Too less/many parameters: You supplied 15??? One or more output arguments not assigned during call to "mexClassRF_train". about randomforest-matlab

[GoogleCodeExporter]

hello, thanks for sending the bug report

i don't have a mac machine currently to test this and will take me till the end 
of 
the week to get a hold of one. 

it will be great if you can post what arguments you were passing, so i can 
setup a 
dummy test. 

like, argument1: size NxD and type, you can get that via whos('argument1')


You can also put in a debug point at line 353 or classRF_Train.m and see if all 
of 
these variables are defined at that point. 
X',int32(Y_new),length(unique(Y)),ntree,mtry,int32(ncat),                       

int32(maxcat), int32(sampsize), strata, Options, int32(ipi),                    

classwt, cutoff, int32(nodesize),int32(nsum), int32(n_size), int32(p_size), 
int32(nsample)

if either one of them is non-existent, it will explain the error message.


I can suggest you compile and run the mex from the source rather than using the 
precompiled mex if you need results in the next few days.

Original comment by abhirana on 16 Feb 2010 at 2:11

[GoogleCodeExporter]

I'm having the same problem; RF classification on the Mac does not work.  All 
variables exist and are defined when passed into mexClassRF_train() but I still 
get the error:


Too less/many parameters: You supplied 15??? One or more output arguments not 
assigned during call to "mexClassRF_train".

Error in ==> classRF_train at 347
    [nrnodes,ntree,xbestsplit,classwt,cutoff,treemap,nodestatus,nodeclass,bestvar,ndbigtree,mtry ...

Original comment by [email protected] on 3 Oct 2011 at 5:03

[GoogleCodeExporter]

i am so sorry. i could never get access to a mac machine to compile the binary 
so uploaded a user-supplied binary.

these instructions may help you compile directly from the source 
http://code.google.com/p/randomforest-matlab/issues/detail?id=8

Original comment by abhirana on 6 Oct 2011 at 12:55