h2o.make_leaderboard: Create a leaderboard from a list of models, grids and/or...

In h2o: R Interface for the 'H2O' Scalable Machine Learning Platform

Create a leaderboard from a list of models, grids and/or automls.

Description

Create a leaderboard from a list of models, grids and/or automls.

Usage

h2o.make_leaderboard(
  object,
  leaderboard_frame,
  sort_metric = "AUTO",
  extra_columns = c(),
  scoring_data = c("AUTO", "train", "valid", "xval")
)

Arguments

object	List of models, automls, or grids; or just single automl/grid object.
leaderboard_frame	Frame used for generating the metrics (optional).
sort_metric	Metric used for sorting the leaderboard.
extra_columns	What extra columns should be calculated (might require leaderboard_frame). Use "ALL" for all available or list of extra columns.
scoring_data	Metrics to be reported in the leaderboard ("xval", "train", or "valid"). Used if no leaderboard_frame is provided.

Value

data.frame

Examples

## Not run: 
library(h2o)
h2o.init()
iris_hf <- as.h2o(iris)
grid <- h2o.grid("gbm", x = c(1:4), y = 5, training_frame = iris_hf,
                 hyper_params = list(ntrees = c(1, 2, 3)))
h2o.make_leaderboard(grid, iris_hf)

## End(Not run)

h2o documentation built on May 29, 2024, 4:26 a.m.