The first step, one loss; the second step, load + resume -1, two loss, and bug comes.

sanghyun-son / EDSR-PyTorch

PyTorch version of the paper 'Enhanced Deep Residual Networks for Single Image Super-Resolution' (CVPRW 2017)

MIT License

2.43k stars 669 forks source link

The first step, one loss; the second step, load + resume -1, two loss, and bug comes. #247

Closed splinter21 closed 4 years ago

splinter21 commented 4 years ago

Traceback (most recent call last): File "main.py", line 25, in t.train() File "/home/liujing/EDSR/src_vqm2/trainer.py", line 41, in train self.loss.start_log() File "/home/liujing/EDSR/src_vqm2/loss/init.py", line 129, in start_log self.log = torch.cat((self.log, torch.zeros(1, len(self.loss)))) RuntimeError: invalid argument 0: Sizes of tensors must match except in dimension 0. Got 1 and 3 in dimension 1 at /pytorch/aten/src/TH/generic/THTensorMath.cpp:3616