DataFrame.to_dict(orient='dict')

DataFrame.to_dict(orient=’dict’)

>>> df = pd.DataFrame({'name':[1,2,3],"class":[11,22,33],"price":[111,222,333]})
>>> df
   class  name  price
0     11     1    111
1     22     2    222
2     33     3    333

orient : str {‘dict’, ‘list’, ‘series’, ‘split’, ‘records’, ‘index’}
Determines the type of the values of the dictionary.

dict (default) : dict like {column -> {index -> value}}

>>> df.to_dict(orient="dict")
{'class': {0: 11, 1: 22, 2: 33}, 'name': {0: 1, 1: 2, 2: 3}, 'price': {0: 111, 1: 222, 2: 333}}

list : dict like {column -> [values]}

>>> df.to_dict(orient="list")
{'class': [11, 22, 33], 'name': [1, 2, 3], 'price': [111, 222, 333]}

series : dict like {column -> Series(values)}

>>> df.to_dict(orient="series")
{'class': 0    11
1    22
2    33
Name: class, dtype: int64, 'name': 0    1
1    2
2    3
Name: name, dtype: int64, 'price': 0    111
1    222
2    333
Name: price, dtype: int64}

split : dict like {index -> [index], columns -> [columns], data -> [values]}

>>> df.to_dict(orient="split")
{'index': [0, 1, 2], 'columns': ['class', 'name', 'price'], 'data': [[11, 1, 111], [22, 2, 222], [33, 3, 333]]}

records : list like [{column -> value}, … , {column -> value}]

>>> df.to_dict(orient="records")
[{'class': 11, 'name': 1, 'price': 111}, {'class': 22, 'name': 2, 'price': 222}, {'class': 33, 'name': 3, 'price': 333}]

index : dict like {index -> {column -> value}}

>>> df.to_dict(orient="index")
{0: {'class': 11, 'name': 1, 'price': 111}, 1: {'class': 22, 'name': 2, 'price': 222}, 2: {'class': 33, 'name': 3, 'price': 333}}