metadata
dataset_info:
features:
- name: src
dtype: string
- name: trg
dtype: string
- name: keywords
dtype: string
splits:
- name: train
num_bytes: 23169340
num_examples: 144715
- name: validation
num_bytes: 331467
num_examples: 2048
- name: test
num_bytes: 397544
num_examples: 2500
download_size: 15153461
dataset_size: 23898351
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*