译者:hijkzzz
torch.Tensor
是一种包含单一数据类型元素的多维矩阵.
Torch定义了八种CPU张量类型和八种GPU张量类型:
Data type | dtype | CPU tensor | GPU tensor |
---|---|---|---|
32-bit floating point | torch.float32 or torch.float |
torch.FloatTensor |
torch.cuda.FloatTensor |
64-bit floating point | torch.float64 or torch.double |
torch.DoubleTensor |
torch.cuda.DoubleTensor |
16-bit floating point | torch.float16 or torch.half |
torch.HalfTensor |
torch.cuda.HalfTensor |
8-bit integer (unsigned) | torch.uint8 |
torch.ByteTensor |
torch.cuda.ByteTensor |
8-bit integer (signed) | torch.int8 |
torch.CharTensor |
torch.cuda.CharTensor |
16-bit integer (signed) | torch.int16 or torch.short |
torch.ShortTensor |
torch.cuda.ShortTensor |
32-bit integer (signed) | torch.int32 or torch.int |
torch.IntTensor |
torch.cuda.IntTensor |
64-bit integer (signed) | torch.int64 or torch.long |
torch.LongTensor |
torch.cuda.LongTensor |
torch.Tensor
是默认的tensor类型 (torch.FloatTensor
) 的简称.
Tensor 可以用torch.tensor()
转换Python的 list
或序列??生成:
>>> torch.tensor([[1., -1.], [1., -1.]])
tensor([[ 1.0000, -1.0000],
[ 1.0000, -1.0000]])
>>> torch.tensor(np.array([[1, 2, 3], [4, 5, 6]]))
tensor([[ 1, 2, 3],
[ 4, 5, 6]])
警告
torch.tensor()
总是拷贝 data
. 如果你有一个 Tensor data
并且仅仅想改变它的 requires_grad
属性, 可用 requires_grad_()
or detach()
来避免拷贝. 如果你有一个 numpy 数组并且想避免拷贝, 请使用 torch.as_tensor()
.
原文地址:https://www.cnblogs.com/wizardforcel/p/10358933.html
时间: 2024-10-28 09:43:05