| Re: Looking for more elegant plot scaling... [message #76650] |
Mon, 20 June 2011 03:12 |
Gray
Messages: 253
Registered: February 2010
|
Senior Member |
|
|
On Jun 19, 9:01 pm, Balt <bindermue...@gmail.com> wrote:
> Both very elegant and just what I was looking for, thanks guys!
>
> - Balt
There's also the MINMAX routine in the NASA IDL Astro library:
IDL> mnmxy = minmax([all,your,arrays])
IDL> oplot, thing, yra=mnmxy
|
|
|
|
|
|
| Re: Looking for more elegant plot scaling... [message #76652 is a reply to message #76651] |
Sat, 18 June 2011 18:57  |
ben.bighair
Messages: 221
Registered: April 2007
|
Senior Member |
|
|
Hi,
On 6/18/11 1:16 AM, Balt wrote:
> Hi all,
>
> I'm OPLOTting 3 curves on a fourth one. Is there a more elegant way to
> scale a plot to YRANGE=[maxy,miny] than this:
>
> ; get the y max for the plots
> maxy = MAX(temp_obsdata)
> IF MAX(splined_wvrdata) GT maxy THEN maxy = MAX(splined_wvrdata)
> IF MAX(residuals) GT maxy THEN maxy = MAX(residuals)
> IF MAX(interp_obsdata) GT maxy THEN maxy = MAX(interp_obsdata)
>
> miny = MIN(temp_obsdata)
> IF MIN(splined_wvrdata) LT miny THEN miny = MIN(splined_wvrdata)
> IF MIN(residuals) LT miny THEN miny = MIN(residuals)
> IF MIN(interp_obsdata) LT miny THEN miny = MIN(interp_obsdata)
>
> Maybe something a bit more C macro-ish... ?
I don't know about C stuff, but IDL's compare-and-assign operators could
be handy here...
IDL> a = randomn(s,10)
IDL> b = randomn(s,10)
IDL> c = randomn(s,10)
IDL> print, a,b,c
0.774450 0.0338028 -2.49289 -0.430953 2.15833
0.868524 -0.145793 1.13590 0.726517 1.28041
-1.14027 -0.520221 -0.283700 -0.849713 0.00405929
-0.525046 -0.335924 -0.548919 1.01423 0.0764202
2.04586 0.00412738 1.72472 0.707579 0.997633
1.72583 0.913348 0.699497 1.20014 0.907764
IDL> maxy = max(a) > max(b) > max(c)
IDL> print, maxy
2.15833
IDL> maxy = max(a, min = mina) > max(b, min = minb) > max(c, min=minc)
IDL> print, mina,minb,minc
-2.49289 -1.14027 0.00412738
IDL> miny = mina < minb < minc
IDL> print, miny
-2.49289
You could also simply concatenate all of the arrays together, but that
seems like a lot of overhead.
Cheers,
Ben
|
|
|
|
| Re: Looking for more elegant plot scaling... [message #76655 is a reply to message #76652] |
Sat, 18 June 2011 05:04 |
wlandsman
Messages: 743
Registered: June 2000
|
Senior Member |
|
|
I'd probably go with the single line
maxy = max([temp_obsdata,splined_wvrdata,residuals,interp_obsdata], min=miny)
You could also use
maxy = max(temp_obsdata) > max(splined_wvrdata) > max(residuals) > max(interp_obsdata)
miny = min(temp_obsdata) < min(spline_wvrdata) < min(residuals) < min(interp_obsdata)
--Wayne
|
|
|
|
comp.lang.idl-pvwave archive
Members
Search
Help
Login
Home
