Hi Alp,<br>- even with correctly programmed back-propagation, it is usually hard to make the net converge. <br>- usually you initialize neuron weights with somewhat random values, when working with back-propagation.<br>- do some debug prints of the net error while training to see how it is going<br>
- xor function cannot be trained with a single layer neural net !!!<br>Cheers,<br>Martin<br>PS: I did not check the back-propagation algorithm itself.<br><br><br><div class="gmail_quote">On Mon, Jun 15, 2009 at 9:58 AM, Alp Mestan <span dir="ltr"><<a href="mailto:alp@mestan.fr">alp@mestan.fr</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex;">Dear List,<br><br>I'm working with a friend of mine on a Neural Net library in Haskell.<br><br>There are 3 files : neuron.hs, layer.hs and net.hs.<br>
neuron.hs defines the Neuron data type and many utility functions, all of which have been tested and work well.<br>
layer.hs defines layer-level functions (computing the output of a whole layer of neurons, etc). Tested and working.<br>net.hs defines net-level functions (computing the output of a whole neural net) and the famous -- but annoying -- back-propagation algorithm.<br>
<br>You can find them there : <a href="http://mestan.fr/haskell/nn/html/" target="_blank">http://mestan.fr/haskell/nn/html/</a> <br><br>The problem is that here when I ask for final_net or test_output (anything after the train call, in net.hs), it seems to loop and loop around, as if it never gets the error under 0.1.<br>
<br>So I was just wondering if there was one or more Neural Nets and Haskell wizard in there to check the back-propagation implementation, given in net.hs, that seems to be wrong.<br><br>Thanks a lot !<br><br>-- <br><font color="#888888">Alp Mestan<br>
</font><br>_______________________________________________<br>
Haskell-Cafe mailing list<br>
<a href="mailto:Haskell-Cafe@haskell.org">Haskell-Cafe@haskell.org</a><br>
<a href="http://www.haskell.org/mailman/listinfo/haskell-cafe" target="_blank">http://www.haskell.org/mailman/listinfo/haskell-cafe</a><br>
<br></blockquote></div><br>