Upload dpr_raw.py with huggingface_hub
Browse files- dpr_raw.py +105 -0
dpr_raw.py
ADDED
@@ -0,0 +1,105 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
# coding=utf-8
|
2 |
+
# Copyright 2020 The TensorFlow Datasets Authors and the HuggingFace Datasets Authors.
|
3 |
+
#
|
4 |
+
# Licensed under the Apache License, Version 2.0 (the "License");
|
5 |
+
# you may not use this file except in compliance with the License.
|
6 |
+
# You may obtain a copy of the License at
|
7 |
+
#
|
8 |
+
# http://www.apache.org/licenses/LICENSE-2.0
|
9 |
+
#
|
10 |
+
# Unless required by applicable law or agreed to in writing, software
|
11 |
+
# distributed under the License is distributed on an "AS IS" BASIS,
|
12 |
+
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
13 |
+
# See the License for the specific language governing permissions and
|
14 |
+
# limitations under the License.
|
15 |
+
|
16 |
+
# Lint as: python3
|
17 |
+
"""The Definite Pronoun Resolution Dataset."""
|
18 |
+
|
19 |
+
|
20 |
+
import datasets
|
21 |
+
|
22 |
+
|
23 |
+
_CITATION = """\
|
24 |
+
@inproceedings{rahman2012resolving,
|
25 |
+
title={Resolving complex cases of definite pronouns: the winograd schema challenge},
|
26 |
+
author={Rahman, Altaf and Ng, Vincent},
|
27 |
+
booktitle={Proceedings of the 2012 Joint Conference on Empirical Methods in Natural Language Processing and Computational Natural Language Learning},
|
28 |
+
pages={777--789},
|
29 |
+
year={2012},
|
30 |
+
organization={Association for Computational Linguistics}
|
31 |
+
}"""
|
32 |
+
|
33 |
+
_DESCRIPTION = """\
|
34 |
+
Composed by 30 students from one of the author's undergraduate classes. These
|
35 |
+
sentence pairs cover topics ranging from real events (e.g., Iran's plan to
|
36 |
+
attack the Saudi ambassador to the U.S.) to events/characters in movies (e.g.,
|
37 |
+
Batman) and purely imaginary situations, largely reflecting the pop culture as
|
38 |
+
perceived by the American kids born in the early 90s. Each annotated example
|
39 |
+
spans four lines: the first line contains the sentence, the second line contains
|
40 |
+
the target pronoun, the third line contains the two candidate antecedents, and
|
41 |
+
the fourth line contains the correct antecedent. If the target pronoun appears
|
42 |
+
more than once in the sentence, its first occurrence is the one to be resolved.
|
43 |
+
"""
|
44 |
+
|
45 |
+
|
46 |
+
_DATA_URL_PATTERN = "https://s3.amazonaws.com/datasets.huggingface.co/definite_pronoun_resolution/{}.c.txt"
|
47 |
+
|
48 |
+
|
49 |
+
class DefinitePronounResolution(datasets.GeneratorBasedBuilder):
|
50 |
+
"""The Definite Pronoun Resolution Dataset."""
|
51 |
+
|
52 |
+
BUILDER_CONFIGS = [
|
53 |
+
datasets.BuilderConfig(
|
54 |
+
name="plain_text",
|
55 |
+
version=datasets.Version("1.0.0", ""),
|
56 |
+
description="Plain text import of the Definite Pronoun Resolution Dataset.", # pylint: disable=line-too-long
|
57 |
+
)
|
58 |
+
]
|
59 |
+
|
60 |
+
def _info(self):
|
61 |
+
return datasets.DatasetInfo(
|
62 |
+
description=_DESCRIPTION,
|
63 |
+
features=datasets.Features(
|
64 |
+
{
|
65 |
+
"sentence": datasets.Value("string"),
|
66 |
+
"pronoun": datasets.Value("string"),
|
67 |
+
"candidates": datasets.features.Sequence(datasets.Value("string"), length=2),
|
68 |
+
"label": datasets.features.ClassLabel(num_classes=2),
|
69 |
+
}
|
70 |
+
),
|
71 |
+
supervised_keys=("sentence", "label"),
|
72 |
+
homepage="http://www.hlt.utdallas.edu/~vince/data/emnlp12/",
|
73 |
+
citation=_CITATION,
|
74 |
+
)
|
75 |
+
|
76 |
+
def _split_generators(self, dl_manager):
|
77 |
+
files = dl_manager.download_and_extract(
|
78 |
+
{
|
79 |
+
"train": _DATA_URL_PATTERN.format("train"),
|
80 |
+
"test": _DATA_URL_PATTERN.format("test"),
|
81 |
+
}
|
82 |
+
)
|
83 |
+
return [
|
84 |
+
datasets.SplitGenerator(name=datasets.Split.TEST, gen_kwargs={"filepath": files["test"]}),
|
85 |
+
datasets.SplitGenerator(name=datasets.Split.TRAIN, gen_kwargs={"filepath": files["train"]}),
|
86 |
+
]
|
87 |
+
|
88 |
+
def _generate_examples(self, filepath):
|
89 |
+
with open(filepath, encoding="utf-8") as f:
|
90 |
+
line_num = -1
|
91 |
+
while True:
|
92 |
+
line_num += 1
|
93 |
+
sentence = f.readline().strip()
|
94 |
+
pronoun = f.readline().strip()
|
95 |
+
candidates = [c.strip() for c in f.readline().strip().split(",")]
|
96 |
+
correct = f.readline().strip()
|
97 |
+
f.readline()
|
98 |
+
if not sentence:
|
99 |
+
break
|
100 |
+
yield line_num, {
|
101 |
+
"sentence": sentence,
|
102 |
+
"pronoun": pronoun,
|
103 |
+
"candidates": candidates,
|
104 |
+
"label": candidates.index(correct),
|
105 |
+
}
|