search

LSSTDESC / descwl-shear-sims

simple simulations for testing weak lensing shear measurement
BSD 3-Clause "New" or "Revised" License
10 stars 9 forks source link

do not generate and return noise field with simulated exposures #141

Closed esheldon closed 3 years ago

esheldon commented 3 years ago

We are successfully generating these downstream now

Along with this, simplify the output to be a list of exposures per band, rather than a list of dicts with keys data and noise

beckermr commented 3 years ago

I'm confused by this. We are now adding sky backgrounds in these sims so that we can estimate noise from the images? Don't we need to handle units properly to get back to electrons or something?

esheldon commented 3 years ago

All I'm saying is that we will no longer return an example noise field from the sims, we require the processing codes to generate it.

I'm already doing this, I'm ignoring the noise field generated by the sim

beckermr commented 3 years ago

I'm already doing this, I'm ignoring the noise field generated by the sim

Right. I am asking how exactly this process works for both the sims and the real data. I guess you use the weight map or variance image?

esheldon commented 3 years ago

Yep; the gain is important if we need to subtract the poisson noise

beckermr commented 3 years ago

Right ok. That all makes sense.