org.apache.spark.mllib.tree.configuration
Class Strategy

Object
  org.apache.spark.mllib.tree.configuration.Strategy

All Implemented Interfaces:

java.io.Serializable

public class Strategy
extends Object
implements scala.Serializable

:: Experimental :: Stores all the configuration options for tree construction param: algo Learning goal. Supported: org.apache.spark.mllib.tree.configuration.Algo.Classification, org.apache.spark.mllib.tree.configuration.Algo.Regression param: impurity Criterion used for information gain calculation. Supported for Classification: Gini, Entropy. Supported for Regression: Variance. param: maxDepth Maximum depth of the tree. E.g., depth 0 means 1 leaf node; depth 1 means 1 internal node + 2 leaf nodes. param: numClasses Number of classes for classification. (Ignored for regression.) Default value is 2 (binary classification). param: maxBins Maximum number of bins used for discretizing continuous features and for choosing how to split on features at each node. More bins give higher granularity. param: quantileCalculationStrategy Algorithm for calculating quantiles. Supported: org.apache.spark.mllib.tree.configuration.QuantileStrategy.Sort param: categoricalFeaturesInfo A map storing information about the categorical variables and the number of discrete values they take. For example, an entry (n -> k) implies the feature n is categorical with k categories 0, 1, 2, ... , k-1. It's important to note that features are zero-indexed. param: minInstancesPerNode Minimum number of instances each child must have after split. Default value is 1. If a split cause left or right child to have less than minInstancesPerNode, this split will not be considered as a valid split. param: minInfoGain Minimum information gain a split must get. Default value is 0.0. If a split has less information gain than minInfoGain, this split will not be considered as a valid split. param: maxMemoryInMB Maximum memory in MB allocated to histogram aggregation. Default value is 256 MB. param: subsamplingRate Fraction of the training data used for learning decision tree. param: useNodeIdCache If this is true, instead of passing trees to executors, the algorithm will maintain a separate RDD of node Id cache for each row. param: checkpointInterval How often to checkpoint when the node Id cache gets updated. E.g. 10 means that the cache will get checkpointed every 10 updates. If the checkpoint directory is not set in SparkContext, this setting is ignored.

See Also:

Serialized Form

Constructor Summary

Strategy(scala.Enumeration.Value algo, Impurity impurity, int maxDepth, int numClasses, int maxBins, scala.Enumeration.Value quantileCalculationStrategy, scala.collection.immutable.Map<Object,Object> categoricalFeaturesInfo, int minInstancesPerNode, double minInfoGain, int maxMemoryInMB, double subsamplingRate, boolean useNodeIdCache, int checkpointInterval)

Strategy(scala.Enumeration.Value algo, Impurity impurity, int maxDepth, int numClasses, int maxBins, java.util.Map<Integer,Integer> categoricalFeaturesInfo)
Java-friendly constructor for Strategy

Method Summary

scala.Enumeration.Value	algo()
scala.collection.immutable.Map<Object,Object>	categoricalFeaturesInfo()
int	checkpointInterval()
Strategy	copy() Returns a shallow copy of this instance.
static Strategy	defaultStategy(scala.Enumeration.Value algo) Construct a default set of parameters for DecisionTree
static Strategy	defaultStrategy(String algo) Construct a default set of parameters for DecisionTree
scala.Enumeration.Value	getAlgo()
scala.collection.immutable.Map<Object,Object>	getCategoricalFeaturesInfo()
int	getCheckpointInterval()
Impurity	getImpurity()
int	getMaxBins()
int	getMaxDepth()
int	getMaxMemoryInMB()
double	getMinInfoGain()
int	getMinInstancesPerNode()
int	getNumClasses()
scala.Enumeration.Value	getQuantileCalculationStrategy()
double	getSubsamplingRate()
boolean	getUseNodeIdCache()
Impurity	impurity()
boolean	isMulticlassClassification()
boolean	isMulticlassWithCategoricalFeatures()
int	maxBins()
int	maxDepth()
int	maxMemoryInMB()
double	minInfoGain()
int	minInstancesPerNode()
int	numClasses()
scala.Enumeration.Value	quantileCalculationStrategy()
void	setAlgo(String algo) Sets Algorithm using a String.
void	setCategoricalFeaturesInfo(java.util.Map<Integer,Integer> categoricalFeaturesInfo) Sets categoricalFeaturesInfo using a Java Map.
void	setCheckpointInterval(int x$1)
void	setImpurity(Impurity x$1)
void	setMaxBins(int x$1)
void	setMaxDepth(int x$1)
void	setMaxMemoryInMB(int x$1)
void	setMinInfoGain(double x$1)
void	setMinInstancesPerNode(int x$1)
void	setNumClasses(int x$1)
void	setQuantileCalculationStrategy(scala.Enumeration.Value x$1)
void	setSubsamplingRate(double x$1)
void	setUseNodeIdCache(boolean x$1)
double	subsamplingRate()
boolean	useNodeIdCache()

Methods inherited from class Object

equals, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Constructor Detail

Strategy

public Strategy(scala.Enumeration.Value algo,
                Impurity impurity,
                int maxDepth,
                int numClasses,
                int maxBins,
                scala.Enumeration.Value quantileCalculationStrategy,
                scala.collection.immutable.Map<Object,Object> categoricalFeaturesInfo,
                int minInstancesPerNode,
                double minInfoGain,
                int maxMemoryInMB,
                double subsamplingRate,
                boolean useNodeIdCache,
                int checkpointInterval)

Strategy

public Strategy(scala.Enumeration.Value algo,
                Impurity impurity,
                int maxDepth,
                int numClasses,
                int maxBins,
                java.util.Map<Integer,Integer> categoricalFeaturesInfo)

Java-friendly constructor for Strategy

Parameters:

algo - (undocumented)

impurity - (undocumented)

maxDepth - (undocumented)

numClasses - (undocumented)

maxBins - (undocumented)

categoricalFeaturesInfo - (undocumented)

Method Detail

defaultStrategy

public static Strategy defaultStrategy(String algo)

Construct a default set of parameters for DecisionTree

Parameters:

algo - "Classification" or "Regression"

Returns:

(undocumented)

defaultStategy

public static Strategy defaultStategy(scala.Enumeration.Value algo)

Construct a default set of parameters for DecisionTree

Parameters:

algo - Algo.Classification or Algo.Regression

Returns:

(undocumented)

algo

public scala.Enumeration.Value algo()

impurity

public Impurity impurity()

setImpurity

public void setImpurity(Impurity x$1)

maxDepth

public int maxDepth()

setMaxDepth

public void setMaxDepth(int x$1)

numClasses

public int numClasses()

setNumClasses

public void setNumClasses(int x$1)

maxBins

public int maxBins()

setMaxBins

public void setMaxBins(int x$1)

quantileCalculationStrategy

public scala.Enumeration.Value quantileCalculationStrategy()

setQuantileCalculationStrategy

public void setQuantileCalculationStrategy(scala.Enumeration.Value x$1)

categoricalFeaturesInfo

public scala.collection.immutable.Map<Object,Object> categoricalFeaturesInfo()

minInstancesPerNode

public int minInstancesPerNode()

setMinInstancesPerNode

public void setMinInstancesPerNode(int x$1)

minInfoGain

public double minInfoGain()

setMinInfoGain

public void setMinInfoGain(double x$1)

maxMemoryInMB

public int maxMemoryInMB()

setMaxMemoryInMB

public void setMaxMemoryInMB(int x$1)

subsamplingRate

public double subsamplingRate()

setSubsamplingRate

public void setSubsamplingRate(double x$1)

useNodeIdCache

public boolean useNodeIdCache()

setUseNodeIdCache

public void setUseNodeIdCache(boolean x$1)

checkpointInterval

public int checkpointInterval()

setCheckpointInterval

public void setCheckpointInterval(int x$1)

isMulticlassClassification

public boolean isMulticlassClassification()

isMulticlassWithCategoricalFeatures

public boolean isMulticlassWithCategoricalFeatures()

setAlgo

public void setAlgo(String algo)

Sets Algorithm using a String.

Parameters:

algo - (undocumented)

setCategoricalFeaturesInfo

public void setCategoricalFeaturesInfo(java.util.Map<Integer,Integer> categoricalFeaturesInfo)

Sets categoricalFeaturesInfo using a Java Map.

Parameters:

categoricalFeaturesInfo - (undocumented)

copy

public Strategy copy()

Returns a shallow copy of this instance.

getAlgo

public scala.Enumeration.Value getAlgo()

getImpurity

public Impurity getImpurity()

getMaxDepth

public int getMaxDepth()

getNumClasses

public int getNumClasses()

getMaxBins

public int getMaxBins()

getQuantileCalculationStrategy

public scala.Enumeration.Value getQuantileCalculationStrategy()

getCategoricalFeaturesInfo

public scala.collection.immutable.Map<Object,Object> getCategoricalFeaturesInfo()

getMinInstancesPerNode

public int getMinInstancesPerNode()

getMinInfoGain

public double getMinInfoGain()

getMaxMemoryInMB

public int getMaxMemoryInMB()

getSubsamplingRate

public double getSubsamplingRate()

getUseNodeIdCache

public boolean getUseNodeIdCache()

getCheckpointInterval

public int getCheckpointInterval()