File:       examples/ex_bmp2.pl
        Author: Josiah Bryan, <jdb@wcoil.com>
        Desc: 

                This demonstrates the ability of a neural net 
                to generalize and predict what the correct 
                result is for inputs that it has never seen before.
                                
                This teaches a network to recognize a 5x7 bitmap of 
                the letter "J" then it presents the network with a 
                corrupted "J" and displays the results of the networks 
                output.