text-classification bool 2 classes | text stringlengths 0 664k |
|---|---|
false | # Dataset Card for "RSSCN7"
## Dataset Description
- **Paper** [Deep Learning Based Feature Selection for Remote Sensing Scene Classification](https://ieeexplore.ieee.org/iel7/8859/7305891/07272047.pdf)
### Licensing Information
For research and academic purposes.
## Citation Information
[Deep Learning Based Feature Selection for Remote Sensing Scene Classification](https://ieeexplore.ieee.org/iel7/8859/7305891/07272047.pdf)
```
@article{7272047,
title = {Deep Learning Based Feature Selection for Remote Sensing Scene Classification},
author = {Zou, Qin and Ni, Lihao and Zhang, Tong and Wang, Qian},
year = 2015,
journal = {IEEE Geoscience and Remote Sensing Letters},
volume = 12,
number = 11,
pages = {2321--2325},
doi = {10.1109/LGRS.2015.2475299}
}
``` |
false |
# MIRACL (ru) embedded with cohere.ai `multilingual-22-12` encoder
We encoded the [MIRACL dataset](https://huggingface.co/miracl) using the [cohere.ai](https://txt.cohere.ai/multilingual/) `multilingual-22-12` embedding model.
The query embeddings can be found in [Cohere/miracl-ru-queries-22-12](https://huggingface.co/datasets/Cohere/miracl-ru-queries-22-12) and the corpus embeddings can be found in [Cohere/miracl-ru-corpus-22-12](https://huggingface.co/datasets/Cohere/miracl-ru-corpus-22-12).
For the orginal datasets, see [miracl/miracl](https://huggingface.co/datasets/miracl/miracl) and [miracl/miracl-corpus](https://huggingface.co/datasets/miracl/miracl-corpus).
Dataset info:
> MIRACL 🌍🙌🌏 (Multilingual Information Retrieval Across a Continuum of Languages) is a multilingual retrieval dataset that focuses on search across 18 different languages, which collectively encompass over three billion native speakers around the world.
>
> The corpus for each language is prepared from a Wikipedia dump, where we keep only the plain text and discard images, tables, etc. Each article is segmented into multiple passages using WikiExtractor based on natural discourse units (e.g., `\n\n` in the wiki markup). Each of these passages comprises a "document" or unit of retrieval. We preserve the Wikipedia article title of each passage.
## Embeddings
We compute for `title+" "+text` the embeddings using our `multilingual-22-12` embedding model, a state-of-the-art model that works for semantic search in 100 languages. If you want to learn more about this model, have a look at [cohere.ai multilingual embedding model](https://txt.cohere.ai/multilingual/).
## Loading the dataset
In [miracl-ru-corpus-22-12](https://huggingface.co/datasets/Cohere/miracl-ru-corpus-22-12) we provide the corpus embeddings. Note, depending on the selected split, the respective files can be quite large.
You can either load the dataset like this:
```python
from datasets import load_dataset
docs = load_dataset(f"Cohere/miracl-ru-corpus-22-12", split="train")
```
Or you can also stream it without downloading it before:
```python
from datasets import load_dataset
docs = load_dataset(f"Cohere/miracl-ru-corpus-22-12", split="train", streaming=True)
for doc in docs:
docid = doc['docid']
title = doc['title']
text = doc['text']
emb = doc['emb']
```
## Search
Have a look at [miracl-ru-queries-22-12](https://huggingface.co/datasets/Cohere/miracl-ru-queries-22-12) where we provide the query embeddings for the MIRACL dataset.
To search in the documents, you must use **dot-product**.
And then compare this query embeddings either with a vector database (recommended) or directly computing the dot product.
A full search example:
```python
# Attention! For large datasets, this requires a lot of memory to store
# all document embeddings and to compute the dot product scores.
# Only use this for smaller datasets. For large datasets, use a vector DB
from datasets import load_dataset
import torch
#Load documents + embeddings
docs = load_dataset(f"Cohere/miracl-ru-corpus-22-12", split="train")
doc_embeddings = torch.tensor(docs['emb'])
# Load queries
queries = load_dataset(f"Cohere/miracl-ru-queries-22-12", split="dev")
# Select the first query as example
qid = 0
query = queries[qid]
query_embedding = torch.tensor(queries['emb'])
# Compute dot score between query embedding and document embeddings
dot_scores = torch.mm(query_embedding, doc_embeddings.transpose(0, 1))
top_k = torch.topk(dot_scores, k=3)
# Print results
print("Query:", query['query'])
for doc_id in top_k.indices[0].tolist():
print(docs[doc_id]['title'])
print(docs[doc_id]['text'])
```
You can get embeddings for new queries using our API:
```python
#Run: pip install cohere
import cohere
co = cohere.Client(f"{api_key}") # You should add your cohere API Key here :))
texts = ['my search query']
response = co.embed(texts=texts, model='multilingual-22-12')
query_embedding = response.embeddings[0] # Get the embedding for the first text
```
## Performance
In the following table we compare the cohere multilingual-22-12 model with Elasticsearch version 8.6.0 lexical search (title and passage indexed as independent fields). Note that Elasticsearch doesn't support all languages that are part of the MIRACL dataset.
We compute nDCG@10 (a ranking based loss), as well as hit@3: Is at least one relevant document in the top-3 results. We find that hit@3 is easier to interpret, as it presents the number of queries for which a relevant document is found among the top-3 results.
Note: MIRACL only annotated a small fraction of passages (10 per query) for relevancy. Especially for larger Wikipedias (like English), we often found many more relevant passages. This is know as annotation holes. Real nDCG@10 and hit@3 performance is likely higher than depicted.
| Model | cohere multilingual-22-12 nDCG@10 | cohere multilingual-22-12 hit@3 | ES 8.6.0 nDCG@10 | ES 8.6.0 acc@3 |
|---|---|---|---|---|
| miracl-ar | 64.2 | 75.2 | 46.8 | 56.2 |
| miracl-bn | 61.5 | 75.7 | 49.2 | 60.1 |
| miracl-de | 44.4 | 60.7 | 19.6 | 29.8 |
| miracl-en | 44.6 | 62.2 | 30.2 | 43.2 |
| miracl-es | 47.0 | 74.1 | 27.0 | 47.2 |
| miracl-fi | 63.7 | 76.2 | 51.4 | 61.6 |
| miracl-fr | 46.8 | 57.1 | 17.0 | 21.6 |
| miracl-hi | 50.7 | 62.9 | 41.0 | 48.9 |
| miracl-id | 44.8 | 63.8 | 39.2 | 54.7 |
| miracl-ru | 49.2 | 66.9 | 25.4 | 36.7 |
| **Avg** | 51.7 | 67.5 | 34.7 | 46.0 |
Further languages (not supported by Elasticsearch):
| Model | cohere multilingual-22-12 nDCG@10 | cohere multilingual-22-12 hit@3 |
|---|---|---|
| miracl-fa | 44.8 | 53.6 |
| miracl-ja | 49.0 | 61.0 |
| miracl-ko | 50.9 | 64.8 |
| miracl-sw | 61.4 | 74.5 |
| miracl-te | 67.8 | 72.3 |
| miracl-th | 60.2 | 71.9 |
| miracl-yo | 56.4 | 62.2 |
| miracl-zh | 43.8 | 56.5 |
| **Avg** | 54.3 | 64.6 |
|
false | # Dataset Card for "cqadupstack"
## Table of Contents
- [Table of Contents](#table-of-contents)
- [Dataset Description](#dataset-description)
- [Dataset Summary](#dataset-summary)
- [Dataset Structure](#dataset-structure)
- [Data Instances](#data-instances)
- [Data Fields](#data-fields)
- [Additional Information](#additional-information)
- [Licensing Information](#licensing-information)
## Dataset Description
- **Homepage:** [http://nlp.cis.unimelb.edu.au/resources/cqadupstack/](http://nlp.cis.unimelb.edu.au/resources/cqadupstack/)
### Dataset Summary
This is a preprocessed version of cqadupstack, to make it easily consumable via huggingface. The original dataset can be found [here](http://nlp.cis.unimelb.edu.au/resources/cqadupstack/).
CQADupStack is a benchmark dataset for community question-answering (cQA) research. It contains threads from twelve StackExchange1 subforums, annotated with duplicate question information and comes with pre-defined training, development, and test splits, both for retrieval and classification experiments.
## Dataset Structure
### Data Instances
An example of 'train' looks as follows.
```json
{
"question": "Very often, when some unknown company is calling me, in couple of seconds I see its name and logo on standard ...",
"answer": "You didn't explicitely mention it, but from the context I assume you're using a device with Android 4.4 (Kitkat). With that ...",
"title": "Why Dialer shows contact name and image, when contact is not in my address book?",
"forum_tag": "android"
}
```
### Data Fields
The data fields are the same among all splits.
- `question`: a `string` feature.
- `answer`: a `string` feature.
- `title`: a `string` feature.
- `forum_tag`: a categorical `string` feature.
## Additional Information
### Licensing Information
This dataset is distributed under the Apache 2.0 licence.
|
true | |
false |
The full dataset card is visible in the JSON file named "original_cacapo_for_e2e_models-02_13_2023_19_30_07", which has been made with GEMs second datacard creation GUI. |
false | # Dataset Card for "RS_C11"
## Dataset Description
- **Paper** [Feature significance-based multibag-of-visual-words model for remote sensing image scene classification](https://www.spiedigitallibrary.org/journals/journal-of-applied-remote-sensing/volume-10/issue-3/035004/Feature-significance-based-multibag-of-visual-words-model-for-remote/10.1117/1.JRS.10.035004.pdf)
### Licensing Information
Free usage without license.
## Citation Information
[Feature significance-based multibag-of-visual-words model for remote sensing image scene classification](https://www.spiedigitallibrary.org/journals/journal-of-applied-remote-sensing/volume-10/issue-3/035004/Feature-significance-based-multibag-of-visual-words-model-for-remote/10.1117/1.JRS.10.035004.pdf)
```
@article{zhao2016feature,
title = {Feature significance-based multibag-of-visual-words model for remote sensing image scene classification},
author = {Zhao, Lijun and Tang, Ping and Huo, Lianzhi},
year = 2016,
journal = {Journal of Applied Remote Sensing},
publisher = {Society of Photo-Optical Instrumentation Engineers},
volume = 10,
number = 3,
pages = {035004--035004}
}
``` |
false | # Dataset Card for "Optimal-31"
## Dataset Description
- **Paper** [Scene classification with recurrent attention of VHR remote sensing images](https://ieeexplore.ieee.org/iel7/5/8045830/07891544.pdf)
### Licensing Information
[No license for now, cite the paper below.]
## Citation Information
[Scene classification with recurrent attention of VHR remote sensing images](https://ieeexplore.ieee.org/iel7/5/8045830/07891544.pdf)
```
@article{wang2018scene,
title = {Scene classification with recurrent attention of VHR remote sensing images},
author = {Wang, Qi and Liu, Shaoteng and Chanussot, Jocelyn and Li, Xuelong},
year = 2018,
journal = {IEEE Transactions on Geoscience and Remote Sensing},
publisher = {IEEE},
volume = 57,
number = 2,
pages = {1155--1167}
}
``` |
false |
# Alpaca GPT4 English-to-Italian Translated Instructions (WIP)
This dataset contains **15209** instructions that have been translated from English to Italian using `gpt-3.5-turbo`.
Alpaca GPT4: the original **alpaca_gpt4_data.json** dataset contains 52K instruction-following data generated by GPT-4 with prompts in Alpaca. The JSON file has the same format as Alpaca data, except the output is generated by GPT-4:
- instruction: str, describes the task the model should perform. Each of the 52K instructions is unique.
- input: str, optional context or input for the task.
- output: str, the answer to the instruction as generated by GPT-4.
The instructions were sourced from the Alpaca GPT4 dataset and translated using the following prompt:
```
Act as an unrivaled English-to-Italian Translator. Your task is to translate the given passage into Italian, as you are a native Italian speaker. Each message passage contains an instruction, with an optional input (preceded by [|IN|]) and an output ([|OUT|]). You MUST provide accurate and fluent Italian translations. When translating the instruction use the second person singular. Translate each section. Keep [|IN|] and [|OUT|] placeholders. If the input or output doesn't make sense in Italian, revise them.
ENGLISH:
"""
{passage}
"""
ITALIAN:
```
where 'passage' represents the original English text. The inputs are structured in a format that allows `gpt-3.5-turbo` to understand the context. They are translated into Italian while preserving the context of the original English instructions, and formatted in the following way: : `<instruction> (optional [|IN|] <input>) [|OUT|] <output>`.
## License
Please note that the original Alpaca GPT4 dataset and the translations generated by `gpt-3.5-turbo` may have their respective licenses, and it is important to comply with any usage restrictions specified by the original data sources. As this dataset contains partially translated data, proper attribution and compliance with relevant licenses is recommended.
The data is intended and licensed for research use only. The dataset is CC BY NC 4.0 (allowing only non-commercial use) and models trained using the dataset should not be used outside of research purposes.
## Citation
```
@article{peng2023instruction,
title={Instruction Tuning with GPT-4},
author={Peng, Baolin and Li, Chunyuan and He, Pengcheng and Galley, Michel and Gao, Jianfeng},
journal={arXiv preprint arXiv:2304.03277},
year={2023}
}
``` |
false |
riffusion manipulated google/MusicCaps |
false | # Summary
This is a Thai 🇹🇭-instructed dataset translated from cleaned version of the original Alpaca Dataset released by Stanford using Google Cloud Translation, contain 52,000 instructions and demonstrations generated by OpenAI's `text-davinci-003` engine.
This instruction data can be used to conduct instruction-tuning for language models and make the language model follow instruction better.
The following issues have been identified in the original release and fixed in this dataset:
1. **Hallucinations:** Many instructions in the original dataset had instructions referencing data on the internet, which just caused GPT3 to hallucinate an answer.
2. **Merged Instructions:** There were many instructions that were merged together in the original dataset for some reason.
3. **Empty outputs:** Some entries in the original dataset had empty outputs.
4. **Empty code examples:** Some descriptions in the original dataset were missing code examples, making it difficult to understand the intended behavior of the code.
5. **Instructions to generate images:** Some descriptions in the original dataset included instructions to generate images, something obviously not possible.
6. **N/A outputs:** Some code snippets in the original dataset had N/A outputs.
7. **Inconsistent input field:** The original dataset had inconsistent usage of the input field when it was supposed to be empty.
8. **Wrong answers:** Some instructions/questions in the original dataset had incorrect answers. About 80% of the math problems are estimated to have incorrect answers.
9. **Non-Sensical/Unclear instructions:** Many instructions are unclear, we try to clarify (or re-write) if instructions are non-sensical. Instructions that are slightly unclear, but where one could deduce the meaning are not altered.
10. **Extraneous escape and control characters:** The original dataset had several entries with extraneous escape and control characters.
### Original Alpaca Dataset Summary
Alpaca is a dataset of 52,000 instructions and demonstrations generated by OpenAI's `text-davinci-003` engine. This instruction data can be used to conduct instruction-tuning for language models and make the language model follow instruction better.
The authors built on the data generation pipeline from [Self-Instruct framework](https://github.com/yizhongw/self-instruct) and made the following modifications:
- The `text-davinci-003` engine to generate the instruction data instead of `davinci`.
- A [new prompt](https://github.com/tatsu-lab/stanford_alpaca/blob/main/prompt.txt) was written that explicitly gave the requirement of instruction generation to `text-davinci-003`.
- Much more aggressive batch decoding was used, i.e., generating 20 instructions at once, which significantly reduced the cost of data generation.
- The data generation pipeline was simplified by discarding the difference between classification and non-classification instructions.
- Only a single instance was generated for each instruction, instead of 2 to 3 instances as in Self-Instruct.
The authors built on the data generation pipeline from [Self-Instruct framework](https://github.com/yizhongw/self-instruct) and made the following modifications:
- The `text-davinci-003` engine to generate the instruction data instead of `davinci`.
- A [new prompt](https://github.com/tatsu-lab/stanford_alpaca/blob/main/prompt.txt) was written that explicitly gave the requirement of instruction generation to `text-davinci-003`.
- Much more aggressive batch decoding was used, i.e., generating 20 instructions at once, which significantly reduced the cost of data generation.
- The data generation pipeline was simplified by discarding the difference between classification and non-classification instructions.
- Only a single instance was generated for each instruction, instead of 2 to 3 instances as in Self-Instruct.
This produced an instruction-following dataset with 52K examples obtained at a much lower cost (less than $500).
In a preliminary study, the authors also found that the 52K generated data to be much more diverse than the data released by [Self-Instruct](https://github.com/yizhongw/self-instruct/blob/main/data/seed_tasks.jsonl).
Supported Tasks:
- Training LLMs
- Synthetic Data Generation
- Data Augmentation
Languages: Thai
Version: 1.0
--- |
false |
# IVA Swift GitHub Code Dataset
## Dataset Description
This is the curated IVA Swift dataset extracted from GitHub.
It contains curated Swift files gathered with the purpose to train a code generation model.
The dataset consists of 383380 swift code files from GitHub totaling ~542MB of data.
The [uncurated](https://huggingface.co/datasets/mvasiliniuc/iva-swift-codeint) dataset was created from the public GitHub dataset on Google BiqQuery.
### How to use it
To download the full dataset:
```python
from datasets import load_dataset
dataset = load_dataset('mvasiliniuc/iva-swift-codeint-clean', split='train')
```
```python
from datasets import load_dataset
dataset = load_dataset('mvasiliniuc/iva-swift-codeint-clean', split='train')
print(dataset[723])
#OUTPUT:
{
"repo_name":"jdkelley/Udacity-OnTheMap-ExampleApps",
"path":"TheMovieManager-v2/TheMovieManager/BorderedButton.swift",
"copies":"2",
"size":"2649",
"content":"...let phoneBorderedButtonExtraPadding: CGFloat = 14.0\n \n var backingColor: UIColor? = nil\n var highlightedBackingColor: UIColor? = nil\n \n // MARK: Initialization\n}",
"license":"mit",
"hash":"db1587fd117e9a835f58cf8203d8bf05",
"line_mean":29.1136363636,
"line_max":87,
"alpha_frac":0.6700641752,
"ratio":5.298,
"autogenerated":false,
"config_or_test":false,
"has_no_keywords":false,
"has_few_assignments":false
}
```
## Data Structure
### Data Fields
|Field|Type|Description|
|---|---|---|
|repo_name|string|name of the GitHub repository|
|path|string|path of the file in GitHub repository|
|copies|string|number of occurrences in dataset|
|content|string|content of source file|
|size|string|size of the source file in bytes|
|license|string|license of GitHub repository|
|hash|string|Hash of content field.|
|line_mean|number|Mean line length of the content.
|line_max|number|Max line length of the content.
|alpha_frac|number|Fraction between mean and max line length of content.
|ratio|number|Character/token ratio of the file with tokenizer.
|autogenerated|boolean|True if the content is autogenerated by looking for keywords in the first few lines of the file.
|config_or_test|boolean|True if the content is a configuration file or a unit test.
|has_no_keywords|boolean|True if a file has none of the keywords for Swift Programming Language.
|has_few_assignments|boolean|True if file uses symbol '=' less than `minimum` times.
### Instance
```json
{
"repo_name":"...",
"path":".../BorderedButton.swift",
"copies":"2",
"size":"2649",
"content":"...",
"license":"mit",
"hash":"db1587fd117e9a835f58cf8203d8bf05",
"line_mean":29.1136363636,
"line_max":87,
"alpha_frac":0.6700641752,
"ratio":5.298,
"autogenerated":false,
"config_or_test":false,
"has_no_keywords":false,
"has_few_assignments":false
}
```
## Languages
The dataset contains only Swift files.
```json
{
"Swift": [".swift"]
}
```
## Licenses
Each entry in the dataset contains the associated license. The following is a list of licenses involved and their occurrences.
```json
{
"agpl-3.0":1695,
"apache-2.0":85514,
"artistic-2.0":207,
"bsd-2-clause":3132,
"bsd-3-clause":6600,
"cc0-1.0":1409,
"epl-1.0":605,
"gpl-2.0":9374,
"gpl-3.0":18920,
"isc":808,
"lgpl-2.1":1122,
"lgpl-3.0":3103,
"mit":240929,
"mpl-2.0":8181,
"unlicense":1781
}
```
## Dataset Statistics
```json
{
"Total size": "~542 MB",
"Number of files": 383380,
"Number of files under 500 bytes": 3680,
"Average file size in bytes": 5942,
}
```
## Curation Process
* Removal of duplication files based on file hash.
* Removal of file templates. File containing the following: `___FILENAME___, ___PACKAGENAME___, ___FILEBASENAME___, ___FILEHEADER___, ___VARIABLE`
* Removal of the files containing the following words in the first 10 lines: `generated, auto-generated", "autogenerated", "automatically generated`
* Removal of the files containing the following words in the first 10 lines with a probability of 0.7: `test", "unit test", "config", "XCTest", "JUnit`
* Removal of file with the rate of alphanumeric characters below 0.3 of the file.
* Removal of near duplication based MinHash and Jaccard similarity.
* Removal of files with mean line length above 100.
* Removal of files without mention of keywords with a probability of 0.7: `struct ", "class ", "for ", "while ", "enum ", "func ", "typealias ", "var ", "let ", "protocol ", "public ", "private ", "internal ", "import "`
* Removal of files that use the assignment operator `=` less than 3 times.
* Removal of files with the ratio between the number of characters and number of tokens after tokenization lower than 1.5.
Curation process is a derivation of the one used in CodeParrot project: https://huggingface.co/codeparrot
## Data Splits
The dataset only contains a train split which is separated into train and valid which can be found here:
* Clean Version Train: https://huggingface.co/datasets/mvasiliniuc/iva-swift-codeint-clean-train
* Clean Version Valid: https://huggingface.co/datasets/mvasiliniuc/iva-swift-codeint-clean-valid
# Considerations for Using the Data
The dataset comprises source code from various repositories, potentially containing harmful or biased code,
along with sensitive information such as passwords or usernames.
|
true | |
false | # Dataset Card for Dataset Name
## Dataset Description
- **Autor:** Rubén Darío Jaramillo
- **Email:** rubend18@hotmail.com
- **WhatsApp:** +593 93 979 6676
### Dataset Summary
This dataset has been generated using [Prompt Generator for OpenAI's DALL-E](https://huggingface.co/spaces/rubend18/Prompt-Generator-for-OpenAI-DALL-E).
### Languages
English
## Dataset Structure
1.000.000 Prompts |
true | # Dataset Card for Dataset Name
## Dataset Description
- **Repository:** https://github.com/msmadi/ABSA-Hotels/tree/master
### Dataset Summary
Aspect based sentiment analysis dataset using hotel reviews in Arabic.
### Languages
Arabic
### Licensing Information
Original dataset was licensed under MIT, so this is also under MIT
### Citation Information
Cite this and the original authors if you want to.
|
false | |
false |
# Dataset Card for RepoBench-C
## Dataset Description
- **Homepage:** https://github.com/Leolty/repobench
- **Paper:** https://arxiv.org/abs/2306.03091
## Dataset Summary
**RepoBench-C (Completion)** is a subtask of **RepoBench**([GitHub](https://github.com/Leolty/repobench), [arXiv](https://arxiv.org/abs/2306.03091)), focuing on the prediction of the next line of code, given in-file context (including several preceding lines and import statements), and cross-file context.
## Settings
- `cff`: short for cross_file_first, indicating the cross-file module in next line is first used in the current file.
- `cfr`: short for cross_file_random, indicating the cross-file module in next line is not first used in the current file.
- `if`: short for in_file, indicating the next line does not contain any cross-file module.
## Supported Tasks
- `python_cff`: python code prediction with cross-file-first setting.
- `python_cfr`: python code prediction with cross-file-random setting.
- `python_if`: python code prediction with in-file setting.
- `java_cff`: java code prediction with cross-file-first setting.
- `java_cfr`: java code prediction with cross-file-random setting.
- `java_if`: java code prediction with in-file setting.
## Loading Data
For example, if you want to load the `test` set to test your model on `Python` code prediction with `cff` setting, you can do the following:
```python
from datasets import load_dataset
dataset = load_dataset("tianyang/repobench-c", "python_cff", split="test")
```
> Note: The `split` argument is optional. If not provided, the entire dataset will be loaded.
## Dataset Structure
```json
{
"repo_name": "repository name of the data point",
"file_path": "path/to/file",
"context": "commented and concatenated cross-file context",
"import_statement": "all import statements in the file",
"code": "the code for next-line prediction",
"prompt": "cross-file context + import statements + in-file code",
"next_line": "the next line of the code"
}
```
## Licensing Information
CC BY-NC-ND 4.0
## Citation Information
```bibtex
@misc{liu2023repobench,
title={RepoBench: Benchmarking Repository-Level Code Auto-Completion Systems},
author={Tianyang Liu and Canwen Xu and Julian McAuley},
year={2023},
eprint={2306.03091},
archivePrefix={arXiv},
primaryClass={cs.CL}
}
```
## Contributions
Thanks to [@Leolty](https://github.com/Leolty) for adding this dataset. |
true | # rudetoxifier_data
## Dataset Description
- **Homepage:** https://github.com/s-nlp/rudetoxifier
- **Repository:** https://github.com/s-nlp/rudetoxifier
- **Paper:** [Methods for Detoxification of Texts for the Russian Language](https://arxiv.org/abs/2105.09052)
- **Point of Contact:** [Daryna Dementieva](mailto:daryna.dementieva@skoltech.ru)
Huggingface copy of Github repo with dataset. |
false |
# Dataset Card for "arxiv-cs-ml-instruct-tune-50k"
### Dataset Summary
The "ArtifactAI/arxiv-cs-ml-instruct-tune-50k" dataset consists of question-answer pairs derived from ArXiv abstracts from the following categories:
"cs.CL", "cs.AI", "cs.CV", "cs.HC", "cs.IR", "cs.RO", "cs.NE", "stat.ML". Questions are generated using the [t5-base model](https://huggingface.co/t5-base), while the answers are generated using the [GPT-3.5-turbo model](https://openai.com/chatgpt).
### Languages
English
## Dataset Structure
### Data Instances
#### train
- **Size of downloaded dataset files:** 38.4 MB
An example of 'train' looks as follows.
```
{
"question": "What is multitask learning?",
"answer": "Multitask learning is a technique in machine learning that allows a single model to simultaneously learn multiple related tasks.
Instead of training separate models for each task, multitask learning leverages shared patterns across tasks to improve overall performance.
This can lead to faster training, better accuracy, and more efficient use of computational resources.
For example, a model might be trained to simultaneously predict the sentiment of customer reviews and categorize them by topic.
Both tasks require analyzing natural language, so the shared knowledge can help improve performance on both tasks."
}
```
### Data Fields
The data fields present in the dataset are as follows:
- question: a string feature representing the question.
- answer: a string feature representing the answer.
#### plain_text
- `text`: a `string` feature.
### Data Splits
train: 50,063 question answer pairs
### Curation Rationale
[More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
### Source Data
Question-answer pairs derived from [ArXiv](https://arxiv.org/) abstracts.
#### Initial Data Collection and Normalization
The "ArtifactAI/arxiv-cs-ml-instruct-tune-50k" dataset consists of question-answer pairs derived from ArXiv abstracts. Questions are generated from ArXiv papers in the following categories:
- cs.CL
- cs.AI
- cs.CV
- cs.HC
- cs.IR
- cs.RO
- cs.NE
- stat.ML
Questions are generated using the [t5-base model](https://huggingface.co/t5-base), while the answers are generated using the [GPT-3.5-turbo model](https://openai.com/chatgpt).
### Annotations
The dataset doesn't contain annotations.
### Personal and Sensitive Information
None
#### Notice policy
Should you consider that our data contains material that is owned by you and should therefore not be reproduced here, please:
Clearly identify yourself, with detailed contact data such as an address, telephone number or email address at which you can be contacted.
Clearly identify the copyrighted work claimed to be infringed.
Clearly identify the material that is claimed to be infringing and information reasonably sufficient to allow us to locate the material.
And contact us at the following email address: matt at artifactai.com and datasets at huggingface.co
#### Take down policy
The original authors will comply to legitimate requests by removing the affected sources from the next release of the corpus.
Hugging Face will also update this repository accordingly.
### Citation Information
```
@misc{arxiv-cs-ml-instruct-tune-50k,
title={arxiv-cs-ml-instruct-tune-50k},
author={Matthew Kenney},
year={2023}
}
```
|
false | |
false |
Explain tuned Dolly-V2 dataset ~15K created using approaches from Orca Research Paper.
We leverage all of the 15 system instructions provided in Orca Research Paper to generate explain tuned datasets, in contrast to vanilla instruction tuning approaches used by original datasets.
This helps student models like orca_mini_13b, orca_mini_7b or orca_mini_3b to learn thought process from teacher model, which is ChatGPT (gpt-3.5-turbo-0301 version).
Please see how the System prompt is added before each instruction. |
false |
# CMMLU: Measuring massive multitask language understanding in Chinese
- **Homepage:** [https://github.com/haonan-li/CMMLU](https://github.com/haonan-li/CMMLU)
- **Repository:** [https://huggingface.co/datasets/haonan-li/cmmlu](https://huggingface.co/datasets/haonan-li/cmmlu)
- **Paper:** [CMMLU: Measuring Chinese Massive Multitask Language Understanding](https://arxiv.org/abs/2306.09212).
## Table of Contents
- [Introduction](#introduction)
- [Leaderboard](#leaderboard)
- [Data](#data)
- [Citation](#citation)
- [License](#license)
## Introduction
CMMLU is a comprehensive Chinese assessment suite specifically designed to evaluate the advanced knowledge and reasoning abilities of LLMs within the Chinese language and cultural context.
CMMLU covers a wide range of subjects, comprising 67 topics that span from elementary to advanced professional levels. It includes subjects that require computational expertise, such as physics and mathematics, as well as disciplines within humanities and social sciences.
Many of these tasks are not easily translatable from other languages due to their specific contextual nuances and wording.
Furthermore, numerous tasks within CMMLU have answers that are specific to China and may not be universally applicable or considered correct in other regions or languages.
## Leaderboard
Latest leaderboard is in our [github](https://github.com/haonan-li/CMMLU).
## Data
We provide development and test dataset for each of 67 subjects, with 5 questions in development set and 100+ quesitons in test set.
Each question in the dataset is a multiple-choice questions with 4 choices and only one choice as the correct answer.
Here are two examples:
```
题目：同一物种的两类细胞各产生一种分泌蛋白，组成这两种蛋白质的各种氨基酸含量相同，但排列顺序不同。其原因是参与这两种蛋白质合成的：
A. tRNA种类不同
B. 同一密码子所决定的氨基酸不同
C. mRNA碱基序列不同
D. 核糖体成分不同
答案是：C
```
```
题目：某种植物病毒V是通过稻飞虱吸食水稻汁液在水稻间传播的。稻田中青蛙数量的增加可减少该病毒在水稻间的传播。下列叙述正确的是：
A. 青蛙与稻飞虱是捕食关系
B. 水稻和病毒V是互利共生关系
C. 病毒V与青蛙是寄生关系
D. 水稻与青蛙是竞争关系
答案是：
```
#### Load data
```python
from datasets import load_dataset
cmmlu=load_dataset(r"haonan-li/cmmlu", 'agronomy')
print(cmmlu['test'][0])
```
#### Load all data at once
```python
task_list = ['agronomy', 'anatomy', 'ancient_chinese', 'arts', 'astronomy', 'business_ethics', 'chinese_civil_service_exam', 'chinese_driving_rule', 'chinese_food_culture', 'chinese_foreign_policy', 'chinese_history', 'chinese_literature',
'chinese_teacher_qualification', 'clinical_knowledge', 'college_actuarial_science', 'college_education', 'college_engineering_hydrology', 'college_law', 'college_mathematics', 'college_medical_statistics', 'college_medicine', 'computer_science',
'computer_security', 'conceptual_physics', 'construction_project_management', 'economics', 'education', 'electrical_engineering', 'elementary_chinese', 'elementary_commonsense', 'elementary_information_and_technology', 'elementary_mathematics',
'ethnology', 'food_science', 'genetics', 'global_facts', 'high_school_biology', 'high_school_chemistry', 'high_school_geography', 'high_school_mathematics', 'high_school_physics', 'high_school_politics', 'human_sexuality',
'international_law', 'journalism', 'jurisprudence', 'legal_and_moral_basis', 'logical', 'machine_learning', 'management', 'marketing', 'marxist_theory', 'modern_chinese', 'nutrition', 'philosophy', 'professional_accounting', 'professional_law',
'professional_medicine', 'professional_psychology', 'public_relations', 'security_study', 'sociology', 'sports_science', 'traditional_chinese_medicine', 'virology', 'world_history', 'world_religions']
from datasets import load_dataset
cmmlu = {k: load_dataset(r"haonan-li/cmmlu", k) for k in task_list}
```
## Citation
```
@misc{li2023cmmlu,
title={CMMLU: Measuring massive multitask language understanding in Chinese},
author={Haonan Li and Yixuan Zhang and Fajri Koto and Yifei Yang and Hai Zhao and Yeyun Gong and Nan Duan and Timothy Baldwin},
year={2023},
eprint={2306.09212},
archivePrefix={arXiv},
primaryClass={cs.CL}
}
```
## License
The CMMLU dataset is licensed under a
[Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License](http://creativecommons.org/licenses/by-nc-sa/4.0/).
|
false |
# Dataset Card for Tilde-MODEL-Catalan
## Table of Contents
- [Table of Contents](#table-of-contents)
- [Dataset Description](#dataset-description)
- [Dataset Summary](#dataset-summary)
- [Supported Tasks and Leaderboards](#supported-tasks-and-leaderboards)
- [Languages](#languages)
- [Dataset Structure](#dataset-structure)
- [Data Instances](#data-instances)
- [Data Fields](#data-fields)
- [Data Splits](#data-splits)
- [Dataset Creation](#dataset-creation)
- [Curation Rationale](#curation-rationale)
- [Source Data](#source-data)
- [Annotations](#annotations)
- [Personal and Sensitive Information](#personal-and-sensitive-information)
- [Considerations for Using the Data](#considerations-for-using-the-data)
- [Social Impact of Dataset](#social-impact-of-dataset)
- [Discussion of Biases](#discussion-of-biases)
- [Other Known Limitations](#other-known-limitations)
- [Additional Information](#additional-information)
- [Dataset Curators](#dataset-curators)
- [Licensing Information](#licensing-information)
- [Citation Information](#citation-information)
- [Contributions](#contributions)
## Dataset Description
- **Homepage:** https://www.softcatala.org/
- **Repository:** https://github.com/Softcatala/Europarl-catalan
- **Paper:**
- **Leaderboard:**
- **Point of Contact:**
### Dataset Summary
This dataset contains two dataset pairs corresponding to the Europarl corpus. Both the English and the German version are aligned with the Catalan translation, which has been obtained using Apertium's RBMT system from the Spanish version of the Spanish-English alignment. Catalan-German alignment has been obtained using this [alignment finder](https://github.com/davidcanovas/alignment-finder-with-pivot-language) from de-en and ca-en.
- Catalan-English: 1 965 735 segments.
- Catalan-German: 1 734 644 segments.
### Supported Tasks and Leaderboards
This dataset can be used to train NMT and SMT systems.
It has been used as a training corpus for the [Softcatalà machine translation engine](https://www.softcatala.org/traductor/).
### Languages
Catalan (`ca`).
German (`de`).
English (`en`).
## Dataset Structure
### Data Instances
[More Information Needed]
### Data Fields
Raw text.
### Data Splits
One file for language.
## Dataset Creation
### Curation Rationale
[More Information Needed]
### Source Data
#### Initial Data Collection and Normalization
[More Information Needed]
#### Who are the source language producers?
[More Information Needed]
### Annotations
#### Annotation process
[More Information Needed]
#### Who are the annotators?
[More Information Needed]
### Personal and Sensitive Information
[More Information Needed]
## Considerations for Using the Data
### Social Impact of Dataset
[More Information Needed]
### Discussion of Biases
[More Information Needed]
### Other Known Limitations
[More Information Needed]
## Additional Information
### Dataset Curators
[@softcatala](https://github.com/Softcatala)
[@jordimas](https://github.com/jordimas)
[@davidcanovas](https://github.com/davidcanovas)
### Licensing Information
[CC BY 4.0](https://creativecommons.org/licenses/by/4.0/).
### Citation Information
[More Information Needed]
### Contributions
[More Information Needed]
|
false |
# Dataset Card for vumichien/common_voice_large_jsut_jsss_css10
|
true | # AutoNLP Dataset for project: cat33
## Table of content
- [Dataset Description](#dataset-description)
- [Languages](#languages)
- [Dataset Structure](#dataset-structure)
- [Data Instances](#data-instances)
- [Data Fields](#data-fields)
- [Data Splits](#data-splits)
## Dataset Descritpion
This dataset has been automatically processed by AutoNLP for project cat33.
### Languages
The BCP-47 code for the dataset's language is zh.
## Dataset Structure
### Data Instances
A sample from this dataset looks as follows:
```json
[
{
"text": "\"\u5341\u56db\u4e94\"\u65f6\u671f\uff0c\u4f9d\u6258\u6d77\u5357\u5730\u7406\u533a\u4f4d\u4f18\u52bf\u548c\u6d77\u6d0b\u8d44\u6e90\u4f18\u52bf\uff0c\u52a0\u5feb\u57f9\u80b2\u58ee\u5927\u6d77\u6d0b\u7ecf\u6d4e\uff0c\u62d3\u5c55\u6d77\u5357\u7ecf\u6d4e\u53d1\u5c55\u84dd\u8272\u7a7a\u95f4\uff0c\u5bf9\u670d\u52a1\u6d77\u6d0b\u5f3a\u56fd\u6218\u7565\u3001\u63a8\u52a8\u6d77\u5357\u81ea\u7531\u8d38\u6613\u6e2f\u5efa\u8bbe\u53ca\u5b9e\u73b0\u81ea\u8eab\u53d1\u5c55\u5177\u6709\u91cd\u8981\u610f\u4e49",
"target": 9
},
{
"text": "\u9010\u6b65\u5b9e\u65bd\u533b\u7597\u5668\u68b0\u552f\u4e00\u6807\u8bc6\uff0c\u52a0\u5f3a\u4e0e\u533b\u7597\u7ba1\u7406\u3001\u533b\u4fdd\u7ba1\u7406\u7b49\u8854\u63a5",
"target": 8
}
]
```
### Dataset Fields
The dataset has the following fields (also called "features"):
```json
{
"text": "Value(dtype='string', id=None)",
"target": "ClassLabel(num_classes=32, names=['\u4e92\u8054\u7f51\u670d\u52a1', '\u4ea4\u901a\u8fd0\u8f93', '\u4f11\u95f2\u670d\u52a1', '\u4f20\u5a92', '\u4fe1\u606f\u6280\u672f', '\u516c\u7528\u4e8b\u4e1a', '\u519c\u4e1a', '\u5316\u5de5\u5236\u9020', '\u533b\u836f\u751f\u7269', '\u5546\u4e1a\u8d38\u6613', '\u56fd\u9632\u519b\u5de5', '\u5bb6\u7528\u7535\u5668', '\u5efa\u7b51\u4e1a', '\u623f\u5730\u4ea7', '\u6559\u80b2', '\u6587\u5316', '\u6709\u8272\u91d1\u5c5e', '\u673a\u68b0\u88c5\u5907\u5236\u9020', '\u6797\u4e1a', '\u6c7d\u8f66\u5236\u9020', '\u6e14\u4e1a', '\u7535\u5b50\u5236\u9020', '\u7535\u6c14\u8bbe\u5907', '\u755c\u7267\u4e1a', '\u7eba\u7ec7\u670d\u88c5\u5236\u9020', '\u8f7b\u5de5\u5236\u9020', '\u901a\u4fe1', '\u91c7\u77ff\u4e1a', '\u94a2\u94c1', '\u94f6\u884c', '\u975e\u94f6\u91d1\u878d', '\u98df\u54c1\u996e\u6599'], id=None)"
}
```
### Dataset Splits
This dataset is split into a train and validation split. The split sizes are as follow:
| Split name | Num samples |
| ------------ | ------------------- |
| train | 1836 |
| valid | 460 |
|
true |
# Dataset Card for [es_tweets_laboral]
## Table of Contents
- [Table of Contents](#table-of-contents)
- [Dataset Description](#dataset-description)
- [Dataset Summary](#dataset-summary)
- [Supported Tasks and Leaderboards](#supported-tasks-and-leaderboards)
- [Languages](#languages)
- [Dataset Structure](#dataset-structure)
- [Data Instances](#data-instances)
- [Data Fields](#data-fields)
- [Data Splits](#data-splits)
- [Dataset Creation](#dataset-creation)
- [Curation Rationale](#curation-rationale)
- [Source Data](#source-data)
- [Annotations](#annotations)
- [Personal and Sensitive Information](#personal-and-sensitive-information)
- [Considerations for Using the Data](#considerations-for-using-the-data)
- [Social Impact of Dataset](#social-impact-of-dataset)
- [Discussion of Biases](#discussion-of-biases)
- [Other Known Limitations](#other-known-limitations)
- [Additional Information](#additional-information)
- [Dataset Curators](#dataset-curators)
- [Licensing Information](#licensing-information)
- [Citation Information](#citation-information)
- [Contributions](#contributions)
## Dataset Description
Dataset creado por @hucruz, @DanielaGarciaQuezada, @hylandude, @BloodBoy21
Etiquetado por @DanielaGarciaQuezada
- **Homepage:**
- **Repository:**
- **Paper:**
- **Leaderboard:**
- **Point of Contact:**
### Dataset Summary
[More Information Needed]
### Supported Tasks and Leaderboards
[More Information Needed]
### Languages
español
## Dataset Structure
### Data Instances
[More Information Needed]
### Data Fields
[More Information Needed]
### Data Splits
[More Information Needed]
## Dataset Creation
### Curation Rationale
[More Information Needed]
### Source Data
#### Initial Data Collection and Normalization
[More Information Needed]
#### Who are the source language producers?
[More Information Needed]
### Annotations
#### Annotation process
[More Information Needed]
#### Who are the annotators?
[More Information Needed]
### Personal and Sensitive Information
[More Information Needed]
## Considerations for Using the Data
### Social Impact of Dataset
[More Information Needed]
### Discussion of Biases
[More Information Needed]
### Other Known Limitations
[More Information Needed]
## Additional Information
### Dataset Curators
[More Information Needed]
### Licensing Information
[More Information Needed]
### Citation Information
[More Information Needed]
### Contributions
Thanks to [@github-username](https://github.com/<github-username>) for adding this dataset. |
true |
# Dataset Card for [readability-es-sentences]
## Dataset Description
Compilation of short Spanish articles for readability assessment.
### Dataset Summary
This dataset is a compilation of short articles from websites dedicated to learn Spanish as a second language. These articles have been compiled from the following sources:
- **Coh-Metrix-Esp corpus (Quispesaravia, et al., 2016):** collection of 100 parallel texts with simple and complex variants in Spanish. These texts include children's and adult stories to fulfill each category.
- **[kwiziq](https://www.kwiziq.com/):** a language learner assistant
- **[hablacultura.com](https://hablacultura.com/):** Spanish resources for students and teachers. We have downloaded the available content in their websites.
### Languages
Spanish
## Dataset Structure
The dataset includes 1019 text entries between 80 and 8714 characters long. The vast majority (97%) are below 4,000 characters long.
### Data Fields
The dataset is formatted as a json lines and includes the following fields:
- **Category:** when available, this includes the level of this text according to the Common European Framework of Reference for Languages (CEFR).
- **Level:** standardized readability level: complex or simple.
- **Level-3:** standardized readability level: basic, intermediate or advanced
- **Text:** original text formatted into sentences.
Not all the entries contain usable values for `category`, `level` and `level-3`, but all of them should contain at least one of `level`, `level-3`. When the corresponding information could not be derived, we use the special `"N/A"` value to indicate so.
## Additional Information
### Licensing Information
https://creativecommons.org/licenses/by-nc-sa/4.0/
### Citation Information
Please cite this page to give credit to the authors :)
### Team
- [Laura Vásquez-Rodríguez](https://lmvasque.github.io/)
- [Pedro Cuenca](https://twitter.com/pcuenq)
- [Sergio Morales](https://www.fireblend.com/)
- [Fernando Alva-Manchego](https://feralvam.github.io/)
|
false |
# Dataset Card for "Abkhaz text"
## Table of Contents
- [Dataset Description](#dataset-description)
- [Dataset Summary](#dataset-summary)
- [Considerations for Using the Data](#considerations-for-using-the-data)
- [Other Known Limitations](#other-known-limitations)
## Dataset Description
- **Point of Contact:** [Nart Tlisha](mailto:daniel.abzakh@gmail.com)
- **Size of the generated dataset:** 176 MB
### Dataset Summary
The Abkhaz language monolingual dataset is a collection of 1,470,480 sentences extracted from different sources. The dataset is available under the Creative Commons Universal Public Domain License. Part of it is also available as part of [Common Voice](https://commonvoice.mozilla.org/ab), another part is from the [Abkhaz National Corpus](https://clarino.uib.no/abnc)
## Dataset Creation
### Source Data
Here is a link to the source of a large part of the data on [github](https://github.com/danielinux7/Multilingual-Parallel-Corpus/blob/master/ebooks/reference.md)
## Considerations for Using the Data
### Other Known Limitations
The accuracy of the dataset is around 95% (gramatical, arthographical errors)
|
false |
# Dataset Card for MTet
## Table of Contents
- [Table of Contents](#table-of-contents)
- [Dataset Description](#dataset-description)
- [Dataset Summary](#dataset-summary)
- [Supported Tasks and Leaderboards](#supported-tasks-and-leaderboards)
- [Languages](#languages)
- [Dataset Structure](#dataset-structure)
- [Data Instances](#data-instances)
- [Data Fields](#data-fields)
- [Data Splits](#data-splits)
- [Dataset Creation](#dataset-creation)
- [Curation Rationale](#curation-rationale)
- [Source Data](#source-data)
- [Annotations](#annotations)
- [Personal and Sensitive Information](#personal-and-sensitive-information)
- [Considerations for Using the Data](#considerations-for-using-the-data)
- [Social Impact of Dataset](#social-impact-of-dataset)
- [Discussion of Biases](#discussion-of-biases)
- [Other Known Limitations](#other-known-limitations)
- [Additional Information](#additional-information)
- [Dataset Curators](#dataset-curators)
- [Licensing Information](#licensing-information)
- [Citation Information](#citation-information)
- [Contributions](#contributions)
## Dataset Description
- **Homepage:** https://translate.vietai.org/
- **Repository:** https://github.com/vietai/mTet
- **Paper:**
- **Leaderboard:**
- **Point of Contact:**
### Dataset Summary
MTet (Multi-domain Translation for English-Vietnamese) dataset contains roughly 4.2 million English-Vietnamese pairs of
texts, ranging across multiple different domains such as medical publications, religious texts, engineering articles,
literature, news, and poems.
This dataset extends our previous SAT (Style Augmented Translation) dataset (v1.0) by adding more high-quality
English-Vietnamese sentence pairs on various domains.
### Supported Tasks and Leaderboards
- Machine Translation
### Languages
The languages in the dataset are:
- Vietnamese (`vi`)
- English (`en`)
## Dataset Structure
### Data Instances
```
{
'translation': {
'en': 'He said that existing restrictions would henceforth be legally enforceable, and violators would be fined.',
'vi': 'Ông nói những biện pháp hạn chế hiện tại sẽ được nâng lên thành quy định pháp luật, và những ai vi phạm sẽ chịu phạt.'
}
}
```
### Data Fields
- `translation`:
- `en`: Parallel text in English.
- `vi`: Parallel text in Vietnamese.
### Data Splits
The dataset is in a single "train" split.
| | train |
|--------------------|--------:|
| Number of examples | 4163853 |
## Dataset Creation
### Curation Rationale
[More Information Needed]
### Source Data
#### Initial Data Collection and Normalization
[More Information Needed]
#### Who are the source language producers?
[More Information Needed]
### Annotations
#### Annotation process
[More Information Needed]
#### Who are the annotators?
[More Information Needed]
### Personal and Sensitive Information
[More Information Needed]
## Considerations for Using the Data
### Social Impact of Dataset
[More Information Needed]
### Discussion of Biases
[More Information Needed]
### Other Known Limitations
[More Information Needed]
## Additional Information
### Dataset Curators
[More Information Needed]
### Licensing Information
[Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International (CC BY-NC-SA 4.0)](https://creativecommons.org/licenses/by-nc-sa/4.0/).
### Citation Information
```bibtex
@article{mTet2022,
author = {Chinh Ngo, Hieu Tran, Long Phan, Trieu H. Trinh, Hieu Nguyen, Minh Nguyen, Minh-Thang Luong},
title = {MTet: Multi-domain Translation for English and Vietnamese},
journal = {https://github.com/vietai/mTet},
year = {2022},
}
```
### Contributions
Thanks to [@albertvillanova](https://huggingface.co/albertvillanova) for adding this dataset.
|
false |
# NEREL dataset
## Table of Contents
- [Dataset Description](#dataset-description)
- [Dataset Structure](#dataset-structure)
- [Citation Information](#citation-information)
- [Contacts](#contacts)
## Dataset Description
NEREL dataset (https://doi.org/10.48550/arXiv.2108.13112) is
a Russian dataset for named entity recognition and relation extraction.
NEREL is significantly larger than existing Russian datasets:
to date it contains 56K annotated named entities and 39K annotated relations.
Its important difference from previous datasets is annotation of nested named
entities, as well as relations within nested entities and at the discourse
level. NEREL can facilitate development of novel models that can extract
relations between nested named entities, as well as relations on both sentence
and document levels. NEREL also contains the annotation of events involving
named entities and their roles in the events.
You can see full entity types list in a subset "ent_types"
and full list of relation types in a subset "rel_types".
## Dataset Structure
There are three "configs" or "subsets" of the dataset.
Using
`load_dataset('MalakhovIlya/NEREL', 'ent_types')['ent_types']`
you can download list of entity types (
Dataset({features: ['type', 'link']})
) where "link" is a knowledge base name used in entity linking task.
Using
`load_dataset('MalakhovIlya/NEREL', 'rel_types')['rel_types']`
you can download list of entity types (
Dataset({features: ['type', 'arg1', 'arg2']})
) where "arg1" and "arg2" are lists of entity types that can take part in such
"type" of relation. \<ENTITY> stands for any type.
Using
`load_dataset('MalakhovIlya/NEREL', 'data')` or `load_dataset('MalakhovIlya/NEREL')`
you can download the data itself,
DatasetDict with 3 splits: "train", "test" and "dev".
Each of them contains text document with annotated entities, relations and
links.
"entities" are used in named-entity recognition task (see https://en.wikipedia.org/wiki/Named-entity_recognition).
"relations" are used in relationship extraction task (see https://en.wikipedia.org/wiki/Relationship_extraction).
"links" are used in entity linking task (see https://en.wikipedia.org/wiki/Entity_linking)
Each entity is represented by a string of the following format:
`"<id>\t<type> <start> <stop>\t<text>"`, where
`<id>` is an entity id,
`<type>` is one of entity types,
`<start>` is a position of the first symbol of entity in text,
`<stop>` is the last symbol position in text +1.
Each relation is represented by a string of the following format:
`"<id>\t<type> Arg1:<arg1_id> Arg2:<arg2_id>"`, where
`<id>` is a relation id,
`<arg1_id>` and `<arg2_id>` are entity ids.
Each link is represented by a string of the following format:
`"<id>\tReference <ent_id> <link>\t<text>"`, where
`<id>` is a link id,
`<ent_id>` is an entity id,
`<link>` is a reference to knowledge base entity (example: "Wikidata:Q1879675" if link exists, else "Wikidata:NULL"),
`<text>` is a name of entity in knowledge base if link exists, else empty string.
## Citation Information
@article{loukachevitch2021nerel,
title={NEREL: A Russian Dataset with Nested Named Entities, Relations and Events},
author={Loukachevitch, Natalia and Artemova, Ekaterina and Batura, Tatiana and Braslavski, Pavel and Denisov, Ilia and Ivanov, Vladimir and Manandhar, Suresh and Pugachev, Alexander and Tutubalina, Elena},
journal={arXiv preprint arXiv:2108.13112},
year={2021}
}
|
true | # Dataset Card for YOSM
## Table of Contents
- [Dataset Description](#dataset-description)
- [Dataset Summary](#dataset-summary)
- [Supported Tasks and Leaderboards](#supported-tasks-and-leaderboards)
- [Languages](#languages)
- [Dataset Structure](#dataset-structure)
- [Data Instances](#data-instances)
- [Data Fields](#data-fields)
- [Data Splits](#data-splits)
- [Dataset Creation](#dataset-creation)
- [Curation Rationale](#curation-rationale)
- [Source Data](#source-data)
- [Annotations](#annotations)
- [Personal and Sensitive Information](#personal-and-sensitive-information)
- [Considerations for Using the Data](#considerations-for-using-the-data)
- [Social Impact of Dataset](#social-impact-of-dataset)
- [Discussion of Biases](#discussion-of-biases)
- [Other Known Limitations](#other-known-limitations)
- [Additional Information](#additional-information)
- [Dataset Curators](#dataset-curators)
- [Licensing Information](#licensing-information)
- [Citation Information](#citation-information)
- [Contributions](#contributions)
## Dataset Description
- **Homepage:**
- **Repository:** [Iyanuoluwa/YOSM](https://github.com/IyanuSh/YOSM)
- **Paper:** [A new Yorùbá Sentiment Corpus for Nigerian/Nollywood Movie Reviews](https://arxiv.org/pdf/2204.09711.pdf)
- **Point of Contact:** [Iyanuoluwa Shode](mailto:shodei1@montclair.edu)
### Dataset Summary
YOSM is the first Yorùbá sentiment corpus for Nollywood movie reviews. The reviews were collected from movie reviews websites - IMDB, Rotten Tomatoes, LetterboxD, Cinemapointer, and Nollyrated.
### Languages
Yorùbá (ISO 639-1: yo) - the third most spoken indigenous African language with over 50 million speakers.
## Dataset Structure
### Data Instances
An instance consists of a movie review and the corresponding class label.
### Data Fields
- `yo_review`: A movie review in Yorùbá
- `sentiment`: The label describing the sentiment of the movie review.
### Data Splits
The YOSM dataset has 3 splits: _train_, _dev_, and _test_. Below are the statistics for Version 3.0.0 of the dataset.
| Dataset Split | Number of Instances in Split |
| ------------- | ------------------------------------------- |
| Train | 800 |
| Development | 200 |
| Test | 500 |
### Data Splits
[More Information Needed]
## Dataset Creation
### Curation Rationale
[More Information Needed]
### Source Data
#### Initial Data Collection and Normalization
[More Information Needed]
#### Who are the source language producers?
[More Information Needed]
### Annotations
#### Annotation process
[More Information Needed]
#### Who are the annotators?
[More Information Needed]
### Personal and Sensitive Information
[More Information Needed]
## Considerations for Using the Data
### Social Impact of Dataset
[More Information Needed]
### Discussion of Biases
[More Information Needed]
### Other Known Limitations
[More Information Needed]
## Additional Information
### Dataset Curators
[More Information Needed]
### Licensing Information
[More Information Needed]
### Citation Information
[More Information Needed]
### Contributions
|
false | |
false |
# Dataset Card for BEIR Benchmark
## Table of Contents
- [Dataset Description](#dataset-description)
- [Dataset Summary](#dataset-summary)
- [Supported Tasks and Leaderboards](#supported-tasks-and-leaderboards)
- [Languages](#languages)
- [Dataset Structure](#dataset-structure)
- [Data Instances](#data-instances)
- [Data Fields](#data-fields)
- [Data Splits](#data-splits)
- [Dataset Creation](#dataset-creation)
- [Curation Rationale](#curation-rationale)
- [Source Data](#source-data)
- [Annotations](#annotations)
- [Personal and Sensitive Information](#personal-and-sensitive-information)
- [Considerations for Using the Data](#considerations-for-using-the-data)
- [Social Impact of Dataset](#social-impact-of-dataset)
- [Discussion of Biases](#discussion-of-biases)
- [Other Known Limitations](#other-known-limitations)
- [Additional Information](#additional-information)
- [Dataset Curators](#dataset-curators)
- [Licensing Information](#licensing-information)
- [Citation Information](#citation-information)
- [Contributions](#contributions)
## Dataset Description
- **Homepage:** https://github.com/UKPLab/beir
- **Repository:** https://github.com/UKPLab/beir
- **Paper:** https://openreview.net/forum?id=wCu6T5xFjeJ
- **Leaderboard:** https://docs.google.com/spreadsheets/d/1L8aACyPaXrL8iEelJLGqlMqXKPX2oSP_R10pZoy77Ns
- **Point of Contact:** nandan.thakur@uwaterloo.ca
### Dataset Summary
BEIR is a heterogeneous benchmark that has been built from 18 diverse datasets representing 9 information retrieval tasks:
- Fact-checking: [FEVER](http://fever.ai), [Climate-FEVER](http://climatefever.ai), [SciFact](https://github.com/allenai/scifact)
- Question-Answering: [NQ](https://ai.google.com/research/NaturalQuestions), [HotpotQA](https://hotpotqa.github.io), [FiQA-2018](https://sites.google.com/view/fiqa/)
- Bio-Medical IR: [TREC-COVID](https://ir.nist.gov/covidSubmit/index.html), [BioASQ](http://bioasq.org), [NFCorpus](https://www.cl.uni-heidelberg.de/statnlpgroup/nfcorpus/)
- News Retrieval: [TREC-NEWS](https://trec.nist.gov/data/news2019.html), [Robust04](https://trec.nist.gov/data/robust/04.guidelines.html)
- Argument Retrieval: [Touche-2020](https://webis.de/events/touche-20/shared-task-1.html), [ArguAna](tp://argumentation.bplaced.net/arguana/data)
- Duplicate Question Retrieval: [Quora](https://www.quora.com/q/quoradata/First-Quora-Dataset-Release-Question-Pairs), [CqaDupstack](http://nlp.cis.unimelb.edu.au/resources/cqadupstack/)
- Citation-Prediction: [SCIDOCS](https://allenai.org/data/scidocs)
- Tweet Retrieval: [Signal-1M](https://research.signal-ai.com/datasets/signal1m-tweetir.html)
- Entity Retrieval: [DBPedia](https://github.com/iai-group/DBpedia-Entity/)
All these datasets have been preprocessed and can be used for your experiments.
```python
```
### Supported Tasks and Leaderboards
The dataset supports a leaderboard that evaluates models against task-specific metrics such as F1 or EM, as well as their ability to retrieve supporting information from Wikipedia.
The current best performing models can be found [here](https://eval.ai/web/challenges/challenge-page/689/leaderboard/).
### Languages
All tasks are in English (`en`).
## Dataset Structure
All BEIR datasets must contain a corpus, queries and qrels (relevance judgments file). They must be in the following format:
- `corpus` file: a `.jsonl` file (jsonlines) that contains a list of dictionaries, each with three fields `_id` with unique document identifier, `title` with document title (optional) and `text` with document paragraph or passage. For example: `{"_id": "doc1", "title": "Albert Einstein", "text": "Albert Einstein was a German-born...."}`
- `queries` file: a `.jsonl` file (jsonlines) that contains a list of dictionaries, each with two fields `_id` with unique query identifier and `text` with query text. For example: `{"_id": "q1", "text": "Who developed the mass-energy equivalence formula?"}`
- `qrels` file: a `.tsv` file (tab-seperated) that contains three columns, i.e. the `query-id`, `corpus-id` and `score` in this order. Keep 1st row as header. For example: `q1 doc1 1`
### Data Instances
A high level example of any beir dataset:
```python
corpus = {
"doc1" : {
"title": "Albert Einstein",
"text": "Albert Einstein was a German-born theoretical physicist. who developed the theory of relativity, \
one of the two pillars of modern physics (alongside quantum mechanics). His work is also known for \
its influence on the philosophy of science. He is best known to the general public for his mass–energy \
equivalence formula E = mc2, which has been dubbed 'the world's most famous equation'. He received the 1921 \
Nobel Prize in Physics 'for his services to theoretical physics, and especially for his discovery of the law \
of the photoelectric effect', a pivotal step in the development of quantum theory."
},
"doc2" : {
"title": "", # Keep title an empty string if not present
"text": "Wheat beer is a top-fermented beer which is brewed with a large proportion of wheat relative to the amount of \
malted barley. The two main varieties are German Weißbier and Belgian witbier; other types include Lambic (made\
with wild yeast), Berliner Weisse (a cloudy, sour beer), and Gose (a sour, salty beer)."
},
}
queries = {
"q1" : "Who developed the mass-energy equivalence formula?",
"q2" : "Which beer is brewed with a large proportion of wheat?"
}
qrels = {
"q1" : {"doc1": 1},
"q2" : {"doc2": 1},
}
```
### Data Fields
Examples from all configurations have the following features:
### Corpus
- `corpus`: a `dict` feature representing the document title and passage text, made up of:
- `_id`: a `string` feature representing the unique document id
- `title`: a `string` feature, denoting the title of the document.
- `text`: a `string` feature, denoting the text of the document.
### Queries
- `queries`: a `dict` feature representing the query, made up of:
- `_id`: a `string` feature representing the unique query id
- `text`: a `string` feature, denoting the text of the query.
### Qrels
- `qrels`: a `dict` feature representing the query document relevance judgements, made up of:
- `_id`: a `string` feature representing the query id
- `_id`: a `string` feature, denoting the document id.
- `score`: a `int32` feature, denoting the relevance judgement between query and document.
### Data Splits
| Dataset | Website| BEIR-Name | Type | Queries | Corpus | Rel D/Q | Down-load | md5 |
| -------- | -----| ---------| --------- | ----------- | ---------| ---------| :----------: | :------:|
| MSMARCO | [Homepage](https://microsoft.github.io/msmarco/)| ``msmarco`` | ``train``<br>``dev``<br>``test``| 6,980 | 8.84M | 1.1 | [Link](https://public.ukp.informatik.tu-darmstadt.de/thakur/BEIR/datasets/msmarco.zip) | ``444067daf65d982533ea17ebd59501e4`` |
| TREC-COVID | [Homepage](https://ir.nist.gov/covidSubmit/index.html)| ``trec-covid``| ``test``| 50| 171K| 493.5 | [Link](https://public.ukp.informatik.tu-darmstadt.de/thakur/BEIR/datasets/trec-covid.zip) | ``ce62140cb23feb9becf6270d0d1fe6d1`` |
| NFCorpus | [Homepage](https://www.cl.uni-heidelberg.de/statnlpgroup/nfcorpus/) | ``nfcorpus`` | ``train``<br>``dev``<br>``test``| 323 | 3.6K | 38.2 | [Link](https://public.ukp.informatik.tu-darmstadt.de/thakur/BEIR/datasets/nfcorpus.zip) | ``a89dba18a62ef92f7d323ec890a0d38d`` |
| BioASQ | [Homepage](http://bioasq.org) | ``bioasq``| ``train``<br>``test`` | 500 | 14.91M | 8.05 | No | [How to Reproduce?](https://github.com/UKPLab/beir/blob/main/examples/dataset#2-bioasq) |
| NQ | [Homepage](https://ai.google.com/research/NaturalQuestions) | ``nq``| ``train``<br>``test``| 3,452 | 2.68M | 1.2 | [Link](https://public.ukp.informatik.tu-darmstadt.de/thakur/BEIR/datasets/nq.zip) | ``d4d3d2e48787a744b6f6e691ff534307`` |
| HotpotQA | [Homepage](https://hotpotqa.github.io) | ``hotpotqa``| ``train``<br>``dev``<br>``test``| 7,405 | 5.23M | 2.0 | [Link](https://public.ukp.informatik.tu-darmstadt.de/thakur/BEIR/datasets/hotpotqa.zip) | ``f412724f78b0d91183a0e86805e16114`` |
| FiQA-2018 | [Homepage](https://sites.google.com/view/fiqa/) | ``fiqa`` | ``train``<br>``dev``<br>``test``| 648 | 57K | 2.6 | [Link](https://public.ukp.informatik.tu-darmstadt.de/thakur/BEIR/datasets/fiqa.zip) | ``17918ed23cd04fb15047f73e6c3bd9d9`` |
| Signal-1M(RT) | [Homepage](https://research.signal-ai.com/datasets/signal1m-tweetir.html)| ``signal1m`` | ``test``| 97 | 2.86M | 19.6 | No | [How to Reproduce?](https://github.com/UKPLab/beir/blob/main/examples/dataset#4-signal-1m) |
| TREC-NEWS | [Homepage](https://trec.nist.gov/data/news2019.html) | ``trec-news`` | ``test``| 57 | 595K | 19.6 | No | [How to Reproduce?](https://github.com/UKPLab/beir/blob/main/examples/dataset#1-trec-news) |
| ArguAna | [Homepage](http://argumentation.bplaced.net/arguana/data) | ``arguana``| ``test`` | 1,406 | 8.67K | 1.0 | [Link](https://public.ukp.informatik.tu-darmstadt.de/thakur/BEIR/datasets/arguana.zip) | ``8ad3e3c2a5867cdced806d6503f29b99`` |
| Touche-2020| [Homepage](https://webis.de/events/touche-20/shared-task-1.html) | ``webis-touche2020``| ``test``| 49 | 382K | 19.0 | [Link](https://public.ukp.informatik.tu-darmstadt.de/thakur/BEIR/datasets/webis-touche2020.zip) | ``46f650ba5a527fc69e0a6521c5a23563`` |
| CQADupstack| [Homepage](http://nlp.cis.unimelb.edu.au/resources/cqadupstack/) | ``cqadupstack``| ``test``| 13,145 | 457K | 1.4 | [Link](https://public.ukp.informatik.tu-darmstadt.de/thakur/BEIR/datasets/cqadupstack.zip) | ``4e41456d7df8ee7760a7f866133bda78`` |
| Quora| [Homepage](https://www.quora.com/q/quoradata/First-Quora-Dataset-Release-Question-Pairs) | ``quora``| ``dev``<br>``test``| 10,000 | 523K | 1.6 | [Link](https://public.ukp.informatik.tu-darmstadt.de/thakur/BEIR/datasets/quora.zip) | ``18fb154900ba42a600f84b839c173167`` |
| DBPedia | [Homepage](https://github.com/iai-group/DBpedia-Entity/) | ``dbpedia-entity``| ``dev``<br>``test``| 400 | 4.63M | 38.2 | [Link](https://public.ukp.informatik.tu-darmstadt.de/thakur/BEIR/datasets/dbpedia-entity.zip) | ``c2a39eb420a3164af735795df012ac2c`` |
| SCIDOCS| [Homepage](https://allenai.org/data/scidocs) | ``scidocs``| ``test``| 1,000 | 25K | 4.9 | [Link](https://public.ukp.informatik.tu-darmstadt.de/thakur/BEIR/datasets/scidocs.zip) | ``38121350fc3a4d2f48850f6aff52e4a9`` |
| FEVER | [Homepage](http://fever.ai) | ``fever``| ``train``<br>``dev``<br>``test``| 6,666 | 5.42M | 1.2| [Link](https://public.ukp.informatik.tu-darmstadt.de/thakur/BEIR/datasets/fever.zip) | ``5a818580227bfb4b35bb6fa46d9b6c03`` |
| Climate-FEVER| [Homepage](http://climatefever.ai) | ``climate-fever``|``test``| 1,535 | 5.42M | 3.0 | [Link](https://public.ukp.informatik.tu-darmstadt.de/thakur/BEIR/datasets/climate-fever.zip) | ``8b66f0a9126c521bae2bde127b4dc99d`` |
| SciFact| [Homepage](https://github.com/allenai/scifact) | ``scifact``| ``train``<br>``test``| 300 | 5K | 1.1 | [Link](https://public.ukp.informatik.tu-darmstadt.de/thakur/BEIR/datasets/scifact.zip) | ``5f7d1de60b170fc8027bb7898e2efca1`` |
| Robust04 | [Homepage](https://trec.nist.gov/data/robust/04.guidelines.html) | ``robust04``| ``test``| 249 | 528K | 69.9 | No | [How to Reproduce?](https://github.com/UKPLab/beir/blob/main/examples/dataset#3-robust04) |
## Dataset Creation
### Curation Rationale
[Needs More Information]
### Source Data
#### Initial Data Collection and Normalization
[Needs More Information]
#### Who are the source language producers?
[Needs More Information]
### Annotations
#### Annotation process
[Needs More Information]
#### Who are the annotators?
[Needs More Information]
### Personal and Sensitive Information
[Needs More Information]
## Considerations for Using the Data
### Social Impact of Dataset
[Needs More Information]
### Discussion of Biases
[Needs More Information]
### Other Known Limitations
[Needs More Information]
## Additional Information
### Dataset Curators
[Needs More Information]
### Licensing Information
[Needs More Information]
### Citation Information
Cite as:
```
@inproceedings{
thakur2021beir,
title={{BEIR}: A Heterogeneous Benchmark for Zero-shot Evaluation of Information Retrieval Models},
author={Nandan Thakur and Nils Reimers and Andreas R{\"u}ckl{\'e} and Abhishek Srivastava and Iryna Gurevych},
booktitle={Thirty-fifth Conference on Neural Information Processing Systems Datasets and Benchmarks Track (Round 2)},
year={2021},
url={https://openreview.net/forum?id=wCu6T5xFjeJ}
}
```
### Contributions
Thanks to [@Nthakur20](https://github.com/Nthakur20) for adding this dataset. |
false |
# NEMO-Corpus - The Hebrew Named Entities and Morphology Corpus
**Disclaimer**: It's just a huggingface datasets convenient interface for research purpose which is fetching the original data from [github](https://github.com/OnlpLab/NEMO-Corpus). I'm not an author of this work.
```python
from datasets import load_dataset
# the main corpus
ds = load_dataset('imvladikon/nemo_corpus')
for sample in ds["train"]:
print(sample)
# the nested corpus
ds = load_dataset('imvladikon/nemo_corpus', "nested")
```
Getting classes and encoding/decoding could be done through these functions:
```
idx2label = dataset["train"].features["ner_tags"].feature.int2str
label2idx = dataset["train"].features["ner_tags"].feature.str2int
```
or just use raw_tags field.
## Fields
available fields (flat):
* "id"
* "sentence"
* "tokens"
* "raw_tags"
* "ner_tags"
* "spans"
Example of the one record for `flat`:
```json
{'id': '0', 'tokens': ['"', 'תהיה', 'נקמה', 'ו', 'בגדול', '.'], 'sentence': '" תהיה נקמה ו בגדול .', 'raw_tags': ['O', 'O', 'O', 'O', 'O', 'O'], 'ner_tags': [24, 24, 24, 24, 24, 24], 'spans': {'span': [], 'start': [], 'end': [], 'entity': [], 'start_char': [], 'end_char': []}}
```
Example of the one record for `nested`:
```json
{'id': '0', 'tokens': ['"', 'תהיה', 'נקמה', 'ו', 'בגדול', '.'], 'ner_tags': [24, 24, 24, 24, 24, 24], 'ner_tags_2': [24, 24, 24, 24, 24, 24], 'ner_tags_3': [24, 24, 24, 24, 24, 24], 'ner_tags_4': [24, 24, 24, 24, 24, 24]}
```
## Dataset Description
it's README.md of the [original repository](https://github.com/OnlpLab/NEMO-Corpus)
Named Entity (NER) annotations of the Hebrew Treebank (Haaretz newspaper) corpus, including: morpheme and token level NER labels, nested mentions, and more.
We publish the NEMO corpus in the TACL paper [*"Neural Modeling for Named Entities and Morphology (NEMO<sup>2</sup>)"*](https://direct.mit.edu/tacl/article/doi/10.1162/tacl_a_00404/107206/Neural-Modeling-for-Named-Entities-and-Morphology) [1], where we use it in extensive experiments and analyses, showing the importance of morphological boundaries for neural modeling of NER in morphologically rich languages. Code for these models and experiments can be found in the [NEMO code repo](https://github.com/OnlpLab/NEMO).
## Main features:
1. Morpheme, token-single and token-multi sequence labels. Morpheme labels provide exact boundaries, token-multi provide partial sub-word morphological but no exact boundaries, token-single provides only token-level information.
1. All annotations are in `BIOSE` format (`B`=Begin, `I`=Inside, `O`=Outside, `S`=Singleton, `E`=End).
1. Widely-used OntoNotes entity category set: `GPE` (geo-political entity), `PER` (person), `LOC` (location), `ORG` (organization), `FAC` (facility), `EVE` (event), `WOA` (work-of-art), `ANG` (language), `DUC` (product).
1. NEMO includes NER annotations for the two major versions of the Hebrew Treebank, UD (Universal Dependency) and SPMRL. These can be aligned to the other morphosyntactic information layers of the treebank using [bclm](https://github.com/OnlpLab/bclm)
1. We provide nested mentions. Only the first, widest, layer is used in the NEMO<sup>2</sup> paper. We invite you to take on this challenge!
1. Guidelines used for annotation are provided [here](./guidelines/).
1. Corpus was annotated by two native Hebrew speakers of academic education, and curated by the project manager. We provide the original annotations made by the annotators as well to promote work on [learning with disagreements](https://sites.google.com/view/semeval2021-task12/home).
1. Annotation was performed using [WebAnno](https://webanno.github.io/webanno/) (version 3.4.5)
## Legend for Files and Folder Structure
1. The two main [data](./data/) folders are [ud](./data/ud/) and [spmrl](./data/spmrl/), corresponding to the relevant Hebrew Treebank corpus version.
1. Both contain a `gold` folder ([spmrl/gold](./data/spmrl/gold/), [ud/gold](./data/ud/gold/)) of gold curated annotations.
1. Each `gold` folder contains files of the three input-output variants (morph, token-multi, token-single), for each of the treebank splits (train,dev,test).
1. Each `gold` folder also contains a `nested` subfolder ([spmrl/nested](./data/spmrl/gold/nested/), [ud/nested](./data/ud/gold/nested/)), which contains all layers of nested mentions (the first layer is the layer used in the non-nested files, and in the NEMO<sup>2</sup> paper [1])
1. The `ud` folder also contains an [ab_annotators](./data/ud/ab_annotators/) folder. This folder contains the original annotations made by each annotator (named `a`, `b`), including first-layer and nested annotatations.
1. *\*UPDATE 2021-09-06\** `ud` folder now contains a [pilot_annotations](./data/ud/pilot_annotations/) folder. This folder contains the original annotations made by each annotator in our two phase pilot (phase I - sentences 1-200 of dev; phase II - sentences 201-400 of dev).
## Basic Corpus Statistics
| | train | dev | test |
|------------------------------| --:| --:| --:|
| Sentences | 4,937 | 500 | 706 |
| Tokens | 93,504 | 8,531 | 12,619 |
| Morphemes | 127,031 | 11,301 | 16,828 |
| All mentions | 6,282 | 499 | 932 |
| Type: Person (PER) | 2,128 | 193 | 267 |
| Type: Organization (ORG) | 2,043 | 119 | 408 |
| Type: Geo-Political (GPE) | 1,377 | 121 | 195 |
| Type: Location (LOC) | 331 | 28 | 41 |
| Type: Facility (FAC) | 163 | 12 | 11 |
| Type: Work-of-Art (WOA) | 114 | 9 | 6 |
| Type: Event (EVE) | 57 | 12 | 0 |
| Type: Product (DUC) | 36 | 2 | 3 |
| Type: Language (ANG) | 33 | 3 | 1 |
## Aligned Treenbank Versions
The NEMO corpus matches the treebank version of [bclm v.1.0.0](https://github.com/OnlpLab/bclm/releases/tag/v1.0.0-alpha).
This version is based on the [HTB UD v2.2](https://github.com/UniversalDependencies/UD_Hebrew-HTB/releases/tag/r2.2) and the [latest SPMRL HTB version](https://github.com/OnlpLab/HebrewResources/tree/102674bb030f5836e1ab827feb63954ad7a6f8fe/HebrewTreebank/hebtb).
The changes contain (but might not be limited to the following):
1. Flagged and dropped duplicate and leaking sentences (between train and test). In addition to the sentences already removed in the bclm v1.0.0 HTB version, the following duplicate sentences were dropped as well (SPMRL sentence IDs): 5438, 5444, 5445, 5446, 5448, 5449, 5450, 5451, 5453, 5459 (in the bclm dataframes, these are marked in the `duplicate_sent_id` column).
To read the treebank (UD/SPMRL) in a way that matches the NEMO corpus, you can use the following:
```python
import bclm
dropped = [5438, 5444, 5445, 5446, 5448, 5449, 5450, 5451, 5453, 5459]
spdf = bclm.read_dataframe('spmrl') # load SPMRL treebank dataframe
global_dropped = [spdf[spdf.sent_id==d].global_sent_id.iat[0] for d in dropped]
uddf = bclm.read_dataframe('ud') # load UD treebank dataframe
uddf = uddf[(~uddf.global_sent_id.isin(global_dropped))] # remove extra duplicates
spdf = spdf[(~spdf.sent_id.isin(dropped))] # remove extra duplicates
# The resulting dataframes contain gold morph NER labels in the `biose_layer0`, `biose_layer1`... columns.
```
2. The UD treebank contains many more duplicates. In this version: all sentences exist in both UD and SPMRL versions, and all sentences and tokens are aligned between UD and SPMRL.
2. Fixed numbers that were originally reversed.
2. Fixed mismatches between tokens and morphemes.
2. Added Binyan feature.
2. No individual morphemes or tokens were added or removed, only complete sentences.
## Evaluation
An evaluation script is provided in the [NEMO code repo](https://github.com/OnlpLab/NEMO#evaluation) along with evaluation instructions.
## Citations
##### [1]
If you use the NEMO corpus in your research, please cite the NEMO<sup>2</sup> paper:
```bibtex
@article{10.1162/tacl_a_00404,
author = {Bareket, Dan and Tsarfaty, Reut},
title = "{Neural Modeling for Named Entities and Morphology (NEMO2)}",
journal = {Transactions of the Association for Computational Linguistics},
volume = {9},
pages = {909-928},
year = {2021},
month = {09},
abstract = "{Named Entity Recognition (NER) is a fundamental NLP task, commonly formulated as classification over a sequence of tokens. Morphologically rich languages (MRLs) pose a challenge to this basic formulation, as the boundaries of named entities do not necessarily coincide with token boundaries, rather, they respect morphological boundaries. To address NER in MRLs we then need to answer two fundamental questions, namely, what are the basic units to be labeled, and how can these units be detected and classified in realistic settings (i.e., where no gold morphology is available). We empirically investigate these questions on a novel NER benchmark, with parallel token- level and morpheme-level NER annotations, which we develop for Modern Hebrew, a morphologically rich-and-ambiguous language. Our results show that explicitly modeling morphological boundaries leads to improved NER performance, and that a novel hybrid architecture, in which NER precedes and prunes morphological decomposition, greatly outperforms the standard pipeline, where morphological decomposition strictly precedes NER, setting a new performance bar for both Hebrew NER and Hebrew morphological decomposition tasks.}",
issn = {2307-387X},
doi = {10.1162/tacl_a_00404},
url = {https://doi.org/10.1162/tacl\_a\_00404},
eprint = {https://direct.mit.edu/tacl/article-pdf/doi/10.1162/tacl\_a\_00404/1962472/tacl\_a\_00404.pdf},
}
```
##### [2]
Please cite the Hebrew Treebank as well, described the following paper:
```bibtex
@article{sima2001building,
title={Building a tree-bank of modern Hebrew text},
author={Sima’an, Khalil and Itai, Alon and Winter, Yoad and Altman, Alon and Nativ, Noa},
journal={Traitement Automatique des Langues},
volume={42},
number={2},
pages={247--380},
year={2001},
publisher={Citeseer}
}
```
##### [3]
The UD version of the Hebrew Treebank is described in:
```bibtex
@inproceedings{sade-etal-2018-hebrew,
title = "The {H}ebrew {U}niversal {D}ependency Treebank: Past Present and Future",
author = "Sade, Shoval and
Seker, Amit and
Tsarfaty, Reut",
booktitle = "Proceedings of the Second Workshop on Universal Dependencies ({UDW} 2018)",
month = nov,
year = "2018",
address = "Brussels, Belgium",
publisher = "Association for Computational Linguistics",
url = "https://www.aclweb.org/anthology/W18-6016",
doi = "10.18653/v1/W18-6016",
pages = "133--143",
abstract = "The Hebrew treebank (HTB), consisting of 6221 morpho-syntactically annotated newspaper sentences, has been the only resource for training and validating statistical parsers and taggers for Hebrew, for almost two decades now. During these decades, the HTB has gone through a trajectory of automatic and semi-automatic conversions, until arriving at its UDv2 form. In this work we manually validate the UDv2 version of the HTB, and, according to our findings, we apply scheme changes that bring the UD HTB to the same theoretical grounds as the rest of UD. Our experimental parsing results with UDv2New confirm that improving the coherence and internal consistency of the UD HTB indeed leads to improved parsing performance. At the same time, our analysis demonstrates that there is more to be done at the point of intersection of UD with other linguistic processing layers, in particular, at the points where UD interfaces external morphological and lexical resources.",
}
``` |
true |
# Dataset Card for "UnpredicTable-phonearena-com" - Dataset of Few-shot Tasks from Tables
## Table of Contents
- [Dataset Description](#dataset-description)
- [Dataset Summary](#dataset-summary)
- [Supported Tasks](#supported-tasks-and-leaderboards)
- [Languages](#languages)
- [Dataset Structure](#dataset-structure)
- [Data Instances](#data-instances)
- [Data Fields](#data-instances)
- [Data Splits](#data-instances)
- [Dataset Creation](#dataset-creation)
- [Curation Rationale](#curation-rationale)
- [Source Data](#source-data)
- [Annotations](#annotations)
- [Personal and Sensitive Information](#personal-and-sensitive-information)
- [Considerations for Using the Data](#considerations-for-using-the-data)
- [Social Impact of Dataset](#social-impact-of-dataset)
- [Discussion of Biases](#discussion-of-biases)
- [Other Known Limitations](#other-known-limitations)
- [Additional Information](#additional-information)
- [Dataset Curators](#dataset-curators)
- [Licensing Information](#licensing-information)
- [Citation Information](#citation-information)
## Dataset Description
- **Homepage:** https://ethanperez.net/unpredictable
- **Repository:** https://github.com/JunShern/few-shot-adaptation
- **Paper:** Few-shot Adaptation Works with UnpredicTable Data
- **Point of Contact:** junshern@nyu.edu, perez@nyu.edu
### Dataset Summary
The UnpredicTable dataset consists of web tables formatted as few-shot tasks for fine-tuning language models to improve their few-shot performance.
There are several dataset versions available:
* [UnpredicTable-full](https://huggingface.co/datasets/MicPie/unpredictable_full): Starting from the initial WTC corpus of 50M tables, we apply our tables-to-tasks procedure to produce our resulting dataset, [UnpredicTable-full](https://huggingface.co/datasets/MicPie/unpredictable_full), which comprises 413,299 tasks from 23,744 unique websites.
* [UnpredicTable-unique](https://huggingface.co/datasets/MicPie/unpredictable_unique): This is the same as [UnpredicTable-full](https://huggingface.co/datasets/MicPie/unpredictable_full) but filtered to have a maximum of one task per website. [UnpredicTable-unique](https://huggingface.co/datasets/MicPie/unpredictable_unique) contains exactly 23,744 tasks from 23,744 websites.
* [UnpredicTable-5k](https://huggingface.co/datasets/MicPie/unpredictable_5k): This dataset contains 5k random tables from the full dataset.
* UnpredicTable data subsets based on a manual human quality rating (please see our publication for details of the ratings):
* [UnpredicTable-rated-low](https://huggingface.co/datasets/MicPie/unpredictable_rated-low)
* [UnpredicTable-rated-medium](https://huggingface.co/datasets/MicPie/unpredictable_rated-medium)
* [UnpredicTable-rated-high](https://huggingface.co/datasets/MicPie/unpredictable_rated-high)
* UnpredicTable data subsets based on the website of origin:
* [UnpredicTable-baseball-fantasysports-yahoo-com](https://huggingface.co/datasets/MicPie/unpredictable_baseball-fantasysports-yahoo-com)
* [UnpredicTable-bulbapedia-bulbagarden-net](https://huggingface.co/datasets/MicPie/unpredictable_bulbapedia-bulbagarden-net)
* [UnpredicTable-cappex-com](https://huggingface.co/datasets/MicPie/unpredictable_cappex-com)
* [UnpredicTable-cram-com](https://huggingface.co/datasets/MicPie/unpredictable_cram-com)
* [UnpredicTable-dividend-com](https://huggingface.co/datasets/MicPie/unpredictable_dividend-com)
* [UnpredicTable-dummies-com](https://huggingface.co/datasets/MicPie/unpredictable_dummies-com)
* [UnpredicTable-en-wikipedia-org](https://huggingface.co/datasets/MicPie/unpredictable_en-wikipedia-org)
* [UnpredicTable-ensembl-org](https://huggingface.co/datasets/MicPie/unpredictable_ensembl-org)
* [UnpredicTable-gamefaqs-com](https://huggingface.co/datasets/MicPie/unpredictable_gamefaqs-com)
* [UnpredicTable-mgoblog-com](https://huggingface.co/datasets/MicPie/unpredictable_mgoblog-com)
* [UnpredicTable-mmo-champion-com](https://huggingface.co/datasets/MicPie/unpredictable_mmo-champion-com)
* [UnpredicTable-msdn-microsoft-com](https://huggingface.co/datasets/MicPie/unpredictable_msdn-microsoft-com)
* [UnpredicTable-phonearena-com](https://huggingface.co/datasets/MicPie/unpredictable_phonearena-com)
* [UnpredicTable-sittercity-com](https://huggingface.co/datasets/MicPie/unpredictable_sittercity-com)
* [UnpredicTable-sporcle-com](https://huggingface.co/datasets/MicPie/unpredictable_sporcle-com)
* [UnpredicTable-studystack-com](https://huggingface.co/datasets/MicPie/unpredictable_studystack-com)
* [UnpredicTable-support-google-com](https://huggingface.co/datasets/MicPie/unpredictable_support-google-com)
* [UnpredicTable-w3-org](https://huggingface.co/datasets/MicPie/unpredictable_w3-org)
* [UnpredicTable-wiki-openmoko-org](https://huggingface.co/datasets/MicPie/unpredictable_wiki-openmoko-org)
* [UnpredicTable-wkdu-org](https://huggingface.co/datasets/MicPie/unpredictable_wkdu-org)
* UnpredicTable data subsets based on clustering (for the clustering details please see our publication):
* [UnpredicTable-cluster00](https://huggingface.co/datasets/MicPie/unpredictable_cluster00)
* [UnpredicTable-cluster01](https://huggingface.co/datasets/MicPie/unpredictable_cluster01)
* [UnpredicTable-cluster02](https://huggingface.co/datasets/MicPie/unpredictable_cluster02)
* [UnpredicTable-cluster03](https://huggingface.co/datasets/MicPie/unpredictable_cluster03)
* [UnpredicTable-cluster04](https://huggingface.co/datasets/MicPie/unpredictable_cluster04)
* [UnpredicTable-cluster05](https://huggingface.co/datasets/MicPie/unpredictable_cluster05)
* [UnpredicTable-cluster06](https://huggingface.co/datasets/MicPie/unpredictable_cluster06)
* [UnpredicTable-cluster07](https://huggingface.co/datasets/MicPie/unpredictable_cluster07)
* [UnpredicTable-cluster08](https://huggingface.co/datasets/MicPie/unpredictable_cluster08)
* [UnpredicTable-cluster09](https://huggingface.co/datasets/MicPie/unpredictable_cluster09)
* [UnpredicTable-cluster10](https://huggingface.co/datasets/MicPie/unpredictable_cluster10)
* [UnpredicTable-cluster11](https://huggingface.co/datasets/MicPie/unpredictable_cluster11)
* [UnpredicTable-cluster12](https://huggingface.co/datasets/MicPie/unpredictable_cluster12)
* [UnpredicTable-cluster13](https://huggingface.co/datasets/MicPie/unpredictable_cluster13)
* [UnpredicTable-cluster14](https://huggingface.co/datasets/MicPie/unpredictable_cluster14)
* [UnpredicTable-cluster15](https://huggingface.co/datasets/MicPie/unpredictable_cluster15)
* [UnpredicTable-cluster16](https://huggingface.co/datasets/MicPie/unpredictable_cluster16)
* [UnpredicTable-cluster17](https://huggingface.co/datasets/MicPie/unpredictable_cluster17)
* [UnpredicTable-cluster18](https://huggingface.co/datasets/MicPie/unpredictable_cluster18)
* [UnpredicTable-cluster19](https://huggingface.co/datasets/MicPie/unpredictable_cluster19)
* [UnpredicTable-cluster20](https://huggingface.co/datasets/MicPie/unpredictable_cluster20)
* [UnpredicTable-cluster21](https://huggingface.co/datasets/MicPie/unpredictable_cluster21)
* [UnpredicTable-cluster22](https://huggingface.co/datasets/MicPie/unpredictable_cluster22)
* [UnpredicTable-cluster23](https://huggingface.co/datasets/MicPie/unpredictable_cluster23)
* [UnpredicTable-cluster24](https://huggingface.co/datasets/MicPie/unpredictable_cluster24)
* [UnpredicTable-cluster25](https://huggingface.co/datasets/MicPie/unpredictable_cluster25)
* [UnpredicTable-cluster26](https://huggingface.co/datasets/MicPie/unpredictable_cluster26)
* [UnpredicTable-cluster27](https://huggingface.co/datasets/MicPie/unpredictable_cluster27)
* [UnpredicTable-cluster28](https://huggingface.co/datasets/MicPie/unpredictable_cluster28)
* [UnpredicTable-cluster29](https://huggingface.co/datasets/MicPie/unpredictable_cluster29)
* [UnpredicTable-cluster-noise](https://huggingface.co/datasets/MicPie/unpredictable_cluster-noise)
### Supported Tasks and Leaderboards
Since the tables come from the web, the distribution of tasks and topics is very broad. The shape of our dataset is very wide, i.e., we have 1000's of tasks, while each task has only a few examples, compared to most current NLP datasets which are very deep, i.e., 10s of tasks with many examples. This implies that our dataset covers a broad range of potential tasks, e.g., multiple-choice, question-answering, table-question-answering, text-classification, etc.
The intended use of this dataset is to improve few-shot performance by fine-tuning/pre-training on our dataset.
### Languages
English
## Dataset Structure
### Data Instances
Each task is represented as a jsonline file and consists of several few-shot examples. Each example is a dictionary containing a field 'task', which identifies the task, followed by an 'input', 'options', and 'output' field. The 'input' field contains several column elements of the same row in the table, while the 'output' field is a target which represents an individual column of the same row. Each task contains several such examples which can be concatenated as a few-shot task. In the case of multiple choice classification, the 'options' field contains the possible classes that a model needs to choose from.
There are also additional meta-data fields such as 'pageTitle', 'title', 'outputColName', 'url', 'wdcFile'.
### Data Fields
'task': task identifier
'input': column elements of a specific row in the table.
'options': for multiple choice classification, it provides the options to choose from.
'output': target column element of the same row as input.
'pageTitle': the title of the page containing the table.
'outputColName': output column name
'url': url to the website containing the table
'wdcFile': WDC Web Table Corpus file
### Data Splits
The UnpredicTable datasets do not come with additional data splits.
## Dataset Creation
### Curation Rationale
Few-shot training on multi-task datasets has been demonstrated to improve language models' few-shot learning (FSL) performance on new tasks, but it is unclear which training tasks lead to effective downstream task adaptation. Few-shot learning datasets are typically produced with expensive human curation, limiting the scale and diversity of the training tasks available to study. As an alternative source of few-shot data, we automatically extract 413,299 tasks from diverse internet tables. We provide this as a research resource to investigate the relationship between training data and few-shot learning.
### Source Data
#### Initial Data Collection and Normalization
We use internet tables from the English-language Relational Subset of the WDC Web Table Corpus 2015 (WTC). The WTC dataset tables were extracted from the July 2015 Common Crawl web corpus (http://webdatacommons.org/webtables/2015/EnglishStatistics.html). The dataset contains 50,820,165 tables from 323,160 web domains. We then convert the tables into few-shot learning tasks. Please see our publication for more details on the data collection and conversion pipeline.
#### Who are the source language producers?
The dataset is extracted from [WDC Web Table Corpora](http://webdatacommons.org/webtables/).
### Annotations
#### Annotation process
Manual annotation was only carried out for the [UnpredicTable-rated-low](https://huggingface.co/datasets/MicPie/unpredictable_rated-low),
[UnpredicTable-rated-medium](https://huggingface.co/datasets/MicPie/unpredictable_rated-medium), and [UnpredicTable-rated-high](https://huggingface.co/datasets/MicPie/unpredictable_rated-high) data subsets to rate task quality. Detailed instructions of the annotation instructions can be found in our publication.
#### Who are the annotators?
Annotations were carried out by a lab assistant.
### Personal and Sensitive Information
The data was extracted from [WDC Web Table Corpora](http://webdatacommons.org/webtables/), which in turn extracted tables from the [Common Crawl](https://commoncrawl.org/). We did not filter the data in any way. Thus any user identities or otherwise sensitive information (e.g., data that reveals racial or ethnic origins, sexual orientations, religious beliefs, political opinions or union memberships, or locations; financial or health data; biometric or genetic data; forms of government identification, such as social security numbers; criminal history, etc.) might be contained in our dataset.
## Considerations for Using the Data
### Social Impact of Dataset
This dataset is intended for use as a research resource to investigate the relationship between training data and few-shot learning. As such, it contains high- and low-quality data, as well as diverse content that may be untruthful or inappropriate. Without careful investigation, it should not be used for training models that will be deployed for use in decision-critical or user-facing situations.
### Discussion of Biases
Since our dataset contains tables that are scraped from the web, it will also contain many toxic, racist, sexist, and otherwise harmful biases and texts. We have not run any analysis on the biases prevalent in our datasets. Neither have we explicitly filtered the content. This implies that a model trained on our dataset may potentially reflect harmful biases and toxic text that exist in our dataset.
### Other Known Limitations
No additional known limitations.
## Additional Information
### Dataset Curators
Jun Shern Chan, Michael Pieler, Jonathan Jao, Jérémy Scheurer, Ethan Perez
### Licensing Information
Apache 2.0
### Citation Information
```
@misc{chan2022few,
author = {Chan, Jun Shern and Pieler, Michael and Jao, Jonathan and Scheurer, Jérémy and Perez, Ethan},
title = {Few-shot Adaptation Works with UnpredicTable Data},
publisher={arXiv},
year = {2022},
url = {https://arxiv.org/abs/2208.01009}
}
```
|
false |
# Dataset Card for Brill Iconclass AI Test Set
## Table of Contents
- [Table of Contents](#table-of-contents)
- [Dataset Description](#dataset-description)
- [Dataset Summary](#dataset-summary)
- [Supported Tasks and Leaderboards](#supported-tasks-and-leaderboards)
- [Languages](#languages)
- [Dataset Structure](#dataset-structure)
- [Data Instances](#data-instances)
- [Data Fields](#data-fields)
- [Data Splits](#data-splits)
- [Dataset Creation](#dataset-creation)
- [Curation Rationale](#curation-rationale)
- [Source Data](#source-data)
- [Annotations](#annotations)
- [Personal and Sensitive Information](#personal-and-sensitive-information)
- [Considerations for Using the Data](#considerations-for-using-the-data)
- [Social Impact of Dataset](#social-impact-of-dataset)
- [Discussion of Biases](#discussion-of-biases)
- [Other Known Limitations](#other-known-limitations)
- [Additional Information](#additional-information)
- [Dataset Curators](#dataset-curators)
- [Licensing Information](#licensing-information)
- [Citation Information](#citation-information)
- [Contributions](#contributions)
## Dataset Description
- **Homepage:** [https://iconclass.org/testset/](https://iconclass.org/testset/)
- **Repository:**[https://iconclass.org/testset/](https://iconclass.org/testset/)
- **Paper:**[https://iconclass.org/testset/ICONCLASS_and_AI.pdf](https://iconclass.org/testset/ICONCLASS_and_AI.pdf)
- **Leaderboard:**
- **Point of Contact:**[info@iconclass.org](mailto:info@iconclass.org)
### Dataset Summary
> A test dataset and challenge to apply machine learning to collections described with the Iconclass classification system.
This dataset contains `87749` images with [Iconclass](https://iconclass.org/) metadata assigned to the images. The [iconclass](https://iconclass.org/) metadata classification system is intended to provide ['the comprehensive classification system for the content of images.'](https://iconclass.org/).
> Iconclass was developed in the Netherlands as a standard classification for recording collections, with the idea of assembling huge databases that will allow the retrieval of images featuring particular details, subjects or other common factors. It was developed in the 1970s and was loosely based on the Dewey Decimal System because it was meant to be used in art library card catalogs. [source](https://en.wikipedia.org/wiki/Iconclass)
The [Iconclass](https://iconclass.org)
> view of the world is subdivided in 10 main categories...An Iconclass concept consists of an alphanumeric class number (“notation”) and a corresponding content definition (“textual correlate”). An object can be tagged with as many concepts as the user sees fit. [source](https://iconclass.org/)
These ten divisions are as follows:
- 0 Abstract, Non-representational Art
- 1 Religion and Magic
- 2 Nature
- 3 Human being, Man in general
- 4 Society, Civilization, Culture
- 5 Abstract Ideas and Concepts
- 6 History
- 7 Bible
- 8 Literature
- 9 Classical Mythology and Ancient History
Within each of these divisions further subdivision's are possible (9 or 10 subdivisions). For example, under `4 Society, Civilization, Culture`, one can find:
- 41 · material aspects of daily life
- 42 · family, descendance
- 43 · recreation, amusement
- 44 · state; law; political life
- ...
See [https://iconclass.org/4](https://iconclass.org/4) for the full list.
To illustrate we can look at some example Iconclass classifications.
`41A12` represents `castle`. This classification is generated via building from the 'base' division `4`, with the following attributes:
- 4 · Society, Civilization, Culture
- 41 · material aspects of daily life
- 41A · housing
- 41A1 · civic architecture; edifices; dwellings
[source](https://iconclass.org/41A12)
The construction of Iconclass of parts makes it particularly interesting (and challenging) to tackle via Machine Learning. Whilst one could tackle this dataset as a (multi) label image classification problem, this is only one way of tackling it. For example in the above label `castle` giving the model the 'freedom' to predict only a partial label could result in the prediction `41A` i.e. housing. Whilst a very particular form of housing this prediction for 'castle' is not 'wrong' so much as it is not as precise as a human cataloguer may provide.
### Supported Tasks and Leaderboards
As discussed above this dataset could be tackled in various ways:
- as an image classification task
- as a multi-label classification task
- as an image to text task
- as a task whereby a model predicts partial sequences of the label.
This list is not exhaustive.
### Languages
This dataset doesn't have a natural language. The labels themselves can be treated as a form of language i.e. the label can be thought of as a sequence of tokens that construct a 'sentence'.
## Dataset Structure
The dataset contains a single configuration.
### Data Instances
An example instance of the dataset is as follows:
``` python
{'image': <PIL.JpegImagePlugin.JpegImageFile image mode=RGB size=390x500 at 0x7FC7FFBBD2D0>,
'label': ['31A235', '31A24(+1)', '61B(+54)', '61B:31A2212(+1)', '61B:31D14']}
```
### Data Fields
The dataset is made up of
- an image
- a sequence of Iconclass labels
### Data Splits
The dataset doesn't provide any predefined train, validation or test splits.
## Dataset Creation
> To facilitate the creation of better models in the cultural heritage domain, and promote the research on tools and techniques using Iconclass, we are making this dataset freely available. All that we ask is that any use is acknowledged and results be shared so that we can all benefit. The content is sampled from the Arkyves database. [source](https://labs.brill.com/ictestset/)
[More Information Needed]
### Curation Rationale
[More Information Needed]
### Source Data
#### Initial Data Collection and Normalization
The images are samples from the [Arkyves database](https://brill.com/view/db/arko?language=en). This collection includes images from
> from libraries and museums in many countries, including the Rijksmuseum in Amsterdam, the Netherlands Institute for Art History (RKD), the Herzog August Bibliothek in Wolfenbüttel, and the university libraries of Milan, Utrecht and Glasgow. [source](https://brill.com/view/db/arko?language=en)
[More Information Needed]
#### Who are the source language producers?
[More Information Needed]
### Annotations
#### Annotation process
The annotations are derived from the source dataset see above. Most annotations were likely created by staff with experience with the Iconclass metadata schema.
#### Who are the annotators?
[More Information Needed]
### Personal and Sensitive Information
[More Information Needed]
## Considerations for Using the Data
### Social Impact of Dataset
[More Information Needed]
### Discussion of Biases
Iconclass as a metadata standard absorbs biases from the time and place of its creation (1940s Netherlands). In particular, '32B human races, peoples; nationalities' has been subject to criticism. '32B36 'primitive', 'pre-modern' peoples' is one example of a category which we may not wish to adopt. In general, there are components of the subdivisions of `32B` which reflect a belief that race is a scientific category rather than socially constructed.
The Iconclass community is actively exploring these limitations; for example, see [Revising Iconclass section 32B human races, peoples; nationalities](https://web.archive.org/web/20210425131753/https://iconclass.org/Updating32B.pdf).
One should be aware of these limitations to Iconclass, and in particular, before deploying a model trained on this data in any production settings.
[More Information Needed]
### Other Known Limitations
[More Information Needed]
## Additional Information
### Dataset Curators
Etienne Posthumus
### Licensing Information
[CC0 1.0](https://creativecommons.org/publicdomain/zero/1.0/)
### Citation Information
```
@MISC{iconclass,
title = {Brill Iconclass AI Test Set},
author={Etienne Posthumus},
year={2020}
}
```
### Contributions
Thanks to [@davanstrien](https://github.com/davanstrien) for adding this dataset. |
true |
## Dataset Description
- **Homepage:**
- **Repository:**
- **Paper:**
- **Leaderboard:**
- **Point of Contact:**
### Dataset Summary
Published version of dataset used for paper 'Towards an automatic requirements classification in a new Spanish dataset'
### Languages
Spanish
## Dataset Structure
### Data Fields
Project: Project's Identifier from which the requirements were obtained.
Requirement: Description of the software requirement.
Final label: Label of the requirement: F (functional requirement) and NF (non-functional requirement).
## Dataset Creation
### Initial Data Collection and Normalization
This dataset was created from a collection of functional and non-functional requirements extracted from 13 final degree and 2 master’s projects carried out from the University of A Coruna. It consist in 300 functional and 89 non-funtcional requirements.
## Additional Information
### Citation Information
https://doi.org/10.5281/zenodo.6556541
|
true | # Dataset Card for lampeter_corpus
## Table of Contents
- [Dataset Description](#dataset-description)
- [Dataset Summary](#dataset-summary)
- [Supported Tasks](#supported-tasks-and-leaderboards)
- [Languages](#languages)
- [Dataset Structure](#dataset-structure)
- [Data Instances](#data-instances)
- [Data Fields](#data-instances)
- [Data Splits](#data-instances)
- [Dataset Creation](#dataset-creation)
- [Curation Rationale](#curation-rationale)
- [Source Data](#source-data)
- [Annotations](#annotations)
- [Personal and Sensitive Information](#personal-and-sensitive-information)
- [Considerations for Using the Data](#considerations-for-using-the-data)
- [Social Impact of Dataset](#social-impact-of-dataset)
- [Discussion of Biases](#discussion-of-biases)
- [Other Known Limitations](#other-known-limitations)
- [Additional Information](#additional-information)
- [Dataset Curators](#dataset-curators)
- [Licensing Information](#licensing-information)
- [Citation Information](#citation-information)
## Dataset Description
- **Homepage:** https://ota.bodleian.ox.ac.uk/repository/xmlui/handle/20.500.12024/3193
- **Repository:** [Needs More Information]
- **Paper:** [Needs More Information]
- **Leaderboard:** [Needs More Information]
- **Point of Contact:** Josef Schmied, Claudia Claridge, Rainer Siemund
### Dataset Summary
The Lampeter Corpus of Early Modern English Tracts is a collection of texts on various subject matter published between 1640 and 1740, a time that was marked by the rise of mass publication, the development of public discourse in many areas of everyday life and last but not least, the standardisation of British English. Each text belongs to one of the following genres: Law, Economy, Religion, Politics, Science, Miscellaneous
### Supported Tasks and Leaderboards
- `text-classification`: This dataset comes with dates and genre classifications for each text which can be used to finetune a model for text classification.
### Languages
The text in the dataset is British English. The associated BCP-47 code is `en-GB`
## Dataset Structure
### Data Instances
A typical data point contains an id, a text, the head of the text (which can be missing on some occasions) and the title. The two features which can be used for classification are `date`, which is the year of publication and `genre` which classifies the text into one of six broad areas.
```
{
'id': 'SciB1735',
'text': '\nI. WHEN I read your Defence of the British Mathematicians, I could not, Sir, but admire your Courage in asserting with such undoubting Assurance things so easily disproved. This to me seemed unaccountable, till I reflected on what you say (p. 32.) when upon my having appealed to every thinking Reader, whether it be possible to frame any clear Conception of Fluxions, you express yourself in the following manner, "Pray, Sir, who are those thinking Readers you ap\npeal to? Are they Geometricians, or Persons wholly ignorant of Geometry? If the former, I leave it to them: If the latter, I ask how well are they qualified to judge of the Method of Fluxions"? It must be acknowledged you seem by this Dilemma secure in the favour of one Part of your Readers, and the ignorance of the other. I am nevertheless persuaded there are fair and candid Men among the Mathematicians. And for those who are not Mathematicians, I shall endeavour so to unveil this Mystery, [TRUNCATED]',
'date': '1735',
'genre': 'Science', '
head': 'A DEFENCE OF FREE-THINKING IN Mathematics; &c.\n',
'title': 'A defence of free-thinking in mathematics [...]'
}
```
### Data Fields
The dataset contains the following fields:
- `id`: Unique identifier("string"),
- `text`: ext in the document("string"),
- `date`: Date of publication("date64"),
- `genre`: Broad classification("string"),
- `head`: Often same as title. Can be missing("string"),
- `title`: Title of document("string")
### Data Splits
Train: 120
## Dataset Creation
### Curation Rationale
The period covered by the Lampeter Corpus, 1640 to 1740, marks a crucial period in English history and the elaboration of English as a multi-purpose language. The texts selected for the corpus reflect the standardisation process of English and historical developments between the outbreak of the Civil War and the beginning of the Industrial Revolution. To meet the needs of linguists and historians alike, the Lampeter project has attempted to create a balanced corpus rather than a randomly chosen archive or collection. A balanced corpus, then, is characterised by several transparent sampling criteria.
### Source Data
#### Initial Data Collection and Normalization
The original data is selected according to the following criteria:
- Complete texts only, including dedications, prefaces, postscripts, etc.
- Texts are of varying length, ranging from c. 3,000 to c. 20,000 words.
- Each author appears only once to avoid idiosyncratic language use.
- Major literary figures of the time were excluded since their writing style can be studied in other, existing collections.
- Generally, only first editions of the texts; later editions only if changes were made by the original authors, thus ensuring the authenticity of the language.
#### Who are the source language producers?
Authors of texts between 1640-1740
### Annotations
#### Annotation process
N/A
#### Who are the annotators?
N/A
### Personal and Sensitive Information
N/A
## Considerations for Using the Data
### Social Impact of Dataset
N/A
### Discussion of Biases
The social biases of the time in terms of race, sex, gender, etc. might be encountered in this dataset
### Other Known Limitations
None
## Additional Information
### Dataset Curators
Josef Schmied, Claudia Claridge, Rainer Siemund
### Licensing Information
Creative Commons Attribution-ShareAlike 3.0 Unported (CC BY-SA 3.0)
### Citation Information
University of Oxford, The Lampeter Corpus of Early Modern English Tracts, Oxford Text Archive, http://hdl.handle.net/20.500.12024/3193. |
false |
This is the text style transfer datasets collected by TextBox, including:
- GYAFC Entertainment & Music (gyafc_em).
- GYAFC Family & Relationships (gyafc_fr).
The detail and leaderboard of each dataset can be found in [TextBox page](https://github.com/RUCAIBox/TextBox#dataset). |
true |
# Dataset Card for Law Stack Exchange Dataset
## Dataset Description
- **Paper: [Parameter-Efficient Legal Domain Adaptation](https://aclanthology.org/2022.nllp-1.10/)**
- **Point of Contact: jxl@queensu.ca**
### Dataset Summary
Dataset from the Law Stack Exchange, as used in "Parameter-Efficient Legal Domain Adaptation".
### Citation Information
```
@inproceedings{li-etal-2022-parameter,
title = "Parameter-Efficient Legal Domain Adaptation",
author = "Li, Jonathan and
Bhambhoria, Rohan and
Zhu, Xiaodan",
booktitle = "Proceedings of the Natural Legal Language Processing Workshop 2022",
month = dec,
year = "2022",
address = "Abu Dhabi, United Arab Emirates (Hybrid)",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2022.nllp-1.10",
pages = "119--129",
}
``` |
false |
# Dataset Card for OLM August 2022 Common Crawl
Cleaned and deduplicated pretraining dataset, created with the OLM repo [here](https://github.com/huggingface/olm-datasets) from 20% of the August 2022 Common Crawl snapshot.
Note: `last_modified_timestamp` was parsed from whatever a website returned in it's `Last-Modified` header; there are likely a small number of outliers that are incorrect, so we recommend removing the outliers before doing statistics with `last_modified_timestamp`. |
false |
# HumanEval-Infilling
## Dataset Description
- **Repository:** https://github.com/openai/human-eval-infilling
- **Paper:** https://arxiv.org/pdf/2207.14255
## Dataset Summary
[HumanEval-Infilling](https://github.com/openai/human-eval-infilling) is a benchmark for infilling tasks, derived from [HumanEval](https://huggingface.co/datasets/openai_humaneval) benchmark for the evaluation of code generation models.
## Dataset Structure
To load the dataset you need to specify a subset. By default `HumanEval-SingleLineInfilling` is loaded.
```python
from datasets import load_dataset
ds = load_dataset("humaneval_infilling", "HumanEval-RandomSpanInfilling")
DatasetDict({
test: Dataset({
features: ['task_id', 'entry_point', 'prompt', 'suffix', 'canonical_solution', 'test'],
num_rows: 1640
})
})
```
## Subsets
This dataset has 4 subsets: HumanEval-MultiLineInfilling, HumanEval-SingleLineInfilling, HumanEval-RandomSpanInfilling, HumanEval-RandomSpanInfillingLight.
The single-line, multi-line, random span infilling and its light version have 1033, 5815, 1640 and 164 tasks, respectively.
## Citation
```
@article{bavarian2022efficient,
title={Efficient Training of Language Models to Fill in the Middle},
author={Bavarian, Mohammad and Jun, Heewoo and Tezak, Nikolas and Schulman, John and McLeavey, Christine and Tworek, Jerry and Chen, Mark},
journal={arXiv preprint arXiv:2207.14255},
year={2022}
}
``` |
false |
# Dataset Card for OLM May 2022 Common Crawl
Cleaned and deduplicated pretraining dataset, created with the OLM repo [here](https://github.com/huggingface/olm-datasets) from 15% of the May 2022 Common Crawl snapshot.
Note: `last_modified_timestamp` was parsed from whatever a website returned in it's `Last-Modified` header; there are likely a small number of outliers that are incorrect, so we recommend removing the outliers before doing statistics with `last_modified_timestamp`. |
false |
# Dataset Card for OLM June/July 2022 Common Crawl
Cleaned and deduplicated pretraining dataset, created with the OLM repo [here](https://github.com/huggingface/olm-datasets) from 16% of the June/July 2022 Common Crawl snapshot.
Note: `last_modified_timestamp` was parsed from whatever a website returned in it's `Last-Modified` header; there are likely a small number of outliers that are incorrect, so we recommend removing the outliers before doing statistics with `last_modified_timestamp`. |
false | # Dataset Card for Rock Glacier Detection
## Table of Contents
- [Table of Contents](#table-of-contents)
- [Dataset Description](#dataset-description)
- [Dataset Summary](#dataset-summary)
- [Supported Tasks and Leaderboards](#supported-tasks-and-leaderboards)
- [Languages](#languages)
- [Dataset Structure](#dataset-structure)
- [Data Instances](#data-instances)
- [Data Fields](#data-fields)
- [Data Splits](#data-splits)
- [Dataset Creation](#dataset-creation)
- [Curation Rationale](#curation-rationale)
- [Source Data](#source-data)
- [Annotations](#annotations)
- [Personal and Sensitive Information](#personal-and-sensitive-information)
- [Considerations for Using the Data](#considerations-for-using-the-data)
- [Social Impact of Dataset](#social-impact-of-dataset)
- [Discussion of Biases](#discussion-of-biases)
- [Other Known Limitations](#other-known-limitations)
- [Additional Information](#additional-information)
- [Dataset Curators](#dataset-curators)
- [Licensing Information](#licensing-information)
- [Citation Information](#citation-information)
- [Contributions](#contributions)
## Dataset Description
- **Homepage:** [RockGlacier Homepage](https://github.com/alcazar90/rock-glacier-detection)
- **Repository:** [alcazar90/rock-glacier-detection](https://github.com/alcazar90/rock-glacier-detection)
- **Paper:** N/A
- **Leaderboard:** N/A
- **Point of Contact:** N/A
### Dataset Summary
Rock Glacier Detection dataset with satelital images of rock glaciers in the Chilean Andes.
### Supported Tasks and Leaderboards
- `image-classification`: Based on a satelitel images (from sentinel2), the goal of this task is to predict a rock glacier in the geographic area, if there any.
- `image-segmentation`: ...
### Languages
Spanish
## Dataset Structure
### Data Instances
A sample from the image-classification training set is provided below:
```
df = load_dataset("alkzar90/rock-glacier-dataset", name="image-classification")
df["train"][666]
> {'image': <PIL.PngImagePlugin.PngImageFile image mode=RGBA size=128x128 at 0x7FB2EC58C6D0>,
'labels': 0,
'path': 'train/cordillera/1512.png'
}
```
A sample from the image-segmentation training set is provided below:
```
df = load_dataset("alkzar90/rock-glacier-dataset", name="image-segmentation")
df["train"][666]
> {'image': <PIL.PngImagePlugin.PngImageFile image mode=RGBA size=128x128 at 0x7FB2EB7C1160>,
'masks': <PIL.PngImagePlugin.PngImageFile image mode=RGBA size=128x128 at 0x7FB2EC5A08E0>,
'path': 'train/cordillera/1512.png'}
```
### Data Fields
The data instances have the following fields:
- `image`: A `PIL.Image.Image` object containing the image. Note that when accessing the image column: `dataset[0]["image"]` the image file is automatically decoded. Decoding of a large number of image files might take a significant amount of time. Thus it is important to first query the sample index before the `"image"` column, *i.e.* `dataset[0]["image"]` should **always** be preferred over `dataset["image"][0]`.
- `labels`: an `int` classification label.
Class Label Mappings:
```json
{
"cordillera": 0
"glaciar": 1,
}
```
### Data Splits
| |train|validation| test|
|-------------|----:|---------:|-----:|
|# of examples|7875 |1125 |2700 |
## Dataset Creation
### Curation Rationale
[More Information Needed]
### Source Data
#### Initial Data Collection and Normalization
[More Information Needed]
#### Who are the source language producers?
[More Information Needed]
### Annotations
#### Annotation process
[More Information Needed]
#### Who are the annotators?
[More Information Needed]
### Personal and Sensitive Information
[More Information Needed]
## Considerations for Using the Data
### Social Impact of Dataset
[More Information Needed]
### Discussion of Biases
[More Information Needed]
### Other Known Limitations
[More Information Needed]
## Additional Information
### Dataset Curators
[More Information Needed]
### Licensing Information
[More Information Needed]
### Citation Information
```
@ONLINE {rock-glacier-dataset,
author="CMM - Glaciares (UChile)",
title="Rock Glacier Dataset",
month="October",
year="2022",
url="https://github.com/alcazar90/rock-glacier-detection"
}
```
### Contributions
Thanks to...
|
false |
# Dataset Card for KsponSpeech
## Table of Contents
- [Table of Contents](#table-of-contents)
- [Dataset Description](#dataset-description)
- [Dataset Summary](#dataset-summary)
- [Supported Tasks and Leaderboards](#supported-tasks-and-leaderboards)
- [Languages](#languages)
- [Dataset Structure](#dataset-structure)
- [Data Instances](#data-instances)
- [Data Fields](#data-fields)
- [Data Splits](#data-splits)
- [Dataset Creation](#dataset-creation)
- [Curation Rationale](#curation-rationale)
- [Source Data](#source-data)
- [Annotations](#annotations)
- [Personal and Sensitive Information](#personal-and-sensitive-information)
- [Considerations for Using the Data](#considerations-for-using-the-data)
- [Social Impact of Dataset](#social-impact-of-dataset)
- [Discussion of Biases](#discussion-of-biases)
- [Other Known Limitations](#other-known-limitations)
- [Additional Information](#additional-information)
- [Dataset Curators](#dataset-curators)
- [Licensing Information](#licensing-information)
- [Citation Information](#citation-information)
- [Contributions](#contributions)
## Dataset Description
- **Homepage:** [AIHub](https://www.aihub.or.kr/aihubdata/data/view.do?currMenu=115&topMenu=100&aihubDataSe=realm&dataSetSn=123)
- **Repository:**
- **Paper:** [KsponSpeech](https://www.mdpi.com/2076-3417/10/19/6936)
- **Leaderboard:**
- **Point of Contact:**
### Dataset Summary
This corpus contains 969 h of general open-domain dialog utterances, spoken by about 2000 native Korean speakers in a clean environment. All data were constructed by recording the dialogue of two people freely conversing on a variety of topics and manually transcribing the utterances. The transcription provides a dual transcription consisting of orthography and pronunciation, and disfluency tags for spontaneity of speech, such as filler words, repeated words, and word fragments. This paper also presents the baseline performance of an end-to-end speech recognition model trained with KsponSpeech. In addition, we investigated the performance of standard end-to-end architectures and the number of sub-word units suitable for Korean. We investigated issues that should be considered in spontaneous speech recognition in Korean. KsponSpeech is publicly available on an open data hub site of the Korea government.
### Supported Tasks and Leaderboards
[More Information Needed]
### Languages
Korean
## Dataset Structure
### Data Instances
```json
{
'id': 'KsponSpeech_E00001',
'audio': {'path': None,
'array': array([0.0010376 , 0.00085449, 0.00097656, ..., 0.00250244, 0.0022583 ,
0.00253296]),
'sampling_rate': 16000},
'text': '어 일단은 억지로 과장해서 이렇게 하는 것보다 진실된 마음으로 이걸 어떻게 전달할 수 있을까 공감을 시킬 수 있을까 해서 좀'
}
```
### Data Fields
- audio: A dictionary containing the path to the downloaded audio file, the decoded audio array, and the sampling rate. Note that when accessing the audio column: `dataset[0]["audio"]` the audio file is automatically decoded and resampled to `dataset.features["audio"].sampling_rate`. Decoding and resampling of a large number of audio files might take a significant amount of time. Thus it is important to first query the sample index before the `"audio"` column, *i.e.* `dataset[0]["audio"]` should **always** be preferred over `dataset["audio"][0]`.
- text: the transcription of the audio file.
- id: unique id of the data sample.
### Data Splits
| | Train | Valid | eval.clean | eval.other |
| ----- | ------ | ----- | ---- | ---- |
| #samples | 620000 | 2545 | 3000 | 3000 |
## Dataset Creation
### Curation Rationale
[More Information Needed]
### Source Data
#### Initial Data Collection and Normalization
[More Information Needed]
#### Who are the source language producers?
[More Information Needed]
### Annotations
#### Annotation process
[More Information Needed]
#### Who are the annotators?
[More Information Needed]
### Personal and Sensitive Information
[More Information Needed]
## Considerations for Using the Data
### Social Impact of Dataset
[More Information Needed]
### Discussion of Biases
[More Information Needed]
### Other Known Limitations
[More Information Needed]
## Additional Information
### Dataset Curators
[More Information Needed]
### Licensing Information
[More Information Needed]
### Citation Information
```bibtex
@Article{app10196936,
AUTHOR = {Bang, Jeong-Uk and Yun, Seung and Kim, Seung-Hi and Choi, Mu-Yeol and Lee, Min-Kyu and Kim, Yeo-Jeong and Kim, Dong-Hyun and Park, Jun and Lee, Young-Jik and Kim, Sang-Hun},
TITLE = {KsponSpeech: Korean Spontaneous Speech Corpus for Automatic Speech Recognition},
JOURNAL = {Applied Sciences},
VOLUME = {10},
YEAR = {2020},
NUMBER = {19},
ARTICLE-NUMBER = {6936},
URL = {https://www.mdpi.com/2076-3417/10/19/6936},
ISSN = {2076-3417},
ABSTRACT = {This paper introduces a large-scale spontaneous speech corpus of Korean, named KsponSpeech. This corpus contains 969 h of general open-domain dialog utterances, spoken by about 2000 native Korean speakers in a clean environment. All data were constructed by recording the dialogue of two people freely conversing on a variety of topics and manually transcribing the utterances. The transcription provides a dual transcription consisting of orthography and pronunciation, and disfluency tags for spontaneity of speech, such as filler words, repeated words, and word fragments. This paper also presents the baseline performance of an end-to-end speech recognition model trained with KsponSpeech. In addition, we investigated the performance of standard end-to-end architectures and the number of sub-word units suitable for Korean. We investigated issues that should be considered in spontaneous speech recognition in Korean. KsponSpeech is publicly available on an open data hub site of the Korea government.},
DOI = {10.3390/app10196936}
}
```
|
false |
# Dataset Card for Czech Named Entity Corpus 2.0
## Dataset Description
The dataset contains Czech sentences and annotated named entities. Total number of sentences is around 9,000 and total number of entities is around 34,000. (Total means train + validation + test)
## Dataset Features
Each sample contains:
- `text`: source sentence
- `entities`: list of selected entities. Each entity contains:
- `category_id`: string identifier of the entity category
- `category_str`: human-friendly category name in Czech (verbalizer)
- `start`: index on which the entity starts in the source sentence
- `end`: index on which the entity ends in the source sentence
- `content`: entity content, it was created as `text[start:end]`
- `entity_id`: unique entity string identifier
- `parent_id`: If entity was selected inside another entity (e.g. house number inside address), `parent_id` is the identifier of the parent entity. None otherwise.
The `entity_id` field was checked to be globally unique (across data samples and dataset splits.)
## Entity categories
The list of the recognized entities (`category_id`, `category_str` pairs):
```python3
{
'A': 'číslo v adrese / kontaktním údaji',
'ah': 'číslo domu',
'at': 'telefonní číslo / fax',
'az': 'PSČ (poštovní směrovací číslo)',
'C': 'reference/bibliografie',
'f': 'cizí výraz',
'g_': 'geografický název - jiný',
'gc': 'stát/země',
'gh': 'jméno vodstva',
'gl': 'přírodní oblast/útvar',
'gq': 'městská čtvrť',
'gr': 'území',
'gs': 'ulice/náměstí',
'gt': 'kontinent',
'gu': 'město/zámek',
'i_': 'instituce - jiná',
'ia': 'konference/soutěž',
'ic': 'kulturní/vzdělávací/vědecká instituce',
'if': 'komerční instituce',
'io': 'vládní/politická instituce',
'me': 'emailová adresa',
'mi': 'URL / internetový odkaz',
'mn': 'časopis',
'ms': 'radio/televizní stanice',
'n_': 'číselný výraz - jiný',
'na': 'věk',
'nb': 'číslo stránky/kapitoly/sekce/objektu',
'nc': 'množství/počet',
'ni': 'číslo položky',
'no': 'pořadí',
'ns': 'sportovní skóre',
'o_': 'artefakt - jiný',
'oa': 'umělecké dílo / kulturní artefakt',
'oe': 'jednotka',
'om': 'měna',
'op': 'produkt/výrobek',
'or': 'zákon/směrnice/listina',
'P': 'celé jméno',
'p_': 'jméno - jiné',
'pc': 'národnost',
'pd': '(akademický) titul',
'pf': 'křestní jméno',
'pm': 'prostřední jméno',
'pp': 'mýtická/historická postava',
'ps': 'příjmení',
's': 'zkratka',
'T': 'čas/datum',
'td': 'den',
'tf': 'svátky',
'th': 'hodiny/minuty',
'tm': 'měsíc',
'ty': 'rok',
}
```
## Dataset Source
The dataset is a preprocessed adaptation of existing CNEC 2.0 dataset [project info](https://ufal.mff.cuni.cz/cnec/cnec2.0), [link to data](https://lindat.mff.cuni.cz/repository/xmlui/handle/11858/00-097C-0000-0023-1B22-8). This adaptation contains (almost) same data, but converted to a convenient format. In addition, we inspected and decided to remove entity categories: `?`, `segm`, `cap`, `lower`, `upper`, which were either undocumented and/or carried little semantic meaning.
The category names (verbalizers) are not in the original dataset. They were added by a Czech native speaker using the available [documentation](https://ufal.mff.cuni.cz/cnec/cnec2.0) and by looking at several occurrences in the data.
## Citation
Cite authors of the [original dataset](https://lindat.mff.cuni.cz/repository/xmlui/handle/11858/00-097C-0000-0023-1B22-8):
```bibtex
@misc{11858/00-097C-0000-0023-1B22-8,
title = {Czech Named Entity Corpus 2.0},
author = {{\v S}ev{\v c}{\'{\i}}kov{\'a}, Magda and {\v Z}abokrtsk{\'y}, Zden{\v e}k and Strakov{\'a}, Jana and Straka, Milan},
url = {http://hdl.handle.net/11858/00-097C-0000-0023-1B22-8},
note = {{LINDAT}/{CLARIAH}-{CZ} digital library at the Institute of Formal and Applied Linguistics ({{\'U}FAL}), Faculty of Mathematics and Physics, Charles University},
copyright = {Attribution-{NonCommercial}-{ShareAlike} 3.0 Unported ({CC} {BY}-{NC}-{SA} 3.0)},
year = {2014}
}
```
|
true | # Dataset Card for The Harvard USPTO Patent Dataset (HUPD)
## Dataset Description
- **Homepage:** [https://patentdataset.org/](https://patentdataset.org/)
- **Repository:** [HUPD GitHub repository](https://github.com/suzgunmirac/hupd)
- **Paper:** [HUPD arXiv Submission](https://arxiv.org/abs/2207.04043)
- **Point of Contact:** Mirac Suzgun
### Dataset Summary
The Harvard USPTO Dataset (HUPD) is a large-scale, well-structured, and multi-purpose corpus of English-language utility patent applications filed to the United States Patent and Trademark Office (USPTO) between January 2004 and December 2018.
### Experiments and Tasks Considered in the Paper
- **Patent Acceptance Prediction**: Given a section of a patent application (in particular, the abstract, claims, or description), predict whether the application will be accepted by the USPTO.
- **Automated Subject (IPC/CPC) Classification**: Predict the primary IPC or CPC code of a patent application given (some subset of) the text of the application.
- **Language Modeling**: Masked/autoregressive language modeling on the claims and description sections of patent applications.
- **Abstractive Summarization**: Given the claims or claims section of a patent application, generate the abstract.
### Languages
The dataset contains English text only.
### Domain
Patents (intellectual property).
### Dataset Curators
The dataset was created by Mirac Suzgun, Luke Melas-Kyriazi, Suproteem K. Sarkar, Scott Duke Kominers, and Stuart M. Shieber.
## Dataset Structure
Each patent application is defined by a distinct JSON file, named after its application number, and includes information about
the application and publication numbers,
title,
decision status,
filing and publication dates,
primary and secondary classification codes,
inventor(s),
examiner,
attorney,
abstract,
claims,
background,
summary, and
full description of the proposed invention, among other fields. There are also supplementary variables, such as the small-entity indicator (which denotes whether the applicant is considered to be a small entity by the USPTO) and the foreign-filing indicator (which denotes whether the application was originally filed in a foreign country).
In total, there are 34 data fields for each application. A full list of data fields used in the dataset is listed in the next section.
### Data Instances
Each patent application in our patent dataset is defined by a distinct JSON file (e.g., ``8914308.json``), named after its unique application number. The format of the JSON files is as follows:
```python
{
"application_number": "...",
"publication_number": "...",
"title": "...",
"decision": "...",
"date_produced": "...",
"date_published": "...",
"main_cpc_label": "...",
"cpc_labels": ["...", "...", "..."],
"main_ipcr_label": "...",
"ipcr_labels": ["...", "...", "..."],
"patent_number": "...",
"filing_date": "...",
"patent_issue_date": "...",
"abandon_date": "...",
"uspc_class": "...",
"uspc_subclass": "...",
"examiner_id": "...",
"examiner_name_last": "...",
"examiner_name_first": "...",
"examiner_name_middle": "...",
"inventor_list": [
{
"inventor_name_last": "...",
"inventor_name_first": "...",
"inventor_city": "...",
"inventor_state": "...",
"inventor_country": "..."
}
],
"abstract": "...",
"claims": "...",
"background": "...",
"summary": "...",
"full_description": "..."
}
```
## Usage
### Loading the Dataset
#### Sample (January 2016 Subset)
The following command can be used to load the `sample` version of the dataset, which contains all the patent applications that were filed to the USPTO during the month of January in 2016. This small subset of the dataset can be used for debugging and exploration purposes.
```python
from datasets import load_dataset
dataset_dict = load_dataset('HUPD/hupd',
name='sample',
data_files="https://huggingface.co/datasets/HUPD/hupd/blob/main/hupd_metadata_2022-02-22.feather",
icpr_label=None,
train_filing_start_date='2016-01-01',
train_filing_end_date='2016-01-21',
val_filing_start_date='2016-01-22',
val_filing_end_date='2016-01-31',
)
```
#### Full Dataset
If you would like to use the **full** version of the dataset, please make sure that change the `name` field from `sample` to `all`, specify the training and validation start and end dates carefully, and set `force_extract` to be `True` (so that you would only untar the files that you are interested in and not squander your disk storage space). In the following example, for instance, we set the training set year range to be [2011, 2016] (inclusive) and the validation set year range to be 2017.
```python
from datasets import load_dataset
dataset_dict = load_dataset('HUPD/hupd',
name='all',
data_files="https://huggingface.co/datasets/HUPD/hupd/blob/main/hupd_metadata_2022-02-22.feather",
icpr_label=None,
force_extract=True,
train_filing_start_date='2011-01-01',
train_filing_end_date='2016-12-31',
val_filing_start_date='2017-01-01',
val_filing_end_date='2017-12-31',
)
```
### Google Colab Notebook
You can also use the following Google Colab notebooks to explore HUPD.
- [](https://colab.research.google.com/drive/1_ZsI7WFTsEO0iu_0g3BLTkIkOUqPzCET?usp=sharing)[ HUPD Examples: Loading the Dataset](https://colab.research.google.com/drive/1_ZsI7WFTsEO0iu_0g3BLTkIkOUqPzCET?usp=sharing)
- [](https://colab.research.google.com/drive/1TzDDCDt368cUErH86Zc_P2aw9bXaaZy1?usp=sharing)[ HUPD Examples: Loading HUPD By Using HuggingFace's Libraries](https://colab.research.google.com/drive/1TzDDCDt368cUErH86Zc_P2aw9bXaaZy1?usp=sharing)
- [](https://colab.research.google.com/drive/1TzDDCDt368cUErH86Zc_P2aw9bXaaZy1?usp=sharing)[ HUPD Examples: Using the HUPD DistilRoBERTa Model](https://colab.research.google.com/drive/11t69BWcAVXndQxAOCpKaGkKkEYJSfydT?usp=sharing)
- [](https://colab.research.google.com/drive/1TzDDCDt368cUErH86Zc_P2aw9bXaaZy1?usp=sharing)[ HUPD Examples: Using the HUPD T5-Small Summarization Model](https://colab.research.google.com/drive/1VkCtrRIryzev_ixDjmJcfJNK-q6Vx24y?usp=sharing)
## Dataset Creation
### Source Data
HUPD synthesizes multiple data sources from the USPTO: While the full patent application texts were obtained from the USPTO Bulk Data Storage System (Patent Application Data/XML Versions 4.0, 4.1, 4.2, 4.3, 4.4 ICE, as well as Version 1.5) as XML files, the bibliographic filing metadata were obtained from the USPTO Patent Examination Research Dataset (in February, 2021).
### Annotations
Beyond our patent decision label, for which construction details are provided in the paper, the dataset does not contain any human-written or computer-generated annotations beyond those produced by patent applicants or the USPTO.
### Data Shift
A major feature of HUPD is its structure, which allows it to demonstrate the evolution of concepts over time. As we illustrate in the paper, the criteria for patent acceptance evolve over time at different rates, depending on category. We believe this is an important feature of the dataset, not only because of the social scientific questions it raises, but also because it facilitates research on models that can accommodate concept shift in a real-world setting.
### Personal and Sensitive Information
The dataset contains information about the inventor(s) and examiner of each patent application. These details are, however, already in the public domain and available on the USPTO's Patent Application Information Retrieval (PAIR) system, as well as on Google Patents and PatentsView.
### Social Impact of the Dataset
The authors of the dataset hope that HUPD will have a positive social impact on the ML/NLP and Econ/IP communities. They discuss these considerations in more detail in [the paper](https://arxiv.org/abs/2207.04043).
### Impact on Underserved Communities and Discussion of Biases
The dataset contains patent applications in English, a language with heavy attention from the NLP community. However, innovation is spread across many languages, cultures, and communities that are not reflected in this dataset. HUPD is thus not representative of all kinds of innovation. Furthermore, patent applications require a fixed cost to draft and file and are not accessible to everyone. One goal of this dataset is to spur research that reduces the cost of drafting applications, potentially allowing for more people to seek intellectual property protection for their innovations.
### Discussion of Biases
Section 4 of [the HUPD paper](https://arxiv.org/abs/2207.04043) provides an examination of the dataset for potential biases. It shows, among other things, that female inventors are notably underrepresented in the U.S. patenting system, that small and micro entities (e.g., independent inventors, small companies, non-profit organizations) are less likely to have positive outcomes in patent obtaining than large entities (e.g., companies with more than 500 employees), and that patent filing and acceptance rates are not uniformly distributed across the US. Our empirical findings suggest that any study focusing on the acceptance prediction task, especially if it is using the inventor information or the small-entity indicator as part of the input, should be aware of the the potential biases present in the dataset and interpret their results carefully in light of those biases.
- Please refer to Section 4 and Section D for an in-depth discussion of potential biases embedded in the dataset.
### Licensing Information
HUPD is released under the CreativeCommons Attribution-NonCommercial-ShareAlike 4.0 International.
### Citation Information
```
@article{suzgun2022hupd,
title={The Harvard USPTO Patent Dataset: A Large-Scale, Well-Structured, and Multi-Purpose Corpus of Patent Applications},
author={Suzgun, Mirac and Melas-Kyriazi, Luke and Sarkar, Suproteem K. and Kominers, Scott Duke and Shieber, Stuart M.},
year={2022},
publisher={arXiv preprint arXiv:2207.04043},
url={https://arxiv.org/abs/2207.04043},
``` |
false | # AutoTrain Dataset for project: auto-arabic-summarization
## Dataset Description
This dataset has been automatically processed by AutoTrain for project auto-arabic-summarization.
### Languages
The BCP-47 code for the dataset's language is unk.
## Dataset Structure
### Data Instances
A sample from this dataset looks as follows:
```json
[
{
"text": "\u0627\u0643\u062f \u0648\u0632\u064a\u0631 \u0627\u0644\u0635\u0646\u0627\u0639\u0647 \u0648\u0627\u0644\u0637\u0627\u0642\u0647 \u0648\u0627\u0644\u0645\u0646\u0627\u062c\u0645 \u0632\u0643\u0631\u064a\u0627 \u062d\u0645\u062f \u0627\u0646\u0647 \u062a\u0645 \u0627\u0644\u064a\u0648\u0645 \u0627\u0644\u062e\u0645\u064a\u0633 \u062e\u0644\u0627\u0644 \u062c\u0644\u0633\u0647 \u0627\u0644\u062a\u0627\u0645\u062a \u0628\u0627\u0644\u0639\u0627\u0635\u0645\u0647 \u0648\u0632\u064a\u0631 \u0627\u0644\u0637\u0627\u0642\u0647 \u0627\u0644\u062c\u0632\u0627\u0626\u064a \u0635\u0627\u0644\u062d \u062e\u0628\u0631\u064a \u0628\u062e\u0635\u0648\u0635 \u0627\u0634\u063a\u0627\u0644 \u0627\u0644\u0644\u062c\u0646\u0647 \u0627\u0644\u062a\u0648\u0646\u0633\u064a\u0647 \u0627\u0644\u062c\u0632\u0627\u0626\u0631\u064a\u0647 \u0645\u062c\u0627\u0644 \u0627\u0644\u0637\u0627\u0642\u0647 \u0644\u062a\u0642\u064a\u064a\u0645 \u0645\u062f\u0649 \u062a\u0637\u0628\u064a\u0642 \u0627\u0644\u0628\u0631\u0627\u0645\u062c \u0627\u0644\u0645\u062a\u0641\u0642 \u0639\u0644\u064a\u0647\u0627 \u062e\u0628\u0631\u0627\u0621 \u0627\u0644\u0628\u0644\u062f\u064a\u0646 \u0627\u0644\u0627\u062a\u0641\u0627\u0642 \u062a\u0632\u0648\u064a\u062f \u0627\u0644\u0645\u0646\u0627\u0637\u0642 \u0627\u0644\u062d\u062f\u0648\u062f\u064a\u0647 \u0627\u0644\u062a\u0648\u0646\u0633\u064a\u0647 \u0628\u0627\u0644\u0643\u0645\u064a\u0627\u062a \u0627\u0644\u0643\u0627\u0641\u064a\u0647 \u0642\u0648\u0627\u0631\u064a\u0631 \u0627\u0644\u063a\u0627\u0632 \u0627\u0644\u0645\u0646\u0632\u0644\u064a \u062a\u0642\u062f\u0631 \u0628\u062d\u0648\u0627\u0644\u064a \u0637\u0646 \u0627\u0644\u0642\u0648\u0627\u0631\u064a\u0631 \u0648\u0627\u0636\u0627\u0641 \u062d\u0645\u062f \u0627\u0646\u0647 \u0627\u0644\u0646\u0642\u0627\u0637 \u062a\u0645 \u0627\u0644\u0627\u062a\u0641\u0627\u0642 \u0628\u0634\u0627\u0646\u0647\u0627 \u062c\u0644\u0633\u0647 \u0627\u0644\u064a\u0648\u0645 \u062a\u0632\u0648\u064a\u062f \u0627\u0644\u0633\u0648\u0642 \u0627\u0644\u062a\u0648\u0646\u0633\u064a\u0647 \u0628\u0627\u0644\u063a\u0627\u0632 \u0627\u0644\u0637\u0628\u064a\u0639\u064a \u0639\u0628\u0631 \u0627\u0644\u0627\u0646\u0627\u0628\u064a\u0628 \u0648\u062a\u0632\u0648\u064a\u062f \u0627\u0644\u0645\u0646\u0627\u0637\u0642 \u0628\u0627\u0644\u0628\u062a\u0631\u0648\u0644 \u0627\u0644\u0645\u0633\u0627\u0644 \u0627\u0636\u0627\u0641\u0647 \u0627\u0644\u0649 \u062f\u0639\u0645 \u0627\u0644\u062a\u0639\u0627\u0648\u0646 \u0627\u0644\u0645\u062c\u0627\u0644 \u0627\u0644\u062a\u062c\u0627\u0631\u064a \u062a\u0645 \u0627\u0645\u0636\u0627\u0621 \u0645\u0630\u0643\u0631\u0647 \u062a\u0641\u0627\u0647\u0645 \u0639\u0642\u062f \u0644\u062a\u0643\u0648\u064a\u0646 \u062a\u0642\u0646\u0646\u064a\u0646 \u062a\u0648\u0646\u0633\u064a\u064a\u0646 \u0627\u0644\u062c\u0632\u0627\u0626\u0631",
"target": "\u0643\u0645\u0627 \u062a\u0645 \u0627\u0645\u0636\u0627\u0621 \u0645\u0630\u0643\u0631\u0629 \u062a\u0641\u0627\u0647\u0645 \u0639\u0642\u062f \u0644\u062a\u0643\u0648\u064a\u0646 \u062a\u0642\u0646\u0646\u064a\u0646 \u062a\u0648\u0646\u0633\u064a\u064a\u0646 \u0641\u064a \u0627\u0644\u062c\u0632\u0627\u0626\u0631 ."
},
{
"text": "\u0642\u0627\u0644 \u0627\u0644\u0648\u0632\u064a\u0631 \u0627\u0644\u0627\u0648\u0644 \u0627\u0644\u062c\u0632\u0627\u0626\u0631\u064a \u0639\u0628\u062f \u0627\u0644\u0645\u0627\u0644\u0643 \u0633\u0644\u0627\u0644 \u0627\u062b\u0631 \u0644\u0642\u0627\u0621 \u062c\u0645\u0639\u0647 \u0628\u0631\u0626\u064a\u0633 \u0645\u062c\u0644\u0633 \u0646\u0648\u0627\u0628 \u0627\u0644\u0634\u0639\u0628 \u0645\u062d\u0645\u062f \u0627\u0644\u0646\u0627\u0635\u0631 \u0627\u0644\u0639\u0644\u0627\u0642\u0627\u062a \u0627\u0644\u062b\u0646\u0627\u0626\u064a\u0647 \u0627\u0644\u0628\u0644\u062f\u064a\u0646 \u0645\u0645\u064a\u0632\u0647 \u0648\u0633\u062a\u0643\u0648\u0646 \u0627\u062d\u0633\u0646 \u062e\u0644\u0627\u0644 \u0627\u0644\u0641\u062a\u0631\u0647 \u0627\u0644\u0642\u0627\u062f\u0645\u0647 \u0648\u0627\u0636\u0627\u0641 \u062a\u0635\u0631\u064a\u062d \u0644\u0645\u0631\u0627\u0633\u0644 \u0627\u0644\u062c\u0648\u0647\u0631\u0647 \u0627\u0641 \u0627\u0645 \u0627\u0646\u0647 \u0639\u0627\u0647\u062f \u0631\u0626\u064a\u0633 \u0627\u0644\u0645\u062c\u0644\u0633 \u0628\u0627\u0644\u0645\u062d\u0627\u0641\u0638\u0647 \u0645\u062a\u0627\u0646\u0647 \u0627\u0644\u0639\u0644\u0627\u0642\u0647 \u0627\u0644\u0628\u0644\u062f\u064a\u0646 \u0648\u0645\u0648\u0627\u0635\u0644\u0647 \u0627\u0644\u062a\u0642\u062f\u0645 \u0648\u0627\u0644\u0639\u0645\u0644 \u0645\u0639\u0627 \u0648\u0627\u0648\u0636\u062d \u0639\u0628\u062f \u0627\u0644\u0645\u0627\u0644\u0643 \u0633\u0644\u0627\u0644 \u0645\u062d\u0645\u062f \u0627\u0644\u0646\u0627\u0635\u0631 \u0627\u0628\u062f\u0649 \u062f\u0639\u0645\u0647 \u0644\u0644\u0645\u0646\u0647\u062c \u062a\u0646\u062a\u0647\u062c\u0647 \u0627\u0644\u062c\u0632\u0627\u0626\u0631 \u0648\u0639\u0645\u0644\u0647\u0627 \u0648\u064a\u0627\u062a\u064a \u0627\u062c\u062a\u0645\u0627\u0639 \u0627\u0644\u0648\u0632\u064a\u0631 \u0627\u0644\u0627\u0648\u0644 \u0627\u0644\u062c\u0632\u0627\u0626\u0631\u064a \u0628\u0631\u0626\u064a\u0633 \u0627\u0644\u0645\u062c\u0644\u0633 \u0647\u0627\u0645\u0634 \u0632\u064a\u0627\u0631\u0647 \u0639\u0645\u0644 \u0627\u062f\u0627\u0647\u0627 \u0627\u0644\u064a\u0648\u0645 \u0627\u0644\u062e\u0645\u064a\u0633 \u062a\u0648\u0646\u0633 \u062a\u0631\u0627\u0633 \u062e\u0644\u0627\u0644\u0647\u0627 \u0627\u0634\u063a\u0627\u0644 \u0627\u0644\u062f\u0648\u0631\u0647 \u0627\u0644 \u0644\u0644\u062c\u0646\u0647 \u0627\u0644\u0645\u062e\u062a\u0644\u0637\u0647 \u0627\u0644\u0639\u0644\u064a\u0627 \u0627\u0644\u062a\u0648\u0646\u0633\u064a\u0647 \u0627\u0644\u062c\u0632\u0627\u0626\u0631\u064a\u0647 \u0631\u0641\u0642\u0647 \u0631\u0626\u064a\u0633 \u0627\u0644\u062d\u0643\u0648\u0645\u0647 \u064a\u0648\u0633\u0641 \u0627\u0644\u0634\u0627\u0647\u062f \u0648\u0627\u0644\u062a\u064a \u0627\u0646\u062a\u0647\u062a \u0628\u0627\u0644\u0645\u0635\u0627\u062f\u0642\u0647 \u0639\u062f\u064a\u062f \u0627\u0644\u0627\u062a\u0641\u0627\u0642\u064a\u0627\u062a \u062a\u0648\u0646\u0633 \u0648\u0627\u0644\u062c\u0632\u0627\u0626\u0631",
"target": "\n\u0642\u0627\u0644 \u0627\u0644\u0648\u0632\u064a\u0631 \u0627\u0644\u0623\u0648\u0644 \u0627\u0644\u062c\u0632\u0627\u0626\u0631\u064a \u0639\u0628\u062f \u0627\u0644\u0645\u0627\u0644\u0643 \u0633\u0644\u0627\u0644 \u0627\u062b\u0631 \u0644\u0642\u0627\u0621 \u062c\u0645\u0639\u0647 \u0628\u0631\u0626\u064a\u0633 \u0645\u062c\u0644\u0633 \u0646\u0648\u0627\u0628 \u0627\u0644\u0634\u0639\u0628 \u0645\u062d\u0645\u062f \u0627\u0644\u0646\u0627\u0635\u0631\u060c \u0625\u0646 \u0627\u0644\u0639\u0644\u0627\u0642\u0627\u062a \u0627\u0644\u062b\u0646\u0627\u0626\u064a\u0629 \u0628\u064a\u0646 \u0627\u0644\u0628\u0644\u062f\u064a\u0646 \u0645\u0645\u064a\u0632\u0629 \u0648\u0633\u062a\u0643\u0648\u0646 \u0623\u062d\u0633\u0646 \u062e\u0644\u0627\u0644 \u0627\u0644\u0641\u062a\u0631\u0629 \u0627\u0644\u0642\u0627\u062f\u0645\u0629."
}
]
```
### Dataset Fields
The dataset has the following fields (also called "features"):
```json
{
"text": "Value(dtype='string', id=None)",
"target": "Value(dtype='string', id=None)"
}
```
### Dataset Splits
This dataset is split into a train and validation split. The split sizes are as follow:
| Split name | Num samples |
| ------------ | ------------------- |
| train | 5102 |
| valid | 1276 |
|
false |
<div align="center">
<img width="640" alt="keremberke/csgo-object-detection" src="https://huggingface.co/datasets/keremberke/csgo-object-detection/resolve/main/thumbnail.jpg">
</div>
### Dataset Labels
```
['ct', 'cthead', 't', 'thead']
```
### Number of Images
```json
{'train': 3879, 'valid': 383, 'test': 192}
```
### How to Use
- Install [datasets](https://pypi.org/project/datasets/):
```bash
pip install datasets
```
- Load the dataset:
```python
from datasets import load_dataset
ds = load_dataset("keremberke/csgo-object-detection", name="full")
example = ds['train'][0]
```
### Roboflow Dataset Page
[https://universe.roboflow.com/asd-culfr/wlots/dataset/1](https://universe.roboflow.com/asd-culfr/wlots/dataset/1?ref=roboflow2huggingface)
### Citation
```
@misc{ wlots_dataset,
title = { wlots Dataset },
type = { Open Source Dataset },
author = { asd },
howpublished = { \\url{ https://universe.roboflow.com/asd-culfr/wlots } },
url = { https://universe.roboflow.com/asd-culfr/wlots },
journal = { Roboflow Universe },
publisher = { Roboflow },
year = { 2022 },
month = { may },
note = { visited on 2023-01-27 },
}
```
### License
CC BY 4.0
### Dataset Summary
This dataset was exported via roboflow.com on December 28, 2022 at 8:08 PM GMT
Roboflow is an end-to-end computer vision platform that helps you
* collaborate with your team on computer vision projects
* collect & organize images
* understand unstructured image data
* annotate, and create datasets
* export, train, and deploy computer vision models
* use active learning to improve your dataset over time
It includes 4454 images.
Ct-cthead-t-thead are annotated in COCO format.
The following pre-processing was applied to each image:
* Auto-orientation of pixel data (with EXIF-orientation stripping)
* Resize to 416x416 (Fill (with center crop))
The following augmentation was applied to create 3 versions of each source image:
* Random brigthness adjustment of between -15 and +15 percent
|
false |
<div align="center">
<img width="640" alt="keremberke/construction-safety-object-detection" src="https://huggingface.co/datasets/keremberke/construction-safety-object-detection/resolve/main/thumbnail.jpg">
</div>
### Dataset Labels
```
['barricade', 'dumpster', 'excavators', 'gloves', 'hardhat', 'mask', 'no-hardhat', 'no-mask', 'no-safety vest', 'person', 'safety net', 'safety shoes', 'safety vest', 'dump truck', 'mini-van', 'truck', 'wheel loader']
```
### Number of Images
```json
{'train': 307, 'valid': 57, 'test': 34}
```
### How to Use
- Install [datasets](https://pypi.org/project/datasets/):
```bash
pip install datasets
```
- Load the dataset:
```python
from datasets import load_dataset
ds = load_dataset("keremberke/construction-safety-object-detection", name="full")
example = ds['train'][0]
```
### Roboflow Dataset Page
[https://universe.roboflow.com/roboflow-universe-projects/construction-site-safety/dataset/1](https://universe.roboflow.com/roboflow-universe-projects/construction-site-safety/dataset/1?ref=roboflow2huggingface)
### Citation
```
@misc{ construction-site-safety_dataset,
title = { Construction Site Safety Dataset },
type = { Open Source Dataset },
author = { Roboflow Universe Projects },
howpublished = { \\url{ https://universe.roboflow.com/roboflow-universe-projects/construction-site-safety } },
url = { https://universe.roboflow.com/roboflow-universe-projects/construction-site-safety },
journal = { Roboflow Universe },
publisher = { Roboflow },
year = { 2023 },
month = { jan },
note = { visited on 2023-01-26 },
}
```
### License
CC BY 4.0
### Dataset Summary
This dataset was exported via roboflow.com on December 29, 2022 at 11:22 AM GMT
Roboflow is an end-to-end computer vision platform that helps you
* collaborate with your team on computer vision projects
* collect & organize images
* understand unstructured image data
* annotate, and create datasets
* export, train, and deploy computer vision models
* use active learning to improve your dataset over time
It includes 398 images.
Construction are annotated in COCO format.
The following pre-processing was applied to each image:
* Auto-orientation of pixel data (with EXIF-orientation stripping)
No image augmentation techniques were applied.
|
false |
# Dataset Card for Bernice Pre-train Data
## Table of Contents
- [Table of Contents](#table-of-contents)
- [Dataset Description](#dataset-description)
- [Dataset Summary](#dataset-summary)
- [Supported Tasks and Leaderboards](#supported-tasks-and-leaderboards)
- [Languages](#languages)
- [Dataset Structure](#dataset-structure)
- [Data Instances](#data-instances)
- [Data Fields](#data-fields)
- [Data Splits](#data-splits)
- [Dataset Creation](#dataset-creation)
- [Curation Rationale](#curation-rationale)
- [Source Data](#source-data)
- [Annotations](#annotations)
- [Personal and Sensitive Information](#personal-and-sensitive-information)
- [Considerations for Using the Data](#considerations-for-using-the-data)
- [Social Impact of Dataset](#social-impact-of-dataset)
- [Discussion of Biases](#discussion-of-biases)
- [Other Known Limitations](#other-known-limitations)
- [Additional Information](#additional-information)
- [Dataset Curators](#dataset-curators)
- [Licensing Information](#licensing-information)
- [Citation Information](#citation-information)
- [Contributions](#contributions)
## Dataset Description
- **Homepage:** N/A
- **Repository:** https://github.com/JHU-CLSP/Bernice-Twitter-encoder
- **Paper:** _Bernice: A Multilingual Pre-trained Encoder for Twitter_ at [EMNLP 2022](https://preview.aclanthology.org/emnlp-22-ingestion/2022.emnlp-main.415)
- **Leaderboard:** N/A
- **Point of Contact:** Alexandra DeLucia aadelucia (at) jhu.edu
### Dataset Summary
Tweet IDs for the 2.5 billion multilingual tweets used to train Bernice, a Twitter encoder.
Read the paper [here](https://preview.aclanthology.org/emnlp-22-ingestion/2022.emnlp-main.415).
The tweets are from the public 1% Twitter API stream from January 2016 to December 2021.
Twitter-provided language metadata is provided with the tweet ID. The data contains 66 unique languages, as identified by [ISO 639 language codes](https://www.wikiwand.com/en/List_of_ISO_639-1_codes), including `und` for undefined languages.
Tweets need to be re-gathered via the Twitter API. We suggest [Hydrator](https://github.com/DocNow/hydrator) or [tweepy](https://www.tweepy.org/).
To load with HuggingFace:
```python
from datasets import load_dataset
dataset = load_dataset("jhu-clsp/bernice-pretrain-data")
for i, row in enumerate(dataset["train"]):
print(row)
if i > 10:
break
```
If you only want Indic languages, use
```python
dataset = load_dataset("jhu-clsp/bernice-pretrain-data", "indic")
```
### Supported Tasks and Leaderboards
N/A
### Languages
65 languages (ISO 639 codes shown below), plus an `und` (undefined) category.
All language identification provided by Twitter API.
| | | | | | | |
|----|-----|----|----|----|-----|----|
| en | ru | ht | zh | bn | ps | lt |
| es | bo | ur | ta | sr | ckb | km |
| pt | it | sv | ro | bg | si | dv |
| ja | th | ca | no | mr | hy | lo |
| ar | de | el | uk | ml | or | ug |
| in | hi | fi | cy | is | pa | |
| ko | pl | cs | ne | te | am | |
| tr | nl | iw | hu | gu | sd | |
| fr | fa | da | eu | kn | my | |
| tl | et | vi | sl | lv | ka | |
## Dataset Structure
### Data Instances
Data is provided in gzip'd files organized by year and month of tweet origin.
Tweets are one per line, with fields separated by tabs.
### Data Fields
* `tweet ID`: ID of tweet
* `lang`: ISO 639 code of language, provided by Twitter metadata. Accuracy of label is not known.
* `year`: Year tweet was created. Year is also provided in the file names.
### Data Splits
[More Information Needed]
## Dataset Creation
### Curation Rationale
Data was gathered to support the training of Bernice, a multilingual pre-trained Twitter encoder.
### Source Data
#### Initial Data Collection and Normalization
Data was gathered via the Twitter API public 1% stream from January 2016 through December 2021.
Tweets with less than three non-username or URL space-delimited words were removed.
All usernames and URLs were replaced with `@USER` and `HTTPURL`, respectively.
#### Who are the source language producers?
Data was produced by users on Twitter.
### Annotations
N/A
### Personal and Sensitive Information
As per Twitter guidelines, only tweet IDs and not full tweets are shared.
Tweets will only be accessible if user has not removed their account (or been banned), tweets were deleted or removed, or a user changed their account access to private.
## Considerations for Using the Data
### Social Impact of Dataset
[More Information Needed]
### Discussion of Biases
[More Information Needed]
### Other Known Limitations
[More Information Needed]
## Additional Information
### Dataset Curators
Dataset gathered and processed by Mark Dredze, Alexandra DeLucia, Shijie Wu, Aaron Mueller, Carlos Aguirre, and Philip Resnik.
### Licensing Information
MIT
### Citation Information
Please cite the Bernice paper if you use this dataset:
> Alexandra DeLucia, Shijie Wu, Aaron Mueller, Carlos Aguirre, Philip Resnik, and Mark Dredze. 2022. Bernice: A Multilingual Pre-trained Encoder for Twitter. In Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing, pages 6191–6205, Abu Dhabi, United Arab Emirates. Association for Computational Linguistics.
### Contributions
Dataset uploaded by [@AADeLucia](https://github.com/AADeLucia).
|
false |
# Dataset Card for UIBert
## Dataset Description
- **Homepage:** https://github.com/google-research-datasets/uibert
- **Repository:** https://github.com/google-research-datasets/uibert
- **Paper:** https://arxiv.org/abs/2107.13731
- **Leaderboard:**
- UIBert: https://arxiv.org/abs/2107.13731
- Pix2Struct: https://arxiv.org/pdf/2210.03347
### Dataset Summary
This is a Hugging Face formatted dataset derived from the [Google UIBert dataset](https://github.com/google-research-datasets/uibert), which is in turn derived from the [RICO dataset](https://interactionmining.org/rico).
### Supported Tasks and Leaderboards
- UI Understanding
- UI Referring Expressions
- UI Action Automation
### Languages
- English
## Dataset Structure
- `screenshot`: blob of pixels.
- `prompt`: Prompt referring to a UI component with an optional action verb. For example "click on search button next to menu drawer."
- `target_bounding_box`: Bounding box of targeted UI components. `[xmin, ymin, xmax, ymax]`
### Data Splits
- train: 15K samples
- validation: 471 samples
- test: 565 samples
## Dataset Creation
### Curation Rationale
[More Information Needed]
### Source Data
#### Initial Data Collection and Normalization
[More Information Needed]
#### Who are the source language producers?
[More Information Needed]
### Annotations
#### Annotation process
[More Information Needed]
#### Who are the annotators?
[More Information Needed]
### Personal and Sensitive Information
[More Information Needed]
## Considerations for Using the Data
### Social Impact of Dataset
[More Information Needed]
### Discussion of Biases
[More Information Needed]
### Other Known Limitations
[More Information Needed]
## Additional Information
### Dataset Curators
[More Information Needed]
### Licensing Information
[More Information Needed]
### Citation Information
[More Information Needed]
### Contributions
[More Information Needed]
|
false |
# Wikipedia (hi) embedded with cohere.ai `multilingual-22-12` encoder
We encoded [Wikipedia (hi)](https://hi.wikipedia.org) using the [cohere.ai](https://txt.cohere.ai/multilingual/) `multilingual-22-12` embedding model.
To get an overview how this dataset was created and pre-processed, have a look at [Cohere/wikipedia-22-12](https://huggingface.co/datasets/Cohere/wikipedia-22-12).
## Embeddings
We compute for `title+" "+text` the embeddings using our `multilingual-22-12` embedding model, a state-of-the-art model that works for semantic search in 100 languages. If you want to learn more about this model, have a look at [cohere.ai multilingual embedding model](https://txt.cohere.ai/multilingual/).
## Further languages
We provide embeddings of Wikipedia in many different languages:
[ar](https://huggingface.co/datasets/Cohere/wikipedia-22-12-ar-embeddings), [de](https://huggingface.co/datasets/Cohere/wikipedia-22-12-de-embeddings), [en](https://huggingface.co/datasets/Cohere/wikipedia-22-12-en-embeddings), [es](https://huggingface.co/datasets/Cohere/wikipedia-22-12-es-embeddings), [fr](https://huggingface.co/datasets/Cohere/wikipedia-22-12-fr-embeddings), [hi](https://huggingface.co/datasets/Cohere/wikipedia-22-12-hi-embeddings), [it](https://huggingface.co/datasets/Cohere/wikipedia-22-12-it-embeddings), [ja](https://huggingface.co/datasets/Cohere/wikipedia-22-12-ja-embeddings), [ko](https://huggingface.co/datasets/Cohere/wikipedia-22-12-ko-embeddings), [simple english](https://huggingface.co/datasets/Cohere/wikipedia-22-12-simple-embeddings), [zh](https://huggingface.co/datasets/Cohere/wikipedia-22-12-zh-embeddings),
You can find the Wikipedia datasets without embeddings at [Cohere/wikipedia-22-12](https://huggingface.co/datasets/Cohere/wikipedia-22-12).
## Loading the dataset
You can either load the dataset like this:
```python
from datasets import load_dataset
docs = load_dataset(f"Cohere/wikipedia-22-12-hi-embeddings", split="train")
```
Or you can also stream it without downloading it before:
```python
from datasets import load_dataset
docs = load_dataset(f"Cohere/wikipedia-22-12-hi-embeddings", split="train", streaming=True)
for doc in docs:
docid = doc['id']
title = doc['title']
text = doc['text']
emb = doc['emb']
```
## Search
A full search example:
```python
#Run: pip install cohere datasets
from datasets import load_dataset
import torch
import cohere
co = cohere.Client(f"<<COHERE_API_KEY>>") # Add your cohere API key from www.cohere.com
#Load at max 1000 documents + embeddings
max_docs = 1000
docs_stream = load_dataset(f"Cohere/wikipedia-22-12-hi-embeddings", split="train", streaming=True)
docs = []
doc_embeddings = []
for doc in docs_stream:
docs.append(doc)
doc_embeddings.append(doc['emb'])
if len(docs) >= max_docs:
break
doc_embeddings = torch.tensor(doc_embeddings)
query = 'Who founded Youtube'
response = co.embed(texts=[query], model='multilingual-22-12')
query_embedding = response.embeddings
query_embedding = torch.tensor(query_embedding)
# Compute dot score between query embedding and document embeddings
dot_scores = torch.mm(query_embedding, doc_embeddings.transpose(0, 1))
top_k = torch.topk(dot_scores, k=3)
# Print results
print("Query:", query)
for doc_id in top_k.indices[0].tolist():
print(docs[doc_id]['title'])
print(docs[doc_id]['text'], "\n")
```
## Performance
You can find performance on the MIRACL dataset (a semantic search evaluation dataset) here: [miracl-en-queries-22-12#performance](https://huggingface.co/datasets/Cohere/miracl-en-queries-22-12#performance) |
false |
<div align="center">
<img width="640" alt="keremberke/protective-equipment-detection" src="https://huggingface.co/datasets/keremberke/protective-equipment-detection/resolve/main/thumbnail.jpg">
</div>
### Dataset Labels
```
['glove', 'goggles', 'helmet', 'mask', 'no_glove', 'no_goggles', 'no_helmet', 'no_mask', 'no_shoes', 'shoes']
```
### Number of Images
```json
{'valid': 3570, 'test': 1935, 'train': 6473}
```
### How to Use
- Install [datasets](https://pypi.org/project/datasets/):
```bash
pip install datasets
```
- Load the dataset:
```python
from datasets import load_dataset
ds = load_dataset("keremberke/protective-equipment-detection", name="full")
example = ds['train'][0]
```
### Roboflow Dataset Page
[https://universe.roboflow.com/personal-protective-equipment/ppes-kaxsi/dataset/7](https://universe.roboflow.com/personal-protective-equipment/ppes-kaxsi/dataset/7?ref=roboflow2huggingface)
### Citation
```
@misc{ ppes-kaxsi_dataset,
title = { PPEs Dataset },
type = { Open Source Dataset },
author = { Personal Protective Equipment },
howpublished = { \\url{ https://universe.roboflow.com/personal-protective-equipment/ppes-kaxsi } },
url = { https://universe.roboflow.com/personal-protective-equipment/ppes-kaxsi },
journal = { Roboflow Universe },
publisher = { Roboflow },
year = { 2022 },
month = { jul },
note = { visited on 2023-01-18 },
}
```
### License
CC BY 4.0
### Dataset Summary
This dataset was exported via roboflow.ai on July 7, 2022 at 3:49 PM GMT
It includes 11978 images.
Ppe-equipements are annotated in COCO format.
The following pre-processing was applied to each image:
* Auto-orientation of pixel data (with EXIF-orientation stripping)
No image augmentation techniques were applied.
|
true |
# Dataset Card for Bundestag-v2
## Table of Contents
- [Table of Contents](#table-of-contents)
- [Dataset Description](#dataset-description)
- [Dataset Summary](#dataset-summary)
- [Supported Tasks](#supported-tasks)
- [Languages](#languages)
- [Dataset Structure](#dataset-structure)
- [Data Instances](#data-instances)
- [Data Fields](#data-fields)
- [Data Splits](#data-splits)
- [Dataset Creation](#dataset-creation)
- [Curation Rationale](#curation-rationale)
- [Source Data](#source-data)
- [Personal and Sensitive Information](#personal-and-sensitive-information)
- [Considerations for Using the Data](#considerations-for-using-the-data)
- [Social Impact of Dataset](#social-impact-of-dataset)
- [Discussion of Biases](#discussion-of-biases)
- [Other Known Limitations](#other-known-limitations)
- [Additional Information](#additional-information)
- [Dataset Curators](#dataset-curators)
- [Licensing Information](#licensing-information)
- [Citation Information](#citation-information)
- [Contributions](#contributions)
## Dataset Description
- Homepage: https://doi.org/10.7910/DVN/L4OAKN
### Dataset Summary
This dataset was generated from the [ParlSpeech V2](https://doi.org/10.7910/DVN/L4OAKN) dataset. It contains speeches from the german parliament from 1990 until 2020 labelled with the party of the speaker.
### Supported Tasks
Text Classification
### Languages
German
## Dataset Structure
### Data Fields
- text: Transcript of the speech in german
- party: Party of the speaker
### Data Splits
- train
- validation
- test
## Dataset Creation
### Curation Rationale
Created to train a language model, which is able to classify speeches by party.
### Source Data
#### Initial Data Collection and Normalization
- [ParlSpeech V2](https://doi.org/10.7910/DVN/L4OAKN)
## Considerations for Using the Data
### Social Impact of Dataset
These are political speeches, therefor the content can be controversial and potentially harmful.
## Additional Information
### Licensing Information
[CCO 1.0](http://creativecommons.org/publicdomain/zero/1.0)
### Citation Information
Bibtex entry:
```
@data{DVN/L4OAKN_2020,
author = {Rauh, Christian and Schwalbach, Jan},
publisher = {Harvard Dataverse},
title = {{The ParlSpeech V2 data set: Full-text corpora of 6.3 million parliamentary speeches in the key legislative chambers of nine representative democracies}},
year = {2020},
version = {V1},
doi = {10.7910/DVN/L4OAKN},
url = {https://doi.org/10.7910/DVN/L4OAKN}
}
``` |
false |
# Dataset Card for OSCAR-2019-Burmese-fix
## Dataset Description
This dataset is a cleand version of Myanmar language in OSCAR 2019 dataset.
### Contributions
[Swan Htet Aung](https://github.com/swanhtet1992)
|
false |
# Dataset Card for "REDDIT_submissions"
## Dataset Description
- **Homepage:**
- **Paper: https://arxiv.org/abs/2001.08435**
### Dataset Summary
Submissions of 50 high-quality subreddits, extracted from the REDDIT PushShift data dumps (from 2006 to Jan 2023).
### Supported Tasks
These submissions can be used for text generation and language modeling, as well as dialogue modeling.
## Dataset Structure
### Data Splits
Each split corresponds to a specific subreddit in the following list: "tifu", "explainlikeimfive", "WritingPrompts", "changemyview", "LifeProTips", "todayilearned", "science", "askscience", "ifyoulikeblank", "Foodforthought", "IWantToLearn", "bestof", "IAmA", "socialskills", "relationship_advice", "philosophy", "YouShouldKnow", "history", "books", "Showerthoughts", "personalfinance", "buildapc", "EatCheapAndHealthy", "boardgames", "malefashionadvice", "femalefashionadvice", "scifi", "Fantasy", "Games", "bodyweightfitness", "SkincareAddiction", "podcasts", "suggestmeabook", "AskHistorians", "gaming", "DIY", "mildlyinteresting", "sports", "space", "gadgets", "Documentaries", "GetMotivated", "UpliftingNews", "technology", "Fitness", "travel", "lifehacks", "Damnthatsinteresting", "gardening", "programming"
## Dataset Creation
### Curation Rationale
All the information fields have been cast to string, as their format change through time from one dump to the following. A reduced number of keys have been kept: "allow_live_comments", "archived", "author", "author_fullname", "banned_by", "category", "content_categories", "contest_mode", "created_utc", "discussion_type", "distinguished", "domain", "edited", "gilded", "hidden", "hide_score", "id", "is_created_from_ads_ui", "is_crosspostable", "is_meta", "is_original_content", "is_reddit_media_domain", "is_robot_indexable", "is_self", "is_video", "locked", "media", "media_embed", "media_only", "name", "no_follow", "num_comments", "num_crossposts", "over_18", "parent_whitelist_status", "permalink", "pinned", "post_hint", "pwls", "quarantine", "removed_by", "removed_by_category", "retrieved_on", "score", "secure_media", "secure_media_embed", "selftext", "send_replies", "spoiler", "stickied", "subreddit", "subreddit_id", "subreddit_name_prefixed", "subreddit_subscribers", "subreddit_type", "suggested_sort", "title", "top_awarded_type", "total_awards_received", "treatment_tags", "upvote_ratio", "url", "url_overridden_by_dest", "view_count", "whitelist_status", "wls".
### Source Data
The [Reddit PushShift data dumps](https://files.pushshift.io/reddit/) are part of a data collection effort which crawls Reddit at regular intervals, to extract and keep all its data.
#### Initial Data Collection and Normalization
See the paper.
#### Who are the source language producers?
Redditors are mostly young (65% below 30), male (70%), and American (50% of the site).
### Personal and Sensitive Information
The data contains Redditor's usernames associated to their content.
## Considerations for Using the Data
This dataset should be anonymized before any processing.
Though the subreddits selected are considered as being of higher quality, they can still reflect what you can find on the internet in terms of expressions of biases and toxicity.
### Contributions
Thanks to [@clefourrier](https://github.com/clefourrier) for adding this dataset. |
false | |
false |
# Dataset Card for pills-sxdht
** The original COCO dataset is stored at `dataset.tar.gz`**
## Dataset Description
- **Homepage:** https://universe.roboflow.com/object-detection/pills-sxdht
- **Point of Contact:** francesco.zuppichini@gmail.com
### Dataset Summary
pills-sxdht
### Supported Tasks and Leaderboards
- `object-detection`: The dataset can be used to train a model for Object Detection.
### Languages
English
## Dataset Structure
### Data Instances
A data point comprises an image and its object annotations.
```
{
'image_id': 15,
'image': <PIL.JpegImagePlugin.JpegImageFile image mode=RGB size=640x640 at 0x2373B065C18>,
'width': 964043,
'height': 640,
'objects': {
'id': [114, 115, 116, 117],
'area': [3796, 1596, 152768, 81002],
'bbox': [
[302.0, 109.0, 73.0, 52.0],
[810.0, 100.0, 57.0, 28.0],
[160.0, 31.0, 248.0, 616.0],
[741.0, 68.0, 202.0, 401.0]
],
'category': [4, 4, 0, 0]
}
}
```
### Data Fields
- `image`: the image id
- `image`: `PIL.Image.Image` object containing the image. Note that when accessing the image column: `dataset[0]["image"]` the image file is automatically decoded. Decoding of a large number of image files might take a significant amount of time. Thus it is important to first query the sample index before the `"image"` column, *i.e.* `dataset[0]["image"]` should **always** be preferred over `dataset["image"][0]`
- `width`: the image width
- `height`: the image height
- `objects`: a dictionary containing bounding box metadata for the objects present on the image
- `id`: the annotation id
- `area`: the area of the bounding box
- `bbox`: the object's bounding box (in the [coco](https://albumentations.ai/docs/getting_started/bounding_boxes_augmentation/#coco) format)
- `category`: the object's category.
#### Who are the annotators?
Annotators are Roboflow users
## Additional Information
### Licensing Information
See original homepage https://universe.roboflow.com/object-detection/pills-sxdht
### Citation Information
```
@misc{ pills-sxdht,
title = { pills sxdht Dataset },
type = { Open Source Dataset },
author = { Roboflow 100 },
howpublished = { \url{ https://universe.roboflow.com/object-detection/pills-sxdht } },
url = { https://universe.roboflow.com/object-detection/pills-sxdht },
journal = { Roboflow Universe },
publisher = { Roboflow },
year = { 2022 },
month = { nov },
note = { visited on 2023-03-29 },
}"
```
### Contributions
Thanks to [@mariosasko](https://github.com/mariosasko) for adding this dataset. |
false |
# Dataset Card for smoke-uvylj
** The original COCO dataset is stored at `dataset.tar.gz`**
## Dataset Description
- **Homepage:** https://universe.roboflow.com/object-detection/smoke-uvylj
- **Point of Contact:** francesco.zuppichini@gmail.com
### Dataset Summary
smoke-uvylj
### Supported Tasks and Leaderboards
- `object-detection`: The dataset can be used to train a model for Object Detection.
### Languages
English
## Dataset Structure
### Data Instances
A data point comprises an image and its object annotations.
```
{
'image_id': 15,
'image': <PIL.JpegImagePlugin.JpegImageFile image mode=RGB size=640x640 at 0x2373B065C18>,
'width': 964043,
'height': 640,
'objects': {
'id': [114, 115, 116, 117],
'area': [3796, 1596, 152768, 81002],
'bbox': [
[302.0, 109.0, 73.0, 52.0],
[810.0, 100.0, 57.0, 28.0],
[160.0, 31.0, 248.0, 616.0],
[741.0, 68.0, 202.0, 401.0]
],
'category': [4, 4, 0, 0]
}
}
```
### Data Fields
- `image`: the image id
- `image`: `PIL.Image.Image` object containing the image. Note that when accessing the image column: `dataset[0]["image"]` the image file is automatically decoded. Decoding of a large number of image files might take a significant amount of time. Thus it is important to first query the sample index before the `"image"` column, *i.e.* `dataset[0]["image"]` should **always** be preferred over `dataset["image"][0]`
- `width`: the image width
- `height`: the image height
- `objects`: a dictionary containing bounding box metadata for the objects present on the image
- `id`: the annotation id
- `area`: the area of the bounding box
- `bbox`: the object's bounding box (in the [coco](https://albumentations.ai/docs/getting_started/bounding_boxes_augmentation/#coco) format)
- `category`: the object's category.
#### Who are the annotators?
Annotators are Roboflow users
## Additional Information
### Licensing Information
See original homepage https://universe.roboflow.com/object-detection/smoke-uvylj
### Citation Information
```
@misc{ smoke-uvylj,
title = { smoke uvylj Dataset },
type = { Open Source Dataset },
author = { Roboflow 100 },
howpublished = { \url{ https://universe.roboflow.com/object-detection/smoke-uvylj } },
url = { https://universe.roboflow.com/object-detection/smoke-uvylj },
journal = { Roboflow Universe },
publisher = { Roboflow },
year = { 2022 },
month = { nov },
note = { visited on 2023-03-29 },
}"
```
### Contributions
Thanks to [@mariosasko](https://github.com/mariosasko) for adding this dataset. |
false |
# Camoscio instruction-tuning dataset
This repository contains the dataset used to train [Camoscio](https://huggingface.co/teelinsan/camoscio-7b-llama).
This dataset is an Italian translation with ChatGPT of the [Stanford Alpaca dataset](https://github.com/tatsu-lab/stanford_alpaca).
Please refer to the [Camoscio repo](https://github.com/teelinsan/camoscio) for more info.
|
false |
# Dataset Card for Dataset Name
## Dataset Description
- **Homepage:**
- **Repository:**
- **Paper:**
- **Leaderboard:**
- **Point of Contact:**
### Dataset Summary
This dataset card aims to be a base template for new datasets. It has been generated using [this raw template](https://github.com/huggingface/huggingface_hub/blob/main/src/huggingface_hub/templates/datasetcard_template.md?plain=1).
### Supported Tasks and Leaderboards
[More Information Needed]
### Languages
[More Information Needed]
## Dataset Structure
### Data Instances
[More Information Needed]
### Data Fields
[More Information Needed]
### Data Splits
[More Information Needed]
## Dataset Creation
### Curation Rationale
[More Information Needed]
### Source Data
#### Initial Data Collection and Normalization
[More Information Needed]
#### Who are the source language producers?
[More Information Needed]
### Annotations
#### Annotation process
[More Information Needed]
#### Who are the annotators?
[More Information Needed]
### Personal and Sensitive Information
[More Information Needed]
## Considerations for Using the Data
### Social Impact of Dataset
[More Information Needed]
### Discussion of Biases
[More Information Needed]
### Other Known Limitations
[More Information Needed]
## Additional Information
### Dataset Curators
[More Information Needed]
### Licensing Information
[More Information Needed]
### Citation Information
[More Information Needed]
### Contributions
[More Information Needed] |
false | |
false |
# lexica-aperture-v2
[lexica-aperture-v3](https://huggingface.co/datasets/hlky/lexica-aperture-v3/)
[scrape script](https://github.com/hlky/scrape/blob/main/lexica.py)
```
910322 rows x 15 columns
'id', 'width', 'height', 'upscaled_width', 'upscaled_height', 'is_upscaled', 'url', 'upscaled_url', 'userid', 'prompt', 'negativePrompt', 'timestamp', 'seed', 'cfg', 'model'
```
is_upscaled = width != upscaled_height
if image is upscaled then url is
```
https://image.lexica.art/md2/{id}
```
and upscaled url is
```
https://image.lexica.art/full_jpg/{id}
```
|
false |
MMC4-130k-chinese是对MMC4中，抽样了130k左右 simliarty较高的图文pair得到的数据集
Chinese版本是对这里所有的caption进行了翻译。
我们会陆续将更多数据集发布到hf，包括
- [ ] Coco Caption的中文翻译
- [ ] CoQA的中文翻译
- [ ] CNewSum的Embedding数据
- [ ] 增广的开放QA数据
- [x] WizardLM的中文翻译
如果你也在做这些数据集的筹备，欢迎来联系我们，避免重复花钱。
# 骆驼(Luotuo): 开源中文大语言模型
[https://github.com/LC1332/Luotuo-Chinese-LLM](https://github.com/LC1332/Luotuo-Chinese-LLM)
骆驼(Luotuo)项目是由[冷子昂](https://blairleng.github.io) @ 商汤科技, 陈启源 @ 华中师范大学 以及 李鲁鲁 @ 商汤科技 发起的中文大语言模型开源项目，包含了一系列语言模型。
( 注意: [陈启源](https://qiyuan-chen.github.io/) 正在寻找2024推免导师，欢迎联系 )
骆驼项目**不是**商汤科技的官方产品。
## Citation
Please cite the repo if you use the data or code in this repo.
```
@misc{alpaca,
author={Ziang Leng, Qiyuan Chen and Cheng Li},
title = {Luotuo: An Instruction-following Chinese Language model, LoRA tuning on LLaMA},
year = {2023},
publisher = {GitHub},
journal = {GitHub repository},
howpublished = {\url{https://github.com/LC1332/Luotuo-Chinese-LLM}},
}
``` |
false |
## Dataset Description
- **Repository:** https://github.com/rahular/varta
- **Paper:** https://arxiv.org/abs/2305.05858
### Dataset Summary
Varta is a diverse, challenging, large-scale, multilingual, and high-quality headline-generation dataset containing 41.8 million news articles in 14 Indic languages and English.
The data is crawled from DailyHunt, a popular news aggregator in India that pulls high-quality articles from multiple trusted and reputed news publishers.
### Languages
Assamese, Bhojpuri, Bengali, English, Gujarati, Hindi, Kannada, Malayalam, Marathi, Nepali, Oriya, Punjabi, Tamil, Telugu, and Urdu.
## Dataset Structure
### Data Instances
```
{
"id":"n400000150",
"langCode":"as",
"source_url":"https://www.etvbharat.com/assamese/assam/bharat/militant-hideout-destroyed-on-srinagar-bandipora-highway/assam20220630074145729729173",
"dh_url":"https://m.dailyhunt.in/news/india/assamese/etvbharatassamese-epaper-dh6b381d65c3344bbcad9a06ee28b4ab2a/boma+nikshepeve+dhbans+kva+hl+santvasabadiv+aatmagopanasthali-newsid-n400000150"
}
```
### Data Fields
- id: unique identifier for the artilce on DailyHunt. This id will be used to recreate the dataset.
- langCode: ISO 639-1 language code
- source_url: the url that points to the article on the website of the original publisher
- dh_url: the url that points to the article on DailyHunt
### Data Splits
From every language, we randomly sample 10,000 articles each for validation and testing. We also ensure that at least 80% of a language’s data is available for training.
Therefore, if a language has less than 100,000 articles, we restrict its validation and test splits to 10% of its size.
We also create a `small` training set by limiting the number of articles from each language to 100K.
This `small` training set with a size of 1.3M is used in all our fine-tuning experiments.
You can find the `small` training set [here](https://huggingface.co/datasets/rahular/varta/blob/main/varta/train/train_100k.json)
## Data Recreation
To recreate the dataset, follow this [README file](https://github.com/rahular/varta/tree/main/crawler#README.md).
## Misc
- Original source: https://m.dailyhunt.in/
- License: CC-BY 4.0
## Citation Information
```
@misc{aralikatte2023varta,
title={V\=arta: A Large-Scale Headline-Generation Dataset for Indic Languages},
author={Rahul Aralikatte and Ziling Cheng and Sumanth Doddapaneni and Jackie Chi Kit Cheung},
year={2023},
eprint={2305.05858},
archivePrefix={arXiv},
primaryClass={cs.CL}
}
``` |
false | 这是一个「英雄联盟」原画的图片数据集，旨在为「英雄联盟」原画风格的图片生成和风格迁移提供训练数据。本数据集中的图片均为高分辨率的「英雄联盟」原画，图片尺寸全部大于 1920 * 1080。 |
false |
# Captcha dataset
## Data
Captcha images with solutions of exactly 6 digit numbers
## Splits
* Train: 6000 images
* Validation: 2000 images
* Test: 2000 images
## Example
 |
false | |
true | # Dataset Card for Dataset Name
## Dataset Description
- **Homepage:**
- **Repository:**
- **Paper:**
- **Leaderboard:**
- **Point of Contact:**
### Dataset Summary
This dataset card aims to be a base template for new datasets. It has been generated using [this raw template](https://github.com/huggingface/huggingface_hub/blob/main/src/huggingface_hub/templates/datasetcard_template.md?plain=1).
### Supported Tasks and Leaderboards
[More Information Needed]
### Languages
[More Information Needed]
## Dataset Structure
### Data Instances
[More Information Needed]
### Data Fields
[More Information Needed]
### Data Splits
[More Information Needed]
## Dataset Creation
### Curation Rationale
[More Information Needed]
### Source Data
#### Initial Data Collection and Normalization
[More Information Needed]
#### Who are the source language producers?
[More Information Needed]
### Annotations
#### Annotation process
[More Information Needed]
#### Who are the annotators?
[More Information Needed]
### Personal and Sensitive Information
[More Information Needed]
## Considerations for Using the Data
### Social Impact of Dataset
[More Information Needed]
### Discussion of Biases
[More Information Needed]
### Other Known Limitations
[More Information Needed]
## Additional Information
### Dataset Curators
[More Information Needed]
### Licensing Information
[More Information Needed]
### Citation Information
[More Information Needed]
### Contributions
[More Information Needed] |
false |
# Dataset Card for Quick, Draw!
This is a processed version of Google's [Quick, Draw](https://huggingface.co/datasets/quickdraw/) dataset to be compatible with the latest versions of 🤗 Datasets that support .parquet files. NOTE: this dataset only contains the "preprocessed_bitmaps" subset of the original dataset.
|
false |
## Dataset Description
- **Homepage:** [Add homepage URL here if available (unless it's a GitHub repository)]()
- **Repository:** [If the dataset is hosted on github or has a github homepage, add URL here]()
- **Paper:** [If the dataset was introduced by a paper or there was a paper written describing the dataset, add URL here (landing page for Arxiv paper preferred)]()
- **Leaderboard:** [If the dataset supports an active leaderboard, add link here]()
- **Point of Contact:** [If known, name and email of at least one person the reader can contact for questions about the dataset.]()
### Dataset Summary
Briefly summarize the dataset, its intended use and the supported tasks. Give an overview of how and why the dataset was created. The summary should explicitly mention the languages present in the dataset (possibly in broad terms, e.g. *translations between several pairs of European languages*), and describe the domain, topic, or genre covered.
### Supported Tasks and Leaderboards
For each of the tasks tagged for this dataset, give a brief description of the tag, metrics, and suggested models (with a link to their HuggingFace implementation if available). Give a similar description of tasks that were not covered by the structured tag set (repace the `task-category-tag` with an appropriate `other:other-task-name`).
- `task-category-tag`: The dataset can be used to train a model for [TASK NAME], which consists in [TASK DESCRIPTION]. Success on this task is typically measured by achieving a *high/low* [metric name](https://huggingface.co/metrics/metric_name). The ([model name](https://huggingface.co/model_name) or [model class](https://huggingface.co/transformers/model_doc/model_class.html)) model currently achieves the following score. *[IF A LEADERBOARD IS AVAILABLE]:* This task has an active leaderboard which can be found at [leaderboard url]() and ranks models based on [metric name](https://huggingface.co/metrics/metric_name) while also reporting [other metric name](https://huggingface.co/metrics/other_metric_name).
### Languages
Provide a brief overview of the languages represented in the dataset. Describe relevant details about specifics of the language such as whether it is social media text, African American English,...
When relevant, please provide [BCP-47 codes](https://tools.ietf.org/html/bcp47), which consist of a [primary language subtag](https://tools.ietf.org/html/bcp47#section-2.2.1), with a [script subtag](https://tools.ietf.org/html/bcp47#section-2.2.3) and/or [region subtag](https://tools.ietf.org/html/bcp47#section-2.2.4) if available.
## Dataset Structure
### Data Instances
Provide an JSON-formatted example and brief description of a typical instance in the dataset. If available, provide a link to further examples.
```
{
'example_field': ...,
...
}
```
Provide any additional information that is not covered in the other sections about the data here. In particular describe any relationships between data points and if these relationships are made explicit.
### Data Fields
List and describe the fields present in the dataset. Mention their data type, and whether they are used as input or output in any of the tasks the dataset currently supports. If the data has span indices, describe their attributes, such as whether they are at the character level or word level, whether they are contiguous or not, etc. If the datasets contains example IDs, state whether they have an inherent meaning, such as a mapping to other datasets or pointing to relationships between data points.
- `example_field`: description of `example_field`
Note that the descriptions can be initialized with the **Show Markdown Data Fields** output of the [Datasets Tagging app](https://huggingface.co/spaces/huggingface/datasets-tagging), you will then only need to refine the generated descriptions.
### Data Splits
Describe and name the splits in the dataset if there are more than one.
Describe any criteria for splitting the data, if used. If there are differences between the splits (e.g. if the training annotations are machine-generated and the dev and test ones are created by humans, or if different numbers of annotators contributed to each example), describe them here.
Provide the sizes of each split. As appropriate, provide any descriptive statistics for the features, such as average length. For example:
| | train | validation | test |
|-------------------------|------:|-----------:|-----:|
| Input Sentences | | | |
| Average Sentence Length | | | |
## Dataset Creation
### Curation Rationale
What need motivated the creation of this dataset? What are some of the reasons underlying the major choices involved in putting it together?
### Source Data
This section describes the source data (e.g. news text and headlines, social media posts, translated sentences,...)
#### Initial Data Collection and Normalization
Describe the data collection process. Describe any criteria for data selection or filtering. List any key words or search terms used. If possible, include runtime information for the collection process.
If data was collected from other pre-existing datasets, link to source here and to their [Hugging Face version](https://huggingface.co/datasets/dataset_name).
If the data was modified or normalized after being collected (e.g. if the data is word-tokenized), describe the process and the tools used.
#### Who are the source language producers?
State whether the data was produced by humans or machine generated. Describe the people or systems who originally created the data.
If available, include self-reported demographic or identity information for the source data creators, but avoid inferring this information. Instead state that this information is unknown. See [Larson 2017](https://www.aclweb.org/anthology/W17-1601.pdf) for using identity categories as a variables, particularly gender.
Describe the conditions under which the data was created (for example, if the producers were crowdworkers, state what platform was used, or if the data was found, what website the data was found on). If compensation was provided, include that information here.
Describe other people represented or mentioned in the data. Where possible, link to references for the information.
### Annotations
If the dataset contains annotations which are not part of the initial data collection, describe them in the following paragraphs.
#### Annotation process
If applicable, describe the annotation process and any tools used, or state otherwise. Describe the amount of data annotated, if not all. Describe or reference annotation guidelines provided to the annotators. If available, provide interannotator statistics. Describe any annotation validation processes.
#### Who are the annotators?
If annotations were collected for the source data (such as class labels or syntactic parses), state whether the annotations were produced by humans or machine generated.
Describe the people or systems who originally created the annotations and their selection criteria if applicable.
If available, include self-reported demographic or identity information for the annotators, but avoid inferring this information. Instead state that this information is unknown. See [Larson 2017](https://www.aclweb.org/anthology/W17-1601.pdf) for using identity categories as a variables, particularly gender.
Describe the conditions under which the data was annotated (for example, if the annotators were crowdworkers, state what platform was used, or if the data was found, what website the data was found on). If compensation was provided, include that information here.
### Personal and Sensitive Information
State whether the dataset uses identity categories and, if so, how the information is used. Describe where this information comes from (i.e. self-reporting, collecting from profiles, inferring, etc.). See [Larson 2017](https://www.aclweb.org/anthology/W17-1601.pdf) for using identity categories as a variables, particularly gender. State whether the data is linked to individuals and whether those individuals can be identified in the dataset, either directly or indirectly (i.e., in combination with other data).
State whether the dataset contains other data that might be considered sensitive (e.g., data that reveals racial or ethnic origins, sexual orientations, religious beliefs, political opinions or union memberships, or locations; financial or health data; biometric or genetic data; forms of government identification, such as social security numbers; criminal history).
If efforts were made to anonymize the data, describe the anonymization process.
## Considerations for Using the Data
### Social Impact of Dataset
Please discuss some of the ways you believe the use of this dataset will impact society.
The statement should include both positive outlooks, such as outlining how technologies developed through its use may improve people's lives, and discuss the accompanying risks. These risks may range from making important decisions more opaque to people who are affected by the technology, to reinforcing existing harmful biases (whose specifics should be discussed in the next section), among other considerations.
Also describe in this section if the proposed dataset contains a low-resource or under-represented language. If this is the case or if this task has any impact on underserved communities, please elaborate here.
### Discussion of Biases
Provide descriptions of specific biases that are likely to be reflected in the data, and state whether any steps were taken to reduce their impact.
For Wikipedia text, see for example [Dinan et al 2020 on biases in Wikipedia (esp. Table 1)](https://arxiv.org/abs/2005.00614), or [Blodgett et al 2020](https://www.aclweb.org/anthology/2020.acl-main.485/) for a more general discussion of the topic.
If analyses have been run quantifying these biases, please add brief summaries and links to the studies here.
### Other Known Limitations
If studies of the datasets have outlined other limitations of the dataset, such as annotation artifacts, please outline and cite them here.
## Additional Information
### Dataset Curators
List the people involved in collecting the dataset and their affiliation(s). If funding information is known, include it here.
### Licensing Information
Provide the license and link to the license webpage if available.
### Citation Information
Provide the [BibTex](http://www.bibtex.org/)-formatted reference for the dataset. For example:
```
@article{article_id,
author = {Author List},
title = {Dataset Paper Title},
journal = {Publication Venue},
year = {2525}
}
```
If the dataset has a [DOI](https://www.doi.org/), please provide it here.
### Contributions
Thanks to [@github-username](https://github.com/<github-username>) for adding this dataset. |
true | # AutoNLP Dataset for project: song-lyrics-demo
## Table of content
- [Dataset Description](#dataset-description)
- [Languages](#languages)
- [Dataset Structure](#dataset-structure)
- [Data Instances](#data-instances)
- [Data Fields](#data-fields)
- [Data Splits](#data-splits)
## Dataset Descritpion
This dataset has been automatically processed by AutoNLP for project song-lyrics-demo.
### Languages
The BCP-47 code for the dataset's language is en.
## Dataset Structure
### Data Instances
A sample from this dataset looks as follows:
```json
[
{
"target": 2,
"text": "[Intro: Method Man w/ sample] + (Sunny valentine). We got butter (8X). (The gun'll go the gun'll go.... The gun'll go...). [Raekwon]. Aiyo one thing for sure keep you of all. Keep a nice crib fly away keep to the point. Keep niggaz outta ya face who snakes. Keep bitches in they place keep the mac in a special place. Keep moving for papes keep cool keep doing what you doing. Keep it fly keep me in the crates. Cuz I will erase shit on the real note you'se a waste. It's right here for you I will lace you. Rip you and brace you put a nice W up on ya face. Word to mother you could get chased. It's nothing to taste blood on a thug if he gotta go. All I know is we be giving grace. This is a place from where we make tapes. We make 'em everywhere still in all we be making base. Y'all be making paste these little niggaz they be making shapes. Our shit is art yours is traced. [Chorus: Sunny Valentine]. This is the way that we rolling in the streets. You know when we roll we be packing that heat. The gun'll go the gun'll go the gun'll go the gun'll go. The gun'll go the gun'll go the gun'll go the gun'll go. The gun'll go the gun'll go.... [Method Man]. This is Poverty Island man these animals don't run. Slums where the ambulance don't come. Who got the best base? Fiends waiting to smoke some. Approach something ask him where he getting that coke from. My dudes hug blocks like samurai shogun. Cuz no V and no ones equalling no fun. Who want a treat they know huh? Body to go numb. My woman need funds plus her hair and her toes done. It is what it is though you fuck with the kid flow. That make it hard to get dough the harder to get gold. Harder the piff blow harder when it snow. The pinky and the wrist glow this here what we live for. Get gwop then get low but first thought. We gotta get the work off the gift and the curse boss. Yeah see I'm the shit yo the dirt in the fit no. Hustling from the get-go the motto is get more. [Chorus]. [Masta Killa]. We was quiet flashy brothers strapped all along. With the dirty .38 long twelve hour shift gate. Took case state to state you think he won't hold his weight?. Put ya money on the plate and watch it get scrapped. We get ape up in that club off that juice and Henn. And it's a no win situation fucking with them. You mean like Ewing at the front at the rim finger roll a Dutch. Million dollar stages touched techs gauges bust. Trust no one the lone shogun rugged Timb boot stomper. Damaging lyrical mass destruction launcher. Nothing can calm the quakeage when I break kid. Peace to my brothers up north doing state bids. [Chorus]. [Chorus 2: Sunny Valentine]. Whoa... this is the way we be rolling in the club. You know when we roll we be packing .32 snubs. The gun'll go the gun'll go the gun'll go the gun'll go. The gun'll go the gun'll go the gun'll go the gun'll go. The gun'll go the gun'll go the gun'll go the gun'll go. [Outro: sample to fade]. We got butter..."
},
{
"target": 4,
"text": "[Sean Paul:]. Aye. It's Sean Paul 'long side. The mandem called Jay Sean. Fi di gal dem. Tellin' 'em again what we tell 'em. [Jay Sean:]. Pass me a drink to the left yeah. Said her name was Delilah. And I'm like \"you should come my way\". I already surrender. Damn girl that body's fire. You gon' remember my name. (She should give it up definite). You need it. I need it. We can jump in the deep end. I wanna get lost in your love. I just wanna be close to you. (Just wanna I just wanna). And do all the things you want me to. I just wanna be close to you. (I just wanna I just wanna). And show you the way I feel. You make my love go. You make my love go. You make my love go. In the morning we gon' do it again wake up. I'mma do it like we just broke up and made up. Get up on top of me and work up a sweat work up a sweat. See we can do it any type of way that you want. I'm thinking maybe you're the right kind of wrong. I'm saying baby you won't ever forget my love. You need it. I need it. We can jump in the deep end. I wanna get lost in your love. I just wanna be close to you. (Just wanna I just wanna). And do all the things you want me to. I just wanna be close to you. (I just wanna I just wanna). And show you the way I feel. You make my love go. You make my love go. You make my love go. [Sean Paul:]. Girl mi wan' figure hundred hundred and fifty. Love how you move you know that I'm with it. Perfect size I know that you fit it. Just let me hit it you know mi not quit it. Pon di Dl like Cassie and Diddy. Mi na wound a mi watch we like Sin City. Full time mi run da ting mi tall legend. If you don't come gimme dat would I be offended my girl. Come here down wan' see something me want in life and then waste time. A you a mi pree every day baby full time when ya de pon on mi mind. So mi wine if you give it to me baby girl so we can play. Stick to the ting now I am your king my girl this is what we say. [Jay Sean:]. I just wanna be close to you. (Just wanna I just wanna). And do all the things you want me to. I just wanna be close to you. (I just wanna I just wanna). And show you the way I feel. You make my love go. You make my love go. You make my love go"
}
]
```
### Dataset Fields
The dataset has the following fields (also called "features"):
```json
{
"target": "ClassLabel(num_classes=6, names=['Dance', 'Heavy Metal', 'Hip Hop', 'Indie', 'Pop', 'Rock'], names_file=None, id=None)",
"text": "Value(dtype='string', id=None)"
}
```
### Dataset Splits
This dataset is split into a train and validation split. The split sizes are as follow:
| Split name | Num samples |
| ------------ | ------------------- |
| train | 48493 |
| valid | 5389 |
|
true | ## Data Description
Long-COVID related articles have been manually collected by information specialists.
Please find further information [here](https://doi.org/10.1093/database/baac048).
## Size
||Training|Development|Test|Total|
|--|--|--|--|--|
Positive Examples|215|76|70|345|
Negative Examples|199|62|68|345|
Total|414|238|138|690|
## Citation
@article{10.1093/database/baac048,
author = {Langnickel, Lisa and Darms, Johannes and Heldt, Katharina and Ducks, Denise and Fluck, Juliane},
title = "{Continuous development of the semantic search engine preVIEW: from COVID-19 to long COVID}",
journal = {Database},
volume = {2022},
year = {2022},
month = {07},
issn = {1758-0463},
doi = {10.1093/database/baac048},
url = {https://doi.org/10.1093/database/baac048},
note = {baac048},
eprint = {https://academic.oup.com/database/article-pdf/doi/10.1093/database/baac048/44371817/baac048.pdf},
} |
false |
# Dataset Card for Beans
## Table of Contents
- [Table of Contents](#table-of-contents)
- [Dataset Description](#dataset-description)
- [Dataset Summary](#dataset-summary)
- [Supported Tasks and Leaderboards](#supported-tasks-and-leaderboards)
- [Languages](#languages)
- [Dataset Structure](#dataset-structure)
- [Data Instances](#data-instances)
- [Data Fields](#data-fields)
- [Data Splits](#data-splits)
- [Dataset Creation](#dataset-creation)
- [Curation Rationale](#curation-rationale)
- [Source Data](#source-data)
- [Annotations](#annotations)
- [Personal and Sensitive Information](#personal-and-sensitive-information)
- [Considerations for Using the Data](#considerations-for-using-the-data)
- [Social Impact of Dataset](#social-impact-of-dataset)
- [Discussion of Biases](#discussion-of-biases)
- [Other Known Limitations](#other-known-limitations)
- [Additional Information](#additional-information)
- [Dataset Curators](#dataset-curators)
- [Licensing Information](#licensing-information)
- [Citation Information](#citation-information)
- [Contributions](#contributions)
## Dataset Description
- **Homepage:**[Beans Homepage](https://github.com/AI-Lab-Makerere/ibean/)
- **Repository:**[AI-Lab-Makerere/ibean](https://github.com/AI-Lab-Makerere/ibean/)
- **Paper:** N/A
- **Leaderboard:** N/A
- **Point of Contact:** N/A
### Dataset Summary
Beans leaf dataset with images of diseased and health leaves.
### Supported Tasks and Leaderboards
- image-classification
### Languages
English
## Dataset Structure
### Data Instances
A sample from the training set is provided below:
```
{
'image_file_path': '/root/.cache/huggingface/datasets/downloads/extracted/0aaa78294d4bf5114f58547e48d91b7826649919505379a167decb629aa92b0a/train/bean_rust/bean_rust_train.109.jpg',
'labels': 1
}
```
### Data Fields
The data instances have the following fields:
- `image_file_path`: a `string` filepath to an image.
- `labels`: an `int` classification label.
### Data Splits
| name |train|validation|test|
|----------|----:|----:|----:|
|beans|1034|133|128|
## Dataset Creation
### Curation Rationale
[More Information Needed]
### Source Data
#### Initial Data Collection and Normalization
[More Information Needed]
#### Who are the source language producers?
[More Information Needed]
### Annotations
#### Annotation process
[More Information Needed]
#### Who are the annotators?
[More Information Needed]
### Personal and Sensitive Information
[More Information Needed]
## Considerations for Using the Data
### Social Impact of Dataset
[More Information Needed]
### Discussion of Biases
[More Information Needed]
### Other Known Limitations
[More Information Needed]
## Additional Information
### Dataset Curators
[More Information Needed]
### Licensing Information
[More Information Needed]
### Citation Information
```
@ONLINE {beansdata,
author="Makerere AI Lab",
title="Bean disease dataset",
month="January",
year="2020",
url="https://github.com/AI-Lab-Makerere/ibean/"
}
```
### Contributions
Thanks to [@nateraw](https://github.com/nateraw) for adding this dataset.
|
false |
# Dataset Card for Cats Vs. Dogs
## Table of Contents
- [Table of Contents](#table-of-contents)
- [Dataset Description](#dataset-description)
- [Dataset Summary](#dataset-summary)
- [Supported Tasks and Leaderboards](#supported-tasks-and-leaderboards)
- [Languages](#languages)
- [Dataset Structure](#dataset-structure)
- [Data Instances](#data-instances)
- [Data Fields](#data-fields)
- [Data Splits](#data-splits)
- [Dataset Creation](#dataset-creation)
- [Curation Rationale](#curation-rationale)
- [Source Data](#source-data)
- [Annotations](#annotations)
- [Personal and Sensitive Information](#personal-and-sensitive-information)
- [Considerations for Using the Data](#considerations-for-using-the-data)
- [Social Impact of Dataset](#social-impact-of-dataset)
- [Discussion of Biases](#discussion-of-biases)
- [Other Known Limitations](#other-known-limitations)
- [Additional Information](#additional-information)
- [Dataset Curators](#dataset-curators)
- [Licensing Information](#licensing-information)
- [Citation Information](#citation-information)
- [Contributions](#contributions)
## Dataset Description
- **Homepage:**[Cats vs Dogs Dataset](https://www.microsoft.com/en-us/download/details.aspx?id=54765)
- **Repository:** N/A
- **Paper:**[Paper](https://www.microsoft.com/en-us/research/wp-content/uploads/2007/10/CCS2007.pdf)
- **Leaderboard:** N/A
- **Point of Contact:** N/A
### Dataset Summary
A large set of images of cats and dogs. There are 1738 corrupted images that are dropped.
### Supported Tasks and Leaderboards
- image-classification
### Languages
English
## Dataset Structure
### Data Instances
A sample from the training set is provided below:
```
{
'image': '/root/.cache/huggingface/datasets/downloads/extracted/6e1e8c9052e9f3f7ecbcb4b90860668f81c1d36d86cc9606d49066f8da8bfb4f/PetImages/Cat/1.jpg',
'label': 0
}
```
### Data Fields
The data instances have the following fields:
- `image_file_path`: a `string` filepath to an image.
- `labels`: an `int` classification label.
### Data Splits
| name |train|
|----------|----:|
|cats_and_dogs|23410|
## Dataset Creation
### Curation Rationale
[More Information Needed]
### Source Data
#### Initial Data Collection and Normalization
[More Information Needed]
#### Who are the source language producers?
[More Information Needed]
### Annotations
#### Annotation process
[More Information Needed]
#### Who are the annotators?
[More Information Needed]
### Personal and Sensitive Information
[More Information Needed]
## Considerations for Using the Data
### Social Impact of Dataset
[More Information Needed]
### Discussion of Biases
[More Information Needed]
### Other Known Limitations
[More Information Needed]
## Additional Information
### Dataset Curators
[More Information Needed]
### Licensing Information
[More Information Needed]
### Citation Information
```
@Inproceedings (Conference){asirra-a-captcha-that-exploits-interest-aligned-manual-image-categorization,
author = {Elson, Jeremy and Douceur, John (JD) and Howell, Jon and Saul, Jared},
title = {Asirra: A CAPTCHA that Exploits Interest-Aligned Manual Image Categorization},
booktitle = {Proceedings of 14th ACM Conference on Computer and Communications Security (CCS)},
year = {2007},
month = {October},
publisher = {Association for Computing Machinery, Inc.},
url = {https://www.microsoft.com/en-us/research/publication/asirra-a-captcha-that-exploits-interest-aligned-manual-image-categorization/},
edition = {Proceedings of 14th ACM Conference on Computer and Communications Security (CCS)},
}
```
### Contributions
Thanks to [@nateraw](https://github.com/nateraw) for adding this dataset.
|
false |
# Dataset Card for Food-101
## Table of Contents
- [Table of Contents](#table-of-contents)
- [Dataset Description](#dataset-description)
- [Dataset Summary](#dataset-summary)
- [Supported Tasks and Leaderboards](#supported-tasks-and-leaderboards)
- [Languages](#languages)
- [Dataset Structure](#dataset-structure)
- [Data Instances](#data-instances)
- [Data Fields](#data-fields)
- [Data Splits](#data-splits)
- [Dataset Creation](#dataset-creation)
- [Curation Rationale](#curation-rationale)
- [Source Data](#source-data)
- [Annotations](#annotations)
- [Personal and Sensitive Information](#personal-and-sensitive-information)
- [Considerations for Using the Data](#considerations-for-using-the-data)
- [Social Impact of Dataset](#social-impact-of-dataset)
- [Discussion of Biases](#discussion-of-biases)
- [Other Known Limitations](#other-known-limitations)
- [Additional Information](#additional-information)
- [Dataset Curators](#dataset-curators)
- [Licensing Information](#licensing-information)
- [Citation Information](#citation-information)
- [Contributions](#contributions)
## Dataset Description
- **Homepage:**[Food-101 Dataset](https://data.vision.ee.ethz.ch/cvl/datasets_extra/food-101/)
- **Repository:** N/A
- **Paper:**[Paper](https://data.vision.ee.ethz.ch/cvl/datasets_extra/food-101/static/bossard_eccv14_food-101.pdf)
- **Leaderboard:** N/A
- **Point of Contact:** N/A
### Dataset Summary
This dataset consists of 101 food categories, with 101'000 images. For each class, 250 manually reviewed test images are provided as well as 750 training images. On purpose, the training images were not cleaned, and thus still contain some amount of noise. This comes mostly in the form of intense colors and sometimes wrong labels. All images were rescaled to have a maximum side length of 512 pixels.
### Supported Tasks and Leaderboards
- image-classification
### Languages
English
## Dataset Structure
### Data Instances
A sample from the training set is provided below:
```
{
'image': '/root/.cache/huggingface/datasets/downloads/extracted/6e1e8c9052e9f3f7ecbcb4b90860668f81c1d36d86cc9606d49066f8da8bfb4f/food-101/images/churros/1004234.jpg',
'label': 23
}
```
### Data Fields
The data instances have the following fields:
- `image`: a `string` filepath to an image.
- `label`: an `int` classification label.
### Data Splits
| name |train|validation|
|----------|----:|---------:|
|food101|75750|25250|
## Dataset Creation
### Curation Rationale
[More Information Needed]
### Source Data
#### Initial Data Collection and Normalization
[More Information Needed]
#### Who are the source language producers?
[More Information Needed]
### Annotations
#### Annotation process
[More Information Needed]
#### Who are the annotators?
[More Information Needed]
### Personal and Sensitive Information
[More Information Needed]
## Considerations for Using the Data
### Social Impact of Dataset
[More Information Needed]
### Discussion of Biases
[More Information Needed]
### Other Known Limitations
[More Information Needed]
## Additional Information
### Dataset Curators
[More Information Needed]
### Licensing Information
[More Information Needed]
### Citation Information
```
@inproceedings{bossard14,
title = {Food-101 -- Mining Discriminative Components with Random Forests},
author = {Bossard, Lukas and Guillaumin, Matthieu and Van Gool, Luc},
booktitle = {European Conference on Computer Vision},
year = {2014}
}
```
### Contributions
Thanks to [@nateraw](https://github.com/nateraw) for adding this dataset.
|
false |
# Dataset Card for "squad_v2_sv"
## Table of Contents
- [Dataset Description](#dataset-description)
- [Dataset Summary](#dataset-summary)
- [Supported Tasks](#supported-tasks)
- [Languages](#languages)
- [Dataset Structure](#dataset-structure)
- [Data Instances](#data-instances)
- [Data Fields](#data-fields)
- [Data Splits Sample Size](#data-splits-sample-size)
- [Dataset Creation](#dataset-creation)
- [Curation Rationale](#curation-rationale)
- [Source Data](#source-data)
- [Annotations](#annotations)
- [Personal and Sensitive Information](#personal-and-sensitive-information)
- [Considerations for Using the Data](#considerations-for-using-the-data)
- [Social Impact of Dataset](#social-impact-of-dataset)
- [Discussion of Biases](#discussion-of-biases)
- [Other Known Limitations](#other-known-limitations)
- [Additional Information](#additional-information)
- [Dataset Curators](#dataset-curators)
- [Licensing Information](#licensing-information)
- [Citation Information](#citation-information)
- [Contributions](#contributions)
## Dataset Description
- **Homepage:** [https://github.com/susumu2357/SQuAD_v2_sv](https://github.com/susumu2357/SQuAD_v2_sv)
- **Repository:** [https://github.com/susumu2357/SQuAD_v2_sv](https://github.com/susumu2357/SQuAD_v2_sv)
- **Paper:** None
- **Point of Contact:** [More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
- **Size of downloaded dataset files:** 10.09 MB
- **Size of the generated dataset:** 113.27 MB
- **Total amount of disk used:** 123.36 MB
### Dataset Summary
SQuAD_v2_sv is a Swedish version of SQuAD2.0. Translation was done automatically using the Google Translate API but it is not so straightforward for the following reasons.
- The span that determines the start and end of the answer in the context may change after translation.
- If the context and the answer are translated independently, the translated answer may not be included in the translated context.
Details on how to handle these dificulties are described in the git hub repo.
### Supported Tasks
[More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
### Languages
Swedish
## Dataset Structure
### Data Fields
The data fields are the same among all splits.
#### squad_v2
- `id`: a `string` feature.
- `title`: a `string` feature.
- `context`: a `string` feature.
- `question`: a `string` feature.
- `answers`: a dictionary feature containing:
- `text`: a `string` feature.
- `answer_start`: a `int32` feature.
### Data Splits Sample Size
| name |train |validation|
|--------|-----:|---------:|
|squad_v2_Sv|113898| 11156|
## Dataset Creation
### Curation Rationale
[More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
### Source Data
[More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
### Annotations
[More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
### Personal and Sensitive Information
[More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
## Considerations for Using the Data
### Social Impact of Dataset
[More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
### Discussion of Biases
[More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
### Other Known Limitations
[More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
## Additional Information
### Dataset Curators
[More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
### Licensing Information
[More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
### Citation Information
```
@misc{squad_v2_sv,
author = {Susumu Okazawa},
title = {Swedish translation of SQuAD2.0},
year = {2021},
publisher = {GitHub},
journal = {GitHub repository},
howpublished = {\url{https://github.com/susumu2357/SQuAD_v2_sv}}
``` |
false |
# Dataset Card for NMSQA(Natural Multi-speaker Spoken Question Answering)
## Table of Contents
- [Table of Contents](#table-of-contents)
- [Dataset Description](#dataset-description)
- [Dataset Summary](#dataset-summary)
- [Supported Tasks and Leaderboards](#supported-tasks-and-leaderboards)
- [Languages](#languages)
- [Dataset Structure](#dataset-structure)
- [Data Instances](#data-instances)
- [Data Fields](#data-fields)
- [Data Splits](#data-splits)
- [Dataset Creation](#dataset-creation)
- [Curation Rationale](#curation-rationale)
- [Source Data](#source-data)
- [Annotations](#annotations)
- [Personal and Sensitive Information](#personal-and-sensitive-information)
- [Considerations for Using the Data](#considerations-for-using-the-data)
- [Social Impact of Dataset](#social-impact-of-dataset)
- [Discussion of Biases](#discussion-of-biases)
- [Other Known Limitations](#other-known-limitations)
- [Additional Information](#additional-information)
- [Dataset Curators](#dataset-curators)
- [Licensing Information](#licensing-information)
- [Citation Information](#citation-information)
- [Contributions](#contributions)
## Dataset Description
- Homepage:
https://github.com/DanielLin94144/DUAL-textless-SQA
- Repository:
https://github.com/DanielLin94144/DUAL-textless-SQA
- Paper:
https://arxiv.org/abs/2203.04911
- Leaderboard:
- Point of Contact:
Download audio data: [https://huggingface.co/datasets/voidful/NMSQA/resolve/main/nmsqa_audio.tar.gz](https://huggingface.co/datasets/voidful/NMSQA/resolve/main/nmsqa_audio.tar.gz)
Unzip audio data: `tar -xf nmsqa_audio.tar.gz`
### Dataset Summary
The Natural Multi-speaker Spoken Question Answering (NMSQA) dataset is designed for the task of textless spoken question answering. It is based on the SQuAD dataset and contains spoken questions and passages. The dataset includes the original text, transcriptions, and audio files of the spoken content. This dataset is created to evaluate the performance of models on textless spoken question answering tasks.
### Supported Tasks and Leaderboards
The primary task supported by this dataset is textless spoken question answering, where the goal is to answer questions based on spoken passages without relying on textual information. The dataset can also be used for automatic speech recognition tasks.
### Languages
The dataset is in English.
## Dataset Structure
### Data Instances
Each instance in the dataset contains the following fields:
- id: Unique identifier for the instance
- title: The title of the passage
- context: The passage text
- question: The question text
- - answer_start: The start index of the answer in the text
- audio_full_answer_end: The end position of the audio answer in seconds
- audio_full_answer_start: The start position of the audio answer in seconds
- audio_full_neg_answer_end: The end position of the audio answer in seconds for an incorrect answer with the same words
- audio_full_neg_answer_start: The start position of the audio answer in seconds for an incorrect answer with the same words
- audio_segment_answer_end: The end position of the audio answer in seconds for the segment
- audio_segment_answer_start: The start position of the audio answer in seconds for the segment
- text: The answer text
- content_segment_audio_path: The audio path for the content segment
- content_full_audio_path: The complete audio path for the content
- content_audio_sampling_rate: The audio sampling rate
- content_audio_speaker: The audio speaker
- content_segment_text: The segment text of the content
- content_segment_normalized_text: The normalized text for generating audio
- question_audio_path: The audio path for the question
- question_audio_sampling_rate: The audio sampling rate
- question_audio_speaker: The audio speaker
- question_normalized_text: The normalized text for generating audio
### Data Fields
The dataset includes the following data fields:
- id
- title
- context
- question
- answers
- content_segment_audio_path
- content_full_audio_path
- content_audio_sampling_rate
- content_audio_speaker
- content_segment_text
- content_segment_normalized_text
- question_audio_path
- question_audio_sampling_rate
- question_audio_speaker
- question_normalized_text
### Data Splits
The dataset is split into train, dev, and test sets.
## Dataset Creation
### Curation Rationale
The NMSQA dataset is created to address the challenge of textless spoken question answering, where the model must answer questions based on spoken passages without relying on textual information.
### Source Data
The NMSQA dataset is based on the SQuAD dataset, with spoken questions and passages created from the original text data.
#### Initial Data Collection and Normalization
The initial data collection involved converting the original SQuAD dataset's text-based questions and passages into spoken audio files. The text was first normalized, and then audio files were generated using text-to-speech methods.
#### Who are the source language producers?
The source language producers are the creators of the SQuAD dataset and the researchers who generated the spoken audio files for the NMSQA dataset.
### Annotations
#### Annotation process
The annotations for the NMSQA dataset are derived from the original SQuAD dataset. Additional annotations, such as audio start and end positions for correct and incorrect answers, as well as audio file paths and speaker information, are added by the dataset creators.
#### Who are the annotators?
The annotators for the NMSQA dataset are the creators of the SQuAD dataset and the researchers who generated the spoken audio files and additional annotations for the NMSQA dataset.
### Personal and Sensitive Information
The dataset does not contain any personal or sensitive information.
## Considerations for Using the Data
### Social Impact of Dataset
The NMSQA dataset contributes to the development and evaluation of models for textless spoken question answering tasks, which can lead to advancements in natural language processing and automatic speech recognition. Applications of these technologies can improve accessibility and convenience in various domains, such as virtual assistants, customer service, and voice-controlled devices.
### Discussion of Biases
The dataset inherits potential biases from the original SQuAD dataset, which may include biases in the selection of passages, questions, and answers. Additionally, biases may be introduced in the text-to-speech process and the choice of speakers used to generate the spoken audio files.
### Other Known Limitations
As the dataset is based on the SQuAD dataset, it shares the same limitations, including the fact that it is limited to the English language and mainly focuses on factual questions. Furthermore, the dataset may not cover a wide range of accents, dialects, or speaking styles.
## Additional Information
### Dataset Curators
The NMSQA dataset is curated by Guan-Ting Lin, Yung-Sung Chuang, Ho-Lam Chung, Shu-Wen Yang, Hsuan-Jui Chen, Shang-Wen Li, Abdelrahman Mohamed, Hung-Yi Lee, and Lin-Shan Lee.
### Licensing Information
The licensing information for the dataset is not explicitly mentioned.
### Citation Information
```css
@article{lin2022dual,
title={DUAL: Textless Spoken Question Answering with Speech Discrete Unit Adaptive Learning},
author={Lin, Guan-Ting and Chuang, Yung-Sung and Chung, Ho-Lam and Yang, Shu-wen and Chen, Hsuan-Jui and Li, Shang-Wen and Mohamed, Abdelrahman and Lee, Hung-yi and Lee, Lin-shan},
journal={arXiv preprint arXiv:2203.04911},
year={2022}
}
```
### Contributions
Thanks to [@voidful](https://github.com/voidful) for adding this dataset. |
true |
# Dataset Card of "unam_tesis"
## Dataset Description
- **Homepage:**
- **Repository:**
- **Paper:**
- **Leaderboard:**
- **Point of Contact:**
- [yiselclavel@gmail.com](mailto:yiselclavel@gmail.com)
- [isaac7isaias@gmail.com](mailto:isaac7isaias@gmail.com)
### Dataset Summary
El dataset unam_tesis cuenta con 1000 tesis de 5 carreras de la Universidad Nacional Autónoma de México (UNAM), 200 por carrera. Se pretende seguir incrementando este dataset con las demás carreras y más tesis.
### Supported Tasks and Leaderboards
text-classification
### Languages
Español (es)
## Dataset Structure
### Data Instances
Las instancias del dataset son de la siguiente forma:
El objetivo de esta tesis es elaborar un estudio de las condiciones asociadas al aprendizaje desde casa a nivel preescolar y primaria en el municipio de Nicolás Romero a partir de la cancelación de clases presenciales ante la contingencia sanitaria del Covid-19 y el entorno familiar del estudiante. En México, la Encuesta para la Medición del Impacto COVID-19 en la Educación (ECOVID-ED) 2020, es un proyecto que propone el INEGI y realiza de manera especial para conocer las necesidades de la población estudiantil de 3 a 29 años de edad, saber qué está sucediendo con su entorno inmediato, las condiciones en las que desarrollan sus actividades académicas y el apoyo que realizan padres, tutores o cuidadores principales de las personas en edad formativa. La ECOVID-ED 2020 se llevó a cabo de manera especial con el objetivo de conocer el impacto de la cancelación provisional de clases presenciales en las instituciones educativas del país para evitar los contagios por la pandemia COVID-19 en la experiencia educativa de niños, niñas, adolescentes y jóvenes de 3 a 29 años, tanto en el ciclo escolar 2019-2020, como en ciclo 2020-2021. En este ámbito de investigación, el Instituto de Investigaciones sobre la Universidad y la Educación (IISUE) de la Universidad Nacional Autónoma de México público en 2020 la obra “Educación y Pandemia: Una visión académica” que se integran 34 trabajos que abordan la muy amplia temática de la educación y la universidad con reflexiones y ejercicios analíticos estrechamente relacionadas en el marco coyuntural de la pandemia COVID-19. La tesis se presenta en tres capítulos: En el capítulo uno se realizará una descripción del aprendizaje de los estudiantes a nivel preescolar y primaria del municipio de NicolásRomero, Estado de México, que por motivo de la contingencia sanitaria contra el Covid-19 tuvieron que concluir su ciclo académico 2019-2020 y el actual ciclo 2020-2021 en su casa debido a la cancelación provisional de clases presenciales y bajo la tutoría de padres, familiar o ser cercano; así como las horas destinadas al estudio y las herramientas tecnológicas como teléfonos inteligentes, computadoras portátiles, computadoras de escritorio, televisión digital y tableta. En el capítulo dos, se presentarán las herramientas necesarias para la captación de la información mediante técnicas de investigación social, a través de las cuales se mencionará, la descripción, contexto y propuestas del mismo, considerando los diferentes tipos de cuestionarios, sus componentes y diseño, teniendo así de manera específica la diversidad de ellos, que llevarán como finalidad realizar el cuestionario en línea para la presente investigación. Posteriormente, se podrá destacar las fases del diseño de la investigación, que se realizarán mediante una prueba piloto tomando como muestra a distintos expertos en el tema. De esta manera se obtendrá la información relevante para estudiarla a profundidad. En el capítulo tres, se realizará el análisis apoyado de las herramientas estadísticas, las cuales ofrecen explorar la muestra de una manera relevante, se aplicará el método inferencial para expresar la información y predecir las condiciones asociadas al autoaprendizaje, la habilidad pedagógica de padres o tutores, la convivencia familiar, la carga académica y actividades escolares y condicionamiento tecnológico,con la finalidad de inferir en la población. Asimismo, se realizarán pruebas de hipótesis, tablas de contingencia y matriz de correlación. Por consiguiente, los resultados obtenidos de las estadísticas se interpretarán para describir las condiciones asociadas y como impactan en la enseñanza de preescolar y primaria desde casa.|María de los Ángeles|Blancas Regalado|Análisis de las condiciones del aprendizaje desde casa en los alumnos de preescolar y primaria del municipio de Nicolás Romero |2022|Actuaría
| Carreras | Número de instancias |
|--------------|----------------------|
| Actuaría | 200 |
| Derecho| 200 |
| Economía| 200 |
| Psicología| 200 |
| Química Farmacéutico Biológica| 200 |
### Data Fields
El dataset está compuesto por los siguientes campos: "texto|titulo|carrera". <br/>
texto: Se refiere al texto de la introducción de la tesis. <br/>
titulo: Se refiere al título de la tesis. <br/>
carrera: Se refiere al nombre de la carrera a la que pertenece la tesis. <br/>
### Data Splits
El dataset tiene 2 particiones: entrenamiento (train) y prueba (test).
| Partición | Número de instancias |
|--------------|-------------------|
| Entrenamiento | 800 |
| Prueba | 200 |
## Dataset Creation
### Curation Rationale
La creación de este dataset ha sido motivada por la participación en el Hackathon 2022 de PLN en Español organizado por Somos NLP, con el objetivo de democratizar el NLP en español y promover su aplicación a buenas causas y, debido a que no existe un dataset de tesis en español.
### Source Data
#### Initial Data Collection and Normalization
El dataset original (dataset_tesis) fue creado a partir de un proceso de scraping donde se extrajeron tesis de la Universidad Nacional Autónoma de México en el siguiente link: https://tesiunam.dgb.unam.mx/F?func=find-b-0&local_base=TES01.
Se optó por realizar un scraper para conseguir la información. Se decidió usar la base de datos TESIUNAM, la cual es un catálogo en donde se pueden visualizar las tesis de los sustentantes que obtuvieron un grado en la UNAM, así como de las tesis de licenciatura de escuelas incorporadas a ella.
Para ello, en primer lugar se consultó la Oferta Académica (http://oferta.unam.mx/indice-alfabetico.html) de la Universidad, sitio de donde se extrajo cada una de las 131 licenciaturas en forma de lista. Después, se analizó cada uno de los casos presente en la base de datos, debido a que existen carreras con más de 10 tesis, otras con menos de 10, o con solo una o ninguna tesis disponible. Se usó Selenium para la interacción con un navegador Web (Edge) y está actualmente configurado para obtener las primeras 20 tesis, o menos, por carrera.
Este scraper obtiene de esta base de datos:
- Nombres del Autor
- Apellidos del Autor
- Título de la Tesis
- Año de la Tesis
- Carrera de la Tesis
A la vez, este scraper descarga cada una de las tesis en la carpeta Downloads del equipo local. En el csv formado por el scraper se añadió el "Resumen/Introduccion/Conclusion de la tesis", dependiendo cual primero estuviera disponible, ya que la complejidad recae en la diferencia de la estructura y formato de cada una de las tesis.
#### Who are the source language producers?
Los datos son creados por humanos de forma manual, en este caso por estudiantes de la UNAM y revisados por sus supervisores.
### Annotations
El dataset fue procesado para eliminar información innecesaria para los clasificadores. El dataset original cuenta con los siguientes campos: "texto|autor_nombre|autor_apellido|titulo|año|carrera".
#### Annotation process
Se extrajeron primeramente 200 tesis de 5 carreras de esta universidad: Actuaría, Derecho, Economía, Psicología y Química Farmacéutico Biológica. De estas se extrajo: introducción, nombre del autor, apellidos de autor, título de la tesis y la carrera. Los datos fueron revisados y limpiados por los autores.
Luego, el dataset fue procesado con las siguientes tareas de Procesamiento de Lenguaje Natural (dataset_tesis_procesado):
- convertir a minúsculas
- tokenización
- eliminar palabras que no son alfanuméricas
- eliminar palabras vacías
- stemming: eliminar plurales
#### Who are the annotators?
Las anotaciones fueron hechas por humanos, en este caso los autores del dataset, usando código de máquina en el lenguaje Python.
### Personal and Sensitive Information
[More Information Needed]
## Considerations for Using the Data
### Social Impact of Dataset
El presente conjunto de datos favorecerá la búsqueda e investigación relacionada con tesis en español, a partir de su categorización automática por un modelo entrenado con este dataset. Esta tarea favorece el cumplimiento del objetivo 4 de Desarrollo Sostenible de la ONU: Educación y Calidad (https://www.un.org/sustainabledevelopment/es/objetivos-de-desarrollo-sostenible/).
### Discussion of Biases
El texto tiene algunos errores en la codificación por lo que algunos caracteres como las tildes no se muestran correctamente. Las palabras con estos caracteres son eliminadas en el procesamiento hasta que se corrija el problema.
### Other Known Limitations
[More Information Needed]
## Additional Information
### Dataset Curators
Miembros del equipo (user de Hugging Face):
[Isacc Isahias López López](https://huggingface.co/MajorIsaiah)
[Yisel Clavel Quintero](https://huggingface.co/clavel)
[Dionis López](https://huggingface.co/inoid)
[Ximena Yeraldin López López](https://huggingface.co/Ximyer)
### Licensing Information
La versión 1.0.0 del dataset unam_tesis está liberada bajo la licencia <a href='http://www.apache.org/licenses/LICENSE-2.0'/> Apache-2.0 License </a>.
### Citation Information
"Esta base de datos se ha creado en el marco del Hackathon 2022 de PLN en Español organizado por Somos NLP patrocinado por Platzi, Paperspace y Hugging Face: https://huggingface.co/hackathon-pln-es."
Para citar este dataset, por favor, use el siguiente formato de cita:
@inproceedings{Hackathon 2022 de PLN en Español,
title={UNAM's Theses with BETO fine-tuning classify},
author={López López, Isaac Isaías; Clavel Quintero, Yisel; López Ramos, Dionis & López López, Ximena Yeraldin},
booktitle={Hackathon 2022 de PLN en Español},
year={2022}
}
### Contributions
Gracias a [@yiselclavel](https://github.com/yiselclavel) y [@IsaacIsaias](https://github.com/IsaacIsaias) por agregar este dataset.
|
true |
This file contains news texts (sentences) belonging to 5 different news categories (political, business, technology, sports and Entertainment). The original dataset was released by Nisansa de Silva (*Sinhala Text Classification: Observations from the Perspective of a Resource Poor Language, 2015*). The original dataset is processed and cleaned of single word texts, English only sentences etc.
If you use this dataset, please cite {*Nisansa de Silva, Sinhala Text Classification: Observations from the Perspective of a Resource Poor Language, 2015*} and {*Dhananjaya et al. BERTifying Sinhala - A Comprehensive Analysis of Pre-trained Language Models for Sinhala Text Classification, 2022*} |
true |
This dataset contains Sinhala news headlines extracted from 9 news sources (websites) (Sri Lanka Army, Dinamina, GossipLanka, Hiru, ITN, Lankapuwath, NewsLK,
Newsfirst, World Socialist Web Site-Sinhala). This is a processed version of the corpus created by *Sachintha, D., Piyarathna, L., Rajitha, C., and Ranathunga, S. (2021). Exploiting parallel corpora to improve multilingual embedding based document and sentence alignment*. Single word sentences, invalid characters have been removed from the originally extracted corpus and also subsampled to handle class imbalance.
If you use this dataset please cite {*Dhananjaya et al. BERTifying Sinhala - A Comprehensive Analysis of Pre-trained Language Models for Sinhala Text Classification, 2022*} |
false |
# Dataset Card for Goud summarization dataset
## Table of Contents
- [Table of Contents](#table-of-contents)
- [Dataset Description](#dataset-description)
- [Dataset Summary](#dataset-summary)
- [Supported Tasks and Leaderboards](#supported-tasks-and-leaderboards)
- [Languages](#languages)
- [Dataset Structure](#dataset-structure)
- [Data Instances](#data-instances)
- [Data Fields](#data-fields)
- [Data Splits](#data-splits)
- [Dataset Creation](#dataset-creation)
- [Curation Rationale](#curation-rationale)
- [Source Data](#source-data)
- [Annotations](#annotations)
- [Personal and Sensitive Information](#personal-and-sensitive-information)
- [Considerations for Using the Data](#considerations-for-using-the-data)
- [Social Impact of Dataset](#social-impact-of-dataset)
- [Discussion of Biases](#discussion-of-biases)
- [Other Known Limitations](#other-known-limitations)
- [Additional Information](#additional-information)
- [Dataset Curators](#dataset-curators)
- [Licensing Information](#licensing-information)
- [Citation Information](#citation-information)
- [Contributions](#contributions)
## Dataset Description
- **Homepage:**[Needs More Information]
- **Repository:**[Needs More Information]
- **Paper:**[Goud.ma: a News Article Dataset for Summarization in Moroccan Darija](https://openreview.net/forum?id=BMVq5MELb9)
- **Leaderboard:**[Needs More Information]
- **Point of Contact:**[Needs More Information]
### Dataset Summary
Goud-sum contains 158k articles and their headlines extracted from [Goud.ma](https://www.goud.ma/) news website. The articles are written in the Arabic script. All headlines are in Moroccan Darija, while articles may be in Moroccan Darija, in Modern Standard Arabic, or a mix of both (code-switched Moroccan Darija).
### Supported Tasks and Leaderboards
Text Summarization
### Languages
* Moroccan Arabic (Darija)
* Modern Standard Arabic
## Dataset Structure
### Data Instances
The dataset consists of article-headline pairs in string format.
### Data Fields
* article: a string containing the body of the news article
* headline: a string containing the article's headline
* categories: a list of string of article categories
### Data Splits
Goud-sum dataset has 3 splits: _train_, _validation_, and _test_. Below are the number of instances in each split.
| Dataset Split | Number of Instances in Split |
| ------------- | ------------------------------------------- |
| Train | 139,288 |
| Validation | 9,497 |
| Test | 9,497 |
## Dataset Creation
### Curation Rationale
[More Information Needed]
### Source Data
#### Initial Data Collection and Normalization
[More Information Needed]
#### Who are the source language producers?
The text was written by journalists at [Goud](https://www.goud.ma/).
### Annotations
The dataset does not contain any additional annotations.
#### Annotation process
[N/A]
#### Who are the annotators?
[N/A]
### Personal and Sensitive Information
[More Information Needed]
## Considerations for Using the Data
### Social Impact of Dataset
[More Information Needed]
### Discussion of Biases
[More Information Needed]
### Other Known Limitations
[More Information Needed]
## Additional Information
### Dataset Curators
[More Information Needed]
### Licensing Information
[More Information Needed]
### Citation Information
```
@inproceedings{issam2022goudma,
title={Goud.ma: a News Article Dataset for Summarization in Moroccan Darija},
author={Abderrahmane Issam and Khalil Mrini},
booktitle={3rd Workshop on African Natural Language Processing},
year={2022},
url={https://openreview.net/forum?id=BMVq5MELb9}
}
```
### Contributions
Thanks to [@issam9](https://github.com/issam9) and [@KhalilMrini](https://github.com/KhalilMrini) for adding this dataset.
|
false |
# VQGAN Pairs
This dataset contains ~2.4 million image pairs intended for improvement of image quality in VQGAN predictions. Each pair consists of:
- A 512x512 crop of an image taken from Open Images.
- A 256x256 image encoded and decoded using VQGAN, corresponding to the same image crop as the original.
This is the VQGAN implementation that was used for encoding and decoding: https://github.com/patil-suraj/vqgan-jax
# License
This dataset is created using Open Images, which has the following license:
The annotations are licensed by Google LLC under CC BY 4.0 license. The images are listed as having a CC BY 2.0 license. Note: while we tried to identify images that are licensed under a Creative Commons Attribution license, we make no representations or warranties regarding the license status of each image and you should verify the license for each image yourself. |
false | ## Table of Contents
- [Dataset Description](#dataset-description)
- [Dataset Summary](#dataset-summary)
- [Supported Tasks and Leaderboards](#supported-tasks-and-leaderboards)
- [Languages](#languages)
- [Dataset Structure](#dataset-structure)
- [Data Instances](#data-instances)
- [Data Fields](#data-fields)
- [Data Splits](#data-splits)
- [Dataset Creation](#dataset-creation)
- [Curation Rationale](#curation-rationale)
- [Source Data](#source-data)
- [Annotations](#annotations)
- [Personal and Sensitive Information](#personal-and-sensitive-information)
- [Considerations for Using the Data](#considerations-for-using-the-data)
- [Social Impact of Dataset](#social-impact-of-dataset)
- [Discussion of Biases](#discussion-of-biases)
- [Other Known Limitations](#other-known-limitations)
- [Additional Information](#additional-information)
- [Dataset Curators](#dataset-curators)
- [Licensing Information](#licensing-information)
- [Citation Information](#citation-information)
- [Contributions](#contributions)
## Dataset Description
- **Repository:** [Github](https://github.com/ncats/epi4GARD/tree/master/EpiExtract4GARD#epiextract4gard)
- **Paper:** Pending
### Dataset Summary
EpiSet4NER-v2 is a gold-standard dataset for epidemiological entity recognition of location, epidemiologic types (e.g. "prevalence", "annual incidence", "estimated occurrence"), and epidemiological rates (e.g. "1.7 per 1,000,000 live births", "2.1:1.000.000", "one in five million", "0.03%") created by the [Genetic and Rare Diseases Information Center (GARD)](https://rarediseases.info.nih.gov/), a program in [the National Center for Advancing Translational Sciences](https://ncats.nih.gov/), one of the 27 [National Institutes of Health](https://www.nih.gov/). It was labeled programmatically using spaCy NER and rule-based methods, then manually validated by biomedical researchers, including a GARD curator (genetic and rare disease expert). This weakly-supervised teaching method allowed us to construct this high quality dataset in an efficient manner and achieve satisfactory performance on a multi-type token classification problem. It was used to train [EpiExtract4GARD-v2](https://huggingface.co/ncats/EpiExtract4GARD-v2), a BioBERT-based model fine-tuned for NER.
### Data Fields
The data fields are the same among all splits.
- `id`: a `string` feature that indicates sentence number.
- `tokens`: a `list` of `string` features.
- `ner_tags`: a `list` of classification labels, with possible values including `O` (0), `B-LOC` (1), `I-LOC` (2), `B-EPI` (3), `I-EPI` (4),`B-STAT` (5),`I-STAT` (6).
### Data Splits
|name |train |validation|test|
|---------|-----:|----:|----:|
|EpiSet \# of abstracts|456|114|50|
|EpiSet \# tokens |117888|31262|13910|
## Dataset Creation
*Figure 1:* Creation of EpiSet4NER by NIH/NCATS
Comparing the programmatically labeled test set to the manually corrected test set allowed us to measure the precision, recall, and F1 of the programmatic labeling.
*Table 1:* Programmatic labeling of EpiSet4NER
| Evaluation Level | Entity | Precision | Recall | F1 |
|:----------------:|:------------------------:|:---------:|:------:|:-----:|
| Entity-Level | Overall | 0.559 | 0.662 | 0.606 |
| | Location | 0.597 | 0.661 | 0.627 |
| | Epidemiologic Type | 0.854 | 0.911 | 0.882 |
| | Epidemiologic Rate | 0.175 | 0.255 | 0.207 |
| Token-Level | Overall | 0.805 | 0.710 | 0.755 |
| | Location | 0.868 | 0.713 | 0.783 |
| | Epidemiologic Type | 0.908 | 0.908 | 0.908 |
| | Epidemiologic Rate | 0.739 | 0.645 | 0.689 |
An example of the text labeling:
*Figure 2:* Text Labeling using spaCy and rule-based labeling. Ideal labeling is bolded on the left. Actual programmatic output is on the right. [\[Figure citation\]](https://pubmed.ncbi.nlm.nih.gov/33649778/)
### Curation Rationale
To train ML/DL models that automate the process of rare disease epidemiological curation. This is crucial information to patients & families, researchers, grantors, and policy makers, primarily for funding purposes.
### Source Data
620 rare disease abstracts classified as epidemiological by a LSTM RNN rare disease epi classifier from 488 diseases. See Figure 1.
#### Initial Data Collection and Normalization
A random sample of 500 disease names were gathered from a list of ~6061 rare diseases tracked by GARD until ≥50 abstracts had been returned for each disease or the EBI RESTful API results were exhausted. Though we called ~25,000 abstracts from PubMed's db, only 7699 unique abstracts were returned for 488 diseases. Out of 7699 abstracts, only 620 were classified as epidemiological by the LSTM RNN epidemiological classifier.
### Annotations
#### Annotation process
Programmatic labeling. See [here](https://github.com/ncats/epi4GARD/blob/master/EpiExtract4GARD/create_labeled_dataset_V2.ipynb) and then [here](https://github.com/ncats/epi4GARD/blob/master/EpiExtract4GARD/modify_existing_labels.ipynb). The test set was manually corrected after creation.
#### Who are the annotators?
Programmatic labeling was done by [@William Kariampuzha](https://github.com/wzkariampuzha), one of the NCATS researchers.
The test set was manually corrected by 2 more NCATS researchers and a GARD curator (genetic and rare disease expert).
### Personal and Sensitive Information
None. These are freely available abstracts from PubMed.
## Considerations for Using the Data
### Social Impact of Dataset
Assisting 25-30 millions Americans with rare diseases. Additionally can be useful for Orphanet or CDC researchers/curators.
### Discussion of Biases and Limitations
- There were errors in the source file that contained rare disease synonyms of names, which may have led to some unrelated abstracts being included in the training, validation, and test sets.
- The abstracts were gathered through the EBI API and is thus subject to any biases that the EBI API had. The NCBI API returns very different results as shown by an API analysis here.
- The [long short-term memory recurrent neural network epi classifier](https://pubmed.ncbi.nlm.nih.gov/34457147/) was used to sift the 7699 rare disease abstracts. This model had a hold-out validation F1 score of 0.886 and a test F1 (which was compared against a GARD curator who used full-text articles to determine truth-value of epidemiological abstract) of 0.701. With 620 epi abstracts filtered from 7699 original rare disease abstracts, there are likely several false positives and false negative epi abstracts.
- Tokenization was done by spaCy which may be a limitation (or not) for current and future models trained on this set.
- The programmatic labeling was very imprecise as seen by Table 1. This is likely the largest limitation of the [BioBERT-based model](https://huggingface.co/ncats/EpiExtract4GARD) trained on this set.
- The test set was difficult to validate even for general NCATS researchers, which is why we relied on a rare disease expert to verify our modifications. As this task of epidemiological information identification is quite difficult for non-expert humans to complete, this set, and especially a gold-standard dataset in the possible future, represents a challenging gauntlet for NLP systems, especially those focusing on numeracy, to compete on.
## Additional Information
### Dataset Curators
[NIH GARD](https://rarediseases.info.nih.gov/about-gard/pages/23/about-gard)
### Licensing Information
[More Information Needed]
### Citation Information
[More Information Needed]
### Contributions
Thanks to [@William Kariampuzha](https://github.com/wzkariampuzha) at NCATS/Axle Informatics for adding this dataset. |
false |
# Dataset Card for UlyssesNER-Br
## Table of Contents
- [Table of Contents](#table-of-contents)
- [Dataset Description](#dataset-description)
- [Dataset Summary](#dataset-summary)
- [Supported Tasks and Leaderboards](#supported-tasks-and-leaderboards)
- [Languages](#languages)
- [Dataset Structure](#dataset-structure)
- [Data Instances](#data-instances)
- [Data Fields](#data-fields)
- [Data Splits](#data-splits)
- [Dataset Creation](#dataset-creation)
- [Curation Rationale](#curation-rationale)
- [Source Data](#source-data)
- [Annotations](#annotations)
- [Personal and Sensitive Information](#personal-and-sensitive-information)
- [Considerations for Using the Data](#considerations-for-using-the-data)
- [Social Impact of Dataset](#social-impact-of-dataset)
- [Discussion of Biases](#discussion-of-biases)
- [Other Known Limitations](#other-known-limitations)
- [Additional Information](#additional-information)
- [Dataset Curators](#dataset-curators)
- [Licensing Information](#licensing-information)
- [Citation Information](#citation-information)
- [Contributions](#contributions)
## Dataset Description
- **Homepage:** [Convenio-Camara-dos-Deputados/ulyssesner-br-propor](https://github.com/Convenio-Camara-dos-Deputados/ulyssesner-br-propor)
- **Repository:** [Convenio-Camara-dos-Deputados/ulyssesner-br-propor](https://github.com/Convenio-Camara-dos-Deputados/ulyssesner-br-propor)
- **Paper:** [UlyssesNER-Br: a Corpus of Brazilian Legislative Documents for Named Entity Recognition](https://link.springer.com/chapter/10.1007/978-3-030-98305-5_1)
- **Leaderboard:**
- **Point of Contact:**
### Dataset Summary
[More Information Needed]
### Supported Tasks and Leaderboards
[More Information Needed]
### Languages
Portuguese (Brazil).
## Dataset Structure
### Data Instances
[More Information Needed]
### Data Fields
[More Information Needed]
### Data Splits
[More Information Needed]
## Dataset Creation
### Curation Rationale
[More Information Needed]
### Source Data
#### Initial Data Collection and Normalization
[More Information Needed]
#### Who are the source language producers?
[More Information Needed]
### Annotations
#### Annotation process
[More Information Needed]
#### Who are the annotators?
[More Information Needed]
### Personal and Sensitive Information
[More Information Needed]
## Considerations for Using the Data
### Social Impact of Dataset
[More Information Needed]
### Discussion of Biases
[More Information Needed]
### Other Known Limitations
[More Information Needed]
## Additional Information
### Dataset Curators
[More Information Needed]
### Licensing Information
[More Information Needed]
### Citation Information
```
@inproceedings{UlyssesNER-Br,
title={UlyssesNER-Br: A Corpus of Brazilian Legislative Documents for Named Entity Recognition},
author={Albuquerque, Hidelberg O. and Costa, Rosimeire and Silvestre, Gabriel and Souza, Ellen and da Silva, Nádia F. F. and Vitório, Douglas and Moriyama, Gyovana and Martins, Lucas and Soezima, Luiza and Nunes, Augusto and Siqueira, Felipe and Tarrega, João P. and Beinotti, Joao V. and Dias, Marcio and Silva, Matheus and Gardini, Miguel and Silva, Vinicius and de Carvalho, André C. P. L. F. and Oliveira, Adriano L. I.},
booktitle={Computational Processing of the Portuguese Language},
year={2022},
publisher={Springer International Publishing},
isbn={978-3-030-98305-5},
doi={https://doi.org/10.1007/978-3-030-98305-5_1}
}
```
### Contributions
Thanks to [@augusnunes](https://github.com/augusnunes) for adding this dataset. |
true |
# Dataset Card for wikitext_linked
## Table of Contents
- [Table of Contents](#table-of-contents)
- [Dataset Description](#dataset-description)
- [Dataset Summary](#dataset-summary)
- [Supported Tasks and Leaderboards](#supported-tasks-and-leaderboards)
- [Languages](#languages)
- [Dataset Structure](#dataset-structure)
- [Data Instances](#data-instances)
- [Data Fields](#data-fields)
- [Data Splits](#data-splits)
- [Dataset Creation](#dataset-creation)
- [Curation Rationale](#curation-rationale)
- [Source Data](#source-data)
- [Annotations](#annotations)
- [Personal and Sensitive Information](#personal-and-sensitive-information)
- [Considerations for Using the Data](#considerations-for-using-the-data)
- [Social Impact of Dataset](#social-impact-of-dataset)
- [Discussion of Biases](#discussion-of-biases)
- [Other Known Limitations](#other-known-limitations)
- [Additional Information](#additional-information)
- [Dataset Curators](#dataset-curators)
- [Licensing Information](#licensing-information)
- [Citation Information](#citation-information)
- [Contributions](#contributions)
## Dataset Description
- **Homepage:** -
- **Repository:** [https://github.com/GabrielKP/svo/](https://github.com/GabrielKP/svo/)
- **Paper:** -
- **Leaderboard:** -
- **Point of Contact:** [gabriel.kressin@dfki.de](mailto:gabriel.kressin@dfki.de)
### Dataset Summary
The WikiText language modeling dataset is a collection of over 100 million tokens extracted from
the set of verified Good and Featured articles on Wikipedia. Dependency Relations, POS, NER tags
are marked with [trankit](https://github.com/nlp-uoregon/trankit), entities are linked with
[entity-fishing](https://nerd.readthedocs.io/en/latest/index.html), which also tags another field
of NER tags. The dataset is available under the Creative Commons Attribution-ShareAlike License.
Compared to the preprocessed version of Penn Treebank (PTB), WikiText-2 is over 2 times larger and
WikiText-103 is over 110 times larger. The WikiText dataset also features a far larger vocabulary
and retains the original case, punctuation and numbers - all of which are removed in PTB. As it is
composed of full articles, the dataset is well suited for models that can take advantage of long
term dependencies.
### Supported Tasks and Leaderboards
- masked-language-modeling
- named-entity-recognition
- part-of-speech
- lemmatization
- parsing
- entity-linking-classification
### Languages
English.
## Dataset Structure
### Data Instances
#### wikitext2
- **Size of downloaded dataset files:** 27.3 MB
- **Size of the generated dataset:** 197.2 MB
- **Total amount of disk used:** 197.2 MB
An example of 'validation' looks as follows.
```json
{
'text': 'It is closely related to the American lobster , H. americanus .',
'original_id': 3,
'tok_span': [[0, 0], [0, 2], [3, 5], [6, 13], [14, 21], [22, 24], [25, 28], [29, 37], [38, 45], [46, 47], [48, 50], [51, 61], [62, 63]],
'tok_upos': ['root', 'PRON', 'AUX', 'ADV', 'ADJ', 'ADP', 'DET', 'ADJ', 'NOUN', 'PUNCT', 'PROPN', 'PROPN', 'PUNCT'],
'tok_xpos': ['root', 'PRP', 'VBZ', 'RB', 'JJ', 'IN', 'DT', 'JJ', 'NN', ',', 'NNP', 'NNP', '.'],
'tok_dephead': [0, 4, 4, 4, 0, 8, 8, 8, 4, 8, 8, 10, 4],
'tok_deprel': ['root', 'nsubj', 'cop', 'advmod', 'root', 'case', 'det', 'amod', 'obl', 'punct', 'appos', 'flat', 'punct'],
'tok_lemma': [None, 'it', 'be', 'closely', 'related', 'to', 'the', 'american', 'lobster', ',', 'H.', 'americanus', '.'],
'tok_ner': [None, 'O', 'O', 'O', 'O', 'O', 'O', 'S-MISC', 'O', 'O', 'O', 'O', 'O'],
'ent_span': [[29, 45]],
'ent_wikipedia_external_ref': ['377397'],
'ent_ner': [None],
'ent_domains': [['Enterprise']],
}
```
#### wikitext103
- **Size of downloaded dataset files:** 1.11 GB
- **Size of the generated dataset:** 7.82 GB
- **Total amount of disk used:** 7.82 GB
An example of 'train' looks as follows.
```json
{
'text': 'Vision for the PlayStation Portable .',
'original_id': 3,
'tok_span': [[0, 0], [0, 6], [7, 10], [11, 14], [15, 26], [27, 35], [36, 37]],
'tok_upos': ['root', 'NOUN', 'ADP', 'DET', 'PROPN', 'PROPN', 'PUNCT'],
'tok_xpos': ['root', 'NN', 'IN', 'DT', 'NNP', 'NNP', '.'],
'tok_dephead': [0, 0, 5, 5, 5, 1, 1],
'tok_deprel': ['root', 'root', 'case', 'det', 'compound', 'nmod', 'punct'],
'tok_lemma': [None, 'vision', 'for', 'the', 'PlayStation', 'Portable', '.'],
'tok_ner': [None, 'O', 'O', 'O', 'B-MISC', 'E-MISC', 'O'],
'ent_span': [[15, 35]],
'ent_wikipedia_external_ref': ['619009'],
'ent_ner': [None],
'ent_domains': [['Electronics', 'Computer_Science']]
}
```
Use following code to print the examples nicely:
```py
def print_tokens_entities(example):
text = example['text']
print(
"Text:\n"
f" {text}"
"\nOrig-Id: "
f"{example['original_id']}"
"\nTokens:"
)
iterator = enumerate(zip(
example["tok_span"],
example["tok_upos"],
example["tok_xpos"],
example["tok_ner"],
example["tok_dephead"],
example["tok_deprel"],
example["tok_lemma"],
))
print(f" Id | {'token':12} | {'upos':8} | {'xpos':8} | {'ner':8} | {'deph':4} | {'deprel':9} | {'lemma':12} | Id")
print("---------------------------------------------------------------------------------------------------")
for idx, (tok_span, upos, xpos, ner, dephead, deprel, lemma) in iterator:
print(f" {idx:3} | {text[tok_span[0]:tok_span[1]]:12} | {upos:8} | {xpos:8} | {str(ner):8} | {str(dephead):4} | {deprel:9} | {str(lemma):12} | {idx}")
iterator = list(enumerate(zip(
example.get("ent_span", []),
example.get("ent_wikipedia_external_ref", []),
example.get("ent_ner", []),
example.get("ent_domains", []),
)))
if len(iterator) > 0:
print("Entities")
print(f" Id | {'entity':21} | {'wiki_ref':7} | {'ner':7} | domains")
print("--------------------------------------------------------------------")
for idx, ((start, end), wiki_ref, ent_ner, ent_domains) in iterator:
print(f" {idx:3} | {text[start:end]:21} | {str(wiki_ref):7} | {str(ent_ner):7} | {ent_domains}")
```
### Data Fields
The data fields are the same among all splits.
* text: string feature.
* original_id: int feature. Mapping to index within original wikitext dataset.
* tok_span: sequence of (int, int) tuples. Denotes token spans (start inclusive, end exclusive)
within each sentence.
**Note that each sentence includes an artificial root node to align dependency relations.**
* tok_upos: string feature. [Universal Dependency POS tag](https://universaldependencies.org/)
tags. Aligned with tok_span. Root node has tag "root".
* tok_xpos: string geature. [XPOS POS tag](https://trankit.readthedocs.io/en/latest/overview.html#token-list).
Aligned with tok_span. Root node has tag "root".
* tok_dephead: int feature.
[Universal Dependency Head Node](https://universaldependencies.org/introduction.html). Int refers
to tokens in tok_span. Root node has head `0` (itself).
* tok_deprel: [Universal Dependency Relation Description](https://universaldependencies.org/introduction.html).
Refers to the relation between this token and head token. Aligned with tok_span. Root node has
dependency relation "root" to itself.
* tok_lemma: string feature. Lemma of token. Aligend with tok_span.
* tok_ner: string feature. NER tag of token. Marked in BIOS schema (e.g. S-MISC, B-LOC, ...)
Aligned with tok_span. Root node has NER tag `None`.
* ent_span: sequence of (int, int) tuples. Denotes entities found by entity-fishing
(start inclusive, end exclusive).
* ent_wikipedia_external_ref: string feature. External Reference to wikipedia page. You can
access the wikipedia page via the url `https://en.wikipedia.org/wiki?curid=<ent_wikipedia_external_ref>`.
Aligend with ent_span. All entities either have this field, or the `ent_ner` field, but not both.
An empty field is denoted by the string `None`. Aligned with ent_span.
* ent_ner: string feature. Denotes NER tags. An empty field is denoted by the string `None`.
Aligned with ent_span.
"ent_domains": sequence of string. Denotes domains of entity. Can be empty sequence. Aligned with
ent_span.
### Data Splits
| name | train |validation| test|
|-------------------|------:|---------:|----:|
|wikitext103 |4076530| 8607|10062|
|wikitext2 | 82649| 8606|10062|
## Dataset Creation
### Curation Rationale
[More Information Needed]
### Source Data
#### Initial Data Collection and Normalization
[https://huggingface.co/datasets/wikitext](https://huggingface.co/datasets/wikitext)
#### Who are the source language producers?
[More Information Needed]
### Annotations
#### Annotation process
1. Started with `wikitext2-raw-v1` and `wikitext103-raw-v1` from [wikitext](https://huggingface.co/datasets/wikitext)
2. Ran datasets through Trankit. Marked all fields starting with `tok`.
In this step, the texts have been split into sentences. To retain the original text sections
you can accumulate over `original_id` (examples are in order).
3. Ran datasets through entity-fishing. Marked all fields starting with `ent`.
#### Who are the annotators?
Machines powered by [DFKI](https://www.dfki.de/web).
### Personal and Sensitive Information
[More Information Needed]
## Considerations for Using the Data
### Social Impact of Dataset
[More Information Needed]
### Discussion of Biases
[More Information Needed]
### Other Known Limitations
[More Information Needed]
## Additional Information
### Dataset Curators
[More Information Needed]
### Licensing Information
Creative Commons Attribution-ShareAlike 4.0 International (CC BY-SA 4.0)
### Citation Information
Please cite the original creators of wikitext, and the great people
developing trankit and entity-fishing.
```
@misc{merity2016pointer,
title={Pointer Sentinel Mixture Models},
author={Stephen Merity and Caiming Xiong and James Bradbury and Richard Socher},
year={2016},
eprint={1609.07843},
archivePrefix={arXiv},
primaryClass={cs.CL}
}
@inproceedings{nguyen2021trankit,
title={Trankit: A Light-Weight Transformer-based Toolkit for Multilingual Natural Language Processing},
author={Nguyen, Minh Van and Lai, Viet Dac and Veyseh, Amir Pouran Ben and Nguyen, Thien Huu},
booktitle="Proceedings of the 16th Conference of the European Chapter of the Association for Computational Linguistics: System Demonstrations",
year={2021}
}
@misc{entity-fishing,
title = {entity-fishing},
howpublished = {\\url{https://github.com/kermitt2/entity-fishing}},
publisher = {GitHub},
year = {2016--2022},
archivePrefix = {swh},
eprint = {1:dir:cb0ba3379413db12b0018b7c3af8d0d2d864139c}
}
```
### Contributions
Thanks to [@GabrielKP](https://github.com/GabrielKP) for adding this dataset.
|
false |
# Dataset Card for BEIR Benchmark
## Table of Contents
- [Dataset Description](#dataset-description)
- [Dataset Summary](#dataset-summary)
- [Supported Tasks and Leaderboards](#supported-tasks-and-leaderboards)
- [Languages](#languages)
- [Dataset Structure](#dataset-structure)
- [Data Instances](#data-instances)
- [Data Fields](#data-fields)
- [Data Splits](#data-splits)
- [Dataset Creation](#dataset-creation)
- [Curation Rationale](#curation-rationale)
- [Source Data](#source-data)
- [Annotations](#annotations)
- [Personal and Sensitive Information](#personal-and-sensitive-information)
- [Considerations for Using the Data](#considerations-for-using-the-data)
- [Social Impact of Dataset](#social-impact-of-dataset)
- [Discussion of Biases](#discussion-of-biases)
- [Other Known Limitations](#other-known-limitations)
- [Additional Information](#additional-information)
- [Dataset Curators](#dataset-curators)
- [Licensing Information](#licensing-information)
- [Citation Information](#citation-information)
- [Contributions](#contributions)
## Dataset Description
- **Homepage:** https://github.com/UKPLab/beir
- **Repository:** https://github.com/UKPLab/beir
- **Paper:** https://openreview.net/forum?id=wCu6T5xFjeJ
- **Leaderboard:** https://docs.google.com/spreadsheets/d/1L8aACyPaXrL8iEelJLGqlMqXKPX2oSP_R10pZoy77Ns
- **Point of Contact:** nandan.thakur@uwaterloo.ca
### Dataset Summary
BEIR is a heterogeneous benchmark that has been built from 18 diverse datasets representing 9 information retrieval tasks:
- Fact-checking: [FEVER](http://fever.ai), [Climate-FEVER](http://climatefever.ai), [SciFact](https://github.com/allenai/scifact)
- Question-Answering: [NQ](https://ai.google.com/research/NaturalQuestions), [HotpotQA](https://hotpotqa.github.io), [FiQA-2018](https://sites.google.com/view/fiqa/)
- Bio-Medical IR: [TREC-COVID](https://ir.nist.gov/covidSubmit/index.html), [BioASQ](http://bioasq.org), [NFCorpus](https://www.cl.uni-heidelberg.de/statnlpgroup/nfcorpus/)
- News Retrieval: [TREC-NEWS](https://trec.nist.gov/data/news2019.html), [Robust04](https://trec.nist.gov/data/robust/04.guidelines.html)
- Argument Retrieval: [Touche-2020](https://webis.de/events/touche-20/shared-task-1.html), [ArguAna](tp://argumentation.bplaced.net/arguana/data)
- Duplicate Question Retrieval: [Quora](https://www.quora.com/q/quoradata/First-Quora-Dataset-Release-Question-Pairs), [CqaDupstack](http://nlp.cis.unimelb.edu.au/resources/cqadupstack/)
- Citation-Prediction: [SCIDOCS](https://allenai.org/data/scidocs)
- Tweet Retrieval: [Signal-1M](https://research.signal-ai.com/datasets/signal1m-tweetir.html)
- Entity Retrieval: [DBPedia](https://github.com/iai-group/DBpedia-Entity/)
All these datasets have been preprocessed and can be used for your experiments.
```python
```
### Supported Tasks and Leaderboards
The dataset supports a leaderboard that evaluates models against task-specific metrics such as F1 or EM, as well as their ability to retrieve supporting information from Wikipedia.
The current best performing models can be found [here](https://eval.ai/web/challenges/challenge-page/689/leaderboard/).
### Languages
All tasks are in English (`en`).
## Dataset Structure
All BEIR datasets must contain a corpus, queries and qrels (relevance judgments file). They must be in the following format:
- `corpus` file: a `.jsonl` file (jsonlines) that contains a list of dictionaries, each with three fields `_id` with unique document identifier, `title` with document title (optional) and `text` with document paragraph or passage. For example: `{"_id": "doc1", "title": "Albert Einstein", "text": "Albert Einstein was a German-born...."}`
- `queries` file: a `.jsonl` file (jsonlines) that contains a list of dictionaries, each with two fields `_id` with unique query identifier and `text` with query text. For example: `{"_id": "q1", "text": "Who developed the mass-energy equivalence formula?"}`
- `qrels` file: a `.tsv` file (tab-seperated) that contains three columns, i.e. the `query-id`, `corpus-id` and `score` in this order. Keep 1st row as header. For example: `q1 doc1 1`
### Data Instances
A high level example of any beir dataset:
```python
corpus = {
"doc1" : {
"title": "Albert Einstein",
"text": "Albert Einstein was a German-born theoretical physicist. who developed the theory of relativity, \
one of the two pillars of modern physics (alongside quantum mechanics). His work is also known for \
its influence on the philosophy of science. He is best known to the general public for his mass–energy \
equivalence formula E = mc2, which has been dubbed 'the world's most famous equation'. He received the 1921 \
Nobel Prize in Physics 'for his services to theoretical physics, and especially for his discovery of the law \
of the photoelectric effect', a pivotal step in the development of quantum theory."
},
"doc2" : {
"title": "", # Keep title an empty string if not present
"text": "Wheat beer is a top-fermented beer which is brewed with a large proportion of wheat relative to the amount of \
malted barley. The two main varieties are German Weißbier and Belgian witbier; other types include Lambic (made\
with wild yeast), Berliner Weisse (a cloudy, sour beer), and Gose (a sour, salty beer)."
},
}
queries = {
"q1" : "Who developed the mass-energy equivalence formula?",
"q2" : "Which beer is brewed with a large proportion of wheat?"
}
qrels = {
"q1" : {"doc1": 1},
"q2" : {"doc2": 1},
}
```
### Data Fields
Examples from all configurations have the following features:
### Corpus
- `corpus`: a `dict` feature representing the document title and passage text, made up of:
- `_id`: a `string` feature representing the unique document id
- `title`: a `string` feature, denoting the title of the document.
- `text`: a `string` feature, denoting the text of the document.
### Queries
- `queries`: a `dict` feature representing the query, made up of:
- `_id`: a `string` feature representing the unique query id
- `text`: a `string` feature, denoting the text of the query.
### Qrels
- `qrels`: a `dict` feature representing the query document relevance judgements, made up of:
- `_id`: a `string` feature representing the query id
- `_id`: a `string` feature, denoting the document id.
- `score`: a `int32` feature, denoting the relevance judgement between query and document.
### Data Splits
| Dataset | Website| BEIR-Name | Type | Queries | Corpus | Rel D/Q | Down-load | md5 |
| -------- | -----| ---------| --------- | ----------- | ---------| ---------| :----------: | :------:|
| MSMARCO | [Homepage](https://microsoft.github.io/msmarco/)| ``msmarco`` | ``train``<br>``dev``<br>``test``| 6,980 | 8.84M | 1.1 | [Link](https://public.ukp.informatik.tu-darmstadt.de/thakur/BEIR/datasets/msmarco.zip) | ``444067daf65d982533ea17ebd59501e4`` |
| TREC-COVID | [Homepage](https://ir.nist.gov/covidSubmit/index.html)| ``trec-covid``| ``test``| 50| 171K| 493.5 | [Link](https://public.ukp.informatik.tu-darmstadt.de/thakur/BEIR/datasets/trec-covid.zip) | ``ce62140cb23feb9becf6270d0d1fe6d1`` |
| NFCorpus | [Homepage](https://www.cl.uni-heidelberg.de/statnlpgroup/nfcorpus/) | ``nfcorpus`` | ``train``<br>``dev``<br>``test``| 323 | 3.6K | 38.2 | [Link](https://public.ukp.informatik.tu-darmstadt.de/thakur/BEIR/datasets/nfcorpus.zip) | ``a89dba18a62ef92f7d323ec890a0d38d`` |
| BioASQ | [Homepage](http://bioasq.org) | ``bioasq``| ``train``<br>``test`` | 500 | 14.91M | 8.05 | No | [How to Reproduce?](https://github.com/UKPLab/beir/blob/main/examples/dataset#2-bioasq) |
| NQ | [Homepage](https://ai.google.com/research/NaturalQuestions) | ``nq``| ``train``<br>``test``| 3,452 | 2.68M | 1.2 | [Link](https://public.ukp.informatik.tu-darmstadt.de/thakur/BEIR/datasets/nq.zip) | ``d4d3d2e48787a744b6f6e691ff534307`` |
| HotpotQA | [Homepage](https://hotpotqa.github.io) | ``hotpotqa``| ``train``<br>``dev``<br>``test``| 7,405 | 5.23M | 2.0 | [Link](https://public.ukp.informatik.tu-darmstadt.de/thakur/BEIR/datasets/hotpotqa.zip) | ``f412724f78b0d91183a0e86805e16114`` |
| FiQA-2018 | [Homepage](https://sites.google.com/view/fiqa/) | ``fiqa`` | ``train``<br>``dev``<br>``test``| 648 | 57K | 2.6 | [Link](https://public.ukp.informatik.tu-darmstadt.de/thakur/BEIR/datasets/fiqa.zip) | ``17918ed23cd04fb15047f73e6c3bd9d9`` |
| Signal-1M(RT) | [Homepage](https://research.signal-ai.com/datasets/signal1m-tweetir.html)| ``signal1m`` | ``test``| 97 | 2.86M | 19.6 | No | [How to Reproduce?](https://github.com/UKPLab/beir/blob/main/examples/dataset#4-signal-1m) |
| TREC-NEWS | [Homepage](https://trec.nist.gov/data/news2019.html) | ``trec-news`` | ``test``| 57 | 595K | 19.6 | No | [How to Reproduce?](https://github.com/UKPLab/beir/blob/main/examples/dataset#1-trec-news) |
| ArguAna | [Homepage](http://argumentation.bplaced.net/arguana/data) | ``arguana``| ``test`` | 1,406 | 8.67K | 1.0 | [Link](https://public.ukp.informatik.tu-darmstadt.de/thakur/BEIR/datasets/arguana.zip) | ``8ad3e3c2a5867cdced806d6503f29b99`` |
| Touche-2020| [Homepage](https://webis.de/events/touche-20/shared-task-1.html) | ``webis-touche2020``| ``test``| 49 | 382K | 19.0 | [Link](https://public.ukp.informatik.tu-darmstadt.de/thakur/BEIR/datasets/webis-touche2020.zip) | ``46f650ba5a527fc69e0a6521c5a23563`` |
| CQADupstack| [Homepage](http://nlp.cis.unimelb.edu.au/resources/cqadupstack/) | ``cqadupstack``| ``test``| 13,145 | 457K | 1.4 | [Link](https://public.ukp.informatik.tu-darmstadt.de/thakur/BEIR/datasets/cqadupstack.zip) | ``4e41456d7df8ee7760a7f866133bda78`` |
| Quora| [Homepage](https://www.quora.com/q/quoradata/First-Quora-Dataset-Release-Question-Pairs) | ``quora``| ``dev``<br>``test``| 10,000 | 523K | 1.6 | [Link](https://public.ukp.informatik.tu-darmstadt.de/thakur/BEIR/datasets/quora.zip) | ``18fb154900ba42a600f84b839c173167`` |
| DBPedia | [Homepage](https://github.com/iai-group/DBpedia-Entity/) | ``dbpedia-entity``| ``dev``<br>``test``| 400 | 4.63M | 38.2 | [Link](https://public.ukp.informatik.tu-darmstadt.de/thakur/BEIR/datasets/dbpedia-entity.zip) | ``c2a39eb420a3164af735795df012ac2c`` |
| SCIDOCS| [Homepage](https://allenai.org/data/scidocs) | ``scidocs``| ``test``| 1,000 | 25K | 4.9 | [Link](https://public.ukp.informatik.tu-darmstadt.de/thakur/BEIR/datasets/scidocs.zip) | ``38121350fc3a4d2f48850f6aff52e4a9`` |
| FEVER | [Homepage](http://fever.ai) | ``fever``| ``train``<br>``dev``<br>``test``| 6,666 | 5.42M | 1.2| [Link](https://public.ukp.informatik.tu-darmstadt.de/thakur/BEIR/datasets/fever.zip) | ``5a818580227bfb4b35bb6fa46d9b6c03`` |
| Climate-FEVER| [Homepage](http://climatefever.ai) | ``climate-fever``|``test``| 1,535 | 5.42M | 3.0 | [Link](https://public.ukp.informatik.tu-darmstadt.de/thakur/BEIR/datasets/climate-fever.zip) | ``8b66f0a9126c521bae2bde127b4dc99d`` |
| SciFact| [Homepage](https://github.com/allenai/scifact) | ``scifact``| ``train``<br>``test``| 300 | 5K | 1.1 | [Link](https://public.ukp.informatik.tu-darmstadt.de/thakur/BEIR/datasets/scifact.zip) | ``5f7d1de60b170fc8027bb7898e2efca1`` |
| Robust04 | [Homepage](https://trec.nist.gov/data/robust/04.guidelines.html) | ``robust04``| ``test``| 249 | 528K | 69.9 | No | [How to Reproduce?](https://github.com/UKPLab/beir/blob/main/examples/dataset#3-robust04) |
## Dataset Creation
### Curation Rationale
[Needs More Information]
### Source Data
#### Initial Data Collection and Normalization
[Needs More Information]
#### Who are the source language producers?
[Needs More Information]
### Annotations
#### Annotation process
[Needs More Information]
#### Who are the annotators?
[Needs More Information]
### Personal and Sensitive Information
[Needs More Information]
## Considerations for Using the Data
### Social Impact of Dataset
[Needs More Information]
### Discussion of Biases
[Needs More Information]
### Other Known Limitations
[Needs More Information]
## Additional Information
### Dataset Curators
[Needs More Information]
### Licensing Information
[Needs More Information]
### Citation Information
Cite as:
```
@inproceedings{
thakur2021beir,
title={{BEIR}: A Heterogeneous Benchmark for Zero-shot Evaluation of Information Retrieval Models},
author={Nandan Thakur and Nils Reimers and Andreas R{\"u}ckl{\'e} and Abhishek Srivastava and Iryna Gurevych},
booktitle={Thirty-fifth Conference on Neural Information Processing Systems Datasets and Benchmarks Track (Round 2)},
year={2021},
url={https://openreview.net/forum?id=wCu6T5xFjeJ}
}
```
### Contributions
Thanks to [@Nthakur20](https://github.com/Nthakur20) for adding this dataset. |
false |
# Dataset Card for BEIR Benchmark
## Table of Contents
- [Dataset Description](#dataset-description)
- [Dataset Summary](#dataset-summary)
- [Supported Tasks and Leaderboards](#supported-tasks-and-leaderboards)
- [Languages](#languages)
- [Dataset Structure](#dataset-structure)
- [Data Instances](#data-instances)
- [Data Fields](#data-fields)
- [Data Splits](#data-splits)
- [Dataset Creation](#dataset-creation)
- [Curation Rationale](#curation-rationale)
- [Source Data](#source-data)
- [Annotations](#annotations)
- [Personal and Sensitive Information](#personal-and-sensitive-information)
- [Considerations for Using the Data](#considerations-for-using-the-data)
- [Social Impact of Dataset](#social-impact-of-dataset)
- [Discussion of Biases](#discussion-of-biases)
- [Other Known Limitations](#other-known-limitations)
- [Additional Information](#additional-information)
- [Dataset Curators](#dataset-curators)
- [Licensing Information](#licensing-information)
- [Citation Information](#citation-information)
- [Contributions](#contributions)
## Dataset Description
- **Homepage:** https://github.com/UKPLab/beir
- **Repository:** https://github.com/UKPLab/beir
- **Paper:** https://openreview.net/forum?id=wCu6T5xFjeJ
- **Leaderboard:** https://docs.google.com/spreadsheets/d/1L8aACyPaXrL8iEelJLGqlMqXKPX2oSP_R10pZoy77Ns
- **Point of Contact:** nandan.thakur@uwaterloo.ca
### Dataset Summary
BEIR is a heterogeneous benchmark that has been built from 18 diverse datasets representing 9 information retrieval tasks:
- Fact-checking: [FEVER](http://fever.ai), [Climate-FEVER](http://climatefever.ai), [SciFact](https://github.com/allenai/scifact)
- Question-Answering: [NQ](https://ai.google.com/research/NaturalQuestions), [HotpotQA](https://hotpotqa.github.io), [FiQA-2018](https://sites.google.com/view/fiqa/)
- Bio-Medical IR: [TREC-COVID](https://ir.nist.gov/covidSubmit/index.html), [BioASQ](http://bioasq.org), [NFCorpus](https://www.cl.uni-heidelberg.de/statnlpgroup/nfcorpus/)
- News Retrieval: [TREC-NEWS](https://trec.nist.gov/data/news2019.html), [Robust04](https://trec.nist.gov/data/robust/04.guidelines.html)
- Argument Retrieval: [Touche-2020](https://webis.de/events/touche-20/shared-task-1.html), [ArguAna](tp://argumentation.bplaced.net/arguana/data)
- Duplicate Question Retrieval: [Quora](https://www.quora.com/q/quoradata/First-Quora-Dataset-Release-Question-Pairs), [CqaDupstack](http://nlp.cis.unimelb.edu.au/resources/cqadupstack/)
- Citation-Prediction: [SCIDOCS](https://allenai.org/data/scidocs)
- Tweet Retrieval: [Signal-1M](https://research.signal-ai.com/datasets/signal1m-tweetir.html)
- Entity Retrieval: [DBPedia](https://github.com/iai-group/DBpedia-Entity/)
All these datasets have been preprocessed and can be used for your experiments.
```python
```
### Supported Tasks and Leaderboards
The dataset supports a leaderboard that evaluates models against task-specific metrics such as F1 or EM, as well as their ability to retrieve supporting information from Wikipedia.
The current best performing models can be found [here](https://eval.ai/web/challenges/challenge-page/689/leaderboard/).
### Languages
All tasks are in English (`en`).
## Dataset Structure
All BEIR datasets must contain a corpus, queries and qrels (relevance judgments file). They must be in the following format:
- `corpus` file: a `.jsonl` file (jsonlines) that contains a list of dictionaries, each with three fields `_id` with unique document identifier, `title` with document title (optional) and `text` with document paragraph or passage. For example: `{"_id": "doc1", "title": "Albert Einstein", "text": "Albert Einstein was a German-born...."}`
- `queries` file: a `.jsonl` file (jsonlines) that contains a list of dictionaries, each with two fields `_id` with unique query identifier and `text` with query text. For example: `{"_id": "q1", "text": "Who developed the mass-energy equivalence formula?"}`
- `qrels` file: a `.tsv` file (tab-seperated) that contains three columns, i.e. the `query-id`, `corpus-id` and `score` in this order. Keep 1st row as header. For example: `q1 doc1 1`
### Data Instances
A high level example of any beir dataset:
```python
corpus = {
"doc1" : {
"title": "Albert Einstein",
"text": "Albert Einstein was a German-born theoretical physicist. who developed the theory of relativity, \
one of the two pillars of modern physics (alongside quantum mechanics). His work is also known for \
its influence on the philosophy of science. He is best known to the general public for his mass–energy \
equivalence formula E = mc2, which has been dubbed 'the world's most famous equation'. He received the 1921 \
Nobel Prize in Physics 'for his services to theoretical physics, and especially for his discovery of the law \
of the photoelectric effect', a pivotal step in the development of quantum theory."
},
"doc2" : {
"title": "", # Keep title an empty string if not present
"text": "Wheat beer is a top-fermented beer which is brewed with a large proportion of wheat relative to the amount of \
malted barley. The two main varieties are German Weißbier and Belgian witbier; other types include Lambic (made\
with wild yeast), Berliner Weisse (a cloudy, sour beer), and Gose (a sour, salty beer)."
},
}
queries = {
"q1" : "Who developed the mass-energy equivalence formula?",
"q2" : "Which beer is brewed with a large proportion of wheat?"
}
qrels = {
"q1" : {"doc1": 1},
"q2" : {"doc2": 1},
}
```
### Data Fields
Examples from all configurations have the following features:
### Corpus
- `corpus`: a `dict` feature representing the document title and passage text, made up of:
- `_id`: a `string` feature representing the unique document id
- `title`: a `string` feature, denoting the title of the document.
- `text`: a `string` feature, denoting the text of the document.
### Queries
- `queries`: a `dict` feature representing the query, made up of:
- `_id`: a `string` feature representing the unique query id
- `text`: a `string` feature, denoting the text of the query.
### Qrels
- `qrels`: a `dict` feature representing the query document relevance judgements, made up of:
- `_id`: a `string` feature representing the query id
- `_id`: a `string` feature, denoting the document id.
- `score`: a `int32` feature, denoting the relevance judgement between query and document.
### Data Splits
| Dataset | Website| BEIR-Name | Type | Queries | Corpus | Rel D/Q | Down-load | md5 |
| -------- | -----| ---------| --------- | ----------- | ---------| ---------| :----------: | :------:|
| MSMARCO | [Homepage](https://microsoft.github.io/msmarco/)| ``msmarco`` | ``train``<br>``dev``<br>``test``| 6,980 | 8.84M | 1.1 | [Link](https://public.ukp.informatik.tu-darmstadt.de/thakur/BEIR/datasets/msmarco.zip) | ``444067daf65d982533ea17ebd59501e4`` |
| TREC-COVID | [Homepage](https://ir.nist.gov/covidSubmit/index.html)| ``trec-covid``| ``test``| 50| 171K| 493.5 | [Link](https://public.ukp.informatik.tu-darmstadt.de/thakur/BEIR/datasets/trec-covid.zip) | ``ce62140cb23feb9becf6270d0d1fe6d1`` |
| NFCorpus | [Homepage](https://www.cl.uni-heidelberg.de/statnlpgroup/nfcorpus/) | ``nfcorpus`` | ``train``<br>``dev``<br>``test``| 323 | 3.6K | 38.2 | [Link](https://public.ukp.informatik.tu-darmstadt.de/thakur/BEIR/datasets/nfcorpus.zip) | ``a89dba18a62ef92f7d323ec890a0d38d`` |
| BioASQ | [Homepage](http://bioasq.org) | ``bioasq``| ``train``<br>``test`` | 500 | 14.91M | 8.05 | No | [How to Reproduce?](https://github.com/UKPLab/beir/blob/main/examples/dataset#2-bioasq) |
| NQ | [Homepage](https://ai.google.com/research/NaturalQuestions) | ``nq``| ``train``<br>``test``| 3,452 | 2.68M | 1.2 | [Link](https://public.ukp.informatik.tu-darmstadt.de/thakur/BEIR/datasets/nq.zip) | ``d4d3d2e48787a744b6f6e691ff534307`` |
| HotpotQA | [Homepage](https://hotpotqa.github.io) | ``hotpotqa``| ``train``<br>``dev``<br>``test``| 7,405 | 5.23M | 2.0 | [Link](https://public.ukp.informatik.tu-darmstadt.de/thakur/BEIR/datasets/hotpotqa.zip) | ``f412724f78b0d91183a0e86805e16114`` |
| FiQA-2018 | [Homepage](https://sites.google.com/view/fiqa/) | ``fiqa`` | ``train``<br>``dev``<br>``test``| 648 | 57K | 2.6 | [Link](https://public.ukp.informatik.tu-darmstadt.de/thakur/BEIR/datasets/fiqa.zip) | ``17918ed23cd04fb15047f73e6c3bd9d9`` |
| Signal-1M(RT) | [Homepage](https://research.signal-ai.com/datasets/signal1m-tweetir.html)| ``signal1m`` | ``test``| 97 | 2.86M | 19.6 | No | [How to Reproduce?](https://github.com/UKPLab/beir/blob/main/examples/dataset#4-signal-1m) |
| TREC-NEWS | [Homepage](https://trec.nist.gov/data/news2019.html) | ``trec-news`` | ``test``| 57 | 595K | 19.6 | No | [How to Reproduce?](https://github.com/UKPLab/beir/blob/main/examples/dataset#1-trec-news) |
| ArguAna | [Homepage](http://argumentation.bplaced.net/arguana/data) | ``arguana``| ``test`` | 1,406 | 8.67K | 1.0 | [Link](https://public.ukp.informatik.tu-darmstadt.de/thakur/BEIR/datasets/arguana.zip) | ``8ad3e3c2a5867cdced806d6503f29b99`` |
| Touche-2020| [Homepage](https://webis.de/events/touche-20/shared-task-1.html) | ``webis-touche2020``| ``test``| 49 | 382K | 19.0 | [Link](https://public.ukp.informatik.tu-darmstadt.de/thakur/BEIR/datasets/webis-touche2020.zip) | ``46f650ba5a527fc69e0a6521c5a23563`` |
| CQADupstack| [Homepage](http://nlp.cis.unimelb.edu.au/resources/cqadupstack/) | ``cqadupstack``| ``test``| 13,145 | 457K | 1.4 | [Link](https://public.ukp.informatik.tu-darmstadt.de/thakur/BEIR/datasets/cqadupstack.zip) | ``4e41456d7df8ee7760a7f866133bda78`` |
| Quora| [Homepage](https://www.quora.com/q/quoradata/First-Quora-Dataset-Release-Question-Pairs) | ``quora``| ``dev``<br>``test``| 10,000 | 523K | 1.6 | [Link](https://public.ukp.informatik.tu-darmstadt.de/thakur/BEIR/datasets/quora.zip) | ``18fb154900ba42a600f84b839c173167`` |
| DBPedia | [Homepage](https://github.com/iai-group/DBpedia-Entity/) | ``dbpedia-entity``| ``dev``<br>``test``| 400 | 4.63M | 38.2 | [Link](https://public.ukp.informatik.tu-darmstadt.de/thakur/BEIR/datasets/dbpedia-entity.zip) | ``c2a39eb420a3164af735795df012ac2c`` |
| SCIDOCS| [Homepage](https://allenai.org/data/scidocs) | ``scidocs``| ``test``| 1,000 | 25K | 4.9 | [Link](https://public.ukp.informatik.tu-darmstadt.de/thakur/BEIR/datasets/scidocs.zip) | ``38121350fc3a4d2f48850f6aff52e4a9`` |
| FEVER | [Homepage](http://fever.ai) | ``fever``| ``train``<br>``dev``<br>``test``| 6,666 | 5.42M | 1.2| [Link](https://public.ukp.informatik.tu-darmstadt.de/thakur/BEIR/datasets/fever.zip) | ``5a818580227bfb4b35bb6fa46d9b6c03`` |
| Climate-FEVER| [Homepage](http://climatefever.ai) | ``climate-fever``|``test``| 1,535 | 5.42M | 3.0 | [Link](https://public.ukp.informatik.tu-darmstadt.de/thakur/BEIR/datasets/climate-fever.zip) | ``8b66f0a9126c521bae2bde127b4dc99d`` |
| SciFact| [Homepage](https://github.com/allenai/scifact) | ``scifact``| ``train``<br>``test``| 300 | 5K | 1.1 | [Link](https://public.ukp.informatik.tu-darmstadt.de/thakur/BEIR/datasets/scifact.zip) | ``5f7d1de60b170fc8027bb7898e2efca1`` |
| Robust04 | [Homepage](https://trec.nist.gov/data/robust/04.guidelines.html) | ``robust04``| ``test``| 249 | 528K | 69.9 | No | [How to Reproduce?](https://github.com/UKPLab/beir/blob/main/examples/dataset#3-robust04) |
## Dataset Creation
### Curation Rationale
[Needs More Information]
### Source Data
#### Initial Data Collection and Normalization
[Needs More Information]
#### Who are the source language producers?
[Needs More Information]
### Annotations
#### Annotation process
[Needs More Information]
#### Who are the annotators?
[Needs More Information]
### Personal and Sensitive Information
[Needs More Information]
## Considerations for Using the Data
### Social Impact of Dataset
[Needs More Information]
### Discussion of Biases
[Needs More Information]
### Other Known Limitations
[Needs More Information]
## Additional Information
### Dataset Curators
[Needs More Information]
### Licensing Information
[Needs More Information]
### Citation Information
Cite as:
```
@inproceedings{
thakur2021beir,
title={{BEIR}: A Heterogeneous Benchmark for Zero-shot Evaluation of Information Retrieval Models},
author={Nandan Thakur and Nils Reimers and Andreas R{\"u}ckl{\'e} and Abhishek Srivastava and Iryna Gurevych},
booktitle={Thirty-fifth Conference on Neural Information Processing Systems Datasets and Benchmarks Track (Round 2)},
year={2021},
url={https://openreview.net/forum?id=wCu6T5xFjeJ}
}
```
### Contributions
Thanks to [@Nthakur20](https://github.com/Nthakur20) for adding this dataset. |
false |
# Dataset Card for BEIR Benchmark
## Table of Contents
- [Dataset Description](#dataset-description)
- [Dataset Summary](#dataset-summary)
- [Supported Tasks and Leaderboards](#supported-tasks-and-leaderboards)
- [Languages](#languages)
- [Dataset Structure](#dataset-structure)
- [Data Instances](#data-instances)
- [Data Fields](#data-fields)
- [Data Splits](#data-splits)
- [Dataset Creation](#dataset-creation)
- [Curation Rationale](#curation-rationale)
- [Source Data](#source-data)
- [Annotations](#annotations)
- [Personal and Sensitive Information](#personal-and-sensitive-information)
- [Considerations for Using the Data](#considerations-for-using-the-data)
- [Social Impact of Dataset](#social-impact-of-dataset)
- [Discussion of Biases](#discussion-of-biases)
- [Other Known Limitations](#other-known-limitations)
- [Additional Information](#additional-information)
- [Dataset Curators](#dataset-curators)
- [Licensing Information](#licensing-information)
- [Citation Information](#citation-information)
- [Contributions](#contributions)
## Dataset Description
- **Homepage:** https://github.com/UKPLab/beir
- **Repository:** https://github.com/UKPLab/beir
- **Paper:** https://openreview.net/forum?id=wCu6T5xFjeJ
- **Leaderboard:** https://docs.google.com/spreadsheets/d/1L8aACyPaXrL8iEelJLGqlMqXKPX2oSP_R10pZoy77Ns
- **Point of Contact:** nandan.thakur@uwaterloo.ca
### Dataset Summary
BEIR is a heterogeneous benchmark that has been built from 18 diverse datasets representing 9 information retrieval tasks:
- Fact-checking: [FEVER](http://fever.ai), [Climate-FEVER](http://climatefever.ai), [SciFact](https://github.com/allenai/scifact)
- Question-Answering: [NQ](https://ai.google.com/research/NaturalQuestions), [HotpotQA](https://hotpotqa.github.io), [FiQA-2018](https://sites.google.com/view/fiqa/)
- Bio-Medical IR: [TREC-COVID](https://ir.nist.gov/covidSubmit/index.html), [BioASQ](http://bioasq.org), [NFCorpus](https://www.cl.uni-heidelberg.de/statnlpgroup/nfcorpus/)
- News Retrieval: [TREC-NEWS](https://trec.nist.gov/data/news2019.html), [Robust04](https://trec.nist.gov/data/robust/04.guidelines.html)
- Argument Retrieval: [Touche-2020](https://webis.de/events/touche-20/shared-task-1.html), [ArguAna](tp://argumentation.bplaced.net/arguana/data)
- Duplicate Question Retrieval: [Quora](https://www.quora.com/q/quoradata/First-Quora-Dataset-Release-Question-Pairs), [CqaDupstack](http://nlp.cis.unimelb.edu.au/resources/cqadupstack/)
- Citation-Prediction: [SCIDOCS](https://allenai.org/data/scidocs)
- Tweet Retrieval: [Signal-1M](https://research.signal-ai.com/datasets/signal1m-tweetir.html)
- Entity Retrieval: [DBPedia](https://github.com/iai-group/DBpedia-Entity/)
All these datasets have been preprocessed and can be used for your experiments.
```python
```
### Supported Tasks and Leaderboards
The dataset supports a leaderboard that evaluates models against task-specific metrics such as F1 or EM, as well as their ability to retrieve supporting information from Wikipedia.
The current best performing models can be found [here](https://eval.ai/web/challenges/challenge-page/689/leaderboard/).
### Languages
All tasks are in English (`en`).
## Dataset Structure
All BEIR datasets must contain a corpus, queries and qrels (relevance judgments file). They must be in the following format:
- `corpus` file: a `.jsonl` file (jsonlines) that contains a list of dictionaries, each with three fields `_id` with unique document identifier, `title` with document title (optional) and `text` with document paragraph or passage. For example: `{"_id": "doc1", "title": "Albert Einstein", "text": "Albert Einstein was a German-born...."}`
- `queries` file: a `.jsonl` file (jsonlines) that contains a list of dictionaries, each with two fields `_id` with unique query identifier and `text` with query text. For example: `{"_id": "q1", "text": "Who developed the mass-energy equivalence formula?"}`
- `qrels` file: a `.tsv` file (tab-seperated) that contains three columns, i.e. the `query-id`, `corpus-id` and `score` in this order. Keep 1st row as header. For example: `q1 doc1 1`
### Data Instances
A high level example of any beir dataset:
```python
corpus = {
"doc1" : {
"title": "Albert Einstein",
"text": "Albert Einstein was a German-born theoretical physicist. who developed the theory of relativity, \
one of the two pillars of modern physics (alongside quantum mechanics). His work is also known for \
its influence on the philosophy of science. He is best known to the general public for his mass–energy \
equivalence formula E = mc2, which has been dubbed 'the world's most famous equation'. He received the 1921 \
Nobel Prize in Physics 'for his services to theoretical physics, and especially for his discovery of the law \
of the photoelectric effect', a pivotal step in the development of quantum theory."
},
"doc2" : {
"title": "", # Keep title an empty string if not present
"text": "Wheat beer is a top-fermented beer which is brewed with a large proportion of wheat relative to the amount of \
malted barley. The two main varieties are German Weißbier and Belgian witbier; other types include Lambic (made\
with wild yeast), Berliner Weisse (a cloudy, sour beer), and Gose (a sour, salty beer)."
},
}
queries = {
"q1" : "Who developed the mass-energy equivalence formula?",
"q2" : "Which beer is brewed with a large proportion of wheat?"
}
qrels = {
"q1" : {"doc1": 1},
"q2" : {"doc2": 1},
}
```
### Data Fields
Examples from all configurations have the following features:
### Corpus
- `corpus`: a `dict` feature representing the document title and passage text, made up of:
- `_id`: a `string` feature representing the unique document id
- `title`: a `string` feature, denoting the title of the document.
- `text`: a `string` feature, denoting the text of the document.
### Queries
- `queries`: a `dict` feature representing the query, made up of:
- `_id`: a `string` feature representing the unique query id
- `text`: a `string` feature, denoting the text of the query.
### Qrels
- `qrels`: a `dict` feature representing the query document relevance judgements, made up of:
- `_id`: a `string` feature representing the query id
- `_id`: a `string` feature, denoting the document id.
- `score`: a `int32` feature, denoting the relevance judgement between query and document.
### Data Splits
| Dataset | Website| BEIR-Name | Type | Queries | Corpus | Rel D/Q | Down-load | md5 |
| -------- | -----| ---------| --------- | ----------- | ---------| ---------| :----------: | :------:|
| MSMARCO | [Homepage](https://microsoft.github.io/msmarco/)| ``msmarco`` | ``train``<br>``dev``<br>``test``| 6,980 | 8.84M | 1.1 | [Link](https://public.ukp.informatik.tu-darmstadt.de/thakur/BEIR/datasets/msmarco.zip) | ``444067daf65d982533ea17ebd59501e4`` |
| TREC-COVID | [Homepage](https://ir.nist.gov/covidSubmit/index.html)| ``trec-covid``| ``test``| 50| 171K| 493.5 | [Link](https://public.ukp.informatik.tu-darmstadt.de/thakur/BEIR/datasets/trec-covid.zip) | ``ce62140cb23feb9becf6270d0d1fe6d1`` |
| NFCorpus | [Homepage](https://www.cl.uni-heidelberg.de/statnlpgroup/nfcorpus/) | ``nfcorpus`` | ``train``<br>``dev``<br>``test``| 323 | 3.6K | 38.2 | [Link](https://public.ukp.informatik.tu-darmstadt.de/thakur/BEIR/datasets/nfcorpus.zip) | ``a89dba18a62ef92f7d323ec890a0d38d`` |
| BioASQ | [Homepage](http://bioasq.org) | ``bioasq``| ``train``<br>``test`` | 500 | 14.91M | 8.05 | No | [How to Reproduce?](https://github.com/UKPLab/beir/blob/main/examples/dataset#2-bioasq) |
| NQ | [Homepage](https://ai.google.com/research/NaturalQuestions) | ``nq``| ``train``<br>``test``| 3,452 | 2.68M | 1.2 | [Link](https://public.ukp.informatik.tu-darmstadt.de/thakur/BEIR/datasets/nq.zip) | ``d4d3d2e48787a744b6f6e691ff534307`` |
| HotpotQA | [Homepage](https://hotpotqa.github.io) | ``hotpotqa``| ``train``<br>``dev``<br>``test``| 7,405 | 5.23M | 2.0 | [Link](https://public.ukp.informatik.tu-darmstadt.de/thakur/BEIR/datasets/hotpotqa.zip) | ``f412724f78b0d91183a0e86805e16114`` |
| FiQA-2018 | [Homepage](https://sites.google.com/view/fiqa/) | ``fiqa`` | ``train``<br>``dev``<br>``test``| 648 | 57K | 2.6 | [Link](https://public.ukp.informatik.tu-darmstadt.de/thakur/BEIR/datasets/fiqa.zip) | ``17918ed23cd04fb15047f73e6c3bd9d9`` |
| Signal-1M(RT) | [Homepage](https://research.signal-ai.com/datasets/signal1m-tweetir.html)| ``signal1m`` | ``test``| 97 | 2.86M | 19.6 | No | [How to Reproduce?](https://github.com/UKPLab/beir/blob/main/examples/dataset#4-signal-1m) |
| TREC-NEWS | [Homepage](https://trec.nist.gov/data/news2019.html) | ``trec-news`` | ``test``| 57 | 595K | 19.6 | No | [How to Reproduce?](https://github.com/UKPLab/beir/blob/main/examples/dataset#1-trec-news) |
| ArguAna | [Homepage](http://argumentation.bplaced.net/arguana/data) | ``arguana``| ``test`` | 1,406 | 8.67K | 1.0 | [Link](https://public.ukp.informatik.tu-darmstadt.de/thakur/BEIR/datasets/arguana.zip) | ``8ad3e3c2a5867cdced806d6503f29b99`` |
| Touche-2020| [Homepage](https://webis.de/events/touche-20/shared-task-1.html) | ``webis-touche2020``| ``test``| 49 | 382K | 19.0 | [Link](https://public.ukp.informatik.tu-darmstadt.de/thakur/BEIR/datasets/webis-touche2020.zip) | ``46f650ba5a527fc69e0a6521c5a23563`` |
| CQADupstack| [Homepage](http://nlp.cis.unimelb.edu.au/resources/cqadupstack/) | ``cqadupstack``| ``test``| 13,145 | 457K | 1.4 | [Link](https://public.ukp.informatik.tu-darmstadt.de/thakur/BEIR/datasets/cqadupstack.zip) | ``4e41456d7df8ee7760a7f866133bda78`` |
| Quora| [Homepage](https://www.quora.com/q/quoradata/First-Quora-Dataset-Release-Question-Pairs) | ``quora``| ``dev``<br>``test``| 10,000 | 523K | 1.6 | [Link](https://public.ukp.informatik.tu-darmstadt.de/thakur/BEIR/datasets/quora.zip) | ``18fb154900ba42a600f84b839c173167`` |
| DBPedia | [Homepage](https://github.com/iai-group/DBpedia-Entity/) | ``dbpedia-entity``| ``dev``<br>``test``| 400 | 4.63M | 38.2 | [Link](https://public.ukp.informatik.tu-darmstadt.de/thakur/BEIR/datasets/dbpedia-entity.zip) | ``c2a39eb420a3164af735795df012ac2c`` |
| SCIDOCS| [Homepage](https://allenai.org/data/scidocs) | ``scidocs``| ``test``| 1,000 | 25K | 4.9 | [Link](https://public.ukp.informatik.tu-darmstadt.de/thakur/BEIR/datasets/scidocs.zip) | ``38121350fc3a4d2f48850f6aff52e4a9`` |
| FEVER | [Homepage](http://fever.ai) | ``fever``| ``train``<br>``dev``<br>``test``| 6,666 | 5.42M | 1.2| [Link](https://public.ukp.informatik.tu-darmstadt.de/thakur/BEIR/datasets/fever.zip) | ``5a818580227bfb4b35bb6fa46d9b6c03`` |
| Climate-FEVER| [Homepage](http://climatefever.ai) | ``climate-fever``|``test``| 1,535 | 5.42M | 3.0 | [Link](https://public.ukp.informatik.tu-darmstadt.de/thakur/BEIR/datasets/climate-fever.zip) | ``8b66f0a9126c521bae2bde127b4dc99d`` |
| SciFact| [Homepage](https://github.com/allenai/scifact) | ``scifact``| ``train``<br>``test``| 300 | 5K | 1.1 | [Link](https://public.ukp.informatik.tu-darmstadt.de/thakur/BEIR/datasets/scifact.zip) | ``5f7d1de60b170fc8027bb7898e2efca1`` |
| Robust04 | [Homepage](https://trec.nist.gov/data/robust/04.guidelines.html) | ``robust04``| ``test``| 249 | 528K | 69.9 | No | [How to Reproduce?](https://github.com/UKPLab/beir/blob/main/examples/dataset#3-robust04) |
## Dataset Creation
### Curation Rationale
[Needs More Information]
### Source Data
#### Initial Data Collection and Normalization
[Needs More Information]
#### Who are the source language producers?
[Needs More Information]
### Annotations
#### Annotation process
[Needs More Information]
#### Who are the annotators?
[Needs More Information]
### Personal and Sensitive Information
[Needs More Information]
## Considerations for Using the Data
### Social Impact of Dataset
[Needs More Information]
### Discussion of Biases
[Needs More Information]
### Other Known Limitations
[Needs More Information]
## Additional Information
### Dataset Curators
[Needs More Information]
### Licensing Information
[Needs More Information]
### Citation Information
Cite as:
```
@inproceedings{
thakur2021beir,
title={{BEIR}: A Heterogeneous Benchmark for Zero-shot Evaluation of Information Retrieval Models},
author={Nandan Thakur and Nils Reimers and Andreas R{\"u}ckl{\'e} and Abhishek Srivastava and Iryna Gurevych},
booktitle={Thirty-fifth Conference on Neural Information Processing Systems Datasets and Benchmarks Track (Round 2)},
year={2021},
url={https://openreview.net/forum?id=wCu6T5xFjeJ}
}
```
### Contributions
Thanks to [@Nthakur20](https://github.com/Nthakur20) for adding this dataset. |
false |
# Dataset Card for BEIR Benchmark
## Table of Contents
- [Dataset Description](#dataset-description)
- [Dataset Summary](#dataset-summary)
- [Supported Tasks and Leaderboards](#supported-tasks-and-leaderboards)
- [Languages](#languages)
- [Dataset Structure](#dataset-structure)
- [Data Instances](#data-instances)
- [Data Fields](#data-fields)
- [Data Splits](#data-splits)
- [Dataset Creation](#dataset-creation)
- [Curation Rationale](#curation-rationale)
- [Source Data](#source-data)
- [Annotations](#annotations)
- [Personal and Sensitive Information](#personal-and-sensitive-information)
- [Considerations for Using the Data](#considerations-for-using-the-data)
- [Social Impact of Dataset](#social-impact-of-dataset)
- [Discussion of Biases](#discussion-of-biases)
- [Other Known Limitations](#other-known-limitations)
- [Additional Information](#additional-information)
- [Dataset Curators](#dataset-curators)
- [Licensing Information](#licensing-information)
- [Citation Information](#citation-information)
- [Contributions](#contributions)
## Dataset Description
- **Homepage:** https://github.com/UKPLab/beir
- **Repository:** https://github.com/UKPLab/beir
- **Paper:** https://openreview.net/forum?id=wCu6T5xFjeJ
- **Leaderboard:** https://docs.google.com/spreadsheets/d/1L8aACyPaXrL8iEelJLGqlMqXKPX2oSP_R10pZoy77Ns
- **Point of Contact:** nandan.thakur@uwaterloo.ca
### Dataset Summary
BEIR is a heterogeneous benchmark that has been built from 18 diverse datasets representing 9 information retrieval tasks:
- Fact-checking: [FEVER](http://fever.ai), [Climate-FEVER](http://climatefever.ai), [SciFact](https://github.com/allenai/scifact)
- Question-Answering: [NQ](https://ai.google.com/research/NaturalQuestions), [HotpotQA](https://hotpotqa.github.io), [FiQA-2018](https://sites.google.com/view/fiqa/)
- Bio-Medical IR: [TREC-COVID](https://ir.nist.gov/covidSubmit/index.html), [BioASQ](http://bioasq.org), [NFCorpus](https://www.cl.uni-heidelberg.de/statnlpgroup/nfcorpus/)
- News Retrieval: [TREC-NEWS](https://trec.nist.gov/data/news2019.html), [Robust04](https://trec.nist.gov/data/robust/04.guidelines.html)
- Argument Retrieval: [Touche-2020](https://webis.de/events/touche-20/shared-task-1.html), [ArguAna](tp://argumentation.bplaced.net/arguana/data)
- Duplicate Question Retrieval: [Quora](https://www.quora.com/q/quoradata/First-Quora-Dataset-Release-Question-Pairs), [CqaDupstack](http://nlp.cis.unimelb.edu.au/resources/cqadupstack/)
- Citation-Prediction: [SCIDOCS](https://allenai.org/data/scidocs)
- Tweet Retrieval: [Signal-1M](https://research.signal-ai.com/datasets/signal1m-tweetir.html)
- Entity Retrieval: [DBPedia](https://github.com/iai-group/DBpedia-Entity/)
All these datasets have been preprocessed and can be used for your experiments.
```python
```
### Supported Tasks and Leaderboards
The dataset supports a leaderboard that evaluates models against task-specific metrics such as F1 or EM, as well as their ability to retrieve supporting information from Wikipedia.
The current best performing models can be found [here](https://eval.ai/web/challenges/challenge-page/689/leaderboard/).
### Languages
All tasks are in English (`en`).
## Dataset Structure
All BEIR datasets must contain a corpus, queries and qrels (relevance judgments file). They must be in the following format:
- `corpus` file: a `.jsonl` file (jsonlines) that contains a list of dictionaries, each with three fields `_id` with unique document identifier, `title` with document title (optional) and `text` with document paragraph or passage. For example: `{"_id": "doc1", "title": "Albert Einstein", "text": "Albert Einstein was a German-born...."}`
- `queries` file: a `.jsonl` file (jsonlines) that contains a list of dictionaries, each with two fields `_id` with unique query identifier and `text` with query text. For example: `{"_id": "q1", "text": "Who developed the mass-energy equivalence formula?"}`
- `qrels` file: a `.tsv` file (tab-seperated) that contains three columns, i.e. the `query-id`, `corpus-id` and `score` in this order. Keep 1st row as header. For example: `q1 doc1 1`
### Data Instances
A high level example of any beir dataset:
```python
corpus = {
"doc1" : {
"title": "Albert Einstein",
"text": "Albert Einstein was a German-born theoretical physicist. who developed the theory of relativity, \
one of the two pillars of modern physics (alongside quantum mechanics). His work is also known for \
its influence on the philosophy of science. He is best known to the general public for his mass–energy \
equivalence formula E = mc2, which has been dubbed 'the world's most famous equation'. He received the 1921 \
Nobel Prize in Physics 'for his services to theoretical physics, and especially for his discovery of the law \
of the photoelectric effect', a pivotal step in the development of quantum theory."
},
"doc2" : {
"title": "", # Keep title an empty string if not present
"text": "Wheat beer is a top-fermented beer which is brewed with a large proportion of wheat relative to the amount of \
malted barley. The two main varieties are German Weißbier and Belgian witbier; other types include Lambic (made\
with wild yeast), Berliner Weisse (a cloudy, sour beer), and Gose (a sour, salty beer)."
},
}
queries = {
"q1" : "Who developed the mass-energy equivalence formula?",
"q2" : "Which beer is brewed with a large proportion of wheat?"
}
qrels = {
"q1" : {"doc1": 1},
"q2" : {"doc2": 1},
}
```
### Data Fields
Examples from all configurations have the following features:
### Corpus
- `corpus`: a `dict` feature representing the document title and passage text, made up of:
- `_id`: a `string` feature representing the unique document id
- `title`: a `string` feature, denoting the title of the document.
- `text`: a `string` feature, denoting the text of the document.
### Queries
- `queries`: a `dict` feature representing the query, made up of:
- `_id`: a `string` feature representing the unique query id
- `text`: a `string` feature, denoting the text of the query.
### Qrels
- `qrels`: a `dict` feature representing the query document relevance judgements, made up of:
- `_id`: a `string` feature representing the query id
- `_id`: a `string` feature, denoting the document id.
- `score`: a `int32` feature, denoting the relevance judgement between query and document.
### Data Splits
| Dataset | Website| BEIR-Name | Type | Queries | Corpus | Rel D/Q | Down-load | md5 |
| -------- | -----| ---------| --------- | ----------- | ---------| ---------| :----------: | :------:|
| MSMARCO | [Homepage](https://microsoft.github.io/msmarco/)| ``msmarco`` | ``train``<br>``dev``<br>``test``| 6,980 | 8.84M | 1.1 | [Link](https://public.ukp.informatik.tu-darmstadt.de/thakur/BEIR/datasets/msmarco.zip) | ``444067daf65d982533ea17ebd59501e4`` |
| TREC-COVID | [Homepage](https://ir.nist.gov/covidSubmit/index.html)| ``trec-covid``| ``test``| 50| 171K| 493.5 | [Link](https://public.ukp.informatik.tu-darmstadt.de/thakur/BEIR/datasets/trec-covid.zip) | ``ce62140cb23feb9becf6270d0d1fe6d1`` |
| NFCorpus | [Homepage](https://www.cl.uni-heidelberg.de/statnlpgroup/nfcorpus/) | ``nfcorpus`` | ``train``<br>``dev``<br>``test``| 323 | 3.6K | 38.2 | [Link](https://public.ukp.informatik.tu-darmstadt.de/thakur/BEIR/datasets/nfcorpus.zip) | ``a89dba18a62ef92f7d323ec890a0d38d`` |
| BioASQ | [Homepage](http://bioasq.org) | ``bioasq``| ``train``<br>``test`` | 500 | 14.91M | 8.05 | No | [How to Reproduce?](https://github.com/UKPLab/beir/blob/main/examples/dataset#2-bioasq) |
| NQ | [Homepage](https://ai.google.com/research/NaturalQuestions) | ``nq``| ``train``<br>``test``| 3,452 | 2.68M | 1.2 | [Link](https://public.ukp.informatik.tu-darmstadt.de/thakur/BEIR/datasets/nq.zip) | ``d4d3d2e48787a744b6f6e691ff534307`` |
| HotpotQA | [Homepage](https://hotpotqa.github.io) | ``hotpotqa``| ``train``<br>``dev``<br>``test``| 7,405 | 5.23M | 2.0 | [Link](https://public.ukp.informatik.tu-darmstadt.de/thakur/BEIR/datasets/hotpotqa.zip) | ``f412724f78b0d91183a0e86805e16114`` |
| FiQA-2018 | [Homepage](https://sites.google.com/view/fiqa/) | ``fiqa`` | ``train``<br>``dev``<br>``test``| 648 | 57K | 2.6 | [Link](https://public.ukp.informatik.tu-darmstadt.de/thakur/BEIR/datasets/fiqa.zip) | ``17918ed23cd04fb15047f73e6c3bd9d9`` |
| Signal-1M(RT) | [Homepage](https://research.signal-ai.com/datasets/signal1m-tweetir.html)| ``signal1m`` | ``test``| 97 | 2.86M | 19.6 | No | [How to Reproduce?](https://github.com/UKPLab/beir/blob/main/examples/dataset#4-signal-1m) |
| TREC-NEWS | [Homepage](https://trec.nist.gov/data/news2019.html) | ``trec-news`` | ``test``| 57 | 595K | 19.6 | No | [How to Reproduce?](https://github.com/UKPLab/beir/blob/main/examples/dataset#1-trec-news) |
| ArguAna | [Homepage](http://argumentation.bplaced.net/arguana/data) | ``arguana``| ``test`` | 1,406 | 8.67K | 1.0 | [Link](https://public.ukp.informatik.tu-darmstadt.de/thakur/BEIR/datasets/arguana.zip) | ``8ad3e3c2a5867cdced806d6503f29b99`` |
| Touche-2020| [Homepage](https://webis.de/events/touche-20/shared-task-1.html) | ``webis-touche2020``| ``test``| 49 | 382K | 19.0 | [Link](https://public.ukp.informatik.tu-darmstadt.de/thakur/BEIR/datasets/webis-touche2020.zip) | ``46f650ba5a527fc69e0a6521c5a23563`` |
| CQADupstack| [Homepage](http://nlp.cis.unimelb.edu.au/resources/cqadupstack/) | ``cqadupstack``| ``test``| 13,145 | 457K | 1.4 | [Link](https://public.ukp.informatik.tu-darmstadt.de/thakur/BEIR/datasets/cqadupstack.zip) | ``4e41456d7df8ee7760a7f866133bda78`` |
| Quora| [Homepage](https://www.quora.com/q/quoradata/First-Quora-Dataset-Release-Question-Pairs) | ``quora``| ``dev``<br>``test``| 10,000 | 523K | 1.6 | [Link](https://public.ukp.informatik.tu-darmstadt.de/thakur/BEIR/datasets/quora.zip) | ``18fb154900ba42a600f84b839c173167`` |
| DBPedia | [Homepage](https://github.com/iai-group/DBpedia-Entity/) | ``dbpedia-entity``| ``dev``<br>``test``| 400 | 4.63M | 38.2 | [Link](https://public.ukp.informatik.tu-darmstadt.de/thakur/BEIR/datasets/dbpedia-entity.zip) | ``c2a39eb420a3164af735795df012ac2c`` |
| SCIDOCS| [Homepage](https://allenai.org/data/scidocs) | ``scidocs``| ``test``| 1,000 | 25K | 4.9 | [Link](https://public.ukp.informatik.tu-darmstadt.de/thakur/BEIR/datasets/scidocs.zip) | ``38121350fc3a4d2f48850f6aff52e4a9`` |
| FEVER | [Homepage](http://fever.ai) | ``fever``| ``train``<br>``dev``<br>``test``| 6,666 | 5.42M | 1.2| [Link](https://public.ukp.informatik.tu-darmstadt.de/thakur/BEIR/datasets/fever.zip) | ``5a818580227bfb4b35bb6fa46d9b6c03`` |
| Climate-FEVER| [Homepage](http://climatefever.ai) | ``climate-fever``|``test``| 1,535 | 5.42M | 3.0 | [Link](https://public.ukp.informatik.tu-darmstadt.de/thakur/BEIR/datasets/climate-fever.zip) | ``8b66f0a9126c521bae2bde127b4dc99d`` |
| SciFact| [Homepage](https://github.com/allenai/scifact) | ``scifact``| ``train``<br>``test``| 300 | 5K | 1.1 | [Link](https://public.ukp.informatik.tu-darmstadt.de/thakur/BEIR/datasets/scifact.zip) | ``5f7d1de60b170fc8027bb7898e2efca1`` |
| Robust04 | [Homepage](https://trec.nist.gov/data/robust/04.guidelines.html) | ``robust04``| ``test``| 249 | 528K | 69.9 | No | [How to Reproduce?](https://github.com/UKPLab/beir/blob/main/examples/dataset#3-robust04) |
## Dataset Creation
### Curation Rationale
[Needs More Information]
### Source Data
#### Initial Data Collection and Normalization
[Needs More Information]
#### Who are the source language producers?
[Needs More Information]
### Annotations
#### Annotation process
[Needs More Information]
#### Who are the annotators?
[Needs More Information]
### Personal and Sensitive Information
[Needs More Information]
## Considerations for Using the Data
### Social Impact of Dataset
[Needs More Information]
### Discussion of Biases
[Needs More Information]
### Other Known Limitations
[Needs More Information]
## Additional Information
### Dataset Curators
[Needs More Information]
### Licensing Information
[Needs More Information]
### Citation Information
Cite as:
```
@inproceedings{
thakur2021beir,
title={{BEIR}: A Heterogeneous Benchmark for Zero-shot Evaluation of Information Retrieval Models},
author={Nandan Thakur and Nils Reimers and Andreas R{\"u}ckl{\'e} and Abhishek Srivastava and Iryna Gurevych},
booktitle={Thirty-fifth Conference on Neural Information Processing Systems Datasets and Benchmarks Track (Round 2)},
year={2021},
url={https://openreview.net/forum?id=wCu6T5xFjeJ}
}
```
### Contributions
Thanks to [@Nthakur20](https://github.com/Nthakur20) for adding this dataset. |
true |
# Dataset Card for Annotated dataset to assess the accuracy of the textual description of cultural heritage records
## Table of Contents
- [Table of Contents](#table-of-contents)
- [Dataset Description](#dataset-description)
- [Dataset Summary](#dataset-summary)
- [Supported Tasks and Leaderboards](#supported-tasks-and-leaderboards)
- [Languages](#languages)
- [Dataset Structure](#dataset-structure)
- [Data Instances](#data-instances)
- [Data Fields](#data-fields)
- [Data Splits](#data-splits)
- [Dataset Creation](#dataset-creation)
- [Curation Rationale](#curation-rationale)
- [Source Data](#source-data)
- [Annotations](#annotations)
- [Personal and Sensitive Information](#personal-and-sensitive-information)
- [Considerations for Using the Data](#considerations-for-using-the-data)
- [Social Impact of Dataset](#social-impact-of-dataset)
- [Discussion of Biases](#discussion-of-biases)
- [Other Known Limitations](#other-known-limitations)
- [Additional Information](#additional-information)
- [Dataset Curators](#dataset-curators)
- [Licensing Information](#licensing-information)
- [Citation Information](#citation-information)
- [Contributions](#contributions)
## Dataset Description
- **Homepage:**[https://doi.org/10.6084/m9.figshare.13359104.v1](https://doi.org/10.6084/m9.figshare.13359104.v1)
- **Repository:**[https://doi.org/10.6084/m9.figshare.13359104.v1](https://doi.org/10.6084/m9.figshare.13359104.v1)
- **Paper:**[https://doi.org/10.1007/s00799-021-00302-1](https://doi.org/10.1007/s00799-021-00302-1)
- **Leaderboard:**
- **Point of Contact:**
### Dataset Summary
The dataset contains more than 100K textual descriptions of cultural items from [Cultura Italia](http://www.culturaitalia.it/opencms/index.jsp?language=en), the Italian National Cultural aggregator. Each of the description is labeled either HIGH or LOW quality, according its adherence to the standard cataloguing guidelines provided by Istituto Centrale per il Catalogo e la Documentazione (ICCD). More precisely, each description is labeled as HIGH quality if the object and subject of the item (for which the description is provided) are both described according to the ICCD guidelines, and as LOW quality in all other cases. Most of the dataset was manually annotated, with ~30K descriptions automatically labeled as LOW quality due to their length (less than 3 tokens) or their provenance from old (pre-2012), not curated, collections. The dataset was developed to support the training and testing of ML text classification approaches for automatically assessing the quality of textual descriptions in digital Cultural Heritage repositories.
### Supported Tasks and Leaderboards
This dataset can be used for text classification tasks. The [paper](https://doi.org/10.1007/s00799-021-00302-1) introducing the dataset achieved an f1 score of `.783` for the task of classifying if a metadata record was low or high quality. Please see the [results table](https://link.springer.com/article/10.1007/s00799-021-00302-1/tables/4) for a full overview of the results reported in the paper.
### Languages
The dataset consists of Italian metadata records. The labels are in English.
## Dataset Structure
The dataset has only one configuration.
### Data Instances
An example instance from the dataset:
``` python
{'metadata_text': 'Figure:putto.Oggetti:ghirlanda di fiori',
'label': 0,
'source': 'OpereArteVisiva'}
```
### Data Fields
The datafields are:
- `metadata_text`: this contains the metadata text which was sourced from [Cultura Italia](http://www.culturaitalia.it/opencms/index.jsp?language=en)
- `label`: this is the label indicating if the record is `High_Quality`, or `Low_Quality`. Most of the dataset was manually annotated, with ~30K descriptions automatically labelled as LOW quality due to their length (less than 3 tokens) or their provenance from old (pre-2012), not curated, collections.
- `source`: the source of the metadata record
### Data Splits
The dataset used 'ten-fold cross-validation' and doesn't report specific splits for train, validation and test data.
## Dataset Creation
The dataset was generated using records from [Cultura Italia](http://www.culturaitalia.it/opencms/index.jsp?language=en). From the paper introducing the dataset:
> By using the textual description encoded by the dc:description element from the Dublin Core metadata schema, we collect a dataset of 100,821 descriptions, after duplicate removal. These records include mainly data from “Musei d’Italia” and “Regione Marche” datasets, which have been chosen because they contain a high number of non-empty dc:description elements. p.221
### Curation Rationale
From the paper:
> Duplicates were removed for two reasons: this reduced annotation effort in the subsequent manual annotation, and avoided that the same example appear both in the training and in the test set, a situation that could make classification biased and lead to inaccurate evaluation in supervised settings.Footnote 10 Duplicated descriptions were mainly short and of low-quality, reporting few generic words to describe an item (e.g. “Mensola.”, “Dipinto.”). p.221
### Source Data
#### Initial Data Collection and Normalization
The dataset was generated using records from [Cultura Italia](http://www.culturaitalia.it/opencms/index.jsp?language=en). This repository is accessible via an OAI-PMH handler or via a [SPARQL endpoint](http://dati.culturaitalia.it/sparql).
As discussed above duplicates were removed from the dataset.
#### Who are the source language producers?
The metadata producers are staff working in Italian cultural heritage institutions.
### Annotations
#### Annotation process
From the paper:
> "Most of the dataset was manually annotated, with ~30K descriptions automatically labeled as LOW quality due to their length (less than 3 tokens) or their provenance from old (pre-2012), not curated, collections."
To determine the quality of the collected descriptions the authors of the paper used guidelines from the [Istituto Centrale per il Catalogo e la Documentazione](http://www.iccd.beniculturali.it/)
From the paper:
> "More precisely, a specific section of the guidelines addresses how to describe any cultural item, clarifying that both the object and the subject of the item must be presented in the description as follows:
> Object: the object typology and shape must be described. To describe the object, the cataloguer must refer to the vocabularies provided by ICCD, using specific terminology (e.g. the technique used for paintings and drawings, or the material for the archaeological items);
> Subject: the cataloguer must report the iconographic and decorative settings of the item, such as the characters of the depicted scene in a painting and their attribution. Other aspects (e.g. the history behind the painting or the painter) should not be included." p.221
[More Information Needed]
#### Who are the annotators?
> "The annotation is carried out by an expert in cultural heritage who collaborated in the past with Cultura Italia and has therefore in-depth knowledge of the data characteristics and of the ICCD guidelines." p.222
### Personal and Sensitive Information
No personal or sensitive information is described in the paper.
## Considerations for Using the Data
### Social Impact of Dataset
[More Information Needed]
### Discussion of Biases
[More Information Needed]
### Other Known Limitations
[More Information Needed]
## Additional Information
### Dataset Curators
- Lorenzini, Matteo
- Rospocher, Marco
- Tonelli, Sara
### Licensing Information
[cc-by-4.0](https://creativecommons.org/licenses/by/4.0/)
### Citation Information
```
@article{Lorenzini2020,
author = "Matteo Lorenzini and Marco Rospocher and Sara Tonelli",
title = "{Annotated dataset to assess the accuracy of the textual description of cultural heritage records}",
year = "2020",
month = "12",
url = "https://figshare.com/articles/dataset/Annotated_dataset_to_assess_the_accuracy_of_the_textual_description_of_cultural_heritage_records/13359104",
doi = "10.6084/m9.figshare.13359104.v1"
}
```
### Contributions
Thanks to [@davanstrien](https://github.com/davanstrien) for adding this dataset.
|
false |
## Table of Contents
- [Dataset Description](#dataset-description)
- [Dataset Summary](#dataset-summary)
- [Supported Tasks and Leaderboards](#supported-tasks-and-leaderboards)
- [Installation](#installation)
- [Install requirements](#install-requirements)
- [Download settings](#download-settings)
- [Dataset Structure](#dataset-structure)
- [Data Instances](#data-instances)
- [Data Fields](#data-fields)
- [Data Splits](#data-splits)
- [Dataset Creation](#dataset-creation)
- [Personal and Sensitive Information](#personal-and-sensitive-information)
- [Considerations for Using the Data](#considerations-for-using-the-data)
- [Social Impact of Dataset](#social-impact-of-dataset)
- [Discussion of Biases](#discussion-of-biases)
- [Other Known Limitations](#other-known-limitations)
- [Additional Information](#additional-information)
- [Dataset Curators](#dataset-curators)
- [Licensing Information](#licensing-information)
- [Citation Information](#citation-information)
- [Contributions](#contributions)
## Dataset Description
- **Homepage:** [https://www.kietzmannlab.org/ecoset](https://www.kietzmannlab.org/ecoset/)
- **Repository:** [https://codeocean.com/capsule/9570390/tree/v1](https://codeocean.com/capsule/9570390/tree/v1)
- **Paper:** [https://www.pnas.org/doi/full/10.1073/pnas.2011417118](https://doi.org/10.1073/pnas.2011417118)
- **Point of Contact:** [tim.kietzmann@uni-osnabrueck.de](tim.kietzmann@uni-osnabrueck.de)
### Dataset Summary
Tired of all the dogs in ImageNet (ILSVRC)? Then ecoset is here for you. 1.5m images
from 565 basic level categories, chosen to be both (i) frequent in linguistic usage,
and (ii) rated by human observers as concrete (e.g. ‘table’ is concrete, ‘romance’
is not).
Ecoset is a typical image recognition dataset, combining images of objects with appropriate
labels (one label per image). Importantly, ecoset is intended to provide higher ecological
validity than its counterparts, with a mislabelling error rate < 5% and filtered for NSFW content.
For more information on the dataset, consider reading the [original publication](https://doi.org/10.1073/pnas.2011417118).
Ecoset consists of a train, test, and validation subset which all are openly available to the user.
### Supported Tasks and Leaderboards
Ecoset is a large multi-class single-label object recognition image dataset (similar to ImageNet).
## Installation
### Install Requirements
In order to work with ecoset, please make sure to install the s3 compatible version of huggingface datasets, which should include the `s3fs`, `botocore` and `boto3` modules:
```bash
pip install datasets[s3]
```
If you want to work with the dataset in `Huggingface.datasets`, you might also want to make sure to install PIL (`pip install Pillow`) in order to work with image input. However, downloading the dataset will work despite not having installed PIL.
### Download Settings
Please set `ignore_verifications=True`. when downloading this dataset, else the download will result in an error:
```python
from datasets import load_dataset
dataset = load_dataset("kietzmannlab/ecoset", ignore_verifications=True)
```
| NOTE: If you get errors like: `FileNotFoundError: [Errno 2] No such file or directory:'<DATASET_PATH>'` this is likely due do having previously downloaded the dataset and then cancelling the download. If this is the case for you, you can fix this error by manually removing the dataset path and reinstalling the dataset. |
| --- |
## Dataset Structure
We show detailed information for all the configurations of the dataset. Currently, there is only one setting (`Full`) available, containing all data.
### Data Instances
#### Full
- **Size of downloaded dataset files:** 155 GB
- **Total amount of disk used:** 311 GB
## Dataset Creation
A total of 565 categories were selected based on the following: 1) their word frequency in American television and film subtitles (SUBTLEX_US), 2) the perceived concreteness by human observers, and 3) the availability of a minimum of 700 images. Images were sourced via the overall ImageNet database (the same resource used for ILSVRC 2012) or obtained under CC BY-NC-SA 2.0 license from Bing image search and Flickr. Thorough data cleaning procedures were put in place to remove duplicates and to assure an expected misclassification rate per category of <4%.
### Curation Rationale
More information on the curation of the dataset can be found in the [original publication](https://doi.org/10.1073/pnas.2011417118).
### Source Data
The source data is available under: [https://codeocean.com/capsule/9570390/tree/v1](https://codeocean.com/capsule/9570390/tree/v1)
### Annotations
Each ecoset image folder is annotated with class labels according to the main object depicted in a class of images. No further annotations are added to the dataset.
### Personal and Sensitive Information
The dataset was tested to exclude sensitive images using Yahoo's Open NSFW detection model, removing all image with an NSFW score above 0.8. For this dataset, only images with secured license information was used, which should prevent the inclusion of images without consent of the image's authors and subjects. Despite these measures, it is possible that the images in the dataset contain personal and sensitive information.
## Considerations for Using the Data
### Social Impact of Dataset
Large-scale image-label datasets such as ImageNet are the backbone of modern Computer Vision. However, such large datasets often suffer from problems like mislabeling, category biases, misrepresentations, and unsafe content. Ecoset was created with the aim to reduce these biases and consequently improve the social impact of Computer Vision techniques trained on the dataset. More information on the social impact of the dataset can be found in the [original publication](https://doi.org/10.1073/pnas.2011417118).
### Discussion of Biases
Despite best efforts to provide an ecologically valid and overall less biased dataset, ecoset is still likely to contain biased data. The category selection of ecoset was based on human concreteness ratings and word frequencies in a corpus consisting of American television and film subtitles. This undoubtedly biases the category selection toward Western cultures. Image inclusion was based on the availability via Bing/Flickr search results as well as the existence of relevant ImageNet categories. Images depicting people, specifically the categories “man,” “woman,” and “child,” were not sampled according to census distributions (age, ethnicity, gender, etc.).
### Other Known Limitations
In addition to points mentioned in [Discussion of Biases](#discussion-of-biases), ecoset image and category distributions do not reflect the naturalistic, egocentric visual input typically encountered in the everyday life of infant and adults.
## Additional Information
### Dataset Curators
The corpus was put together by Johannes Mehrer, Courtney J. Spoerer, Emer C. Jones, Nikolaus Kriegeskorte, and Tim C. Kietzmann.
### Licensing Information
Ecoset is licensed under Creative Commons Attribution-NonCommercial-ShareAlike 2.0 license (cc-by-nc-sa-2.0).
### Citation Information
```
@article{mehrer2021ecologically,
title={An ecologically motivated image dataset for deep learning yields better models of human vision},
author={Mehrer, Johannes and Spoerer, Courtney J and Jones, Emer C and Kriegeskorte, Nikolaus and Kietzmann, Tim C},
journal={Proceedings of the National Academy of Sciences},
volume={118},
number={8},
pages={e2011417118},
year={2021},
publisher={National Acad Sciences}
}
```
### Contributions
The ecoset dataloader and dataset card was created by [@DiGyt](https://github.com/DiGyt) on behalf of [@kietzmannlab](https://huggingface.co/kietzmannlab).
For questions and suggestions feel free to reach out.
|
false |
This is the story generation datasets collected by TextBox, including:
- ROCStories (roc)
- WritingPrompts (wp)
- Hippocorpus (hc)
- WikiPlots (wikip)
- ChangeMyView (cmv).
The detail and leaderboard of each dataset can be found in [TextBox page](https://github.com/RUCAIBox/TextBox#dataset). |
false |
# TeTIm-Eval
|
false |
# Dataset Card for Swedish Citesum Dataset
The Swedish citesum dataset has only been machine-translated to improve downstream fine-tuning on Swedish summarization tasks.
## Dataset Summary
Read about the full details at original English version: https://huggingface.co/datasets/citesum
### Paper
https://arxiv.org/abs/2205.06207
### Authors
Yuning Mao, Ming Zhong, Jiawei Han
University of Illinois Urbana-Champaign
{yuningm2, mingz5, hanj}@illinois.edu
## Data details
- src (string): source text. long description of paper
- tgt (string): target text. tldr of paper
- paper_id (string): unique id for the paper
- title (string): title of the paper
- discipline (dict):
- venue (string): Where the paper was published (conference)
- journal (string): Journal in which the paper was published
- mag_field_of_study (list[str]): scientific fields that the paper falls under.
### Data Splits
The Swedish xsum dataset follows the same splits as the original English version and has 3 splits: _train_, _validation_, and _test_.
| Dataset Split | Number of Instances in Split |
| ------------- | ------------------------------------------- |
| Train | 83,304 |
| Validation | 4,721 |
| Test | 4,921 |
|
false |
### dataset description
We downloaded PubChem-10m dataset from [here](https://deepchemdata.s3-us-west-1.amazonaws.com/datasets/pubchem_10m.txt.zip) and canonicalized it.
We used the following function to canonicalize the data and removed some SMILES that cannot be read by RDKit.
```python:
from rdkit import Chem
def canonicalize(mol):
mol = Chem.MolToSmiles(Chem.MolFromSmiles(mol),True)
return mol
```
We randomly split the preprocessed data into train and validation. The ratio is 9 : 1. |
true |
# Dataset Description
This is a dataset of emotional contexts that was retrieved from the original EmpatheticDialogues (ED) dataset. Respondents were asked to describe an event that was associated with a particular emotion label (i.e. p(event|emotion).
There are 32 emotion labels in total.
There are 19209, 2756, and 2542 instances of emotional descriptions in the train, valid, and test set, respectively. |
false | Initial annotated dataset derived from `ImageIN/IA_unlabelled` |
false |
# KPBiomed, A Large-Scale Dataset for keyphrase generation
## About
This dataset is made of 5.6 million abstracts with author assigned keyphrases.
Details about the dataset can be found in the original paper:
Maël Houbre, Florian Boudin and Béatrice Daille. 2022. [A Large-Scale Dataset for Biomedical Keyphrase Generation](https://arxiv.org/abs/2211.12124). In Proceedings of the 13th International Workshop on Health Text Mining and Information Analysis (LOUHI 2022).
Reference (author-assigned) keyphrases are also categorized under the PRMU (<u>P</u>resent-<u>R</u>eordered-<u>M</u>ixed-<u>U</u>nseen) scheme as proposed in the following paper:
- Florian Boudin and Ygor Gallina. 2021.
[Redefining Absent Keyphrases and their Effect on Retrieval Effectiveness](https://aclanthology.org/2021.naacl-main.330/).
In Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, pages 4185–4193, Online. Association for Computational Linguistics.
Text pre-processing (tokenization) is carried out using spacy (en_core_web_sm model) with a special rule to avoid splitting words with hyphens (e.g. graph-based is kept as one token). Stemming (Porter's stemmer implementation provided in nltk) is applied before reference keyphrases are matched against the source text.
## Content
The details of the dataset are in the table below:
| Split | # documents | # keyphrases by document (average) | % Present | % Reordered | % Mixed | % Unseen |
| :----------- | ----------: | ---------------------------------: | --------: | ----------: | ------: | -------: |
| Train small | 500k | 5.24 | 66.31 | 7.16 | 12.60 | 13.93 |
| Train medium | 2M | 5.24 | 66.30 | 7.18 | 12.57 | 13.95 |
| Train large | 5.6M | 5.23 | 66.32 | 7.18 | 12.55 | 13.95 |
| Validation | 20k | 5.25 | 66.44 | 7.07 | 12.45 | 14.05 |
| Test | 20k | 5.22 | 66.59 | 7.22 | 12.44 | 13.75 |
The following data fields are available:
- **id**: unique identifier of the document.
- **title**: title of the document.
- **abstract**: abstract of the document.
- **keyphrases**: list of reference keyphrases.
- **mesh terms**: list of indexer assigned MeSH terms if available (around 68% of the articles)
- **prmu**: list of <u>P</u>resent-<u>R</u>eordered-<u>M</u>ixed-<u>U</u>nseen categories for reference keyphrases.
- **authors**: list of the article's authors
- **year**: publication year
**NB**: The present keyphrases (represented by the "P" label in the PRMU column) are sorted by their apparition order in the text (title + text).
|
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.