从 pandas dataframe 创建字典，其值为元组

Question

I have a dataframe like this:

>>> df = pd.DataFrame(np.array([[1, 2], ['a', 'b']]), columns=['col1', 'col2'])
>>> df
  col1 col2
0    1    2
1    a    b

What I am trying to do is to create a dict from this dataframe where row index is the key and col1 and col2 are a value in a tuple form for my dict. Here is what I am doing but this returns a list:

>>> import numpy as np
>>> import pandas as pd
>>> df.T.to_dict('list')
{0: ['1', '2'], 1: ['a', 'b']}

Here is what I am trying to get:

{0: ('1', '2'), 1: ('a', 'b')}

Answer 1

Lets try

df.agg(tuple,1).to_dict()

Answer 2

Another way is to take the dict you have apply a dict comprehension:

>>> {k: tuple(v) for k, v in df.T.to_dict("list").items()}
{0: ('1', '2'), 1: ('a', 'b')}

Answer 3

Not sure about the speed but with itertuples

dict(zip(df.index, df.itertuples(index=False, name=None)))
{0: ('1', '2'), 1: ('a', 'b')}

Or

dict(zip(df.index,map(tuple,df.values)))