Lsqcurvefit for 2D gaussian: no error! but not fitting well to the data!

I am addding an example: please refer to XYZ from the file.

Code:

x0 = [10,0,2,0,2]; %Inital guess parameters

%% --- Fit---------------------

[x,resnorm,residual,~] = lsqcurvefit(@D2GaussFunction,x0,xdata,Z);

Zfit=D2GaussFunction(x,xdata);

figure(1)

subplot(1,2,1)

surf(X,Y,Z)

subplot(1,2,2)

surf(X,Y,Zfit)

Here is my D2Gauss Function: based on: https://www.mathworks.com/matlabcentral/fileexchange/37087-fit-2d-gaussian-function-to-data

function F = D2GaussFunction(x,xdata)

F = x(1)*exp(-((xdata(:,:,1)-x(2)).^2/(2*x(3).^2)+(xdata(:,:,2)-x(4)).^2/(2*x(5).^2)));

I am getting the following figure where the fitting is clearly not good (subplot 1 is raw data, subplot2 is fitting). Could you please suggest what is going wrong here?

[Alex Sha]

Hi, Sujay Ray, if adding one more to your original function like below:

F = x(1)*exp(-((xdata(:,:,1)-x(2)).^2/(2*x(3).^2)+(xdata(:,:,2)-x(4)).^2/(2*x(5).^2)))+x(6);

The result will be much better:

Root of Mean Square Error (RMSE): 15.8689661201125
Sum of Squared Residual: 72777.1607734494
Correlation Coef. (R): 0.975082856345198
R-Square: 0.95078657673831

Parameter	Best Estimate
----------	-------------
x1	501.134541206993
x2	-0.471981642148694
x3	-1.43387418591403
x4	0.621256518406971
x5	-1.35781705474931
x6	413.42087378625

[Sujay Ray]

Thanks Alex. This fixes the issue...