Skip to content

Nelson-Jnrnd/f1-pit-predictor-dataset

Repository files navigation

f1-pit-predictor-dataset

This repository contains a Jupyter notebook for collecting telemetry data from F1 races using the FastF1 API and storing them in CSV files.

The notebook collects data for a whole F1 season (from 2018 to 2022) and can be easily modified to include additional seasons.

The data were intented to be used to train a machine learning model for predicting when a driver will make a pit stop during a race.

Data

The data is organized into separate CSV files for each race, as well as a single CSV file containing all the data for the season.

This dataset contains lap-by-lap data for each drivers during races and contains the following columns:

  • LapStartTime: The start time of the lap in seconds.
  • LapNumber: The number of the lap.
  • LapTime: The duration of the lap in seconds.
  • DriverNumber: The number of the driver who set the lap (44, 77, etc).
  • Compound: The type of tire compound used for the lap.
  • TyreLife: The number of laps made on this set of tire.
  • TrackStatus: Single digit status codes (‘1’: Track clear, ‘2’: Yellow flag, ‘4’: Safety Car, ‘5’: Red Flag, ‘6’: Virtual Safety Car deployed, ‘7’: Virtual Safety Car ending).
  • Stint: The stint number of the race for the driver (2 after the first stop, 3 after the second etc).
  • DistanceToDriverAhead: The distance to the driver ahead in meters.
  • DriverAhead: The name tag (HAM, ALO, etc) of the driver ahead.
  • Track: The name of the track where the race took place.
  • AirTemp: The temperature of the air in degrees Celsius.
  • Humidity: The humidity in percentage.
  • Pressure: The pressure in hPa.
  • Rainfall: Whether it rained during the lap (True/False).
  • TrackTemp: The temperature of the track in degrees Celsius.
  • WindDirection: The direction of the wind.
  • WindSpeed: The speed of the wind in meters per second.

About

No description, website, or topics provided.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published