five

AUTOMATIC/jaicards

收藏
Hugging Face2024-02-04 更新2024-03-04 收录
下载链接:
https://hf-mirror.com/datasets/AUTOMATIC/jaicards
下载链接
链接失效反馈
官方服务:
资源简介:
--- license: mit task_categories: - conversational - text-generation size_categories: - 100K<n<1M --- # janitorai-cards This dataset contains 190k cards that I received from janitorai, from a source that wished to remain anonymous. My addition to this data is conversion of cards to [v2 character card](https://github.com/malfoyslastname/character-card-spec-v2/blob/main/README.md) format, and a local webpage that can be used to explore the dataset. ### Webpage ![](screenshot.png) Ther webpage lets you browse cards, search by text, fitler by tags and order by date/name/popularity. To use the webpage, put [index.html](index.html) into a directory, and download and extract archives into same directory: [0123.zip](0123.zip), [4567.zip](4567.zip), [89ab.zip](89ab.zip), [cdef.zip](cdef.zip), and [html.zip](html.zip). After that, just open [index.html](index.html) in the browser. The directory structure should look like this: ``` 📁 ┣━━ 📄 index.html ┣━━ 📁 cards ┃ ┣━━ 📁 0 ┃ ┣━━ 📁 1 ┃ ┃ ... ┃ ┗━━ 📁 f ┗━━ 📁 html ┣━━ 📄 allcards.js ┣━━ 📄 cards.js ┗━━ 📄 cardsmeta.js ``` For performance reasons, the webpage only loads 10000 most popular cards when you open it. To view all, click the "Load all" button in the top row. Caveat: instead of downloading the card, it opens it in a new page—you have to save it yourself. I can't figure out how to get the download to work. ### Files - [0123.zip](0123.zip), [4567.zip](4567.zip), [89ab.zip](89ab.zip), [cdef.zip](cdef.zip) - archives with v2 character cards, tested to work with SillyTavern. - [cards-js.7z](cards-js.7z) - all v2 character cards in json format, without images, tested to work with SillyTavern. - [index.html](index.html) - webpage for browsing cards. - [html.zip](html.zip) - files with information about cards - it's needed for the webpage to function. - [orig.7z](orig.7z) - original json files with cards from janitorai - not compatible with any software.
提供机构:
AUTOMATIC
原始信息汇总

janitorai-cards 数据集概述

数据集描述

  • 名称: janitorai-cards
  • 包含内容: 190,000张卡片
  • 来源: 匿名来源
  • 格式转换: 转换为v2 character card格式

文件列表

  • 0123.zip, 4567.zip, 89ab.zip, cdef.zip: 包含v2角色卡片的压缩包,已测试可与SillyTavern兼容。
  • cards-js.7z: 所有v2角色卡片的JSON格式文件,不含图片,已测试可与SillyTavern兼容。
  • index.html: 用于浏览卡片的网页文件。
  • html.zip: 包含卡片信息的文件,用于网页功能。
  • orig.7z: 来自janitorai的原始JSON文件,不兼容任何软件。

网页功能

  • 浏览卡片: 支持通过文本搜索、标签过滤、按日期/名称/流行度排序。
  • 性能优化: 网页初始加载10,000张最流行的卡片,点击"Load all"按钮可查看所有卡片。
  • 注意事项: 卡片在新页面中打开,需手动保存。
5,000+
优质数据集
54 个
任务类型
进入经典数据集
二维码
社区交流群

面向社区/商业的数据集话题

二维码
科研交流群

面向高校/科研机构的开源数据集话题

数据驱动未来

携手共赢发展

商业合作