Merge pull request #36 from adamlilith/solstice_2022_2023

Solstice 2022 2023

DESCRIPTION

@@ -2,7 +2,7 @@ Package: enmSdmX
 Type: Package
 Title: Species Distribution Modeling and Ecological Niche Modeling
 Version: 1.1.3
-Date: 2023-09-15
+Date: 2023-03-06
 Authors@R: 
 	c(
 		person(
@@ -13,7 +13,6 @@ Authors@R:
 			comment = c(ORCID = '0000-0002-6420-1659')
 		)
 	)
-Maintainer: Adam B. Smith <[email protected]>
 Description: Implements species distribution modeling and ecological niche
 	modeling, including: bias correction, spatial cross-validation, model
 	evaluation, raster interpolation, biotic "velocity" (speed and
@@ -32,7 +31,6 @@ Depends:
 Imports:
 	boot,
 	data.table,
-	dismo,
 	doParallel,
 	DT,
 	foreach,
@@ -45,11 +43,13 @@ Imports:
 	MuMIn,
 	omnibus,
 	parallel,
+	predicts,
 	ranger,
 	rJava,
 	scales,
 	sf,
 	shiny,
+	sp,
 	statisfactory,
 	stats,
 	terra,
@@ -62,4 +62,4 @@ URL: https://github.com/adamlilith/enmSdmX
 BugReports: https://github.com/adamlilith/enmSdmX
 Encoding: UTF-8
 License: MIT + file LICENSE
-RoxygenNote: 7.2.3
+RoxygenNote: 7.3.0

NAMESPACE

@@ -49,3 +49,4 @@ export(trainNS)
 export(trainRF)
 export(weightByDist)
 import(data.table)
+importFrom(stats,predict)

NEWS.md

@@ -1,5 +1,8 @@
-# enmSdmX 1.1.3 2023-09-15
-- Added European Datrim 1950 (ED50) to coordinate reference systems table (mea cupla!)
+# enmSdmX 1.1.3 2023-02-02
+- Removed dependency on `dismo`, replaced where possible by `predicts`; copied `gbm.step()` and `predict()` method for MaxEnt to `enmSdmX` as a momentary fix; would love a professional solution!
+- Added European Datum 1950 (ED50) to coordinate reference systems table
+- Added ability to export intermediate values for plotting in `evalContBoyce()`
+- Fixed spelling errors in `crss` data table (coordinate reference system names)
 
 # enmSdmX 1.1.2 2023-09-07
 - Minor issue fixes with undocumented arguments for non-exported functions

R/01_imports.r

@@ -0,0 +1 @@
+#' @import predicts::MaxEnt_model

R/aaa.r

@@ -6,8 +6,8 @@
 
 	ver <- read.dcf(file=system.file('DESCRIPTION', package=pkg), fields='Version')
 	packageStartupMessage(paste0('This is ', pkg, ' ', ver, '.'))
-	packageStartupMessage(paste0('* Back-incompatible changes starting starting with enmSdmX version 1.1.1:'))
+	packageStartupMessage(paste0('Back-incompatible changes starting starting with enmSdmX version 1.1.1:'))
 	packageStartupMessage(paste0('* trainRF() replaces use of the randomForest package with the faster ranger package, which produces random forests that are statistically equivalent.'))
-	packageStartupMessage(paste0('geoFold() uses ', dQuote('complete'), ' clustering by default (was ', dQuote('single'), ').'))
+	packageStartupMessage(paste0('* geoFold() uses ', dQuote('complete'), ' clustering by default (was ', dQuote('single'), ').'))
 
 }

R/evalAUC.r

@@ -9,7 +9,7 @@
 #' @param ... Other arguments (unused).
 #' @return A Numeric value.
 #'
-#' @seealso \code{\link[dismo]{evaluate}}, \code{\link{evalMultiAUC}}
+#' @seealso \code{\link[predicts]{pa_evaluate}}, \code{\link{evalMultiAUC}}
 #' 
 #' @examples
 #' pres <- seq(0.5, 1, by=0.1)

R/evalContBoyce.r

@@ -12,17 +12,18 @@
 #' @param method Character. Type of correlation to calculate. The default is \code{'spearman'}, the Spearman rank correlation coefficient used by Boyce et al. (2002) and Hirzel et al. (2006), which is the "traditional" CBI. In contrast, \code{'pearson'} or \code{'kendall'} can be used instead.  See \code{\link[stats]{cor}} for more details.
 #' @param dropZeros Logical. If \code{TRUE} then drop all bins in which the frequency of presences is 0.
 #' @param graph Logical. If \code{TRUE} then plot P vs E and P/E versus bin.
+#' @param table Logical. If \code{TRUE}, the function will also return a table of P (proportion of presence weights per bin), E (expected proportion of presence weights per bin--from contrast sites), and the ratio of the two.
 #' @param na.rm Logical. If \code{TRUE} then remove any presences and associated weights and background predictions and associated weights with \code{NA}s.
 #' @param ... Other arguments (not used).
 #'
-#' @return Numeric value.
+#' @return Numeric value, or if \code{table} is \code{TRUE}, then a list object with CBI plus a data frame with P (proportion of presence weights per bin), E (expected proportion of presence weights per bin--from contrast sites), and the ratio of the two.
 #' 
 #' @details CBI is the Spearman rank correlation coefficient between the proportion of sites in each prediction class and the expected proportion of predictions in each prediction class based on the proportion of the landscape that is in that class.  The index ranges from -1 to 1. Values >0 indicate the model's output is positively correlated with the true probability of presence.  Values <0 indicate it is negatively correlated with the true probability of presence.
 #' 
 #' @references Boyce, M.S., Vernier, P.R., Nielsen, S.E., and Schmiegelow, F.K.A.  2002.  Evaluating resource selection functions.  \emph{Ecological Modeling} 157:281-300. \doi{https://doi.org/10.1016/S0304-3800(02)00200-4}
 #' @references Hirzel, A.H., Le Lay, G., Helfer, V., Randon, C., and Guisan, A.  2006.  Evaluating the ability of habitat suitability models to predict species presences.  \emph{Ecological Modeling} 199:142-152. \doi{10.1016/j.ecolmodel.2006.05.017}
 #'
-#' @seealso \code{\link[stats]{cor}}, \code{\link[dismo]{evaluate}}, \code{\link{evalAUC}}, \code{\link{evalMultiAUC}}, \code{\link{evalContBoyce}}, \code{\link{evalThreshold}}, \code{\link{evalThresholdStats}}, \code{\link{evalTjursR2}}, \code{\link{evalTSS}}
+#' @seealso \code{\link[stats]{cor}}, \code{\link[predicts]{pa_evaluate}}, \code{\link{evalAUC}}, \code{\link{evalMultiAUC}}, \code{\link{evalContBoyce}}, \code{\link{evalThreshold}}, \code{\link{evalThresholdStats}}, \code{\link{evalTjursR2}}, \code{\link{evalTSS}}
 #'
 #' @examples
 #'
@@ -47,6 +48,7 @@ evalContBoyce <- function(
 	method = 'spearman',
 	dropZeros = TRUE,
 	graph = FALSE,
+	table = FALSE,
 	na.rm = FALSE,
 	...
 ) {
@@ -154,6 +156,16 @@ evalContBoyce <- function(
 
 	# calculate continuous Boyce index (cbi)
 	cbi <- stats::cor(x=meanPred, y=PE, method=method)
-	cbi
+
+	if (table) {
+		out <- list(
+			table = data.frame(bin = 1:numBins, propPres = P, propExpected = E, PtoERatio = PE),
+			cbi = cbi
+		)
+	} else {
+		out <- cbi
+	}
+
+	out
 
 }

R/evalMultiAUC.r

@@ -8,7 +8,7 @@
 #'
 #' @return Named numeric vector. The names will appear as \code{case2_over_case1} (which in this example means the AUC of item #1 in the \code{...} when compared to the second item in \code{...}), plus \code{multivariate} (which is the multivariate AUC).
 #' 
-#' @seealso \code{\link[dismo]{evaluate}}, \code{\link{evalAUC}}, \code{\link{evalContBoyce}}, \code{\link{evalThreshold}}, \code{\link{evalThresholdStats}}, \code{\link{evalTjursR2}}, \code{\link{evalTSS}}
+#' @seealso \code{\link[predicts]{pa_evaluate}}, \code{\link{evalAUC}}, \code{\link{evalContBoyce}}, \code{\link{evalThreshold}}, \code{\link{evalThresholdStats}}, \code{\link{evalTjursR2}}, \code{\link{evalTSS}}
 #'
 #' @references
 #' Hand, DJ and Till, RJ. 2001. A simple generalisation of the area under the ROC curve for multiple class classification problems. \emph{Machine Learning} 45:171-186 \doi{10.1023/A:1010920819831}.

R/evalTSS.r

@@ -12,7 +12,7 @@
 #' @details This function calculates the maximum value of the True Skill Statistic (i.e., across all thresholds, the values that maximizes sensitivity plus specificity).
 #' @references  See Allouche, O., Tsoar, A., and Kadmon, R. 2006. Assessing the accuracy of species distribution models: Prevalence, kappa and the true skill statistic (TSS). \emph{Journal of Applied Ecology} 43:1223-1232. \doi{10.1111/j.1365-2664.2006.01214.x}
 #'
-#' @seealso \code{\link[dismo]{evaluate}}, \code{\link{evalAUC}}, \code{\link{evalMultiAUC}}, \code{\link{evalContBoyce}}, \code{\link{evalThreshold}}, \code{\link{evalThresholdStats}}, \code{\link{evalTjursR2}}
+#' @seealso \code{\link[predicts]{pa_evaluate}}, \code{\link{evalAUC}}, \code{\link{evalMultiAUC}}, \code{\link{evalContBoyce}}, \code{\link{evalThreshold}}, \code{\link{evalThresholdStats}}, \code{\link{evalTjursR2}}
 #'
 #' @examples
 #' set.seed(123)
@@ -34,14 +34,6 @@
 #' presWeight <- c(rep(0.1, 30), rep(1, 30))
 #' evalTSS(pres, contrast, presWeight=presWeight)
 #' 
-#' e <- dismo::evaluate(pres, contrast)
-#' max(e@TPR + e@TNR) - 1
-#' 
-#' # Why different values from dismo's evaluate() function?
-#' # Because dismo's function uses thresholds based on presence/non-presence
-#' # values, whereas evalTSS uses equall-spaced thresholds.
-#' head(e@t)
-#'
 #' @export
 evalTSS <- function(
 	pres,

R/evalThreshold.r

@@ -1,6 +1,6 @@
 #' Weighted thresholds for predictions
 #'
-#' This function is similar to the \code{\link[dismo]{threshold}} function in the \pkg{dismo} package, which calculates thresholds to create binary predictions from continuous values. However, unlike that function, it allows the user to specify weights for presences and absence/background predictions. The output will thus be the threshold that best matches the specified criterion taking into account the relative weights of the input values.
+#' This function is similar to the \code{\link[predicts]{threshold}} function in the \pkg{predicts} package, which calculates thresholds to create binary predictions from continuous values. However, unlike that function, it allows the user to specify weights for presences and absence/background predictions. The output will thus be the threshold that best matches the specified criterion taking into account the relative weights of the input values.
 #' @param pres Numeric vector. Predicted values at test presences.
 #' @param contrast Numeric vector. Predicted values at background/absence sites.
 #' @param presWeight Numeric vector same length as \code{pres}. Relative weights of presence sites. The default is to assign each presence a weight of 1.
@@ -20,7 +20,7 @@
 #' @return Named numeric vector.
 #' Fielding, A.H. and J.F. Bell. 1997. A review of methods for the assessment of prediction errors in conservation presence/absence models. \emph{Environmental Conservation} 24:38-49. \doi{10.1017/S0376892997000088}
 #'
-#' @seealso \code{\link[dismo]{threshold}}, \code{\link[dismo]{evaluate}}, \code{\link{evalAUC}}, \code{\link{evalMultiAUC}}, \code{\link{evalContBoyce}}, \code{\link{evalThresholdStats}}, \code{\link{evalTjursR2}}, \code{\link{evalTSS}}
+#' @seealso \code{\link[predicts]{threshold}}, \code{\link[predicts]{pa_evaluate}}, \code{\link{evalAUC}}, \code{\link{evalMultiAUC}}, \code{\link{evalContBoyce}}, \code{\link{evalThresholdStats}}, \code{\link{evalTjursR2}}, \code{\link{evalTSS}}
 #'
 #' @examples
 #' set.seed(123)

R/evalThresholdStats.r

@@ -23,7 +23,7 @@
 #' }
 #' Fielding, A.H. and J.F. Bell. 1997. A review of methods for the assessment of prediction errors in conservation presence/absence models. \emph{Environmental Conservation} 24:38-49. \doi{10.1017/S0376892997000088}
 #'
-#' @seealso \code{\link[dismo]{threshold}}, \code{\link[dismo]{evaluate}}, \code{\link{evalAUC}}, \code{\link{evalMultiAUC}}, \code{\link{evalContBoyce}}, \code{\link{evalThreshold}}, \code{\link{evalTjursR2}}, \code{\link{evalTSS}}
+#' @seealso \code{\link[predicts]{threshold}}, \code{\link[predicts]{pa_evaluate}}, \code{\link{evalAUC}}, \code{\link{evalMultiAUC}}, \code{\link{evalContBoyce}}, \code{\link{evalThreshold}}, \code{\link{evalTjursR2}}, \code{\link{evalTSS}}
 #'
 #' @examples
 #' set.seed(123)

R/evalTjursR2.r

@@ -12,7 +12,7 @@
 #'
 #' @references Tjur, T. 2009. Coefficients of determination in logistic regression models-A new proposal: The coefficient of discrimination. \emph{The American Statistician} 63:366-372. \doi{10.1198/tast.2009.08210}
 #'
-#' @seealso \code{\link[dismo]{evaluate}}, \code{\link{evalAUC}}, \code{\link{evalMultiAUC}}, \code{\link{evalContBoyce}}, \code{\link{evalThreshold}}, \code{\link{evalThresholdStats}}, \code{\link{evalTSS}}
+#' @seealso \code{\link[predicts]{pa_evaluate}}, \code{\link{evalAUC}}, \code{\link{evalMultiAUC}}, \code{\link{evalContBoyce}}, \code{\link{evalThreshold}}, \code{\link{evalThresholdStats}}, \code{\link{evalTSS}}
 #'
 #' @examples
 #' pres <- seq(0.5, 1, by=0.1)