LLukas22/lfqa_preprocessed

---
license: mit
task_categories:
- question-answering
- sentence-similarity
language:
- en
size_categories:
- 100K<n<1M
---
# Dataset Card for "lfqa_preprocessed"

## Table of Contents
- [Table of Contents](#table-of-contents)
- [Dataset Description](#dataset-description)
- [Dataset Summary](#dataset-summary)
- [Dataset Structure](#dataset-structure)
- [Data Instances](#data-instances)
- [Data Fields](#data-fields)
- [Data Splits](#data-splits)
- [Additional Information](#additional-information)
- [Licensing Information](#licensing-information)

## Dataset Description

- **Homepage:** [https://towardsdatascience.com/long-form-qa-beyond-eli5-an-updated-dataset-and-approach-319cb841aabb](https://towardsdatascience.com/long-form-qa-beyond-eli5-an-updated-dataset-and-approach-319cb841aabb)

### Dataset Summary

This is a simplified version of [vblagoje's](https://huggingface.co/vblagoje) *[lfqa_support_docs](https://huggingface.co/datasets/vblagoje/lfqa_support_docs)* and *[lfqa](https://huggingface.co/datasets/vblagoje/lfqa)* datasets.
It was generated by me to have a more straight forward way to train Seq2Seq models on context based long form question answering tasks.

## Dataset Structure

### Data Instances

An example of 'train' looks as follows.

```json
{
"question": "what's the difference between a forest and a wood?",
"answer": "They're used interchangeably a lot. You'll get different answers from different resources, but the ...",
"context": [
"Wood is divided, according to its botanical origin, into two kinds: softwoods, ...",
"Processing and products differs especially with regard to the distinction between softwood and hardwood ..."
]
}
```

### Data Fields

The data fields are the same among all splits.

- `question`: a `string` feature.
- `answer`: a `string` feature.
- `context`: a list feature containing `string` features.

### Data Splits

| name |train|validation|
|----------|----:|---------:|
| |226147| 3020|

## Additional Information

### Licensing Information

This dataset is distributed under the MIT licence.

原始信息汇总

数据集概述

数据集描述

数据集总结

• 本数据集是vblagoje的*lfqa_support_docs和lfqa*数据集的简化版本。
• 旨在为基于上下文的长形式问答任务训练Seq2Seq模型提供更直接的方式。

数据集结构

数据实例

• 示例包括：
  • question: 问题，字符串类型。
  • answer: 答案，字符串类型。
  • context: 上下文，列表类型，包含字符串类型元素。

数据字段

• question: 字符串类型。
• answer: 字符串类型。
• context: 列表类型，包含字符串类型元素。

数据分割

• train: 226147条记录。
• validation: 3020条记录。

附加信息

许可信息

• 本数据集遵循MIT许可。