nojhan/eodev
Archived
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity
This repository has been archived on 2026-03-28. You can view files and clone it, but you cannot make any changes to its state, such as pushing and creating new issues, pull requests or comments.
eodev/eo/tutorial/html/eoLesson4.html

237 lines
15 KiB
HTML
Raw Blame History

<!doctype html public "-//w3c//dtd html 4.0 transitional//en">
<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1">
<meta name="GENERATOR" content="Mozilla/4.73 [en] (X11; I; Linux 2.2.15-4mdk i686) [Netscape]">
<title>Tutorial: Lesson 3</title>
</head>
<body text="#000000" link="#0000EE" vlink="#551A8B" alink="#FF0000" background="beige009.jpg">
<a href="eoLesson2.html">Lesson 3</a> -
<a href="eoLesson4.html">Lesson
5</a> -
<a href="eoTutorial.html">Main page</a> -
<a href="eoTopDown.html">Algorithm-Based</a>
- <a href="eoBottomUp.html">Component-Based</a> - <a href="eoProgramming.html">Hints</a>
- <b><font face="Arial,Helvetica"><font size=+1><a href="../../doc/html/index.html">EO
documentation</a></font></font></b>
<br>
<hr WIDTH="100%"><!-- -------------- End of header ------------------ --><!-- ----------------------------------------------- -->
<center>
<h1>
<font color="#FF0000">Tutorial Lesson 4: fully operational EA</font></h1></center>
In this lesson, you will still use the same Evolutionary Algorithm. But
this time you will have <b><font color="#FF6600">full control of all components</font></b>
from the c<b><font color="#FF6600">ommand-line or a parameter file</font></b>.
You can even use the algorithm decribed here without any other knowledge
of EO, just by writing your fitness function as a plain C++ function. This
is why this lesson starts with a <a href="#programmerguide">user's guide</a>,
most of it being representation-independent, with some parts that are specific
of respectively the <a href="#binary">binary</a> and the <a href="#real">real</a>
algorithms.
<br>However, the ultimate purpose of this tutorial is to be able to do
your own experiments - and these these will likely fall outside the scope
of these two programs. This is why you should also read the programmer's
guides, as the structure and memory managements are here radically different
that in the 3 previous lessons - though relying of course on the same objects.
<br>&nbsp;
<p>
<hr WIDTH="100%">
<br><a NAME="userguide"></a><b><font color="#000099"><font size=+2>User's
guide</font></font></b>
<br><font color="#000000">As already said, the behavior of the algorithms
will be exactly the same as the previous one as far as optimization is
concerned. Only now you will be able to tune every component of the algorithms
- except the type of genotype - using run-time parameters.</font>
<br><font color="#000000">Also, as in previous lessons, most of the code
is representation-independent, i.e. is the same for both the binary genotypes
and the real-valued genotypes. This small user's guide reflects that, but
you can go directly to the <a href="#binary">binary</a> or the <a href="#real">real</a>
parts if you wish. Parameters input The way to input parameters has already
be described in <a href="eoLEsson3.html#paraminput">Lesson 3</a>. To get
a list of parameters, type the command with option --help (or -h): with
both testBit and testReal this will result in</font>
<ul>
<li>
<font color="#000000">Printing the list of keywords on the standard output</font></li>
<li>
<font color="#000000">Creating (or overwriting) a file name testBit.status
or testReal.status that contains the list of all recognized parameters
and has the format of an input parameter file.</font></li>
</ul>
<b><font color="#000099"><font size=+1>User's guide:</font></font></b><font color="#000000">
</font><b><font color="#FF0000">The status file</font></b>
<br><font color="#000000">This file will always contain the list of the
parameters that have been actually used by the last run of the program,
however thay have been entered (try </font><b><tt><font color="#FF6666"><font size=+1>testBit
-G1</font></font></tt></b><font color="#000000"> and take a look a the
status file). The parameters that are commented out (a # character comments
out the rest of the line) in the file were not specified by the user.</font><font color="#000000"></font>
<p><b><font color="#000099"><font size=+1>User's guide:</font></font></b><font color="#000000">
</font><b><font color="#FF0000">Representation-independent parameters</font></b><font color="#000000"></font>
<br><font color="#000000">The parameters are organized in sections.</font>
<br><font color="#000000">The first section only contains the random seed.</font><font color="#000000"></font>
<p><b><tt><font color="#993300"><font size=+1>###### General ######</font></font></tt></b>
<br><b><tt><font color="#993300"><font size=+1># --help=0 # -h : Prints
this message</font></font></tt></b>
<br><font color="#000000">Whether or not help was requested is handled
through a boolean parameter</font>
<br><b><tt><font color="#993300"><font size=+1># --seed=988700289 # -S
: Random number seed</font></font></tt></b>
<br><font color="#000000">The seed for the <a href="eoProgramming.html#random">Random
Number Generator</a></font>
<p><b><tt><font color="#3333FF"><font size=+1>###### Output ######</font></font></tt></b>
<br><font color="#000000">This section contains parameters related to output
(to screen, to files, graphical, ...).</font>
<br><b><tt><font color="#3333FF"><font size=+1># --useEval=1 # Use nb of
eval. as counter (vs nb of gen.)</font></font></tt></b>
<br><font color="#000000">Boolean parameter, whether or not you want the
nb of evluations to be displayed and used as counter in plots</font>
<br><b><tt><font color="#3333FF"><font size=+1># --printBestStat=1 # Print
Best/avg/stdev every gen.</font></font></tt></b>
<br><font color="#000000">Boolean parameter, toggles screen output of indicated
statistics</font>
<br><b><tt><font color="#3333FF"><font size=+1># --plotBestStat=0 # Plot
Best/avg Stat</font></font></tt></b>
<br><font color="#000000">Boolean parameter, toggles gnuplot output of
best and average plots (Linux only at the moment)</font>
<br><b><tt><font color="#3333FF"><font size=+1># --BestFileName=best.xg
# Name of file for Best/avg/stdev</font></font></tt></b>
<br><font color="#000000">String parameter, if present, the statistics
are stored in that file (no default)</font>
<br><b><tt><font color="#3333FF"><font size=+1># --printPop=0 # Print sorted
pop. every gen.</font></font></tt></b>
<br><font color="#000000">Boolean parameter, adds a dump of the whole population
to the screen every generation</font>
<br><b><tt><font color="#3333FF"><font size=+1># --printFDC=1 # Print FDC
coeff. every gen.</font></font></tt></b>
<br><font color="#000000">Boolean parameter, adds Fitness Distance Correlation
to output every generation</font>
<br><b><tt><font color="#3333FF"><font size=+1># --plotFDCStat=0 # Plot
FDC scatter plot</font></font></tt></b>
<br><font color="#000000">Boolean parameter, toggles the Fitness Distance
Correlation plot (Fitness vs distance to best)</font>
<br><b><tt><font color="#3333FF"><font size=+1># --plotHisto=0 # Plot histogram
of fitnesses</font></font></tt></b>
<br><font color="#000000">Boolean parameter: if on, gnuplot is used to
plot the sorted population (fitness vs rank). Gives a graphical idea of
the diversity.</font><font color="#000000"></font>
<p><b><tt><font color="#3333FF"><font size=+1>###### Persistence ######</font></font></tt></b>
<br><font color="#000000">This section contains parameters handling job
restart</font>
<br><b><tt><font color="#3333FF"><font size=+1># --Load= # -L : A save
file to restart from</font></font></tt></b>
<br><font color="#000000">String parameter: if present, the initial population
(and the RNG status) is read from indicated file. That file </font><b><font color="#FF0000">must</font></b><font color="#000000">
come from a previous save (or must be in same format!), i.e. must contain
a popualtion, the RNG and all parameters. If no other parameter is modified,
using a previously saved population and RNG will give exactly the same
results than having run that previous run longer. And a way to be sure
to re-use the same parameters is to ... use that very save file as parameter
file, as it contains all actual parameters in the right format.</font>
<br><font color="#000000">Note that if not enough individuals are read,
the remaining are randomly initialized. No default value.</font>
<br><b><tt><font color="#3333FF"><font size=+1># --recomputeFitness=0 #
-r : Recompute the fitness after re-loading the pop.?</font></font></tt></b>
<br><font color="#000000">Boolean parameter: in case some individuals are
read from a file, their fitness is read too. If this one is true, it is
nevertheless recomputed.</font>
<br><b><tt><font color="#3333FF"><font size=+1># --saveFrequency=0 # Save
every F generation (0 = only final state, absent = never)</font></font></tt></b>
<br><font color="#000000">Integer parameter: interval between two dump
to disk of the whole population (+RNG + parameters) to disk, in a file
named genNN.sav, where NN&nbsp;is the generation number. If this prameter
is present (even with 0 or negative value), the final population will always
be saved, whatever the reason for stopping. Hence the only way to avoid
all saves is to omit the parameter (there is no default value).</font>
<br><b><tt><font color="#3333FF"><font size=+1># --saveTimeInterval=0 #
Save every T seconds (0 or absent = never)</font></font></tt></b>
<br><font color="#000000">Integer parameter: time interval between two
population (+RNG + parameters) dumps to disks. Files are names timeNN.sav.
See pervious parameter description for ore details. No default value.</font>
<br><b><tt><font color="#3333FF"><font size=+1># --status=t-eoGA.status
# Status file</font></font></tt></b>
<br><font color="#000000">String parameter: name of the status file (that
contains all parameters in the input format). There is no way to avoid
creating that file except recompiling ... or giving the name /dev/null
(Unix).</font><b><tt><font color="#3333FF"><font size=+1></font></font></tt></b>
<p><b><tt><font color="#3333FF"><font size=+1>###### Stopping criterion
######</font></font></tt></b>
<br><font color="#000000">This section allows to decide when the algorithm
will stop.</font>
<br><b><tt><font color="#3333FF"><font size=+1># --maxGen=100 # -G : Maximum
number of generations (0 = none)</font></font></tt></b>
<br><font color="#000000">Integer parameter: maximum number of generations.
Default is 0 which is equivalent to infinity.</font>
<br><b><tt><font color="#3333FF"><font size=+1># --steadyGen=100 # -s :
Number of generations with no improvement</font></font></tt></b>
<br><font color="#000000">Integer parameter: stops whenever that number
of generations is passed without any improvement of the best fitness in
the population, provided the following minimum number of generations has
been done. No default value.</font>
<br><b><tt><font color="#3333FF"><font size=+1># --minGen=0 # -g : Minimum
number of generations</font></font></tt></b>
<br><font color="#000000">Integer parameter: the above steadyGen parameter
starts its job only after that minimum nuber of generations is passed.
No default value.</font>
<br><b><tt><font color="#3333FF"><font size=+1># --maxEval=0 # -E : Maximum
number of evaluations (0 = none)</font></font></tt></b>
<br><font color="#000000">Integer parameter: maximum number of generations.
No default value.</font>
<br><b><tt><font color="#3333FF"><font size=+1># --targetFitness=0 # -T
: Stop when fitness reaches</font></font></tt></b>
<br><font color="#000000">Real-valued parameter: the algorithm stops whenever
the best fitness reaches that target. No default value.</font>
<br><b><tt><font color="#3333FF"><font size=+1># --CtrlC=0 # -C : Terminate
current generation upon Ctrl C</font></font></tt></b>
<br><font color="#000000">Boolean parameter: if true, Ctrl C only stops
after the current generation as completed (eventually dumping population
to a file if some saver is active).</font>
<p><b><tt><font color="#009900"><font size=+1>###### engine ######</font></font></tt></b>
<br><font color="#000000">In this section, one chooses all components of
the Evolution Engine (selection, replacemenet and the like).</font>
<br><b><tt><font color="#009900"><font size=+1># --selection=DetTour(2)
# -S : Selection: Roulette, DetTour(T), StochTour(t) or Sequential(ordered/unordered)</font></font></tt></b>
<br><font color="#000000">String parameter: Name of selection procedure.
Availabable are the </font><b><font color="#FF6600">roulette wheel</font></b><font color="#000000">
(name </font><b><tt><font color="#009900"><font size=+1>Roulette</font></font></tt></b><font color="#000000">,
fitness scaling coming soon), </font><b><font color="#FF6600">deterministic
tournament</font></b><font color="#000000"> (name </font><b><tt><font color="#009900"><font size=+1>DetTour</font></font></tt></b><font color="#000000">
with size - integer > 2 - in parentheses right after the name, use double
quotes on the command line),&nbsp; stochastic tournament (name </font><b><tt><font color="#009900"><font size=+1>StochTour</font></font></tt></b><font color="#000000">
with probability - float in [0.5, 1] - in parentheses), sequential (name
</font><b><tt><font color="#009900"><font size=+1>Sequential</font></font></tt></b><font color="#000000">,
all individuals in turn), either from best to worst (option </font><b><tt><font color="#009900"><font size=+1>ordered</font></font></tt></b><font color="#000000">
in parentheses), or in random ordered (option </font><b><tt><font color="#009900"><font size=+1>unordered</font></font></tt></b><font color="#000000">)
or finally repeated independent uniform choices&nbsp; (name </font><b><tt><font color="#009900"><font size=+1>Random</font></font></tt></b><font color="#000000">).</font>
<br><b><tt><font color="#009900"><font size=+1># --offspringRate=100% #
-O : Nb of offspring (percentage or absolute)</font></font></tt></b><b><tt><font color="#009900"><font size=+1></font></font></tt></b>
<p><b><tt><font color="#009900"><font size=+1># --replacement=Comma # -R
: Replacement: Comma, Plus or EPTour(T)</font></font></tt></b>
<br><font color="#000000">String parameter: Name of replacement procedure.
Availabable are</font>
<br><b><tt><font color="#009900"><font size=+1># --weakElitism=0 # -w :
Old best parent replaces new worst offspring *if necessary*</font></font></tt></b>
<br><font color="#009900"></font>&nbsp;
<p>
<hr WIDTH="100%">
<br><a NAME="programmerguide"></a><b><font color="#000099"><font size=+2>Programmer's
guide
<hr WIDTH="100%"></font></font></b><font color="#000000"><a href="eoLesson2.html">Lesson
3</a> -
<a href="eoLesson4.html">Lesson 5</a> -
<a href="eoTutorial.html">Main
page</a> -
<a href="eoTopDown.html">Algorithm-Based</a> - <a href="eoBottomUp.html">Component-Based</a>
- <a href="eoProgramming.html">Hints</a> -<b> <font face="Arial,Helvetica"><font size=+1><a href="../../doc/html/index.html">EO
documentation</a></font></font></b></font>
<br>
<hr>
<address>
<font color="#000000"><a href="mailto:Marc.Schoenauer@polytechnique.fr">Marc
Schoenauer</a></font></address>
<br><!-- Created: Tue May 1 14:49:12 CET 2001 --><!-- hhmts start --><font color="#000000">Last
modified: None of your business!</font><!-- hhmts end -->
</body>
</html>
Reference in a new issue View git blame Copy permalink