From Training to Deployment: End-to-End MLOps Pipeline on AWS | Project Demo скачать в хорошем качестве

From Training to Deployment: End-to-End MLOps Pipeline on AWS | Project Demo

In this video, I present my end-to-end MLOps term project: U.S. Visa Approval Classification System built with a full production-style machine learning workflow. This project goes beyond model training and covers the complete MLOps lifecycle, including: Data ingestion Data validation Data transformation Model training and selection Model artifact storage and versioning FastAPI deployment Docker containerization GitHub Actions CI/CD AWS ECR and EC2 deployment Monitoring with Evidently AI In this video, I also explained the project structure , the MLOps pipeline flow, and the deployment process where I connect GitHub, GitHub Actions, a self-hosted runner on EC2, AWS ECR, and Docker to automatically deploy the application. Tech Stack Python Scikit-learn XGBoost CatBoost FastAPI Docker GitHub Actions AWS EC2 AWS ECR AWS S3 Evidently AI MongoDB Topics Covered End-to-end MLOps pipeline design Folder structure explanation CI/CD automation with GitHub Actions Self-hosted GitHub runner setup on EC2 Docker image build and push to ECR Pulling and running the latest image on EC2 Monitoring and future improvements This project shows how to move from a notebook-based ML workflow to a maintainable, scalable, and production-oriented MLOps system Author: Pankaj Kumar Pramanik Role: Data, AI & MLOps Engineer Website: pankajpramanik.com #MLOps #MachineLearning #AWS #Docker #FastAPI #GitHubActions #DataScience #CICD #ECR #EC2