求熵 python 代码

#coding=gbk

import nltk

import math

def entropy(labels):

    freqdist = nltk.FreqDist(labels)  #Frequency Distributions

    probs = [freqdist.freq(l) for l in freqdist] #freqdist.freq(l)  是将频率转换成概率。即频率除以总个数等于概率

    return -sum([p * math.log(p,2) for p in probs])



print entropy(["u","m","m","u"]) #结果为1

print entropy(["m","m","m","m"])  #结果为-0.0