metadata
dataset_info:
features:
- name: input
dtype: string
- name: output
sequence: string
- name: id
dtype: string
splits:
- name: train
num_bytes: 198154.34343434343
num_examples: 158
- name: test
num_bytes: 25082.828282828283
num_examples: 20
- name: valid
num_bytes: 25082.828282828283
num_examples: 20
download_size: 66025
dataset_size: 248320
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*
- split: valid
path: data/valid-*