训练卷积神经网络的一些合成图像的手写数字，然后在测试集上运行训练好的网络，计算其准确性。

[XTrain, YTrain] = digitTrain4DArrayData;%导入训练数据
layers = [ ...
  imageInputLayer([28 28 1])
  convolution2dLayer(5,20)
  reluLayer
  maxPooling2dLayer(2,'Stride',2)
  fullyConnectedLayer(10)
  softmaxLayer
  classificationLayer];%定义CNN神经网络：输入为图像并给定图像大小；卷积核为5x5,20个；激励函数采用relu;采用2x2步长为2的最大池化
options = trainingOptions('sgdm', 'Plots', 'training-progress');%采用SGDM求解器并画出网络训练过程
net = trainNetwork(XTrain, YTrain, layers, options);
[XTest, YTest] = digitTest4DArrayData;%导入测试数据
YPred = classify(net, XTest);
accuracy = sum(YTest == YPred)/numel(YTest)

accuracy =
    0.9770