Predict Falcon 9 Booster Landing Success
Real-time predictions using machine learning on SpaceX launch data
This web app predicts whether the first stage of a SpaceX Falcon 9 rocket will land successfully based on key mission parameters. The predictions are powered by a machine learning model trained on historical launch data from SpaceX.
Key Features
- Data-Driven Predictions:
Trained on real SpaceX launch data. - ML Models:
Multiple classification models such as Logistic Regression, Decision Tree, KNN, Random Forest, AdaBoost, Gradient Boosting, SVM, XGBoost, and CatBoost were trained and evaluated. - Best Model:
XGBoost was selected as the final model based on its performance (accuracy ≈ 88.235%) on test data. - Interactive Web App:
Input launch parameters and get real-time predictions.
How It Works
The project follows an end-to-end machine learning pipeline:
- Data Collection from SpaceX API
- Exploratory Data Analysis (EDA) and Feature Engineering
- Model Training with multiple classification algorithms
- Hyperparameter Tuning and Best Model Selection
- Flask App Deployment for interactive user predictions
Why It Matters
- Helps companies estimate risk for reusable rockets.
- Provides insights for future booster designs and operational decisions.
- Enables competitive analysis against SpaceX and other providers.