ClassificationOverview Class Reference

Public Member Functions
def	__init__ (self, output_file_name, truth_name=None, select=[], exclude=[], groupbys=[], auxiliaries=[], filters=[])
def	train (self, input_tree)

Public Attributes
	output_file_name
	cached output filename
	truth_name
	cached truth name
	select
	cached selection-specifier array
	exclude
	cached exclusion-specifier array
	groupbys
	cached groupby-specifier array
	auxiliaries
	cached auxiliary-specifier array
	filters
	cached filter-specifier array
	classification_analyses
	array of classification analyses

Detailed Description

Class to generate overview plots for the classification power of various variables from a TTree.

In order to get an overview, which variables from a diverse set, generated from a recording filter
or some other sort of validation, perform well in classification task.

Definition at line 29 of file train.py.

Constructor & Destructor Documentation

__init__()

def __init__	(		self,
			output_file_name,
			truth_name = None,
			select = [],
			exclude = [],
			groupbys = [],
			auxiliaries = [],
			filters = []
	)

Constructor

Definition at line 37 of file train.py.

    def __init__(self,
                 output_file_name,
                 truth_name=None,
                 select=[],
                 exclude=[],
                 groupbys=[],
                 auxiliaries=[],
                 filters=[]):
        """Constructor"""
        
        self.output_file_name = output_file_name
        
        self.truth_name = truth_name
        
        self.select = select
        
        self.exclude = exclude
        
        self.groupbys = groupbys
        
        self.auxiliaries = auxiliaries
        
        self.filters = filters
 
        
        self.classification_analyses = []
 

Member Function Documentation

train()

def train	(		self,
			input_tree
	)

Main method feed with a TTree containing the truth variable and the variables to be investigated.

Branches that contain "truth" in the name are considered to directly contain information about
true classification target and are not analysed here.

Args:
    input_tree (ROOT.TTree) : Tree containing the variables to be investigated
as well as the classification target.
    truth_name (str, optional) : Name of the branch of the classification target.
If not given the Branch with the shortest name containing "truth" is selected.

Definition at line 64 of file train.py.

---

The documentation for this class was generated from the following file:

• tracking/trackFindingCDC/scripts/trackfindingcdc/mva/train.py