SpaceX Launch Prediction – Project Overview

A complete Data Science & MLOps project: from raw data collection to model deployment with Docker and Railway.

Objective

Predict whether a Falcon 9 launch will succeed using mission and operational data.

Model

Random Forest Classifier
Accuracy: 0.95
Macro F1: 0.94

Deployment

Django web application containerized with Docker and deployed on Railway.

MLOps Pipeline

DVC – Version control for datasets and models.
MLflow – Experiment tracking (parameters, metrics, artifacts).
scikit-learn – Random Forest pipeline with preprocessing.
Django – Web interface for predictions.
Docker – Containerization for reproducibility.
Railway – Cloud deployment of the application.

Technology Stack

Python Django scikit-learn MLflow DVC Docker Railway

Conclusion

This project demonstrates how Data Science, MLOps, and Web Development can be combined to deliver a full end-to-end solution. The pipeline ensures reproducibility with DVC, experiment tracking with MLflow, and deployment with Docker and Railway. Future work includes extending to regression tasks such as cost prediction, adding real-time APIs, and registering production models with MLflow Model Registry.