참고소스 수정본

참고/neo4j-graphrag-python-main/examples/customize/retrievers/external/weaviate/README.md

@@ -0,0 +1,38 @@
+### Start services locally
+
+This is a manual task you need to do in the terminal.
+
+This spins up Neo4j and Weaviate containers and is configuring Weaviate to use embeddings from Hugging Face's Sentence Transformers using the "all-MiniLM-L6-v2" model, which has 384 dimensions.
+
+```bash
+docker compose -f tests/e2e/docker-compose.yml up
+```
+
+### Write data (once)
+
+Run this from the project root to write data to both dbs.
+
+```
+uv run python -m tests/e2e/weaviate_e2e/populate_dbs.py
+```
+
+### Install Weaviate client
+
+You need to install the `weaviate-client` package to use this retriever.
+
+```bash
+pip install weaviate-client
+```
+
+### Search
+
+```
+# search by vector
+uv run python -m examples.customize.retrievers.external.weaviate.vector_search
+
+# search by text, with embeddings generated locally (via embedder argument)
+uv run python -m examples.customize.retrievers.external.weaviate.text_search_local_embedder
+
+# search by text, with embeddings generated on the Weaviate side, via configured vectorizer
+uv run python -m examples.customize.retrievers.external.weaviate.text_search_remote_embedder
+```

참고/neo4j-graphrag-python-main/examples/customize/retrievers/external/weaviate/weaviate_text_search_local_embedder.py

@@ -0,0 +1,40 @@
+"""This example demonstrates how to use WeaviateNeo4jRetriever, ie vectors are
+stored in the Weaviate database.
+
+See the [README](./README.md) for more
+information about how spin up a Weaviate and Neo4j databases if needed.
+
+In this example, we are embeddings a text and provide a local embeder to the
+WeaviateNeo4jRetriever.
+"""
+
+from neo4j import GraphDatabase
+from neo4j_graphrag.embeddings.sentence_transformers import (
+    SentenceTransformerEmbeddings,
+)
+from neo4j_graphrag.retrievers import WeaviateNeo4jRetriever
+from weaviate.connect.helpers import connect_to_local
+
+NEO4J_URL = "neo4j://localhost:7687"
+NEO4J_AUTH = ("neo4j", "password")
+
+
+def main() -> None:
+    with GraphDatabase.driver(NEO4J_URL, auth=NEO4J_AUTH) as neo4j_driver:
+        with connect_to_local() as w_client:
+            embedder = SentenceTransformerEmbeddings(model="all-MiniLM-L6-v2")
+            retriever = WeaviateNeo4jRetriever(
+                driver=neo4j_driver,
+                client=w_client,
+                collection="Jeopardy",
+                id_property_external="neo4j_id",
+                id_property_neo4j="id",
+                embedder=embedder,
+            )
+
+            res = retriever.search(query_text="biology", top_k=2)
+            print(res)
+
+
+if __name__ == "__main__":
+    main()

참고/neo4j-graphrag-python-main/examples/customize/retrievers/external/weaviate/weaviate_text_search_remote_embedder.py

@@ -0,0 +1,37 @@
+"""This example demonstrates how to use WeaviateNeo4jRetriever, ie vectors are
+stored in the Weaviate database.
+
+See the [README](./README.md) for more
+information about how spin up a Weaviate and Neo4j databases if needed.
+
+In this example, we are embeddings a text and provide a remote embeder to the
+WeaviateNeo4jRetriever.
+"""
+
+from neo4j import GraphDatabase
+from neo4j_graphrag.retrievers import WeaviateNeo4jRetriever
+from weaviate.connect.helpers import connect_to_local
+
+NEO4J_URL = "neo4j://localhost:7687"
+NEO4J_AUTH = ("neo4j", "password")
+
+
+def main() -> None:
+    with GraphDatabase.driver(NEO4J_URL, auth=NEO4J_AUTH) as neo4j_driver:
+        with connect_to_local() as w_client:
+            retriever = WeaviateNeo4jRetriever(
+                driver=neo4j_driver,
+                client=w_client,
+                collection="Jeopardy",
+                id_property_external="neo4j_id",
+                id_property_neo4j="id",
+            )
+
+            # Weaviate is configured to embed the text on the server side
+            # see populate_dbs.py for the configuration
+            res = retriever.search(query_text="biology", top_k=2)
+            print(res)
+
+
+if __name__ == "__main__":
+    main()

참고/neo4j-graphrag-python-main/examples/customize/retrievers/external/weaviate/weaviate_vector_search.py

@@ -0,0 +1,34 @@
+"""This example demonstrates how to use WeaviateNeo4jRetriever, ie vectors are
+stored in the Weaviate database.
+
+See the [README](./README.md) for more
+information about how spin up a Weaviate and Neo4j databases if needed.
+
+In this example, search is performed from an already existing vector.
+"""
+
+from embedding_biology import EMBEDDING_BIOLOGY
+from neo4j import GraphDatabase
+from neo4j_graphrag.retrievers import WeaviateNeo4jRetriever
+from weaviate.connect.helpers import connect_to_local
+
+NEO4J_URL = "neo4j://localhost:7687"
+NEO4J_AUTH = ("neo4j", "password")
+
+
+def main() -> None:
+    with GraphDatabase.driver(NEO4J_URL, auth=NEO4J_AUTH) as neo4j_driver:
+        with connect_to_local() as w_client:
+            retriever = WeaviateNeo4jRetriever(
+                driver=neo4j_driver,
+                client=w_client,
+                collection="Jeopardy",
+                id_property_external="neo4j_id",
+                id_property_neo4j="id",
+            )
+            res = retriever.search(query_vector=EMBEDDING_BIOLOGY, top_k=2)
+            print(res)
+
+
+if __name__ == "__main__":
+    main()