MATLAB实现波士顿房价预测使用BP神经网络

MATLAB实现波士顿房价预测，使用BP神经网络

代码如下(包括下载数据和训练网络)：

%%Download Housing Prices

filename = 'housing.txt';

%下载

urlwrite('http://archive.ics.uci.edu/ml/machine-learning-databases/housing/housing.data',filename);

%指定名字

inputNames = {'CRIM','ZN','INDUS','CHAS','NOX','RM','AGE','DIS','RAD','TAX','PTRATIO','B','LSTAT'};

outputNames = {'MEDV'};

housingAttributes = [inputNames,outputNames];

%%Import Data

%格式规范

formatSpec = '%8f%7f%8f%3f%8f%8f%7f%8f%4f%7f%7f%7f%7f%f%[^\n\r]';

fileID = fopen(filename,'r');

%读取大文件，比起testread方便

dataArray = textscan(fileID, formatSpec, 'Delimiter', '', 'WhiteSpace', '', 'ReturnOnError', false);

fclose(fileID);

%写进表格

housing = table(dataArray{1:end-1}, 'VariableNames', {'VarName1','VarName2','VarName3','VarName4','VarName5','VarName6','VarName7','VarName8','VarName9',...

'VarName10','VarName11','VarName12','VarName13','VarName14'});

%Delete the file and clear temporary variables

clearvars filename formatSpec fileID dataArray ans;

%%delete housing.txt

%%Read into a Table

%重新定义变量名字

housing.Properties.VariableNames = housingAttributes;

%X特征向量 Y房价

X = housing{:,inputNames};

Y = housing{:,outputNames};

%数据处理好了，开始训练

features=X;prices=Y;len = length(prices);

index = randperm(len);%生成1~len 的随机数

%%产生训练集和数据集

%训练集——前70%

p_train = features(index(1:round(len*0.7)),:);%训练样本输入

t_train = prices(index(1:round(len*0.7)),:);%训练样本输出

%测试集——后30%

p_test = features(index(round(len*0.7)+1:end),:);%测试样本输入

t_test = prices(index(round(len*0.7)+1:end),:);%测试样本输出

%%数据归一化

%输入样本归一化

[pn_train,ps1] = mapminmax(p_train');

pn_test = mapminmax('apply',p_test',ps1);

%输出样本归一化

[tn_train,ps2] = mapminmax(t_train');

%tn_test = mapminmax('apply',t_test',ps2);

%%神经网络

%创建和训练

net = feedforwardnet(5,'trainlm');%创建网络

net.trainParam.epochs = 5000;%设置训练次数

net.trainParam.goal=0.0000001;%设置收敛误差

[net,tr]=train(net,pn_train,tn_train);%训练网络

%网络仿真,测试数据

b=sim(net,pn_test);%放入到网络输出数据

%%结果反归一化，预测的价格

predict_prices = mapminmax('reverse',b,ps2);

%%结果分析

t_test = t_test';

err_prices = t_test-predict_prices;%误差

[mean(err_prices) std(err_prices)]%求平均，标准差

figure(1);

plot(t_test);

hold on;

plot(predict_prices,'r');

xlim([1 length(t_test)]);

hold off;

legend({'Actual','Predicted'})

xlabel('Training Data point');

ylabel('Median house price');

结果：

MATLAB实现波士顿房价预测使用BP神经网络相关教程