From: Eray Ozkural <erayo@cs.bilkent.edu.tr>
To: Oleg <oleg_inconnu@myrealbox.com>
Cc: caml-list <caml-list@inria.fr>
Subject: Re: [Caml-list] productivity improvement
Date: Tue, 15 Oct 2002 18:08:17 +0300 [thread overview]
Message-ID: <200210151808.17342.erayo@cs.bilkent.edu.tr> (raw)
In-Reply-To: <200210151233.IAA08545@apakabar.cc.columbia.edu>
On Tuesday 15 October 2002 15:34, Oleg wrote:
> On Tuesday 15 October 2002 05:31 am, Eray Ozkural wrote:
> > I have a feeling I can beat any ANN implementation written in C++ for
> > that matter ;) I was writing a generic ANN library in C++ and found it to
> > be quite difficult to put together different kinds of networks and
> > algorithms in the same basket. It would really benefit from a well
> > designed generic graph library which I can imagine would be possible only
> > in a functional language.
>
> "Beat" in what sense? Should you decide to write an O'Caml ANN library that
> learns better or faster than PDP++, SNNS and Torch, I can't imagine anyone
> trying to keep you from doing it.
>
I think better in the sense of extensibility, it could be made to allow more
sophisticated learning algorithms or ANN models. It could be made just as
efficient as any C code, or even faster who knows. Actually, what I have in
mind is a general purpose machine learning library which has all the standard
networks under the "ANN" module: single layer, multi layer feed forward
(together with BP), hopfield and kohonen nets... I wrote all that in C++ for
a grad course but I think it has its shortcomings, so I intend to rewrite it
in ocaml so that I can have a convenient machine learning shell.
Nevertheless, it's an awful lot of work if you want to have your interfaces
tidy.
> BTW BOOST has some sort of template graph library by Jeremy Siek.
Yes I know, but I prefer to use my own stuff. Even though that library is
supposed to go into C++ standard some time in the future ;) I think C++ will
be obsolete by then ;)
Here is some C++ client code for character recognition to give you a feel of
the approach I have in mind. I would like to have an ANN library that is more
generic than the one I crafted in C++ ;) Any ideas welcome.
typedef Neuron< Bipolar_Sigmoidal_Custom > Neuron;
typedef Sqr_Matrix<double,5> Matrix;
typedef Matrix_Source< Matrix > Source;
typedef pair<Source *, vector<double> > Training_Pair;
// use a multi layered neural network
Feed_Forward_Net<Neuron, Raw_Input_Neuron, Neuron> ff_net;
// add hidden neurons initialized at small random values.
for (int i=0; i<11; i++)
ff_net.add_hidden(Neuron(Rand::rand_double(-0.005, 0.005), 0));
// initialize with random values
ff_net.init_random();
// a square 5x5 matrix
Matrix Amtx;
// we now configure our network for ebp
Source source(Amtx);
ff_net.connect_input(source);
ff_net.connect_output(10); // our coding requires ten outputs
// the net has been put to required topology
// read the training sets into this training pairs list
vector< Training_Pair > pairs;
// this is all hardwired, not much config
// is required.
list< Matrix* > matrices;
// list< auto_ptr<Matrix> > matrices;
list< Source > sources;
for (int character = 0; character < 10; character++) {
ostrstream name_stream;
name_stream << "data/char-" << character << ".txt" << ends;
ifstream file_in( name_stream.str() );
// 4 patterns each
for (int i = 0; i<4; i++) {
Matrix *Amtx = new Matrix;
matrices.push_back( Amtx );
sources.push_back( Source(*Amtx) );
vector<double> desired = cons_max(10, character);
pairs.push_back( Training_Pair(&sources.back(), desired) );
file_in >> *Amtx;
}
}
// train the network with this data
// the learning coefficient is 0.02
CPU_Time start_time;
ff_net.train(pairs, 0.02, 0.00002, 100000);
cout << "Trained in " << CPU_Time() - start_time << endl;
// voila
nlog << ff_net << endl;
// now testing with the original training set
list< Source >::iterator source_it = sources.begin();
int correct_results = 0;
for (int character = 0; character < 10; character++) {
cout << "testing character " << character << endl;
for (int i = 0; i<4; i++) {
ff_net.compute(*source_it++);
vector<double> result_vec(10);
for (int res=0; res<10; res++)
result_vec[res] = ff_net.output_layer[res].read();
double result =select_max(result_vec);
cout << " pattern " << i
<< " : " << result << endl;
if (result == character)
correct_results++;
}
}
double train_success = double(correct_results) / 40 * 100;
cout << "Training success is " << train_success << "%" << endl;
and so forth...
--
Eray Ozkural (exa) <erayo@cs.bilkent.edu.tr>
Comp. Sci. Dept., Bilkent University, Ankara
www: http://www.cs.bilkent.edu.tr/~erayo Malfunction: http://mp3.com/ariza
GPG public key fingerprint: 360C 852F 88B0 A745 F31B EA0F 7C07 AE16 874D 539C
-------------------
To unsubscribe, mail caml-list-request@inria.fr Archives: http://caml.inria.fr
Bug reports: http://caml.inria.fr/bin/caml-bugs FAQ: http://caml.inria.fr/FAQ/
Beginner's list: http://groups.yahoo.com/group/ocaml_beginners
next prev parent reply other threads:[~2002-10-15 17:11 UTC|newest]
Thread overview: 77+ messages / expand[flat|nested] mbox.gz Atom feed top
[not found] <20020716172916.4903.qmail@web10702.mail.yahoo.com>
2002-07-18 23:14 ` Oleg
2002-07-18 23:27 ` Brian Smith
2002-07-18 23:54 ` William Lovas
2002-07-19 3:59 ` Oleg
[not found] ` <20020719010318.B3631@boson.den.co.bbnow.net>
2002-07-19 8:22 ` Oleg
2002-07-19 8:57 ` Andreas Rossberg
2002-07-19 10:14 ` Alessandro Baretta
2002-07-19 18:15 ` John Max Skaller
2002-07-19 18:33 ` Brian Smith
2002-07-20 17:30 ` John Max Skaller
2002-07-19 19:06 ` Alessandro Baretta
2002-07-20 17:49 ` John Max Skaller
2002-07-19 10:34 ` Oleg
2002-07-19 17:25 ` Andreas Rossberg
2002-07-20 16:58 ` John Max Skaller
2002-07-19 16:35 ` Brian Rogoff
2002-10-16 23:24 ` Eray Ozkural
2002-07-19 1:25 ` Alessandro Baretta
2002-07-19 4:04 ` Oleg
2002-07-19 15:46 ` [Caml-list] Rule based language [was: productivity improvement] Alessandro Baretta
2002-07-19 17:20 ` [Caml-list] compact.c Julie Farago
2002-10-15 9:31 ` [Caml-list] productivity improvement Eray Ozkural
2002-10-15 12:34 ` Oleg
2002-10-15 15:08 ` Eray Ozkural [this message]
2002-07-19 4:42 ` Emmanuel Renieris
2002-07-19 9:57 ` Oleg
2002-07-19 10:43 ` Alessandro Baretta
2002-07-19 10:52 ` Daniel de Rauglaudre
2002-07-19 11:36 ` Alessandro Baretta
2002-07-19 11:10 ` Xavier Leroy
2002-10-15 9:24 ` Eray Ozkural
2002-10-15 18:47 ` Pal-Kristian Engstad
2002-10-17 0:12 ` Eray Ozkural
2002-10-17 9:34 ` Diego Olivier Fernandez Pons
2002-10-17 15:55 ` Jeffrey Palmer
2002-10-17 16:15 ` brogoff
2002-10-17 18:21 ` [Caml-list] Re: Camlp4 optimizations (was: productivity improvement) Christophe TROESTLER
2002-10-17 18:32 ` Chris Hecker
2002-10-17 19:08 ` Shivkumar Chandrasekaran
2002-10-17 20:01 ` Chris Hecker
2002-10-17 19:36 ` Daniel de Rauglaudre
2002-10-17 19:59 ` Brian Hurt
2002-10-17 20:22 ` Chris Hecker
2002-10-17 21:19 ` Brian Hurt
2002-10-17 21:37 ` Jeffrey Palmer
2002-10-17 23:55 ` Alessandro Baretta
2002-10-18 0:57 ` Jeffrey Palmer
2002-10-18 4:21 ` Alessandro Baretta
2002-10-18 8:23 ` Remi VANICAT
2002-10-18 8:46 ` Sven Luther
2002-10-18 1:47 ` Brian Hurt
2002-10-17 23:03 ` Chris Hecker
2002-10-18 23:55 ` brogoff
2002-10-18 10:43 ` [Caml-list] productivity improvement Diego Olivier Fernandez Pons
2002-10-21 8:57 ` Francois Pottier
[not found] ` <200207200640.CAA11477@dewberry.cc.columbia.edu>
[not found] ` <3D391B41.50900@baretta.com>
[not found] ` <200207210059.UAA17003@dewberry.cc.columbia.edu>
2002-07-21 13:00 ` [Caml-list] Rule based language [was: productivity improvement] Alessandro Baretta
2002-07-23 9:53 ` Oleg
2002-07-24 8:07 ` Alessandro Baretta
[not found] <200207092004.QAA09587@psi-phi.mit.edu>
2002-07-09 20:16 ` [Caml-list] productivity improvement Oleg
2002-07-08 19:53 Oleg
2002-07-08 20:14 ` Michael Vanier
2002-07-10 15:50 ` John Max Skaller
2002-07-10 18:56 ` Alessandro Baretta
2002-07-10 19:09 ` Jun P.FURUSE
2002-07-11 23:43 ` Pierre Weis
2002-07-09 12:45 ` Basile STARYNKEVITCH
2002-07-09 18:20 ` Shannon --jj Behrens
2002-07-09 19:16 ` Oleg
2002-07-09 20:31 ` Shannon --jj Behrens
2002-07-10 10:02 ` sebastien FURIC
2002-07-10 11:58 ` Dave Mason
2002-07-10 13:11 ` sebastien FURIC
2002-07-10 19:22 ` nadji
2002-07-10 15:39 ` John Max Skaller
2002-07-11 8:57 ` Nicolas barnier
2002-07-16 3:34 ` Oleg
2002-10-18 3:13 ` Eray Ozkural
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=200210151808.17342.erayo@cs.bilkent.edu.tr \
--to=erayo@cs.bilkent.edu.tr \
--cc=caml-list@inria.fr \
--cc=oleg_inconnu@myrealbox.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).