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 | """All Spanish dataset configurations used in EuroEval."""
from ..data_models import DatasetConfig
from ..languages import ES
from ..tasks import COMMON_SENSE, EUROPEAN_VALUES, KNOW, LA, MCRC, NER, RC, SENT, SUMM
### Official datasets ###
SENTIMENT_HEADLINES_CONFIG = DatasetConfig(
name="sentiment-headlines-es",
pretty_name="the truncated version of the Spanish sentiment classification dataset "
"SentimentHeadlines",
huggingface_id="EuroEval/sentiment-headlines-es",
task=SENT,
languages=[ES],
)
SCALA_ES_CONFIG = DatasetConfig(
name="scala-es",
pretty_name="the Spanish part of the linguistic acceptability dataset ScaLA",
huggingface_id="EuroEval/scala-es",
task=LA,
languages=[ES],
)
CONLL_ES_CONFIG = DatasetConfig(
name="conll-es",
pretty_name="the Spanish part of the truncated version of the named entity "
"recognition dataset CoNLL 2002",
huggingface_id="EuroEval/conll-es-mini",
task=NER,
languages=[ES],
)
MLQA_ES_CONFIG = DatasetConfig(
name="mlqa-es",
pretty_name="the Spanish version of the reading comprehension dataset MLQA",
huggingface_id="EuroEval/mlqa-es",
task=RC,
languages=[ES],
)
MLSUM_ES_CONFIG = DatasetConfig(
name="mlsum-es",
pretty_name="the truncated version of the Spanish summarisation dataset MLSum-es",
huggingface_id="EuroEval/mlsum-es-mini",
task=SUMM,
languages=[ES],
)
MMLU_ES_CONFIG = DatasetConfig(
name="mmlu-es",
pretty_name="the truncated version of the Spanish knowledge dataset MMLU-es, "
"translated from the English MMLU dataset",
huggingface_id="EuroEval/mmlu-es-mini",
task=KNOW,
languages=[ES],
)
HELLASWAG_ES_CONFIG = DatasetConfig(
name="hellaswag-es",
pretty_name="the truncated version of the Spanish common-sense reasoning dataset "
"HellaSwag-es, translated from the English HellaSwag dataset",
huggingface_id="EuroEval/hellaswag-es-mini",
task=COMMON_SENSE,
languages=[ES],
)
EUROPEAN_VALUES_ES_CONFIG = DatasetConfig(
name="european-values-es",
pretty_name="the Spanish version of the European values evaluation dataset",
huggingface_id="EuroEval/european-values-es",
task=EUROPEAN_VALUES,
languages=[ES],
splits=["test"],
bootstrap_samples=False,
_instruction_prompt="{text}",
)
### Unofficial datasets ###
XQUAD_ES_CONFIG = DatasetConfig(
name="xquad-es",
pretty_name="the Spanish version of the reading comprehension dataset XQuAD",
huggingface_id="EuroEval/xquad-es",
task=RC,
languages=[ES],
unofficial=True,
)
BELEBELE_ES_CONFIG = DatasetConfig(
name="belebele-es",
pretty_name="the Spanish multiple choice reading comprehension dataset "
"BeleBele-es, translated from the English BeleBele dataset",
huggingface_id="EuroEval/belebele-es-mini",
task=MCRC,
languages=[ES],
unofficial=True,
)
MULTI_WIKI_QA_ES_CONFIG = DatasetConfig(
name="multi-wiki-qa-es",
pretty_name="the truncated version of the Spanish part of the reading "
"comprehension dataset MultiWikiQA",
huggingface_id="EuroEval/multi-wiki-qa-es-mini",
task=RC,
languages=[ES],
unofficial=True,
)
GOLDENSWAG_ES_CONFIG = DatasetConfig(
name="goldenswag-es",
pretty_name="the truncated version of the Spanish common-sense reasoning "
"dataset GoldenSwag-es, translated from the English GoldenSwag dataset",
huggingface_id="EuroEval/goldenswag-es-mini",
task=COMMON_SENSE,
languages=[ES],
unofficial=True,
)
EUROPEAN_VALUES_SITUATIONAL_ES_CONFIG = DatasetConfig(
name="european-values-situational-es",
pretty_name="the Spanish version of the European values evaluation dataset, where "
"the questions are phrased in a situational way",
huggingface_id="EuroEval/european-values-situational-es",
task=EUROPEAN_VALUES,
languages=[ES],
splits=["test"],
bootstrap_samples=False,
_instruction_prompt="{text}",
unofficial=True,
)
EUROPEAN_VALUES_COMPLETIONS_ES_CONFIG = DatasetConfig(
name="european-values-completions-es",
pretty_name="the Spanish version of the European values evaluation dataset, where "
"the questions are phrased as sentence completions",
huggingface_id="EuroEval/european-values-completions-es",
task=EUROPEAN_VALUES,
languages=[ES],
splits=["test"],
bootstrap_samples=False,
_instruction_prompt="{text}",
unofficial=True,
)
|