RuntimeError: The size of tensor a (17) must match the size of tensor b (18) at non-singleton dimension 1

[marma1ade]

I have this issue in function embedding_rep_loss, the distance_matrix size is 1818, the trans_matrix size is 1717,
when doing :
" rep_loss = torch.sum(torch.tensor(trans_matrix).cuda() * distance_matrix) "
this error occured

[sxlong0205]

同问，这个问题要怎么解决

[18582088138]

This issue is due to the use of "args.embedding_emd = True"
In "emd_task_distill.py" +402 , use embedding_rep_loss function that make the tensor shape doesn't match
So make the args.embedding_emd = False can fix this issue

[lxk00]

Sorry there is indeed an error here. Code are fixed.