eolesson4.html

这是linux下的进化计算的源代码。 === === === === === === === === === === === ===== check latest news at http:

Name of replacement procedure. Availabable are the </font><b><font color="#FF6600">ES
plus and comma</font></b><font color="#000000"> deterministic replacement
strategies (named respectively </font><b><tt><font color="#009900"><font size=+1>Plus</font></font></tt></b><font color="#000000">
and&nbsp; </font><b><tt><font color="#009900"><font size=+1>Comma</font></font></tt></b><font color="#000000">);
</font><b><font color="#FF6600">EP
stochastic tournament</font></b><font color="#000000"> (name </font><b><tt><font color="#009900"><font size=+1>EPTour</font></font></tt></b><font color="#000000">
with tournament size in parentheses); and the </font><b><font color="#FF6600">steady-state
</font></b><font color="#000000">replacement
procedures, at the moment only based on fitnesses, replacement being either
</font><font color="#FF6600">deterministic</font><font color="#000000">
(new born replace worst parent, name </font><b><tt><font color="#009900"><font size=+1>SSGAWorst</font></font></tt></b><font color="#000000">),&nbsp;
or based on a tournament&nbsp; (name </font><b><tt><font color="#009900"><font size=+1>SSGADet</font></font></tt></b><font color="#000000">
for deterministic tournament, size in parentheses, and </font><b><tt><font color="#009900"><font size=+1>SSGAStoch</font></font></tt></b><font color="#000000">&nbsp;
for stochastic tournament, probability in parentheses). </font><font color="#FF0000">Default
is Comma</font><font color="#000000"> (which btw is also SGA </font><font color="#FF0000">generational
replacement</font><font color="#000000"> whenever there are as many offspring
as parents).</font>
<p><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="#FF0000">Boolean parameter:</font><font color="#000000">
if true, weak elitism is added to the replacement procedure (i.e. if the
best fitness among the offspring is less than the best fitness, the best
parent replaces the worst offspring). </font><font color="#FF0000">Default
is false.</font>
<p>
<hr SIZE=5 WIDTH="30%">
<p><b><font size=+1><font color="#FF0000">Section </font><tt><font color="#3333FF">######
Output ######</font></tt></font></b>
<br><font color="#000000">This first section on Output contains parameters
related to screen text output.</font>
<p><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="#FF0000">Boolean parameter:</font><font color="#000000">
whether or not you want the nb of evluations to be displayed and used as
counter in statistics outputs and plots. </font><font color="#FF0000">Default
is true</font><font color="#000000">.</font>
<p><b><tt><font color="#3333FF"><font size=+1># --printBestStat=1 # Print
Best/avg/stdev every gen.</font></font></tt></b>
<br><font color="#FF0000">Boolean parameter:</font><font color="#000000">
toggles screen output of indicated statistics. </font><font color="#FF0000">Default
is true</font><font color="#000000">.</font>
<p><b><tt><font color="#3333FF"><font size=+1># --printPop=0 # Print sorted
pop. every gen.</font></font></tt></b>
<br><font color="#FF0000">Boolean parameter:</font><font color="#000000">
adds a dump of the whole population to the screen every generation. Is
likely to generate </font><b><font color="#FF6600">huge</font></b><font color="#000000">
output! </font><font color="#FF0000">Default is false.</font>
<p><b><tt><font color="#3333FF"><font size=+1># --printFDC=1 # Print FDC
coeff. every gen.</font></font></tt></b>
<br><font color="#FF0000">Boolean parameter:</font><font color="#000000">
adds Fitness Distance Correlation to output every generation. </font><font color="#FF0000">Default
is false.</font>
<p>
<hr SIZE=5 WIDTH="30%">
<p><b><font size=+1><font color="#FF0000">Section </font><tt><font color="#3333FF">######
Output - Disk ######</font></tt></font></b>
<br><font color="#000000">This second section on Output contains parameters
related to DISK output.</font>
<p><b><tt><font color="#3333FF"><font size=+1># --resDir=Res # Directory
to store DISK outputs</font></font></tt></b>
<br><font color="#FF0000">String parameter: </font><font color="#000000">All
</font><font color="#FF6600">DISK
output</font><font color="#000000"> will be stored in a separate directory
-this is its name. If the directory does not exist, it is created. Note
that </font><font color="#FF6600">all graphical displays</font><font color="#000000">
will use that directory for their temporary files. Also </font><font color="#FF6600">all
job dump</font><font color="#000000"> (see section </font><b><font color="#3333FF">Persistence</font></b><font color="#000000">
below) store their files there too.</font>
<p><b><tt><font color="#3333FF"><font size=+1># --eraseDir=0 # erase files
in dirName if any</font></font></tt></b>
<br><font color="#FF0000">Boolean parameter:</font><font color="#000000">
in order not to mix up files from different runs, it is mandatory to ensure
that the directory where all files will be stored is empty. However, if
this parameter is not set and the directory already exists, an exception
is thrown and the program stops. It it is set, </font><b><font color="#FF6600">all
files in the result directory are erased</font></b><font color="#000000">.</font>
<p><b><tt><font color="#3333FF"><font size=+1># --fileBestStat=0 # Output
Best/avg/stdev to a file</font></font></tt></b>
<br><font color="#FF0000">Boolean parameter:</font><font color="#000000">
if present, the best, average and standard deviation statistics are stored
in file </font><b><tt><font color="#993300"><font size=+1>resDir/best.xg</font></font></tt></b><font color="#000000">.
Each line contains the generation number, eventualy the evaluation count
(depending on parameter </font><b><tt><font color="#3333FF"><font size=+1>useEval</font></font></tt></b><font color="#000000">
then the statistics. </font><font color="#FF0000">Default is false.</font>
<br>&nbsp;
<p>
<hr SIZE=5 WIDTH="30%">
<p><b><font size=+1><font color="#FF0000">Section </font><tt><font color="#3333FF">######
Output - Graphical ######</font></tt></font></b>
<br><font color="#000000">This last section on Output contains parameters
related to graphical output (only available in Unix through gnuplot at
the moment).</font>
<p><b><tt><font color="#3333FF"><font size=+1># --plotBestStat=0 # Plot
Best/avg Stat</font></font></tt></b>
<br><font color="#FF0000">Boolean parameter:</font><font color="#000000">
toggles gnuplot output of best and average plots (Linux only at the moment).
</font><font color="#FF0000">Default
is false.</font>
<p><b><tt><font color="#3333FF"><font size=+1># --plotFDCStat=0 # Plot
FDC scatter plot</font></font></tt></b>
<br><font color="#FF0000">Boolean parameter:</font><font color="#000000">
toggles the Fitness Distance Correlation plot (Fitness vs distance to best).
</font><font color="#FF0000">Default
is false.</font>
<p><b><tt><font color="#3333FF"><font size=+1># --plotHisto=0 # Plot histogram
of fitnesses</font></font></tt></b>
<br><font color="#FF0000">Boolean parameter:</font><font color="#000000">
if on, gnuplot is used to plot the sorted population (fitness vs rank).
Gives a graphical idea of the diversity. </font><font color="#FF0000">Default
is false.</font>
<p>
<hr SIZE=5 WIDTH="30%">
<p><b><font size=+1><font color="#FF0000">Section </font><tt><font color="#3333FF">######
Persistence ######</font></tt></font></b>
<br><font color="#000000">This section contains parameters handling job
dump and restart mechanism.</font>
<p><b><tt><font color="#3333FF"><font size=+1># --Load= # -L : A save file
to restart from</font></font></tt></b>
<br><font color="#FF0000">String parameter:</font><font color="#000000">
if present, the initial population (and the RNG) 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. </font><font color="#FF0000">No
default value</font><font color="#000000">.</font>
<p><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="#FF0000">Boolean parameter:</font><font color="#000000">
in case some individuals are read from a file, their fitness is read too.
If this one is true, it is nevertheless recomputed. </font><font color="#FF0000">Default
is false</font><font color="#000000"> i.e. use fitnes that's in the file.</font>
<p><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="#FF0000">Integer parameter:</font><font color="#000000">
interval between two dump to disk of the whole population (+RNG + parameters),
in a file named genNN.sav in the </font><b><tt><font color="#993300"><font size=+1>dirRes</font></font></tt></b><font color="#000000">
directory, where NN 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 </font><font color="#FF0000">no default
value</font><font color="#000000">).</font>
<p><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="#FF0000">Integer parameter:</font><font color="#000000">
time interval between two population (+RNG + parameters) dumps to disks.
Files are names timeNN.sav. See pervious parameter description for ore
details. </font><font color="#FF0000">No default value.</font>
<p><b><tt><font color="#3333FF"><font size=+1># --status=t-eoGA.status
# Status file</font></font></tt></b>
<br><font color="#FF0000">String parameter:</font><font color="#000000">
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). Default value is </font><font color="#FF0000">ProgramName.status</font>
<p>
<hr SIZE=5 WIDTH="30%">
<p><b><font size=+1><font color="#FF0000">Section </font><tt><font color="#3333FF">######
Stopping criterion ######</font></tt></font></b>
<br><font color="#000000">This section allows to decide when the algorithm
will stop.</font>
<p><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.
A value of 0 disables that stopping criterion. </font><font color="#FF0000">Default
is 100.</font>
<p><b><tt><font color="#3333FF"><font size=+1># --steadyGen=100 # -s :
Number of generations with no improvement</font></font></tt></b>
<br><font color="#FF0000">Integer parameter:</font><font color="#000000">
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. </font><font color="#FF0000">No default value.</font>
<p><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.
</font><font color="#FF0000">No
default value.</font>
<p><b><tt><font color="#3333FF"><font size=+1># --maxEval=0 # -E : Maximum
number of evaluations (0 = none)</font></font></tt></b>