Jsevisal/gesture_pred

该数据集包含手势预测任务的数据，涵盖了多种手势和情感表达，如BUT、CALM_DOWN、COME_ON等。每个标签都有B-和I-两种形式，表示开始和内部。数据集分为训练集、测试集和验证集，分别包含2339、130和130个样本。任务类别为token分类，语言为英语。

This dataset contains data for the gesture prediction task, covering various gestures and emotional expressions such as BUT, CALM_DOWN, COME_ON, etc. Each label has two forms: B- and I-, which represent the beginning and inside of a token respectively. The dataset is divided into training, test, and validation sets, containing 2339, 130, and 130 samples respectively. The task category is token classification, and the language used is English.

原始信息汇总

数据集概述

数据集名称

• 名称：gesture_pred

数据集特征

• id: 整数类型 (int64)
• tokens: 字符串序列
• gestures: 字符串序列
• label: 分类标签序列
  • 类别名称:
    • 0: B-BUT
    • 1: I-BUT
    • 2: B-CALM_DOWN
    • 3: I-CALM_DOWN
    • 4: B-COME_ON
    • 5: I-COME_ON
    • 6: B-EMPHATIC
    • 7: I-EMPHATIC
    • 8: B-ENTHUSIASTIC
    • 9: I-ENTHUSIASTIC
    • 10: B-EXPLAIN
    • 11: I-EXPLAIN
    • 12: B-FRONT
    • 13: I-FRONT
    • 14: B-GREET
    • 15: I-GREET
    • 16: B-ITERATE
    • 17: I-ITERATE
    • 18: B-NEUTRAL
    • 19: I-NEUTRAL
    • 20: B-NO
    • 21: I-NO
    • 22: B-NO_GESTURE
    • 23: I-NO_GESTURE
    • 24: B-OTHER_PEER
    • 25: I-OTHER_PEER
    • 26: B-PLEASE
    • 27: I-PLEASE
    • 28: B-QUESTION
    • 29: I-QUESTION
    • 30: B-SELF
    • 31: I-SELF
    • 32: B-SORRY
    • 33: I-SORRY
    • 34: B-THANKS
    • 35: I-THANKS
    • 36: B-THINKING
    • 37: I-THINKING
    • 38: B-THIRD_PERSON
    • 39: I-THIRD_PERSON
    • 40: B-YES
    • 41: I-YES

数据集划分

• train: 2339个样本，714214字节
• test: 130个样本，40730字节
• validation: 130个样本，41891字节

数据集大小

• 下载大小: 140110字节
• 数据集总大小: 796835字节

任务类别

• 任务类型: 词元分类

语言

• 语言: 英语 (en)