Pretrained-IPT icon indicating copy to clipboard operation
Pretrained-IPT copied to clipboard

MACOS系统想复现图像去雨功能,如何修改代码可以绕过使用cuda的部分?

Open fuhao66 opened this issue 2 years ago • 2 comments

fuhao66 avatar May 07 '22 06:05 fuhao66

Traceback (most recent call last): File "/Users/liufuhao/Pretrained-IPT/main.py", line 37, in main() File "/Users/liufuhao/Pretrained-IPT/main.py", line 33, in main t.test() File "/Users/liufuhao/Pretrained-IPT/trainer.py", line 42, in test sr = self.model(rain, idx_scale) File "/Users/liufuhao/Anaconda/anaconda3/envs/pytorch/lib/python3.9/site-packages/torch/nn/modules/module.py", line 1110, in _call_impl return forward_call(input, **kwargs) File "/Users/liufuhao/Pretrained-IPT/model/init.py", line 58, in forward return forward_function(x) File "/Users/liufuhao/Pretrained-IPT/model/init.py", line 169, in forward_chop y_h_cut = self.cut_h(x_h_cut, h, w, h_cut, w_cut, padsize, shave, scale, batchsize) File "/Users/liufuhao/Pretrained-IPT/model/init.py", line 215, in cut_h y_h_cut_unfold.append(P.data_parallel(self.model, x_h_cut_unfold[ibatchsize:(i+1)*batchsize,...], range(self.n_GPUs)).cpu()) File "/Users/liufuhao/Anaconda/anaconda3/envs/pytorch/lib/python3.9/site-packages/torch/nn/parallel/data_parallel.py", line 208, in data_parallel output_device = device_ids[0] IndexError: range object index out of range

fuhao66 avatar May 07 '22 07:05 fuhao66

把代码中 .cuda() 全部去掉即可。

HantingChen avatar Jun 06 '22 06:06 HantingChen