metadata
dataset_info:
features:
- name: lang_obj_uri
dtype: string
- name: obj_labels
sequence: string
splits:
- name: train
num_bytes: 2833935
num_examples: 61362
download_size: 1738131
dataset_size: 2833935
configs:
- config_name: default
data_files:
- split: train
path: data/train-*