Returning the index value when using groupby

Question

Im using grouby to rearrange a dataframe that has a date time index and I'd like to be able to return the datetime index value at the first grouping. For example, if I group by the Number column, i'd like to return the index value at the first instance of each number. I'm able to use the Trade Type with the .iat[0] to get the Type at the first occurrence of number. How would i return the index value?

                  Number Type
01/02/2007 - 05:30  1   Down
01/02/2007 - 10:30  2   Up
01/02/2007 - 15:30  2   Up
01/02/2007 - 17:00  2   Up
01/02/2007 - 18:30  2   Down
01/02/2007 - 22:00  3   Down
01/03/2007 - 06:30  3   Down
01/03/2007 - 11:00  3   Down
01/04/2007 - 08:30  3   Down
01/04/2007 - 09:30  4   Down
01/04/2007 - 10:30  4   Down
01/05/2007 - 10:00  4   Down
01/07/2007 - 23:00  4   Down
01/08/2007 - 10:30  5   Down

this is the function i'm using to build a dictionary which will return a series that I can then apply to my dataframe.

def compact_view(groupby):
agg_dict = {'Type': groupby['Type'].iat[0],
            'Time': groupby['Date_Time'].iat[0]<----?


            }

return pd.Series(agg_dict, index=['Type','Time'])

So i'd want these returned:

01/02/2007 - 05:30  1
01/02/2007 - 10:30  2
01/02/2007 - 22:00  3
01/04/2007 - 09:30  4
01/08/2007 - 10:30  5

Hopefully this makes sense

Answer 1

This is head

df.groupby('Number').head(1)