metadata
dataset_info:
features:
- name: username
dtype: string
- name: score
dtype: float64
- name: timestamp
dtype: string
splits:
- name: train
num_bytes: 198990
num_examples: 3879
download_size: 119378
dataset_size: 198990
configs:
- config_name: default
data_files:
- split: train
path: data/train-*