Glmnet varimp fix

models/files/glmnet.R

@@ -116,7 +116,7 @@ modelInfo <- list(label = "glmnet",
                       if(length(lambda) > 1) stop("Only one value of lambda is allowed right now")
                       if(!is.null(x$lambdaOpt)) {
                         lambda <- x$lambdaOpt
-                      } else stop("must supply a vaue of lambda")
+                      } else stop("must supply a value of lambda")
                     }
                     allVar <- if(is.list(x$beta)) rownames(x$beta[[1]]) else rownames(x$beta)
                     out <- unlist(predict(x, s = lambda, type = "nonzero"))
@@ -132,14 +132,14 @@ modelInfo <- list(label = "glmnet",
                       if(length(lambda) > 1) stop("Only one value of lambda is allowed right now")
                       if(!is.null(object$lambdaOpt)) {
                         lambda <- object$lambdaOpt
-                      } else stop("must supply a vaue of lambda")
+                      } else stop("must supply a value of lambda")
                     }
                     beta <- predict(object, s = lambda, type = "coef")
                     if(is.list(beta)) {
                       out <- do.call("cbind", lapply(beta, function(x) x[,1]))
                       out <- as.data.frame(out)
                     } else out <- data.frame(Overall = beta[,1])
-                    out <- out[rownames(out) != "(Intercept)",,drop = FALSE]
+                    out <- abs(out[rownames(out) != "(Intercept)",,drop = FALSE])
                     out
                   },
                   levels = function(x) if(any(names(x) == "obsLevels")) x$obsLevels else NULL,

pkg/caret/DESCRIPTION

@@ -1,6 +1,6 @@
 Package: caret
-Version: 6.0-51
-Date: 2015-07-12
+Version: 6.0-50
+Date: 2015-07-04
 Title: Classification and Regression Training
 Author: Max Kuhn. Contributions from Jed Wing, Steve Weston, Andre
     Williams, Chris Keefer, Allan Engelhardt, Tony Cooper, Zachary Mayer,

pkg/caret/inst/NEWS.Rd

@@ -6,7 +6,7 @@
 	\itemize{
 		\item A new model using the \cpkg{randomForest} and \cpkg{inTrees} packages called \code{rfRules} was added. A basic random forest model is used and then is decomposed into rules (of user-specified complexity). The \cpkg{inTrees} package is used to prune and optimize the rules. Thanks to Mirjam Jenny who suggested the workflow. 
 		\item Other new models (and their packages): \code{bartMachine} (\cpkg{bartMachine}), \code{rotationForest} (\cpkg{rotationForest}), \code{sdwd} (\cpkg{sdwd}), \code{loclda} (\cpkg{klaR}), \code{nnls} (\cpkg{nnls}), \code{svmLinear2} (\cpkg{e1071}), \code{rqnc} (\cpkg{rqPen}), and \code{rqlasso} (\cpkg{rqPen})
-		\item When specifying your own resampling indices, a value of \code{method = "custom"} can be used with \code{trainControl} for better printing. 
+		\item When specifying your own resampling indicies, a value of \code{method = "custom"} can be used with \code{trainControl} for better printing. 
 		\item Tim Lucas fixed a bug in \code{avNNet} when \code{bag = TRUE}
 		\item Fixed a bug found by ruggerorossi in \code{method = "dnn"} with classification.
 		\item A new option called \code{sampling} was added to \code{trainControl} that allows users to subsample their data in the case of a class imbalance. Another \href{http://topepo.github.io/caret/sampling.html}{help page} was added to explain the features. 
@@ -16,7 +16,7 @@
 		\item An internal function (\code{class2ind}) that can be used to make dummy variables for a single factor vector is now documented and exported.  
 		\item A bug was fixed in the \code{xyplot.lift} where the reference line was incorrectly computed. Thanks to Einat Sitbon for finding this. 
 		\item A bug related to calculating the Box-Cox transformation found by John Johnson was fixed.  
-		} 
+		}
 }
 
 \section{Changes in version 6.0-47}{

pkg/caret/tests/testthat/test_varImp.R

@@ -0,0 +1,26 @@
+library(caret)
+
+context('Testing varImp')
+
+test_that('glmnet varImp returns non-negative values', {
+  skip_on_cran()
+#  library(glmnet, verbose=TRUE)
+  skip_if_not_installed('glmnet')
+  set.seed(1)
+  dat <- SLC14_1(200)
+
+  reg <- train(y ~ ., data = dat,
+                      method = "glmnet",
+                      tuneGrid = data.frame(lambda = .1, alpha = .5),
+                      trControl = trainControl(method = "none"))
+
+  # this checks that some coefficients are negative
+  coefs <- predict(reg$finalModel, s=0.1, type="coef")
+  expect_less_than(0, sum(0 > coefs))
+  # now check that all elements of varImp are nonnegative,
+  # in spite of negative coefficients
+  vis <- varImp(reg, s=0.1, scale=F)$importance
+  expect_equal(0, sum(0 > vis))
+})
+
+