a = torch.tensor([[1, 2, 3], [4, 5, 6]], dtype=torch.float) # torch.float默认为torch.float32
print(a.shape) # torch.Size([2, 3])
print(a.size()) # torch.Size([2, 3])
print(a.dim()) # 2
print(a.dtype) # torch.float32
print(a.type()) # torch.FloatTensor
print(type(a)) # <class 'torch.Tensor'>
b = cv2.imread(r'E:\000.jpg')
print(b.shape) # (900, 1440, 3)
print(b.size) # 3888000
print(b.ndim) # 3
print(b.dtype) # uint8
print(type(b)) # <class 'numpy.ndarray'>
Propiedades de los tensores frente a las matrices
Supongo que te gusta
Origin blog.csdn.net/weixin_48158964/article/details/132401657
Recomendado
Clasificación