-
Notifications
You must be signed in to change notification settings - Fork 0
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Redshifts from sky fiber spectra #6
Comments
This is a good suggestion @julienguy. @forero can I assign this to you? One idea would be to use the same set of tiles chosen by |
@moustakas @julienguy Sure, I can work on this, but I am not entirely sure about what I should do exactly. i.e., we start by taking the fibers with |
I would simply record the fraction of sky fibers with delta_chi2 > 9 and delta_chi2 > 16 and in the redshift range of each target class in order to get a measurement of the false positive rate. |
Actually, @forero, I think this is relatively easy for me to do. I'll report back here once I've implemented what I have in mind. |
OK, I have a sample of ~33k sky-fiber spectra written out based on the same set of representative main-survey (iron) tiles we're using for our nominal tests. The catalog is here:
For the record, the coadded (cumulative-tile) spectra and (iron) Redrock results for this sample were gathered with:
But let's keep this ticket open until I've coded up the summary statistics @julienguy wants to see based on all the different template sets. |
@moustakas Did you rerun to code to generate the representative main-survey (iron) tiles? The list of tiles I shared with you ages ago was not reproducible (I did not initialize the random number generator). The committed version of https://github.com/desihub/rrevaluator/blob/main/bin/select_test_tiles generates reproducible results. |
Consider adding a false positive score based on the redrock fits of sky fiber spectra.
It helps making decision on the balance between flexibility and realism of the spectral models.
The text was updated successfully, but these errors were encountered: