README.md

metadata

license: zlib
task_categories:
  - text-generation
  - text2text-generation
  - translation
language:
  - zh
tags:
  - Novel
  - Light-Novel
  - Japanese
  - Chinese
size_categories:
  - 10B<n<100B

Light novels translated in Chinese - crawled from public websites that do not prohibit crawlers

脚盆轻小说汉化 - 从未禁止爬虫的公共网站爬取

---

Version 1 (2025-05-03)

版本 1 (2025-05-03)

Contains around 1500 light novels, including PDF with illustration and txt text files.

• It may be a good source of data that can be used to train your stylish LLM.

• Kindly note that the author has partially clean the text BUT DOES NOT GUARANTEE that it is fully cleaned up.

• 包含约 1500 部轻小说，包括带插图的 PDF 和 txt 文本文件。

• 这可以是训练你的具有风格化的大语言模型 (LLM) 的良好数据来源。

• 请注意，作者已部分清理文本，但不保证已完全清理。

---

File Structure

文件结构

• The things shown in the Data Studio is ONLY A VERY SMALL PART OF DATA available to be reviewed. Please download the archive to access full datasets.

• 在Data Studio中显示预览的内容只是可供查看的数据的一小部分。请下载存档以访问完整的数据集。

• In Version 1, we only provide around 1500 light novels, but we will update to 5000 in Version 2.

• 在版本 1 中，我们仅提供大约 1500 部轻小说，但我们将在最终在版本 2 中更新到 5000 部。

---

• /passcode Contains a file that shows the passcode to decipher the main text and pdf data archives.

• /examples Contains examples of pdf and txt files, for users to be famalier with the format of this dataset.

• /pdf Contains a series of encrypted indivudial novel pdfs.

• /txt Contains a series of encrypted indivudial novel txts.

• /concat.txt Contains the combined complete set of novels.

• /index.csv Contains the index-title mapping of the included novels.

• /description.xlsx Contains the index-title-description mapping of the included novels.

• /passcode 包含一个文件，显示用于解密正文和 PDF 数据档案的密码。

• /examples 包含 PDF 和 txt 文件的示例，以便用户熟悉此数据集的格式。

• /pdf 包含一系列加密的独立的小说 PDF。

• /txt 包含一系列加密的独立的小说 txt。

• /concat.txt 包含合并后的所有的小说。

• /index.csv 包含所收录小说的索引-标题映射。

• /description.xlsx 包含所收录小说的索引-标题-描述映射。