<p align='center'>
<img src='https://user-images.githubusercontent.com/50035229/233825222-0add62d1-e12f-41ef-86d5-4bf3d0f96d84.png' alt='' width='800'/>
</p>
<p align='center'>
Feishu-OpenAI à ç§æç¥è¯åº
<br>
<br>
ð Feishu Vector Knowledge Management ð
</p>
## ð» æºå¨äººåè½
é¤äº [Feishu-OpenAI](https://github.com/ConnectAI-E/Feishu-OpenAI) æ¯æå
åè½å¤ï¼
å¢å äºç¥è¯åºåè½ï¼å
¶ä¸å
·ä½æï¼
ð¬ ç¥è¯åºé®çï¼ä½¿ç¨ /faq æ ç¥è¯åº ä½ä¸ºæ¥è¯¢æ令
ð£ ç¥è¯åº CSV 导å
¥ï¼æ¯æ导å
¥ CSV æ ¼å¼åéæ°æ®
ð ç¥è¯åº CSV å建ï¼æ¯æé建 CSV æ ¼å¼åéæ件ï¼éä½ token ææ¬
ð ç¥è¯åºç®¡çï¼æ¯æ导å
¥ URL ç½é¡µãæ件æ°æ® ð§
ð ç¥è¯åºç®¡çï¼æ¯ææ¥è¯¢æ°æ®åºè®°å½å¹¶å¢å æ¹è®°å½ ð§
ð ç¼åé®é¢åéï¼éä½ token ææ¬ï¼åå°éå¤æ¥è¯¢
## ð 项ç®ç¹ç¹
- ð¥ åºäº Embeddings + Qdrant æ¥è¯¢ä¸ä¸æ
## 项ç®é¨ç½²
### 项ç®åå§å
```sh
git clone https://github.com/ConnectAI-E/Feishu-Vector-Knowledge-Management
cd Feishu-Vector-Knowledge-Management
```
### 导å
¥æ°æ®
```sh
# åå²qaæ°æ®ä¸ºcsvæ件 demo:raw.txt
go run ./cmd - prepare split -f ./data/demo/raw.txt -o ./data/demo/raw.csv
# å°csvæ件转æ¢ä¸ºåéæ°æ®(è°ç¨openai-embedding-api), raw.csv å¿
é¡»å
å«titleåcontentå段
go run ./cmd - prepare analyze -f ./data/demo/raw.csv -o ./data/demo/vector.csv
# 导å
¥æ°æ®csv(åé)æ°æ®
go run ./cmd - prepare import -f ./data/demo/vector.csv
```
#### CSV æ件表头
```csv
id,url,title,content,title_vector,content_vector,vector_id
```
[æ ·ä¾æ°æ®ä¸è½½](./data/demo/data.csv)
#### Qdrant æ¥å£è°è¯æµè¯
å¨çº¿swaggerææ¡£ï¼https://ui.qdrant.tech/#/
#### é¨ç½²
<details>
<summary>docker-compose é¨ç½²</summary>
<br>
ç¼è¾ docker-compose.yamlï¼éè¿ environment é
ç½®ç¸åºç¯å¢åéï¼æè
éè¿ volumes æè½½ç¸åºé
ç½®æ件ï¼ï¼ç¶åè¿è¡ä¸é¢çå½ä»¤å³å¯
```bash
# æ建éå
docker compose build
# å¯å¨æå¡
docker compose up -d
# åæ¢æå¡
docker compose down
```
äºä»¶åè°å°å: http://IP:9000/webhook/event
å¡çåè°å°å: http://IP:9000/webhook/card
</details>
## æ´å¤äº¤æµ
æ´å¤ç»è请访é®é¡¹ç® [Feishu-OpenAI](https://github.com/ConnectAI-E/Feishu-OpenAI)
## èµå©æè°¢
åæ
æè°¢ 'Find My Ai' æä¾çé¨åç»è´¹èµå©ï¼
私有知识库机器人，基于向量数据库和openai实现。.zip
需积分: 0 16 浏览量
2024-02-10
16:08:34
上传
评论
收藏 6.87MB ZIP 举报
私有知识库机器人，基于向量数据库和openai实现。.zip （62个子文件） 
资料总结 
go.mod 2KB go.sum 54KB role_list.yaml 8KB data qdrant .gitignore 14B demo raw.csv 9KB vector.csv 1.01MB data.csv 15.91MB
raw.txt 9KB .gitignore 20B LICENSE 34KB config.example.yaml 629B cmd lark-vkm.go 420B prepare analyze.go 5KB cmd.go 1KB split.go 2KB import.go 2KB server cmd.go 2KB docker-compose.yml 1KB internal initialization lark_client.go 279B config.go 4KB role_list.go 2KB gin.go 2KB handlers event_msg_action.go 1021B card_clear_action.go 1KB init.go 2KB common.go 2KB card_role_action.go 2KB event_common_action.go 4KB card_pic_action.go 3KB msg.go 20KB handler.go 3KB card_common_action.go 1KB event_audio_action.go 1KB event_vkm_action.go 4KB event_pic_action.go 3KB Dockerfile 434B pkg services sessionCache.go 4KB msgCache.go 718B vectorCache.go 1007B openai embeddings_test.go 1KB picture.go 7KB gpt3_test.go 3KB common.go 4KB billing.go 2KB embeddings.go 1KB test_file 
test.jpg 83KB test.wav 39KB
img.png 13KB gpt3.go 2KB audio.go 2KB utils audio wav.go 2KB ogg.go 994B strings.go 791B strings_test.go 2KB loadbalancer loadbalancer.go 2KB qdrantkit collection.go 2KB point.go 3KB common.go 616B collection_test.go 1KB qdrant.go 211B .gitignore 643B README.md 2KB资源评论
