metadata
dataset_info:
features:
- name: translation
struct:
- name: en
dtype: string
- name: th
dtype: string
splits:
- name: train
num_bytes: 13578996
num_examples: 4998
- name: validation
num_bytes: 2518289
num_examples: 624
- name: test
num_bytes: 2783734
num_examples: 624
download_size: 7114435
dataset_size: 18881019
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*