Check for incorrect argument data type or missing argument in call to function "minimize"

I have recently updated my Matlab fron 2019b to 2022a. I have problem when I use Gaussian Process toolbox (see here: Link). I did not have problem using old version of Matlab. Let consider this simple example:

x = gpml_randn(0.8, 20, 1);                 % 20 training inputs
y = sin(3*x) + 0.1*gpml_randn(0.9, 20, 1);  % 20 noisy training targets
xs = linspace(-3, 3, 61)';                  % 61 test inputs 

meanfunc = [];                    % empty: don't use a mean function
covfunc = @covSEiso;              % Squared Exponental covariance function
likfunc = @likGauss;              % Gaussian likelihood

hyp = struct('mean', [], 'cov', [0 0], 'lik', -1);
hyp2 = minimize(hyp, @gp, -100, @infGaussLik, meanfunc, covfunc, likfunc, x, y);

When I run this code, it shows the error message:

Check for incorrect argument data type or missing argument in call to function 'minimize'.

I do not know why the new version of Matlab returns this error. Can anyone tell me what is wrong with this code?

The minimize function: EDIT: Copyright code removed.

Answer 1

[Torsten]

There is no valid call to "minimize" with 9 arguments as you do above.

The first argument must be a function handle - it's not the case in your call above.

I guess you are talking about the "minimize" function from the file exchange here:

https://de.mathworks.com/matlabcentral/fileexchange/24298-minimize