bourdoiscatie
commited on
Commit
•
b6bfa21
1
Parent(s):
e18c39c
Create README.md
Browse files
README.md
ADDED
@@ -0,0 +1,91 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
language:
|
3 |
+
- fr
|
4 |
+
license: cc-by-sa-4.0
|
5 |
+
size_categories:
|
6 |
+
- 10K<n<100K
|
7 |
+
task_categories:
|
8 |
+
- text-generation
|
9 |
+
---
|
10 |
+
|
11 |
+
# orange_sum_fr_prompt_text_generation
|
12 |
+
## Summary
|
13 |
+
|
14 |
+
**orange_sum_fr_prompt_text_generation** is a subset of the [**Dataset of French Prompts (DFP)**]().
|
15 |
+
It contains **48,069** rows that can be used for a part-of-speech task.
|
16 |
+
The original data (without prompts) comes from the dataset [orange_sum](https://huggingface.co/datasets/orange_sum) by Eddine et al.
|
17 |
+
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](https://huggingface.co/datasets/bigscience/xP3) dataset by Muennighoff et al.
|
18 |
+
|
19 |
+
|
20 |
+
|
21 |
+
## Prompts used
|
22 |
+
### List
|
23 |
+
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.
|
24 |
+
|
25 |
+
```
|
26 |
+
'"'+document+'"\n Continuer le texte sur 4000 caractères maximum :',
|
27 |
+
'"'+document+'"\n Continue le texte sur 4000 caractères maximum :',
|
28 |
+
'"'+document+'"\n Continuez le texte sur 4000 caractères maximum :',
|
29 |
+
'"'+document+'"\n Poursuivre le texte sur 4000 caractères maximum :',
|
30 |
+
'"'+document+'"\n Poursuis le texte sur 4000 caractères maximum :',
|
31 |
+
'"'+document+'"\n Poursuivez le texte sur 4000 caractères maximum :',
|
32 |
+
'"'+document+'"\n Prolonger le texte sur 4000 caractères maximum :',
|
33 |
+
'"'+document+'"\n Prolonge le texte sur 4000 caractères maximum :',
|
34 |
+
'"'+document+'"\n Prolongez le texte sur 4000 caractères maximum :',
|
35 |
+
'"'+document+'"\n Rédiger la suite du texte : ',
|
36 |
+
'"'+document+'"\n Rédige la suite du texte : ',
|
37 |
+
'"'+document+'"\n Rédigez la suite du texte : ',
|
38 |
+
'"'+document+'"\n Imaginer la suite du texte : ',
|
39 |
+
'"'+document+'"\n Imagine la suite du texte : ',
|
40 |
+
'"'+document+'"\n Imaginez la suite du texte : ',
|
41 |
+
'"'+document+'"\n Ecrire la suite du texte : ',
|
42 |
+
'"'+document+'"\n Ecris la suite du texte : ',
|
43 |
+
'"'+document+'"\n Ecriver la suite du texte : ',
|
44 |
+
'"'+document+'"\n Développer la suite du texte : ',
|
45 |
+
'"'+document+'"\n Développe la suite du texte : ',
|
46 |
+
'"'+document+'"\n Développez la suite du texte : ',
|
47 |
+
'"'+document+'"\nGénérer la suite du texte : ',
|
48 |
+
'"'+document+'"\nGénère la suite du texte : ',
|
49 |
+
'"'+document+'"\n Générez la suite du texte : ',
|
50 |
+
```
|
51 |
+
|
52 |
+
### Features used in the prompts
|
53 |
+
In the prompt list above, `text` and `targets` have been constructed from:
|
54 |
+
```
|
55 |
+
orange_sum = load_dataset('orange_sum','abstract')
|
56 |
+
if len(orange_sum['train'][i]['text']) > 1000:
|
57 |
+
document = orange_sum['train'][i]['text'][:1000]
|
58 |
+
targets = orange_sum['train'][i]['summary'][1000:]
|
59 |
+
```
|
60 |
+
|
61 |
+
|
62 |
+
|
63 |
+
# Splits
|
64 |
+
- train with 48,069 samples
|
65 |
+
- no valid and test splits in the original data
|
66 |
+
|
67 |
+
|
68 |
+
# How to use?
|
69 |
+
```
|
70 |
+
from datasets import load_dataset
|
71 |
+
dataset = load_dataset("CATIE-AQ/orange_sum_fr_prompt_text_generation")
|
72 |
+
```
|
73 |
+
|
74 |
+
# Citation
|
75 |
+
## Original data
|
76 |
+
> @article{eddine2020barthez,
|
77 |
+
title={BARThez: a Skilled Pretrained French Sequence-to-Sequence Model},
|
78 |
+
author={Eddine, Moussa Kamal and Tixier, Antoine J-P and Vazirgiannis, Michalis},
|
79 |
+
journal={arXiv preprint arXiv:2010.12321},
|
80 |
+
year={2020}
|
81 |
+
}
|
82 |
+
|
83 |
+
|
84 |
+
|
85 |
+
|
86 |
+
## This Dataset
|
87 |
+
|
88 |
+
|
89 |
+
|
90 |
+
## License
|
91 |
+
CC-BY-SA-4.0
|