|
From: | Dmitri A. Sergatskov |
Subject: | False alarm (Re: reshape slowdown?) |
Date: | Tue, 22 Feb 2005 10:48:21 -0700 |
User-agent: | Mozilla Thunderbird 0.9 (X11/20041127) |
John W. Eaton wrote:
On 22-Feb-2005, David Bateman <address@hidden> wrote:| I did change the reshape function between 2.1.57 and 2.1.58 for NDArrays | as you say. If it is two times slower for 2D arrays, we should probably | special case the 2D version and use the old code... Want to propose a | change?Or at least some test cases so we can gather some data about the performance? What size matrices does it really affect, etc.?
I am sorry -- the reshape really does not take any time. I overlookedthat I have a=b' in the loop that does go slower than before, but that, I think, relates to the different lapack/blas version. It is still somewhat weird, as most other lapack functions perform the same or even faster than before...
jwe
Dmitri. -- ------------------------------------------------------------- Octave is freely available under the terms of the GNU GPL. Octave's home on the web: http://www.octave.org How to fund new projects: http://www.octave.org/funding.html Subscription information: http://www.octave.org/archive.html -------------------------------------------------------------
[Prev in Thread] | Current Thread | [Next in Thread] |