{"id":2362,"date":"2008-12-19T16:20:38","date_gmt":"2008-12-19T16:20:38","guid":{"rendered":"https:\/\/blogs.mathworks.com\/pick\/2008\/12\/19\/counting-elements\/"},"modified":"2008-12-19T16:20:38","modified_gmt":"2008-12-19T16:20:38","slug":"counting-elements","status":"publish","type":"post","link":"https:\/\/blogs.mathworks.com\/pick\/2008\/12\/19\/counting-elements\/","title":{"rendered":"Counting Elements"},"content":{"rendered":"
\r\n \r\n

Jiro<\/a>'s pick this week is COUNTMEMBER<\/a> by Jos<\/a>.\r\n <\/p>\r\n <\/introduction>\r\n

A while back, I wrote a Pick of the Week post<\/a> on a file that created a MATLAB file template. In that post, I talked a bit about a \"best practice\" for formatting your MATLAB\r\n files: things like, having an H1 line, including help text, including examples, etc.\r\n <\/p>\r\n

How is that related to this week's pick? Well, I chose countmember<\/tt> because Jos writes very well-written files. (I could have chosen some of his other files, but I liked the code also.) Open\r\n the file up and see it for yourself. He has a simple, but descriptive H1 line so that LOOKFOR<\/tt> will be able to find it. The help block follows the format that many MathWorks MATLAB files use: capitalized syntax (but\r\n case-sensitive example code), the examples can be directly executed (highlight and F9), and he has a \"See also\" section for\r\n related functions.\r\n <\/p>

help countmember<\/span><\/pre>
  COUNTMEMBER - count members\r\n \r\n    C = COUNTMEMBER(A,B) counts the number of times the elements of A occurs\r\n    in B, such the Ci = sum(B==Ai). C has the same size and shape as A.\r\n    A may contain non-unique elements. A and B should be of the same type. \r\n    A can be be a cell array.\r\n \r\n    Examples:\r\n      countmember([1 2 1 3],[1 2 2 2 2]) \r\n         -> 1     4     1     0\r\n      countmember({'a','b','c'},{'a','x','a'}) \r\n         -> 2     0     0\r\n \r\n    See also ISMEMBER, UNIQUE, HISTC\r\n \r\n    2005 Jos van der Geest\r\n\r\n<\/pre>
Of course, the other reason I selected this was his elegant solution to the task of counting the occurrence of elements. The\r\n      code has a nice error checking mechanism and deals with empty input arguments. The code is vectorized (no for-loops) and uses\r\n      UNIQUE<\/tt> to eliminate redundant search. Also, notice the little trick to reduce overhead in his lines of code for UNIQUE<\/tt> and ISMEMBER<\/tt>.\r\n   <\/p>\r\n   
Comments<\/a>?\r\n   <\/p>\r\n   
We will be taking next Friday off for holiday, but we'll be back next year on the 2nd of January. Have a safe and happy holiday!<\/b><\/i><\/p>