com.hankcs.hanlp.classification.features

Class BaseFeatureData

java.lang.Object

com.hankcs.hanlp.classification.features.BaseFeatureData

Direct Known Subclasses:

DfFeatureData

public class BaseFeatureData
extends Object

储存所有必需的统计数据,尽量不要存太多东西在这里,因为多个分类器都用这个结构,所以里面的数据仅保留必需的数据

Field Summary

Fields

Modifier and Type	Field and Description
int[]	categoryCounts 每个类目中的文档数量
int[][]	featureCategoryJointCount 一个特征在类目中分别出现几次(键是特征,值的键是类目)
int	n 样本数量
BinTrie<Integer>	wordIdTrie 新的特征映射表

Constructor Summary

Constructors

Constructor and Description
BaseFeatureData(IDataSet dataSet) 构造一个空白的统计对象

Method Summary

Methods inherited from class java.lang.Object

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

Field Detail

n

public int n

样本数量

featureCategoryJointCount

public int[][] featureCategoryJointCount

一个特征在类目中分别出现几次(键是特征,值的键是类目)

categoryCounts

public int[] categoryCounts

每个类目中的文档数量

wordIdTrie

public BinTrie<Integer> wordIdTrie

新的特征映射表

Constructor Detail

BaseFeatureData

public BaseFeatureData(IDataSet dataSet)

构造一个空白的统计对象