MATLAB提供了Deep Learning Toolbox，对LSTM调用，示例

是的，MATLAB提供了Deep Learning Toolbox，以下是一个简单的LSTM网络示例，用于预测时间序列数据：

% 加载数据
load sunspot_dataset
data = double(sunspotdata);

% 将数据分为训练集和测试集
numTimeStepsTrain = floor(0.9*numel(data));
dataTrain = data(1:numTimeStepsTrain+1);
dataTest = data(numTimeStepsTrain+1:end);

% 标准化数据
mu = mean(dataTrain);
sig = std(dataTrain);
dataTrainStandardized = (dataTrain - mu) / sig;

% 创建输入序列和响应序列
XTrain = dataTrainStandardized(1:end-1);
YTrain = dataTrainStandardized(2:end);

% 创建LSTM网络
numFeatures = 1;
numResponses = 1;
numHiddenUnits = 200;
layers = [ ...
    sequenceInputLayer(numFeatures)
    lstmLayer(numHiddenUnits)
    fullyConnectedLayer(numResponses)
    regressionLayer];

% 配置LSTM网络的训练选项
options = trainingOptions('adam', ...
    'MaxEpochs', 100, ...
    'GradientThreshold', 1, ...
    'InitialLearnRate', 0.01, ...
    'LearnRateSchedule','piecewise', ...
    'LearnRateDropFactor', 0.1, ...
    'LearnRateDropPeriod', 50, ...
    'Verbose', 0, ...
    'Plots','training-progress');

% 训练LSTM网络
net = trainNetwork(XTrain',YTrain',layers,options);

% 使用测试集评估训练好的LSTM网络
dataTestStandardized = (dataTest - mu) / sig;
XTest = dataTestStandardized(1:end-1);
YTest = dataTest(2:end);
YPred = predict(net,XTest');
YPred = sig*YPred + mu;

% 可视化结果
figure
plot(dataTrain(1:end-1))
hold on
idx = numTimeStepsTrain+1:(numTimeStepsTrain+numel(YTest));
plot(idx,YTest)
plot(idx,YPred,'.-')
hold off
xlabel("Year")
ylabel("Sunspots")
title("Forecast")
legend(["Observed" "Test" "Forecast"])

在此示例中，我们加载了太阳黑子时间序列数据，将其分为训练集和测试集，并标准化了训练集数据。然后，我们创建了一个简单的LSTM网络，并使用Adam优化器训练了100个epoch，最后使用测试集评估了网络，并将结果可视化。