分类预测 | MATLAB实现PSO-DBN粒子群优化深度置信网络多输入分类预测

分类预测 | MATLAB实现PSO-DBN粒子群优化深度置信网络多输入分类预测

效果一览

基本介绍

Matlab实现PSO-DBN粒子群优化深度置信网络多输入分类预测
多特征输入单输出的二分类及多分类模型。程序内注释详细，直接替换数据就可以用。程序语言为matlab，程序可出分类效果图，迭代优化图，混淆矩阵图。
粒子群优化学习率、迭代次数和隐藏层单元数目。

程序设计

• 完整源码和数据下载：MATLAB实现PSO-DBN粒子群优化深度置信网络多输入分类预测

%-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
%%  清空环境变量
clc;
clear;
warning off
close all
%-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
%%  添加路径
addpath("Toolbox\")
%-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
%%  读取数据
res = xlsread('数据集.xlsx');



%%  性能评价
error1 = sum((T_sim1' == T_train)) / M * 100 ;
error2 = sum((T_sim2' == T_test )) / N * 100 ;
%-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
%%  适应度曲线
figure
plot(1: length(curve), curve, 'LineWidth', 1.5);
title('PSO-DBN适应度变化曲线', 'FontSize', 10);
xlabel('迭代次数', 'FontSize', 10);
ylabel('适应度值', 'FontSize', 10);
xlim([1, length(curve)])
grid on

%%  损失函数曲线
figure
plot(1: length(accu), accu, 'r-', 'LineWidth', 1)
xlabel('迭代次数')
ylabel('准确率')
legend('训练集正确率')
title ('训练集正确率曲线')
xlim([1, length(accu)])
grid
    
figure
plot(1 : length(loss), loss, 'b-', 'LineWidth', 1)
xlabel('迭代次数')
ylabel('损失函数')
legend('训练集损失值')
title ('训练集损失函数曲线')
xlim([1, length(loss)])
grid

%%  绘图
figure
plot(1: M, T_train, 'r-*', 1: M, T_sim1, 'b-o', 'LineWidth', 1)
legend('真实值', 'PSO-DBN预测值')
xlabel('预测样本')
ylabel('预测结果')
string = {'训练集预测结果对比'; ['准确率=' num2str(error1) '%']};
title(string)
grid

figure
plot(1: N, T_test, 'r-*', 1: N, T_sim2, 'b-o', 'LineWidth', 1)
legend('真实值', 'PSO-DBN预测值')
xlabel('预测样本')
ylabel('预测结果')
string = {'测试集预测结果对比'; ['准确率=' num2str(error2) '%']};
title(string)
grid
%-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
%%  混淆矩阵
if flag_conusion == 1

    figure
    cm = confusionchart(T_train, T_sim1);
    cm.Title = 'Confusion Matrix for Train Data';
    cm.ColumnSummary = 'column-normalized';
    cm.RowSummary = 'row-normalized';
    
    figure
    cm = confusionchart(T_test, T_sim2);
    cm.Title = 'Confusion Matrix for Test Data';
    cm.ColumnSummary = 'column-normalized';
    cm.RowSummary = 'row-normalized';
end

参考资料

[1] https://download.csdn.net/download/kjm13182345320/87899283?spm=1001.2014.3001.5503
[2] https://download.csdn.net/download/kjm13182345320/87899230?spm=1001.2014.3001.5503