five

cjfcsjt/MOTIF_automation_preprocess

收藏
Hugging Face2024-05-05 更新2024-06-12 收录
下载链接:
https://hf-mirror.com/datasets/cjfcsjt/MOTIF_automation_preprocess
下载链接
链接失效反馈
官方服务:
资源简介:
--- dataset_info: features: - name: image struct: - name: bytes dtype: binary - name: path dtype: 'null' - name: image_id dtype: string - name: verb_str dtype: string - name: action dtype: string - name: instr dtype: string - name: trace_id dtype: string - name: goal dtype: string - name: app dtype: string - name: screen_w dtype: int64 - name: screen_h dtype: int64 - name: vh_w dtype: int64 - name: vh_h dtype: int64 - name: scale_x dtype: float64 - name: scale_y dtype: float64 - name: types_mc sequence: int64 - name: desc_mc sequence: string - name: bbox_mc sequence: sequence: float64 - name: __index_level_0__ dtype: string - name: __index_level_1__ dtype: int64 splits: - name: train_au_tu num_bytes: 3754647103 num_examples: 8388 - name: test_au_tu num_bytes: 418681835 num_examples: 1019 download_size: 4299361585 dataset_size: 4173328938 configs: - config_name: default data_files: - split: test_au_tu path: data/test_au_tu-* - split: train_au_tu path: data/train_au_tu-* ---
提供机构:
cjfcsjt
原始信息汇总

数据集概述

数据集特征

  • image
    • bytes: binary
    • path: null
  • image_id: string
  • verb_str: string
  • action: string
  • instr: string
  • trace_id: string
  • goal: string
  • app: string
  • screen_w: int64
  • screen_h: int64
  • vh_w: int64
  • vh_h: int64
  • scale_x: float64
  • scale_y: float64
  • types_mc: sequence of int64
  • desc_mc: sequence of string
  • bbox_mc: sequence of sequence of float64
  • index_level_0: string
  • index_level_1: int64

数据集分割

  • train_au_tu
    • num_bytes: 3754647103
    • num_examples: 8388
  • test_au_tu
    • num_bytes: 418681835
    • num_examples: 1019

数据集大小

  • download_size: 4299361585
  • dataset_size: 4173328938

配置文件

  • config_name: default
    • data_files
      • split: test_au_tu
        • path: data/test_au_tu-*
      • split: train_au_tu
        • path: data/train_au_tu-*
5,000+
优质数据集
54 个
任务类型
进入经典数据集
二维码
社区交流群

面向社区/商业的数据集话题

二维码
科研交流群

面向高校/科研机构的开源数据集话题

数据驱动未来

携手共赢发展

商业合作