Rodekool commited on
Commit
ef81ebc
1 Parent(s): 0ecceee

Create new file

Browse files
Files changed (1) hide show
  1. ornl20.py +73 -0
ornl20.py ADDED
@@ -0,0 +1,73 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ # coding=utf-8
2
+ # Copyright 2020 The TensorFlow Datasets Authors and the HuggingFace NLP 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
+ """Open Data Rechtspraak dutch topic classification dataset."""
18
+
19
+ from __future__ import absolute_import, division, print_function
20
+
21
+ import csv
22
+
23
+ # import nlp # old loader
24
+ import datasets
25
+ from datasets.tasks import TextClassification
26
+
27
+
28
+ _DESCRIPTION = """\
29
+ still a WIP, Dataset originally comes from Open Data van de Rechtspraak"
30
+ """
31
+
32
+ _TRAIN_DOWNLOAD_URL = "https://huggingface.co/datasets/Rodekool/ornl20/resolve/main/train.csv"
33
+ _TEST_DOWNLOAD_URL = "https://huggingface.co/datasets/Rodekool/ornl20/resolve/main/test.csv"
34
+
35
+
36
+ class ORnl(datasets.GeneratorBasedBuilder):
37
+ """Open Data van de Rechtspraak dutch topic classification dataset."""
38
+
39
+ def _info(self):
40
+ return datasets.DatasetInfo(
41
+ description=_DESCRIPTION,
42
+ features=datasets.Features(
43
+ {
44
+ "text": datasets.Value("string"),
45
+ "label": datasets.features.ClassLabel(names=['Ambtenarenrecht', 'Arbeidsrecht', 'Belastingrecht', 'Omgevingsrecht', 'Personen- en familierecht', 'Socialezekerheidsrecht', 'Verbintenissenrecht', 'Vreemdelingenrecht']),
46
+ }
47
+ ),
48
+ homepage="https://www.rechtspraak.nl/Uitspraken/paginas/open-data.aspx",
49
+ )
50
+
51
+ def _split_generators(self, dl_manager):
52
+ train_path = dl_manager.download_and_extract(_TRAIN_DOWNLOAD_URL)
53
+ test_path = dl_manager.download_and_extract(_TEST_DOWNLOAD_URL)
54
+ return [
55
+ datasets.SplitGenerator(name=datasets.Split.TRAIN, gen_kwargs={"filepath": train_path}),
56
+ datasets.SplitGenerator(name=datasets.Split.TEST, gen_kwargs={"filepath": test_path}),
57
+ ]
58
+
59
+ def _generate_examples(self, filepath):
60
+ """Generate Rechtspraak examples."""
61
+ with open(filepath, encoding="utf-8") as csv_file:
62
+ csv_reader = csv.reader(
63
+ csv_file, quotechar='"', delimiter=",", quoting=csv.QUOTE_ALL, skipinitialspace=True
64
+ )
65
+ for id_, row in enumerate(csv_reader):
66
+ label, title, description = row
67
+
68
+ # Original labels are [1, 2, ... 25, 26] ->
69
+ # Re-map to [0, 1, ... 24, 25].
70
+
71
+ label = int(label) - 1
72
+ text = " ".join((title, description))
73
+ yield id_, {"text": text, "label": label}