assess.glmnet: assess performance of a 'glmnet' object using test data.

In glmnet: Lasso and Elastic-Net Regularized Generalized Linear Models

assess performance of a 'glmnet' object using test data.

Description

Given a test set, produce summary performance measures for the glmnet model(s)

Usage

assess.glmnet(
  object,
  newx = NULL,
  newy,
  weights = NULL,
  family = c("gaussian", "binomial", "poisson", "multinomial", "cox", "mgaussian"),
  ...
)

confusion.glmnet(
  object,
  newx = NULL,
  newy,
  family = c("binomial", "multinomial"),
  ...
)

roc.glmnet(object, newx = NULL, newy, ...)

Arguments

object	Fitted "glmnet" or "cv.glmnet", "relaxed" or "cv.relaxed" object, OR a matrix of predictions (for roc.glmnet or assess.glmnet). For roc.glmnet the model must be a 'binomial', and for confusion.glmnet must be either 'binomial' or 'multinomial'
newx	If predictions are to made, these are the 'x' values. Required for confusion.glmnet
newy	required argument for all functions; the new response values
weights	For observation weights for the test observations
family	The family of the model, in case predictions are passed in as 'object'
...	additional arguments to predict.glmnet when "object" is a "glmnet" fit, and predictions must be made to produce the statistics.

Details

assess.glmnet produces all the different performance measures provided by cv.glmnet for each of the families. A single vector, or a matrix of predictions can be provided, or fitted model objects or CV objects. In the case when the predictions are still to be made, the ... arguments allow, for example, 'offsets' and other prediction parameters such as values for 'gamma' for 'relaxed' fits. roc.glmnet produces for a single vector a two column matrix with columns TPR and FPR (true positive rate and false positive rate). This object can be plotted to produce an ROC curve. If more than one predictions are called for, then a list of such matrices is produced. confusion.glmnet produces a confusion matrix tabulating the classification results. Again, a single table or a list, with a print method.

Value

assess.glmnet produces a list of vectors of measures. roc.glmnet a list of 'roc' two-column matrices, and confusion.glmnet a list of tables. If a single prediction is provided, or predictions are made from a CV object, the latter two drop the list status and produce a single matrix or table.

Author(s)

Trevor Hastie and Rob Tibshirani
Maintainer: Trevor Hastie [e-mail]

See Also

cv.glmnet, glmnet.measures and vignette("relax",package="glmnet")

Examples

data(QuickStartExample)
x <- QuickStartExample$x; y <- QuickStartExample$y
set.seed(11)
train = sample(seq(length(y)),70,replace=FALSE)
fit1 = glmnet(x[train,], y[train])
assess.glmnet(fit1, newx = x[-train,], newy = y[-train])
preds = predict(fit1, newx = x[-train, ], s = c(1, 0.25))
assess.glmnet(preds, newy = y[-train], family = "gaussian")
fit1c = cv.glmnet(x, y, keep = TRUE)
fit1a = assess.glmnet(fit1c$fit.preval, newy=y,family="gaussian")
plot(fit1c$lambda, log="x",fit1a$mae,xlab="Log Lambda",ylab="Mean Absolute Error")
abline(v=fit1c$lambda.min, lty=2, col="red")
data(BinomialExample)
x <- BinomialExample$x; y <- BinomialExample$y
fit2 = glmnet(x[train,], y[train], family = "binomial")
assess.glmnet(fit2,newx = x[-train,], newy=y[-train], s=0.1)
plot(roc.glmnet(fit2, newx = x[-train,], newy=y[-train])[[10]])
fit2c = cv.glmnet(x, y, family = "binomial", keep=TRUE)
idmin = match(fit2c$lambda.min, fit2c$lambda)
plot(roc.glmnet(fit2c$fit.preval, newy = y)[[idmin]])
data(MultinomialExample)
x <- MultinomialExample$x; y <- MultinomialExample$y
set.seed(103)
train = sample(seq(length(y)),100,replace=FALSE)
fit3 = glmnet(x[train,], y[train], family = "multinomial")
confusion.glmnet(fit3, newx = x[-train, ], newy = y[-train], s = 0.01)
fit3c = cv.glmnet(x, y, family = "multinomial", type.measure="class", keep=TRUE)
idmin = match(fit3c$lambda.min, fit3c$lambda)
confusion.glmnet(fit3c$fit.preval, newy = y, family="multinomial")[[idmin]]

glmnet documentation built on Aug. 22, 2023, 9:12 a.m.