Python/Pandas dataframe - return column name

Question

Is there a way to return the name/header of a column into a string in a pandas dataframe? I want to work with a row of data which has the same prefix. The dataframe header looks like this:

col_00 | col_01 | ... | col_51 | bc_00 | cd_00 | cd_01 | ... | cd_90

I'd like to apply a function to each row, but only from col_00 to col_51 and to cd_00 to cd_90 separately. To do this, I thought I'd collect the column names into a list, fe. to_work_with would be the list of columns starting with the prefix 'col', apply the function to df[to_work_with]. Then I'd change the to_work_with and it would contain the list of columns starting with the 'cd' prefix et cetera. But I don't know how to iterate through the column names.

So basically, the thing I'm looking for is this function:

to_work_with = column names in the df that start with "thisstring"

How can I do that? Thank you!

Do you want [col for col in df.columns.values if col.startswith("thisstring")]? — Michael Hoff
– Michael Hoff, Commented Jul 3, 2016 at 11:16

jezrael · Accepted Answer · 2017-11-14 06:59:55Z

2

You can use boolean indexing with str.startswith:

cols = df.columns[df.columns.str.startswith('cd')]
print (cols)
Index(['cd_00', 'cd_01', 'cd_02', 'cd_90'], dtype='object')

Sample:

print (df)
   col_00  col_01  col_02  col_51  bc_00  cd_00  cd_01  cd_02  cd_90
0       1       2       3       4      5      6      7      8      9

cols = df.columns[df.columns.str.startswith('cd')]
print (cols)
Index(['cd_00', 'cd_01', 'cd_02', 'cd_90'], dtype='object')

#if want apply some function for filtered columns only
def f(x):
    return x + 1

df[cols] = df[cols].apply(f)    
print (df)
   col_00  col_01  col_02  col_51  bc_00  cd_00  cd_01  cd_02  cd_90
0       1       2       3       4      5      7      8      9     10

Another solution with list comprehension:

cols = [col for col in df.columns if col.startswith("cd")]
print (cols)
['cd_00', 'cd_01', 'cd_02', 'cd_90']

edited Nov 14, 2017 at 6:59

answered Jul 3, 2016 at 11:16

jezrael

868k103 gold badges1.4k silver badges1.3k bronze badges

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

5 Comments

lte__ Over a year ago

Amazing! Thank you! :)

jezrael Over a year ago

Glad can help you! Good luck!

JKC Over a year ago

Hi @jezrael very good one. I am able to create the Dataframe with values by just giving df[cols]. Can you please let me know the use of apply function there ? Sorry I am still in a learning process. So this might be a basic question to you.

jezrael Over a year ago

@JKC - It is only a example for apply some function for columns filtered by startswith :)

jezrael Over a year ago

No problem, I add ot to answer ;)

Collectives™ on Stack Overflow

Python/Pandas dataframe - return column name

1 Answer 1

5 Comments

Your Answer

Linked

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

5 Comments

Your Answer

Sign up or log in

Post as a guest

Linked

Related