Single-card inference - Githubissues

After training on server, I got DS-GAN-Epoch300.pth, I wanted to use this weight for single-card inference.

I changed device_ids = [0, 1, 2, 3] to device_ids = [0], and ran infer.py, then I got this error:

Traceback (most recent call last):
  File "/data/PosterLayout-CVPR2023/infer.py", line 114, in <module>
    main()
  File "/data/PosterLayout-CVPR2023/infer.py", line 111, in main
    test(G, testing_dl, 1)
  File "/data/PosterLayout-CVPR2023/infer.py", line 58, in test
    cls, box = G(imgs, fix_noise)
  File "/data/anaconda3/envs/pl/lib/python3.9/site-packages/torch/nn/modules/module.py", line 1130, in _call_impl
    return forward_call(*input, **kwargs)
  File "/data/anaconda3/envs/pl/lib/python3.9/site-packages/torch/nn/parallel/data_parallel.py", line 166, in forward
    return self.module(*inputs[0], **kwargs[0])
  File "/data/anaconda3/envs/pl/lib/python3.9/site-packages/torch/nn/modules/module.py", line 1130, in _call_impl
    return forward_call(*input, **kwargs)
  File "/data/PosterLayout-CVPR2023/model.py", line 99, in forward
    lstm_output = self.cnnlstm(layout, h0)
  File "/data/anaconda3/envs/pl/lib/python3.9/site-packages/torch/nn/modules/module.py", line 1130, in _call_impl
    return forward_call(*input, **kwargs)
  File "/data/PosterLayout-CVPR2023/model.py", line 78, in forward
    output, _ = self.lstm(x, (torch.zeros_like(h0).contiguous(), h0.contiguous()))
  File "/data/anaconda3/envs/pl/lib/python3.9/site-packages/torch/nn/modules/module.py", line 1130, in _call_impl
    return forward_call(*input, **kwargs)
  File "/data/anaconda3/envs/pl/lib/python3.9/site-packages/torch/nn/modules/rnn.py", line 767, in forward
    self.check_forward_args(input, hx, batch_sizes)
  File "/data/anaconda3/envs/pl/lib/python3.9/site-packages/torch/nn/modules/rnn.py", line 693, in check_forward_args
    self.check_hidden_size(hidden[0], self.get_expected_hidden_size(input, batch_sizes),
  File "/data/anaconda3/envs/pl/lib/python3.9/site-packages/torch/nn/modules/rnn.py", line 226, in check_hidden_size
    raise RuntimeError(msg.format(expected_hidden_size, list(hx.size())))
RuntimeError: Expected hidden[0] size (8, 4, 256), got [8, 1, 256]

Would you please try this, or offer some tips and instructions on single-card inference? Thank you!

PKU-ICST-MIPL / PosterLayout-CVPR2023

Single-card inference #5