How to pass additional inputs to function in 'fminbnd'?

9 views (last 30 days)
Abhinav
Abhinav on 5 Feb 2018
Edited: Stephen23 on 19 Mar 2020
I have a function to optimize with respect to 'lambda':
function v = minfunction(lambda)
data = xlsread('data.xlsx');
data = data(:,2:3);
idx = xlsread('idx.xlsx');
v = 0;
for i = 1:size(idx,1)
v = v + max(0,(lambda*data(idx(i,1),1)+(1-lambda)*data(idx(i,1),2)-(lambda*data(idx(i,2),1)+(1-lambda)*data(idx(i,2),2))));
end
end
The optimization is of following form
lambda = fminbnd(@minfunction,0,1)
However, the function 'minfunction' asks for other arguments 'data' and 'idx'. These matrices are updated after each iteration. Currently, I am using excel data to pass these arguments. But I want to pass arguments 'data' and 'idx' in the 'minfunction' it self. If I am writing
function v = minfunction(data,idx,lambda)
and calling the optimizer as
lambda = fminbnd(@minfunction,0,1)
the MATLAB is showing error as too many input arguments.
Can someone help me with this? I am attaching the excel files with question containing idx and data.

Sign in to answer this question.

Accepted Answer

Stephen23
Stephen23 on 5 Feb 2018
Edited: Stephen23 on 19 Mar 2020
Do NOT call xlsread inside the function! Doing so will result in slow and inefficient code due to repeated file access.
The proper solution to your question is to define your function with as many arguments as you require:
function v = minfunction(data,idx,lambda)
and then pass it to the optimizer using an anonymous function:
idx = ...
lam = ...
v = fminbnd(@(t)minfunction(t,idx,lam), 0, 1);
Using this method the values of idx and lam are defined just once, and are then used for all iterations of the optimizer. You can read more this method by reading the MATLAB documentation:
https://www.mathworks.com/help/matlab/math/parameterizing-functions.html
  4 Comments
Show 2 older comments
ahmed elakhdar
ahmed elakhdar on 18 Mar 2020
lambda = fminbnd(@(lambda)minfunction(data,idx,lambda),0,1)

Sign in to comment.

More Answers (0)

Categories

Find more on Problem-Based Optimization Setup in Help Center and File Exchange

Tags

Community Treasure Hunt

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

Start Hunting!