Datasets:
File size: 5,087 Bytes
a9293ec |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 |
import os
import re
import xml.etree.ElementTree as ET
import datasets
XML_NAMESPACE = "{http://www.w3.org/XML/1998/namespace}"
DEFAULT_LANG = "default"
_CITATION = """\
@misc{janes_preklop,
title = {Tweet code-switching corpus Janes-Preklop 1.0},
author = {Reher, {\v S}pela and Erjavec, Toma{\v z} and Fi{\v s}er, Darja},
url = {http://hdl.handle.net/11356/1154},
note = {Slovenian language resource repository {CLARIN}.{SI}},
copyright = {Creative Commons - Attribution-{ShareAlike} 4.0 International ({CC} {BY}-{SA} 4.0)},
issn = {2820-4042},
year = {2017}
}
"""
_DESCRIPTION = """\
Janes-Preklop is a corpus of Slovene tweets that is manually annotated for code-switching (the use of words from two
or more languages within one sentence or utterance), according to the supplied typology.
"""
_HOMEPAGE = "http://nl.ijs.si/janes/"
_LICENSE = "Creative Commons - Attribution-ShareAlike 4.0 International (CC BY-SA 4.0)"
_URLS = {
"janes_preklop_tei": "https://www.clarin.si/repository/xmlui/bitstream/handle/11356/1154/Janes-Preklop.TEI.zip"
}
def namespace(element):
# https://stackoverflow.com/a/12946675
m = re.match(r'\{.*\}', element.tag)
return m.group(0) if m else ''
def word_info(wordlike_tag, _namespace):
if wordlike_tag.tag == f"{_namespace}c":
return None, None
elif wordlike_tag.tag in {f"{_namespace}w", f"{_namespace}pc"}:
_word = wordlike_tag.text.strip()
return [_word], None
words, lang = [], []
if wordlike_tag.tag == f"{_namespace}choice":
# Ignore <orig>, keep <reg>ularized text
reg_tag = wordlike_tag.find(f"{_namespace}reg")
for _child in reg_tag:
_child_words, _child_langs = word_info(_child, _namespace)
if _child_words is None:
continue
words.extend(_child_words)
if _child_langs is None:
_child_langs = [DEFAULT_LANG for _ in range(len(_child_words))]
lang.extend(_child_langs)
return words, lang
elif wordlike_tag.tag == f"{_namespace}seg":
codeswitch_lang = wordlike_tag.attrib.get(f"{XML_NAMESPACE}lang", DEFAULT_LANG)
_seg_words, _seg_langs = [], []
for _child in wordlike_tag:
_child_words, _child_langs = word_info(_child, _namespace)
if _child_words is None:
continue
_seg_words.extend(_child_words)
if _child_langs is None:
_child_langs = [codeswitch_lang for _ in range(len(_child_words))]
_seg_langs.extend(_child_langs)
if codeswitch_lang != DEFAULT_LANG:
_seg_langs[0] = f"B-{_seg_langs[0]}"
for _i in range(1, len(_seg_words)):
_seg_langs[_i] = f"I-{_seg_langs[_i]}"
words.extend(_seg_words)
lang.extend(_seg_langs)
return words, lang
class JanesPreklop(datasets.GeneratorBasedBuilder):
"""Code-switched Slovene tweet dataset. """
VERSION = datasets.Version("1.0.0")
def _info(self):
features = datasets.Features(
{
"id": datasets.Value("string"),
"words": datasets.Sequence(datasets.Value("string")),
"language": datasets.Sequence(datasets.Value("string"))
}
)
return datasets.DatasetInfo(
description=_DESCRIPTION,
features=features,
homepage=_HOMEPAGE,
license=_LICENSE,
citation=_CITATION,
)
def _split_generators(self, dl_manager):
urls = _URLS["janes_preklop_tei"]
data_dir = dl_manager.download_and_extract(urls)
return [
datasets.SplitGenerator(
name=datasets.Split.TRAIN,
gen_kwargs={"file_path": os.path.join(data_dir, "Janes-Preklop.TEI", "janes.preklop.body.xml")},
)
]
def _generate_examples(self, file_path):
curr_doc = ET.parse(file_path)
root = curr_doc.getroot()
NAMESPACE = namespace(root)
idx_ex = 0
for curr_ex in root.iterfind(f"{NAMESPACE}ab"):
ex_words, ex_langs = [], []
for child_tag in curr_ex:
if child_tag.tag not in {f"{NAMESPACE}s", f"{NAMESPACE}c"}:
continue
if child_tag.tag == f"{NAMESPACE}c":
continue
# Iterate over elements of a <s>entence
for word_or_seg_tag in child_tag:
_words, _langs = word_info(word_or_seg_tag, NAMESPACE)
if _words is None:
continue
if _langs is None:
_langs = [DEFAULT_LANG for _ in range(len(_words))]
ex_words.extend(_words)
ex_langs.extend(_langs)
yield idx_ex, {
"id": curr_ex.attrib[f"{XML_NAMESPACE}id"],
"words": ex_words,
"language": ex_langs
}
idx_ex += 1
|