Pandas subtract DataFrame column from lookup DataFrame column

user_13390673 0
Viewed 106

Given a dataframe test and another dataframe lookup

test = pd.DataFrame( [['a',1],
                      ['a',2], 
                      ['b',9]], columns = ['id', 'n'])

lookup_mins = pd.DataFrame( [['a',1], 
                             ['b',9], 
                             ['c',7]] , columns = ['id', 'n'])

trying to subtract each value n in test by the value n with matching id in lookup_mins using

s = lookup_mins.groupby(['id'])['n'].transform('min')
test['n2'] = test['n'] - s

Expected result,

id n n2
a  1 0
a  2 1
b  9 0

but getting instead

id n n2
a  1 0
a  2 -7
b  9  9

How to subtract test and lookup_mins to get the expected result as stated above?

pandas lookup-tables subtraction dataframe python
1 Answers

Use Series.map with aggregate min:

s = lookup_mins.groupby(['id'])['n'].min()
test['n2'] = test['n'] - test['id'].map(s)
print (test)
  id  n  n2
0  a  1   0
1  a  2   1
2  b  9   0
user_29010020
Related