Why do I get a warning when I load a neural network object in a compiled application?

I have stored neural network objects in a MAT file and have created and compiled the following function:

function try_to_open

load my_nn_objs.mat;

Executing try_to_open inside of MATLAB works as expected. However if I compile the function using the following command:

mcc -m try_to_open.m

I get the following warning:

 Warning: Class ':all:' is an unknown object class.  Object '' of this 

          class has been converted to a structure.

Answer 1

[MathWorks Support Team]

To make the compiler aware of the neural network objects, modify the following directive in your function:

 %#function network

For example:

function try_to_open

%#function network

load my_nn_objs.mat;

Answer 2

[Oleksii Sidorov]

I faced the same issue installing Inception-ResNet-v2,

most probably it is caused by the fact that definition of DAGNetwork contains layers which are NOT default in NN Toolbox. These "new" layers are downloaded and installed together with add-on. So, logically, without them network model has a "reference" on layers which are not exist.

I solved this issue by copying whole add-on's directory and running inceptionresnetv2.m file instead of just loading data from inceptionresnetv2.mat. In this way model is imported together with layers.

[Oleksii Sidorov]

You also need to copy that custom layers to your working dir, and add it to matlab path like following:

 exampleFolder = genpath(fullfile(pwd));
 addpath(exampleFolder)