AttributeError: 'DataFrameGroupBy' object 没有属性 'colnames'

Question

I have a list of column names as shown below:

colnames = ['split1', 'split2', 'split3', 'split4', 'split5', 'split6', 'split7']

I have used this in a line of code where I iterate over the list items as shown below:

for i in range(0,x+2):
    df[d[i+1]] = df.groupby(d[i]).colnames[i].transform(lambda x : x.factorize()[0]+1)

This is giving me the attribute error:

AttributeError: 'DataFrameGroupBy' object has no attribute 'columns'

But if I individually give the colname items to the code, it works fine:

df['dot2'] = df.groupby('dot1').split1.transform(lambda x : x.factorize()[0]+1)

Answer 1

In df.groupby(d[i]).colnames[i] , colnames[i] is a variable. You may need do

df.groupby(d[i])[f'{colnames[i]}']