Displaying 20 results from an estimated 2000 matches similar to: "Problem in getting 632plus error using randomForest by ipred!"
2005 Jan 06
1
different result from the same errorest() in library( ipred)
Dear all,
Does anybody can explain this: different results got when all the same parameters are used in the errorest() in library ipred, as the following?
errorest(Species ~ ., data=iris, model=randomForest, estimator = "cv", est.para=control.errorest(k=3), mtry=2)$err
[1] 0.03333333
> errorest(Species ~ ., data=iris, model=randomForest, estimator = "cv",
2005 Jun 23
1
errorest
Hi,
I am using errorest function from ipred package.
I am hoping to perform "bootstrap 0.632+" and "bootstrap leave one out".
According to the manual page for errorest, i use the following command:
ce632[i]<-errorest(ytrain ~., data=mydata, model=lda,
estimator=c("boot","632plus"), predict=mypredict.lda)$error
It didn't work. I then tried the
2009 Apr 25
1
Overlapping parameters "k" in different functions in "ipred"
Dear List,
I have a question regarding "ipred" package. Under 10-fold cv, for different knn ( = 1,3,...25), I am getting same misclassification errors:
#############################################
library(ipred)
data(iris)
cv.k = 10 ## 10-fold cross-validation
bwpredict.knn <- function(object, newdata) predict.ipredknn(object, newdata, type="class")
for (i in
2005 Jan 10
0
Stadard errors and boxplots with 632plus error estimator, "errorest"
Dear R-users,
I'd like to estimate standard errors (for lda) and make a boxplot with the
"632plus" and "boot" error estimators included in package ipred (method:
errorest). The "boot" estimator returns only a standard deviation but not
the whole error data.
Thank you in advance,
regards,
Antoine
2009 Nov 02
1
modifying predict.nnet() to function with errorest()
Greetings,
I am having trouble calculating artificial neural network
misclassification errors using errorest() from the ipred package.
I have had no problems estimating the values with randomForest()
or svm(), but can't seem to get it to work with nnet(). I believe
this is due to the output of the predict.nnet() function within
cv.factor(). Below is a quick example of the problem I'm
2004 Jan 09
3
ipred and lda
Dear all,
can anybody help me with the program below? The function predict.lda
seems to be defined but cannot be used by errortest.
The R version is 1.7.1
Thanks in advance,
Stefan
----------------
library("MASS");
library("ipred");
data(iris3);
tr <- sample(1:50, 25);
train <- rbind(iris3[tr,,1], iris3[tr,,2], iris3[tr,,3]);
test <- rbind(iris3[-tr,,1],
2012 Nov 09
0
10-Fold Cross Validation AND Random Forest
Hi,
I am using the Random Forest package to classify observations into one of two classes. My data is unbalanced with the minority class accounting for 7% of total data set.
I have heard the 10-Fold Cross validation can help me with improving classification. But being new at most of this it's not something I can do from scratch on my own. So I have spent all this morning trying to find a
2005 Mar 18
2
logistic model cross validation resolved
This post is NOT a question, but an answer. For readers please disregard all earlier posts by myself about this question.
I'm posting for two reasons. First to say thanks, especially to Dimitris, for suggesting the use of errorest in the ipred library. Second, so that the solution to this problem is in the archives in case it gets asked again.
If one wants to run a k-fold cross-validation
2005 Jan 06
1
leave-one-out cross validation for randomForest
Dear all,
Can I get the leave-one-out cross validation error of randomForest in
R? I only found tune(), which got the 10-fold cross validation error.
Thanks for any information.
Xin LIU
This e-mail is from ArraDx Ltd
The e-mail and any files transmitted with it are confidentia...{{dropped}}
2006 Feb 02
0
crossvalidation in svm regression in e1071 gives incorrect results (PR#8554)
Full_Name: Noel O'Boyle
Version: 2.1.0
OS: Debian GNU/Linux Sarge
Submission from: (NULL) (131.111.8.96)
(1) Description of error
The 10-fold CV option for the svm function in e1071 appears to give incorrect
results for the rmse.
The example code in (3) uses the example regression data in the svm
documentation. The rmse for internal prediction is 0.24. It is expected the
10-fold CV rmse
2006 Feb 02
0
crossvalidation in svm regression in e1071 gives incorre ct results (PR#8554)
1. This is _not_ a bug in R itself. Please don't use R's bug reporting
system for contributed packages.
2. This is _not_ a bug in svm() in `e1071'. I believe you forgot to take
sqrt.
3. You really should use the `tot.MSE' component rather than the mean of
the `MSE' component, but this is only a very small difference.
So, instead of spread[i] <- mean(mysvm$MSE), you
2005 Jun 24
1
mypredict.
Hi,
I am wondering what does "mypredict.lda<-function(object,
newdata)predict(object, newdata=newdata)$class" actually do?
I run a few errorest commands in the same function on the same dataset using
the same classifier lda. The only difference is some use "cv", other use
"boot" and "632plus". They all share one mypredict.lda.
Will it cause any
2007 Aug 30
0
rpart's loss matrix in ipred
Dear R users,
I have been using the rpart procedure to predict the occurrence of
depression in a large data file. Since the prevalence is very low (5%),
I have been using classification trees with a loss matrix that penalized
false negatives more than false positives. I have become interesested in
bagging these (successful!) classification trees, and have been using
the ipred package for
2003 Apr 16
2
Jackknife and rpart
Hi,
First, thanks to those who helped me see my gross misunderstanding of
randomForest. I worked through a baging tutorial and now understand the
"many tree" approach. However, it is not what I want to do! My bagged
errors are accpetable but I need to use the actual tree and need a single
tree application.
I am using rpart for a classification tree but am interested in a more
unbaised
2009 Sep 11
0
ipred bagging segfault on 64 bit linux build
I wanted to report this issue here so others may not find themselves alone
and as the author is apparently active on the list.
I havent done an exhaustive test by any means, cause I dont have time. But
here's a small example. Apparently the "ns" argument is the one that is
killing it. I've gotten several different segfault messages, the only other
one I remember said "out
2002 Apr 10
1
New Package: ipred - Improved predictors
The package ipred is uploaded to CRAN.
The main focus of the package is the calculation of improved predictors
in classification tasks. Misclassification error can be improved by
bootstrap aggregated classification trees and/or
the framework of indirect classification. Furthermore, a unified
interface for the estimation of misclassification error
completes the features of ipred.
We try to make
2002 Apr 10
1
New Package: ipred - Improved predictors
The package ipred is uploaded to CRAN.
The main focus of the package is the calculation of improved predictors
in classification tasks. Misclassification error can be improved by
bootstrap aggregated classification trees and/or
the framework of indirect classification. Furthermore, a unified
interface for the estimation of misclassification error
completes the features of ipred.
We try to make
2006 Aug 01
0
rsurv in ipred
Hi,
I'm trying to find information about rsurv "Simulating Survival data" in
the IPred package, without luck this far. In the description of this
function we are asked to consult Hothorn et al. (2003) for the details.
This paper is not in the reference list. Should it be same authors (2004)?
In that case I will try my library, in any other case could someone please
give me
2009 Jan 22
4
dimnames in pkg "ipred"
Hello List,
I`m trying to make prediction using a bagged tree with the package ipred. I tried to follow the manual but I`m getting an error message. Also browsing through the list-archive I didn`t find any hint.
Maybe someone can help me?
selbag <- bagging(SOIL_UNIT ~., data=traindat.bin, coob=TRUE)
Error in dimnames(X) <- list(dn[[1L]], unlist(collabs, use.names = FALSE)) :
2005 Jul 29
0
PLS component selection for GPLS question
How to select the number of PLS components for GPLS for data sets
with few samples?
Concrete problem:
My data set: 9 samples of class A and 37 of class B with 254
descriptors.
In the paper: "Classification Using Generalized Partial Least
Squares", Beiying Ding, Robert Gentleman, Bioconductor
Project Working Papers, year 2004, paper 5
Section 2.6 Assessing Prediction:
Cite: