[ad_1]
\n<\/p>\n
Another approach is through The second line of code is actually quite elegant. Let’s start with the first Calling the above code gives us:<\/p>\nbsxfun<\/code><\/a> and unique<\/code><\/a>. Assuming that A<\/code> is a row vector, you would do the following:<\/p>\nun = unique(A.', 'stable');\nout = bsxfun(@times, bsxfun(@eq, A, un), un);\n<\/code><\/pre>\nout<\/code> contains your desired result. This code deserves some explanation. The first line of code determines all unique entries in A<\/code> stored in un<\/code>, but provides them in stable<\/code> order, or in the order in which they are encountered. Not doing this would also sort<\/strong> the values too. I figure that the case of values from 1 to 4 is a simplified example and you’d want to do this for numbers in any arbitrary order and them appearing anywhere in the vector as well.<\/p>\nbsxfun<\/code> call that’s nested inside:<\/p>\nbsxfun(@eq, A, un)\n<\/code><\/pre>\nbsxfun<\/code> is a very nice function that broadcasts<\/strong> over singleton dimensions. What this means for our case is that A<\/code> is a row vector and un<\/code> is a column vector, and this would produce a matrix<\/strong> where we find the element-wise equality<\/strong> of two matrices – one matrix which duplicates row vectors of A<\/code> for as many elements as we have in un<\/code> and another matrix which duplicates column vectors for as many elements as there are in A<\/code>. The result of this is a matrix where each row tells you which locations in A<\/code> matched a particular value in un<\/code> starting from the beginning of un<\/code> (first row) to the end of un<\/code> (last row). The last part of this is to take this equality matrix and do an element-wise multiplication with un<\/code> using another bsxfun<\/code> call so that un<\/code> broadcasts itself again column-wise like what was done in bsxfun(@eq, A, un)<\/code>. The result would be to zero out the exact locations for each row that did not equal to the corresponding value in un<\/code> that is represented and we thus have our result.<\/p>\nExample<\/h1>\n
>> A = [1 1 1 1 1 2 2 2 3 3 3 4 4 4]\n\nA =\n\n Columns 1 through 13\n\n 1 1 1 1 1 2 2 2 3 3 3 4 4\n\n Column 14\n\n 4\n<\/code><\/pre>\n>> out\n\nout =\n\n Columns 1 through 13\n\n 1 1 1 1 1 0 0 0 0 0 0 0 0\n 0 0 0 0 0 2 2 2 0 0 0 0 0\n 0 0 0 0 0 0 0 0 3 3 3 0 0\n 0 0 0 0 0 0 0 0 0 0 0 4 4\n\n Column 14\n\n 0\n 0\n 0\n 4\n<\/code><\/pre>\n<\/p><\/div>\n