欢迎您访问程序员文章站本站旨在为大家提供分享程序员计算机编程知识!
您现在的位置是: 首页  >  IT编程

Python使用numpy实现BP神经网络

程序员文章站 2022-04-22 08:51:30
本文完全利用numpy实现一个简单的BP神经网络,由于是做regression而不是classification,因此在这里输出层选取的激励函数就是f(x)=x。BP神经网...

本文完全利用numpy实现一个简单的BP神经网络,由于是做regression而不是classification,因此在这里输出层选取的激励函数就是f(x)=x。BP神经网络的具体原理此处不再介绍。

import numpy as np 
 
class NeuralNetwork(object): 
  def __init__(self, input_nodes, hidden_nodes, output_nodes, learning_rate): 
    # Set number of nodes in input, hidden and output layers.设定输入层、隐藏层和输出层的node数目 
    self.input_nodes = input_nodes 
    self.hidden_nodes = hidden_nodes 
    self.output_nodes = output_nodes 
 
    # Initialize weights,初始化权重和学习速率 
    self.weights_input_to_hidden = np.random.normal(0.0, self.hidden_nodes**-0.5,  
                    ( self.hidden_nodes, self.input_nodes)) 
 
    self.weights_hidden_to_output = np.random.normal(0.0, self.output_nodes**-0.5,  
                    (self.output_nodes, self.hidden_nodes)) 
    self.lr = learning_rate 
     
    # 隐藏层的激励函数为sigmoid函数,Activation function is the sigmoid function 
    self.activation_function = (lambda x: 1/(1 + np.exp(-x))) 
   
  def train(self, inputs_list, targets_list): 
    # Convert inputs list to 2d array 
    inputs = np.array(inputs_list, ndmin=2).T  # 输入向量的shape为 [feature_diemension, 1] 
    targets = np.array(targets_list, ndmin=2).T  
 
    # 向前传播,Forward pass 
    # TODO: Hidden layer 
    hidden_inputs = np.dot(self.weights_input_to_hidden, inputs) # signals into hidden layer 
    hidden_outputs = self.activation_function(hidden_inputs) # signals from hidden layer 
 
     
    # 输出层,输出层的激励函数就是 y = x 
    final_inputs = np.dot(self.weights_hidden_to_output, hidden_outputs) # signals into final output layer 
    final_outputs = final_inputs # signals from final output layer 
     
    ### 反向传播 Backward pass,使用梯度下降对权重进行更新 ### 
     
    # 输出误差 
    # Output layer error is the difference between desired target and actual output. 
    output_errors = (targets_list-final_outputs) 
 
    # 反向传播误差 Backpropagated error 
    # errors propagated to the hidden layer 
    hidden_errors = np.dot(output_errors, self.weights_hidden_to_output)*(hidden_outputs*(1-hidden_outputs)).T 
 
    # 更新权重 Update the weights 
    # 更新隐藏层与输出层之间的权重 update hidden-to-output weights with gradient descent step 
    self.weights_hidden_to_output += output_errors * hidden_outputs.T * self.lr 
    # 更新输入层与隐藏层之间的权重 update input-to-hidden weights with gradient descent step 
    self.weights_input_to_hidden += (inputs * hidden_errors * self.lr).T 
  
  # 进行预测   
  def run(self, inputs_list): 
    # Run a forward pass through the network 
    inputs = np.array(inputs_list, ndmin=2).T 
     
    #### 实现向前传播 Implement the forward pass here #### 
    # 隐藏层 Hidden layer 
    hidden_inputs = np.dot(self.weights_input_to_hidden, inputs) # signals into hidden layer 
    hidden_outputs = self.activation_function(hidden_inputs) # signals from hidden layer 
     
    # 输出层 Output layer 
    final_inputs = np.dot(self.weights_hidden_to_output, hidden_outputs) # signals into final output layer 
    final_outputs = final_inputs # signals from final output layer  
     
    return final_outputs 

以上就是本文的全部内容,希望对大家的学习有所帮助,也希望大家多多支持。