知识卡片
阈值像素溯因须用事实像素
若像素B:=I(U≤p),U原先Uniform(0,1),给定该次p及B=1时U在[0,p]上均匀,给定B=0时则在(p,1]上均匀。两种事实需要不同区间;原例只用Uniform(0,p)而忽略像素0,不能保证不干预时重现原图。
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
sim_img = (n_image_unif <= p_output).int() #9 return sim_img #1 A function that generates a variate from the N_image exogenous distribution #2 The parameters of N_image’s distribution include location and scale parameters for a normal distribution and the upper bound of a uniform distribution. #3 Sample a normal random variate from the normal distribution. #4 Sample a uniform random variate from a uniform distribution. #5 Combine
参考来源
- 位置:《Causal AI》第09章《通用反事实推断算法》(来源项目:a19c7933-4eda-4176-aebd-30fe7e1ad292;源文件:extracted/Causal AI/_epub-src/OEBPS/Text/chapter-9.xhtml)
- 结论依据:摘录中关于阈值像素溯因须用事实像素的定义、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:sim_img = (n_image_unif <= p_output).int() #9 return sim_img #1 A function that generates a variate from the N_image exogenous distribution #2 The parameters of N_image’s distribution include location and scale parameters for a normal distribution and the upper bound of a uniform distribution. #3 Sample a normal random variate from the normal distribution. #4 Sample a uniform random variate from a uniform distribution. #5 Combine