A more updated version of the static probes DETF figure of merit. A previous version exists in MAF under summaryMetrics, but this needed to go on maf_contrib due to dependencies (I was not sure if pip-installing them is supposed to go on some requirements file in this repo or not?)
Description from old PR:
Metric now takes into account systematics (multiplicative shear bias, photometric redshift uncertainties, galaxy bias, intrinsic alignments, and baryonic physics effects). All of these have default values provided. In the future, some (e.g. photo-z uncertainties) might have their own MAF metrics, in which case this metric should be modified to take that as input.
Emulator now uses a gaussian process instead of a simple linear interpolation between the 6-D grid points at which FoM are calculated.
Metric can otherwise be run in the same way, the FoM values are much lower than the previous implementation due to the inclusion of systematics, but the relative trends should still be very similar (difference is below 2% on v1.4 OpSim runs.
Exactly this attempted PR to sims_maf https://github.com/lsst/sims_maf/pull/198
A more updated version of the static probes DETF figure of merit. A previous version exists in MAF under summaryMetrics, but this needed to go on maf_contrib due to dependencies (I was not sure if pip-installing them is supposed to go on some requirements file in this repo or not?)
Description from old PR:
Emulator now uses a gaussian process instead of a simple linear interpolation between the 6-D grid points at which FoM are calculated.
Metric can otherwise be run in the same way, the FoM values are much lower than the previous implementation due to the inclusion of systematics, but the relative trends should still be very similar (difference is below 2% on v1.4 OpSim runs.