comp.lang.idl-pvwave archive: archive » Re: hist

Home » Public Forums » archive » Re: hist_nd question

Show: Today's Messages :: Show Polls :: Message Navigator
E-mail to friend

Re: hist_nd question [message #52931 is a reply to message #52930]

Mon, 05 March 2007 11:16

JD Smith
Messages: 850
Registered: December 1999

Senior Member

On Mon, 05 Mar 2007 16:17:09 +0100, Wox wrote:

> Hi all,
>
> I have been using JD's hist_nd before, but I just found something strange.
> Maybe this is already corrected somewhere or maybe this is a "feature",
> but N-dimensional points falling outside of the given min/max boundries,
> are mapped to the first and the last bin. To clarify:

Good catch. The problem is HIST_ND assumes that the 1D indices are
unique, such that things outside the bounds of one axis "wrap into"
the next dimension. This is not a problem if you let the code derive
the N-dimensional bounding box itself, but if you hand specify a range
with MIN/MAX, you can feel this issue. HIST_2D solves this by brute
force setting to -1 any location which falls outside the bounds. I've
modified HIST_ND to do something similar. Please find (and test) the
updated version here:

turtle.as.arizona.edu/idl/hist_nd.pro

This version requires IDL v6.1 or later.

An interesting issue arose while making these changes. Comparing the
newly slimmed-down loop-based version of the single index creation:

h=long((V[s[0]-1,*]-mn[s[0]-1])/bs[s[0]-1])
for i=s[0]-2,0,-1 do h=nbins[i]*h + long((V[i,*]-mn[i])/bs[i])

to the fully threaded, loop-free version:

h=total(long((V-rebin(mn,s,/SAMP))/rebin(bs,s,/SAMP)) * $
rebin([1L,product(nbins[0:s[0]-2],/CUMULATIVE,/PRESERVE_TYPE )], $
s,/SAMP), 1,/PRESERVE_TYPE)

I found the former to perform better by 20% or more for large arrays,
which results purely from the overhead of REBIN'ing small arrays 3
times, to the size of the input array (e.g. 3 x 5 million for a large
array), with it's resulting greater demands on memory. As is often
pointed out, small loops which perform lots of work per iteration do
not feel the loop penalty, and depending on the exact memory
requirements, can actually excel over loop-free methods (yes, you hear
me saying this).

JD

Report message to a moderator

[Message index]

		Re: hist_nd question By: Craig Markwardt on Wed, 07 March 2007 06:17
		Re: hist_nd question By: Wox on Tue, 06 March 2007 01:42
		Re: hist_nd question By: JD Smith on Mon, 05 March 2007 12:49
		Re: hist_nd question By: Kenneth Bowman on Mon, 05 March 2007 11:59
		Re: hist_nd question By: David Fanning on Mon, 05 March 2007 11:35
		Re: hist_nd question By: JD Smith on Mon, 05 March 2007 11:16
		Re: hist_nd question By: Wox on Mon, 05 March 2007 08:05
		Re: hist_nd question By: Wox on Mon, 05 March 2007 07:42

Previous Topic:	Checking for perl
Next Topic:	Re: Checking for perl

-=] Back to Top [=-

[ Syndicate this forum (XML) ] [

] [

]

Current Time: Wed Oct 15 16:28:51 PDT 2025

Total time taken to generate the page: 1.28246 seconds