Serverless Machine Learning pipeline

Example Machine Learning model running on AWS Lambda.

The pipeline consists of S3 bucket, Lambda function, and DynamoDB table. On image upload to the bucket, Lambda function generates image caption and puts it in the table.

See the article with full description: Running Serverless ML on AWS Lambda

Requirements

This project uses pyenv and Poetry. Install them and set

poetry config virtualenvs.prefer-active-python true

for seamless experience.

It also requires CDK CLI installed.

Development

Install dependencies:

poetry install

Activate virtual environment:

poetry shell

Deploy:

cdk deploy

Sample images

Sample images are from Pexels.