metadata
license: mit
dataset_info:
features:
- name: input_features
sequence:
sequence:
sequence: float32
- name: labels
sequence: int64
splits:
- name: train
num_bytes: 38452555400
num_examples: 40032
- name: test
num_bytes: 10206690080
num_examples: 10626
download_size: 9099615221
dataset_size: 48659245480
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*