Main Content

squeezenet

(Not recommended) SqueezeNet convolutional neural network

  • SqueezeNet network architecture

squeezenet is not recommended. Use the imagePretrainedNetwork function instead. For more information, see Version History.

Description

SqueezeNet is a convolutional neural network that is 18 layers deep. You can load a pretrained version of the network trained on more than a million images from the ImageNet database [1]. The pretrained network can classify images into 1000 object categories, such as keyboard, mouse, pencil, and many animals. As a result, the network has learned rich feature representations for a wide range of images. This function returns a SqueezeNet v1.1 network, which has similar accuracy to SqueezeNet v1.0 but requires fewer floating-point operations per prediction [3]. The network has an image input size of 227-by-227. For more pretrained networks in MATLAB®, see Pretrained Deep Neural Networks.

net = squeezenet returns a SqueezeNet network trained on the ImageNet data set.

example

net = squeezenet('Weights','imagenet') returns a SqueezeNet network trained on the ImageNet data set. This syntax is equivalent to net = squeezenet.

lgraph = squeezenet('Weights','none') returns the untrained SqueezeNet network architecture.

Examples

collapse all

Load a pretrained SqueezeNet network.

net = squeezenet
net = 

  DAGNetwork with properties:

         Layers: [68×1 nnet.cnn.layer.Layer]
    Connections: [75×2 table]

This function returns a DAGNetwork object.

SqueezeNet is included within Deep Learning Toolbox™. To load other networks, use functions such as googlenet to get links to download pretrained networks from the Add-On Explorer.

Output Arguments

collapse all

Pretrained SqueezeNet convolutional neural network, returned as a DAGNetwork object.

Untrained SqueezeNet convolutional neural network architecture, returned as a LayerGraph object.

References

[1] ImageNet. http://www.image-net.org.

[2] Iandola, Forrest N., Song Han, Matthew W. Moskewicz, Khalid Ashraf, William J. Dally, and Kurt Keutzer. “SqueezeNet: AlexNet-Level Accuracy with 50x Fewer Parameters and <0.5MB Model Size.” Preprint, submitted November 4, 2016. https://arxiv.org/abs/1602.07360.

[3] Iandola, Forrest N. "SqueezeNet." https://github.com/forresti/SqueezeNet.

Extended Capabilities

expand all

Version History

Introduced in R2018a

collapse all