Failed to using syntax 'genFunction' to simulate

Question

Tien Tran 2016-4-11

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/278281-failed-to-using-syntax-genfunction-to-simulate

回答： Walter Roberson 2016-4-11

采纳的回答： Walter Roberson

netFcn.m

I have trained a ANN, then using 'genFunction' to create a function file.m named 'netFcn'. However, I have not yet understood about syntax of function: netFcn(X,~,~).
For example: input = [-0.993243243243243;-1] ; target = [-0.148851286284220]
the result ANN trained from this input is: output = [-0.138566963076372]
while using syntax : netFcn(-0.993243243243243,-1) = -0.131790888633589Why it is different with the output of ANN ? How to use a syntax netFcn(X,~,~) to predict correctly?

0 个评论
显示 -2更早的评论隐藏 -2更早的评论

请先登录，再进行评论。

请先登录，再回答此问题。

Answer 1

Walter Roberson 2016-4-11

1
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/278281-failed-to-using-syntax-genfunction-to-simulate#answer_217362

在 MATLAB Online 中打开

The syntax

function result = netFcn(X,~,~)

means that netFcn expects three inputs to be passed in, but that it is going to ignore the second and third inputs.

The syntax

netFcn(-0.993243243243243,-1)

passes in the scalar -0.993243243243243 as the argument X to netFcn, and passes the scalar -1 as an additional parameter that netFcn is going to ignore. This is not the same as

input = [-0.993243243243243;-1]

for your training, because that input is a vector of two values, not a scalar. The equivalent would be

netFcn([-0.993243243243243;-1])