How to use of 'prediction block' in Simulink

% One-dimension data was generated and the tranined net was save to 'TrainedNetwork_for_Simulation.mat'

% Generate data

t = 0:0.1:50;

y = sin(t) + 0.1 * randn(size(t));

% Train the network

net= feedforwardnet;

net = train(net,t,y);

% Save the trained network

save('TrainedNetwork_for_simulink', 'net');

% I'm trying to use a trained network in a Simulink model(TestPredict.slx), but I'm getting an error. The error message indicates that

% Error in 'TestPredict/Predict': Failed to evaluate mask initialization commands.

% Caused by:

% MAT file TrainedNetwork_for_simulink.mat does not contain a valid SeriesNetwork, DAGNetwork, dlnetwork, yolov2ObjectDetector or an %% % ssdObjectDetector object.

[AJ Ibraheem]

The 'Predict' block does not support the network type that you've trained.

Try training your network using trainnet instead, see https://uk.mathworks.com/help/deeplearning/ref/trainnet.html