Nabble - JGAP

Re: crossover probability

2009-12-04T01:18:32Z

Krzysztof,

the JGAP documentation refers to the probability of selecting individuals
into the next generation, NOT to the probability of crossing over.
So your understanding in 1) is correct

Regarding your point 2):
This is not correct, I think: You can write your own implementation that
controls the probability of crossing over dependent on the individuals'
fitness .

In general: It only makes sense to crossover individuals that are selected
beforehand.

Best

Klaus
www.klaus-meffert.com

> -----Original Message-----
> From: Krzysztof Chmielewski [mailto:krzysiek.chmielewski@...]
> Sent: Friday, December 04, 2009 1:50 AM
> To: jgap-users@...
> Subject: [jgap-users] crossover probability
>
> Hi, I have a question to crossover probability. On JGAP web
> site we can read that: *
>
> *Natural selection in JGAP is statistical, so chromosomes
> that are more fit have a better statistical chance of being
> selected over chromosomes that are less fit, but it's not guaranteed.
>
> So I assumed, that probability of choosing chromosome to
> crossover depending on fitness value of this chromosome. But
> when in studied the CrossOverOperator i have noticed
> something like that:
>
> public void operate(final Population a_population,
> final List a_candidateChromosomes) {
> // Work out the number of crossovers that should be performed.
> // -----------------------------------------------------------
> int size = Math.min(getConfiguration().getPopulationSize(),
> a_population.size());
> int numCrossovers = 0;
> if (m_crossoverRate >= 0) {
> numCrossovers = size / m_crossoverRate;
> }
> else if (m_crossoverRateCalc != null) {
> numCrossovers = size /
> m_crossoverRateCalc.calculateCurrentRate();
> }
> else {
> numCrossovers = (int) (size * m_crossoverRatePercent);
> }
>
> Where calculateCurrentRate() return int.
>
> So:
> 1. The fitness value of chromosome does not impact on the
> probability of choosing this chromosome to crossover right??
> The only place where the fitness value is used is in
> NaturalSelector where we choose "n" best chromosomes, yes?
> 2. Even if I wrote my own UniversalRateCalc, the
> rate(probability) of crossover is the same for every
> chromosome in this evolution, yes?
>
> So what should I do if I want that probability of crossover
> of particular chromosome is an individual value for each
> chromosome in this population and depends of fitness value of
> this chromosome?
>
>
>
> --------------------------------------------------------------
> ----------------
> Join us December 9, 2009 for the Red Hat Virtual Experience,
> a free event focused on virtualization and cloud computing.
> Attend in-depth sessions from your desk. Your couch. Anywhere.
> http://p.sf.net/sfu/redhat-sfdev2dev
> _______________________________________________
> jgap-users mailing list
> jgap-users@...
> https://lists.sourceforge.net/lists/listinfo/jgap-users

------------------------------------------------------------------------------
Join us December 9, 2009 for the Red Hat Virtual Experience,
a free event focused on virtualization and cloud computing.
Attend in-depth sessions from your desk. Your couch. Anywhere.
http://p.sf.net/sfu/redhat-sfdev2dev
_______________________________________________
jgap-users mailing list
jgap-users@...
https://lists.sourceforge.net/lists/listinfo/jgap-users

From forum: jgap-users

crossover probability

2009-12-03T16:50:18Z

Hi, I have a question to crossover probability. On JGAP web site we can
read that: *

*Natural selection in JGAP is statistical, so chromosomes that are more
fit have a better statistical chance of being selected over chromosomes
that are less fit, but it's not guaranteed.

So I assumed, that probability of choosing chromosome to crossover
depending on fitness value of this chromosome. But when in studied the
CrossOverOperator i have noticed something like that:

public void operate(final Population a_population,
final List a_candidateChromosomes) {
// Work out the number of crossovers that should be performed.
// -----------------------------------------------------------
int size = Math.min(getConfiguration().getPopulationSize(),
a_population.size());
int numCrossovers = 0;
if (m_crossoverRate >= 0) {
numCrossovers = size / m_crossoverRate;
}
else if (m_crossoverRateCalc != null) {
numCrossovers = size / m_crossoverRateCalc.calculateCurrentRate();
}
else {
numCrossovers = (int) (size * m_crossoverRatePercent);
}

Where calculateCurrentRate() return int.

So:
1. The fitness value of chromosome does not impact on the probability of
choosing this chromosome to crossover right?? The only place where the
fitness value is used is in NaturalSelector where we choose "n" best
chromosomes, yes?
2. Even if I wrote my own UniversalRateCalc, the rate(probability) of
crossover is the same for every chromosome in this evolution, yes?

So what should I do if I want that probability of crossover of
particular chromosome is an individual value for each chromosome in this
population and depends of fitness value of this chromosome?

------------------------------------------------------------------------------
Join us December 9, 2009 for the Red Hat Virtual Experience,
a free event focused on virtualization and cloud computing.
Attend in-depth sessions from your desk. Your couch. Anywhere.
http://p.sf.net/sfu/redhat-sfdev2dev
_______________________________________________
jgap-users mailing list
jgap-users@...
https://lists.sourceforge.net/lists/listinfo/jgap-users

From forum: jgap-users

grid fitness distribution question

2009-11-30T09:12:32Z

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

- From following the examples I have created an application that should
distribute the fitness calculations across multiple workers. The fitness
calculation takes about 3 minutes to run per chromosome so it is
essential it is distributed.

However in the example for fitness distribution it uses the line
pop.sortByFitness() which causes all the fitness calculations to then be
done for each chromosome on the client before the requests are split up
for the workers? Removing this line does send the fitness requests to
the workers. This line is need though as pointed out in the example
snce they need to be sorted here in order to keep the best chromosomes
for when the population gets bigger. As such I am wondering if this
might be a bug in the example or if I am missing something.

I have a similar issue when it is working on the evolving. I would have
thought that the way that the fitness distribution was implemented that
any time it needed a fitness it would pass it to a worker. However
during the evolving it calculates the fitness of new chromosomes as they
are generated, which by default is handled on the client. I see some
examples on distributing the evolving, but this seems like it should be
for a different purpose.

Any advice on how to fix these issues would be appreciated.

Thanks,
Bryan
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.10 (MingW32)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/

iQEcBAEBAgAGBQJLE/0AAAoJEIB5nnZY0MtS9jAH+wVjezxDyorOfS4gyLmyKcgc
YJskmUPj9Sts8N4iZfLP0xCgGo2DNjAqIIpbNUPhb3j59pTwmoTXYwEWp4PuO2Zi
BofSGoNlh2C8SPl1Ycl8kAMUhuMCvBkFHMFnxdRByoD1pkCMHlzk49Jjt+P51yik
irw2UTzjrfYh3wyFaL/vCnJTELncdqDkLRcPHPwlBijrQxL6fB6xbbVzrbgILQ7d
aqdXrJKgsgzSYUBajky/pCo15kqpnS8HeRPu6/fhQAAXuT8svyt5U9z9zqKpHf1K
7Klp0RYiTdQvQHtxoYaZP4ML8py3H515KCUVGaFecxTuXxQwaU91L/q7IxKYgGk=
=0BaT
-----END PGP SIGNATURE-----

------------------------------------------------------------------------------
Join us December 9, 2009 for the Red Hat Virtual Experience,
a free event focused on virtualization and cloud computing.
Attend in-depth sessions from your desk. Your couch. Anywhere.
http://p.sf.net/sfu/redhat-sfdev2dev
_______________________________________________
jgap-users mailing list
jgap-users@...
https://lists.sourceforge.net/lists/listinfo/jgap-users

From forum: jgap-users

Re: HOW TO TEST AN ALGORITHM FOR GSM APPLICABILITY?

2009-11-23T02:35:43Z

Message

Halil,

could you tell me what you exactly mean with GSM in your case?

Best

Klaus

----- original Nachricht --------

Betreff: [jgap-users] HOW TO TEST AN ALGORITHM FOR GSM APPLICABILITY?
Gesendet: So, 22. Nov 2009
Von: Halil Memeti

I have been wondering how do I test whether my algorithm is applicable for the GSM.

I mean, I have been implementing an algorithm that would allow same frequency operating within a cell. However, I would like to try to test whether that is applicable for GSM.

Any help in this area would be highly appreciated!

Thanks in advance!

--- original Nachricht Ende ----

------------------------------------------------------------------------------
Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
trial. Simplify your report design, integration and deployment - and focus on
what you do best, core application coding. Discover what's new with
Crystal Reports now. http://p.sf.net/sfu/bobj-july
_______________________________________________
jgap-users mailing list
jgap-users@...
https://lists.sourceforge.net/lists/listinfo/jgap-users

From forum: jgap-users

Re: Chromosomes with varient number of genes

2009-11-23T02:29:23Z

Gianpaolo,

having chromosomes with a varying number of genes is quite easy. Just use a CompositeGene and inside it the desired genes you need. But the thing is that your genetic operators (mainly the crossover operator, or also a custom one) must be able to handle this case.
I think, adding an example for this may be helpful. For that it would be great to have a problem description, or a source code from you, that allows setting up an example quite easily. Please also describe which genetic operators are applicable and how they should work. Maybe a reference to the GABIL source (hopefully in Java) also helps.

Best

Klaus
----- original Nachricht --------

Betreff: Re: [jgap-users] Chromosomes with varient number of genes
Gesendet: So, 22. Nov 2009
Von: Gianpaolo Valero<alexander.gvc@...>

>
> Hi all,
>
> I'm using GABIL for GA. I wanted to use JGAP to solve this problem. So,
> i have the same problem that Sajith has. I just read the solution you
> posted
> and there's something that's not clear.
>
> You wrote "you have to do is create a String Gene". So, my choromosomes
> would have only one gene with different size?. Is that the solution you're
> giving?
>
> andycrack wrote:
> >
> > Hi Sajith
> > From my experience with JGAP I think you have to do is
> > create a String Gene, because in this class you can set a minimal and
> > maximal Length of the string, maybe if you use a random generator of
> > integers also include in this library you can simulate diferent
> > chromosomes with genes dynamically.
> >
> > I hope this is a good answer for you
> >
> > best regards
> >
> > Christian
> >
> > On 11/14/09, Sajith Vimukthi <sajith.vim@...> wrote:
> >> Hi,
> >>
> >> I am very new to GA and JGAP. Indeed I want to select an optimal set if
> >> words out of a huge text. For that I though of using a GA. When I
> thought
> >> about that in the context of GA, I felt that there should be some
> ability
> >> where I should be able to define the number of genes in a chromosome
> >> dynamically. Is it possible with GA? Does JGAP support that?
> >> --
> >> Thanks,
> >> Regards,
> >> Sajith Vimukthi Weerakoon,
> >> T .P No : ++94-716102392
> >>
> >
> >
> ----------------------------------------------------------------------------
> --
> > Let Crystal Reports handle the reporting - Free Crystal Reports 2008
> > 30-Day
> > trial. Simplify your report design, integration and deployment - and
> focus
> > on
> > what you do best, core application coding. Discover what's new with
> > Crystal Reports now. http://p.sf.net/sfu/bobj-july
> > _______________________________________________
> > jgap-users mailing list
> > jgap-users@...
> > https://lists.sourceforge.net/lists/listinfo/jgap-users
> >
> >
>
> --
> View this message in context:
> http://old.nabble.com/Chromosomes-with-varient-number-of-genes-tp26349456p26
> 462191.html
> Sent from the jgap-users mailing list archive at Nabble.com.
>
>
> ----------------------------------------------------------------------------
> --
> Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
>
> trial. Simplify your report design, integration and deployment - and focus
> on
> what you do best, core application coding. Discover what's new with
> Crystal Reports now. http://p.sf.net/sfu/bobj-july
> _______________________________________________
> jgap-users mailing list
> jgap-users@...
> https://lists.sourceforge.net/lists/listinfo/jgap-users
>

From forum: jgap-users

GABIL with JGAP

2009-11-21T17:57:01Z

Hi,

I just wanted to know if there's a way of using GABIL with JGAP. The biggest problem i've found is that GABIL uses non-fixed-length chromosomes and chromosomes in JGAP are fixed-lenght.

If there's a way how?

From forum: jgap-devl

Re: Chromosomes with varient number of genes

2009-11-21T17:50:59Z

Hi all,

I'm using GABIL for GA. I wanted to use JGAP to solve this problem. So, i have the same problem that Sajith has. I just read the solution you posted and there's something that's not clear.

You wrote "you have to do is create a String Gene". So, my choromosomes would have only one gene with different size?. Is that the solution you're giving?

andycrack wrote:

Hi Sajith
From my experience with JGAP I think you have to do is
create a String Gene, because in this class you can set a minimal and
maximal Length of the string, maybe if you use a random generator of
integers also include in this library you can simulate diferent
chromosomes with genes dynamically.

I hope this is a good answer for you

best regards

Christian

On 11/14/09, Sajith Vimukthi <sajith.vim@gmail.com> wrote:
> Hi,
>
> I am very new to GA and JGAP. Indeed I want to select an optimal set if
> words out of a huge text. For that I though of using a GA. When I thought
> about that in the context of GA, I felt that there should be some ability
> where I should be able to define the number of genes in a chromosome
> dynamically. Is it possible with GA? Does JGAP support that?
> --
> Thanks,
> Regards,
> Sajith Vimukthi Weerakoon,
> T .P No : ++94-716102392
>

------------------------------------------------------------------------------
Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
trial. Simplify your report design, integration and deployment - and focus on
what you do best, core application coding. Discover what's new with
Crystal Reports now. http://p.sf.net/sfu/bobj-july
_______________________________________________
jgap-users mailing list
jgap-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/jgap-users

From forum: jgap-users

GABIL with JGAP

2009-11-21T17:46:43Z

Hi,

Next monday I have to turned in a java project for acadamy porpuse that uses GABIL. Actually I'm using GABIL to do machine learning.

First of all, for what i've seen JGAP is a nice package for solvin almost every genetic problem. Although it doesn't seem to work with GABIL framework.

The basis of GABIL is that chromosomes are not fixed-length. That is, that every chromosomes can have different numbers of genes. I've looked around in your code and i noticed that your BaseChromosome and IChromosome worked with arrays of genes. Is too dificult to change this classes to work whith vector or ArrayList for example?. If not, how to do that?. If it is, Do you know where I can find a package that handles GABIL solving methods?.

Thanks for your time, and i'll keep looking into your code to see if there is a way I can do what i wan.

Sincerely yours,

Gianpaolo A. Valero @Venezuela

------------------------------------------------------------------------------
Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
trial. Simplify your report design, integration and deployment - and focus on
what you do best, core application coding. Discover what's new with
Crystal Reports now. http://p.sf.net/sfu/bobj-july
_______________________________________________
jgap-users mailing list
jgap-users@...
https://lists.sourceforge.net/lists/listinfo/jgap-users

From forum: jgap-users

HOW TO TEST AN ALGORITHM FOR GSM APPLICABILITY?

2009-11-21T12:57:29Z

From forum: jgap-users

Two bugs(?) in AbstractSuperGene

2009-11-16T05:19:24Z

Hi all,

I have discovered two "inconveniences" in the AbstractSuperGene implementation. I'm not sure whether
they can outright be called bugs, but they're certainly exposing rather unexpected behaviour.

First:
The constructor AbstractSupergene(Configuration, Gene[]), in conjunction with the default
newGeneInternal() implementation.

This is implemented in a slightly odd way, because new Objects get instantiated with the *old*,
uncloned subgenes - which are replaced by the newGeneInternal() method *after* the constructor of
the copy had been invoked. IMO this goes against intuition (and usability).

To give an example where it becomes obvious: I need to reference the "parent" supergene from within
my subgenes at times. So the intuitive approach would be (simplified pseudo-code):

class MySuperGene extends AbstractSuperGene {
public MySuperGene(Configuration c, Gene[] subGenes) {
for (Gene g: subGenes) {
((SubGene)g).setParent(this);
}
}
}

With the current implementation, one needs to create workarounds for this (the easiest being to
"adjust" the newGeneInternal() method. (Which I did anyway for performance reasons, because the
constructor invocation through reflection seems to quickly become a bottleneck). In any case, I
would suggest to minimally change the newGeneInternal() method, so that it uses the copied gene
array for the constructor (pseudo-diff):

AbstractSuperGene.java:200
-AbstractSupergene age = (AbstractSupergene) constr.newInstance(new Object[] {getConfiguration(),
getGenes()});
+AbstractSupergene age = (AbstractSupergene) constr.newInstance(new Object[] {getConfiguration(), g});
if (m_validator != this) {
age.setValidator(m_validator);
}
- age.m_genes = g;

I don't think this would break any existing functionality, but IMO makes life easier and exposes the
intuitively expected behaviour.

Second:
I'm heavily using SuperGenes in my project, including SuperGenes composed of SuperGenes composed of
SuperGenes.... And I realized that not everything mutates ;-)
While trying to figure out the reason, I took a closer look at the AbstractSuperGene.applyMutation()
method. Indeed, it's even described in the Javadoc: The method is delegated to the first element of
the gene indexed by 'index'.

Which left me wondering whether that really makes sense... Shouldn't it be mutating the other
elements as well, at times? Or, putting it the other way around, and a bit bluntly: if the other
subgenes are never mutated, why would one implement them as genes in the first place? They could
just as well go (for example) into the applicationData as constants.

I thus changed my "private" implementation so that it randomly picks one of the subgenes to be
mutated. (I'm not sure as to how this should be affected by the mutation rate and such, but that's a
different story...) There's one more catch to this however, in that one can't simply use the
RandomGenerator from the Configuration, because this one doesn't necessarily give a uniform
distribution...

Ok, I hope that what I wrote is half-way understandable and makes sense as changes to the
AbstractSuperGene. Any feedback or comments on these issues is highly appreciated. :-)

Thanks a lot & cheers
Christoph

------------------------------------------------------------------------------
Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
trial. Simplify your report design, integration and deployment - and focus on
what you do best, core application coding. Discover what's new with
Crystal Reports now. http://p.sf.net/sfu/bobj-july
_______________________________________________
jgap-users mailing list
jgap-users@...
https://lists.sourceforge.net/lists/listinfo/jgap-users

smime.p7s (9K) Download Attachment

From forum: jgap-users

Re: Problem generating population

2009-11-16T00:44:21Z

Bruno,

sorry for the late reply.

Is it possible in your case to increase the maxInitDepth?
Do you believe there is a program error or bad program behaviour?

Is it possible to get the configuration of your problem to dig in deeper and
develop a good solution for your issue?

Best

Klaus
www.klaus-meffert.com

> -----Original Message-----
> From: Bruno Fernández [mailto:bfg84@...]
> Sent: Monday, October 26, 2009 5:49 PM
> To: jgap-users@...
> Subject: [jgap-users] Problem generating population
>
> Hello. I'm using genetic programming with jgap and I have a
> problem while generating the initial population. I have a
> type which is output of only one function but it's input of
> many of them and there are no terminals of this type. The
> problem is, when I try to generate the initial population
> with "GPGenotype.randomInitialGenotype", it can't find
> suitable nets and clone only a few. As a result, I have no
> enough variability.
>
> The problem seems to be that it can't close the net if one
> node with the problematic type as an input appears in the
> (maxInitDepth-1) level. The bigger the maxInitDepth constant
> is, the bigger is the time it takes to generate the
> population and with worst results.
>
> Is there any posibility to make laxer the condition so, if
> this situation happens, it will try to close the branch as
> soon as possible even if it breaks the max depth restriction?
> Any other solution will be welcome too.
>
> Thank you very much.
>
> Sorry if my english is not good enough, I hope the message
> were understandable.
>
>
>
>
> --------------------------------------------------------------
> ----------------
> Come build with us! The BlackBerry(R) Developer Conference in
> SF, CA is the only developer event you need to attend this
> year. Jumpstart your developing skills, take BlackBerry
> mobile applications to market and stay ahead of the curve.
> Join us from November 9 - 12, 2009. Register now!
> http://p.sf.net/sfu/devconference
> _______________________________________________
> jgap-users mailing list
> jgap-users@...
> https://lists.sourceforge.net/lists/listinfo/jgap-users

From forum: jgap-users

Require suggestion for given problem.

2009-11-15T21:23:38Z

Hello,

We are developing system for stacking Circle Cylinder shape items on containers. (Like coin stack..)
And this problem have many constraints based on container type. (i.e Weight, Diameter, Type, Color)
We have defined these constraints into two categories . Hard and Soft (From analogy of JBoss Drools Solver, I tried it before JGAP)

Hard constraints are compulsory. If it does not fulfills then solution cannot be determined.
Soft constraints are constraints which should be optimized (I have set to Lower score means better score)

And Input will be “List of Containers” and “List of Cylinders”. Usual input will be 10 to 30 Cylinders and 4 to 8 containers
And Output should be filling up all the containers + [remaining unassigned cylinders] OR using all the cylinders
[AND Hard constraints must be adhered and soft constrains should be near to optimum]

Right now, I am giving Hard constraint much higher fitness then soft constraint. So if ONE hard constraint is broken then fitness will be say 30000, and if ONE soft constraint is broken then fitness will be 2500

Let me give you a hypothetical example for better understanding.
Circles:
C1[height: 5,diameter:6,group-A],
C2[height: 7,diameter:4,group-B] ,
C3[height: 3,diameter:3,group-D],
C4[height: 4,diameter:8,group-C],
C5[height: 2,diameter:6,group-A],
C6[height: 8,diameter:9,group-A],
C7[height: 3,diameter:2,group-D]
C8[height: 6,diameter:2,group-B]

Containers:
CON1[max-diameter:5, height: 10, max-capacity:3],
CON2[max-diameter:5, height: 10, max-capacity:3],
CON3[max-diameter:10, height: 8, max-capacity:3],
CON4[max-diameter:10, height: 8, max-capacity:3],

Hard Constraints
1. Container height
2. Container diameter
3. Max capacity

Soft Constraints
1. Minimum capacity wastage of containers
2. Lower diameter circle cannot be placed on higher diameter circle (i.e C3 on layer-1 and C2 on layer-2 should be avoided)
3. Circles with group [A,B] or [C,D] should be kept in one base. (i.e. rather than putting C1,C3 together prefer, C1,C5 together)

So in above detail, Input is CIRCLES + CONTAINERS and output should be (it’s just a representation, not a proper solution);
CON1 [0:C1, 1:C5,2:C6], (0,1,2 denotes layers)
CON2 [0:C2, 1:C4,2:C7],
CON3 [0:C3, 1:C5]
CON4 []

In this result, CON4 will be empty.. as there are no Circles available.
Sometimes in input number of circles are more around 30 and number of Containers are lesser (like 3 or 4) so, in that case.. elements list will remain unassigned.

I am maintaining a index based list and mapping it with IntegerGene. So if my chromosome is [2,4,6,1,0,3,5] then I’ll create a new List<Cylinder> for fitness calculation.

Fitness calculation function

It start filling Containers based on height and weight capacity. From given List<Cylinder> (which is created based on chromosome). After setting all Cylinders in Containers. Fitness value will be calculated on wastage of weight capacity and height capacity.

So please suggest better options for my settings. Here is my configuration.

Configuration gaConf = new DefaultConfiguration();
Configuration.resetProperty(Configuration.PROPERTY_FITEVAL_INST);
gaConf.setFitnessEvaluator(new DeltaFitnessEvaluator());

gaConf.getGeneticOperators().clear();

MutationOperator swapper = new SwappingMutationOperator(gaConf);
gaConf.addGeneticOperator(swapper);

gaConf.setPreservFittestIndividual(true);
gaConf.setKeepPopulationSizeConstant(false);
gaConf.setPopulationSize(getSizeOfPopulation());

IChromosome sampleChromosome = new Chromosome(config, new IntegerGene(config), chromozomeSize);
config.setSampleChromosome(sampleChromosome);
config.setFitnessFunction(getFitnessFunction());
Genotype genotype = Genotype.randomInitialGenotype(config);
List chromosomes = genotype.getPopulation().getChromosomes();
for (Object chromosome : chromosomes) {
     IChromosome chrom = (IChromosome) chromosome;
     for (int j = 0; j < chrom.size(); j++) {
         Gene gene = chrom.getGene(j);
         gene.setAllele(j);
     }
}

· Is my configuration okay? OR I am missing something?

· Is IntegerGene appropriate for this problem?

· I am only using SwappingMutationOperation.

I am giving 250 NUMBER_OF_EVOLUTION and 50 SIZE_OF_POPULATION..
But after 100 to 120 number of evolutions.. I do not get better fitness chromosome.

Please let me know if you required more information..

Regards,

P Please don't print this e-mail unless you really need to.

Disclaimer:
The information contained in this electronic message and in any attachments to this message is confidential, legally privileged and intended only for the person or entity to which this electronic message is addressed. If you are not the intended recipient, please notify the system manager and you are hereby notified that any distribution, copying, review, retransmission, dissemination or other use of this electronic transmission or the information contained in it is strictly prohibited. Please accordingly also note that any views or opinions presented in this email are solely those of the author and may not represent those of the Company or bind the Company. This message has been scanned for viruses and dangerous content by Mail Scanner, and is believed to be clean. The Company accepts no liability for any damage caused by any virus transmitted by this email. www.essar.com < http://www.essar.com/ > .

------------------------------------------------------------------------------
Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
trial. Simplify your report design, integration and deployment - and focus on
what you do best, core application coding. Discover what's new with
Crystal Reports now. http://p.sf.net/sfu/bobj-july
_______________________________________________
jgap-users mailing list
jgap-users@...
https://lists.sourceforge.net/lists/listinfo/jgap-users

From forum: jgap-users

Re: Chromosomes with varient number of genes

2009-11-14T05:50:44Z

Hi Sajith
From my experience with JGAP I think you have to do is
create a String Gene, because in this class you can set a minimal and
maximal Length of the string, maybe if you use a random generator of
integers also include in this library you can simulate diferent
chromosomes with genes dynamically.

I hope this is a good answer for you

best regards

Christian

On 11/14/09, Sajith Vimukthi <sajith.vim@...> wrote:

> Hi,
>
> I am very new to GA and JGAP. Indeed I want to select an optimal set if
> words out of a huge text. For that I though of using a GA. When I thought
> about that in the context of GA, I felt that there should be some ability
> where I should be able to define the number of genes in a chromosome
> dynamically. Is it possible with GA? Does JGAP support that?
> --
> Thanks,
> Regards,
> Sajith Vimukthi Weerakoon,
> T .P No : ++94-716102392
>

From forum: jgap-users

Question about Genetic engine

2009-11-14T01:57:37Z

Hi, my name is Krzysztof Chmielewski. First I would like to thank You
for doing great Job, JGAP is something very, very helpful. For last few
months I’m working on school project where my task is to find optimal
resource allocation in telecommunication network, and I want to use
Genetic Algorithms. But I have some questions about Your GP Engine.

I Would like to know what are the values of most important parameters,
like probability of reproduction, crossover, mutation, and can I change
them. Then, after the reproduction, does GP engine use only new
chromosomes to compare with fitness function or does it use parents and
children? And if the chromosome fits better to fitness function, would
it make any influence on the probability of choosing this chromosome to
reproduce, and if yes then how, what is the formula? Is there any
documentation for this?

Best regards

Krzysztof Chmielewski

------------------------------------------------------------------------------
Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
trial. Simplify your report design, integration and deployment - and focus on
what you do best, core application coding. Discover what's new with
Crystal Reports now. http://p.sf.net/sfu/bobj-july
_______________________________________________
jgap-users mailing list
jgap-users@...
https://lists.sourceforge.net/lists/listinfo/jgap-users

From forum: jgap-users

Chromosomes with varient number of genes

2009-11-13T22:03:10Z

Hi,

I am very new to GA and JGAP. Indeed I want to select an optimal set if words out of a huge text. For that I though of using a GA. When I thought about that in the context of GA, I felt that there should be some ability where I should be able to define the number of genes in a chromosome dynamically. Is it possible with GA? Does JGAP support that?
--
Thanks,
Regards,
Sajith Vimukthi Weerakoon,
T .P No : ++94-716102392

------------------------------------------------------------------------------
Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
trial. Simplify your report design, integration and deployment - and focus on
what you do best, core application coding. Discover what's new with
Crystal Reports now. http://p.sf.net/sfu/bobj-july
_______________________________________________
jgap-users mailing list
jgap-users@...
https://lists.sourceforge.net/lists/listinfo/jgap-users

From forum: jgap-users

explicit reduction of population size on each generation

2009-10-29T18:18:37Z

Hi,

I want to be able to explicitly control my population size with each generation. I will be starting with a large population and reducing at some rate. I am unclear on how to do this. I tried the following:

private static Genotype reduce (Genotype genotype, int nsize)
{
try
{
List<IChromosome> chromosomes = genotype.getFittestChromosomes (nsize);

Configuration gaconfig = genotype.getConfiguration ();
gaconfig.reset();
gaconfig.setPopulationSize (nsize);

Population population = new Population (gaconfig, (IChromosome[])chromosomes.toArray ());
return new Genotype (gaconfig, population);
}
catch (InvalidConfigurationException e)
{ throw new RuntimeException ("failed to configure GA", e); }
}

However JGAP does not allow the configuration to be modified after it has been in use. What is the best approach here?

---
Jonathan Shore
Systematic Trading Group

From forum: jgap-users

checkpointing JGAP

2009-10-29T18:14:14Z

Hi,

I have a long running calibration using JGAP. I have on occassion had faults with hardware or power, and/or would like to stop a run and be able to restart at the last "checkpoint".

I note that JGAP supports various forms of serialization for classes. As it stands is it possible for me to save the state of JGAP (ie the chromosomes and configuration) such that I can restart JGAP at the start of generation "i". I would be saving checkpoints of the state at the end of each generation.

Of course I would restore my own state. Just want to know the approach needed with JGAP to saftely restore state.

regards

Jonathan Shore
Systematic Trading Group

From forum: jgap-users

Genotype.randomInitialGenotype() does not clone the sample chromosome application data

2009-10-27T15:13:29Z

I expected the app data from the sample chromosome to be copied whenever the new chromosomes are generated from the sample one.
Is this a valid expectation?

I tracked this down to Chromosome.randomInitialChromosome(Configuration). It creates a new chromosome with new Chromosome(a_configuration, newGenes). Would it be better to clone the chromosome and then set the new genes and configuration?

Thanks,
Andriy

------------------------------------------------------------------------------
Come build with us! The BlackBerry(R) Developer Conference in SF, CA
is the only developer event you need to attend this year. Jumpstart your
developing skills, take BlackBerry mobile applications to market and stay
ahead of the curve. Join us from November 9 - 12, 2009. Register now!
http://p.sf.net/sfu/devconference
_______________________________________________
jgap-users mailing list
jgap-users@...
https://lists.sourceforge.net/lists/listinfo/jgap-users

From forum: jgap-users

Problem generating population

2009-10-26T09:48:50Z

Hello. I'm using genetic programming with jgap and I have a problem while generating the initial population. I have a type which is output of only one function but it's input of many of them and there are no terminals of this type. The problem is, when I try to generate the initial population with "GPGenotype.randomInitialGenotype", it can't find suitable nets and clone only a few. As a result, I have no enough variability.

The problem seems to be that it can't close the net if one node with the problematic type as an input appears in the (maxInitDepth-1) level. The bigger the maxInitDepth constant is, the bigger is the time it takes to generate the population and with worst results.

Is there any posibility to make laxer the condition so, if this situation happens, it will try to close the branch as soon as possible even if it breaks the max depth restriction? Any other solution will be welcome too.

Thank you very much.

Sorry if my english is not good enough, I hope the message were understandable.

------------------------------------------------------------------------------
Come build with us! The BlackBerry(R) Developer Conference in SF, CA
is the only developer event you need to attend this year. Jumpstart your
developing skills, take BlackBerry mobile applications to market and stay
ahead of the curve. Join us from November 9 - 12, 2009. Register now!
http://p.sf.net/sfu/devconference
_______________________________________________
jgap-users mailing list
jgap-users@...
https://lists.sourceforge.net/lists/listinfo/jgap-users

From forum: jgap-users

Re: GP grid examples

2009-10-21T02:02:55Z

Simone,

in the JGAP installation directory there are three batch (.bat) files
related to the grid functionality.

However, this example is only for GA and not for GP, as you already pointed
out.

I will add some batch files to the next release that show how to use GP in a
grid.
Basically, the class JGAPClientGP has to be used together with a
configuration like examples.grid.mathProblemDistributed.GridConfiguration.

If I can manage, I will send some example batch files before the next
release.

Best

Klaus
www.klaus-meffert.com

> -----Original Message-----
> From: Simone Bronzin [mailto:simone.bronzin@...]
> Sent: Tuesday, October 20, 2009 5:38 PM
> To: jgap-users@...
> Subject: [jgap-users] GP grid examples
>
>
> Is there any working example for using the grid feature in a
> gp context? I see specific packages for the gp case, but no examples.
>
> TIA
>
> --
>
> Simone Bronzin
> Unitary
> Executive
>
> mobile: +39.3939459672
>
>
>
> --------------------------------------------------------------
> ----------------
> Come build with us! The BlackBerry(R) Developer Conference in
> SF, CA is the only developer event you need to attend this
> year. Jumpstart your developing skills, take BlackBerry
> mobile applications to market and stay ahead of the curve.
> Join us from November 9 - 12, 2009. Register now!
> http://p.sf.net/sfu/devconference
> _______________________________________________
> jgap-users mailing list
> jgap-users@...
> https://lists.sourceforge.net/lists/listinfo/jgap-users

------------------------------------------------------------------------------
Come build with us! The BlackBerry(R) Developer Conference in SF, CA
is the only developer event you need to attend this year. Jumpstart your
developing skills, take BlackBerry mobile applications to market and stay
ahead of the curve. Join us from November 9 - 12, 2009. Register now!
http://p.sf.net/sfu/devconference
_______________________________________________
jgap-users mailing list
jgap-users@...
https://lists.sourceforge.net/lists/listinfo/jgap-users

From forum: jgap-users

GP grid examples

2009-10-20T08:37:55Z

Is there any working example for using the grid feature in a gp
context? I see specific packages for the gp case, but no examples.

TIA

--

Simone Bronzin
Unitary
Executive

mobile: +39.3939459672

------------------------------------------------------------------------------
Come build with us! The BlackBerry(R) Developer Conference in SF, CA
is the only developer event you need to attend this year. Jumpstart your
developing skills, take BlackBerry mobile applications to market and stay
ahead of the curve. Join us from November 9 - 12, 2009. Register now!
http://p.sf.net/sfu/devconference
_______________________________________________
jgap-users mailing list
jgap-users@...
https://lists.sourceforge.net/lists/listinfo/jgap-users

From forum: jgap-users

Re: Population generation and verbose level doubts

2009-10-14T06:59:18Z

Hello Rodrigo,

egarding 1):

This is an issue with random program generation. In case no program can be evolved randomly, that sticks to the syntax of the GP commands and terminals, this message is returned.

Regarding 2)

You can use the log4j.properties file. Open the file and set the log level to DEBUG (instead of INFO, see log4.rootCategory).

Besides, you can add logging to the code. In case you need something special that could be of common interest, please tell me, maybe we can integrate into the JGAP core.

Best

Klaus
www.klaus-meffert.com

From: Rodrigo Tripodi [mailto:rtripodi@...]
Sent: Wednesday, September 30, 2009 2:38 PM
To: jgap-users@...
Subject: [jgap-users] Population generation and verbose level doubts

Hi!

I'm using jgap for my mater's project. It's a math problem and I've some doubts.

1) When I get the warning "WARN GPPopulation - Prototype program reused because random program did not satisfy constraints" what does it mean? If the size of my population is, for example, 10 and I get 10 warnings like that it means that all the generated individuals are equal? How could I know which program is this prototype one?

2) How can I set the maximum verbose level for jgap?

thanks in advance.

best,

--
Rodrigo Tripodi
UNICAMP - BRAZIL

From forum: jgap-users

Re: Problem with Custom Quarter Gene

2009-10-13T05:32:03Z

Christian,

the code sample in the documentation was outdated. I updated it. Please try again.

In case of further issues please don't hesitate coming back again.

Thanx for your feedback and best

Klaus

www.klaus-meffert.com

From: Christian Avendaño [mailto:andycrack@...]
Sent: Tuesday, October 13, 2009 4:07 AM
To: jgap-users@...
Subject: Re: [jgap-users] Problem with Custom Quarter Gene

Hi
When i try to compile te class Quarter Gene
i have the following errors

QuarterGene is not abstract and does not override abstract method newGeneInternal() in org.jgap.BaseGene Line 14
cannot find symbol constructor BaseGene() Line 26
cannot find symbol constructor BaseGene() Line 42

the source code i taken from http://jgap.sourceforge.net/doc/genes.html

If you can help me it would be fantastic
thanks a lot

regards
Christian

From forum: jgap-users

Re: Problem with Custom Quarter Gene

2009-10-12T19:07:16Z

Hi
When i try to compile te class Quarter Gene
i have the following errors

QuarterGene is not abstract and does not override abstract method newGeneInternal() in org.jgap.BaseGene Line 14
cannot find symbol constructor BaseGene() Line 26
cannot find symbol constructor BaseGene() Line 42

the source code i taken from http://jgap.sourceforge.net/doc/genes.html

If you can help me it would be fantastic
thanks a lot

regards
Christian

From forum: jgap-users

help : error DecoderException.java not found

2009-10-12T10:16:23Z

hi all,

i'm new a user of JGAP, i want to execute the example
multidimension.KnapsackMain and i found this error :

Exception in thread "main" java.lang.NoClassDefFoundError:
org/apache/commons/codec/DecoderException

could any one help me,
thanks in advance
regards
Amel

------------------------------------------------------------------------------
Come build with us! The BlackBerry(R) Developer Conference in SF, CA
is the only developer event you need to attend this year. Jumpstart your
developing skills, take BlackBerry mobile applications to market and stay
ahead of the curve. Join us from November 9 - 12, 2009. Register now!
http://p.sf.net/sfu/devconference
_______________________________________________
jgap-announce mailing list
jgap-announce@...
https://lists.sourceforge.net/lists/listinfo/jgap-announce

From forum: jgap-announce

Re: Bulk fitness functions (again)

2009-10-10T05:17:08Z

Christoph,

your thoughts sound logical. To be honest, I don't know how the position of
the call to the bulk fitness function evolved.
I changed the call sequence as you suggested. After some tests I will check
in the changes to CVS.
A new release of JGAP should appear within the next few weeks.

Best

Klaus
www.klaus-meffert.com

> -----Original Message-----
> From: Christoph Langguth [mailto:christoph.langguth@...]
> Sent: Wednesday, October 07, 2009 8:28 PM
> To: jgap-users@...
> Subject: [jgap-users] Bulk fitness functions (again)
>
> Hi,
>
> I'm currently taking a look at JGAP to replace my
> "home-grown" GA mini-setup for a project I'm working on. So
> far, I like it a lot :-)
>
> However, while investigating the bulk fitness functions, I
> ran into a problem - namely that essentially nothing evolves
> at all. Taking a closer look at the sources and browsing the
> latest archive, I ended up at this post:
>
> http://sourceforge.net/mailarchive/forum.php?thread_name=4ABD5
> ED8.8090202%40lizanddavid.com&forum_name=jgap-users
>
> I couldn't see or download the attachments from the post, but
> I think that my approach and conclusion are pretty much the
> same as the original author's:
>
> Why is the bulk fitness function applied *after* the selection?
> In other words: determining the current fitness should be a
> prerequisite for the selection -- what good is it to know the
> theoretical fitness of an individual after it has been thrown
> out of the population, based on old facts? ;-)
>
> I simply took the stock GABreeder, put it into my own class
> and, inside the evolve() method, changed the order to be
> bulkFunction.evaluate(); updateChromosomes();
> applyNaturalSelectors(); (that's around lines 115-130 of
> GABreeder.java).
>
> Then things looked correct to me, whereas, as said above,
> with the original GABreeder, no evolution at all took place.
>
> Could someone please comment on whether this is indeed a bug,
> or whether just a couple of people are misunderstanding the
> purpose of the BulkFitnessFunction?
>
> Even browsing through the Javadoc and looking at things like
> BulkFitnessOffsetRemover, it seems like these classes are
> intended to be applied before the selection, aren't they?
>
> Thanks & cheers
> Chris
>

From forum: jgap-users

Bulk fitness functions (again)

2009-10-07T11:27:30Z

Hi,

I'm currently taking a look at JGAP to replace my "home-grown" GA mini-setup for a project I'm
working on. So far, I like it a lot :-)

However, while investigating the bulk fitness functions, I ran into a problem - namely that
essentially nothing evolves at all. Taking a closer look at the sources and browsing the latest
archive, I ended up at this post:

http://sourceforge.net/mailarchive/forum.php?thread_name=4ABD5ED8.8090202%40lizanddavid.com&forum_name=jgap-users

I couldn't see or download the attachments from the post, but I think that my approach and
conclusion are pretty much the same as the original author's:

Why is the bulk fitness function applied *after* the selection?
In other words: determining the current fitness should be a prerequisite for the selection -- what
good is it to know the theoretical fitness of an individual after it has been thrown out of the
population, based on old facts? ;-)

I simply took the stock GABreeder, put it into my own class and, inside the evolve() method, changed
the order to be bulkFunction.evaluate(); updateChromosomes(); applyNaturalSelectors(); (that's
around lines 115-130 of GABreeder.java).

Then things looked correct to me, whereas, as said above, with the original GABreeder, no evolution
at all took place.

Could someone please comment on whether this is indeed a bug, or whether just a couple of people are
misunderstanding the purpose of the BulkFitnessFunction?

Even browsing through the Javadoc and looking at things like BulkFitnessOffsetRemover, it seems like
these classes are intended to be applied before the selection, aren't they?

Thanks & cheers
Chris

------------------------------------------------------------------------------
Come build with us! The BlackBerry(R) Developer Conference in SF, CA
is the only developer event you need to attend this year. Jumpstart your
developing skills, take BlackBerry mobile applications to market and stay
ahead of the curve. Join us from November 9 - 12, 2009. Register now!
http://p.sf.net/sfu/devconference
_______________________________________________
jgap-users mailing list
jgap-users@...
https://lists.sourceforge.net/lists/listinfo/jgap-users

smime.p7s (9K) Download Attachment

From forum: jgap-users

Population generation and verbose level doubts

2009-09-30T05:38:16Z

Hi!

I'm using jgap for my mater's project. It's a math problem and I've some doubts.

1) When I get the warning "WARN GPPopulation - Prototype program reused because random program did not satisfy constraints" what does it mean? If the size of my population is, for example, 10 and I get 10 warnings like that it means that all the generated individuals are equal? How could I know which program is this prototype one?

2) How can I set the maximum verbose level for jgap?

thanks in advance.

best,

--
Rodrigo Tripodi
UNICAMP - BRAZIL

------------------------------------------------------------------------------
Come build with us! The BlackBerry® Developer Conference in SF, CA
is the only developer event you need to attend this year. Jumpstart your
developing skills, take BlackBerry mobile applications to market and stay
ahead of the curve. Join us from November 9-12, 2009. Register now!
http://p.sf.net/sfu/devconf
_______________________________________________
jgap-users mailing list
jgap-users@...
https://lists.sourceforge.net/lists/listinfo/jgap-users

From forum: jgap-users

Re: Bug in CrossoverOperator with ICompositeGene

2009-09-28T06:53:41Z

Sebastien,

you are right, I fixed the code according to your suggestion and checked it in to CVS.

In the future, a more sophisticated x-over could be used/provided for CompositeGenes: The second gene can be potentially any one with the same type as the already chosen gene.

Best

Klaus
www.klaus-meffert.com

From: Sebastien Huet [mailto:sebhuet@...]
Sent: Monday, September 28, 2009 9:31 AM
To: jgap-users@...
Subject: [jgap-users] Bug in CrossoverOperator with ICompositeGene

Hi,

First thanks a lot for this incredible work.

He is the problem i faced.

I have a composite gene made with integers & booleans genes.
my program crashes when crossoving ( can not cast integer to boolean or boolean to integer).

if think the bug comes from CrossoverOperator.doCrossover:

the random index over the gene is generated two times and may choose two differents kind of gene.

// Make a distinction for ICompositeGene for the first gene.
      // ---------------------------------------------------------
      if (firstGenes[j] instanceof ICompositeGene) {
        // Randomly determine gene to be considered.
        // -----------------------------------------
        int index1 = generator.nextInt(firstGenes[j].size());
        gene1 = ( (ICompositeGene) firstGenes[j]).geneAt(index1);
      }
      else {
        gene1 = firstGenes[j];
      }
      // Make a distinction for the second gene if CompositeGene.
      // --------------------------------------------------------
      if (secondGenes[j] instanceof ICompositeGene) {
        // Randomly determine gene to be considered.
        // -----------------------------------------
        int index2 = generator.nextInt(secondGenes[j].size());
        gene2 = ( (ICompositeGene) secondGenes[j]).geneAt(index2);
      }
      else {
        gene2 = secondGenes[j];
      }

I think the code should be:

// Make a distinction for ICompositeGene for the first gene.

      int index = 0;
      // ---------------------------------------------------------
      if (firstGenes[j] instanceof ICompositeGene) {
        // -----------------------------------------
        // Randomly determine gene to be considered.
        index = generator.nextInt(firstGenes[j].size()); //Assuming the two genes have the same size
        gene1 = ( (ICompositeGene) firstGenes[j]).geneAt(index);
      }
      else {
        gene1 = firstGenes[j];
      }
      // Make a distinction for the second gene if CompositeGene.
      // --------------------------------------------------------
      if (secondGenes[j] instanceof ICompositeGene) {
        gene2 = ( (ICompositeGene) secondGenes[j]).geneAt(index);
      }
      else {
        gene2 = secondGenes[j];
      }

Regards.

Seb

------------------------------------------------------------------------------
Come build with us! The BlackBerry® Developer Conference in SF, CA
is the only developer event you need to attend this year. Jumpstart your
developing skills, take BlackBerry mobile applications to market and stay
ahead of the curve. Join us from November 9-12, 2009. Register now!
http://p.sf.net/sfu/devconf
_______________________________________________
jgap-users mailing list
jgap-users@...
https://lists.sourceforge.net/lists/listinfo/jgap-users

From forum: jgap-users

Bug in CrossoverOperator with ICompositeGene

2009-09-28T00:30:31Z

Hi,

First thanks a lot for this incredible work.

He is the problem i faced.

I have a composite gene made with integers & booleans genes.
my program crashes when crossoving ( can not cast integer to boolean or boolean to integer).

if think the bug comes from CrossoverOperator.doCrossover:

the random index over the gene is generated two times and may choose two differents kind of gene.

// Make a distinction for ICompositeGene for the first gene.
      // ---------------------------------------------------------
      if (firstGenes[j] instanceof ICompositeGene) {
        // Randomly determine gene to be considered.
        // -----------------------------------------
        int index1 = generator.nextInt(firstGenes[j].size());
        gene1 = ( (ICompositeGene) firstGenes[j]).geneAt(index1);
      }
      else {
        gene1 = firstGenes[j];
      }
      // Make a distinction for the second gene if CompositeGene.
      // --------------------------------------------------------
      if (secondGenes[j] instanceof ICompositeGene) {
        // Randomly determine gene to be considered.
        // -----------------------------------------
        int index2 = generator.nextInt(secondGenes[j].size());
        gene2 = ( (ICompositeGene) secondGenes[j]).geneAt(index2);
      }
      else {
        gene2 = secondGenes[j];
      }

I think the code should be:

// Make a distinction for ICompositeGene for the first gene.

      int index = 0;
      // ---------------------------------------------------------
      if (firstGenes[j] instanceof ICompositeGene) {
        // -----------------------------------------
        // Randomly determine gene to be considered.
        index = generator.nextInt(firstGenes[j].size()); //Assuming the two genes have the same size
        gene1 = ( (ICompositeGene) firstGenes[j]).geneAt(index);
      }
      else {
        gene1 = firstGenes[j];
      }
      // Make a distinction for the second gene if CompositeGene.
      // --------------------------------------------------------
      if (secondGenes[j] instanceof ICompositeGene) {
        gene2 = ( (ICompositeGene) secondGenes[j]).geneAt(index);
      }
      else {
        gene2 = secondGenes[j];
      }

Regards.

Seb

------------------------------------------------------------------------------
Come build with us! The BlackBerry® Developer Conference in SF, CA
is the only developer event you need to attend this year. Jumpstart your
developing skills, take BlackBerry mobile applications to market and stay
ahead of the curve. Join us from November 9-12, 2009. Register now!
http://p.sf.net/sfu/devconf
_______________________________________________
jgap-users mailing list
jgap-users@...
https://lists.sourceforge.net/lists/listinfo/jgap-users

From forum: jgap-users

Problems using the bulk fitness function facility

2009-09-25T17:22:48Z

For my particular problem, I need access to the whole population before
I can determine the fitness of an individual, and so I thought that the
easiest approach would be to implement a BulkFitnessFunction. I had
trouble getting it working, did some debugging, and decided that either
I have misunderstood the intention behind BulkFitnessFunction, or there
is a bug in GABreeder.

After a quick search of this users mailing list, I saw that I am not the
only one (See: Mike Lechner <mike.lechner@gm...>, July 22, 2009,
BulkFitFn doesn't mutate or breed?). This thread did not seem to come
to a resolution.

Attached is the simpest example that I could think of. So simple that
it does not actually need to use a bulk fitness function, but it shows
the problem. The population consists of two individuals, with one
integer gene each, and a bulk fitness function that simply uses the
value of the gene as its fitness value. If you run it as it is, the two
individuals at the end of the 100 iterations are the same as the two you
start with.

If you uncomment the line where I set the breeder to be my breeder then,
unless you are really unlucky, at least one of the individuals should
have been replaced with a "fitter" one at some stage during the 100
iterations.

Cheers,
David Kemp

package kemp.jgapexample;

import java.util.List;

import org.jgap.BulkFitnessFunction;
import org.jgap.Chromosome;
import org.jgap.Configuration;
import org.jgap.Gene;
import org.jgap.Genotype;
import org.jgap.IChromosome;
import org.jgap.Population;
import org.jgap.impl.DefaultConfiguration;
import org.jgap.impl.GABreeder;
import org.jgap.impl.IntegerGene;

public class ExampleWithBulkFitnessFunction {

public static void main(String[] args) throws Exception {
Configuration conf = new DefaultConfiguration();
// conf.setBreeder(new MyBreeder());
conf.setBulkFitnessFunction(new MyFitnessFunction());
conf.setPopulationSize(2);
conf.setSampleChromosome(new Chromosome(conf, new Gene[] {new IntegerGene(conf, 0, 1000)}));
Genotype population = Genotype.randomInitialGenotype(conf);
evolve(population);
}

private static void evolve(Genotype population) {
for (int i = 0; i < 100; i++) {
population.evolve();
List<IChromosome> chromosomes = population.getPopulation().getChromosomes();
System.out.println(String.format("%s, %s",
chromosomes.get(0).getGenes()[0].getAllele(),
chromosomes.get(1).getGenes()[0].getAllele()));
}
}

private static class MyFitnessFunction extends BulkFitnessFunction {
@Override
public void evaluate(Population population) {
List<IChromosome> chromosomes = population.getChromosomes();
for (IChromosome chromosome : chromosomes) {
chromosome.setFitnessValue((Integer) ((IntegerGene) chromosome.getGenes()[0]).getAllele());
}
}
}

private static class MyBreeder extends GABreeder {
@Override
protected void updateChromosomes(Population a_pop, Configuration a_conf) {
int currentPopSize = a_pop.size();
// Ensure all chromosomes are updated.
// -----------------------------------
BulkFitnessFunction bulkFunction = a_conf.getBulkFitnessFunction();
boolean bulkFitFunc = (bulkFunction != null);
if (bulkFitFunc) {
bulkFunction.evaluate(a_pop);
} else {
for (int i = 0; i < currentPopSize; i++) {
IChromosome chrom = a_pop.getChromosome(i);
chrom.getFitnessValue();
}
}
}
}
}

------------------------------------------------------------------------------
Come build with us! The BlackBerry® Developer Conference in SF, CA
is the only developer event you need to attend this year. Jumpstart your
developing skills, take BlackBerry mobile applications to market and stay
ahead of the curve. Join us from November 9-12, 2009. Register now!
http://p.sf.net/sfu/devconf
_______________________________________________
jgap-users mailing list
jgap-users@...
https://lists.sourceforge.net/lists/listinfo/jgap-users

From forum: jgap-users

MathProblem - Multiple X (input) values?

2009-09-17T22:37:35Z

Hi,

I'm working with the MathProblem example, but I can't figure out how to have several input variables -- for example instead of just "X", have two variables X1, X2 such that Y = X1 * X2. I added the variable vx2 to the nodeSets but when I print out the optimal solution, it just contains "X1" and not "X2". Please advise.

CommandGene[][] nodeSets = { {
// We use a variable that can be set in the fitness function.
// ----------------------------------------------------------
vx1 = Variable.create(conf, "X1", CommandGene.FloatClass),
vx2 = Variable.create(conf, "X2", CommandGene.FloatClass),
....

Thank you!!

From forum: jgap-users

Re: marshalling GPPopulation

2009-09-15T08:01:22Z

Simone,

the main cause is that either result is null or the GP configuration of
result. Please ensure that result is not null when writing to a file.
I would also suggest to not append to the file in this case.

Normally, you should not marhal the GP configuration as it could become
quite lengthy as XML, maybe a circular reference could occur. Better would
be rebuilding the GP config. After unmarshalling a GPProgram and setting the
config. to it.

Best

Klaus
www.klaus-meffert.com

> -----Original Message-----
> From: Simone Bronzin [mailto:simone.bronzin@...]
> Sent: Tuesday, September 15, 2009 11:55 AM
> To: Klaus Meffert
> Cc: jgap-users@...
> Subject: RE: [jgap-users] marshalling GPPopulation
>
> Klaus,
>
> thank you for your reply. I got an error. Running this
>
> ...
> XStream xstream = new XStream();
>
> PrintWriter outFile = new PrintWriter(new
> FileOutputStream("gpprogram.xml", true));
>
> String xml = xstream.toXML(gp.getAllTimeBest());
>
> outFile.print(xml);
> outFile.close();
>
> File f = new File("gpprogram.xml");
> InputStream oi = new FileInputStream(f);
> GPProgram result = (GPProgram) xstream.fromXML(oi);
>
> double gpFitness = fitness.evaluate(gp);
>
> a NullPointerException is raised at the line
>
> protected double evaluate(final IGPProgram gpProgram) { ...
> gpProgram.getGPConfiguration().clearMemory(); // HERE ...
> }
>
>
> it seems that GPConfiguration.m_memory is not correctly set
> by the unmarshaling process, and remains null. Before diving
> into xstream documentation I was wondering if I missed something.
>
> Thank you,
>
> Simone
>
>
> Def. Quota Klaus Meffert <jgap@...>:
>
> > Simone,
> >
> > You could use Xstream, which is includes with JGAP. E.g.:
> >
> > //To read an object from file
> > XStream xstream = new XStream();
> > File f = new File(a_filename);
> > InputStream oi = new FileInputStream(f);
> > GPGenotype result = (GPGenotype) xstream.fromXML(oi);
> >
> > // To write, use xstream.toXML
> >
> > Or use xstream.marshal(..) and xstream.unmarshal --> also see
> > Internet-documentation
> >
> > Best
> >
> > Klaus
> > www.klaus-meffert.com
> >
> >
> >
> >> -----Original Message-----
> >> From: Simone Bronzin [mailto:simone.bronzin@...]
> >> Sent: Friday, September 11, 2009 11:30 AM
> >> To: jgap-users@...
> >> Subject: [jgap-users] marshalling GPPopulation
> >>
> >> Hi all,
> >>
> >> Is there any easy way to marshal and unmarshal a GPPopulation?
> >>
> >> thank you,
> >>
> >> Simone Bronzin
> >> Unitary
> >> Executive
> >>
> >> mobile: +39.3939459672
> >>
> >>
> >>
> >> --------------------------------------------------------------
> >> ----------------
> >> Let Crystal Reports handle the reporting - Free Crystal
> Reports 2008
> >> 30-Day trial. Simplify your report design, integration and
> deployment
> >> - and focus on what you do best, core application coding. Discover
> >> what's new with Crystal Reports now.
> http://p.sf.net/sfu/bobj-july
> >> _______________________________________________
> >> jgap-users mailing list
> >> jgap-users@...
> >> https://lists.sourceforge.net/lists/listinfo/jgap-users
> >
> >
>
>
>
> Simone Bronzin
> Unitary
> Executive
>
> mobile: +39.3939459672
> phone: +39.
>

From forum: jgap-users

Re: marshalling GPPopulation

2009-09-12T06:37:25Z

Simone,

You could use Xstream, which is includes with JGAP. E.g.:

//To read an object from file
XStream xstream = new XStream();
File f = new File(a_filename);
InputStream oi = new FileInputStream(f);
GPGenotype result = (GPGenotype) xstream.fromXML(oi);

// To write, use xstream.toXML

Or use xstream.marshal(..) and xstream.unmarshal --> also see
Internet-documentation

Best

Klaus
www.klaus-meffert.com

> -----Original Message-----
> From: Simone Bronzin [mailto:simone.bronzin@...]
> Sent: Friday, September 11, 2009 11:30 AM
> To: jgap-users@...
> Subject: [jgap-users] marshalling GPPopulation
>
> Hi all,
>
> Is there any easy way to marshal and unmarshal a GPPopulation?
>
> thank you,
>
> Simone Bronzin
> Unitary
> Executive
>
> mobile: +39.3939459672
>
>
>
> --------------------------------------------------------------
> ----------------
> Let Crystal Reports handle the reporting - Free Crystal
> Reports 2008 30-Day trial. Simplify your report design,
> integration and deployment - and focus on what you do best,
> core application coding. Discover what's new with Crystal
> Reports now. http://p.sf.net/sfu/bobj-july
> _______________________________________________
> jgap-users mailing list
> jgap-users@...
> https://lists.sourceforge.net/lists/listinfo/jgap-users

From forum: jgap-users

marshalling GPPopulation

2009-09-11T02:30:10Z

Hi all,

Is there any easy way to marshal and unmarshal a GPPopulation?

thank you,

Simone Bronzin
Unitary
Executive

mobile: +39.3939459672

------------------------------------------------------------------------------
Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
trial. Simplify your report design, integration and deployment - and focus on
what you do best, core application coding. Discover what's new with
Crystal Reports now. http://p.sf.net/sfu/bobj-july
_______________________________________________
jgap-users mailing list
jgap-users@...
https://lists.sourceforge.net/lists/listinfo/jgap-users

From forum: jgap-users