如果熊猫中的标签不相同，我可以组合两行的值吗

Question

Here's the 2 dataframes I want to combine. But the labels are different from each other

df1
Date        Campaign                              Sales
11/07/2020  AMZ CT BR Leather Shoes ABCDEFG1234   $10
11/07/2020  AMZ CT NB Leather Shoes ABCDEFG1234   $20
11/07/2020  AMZ OG BR Bag HGIJK567                $30
11/07/2020  AMZ OG NB Bag HGIJK567 Desktop        $40

df2
Date        Campaign                              Spend        
11/07/2020  GA BR Leather Shoes ABCDEFG1234       $5
11/07/2020  GA NB Leather Shoes ABCDEFG1234       $6
11/07/2020  GA BR Bag HGIJK567                    $7
11/07/2020  GA NB Bag HGIJK567 Desktop            $8

Here's the output I want

df3
Date        Campaign                         Spend   Sales
11/07/2020  CT BR Leather Shoes ABCDEFG1234  $5      $10
11/07/2020  CT NB Leather Shoes ABCDEFG1234  $6      $20
11/07/2020  OG BR Bag HGIJK567               $7      $30
11/07/2020  OG NB Bag HGIJK567 Desktop       $8      $40

 

Answer 1

I would create an extra column to perform the merge on. For what I can see, merging is done based on the product name without the first acronyms.

df1['Campaign_j'] = df1['Campaign'].map(lambda x: ' '.join(x.split()[3:]))
df2['Campaign_j'] = df2['Campaign'].map(lambda x: ' '.join(x.split()[2:]))
print(df1)
print(df2)
df3 = df1.merge(df2,how='left',on=['Campaign_j'],suffixes=('','_x')).drop_duplicates('Campaign_x')[['Campaign','Sales','Spend']]

After the joining, we will drop the duplicates from the first Campaign column (Campaign_x) and finally select the desired columns. I have not added the date column because it has no effect in this problem. Output:

                              Campaign Sales Costs
0  AMZ CT BR Leather Shoes ABCDEFG1234    10     5
2  AMZ CT NB Leather Shoes ABCDEFG1234    20     6
4               AMZ OG BR Bag HGIJK567    30     7
5       AMZ OG NB Bag HGIJK567 Desktop    40     8

Answer 2

If I understand your question correctly
Yes, you can. But rows that are not in one of dataframes are left blank.
Let me give you an example: If you have two dataframe First.csv and Second.csv as follows:
First dataframe:

A, B, C
1, 2, 3
2, 3, 4

Second dataframe:

A, C
1, 3
2, 4

import pandas as pd
df_a = pd.read_csv('First.csv')
df_b = pd.read_csv('Second.csv')

You can use:

df_row_merged = pd.concat([df_a, df_b], ignore_index=True).

to merge two dataframes. df_row_merged will be as follows:

A, B,   C
1, 2.0, 3
2, 3.0, 4
1,    , 3
2,    , 4

I hope this help you.