知识卡片
数据驱动改变设计工作的重点
训练样例能让模型学习表面变体和统计关系,减少逐条手写规则。工作转向样本代表性、标签、数据划分与错误分析,并未自动完成评价;更多数据或更深模型也不必然更准、更安全或更省维护成本。
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
Machine learning gives you a shortcut to quickly build machines that behave as if a team of programmers had spent years programming them with hundreds of complex hardcoded algorithms and decision tree branches. With a modern NLP pipeline, you can teach a machine to respond effectively to patterns in text without entering regular expression hell; all you need are examples of user statements and the
参考来源
- 位置:《Natural Language Processing in Action, Second Edition》第01章《会读写的机器与自然语言处理》(来源项目:a19c7933-4eda-4176-aebd-30fe7e1ad292;源文件:extracted/Natural Language Processing in Action, Second Edition/_epub-src/OEBPS/Text/chapter-1.html)
- 结论依据:摘录中关于数据驱动改变设计工作的重点的定义、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:Machine learning gives you a shortcut to quickly build machines that behave as if a team of programmers had spent years programming them with hundreds of complex hardcoded algorithms and decision tree branches. With a modern NLP pipeline, you can teach a machine to respond effectively to patterns in text without entering regular expression hell; all you need are examples of user statements and the