HibCV — HibCV • hibiiscus

This is the template object for a an R6 class containing the results of cross-validation. It contains functions for summarising and visualising the results.

Details

Template object for cross-validation results

Public fields

last_update: (character)
Identifies the most recent update to the object (YYYY-MM-DD).
model_type: (character)
Identifies the type of model; one of "HibModelLME4", "HibModelINLA", "HibModelBRMS".
original_model_obj: (list)
Contains a list of core model information
cv_results: (list)
Contains the list of results from cross-validation
insample_results: (list)
Contains the list of results from in-sample validation
re_form: (list)
A list of the re_form model structures used in evaluation.
use_original_mesh: (logical or NULL)
Indicates use of original mesh for spatial INLA models.

Methods

`HibCV$new()`

Creates a new instance of this R6 class. Note, this object is constructed via the hib_cv_model function found within this package.

Usage

HibCV$new(mod_obj, cv_res, insamp_res, re_form, use_orig_mesh)

Arguments

mod_obj: (HibModelXXXX)
The model object used in cross-validation.
cv_res: (list)
The results of cross-validation.
insamp_res: (list)
The results of ins-sample validation.
re_form: (list)
The list of random effects structures used in model evaluation.
use_orig_mesh: (logical or NULL)
Indicates use of original mesh for spatial INLA models

Returns

(HibCV)

`HibCV$print()`

Base printout function presents core information about the object.

Usage

HibCV$print()

Returns

(NULL).

`HibCV$plot()`

Function to generate figures summarising the cross- validation results.

Usage

HibCV$plot(type = "metrics", metrics = c("rmse", "rsq"))

Arguments

type: (character()) Options are: "coefficients", "coefficients-summary", "obs-v-pred", "obs-v-error", "pred-v-error", "metrics", "map-sites".
metrics: (character) Options are the lower-case abbreviations of calculated error metrics. - Only needed if type == "metrics".

Returns

(ggplot).

`HibCV$clone()`

The objects of this class are cloneable with this method.

Usage

HibCV$clone(deep = FALSE)

Arguments

deep: Whether to make a deep clone.