Skip to content

Fix Task Lang Table #2487

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.

Sign up for GitHub

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
Samoed merged 3 commits into from
Apr 4, 2025
Merged

Fix Task Lang Table #2487

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
76b12a5
Fix Task Lang Table
ayush1298 Apr 4, 2025
c5b921b
added tasks.md
ayush1298 Apr 4, 2025
ed04858
fix
ayush1298 Apr 4, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
21 changes: 13 additions & 8 deletions docs/create_tasks_table.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -106,13 +106,14 @@ def create_task_lang_table(tasks: list[mteb.AbsTask], sort_by_sum=False) -> str:
df = df.sort(by="sum", descending=True)

total = df.sum()

task_names_md = " | ".join(sorted(get_args(TASK_TYPE)))
horizontal_line_md = "---|---" * (len(sorted(get_args(TASK_TYPE))) + 1)
table = f"""
| ISO Code | Language | Family | {task_names_md} | Sum |
|{horizontal_line_md}|
"""
task_names = sorted(get_args(TASK_TYPE))
headers = ["ISO Code", "Language", "Family"] + task_names + ["Sum"]
table_header = "| " + " | ".join(headers) + " |"
separator_line = "|"
for header in headers:
width = len(header) + 2
separator_line += "-" * width + "|"
table = table_header + "\n" + separator_line + "\n"

for row in df.iter_rows():
table += f"| {row[0]} "
Expand All @@ -138,7 +139,11 @@ def insert_tables(
for table, tag in zip(tables, tags):
start = f"<!-- {tag} START -->"
end = f"<!-- {tag} END -->"
md = md.replace(md[md.index(start) + len(start) : md.index(end)], table)
# Ensure a newline after the start tag
md = md.replace(
md[md.index(start) + len(start) : md.index(end)],
f"\n{table}\n",
)

Path(file_path).write_text(md, encoding="utf-8")

Expand Down
5 changes: 4 additions & 1 deletion docs/tasks.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -5,6 +5,7 @@ The following tables give you an overview of the tasks in MTEB.

<!-- This allows the table to be autogenerated in the future: -->
<!-- TASKS TABLE START -->

| Name | Languages | Type | Category | Domains | # Samples | Dataset statistics |
|------|-----------|------|----------|---------|-----------|--------------------|
| [AFQMC](https://aclanthology.org/2021.emnlp-main.357) | ['cmn'] | STS | s2s | | None | None |
Expand Down Expand Up @@ -893,6 +894,7 @@ The following tables give you an overview of the tasks in MTEB.
| [mFollowIRCrossLingualInstructionRetrieval](https://neuclir.github.io/) (Weller et al., 2024) | ['eng', 'fas', 'rus', 'zho'] | Retrieval | s2p | [News, Written] | {'test': 121758} | {'test': {'num_samples': 121758, 'num_docs': 121635, 'num_queries': 123, 'number_of_characters': 283654099, 'min_document_length': 74, 'average_document_length': 2331.08, 'max_document_length': 24179, 'unique_docs': 121635, 'min_query_length': 32, 'average_query_length': 81.88, 'max_query_length': 173, 'unique_queries': 75, 'min_instruction_length': 93, 'average_instruction_length': 389.95, 'max_instruction_length': 887, 'unique_instructions': 75, 'min_changed_instruction_length': 180, 'average_changed_instruction_length': 450.55, 'max_changed_instruction_length': 974, 'unique_changed_instructions': 123, 'min_average_relevant_docs_per_query': 0, 'average_relevant_docs_per_query': 10.43, 'max_average_relevant_docs_per_query': 24, 'min_average_top_ranked_per_query': 1000, 'average_top_ranked_per_query': 1000.0, 'max_average_top_ranked_per_query': 1000, 'hf_subset_descriptive_stats': {'eng-fas': {'num_samples': 41229, 'num_docs': 41189, 'num_queries': 40, 'number_of_characters': 129597567, 'min_document_length': 99, 'average_document_length': 3145.5, 'max_document_length': 24179, 'unique_docs': 41189, 'min_query_length': 34, 'average_query_length': 80.08, 'max_query_length': 124, 'unique_queries': 40, 'min_instruction_length': 150, 'average_instruction_length': 396.88, 'max_instruction_length': 887, 'unique_instructions': 40, 'min_changed_instruction_length': 205, 'average_changed_instruction_length': 463.18, 'max_changed_instruction_length': 974, 'unique_changed_instructions': 40, 'min_average_relevant_docs_per_query': 1, 'average_relevant_docs_per_query': 10.85, 'max_average_relevant_docs_per_query': 22, 'min_average_top_ranked_per_query': 1000, 'average_top_ranked_per_query': 1000.0, 'max_average_top_ranked_per_query': 1000}, 'eng-rus': {'num_samples': 39366, 'num_docs': 39326, 'num_queries': 40, 'number_of_characters': 109522175, 'min_document_length': 75, 'average_document_length': 2784.08, 'max_document_length': 24061, 'unique_docs': 39326, 'min_query_length': 32, 'average_query_length': 81.88, 'max_query_length': 173, 'unique_queries': 40, 'min_instruction_length': 93, 'average_instruction_length': 371.12, 'max_instruction_length': 887, 'unique_instructions': 40, 'min_changed_instruction_length': 180, 'average_changed_instruction_length': 431.8, 'max_changed_instruction_length': 957, 'unique_changed_instructions': 40, 'min_average_relevant_docs_per_query': 0, 'average_relevant_docs_per_query': 9.78, 'max_average_relevant_docs_per_query': 24, 'min_average_top_ranked_per_query': 1000, 'average_top_ranked_per_query': 1000.0, 'max_average_top_ranked_per_query': 1000}, 'eng-zho': {'num_samples': 41163, 'num_docs': 41120, 'num_queries': 43, 'number_of_characters': 44534357, 'min_document_length': 74, 'average_document_length': 1082.05, 'max_document_length': 23840, 'unique_docs': 41120, 'min_query_length': 32, 'average_query_length': 83.56, 'max_query_length': 159, 'unique_queries': 43, 'min_instruction_length': 157, 'average_instruction_length': 401.02, 'max_instruction_length': 731, 'unique_instructions': 43, 'min_changed_instruction_length': 209, 'average_changed_instruction_length': 456.26, 'max_changed_instruction_length': 822, 'unique_changed_instructions': 43, 'min_average_relevant_docs_per_query': 1, 'average_relevant_docs_per_query': 10.65, 'max_average_relevant_docs_per_query': 24, 'min_average_top_ranked_per_query': 1000, 'average_top_ranked_per_query': 1000.0, 'max_average_top_ranked_per_query': 1000}}}} |
| [mFollowIRInstructionRetrieval](https://neuclir.github.io/) (Weller et al., 2024) | ['fas', 'rus', 'zho'] | Retrieval | s2p | [News, Written] | {'test': 121758} | {'test': {'num_samples': 121758, 'num_docs': 121635, 'num_queries': 123, 'number_of_characters': 283622456, 'min_document_length': 74, 'average_document_length': 2331.08, 'max_document_length': 24179, 'unique_docs': 121635, 'min_query_length': 10, 'average_query_length': 57.11, 'max_query_length': 136, 'unique_queries': 123, 'min_instruction_length': 37, 'average_instruction_length': 281.07, 'max_instruction_length': 1009, 'unique_instructions': 123, 'min_changed_instruction_length': 44, 'average_changed_instruction_length': 326.94, 'max_changed_instruction_length': 1083, 'unique_changed_instructions': 123, 'min_average_relevant_docs_per_query': 0, 'average_relevant_docs_per_query': 10.43, 'max_average_relevant_docs_per_query': 24, 'min_average_top_ranked_per_query': 1000, 'average_top_ranked_per_query': 1000.0, 'max_average_top_ranked_per_query': 1000, 'hf_subset_descriptive_stats': {'fas': {'num_samples': 41229, 'num_docs': 41189, 'num_queries': 40, 'number_of_characters': 129593838, 'min_document_length': 99, 'average_document_length': 3145.5, 'max_document_length': 24179, 'unique_docs': 41189, 'min_query_length': 34, 'average_query_length': 72.65, 'max_query_length': 124, 'unique_queries': 40, 'min_instruction_length': 121, 'average_instruction_length': 358.93, 'max_instruction_length': 759, 'unique_instructions': 40, 'min_changed_instruction_length': 163, 'average_changed_instruction_length': 415.32, 'max_changed_instruction_length': 842, 'unique_changed_instructions': 40, 'min_average_relevant_docs_per_query': 1, 'average_relevant_docs_per_query': 10.85, 'max_average_relevant_docs_per_query': 22, 'min_average_top_ranked_per_query': 1000, 'average_top_ranked_per_query': 1000.0, 'max_average_top_ranked_per_query': 1000}, 'rus': {'num_samples': 39366, 'num_docs': 39326, 'num_queries': 40, 'number_of_characters': 109523683, 'min_document_length': 75, 'average_document_length': 2784.08, 'max_document_length': 24061, 'unique_docs': 39326, 'min_query_length': 26, 'average_query_length': 77.5, 'max_query_length': 136, 'unique_queries': 40, 'min_instruction_length': 78, 'average_instruction_length': 387.0, 'max_instruction_length': 1009, 'unique_instructions': 40, 'min_changed_instruction_length': 187, 'average_changed_instruction_length': 458.0, 'max_changed_instruction_length': 1083, 'unique_changed_instructions': 40, 'min_average_relevant_docs_per_query': 0, 'average_relevant_docs_per_query': 9.78, 'max_average_relevant_docs_per_query': 24, 'min_average_top_ranked_per_query': 1000, 'average_top_ranked_per_query': 1000.0, 'max_average_top_ranked_per_query': 1000}, 'zho': {'num_samples': 41163, 'num_docs': 41120, 'num_queries': 43, 'number_of_characters': 44504935, 'min_document_length': 74, 'average_document_length': 1082.05, 'max_document_length': 23840, 'unique_docs': 41120, 'min_query_length': 10, 'average_query_length': 23.7, 'max_query_length': 44, 'unique_queries': 43, 'min_instruction_length': 37, 'average_instruction_length': 110.09, 'max_instruction_length': 209, 'unique_instructions': 43, 'min_changed_instruction_length': 44, 'average_changed_instruction_length': 122.81, 'max_changed_instruction_length': 229, 'unique_changed_instructions': 43, 'min_average_relevant_docs_per_query': 1, 'average_relevant_docs_per_query': 10.65, 'max_average_relevant_docs_per_query': 24, 'min_average_top_ranked_per_query': 1000, 'average_top_ranked_per_query': 1000.0, 'max_average_top_ranked_per_query': 1000}}}} |
| [mMARCO-NL](https://github.com/unicamp-dl/mMARCO) (Luiz Bonifacio and Israel Campiotti and Roberto de Alencar Lotufo and Rodrigo Frassetto Nogueira, 2021) | ['nld'] | Retrieval | s2p | [Web, Written] | None | None |

<!-- TASKS TABLE END -->

</details>
Expand All @@ -905,7 +907,7 @@ The following tables give you an overview of the tasks in MTEB.

<!-- TASK LANG TABLE START -->
| ISO Code | Language | Family | Any2AnyMultiChoice | Any2AnyMultilingualRetrieval | Any2AnyRetrieval | BitextMining | Classification | Clustering | Compositionality | DocumentUnderstanding | ImageClassification | ImageClustering | ImageMultilabelClassification | InstructionRetrieval | MultilabelClassification | PairClassification | Reranking | Retrieval | STS | Speed | Summarization | VisionCentricQA | VisualSTS(eng) | VisualSTS(multi) | ZeroShotClassification | Sum |
|---|------|------|------|------|------|------|------|------|------|------|------|------|------|------|------|------|------|------|------|------|------|------|------|---|
|----------|----------|--------|--------------------|------------------------------|------------------|--------------|----------------|------------|------------------|-----------------------|---------------------|-----------------|-------------------------------|----------------------|--------------------------|--------------------|-----------|-----------|-----|-------|---------------|-----------------|----------------|------------------|------------------------|-----|
| aai | Arifama-Miniafia | Austronesian | 0 | 0 | 0 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 |
| aak | Ankave | Angan | 0 | 0 | 0 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 |
| aau | Abau | Sepik | 0 | 0 | 0 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 |
Expand Down Expand Up @@ -1959,6 +1961,7 @@ The following tables give you an overview of the tasks in MTEB.
| zul | Zulu | Atlantic-Congo | 0 | 0 | 0 | 2 | 3 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 7 |
| zyp | Zyphe Chin | Sino-Tibetan | 0 | 0 | 0 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 |
| Total | None | None | None | 0 | 55 | 49 | 1492 | 836 | 316 | 7 | 10 | 22 | 5 | 0 | 3 | 28 | 91 | 56 | 591 | 88 | 2 | 2 | 6 | 7 | 37 | 24 |

<!-- TASK LANG TABLE END -->

</details>