Merge pull request #28 from Laksafoss/v0.1.1

Update v0.1.0 for CRAN submission
Laksafoss · Feb 26, 2024 · e8c0c87 · e8c0c87
2 parents 04923b4 + f490c1d
commit e8c0c87
Show file tree

Hide file tree

Showing 21 changed files with 271 additions and 576 deletions.
diff --git a/DESCRIPTION b/DESCRIPTION
@@ -1,10 +1,14 @@
 Package: EpiForsk
 Title: Code Sharing at the Department of Epidemiological Research at Statens Serum Institut 
-Version: 0.1.0
+Version: 0.1.1
 Authors@R: 
-    c(person("Anna", "Laksafoss", , "adls@ssi.dk", role = c("aut", "cre"),
+    c(person("Anders", "Husby", , "andh@ssi.dk", role = c("aut"),
+             comment = c(ORCID = "0000-0002-7634-8455")),
+      person("Anna", "Laksafoss", , "adls@ssi.dk", role = c("aut"),
              comment = c(ORCID = "0000-0002-9898-2924")),
-      person("Kim Daniel", "Jakobsen", , "kija@ssi.dk", role = c("aut"),
+      person("Emilia Myrup", "Thiesson", , "emth@ssi.dk", role = c("aut"),
+             comment = c(ORCID = "0000-0001-6258-4177")),
+      person("Kim Daniel", "Jakobsen", , "kija@ssi.dk", role = c("aut", "cre"),
              comment = c(ORCID = "0000-0003-0086-9980")),
       person("Mikael", "Andersson", , "aso@ssi.dk", role = c("aut"),
              comment = c(ORCID = "0000-0002-0114-2057")),
@@ -18,7 +22,7 @@ Description: This is a collection of assorted functions and examples collected
     regression functionalities a method for calculating the confidence intervals 
     for functions of parameters from a GLM, Bayes equivalent for hypothesis 
     testing with asymptotic Bayes factor, and several help functions for 
-    generalized random forest analysis using the grf package. 
+    generalized random forest analysis using 'grf'. 
 License: MIT + file LICENSE
 Encoding: UTF-8
 Roxygen: list(markdown = TRUE)
@@ -39,7 +43,7 @@ Imports:
     nnet,
     patchwork,
     policytree,
-    progress,
+    progressr,
     purrr,
     rlang,
     stringr,
@@ -54,14 +58,12 @@ LazyData: true
 Suggests: 
     cli,
     CVXR,
-    doParallel,
-    doSNOW,
-    foreach,
+    furrr,
+    future,
     ggsci,
     knitr,
     parallel,
     rmarkdown,
-    snow,
     testthat (>= 3.0.0)
 VignetteBuilder: knitr
 Config/testthat/edition: 3
diff --git a/R/EpiForsk_package.R b/R/EpiForsk_package.R
@@ -10,16 +10,6 @@
 #' testing with asymptotic Bayes factor, and several help functions for
 #' generalized random forest analysis using the grf package.
 #'
-#' In the package there are contributions from
-#' \itemize{
-#'   \item{ADLS : }{Anna Damkjær Laksafoss (https://orcid.org/0000-0002-9898-2924)}
-#'   \item{ANDH : }{Anders Husby (https://orcid.org/0000-0002-7634-8455)}
-#'   \item{ASO  : }{Mikael Andersson (https://orcid.org/0000-0002-0114-2057)}
-#'   \item{EMTH : }{Emilia Myrup Thiesson (https://orcid.org/0000-0001-6258-4177)}
-#'   \item{KIJA : }{Kim Daniel Jakobsen (https://orcid.org/0000-0003-0086-9980)}
-#'   \item{KLP  : }{Klaus Rostgaard (https://orcid.org/0000-0001-6220-9414)}
-#' }
-#'
 #' @importFrom gridExtra arrangeGrob
 #' @importFrom methods hasArg
 #' @importFrom dplyr .data
@@ -46,5 +36,10 @@ globalVariables(c("y"))
 globalVariables(c(".SD"))
 globalVariables(c(".N"))
 
+#' make package data table aware
+#'
+#' This package uses data.table as a fast alternative to dplyr in cases where
+#' performance is essential.
+#'
 #' @export
 .datatable.aware = TRUE
diff --git a/R/aso_odds_ratio_function.R b/R/aso_odds_ratio_function.R
@@ -82,6 +82,7 @@
 #'   expvars = c("tocc", "education", "tocc:education")
 #' )
 #'
+#' \donttest{
 #' # Conditional binomial logistic regression with some extra text added:
 #' func_est2 <- odds_ratio_function(
 #'   logan2,
@@ -90,6 +91,7 @@
 #'   matchgroup = "id",
 #'   textvar = "Testing function"
 #' )
+#' }
 #'
 #' # Standard binomial logistic regression as survey data with no prepared
 #' # weights:

diff --git a/R/kija_cate_surface.R b/R/kija_cate_surface.R
@@ -57,12 +57,12 @@
 #' @author KIJA
 #'
 #' @examples
-#' n <- 1500
-#' p <- 5
+#' n <- 1000
+#' p <- 3
 #' X <- matrix(rnorm(n * p), n, p) |> as.data.frame()
 #' X_d <- data.frame(
-#'   X_d1 = factor(sample(1:5, n, replace = TRUE)),
-#'   X_d2 = factor(sample(1:5, n, replace = TRUE))
+#'   X_d1 = factor(sample(1:3, n, replace = TRUE)),
+#'   X_d2 = factor(sample(1:3, n, replace = TRUE))
 #' )
 #' X_d <- DiscreteCovariatesToOneHot(X_d)
 #' X <- cbind(X, X_d)
@@ -72,13 +72,11 @@
 #' cf <- grf::causal_forest(X, Y, W)
 #' cate_surface <- CATESurface(
 #'   cf,
-#'   continuous_covariates = paste0("V", 1:4),
+#'   continuous_covariates = paste0("V", 1:2),
 #'   discrete_covariates = "X_d1",
 #'   grid = list(
 #'     V1 = 10,
-#'     V2 = 5,
-#'     V3 = -5:5,
-#'     V4 = 2
+#'     V2 = -5:5
 #'   ),
 #'   other_discrete = data.frame(
 #'     covs = "X_d2",

diff --git a/R/kija_causal_forest_dynamic_subgroups.R b/R/kija_causal_forest_dynamic_subgroups.R
@@ -38,8 +38,8 @@
 #' @author KIJA
 #'
 #' @examples
-#' n <- 1000
-#' p <- 5
+#' n <- 800
+#' p <- 3
 #' X <- matrix(rnorm(n * p), n, p) |> as.data.frame()
 #' W <- rbinom(n, 1, 0.5)
 #' event_prob <- 1 / (1 + exp(2 * (pmax(2 * X[, 1], 0) * W - X[, 2])))

diff --git a/R/kija_covariate_balance.R b/R/kija_covariate_balance.R
@@ -89,6 +89,7 @@
 #' @author KIJA
 #'
 #' @examples
+#' \donttest{
 #' n <- 1000
 #' p <- 5
 #' X <- matrix(rnorm(n * p), n, p) |>
@@ -152,6 +153,7 @@
 #'   cd_x_scale_width = 1,
 #'   cd_bar_width = 0.3
 #' )
+#' }
 #'
 #' @export
 

diff --git a/R/kija_discrete_covariates_to_onehot.R b/R/kija_discrete_covariates_to_onehot.R
@@ -26,7 +26,7 @@
 
 DiscreteCovariatesToOneHot <- function(df,
                                        factors = dplyr::everything()) {
-  if (!is.data.frame(df)) {
+  if (!inherits(df, "data.frame")) {
     stop("df must be a data.frame or data.frame like object.")
   }
   df_f <- df |> dplyr::select({{ factors }})