metadata
dataset_info:
features:
- name: input
dtype: string
- name: output
sequence: string
- name: id
dtype: string
splits:
- name: train
num_bytes: 132886.4
num_examples: 120
- name: test
num_bytes: 16610.8
num_examples: 15
- name: valid
num_bytes: 16610.8
num_examples: 15
download_size: 46260
dataset_size: 166107.99999999997
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*
- split: valid
path: data/valid-*