Initial dMag array being set to all nan's, interpolation error in completeness calculation #414
Description
Describe the bug
During the normal caching process, EXOSIMS calculates the integration cutoff dMag list and stores them in a cache file. However, these values are all being set to nan. Then, when calculating the integration cutoff time completeness, the required interpolation that uses the dMag values hits a ValueError.
To Reproduce
Use Nemati.py for the OpticalSystem and run a simple simulation. Can choose to warm the cache, this will be sufficient to reproduce the result.
Expected behavior
The expected behavior is to first set up the integration cutoff dMags to have the correct values and not nans. Then the interpolation in the next step will work.
JSON script:
See attached.
EXOSIMS version:
- 3.4.1 master
Additional context
The issue is circumvented by skipping the 'continue' in Nemati.py line 369. This will have the following effect:
The dMag is initially set to nan. Then we do not go back to the beginning of the loop but rather move on to the following 'else' statement, where the dMag value is overwritten with another value. Doing this makes the simulation complete with reasonable results, but is not intended to fix the core issue.
Cor1_ppf0.10_iwa1.5_ctmax0.4_alpha-6_rciwa1.00e-10_beta-7.00_gamma-3_AYO.json