about transpose in line 619-629 of utils.py about dag-gnn HOT 3 OPEN

@Wzz666 are you sure? I have been using it and it seems to work fine.

from dag-gnn.

@Wzz666 are you sure? I have been using it and it seems to work fine.
@ItsyPetkov
hello！
Thank you for your comment!
I'm confused because after running the code, I got the following results.
A = np.array([[0., 0., 0., 2.], [0., 0., 3., 0.],[1., 0., 0., 0.],[0., 0., 6., 0.]])

def preprocess_adj_new(adj):
adj_normalized = (torch.eye(adj.shape[0]).double() - (adj.transpose(0,1)))
return adj_normalized
print(A)

array([[0., 0., 0., 2.],
[0., 0., 3., 0.],
[1., 0., 0., 0.],
[0., 0., 6., 0.]])

print(preprocess_adj_new(A))

tensor([[ 1., 0., 0., -2.],
[ 0., 1., -3., 0.],
[-1., 0., 1., 0.],
[ 0., 0., -6., 1.]], dtype=torch.float64)

It seem that i got (I - A) not (I - A^T).

from dag-gnn.

@Wzz666 you should read how to method works. It does exactly what it is supposed to do

from dag-gnn.