Skip to content

Latest commit

 

History

History
18 lines (13 loc) · 742 Bytes

readme.md

File metadata and controls

18 lines (13 loc) · 742 Bytes

Flow Vision

  1. Clone repository and go to the root directory.
  2. Use conda or python3 -m venv venv to create a virtual environment. Activate the virtual environment.
  3. Run pip install -r requirements.txt to install all the required dependencies.
  4. Run python src/run.py to start ASGI server.
  5. The following are the endpoints exposed from the service (a) POST /flowvision/v1/extract-reading (b) POST /flowvision/v1/feedback
  6. The API specification will give more details about the request and response structure

Use Qwen2-VL model

  • Choose the Qwen/Qwen2-VL-2B-Instruct model in the conf/config.yaml file.

Use OpenAI model

  • Choose the gpt-4o model in the conf/config.yaml file.