README.md

metadata

license: mit
dataset_info:
  features:
    - name: id
      dtype: string
    - name: translation
      struct:
        - name: nepali
          dtype: string
        - name: roman
          dtype: string
  splits:
    - name: train
      num_bytes: 148156341
      num_examples: 2397414
    - name: validation
      num_bytes: 158422
      num_examples: 2804
  download_size: 86185823
  dataset_size: 148314763
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*
      - split: validation
        path: data/validation-*
task_categories:
  - translation
language:
  - ne
  - en
tags:
  - nepali
  - roman-english2nepali
  - translation
size_categories:
  - 1M<n<10M

Use this dataset

from datasets import load_dataset

data = load_dataset("syubraj/roman2nepali-transliteration")

Source Dataset:

Saugatkafley/Nepali-Roman-Transliteration