【LSTM回归预测】基于长短期记忆网络BiLSTM+Adboost实现数据回归预测附matlab代码.rar

%% 清空环境变量 warning off % 关闭报警信息 close all % 关闭开启的图窗 clear % 清空变量 clc % 清空命令行 %% 导入数据 res = xlsread('数据集.xlsx'); %% 划分训练集和测试集 temp = randperm(103); P_train = res(temp(1: 80), 1: 7)'; T_train = res(temp(1: 80), 8)'; M = size(P_train, 2); P_test = res(temp(81: end), 1: 7)'; T_test = res(temp(81: end), 8)'; N = size(P_test, 2); %% 数据归一化 [P_train, ps_input] = mapminmax(P_train, 0, 1); P_test = mapminmax('apply', P_test, ps_input); [t_train, ps_output] = mapminmax(T_train, 0, 1); t_test = mapminmax('apply', T_test, ps_output); %% 数据平铺 P_train = double(reshape(P_train, 7, 1, 1, M)); P_test = double(reshape(P_test , 7, 1, 1, N)); t_train = t_train'; t_test = t_test' ; %% 数据格式转换 for i = 1 : M p_train{i, 1} = P_train(:, :, 1, i); end for i = 1 : N p_test{i, 1} = P_test( :, :, 1, i); end %% 创建模型 layers = [ sequenceInputLayer(7) % 建立输入层 bilstmLayer(4, 'OutputMode', 'last') % LSTM层 reluLayer % Relu激活层 fullyConnectedLayer(1) % 全连接层 regressionLayer]; % 回归层 %% 参数设置 options = trainingOptions('adam', ... % Adam 梯度下降算法 'MaxEpochs', 1500, ... % 最大迭代次数 'InitialLearnRate', 0.01, ... % 初始学习率为 0.01 'LearnRateSchedule', 'piecewise', ... % 学习率下降 'LearnRateDropFactor', 0.1, ... % 学习率下降因子 'LearnRateDropPeriod', 1200, ... % 经过 1200 次训练后 学习率为 0.01 * 0.1 'Shuffle', 'every-epoch', ... % 每次训练打乱数据集 'Plots', 'training-progress', ... % 画出曲线 'Verbose', false); %% 训练模型 % 权重初始化 D = ones(1, M) / M; %% 参数设置 K = 10; % 弱回归器个数 H = 6; % 隐藏层节点个数 %% 弱回归器回归 for i = 1 : K %% 训练网络 net = trainNetwork(p_train, t_train, layers, options); %% 仿真预测 t_sim1{i} = predict(net, p_train); t_sim2{i} = predict(net, p_test ); %% 数据格式转换 ztrain = t_train; t_sim1{i} = double(t_sim1{i}); t_sim2{i} = double(t_sim2{i}); %% 数据反归一化 T_sim11 = mapminmax('reverse', t_sim1{i}', ps_output); T_sim22 = mapminmax('reverse', t_sim2{i}', ps_output); %% 均方根误差 error1(i,:) = T_sim11 - T_train; %% 调整D值 weight(i) = 0; for j = 1 : M if abs(error1(i, j)) > 0.1 weight(i) = weight(i) + D(i, j); D(i + 1, j) = D(i, j) * 1.1; else D(i + 1, j) = D(i, j); end end %% 弱分类器i权重 weight(i) = 0.5 / exp(abs(weight(i))); %% D值归一化 D(i + 1, :) = D(i + 1, :) / sum(D(i + 1, :)); end %% 强预测器预测 weight = weight / sum(weight); %% 强分类器分类结果 T_sim1 = zeros(M,1); T_sim2 = zeros(N,1); for i = 1 : K output1 = (weight(i) * t_sim1{i}); output2 = (weight(i) * t_sim2{i}); T_sim1 = output1 + T_sim1; T_sim2 = output2 + T_sim2; end %% 数据反归一化 T_sim1 = mapminmax('reverse', T_sim1', ps_output); T_sim2 = mapminmax('reverse', T_sim2', ps_output); %% 均方根误差 error1 = sqrt(sum((T_sim1 - T_train).^2) ./ M); error2 = sqrt(sum((T_sim2 - T_test ).^2) ./ N); %% 查看网络结构 analyzeNetwork(net) %% 绘图 figure plot(1: M, T_train, 'r-*', 1: M, T_sim1, 'b-o', 'LineWidth', 1) legend('真实值', '预测值') xlabel('预测样本') ylabel('预测结果') string = {'训练集预测结果对比'; ['RMSE=' num2str(error1)]}; title(string) xlim([1, M]) grid figure plot(1: N, T_test, 'r-*', 1: N, T_sim2, 'b-o', 'LineWidth', 1) legend('真实值', '预测值') xlabel('预测样本') ylabel('预测结果') string = {'测试集预测结果对比'; ['RMSE=' num2str(error2)]}; title(string) xlim([1, N]) grid %% 相关指标计算 % R2 R1 = 1 - norm(T_train - T_sim1')^2 / norm(T_train - mean(T_train))^2; R2 = 1 - norm(T_test - T_sim2')^2 / norm(T_test - mean(T_test ))^2; disp(['训练集数据的R2为：', num2str(R1)]) disp(['测试集数据的R2为：', num2str(R2)]) % MAE mae1 = sum(abs(T_sim1' - T_train)) ./ M ; mae2 = sum(abs(T_sim2' - T_test )) ./ N ; disp(['训练集数据的MAE为：', num2str(mae1)]) disp(['测试集数据的MAE为：', num2str(mae2)]) % MBE mbe1 = sum(T_sim1' - T_train) ./ M ; mbe2 = sum(T_sim2' - T_test ) ./ N ; disp(['训练集数据的MBE为：', num2str(mbe1)]) disp(['测试集数据的MBE为：', num2str(mbe2)]) %% 绘制散点图 sz = 25; c = 'b'; figure scatter(T_train, T_sim1, sz, c) hold on plot(xlim, ylim, '--k') xlabel('训练集真实值'); ylabel('训练集预测值'); xlim([min(T_train) max(T_train)]) ylim([min(T_sim1) max(T_sim1)]) title('训练集预测值 vs. 训练集真实值') figure scatter(T_test, T_sim2, sz, c) hold on plot(xlim, ylim, '--k') xlabel('测试集真实值'); ylabel('测试集预测值'); xlim([min(T_test) max(T_test)]) ylim([min(T_sim2) max(T_sim2)]) title('测试集预测值 vs. 测试集真实值')