metadata
dataset_info:
features:
- name: text
dtype: string
- name: file
dtype: string
- name: 'y'
sequence: float32
splits:
- name: train
num_bytes: 11408261281
num_examples: 46993
download_size: 11437777844
dataset_size: 11408261281
configs:
- config_name: default
data_files:
- split: train
path: data/train-*