Manually adjusting SEG or EE will trigger numpy errors

[sebvetter]

In the Electron Extraction plugin we sample the number of extracted electrons according to a binomial distribution with a probability distribution according to the CY.

If we adjust the SE gain, or manually set the EE, this number can be arbitrary (and therefore no longer fit as a probability for the binomial) resulting in this numpy error:

File /opt/XENONnT/anaconda/envs/XENONnT_el7.2025.03.1/lib/python3.11/site-packages/fuse/plugins/detector_physics/electron_extraction.py:121, in ElectronExtraction.compute(self, interactions_in_roi)
118 else:
119 cy = self.electron_extraction_yield
--> 121 n_electron = self.rng.binomial(n=interactions_in_roi[mask]["n_electron_interface"], p=cy)
123 result = np.zeros(len(interactions_in_roi), dtype=self.dtype)
124 result["n_electron_extracted"][mask] = n_electron

File numpy/random/_generator.pyx:3089, in numpy.random._generator.Generator.binomial()

File _common.pyx:396, in numpy.random._common.check_array_constraint()

File _common.pyx:382, in numpy.random._common._check_array_cons_bounded_0_1()

ValueError: p < 0, p > 1 or p contains NaNs

An if + normalizing factor should solve this I believe.

[sebvetter]

I was misunderstanding the ext_eff_from_map flag. If set to True fuse will compute the EE from the SEG value (or map). To override the EE with your specific value, this flag has to be set to False.

This issue can be solved by clipping the CY variable to (0, 1), but I am not sure we want that. Given that this error will occur when someone misunderstands the g2/EE/SEG connection like I did.