Package org.knime.base.data.neural

This package contains all the necessary classes to build your own neural network.

See:
          Description

Class Summary
Architecture This class represents the general architecture of a neural network and specifies how much layers, neurons constitute the neural network.
HiddenLayer This class represents a hidden layer in a MultiLayerPerceptron.
InputLayer A special layer holding the input perceptrons.
InputPerceptron Class representing an input perceptron.
Layer An abstract class defining a layer in a Neural Network.
MultiLayerPerceptron Representation of a MultiLayer Perceptron, a neural net with one or more hidden layers.
Perceptron Abstract class defining the behaviour of a perceptron in a neural network.
SigmoidPerceptron A hidden layer perceptron with a sigmoid activation function.
 

Package org.knime.base.data.neural Description

This package contains all the necessary classes to build your own neural network.

MultiLayerPerceptron is a class that holds a neural network. Architecture defines the architecture of the network, e.g. the number of inputs, outputs. Perceptron is the abstract superclass for all perceptrons in a neural network. Layer is an abstrat superclass defining a layer in a neural network.



Copyright, 2003 - 2010. All rights reserved.
University of Konstanz, Germany.
Chair for Bioinformatics and Information Mining, Prof. Dr. Michael R. Berthold.
You may not modify, publish, transmit, transfer or sell, reproduce, create derivative works from, distribute, perform, display, or in any way exploit any of the content, in whole or in part, except as otherwise expressly permitted in writing by the copyright owner or as specified in the license file distributed with this product.