Propagator
< back to parent (FitterEngine)
Description
The propagator handles the engine in charge of applying parameters to the samples. It owns the parameters, the samples, and the data loader.
Config options
Option | Type | Description | Default |
---|---|---|---|
fitSampleSetConfig | json | SampleSet config | |
parameterSetListConfig | json | ParameterSetList config | |
dataSetList | json | DatasetManager config | |
plotGeneratorConfig | json | PlotGenerator config | |
eventTreeWriter | json | EventTreeWriter config | |
showEventBreakdown | bool | Print sample total weight | true |
enableStatThrowInToys | bool | Throw statistical error with a poisson distribution | true |
enableEventMcThrow | bool | Each MC event get reweighted with Poisson(1) | true |
gaussStatThrowInToys | bool | Throw statistical error with a gaussian distribution instead | false |
throwAsimovFitParameters | bool | Throw parameters of MC before fit (used to test fitter convergence) | false |
reThrowParSetIfOutOfBounds | bool | If any thrown parameter of the set is out of bounds, throw again | true |
globalEventReweightCap | double | Will cap the weight applied by the parameters: evWeight = baseWeight * min(parWeight, cap) | nan |