importTensorFlowLayers

Import a pretrained TensorFlow network in the saved model format as a LayerGraph object. Then, assemble the imported layers into a DAGNetwork object, and use the assembled network to classify an image.

Specify the model folder.

if ~exist('digitsDAGnet','dir')
    unzip('digitsDAGnet.zip')
end
modelFolder = './digitsDAGnet';

Specify the class names.

classNames = {'0','1','2','3','4','5','6','7','8','9'};

Import the layers and weights of a TensorFlow network in the saved model format. By default, importTensorFlowLayers imports the network as a LayerGraph object compatible with a DAGNetwork object. Specify the output layer type for an image classification problem.

lgraph = importTensorFlowLayers(modelFolder,'OutputLayerType','classification')

Importing the saved model...
Translating the model, this may take a few minutes...
Finished translation.

lgraph = 
  LayerGraph with properties:

         Layers: [13×1 nnet.cnn.layer.Layer]
    Connections: [13×2 table]
     InputNames: {'input_1'}
    OutputNames: {'ClassificationLayer_activation_1'}

Display the last layer of the imported network. The output shows that importTensorFlowLayers appends a ClassificationOutputLayer to the end of the network architecture.

lgraph.Layers(end)

ans = 
  ClassificationOutputLayer with properties:

            Name: 'ClassificationLayer_activation_1'
         Classes: 'auto'
    ClassWeights: 'none'
      OutputSize: 'auto'

   Hyperparameters
    LossFunction: 'crossentropyex'

The classification layer does not contain the classes, so you must specify these before assembling the network. If you do not specify the classes, then the software automatically sets the classes to 1, 2, ..., N, where N is the number of classes.

The classification layer has the name 'ClassificationLayer_activation_1'. Set the classes to classNames and then replace the imported classification layer with the new one.

cLayer = lgraph.Layers(end);
cLayer.Classes = classNames;
lgraph = replaceLayer(lgraph,'ClassificationLayer_activation_1',cLayer);

Assemble the layer graph using assembleNetwork to return a DAGNetwork object.

net = assembleNetwork(lgraph)

net = 
  DAGNetwork with properties:

         Layers: [13×1 nnet.cnn.layer.Layer]
    Connections: [13×2 table]
     InputNames: {'input_1'}
    OutputNames: {'ClassificationLayer_activation_1'}

Read the image you want to classify.

digitDatasetPath = fullfile(toolboxdir('nnet'),'nndemos','nndatasets','DigitDataset');
I = imread(fullfile(digitDatasetPath,'5','image4009.png'));

Classify the image using the imported network.

label = classify(net,I);

Display the image and the classification result.

imshow(I)
title(['Classification result ' classNames{label}])

Import a pretrained TensorFlow network in the saved model format as a LayerGraph object compatible with a dlnetwork object. Then, convert the LayerGraph object to a dlnetwork to classify an image.

Specify the model folder.

if ~exist('digitsDAGnet','dir')
    unzip('digitsDAGnet.zip')
end
modelFolder = './digitsDAGnet';

Specify the class names.

classNames = {'0','1','2','3','4','5','6','7','8','9'};

Import the TensorFlow network as layers compatible with a dlnetwork object.

lgraph = importTensorFlowLayers(modelFolder,'TargetNetwork','dlnetwork')

Importing the saved model...
Translating the model, this may take a few minutes...
Finished translation.

lgraph = 
  LayerGraph with properties:

         Layers: [12×1 nnet.cnn.layer.Layer]
    Connections: [12×2 table]
     InputNames: {'input_1'}
    OutputNames: {1×0 cell}

Read the image you want to classify and display the size of the image. The image is a grayscale (one-channel) image with size 28-by-28 pixels.

digitDatasetPath = fullfile(toolboxdir('nnet'),'nndemos','nndatasets','DigitDataset');
I = imread(fullfile(digitDatasetPath,'5','image4009.png'));
size(I)

ans = 1×2

    28    28

Convert the imported layer graph to a dlnetwork object.

dlnet = dlnetwork(lgraph);

Display the input size of the network. In this case, the image size matches the network input size. If they do not match, you must resize the image by using imresize(I, netInputSize(1:2)).

dlnet.Layers(1).InputSize

ans = 1×3

    28    28     1

Convert the image to a dlarray. Format the images with the dimensions 'SSCB' (spatial, spatial, channel, batch). In this example, the batch size is 1 and you can omit it ('SSC').

I_dlarray = dlarray(single(I),'SSCB');

Classify the sample image and find the predicted label.

prob = predict(dlnet,I_dlarray);
[~,label] = max(prob);

Display the image and the classification result.

imshow(I)
title(['Classification result ' classNames{label}])

Import a pretrained TensorFlow network in the saved model format as a LayerGraph object. Then, assemble the imported layers into a DAGNetwork object. The imported network contains layers that are not supported for conversion into built-in MATLAB layers. The software automatically generates custom layers when you import these layers.

This example uses the helper function findCustomLayers. To view the code for this function, see Helper Function.

Specify the model folder.

if ~exist('digitsDAGnetwithnoise','dir')
    unzip('digitsDAGnetwithnoise.zip')
end
modelFolder = './digitsDAGnetwithnoise';

Specify the class names.

classNames = {'0','1','2','3','4','5','6','7','8','9'};

Import the layers and weights of a TensorFlow network in the saved model format. By default, importTensorFlowLayers imports the network as a LayerGraph object compatible with a DAGNetwork object. Specify the output layer type for an image classification problem.

lgraph = importTensorFlowLayers(modelFolder,'OutputLayerType','classification');

Warning: 'importTensorFlowLayers' is not recommended and will be removed in a future release. To import TensorFlow models, use importNetworkFromTensorFlow function.

Importing the saved model...
Translating the model, this may take a few minutes...
Finished translation.

If the imported network contains layers not supported for conversion into built-in MATLAB layers, then importTensorFlowLayers can automatically generate custom layers in place of these layers. importTensorFlowLayers saves each generated custom layer to a separate .m file in the namespace +digitsDAGnetwithnoise in the current folder.

Find the indices of the automatically generated custom layers, using the helper function findCustomLayers, and display the custom layers.

ind = findCustomLayers(lgraph.Layers,'+digitsDAGnetwithnoise');
lgraph.Layers(ind)

ans = 
  3×1 Layer array with layers:

     1   'concatenate_1'      Concatenate     digitsDAGnetwithnoise.kConcatenate1Layer3826
     2   'gaussian_noise_1'   GaussianNoise   digitsDAGnetwithnoise.kGaussianNoise1Layer3766
     3   'gaussian_noise_2'   GaussianNoise   digitsDAGnetwithnoise.kGaussianNoise2Layer3791

The classification layer does not contain the classes, so you must specify these before assembling the network. If you do not specify the classes, then the software automatically sets the classes to 1, 2, ..., N, where N is the number of classes.

The classification layer has the name 'ClassificationLayer_activation_1'. Set the classes to classNames and then replace the imported classification layer with the new one.

cLayer = lgraph.Layers(end);
cLayer.Classes = classNames;
lgraph = replaceLayer(lgraph,'ClassificationLayer_activation_1',cLayer);

Assemble the layer graph using assembleNetwork. The function returns a DAGNetwork object that is ready to use for prediction.

net = assembleNetwork(lgraph)

net = 
  DAGNetwork with properties:

         Layers: [15×1 nnet.cnn.layer.Layer]
    Connections: [15×2 table]
     InputNames: {'input_1'}
    OutputNames: {'ClassificationLayer_activation_1'}

function indices = findCustomLayers(layers,Namespace)

s = what(['.\' Namespace]);

indices = zeros(1,length(s.m));
for i = 1:length(layers)
    for j = 1:length(s.m)
        if strcmpi(class(layers(i)),[Namespace(2:end) '.' s.m{j}(1:end-2)])
            indices(j) = i;
        end
    end
end

end