Skip to content

Backend

Deniz BT edited this page Jul 21, 2024 · 8 revisions

backend-architecture-sp24

Features

  • Serverless Deployment with AWS Lambda
  • Continuous, Efficient Deployment with AWS SageMaker: Ensure streamlined, fine-tuned deployments of the YoloV8 model.
  • Real-Time Frontend-Backend Communication with WebSockets: Maintain a live, persistent connection to poll for results.
  • Concurrent User Request Handling with AWS SQS: Manage and process multiple user requests simultaneously.

Important Files

  • lambda-container: Directory with Dockerfile to build Docker container that is deployed to AWS Lambda and runs entire ML and backend pipeline, after being triggered request from frontend.
  • lambda-container/lambda_function.py: Runs the entire ML pipeline, and communicates with AWS components. Deployed via Docker container to our main AWS Lambda function.
Clone this wiki locally