Quantcast

Duplicated genes in a solution/individual

classic Classic list List threaded Threaded
4 messages Options
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Duplicated genes in a solution/individual

Bernardo Pereira Nunes
Hi All,

I would like to create solutions without duplicated genes.

For example:

When the set of solutions are created I get the following individuals:
x + x + y + z or y + y + x or z + z + y + y ...

However, the problem that I am addressing will never have duplicated
genes. Thus, I would like to constrain the creation of the individuals
to obtain only indivuals without repeated genes. For example:
x + y + z or x + y or z + y

Is there a way to configure JGap to do not use a gene that was already
used in the creation of one individual?

Thanks!
Bernardo




------------------------------------------------------------------------------
Everyone hates slow websites. So do we.
Make your web apps faster with AppDynamics
Download AppDynamics Lite for free today:
http://ad.doubleclick.net/clk;258768047;13503038;j?
http://info.appdynamics.com/FreeJavaPerformanceDownload.html
_______________________________________________
jgap-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/jgap-users
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Duplicated genes in a solution/individual

Mauricio Noda
Try setting "doublette chromosomes allowed" to false in natural selectors:

http://jgap.sourceforge.net/javadoc/3.6/org/jgap/NaturalSelectorExt.html#setDoubletteChromosomesAllowed%28boolean%29


> Date: Wed, 19 Sep 2012 18:16:16 -0300

> From: [hidden email]
> To: [hidden email]
> Subject: [jgap-users] Duplicated genes in a solution/individual
>
> Hi All,
>
> I would like to create solutions without duplicated genes.
>
> For example:
>
> When the set of solutions are created I get the following individuals:
> x + x + y + z or y + y + x or z + z + y + y ...
>
> However, the problem that I am addressing will never have duplicated
> genes. Thus, I would like to constrain the creation of the individuals
> to obtain only indivuals without repeated genes. For example:
> x + y + z or x + y or z + y
>
> Is there a way to configure JGap to do not use a gene that was already
> used in the creation of one individual?
>
> Thanks!
> Bernardo
>
>
>
>
> ------------------------------------------------------------------------------
> Everyone hates slow websites. So do we.
> Make your web apps faster with AppDynamics
> Download AppDynamics Lite for free today:
> http://ad.doubleclick.net/clk;258768047;13503038;j?
> http://info.appdynamics.com/FreeJavaPerformanceDownload.html
> _______________________________________________
> jgap-users mailing list
> [hidden email]
> https://lists.sourceforge.net/lists/listinfo/jgap-users

------------------------------------------------------------------------------
Live Security Virtual Conference
Exclusive live event will cover all the ways today's security and
threat landscape has changed and how IT managers can respond. Discussions
will include endpoint security, mobile security and the latest in malware
threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/
_______________________________________________
jgap-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/jgap-users
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Duplicated genes in a solution/individual

Bernardo Pereira Nunes
Hi Mauricio,

Thanks for your answer. I have already tried your suggestion, but it is not what I really want, although it helps too. What I mean is that I dont want to have a duplicated gene in a solution/chromossome.

For instance, imagine that I have the following genes: a, b and c

Possibly, the JGAP can give me a solution like: f(a) = a + a, but as you can see (a) is duplicated. Thus, I want to constrain my chromossome to accept singletons genes. In this manner, I would expect that a possible solution should look like f(a) = a. Do you know how to constrain the number of times that a gene can appear in a solution? In my case, I want to constrain to one appearance.

Thanks!
Bernardo

Em 21/09/2012 17:23, Mauricio Noda escreveu:
Try setting "doublette chromosomes allowed" to false in natural selectors:

http://jgap.sourceforge.net/javadoc/3.6/org/jgap/NaturalSelectorExt.html#setDoubletteChromosomesAllowed%28boolean%29


> Date: Wed, 19 Sep 2012 18:16:16 -0300
> From: [hidden email]
> To: [hidden email]
> Subject: [jgap-users] Duplicated genes in a solution/individual
>
> Hi All,
>
> I would like to create solutions without duplicated genes.
>
> For example:
>
> When the set of solutions are created I get the following individuals:
> x + x + y + z or y + y + x or z + z + y + y ...
>
> However, the problem that I am addressing will never have duplicated
> genes. Thus, I would like to constrain the creation of the individuals
> to obtain only indivuals without repeated genes. For example:
> x + y + z or x + y or z + y
>
> Is there a way to configure JGap to do not use a gene that was already
> used in the creation of one individual?
>
> Thanks!
> Bernardo
>
>
>
>
> ------------------------------------------------------------------------------
> Everyone hates slow websites. So do we.
> Make your web apps faster with AppDynamics
> Download AppDynamics Lite for free today:
> http://ad.doubleclick.net/clk;258768047;13503038;j?
> http://info.appdynamics.com/FreeJavaPerformanceDownload.html
> _______________________________________________
> jgap-users mailing list
> [hidden email]
> https://lists.sourceforge.net/lists/listinfo/jgap-users

--
Untitled Document

Bernardo Pereira Nunes
Tecnologia da Informação
CCEAD PUC-RIO

Telefones: (21) 3527-1454/1455 - Ramal: 222
Rua Marquês de São Vicente, 225 - Ed. Padre Leonel Franca, 2o andar
Gávea - Rio de Janeiro - RJ - CEP 22453-900


------------------------------------------------------------------------------
Live Security Virtual Conference
Exclusive live event will cover all the ways today's security and
threat landscape has changed and how IT managers can respond. Discussions
will include endpoint security, mobile security and the latest in malware
threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/
_______________________________________________
jgap-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/jgap-users
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Duplicated genes in a solution/individual

Mauricio Noda
Untitled Document
There are supergenes, which is a validation check over generated chromosomes. Invalid chromosomes are deleted from the pool.
 
You can create custom genetic operators which avoid duplicated genes.
 
Or design the chromosome in a way that it never have duplicated variables. i.e. the chromosome have 5 genes A, B, C, D, E... and the function is: f(a,b,c,d,e) = Aa + Bb + Cc + Dd + Ee
 
Or... simply allow duplicated variables, and then you simplify the function afterwards using another non-genetic solution.
 
Sent: Monday, September 24, 2012 8:34 AM
Subject: Re: [jgap-users] Duplicated genes in a solution/individual
 
Hi Mauricio,

Thanks for your answer. I have already tried your suggestion, but it is not what I really want, although it helps too. What I mean is that I dont want to have a duplicated gene in a solution/chromossome.

For instance, imagine that I have the following genes: a, b and c

Possibly, the JGAP can give me a solution like: f(a) = a + a, but as you can see (a) is duplicated. Thus, I want to constrain my chromossome to accept singletons genes. In this manner, I would expect that a possible solution should look like f(a) = a. Do you know how to constrain the number of times that a gene can appear in a solution? In my case, I want to constrain to one appearance.

Thanks!
Bernardo

Em 21/09/2012 17:23, Mauricio Noda escreveu:
Try setting "doublette chromosomes allowed" to false in natural selectors:

http://jgap.sourceforge.net/javadoc/3.6/org/jgap/NaturalSelectorExt.html#setDoubletteChromosomesAllowed%28boolean%29


> Date: Wed, 19 Sep 2012 18:16:16 -0300

> From: [hidden email]
> To: [hidden email]
> Subject: [jgap-users] Duplicated genes in a solution/individual
>
> Hi All,
>
> I would like to create solutions without duplicated genes.
>
> For example:
>
> When the set of solutions are created I get the following individuals:
> x + x + y + z or y + y + x or z + z + y + y ...
>
> However, the problem that I am addressing will never have duplicated
> genes. Thus, I would like to constrain the creation of the individuals
> to obtain only indivuals without repeated genes. For example:
> x + y + z or x + y or z + y
>
> Is there a way to configure JGap to do not use a gene that was already
> used in the creation of one individual?
>
> Thanks!
> Bernardo
>
>
>
>
> ------------------------------------------------------------------------------
> Everyone hates slow websites. So do we.
> Make your web apps faster with AppDynamics
> Download AppDynamics Lite for free today:
> http://ad.doubleclick.net/clk;258768047;13503038;j?
> http://info.appdynamics.com/FreeJavaPerformanceDownload.html
> _______________________________________________
> jgap-users mailing list
> [hidden email]
> https://lists.sourceforge.net/lists/listinfo/jgap-users

--

Bernardo Pereira Nunes
Tecnologia da Informação
CCEAD PUC-RIO

Telefones: (21) 3527-1454/1455 - Ramal: 222
Rua Marquês de São Vicente, 225 - Ed. Padre Leonel Franca, 2o andar
Gávea - Rio de Janeiro - RJ - CEP 22453-900


------------------------------------------------------------------------------
Live Security Virtual Conference
Exclusive live event will cover all the ways today's security and
threat landscape has changed and how IT managers can respond. Discussions
will include endpoint security, mobile security and the latest in malware
threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/
_______________________________________________
jgap-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/jgap-users
Loading...