You don't need fipy at all to use lsmlib. It's just convenient for plotting stuff. Just pass numpy arrays to lsmlib, reshape and plot the results with matplotlib's contourf. No need for fipy at all. In fact just remove all fipy imports and go from there.
On Thu, Feb 23, 2012 at 12:06 PM, wang yunbo <[email protected]> wrote: > Hi Dr. Wheeler, > > After making those changes, the codes works. However, the graph generated > is weird and different from that generated with fipy trunk. > > Hence I decide to use trunk. Is there any manual in PDF I can refer to? > > I tried to build the manual by python setup.py build_docs --pdf and also > using --html. But I am getting errors from sphinx, so I was wondering if > there is a manual you could provide me. > > > Best, > > Yunbo > > On Feb 20, 2012, at 4:43 PM, Daniel Wheeler wrote: > > > > On Mon, Feb 20, 2012 at 1:38 PM, wang yunbo <[email protected]> wrote: > >> >> Traceback (most recent call last): >> File "circle.py", line 11, in <module> >> phi[:] = (m.x - L / 2)**2 + (m.y - L / 2)**2 - 0.25**2 >> AttributeError: UniformGrid2D instance has no attribute 'x' >> > > > Use > > x, y = m.getCellCenters() > > instead of m.x and m.y and it should work. The changes to get it ti work > with 2.1.3 are trivial so I wouldn't update your version of fipy, > > > -- > Daniel Wheeler > > > -- Daniel Wheeler
_______________________________________________ fipy mailing list [email protected] http://www.ctcms.nist.gov/fipy [ NIST internal ONLY: https://email.nist.gov/mailman/listinfo/fipy ]
