metadata

dataset_info:
  features:
    - name: image
      dtype: PIL.Image.Image
    - name: label
      dtype: int
  class_label:
    names:
      '0': bowtie
      '1': windmill
      '2': tree
      '3': river
      '4': ice cream
      '5': eye
      '6': book
      '7': sun
      '8': star
      '9': airplane
      '10': butterfly
      '11': clock
      '12': car
      '13': fish
      '14': face
      '15': umbrella
      '16': cat
      '17': bicycle
      '18': pizza
      '19': house
      '20': cake
      '21': bucket
      '22': crown
      '23': light bulb
      '24': cell phone
      '25': t-shirt
  splits:
    - name: train
      num_bytes: 174683075.2
      num_examples: 416000
    - name: val
      num_bytes: 21851140.4
      num_examples: 52000
    - name: test
      num_bytes: 21675900.4
      num_examples: 52000
  download_size: 218844448
  dataset_size: 218210116
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*
      - split: val
        path: data/val-*
      - split: test
        path: data/test-*
task_categories:
  - image-classification
tags:
  - art
size_categories:
  - 100K<n<1M

Quick! Draw 26 Class Dataset

This dataset is derived from the Google Quick! Draw dataset and contains 26 classes of doodle images drawn by users. The classes include common objects and entities like animals, vehicles, food items, and everyday objects.

Dataset Details

Number of Classes: 26
Total Images: 520,000 (416,000 train, 52,000 val, 52,000 test)
Image Format: PNG images of size 28x28 pixels (grayscale)
Data Fields:
- image: PIL Image object
- label: Integer label corresponding to class

Class Labels

0: bowtie, 1: windmill, 2: tree, 3: river, 4: ice cream, 5: eye, 6: book, 7: sun, 8: star, 9: airplane, 10: butterfly, 11: clock, 12: car, 13: fish, 14: face, 15: umbrella, 16: cat, 17: bicycle, 18: pizza, 19: house, 20: cake, 21: bucket, 22: crown, 23: light bulb, 24: cell phone, 25: t-shirt

Download and Loading

You can load this dataset using the load_dataset function from the datasets library:

from datasets import load_dataset

dataset = load_dataset("OmAlve/quickdraw_26_classes")

This will download and cache the dataset locally.

Maintainers

Om Alve