Closed mrawls closed 7 years ago
I should add, the one thing this doesn't do is work with the current spec.continuum.fit
function. It looks like that is currently very hard-wired with the default wavelength grid; is there any way you could generalize it a bit to work with any wavelength grid?
Thanks, that was quick! I'm busy today, but I will take a look at whether this works for me ASAP.
Yes, spec.continuum.fit
is somewhat specific to the apStar
grid, because it treats every detector separately. I don't know enough about the apVisit
format to know how to edit it for those. Of course, it would be relatively easy to write the continuum fit for just a single wavelength region, which would be general.
Looked at this briefly and it seems to largely work. I'll merge it once I can do some more testing (probably only early next week), but a few comments:
FIBERID
needs to be a string. It might be better to allow an int and cast it to %03i
. Same for MJD
and loc_id
.apogee_id
is currently not used, so it probably shouldn't be a required input. Maybe make it a keyword if you plan to allow downloading based on apogee_id
.Also would be great if you could include an example in the README.rst
, similar to the apStar
example.
I will add some examples in the README shortly.
Sorry for the delay (was at a conference without access to GitHub last week!), but I merged this now. I slightly changed the code to allow fiberid
to be a regular integer (it's an integer in the allVisit
file, so I think your version would not have worked if you get values from allVisit
).
Thanks!
This should do the trick, but goodness do I have some words for whoever decided to sort things by location, date, and fiber rather than apogee_id.