Re: voigt errors

On Fri, Nov 16, 2007 at 13:45 -0500, rgibson_at_email.domain.hidden> I frequently get errors using the new voigt parameter with ISIS 1.4.8.  I can't
> tell what the problem is from the error output.  It seems related to the
> parameters I set before calling fit_counts, though.
> 
> Thanks,
> Rob
> 

I think you could avoid these errors by providing more
reasonable, narrower limits on the parameter values.

For example, it seems unreasonable to allow the minimizer to
explore negative values of the photon energy, and thermal
velocity.

-John


> 
> isis> list_par;
> poly5aScale(1) + voigt(1)
>  idx  param              tie-to  freeze         value         min         max
>   1  poly5aScale(1).a0       0     1    -6.682142e+28     -1e+100      1e+100
>   2  poly5aScale(1).a1d1e1   0     1     2.238173e+27     -1e+100      1e+100
>   3  poly5aScale(1).a2d1e2   0     1      0.003270774     -1e+100      1e+100
>   4  poly5aScale(1).a3d1e3   0     1                0     -1e+100      1e+100
>   5  poly5aScale(1).a4d1e4   0     1                0     -1e+100      1e+100
>   6  poly5aScale(1).a5d1e5   0     1                0     -1e+100      1e+100
>   7  poly5aScale(1).scale    0     1                1           0           0
>   8  voigt(1).norm           0     0     3.834939e+31     -1e+100      1e+100 
> photons/s/cm^2
>   9  voigt(1).energy         0     0      0.007998968     -1e+100      1e+100 
> keV
>  10  voigt(1).fwhm           0     0            0.001     -1e+100      1e+100 
> keV
>  11  voigt(1).vtherm         0     0             4000     -1e+100      1e+100 
> km/s
> 
> isis> fit_counts();
> 
> ... stuff happens...
> statistic: 6.063471e+02 p[0]=1.242990e+32       p[1]=7.998968e-03      
> p[2]=1.000000e-03       p[3]=9.420166e+01
> statistic: 6.063471e+02 p[0]=1.242990e+32       p[1]=7.998968e-03      
> p[2]=1.000000e-03       p[3]=9.420166e+01
> subplex[245]:  succeeded
>  Parameters[Variable] = 11[4]
>             Data bins = 100
>            Chi-square = 706.2502
>    Reduced chi-square = 7.356772
> Error: function evaluation failed
> ***string***:1:voigt:Isis Error
> Struct_Type with 3 fields
> 100
> 4
> 706.25
> 0
> isis> get_fit_method;
> subplex;tol=1.e-5;scale_factor=0.1;maxnfe=2000
> 
> ----
> You received this message because you are
> subscribed to the isis-users list.
> To unsubscribe, send a message to
> isis-users-request_at_email.domain.hidden> with the first line of the message as:
> unsubscribe

----
You received this message because you are
subscribed to the isis-users list.
To unsubscribe, send a message to
isis-users-request_at_email.domain.hiddenwith the first line of the message as:
unsubscribe