Open thirteenbird opened 6 years ago
Hello.
Did you set --n_GPUs
larger than 1?
Please let me know the input argument you used to execute the code.
Thank you.
I didn't change the setting of --n_GPUs, so it isn't larger than 1. i just use this : python main.py --model EDSR --scale 2 --patch_size 96 --save EDSR_baseline_x2 --reset --cpu and run
I only modified the places you said need to be modified, only --dir_data and --ext.
Can I ask your OS and PyTorch version?
I checked that --cpu
works well with my environment (Ubuntu 16.04, PyTorch 0.4.1)
After i run the code, cmd write out: Making model... Preparing loss function: 1.000 * L1 Traceback (most recent call last): File "main.py", line 19, in
t.train()
File "/home/ubuntu/Desktop/EDSR-PyTorch-2/src/trainer.py", line 34, in train
self.loss.step()
File "/home/ubuntu/Desktop/EDSR-PyTorch-2/src/loss/init.py", line 87, in step
for l in self.get_loss_module():
File "/home/ubuntu/Desktop/EDSR-PyTorch-2/src/loss/init.py", line 123, in get_loss_module
return self.loss_module.module
File "/home/ubuntu/anaconda3/lib/python3.6/site-packages/torch/nn/modules/module.py", line 518, in getattr
type(self).name, name))
AttributeError: 'ModuleList' object has no attribute 'module'
i don't know why it comes out this problem