metadata
dataset_info:
features:
- name: en
dtype: string
- name: hi_ng
dtype: string
- name: source
dtype: int64
splits:
- name: train
num_bytes: 18814411
num_examples: 178701
- name: test
num_bytes: 1098000
num_examples: 10401
download_size: 11924718
dataset_size: 19912411
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*
Fork of findnitai/english-to-hinglish that splits the training set into train/test.