metadata
dataset_info:
features:
- name: input
sequence: float32
splits:
- name: train
num_bytes: 90001800
num_examples: 450
download_size: 90636595
dataset_size: 90001800
configs:
- config_name: default
data_files:
- split: train
path: data/train-*