Seongill/Trivia_missing_5_similar_substitution_full
收藏Hugging Face2023-12-13 更新2024-03-04 收录
下载链接:
https://hf-mirror.com/datasets/Seongill/Trivia_missing_5_similar_substitution_full
下载链接
链接失效反馈官方服务:
资源简介:
---
dataset_info:
features:
- name: question
dtype: string
- name: answers
sequence: string
- name: ctxs
list:
- name: answer_sent
sequence: string
- name: hasanswer
dtype: bool
- name: id
dtype: string
- name: score
dtype: float64
- name: text
dtype: string
- name: title
dtype: string
- name: has_answer
dtype: bool
- name: random_sub
dtype: string
- name: similar_sub
dtype: string
- name: ent_type
dtype: string
- name: new_ctxs
list:
- name: answer_sent
sequence: string
- name: hasanswer
dtype: bool
- name: id
dtype: string
- name: is_adv
dtype: bool
- name: new_answer_sent
dtype: string
- name: original_text
dtype: string
- name: score
dtype: float64
- name: text
dtype: string
- name: title
dtype: string
- name: num_advs
dtype: int64
splits:
- name: train
num_bytes: 92783901
num_examples: 11313
download_size: 53918860
dataset_size: 92783901
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
---
This dataset is primarily used for question answering systems, containing questions, answers, and related context information. The dataset features are rich, including the text of the question, answer sequences, context lists (containing answer sentences, has-answer indicators, context IDs, scores, texts, and titles), has-answer indicators, random sub-items, similar sub-items, entity types, new context lists (containing answer sentences, has-answer indicators, context IDs, whether they are adversarial items, new answer sentences, original texts, scores, texts, and titles), and the number of adversarial items. The dataset is divided into a training set with 11313 samples.
提供机构:
Seongill
原始信息汇总
数据集信息
特征
- question: 类型为字符串。
- answers: 类型为字符串序列。
- ctxs: 列表类型,包含以下字段:
- answer_sent: 类型为字符串序列。
- hasanswer: 类型为布尔值。
- id: 类型为字符串。
- score: 类型为浮点数(float64)。
- text: 类型为字符串。
- title: 类型为字符串。
- has_answer: 类型为布尔值。
- random_sub: 类型为字符串。
- similar_sub: 类型为字符串。
- ent_type: 类型为字符串。
- new_ctxs: 列表类型,包含以下字段:
- answer_sent: 类型为字符串序列。
- hasanswer: 类型为布尔值。
- id: 类型为字符串。
- is_adv: 类型为布尔值。
- new_answer_sent: 类型为字符串。
- original_text: 类型为字符串。
- score: 类型为浮点数(float64)。
- text: 类型为字符串。
- title: 类型为字符串。
- num_advs: 类型为整数(int64)。
数据分割
- train: 包含11313个样本,占用92783901字节。
数据集大小
- 下载大小: 53918860字节。
- 数据集大小: 92783901字节。
配置
- default: 包含训练数据文件,路径为
data/train-*。



