Skip to content

Adds family of NeuML/pubmedbert-base-embedding models #2443

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
KennethEnevoldsen merged 4 commits into from
Apr 1, 2025
Merged

Adds family of NeuML/pubmedbert-base-embedding models #2443

Changes from all commits
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
9c30967
feat: added pubmedbert model2vec models
nadshe Mar 25, 2025
da4fadf
fix: attribute model_name
nadshe Mar 26, 2025
7877a84
fix: fixed commit hash for pubmed_bert model2vec models
nadshe Mar 26, 2025
79154b8
fix: changes requested in PR 2443
nadshe Mar 31, 2025
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
119 changes: 119 additions & 0 deletions mteb/models/model2vec_models.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -235,3 +235,122 @@ def encode(
public_training_code="https://github.com/MinishLab/model2vec",
public_training_data=None,
)

pubmed_bert_100k = ModelMeta(
loader=partial(
Model2VecWrapper, model_name="NeuML/pubmedbert-base-embeddings-100K"
),
name="NeuML/pubmedbert-base-embeddings-100K",
languages=["eng_Latn"],
open_weights=True,
revision="bac5e3b12fb8c650e92a19c41b436732c4f16e9e",
release_date="2025-01-03",
n_parameters=1 * 1e5,
memory_usage_mb=0,
max_tokens=np.inf,
embed_dim=64,
license="apache-2.0",
similarity_fn_name="cosine",
framework=["NumPy"],
reference="https://huggingface.co/NeuML/pubmedbert-base-embeddings-100K",
use_instructions=False,
adapted_from="NeuML/pubmedbert-base-embeddings",
superseded_by=None,
training_datasets={},
public_training_code="https://huggingface.co/NeuML/pubmedbert-base-embeddings-100K#training",
public_training_data="https://pubmed.ncbi.nlm.nih.gov/download/",
)

pubmed_bert_500k = ModelMeta(
loader=partial(
Model2VecWrapper, model_name="NeuML/pubmedbert-base-embeddings-500K"
),
name="NeuML/pubmedbert-base-embeddings-500K",
languages=["eng_Latn"],
open_weights=True,
revision="34ba71e35c393fdad7ed695113f653feb407b16b",
release_date="2025-01-03",
n_parameters=5 * 1e5,
memory_usage_mb=2,
max_tokens=np.inf,
embed_dim=64,
license="apache-2.0",
similarity_fn_name="cosine",
framework=["NumPy"],
reference="https://huggingface.co/NeuML/pubmedbert-base-embeddings-500K",
use_instructions=False,
adapted_from="NeuML/pubmedbert-base-embeddings",
superseded_by=None,
training_datasets={},
public_training_code="https://huggingface.co/NeuML/pubmedbert-base-embeddings-500K#training",
public_training_data="https://pubmed.ncbi.nlm.nih.gov/download/",
)

pubmed_bert_1m = ModelMeta(
loader=partial(Model2VecWrapper, model_name="NeuML/pubmedbert-base-embeddings-1M"),
name="NeuML/pubmedbert-base-embeddings-1M",
languages=["eng_Latn"],
open_weights=True,
revision="2b7fed222594708da6d88bcda92ae9b434b7ddd1",
release_date="2025-01-03",
n_parameters=1 * 1e6,
memory_usage_mb=2,
max_tokens=np.inf,
embed_dim=64,
license="apache-2.0",
similarity_fn_name="cosine",
framework=["NumPy"],
reference="https://huggingface.co/NeuML/pubmedbert-base-embeddings-1M",
use_instructions=False,
adapted_from="NeuML/pubmedbert-base-embeddings",
superseded_by=None,
training_datasets={},
public_training_code="https://huggingface.co/NeuML/pubmedbert-base-embeddings-1M#training",
public_training_data="https://pubmed.ncbi.nlm.nih.gov/download/",
)

pubmed_bert_2m = ModelMeta(
loader=partial(Model2VecWrapper, model_name="NeuML/pubmedbert-base-embeddings-2M"),
name="NeuML/pubmedbert-base-embeddings-2M",
languages=["eng_Latn"],
open_weights=True,
revision="1d7bbe04d6713e425161146bfdc71473cbed498a",
release_date="2025-01-03",
n_parameters=1.95 * 1e6,
memory_usage_mb=7,
max_tokens=np.inf,
embed_dim=64,
license="apache-2.0",
similarity_fn_name="cosine",
framework=["NumPy"],
reference="https://huggingface.co/NeuML/pubmedbert-base-embeddings-2M",
use_instructions=False,
adapted_from="NeuML/pubmedbert-base-embeddings",
superseded_by=None,
training_datasets={},
public_training_code="https://huggingface.co/NeuML/pubmedbert-base-embeddings-2M#training",
public_training_data="https://pubmed.ncbi.nlm.nih.gov/download/",
)

pubmed_bert_8m = ModelMeta(
loader=partial(Model2VecWrapper, model_name="NeuML/pubmedbert-base-embeddings-8M"),
name="NeuML/pubmedbert-base-embeddings-8M",
languages=["eng_Latn"],
open_weights=True,
revision="387d350015e963744f4fafe56a574b7cd48646c9",
release_date="2025-01-03",
n_parameters=7.81 * 1e6,
memory_usage_mb=30,
max_tokens=np.inf,
embed_dim=256,
license="apache-2.0",
similarity_fn_name="cosine",
framework=["NumPy"],
reference="https://huggingface.co/NeuML/pubmedbert-base-embeddings-8M",
use_instructions=False,
adapted_from="NeuML/pubmedbert-base-embeddings",
superseded_by=None,
training_datasets={},
public_training_code="https://huggingface.co/NeuML/pubmedbert-base-embeddings-8M#training",
public_training_data="https://pubmed.ncbi.nlm.nih.gov/download/",
)
Loading