Displaying 20 results from an estimated 700 matches similar to: "nnet and weights: error analysis using V&R example"
2004 Sep 23
0
nnet with weights parameter: odd error
Dear R-users
I use nnet for a classification (2 classes) problem. I use the code
CVnn1, CVnn2 as described in V&R.
The thing I changed to the code is: I define the (class) weight for each
observation in each cv 'bag' and give the vector of weights as parameter
of nnet(..weights = weight.vector...)
Unfortunately I get an error during some (but not all!) inner-fold cv runs:
2004 Mar 30
1
classification with nnet: handling unequal class sizes
I hope this question is adequate for this list
I use the nnet code from V&R p. 348: The very nice and general function
CVnn2() to choose the number of hidden units and the amount of weight
decay by an inner cross-validation- with a slight modification to use it
for classification (see below).
My data has 2 classes with unequal size: 45 observations for classI and
116 obs. for classII
With
2011 Jan 05
0
Nnet and AIC: selection of a parsimonious parameterisation
Hi All,
I am trying to use a neural network for my work, but I am not sure about my
approach to select a parsimonious model. In R with nnet, the IAC has
not been defined for a feed-forward neural network with a single hidden layer.
Is this because it does not make sens mathematically in this case?
For example, is this pseudo code sensible?
Thanks in advance for your help. I am sorry if this
2004 Jun 14
2
CVnn2 + nnet question
Hi,
I am trying to determine the number of units in the hidden layer
and the decay rate using the CVnn2 script found in MASS directory
(reference: pg 348,MASS-4).
The model that I am using is in the form of Y ~ X1 + X2 + X3...
+ X11 and the underlying data is time-series in nature.
I found the MASS and nnet package extremely useful (many thanks
to the contributors).
However I am getting
1998 Aug 18
1
Problem in "configure" for Solaris (cc) -- solved (partly) --
> From: Peter Dalgaard BSA <p.dalgaard@biostat.ku.dk>
>
> Martin Maechler <maechler@stat.math.ethz.ch> writes:
>
> > Reading the long output of ``cc -flags'',
> > I see that more than average optimization is done using
> > cc -xO[1-4]
> > #
And -O is -xO2. If you do want higher speed, you need to use
other flags too, and -fast is a
2005 Sep 08
0
Effect of data set size on calculation
Dear listers,
I have a piece of code which performs an ANOVA type of analysis on 2D GC
data. The code is shown below:
# ANOVA 2D GC analysis
# maxc <- number of samples
# nreps <- number of samples
maxc <- 2
nreps <- 4
sscl <- NULL
cmean <- NULL
#
# Initial stat. variable
#
dftot <- nrow(mat)-1
dfcl <- maxc - 1
dferr <- dftot - dfcl
totmean
2008 Nov 17
1
An array of an array of boxplots in lattice
Using the data set fgl in MASS the following code
layout(matrix(1:9,3,3))
for(i in 1:9){
boxplot(fgl[,i] ~ type, data = fgl,main=dimnames(fgl)[[2]][i])}
produces a 3 by 3 array of plots, each one of which consists of six
boxplots.
Is it possible to do this in lattice?
Steve
"R version 2.7.2 (2008-08-25)" on Ubuntu 6.06
2020 Nov 03
0
[PATCH 1/2] Revert "vhost-vdpa: fix page pinning leakage in error path"
On 2020/10/30 ??3:45, Si-Wei Liu wrote:
> This reverts commit 7ed9e3d97c32d969caded2dfb6e67c1a2cc5a0b1.
>
> Signed-off-by: Si-Wei Liu <si-wei.liu at oracle.com>
> ---
> drivers/vhost/vdpa.c | 119 +++++++++++++++++++++------------------------------
> 1 file changed, 48 insertions(+), 71 deletions(-)
I saw this has been reverted there
2020 Oct 01
0
[PATCH] vhost-vdpa: fix page pinning leakage in error path
Pinned pages are not properly accounted particularly when
mapping error occurs on IOTLB update. Clean up dangling
pinned pages for the error path. As the inflight pinned
pages, specifically for memory region that strides across
multiple chunks, would need more than one free page for
book keeping and accounting. For simplicity, pin pages
for all memory in the IOVA range in one go rather than
have
2020 Oct 01
0
[PATCH v2] vhost-vdpa: fix page pinning leakage in error path
Pinned pages are not properly accounted particularly when
mapping error occurs on IOTLB update. Clean up dangling
pinned pages for the error path. As the inflight pinned
pages, specifically for memory region that strides across
multiple chunks, would need more than one free page for
book keeping and accounting. For simplicity, pin pages
for all memory in the IOVA range in one go rather than
have
2005 Nov 10
1
question about the dataset fgl
Dear sir or ma'am,
I have a question about the dataset "fgl." The dataset seems to be in
the "VR" package, so I tried to download it from CRAN. However, after
downloading, when I tried to load the package, it was not in my package
list. I am wondering what is wrong.
Any advice on how to access the fgl dataset would be appreciated.
Thanks.
Seungho Huh
2012 Jan 13
0
Example of "task seeds" with R parallel. Critique?
Greetings:
In R parallel's vignette, there is a comment "It would however take
only slightly more work to allocate a stream to each task." (p.6).
I've written down a working example that can allocate not just one,
but several separate seeds for each task. (We have just a few project
here that need multiple streams). I would like to help work that up
for inclusion in the
2005 Apr 20
1
negative p-values from fisher's test (PR#7801)
Full_Name: Martha Nason
Version: 2.0.1
OS: Windows XP
Submission from: (NULL) (137.187.154.154)
I am running simulations using fisher's test on 2 x c tables and a very small
p.value from fisher's test (<2.2e-16) is returned as a negative number. Code
follows.
> set.seed(0)
> nreps.outer <-7
> pvalue.fisher <- rep(NA,nreps.outer)
>
> population1 <- c(
2006 Aug 24
2
Why are lagged correlations typically negative?
Recently, I was working with some lagged designs where a vector of
observations at one time was used to predict a vector of observations at
another time using a lag 1 design. In the work, I noticed a lot of
negative correlations, so I ran a simple simulation with 2 matched
points. The crude simulation example below shows that the correlation
can be -1 or +1, but interestingly if you do this
2012 Oct 20
1
Logistic regression/Cut point? predict ??
I am new to R and I am trying to do a monte carlo simulation where I
generate data and interject error then test various cut points; however, my
output was garbage (at x equal zero, I did not get .50)
I am basically testing the performance of classifiers.
Here is the code:
n <- 1000; # Sample size
fitglm <- function(sigma,tau){
x <- rnorm(n,0,sigma)
intercept <- 0
beta
2003 Apr 03
0
debugging question
Dear R-devel,
A user reported a strange problem with predict.randomForest in the
randomForest package yesterday, and I'm baffled by it. The code at the end
of the message produces the error. The problem is that, in
predict.randomForest, there's a .Fortran call to "runforest". One of the
arguments passed in is "countts", which is a vector of doubles. The error
2009 Feb 18
1
Training nnet in two ways, trying to understand the performance difference - with (i hope!) commented, minimal, self-contained, reproducible code
Dear all,
Objective: I am trying to learn about neural networks. I want to see
if i can train an artificial neural network model to discriminate
between spam and nonspam emails.
Problem: I created my own model (example 1 below) and got an error of
about 7.7%. I created the same model using the Rattle package (example
2 below, based on rattles log script) and got a much better error of
about
2012 Dec 01
0
[LLVMdev] problem trying to write an LLVM register-allocation pass
On 11/30/2012 6:36 PM, Lang Hames wrote:
>
>
> RBP is used as the frame pointer on x86 (hence its automatic
> appearance in your code), and shouldn't be allocated to any vreg in
> function bar. Loading/saving RBP should be managed by the stack frame
> setup/teardown code.
> If it doesn't already, your allocator should filter out reserved
> registers (See
1998 Jul 20
0
R-beta: V&R libraries, tree for 0.62.2, R Complements for V&R
Versions of the V&R libraries, VR_5.3pl025-1.tar.gz, and of
tree_0.2-2.tar.gz (giving most of the functionality of S's tree system
including all that I have ever used) are now available on CRAN's Vienna
node, and will propagate.
There is an early draft of an R user's complements on Venables & Ripley
available from
http://www.stats.ox.ac.uk/pub/MASS2/Compl.shtml
This is
1998 Jul 20
0
R-beta: V&R libraries, tree for 0.62.2, R Complements for V&R
Versions of the V&R libraries, VR_5.3pl025-1.tar.gz, and of
tree_0.2-2.tar.gz (giving most of the functionality of S's tree system
including all that I have ever used) are now available on CRAN's Vienna
node, and will propagate.
There is an early draft of an R user's complements on Venables & Ripley
available from
http://www.stats.ox.ac.uk/pub/MASS2/Compl.shtml
This is