Datasets:
ShkalikovOleh
commited on
Commit
•
3e3575b
1
Parent(s):
ef1dbcb
Convert dataset to Parquet (#9)
Browse files- Convert dataset to Parquet (549279667653b12a7377001295615ea6cec3d369)
- Delete loading script (c8c9ca6ed35d48368a8e835afa15626ae6bbfd9e)
- README.md +38 -8
- en/test-00000-of-00001.parquet +3 -0
- europarl-ner.py +0 -137
README.md
CHANGED
@@ -1,17 +1,47 @@
|
|
1 |
---
|
2 |
-
license: other
|
3 |
-
license_name: europarl-custom
|
4 |
-
license_link: https://www.statmt.org/europarl/
|
5 |
-
task_categories:
|
6 |
-
- token-classification
|
7 |
language:
|
8 |
- en
|
9 |
- it
|
10 |
- de
|
11 |
- es
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
12 |
tags:
|
13 |
- NER
|
14 |
- Europarl
|
15 |
-
|
16 |
-
|
17 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
---
|
|
|
|
|
|
|
|
|
|
|
2 |
language:
|
3 |
- en
|
4 |
- it
|
5 |
- de
|
6 |
- es
|
7 |
+
license: other
|
8 |
+
size_categories:
|
9 |
+
- 1K<n<10K
|
10 |
+
task_categories:
|
11 |
+
- token-classification
|
12 |
+
license_name: europarl-custom
|
13 |
+
license_link: https://www.statmt.org/europarl/
|
14 |
tags:
|
15 |
- NER
|
16 |
- Europarl
|
17 |
+
dataset_info:
|
18 |
+
config_name: en
|
19 |
+
features:
|
20 |
+
- name: tokens
|
21 |
+
sequence: string
|
22 |
+
- name: ner_tags
|
23 |
+
sequence:
|
24 |
+
class_label:
|
25 |
+
names:
|
26 |
+
'0': O
|
27 |
+
'1': B-PER
|
28 |
+
'2': I-PER
|
29 |
+
'3': B-ORG
|
30 |
+
'4': I-ORG
|
31 |
+
'5': B-LOC
|
32 |
+
'6': I-LOC
|
33 |
+
'7': B-MISC
|
34 |
+
'8': I-MISC
|
35 |
+
splits:
|
36 |
+
- name: test
|
37 |
+
num_bytes: 374649
|
38 |
+
num_examples: 799
|
39 |
+
download_size: 64713
|
40 |
+
dataset_size: 374649
|
41 |
+
configs:
|
42 |
+
- config_name: en
|
43 |
+
data_files:
|
44 |
+
- split: test
|
45 |
+
path: en/test-*
|
46 |
+
default: true
|
47 |
+
---
|
en/test-00000-of-00001.parquet
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:ac5dd43f41ca544dfbcd4b1ad643626b8d6973313b2c5bf6655e22798b35f742
|
3 |
+
size 64713
|
europarl-ner.py
DELETED
@@ -1,137 +0,0 @@
|
|
1 |
-
# coding=utf-8
|
2 |
-
|
3 |
-
"""The HF Datasets adapter for Evaluation Corpus for Named Entity Recognition using Europarl"""
|
4 |
-
|
5 |
-
import datasets
|
6 |
-
|
7 |
-
_CITATION = """@inproceedings{agerri-etal-2018-building,
|
8 |
-
title = "Building Named Entity Recognition Taggers via Parallel Corpora",
|
9 |
-
author = "Agerri, Rodrigo and
|
10 |
-
Chung, Yiling and
|
11 |
-
Aldabe, Itziar and
|
12 |
-
Aranberri, Nora and
|
13 |
-
Labaka, Gorka and
|
14 |
-
Rigau, German",
|
15 |
-
editor = "Calzolari, Nicoletta and
|
16 |
-
Choukri, Khalid and
|
17 |
-
Cieri, Christopher and
|
18 |
-
Declerck, Thierry and
|
19 |
-
Goggi, Sara and
|
20 |
-
Hasida, Koiti and
|
21 |
-
Isahara, Hitoshi and
|
22 |
-
Maegaard, Bente and
|
23 |
-
Mariani, Joseph and
|
24 |
-
Mazo, H{\'e}l{\`e}ne and
|
25 |
-
Moreno, Asuncion and
|
26 |
-
Odijk, Jan and
|
27 |
-
Piperidis, Stelios and
|
28 |
-
Tokunaga, Takenobu",
|
29 |
-
booktitle = "Proceedings of the Eleventh International Conference on Language Resources and Evaluation ({LREC} 2018)",
|
30 |
-
month = may,
|
31 |
-
year = "2018",
|
32 |
-
address = "Miyazaki, Japan",
|
33 |
-
publisher = "European Language Resources Association (ELRA)",
|
34 |
-
url = "https://aclanthology.org/L18-1557",
|
35 |
-
}"""
|
36 |
-
|
37 |
-
_DESCRIPTION = """This dataset contains a gold-standard test set created from the
|
38 |
-
Europarl corpus. The test set consists of 799 sentences manually annotated using
|
39 |
-
four entity types and following the CoNLL 2002 and 2003 guidelines for 4 languages:
|
40 |
-
English, German, Italian and Spanish."""
|
41 |
-
|
42 |
-
_DATA_URLs = {
|
43 |
-
"en": "https://github.com/ixa-ehu/ner-evaluation-corpus-europarl/raw/master/en-europarl.test.conll02",
|
44 |
-
"de": "https://github.com/ixa-ehu/ner-evaluation-corpus-europarl/raw/master/de-europarl.test.conll02",
|
45 |
-
"es": "https://github.com/ixa-ehu/ner-evaluation-corpus-europarl/raw/master/es-europarl.test.conll02",
|
46 |
-
"it": "https://github.com/ixa-ehu/ner-evaluation-corpus-europarl/raw/master/it-europarl.test.conll02",
|
47 |
-
}
|
48 |
-
_HOMEPAGE = "https://github.com/ixa-ehu/ner-evaluation-corpus-europarl"
|
49 |
-
_VERSION = "1.0.0"
|
50 |
-
_LANGS = ["en", "de", "es", "it"]
|
51 |
-
|
52 |
-
|
53 |
-
class EuroparlNERConfig(datasets.BuilderConfig):
|
54 |
-
def __init__(self, **kwargs):
|
55 |
-
super(EuroparlNERConfig, self).__init__(
|
56 |
-
version=datasets.Version(_VERSION, ""), **kwargs
|
57 |
-
)
|
58 |
-
|
59 |
-
|
60 |
-
class EuroparlNER(datasets.GeneratorBasedBuilder):
|
61 |
-
"""EuroparlNER is a multilingual named entity recognition dataset consisting of
|
62 |
-
manualy anotated part of the European Parliament Proceedings Parallel Corpus
|
63 |
-
1996-2011 with LOC, PER, ORG and MISC tags"""
|
64 |
-
|
65 |
-
VERSION = datasets.Version(_VERSION)
|
66 |
-
BUILDER_CONFIGS = [
|
67 |
-
EuroparlNERConfig(
|
68 |
-
name=lang, description=f"EuroparlNER examples in language {lang}"
|
69 |
-
)
|
70 |
-
for lang in _LANGS
|
71 |
-
]
|
72 |
-
DEFAULT_CONFIG_NAME = "en"
|
73 |
-
|
74 |
-
def _info(self):
|
75 |
-
features = datasets.Features(
|
76 |
-
{
|
77 |
-
"tokens": datasets.Sequence(datasets.Value("string")),
|
78 |
-
"ner_tags": datasets.Sequence(
|
79 |
-
datasets.features.ClassLabel(
|
80 |
-
names=[
|
81 |
-
"O",
|
82 |
-
"B-PER",
|
83 |
-
"I-PER",
|
84 |
-
"B-ORG",
|
85 |
-
"I-ORG",
|
86 |
-
"B-LOC",
|
87 |
-
"I-LOC",
|
88 |
-
"B-MISC",
|
89 |
-
"I-MISC",
|
90 |
-
]
|
91 |
-
)
|
92 |
-
),
|
93 |
-
}
|
94 |
-
)
|
95 |
-
return datasets.DatasetInfo(
|
96 |
-
description=_DESCRIPTION,
|
97 |
-
features=features,
|
98 |
-
supervised_keys=None,
|
99 |
-
homepage=_HOMEPAGE,
|
100 |
-
citation=_CITATION,
|
101 |
-
)
|
102 |
-
|
103 |
-
def _split_generators(self, dl_manager):
|
104 |
-
lang = self.config.name
|
105 |
-
dl_dir = dl_manager.download(_DATA_URLs[lang])
|
106 |
-
|
107 |
-
return [
|
108 |
-
datasets.SplitGenerator(
|
109 |
-
name=datasets.Split.TEST,
|
110 |
-
gen_kwargs={"filepath": dl_dir},
|
111 |
-
),
|
112 |
-
]
|
113 |
-
|
114 |
-
def _generate_examples(self, filepath):
|
115 |
-
guid_index = 1
|
116 |
-
with open(filepath, encoding="utf-8") as f:
|
117 |
-
tokens = []
|
118 |
-
ner_tags = []
|
119 |
-
for line in f:
|
120 |
-
if line == "" or line == "\n":
|
121 |
-
if tokens:
|
122 |
-
yield guid_index, {
|
123 |
-
"tokens": tokens,
|
124 |
-
"ner_tags": ner_tags,
|
125 |
-
}
|
126 |
-
guid_index += 1
|
127 |
-
tokens = []
|
128 |
-
ner_tags = []
|
129 |
-
else:
|
130 |
-
# EuroparlNER data is tab separated
|
131 |
-
splits = line.split("\t")
|
132 |
-
tokens.append(splits[0])
|
133 |
-
if len(splits) > 1:
|
134 |
-
ner_tags.append(splits[1].replace("\n", ""))
|
135 |
-
else:
|
136 |
-
# examples have no label in test set
|
137 |
-
ner_tags.append("O")
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|