知识卡片
小提交保留单一变更意图
把不相关修改拆成可理解的逻辑单元,可以更容易审阅、定位失败和回退。小提交追求的是意图清晰和可验证,而非机械限制文件数或行数。若一次构建失败,团队应能判断究竟是哪项行为改变需要调查。
这张卡片看不懂?
这张卡片已通过人工审查,不再接受失败反馈。
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
make small and frequent code commits. We often see ML practitioners lumping multiple unrelated changes (even across 10–20 files) into a single commit. If that commit breaks the build on CI (i.e., tests fail), which logical set of changes caused the error? It’s hard to tell when change sets are large. So, in addition to using version control, we should also use it well by
参考来源
- 位置:《Effective Machine Learning Teams (David Tan, Ada Leung, David Colls)》第09章《MLOps与持续交付》(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/ch09.html)
- 结论依据:摘录中关于小提交保留单一变更意图的定义、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:make small and frequent code commits. We often see ML practitioners lumping multiple unrelated changes (even across 10–20 files) into a single commit. If that commit breaks the build on CI (i.e., tests fail), which logical set of changes caused the error? It’s hard to tell when change sets are large. So, in addition to using version control, we should also use it well by