Propagator Configuration
< back to parent (likelihoodInterfaceConfig)
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 |
---|---|---|---|
sampleSetConfig | json | Configure the samples that will be compared data vs model | |
parametersManagerConfig/parameterSetList | json | Configure the parameters that will affect the model | |
parameterInjection/parameterSetList | list(json) | Override values for the parameters | |
showNbEventParameterBreakdown | bool | Print the number of event affected by each parameter | false |
showNbEventPerSampleParameterBreakdown | bool | Print the number of event per sample affected by each parameter | false |
globalEventReweightCap | double | Will cap the weight applied by the parameters: evWeight = baseWeight * min(parWeight, cap) | nan |
debugPrintLoadedEvents | bool | If true then print contents for a set of events. | false |
debugPrintLoadedEventsNbPerSample | int | Number of events to dump in the debugPrintLoadedEvents output | 5 |
devSingleThreadReweight | bool | Use a single thread for reweighting events with the CPU | false |
devSingleThreadHistFill | bool | Use a single thread for filling the histograms with the CPU | false |