metadata
dataset_info:
features:
- name: hash
dtype: string
- name: category
dtype: string
- name: subcategory
dtype: string
- name: task
dtype: string
- name: description
dtype: string
- name: ground_truth
dtype: string
- name: file_hash
dtype: string
- name: slide_number
dtype: int32
- name: content_images
dtype: string
- name: image
dtype: image
- name: json_content
dtype: string
splits:
- name: train
num_bytes: 435612376
num_examples: 800
download_size: 422280230
dataset_size: 435612376
configs:
- config_name: default
data_files:
- split: train
path: data/train-*