mteb 2.7.16__py3-none-any.whl → 2.7.18__py3-none-any.whl

mteb/models/model_implementations/voyage_models.py

@@ -150,7 +150,7 @@ class VoyageModel(AbsEncoder):
         sentences: list[str],
         batch_size: int,
         input_type: Literal["query", "document"],
-    ) -> np.ndarray:
+    ) -> Array:
         embeddings, index = [], 0
 
         output_dtype = VOYAGE_DTYPE_TRANSLATION.get(

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
 

mteb/tasks/retrieval/code/code_rag.py

@@ -51,7 +51,7 @@ class CodeRAGProgrammingSolutionsRetrieval(AbsTaskRetrieval):
         **common_args,
     )
 
-    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
@@ -59,7 +59,7 @@ class CodeRAGProgrammingSolutionsRetrieval(AbsTaskRetrieval):
         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:
         """And transform to a retrieval dataset, which have the following attributes
 
         self.corpus = Dict[doc_id, Dict[str, str]] #id => dict with document data like title and text
@@ -108,7 +108,7 @@ class CodeRAGOnlineTutorialsRetrieval(AbsTaskRetrieval):
         **common_args,
     )
 
-    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
@@ -116,7 +116,7 @@ class CodeRAGOnlineTutorialsRetrieval(AbsTaskRetrieval):
         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:
         """And transform to a retrieval dataset, which have the following attributes
 
         self.corpus = Dict[doc_id, Dict[str, str]] #id => dict with document data like title and text
@@ -168,7 +168,7 @@ class CodeRAGLibraryDocumentationSolutionsRetrieval(AbsTaskRetrieval):
         **common_args,
     )
 
-    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
@@ -176,7 +176,7 @@ class CodeRAGLibraryDocumentationSolutionsRetrieval(AbsTaskRetrieval):
         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:
         """And transform to a retrieval dataset, which have the following attributes
 
         self.corpus = Dict[doc_id, Dict[str, str]] #id => dict with document data like title and text
@@ -225,7 +225,7 @@ class CodeRAGStackoverflowPostsRetrieval(AbsTaskRetrieval):
         **common_args,
     )
 
-    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
@@ -233,7 +233,7 @@ class CodeRAGStackoverflowPostsRetrieval(AbsTaskRetrieval):
         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:
         """And transform to a retrieval dataset, which have the following attributes
 
         self.corpus = Dict[doc_id, Dict[str, str]] #id => dict with document data like title and text

mteb/tasks/retrieval/code/code_search_net_cc_retrieval.py

@@ -99,7 +99,7 @@ class CodeSearchNetCCRetrieval(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/coir_code_search_net_retrieval.py

@@ -97,7 +97,7 @@ class COIRCodeSearchNetRetrieval(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/ds1000_retrieval.py

@@ -34,7 +34,7 @@ class DS1000Retrieval(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/fresh_stack_retrieval.py

@@ -37,7 +37,7 @@ class FreshStackRetrieval(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/human_eval_retrieval.py

@@ -34,7 +34,7 @@ class HumanEvalRetrieval(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/mbpp_retrieval.py

@@ -34,7 +34,7 @@ class MBPPRetrieval(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/wiki_sql_retrieval.py

@@ -36,7 +36,7 @@ class WikiSQLRetrieval(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