search

xinntao / Real-ESRGAN

Real-ESRGAN aims at developing Practical Algorithms for General Image/Video Restoration.
BSD 3-Clause "New" or "Revised" License
28.55k stars 3.58k forks source link

Train scale x2 model #81

Open yangmingwanli opened 3 years ago

yangmingwanli commented 3 years ago

Hi, thanks for the great works!

I was trying to finetune a x2 model with own dataset, got these errors of tensor size mismatch. Scale x4 works fine though.

Data prep included multi-scale images step. Opt file was modified from finetune_realesrgan_x4plus.yml with scale set to 4, data path updated and batch_size_per_gpu changed per gpu limit.

Any suggestions on what might be missing?

python realesrgan/train.py -opt options/finetune_realesrgan_x2plus.yml

Log 2021-09-16 19:54:44,375 INFO: Start training from epoch: 0, iter: 0 C:\Users\xxxxx.conda\envs\torch\lib\site-packages\torch\nn\functional.py:3063: UserWarning: Default upsampling behavior when mode=bilinear is changed to align_corners=False since 0.4.0. Please specify align_corners=True if the old behavior is desired. See the documentation of nn.Upsample for details. "See the documentation of nn.Upsample for details.".format(mode)) C:\Users\xxxxx.conda\envs\torch\lib\site-packages\torch\nn\functional.py:3103: UserWarning: The default behavior for interpolate/upsample with float scale_factor changed in 1.6.0 to align with other frameworks/libraries, and now uses scale_factor directly, instead of relying on the computed output size. If you wish to restore the old behavior, please set recompute_scale_factor=True. See the documentation of nn.Upsample for details. warnings.warn("The default behavior for interpolate/upsample with float scale_factor changed " C:\Users\xxxxx.conda\envs\torch\lib\site-packages\torch\nn\functional.py:3063: UserWarning: Default upsampling behavior when mode=bicubic is changed to align_corners=False since 0.4.0. Please specify align_corners=True if the old behavior is desired. See the documentation of nn.Upsample for details. "See the documentation of nn.Upsample for details.".format(mode)) C:\Users\xxxxx.conda\envs\torch\lib\site-packages\basicsr\losses\losses.py:16: UserWarning: Using a target size (torch.Size([2, 3, 256, 256])) that is different to the input size (torch.Size([2, 3, 512, 512])). This will likely lead to incorrect results due to broadcasting. Please ensure they have the same size. return F.l1_loss(pred, target, reduction='none') Traceback (most recent call last): File "realesrgan/train.py", line 11, in train_pipeline(root_path) File "C:\Users\xxxxx.conda\envs\torch\lib\site-packages\basicsr\train.py", line 167, in train_pipeline model.optimize_parameters(current_iter) File "c:\users\xxxxx\real-esrgan\realesrgan\models\realesrgan_model.py", line 200, in optimize_parameters l_g_pix = self.cri_pix(self.output, l1_gt) File "C:\Users\xxxxx.conda\envs\torch\lib\site-packages\torch\nn\modules\module.py", line 727, in _call_impl result = self.forward(*input, *kwargs) File "C:\Users\xxxxx.conda\envs\torch\lib\site-packages\basicsr\losses\losses.py", line 55, in forward return self.loss_weight l1_loss(pred, target, weight, reduction=self.reduction) File "C:\Users\xxxxx.conda\envs\torch\lib\site-packages\basicsr\losses\loss_util.py", line 91, in wrapper loss = loss_func(pred, target, **kwargs) File "C:\Users\xxxxx.conda\envs\torch\lib\site-packages\basicsr\losses\losses.py", line 16, in l1_loss return F.l1_loss(pred, target, reduction='none') File "C:\Users\xxxxx.conda\envs\torch\lib\site-packages\torch\nn\functional.py", line 2633, in l1_loss expanded_input, expanded_target = torch.broadcast_tensors(input, target) File "C:\Users\xxxxx.conda\envs\torch\lib\site-packages\torch\functional.py", line 71, in broadcast_tensors return _VF.broadcast_tensors(tensors) # type: ignore RuntimeError: The size of tensor a (512) must match the size of tensor b (256) at non-singleton dimension 3

xinntao commented 3 years ago

It seems that your data have the wrong size.

Try to check the shape of your network output and the data.