Upload becasv2.py
Browse files- becasv2.py +158 -0
becasv2.py
ADDED
@@ -0,0 +1,158 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
from datasets.utils import version
|
2 |
+
"""TODO(squad_es): Add a description here."""
|
3 |
+
|
4 |
+
|
5 |
+
import json
|
6 |
+
|
7 |
+
import datasets
|
8 |
+
|
9 |
+
|
10 |
+
# TODO(squad_es): BibTeX citation
|
11 |
+
_CITATION = """\
|
12 |
+
@article{2016arXiv160605250R,
|
13 |
+
author = {Casimiro Pio , Carrino and Marta R. , Costa-jussa and Jose A. R. , Fonollosa},
|
14 |
+
title = "{Automatic Spanish Translation of the SQuAD Dataset for Multilingual
|
15 |
+
Question Answering}",
|
16 |
+
journal = {arXiv e-prints},
|
17 |
+
year = 2019,
|
18 |
+
eid = {arXiv:1912.05200v1},
|
19 |
+
pages = {arXiv:1912.05200v1},
|
20 |
+
archivePrefix = {arXiv},
|
21 |
+
eprint = {1912.05200v2},
|
22 |
+
}
|
23 |
+
"""
|
24 |
+
|
25 |
+
# TODO(squad_es_v1):
|
26 |
+
_DESCRIPTION = """\
|
27 |
+
automatic translation of the Stanford Question Answering Dataset (SQuAD) v2 into Spanish
|
28 |
+
"""
|
29 |
+
|
30 |
+
_URL = "https://raw.githubusercontent.com/EvelynQuevedo/becas/main/"
|
31 |
+
print(_URL)
|
32 |
+
_URLS_V1 = {
|
33 |
+
"train": _URL + "datos/data.json"
|
34 |
+
|
35 |
+
#"dev": _URL + "SQuAD-es-v1.1/dev-v1.1-es.json",
|
36 |
+
}
|
37 |
+
print(_URLS_V1)
|
38 |
+
#_URLS_V2 = {
|
39 |
+
# "train": _URL + "SQuAD-es-v2.0/train-v2.0-es.json",
|
40 |
+
#"dev": _URL + "SQuAD-es-v2.0/dev-v2.0-es.json",
|
41 |
+
#}
|
42 |
+
|
43 |
+
|
44 |
+
class SquadEsConfig(datasets.BuilderConfig):
|
45 |
+
"""BuilderConfig for SQUADEsV2."""
|
46 |
+
|
47 |
+
def __init__(self, **kwargs):
|
48 |
+
"""BuilderConfig for SQUADEsV2.
|
49 |
+
Args:
|
50 |
+
**kwargs: keyword arguments forwarded to super.
|
51 |
+
"""
|
52 |
+
super(SquadEsConfig, self).__init__(**kwargs)
|
53 |
+
|
54 |
+
|
55 |
+
class SquadEs(datasets.GeneratorBasedBuilder):
|
56 |
+
"""TODO(squad_es): Short description of my dataset."""
|
57 |
+
|
58 |
+
# TODO(squad_es): Set up version.
|
59 |
+
VERSION = datasets.Version("0.1.0")
|
60 |
+
BUILDER_CONFIGS = [
|
61 |
+
SquadEsConfig(
|
62 |
+
name="v1.1.0",
|
63 |
+
version=datasets.Version("1.1.0", ""),
|
64 |
+
description="Plain text Spanish squad version 1",
|
65 |
+
),
|
66 |
+
#SquadEsConfig(
|
67 |
+
# name="v2.0.0",
|
68 |
+
# version=datasets.Version("2.0.0", ""),
|
69 |
+
# description="Plain text Spanish squad version 2",
|
70 |
+
#),
|
71 |
+
]
|
72 |
+
|
73 |
+
def _info(self):
|
74 |
+
# TODO(squad_es): Specifies the datasets.DatasetInfo object
|
75 |
+
return datasets.DatasetInfo(
|
76 |
+
# This is the description that will appear on the datasets page.
|
77 |
+
description=_DESCRIPTION,
|
78 |
+
# datasets.features.FeatureConnectors
|
79 |
+
features=datasets.Features(
|
80 |
+
{
|
81 |
+
|
82 |
+
# These are the features of your dataset like images, labels ...
|
83 |
+
"id": datasets.Value("string"),
|
84 |
+
"title": datasets.Value("string"),
|
85 |
+
"context": datasets.Value("string"),
|
86 |
+
"question": datasets.Value("string"),
|
87 |
+
"answers": datasets.features.Sequence(
|
88 |
+
{
|
89 |
+
"text": datasets.Value("string"),
|
90 |
+
"answer_start": datasets.Value("int32"),
|
91 |
+
}
|
92 |
+
),
|
93 |
+
}
|
94 |
+
),
|
95 |
+
# If there's a common (input, target) tuple from the features,
|
96 |
+
# specify them here. They'll be used if as_supervised=True in
|
97 |
+
# builder.as_dataset.
|
98 |
+
supervised_keys=None,
|
99 |
+
# Homepage of the dataset for documentation
|
100 |
+
homepage="https://github.com/EvelynQuevedo/becas",
|
101 |
+
|
102 |
+
citation=_CITATION,
|
103 |
+
)
|
104 |
+
|
105 |
+
def _split_generators(self, dl_manager):
|
106 |
+
"""Returns SplitGenerators."""
|
107 |
+
# TODO(squad_es): Downloads the data and defines the splits
|
108 |
+
# dl_manager is a datasets.download.DownloadManager that can be used to
|
109 |
+
|
110 |
+
# download and extract URLs
|
111 |
+
if self.config.name == "v1.1.0":
|
112 |
+
dl_dir = dl_manager.download_and_extract(_URLS_V1)
|
113 |
+
print(dl_dir)
|
114 |
+
# elif self.config.name == "v2.0.0":
|
115 |
+
# dl_dir = dl_manager.download_and_extract(_URLS_V2)
|
116 |
+
else:
|
117 |
+
raise Exception("version does not match any existing one")
|
118 |
+
return [
|
119 |
+
datasets.SplitGenerator(
|
120 |
+
name=datasets.Split.TRAIN,
|
121 |
+
# These kwargs will be passed to _generate_examples
|
122 |
+
gen_kwargs={"filepath": dl_dir["train"]},
|
123 |
+
),
|
124 |
+
# datasets.SplitGenerator(
|
125 |
+
# name=datasets.Split.VALIDATION,
|
126 |
+
# These kwargs will be passed to _generate_examples
|
127 |
+
# gen_kwargs={"filepath": dl_dir["dev"]},
|
128 |
+
#),
|
129 |
+
]
|
130 |
+
|
131 |
+
def _generate_examples(self, filepath):
|
132 |
+
"""Yields examples."""
|
133 |
+
# TODO(squad_es): Yields (key, example) tuples from the dataset
|
134 |
+
with open(filepath, encoding="utf-8") as f:
|
135 |
+
data = json.load(f)
|
136 |
+
for example in data["data"]:
|
137 |
+
title = example.get("title", "").strip()
|
138 |
+
for paragraph in example["paragraphs"]:
|
139 |
+
context = paragraph["context"].strip()
|
140 |
+
for qa in paragraph["qas"]:
|
141 |
+
question = qa["question"].strip()
|
142 |
+
id_ = qa["id"]
|
143 |
+
|
144 |
+
answer_starts = [answer["answer_start"] for answer in qa["answers"]]
|
145 |
+
answers = [answer["text"].strip() for answer in qa["answers"]]
|
146 |
+
|
147 |
+
yield id_, {
|
148 |
+
"title": title,
|
149 |
+
"context": context,
|
150 |
+
"question": question,
|
151 |
+
"id": id_,
|
152 |
+
"answers": {
|
153 |
+
"answer_start": answer_starts,
|
154 |
+
"text": answers,
|
155 |
+
|
156 |
+
},
|
157 |
+
|
158 |
+
}
|