IndoNLP / nusa-crowd

A collaborative project to collect datasets in Indonesian languages.

Apache License 2.0

262 stars 62 forks source link

Create dataset loader for XCOPA #153

Closed SamuelCahyawijaya closed 2 years ago

SamuelCahyawijaya commented 2 years ago

https://indonlp.github.io/nusa-catalogue/card.html?xcopa

SamuelCahyawijaya commented 2 years ago

XCOPA would fit MCQA format with id as idx, context as premises, question as question, choices is a list [choice_1, choice_2], and label is the index label from the dataset

yana-xuyan commented 2 years ago

self-assign

yana-xuyan commented 2 years ago

question is not used in this task, so I think question should be empty.