Loading ...
Sorry, an error occurred while loading the content.

JAGA ? ? ?

Expand Messages
  • Aditya
    Hello all I am an under-graduate IT student trying out GAs for the first time. While searching on the net, I came across this site -- www.jaga.org . JAGA
    Message 1 of 6 , Nov 15, 2004
    • 0 Attachment
      Hello all

      I am an under-graduate IT student trying out GAs for the first time. While searching on the net, I came across this site -- www.jaga.org . JAGA stands for Java Api for Genetic Algorithms. We all are aware of the power of JAVA. Myself being more familiar with JAVA would prefer to work in JAVA than say C or C++. I had two questions in mind-

      1. Why does the researchers prefer C or C++ over JAVA for working on GAs. (Atleast the search results indicated so.)

      2. Has anyone tested these APIs for GAs? Do they follow the standards?

      Any advices for a novice will be extremely appreciated.

      Aditya Sharma B. Tech (ICT) DAIICT

      Gandhinagar - India


      ---------------------------------
      ALL-NEW Yahoo! Messenger - all new features - even more fun!

      [Non-text portions of this message have been removed]
    • woodsc@f00f.net
      1) I m not sure if people do prefer C or C++ over java for GAs. A lot of people use lisp for GAs. I did some work with the scripting language ruby on GPs.
      Message 2 of 6 , Nov 18, 2004
      • 0 Attachment
        1) I'm not sure if people do prefer C or C++ over java for GAs. A lot of
        people use lisp for GAs. I did some work with the scripting language
        ruby on GPs. Speedwise, there isn't much difference between java and C
        so I'd work with whichever is easiest for you :)

        2) I haven't tried any particular API's for GA's, but I don't think
        there is a standard for it. Its probably best to understand what the
        code is doing by reading through some of it to make sure its compatible
        with what you want to do.

        On Mon, Nov 15, 2004 at 11:26:43AM +0000, Aditya wrote:
        >
        >
        >
        > Hello all
        >
        > I am an under-graduate IT student trying out GAs for the first time. While searching on the net, I came across this site -- www.jaga.org . JAGA stands for Java Api for Genetic Algorithms. We all are aware of the power of JAVA. Myself being more familiar with JAVA would prefer to work in JAVA than say C or C++. I had two questions in mind-
        >
        > 1. Why does the researchers prefer C or C++ over JAVA for working on GAs. (Atleast the search results indicated so.)
        >
        > 2. Has anyone tested these APIs for GAs? Do they follow the standards?
        >
        > Any advices for a novice will be extremely appreciated.
        >
        > Aditya Sharma B. Tech (ICT) DAIICT
        >
        > Gandhinagar - India
        >
        >
        > ---------------------------------
        > ALL-NEW Yahoo! Messenger - all new features - even more fun!
        >
        > [Non-text portions of this message have been removed]
        >
        >
        >
        >
        >
        >
        >
        >
        > Yahoo! Groups Links
        >
        >
        >
        >
        >
        >
      • Greg Paperin
        Hi, I was leading the development of JAGA, so I think I can comment on it quite well)). We are not aware of some sort of GA-platform standard, our guidelines
        Message 3 of 6 , Nov 19, 2004
        • 0 Attachment
          Hi,

          I was leading the development of JAGA, so I think I can comment on it quite
          well)).

          We are not aware of some sort of GA-platform standard, our guidelines during
          the specification were as follows:
          - provide an extendable and flexible core API so that us or others can
          implement extensions or particular algorithms easily
          - provide a standard set of widely used algorithms, and encodings with
          appropriate operators
          - add encodings and operators used for our own experiments.

          The reason we decided to create JAGA is because we were not able to find any
          good Java GA platforms which would be extensible enough for our needs. There
          are several good platforms, but they are mostly in C, C++ or Lisp.
          We don't know why that is, but I personally think that the reason is that Java
          has been a really widely used language for only the last 6 or 7 years, so only
          researchers who started their undergraduate degrees after 1998 or so really
          "grew up" with Java. More experienced researchers might feel more accustomed
          with older languages, so they use them.
          We hope, JAGA will become the new generation experimentation platform.

          At present, the work on JAGA has been suspended since this summer, after a
          first final release was made, because the key contributors are busy with other
          projects, but we will resume the work soon.

          Please, feel free to get in touch if you have any questions or need any help.
          If you prefer to contact us directly rather than posting to this mailing list,
          write to gp@....

          Best Regards,

          Greg.

          -------------------------------------------------------------------------------
          Gregory Paperin
          Research Fellow of Computer Science
          University College London
          Gower Street
          London WC1E 6BT
          United Kingdom
          http://www.paperin.org


          > -----Original Message-----
          > From: Aditya [mailto:adi_sun3@...]
          > Sent: Monday, November 15, 2004 11:27 AM
          > To: genetic_programming@yahoogroups.com
          > Subject: [GP] JAGA ? ? ?
          >
          >
          >
          >
          > Hello all
          >
          > I am an under-graduate IT student trying out GAs for the
          > first time. While searching on the net, I came across this
          > site -- www.jaga.org . JAGA stands for Java Api for Genetic
          > Algorithms. We all are aware of the power of JAVA. Myself
          > being more familiar with JAVA would prefer to work in JAVA
          > than say C or C++. I had two questions in mind-
          >
          > 1. Why does the researchers prefer C or C++ over JAVA for
          > working on GAs. (Atleast the search results indicated so.)
          >
          > 2. Has anyone tested these APIs for GAs? Do they follow the
          > standards?
          >
          > Any advices for a novice will be extremely appreciated.
          >
          > Aditya Sharma
          > B. Tech (ICT)
          > DAIICT
          >
          > Gandhinagar - India
          >
          >
          > ---------------------------------
          > ALL-NEW Yahoo! Messenger - all new features - even more fun!
          >
          > [Non-text portions of this message have been removed]
          >
          >
          >
          >
          >
          >
          > ------------------------ Yahoo! Groups Sponsor
          > --------------------~-->
          > $9.95 domain names from Yahoo!. Register anything.
          > http://us.click.yahoo.com/J8kdrA/y20IAA/yQLSAA/7brrlB/TM
          > --------------------------------------------------------------
          > ------~->
          >
          >
          > Yahoo! Groups Links
          >
          >
          >
          >
          >
          >
          >
          >
          >
          >
        • Sean Luke
          You all KNOW this would bring me out. :-) ... Wow, ECJ was not extensible enough for your needs? What did you need? ... Some Java EC toolkits you might
          Message 4 of 6 , Nov 21, 2004
          • 0 Attachment
            You all KNOW this would bring me out. :-)


            On Nov 19, 2004, at 10:46 AM, Greg Paperin wrote:

            > The reason we decided to create JAGA is because we were not able to
            > find any
            > good Java GA platforms which would be extensible enough for our needs.

            Wow, ECJ was not extensible enough for your needs? What did you need?


            > There are several good platforms, but they are mostly in C, C++ or
            > Lisp.

            Some Java EC toolkits you might examine:

            ECJ
            JEO/DREAM
            ECKit (ultralightweight)
            JDEAL
            EVALife's EA Library
            GA Playground (GAA)


            > We don't know why that is, but I personally think that the reason is
            > that Java
            > has been a really widely used language for only the last 6 or 7 years,
            > so only
            > researchers who started their undergraduate degrees after 1998 or so
            > really
            > "grew up" with Java. More experienced researchers might feel more
            > accustomed
            > with older languages, so they use them. We hope, JAGA will become the
            > new
            > generation experimentation platform.

            This is an odd comment indeed to make in a mailing list where several
            Java toolkits are among the very most popular. If I were to make a
            guess, I'd say over 80% of its readership is very experienced with
            Java, and at least a third of the readership uses Java as their primary
            research language.

            Sean, who'd like to know from everyone what stupidities should be
            excised from ECJ to make it a better system. [oh, btw, we're working
            on an applet, a GUI front-end, and a client/server evaluator]
          • Greg Paperin
            Hi, When we started on JAGA, we did not know about ECJ. At that time everyone involved was a student and we were not perfectly informed about the available
            Message 5 of 6 , Nov 21, 2004
            • 0 Attachment
              Hi,

              When we started on JAGA, we did not know about ECJ. At that time everyone
              involved was a student and we were not perfectly informed about the available
              options. We looked hard though before deciding to "reinvent the wheel", we
              found several APIs but none of those was good enough. While I do not know about
              the present situation, back then ECJ was not very well promoted on the web.

              Since then JAGA was extended and used for advanced research; I do believe that
              JAGA is superior to some of the APIs you stated and maybe not superior to
              others, but I would like to avoid picking on specific projects. Of course, I
              might be wrong altogether)) .

              I also believe that I was quite clear that the explanation I suggested for Java
              GA/GP APIs not to be available/easy to find was just a guess.
              But if Sean believes that it is so odd, I would be happy to be told better
              explanation.
              Fact is nonetheless that Aditya Sharma's original mail confirms that Java APIs
              are still hard to find (whether they don't exist or don't show up in Google is
              a different question).

              Finally, in hope that this is not going to become an argument about "the best
              API", I d like to say that I really think that in this sort of areas, there is
              never a perfect solution. One API will be better suited for certain situations
              or programmer-styles, another will be better suited for others. Only being
              experienced with all of them (or being lucky) can make sure that you take the
              best choice. ))

              Cheers,

              Greg.


              -------------------------------------------------------------------------------
              Gregory Paperin
              Research Fellow of Computer Science
              University College London
              Gower Street
              London WC1E 6BT
              United Kingdom
              http://www.paperin.org




              > -----Original Message-----
              > From: Sean Luke [mailto:sean@...]
              > Sent: Sunday, November 21, 2004 6:08 PM
              > To: genetic_programming@yahoogroups.com
              > Subject: Re: [GP] JAGA ? ? ?
              >
              >
              > You all KNOW this would bring me out. :-)
              >
              >
              > On Nov 19, 2004, at 10:46 AM, Greg Paperin wrote:
              >
              > > The reason we decided to create JAGA is because we were not able to
              > > find any good Java GA platforms which would be extensible
              > enough for
              > > our needs.
              >
              > Wow, ECJ was not extensible enough for your needs? What did you need?
              >
              >
              > > There are several good platforms, but they are mostly in C, C++ or
              > > Lisp.
              >
              > Some Java EC toolkits you might examine:
              >
              > ECJ
              > JEO/DREAM
              > ECKit (ultralightweight)
              > JDEAL
              > EVALife's EA Library
              > GA Playground (GAA)
              >
              >
              > > We don't know why that is, but I personally think that the
              > reason is
              > > that Java has been a really widely used language for only
              > the last 6
              > > or 7 years, so only researchers who started their undergraduate
              > > degrees after 1998 or so really "grew up" with Java. More
              > experienced
              > > researchers might feel more accustomed with older
              > languages, so they
              > > use them. We hope, JAGA will become the new generation
              > experimentation
              > > platform.
              >
              > This is an odd comment indeed to make in a mailing list where
              > several Java toolkits are among the very most popular. If I
              > were to make a guess, I'd say over 80% of its readership is
              > very experienced with Java, and at least a third of the
              > readership uses Java as their primary research language.
              >
              > Sean, who'd like to know from everyone what stupidities
              > should be excised from ECJ to make it a better system. [oh,
              > btw, we're working on an applet, a GUI front-end, and a
              > client/server evaluator]
              >
              >
              >
              > ------------------------ Yahoo! Groups Sponsor
              > --------------------~--> Make a clean sweep of pop-up ads.
              > Yahoo! Companion Toolbar.
              > Now with Pop-Up Blocker. Get it for free!
              > http://us.click.yahoo.com/L5YrjA/eSIIAA/yQLSAA/7brrlB/TM
              > --------------------------------------------------------------
              > ------~->
              >
              >
              > Yahoo! Groups Links
              >
              >
              >
              >
              >
              >
              >
              >
              >
              >
            • samuel sharaf
              hi aditya, good questions, first, most of the programs are written in C/C++ because most of the engineering course work in usa is still being done in either C
              Message 6 of 6 , Dec 30, 2004
              • 0 Attachment
                hi aditya, good questions, first, most of the programs are written in C/C++ because most of the engineering course work in usa is still being done in either C or C++. Engineers and scientists working on GA algo's find it easier to code in C/C++.
                however, i am one of the engineers who have realized that java is much simpler to code than other low level languages. i haven't tried jaga.org, it looks interesting though.

                are u doing CS majors ?

                regards,
                Sam

                Aditya <adi_sun3@...> wrote:


                Hello all

                I am an under-graduate IT student trying out GAs for the first time. While searching on the net, I came across this site -- www.jaga.org . JAGA stands for Java Api for Genetic Algorithms. We all are aware of the power of JAVA. Myself being more familiar with JAVA would prefer to work in JAVA than say C or C++. I had two questions in mind-

                1. Why does the researchers prefer C or C++ over JAVA for working on GAs. (Atleast the search results indicated so.)

                2. Has anyone tested these APIs for GAs? Do they follow the standards?

                Any advices for a novice will be extremely appreciated.

                Aditya Sharma B. Tech (ICT) DAIICT

                Gandhinagar - India


                ---------------------------------
                ALL-NEW Yahoo! Messenger - all new features - even more fun!

                [Non-text portions of this message have been removed]





                Yahoo! Groups Sponsor
                Get unlimited calls to

                U.S./Canada


                ---------------------------------
                Yahoo! Groups Links

                To visit your group on the web, go to:
                http://groups.yahoo.com/group/genetic_programming/

                To unsubscribe from this group, send an email to:
                genetic_programming-unsubscribe@yahoogroups.com

                Your use of Yahoo! Groups is subject to the Yahoo! Terms of Service.



                [Non-text portions of this message have been removed]
              Your message has been successfully submitted and would be delivered to recipients shortly.