NN Data Fitting - minimum

3 次查看（过去 30 天）
LuC
LuC 2016-4-11
评论： LuC 2016-4-22
I did the data fitting with neural networks and I got the function that looks very nice, i.e. there is a unique minimum and maximum on a given domain. How can I find the minimum and maximum?
EDIT: Added questions:
How would it be possible to find the x-axis and y-axis intercepts, i.e. f(x) = 0 and f(0)?
Furthermore, I want to estimate these data with the following model y = Dsin[C cos{(1 − E)Bx + E cos(Bx + 1)}]
How can I estimate parameters B, C, D and E (not constants) using NNs? Or it would be better to use Global Optimisation Algorithms, for instance GA?

请先登录，再回答此问题。

采纳的回答

Greg Heath
Greg Heath 2016-4-12
See the help and doc documentation for
1. min
2. max
3. minmax
Hope this helps.
Thank you for formally accepting my answer
Greg

更多回答（1 个）

LuC
LuC 2016-4-19
Thank you.
How would it be possible to find the x-axis and y-axis intercepts, i.e. f(x) = 0 and f(0)?
Furthermore, I want to estimate these data with the following model y = Dsin[C cos{(1 − E)Bx + E cos(Bx + 1)}]
How can I estimate parameters B, C, D and E (not constants) using NNs? Or it would be better to use Global Optimisation Algorithms such as GA?
  2 个评论
Greg Heath
Greg Heath 2016-4-20
I do not see a NN being appropriate for those type of tasks.
Greg
LuC
LuC 2016-4-22
So I should try with the Global Optimisation Tools?

请先登录，再进行评论。

类别

在 Help Center 和 File Exchange 中查找有关 Deep Learning Toolbox 的更多信息

标签

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by