neuralnet.cpp

用C#写的一个神经网络结合遗传算法优化的程序

#include "stdafx.h"
#include "NeuralNet.h"
SNeuron::SNeuron(int iNumInputs) : m_iNumInputs(iNumInputs+1)
{
	for(int i=0; i<iNumInputs+1;i++)//附加偏移（bias）
	{//权值初始化为-1--1之间的随机值
		m_vecWeight.push_back(RandomClamped());
	}
}
SNeuronLayer::SNeuronLayer(int iNumNeurons, int iNumInputsPerNeuron):m_iNumNeurons(iNumNeurons)                          
{
	for (int i=0; i<iNumNeurons; ++i)
		m_vecNeurons.push_back(SNeuron(iNumInputsPerNeuron));
}
CNeuralNet::CNeuralNet()
{
	m_iNumInputs=14;
	m_iNumOutputs=3;
	m_iNumHiddenLayers=1;
	m_iNeuronPerHiddenLayer=6;
	CreateNet();
}
void CNeuralNet::CreateNet()
{
	if(m_iNumHiddenLayers>0)
	{
		m_vecLayers.push_back(SNeuronLayer(m_iNeuronPerHiddenLayer,m_iNumInputs));//第一隐层
		for(int i=0;i<m_iNumHiddenLayers-1;i++)
		{
			m_vecLayers.push_back(SNeuronLayer(m_iNeuronPerHiddenLayer,m_iNeuronPerHiddenLayer));//每个隐层的神经元数目都相同
		}
		m_vecLayers.push_back(SNeuronLayer(m_iNumOutputs,m_iNeuronPerHiddenLayer));
	}
	else
	{
		m_vecLayers.push_back(SNeuronLayer(m_iNumOutputs,m_iNumInputs));
	}
}
vector<double> CNeuralNet::Update(vector<double> &inputs)
{
	vector<double> outputs;
	int cWeight=0;
	if(inputs.size()!=m_iNumInputs)
	{
		return outputs;//返回一个空
	}
	for(int i=0; i<m_iNumHiddenLayers+1; i++)//隐层+输出层 对每一层
	{
		if(i>0)
		{
			inputs=outputs;
		}
		outputs.clear();
		cWeight=0;
		//for each neuron sum the (inputs * corresponding weights).Throw the total at our sigmoid function to get the output.
		for(int j=0; j<m_vecLayers[i].m_iNumNeurons; j++)//对每一个神经元
		{
			double netinput=0;
			int NumInputs=m_vecLayers[i].m_vecNeurons[j].m_iNumInputs;
			for(int k=0; k<NumInputs-1; k++)//对每一个输入
			{
				netinput+=m_vecLayers[i].m_vecNeurons[j].m_vecWeight[k]*inputs[cWeight++];//权值*输入

			}
			netinput+=m_vecLayers[i].m_vecNeurons[j].m_vecWeight[NumInputs-1]*(-1);//权值*阈值 输入为-1的权值，称为偏移（bias）
			outputs.push_back(Sigmoid(netinput,1));
			cWeight=0;
		}
	}
	return outputs;
}

vector<double> CNeuralNet::GetWeights() const
{
	vector<double> weights;
	for(int i=0; i<m_iNumHiddenLayers+1; i++)
	{
		for(int j=0; j<m_vecLayers[i].m_iNumNeurons; j++)
		{
			for(int k=0; k<m_vecLayers[i].m_vecNeurons[j].m_iNumInputs; k++)
			{
				weights.push_back(m_vecLayers[i].m_vecNeurons[j].m_vecWeight[k]);
			}
		}
	}
	return weights;
}
void CNeuralNet::PutWeights(std::vector<double> &weights)
{
	int cWeight=0;
	for(int i=0; i<m_iNumHiddenLayers+1; i++)
	{
		for(int j=0; j<m_vecLayers[i].m_iNumNeurons; j++)
		{
			for(int k=0; k<m_vecLayers[i].m_vecNeurons[j].m_iNumInputs; k++)
			{
				m_vecLayers[i].m_vecNeurons[j].m_vecWeight[k]=weights[cWeight++];
			}
		}
	}
	return;
}
int CNeuralNet::GetNumberOfWeights() const
{
	int weights=0;
	for(int i=0; i<m_iNumHiddenLayers+1; i++)
	{
		for(int j=0; j<m_vecLayers[i].m_iNumNeurons; j++)
		{
			for(int k=0; k<m_vecLayers[i].m_vecNeurons[j].m_iNumInputs; k++)
			{
				weights++;
			}
		}
	}
	return weights;
}
double CNeuralNet::Sigmoid(double activation, double response)
{
	return 1/(1+exp(-activation/response));
}
void CNeuralNet::OutputWeight()
{
	ofstream weight("weight.txt");
	int numofweights=0;
	for(int i=0; i<m_iNumHiddenLayers+1; i++)
	{
		for(int n=0; n<m_vecLayers[i].m_iNumNeurons; n++)
		{
			for(int k=0; k<m_vecLayers[i].m_vecNeurons[n].m_iNumInputs; k++)
			{
				weight<<m_vecLayers[i].m_vecNeurons[n].m_vecWeight[k]<<endl;
				numofweights++;
			}
		}
	}
	weight<<"权值数目："<<numofweights;
}