How to check the Algorithm progress?

1 view (last 30 days)
승화 이
승화 이 on 31 Jul 2020
Hellow. I currently make the answer of my optimze problem.
but It can't be used until I confirmed that Algorithm is the same as paper.
The paper is Coleman, T. F. and Y. Li. “An Interior, Trust Region Approach for Nonlinear Minimization Subject to Bounds.” SIAM Journal on Optimization, Vol. 6, 1996, pp. 418–445.
I want to check if it has been done in the same way as the paper's algorithm 2.
But I can't find or understand how to figure it
Below is the formula that I got
options = optimoptions(@lsqnonlin,'Algorithm', 'trust-region-reflective','Display', 'iter', 'MaxFunctionEvaluations', 1000000, 'MaxIterations', 100000);
D_init = zeros(1,1);
lower_bound = 10^(-10);
upper_bound = 10^(-8);
[D_init,~,~,exitflag,output] = lsqnonlin(@myObjective, D_init, lower_bound, upper_bound, options)
Norm of First-order
Iteration Func-count f(x) step optimality
0 2 1.49051 0.873
1 4 0.499246 4.36598e-05 0.554
2 6 0.378797 1.9807e-05 0.24
3 8 0.360547 4.95176e-06 0.195
4 10 0.356504 1.23794e-06 0.185
5 12 0.355523 3.09485e-07 0.183
There may be a local minimum.
Current step size tolerance for step size
Lsqnonlin has been stopped because it is less than the value.
<Stop Criteria Details>
D_init =
9.1835e-09
exitflag =
2
output =
Including the following fields struct:
firstorderopt: 0.1826
iterations: 5
funcCount: 12
cgiterations: 0
algorithm: 'trust-region-reflective'
stepsize: 3.0948e-07
message: '↵There may be a local minimum.↵Current step size tolerance for step size↵
Lsqnonlin has been stopped because it is less than the value.↵↵<Stop Criteria Details>↵↵ Norm of step 3.094848e-07 is lower then options.StepTolerance = 1.000000e-06 that Optimization has been stopped.↵↵'
Becuse of tanslation it's not good to read
I attach the formula
Thank you for your time.

Answers (0)

Products


Release

R2020a

Community Treasure Hunt

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

Start Hunting!