Closed sybenzvi closed 4 years ago
Fixed; see PR #541. Turns out this had nothing to do with bad templates or issue #537, but rather an incorrectly defined redshift array passed to sim_spectra. I know longer remember why I defined the array in the first place; February Segev made July Segev's life harder by not providing comments.
The user-level script which generates galaxy + transient spectra through desisim and desitarget is letting through failed spectra (flux values all zero or containing NaN). The script could catch and skip these failures. It's possible this issue is duplicating #537.