README.md

---
dataset_info:
  features:
  - name: chars
    sequence: string
  - name: labels
    sequence:
      class_label:
        names:
          '0': D
          '1': I
          '2': P
          '3': S
  - name: logits
    sequence:
      sequence: float32
      length: 4
  splits:
  - name: train
    num_bytes: 175352249
    num_examples: 46033
  - name: validation
    num_bytes: 9697876
    num_examples: 2557
  - name: test
    num_bytes: 9693002
    num_examples: 2558
  download_size: 144555783
  dataset_size: 194743127
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: validation
    path: data/validation-*
  - split: test
    path: data/test-*
language:
- yue
license: mit
task_categories:
- token-classification
---

This segmentation dataset was generated with [AlienKevin/electra-hongkongese-base-hkcancor-multi](https://huggingface.co/AlienKevin/electra-hongkongese-base-hkcancor-multi) on [R5dwMg/zh-wiki-yue-long](https://huggingface.co/datasets/R5dwMg/zh-wiki-yue-long).

**See https://github.com/AlienKevin/dips for details.**