larryvrh's picture
Upload README.md with huggingface_hub
e4c17f9
metadata
license: cc-by-sa-4.0
dataset_info:
  features:
    - name: ja
      dtype: string
    - name: zh
      dtype: string
  splits:
    - name: train
      num_bytes: 149036235
      num_examples: 690095
  download_size: 115870646
  dataset_size: 149036235
task_categories:
  - translation
language:
  - ja
  - zh
size_categories:
  - 100K<n<1M

Filtered and modified version of Japanese/Chinese language pair data from WikiMatrix v1.

Process steps: 1. Basic regex based filtering / length checking to remove abnormal pairs. 2. Semantic similarity filtering with a threshold value of 0.6, based on sentence-transformers/LaBSE. 3. Convert all Traditional Chinese sentences into Simplified Chinese with zhconv.

经过过滤和修改的日语/中文语言对数据,来自WikiMatrix v1

处理步骤: 1. 基本的基于正则表达式的过滤/长度检查,以删除异常对。 2. 基于sentence-transformers/LaBSE的语义相似性过滤,阈值为0.6。 3. 使用zhconv将所有繁体中文句子转换为简体中文。

以下はフィルタリングされ修正された日本語/中国語のペアデータです。データ元はWikiMatrix v1です。

処理手順:

  1. 正規表現に基づくフィルタリング/長さのチェックを行い、異常なペアを削除します。
  2. sentence-transformers/LaBSEに基づくセマンティック類似性フィルタリングを行い、閾値は0.6です。
  3. zhconvを使って、すべての繁体字中国語の文を簡体字中国語に変換します。