AnnaSallesRius
commited on
Commit
•
21060c8
1
Parent(s):
c076f5a
Upload 4 files
Browse files- OLD/CoQCat.py +80 -0
- OLD/dev.json +0 -0
- OLD/test.json +0 -0
- OLD/train.json +3 -0
OLD/CoQCat.py
ADDED
@@ -0,0 +1,80 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
"""CoQA dataset."""
|
2 |
+
|
3 |
+
|
4 |
+
import json
|
5 |
+
|
6 |
+
import datasets
|
7 |
+
|
8 |
+
|
9 |
+
_HOMEPAGE = ""
|
10 |
+
|
11 |
+
_CITATION = """\
|
12 |
+
|
13 |
+
"""
|
14 |
+
|
15 |
+
_DESCRIPTION = """\
|
16 |
+
CoQCat - Conversational Question Answering in Catalan
|
17 |
+
"""
|
18 |
+
|
19 |
+
_TRAIN_FILE = "train.json"
|
20 |
+
_DEV_FILE = "dev.json"
|
21 |
+
_TEST_FILE = "test.json"
|
22 |
+
|
23 |
+
class Coqa(datasets.GeneratorBasedBuilder):
|
24 |
+
|
25 |
+
VERSION = datasets.Version("1.0.0")
|
26 |
+
|
27 |
+
def _info(self):
|
28 |
+
return datasets.DatasetInfo(
|
29 |
+
description=_DESCRIPTION,
|
30 |
+
features=datasets.Features(
|
31 |
+
{
|
32 |
+
"source": datasets.Value("string"),
|
33 |
+
"story": datasets.Value("string"),
|
34 |
+
"questions": datasets.features.Sequence(datasets.Value("string")),
|
35 |
+
"answers": datasets.features.Sequence(
|
36 |
+
{
|
37 |
+
"input_text": datasets.Value("string"),
|
38 |
+
"answer_start": datasets.Value("int32"),
|
39 |
+
"answer_end": datasets.Value("int32"),
|
40 |
+
}
|
41 |
+
),
|
42 |
+
}
|
43 |
+
),
|
44 |
+
homepage=_HOMEPAGE,
|
45 |
+
citation=_CITATION,
|
46 |
+
)
|
47 |
+
|
48 |
+
|
49 |
+
def _split_generators(self, dl_manager):
|
50 |
+
"""Returns SplitGenerators."""
|
51 |
+
urls_to_download = {
|
52 |
+
"train": f"{_TRAIN_FILE}",
|
53 |
+
"dev": f"{_DEV_FILE}",
|
54 |
+
"test": f"{_TEST_FILE}",
|
55 |
+
}
|
56 |
+
downloaded_files = dl_manager.download_and_extract(urls_to_download)
|
57 |
+
|
58 |
+
return [
|
59 |
+
datasets.SplitGenerator(name=datasets.Split.TRAIN, gen_kwargs={"filepath": downloaded_files["train"], "split": "train"}),
|
60 |
+
datasets.SplitGenerator(name=datasets.Split.VALIDATION, gen_kwargs={"filepath": downloaded_files["dev"], "split": "validation"}),
|
61 |
+
datasets.SplitGenerator(name=datasets.Split.TEST, gen_kwargs={"filepath": downloaded_files["test"], "split": "test"}),
|
62 |
+
]
|
63 |
+
|
64 |
+
def _generate_examples(self, filepath, split):
|
65 |
+
"""Yields examples."""
|
66 |
+
with open(filepath, encoding="utf-8") as f:
|
67 |
+
data = json.load(f)
|
68 |
+
for row in data["data"]:
|
69 |
+
questions = [question["input_text"] for question in row["questions"]]
|
70 |
+
story = row["story"]
|
71 |
+
source = row["source"]
|
72 |
+
answers_start = [answer["span_start"] for answer in row["answers"]]
|
73 |
+
answers_end = [answer["span_end"] for answer in row["answers"]]
|
74 |
+
answers = [answer["input_text"] for answer in row["answers"]]
|
75 |
+
yield row["id"], {
|
76 |
+
"source": source,
|
77 |
+
"story": story,
|
78 |
+
"questions": questions,
|
79 |
+
"answers": {"input_text": answers, "answer_start": answers_start, "answer_end": answers_end},
|
80 |
+
}
|
OLD/dev.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|
OLD/test.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|
OLD/train.json
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:0e91c9e753e4a747b8aae80f2402df41d2cfaef2add7296e314f57b1ec7e787c
|
3 |
+
size 39968225
|