Python使用numpy实现BP神经网络

Python使用使用numpy实现实现BP神经网络神经网络

主要为大家详细介绍了Python使用numpy实现BP神经网络，具有一定的参考价值，感兴趣的小伙伴们可以参考

一下

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):

# 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)))

# 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

# 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我们。