lsqcurvefit Jacobian returns a scalar

I am performing nonlinear regression using the lsqcurvefit function. I want to calculate the covariance matrix; however, the Jacobian that my code returns is a scalar. Is there something obviously wrong with my code here? Thank you.

F = @(x,f) x(1)/2 * sin(x(2)*pi/2)./(cosh(x(2)*log(2*pi*f*x(3)))+cos(x(2)*pi/2));

    options = optimoptions('lsqcurvefit','OptimalityTolerance', 1e-30 , 'StepTolerance', 1e-30 , 'FunctionTolerance', 1e-30 , 'MaxFunctionEvaluations', 4e100 );
    lb = [0 0 0]; 
    ub = [1 1 1];
    x0 = [5e-7 0.5 5e-5];


    [x,resnorm,~,jacobian] = lsqcurvefit(F,x0,f,y,lb,ub,options);

    format long
    disp(s + "degrees")

    %Display results and plot fit 
    resnorm;
    p = x;
    %disp(p(1)+"   "+p(2)+"   "+ p(3));
    chi(i) = p(1);
    alpha(i) = p(2);
    tau(i) = p(3);

    jacobian

Answer 1

[Matt J]

Check your calling syntax.

[x,resnorm,~,~,~,~,jacobian] = lsqcurvefit(F,x0,f,y,lb,ub,options);