README.md

metadata

license: apache-2.0
language:
  - en
  - am
  - ig
  - fr
  - sn
  - ln
  - lug
  - wo
  - ee
  - xh
  - kin
  - tw
  - zu
  - orm
  - yo
  - ha
  - sot
  - sw
size_categories:
  - n<1K
multilinguality:
  - multilingual
pretty_name: afrixnli
language_details: >-
  eng, amh, ibo, fra, sna, lin, wol, ewe, lug, xho, kin, twi, zul, orm, yor,
  hau, sot, swa
source_datasets:
  - xnli
tags:
  - afrixnli
  - afri-xnli
  - africanxnli
task_categories:
  - text-classification
task_ids:
  - natural-language-inference
configs:
  - config_name: amh
    data_files:
      - split: validation
        path: data/amh/dev.tsv
      - split: test
        path: data/amh/test.tsv
  - config_name: eng
    data_files:
      - split: validation
        path: data/eng/dev.tsv
      - split: test
        path: data/eng/test.tsv
  - config_name: ewe
    data_files:
      - split: validation
        path: data/ewe/dev.tsv
      - split: test
        path: data/ewe/test.tsv
  - config_name: fra
    data_files:
      - split: validation
        path: data/fra/dev.tsv
      - split: test
        path: data/fra/test.tsv
  - config_name: hau
    data_files:
      - split: validation
        path: data/hau/dev.tsv
      - split: test
        path: data/hau/test.tsv
  - config_name: ibo
    data_files:
      - split: validation
        path: data/ibo/dev.tsv
      - split: test
        path: data/ibo/test.tsv
  - config_name: kin
    data_files:
      - split: validation
        path: data/kin/dev.tsv
      - split: test
        path: data/kin/test.tsv
  - config_name: lin
    data_files:
      - split: validation
        path: data/lin/dev.tsv
      - split: test
        path: data/lin/test.tsv
  - config_name: lug
    data_files:
      - split: validation
        path: data/lug/dev.tsv
      - split: test
        path: data/lug/test.tsv
  - config_name: orm
    data_files:
      - split: validation
        path: data/orm/dev.tsv
      - split: test
        path: data/orm/test.tsv
  - config_name: sna
    data_files:
      - split: validation
        path: data/sna/dev.tsv
      - split: test
        path: data/sna/test.tsv
  - config_name: sot
    data_files:
      - split: validation
        path: data/sot/dev.tsv
      - split: test
        path: data/sot/test.tsv
  - config_name: swa
    data_files:
      - split: validation
        path: data/swa/dev.tsv
      - split: test
        path: data/swa/test.tsv
  - config_name: twi
    data_files:
      - split: validation
        path: data/twi/dev.tsv
      - split: test
        path: data/twi/test.tsv
  - config_name: wol
    data_files:
      - split: validation
        path: data/wol/dev.tsv
      - split: test
        path: data/wol/test.tsv
  - config_name: xho
    data_files:
      - split: validation
        path: data/xho/dev.tsv
      - split: test
        path: data/xho/test.tsv
  - config_name: yor
    data_files:
      - split: validation
        path: data/yor/dev.tsv
      - split: test
        path: data/yor/test.tsv
  - config_name: zul
    data_files:
      - split: validation
        path: data/zul/dev.tsv
      - split: test
        path: data/zul/test.tsv

Dataset Card for afrixnli

Table of Contents

• Table of Contents
• Dataset Description
  • Dataset Summary
  • Languages
• Dataset Structure
  • Data Instances
  • Data Fields
  • Data Splits

Dataset Description

• Point of Contact: d.adelani@ucl.ac.uk

Dataset Summary

AFRIXNLI is an evaluation dataset comprising translations of a subset of the XNLI dataset into 16 African languages. It includes both validation and test sets across all 18 languages, maintaining the English and French subsets from the original XNLI dataset.

Languages

There are 18 languages available :

Dataset Structure

Data Instances

The examples look like this for English:

from datasets import load_dataset
data = load_dataset('masakhane/afrixnli', 'eng') 
# Please, specify the language code
# A data point example is below:
{
'premise': 'The doors were locked when we went in.',
'hypothesis': 'All of the doors were open.',
'label': 0
}

Data Fields

• premise: a multilingual string variable,
• hypothesis: a multilingual string variable,
• label: a classification label, with possible values including entailment (0), neutral (1), contradiction (2).

Data Splits

All languages has two splits, dev and test a subset of the original dev and test splits of the XNLI dataset.

The splits have the following sizes :

Language	validation	test
English	450	600