how can determine a best theshold value from a vector

3 次查看（过去 30 天）

hanadi abbas 2019-12-3

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/494573-how-can-determine-a-best-theshold-value-from-a-vector

评论： hanadi abbas 2019-12-3

I have vector of 8 element , let A=[0.5807 0.7627 0.6584 0.6274 0.6240 0.6195 0.6012 0.7334] , the corresponding func for each element of a is B=[ 1 0.7 0.8 1 1 1 1 1] .how can find the threshold value of a such that B=1 always. vector A is changed continouly so, i cannot take a specific value as a threshold. how can do it in matlab.

2 个评论
显示无隐藏无

Image Analyst 2019-12-3

在 MATLAB Online 中打开

No idea. How did you get the first B vector? What do you mean that the elements of B are a "func" for the elements of A? How are the different A created? If A is thresholded to create B then B would have only true or false values, not floating point values. If you want B to always = 1 then just do this

B = ones(1, length(A));

tutorial-how-to-ask-a-question-on-answers-and-get-a-fast-answer

hanadi abbas 2019-12-3

sorry, my question may be not clear.

let B vector is an accuracy for each value of A vector. I want threshold value from vector A such that take the elements of A that maximize accuracy and ignor the others. B is not a function of A , so i canot use it as objective function. can I use PSO ?

I have parts of program, in each part there is new vector A . consequently, new accuracy vector B is obtained. so, in each part i want new threshold value that depend on accuracy.

请先登录，再进行评论。

请先登录，再回答此问题。