Skip to content

fix case insensitive lookup for custom text models #645

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
joein merged 2 commits into from
Sep 16, 2026
Merged

fix case insensitive lookup for custom text models #645

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
ff91ff6
fix case insensitive lookup for custom text models
CODING-DARSH Jun 6, 2026
d844fec
refactor: refactor a bit
joein Sep 16, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
5 changes: 3 additions & 2 deletions fastembed/text/custom_text_embedding.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -50,8 +50,9 @@ def __init__(
specific_model_path=specific_model_path,
**kwargs,
)
self._pooling = self.POSTPROCESSING_MAPPING[model_name].pooling
self._normalization = self.POSTPROCESSING_MAPPING[model_name].normalization
postprocessing_config = self.POSTPROCESSING_MAPPING[self.model_description.model]
self._pooling = postprocessing_config.pooling
self._normalization = postprocessing_config.normalization

@classmethod
def _list_supported_models(cls) -> list[DenseModelDescription]:
Expand Down
32 changes: 20 additions & 12 deletions tests/test_custom_models.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -21,9 +21,11 @@
@pytest.fixture(autouse=True)
def restore_custom_models_fixture():
CustomTextEmbedding.SUPPORTED_MODELS = []
CustomTextEmbedding.POSTPROCESSING_MAPPING = {}
CustomTextCrossEncoder.SUPPORTED_MODELS = []
yield
CustomTextEmbedding.SUPPORTED_MODELS = []
CustomTextEmbedding.POSTPROCESSING_MAPPING = {}
CustomTextCrossEncoder.SUPPORTED_MODELS = []


Expand Down Expand Up @@ -74,9 +76,6 @@ def test_text_custom_model():
if is_ci:
delete_model_cache(model.model._model_dir)

CustomTextEmbedding.SUPPORTED_MODELS.clear()
CustomTextEmbedding.POSTPROCESSING_MAPPING.clear()


def test_cross_encoder_custom_model():
is_ci = os.getenv("CI")
Expand Down Expand Up @@ -114,8 +113,6 @@ def test_cross_encoder_custom_model():
if is_ci:
delete_model_cache(model.model._model_dir)

CustomTextCrossEncoder.SUPPORTED_MODELS.clear()


def test_mock_add_custom_models():
dim = 5
Expand Down Expand Up @@ -175,8 +172,24 @@ def test_mock_add_custom_models():
)
assert np.allclose(post_processed_output, expected_output[model_name], atol=1e-3)

CustomTextEmbedding.SUPPORTED_MODELS.clear()
CustomTextEmbedding.POSTPROCESSING_MAPPING.clear()

def test_custom_text_model_lookup_is_case_insensitive():
model_name = "Org/Model"

TextEmbedding.add_custom_model(
model_name,
pooling=PoolingType.MEAN,
normalization=True,
sources=ModelSource(hf="artificial"),
dim=5,
size_in_gb=0.1,
)

model = TextEmbedding("org/model", lazy_load=True, specific_model_path="./")

assert isinstance(model.model, CustomTextEmbedding)
assert model.model._pooling == PoolingType.MEAN
assert model.model._normalization is True


def test_do_not_add_existing_model():
Expand Down Expand Up @@ -212,9 +225,6 @@ def test_do_not_add_existing_model():
size_in_gb=0.47,
)

CustomTextEmbedding.SUPPORTED_MODELS.clear()
CustomTextEmbedding.POSTPROCESSING_MAPPING.clear()


def test_do_not_add_existing_cross_encoder():
existing_base_model = "Xenova/ms-marco-MiniLM-L-6-v2"
Expand All @@ -239,5 +249,3 @@ def test_do_not_add_existing_cross_encoder():
sources=ModelSource(hf=custom_model_name),
size_in_gb=0.08,
)

CustomTextCrossEncoder.SUPPORTED_MODELS.clear()