Airflow 2.x for ML Pipeline

This is the git repo for learning airflow pipeline implementation for machine learning use case. The code has been improvised from the source repositories:

e2e-ml-pipeline-airflow - https://github.com/NicoloAlbanese/airflow-ml-pipeline-mvp
pycon-sweden-airflow-ml-pipelines - https://github.com/pycon-ml/airflow_workshop

Pre-requisites

You should have docker and docker-compose installed on your machine !

The easiest way to have everything ready for the workshop is to install Docker Desktop

Docker resource requirement

Minimum resource requirement for docker to start all the services is mentioned below:

Resource	Recommendation
Memory	3 GB
CPU	2 CPU

Getting started

1. Clone this repo

Clone the repo and cd into corresponding folder.

2. Use docker-compose to start the applications:

docker-compose up

4. Access services from browser

Airflow

UI: http://localhost:8080

Username: airflow

Password: airflow

MLflow

http://localhost:5000

Celery Flower

http://localhost:5555

Tear down

Stop and remove containers, networks, images, and volumes

docker-compose down

ShrikanthSingh/airflow-mlflow-ml