回歸預(yù)測 | MATLAB實(shí)現(xiàn)GRU門控循環(huán)單元多輸入多輸出

預(yù)測效果

基本介紹

MATLAB實(shí)現(xiàn)GRU門控循環(huán)單元多輸入多輸出，數(shù)據(jù)為多輸入多輸出預(yù)測數(shù)據(jù)，輸入10個特征，輸出3個變量,程序亂碼是由于版本不一致導(dǎo)致，可以用記事本打開復(fù)制到你的文件，運(yùn)行環(huán)境MATLAB2020b及以上。命令窗口輸出MAE和R2，可在下載區(qū)獲取數(shù)據(jù)和程序內(nèi)容。

程序設(shè)計(jì)

• 完整程序和數(shù)據(jù)下載方式(資源處直接下載)：MATLAB實(shí)現(xiàn)GRU門控循環(huán)單元多輸入多輸出

%-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
layers = [ ...sequenceInputLayer(numFeatures)fullyConnectedLayer(numResponses)regressionLayer];
options = trainingOptions('adam', ...'MaxEpochs',250, ...'GradientThreshold',1, ...'InitialLearnRate',0.005, ...'LearnRateSchedule','piecewise', ...'LearnRateDropPeriod',125, ...'LearnRateDropFactor',0.2, ...'ExecutionEnvironment','cpu', ...'Verbose',0, ...'Plots','training-progress');
%-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------   
net = trainNetwork(XTrain,YTrain,layers,options);
dataTestStandardized = (dataTest - mu) / sig;
XTest = dataTestStandardized(1:end-1);
net = predictAndUpdateState(net,XTrain);
[net,YPred] = predictAndUpdateState(net,YTrain(end));
numTimeStepsTest = numel(XTest);
%-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

往期精彩

MATLAB實(shí)現(xiàn)RBF徑向基神經(jīng)網(wǎng)絡(luò)多輸入多輸出預(yù)測
MATLAB實(shí)現(xiàn)BP神經(jīng)網(wǎng)絡(luò)多輸入多輸出預(yù)測
MATLAB實(shí)現(xiàn)DNN神經(jīng)網(wǎng)絡(luò)多輸入多輸出預(yù)測
MATLAB實(shí)現(xiàn)GRNN廣義回歸神經(jīng)網(wǎng)絡(luò)多輸入多輸出預(yù)測

參考資料

[1] https://blog.csdn.net/kjm13182345320/article/details/116377961
[2] https://blog.csdn.net/kjm13182345320/article/details/127931217
[3] https://blog.csdn.net/kjm13182345320/article/details/127894261