faster Array.take( floatindices.astype(int) ) ?
Folks, is there a way to speed up Array.take( floatindices.astype(int) ) ? astype(int) makes a copy, floor() returns floats. (Is there a wiki of NumPy one-liners / various tricks ? would sure beat googling.) Thanks, cheers -- denis
On Fri, Feb 1, 2013 at 6:24 AM, denis <denis-bz-gg@t-online.de> wrote:
Folks, is there a way to speed up Array.take( floatindices.astype(int) ) ? astype(int) makes a copy, floor() returns floats.
(Is there a wiki of NumPy one-liners / various tricks ? would sure beat googling.)
You can use floats for the indices in the take method, they are floored before use. That said, the Fortran FLOOR function returns integers and it would be useful if numpy had integer versions of floor, ceil, divmod, say, ifloor, iceil, idivmod that raised errors if the resulting integers overflowed. Chuck
Charles R Harris <charlesr.harris <at> gmail.com> writes:
You can use floats for the indices in the take method, they are floored before use.
Chuck, that's in 1.7 ? In 1.6.2 x = np.arange(10) x.take([3.14]) array([3]) x.take(np.array([3.14])) TypeError: array cannot be safely cast to required type cheers -- denis
participants (2)
-
Charles R Harris -
denis