EuroEval 15.12.0__py3-none-any.whl → 16.7.1__py3-none-any.whl

euroeval/dataset_configs/latvian.py

@@ -0,0 +1,87 @@
+"""All Latvian dataset configurations used in EuroEval."""
+
+from ..data_models import DatasetConfig
+from ..languages import LATVIAN
+from ..tasks import COMMON_SENSE, KNOW, LA, NER, RC, SENT, SUMM
+
+### Official datasets ###
+
+LATVIAN_TWITTER_SENTIMENT_CONFIG = DatasetConfig(
+    name="latvian-twitter-sentiment",
+    pretty_name="Latvian Twitter Sentiment",
+    source="EuroEval/latvian-twitter-sentiment-mini",
+    task=SENT,
+    languages=[LATVIAN],
+)
+
+SCALA_LV_CONFIG = DatasetConfig(
+    name="scala-lv",
+    pretty_name="ScaLA-lv",
+    source="EuroEval/scala-lv",
+    task=LA,
+    languages=[LATVIAN],
+)
+
+FULLSTACK_NER_LV_CONFIG = DatasetConfig(
+    name="fullstack-ner-lv",
+    pretty_name="FullStack NER-lv",
+    source="EuroEval/fullstack-ner-lv-mini",
+    task=NER,
+    languages=[LATVIAN],
+)
+
+MULTI_WIKI_QA_LV_CONFIG = DatasetConfig(
+    name="multi-wiki-qa-lv",
+    pretty_name="MultiWikiQA-lv",
+    source="EuroEval/multi-wiki-qa-lv-mini",
+    task=RC,
+    languages=[LATVIAN],
+)
+
+LSM_CONFIG = DatasetConfig(
+    name="lsm",
+    pretty_name="LSM",
+    source="EuroEval/lsm-mini",
+    task=SUMM,
+    languages=[LATVIAN],
+)
+
+
+MMLU_LV_CONFIG = DatasetConfig(
+    name="mmlu-lv",
+    pretty_name="MMLU-lv",
+    source="EuroEval/mmlu-lv-mini",
+    task=KNOW,
+    languages=[LATVIAN],
+)
+
+COPA_LV_CONFIG = DatasetConfig(
+    name="copa-lv",
+    pretty_name="COPA-lv",
+    source="EuroEval/copa-lv",
+    task=COMMON_SENSE,
+    languages=[LATVIAN],
+    _labels=["a", "b"],
+)
+
+
+### Unofficial datasets ###
+
+WIKIANN_LV_CONFIG = DatasetConfig(
+    name="wikiann-lv",
+    pretty_name="WikiANN-lv",
+    source="EuroEval/wikiann-lv-mini",
+    task=NER,
+    languages=[LATVIAN],
+    unofficial=True,
+)
+
+WINOGRANDE_LV_CONFIG = DatasetConfig(
+    name="winogrande-lv",
+    pretty_name="Winogrande-lv",
+    source="EuroEval/winogrande-lv",
+    task=COMMON_SENSE,
+    languages=[LATVIAN],
+    _labels=["a", "b"],
+    unofficial=True,
+)

euroeval/dataset_configs/lithuanian.py

@@ -0,0 +1,64 @@
+"""All Lithuanian dataset configurations used in EuroEval."""
+
+from ..data_models import DatasetConfig
+from ..languages import LITHUANIAN
+from ..tasks import COMMON_SENSE, KNOW, LA, NER, RC, SENT, SUMM
+
+### Official datasets ###
+
+LITHUANIAN_EMOTIONS_CONFIG = DatasetConfig(
+    name="lithuanian-emotions",
+    pretty_name="Lithuanian Emotions",
+    source="EuroEval/lithuanian-emotions-mini",
+    task=SENT,
+    languages=[LITHUANIAN],
+)
+
+SCALA_LT_CONFIG = DatasetConfig(
+    name="scala-lt",
+    pretty_name="ScaLA-lt",
+    source="EuroEval/scala-lt",
+    task=LA,
+    languages=[LITHUANIAN],
+)
+
+WIKIANN_LT_CONFIG = DatasetConfig(
+    name="wikiann-lt",
+    pretty_name="WikiANN-lt",
+    source="EuroEval/wikiann-lt-mini",
+    task=NER,
+    languages=[LITHUANIAN],
+)
+
+MULTI_WIKI_QA_LT_CONFIG = DatasetConfig(
+    name="multi-wiki-qa-lt",
+    pretty_name="MultiWikiQA-lt",
+    source="EuroEval/multi-wiki-qa-lt-mini",
+    task=RC,
+    languages=[LITHUANIAN],
+)
+
+LRYTAS_CONFIG = DatasetConfig(
+    name="lrytas",
+    pretty_name="Lrytas",
+    source="EuroEval/lrytas-mini",
+    task=SUMM,
+    languages=[LITHUANIAN],
+)
+
+LT_HISTORY_CONFIG = DatasetConfig(
+    name="lt-history",
+    pretty_name="LT-History",
+    source="EuroEval/lt-history",
+    task=KNOW,
+    languages=[LITHUANIAN],
+)
+
+WINOGRANDE_LT_CONFIG = DatasetConfig(
+    name="winogrande-lt",
+    pretty_name="Winogrande-lt",
+    source="EuroEval/winogrande-lt",
+    task=COMMON_SENSE,
+    languages=[LITHUANIAN],
+    _labels=["a", "b"],
+)

euroeval/dataset_configs/norwegian.py

@@ -1,186 +1,212 @@
 """All Norwegian dataset configurations used in EuroEval."""
 
 from ..data_models import DatasetConfig
-from ..languages import NB, NN, NO
-from ..tasks import COMMON_SENSE, KNOW, LA, MCRC, NER, RC, SENT, SUMM
+from ..languages import NORWEGIAN, NORWEGIAN_BOKMÅL, NORWEGIAN_NYNORSK
+from ..tasks import COMMON_SENSE, EUROPEAN_VALUES, KNOW, LA, MCRC, NER, RC, SENT, SUMM
 
 ### Official datasets ###
 
 NOREC_CONFIG = DatasetConfig(
     name="norec",
-    pretty_name="the truncated version of the Norwegian sentiment classification "
-    "dataset NoReC",
-    huggingface_id="EuroEval/norec-mini",
+    pretty_name="NoReC",
+    source="EuroEval/norec-mini",
     task=SENT,
-    languages=[NB, NN, NO],
+    languages=[NORWEGIAN_BOKMÅL, NORWEGIAN_NYNORSK, NORWEGIAN],
 )
 
 SCALA_NB_CONFIG = DatasetConfig(
     name="scala-nb",
-    pretty_name="the Bokmål part of the linguistic acceptability dataset ScaLA",
-    huggingface_id="EuroEval/scala-nb",
+    pretty_name="ScaLA-nb",
+    source="EuroEval/scala-nb",
     task=LA,
-    languages=[NB, NO],
+    languages=[NORWEGIAN_BOKMÅL, NORWEGIAN],
 )
 
 SCALA_NN_CONFIG = DatasetConfig(
     name="scala-nn",
-    pretty_name="the Nynorsk part of the linguistic acceptability dataset ScaLA",
-    huggingface_id="EuroEval/scala-nn",
+    pretty_name="ScaLA-nn",
+    source="EuroEval/scala-nn",
     task=LA,
-    languages=[NN],
+    languages=[NORWEGIAN_NYNORSK],
 )
 
 NORNE_NB_CONFIG = DatasetConfig(
     name="norne-nb",
-    pretty_name="the truncated version of the Bokmål part of the Norwegian named "
-    "entity recognition dataset NorNE",
-    huggingface_id="EuroEval/norne-nb-mini",
+    pretty_name="NorNE-nb",
+    source="EuroEval/norne-nb-mini",
     task=NER,
-    languages=[NB, NO],
+    languages=[NORWEGIAN_BOKMÅL, NORWEGIAN],
 )
 
 NORNE_NN_CONFIG = DatasetConfig(
     name="norne-nn",
-    pretty_name="the truncated version of the Nynorsk part of the Norwegian named "
-    "entity recognition dataset NorNE",
-    huggingface_id="EuroEval/norne-nn-mini",
+    pretty_name="NorNE-nn",
+    source="EuroEval/norne-nn-mini",
     task=NER,
-    languages=[NN],
+    languages=[NORWEGIAN_NYNORSK],
 )
 
 NORQUAD_CONFIG = DatasetConfig(
     name="norquad",
-    pretty_name="the truncated version of the Norwegian question answering "
-    "dataset NorQuAD",
-    huggingface_id="EuroEval/norquad-mini",
+    pretty_name="NorQuAD",
+    source="EuroEval/norquad-mini",
     task=RC,
-    languages=[NB, NN, NO],
+    languages=[NORWEGIAN_BOKMÅL, NORWEGIAN_NYNORSK, NORWEGIAN],
     _num_few_shot_examples=2,
 )
 
 NO_SAMMENDRAG_CONFIG = DatasetConfig(
     name="no-sammendrag",
-    pretty_name="the truncated version of the Norwegian summarisation dataset "
-    "Norske Sammendrag",
-    huggingface_id="EuroEval/no-sammendrag-mini",
+    pretty_name="NoSammendrag",
+    source="EuroEval/no-sammendrag-mini",
     task=SUMM,
-    languages=[NB, NN, NO],
+    languages=[NORWEGIAN_BOKMÅL, NORWEGIAN_NYNORSK, NORWEGIAN],
 )
 
 NRK_QUIZ_QA_CONFIG = DatasetConfig(
     name="nrk-quiz-qa",
-    pretty_name="the truncated version of the Norwegian knowledge dataset NRK Quiz QA",
-    huggingface_id="EuroEval/nrk-quiz-qa-mini",
+    pretty_name="NRK Quiz QA",
+    source="EuroEval/nrk-quiz-qa-mini",
     task=KNOW,
-    languages=[NB, NN, NO],
+    languages=[NORWEGIAN_BOKMÅL, NORWEGIAN_NYNORSK, NORWEGIAN],
 )
 
 IDIOMS_NO_CONFIG = DatasetConfig(
     name="idioms-no",
-    pretty_name="the Norwegian knowledge dataset Idioms-no",
-    huggingface_id="EuroEval/idioms-no",
+    pretty_name="Idioms-no",
+    source="EuroEval/idioms-no",
     task=KNOW,
-    languages=[NB, NN, NO],
+    languages=[NORWEGIAN_BOKMÅL, NORWEGIAN_NYNORSK, NORWEGIAN],
 )
 
 NOR_COMMON_SENSE_QA_CONFIG = DatasetConfig(
     name="nor-common-sense-qa",
-    pretty_name="the truncated version of the Norwegian common-sense reasoning dataset "
-    "NorCommonSenseQA",
-    huggingface_id="EuroEval/nor-common-sense-qa",
+    pretty_name="NorCommonSenseQA",
+    source="EuroEval/nor-common-sense-qa",
     task=COMMON_SENSE,
-    languages=[NB, NN, NO],
+    languages=[NORWEGIAN_BOKMÅL, NORWEGIAN_NYNORSK, NORWEGIAN],
     _labels=["a", "b", "c", "d", "e"],
 )
 
+VALEU_NO_CONFIG = DatasetConfig(
+    name="valeu-no",
+    pretty_name="VaLEU-no",
+    source="EuroEval/european-values-no",
+    task=EUROPEAN_VALUES,
+    languages=[NORWEGIAN_BOKMÅL, NORWEGIAN_NYNORSK, NORWEGIAN],
+    splits=["test"],
+    bootstrap_samples=False,
+    _instruction_prompt="{text}",
+)
+
 
 ### Unofficial datasets ###
 
 NO_COLA_CONFIG = DatasetConfig(
     name="no-cola",
-    pretty_name="the truncated version of the Norwegian linguistic acceptability "
-    "dataset NoCoLA",
-    huggingface_id="EuroEval/no-cola-mini",
+    pretty_name="NoCoLA",
+    source="EuroEval/no-cola-mini",
     task=LA,
-    languages=[NB, NO],
+    languages=[NORWEGIAN_BOKMÅL, NORWEGIAN],
     unofficial=True,
 )
 
 NORGLM_MULTI_QA = DatasetConfig(
     name="norglm-multi-qa",
-    pretty_name="the question answering part of the Norwegian NorGLM multi-task human "
-    "annotated dataset NO-Multi-QA-Sum",
-    huggingface_id="EuroEval/norglm-multi-qa",
+    pretty_name="NorGLM-Multi-QA",
+    source="EuroEval/norglm-multi-qa",
     task=RC,
-    languages=[NB, NN, NO],
+    languages=[NORWEGIAN_BOKMÅL, NORWEGIAN_NYNORSK, NORWEGIAN],
     unofficial=True,
 )
 
 NORGLM_MULTI_SUM = DatasetConfig(
     name="norglm-multi-sum",
-    pretty_name="the summarisation part of the Norwegian NorGLM multi-task human "
-    "annotated dataset NO-Multi-QA-Sum",
-    huggingface_id="EuroEval/norglm-multi-sum",
+    pretty_name="NorGLM-Multi-Sum",
+    source="EuroEval/norglm-multi-sum",
     task=SUMM,
-    languages=[NB, NN, NO],
+    languages=[NORWEGIAN_BOKMÅL, NORWEGIAN_NYNORSK, NORWEGIAN],
     unofficial=True,
 )
 
 SCHIBSTED_NO_CONFIG = DatasetConfig(
     name="schibsted-no",
-    pretty_name="the Norwegian summarisation dataset Schibsted-no",
-    huggingface_id="EuroEval/schibsted-article-summaries-no",
+    pretty_name="Schibsted-no",
+    source="EuroEval/schibsted-article-summaries-no",
     task=SUMM,
-    languages=[NB, NN, NO],
+    languages=[NORWEGIAN_BOKMÅL, NORWEGIAN_NYNORSK, NORWEGIAN],
     unofficial=True,
 )
 
 PERSONAL_SUM_CONFIG = DatasetConfig(
     name="personal-sum",
-    pretty_name="the Norwegian summarisation dataset personal-sum",
-    huggingface_id="EuroEval/personal-sum",
+    pretty_name="Personal Sum",
+    source="EuroEval/personal-sum",
     task=SUMM,
-    languages=[NB, NN, NO],
+    languages=[NORWEGIAN_BOKMÅL, NORWEGIAN_NYNORSK, NORWEGIAN],
     unofficial=True,
 )
 
 MMLU_NO_CONFIG = DatasetConfig(
     name="mmlu-no",
-    pretty_name="the truncated version of the Norwegian knowledge dataset MMLU-no, "
-    "translated from the English MMLU dataset",
-    huggingface_id="EuroEval/mmlu-no-mini",
+    pretty_name="MMLU-no",
+    source="EuroEval/mmlu-no-mini",
     task=KNOW,
-    languages=[NB, NN, NO],
+    languages=[NORWEGIAN_BOKMÅL, NORWEGIAN_NYNORSK, NORWEGIAN],
     unofficial=True,
 )
 
 ARC_NO_CONFIG = DatasetConfig(
     name="arc-no",
-    pretty_name="the truncated version of the Norwegian knowledge dataset ARC-no, "
-    "translated from the English ARC dataset",
-    huggingface_id="EuroEval/arc-no-mini",
+    pretty_name="ARC-no",
+    source="EuroEval/arc-no-mini",
     task=KNOW,
-    languages=[NB, NN, NO],
+    languages=[NORWEGIAN_BOKMÅL, NORWEGIAN_NYNORSK, NORWEGIAN],
     unofficial=True,
 )
 
 HELLASWAG_NO_CONFIG = DatasetConfig(
     name="hellaswag-no",
-    pretty_name="the truncated version of the Norwegian common-sense reasoning dataset "
-    "HellaSwag-no, translated from the English HellaSwag dataset",
-    huggingface_id="EuroEval/hellaswag-no-mini",
+    pretty_name="HellaSwag-no",
+    source="EuroEval/hellaswag-no-mini",
     task=COMMON_SENSE,
-    languages=[NB, NN, NO],
+    languages=[NORWEGIAN_BOKMÅL, NORWEGIAN_NYNORSK, NORWEGIAN],
     unofficial=True,
 )
 
 BELEBELE_NO_CONFIG = DatasetConfig(
     name="belebele-no",
-    pretty_name="the Norwegian multiple choice reading comprehension dataset "
-    "BeleBele-no, translated from the English BeleBele dataset",
-    huggingface_id="EuroEval/belebele-no-mini",
+    pretty_name="Belebele-no",
+    source="EuroEval/belebele-no-mini",
     task=MCRC,
-    languages=[NB, NN, NO],
+    languages=[NORWEGIAN_BOKMÅL, NORWEGIAN_NYNORSK, NORWEGIAN],
+    unofficial=True,
+)
+
+MULTI_WIKI_QA_NB_CONFIG = DatasetConfig(
+    name="multi-wiki-qa-nb",
+    pretty_name="MultiWikiQA-nb",
+    source="EuroEval/multi-wiki-qa-no-mini",
+    task=RC,
+    languages=[NORWEGIAN_BOKMÅL, NORWEGIAN],
+    unofficial=True,
+)
+
+MULTI_WIKI_QA_NN_CONFIG = DatasetConfig(
+    name="multi-wiki-qa-nn",
+    pretty_name="MultiWikiQA-nn",
+    source="EuroEval/multi-wiki-qa-nn-mini",
+    task=RC,
+    languages=[NORWEGIAN_NYNORSK],
+    unofficial=True,
+)
+
+WINOGRANDE_NO_CONFIG = DatasetConfig(
+    name="winogrande-no",
+    pretty_name="Winogrande-no",
+    source="EuroEval/winogrande-no",
+    task=COMMON_SENSE,
+    languages=[NORWEGIAN_BOKMÅL, NORWEGIAN_NYNORSK, NORWEGIAN],
+    _labels=["a", "b"],
     unofficial=True,
 )

euroeval/dataset_configs/polish.py

@@ -0,0 +1,96 @@
+"""All Polish dataset configurations used in EuroEval."""
+
+from ..data_models import DatasetConfig
+from ..languages import POLISH
+from ..tasks import COMMON_SENSE, EUROPEAN_VALUES, KNOW, LA, NER, RC, SENT, SUMM
+
+### Official datasets ###
+
+POLEMO2_CONFIG = DatasetConfig(
+    name="polemo2",
+    pretty_name="Polemo2",
+    source="EuroEval/polemo2-mini",
+    task=SENT,
+    languages=[POLISH],
+)
+
+SCALA_PL_CONFIG = DatasetConfig(
+    name="scala-pl",
+    pretty_name="ScaLA-pl",
+    source="EuroEval/scala-pl",
+    task=LA,
+    languages=[POLISH],
+)
+
+KPWR_NER_CONFIG = DatasetConfig(
+    name="kpwr-ner",
+    pretty_name="KPWr-NER",
+    source="EuroEval/kpwr-ner",
+    task=NER,
+    languages=[POLISH],
+)
+
+POQUAD_CONFIG = DatasetConfig(
+    name="poquad",
+    pretty_name="PoQuAD",
+    source="EuroEval/poquad-mini",
+    task=RC,
+    languages=[POLISH],
+)
+
+PSC_CONFIG = DatasetConfig(
+    name="psc",
+    pretty_name="PSC",
+    source="EuroEval/psc-mini",
+    task=SUMM,
+    languages=[POLISH],
+)
+
+LLMZSZL_CONFIG = DatasetConfig(
+    name="llmzszl",
+    pretty_name="LLMzSzŁ",
+    source="EuroEval/llmzszl-mini",
+    task=KNOW,
+    languages=[POLISH],
+)
+
+WINOGRANDE_PL_CONFIG = DatasetConfig(
+    name="winogrande-pl",
+    pretty_name="Winogrande-pl",
+    source="EuroEval/winogrande-pl",
+    task=COMMON_SENSE,
+    languages=[POLISH],
+    _labels=["a", "b"],
+)
+
+VALEU_PL_CONFIG = DatasetConfig(
+    name="valeu-pl",
+    pretty_name="VaLEU-pl",
+    source="EuroEval/european-values-pl",
+    task=EUROPEAN_VALUES,
+    languages=[POLISH],
+    splits=["test"],
+    bootstrap_samples=False,
+    _instruction_prompt="{text}",
+)
+
+
+### Unofficial datasets ###
+
+MULTI_WIKI_QA_PL_CONFIG = DatasetConfig(
+    name="multi-wiki-qa-pl",
+    pretty_name="MultiWikiQA-pl",
+    source="EuroEval/multi-wiki-qa-pl-mini",
+    task=RC,
+    languages=[POLISH],
+    unofficial=True,
+)
+
+GOLDENSWAG_PL_CONFIG = DatasetConfig(
+    name="goldenswag-pl",
+    pretty_name="GoldenSwag-pl",
+    source="EuroEval/goldenswag-pl-mini",
+    task=COMMON_SENSE,
+    languages=[POLISH],
+    unofficial=True,
+)

euroeval/dataset_configs/portuguese.py

@@ -1,64 +1,77 @@
 """All Portuguese dataset configurations used in EuroEval."""
 
 from ..data_models import DatasetConfig
-from ..languages import PT
-from ..tasks import COMMON_SENSE, KNOW, LA, MCRC, NER, SENT, SUMM
+from ..languages import EUROPEAN_PORTUGUESE, PORTUGUESE
+from ..tasks import COMMON_SENSE, EUROPEAN_VALUES, KNOW, LA, MCRC, NER, RC, SENT, SUMM
 
 ### Official datasets ###
 
 SST2_PT_CONFIG = DatasetConfig(
     name="sst2-pt",
-    pretty_name="the truncated version of the Portuguese sentiment classification "
-    "dataset SST2-pt, translated from the English SST2 dataset",
-    huggingface_id="EuroEval/sst2-pt-mini",
+    pretty_name="SST2-pt",
+    source="EuroEval/sst2-pt-mini",
     task=SENT,
-    languages=[PT],
+    languages=[PORTUGUESE, EUROPEAN_PORTUGUESE],
     _labels=["positive", "negative"],
 )
 
-
-MMLU_PT_CONFIG = DatasetConfig(
-    name="mmlu-pt",
-    pretty_name="the truncated version of the Portuguese knowledge dataset MMLU-pt, "
-    "translated from the English MMLU dataset",
-    huggingface_id="EuroEval/mmlu-pt-mini",
-    task=KNOW,
-    languages=[PT],
-)
-
-
-GOLDENSWAG_PT_CONFIG = DatasetConfig(
-    name="goldenswag-pt",
-    pretty_name="the truncated version of the Portuguese common-sense reasoning "
-    "dataset GoldenSwag-pt, translated from the English GoldenSwag dataset",
-    huggingface_id="EuroEval/goldenswag-pt-mini",
-    task=COMMON_SENSE,
-    languages=[PT],
-)
-
-
 SCALA_PT = DatasetConfig(
     name="scala-pt",
-    pretty_name="the Portuguese part of the linguistic acceptability dataset ScaLA",
-    huggingface_id="EuroEval/scala-pt",
+    pretty_name="ScaLA-pt",
+    source="EuroEval/scala-pt",
     task=LA,
-    languages=[PT],
+    languages=[PORTUGUESE, EUROPEAN_PORTUGUESE],
 )
 
 HAREM_CONFIG = DatasetConfig(
     name="harem",
-    pretty_name="the Portuguese named entity recognition dataset HAREM",
-    huggingface_id="EuroEval/harem",
+    pretty_name="HAREM",
+    source="EuroEval/harem",
     task=NER,
-    languages=[PT],
+    languages=[PORTUGUESE, EUROPEAN_PORTUGUESE],
+)
+
+MULTI_WIKI_QA_PT_CONFIG = DatasetConfig(
+    name="multi-wiki-qa-pt",
+    pretty_name="MultiWikiQA-pt",
+    source="EuroEval/multi-wiki-qa-pt-pt-mini",
+    task=RC,
+    languages=[PORTUGUESE, EUROPEAN_PORTUGUESE],
 )
 
 PUBLICO_CONFIG = DatasetConfig(
     name="publico",
-    pretty_name="the truncated version of the Portuguese summarisation dataset Público",
-    huggingface_id="EuroEval/publico-mini",
+    pretty_name="Publico",
+    source="EuroEval/publico-mini",
     task=SUMM,
-    languages=[PT],
+    languages=[PORTUGUESE, EUROPEAN_PORTUGUESE],
+)
+
+MMLU_PT_CONFIG = DatasetConfig(
+    name="mmlu-pt",
+    pretty_name="MMLU-pt",
+    source="EuroEval/mmlu-pt-mini",
+    task=KNOW,
+    languages=[PORTUGUESE, EUROPEAN_PORTUGUESE],
+)
+
+GOLDENSWAG_PT_CONFIG = DatasetConfig(
+    name="goldenswag-pt",
+    pretty_name="GoldenSwag-pt",
+    source="EuroEval/goldenswag-pt-mini",
+    task=COMMON_SENSE,
+    languages=[PORTUGUESE, EUROPEAN_PORTUGUESE],
+)
+
+VALEU_PT_CONFIG = DatasetConfig(
+    name="valeu-pt",
+    pretty_name="VaLEU-pt",
+    source="EuroEval/european-values-pt",
+    task=EUROPEAN_VALUES,
+    languages=[PORTUGUESE, EUROPEAN_PORTUGUESE],
+    splits=["test"],
+    bootstrap_samples=False,
+    _instruction_prompt="{text}",
 )
 
 
@@ -66,9 +79,19 @@ PUBLICO_CONFIG = DatasetConfig(
 
 BOOLQ_PT_CONFIG = DatasetConfig(
     name="boolq-pt",
-    pretty_name="the Portuguese multiple choice reading comprehension dataset "
-    "BoolQ-pt, translated from the English BoolQ dataset",
-    huggingface_id="EuroEval/boolq-pt",
+    pretty_name="BoolQ-pt",
+    source="EuroEval/boolq-pt",
     task=MCRC,
-    languages=[PT],
+    languages=[PORTUGUESE, EUROPEAN_PORTUGUESE],
+    unofficial=True,
+)
+
+WINOGRANDE_PT_CONFIG = DatasetConfig(
+    name="winogrande-pt",
+    pretty_name="Winogrande-pt",
+    source="EuroEval/winogrande-pt",
+    task=COMMON_SENSE,
+    languages=[PORTUGUESE, EUROPEAN_PORTUGUESE],
+    _labels=["a", "b"],
+    unofficial=True,
 )