Skip to content

Latest commit

 

History

History
35 lines (25 loc) · 1.58 KB

File metadata and controls

35 lines (25 loc) · 1.58 KB

Message Stream Classifier

A websocket client that processes messages from a socket and tries to dentable conversations and categorize them as possible calendar events

High level Information flow

high level flow

Running With Docker

  • setup .env file with cp .env.example .env and set the WS_SOCK variable, this is the websocket that gets ingested.
  • setup docker if you don't have it yet
  • run sh build_and_run.sh
    • what it does
      • creates model (this is where the model gets stored) and results (the output will be generated here) folder
      • it checks if you have the model files if not downloads them
      • run docker build
      • run docker run command with the required attached volumes
      • execute the client.py in docker, this will start processing the stream

Running with manual setup / seeting up for development

running tests

  • after setting up uv, you can run uv run pytest