README.md

metadata

language:
  - bo
  - en
license: cc-by-sa-4.0
size_categories:
  - 100K<n<1M
task_categories:
  - translation
dataset_info:
  features:
    - name: bo
      dtype: string
    - name: en
      dtype: string
    - name: topic
      dtype: string
  splits:
    - name: train
      num_bytes: 81988008
      num_examples: 127921
  download_size: 16262117
  dataset_size: 81988008
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*

This dataset is the Tibetan-English sentence pairs from the 84000 dataset.

The dataset was re-formatted for Hugging Face using the code found here.

The topic labels were generated with easy_text_clustering using the code found here.