
Re: ClassificationKNN.fit with cell array of strings as Y
Posted:
May 7, 2013 1:04 PM


On 5/7/2013 12:55 PM, Alan_Weiss wrote: > On 5/7/2013 11:07 AM, Jay wrote: >> >>> I am not sure where in the documentation you found your statement 1. >>> Here's what I see when I look at the current documentation: >>> http://www.mathworks.com/help/stats/classificationknn.fit.html >>> >>> Y  Grouping variables of response values with the same number of >>> elements (rows) as X. Each entry in Y is the response to the data in >>> the corresponding row of X. >>> >>> This does not say that the variable must be a numeric vector, but a >>> grouping variable >>> http://www.mathworks.com/help/stats/groupingvariables.html >> >> Hm, yes, correct. Here is what I was referring to: >> http://www.mathworks.de/de/help/stats/classificationknnclass.html >> >> Y Numeric vector of response values with the same number of rows as >> X. Each entry in Y is the response to the data in the corresponding >> row of X. >> >>> As far as your problem 2, I am not sure what is going on. I wonder >>> if your function myDistance is giving you a problem somehow. >> >> So you believe everything should work fine when using a cell array of >> strings. I don't see how my distance function could cause such an >> error. My distance function is defined as [ distance ] = myDistance( >> obj1, objects) >> and will always return at least "distance = zeros(size(objects, >> 1),1); " or higher values. > > There must be something strange about the German web site, because > when I try to look at the documentation page for the ClassificationKNN > class, I do not see the "Numeric vector..." that you do, even when I > look at the R2012b version of the documentation. I will try to > investigate this discrepancy. Thanks for reporting the issue. > > And, as I said, I really do not know what is going on with the predict > method that is causing those errors. I will try to look into it, but > it might take a few days. > > Alan Weiss > MATLAB mathematical toolbox documentation
On second thought, perhaps you should contact Technical Support. If you really found a bug (and I am satisfied that, if there is a bug, it is not in the documentation), then tech support is probably in the best position to research the problem and help to find a solution.
Alan Weiss MATLAB mathematical toolbox documentation

