quick_analyse: Quickly analyse a data frame using h2o.ai

In lewinfox/quickanalyser: Quickly Analyse Data Using h2o.ai

Description

Uses h2o.ai to quickly run a random forest, gradient boosting machine and deep learning model on an input data frame, and returns information on the performance of the models.

Usage

quick_analyse(data, response_variable = "y", analysis_mode = c("light",
  "full"))

Arguments

data	input data frame to be analysed
response_variable	name (string) of column in the data frame that contains the response / dependent variable. If missing, defaults to "y".
analysis_mode	one of either "light" (default) or "full". This determines the options that will be passed to the internal models. Light mode aims to reduce run time at the expense of accuracy, full mode is more heavyweight. Your mileage may vary.

Value

a list, each item of which is another list containing the h2o model objects, and the h2o.performance objects extracted from those objects, respectively.

Examples

data <- datasets::iris
quick_result <- quick_analyse(data = data, response_variable = "Species")
full_result <- quick_analyse(data = data, response_variable = "Species", analysis_mode = "full")

lewinfox/quickanalyser documentation built on May 13, 2019, 1:37 p.m.