Daniel,

Thanks for the report.

Hmm, I have not seen this before and it seems to work for me. It looks
like the python wrapper is creating an integer array the c extension
does not like. The call to PyArray_FROMANY() is failing and giving
this message. I will do some more testing and get back to you.

What version of numpy are you using and what platform are you on?

Thanks,
Jason



On Thu, May 31, 2012 at 11:13 AM, Daniel Wheeler
<[email protected]> wrote:
> On Thu, May 31, 2012 at 12:02 PM, Jason Furtney <[email protected]> wrote:
>> Daniel,
>>
>> Thanks for the reply and for your interest in scikit-fmm. Yes, I think
>> I have the extension velocities implemented correctly according to the
>> Sethian paper/book. Everything seems to be working but one issue
>> remains: I seem to have used the opposite array order (FORTRAN vs C)
>> in scikit-fmm  from the rest of numpy/scipy/fipy. I am still confused
>> on this and need to give it more thought.
>>
>> I would be pleased if you could give it a try and get back to me with
>> any problems.
>
> Thanks Jason. First problem! I installed and then this...
>
> In [1]: import skfmm
>
> In [2]: import numpy as np
>
> In [3]: phi = np.ones((3, 3))
>
> In [4]: phi[1, 1] = -1
>
> In [5]: skfmm.distance(phi)
> ---------------------------------------------------------------------------
> ValueError                                Traceback (most recent call last)
> /users/wd15/packages/<ipython-input-5-ceafd4bf2e24> in <module>()
> ----> 1 skfmm.distance(phi)
>
> /users/wd15/.virtualenvs/lsmlib/lib/python2.6/site-packages/skfmm/pfmm.pyc
> in distance(phi, dx, self_test)
>     71     """
>     72     phi, dx, flag = pre_process_args(phi, dx)
> ---> 73     d = cFastMarcher(phi, dx, flag, None, int(self_test))
>     74     d = post_process_result(d)
>     75     return d
>
> ValueError: flag must be a 1D to 12-D array of integers
>
> Any ideas?
>
>> I have a paper submission deadline today so will reply with more details 
>> soon.
>
> Good luck with that.
>
> I should let you know that I'm presenting a poster at the scipy
> conference on my experiences with lsmlib/scikit-fmm and modeling
> electrochemical deposition problems.
>
> --
> Daniel Wheeler
>
> _______________________________________________
> fipy mailing list
> [email protected]
> http://www.ctcms.nist.gov/fipy
>  [ NIST internal ONLY: https://email.nist.gov/mailman/listinfo/fipy ]



-- 
--
  Jason K. Furtney
  Itasca Consulting Group
  111 3rd Ave. South, Suite 450
  Minneapolis, MN 55401 USA
    (612) 371-4711

_______________________________________________
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