mteb 2.7.1__py3-none-any.whl → 2.7.3__py3-none-any.whl

mteb/tasks/aggregated_tasks/eng/sts17_multilingual_visual_sts_eng.py

@@ -1,10 +1,10 @@
-from mteb.abstasks.abstask import AbsTask
-from mteb.abstasks.aggregated_task import AbsTaskAggregate, AggregateTaskMetadata
+from mteb.abstasks.aggregate_task_metadata import AggregateTaskMetadata
+from mteb.abstasks.aggregated_task import AbsTaskAggregate
 from mteb.tasks.sts.multilingual.sts17_multilingual_visual_sts import (
     STS17MultilingualVisualSTS,
 )
 
-task_list_sts17: list[AbsTask] = [
+task_list_sts17 = [
     STS17MultilingualVisualSTS().filter_languages(
         languages=["eng"], hf_subsets=["en-en"]
     )

mteb/tasks/aggregated_tasks/eng/sts_benchmark_multilingual_visual_sts_eng.py

@@ -1,10 +1,10 @@
-from mteb.abstasks.abstask import AbsTask
-from mteb.abstasks.aggregated_task import AbsTaskAggregate, AggregateTaskMetadata
+from mteb.abstasks.aggregate_task_metadata import AggregateTaskMetadata
+from mteb.abstasks.aggregated_task import AbsTaskAggregate
 from mteb.tasks.sts.multilingual.sts_benchmark_multilingual_visual_sts import (
     STSBenchmarkMultilingualVisualSTS,
 )
 
-task_list_stsb: list[AbsTask] = [
+task_list_stsb = [
     STSBenchmarkMultilingualVisualSTS().filter_languages(
         languages=["eng"], hf_subsets=["en"]
     )

mteb/tasks/aggregated_tasks/fas/cqadupstack_retrieval_fa.py

@@ -1,5 +1,5 @@
-from mteb.abstasks import AbsTask
-from mteb.abstasks.aggregated_task import AbsTaskAggregate, AggregateTaskMetadata
+from mteb.abstasks.aggregate_task_metadata import AggregateTaskMetadata
+from mteb.abstasks.aggregated_task import AbsTaskAggregate
 from mteb.tasks.retrieval import (
     CQADupstackAndroidRetrievalFa,
     CQADupstackEnglishRetrievalFa,
@@ -15,7 +15,7 @@ from mteb.tasks.retrieval import (
     CQADupstackWordpressRetrievalFa,
 )
 
-task_list_cqa: list[AbsTask] = [
+task_list_cqa = [
     CQADupstackAndroidRetrievalFa(),
     CQADupstackEnglishRetrievalFa(),
     CQADupstackGamingRetrievalFa(),

mteb/tasks/aggregated_tasks/fas/syn_per_chatbot_conv_sa_classification.py

@@ -1,5 +1,5 @@
-from mteb.abstasks import AbsTask
-from mteb.abstasks.aggregated_task import AbsTaskAggregate, AggregateTaskMetadata
+from mteb.abstasks.aggregate_task_metadata import AggregateTaskMetadata
+from mteb.abstasks.aggregated_task import AbsTaskAggregate
 from mteb.tasks.classification import (
     SynPerChatbotConvSAAnger,
     SynPerChatbotConvSAFear,
@@ -12,7 +12,7 @@ from mteb.tasks.classification import (
     SynPerChatbotConvSASurprise,
 )
 
-task_list_cqa: list[AbsTask] = [
+task_list_cqa = [
     SynPerChatbotConvSAAnger(),
     SynPerChatbotConvSASatisfaction(),
     SynPerChatbotConvSAFriendship(),

mteb/tasks/aggregated_tasks/multilingual/sts17_multilingual_vision_sts.py

@@ -1,10 +1,10 @@
-from mteb.abstasks.abstask import AbsTask
-from mteb.abstasks.aggregated_task import AbsTaskAggregate, AggregateTaskMetadata
+from mteb.abstasks.aggregate_task_metadata import AggregateTaskMetadata
+from mteb.abstasks.aggregated_task import AbsTaskAggregate
 from mteb.tasks.sts.multilingual.sts17_multilingual_visual_sts import (
     STS17MultilingualVisualSTS,
 )
 
-task_list_sts17_multi: list[AbsTask] = [
+task_list_sts17_multi = [
     STS17MultilingualVisualSTS().filter_languages(
         languages=["ara", "eng", "spa", "kor"],
         hf_subsets=[

mteb/tasks/aggregated_tasks/multilingual/sts_benchmark_multilingual_visual_sts.py

@@ -1,10 +1,10 @@
-from mteb.abstasks.abstask import AbsTask
-from mteb.abstasks.aggregated_task import AbsTaskAggregate, AggregateTaskMetadata
+from mteb.abstasks.aggregate_task_metadata import AggregateTaskMetadata
+from mteb.abstasks.aggregated_task import AbsTaskAggregate
 from mteb.tasks.sts.multilingual.sts_benchmark_multilingual_visual_sts import (
     STSBenchmarkMultilingualVisualSTS,
 )
 
-task_list_multi: list[AbsTask] = [
+task_list_multi = [
     STSBenchmarkMultilingualVisualSTS().filter_languages(
         languages=[
             "deu",

mteb/tasks/aggregated_tasks/nld/cqadupstack_nl_retrieval.py

@@ -1,5 +1,5 @@
-from mteb.abstasks import AbsTask
-from mteb.abstasks.aggregated_task import AbsTaskAggregate, AggregateTaskMetadata
+from mteb.abstasks.aggregate_task_metadata import AggregateTaskMetadata
+from mteb.abstasks.aggregated_task import AbsTaskAggregate
 from mteb.tasks.retrieval import (
     CQADupstackAndroidNLRetrieval,
     CQADupstackEnglishNLRetrieval,
@@ -15,7 +15,7 @@ from mteb.tasks.retrieval import (
     CQADupstackWordpressNLRetrieval,
 )
 
-task_list_cqa: list[AbsTask] = [
+task_list_cqa = [
     CQADupstackAndroidNLRetrieval(),
     CQADupstackEnglishNLRetrieval(),
     CQADupstackGamingNLRetrieval(),

mteb/tasks/aggregated_tasks/pol/cqadupstack_retrieval_pl.py

@@ -1,5 +1,5 @@
-from mteb.abstasks import AbsTask
-from mteb.abstasks.aggregated_task import AbsTaskAggregate, AggregateTaskMetadata
+from mteb.abstasks.aggregate_task_metadata import AggregateTaskMetadata
+from mteb.abstasks.aggregated_task import AbsTaskAggregate
 from mteb.tasks.retrieval.pol.cqadupstack_pl_retrieval import (
     CQADupstackAndroidRetrievalPL,
     CQADupstackEnglishRetrievalPL,
@@ -15,7 +15,7 @@ from mteb.tasks.retrieval.pol.cqadupstack_pl_retrieval import (
     CQADupstackWordpressRetrievalPL,
 )
 
-task_list_cqa: list[AbsTask] = [
+task_list_cqa = [
     CQADupstackAndroidRetrievalPL(),
     CQADupstackEnglishRetrievalPL(),
     CQADupstackGamingRetrievalPL(),

mteb/tasks/clustering/nob/snl_clustering.py

@@ -1,13 +1,18 @@
+from __future__ import annotations
+
 import random
-from collections.abc import Iterable
 from itertools import islice
-from typing import TypeVar
+from typing import TYPE_CHECKING, TypeVar
 
 import datasets
 
 from mteb.abstasks.clustering_legacy import AbsTaskClusteringLegacy
 from mteb.abstasks.task_metadata import TaskMetadata
 
+if TYPE_CHECKING:
+    from collections.abc import Iterable
+
+
 T = TypeVar("T")
 
 

mteb/tasks/clustering/nob/vg_clustering.py

@@ -1,13 +1,18 @@
+from __future__ import annotations
+
 import random
-from collections.abc import Iterable
 from itertools import islice
-from typing import TypeVar
+from typing import TYPE_CHECKING, TypeVar
 
 import datasets
 
 from mteb.abstasks.clustering_legacy import AbsTaskClusteringLegacy
 from mteb.abstasks.task_metadata import TaskMetadata
 
+if TYPE_CHECKING:
+    from collections.abc import Iterable
+
+
 T = TypeVar("T")
 
 

mteb/tasks/retrieval/eng/limit_retrieval.py

@@ -1,8 +1,13 @@
-from collections.abc import Sequence
+from __future__ import annotations
+
+from typing import TYPE_CHECKING
 
 from mteb.abstasks.retrieval import AbsTaskRetrieval
 from mteb.abstasks.task_metadata import TaskMetadata
 
+if TYPE_CHECKING:
+    from collections.abc import Sequence
+
 _CITATION = """
 @misc{weller2025theoreticallimit,
   archiveprefix = {arXiv},

mteb/tasks/retrieval/multilingual/ru_sci_bench_retrieval.py

@@ -30,15 +30,15 @@ def load_ruscibench_data(
 
     for lang in langs:
         lang_corpus = cast(
-            datasets.Dataset,
+            "datasets.Dataset",
             datasets.load_dataset(path, f"corpus-{lang}", revision=revision),
         )["corpus"]
         lang_queries = cast(
-            datasets.Dataset,
+            "datasets.Dataset",
             datasets.load_dataset(path, f"queries-{lang}", revision=revision),
         )["queries"]
         lang_qrels = cast(
-            datasets.Dataset,
+            "datasets.Dataset",
             datasets.load_dataset(path, f"{lang}", revision=revision),
         )["test"]
         corpus[lang] = {

mteb/types/_encoder_io.py

@@ -7,10 +7,10 @@ from typing import TYPE_CHECKING, TypedDict
 import numpy as np
 import torch
 from datasets import Dataset
-from typing_extensions import NotRequired
 
 if TYPE_CHECKING:
     from PIL import Image
+    from typing_extensions import NotRequired
 
 
 class EncodeKwargs(TypedDict):

mteb/types/statistics.py

@@ -1,6 +1,13 @@
-from typing_extensions import NotRequired, TypedDict
+from __future__ import annotations
 
-from mteb.types import HFSubset
+from typing import TYPE_CHECKING
+
+from typing_extensions import TypedDict
+
+if TYPE_CHECKING:
+    from typing_extensions import NotRequired
+
+    from mteb.types import HFSubset
 
 
 class SplitDescriptiveStatistics(TypedDict):

{mteb-2.7.1.dist-info → mteb-2.7.3.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: mteb
-Version: 2.7.1
+Version: 2.7.3
 Summary: Massive Text Embedding Benchmark
 Author-email: MTEB Contributors <niklas@huggingface.co>, Kenneth Enevoldsen <kenneth.enevoldsen@cas.au.dk>, Nouamane Tazi <nouamane@huggingface.co>, Nils Reimers <info@nils-reimers.de>
 Maintainer-email: Kenneth Enevoldsen <kenneth.enevoldsen@cas.au.dk>, Roman Solomatin <risolomatin@gmail.com>, Isaac Chung <chungisaac1217@gmail.com>