|
--- |
|
tags: |
|
- augmented-indonli |
|
license: bigscience-openrail-m |
|
datasets: |
|
- indonli |
|
--- |
|
|
|
On this repository, I just downloaded and processed the `translate_train.tar.gz` file with `train.jsonl` & `dev.jsonl` here: `https://github.com/ir-nlp-csui/indonli/tree/main/data` |
|
|
|
How to use? As simple as this: |
|
|
|
```python |
|
!wget https://huggingface.co/datasets/muhammadravi251001/augmented-indo-nli/raw/main/dev_augmented.jsonl |
|
!wget https://huggingface.co/datasets/muhammadravi251001/augmented-indo-nli/resolve/main/train_augmented.jsonl |
|
|
|
import pandas as pd |
|
data_train_augmented_indonli = pd.read_json(path_or_buf='train.jsonl', lines=True) |
|
data_dev_augmented_indonli = pd.read_json(path_or_buf='dev.jsonl', lines=True) |
|
) |
|
``` |
|
Voila~! |
|
|
|
## Reference |
|
|
|
The dataset I used is by IndoNLI. |
|
|
|
``` |
|
@inproceedings{indonli, |
|
title = "IndoNLI: A Natural Language Inference Dataset for Indonesian", |
|
author = "Mahendra, Rahmad and Aji, Alham Fikri and Louvan, Samuel and Rahman, Fahrurrozi and Vania, Clara", |
|
booktitle = "Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing", |
|
month = nov, |
|
year = "2021", |
|
publisher = "Association for Computational Linguistics", |
|
} |
|
``` |