Pandas: Compare all values within a dataframe by row

Question

I am trying to match rows and aggregate them in a single row.

For example for the table below, I want to aggregate the first three rows because they are similar. 4th isnt similar. In my check, I do nothing for any row that has col 1 as B. And then again aggregation for final two rows:

|---------------------|------------------|------------------|
|      Col 1          |     Col 2        |       Col 3      |
|---------------------|------------------|------------------|
|        A            |       12st       |        13        |
|---------------------|------------------|------------------|
|        A            |       12st       |        13        |
|---------------------|------------------|------------------|
|        A            |       12st       |        13        |
|---------------------|------------------|------------------|
|        A            |       12st       |        17        |
|---------------------|------------------|------------------|
|        B            |       11aa       |        10        |
|---------------------|------------------|------------------|
|        C            |       10ee       |        10        |
|---------------------|------------------|------------------|
|        C            |       10ee       |        10        |
|---------------------|------------------|------------------|

df = pd.DataFrame({'Col 1': ['A', 'A', 'A','A', 'B', 'C', 'C'],'Col 2': ['12st', '12st', '12st', '12st', '11aa' ,'10ee','10ee'],'Col 3': [13, 13, 13, 17, 10, 10, 10 ]})

I want to get the following output:

|---------------------|------------------|------------------|---------------|
|      Col 1          |     Col 2        |       Col 3      |    Col 4      |
|---------------------|------------------|------------------|---------------|
|        A            |       12st       |        13        |      3        |
|---------------------|------------------|------------------|---------------|
|        A            |       12st       |        17        |      1        |
|---------------------|------------------|------------------|---------------|
|        B            |       11a        |        10        |      1        |
|---------------------|------------------|------------------|---------------|
|        C            |       10ee       |        10        |      2        |
|---------------------|------------------|------------------|---------------|

I have tried simpler things like df.shift() but that seems to only work for a specific col and not row. Plus I want to do this iteratively for the rows (i) it keeps on matching (i==i+1==i+2).

Thanks

Answer 1

I think groupby.size can do it like:

print (df.groupby(['Col 1','Col 2', 'Col 3']).size().reset_index(name='Col 4'))
  Col 1 Col 2  Col 3  Col 4
0     A  12st     13      3
1     A  12st     17      1
2     B  11aa     10      1
3     C  10ee     10      2

Answer 2

I think you probably do something like this:

output_data = []

for i,row in range(rows):
    current_row = df.iloc[i]
    try:
        # check if row is in output_data
        output_data.index(current_row)
    except:
        output_data.append(current_row)

# Create a new dataframe
new_df = pd.DataFrame(output_data)

Please let me know if this helps: :D Thanks!