Using index data to fill a matrix

I have a matrix which I have extracted from an Excel file that contains [row_location, column_location, data]. The row and column are non conituous, so the matrix looks like this:

ExtractedData = 
                [8]    [93]    [0.1981]
                [8]    [94]    [0.2209]
                [8]    [95]    [0.2276]
                [8]    [96]    [0.2276]
                [9]    [49]    [0.2773]
                [9]    [50]    [0.3941]
                [9]    [51]    [0.3256]
                [9]    [52]    [0.3135]
                [9]    [53]    [0.2343]
                [9]    [54]    [0.3337]

I know how big the full matrix would be from

    MaxRow = max(ExtractedData(:, 1))
    MaxCol = max(ExtractedData(:, 2))

So I create a NaN matrix of the size

    DataToPlot = NaN(MaxRow MaxCol)

& then I would like to populate the correct entry in the DataToPlot matrix with the data from the spreadsheet i.e.

DataToPlot(8, 93) = 0.1981
DataToPlot(8, 94) = 0.2209
DataToPlot(8, 95) = 0.2276
etc

I can see that I want to use a loop to step through the ExtractedData one row at a time, but then I'm unsure of how to extract & use the index data to populate the correct location with the data.

The end goal is to generate a surface plot of the data so if there are any shortcuts or 'best practice' ways of going from extacted data to surface plot I'd appreciate a steer in the right direction.

Answer 1

[Jan]

This works with a loop easily:

D      = ExtractedData;  % Easier to read...
MaxRow = max(D(:, 1));
MaxCol = max(D(:, 2));
ToPlot = NaN(MaxRow, MaxCol);
for k = 1:size(D, 1)
    ToPlot(D(k, 1), D(k, 2)) = D(k, 3);
end

But it is smarter to do this by sub2ind (link):

MaxSize = max(D(:, 1:2), [], 1);  % Get size as vector 
ToPlot  = NaN(MaxSize);
index   = sub2ind(MaxSize, D(:, 1), D(:, 2));  % [EDITED, Typo: D(:,1) -> D(:,2)]
ToPlot(index) = D(:, 3);

[Jan]

[MOVED from section for answers] DrEamonn wrote:

Hi Jan

MaxSize = max(D(:, 1:2), [], 1);  % Get size as vector 
ToPlot  = NaN(MaxSize);
index   = sub2ind(MaxSize, D(:, 1), D(:, 1));
ToPlot(index) = D(:, 3);

I've incorporated your code in to mine , & I get an error

Index exceeds matrix dimensions.

Should the second D statement in this lline read D(:,2)?

index   = sub2ind(MaxSize, D(:, 1), D(:, 1));

I've tried that as a change but I still get the error

[Jan]

[MOVED from section for answers] DrEamonn wrote:

I think I fixed the problem - it was a resize function elsewhere in my code to remove a text header

Thanks for you help Jan

Eamonn

[Jan]

You are welcome. I've fixed the typo now.

Please use the section from comments to post a comment.