MachineShop (version 1.3.0)

summary: Model Performance Summary

Description

Summary statistics for resampled model performance metrics.

Usage

# S3 method for Performance
summary(object, stats = c(Mean = base::mean, Median
  = stats::median, SD = stats::sd, Min = base::min, Max = base::max),
  na.rm = TRUE, ...)
# S3 method for Resamples
summary(object, stats = c(Mean = base::mean, Median =
  stats::median, SD = stats::sd, Min = base::min, Max = base::max),
  na.rm = TRUE, ...)
# S3 method for MLModelTune
summary(object, stats = c(Mean = base::mean, Median
  = stats::median, SD = stats::sd, Min = base::min, Max = base::max),
  na.rm = TRUE, ...)
# S3 method for Confusion
summary(object, ...)
# S3 method for ConfusionMatrix
summary(object, ...)
# S3 method for Curves
summary(object, stat = base::mean, ...)

Arguments

object

object to summarize.

stats

function, one or more function names, or list of named functions to include in the calculation of summary statistics.

na.rm

logical indicating whether to exclude missing values.

...

arguments passed to other methods.

stat

function to compute a summary statistic at each cutoff value of resampled metrics in Curves, or NULL for resample-specific metrics.

Value

array with summmary statistics in the second dimension, metrics in the first for single models, and models and metrics in the first and third, respectively, for multiple models.

See Also

performance, resample, diff, tune, confusion

Examples

# NOT RUN {
## Factor response example

fo <- Species ~ .
control <- CVControl()

gbmres1 <- resample(fo, iris, GBMModel(n.trees = 25), control)
gbmres2 <- resample(fo, iris, GBMModel(n.trees = 50), control)
gbmres3 <- resample(fo, iris, GBMModel(n.trees = 100), control)
summary(gbmres3)

res <- Resamples(GBM1 = gbmres1, GBM2 = gbmres2, GBM3 = gbmres3)
summary(res)

# }