metadata
dataset_info:
features:
- name: sample_id
dtype: string
- name: dataset_index
dtype: int64
- name: source_image
dtype: image
- name: document_with_boxes_image
dtype: image
- name: document_markdown
dtype: string
- name: extracted_figures
list: image
- name: extracted_figures_metadata
list: string
- name: document_final_markdown
dtype: string
splits:
- name: train
num_bytes: 993543965
num_examples: 1021
download_size: 986827168
dataset_size: 993543965
configs:
- config_name: default
data_files:
- split: train
path: data/train-*