Конвертировать Pandas DataFrame с дубликатом в словарь

Question

У меня есть Pandas Dataframe (пример):

+-----------+---------+------------+-------------------+-------------+---------+
| main_name |  from   |   where    |       time        |    field    | measure |
+-----------+---------+------------+-------------------+-------------+---------+
| Students  | table_s | source = 4 | current_date - 1  | students_id | count   |
| Students  | table_s | type = 5   | current_date - 1  | students_id | count   |
| Students  | table_s | age = 20   | current_date - 1  | students_id | count   |
| Teacher   | table_t | source = 6 | current_date - 30 | b           | sum     |
| Teacher   | table_t | type = 6   | current_date - 30 | b           | sum     |
| Teacher   | table_t | age = 40   | current_date - 30 | b           | sum     |
+-----------+---------+------------+-------------------+-------------+---------+

Мне нужно преобразовать его в словарь

Ожидаемый результат:

{'students': {'from':[tables_s] , 'where':[source = 4, type =5 , age=20], 'field' = [students_id], 'measure' = [count], 'time': [current_date - 1]}

Я пытался использовать groupby и to_dict () .Но я не понимаю, что Thx