Spatial template use shows template bin structure now in the simulation

[CristianAntochi]

Hello all,
Back in February the use of the spatial templates for spatial dependent sources was a bit more complicated than now, but the FD simulation of the source using the template did not show bin structures (say r, binned uniformly in r^2), plot for reference:

Now it does!

[JelleAalbers]

Interesting Cristian. Indeed, your two figures are clearly different: the second seems to have way more events overall, especially at higher r, and fewer events at lower r. The higher density causes the bin structure to appear more prominent -- whether that's the only reason why the bin structure appears more prominent or whether there is something else going on as well is hard to say.

The change you're referring to between February and now is #100. As described there, in February, you had to pass some funnily normalized histogram that was neither a PDF nor the expected events/bin. Since March, you have to pass an events/bin histogram.

There are fifteen possible explanations for the difference you're seeing, corresponding to the yes/no answers to the four questions below -- except 'yes to all', which is ruled out because the plots are different.

1. Did you give flamedisx the template correctly in February?
2. Are you giving flamedisx the template correctly now?
3. Did flamedisx work correctly in February?
4. Does flamedisx work correctly now?

So I don't think we can say there is a flamedisx bug now just yet. It's possible, but so are other explanations. For example, if you are giving exactly the same template to your code without changing your normalization, the answer to either 1 or 2 is 'No' because the convention on what to pass has changed (see above).

You probably already saw the notebook at https://github.com/FlamTeam/flamedisx-notebooks/blob/master/spatial_variation_demo.ipynb, but if not, that might help.