Main Content


Generate Simulink block for shallow neural network simulation



gensim(net,st) creates a Simulink® system containing a block that simulates neural network net with a sampling time of st.

If net has no input or layer delays (net.numInputDelays and net.numLayerDelays are both 0), you can use –1 for st to get a network that samples continuously.

gensim does not support deep learning networks such as convolutional or LSTM networks. For more information on code generation for deep learning, see Code Generation.

For more information on gensim, at the MATLAB command prompt, enter help network/gensim.


collapse all

This example shows how to generate a Simulink block for a feedforward network.

Create a feed-forward network using the data from the simple fit data set and generate the Simulink block.

[x,t] = simplefit_dataset;
net = feedforwardnet(10);
net = train(net,x,t)

This example shows how to generate a Simulink block for a NARX network.

Create a NARX network.

[x,t] = simplenarx_dataset;
net = narxnet(1:2,1:2,20);
[xs,xi,ai,ts] = preparets(net,x,{},t);
net = train(net,xs,ts,xi,ai);
y = net(xs,xi,ai);

Convert the network to closed loop.

net = closeloop(net);

Prepare the data and simulate the network’s closed loop response.

[xs,xi,ai,ts] = preparets(net,x,{},t);
y = net(xs,xi,ai);

Convert the network to a Simulink system with workspace input and output ports.

[sysName,netName] = gensim(net,'InputMode','Workspace',...

Initialize the delay states. Note that this is an important step to obtain the same output as in MATLAB®.


Define the model input X1 in the workspace, simulate the system programmatically.

x1 = nndata2sim(xs,1,1);
out = sim(sysName,'ReturnWorkspaceOutputs','on','StopTime',num2str(x1.time(end)));
ysim = sim2nndata(out.y1);

Input Arguments

collapse all

Input network, specified as a network object. To create a network object, use for example, feedforwardnet or narxnet.

Specify the sample time as a value other than -1. For more information, see Specify Sample Time (Simulink).

Version History

Introduced before R2006a

See Also