metadata
dataset_info:
features:
- name: text
dtype: string
- name: motion
sequence:
sequence: float32
splits:
- name: train
num_bytes: 440563896
num_examples: 1470
download_size: 440531956
dataset_size: 440563896
configs:
- config_name: default
data_files:
- split: train
path: data/train-*