Error in activations() call when running 'Image Category Classification Using Deep Learning' Example

5 views (last 30 days)
I've been walking through the following example and I get an error in the first call to the activations() function:
The error says:
"Check for missing argument or incorrect argument data type in call to function 'activations'."
If Iuse the image data set supplied in the link, the error becomes:
"Undefined function 'activations' for input arguments of type 'nnet.cnn.LayerGraph'.
Error in Untitled7 (line 71)
trainingFeatures = activations( net, augmentedTrainingSet, featureLayer, ..."
I haven't changed any of the code, except my images are rgb, which I don't think affects this function call. Which of my inputs could be incorrect? Or am I missing necessary toolbox?
  1 Comment
Dominik Mattioli
Dominik Mattioli on 10 Jun 2021
I believe I've figured it out - the example says that you can use
net = resnet50( 'Weights', 'None' );
if you don't have the resnet pretrained model installed. However, I just installed the toolbox and changed the network instantiation to
net = resnet50();
and now it works. Perhaps the documentation could be edited to note this.

Sign in to comment.

Answers (0)

Categories

Find more on Install Products in Help Center and File Exchange

Community Treasure Hunt

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

Start Hunting!