metadata
dataset_info:
features:
- name: feature
sequence:
sequence: float32
- name: labels
sequence: int64
splits:
- name: train
num_bytes: 7078298528
num_examples: 115012
download_size: 1411105345
dataset_size: 7078298528
configs:
- config_name: default
data_files:
- split: train
path: data/train-*