Skip to content

Latest commit

 

History

History

11_Deployment_with_Serverless

README.md

EMLOV4-Session-11 Assignment -Deployment w/ Serverless

Contents

Requirements

  • Create a CI/CD Pipeline to deploy/update the model to AWS Lambda
  • Submit the deployed function URL

Development Method

Installations

    # aws cli
    curl "https://awscli.amazonaws.com/awscli-exe-linux-x86_64.zip" -o "awscliv2.zip"
    unzip awscliv2.zip
    sudo ./aws/install

    # python requirements
    pip install -r requirements-dev.txt
    pip install -r requirements.txt

    # aws-cdk cli
    npm install -g aws-cdk

Build Command

  • python3 export_onnx.py

Deploy locally

  • uvicorn --host 0.0.0.0 --port 8000 app:app

Test locally

  • pytest test.py

Docker test

  • docker build -t fastapi-classifier .

  • docker run -p 8000:8000 fastapi-classifier

CDK Usage

  • cdk bootstrap

  • cdk deploy --require-approval=never

  • cdk destroy

Debug Command

In case if you have deleted aws lambda function manually in UI, use below to clean completely else it would throw error while deploying agan

aws cloudformation delete-stack --stack-name CatDogClassifierFastAPIStack

AWS CDK Integration

  • In cdk.py file CatDogClassifierFastAPIStack class is used to fetch a Dockerfile and create a container like service with aws lambda service.
  • While developing in local make sure that you have set all the environment variables
    • aws configure -> AWS_ACCESS_KEY, AWS_SECRET_KEY, AWS_REGION
    • CDK_DEFAULT_ACCOUNT, CDK_DEFAULT_REGION
  • After running cdk bootstrap command is used to set up the necessary AWS resources that AWS CDK needs to deploy your infrastructure. It prepares your AWS environment for the first time deployment or when you need to deploy infrastructure that requires certain resources like storing docker image, I am roles, CloudFormation.
  • cdk deploy - Deploys your infrastructure to AWS as defined by your CDK stack.
  • cdk destroy - Destroy all resources created about without any traces left.

Github Actions Pipeline

  • Configure the following env variables in github secrets

    • AWS_ACCESS_KEY, AWS_SECRET_KEY, AWS_REGION
    • CDK_DEFAULT_ACCOUNT, CDK_DEFAULT_REGION

  • Use actions/checkout@v4, actions/setup-python@v5, actions/setup-node@v4 to checkout the code, install python env and node.

  • Export the onnx model from torch script and use cdk to deploy the fast api service to serveless lambda service

Learnings

  • Use AWS CDK and create AWS cloud resources programatically.

Results Screenshots

Deployed URL

CI CD run

Group Members

  1. Ajith Kumar V (myself)
  2. Pravin Sagar
  3. Pratyush