---
id: "vehicle"
slug: "vehicle"
name: "Vehicle"
domain: "Automotive Engineering & Computer Vision"
taskType: "multiclass_classification"
numRows: 846
numColumns: 19
targetColumn: "vehicle_type"
primaryMetric: "Log-Loss"
primaryScore: "0.2401"
parquetUrl: "https://huggingface.co/datasets/carlahq/demo-tabular-benchmarks/resolve/main/data/vehicle/vehicle.parquet"
spreadsheetUrl: "https://docs.google.com/spreadsheets/d/1SklgGHHdrtC8k7VGtYcLn7TzoTlnMBaWz5_gIwqO9kI/edit?usp=sharing"
canonicalUrl: "https://carlahq.eu/datasets/vehicle"
---

# Dataset: Vehicle (vehicle)

> The Statlog Vehicle Silhouettes dataset comprises 846 instances of extracted 2D geometric and statistical shape moments used to classify silhouettes into four distinct vehicle types: double-decker bus, Chevrolet van, Saab 9000, and Opel Manta 400.

## Dataset Overview

| Attribute | Value |
| :--- | :--- |
| **Task Type** | `multiclass_classification` |
| **Business Domain** | Automotive Engineering & Computer Vision |
| **Target Column** | `vehicle_type` |
| **Rows (Instances)** | 846 |
| **Features (Columns)** | 19 |
| **Primary Evaluation Metric** | Log-Loss |
| **Best Benchmark Score** | **0.2401** |

## Business Decision Context
**Context:** Automated traffic monitoring, tolling systems, and optical surveillance require robust classification of 3D motor vehicles from 2D camera silhouettes across varying angles of rotation and elevation.
**Objective:** Accurately classify binary silhouette shape features into one of four vehicle categories (bus, opel, saab, van), resolving subtle structural distinctions between similar passenger car profiles.
**Decision Impact:** Automates vehicle categorization for smart infrastructure, reduces human review overhead in tolling and traffic enforcement, and prevents revenue loss from vehicle misclassification where false classification of vehicle classes triggers incorrect toll rates or access violations.

## Feature Schema
| Column Name | Data Type | Role | Distinct Values | Missing Values (%) | Description |
| :--- | :--- | :--- | :--- | :--- | :--- |
| `vehicle_type` | `categorical` | target | 4 | 0 | Target categorical class label representing the vehicle model: 'bus' (double decker bus), 'van' (Chevrolet van), 'saab' (Saab 9000), or 'opel' (Opel Manta 400). |
| `shape_compactness` | `numeric` | feature | 44 | 0 | Geometric compactness of the silhouette contour, computed as (average perimeter squared) / area. |
| `shape_circularity` | `numeric` | feature | 27 | 0 | Circularity ratio of the 2D silhouette, computed as (average radius squared) / area. |
| `distance_circularity` | `numeric` | feature | 63 | 0 | Boundary distance circularity, calculated as area / (average distance from border squared). |
| `radius_ratio` | `numeric` | feature | 134 | 0 | Radial deviation ratio defined as (maximum radius - minimum radius) / average radius. |
| `principal_axis_aspect_ratio` | `numeric` | feature | 37 | 0 | Aspect ratio of the primary inertia axes, calculated as minor axis length / major axis length. |
| `max_length_aspect_ratio` | `numeric` | feature | 21 | 0 | Ratio of the dimension perpendicular to the maximum length divided by the maximum length. |
| `scatter_ratio` | `numeric` | feature | 131 | 0 | Inertia scatter ratio, computed as moment of inertia about the minor axis / moment of inertia about the major axis. |
| `shape_elongatedness` | `numeric` | feature | 35 | 0 | Measure of silhouette elongation calculated as area / (shrink width squared). |
| `principal_axis_rectangularity` | `numeric` | feature | 13 | 0 | Rectangularity relative to principal axes, computed as area / (principal axis length * principal axis width). |
| `max_length_rectangularity` | `numeric` | feature | 66 | 0 | Rectangularity relative to bounding max length, computed as area / (maximum length * perpendicular length). |
| `scaled_variance_major_axis` | `numeric` | feature | 128 | 0 | Normalized 2nd-order moment of the silhouette area distribution along the major principal axis. |
| `scaled_variance_minor_axis` | `numeric` | feature | 424 | 0 | Normalized 2nd-order moment of the silhouette area distribution along the minor principal axis. |
| `scaled_radius_of_gyration` | `numeric` | feature | 143 | 0 | Normalized radius of gyration computed as (major axis variance + minor axis variance) / area. |
| `skewness_major_axis` | `numeric` | feature | 39 | 0 | Normalized 3rd-order central moment measuring silhouette asymmetry about the major axis. |
| `skewness_minor_axis` | `numeric` | feature | 23 | 0 | Normalized 3rd-order central moment measuring silhouette asymmetry about the minor axis. |
| `kurtosis_major_axis` | `numeric` | feature | 41 | 0 | Normalized 4th-order central moment measuring silhouette distribution peakedness about the major axis. |
| `kurtosis_minor_axis` | `numeric` | feature | 30 | 0 | Normalized 4th-order central moment measuring silhouette distribution peakedness about the minor axis. |
| `hollows_ratio` | `numeric` | feature | 31 | 0 | Ratio quantifying internal voids, concavities, or hollow regions within the vehicle silhouette outline. |

## Empirical Benchmark Leaderboard
| Rank | Algorithm / Model | Model Family | Runtime | Accuracy | ROC AUC | F1-Score |
| :--- | :--- | :--- | :--- | :--- | :--- | :--- |
| 1 | **Google TabFM v1.0** | Tabular Foundation Model | PyTorch/Python | **90.77%** | 0.9673 | 0.8101 |
| 2 | **TabPFN v3** | Tabular Foundation Model | PyTorch/Python | **88.87%** | 0.9573 | 0.7721 |
| 3 | **EXAONE Tabular** | Tabular Foundation Model | PyTorch/Python | **88.10%** | 0.9521 | 0.7516 |
| 4 | **Carla Engine (TabICLv2 WebGPU)** | Tabular Foundation Model | Browser | **88.10%** | 0.9510 | 0.7532 |
| 5 | **Streaming nanotabicl** | Tabular Foundation Model | PyTorch/Python | **88.01%** | 0.9511 | 0.7522 |
| 6 | **Full TabICLv2 (PyTorch Reference)** | Tabular Foundation Model | PyTorch/Python | **88.01%** | 0.9511 | 0.7522 |
| 7 | **nanotabicl Vanilla** | Tabular Foundation Model | PyTorch/Python | **87.99%** | 0.9511 | 0.7514 |
| 8 | **LMT** | Machine Learning Model | Python | **82.98%** | 0.9603 | 0.8285 |
| 9 | **ClassificationViaRegression M5P** | Decision Tree | Python | **81.21%** | 0.9497 | 0.8095 |
| 10 | **AdaBoostM1 J48** | AdaBoost | Python | **76.12%** | 0.9220 | 0.7611 |
| 11 | **RandomForest** | Random Forest | Python | **75.18%** | 0.9312 | 0.7456 |
| 12 | **AdaBoostM1 REPTree** | AdaBoost | Python | **74.82%** | 0.9192 | 0.7425 |
| 13 | **Bagging J48** | Decision Tree | Python | **74.59%** | 0.9185 | 0.7401 |
| 14 | **RandomForest** | Random Forest | Python | **74.00%** | 0.9089 | 0.7356 |
| 15 | **Bagging RandomTree** | Decision Tree | Python | **74.00%** | 0.9089 | 0.7356 |

### Repeated Cross-Validation Fold Scores (TabICLv2 Reference)
| Fold | Train Samples | Test Samples | Metric Score | Latency |
| :--- | :--- | :--- | :--- | :--- |
| Fold 0 | 564 | 282 | 0.2442 | 868ms |
| Fold 1 | 564 | 282 | 0.2338 | 812ms |
| Fold 2 | 564 | 282 | 0.2091 | 824ms |
| Fold 0 | 564 | 282 | 0.2230 | 815ms |
| Fold 1 | 564 | 282 | 0.2586 | 822ms |
| Fold 2 | 564 | 282 | 0.2620 | 813ms |
| Fold 0 | 564 | 282 | 0.2359 | 821ms |
| Fold 1 | 564 | 282 | 0.2423 | 815ms |
| Fold 2 | 564 | 282 | 0.2262 | 819ms |
| Fold 0 | 564 | 282 | 0.2147 | 811ms |
| Fold 1 | 564 | 282 | 0.2608 | 827ms |
| Fold 2 | 564 | 282 | 0.2526 | 810ms |
| Fold 0 | 564 | 282 | 0.2414 | 832ms |
| Fold 1 | 564 | 282 | 0.2437 | 810ms |
| Fold 2 | 564 | 282 | 0.2503 | 830ms |
| Fold 0 | 564 | 282 | 0.2692 | 801ms |
| Fold 1 | 564 | 282 | 0.2096 | 815ms |
| Fold 2 | 564 | 282 | 0.2073 | 799ms |
| Fold 0 | 564 | 282 | 0.2304 | 817ms |
| Fold 1 | 564 | 282 | 0.2493 | 805ms |
| Fold 2 | 564 | 282 | 0.2513 | 827ms |
| Fold 0 | 564 | 282 | 0.2600 | 815ms |
| Fold 1 | 564 | 282 | 0.2296 | 818ms |
| Fold 2 | 564 | 282 | 0.2323 | 819ms |
| Fold 0 | 564 | 282 | 0.2386 | 820ms |
| Fold 1 | 564 | 282 | 0.2733 | 816ms |
| Fold 2 | 564 | 282 | 0.2160 | 815ms |
| Fold 0 | 564 | 282 | 0.2504 | 807ms |
| Fold 1 | 564 | 282 | 0.2614 | 826ms |
| Fold 2 | 564 | 282 | 0.2246 | 822ms |

## Provenance & Downloadable Artifacts
- **Dataset Parquet File:** [Direct Parquet Download](https://huggingface.co/datasets/carlahq/demo-tabular-benchmarks/resolve/main/data/vehicle/vehicle.parquet)
- **Google Sheets Interactive Demo:** [Open in Google Sheets](https://docs.google.com/spreadsheets/d/1SklgGHHdrtC8k7VGtYcLn7TzoTlnMBaWz5_gIwqO9kI/edit?usp=sharing)
- **OpenML Source:** [OpenML Dataset](https://www.openml.org/d/54)
- **Academic Paper / Citation:** [Read Research Paper](https://archive.ics.uci.edu/dataset/149/statlog+vehicle+silhouettes)
- **Canonical HTML Page:** [https://carlahq.eu/datasets/vehicle](https://carlahq.eu/datasets/vehicle)