CODEWITHSUNDEEP
pythonpandasdataframereplace
julia_3010
julia_3010

Reputation: 255

Python replace all values in dataframe with values from other dataframe

I'm quite new to python (and pandas) and a have a replace task for a large dataframe i couldn't find a solution for.

So i have two dataframes, one (df1) which looks something like this:

Id      Id      Id      
4954733 3929949 515674  
2950086 1863885 4269069 
1241018 3711213 4507609 
3806276 2035233 4968071 
4437138 1248817 1167192 
5468160 4726010 2851685 
1211786 2604463 5172095 
2914539 5235788 4130808 
4730974 5835757 1536235 
2201352 5779683 5771612 
3864854 4784259 2928288

the other dataframe (df2) containing all the 'old' id's and the corresponding new ones in the next column (from 1 to 20,000), which looks something like this:

Id       Id_new
5774290  1
761000   2
3489755  3
1084156  4
2188433  5
3456900  6
4364416  7
3518181  8
3926684  9
5797492  10
4435820  11

what i would like to do is replace all the id's (all columns) in df1 with the corresponding Id_new from df2. I guess ideally without having to do a merge or join for each column, given the size of the dataset?

The result should be like this: df_new

Id_new  Id_new  Id_new
8        12     22
16       9      8
21       25     10
10       15     13
29       6      4
22       7      22
30       3      3
11       31     29
32       29     27
12       3      4
14       6      24

Any tips would be great, thanks in advance!

Upvotes: 1

Views: 276

Answers (1)

jezrael
jezrael

Reputation: 862681

I think you need replace by Series created by set_index:

print (df1)
         Id     Id.1     Id.2
0   4954733  3929949   515674 <-first value changed for match data
1   2950086  1863885  4269069
2   1241018  3711213  4507609
3   3806276  2035233  4968071
4   4437138  1248817  1167192
5   5468160  4726010  2851685
6   1211786  2604463  5172095
7   2914539  5235788  4130808
8   4730974  5835757  1536235
9   2201352  5779683  5771612
10  3864854  4784259  2928288

df = df1.replace(df2.set_index('Id')['Id_new'])
print (df)
         Id     Id.1     Id.2
0         1  3929949   515674
1   2950086  1863885  4269069
2   1241018  3711213  4507609
3   3806276  2035233  4968071
4   4437138  1248817  1167192
5   5468160  4726010  2851685
6   1211786  2604463  5172095
7   2914539  5235788  4130808
8   4730974  5835757  1536235
9   2201352  5779683  5771612
10  3864854  4784259  2928288

Upvotes: 3

Related Questions