keraflow.layers.core.Dropout Class Reference

Applies Dropout to the input. More...

Inheritance diagram for keraflow.layers.core.Dropout:

Public Member Functions
def	__init__

Detailed Description

Applies Dropout to the input.

In training phase, for each unit in the input tensor, Dropout sets it to 0 with probability drop_rate and sets it to 1/(1-drop_rate) times with probability 1-drop_rate. In testing phase, Dropout simply returns the input tensor.

• input_shape: nD, (nb_samples, ...)
• output_shape: nD, (nb_samples, ...)
• Reference: Dropout: A Simple Way to Prevent Neural Networks from Overfitting

Constructor & Destructor Documentation

def keraflow.layers.core.Dropout.__init__	(		self,
			drop_rate,
			kwargs
	)

Parameters

drop_rate	float between 0 and 1. Fraction of the input units to drop.
kwargs	see Layer.__init__.