On Apr 25, 2012, at 4:22 PM, Yun Tao wrote: > I'm trying to generate the diffusion pattern of a bivariate_normal. However, > so far a main obstacle is to form the distribution on top of a 2D mesh. The > bivariate_normal function from matplotlib.mlab takes axial arguments that are > spit out by numpy.meshgrid. But it's clear that the underlying mesh > construction of meshgrid is different from Grid2D from fipy:
There are several examples that show how to obtain the coordinates of the cell centers, e.g., http://www.ctcms.nist.gov/fipy/examples/diffusion/generated/examples.diffusion.anisotropy.html?highlight=getcellcenters http://www.ctcms.nist.gov/fipy/examples/diffusion/generated/examples.diffusion.circle.html?highlight=getcellcenters http://www.ctcms.nist.gov/fipy/examples/levelSet/generated/examples.levelSet.advection.circle.html?highlight=getcellcenters http://www.ctcms.nist.gov/fipy/examples/phase/generated/examples.phase.anisotropy.html?highlight=getcellcenters The only difference I can see is that np.meshgrid() returns a shaped result, where as FiPy's .getCellCenters() just returns lists of numbers. That's just fine; matplotlib.mlab.bivariate_normal() will return z in the same shape as X and Y and FiPy wants its values unshaped. _______________________________________________ fipy mailing list [email protected] http://www.ctcms.nist.gov/fipy [ NIST internal ONLY: https://email.nist.gov/mailman/listinfo/fipy ]
