Hi All, I am looking for some help figuring out what is causing an error in my attempt to fit a regularized logistic regression (specifically finding the optimal lambda value using cv.glmnet). Running the following command: RegLR_CV<-cv.glmnet(x=train.sub.clean[,-c(431)],y=as.factor(train.sub$finalAttrite),family="binomial") where the table train.sub.clean are all numeric and has had all zero variance variables removed (using the carat package). This function runs for a while then produces the following error. Anyone recommend what I should be looking for to troubleshoot? It must be my data set as the toy examples in the package run fine. Error: Error in as.matrix(cbind2(1, newx) %*% nbeta) : error in evaluating the argument 'x' in selecting a method for function 'as.matrix': Error in t(.Call(Csparse_dense_crossprod, y, t(x))) : error in evaluating the argument 'x' in selecting a method for function 't': Error: invalid class 'NA' to dup_mMatrix_as_dgeMatrix R: R version 2.13.0 (2011-04-13) Platform: x86_64-pc-mingw32/x64 (64-bit) glmnet: 1.7 [[alternative HTML version deleted]]