xujli / mie-acl-2020-torch

Torch implementation of MIE approach in ACL 2020

MIE-ACL-2020-torch

Torch implementation of MIE approach in ACL 2020

mie-acl-2020-torch's Issues

Potential Bug?

Epoch 1:: 0%|▎ | 1/370 [00:03<23:41, 3.85s/it]
Traceback (most recent call last):
File "/root/autodl-nas/MIE-ACL-2020-torch/main.py", line 51, in
model.train(
File "/root/autodl-nas/MIE-ACL-2020-torch/src/network.py", line 589, in train
slot_pred_labels, num = self.Dmodels[slot](self.slot_utt_hs_dict[slot], self.slot_candidate_cs_dict[slot],
File "/root/miniconda3/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1110, in _call_impl
return forward_call(*input, **kwargs)
File "/root/autodl-nas/MIE-ACL-2020-torch/src/network.py", line 188, in forward
torch.matmul(
RuntimeError: The size of tensor a (35) must match the size of tensor b (1225) at non-singleton dimension 0
root@container-893311b7ac-b1877a90:~/autodl-nas/MIE-ACL-2020-torch#

When I try to run the code, I got the message above. Could you check that please? Thanks!