Count the occurrences of entire row in pandas DataFrame

Question

I need to count the occurrence of entire row in pandas DataFrame

For example if I have the Data Frame:

A = pd.DataFrame([['a','b','c'],['b','a','c'],['a','b','c']])

The expected result should be:

'a','b','c' : 2

'b','a','c' : 1

value_counts only counts the occurrence of one element in a series (one column)

I can create new column that will have all the element of that row and count values in that column, but I hope for a better solution.

Please provide an [mcve] to show what you have tried so far and where you are stuck at the moment. In addition, you might want to have a look at this answer and _adopt_ it to your needs: https://stackoverflow.com/questions/15943769/how-do-i-get-the-row-count-of-a-pandas-dataframe — albert, Aug 03 '21 at 07:17
thank. I want to count occurrent of specific row, not all rows in Data Frame. I edited my question — tumir, Aug 03 '21 at 07:25

score 0 · Accepted Answer · answered Aug 03 '21 at 07:37

0

You can do this:

A = pd.DataFrame([['a','b','c'],['b','a','c'],['a','b','c']])
A.groupby(A.columns.tolist(),as_index=False).size()

which returns:

 0  1  2  size
0  a  b  c     2
1  b  a  c     1

answered Aug 03 '21 at 07:37

thank you. that looks good. basically doing groupby for all columns and counting the size of each group. thanks a lot – tumir Aug 03 '21 at 08:07
If this is the answer you were looking for, please to mark it as accepted so it disappears from the list of unanswered questions. :-) – Serge de Gosson de Varennes Aug 03 '21 at 09:01
@SergedeGossondeVarennes isn't this equivalent to `A["size"] = A.value_counts()` ? – ljmc Jan 03 '23 at 21:57

score 0 · Answer 2 · edited Aug 04 '21 at 05:05

0

Below code would provide the result you are expecting.

df = pd.DataFrame([['a','b','c'],['b','a','c'],['a','b','c']])

df.groupby(list(df.columns))[0].count().to_frame('count').reset_index()

above code results in following dataframe.

    0   1   2   count

 0  a   b   c   2

 1  b   a   c   1

edited Aug 04 '21 at 05:05

R_Dax

answered Aug 03 '21 at 07:56

that is a good solution for that specific example, but I got the base idea. groupby for all needed columns (in my case all columns) and counting the size of each group. thanks – tumir Aug 03 '21 at 08:11

2 Answers2