I have to develop a Radial Basis Function network for handwritten digits classification and have some problem of how many neurons on the output neuron should I use. I have two ideas: -1 10 hidden neurons (one for each digit) and just one output neuron that computes the linear combination (e.g the weight are arranged in a way that the sum outputs the number) -2 use 10 output neurons, one for each digit and the output is a vector with 0s and 1 in the correct position. The number of hidden neurons has to be chose


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.