comp.lang.idl-pvwave archive
Messages from Usenet group comp.lang.idl-pvwave, compiled by Paulo Penteado

Home » Public Forums » archive » MPMINSTEP in mpfit?
Show: Today's Messages :: Show Polls :: Message Navigator
E-mail to friend 
Switch to threaded view of this topic Create a new topic Submit Reply
MPMINSTEP in mpfit? [message #94068] Thu, 12 January 2017 23:08 Go to next message
Maryam is currently offline  Maryam
Messages: 16
Registered: July 2012
Junior Member
Hello.

I was wondering if we can specify the minimum change to be made in the parameter value in mpfit (i.e. similar to MPMAXSTEP but for min step)? The keyword .step specifies the minimum step size in the calculation of the derivatives but that's not what I want. I want to be able to tell mpfit what the minimum change in the parameters should be from one iteration to the next and let it figure out what step size to use for the calculation of the derivatives based on that. I noticed that earlier versions of mpfit.pro apparently had a MPMINSTEP but it seems to have been removed.

Any help is appreciated.
Maryam
Re: MPMINSTEP in mpfit? [message #94073 is a reply to message #94068] Fri, 13 January 2017 07:15 Go to previous messageGo to next message
Craig Markwardt is currently offline  Craig Markwardt
Messages: 1869
Registered: November 1996
Senior Member
On Friday, January 13, 2017 at 2:08:12 AM UTC-5, Maryam wrote:
> Hello.
>
> I was wondering if we can specify the minimum change to be made in the parameter value in mpfit (i.e. similar to MPMAXSTEP but for min step)? The keyword .step specifies the minimum step size in the calculation of the derivatives but that's not what I want. I want to be able to tell mpfit what the minimum change in the parameters should be from one iteration to the next and let it figure out what step size to use for the calculation of the derivatives based on that. I noticed that earlier versions of mpfit.pro apparently had a MPMINSTEP but it seems to have been removed.

The short answer is, probably not.

If your concern is about how derivatives are calculated, use STEP. That will set a minimum size for the finite difference derivative step size.

If your concern is about parameter step size from one iteration to the next, then no, you can't set that. I experimented with it early on, but it never functioned properly.

Craig
Re: MPMINSTEP in mpfit? [message #94079 is a reply to message #94073] Fri, 13 January 2017 11:39 Go to previous message
Maryam is currently offline  Maryam
Messages: 16
Registered: July 2012
Junior Member
Thank you!
  Switch to threaded view of this topic Create a new topic Submit Reply
Previous Topic: boundary of a masked region
Next Topic: reform a transposed array

-=] Back to Top [=-
[ Syndicate this forum (XML) ] [ RSS ] [ PDF ]

Current Time: Wed Oct 08 07:17:30 PDT 2025

Total time taken to generate the page: 0.00384 seconds