---
language:
- en
dataset_info:
  features:
  - name: homograph
    dtype: string
  - name: wordid
    dtype: string
  - name: sentence
    dtype: string
  - name: start
    dtype: int64
  - name: end
    dtype: int64
  splits:
  - name: train
    num_bytes: 2049515
    num_examples: 14487
  - name: test
    num_bytes: 228724
    num_examples: 1615
  download_size: 1159431
  dataset_size: 2278239
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: test
    path: data/test-*
---


A 🤗-datasets compatible clone of https://github.com/google-research-datasets/WikipediaHomographData.