CODEWITHSUNDEEP
pythonpython-3.xpandasdataframe
Adept
Adept

Reputation: 554

Complex link between 2 dataframes

I have 2 dataframes. One first is my reference

df_ref
ID  REF  VALUE
A   1    12
A   2    36
A   3    95
B   1    54
B   2    67
B   3    81
C   1    89
C   2    123
C   3    14

And the second is my restricted :

df_restrict
ID  V1  V2
A   1   2
B   3   2
C   2   1

What I want is to replace V1 and V2 in df_restrict by the values corresponding to the (ID, REF) from df_ref

df_restrict
ID  V1  V2
A   12  36
B   81  67
C   123 89

Thanks in advance

Upvotes: 2

Views: 83

Answers (4)

BENY
BENY

Reputation: 323226

Try something

out = df_restrict.set_index('ID').T.replace(df_ref.pivot(*df_ref.columns).T).T
Out[188]: 
     V1  V2
ID         
A    12  36
B    81  67
C   123  89

Upvotes: 1

Shubham Sharma
Shubham Sharma

Reputation: 71689

We can pivot df_ref to reshape, then set the index of df_restrict to ID, then transpose and replace the the values from the reshaped df_ref

r = df_ref.pivot('REF', 'ID', 'VALUE')
df_restrict.set_index('ID').T.replace(r).T

     V1  V2
ID         
A    12  36
B    81  67
C   123  89

Upvotes: 5

IoaTzimas
IoaTzimas

Reputation: 10624

Here is my suggestion:

df_restrict['V1']=df_restrict.merge(df_ref, left_on=['ID', 'V1'], right_on=['ID', 'REF'])['VALUE']

df_restrict['V2']=df_restrict.merge(df_ref, left_on=['ID', 'V2'], right_on=['ID', 'REF'])['VALUE']

>>> print(df_restrict)
  ID   V1  V2
0  A   12  36
1  B   81  67
2  C  123  89

Upvotes: 1

ALollz
ALollz

Reputation: 59539

melt the second DataFrame so you can merge the values on both columns, ['ID', 'REF'], and then pivot back to your original shape.

df1 = (df_restrict.melt(id_vars=['ID'], value_name='REF')
        .merge(df_ref, how='left')
        .pivot(index='ID', columns='variable', values='VALUE')
        .rename_axis(columns=None))

#     V1  V2
#ID         
#A    12  36
#B    81  67
#C   123  89

Upvotes: 3

Related Questions