This repository showcases a hands-on, modular approach to building and deploying a Loan Approval Prediction system using MLOps best practices. It covers the full machine learning lifecycle from data preprocessing and model training to deployment, monitoring, and CI/CD integration making it ideal for learners and practitioners aiming to master MLOps.
- Languages: Python, Jupyter Notebook
- Frameworks: Flask, FastAPI, Streamlit
- MLOps Tools: MLflow, DVC, GitHub, Jenkins, Docker, Kubernetes
- Monitoring: Prometheus, Grafana, WhyLogs
- Cloud: AWS
mlops/
├── Build-ML-App-FASTAPI/ # FastAPI-based ML app
├── Build-ML-App-Flask/ # Flask-based ML app
├── Build-ML-App-Streamlit/ # Streamlit-based ML app
├── Continuous-Monitoring-Prometheus-Grafana/
├── Deploy-Applications-Docker-Compose/
├── Docker-for-ML/
├── Getting-Started-with-AWS/
├── Git-For-MLOps/
├── Kubernetes-101/
├── Linux-Basics/
├── ML-Monitoring-WhyLogs/
├── MLFlow-Manage-ML-Experiments/
├── Packaging-ML-Model/
├── Python-For-DataScience-MLOps/
├── Python-for-mlops/
├── YAML-Basics/
├── ml-ci-cd-jenkins/
├── .gitignore
└── README.md
Clone the repository
git clone https://github.com/ispr08/mlops.git
cd mlops
(Optional) Create a virtual environment
python -m venv venv
source venv/bin/activate
Install required packages
pip install -r requirements.txt- Multiple deployment frameworks: Flask, FastAPI, Streamlit
- Model packaging and versioning with MLflow and DVC
- CI/CD pipeline setup using Jenkins
- Real-time monitoring with Prometheus, Grafana, and WhyLogs
- Cloud integration walkthrough with AWS
- Containerization and orchestration using Docker and Kubernetes
Each subfolder contains modular implementations. Refer to the respective README or scripts inside:
cd Build-ML-App-FastAPI
python app.py- Prometheus & Grafana: Located in
Continuous-Monitoring-Prometheus-Grafana/ - WhyLogs: For data drift and model performance tracking