Hi Edwin, See this,
<http://matdl-osi.org/fipy/ticket/129> If you need to use the gist viewer, try creating a new viewer everytime you want to plot. So plot like this, Viewer(variable).plot() i.e. create a new viewer everytime you plot. That will overcome the memory issue. Cheers On Nov 22, 2007 10:27 AM, R. Edwin Garcia <[EMAIL PROTECTED]> wrote: > > Hi guys, > > I have a similar problem. I ran a couple of nights ago one of the KWC models > for grain growth and coarsening, (Acta Materialia 51 (2003) 6035–6058) , and > while the initial memory usage is 300MB (500x500 mesh), after one hour it > had used 2.1GB of my available 3GB. I was quite surprised because the code > was only looping through the solver and updating the viewers. The python > script crashed eventually because running out of RAM. > > I was wondering if there were any known leaks in the libraries, or if there > was a way to force the garbage collector to free any of the temporary > variables, or a quick fix because the problem is going to worsen when I > decide to run it in 3D... I will attach an example (somewhat simplified) > script in the next email. > > Thanks, > Edwin > > > > On Nov 21, 2007, at 11:55 AM, joelvuillon wrote: > > Hi everybody, > > Using Fipy for solving elasticity problems, I have noticed a loss of RAM > using fipy/tools/numerix library and, in particular, the numerix.MAtake > function when indices is a masked array. In addition, MA.take does not work > in such a case (cf documentation on fipy.tools). > > Do you have already experienced such loss of RAM problem ? > > Is there a convenient mean to avoid such a problem while keeping indices as > a masked array ? > > Bests, > > Joël > > -- Daniel Wheeler
