edu.illinois.cs.cogcomp.lbjava

Class Train.TrainingThread

java.lang.Object

java.lang.Thread

edu.illinois.cs.cogcomp.lbjava.Train.TrainingThread

All Implemented Interfaces:

Runnable

Enclosing class:

Train

protected class Train.TrainingThread
extends Thread

This class contains the code that trains a learning classifier. It is a subclass of Thread so that it may be executed concurrently.

Author:

Nick Rizzolo

Nested Class Summary

Nested classes/interfaces inherited from class java.lang.Thread

Thread.State, Thread.UncaughtExceptionHandler

Field Summary

Fields

Modifier and Type	Field and Description
int	byteOffset The byte offset at which the learner appeared.
protected String	classDir The directory into which class files, model files, etc are written.
protected String	exFilePath The file into which training examples are extracted.
protected LearningClassifierExpression	lce The expression that specified the learner.
protected Learner	learner The learning classifier being trained.
protected Class	learnerClass The class of learner.
protected Class	parametersClass learner's Parameters class.
protected boolean	preExtract Whether or not example vectors should be pre-extracted.
protected boolean	preExtractZip Whether or not pre-extracted example files should be compressed.
protected String	testExFilePath The file into which testing examples are extracted.
protected TestingMetric	testingMetric The metric with which to measure the learner's performance on a test set.
protected Parser	testParser The parser from which testing objects are obtained.
protected BatchTrainer	trainer Actually does the training.

Fields inherited from class java.lang.Thread

MAX_PRIORITY, MIN_PRIORITY, NORM_PRIORITY

Constructor Summary

Constructors

Constructor and Description
TrainingThread(String n, int b, LearningClassifierExpression lce) Initializing constructor.

Method Summary

Modifier and Type	Method and Description
void	run() Performs the training and then generates the new code.

Methods inherited from class java.lang.Thread

activeCount, checkAccess, clone, countStackFrames, currentThread, destroy, dumpStack, enumerate, getAllStackTraces, getContextClassLoader, getDefaultUncaughtExceptionHandler, getId, getName, getPriority, getStackTrace, getState, getThreadGroup, getUncaughtExceptionHandler, holdsLock, interrupt, interrupted, isAlive, isDaemon, isInterrupted, join, join, join, resume, setContextClassLoader, setDaemon, setDefaultUncaughtExceptionHandler, setName, setPriority, setUncaughtExceptionHandler, sleep, sleep, start, stop, stop, suspend, toString, yield

Methods inherited from class java.lang.Object

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

Field Detail

byteOffset

public int byteOffset

The byte offset at which the learner appeared.

lce

protected LearningClassifierExpression lce

The expression that specified the learner.

learner

protected Learner learner

The learning classifier being trained.

learnerClass

protected Class learnerClass

The class of learner.

parametersClass

protected Class parametersClass

learner's Parameters class.

exFilePath

protected String exFilePath

The file into which training examples are extracted.

testExFilePath

protected String testExFilePath

The file into which testing examples are extracted.

classDir

protected String classDir

The directory into which class files, model files, etc are written.

preExtract

protected boolean preExtract

Whether or not example vectors should be pre-extracted.

preExtractZip

protected boolean preExtractZip

Whether or not pre-extracted example files should be compressed.

trainer

protected BatchTrainer trainer

Actually does the training.

testParser

protected Parser testParser

The parser from which testing objects are obtained.

testingMetric

protected TestingMetric testingMetric

The metric with which to measure the learner's performance on a test set.

Constructor Detail

TrainingThread

public TrainingThread(String n,
                      int b,
                      LearningClassifierExpression lce)

Initializing constructor.

Parameters:

n - The name of the learner.

b - The byte offset at which the learner appeared.

lce - The expression that specified the learner.

Method Detail

run

public void run()

Performs the training and then generates the new code.

Specified by:

run in interface Runnable

Overrides:

run in class Thread