search for: jhibschman

Hi all, I'm trying to make an integer-backed quarter (as in fraction of year) class, but I can't quite it to work. I want integer-backed so I don't have to worry about floating-point effects when doing math, and so that I can use it as in data.table. First of all, is there a good reference for this anywhere? All of the S4 tutorials that I've found have been too high-level, and

What is the current best package for manipulating HDF5 data files? I tried "hdf5" a long time ago, but I ran into memory problems. "h5r" is on CRAN now, and "rhdf5" is part of bioconductor. Ideally, I'd like to read simple vectors or tables, either the entire thing or a subset of rows. I don't need much writing support, but it would be nice. Compression is a

Is there any way to get the current memory used by R without running gc()? I'd like to include the memory usage in logging output, but calling gc() to get that information takes long enough to be noticeable (~ 6 s with ~ 20 GB of ram in use), which goes against the point of logging. Thanks, Johann

I'm using R 2.10.0, with zoo 1.5-8. The release notes for zoo 1.5-8 claim a bug with unique for yearmon objects has been fixed, but I'm still having problems. Browse[1]> tmp2 [1] "Dec 1996" "Dec 1996" Browse[1]> unique(tmp2) [1] "Dec 1996" "Dec 1996" Browse[1]> unique(unique(tmp2)) [1] "Dec 1996" Browse[1]> as.numeric(tmp2) -

I'd like to access data in my R session from elsewhere via HTTP. My particular use case would be R on Linux, accessing data from Windows, all on the same intranet. Naively, when I say this, I imagine something like: > theData <- big.calculation.returning.data.frame() > startHttpServer(port=8675) (blocks until I C-c it) Elsewhere... http://linuxmachine:8675/htmlDF?theData

When I run predict.Arima in my code, I get warnings like: Warning message: In cbind(intercept = rep(1, n), xreg) : number of rows of result is not a multiple of vector length (arg 1) I think this is because I'm not running predict.Arima in the same environment that I did the fit, so the data object used in the fit is no longer present. Looking at the predict.Arima source,

I'm having trouble calling randomForest::partialPlot programmatically. It tries to use name of the (R) variable as the data column name. Example: library(randomForest) iris.rf <- randomForest(Species ~ ., data=iris, importance=TRUE, proximity=TRUE) partialPlot(iris.rf, iris, Sepal.Width) # works partialPlot(iris.rf, iris, "Sepal.Width") # works (function(var.name)

I'm using 2.10.0 on Linux (64 bit), and I just noticed that random numbers are occasionally added to the text of names in vectors. It's happened to me in two separate, long-running R sessions, but I can't find a way to reproduce it in a smaller setting. The code I'm using is > diag.gam.2 <- mdl.run.diag(fit.gam.2, ds.valid) > diag.gam.2 rmse mdae.1413751

I'm experimenting with a few model-fitting classes of my own. I'm leaning towards using S4 for my classes, but the R functions I'd want to override (residuals, predict, etc.) are all S3 methods. As I understand it, I could do setGeneric("residuals"), then add S4 specializations to it. However, I don't understand the full consequences of doing this. Are there any

I'm trying to plot the "marginals" of a fit: the aggregated value of the actual and predicted vs. a cut/bucketed dimension. (The data set is huge, so just plotting the raw points would be unintelligible.) I'd also like to plot the number of points in each bucket (or, rather, the sum of the weights in each bucket), so I can mentally discount crazy behavior at low weights. To do

Is there any package that assists in saving and reconstituting glm and nls fits without bringing along the accompanying data? A quick search on CRAN didn't turn up anything. If not, how do other people deal with saving the coefficients of model fits? For example, I've run a glm fit that has 23 coefficents on data set that had 193,008 rows, by the time the fit was called. When I save