Vinay Setty
commited on
Commit
·
235d34e
1
Parent(s):
a2bfa22
Initial arabic data commit
Browse files- clef24_dataset_twitter.py +80 -0
- data/dev.tsv +0 -0
- data/test.tsv +0 -0
- data/train.tsv +0 -0
clef24_dataset_twitter.py
ADDED
@@ -0,0 +1,80 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
"""Multilang Dataset loading script."""
|
2 |
+
|
3 |
+
from datasets import DatasetInfo, BuilderConfig, Version, GeneratorBasedBuilder, DownloadManager
|
4 |
+
from datasets import SplitGenerator, Split, Features, Value
|
5 |
+
from typing import Generator, Tuple, Union
|
6 |
+
|
7 |
+
import os
|
8 |
+
|
9 |
+
_DESCRIPTION = """
|
10 |
+
This dataset includes Arabic/Dutch/Spanish Twitter data for CLEF 2024 CheckThat! Lab task1.
|
11 |
+
"""
|
12 |
+
|
13 |
+
_CITATION = """\
|
14 |
+
@inproceedings{barron2024clef,
|
15 |
+
title={The CLEF-2024 CheckThat! Lab: Check-Worthiness, Subjectivity, Persuasion, Roles, Authorities, and Adversarial Robustness},
|
16 |
+
author={Barr{\'o}n-Cede{\~n}o, Alberto and Alam, Firoj and Chakraborty, Tanmoy and Elsayed, Tamer and Nakov, Preslav and Przyby{\l}a, Piotr and Stru{\ss}, Julia Maria and Haouari, Fatima and Hasanain, Maram and Ruggeri, Federico and others},
|
17 |
+
booktitle={European Conference on Information Retrieval},
|
18 |
+
pages={449--458},
|
19 |
+
year={2024},
|
20 |
+
organization={Springer}
|
21 |
+
}
|
22 |
+
"""
|
23 |
+
|
24 |
+
_LICENSE = "Your dataset's license here."
|
25 |
+
|
26 |
+
class CLEF24EsData(GeneratorBasedBuilder):
|
27 |
+
"""A multilingual text dataset."""
|
28 |
+
|
29 |
+
BUILDER_CONFIGS = [
|
30 |
+
BuilderConfig(name="clef24_tweet_data", version=Version("1.0.0"), description="Multilingual dataset for text classification."),
|
31 |
+
]
|
32 |
+
|
33 |
+
DEFAULT_CONFIG_NAME = "clef24_tweet_data" # Default configuration name.
|
34 |
+
|
35 |
+
def _info(self):
|
36 |
+
"""Construct the DatasetInfo object."""
|
37 |
+
return DatasetInfo(
|
38 |
+
description=_DESCRIPTION,
|
39 |
+
features=Features({
|
40 |
+
"tweet_id": Value("string"),
|
41 |
+
"tweet_url": Value("string"),
|
42 |
+
"tweet_text": Value("string"),
|
43 |
+
"class_label": Value("string"),
|
44 |
+
}),
|
45 |
+
supervised_keys=("tweet_text", "class_label"),
|
46 |
+
homepage="https://gitlab.com/checkthat_lab/clef2024-checkthat-lab/-/tree/main/task1",
|
47 |
+
citation=_CITATION,
|
48 |
+
license=_LICENSE,
|
49 |
+
)
|
50 |
+
|
51 |
+
def _split_generators(self, dl_manager: DownloadManager) -> list[SplitGenerator]:
|
52 |
+
"""Returns SplitGenerators."""
|
53 |
+
# Assumes your dataset is located in "data"
|
54 |
+
data_dir = os.path.abspath("data")
|
55 |
+
splits = {"train": Split.TRAIN, "dev": Split.VALIDATION, "test": Split.TEST}
|
56 |
+
|
57 |
+
return [
|
58 |
+
SplitGenerator(
|
59 |
+
name=splits[split],
|
60 |
+
gen_kwargs={
|
61 |
+
"filepath": os.path.join(data_dir, f"{split}.tsv"),
|
62 |
+
"split": splits[split]
|
63 |
+
},
|
64 |
+
)
|
65 |
+
for split in splits.keys()
|
66 |
+
]
|
67 |
+
|
68 |
+
def _generate_examples(self, filepath: Union[str, os.PathLike], split: str) -> Generator[Tuple[str, dict], None, None]:
|
69 |
+
"""Yields examples."""
|
70 |
+
with open(filepath, encoding="utf-8") as f:
|
71 |
+
for id_, row in enumerate(f):
|
72 |
+
if id_ == 0: # Optionally skip header
|
73 |
+
continue
|
74 |
+
cols = row.strip().split('\t')
|
75 |
+
yield f"{split}_{id_}", {
|
76 |
+
"tweet_id": cols[0],
|
77 |
+
"tweet_url": cols[1],
|
78 |
+
"tweet_text": cols[2],
|
79 |
+
"class_label": cols[3],
|
80 |
+
}
|
data/dev.tsv
ADDED
The diff for this file is too large to render.
See raw diff
|
|
data/test.tsv
ADDED
The diff for this file is too large to render.
See raw diff
|
|
data/train.tsv
ADDED
The diff for this file is too large to render.
See raw diff
|
|