Bump redisvl and fix bugs (#26)

* round scores to handle any float match discrepencies

* bump redisvl

* add dtype param to custom vectorizer in cache

* fix spacing and linting

* fix formatting

* update formatting and linting

* fix linting

libs/redis/langchain_redis/cache.py

@@ -19,6 +19,7 @@
 from redisvl.extensions.llmcache import (  # type: ignore[import]
     SemanticCache as RedisVLSemanticCache,
 )
+from redisvl.schema.fields import VectorDataType  # type: ignore[import]
 from redisvl.utils.vectorize import BaseVectorizer  # type: ignore[import]
 
 from langchain_redis.version import __full_lib_name__
@@ -36,22 +37,49 @@ def __init__(self, embeddings: Embeddings):
         dims = len(embeddings.embed_query("test"))
         super().__init__(model="custom_embeddings", dims=dims, embeddings=embeddings)
 
-    def encode(self, texts: Union[str, List[str]]) -> np.ndarray:
+    def encode(
+        self,
+        texts: Union[str, List[str]],
+        dtype: Union[str, VectorDataType],
+        **kwargs: Any,
+    ) -> np.ndarray:
+        if isinstance(dtype, VectorDataType):
+            dtype = dtype.value.lower()
         if isinstance(texts, str):
-            return np.array(self.embeddings.embed_query(texts), dtype=np.float32)
-        return np.array(self.embeddings.embed_documents(texts), dtype=np.float32)
+            return np.array(self.embeddings.embed_query(texts), dtype=dtype)
+        return np.array(self.embeddings.embed_documents(texts), dtype=dtype)
 
-    def embed(self, text: str) -> List[float]:
-        return self.encode(text).tolist()
+    def embed(
+        self,
+        text: str,
+        dtype: Union[str, VectorDataType] = "float32",
+        **kwargs: Any,
+    ) -> List[float]:
+        return self.encode(text, dtype, **kwargs).tolist()
 
-    def embed_many(self, texts: List[str]) -> List[List[float]]:
-        return self.encode(texts).tolist()
+    def embed_many(
+        self,
+        texts: List[str],
+        dtype: Union[str, VectorDataType] = "float32",
+        **kwargs: Any,
+    ) -> List[List[float]]:
+        return self.encode(texts, dtype, **kwargs).tolist()
 
-    async def aembed(self, text: str) -> List[float]:
-        return await asyncio.to_thread(self.embed, text)
+    async def aembed(
+        self,
+        text: str,
+        dtype: Union[str, VectorDataType] = "float32",
+        **kwargs: Any,
+    ) -> List[float]:
+        return await asyncio.to_thread(self.embed, text, dtype, **kwargs)
 
-    async def aembed_many(self, texts: List[str]) -> List[List[float]]:
-        return await asyncio.to_thread(self.embed_many, texts)
+    async def aembed_many(
+        self,
+        texts: List[str],
+        dtype: Union[str, VectorDataType] = "float32",
+        **kwargs: Any,
+    ) -> List[List[float]]:
+        return await asyncio.to_thread(self.embed_many, texts, dtype, **kwargs)
 
 
 class RedisCache(BaseCache):

libs/redis/langchain_redis/vectorstores.py

@@ -965,7 +965,10 @@ def similarity_search_with_score_by_vector(
                 doc_embeddings_dict = {
                     doc_id: doc[self.config.embedding_field]
                     if self.config.storage_type == StorageType.JSON.value
-                    else buffer_to_array(doc[self.config.embedding_field])
+                    else buffer_to_array(
+                        doc[self.config.embedding_field],
+                        dtype=self.config.vector_datatype,
+                    )
                     for doc_id, doc in zip(doc_ids, docs_from_storage)
                 }
 
@@ -1039,7 +1042,10 @@ def similarity_search_with_score_by_vector(
                                 },
                             ),
                             float(result.get("vector_distance", 0)),
-                            buffer_to_array(doc.get(self.config.embedding_field)),
+                            buffer_to_array(
+                                doc.get(self.config.embedding_field),
+                                dtype=self.config.vector_datatype,
+                            ),
                         )
                         for doc, result in zip(full_docs, results)
                     ]

libs/redis/pyproject.toml

@@ -13,7 +13,7 @@ license = "MIT"
 [tool.poetry.dependencies]
 python = ">=3.9,<4.0"
 langchain-core = "^0.3"
-redisvl = "^0.3.3"
+redisvl = "^0.3.5"
 numpy = "^1"
 python-ulid = "^2.7.0"
 

libs/redis/tests/integration_tests/test_vectorstores_hash.py

@@ -867,7 +867,7 @@ def test_similarity_search_with_score_with_limit_distance(redis_url: str) -> Non
 
     # Print and verify the scores
     for doc, score in output:  # type: ignore[misc]
-        assert score >= 0  # Ensure score is non-negative
+        assert float(round(score, 4)) >= 0  # Ensure score is non-negative
 
     # Clean up
     vector_store.index.delete(drop=True)

libs/redis/tests/integration_tests/test_vectorstores_json.py

@@ -897,7 +897,7 @@ def test_similarity_search_with_score_with_limit_distance(redis_url: str) -> Non
 
     # Print and verify the scores
     for doc, score in output:  # type: ignore[misc]
-        assert score >= 0  # Ensure score is non-negative
+        assert float(round(score, 4)) >= 0  # Ensure score is non-negative
 
     # Clean up
     vector_store.index.delete(drop=True)