metadata
dataset_info:
features:
- name: is
dtype: string
- name: en
dtype: string
- name: sco
dtype: float64
splits:
- name: train
num_bytes: 1284418
num_examples: 3992
download_size: 793776
dataset_size: 1284418
configs:
- config_name: default
data_files:
- split: train
path: data/train-*