SpaceX Launch Prediction – Data Science & MLOps

This project was developed as part of the IBM Data Science program and represents a complete end-to-end solution. It demonstrates the application of modern Data Science practices to real-world aerospace data, integrating data collection, analysis, machine learning, and deployment.

Visitors can explore the different sections of the project, including data preparation, exploratory analysis, dashboards, and interactive visualizations. A dedicated form will also allow new launch data to be submitted in order to generate predictions using the trained machine learning model.

Data Pipeline

Collection of launch records, preprocessing, and versioning with DVC for reproducibility.

Exploratory Analysis

Statistical exploration and visual insights to identify trends and relationships in the data.

Predictive Modeling

A Random Forest model trained and tracked with MLflow, achieving strong predictive accuracy. Predictions can be generated through the upcoming web form.