MATLAB: How to get a neural network without hiddenlayer

neural networkno hiddenlayer

hello,

I try to initialize a neural network without a hiddenlayer to estimate the distance from a wall with to circles on it. the neural network should be as simple as possible ( I don´t need to consider of a rotation or angle deviation). The inputneurons are the pixels of the images which i transform in a vector, so that it isn´t a matrix anymore. The network should connect directly between 100 inputneurons and just 1 outputeuron.

I hope you can help me with my issue.

Thanks for your help!

Best Answer

If regression, use:

     net = fitnet([]);     % For regression
     view(net)

If classification, use:

   net = patternnet([]); % For classification
   view(net)

Related Solutions

MATLAB: How to train a feedfordward neural network with error weights

 Use the command window
 help mse
 doc mse

Hope this helps.

Thank you for formally accepting my answer

Greg

MATLAB: How to create a neural network with 1 layer only (no hidden layers)

 net = fitnet([]);     % For regression
 net = patternnet([]); % For classification