project1.m

clear;clc;

% https://www.mathworks.com/help/deeplearning/examples/train-deep-learning-network-to-classify-new-images.html

%% Loading monkey data
imdsValidation = imageDatastore('10-monkey-species/validation', 'IncludeSubfolders',true,'LabelSource','foldernames');
imdsTrain = imageDatastore('10-monkey-species/training', 'IncludeSubfolders',true,'LabelSource','foldernames');

%% Selecting cnn model
net = resnet50;
inputSize = net.Layers(1).InputSize;
% analyzeNetwork(net);

%% Extract the layer graph from the trained network. 
% If the network is a SeriesNetwork object, such as AlexNet, 
% VGG-16, or VGG-19, then convert the list of layers in net.
% Layers to a layer graph.
if isa(net,'SeriesNetwork') 
  lgraph = layerGraph(net.Layers); 
else
  lgraph = layerGraph(net);
end 

[learnableLayer,classLayer] = findLayersToReplace(lgraph);
[learnableLayer,classLayer] 

%% podmiana FullyConnectedLayer
numClasses = numel(categories(imdsTrain.Labels));

if isa(learnableLayer,'nnet.cnn.layer.FullyConnectedLayer')
    newLearnableLayer = fullyConnectedLayer(numClasses, ...
        'Name','new_fc', ...
        'WeightLearnRateFactor',10, ...
        'BiasLearnRateFactor',10);
    
elseif isa(learnableLayer,'nnet.cnn.layer.Convolution2DLayer')
    newLearnableLayer = convolution2dLayer(1,numClasses, ...
        'Name','new_conv', ...
        'WeightLearnRateFactor',10, ...
        'BiasLearnRateFactor',10);
end
lgraph = replaceLayer(lgraph,learnableLayer.Name,newLearnableLayer);

%% podmiana wyjscia
newClassLayer = classificationLayer('Name','new_classoutput');
lgraph = replaceLayer(lgraph,classLayer.Name,newClassLayer);

%analyzeNetwork(lgraph);

%% zamrożenie warstw
layers = lgraph.Layers;
connections = lgraph.Connections;

layers(1:end-3) = freezeWeights(layers(1:end-3));
lgraph = createLgraphUsingConnections(layers,connections);

%% augmented image datastore 
pixelRange = [-30 30];
scaleRange = [0.9 1.1];
imageAugmenter = imageDataAugmenter( ...
    'RandXReflection',true, ...
    'RandXTranslation',pixelRange, ...
    'RandYTranslation',pixelRange, ...
    'RandXScale',scaleRange, ...
    'RandYScale',scaleRange);
augimdsTrain = augmentedImageDatastore(inputSize(1:2),imdsTrain, ...
    'DataAugmentation',imageAugmenter);
augimdsValidation = augmentedImageDatastore(inputSize(1:2),imdsValidation);

%% training options
miniBatchSize = 10;
valFrequency = floor(numel(augimdsTrain.Files)/miniBatchSize);
options = trainingOptions('sgdm', ...
    'MiniBatchSize',miniBatchSize, ...
    'MaxEpochs',6, ...
    'InitialLearnRate',3e-4, ...
    'Shuffle','every-epoch', ...
    'ValidationData',augimdsValidation, ...
    'ValidationFrequency',valFrequency, ...
    'Verbose',false, ...
    'Plots','training-progress');

%% training
trainedNet = trainNetwork(augimdsTrain,lgraph,options);

[YPredValid,scoresValid] = classify(trainedNet, augimdsValidation);
validationLabels = grp2idx(imdsValidation.Labels);
validationResult = grp2idx(YPredValid);

[YPredTrain,scoresTrain] = classify(trainedNet, augimdsTrain);
trainLabels = grp2idx(imdsTrain.Labels);
trainResult = grp2idx(YPredTrain);