Giguru Scheuer commited on
Commit
8205a6a
1 Parent(s): e5b5d86

Added datasetinfo

Browse files
Files changed (1) hide show
  1. canard_quretec.py +161 -0
canard_quretec.py ADDED
@@ -0,0 +1,161 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ # coding=utf-8
2
+ # Copyright 2020 The HuggingFace Datasets Authors and the current dataset script contributor.
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
+
17
+ import csv
18
+ import json
19
+ import os
20
+
21
+ import datasets
22
+
23
+
24
+ # TODO: Add BibTeX citation
25
+ # Find for instance the citation on arxiv or on the dataset repo/website
26
+ _CITATION = """\
27
+ @inproceedings{Elgohary:Peskov:Boyd-Graber-2019,
28
+ Title = {Can You Unpack That? Learning to Rewrite Questions-in-Context},
29
+ Author = {Ahmed Elgohary and Denis Peskov and Jordan Boyd-Graber},
30
+ Booktitle = {Empirical Methods in Natural Language Processing},
31
+ Year = {2019}
32
+ }
33
+ """
34
+
35
+ # You can copy an official description
36
+ _DESCRIPTION = """\
37
+ CANARD has been preprocessed by Voskarides et al. to train and evaluate their Query Resolution Term Classification
38
+ model (QuReTeC).
39
+
40
+ CANARD is a dataset for question-in-context rewriting that consists of questions each given in a dialog context
41
+ together with a context-independent rewriting of the question. The context of each question is the dialog utterences
42
+ that precede the question. CANARD can be used to evaluate question rewriting models that handle important linguistic
43
+ phenomena such as coreference and ellipsis resolution.
44
+ """
45
+
46
+ _HOMEPAGE = "https://sites.google.com/view/qanta/projects/canard"
47
+
48
+ _LICENSE = "CC BY-SA 4.0"
49
+
50
+ # The HuggingFace dataset library don't host the datasets but only point to the original files
51
+ # This can be an arbitrary nested dict/list of URLs (see below in `_split_generators` method)
52
+ _URLs = {
53
+ 'voskarides': "https://drive.google.com/drive/folders/1e3s-V6VQqOKHrmn_kBStNsV0gGHPeJVf",
54
+ }
55
+
56
+
57
+ class CanardQuretec(datasets.GeneratorBasedBuilder):
58
+ """
59
+ Voskarides et al. have preprocessed CANARD in different ways depending on their experiment.
60
+ """
61
+
62
+ VERSION = datasets.Version("1.0.0")
63
+
64
+ # This is an example of a dataset with multiple configurations.
65
+ # If you don't want/need to define several sub-sets in your dataset,
66
+ # just remove the BUILDER_CONFIG_CLASS and the BUILDER_CONFIGS attributes.
67
+
68
+ # If you need to make complex sub-parts in the datasets with configurable options
69
+ # You can create your own builder configuration class to store attribute, inheriting from datasets.BuilderConfig
70
+ # BUILDER_CONFIG_CLASS = MyBuilderConfig
71
+
72
+ # You will be able to load one or the other configurations in the following list with
73
+ # data = datasets.load_dataset('my_dataset', 'first_domain')
74
+ # data = datasets.load_dataset('my_dataset', 'second_domain')
75
+ BUILDER_CONFIGS = [
76
+ datasets.BuilderConfig(name="gold_supervision", version=VERSION, description="Was used for training quretec with gold supervision"),
77
+ # datasets.BuilderConfig(name="original_all", version=VERSION, description="Was used for creating dataset statistics"),
78
+ ]
79
+
80
+ # It's not mandatory to have a default configuration. Just use one if it make sense.
81
+ DEFAULT_CONFIG_NAME = "gold_supervision"
82
+
83
+ def _info(self):
84
+ # This is the name of the configuration selected in BUILDER_CONFIGS above
85
+ # if self.config.name == "gold_supervision" or self.config.name == "original_all":
86
+ features = datasets.Features(
87
+ {
88
+ "id": datasets.Value("string"),
89
+ "prev_questions": datasets.Value("string"),
90
+ "cur_question": datasets.Value("string"),
91
+ "gold_terms": datasets.features.Sequence(feature=datasets.Value('string')),
92
+ "semantic_terms": datasets.features.Sequence(feature=datasets.Value('string')),
93
+ "overlapping_terms": datasets.features.Sequence(feature=datasets.Value('string')),
94
+ "answer_text_with_window": datasets.Value("string"),
95
+ "answer_text": datasets.Value("string"),
96
+ "bert_ner_overlap": datasets.Array2D(shape=(2,), dtype="string")
97
+ }
98
+ )
99
+ return datasets.DatasetInfo(
100
+ # This is the description that will appear on the datasets page.
101
+ description=_DESCRIPTION,
102
+ # This defines the different columns of the dataset and their types
103
+ features=features, # Here we define them above because they are different between the two configurations
104
+ # If there's a common (input, target) tuple from the features,
105
+ # specify them here. They'll be used if as_supervised=True in
106
+ # builder.as_dataset.
107
+ supervised_keys=None,
108
+ # Homepage of the dataset for documentation
109
+ homepage=_HOMEPAGE,
110
+ # License for the dataset if available
111
+ license=_LICENSE,
112
+ # Citation for the dataset
113
+ citation=_CITATION,
114
+ )
115
+
116
+ def _split_generators(self, dl_manager):
117
+ """Returns SplitGenerators."""
118
+ # TODO: This method is tasked with downloading/extracting the data and defining the splits depending on the configuration
119
+ # If several configurations are possible (listed in BUILDER_CONFIGS), the configuration selected by the user is in self.config.name
120
+
121
+ # dl_manager is a datasets.download.DownloadManager that can be used to download and extract URLs
122
+ # It can accept any type or nested list/dict and will give back the same structure with the url replaced with path to local files.
123
+ # By default the archives will be extracted and a path to a cached folder where they are extracted is returned instead of the archive
124
+ my_urls = _URLs[self.config.name]
125
+ data_dir = dl_manager.download_and_extract(my_urls)
126
+ return [
127
+ datasets.SplitGenerator(
128
+ name=datasets.Split.TRAIN,
129
+ gen_kwargs={ # These kwargs will be passed to _generate_examples
130
+ "filepath": os.path.join(data_dir, "train_gold_supervision.json"),
131
+ "split": "train",
132
+ },
133
+ ),
134
+ datasets.SplitGenerator(
135
+ name=datasets.Split.TEST,
136
+ gen_kwargs={ # These kwargs will be passed to _generate_examples
137
+ "filepath": os.path.join(data_dir, "test_gold_supervision.json"),
138
+ "split": "test"
139
+ },
140
+ ),
141
+ datasets.SplitGenerator(
142
+ name=datasets.Split.VALIDATION,
143
+ gen_kwargs={ # These kwargs will be passed to _generate_examples
144
+ "filepath": os.path.join(data_dir, "dev_gold_supervision.json"),
145
+ "split": "dev",
146
+ },
147
+ ),
148
+ ]
149
+
150
+ def _generate_examples(
151
+ self, filepath, split # method parameters are unpacked from `gen_kwargs` as given in `_split_generators`
152
+ ):
153
+ """ Yields examples as (key, example) tuples. """
154
+ # This method handles input defined in _split_generators to yield (key, example) tuples from the dataset.
155
+ # The `key` is here for legacy reason (tfds) and is not important in itself.
156
+
157
+ with open(filepath) as f:
158
+ data = json.load(f)
159
+ for id_, row in data:
160
+ # if self.config.name == "first_domain":
161
+ yield id_, row