EphronM/usa_passports_ocr_funsd

---
dataset_info:
features:
- name: name
dtype: string
- name: page_no
dtype: int64
- name: width
dtype: int64
- name: height
dtype: int64
- name: text
sequence: string
- name: bbox
sequence:
sequence: int64
- name: segment_bbox
sequence:
sequence: int64
- name: segment_id
sequence: int64
- name: qas
struct:
- name: answers
list:
- name: answer_end
sequence: int64
- name: answer_start
sequence: int64
- name: text
sequence: string
- name: question
sequence: string
- name: question_id
sequence: int64
- name: image
dtype: string
- name: md5sum
dtype: string
splits:
- name: train
num_bytes: 235094122
num_examples: 72
- name: test
num_bytes: 65985649
num_examples: 20
download_size: 300115943
dataset_size: 301079771
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*
---

dataset_info:
features:
- name: 名称
dtype: 字符串
- name: 页码
dtype: int64
- name: 宽度
dtype: int64
- name: 高度
dtype: int64
- name: 文本
sequence: 字符串
- name: 边界框（bbox）
sequence:
sequence: int64
- name: 段边界框（segment_bbox）
sequence:
sequence: int64
- name: 段ID（segment_id）
sequence: int64
- name: 问答对（qas）
struct:
- name: 答案列表
list:
- name: 答案结束位置
sequence: int64
- name: 答案开始位置
sequence: int64
- name: 答案文本
sequence: 字符串
- name: 问题
sequence: 字符串
- name: 问题ID
sequence: int64
- name: 图片路径
dtype: 字符串
- name: MD5校验和
dtype: 字符串
splits:
- name: 训练集
num_bytes: 235094122
num_examples: 72
- name: 测试集
num_bytes: 65985649
num_examples: 20
download_size: 300115943
dataset_size: 301079771
configs:
- config_name: 默认配置
data_files:
- split: 训练集
path: data/train-*
- split: 测试集
path: data/test-*

原始信息汇总

数据集概述

数据集特征

• name: 字符串类型
• page_no: 整数类型
• width: 整数类型
• height: 整数类型
• text: 字符串序列
• bbox: 整数序列的序列
• segment_bbox: 整数序列的序列
• segment_id: 整数序列
• qas: 结构体，包含：
  • answers: 列表，包含：
    • answer_end: 整数序列
    • answer_start: 整数序列
    • text: 字符串序列
  • question: 字符串序列
  • question_id: 整数序列
• image: 字符串类型
• md5sum: 字符串类型

数据集分割

• train: 72个样本，占用235094122字节
• test: 20个样本，占用65985649字节

数据集大小

• 下载大小: 300115943字节
• 数据集大小: 301079771字节

数据文件配置

• default 配置:
  • train: 路径为 data/train-*
  • test: 路径为 data/test-*