Dark Mode
Formula E Driver Performance Metrics
Data Science and Analytics
Tags and Keywords
Trusted By
"No reviews yet"
Free
About
Data details performance metrics collected during Formula E racing events, specifically designed for predicting driver lap times. The challenge uses these metrics to build a machine learning model capable of predicting the Envision Racing drivers’ lap times for critical qualifying sessions. Access to these fast insights helps drivers make swift decisions during a race. Effective analysis requires considering several performance factors such as weather, track conditions, and the driver’s familiarity with the specific circuit. This material is made available through a collaboration between Genpact and Envision Racing.
Columns
The dataset contains 25 columns, providing detailed telemetry and session information:
- NUMBER: Number in sequence.
- DRIVER_NUMBER: The assigned number for the driver.
- LAP_NUMBER: The sequential lap number.
- LAP_TIME: The time taken to finish the lap in seconds (the target column).
- LAP_IMPROVEMENT: The count of lap improvements.
- CROSSING_FINISH_LINE_IN_PIT: An indicator of crossing the line in the pits.
- S1: Sector 1 time, recorded in [min:sec.microseconds].
- S1_IMPROVEMENT: Improvement status in the first sector.
- S2: Sector 2 time, recorded in [min:sec.microseconds].
- S2_IMPROVEMENT: Improvement status in the second sector.
- S3: Sector 3 time, recorded in [min:sec.microseconds].
- S3_IMPROVEMENT: Improvement status in the third sector.
- KPH: Speed measurement in kilometres per hour.
- ELAPSED: Total time elapsed, recorded in [min:sec.microseconds].
- HOUR: Time, recorded in [min:sec.microseconds].
- S1_LARGE, S2_LARGE, S3_LARGE: Additional sector time measurements, recorded in [min:sec.microseconds].
- DRIVER_NAME: Name of the Envision Racing driver.
- PIT_TIME: Time taken when the car stops in the pits.
- GROUP: The qualifying group of the driver.
- TEAM: The name of the racing team.
- POWER: Brake Horsepower (bhp).
- LOCATION: The location of the race event.
- EVENT: Specifies whether the session was free practice or qualifying.
Distribution
The material is distributed across three files. The training data (
train.csv) contains 10,276 rows with 25 columns, including the target variable LAP_TIME. The testing data (test.csv) contains 420 rows, also with 25 columns. A sample submission file (submission.csv) is provided for structure and is very small (849 B). The test data for prediction covers 420 entries that are 100% valid, with no missing or mismatched values.Usage
This resource is ideally suited for building and testing predictive machine learning models. The primary application is multivariate regression aimed at predicting lap times. Analysts should focus on optimizing the Root Mean Square Logarithmic Error (RMSLE) metric to ensure the model generalises effectively on unseen data. The data allows for detailed analysis of how factors like speed, sector times, and pit stops influence overall lap performance. The challenge specifically asks for predictions of
LAP_TIME for qualifying groups in locations 6, 7, and 8.Coverage
The scope covers performance data from the Envision Racing team in the Formula E championship. The data details parameters from various race locations and events (free practice and qualifying). It includes statistical information related to the car's power, time elapsed, and speed, measured over multiple laps and three distinct sectors per circuit.
License
CC0: Public Domain
Who Can Use It
The dataset is intended for data science professionals, machine learning engineers, artificial intelligence practitioners, and technology enthusiasts. It serves as an excellent platform for those looking to showcase their data science skills by applying advanced analytical techniques to real-world auto racing dynamics.
Dataset Name Suggestions
- Envision Racing Formula E Lap Time Prediction Data
- Formula E Driver Performance Metrics
- Qualifying Session Lap Time Analysis
- Auto Racing Telemetry Data
Attributes
Original Data Source: Formula E Driver Performance Metrics
QUALITYLoading...