Small_Lemon_Tree的博客

示例文件 test.pyimport torchA = torch.arange(6)A = A.reshape(1, 1, 2, 3, -1)B = A.squeeze()print(A)print(B)C = B.unsqueeze(0)D = B.unsqueeze(1)E = B.unsqueeze(2)print(C)print(D)print(E)终端命令行和运行结果<user>python test.pytensor([[[[[0],

tensor.view()和tensor.reshape()示例文件 test.pyimport torcha = torch.randint(0, 10, (3, 4))b = a.view(2, 6)c = a.reshape(2, 6)print(id(a)==id(b), id(a)==id(c), id(b)==id(c))print(id(a.storage())==id(b.storage()), id(a.storage())==id(c.storage()),

示例文件 test.pyimport torchcor1 = torch.arange(5).float()cor2 = torch.arange(4).float()print(cor1)print(cor2)X, Y = torch.meshgrid(cor1, cor2)print(X)print(X.permute(0, 1))print(X.permute(1, 0))print(X.transpose(0, 1))print(X.transpose(1, 0))终端

示例文件 test.pyimport torchcor1 = torch.arange(5).float()cor2 = torch.arange(4).float()print(cor1)print(cor2)X, Y = torch.meshgrid(cor1, cor2)print(X)print(Y)cor1 = torch.arange(3).float()cor2 = torch.arange(5).float()cor3 = torch.arange(4).floa

示例文件 test.pyimport torchcor1 = torch.arange(5).float()cor2 = torch.arange(4).float()print(cor1)print(cor2)X, Y = torch.meshgrid(cor1, cor2)print(X)print(Y)终端命令行及运行结果<user>python test.pytensor([0., 1., 2., 3., 4.])tensor([0., 1., 2., 3.

示例文件 test.pyimport torchx = torch.arange(0, 6)print(x)print(x.type())y = torch.range(0, 6)print(y)print(y.type())终端命令行和运行结果<user>python test.pytensor([0, 1, 2, 3, 4, 5])torch.LongTensortensor([0., 1., 2., 3., 4., 5., 6.])torch.FloatTen

Tensor类型的转换Pytorch数据类型的转换可以通过三个方式：1）调用Tensor成员函数long()，int()，double()，float()，byte()2）调用Tensor成员函数type()，传入数据类型3）调用Tensor成员函数type_as()，传入实例对象示例文件 test.pyimport torcha = torch.randn(2, 3)print(a.type())b = a.int()c = a.type(torch.LongTensor)pri

基本类型Data typeCPU tensorGPU tensor32-bit floating pointtorch.FloatTensortorch.cuda.FloatTensor64-bit floating pointtorch.DoubleTensortorch.cuda.DoubleTensor16-bit floating pointN/Atorch.cuda.HalfTensor8-bit integer(unsigned)torch

(1) CPU张量 ----> GPU张量，使用Tensor.cuda()(2) GPU张量 ----> CPU张量，使用Tensor.cpu()我们可以通过torch.cuda.is_available()函数来判断当前的环境是否支持GPU，如果支持，则返回GPU类型，否则返回CPU类型。def commonType(Tensor): if torch.cuda.is_available(): cuda = "cuda:0" return Tens