3-17Pytorch与线性代数运算

二范数：欧氏距离
核范数：低秩问题的求解
计算机技术中有一种特点，图像有一种区域性，反应到数据上就是有相关性
范数作用：定义loss,参数约束

import torch
a = torch.rand(1,1)
b = torch.rand(1,1)

print(a,b)

print(torch.dist(a,b,p = 0))
print(torch.dist(a,b,p = 1))
print(torch.dist(a,b,p = 2))
print(torch.dist(a,b,p = 3))
print('-------------------------------------------------')
print(torch.norm(a))
print(torch.norm(a,p = 0))
print(torch.norm(a,p = 1))
print(torch.norm(a,p = 2))
print(torch.norm(a,p = 3))
#计算核范数
print('核范数：',torch.norm(a,p = 'fro'))

tensor([[0.9197]]) tensor([[0.8488]])
tensor(1.)
tensor(0.0708)
tensor(0.0708)
tensor(0.0708)
-------------------------------------------------
tensor(0.9197)
tensor(1.)
tensor(0.9197)
tensor(0.9197)
tensor(0.9197)
核范数： tensor(0.9197)