运行时错误：“Byte”未实现“reflection_pad2d”

Question

padding = (2, 2, 2, 2)
img = torch.nn.functional.pad(img, padding, mode='reflect')
out = torch.nn.functional.conv2d(img, kernel, groups=img.shape[1])

以下是完整的错误：

  File "/home/amir/PycharmProjects/LPTN/loadPretrainedModel.py", line 57, in conv_gauss
    img = torch.nn.functional.pad(img, padding, mode='reflect')
  File "/usr/local/lib/python3.8/dist-packages/torch/nn/functional.py", line 4017, in _pad
    return torch._C._nn.reflection_pad2d(input, pad)
RuntimeError: "reflection_pad2d" not implemented for 'Byte'

你认为问题出在哪里？我想不出来。提前谢谢

Answer 1

您需要将img的数据类型更改为浮动，例如img.float()。许多操作（如reflection_pad2d）仅针对浮点张量实现