org.grouplens.lenskit.eval.metrics.predict

Class HLUtilityPredictMetric

java.lang.Object

org.grouplens.lenskit.eval.metrics.AbstractMetric<TrainTestEvalTask>

org.grouplens.lenskit.eval.metrics.AbstractTestUserMetric

org.grouplens.lenskit.eval.metrics.predict.HLUtilityPredictMetric

All Implemented Interfaces:

Metric<TrainTestEvalTask>, TestUserMetric

public class HLUtilityPredictMetric
extends AbstractTestUserMetric

Nested Class Summary

Nested Classes

Modifier and Type	Class and Description
class	HLUtilityPredictMetric.Accum

Constructor Summary

Constructors

Constructor and Description
HLUtilityPredictMetric()
HLUtilityPredictMetric(double newAlpha)

Method Summary

Methods

Modifier and Type	Method and Description
java.util.List<java.lang.String>	getColumnLabels() Get labels for the aggregate columns output by this evaluator.
java.util.List<java.lang.String>	getUserColumnLabels() Get labels for the per-user columns output by this evaluator.
HLUtilityPredictMetric.Accum	makeAccumulator(Attributed algo, TTDataSet ds) Create a result accumulator for a single row for this evaluation.

Methods inherited from class org.grouplens.lenskit.eval.metrics.AbstractTestUserMetric

finalRow, userRow

Methods inherited from class org.grouplens.lenskit.eval.metrics.AbstractMetric

finishEvaluation, getCurrentEvaluation, startEvaluation

Methods inherited from class java.lang.Object

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

Methods inherited from interface org.grouplens.lenskit.eval.metrics.Metric

finishEvaluation, startEvaluation

Constructor Detail

HLUtilityPredictMetric

public HLUtilityPredictMetric(double newAlpha)

HLUtilityPredictMetric

public HLUtilityPredictMetric()

Method Detail

makeAccumulator

public HLUtilityPredictMetric.Accum makeAccumulator(Attributed algo,
                                           TTDataSet ds)

Description copied from interface: TestUserMetric

Create a result accumulator for a single row for this evaluation. The accumulator will be passed the predictions for each user in turn, then asked for the results from the evaluation to insert into the results table.

One accumulator is created and used per evaluation (data set × algorithmInfo).

Individual accumulators do not need to be thread-safe, but it must be possible to have multiple accumulators in separate threads in use concurrently.

Parameters:

algo - The algorithmInfo to be evaluated.

ds - The data set being evaluated — used if the evaluator needs something from it (such as the preference domain).

Returns:

The result accumulator for aggregating prediction results over a single evaluation.

getColumnLabels

public java.util.List<java.lang.String> getColumnLabels()

Description copied from interface: TestUserMetric

Get labels for the aggregate columns output by this evaluator.

Returns:

The labels for this evaluator's output, used as column headers when outputting the results table.

getUserColumnLabels

public java.util.List<java.lang.String> getUserColumnLabels()

Description copied from interface: TestUserMetric

Get labels for the per-user columns output by this evaluator.

Returns:

The labels for this evaluator's per-user output, used as column headers when outputting the results table.

See Also:

TestUserMetricAccumulator.evaluate(org.grouplens.lenskit.eval.traintest.TestUser)