| Re: IDL FOR Loop variable increments [message #62575] |
Mon, 22 September 2008 08:54  |
raghuram
Messages: 32
Registered: February 2008
|
Member |
|
|
On Sep 21, 3:37 pm, Raghu <raghuram.narasim...@gmail.com> wrote:
> On Sep 21, 6:44 am, Bulrush <Wasit.Weat...@gmail.com> wrote:
>
>
>
>> On Sep 20, 12:51 am, Raghu <raghuram.narasim...@gmail.com> wrote:
>
>>> On Sep 19, 12:09 pm, pgri...@gmail.com wrote:
>
>>>> R.G. Stockwell wrote:
>>>> > "Jean H" <jghas...@DELTHIS.ucalgary.ANDTHIS.ca> wrote in message
>>>> >news:gauiil$u32$1@news.ucalgary.ca...
>>>> > ...
>>>> > > Could you comment on the "risk" of changing the loop counter within the
>>>> > > loop?
>
>>>> > my 2 cents.
>
>>>> > First, it is in changing the counter of a for loop.
>>>> > A for loop explicitly outlines what all counter variables will be.
>
>>>> > There are two things:
>
>>>> > 1) infinite loop, one could easily change the counter to never
>>>> > reach the end condition. A (valid) for loop will always reach the end
>>>> > condition.
>
>>>> > 2) more insidious, you could inadvertantly cast the counter to a float from
>>>> > an int, and then have one extra (and unintended ) statement executed.
>
>>>> This seems not to be possible in IDL, as loop counters, unlike normal
>>>> variables, cannot change their type.
>
>>>> Ciao,
>>>> Paolo
>
>>>> > instead of 0,1,2,3,4,5,6 (and not executing i = 7) you could get
>>>> > 0,1,2,3,4,4.99999999,5.99999,6.99999999, (and effectively executing the
>>>> > extra i ~ 7 step).
>
>>>> > Cheers,
>>>> > bob
>
>>> Hi all,
>
>>> Thanks for your replies. Just as David mentioned in his first
>>> response, a while loop worked out much better. Within a single while
>>> loop, i was able to accomplish the task, albeit a bit slowly because
>>> of the non-array operation.
>
>>> Thanks !- Hide quoted text -
>
>>> - Show quoted text -
>
>> Why not you do not share your final results with us to close this
>> post.
>> Elkunn
>
> Hi,
>
> I will. I don't have the code with me this weekend. I'll post it on
> Monday at work.
>
> Thanks,
> Raghu
Hello,
here is the while loop piece of code that works now.
b=0
k=1
while (b+k lt nb) do begin
if finite(ndvislice[s,b+k]) eq 0 or finite (ndsislice[s,b+k]) eq 0
then begin
ndvi[s,b+k]=mask[s,r]
ndsi[s,b+k]=mask[s,r]
k=k+1
endif else begin
if (ndvislice[s,b+k] lt ndvislice[s,b])then begin;
ndvi[s,b+k]=mask[s,r]
ndsi[s,b+k]=mask[s,r]
k=k+1
endif else begin
ndvi[s,b+k]=ndvislice[s,b+k]
ndsi[s,b+k]=ndsislice[s,b+k]
b=b+k
k=1
endelse
endelse
endwhile
So, with two conditions, i can change the number of bands (b) or i can
change the counter (k) depending on which condition is satisfied
during an iteration. Although it works, it does take some time which
might suggest that an array oriented code might work faster. But
anyway, it works.
Thanks!
Raghu
|
|
|
|
| Re: IDL FOR Loop variable increments [message #62578 is a reply to message #62575] |
Sun, 21 September 2008 15:37  |
raghuram
Messages: 32
Registered: February 2008
|
Member |
|
|
On Sep 21, 6:44 am, Bulrush <Wasit.Weat...@gmail.com> wrote:
> On Sep 20, 12:51 am, Raghu <raghuram.narasim...@gmail.com> wrote:
>
>
>
>> On Sep 19, 12:09 pm, pgri...@gmail.com wrote:
>
>>> R.G. Stockwell wrote:
>>>> "Jean H" <jghas...@DELTHIS.ucalgary.ANDTHIS.ca> wrote in message
>>>> news:gauiil$u32$1@news.ucalgary.ca...
>>>> ...
>>>> > Could you comment on the "risk" of changing the loop counter within the
>>>> > loop?
>
>>>> my 2 cents.
>
>>>> First, it is in changing the counter of a for loop.
>>>> A for loop explicitly outlines what all counter variables will be.
>
>>>> There are two things:
>
>>>> 1) infinite loop, one could easily change the counter to never
>>>> reach the end condition. A (valid) for loop will always reach the end
>>>> condition.
>
>>>> 2) more insidious, you could inadvertantly cast the counter to a float from
>>>> an int, and then have one extra (and unintended ) statement executed.
>
>>> This seems not to be possible in IDL, as loop counters, unlike normal
>>> variables, cannot change their type.
>
>>> Ciao,
>>> Paolo
>
>>>> instead of 0,1,2,3,4,5,6 (and not executing i = 7) you could get
>>>> 0,1,2,3,4,4.99999999,5.99999,6.99999999, (and effectively executing the
>>>> extra i ~ 7 step).
>
>>>> Cheers,
>>>> bob
>
>> Hi all,
>
>> Thanks for your replies. Just as David mentioned in his first
>> response, a while loop worked out much better. Within a single while
>> loop, i was able to accomplish the task, albeit a bit slowly because
>> of the non-array operation.
>
>> Thanks !- Hide quoted text -
>
>> - Show quoted text -
>
> Why not you do not share your final results with us to close this
> post.
> Elkunn
Hi,
I will. I don't have the code with me this weekend. I'll post it on
Monday at work.
Thanks,
Raghu
|
|
|
|
| Re: IDL FOR Loop variable increments [message #62580 is a reply to message #62578] |
Sun, 21 September 2008 06:44 |
Wasit.Weather
Messages: 62
Registered: February 2008
|
Member |
|
|
On Sep 20, 12:51 am, Raghu <raghuram.narasim...@gmail.com> wrote:
> On Sep 19, 12:09 pm, pgri...@gmail.com wrote:
>
>
>
>
>
>> R.G. Stockwell wrote:
>>> "Jean H" <jghas...@DELTHIS.ucalgary.ANDTHIS.ca> wrote in message
>>> news:gauiil$u32$1@news.ucalgary.ca...
>>> ...
>>>> Could you comment on the "risk" of changing the loop counter within the
>>>> loop?
>
>>> my 2 cents.
>
>>> First, it is in changing the counter of a for loop.
>>> A for loop explicitly outlines what all counter variables will be.
>
>>> There are two things:
>
>>> 1) infinite loop, one could easily change the counter to never
>>> reach the end condition. A (valid) for loop will always reach the end
>>> condition.
>
>>> 2) more insidious, you could inadvertantly cast the counter to a float from
>>> an int, and then have one extra (and unintended ) statement executed.
>
>> This seems not to be possible in IDL, as loop counters, unlike normal
>> variables, cannot change their type.
>
>> Ciao,
>> Paolo
>
>>> instead of 0,1,2,3,4,5,6 (and not executing i = 7) you could get
>>> 0,1,2,3,4,4.99999999,5.99999,6.99999999, (and effectively executing the
>>> extra i ~ 7 step).
>
>>> Cheers,
>>> bob
>
> Hi all,
>
> Thanks for your replies. Just as David mentioned in his first
> response, a while loop worked out much better. Within a single while
> loop, i was able to accomplish the task, albeit a bit slowly because
> of the non-array operation.
>
> Thanks !- Hide quoted text -
>
> - Show quoted text -
Why not you do not share your final results with us to close this
post.
Elkunn
|
|
|
|
| Re: IDL FOR Loop variable increments [message #62583 is a reply to message #62580] |
Fri, 19 September 2008 22:51 |
raghuram
Messages: 32
Registered: February 2008
|
Member |
|
|
On Sep 19, 12:09 pm, pgri...@gmail.com wrote:
> R.G. Stockwell wrote:
>> "Jean H" <jghas...@DELTHIS.ucalgary.ANDTHIS.ca> wrote in message
>> news:gauiil$u32$1@news.ucalgary.ca...
>> ...
>>> Could you comment on the "risk" of changing the loop counter within the
>>> loop?
>
>> my 2 cents.
>
>> First, it is in changing the counter of a for loop.
>> A for loop explicitly outlines what all counter variables will be.
>
>> There are two things:
>
>> 1) infinite loop, one could easily change the counter to never
>> reach the end condition. A (valid) for loop will always reach the end
>> condition.
>
>> 2) more insidious, you could inadvertantly cast the counter to a float from
>> an int, and then have one extra (and unintended ) statement executed.
>
> This seems not to be possible in IDL, as loop counters, unlike normal
> variables, cannot change their type.
>
> Ciao,
> Paolo
>
>
>
>> instead of 0,1,2,3,4,5,6 (and not executing i = 7) you could get
>> 0,1,2,3,4,4.99999999,5.99999,6.99999999, (and effectively executing the
>> extra i ~ 7 step).
>
>> Cheers,
>> bob
Hi all,
Thanks for your replies. Just as David mentioned in his first
response, a while loop worked out much better. Within a single while
loop, i was able to accomplish the task, albeit a bit slowly because
of the non-array operation.
Thanks !
|
|
|
|
| Re: IDL FOR Loop variable increments [message #62591 is a reply to message #62583] |
Fri, 19 September 2008 12:09 |
pgrigis
Messages: 436
Registered: September 2007
|
Senior Member |
|
|
R.G. Stockwell wrote:
> "Jean H" <jghasban@DELTHIS.ucalgary.ANDTHIS.ca> wrote in message
> news:gauiil$u32$1@news.ucalgary.ca...
> ...
>> Could you comment on the "risk" of changing the loop counter within the
>> loop?
>
> my 2 cents.
>
> First, it is in changing the counter of a for loop.
> A for loop explicitly outlines what all counter variables will be.
>
> There are two things:
>
> 1) infinite loop, one could easily change the counter to never
> reach the end condition. A (valid) for loop will always reach the end
> condition.
>
> 2) more insidious, you could inadvertantly cast the counter to a float from
> an int, and then have one extra (and unintended ) statement executed.
This seems not to be possible in IDL, as loop counters, unlike normal
variables, cannot change their type.
Ciao,
Paolo
>
> instead of 0,1,2,3,4,5,6 (and not executing i = 7) you could get
> 0,1,2,3,4,4.99999999,5.99999,6.99999999, (and effectively executing the
> extra i ~ 7 step).
>
> Cheers,
> bob
|
|
|
|
| Re: IDL FOR Loop variable increments [message #62592 is a reply to message #62591] |
Fri, 19 September 2008 12:07 |
raghuram
Messages: 32
Registered: February 2008
|
Member |
|
|
On Sep 19, 9:53 am, Ed Hyer <ejh...@gmail.com> wrote:
> Raghu,
>
> Not knowing the makeup of NDVI, NDVISLICE, and MASK in your example,
> it's very difficult to determine what you're up to here, but it
> strongly looks to me like some higher-order functions are in order
> here. You are aware that IDL can give you an array MAX(VAR,DIM=N) or
> MIN, TOTAL, or can do any boolean comparison you choose across an
> entire array?
>
> As I said, I don't really know what the outcome of your code should
> be, but even in this economy, I'll bet $5 you can do it with a much
> smaller, more comprehensible set of array operations.
>
> --Edward H.
You are right. I just haven't thought about using array operations
probably because
a working using these loops doesn't take very long to run. From what i
read here,
it looks a FOR loop may not be the best option for such operations.
Maybe i'll
use a while loop and see what happens or i'll rethink how the code can
be
written in a much more efficient manner. I do know that IDL can do
things like
MAx, TOtal etc. But in my case, all i'm trying to do is to assign
values from
one array(ndvislice) to another(ndvi) based on ndvislice's previious
values.
The outcome i'd expect for each output would be a finite value or a
NaN value.
So i assign my output(ndvi) to ndvislice if conditions are met, else
i'll assign
a NaN value( Mask).
Raghu
|
|
|
|
| Re: IDL FOR Loop variable increments [message #62593 is a reply to message #62592] |
Fri, 19 September 2008 11:53 |
R.G. Stockwell
Messages: 363
Registered: July 1999
|
Senior Member |
|
|
"Jean H" <jghasban@DELTHIS.ucalgary.ANDTHIS.ca> wrote in message
news:gauiil$u32$1@news.ucalgary.ca...
...
> Could you comment on the "risk" of changing the loop counter within the
> loop?
my 2 cents.
First, it is in changing the counter of a for loop.
A for loop explicitly outlines what all counter variables will be.
There are two things:
1) infinite loop, one could easily change the counter to never
reach the end condition. A (valid) for loop will always reach the end
condition.
2) more insidious, you could inadvertantly cast the counter to a float from
an int, and then have one extra (and unintended ) statement executed.
instead of 0,1,2,3,4,5,6 (and not executing i = 7) you could get
0,1,2,3,4,4.99999999,5.99999,6.99999999, (and effectively executing the
extra i ~ 7 step).
Cheers,
bob
|
|
|
|
| Re: IDL FOR Loop variable increments [message #62599 is a reply to message #62593] |
Fri, 19 September 2008 09:53 |
MarioIncandenza
Messages: 231
Registered: February 2005
|
Senior Member |
|
|
Raghu,
Not knowing the makeup of NDVI, NDVISLICE, and MASK in your example,
it's very difficult to determine what you're up to here, but it
strongly looks to me like some higher-order functions are in order
here. You are aware that IDL can give you an array MAX(VAR,DIM=N) or
MIN, TOTAL, or can do any boolean comparison you choose across an
entire array?
As I said, I don't really know what the outcome of your code should
be, but even in this economy, I'll bet $5 you can do it with a much
smaller, more comprehensible set of array operations.
--Edward H.
|
|
|
|
|
|
|
|
|
|
|
|
| Re: IDL FOR Loop variable increments [message #62605 is a reply to message #62603] |
Thu, 18 September 2008 18:31 |
raghuram
Messages: 32
Registered: February 2008
|
Member |
|
|
On Sep 18, 2:52 pm, Jean H <jghas...@DELTHIS.ucalgary.ANDTHIS.ca>
wrote:
> Could you comment on the "risk" of changing the loop counter within the
> loop?
>
> Jean
Ok here's the scene. In my example, lets say b=0 and k=2 (i.e. b+k=
0+2 = 2) satisfies my condition, thats like saying i'm comparing
Bands '0' and '2' as in line 9, "if ndvislice[s,b+k] ge ndvislice[s,b]
then begin". The next thing i want to do is to
compare bands '2' and '3'. Its because band 1 does not satisy my
condition and i don't want the code to look at band 1 again.
Hence i want to skip from band 0 to band 2. This is primarily because
when k was 1 (b+k=0+1=1), the condition was not satisfied which
is why 'k' changes to 2. The increments are not necessarily consistent
either. I may want to skip from band 4 to band 7 if bands 5 and 6
do not satisfy my condition which in the above code is on the 9th line
"if ndvislice[s,b+k] ge ndvislice[s,b] then begin ".
So, the reason why i try to set b to b+k is just so that 'b' turns to
'b+k' as in, b=0 to 'b+2" which means i want the new b,
or the b in the next FOR loop iteration to change to 2 from 0 instead
of changing to 1.
Maybe there's a better way to do this. I hope i'm being clear.
Looking forward to a reply,
Raghu
|
|
|
|
| Re: IDL FOR Loop variable increments [message #62607 is a reply to message #62605] |
Thu, 18 September 2008 15:46 |
David Fanning
Messages: 11724
Registered: August 2001
|
Senior Member |
|
|
Jean H writes:
>> I think you have clearly chosen the wrong program
>> construction. You should NEVER be changing a loop
>> variable inside the loop programmatically. (I don't
>> think IDL 7 even allows it, since I was inadvertently
>> doing it the other day and got my hand slapped.) Maybe
>> you want a WHILE loop.
>
> No problem to do that here..
> { x86 Win32 Windows Microsoft Windows 7.0.1 Mar 20 2008 32 64}
Humm. I was afraid of that. I was too much in the middle
of something to stop and see what the real problem was.
I was passing a loop variable into another program as a
parameter, and that other program was, on occasion, changing
the passed-in variable to something else, based on some
programming logic. I was getting some kind of error about
"cannot change this kind of variable".
I don't know. I've never seen the error before, and I can't
reproduce it now in a quick test program, so... Probably
didn't happen. :-)
> Could you comment on the "risk" of changing the loop counter within the
> loop?
I was in a hurry and late for a meeting when I wrote this,
but as I was going on the door, I realized that of course
we often change "loop variables" inside of loops. In fact,
the loop I suggested, a WHILE loop, nearly *always* does that.
So, nonsense.
The point I was trying to make, though, is that in
a counting loop, like a FOR loop, it seems to me that
changing the counter in the program is a terrible idea.
I just can't think of any time it is a good idea. I think
it is a programming style that just flat out leads to
programming errors. If you think you have to do it, then
I think you ought to re-think what you are doing, and choose
another programming structure that lends itself more easily
to the task.
I've just never seen anything good come of it.
Cheers,
David
P.S. Let's just say that since I now have to work for a living, I
find there is a LOT less time to actually think about what I am
writing. I am MUCH more sympathetic to the kind of advice that
is typical on any newsgroup except this one. :-)
--
David Fanning, Ph.D.
Coyote's Guide to IDL Programming (www.dfanning.com)
Sepore ma de ni thui. ("Perhaps thou speakest truth.")
|
|
|
|
| Re: IDL FOR Loop variable increments [message #62608 is a reply to message #62607] |
Thu, 18 September 2008 15:00 |
Jean H.
Messages: 472
Registered: July 2006
|
Senior Member |
|
|
> k=1
> for b=0,8 do begin ; Number of bands
> j=7
> c=b+k
> if b gt 0 and b ne (c) then continue
So, in this case, you go back to the for loop and increase b...
so, you have the first iteration with b=0 (this above statement is not
executed when b=0). Then, you increase b, so 'b gt 0' is true, and 'b ne
c' is also always true... k >= 1, so b+ k will NEVER be equal to b ...
so the above statement is always executed, until b > 8.
Jean
> for k=1, j do begin ; Iterations within bands
> if c lt 7 then begin
> if finite(ndvislice[s,b+k]) eq 1 then begin
> if ndvislice[s,b+k] ge ndvislice[s,b] then begin
> ndvi[s,b+k]=ndvislice[s,b+k]
> break
> endif else begin
> if ndvislice[s,b+k] lt ndvislice[s,b] then begin
> ndvi[s,b+k]=mask[s,r]
> endif
> endelse
> endif else begin
> ndvi[s,b+k]=mask[s,r]
> endelse
> endif
> endfor
> ENDFOR
>
> In the code above, i want b ( the number of bands) to be incremented
> by the value (b+k) instead of standard consistent increments of 1 or
> 2.. So, instead of b progressing from 0 to 8 as 0,1,2,3 etc..i want it
> to iterate based on the value of (b+k) derived out of the inner FOR
> loop (for k=1, j). My results run correctly but those pixels for which
> 'b' needs a (b+k) increment are not recogized by the line "if b gt 0
> and b ne (c) then continue" and so it keeps skipping all 'b'
> iterations. 'c' here is just a variable assigned to the value of b+k.
>
> Where am i going wrong ? Please let me know if you need more
> information.
>
> Thanks,
> Raghu
|
|
|
|
|
|
| Re: IDL FOR Loop variable increments [message #62612 is a reply to message #62609] |
Thu, 18 September 2008 13:23 |
pgrigis
Messages: 436
Registered: September 2007
|
Senior Member |
|
|
Raghu wrote:
> Hi all,
>
> I'm having a problem with FOR loop increments in IDL which i'm not
> able to fix. I have pasted the code and a small description about it
> follows.
> k=1
> for b=0,8 do begin ; Number of bands
> j=7
> c=b+k
k here is going to be equal 1 for the first iteration, and equal j
afterwards
Paolo
> if b gt 0 and b ne (c) then continue
> for k=1, j do begin ; Iterations within bands
> if c lt 7 then begin
> if finite(ndvislice[s,b+k]) eq 1 then begin
> if ndvislice[s,b+k] ge ndvislice[s,b] then begin
> ndvi[s,b+k]=ndvislice[s,b+k]
> break
> endif else begin
> if ndvislice[s,b+k] lt ndvislice[s,b] then begin
> ndvi[s,b+k]=mask[s,r]
> endif
> endelse
> endif else begin
> ndvi[s,b+k]=mask[s,r]
> endelse
> endif
> endfor
> ENDFOR
>
> In the code above, i want b ( the number of bands) to be incremented
> by the value (b+k) instead of standard consistent increments of 1 or
> 2.. So, instead of b progressing from 0 to 8 as 0,1,2,3 etc..i want it
> to iterate based on the value of (b+k) derived out of the inner FOR
> loop (for k=1, j). My results run correctly but those pixels for which
> 'b' needs a (b+k) increment are not recogized by the line "if b gt 0
> and b ne (c) then continue" and so it keeps skipping all 'b'
> iterations. 'c' here is just a variable assigned to the value of b+k.
>
> Where am i going wrong ? Please let me know if you need more
> information.
>
> Thanks,
> Raghu
|
|
|
|
| Re: IDL FOR Loop variable increments [message #62613 is a reply to message #62612] |
Thu, 18 September 2008 13:03 |
David Fanning
Messages: 11724
Registered: August 2001
|
Senior Member |
|
|
Raghu writes:
> In the code above, i want b ( the number of bands) to be incremented
> by the value (b+k) instead of standard consistent increments of 1 or
> 2.. So, instead of b progressing from 0 to 8 as 0,1,2,3 etc..i want it
> to iterate based on the value of (b+k) derived out of the inner FOR
> loop (for k=1, j). My results run correctly but those pixels for which
> 'b' needs a (b+k) increment are not recogized by the line "if b gt 0
> and b ne (c) then continue" and so it keeps skipping all 'b'
> iterations. 'c' here is just a variable assigned to the value of b+k.
>
> Where am i going wrong ?
I think you have clearly chosen the wrong program
construction. You should NEVER be changing a loop
variable inside the loop programmatically. (I don't
think IDL 7 even allows it, since I was inadvertently
doing it the other day and got my hand slapped.) Maybe
you want a WHILE loop.
Cheers,
David
--
David Fanning, Ph.D.
Coyote's Guide to IDL Programming (www.dfanning.com)
Sepore ma de ni thui. ("Perhaps thou speakest truth.")
|
|
|
|
| Re: IDL FOR Loop variable increments [message #62712 is a reply to message #62575] |
Tue, 23 September 2008 07:48 |
Jeremy Bailin
Messages: 618
Registered: April 2008
|
Senior Member |
|
|
On Sep 22, 11:54 am, Raghu <raghuram.narasim...@gmail.com> wrote:
> On Sep 21, 3:37 pm, Raghu <raghuram.narasim...@gmail.com> wrote:
>
>
>
>> On Sep 21, 6:44 am, Bulrush <Wasit.Weat...@gmail.com> wrote:
>
>>> On Sep 20, 12:51 am, Raghu <raghuram.narasim...@gmail.com> wrote:
>
>>>> On Sep 19, 12:09 pm, pgri...@gmail.com wrote:
>
>>>> > R.G. Stockwell wrote:
>>>> > > "Jean H" <jghas...@DELTHIS.ucalgary.ANDTHIS.ca> wrote in message
>>>> > >news:gauiil$u32$1@news.ucalgary.ca...
>>>> > > ...
>>>> > > > Could you comment on the "risk" of changing the loop counter within the
>>>> > > > loop?
>
>>>> > > my 2 cents.
>
>>>> > > First, it is in changing the counter of a for loop.
>>>> > > A for loop explicitly outlines what all counter variables will be.
>
>>>> > > There are two things:
>
>>>> > > 1) infinite loop, one could easily change the counter to never
>>>> > > reach the end condition. A (valid) for loop will always reach the end
>>>> > > condition.
>
>>>> > > 2) more insidious, you could inadvertantly cast the counter to a float from
>>>> > > an int, and then have one extra (and unintended ) statement executed.
>
>>>> > This seems not to be possible in IDL, as loop counters, unlike normal
>>>> > variables, cannot change their type.
>
>>>> > Ciao,
>>>> > Paolo
>
>>>> > > instead of 0,1,2,3,4,5,6 (and not executing i = 7) you could get
>>>> > > 0,1,2,3,4,4.99999999,5.99999,6.99999999, (and effectively executing the
>>>> > > extra i ~ 7 step).
>
>>>> > > Cheers,
>>>> > > bob
>
>>>> Hi all,
>
>>>> Thanks for your replies. Just as David mentioned in his first
>>>> response, a while loop worked out much better. Within a single while
>>>> loop, i was able to accomplish the task, albeit a bit slowly because
>>>> of the non-array operation.
>
>>>> Thanks !- Hide quoted text -
>
>>>> - Show quoted text -
>
>>> Why not you do not share your final results with us to close this
>>> post.
>>> Elkunn
>
>> Hi,
>
>> I will. I don't have the code with me this weekend. I'll post it on
>> Monday at work.
>
>> Thanks,
>> Raghu
>
> Hello,
>
> here is the while loop piece of code that works now.
>
> b=0
> k=1
>
> while (b+k lt nb) do begin
>
> if finite(ndvislice[s,b+k]) eq 0 or finite (ndsislice[s,b+k]) eq 0
> then begin
> ndvi[s,b+k]=mask[s,r]
> ndsi[s,b+k]=mask[s,r]
> k=k+1
> endif else begin
> if (ndvislice[s,b+k] lt ndvislice[s,b])then begin;
> ndvi[s,b+k]=mask[s,r]
> ndsi[s,b+k]=mask[s,r]
> k=k+1
> endif else begin
> ndvi[s,b+k]=ndvislice[s,b+k]
> ndsi[s,b+k]=ndsislice[s,b+k]
> b=b+k
> k=1
> endelse
> endelse
> endwhile
>
> So, with two conditions, i can change the number of bands (b) or i can
> change the counter (k) depending on which condition is satisfied
> during an iteration. Although it works, it does take some time which
> might suggest that an array oriented code might work faster. But
> anyway, it works.
>
> Thanks!
>
> Raghu
Okay, I think I finally understand what you're trying to do. Let me
try to encapsulate it in words:
As long as ndvislice is not dropping (with NaNs treated as missing
data), ndvi and ndsi get set to ndvislice and ndsislice respectively,
otherwise they get set to the mask value.
(and ndvi[0] and ndsi[0] are unchanged from the input)
In which case, I think that this is a vectorized version:
; mask out all values at first, and only set the relevant ones
ndvi[s,1:*] = mask[s,r]
ndsi[s,1:*] = mask[s,r]
; ignore missing data
wherefinite = where(finite(ndvislice[s,*]) ne 0 and
finite(ndsislice[s,*]) ne 0, nfinite)
if nfinite gt 0 then begin
; which ones are greater than or equal to previous good value?
wherenodrop = where(ndvislice[s,wherefinite[1:*]] ge
ndvislice[s,wherefinite], nnodrop)
if nnodrop gt 0 then begin
; get the index back into the original array
nodrop_i = wherefinite[wherenodrop]+1
ndvi[s,nodrop_i] = ndvislice[s,nodrop_i]
ndsi[s,nodrop_i] = ndsislice[s,nodrop_i]
endif
endif
-Jeremy.
|
|
|
|
comp.lang.idl-pvwave archive
Members
Search
Help
Login
Home
