LSTM Error in Index in array bounds

2 次查看（过去 30 天）

显示更早的评论

Yossi 2019-2-24

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/446705-lstm-error-in-index-in-array-bounds

评论： Jan 2019-2-28

aapl.mat

Hello,

I'm trying to run LSTM , but it seems that the code is exceed the boundtry (.mat file attached).

close all; clear all
load aapl.mat
date=(AAPL{:,1});
Low=(AAPL{:,4})';
closeP=(AAPL{:,6})';
data={Low closeP}';
%data=[closeP]';
figure
plot(date,closeP)
xlabel("Month")
ylabel("Value")
title("S&P500")
numTimeStepsTrain = floor(0.80*length(data{1}));
for i=1:numel(data)
    dataTrain{i} = data{i}(1:numTimeStepsTrain+1);
    dataTest{i} = data{i}(numTimeStepsTrain+1:end);
end
for i=1:numel(dataTrain)
    mu_train{i} = mean(dataTrain{i},2);
    sig_train{i} = std(dataTrain{i});
    dataTrainStandardized{i}=(dataTrain{i}-mu_train{i})./sig_train{i};
end
for i=1:numel(dataTrainStandardized)
    XTrain{i} = dataTrainStandardized{i}(:,1:end-1);
    YTrain{i} = dataTrainStandardized{2}(:,2:end);
end
numFeatures = 1;
numResponses = 1;
numHiddenUnits = 200;
layers = [ ...
    sequenceInputLayer(numFeatures)
    lstmLayer(numHiddenUnits)
    fullyConnectedLayer(numResponses)
    regressionLayer];
options = trainingOptions('adam', ...
    'MaxEpochs',100, ...
    'GradientThreshold',1, ...
    'InitialLearnRate',0.005, ...
    'LearnRateSchedule','piecewise', ...
    'LearnRateDropPeriod',125, ...
    'LearnRateDropFactor',0.2, ...
    'Verbose',0, ...
    'Plots','training-progress');
net = trainNetwork(XTrain,YTrain,layers,options);
for i=1:numel(dataTest)
    %mu_test{i} = mean(dataTest{i},2);
    %sig_test{i} = std(dataTest{i});
    dataTestStandardized{i}=(dataTest{i}-mu_train{i})./sig_train{i};
end
for i=1:numel(dataTestStandardized)
    XTest{i} = dataTestStandardized{i}(:,1:end-1);
    YTest{i}=dataTestStandardized{i}(:,2:end);
    YTestReg{i}=dataTestStandardized{i}(:,end);
end
%XTest = dataTestStandardized(:,1:end-1);
net = predictAndUpdateState(net,XTrain);
[net,YPred] = predictAndUpdateState(net,YTestReg);
numTimeStepsTest = numel(XTest);
%for j=1:numel(dataTestStandardized)
    for i = 2:251
        [net,YPred(:,1)] = predictAndUpdateState(net,YPred(:,i-1),'ExecutionEnvironment','cpu');
    end