学习记录 --- Pytorch 张量创建

个人学习总结，持续更新中……

参考文献

[1] pytorch自动求导Autograd系列教程（一）

张量创建

torch.tensor(array_like)方法

torch.tensor()仅仅是python函数，函数原型是：

tensor(data, dtype=None, device=None, requires_grad=False) -> Tensor
参数:
data： (array_like): tensor的初始值. 可以是列表，元组，numpy数组，标量等;
dtype： tensor元素的数据类型
device： 指定CPU或者是GPU设备，默认是None
requires_grad：是否可以求导，即求梯度，默认是False，即不可导的

torch.tensor会从data中的数据部分做拷贝（而不是直接引用），根据原始数据类型生成相应的torch.LongTensor、torch.FloatTensor和torch.DoubleTensor。

import torch

'''
def tensor(data: Any,
           dtype: Optional[_dtype] = None,
           device: Union[_device, str, None] = None,
           requires_grad: _bool = False) -> Tensor
'''
import torch

a = torch.tensor(data=[[1., 2.], [3., 4.]], dtype=torch.float, device='cpu', requires_grad=True)
print(a)
'''
tensor([[1., 2.],
        [3., 4.]], requires_grad=True)
'''
print(a.dtype)
# torch.float32
print(a.type())
# torch.FloatTensor
print(a.device)
# cpu
print(a.requires_grad)
# True

torch.Tensor()

torch.Tensor(n, m)方法

torch.Tensor()是python类，更明确地说，是默认张量类型torch.FloatTensor()的别名，torch.Tensor([1,2])会调用Tensor类的构造函数__init__，生成单精度浮点类型的张量。
其他类似的有：torch.DoubleTensor(n, m)、torch.LongTensor(n, m)等。

最基础的Tensor()函数创建方法，参数为Tensor的每一维大小。

import torch

# 最基础的Tensor()函数创建方法，参数为Tensor的每一维大小
a = torch.Tensor(2, 2)
print(a)
'''
tensor([[0., 0.],
        [0., 0.]])
'''
print(a.dtype)
# torch.float32
print(a.type())
# torch.FloatTensor
print(a.device)
# cpu
print(a.requires_grad)
# False

torch.Tensor(array_like)

import torch
# 使用Python的list序列进行创建
a = torch.Tensor([[1, 2], [3, 4]],device='cpu')
print(a)
'''
tensor([[1., 2.],
        [3., 4.]])
'''
print(a.dtype)
# torch.float32
print(a.type())
# torch.FloatTensor
print(a.device)
# cpu
print(a.requires_grad)
# False

张量类型

Torch defines 10 tensor types with CPU and GPU variants which are as follows:

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 1	torch.float16 or torch.half	torch.HalfTensor	torch.cuda.HalfTensor
16-bit floating point 2	torch.bfloat16	torch.BFloat16Tensor	torch.cuda.BFloat16Tensor
32-bit complex	torch.complex32
64-bit complex	torch.complex64
128-bit complex	torch.complex128 or torch.cdouble
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
Boolean	torch.bool	torch.BoolTensor	torch.cuda.BoolTensor
quantized 8-bit integer (unsigned)	torch.quint8	torch.ByteTensor
quantized 8-bit integer (signed)	torch.qint8	torch.CharTensor
quantized 32-bit integer (signed)	torch.qfint32	torch.IntTensor
quantized 4-bit integer (unsigned) 3	torch.quint4x2	torch.ByteTensor