跳到主要内容

dsh-data-cleaning

已验证

@liuhange/dsh-data-cleaning · v3.0.4 · MIT

Data cleaning plugin: deduplication, format standardization, anomaly detection Also known as @deepseek-ai/dsh-data-cleaning.

安装

dsh plugin add @liuhange/dsh-data-cleaning

用 dsh --profile default --dump-config 确认 layer 已生效 —— 参见安装指南。

源码

发布到 npm 但没有公开仓库。安装前请检查包内容。

标签

说明文档

@liuhange/dsh-data-cleaning

Data cleaning plugin for DeepSeek Harness. Deduplication, format standardization, anomaly detection.

安装

dsh plugin --profile web add @liuhange/dsh-data-cleaning

Tool: clean_data

  • Parameters: filePath (required), removeDuplicates (default true), standardizeFormat (default true)
  • Output: Cleaning report with dedup stats and anomaly details
  • Output file: {originalName}_cleaned.{ext}

Known Limitations and Deferred Work

  • Anomaly detection rules are basic (short content, null values); extensible via config