LightGBMRegressor

class LightGBMRegressor.LightGBMRegressor(alpha=0.9, application='regression', baggingFraction=1.0, baggingFreq=0, baggingSeed=3, defaultListenPort=12400, earlyStoppingRound=0, featureFraction=1.0, featuresCol='features', labelCol='label', learningRate=0.1, maxBin=255, maxDepth=-1, minSumHessianInLeaf=0.001, numIterations=100, numLeaves=31, parallelism='data_parallel', predictionCol='prediction', timeout=120.0)

Bases: mmlspark.Utils.ComplexParamsMixin, pyspark.ml.util.JavaMLReadable, pyspark.ml.util.JavaMLWritable, pyspark.ml.wrapper.JavaEstimator

Parameters:

• alpha (double) – parameter for Huber loss and Quantile regression (default: 0.9)
• application (str) – Regression application, regression_l2, regression_l1, huber, fair, poisson, quantile, mape, gamma or tweedie (default: regression)
• baggingFraction (double) – Bagging fraction (default: 1.0)
• baggingFreq (int) – Bagging frequence (default: 0)
• baggingSeed (int) – Bagging seed (default: 3)
• defaultListenPort (int) – The default listen port on executors, used for testing (default: 12400)
• earlyStoppingRound (int) – Early stopping round (default: 0)
• featureFraction (double) – Feature fraction (default: 1.0)
• featuresCol (str) – features column name (default: features)
• labelCol (str) – label column name (default: label)
• learningRate (double) – Learning rate or shrinkage rate (default: 0.1)
• maxBin (int) – Max bin (default: 255)
• maxDepth (int) – Max depth (default: -1)
• minSumHessianInLeaf (double) – Minimal sum hessian in one leaf (default: 0.001)
• numIterations (int) – Number of iterations, LightGBM constructs num_class * num_iterations trees (default: 100)
• numLeaves (int) – Number of leaves (default: 31)
• parallelism (str) – Tree learner parallelism, can be set to data_parallel or voting_parallel (default: data_parallel)
• predictionCol (str) – prediction column name (default: prediction)
• timeout (double) – Timeout in seconds (default: 120.0)

getAlpha()

Returns:parameter for Huber loss and Quantile regression (default: 0.9) Return type:double

getApplication()

Returns:Regression application, regression_l2, regression_l1, huber, fair, poisson, quantile, mape, gamma or tweedie (default: regression) Return type:str

getBaggingFraction()

Returns:Bagging fraction (default: 1.0) Return type:double

getBaggingFreq()

Returns:Bagging frequence (default: 0) Return type:int

getBaggingSeed()

Returns:Bagging seed (default: 3) Return type:int

getDefaultListenPort()

Returns:The default listen port on executors, used for testing (default: 12400) Return type:int

getEarlyStoppingRound()

Returns:Early stopping round (default: 0) Return type:int

getFeatureFraction()

Returns:Feature fraction (default: 1.0) Return type:double

getFeaturesCol()

Returns:features column name (default: features) Return type:str

static getJavaPackage()

Returns package name String.

getLabelCol()

Returns:label column name (default: label) Return type:str

getLearningRate()

Returns:Learning rate or shrinkage rate (default: 0.1) Return type:double

getMaxBin()

Returns:Max bin (default: 255) Return type:int

getMaxDepth()

Returns:Max depth (default: -1) Return type:int

getMinSumHessianInLeaf()

Returns:Minimal sum hessian in one leaf (default: 0.001) Return type:double

getNumIterations()

Returns:Number of iterations, LightGBM constructs num_class * num_iterations trees (default: 100) Return type:int

getNumLeaves()

Returns:Number of leaves (default: 31) Return type:int

getParallelism()

Returns:Tree learner parallelism, can be set to data_parallel or voting_parallel (default: data_parallel) Return type:str

getPredictionCol()

Returns:prediction column name (default: prediction) Return type:str

getTimeout()

Returns:Timeout in seconds (default: 120.0) Return type:double

classmethod read()

Returns an MLReader instance for this class.

setAlpha(value)

Parameters:alpha (double) – parameter for Huber loss and Quantile regression (default: 0.9)

setApplication(value)

Parameters:application (str) – Regression application, regression_l2, regression_l1, huber, fair, poisson, quantile, mape, gamma or tweedie (default: regression)

setBaggingFraction(value)

Parameters:baggingFraction (double) – Bagging fraction (default: 1.0)

setBaggingFreq(value)

Parameters:baggingFreq (int) – Bagging frequence (default: 0)

setBaggingSeed(value)

Parameters:baggingSeed (int) – Bagging seed (default: 3)

setDefaultListenPort(value)

Parameters:defaultListenPort (int) – The default listen port on executors, used for testing (default: 12400)

setEarlyStoppingRound(value)

Parameters:earlyStoppingRound (int) – Early stopping round (default: 0)

setFeatureFraction(value)

Parameters:featureFraction (double) – Feature fraction (default: 1.0)

setFeaturesCol(value)

Parameters:featuresCol (str) – features column name (default: features)

setLabelCol(value)

Parameters:labelCol (str) – label column name (default: label)

setLearningRate(value)

Parameters:learningRate (double) – Learning rate or shrinkage rate (default: 0.1)

setMaxBin(value)

Parameters:maxBin (int) – Max bin (default: 255)

setMaxDepth(value)

Parameters:maxDepth (int) – Max depth (default: -1)

setMinSumHessianInLeaf(value)

Parameters:minSumHessianInLeaf (double) – Minimal sum hessian in one leaf (default: 0.001)

setNumIterations(value)

Parameters:numIterations (int) – Number of iterations, LightGBM constructs num_class * num_iterations trees (default: 100)

setNumLeaves(value)

Parameters:numLeaves (int) – Number of leaves (default: 31)

setParallelism(value)

Parameters:parallelism (str) – Tree learner parallelism, can be set to data_parallel or voting_parallel (default: data_parallel)

setParams(alpha=0.9, application='regression', baggingFraction=1.0, baggingFreq=0, baggingSeed=3, defaultListenPort=12400, earlyStoppingRound=0, featureFraction=1.0, featuresCol='features', labelCol='label', learningRate=0.1, maxBin=255, maxDepth=-1, minSumHessianInLeaf=0.001, numIterations=100, numLeaves=31, parallelism='data_parallel', predictionCol='prediction', timeout=120.0)

Set the (keyword only) parameters

Parameters:

• alpha (double) – parameter for Huber loss and Quantile regression (default: 0.9)
• application (str) – Regression application, regression_l2, regression_l1, huber, fair, poisson, quantile, mape, gamma or tweedie (default: regression)
• baggingFraction (double) – Bagging fraction (default: 1.0)
• baggingFreq (int) – Bagging frequence (default: 0)
• baggingSeed (int) – Bagging seed (default: 3)
• defaultListenPort (int) – The default listen port on executors, used for testing (default: 12400)
• earlyStoppingRound (int) – Early stopping round (default: 0)
• featureFraction (double) – Feature fraction (default: 1.0)
• featuresCol (str) – features column name (default: features)
• labelCol (str) – label column name (default: label)
• learningRate (double) – Learning rate or shrinkage rate (default: 0.1)
• maxBin (int) – Max bin (default: 255)
• maxDepth (int) – Max depth (default: -1)
• minSumHessianInLeaf (double) – Minimal sum hessian in one leaf (default: 0.001)
• numIterations (int) – Number of iterations, LightGBM constructs num_class * num_iterations trees (default: 100)
• numLeaves (int) – Number of leaves (default: 31)
• parallelism (str) – Tree learner parallelism, can be set to data_parallel or voting_parallel (default: data_parallel)
• predictionCol (str) – prediction column name (default: prediction)
• timeout (double) – Timeout in seconds (default: 120.0)

setPredictionCol(value)

Parameters:predictionCol (str) – prediction column name (default: prediction)

setTimeout(value)

Parameters:timeout (double) – Timeout in seconds (default: 120.0)

class LightGBMRegressor.M(java_model=None)

Bases: mmlspark.Utils.ComplexParamsMixin, pyspark.ml.wrapper.JavaModel, pyspark.ml.util.JavaMLWritable, pyspark.ml.util.JavaMLReadable

Model fitted by LightGBMRegressor.

This class is left empty on purpose. All necessary methods are exposed through inheritance.

static getJavaPackage()

Returns package name String.

classmethod read()

Returns an MLReader instance for this class.