RacingLineAI is an advanced interactive dashboard designed to analyze, visualize, and predict Formula 1 race data leveraging machine learning and telemetry data. The platform provides detailed race pace insights, tyre degradation analysis, sector dominance visualization, and predictive intelligence models like LSTM-based tyre life forecasting, lap time regression, and strategy prediction.

• Season & Race Filtering: Select specific seasons (2018-2025) and Grand Prix events to analyze.
• Driver & Tyre Compound Filters: Compare performance for chosen drivers and tyre compounds.
• Circuit Layout Visualization: Displays the track layout with telemetry from fastest laps.
• Race Pace Summary: Aggregates average lap times, fastest laps, and pit stops per driver.
• Gap to Leader Visualization: Lap-by-lap gap analysis to the race leader.
• Tyre Degradation Curves: Visualizes lap time increase relative to tyre life.
• Sector Dominance Charts: Shows driver strengths across different track sectors.
• Delta to Fastest Lap: Tracks lap delta compared to driver’s best lap.
• Stint Type Lap Time Distribution: Analyze opening, mid, and closing stint pace.
• Predictive Intelligence:
  • LSTM Tyre Forecast: Forecasts lap times over tyre life using an LSTM model.
  • Lap Time Regression: Predicts lap times based on tyre life, compound, and track temperature.
  • Strategy Predictor: Estimates optimal tyre stint length based on historical data.

Telemetry-based visualization of the selected Grand Prix circuit.

Table showing average lap times, fastest laps, and pit counts per driver.

Line chart visualizing lap-by-lap gap to the race leader.

Shows lap time progression as tyres wear out.

Bar chart of best sector times by driver, color-coded by team.

Line chart showing lap time delta relative to each driver’s fastest lap.

Box plots showing lap time distribution for opening, mid, and closing tyre stints.

Forecasted lap times vs. actual lap times based on tyre degradation.

Scatter plot comparing actual vs predicted lap times from regression model.

Box plots showing historical stint lengths for different tyre compounds.

Uses a Long Short-Term Memory neural network to predict future lap times as tyre performance degrades over time. This helps anticipate how a driver’s pace will evolve beyond recorded laps.

A linear regression model that predicts lap time by considering factors like tyre compound, tyre life, and track temperature. Useful for quick estimations of lap performance under varying conditions.

Analyzes historical stint lengths and tyre compounds to provide insights on optimal tyre change windows, helping with race strategy planning.

• Python for backend data processing and machine learning.
• Streamlit for interactive dashboard frontend.
• FastF1 for fetching and processing official Formula 1 telemetry and lap data.
• PyTorch for building and training the LSTM predictive model.
• Scikit-learn for regression modeling and metrics.
• Plotly and Matplotlib for rich interactive and static visualizations.
• Pandas and NumPy for data manipulation.

1. Clone the repository

git clone https://github.com/osp06/racinglinai.git
cd racinglineai

2. **Create and activate a virtual environment (recommended)**
python -m venv venv
source venv/bin/activate   # Linux/macOS
venv\Scripts\activate      # Windows

3. **Install required packages**
pip install -r requirements.txt

4. **Download processed race data**
 Ensure the folder data/processed/ contains the CSV files for all races from 2018 to 2025 named like all_races_combined_2018.csv, ..., all_races_combined_2025.csv. 

5. ** Run app **
streamlit run streamlit_app.py