MATLAB: How to apply Back propagation for 3 class problem

multi-classnewffr2012a

Hello. I want to solve a classification problem with 3 classes using multi layer neural network with back propagation algorithm. I'm using matlab 2012a. I'm facing trouble with newff function. I want to build a network with one hidden layer and there will be 3 neurons in the output layer, one for each class. Please advise me with example. Thanks.

Best Answer

Newff is obsolete. Switch to fitnet for regression and curve fitting and patternnet for pattern recognition and classification.
help patternnet.
doc patternnet.
Use zscore or mapstd to standardize inputs. Use tansig, softmax, and trainscg with columns of eye(3) as targets. Use trial and error to find a suitale value for H, the number of hidden nodes.
See all of the documentation, examples and demos for patternnet, fitnet and feedforward net (replacing newpr, newfit and newff)
Hope this helps.
Greg