h2o.kolmogorov_smirnov function

Kolmogorov-Smirnov metric for binomial models

Kolmogorov-Smirnov metric for binomial models

Retrieves a Kolmogorov-Smirnov metric for given binomial model. The number returned is in range between 0 and 1. K-S metric represents the degree of separation between the positive (1) and negative (0) cumulative distribution functions. Detailed metrics per each group are to be found in the gains-lift table.

h2o.kolmogorov_smirnov(object) ## S4 method for signature 'H2OModelMetrics' h2o.kolmogorov_smirnov(object) ## S4 method for signature 'H2OModel' h2o.kolmogorov_smirnov(object)

Arguments

  • object: Either an H2OModel object or an H2OModelMetrics object.

Returns

Kolmogorov-Smirnov metric, a number between 0 and 1.

Details

The H2OModelMetrics version of this function will only take H2OBinomialMetrics objects.

Examples

## Not run: library(h2o) h2o.init() data <- h2o.importFile( path = "https://s3.amazonaws.com/h2o-public-test-data/smalldata/airlines/allyears2k_headers.zip") model <- h2o.gbm(x = c("Origin", "Distance"), y = "IsDepDelayed", training_frame = data, ntrees = 1) h2o.kolmogorov_smirnov(model) ## End(Not run)

See Also

h2o.gainsLift to see detailed K-S metrics per group

  • Maintainer: Tomas Fryda
  • License: Apache License (== 2.0)
  • Last published: 2024-01-11