metadata
dataset_info:
features:
- name: usc
dtype: string
- name: text
sequence: string
- name: svg
dtype: string
- name: image
dtype: image
- name: caption
dtype: string
splits:
- name: train
num_bytes: 162724007.625
num_examples: 6395
download_size: 151359146
dataset_size: 162724007.625
configs:
- config_name: default
data_files:
- split: train
path: data/train-*