An AI-powered Formula 1 prediction system that uses machine learning to forecast race winners and championship outcomes for the 2025 F1 season. Built with Python, scikit-learn, and Streamlit.
- Single Race Predictions: Predict the outcome of any 2025 F1 race with detailed win probabilities
- Championship Simulation: Simulate the entire 2025 season and get final standings predictions
- Real-time Updates: Incorporates actual 2025 race results into predictions
- Interactive UI: Modify grid positions to see how qualifying affects race outcomes
- Comprehensive Stats: View driver and constructor championship predictions
- Includes real results from the first two races:
- Australian GP (Melbourne): Won by Lando Norris (McLaren)
- Chinese GP (Shanghai): Won by Oscar Piastri (McLaren)
- Updated 2025 driver lineup with all mid-season changes
- Complete 24-race calendar with correct dates and venues
- Clone the repository:
git clone https://github.com/AndreaZero/f1-2025-winner.git
cd f1-2025-winner- Install required packages:
pip install -r requirements.txt- Run the Streamlit app:
python -m streamlit run f1_predictor.py- Go to the "2025 Predictions" tab
- Select a race from the dropdown menu
- (Optional) Modify grid positions
- Click "Predict Race Results"
- View detailed predictions including win probabilities
- Navigate to the "Championship Prediction" tab
- Click "Simulate Remaining Races"
- View:
- Final Driver's Championship standings
- Constructor's Championship standings
- Predicted podiums for all remaining races
- Points distribution visualization
The prediction system uses a Random Forest model trained on historical F1 data from 1950-2024, considering factors such as:
- Qualifying position
- Recent driver performance
- Team performance
- Circuit-specific statistics
- Championship position
- Historical results
- Historical F1 data (1950-2024)
- Real-time 2025 season results
- Current driver and constructor information
- Official 2025 F1 calendar
Feel free to open issues or submit pull requests with improvements.
- Formula 1 for providing historical data
- Streamlit for the amazing web framework
- The F1 community for inspiration and feedback
This is a predictive model for entertainment purposes. Actual race results may vary significantly from predictions.