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: 4714091936
num_examples: 4904
- name: test
num_bytes: 2126305304
num_examples: 2212
download_size: 1258107442
dataset_size: 6840397240
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*