Datasets:
Tasks:
Text Retrieval
Sub-tasks:
entity-linking-retrieval
Languages:
Chinese
Size:
1M - 10M
ArXiv:
License:
Delete hansel.py
Browse files
hansel.py
DELETED
@@ -1,171 +0,0 @@
|
|
1 |
-
# coding=utf-8
|
2 |
-
|
3 |
-
"""Hansel: A Chinese Few-Shot and Zero-Shot Entity Linking Benchmark"""
|
4 |
-
|
5 |
-
|
6 |
-
import json
|
7 |
-
import os
|
8 |
-
|
9 |
-
import datasets
|
10 |
-
|
11 |
-
|
12 |
-
|
13 |
-
_HANSEL_CITATION = """\
|
14 |
-
@misc{xu2022hansel,
|
15 |
-
title = {Hansel: A Chinese Few-Shot and Zero-Shot Entity Linking Benchmark},
|
16 |
-
author = {Xu, Zhenran and Shan, Zifei and Li, Yuxin and Hu, Baotian and Qin, Bing},
|
17 |
-
publisher = {arXiv},
|
18 |
-
year = {2022},
|
19 |
-
url = {https://arxiv.org/abs/2207.13005}
|
20 |
-
}
|
21 |
-
|
22 |
-
"""
|
23 |
-
|
24 |
-
_HANSEL_DESCRIPTION = """\
|
25 |
-
Hansel is a high-quality human-annotated Chinese entity linking (EL) dataset, used for testing Chinese EL systems' generalization ability to tail entities and emerging entities.
|
26 |
-
The test set contains Few-shot (FS) and zero-shot (ZS) slices, has 10K examples and uses Wikidata as the corresponding knowledge base.
|
27 |
-
The training and validation sets are from Wikipedia hyperlinks, useful for large-scale pretraining of Chinese EL systems.
|
28 |
-
|
29 |
-
"""
|
30 |
-
|
31 |
-
|
32 |
-
_URLS = {
|
33 |
-
"train": "hansel-train.jsonl",
|
34 |
-
"val": "hansel-val.jsonl",
|
35 |
-
"hansel-fs": "hansel-few-shot-v1.jsonl",
|
36 |
-
"hansel-zs": "hansel-zero-shot-v1.jsonl",
|
37 |
-
}
|
38 |
-
|
39 |
-
logger = datasets.logging.get_logger(__name__)
|
40 |
-
|
41 |
-
|
42 |
-
class HanselConfig(datasets.BuilderConfig):
|
43 |
-
"""BuilderConfig for HanselConfig."""
|
44 |
-
|
45 |
-
def __init__(self, features, data_url, citation, url, **kwargs):
|
46 |
-
"""BuilderConfig for Hansel.
|
47 |
-
|
48 |
-
Args:
|
49 |
-
features: `list[string]`, list of the features that will appear in the
|
50 |
-
feature dict. Should not include "label".
|
51 |
-
data_url: `string`, url to download the zip file from.
|
52 |
-
citation: `string`, citation for the data set.
|
53 |
-
url: `string`, url for information about the data set.
|
54 |
-
**kwargs: keyword arguments forwarded to super.
|
55 |
-
"""
|
56 |
-
super(HanselConfig, self).__init__(version=datasets.Version("1.0.0"), **kwargs)
|
57 |
-
self.features = features
|
58 |
-
self.data_url = data_url
|
59 |
-
self.citation = citation
|
60 |
-
self.url = url
|
61 |
-
|
62 |
-
|
63 |
-
class Hansel(datasets.GeneratorBasedBuilder):
|
64 |
-
"""The Hansel benchmark."""
|
65 |
-
|
66 |
-
BUILDER_CONFIGS = [
|
67 |
-
HanselConfig(
|
68 |
-
name="wiki",
|
69 |
-
description=_HANSEL_DESCRIPTION,
|
70 |
-
features=["id", "text", "start", "end", "mention", "gold_id"],
|
71 |
-
data_url="https://huggingface.co/datasets/HIT-TMG/Hansel/blob/main/",
|
72 |
-
citation=_HANSEL_CITATION,
|
73 |
-
url="https://github.com/HITsz-TMG/Hansel",
|
74 |
-
)
|
75 |
-
HanselConfig(
|
76 |
-
name="hansel-few-shot",
|
77 |
-
description=_HANSEL_DESCRIPTION,
|
78 |
-
features=["id", "text", "start", "end", "mention", "gold_id", "source", "domain"],
|
79 |
-
data_url="https://huggingface.co/datasets/HIT-TMG/Hansel/blob/main/",
|
80 |
-
citation=_HANSEL_CITATION,
|
81 |
-
url="https://github.com/HITsz-TMG/Hansel",
|
82 |
-
)
|
83 |
-
HanselConfig(
|
84 |
-
name="hansel-zero-shot",
|
85 |
-
description=_HANSEL_DESCRIPTION,
|
86 |
-
features=["id", "text", "start", "end", "mention", "gold_id", "source", "domain"],
|
87 |
-
data_url="https://huggingface.co/datasets/HIT-TMG/Hansel/blob/main/",
|
88 |
-
citation=_HANSEL_CITATION,
|
89 |
-
url="https://github.com/HITsz-TMG/Hansel",
|
90 |
-
)
|
91 |
-
]
|
92 |
-
|
93 |
-
def _info(self):
|
94 |
-
features = {feature: datasets.Value("string") for feature in self.config.features}
|
95 |
-
features["start"] = datasets.Value("int64")
|
96 |
-
features["end"] = datasets.Value("int64")
|
97 |
-
|
98 |
-
return datasets.DatasetInfo(
|
99 |
-
description=self.config.description,
|
100 |
-
features=datasets.Features(features),
|
101 |
-
homepage=self.config.url,
|
102 |
-
citation=self.config.citation
|
103 |
-
)
|
104 |
-
|
105 |
-
|
106 |
-
def _split_generators(self, dl_manager):
|
107 |
-
|
108 |
-
urls_to_download = self._URLS
|
109 |
-
downloaded_files = dl_manager.download_and_extract(urls_to_download)
|
110 |
-
|
111 |
-
if "hansel-few" in self.config.name:
|
112 |
-
return [
|
113 |
-
datasets.SplitGenerator(
|
114 |
-
name=datasets.Split.TEST,
|
115 |
-
gen_kwargs={
|
116 |
-
"data_file": downloaded_files["hansel-fs"]),
|
117 |
-
"split": datasets.Split.TEST,
|
118 |
-
},
|
119 |
-
),
|
120 |
-
]
|
121 |
-
if "hansel-zero" in self.config.name:
|
122 |
-
return [
|
123 |
-
datasets.SplitGenerator(
|
124 |
-
name=datasets.Split.TEST,
|
125 |
-
gen_kwargs={
|
126 |
-
"data_file": downloaded_files["hansel-zs"],
|
127 |
-
"split": datasets.Split.TEST,
|
128 |
-
},
|
129 |
-
),
|
130 |
-
]
|
131 |
-
return [
|
132 |
-
datasets.SplitGenerator(
|
133 |
-
name=datasets.Split.TRAIN,
|
134 |
-
gen_kwargs={
|
135 |
-
"data_file": downloaded_files["train"],
|
136 |
-
"split": datasets.Split.TRAIN,
|
137 |
-
},
|
138 |
-
),
|
139 |
-
datasets.SplitGenerator(
|
140 |
-
name=datasets.Split.VALIDATION,
|
141 |
-
gen_kwargs={
|
142 |
-
"data_file": downloaded_files["val"],
|
143 |
-
"split": datasets.Split.VALIDATION,
|
144 |
-
},
|
145 |
-
),
|
146 |
-
]
|
147 |
-
|
148 |
-
def _generate_examples(self, data_file, split):
|
149 |
-
logger.info("generating examples from = %s", data_file)
|
150 |
-
with open(data_file, encoding="utf-8") as f:
|
151 |
-
for line in f:
|
152 |
-
temDict = json.loads(line)
|
153 |
-
key = temDict["id"]
|
154 |
-
if "hansel" in self.config.name:
|
155 |
-
yield key, {
|
156 |
-
"text": temDict["text"],
|
157 |
-
"start": temDict["start"],
|
158 |
-
"end": temDict["end"],
|
159 |
-
"mention": temDict["mention"],
|
160 |
-
"gold_id": temDict["gold_id"],
|
161 |
-
"source": temDict["source"],
|
162 |
-
"domain": temDict["domain"],
|
163 |
-
}
|
164 |
-
else:
|
165 |
-
yield key, {
|
166 |
-
"text": temDict["text"],
|
167 |
-
"start": temDict["start"],
|
168 |
-
"end": temDict["end"],
|
169 |
-
"mention": temDict["mention"],
|
170 |
-
"gold_id": temDict["gold_id"],
|
171 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|