CODEWITHSUNDEEP
pandasdataframereplace
Bimons
Bimons

Reputation: 241

Replace string taken from another column in pandas

I am trying to do a replace across one column of a pandas dataframe like the below.

From:

a          b
house      ho
cheese     ee
king       ng

To:

a        b
use      ho
chse     ee
ki       ng

My attempt is to use:

df['a'] = df['a'].str.replace(df['b'], "")

but I get TypeError: 'Series' objects are mutable, thus they cannot be hashed

I have done it by iterating row by row across the dataframe but its 200,000 rows so would take hours. Does anyone know how I can make this work?

Upvotes: 1

Views: 48

Answers (1)

jezrael
jezrael

Reputation: 863781

Because performance is important here is possible use list comprehension with replace for replace per rows:

df['a'] = [a.replace(b, "") for a, b in df[['a','b']].values]

Another solution is slowier with DataFrame.apply:

df['a'] = df.apply(lambda x: x.a.replace(x.b, ""), axis=1)

print (df)
      a   b
0   use  ho
1  chse  ee
2    ki  ng

Upvotes: 3

Related Questions