Need help using mink and maxk

I am simplifying coordinates, MwCoords. The left column is the x values and the right column is y values. I do not want to alter the postition of the x or y values.

MwCoords =

  241.5493  360.1065
  241.7508  251.1975
  246.9076  148.8279
  350.6785  144.9718
  349.2095  255.6598
  352.6571  364.5513
  456.4168  149.5675
  459.3270  256.3704
  460.4547  360.4354

I have been using mink to find the three minimum values in the x and maxk to find the 3 max values in the x

MwCoordsMins = mink(MwCoords(:,1),3)
MwCoordsMaxs = maxk(MwCoords(:,1),3)

This outputs

MwCoordsMins =
  241.5493
  241.7508
  246.9076
  
MwCoordsMaxs =
  460.4547
  459.3270
  456.4168

I need to put 1's in the mins, and 3's in the max of the origional MwCoords

From there I need to assign the 3 middle values to 2's

Then I need to plug it back into the origional matrix without disrupting the positioning of these. The postitioning will change every now and then, thats why it has to be in the same place.

After I figure out the x values, I move on to the y values.

If anyone can help, or offer insight that would be great!

Edit - Whatever I added, I put it in bold

[dpb]

Return the optional location vector from mink, maxk. Then it's trivial to replace those locations with a constant.

The combination of all these operations if going to be done repeatedly would be a prime candidate for a user function to operate on the array passed to it.

[Conner Carriere]

I guess I am not sure how to return the outputs to their origional locations. It will be done repeatadly, so I will attempt to make a function out of it.

Answer 1

[Star Strider]

The mink and maxk functions can give the indices of those values as the second output.

If I understand correctly what you want to do, I would do something like this:

[MwCoordsMins,ix1] = mink(MwCoords(:,1),3);
[MwCoordsMaxs,ix2] = maxk(MwCoords(:,1),3);

Then assign the elements accordingly as ‘1’, ‘2’, or ‘3’.

EDIT — I originally posted the full answer before noticing that this was tagged as homework. That part is now gone.

.

[Conner Carriere]

This is for my final project that is not due for a bit. I guess I am a bit confused how to assign certain parts of a matrix. Also I do not understand why the ix1 outputs 1,2,3 when there are 9 variables that it is reading.

[MwCoordsMins,ix1] = mink(MwCoords(:,1),3);
ix1 = 1
      2
      3

[Star Strider]

If it’s not actually homework, I’ll post the rest of it:

‘I need to put 1's in the mins, and 3's in the max of the origional MwCoords

From there I need to assign the 3 middle values to 2's’

So:

[MwCoordsMins,ix1] = mink(MwCoords(:,1),3);
[MwCoordsMaxs,ix2] = maxk(MwCoords(:,1),3);

MwCoords(:,1) = 2;
MwCoords(ix1,1) = 1;
MwCoords(ix2,1) = 3;

produces:

MwCoords =
            1       360.11
            1        251.2
            1       148.83
            2       144.97
            2       255.66
            2       364.55
            3       149.57
            3       256.37
            3       360.44

I’m not certain what you want to do with the rest, so I’ll leave that to you.

[Conner Carriere]

Thank you for your help!

[Star Strider]

As always, my pleasure!

[Conner Carriere]

Sorry to bother, but I got everthing working, but some issue pops up and im unlcear why.

I am creating a function for this

function MCoords = CoordsMinMax(g)

%This is for the x values

[MwCoordsMinsx,ix1] = mink(g(:,1),3);
[MwCoordsMaxsx,ix2] = maxk(g(:,1),3);
MCoords(:,1) = 2;
MCoords(ix1,1) = 1;
MCoords(ix2,1) = 3;

%this is for the y values
[MCoordsMinsy,iy1] = mink(g(:,2),3);
[MCoordsMaxsy,iy2] = maxk(g(:,2),3);
MCoords(:,2) = 2;
MCoords(iy1,2) = 1;
MCoords(iy2,2) = 3;

end

This is how I call it in the script

Mw2Coords = CoordsMinMax(MwCoords)

It is just called Mw2Coords now, so I dont overwrite MwCoords while working on it.

Here is the original Mw Coords

MwCoords =
  241.5493  360.1065
  241.7508  251.1975
  246.9076  148.8279
  350.6785  144.9718
  349.2095  255.6598
  352.6571  364.5513
  456.4168  149.5675
  459.3270  256.3704
  460.4547  360.4354
  

The output of the function is

  Mw2Coords =
     1     3
     1     2
     1     1
     0     1
     0     2
     0     3
     3     1
     3     2
     3     3

Why do those 3 0's come up?

[Star Strider]

I’m not exactly certain what the problem is since I didn’t see the function call itself.

This code (with your original matrix):

[MwCoordsMins,ix1] = mink(MwCoords(:,1),3);
[MwCoordsMaxs,ix2] = maxk(MwCoords(:,1),3);

MwCoords(:,1) = 2;
MwCoords(ix1,1) = 1;
MwCoords(ix2,1) = 3;

[MCoordsMinsy,iy1] = mink(MwCoords(:,2),3);
[MCoordsMaxsy,iy2] = maxk(MwCoords(:,2),3);

MwCoords(:,2) = 2;
MwCoords(iy1,2) = 1;
MwCoords(iy2,2) = 3;

produces this result:

MwCoords =
     1     3
     1     2
     1     1
     2     1
     2     2
     2     3
     3     1
     3     2
     3     3

.

[dpb]

function MCoords = CoordsMinMax(g)
  % Replace 3 min, max X,Y coordinates --> 1,3 respectively; remainder --> 2
  [~,imnx] = mink(g(:,1),3);    % indices min 3 in X
  [~,imxx] = maxk(g(:,1),3);    % indices max 3 in X
  [~,imny] = mink(g(:,2),3);    % ditto for Y
  [~,imxy] = maxk(g(:,2),3);
    
  MCoords=2*ones(size(g));      % set output to 2 same size as input
  MCoords(imnx,1) = 1;          % minima are 1
  MCoords(imxx,1) = 3;          % maxima are 3
  MCoords(imny,1) = 1;          % ditto
  MCoords(imxy,1) = 3;
end

[Star Strider]

Conner Carriere —

Put this assignment as the first line in your function, the problem then disappears, and the output is correct:

MCoords = g;

(Note that dpb’s solution does essentially the same thing.)

[dpb]

NB: original had column 1 in all; now fixed above -- dpb