知识卡片
概率区间不同于二元结果区间
逻辑拟合曲线周围的带通常描述条件概率估计的不确定性,不是宣称一次二元结果会落在连续概率带内。系数、阈值及概率校准也承担不同问题,不能只看曲线贴合判断模型可靠。
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
- · 监督学习模型
原书出处 · 1 处
- · 查看原书依据
The horizontal axis is labeled score and ranges from 100 to 700 in increments of 100. The vertical axis is labeled probability of event and ranges from 0.0 to 1.0 in increments of 0.2. The plot has a logistic regression curve fitted to data points. The equation of the logistic regression model reads, P of 1 equals exp of negative 5.58 plus 0.01505 Score divided
参考来源
- 位置:《Machine Learning Fundamentals: Concepts, Models, and Applications》第06章《监督学习模型》(来源项目:a19c7933-4eda-4176-aebd-30fe7e1ad292;源文件:extracted/Machine Learning Fundamentals: Concepts, Models, and Applications/_epub-src/OEBPS/xhtml/19_chapter06.xhtml)
- 结论依据:摘录中关于概率区间不同于二元结果区间的定义、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:The horizontal axis is labeled score and ranges from 100 to 700 in increments of 100. The vertical axis is labeled probability of event and ranges from 0.0 to 1.0 in increments of 0.2. The plot has a logistic regression curve fitted to data points. The equation of the logistic regression model reads, P of 1 equals exp of negative 5.58 plus 0.01505 Score divided