On Tue, Aug 21, 2012 at 5:45 PM, Fausto Arinos de A. Barbuto <
[email protected]> wrote:

>
>
>
>
> I have pysparse 64-bit whose binary (installer) I downloaded from
> Christopher Gohlke's useful
> website at http://www.lfd.uci.edu/~gohlke/pythonlibs/  Chris makes
> himself very clear when he
> writes that all his 64-bit installers are "unofficial".  But his binaries
> always worked on previous
> FiPy installations.  I even dare to say that I would never have FiPy
> installed on this W7 box if
> it weren't for his binaries.
>

That site looks really good. Thanks for sharing that.


> As for the fipy.test('--scipy') output, I really don't know what to do
> with it.  It has 2300+ lines
> and contains lots of information.  Apparently, its first lines are
> similar, if not equal, to the ones
> produced by regular test.  I zipped the lines and attached it to this
> note, should you wish to
> give it a look.
>

I gave it a look as it's throwing lots of these

   File "C:\Program
Files\Python\lib\site-packages\fipy-3.0-py2.7.egg\fipy\tools\numerix.py",
line 521, in nearest
        nearestIndices[chunk] = argmin(tmp, axis=0)
    TypeError: array cannot be safely cast to required type

and errors like this

      print _FaceGradContributions(v0).globalValue.shape
  Expected:
      (2, 24)
  Got:
      (2L, 24L)

Also it seems that fipy is still the pysparse solver. I think the
fipy.test('--scipy') syntax doesn't quite work as intended. Anyway, try
this.

   $ set FIPY_SOLVERS=scipy
   $ python
   >>> import fipy
   >>> fipy.test()

That should rid you of the matrix populating errors assuming scipy is
installed. You may still have some of the other errors, but they are
nothing to be concerned about.

Cheers



-- 
Daniel Wheeler
_______________________________________________
fipy mailing list
[email protected]
http://www.ctcms.nist.gov/fipy
  [ NIST internal ONLY: https://email.nist.gov/mailman/listinfo/fipy ]

Reply via email to