help with curve fitting with a data function

Question

sofia 2023-4-6

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/1942544-help-with-curve-fitting-with-a-data-function

回答： Davide Masiello 2023-4-6

Hello,

I'm new to matlab and I'm trying to do a curve fit on my data. However the fit I get isn't what I want, I'm expecting a sinusoidal kind of fit, the graph should look something like how it would if it had as parameter a=0.22 and b=2.69.

this are the data in the 'ottica10.dat' , followed by the script

0 0.93662268

0.166666667 0.768673057

0.333333333 0.440962327

0.5 0.210683072

0.666666667 0.13586155

0.833333333 0.350939133

0.916666667 0.595077789

1 0.741090275

1.166666667 0.924650127

1.333333333 0.798268208

1.67 0.149390082

1.836666667 0.117268597

2.003333333 0.333450767

2.17 0.631289664

2.336666667 0.739983858

2.503333333 0.603385353

2.67 0.396734567

2.836666667 0.113693139

3.003333333 0.130192654

% Load data from file
data = load('ottica10.dat');
Error using load
Unable to find file or directory 'ottica10.dat'.
% Define the model function
M = 2;
myfun = @(p,x) 1/M * exp(-p(1) * 0.5 .* x) .* abs(sin(M*p(2) .* x * 0.5)./(sin(0.5 * p(2) .* x) + 1e-10));
% Set initial values for parameters
p0 = [1, 1];
% Fit the model to the data
params = lsqcurvefit(myfun, p0, data(:,1), data(:,2));
% Generate points for curve plot
x = linspace(min(data(:,1)), max(data(:,1)), 100);
y = myfun(params, x);
% Plot the data and the fitted curve
plot(data(:,1), data(:,2), 'o', x, y);
xlabel('x');
ylabel('y');
legend('Data', 'Fitted curve');

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

Davide Masiello 2023-4-6

0
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/1942544-help-with-curve-fitting-with-a-data-function#answer_1210634

在 MATLAB Online 中打开

ottica10.txt

Just needed better initial guesses, I think

data = readmatrix('ottica10.txt');

M = 2;

myfun = @(p,x) 1/M*exp(-p(1)*0.5*x).*abs(sin(M*p(2).* x * 0.5)./(sin(0.5 * p(2) .* x) + 1e-10));

p0 = [0.5, 5];

params = lsqcurvefit(myfun, p0, data(:,1), data(:,2))

Local minimum possible. lsqcurvefit stopped because the final change in the sum of squares relative to its initial value is less than the value of the function tolerance.

params = 1×2

0.3419 5.3360

% Generate points for curve plot

x = linspace(data(1,1),data(end,1), 100);

y = myfun(params,x);

% Plot the data and the fitted curve

plot(data(:,1), data(:,2), 'o', x, y); hold on

xlabel('x');

ylabel('y');

legend('Data', 'Fitted curve');