Interact your data and environment using the local GPT, no data leaks, 100% privately, 100% security
Go to file
2023-05-11 15:40:12 +08:00
.github/workflows Create python-publish.yml 2023-04-29 23:30:55 +08:00
.vscode run a demo 2023-04-29 15:17:48 +08:00
asserts Add: config 2023-05-09 11:18:06 +08:00
docs add file 2023-04-28 22:50:44 +08:00
examples fix 2023-05-11 15:40:12 +08:00
pilot fix 2023-05-11 15:40:12 +08:00
.env.template Add: config 2023-05-09 11:18:06 +08:00
.gitignore Add: config 2023-05-09 11:18:06 +08:00
CONTRIBUTING.md rm oceanbase document 2023-05-08 23:01:52 +08:00
environment.yml rm fschat relay 2023-05-11 10:59:08 +08:00
LICENSE Initial commit 2023-04-13 22:52:44 +08:00
README.md Update README.md 2023-05-09 23:49:37 +08:00
requirements.txt add token size 2023-05-11 13:30:57 +08:00
run.sh rm oceanbase document 2023-05-08 23:01:52 +08:00

DB-GPT

A Open Database-GPT Experiment, A fully localized project.

GitHub Repo stars

一个数据库相关的GPT实验项目, 模型与数据全部本地化部署, 绝对保障数据的隐私安全。 同时此GPT项目可以直接本地部署连接到私有数据库, 进行私有数据处理。

DB-GPT 是一个实验性的开源应用程序,它基于FastChat,并使用vicuna-13b作为基础模型。此外,此程序结合了langchainllama-index基于现有知识库进行In-Context Learning来对其进行数据库相关知识的增强。它可以进行SQL生成、SQL诊断、数据库知识问答等一系列的工作。

项目方案

DB-GPT is an experimental open-source application that builds upon the FastChat model and uses vicuna as its base model. Additionally, it looks like this application incorporates langchain and llama-index embedding knowledge to improve Database-QA capabilities.

Overall, it appears to be a sophisticated and innovative tool for working with databases. If you have any specific questions about how to use or implement DB-GPT in your work, please let me know and I'll do my best to assist you.

运行效果演示

Run on an RTX 4090 GPU (The origin mov not sped up!, YouTube地址)

  • 运行演示

  • SQL生成示例 首先选择对应的数据库, 然后模型即可根据对应的数据库Schema信息生成SQL

The Generated SQL is runable.

  • 数据库QA示例

基于默认内置知识库QA

Dependencies

  1. First you need to install python requirements.
python>=3.9
pip install -r requirements.txt

or if you use conda envirenment, you can use this command

cd DB-GPT
conda env create -f environment.yml
  1. MySQL Install

In this project examples, we connect mysql and run SQL-Generate. so you need install mysql local for test. recommand docker

docker run --name=mysql -p 3306:3306 -e MYSQL_ROOT_PASSWORD=aa123456 -dit mysql:latest

The password just for test, you can change this if necessary

Install

  1. 基础模型下载 关于基础模型, 可以根据vicuna合成教程进行合成。 如果此步有困难的同学,也可以直接使用Hugging Face上的模型进行替代. 替代模型

  2. Run model server

cd pilot/server
python vicuna_server.py
  1. Run gradio webui
python webserver.py 

Featurs

  • SQL-Generate
  • Database-QA Based Knowledge
  • SQL-diagnosis

总的来说它是一个用于数据库的复杂且创新的AI工具。如果您对如何在工作中使用或实施DB-GPT有任何具体问题请联系我, 我会尽力提供帮助, 同时也欢迎大家参与到项目建设中, 做一些有趣的事情。

Contribute

Contribute

Licence

MIT