https://github.com/frontmage/fishpond
鱼塘
https://github.com/frontmage/fishpond
Last synced: 10 months ago
JSON representation
鱼塘
- Host: GitHub
- URL: https://github.com/frontmage/fishpond
- Owner: FrontMage
- Created: 2020-10-12T14:52:59.000Z (over 5 years ago)
- Default Branch: master
- Last Pushed: 2020-10-12T14:53:26.000Z (over 5 years ago)
- Last Synced: 2025-05-31T11:35:18.402Z (about 1 year ago)
- Language: JavaScript
- Size: 161 KB
- Stars: 2
- Watchers: 2
- Forks: 2
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
鱼塘
基于网页微信的养鱼插件
## Table of Contents
- [Table of Contents](#table-of-contents)
- [About The Project](#about-the-project)
- [Built With](#built-with)
- [Getting Started](#getting-started)
- [Prerequisites](#prerequisites)
- [Roadmap](#roadmap)
- [Contributing](#contributing)
- [License](#license)
## About The Project
常年被养鱼突然有感而发，为何不把 NLP 技术用于养鱼，通过模型评测每一句聊天，给出对应的情感分析数据
### Built With
- [fastapi](https://fastapi.tiangolo.com/)
- [snownlp](https://pypi.org/project/snownlp/)
- [docker](https://www.docker.com/)
## Getting Started
先要安装[tampermonkey](https://chrome.google.com/webstore/detail/tampermonkey)或者其他类似的工具。
然后把 repo 里的`fishpond.js`丢到用户脚本里面并启用。
然后需要启动 NLP 模型服务器
```bash
clone the repo
cd fishpond_sentiment_analysis
uvicorn main:app --host 0.0.0.0 --log-level debug
```
然后登录网页微信即可使用。
### Prerequisites
- snownlp
```sh
pip install snownlp
```
- fastapi
```sh
pip install fastapi
```
## Roadmap
See the [open issues](https://github.com/frontmage/fishpond/issues) for a list of proposed features (and known issues).
目前得做个看板用来统计各个鱼儿的情感曲线以及平均回复时间，统计这个是想看看塘主有没有什么特征，例如平均回复时间可能>2 小时。
## Contributing
现在模型是用的`snownlp`，它的情感分析是基于商品评价语料库，我们需要真正的养鱼语料库，所以各位塘主和鱼儿们，把自己的聊天贡献出来吧～
在`data/pond/sample.csv`可以参考如何标注自己的聊天记录为 csv，我每周用 bert 训练一发，模型会以 docker 镜像的形式发布。
## License
Distributed under the MIT License. See `LICENSE` for more information.