Datasets:

billingsmoore
/

Hopkins-bo-en

language:
  - bo
  - en
license: cc-by-sa-3.0
size_categories:
  - 10K<n<100K
task_categories:
  - translation
dataset_info:
  features:
    - name: bo
      dtype: string
    - name: en
      dtype: string
    - name: topic
      dtype: string
  splits:
    - name: train
      num_bytes: 2822068
      num_examples: 27730
  download_size: 577232
  dataset_size: 2822068
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*

This data consists of Tibetan-English text pairs for machine translation. This data was extracted from Jeffrey Hopkins' Tibetan-Sanskrit-English Dictionary as prepared as a '.xml' file by in manyoso/tibetan-corpus.

It was reformatted using the code found here.

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