import pandas as pd
import numpy as np
import scipy.stats as ss
import matplotlib.pyplot as plt
import seaborn as sns

sns.set_context(font_scale=1.5)
# df = pd.read_csv("./data/HR.csv")
# sns.heatmap(df.corr(),vmin=-1,vmax=1,cmap=sns.color_palette("RdBu",n_colors=128))
# plt.show()

s1 = pd.Series(["X1","X1","X2","X2","X2","X2"])
s2 = pd.Series(["Y1","Y1","Y1","Y2","Y2","Y2"])

def getEntropy(s):
   #先判断一下是否是一个Series
   if not isinstance(s,pd.core.series.Series):#如果不是的话就转换成
       s = pd.Series(s)
   #知道熵,就要了解其分布 除以数量,获取其概率分布
   prt_ary = pd.groupby(s,by=s).count().values/float(len(s))
   #以2为底求对数 * 概率 然后求和 就可以得到熵
   return -(np.log2(prt_ary)*prt_ary).sum()

print("Entropy:",getEntropy(s1))
#0.5 0.5的分布 熵就是1
print("Entropy:",getEntropy(s2))

def getCondEntropy(s1,s2):
   d=dict()
   #求出分布
   for i in list(range(len(s1))):
       #准备了一个结构体 value 是一个数组 s1值之下 s2 的分布
       d[s1[i]]=d.get(s1[i],[])+[s2[i]]
   return sum([getEntropy(d[k]) * len(d[k])/float(len(s1)) for k in d])

print("CondEntropy:" + getCondEntropy(s1,s2))

Traceback (most recent call last):

  File "C:/Users/5838/Untitled Folder/pycharm/cov.py", line 36, in <module>

    print("CondEntropy:" + getCondEntropy(s1,s2))

TypeError: ufunc 'add' did not contain a loop with signature matching types dtype('<U32') dtype('<U32') dtype('<U32')