search

sanssouci-org / sanssouci.python

Post hoc inference via multiple testing
GNU General Public License v3.0
6 stars 3 forks source link

Wrong assumption in calibrate_jer #32

Closed alexblnn closed 2 years ago

alexblnn commented 2 years ago

calibrate_jer assumes that there exists a template in learned_templates such that JER<alpha, which is not always the case. If it is not the case, the dichotomy needs to be aborted (and raise an error) and this is not the case right now.

pneuvial commented 2 years ago

Similarly, should there be a special treatment for the case where all templates in learned_templates are such that JER<alpha ? In that case the JER budget is not exhausted by the templates, it may be useful to warn users for this?