
% source /home/artific/broggi/bp.setup
% cd bp
% ls 
esempi/     esercizio/
% cd esempi
% ls
README       circle.bp    circle.tra   rsin.bp      xor.bp
bp*          circle.tes   ric.bp       sin.bp
% bp xor.bp 
 
Basis of AI Backprop (c) 1990-96 by Donald R. Tveter
   drt@mcs.com - http://www.mcs.com/~drt/home.html
              April 10, 1996 version.
seed = 7, range = - 1.00 to + 1.00
4 training patterns read

taking commands from stdin now
[ACDFGMNPQTW?!acdefhlmopqrstw]? t 0.01
[ACDFGMNPQTW?!acdefhlmopqrstw]? r
running . . .
   10      0.00 % 0.50000   
   20      0.00 % 0.49992   
   30      0.00 % 0.49956   
   40      0.00 % 0.49704   
   50      0.00 % 0.46327   
   60      0.00 % 0.24057   
   70      0.00 % 0.07574   
   80      0.00 % 0.03375   
   90      0.00 % 0.02140   
  100      0.00 % 0.01639   
[ACDFGMNPQTW?!acdefhlmopqrstw]? r
running . . .
  110     25.00 % 0.01375   
  120     25.00 % 0.01207   
  130     25.00 % 0.01086   
  140     50.00 % 0.00991   
  150     50.00 % 0.00913   
  160     75.00 % 0.00847   
  170     75.00 % 0.00790   
  172    100.00 % 0.00780    DONE
[ACDFGMNPQTW?!acdefhlmopqrstw]? p
    1  0.995  e 0.005 ok
    2  0.009  e 0.009 ok
    3  0.990  e 0.010 ok
    4  0.007  e 0.007 ok
  172    (TOL) 100.00 % (4 right  0 wrong)  0.00780 err/unit
[ACDFGMNPQTW?!acdefhlmopqrstw]? 0 0
       0.009 

[ACDFGMNPQTW?!acdefhlmopqrstw]? 0 1
       0.990 

[ACDFGMNPQTW?!acdefhlmopqrstw]? 1 1
       0.007 

[ACDFGMNPQTW?!acdefhlmopqrstw]? 1 0
       0.995 

[ACDFGMNPQTW?!acdefhlmopqrstw]? w 3 1
layer unit  inuse  unit value    weight   inuse   input from unit
  1     1     1      1.00000    -9.55455     1       -9.55455
  1     2     1      0.00000     9.42363     1        0.00000
  2     1     1      0.98952    19.87322     1       19.66496
  3     b     1      1.00000    -4.82407     2       -4.82407
                                              sum =   5.28634

[ACDFGMNPQTW?!acdefhlmopqrstw]? w 2 1
layer unit  inuse  unit value    weight   inuse   input from unit
  1     1     1      1.00000     9.93119     1        9.93119
  1     2     1      0.00000   -10.21070     1       -0.00000
  2     b     1      1.00000    -5.38342     2       -5.38342
                                              sum =   4.54777

[ACDFGMNPQTW?!acdefhlmopqrstw]? q
%


 ================== xor.bp ===========================

m 2 1 1 x   * make a 2-1-1 network
s7          * seed the random number generator
ci 1        * giving the network a "kick" sets new random weights
a as        * standard smooth sigmoid

rt {        * there are four patterns following
1 0 1
0 0 0
0 1 1
1 1 0 }

e 0.5       * set the learning rate, eta to 0.5
a 0.9       * set the momentum, alpha to 0.9

 =====================================================

	CLASSIFICAZIONE IN DUE CLASSI

% bp circle.bp

per leggere il test set:   tf circle.tes
per vedere i training patterns: p
per vedere i test patterns: t

 =====================================================

	LEARNING DELLA FUNZIONE sin(x)

% bp sin.bp

 =====================================================

	LEARNING DELLA FUNZIONE sin(x) 
	CON RETE RICORRENTE

% bp rsin.bp

per inserire un pattern:  <real> H

 =====================================================

	ALTRO ESEMPIO DI RETE RICORRENTE

% bp ric.cp

provare con soglia 0.01  (comando: t 0.01)

 =====================================================


