Displaying 5 results from an estimated 5 matches for "numclass".
Did you mean:
mclass
2009 Apr 30
1
unexpected behavior of rpart 3.1-43, loss matrix
...p = ".")))(Y, offset, parms, wt)
: Wrong length for loss matrix
note that while the levels of the factor range from 1:6, for the
concrete obseration data, only levels 1, 2, 3, 5 do occur.
the error is caused by the code of rpart.class:
fy <- as.factor(y)
y <- as.integer(fy)
numclass <- max(y[!is.na(y)])
...
temp2 <- parms$loss
if (length(temp2) != numclass^2)
stop("Wrong length for loss matrix")
for the example, numclass is set to 5 instead of 6.
while for that small example, it may be discussable whether or not
numclass should be 6, consider a set of da...
2002 Jan 25
0
rpart subsets
...ro. Here is the change to gini.c:
#if 1 // Bob's simple fix
// This fixes a problem when predicting a categorical using a
subset.
// The subset may be missing one or more of the categories. Here
// we make any zeros a very small number so that the rpart run will
not fail.
for (i=0; i<numclass; i++) {
if (freq[i] == 0) freq[i] = 0.00000000001;
}
#endif
for (i=0; i<numclass; i++) {
prior[i] /= freq[i];
aprior[i] /= (temp * freq[i]); /* pi_i / n_i */
}
}
The calculation of the priors (existing code) above shows where I put the
fix in giniinit (in gini.c...
2002 Jan 28
0
rpart subset fix
...ro. Here is the change to gini.c:
#if 1 // Bob's simple fix
// This fixes a problem when predicting a categorical using a
subset.
// The subset may be missing one or more of the categories. Here
// we make any zeros a very small number so that the rpart run will
not fail.
for (i=0; i<numclass; i++) {
if (freq[i] == 0) freq[i] = 0.00000000001;
}
#endif
for (i=0; i<numclass; i++) {
prior[i] /= freq[i];
aprior[i] /= (temp * freq[i]); /* pi_i / n_i */
}
}
The calculation of the priors (existing code) above shows where I put the
fix in giniinit (in gini.c...
2002 Mar 13
0
rpart error with 0-frequency factor levels (with partial fix) (PR#1378)
...-----------
I tried to trace the problem, and took a look at the beginning of
rpart.class:
"rpart.class" <-
function (y, offset, parms, wt)
{
if (!is.null(offset))
stop("No offset allowed in classification models")
fy <- as.factor(y)
y <- as.integer(fy)
numclass <- max(y[!is.na(y)])
...
}
I thought the problem might have to do with the fact that numclass would
register 3 for my examples t3 and t5, but 2 for my examples t2 and t4. So I
tried substituting
numclass <- length(unique(y[!is.na(y)]))
to get the number of factors present rather than the...
2006 Nov 30
0
extending data.frame with S4 class
...te an instance, I get something
that looks and acts like a numeric. When I extend "data.frame" and
create an instance, I get a NULL 'core'. Why is this, and can it be
fixed? setOldClass(), prototype=, and initialize() don't seem to help.
Thanks,
Tim
----
setClass("numClass",representation("numeric",num="numeric"))
nn <- new("numClass",2,num=1)
nn
nn+1
setClass("dfClass",representation("data.frame",num="numeric"))
df <- new("dfClass",as.data.frame(BOD),num=1)
df