Attend the Women in ML Symposium on December 7 Register now

pragmeval

References:

verifiability

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:pragmeval/verifiability')
  • Description:
Evaluation of language understanding with a 11 datasets benchmark focusing on discourse and pragmatics
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'test' 2424
'train' 5712
'validation' 634
  • Features:
{
    "sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "label": {
        "num_classes": 3,
        "names": [
            "experiential",
            "unverifiable",
            "non-experiential"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    },
    "idx": {
        "dtype": "int32",
        "id": null,
        "_type": "Value"
    }
}

emobank-arousal

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:pragmeval/emobank-arousal')
  • Description:
Evaluation of language understanding with a 11 datasets benchmark focusing on discourse and pragmatics
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'test' 683
'train' 5470
'validation' 684
  • Features:
{
    "sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "label": {
        "num_classes": 2,
        "names": [
            "low",
            "high"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    },
    "idx": {
        "dtype": "int32",
        "id": null,
        "_type": "Value"
    }
}

switchboard

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:pragmeval/switchboard')
  • Description:
Evaluation of language understanding with a 11 datasets benchmark focusing on discourse and pragmatics
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'test' 649
'train' 18930
'validation' 2113
  • Features:
{
    "sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "label": {
        "num_classes": 41,
        "names": [
            "Response Acknowledgement",
            "Uninterpretable",
            "Or-Clause",
            "Reject",
            "Statement-non-opinion",
            "3rd-party-talk",
            "Repeat-phrase",
            "Hold Before Answer/Agreement",
            "Signal-non-understanding",
            "Offers, Options Commits",
            "Agree/Accept",
            "Dispreferred Answers",
            "Hedge",
            "Action-directive",
            "Tag-Question",
            "Self-talk",
            "Yes-No-Question",
            "Rhetorical-Question",
            "No Answers",
            "Open-Question",
            "Conventional-closing",
            "Other Answers",
            "Acknowledge (Backchannel)",
            "Wh-Question",
            "Declarative Wh-Question",
            "Thanking",
            "Yes Answers",
            "Affirmative Non-yes Answers",
            "Declarative Yes-No-Question",
            "Backchannel in Question Form",
            "Apology",
            "Downplayer",
            "Conventional-opening",
            "Collaborative Completion",
            "Summarize/Reformulate",
            "Negative Non-no Answers",
            "Statement-opinion",
            "Appreciation",
            "Other",
            "Quotation",
            "Maybe/Accept-part"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    },
    "idx": {
        "dtype": "int32",
        "id": null,
        "_type": "Value"
    }
}

persuasiveness-eloquence

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:pragmeval/persuasiveness-eloquence')
  • Description:
Evaluation of language understanding with a 11 datasets benchmark focusing on discourse and pragmatics
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'test' 90
'train' 725
'validation' 91
  • Features:
{
    "sentence1": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence2": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "label": {
        "num_classes": 2,
        "names": [
            "low",
            "high"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    },
    "idx": {
        "dtype": "int32",
        "id": null,
        "_type": "Value"
    }
}

mrda

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:pragmeval/mrda')
  • Description:
Evaluation of language understanding with a 11 datasets benchmark focusing on discourse and pragmatics
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'test' 6459
'train' 14484
'validation' 1630
  • Features:
{
    "sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "label": {
        "num_classes": 51,
        "names": [
            "Declarative-Question",
            "Statement",
            "Reject",
            "Or-Clause",
            "3rd-party-talk",
            "Continuer",
            "Hold Before Answer/Agreement",
            "Assessment/Appreciation",
            "Signal-non-understanding",
            "Floor Holder",
            "Sympathy",
            "Dispreferred Answers",
            "Reformulate/Summarize",
            "Exclamation",
            "Interrupted/Abandoned/Uninterpretable",
            "Expansions of y/n Answers",
            "Action-directive",
            "Tag-Question",
            "Accept",
            "Rhetorical-question Continue",
            "Self-talk",
            "Rhetorical-Question",
            "Yes-No-question",
            "Open-Question",
            "Rising Tone",
            "Other Answers",
            "Commit",
            "Wh-Question",
            "Repeat",
            "Follow Me",
            "Thanking",
            "Offer",
            "About-task",
            "Reject-part",
            "Affirmative Non-yes Answers",
            "Apology",
            "Downplayer",
            "Humorous Material",
            "Accept-part",
            "Collaborative Completion",
            "Mimic Other",
            "Understanding Check",
            "Misspeak Self-Correction",
            "Or-Question",
            "Topic Change",
            "Negative Non-no Answers",
            "Floor Grabber",
            "Correct-misspeaking",
            "Maybe",
            "Acknowledge-answer",
            "Defending/Explanation"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    },
    "idx": {
        "dtype": "int32",
        "id": null,
        "_type": "Value"
    }
}

gum

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:pragmeval/gum')
  • Description:
Evaluation of language understanding with a 11 datasets benchmark focusing on discourse and pragmatics
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'test' 248
'train' 1700
'validation' 259
  • Features:
{
    "sentence1": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence2": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "label": {
        "num_classes": 17,
        "names": [
            "preparation",
            "evaluation",
            "circumstance",
            "solutionhood",
            "justify",
            "result",
            "evidence",
            "purpose",
            "concession",
            "elaboration",
            "background",
            "condition",
            "cause",
            "restatement",
            "motivation",
            "antithesis",
            "no_relation"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    },
    "idx": {
        "dtype": "int32",
        "id": null,
        "_type": "Value"
    }
}

emergent

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:pragmeval/emergent')
  • Description:
Evaluation of language understanding with a 11 datasets benchmark focusing on discourse and pragmatics
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'test' 259
'train' 2076
'validation' 259
  • Features:
{
    "sentence1": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence2": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "label": {
        "num_classes": 3,
        "names": [
            "observing",
            "for",
            "against"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    },
    "idx": {
        "dtype": "int32",
        "id": null,
        "_type": "Value"
    }
}

persuasiveness-relevance

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:pragmeval/persuasiveness-relevance')
  • Description:
Evaluation of language understanding with a 11 datasets benchmark focusing on discourse and pragmatics
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'test' 90
'train' 725
'validation' 91
  • Features:
{
    "sentence1": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence2": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "label": {
        "num_classes": 2,
        "names": [
            "low",
            "high"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    },
    "idx": {
        "dtype": "int32",
        "id": null,
        "_type": "Value"
    }
}

persuasiveness-specificity

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:pragmeval/persuasiveness-specificity')
  • Description:
Evaluation of language understanding with a 11 datasets benchmark focusing on discourse and pragmatics
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'test' 62
'train' 504
'validation' 62
  • Features:
{
    "sentence1": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence2": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "label": {
        "num_classes": 2,
        "names": [
            "low",
            "high"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    },
    "idx": {
        "dtype": "int32",
        "id": null,
        "_type": "Value"
    }
}

persuasiveness-strength

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:pragmeval/persuasiveness-strength')
  • Description:
Evaluation of language understanding with a 11 datasets benchmark focusing on discourse and pragmatics
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'test' 46
'train' 371
'validation' 46
  • Features:
{
    "sentence1": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence2": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "label": {
        "num_classes": 2,
        "names": [
            "low",
            "high"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    },
    "idx": {
        "dtype": "int32",
        "id": null,
        "_type": "Value"
    }
}

emobank-dominance

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:pragmeval/emobank-dominance')
  • Description:
Evaluation of language understanding with a 11 datasets benchmark focusing on discourse and pragmatics
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'test' 798
'train' 6392
'validation' 798
  • Features:
{
    "sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "label": {
        "num_classes": 2,
        "names": [
            "low",
            "high"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    },
    "idx": {
        "dtype": "int32",
        "id": null,
        "_type": "Value"
    }
}

squinky-implicature

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:pragmeval/squinky-implicature')
  • Description:
Evaluation of language understanding with a 11 datasets benchmark focusing on discourse and pragmatics
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'test' 465
'train' 3724
'validation' 465
  • Features:
{
    "sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "label": {
        "num_classes": 2,
        "names": [
            "low",
            "high"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    },
    "idx": {
        "dtype": "int32",
        "id": null,
        "_type": "Value"
    }
}

sarcasm

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:pragmeval/sarcasm')
  • Description:
Evaluation of language understanding with a 11 datasets benchmark focusing on discourse and pragmatics
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'test' 469
'train' 3754
'validation' 469
  • Features:
{
    "sentence1": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence2": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "label": {
        "num_classes": 2,
        "names": [
            "notsarc",
            "sarc"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    },
    "idx": {
        "dtype": "int32",
        "id": null,
        "_type": "Value"
    }
}

squinky-formality

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:pragmeval/squinky-formality')
  • Description:
Evaluation of language understanding with a 11 datasets benchmark focusing on discourse and pragmatics
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'test' 452
'train' 3622
'validation' 453
  • Features:
{
    "sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "label": {
        "num_classes": 2,
        "names": [
            "low",
            "high"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    },
    "idx": {
        "dtype": "int32",
        "id": null,
        "_type": "Value"
    }
}

stac

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:pragmeval/stac')
  • Description:
Evaluation of language understanding with a 11 datasets benchmark focusing on discourse and pragmatics
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'test' 1304
'train' 11230
'validation' 1247
  • Features:
{
    "sentence1": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence2": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "label": {
        "num_classes": 18,
        "names": [
            "Comment",
            "Contrast",
            "Q_Elab",
            "Parallel",
            "Explanation",
            "Narration",
            "Continuation",
            "Result",
            "Acknowledgement",
            "Alternation",
            "Question_answer_pair",
            "Correction",
            "Clarification_question",
            "Conditional",
            "Sequence",
            "Elaboration",
            "Background",
            "no_relation"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    },
    "idx": {
        "dtype": "int32",
        "id": null,
        "_type": "Value"
    }
}

pdtb

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:pragmeval/pdtb')
  • Description:
Evaluation of language understanding with a 11 datasets benchmark focusing on discourse and pragmatics
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'test' 1085
'train' 12907
'validation' 1204
  • Features:
{
    "sentence1": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence2": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "label": {
        "num_classes": 16,
        "names": [
            "Synchrony",
            "Contrast",
            "Asynchronous",
            "Conjunction",
            "List",
            "Condition",
            "Pragmatic concession",
            "Restatement",
            "Pragmatic cause",
            "Alternative",
            "Pragmatic condition",
            "Pragmatic contrast",
            "Instantiation",
            "Exception",
            "Cause",
            "Concession"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    },
    "idx": {
        "dtype": "int32",
        "id": null,
        "_type": "Value"
    }
}

persuasiveness-premisetype

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:pragmeval/persuasiveness-premisetype')
  • Description:
Evaluation of language understanding with a 11 datasets benchmark focusing on discourse and pragmatics
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'test' 70
'train' 566
'validation' 71
  • Features:
{
    "sentence1": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence2": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "label": {
        "num_classes": 8,
        "names": [
            "testimony",
            "warrant",
            "invented_instance",
            "common_knowledge",
            "statistics",
            "analogy",
            "definition",
            "real_example"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    },
    "idx": {
        "dtype": "int32",
        "id": null,
        "_type": "Value"
    }
}

squinky-informativeness

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:pragmeval/squinky-informativeness')
  • Description:
Evaluation of language understanding with a 11 datasets benchmark focusing on discourse and pragmatics
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'test' 464
'train' 3719
'validation' 465
  • Features:
{
    "sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "label": {
        "num_classes": 2,
        "names": [
            "low",
            "high"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    },
    "idx": {
        "dtype": "int32",
        "id": null,
        "_type": "Value"
    }
}

persuasiveness-claimtype

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:pragmeval/persuasiveness-claimtype')
  • Description:
Evaluation of language understanding with a 11 datasets benchmark focusing on discourse and pragmatics
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'test' 19
'train' 160
'validation' 20
  • Features:
{
    "sentence1": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence2": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "label": {
        "num_classes": 3,
        "names": [
            "Value",
            "Fact",
            "Policy"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    },
    "idx": {
        "dtype": "int32",
        "id": null,
        "_type": "Value"
    }
}

emobank-valence

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:pragmeval/emobank-valence')
  • Description:
Evaluation of language understanding with a 11 datasets benchmark focusing on discourse and pragmatics
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'test' 643
'train' 5150
'validation' 644
  • Features:
{
    "sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "label": {
        "num_classes": 2,
        "names": [
            "low",
            "high"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    },
    "idx": {
        "dtype": "int32",
        "id": null,
        "_type": "Value"
    }
}