comp.lang.idl-pvwave archive: archive

Home » Public Forums » archive » CURVEFIT function

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

CURVEFIT function [message #85101]

Thu, 04 July 2013 04:47

fd_luni
Messages: 66
Registered: January 2013

Member

Hi,

I was wondering if there is any command that I can put in the function CURVEFIT in order not print out the following:

CURVEFIT: Failed to converge- CHISQ increasing without bound.

Many Thanks
M

Report message to a moderator

Re: CURVEFIT function [message #85102 is a reply to message #85101]

Thu, 04 July 2013 05:34

Petros Syntelis
Messages: 32
Registered: June 2013

Member

On Thursday, July 4, 2013 2:47:14 PM UTC+3, fd_...@mail.com wrote:
> Hi,
>
>
>
> I was wondering if there is any command that I can put in the function CURVEFIT in order not print out the following:
>
>
>
> CURVEFIT: Failed to converge- CHISQ increasing without bound.
>
>
>
> Many Thanks
>
> M

You can edit the curvefit file and comment the message, or change it in a way that is tuned off by a keyword.

By my little experience in nonlinear fitting,i would definitely NOT recommend to do so if you are not absolutely sure of why you are doing so..
Non linear fitting is tricky as it is. You shouldn't make it more troublesome.

Report message to a moderator

Re: CURVEFIT function [message #85103 is a reply to message #85102]

Thu, 04 July 2013 07:19

Moritz Fischer
Messages: 32
Registered: June 2013

Member

Hi Maria,
looking at the code ( release 8.1, lines 367 ) it seems to me you could
set the status keyword to suppress the message.

cheers
m

Am 04.07.2013 14:34, schrieb Petros Syntelis:
> On Thursday, July 4, 2013 2:47:14 PM UTC+3, fd_...@mail.com wrote:
>> Hi,
>>
>>
>>
>> I was wondering if there is any command that I can put in the function CURVEFIT in order not print out the following:
>>
>>
>>
>> CURVEFIT: Failed to converge- CHISQ increasing without bound.
>>
>>
>>
>> Many Thanks
>>
>> M
>
> You can edit the curvefit file and comment the message, or change it in a way that is tuned off by a keyword.
>
> By my little experience in nonlinear fitting,i would definitely NOT recommend to do so if you are not absolutely sure of why you are doing so..
> Non linear fitting is tricky as it is. You shouldn't make it more troublesome.
>

Report message to a moderator

Re: CURVEFIT function [message #85104 is a reply to message #85103]

Thu, 04 July 2013 07:28

fd_luni
Messages: 66
Registered: January 2013

Member

On Thursday, 4 July 2013 15:19:10 UTC+1, Moritz Fischer wrote:
> Hi Maria, looking at the code ( release 8.1, lines 367 ) it seems to me you could set the status keyword to suppress the message. cheers m Am 04.07.2013 14:34, schrieb Petros Syntelis: > On Thursday, July 4, 2013 2:47:14 PM UTC+3, fd_...@mail.com wrote: >> Hi, >> >> >> >> I was wondering if there is any command that I can put in the function CURVEFIT in order not print out the following: >> >> >> >> CURVEFIT: Failed to converge- CHISQ increasing without bound. >> >> >> >> Many Thanks >> >> M > > You can edit the curvefit file and comment the message, or change it in a way that is tuned off by a keyword. > > By my little experience in nonlinear fitting,i would definitely NOT recommend to do so if you are not absolutely sure of why you are doing so.. > Non linear fitting is tricky as it is. You shouldn't make it more troublesome. >

I have the version 6.4. I tried to put the keyword STATUS=1 but it still appear the message.

Report message to a moderator

Re: CURVEFIT function [message #85111 is a reply to message #85101]

Fri, 05 July 2013 02:51

Petros Syntelis
Messages: 32
Registered: June 2013

Member

On Thursday, July 4, 2013 2:47:14 PM UTC+3, fd_...@mail.com wrote:
> Hi,
>
>
>
> I was wondering if there is any command that I can put in the function CURVEFIT in order not print out the following:
>
>
>
> CURVEFIT: Failed to converge- CHISQ increasing without bound.
>
>
>
> Many Thanks
>
> M

You used it in a wrong way. you set
status=status
and after you run
yfit=curvefit(...., status=status)
you see the result of the computation by
print, status
Then the status variable takes the values of 0,1,2 which stand for one of the three cases.

Petros

Report message to a moderator

Re: CURVEFIT function [message #85142 is a reply to message #85101]

Tue, 09 July 2013 13:15

Paul Van Delst[1]
Messages: 1157
Registered: April 2002

Senior Member

Yes, there is.

I believe the command is: MPFIT

:o)

See http://www.physics.wisc.edu/~craigm/idl/fitting.html

cheers,

paulv

On 07/04/13 07:47, fd_luni@mail.com wrote:
> Hi,
>
> I was wondering if there is any command that I can put in the function CURVEFIT in order not print out the following:
>
> CURVEFIT: Failed to converge- CHISQ increasing without bound.
>
> Many Thanks
> M
>

Report message to a moderator

Previous Topic:	Fast way to collapse data cube with median
Next Topic:	Most Important and urgent_clipping out a portion of image

-=] Back to Top [=-

[ Syndicate this forum (XML) ] [

] [

]

Current Time: Wed Oct 08 11:40:19 PDT 2025

Total time taken to generate the page: 0.00739 seconds