dnidever / doppler Goto Github PK

Hey @dnidever! This may be an issue for @holtzmanjon (because I don't know what version of doppler he uses for APOGEE), but I was just writing a sentence about doppler in a planned APOGEE paper, and realized that the journal will probably ask for a citable record for this software. This isn't urgent, but it would probably be good to connect this repo to Zenodo (or some other similar service) and tag a release or version for the software version that was used to produce the APOGEE DR17 reduction. I'm happy to help with any of this if you haven't done it before (it should only take a few minutes).

Hello @dnidever,

When I run doppler with the Payne model option and specify which parameters I want to fit (similar to your second Payne model example in the documentation) I am running into problems. It works fine if I use the default values of teff,logg,fe_h,alpha_h,rv but if I use other values I get one of a few errors.

If I add one more parameter, say vmacro, or use the exact list in your example (teff,logg,fe_h,alpha_h,c_h,rv), I get the error:

UnboundLocalError: local variable 'initparlabels' referenced before assignment

that corresponds to line 1434 in the rv.py file. Finding that line, there are 2 identical "if" conditions in lines 1430 and 1432, and since I give 6 parameters and not 5, neither is True, so the initparlabels variable doesn't get set.

Knowing this, if I fit only 5 parameters, but I do not include rv in the list (teff,logg,fe_h,alpha_h,vmacro), I get the error:

ValueError: not enough values to unpack (expected 1, got 0)

corresponding to line 1771 in the rv.py file.

If I fit for the parameters teff,logg,c_h,rv,vmacro I get the notification "THIS SHOULD NOT HAPPEN!" and the following:

> /Users/kcarrell/Library/Python/3.8/lib/python/site-packages/doppler/payne.py(1517)mklabels()
-> for k,name in enumerate(labelnames):
(Pdb)

and it waits for a keyboard prompt.

If I remove teff or logg from the list I get an error saying they are required input parameters.

If I replace alpha_h with vmacro everything runs fine, but if I replace it with vmicro it gives me the error:

ValueError: Each lower bound must be strictly less than each upper bound.

corresponding to line 793 in the least_squares.py file.

If I replace alpha_h with c_h it seems to run fine and produce results.

I suspect one of the two "if" statements should be changed in lines 1430 and 1432, although I'm not positive.

(Also, on a minor, nitpicky note, in line 1407 the comment says you are loading the Cannon models and not the Payne models.)

Thanks.
Kenneth