Page 1 of 1

minimum maximum misfit

Posted: Mon May 26, 2008 11:02 am
by gdg
Hi Marc,
the dinver in my pc (last snapshot distributed in May) allows to select the maximum misfit limited to the forth decimal numbers (0.000?).
Now the problem is that, after the inversion, I have a misfit very low (2.1e-6), and the viewer to see the inversion results (I mean ground profiles viewer, dispersion viewer etc etc) not accept an exponential format for the misfit.

Maybe I can select the models at very low misfits with the combined use of gpdcreport and figue,
I suppose something like that
gpdcreport -m 2.2e-6 | figue -c dc.mkup

Giuseppe

Posted: Thu Jun 05, 2008 12:49 pm
by admin
You can also use gpviewdcreport to view report files like in dinver.

You can also use gpdcreport but note the minor changes:
gpdcreport -m 2.2e-6 -Rp 1 | figue -mc dc.mkup
Without -Rp 1, the default output is the model files (option -gm). With -Rp you can extract the phase slowness of Rayleigh fundamental mode. Use options -vs or -vp instead to get the Vs or Vp profile. Option -mc stands for "many or multi curves". It is better than -c when there are many curves to plot like in the inversion results.

Third alternative, you can easily change the precision of spin boxes (the little wigdet with arrows up and down used to specify values like the maximum misfit). There is a special post about that: Changing the precision in spin boxes for real values