CODEWITHSUNDEEP
pythonpandasdataframe
Miguel Gonzalez
Miguel Gonzalez

Reputation: 773

How to change column level in df.pivot()

I'd like that the column introduced as argument in df.pivot() act as outter level, not inner.

This is how my dataframe looks like:

import pandas as pd

d1 = {'Zone':'Zone 00', 'Area':1,'A':2, 'B':51,'C':5,'D':1}
d2 = {'Zone':'Zone 00', 'Area':2,'A':6, 'B':5,'C':36,'D':2}
d3 = {'Zone':'Zone 01', 'Area':1,'A':2, 'B':8,'C':9,'D':22}
d4 = {'Zone':'Zone 01', 'Area':2,'A':8, 'B':55,'C':19,'D':42}
d5 = {'Zone':'Zone 02', 'Area':1,'A':14, 'B':42,'C':8,'D':23}
d6 = {'Zone':'Zone 02', 'Area':2,'A':23, 'B':96,'C':75,'D':12}

dics = (d1, d2, d3, d4, d5, d6)
df = pd.DataFrame([i for i in dics])

enter image description here

I pivot the dataframe to display the params (A, B, C, D) by zone's areas and not by the own params.

df.pivot(columns='Area', index='Zone',values=['A','B','C','D'])

But I obtain the column's level ordered in the inverse way:

enter image description here

This is what I'd like to obtain: enter image description here

I have tried df.swaplevel(axis=1) method, but no inverse grouper is set. It only changes level order.

Does anybody knows how to do it? Thanks in advance.

Upvotes: 3

Views: 1225

Answers (1)

Quang Hoang
Quang Hoang

Reputation: 150765

Try swaplevel:

(df.pivot(columns='Area', index='Zone',values=['A','B','C','D'])
   .swaplevel(0,1, axis=1)
   .sort_index(axis=1)
)

Output:

Area      1              2            
          A   B  C   D   A   B   C   D
Zone                                  
Zone 00   2  51  5   1   6   5  36   2
Zone 01   2   8  9  22   8  55  19  42
Zone 02  14  42  8  23  23  96  75  12

Upvotes: 5

Related Questions