NAFNet

NAFNet copied to clipboard

你好， 请问NAFNet在Denoise中如何测试大图像呢，要切成多个小图像再进行测试吗？

958984240， 您好，

Denoise的大图像测试和其他任务并无不同， 一般来说建议整图测试。

958984240， 您好，

Denoise的大图像测试和其他任务并无不同， 一般来说建议整图测试。

您好，我现在用自己的数据集，按原train.py代码在训练的时候进行val的测试，但是在测试部分图片后会报这样的错误 File "～/Projects/NAFNet/basicsr/train.py", line 305, in main() File "～/Projects/NAFNet/basicsr/train.py", line 270, in main model.validation(val_loader, current_iter, tb_logger, File "～/Projects/NAFNet/basicsr/models/base_model.py", line 55, in validation return self.dist_validation(dataloader, current_iter, tb_logger, save_img, rgb2bgr, use_image) File "～/Projects/NAFNet/basicsr/models/image_restoration_model.py", line 306, in dist_validation self.test() File "～/Projects/NAFNet/basicsr/models/image_restoration_model.py", line 268, in test pred = self.net_g(self.lq[i:j]) File "～/anaconda3/envs/pytorch111_cuda113/lib/python3.9/site-packages/torch/nn/modules/module.py", line 1110, in _call_impl return forward_call(*input, **kwargs) File "～/anaconda3/envs/pytorch111_cuda113/lib/python3.9/site-packages/torch/nn/parallel/distributed.py", line 963, in forward output = self.module(*inputs[0], **kwargs[0]) File "～/anaconda3/envs/pytorch111_cuda113/lib/python3.9/site-packages/torch/nn/modules/module.py", line 1110, in _call_impl return forward_call(*input, **kwargs) File "～/Projects/NAFNet/basicsr/models/archs/NAFNet_arch.py", line 173, in forward x = encoder(x) File "～/anaconda3/envs/pytorch111_cuda113/lib/python3.9/site-packages/torch/nn/modules/module.py", line 1110, in _call_impl return forward_call(*input, **kwargs) File "～/anaconda3/envs/pytorch111_cuda113/lib/python3.9/site-packages/torch/nn/modules/container.py", line 141, in forward input = module(input) File "～/anaconda3/envs/pytorch111_cuda113/lib/python3.9/site-packages/torch/nn/modules/module.py", line 1110, in _call_impl return forward_call(*input, **kwargs) File "～/Projects/NAFNet/basicsr/models/archs/NAFNet_arch.py", line 80, in forward x = self.conv2(x) File "～/anaconda3/envs/pytorch111_cuda113/lib/python3.9/site-packages/torch/nn/modules/module.py", line 1110, in _call_impl return forward_call(*input, **kwargs) File "～/anaconda3/envs/pytorch111_cuda113/lib/python3.9/site-packages/torch/nn/modules/conv.py", line 447, in forward return self._conv_forward(input, self.weight, self.bias) File "～/anaconda3/envs/pytorch111_cuda113/lib/python3.9/site-packages/torch/nn/modules/conv.py", line 443, in _conv_forward return F.conv2d(input, weight, bias, self.stride, RuntimeError: Expected canUse32BitIndexMath(input) && canUse32BitIndexMath(output) to be true, but got false. (Could this error message be improved? If so, please report an enhancement request to PyTorch.)

查资料后没找到具体解决方法，不知道是否由于图片的分辨率大的原因。（测试的图片分辨率不同，类似1683，1230，3这种宽高都是1000多*1000多的图片）。