反向传播算法

神经网络

**反向传播第一层代码

import numpy as np
class FullyConnect:
    def _init_(self,l_x,l_y):
        self.weights = np.random.randn(l_y,l_x)
        self.bias =np.random.randn(l)
        
    def forward(self,x):
        self.x=x
        self.y=np.dot(self.weights.x)+self.bias
        return self.y
    def backward(self,x):
        self.dw = d*self.x
        self.db =d
        self.dx = d*self.weights
        return self.dw,self.db

反向传播算法第二层：

	class Simoid:
    def _init_(self):
        pass
    def sigmoid(self,x):
        return 1/(1+np.exp(-x))
    def forward (self,x):
        self.x=x
        self.y=self.sigmoid(x)
        return self.y
    def backward(self):
        sig=self.sigmoid(sig.x)
        self.dx = sig*(1-sig)
        return self.dx

其实，反向传播算法就是梯度下降算法