[Numpy-discussion] numpy.ma.flatnotmasked_contiguous error in slice.stop values?

Geordie McBain gdmcbain at freeshell.org
Sun Aug 8 23:54:32 EDT 2010


2010/8/9 Pierre GM <pgmdevlist at gmail.com>:
> OK, mind opening a ticket ? I'll try to take care of that tmw and I don't want the bug report to fall between the cracks.

Done. http://projects.scipy.org/numpy/ticket/1576.  Thank you for your
very prompt and helpful response.  I wasn't aware of the ticket system
previously.

> Mmh, tuples are nicer than slices... We'll see.

O.K.  I'm fairly new to NumPy and Matplotlib, so I'll leave the design
decision to the experts, but I did find

segments = np.ma.flatnotmasked_contiguous (am)
print am[segments[0]], am[segments[1]]

easier than

segments1 = matplotlib.mlab.contiguous_regions (am)
print am[slice (*segments1[0])], am[slice (*segments1[1])]

but perhaps there are more elegant ways of dealing with tuples
representing slices?

> print numpy.version.version

Thanks.



More information about the NumPy-Discussion mailing list