Datasets:

Languages:
English
Size Categories:
100K<n<1M
ArXiv:
Tags:
License:
yuvalr commited on
Commit
ee2b54f
1 Parent(s): 99ea19b

Upload 13 files

Browse files
.gitattributes CHANGED
@@ -53,3 +53,7 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
53
  *.jpg filter=lfs diff=lfs merge=lfs -text
54
  *.jpeg filter=lfs diff=lfs merge=lfs -text
55
  *.webp filter=lfs diff=lfs merge=lfs -text
 
 
 
 
 
53
  *.jpg filter=lfs diff=lfs merge=lfs -text
54
  *.jpeg filter=lfs diff=lfs merge=lfs -text
55
  *.webp filter=lfs diff=lfs merge=lfs -text
56
+ minority_examples/train.anti_biased.jsonl filter=lfs diff=lfs merge=lfs -text
57
+ minority_examples/train.biased.jsonl filter=lfs diff=lfs merge=lfs -text
58
+ partial_input/train.anti_biased.jsonl filter=lfs diff=lfs merge=lfs -text
59
+ partial_input/train.biased.jsonl filter=lfs diff=lfs merge=lfs -text
anli.py ADDED
@@ -0,0 +1,137 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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 Adversarial NLI Corpus."""
18
+
19
+
20
+ import json
21
+ import os
22
+
23
+ import datasets
24
+
25
+
26
+ _CITATION = """\
27
+ @InProceedings{nie2019adversarial,
28
+ title={Adversarial NLI: A New Benchmark for Natural Language Understanding},
29
+ author={Nie, Yixin
30
+ and Williams, Adina
31
+ and Dinan, Emily
32
+ and Bansal, Mohit
33
+ and Weston, Jason
34
+ and Kiela, Douwe},
35
+ booktitle = "Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics",
36
+ year = "2020",
37
+ publisher = "Association for Computational Linguistics",
38
+ }
39
+ """
40
+
41
+ _DESCRIPTION = """\
42
+ The Adversarial Natural Language Inference (ANLI) is a new large-scale NLI benchmark dataset,
43
+ The dataset is collected via an iterative, adversarial human-and-model-in-the-loop procedure.
44
+ ANLI is much more difficult than its predecessors including SNLI and MNLI.
45
+ It contains three rounds. Each round has train/dev/test splits.
46
+ """
47
+
48
+ stdnli_label = {
49
+ "e": "entailment",
50
+ "n": "neutral",
51
+ "c": "contradiction",
52
+ }
53
+
54
+
55
+ class ANLIConfig(datasets.BuilderConfig):
56
+ """BuilderConfig for ANLI."""
57
+
58
+ def __init__(self, **kwargs):
59
+ """BuilderConfig for ANLI.
60
+
61
+ Args:
62
+ .
63
+ **kwargs: keyword arguments forwarded to super.
64
+ """
65
+ super(ANLIConfig, self).__init__(version=datasets.Version("0.1.0", ""), **kwargs)
66
+
67
+
68
+ class ANLI(datasets.GeneratorBasedBuilder):
69
+ """ANLI: The ANLI Dataset."""
70
+
71
+ BUILDER_CONFIGS = [
72
+ ANLIConfig(
73
+ name=bias_amplified_splits_type,
74
+ description="",
75
+ ) for bias_amplified_splits_type in ["minority_examples", "partial_input"]
76
+ ]
77
+
78
+ def _info(self):
79
+ return datasets.DatasetInfo(
80
+ description=_DESCRIPTION,
81
+ features=datasets.Features(
82
+ {
83
+ "round": datasets.Value("string"),
84
+ "uid": datasets.Value("string"),
85
+ "premise": datasets.Value("string"),
86
+ "hypothesis": datasets.Value("string"),
87
+ "label": datasets.features.ClassLabel(names=["entailment", "neutral", "contradiction"]),
88
+ "reason": datasets.Value("string"),
89
+ }
90
+ ),
91
+ # No default supervised_keys (as we have to pass both premise
92
+ # and hypothesis as input).
93
+ supervised_keys=None,
94
+ homepage="https://github.com/facebookresearch/anli/",
95
+ citation=_CITATION,
96
+ )
97
+
98
+ def _vocab_text_gen(self, filepath):
99
+ for _, ex in self._generate_examples(filepath):
100
+ yield " ".join([ex["premise"], ex["hypothesis"]])
101
+
102
+ def _split_generators(self, dl_manager):
103
+ return [
104
+ datasets.SplitGenerator(name="train.biased", gen_kwargs={"filepath": dl_manager.download(os.path.join(self.config.name, "train.biased.jsonl"))}),
105
+ datasets.SplitGenerator(name="train.anti-biased", gen_kwargs={"filepath": dl_manager.download(os.path.join(self.config.name, "train.anti-biased.jsonl"))}),
106
+ datasets.SplitGenerator(name="validation.biased", gen_kwargs={"filepath": dl_manager.download(os.path.join(self.config.name, "validation.biased.jsonl"))}),
107
+ datasets.SplitGenerator(name="validation.anti-biased", gen_kwargs={"filepath": dl_manager.download(os.path.join(self.config.name, "validation.anti-biased.jsonl"))}),
108
+ datasets.SplitGenerator(name="test.biased", gen_kwargs={"filepath": dl_manager.download(os.path.join(self.config.name, "test.biased.jsonl"))}),
109
+ datasets.SplitGenerator(name="test.anti-biased", gen_kwargs={"filepath": dl_manager.download(os.path.join(self.config.name, "test.anti-biased.jsonl"))})
110
+ ]
111
+
112
+ def _generate_examples(self, filepath):
113
+ """Generate examples.
114
+
115
+ Args:
116
+ filepath: a string
117
+
118
+ Yields:
119
+ dictionaries containing "premise", "hypothesis" and "label" strings
120
+ """
121
+ for idx, line in enumerate(open(filepath, "rb")):
122
+ if line is not None:
123
+ line = line.strip().decode("utf-8")
124
+ item = json.loads(line)
125
+
126
+ reason_text = ""
127
+ if "reason" in item:
128
+ reason_text = item["reason"]
129
+
130
+ yield f'{item["round"]}-{item["uid"]}', {
131
+ "round": item["round"],
132
+ "uid": item["uid"],
133
+ "premise": item["context"],
134
+ "hypothesis": item["hypothesis"],
135
+ "label": stdnli_label[item["label"]],
136
+ "reason": reason_text,
137
+ }
minority_examples/test.anti_biased.jsonl ADDED
The diff for this file is too large to render. See raw diff
 
minority_examples/test.biased.jsonl ADDED
The diff for this file is too large to render. See raw diff
 
minority_examples/train.anti_biased.jsonl ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:22a8c58c3410478a97dd15ce7e68bca4348698f23c142bd9c10256b9a25ca9fe
3
+ size 14636850
minority_examples/train.biased.jsonl ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6f794c9947a7ebab423662534e0ad23e2bd4c4c4a4eb934ebad64647b2c42e7f
3
+ size 67780649
minority_examples/validation.anti_biased.jsonl ADDED
The diff for this file is too large to render. See raw diff
 
minority_examples/validation.biased.jsonl ADDED
The diff for this file is too large to render. See raw diff
 
partial_input/test.anti_biased.jsonl ADDED
The diff for this file is too large to render. See raw diff
 
partial_input/test.biased.jsonl ADDED
The diff for this file is too large to render. See raw diff
 
partial_input/train.anti_biased.jsonl ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e6e671d8dde3bf6df3925e8e65b694060f95e1dd061cbc5d46a526b1202156ff
3
+ size 15141511
partial_input/train.biased.jsonl ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:35cfb7adf1b8cbbac62fc10bcb5ad6f65ae467c6703a226a533a378cb8a65753
3
+ size 67275988
partial_input/validation.anti_biased.jsonl ADDED
The diff for this file is too large to render. See raw diff
 
partial_input/validation.biased.jsonl ADDED
The diff for this file is too large to render. See raw diff