  MLP Network Sizing
 
  1. Functions of the Network Sizing Program
   a. requests a training data file
   b. requests the number of inputs
   c. requests the number of outputs
   d. requests a maximum number of hidden units to try from the user.
      A typical number is 40.
   e. Estimates the MSE vs No. of hidden units and plot it.

  2. Example


  TWOD.TRA            ! input training data filename
  8                   ! number of inputs per pattern 
  7                   ! number of outputs per pattern 
  40                  ! maximum number of hidden units to try

      Here, we will estimate the MSE vs hidden units required for 
      a remote sensing problem.
      The program predicts that a network with different hidden units vs MSE. After training
      user can choose a suitable network structure i.e No. of hidden units.

      You can run this program on your own data.

