Add multiple index levels to existing pandas index

Question

I have a df with a regular datetime index. I need to add multiple index levels. What's a pythonic way to achieve this?

import numpy as np
import pandas as pd

idx = pd.date_range(start="2020-01-01", end="2020-01-10")
vals = {"values": np.random.randint(low=1, high=100, size=10)}
df = pd.DataFrame(data=vals, index=idx)
df.index.name = "time"

I need to add, for example, two new index levels. They must prepend the current index level "time". The result should look like this.

                              values
L0_name  L1_name  time              
L0 value L1 value 2020-01-01      87
                  2020-01-02      46
                  2020-01-03      19
                  2020-01-04      44
                  2020-01-05      94
                  2020-01-06      58
                  2020-01-07      74
                  2020-01-08      32
                  2020-01-09      64
                  2020-01-10      84

jezrael · Accepted Answer · 2020-06-19 08:44:24Z

Create new columns by DataFrame.assign, append to existing index by DataFrame.set_index and then change order of them by DataFrame.reorder_levels:

df1 = (df.assign(L0_name = 'L0 value', L1_name='L1 value')
         .set_index(['L0_name','L1_name'], append=True)
         .reorder_levels([1,2,0]))

Or create new MultiIndex, e.g. by MultiIndex.from_product and overwrite existing index by DataFrame.set_index:

idx = pd.MultiIndex.from_product([['L0 value'],['L1 value'], df.index], 
                                 names=['L0_name','L1_name','time'])

df1 = df.set_index(idx)

print (df1)
                              values
L0_name  L1_name  time              
L0 value L1 value 2020-01-01      84
                  2020-01-02      31
                  2020-01-03       4
                  2020-01-04      49
                  2020-01-05      77
                  2020-01-06      49
                  2020-01-07      15
                  2020-01-08      24
                  2020-01-09      49
                  2020-01-10      62

Collectives™ on Stack Overflow

Add multiple index levels to existing pandas index

1 Answer 1

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

Comments

Your Answer

Sign up or log in

Post as a guest

Related