python计算两张图像的L1和L2损失

版权声明：本文为博主原创文章，未经博主允许不得转载。 https://blog.csdn.net/yql_617540298/article/details/87904860

理解

损失函数中正则化项L1和L2的理解：

https://blog.csdn.net/fjssharpsword/article/details/78842374

过拟合的解释： 
https://hit-scir.gitbooks.io/neural-networks-and-deep-learning-zh_cn/content/chap3/c3s5ss2.html

正则化的解释： 
https://hit-scir.gitbooks.io/neural-networks-and-deep-learning-zh_cn/content/chap3/c3s5ss1.html

正则化的解释： 
http://blog.csdn.net/u012162613/article/details/44261657

正则化的数学解释（一些图来源于这里）： 
http://blog.csdn.net/zouxy09/article/details/24971995

源码

import numpy as np
from PIL import Image

def L1(yhat, y):
    loss = np.mean(np.abs(y - yhat))
    return loss

def L2(yhat, y):
    loss =np.sum(np.power((y - yhat), 2))
    return loss
#调用
for i in range(1,11):
    imgA = Image.open("C:/Users/Administrator/Desktop/A/"+str(i).zfill(4)+".jpg")
    yhat = np.array(imgA)
    
    imgB = Image.open("C:/Users/Administrator/Desktop/B/"+str(i).zfill(4)+".jpg")
    y = np.array(imgB)
    
    print("L1 = " ,(L1(yhat,y)))
    print("L2 = " ,(L2(yhat,y)))