Hi Eric,<div><br></div><div>Is it possible to fit more than 20 sources at a time? I have a few channels where there are up to 40 sources.</div><div><br></div><div>Thanks,</div><div>Tanmoy</div><div><br><br><div class="gmail_quote">
On Mon, Jan 9, 2012 at 6:39 PM, Eric Greisen <span dir="ltr"><<a href="mailto:egreisen@nrao.edu" target="_blank">egreisen@nrao.edu</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<div>Tanmoy Laskar wrote:<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
Thanks, Eric. That works brilliantly.<br>
<br>
I'm running fits on my data set, and for certain channels, both the pre- and post-fit RMS sometimes come out to be NaN or Infinity. The "CHI squares" are all finite, though! Why do you suppose this could be? Here is a sample INLIST file that does this:<br>
<br>
</blockquote></div>
I don't think your model is all that good. It needs a source at the origin.<br>
<br>
But UVFIT had lots of errors affecting it when there were > 4 components fit. I have fixed the dimensions of lots of variables and things are much better. I also fixed the output file a bit for times and rms/chisq formats.<br>
<br>
Tomorrow's MNJ in 31DEC12 will fix.<br><font color="#888888">
<br>
Eric Greisen<br>
</font></blockquote></div><br></div>