get_best_model_results: Extract best model performances

In boyanangelov/sdmbench: Benchmark Species Distribution Models

Description

Extract best model performances

Usage

get_best_model_results(bmr)

Arguments

bmr	A bmr object, result of the benchmark_sdm function.

Value

A dataframe containing the best model results (learner name, optimal iteration number and associated AUC).

Examples

## Not run: 
# download benchmarking data
benchmarking_data <- get_benchmarking_data("Lynx lynx",
                                           limit = 1500)

# create a list of algorithms to compare
learners <- list(mlr::makeLearner("classif.randomForest",
                                  predict.type = "prob"),
                 mlr::makeLearner("classif.logreg",
                                  predict.type = "prob"))

# run the model benchmarking process
bmr <- benchmark_sdm(benchmarking_data$df_data,
                     learners = learners,
                     dataset_type = "default",
                     sample = FALSE)

# get best model results
# you should obtain a dataframe containing the highest performing (by AUC)
# algorithm name, iteration and associated AUC
best_results <- get_best_model_results(bmr)
best_results

## End(Not run)

boyanangelov/sdmbench documentation built on Dec. 14, 2020, 1:08 a.m.