mteb 2.1.14__py3-none-any.whl → 2.1.16__py3-none-any.whl
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- mteb/cache.py +6 -2
- mteb/evaluate.py +24 -4
- mteb/models/instruct_wrapper.py +3 -0
- mteb/models/model_implementations/jasper_models.py +206 -6
- mteb/models/model_implementations/spartan8806_atles_champion.py +26 -0
- mteb/results/task_result.py +13 -9
- {mteb-2.1.14.dist-info → mteb-2.1.16.dist-info}/METADATA +2 -2
- {mteb-2.1.14.dist-info → mteb-2.1.16.dist-info}/RECORD +12 -11
- {mteb-2.1.14.dist-info → mteb-2.1.16.dist-info}/WHEEL +0 -0
- {mteb-2.1.14.dist-info → mteb-2.1.16.dist-info}/entry_points.txt +0 -0
- {mteb-2.1.14.dist-info → mteb-2.1.16.dist-info}/licenses/LICENSE +0 -0
- {mteb-2.1.14.dist-info → mteb-2.1.16.dist-info}/top_level.txt +0 -0
mteb/cache.py
CHANGED
|
@@ -62,7 +62,11 @@ class ResultCache:
|
|
|
62
62
|
Returns:
|
|
63
63
|
The path to the results of the task.
|
|
64
64
|
"""
|
|
65
|
-
results_folder =
|
|
65
|
+
results_folder = (
|
|
66
|
+
self.cache_path / "results"
|
|
67
|
+
if not remote
|
|
68
|
+
else self.cache_path / "remote" / "results"
|
|
69
|
+
)
|
|
66
70
|
|
|
67
71
|
if isinstance(model_name, ModelMeta):
|
|
68
72
|
if model_revision is not None:
|
|
@@ -74,7 +78,7 @@ class ResultCache:
|
|
|
74
78
|
elif isinstance(model_name, str):
|
|
75
79
|
model_name = model_name.replace("/", "__").replace(" ", "_")
|
|
76
80
|
|
|
77
|
-
model_path =
|
|
81
|
+
model_path = results_folder / model_name
|
|
78
82
|
|
|
79
83
|
if model_revision is None:
|
|
80
84
|
logger.warning(
|
mteb/evaluate.py
CHANGED
|
@@ -256,6 +256,20 @@ def _check_model_modalities(
|
|
|
256
256
|
logger.warning(msg)
|
|
257
257
|
|
|
258
258
|
|
|
259
|
+
def _requires_merge(task: AbsTask, existing_results: TaskResult) -> bool:
|
|
260
|
+
"""Check if the existing results require merging with new results."""
|
|
261
|
+
# If the task has multiple eval splits and existing results cover only a subset, we need to merge
|
|
262
|
+
required_evals = dict.fromkeys(task.eval_splits, task.hf_subsets)
|
|
263
|
+
for split, subsets in required_evals.items():
|
|
264
|
+
res = existing_results.scores.get(split, None)
|
|
265
|
+
if res is None:
|
|
266
|
+
return True
|
|
267
|
+
hf_subsets = [r["hf_subset"] for r in res]
|
|
268
|
+
if not set(subsets).issubset(set(hf_subsets)):
|
|
269
|
+
return True
|
|
270
|
+
return False
|
|
271
|
+
|
|
272
|
+
|
|
259
273
|
def evaluate(
|
|
260
274
|
model: ModelMeta | MTEBModels | SentenceTransformer | CrossEncoder,
|
|
261
275
|
tasks: AbsTask | Iterable[AbsTask],
|
|
@@ -388,13 +402,18 @@ def evaluate(
|
|
|
388
402
|
|
|
389
403
|
if (
|
|
390
404
|
existing_results
|
|
391
|
-
and overwrite_strategy
|
|
392
|
-
|
|
393
|
-
and
|
|
405
|
+
and overwrite_strategy
|
|
406
|
+
not in (OverwriteStrategy.ALWAYS, OverwriteStrategy.NEVER)
|
|
407
|
+
and (
|
|
408
|
+
not _requires_merge(task, existing_results)
|
|
409
|
+
or existing_results.is_mergeable(task)
|
|
410
|
+
)
|
|
394
411
|
):
|
|
395
412
|
missing_eval = existing_results.get_missing_evaluations(task)
|
|
396
413
|
else:
|
|
397
414
|
missing_eval = dict.fromkeys(task.eval_splits, task.hf_subsets)
|
|
415
|
+
# Will be fully recomputed so we set it to None to avoid merging:
|
|
416
|
+
existing_results = None
|
|
398
417
|
|
|
399
418
|
if (
|
|
400
419
|
existing_results
|
|
@@ -415,7 +434,8 @@ def evaluate(
|
|
|
415
434
|
OverwriteStrategy.ONLY_CACHE,
|
|
416
435
|
]:
|
|
417
436
|
raise ValueError(
|
|
418
|
-
f"overwrite_strategy is set to '{overwrite_strategy.value}' and the results file exists
|
|
437
|
+
f"overwrite_strategy is set to '{overwrite_strategy.value}' and the results file exists for task {task.metadata.name}. "
|
|
438
|
+
+ f"However there are the following missing splits (and subsets): {missing_eval}. To rerun these set overwrite_strategy to 'only-missing'."
|
|
419
439
|
)
|
|
420
440
|
|
|
421
441
|
if existing_results:
|
mteb/models/instruct_wrapper.py
CHANGED
|
@@ -153,6 +153,9 @@ class InstructSentenceTransformerModel(AbsEncoder):
|
|
|
153
153
|
|
|
154
154
|
self.model_name = model_name
|
|
155
155
|
self.model = SentenceTransformer(model_name, revision=revision, **kwargs)
|
|
156
|
+
if max_seq_length:
|
|
157
|
+
# https://github.com/huggingface/sentence-transformers/issues/3575
|
|
158
|
+
self.model.max_seq_length = max_seq_length
|
|
156
159
|
self.apply_instruction_to_passages = apply_instruction_to_passages
|
|
157
160
|
self.prompts_dict = prompts_dict
|
|
158
161
|
|
|
@@ -21,6 +21,198 @@ from mteb.types import Array, BatchedInput, PromptType
|
|
|
21
21
|
|
|
22
22
|
logger = logging.getLogger(__name__)
|
|
23
23
|
|
|
24
|
+
jasper_token_compression_600m_prompts_dict = {
|
|
25
|
+
"AFQMC": "Retrieve semantically similar text",
|
|
26
|
+
"AILACasedocs": {
|
|
27
|
+
"query": "Given a web search query, retrieve relevant passages that answer the query",
|
|
28
|
+
"document": "",
|
|
29
|
+
},
|
|
30
|
+
"AILAStatutes": {
|
|
31
|
+
"query": "Given a web search query, retrieve relevant passages that answer the query",
|
|
32
|
+
"document": "",
|
|
33
|
+
},
|
|
34
|
+
"ATEC": "Retrieve semantically similar text",
|
|
35
|
+
"AmazonCounterfactualClassification": "Classify a given Amazon customer review text as either counterfactual or not-counterfactual",
|
|
36
|
+
"ArXivHierarchicalClusteringP2P": "Identify the main and secondary category of Arxiv papers based on the titles and abstracts",
|
|
37
|
+
"ArXivHierarchicalClusteringS2S": "Identify the main and secondary category of Arxiv papers based on the titles",
|
|
38
|
+
"ArguAna": {
|
|
39
|
+
"query": "Given a claim, find documents that refute the claim",
|
|
40
|
+
"document": "Given a claim, find documents that refute the claim",
|
|
41
|
+
},
|
|
42
|
+
"AskUbuntuDupQuestions": {
|
|
43
|
+
"query": "Retrieve duplicate questions from AskUbuntu forum",
|
|
44
|
+
"document": "",
|
|
45
|
+
},
|
|
46
|
+
"BIOSSES": "Retrieve semantically similar text",
|
|
47
|
+
"BQ": "Retrieve semantically similar text",
|
|
48
|
+
"Banking77Classification": "Given a online banking query, find the corresponding intents",
|
|
49
|
+
"BiorxivClusteringP2P.v2": "Identify the main category of Biorxiv papers based on the titles and abstracts",
|
|
50
|
+
"CLSClusteringP2P": "Identify the main category of scholar papers based on the titles and abstracts",
|
|
51
|
+
"CLSClusteringS2S": "Identify the main category of scholar papers based on the titles",
|
|
52
|
+
"CMedQAv1-reranking": {
|
|
53
|
+
"query": "Given a Chinese community medical question, retrieve replies that best answer the question",
|
|
54
|
+
"document": "",
|
|
55
|
+
},
|
|
56
|
+
"CMedQAv2-reranking": {
|
|
57
|
+
"query": "Given a Chinese community medical question, retrieve replies that best answer the question",
|
|
58
|
+
"document": "",
|
|
59
|
+
},
|
|
60
|
+
"CQADupstackGamingRetrieval": {
|
|
61
|
+
"query": "Given a question, retrieve detailed question descriptions from Stackexchange that are duplicates to the given question",
|
|
62
|
+
"document": "Given a question, retrieve detailed question descriptions from Stackexchange that are duplicates to the given question",
|
|
63
|
+
},
|
|
64
|
+
"CQADupstackUnixRetrieval": {
|
|
65
|
+
"query": "Given a question, retrieve detailed question descriptions from Stackexchange that are duplicates to the given question",
|
|
66
|
+
"document": "Given a question, retrieve detailed question descriptions from Stackexchange that are duplicates to the given question",
|
|
67
|
+
},
|
|
68
|
+
"ClimateFEVERHardNegatives": {
|
|
69
|
+
"query": "Given a claim about climate change, retrieve documents that support or refute the claim",
|
|
70
|
+
"document": "",
|
|
71
|
+
},
|
|
72
|
+
"CmedqaRetrieval": {
|
|
73
|
+
"query": "Given a Chinese community medical question, retrieve replies that best answer the question",
|
|
74
|
+
"document": "",
|
|
75
|
+
},
|
|
76
|
+
"Cmnli": "Retrieve semantically similar text.",
|
|
77
|
+
"CovidRetrieval": {
|
|
78
|
+
"query": "Given a question on COVID-19, retrieve news articles that answer the question",
|
|
79
|
+
"document": "",
|
|
80
|
+
},
|
|
81
|
+
"DuRetrieval": {
|
|
82
|
+
"query": "Given a Chinese search query, retrieve web passages that answer the question",
|
|
83
|
+
"document": "",
|
|
84
|
+
},
|
|
85
|
+
"EcomRetrieval": {
|
|
86
|
+
"query": "Given a user query from an e-commerce website, retrieve description sentences of relevant products",
|
|
87
|
+
"document": "",
|
|
88
|
+
},
|
|
89
|
+
"FEVERHardNegatives": {
|
|
90
|
+
"query": "Given a claim, retrieve documents that support or refute the claim",
|
|
91
|
+
"document": "",
|
|
92
|
+
},
|
|
93
|
+
"FiQA2018": {
|
|
94
|
+
"query": "Given a financial question, retrieve user replies that best answer the question",
|
|
95
|
+
"document": "",
|
|
96
|
+
},
|
|
97
|
+
"GerDaLIRSmall": {
|
|
98
|
+
"query": "Given a web search query, retrieve relevant passages that answer the query",
|
|
99
|
+
"document": "",
|
|
100
|
+
},
|
|
101
|
+
"HotpotQAHardNegatives": {
|
|
102
|
+
"query": "Given a multi-hop question, retrieve documents that can help answer the question",
|
|
103
|
+
"document": "",
|
|
104
|
+
},
|
|
105
|
+
"IFlyTek": "Given an App description text, find the appropriate fine-grained category",
|
|
106
|
+
"ImdbClassification": "Classify the sentiment expressed in the given movie review text from the IMDB dataset",
|
|
107
|
+
"JDReview": "Classify the customer review for iPhone on e-commerce platform into positive or negative",
|
|
108
|
+
"LCQMC": "Retrieve semantically similar text",
|
|
109
|
+
"LeCaRDv2": {
|
|
110
|
+
"query": "Given a web search query, retrieve relevant passages that answer the query",
|
|
111
|
+
"document": "",
|
|
112
|
+
},
|
|
113
|
+
"LegalBenchConsumerContractsQA": {
|
|
114
|
+
"query": "Given a web search query, retrieve relevant passages that answer the query",
|
|
115
|
+
"document": "",
|
|
116
|
+
},
|
|
117
|
+
"LegalBenchCorporateLobbying": {
|
|
118
|
+
"query": "Given a web search query, retrieve relevant passages that answer the query",
|
|
119
|
+
"document": "",
|
|
120
|
+
},
|
|
121
|
+
"LegalQuAD": {
|
|
122
|
+
"query": "Given a web search query, retrieve relevant passages that answer the query",
|
|
123
|
+
"document": "",
|
|
124
|
+
},
|
|
125
|
+
"LegalSummarization": {
|
|
126
|
+
"query": "Given a web search query, retrieve relevant passages that answer the query",
|
|
127
|
+
"document": "",
|
|
128
|
+
},
|
|
129
|
+
"MMarcoReranking": {
|
|
130
|
+
"query": "Given a Chinese search query, retrieve web passages that answer the question",
|
|
131
|
+
"document": "",
|
|
132
|
+
},
|
|
133
|
+
"MMarcoRetrieval": {
|
|
134
|
+
"query": "Given a web search query, retrieve relevant passages that answer the query",
|
|
135
|
+
"document": "",
|
|
136
|
+
},
|
|
137
|
+
"MTOPDomainClassification": "Classify the intent domain of the given utterance in task-oriented conversation",
|
|
138
|
+
"MassiveIntentClassification": "Given a user utterance as query, find the user intents",
|
|
139
|
+
"MassiveScenarioClassification": "Given a user utterance as query, find the user scenarios",
|
|
140
|
+
"MedicalRetrieval": {
|
|
141
|
+
"query": "Given a medical question, retrieve user replies that best answer the question",
|
|
142
|
+
"document": "",
|
|
143
|
+
},
|
|
144
|
+
"MedrxivClusteringP2P.v2": "Identify the main category of Medrxiv papers based on the titles and abstracts",
|
|
145
|
+
"MedrxivClusteringS2S.v2": "Identify the main category of Medrxiv papers based on the titles",
|
|
146
|
+
"MindSmallReranking": {
|
|
147
|
+
"query": "Retrieve relevant news articles based on user browsing history",
|
|
148
|
+
"document": "",
|
|
149
|
+
},
|
|
150
|
+
"MultilingualSentiment": "Classify sentiment of the customer review into positive, neutral, or negative",
|
|
151
|
+
"Ocnli": "Retrieve semantically similar text.",
|
|
152
|
+
"OnlineShopping": "Classify the customer review for online shopping into positive or negative",
|
|
153
|
+
"PAWSX": "Retrieve semantically similar text",
|
|
154
|
+
"QBQTC": "Retrieve semantically similar text",
|
|
155
|
+
"SCIDOCS": {
|
|
156
|
+
"query": "Given a scientific paper title, retrieve paper abstracts that are cited by the given paper",
|
|
157
|
+
"document": "",
|
|
158
|
+
},
|
|
159
|
+
"SICK-R": "Retrieve semantically similar text",
|
|
160
|
+
"STS12": "Retrieve semantically similar text",
|
|
161
|
+
"STS13": "Retrieve semantically similar text",
|
|
162
|
+
"STS14": "Retrieve semantically similar text",
|
|
163
|
+
"STS15": "Retrieve semantically similar text",
|
|
164
|
+
"STS17": "Retrieve semantically similar text",
|
|
165
|
+
"STS22.v2": "Retrieve semantically similar text",
|
|
166
|
+
"STSB": "Retrieve semantically similar text",
|
|
167
|
+
"STSBenchmark": "Retrieve semantically similar text",
|
|
168
|
+
"SprintDuplicateQuestions": "Retrieve duplicate questions from Sprint forum",
|
|
169
|
+
"StackExchangeClustering.v2": "Identify the topic or theme of StackExchange posts based on the titles",
|
|
170
|
+
"StackExchangeClusteringP2P.v2": "Identify the topic or theme of StackExchange posts based on the given paragraphs",
|
|
171
|
+
"SummEvalSummarization.v2": "Retrieve semantically similar text",
|
|
172
|
+
"T2Reranking": {
|
|
173
|
+
"query": "Given a Chinese search query, retrieve web passages that answer the question",
|
|
174
|
+
"document": "",
|
|
175
|
+
},
|
|
176
|
+
"T2Retrieval": {
|
|
177
|
+
"query": "Given a Chinese search query, retrieve web passages that answer the question",
|
|
178
|
+
"document": "",
|
|
179
|
+
},
|
|
180
|
+
"TNews": "Classify the fine-grained category of the given news title",
|
|
181
|
+
"TRECCOVID": {
|
|
182
|
+
"query": "Given a query on COVID-19, retrieve documents that answer the query",
|
|
183
|
+
"document": "",
|
|
184
|
+
},
|
|
185
|
+
"ThuNewsClusteringP2P": "Identify the topic or theme of the given news articles based on the titles and contents",
|
|
186
|
+
"ThuNewsClusteringS2S": "Identify the topic or theme of the given news articles based on the titles",
|
|
187
|
+
"Touche2020Retrieval.v3": {
|
|
188
|
+
"query": "Given a question, retrieve detailed and persuasive arguments that answer the question",
|
|
189
|
+
"document": "",
|
|
190
|
+
},
|
|
191
|
+
"ToxicConversationsClassification": "Classify the given comments as either toxic or not toxic",
|
|
192
|
+
"TweetSentimentExtractionClassification": "Classify the sentiment of a given tweet as either positive, negative, or neutral",
|
|
193
|
+
"TwentyNewsgroupsClustering.v2": "Identify the topic or theme of the given news articles",
|
|
194
|
+
"TwitterSemEval2015": "Retrieve tweets that are semantically similar to the given tweet",
|
|
195
|
+
"TwitterURLCorpus": "Retrieve tweets that are semantically similar to the given tweet",
|
|
196
|
+
"VideoRetrieval": {
|
|
197
|
+
"query": "Given a video search query, retrieve the titles of relevant videos",
|
|
198
|
+
"document": "",
|
|
199
|
+
},
|
|
200
|
+
"Waimai": "Classify the customer review from a food takeaway platform into positive or negative",
|
|
201
|
+
}
|
|
202
|
+
jasper_token_compression_600m_loader_kwargs = dict(
|
|
203
|
+
model_kwargs={
|
|
204
|
+
"attn_implementation": "sdpa",
|
|
205
|
+
"torch_dtype": "bfloat16",
|
|
206
|
+
"trust_remote_code": True,
|
|
207
|
+
},
|
|
208
|
+
tokenizer_kwargs={"padding_side": "left"},
|
|
209
|
+
trust_remote_code=True,
|
|
210
|
+
prompts_dict=jasper_token_compression_600m_prompts_dict,
|
|
211
|
+
apply_instruction_to_passages=True,
|
|
212
|
+
instruction_template="Instruct: {instruction}\nQuery: ",
|
|
213
|
+
max_seq_length=1024,
|
|
214
|
+
)
|
|
215
|
+
|
|
24
216
|
|
|
25
217
|
def instruction_template(
|
|
26
218
|
instruction: str, prompt_type: PromptType | None = None
|
|
@@ -135,13 +327,10 @@ jasper_en_v1 = ModelMeta(
|
|
|
135
327
|
""",
|
|
136
328
|
)
|
|
137
329
|
|
|
330
|
+
|
|
138
331
|
Jasper_Token_Compression_600M = ModelMeta(
|
|
139
332
|
loader=InstructSentenceTransformerModel,
|
|
140
|
-
loader_kwargs=
|
|
141
|
-
instruction_template=instruction_template,
|
|
142
|
-
apply_instruction_to_passages=False,
|
|
143
|
-
trust_remote_code=True,
|
|
144
|
-
),
|
|
333
|
+
loader_kwargs=jasper_token_compression_600m_loader_kwargs,
|
|
145
334
|
name="infgrad/Jasper-Token-Compression-600M",
|
|
146
335
|
languages=["eng-Latn", "zho-Hans"],
|
|
147
336
|
open_weights=True,
|
|
@@ -156,7 +345,7 @@ Jasper_Token_Compression_600M = ModelMeta(
|
|
|
156
345
|
similarity_fn_name="cosine",
|
|
157
346
|
framework=["Sentence Transformers", "PyTorch"],
|
|
158
347
|
use_instructions=True,
|
|
159
|
-
public_training_code=
|
|
348
|
+
public_training_code="https://github.com/DunZhang/Jasper-Token-Compression-Training",
|
|
160
349
|
# public_training_data: unsupervised data for distillation
|
|
161
350
|
public_training_data="https://huggingface.co/datasets/infgrad/jasper_text_distill_dataset",
|
|
162
351
|
training_datasets=bge_m3_training_data
|
|
@@ -164,4 +353,15 @@ Jasper_Token_Compression_600M = ModelMeta(
|
|
|
164
353
|
| bge_full_data
|
|
165
354
|
| E5_MISTRAL_TRAINING_DATA
|
|
166
355
|
| qzhou_training_data,
|
|
356
|
+
citation="""
|
|
357
|
+
@misc{zhang2025jaspertokencompression600mtechnicalreport,
|
|
358
|
+
title={Jasper-Token-Compression-600M Technical Report},
|
|
359
|
+
author={Dun Zhang and Ziyang Zeng and Yudong Zhou and Shuyang Lu},
|
|
360
|
+
year={2025},
|
|
361
|
+
eprint={2511.14405},
|
|
362
|
+
archivePrefix={arXiv},
|
|
363
|
+
primaryClass={cs.IR},
|
|
364
|
+
url={https://arxiv.org/abs/2511.14405},
|
|
365
|
+
}
|
|
366
|
+
""",
|
|
167
367
|
)
|
|
@@ -0,0 +1,26 @@
|
|
|
1
|
+
"""ATLES Champion Embedding Model for MTEB."""
|
|
2
|
+
|
|
3
|
+
from mteb.models.model_meta import ModelMeta
|
|
4
|
+
from mteb.models.sentence_transformer_wrapper import sentence_transformers_loader
|
|
5
|
+
|
|
6
|
+
spartan8806_atles_champion_embedding = ModelMeta(
|
|
7
|
+
loader=sentence_transformers_loader,
|
|
8
|
+
name="spartan8806/atles-champion-embedding",
|
|
9
|
+
languages=["eng-Latn"],
|
|
10
|
+
open_weights=True,
|
|
11
|
+
revision="d4c74d7000bbd25f3597fc0f2dcde59ef1386e8f",
|
|
12
|
+
release_date="2025-11-15",
|
|
13
|
+
n_parameters=110_000_000,
|
|
14
|
+
memory_usage_mb=420,
|
|
15
|
+
max_tokens=512,
|
|
16
|
+
embed_dim=768,
|
|
17
|
+
license="apache-2.0",
|
|
18
|
+
similarity_fn_name="cosine",
|
|
19
|
+
framework=["Sentence Transformers"],
|
|
20
|
+
reference="https://huggingface.co/spartan8806/atles-champion-embedding",
|
|
21
|
+
use_instructions=False,
|
|
22
|
+
training_datasets={"STSBenchmark"},
|
|
23
|
+
adapted_from="sentence-transformers/all-mpnet-base-v2",
|
|
24
|
+
public_training_code=None,
|
|
25
|
+
public_training_data=None,
|
|
26
|
+
)
|
mteb/results/task_result.py
CHANGED
|
@@ -698,27 +698,31 @@ class TaskResult(BaseModel):
|
|
|
698
698
|
name = result.metadata.name
|
|
699
699
|
revision = result.metadata.revision
|
|
700
700
|
else:
|
|
701
|
+
msg = "result must be a TaskResult or AbsTask object"
|
|
702
|
+
if raise_error:
|
|
703
|
+
raise ValueError(msg)
|
|
704
|
+
logger.debug(msg)
|
|
701
705
|
return False
|
|
702
706
|
|
|
703
707
|
if self.task_name != name:
|
|
708
|
+
msg = f"Cannot merge TaskResult objects as they are derived from different tasks ({self.task_name} and {name})"
|
|
704
709
|
if raise_error:
|
|
705
|
-
raise ValueError(
|
|
706
|
-
|
|
707
|
-
)
|
|
710
|
+
raise ValueError(msg)
|
|
711
|
+
logger.debug(msg)
|
|
708
712
|
return False
|
|
709
713
|
|
|
710
714
|
if Criteria.MTEB_VERSION in criteria and self.mteb_version != mteb_version:
|
|
715
|
+
msg = f"Cannot merge TaskResult objects as they are derived from different MTEB versions ({self.mteb_version} (loaded) and {mteb_version} (current))"
|
|
711
716
|
if raise_error:
|
|
712
|
-
raise ValueError(
|
|
713
|
-
|
|
714
|
-
)
|
|
717
|
+
raise ValueError(msg)
|
|
718
|
+
logger.debug(msg)
|
|
715
719
|
return False
|
|
716
720
|
|
|
717
721
|
if Criteria.DATASET_REVISION in criteria and self.dataset_revision != revision:
|
|
722
|
+
msg = f"Cannot merge TaskResult objects as they are derived from different dataset revisions ({self.dataset_revision} and {revision})"
|
|
718
723
|
if raise_error:
|
|
719
|
-
raise ValueError(
|
|
720
|
-
|
|
721
|
-
)
|
|
724
|
+
raise ValueError(msg)
|
|
725
|
+
logger.debug(msg)
|
|
722
726
|
return False
|
|
723
727
|
|
|
724
728
|
return True
|
|
@@ -1,6 +1,6 @@
|
|
|
1
1
|
Metadata-Version: 2.4
|
|
2
2
|
Name: mteb
|
|
3
|
-
Version: 2.1.
|
|
3
|
+
Version: 2.1.16
|
|
4
4
|
Summary: Massive Text Embedding Benchmark
|
|
5
5
|
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>
|
|
6
6
|
Maintainer-email: Kenneth Enevoldsen <kenneth.enevoldsen@cas.au.dk>, Roman Solomatin <risolomatin@gmail.com>, Isaac Chung <chungisaac1217@gmail.com>
|
|
@@ -37,7 +37,7 @@ Requires-Dist: torchvision>0.2.1; extra == "image"
|
|
|
37
37
|
Provides-Extra: codecarbon
|
|
38
38
|
Requires-Dist: codecarbon<3.0.0,>=2.0.0; extra == "codecarbon"
|
|
39
39
|
Provides-Extra: leaderboard
|
|
40
|
-
Requires-Dist: gradio==5.
|
|
40
|
+
Requires-Dist: gradio==5.49.1; extra == "leaderboard"
|
|
41
41
|
Requires-Dist: plotly<6.0.0,>=5.24.0; extra == "leaderboard"
|
|
42
42
|
Requires-Dist: cachetools>=5.2.0; extra == "leaderboard"
|
|
43
43
|
Requires-Dist: matplotlib>=3.9.4; extra == "leaderboard"
|
|
@@ -5,9 +5,9 @@ mteb/_helpful_enum.py,sha256=jh73N1jlcpg7RGz4bj8UpctiMNvqvHpp9wrB7SYEzIU,510
|
|
|
5
5
|
mteb/_log_once.py,sha256=-tUKzxGQzf2LZSuQXi97oYFXMta1B6GEYXd7BPqssvY,1095
|
|
6
6
|
mteb/_requires_package.py,sha256=eHg_TD9BVZRzNCcQQrUP17d8M1DF_vOd_tVx54AmAnM,3017
|
|
7
7
|
mteb/_set_seed.py,sha256=HPlPRl__Pe6IG-4UgJqTfplcivJ_wA2kaClbXoHQedM,1178
|
|
8
|
-
mteb/cache.py,sha256=
|
|
8
|
+
mteb/cache.py,sha256=AFCxgjODLg1-BJDiBC0F_fpc-PVNNgbugo84i4Ft0ZE,20088
|
|
9
9
|
mteb/deprecated_evaluator.py,sha256=t13Eluvm5ByVIOqgT7fqiVfLb8Ud3A4bbF2djRfs8iA,26901
|
|
10
|
-
mteb/evaluate.py,sha256=
|
|
10
|
+
mteb/evaluate.py,sha256=Pm0b9cfDHKVJ1fluNM3M9A4V3bO_FPvLNvTvwd2_tWk,17996
|
|
11
11
|
mteb/filter_tasks.py,sha256=5XE1OYmgDDoJYnXwFf4ma_PIT_Lekzs420sQF_kpCiY,7240
|
|
12
12
|
mteb/get_tasks.py,sha256=6Gc18a2bZoLQV1Ms_qdr2KieAqIXg8TDg4l7ZN8rW2I,14218
|
|
13
13
|
mteb/load_results.py,sha256=Xw2ZX7BToU92WwUTQUQKPAgPhX7ucyRRdoCrxAoPHdI,6414
|
|
@@ -1430,7 +1430,7 @@ mteb/leaderboard/text_segments.py,sha256=iMIkS04QQjPbT-SkU0x6fOcS8xRbUYevryu9Hyd
|
|
|
1430
1430
|
mteb/models/__init__.py,sha256=ycGU-x60LT0OFyP4CYa5pQhM7J5hCimubuT56va9wfM,741
|
|
1431
1431
|
mteb/models/abs_encoder.py,sha256=m0JkRfRPMYadDgBR9eozRloI31ZSWkSzDFINpwbfLZk,16533
|
|
1432
1432
|
mteb/models/get_model_meta.py,sha256=VpZZNINk-QrNeVpPZnlqzlLhtBs8G84eRwTzAb_gRD4,9108
|
|
1433
|
-
mteb/models/instruct_wrapper.py,sha256=
|
|
1433
|
+
mteb/models/instruct_wrapper.py,sha256=Ty4nfEvioycL_uATkhd0PGuyeB5Xc9xrRd6HOGgb-tc,9005
|
|
1434
1434
|
mteb/models/model_meta.py,sha256=b-Nel9nX5bJk4cgJnqkBzEKyMY7uXvxlCBSxmmH1Ios,14769
|
|
1435
1435
|
mteb/models/models_protocols.py,sha256=D2hYWn_UBGMaKtRwBx3u0B0ni6lHJjSzTxX21XFNwIc,8917
|
|
1436
1436
|
mteb/models/search_wrappers.py,sha256=qe2APunvRfPREdrq1moSi44mFXV6uaHvGHcLnaza-Sc,15483
|
|
@@ -1483,7 +1483,7 @@ mteb/models/model_implementations/hinvec_models.py,sha256=I_d_dSNVaGIwMIwyvTlaPA
|
|
|
1483
1483
|
mteb/models/model_implementations/human.py,sha256=klMpuMAtYH92EIEwNMEhne_Baf9fNiTg1DNWYD11P44,532
|
|
1484
1484
|
mteb/models/model_implementations/ibm_granite_models.py,sha256=YCT0jbgawy19ps5l8QlxpQoJLjq8Nh-3R-e6yxS0DRM,7902
|
|
1485
1485
|
mteb/models/model_implementations/inf_models.py,sha256=lvXUFhAYDltq2_Xa9MHcwfhh1V20rbJLSgON76tkj6w,2906
|
|
1486
|
-
mteb/models/model_implementations/jasper_models.py,sha256=
|
|
1486
|
+
mteb/models/model_implementations/jasper_models.py,sha256=ZY7qRRpBpD3eVryQb4rLs5E3KDXlgFBvyelataqLIWs,16213
|
|
1487
1487
|
mteb/models/model_implementations/jina_clip.py,sha256=CfiIxbhKspjQajNtObCfGPHOWPk6uLn4cuwydQHFTMo,5118
|
|
1488
1488
|
mteb/models/model_implementations/jina_models.py,sha256=QWoesiTygdFTLcdGpdx26wOUI1AXRz3jLmxGHJ0WMNE,29919
|
|
1489
1489
|
mteb/models/model_implementations/kalm_models.py,sha256=FmW7Z5Qs6WYBLuKvql3u4IJW36kj4k-Ypah8qTBEBkg,59837
|
|
@@ -1536,6 +1536,7 @@ mteb/models/model_implementations/sentence_transformers_models.py,sha256=EtEaXg1
|
|
|
1536
1536
|
mteb/models/model_implementations/shuu_model.py,sha256=KkcuVYjIzoha3Fvxh8ppqHQ9BfNMWeqDqn9dGCRKUjg,1167
|
|
1537
1537
|
mteb/models/model_implementations/siglip_models.py,sha256=tvi8QB2ayBoeXsxwHrl5RFlkknvE6FM9N06zSBWGQD0,12602
|
|
1538
1538
|
mteb/models/model_implementations/sonar_models.py,sha256=Nc6kAJRWSrxA57DPRrgOPHqS1dNhz2vsE_1ZA2JtigQ,4784
|
|
1539
|
+
mteb/models/model_implementations/spartan8806_atles_champion.py,sha256=9sWQH7tOT0uxXA7sbQcnqGt2f5O9xcw9HqFpRCzoQAA,918
|
|
1539
1540
|
mteb/models/model_implementations/stella_models.py,sha256=NL3tk-rnuBdznsQ-nmelqun4tFO2xKoNPPOOVKqnPGU,8062
|
|
1540
1541
|
mteb/models/model_implementations/tarka_models.py,sha256=xC6olJs9PSe_lrYsScw5hDHTjYSjcxgbvfK_7IoBFnk,27397
|
|
1541
1542
|
mteb/models/model_implementations/text2vec_models.py,sha256=zaHWRc2W0RYZAOetinqRzug9UGW0HmY5U-jYsLXA8wo,4160
|
|
@@ -1552,7 +1553,7 @@ mteb/models/model_implementations/youtu_models.py,sha256=NB74E6z-_36HyXb8GXKn8Cr
|
|
|
1552
1553
|
mteb/results/__init__.py,sha256=EXQqK4Am5eIYzD52dpcGAFSdqnC38oE6JHN302oidHc,158
|
|
1553
1554
|
mteb/results/benchmark_results.py,sha256=OWqeBxbNsPmOKRhxY980N5CikpdJXToDGJGTXUe64Lw,18209
|
|
1554
1555
|
mteb/results/model_result.py,sha256=Wdbkpxq7_geliYDr4558i6txDVdsHL-Y9WAv_u7thlI,13689
|
|
1555
|
-
mteb/results/task_result.py,sha256=
|
|
1556
|
+
mteb/results/task_result.py,sha256=RqUhmWqxraQM7KP8EDvEcFzVf80it4tMkQNq9YVcsME,32005
|
|
1556
1557
|
mteb/tasks/__init__.py,sha256=izAxU0ip1F_YUwx0dFCuN35BaktdmePh6vlDiHC0kLo,503
|
|
1557
1558
|
mteb/tasks/aggregated_tasks/__init__.py,sha256=Ufgbh1AirxCQkojO3AUhUFWM8zQG10cfdVTkj_PeyLI,104
|
|
1558
1559
|
mteb/tasks/aggregated_tasks/eng/__init__.py,sha256=HgaSyAX8Is5CGE006RgJkLQQVxrx2FmMnm6NHQBDi-4,358
|
|
@@ -2554,9 +2555,9 @@ mteb/types/_metadata.py,sha256=NN-W0S6a5TDV7UkpRx1pyWtGF4TyyCyoPUfHOwdeci8,2290
|
|
|
2554
2555
|
mteb/types/_result.py,sha256=CRAUc5IvqI3_9SyXDwv-PWLCXwXdZem9RePeYESRtuw,996
|
|
2555
2556
|
mteb/types/_string_validators.py,sha256=PY-dYq4E8O50VS3bLYdldPWp400fl_WzUjfVSkNWe8U,523
|
|
2556
2557
|
mteb/types/statistics.py,sha256=YwJsxTf1eaCI_RE-J37a-gK5wDeGAsmkeZKoZCFihSo,3755
|
|
2557
|
-
mteb-2.1.
|
|
2558
|
-
mteb-2.1.
|
|
2559
|
-
mteb-2.1.
|
|
2560
|
-
mteb-2.1.
|
|
2561
|
-
mteb-2.1.
|
|
2562
|
-
mteb-2.1.
|
|
2558
|
+
mteb-2.1.16.dist-info/licenses/LICENSE,sha256=xx0jnfkXJvxRnG63LTGOxlggYnIysveWIZ6H3PNdCrQ,11357
|
|
2559
|
+
mteb-2.1.16.dist-info/METADATA,sha256=fTc0lzu8SewlGWKPCZdDDto9JVN_Rp3xOoQEN4pOuIs,13574
|
|
2560
|
+
mteb-2.1.16.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
|
|
2561
|
+
mteb-2.1.16.dist-info/entry_points.txt,sha256=8IJoEJFKoDHmVnNev-qJ9pp4Ln7_1-ma9QsXnzVCzGU,39
|
|
2562
|
+
mteb-2.1.16.dist-info/top_level.txt,sha256=OLVIjcQAlWBz0bdmutKlWHLF42FF0hp4uVAg3ZyiG4U,5
|
|
2563
|
+
mteb-2.1.16.dist-info/RECORD,,
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|