metadata
dataset_info:
features:
- name: id
dtype: int64
- name: translation
struct:
- name: ar
dtype: string
- name: en
dtype: string
- name: shu
dtype: string
splits:
- name: train
num_bytes: 17115363
num_examples: 24865
- name: test
num_bytes: 4290613
num_examples: 6217
download_size: 10875211
dataset_size: 21405976
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*