Problem loading pretrained fuzzy model

I have next problem runing the "

Explain Black-Box Model Using Fuzzy

" example:

>> data = load('dataExplainDNN.mat');

dnnLKA = data.trainedDNN;

Warning: Unable to load instances of class rl.layer.ScalingLayer into a heterogeneous array. The definition of

rl.layer.ScalingLayer could be missing or contain an error. Default objects will be substituted.

Warning: While loading an object of class 'SeriesNetwork':

Error using the predict function in layer nnet.cnn.layer.MissingLayer. The function threw an error and could not be executed.

>> steeringAngle = predict(dnnLKA,zeros(1,6))

Dot indexing is not supported for variables of this type.

Error in SeriesNetwork/predict (line 320)

Y = this.UnderlyingDAGNetwork.predict(X, varargin{:});

The class "rl" is not load correctly, load a structure but is not the original data (SacalingLayer object maybe?), then fail when try to use as a predict model.

I need to add some function of the original model like a add-on?

Answer 1

[Aman]

Hi,

I understand that you are trying to execute the "Explain Black-Box Model Using Fuzzy Support System" example and are facing an issue with the "dnnLKA" model.

This issue occurs when the class definition for that network is not on the MATLAB path when the network is being used.

Please refer to the following documentation to learn about how to solve this issue by including the class definition.

https://www.mathworks.com/support/search.html/answers/1949043-compiled-matlab-function-does-not-recognize-predict-function-from-the-statistics-and-machine-learn.html?fq%5B%5D=asset_type_name:answer&fq%5B%5D=category:stats/gaussian-process-regression&page=1

I hope it helps!

[Rodrigo Salas]

Hi, a temporal solution was try in another computer with more ToolBox and the example run without problem. I checked what LKA is and is a Lane Keeping Assist, this model require another ToolBox that I haven't buy in my current pc. Thanks for the answer.