mteb 2.7.17__py3-none-any.whl → 2.7.19__py3-none-any.whl

mteb/models/model_implementations/ict_time_and_querit_models.py

@@ -0,0 +1,115 @@
+from __future__ import annotations
+
+from mteb.models.instruct_wrapper import InstructSentenceTransformerModel
+from mteb.models.model_meta import ModelMeta
+from mteb.types import PromptType
+
+
+def instruction_template(
+    instruction: str | dict, prompt_type: PromptType | None = None
+) -> str:
+    """Format instruction for the model."""
+    if isinstance(instruction, dict):
+        instruction = instruction.get(prompt_type.value if prompt_type else "", "")
+    elif prompt_type == PromptType.document:
+        return ""
+
+    if not instruction:
+        return ""
+    return f"Instruct: {instruction}\nQuery:"
+
+
+multilingual_langs = [
+    "deu-Latn",
+    "ita-Latn",
+    "ara-Arab",
+    "fas-Arab",
+    "fra-Latn",
+    "hin-Deva",
+    "spa-Latn",
+    "zho-Hans",
+    "ben-Beng",
+    "eng-Latn",
+    "fin-Latn",
+    "ind-Latn",
+    "jpn-Jpan",
+    "kor-Hang",
+    "rus-Cyrl",
+    "swh-Latn",
+    "tel-Telu",
+    "tha-Thai",
+]
+
+training_data = [
+    "FEVER",
+    "DuRetrieval",
+    "HotpotQA",
+    "MSMARCO",
+    "T2Retrieval",
+    "NQ",
+    "MIRACLRetrieval",
+    "MrTidyRetrieval",
+    "AmazonCounterfactualClassification",
+    "Banking77Classification",
+    "ImdbClassification",
+    "MTOPDomainClassification",
+    "ToxicConversationsClassification",
+    "TweetSentimentExtractionClassification",
+]
+
+boom_4b_instructions = {
+    "AmazonCounterfactualClassification": "Classify a given Amazon customer review text as either counterfactual or not-counterfactual.",
+    "AmazonPolarityClassification": "Classify Amazon reviews into positive or negative sentiment.",
+    "AmazonReviewsClassification": "Classify the given Amazon review into its appropriate rating category.",
+    "Banking77Classification": "Given a online banking query, find the corresponding intents.",
+    "EmotionClassification": "Classify the emotion expressed in the given Twitter message into one of the six emotions: anger, fear, joy, love, sadness, and surprise.",
+    "ImdbClassification": "Classify the sentiment expressed in the given movie review text from the IMDB dataset.",
+    "MassiveIntentClassification": "Given a user utterance as query, find the user intents.",
+    "MassiveScenarioClassification": "Given a user utterance as query, find the user scenarios.",
+    "MTOPDomainClassification": "Classify the intent domain of the given utterance in task-oriented conversation.",
+    "MTOPIntentClassification": "Classify the intent of the given utterance in task-oriented conversation.",
+    "ToxicConversationsClassification": "Classify the given comments as either toxic or not toxic.",
+    "TweetSentimentExtractionClassification": "Classify the sentiment of a given tweet as either positive, negative, or neutral.",
+    "TNews": "Classify the fine-grained category of the given news title.",
+    "ClimateFEVER": "Given a claim about climate change, retrieve documents that support or refute the claim.",
+    "ClimateFEVERHardNegatives": "Given a claim about climate change, retrieve documents that support or refute the claim.",
+    "DBPedia": "Given a query, retrieve relevant entity descriptions from DBPedia.",
+    "FEVER": "Given a claim, retrieve documents that support or refute the claim.",
+    "FEVERHardNegatives": "Given a claim, retrieve documents that support or refute the claim.",
+    "FiQA2018": "Given a financial question, retrieve user replies that best answer the question.",
+    "HotpotQA": "Given a multi-hop question, retrieve documents that can help answer the question.",
+    "HotpotQAHardNegatives": "Given a multi-hop question, retrieve documents that can help answer the question.",
+    "MSMARCO": "Given a web search query, retrieve relevant passages that answer the query.",
+    "NFCorpus": "Given a question, retrieve relevant documents that best answer the question.",
+    "NQ": "Given a question, retrieve Wikipedia passages that answer the question.",
+}
+# How the template actually renders each one at inference time:
+#   instruction_template(boom_4b_instructions["Banking77Classification"], PromptType.query)
+#   -> "Instruct: Given a online banking query, find the corresponding intents.\nQuery:"
+
+boom_4b_v1 = ModelMeta(
+    loader=InstructSentenceTransformerModel,
+    loader_kwargs=dict(
+        instruction_template=instruction_template,
+    ),
+    name="ICT-TIME-and-Querit/BOOM_4B_v1",
+    model_type=["dense"],
+    languages=multilingual_langs,
+    open_weights=True,
+    adapted_from="Qwen/Qwen3-4B",
+    revision="447ab88574d27e67c428acc2b429d7d4580a4ea7",
+    release_date="2026-01-31",
+    n_parameters=4021774336,
+    n_embedding_parameters=None,
+    memory_usage_mb=7671,
+    embed_dim=2560,
+    max_tokens=32768,
+    license="apache-2.0",
+    reference="https://huggingface.co/ICT-TIME-and-Querit/BOOM_4B_v1",
+    similarity_fn_name="cosine",
+    framework=["Sentence Transformers", "PyTorch", "safetensors", "Transformers"],
+    use_instructions=True,
+    public_training_code=None,
+    public_training_data=None,
+    training_datasets=training_data,
+)

mteb/models/model_implementations/pylate_models.py

@@ -53,7 +53,7 @@ class PylateSearchEncoder:
         hf_split: str,
         hf_subset: str,
         encode_kwargs: EncodeKwargs,
-        num_proc: int,
+        num_proc: int | None,
     ) -> None:
         """Index the corpus for retrieval.
 
@@ -89,7 +89,7 @@ class PylateSearchEncoder:
         top_k: int,
         encode_kwargs: EncodeKwargs,
         top_ranked: TopRankedDocumentsType | None = None,
-        num_proc: int,
+        num_proc: int | None,
     ) -> RetrievalOutputType:
         queries_dataloader = create_dataloader(
             queries,
@@ -150,7 +150,7 @@ class PylateSearchEncoder:
         hf_split: str,
         top_k: int,
         encode_kwargs: EncodeKwargs,
-        num_proc: int,
+        num_proc: int | None,
     ) -> dict[str, list[tuple[float, str]]]:
         from pylate import indexes, retrieve
 
@@ -216,7 +216,7 @@ class PylateSearchEncoder:
         hf_subset: str,
         hf_split: str,
         encode_kwargs: EncodeKwargs,
-        num_proc: int = 1,
+        num_proc: int | None = None,
     ) -> dict[str, list[tuple[float, str]]]:
         """Rerank with PyLate's rank.rerank using per-query candidates.
 

mteb/models/models_protocols.py

@@ -32,7 +32,7 @@ class SearchProtocol(Protocol):
         hf_split: str,
         hf_subset: str,
         encode_kwargs: EncodeKwargs,
-        num_proc: int,
+        num_proc: int | None,
     ) -> None:
         """Index the corpus for retrieval.
 
@@ -56,7 +56,7 @@ class SearchProtocol(Protocol):
         top_k: int,
         encode_kwargs: EncodeKwargs,
         top_ranked: TopRankedDocumentsType | None = None,
-        num_proc: int,
+        num_proc: int | None,
     ) -> RetrievalOutputType:
         """Search the corpus using the given queries.
 

mteb/models/search_wrappers.py

@@ -59,7 +59,7 @@ class SearchEncoderWrapper:
         hf_split: str,
         hf_subset: str,
         encode_kwargs: EncodeKwargs,
-        num_proc: int = 1,
+        num_proc: int | None = None,
     ) -> None:
         """Index the corpus for retrieval.
 
@@ -101,7 +101,7 @@ class SearchEncoderWrapper:
         top_k: int,
         encode_kwargs: EncodeKwargs,
         top_ranked: TopRankedDocumentsType | None = None,
-        num_proc: int = 1,
+        num_proc: int | None = None,
     ) -> RetrievalOutputType:
         """Search the corpus for the given queries.
 
@@ -485,7 +485,7 @@ class SearchCrossEncoderWrapper:
         hf_split: str,
         hf_subset: str,
         encode_kwargs: EncodeKwargs,
-        num_proc: int = 1,
+        num_proc: int | None = None,
     ) -> None:
         """Index the corpus for retrieval.
 
@@ -509,7 +509,7 @@ class SearchCrossEncoderWrapper:
         top_k: int,
         encode_kwargs: EncodeKwargs,
         top_ranked: TopRankedDocumentsType | None = None,
-        num_proc: int = 1,
+        num_proc: int | None = None,
     ) -> RetrievalOutputType:
         """Search the corpus using the given queries.
 

mteb/tasks/bitext_mining/multilingual/bible_nlp_bitext_mining.py

@@ -914,7 +914,7 @@ class BibleNLPBitextMining(AbsTaskBitextMining):
         self.dataset_transform()
         self.data_loaded = True
 
-    def dataset_transform(self, num_proc: int = 1, **kwargs) -> None:
+    def dataset_transform(self, num_proc: int | None = None, **kwargs) -> None:
         # Convert to standard format
         for lang in self.hf_subsets:
             l1, l2 = (l.split("_")[0] for l in lang.split("-"))

mteb/tasks/bitext_mining/multilingual/flores_bitext_mining.py

@@ -265,7 +265,7 @@ class FloresBitextMining(AbsTaskBitextMining):
 """,
     )
 
-    def load_data(self, num_proc: int = 1, **kwargs) -> None:
+    def load_data(self, num_proc: int | None = None, **kwargs) -> None:
         if self.data_loaded:
             return
 

mteb/tasks/bitext_mining/multilingual/in22_conv_bitext_mining.py

@@ -99,7 +99,7 @@ class IN22ConvBitextMining(AbsTaskBitextMining):
 """,
     )
 
-    def load_data(self, num_proc: int = 1, **kwargs) -> None:
+    def load_data(self, num_proc: int | None = None, **kwargs) -> None:
         if self.data_loaded:
             return
 

mteb/tasks/bitext_mining/multilingual/in22_gen_bitext_mining.py

@@ -93,7 +93,7 @@ class IN22GenBitextMining(AbsTaskBitextMining):
 """,
     )
 
-    def load_data(self, num_proc: int = 1, **kwargs) -> None:
+    def load_data(self, num_proc: int | None = None, **kwargs) -> None:
         if self.data_loaded:
             return
 

mteb/tasks/bitext_mining/multilingual/ntrex_bitext_mining.py

@@ -280,7 +280,7 @@ class NTREXBitextMining(AbsTaskBitextMining):
 """,
     )
 
-    def load_data(self, num_proc: int = 1, **kwargs) -> None:
+    def load_data(self, num_proc: int | None = None, **kwargs) -> None:
         if self.data_loaded:
             return
 

mteb/tasks/bitext_mining/multilingual/roma_tales_bitext_mining.py

@@ -32,7 +32,7 @@ class RomaTalesBitextMining(AbsTaskBitextMining):
         bibtex_citation="",
     )
 
-    def load_data(self, num_proc: int = 1, **kwargs) -> None:
+    def load_data(self, num_proc: int | None = None, **kwargs) -> None:
         """Load dataset from HuggingFace hub and convert it to the standard format."""
         if self.data_loaded:
             return

mteb/tasks/classification/ben/bengali_document_classification.py

@@ -43,7 +43,7 @@ Islam, Tanvir},
         superseded_by="BengaliDocumentClassification.v2",
     )
 
-    def dataset_transform(self, num_proc: int = 1, **kwargs) -> None:
+    def dataset_transform(self, num_proc: int | None = None, **kwargs) -> None:
         self.dataset = self.dataset.rename_columns(
             {"article": "text", "category": "label"}
         )
@@ -92,7 +92,7 @@ Islam, Tanvir},
 """,
     )
 
-    def dataset_transform(self, num_proc: int = 1, **kwargs) -> None:
+    def dataset_transform(self, num_proc: int | None = None, **kwargs) -> None:
         self.dataset = self.stratified_subsampling(
             self.dataset, seed=self.seed, splits=["test"]
         )

mteb/tasks/classification/ces/czech_product_review_sentiment_classification.py

@@ -46,7 +46,7 @@ Montoyo, Andres},
     )
     samples_per_label = 16
 
-    def dataset_transform(self, num_proc: int = 1, **kwargs) -> None:
+    def dataset_transform(self, num_proc: int | None = None, **kwargs) -> None:
         self.dataset = self.dataset.rename_columns(
             {"comment": "text", "rating_str": "label"}
         )
@@ -99,7 +99,7 @@ Montoyo, Andres},
     )
     samples_per_label = 16
 
-    def dataset_transform(self, num_proc: int = 1, **kwargs) -> None:
+    def dataset_transform(self, num_proc: int | None = None, **kwargs) -> None:
         self.dataset = self.stratified_subsampling(
             self.dataset, seed=self.seed, splits=["test"]
         )

mteb/tasks/classification/ces/czech_so_me_sentiment_classification.py

@@ -46,7 +46,7 @@ Montoyo, Andres},
     )
     samples_per_label = 16
 
-    def dataset_transform(self, num_proc: int = 1, **kwargs) -> None:
+    def dataset_transform(self, num_proc: int | None = None, **kwargs) -> None:
         self.dataset = self.dataset.rename_columns(
             {"comment": "text", "sentiment_int": "label"}
         )

mteb/tasks/classification/multilingual/hin_dialect_classification.py

@@ -60,7 +60,7 @@ class HinDialectClassification(AbsTaskClassification):
 """,
     )
 
-    def dataset_transform(self, num_proc: int = 1, **kwargs) -> None:
+    def dataset_transform(self, num_proc: int | None = None, **kwargs) -> None:
         self.dataset = self.dataset.rename_columns(
             {"folksong": "text", "language": "label"}
         )

mteb/tasks/classification/multilingual/indic_lang_classification.py

@@ -137,6 +137,6 @@ Okazaki, Naoaki},
         self.dataset_transform()
         self.data_loaded = True
 
-    def dataset_transform(self, num_proc: int = 1, **kwargs) -> None:
+    def dataset_transform(self, num_proc: int | None = None, **kwargs) -> None:
         self.dataset = self.dataset.remove_columns(["language", "script"])
         self.dataset = self.dataset.rename_columns({"native sentence": "text"})

mteb/tasks/classification/multilingual/indic_sentiment_classification.py

@@ -52,7 +52,7 @@ class IndicSentimentClassification(AbsTaskClassification):
 """,
     )
 
-    def dataset_transform(self, num_proc: int = 1, **kwargs) -> None:
+    def dataset_transform(self, num_proc: int | None = None, **kwargs) -> None:
         label_map = {"Negative": 0, "Positive": 1}
         # Convert to standard format
         for lang in self.hf_subsets:

mteb/tasks/classification/multilingual/language_classification.py

@@ -66,7 +66,7 @@ in Natural Language Processing},
 """,
     )
 
-    def dataset_transform(self, num_proc: int = 1, **kwargs) -> None:
+    def dataset_transform(self, num_proc: int | None = None, **kwargs) -> None:
         self.dataset = self.dataset.rename_columns({"labels": "label"})
         self.dataset = self.stratified_subsampling(
             self.dataset, seed=self.seed, splits=["test"]

mteb/tasks/classification/multilingual/south_african_lang_classification.py

@@ -49,7 +49,7 @@ class SouthAfricanLangClassification(AbsTaskClassification):
 """,
     )
 
-    def dataset_transform(self, num_proc: int = 1, **kwargs) -> None:
+    def dataset_transform(self, num_proc: int | None = None, **kwargs) -> None:
         self.dataset = self.dataset.rename_columns(
             {" text": "text", "lang_id": "label"}
         )

mteb/tasks/classification/multilingual/turkic_classification.py

@@ -49,7 +49,7 @@ class TurkicClassification(AbsTaskClassification):
         )
         return dataset_lang["train"]
 
-    def load_data(self, num_proc: int = 1, **kwargs) -> None:
+    def load_data(self, num_proc: int | None = None, **kwargs) -> None:
         """Load dataset from HuggingFace hub"""
         if self.data_loaded:
             return

mteb/tasks/classification/slk/slovak_movie_review_sentiment_classification.py

@@ -35,7 +35,7 @@ class SlovakMovieReviewSentimentClassification(AbsTaskClassification):
         superseded_by="SlovakMovieReviewSentimentClassification.v2",
     )
 
-    def dataset_transform(self, num_proc: int = 1, **kwargs) -> None:
+    def dataset_transform(self, num_proc: int | None = None, **kwargs) -> None:
         self.dataset = self.dataset.rename_columns({"comment": "text"})
 
         self.dataset = self.stratified_subsampling(
@@ -76,7 +76,7 @@ class SlovakMovieReviewSentimentClassificationV2(AbsTaskClassification):
         adapted_from=["SlovakMovieReviewSentimentClassification"],
     )
 
-    def dataset_transform(self, num_proc: int = 1, **kwargs) -> None:
+    def dataset_transform(self, num_proc: int | None = None, **kwargs) -> None:
         self.dataset = self.stratified_subsampling(
             self.dataset, seed=self.seed, splits=["test"]
         )

mteb/tasks/classification/swa/swahili_news_classification.py

@@ -37,7 +37,7 @@ class SwahiliNewsClassification(AbsTaskClassification):
         superseded_by="SwahiliNewsClassification.v2",
     )
 
-    def dataset_transform(self, num_proc: int = 1, **kwargs) -> None:
+    def dataset_transform(self, num_proc: int | None = None, **kwargs) -> None:
         self.dataset = self.dataset.rename_columns(
             {"content": "text", "category": "label"}
         )
@@ -81,7 +81,7 @@ class SwahiliNewsClassificationV2(AbsTaskClassification):
         adapted_from=["SwahiliNewsClassification"],
     )
 
-    def dataset_transform(self, num_proc: int = 1, **kwargs) -> None:
+    def dataset_transform(self, num_proc: int | None = None, **kwargs) -> None:
         self.dataset = self.stratified_subsampling(
             self.dataset, seed=self.seed, splits=["train"]
         )

mteb/tasks/clustering/deu/ten_k_gnad_clustering_p2p.py

@@ -63,7 +63,7 @@ class TenKGnadClusteringP2PFast(AbsTaskClustering):
         adapted_from=["TenKGnadClusteringP2P"],
     )
 
-    def dataset_transform(self, num_proc: int = 1, **kwargs) -> None:
+    def dataset_transform(self, num_proc: int | None = None, **kwargs) -> None:
         ds = _convert_to_fast(
             self.dataset, self.input_column_name, self.label_column_name, self.seed
         )

mteb/tasks/clustering/deu/ten_k_gnad_clustering_s2s.py

@@ -63,7 +63,7 @@ class TenKGnadClusteringS2SFast(AbsTaskClustering):
         adapted_from=["TenKGnadClusteringS2S"],
     )
 
-    def dataset_transform(self, num_proc: int = 1, **kwargs) -> None:
+    def dataset_transform(self, num_proc: int | None = None, **kwargs) -> None:
         ds = _convert_to_fast(
             self.dataset, self.input_column_name, self.label_column_name, self.seed
         )

mteb/tasks/clustering/multilingual/mlsum_clustering_p2p.py

@@ -51,7 +51,7 @@ class MLSUMClusteringP2P(AbsTaskClusteringLegacy):
         superseded_by="MLSUMClusteringP2P.v2",
     )
 
-    def load_data(self, num_proc: int = 1, **kwargs) -> None:
+    def load_data(self, num_proc: int | None = None, **kwargs) -> None:
         """Load dataset from HuggingFace hub and convert it to the standard format."""
         if self.data_loaded:
             return
@@ -124,7 +124,7 @@ class MLSUMClusteringP2PFast(AbsTaskClustering):
         adapted_from=["MLSUMClusteringP2P"],
     )
 
-    def load_data(self, num_proc: int = 1, **kwargs) -> None:
+    def load_data(self, num_proc: int | None = None, **kwargs) -> None:
         """Load dataset from HuggingFace hub and convert it to the standard format."""
         if self.data_loaded:
             return

mteb/tasks/clustering/multilingual/mlsum_clustering_s2s.py

@@ -51,7 +51,7 @@ class MLSUMClusteringS2S(AbsTaskClusteringLegacy):
         superseded_by="MLSUMClusteringS2S.v2",
     )
 
-    def load_data(self, num_proc: int = 1, **kwargs) -> None:
+    def load_data(self, num_proc: int | None = None, **kwargs) -> None:
         """Load dataset from HuggingFace hub and convert it to the standard format."""
         if self.data_loaded:
             return
@@ -119,7 +119,7 @@ class MLSUMClusteringS2SFast(AbsTaskClustering):
         adapted_from=["MLSUMClusteringS2S"],
     )
 
-    def load_data(self, num_proc: int = 1, **kwargs) -> None:
+    def load_data(self, num_proc: int | None = None, **kwargs) -> None:
         """Load dataset from HuggingFace hub and convert it to the standard format."""
         if self.data_loaded:
             return

mteb/tasks/clustering/nob/vg_hierarchical_clustering.py

@@ -45,7 +45,7 @@ class VGHierarchicalClusteringP2P(AbsTaskClustering):
         prompt="Identify the categories (e.g. sports) of given articles in Norwegian",
     )
 
-    def dataset_transform(self, num_proc: int = 1, **kwargs) -> None:
+    def dataset_transform(self, num_proc: int | None = None, **kwargs) -> None:
         self.dataset = self.dataset.rename_columns(
             {"article": "sentences", "classes": "labels"}
         )
@@ -92,7 +92,7 @@ class VGHierarchicalClusteringS2S(AbsTaskClustering):
         prompt="Identify the categories (e.g. sports) of given articles in Norwegian",
     )
 
-    def dataset_transform(self, num_proc: int = 1, **kwargs) -> None:
+    def dataset_transform(self, num_proc: int | None = None, **kwargs) -> None:
         self.dataset = self.dataset.rename_columns(
             {"ingress": "sentences", "classes": "labels"}
         )

mteb/tasks/image_text_pair_classification/eng/image_co_de.py

@@ -53,7 +53,7 @@ class ImageCoDe(AbsTaskImageTextPairClassification):
 """,
     )
 
-    def load_data(self, num_proc: int = 1, **kwargs) -> None:
+    def load_data(self, num_proc: int | None = None, **kwargs) -> None:
         if self.data_loaded:
             return
 

mteb/tasks/image_text_pair_classification/eng/sugar_crepe.py

@@ -45,7 +45,7 @@ class SugarCrepe(AbsTaskImageTextPairClassification):
 """,
     )
 
-    def load_data(self, num_proc: int = 1, **kwargs) -> None:
+    def load_data(self, num_proc: int | None = None, **kwargs) -> None:
         """Load dataset from HuggingFace hub"""
         if self.data_loaded:
             return

mteb/tasks/instruction_reranking/multilingual/m_follow_ir.py

@@ -175,7 +175,7 @@ class mFollowIRCrossLingual(AbsTaskRetrieval):  # noqa: N801
 """,
     )
 
-    def load_data(self, num_proc: int = 1, **kwargs) -> None:
+    def load_data(self, num_proc: int | None = None, **kwargs) -> None:
         if self.data_loaded:
             return
 
@@ -243,7 +243,7 @@ class mFollowIR(AbsTaskRetrieval):  # noqa: N801
 """,
     )
 
-    def load_data(self, num_proc: int = 1, **kwargs) -> None:
+    def load_data(self, num_proc: int | None = None, **kwargs) -> None:
         if self.data_loaded:
             return
 

mteb/tasks/multichoice/eng/cv_bench.py

@@ -123,7 +123,7 @@ class CVBenchCount(AbsTaskRetrieval):
 """,
     )
 
-    def load_data(self, num_proc: int = 1, **kwargs) -> None:
+    def load_data(self, num_proc: int | None = None, **kwargs) -> None:
         self.corpus, self.queries, self.relevant_docs, self.top_ranked = _load_data(
             path=self.metadata.dataset["path"],
             splits=self.metadata.eval_splits,
@@ -165,7 +165,7 @@ class CVBenchRelation(AbsTaskRetrieval):
 """,
     )
 
-    def load_data(self, num_proc: int = 1, **kwargs) -> None:
+    def load_data(self, num_proc: int | None = None, **kwargs) -> None:
         self.corpus, self.queries, self.relevant_docs, self.top_ranked = _load_data(
             path=self.metadata.dataset["path"],
             splits=self.metadata.eval_splits,
@@ -207,7 +207,7 @@ class CVBenchDepth(AbsTaskRetrieval):
 """,
     )
 
-    def load_data(self, num_proc: int = 1, **kwargs) -> None:
+    def load_data(self, num_proc: int | None = None, **kwargs) -> None:
         self.corpus, self.queries, self.relevant_docs, self.top_ranked = _load_data(
             path=self.metadata.dataset["path"],
             splits=self.metadata.eval_splits,
@@ -249,7 +249,7 @@ class CVBenchDistance(AbsTaskRetrieval):
 """,
     )
 
-    def load_data(self, num_proc: int = 1, **kwargs) -> None:
+    def load_data(self, num_proc: int | None = None, **kwargs) -> None:
         self.corpus, self.queries, self.relevant_docs, self.top_ranked = _load_data(
             path=self.metadata.dataset["path"],
             splits=self.metadata.eval_splits,

mteb/tasks/multilabel_classification/nld/covid_disinformation_nl_multi_label_classification.py

@@ -66,7 +66,7 @@ Yih, Scott Wen-tau},
         },
     )
 
-    def dataset_transform(self, num_proc: int = 1, **kwargs) -> None:
+    def dataset_transform(self, num_proc: int | None = None, **kwargs) -> None:
         labels = [
             "q2_label",
             "q3_label",

mteb/tasks/pair_classification/eng/pub_chem_smilespc.py

@@ -76,7 +76,7 @@ class PubChemSMILESPC(AbsTaskPairClassification):
 """,
     )
 
-    def load_data(self, num_proc: int = 1, **kwargs: Any) -> None:
+    def load_data(self, num_proc: int | None = None, **kwargs: Any) -> None:
         if self.data_loaded:
             return
 

mteb/tasks/pair_classification/multilingual/pub_chem_wiki_pair_classification.py

@@ -60,7 +60,7 @@ class PubChemWikiPairClassification(AbsTaskPairClassification):
 """,
     )
 
-    def dataset_transform(self, num_proc: int = 1, **kwargs) -> None:
+    def dataset_transform(self, num_proc: int | None = None, **kwargs) -> None:
         _dataset = {}
         for lang in self.hf_subsets:
             _dataset[lang] = {}

mteb/tasks/pair_classification/multilingual/rte3.py

@@ -52,7 +52,7 @@ Dolan, Bill},
         # sum of 4 languages after neutral filtering
     )
 
-    def load_data(self, num_proc: int = 1, **kwargs) -> None:
+    def load_data(self, num_proc: int | None = None, **kwargs) -> None:
         """Load dataset from HuggingFace hub"""
         if self.data_loaded:
             return

mteb/tasks/retrieval/ara/sadeem_question_retrieval.py

@@ -37,7 +37,7 @@ class SadeemQuestionRetrieval(AbsTaskRetrieval):
 """,
     )
 
-    def load_data(self, num_proc: int = 1, **kwargs) -> None:
+    def load_data(self, num_proc: int | None = None, **kwargs) -> None:
         if self.data_loaded:
             return
 

mteb/tasks/retrieval/code/code_edit_search_retrieval.py

@@ -53,7 +53,7 @@ class CodeEditSearchRetrieval(AbsTaskRetrieval):
 """,
     )
 
-    def load_data(self, num_proc: int = 1, **kwargs) -> None:
+    def load_data(self, num_proc: int | None = None, **kwargs) -> None:
         if self.data_loaded:
             return