metadata
dataset_info:
features:
- name: pred
sequence: float32
- name: label
dtype: float64
splits:
- name: train
num_bytes: 16000
num_examples: 1000
download_size: 9796
dataset_size: 16000
configs:
- config_name: default
data_files:
- split: train
path: data/train-*