Apply pandas to_numeric() to a some columns in a dataframe

Question

Given a dataframe such as:

page            source_url      count   yes   no   ii  dk

google.com      youtube.com        3      0    0    1   2

facebook.com    whatsapp.com       3      0    0    1   2

microsoft.com   stackoverflow.com  3      0    0    1   2

I want to apply pd.to_numeric on the last 4 columns only.

I tried doing :

df['count'] = pd.to_numeric(df['count'])
df['yes'] = pd.to_numeric(df['yes'])
df['no'] = pd.to_numeric(df['no'])
df['ii'] = pd.to_numeric(df['i'])
df['dk'] = pd.to_numeric(df['dk'])

But the first line keeps giving me:

ValueError at /
Unable to parse string "<bound method Series.count of page          https://www.snopes.com/fact-check/trump-suicid...
source_url    https://www.facebook.com/candace.witte/posts/1...
count                                                         0
Name: 7909, dtype: object>" at position 0

How can I use to_numeric() on some columns of a dataframe?

Did you perhaps use df.count in your code, either currently or in a prior function or iteration, to assign to the count column? It looks like you may have overwritten something in your namespace with the dataframe.count function signature, based on string <bound method Series.count... — G. Anderson
– G. Anderson, Commented Jul 19, 2019 at 16:29

Juan C · Accepted Answer · 2019-07-19 16:32:12Z

1

It seems your column count has at least one non-numeric value, in this case one that starts with "bound method Series.count...". To turn errors into nan you could do the following:

df['count'] =  df.count.apply(pd.to_numeric, errors='coerce')

If more columns have the same problem, you could do the same for them

answered Jul 19, 2019 at 16:32

Juan C

6,1484 gold badges27 silver badges64 bronze badges

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

1 Comment

Mohamed Moustafa Over a year ago

Thx man. I checked the csv and there was an issue with the script that generated it. I fixed it and now the code works

Collectives™ on Stack Overflow

Apply pandas to_numeric() to a some columns in a dataframe

1 Answer 1

1 Comment

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

1 Comment

Your Answer

Sign up or log in

Post as a guest

Related