metadata

configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*
      - split: test
        path: data/test-*
      - split: valid
        path: data/valid-*
dataset_info:
  features:
    - name: id
      dtype: int64
    - name: translation
      struct:
        - name: en
          dtype: string
        - name: ko
          dtype: string
  splits:
    - name: train
      num_bytes: 5061272.804687347
      num_examples: 46158
    - name: test
      num_bytes: 281254.92317741335
      num_examples: 2565
    - name: valid
      num_bytes: 281145.272135239
      num_examples: 2564
  download_size: 4172120
  dataset_size: 5623673
task_categories:
  - translation
language:
  - en
  - ko
pretty_name: en2ko_hiphop
size_categories:
  - 10K<n<100K

Dataset Card for "en2ko_hiphop"

Acknowledgements

I gratefully acknowledge DanceD(http://danced.co.kr/) of Korean Hiphop community HIPHOPLE(https://hiphople.com/). All English-to-Korean translations have been provided by DanceD.