Pandas dataframe of sums from another dataframe -

i have pandas dataframe. want make dataframe, columns sums of selected columns first frame.

for example in first dataframe have columns named 'a', 'b', 'c' , 'd'. new columns 'x' , 'y' , 'x' has them sum of 'a','b' , 'c' , 'y' contains 'd'. "mapping" may saved in dictionary: {'x': ['a', 'b', 'c'], y: ['d']}, not need be.

shall initialize empty data frame keys 'x' , 'y' , add columns or there quicker way?

you can directly initialize dataframe data in it:

df2 = pd.dataframe({'x': df[['a', 'b', 'c']].sum(axis=1), 'y': df['d']}) 

or can build column column:

df2 = df[['a', 'b', 'c']].sum(axis=1).to_frame(name='x') df2['y'] = df['d'] 

for first column, you'll need call to_frame name desired column name. otherwise, df2 series instead of dataframe , you'll have trouble adding columns. after the initial column, can assign columns usual.

sample input:

df = pd.dataframe({'a':[3,1,4],'b':[1,5,9],'c':[2,6,5],'d':[3,5,8]})      b  c  d 0  3  1  2  3 1  1  5  6  5 2  4  9  5  8 

sample output (for either method):

    x  y 0   6  3 1  12  5 2  18  8