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
164
165
166
167
168
169
170
171
172
173
174
175
176
177 | """All Danish dataset configurations used in EuroEval."""
from ..data_models import DatasetConfig
from ..languages import DA
from ..tasks import COMMON_SENSE, EUROPEAN_VALUES, KNOW, LA, MCRC, NER, RC, SENT, SUMM
### Official datasets ###
ANGRY_TWEETS_CONFIG = DatasetConfig(
name="angry-tweets",
pretty_name="the truncated version of the Danish sentiment classification "
"dataset AngryTweets",
huggingface_id="EuroEval/angry-tweets-mini",
task=SENT,
languages=[DA],
)
SCALA_DA_CONFIG = DatasetConfig(
name="scala-da",
pretty_name="the Danish part of the linguistic acceptability dataset ScaLA",
huggingface_id="EuroEval/scala-da",
task=LA,
languages=[DA],
)
DANSK_CONFIG = DatasetConfig(
name="dansk",
pretty_name="the truncated version of the Danish named entity recognition "
"dataset DANSK",
huggingface_id="EuroEval/dansk-mini",
task=NER,
languages=[DA],
)
SCANDIQA_DA_CONFIG = DatasetConfig(
name="scandiqa-da",
pretty_name="the Danish part of the truncated version of the question answering "
"dataset ScandiQA",
huggingface_id="EuroEval/scandiqa-da-mini",
task=RC,
languages=[DA],
)
NORDJYLLAND_NEWS_CONFIG = DatasetConfig(
name="nordjylland-news",
pretty_name="the truncated version of the Danish summarisation dataset "
"Nordjylland News",
huggingface_id="EuroEval/nordjylland-news-mini",
task=SUMM,
languages=[DA],
)
DANSKE_TALEMAADER_CONFIG = DatasetConfig(
name="danske-talemaader",
pretty_name="the truncated version of the Danish knowledge dataset Danske "
"Talemåder",
huggingface_id="EuroEval/danske-talemaader",
task=KNOW,
languages=[DA],
)
DANISH_CITIZEN_TESTS_CONFIG = DatasetConfig(
name="danish-citizen-tests",
pretty_name="the Danish knowledge dataset Danish Citizen Tests",
huggingface_id="EuroEval/danish-citizen-tests-updated",
task=KNOW,
languages=[DA],
)
HELLASWAG_DA_CONFIG = DatasetConfig(
name="hellaswag-da",
pretty_name="the truncated version of the Danish common-sense reasoning dataset "
"HellaSwag-da, translated from the English HellaSwag dataset",
huggingface_id="EuroEval/hellaswag-da-mini",
task=COMMON_SENSE,
languages=[DA],
)
EUROPEAN_VALUES_DA_CONFIG = DatasetConfig(
name="european-values-da",
pretty_name="the Danish version of the European values evaluation dataset",
huggingface_id="EuroEval/european-values-da",
task=EUROPEAN_VALUES,
languages=[DA],
splits=["test"],
bootstrap_samples=False,
_instruction_prompt="{text}",
)
### Unofficial datasets ###
DANE_CONFIG = DatasetConfig(
name="dane",
pretty_name="the truncated version of the Danish named entity recognition "
"dataset DaNE",
huggingface_id="EuroEval/dane-mini",
task=NER,
languages=[DA],
unofficial=True,
)
MMLU_DA_CONFIG = DatasetConfig(
name="mmlu-da",
pretty_name="the truncated version of the Danish knowledge dataset MMLU-da, "
"translated from the English MMLU dataset",
huggingface_id="EuroEval/mmlu-da-mini",
task=KNOW,
languages=[DA],
unofficial=True,
)
ARC_DA_CONFIG = DatasetConfig(
name="arc-da",
pretty_name="the truncated version of the Danish knowledge dataset ARC-da, "
"translated from the English ARC dataset",
huggingface_id="EuroEval/arc-da-mini",
task=KNOW,
languages=[DA],
unofficial=True,
)
BELEBELE_DA_CONFIG = DatasetConfig(
name="belebele-da",
pretty_name="the Danish multiple choice reading comprehension dataset BeleBele-da, "
"translated from the English BeleBele dataset",
huggingface_id="EuroEval/belebele-da-mini",
task=MCRC,
languages=[DA],
unofficial=True,
)
MULTI_WIKI_QA_DA_CONFIG = DatasetConfig(
name="multi-wiki-qa-da",
pretty_name="the truncated version of the Danish part of the reading "
"comprehension dataset MultiWikiQA",
huggingface_id="EuroEval/multi-wiki-qa-da-mini",
task=RC,
languages=[DA],
unofficial=True,
)
GOLDENSWAG_DA_CONFIG = DatasetConfig(
name="goldenswag-da",
pretty_name="the truncated version of the Danish common-sense reasoning "
"dataset GoldenSwag-da, translated from the English GoldenSwag dataset",
huggingface_id="EuroEval/goldenswag-da-mini",
task=COMMON_SENSE,
languages=[DA],
unofficial=True,
)
EUROPEAN_VALUES_SITUATIONAL_DA_CONFIG = DatasetConfig(
name="european-values-situational-da",
pretty_name="the Danish version of the European values evaluation dataset, where "
"the questions are phrased in a situational way",
huggingface_id="EuroEval/european-values-situational-da",
task=EUROPEAN_VALUES,
languages=[DA],
splits=["test"],
bootstrap_samples=False,
_instruction_prompt="{text}",
unofficial=True,
)
EUROPEAN_VALUES_COMPLETIONS_DA_CONFIG = DatasetConfig(
name="european-values-completions-da",
pretty_name="the Danish version of the European values evaluation dataset, where "
"the questions are phrased as sentence completions",
huggingface_id="EuroEval/european-values-completions-da",
task=EUROPEAN_VALUES,
languages=[DA],
splits=["test"],
bootstrap_samples=False,
_instruction_prompt="{text}",
unofficial=True,
)
|