Neural Network Program - Revision history

Thomas Linder Puls: Code layout

2009-05-01T11:52:25Z

Code layout

← Older revision		Revision as of 13:52, 1 May 2009
Line 43:		Line 43:

	<vip>clauses		<vip>clauses
	i(0).		i(0).
	i(1).		i(1).
	i(2).		i(2).
	i(3).		i(3).
	i(4).		i(4).
	i(8).		i(8).
	i(7).		i(7).
	i(6).		i(6).
	i(5).		i(5).
	o(9).
	c(5,4,0.25).		clauses
	c(6,4,0.25).		o(9).
	c(7,4,0.25).
	c(8,4,0.25).		clauses
	c(3,4,0.25).		c(5,4,0.25).
	c(2,4,0.25).		c(6,4,0.25).
	c(1,4,0.25).		c(7,4,0.25).
	c(0,4,0.25).		c(8,4,0.25).
	c(4,9,0.5).</vip> First is the numbers of the input neurons, then the numbers of the output neurons (here just a single) and finally the connections (there is a signal path from neuron <vp>5</vp> to neuron <vp>4</vp> with weight <vp>0.25</vp>).		c(3,4,0.25).
			c(2,4,0.25).
			c(1,4,0.25).
			c(0,4,0.25).
			c(4,9,0.5).</vip>

			First is the numbers of the input neurons, then the numbers of the output neurons (here just a single) and finally the connections (there is a signal path from neuron <vp>5</vp> to neuron <vp>4</vp> with weight <vp>0.25</vp>).

	<vp>netSaver</vp> and <vp>netLoader</vp> uses red-black-trees (in exactly the same way) to speed up creation of the other representation of the net. When saving we need to translate neurons to numbers. So whenever we meet a neuron we call a predicate <vp>getOrCreate</vp> which will first look in the red-black-tree to see of that neuron has already been translated, or otherwise create the translation and insert it in the red-black-tree. The important thing about red-black-trees is that look-up and insert are proportional to the depth of the tree. And when the tree contains <vp>N</vp> elements it has a maximal depth of <vp>2*log2(N)</vp>. So if there are for example <vp>65536</vp> elements in the tree the maximal depth is <vp>32</vp>.		<vp>netSaver</vp> and <vp>netLoader</vp> uses red-black-trees (in exactly the same way) to speed up creation of the other representation of the net. When saving we need to translate neurons to numbers. So whenever we meet a neuron we call a predicate <vp>getOrCreate</vp> which will first look in the red-black-tree to see of that neuron has already been translated, or otherwise create the translation and insert it in the red-black-tree. The important thing about red-black-trees is that look-up and insert are proportional to the depth of the tree. And when the tree contains <vp>N</vp> elements it has a maximal depth of <vp>2*log2(N)</vp>. So if there are for example <vp>65536</vp> elements in the tree the maximal depth is <vp>32</vp>.

Thomas Linder Puls: Categorized

2008-10-29T20:09:07Z

Categorized

← Older revision		Revision as of 22:09, 29 October 2008
Line 88:		Line 88:

	The bottom layer is the input layer, and it can for example be feed with the pixel values from an image.		The bottom layer is the input layer, and it can for example be feed with the pixel values from an image.

			[[Category:Examples]]

Thomas Linder Puls: First version

2008-06-30T09:30:50Z

First version

@@ Line 1: / Line 1: @@
-This article describes a little neural network program [http://discuss.visual-prolog.com].  The program nor the article can be considered finished.  I hope that other persons interested in neural networks will help finishing it.
+This article describes a little neural network program that can be downloaded from [http://discuss.visual-prolog.com/viewtopic.php?t=7819 Neural Network Program].
-The program can:
+The program is not finished, I hope that people with interest in neural networks will help extend it and maintain this article.
 * Build a network
 * Save it to a file
@@ Line 7: / Line 9: @@
 * Calculate the network
 But it cannot train the net, as I don't know how training works.

Thomas Linder Puls: New page: This article describes a little neural network program [http://discuss.visual-prolog.com]. The program nor the article can be considered finished. I hope that other persons interested in...

2008-06-30T07:28:55Z

New page: This article describes a little neural network program [http://discuss.visual-prolog.com]. The program nor the article can be considered finished. I hope that other persons interested in...

New page

This article describes a little neural network program [http://discuss.visual-prolog.com]. The program nor the article can be considered finished. I hope that other persons interested in neural networks will help finishing it.

The program can:
* Build a network
* Save it to a file
* Load it from a file
* Calculate the network
But it cannot train the net, as I don't know how training works.