Skip to content

Commit 28d2e33

Browse files
authored
docs: update description and dbgpts (#1162)
1 parent d5afa6e commit 28d2e33

File tree

3 files changed

+12
-100
lines changed

3 files changed

+12
-100
lines changed

README.md

Lines changed: 5 additions & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -39,11 +39,13 @@
3939

4040
## What is DB-GPT?
4141

42-
DB-GPT is an open-source, data-domain large model framework. Its purpose is to build the infrastructure for the large model domain by developing a variety of technical capabilities, including multi-model management, Text2SQL performance optimization, RAG framework and optimization, and Multi-Agents framework collaboration. These capabilities aim to simplify and facilitate the construction of large model applications around databases.
42+
DB-GPT is an open source AI native data app development framework with AWEL(Agentic Workflow Expression Language) and agents. The purpose is to build infrastructure in the field of large models, through the development of multiple technical capabilities such as multi-model management (SMMF), Text2SQL effect optimization, RAG framework and optimization, Multi-Agents framework collaboration, AWEL (agent workflow orchestration), etc. Which makes large model applications with data simpler and more convenient.
4343

4444
In the Data 3.0 era, based on models and databases, enterprises and developers can build their own bespoke applications with less code.
4545

4646
### Data Agents
47+
![awel_web_search_flow](https://github.com/eosphoros-ai/DB-GPT/assets/17919400/fe1eaf9e-5a0f-4d09-b1a7-f8dc75682695)
48+
4749
![chat_excel](https://github.com/eosphoros-ai/DB-GPT/assets/17919400/03d67da4-b9b9-4df3-8890-176a0941dab8)
4850

4951
![data agents](https://github.com/eosphoros-ai/DB-GPT/assets/17919400/ced393b4-9180-437a-90c5-b43633cda8cb)
@@ -79,6 +81,8 @@ The core capabilities include the following parts:
7981
### SubModule
8082
- [DB-GPT-Hub](https://github.com/eosphoros-ai/DB-GPT-Hub) Text-to-SQL workflow with high performance by applying Supervised Fine-Tuning (SFT) on Large Language Models (LLMs).
8183

84+
- [dbgpts](https://github.com/eosphoros-ai/dbgpts) dbgpts is the official repository which contains some data apps、AWEL operators、AWEL workflow templates and agents which build upon DB-GPT.
85+
8286
#### Text2SQL Finetune
8387
- support llms
8488
- [x] LLaMA

README.zh.md

Lines changed: 7 additions & 99 deletions
Original file line numberDiff line numberDiff line change
@@ -37,12 +37,16 @@
3737
</div>
3838

3939
## DB-GPT 是什么?
40-
DB-GPT是一个开源的数据域大模型框架。目的是构建大模型领域的基础设施,通过开发多模型管理、Text2SQL效果优化、RAG框架以及优化、Multi-Agents框架协作等多种技术能力,让围绕数据库构建大模型应用更简单,更方便。
40+
DB-GPT是一个开源的AI原生数据应用开发框架(AI Native Data App Development framework with AWEL(Agentic Workflow Expression Language) and Agents)。目的是构建大模型领域的基础设施,通过开发多模型管理(SMMF)、Text2SQL效果优化、RAG框架以及优化、Multi-Agents框架协作、AWEL(智能体工作流编排)等多种技术能力,让围绕数据库构建大模型应用更简单,更方便。
41+
4142
数据3.0 时代,基于模型、数据库,企业/开发者可以用更少的代码搭建自己的专属应用。
4243

4344
## 效果演示
4445

4546
### Data Agents
47+
48+
![awel_web_search_flow](https://github.com/eosphoros-ai/DB-GPT/assets/17919400/fe1eaf9e-5a0f-4d09-b1a7-f8dc75682695)
49+
4650
![chat_excel](https://github.com/eosphoros-ai/DB-GPT/assets/17919400/03d67da4-b9b9-4df3-8890-176a0941dab8)
4751

4852
![data agents](https://github.com/eosphoros-ai/DB-GPT/assets/17919400/ced393b4-9180-437a-90c5-b43633cda8cb)
@@ -85,6 +89,8 @@ DB-GPT是一个开源的数据域大模型框架。目的是构建大模型领
8589
- [DB-GPT-Plugins](https://github.com/eosphoros-ai/DB-GPT-Plugins) DB-GPT 插件仓库, 兼容Auto-GPT
8690
- [GPT-Vis](https://github.com/eosphoros-ai/DB-GPT-Web) 可视化协议
8791

92+
- [dbgpts](https://github.com/eosphoros-ai/dbgpts) dbgpts 是官方提供的数据应用仓库, 包含数据智能应用, 智能体编排流程模版, 通用算子等构建在DB-GPT之上的资源。
93+
8894
## 安装
8995

9096
![Docker](https://img.shields.io/badge/docker-%230db7ed.svg?style=for-the-badge&logo=docker&logoColor=white)
@@ -192,104 +198,6 @@ DB-GPT是一个开源的数据域大模型框架。目的是构建大模型领
192198

193199
The MIT License (MIT)
194200

195-
# 路线图
196-
197-
### 知识库RAG检索优化
198-
199-
- [x] Multi Documents
200-
- [x] PDF
201-
- [x] Excel, csv
202-
- [x] Word
203-
- [x] Text
204-
- [x] MarkDown
205-
- [ ] Code
206-
- [ ] Images
207-
- [x] RAG
208-
- [ ] Graph Database
209-
- [ ] Neo4j Graph
210-
- [ ] Nebula Graph
211-
- [x] Multi Vector Database
212-
- [x] Chroma
213-
- [x] Milvus
214-
- [x] Weaviate
215-
- [x] PGVector
216-
- [ ] Elasticsearch
217-
- [ ] ClickHouse
218-
- [ ] Faiss
219-
220-
### 多数据源支持
221-
222-
- 支持数据源
223-
224-
- [x] MySQL
225-
- [x] PostgresSQL
226-
- [x] Spark
227-
- [x] DuckDB
228-
- [x] Sqlite
229-
- [x] MSSQL
230-
- [x] ClickHouse
231-
- [x] StarRocks
232-
- [ ] Oracle
233-
- [ ] Redis
234-
- [ ] MongoDB
235-
- [ ] HBase
236-
- [x] Doris
237-
- [ ] DB2
238-
- [ ] Couchbase
239-
- [ ] Elasticsearch
240-
- [ ] OceanBase
241-
- [ ] TiDB
242-
243-
244-
### 多模型管理与推理优化
245-
- [x] [集群部署](https://www.yuque.com/eosphoros/dbgpt-docs/gmbp9619ytyn2v1s)
246-
- [x] [fastchat支持](https://github.com/lm-sys/FastChat)
247-
- [x] [vLLM 支持](https://www.yuque.com/eosphoros/dbgpt-docs/bhy9igdvanx1uluf)
248-
- [x] 上层接口兼容Openai
249-
- [ ] 云原生环境与Ray环境支持
250-
- [ ] 注册中心引入nacos
251-
- [ ] Embedding模型扩充,优化
252-
253-
### Agents与插件市场
254-
- [x] 多Agents框架
255-
- [x] 自定义Agents
256-
- [x] 插件市场
257-
- [ ] CoT集成
258-
- [ ] 丰富插件样本库
259-
- [ ] 支持AutoGPT协议
260-
- [ ] Multi-agents & 可视化能力打通,定义LLM+Vis新标准
261-
262-
263-
### 测试评估能力建设
264-
- [ ] 知识库的数据文本集
265-
- [ ] 问题集合 [easy、medium、hard]
266-
- [ ] 评分机制
267-
- [ ] Excel + DB库表的测试评估
268-
269-
### 成本与可观测性
270-
- [x] [debugging](https://db-gpt.readthedocs.io/en/latest/getting_started/observability.html)
271-
- [ ] 可观测性
272-
- [ ] 推理预算
273-
274-
### Text2SQL微调
275-
- support llms
276-
- [x] LLaMA
277-
- [x] LLaMA-2
278-
- [x] BLOOM
279-
- [x] BLOOMZ
280-
- [x] Falcon
281-
- [x] Baichuan
282-
- [x] Baichuan2
283-
- [x] InternLM
284-
- [x] Qwen
285-
- [x] XVERSE
286-
- [x] ChatGLM2
287-
288-
- SFT模型准确率
289-
截止20231010,我们利用本项目基于开源的13B大小的模型微调后,在Spider的评估集上的执行准确率,已经超越GPT-4!
290-
291-
[More Information about Text2SQL finetune](https://github.com/eosphoros-ai/DB-GPT-Hub)
292-
293201
## 引用
294202
如果您发现`DB-GPT`对您的研究或开发有用,请引用以下<a href="https://arxiv.org/abs/2312.17449" target="_blank">论文</a>:
295203

assets/wechat.jpg

-15.4 KB
Loading

0 commit comments

Comments
 (0)