Count Partial match in each DataFrame Column in Pandas

Question

I have Data frame in which I want to count occurrence of some word per each column. Per one coumn I can do:

df['Col1'].str.contains('test').value_couns()

or

df[df['Col1'].str.contains('test')]['Col1'].count()

and i get count for particular column.

How can I get it for all columns? I would like to avoid do it manually per each column since there are quite a few of them.

Expected output

I believe this is a duplicate of https://stackoverflow.com/questions/32589829/how-to-get-value-counts-for-multiple-columns-at-once-in-pandas-dataframe — Dillon, Jun 22 '18 at 07:21
I think I have something which is work almost as I wanted: df.astype(str).apply(lambda x: x.str.contains('test').value_counts()) — Submi, Jun 22 '18 at 07:43

Mohamed Thasin ah · Accepted Answer · 2018-06-22T07:53:18.987

1

one way to solve this, As Submi Tried,

print (df.astype(str).apply(lambda x: x.str.contains('test').value_counts()).loc[True].fillna(0)).to_frame().T.reset_index(drop=True)

Output:

   col1  col2  col3
0   1.0   0.0   2.0

edited Jun 22 '18 at 07:53

answered Jun 22 '18 at 07:42

Mohamed Thasin ah

First option I would like to avoid, I have a lot of dfs and a lot of different columns in those. Second option works. only difference with @theletz solution is floats instead of int but this is not a problem. Thank you. – Submi Jun 22 '18 at 07:52
@Submi - To get exactly same output use reset_index. updated to the solution – Mohamed Thasin ah Jun 22 '18 at 07:57

score 1 · Answer 2 · answered Jun 22 '18 at 07:45

1

I think you are looking for this:

df.applymap(lambda x: 'test' in str(x)).sum()

answered Jun 22 '18 at 07:45

theletz

2 Answers2