Create codegauntlt.py
Browse files- codegauntlt.py +69 -0
codegauntlt.py
ADDED
@@ -0,0 +1,69 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
|
2 |
+
# coding=utf-8
|
3 |
+
# CodeGauntlt dataset loading script for Hugging Face Datasets
|
4 |
+
# path: codegauntlt.py
|
5 |
+
|
6 |
+
import json
|
7 |
+
import datasets
|
8 |
+
|
9 |
+
_DESCRIPTION = """\
|
10 |
+
CodeGauntlt is a multi-source dataset designed for evaluating and enhancing the robustness of AI code repair and generation agents. It introduces adversarially-constructed, obfuscated, or deceptive bugs across several programming languages, based on real-world and synthetic sources.
|
11 |
+
"""
|
12 |
+
|
13 |
+
_HOMEPAGE = "https://huggingface.co/datasets/HackerHardware/CodeGauntlt"
|
14 |
+
|
15 |
+
_CITATION = """\
|
16 |
+
@misc{codegauntlt2025,
|
17 |
+
title={CodeGauntlt: A Dataset for Adversarial Evaluation of Code Repair Models},
|
18 |
+
author={Esteban and Collaborators},
|
19 |
+
year={2025},
|
20 |
+
howpublished={\\url{https://huggingface.co/datasets/HackerHardware/CodeGauntlt}},
|
21 |
+
}
|
22 |
+
"""
|
23 |
+
|
24 |
+
class CodeGauntlt(datasets.GeneratorBasedBuilder):
|
25 |
+
VERSION = datasets.Version("1.0.0")
|
26 |
+
|
27 |
+
def _info(self):
|
28 |
+
features = datasets.Features(
|
29 |
+
{
|
30 |
+
"id": datasets.Value("string"),
|
31 |
+
"source": datasets.Value("string"),
|
32 |
+
"description": datasets.Value("string"),
|
33 |
+
"code_buggy": datasets.Value("string"),
|
34 |
+
"code_fixed": datasets.Value("string"),
|
35 |
+
"bug_type": datasets.Value("string"),
|
36 |
+
"tags": datasets.Value("string"),
|
37 |
+
"metadata": datasets.Value("string")
|
38 |
+
}
|
39 |
+
)
|
40 |
+
return datasets.DatasetInfo(
|
41 |
+
description=_DESCRIPTION,
|
42 |
+
features=features,
|
43 |
+
homepage=_HOMEPAGE,
|
44 |
+
citation=_CITATION,
|
45 |
+
license="apache-2.0"
|
46 |
+
)
|
47 |
+
|
48 |
+
def _split_generators(self, dl_manager):
|
49 |
+
data_dir = dl_manager.download_and_extract("./data")
|
50 |
+
return [
|
51 |
+
datasets.SplitGenerator(
|
52 |
+
name=datasets.Split.TRAIN,
|
53 |
+
gen_kwargs={"filepath": f"{data_dir}/train.jsonl"}
|
54 |
+
),
|
55 |
+
datasets.SplitGenerator(
|
56 |
+
name=datasets.Split.VALIDATION,
|
57 |
+
gen_kwargs={"filepath": f"{data_dir}/validation.jsonl"}
|
58 |
+
),
|
59 |
+
datasets.SplitGenerator(
|
60 |
+
name=datasets.Split.TEST,
|
61 |
+
gen_kwargs={"filepath": f"{data_dir}/test.jsonl"}
|
62 |
+
),
|
63 |
+
]
|
64 |
+
|
65 |
+
def _generate_examples(self, filepath):
|
66 |
+
with open(filepath, encoding="utf-8") as f:
|
67 |
+
for i, line in enumerate(f):
|
68 |
+
record = json.loads(line)
|
69 |
+
yield i, record
|