Iterative numerical optimizaiton routines?

Collapse
This topic is closed.
X
X
 
  • Time
  • Show
Clear All
new posts
  • Anthony Calabria

    Iterative numerical optimizaiton routines?

    Is anyone aware of iterative numerical optimzaiton routines in Java
    and/or Python? I'm looking for a routine similar to the fmincon in
    Matlab, powell in IDL, solver in Excel. There appears to be an
    unconstrained optimization routine in the book "A Numerical Library in
    Java for Scientists and Engineers" although I'd prefer a constrained
    routine. Any comments on the book would be helpful too since I'll
    probably buy it shortly.

    Thanks,
    Anthony
  • Robin Becker

    #2
    Re: Iterative numerical optimizaiton routines?

    In article <e0dbd886.03102 61817.4d3820ea@ posting.google. com>, Anthony
    Calabria <anthonycalabri a@alum.rit.edu> writes[color=blue]
    >Is anyone aware of iterative numerical optimzaiton routines in Java
    >and/or Python? I'm looking for a routine similar to the fmincon in
    >Matlab, powell in IDL, solver in Excel. There appears to be an
    >unconstraine d optimization routine in the book "A Numerical Library in
    >Java for Scientists and Engineers" although I'd prefer a constrained
    >routine. Any comments on the book would be helpful too since I'll
    >probably buy it shortly.
    >
    >Thanks,
    >Anthony[/color]
    There is Aaron Watters LP and netflow optimisation on Python Starship
    (user arw I beleieve). There are also a Nelder Mead algorithm and others
    in Python SciPy I think. There is a java LP at Washington university
    somewhere google for 'cassowary'.
    --
    Robin Becker

    Comment

    • Frithiof Andreas Jensen

      #3
      Re: Iterative numerical optimizaiton routines?


      "Anthony Calabria" <anthonycalabri a@alum.rit.edu> wrote in message
      news:e0dbd886.0 310261817.4d382 0ea@posting.goo gle.com...[color=blue]
      > Is anyone aware of iterative numerical optimzaiton routines in Java
      > and/or Python?[/color]

      SciPy contains many numeric optimisation algorithms - not written in Python,
      but as extention modules - unfortunately SciPy is still Python 2.2.


      Comment

      • Cousin Stanley

        #4
        Re: Iterative numerical optimizaiton routines?

        | SciPy contains many numeric optimisation algorithms
        | - not written in Python, but as extention modules
        | - unfortunately SciPy is still Python 2.2.

        The Enthought Python 2.3 distribution contains SciPy ....
        At Enthought, we power digital transformation for science to deliver new business value and faster scientific discovery.


        --
        Cousin Stanley
        Human Being
        Phoenix, Arizona

        Comment

        • Frithiof Andreas Jensen

          #5
          Re: Iterative numerical optimizaiton routines?


          "Cousin Stanley" <CousinStanley@ hotmail.com> wrote in message
          news:bnjcia$126 n0l$1@ID-130333.news.uni-berlin.de...
          [color=blue]
          > The Enthought Python 2.3 distribution contains SciPy ....
          > http://www.enthought.com/[/color]

          Yes, indeed it does - but it *also* contains every science/numeric tool on
          the planet; For my part I just need the SciPy tools not an entirely new
          universe ;-)


          Comment

          • Robert Kern

            #6
            Re: Iterative numerical optimizaiton routines?

            Frithiof Andreas Jensen <frithiof.jense n@removethis.te d.ericsson.dk> wrote:
            [color=blue]
            > "Cousin Stanley" <CousinStanley@ hotmail.com> wrote in message
            > news:bnjcia$126 n0l$1@ID-130333.news.uni-berlin.de...
            >[color=green]
            >> The Enthought Python 2.3 distribution contains SciPy ....
            >> http://www.enthought.com/[/color]
            >
            > Yes, indeed it does - but it *also* contains every science/numeric tool on
            > the planet; For my part I just need the SciPy tools not an entirely new
            > universe ;-)[/color]

            I think his point was that SciPy does indeed run on Python 2.3, as I can
            happily confirm.

            :-)

            --
            Robert Kern
            kern@ugcs.calte ch.edu

            "In the fields of Hell where the grass grows high
            Are the graves of dreams allowed to die."
            -- Richard Harter

            Comment

            • Frithiof Andreas Jensen

              #7
              Re: Iterative numerical optimizaiton routines?


              "Robert Kern" <kern@ugcs.calt ech.edu> wrote in message
              news:bnmjco$ok5 $1@naig.caltech .edu...
              [color=blue]
              > I think his point was that SciPy does indeed run on Python 2.3, as I can
              > happily confirm.[/color]

              Do Tell - How did you get it to install? On my distribution known as:

              "Python 2.3 (#46, Jul 29 2003, 18:54:32) [MSC v.1200 32 bit (Intel)] on
              win32" the installer refuses to run, rudely demanding Python 2.2.

              PS: I am disinclined to build from source on Windows because I have *enough*
              company-mandated crud on my win32 box already - Linux would of course be a
              different thing; but I haven't got a Linux box at the moment.


              Comment

              • Robert Kern

                #8
                Re: Iterative numerical optimizaiton routines?

                In article <bnoi68$men$1@n ewstree.wise.ed t.ericsson.se> you wrote:
                [color=blue]
                > "Robert Kern" <kern@ugcs.calt ech.edu> wrote in message
                > news:bnmjco$ok5 $1@naig.caltech .edu...
                >[color=green]
                >> I think his point was that SciPy does indeed run on Python 2.3, as I can
                >> happily confirm.[/color]
                >
                > Do Tell - How did you get it to install? On my distribution known as:
                >
                > "Python 2.3 (#46, Jul 29 2003, 18:54:32) [MSC v.1200 32 bit (Intel)] on
                > win32" the installer refuses to run, rudely demanding Python 2.2.[/color]

                Ah, Windows binaries are another matter. The ones provided at scipy.org do
                indeed only go up to 2.2. I've only built SciPy on UN*X-type OSs.
                [color=blue]
                > PS: I am disinclined to build from source on Windows because I have *enough*
                > company-mandated crud on my win32 box already - Linux would of course be a
                > different thing; but I haven't got a Linux box at the moment.[/color]

                Your best bet then is to download the Python 2.3 Enthought Edition and rip out
                the parts you don't need. Or wait until 0.20 release of SciPy when I'm sure they
                will make 2.3 binaries. Or you could ask them if they have prerelease 2.3
                binaries (well, they obviously do for the Enthought Edition) ready to package up
                and put on the downloads page.

                --
                Robert Kern
                kern@ugcs.calte ch.edu

                "In the fields of Hell where the grass grows high
                Are the graves of dreams allowed to die."
                -- Richard Harter

                Comment

                Working...