On Mon, May 21, 2012 at 7:32 PM, Thouis (Ray) Jones wrote:
> I submitted a PR for this functionality after cleaning it up a bit:
> https://github.com/numpy/numpy/pull/284
I meant to ask here (and Travis reminded me in the PR):
Currently, the trace_data_allocations() function (the only one added
I submitted a PR for this functionality after cleaning it up a bit:
https://github.com/numpy/numpy/pull/284
I've attached an example that produces HTML for a a sortable table
tracking allocations while running through numpy.test().
Ray Jones
sorttable.js
Description: JavaScript source
track_a
On Thu, May 17, 2012 at 9:52 PM, Nathaniel Smith wrote:
> I'd be tempted to just see if I could get by with massif or another
> "real" heap profiler -- unfortunately the ones I know are C oriented,
> but might still be useful...
I got some very useful information from Fabien's technique, which le
On Thu, May 17, 2012 at 7:50 PM, Stéfan van der Walt wrote:
> On Wed, May 16, 2012 at 12:34 PM, Thouis Jones wrote:
>> I wondered, however, if there were a better way to accomplish the same
>> goal, preferably in pure python.
>
> Fabien recently posted this; not sure if it addresses your use case
On Wed, May 16, 2012 at 12:34 PM, Thouis Jones wrote:
> I wondered, however, if there were a better way to accomplish the same
> goal, preferably in pure python.
Fabien recently posted this; not sure if it addresses your use case:
http://fseoane.net/blog/2012/line-by-line-report-of-memory-usage/
I recently had need of tracing numpy data allocation/deallocation. I
was unable to find a simple way to do so, and so ended up putting the
code below into ndarraytypes.h to allow me to trace allocations. A
key part is that this jumps back into python, so I can inspect the
stack and find out where
