ID3 PYTHON实现，离散特征值

from math import log

import numpy as np
import pandas as pd

gMinSubsetSize = 1

class ID3Node:
    def __init__(self,pType=None,pFeature=None,pData=None,pGain=None,pParent=None):
        self.type = pType
        self.data = pData
        self.splitFeature = pFeature
        self.gain = pGain
        self.parent = pParent
        self.childs = []

def calcEntropy(pDataset,pFeature):
    dataset = pd.DataFrame(pDataset)
    classInfo = dataset[pFeature].value_counts()
    '''
    classLabel = classLabel.drop_duplicates()
    totalSamples = dataset.shape[0]