We are trying to track down what is causing an unusually large additive shear bias in our recent runs. To debug this, it would be useful to have a mode of running where we skip the PSF modeling step and instead render the true PSF used in image simulation creation into the MEDS file directly.
mcclearyj
commented
2 years ago
We are trying to track down what is causing an unusually large additive shear bias in our recent runs. To debug this, it would be useful to have a mode of running where we skip the PSF modeling step and instead render the true PSF used in image simulation creation into the MEDS file directly.
NOTE: Not ready for code review or merger yet.