Datasets:

Rakesh7n
/

Formula_One_Cars

dataset_info:
  features:
    - name: image
      dtype: image
    - name: label
      dtype: string
  splits:
    - name: train
      num_bytes: 595074867
      num_examples: 1925
    - name: test
      num_bytes: 180776681
      num_examples: 482
  download_size: 806383581
  dataset_size: 775851548
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*
      - split: test
        path: data/test-*
license: mit

🏎️ Formula 1 Cars Dataset

A curated computer vision and multi-modal dataset consisting of 2,407 high-resolution images across 8 prominent Formula 1 teams.

📌 Dataset Summary

Total Valid Images: 2,407
Number of Classes: 8 F1 Teams
Format: Hugging Face datasets.Image feature + Class Label
Train / Test Split: 80% Train (1,925 images) / 20% Test (482 images)

🗂️ Dataset Structure

Features

Each instance contains the exact image object and its corresponding class label:

Features({
    'image': Image(decode=True),
    'label': ClassLabel(names=[
        'AlphaTauri F1 car',
        'Ferrari F1 car',
        'McLaren F1 car',
        'Mercedes F1 car',
        'Racing Point F1 car',
        'Red Bull Racing F1 car',
        'Renault F1 car',
        'Williams F1 car'
    ])
})

Data Sample

{
    'image': <PIL.JpegImagePlugin.JpegImageFile image mode=RGB size=1920x1080>,
    'label': 'Ferrari F1 car'
}

🚀 Quickstart & Usage

Load the Dataset

from datasets import load_dataset

# Load directly from Hugging Face Hub
dataset = load_dataset("Rakesh7n/Formula_One_Cars")

print(dataset)
# Output:
# DatasetDict({
#     train: Dataset({ features: ['image', 'label'], num_rows: 1925 }),
#     test: Dataset({ features: ['image', 'label'], num_rows: 482 })
# })

📋 Acknowledgements & Citation

Original Dataset

Creator: Sagar Khanna
Kaggle Dataset: Formula One Cars

ℹ️ Source Attribution & Modifications: This dataset is a modified version of Sagar Khanna's Formula One Cars Kaggle Dataset.

Hierarchical Directory Removal: The original nested folder structure (Formula One Cars/<team>/<image>) has been removed and restructured into a single flat directory with standardized, unique sequential indexing (0000.jpg, 0001.jpg, ...).

Data Cleaning & Validation: Verified with PIL.Image.verify() to remove corrupt/invalid non-image files.

Hugging Face Native Format: Formatted with native Hugging Face Image() feature and standard 80/20 train/test split for one-line loading and training.

Citation

@misc{khanna_formula_one_cars,
  author = {Sagar Khanna},
  title = {Formula One Cars},
  year = {2021},
  publisher = {Kaggle},
  howpublished = {\url{https://www.kaggle.com/datasets/sagarplanner/formula-one-cars}}
}

@dataset{formula_one_cars_hf,
  author = {Rakesh},
  title = {Formula 1 Cars Dataset (Modified - Flattened)},
  year = {2026},
  publisher = {Hugging Face},
  howpublished = {\url{https://huggingface.co/datasets/Rakesh7n/Formula_One_Cars}}
}

License

Distributed under the MIT License.