Create SemEval2020Task9CodeSwitch.py
Browse files- SemEval2020Task9CodeSwitch.py +137 -0
SemEval2020Task9CodeSwitch.py
ADDED
@@ -0,0 +1,137 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
import datasets
|
2 |
+
|
3 |
+
|
4 |
+
logger = datasets.logging.get_logger(__name__)
|
5 |
+
|
6 |
+
|
7 |
+
_CITATION = """\
|
8 |
+
@inproceedings{tjong-kim-sang-2002-introduction,
|
9 |
+
title = "Introduction to the {C}o{NLL}-2002 Shared Task: Language-Independent Named Entity Recognition",
|
10 |
+
author = "Tjong Kim Sang, Erik F.",
|
11 |
+
booktitle = "{COLING}-02: The 6th Conference on Natural Language Learning 2002 ({C}o{NLL}-2002)",
|
12 |
+
year = "2002",
|
13 |
+
url = "https://www.aclweb.org/anthology/W02-2024",
|
14 |
+
}
|
15 |
+
"""
|
16 |
+
|
17 |
+
_DESCRIPTION = """\
|
18 |
+
Named entities are phrases that contain the names of persons, organizations, locations, times and quantities.
|
19 |
+
Example:
|
20 |
+
[PER Wolff] , currently a journalist in [LOC Argentina] , played with [PER Del Bosque] in the final years of the seventies in [ORG Real Madrid] .
|
21 |
+
The shared task of CoNLL-2002 concerns language-independent named entity recognition.
|
22 |
+
We will concentrate on four types of named entities: persons, locations, organizations and names of miscellaneous entities that do not belong to the previous three groups.
|
23 |
+
The participants of the shared task will be offered training and test data for at least two languages.
|
24 |
+
They will use the data for developing a named-entity recognition system that includes a machine learning component.
|
25 |
+
Information sources other than the training data may be used in this shared task.
|
26 |
+
We are especially interested in methods that can use additional unannotated data for improving their performance (for example co-training).
|
27 |
+
The train/validation/test sets are available in Spanish and Dutch.
|
28 |
+
For more details see https://www.clips.uantwerpen.be/semeval2016/ner/ and https://www.aclweb.org/anthology/W02-2024/
|
29 |
+
"""
|
30 |
+
|
31 |
+
_URL = "https://raw.githubusercontent.com/YaxinCui/Semeval_2020_task9_data/main/Spanglish/"
|
32 |
+
|
33 |
+
TRAINING_FILE_Dict = {
|
34 |
+
'Spanglish': "Spanglish_train.conll",
|
35 |
+
|
36 |
+
}
|
37 |
+
|
38 |
+
TEST_FILE_Dict = {
|
39 |
+
'Spanglish': "Spanglish_dev.conll",
|
40 |
+
}
|
41 |
+
|
42 |
+
class Semeval2016Config(datasets.BuilderConfig):
|
43 |
+
"""BuilderConfig for Semeval2016"""
|
44 |
+
|
45 |
+
def __init__(self, **kwargs):
|
46 |
+
"""BuilderConfig forSemeval2016.
|
47 |
+
Args:
|
48 |
+
**kwargs: keyword arguments forwarded to super.
|
49 |
+
"""
|
50 |
+
super(Semeval2016Config, self).__init__(**kwargs)
|
51 |
+
|
52 |
+
|
53 |
+
class Semeval2016(datasets.GeneratorBasedBuilder):
|
54 |
+
"""Semeval2016 dataset."""
|
55 |
+
|
56 |
+
BUILDER_CONFIGS = [
|
57 |
+
Semeval2016Config(name="Spanglish", version=datasets.Version("1.0.0"), description="Semeval2016 Spanish dataset"),
|
58 |
+
]
|
59 |
+
|
60 |
+
def _info(self):
|
61 |
+
return datasets.DatasetInfo(
|
62 |
+
description=_DESCRIPTION,
|
63 |
+
features=datasets.Features(
|
64 |
+
{
|
65 |
+
"id": datasets.Value("string"),
|
66 |
+
"meta": datasets.Value("string"),
|
67 |
+
"tokens": datasets.Sequence(datasets.Value("string")),
|
68 |
+
# "langs": datasets.Sequence(datasets.features.ClassLabel(names=["lang1","lang2","ambiguous","other","ne","unk","mixed","fw","8","9","10","11",] ) ),
|
69 |
+
"label": datasets.features.ClassLabel(
|
70 |
+
names=[
|
71 |
+
"positive",
|
72 |
+
"neutral",
|
73 |
+
"negative",
|
74 |
+
]
|
75 |
+
),
|
76 |
+
}
|
77 |
+
),
|
78 |
+
supervised_keys=None,
|
79 |
+
homepage="/",
|
80 |
+
citation=_CITATION,
|
81 |
+
)
|
82 |
+
|
83 |
+
def _split_generators(self, dl_manager):
|
84 |
+
"""Returns SplitGenerators."""
|
85 |
+
|
86 |
+
if self.config.name=="Spanglish":
|
87 |
+
urls_to_download = {
|
88 |
+
"train": f"{_URL}{TRAINING_FILE_Dict[self.config.name]}",
|
89 |
+
"test": f"{_URL}{TEST_FILE_Dict[self.config.name]}",
|
90 |
+
}
|
91 |
+
|
92 |
+
downloaded_files = dl_manager.download_and_extract(urls_to_download)
|
93 |
+
|
94 |
+
return [
|
95 |
+
datasets.SplitGenerator(name=datasets.Split.TRAIN, gen_kwargs={"filepath": downloaded_files["train"]}),
|
96 |
+
datasets.SplitGenerator(name=datasets.Split.TEST, gen_kwargs={"filepath": downloaded_files["test"]}),
|
97 |
+
]
|
98 |
+
|
99 |
+
def _generate_examples(self, filepath):
|
100 |
+
logger.info("⏳ Generating examples from = %s", filepath)
|
101 |
+
prev_pos = '$$$'
|
102 |
+
with open(filepath, encoding="utf-8") as f:
|
103 |
+
guid = 0
|
104 |
+
meta = None
|
105 |
+
tokens = []
|
106 |
+
langs = []
|
107 |
+
label = None
|
108 |
+
for line in f:
|
109 |
+
if len(tokens) and (line == "" or line == "\n"):
|
110 |
+
yield guid, {
|
111 |
+
"id": str(guid),
|
112 |
+
"meta": str(meta),
|
113 |
+
"tokens": tokens,
|
114 |
+
"label": label,
|
115 |
+
}
|
116 |
+
guid += 1
|
117 |
+
tokens = []
|
118 |
+
langs = []
|
119 |
+
labels = []
|
120 |
+
else:
|
121 |
+
line = line.strip()
|
122 |
+
# semeval2016 tokens are space separated
|
123 |
+
splits = [s.rstrip() for s in line.split(" ")]
|
124 |
+
if len(tokens)==0 and line.startswith("meta "):
|
125 |
+
meta = splits[1]
|
126 |
+
label = splits[2]
|
127 |
+
else:
|
128 |
+
tokens.append(splits[0])
|
129 |
+
langs.append(splits[1])
|
130 |
+
# last example
|
131 |
+
|
132 |
+
yield guid, {
|
133 |
+
"id": str(guid),
|
134 |
+
"meta": str(meta),
|
135 |
+
"tokens": tokens,
|
136 |
+
"label": label,
|
137 |
+
}
|