metadata
dataset_info:
features:
- name: source
dtype: string
- name: target
dtype: string
splits:
- name: train
num_bytes: 11052207
num_examples: 12460
- name: validation
num_bytes: 432669
num_examples: 500
- name: test
num_bytes: 1319768
num_examples: 1500
download_size: 6824704
dataset_size: 12804644
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*