How to calculate percentage of each class variable after grouping it by another variable?

I have a data that initially looks like this:

column1	Time
yes	271-273
no	271-273
neutral	271-273
no	274-276
...	...

I want the percentage of yes, percentage of no, and percentage of neutral across each Time range. I was able to get the count of each one of these categories (yes,no,neutral) across every time range using the following code:

df['COUNTER'] =1    
group_data = df.groupby(['Time','column1'])['COUNTER'].sum()

I am not sure how to calculate the percentage of counts from this.

Solution

Use SeriesGroupBy.value_counts with parameter normalize=True:

print (df)
   column1     Time
0      yes  271-273
1       no  271-273
2  neutral  271-273
3       no  271-273 <- changed data for better sample

group_data = (df.groupby(['Time'])['column1']
                .value_counts(normalize=True)
                .reset_index(name='%') )

print (group_data)
      Time  column1     %
0  271-273       no  0.50
1  271-273  neutral  0.25
2  271-273      yes  0.25

Another idea with divide by sum of counts with DataFrameGroupBy.size and Series.div:

s = df.groupby(['Time','column1']).size()
group_data = s.div(s.groupby(level=0).sum()).reset_index(name='%') 
print (group_data)

      Time  column1     %
0  271-273  neutral  0.25
1  271-273       no  0.50
2  271-273      yes  0.25