Datasets:
Tasks:
Question Answering
Modalities:
Text
Formats:
parquet
Sub-tasks:
extractive-qa
Languages:
English
Size:
1K - 10K
ArXiv:
Tags:
conversational-qa
License:
Commit
•
9df4273
1
Parent(s):
7315a18
Delete loading script
Browse files
coqa.py
DELETED
@@ -1,91 +0,0 @@
|
|
1 |
-
"""CoQA dataset."""
|
2 |
-
|
3 |
-
|
4 |
-
import json
|
5 |
-
|
6 |
-
import datasets
|
7 |
-
|
8 |
-
|
9 |
-
_HOMEPAGE = "https://stanfordnlp.github.io/coqa/"
|
10 |
-
|
11 |
-
_CITATION = """\
|
12 |
-
@article{reddy-etal-2019-coqa,
|
13 |
-
title = "{C}o{QA}: A Conversational Question Answering Challenge",
|
14 |
-
author = "Reddy, Siva and
|
15 |
-
Chen, Danqi and
|
16 |
-
Manning, Christopher D.",
|
17 |
-
journal = "Transactions of the Association for Computational Linguistics",
|
18 |
-
volume = "7",
|
19 |
-
year = "2019",
|
20 |
-
address = "Cambridge, MA",
|
21 |
-
publisher = "MIT Press",
|
22 |
-
url = "https://aclanthology.org/Q19-1016",
|
23 |
-
doi = "10.1162/tacl_a_00266",
|
24 |
-
pages = "249--266",
|
25 |
-
}
|
26 |
-
"""
|
27 |
-
|
28 |
-
_DESCRIPTION = """\
|
29 |
-
CoQA: A Conversational Question Answering Challenge
|
30 |
-
"""
|
31 |
-
|
32 |
-
_TRAIN_DATA_URL = "https://nlp.stanford.edu/data/coqa/coqa-train-v1.0.json"
|
33 |
-
_DEV_DATA_URL = "https://nlp.stanford.edu/data/coqa/coqa-dev-v1.0.json"
|
34 |
-
|
35 |
-
|
36 |
-
class Coqa(datasets.GeneratorBasedBuilder):
|
37 |
-
|
38 |
-
VERSION = datasets.Version("1.0.0")
|
39 |
-
|
40 |
-
def _info(self):
|
41 |
-
return datasets.DatasetInfo(
|
42 |
-
description=_DESCRIPTION,
|
43 |
-
features=datasets.Features(
|
44 |
-
{
|
45 |
-
"source": datasets.Value("string"),
|
46 |
-
"story": datasets.Value("string"),
|
47 |
-
"questions": datasets.features.Sequence(datasets.Value("string")),
|
48 |
-
"answers": datasets.features.Sequence(
|
49 |
-
{
|
50 |
-
"input_text": datasets.Value("string"),
|
51 |
-
"answer_start": datasets.Value("int32"),
|
52 |
-
"answer_end": datasets.Value("int32"),
|
53 |
-
}
|
54 |
-
),
|
55 |
-
}
|
56 |
-
),
|
57 |
-
homepage=_HOMEPAGE,
|
58 |
-
citation=_CITATION,
|
59 |
-
)
|
60 |
-
|
61 |
-
def _split_generators(self, dl_manager):
|
62 |
-
"""Returns SplitGenerators."""
|
63 |
-
urls_to_download = {"train": _TRAIN_DATA_URL, "dev": _DEV_DATA_URL}
|
64 |
-
downloaded_files = dl_manager.download_and_extract(urls_to_download)
|
65 |
-
|
66 |
-
return [
|
67 |
-
datasets.SplitGenerator(
|
68 |
-
name=datasets.Split.TRAIN, gen_kwargs={"filepath": downloaded_files["train"], "split": "train"}
|
69 |
-
),
|
70 |
-
datasets.SplitGenerator(
|
71 |
-
name=datasets.Split.VALIDATION, gen_kwargs={"filepath": downloaded_files["dev"], "split": "validation"}
|
72 |
-
),
|
73 |
-
]
|
74 |
-
|
75 |
-
def _generate_examples(self, filepath, split):
|
76 |
-
"""Yields examples."""
|
77 |
-
with open(filepath, encoding="utf-8") as f:
|
78 |
-
data = json.load(f)
|
79 |
-
for row in data["data"]:
|
80 |
-
questions = [question["input_text"] for question in row["questions"]]
|
81 |
-
story = row["story"]
|
82 |
-
source = row["source"]
|
83 |
-
answers_start = [answer["span_start"] for answer in row["answers"]]
|
84 |
-
answers_end = [answer["span_end"] for answer in row["answers"]]
|
85 |
-
answers = [answer["input_text"] for answer in row["answers"]]
|
86 |
-
yield row["id"], {
|
87 |
-
"source": source,
|
88 |
-
"story": story,
|
89 |
-
"questions": questions,
|
90 |
-
"answers": {"input_text": answers, "answer_start": answers_start, "answer_end": answers_end},
|
91 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|