metadata
dataset_info:
features:
- name: input_text
dtype: string
- name: question_id
dtype: int64
- name: answer
dtype: string
- name: answer_type
dtype: int64
- name: image
dtype: image
- name: pdf_filepath
dtype: string
- name: question_page_number
dtype: string
splits:
- name: test
num_bytes: 1461659890.23
num_examples: 1155
download_size: 1190766680
dataset_size: 1461659890.23
configs:
- config_name: default
data_files:
- split: test
path: data/test-*
license: cc-by-sa-4.0
language:
- ja
source_datasets:
- shunk031/JDocQA
JDocQA_SingleImage Dataset
Dataset Summary
JDocQA_SingleImageは、shunk031/JDocQAのtestサブセットを基に作成されたデータセットで、PDFファイルを200dpiの画像に変換し、画像が取得できない設問と複数画像が必要な設問を除外しています。元のデータセットが37GBと大きすぎるため、サイズを削減しつつ実用性を保つことを目的としています。
- 元データ:
shunk031/JDocQA(test split, 1,176 instances) - 変換後: PDFを画像に置き換え、画像が取得できてかつ単一画像入力の設問のみを含む
- 言語: 日本語 (BCP-47 ja-JP)
Data Fields
input_text: 質問テキストquestion_id: ユニークなIDanswer: 回答answer_type: 回答タイプ（0: Yes/No, 1: Factoid, 2: Numerical, 3: Open-ended）image: 200dpiのPNG画像（バイト形式）pdf_filepath: 元のPDFファイルパス（デバッグ用）question_page_number: 質問に関連するページ番号（デバッグ用）
Usage
from datasets import load_dataset
dataset = load_dataset("your_username/JDocQA_Image", split="test")
Additional Information
Dataset Curators
[More Information Needed]
Licensing Information
From JDocQA's README.md:
JDocQA dataset annotations are distributed under CC BY-SA 4.0. We are delighted to see many derivations from JDocQA! When you create any derivations, e.g., datasets, papers, etc, from JDocQA, please cite our paper accordingly. If your derivations are web-based projects, please cite our paper and include the link to this github page.
Citation Information
@inproceedings{onami2024jdocqa,
title={JDocQA: Japanese Document Question Answering Dataset for Generative Language Models},
author={Onami, Eri and Kurita, Shuhei and Miyanishi, Taiki and Watanabe, Taro},
booktitle={Proceedings of the 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation (LREC-COLING 2024)},
pages={9503--9514},
year={2024}
}