Deep Learning Toolbox Model for SqueezeNet Network

Pretrained SqueezeNet model for image classification

The SqueezeNet pretrained model for image classification is a part of the Deep Learning Toolbox in R2020a and does not require a separate installation.

• If you are using the R2024a version of the Deep Learning Toolbox, you can type the following in the command line or access the model directly without installation from the Deep Network Designer app:

[net, classes] = imagePretrainedNetwork("squeezenet");

• If you are using a version of the Deep Learning Toolbox between R2020a and R2023b, you can type squeezenet in the command line or access the model directly without installation from the Deep Network Designer App. If you are using R2018a to R2019b, you'll need to download and install this support package.

SqueezeNet is a pretrained model that has been trained on a subset of the ImageNet database. The model is trained on more than a million images, and can classify images into 1000 object categories (e.g. keyboard, mouse, pencil, and many animals).

Opening the squeezenet.mlpkginstall file from your operating system or from within MATLAB will initiate the installation process for the release you have.

This mlpkginstall file is functional for R2018a and beyond. Use squeezenet instead of imagePretrainedNetwork if using a release prior to R2024a.

Usage Example:

% Access the trained model
[net, classes] = imagePretrainedNetwork("squeezenet");

% See details of the architecture
net.Layers

% Read the image to classify
I = imread('peppers.png');

% Adjust size of the image
sz = net.Layers(1).InputSize
I = I(1:sz(1),1:sz(2),1:sz(3));

% Classify the image using SqueezeNet
scores = predict(net, single(I));
label = scores2label(scores, classes)

% Show the image and the classification results
figure
imshow(I)
text(10,20,char(label),'Color','white')