Description: 👉 Download the dataset here The Doodle Classifier Prepared Dataset comprises over 1 million images spanning 340 classes of doodles. Sourced from the Quick, Draw! dataset, it features grayscale images of hand-drawn sketches organized by class. Each image has been processed to facilitate machine learning tasks, ensuring a clean and easy-to-use version of the original dataset. Download Dataset Images: Grayscale images of doodles, each sized at 255×255 pixels. Classes: 340 distinct… See the full description on the dataset page: https://huggingface.co/datasets/gtsaidata/Doodle-Dataset.

The Doodle Classifier Prepared Dataset features over 1 million grayscale images of doodles across 340 categories, sourced from Google's Quick, Draw! dataset. Each 255x255 pixel image is organized by class and processed for easy use in machine learning tasks.

Description:

👉 Download the dataset here

The Doodle Dataset Classifier Prepared Dataset comprises over 1 million images spanning 340 classes of doodles. Sourced from the Quick, Draw! dataset, it features grayscale images of hand-drawn sketches organized by class. Each image has been processed to facilitate machine learning tasks, ensuring a clean and easy-to-use version of the original dataset.

Download Dataset

Images: Grayscale images of doodles, each sized at 255×255 pixels.

Classes: 340 distinct categories of doodles, each stored in its own directory.

Total Images: 1,020,000 images, with each class containing precisely 3,000 images.

This refined dataset provides a manageable and structured subset of the original Quick, Draw! dataset by Google, which contains approximately 50 million images

This dataset is sourced from Kaggle.

Dataset Description: Doodle Classifier Prepared Dataset

Overview

This dataset consists of over 1 million images covering 340 classes of doodles. It contains grayscale images of doodles, organized by class, extracted from the Quick, Draw! dataset. Each image represents a hand-drawn sketch from various categories, processed to be ready for machine learning tasks.

This dataset is a clean, processed, and easy-to-use version of the original Quick, Draw! dataset by Google, which has approximately 50 million images.

Content

• Images: Grayscale images of doodles, each 255x255 pixels.
• Classes: 340 categories of doodles, each stored in its directory.
• Total Images: 1,020,000 images, with each class containing exactly 3,000 images.

Structure

• Doodle Folder: Contains 340 subfolders, each representing a different doodle class. Each subfolder includes exactly 3,000 images.
• CSV File (master_doodle_dataframe.csv): Contains additional metadata about the images, including:
  • countrycode: The country code of the user who drew the doodle.
  • drawing: The drawing data is in JSON format.
  • key_id: Unique identifier for each doodle.
  • recognized: Boolean indicating whether the doodle was recognized.
  • word: The class label (e.g., "traffic light").
  • image_path: The file path where the image is stored.

Usage

This dataset is your playground for: - Training and evaluating machine learning models, especially for image classification tasks. - Conducting research and educational activities with a well-organized set of doodle images. - Benchmarking doodle recognition algorithms.

Acknowledgements

This dataset is a clean and processed version of the original Quick, Draw! dataset by Google, which contains approximately 50 million images. Special thanks to the original creators and contributors of the dataset.

License

This dataset is shared under the CC BY 4.0 license. Please attribute the source when using this dataset in your work.

Best of Luck

We hope this dataset serves as a valuable resource for your projects. Happy coding and may your models achieve high accuracy!