support default values for query featureshavardpe/extract-default-query-feature-values

4 files changed, 50 insertions, 42 deletions

diff --git a/container-search/src/main/java/com/yahoo/search/ranking/GlobalPhaseRanker.java b/container-search/src/main/java/com/yahoo/search/ranking/GlobalPhaseRanker.java
index 829d0c268e5..91acc883803 100644
--- a/container-search/src/main/java/com/yahoo/search/ranking/GlobalPhaseRanker.java
+++ b/container-search/src/main/java/com/yahoo/search/ranking/GlobalPhaseRanker.java
@@ -13,10 +13,7 @@ import com.yahoo.tensor.Tensor;
 import com.yahoo.data.access.helpers.MatchFeatureData;
 import com.yahoo.data.access.helpers.MatchFeatureFilter;
 
-import java.util.ArrayList;
-import java.util.Collection;
-import java.util.List;
-import java.util.Optional;
+import java.util.*;
 import java.util.function.Supplier;
 import java.util.logging.Logger;
 
@@ -52,12 +49,12 @@ public class GlobalPhaseRanker {
 
     static void rerankHitsImpl(GlobalPhaseSetup setup, Query query, Result result) {
         var mainSpec = setup.globalPhaseEvalSpec;
-        var mainSrc = withQueryPrep(mainSpec.evalSource(), mainSpec.fromQuery(), query);
+        var mainSrc = withQueryPrep(mainSpec.evalSource(), mainSpec.fromQuery(), setup.defaultValues, query);
         int rerankCount = resolveRerankCount(setup, query);
         var normalizers = new ArrayList<NormalizerContext>();
         for (var nSetup : setup.normalizers) {
             var normSpec = nSetup.inputEvalSpec();
-            var normEvalSrc = withQueryPrep(normSpec.evalSource(), normSpec.fromQuery(), query);
+            var normEvalSrc = withQueryPrep(normSpec.evalSource(), normSpec.fromQuery(), setup.defaultValues, query);
             normalizers.add(new NormalizerContext(nSetup.name(), nSetup.supplier().get(), normEvalSrc, normSpec.fromMF()));
         }
         var rescorer = new HitRescorer(mainSrc, mainSpec.fromMF(), normalizers);
@@ -73,8 +70,8 @@ public class GlobalPhaseRanker {
         }
     }
 
-    static Supplier<Evaluator> withQueryPrep(Supplier<Evaluator> evalSource, List<String> queryFeatures, Query query) {
-        var prepared = PreparedInput.findFromQuery(query, queryFeatures);
+    static Supplier<Evaluator> withQueryPrep(Supplier<Evaluator> evalSource, List<String> queryFeatures, Map<String, Tensor> defaultValues, Query query) {
+        var prepared = PreparedInput.findFromQuery(query, queryFeatures, defaultValues);
         Supplier<Evaluator> supplier = () -> {
             var evaluator = evalSource.get();
             for (var entry : prepared) {
diff --git a/container-search/src/main/java/com/yahoo/search/ranking/GlobalPhaseSetup.java b/container-search/src/main/java/com/yahoo/search/ranking/GlobalPhaseSetup.java
index 31a676e4c8e..e5cd09d3a18 100644
--- a/container-search/src/main/java/com/yahoo/search/ranking/GlobalPhaseSetup.java
+++ b/container-search/src/main/java/com/yahoo/search/ranking/GlobalPhaseSetup.java
@@ -3,15 +3,10 @@ package com.yahoo.search.ranking;
 
 import ai.vespa.models.evaluation.FunctionEvaluator;
 
+import com.yahoo.tensor.Tensor;
 import com.yahoo.vespa.config.search.RankProfilesConfig;
 
-import java.util.ArrayList;
-import java.util.Collection;
-import java.util.HashSet;
-import java.util.List;
-import java.util.Set;
-import java.util.Map;
-import java.util.HashMap;
+import java.util.*;
 import java.util.function.Supplier;
 
 class GlobalPhaseSetup {
@@ -20,16 +15,19 @@ class GlobalPhaseSetup {
     final int rerankCount;
     final Collection<String> matchFeaturesToHide;
     final List<NormalizerSetup> normalizers;
+    final Map<String, Tensor> defaultValues;
 
     GlobalPhaseSetup(FunEvalSpec globalPhaseEvalSpec,
                      final int rerankCount,
                      Collection<String> matchFeaturesToHide,
-                     List<NormalizerSetup> normalizers)
+                     List<NormalizerSetup> normalizers,
+                     Map<String, Tensor> defaultValues)
     {
         this.globalPhaseEvalSpec = globalPhaseEvalSpec;
         this.rerankCount = rerankCount;
         this.matchFeaturesToHide = matchFeaturesToHide;
         this.normalizers = normalizers;
+        this.defaultValues = defaultValues;
     }
 
     static GlobalPhaseSetup maybeMakeSetup(RankProfilesConfig.Rankprofile rp, RankProfilesEvaluator modelEvaluator) {
@@ -106,7 +104,7 @@ class GlobalPhaseSetup {
             }
             Supplier<Evaluator> supplier = SimpleEvaluator.wrap(functionEvaluatorSource);
             var gfun = new FunEvalSpec(supplier, fromQuery, fromMF);
-            return new GlobalPhaseSetup(gfun, rerankCount, namesToHide, normalizers);
+            return new GlobalPhaseSetup(gfun, rerankCount, namesToHide, normalizers, Collections.emptyMap());
         }
         return null;
     }
diff --git a/container-search/src/main/java/com/yahoo/search/ranking/PreparedInput.java b/container-search/src/main/java/com/yahoo/search/ranking/PreparedInput.java
index 5491724cc08..914635fef59 100644
--- a/container-search/src/main/java/com/yahoo/search/ranking/PreparedInput.java
+++ b/container-search/src/main/java/com/yahoo/search/ranking/PreparedInput.java
@@ -13,16 +13,13 @@ import com.yahoo.tensor.Tensor;
 import com.yahoo.data.access.helpers.MatchFeatureData;
 import com.yahoo.data.access.helpers.MatchFeatureFilter;
 
-import java.util.ArrayList;
-import java.util.Collection;
-import java.util.List;
-import java.util.Optional;
+import java.util.*;
 import java.util.function.Supplier;
 import java.util.logging.Logger;
 
 record PreparedInput(String name, Tensor value) {
 
-    static List<PreparedInput> findFromQuery(Query query, Collection<String> queryFeatures) {
+    static List<PreparedInput> findFromQuery(Query query, Collection<String> queryFeatures, Map<String, Tensor> defaultValues) {
         List<PreparedInput> result = new ArrayList<>();
         var ranking = query.getRanking();
         var rankFeatures = ranking.getFeatures();
@@ -36,6 +33,12 @@ record PreparedInput(String name, Tensor value) {
                 feature = rankFeatures.getTensor(needed);
             }
             if (feature.isEmpty()) {
+                var t = defaultValues.get(needed);
+                if (t != null) {
+                    feature = Optional.of(t);
+                }
+            }
+            if (feature.isEmpty()) {
                 throw new IllegalArgumentException("missing query feature: " + queryFeatureName);
             }
             result.add(new PreparedInput(needed, feature.get()));
diff --git a/container-search/src/test/java/com/yahoo/search/ranking/GlobalPhaseRerankHitsImplTest.java b/container-search/src/test/java/com/yahoo/search/ranking/GlobalPhaseRerankHitsImplTest.java
index ce9ac377908..f55130c0c93 100644
--- a/container-search/src/test/java/com/yahoo/search/ranking/GlobalPhaseRerankHitsImplTest.java
+++ b/container-search/src/test/java/com/yahoo/search/ranking/GlobalPhaseRerankHitsImplTest.java
@@ -60,17 +60,20 @@ public class GlobalPhaseRerankHitsImplTest {
     static NormalizerSetup makeNormalizer(String name, List<Double> expected, FunEvalSpec evalSpec) {
         return new NormalizerSetup(name, () -> new ExpectingNormalizer(expected), evalSpec);
     }
-    static GlobalPhaseSetup makeFullSetup(FunEvalSpec mainSpec, int rerankCount,
-                                          List<String> hiddenMF, List<NormalizerSetup> normalizers)
-    {
-        return new GlobalPhaseSetup(mainSpec, rerankCount, hiddenMF, normalizers);
-    }
-    static GlobalPhaseSetup makeSimpleSetup(FunEvalSpec mainSpec, int rerankCount) {
-        return makeFullSetup(mainSpec, rerankCount, Collections.emptyList(), Collections.emptyList());
-    }
-    static GlobalPhaseSetup makeNormSetup(FunEvalSpec mainSpec, List<NormalizerSetup> normalizers) {
-        return makeFullSetup(mainSpec, 100, Collections.emptyList(), normalizers);
-    }
+    static class SetupBuilder {
+        FunEvalSpec mainSpec = makeConstSpec(0.0);
+        int rerankCount = 100;
+        List<String> hiddenMF = new ArrayList<>();
+        List<NormalizerSetup> normalizers = new ArrayList<>();
+        Map<String, Tensor> defaultValues = new HashMap<>();
+        SetupBuilder eval(FunEvalSpec spec) { mainSpec = spec; return this; }
+        SetupBuilder rerank(int value) { rerankCount = value; return this; }
+        SetupBuilder hide(String mf) { hiddenMF.add(mf); return this; }
+        SetupBuilder addNormalizer(NormalizerSetup normalizer) { normalizers.add(normalizer); return this; }
+        SetupBuilder addDefault(String name, Tensor value) { defaultValues.put(name, value); return this; }
+        GlobalPhaseSetup build() { return new GlobalPhaseSetup(mainSpec, rerankCount, hiddenMF, normalizers, defaultValues); }
+    }
+    static SetupBuilder setup() { return new SetupBuilder(); }
     static record NamedValue(String name, double value) {}
     NamedValue value(String name, double value) {
         return new NamedValue(name, value);
@@ -167,7 +170,7 @@ public class GlobalPhaseRerankHitsImplTest {
         }
     }
     @Test void partialRerankWithRescaling() {
-        var setup = makeSimpleSetup(makeConstSpec(3.0), 2);
+        var setup = setup().rerank(2).eval(makeConstSpec(3.0)).build();
         var query = makeQuery(Collections.emptyList());
         var result = makeResult(query, List.of(hit("a", 3), hit("b", 4), hit("c", 5), hit("d", 6)));
         var expect = Expect.make(List.of(hit("a", 1), hit("b", 2), hit("c", 3), hit("d", 3)));
@@ -175,8 +178,7 @@ public class GlobalPhaseRerankHitsImplTest {
         expect.verifyScores(result);
     }
     @Test void matchFeaturesCanBePartiallyHidden() {
-        var setup = makeFullSetup(makeSumSpec(Collections.emptyList(), List.of("public_value", "private_value")), 2,
-                List.of("private_value"), Collections.emptyList());
+        var setup = setup().eval(makeSumSpec(Collections.emptyList(), List.of("public_value", "private_value"))).hide("private_value").build();
         var query = makeQuery(Collections.emptyList());
         var factory = new HitFactory(List.of("public_value", "private_value"));
         var result = makeResult(query, List.of(factory.create("a", 1, List.of(value("public_value", 2), value("private_value", 3))),
@@ -188,8 +190,7 @@ public class GlobalPhaseRerankHitsImplTest {
         verifyDoesNotHaveMF(result, "private_value");
     }
     @Test void matchFeaturesCanBeRemoved() {
-        var setup = makeFullSetup(makeSumSpec(Collections.emptyList(), List.of("private_value")), 2,
-                List.of("private_value"), Collections.emptyList());
+        var setup = setup().eval(makeSumSpec(Collections.emptyList(), List.of("private_value"))).hide("private_value").build();
         var query = makeQuery(Collections.emptyList());
         var factory = new HitFactory(List.of("private_value"));
         var result = makeResult(query, List.of(factory.create("a", 1, List.of(value("private_value", 3))),
@@ -200,7 +201,7 @@ public class GlobalPhaseRerankHitsImplTest {
         verifyDoesNotHaveMatchFeaturesField(result);
     }
     @Test void queryFeaturesCanBeUsed() {
-        var setup = makeSimpleSetup(makeSumSpec(List.of("foo"), List.of("bar")), 2);
+        var setup = setup().eval(makeSumSpec(List.of("foo"), List.of("bar"))).build();
         var query = makeQuery(List.of(value("query(foo)", 7)));
         var factory = new HitFactory(List.of("bar"));
         var result = makeResult(query, List.of(factory.create("a", 1, List.of(value("bar", 2))),
@@ -211,7 +212,7 @@ public class GlobalPhaseRerankHitsImplTest {
         verifyHasMF(result, "bar");
     }
     @Test void queryFeaturesCanBeUsedWhenPrepared() {
-        var setup = makeSimpleSetup(makeSumSpec(List.of("foo"), List.of("bar")), 2);
+        var setup = setup().eval(makeSumSpec(List.of("foo"), List.of("bar"))).build();
         var query = makeQueryWithPrepare(List.of(value("query(foo)", 7)));
         var factory = new HitFactory(List.of("bar"));
         var result = makeResult(query, List.of(factory.create("a", 1, List.of(value("bar", 2))),
@@ -221,9 +222,18 @@ public class GlobalPhaseRerankHitsImplTest {
         expect.verifyScores(result);
         verifyHasMF(result, "bar");
     }
+    @Test void queryFeaturesCanBeDefaultValues() {
+        var setup = setup().eval(makeSumSpec(List.of("foo", "bar"), Collections.emptyList()))
+                .addDefault("query(bar)", Tensor.from(5.0)).build();
+        var query = makeQuery(List.of(value("query(foo)", 7)));
+        var result = makeResult(query, List.of(hit("a", 1)));
+        var expect = Expect.make(List.of(hit("a", 12)));
+        GlobalPhaseRanker.rerankHitsImpl(setup, query, result);
+        expect.verifyScores(result);
+    }
     @Test void withNormalizer() {
-        var setup = makeNormSetup(makeSumSpec(Collections.emptyList(), List.of("bar")),
-                List.of(makeNormalizer("foo", List.of(115.0, 65.0, 55.0, 45.0, 15.0), makeSumSpec(List.of("x"), List.of("bar")))));
+        var setup = setup().eval(makeSumSpec(Collections.emptyList(), List.of("bar")))
+                .addNormalizer(makeNormalizer("foo", List.of(115.0, 65.0, 55.0, 45.0, 15.0), makeSumSpec(List.of("x"), List.of("bar")))).build();
         var query = makeQuery(List.of(value("query(x)", 5)));
         var factory = new HitFactory(List.of("bar"));
         var result = makeResult(query, List.of(factory.create("a", 1, List.of(value("bar", 10))),