yhavinga commited on
Commit
c574d0a
1 Parent(s): e6fc625

Add dataset script

Browse files
test.jsonl.gz → dataset/test.jsonl.gz RENAMED
File without changes
train.jsonl.gz → dataset/train.jsonl.gz RENAMED
File without changes
unsupervised.jsonl.gz → dataset/unsupervised.jsonl.gz RENAMED
File without changes
imdb_dutch.py ADDED
@@ -0,0 +1,127 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
+ """IMDB movie reviews dataset."""
18
+ import gzip
19
+ import json
20
+
21
+ import datasets
22
+ from datasets.tasks import TextClassification
23
+
24
+
25
+ _DESCRIPTION = """\
26
+ Large Movie Review Dataset translated to Dutch.
27
+
28
+ This is a dataset for binary sentiment classification containing substantially \
29
+ more data than previous benchmark datasets. We provide a set of 24,992 highly \
30
+ polar movie reviews for training, and 24,992 for testing. There is additional \
31
+ unlabeled data for use as well.\
32
+ """
33
+
34
+ _CITATION = """\
35
+ @InProceedings{maas-EtAl:2011:ACL-HLT2011,
36
+ author = {Maas, Andrew L. and Daly, Raymond E. and Pham, Peter T. and Huang, Dan and Ng, Andrew Y. and Potts, Christopher},
37
+ title = {Learning Word Vectors for Sentiment Analysis},
38
+ booktitle = {Proceedings of the 49th Annual Meeting of the Association for Computational Linguistics: Human Language Technologies},
39
+ month = {June},
40
+ year = {2011},
41
+ address = {Portland, Oregon, USA},
42
+ publisher = {Association for Computational Linguistics},
43
+ pages = {142--150},
44
+ url = {http://www.aclweb.org/anthology/P11-1015}
45
+ }
46
+ """
47
+
48
+ _DOWNLOAD_URL = "https://huggingface.co/datasets/yhavinga/imdb_dutch/resolve/main/dataset/{split}.gz"
49
+
50
+
51
+ class IMDBReviewsConfig(datasets.BuilderConfig):
52
+ """BuilderConfig for IMDBReviews."""
53
+
54
+ def __init__(self, **kwargs):
55
+ """BuilderConfig for IMDBReviews.
56
+
57
+ Args:
58
+ **kwargs: keyword arguments forwarded to super.
59
+ """
60
+ super(IMDBReviewsConfig, self).__init__(
61
+ version=datasets.Version("1.0.0", ""), **kwargs
62
+ )
63
+
64
+
65
+ class Imdb(datasets.GeneratorBasedBuilder):
66
+ """IMDB movie reviews dataset."""
67
+
68
+ BUILDER_CONFIGS = [
69
+ IMDBReviewsConfig(
70
+ name="plain_text",
71
+ description="Plain text",
72
+ )
73
+ ]
74
+
75
+ def _info(self):
76
+ return datasets.DatasetInfo(
77
+ description=_DESCRIPTION,
78
+ features=datasets.Features(
79
+ {
80
+ "text": datasets.Value("string"),
81
+ "text_en": datasets.Value("string"),
82
+ "label": datasets.features.ClassLabel(names=["neg", "pos"]),
83
+ }
84
+ ),
85
+ supervised_keys=None,
86
+ homepage="http://ai.stanford.edu/~amaas/data/sentiment/",
87
+ citation=_CITATION,
88
+ task_templates=[
89
+ TextClassification(text_column="text", label_column="label")
90
+ ],
91
+ )
92
+
93
+ def _split_generators(self, dl_manager):
94
+ return [
95
+ datasets.SplitGenerator(
96
+ name=datasets.Split.TRAIN,
97
+ gen_kwargs={
98
+ "files": dl_manager.download(_DOWNLOAD_URL.format(split="train")),
99
+ "split": "train",
100
+ },
101
+ ),
102
+ datasets.SplitGenerator(
103
+ name=datasets.Split.TEST,
104
+ gen_kwargs={
105
+ "files": dl_manager.download(_DOWNLOAD_URL.format(split="test")),
106
+ "split": "test",
107
+ },
108
+ ),
109
+ datasets.SplitGenerator(
110
+ name=datasets.Split("unsupervised"),
111
+ gen_kwargs={
112
+ "files": dl_manager.download(
113
+ _DOWNLOAD_URL.format(split="unsupervised")
114
+ ),
115
+ "split": "unsupervised",
116
+ "labeled": False,
117
+ },
118
+ ),
119
+ ]
120
+
121
+ def _generate_examples(self, files, split, labeled=True):
122
+ """Generate aclImdb examples."""
123
+ for filepath in files:
124
+ with gzip.open(open(filepath, "rb"), "rt", encoding="utf-8") as f:
125
+ for _, line in enumerate(f):
126
+ example = json.loads(line)
127
+ yield _, example