fast_equal_frequency_binning Class Reference

Public Member Functions
	__init__ (self, state=None)
	fit (self, x, number_of_bins=100)
	apply (self, x)
	export_state (self)

Public Attributes
dict	state = {'binning_array': [], 'number_of_bins': 0}
	State of the class.

Detailed Description

This class provides a fast implementation of equal frequency binning.
In Equal frequency binning the binning is chosen in a way that every bin has the same number of entries.
An example with a Neural Network can be found in: mva/examples/keras/preprocessing.py

Definition at line 14 of file preprocessing.py.

Constructor & Destructor Documentation

__init__()

__init__	(		self,
			state = None )

Init the class.
If you saved a state before and wants to rebuild the class use the state parameter.

Definition at line 21 of file preprocessing.py.

    def __init__(self, state=None):
        """
        Init the class.
        If you saved a state before and wants to rebuild the class use the state parameter.
        """
        if state is None:
            
            self.state = {'binning_array': [], 'number_of_bins': 0}
        else:
            self.state = state
 

Member Function Documentation

apply()

apply	(		self,
			x )

Bin a dataset

Definition at line 41 of file preprocessing.py.

    def apply(self, x):
        """
        Bin a dataset
        """
        for variable in range(len(x[0, :])):
            x[:, variable] = np.digitize(np.nan_to_num(x[:, variable]),
                                         self.state['binning_array'][variable][1:-1]) / self.state['number_of_bins']
        return x
 

export_state()

export_state

(

self

)

Returns a pickable dictionary to save the state of the class in a mva weightfile

Definition at line 50 of file preprocessing.py.

    def export_state(self):
        """
        Returns a pickable dictionary to save the state of the class in a mva weightfile
        """
        return self.state

fit()

fit	(		self,
			x,
			number_of_bins = 100 )

Do the fitting -> calculate binning boundaries

Definition at line 32 of file preprocessing.py.

    def fit(self, x, number_of_bins=100):
        """
        Do the fitting -> calculate binning boundaries
        """
        for variable in range(len(x[0, :])):
            self.state['binning_array'].append(np.percentile(np.nan_to_num(x[:, variable]),
                                                             np.linspace(0, 100, number_of_bins + 1)))
        self.state['number_of_bins'] = number_of_bins
 

Member Data Documentation

state

dict state = {'binning_array': [], 'number_of_bins': 0}

State of the class.

This will be saved

Definition at line 28 of file preprocessing.py.

---

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

• mva/scripts/basf2_mva_extensions/preprocessing.py