Skip to contents

Output summary statistics for a MaxEnt model ('SDMmodelCV' object)

Source: R/compute_MaxEnt_summary_statistics_CV.R
compute_MaxEnt_summary_statistics_CV.Rd

This function will create a directory for and save a MaxEnt model that was run using the SDMtune R package. It will use the model to calculate a list of summary statistics based on the test data and covariates given. See 'return' for a list of summary statistics outputs.

Usage

compute_MaxEnt_summary_statistics_CV(
  model.obj,
  model.name = "MODEL",
  mypath,
  create.dir = FALSE,
  env.covar.obj,
  train.obj,
  trainFolds.obj,
  test.obj,
  plot.fun = "mean",
  plot.type = "cloglog",
  jk.test.type = "test"
)

Arguments

model.obj

A model object created by the package 'SDMtune', should be of class 'SDMmodelCV'.

model.name

Character. A string matching the name of the object set for model.obj. Exclude unnecessary phrases, such as the "_model" ending.

mypath

Character.A file path to the sub directory where the model output will be stored. Should be used with the file.path() function (i.e. with '/' instead of '\'). If this sub directory does not already exist and should be created by the function, set create.dir = TRUE. This will create a folder from the last part of the filepath in mypath.

create.dir

Logical. Should the last element of mypath create a sub directory for the model output? If TRUE, the main folder will be created for the model output. If FALSE (ie, the sub directory already exists), only the "plots" folder within the model output sub directory will be created.

env.covar.obj

A stack of rasters of environmental covariates. These covariates are used to train and test the MaxEnt model, as well as to make predictions. These should be the same covariates that you used to train the model. This must a SpatRaster object created using terra::rast().

train.obj

The main group of presence and background points used to train the model. Should be a SWD object, created using the SDMtune::prepareSWD() function.

trainFolds.obj

A list of two matrices that specify the fold of the training and testing points. This object is used to create k-fold partitions from presence and background datasets to train a SWDmodelCV object. This is created using the SDMtune::randomFolds() function.

test.obj

A withheld group of presence and background points used to test the model after training. Should be a SWD object, created using the SDMtune::prepareSWD() function.

plot.fun

Character. Default is "mean". The function used to determine the level of the other variables when creating the marginal curve. Can be one of: min, mean, median, max, or sd (standard deviation).

plot.type

Character. Default is "cloglog".The type of output desired for the marginal and univariate response curves. Choices are "logistic" and "cloglog".If both are used, must be concatenated in the form: c("logistic", "cloglog").

jk.test.type

Character. Default is "test". When a jackknife test is conducted, this specifies whether to conduct the test on the training or the test dataset. Choices are one of both of "train" and "test". If both are used, must be concatenated in the form: c("train", "test").

Value

The output of this function includes the following:

  • training and test datasets used for the model

  • listed model parameters and suitability thresholds

  • K-folds and which samples were included per fold

  • variable contributions, permutation importance and SD

  • confusion matrices per iteration

  • jackknife tests for both training and testing data, per iteration

  • jackknife plots

  • AUC / TSS

  • ROC plots

  • marginal and univariate response curves

Details

The function requires the packages 'tidyverse', 'here', 'devtools', 'rJava', 'dismo', 'SDMtune', 'viridis', 'plotROC', 'cli', and 'terra'.

Examples