Datasets:

Multilinguality:
translation
Size Categories:
1K<n<10K
Language Creators:
found
Annotations Creators:
found
ArXiv:
License:
flores / README.md
lhoestq's picture
lhoestq HF staff
rename configs to config_name
33a2ca5
metadata
pretty_name: Flores
annotations_creators:
  - found
language_creators:
  - found
language:
  - en
  - ne
  - si
license:
  - cc-by-4.0
multilinguality:
  - translation
size_categories:
  - 1K<n<10K
source_datasets:
  - extended|wikipedia
  - extended|opus_gnome
  - extended|opus_ubuntu
  - extended|open_subtitles
  - extended|paracrawl
  - extended|bible_para
  - extended|kde4
  - extended|other-global-voices
  - extended|other-common-crawl
task_categories:
  - translation
task_ids: []
paperswithcode_id: flores
dataset_info:
  - config_name: neen
    features:
      - name: translation
        dtype:
          translation:
            languages:
              - ne
              - en
    splits:
      - name: validation
        num_bytes: 849380
        num_examples: 2560
      - name: test
        num_bytes: 999063
        num_examples: 2836
    download_size: 1542781
    dataset_size: 1848443
  - config_name: sien
    features:
      - name: translation
        dtype:
          translation:
            languages:
              - si
              - en
    splits:
      - name: validation
        num_bytes: 1031158
        num_examples: 2899
      - name: test
        num_bytes: 983563
        num_examples: 2767
    download_size: 1542781
    dataset_size: 2014721
config_names:
  - neen
  - sien

Dataset Card for "flores"

Table of Contents

Dataset Description

Dataset Summary

Evaluation datasets for low-resource machine translation: Nepali-English and Sinhala-English.

Supported Tasks and Leaderboards

More Information Needed

Languages

More Information Needed

Dataset Structure

Data Instances

neen

  • Size of downloaded dataset files: 1.54 MB
  • Size of the generated dataset: 1.86 MB
  • Total amount of disk used: 3.40 MB

An example of 'validation' looks as follows.

This example was too long and was cropped:

{
    "translation": "{\"en\": \"This is the wrong translation!\", \"ne\": \"यस वाहेक आगम पूजा, तारा पूजा, व्रत आदि पनि घरभित्र र वाहिर दुवै स्थानमा गरेको पा..."
}

sien

  • Size of downloaded dataset files: 1.54 MB
  • Size of the generated dataset: 2.01 MB
  • Total amount of disk used: 3.57 MB

An example of 'validation' looks as follows.

This example was too long and was cropped:

{
    "translation": "{\"en\": \"This is the wrong translation!\", \"si\": \"එවැනි ආවරණයක් ලබාදීමට රක්ෂණ සපයන්නෙකු කැමති වුවත් ඒ සාමාන් යයෙන් බොහෝ රටවල පොදු ..."
}

Data Fields

The data fields are the same among all splits.

neen

  • translation: a multilingual string variable, with possible languages including ne, en.

sien

  • translation: a multilingual string variable, with possible languages including si, en.

Data Splits

name validation test
neen 2560 2836
sien 2899 2767

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

@misc{guzmn2019new,
    title={Two New Evaluation Datasets for Low-Resource Machine Translation: Nepali-English and Sinhala-English},
    author={Francisco Guzman and Peng-Jen Chen and Myle Ott and Juan Pino and Guillaume Lample and Philipp Koehn and Vishrav Chaudhary and Marc'Aurelio Ranzato},
    year={2019},
    eprint={1902.01382},
    archivePrefix={arXiv},
    primaryClass={cs.CL}
}

Contributions

Thanks to @thomwolf, @patrickvonplaten, @lewtun for adding this dataset.