nikchar/retrieval_verification_squeezebert
收藏Hugging Face2023-10-01 更新2024-03-04 收录
下载链接:
https://hf-mirror.com/datasets/nikchar/retrieval_verification_squeezebert
下载链接
链接失效反馈官方服务:
资源简介:
---
dataset_info:
features:
- name: claim
dtype: string
- name: evidence_wiki_url
dtype: string
- name: text
dtype: string
- name: retrieved_evidence_title
sequence: string
- name: retrieved_evidence_text
sequence: string
- name: labels
dtype: int64
- name: Retrieval_Success
dtype: bool
- name: Predicted_Labels
dtype: int64
- name: Predicted_Labels_Each_doc
sequence: int64
splits:
- name: train
num_bytes: 73601741
num_examples: 11073
download_size: 34426520
dataset_size: 73601741
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
---
# Dataset Card for "retrieval_verification_squeezebert"
[More Information needed](https://github.com/huggingface/datasets/blob/main/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
This dataset is used for retrieval verification tasks and includes multiple features such as claim, evidence Wikipedia URL, text, retrieved evidence title and text, labels, retrieval success flag, predicted labels, and predicted labels for each document. The dataset is divided into a training set with 11073 samples.
提供机构:
nikchar
原始信息汇总
数据集信息
特征
- claim: 数据类型为字符串。
- evidence_wiki_url: 数据类型为字符串。
- text: 数据类型为字符串。
- retrieved_evidence_title: 数据类型为字符串序列。
- retrieved_evidence_text: 数据类型为字符串序列。
- labels: 数据类型为64位整数。
- Retrieval_Success: 数据类型为布尔值。
- Predicted_Labels: 数据类型为64位整数。
- Predicted_Labels_Each_doc: 数据类型为64位整数序列。
数据分割
- train: 包含11073个样本,总字节数为73601741。
数据集大小
- 下载大小: 34426520字节。
- 数据集大小: 73601741字节。
配置
- default: 包含训练数据文件,路径为
data/train-*。



