MLJ: Machine Learning in Java API Manual: Class SplitScore

Overview

Package

Class

Tree

Deprecated

Index

Help

PREV CLASS NEXT CLASS

FRAMES NO FRAMES

SUMMARY: INNER | FIELD | CONSTR | METHOD

DETAIL: FIELD | CONSTR | METHOD

shared
Class SplitScore

java.lang.Object
  |
  +--shared.SplitScore

Direct Known Subclasses:: SplitAttr

public class SplitScore
extends java.lang.Object

A class for determining, holding, and returning the information associated with an attribute split. Uses methods from the Entropy class for the determination of the scores. The scores for a specific item are cached in an internal cache structure. This structure contains the entropy, conditional entropy, split entropy, mutual information, gain ratio, split distribution, label distribution, and the number of instances that these scores have been evaluated on. Access to scores that have been cached are constant time, while items that need refreshment have a longer access time for recalculation.

Field Summary

static byte defaultSplitScoreCriterion
          Default criterion for determining the score of a particular split.

static byte externalScore
          Value indicating how splits are scored.

static byte gainRatio
          Value indicating how splits are scored.

protected LogOptions logOptions
          Logging options for this class.

static byte mutualInfo
          Value indicating how splits are scored.

static byte mutualInfoRatio
          Value indicating how splits are scored.

static byte normalizedMutualInfo
          Value indicating how splits are scored.

static java.lang.String[] splitScoreCriterionEnum
          String names for each form of split criterion.

Constructor Summary

SplitScore()
          Constructor.

SplitScore(SplitScore source)
          Copy Constructor.

Method Summary

SplitScore assign(SplitScore rhs)
          Assigns the given SplitScore data to this SplitScore.

void display()
          Produces formatted display of contents of object.

void display(java.io.Writer stream)
          Produces formatted display of contents of object.

void display(java.io.Writer stream, DisplayPref dp)
          Produces formatted display of contents of object.

double get_cond_entropy()
          Returns cache.condEntropy, first checking to see if it has yet been set.

double get_entropy()
          Returns cache.entropy, first checking to see if it has yet been set.

double get_external_score()
          Returns the value, set externally, for the score.

double get_gain_ratio()
          Determines, and returns, cache.gainRatio.

double[] get_label_dist()
          The label distribution is calculated from the split and label distribution.

int get_log_level()
          Returns the logging level for this object.

LogOptions get_log_options()
          Returns the LogOptions object for this object.

java.io.Writer get_log_stream()
          Returns the stream to which logs for this object are written.

double get_mutual_info_ratio()
          Returns the mutual information ratio, which is the ratio between the mutual info and entropy.

double get_mutual_info(boolean normalize)
          Returns the mutual info (information gain) score for the split this SplitScore object represents.

double[][] get_split_and_label_dist()
          Returns a reference to the requested distribution array.

double[] get_split_dist()
          The split distribution is calculated from the split and label distribution.

double get_split_entropy()
          Returns cache.splitEntropy, first checking to see if it has yet been set.

byte get_split_score_criterion()
          Returns the type of criterion used in scoring splits.

double get_unnormalized_mutual_info()
          Returns the mutual info (information gain) score for the split this SplitScore object represents.

boolean has_distribution()
          Checks if there exists a splitAndLabel distribution.

boolean has_distribution(boolean fatalOnFalse)
          Checks if there exists a splitAndLabel distribution.

boolean has_external_score()
          Checks if an external score has been set.

double normalize_by_num_splits(double score)
          Normalize by the number of splits.

int num_splits()
          Returns the number of splits--not including unknowns.

double[] release_label_dist()
          Returns the label distribution array and releases ownership.

double[][] release_split_and_label_dist()
          Returns the split and label distribution array and releases ownership.

double[] release_split_dist()
          Returns the split distribution array and releases ownership.

void reset()
          Clear (delete) distribution array data.

double score()
          The criterion calculation depends on the score criterion.

double score(double[][] sAndLDist, double[] sDist, double[] lDist, double passedEntropy, double passedWeight)
          Computes the scores and updates the cache when there are being computed many times for the same number of instances and entropy.

void set_external_score(double extScore)
          Set the external score.

void set_log_level(int level)
          Sets the logging level for this object.

void set_log_options(LogOptions opt)
          Sets the LogOptions object for this object.

void set_log_prefixes(java.lang.String file, int line, int lvl1, int lvl2)
          Sets the logging message prefix for this object.

void set_log_stream(java.io.Writer strm)
          Sets the stream to which logging options are displayed.

double[][] set_split_and_label_dist(double[][] sAndLDist)
          Stores the splitAndLabelDist array.

void set_split_dist(double[] sDist)
          Stores the cache.splitDist array.

void set_split_score_criterion(byte choice)
          Sets the split score criterion.

double total_weight()
          Returns the total weight from the cache.

Methods inherited from class java.lang.Object

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

Field Detail