Add 2 new columns to existing dataframe using appl

问题:

I want to use the apply function that: - Takes 2 columns as inputs - Outputs two new columns based on a function.

An example is with this add_multiply function.

#function with 2 column inputs and 2 outputs
def add_multiply (a,b):
  return (a+b, a*b )

#example dataframe
df = pd.DataFrame({'col1': [1, 2], 'col2': [3, 4]})

#this doesn't work
df[['add', 'multiply']] = df.apply(lambda x: add_multiply(x['col1'], x['col2']), axis=1)

ideal result:

col1  col2  add  multiply
1     3     4    3
2     4     6    8

回答1:

You can add result_type='expand' in the apply:

‘expand’ : list-like results will be turned into columns.

df[['add', 'multiply']]=df.apply(lambda x: add_multiply(x['col1'], x['col2']),axis=1,
                             result_type='expand')

Or call a dataframe constructor:

df[['add', 'multiply']]=pd.DataFrame(df.apply(lambda x: add_multiply(x['col1'], 
                                    x['col2']), axis=1).tolist())

---

   col1  col2  add  multiply
0     1     3    4         3
1     2     4    6         8

回答2:

anky_91's answer highlights a useful option in apply.

For this particular case however, apply is not even required,

df['add'], df['multiply'] = add_multiply(df['col1'],df['col2'])