如何对 Pandas GroupBy 对象中的组进行排序？

Question

I have a pandas DataFrame with 3 columns: Date, Concept and Value. I want to add a fourth column "Rank" that contains the order statistic of Value when grouped by Date. So for example, consider the following table:

Date	Concept	Value
2021-01-01	A	20
2021-01-01	B	30
2021-01-01	C	25
2021-02-01	A	17
2021-02-01	B	31
2021-02-01	C	42

In this case, my final table should be this:

Date	Concept	Value	Rank
2021-01-01	A	20	1
2021-01-01	B	30	3
2021-01-01	C	25	2
2021-02-01	A	17	1
2021-02-01	B	31	2
2021-02-01	C	42	3

Is there any elegant way to do it with pandas?

Answer 1

Use transform with pd.Series.rank :

df = pd.DataFrame({'Date': {0: '2021-01-01 ', 1: '2021-01-01 ', 2: '2021-01-01 ', 3: '2021-02-01 ', 4: '2021-02-01 ', 5: '2021-02-01 '}, 
                   'Concept ': {0: 'A ', 1: 'B ', 2: 'C ', 3: 'A ', 4: 'B ', 5: 'C '}, 
                   'Value': {0: 20, 1: 30, 2: 25, 3: 17, 4: 31, 5: 42}})

df['Rank'] = df.groupby(['Date'])['Value'].transform(pd.Series.rank)

          Date Concept   Value  Rank
0  2021-01-01        A      20     1
1  2021-01-01        B      30     3
2  2021-01-01        C      25     2
3  2021-02-01        A      17     1
4  2021-02-01        B      31     2
5  2021-02-01        C      42     3