README.md

metadata

task_categories:
  - feature-extraction
language:
  - en
dataset_info:
  features:
    - name: image
      dtype: image
    - name: ground_truth
      dtype: string
  splits:
    - name: train
      num_bytes: 235013794
      num_examples: 426
    - name: validation
      num_bytes: 26678659
      num_examples: 50
    - name: test
      num_bytes: 15053216
      num_examples: 26
  download_size: 197949185
  dataset_size: 276745669