spec error in frame 3, not 2?

[saurabhwjha]

I was getting negative uncertainties that were messing up the specombine step. These lines get the error spectrum from frame 2

866: specerr = hdu[0].data[2, ap]
1017: specerrs[i] = interp(lamgrid, lam, hdu[0].data[2][ap] ** 2.0) ** 0.5

Looking at the spectra though, I think the uncertainty is in frame 3?

[cmccully]

Python is 0 indexed. I believe that the uncertainties are in the third extension which corresponds to index 2. The negative uncertainties is a significant issue though.

[saurabhwjha]

It is definitely in BANDID 4 for my spectra:

% imhead sci13.25x1d001.fits | grep BANDID
BANDID1 = 'spectrum - background fit, weights variance, clean no'
BANDID2 = 'raw - background fit, weights none, clean no'
BANDID3 = 'background - background fit'
BANDID4 = 'sigma - background fit, weights variance, clean no'

On Sep 23, 2015, at 5:51 pm, Curtis McCully notifications@github.com wrote:

Python is 0 indexed. I believe that the uncertainties are in the third extension which corresponds to index 2. The negative uncertainties is a significant issue though.

—
Reply to this email directly or view it on GitHub.