-
Notifications
You must be signed in to change notification settings - Fork 466
Remove duplicated string in docstring of TaskMetadata class #2087
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.
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
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Samoed
approved these changes
Feb 17, 2025
silky1708
pushed a commit
to silky1708/mteb
that referenced
this pull request
Mar 10, 2025
…gs-benchmark#2087) * Remove duplicated string in docstring of TaskMetadata class * Remove duplicated dataset field
isaac-chung
added a commit
that referenced
this pull request
Mar 13, 2025
* misc: Add image classification descriptive stats implementation (#2045) * add ImageClassificationDescriptiveStatistics * add MNIST descriptive stats * use tuples instead * add label count and update docstrings * update MNIST example * Update tasks table * fix: Add column descriptions to leaderboard (#2039) * fix: Add column descriptions to leaderboard * removed existing overlap * fix: Add BRIGHT (long) and fix bug in TaskResult.filter_and_validate() (#2041) * fix: Add BRIGHT Long Fixes #1978 * fix: Add BRIGHT(long) * fix bug in task results * updated bright * updated tests for TaskResults * 1.34.12 Automatically generated by python-semantic-release * misc: Add image clustering descriptive stats implementation (#2057) * add image clustering descirptive stats and run * finish off last one * remove script * fix: Update embed_dim for jina models (#2058) see embeddings-benchmark/results#117 * Update tasks table * 1.34.13 Automatically generated by python-semantic-release * Add giga embeddings (#1741) * add gigaembeddings * use jasper * fix name * create sentence_transformer instruct wrapper * apply instruction template * fix jasper * update meta * misc: Add ZS and multilabel image classification descriptive stats implementation (#2059) * add image clustering descirptive stats and run * finish off last one * remove script * add ImageMultilabelClassificationDescriptiveStatistics * add VOC2007 * add zeroshot and mnist example * Update tasks table * Rename MIEB task classes with duplicated names (#2061) fix class names * misc: Add VisualSTS descriptive stats (#2062) * add visualsts stats * add last dataset * Update tasks table * fix: Added gte models (#1539) * fix: Added gte models * fix: Add mixbai models (#1540) for #1515 * fix: Add climate fever v2 (#1873) * Updated ClimateFEVER dataset with new version * Adds Fill in the empty metadata. * Updates the date tuple * Update class name Co-authored-by: Kenneth Enevoldsen <[email protected]> * Update domains Co-authored-by: Kenneth Enevoldsen <[email protected]> * Update task_subtypes * Update annotations_creators for the first version * Update date Co-authored-by: Kenneth Enevoldsen <[email protected]> * Update task subtypes * Update path * Update description --------- Co-authored-by: Kenneth Enevoldsen <[email protected]> Co-authored-by: Mina Parham <[email protected]> * Update tasks table * fix: Updating paper scripts (#1958) * change reference revisions to align with paper * Update author list * Added code for main results table * updated minor changes * added external as a "no_revision_available" case * revert unintended changes * format * 1.34.14 Automatically generated by python-semantic-release * Add datasets for a benchmark newly introduced for "Engineering" domain (#1911) * adding clustering tasks (built-bench-clustering S2S & P2P) * updated built-bench-clustering tasks * Updated BuiltBenchClustering tasks * Added "Engineering" as new domain to TaskMetadata.py * Updated tasks table in docs * Updated task metadata for BuiltBenchClustering S2S and P2P * updated metadata for clustering tasks * Add/update BuiltBench tasks - Add BuiltBenchRetrieval task - Add BuiltBenchReranking task - Update metadata for BuiltBenchClusterinP2P - Update metadata for BuiltBenchClusterinS2S * update BuiltBench benchmark * Update mteb/benchmarks/benchmarks.py Co-authored-by: Roman Solomatin <[email protected]> * Update mteb/tasks/Clustering/eng/BuiltBenchClusteringS2S.py Co-authored-by: Roman Solomatin <[email protected]> * Update mteb/tasks/Clustering/eng/BuiltBenchClusteringP2P.py Co-authored-by: Roman Solomatin <[email protected]> * Update mteb/benchmarks/benchmarks.py Co-authored-by: Isaac Chung <[email protected]> * Fix formatting via ruff --------- Co-authored-by: Roman Solomatin <[email protected]> Co-authored-by: Isaac Chung <[email protected]> * Update tasks table * misc: update model names to adjust for adding to results repo (#2074) * update model names to adjust for adding to results repo * update model meta script * misc: Add all image classification descriptive stats (#2073) * add most image classification descr stats * revert changes to encoder * add stats --------- Co-authored-by: Roman Solomatin <[email protected]> * Update tasks table * ci: Rerun tests that fail due to networking issues. (#2029) * fix: rerun tests that fail - Networking * update tests to use tmp_path * set versions for dev dependencies * add pytest options to pyproject.toml * add rerun json.decoder.JSONDecodeError * remove JSONDecodeError from pyproject.toml * add huggingface_hub.errors.HfHubHTTPError * add huggingface_hub.errors.LocalEntryNotFoundError https://github.com/embeddings-benchmark/mteb/actions/runs/13298535701/job/37139767443?pr=2044 * FileNotFoundError https://github.com/embeddings-benchmark/mteb/actions/runs/13302915091/job/37147507251?pr=2029 * add doc to pytest rerun --------- Co-authored-by: sam021313 <[email protected]> * fix: generate metadata (#2063) * fix: generate metadata * use logging not print for script * lint * add iso639 to dev pyproject * fix import * add memory_usage_mb * set version for iso639 Co-authored-by: Kenneth Enevoldsen <[email protected]> --------- Co-authored-by: sam021313 <[email protected]> Co-authored-by: Kenneth Enevoldsen <[email protected]> Co-authored-by: Roman Solomatin <[email protected]> * 1.34.15 Automatically generated by python-semantic-release * fix: add missing `e5` training datasets (#2065) add missing training datasets * 1.34.16 Automatically generated by python-semantic-release * fix: Ensure voyage model uses different naming scheme (#2083) * fix: Added make command for running leaderboard locally * fix: Ensure voyage models doesn't re-use the name * 1.34.17 Automatically generated by python-semantic-release * fix: Freeze model/rank columns in leaderboard (#2044) * fix: freeze model/rank columns in leaderboard * freezing zero-shot column * update min gradio version to 5.16.0 in pyproject.toml --------- Co-authored-by: Shikhar Shiromani <[email protected]> * 1.34.18 Automatically generated by python-semantic-release * fix: Fixed previous incorrect specification of splits for CMTEB ( MTEB(cmn, v1) ) (#2086) Fixes #2064 * 1.34.19 Automatically generated by python-semantic-release * Remove duplicated string in docstring of TaskMetadata class (#2087) * Remove duplicated string in docstring of TaskMetadata class * Remove duplicated dataset field * fix: Smarter leaderboard caching with cachetools (#2085) * Added smarter caching to callbacks * Added cachetools as a dependency * Ran linting * Removed debugging print statement * Bumped Gradio version * Dependency fixes * Dependency fixes --------- Co-authored-by: Kenneth Enevoldsen <[email protected]> * fix: Missing fixes for #2086 - change MultilingualSentiment split from test to validation in CMTEB (#2088) * fix: Fixed previous incorrect specification of splits for CMTEB ( MTEB(cmn, v1) ) Fixes #2064 * change MultilingualSentiment split from test to validation in CMTEB * 1.34.20 Automatically generated by python-semantic-release * merge gme models (#2089) * fix: Add back task filtering by modalities (#2080) * add back task filtering by modalities * add unit test * check if task modalities is a subset of model modalities and fix tests * add model_modalities_more_than_task_modalities case * 1.34.21 Automatically generated by python-semantic-release * Added gtr-t5-base/large/xl/xxl metadata to mteb (#2092) * Added GTR Models to codebase * Linted gtr models file. * Added gtr-base/large/xl/xxl to sentence_transformers_models.py * Added memory_usage_mb and training_datasets * Reformatted training dataset names * Reformatted training dataset names * Reformatted training dataset names --------- Co-authored-by: sufen <[email protected]> * misc: Add Any2TextMutipleChoice Descriptive Statistics (#2095) * add Any2TextMutipleChoiceDescriptiveStatistics * run on all tasks * Update tasks table * fix: Updated model annotations for GTE, e5, gritlm, and SFR models (#2101) Reported with references to paper + qoutes. * fix: Update links (#2098) * Fix link * Fix link * 1.34.22 Automatically generated by python-semantic-release * Add model inf-retriever-v1-1.5b (#2106) Add inf-retriever-v1-1.5b model * docs: Fix typos & refine text (#2102) * Update app.py * Fix typos * misc: Run Zeroshot Classification Descriptive Stats (#2105) * add most datasets * add birdsnap and imgnet1k * add scimmir and sun397 * add uck101 zs * Update tasks table * fix: add warning about task category conversion (#2108) add warning about task category conversion * 1.34.23 Automatically generated by python-semantic-release * fix: Add codesage-large-v2 (#2090) * Add codesage-large-v2 * Address comments * Add training dataset * Fix issues * Format code * Remove unnecessary wrapper * 1.34.24 Automatically generated by python-semantic-release * fix: add training data to BGE-m3-custom-fr (#2110) This ensure that is it correctly filtered as non-zero-shot * 1.34.25 Automatically generated by python-semantic-release * fix: Upgrade ruff to be gradio compatible (#2111) * fix: update ruff to be gradio compatible (>=0.9.3) * format * fix: upgrade ruff to latests (same as gradio compatible) * 1.34.26 Automatically generated by python-semantic-release * docs: Follow google docstring format (#2115) Fixes #2113 * Update leaderboard_refresh.yaml (#2121) * fix InstructSentenceTransformer Model name (#2125) fix params * fix voyage (#2127) * fix: update e5 instruct training data (#2129) update e5 training data * 1.34.27 Automatically generated by python-semantic-release * format * Update tasks table * fix: Add 2 new Static Sentence Transformer models (#2112) * Add 2 new Static Sentence Transformer models * Add Tatoeba Co-authored-by: Roman Solomatin <[email protected]> --------- Co-authored-by: Roman Solomatin <[email protected]> * 1.34.28 Automatically generated by python-semantic-release * add is_cross_encoder (#1869) * add is_cross_encoder * Update mteb/model_meta.py Co-authored-by: Kenneth Enevoldsen <[email protected]> * change value --------- Co-authored-by: Kenneth Enevoldsen <[email protected]> * Qodo embed 1 1.5 b (#2137) * feat: Add Qodo-Embed-1-1.5B model metadata * fix: Add Qodo models to overview imports * fix: Add adapted_from field to Qodo model metadata * Update mteb/models/qodo_models.py Co-authored-by: Roman Solomatin <[email protected]> * relint --------- Co-authored-by: Tal Sheffer <[email protected]> Co-authored-by: Roman Solomatin <[email protected]> * misc: merge summary retrieval into bitext mining (#2140) merge summary retrieval into bitext mining * test: fix dataset availability test (#2141) This simplified the test and also make it a lot simpler. It also removed about 100 test cases which where all to the same API call. * fix: Update NVIDIA-Embed training data (#2143) Added a few missing annotations for nvidia-embed * 1.34.29 Automatically generated by python-semantic-release * fix: Add annotations for Voyage exp (#2144) * fix: Update NVIDIA-Embed training data Added a few missing annotations for nvidia-embed * fix update annotationf for voyage exp * 1.34.30 Automatically generated by python-semantic-release * Fix tokens num in cde models (#2148) fix tokens * feat: Add Qodo-Embed-1-7B model metadata and rename existing model (#2146) * feat: Add Qodo-Embed-1-7B model metadata and rename existing model * lint * fix revision * update license name --------- Co-authored-by: Tal Sheffer <[email protected]> * 1.35.0 Automatically generated by python-semantic-release * misc: add Any2AnyRetrievalDescriptiveStatistics (#2139) add Any2AnyRetrievalDescriptiveStatistics * Update tasks table * Added zero-shot percentages and different filtering scheme (#2153) * Added zero-shot percentages and different filtering scheme * Update mteb/model_meta.py Co-authored-by: Roman Solomatin <[email protected]> --------- Co-authored-by: Roman Solomatin <[email protected]> * fix: Incorrect annotations for Mistral-based embedding models (#2157) Fixes #2155 * 1.35.1 Automatically generated by python-semantic-release * Update FaMTEBRetrieval.py (#2171) The URL pointed to the settings page instead of the main repo URL. Now it is fixed. * Update tasks table * fix: Add Training data annotations (#2173) * redo to voyage to only training data * Add training data annotation for Kalm embeddings #2168 * Add correct training data annotations to Stella #2164 * removed fiqa PL as it does not exist * remove ArxivClusteringS2S.v2 as it does not exist * Add training data annotation for GIST embedding #2166 * fix max tokens for kalm models #2162 * remove eli 5 * 1.35.2 Automatically generated by python-semantic-release * feat: Add MIEB and MIEB-lite as benchmarks (#2035) * add mieb and mieb-lite to benchmarks * add CompositionalityEvaluation and DocumentUnderstanding types * add VisionCentric type * add missing comma * split STS17MultilingualVisualSTS and STSBenchmarkMultilingualSTS to eng and non-eng * use aggregate task instead so we can name the subsets * shorten names * fix import * alternative strategy to avoid using get_task * follow other aggregate tasks and skip metadata test * run LB without errors when selecting MIEB(-lite) * add back the capability as taks type * typo * extend description * split into mieb(eng) and mieb(multilingual) * remove unneeded files * remove aggtask additions for test * edit descriptions based on screenshots * shorten * rename to Compositionality and include ImageCoDeT2IMultiChoice * re-tag missing VisionCentric tasks * re-tag rparis and roxford as retrieval and include fixes * re-tag voc2007 as image cls * make lint * correct num task types in descriptions * add one model to models_to_annotate * add mieb reference models * update task types * relabel to multilingual retrieval task type to align with paper * fix reference and bibtex * edit task list to match with final list * add back agg task to reproduce table column in paper * fix filtering and import * update tests * mieb lite add back missing tasks * fix metadata test * multi should have all 4 variants * fix task counts * lite has 10 task types * fix visualSTS-17 lang splits * Aggregate task can now use subsets & eval langs to filter TaskResults * fix test and mark VisualSTS17 as multilingual * fix tests * add agg task running script * add voyage meta * fix citations * capitalize * add coarse/fine labels --------- Co-authored-by: gowitheflow-1998 <[email protected]> * Update tasks table * 1.36.0 Automatically generated by python-semantic-release * fix: update training datasets and revision for jina models (#2179) * feat: update training datasets and revision for jina models * feat: update training datasets and revision for jina models * fix: Add more training data annotations (#2178) * redo to voyage to only training data * Add training data annotation for Kalm embeddings #2168 * Add correct training data annotations to Stella #2164 * removed fiqa PL as it does not exist * remove ArxivClusteringS2S.v2 as it does not exist * Add training data annotation for GIST embedding #2166 * fix max tokens for kalm models #2162 * remove eli 5 * fix: add training data for Bilingual Embeddings fixes #2167 * 1.36.1 Automatically generated by python-semantic-release * Added training data annotation for e5-base-4k (#2186) * fix: Added training data annotations to MXBAI (#2185) * fix: Update MTEB(Scandinavian) to use new DanFEVER (#2180) This also resolves the missing data in the leaderboard. Fixes #2172 * fix: Added training data annotation for MMLW models (#2188) * Added training data annotation for MMLW models * Added GIST annotations Kenneth missed * Added Stella en 400m training data' * 1.36.2 Automatically generated by python-semantic-release * fix: Added training data for sentence-croissant (#2189) * 1.36.3 Automatically generated by python-semantic-release * fix: update ru models annotation (#2181) * 1.36.4 Automatically generated by python-semantic-release * fix: Alphabetical ordering of tasks in dropdowns (#2191) * 1.36.5 Automatically generated by python-semantic-release * misc: Speed up qrel creation in any2anyretrieval (#2196) * use numpy vectorized operations instead of row-by-row * scores are int * use 'mteb.MTEB' instead of 'MTEB' for custom model (#2199) * add base models for e5 (#2183) * add similar datasets (#2205) * add similar datasets * add nano * update is filled * Update mteb/abstasks/TaskMetadata.py Co-authored-by: Kenneth Enevoldsen <[email protected]> --------- Co-authored-by: Kenneth Enevoldsen <[email protected]> * add labse annotation (#2182) * add labse annotation * Update mteb/models/sentence_transformers_models.py Co-authored-by: Kenneth Enevoldsen <[email protected]> --------- Co-authored-by: Kenneth Enevoldsen <[email protected]> * fix: Fixed leaderboard crash (#2221) * Fixed leaderboard crash * Fixed language selection error * Ran linting * 1.36.6 Automatically generated by python-semantic-release * fix: More training data annotations (#2220) * Added training data annotation for bge-gemma * Added missing annotations for Voyage models * Added training data for sts-multilingual-mpnet * Added all mteb datasets to STS-multilingual training data * 1.36.7 Automatically generated by python-semantic-release * Add LLM2CLIP (OpenAI variants) (#2222) * model loading and get_text_embeddings * add image_emb, fused_emb, and calc probs methods * add b16 model * add llm2clip_openai_l_14_224 (not working yet) * got llm2clip_openai_l_14_224 working * make lint * add training sets and allow py files * Change `dataset on HF` test to use official api (#2213) * refactor dataset checking * increase timeout * increase timeout * remove timeout * Descriptive stats functions for Any2AnyMC and ImageTextPC (#2197) * Add Any2AnyMC descriptive stats * Add descriptive stats function for ImageTextPC * add descriptive stats examples * linter * update multi choice descriptive stats * Update tasks table * fix: Add training data annotations to uderver-bloom models (#2210) * fix: Add training data annotations to uderver-bloom models fixes #2193 * fix: add mixedbread --------- Co-authored-by: Márton Kardos <[email protected]> * 1.36.8 Automatically generated by python-semantic-release * Add comment to `voyage-3-m-exp` model (#2229) * remove model size from voyage-3-m-exp model * Update mteb/models/voyage_models.py * Update mteb/models/voyage_models.py * docs: Update description of EURLex (#2231) * Automatically add similar tasks to training_tasks (#2228) * refactor dataset checking * increase timeout * increase timeout * remove timeout * start * automatically find datasets * update comment * fix aggregate task metadata * fixes * lint * rename * update fetch check * Remove overlapping legends from radar chart (#2195) * Remove overlapping legends from radar chart * ensure graph is not blocked * Overlapping legend issue of Radar Chart * misc: Run Any2AnyRetrieval descriptive stats (#2223) * run a few datasets * add a few more * run more tasks * add more datasets * remove pdb * remove newline * add more datasets * Update tasks table * misc: Add rest of the vision centric and compositionality descriptive stats (#2267) add the rest * Update tasks table * Fix `calculate_memory_usage_mb` in adding_a_model.md (#2271) * Add Arabic-Triplet-Matryoshka-V2 model metadata to MTEB (#2270) * Add Arabic-Triplet-Matryoshka-V2 model metadata to MTEB * Update memory_usage_mb with correct calculated value * Update mteb/models/Arabic_Triplet_Matryoshka_V2.py Co-authored-by: Roman Solomatin <[email protected]> * Update mteb/models/Arabic_Triplet_Matryoshka_V2.py Co-authored-by: Roman Solomatin <[email protected]> * remove comments * added correct memory usage * Update mteb/models/Arabic_Triplet_Matryoshka_V2.py Co-authored-by: Roman Solomatin <[email protected]> * Apply linter fixes with ruff * Update mteb/models/Arabic_Triplet_Matryoshka_V2.py Co-authored-by: Roman Solomatin <[email protected]> * Update mteb/models/Arabic_Triplet_Matryoshka_V2.py Co-authored-by: Roman Solomatin <[email protected]> * Add Arabic_Triplet_Matryoshka_V2 to overview.py * Rename model file to ara_models.py and update imports --------- Co-authored-by: Roman Solomatin <[email protected]> * fix: Add WebFAQ Retrieval dataset (#2236) * Add WebFAQ Retrieval dataset Signed-off-by: Michael Dinzinger <[email protected]> * Small change WebFAQRetrieval.py Signed-off-by: Michael Dinzinger <[email protected]> * Add remaining languages to WebFAQ Retrieval task Signed-off-by: Michael Dinzinger <[email protected]> * Add descriptive stats Signed-off-by: Michael Dinzinger <[email protected]> --------- Signed-off-by: Michael Dinzinger <[email protected]> * Update tasks table * 1.36.9 Automatically generated by python-semantic-release * fix: Formatting issue in Performance Plot (#2237) * Formatting issue in Performance Plot * make lint * added function for better code readability * 1.36.10 Automatically generated by python-semantic-release * ci: run test_dataset_on_hf separately (#2201) * dont run test_dataset_on_hf in every pr * lint * Update call pytest test_datasets Co-authored-by: Kenneth Enevoldsen <[email protected]> * Update tests/test_tasks/test_all_abstasks.py Co-authored-by: Kenneth Enevoldsen <[email protected]> * not datasets for test * run dataset loading test for push or pull_request * apply feedback --------- Co-authored-by: sam021313 <[email protected]> Co-authored-by: Kenneth Enevoldsen <[email protected]> * add gemini-embedding-exp-03-07 (#2279) * add gemini-embedding-exp-03-07 * remove space for lint * lint fix * update link (#2281) * fix: Run remaining MIEB desc stats (#2288) * run Vidore * GLDv2 * run the rest --------- Co-authored-by: Isaac Chung <[email protected]> * Update tasks table * 1.36.11 Automatically generated by python-semantic-release * fix: Added Filter Modality (#2262) * Added Filter Modality * resolve suggestions * make lint * make sure test pass * make lint * added exclusive_modality_filter and unit tests * Integrate tests on overview.py * Update tests/test_overview.py Co-authored-by: Roman Solomatin <[email protected]> * added task related to image modality * Update mteb/abstasks/AbsTask.py Co-authored-by: Kenneth Enevoldsen <[email protected]> * Update mteb/abstasks/AbsTask.py Co-authored-by: Kenneth Enevoldsen <[email protected]> * update overview..py * make lint * update documentation --------- Co-authored-by: Roman Solomatin <[email protected]> Co-authored-by: Kenneth Enevoldsen <[email protected]> * 1.36.12 Automatically generated by python-semantic-release * fix: Add `ModelMeta` license & custom validations (#2293) * license validation * move licenses * update imports --------- Co-authored-by: Isaac Chung <[email protected]> * 1.36.13 Automatically generated by python-semantic-release * ci: Add pre-commit hook (#2194) * make dev life nicer - pre-commit hooks * add pre-commit to install * update precommit * update ruff pre-commit * lint * lint --------- Co-authored-by: sam021313 <[email protected]> * Update tasks table * fix: bug in voyage implementation (#2304) * fix: Fix bug in voyage implementation "passage" is not a valid input for the voyage API. Remapped to "document". * Update mteb/models/voyage_models.py Co-authored-by: Roman Solomatin <[email protected]> --------- Co-authored-by: Roman Solomatin <[email protected]> * 1.36.14 Automatically generated by python-semantic-release * fix: Update voyage name to include Org. (#2322) * 1.36.15 Automatically generated by python-semantic-release * Added VDR Model (#2290) * Added VDR Model * change custom wrapper to SentenceTransformer Wrapper * remove kwargs and add TODO for Image Modality * Update mteb/models/vdr_models.py Co-authored-by: Roman Solomatin <[email protected]> --------- Co-authored-by: Roman Solomatin <[email protected]> * fix: Resolve conflicting dependencies (#2323) These errors where discovered when trying to install the package using `uv`. We have a problem with salesforce-lavis, which is not compatible with the current set of dependencies. * 1.36.16 Automatically generated by python-semantic-release * fix: remove SyntaxWarnings in py312 (#2325) * fix: Resolve conflicting dependencies These errors where discovered when trying to install the package using `uv`. We have a problem with salesforce-lavis, which is not compatible with the current set of dependencies. * fix: Remove syntax warnings occuring in python 3.12 ``` Python 3.12.0 (main, Oct 2 2023, 20:56:14) [Clang 16.0.3 ] on darwin Type "help", "copyright", "credits" or "license" for more information. >>> import mteb # no syntax warnings >>> ``` * 1.36.17 Automatically generated by python-semantic-release * fix: add annotation models for stella zh (#2277) * fix: add annotation models for stella zh Additionally fixed a few annotation errors * format * Update mteb/models/stella_models.py Co-authored-by: Isaac Chung <[email protected]> --------- Co-authored-by: Isaac Chung <[email protected]> * 1.36.18 Automatically generated by python-semantic-release * fix: Add ModelMeta rubert-mini-frida, BERTA (#2330) * Add rubert-mini-frida model meta * Add BERTA model meta * docs: fix typos * 1.36.19 Automatically generated by python-semantic-release * fix: Add WebFAQ bitext mining tasks (#2326) * Add WebFAQ bitext mining tasks Signed-off-by: Michael Dinzinger <[email protected]> * Lower number of language pairs in WebFAQBitextMining Signed-off-by: Michael Dinzinger <[email protected]> --------- Signed-off-by: Michael Dinzinger <[email protected]> * Update tasks table * 1.36.20 Automatically generated by python-semantic-release * make lint * fix validation for license * fix remaining validation errors --------- Signed-off-by: Michael Dinzinger <[email protected]> Co-authored-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com> Co-authored-by: Kenneth Enevoldsen <[email protected]> Co-authored-by: github-actions <[email protected]> Co-authored-by: Roman Solomatin <[email protected]> Co-authored-by: Mina Parham <[email protected]> Co-authored-by: Mina Parham <[email protected]> Co-authored-by: Mehrzad Shahin-Moghadam <[email protected]> Co-authored-by: Roman Solomatin <[email protected]> Co-authored-by: Sam <[email protected]> Co-authored-by: sam021313 <[email protected]> Co-authored-by: Shikhar Shiromani <[email protected]> Co-authored-by: Shikhar Shiromani <[email protected]> Co-authored-by: Ruslan Bel'kov <[email protected]> Co-authored-by: Márton Kardos <[email protected]> Co-authored-by: sufen-f <[email protected]> Co-authored-by: sufen <[email protected]> Co-authored-by: Niklas Muennighoff <[email protected]> Co-authored-by: Samuel Yang <[email protected]> Co-authored-by: Aradhye Agarwal <[email protected]> Co-authored-by: Tom Aarsen <[email protected]> Co-authored-by: talshef <[email protected]> Co-authored-by: Tal Sheffer <[email protected]> Co-authored-by: garciasces <[email protected]> Co-authored-by: gowitheflow-1998 <[email protected]> Co-authored-by: Wang Bo <[email protected]> Co-authored-by: Munot Ayush Sunil <[email protected]> Co-authored-by: Yaya Sy <[email protected]> Co-authored-by: Imene Kerboua <[email protected]> Co-authored-by: Eng. Omar Najar <[email protected]> Co-authored-by: Michael Dinzinger <[email protected]> Co-authored-by: Jinhyuk Lee <[email protected]> Co-authored-by: Isaac Chung <[email protected]> Co-authored-by: sergeyz-zh <[email protected]>
isaac-chung
added a commit
that referenced
this pull request
Apr 1, 2025
* misc: Add image classification descriptive stats implementation (#2045) * add ImageClassificationDescriptiveStatistics * add MNIST descriptive stats * use tuples instead * add label count and update docstrings * update MNIST example * Update tasks table * fix: Add column descriptions to leaderboard (#2039) * fix: Add column descriptions to leaderboard * removed existing overlap * fix: Add BRIGHT (long) and fix bug in TaskResult.filter_and_validate() (#2041) * fix: Add BRIGHT Long Fixes #1978 * fix: Add BRIGHT(long) * fix bug in task results * updated bright * updated tests for TaskResults * 1.34.12 Automatically generated by python-semantic-release * misc: Add image clustering descriptive stats implementation (#2057) * add image clustering descirptive stats and run * finish off last one * remove script * fix: Update embed_dim for jina models (#2058) see embeddings-benchmark/results#117 * Update tasks table * 1.34.13 Automatically generated by python-semantic-release * Add giga embeddings (#1741) * add gigaembeddings * use jasper * fix name * create sentence_transformer instruct wrapper * apply instruction template * fix jasper * update meta * misc: Add ZS and multilabel image classification descriptive stats implementation (#2059) * add image clustering descirptive stats and run * finish off last one * remove script * add ImageMultilabelClassificationDescriptiveStatistics * add VOC2007 * add zeroshot and mnist example * Update tasks table * Rename MIEB task classes with duplicated names (#2061) fix class names * misc: Add VisualSTS descriptive stats (#2062) * add visualsts stats * add last dataset * Update tasks table * fix: Added gte models (#1539) * fix: Added gte models * fix: Add mixbai models (#1540) for #1515 * fix: Add climate fever v2 (#1873) * Updated ClimateFEVER dataset with new version * Adds Fill in the empty metadata. * Updates the date tuple * Update class name Co-authored-by: Kenneth Enevoldsen <[email protected]> * Update domains Co-authored-by: Kenneth Enevoldsen <[email protected]> * Update task_subtypes * Update annotations_creators for the first version * Update date Co-authored-by: Kenneth Enevoldsen <[email protected]> * Update task subtypes * Update path * Update description --------- Co-authored-by: Kenneth Enevoldsen <[email protected]> Co-authored-by: Mina Parham <[email protected]> * Update tasks table * fix: Updating paper scripts (#1958) * change reference revisions to align with paper * Update author list * Added code for main results table * updated minor changes * added external as a "no_revision_available" case * revert unintended changes * format * 1.34.14 Automatically generated by python-semantic-release * Add datasets for a benchmark newly introduced for "Engineering" domain (#1911) * adding clustering tasks (built-bench-clustering S2S & P2P) * updated built-bench-clustering tasks * Updated BuiltBenchClustering tasks * Added "Engineering" as new domain to TaskMetadata.py * Updated tasks table in docs * Updated task metadata for BuiltBenchClustering S2S and P2P * updated metadata for clustering tasks * Add/update BuiltBench tasks - Add BuiltBenchRetrieval task - Add BuiltBenchReranking task - Update metadata for BuiltBenchClusterinP2P - Update metadata for BuiltBenchClusterinS2S * update BuiltBench benchmark * Update mteb/benchmarks/benchmarks.py Co-authored-by: Roman Solomatin <[email protected]> * Update mteb/tasks/Clustering/eng/BuiltBenchClusteringS2S.py Co-authored-by: Roman Solomatin <[email protected]> * Update mteb/tasks/Clustering/eng/BuiltBenchClusteringP2P.py Co-authored-by: Roman Solomatin <[email protected]> * Update mteb/benchmarks/benchmarks.py Co-authored-by: Isaac Chung <[email protected]> * Fix formatting via ruff --------- Co-authored-by: Roman Solomatin <[email protected]> Co-authored-by: Isaac Chung <[email protected]> * Update tasks table * misc: update model names to adjust for adding to results repo (#2074) * update model names to adjust for adding to results repo * update model meta script * misc: Add all image classification descriptive stats (#2073) * add most image classification descr stats * revert changes to encoder * add stats --------- Co-authored-by: Roman Solomatin <[email protected]> * Update tasks table * ci: Rerun tests that fail due to networking issues. (#2029) * fix: rerun tests that fail - Networking * update tests to use tmp_path * set versions for dev dependencies * add pytest options to pyproject.toml * add rerun json.decoder.JSONDecodeError * remove JSONDecodeError from pyproject.toml * add huggingface_hub.errors.HfHubHTTPError * add huggingface_hub.errors.LocalEntryNotFoundError https://github.com/embeddings-benchmark/mteb/actions/runs/13298535701/job/37139767443?pr=2044 * FileNotFoundError https://github.com/embeddings-benchmark/mteb/actions/runs/13302915091/job/37147507251?pr=2029 * add doc to pytest rerun --------- Co-authored-by: sam021313 <[email protected]> * fix: generate metadata (#2063) * fix: generate metadata * use logging not print for script * lint * add iso639 to dev pyproject * fix import * add memory_usage_mb * set version for iso639 Co-authored-by: Kenneth Enevoldsen <[email protected]> --------- Co-authored-by: sam021313 <[email protected]> Co-authored-by: Kenneth Enevoldsen <[email protected]> Co-authored-by: Roman Solomatin <[email protected]> * 1.34.15 Automatically generated by python-semantic-release * fix: add missing `e5` training datasets (#2065) add missing training datasets * 1.34.16 Automatically generated by python-semantic-release * fix: Ensure voyage model uses different naming scheme (#2083) * fix: Added make command for running leaderboard locally * fix: Ensure voyage models doesn't re-use the name * 1.34.17 Automatically generated by python-semantic-release * fix: Freeze model/rank columns in leaderboard (#2044) * fix: freeze model/rank columns in leaderboard * freezing zero-shot column * update min gradio version to 5.16.0 in pyproject.toml --------- Co-authored-by: Shikhar Shiromani <[email protected]> * 1.34.18 Automatically generated by python-semantic-release * fix: Fixed previous incorrect specification of splits for CMTEB ( MTEB(cmn, v1) ) (#2086) Fixes #2064 * 1.34.19 Automatically generated by python-semantic-release * Remove duplicated string in docstring of TaskMetadata class (#2087) * Remove duplicated string in docstring of TaskMetadata class * Remove duplicated dataset field * fix: Smarter leaderboard caching with cachetools (#2085) * Added smarter caching to callbacks * Added cachetools as a dependency * Ran linting * Removed debugging print statement * Bumped Gradio version * Dependency fixes * Dependency fixes --------- Co-authored-by: Kenneth Enevoldsen <[email protected]> * fix: Missing fixes for #2086 - change MultilingualSentiment split from test to validation in CMTEB (#2088) * fix: Fixed previous incorrect specification of splits for CMTEB ( MTEB(cmn, v1) ) Fixes #2064 * change MultilingualSentiment split from test to validation in CMTEB * 1.34.20 Automatically generated by python-semantic-release * merge gme models (#2089) * fix: Add back task filtering by modalities (#2080) * add back task filtering by modalities * add unit test * check if task modalities is a subset of model modalities and fix tests * add model_modalities_more_than_task_modalities case * 1.34.21 Automatically generated by python-semantic-release * Added gtr-t5-base/large/xl/xxl metadata to mteb (#2092) * Added GTR Models to codebase * Linted gtr models file. * Added gtr-base/large/xl/xxl to sentence_transformers_models.py * Added memory_usage_mb and training_datasets * Reformatted training dataset names * Reformatted training dataset names * Reformatted training dataset names --------- Co-authored-by: sufen <[email protected]> * misc: Add Any2TextMutipleChoice Descriptive Statistics (#2095) * add Any2TextMutipleChoiceDescriptiveStatistics * run on all tasks * Update tasks table * fix: Updated model annotations for GTE, e5, gritlm, and SFR models (#2101) Reported with references to paper + qoutes. * fix: Update links (#2098) * Fix link * Fix link * 1.34.22 Automatically generated by python-semantic-release * Add model inf-retriever-v1-1.5b (#2106) Add inf-retriever-v1-1.5b model * docs: Fix typos & refine text (#2102) * Update app.py * Fix typos * misc: Run Zeroshot Classification Descriptive Stats (#2105) * add most datasets * add birdsnap and imgnet1k * add scimmir and sun397 * add uck101 zs * Update tasks table * fix: add warning about task category conversion (#2108) add warning about task category conversion * 1.34.23 Automatically generated by python-semantic-release * fix: Add codesage-large-v2 (#2090) * Add codesage-large-v2 * Address comments * Add training dataset * Fix issues * Format code * Remove unnecessary wrapper * 1.34.24 Automatically generated by python-semantic-release * fix: add training data to BGE-m3-custom-fr (#2110) This ensure that is it correctly filtered as non-zero-shot * 1.34.25 Automatically generated by python-semantic-release * fix: Upgrade ruff to be gradio compatible (#2111) * fix: update ruff to be gradio compatible (>=0.9.3) * format * fix: upgrade ruff to latests (same as gradio compatible) * 1.34.26 Automatically generated by python-semantic-release * docs: Follow google docstring format (#2115) Fixes #2113 * Update leaderboard_refresh.yaml (#2121) * fix InstructSentenceTransformer Model name (#2125) fix params * fix voyage (#2127) * fix: update e5 instruct training data (#2129) update e5 training data * 1.34.27 Automatically generated by python-semantic-release * format * Update tasks table * fix: Add 2 new Static Sentence Transformer models (#2112) * Add 2 new Static Sentence Transformer models * Add Tatoeba Co-authored-by: Roman Solomatin <[email protected]> --------- Co-authored-by: Roman Solomatin <[email protected]> * 1.34.28 Automatically generated by python-semantic-release * add is_cross_encoder (#1869) * add is_cross_encoder * Update mteb/model_meta.py Co-authored-by: Kenneth Enevoldsen <[email protected]> * change value --------- Co-authored-by: Kenneth Enevoldsen <[email protected]> * Qodo embed 1 1.5 b (#2137) * feat: Add Qodo-Embed-1-1.5B model metadata * fix: Add Qodo models to overview imports * fix: Add adapted_from field to Qodo model metadata * Update mteb/models/qodo_models.py Co-authored-by: Roman Solomatin <[email protected]> * relint --------- Co-authored-by: Tal Sheffer <[email protected]> Co-authored-by: Roman Solomatin <[email protected]> * misc: merge summary retrieval into bitext mining (#2140) merge summary retrieval into bitext mining * test: fix dataset availability test (#2141) This simplified the test and also make it a lot simpler. It also removed about 100 test cases which where all to the same API call. * fix: Update NVIDIA-Embed training data (#2143) Added a few missing annotations for nvidia-embed * 1.34.29 Automatically generated by python-semantic-release * fix: Add annotations for Voyage exp (#2144) * fix: Update NVIDIA-Embed training data Added a few missing annotations for nvidia-embed * fix update annotationf for voyage exp * 1.34.30 Automatically generated by python-semantic-release * Fix tokens num in cde models (#2148) fix tokens * feat: Add Qodo-Embed-1-7B model metadata and rename existing model (#2146) * feat: Add Qodo-Embed-1-7B model metadata and rename existing model * lint * fix revision * update license name --------- Co-authored-by: Tal Sheffer <[email protected]> * 1.35.0 Automatically generated by python-semantic-release * misc: add Any2AnyRetrievalDescriptiveStatistics (#2139) add Any2AnyRetrievalDescriptiveStatistics * Update tasks table * Added zero-shot percentages and different filtering scheme (#2153) * Added zero-shot percentages and different filtering scheme * Update mteb/model_meta.py Co-authored-by: Roman Solomatin <[email protected]> --------- Co-authored-by: Roman Solomatin <[email protected]> * fix: Incorrect annotations for Mistral-based embedding models (#2157) Fixes #2155 * 1.35.1 Automatically generated by python-semantic-release * Update FaMTEBRetrieval.py (#2171) The URL pointed to the settings page instead of the main repo URL. Now it is fixed. * Update tasks table * fix: Add Training data annotations (#2173) * redo to voyage to only training data * Add training data annotation for Kalm embeddings #2168 * Add correct training data annotations to Stella #2164 * removed fiqa PL as it does not exist * remove ArxivClusteringS2S.v2 as it does not exist * Add training data annotation for GIST embedding #2166 * fix max tokens for kalm models #2162 * remove eli 5 * 1.35.2 Automatically generated by python-semantic-release * feat: Add MIEB and MIEB-lite as benchmarks (#2035) * add mieb and mieb-lite to benchmarks * add CompositionalityEvaluation and DocumentUnderstanding types * add VisionCentric type * add missing comma * split STS17MultilingualVisualSTS and STSBenchmarkMultilingualSTS to eng and non-eng * use aggregate task instead so we can name the subsets * shorten names * fix import * alternative strategy to avoid using get_task * follow other aggregate tasks and skip metadata test * run LB without errors when selecting MIEB(-lite) * add back the capability as taks type * typo * extend description * split into mieb(eng) and mieb(multilingual) * remove unneeded files * remove aggtask additions for test * edit descriptions based on screenshots * shorten * rename to Compositionality and include ImageCoDeT2IMultiChoice * re-tag missing VisionCentric tasks * re-tag rparis and roxford as retrieval and include fixes * re-tag voc2007 as image cls * make lint * correct num task types in descriptions * add one model to models_to_annotate * add mieb reference models * update task types * relabel to multilingual retrieval task type to align with paper * fix reference and bibtex * edit task list to match with final list * add back agg task to reproduce table column in paper * fix filtering and import * update tests * mieb lite add back missing tasks * fix metadata test * multi should have all 4 variants * fix task counts * lite has 10 task types * fix visualSTS-17 lang splits * Aggregate task can now use subsets & eval langs to filter TaskResults * fix test and mark VisualSTS17 as multilingual * fix tests * add agg task running script * add voyage meta * fix citations * capitalize * add coarse/fine labels --------- Co-authored-by: gowitheflow-1998 <[email protected]> * Update tasks table * 1.36.0 Automatically generated by python-semantic-release * fix: update training datasets and revision for jina models (#2179) * feat: update training datasets and revision for jina models * feat: update training datasets and revision for jina models * fix: Add more training data annotations (#2178) * redo to voyage to only training data * Add training data annotation for Kalm embeddings #2168 * Add correct training data annotations to Stella #2164 * removed fiqa PL as it does not exist * remove ArxivClusteringS2S.v2 as it does not exist * Add training data annotation for GIST embedding #2166 * fix max tokens for kalm models #2162 * remove eli 5 * fix: add training data for Bilingual Embeddings fixes #2167 * 1.36.1 Automatically generated by python-semantic-release * Added training data annotation for e5-base-4k (#2186) * fix: Added training data annotations to MXBAI (#2185) * fix: Update MTEB(Scandinavian) to use new DanFEVER (#2180) This also resolves the missing data in the leaderboard. Fixes #2172 * fix: Added training data annotation for MMLW models (#2188) * Added training data annotation for MMLW models * Added GIST annotations Kenneth missed * Added Stella en 400m training data' * 1.36.2 Automatically generated by python-semantic-release * fix: Added training data for sentence-croissant (#2189) * 1.36.3 Automatically generated by python-semantic-release * fix: update ru models annotation (#2181) * 1.36.4 Automatically generated by python-semantic-release * fix: Alphabetical ordering of tasks in dropdowns (#2191) * 1.36.5 Automatically generated by python-semantic-release * misc: Speed up qrel creation in any2anyretrieval (#2196) * use numpy vectorized operations instead of row-by-row * scores are int * use 'mteb.MTEB' instead of 'MTEB' for custom model (#2199) * add base models for e5 (#2183) * add similar datasets (#2205) * add similar datasets * add nano * update is filled * Update mteb/abstasks/TaskMetadata.py Co-authored-by: Kenneth Enevoldsen <[email protected]> --------- Co-authored-by: Kenneth Enevoldsen <[email protected]> * add labse annotation (#2182) * add labse annotation * Update mteb/models/sentence_transformers_models.py Co-authored-by: Kenneth Enevoldsen <[email protected]> --------- Co-authored-by: Kenneth Enevoldsen <[email protected]> * fix: Fixed leaderboard crash (#2221) * Fixed leaderboard crash * Fixed language selection error * Ran linting * 1.36.6 Automatically generated by python-semantic-release * fix: More training data annotations (#2220) * Added training data annotation for bge-gemma * Added missing annotations for Voyage models * Added training data for sts-multilingual-mpnet * Added all mteb datasets to STS-multilingual training data * 1.36.7 Automatically generated by python-semantic-release * Add LLM2CLIP (OpenAI variants) (#2222) * model loading and get_text_embeddings * add image_emb, fused_emb, and calc probs methods * add b16 model * add llm2clip_openai_l_14_224 (not working yet) * got llm2clip_openai_l_14_224 working * make lint * add training sets and allow py files * Change `dataset on HF` test to use official api (#2213) * refactor dataset checking * increase timeout * increase timeout * remove timeout * Descriptive stats functions for Any2AnyMC and ImageTextPC (#2197) * Add Any2AnyMC descriptive stats * Add descriptive stats function for ImageTextPC * add descriptive stats examples * linter * update multi choice descriptive stats * Update tasks table * fix: Add training data annotations to uderver-bloom models (#2210) * fix: Add training data annotations to uderver-bloom models fixes #2193 * fix: add mixedbread --------- Co-authored-by: Márton Kardos <[email protected]> * 1.36.8 Automatically generated by python-semantic-release * Add comment to `voyage-3-m-exp` model (#2229) * remove model size from voyage-3-m-exp model * Update mteb/models/voyage_models.py * Update mteb/models/voyage_models.py * docs: Update description of EURLex (#2231) * Automatically add similar tasks to training_tasks (#2228) * refactor dataset checking * increase timeout * increase timeout * remove timeout * start * automatically find datasets * update comment * fix aggregate task metadata * fixes * lint * rename * update fetch check * Remove overlapping legends from radar chart (#2195) * Remove overlapping legends from radar chart * ensure graph is not blocked * Overlapping legend issue of Radar Chart * misc: Run Any2AnyRetrieval descriptive stats (#2223) * run a few datasets * add a few more * run more tasks * add more datasets * remove pdb * remove newline * add more datasets * Update tasks table * misc: Add rest of the vision centric and compositionality descriptive stats (#2267) add the rest * Update tasks table * Fix `calculate_memory_usage_mb` in adding_a_model.md (#2271) * Add Arabic-Triplet-Matryoshka-V2 model metadata to MTEB (#2270) * Add Arabic-Triplet-Matryoshka-V2 model metadata to MTEB * Update memory_usage_mb with correct calculated value * Update mteb/models/Arabic_Triplet_Matryoshka_V2.py Co-authored-by: Roman Solomatin <[email protected]> * Update mteb/models/Arabic_Triplet_Matryoshka_V2.py Co-authored-by: Roman Solomatin <[email protected]> * remove comments * added correct memory usage * Update mteb/models/Arabic_Triplet_Matryoshka_V2.py Co-authored-by: Roman Solomatin <[email protected]> * Apply linter fixes with ruff * Update mteb/models/Arabic_Triplet_Matryoshka_V2.py Co-authored-by: Roman Solomatin <[email protected]> * Update mteb/models/Arabic_Triplet_Matryoshka_V2.py Co-authored-by: Roman Solomatin <[email protected]> * Add Arabic_Triplet_Matryoshka_V2 to overview.py * Rename model file to ara_models.py and update imports --------- Co-authored-by: Roman Solomatin <[email protected]> * fix: Add WebFAQ Retrieval dataset (#2236) * Add WebFAQ Retrieval dataset Signed-off-by: Michael Dinzinger <[email protected]> * Small change WebFAQRetrieval.py Signed-off-by: Michael Dinzinger <[email protected]> * Add remaining languages to WebFAQ Retrieval task Signed-off-by: Michael Dinzinger <[email protected]> * Add descriptive stats Signed-off-by: Michael Dinzinger <[email protected]> --------- Signed-off-by: Michael Dinzinger <[email protected]> * Update tasks table * 1.36.9 Automatically generated by python-semantic-release * fix: Formatting issue in Performance Plot (#2237) * Formatting issue in Performance Plot * make lint * added function for better code readability * 1.36.10 Automatically generated by python-semantic-release * ci: run test_dataset_on_hf separately (#2201) * dont run test_dataset_on_hf in every pr * lint * Update call pytest test_datasets Co-authored-by: Kenneth Enevoldsen <[email protected]> * Update tests/test_tasks/test_all_abstasks.py Co-authored-by: Kenneth Enevoldsen <[email protected]> * not datasets for test * run dataset loading test for push or pull_request * apply feedback --------- Co-authored-by: sam021313 <[email protected]> Co-authored-by: Kenneth Enevoldsen <[email protected]> * add gemini-embedding-exp-03-07 (#2279) * add gemini-embedding-exp-03-07 * remove space for lint * lint fix * update link (#2281) * fix: Run remaining MIEB desc stats (#2288) * run Vidore * GLDv2 * run the rest --------- Co-authored-by: Isaac Chung <[email protected]> * Update tasks table * 1.36.11 Automatically generated by python-semantic-release * fix: Added Filter Modality (#2262) * Added Filter Modality * resolve suggestions * make lint * make sure test pass * make lint * added exclusive_modality_filter and unit tests * Integrate tests on overview.py * Update tests/test_overview.py Co-authored-by: Roman Solomatin <[email protected]> * added task related to image modality * Update mteb/abstasks/AbsTask.py Co-authored-by: Kenneth Enevoldsen <[email protected]> * Update mteb/abstasks/AbsTask.py Co-authored-by: Kenneth Enevoldsen <[email protected]> * update overview..py * make lint * update documentation --------- Co-authored-by: Roman Solomatin <[email protected]> Co-authored-by: Kenneth Enevoldsen <[email protected]> * 1.36.12 Automatically generated by python-semantic-release * fix: Add `ModelMeta` license & custom validations (#2293) * license validation * move licenses * update imports --------- Co-authored-by: Isaac Chung <[email protected]> * 1.36.13 Automatically generated by python-semantic-release * ci: Add pre-commit hook (#2194) * make dev life nicer - pre-commit hooks * add pre-commit to install * update precommit * update ruff pre-commit * lint * lint --------- Co-authored-by: sam021313 <[email protected]> * Update tasks table * fix: bug in voyage implementation (#2304) * fix: Fix bug in voyage implementation "passage" is not a valid input for the voyage API. Remapped to "document". * Update mteb/models/voyage_models.py Co-authored-by: Roman Solomatin <[email protected]> --------- Co-authored-by: Roman Solomatin <[email protected]> * 1.36.14 Automatically generated by python-semantic-release * fix: Update voyage name to include Org. (#2322) * 1.36.15 Automatically generated by python-semantic-release * Added VDR Model (#2290) * Added VDR Model * change custom wrapper to SentenceTransformer Wrapper * remove kwargs and add TODO for Image Modality * Update mteb/models/vdr_models.py Co-authored-by: Roman Solomatin <[email protected]> --------- Co-authored-by: Roman Solomatin <[email protected]> * fix: Resolve conflicting dependencies (#2323) These errors where discovered when trying to install the package using `uv`. We have a problem with salesforce-lavis, which is not compatible with the current set of dependencies. * 1.36.16 Automatically generated by python-semantic-release * fix: remove SyntaxWarnings in py312 (#2325) * fix: Resolve conflicting dependencies These errors where discovered when trying to install the package using `uv`. We have a problem with salesforce-lavis, which is not compatible with the current set of dependencies. * fix: Remove syntax warnings occuring in python 3.12 ``` Python 3.12.0 (main, Oct 2 2023, 20:56:14) [Clang 16.0.3 ] on darwin Type "help", "copyright", "credits" or "license" for more information. >>> import mteb # no syntax warnings >>> ``` * 1.36.17 Automatically generated by python-semantic-release * fix: add annotation models for stella zh (#2277) * fix: add annotation models for stella zh Additionally fixed a few annotation errors * format * Update mteb/models/stella_models.py Co-authored-by: Isaac Chung <[email protected]> --------- Co-authored-by: Isaac Chung <[email protected]> * 1.36.18 Automatically generated by python-semantic-release * fix: Add ModelMeta rubert-mini-frida, BERTA (#2330) * Add rubert-mini-frida model meta * Add BERTA model meta * docs: fix typos * 1.36.19 Automatically generated by python-semantic-release * fix: Add WebFAQ bitext mining tasks (#2326) * Add WebFAQ bitext mining tasks Signed-off-by: Michael Dinzinger <[email protected]> * Lower number of language pairs in WebFAQBitextMining Signed-off-by: Michael Dinzinger <[email protected]> --------- Signed-off-by: Michael Dinzinger <[email protected]> * Update tasks table * 1.36.20 Automatically generated by python-semantic-release * fix: Add `trust_remote_code` to MIRACLRetrieval * fix: Add `trust_remote_code` to MIRACLRetrieval (#2344) * 1.36.21 Automatically generated by python-semantic-release * fix: Correctly pass trust remote code to Miracl * fix: Ensure MIRACL pass trust_remote_code (#2346) * fix: Add `trust_remote_code` to MIRACLRetrieval * fix: Correctly pass trust remote code to Miracl * fix * 1.36.22 Automatically generated by python-semantic-release * add-Data Korean Clustering dataset (KLUE-modified) (#2283) * add PatentFnBClustering.py * do make lint and revise * rollback Makefile * Update mteb/tasks/Clustering/kor/PatentFnBClustering.py Co-authored-by: Roman Solomatin <[email protected]> * klue_mrc_domain * make lint * klue_modified_clustering_dataset --------- Co-authored-by: Roman Solomatin <[email protected]> * Rename dunzhang and Jasper models to NovaResearch (#2373) * Rename dunzhang and Jasper models to NovaResearch * rename model in tests * correct reference link * correct MIEB dataset stats (#2374) * correct stats * update Any2AnyMultiChoice qrels stats compute logic * final correction * Update tasks table * Correct -1 to No information in Zero shot (#2381) * fix leaderboard (#2385) * fix: Reduce logging and Warnings (#2349) * Reduce logging and Warnings * make lint * format license to lowercase * Address all comments * Update mteb/leaderboard/app.py Co-authored-by: Roman Solomatin <[email protected]> --------- Co-authored-by: Roman Solomatin <[email protected]> * 1.36.23 Automatically generated by python-semantic-release * fix: b1ade (#2386) * fix: added b1ade_models.py (#2340) * added b1ade_models.py * changing based on requested * Update mteb/models/b1ade_models.py Co-authored-by: Kenneth Enevoldsen <[email protected]> --------- Co-authored-by: Kenneth Enevoldsen <[email protected]> * fix: missing import and formatting --------- Co-authored-by: Shreyas Subramanian <[email protected]> * 1.36.24 Automatically generated by python-semantic-release * fix: pin gradio dependency to ensure leaderboards works (#2387) * 1.36.25 Automatically generated by python-semantic-release * fix: Ensure BrightRetrieval is valid to run (#2334) * fix: Ensure BrightRetrieval is valid to run Not sure this is the best way to fix this. Let me know if you can find a better fix. fixes #2327 * fix: convert brightretrieval to two tasks * fix collecting error * Update tasks table * 1.36.26 Automatically generated by python-semantic-release * Pass task name to all evaluators (#2389) * pass task name to all tasks * add test * fix loader * fix: renaming Zeroshot -> ZeroShot (#2395) * fix: renaming Zeroshot -> ZeroShot Adresses #2078 * rename 1 * rename 2 * format * fixed error * 1.36.27 Automatically generated by python-semantic-release * fix: Update AmazonPolarityClassification license (#2402) Update AmazonPolarityClassification.py * fix b1ade name (#2403) * 1.36.28 Automatically generated by python-semantic-release * Minor style changes (#2396) * fix: renaming Zeroshot -> ZeroShot Adresses #2078 * fix: minor style changes Adresses #2078 * rename 1 * rename 2 * format * fixed error --------- Co-authored-by: Isaac Chung <[email protected]> * Added new dataset and tasks - ClusTREC-covid , clustering of thematic covid related scientific papers (#2302) * Clustrec covid new dataset and task * fix * fix * fix * fix * fix * descriptive stats * change all mentions of clustrec-covidp2p to clustrec-covid * change ' to " * Update tasks table * fix: Major updates to docs + make mieb dep optional (#2397) * fix: renaming Zeroshot -> ZeroShot Adresses #2078 * fix: minor style changes Adresses #2078 * fix: Major updates to documentation This PR does the following: - This introduced other modalities more clearly in the documentation as well as make it easier to transition to a full on documentation site later. - added minor code updates due to discovered inconsistencies in docs and code. - Added the MMTEB citation where applicable - makes the docs ready to move torchvision to an optional dependency * Moved VISTA example * rename 1 * rename 2 * format * fixed error * fix: make torchvision optional (#2399) * fix: make torchvision optional * format * add docs * minor fix * remove transform from Any2TextMultipleChoiceEvaluator --------- Co-authored-by: Isaac Chung <[email protected]> * move Running SentenceTransformer model with prompts to usage --------- Co-authored-by: Isaac Chung <[email protected]> * 1.36.29 Automatically generated by python-semantic-release * remove Arabic_Triplet_Matryoshka_V2.py (#2405) * Min torchvision>0.2.1 (#2410) matching torch>1.0.0 * fix: Add validation to model_name in `ModelMeta` (#2404) * add test for name validation * upd docs * upd cohere name * fix tests * fix name for average_word_embeddings_komninos * fix name for average_word_embeddings_komninos * fix reranker test * fix reranker test * 1.36.30 Automatically generated by python-semantic-release * [MIEB] "capability measured"-Abstask 1-1 matching refactor [1/3]: reimplement CV-Bench (#2414) * refactor CV-Bench * reimplement CV Bench * remove abstask/evaluator/tests for Any2TextMultipleChoice * rerun descriptive stats * Update tasks table * fix: Add option to remove benchmark from leaderboard (#2417) fix: Add option to remove leaderboard from leaderboard fixes #2413 This only removed the benchmark from the leaderboard but keep it in MTEB. * 1.36.31 Automatically generated by python-semantic-release * fix: Add VDR Multilingual Dataset (#2408) * Added VDR Multilingual Dataset * address comments * make lint * Formated Dataset for retrieval * Update mteb/tasks/Retrieval/multilingual/VdrMultilingualRetrieval.py Co-authored-by: Roman Solomatin <[email protected]> * Update mteb/tasks/Retrieval/multilingual/VdrMultilingualRetrieval.py Co-authored-by: Roman Solomatin <[email protected]> * make lint * corrected date * fix dataset building * move to image folder --------- Co-authored-by: Roman Solomatin <[email protected]> Co-authored-by: Isaac Chung <[email protected]> * Update tasks table * 1.36.32 Automatically generated by python-semantic-release * HOTFIX: pin setuptools (#2423) * pin setuptools * pin setuptools * pin setuptools in makefile * try ci * fix ci * remove speed from installs * add __init__.py Clustering > kor folder, And edit __init__.py in Clustering folder (#2422) * add PatentFnBClustering.py * do make lint and revise * rollback Makefile * Update mteb/tasks/Clustering/kor/PatentFnBClustering.py Co-authored-by: Roman Solomatin <[email protected]> * klue_mrc_domain * make lint * klue_modified_clustering_dataset * clustering & kor folder add __init.py * clustering & kor folder add __init__.py * task.py roll-back * correct text_creation to sample_creation & delete form in MetaData * correct task_subtype in TaskMetaData * delete space * edit metadata * edit task_subtypes --------- Co-authored-by: Roman Solomatin <[email protected]> * Update tasks table * Update speed dependencies with new setuptools release (#2429) * add richinfoai models (#2427) * add richinfoai models add richinfoai models * format codes by linter format codes by linter * Added Memory Usage column on leaderboard (#2428) * docs: typos; Standardize spacing; Chronological order (#2436) * Fix typos; add chrono order * Fix spacing * fix: Add model specific dependencies in pyproject.toml (#2424) * Add model specific dependencies in pyproject.toml * Update documentation * 1.36.33 Automatically generated by python-semantic-release * [MIEB] "capability measured"-Abstask 1-1 matching refactor [2/3]: reimplement r-Oxford and r-Paris (#2442) * MutipleChoiceEvaluationMixin; reimplement r-Oxford and r-Paris; rerun stats * modify benchmark list * fix citation * Update tasks table * Error while evaluating MIRACLRetrievalHardNegatives: 'trust_remote_code' (#2445) Fixes #2444 * Feat/searchmap preview (#2420) * Added meta information about SearchMap_Preview model to the model_dir * Added meta information about SearchMap_Preview model to the model_dir * updated revision name * Device loading and cuda cache cleaning step left out * removed task instructions since it's not necessary * changed sentence transformer loader to mteb default loader and passed instructions s model prompts * Included searchmap to the models overview page * Included searchmap to the models overview page * added meta data information about where model was adpated from * Update mteb/models/searchmap_models.py * fix lint * lint --------- Co-authored-by: Roman Solomatin <[email protected]> Co-authored-by: Roman Solomatin <[email protected]> * Add Background Gradients in Summary and Task Table (#2392) * Add Background Gradients in Summary and Task Table * Remove warnings and add light green cmap * Address comments * Separate styling function * address comments * added comments * add ops_moa_models (#2439) * add ops_moa_models * add custom implementations * Simplify custom implementation and format the code * support SentenceTransformers * add training datasets * Update mteb/models/ops_moa_models.py Co-authored-by: Roman Solomatin <[email protected]> * update training_datasets --------- Co-authored-by: kunka.xgw <[email protected]> Co-authored-by: Roman Solomatin <[email protected]> * leaderboard fix (#2456) * ci: cache `~/.cache/huggingface` (#2464) ci: cache ~/.cache/huggingface Co-authored-by: sam021313 <[email protected]> * [MIEB] "capability measured"-Abstask 1-1 matching refactor [3/3]: reimplement ImageCoDe (#2468) * reimplement ImageCoDe with ImageTextPairClassification * add missing stats file * Update tasks table * fix: Adds family of NeuML/pubmedbert-base-embedding models (#2443) * feat: added pubmedbert model2vec models * fix: attribute model_name * fix: fixed commit hash for pubmed_bert model2vec models * fix: changes requested in PR 2443 * fix: add nb_sbert model (#2339) * add_nb_sbert_model * Update nb_sbert.py added n_parameters and release_date * Update mteb/models/nb_sbert.py Co-authored-by: Roman Solomatin <[email protected]> * Update nb_sbert.py fix: make lint * added nb_sbert to overview.py + ran make lint * Update nb_sbert.py Fix error: Input should be a valid date or datetime, month value is outside expected range of 1-12 --------- Co-authored-by: Roman Solomatin <[email protected]> * 1.36.34 Automatically generated by python-semantic-release * fix test --------- Signed-off-by: Michael Dinzinger <[email protected]> Co-authored-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com> Co-authored-by: Kenneth Enevoldsen <[email protected]> Co-authored-by: github-actions <[email protected]> Co-authored-by: Roman Solomatin <[email protected]> Co-authored-by: Mina Parham <[email protected]> Co-authored-by: Mina Parham <[email protected]> Co-authored-by: Mehrzad Shahin-Moghadam <[email protected]> Co-authored-by: Roman Solomatin <[email protected]> Co-authored-by: Sam <[email protected]> Co-authored-by: sam021313 <[email protected]> Co-authored-by: Shikhar Shiromani <[email protected]> Co-authored-by: Shikhar Shiromani <[email protected]> Co-authored-by: Ruslan Bel'kov <[email protected]> Co-authored-by: Márton Kardos <[email protected]> Co-authored-by: sufen-f <[email protected]> Co-authored-by: sufen <[email protected]> Co-authored-by: Niklas Muennighoff <[email protected]> Co-authored-by: Samuel Yang <[email protected]> Co-authored-by: Aradhye Agarwal <[email protected]> Co-authored-by: Tom Aarsen <[email protected]> Co-authored-by: talshef <[email protected]> Co-authored-by: Tal Sheffer <[email protected]> Co-authored-by: garciasces <[email protected]> Co-authored-by: gowitheflow-1998 <[email protected]> Co-authored-by: Wang Bo <[email protected]> Co-authored-by: Munot Ayush Sunil <[email protected]> Co-authored-by: Yaya Sy <[email protected]> Co-authored-by: Imene Kerboua <[email protected]> Co-authored-by: Eng. Omar Najar <[email protected]> Co-authored-by: Michael Dinzinger <[email protected]> Co-authored-by: Jinhyuk Lee <[email protected]> Co-authored-by: Isaac Chung <[email protected]> Co-authored-by: sergeyz-zh <[email protected]> Co-authored-by: OnandOn <[email protected]> Co-authored-by: chenghao xiao <[email protected]> Co-authored-by: Shreyas Subramanian <[email protected]> Co-authored-by: Uri K <[email protected]> Co-authored-by: richinfo-ai <[email protected]> Co-authored-by: Adewole Babatunde <[email protected]> Co-authored-by: ahxgw <[email protected]> Co-authored-by: kunka.xgw <[email protected]> Co-authored-by: Nadia Sheikh <[email protected]> Co-authored-by: theatollersrud <[email protected]>
isaac-chung
added a commit
that referenced
this pull request
Apr 4, 2025
* misc: Add image classification descriptive stats implementation (#2045) * add ImageClassificationDescriptiveStatistics * add MNIST descriptive stats * use tuples instead * add label count and update docstrings * update MNIST example * Update tasks table * fix: Add column descriptions to leaderboard (#2039) * fix: Add column descriptions to leaderboard * removed existing overlap * fix: Add BRIGHT (long) and fix bug in TaskResult.filter_and_validate() (#2041) * fix: Add BRIGHT Long Fixes #1978 * fix: Add BRIGHT(long) * fix bug in task results * updated bright * updated tests for TaskResults * 1.34.12 Automatically generated by python-semantic-release * misc: Add image clustering descriptive stats implementation (#2057) * add image clustering descirptive stats and run * finish off last one * remove script * fix: Update embed_dim for jina models (#2058) see embeddings-benchmark/results#117 * Update tasks table * 1.34.13 Automatically generated by python-semantic-release * Add giga embeddings (#1741) * add gigaembeddings * use jasper * fix name * create sentence_transformer instruct wrapper * apply instruction template * fix jasper * update meta * misc: Add ZS and multilabel image classification descriptive stats implementation (#2059) * add image clustering descirptive stats and run * finish off last one * remove script * add ImageMultilabelClassificationDescriptiveStatistics * add VOC2007 * add zeroshot and mnist example * Update tasks table * Rename MIEB task classes with duplicated names (#2061) fix class names * misc: Add VisualSTS descriptive stats (#2062) * add visualsts stats * add last dataset * Update tasks table * fix: Added gte models (#1539) * fix: Added gte models * fix: Add mixbai models (#1540) for #1515 * fix: Add climate fever v2 (#1873) * Updated ClimateFEVER dataset with new version * Adds Fill in the empty metadata. * Updates the date tuple * Update class name Co-authored-by: Kenneth Enevoldsen <[email protected]> * Update domains Co-authored-by: Kenneth Enevoldsen <[email protected]> * Update task_subtypes * Update annotations_creators for the first version * Update date Co-authored-by: Kenneth Enevoldsen <[email protected]> * Update task subtypes * Update path * Update description --------- Co-authored-by: Kenneth Enevoldsen <[email protected]> Co-authored-by: Mina Parham <[email protected]> * Update tasks table * fix: Updating paper scripts (#1958) * change reference revisions to align with paper * Update author list * Added code for main results table * updated minor changes * added external as a "no_revision_available" case * revert unintended changes * format * 1.34.14 Automatically generated by python-semantic-release * Add datasets for a benchmark newly introduced for "Engineering" domain (#1911) * adding clustering tasks (built-bench-clustering S2S & P2P) * updated built-bench-clustering tasks * Updated BuiltBenchClustering tasks * Added "Engineering" as new domain to TaskMetadata.py * Updated tasks table in docs * Updated task metadata for BuiltBenchClustering S2S and P2P * updated metadata for clustering tasks * Add/update BuiltBench tasks - Add BuiltBenchRetrieval task - Add BuiltBenchReranking task - Update metadata for BuiltBenchClusterinP2P - Update metadata for BuiltBenchClusterinS2S * update BuiltBench benchmark * Update mteb/benchmarks/benchmarks.py Co-authored-by: Roman Solomatin <[email protected]> * Update mteb/tasks/Clustering/eng/BuiltBenchClusteringS2S.py Co-authored-by: Roman Solomatin <[email protected]> * Update mteb/tasks/Clustering/eng/BuiltBenchClusteringP2P.py Co-authored-by: Roman Solomatin <[email protected]> * Update mteb/benchmarks/benchmarks.py Co-authored-by: Isaac Chung <[email protected]> * Fix formatting via ruff --------- Co-authored-by: Roman Solomatin <[email protected]> Co-authored-by: Isaac Chung <[email protected]> * Update tasks table * misc: update model names to adjust for adding to results repo (#2074) * update model names to adjust for adding to results repo * update model meta script * misc: Add all image classification descriptive stats (#2073) * add most image classification descr stats * revert changes to encoder * add stats --------- Co-authored-by: Roman Solomatin <[email protected]> * Update tasks table * ci: Rerun tests that fail due to networking issues. (#2029) * fix: rerun tests that fail - Networking * update tests to use tmp_path * set versions for dev dependencies * add pytest options to pyproject.toml * add rerun json.decoder.JSONDecodeError * remove JSONDecodeError from pyproject.toml * add huggingface_hub.errors.HfHubHTTPError * add huggingface_hub.errors.LocalEntryNotFoundError https://github.com/embeddings-benchmark/mteb/actions/runs/13298535701/job/37139767443?pr=2044 * FileNotFoundError https://github.com/embeddings-benchmark/mteb/actions/runs/13302915091/job/37147507251?pr=2029 * add doc to pytest rerun --------- Co-authored-by: sam021313 <[email protected]> * fix: generate metadata (#2063) * fix: generate metadata * use logging not print for script * lint * add iso639 to dev pyproject * fix import * add memory_usage_mb * set version for iso639 Co-authored-by: Kenneth Enevoldsen <[email protected]> --------- Co-authored-by: sam021313 <[email protected]> Co-authored-by: Kenneth Enevoldsen <[email protected]> Co-authored-by: Roman Solomatin <[email protected]> * 1.34.15 Automatically generated by python-semantic-release * fix: add missing `e5` training datasets (#2065) add missing training datasets * 1.34.16 Automatically generated by python-semantic-release * fix: Ensure voyage model uses different naming scheme (#2083) * fix: Added make command for running leaderboard locally * fix: Ensure voyage models doesn't re-use the name * 1.34.17 Automatically generated by python-semantic-release * fix: Freeze model/rank columns in leaderboard (#2044) * fix: freeze model/rank columns in leaderboard * freezing zero-shot column * update min gradio version to 5.16.0 in pyproject.toml --------- Co-authored-by: Shikhar Shiromani <[email protected]> * 1.34.18 Automatically generated by python-semantic-release * fix: Fixed previous incorrect specification of splits for CMTEB ( MTEB(cmn, v1) ) (#2086) Fixes #2064 * 1.34.19 Automatically generated by python-semantic-release * Remove duplicated string in docstring of TaskMetadata class (#2087) * Remove duplicated string in docstring of TaskMetadata class * Remove duplicated dataset field * fix: Smarter leaderboard caching with cachetools (#2085) * Added smarter caching to callbacks * Added cachetools as a dependency * Ran linting * Removed debugging print statement * Bumped Gradio version * Dependency fixes * Dependency fixes --------- Co-authored-by: Kenneth Enevoldsen <[email protected]> * fix: Missing fixes for #2086 - change MultilingualSentiment split from test to validation in CMTEB (#2088) * fix: Fixed previous incorrect specification of splits for CMTEB ( MTEB(cmn, v1) ) Fixes #2064 * change MultilingualSentiment split from test to validation in CMTEB * 1.34.20 Automatically generated by python-semantic-release * merge gme models (#2089) * fix: Add back task filtering by modalities (#2080) * add back task filtering by modalities * add unit test * check if task modalities is a subset of model modalities and fix tests * add model_modalities_more_than_task_modalities case * 1.34.21 Automatically generated by python-semantic-release * Added gtr-t5-base/large/xl/xxl metadata to mteb (#2092) * Added GTR Models to codebase * Linted gtr models file. * Added gtr-base/large/xl/xxl to sentence_transformers_models.py * Added memory_usage_mb and training_datasets * Reformatted training dataset names * Reformatted training dataset names * Reformatted training dataset names --------- Co-authored-by: sufen <[email protected]> * misc: Add Any2TextMutipleChoice Descriptive Statistics (#2095) * add Any2TextMutipleChoiceDescriptiveStatistics * run on all tasks * Update tasks table * fix: Updated model annotations for GTE, e5, gritlm, and SFR models (#2101) Reported with references to paper + qoutes. * fix: Update links (#2098) * Fix link * Fix link * 1.34.22 Automatically generated by python-semantic-release * Add model inf-retriever-v1-1.5b (#2106) Add inf-retriever-v1-1.5b model * docs: Fix typos & refine text (#2102) * Update app.py * Fix typos * misc: Run Zeroshot Classification Descriptive Stats (#2105) * add most datasets * add birdsnap and imgnet1k * add scimmir and sun397 * add uck101 zs * Update tasks table * fix: add warning about task category conversion (#2108) add warning about task category conversion * 1.34.23 Automatically generated by python-semantic-release * fix: Add codesage-large-v2 (#2090) * Add codesage-large-v2 * Address comments * Add training dataset * Fix issues * Format code * Remove unnecessary wrapper * 1.34.24 Automatically generated by python-semantic-release * fix: add training data to BGE-m3-custom-fr (#2110) This ensure that is it correctly filtered as non-zero-shot * 1.34.25 Automatically generated by python-semantic-release * fix: Upgrade ruff to be gradio compatible (#2111) * fix: update ruff to be gradio compatible (>=0.9.3) * format * fix: upgrade ruff to latests (same as gradio compatible) * 1.34.26 Automatically generated by python-semantic-release * docs: Follow google docstring format (#2115) Fixes #2113 * Update leaderboard_refresh.yaml (#2121) * fix InstructSentenceTransformer Model name (#2125) fix params * fix voyage (#2127) * fix: update e5 instruct training data (#2129) update e5 training data * 1.34.27 Automatically generated by python-semantic-release * format * Update tasks table * fix: Add 2 new Static Sentence Transformer models (#2112) * Add 2 new Static Sentence Transformer models * Add Tatoeba Co-authored-by: Roman Solomatin <[email protected]> --------- Co-authored-by: Roman Solomatin <[email protected]> * 1.34.28 Automatically generated by python-semantic-release * add is_cross_encoder (#1869) * add is_cross_encoder * Update mteb/model_meta.py Co-authored-by: Kenneth Enevoldsen <[email protected]> * change value --------- Co-authored-by: Kenneth Enevoldsen <[email protected]> * Qodo embed 1 1.5 b (#2137) * feat: Add Qodo-Embed-1-1.5B model metadata * fix: Add Qodo models to overview imports * fix: Add adapted_from field to Qodo model metadata * Update mteb/models/qodo_models.py Co-authored-by: Roman Solomatin <[email protected]> * relint --------- Co-authored-by: Tal Sheffer <[email protected]> Co-authored-by: Roman Solomatin <[email protected]> * misc: merge summary retrieval into bitext mining (#2140) merge summary retrieval into bitext mining * test: fix dataset availability test (#2141) This simplified the test and also make it a lot simpler. It also removed about 100 test cases which where all to the same API call. * fix: Update NVIDIA-Embed training data (#2143) Added a few missing annotations for nvidia-embed * 1.34.29 Automatically generated by python-semantic-release * fix: Add annotations for Voyage exp (#2144) * fix: Update NVIDIA-Embed training data Added a few missing annotations for nvidia-embed * fix update annotationf for voyage exp * 1.34.30 Automatically generated by python-semantic-release * Fix tokens num in cde models (#2148) fix tokens * feat: Add Qodo-Embed-1-7B model metadata and rename existing model (#2146) * feat: Add Qodo-Embed-1-7B model metadata and rename existing model * lint * fix revision * update license name --------- Co-authored-by: Tal Sheffer <[email protected]> * 1.35.0 Automatically generated by python-semantic-release * misc: add Any2AnyRetrievalDescriptiveStatistics (#2139) add Any2AnyRetrievalDescriptiveStatistics * Update tasks table * Added zero-shot percentages and different filtering scheme (#2153) * Added zero-shot percentages and different filtering scheme * Update mteb/model_meta.py Co-authored-by: Roman Solomatin <[email protected]> --------- Co-authored-by: Roman Solomatin <[email protected]> * fix: Incorrect annotations for Mistral-based embedding models (#2157) Fixes #2155 * 1.35.1 Automatically generated by python-semantic-release * Update FaMTEBRetrieval.py (#2171) The URL pointed to the settings page instead of the main repo URL. Now it is fixed. * Update tasks table * fix: Add Training data annotations (#2173) * redo to voyage to only training data * Add training data annotation for Kalm embeddings #2168 * Add correct training data annotations to Stella #2164 * removed fiqa PL as it does not exist * remove ArxivClusteringS2S.v2 as it does not exist * Add training data annotation for GIST embedding #2166 * fix max tokens for kalm models #2162 * remove eli 5 * 1.35.2 Automatically generated by python-semantic-release * feat: Add MIEB and MIEB-lite as benchmarks (#2035) * add mieb and mieb-lite to benchmarks * add CompositionalityEvaluation and DocumentUnderstanding types * add VisionCentric type * add missing comma * split STS17MultilingualVisualSTS and STSBenchmarkMultilingualSTS to eng and non-eng * use aggregate task instead so we can name the subsets * shorten names * fix import * alternative strategy to avoid using get_task * follow other aggregate tasks and skip metadata test * run LB without errors when selecting MIEB(-lite) * add back the capability as taks type * typo * extend description * split into mieb(eng) and mieb(multilingual) * remove unneeded files * remove aggtask additions for test * edit descriptions based on screenshots * shorten * rename to Compositionality and include ImageCoDeT2IMultiChoice * re-tag missing VisionCentric tasks * re-tag rparis and roxford as retrieval and include fixes * re-tag voc2007 as image cls * make lint * correct num task types in descriptions * add one model to models_to_annotate * add mieb reference models * update task types * relabel to multilingual retrieval task type to align with paper * fix reference and bibtex * edit task list to match with final list * add back agg task to reproduce table column in paper * fix filtering and import * update tests * mieb lite add back missing tasks * fix metadata test * multi should have all 4 variants * fix task counts * lite has 10 task types * fix visualSTS-17 lang splits * Aggregate task can now use subsets & eval langs to filter TaskResults * fix test and mark VisualSTS17 as multilingual * fix tests * add agg task running script * add voyage meta * fix citations * capitalize * add coarse/fine labels --------- Co-authored-by: gowitheflow-1998 <[email protected]> * Update tasks table * 1.36.0 Automatically generated by python-semantic-release * fix: update training datasets and revision for jina models (#2179) * feat: update training datasets and revision for jina models * feat: update training datasets and revision for jina models * fix: Add more training data annotations (#2178) * redo to voyage to only training data * Add training data annotation for Kalm embeddings #2168 * Add correct training data annotations to Stella #2164 * removed fiqa PL as it does not exist * remove ArxivClusteringS2S.v2 as it does not exist * Add training data annotation for GIST embedding #2166 * fix max tokens for kalm models #2162 * remove eli 5 * fix: add training data for Bilingual Embeddings fixes #2167 * 1.36.1 Automatically generated by python-semantic-release * Added training data annotation for e5-base-4k (#2186) * fix: Added training data annotations to MXBAI (#2185) * fix: Update MTEB(Scandinavian) to use new DanFEVER (#2180) This also resolves the missing data in the leaderboard. Fixes #2172 * fix: Added training data annotation for MMLW models (#2188) * Added training data annotation for MMLW models * Added GIST annotations Kenneth missed * Added Stella en 400m training data' * 1.36.2 Automatically generated by python-semantic-release * fix: Added training data for sentence-croissant (#2189) * 1.36.3 Automatically generated by python-semantic-release * fix: update ru models annotation (#2181) * 1.36.4 Automatically generated by python-semantic-release * fix: Alphabetical ordering of tasks in dropdowns (#2191) * 1.36.5 Automatically generated by python-semantic-release * misc: Speed up qrel creation in any2anyretrieval (#2196) * use numpy vectorized operations instead of row-by-row * scores are int * use 'mteb.MTEB' instead of 'MTEB' for custom model (#2199) * add base models for e5 (#2183) * add similar datasets (#2205) * add similar datasets * add nano * update is filled * Update mteb/abstasks/TaskMetadata.py Co-authored-by: Kenneth Enevoldsen <[email protected]> --------- Co-authored-by: Kenneth Enevoldsen <[email protected]> * add labse annotation (#2182) * add labse annotation * Update mteb/models/sentence_transformers_models.py Co-authored-by: Kenneth Enevoldsen <[email protected]> --------- Co-authored-by: Kenneth Enevoldsen <[email protected]> * fix: Fixed leaderboard crash (#2221) * Fixed leaderboard crash * Fixed language selection error * Ran linting * 1.36.6 Automatically generated by python-semantic-release * fix: More training data annotations (#2220) * Added training data annotation for bge-gemma * Added missing annotations for Voyage models * Added training data for sts-multilingual-mpnet * Added all mteb datasets to STS-multilingual training data * 1.36.7 Automatically generated by python-semantic-release * Add LLM2CLIP (OpenAI variants) (#2222) * model loading and get_text_embeddings * add image_emb, fused_emb, and calc probs methods * add b16 model * add llm2clip_openai_l_14_224 (not working yet) * got llm2clip_openai_l_14_224 working * make lint * add training sets and allow py files * Change `dataset on HF` test to use official api (#2213) * refactor dataset checking * increase timeout * increase timeout * remove timeout * Descriptive stats functions for Any2AnyMC and ImageTextPC (#2197) * Add Any2AnyMC descriptive stats * Add descriptive stats function for ImageTextPC * add descriptive stats examples * linter * update multi choice descriptive stats * Update tasks table * fix: Add training data annotations to uderver-bloom models (#2210) * fix: Add training data annotations to uderver-bloom models fixes #2193 * fix: add mixedbread --------- Co-authored-by: Márton Kardos <[email protected]> * 1.36.8 Automatically generated by python-semantic-release * Add comment to `voyage-3-m-exp` model (#2229) * remove model size from voyage-3-m-exp model * Update mteb/models/voyage_models.py * Update mteb/models/voyage_models.py * docs: Update description of EURLex (#2231) * Automatically add similar tasks to training_tasks (#2228) * refactor dataset checking * increase timeout * increase timeout * remove timeout * start * automatically find datasets * update comment * fix aggregate task metadata * fixes * lint * rename * update fetch check * Remove overlapping legends from radar chart (#2195) * Remove overlapping legends from radar chart * ensure graph is not blocked * Overlapping legend issue of Radar Chart * misc: Run Any2AnyRetrieval descriptive stats (#2223) * run a few datasets * add a few more * run more tasks * add more datasets * remove pdb * remove newline * add more datasets * Update tasks table * misc: Add rest of the vision centric and compositionality descriptive stats (#2267) add the rest * Update tasks table * Fix `calculate_memory_usage_mb` in adding_a_model.md (#2271) * Add Arabic-Triplet-Matryoshka-V2 model metadata to MTEB (#2270) * Add Arabic-Triplet-Matryoshka-V2 model metadata to MTEB * Update memory_usage_mb with correct calculated value * Update mteb/models/Arabic_Triplet_Matryoshka_V2.py Co-authored-by: Roman Solomatin <[email protected]> * Update mteb/models/Arabic_Triplet_Matryoshka_V2.py Co-authored-by: Roman Solomatin <[email protected]> * remove comments * added correct memory usage * Update mteb/models/Arabic_Triplet_Matryoshka_V2.py Co-authored-by: Roman Solomatin <[email protected]> * Apply linter fixes with ruff * Update mteb/models/Arabic_Triplet_Matryoshka_V2.py Co-authored-by: Roman Solomatin <[email protected]> * Update mteb/models/Arabic_Triplet_Matryoshka_V2.py Co-authored-by: Roman Solomatin <[email protected]> * Add Arabic_Triplet_Matryoshka_V2 to overview.py * Rename model file to ara_models.py and update imports --------- Co-authored-by: Roman Solomatin <[email protected]> * fix: Add WebFAQ Retrieval dataset (#2236) * Add WebFAQ Retrieval dataset Signed-off-by: Michael Dinzinger <[email protected]> * Small change WebFAQRetrieval.py Signed-off-by: Michael Dinzinger <[email protected]> * Add remaining languages to WebFAQ Retrieval task Signed-off-by: Michael Dinzinger <[email protected]> * Add descriptive stats Signed-off-by: Michael Dinzinger <[email protected]> --------- Signed-off-by: Michael Dinzinger <[email protected]> * Update tasks table * 1.36.9 Automatically generated by python-semantic-release * fix: Formatting issue in Performance Plot (#2237) * Formatting issue in Performance Plot * make lint * added function for better code readability * 1.36.10 Automatically generated by python-semantic-release * ci: run test_dataset_on_hf separately (#2201) * dont run test_dataset_on_hf in every pr * lint * Update call pytest test_datasets Co-authored-by: Kenneth Enevoldsen <[email protected]> * Update tests/test_tasks/test_all_abstasks.py Co-authored-by: Kenneth Enevoldsen <[email protected]> * not datasets for test * run dataset loading test for push or pull_request * apply feedback --------- Co-authored-by: sam021313 <[email protected]> Co-authored-by: Kenneth Enevoldsen <[email protected]> * add gemini-embedding-exp-03-07 (#2279) * add gemini-embedding-exp-03-07 * remove space for lint * lint fix * update link (#2281) * fix: Run remaining MIEB desc stats (#2288) * run Vidore * GLDv2 * run the rest --------- Co-authored-by: Isaac Chung <[email protected]> * Update tasks table * 1.36.11 Automatically generated by python-semantic-release * fix: Added Filter Modality (#2262) * Added Filter Modality * resolve suggestions * make lint * make sure test pass * make lint * added exclusive_modality_filter and unit tests * Integrate tests on overview.py * Update tests/test_overview.py Co-authored-by: Roman Solomatin <[email protected]> * added task related to image modality * Update mteb/abstasks/AbsTask.py Co-authored-by: Kenneth Enevoldsen <[email protected]> * Update mteb/abstasks/AbsTask.py Co-authored-by: Kenneth Enevoldsen <[email protected]> * update overview..py * make lint * update documentation --------- Co-authored-by: Roman Solomatin <[email protected]> Co-authored-by: Kenneth Enevoldsen <[email protected]> * 1.36.12 Automatically generated by python-semantic-release * fix: Add `ModelMeta` license & custom validations (#2293) * license validation * move licenses * update imports --------- Co-authored-by: Isaac Chung <[email protected]> * 1.36.13 Automatically generated by python-semantic-release * ci: Add pre-commit hook (#2194) * make dev life nicer - pre-commit hooks * add pre-commit to install * update precommit * update ruff pre-commit * lint * lint --------- Co-authored-by: sam021313 <[email protected]> * Update tasks table * fix: bug in voyage implementation (#2304) * fix: Fix bug in voyage implementation "passage" is not a valid input for the voyage API. Remapped to "document". * Update mteb/models/voyage_models.py Co-authored-by: Roman Solomatin <[email protected]> --------- Co-authored-by: Roman Solomatin <[email protected]> * 1.36.14 Automatically generated by python-semantic-release * fix: Update voyage name to include Org. (#2322) * 1.36.15 Automatically generated by python-semantic-release * Added VDR Model (#2290) * Added VDR Model * change custom wrapper to SentenceTransformer Wrapper * remove kwargs and add TODO for Image Modality * Update mteb/models/vdr_models.py Co-authored-by: Roman Solomatin <[email protected]> --------- Co-authored-by: Roman Solomatin <[email protected]> * fix: Resolve conflicting dependencies (#2323) These errors where discovered when trying to install the package using `uv`. We have a problem with salesforce-lavis, which is not compatible with the current set of dependencies. * 1.36.16 Automatically generated by python-semantic-release * fix: remove SyntaxWarnings in py312 (#2325) * fix: Resolve conflicting dependencies These errors where discovered when trying to install the package using `uv`. We have a problem with salesforce-lavis, which is not compatible with the current set of dependencies. * fix: Remove syntax warnings occuring in python 3.12 ``` Python 3.12.0 (main, Oct 2 2023, 20:56:14) [Clang 16.0.3 ] on darwin Type "help", "copyright", "credits" or "license" for more information. >>> import mteb # no syntax warnings >>> ``` * 1.36.17 Automatically generated by python-semantic-release * fix: add annotation models for stella zh (#2277) * fix: add annotation models for stella zh Additionally fixed a few annotation errors * format * Update mteb/models/stella_models.py Co-authored-by: Isaac Chung <[email protected]> --------- Co-authored-by: Isaac Chung <[email protected]> * 1.36.18 Automatically generated by python-semantic-release * fix: Add ModelMeta rubert-mini-frida, BERTA (#2330) * Add rubert-mini-frida model meta * Add BERTA model meta * docs: fix typos * 1.36.19 Automatically generated by python-semantic-release * fix: Add WebFAQ bitext mining tasks (#2326) * Add WebFAQ bitext mining tasks Signed-off-by: Michael Dinzinger <[email protected]> * Lower number of language pairs in WebFAQBitextMining Signed-off-by: Michael Dinzinger <[email protected]> --------- Signed-off-by: Michael Dinzinger <[email protected]> * Update tasks table * 1.36.20 Automatically generated by python-semantic-release * fix: Add `trust_remote_code` to MIRACLRetrieval * fix: Add `trust_remote_code` to MIRACLRetrieval (#2344) * 1.36.21 Automatically generated by python-semantic-release * fix: Correctly pass trust remote code to Miracl * fix: Ensure MIRACL pass trust_remote_code (#2346) * fix: Add `trust_remote_code` to MIRACLRetrieval * fix: Correctly pass trust remote code to Miracl * fix * 1.36.22 Automatically generated by python-semantic-release * add-Data Korean Clustering dataset (KLUE-modified) (#2283) * add PatentFnBClustering.py * do make lint and revise * rollback Makefile * Update mteb/tasks/Clustering/kor/PatentFnBClustering.py Co-authored-by: Roman Solomatin <[email protected]> * klue_mrc_domain * make lint * klue_modified_clustering_dataset --------- Co-authored-by: Roman Solomatin <[email protected]> * Rename dunzhang and Jasper models to NovaResearch (#2373) * Rename dunzhang and Jasper models to NovaResearch * rename model in tests * correct reference link * correct MIEB dataset stats (#2374) * correct stats * update Any2AnyMultiChoice qrels stats compute logic * final correction * Update tasks table * Correct -1 to No information in Zero shot (#2381) * fix leaderboard (#2385) * fix: Reduce logging and Warnings (#2349) * Reduce logging and Warnings * make lint * format license to lowercase * Address all comments * Update mteb/leaderboard/app.py Co-authored-by: Roman Solomatin <[email protected]> --------- Co-authored-by: Roman Solomatin <[email protected]> * 1.36.23 Automatically generated by python-semantic-release * fix: b1ade (#2386) * fix: added b1ade_models.py (#2340) * added b1ade_models.py * changing based on requested * Update mteb/models/b1ade_models.py Co-authored-by: Kenneth Enevoldsen <[email protected]> --------- Co-authored-by: Kenneth Enevoldsen <[email protected]> * fix: missing import and formatting --------- Co-authored-by: Shreyas Subramanian <[email protected]> * 1.36.24 Automatically generated by python-semantic-release * fix: pin gradio dependency to ensure leaderboards works (#2387) * 1.36.25 Automatically generated by python-semantic-release * fix: Ensure BrightRetrieval is valid to run (#2334) * fix: Ensure BrightRetrieval is valid to run Not sure this is the best way to fix this. Let me know if you can find a better fix. fixes #2327 * fix: convert brightretrieval to two tasks * fix collecting error * Update tasks table * 1.36.26 Automatically generated by python-semantic-release * Pass task name to all evaluators (#2389) * pass task name to all tasks * add test * fix loader * fix: renaming Zeroshot -> ZeroShot (#2395) * fix: renaming Zeroshot -> ZeroShot Adresses #2078 * rename 1 * rename 2 * format * fixed error * 1.36.27 Automatically generated by python-semantic-release * fix: Update AmazonPolarityClassification license (#2402) Update AmazonPolarityClassification.py * fix b1ade name (#2403) * 1.36.28 Automatically generated by python-semantic-release * Minor style changes (#2396) * fix: renaming Zeroshot -> ZeroShot Adresses #2078 * fix: minor style changes Adresses #2078 * rename 1 * rename 2 * format * fixed error --------- Co-authored-by: Isaac Chung <[email protected]> * Added new dataset and tasks - ClusTREC-covid , clustering of thematic covid related scientific papers (#2302) * Clustrec covid new dataset and task * fix * fix * fix * fix * fix * descriptive stats * change all mentions of clustrec-covidp2p to clustrec-covid * change ' to " * Update tasks table * fix: Major updates to docs + make mieb dep optional (#2397) * fix: renaming Zeroshot -> ZeroShot Adresses #2078 * fix: minor style changes Adresses #2078 * fix: Major updates to documentation This PR does the following: - This introduced other modalities more clearly in the documentation as well as make it easier to transition to a full on documentation site later. - added minor code updates due to discovered inconsistencies in docs and code. - Added the MMTEB citation where applicable - makes the docs ready to move torchvision to an optional dependency * Moved VISTA example * rename 1 * rename 2 * format * fixed error * fix: make torchvision optional (#2399) * fix: make torchvision optional * format * add docs * minor fix * remove transform from Any2TextMultipleChoiceEvaluator --------- Co-authored-by: Isaac Chung <[email protected]> * move Running SentenceTransformer model with prompts to usage --------- Co-authored-by: Isaac Chung <[email protected]> * 1.36.29 Automatically generated by python-semantic-release * remove Arabic_Triplet_Matryoshka_V2.py (#2405) * Min torchvision>0.2.1 (#2410) matching torch>1.0.0 * fix: Add validation to model_name in `ModelMeta` (#2404) * add test for name validation * upd docs * upd cohere name * fix tests * fix name for average_word_embeddings_komninos * fix name for average_word_embeddings_komninos * fix reranker test * fix reranker test * 1.36.30 Automatically generated by python-semantic-release * [MIEB] "capability measured"-Abstask 1-1 matching refactor [1/3]: reimplement CV-Bench (#2414) * refactor CV-Bench * reimplement CV Bench * remove abstask/evaluator/tests for Any2TextMultipleChoice * rerun descriptive stats * Update tasks table * fix: Add option to remove benchmark from leaderboard (#2417) fix: Add option to remove leaderboard from leaderboard fixes #2413 This only removed the benchmark from the leaderboard but keep it in MTEB. * 1.36.31 Automatically generated by python-semantic-release * fix: Add VDR Multilingual Dataset (#2408) * Added VDR Multilingual Dataset * address comments * make lint * Formated Dataset for retrieval * Update mteb/tasks/Retrieval/multilingual/VdrMultilingualRetrieval.py Co-authored-by: Roman Solomatin <[email protected]> * Update mteb/tasks/Retrieval/multilingual/VdrMultilingualRetrieval.py Co-authored-by: Roman Solomatin <[email protected]> * make lint * corrected date * fix dataset building * move to image folder --------- Co-authored-by: Roman Solomatin <[email protected]> Co-authored-by: Isaac Chung <[email protected]> * Update tasks table * 1.36.32 Automatically generated by python-semantic-release * HOTFIX: pin setuptools (#2423) * pin setuptools * pin setuptools * pin setuptools in makefile * try ci * fix ci * remove speed from installs * add __init__.py Clustering > kor folder, And edit __init__.py in Clustering folder (#2422) * add PatentFnBClustering.py * do make lint and revise * rollback Makefile * Update mteb/tasks/Clustering/kor/PatentFnBClustering.py Co-authored-by: Roman Solomatin <[email protected]> * klue_mrc_domain * make lint * klue_modified_clustering_dataset * clustering & kor folder add __init.py * clustering & kor folder add __init__.py * task.py roll-back * correct text_creation to sample_creation & delete form in MetaData * correct task_subtype in TaskMetaData * delete space * edit metadata * edit task_subtypes --------- Co-authored-by: Roman Solomatin <[email protected]> * Update tasks table * Update speed dependencies with new setuptools release (#2429) * add richinfoai models (#2427) * add richinfoai models add richinfoai models * format codes by linter format codes by linter * Added Memory Usage column on leaderboard (#2428) * docs: typos; Standardize spacing; Chronological order (#2436) * Fix typos; add chrono order * Fix spacing * fix: Add model specific dependencies in pyproject.toml (#2424) * Add model specific dependencies in pyproject.toml * Update documentation * 1.36.33 Automatically generated by python-semantic-release * [MIEB] "capability measured"-Abstask 1-1 matching refactor [2/3]: reimplement r-Oxford and r-Paris (#2442) * MutipleChoiceEvaluationMixin; reimplement r-Oxford and r-Paris; rerun stats * modify benchmark list * fix citation * Update tasks table * Error while evaluating MIRACLRetrievalHardNegatives: 'trust_remote_code' (#2445) Fixes #2444 * Feat/searchmap preview (#2420) * Added meta information about SearchMap_Preview model to the model_dir * Added meta information about SearchMap_Preview model to the model_dir * updated revision name * Device loading and cuda cache cleaning step left out * removed task instructions since it's not necessary * changed sentence transformer loader to mteb default loader and passed instructions s model prompts * Included searchmap to the models overview page * Included searchmap to the models overview page * added meta data information about where model was adpated from * Update mteb/models/searchmap_models.py * fix lint * lint --------- Co-authored-by: Roman Solomatin <[email protected]> Co-authored-by: Roman Solomatin <[email protected]> * Add Background Gradients in Summary and Task Table (#2392) * Add Background Gradients in Summary and Task Table * Remove warnings and add light green cmap * Address comments * Separate styling function * address comments * added comments * add ops_moa_models (#2439) * add ops_moa_models * add custom implementations * Simplify custom implementation and format the code * support SentenceTransformers * add training datasets * Update mteb/models/ops_moa_models.py Co-authored-by: Roman Solomatin <[email protected]> * update training_datasets --------- Co-authored-by: kunka.xgw <[email protected]> Co-authored-by: Roman Solomatin <[email protected]> * leaderboard fix (#2456) * ci: cache `~/.cache/huggingface` (#2464) ci: cache ~/.cache/huggingface Co-authored-by: sam021313 <[email protected]> * [MIEB] "capability measured"-Abstask 1-1 matching refactor [3/3]: reimplement ImageCoDe (#2468) * reimplement ImageCoDe with ImageTextPairClassification * add missing stats file * Update tasks table * fix: Adds family of NeuML/pubmedbert-base-embedding models (#2443) * feat: added pubmedbert model2vec models * fix: attribute model_name * fix: fixed commit hash for pubmed_bert model2vec models * fix: changes requested in PR 2443 * fix: add nb_sbert model (#2339) * add_nb_sbert_model * Update nb_sbert.py added n_parameters and release_date * Update mteb/models/nb_sbert.py Co-authored-by: Roman Solomatin <[email protected]> * Update nb_sbert.py fix: make lint * added nb_sbert to overview.py + ran make lint * Update nb_sbert.py Fix error: Input should be a valid date or datetime, month value is outside expected range of 1-12 --------- Co-authored-by: Roman Solomatin <[email protected]> * 1.36.34 Automatically generated by python-semantic-release * suppress logging warnings on leaderboard (#2406) * supress logging warnings * remove loggers * return blocks * rename function * fix gme models * add server name * update after merge * fix ruff * fix: E5 instruct now listed as sbert compatible (#2475) Fixes #1442 * 1.36.35 Automatically generated by python-semantic-release * [MIEB] rename VisionCentric to VisionCentricQA (#2479) rename VisionCentric to VisionCentricQA * ci: Run dataset loading only when pushing to main (#2480) Update dataset_loading.yml * fix table in tasks.md (#2483) * Update tasks table * fix: add prompt to NanoDBPedia (#2486) * 1.36.36 Automatically generated by python-semantic-release * Fix Task Lang Table (#2487) * Fix Task Lang Table * added tasks.md * fix * fix: Ignore datasets not available in tests (#2484) * add back MockAudioEncoder --------- Signed-off-by: Michael Dinzinger <[email protected]> Co-authored-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com> Co-authored-by: Kenneth Enevoldsen <[email protected]> Co-authored-by: github-actions <[email protected]> Co-authored-by: Roman Solomatin <[email protected]> Co-authored-by: Mina Parham <[email protected]> Co-authored-by: Mina Parham <[email protected]> Co-authored-by: Mehrzad Shahin-Moghadam <[email protected]> Co-authored-by: Roman Solomatin <[email protected]> Co-authored-by: Sam <[email protected]> Co-authored-by: sam021313 <[email protected]> Co-authored-by: Shikhar Shiromani <[email protected]> Co-authored-by: Shikhar Shiromani <[email protected]> Co-authored-by: Ruslan Bel'kov <[email protected]> Co-authored-by: Márton Kardos <[email protected]> Co-authored-by: sufen-f <[email protected]> Co-authored-by: sufen <[email protected]> Co-authored-by: Niklas Muennighoff <[email protected]> Co-authored-by: Samuel Yang <[email protected]> Co-authored-by: Aradhye Agarwal <[email protected]> Co-authored-by: Tom Aarsen <[email protected]> Co-authored-by: talshef <[email protected]> Co-authored-by: Tal Sheffer <[email protected]> Co-authored-by: garciasces <[email protected]> Co-authored-by: gowitheflow-1998 <[email protected]> Co-authored-by: Wang Bo <[email protected]> Co-authored-by: Munot Ayush Sunil <[email protected]> Co-authored-by: Yaya Sy <[email protected]> Co-authored-by: Imene Kerboua <[email protected]> Co-authored-by: Eng. Omar Najar <[email protected]> Co-authored-by: Michael Dinzinger <[email protected]> Co-authored-by: Jinhyuk Lee <[email protected]> Co-authored-by: Isaac Chung <[email protected]> Co-authored-by: sergeyz-zh <[email protected]> Co-authored-by: OnandOn <[email protected]> Co-authored-by: chenghao xiao <[email protected]> Co-authored-by: Shreyas Subramanian <[email protected]> Co-authored-by: Uri K <[email protected]> Co-authored-by: richinfo-ai <[email protected]> Co-authored-by: Adewole Babatunde <[email protected]> Co-authored-by: ahxgw <[email protected]> Co-authored-by: kunka.xgw <[email protected]> Co-authored-by: Nadia Sheikh <[email protected]> Co-authored-by: theatollersrud <[email protected]> Co-authored-by: hongst <[email protected]>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Code Quality
make lint
to maintain consistent style.Documentation
Testing
make test-with-coverage
.make test
ormake test-with-coverage
to ensure no existing functionality is broken.Adding datasets checklist
Reason for dataset addition: ...
mteb -m {model_name} -t {task_name}
command.sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2
intfloat/multilingual-e5-small
self.stratified_subsampling() under dataset_transform()
make test
.make lint
.Adding a model checklist
mteb.get_model(model_name, revision)
andmteb.get_model_meta(model_name, revision)