← 返回刚才阅读的位置

知识卡片

普通读书笔记卡 · 8759.a · 生成模型:GPT-6

非确定输出仍可有确定契约

理解等级:中级

模型具体预测可变时,仍可检查字段类型、允许标签、范围及不变字段。schema检查回答的是协议有效,不代替准确性评价;不能因预测属于合法集合就判模型学得正确。

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

      This tool and technique can help us write readable and holistic assertions, even for situations when the subject of our assertions contains nondeterministic values. Note In organizations with mature software engineering practices, there may be established patterns on how to test APIs or even other aspects of a holistic testing strategy. If that’s the case for you, we recommend that you find the relevant documentation

    参考来源

    - 位置:《Effective Machine Learning Teams (David Tan, Ada Leung, David Colls)》第05章《软件自动化测试》(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/ch05.html) - 结论依据:摘录中关于非确定输出仍可有确定契约的定义、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:This tool and technique can help us write readable and holistic assertions, even for situations when the subject of our assertions contains nondeterministic values. Note In organizations with mature software engineering practices, there may be established patterns on how to test APIs or even other aspects of a holistic testing strategy. If that’s the case for you, we recommend that you find the relevant documentation