metadata
dataset_info:
features:
- name: input
sequence: float64
- name: output
sequence: float64
splits:
- name: train
num_bytes: 16320000
num_examples: 40000
- name: test
num_bytes: 4080000
num_examples: 10000
download_size: 14052169
dataset_size: 20400000
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*