imprés

Referencias:

presuposición_todas_n_presuposición

Utilice el siguiente comando para cargar este conjunto de datos en TFDS:

ds = tfds.load('huggingface:imppres/presupposition_all_n_presupposition')
  • Descripción :
Over >25k semiautomatically generated sentence pairs illustrating well-studied pragmatic inference types. IMPPRES is an NLI dataset following the format of SNLI (Bowman et al., 2015), MultiNLI (Williams et al., 2018) and XNLI (Conneau et al., 2018), which was created to evaluate how well trained NLI models recognize several classes of presuppositions and scalar implicatures.
  • Licencia : Licencia pública internacional Creative Commons Atribución-No comercial 4.0
  • Versión : 1.1.0
  • Divisiones :
Dividir Ejemplos
'all_n_presupposition' 1900
  • Características :
{
    "premise": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "hypothesis": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "trigger": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "trigger1": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "trigger2": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "presupposition": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "gold_label": {
        "num_classes": 3,
        "names": [
            "entailment",
            "neutral",
            "contradiction"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    },
    "UID": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "pairID": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "paradigmID": {
        "dtype": "int16",
        "id": null,
        "_type": "Value"
    }
}

presuposición_ambas_presuposición

Utilice el siguiente comando para cargar este conjunto de datos en TFDS:

ds = tfds.load('huggingface:imppres/presupposition_both_presupposition')
  • Descripción :
Over >25k semiautomatically generated sentence pairs illustrating well-studied pragmatic inference types. IMPPRES is an NLI dataset following the format of SNLI (Bowman et al., 2015), MultiNLI (Williams et al., 2018) and XNLI (Conneau et al., 2018), which was created to evaluate how well trained NLI models recognize several classes of presuppositions and scalar implicatures.
  • Licencia : Licencia pública internacional Creative Commons Atribución-No comercial 4.0
  • Versión : 1.1.0
  • Divisiones :
Dividir Ejemplos
'both_presupposition' 1900
  • Características :
{
    "premise": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "hypothesis": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "trigger": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "trigger1": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "trigger2": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "presupposition": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "gold_label": {
        "num_classes": 3,
        "names": [
            "entailment",
            "neutral",
            "contradiction"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    },
    "UID": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "pairID": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "paradigmID": {
        "dtype": "int16",
        "id": null,
        "_type": "Value"
    }
}

presuposición_cambio_de_estado

Utilice el siguiente comando para cargar este conjunto de datos en TFDS:

ds = tfds.load('huggingface:imppres/presupposition_change_of_state')
  • Descripción :
Over >25k semiautomatically generated sentence pairs illustrating well-studied pragmatic inference types. IMPPRES is an NLI dataset following the format of SNLI (Bowman et al., 2015), MultiNLI (Williams et al., 2018) and XNLI (Conneau et al., 2018), which was created to evaluate how well trained NLI models recognize several classes of presuppositions and scalar implicatures.
  • Licencia : Licencia pública internacional Creative Commons Atribución-No comercial 4.0
  • Versión : 1.1.0
  • Divisiones :
Dividir Ejemplos
'change_of_state' 1900
  • Características :
{
    "premise": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "hypothesis": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "trigger": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "trigger1": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "trigger2": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "presupposition": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "gold_label": {
        "num_classes": 3,
        "names": [
            "entailment",
            "neutral",
            "contradiction"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    },
    "UID": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "pairID": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "paradigmID": {
        "dtype": "int16",
        "id": null,
        "_type": "Value"
    }
}

presuposición_hendidura_existencia

Utilice el siguiente comando para cargar este conjunto de datos en TFDS:

ds = tfds.load('huggingface:imppres/presupposition_cleft_existence')
  • Descripción :
Over >25k semiautomatically generated sentence pairs illustrating well-studied pragmatic inference types. IMPPRES is an NLI dataset following the format of SNLI (Bowman et al., 2015), MultiNLI (Williams et al., 2018) and XNLI (Conneau et al., 2018), which was created to evaluate how well trained NLI models recognize several classes of presuppositions and scalar implicatures.
  • Licencia : Licencia pública internacional Creative Commons Atribución-No comercial 4.0
  • Versión : 1.1.0
  • Divisiones :
Dividir Ejemplos
'cleft_existence' 1900
  • Características :
{
    "premise": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "hypothesis": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "trigger": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "trigger1": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "trigger2": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "presupposition": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "gold_label": {
        "num_classes": 3,
        "names": [
            "entailment",
            "neutral",
            "contradiction"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    },
    "UID": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "pairID": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "paradigmID": {
        "dtype": "int16",
        "id": null,
        "_type": "Value"
    }
}

presuposición_hendidura_uniqueness

Utilice el siguiente comando para cargar este conjunto de datos en TFDS:

ds = tfds.load('huggingface:imppres/presupposition_cleft_uniqueness')
  • Descripción :
Over >25k semiautomatically generated sentence pairs illustrating well-studied pragmatic inference types. IMPPRES is an NLI dataset following the format of SNLI (Bowman et al., 2015), MultiNLI (Williams et al., 2018) and XNLI (Conneau et al., 2018), which was created to evaluate how well trained NLI models recognize several classes of presuppositions and scalar implicatures.
  • Licencia : Licencia pública internacional Creative Commons Atribución-No comercial 4.0
  • Versión : 1.1.0
  • Divisiones :
Dividir Ejemplos
'cleft_uniqueness' 1900
  • Características :
{
    "premise": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "hypothesis": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "trigger": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "trigger1": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "trigger2": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "presupposition": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "gold_label": {
        "num_classes": 3,
        "names": [
            "entailment",
            "neutral",
            "contradiction"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    },
    "UID": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "pairID": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "paradigmID": {
        "dtype": "int16",
        "id": null,
        "_type": "Value"
    }
}

presuposición_solo_presuposición

Utilice el siguiente comando para cargar este conjunto de datos en TFDS:

ds = tfds.load('huggingface:imppres/presupposition_only_presupposition')
  • Descripción :
Over >25k semiautomatically generated sentence pairs illustrating well-studied pragmatic inference types. IMPPRES is an NLI dataset following the format of SNLI (Bowman et al., 2015), MultiNLI (Williams et al., 2018) and XNLI (Conneau et al., 2018), which was created to evaluate how well trained NLI models recognize several classes of presuppositions and scalar implicatures.
  • Licencia : Licencia pública internacional Creative Commons Atribución-No comercial 4.0
  • Versión : 1.1.0
  • Divisiones :
Dividir Ejemplos
'only_presupposition' 1900
  • Características :
{
    "premise": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "hypothesis": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "trigger": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "trigger1": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "trigger2": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "presupposition": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "gold_label": {
        "num_classes": 3,
        "names": [
            "entailment",
            "neutral",
            "contradiction"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    },
    "UID": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "pairID": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "paradigmID": {
        "dtype": "int16",
        "id": null,
        "_type": "Value"
    }
}

presuposición_poseída_existencia_definida

Utilice el siguiente comando para cargar este conjunto de datos en TFDS:

ds = tfds.load('huggingface:imppres/presupposition_possessed_definites_existence')
  • Descripción :
Over >25k semiautomatically generated sentence pairs illustrating well-studied pragmatic inference types. IMPPRES is an NLI dataset following the format of SNLI (Bowman et al., 2015), MultiNLI (Williams et al., 2018) and XNLI (Conneau et al., 2018), which was created to evaluate how well trained NLI models recognize several classes of presuppositions and scalar implicatures.
  • Licencia : Licencia pública internacional Creative Commons Atribución-No comercial 4.0
  • Versión : 1.1.0
  • Divisiones :
Dividir Ejemplos
'possessed_definites_existence' 1900
  • Características :
{
    "premise": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "hypothesis": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "trigger": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "trigger1": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "trigger2": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "presupposition": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "gold_label": {
        "num_classes": 3,
        "names": [
            "entailment",
            "neutral",
            "contradiction"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    },
    "UID": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "pairID": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "paradigmID": {
        "dtype": "int16",
        "id": null,
        "_type": "Value"
    }
}

presuposición_possessed_definites_uniqueness

Utilice el siguiente comando para cargar este conjunto de datos en TFDS:

ds = tfds.load('huggingface:imppres/presupposition_possessed_definites_uniqueness')
  • Descripción :
Over >25k semiautomatically generated sentence pairs illustrating well-studied pragmatic inference types. IMPPRES is an NLI dataset following the format of SNLI (Bowman et al., 2015), MultiNLI (Williams et al., 2018) and XNLI (Conneau et al., 2018), which was created to evaluate how well trained NLI models recognize several classes of presuppositions and scalar implicatures.
  • Licencia : Licencia pública internacional Creative Commons Atribución-No comercial 4.0
  • Versión : 1.1.0
  • Divisiones :
Dividir Ejemplos
'possessed_definites_uniqueness' 1900
  • Características :
{
    "premise": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "hypothesis": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "trigger": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "trigger1": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "trigger2": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "presupposition": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "gold_label": {
        "num_classes": 3,
        "names": [
            "entailment",
            "neutral",
            "contradiction"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    },
    "UID": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "pairID": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "paradigmID": {
        "dtype": "int16",
        "id": null,
        "_type": "Value"
    }
}

presuposición_pregunta_presuposición

Utilice el siguiente comando para cargar este conjunto de datos en TFDS:

ds = tfds.load('huggingface:imppres/presupposition_question_presupposition')
  • Descripción :
Over >25k semiautomatically generated sentence pairs illustrating well-studied pragmatic inference types. IMPPRES is an NLI dataset following the format of SNLI (Bowman et al., 2015), MultiNLI (Williams et al., 2018) and XNLI (Conneau et al., 2018), which was created to evaluate how well trained NLI models recognize several classes of presuppositions and scalar implicatures.
  • Licencia : Licencia pública internacional Creative Commons Atribución-No comercial 4.0
  • Versión : 1.1.0
  • Divisiones :
Dividir Ejemplos
'question_presupposition' 1900
  • Características :
{
    "premise": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "hypothesis": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "trigger": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "trigger1": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "trigger2": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "presupposition": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "gold_label": {
        "num_classes": 3,
        "names": [
            "entailment",
            "neutral",
            "contradiction"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    },
    "UID": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "pairID": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "paradigmID": {
        "dtype": "int16",
        "id": null,
        "_type": "Value"
    }
}

implicaturas_conectivas

Utilice el siguiente comando para cargar este conjunto de datos en TFDS:

ds = tfds.load('huggingface:imppres/implicature_connectives')
  • Descripción :
Over >25k semiautomatically generated sentence pairs illustrating well-studied pragmatic inference types. IMPPRES is an NLI dataset following the format of SNLI (Bowman et al., 2015), MultiNLI (Williams et al., 2018) and XNLI (Conneau et al., 2018), which was created to evaluate how well trained NLI models recognize several classes of presuppositions and scalar implicatures.
  • Licencia : Licencia pública internacional Creative Commons Atribución-No comercial 4.0
  • Versión : 1.1.0
  • Divisiones :
Dividir Ejemplos
'connectives' 1200
  • Características :
{
    "premise": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "hypothesis": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "gold_label_log": {
        "num_classes": 3,
        "names": [
            "entailment",
            "neutral",
            "contradiction"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    },
    "gold_label_prag": {
        "num_classes": 3,
        "names": [
            "entailment",
            "neutral",
            "contradiction"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    },
    "spec_relation": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "item_type": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "trigger": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "lexemes": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

implicatura_gradable_adjetivo

Utilice el siguiente comando para cargar este conjunto de datos en TFDS:

ds = tfds.load('huggingface:imppres/implicature_gradable_adjective')
  • Descripción :
Over >25k semiautomatically generated sentence pairs illustrating well-studied pragmatic inference types. IMPPRES is an NLI dataset following the format of SNLI (Bowman et al., 2015), MultiNLI (Williams et al., 2018) and XNLI (Conneau et al., 2018), which was created to evaluate how well trained NLI models recognize several classes of presuppositions and scalar implicatures.
  • Licencia : Licencia pública internacional Creative Commons Atribución-No comercial 4.0
  • Versión : 1.1.0
  • Divisiones :
Dividir Ejemplos
'gradable_adjective' 1200
  • Características :
{
    "premise": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "hypothesis": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "gold_label_log": {
        "num_classes": 3,
        "names": [
            "entailment",
            "neutral",
            "contradiction"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    },
    "gold_label_prag": {
        "num_classes": 3,
        "names": [
            "entailment",
            "neutral",
            "contradiction"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    },
    "spec_relation": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "item_type": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "trigger": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "lexemes": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

implicatura_gradable_verbo

Utilice el siguiente comando para cargar este conjunto de datos en TFDS:

ds = tfds.load('huggingface:imppres/implicature_gradable_verb')
  • Descripción :
Over >25k semiautomatically generated sentence pairs illustrating well-studied pragmatic inference types. IMPPRES is an NLI dataset following the format of SNLI (Bowman et al., 2015), MultiNLI (Williams et al., 2018) and XNLI (Conneau et al., 2018), which was created to evaluate how well trained NLI models recognize several classes of presuppositions and scalar implicatures.
  • Licencia : Licencia pública internacional Creative Commons Atribución-No comercial 4.0
  • Versión : 1.1.0
  • Divisiones :
Dividir Ejemplos
'gradable_verb' 1200
  • Características :
{
    "premise": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "hypothesis": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "gold_label_log": {
        "num_classes": 3,
        "names": [
            "entailment",
            "neutral",
            "contradiction"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    },
    "gold_label_prag": {
        "num_classes": 3,
        "names": [
            "entailment",
            "neutral",
            "contradiction"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    },
    "spec_relation": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "item_type": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "trigger": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "lexemes": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

implicatura_modales

Utilice el siguiente comando para cargar este conjunto de datos en TFDS:

ds = tfds.load('huggingface:imppres/implicature_modals')
  • Descripción :
Over >25k semiautomatically generated sentence pairs illustrating well-studied pragmatic inference types. IMPPRES is an NLI dataset following the format of SNLI (Bowman et al., 2015), MultiNLI (Williams et al., 2018) and XNLI (Conneau et al., 2018), which was created to evaluate how well trained NLI models recognize several classes of presuppositions and scalar implicatures.
  • Licencia : Licencia pública internacional Creative Commons Atribución-No comercial 4.0
  • Versión : 1.1.0
  • Divisiones :
Dividir Ejemplos
'modals' 1200
  • Características :
{
    "premise": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "hypothesis": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "gold_label_log": {
        "num_classes": 3,
        "names": [
            "entailment",
            "neutral",
            "contradiction"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    },
    "gold_label_prag": {
        "num_classes": 3,
        "names": [
            "entailment",
            "neutral",
            "contradiction"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    },
    "spec_relation": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "item_type": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "trigger": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "lexemes": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

implicatura_numerales_10_100

Utilice el siguiente comando para cargar este conjunto de datos en TFDS:

ds = tfds.load('huggingface:imppres/implicature_numerals_10_100')
  • Descripción :
Over >25k semiautomatically generated sentence pairs illustrating well-studied pragmatic inference types. IMPPRES is an NLI dataset following the format of SNLI (Bowman et al., 2015), MultiNLI (Williams et al., 2018) and XNLI (Conneau et al., 2018), which was created to evaluate how well trained NLI models recognize several classes of presuppositions and scalar implicatures.
  • Licencia : Licencia pública internacional Creative Commons Atribución-No comercial 4.0
  • Versión : 1.1.0
  • Divisiones :
Dividir Ejemplos
'numerals_10_100' 1200
  • Características :
{
    "premise": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "hypothesis": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "gold_label_log": {
        "num_classes": 3,
        "names": [
            "entailment",
            "neutral",
            "contradiction"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    },
    "gold_label_prag": {
        "num_classes": 3,
        "names": [
            "entailment",
            "neutral",
            "contradiction"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    },
    "spec_relation": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "item_type": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "trigger": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "lexemes": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

implicatura_numerales_2_3

Utilice el siguiente comando para cargar este conjunto de datos en TFDS:

ds = tfds.load('huggingface:imppres/implicature_numerals_2_3')
  • Descripción :
Over >25k semiautomatically generated sentence pairs illustrating well-studied pragmatic inference types. IMPPRES is an NLI dataset following the format of SNLI (Bowman et al., 2015), MultiNLI (Williams et al., 2018) and XNLI (Conneau et al., 2018), which was created to evaluate how well trained NLI models recognize several classes of presuppositions and scalar implicatures.
  • Licencia : Licencia pública internacional Creative Commons Atribución-No comercial 4.0
  • Versión : 1.1.0
  • Divisiones :
Dividir Ejemplos
'numerals_2_3' 1200
  • Características :
{
    "premise": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "hypothesis": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "gold_label_log": {
        "num_classes": 3,
        "names": [
            "entailment",
            "neutral",
            "contradiction"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    },
    "gold_label_prag": {
        "num_classes": 3,
        "names": [
            "entailment",
            "neutral",
            "contradiction"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    },
    "spec_relation": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "item_type": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "trigger": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "lexemes": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

implicaturas_cuantificadores

Utilice el siguiente comando para cargar este conjunto de datos en TFDS:

ds = tfds.load('huggingface:imppres/implicature_quantifiers')
  • Descripción :
Over >25k semiautomatically generated sentence pairs illustrating well-studied pragmatic inference types. IMPPRES is an NLI dataset following the format of SNLI (Bowman et al., 2015), MultiNLI (Williams et al., 2018) and XNLI (Conneau et al., 2018), which was created to evaluate how well trained NLI models recognize several classes of presuppositions and scalar implicatures.
  • Licencia : Licencia pública internacional Creative Commons Atribución-No comercial 4.0
  • Versión : 1.1.0
  • Divisiones :
Dividir Ejemplos
'quantifiers' 1200
  • Características :
{
    "premise": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "hypothesis": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "gold_label_log": {
        "num_classes": 3,
        "names": [
            "entailment",
            "neutral",
            "contradiction"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    },
    "gold_label_prag": {
        "num_classes": 3,
        "names": [
            "entailment",
            "neutral",
            "contradiction"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    },
    "spec_relation": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "item_type": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "trigger": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "lexemes": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}