Source: EasyReflectometryApp/Backends/Py/logic/fitting.py:309-334 — DEEP_ANALYSIS.md §5 item 8
The method name says selected, but the body uses all experiments (_ordered_experiments()). As a result Bayesian sampling always runs over all experiments regardless of the user's experiment selection.
The classical path also fits all experiments (prepare_threaded_fit), so selection currently affects only plotting. The asymmetry between the name and behaviour should either be fixed (honour the selection) or the scope decision should be documented and the method renamed.
See also §8 recommendation 3 (decide and document fit/sampling scope vs experiment selection).
Source:
EasyReflectometryApp/Backends/Py/logic/fitting.py:309-334— DEEP_ANALYSIS.md §5 item 8The method name says selected, but the body uses all experiments (
_ordered_experiments()). As a result Bayesian sampling always runs over all experiments regardless of the user's experiment selection.The classical path also fits all experiments (
prepare_threaded_fit), so selection currently affects only plotting. The asymmetry between the name and behaviour should either be fixed (honour the selection) or the scope decision should be documented and the method renamed.See also §8 recommendation 3 (decide and document fit/sampling scope vs experiment selection).