Merge branch 'main' into feature/ollama_integration

deepset-ai · Dec 20, 2023 · 23f85b9 · 23f85b9
2 parents b53ab7c + e7d79e7
commit 23f85b9
Show file tree

Hide file tree

Showing 2 changed files with 28 additions and 1 deletion.
diff --git a/integrations/elasticsearch/src/elasticsearch_haystack/document_store.py b/integrations/elasticsearch/src/elasticsearch_haystack/document_store.py
@@ -263,7 +263,7 @@ def _bm25_retrieval(
                                 "query": query,
                                 "fuzziness": fuzziness,
                                 "type": "most_fields",
-                                "operator": "AND",
+                                "operator": "OR",
                             }
                         }
                     ]

diff --git a/integrations/elasticsearch/tests/test_document_store.py b/integrations/elasticsearch/tests/test_document_store.py
@@ -182,6 +182,33 @@ def test_bm25_retrieval_with_fuzziness(self, document_store: ElasticsearchDocume
         assert "functional" in res[1].content
         assert "functional" in res[2].content
 
+    def test_bm25_not_all_terms_must_match(self, document_store: ElasticsearchDocumentStore):
+        """
+        Test that not all terms must mandatorily match for BM25 retrieval to return a result.
+        """
+        documents = [
+            Document(id=1, content="There are over 7,000 languages spoken around the world today."),
+            Document(
+                id=2,
+                content=(
+                    "Elephants have been observed to behave in a way that indicates a high level of self-awareness"
+                    " such as recognizing themselves in mirrors."
+                ),
+            ),
+            Document(
+                id=3,
+                content=(
+                    "In certain parts of the world, like the Maldives, Puerto Rico, and San Diego, you can witness"
+                    " the phenomenon of bioluminescent waves."
+                ),
+            ),
+        ]
+        document_store.write_documents(documents)
+
+        res = document_store._bm25_retrieval("How much self awareness do elephants have?", top_k=3)
+        assert len(res) == 1
+        assert res[0].id == 2
+
     def test_embedding_retrieval(self, document_store: ElasticsearchDocumentStore):
         docs = [
             Document(content="Most similar document", embedding=[1.0, 1.0, 1.0, 1.0]),