h2o (version 3.44.0.3)

h2o.multinomial_aucpr_table: Retrieve the all PR AUC values in a table (One to Rest, One to One, macro and weighted average) for mutlinomial classification.

Description

Retrieves the PR AUC table from an H2OMultinomialMetrics. If "train", "valid", and "xval" parameters are FALSE (default), then the training PR AUC table is returned. If more than one parameter is set to TRUE, then a named vector of PR AUC tables are returned, where the names are "train", "valid" or "xval".

Usage

h2o.multinomial_aucpr_table(object, train = FALSE, valid = FALSE, xval = FALSE)

Arguments

object

An H2OMultinomialMetrics object.

train

Retrieve the training PR AUC table

valid

Retrieve the validation PR AUC table

xval

Retrieve the cross-validation PR AUC table

See Also

h2o.giniCoef for the Gini coefficient, h2o.mse for MSE, and h2o.metric for the various threshold metrics. See h2o.performance for creating H2OModelMetrics objects.

Examples

if (FALSE) {
library(h2o)
h2o.init()

prostate_path <- system.file("extdata", "prostate.csv", package = "h2o")
prostate <- h2o.uploadFile(prostate_path)

prostate[, 2] <- as.factor(prostate[, 2])
model <- h2o.gbm(x = 3:9, y = 2, training_frame = prostate, distribution = "bernoulli")
perf <- h2o.performance(model, prostate)
h2o.multinomial_aucpr_table(perf)
}