demo.py does not working

[yustiks]

I try to run demo.py after several epochs of training, and I have the following error:
lib/python3.7/site-packages/torch/nn/modules/module.py", line 1045, in load_state_dict self.__class__.__name__, "\n\t".join(error_msgs))) RuntimeError: Error(s) in loading state_dict for DataParallel: Missing key(s) in state_dict:

any ideas what is it?

[WenjiaWang0312]

You said you only have 1 GPU? That's just the conflicts between multi-gpu model with single-gpu model. could you change the config in config/super_resolution.yaml as ngpu: 1 and try again

[yustiks]

thank you, @JasonBoy1 ! I set it to be one. And it still has errors!

  batch_size: 64
  width: 128
  height: 32
  epochs: 500
  cuda: True
  ngpu: 1
  workers: 1
  resume: ''
  ckpt_dir: './ckpt/'
  voc_type: 'all' #'digits lower upper all'
  saveInterval: 200
  displayInterval: 100 #display loss
  adadelta: False
  lr: 0.001
  adam: True
  beta1: 0.5
  manualSeed: 1234
  max_len: 100
  keep_ratio: False
  down_sample_scale: 2

[yustiks]

ok, I solved the problem by the next code in the interfaces/base.py:

                if self.config.TRAIN.ngpu == 1:
                    model.load_state_dict(torch.load(self.resume)['state_dict_G'])

so, I changed it to be:

                if self.config.TRAIN.ngpu == 1:
                    state_dict = torch.load(self.resume)['state_dict_G']
                    from collections import OrderedDict
                    new_state_dict = OrderedDict()

                    for k, v in state_dict.items():
                        if 'module' not in k:
                            k = 'module.'+k
                        else:
                            k = k.replace('features.module.', 'module.features.')
                        new_state_dict[k]=v

                    model.load_state_dict(new_state_dict)
                    # model.load_state_dict(torch.load(self.resume)['state_dict_G'])

[LMY199]

ok, I solved the problem by the next code in the interfaces/base.py:

                if self.config.TRAIN.ngpu == 1:
                    model.load_state_dict(torch.load(self.resume)['state_dict_G'])

so, I changed it to be:

                if self.config.TRAIN.ngpu == 1:
                    state_dict = torch.load(self.resume)['state_dict_G']
                    from collections import OrderedDict
                    new_state_dict = OrderedDict()

                    for k, v in state_dict.items():
                        if 'module' not in k:
                            k = 'module.'+k
                        else:
                            k = k.replace('features.module.', 'module.features.')
                        new_state_dict[k]=v

                    model.load_state_dict(new_state_dict)
                    # model.load_state_dict(torch.load(self.resume)['state_dict_G'])

really thanku，i have the same problem and solved now by your code,good for u