Combining matrices of different sizes

Hi

How can i combine two matrices of different sizes?

e.g,

x = [1 2 3 4;5 6 7 8;9 10 11 12;13 14 15 16]; y = [1 2 3;4 5 6;7 8 9];

and

z = [x y]; % where z is also a matrix

how should i combine to x & y to get z?

I tried using the cat function but it gives an error:"Dimensions of matrices being concatenated are not consistent".

[Jan]

What does "combine" mean? How should the result look like? [x,y] cannot work, because the result must be a matrix again. And in matrices all columns and all rows must have the same length, respectively.

[Syed]

In the matix, z = [x y], the first element of the first row is the matrix x while the second element in the same row is y!

[Jan]

@Syed: I do not understand your comment. What is the desired output z for the shown inputs?

[Syed]

z should be like this:

z = [1 2 3 4 1 2 3;5 6 7 8 4 5 6;9 10 11 12 7 8 9;13 14 15 16 0 0 0]

[Gabriel Mühlebach]

Or you can use cell. For ex: a = [1,2;3,4]; b = ones(3,3); c = {a,b}; Like that you can put different matrix size in a same variable...

Answer 1

[Andrei Bobrov]

x = [1 2 3 4;5 6 7 8;9 10 11 12;13 14 15 16]; 
y = [1 2 3;4 5 6;7 8 9];

[i1,j1] = ndgrid(1:size(x,1),1:size(x,2));
[i2,j2] = ndgrid(1:size(y,1),(1:size(y,2))+size(x,2));
z = accumarray([i1(:),j1(:);i2(:),j2(:)],[x(:);y(:)]);

or

sx = size(x);
sy = size(y);
a = max(sx(1),sy(1))
z = [[x;zeros(abs([a 0]-sx))],[y;zeros(abs([a,0]-sy))]]

[Syed]

Thanks Andrei, but can u explain the last line of the second solution u provided? i.e z = [[x;zeros(abs([a 0]-sx))],[y;zeros(abs([a,0]-sy))]]

[sumit kumar]

if i have more than two matrix of different size than in that case.How will i proceed????

[Ansuman Mahapatra]

Thank you. That works. Really helpful code.

[Claudio Iturra]

that's a powerfull code Sr.! Thanks

[Akshat Rastogi]

Thanks for the code. But I have another complication. I don't know beforehand how many matrices I have to concatenate. So I have to do it in a loop. But I'm unable to do it.

Basically I have to use them in polyshape function to create geometry. And I have to create a framework such that for any set of coordinates i.e. x1,x2,x3,x4.... and y1,y2,y3,y4; I'm able to use polyshape. Note that the set of coordinates can change for eg: only x1,x2 & y1,y2 can be be there OR only x1,x2,x3,x4 & y1, y2,y3,y4.

Also my coordinates are in the struct format with field 'points', for eg: region(1).points(:,1) gives the x-coordinate of feature-1 and region(1).points(:,end) gives the y-coordinate of feature-1.

I tried to concatenate using structures in a for loop, but that results in nested structure. i.e.

 X_coord={region(1).points(:,1)};
for i=2:size(region,2)
    X_coord={X_coord, region(i).points(:,1)}
end

As a result, polyshape doesn't work as intented

[Akshat Rastogi]

Hello,

I managed to solve the problem myself.

     % Initialization
    X={region(1).points(:,1)};
    Y={region(1).points(:,end)};

    % Loop over the set of coordinates
    for i=2:size(region,2)
        X=[X,{region(i).points(:,1)}];
        Y=[Y,{region(i).points(:,end)}];
    end

Subsequently X,Y can be used to create Polyshape object.

[kevin harianto]

for some reason im still getting Arrays have incompatible sizes for this operation error.

LocationNew = [[appendingArray;zeros(abs([a 0]-sA))], [Location;zeros(abs([a,0]-sL))]];

I am trying to add the 1D array to the 3D array together in order to match the imageInput of 64, 1856, 3.

classdef LidarSemanticSegmentation < lidar.labeler.AutomationAlgorithm
    % LidarSemanticSegmentation Automation algorithm performs semantic
    % segmentation in the point cloud.
    %   LidarSemanticSegmentation is an automation algorithm for segmenting
    %   a point cloud using SqueezeSegV2 semantic segmentation network
    %   which is trained on Pandaset data set.
    %
    %   See also lidarLabeler, groundTruthLabeler
    %   lidar.labeler.AutomationAlgorithm.
    %   Copyright 2021 The MathWorks, Inc.
    % ----------------------------------------------------------------------
    % Step 1: Define the required properties describing the algorithm. This
    % includes Name, Description, and UserDirections.
    properties(Constant)
        % Name Algorithm Name
        %   Character vector specifying the name of the algorithm.
        Name = 'Lidar Semantic Segmentation';
        % Description Algorithm Description
        %   Character vector specifying the short description of the algorithm.
        Description = 'Segment the point cloud using SqueezeSegV2 network.';
        % UserDirections Algorithm Usage Directions
        %   Cell array of character vectors specifying directions for
        %   algorithm users to follow to use the algorithm.
        UserDirections = {['ROI Label Definition Selection: select one of ' ...
            'the ROI definitions to be labeled'], ...
            'Run: Press RUN to run the automation algorithm. ', ...
            ['Review and Modify: Review automated labels over the interval ', ...
            'using playback controls. Modify/delete/add ROIs that were not ' ...
            'satisfactorily automated at this stage. If the results are ' ...
            'satisfactory, click Accept to accept the automated labels.'], ...
            ['Accept/Cancel: If the results of automation are satisfactory, ' ...
            'click Accept to accept all automated labels and return to ' ...
            'manual labeling. If the results of automation are not ' ...
            'satisfactory, click Cancel to return to manual labeling ' ...
            'without saving the automated labels.']};
    end
    % ---------------------------------------------------------------------
    % Step 2: Define properties you want to use during the algorithm
    % execution.
    properties
        % AllCategories
        % AllCategories holds the default 'unlabelled', 'Vegetation',
        % 'Ground', 'Road', 'RoadMarkings', 'SideWalk', 'Car', 'Truck',
        % 'OtherVehicle', 'Pedestrian', 'RoadBarriers', 'Signs',
        % 'Buildings' categorical types.
        AllCategories = {'unlabelled'};
        % PretrainedNetwork
        %   PretrainedNetwork saves the pretrained SqueezeSegV2 network.
        PretrainedNetwork
    end
    %----------------------------------------------------------------------
    % Note: this method needs to be included for lidarLabeler app to
    % recognize it as using pointcloud
    methods (Static)
        % This method is static to allow the apps to call it and check the
        % signal type before instantiation. When users refresh the
        % algorithm list, we can quickly check and discard algorithms for
        % any signal that is not support in a given app.
        function isValid = checkSignalType(signalType)
            isValid = (signalType == vision.labeler.loading.SignalType.PointCloud);
        end
    end
    %----------------------------------------------------------------------
    % Step 3: Define methods used for setting up the algorithm.
    methods
        function isValid = checkLabelDefinition(algObj, labelDef)
            % Only Voxel ROI label definitions are valid for the Lidar
            %  semantic segmentation algorithm.
            isValid = labelDef.Type == lidarLabelType.Voxel;
            if isValid
                algObj.AllCategories{end+1} = labelDef.Name;
            end
        end
        function isReady = checkSetup(algObj)
            % Is there one selected ROI Label definition to automate.
            isReady = ~isempty(algObj.SelectedLabelDefinitions);
        end
    end
    %----------------------------------------------------------------------
    % Step 4: Specify algorithm execution. This controls what happens when
    %         the user presses RUN. Algorithm execution proceeds by first
    %         executing initialize on the first frame, followed by run on
    %         every frame, and terminate on the last frame.
    methods
        function initialize(algObj,~)
            % Load the pretrained SqueezeSegV2 semantic segmentation network.
            outputFolder = fullfile(tempdir, 'Pandaset');
            pretrainedSqueezeSeg = load(fullfile(outputFolder,'trainedSqueezeSegV2PandasetNet.mat'));
            % Store the network in the 'PretrainedNetwork' property of this object.
            algObj.PretrainedNetwork = pretrainedSqueezeSeg.net;
        end
        function autoLabels = run(algObj, pointCloud)
            % Setup categorical matrix with categories including
            % 'Vegetation', 'Ground', 'Road', 'RoadMarkings', 'SideWalk',
            % 'Car', 'Truck', 'OtherVehicle', 'Pedestrian', 'RoadBarriers',
            % and 'Signs'.
            autoLabels = categorical(zeros(size(pointCloud.Location,1), size(pointCloud.Location,2)), ...
                0:12,algObj.AllCategories);
            %A = zeros(10000,10000);
            %filling in the minimum required resolution
            % to meet the neural network's specification.
  %(first iteration failed)   pointCloud.Location = zeros(65,1856,5);
            %Due to an error we must append the various point cloud data
            %first.
           
           
            %next we can add in the ptCloud locations
       %     Location(:,:,1) = pointCloud.Location;
         %   Location = zeros(65,1856,5);
         % adding the additional elements to the array.
         appendingArray = zeros(64,1856,3);
         Location = [pointCloud];



         %We now have to convert the location 1D array to 3D array
         % permute(reshape(TheList, 300, 300, 400), [2 1 3]).
         %reshape(Location, 1,[]);

        % Location = permute(reshape(Location, 3, 1856, 64), [3 2 1]);


        %using the concatenation to add in the third dimention
        %C = cat(3, A, B);
        %Location = cat(3,Location, appendingArray)
        %adding arrays of different sizes
        % sx = size(x);
%sy = size(y);
%a = max(sx(1),sy(1))
%z = [[x;zeros(abs([a 0]-sx))],[y;zeros(abs([a,0]-sy))]]:
sA = size(appendingArray);
sL = size(Location);
a = max(sA(1),sL(1)); 

%Note: we are trying to add in the elements together to meet the imageInput
%size requirement.
LocationNew = [[appendingArray;zeros(abs([a 0]-sA))], [Location;zeros(abs([a,0]-sL))]];
        % Location =[Location, appendingArray];
            
           
           
           
           
           %This will also be applied to the pointCloud Intensity levels
            % as these are also analyzed by the machine learning algorithm.
  %(Pushed aside for later modifications)          pointCloud.Intensity = zeros(64,1865);
           
  pointCloud=LocationNew;

            % Convert the input point cloud to five channel image.
            I = helperPointCloudToImage(pointCloud);
            % Predict the segmentation result.
            predictedResult = semanticseg(I, algObj.PretrainedNetwork);
            autoLabels(:) = predictedResult;
            %using this area we would be able to continuously update the latest file on
            % sending the output towards the CAN Network or atleast ensure that the
            % item is obtainable
            % This area would work the best.
            %first we must
        end
    end
end
function helperDisplayLabelOverlaidPointCloud(I,predictedResult)
    % helperDisplayLabelOverlaidPointCloud Overlay labels over point cloud object.
    %  helperDisplayLabelOverlaidPointCloud(I,predictedResult)
    %  displays the overlaid pointCloud object. I is the 5 channels organized
    %  input image. predictedResult contains pixel labels.
    
    ptCloud = pointCloud(I(:,:,1:3),Intensity = I(:,:,4));
    cmap = helperPandasetColorMap;
    B = ...
        labeloverlay(uint8(ptCloud.Intensity),predictedResult,Colormap = cmap,Transparency = 0.4);
    pc = pointCloud(ptCloud.Location,Color = B);
    ax = pcshow(pc);
    set(ax,XLim = [-70 70],YLim = [-70 70])
    zoom(ax,3.5)
end
function cmap = helperPandasetColorMap
    cmap = [[30 30 30];  % Unlabeled
        [0 255 0];       % Vegetation
        [255 150 255];   % Ground
        [237 117 32];    % Road
        [255 0 0];       % Road Markings
        [90 30 150];     % Sidewalk
        [255 255 30];    % Car
        [245 150 100];   % Truck
        [150 60 30];     % Other Vehicle
        [255 255 0];     % Pedestrian
        [0 200 255];     % Road Barriers
        [170 100 150];   % Signs
        [255 0 255]];    % Building
    cmap = cmap./255;
end
function image = helperPointCloudToImage(ptcloud)
% helperPointCloudToImage converts the point cloud to 5 channel image
image = ptcloud.Location;
image(:,:,5) = ptcloud.Intensity;
rangeData = iComputeRangeData(image(:,:,1),image(:,:,2),image(:,:,3));
image(:,:,4) = rangeData;
index = isnan(image);
image(index) = 0;
end
function rangeData = iComputeRangeData(xChannel,yChannel,zChannel)
rangeData = sqrt(xChannel.*xChannel+yChannel.*yChannel+zChannel.*zChannel);
end

Answer 2

[suresh s]

hi Syed,

If the Size of both x and y matrix are same than only you can concatenated, otherwise we can't concatenate the matrix in Matlab

[Walter Roberson]

Perturabo comments to suresh s:

Unhelpful answer.

[Walter Roberson]

Perturabo, I disagree. suresh s correctly answered the question in the form it was asked. What was asked for originally at the time was something that cannot be done, and telling someone that what they want to do cannot be done is being helpful.

It was not until slightly later that the original poster gave a sample output that permitted the question to be extended into something that could be solved.

[Dmitriy Antselevich]

Ah yes, let's argue semantics on a Matlab forum.

People are not computers, if someone is asking a question they clearly have trouble with the concept and need the answer to be helpful not 1:1 correct.

[Walter Roberson]

If someone asks to do X, then us telling them that X is not possible is correct and helpful. There is no requirement that we guess at all the different things that they might maybe have wanted to do.

I've done that, you know: taken a vague question and listed off pages and pages of things that the person might have meant, and how to do each of the possibilities, and the advantages and disadvantages of each of them and the contexts in which you might need each one. The response I got back was "Thanks." No vote, no acceptance of the my Answer, and in particular, no clarification of which one they had intended and no highlighting of what particular parts of what I said had especially clarified the situation for the person (so no feedback as to how my answer might have been improved). It took me hours to write up.

Other times, I have taken a few hours to write up responses to all the various things someone might have meant, only for the person to say the question was about something else completely that no-one could ever reasonably have perceived from what they had written.

I notice, Dmitriy, that you have posted no Answers at all. It is not clear that you have any significant experience in answering questions, deducing all of the various things that someone might mean, and explaining all of the possibilities.