Replace Pandas DataFrame column values based on containing dictionary keys

Question

Here is an example, where column is assigned when the row is a dictionary key: https://stackoverflow.com/a/20250996/12603542

What I am looking for is case, where column is assigned when the row contains a dictionary key.

For example (based on example above):

import pandas as pd
df = pd.DataFrame({'col1': {0: 'aba', 1: 'abc', 2: 'abx'}})

#gives me a DataFrame
     col1 
0    aba #contains 'ba'
1    abc #will NOT be replaced
2    abx #contains 'bx'

dictionary = {'ba': 5, 'bx': 8}

#and I need to get:

     col1 
0    5
1    abc
2    8

Ch3steR · Accepted Answer · 2022-02-06T16:17:40.800

5

You could use DataFrame.replace with regex parameter set to True and pass the mapping dictionary.

df.replace(dictionary, regex=True)

#   col2
# 0    5
# 1  abc
# 2    8

This usage of df.replace is less known. You can read more about it here.

edited Feb 06 '22 at 16:17

answered Feb 06 '22 at 16:03

Ch3steR

20,090
4
28
58

1

This is pretty good. +1 – Mayank Porwal Feb 06 '22 at 16:05
@MayankPorwal Thank you. `df.replace` is very handy, though can be slow at times. – Ch3steR Feb 06 '22 at 16:08
Some info why this work ~ https://stackoverflow.com/questions/46944650/replace-value-by-using-regex-to-np-nan – BENY Feb 06 '22 at 16:09
@BENY Thank you. Added it to my answer as comments might get deleted. – Ch3steR Feb 06 '22 at 16:18
This is what I was looking for. Thank you. I would just add `inplace=True` parameter to the line. – bakunet Feb 06 '22 at 16:29

Replace Pandas DataFrame column values based on containing dictionary keys

1 Answers1