Row divide by their column sum in pandas dataframe-CodePudding

I am trying to do a row value divide by their column sum. I think I could do it row by row and then concat them together but I am wondering if there's a more efficient way to do it in pandas.

abc = {
    'fruits': {0: 'apple', 1: 'orange', 2: 'total'},
    'Monday': {0: 2, 1: 4, 2: 6},
    'Tuesday': {0: -2, 1: -6, 2: -8},
    'Wednesday': {0: -40, 1: -65, 2: -105}
}

pd.DataFrame.from_dict(abc)

   fruits  Monday  Tuesday  Wednesday
0   apple       2       -2        -40
1  orange       4       -6        -65
2   total       6       -8       -105

Expected output:

   fruits  Monday  Tuesday  Wednesday
0   apple    0.33     0.25       0.38
1  orange    0.67     0.75       0.62

CodePudding user response：

df = pd.DataFrame.from_dict(abc)
df.iloc[:,1:] = df.iloc[:,1:] / df.iloc[-1,1:]
df = df[:-1]
df

df.iloc[:,1:] takes all columns except fruits. df.iloc[-1,1:] is the last row of the dataframe, which is total. df = df[:-1] assigns all rows except the total row as a df