Removing Rpeated Values from a Matrix

Hi

I have a matrix below:

NaN NaN

518460 3.3220

518460 -12.9840

518520 3.7890

518520 -10.2800

518580 7.6730

518580 16.8510

518640 5.9590

.

.

.

the values in the 1st column are repeated (2-4 times). i want to write a code which checks the values in the 1st column, if it is repeated, then it will read the respective 2nd column values and only keep the row with the largest 2nd column value while deleting the other rows. so the matrix above when ran through the code should be:

518460 3.3220

518520 3.7890

518580 7.6730

518580 16.8510 .

.

.

the data exceeds over 1000 entries but i have only given a small piece of it above.

how can i write the code so i get my second matrix from the first.

Answer 1

[Stephen23]

>> M = [518460,3.3220;518460,-12.9840;518520,3.7890;518520,-10.2800;518580,7.6730;518580,16.8510;518640,5.9590]
M =
   518460        3.32200
   518460      -12.98400
   518520        3.78900
   518520      -10.28000
   518580        7.67300
   518580       16.85100
   518640        5.95900
>> [uni,~,idx] = unique(M(:,1));
>> val = accumarray(idx,M(:,2),[],@max);
>>  [uni,val]
ans =
   518460        3.32200
   518520        3.78900
   518580       16.85100
   518640        5.95900

[Sarvesh Kumar]

thank you very much. it worked like a charm! I thought it would be a long code, did not expect a 3 line command! thanks alot once again.