Skip to content

The agent engineering platform.

Bug-Fix[Community] Fix `FastEmbedEmbeddings` (#26764)

ZhangShenaoe317d457cfe8
diff --git a/libs/community/langchain_community/embeddings/fastembed.py b/libs/community/langchain_community/embeddings/fastembed.py
index 288061b67c..28105dd720 100644
--- a/libs/community/langchain_community/embeddings/fastembed.py
+++ b/libs/community/langchain_community/embeddings/fastembed.py
@@ -65,7 +65,7 @@ class FastEmbedEmbeddings(BaseModel, Embeddings):
Defaults to `None`.
"""
- _model: Any = None # : :meta private:
+ model: Any = None # : :meta private:
model_config = ConfigDict(extra="allow", protected_namespaces=())
@@ -91,7 +91,7 @@ class FastEmbedEmbeddings(BaseModel, Embeddings):
'FastEmbedEmbeddings requires `pip install -U "fastembed>=0.2.0"`.'
)
- values["_model"] = fastembed.TextEmbedding(
+ values["model"] = fastembed.TextEmbedding(
model_name=model_name,
max_length=max_length,
cache_dir=cache_dir,
@@ -110,11 +110,11 @@ class FastEmbedEmbeddings(BaseModel, Embeddings):
"""
embeddings: List[np.ndarray]
if self.doc_embed_type == "passage":
- embeddings = self._model.passage_embed(
+ embeddings = self.model.passage_embed(
texts, batch_size=self.batch_size, parallel=self.parallel
)
else:
- embeddings = self._model.embed(
+ embeddings = self.model.embed(
texts, batch_size=self.batch_size, parallel=self.parallel
)
return [e.tolist() for e in embeddings]
@@ -129,7 +129,7 @@ class FastEmbedEmbeddings(BaseModel, Embeddings):
Embeddings for the text.
"""
query_embeddings: np.ndarray = next(
- self._model.query_embed(
+ self.model.query_embed(
text, batch_size=self.batch_size, parallel=self.parallel
)
)
diff --git a/libs/community/tests/integration_tests/embeddings/test_fastembed.py b/libs/community/tests/integration_tests/embeddings/test_fastembed.py
index 09cede659b..f39c4cedfb 100644
--- a/libs/community/tests/integration_tests/embeddings/test_fastembed.py
+++ b/libs/community/tests/integration_tests/embeddings/test_fastembed.py
@@ -80,3 +80,11 @@ async def test_fastembed_async_embedding_query(
embedding = FastEmbedEmbeddings(model_name=model_name, max_length=max_length) # type: ignore[call-arg]
output = await embedding.aembed_query(document)
assert len(output) == 384
+
+
+def test_fastembed_embedding_query_with_default_params() -> None:
+ """Test fastembed embeddings for query with default model params"""
+ document = "foo bar"
+ embedding = FastEmbedEmbeddings() # type: ignore[call-arg]
+ output = embedding.embed_query(document)
+ assert len(output) == 384