1 torch.rand: Construct a uniformly distributed tensor
torch.rand is a function used to generate a uniform random distribution tensor from a uniform distribution on the interval [0,1) Randomly extract a random number to generate a tensor, and its calling method is as follows:
torch.rand(sizes, out=None) ➡️ Tensorparameter:
-
sizes: used to define the shape of the output tensor
Sample code:
import torch
# 生成一个每个元素服从0-1均匀分布的4行3列随机张量
random_tensor = torch.rand(4, 3)
print('tensor:', random_tensor)
print('type:', random_tensor.type())
print('shape:', random_tensor.shape)Running the code shows:
tensor: tensor([[0.4349, 0.8567, 0.7321],
[0.4057, 0.0222, 0.3444],
[0.9679, 0.0980, 0.8152],
[0.1998, 0.7888, 0.5478]])
type: torch.FloatTensor
shape: torch.Size([4, 3])2 torch.randn: Construct standard normal distribution tensor
torch.randn() is a function used to generate normal random distribution tensor. A random number is randomly selected from the standard normal distribution to generate a tensor. Amount, its calling method is as follows:
torch.randn(sizes, out=None) ➡️ Tensorparameter:
-
sizes: used to define the shape of the output tensor
Sample code:
import torch
# 生成一个每个元素均为标准正态分布的4行3列随机张量
random_tensor = torch.randn(4, 3)
print('tensor:', random_tensor)
print('type:', random_tensor.type())
print('shape:', random_tensor.shape)Running the code shows:
tensor: tensor([[ 0.7776, 0.6305, 0.1961],
[ 0.1831, -0.4187, 0.1245],
[ 0.3092, -1.0463, -0.6656],
[-1.0098, 1.3861, -0.2600]])
type: torch.FloatTensor
shape: torch.Size([4, 3])3 torch.randn_like: Construct a normal distribution tensor with the same shape as the input
torch.randn_like() is used to generate a tensor of the same size as the input tensor, filled with random values from a normal distribution with mean 0 and variance 1. The calling method is as follows:
torch.randn_like(input_tensor, dtype=None, layout=None, device=None, requires_grad=False) ➡️ Tensorparameter:
-
input_tensor (required) – The input tensor whose size will be used to generate the output tensor.
-
dtype (optional) - The required data type for the output tensor. Defaults to None, which means the data type of the input tensor will be used.
-
layout (optional) – The desired memory layout for the output tensor. Defaults to None, which means the memory layout of the input tensor will be used.
-
device (optional) – The device required for the output tensor. Defaults to None, which means the device of the input tensor will be used.
-
requires_grad (optional) - Whether the output tensor should have its gradient calculated during backpropagation. Default is False.
Sample code:
import torch
# 生成一个每个元素均为标准正态分布的4行3列随机张量
tensor_x = torch.randn(4, 3)
tensor_y = torch.randn_like(tensor_x)
print('tensor_x:', tensor_x)
print('type:', tensor_x.type())
print('shape:', tensor_x.shape)
print('tensor_y:', tensor_y)
print('type:', tensor_y.type())
print('shape:', tensor_y.shape)Running the code shows:
tensor_x: tensor([[ 5.5292e-01, 6.5111e-01, -6.0329e-04],
[ 1.0402e+00, -7.4630e-01, 7.5701e-01],
[ 8.8160e-02, -1.2581e+00, -1.8089e-01],
[-4.2769e-01, -8.5043e-01, -5.8388e-01]])
type: torch.FloatTensor
shape: torch.Size([4, 3])
tensor_y: tensor([[ 0.2308, 0.3297, -0.6633],
[ 1.7389, 0.6372, -1.1069],
[-0.2415, -0.8585, 0.3343],
[-1.2581, -0.5001, 0.0317]])
type: torch.FloatTensor
shape: torch.Size([4, 3])4 torch.randint: Construct interval distribution tensor
torch.randint() is a function used to generate arbitrary interval distribution tensor. A random number is randomly selected from the standard normal distribution to generate a tensor. , its calling method is as follows:
torch.randint(low=0, high, sizes, out=None) ➡️ Tensorparameter:
-
low~high: the range of random numbers -
sizes: used to define the shape of the output tensor
Sample code:
import torch
# 生成一个每个元素均为[1-10]均匀分布的4行3列随机张量
tensor_int = torch.randint(1, 10, (4, 3))
print('tensor_int:', tensor_int)
print('type:', tensor_int.type())
print('shape:', tensor_int.shape)
Running the code shows:
tensor_int: tensor([[1, 7, 1],
[3, 8, 7],
[5, 2, 1],
[5, 3, 6]])
type: torch.LongTensor
shape: torch.Size([4, 3])5 torch.randperm: Randomly sort tensors according to the generated random sequence number
torch.randint() is a function used to randomly sort tensor numbers according to the generated random sequence. Its calling format is as follows Show:
torch.randperm(n, out=None, dtype=torch.int64) ➡️ LongTensorparameter:
-
n: an integer, which can be understood as the dimension of a tensor in a certain direction -
dtype: Returned data type (torch.int64)
Sample code:
import torch
# 生成一个0~3的随机整数排序
idx = torch.randperm(4)
# 生成一个4行3列的张量
tensor_4 = torch.Tensor(4, 3)
# 为了方便对比，首先输出tensor_4的结果
print("原始张量\n", tensor_4)
# 下面输出随机生成的行序号
print("\n生成的随机序号\n", idx)
# 下面的指令实现了在行的方向上，对tensor_4进行随机排序，并输出结果
print("\n随机排序后的张量\n", tensor_4[idx])Running the code shows:
原始张量
tensor([[0., 0., 0.],
[0., 0., 0.],
[0., 0., 0.],
[0., 0., 0.]])
生成的随机序号
tensor([3, 0, 2, 1])
随机排序后的张量
tensor([[0., 0., 0.],
[0., 0., 0.],
[0., 0., 0.],
[0., 0., 0.]])