confidenceLevel=0.975,
confounderVecCol="XW",
featuresCol=None,
heterogeneityVecCol="X",
maxDepth=5,
maxIter=1,
minSamplesLeaf=10,
numTrees=20,
outcomeCol=None,
outcomeModel=None,
outcomeResidualCol="OutcomeResidual",
outputCol="EffectAverage",
outputHighCol="EffectUpperBound",
outputLowCol="EffectLowerBound",
parallelism=10,
sampleSplitRatio=[0.5,0.5],
treatmentCol=None,
treatmentModel=None,
treatmentResidualCol="TreatmentResidual",
weightCol=None
):
super(OrthoForestDMLEstimator, self).__init__()
if java_obj is None:
self._java_obj = self._new_java_obj("com.microsoft.azure.synapse.ml.causal.OrthoForestDMLEstimator", self.uid)
else:
self._java_obj = java_obj
self._setDefault(confidenceLevel=0.975)
self._setDefault(confounderVecCol="XW")
self._setDefault(heterogeneityVecCol="X")
self._setDefault(maxDepth=5)
self._setDefault(maxIter=1)
self._setDefault(minSamplesLeaf=10)
self._setDefault(numTrees=20)
self._setDefault(outcomeResidualCol="OutcomeResidual")
self._setDefault(outputCol="EffectAverage")
self._setDefault(outputHighCol="EffectUpperBound")
self._setDefault(outputLowCol="EffectLowerBound")
self._setDefault(parallelism=10)
self._setDefault(sampleSplitRatio=[0.5,0.5])
self._setDefault(treatmentResidualCol="TreatmentResidual")
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,
confidenceLevel=0.975,
confounderVecCol="XW",
featuresCol=None,
heterogeneityVecCol="X",
maxDepth=5,
maxIter=1,
minSamplesLeaf=10,
numTrees=20,
outcomeCol=None,
outcomeModel=None,
outcomeResidualCol="OutcomeResidual",
outputCol="EffectAverage",
outputHighCol="EffectUpperBound",
outputLowCol="EffectLowerBound",
parallelism=10,
sampleSplitRatio=[0.5,0.5],
treatmentCol=None,
treatmentModel=None,
treatmentResidualCol="TreatmentResidual",
weightCol=None
):
"""
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.causal.OrthoForestDMLEstimator"
@staticmethod
def _from_java(java_stage):
module_name=OrthoForestDMLEstimator.__module__
module_name=module_name.rsplit(".", 1)[0] + ".OrthoForestDMLEstimator"
return from_java(java_stage, module_name)
[docs] def setConfidenceLevel(self, value):
"""
Args:
confidenceLevel: confidence level, default value is 0.975
"""
self._set(confidenceLevel=value)
return self
[docs] def setConfounderVecCol(self, value):
"""
Args:
confounderVecCol: Confounders to control for
"""
self._set(confounderVecCol=value)
return self
[docs] def setFeaturesCol(self, value):
"""
Args:
featuresCol: The name of the features column
"""
self._set(featuresCol=value)
return self
[docs] def setHeterogeneityVecCol(self, value):
"""
Args:
heterogeneityVecCol: Vector to divide the treatment by
"""
self._set(heterogeneityVecCol=value)
return self
[docs] def setMaxDepth(self, value):
"""
Args:
maxDepth: Max Depth of Tree
"""
self._set(maxDepth=value)
return self
[docs] def setMaxIter(self, value):
"""
Args:
maxIter: maximum number of iterations (>= 0)
"""
self._set(maxIter=value)
return self
[docs] def setMinSamplesLeaf(self, value):
"""
Args:
minSamplesLeaf: Max Depth of Tree
"""
self._set(minSamplesLeaf=value)
return self
[docs] def setNumTrees(self, value):
"""
Args:
numTrees: Number of trees
"""
self._set(numTrees=value)
return self
[docs] def setOutcomeCol(self, value):
"""
Args:
outcomeCol: outcome column
"""
self._set(outcomeCol=value)
return self
[docs] def setOutcomeModel(self, value):
"""
Args:
outcomeModel: outcome model to run
"""
self._set(outcomeModel=value)
return self
[docs] def setOutcomeResidualCol(self, value):
"""
Args:
outcomeResidualCol: Outcome Residual Column
"""
self._set(outcomeResidualCol=value)
return self
[docs] def setOutputCol(self, value):
"""
Args:
outputCol: The name of the output column
"""
self._set(outputCol=value)
return self
[docs] def setOutputHighCol(self, value):
"""
Args:
outputHighCol: Output Confidence Interval Low
"""
self._set(outputHighCol=value)
return self
[docs] def setOutputLowCol(self, value):
"""
Args:
outputLowCol: Output Confidence Interval Low
"""
self._set(outputLowCol=value)
return self
[docs] def setParallelism(self, value):
"""
Args:
parallelism: the number of threads to use when running parallel algorithms
"""
self._set(parallelism=value)
return self
[docs] def setSampleSplitRatio(self, value):
"""
Args:
sampleSplitRatio: Sample split ratio for cross-fitting. Default: [0.5, 0.5].
"""
self._set(sampleSplitRatio=value)
return self
[docs] def setTreatmentCol(self, value):
"""
Args:
treatmentCol: treatment column
"""
self._set(treatmentCol=value)
return self
[docs] def setTreatmentModel(self, value):
"""
Args:
treatmentModel: treatment model to run
"""
self._set(treatmentModel=value)
return self
[docs] def setTreatmentResidualCol(self, value):
"""
Args:
treatmentResidualCol: Treatment Residual Column
"""
self._set(treatmentResidualCol=value)
return self
[docs] def setWeightCol(self, value):
"""
Args:
weightCol: The name of the weight column
"""
self._set(weightCol=value)
return self
[docs] def getConfidenceLevel(self):
"""
Returns:
confidenceLevel: confidence level, default value is 0.975
"""
return self.getOrDefault(self.confidenceLevel)
[docs] def getConfounderVecCol(self):
"""
Returns:
confounderVecCol: Confounders to control for
"""
return self.getOrDefault(self.confounderVecCol)
[docs] def getFeaturesCol(self):
"""
Returns:
featuresCol: The name of the features column
"""
return self.getOrDefault(self.featuresCol)
[docs] def getHeterogeneityVecCol(self):
"""
Returns:
heterogeneityVecCol: Vector to divide the treatment by
"""
return self.getOrDefault(self.heterogeneityVecCol)
[docs] def getMaxDepth(self):
"""
Returns:
maxDepth: Max Depth of Tree
"""
return self.getOrDefault(self.maxDepth)
[docs] def getMaxIter(self):
"""
Returns:
maxIter: maximum number of iterations (>= 0)
"""
return self.getOrDefault(self.maxIter)
[docs] def getMinSamplesLeaf(self):
"""
Returns:
minSamplesLeaf: Max Depth of Tree
"""
return self.getOrDefault(self.minSamplesLeaf)
[docs] def getNumTrees(self):
"""
Returns:
numTrees: Number of trees
"""
return self.getOrDefault(self.numTrees)
[docs] def getOutcomeCol(self):
"""
Returns:
outcomeCol: outcome column
"""
return self.getOrDefault(self.outcomeCol)
[docs] def getOutcomeModel(self):
"""
Returns:
outcomeModel: outcome model to run
"""
return JavaParams._from_java(self._java_obj.getOutcomeModel())
[docs] def getOutcomeResidualCol(self):
"""
Returns:
outcomeResidualCol: Outcome Residual Column
"""
return self.getOrDefault(self.outcomeResidualCol)
[docs] def getOutputCol(self):
"""
Returns:
outputCol: The name of the output column
"""
return self.getOrDefault(self.outputCol)
[docs] def getOutputHighCol(self):
"""
Returns:
outputHighCol: Output Confidence Interval Low
"""
return self.getOrDefault(self.outputHighCol)
[docs] def getOutputLowCol(self):
"""
Returns:
outputLowCol: Output Confidence Interval Low
"""
return self.getOrDefault(self.outputLowCol)
[docs] def getParallelism(self):
"""
Returns:
parallelism: the number of threads to use when running parallel algorithms
"""
return self.getOrDefault(self.parallelism)
[docs] def getSampleSplitRatio(self):
"""
Returns:
sampleSplitRatio: Sample split ratio for cross-fitting. Default: [0.5, 0.5].
"""
return self.getOrDefault(self.sampleSplitRatio)
[docs] def getTreatmentCol(self):
"""
Returns:
treatmentCol: treatment column
"""
return self.getOrDefault(self.treatmentCol)
[docs] def getTreatmentModel(self):
"""
Returns:
treatmentModel: treatment model to run
"""
return JavaParams._from_java(self._java_obj.getTreatmentModel())
[docs] def getTreatmentResidualCol(self):
"""
Returns:
treatmentResidualCol: Treatment Residual Column
"""
return self.getOrDefault(self.treatmentResidualCol)
[docs] def getWeightCol(self):
"""
Returns:
weightCol: The name of the weight column
"""
return self.getOrDefault(self.weightCol)
def _create_model(self, java_model):
try:
model = OrthoForestDMLModel(java_obj=java_model)
model._transfer_params_from_java()
except TypeError:
model = OrthoForestDMLModel._from_java(java_model)
return model
def _fit(self, dataset):
java_model = self._fit_java(dataset)
return self._create_model(java_model)