从 pandas dataframe 创建相对熵矩阵

Question

I have a dataframe of values, say:

df = pd.DataFrame(np.array([[0.2, 0.5, 0.3], [0.1, 0.2, 0.5], [0.4, 0.3, 0.3]]),
                   columns=['a', 'b', 'c'])

in which every row is a vector of probabilities. I want to compute something like the correlation matrix of df.corr() , but instead of correlation, I want to compute the relative entropy .

What is the best way to do this, as I can't find a way to get inside the .corr() method and simply change the function it uses?

Answer 1

IIUC, use .corr as follows:

import pandas as pd
import numpy as np

from scipy.stats import entropy

df = pd.DataFrame(np.array([[0.2, 0.5, 0.3], [0.1, 0.2, 0.5], [0.4, 0.3, 0.3]]),
                   columns=['a', 'b', 'c'])

res = df.corr(method=entropy)
print(res)

Output

          a         b         c
a  1.000000  0.160246  0.270608
b  0.160246  1.000000  0.167465
c  0.270608  0.167465  1.000000

From the documentation:

callable: callable with input two 1d ndarrays and returning a float. Note that the returned matrix from corr will have 1 along the diagonals and will be symmetric regardless of the callable's behavior.