| Re: uniq slow for loop [message #67263] |
Fri, 17 July 2009 08:59 |
newerjazz
Messages: 5
Registered: July 2009
|
Junior Member |
|
|
Thanks Ben
I rewrote using histogram so as to eliminate calling where inside for
loop. Program is running now. Am sire it will be faster. Any comments
on how to improve further is appreciated.
h=histogram(data(8,*),binsize=1)
hmax=size(h)
hmax=hmax(1)-1
FOR j=0, hmax DO BEGIN
blah=mean(data(*,i:i+h(j)-1))
etc,etc
i=h(j)+i
ENDFOR
On Jul 16, 6:40 pm, "ben.bighair" <ben.bigh...@gmail.com> wrote:
> On Jul 16, 5:52 pm, newerjazz <shang...@gmail.com> wrote:
>
>
>
>> Hi All,
>
>> I have an array of [9,1,000,000]. In one of the columns are sorted
>> numbers (eg. 0,1,1,2,3,3,3,3,3,4,4,5,5,5,6,7,7,7,etc). I want to pick
>> out all the rows where column 9 is equal to some number and average
>> all the quantities in those rows.
>
>> Right now I use a for loop and the where function to pick out these
>> rows. This is of course very slow. Anyway to increase the speed of
>> this would be greatly appreciated. Thank you very much!
>
>> Cheers,
>> newerjazz
>
>> nmax = max(data(8,*))
>> FOR j=0, nmax DO BEGIN
>> w=where(data(8,*) eq j, nw)
>> blah=mean(data(*,w))
>> etc,etc
>> ENDFOR
>
> Hi,
>
> I am pretty sure that you want a variant of histogram. I think you
> might be able to harness something from the following discussion.
>
> http://tinyurl.com/nggcx6
>
> Cheers,
> Ben
|
|
|
|
| Re: uniq slow for loop [message #67269 is a reply to message #67263] |
Thu, 16 July 2009 18:40  |
ben.bighair
Messages: 221
Registered: April 2007
|
Senior Member |
|
|
On Jul 16, 5:52 pm, newerjazz <shang...@gmail.com> wrote:
> Hi All,
>
> I have an array of [9,1,000,000]. In one of the columns are sorted
> numbers (eg. 0,1,1,2,3,3,3,3,3,4,4,5,5,5,6,7,7,7,etc). I want to pick
> out all the rows where column 9 is equal to some number and average
> all the quantities in those rows.
>
> Right now I use a for loop and the where function to pick out these
> rows. This is of course very slow. Anyway to increase the speed of
> this would be greatly appreciated. Thank you very much!
>
> Cheers,
> newerjazz
>
> nmax = max(data(8,*))
> FOR j=0, nmax DO BEGIN
> w=where(data(8,*) eq j, nw)
> blah=mean(data(*,w))
> etc,etc
> ENDFOR
Hi,
I am pretty sure that you want a variant of histogram. I think you
might be able to harness something from the following discussion.
http://tinyurl.com/nggcx6
Cheers,
Ben
|
|
|
|
comp.lang.idl-pvwave archive
Members
Search
Help
Login
Home
