%% LSTM network
%% import data
clc
clear
close all
%读取double格式数据
training_data = xlsread('a.xlsx',1);
training_data(:,8) = [];%去除空列
testing_data = xlsread('a.xlsx',2);
testing_data(:,8) = [];%去除空列
% 把table格式的数据转换为doulbe格式的数据，这样大部分MATLAB版本都可以跑
input_train =training_data(:,1:7);%训练数据输出数据
output_train = training_data(:,8);%训练数据输入数据
input_test = testing_data(:,1:7);%测试数据输出数据
output_test = testing_data(:,8);%测试数据输入数据
[inputn,inputps]=mapminmax(input_train',-1,1);%训练数据的输入数据的归一化
[outputn,outputps]=mapminmax(output_train',-1,1);%训练数据的输出数据的归一化de
inputn_test=mapminmax('apply',input_test',inputps);
%% Define Network Architecture
% Define the network architecture.
numFeatures = size(training_data(:,1:7),2);%输入层维度
numResponses = size(training_data(:,end),2);%输出维度
% 200 hidden units
numHiddenUnits = 60;%第一层维度
% a fully connected layer of size 50 & a dropout layer with dropout probability 0.5
layers = [ ...
sequenceInputLayer(numFeatures)%输入层
lstmLayer(numHiddenUnits,'OutputMode','sequence')%第一层
fullyConnectedLayer(30)%链接层
dropoutLayer(0.5)%遗忘层
fullyConnectedLayer(numResponses)%链接层
regressionLayer];%回归层
% Specify the training options.
% Train for 60 epochs with mini-batches of size 20 using the solver 'adam'
maxEpochs = 60;%最大迭代次数
miniBatchSize = 20;%最小批量
% the learning rate == 0.01
% set the gradient threshold to 1
% set 'Shuffle' to 'never'
options = trainingOptions('adam', ... %解算器
'MaxEpochs',maxEpochs, ... %最大迭代次数
'MiniBatchSize',miniBatchSize, ... %最小批次
'InitialLearnRate',0.01, ... %初始学习率
'GradientThreshold',inf, ... %梯度阈值
'Shuffle','every-epoch', ... %打乱顺序
'Plots','training-progress',... %画图
'Verbose',0); %不输出训练过程
%% Train the Network
net = trainNetwork(inputn,outputn,layers,options);%开始训练
%% Test the Network
y_pred = predict(net,inputn_test,'MiniBatchSize',1)';%测试仿真输出
y_pred=(mapminmax('reverse',y_pred,outputps))';
y_pred0 = predict(net,inputn,'MiniBatchSize',1)';%训练拟合值
y_pred0=(mapminmax('reverse',y_pred0,outputps))';
nx = 1;
npan = (nx-1)*300+1:nx*300;
figure%打开一个图像窗口
plot(y_pred(npan),'k-*')%黑色实线，点的形状为*
hold on%继续画图
plot(output_test(npan),'r-o')%红色实线，点的形状为o
hold off%停止画图
title('测试图')%标题
ylabel('y')%Y轴名称
legend('测试值','实际值')%标签
error1 = y_pred'-output_test;%误差
figure
plot(error1(npan),'k-*')
title('测试误差图')
ylabel('误差')
figure
plot(y_pred0(npan),'k-*')
hold on
plot(output_train(npan),'r-o')
hold off
title('训练图')
ylabel('y')
legend('测试值','实际值')
error0 = y_pred0'-output_train;
figure
plot(error0(npan),'k-*')
title('训练误差图')
ylabel('误差')
%%
%2
nx = 2;
npan = (nx-1)*300+1:nx*300;
figure%打开一个图像窗口
plot(y_pred(npan),'k-*')%黑色实线，点的形状为*
hold on%继续画图
plot(output_test(npan),'r-o')%红色实线，点的形状为o
hold off%停止画图
title('测试图')%标题
ylabel('y')%Y轴名称
legend('测试值','实际值')%标签
error1 = y_pred'-output_test;%误差
figure
plot(error1(npan),'k-*')
title('测试误差图')
ylabel('误差')
figure
plot(y_pred0(npan),'k-*')
hold on
plot(output_train(npan),'r-o')
hold off
title('训练图')
ylabel('y')
legend('测试值','实际值')
error0 = y_pred0'-output_train;
figure
plot(error0(npan),'k-*')
title('训练误差图')
ylabel('误差')
%%
%3
nx = 3;
npan = (nx-1)*300+1:nx*300;
figure%打开一个图像窗口
plot(y_pred(npan),'k-*')%黑色实线，点的形状为*
hold on%继续画图
plot(output_test(npan),'r-o')%红色实线，点的形状为o
hold off%停止画图
title('测试图')%标题
ylabel('y')%Y轴名称
legend('测试值','实际值')%标签
error1 = y_pred'-output_test;%误差
figure
plot(error1(npan),'k-*')
title('测试误差图')
ylabel('误差')
figure
plot(y_pred0(npan),'k-*')
hold on
plot(output_train(npan),'r-o')
hold off
title('训练图')
ylabel('y')
legend('测试值','实际值')
error0 = y_pred0'-output_train;
figure
plot(error0(npan),'k-*')
title('训练误差图')
ylabel('误差')
%%
%4
nx = 4;
npan = (nx-1)*300+1:951;
figure%打开一个图像窗口
plot(y_pred(npan),'k-*')%黑色实线，点的形状为*
hold on%继续画图
plot(output_test(npan),'r-o')%红色实线，点的形状为o
hold off%停止画图
title('测试图')%标题
ylabel('y')%Y轴名称
legend('测试值','实际值')%标签
error1 = y_pred'-output_test;%误差
figure
plot(error1(npan),'k-*')
title('测试误差图')
ylabel('误差')
figure
plot(y_pred0(npan),'k-*')
hold on
plot(output_train(npan),'r-o')
hold off
title('训练图')
ylabel('y')
legend('测试值','实际值')
error0 = y_pred0'-output_train;
figure
plot(error0(npan),'k-*')
title('训练误差图')
ylabel('误差')
[MSE,RMSE,MBE,MAE ] =MSE_RMSE_MBE_MAE(output_test',y_pred)
R2 = R_2(output_test',y_pred)
