Minor patch

• Fixed an error in compute_extrapolation (and by extension compute_nearby) when the number of analogue cells is less than 2 (i.e., near 100% extrapolation). Many thanks to James Smith for flagging the issue.

Minor patch

Corrected compute_extrapolation so that the resolution argument can be actioned even when the supplied prediction.grid is regular.

Minor updates

• Added new extrapolation.object argument to compare_covariates and map_extrapolation, which allows required inputs to be retrieved directly from the list object returned by compute_extrapolation rather than being manually specified for each function. This was done to streamline function use, minimise redundancy, and increase alignment with the package workflow described in Bouchet et al. (2020). DOI: 10.1111/2041-210X.13469.
• Implemented function ellipsis (...) in compare_covariates so that users who wish to use the function without running compute_extrapolation first can do so.
• Added option to suppress warnings and messages via the verbose argument.
• Update package vignettes to reflect the above changes.

Minor release

• Removed unstable package dependencies.
• Updated package documentation for internal functions.
• Added two new tutorial vignettes: (1) demonstrating the use of compute_nearby on a large dataset, and (2) showcasing the use of dsmextra for a priori covariate selection.
• Updated package website

Minor release

• Updated documentation, including the renaming of the segments argument to samples.
• Added a new dataset (acacia), which supports an additional example of the use of compute_extrapolation.
• Updated test files and vignette to reflect the above changes.

Minor release v1.1.0

• Implemented a summary method that replaces summarise_extrapolation (now internalised) [courtesy of @dill].
• Update package vignette.
• Changed ggplot theme to theme_minimal in outputs of compare_covariates.
• Fixed dependency issues arising from the removal of WhatIf from CRAN (#6 (comment))
• Fixed rounding issues in summary.
• Fixed issue with progress bar when running compute_nearby on large datasets.
• Updated package index and documentation.

Patch v1.0.2

(1) Fixed errors arising when segments and/or prediction grids contain NA values.
(2) Minor update to compare_covariates() to increase efficiency when input prediction grid is irregular and has to be rasterised.
Thank you to Andrew DiMatteo for flagging these issues.

Patch v1.0.1

(1) Fixed spelling errors in compare_covariates.R
(2) Updated summarise_extrapolation.R to avoid errors associated with variable outputs from apply().
These updates address issue #3. Many thanks to David Fifield for the thorough bug report (and accompanying solutions!).

dsmextra v1.0.0

To install this release of the dsmextra package, run the code below:

if (!require("remotes")) {
  install.packages("remotes")
}
remotes::install_github("densitymodelling/dsmextra")