July 21, 2010
6:36 p.m.
On Wed, Jul 21, 2010 at 7:25 PM, David Warde-Farley <dwf@cs.toronto.edu> wrote:
On 2010-07-20, at 10:16 PM, Skipper Seabold wrote:
Out of curiosity, is there an explicit way to check if these share memory?
You could do the exact calculations (I think) but this isn't actually implemented in NumPy, though np.may_share_memory is a conservative test for it that will err on the side of false positives.
Ah, that's good to know. In [5]: np.may_share_memory(b,c) Out[5]: False In [6]: a = b.view() In [7]: np.may_share_memory(a,b) Out[7]: True In [8]: np.may_share_memory(a,c) Out[8]: False