In [14]:

    

import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
import seaborn as sns
from collections import Counter

import matplotlib
%matplotlib inline


    

In [15]:

    

data = pd.read_csv("../data/vgsales.csv", warn_bad_lines = True, error_bad_lines = False, verbose = True)
data.drop(data.columns[[0, 11]], axis = 1, inplace = True)
data["Year"] = data["Year"].astype('category')

plt.scatter(data['NA_Sales'], data['JP_Sales'])


    

    




Tokenization took: 10.30 ms
Type conversion took: 10.86 ms
Parser memory cleanup took: 0.00 ms






    
Out[15]:





<matplotlib.collections.PathCollection at 0x7fe36c5053d0>






    

In [16]:

    

pd.scatter_matrix(data, diagonal = 'kde', color = 'k', alpha = 1, figsize=(8,8))


    

    
Out[16]:





array([[<matplotlib.axes._subplots.AxesSubplot object at 0x7fe3691e7050>,
        <matplotlib.axes._subplots.AxesSubplot object at 0x7fe369174e50>,
        <matplotlib.axes._subplots.AxesSubplot object at 0x7fe3690f8dd0>,
        <matplotlib.axes._subplots.AxesSubplot object at 0x7fe369067510>,
        <matplotlib.axes._subplots.AxesSubplot object at 0x7fe368fe8590>],
       [<matplotlib.axes._subplots.AxesSubplot object at 0x7fe369080e90>,
        <matplotlib.axes._subplots.AxesSubplot object at 0x7fe368ed47d0>,
        <matplotlib.axes._subplots.AxesSubplot object at 0x7fe368e57750>,
        <matplotlib.axes._subplots.AxesSubplot object at 0x7fe368e451d0>,
        <matplotlib.axes._subplots.AxesSubplot object at 0x7fe368dc7250>],
       [<matplotlib.axes._subplots.AxesSubplot object at 0x7fe368d2a110>,
        <matplotlib.axes._subplots.AxesSubplot object at 0x7fe368cac490>,
        <matplotlib.axes._subplots.AxesSubplot object at 0x7fe368c2c410>,
        <matplotlib.axes._subplots.AxesSubplot object at 0x7fe368b993d0>,
        <matplotlib.axes._subplots.AxesSubplot object at 0x7fe368b1c690>],
       [<matplotlib.axes._subplots.AxesSubplot object at 0x7fe368aff650>,
        <matplotlib.axes._subplots.AxesSubplot object at 0x7fe368a7ebd0>,
        <matplotlib.axes._subplots.AxesSubplot object at 0x7fe368aaadd0>,
        <matplotlib.axes._subplots.AxesSubplot object at 0x7fe36896d910>,
        <matplotlib.axes._subplots.AxesSubplot object at 0x7fe3688efbd0>],
       [<matplotlib.axes._subplots.AxesSubplot object at 0x7fe36885c450>,
        <matplotlib.axes._subplots.AxesSubplot object at 0x7fe3687de4d0>,
        <matplotlib.axes._subplots.AxesSubplot object at 0x7fe3687c2290>,
        <matplotlib.axes._subplots.AxesSubplot object at 0x7fe368742610>,
        <matplotlib.axes._subplots.AxesSubplot object at 0x7fe3686c5590>]], dtype=object)






    

In [17]:

    

data.corr(method = 'pearson')


    

    
Out[17]:






  

    

      

      
NA_Sales
      
EU_Sales
      
JP_Sales
      
Other_Sales
      
Global_Sales
    
  
  

    

      
NA_Sales
      
1.000000
      
0.767727
      
0.449788
      
0.634737
      
0.941047
    
    

      
EU_Sales
      
0.767727
      
1.000000
      
0.435585
      
0.726385
      
0.902836
    
    

      
JP_Sales
      
0.449788
      
0.435585
      
1.000000
      
0.290187
      
0.611816
    
    

      
Other_Sales
      
0.634737
      
0.726385
      
0.290187
      
1.000000
      
0.748331
    
    

      
Global_Sales
      
0.941047
      
0.902836
      
0.611816
      
0.748331
      
1.000000
    
  

In [20]:

    

from scipy.stats import norm, entropy

def mutual_info(x, y, bins=10):
    counts_xy, bins_x, bins_y = np.histogram2d(x, y, bins=(bins, bins))
    counts_x, bins = np.histogram(x, bins=bins)
    counts_y, bins = np.histogram(y, bins=bins)

    counts_xy += 1
    counts_x += 1
    counts_y += 1
    P_xy = counts_xy / np.sum(counts_xy, dtype=float)
    P_x = counts_x / np.sum(counts_x, dtype=float)
    P_y = counts_y / np.sum(counts_y, dtype=float)

    I_xy = np.sum(P_xy * np.log2(P_xy / (P_x.reshape(-1, 1) * P_y)))

    return I_xy / (entropy(counts_x) + entropy(counts_y))


val = data._get_numeric_data().values
col = data._get_numeric_data().columns

for i in range(len(col)):
    for j in range(i):
        ni = mutual_info(val[:][i], val[:][j])
        print "NI between %s and %s is: %f" % (col[i], col[j], ni)


    

    




NI between EU_Sales and NA_Sales is: 0.042220
NI between JP_Sales and NA_Sales is: 0.043833
NI between JP_Sales and EU_Sales is: 0.037945
NI between Other_Sales and NA_Sales is: 0.043833
NI between Other_Sales and EU_Sales is: 0.040083
NI between Other_Sales and JP_Sales is: 0.051139
NI between Global_Sales and NA_Sales is: 0.040083
NI between Global_Sales and EU_Sales is: 0.037945
NI between Global_Sales and JP_Sales is: 0.049511
NI between Global_Sales and Other_Sales is: 0.049511