Failure for lightcurves with gaps

[philastrophist]

I've run Kali with a reprocessed light curve from Smith+18 and it fails with error:

https://gist.github.com/philastrophist/02a1444ab8a51f29e37481f7b40650ea

notice: kali tools have been set up.
/opt/conda/lib/python2.7/site-packages/numpy/lib/nanfunctions.py:1628: RuntimeWarning: Degrees of freedom <= 0 for slice.
  keepdims=keepdims)
Traceback (most recent call last):
  File "/project/smith18_data/fit_carma.py", line 82, in <module>
    task.fit(lc)
  File "/home/kali/python/kali/carma.py", line 1048, in fit
    self.bestTheta
  File "/home/kali/python/kali/carma.py", line 1058, in bestTheta
    bestWalker = np.where(np.max(self.LnPosterior) == self.LnPosterior)[0][0]
IndexError: index 0 is out of bounds for axis 0 with size 0

Is this because there are gaps? If so, then how does the SDSS lightcurve in your guide/introduction.ipynb work?

Thanks

[AstroVPK]

I'll try to take a look at it either later today or tomorrow - sorry, but there's some personal stuff that's keeping me busy. Vishal Kasliwal Senior Deep Learning Software Engineer Intel Corporation <https://www.intel.com/content/www/us/en/homepage.html>

…

On Tue, Oct 8, 2019 at 4:03 AM Shaun Read ***@***.***> wrote: I've run Kali with a reprocessed light curve from Smith+18 and it fails with error: https://gist.github.com/philastrophist/02a1444ab8a51f29e37481f7b40650ea notice: kali tools have been set up. /opt/conda/lib/python2.7/site-packages/numpy/lib/nanfunctions.py:1628: RuntimeWarning: Degrees of freedom <= 0 for slice. keepdims=keepdims) Traceback (most recent call last): File "/project/smith18_data/fit_carma.py", line 82, in <module> task.fit(lc) File "/home/kali/python/kali/carma.py", line 1048, in fit self.bestTheta File "/home/kali/python/kali/carma.py", line 1058, in bestTheta bestWalker = np.where(np.max(self.LnPosterior) == self.LnPosterior)[0][0] IndexError: index 0 is out of bounds for axis 0 with size 0 Is this because there are gaps? If so, then how does the SDSS lightcurve in your guide/introduction.ipynb work? Thanks — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub <#8?email_source=notifications&email_token=ACWW2J7XA6J4I6DKTCGNHATQNRSJ5A5CNFSM4I6QHYZ2YY3PNVWWK3TUL52HS4DFUVEXG43VMWVGG33NNVSW45C7NFSM4HQKCLMQ>, or mute the thread <https://github.com/notifications/unsubscribe-auth/ACWW2JYLADPD4OW5ICQLPKLQNRSJ5ANCNFSM4I6QHYZQ> .

[philastrophist]

Ok, thanks!

[philastrophist]

Ok, it runs now and returns some results!

However, they look a bit weird: The psd doesn't really track the data that well and nor does the SF.
Have a look at the pdf in the gist if you can. Is it expected to be like this? I know that at the high freq end we're observationally limited so that could explain the PSD but not the SF! I'd been interested to know what you think...
Thanks for your help