Group a dataset through python. for eg. these 4 rows : REPORT_DAT HOOD_140 REPORT_DOW DIVISION LOCATION_TYPE PREMISES_TYPE

Fantastic news! We've Found the answer you've been seeking!

Question:

Group a dataset through python.

for eg. these 4 rows :

REPORT_DAT	HOOD_140	REPORT_DOW	DIVISION	LOCATION_TYPE	PREMISES_TYPE	OFFENCE	Neighbourhood	Count_of_police_force	Average Individual Employment Income	Population	Unemployment Rate	Low Income population(LIM-AT)	EVENT_UNIQUE_ID
2018-01-01	76	Monday	D52	Commercial Dwelling Unit (Hotel, Motel, B & B, Short Term Rental)	Commercial	Assault	Bay Street Corridor	259	63,009	25,797	10.2	9,665	1
2018-01-01	76	Monday	D52	Other Commercial / Corporate Places (For Profit, Warehouse, Corp. Bldg	Commercial	Assault	Bay Street Corridor	259	63,009	25,797	10.2	9,665	1
2018-01-02	76	Tuesday	D52	Apartment (Rooming House, Condo)	Apartment	Assault	Bay Street Corridor	259	63,009	25,797	10.2	9,665	1
2018-01-02	76	Tuesday	D52	Streets, Roads, Highways (Bicycle Path, Private Road)	Outside	Assault	Bay Street Corridor	259	63,009	25,797	10.2	9,665	1

I want to group the 2 dates 1st jan had sum(event_unique_id) =2 , same way 2nd jan had sum(event_unique_id) =2.

i use this code :

grouped_crimedf= crime_df.groupby(['HOOD_140','REPORT_DAT','REPORT_DOW','DIVISION','LOCATION_TYPE','PREMISES_TYPE','OFFENCE','Neighbourhood','Count_of_police_force','Average Individual Employment Income','Population','Unemployment Rate','Low Income population(LIM-AT)']).agg({'EVENT_UNIQUE_ID': 'count'}).reset_index()

I have to use all the columns because if i drop some of them from the above statement then i dont get them in the grouped_df dataset which means i cannot apply LR later on on these columns. how can i get all the columns and at the same time group it based on only date and hood_140.