RuntimeError: Sizes of tensors must match except in dimension 2. Got 31 and 30 (The offending index is 0)

[JingHSY-dev]

Traceback (most recent call last):
File "inference_for_demo_video.py", line 238, in
no_move=False,
File "/home/yaohs/miniconda3/envs/dreamtalk/lib/python3.7/site-packages/torch/autograd/grad_mode.py", line 27, in decorate_context
return func(*args, **kwargs)
File "/home/yaohs/Work/dreamtalk/generators/utils.py", line 102, in render_video
output_dict = net_G(cur_src_img, win_exp)
File "/home/yaohs/miniconda3/envs/dreamtalk/lib/python3.7/site-packages/torch/nn/modules/module.py", line 889, in _call_impl
result = self.forward(*input, **kwargs)
File "/home/yaohs/Work/dreamtalk/generators/face_model.py", line 35, in forward
output = self.warpping_net(input_image, descriptor)
File "/home/yaohs/miniconda3/envs/dreamtalk/lib/python3.7/site-packages/torch/nn/modules/module.py", line 889, in _call_impl
result = self.forward(*input, **kwargs)
File "/home/yaohs/Work/dreamtalk/generators/face_model.py", line 94, in forward
output = self.hourglass(input_image, descriptor)
File "/home/yaohs/miniconda3/envs/dreamtalk/lib/python3.7/site-packages/torch/nn/modules/module.py", line 889, in _call_impl
result = self.forward(*input, **kwargs)
File "/home/yaohs/Work/dreamtalk/generators/base_function.py", line 39, in forward
return self.decoder(self.encoder(x, z), z)
File "/home/yaohs/miniconda3/envs/dreamtalk/lib/python3.7/site-packages/torch/nn/modules/module.py", line 889, in _call_impl
result = self.forward(*input, **kwargs)
File "/home/yaohs/Work/dreamtalk/generators/base_function.py", line 89, in forward
out = torch.cat([out, x.pop()], 1) if self.skip_connect else out
RuntimeError: Sizes of tensors must match except in dimension 2. Got 31 and 30 (The offending index is 0)