metadata
dataset_info:
features:
- name: input_features
sequence:
sequence:
sequence: float32
- name: labels
sequence: int64
- name: input_length
dtype: float64
splits:
- name: train
num_bytes: 16199849336
num_examples: 16860
- name: test
num_bytes: 959759636
num_examples: 999
download_size: 4493822960
dataset_size: 17159608972
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*