americas_nli

参考:

aym

使用以下命令在 TFDS 中加载此数据集:

ds = tfds.load('huggingface:americas_nli/aym')
  • 说明
AmericasNLI is an extension of XNLI (Conneau et al., 2018) – a natural language inference (NLI) dataset covering 15 high-resource languages – to 10 low-resource indigenous languages spoken in the Americas: Ashaninka, Aymara, Bribri, Guarani, Nahuatl, Otomi, Quechua, Raramuri, Shipibo-Konibo, and Wixarika. As with MNLI, the goal is to predict textual entailment (does sentence A imply/contradict/neither sentence B) and is a classification task (given two sentences, predict one of three labels).
  • 许可:无已知许可
  • 版本:1.0.0
  • 拆分
拆分 样本
'test' 750
'validation' 743
  • 特征
{
    "premise": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "hypothesis": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "label": {
        "num_classes": 3,
        "names": [
            "entailment",
            "neutral",
            "contradiction"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    }
}

bzd

使用以下命令在 TFDS 中加载此数据集:

ds = tfds.load('huggingface:americas_nli/bzd')
  • 说明
AmericasNLI is an extension of XNLI (Conneau et al., 2018) – a natural language inference (NLI) dataset covering 15 high-resource languages – to 10 low-resource indigenous languages spoken in the Americas: Ashaninka, Aymara, Bribri, Guarani, Nahuatl, Otomi, Quechua, Raramuri, Shipibo-Konibo, and Wixarika. As with MNLI, the goal is to predict textual entailment (does sentence A imply/contradict/neither sentence B) and is a classification task (given two sentences, predict one of three labels).
  • 许可:无已知许可
  • 版本:1.0.0
  • 拆分
拆分 样本
'test' 750
'validation' 743
  • 特征
{
    "premise": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "hypothesis": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "label": {
        "num_classes": 3,
        "names": [
            "entailment",
            "neutral",
            "contradiction"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    }
}

cni

使用以下命令在 TFDS 中加载此数据集:

ds = tfds.load('huggingface:americas_nli/cni')
  • 说明
AmericasNLI is an extension of XNLI (Conneau et al., 2018) – a natural language inference (NLI) dataset covering 15 high-resource languages – to 10 low-resource indigenous languages spoken in the Americas: Ashaninka, Aymara, Bribri, Guarani, Nahuatl, Otomi, Quechua, Raramuri, Shipibo-Konibo, and Wixarika. As with MNLI, the goal is to predict textual entailment (does sentence A imply/contradict/neither sentence B) and is a classification task (given two sentences, predict one of three labels).
  • 许可:无已知许可
  • 版本:1.0.0
  • 拆分
拆分 样本
'test' 750
'validation' 658
  • 特征
{
    "premise": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "hypothesis": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "label": {
        "num_classes": 3,
        "names": [
            "entailment",
            "neutral",
            "contradiction"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    }
}

gn

使用以下命令在 TFDS 中加载此数据集:

ds = tfds.load('huggingface:americas_nli/gn')
  • 说明
AmericasNLI is an extension of XNLI (Conneau et al., 2018) – a natural language inference (NLI) dataset covering 15 high-resource languages – to 10 low-resource indigenous languages spoken in the Americas: Ashaninka, Aymara, Bribri, Guarani, Nahuatl, Otomi, Quechua, Raramuri, Shipibo-Konibo, and Wixarika. As with MNLI, the goal is to predict textual entailment (does sentence A imply/contradict/neither sentence B) and is a classification task (given two sentences, predict one of three labels).
  • 许可:无已知许可
  • 版本:1.0.0
  • 拆分
拆分 样本
'test' 750
'validation' 743
  • 特征
{
    "premise": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "hypothesis": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "label": {
        "num_classes": 3,
        "names": [
            "entailment",
            "neutral",
            "contradiction"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    }
}

hch

使用以下命令在 TFDS 中加载此数据集:

ds = tfds.load('huggingface:americas_nli/hch')
  • 说明
AmericasNLI is an extension of XNLI (Conneau et al., 2018) – a natural language inference (NLI) dataset covering 15 high-resource languages – to 10 low-resource indigenous languages spoken in the Americas: Ashaninka, Aymara, Bribri, Guarani, Nahuatl, Otomi, Quechua, Raramuri, Shipibo-Konibo, and Wixarika. As with MNLI, the goal is to predict textual entailment (does sentence A imply/contradict/neither sentence B) and is a classification task (given two sentences, predict one of three labels).
  • 许可:无已知许可
  • 版本:1.0.0
  • 拆分
拆分 样本
'test' 750
'validation' 743
  • 特征
{
    "premise": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "hypothesis": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "label": {
        "num_classes": 3,
        "names": [
            "entailment",
            "neutral",
            "contradiction"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    }
}

nah

使用以下命令在 TFDS 中加载此数据集:

ds = tfds.load('huggingface:americas_nli/nah')
  • 说明
AmericasNLI is an extension of XNLI (Conneau et al., 2018) – a natural language inference (NLI) dataset covering 15 high-resource languages – to 10 low-resource indigenous languages spoken in the Americas: Ashaninka, Aymara, Bribri, Guarani, Nahuatl, Otomi, Quechua, Raramuri, Shipibo-Konibo, and Wixarika. As with MNLI, the goal is to predict textual entailment (does sentence A imply/contradict/neither sentence B) and is a classification task (given two sentences, predict one of three labels).
  • 许可:无已知许可
  • 版本:1.0.0
  • 拆分
拆分 样本
'test' 738
'validation' 376
  • 特征
{
    "premise": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "hypothesis": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "label": {
        "num_classes": 3,
        "names": [
            "entailment",
            "neutral",
            "contradiction"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    }
}

oto

使用以下命令在 TFDS 中加载此数据集:

ds = tfds.load('huggingface:americas_nli/oto')
  • 说明
AmericasNLI is an extension of XNLI (Conneau et al., 2018) – a natural language inference (NLI) dataset covering 15 high-resource languages – to 10 low-resource indigenous languages spoken in the Americas: Ashaninka, Aymara, Bribri, Guarani, Nahuatl, Otomi, Quechua, Raramuri, Shipibo-Konibo, and Wixarika. As with MNLI, the goal is to predict textual entailment (does sentence A imply/contradict/neither sentence B) and is a classification task (given two sentences, predict one of three labels).
  • 许可:无已知许可
  • 版本:1.0.0
  • 拆分
拆分 样本
'test' 748
'validation' 222
  • 特征
{
    "premise": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "hypothesis": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "label": {
        "num_classes": 3,
        "names": [
            "entailment",
            "neutral",
            "contradiction"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    }
}

quy

使用以下命令在 TFDS 中加载此数据集:

ds = tfds.load('huggingface:americas_nli/quy')
  • 说明
AmericasNLI is an extension of XNLI (Conneau et al., 2018) – a natural language inference (NLI) dataset covering 15 high-resource languages – to 10 low-resource indigenous languages spoken in the Americas: Ashaninka, Aymara, Bribri, Guarani, Nahuatl, Otomi, Quechua, Raramuri, Shipibo-Konibo, and Wixarika. As with MNLI, the goal is to predict textual entailment (does sentence A imply/contradict/neither sentence B) and is a classification task (given two sentences, predict one of three labels).
  • 许可:无已知许可
  • 版本:1.0.0
  • 拆分
拆分 样本
'test' 750
'validation' 743
  • 特征
{
    "premise": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "hypothesis": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "label": {
        "num_classes": 3,
        "names": [
            "entailment",
            "neutral",
            "contradiction"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    }
}

shp

使用以下命令在 TFDS 中加载此数据集:

ds = tfds.load('huggingface:americas_nli/shp')
  • 说明
AmericasNLI is an extension of XNLI (Conneau et al., 2018) – a natural language inference (NLI) dataset covering 15 high-resource languages – to 10 low-resource indigenous languages spoken in the Americas: Ashaninka, Aymara, Bribri, Guarani, Nahuatl, Otomi, Quechua, Raramuri, Shipibo-Konibo, and Wixarika. As with MNLI, the goal is to predict textual entailment (does sentence A imply/contradict/neither sentence B) and is a classification task (given two sentences, predict one of three labels).
  • 许可:无已知许可
  • 版本:1.0.0
  • 拆分
拆分 样本
'test' 750
'validation' 743
  • 特征
{
    "premise": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "hypothesis": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "label": {
        "num_classes": 3,
        "names": [
            "entailment",
            "neutral",
            "contradiction"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    }
}

tar

使用以下命令在 TFDS 中加载此数据集:

ds = tfds.load('huggingface:americas_nli/tar')
  • 说明
AmericasNLI is an extension of XNLI (Conneau et al., 2018) – a natural language inference (NLI) dataset covering 15 high-resource languages – to 10 low-resource indigenous languages spoken in the Americas: Ashaninka, Aymara, Bribri, Guarani, Nahuatl, Otomi, Quechua, Raramuri, Shipibo-Konibo, and Wixarika. As with MNLI, the goal is to predict textual entailment (does sentence A imply/contradict/neither sentence B) and is a classification task (given two sentences, predict one of three labels).
  • 许可:无已知许可
  • 版本:1.0.0
  • 拆分
拆分 样本
'test' 750
'validation' 743
  • 特征
{
    "premise": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "hypothesis": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "label": {
        "num_classes": 3,
        "names": [
            "entailment",
            "neutral",
            "contradiction"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    }
}

all_languages

使用以下命令在 TFDS 中加载此数据集:

ds = tfds.load('huggingface:americas_nli/all_languages')
  • 说明
AmericasNLI is an extension of XNLI (Conneau et al., 2018) – a natural language inference (NLI) dataset covering 15 high-resource languages – to 10 low-resource indigenous languages spoken in the Americas: Ashaninka, Aymara, Bribri, Guarani, Nahuatl, Otomi, Quechua, Raramuri, Shipibo-Konibo, and Wixarika. As with MNLI, the goal is to predict textual entailment (does sentence A imply/contradict/neither sentence B) and is a classification task (given two sentences, predict one of three labels).
  • 许可:无已知许可
  • 版本:1.0.0
  • 拆分
拆分 样本
'test' 7486
'validation' 6457
  • 特征
{
    "language": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "premise": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "hypothesis": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "label": {
        "num_classes": 3,
        "names": [
            "entailment",
            "neutral",
            "contradiction"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    }
}