Datasets:

goaicorp
/

flores-plus-fra-dyu

Modalities:

Formats:

Size:

Libraries:

Dataset card Data Studio Files Files and versions

You need to agree to share your contact information to access this dataset

This repository is publicly accessible, but you have to accept the conditions to access its files and content.

Log in or Sign Up to review the conditions and access this dataset content.

FLORES+ French ↔ Dioula Benchmark Dataset

A parallel sentence dataset aligned from FLORES+ for the fra_Latn (French) and dyu_Latn (Dioula) language pairs.

Dataset Structure

Splits: test (997 sentences, from FLORES+ dev) and devtest (1012 sentences)
Languages: French (fra_Latn) · Dioula (dyu_Latn)

Each record contains:

sentence_fra_Latn: French sentence
sentence_dyu_Latn: Dioula sentence
id, domain, topic, url, has_image, has_hyperlink: metadata from FLORES+

Intended Use

This dataset is intended for benchmarking machine translation and multilingual NLP models on the French ↔ Dioula language pair.

Source & License

Source: openlanguagedata/flores_plus
License: CC BY-SA 4.0

Downloads last month: 93

Total file size: