1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
|
function demhint(nin, nhidden, nout)
%DEMHINT Demonstration of Hinton diagram for 2-layer feed-forward network.
%
% Description
%
% DEMHINT plots a Hinton diagram for a 2-layer feedforward network with
% 5 inputs, 4 hidden units and 3 outputs. The weight vector is chosen
% from a Gaussian distribution as described under MLP.
%
% DEMHINT(NIN, NHIDDEN, NOUT) allows the user to specify the number of
% inputs, hidden units and outputs.
%
% See also
% HINTON, HINTMAT, MLP, MLPPAK, MLPUNPAK
%
% Copyright (c) Ian T Nabney (1996-2001)
if nargin < 1 nin = 5; end
if nargin < 2 nhidden = 7; end
if nargin < 3 nout = 3; end
% Fix the seed for reproducible results
randn('state', 42);
clc
disp('This demonstration illustrates the plotting of Hinton diagrams')
disp('for Multi-Layer Perceptron networks.')
disp(' ')
disp('Press any key to continue.')
pause
net = mlp(nin, nhidden, nout, 'linear');
[h1, h2] = mlphint(net);
clc
disp('The MLP has been created with')
disp([' ' int2str(nin) ' inputs'])
disp([' ' int2str(nhidden) ' hidden units'])
disp([' ' int2str(nout) ' outputs'])
disp(' ')
disp('One figure is produced for each layer of weights.')
disp('For each layer the fan-in weights are arranged in rows for each unit.')
disp('The bias weight is separated from the rest by a red vertical line.')
disp('The area of each box is proportional to the weight value: positive')
disp('values are white, and negative are black.')
disp(' ')
disp('Press any key to exit.');
pause;
delete(h1);
delete(h2);
|
