drvenabili commited on
Commit
26896f5
1 Parent(s): a1a9f23
Files changed (1) hide show
  1. kubhist2.py +131 -0
kubhist2.py ADDED
@@ -0,0 +1,131 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ # Config file by Simon Hengchen, https://hengchen.net
2
+
3
+ import os
4
+ import datasets
5
+
6
+ logger = datasets.logging.get_logger(__name__)
7
+
8
+ _CITATION = """
9
+ @misc{botenanna,
10
+ title = {"Jag känner en bot, hon heter [MASK]. A BERT for older Swedish, and a more usable dataset for historical newspapers"},
11
+ author = {Simon Hengchen}
12
+ year={2023},
13
+ }
14
+ """
15
+
16
+ _DESCRIPTION = """
17
+ This is a version of the Kubhist 2 dataset created, curated and made available by Språkbanken Text (SBX) at the University of Gothenburg (Sweden) under the CC BY 4.0 license.
18
+ This is a a corpus of OCRed newspapers from Sweden spanning the 1640s to the 1900s.
19
+ The original data is available with many types of annotation in XML at https://spraakbanken.gu.se/en/resources/kubhist2.
20
+ A good description of the data is available in this blog entry by Dana Dannélls: https://spraakbanken.gu.se/blogg/index.php/2019/09/15/the-kubhist-corpus-of-swedish-newspapers/
21
+
22
+ In a nutshell, this hugginface dataset version offers:
23
+ - only the OCRed text
24
+ - available in decadal subsets
25
+
26
+ License is CC BY 4.0 with attribution.
27
+ """
28
+
29
+ _URL = "https://github.com/ChangeIsKey/kubhist2"
30
+
31
+ _URLS = {'1640': './text/1640/1640.txt.gz',
32
+ '1650': './text/1650/1650.txt.gz',
33
+ '1660': './text/1660/1660.txt.gz',
34
+ '1670': './text/1670/1670.txt.gz',
35
+ '1680': './text/1680/1680.txt.gz',
36
+ '1690': './text/1690/1690.txt.gz',
37
+ '1700': './text/1700/1700.txt.gz',
38
+ '1710': './text/1710/1710.txt.gz',
39
+ '1720': './text/1720/1720.txt.gz',
40
+ '1730': './text/1730/1730.txt.gz',
41
+ '1740': './text/1740/1740.txt.gz',
42
+ '1750': './text/1750/1750.txt.gz',
43
+ '1760': './text/1760/1760.txt.gz',
44
+ '1770': './text/1770/1770.txt.gz',
45
+ '1780': './text/1780/1780.txt.gz',
46
+ '1790': './text/1790/1790.txt.gz',
47
+ '1800': './text/1800/1800.txt.gz',
48
+ '1810': './text/1810/1810.txt.gz',
49
+ '1820': './text/1820/1820.txt.gz',
50
+ '1830': './text/1830/1830.txt.gz',
51
+ '1840': './text/1840/1840.txt.gz',
52
+ '1850': './text/1850/1850.txt.gz',
53
+ '1860': './text/1860/1860.txt.gz',
54
+ '1870': './text/1870/1870.txt.gz',
55
+ '1880': './text/1880/1880.txt.gz',
56
+ '1890': './text/1890/1890.txt.gz',
57
+ '1900': './text/1900/1900.txt.gz'
58
+ }
59
+
60
+
61
+ class kubhist2Config(datasets.BuilderConfig):
62
+ """BuilderConfig for kubhist2."""
63
+
64
+ def __init__(self, period="all", **kwargs):
65
+ """Constructs a kubhist2Dataset.
66
+ Args:
67
+ period: can be any key in _URLS, `all` takes all
68
+ **kwargs: keyword arguments forwarded to super.
69
+ """
70
+ if str(period) not in _URLS.keys():
71
+ #logger.warning("No period specified or wrong period, getting everything instead.")
72
+ self.period = "all"
73
+ else:
74
+ self.period = str(period)
75
+ super(kubhist2Config, self).__init__(**kwargs)
76
+
77
+
78
+ class kubhist2(datasets.GeneratorBasedBuilder):
79
+ BUILDER_CONFIG_CLASS = kubhist2Config
80
+
81
+ BUILDER_CONFIGS = []
82
+ for key in _URLS:
83
+ BUILDER_CONFIGS.append(
84
+ kubhist2Config(
85
+ name=key,
86
+ version=datasets.Version("1.0.0", ""),
87
+ description=f"Kubhist2: {key}",
88
+ period=key,
89
+ )
90
+ )
91
+ BUILDER_CONFIGS.append(
92
+ kubhist2Config(
93
+ name="all",
94
+ version=datasets.Version("1.0.0", ""),
95
+ description=f"Kubhist2: all",
96
+ period="all",
97
+ )
98
+ )
99
+ DEFAULT_CONFIG_NAME = "all"
100
+
101
+ def _info(self):
102
+ f = datasets.Features(
103
+ {
104
+ "text": datasets.Value("string")
105
+ }
106
+ )
107
+
108
+ return datasets.DatasetInfo(
109
+ features=f,
110
+ supervised_keys=None,
111
+ homepage="https://github.com/ChangeIsKey",
112
+ citation=_CITATION,
113
+ )
114
+
115
+ def _split_generators(self, dl_manager):
116
+ if self.config.period != "all":
117
+ url = {"train" : _URLS[self.config.period]}
118
+ downloaded_files = dl_manager.download_and_extract(url)
119
+ return [datasets.SplitGenerator(name=datasets.Split.TRAIN, gen_kwargs={"filepath": downloaded_files["train"]})]
120
+
121
+ elif self.config.period == "all":
122
+ url = {"train" : './text/all/all.txt.gz'}
123
+ #print(url)
124
+ downloaded_files = dl_manager.download_and_extract(url)
125
+ return [datasets.SplitGenerator(name=datasets.Split.TRAIN, gen_kwargs={"filepath": downloaded_files["train"]})]
126
+
127
+ def _generate_examples(self, filepath):
128
+ """Yields examples."""
129
+ with open(filepath) as f:
130
+ for i, line in enumerate(f):
131
+ yield i, {"text" : line.rstrip()}