batch_generator Class Reference

Public Member Functions
def	__init__ (self, X, Y, Z)
def	next_batch (self, batch_size)

Public Attributes
	X
	Input Features.
	Y
	Label data.
	Z
	Spectatirs/Quantity to be uncorrelated to.
	len
	Length of the data.
	index_array
	Index array containing indices from 0 to len.
	pointer
	Pointer to the current start of the batch.

Detailed Description

Class to create batches for training the Adversary Network.
Once the steps_per_epoch argument is available for the fit function in keras, this class will become obsolete.

Definition at line 38 of file adversary_network.py.

Constructor & Destructor Documentation

__init__()

def __init__	(		self,
			X,
			Y,
			Z
	)

Init the class
:param X: Input Features
:param Y: Label Data
:param Z: Spectators/Quantity to be uncorrelated to

Definition at line 44 of file adversary_network.py.

    def __init__(self, X, Y, Z):
        """
        Init the class
        :param X: Input Features
        :param Y: Label Data
        :param Z: Spectators/Quantity to be uncorrelated to
        """
        
        self.X = X
        
        self.Y = Y
        
        self.Z = Z
        
        self.len = len(Y)
        
        self.index_array = np.arange(self.len)
        
        self.pointer = 0
 

Member Function Documentation

next_batch()

def next_batch	(		self,
			batch_size
	)

Getting next batch of training data

Definition at line 64 of file adversary_network.py.

---

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

• mva/examples/keras/adversary_network.py