metadata
dataset_info:
features:
- name: duration_sec
dtype: int64
- name: path
sequence: string
- name: finished
dtype: bool
- name: rating
dtype: string
splits:
- name: train
num_bytes: 5688990
num_examples: 51318
download_size: 1526382
dataset_size: 5688990
configs:
- config_name: default
data_files:
- split: train
path: data/train-*