kg-summarizer

Knowledge graph summarization using LLMs with a FastAPI interface

Create Environment

python -m venv venv
venv\Scripts\activate # on Unix/Mac: source venv/bin/activate
pip install -r .\requirements.txt

Start FastAPI/Uvicorn Server

uvicorn main:app --reload # Using local library: ./venv/bin/uvicorn main:app --reload

Local Development

This service requires a valid OpenAI API key and organization ID to be passed as environment variables. See the sample.env file for a template.

For webapps developed against a local deployment of this service, the PYTHON_ENV env variable must be set to dev. This setting enables browsers to interact with the local server by ensuring responses to preflight CORS requests include the Access-Control-Allow-Private-Network: true header, facilitating access within private networks. See the Chrome blog post for more details. To avoid CSRF attacks, this variable shouldn't be set to dev in production.

Server Update Steps

1. Push changes and make new release on github
   • Make sure release version matches FastAPI docs version in kg_summarizer/server.py and setup.py
2. Update Helm chart
   • Update values.yaml with new release version
   • cd translator-devops/helm/kg-summarizer
   • helm -n translator-dev upgrade -f values-populated.yaml kg-summarizer .

Redis server

Start server: redis-server --dbfilename aragorn_cache.rdb --dir /home/joeyr/data/kg_summarizer

API Location

https://kg-summarizer.apps.renci.org/docs

Todo

• Redis database indices changed???
• This example is interesting because Cisplatin is shown to treat multiple subclasses of mucin-producing carcinoma. Can GPT summarize this well? Don't have example indices anymore
• Add pyunit tests
  • Aragorn and strider have unit tests
• Add abstract summary preprocess flag to server
• Create database dataclass and sort keys as initialization