Pandas dataframe to dictionary with value as a list

Question

We have the following dataframe df,

   0   1  2  3
0 'a' 'b' 1  2
1 'a' 'c' 2  4

I want output as

{('a', 'b'): [1, 2], ('a', 'c'): [2, 4]}

How to use .to_dict() to generate the output?

I tried

df.set_index([0,1])[[2,3]].to_dict()

and get the result

{2: {('a', 'b'): 1, ('a', 'c'): 2}, 3: {('a', 'b'): 2, ('a', 'c'): 4}}

Answer 1

After setting the index, take the transpose and then use to_dict with orient='list' :

df.set_index([0,1]).T.to_dict(orient='list')

The resulting output:

{('a', 'b'): [1, 2], ('a', 'c'): [2, 4]}

Answer 2

Try:

{k: v.tolist() for k, v in df.set_index([0, 1]).iterrows()}

{('a', 'b'): [1, 2], ('a', 'c'): [2, 4]}