org.knime.base.node.mine.neural.rprop
Class RPropNodeModel

java.lang.Object
  extended by org.knime.core.node.NodeModel
      extended by org.knime.base.node.mine.neural.rprop.RPropNodeModel

public class RPropNodeModel
extends NodeModel

RPropNodeModel trains a MultiLayerPerceptron with resilient backpropagation.

Author:
Nicolas Cebron, University of Konstanz

Field Summary
static String CLASSCOL_KEY
          Key to store the class column.
static int DEFAULTHIDDENLAYERS
          The default number of iterations.
static int DEFAULTITERATIONS
          The default number of iterations.
static int DEFAULTNEURONSPERLAYER
          The default number of iterations.
static String HIDDENLAYER_KEY
          Key to store the number of hidden layer.
static String IGNOREMV_KEY
          Key to store whether missing values should be ignoted.
static int INPORT
          Inport of the NodeModel for the examples.
static String MAXITER_KEY
          Key to store the number of maximum iterations.
static int MAXNRITERATIONS
          The maximum number of possible iterations.
static String NRHNEURONS_KEY
          Key to store the number of neurons per hidden layer.
 
Constructor Summary
RPropNodeModel()
          The RPropNodeModel has 2 inputs, one for the positive examples and one for the negative ones.
 
Method Summary
protected  PortObjectSpec[] configure(PortObjectSpec[] inSpecs)
          returns null.
protected  PortObject[] execute(PortObject[] inData, ExecutionContext exec)
          The execution consists of three steps: A neural network is build with the inputs and outputs according to the input datatable, number of hidden layers as specified. Input DataTables are converted into double-arrays so they can be attached to the neural net. The neural net is trained. Execute method for general port types.
 double[] getErrors()
           
protected  void loadInternals(File internDir, ExecutionMonitor exec)
          Load internals into the derived NodeModel.
protected  void loadValidatedSettingsFrom(NodeSettingsRO settings)
          Sets new settings from the passed object in the model.
protected  void reset()
          Override this function in the derived model and reset your NodeModel.
protected  void saveInternals(File internDir, ExecutionMonitor exec)
          Save internals of the derived NodeModel.
protected  void saveSettingsTo(NodeSettingsWO settings)
          Adds to the given NodeSettings the model specific settings.
protected  void validateSettings(NodeSettingsRO settings)
          Validates the settings in the passed NodeSettings object.
 
Methods inherited from class org.knime.core.node.NodeModel
addWarningListener, configure, continueLoop, execute, executeModel, getInHiLiteHandler, getLoopEndNode, getLoopStartNode, getNrInPorts, getNrOutPorts, getOutHiLiteHandler, getWarningMessage, notifyViews, notifyWarningListeners, peekFlowVariableDouble, peekFlowVariableInt, peekFlowVariableString, peekScopeVariableDouble, peekScopeVariableInt, peekScopeVariableString, pushFlowVariableDouble, pushFlowVariableInt, pushFlowVariableString, pushScopeVariableDouble, pushScopeVariableInt, pushScopeVariableString, removeWarningListener, setInHiLiteHandler, setWarningMessage, stateChanged
 
Methods inherited from class java.lang.Object
clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait
 

Field Detail

INPORT

public static final int INPORT
Inport of the NodeModel for the examples.

See Also:
Constant Field Values

MAXNRITERATIONS

public static final int MAXNRITERATIONS
The maximum number of possible iterations.

See Also:
Constant Field Values

DEFAULTITERATIONS

public static final int DEFAULTITERATIONS
The default number of iterations.

See Also:
Constant Field Values

DEFAULTHIDDENLAYERS

public static final int DEFAULTHIDDENLAYERS
The default number of iterations.

See Also:
Constant Field Values

DEFAULTNEURONSPERLAYER

public static final int DEFAULTNEURONSPERLAYER
The default number of iterations.

See Also:
Constant Field Values

MAXITER_KEY

public static final String MAXITER_KEY
Key to store the number of maximum iterations.

See Also:
Constant Field Values

IGNOREMV_KEY

public static final String IGNOREMV_KEY
Key to store whether missing values should be ignoted.

See Also:
Constant Field Values

HIDDENLAYER_KEY

public static final String HIDDENLAYER_KEY
Key to store the number of hidden layer.

See Also:
Constant Field Values

NRHNEURONS_KEY

public static final String NRHNEURONS_KEY
Key to store the number of neurons per hidden layer.

See Also:
Constant Field Values

CLASSCOL_KEY

public static final String CLASSCOL_KEY
Key to store the class column.

See Also:
Constant Field Values
Constructor Detail

RPropNodeModel

public RPropNodeModel()
The RPropNodeModel has 2 inputs, one for the positive examples and one for the negative ones. The output is the model of the constructed and trained neural network.

Method Detail

configure

protected PortObjectSpec[] configure(PortObjectSpec[] inSpecs)
                              throws InvalidSettingsException
returns null. Configure method for general port types. The argument specs represent the input object specs and are guaranteed to be subclasses of the PortObjectSpecs that are defined through the PortTypes given in the constructor. Similarly, the returned output specs need to comply with their port types spec class (otherwise an error is reported by the framework). They may also be null.

For a general description of the configure method refer to the description of the specialized NodeModel.configure(DataTableSpec[]) methods as it addresses more use cases.

Overrides:
configure in class NodeModel
Parameters:
inSpecs - The input object specs.
Returns:
The output objects specs or null.
Throws:
InvalidSettingsException - If this node can't be configured.

execute

protected PortObject[] execute(PortObject[] inData,
                               ExecutionContext exec)
                        throws Exception
The execution consists of three steps:
  1. A neural network is build with the inputs and outputs according to the input datatable, number of hidden layers as specified.
  2. Input DataTables are converted into double-arrays so they can be attached to the neural net.
  3. The neural net is trained.
Execute method for general port types. The argument objects represent the input objects and are guaranteed to be subclasses of the PortObject classes that are defined through the PortTypes given in the constructor. Similarly, the returned output objects need to comply with their port types object class (otherwise an error is reported by the framework).

For a general description of the execute method refer to the description of the specialized NodeModel.execute(BufferedDataTable[], ExecutionContext) methods as it addresses more use cases.

Overrides:
execute in class NodeModel
Parameters:
inData - The input objects.
exec - For BufferedDataTable creation and progress.
Returns:
The output objects.
Throws:
Exception - If the node execution fails for any reason.

reset

protected void reset()
Override this function in the derived model and reset your NodeModel. All components should unregister themselves from any observables (at least from the hilite handler right now). All internally stored data structures should be released. User settings should not be deleted/reset though.

Specified by:
reset in class NodeModel

saveSettingsTo

protected void saveSettingsTo(NodeSettingsWO settings)
Adds to the given NodeSettings the model specific settings. The settings don't need to be complete or consistent. If, right after startup, no valid settings are available this method can write either nothing or invalid settings.

Method is called by the Node if the current settings need to be saved or transfered to the node's dialog.

Specified by:
saveSettingsTo in class NodeModel
Parameters:
settings - The object to write settings into.
See Also:
NodeModel.loadValidatedSettingsFrom(NodeSettingsRO), NodeModel.validateSettings(NodeSettingsRO)

validateSettings

protected void validateSettings(NodeSettingsRO settings)
                         throws InvalidSettingsException
Validates the settings in the passed NodeSettings object. The specified settings should be checked for completeness and consistency. It must be possible to load a settings object validated here without any exception in the #loadValidatedSettings(NodeSettings) method. The method must not change the current settings in the model - it is supposed to just check them. If some settings are missing, invalid, inconsistent, or just not right throw an exception with a message useful to the user.

Specified by:
validateSettings in class NodeModel
Parameters:
settings - The settings to validate.
Throws:
InvalidSettingsException - If the validation of the settings failed.
See Also:
NodeModel.saveSettingsTo(NodeSettingsWO), NodeModel.loadValidatedSettingsFrom(NodeSettingsRO)

loadValidatedSettingsFrom

protected void loadValidatedSettingsFrom(NodeSettingsRO settings)
                                  throws InvalidSettingsException
Sets new settings from the passed object in the model. You can safely assume that the object passed has been successfully validated by the #validateSettings(NodeSettings) method. The model must set its internal configuration according to the settings object passed.

Specified by:
loadValidatedSettingsFrom in class NodeModel
Parameters:
settings - The settings to read.
Throws:
InvalidSettingsException - If a property is not available.
See Also:
NodeModel.saveSettingsTo(NodeSettingsWO), NodeModel.validateSettings(NodeSettingsRO)

getErrors

public double[] getErrors()
Returns:
error plot.

loadInternals

protected void loadInternals(File internDir,
                             ExecutionMonitor exec)
                      throws IOException
Load internals into the derived NodeModel. This method is only called if the Node was executed. Read all your internal structures from the given file directory to create your internal data structure which is necessary to provide all node functionalities after the workflow is loaded, e.g. view content and/or hilite mapping.

Specified by:
loadInternals in class NodeModel
Parameters:
internDir - The directory to read from.
exec - Used to report progress and to cancel the load process.
Throws:
IOException - If an error occurs during reading from this dir.
See Also:
NodeModel.saveInternals(File,ExecutionMonitor)

saveInternals

protected void saveInternals(File internDir,
                             ExecutionMonitor exec)
                      throws IOException
Save internals of the derived NodeModel. This method is only called if the Node is executed. Write all your internal structures into the given file directory which are necessary to recreate this model when the workflow is loaded, e.g. view content and/or hilite mapping.

Specified by:
saveInternals in class NodeModel
Parameters:
internDir - The directory to write into.
exec - Used to report progress and to cancel the save process.
Throws:
IOException - If an error occurs during writing to this dir.
See Also:
NodeModel.loadInternals(File,ExecutionMonitor)


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.