mirror of
https://github.com/hwchase17/langchain.git
synced 2025-05-01 13:26:15 +00:00
a017f49fd3
it fixes two issues: ### YGPTs are broken #25575 ``` File ....conda/lib/python3.11/site-packages/langchain_community/embeddings/yandex.py:211, in _make_request(self, texts, **kwargs) .. --> 211 res = stub.TextEmbedding(request, metadata=self._grpc_metadata) # type: ignore[attr-defined] AttributeError: 'YandexGPTEmbeddings' object has no attribute '_grpc_metadata' ``` My gut feeling that #23841 is the cause. I have to drop leading underscore from `_grpc_metadata` for quickfix, but I just don't know how to do it _pydantic_ enough. ### minor issue: if we use `api_key`, which is not the best practice the code fails with ``` File ~/git/...../python3.11/site-packages/langchain_community/embeddings/yandex.py:119, in YandexGPTEmbeddings.validate_environment(cls, values) ... AttributeError: 'tuple' object has no attribute 'append' ``` - Added new integration test. But it requires YGPT env available and active account. I don't know how int tests dis\enabled in CI. - added small unit tests with mocks. Should be fine. --------- Co-authored-by: mikhail-khludnev <mikhail_khludnev@rntgroup.com>
32 lines
1.1 KiB
Python
32 lines
1.1 KiB
Python
import pytest
|
|
|
|
from langchain_community.embeddings.yandex import YandexGPTEmbeddings
|
|
|
|
|
|
@pytest.mark.parametrize(
|
|
"constructor_args",
|
|
[
|
|
dict(),
|
|
dict(disable_request_logging=True),
|
|
],
|
|
)
|
|
# @pytest.mark.scheduled - idk what it means
|
|
# requires YC_* env and active service
|
|
def test_yandex_embedding(constructor_args: dict) -> None:
|
|
documents = ["exactly same", "exactly same", "different"]
|
|
embedding = YandexGPTEmbeddings(**constructor_args)
|
|
doc_outputs = embedding.embed_documents(documents)
|
|
assert len(doc_outputs) == 3
|
|
for i in range(3):
|
|
assert len(doc_outputs[i]) >= 256 # there are many dims
|
|
assert len(doc_outputs[0]) == len(doc_outputs[i]) # dims are te same
|
|
assert doc_outputs[0] == doc_outputs[1] # same input, same embeddings
|
|
assert doc_outputs[2] != doc_outputs[1] # different input, different embeddings
|
|
|
|
qry_output = embedding.embed_query(documents[0])
|
|
assert len(qry_output) >= 256
|
|
assert len(doc_outputs[0]) == len(
|
|
qry_output
|
|
) # query and doc models have same dimensions
|
|
assert doc_outputs[0] != qry_output # query and doc models are different
|