metadata
dataset_info:
features:
- name: docid
dtype: string
- name: video
dtype: string
splits:
- name: train
num_bytes: 297780
num_examples: 10000
download_size: 128893
dataset_size: 297780
configs:
- config_name: default
data_files:
- split: train
path: data/train-*