public class LabelSequenceEvaluator extends AbstractSequenceEvaluator<Label,LabelMetric.Context,LabelSequenceEvaluation,LabelMetric>
Constructor and Description |
---|
LabelSequenceEvaluator() |
Modifier and Type | Method and Description |
---|---|
protected LabelMetric.Context |
createContext(SequenceModel<Label> model,
List<List<Prediction<Label>>> predictions)
Create the context needed for evaluation.
|
protected LabelSequenceEvaluation |
createEvaluation(LabelMetric.Context ctx,
Map<MetricID<Label>,Double> results,
EvaluationProvenance provenance)
Create an evaluation for the given results
|
protected Set<LabelMetric> |
createMetrics(SequenceModel<Label> model)
Creates the appropriate set of metrics for this model, by querying for it's
OutputInfo . |
computeResults, evaluate, evaluate, evaluate
protected Set<LabelMetric> createMetrics(SequenceModel<Label> model)
AbstractSequenceEvaluator
OutputInfo
.createMetrics
in class AbstractSequenceEvaluator<Label,LabelMetric.Context,LabelSequenceEvaluation,LabelMetric>
model
- The model to inspect.protected LabelMetric.Context createContext(SequenceModel<Label> model, List<List<Prediction<Label>>> predictions)
AbstractSequenceEvaluator
createContext
in class AbstractSequenceEvaluator<Label,LabelMetric.Context,LabelSequenceEvaluation,LabelMetric>
model
- the model that will be evaluatedpredictions
- the predictions that will be evaluatedprotected LabelSequenceEvaluation createEvaluation(LabelMetric.Context ctx, Map<MetricID<Label>,Double> results, EvaluationProvenance provenance)
AbstractSequenceEvaluator
createEvaluation
in class AbstractSequenceEvaluator<Label,LabelMetric.Context,LabelSequenceEvaluation,LabelMetric>
ctx
- the context that was used to compute these resultsresults
- the resultsprovenance
- the provenance of the results (including information about the model and dataset)Copyright © 2015–2021 Oracle and/or its affiliates. All rights reserved.