Open qingfengmingyue opened 5 years ago
@qingfengmingyue
i found the working fix, please add the below code will fix this problem(just comment the add_module for scale_kernel instead create scale_kernel variable).
#self.add_module('scale_kernel', kernel_weight) self.scale_kernel = torch.tensor(np.tile(kernel, (3, 1, 1, 1))).float()
@Dayananda-V Can you show the result? I just figure out the mosaic.
Traceback (most recent call last): File "train.py", line 2, in
main(mode=1)
File "/root/edge-informed-sisr/main.py", line 42, in main
model = EdgeMatch(config)
File "/root/edge-informed-sisr/src/edge_match.py", line 27, in init
self.sr_model = SRModel(config).to(config.DEVICE)
File "/root/edge-informed-sisr/src/models.py", line 179, in init
self.add_module('scale_kernel', kernel_weight)
File "/root/anaconda3/lib/python3.7/site-packages/torch/nn/modules/module.py", line 179, in add_module
torch.typename(module)))
TypeError: torch.FloatTensor is not a Module subclass