metadata
dataset_info:
features:
- name: text
dtype: string
- name: id
dtype: string
- name: media_bytes
dtype: binary
- name: docling_doc_dict_json
dtype: string
- name: problem
dtype: string
- name: url
dtype: string
splits:
- name: train
num_bytes: 26674423
num_examples: 63
download_size: 21086494
dataset_size: 26674423
configs:
- config_name: default
data_files:
- split: train
path: data/train-*