知识卡片

普通读书笔记卡 · 8776.c · 生成模型:GPT-6

特征测试记录现实行为非理想行为

理解等级:中级

Characterization test捕捉现有程序在选定输入下的实际表现,给结构变化提供保护。它可能保留已有缺陷,应注明已知问题;单一准确率门槛也不足证明每个转换与预测完全一致。

这张卡片看不懂?

已有反馈 · 0 条

    还没有人提交反馈。

    在整书笔记中的位置

    原书出处 · 1 处

    • · 查看原书依据

      In this section, we’ll share useful guiding principles when refactoring, but before you can even refactor, you must understand two things. First, refactoring without automated tests is highly risky, and we wouldn’t recommend it. In the hands-on exercise, we will demonstrate how you can define characterization tests to first create a safety harness before refactoring.

    参考来源

    - 位置:《Effective Machine Learning Teams (David Tan, Ada Leung, David Colls)》第08章《重构与技术债》(来源项目:a19c7933-4eda-4176-aebd-30fe7e1ad292;源文件:extracted/Effective Machine Learning Teams (David Tan, Ada Leung, David Colls)/_epub-src/OEBPS/ch08.html) - 结论依据:摘录中关于特征测试记录现实行为非理想行为的定义、比较或条件支撑卡片;原书表述不严谨的范围已在覆盖记录逐项限定,不把案例结果外推为普遍结论。 - 原始内容:In this section, we’ll share useful guiding principles when refactoring, but before you can even refactor, you must understand two things. First, refactoring without automated tests is highly risky, and we wouldn’t recommend it. In the hands-on exercise, we will demonstrate how you can define characterization tests to first create a safety harness before refactoring.