Home - News ( United States | United Kingdom | Italy | Germany ) - Football scores

Showing content from https://stackoverflow.com/questions/14964493/multiindex-based-indexing-in-pandas below:

MultiIndex-based indexing in pandas - Stack Overflow

This question already has answers here:

Closed 6 years ago.

If I define a hierarchically-indexed dataframe like this:

import itertools
import pandas as pd
import numpy as np
a = ('A', 'B')
i = (0, 1, 2)
b = (True, False)
idx = pd.MultiIndex.from_tuples(list(itertools.product(a, i, b)),
                                names=('Alpha', 'Int', 'Bool'))
df = pd.DataFrame(np.random.randn(len(idx), 7), index=idx,
                  columns=('I', 'II', 'III', 'IV', 'V', 'VI', 'VII'))

the contents look like this:

In [19]: df
Out[19]: 
                        I        II       III        IV         V        VI       VII
Alpha Int Bool                                                                       
A     0   True  -0.462924  1.210442  0.306737  0.325116 -1.320084 -0.831699  0.892865
          False -0.850570 -0.949779  0.022074 -0.205575 -0.684794 -0.214307 -1.133833
      1   True   0.603602  1.387020 -0.830780 -1.242000 -0.321938  0.484271  0.171738
          False -1.591730  1.282136  0.095159 -1.239882  0.760880 -0.606444 -0.485957
      2   True  -1.346883  1.650247 -1.476443  2.092067  1.344689  0.177083  0.100844
          False  0.001407 -1.127299 -0.417828  0.143595 -0.277838 -0.478262 -0.350906
B     0   True   0.722781 -1.093182  0.237536  0.457614 -2.500885  0.338257  0.009128
          False  0.321022  0.419357  1.161140 -1.371035  1.093696  0.250517 -1.125612
      1   True   0.237441  1.739933  0.029653  0.327823 -0.384647  1.523628 -0.009053
          False -0.459148 -0.598577 -0.593486 -0.607447  1.478399  0.504028 -0.329555
      2   True  -0.583052 -0.986493 -0.057788 -0.639798  1.400311  0.076471 -0.212513
          False  0.896755  2.583520  1.520151  2.367336 -1.084994 -1.233548 -2.414215

I know how to extract the data corresponding to a given column. E.g. for column 'VII':

In [20]: df['VII']
Out[20]: 
Alpha  Int  Bool 
A      0    True     0.892865
            False   -1.133833
       1    True     0.171738
            False   -0.485957
       2    True     0.100844
            False   -0.350906
B      0    True     0.009128
            False   -1.125612
       1    True    -0.009053
            False   -0.329555
       2    True    -0.212513
            False   -2.414215
Name: VII

How do I extract the data matching the following sets of criteria:

1. Alpha=='B'
2. Alpha=='B', Bool==False
3. Alpha=='B', Bool==False, column 'I'
4. Alpha=='B', Bool==False, columns 'I' and 'III'
5. Alpha=='B', Bool==False, columns 'I', 'III', and all columns from 'V' onwards
6. Int is even

(BTW, I did rtfm, more than once even, but I really find it incomprehensible.)

RetroSearch is an open source project built by @garambo | Open a GitHub Issue

Search and Browse the WWW like it's 1997 | Search results from DuckDuckGo

HTML: 3.2 | Encoding: UTF-8 | Version: 0.7.4