深度学习：MATLAB 入门系列（二）

本系列来自 [ Mathworks ]。

使用MATLAB深度学习：Transfer Learning 10行MATLAB代码

%% Load a pre-trained, deep, convolutional network
alex = alexnet;
layers = alex.Layers

%% Modify the network to use five categories
layers(23) = fullyConnectedLayer(5); 
layers(25) = classificationLayer

%% Set up our training data
allImages = imageDatastore('myImages', 'IncludeSubfolders', true, 'LabelSource', 'foldernames');
[trainingImages, testImages] = splitEachLabel(allImages, 0.8, 'randomize');

%% Re-train the Network
opts = trainingOptions('sgdm', 'InitialLearnRate', 0.001, 'MaxEpochs', 20, 'MiniBatchSize', 64);
myNet = trainNetwork(trainingImages, layers, opts);

%% Measure network accuracy
predictedLabels = classify(myNet, testImages); 
accuracy = mean(predictedLabels == testImages.Labels)

使用MATLAB ® Transfer learning，看看它是如何让深度学习应用于您的实际问题。

该示例使用迁移学习（transfer learning）来重新训练AlexNet，一种预先训练的深卷积神经网络（CNN或ConvNet）来识别零食，如热狗，蛋糕和苹果派。