search for: numresp

...cptable = matrix(double(numcp*cpcol), nrow=cpcol), dsplit = matrix(double(1), nsplit,3), isplit = matrix(integer(1), nsplit,3), csplit = catmat, dnode = matrix(double(1), nodes, 3+numresp), inode = matrix(integer(1), nodes, 6), PACKAGE = "rpart") Which in communicates with the C code in "s_to_rp.c" void s_to_rp2(Sint *n, Sint *nsplit, Sint *nnode, Sint *ncat, Sint *numcat, Sint *...

...rop[i]) direc[i] <- -1; } goodness <- (lprop + rprop); ret <- list(goodness= goodness, direction=direc) print("***** END: TEMP2 *****"); ret } # The init function: # fix up y to deal with offsets # return a dummy parms list # numresp is the number of values produced by the eval routine's "label" # numy is the number of columns for y # summary is a function used to print one line in summary.rpart # text is a function used to put text on the plot in text.rpart # In general, this function would also check for b...

Dear R community, I am trying to write my own user defined split function for rpart. I read the example in the tests directory and I understand the general idea of the how to implement user defined splitting functions. However, I am having troubles with addressing the data frame used in calling rpart in my split functions. For example, in the evaluation function that is called once per node,

...<=i]~-1))^2)) gd[x==i]<-ndev-(lft+rt) #dir[x==i]<-1 } list(goodness=gd , direction=) } else{ list(goodness= , direction = ) } } tst.init<-function(y, offset, parms, wt){ browser() if (!is.null(offset)) y <- y-offset list(y=y, parms=0, numresp=2, numy=2, summary= function(yval, dev, wt, ylevel, digits ) { paste(" mean=", format(signif(yval, digits)), ", MSE=" , format(signif(dev/wt, digits)), sep='') }, text= function(yval, dev, wt, ylevel, digits, n, use.n ) { if(use.n) {pa...

...cptable = matrix(double(numcp*cpcol), nrow=cpcol), dsplit = matrix(double(1), nsplit,3), isplit = matrix(integer(1), nsplit,3), csplit = catmat, dnode = matrix(double(1), nodes, 3+numresp), deltaI = matrix(double(1), nodes, nsplit), inode = matrix(integer(1), nodes, 6)) -- Clayton Springer, Ph. D. Sandia National Laboratories csprin@ca.sandia.gov Biosystems Research Department (925) 29...

...####################################User defined function ##################################################################### temp.init<-function(y,offset,parms,wt){ if (!is.null(offset)) y<-y-offset if (is.matrix(y))stop ("response must be a vector") list(y=y,parms=0,numy=1,numresp=1, summary=function(yval,dev,wt,ylevel,digits){ paste("mean=",format(signif(yval,digits)), "MSE=",format(signif(dev/wt,digits)), sep='') }) } temp.eval<-function(y,wt,parms){ n<-length(y) r<-wt for (i in 1:n-1) {r[i+1]=(...

...is point the rpart run will complete but will have many NaN's in the rpart$frame for yval's. This is because of an oversight in rpart.s when calculating the probabilities for each node. The change to the function rpart in rpart.s is: if (method.int ==3 ) { numclass <- init$numresp -1 temp <- rp$dnode[,-(1:4)] %*% diag(init$parms$prior* sum(init$counts)/init$counts) # Bob's clumsy attempt to remove the NaN's that result when # calculating the probabilities tempx <- matrix(0, nrow=nrow(temp), ncol=ncol(temp)) tempx[which(is.finite(temp))] <...

...is point the rpart run will complete but will have many NaN's in the rpart$frame for yval's. This is because of an oversight in rpart.s when calculating the probabilities for each node. The change to the function rpart in rpart.s is: if (method.int ==3 ) { numclass <- init$numresp -1 temp <- rp$dnode[,-(1:4)] %*% diag(init$parms$prior* sum(init$counts)/init$counts) # Bob's clumsy attempt to remove the NaN's that result when # calculating the probabilities tempx <- matrix(0, nrow=nrow(temp), ncol=ncol(temp)) tempx[which(is.finite(temp))] <...