Pandas Merge Dataframes

Question

I have some problem about Pandas.

I have

   df1 = pd.DataFrame({'col1': ['t1' ,'t2','t3','t4','t5'],
            'col2': ['1' ,'2','3','5','6']})

   df2 = pd.DataFrame({'col1': ['t1' ,'t2','t4','t5'],
            'col2': ['5' ,'7','8','2']})

   df3 = pd.DataFrame({'col1': ['t1' ,'t2','t4','t5','t6','t7'],
            'col2': ['6' ,'3','2','5','3','12']})

and result is

   t   d1  d2  d3

   t1   1  5   6
   t2   2  7   3
   t3   3  0   0
   t4   5  8   2
   t5   6  2   5
   t6   0  0   3
   t7   0  0   12

Thanks in advance!

Just rename the columns and merge (outer) and fill nans with 0 — Muhammad Hassan, Mar 09 '22 at 04:47
Use `df1.merge(df2, on='col1', how='outer').merge(df3, how='outer').fillna(0).rename(columns={'col2_x': 'd1', 'col2_y': 'd2', 'col2': 'd3'})` — Mayank Porwal, Mar 09 '22 at 05:25

score 2 · Accepted Answer · answered Mar 09 '22 at 05:04

rename the columns and then merge the data frames.

#rename the comlumns
df1.rename(columns={'col2':'d1'},inplace=True)
df2.rename(columns={'col2':'d2'},inplace=True)
df3.rename(columns={'col2':'d3'},inplace=True)

#merge all the dfs; outer merge
from functools import reduce
dfs = [df1, df2, df3]
dfs = reduce(lambda left,right: pd.merge(left,right,how='outer', on='col1'), dfs)

#fill NaN values as 0
dfs.fillna(0,inplace=True)

Pandas Merge Dataframes

1 Answers1