Fix some links to online documentation

bookdown/19-GA.Rmd

@@ -82,8 +82,8 @@ Other options, such as `preProcess`, can be passed in as well.
 
 Some important options to `gafsControl` are:
 
- - `method`,  `number`, `repeats`, `index`,  `indexOut`, etc: options similar to those for [`train`](http://topepo.github.io/caret/training.html#control) top control resampling.
- - `metric`: this is similar to [`train`](http://topepo.github.io/caret/training.html#control)'s option but, in this case, the value should be a named vector with values for the internal and external metrics. If none are specified, the first value returned by the summary functions (see details below) are used and a warning is issued. A similar two-element vector for the option  `maximize` is also required. See the [last example here](#example2) for an illustration.
+ - `method`,  `number`, `repeats`, `index`,  `indexOut`, etc: options similar to those for [`train`](http://topepo.github.io/caret/model-training-and-tuning.html#control) top control resampling.
+ - `metric`: this is similar to [`train`](http://topepo.github.io/caret/model-training-and-tuning.html#control)'s option but, in this case, the value should be a named vector with values for the internal and external metrics. If none are specified, the first value returned by the summary functions (see details below) are used and a warning is issued. A similar two-element vector for the option  `maximize` is also required. See the [last example here](#example2) for an illustration.
  - `holdout`: this is a number between `[0, 1)` that can be used to hold out samples for computing the internal fitness value. Note that this is independent of the external resampling step. Suppose 10-fold CV is being used. Within a resampling iteration, `holdout` can be used to sample an additional proportion of the 90% resampled data to use for estimating fitness. This may not be a good idea unless you have a very large training set and want to avoid an internal resampling procedure to estimate fitness.
  - `allowParallel` and  `genParallel`: these are logicals to control where parallel processing should be used (if at all). The former will parallelize the external resampling while the latter parallelizes the fitness calculations within a generation. `allowParallel` will almost always be more advantageous.
 

bookdown/20-SA.Rmd

@@ -75,8 +75,8 @@ Other options, such as `preProcess`, can be passed in as well.
 
 Some important options to `safsControl` are:
 
- - `method`, `number`, `repeats`, `index`,  `indexOut`, etc: options similar to those for [`train`](http://topepo.github.io/caret/training.html#control) top control resampling.
- - `metric`: this is similar to [`train`](http://topepo.github.io/caret/training.html#control)'s option but, in this case, the value should be a named vector with values for the internal and external metrics. If none are specified, the first value returned by the summary functions (see details below) are used and a warning is issued. A similar two-element vector for the option `maximize` is also required. See the [last example here](#example2) for an illustration.
+ - `method`, `number`, `repeats`, `index`,  `indexOut`, etc: options similar to those for [`train`](http://topepo.github.io/caret/model-training-and-tuning.html#control) top control resampling.
+ - `metric`: this is similar to [`train`](http://topepo.github.io/caret/model-training-and-tuning.html#control)'s option but, in this case, the value should be a named vector with values for the internal and external metrics. If none are specified, the first value returned by the summary functions (see details below) are used and a warning is issued. A similar two-element vector for the option `maximize` is also required. See the [last example here](#example2) for an illustration.
  - `holdout`: this is a number between `[0, 1)` that can be used to hold out samples for computing the internal fitness value. Note that this is independent of the external resampling step. Suppose 10-fold CV is being used. Within a resampling iteration, `holdout` can be used to sample an additional proportion of the 90% resampled data to use for estimating fitness. This may not be a good idea unless you have a very large training set and want to avoid an internal resampling procedure to estimate fitness.
  - `improve`: an integer (or infinity) defining how many iterations should pass without an improvement in fitness before the current subset is reset to the last known improvement.
  - `allowParallel`: should the external resampling loop be run in parallel?.

html/GA.Rhtml

@@ -171,10 +171,10 @@ Some important options to <span class="mx funCall">gafsControl</span> are:
 </p>
 <ul>
   <li> <span class="mx arg">method</span>, <span class="mx arg">number</span>, <span class="mx arg">repeats</span>, <span class="mx arg">index</span>, <span class="mx arg">indexOut</span>, etc: 
-  options similar to those for <a href="http://topepo.github.io/caret/training.html#control"><span class="mx funCall">train</span></a> top control resampling.
+  options similar to those for <a href="http://topepo.github.io/caret/model-training-and-tuning.html#control"><span class="mx funCall">train</span></a> top control resampling.
   </li> 
   <li> <span class="mx arg">metric</span>: 
-  this is similar to <a href="http://topepo.github.io/caret/training.html#control"><span class="mx funCall">train</span></a>'s option but, in this case, the value should be a named vector with values for the internal and external metrics. If none are specified, the first value returned by the summary functions (see details below) are used and a warning is issued. A similar two-element vector for the option <span class="mx arg">maximize</span> is also required. See the <a href="#example2">last example here</a> for an illustration.
+  this is similar to <a href="http://topepo.github.io/caret/model-training-and-tuning.html#control"><span class="mx funCall">train</span></a>'s option but, in this case, the value should be a named vector with values for the internal and external metrics. If none are specified, the first value returned by the summary functions (see details below) are used and a warning is issued. A similar two-element vector for the option <span class="mx arg">maximize</span> is also required. See the <a href="#example2">last example here</a> for an illustration.
   </li> 
   <li> <span class="mx arg">holdout</span>: 
   this is a number between [0, 1) that can be used to hold out samples for computing the internal fitness value. Note that this is independent of the external resampling step. Suppose 10-fold CV is being used. Within a resampling iteration, <span class="mx arg">holdout</span> can be used to sample an additional proportion of the 90% resampled data to use for estimating fitness. This may not be a good idea unless you have a very large training set and want to avoid an internal resampling procedure to estimate fitness. 

html/SA.Rhtml

@@ -163,10 +163,10 @@ Some important options to <span class="mx funCall">safsControl</span> are:
 </p>
 <ul>
   <li> <span class="mx arg">method</span>, <span class="mx arg">number</span>, <span class="mx arg">repeats</span>, <span class="mx arg">index</span>, <span class="mx arg">indexOut</span>, etc: 
-  options similar to those for <a href="http://topepo.github.io/caret/training.html#control"><span class="mx funCall">train</span></a> top control resampling.
+  options similar to those for <a href="http://topepo.github.io/caret/model-training-and-tuning.html#control"><span class="mx funCall">train</span></a> top control resampling.
   </li> 
   <li> <span class="mx arg">metric</span>: 
-  this is similar to <a href="http://topepo.github.io/caret/training.html#control"><span class="mx funCall">train</span></a>'s option but, in this case, the value should be a named vector with values for the internal and external metrics. If none are specified, the first value returned by the summary functions (see details below) are used and a warning is issued. A similar two-element vector for the option <span class="mx arg">maximize</span> is also required. See the <a href="#example2">last example here</a> for an illustration.
+  this is similar to <a href="http://topepo.github.io/caret/model-training-and-tuning.html#control"><span class="mx funCall">train</span></a>'s option but, in this case, the value should be a named vector with values for the internal and external metrics. If none are specified, the first value returned by the summary functions (see details below) are used and a warning is issued. A similar two-element vector for the option <span class="mx arg">maximize</span> is also required. See the <a href="#example2">last example here</a> for an illustration.
   </li> 
   <li> <span class="mx arg">holdout</span>: 
   this is a number between [0, 1) that can be used to hold out samples for computing the internal fitness value. Note that this is independent of the external resampling step. Suppose 10-fold CV is being used. Within a resampling iteration, <span class="mx arg">holdout</span> can be used to sample an additional proportion of the 90% resampled data to use for estimating fitness. This may not be a good idea unless you have a very large training set and want to avoid an internal resampling procedure to estimate fitness. 

pkg/caret/R/gafs.R

@@ -56,7 +56,7 @@ ga_func_check <- function(x) {
 #'
 #' These functions are used with the \code{functions} argument of the
 #' \code{\link{gafsControl}} function. More information on the details of these
-#' functions are at \url{http://topepo.github.io/caret/GA.html}.
+#' functions are at \url{http://topepo.github.io/caret/feature-selection-using-genetic-algorithms.html}.
 #'
 #' Most of the \code{gafs_*} functions are based on those from the GA package
 #' by Luca Scrucca. These functions here are small re-writes to work outside of
@@ -102,7 +102,7 @@ ga_func_check <- function(x) {
 #'
 #' \url{cran.r-project.org/web/packages/GA/}
 #'
-#' \url{http://topepo.github.io/caret/GA.html}
+#' \url{http://topepo.github.io/caret/feature-selection-using-genetic-algorithms.html}
 #' @examples
 #'
 #' pop <- gafs_initial(vars = 10, popSize = 10)

pkg/caret/R/misc.R

@@ -437,7 +437,7 @@ parse_sampling <- function(x) {
   if(!(x_class %in% c("character", "function", "list"))) {
     stop(paste("The sampling argument should be either a",
                "string, function, or list. See",
-               "http://topepo.github.io/caret/training.html"))
+               "http://topepo.github.io/caret/model-training-and-tuning.html"))
   }
   if(x_class == "character") {
     x <- x[1]

pkg/caret/R/preProcess.R

@@ -150,7 +150,7 @@ invHyperbolicSineFunc <- function(x) log(x+sqrt(x^2+1))
 #' @seealso \code{\link{BoxCoxTrans}}, \code{\link{expoTrans}}
 #' \code{\link[MASS]{boxcox}}, \code{\link[stats]{prcomp}},
 #' \code{\link[fastICA]{fastICA}}, \code{\link{spatialSign}}
-#' @references \url{http://topepo.github.io/caret/preprocess.html}
+#' @references \url{http://topepo.github.io/caret/pre-processing.html}
 #' 
 #' Kuhn and Johnson (2013), Applied Predictive Modeling, Springer, New York
 #' (chapter 4)

pkg/caret/R/rfe.R

@@ -1,16 +1,16 @@
 #' Backwards Feature Selection
 #'
-#' A simple backwards selection, a.k.a. recursive feature selection (RFE),
+#' A simple backwards selection, a.k.a. recursive feature elimination (RFE),
 #' algorithm
 #'
 #' More details on this function can be found at
-#' \url{http://topepo.github.io/caret/featureselection.html}.
+#' \url{http://topepo.github.io/caret/recursive-feature-elimination.html}.
 #'
 #' This function implements backwards selection of predictors based on
 #' predictor importance ranking. The predictors are ranked and the less
 #' important ones are sequentially eliminated prior to modeling. The goal is to
 #' find a subset of predictors that can be used to produce an accurate model.
-#' The web page \url{http://topepo.github.io/caret/featureselection.html#rfe}
+#' The web page \url{http://topepo.github.io/caret/recursive-feature-elimination.html#rfe}
 #' has more details and examples related to this function.
 #'
 #' \code{rfe} can be used with "explicit parallelism", where different
@@ -48,7 +48,7 @@
 #' @param maximize a logical: should the metric be maximized or minimized?
 #' @param rfeControl a list of options, including functions for fitting and
 #' prediction. The web page
-#' \url{http://topepo.github.io/caret/featureselection.html#rfe} has more
+#' \url{http://topepo.github.io/caret/recursive-feature-elimination.html#rfe} has more
 #' details and examples related to this function.
 #' @param object an object of class \code{rfe}
 #' @param size a single integers corresponding to the number of features that
@@ -589,7 +589,7 @@ plot.rfe <- function (x,
 #' details of the feature selection algorithms used in this package.
 #'
 #' More details on this function can be found at
-#' \url{http://topepo.github.io/caret/featureselection.html#rfe}.
+#' \url{http://topepo.github.io/caret/recursive-feature-elimination.html#rfe}.
 #'
 #' Backwards selection requires function to be specified for some operations.
 #'
@@ -648,7 +648,7 @@ plot.rfe <- function (x,
 #' \code{\link{nbFuncs}}.
 #'
 #' Model details about these functions, including examples, are at
-#' \url{http://topepo.github.io/caret/featureselection.html}. .
+#' \url{http://topepo.github.io/caret/recursive-feature-elimination.html}. .
 #'
 #' @param functions a list of functions for model fitting, prediction and
 #' variable importance (see Details below)

pkg/caret/R/safs.R

@@ -177,8 +177,8 @@ predict.safs <- function (object, newdata, ...) {
 #' Many of these options are the same as those described for
 #' \code{\link[caret]{trainControl}}. More extensive documentation and examples
 #' can be found on the \pkg{caret} website at
-#' \url{http://topepo.github.io/caret/GA.html#syntax} and
-#' \url{http://topepo.github.io/caret/SA.html#syntax}.
+#' \url{http://topepo.github.io/caret/feature-selection-using-genetic-algorithms.html#syntax} and
+#' \url{http://topepo.github.io/caret/feature-selection-using-simulated-annealing.html#syntax}.
 #'
 #' The \code{functions} component contains the information about how the model
 #' should be fit and summarized. It also contains the elements needed for the
@@ -227,8 +227,8 @@ predict.safs <- function (object, newdata, ...) {
 #' \code{new}, and \code{iteration}, computes the acceptance probabilities
 #' }
 #'
-#' The pages \url{http://topepo.github.io/caret/GA.html} and
-#' \url{http://topepo.github.io/caret/SA.html} have more details about each of
+#' The pages \url{http://topepo.github.io/caret/feature-selection-using-genetic-algorithms.html} and
+#' \url{http://topepo.github.io/caret/feature-selection-using-simulated-annealing.html} have more details about each of
 #' these functions.
 #'
 #' \code{holdout} can be used to hold out samples for computing the internal
@@ -301,8 +301,8 @@ predict.safs <- function (object, newdata, ...) {
 #' @seealso \code{\link{safs}}, \code{\link{safs}}, , \code{\link{caretGA}},
 #' \code{\link{rfGA}}, \code{\link{treebagGA}}, \code{\link{caretSA}},
 #' \code{\link{rfSA}}, \code{\link{treebagSA}}
-#' @references \url{http://topepo.github.io/caret/GA.html},
-#' \url{http://topepo.github.io/caret/SA.html}
+#' @references \url{http://topepo.github.io/caret/feature-selection-using-genetic-algorithms.html},
+#' \url{http://topepo.github.io/caret/feature-selection-using-simulated-annealing.html}
 #' @keywords utilities
 #' @export safsControl
 safsControl <- function(functions = NULL,
@@ -447,9 +447,9 @@ safsControl <- function(functions = NULL,
 #' @return an object of class \code{safs}
 #' @author Max Kuhn
 #' @seealso \code{\link{safsControl}}, \code{\link{predict.safs}}
-#' @references \url{http://topepo.github.io/caret/GA.html}
+#' @references \url{http://topepo.github.io/caret/feature-selection-using-genetic-algorithms.html}
 #'
-#' \url{http://topepo.github.io/caret/SA.html}
+#' \url{http://topepo.github.io/caret/feature-selection-using-simulated-annealing.html}
 #'
 #' Kuhn and Johnson (2013), Applied Predictive Modeling, Springer
 #'
@@ -677,7 +677,7 @@ safs <- function (x, ...) UseMethod("safs")
 #'
 #' These functions are used with the \code{functions} argument of the
 #' \code{\link{safsControl}} function. More information on the details of these
-#' functions are at \url{http://topepo.github.io/caret/SA.html}.
+#' functions are at \url{http://topepo.github.io/caret/feature-selection-using-simulated-annealing.html}.
 #'
 #' The \code{initial} function is used to create the first predictor subset.
 #' The function \code{safs_initial} randomly selects 20\% of the predictors.
@@ -732,7 +732,7 @@ safs <- function (x, ...) UseMethod("safs")
 #' regression).
 #' @author Max Kuhn
 #' @seealso \code{\link{safs}}, \code{\link{safsControl}}
-#' @references \url{http://topepo.github.io/caret/SA.html}
+#' @references \url{http://topepo.github.io/caret/feature-selection-using-simulated-annealing.html}
 #' @examples
 #'
 #' selected_vars <- safs_initial(vars = 10 , prob = 0.2)

pkg/caret/R/selection.R

@@ -12,7 +12,7 @@
 #' distributions.
 #'
 #' More details on these functions can be found at
-#' \url{http://topepo.github.io/caret/training.html#custom}.
+#' \url{http://topepo.github.io/caret/model-training-and-tuning.html#custom}.
 #'
 #' By default, \code{\link{train}} uses \code{best}.
 #'