Re: Vector Calculation Q [message #7794] |
Tue, 14 January 1997 00:00 |
aspinelli
Messages: 4 Registered: August 1996
|
Junior Member |
|
|
handy@gemini.oscs.montana.edu (Brian Handy) wrote:
> Hi,
> I've got a puzzler I'd like to figure out. I want to do
> sort of a "running integral on a vector, e.g. if I'm given
> a = [1, 4, 2, 6, 3, 4, 2, 6]
> Then I can easily do
> for i=0, (size(a))(1)-1
> b(i) = total(a(0:i)))
> to get...
> b = [1, 5, 7, 13, 16, 20, 22, 28]
> ...I'd like to do this with a vector instead of the loop.
> Any ideas?
My suggestion is not sensible if you have to compute
the thing just one time. If, on the other hand,
you perform the computation several times, the following may prove
useful:
;;;;;;;;
pro init
;;;;;;;;
common globalKernel, m, n
n = 8
m = intarr( n, n )
for i=0,n-1 do begin
for j=i, n-1 do begin
m(i,j) = 1
endfor
endfor
end
;;;;;;;;;;;
pro compute
;;;;;;;;;;;
common globalKernel, m, n
a = [1, 4, 2, 6, 3, 4, 2, 6]
print, a # m
end
;;;;;;;;
pro test
;;;;;;;;
init
compute
end
;;;;;;;;;;;;;
Well, it reduces to building an upper triangular matrix of all ones in
the upper half; maybe there is some better way of initializing it
than mine!
I suspect, however, that you are not satisfied with my answer :(
Cheers
Andrea
Andrea Spinelli - Ismes Spa. 9, Via Pastrengo, 24068 Seriate BG, Italy
tel.: +39-35-307777 fax.: +39-35-302999 e-mail: aspinelli@ismes.it
"Truth hurts, but pimples much more"
|
|
|