RMS ellipticity versus other quantities

msimet / Stile

Stile: the Systematics Tests In Lensing pipeline

BSD 3-Clause "New" or "Revised" License

9 stars 6 forks source link

RMS ellipticity versus other quantities #68

Open msimet opened 9 years ago

msimet commented 9 years ago

We'd like to be able to do this as well, but I think there are some questions about how to implement this. As far as I can see there are three options:

As a subclass of ScatterPlot
As a new test type, RMSEllipticity (or something), which would do this test only
As a new test type, BinnedScatterPlot (or something), which would let you bin one quantity (with a user-defined function to give you a value for the bin--RMS for ellipticity, average for something else, etc) in bins based on a different quantity and plot the results

Any preferences or other ideas?

HironaoMiyatake commented 9 years ago

What we want to do is plotting e_rms and sqrt(e_rms^2 - e_err^2)? If so, I think a subclass of ScatterPlot works fine. If we want binning, we need BinnedScatterPlot.

HironaoMiyatake commented 9 years ago

Just a note from our discussion in email: we might be able to do a similar thing to make a plot of "Number density as a function of magnitude"

rmandelb commented 9 years ago

It seems to me that a new BinnedScatterPlot test would be handy for many things in addition to RMS ellipticities. For example, Hironao and others have been plotting the mean value of PSF model size - star size as a function of magnitude using their own software, but a BinnedScatterPlot class could do that for us.

msimet commented 9 years ago

All right; that was my feeling as well, so it sounds like a plan.