1234567891011121314151617181920212223242526272829 |
- # 直接Gunicorn服务部署指令:
- gunicorn -w 1 -b 0.0.0.0:1111 app:app
- # 基于Gunicorn配置文件服务部署指令:
- gunicorn -c gunicorn_config.py app:app
- # 主程序运行:
- conda activate tool
- python online_run.py
- # 模型配置:
- ## doubao:
- "llm_model_name": "ep-20241018084532-cgm84", deepseek-v3-241226
- "llm_api_key": "817dff39-5586-4f9b-acba-55004167c0b1",
- "llm_base_url": "https://ark.cn-beijing.volces.com/api/v3",
- ## ds_r1:
- "llm_model_name": "deepseek-r1-250120",
- "llm_api_key": "817dff39-5586-4f9b-acba-55004167c0b1",
- "llm_base_url": "https://ark.cn-beijing.volces.com/api/v3",
- ## TODO:
- 1、知識庫向量持久化。【DONE】(√)
- 2、通过意图类别来命中各类问题的系统提示词,而不需要开发多个机器人。(√)
- 3、加入记忆模块。ConversationBufferMemory / ConversationBufferWindowMemory / ConversationSummaryBufferMemory (√)
- 4、加入检察官机器人。
- 5、修改rag_config.py,完善读取知识库文件方法
- ## 虚拟环境:tool
|