i want to create a cascadeforwardnet Neural Networks in MATLAB? how can i create?

%traincgf Fletcher-Powell Conjugate Gradient %traincgp Polak-Ribiére Conjugate Gradient %trainoss One Step Secant %traingdx Variable Learning Rate Gradient Descent %traingdm Gradient Descent with Momentum %traingd Gradient Descent

Answer 1

[Abhipsa]

Hello @sehrish shah,

You can do this by using “cascadeforwadnet” function available in the “Deep Learning Toolbox” in MATLAB R2024b.

You can follow the below steps to create a “casacadeforwardnet”:

1. Prepare the raw data: The input data should be in a matrix form where columns represent the feature and rows represent observations. The target data should also be in a matrix format where each row corresponds to output for the respective input.
2. Create the network: Use the “cascadeforwardnet” function to create a network of desired number of neurons in the hidden layer.
3. Configure the network: You can use the “configure” function to configure the network This step is optional, as unconfigured networks are automatically configured when the “train” function is called for the first time.
4. Train the network: Use “train” function to train the network.
5. Simulate the network: Use “sim” function or call the network object with the inputs to simulate the network.

% Example input and target raw data 
inputs = [0 1 2 3 4 5; 5 4 3 2 1 0];  
targets = [0 1 2 3 4 5];  

% Create a Cascade Forward Network 
hiddenLayerSize = 10; % Number of neurons in the hidden layer, you can change it as per your requirements 
net = cascadeforwardnet(hiddenLayerSize); 

% Configure the network 
net = configure(net, inputs, targets); 

% Train the network 
[net, tr] = train(net, inputs, targets); 

% Test the network 
outputs = net(inputs); 
errors = gsubtract(targets, outputs); 
performance = perform(net, targets, outputs); 

% View the network (optional) 
view(net); 

% Simulate the network with new raw data (example) 
newInputs = [1.5; 3.5]; 
newOutputs = net(newInputs); 
disp(newOutputs); 

For more details, you can refer to the following MATLAB documentation links:

• cascadeforwardnet: https://www.mathworks.com/help/releases/R2024b/deeplearning/ref/cascadeforwardnet.html
• configure: https://www.mathworks.com/help/releases/R2024b/deeplearning/ref/network.configure.html
• train: https://www.mathworks.com/help/releases/R2024b/deeplearning/ref/network.train.html
• sim: https://www.mathworks.com/help/releases/R2024b/deeplearning/ref/sim.html
• view: https://www.mathworks.com/help/releases/R2024b/deeplearning/ref/view.html
• perform: https://www.mathworks.com/help/releases/R2024b/deeplearning/ref/network.perform.html