How to solve the RuntimeError: data set to a tensor that requires gradients must be floating point or complex dtype

[liaoyil]

my test_A and test_B is png photo ,
when I use "python test.py --name arch2cons --netG local --ngf 32 --resize_or_crop none --label_nc 0 --dataroot "datasets/arch2cons" --input_nc 3 --no_instance ",
my terminal show me "
Pretrained network G has fewer layers; The following are not initialized:
['model', 'model1_1', 'model1_2', 'model2_1', 'model2_2', 'model3_1', 'model3_2']
Traceback (most recent call last):
File "D:\DC\Code\Code\pix2pixHD-master\test.py", line 33, in
model.type(torch.uint8)
File "D:\anaconda3\envs\pytorch\Lib\site-packages\torch\nn\modules\module.py", line 974, in type
return self._apply(lambda t: t.type(dst_type))
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "D:\anaconda3\envs\pytorch\Lib\site-packages\torch\nn\modules\module.py", line 802, in _apply
module._apply(fn)
File "D:\anaconda3\envs\pytorch\Lib\site-packages\torch\nn\modules\module.py", line 802, in _apply
module._apply(fn)
File "D:\anaconda3\envs\pytorch\Lib\site-packages\torch\nn\modules\module.py", line 802, in _apply
module._apply(fn)
File "D:\anaconda3\envs\pytorch\Lib\site-packages\torch\nn\modules\module.py", line 828, in _apply
param.data = param_applied
^^^^^^^^^^
RuntimeError: data set to a tensor that requires gradients must be floating point or complex dtype"
I'm just a sophomore with very little understanding about this , who can help me?