| Re: Random selection of pairs [message #51244] |
Sun, 12 November 2006 04:55 |
![Julio[1] is currently offline](theme/default/images/xoffline.png.pagespeed.ic.XRkd1fkXye.png "Julio[1] is currently offline")
Julio[1]
Messages: 52
Registered: May 2005
|
Member |
|
|
Thanks guys... it will help.
Cheers,
Julio
Michael Galloy escreveu:
> Kenneth P. Bowman wrote:
>> It will be much simpler, and probably much quicker, to read all
>> of the pairs and then pick 100 randomly, like this script named
>> select_random:
>>
>> x = findgen(2, 1000) ;create fake lon-lat data
>> r = RANDOMU(seed, 1000) ;generate random numbers
>> j = (SORT(r))[0:9] ;find indices of 10 smallest numbers
>
> Yes, I think this is the way to do it.
>
> It won't help here with only 1000 elements, but I have a routine to get
> the indices of the n smallest (or largest) values in an array. It uses
> HISTOGRAM and REVERSE_INDICES instead of the SORT in the line
>
> j = (SORT(r))[0:9]
>
> The article describing it is at:
>
> http://michaelgalloy.com/2006/06/02/finding-the-n-smallest-e lements-in-an-array.html
>
>
> Mike
> --
> www.michaelgalloy.com
|
|
|
|
| Re: Random selection of pairs [message #51246 is a reply to message #51244] |
Sat, 11 November 2006 20:30  |
Michael Galloy
Messages: 1114
Registered: April 2006
|
Senior Member |
|
|
Kenneth P. Bowman wrote:
> It will be much simpler, and probably much quicker, to read all
> of the pairs and then pick 100 randomly, like this script named
> select_random:
>
> x = findgen(2, 1000) ;create fake lon-lat data
> r = RANDOMU(seed, 1000) ;generate random numbers
> j = (SORT(r))[0:9] ;find indices of 10 smallest numbers
Yes, I think this is the way to do it.
It won't help here with only 1000 elements, but I have a routine to get
the indices of the n smallest (or largest) values in an array. It uses
HISTOGRAM and REVERSE_INDICES instead of the SORT in the line
j = (SORT(r))[0:9]
The article describing it is at:
http://michaelgalloy.com/2006/06/02/finding-the-n-smallest-e lements-in-an-array.html
Mike
--
www.michaelgalloy.com
|
|
|
|
| Re: Random selection of pairs [message #51252 is a reply to message #51246] |
Sat, 11 November 2006 06:58 |
Kenneth P. Bowman
Messages: 585
Registered: May 2000
|
Senior Member |
|
|
In article <1163249745.126148.50070@h48g2000cwc.googlegroups.com>,
"Julio" <julio@cpa.unicamp.br> wrote:
> I have an array with two columns and a thousand of rows. Each row has a
> pair of latitude and longitude.
>
> Array=float(2, 1000)
>
> -27.3456 -54.6529
> -23.4546 -56.7263
> ... and so on...
>
> I need to retrieve only 100 pairs, randomly. How can I do that?
> Any comments welcome!
> Julio
It will be much simpler, and probably much quicker, to read all
of the pairs and then pick 100 randomly, like this script named
select_random:
x = findgen(2, 1000) ;create fake lon-lat data
r = RANDOMU(seed, 1000) ;generate random numbers
j = (SORT(r))[0:9] ;find indices of 10 smallest numbers
PRINT, j
PRINT, r[j]
FOR k = 0, 9 DO PRINT, x[*,j[k]] ;print 10 random pairs
IDL> @select_random
952 848 497 935 490 678 853 313
365 976
0.000964510 0.00113783 0.00255171 0.00340872 0.00390364 0.00452789 0.00609691
0.00676597 0.0117514 0.0121868
1904.00 1905.00
1696.00 1697.00
994.000 995.000
1870.00 1871.00
980.000 981.000
1356.00 1357.00
1706.00 1707.00
626.000 627.000
730.000 731.000
1952.00 1953.00
Ken Bowman
|
|
|
|
comp.lang.idl-pvwave archive
Members
Search
Help
Login
Home
