Want to read Slashdot from your mobile device? Point it at m.slashdot.org and keep reading!

 


Forgot your password?
Close
typodupeerror
×
Programming IT Technology

Build Your Own Neural Network 53

Posted by michael from the skynet dept.
windowpain writes "I just discovered Joone. It's an LGPL neural net development environment for creating, training and testing neural networks. The aim is to create a powerful environment both for enthusiasts and professional users, based on the newest Java technologies. Joone is composed by a central engine that is the fulcrum of all applications that already exist or will be developed. It's available in Linux, MacOS and Windows versions."
This discussion has been archived. No new comments can be posted.

Build Your Own Neural Network More

Build Your Own Neural Network

Comments Filter:

  • Whats wrong with SNNS (Score:5, Interesting)

    by Directrix1 ( 157787 ) on Thursday September 18, 2003 @01:27PM (#6996129)
    The Stuttguart Neural Network Simulator has been available for free for a long time now.

    • Re:Whats wrong with SNNS (Score:5, Informative)

      by KMAPSRULE ( 639889 ) on Thursday September 18, 2003 @02:16PM (#6996516)
      From the SNNS licensing terms:
      SNNSv4.1 is available NOW free of charge for research purposes under a...

      LGPL(Joone's Licensing) is a VERY attractive License for commercial users, "free for research purposes" would preclude some from being able to use it,
      other wise I agree this Joone seems to have less features/algorithm support than SNNS

  • No Thanks (Score:3, Funny)

    by Sunlighter ( 177996 ) on Thursday September 18, 2003 @01:48PM (#6996300)

    I don't have time to play with such a thing. I'm too busy developing the neural network in my skull.

  • Very limited number of algorithms (Score:5, Interesting)

    by a_ghostwheel ( 699776 ) on Thursday September 18, 2003 @01:48PM (#6996304)
    Looks rather limited to me. It's again implements only BP algorithms and their variants. Why not include ART-based (adaptive resonance theory) and LVQ-based (linear vector quantization) algorithms? They are much more efficient than BP in many instances (e.g. classification problems).

    In other words, sounds very limited to me.

    • Re:Very limited number of algorithms (Score:1, Funny)

      by Anonymous Coward
      I have no idea what you just said but I'll trust that you know what you are talking about..

      Remember the Norns from Creatures ? What kind of neural networks should I be reading about to replicate that kind of learning ?

    • Re:Very limited number of algorithms (Score:4, Insightful)

      by elmegil ( 12001 ) on Thursday September 18, 2003 @03:24PM (#6997125) Homepage Journal
      It's open source. Scratch your itch.

      BTW implementing ART would be really cool. I always had trouble getting the full understanding of Grossberg's papers and didn't find many simulators that did ART 10 years ago when I was actually looking at all this stuff.

      • From my experience, one of the problems is that nobody really seems to agree on exactly how to interpret his models; especially ART2 seems to be something nobody really bothers with in practice. I have seen a couple of pretty complete implementations, but they differ pretty wildly on the results you get. Rather, people seem to take his conceptual ideas and incorporating the ones that make sense into their own models.

    • why? because you whine instead of work (Score:1, Funny)

      by Anonymous Coward
      why help a project when you can whine instead! way to go slashdot fool!

  • Can also be combined with genetic algorithms (Score:5, Informative)

    by mrblah ( 229865 ) on Thursday September 18, 2003 @02:28PM (#6996612)
    For those who want to combine genetic algorithms with neural networks, there's also a project [sourceforge.net] that combines joone with JGAP [sourceforge.net] (a Java genetic algorithms framework).

    • Re:Can also be combined with genetic algorithms (Score:2, Informative)

      by Anonymous Coward
      "For those who want to combine genetic algorithms with neural networks, there's also a project that combines joone with JGAP (a Java genetic algorithms framework)."

      Lots of neuroevolution code, including probably the state-of-the-art algorithms (ESP, NEAT) can be downloaded from the NN research Web site at UT-Austin [utexas.edu]. Some of it is available in both C++ and Java. Papers describing the algorithms and various applications can be downloaded from the same site.

  • Anyone remember a language dedicated to NNs? (Score:3, Interesting)

    by RevAaron ( 125240 ) <revaaron AT hotmail DOT com> on Thursday September 18, 2003 @02:46PM (#6996761) Homepage
    Low chance anyone here will know what I was talking about, but in the case someone does ...

    Does anyone remember a programming language that was specifically for creating, training and using neural networks?

    I've always been a big programming language addict. Back in the early 90s, being 12-14 years old and excited to finally have a modem, I was downloading every programming language for DOS I could find from all the BBSes I called.

    I can't remember what it was called. I remember roughly what the IDE looked like, but very little else. It was a fun system- it had general-purpsose programming constructs, but was especially built for creating, training and using neural networks. I seem to recall the syntax having a semi-familiar pascal/algol/C-ish syntax; it wasn't just a library for Lisp or Scheme.

    I've check the SimTel archives, and haven't been able to find it again. Oh, how awesome were BBSes... Stronger sense of community than the 'internet' seems to have. Anywho, thanks for any tips!

    • Someone remembers Ralph!?! (Score:4, Interesting)

      by MarkusQ ( 450076 ) on Friday September 19, 2003 @01:00AM (#7001278) Journal

      Hard as it is for me to believe, you may be talking about a language I wrote in the late 80's. "Ralph" was our internal name for it (inside joke); I think the market droids pushed it as NNSim or something like that. We released a full function version on a bunch of BBSs & talked it up on geni, compuserve, news groups, etc. to promote a hardware accelerator board (DSP based). The idea was people would get interested and then (as their models got larger) they'd want more speed and buy our accelerator board.

      The core language was a based on pascal, but with salient structure (like python) and a bunch of (at the time) interesting extensions. You could declare networks and treat them like an array (for messing with the weights) or like a procedure (for training) or like a function (for using them).

      Does this sound like what you're remembering?

      -- MarkusQ

      P.S. In case you can't guess how the story ends, it turned out that for really interesting networks you'd need a lot more oomph than our boards could provide. The product died, as did several others, and the company sank beneath the waves.

      • Holy balls, man! I've asked this question a dozen times in a dozen different places- I never thought I'd find a possible answer!

        It's been a long time, around 10 years, since I've seen this project. I didn't use it a ton when I did have it on DOS, and consequently don't have a really clear picture of it in my head. However, what you've described soudns like what I remember. Hell yeah!

        I don't suppose you have a copy of the app/system/language aroudn anywhere, do you? I don't own any DOS or Windows mac

        • I found a partial copy from 1989--no idea if it was developmental or stable (but given that it was on my machine I know which way I'd guess). It at least seems to come up under dosemu. If you e-mail me ("MarkusQLreality.com" at: 8 put: "@") I can send you what I've got.

          -- MarkusQ

          P.S. IIRC (based on where I found it), this is a chance survivor of a my-hard-drive-is-dying snapshot that got carried over for all this time because harddrives get larger & cheaper faster than I can type. So don't plan on

  • skynet dept? (Score:2, Funny)

    by mraymer ( 516227 )
    Let's hope we don't make a neural net CPU like SKYNET, because, let's be honest here, SKYNET is pretty dumb:

    1) It has an obvious preference for the dramatic... For example, rather than having the original Terminator simply walk over and kill Sarah Connor when she is stuck in an overturned vehicle, it instead has the Terminator commandeer a semi and attempt to run her over. Brilliant!

    2) It doesn't believe in rushing into a kill as quickly as cyborg-ly possible, but instead has its Terminators advance sl

    • Re:skynet dept? (Score:2, Offtopic)

      by NanoGator ( 522640 )
      "So, let us hope that if we ever built a SKYNET, it wouldn't be stupid enough to conquer time, and then have no idea what the next logical step is. "

      I'd prefer Skynet over the Matrix. Those idiots adopted of the least efficient batteries ever.

    • Re:skynet dept? (Score:1, Funny)

      by Anonymous Coward
      Don't you get it? SkyNet is'nt some vast computer system that uses AI technology for defense purposes. SkyNet is George Bush's secret plan for taking over the world. Here's the plan :
      - Arnie will rule California later this year.
      - A "terrorist" strike from North Korea will strike LA in the same year
      - Arnie will rip off his suit revealing his sleek tight fitting leather terminator outfit underneath
      - Pilotless UAVs will start to bomb LA
      well you've seen the movies you can figure out the rest. Don't say I did'nt
  • True, it's not free (beer or speech), but this [nd.com] really is the most impressive NN development platform I've ever worked with. It has a VB-like interface for creating a custom network, and powerful wizards to handle the common problems.

    Plus, the only limit on their demo is that you can only save a networks structure, but not the trained weights or output. This means you can use the demo to determine if a problem is solvable through a NN, and only get your company to buy it if it works for your project.

  • Another kind of solution. (Score:3, Interesting)

    by DoctorRad ( 608319 ) on Thursday September 18, 2003 @03:35PM (#6997207)
    Self-Organising Maps can be used to solve many similar problems to those for which ANNs are appropriate. Check out the SOMPAK [cis.hut.fi] software. A shame my own research into data visualisation using this technique is company confidential to an ex-employer, some very pretty pictures :-)

    The package inlcudes source code to produce Sammon Maps in Postscript format. These can be very useful tools for finding clusters in data. What they revealed about UK higher education institutes was eye opening.

    Matt...
    --
    A man hears what he wants to hear and disregards the rest.

    • SOMs.... Such fond memories. I too wrangled with one of these a few years back. In my case it was being used to correlate documents. I created a method to group similar nodes which I dubbed "Amoeba Love Fest". It seemed to be working fairly well right up to the point we lost VC and they nuked half the company.

      We should compare notes sometime.

  • Write once run everywhere ? (Score:3, Funny)

    by DrSkwid ( 118965 ) on Thursday September 18, 2003 @04:13PM (#6997549) Journal
    Linux, MacOS and Windows versions

    hmm, almost everywhere

  • This is clearly an adertisement by the author (Score:1, Funny)

    by Anonymous Coward
    I just discovered Joone. It's an LGPL neural net development environment for creating, training and testing neural networks. The aim is to create a powerful environment both for enthusiasts and professional users, based on the newest Java technologies. Joone is composed by a central engine that is the fulcrum of all applications that already exist or will be developed. It's available in Linux, MacOS and Windows versions.

    I've hilighted some terms to help prove my theory. ;)

    So there you have it:

    An amazing

  • Java? (Score:2, Funny)

    by Anonymous Coward
    Let's see...we'll be using a virtual machine to emulate a virtual network of simulated neurons...no thanks!

    I can understand the desire to have portability, but this just reeks. It's like using basic to emulate a 486. The part of me that appreciates elegant, efficient design is puking on the floor.
    • The part of me that appreciates a good hack is thinking of running this under UML :-)

      Is there a version of Joone that runs on a Beowulf cluster? Then I could create a virtual Beowulf cluster of a bunch of UMLs and use it to run "a virtual machine to emulate a virtual network of simulated neurons" -- Cool!

  • Java is way too slow for serious NN code (Score:2, Informative)

    by Anonymous Coward

    My current neural network implementation (on generic CPU) with order of 10^6 neurons use CBLAS ATLAS fast linear algebra library as a number crunching part. When you have to do some down to metal optimizations Jave is waste of time.

    The best property of NN's are when compared to other learning machines is that computation can be parallelized easily. For small problems one should maybe use other ML machines (SVM doesn't overfit) which isn't as easily/well parallelized (-> doesn't scale up so well).

  • Shameless plug (Score:3, Informative)

    by JanneM ( 7445 ) on Friday September 19, 2003 @08:26AM (#7002655) Homepage
    If you want a simulation framework that is more flexible than a dedicated NN simulator, we are developing Ikaros [lucs.lu.se], a discrete-time modular simulator. Runs on Linux, OSX and Windows. Implement your modules in C or C++ (and implementing bindings for other languages would be easy), then specify a connection matrix for inputs and outputs between modules to form a complete simulation.

    The next step in development is creating some graphical visualizing tools, and to make it run multiple instances transparently across a network.

Slashdot Top Deals

Never test for an error condition you don't know how to handle. -- Steinbach

Close