Package org.apache.solr.ltr.feature

Class Feature.FeatureWeight

java.lang.Object

org.apache.lucene.search.Weight

org.apache.solr.ltr.feature.Feature.FeatureWeight

All Implemented Interfaces:

org.apache.lucene.search.SegmentCacheable

Direct Known Subclasses:

FieldLengthFeature.FieldLengthFeatureWeight, FieldValueFeature.FieldValueFeatureWeight, OriginalScoreFeature.OriginalScoreWeight, SolrFeature.SolrFeatureWeight, ValueFeature.ValueFeatureWeight

Enclosing class:

Feature

public abstract class Feature.FeatureWeight
extends org.apache.lucene.search.Weight

Weight for a feature

Nested Class Summary

Nested Classes

Modifier and Type	Class	Description
static class	Feature.FeatureWeight.FeatureScorer	A 'recipe' for computing a feature
static class	Feature.FeatureWeight.FilterFeatureScorer	A FeatureScorer that contains a Scorer, which it delegates to where appropriate.
static class	Feature.FeatureWeight.ValueFeatureScorer	Default FeatureScorer class that returns the score passed in.

Nested classes/interfaces inherited from class org.apache.lucene.search.Weight

org.apache.lucene.search.Weight.DefaultBulkScorer, org.apache.lucene.search.Weight.DefaultScorerSupplier

Field Summary

Fields

Modifier and Type	Field	Description
protected final Map<String,String[]>	efi
protected final MacroExpander	macroExpander
protected final org.apache.lucene.search.Query	originalQuery
protected final SolrQueryRequest	request
protected final org.apache.lucene.search.IndexSearcher	searcher

Fields inherited from class org.apache.lucene.search.Weight

parentQuery

Constructor Summary

Constructors

Constructor	Description
FeatureWeight(org.apache.lucene.search.Query q, org.apache.lucene.search.IndexSearcher searcher, SolrQueryRequest request, org.apache.lucene.search.Query originalQuery, Map<String,String[]> efi)	Initialize a feature without the normalizer from the feature file.

Method Summary

Modifier and Type	Method	Description
org.apache.lucene.search.Explanation	explain(org.apache.lucene.index.LeafReaderContext context, int doc)
abstract Feature.FeatureWeight.FeatureScorer	featureScorer(org.apache.lucene.index.LeafReaderContext context)
float	getDefaultValue()
int	getIndex()
String	getName()
boolean	isCacheable(org.apache.lucene.index.LeafReaderContext ctx)
org.apache.lucene.search.ScorerSupplier	scorerSupplier(org.apache.lucene.index.LeafReaderContext context)
String	toString()	Used in the FeatureWeight's explain.

Methods inherited from class org.apache.lucene.search.Weight

bulkScorer, count, getQuery, matches, scorer

Methods inherited from class java.lang.Object

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

Field Details

searcher

protected final org.apache.lucene.search.IndexSearcher searcher

request

protected final SolrQueryRequest request

efi

protected final Map<String,String[]> efi

macroExpander

protected final MacroExpander macroExpander

originalQuery

protected final org.apache.lucene.search.Query originalQuery

Constructor Details

FeatureWeight

public FeatureWeight(org.apache.lucene.search.Query q,
 org.apache.lucene.search.IndexSearcher searcher,
 SolrQueryRequest request,
 org.apache.lucene.search.Query originalQuery,
 Map<String,String[]> efi)

Initialize a feature without the normalizer from the feature file. This is called on initial construction since multiple models share the same features, but have different normalizers. A concrete model's feature is copied through featForNewModel().

Parameters:

q - Solr query associated with this FeatureWeight

searcher - Solr searcher available for features if they need them

Method Details

getName

public String getName()

getIndex

public int getIndex()

getDefaultValue

public float getDefaultValue()

featureScorer

public abstract Feature.FeatureWeight.FeatureScorer featureScorer(org.apache.lucene.index.LeafReaderContext context)
                                                           throws IOException

Throws:

IOException

scorerSupplier

public org.apache.lucene.search.ScorerSupplier scorerSupplier(org.apache.lucene.index.LeafReaderContext context)
                                                       throws IOException

Specified by:

scorerSupplier in class org.apache.lucene.search.Weight

Throws:

IOException

isCacheable

public boolean isCacheable(org.apache.lucene.index.LeafReaderContext ctx)

explain

public org.apache.lucene.search.Explanation explain(org.apache.lucene.index.LeafReaderContext context,
 int doc)
                                             throws IOException

Specified by:

explain in class org.apache.lucene.search.Weight

Throws:

IOException

toString

public String toString()

Used in the FeatureWeight's explain. Each feature should implement this returning properties of the specific scorer useful for an explain. For example "MyCustomClassFeature [name=" + name + "myVariable:" + myVariable + "]"; If not provided, a default implementation will return basic feature properties, which might not include query time specific values.

Overrides:

toString in class Object