TrainKsSelector.py

#!/usr/bin/env python3
 

 

 
import basf2 as b2
import basf2_mva
import ksSelector as ksSelector
 
# Train weight file for V0 Selection.
# name of output file is specified by the argument 'mva_identifier'.
ksSelector.V0Selector_Training(
    train_data=b2.find_file('KsSelector_train_V0Selector.root', 'examples', False),
    tree_name="tree",
    mva_identifier="MVAFastBDT_V0Selector.root",
    target_variable="isSignal",
    options={'max_events': 1000}  # number of training events manually reduced to speed up test, set to 0 for full statistics
)
 
# Train weight file for V0 Selection.
# name of output file is specified by the argument 'mva_identifier'.
ksSelector.LambdaVeto_Training(
    train_data=b2.find_file('KsSelector_train_LambdaVeto.root', 'examples', False),
    tree_name="tree",
    mva_identifier="MVAFastBDT_LambdaVeto.root",
    target_variable="isSignal",
    options={'max_events': 100}  # number of training events manually reduced to speed up test, set to 0 for full statistics
)
 
# apply mva to the training data
basf2_mva.expert(basf2_mva.vector('MVAFastBDT_V0Selector.root'),
                 basf2_mva.vector(b2.find_file('KsSelector_train_V0Selector.root', 'examples', False)),
                 'tree', 'MVAExpert_V0Selector.root')
basf2_mva.expert(basf2_mva.vector('MVAFastBDT_LambdaVeto.root'),
                 basf2_mva.vector(b2.find_file('KsSelector_train_LambdaVeto.root', 'examples', False)),
                 'tree', 'MVAExpert_LambdaVeto.root')