comp.lang.idl-pvwave archive: archive » "infinite" nested for

Home » Public Forums » archive » "infinite" nested for

Show: Today's Messages :: Show Polls :: Message Navigator
E-mail to friend

Re: "infinite" nested for [message #86645 is a reply to message #86642]

Mon, 25 November 2013 10:25

Mats Löfdahl
Messages: 263
Registered: January 2012

Senior Member

On 2013-11-25 17:18, Chrisss wrote:
> Hi Mats Löfdahl, thanks for your interest!
> my problem is that my "Im" is a 3D matrix in which every pixel is a curve with 60 elements.

That much I got. :o)

> Inner loop is for these ones: it does the fitting after the exponential curve inversion. I must follow a scholastic approach that tests all of 60 inversion time T0 within inner loop.

In the loop you do

T0=t
yCurr=y
if (t ne 0) then yCurr[0:T0]=-y[0:To]

So you are changing the sign of the first t elements in y. (I'm assuming
you don't have To in your program and just happened to write it instead
of T0.)

What I don't really see is why you'd want to try fitting an exponential
to negative numbers, hence the question about just using abs(). Is there
some noise that can make the measured (I assume) numbers be negative?

I guess I don't know what an "exponential curve inversion" is. Maybe you
should make t0 a parameter that you can use in myfunc instead, so
instead of testing all possible points where you can flip the sign of
your data, you fit your unchanged data to a function that is an
exponential where you flip the sign at some point t0, that is part of
the fit.

Report message to a moderator

[Message index]

		"infinite" nested for By: Chrisss on Mon, 25 November 2013 00:13
		Re: "infinite" nested for By: Kai Muehlbauer on Mon, 25 November 2013 00:27
		Re: "infinite" nested for By: Chrisss on Mon, 25 November 2013 02:24
		Re: "infinite" nested for By: Kai Muehlbauer on Mon, 25 November 2013 03:22
		Re: "infinite" nested for By: Chrisss on Mon, 25 November 2013 03:53
		Re: "infinite" nested for By: Kai Muehlbauer on Mon, 25 November 2013 04:12
		Re: "infinite" nested for By: Chrisss on Mon, 25 November 2013 06:56
		Re: "infinite" nested for By: Mats Löfdahl on Mon, 25 November 2013 07:38
		Re: "infinite" nested for By: Chrisss on Mon, 25 November 2013 08:18
		Re: "infinite" nested for By: David Fanning on Mon, 25 November 2013 08:37
		Re: "infinite" nested for By: Chrisss on Mon, 25 November 2013 10:00
		Re: "infinite" nested for By: Mats Löfdahl on Mon, 25 November 2013 10:25
		Re: "infinite" nested for By: Chrisss on Mon, 25 November 2013 11:24
		Re: "infinite" nested for By: Craig Markwardt on Tue, 26 November 2013 08:04
		Re: "infinite" nested for By: Chrisss on Tue, 26 November 2013 09:05

Previous Topic:	lamda default value in CURVEFIT function
Next Topic:	Secret of Failed IDL Business Plan

-=] Back to Top [=-

[ Syndicate this forum (XML) ] [

] [

]

Current Time: Wed Oct 08 17:06:51 PDT 2025

Total time taken to generate the page: 0.00481 seconds