bourdoiscatie's picture
Update README.md
4b795fc
metadata
language:
  - fr
license: cc-by-nc-sa-4.0
size_categories:
  - 100K<n<1M
task_categories:
  - text-generation
tags:
  - DFP
  - french prompts
annotations_creators:
  - found
language_creators:
  - found
multilinguality:
  - monolingual
source_datasets:
  - newsquadfr

newsquadfr_fr_prompt_context_generation_with_answer

Summary

newsquadfr_fr_prompt_context_generation_with_answer is a subset of the Dataset of French Prompts (DFP).
It contains 101,040 rows that can be used for a context-generation (with answer)task.
The original data (without prompts) comes from the dataset newsquadfr and was augmented by questions in SQUAD 2.0 format in the FrenchQA dataset. A list of prompts (see below) was then applied in order to build the input and target columns and thus obtain the same format as the xP3 dataset by Muennighoff et al.

Prompts used

List

24 prompts were created for this dataset. The logic applied consists in proposing prompts in the indicative tense, in the form of tutoiement and in the form of vouvoiement.

'Étant donné la réponse "'+ answer+'", écrire un texte explicatif.\nTexte : ',  
'Étant donné la réponse "'+ answer+'", écris un texte explicatif.\nTexte : ',  
'Étant donné la réponse "'+ answer+'", écrivez un texte explicatif.\nTexte : ',  
'Étant donné la réponse "'+ answer+'", rédiger un texte explicatif.\nTexte : ',  
'Étant donné la réponse "'+ answer+'", rédige un texte explicatif.\nTexte : ',  
'Étant donné la réponse "'+ answer+'", rédigez un texte explicatif.\nTexte : ',  
'Étant donné la réponse "'+ answer+'", générer un texte explicatif.\nTexte : ',  
'Étant donné la réponse "'+ answer+'", génère un texte explicatif.\nTexte : ',  
'Étant donné la réponse "'+ answer+'", générez un texte explicatif.\nTexte : ',  
'Étant donné la réponse "'+ answer+'", créer un texte explicatif.\nTexte : ',  
'Étant donné la réponse "'+ answer+'", crée un texte explicatif.\nTexte : ',  
'Étant donné la réponse "'+ answer+'", créez un texte explicatif.\nTexte : ',  
'Ecrire un texte comme contexte de la réponse "'+ answer+'" \nTexte : ',  
'Ecris un texte comme contexte de la réponse "'+ answer+'" \nTexte : ',  
'Ecrivez un texte comme contexte de la réponse "'+ answer+'" \nTexte : ',  
'Rédiger un texte comme contexte de la réponse "'+ answer+'" \nTexte : ',  
'Rédige un texte comme contexte de la réponse "'+ answer+'" \nTexte : ',  
'Rédigez un texte comme contexte de la réponse "'+ answer+'" \nTexte : ',  
'Générer un texte comme contexte de la réponse "'+ answer+'" \nTexte : ',  
'Génère un texte comme contexte de la réponse "'+ answer+'" \nTexte : ',  
'Générez un texte comme contexte de la réponse "'+ answer+'" \nTexte : ',  
'Créer un texte comme contexte de la réponse "'+ answer+'" \nTexte : ',  
'Crée un texte comme contexte de la réponse "'+ answer+'" \nTexte : ',  
'Créez un texte comme contexte de la réponse "'+ answer+'" \nTexte : ',  

Splits

  • train with 79,200 samples
  • valid with 21,800 samples
  • no test split

How to use?

from datasets import load_dataset
dataset = load_dataset("CATIE-AQ/newsquadfr_fr_prompt_context_generation_with_answer")

Citation

Original data

Hugging Face repository: https://huggingface.co/datasets/lincoln/newsquadfr

This Dataset

@misc {centre_aquitain_des_technologies_de_l'information_et_electroniques_2023,
author = { {Centre Aquitain des Technologies de l'Information et Electroniques} },
title = { DFP (Revision 1d24c09) },
year = 2023,
url = { https://huggingface.co/datasets/CATIE-AQ/DFP },
doi = { 10.57967/hf/1200 },
publisher = { Hugging Face }
}

License

CC BY-NC-SA 4.0