CODEWITHSUNDEEP
pythonpython-3.xpandas
user12229568
user12229568

Reputation: 15

Combining / merging two datasets with duplicated names

I tried merging two datasets (DataFrames) as follows:

D1 = pd.DataFrame({'Village':['Ampil','Ampil','Ampil','Bachey','Bachey','Center','Center','Center','Center'], 'Code':[123,324,190,453,321,786,456,234,987]})

D2 = pd.DataFrame({'Village':['Ampil','Ampil','Bachey','Bachey','Center','Center'],'Lat':[11.563,13.278,12.637,11.356,12.736,13.456], 'Long':[102.234,103.432,105.673,103.539,103.873,102.983]})

I want to merge the two based on the Village column. I want the output to look like the following:

D3 = pd.DataFrame({'Village': ['Ampil','Ampil','Bachey','Bachey','Center','Center'],'Code':[123,324,453,321,786,456],'Lat':[11.563,13.278,12.637,11.356,12.736,13.456], 'Long':[102.234,103.432,105.673,103.539,103.873,102.983]})

I have tried join, merge, and concat but none fit the purpose. I need a code that would apply to a larger data. Really appreciate it if some could help.

Upvotes: 1

Views: 58

Answers (1)

Henry Yik
Henry Yik

Reputation: 22503

One way is to first create a running cumcount for both your initial dfs by Village, and then merge by both Village and count:

df1['count'] = df1.groupby('Village').cumcount()
df2["count"] = df2.groupby('Village').cumcount()

print (df2.merge(df1,on=["Village","count"],how="left").drop("count",axis=1))

#
      Village     Lat     Long  Code
0   Ampil  11.563  102.234   123
1   Ampil  13.278  103.432   324
2  Bachey  12.637  105.673   453
3  Bachey  11.356  103.539   321
4  Center  12.736  103.873   786
5  Center  13.456  102.983   456

Upvotes: 1

Related Questions