1. install requirements. Note: when you install torch, make sure you install correct cuda version.
2. go to config.json to set openai => api_key
3. (Optional) if you wanna use stable diffusion as img generation api, make sure you have start AUTOMATIC1111/stable-diffusion-webui and has start flag --api. And then go to config.json change "image_generate_api": "dall-e" as "image_generate_api": "sd".
4. if you wanna use line bot function, don't forget install ffmpeg
5. start server

python app.py [-h] [-l {CRITICAL,FATAL,ERROR,WARN,WARNING,INFO,DEBUG,NOTSET,critical,fatal,error,warn,warning,info,debug,notset}] [--env ENV] [--host HOST] [--port PORT]

• app.py: 程式碼進入點
• util.py: 輔助函數
• test.py: 用來測試用的code
• config.json: 設定檔

You can start the server and go to http://127.0.0.1:5000 to read the doc