Problem with reshape and Surf plot

Hello MATLAB Community,

I have the following script and data where i would like to plot a 3d Surface and it works well for some data and it doesn't work because of the dimensions in some cases. Can someone please suggest where is the mistake in the script and how to fix it?

opts = detectImportOptions('Data123.xlsx');
opts = opts.setvartype('Y','double');
opts = opts.setvartype('Z','double');

% Load data
data = readtable('Data123.xlsx', opts);

data.Properties.VariableNames = ["X","F_sw","Y","Z"];
% order the data
data = sortrows(data,["X","Y"]);
X = unique(data.X);
Y = unique(data.Y);
Z = reshape(data.Z,[length(Y),length(X)]);
% View
figure(1)
scatter3(data.X,data.Y,data.Z,12,[1 0 0],"filled","MarkerEdgeColor",[0 0 0])
hold on
s = surf(X,Y,Z,"FaceColor",'interp');
hold off

[Steven Lord]

What does "does not work" mean?

• Do you receive warning and/or error messages? If so the full and exact text of those messages (all the text displayed in orange and/or red in the Command Window) may be useful in determining what's going on and how to avoid the warning and/or error.
• Does it do something different than what you expected? If so, what did it do and what did you expect it to do?
• Did MATLAB crash? If so please send the crash log file (with a description of what you were running or doing in MATLAB when the crash occured) to Technical Support using the telephone icon in the upper-right corner of this page so we can investigate.

[Star Strider]

Exploring the data with:

[X,ix1] = unique(data.X)
seg_len = diff([ix1;numel(data.X)])

produces results: [32; 56; 56; 39]

It is going to be difficult to use reshape with that and get any consistent results.

.

[ragnor]

Dear Mr. Steven Lord,

When i run the script with the attached data set, i get the following error related to reshape. I am not able to figure out the exact reason for this. I suppose

And if i have to fix this error so that I can make a surface plot according to this script, how do i proceed?

Error using reshape
Number of elements must not change. Use [] as one of the size inputs to automatically calculate the appropriate size for that dimension.

Error in Plot_fixDC (line 14)
Z = reshape(data.Z,[length(Y),length(X)]);

[Nikhil Sapre]

I see a error in the way you're using reshape

Error using reshape
Number of elements must not change. Use [] as one of the size inputs to automatically calculate the appropriate size for that dimension.

You are trying to reshae the Z array into a matrix thant changes the number of elements. Please read the dosucmentation on reshape. Also, please let give more details on what you're trying to achieve here.

[ragnor]

Dear @Star Strider,

If i don't use unique, is there any other way, i can create a surface plot with the data.

I just want to make a 3d surface plot to display the results.

Thanks in advance.

[ragnor]

Dear @Nikhil Sapre,

I just want to make a 3D surface plot as displayed in the attached figure.

Answer 1

[Star Strider]

The only solution I can propose is to use either griddata or scatteredInterpolant .

Try this —

opts = detectImportOptions('https://www.mathworks.com/matlabcentral/answers/uploaded_files/695269/Data123.xlsx');
opts = opts.setvartype('Y','double');
opts = opts.setvartype('Z','double');

% Load data
data = readtable('https://www.mathworks.com/matlabcentral/answers/uploaded_files/695269/Data123.xlsx', opts)

data = 184×4 table
      X      F_sw     Y       Z   
    _____    ____    ___    ______

    10000    350     0.5    2.9008
    10000    350       1    3.0802
    10000    350     1.5    3.2261
    10000    350       2    3.3174
    10000    350     2.5    3.3299
    10000    350       3    3.3618
    10000    350     3.5    3.4404
    10000    350       4    3.5091
    10000    350     4.5    3.5816
    10000    350       5    3.6302
    10000    350     5.5    3.6945
    10000    350       6    3.7262
    10000    350     6.5    3.7544
    10000    350       7    3.7761
    10000    350     7.5    3.7944
    10000    350       8    3.8067

data.Properties.VariableNames = ["X","F_sw","Y","Z"];
% order the data
data = sortrows(data,["X","Y"]);
% X = unique(data.X);
% Y = unique(data.Y);
X = data.X;
Y = data.Y;
Z = data.Z;

N = 50;
xv = linspace(min(X), max(X), N);
yv = linspace(min(Y), max(Y), N);
[Xm,Ym] = ndgrid(xv, yv);
Zm = griddata(X,Y,Z,Xm,Ym);
% Z = reshape(data.Z,[length(Y),length(X)]);
% View
figure(1)
scatter3(data.X,data.Y,data.Z,12,[1 0 0],"filled","MarkerEdgeColor",[0 0 0])
hold on
s = surf(Xm,Ym,Zm,"FaceColor",'interp', 'EdgeColor','none');
hold off

If it is not the result you want, I will delete my Answer.

.

[ragnor]

@Star Strider: Thank you very much for the solution. I was looking for a similar result. Just a small question, is it possible to join the red dots so that it looks like 4 tansig curves.

Once again thank you.

[Star Strider]

As always, my pleasure!

Joining the red dots rerquires switching from scatter3 to plot3 —

opts = detectImportOptions('https://www.mathworks.com/matlabcentral/answers/uploaded_files/695269/Data123.xlsx');
opts = opts.setvartype('Y','double');
opts = opts.setvartype('Z','double');

% Load data
data = readtable('https://www.mathworks.com/matlabcentral/answers/uploaded_files/695269/Data123.xlsx', opts)

data = 184×4 table
      X      F_sw     Y       Z   
    _____    ____    ___    ______

    10000    350     0.5    2.9008
    10000    350       1    3.0802
    10000    350     1.5    3.2261
    10000    350       2    3.3174
    10000    350     2.5    3.3299
    10000    350       3    3.3618
    10000    350     3.5    3.4404
    10000    350       4    3.5091
    10000    350     4.5    3.5816
    10000    350       5    3.6302
    10000    350     5.5    3.6945
    10000    350       6    3.7262
    10000    350     6.5    3.7544
    10000    350       7    3.7761
    10000    350     7.5    3.7944
    10000    350       8    3.8067

data.Properties.VariableNames = ["X","F_sw","Y","Z"];
% order the data
data = sortrows(data,["X","Y"]);
X = data.X;
Y = data.Y;
Z = data.Z;

N = 50;
xv = linspace(min(X), max(X), N);
yv = linspace(min(Y), max(Y), N);
[Xm,Ym] = ndgrid(xv, yv);
Zm = griddata(X,Y,Z,Xm,Ym);
% Z = reshape(data.Z,[length(Y),length(X)]);
% View
[Xu,Xix] = unique(data.X,'stable');
Xix = [Xix; numel(X)];
for k = 1:numel(Xix)-1
    C{k} = data{(Xix(k):Xix(k+1)-1),[1 3 4]};
end
% C

figure(1)
hold on
% scatter3(data.X,data.Y,data.Z,12,[1 0 0],"filled","MarkerEdgeColor",[0 0 0])
for k = 1:numel(C)
    plot3(C{k}(:,1),C{k}(:,2),C{k}(:,3),'o-r', 'MarkerSize',6, 'MarkerFaceColor',[1 0 0],"MarkerEdgeColor",[0 0 0], 'LineWidth',2.5)
end
s = surf(Xm,Ym,Zm,"FaceColor",'interp', 'EdgeColor','none', 'FaceAlpha',0.8);
hold off
xlabel('X')
ylabel('Y')
zlabel('Z')
grid on
view(-30,30)

However this did not turn out to be straightforward, since there were connecting lines between the plotted line segments. Sectioning them in the ‘C’ cell array and then plotting them separately in the loop solved this. (An alternative approach would have been to section them, concatenating ‘NaN(1,3)’ to the end of each section, then recombining them into a single matrix. Thie approach I took here is easier.)

Experiment with the various line properties to get the result you want.

.

[ragnor]

@Star Strider: Thank you very much. It was indeed very uselful and it is something i wanted to plot.

Thanks again.

[Star Strider]

As always, my pleasure!