  Network Modeling and Pruning Programs
 

  1. Given a trained MLP network, i.e data file name and its weights;
 
   a. The pruning program measures the network' performance as a function of 
      the number of hidden units, and prunes the network of useless 
      hidden units. A network diagram is plotted, as is the mapping 
      error versus number of hidden units.
  
  2. The pruning program Prunes the net, then plot the No. of hidden units vs MSE and saved 
     the pruned network to a file name user specifed.

  3.Example

  twod.tra            ! data file
  weights.wts         ! weight file of trained network
     
      The program will prune the network for the file twod.tra, from the weights 
      file the program can get the network structure so this program do not need user
      give the structure. after pruning the program will ask user if he or she want to
      generate a pruned network and how many hidden units does it need.

      You can run this program on your own data.



