pytorch错误记录:TypeError: Cannot handle this data type: (1, 1, 64),
经过测试:
pairwise_distance函数在pytorch1.7上的表现与pytorch1.10/pytorch1.11上有区别,代码如下
pytorch1.7上
feat_A的size是1,64,64,64,feat_B的size是1,64,64,64
出来应该size是1,1,64,64
而在pytorch1.10/pytorch1.11
出来的size是1,64,64,1
import torch.nn.functional as Fself.dist = F.pairwise_distance(self.feat_A, self.feat_B, keepdim=True)
直接导致了进一步的计算图像大小存不出来
image_pil = Image.fromarray(image_numpy)File "D:\software\Anaconda3\envs\pt3.8\lib\site-packages\PIL\Image.py", line 2815, in fromarrayraise TypeError("Cannot handle this data type: %s, %s" % typekey) from e
TypeError: Cannot handle this data type: (1, 1, 64), |u1
解决方式只能是版本回退
pytorch错误记录:TypeError: Cannot handle this data type: (1, 1, 64),
经过测试:
pairwise_distance函数在pytorch1.7上的表现与pytorch1.10/pytorch1.11上有区别,代码如下
pytorch1.7上
feat_A的size是1,64,64,64,feat_B的size是1,64,64,64
出来应该size是1,1,64,64
而在pytorch1.10/pytorch1.11
出来的size是1,64,64,1
import torch.nn.functional as Fself.dist = F.pairwise_distance(self.feat_A, self.feat_B, keepdim=True)
直接导致了进一步的计算图像大小存不出来
image_pil = Image.fromarray(image_numpy)File "D:\software\Anaconda3\envs\pt3.8\lib\site-packages\PIL\Image.py", line 2815, in fromarrayraise TypeError("Cannot handle this data type: %s, %s" % typekey) from e
TypeError: Cannot handle this data type: (1, 1, 64), |u1
解决方式只能是版本回退