File Exchange

image thumbnail

Deep Learning Toolbox Model for ResNet-101 Network

Pretrained Resnet-101 network model for image classification


Updated 10 Mar 2021

ResNet-101 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, has 347 layers in total, corresponding to a 101 layer residual network, and can classify images into 1000 object categories (e.g. keyboard, mouse, pencil, and many animals).
Opening the resnet101.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 R2017b and beyond.
Usage Example:
% Access the trained model
net = resnet101();
% See details of the architecture
% 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 Resnet-101
label = classify(net, I)
% Show the image and the classification results

Comments and Ratings (9)

sinan salim

cant download it ,,error



Thank you for your share.
I want to know, if I use this model to traing and test my data, can I plot the training progress?
What should I do?


Tunai Marques

Thank you for this implementation.

For all the users, note that if you are adpating this model for your classification task following the MATLAB tutorials, the classification layer is called 'ClassificationLayer_predictions', instead of Resnet50's 'ClassificationLayer_fc1000'. To change the last two layers (ignoring the softmax layer in-between that will shape itself accordangly):

net = resnet101;
numClasses = numel(categories(imdsTrain.Labels))
lgraph = layerGraph(net);
newFCLayer = fullyConnectedLayer(numClasses,'Name','new_fc','WeightLearnRateFactor',10,'BiasLearnRateFactor',10);
lgraph = replaceLayer(lgraph,'fc1000',newFCLayer);
newClassLayer = classificationLayer('Name','new_classoutput');
lgraph = replaceLayer(lgraph,'ClassificationLayer_predictions',newClassLayer);

Again, thank you for this add-on.

adel adel

vikas bhangdiya

can u help how to install resnet101.mlpkginstall file.
I tried Environment -> Add-ons -> Get Hardware Support Package.
but unable to install the file .
waiting for your reply..


nice! I hope in the future also support other types of networks, such as target detection network

adel adel

MATLAB Release Compatibility
Created with R2017b
Compatible with R2017b to R2021a
Platform Compatibility
Windows macOS Linux

Inspired: Pre-trained 3D ResNet-101

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!