Merge pull request #110 from ModelOriented/CRAN-update

Preparation of CRAN release

Author: mayer79

CRAN-SUBMISSION

@@ -1,3 +1,3 @@
-Version: 0.3.8
-Date: 2023-09-23 12:06:38 UTC
-SHA: 7ca4f7e46e925abd8cf0155227d126ebd9ea2364
+Version: 0.4.0
+Date: 2023-11-10 18:24:23 UTC
+SHA: 4b25bb0b9f2223185eac1fe01a19d6014d01eb12

NAMESPACE

@@ -9,6 +9,7 @@ S3method(permshap,ranger)
 S3method(print,kernelshap)
 S3method(print,permshap)
 S3method(summary,kernelshap)
+S3method(summary,permshap)
 export(is.kernelshap)
 export(is.permshap)
 export(kernelshap)

R/kernelshap.R

@@ -117,7 +117,7 @@
 #'   see README for an example. Parallelization automatically disables the progress bar.
 #' @param parallel_args Named list of arguments passed to [foreach::foreach()]. 
 #'   Ideally, this is `NULL` (default). Only relevant if `parallel = TRUE`. 
-#'   Example on Windows: if `object` is a GAM fitted with package {mgcv}, 
+#'   Example on Windows: if `object` is a GAM fitted with package 'mgcv', 
 #'   then one might need to set `parallel_args = list(.packages = "mgcv")`.
 #' @param verbose Set to `FALSE` to suppress messages and the progress bar.
 #' @param ... Additional arguments passed to `pred_fun(object, X, ...)`.
@@ -165,7 +165,7 @@
 #' s
 #' 
 #' # MODEL TWO: Multi-response linear regression
-#' fit <- lm(as.matrix(iris[1:2]) ~ Petal.Length + Petal.Width + Species, data = iris)
+#' fit <- lm(as.matrix(iris[, 1:2]) ~ Petal.Length + Petal.Width + Species, data = iris)
 #' s <- kernelshap(fit, iris[1:4, 3:5], bg_X = bg_X)
 #' summary(s)
 #' 

R/methods.R

@@ -1,15 +1,13 @@
-#' Print Method
-#' 
-#' Prints the first two rows of the matrix (or matrices) of SHAP values. 
+#' Prints "kernelshap" Object
 #'
 #' @param x An object of class "kernelshap".
 #' @param n Maximum number of rows of SHAP values to print.
 #' @param ... Further arguments passed from other methods.
 #' @returns Invisibly, the input is returned.
 #' @export
 #' @examples
-#' fit <- stats::lm(Sepal.Length ~ ., data = iris)
-#' s <- kernelshap(fit, iris[1:3, -1], bg_X = iris[-1])
+#' fit <- lm(Sepal.Length ~ ., data = iris)
+#' s <- kernelshap(fit, iris[1:3, -1], bg_X = iris[, -1])
 #' s
 #' @seealso [kernelshap()]
 print.kernelshap <- function(x, n = 2L, ...) {
@@ -18,25 +16,32 @@ print.kernelshap <- function(x, n = 2L, ...) {
   invisible(x)
 }
 
-#' @describeIn print.kernelshap Print method for "permshap" object
+#' Prints "permshap" Object
+#' 
+#' @param x An object of class "permshap".
+#' @inheritParams print.kernelshap
+#' @inherit print.kernelshap return
 #' @export
+#' @examples
+#' fit <- lm(Sepal.Length ~ ., data = iris)
+#' s <- permshap(fit, iris[1:3, -1], bg_X = iris[, -1])
+#' s
+#' @seealso [permshap()]
 print.permshap <- function(x, n = 2L, ...) {
   print.kernelshap(x, n = n, ...)
 }
 
-#' Summary Method
+#' Summarizes "kernelshap" Object
 #'
 #' @param object An object of class "kernelshap".
-#' @param compact Set to `TRUE` to hide printing the top n SHAP values, 
-#'   standard errors and feature values. 
-#' @param n Maximum number of rows of SHAP values, standard errors and feature values 
-#'   to print.
+#' @param compact Set to `TRUE` for a more compact summary. 
+#' @param n Maximum number of rows of SHAP values etc. to print.
 #' @param ... Further arguments passed from other methods.
 #' @returns Invisibly, the input is returned.
 #' @export
 #' @examples
-#' fit <- stats::lm(Sepal.Length ~ ., data = iris)
-#' s <- kernelshap(fit, iris[1:3, -1], bg_X = iris[-1])
+#' fit <- lm(Sepal.Length ~ ., data = iris)
+#' s <- kernelshap(fit, iris[1:3, -1], bg_X = iris[, -1])
 #' summary(s)
 #' @seealso [kernelshap()]
 summary.kernelshap <- function(object, compact = FALSE, n = 2L, ...) {
@@ -74,6 +79,21 @@ summary.kernelshap <- function(object, compact = FALSE, n = 2L, ...) {
   invisible(object)
 }
 
+#' Summarizes "permshap" Object
+#'
+#' @param object An object of class "permshap".
+#' @inheritParams summary.kernelshap
+#' @inherit summary.kernelshap return
+#' @export
+#' @examples
+#' fit <- lm(Sepal.Length ~ ., data = iris)
+#' s <- permshap(fit, iris[1:3, -1], bg_X = iris[, -1])
+#' summary(s)
+#' @seealso [permshap()]
+summary.permshap <- function(object, compact = FALSE, n = 2L, ...) {
+  summary.kernelshap(object, compact = compact, n = n, ...)
+}
+
 #' Check for kernelshap
 #'
 #' Is object of class "kernelshap"?
@@ -83,7 +103,7 @@ summary.kernelshap <- function(object, compact = FALSE, n = 2L, ...) {
 #' @export
 #' @examples
 #' fit <- lm(Sepal.Length ~ ., data = iris)
-#' s <- kernelshap(fit, iris[1:2, -1], bg_X = iris[-1])
+#' s <- kernelshap(fit, iris[1:2, -1], bg_X = iris[, -1])
 #' is.kernelshap(s)
 #' is.kernelshap("a")
 #' @seealso [kernelshap()]
@@ -100,7 +120,7 @@ is.kernelshap <- function(object){
 #' @export
 #' @examples
 #' fit <- lm(Sepal.Length ~ ., data = iris)
-#' s <- permshap(fit, iris[1:2, -1], bg_X = iris[-1])
+#' s <- permshap(fit, iris[1:2, -1], bg_X = iris[, -1])
 #' is.permshap(s)
 #' is.permshap("a")
 #' @seealso [kernelshap()]

R/permshap.R

@@ -33,7 +33,7 @@
 #' s
 #'
 #' # MODEL TWO: Multi-response linear regression
-#' fit <- lm(as.matrix(iris[1:2]) ~ Petal.Length + Petal.Width + Species, data = iris)
+#' fit <- lm(as.matrix(iris[, 1:2]) ~ Petal.Length + Petal.Width + Species, data = iris)
 #' s <- permshap(fit, iris[1:4, 3:5], bg_X = bg_X)
 #' s
 #'

README.md

@@ -20,7 +20,7 @@ The package contains two workhorses to calculate SHAP values for any model:
 
 ### Kernel SHAP or permutation SHAP?
 
-- Exact Kernel SHAP and exact permutation SHAP values agree for additive models, and differ for models with interactions. 
+- Exact Kernel SHAP and exact permutation SHAP values (usually) agree for additive models. 
 - If the number of features is sufficiently small, we recommend `permshap()` over `kernelshap()`.
 
 ### Typical workflow to explain any model

cran-comments.md

@@ -1,75 +1,33 @@
-# Resubmission II
+# kernelshap 0.4.0
 
-I have now removed the non-registered DOI from the description file.
+Dear CRAN-team
 
-This was Uwe's comment:
+I have added exact permutation SHAP values, and the possibility to use factor-valued predictions.
 
-   Found the following (possibly) invalid DOIs:
-     DOI: 10.5555/3295222.3295230
-       From: DESCRIPTION
-       Status: 404
-       Message: Not Found
-
-and looking for the DOI shows it is not registered.
-Even when going to
-https://dl.acm.org/doi/10.5555/3295222.3295230
-and clicking on the "Publisher Site" link leads us into nirvana.
-
-So I guess the is ill registered and you have to revert the change. My 
-apologoies, I had not checked whether the DOI given is registreed or not.
-
-Best,
-
-
-# Resubmission I
-
-I stumbled over a DOI: Uwe gently pointed this out:
-
-"
-   The Description field contains
-     <https://dl.acm.org/doi/10.5555/3295222.3295230>, and Covert and Lee
-   Please use permanent DOI markup for linking to publications as in 
-<doi:prefix/suffix>.
-"
-
-This resubmission fixes this.
-
-# Original message
-
-Hello CRAN
-
-This is a small maintenance release only.
-
-Thanks a lot
+Thanks a lot for running CRAN.
 
 Michael
 
 ## Checks
 
 ### Revdep
 
-survex 1.1.3                                                                             
+survex 1.2.0                                                                             
 - OK: 1
 - BROKEN: 0
 
-### Local check with innocent NOTE
+### Local check with usual NOTE
 
-❯ checking HTML version of manual ... NOTE
+checking HTML version of manual ... NOTE
   Skipping checking HTML validation: no command 'tidy' found
 
 ### `check_win_devel()` NOTE
 
-- R Under development (unstable) (2023-09-11 r85126 ucrt)
+- R Under development (unstable) (2023-11-09 r85497 ucrt)
 
 ### `check_rhub()` NOTES
 
--> Note sure where the 403 problem comes from. Is it relevant?
-
-Found the following (possibly) invalid URLs:
-  URL: https://dl.acm.org/doi/10.5555/3295222.3295230
-    From: DESCRIPTION
-    Status: 403
-    Message: Forbidden
 * checking HTML version of manual ... NOTE
 Skipping checking HTML validation: no command 'tidy' found
 Skipping checking math rendering: package 'V8' unavailable
+