metadata
dataset_info:
features:
- name: translation
struct:
- name: en
dtype: string
- name: hi
dtype: string
splits:
- name: train
num_bytes: 580676.8
num_examples: 400
- name: validation
num_bytes: 145169.2
num_examples: 100
download_size: 304659
dataset_size: 725846
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*