org.encog.engine
Class EncogEngine

java.lang.Object
  org.encog.engine.EncogEngine

public class EncogEngine
extends Object

The Encog Engine. The Engine forms the core of Encog's neural network calculation and training.

Field Summary

static double

DEFAULT_ZERO_TOLERANCE The default zero tolerance.

Constructor Summary

EncogEngine()

Method Summary

EncogCL	getCL()
static EncogEngine	getInstance() Get the instance to the singleton.
void	initCL() Enable OpenCL processing.
void	shutdown() Provides any shutdown that Encog may need.

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Field Detail

DEFAULT_ZERO_TOLERANCE

public static final double DEFAULT_ZERO_TOLERANCE

The default zero tolerance.

See Also:

Constant Field Values

Constructor Detail

EncogEngine

public EncogEngine()

Method Detail

getInstance

public static EncogEngine getInstance()

Get the instance to the singleton.

Returns:

The instance.

initCL

public void initCL()

Enable OpenCL processing. OpenCL processing allows Encog to use GPU devices to speed calculations. Not all areas of Encog can use this, however, GPU's can currently accelerate the training of Feedforward neural networks. To make use of the GPU you must have OpenCL drivers installed. For more information on getting OpenCL drivers, visit the following URL. http://www.heatonresearch.com/encog/opencl

shutdown

public void shutdown()

Provides any shutdown that Encog may need. Currently this shuts down the thread pool.

getCL

public EncogCL getCL()

Returns:

If Encog is not using GPU/CL processing this attribute will be null. Otherwise it holds the Encog CL object.