metadata
dataset_info:
features:
- name: video_id
dtype: string
- name: audio
dtype: audio
- name: labels
sequence: string
- name: human_labels
sequence: string
splits:
- name: train
num_bytes: 4190889503.038266
num_examples: 3010
- name: test
num_bytes: 2984602953.9159956
num_examples: 2153
download_size: 7180127332
dataset_size: 7175492456.954262
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*