public abstract class BaseGridSearch extends WeightLearningApplication
Modifier and Type | Field and Description |
---|---|
protected String |
currentLocation
The current location we are investigating.
|
protected int |
maxNumLocations
The number of actual possible locations.
|
protected int |
numLocations
The number of locations we will search.
|
protected Map<String,Double> |
objectives
The objectives at each location.
|
allRules, evaluator, inference, inMPEState, mutableRules, observedDB, rvDB, trainingMap
Constructor and Description |
---|
BaseGridSearch(List<Rule> rules,
Database rvDB,
Database observedDB) |
BaseGridSearch(Model model,
Database rvDB,
Database observedDB) |
Modifier and Type | Method and Description |
---|---|
protected abstract boolean |
chooseNextLocation()
Choose the next location we will search.
|
protected void |
doLearn()
Do the actual learning procedure.
|
protected abstract void |
getWeights(double[] weights)
Get the weight configuration at the current location.
|
protected double |
inspectLocation(double[] weights)
Inspect the location defined by the given weights and give back its score (lower is better).
|
close, computeMPEState, getInferenceApplication, getWLA, initGroundModel, initGroundModel, learn, postInitGroundModel, setBudget
protected String currentLocation
protected int maxNumLocations
protected int numLocations
protected void doLearn()
WeightLearningApplication
doLearn
in class WeightLearningApplication
protected double inspectLocation(double[] weights)
protected abstract void getWeights(double[] weights)
protected abstract boolean chooseNextLocation()
Copyright © 2020 University of California, Santa Cruz. All rights reserved.