Ssunbell/boostcamp-docvqa-marker

---
dataset_info:
features:
- name: questionId
dtype: int64
- name: question
dtype: string
- name: image
sequence:
sequence:
sequence: uint8
- name: docId
dtype: int64
- name: ucsf_document_id
dtype: string
- name: ucsf_document_page_no
dtype: string
- name: answers
sequence: string
- name: data_split
dtype: string
- name: words
sequence: string
- name: boxes
sequence:
sequence: int64
- name: question_m
dtype: string
splits:
- name: train
num_bytes: 6384754744
num_examples: 39454
- name: val
num_bytes: 869779916
num_examples: 5349
download_size: 2580109440
dataset_size: 7254534660
---
# Dataset Card for "boostcamp-docvqa-marker"

[More Information needed](https://github.com/huggingface/datasets/blob/main/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)

原始信息汇总

数据集概述

数据集信息

特征

• questionId: 整数类型
• question: 字符串类型
• image: 序列类型，包含多个无符号8位整数
• docId: 整数类型
• ucsf_document_id: 字符串类型
• ucsf_document_page_no: 字符串类型
• answers: 字符串序列类型
• data_split: 字符串类型
• words: 字符串序列类型
• boxes: 序列类型，包含多个整数序列
• question_m: 字符串类型

数据分割

• train: 包含39454个样本，大小为6384754744字节
• val: 包含5349个样本，大小为869779916字节

数据集大小

• 下载大小: 2580109440字节
• 数据集总大小: 7254534660字节