# help with curve fitting with a data function

4 vues (au cours des 30 derniers jours)
sofia le 6 Avr 2023
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
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 commentairesAfficher -2 commentaires plus anciensMasquer -2 commentaires plus anciens

Connectez-vous pour commenter.

### Réponse acceptée

Davide Masiello le 6 Avr 2023
Just needed better initial guesses, I think
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');
##### 0 commentairesAfficher -2 commentaires plus anciensMasquer -2 commentaires plus anciens

Connectez-vous pour commenter.

### Catégories

En savoir plus sur Get Started with Curve Fitting Toolbox dans Help Center et File Exchange

R2023a

### Community Treasure Hunt

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

Start Hunting!

Translated by