diff --git a/CRAN-SUBMISSION b/CRAN-SUBMISSION
index 66ec04c..ab6e4fc 100644
--- a/CRAN-SUBMISSION
+++ b/CRAN-SUBMISSION
@@ -1,3 +1,3 @@
-Version: 0.8.0
-Date: 2023-05-09 17:34:27 UTC
-SHA: e2f9b361e5cb79eb0faa79faabceb98a4bd1726a
+Version: 0.9.0
+Date: 2023-06-09 14:22:29 UTC
+SHA: 3f6bd1f781851c169a080adb317142b133f58147
diff --git a/R/methods.R b/R/methods.R
index f32f19a..8b778e6 100644
--- a/R/methods.R
+++ b/R/methods.R
@@ -271,11 +271,13 @@ c.shapviz <- function(...) {
 #' @param ... Arguments passed to `split()`.
 #' @returns A "mshapviz" object.
 #' @examples
+#' \dontrun{
 #' dtrain <- xgboost::xgb.DMatrix(data.matrix(iris[, -1]), label = iris[, 1])
-#' fit <- xgboost::xgb.train(data = dtrain, nrounds = 50, nthread = 1)
+#' fit <- xgboost::xgb.train(data = dtrain, nrounds = 10, nthread = 1)
 #' sv <- shapviz(fit, X_pred = dtrain, X = iris)
 #' mx <- split(sv, f = iris$Species)
 #' sv_dependence(mx, "Petal.Length")
+#' }
 #' @export
 #' @seealso [shapviz()], [rbind.shapviz()]
 split.shapviz <- function(x, f, ...) {
diff --git a/R/shapviz.R b/R/shapviz.R
index 6ec0aab..e1f61a2 100644
--- a/R/shapviz.R
+++ b/R/shapviz.R
@@ -63,7 +63,7 @@
 #'     scale of the SHAP values.
 #'   - `S_inter`: Numeric array of SHAP interaction values (or `NULL`).
 #' @seealso
-#'   [sv_importance()], [sv_dependence()], [sv_interaction()],
+#'   [sv_importance()], [sv_dependence()], [sv_dependence2D()], [sv_interaction()],
 #'   [sv_waterfall()], [sv_force()], [collapse_shap()]
 #' @examples
 #' S <- matrix(c(1, -1, -1, 1), ncol = 2, dimnames = list(NULL, c("x", "y")))
@@ -110,11 +110,11 @@ shapviz.matrix = function(object, X, baseline = 0, collapse = NULL,
 #'   Creates a "shapviz" object from an XGBoost model.
 #' @export
 #' @examples
-#'
+#' \dontrun{
 #' # XGBoost models
-#' X_pred <- data.matrix(iris[, -1L])
-#' dtrain <- xgboost::xgb.DMatrix(X_pred, label = iris[, 1L])
-#' fit <- xgboost::xgb.train(data = dtrain, nrounds = 50L, nthread = 1L)
+#' X_pred <- data.matrix(iris[, -1])
+#' dtrain <- xgboost::xgb.DMatrix(X_pred, label = iris[, 1])
+#' fit <- xgboost::xgb.train(data = dtrain, nrounds = 10, nthread = 1)
 #'
 #' # Will use numeric matrix "X_pred" as feature matrix
 #' x <- shapviz(fit, X_pred = X_pred)
@@ -129,15 +129,13 @@ shapviz.matrix = function(object, X, baseline = 0, collapse = NULL,
 #' x <- shapviz(fit, X_pred = dtrain, X = iris)
 #'
 #' # Multiclass setting
-#' params <- list(objective = "multi:softprob", num_class = 3L)
-#' X_pred <- data.matrix(iris[, -5L])
-#' dtrain <- xgboost::xgb.DMatrix(X_pred, label = as.integer(iris[, 5L]) - 1L)
-#' fit <- xgboost::xgb.train(
-#'   params = params, data = dtrain, nrounds = 50L, nthread = 1L
-#' )
+#' params <- list(objective = "multi:softprob", num_class = 3, nthread = 1)
+#' X_pred <- data.matrix(iris[, -5])
+#' dtrain <- xgboost::xgb.DMatrix(X_pred, label = as.integer(iris[, 5]) - 1)
+#' fit <- xgboost::xgb.train(params = params, data = dtrain, nrounds = 10)
 #'
 #' # Select specific class
-#' x <- shapviz(fit, X_pred = X_pred, which_class = 3L)
+#' x <- shapviz(fit, X_pred = X_pred, which_class = 3)
 #' x
 #'
 #' # Or combine all classes to "mshapviz" object
@@ -145,9 +143,9 @@ shapviz.matrix = function(object, X, baseline = 0, collapse = NULL,
 #' x
 #'
 #' # What if we would have one-hot-encoded values and want to explain the original column?
-#' X_pred <- stats::model.matrix(~ . -1, iris[, -1L])
-#' dtrain <- xgboost::xgb.DMatrix(X_pred, label = as.integer(iris[, 1L]))
-#' fit <- xgboost::xgb.train(data = dtrain, nrounds = 50L)
+#' X_pred <- stats::model.matrix(~ . -1, iris[, -1])
+#' dtrain <- xgboost::xgb.DMatrix(X_pred, label = as.integer(iris[, 1]))
+#' fit <- xgboost::xgb.train(data = dtrain, nrounds = 10, nthread = 1)
 #' x <- shapviz(
 #'   fit,
 #'   X_pred = X_pred,
@@ -159,29 +157,30 @@ shapviz.matrix = function(object, X, baseline = 0, collapse = NULL,
 #' # Similarly with LightGBM
 #' if (requireNamespace("lightgbm", quietly = TRUE)) {
 #'   fit <- lightgbm::lgb.train(
-#'     params = list(objective = "regression", num_thread = 1L),
-#'     data = lightgbm::lgb.Dataset(X_pred, label = iris[, 1L]),
-#'     nrounds = 50L,
-#'     verbose = -2L
+#'     params = list(objective = "regression", num_thread = 1),
+#'     data = lightgbm::lgb.Dataset(X_pred, label = iris[, 1]),
+#'     nrounds = 10,
+#'     verbose = -2
 #'   )
 #'
 #'   x <- shapviz(fit, X_pred = X_pred)
 #'   x
 #'
 #'   # Multiclass
-#'   params <- list(objective = "multiclass", num_class = 3L, num_thread = 1L)
-#'   X_pred <- data.matrix(iris[, -5L])
-#'   dtrain <- lightgbm::lgb.Dataset(X_pred, label = as.integer(iris[, 5L]) - 1L)
-#'   fit <- lightgbm::lgb.train(params = params, data = dtrain, nrounds = 50L)
+#'   params <- list(objective = "multiclass", num_class = 3, num_thread = 1)
+#'   X_pred <- data.matrix(iris[, -5])
+#'   dtrain <- lightgbm::lgb.Dataset(X_pred, label = as.integer(iris[, 5]) - 1)
+#'   fit <- lightgbm::lgb.train(params = params, data = dtrain, nrounds = 10)
 #'
 #'   # Select specific class
-#'   x <- shapviz(fit, X_pred = X_pred, which_class = 3L)
+#'   x <- shapviz(fit, X_pred = X_pred, which_class = 3)
 #'   x
 #'
 #'   # Or combine all classes to a "mshapviz" object
 #'   mx <- shapviz(fit, X_pred = X_pred)
 #'   mx
-#'   all.equal(mx[[3L]], x)
+#'   all.equal(mx[[3]], x)
+#' }
 #' }
 shapviz.xgb.Booster = function(object, X_pred, X = X_pred, which_class = NULL,
                                collapse = NULL, interactions = FALSE, ...) {
diff --git a/R/sv_dependence.R b/R/sv_dependence.R
index fb15f82..20e7c2d 100644
--- a/R/sv_dependence.R
+++ b/R/sv_dependence.R
@@ -34,8 +34,9 @@
 #' @param ... Arguments passed to [ggplot2::geom_jitter()].
 #' @returns An object of class "ggplot" (or "patchwork") representing a dependence plot.
 #' @examples
+#' \dontrun{
 #' dtrain <- xgboost::xgb.DMatrix(data.matrix(iris[, -1]), label = iris[, 1])
-#' fit <- xgboost::xgb.train(data = dtrain, nrounds = 50, nthread = 1)
+#' fit <- xgboost::xgb.train(data = dtrain, nrounds = 10, nthread = 1)
 #' x <- shapviz(fit, X_pred = dtrain, X = iris)
 #' sv_dependence(x, "Petal.Length")
 #' sv_dependence(x, "Petal.Length", color_var = "Species")
@@ -43,17 +44,13 @@
 #' sv_dependence(x, c("Species", "Petal.Length"))
 #' sv_dependence(x, "Petal.Width", color_var = c("Species", "Petal.Length"))
 #'
-#' # SHAP interaction values
+#' # SHAP interaction values/main effects
 #' x2 <- shapviz(fit, X_pred = dtrain, X = iris, interactions = TRUE)
 #' sv_dependence(x2, "Petal.Length", interactions = TRUE)
-#' sv_dependence(x2, c("Petal.Length", "Species"), color_var = NULL, interactions = TRUE)
-#'
-#' # Show main effect of "Petal.Length" for setosa and virginica separately
-#' mx <- c(
-#'   setosa = x2[x2$X$Species == "setosa"],
-#'   virginica = x2[x2$X$Species == "virginica"]
+#' sv_dependence(
+#'   x2, c("Petal.Length", "Species"), color_var = NULL, interactions = TRUE
 #' )
-#' sv_dependence(mx, "Petal.Length", color_var = NULL, interactions = TRUE)
+#' }
 #' @export
 #' @seealso [potential_interactions()]
 sv_dependence <- function(object, ...) {
diff --git a/R/sv_dependence2D.R b/R/sv_dependence2D.R
index d7d6167..6f1e6f6 100644
--- a/R/sv_dependence2D.R
+++ b/R/sv_dependence2D.R
@@ -28,15 +28,12 @@
 #' @param ... Arguments passed to [ggplot2::geom_jitter()].
 #' @returns An object of class "ggplot" (or "patchwork") representing a dependence plot.
 #' @examples
+#' \dontrun{
 #' dtrain <- xgboost::xgb.DMatrix(data.matrix(iris[, -1]), label = iris[, 1])
-#' fit <- xgboost::xgb.train(data = dtrain, nrounds = 50, nthread = 1)
+#' fit <- xgboost::xgb.train(data = dtrain, nrounds = 10, nthread = 1)
 #' sv <- shapviz(fit, X_pred = dtrain, X = iris)
 #' sv_dependence2D(sv, x = "Petal.Length", y = "Species")
-#' sv_dependence2D(sv, x = "Petal.Length", y = "Sepal.Width")
 #' sv_dependence2D(sv, x = c("Petal.Length", "Species"), y = "Sepal.Width")
-#' sv_dependence2D(
-#'   sv, x = c("Petal.Length", "Species"), y = c("Petal.Width", "Sepal.Width")
-#' )
 #'
 #' # SHAP interaction values
 #' sv2 <- shapviz(fit, X_pred = dtrain, X = iris, interactions = TRUE)
@@ -48,6 +45,7 @@
 #' # mshapviz object
 #' mx <- split(sv, f = iris$Species)
 #' sv_dependence2D(mx, x = "Petal.Length", y = "Sepal.Width")
+#' }
 #' @export
 #' @seealso [sv_dependence()]
 sv_dependence2D <- function(object, ...) {
diff --git a/R/sv_force.R b/R/sv_force.R
index 6d99b11..1d8a463 100644
--- a/R/sv_force.R
+++ b/R/sv_force.R
@@ -11,17 +11,16 @@
 #'   (via [ggrepel::geom_text_repel()]).
 #' @returns An object of class "ggplot" (or "patchwork") representing a force plot.
 #' @examples
+#' \dontrun{
 #' dtrain <- xgboost::xgb.DMatrix(data.matrix(iris[, -1]), label = iris[, 1])
-#' fit <- xgboost::xgb.train(data = dtrain, nrounds = 50, nthread = 1)
+#' fit <- xgboost::xgb.train(data = dtrain, nrounds = 20, nthread = 1)
 #' x <- shapviz(fit, X_pred = dtrain, X = iris[, -1])
 #' sv_force(x)
 #' sv_force(x, row_id = 65, max_display = 3, size = 9, fill_colors = 4:5)
 #'
 #' # Aggregate over all observations with Petal.Length == 1.4
 #' sv_force(x, row_id = x$X$Petal.Length == 1.4)
-#'
-#' # Combine two force plots via {patchwork}
-#' sv_force(c(Obs1 = x[1], Obs2 = x[2]))
+#' }
 #' @export
 #' @seealso [sv_waterfall()]
 sv_force <- function(object, ...) {
diff --git a/R/sv_importance.R b/R/sv_importance.R
index 1e9f2fe..fd53c65 100644
--- a/R/sv_importance.R
+++ b/R/sv_importance.R
@@ -42,19 +42,16 @@
 #'   `kind = "no"` - a named numeric vector of sorted SHAP feature importances
 #'   (or a list of such vectors in case of an object of class "mshapviz").
 #' @examples
+#' \dontrun{
 #' X_train <- data.matrix(iris[, -1])
 #' dtrain <- xgboost::xgb.DMatrix(X_train, label = iris[, 1])
-#' fit <- xgboost::xgb.train(data = dtrain, nrounds = 50, nthread = 1)
+#' fit <- xgboost::xgb.train(data = dtrain, nrounds = 10, nthread = 1)
 #' x <- shapviz(fit, X_pred = X_train)
 #' sv_importance(x)
-#' sv_importance(x, kind = "beeswarm", show_numbers = TRUE)
 #' sv_importance(x, kind = "no")
+#' sv_importance(x, kind = "beeswarm", show_numbers = TRUE)
+#' }
 #'
-#' X <- data.frame(matrix(rnorm(1000), ncol = 20))
-#' S <- as.matrix(X)
-#' x2 <- shapviz(S, X)
-#' sv_importance(x2)
-#' sv_importance(x2, max_display = 5)
 #' @seealso \code{\link{sv_interaction}}
 #' @export
 sv_importance <- function(object, ...) {
diff --git a/R/sv_interaction.R b/R/sv_interaction.R
index 8f2f489..36c30d4 100644
--- a/R/sv_interaction.R
+++ b/R/sv_interaction.R
@@ -18,12 +18,13 @@
 #'   numeric matrix of average absolute SHAP interactions sorted by the average
 #'   absolute SHAP values (or a list of such matrices in case of "mshapviz" object).
 #' @examples
+#' \dontrun{
 #' dtrain <- xgboost::xgb.DMatrix(data.matrix(iris[, -1]), label = iris[, 1])
-#' fit <- xgboost::xgb.train(data = dtrain, nrounds = 50, nthread = 1)
+#' fit <- xgboost::xgb.train(data = dtrain, nrounds = 10, nthread = 1)
 #' x <- shapviz(fit, X_pred = dtrain, X = iris, interactions = TRUE)
-#' sv_interaction(x)
-#' sv_interaction(x, max_display = 2, size = 3, alpha = 0.1)
 #' sv_interaction(x, kind = "no")
+#' sv_interaction(x, max_display = 2, size = 3)
+#' }
 #' @seealso [sv_importance()]
 #' @export
 sv_interaction <- function(object, ...) {
diff --git a/R/sv_waterfall.R b/R/sv_waterfall.R
index 0222053..21e8783 100644
--- a/R/sv_waterfall.R
+++ b/R/sv_waterfall.R
@@ -35,8 +35,9 @@
 #'   will altogether suppress adding text to the bars.
 #' @returns An object of class "ggplot" (or "patchwork") representing a waterfall plot.
 #' @examples
+#' \dontrun{
 #' dtrain <- xgboost::xgb.DMatrix(data.matrix(iris[, -1]), label = iris[, 1])
-#' fit <- xgboost::xgb.train(data = dtrain, nrounds = 50, nthread = 1)
+#' fit <- xgboost::xgb.train(data = dtrain, nrounds = 20, nthread = 1)
 #' x <- shapviz(fit, X_pred = dtrain, X = iris[, -1])
 #' sv_waterfall(x)
 #' sv_waterfall(x, row_id = 123, max_display = 2, size = 9, fill_colors = 4:5)
@@ -48,16 +49,7 @@
 #'
 #' # Aggregate over all observations with Petal.Length == 1.4
 #' sv_waterfall(x, row_id = x$X$Petal.Length == 1.4)
-#'
-#' # More features
-#' X <- as.data.frame(matrix(1:100, nrow = 10))
-#' S <- as.matrix(X)
-#' shp <- shapviz(S, X)
-#' sv_waterfall(shp)
-#'
-#' # Combine two waterfall plots via {patchwork}
-#' sv_waterfall(c(Obs1 = x[1], Obs2 = x[2])) +
-#'   patchwork::plot_layout(ncol = 1)
+#' }
 #' @export
 #' @seealso [sv_force()]
 sv_waterfall <- function(object, ...) {
diff --git a/cran-comments.md b/cran-comments.md
index 90c84c6..c1cd7fe 100644
--- a/cran-comments.md
+++ b/cran-comments.md
@@ -1,3 +1,18 @@
+# Re-resubmission
+
+Examples that take 0.09 seconds on my Windows laptop take 6 seconds on Linux. I don't know how to fix this, so I am resorting to dontruns around all examples involving XGBoost.
+
+# Re-resubmission
+
+Examples still taking too long on Linux. No idea how to fix this except using dontruns...
+
+# Resubmission
+
+Examples taking too long on Linux. 
+
+- I have now reduced the number of examples.
+- And the number of boosting rounds.
+
 # shapviz 0.9.0
 
 Hello CRAN team
@@ -8,29 +23,21 @@ Hello CRAN team
 
 ## Checks look good
 
-### check(manual = TRUE, cran = TRUE) 
-
-> checking data for ASCII and uncompressed saves ... OK
-   WARNING
-  'qpdf' is needed for checks on size reduction of PDFs
+### check(manual = TRUE, cran = TRUE)
 
 > checking HTML version of manual ... NOTE
   Skipping checking HTML validation: no command 'tidy' found
 
-### check_rhub(): Some usual notes...
+### RHub
 
 * checking package dependencies ... NOTE
 Packages which this enhances but not available for checking:
   'fastshap', 'h2o', 'lightgbm'
 * checking HTML version of manual ... NOTE
+Skipping checking HTML validation: no command 'tidy' found
 Skipping checking math rendering: package 'V8' unavailable
-* checking for non-standard things in the check directory ... NOTE
-Found the following files/directories:
-  ''NULL''
-* checking for detritus in the temp directory ... NOTE
-Found the following files/directories:
-  'lastMiKTeXException'
-
-### check_win_devel()
+
+
+### Winbuilder()
 
 Status: OK
diff --git a/man/shapviz.Rd b/man/shapviz.Rd
index da5c9b6..3ef9642 100644
--- a/man/shapviz.Rd
+++ b/man/shapviz.Rd
@@ -168,11 +168,11 @@ return a "mshapviz" object, containing a "shapviz" object per output.
 S <- matrix(c(1, -1, -1, 1), ncol = 2, dimnames = list(NULL, c("x", "y")))
 X <- data.frame(x = c("a", "b"), y = c(100, 10))
 shapviz(S, X, baseline = 4)
-
+\dontrun{
 # XGBoost models
-X_pred <- data.matrix(iris[, -1L])
-dtrain <- xgboost::xgb.DMatrix(X_pred, label = iris[, 1L])
-fit <- xgboost::xgb.train(data = dtrain, nrounds = 50L, nthread = 1L)
+X_pred <- data.matrix(iris[, -1])
+dtrain <- xgboost::xgb.DMatrix(X_pred, label = iris[, 1])
+fit <- xgboost::xgb.train(data = dtrain, nrounds = 10, nthread = 1)
 
 # Will use numeric matrix "X_pred" as feature matrix
 x <- shapviz(fit, X_pred = X_pred)
@@ -187,15 +187,13 @@ sv_dependence(x, "Species")
 x <- shapviz(fit, X_pred = dtrain, X = iris)
 
 # Multiclass setting
-params <- list(objective = "multi:softprob", num_class = 3L)
-X_pred <- data.matrix(iris[, -5L])
-dtrain <- xgboost::xgb.DMatrix(X_pred, label = as.integer(iris[, 5L]) - 1L)
-fit <- xgboost::xgb.train(
-  params = params, data = dtrain, nrounds = 50L, nthread = 1L
-)
+params <- list(objective = "multi:softprob", num_class = 3, nthread = 1)
+X_pred <- data.matrix(iris[, -5])
+dtrain <- xgboost::xgb.DMatrix(X_pred, label = as.integer(iris[, 5]) - 1)
+fit <- xgboost::xgb.train(params = params, data = dtrain, nrounds = 10)
 
 # Select specific class
-x <- shapviz(fit, X_pred = X_pred, which_class = 3L)
+x <- shapviz(fit, X_pred = X_pred, which_class = 3)
 x
 
 # Or combine all classes to "mshapviz" object
@@ -203,9 +201,9 @@ x <- shapviz(fit, X_pred = X_pred)
 x
 
 # What if we would have one-hot-encoded values and want to explain the original column?
-X_pred <- stats::model.matrix(~ . -1, iris[, -1L])
-dtrain <- xgboost::xgb.DMatrix(X_pred, label = as.integer(iris[, 1L]))
-fit <- xgboost::xgb.train(data = dtrain, nrounds = 50L)
+X_pred <- stats::model.matrix(~ . -1, iris[, -1])
+dtrain <- xgboost::xgb.DMatrix(X_pred, label = as.integer(iris[, 1]))
+fit <- xgboost::xgb.train(data = dtrain, nrounds = 10, nthread = 1)
 x <- shapviz(
   fit,
   X_pred = X_pred,
@@ -217,32 +215,33 @@ summary(x)
 # Similarly with LightGBM
 if (requireNamespace("lightgbm", quietly = TRUE)) {
   fit <- lightgbm::lgb.train(
-    params = list(objective = "regression", num_thread = 1L),
-    data = lightgbm::lgb.Dataset(X_pred, label = iris[, 1L]),
-    nrounds = 50L,
-    verbose = -2L
+    params = list(objective = "regression", num_thread = 1),
+    data = lightgbm::lgb.Dataset(X_pred, label = iris[, 1]),
+    nrounds = 10,
+    verbose = -2
   )
 
   x <- shapviz(fit, X_pred = X_pred)
   x
 
   # Multiclass
-  params <- list(objective = "multiclass", num_class = 3L, num_thread = 1L)
-  X_pred <- data.matrix(iris[, -5L])
-  dtrain <- lightgbm::lgb.Dataset(X_pred, label = as.integer(iris[, 5L]) - 1L)
-  fit <- lightgbm::lgb.train(params = params, data = dtrain, nrounds = 50L)
+  params <- list(objective = "multiclass", num_class = 3, num_thread = 1)
+  X_pred <- data.matrix(iris[, -5])
+  dtrain <- lightgbm::lgb.Dataset(X_pred, label = as.integer(iris[, 5]) - 1)
+  fit <- lightgbm::lgb.train(params = params, data = dtrain, nrounds = 10)
 
   # Select specific class
-  x <- shapviz(fit, X_pred = X_pred, which_class = 3L)
+  x <- shapviz(fit, X_pred = X_pred, which_class = 3)
   x
 
   # Or combine all classes to a "mshapviz" object
   mx <- shapviz(fit, X_pred = X_pred)
   mx
-  all.equal(mx[[3L]], x)
+  all.equal(mx[[3]], x)
+}
 }
 }
 \seealso{
-\code{\link[=sv_importance]{sv_importance()}}, \code{\link[=sv_dependence]{sv_dependence()}}, \code{\link[=sv_interaction]{sv_interaction()}},
+\code{\link[=sv_importance]{sv_importance()}}, \code{\link[=sv_dependence]{sv_dependence()}}, \code{\link[=sv_dependence2D]{sv_dependence2D()}}, \code{\link[=sv_interaction]{sv_interaction()}},
 \code{\link[=sv_waterfall]{sv_waterfall()}}, \code{\link[=sv_force]{sv_force()}}, \code{\link[=collapse_shap]{collapse_shap()}}
 }
diff --git a/man/split.shapviz.Rd b/man/split.shapviz.Rd
index eb0e527..35227f2 100644
--- a/man/split.shapviz.Rd
+++ b/man/split.shapviz.Rd
@@ -20,12 +20,14 @@ A "mshapviz" object.
 Splits "shapviz" object along a vector \code{f} into an object of class "mshapviz".
 }
 \examples{
+\dontrun{
 dtrain <- xgboost::xgb.DMatrix(data.matrix(iris[, -1]), label = iris[, 1])
-fit <- xgboost::xgb.train(data = dtrain, nrounds = 50, nthread = 1)
+fit <- xgboost::xgb.train(data = dtrain, nrounds = 10, nthread = 1)
 sv <- shapviz(fit, X_pred = dtrain, X = iris)
 mx <- split(sv, f = iris$Species)
 sv_dependence(mx, "Petal.Length")
 }
+}
 \seealso{
 \code{\link[=shapviz]{shapviz()}}, \code{\link[=rbind.shapviz]{rbind.shapviz()}}
 }
diff --git a/man/sv_dependence.Rd b/man/sv_dependence.Rd
index 283c1b1..b486032 100644
--- a/man/sv_dependence.Rd
+++ b/man/sv_dependence.Rd
@@ -86,8 +86,9 @@ to focus on pure interaction effects (multiplied by two) or on pure main effects
 
 }}
 \examples{
+\dontrun{
 dtrain <- xgboost::xgb.DMatrix(data.matrix(iris[, -1]), label = iris[, 1])
-fit <- xgboost::xgb.train(data = dtrain, nrounds = 50, nthread = 1)
+fit <- xgboost::xgb.train(data = dtrain, nrounds = 10, nthread = 1)
 x <- shapviz(fit, X_pred = dtrain, X = iris)
 sv_dependence(x, "Petal.Length")
 sv_dependence(x, "Petal.Length", color_var = "Species")
@@ -95,17 +96,13 @@ sv_dependence(x, "Petal.Length", color_var = NULL)
 sv_dependence(x, c("Species", "Petal.Length"))
 sv_dependence(x, "Petal.Width", color_var = c("Species", "Petal.Length"))
 
-# SHAP interaction values
+# SHAP interaction values/main effects
 x2 <- shapviz(fit, X_pred = dtrain, X = iris, interactions = TRUE)
 sv_dependence(x2, "Petal.Length", interactions = TRUE)
-sv_dependence(x2, c("Petal.Length", "Species"), color_var = NULL, interactions = TRUE)
-
-# Show main effect of "Petal.Length" for setosa and virginica separately
-mx <- c(
-  setosa = x2[x2$X$Species == "setosa"],
-  virginica = x2[x2$X$Species == "virginica"]
+sv_dependence(
+  x2, c("Petal.Length", "Species"), color_var = NULL, interactions = TRUE
 )
-sv_dependence(mx, "Petal.Length", color_var = NULL, interactions = TRUE)
+}
 }
 \seealso{
 \code{\link[=potential_interactions]{potential_interactions()}}
diff --git a/man/sv_dependence2D.Rd b/man/sv_dependence2D.Rd
index e8170ce..7a685ee 100644
--- a/man/sv_dependence2D.Rd
+++ b/man/sv_dependence2D.Rd
@@ -93,15 +93,12 @@ to focus on pure interaction effects (multiplied by two).
 
 }}
 \examples{
+\dontrun{
 dtrain <- xgboost::xgb.DMatrix(data.matrix(iris[, -1]), label = iris[, 1])
-fit <- xgboost::xgb.train(data = dtrain, nrounds = 50, nthread = 1)
+fit <- xgboost::xgb.train(data = dtrain, nrounds = 10, nthread = 1)
 sv <- shapviz(fit, X_pred = dtrain, X = iris)
 sv_dependence2D(sv, x = "Petal.Length", y = "Species")
-sv_dependence2D(sv, x = "Petal.Length", y = "Sepal.Width")
 sv_dependence2D(sv, x = c("Petal.Length", "Species"), y = "Sepal.Width")
-sv_dependence2D(
-  sv, x = c("Petal.Length", "Species"), y = c("Petal.Width", "Sepal.Width")
-)
 
 # SHAP interaction values
 sv2 <- shapviz(fit, X_pred = dtrain, X = iris, interactions = TRUE)
@@ -114,6 +111,7 @@ sv_dependence2D(
 mx <- split(sv, f = iris$Species)
 sv_dependence2D(mx, x = "Petal.Length", y = "Sepal.Width")
 }
+}
 \seealso{
 \code{\link[=sv_dependence]{sv_dependence()}}
 }
diff --git a/man/sv_force.Rd b/man/sv_force.Rd
index bc1079c..3621732 100644
--- a/man/sv_force.Rd
+++ b/man/sv_force.Rd
@@ -96,17 +96,16 @@ baseline SHAP value.
 
 }}
 \examples{
+\dontrun{
 dtrain <- xgboost::xgb.DMatrix(data.matrix(iris[, -1]), label = iris[, 1])
-fit <- xgboost::xgb.train(data = dtrain, nrounds = 50, nthread = 1)
+fit <- xgboost::xgb.train(data = dtrain, nrounds = 20, nthread = 1)
 x <- shapviz(fit, X_pred = dtrain, X = iris[, -1])
 sv_force(x)
 sv_force(x, row_id = 65, max_display = 3, size = 9, fill_colors = 4:5)
 
 # Aggregate over all observations with Petal.Length == 1.4
 sv_force(x, row_id = x$X$Petal.Length == 1.4)
-
-# Combine two force plots via {patchwork}
-sv_force(c(Obs1 = x[1], Obs2 = x[2]))
+}
 }
 \seealso{
 \code{\link[=sv_waterfall]{sv_waterfall()}}
diff --git a/man/sv_importance.Rd b/man/sv_importance.Rd
index cbb026d..596ec13 100644
--- a/man/sv_importance.Rd
+++ b/man/sv_importance.Rd
@@ -111,19 +111,16 @@ are sorted in decreasing order of importance.
 
 }}
 \examples{
+\dontrun{
 X_train <- data.matrix(iris[, -1])
 dtrain <- xgboost::xgb.DMatrix(X_train, label = iris[, 1])
-fit <- xgboost::xgb.train(data = dtrain, nrounds = 50, nthread = 1)
+fit <- xgboost::xgb.train(data = dtrain, nrounds = 10, nthread = 1)
 x <- shapviz(fit, X_pred = X_train)
 sv_importance(x)
-sv_importance(x, kind = "beeswarm", show_numbers = TRUE)
 sv_importance(x, kind = "no")
+sv_importance(x, kind = "beeswarm", show_numbers = TRUE)
+}
 
-X <- data.frame(matrix(rnorm(1000), ncol = 20))
-S <- as.matrix(X)
-x2 <- shapviz(S, X)
-sv_importance(x2)
-sv_importance(x2, max_display = 5)
 }
 \seealso{
 \code{\link{sv_interaction}}
diff --git a/man/sv_interaction.Rd b/man/sv_interaction.Rd
index 37660aa..a8390fa 100644
--- a/man/sv_interaction.Rd
+++ b/man/sv_interaction.Rd
@@ -87,12 +87,13 @@ The features are sorted in decreasing order of usual SHAP importance.
 
 }}
 \examples{
+\dontrun{
 dtrain <- xgboost::xgb.DMatrix(data.matrix(iris[, -1]), label = iris[, 1])
-fit <- xgboost::xgb.train(data = dtrain, nrounds = 50, nthread = 1)
+fit <- xgboost::xgb.train(data = dtrain, nrounds = 10, nthread = 1)
 x <- shapviz(fit, X_pred = dtrain, X = iris, interactions = TRUE)
-sv_interaction(x)
-sv_interaction(x, max_display = 2, size = 3, alpha = 0.1)
 sv_interaction(x, kind = "no")
+sv_interaction(x, max_display = 2, size = 3)
+}
 }
 \seealso{
 \code{\link[=sv_importance]{sv_importance()}}
diff --git a/man/sv_waterfall.Rd b/man/sv_waterfall.Rd
index 48b87f1..35ac9ec 100644
--- a/man/sv_waterfall.Rd
+++ b/man/sv_waterfall.Rd
@@ -102,8 +102,9 @@ baseline SHAP value.
 
 }}
 \examples{
+\dontrun{
 dtrain <- xgboost::xgb.DMatrix(data.matrix(iris[, -1]), label = iris[, 1])
-fit <- xgboost::xgb.train(data = dtrain, nrounds = 50, nthread = 1)
+fit <- xgboost::xgb.train(data = dtrain, nrounds = 20, nthread = 1)
 x <- shapviz(fit, X_pred = dtrain, X = iris[, -1])
 sv_waterfall(x)
 sv_waterfall(x, row_id = 123, max_display = 2, size = 9, fill_colors = 4:5)
@@ -115,16 +116,7 @@ sv_waterfall(
 
 # Aggregate over all observations with Petal.Length == 1.4
 sv_waterfall(x, row_id = x$X$Petal.Length == 1.4)
-
-# More features
-X <- as.data.frame(matrix(1:100, nrow = 10))
-S <- as.matrix(X)
-shp <- shapviz(S, X)
-sv_waterfall(shp)
-
-# Combine two waterfall plots via {patchwork}
-sv_waterfall(c(Obs1 = x[1], Obs2 = x[2])) +
-  patchwork::plot_layout(ncol = 1)
+}
 }
 \seealso{
 \code{\link[=sv_force]{sv_force()}}
diff --git a/packaging.R b/packaging.R
index 1a5681f..4f3ddec 100644
--- a/packaging.R
+++ b/packaging.R
@@ -101,7 +101,7 @@ library(devtools)
 
 document()
 test()
-check(manual = TRUE, cran = TRUE, vignettes = TRUE)
+check(manual = TRUE, cran = TRUE, vignettes = FALSE)
 build()
 # build(binary = TRUE)
 install(upgrade = FALSE)
@@ -109,7 +109,7 @@ install(upgrade = FALSE)
 # Run only if package is public(!) and should go to CRAN
 if (FALSE) {
   check_win_devel()
-  check_rhub()
+  check_rhub(platforms = "debian-gcc-devel")
 
   # Wait until above checks are passed without relevant notes/warnings
   # then submit to CRAN
diff --git a/vignettes/geographic.Rmd b/vignettes/geographic.Rmd
index 9c91c4d..6814b7c 100644
--- a/vignettes/geographic.Rmd
+++ b/vignettes/geographic.Rmd
@@ -23,7 +23,7 @@ knitr::opts_chunk$set(
 
 ## Setting
 
-In a model with geographic components, we want to express a functional $T$ (usually the expectation or a quantile) of a response $Y$ as a function $f$ of a set of geographic features (latitude/longitude and/or postal code and/or other features varying with location):
+In a model with geographic components, we want to express a functional $T$ (usually the expectation or a quantile) of a response $Y$ as a function $f$ of a set of geographic features (latitude/longitude and/or postal code and/or other features varying with location), and other features:
 
 $$
   T(Y \mid X^\textrm{geo}, X^\textrm{other}) \approx f(X^\textrm{geo}, X^\textrm{other})
@@ -34,7 +34,7 @@ That's where the additivity of SHAP values comes into play: The sum of SHAP valu
 
 ## A first example
 
-For illustration, we will use a beautiful house price dataset containing information on about 14'000 houses sold in 2016 in Miami-Date County. Some of the columns are as follows:
+For illustration, we will use a beautiful house price dataset containing information on about 14'000 houses sold in 2016 in Miami-Dade County. Some of the columns are as follows:
 
 - **SALE_PRC**: Sale price in USD: Its logarithm will be our model response.
 - *LATITUDE*, *LONGITUDE*: Coordinates