CODEWITHSUNDEEP
pythonpandasreplacedataframe
Alejandro Simkievich
Alejandro Simkievich

Reputation: 3792

pandas replace inner workings

I have a pandas dataframe train with the following content in line 24549

train.iloc[24549,:]
>>> 
id                                                           79571
product_uid                                                 123177
product_title    Charlotte Pipe 3/4 in. PVC Sch. 40 90-Degree S...
search_term                                  3/4 in pvc assesories
relevance                                                     2.67
Name: 24549, dtype: object

I want to to replace 'assesories' with 'accessories' in train.iloc[24549,3]

If I do:

train = train.replace('assessories','accessories')

Whenever I check the df value after this command, the word in the above dataframe element is still 'assessories'.

However, if I do:

c =  train.iloc[24549,3]

c = c.replace('assessories','accessories')

c
>>>>  '3/4 in pvc accessories'

I am still scratching my head over this. I do not want to iterate through the df but rather do a vectorized implementation. Do you have any ideas how to do that?

Upvotes: 0

Views: 69

Answers (1)

Alejandro Simkievich
Alejandro Simkievich

Reputation: 3792

the right command is not

train = train.replace('assessories','accessories')

but 

train = train.str.replace('assessories','accessories')

Thanks to my friend Lucas Eustaquio for pointing this out.

Upvotes: 1

Related Questions