public class GaussianProcessPrior extends WeightLearningApplication
Modifier and Type | Class and Description |
---|---|
protected class |
GaussianProcessPrior.ValueAndStd |
protected class |
GaussianProcessPrior.WeightConfig |
Modifier and Type | Field and Description |
---|---|
static int |
MAX_RAND_INT_VAL |
static float |
SMALL_VALUE |
allRules, evaluator, inference, inMPEState, mutableRules, observedDB, rvDB, trainingMap
Constructor and Description |
---|
GaussianProcessPrior(List<Rule> rules,
Database rvDB,
Database observedDB) |
GaussianProcessPrior(Model model,
Database rvDB,
Database observedDB) |
Modifier and Type | Method and Description |
---|---|
protected void |
doLearn()
Do the actual learning procedure.
|
protected List<GaussianProcessPrior.WeightConfig> |
getConfigs() |
protected float |
getFunctionValue(GaussianProcessPrior.WeightConfig config) |
protected int |
getNextPoint(List<GaussianProcessPrior.WeightConfig> configs,
int iteration) |
protected GaussianProcessPrior.ValueAndStd |
predictFnValAndStd(float[] x,
List<GaussianProcessPrior.WeightConfig> xKnown)
predictFnValAndStd, but no memory sharing.
|
protected GaussianProcessPrior.ValueAndStd |
predictFnValAndStd(float[] x,
List<GaussianProcessPrior.WeightConfig> xKnown,
float[] xyStdData,
float[] kernelBuffer1,
float[] kernelBuffer2,
FloatMatrix kernelMatrixShell1,
FloatMatrix kernelMatrixShell2,
FloatMatrix xyStdMatrixShell,
FloatMatrix mulBuffer)
Do the prediction.
|
protected void |
setBlasYKnownForTest(FloatMatrix blasYKnown)
Only for testing.
|
protected void |
setKernelForTest(GaussianProcessKernel kernel)
Only for testing.
|
protected void |
setKnownDataStdInvForTest(FloatMatrix data)
Only for testing.
|
close, computeMPEState, getInferenceApplication, getWLA, initGroundModel, initGroundModel, learn, postInitGroundModel, setBudget
public static final int MAX_RAND_INT_VAL
public static final float SMALL_VALUE
public GaussianProcessPrior(List<Rule> rules, Database rvDB, Database observedDB)
protected void setKnownDataStdInvForTest(FloatMatrix data)
protected void setKernelForTest(GaussianProcessKernel kernel)
protected void setBlasYKnownForTest(FloatMatrix blasYKnown)
protected void doLearn()
WeightLearningApplication
doLearn
in class WeightLearningApplication
protected List<GaussianProcessPrior.WeightConfig> getConfigs()
protected GaussianProcessPrior.ValueAndStd predictFnValAndStd(float[] x, List<GaussianProcessPrior.WeightConfig> xKnown)
protected GaussianProcessPrior.ValueAndStd predictFnValAndStd(float[] x, List<GaussianProcessPrior.WeightConfig> xKnown, float[] xyStdData, float[] kernelBuffer1, float[] kernelBuffer2, FloatMatrix kernelMatrixShell1, FloatMatrix kernelMatrixShell2, FloatMatrix xyStdMatrixShell, FloatMatrix mulBuffer)
xyStdData
- A correctly-sized buffer to perform computations with.
Will get modified.protected float getFunctionValue(GaussianProcessPrior.WeightConfig config)
protected int getNextPoint(List<GaussianProcessPrior.WeightConfig> configs, int iteration)
Copyright © 2020 University of California, Santa Cruz. All rights reserved.