Hi! I've just released version 2.1.4 of Open BEAGLE. this version is mainly a minor features addition, with support for weighted GP primitives sets, HFC...
Hello all, Since the release of Distributed BEAGLE 0.8 one month ago, more than 50 persons downloaded it. We hope you are enjoying Distributed BEAGLE. We are...
Hi. I'm planning on using OpenBeagle to implement several Neural Network training schemes. I want to start with a simple feedforward net, fixed network...
... I am not aware that anyone use Open BEAGLE to evolve weights of neural nets. But as far as you have a good NN C++ class, it might not be very difficult to...
Hello, I would like to define my own PG operators and I wonder how individuals, called in mutate(Beagle::Individual& ioIndividual, Beagle::Context& ioContext) ...
Hi Ewa, ... The indpb parameter indicates the probability of each individual to be choosen for mutation in a population. The selection of mutated is random,...
Hi Tim, ... It's both. To differentiate between an evolver and a cruncher, Distributed BEAGLE use configuration files. For all examples, they are located in...
Hi Ewa, This seems a weird problem for me. First be sure to scope the arguments of MyCrossoverOp::operate to the good deme and context types, i.e. virtual void...
Hi everyone! I just released version 2.1.5 of Open BEAGLE. This new version comes with some small improvements and several bug fixes compared to previous ...
Hi Christian, I'm having a problem with some evolutionary runs and would like to know some per-generation statistics on the use of each primitive. In each run...
Hello, Is it possible to add range constraints to real valued parameters? I have two parameters (x, y), which have the constraints 0 <= x <= 90 10 <= y <= 90 ...
Hello all, I'm experiencing uneven load balancing with dbeagle 0.8. I'm using a GA to generate parameters which control a geometrical model, the evaluation of...
Hi David, The questions you ask will surely help making a better manual. ... I'm not sure if I understand clearly. Tell me if that's what you think is ...
Marc, ... It's somehow nice to know that my knowledge gaps are helping someone ;) [snip] ... while ... lies ... My problem setup: pe-evolver.conf (all other...
Hi, sorry for the late answer, I'm quite in a rush right now. [snip] ... One of the cruncher should get ALL available jobs, since only 6 jobs needs to be...
Hi Daniel, Which gcc version are you using? What is exactly your Linux distro/version? christian ... -- ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ~> Christian...
Hello Christian, I am using a derivative of Knoppix that is Debian based. My gcc version is 3.3.4. Thanks! Daniel. ... What is exactly your Linux ... ...
... This is quite strange, it compiles well on my Linux box (gcc 3.3.2 - Fedora Core 1). It seems not to find the header for std::random_shuffle. Try adding ...
... Hi Christian, I took a diferent route to your suggestion, and went through each node of every tree of every individual. I thought this would allow me to ...
... Yes that sound good. I would name the method "haveValue" instead, as I think it is better to use the infinitive form of verbs in method naming. Otherwize...
Hello everyone, The Open BEAGLE team is proud to announce the joint release of Distributed BEAGLE 0.9 and BEAGLE Visualizer 0.2. ... Distributed BEAGLE is an...
... You can seed your population with individuals. To do so, you must craft a XML file with the list of individuals in the tags "Seeds", like in the following:...
... Yes, it is possible. To do so, you need to use an explicit generational evolver configuration. For an exemple of this, see file 'symbreg-generational.conf'...
... First, please send your e-mails directly to the list (openbeagle@yahoogroups.com), not to the list owners (openbeagle-owner@yahoogroups.com) as you did....
Hi Christian, Previously I worked on an extension to the Beagle::GP::Individual class that stored genealogical information. I'm thinking about updating that ...
Having problems with message search?
Fill out this form to ensure your group is one of the first to be migrated to the new message search system.
