Replace cells with specific terms

Question

I want to replace the words contains "conference" and "group" with "N/A" in the dataframe. E.g. "AAAI Conference"->"N/A" "Alibaba Group" -> "N/A"

The dataframe is called name, I try two ways to do this:

columns=['nameCurrentEmployer',
       'name2ndEmployer', 'name3rdEmployer',
       'name4thEmployer', 'name5thEmployer',
       'name6thEmployer', 'name7thEmployer',
       'name8thEmployer', 'name9thEmployer',
       'name10thEmployer'] 
name.loc[name.str.contains(['conference','group'], case=False), columns] = 'N/A'

Prompt error AttributeError: 'DataFrame' object has no attribute 'str'

NAMES = pd.Series(name.values.flatten())
NAMES.loc[NAMES.str.contains(['conference','group'], case=False), columns] = 'N/A'

Now the error is

TypeError: unhashable type: 'list'

Thank you very much.

i'd suggest you use pandas str replace instead and possibly use a regex expression containing the words 'conference' or 'group' — sammywemmy
– sammywemmy, Commented Feb 3, 2020 at 22:35
What are you using the string 'N/A' for? Why are you doing pd.Series(name.values.flatten()) ? Can you share more of your program? Variable and function names should follow the lower_case_with_underscores style. Always share the entire error message. Do you not have a minimal reproducible example? — AMC
– AMC, Commented Feb 3, 2020 at 22:54
Also, is this not just a worse duplicate of stackoverflow.com/questions/39602824/… ? — AMC
– AMC, Commented Feb 4, 2020 at 0:26

Giorgos Myrianthous · Accepted Answer · 2020-02-04 09:27:34Z

0

str.contains() takes

Character sequence or regular expression.

So instead of ['conference','group'] you should use 'conference|group':

NAMES.loc[NAMES.str.contains('conference|group', case=False), columns] = 'N/A'

Alternatively, I would suggest to use either apply():

NAMES.name = NAMES.name.apply(lambda x: 'N/A' if 'conference' in x else x)

or str.replace()

edited Feb 4, 2020 at 9:27

answered Feb 3, 2020 at 22:42

Giorgos Myrianthous

40.4k21 gold badges155 silver badges174 bronze badges

Sign up to request clarification or add additional context in comments.

4 Comments

KDWB Over a year ago

Thank you, Giorgos. But it still prompts two errors: 1.TypeError: 'Series' objects are mutable, thus they cannot be hashed 2.Indexing Error How should I fix it or is there any other way that can do the same job? Thanks again.

AMC Over a year ago

Why not recommend the use of DataFrame.replace, instead of the needlessly awkward .loc[] method?

AMC Over a year ago

@RenzhiZhao Can you address the points made by the first commenter and myself?

KDWB Over a year ago

@ AMC Thank you, AMC. Replace works. You really help me out and give me a good lesson about raising good questions.

Collectives™ on Stack Overflow

Replace cells with specific terms

1 Answer 1

4 Comments

Your Answer

Linked

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

4 Comments

Your Answer

Sign up or log in

Post as a guest

Linked

Related