version 0.4.0

cran · Feb 10, 2022 · 3752909 · 3752909
1 parent 4eb2398
commit 3752909
Show file tree

Hide file tree

Showing 27 changed files with 362 additions and 311 deletions.
diff --git a/DESCRIPTION b/DESCRIPTION
@@ -1,6 +1,6 @@
 Package: rMIDAS
 Title: Multiple Imputation with Denoising Autoencoders
-Version: 0.3.0
+Version: 0.4.0
 Authors@R: c(
     person(given = "Thomas",
            family = "Robinson",
@@ -18,21 +18,21 @@ Authors@R: c(
     )
 Description: A tool for multiply imputing missing data using 'MIDAS', a deep learning method based on denoising autoencoder neural networks. This algorithm offers significant accuracy and efficiency advantages over other multiple imputation strategies, particularly when applied to large datasets with complex features. Alongside interfacing with 'Python' to run the core algorithm, this package contains functions for processing data before and after model training, running imputation model diagnostics, generating multiple completed datasets, and estimating regression models on these datasets.
 Encoding: UTF-8
-LazyData: true
 RoxygenNote: 7.1.1
 Depends: R (>= 3.6.0), data.table, mltools, reticulate
 Imports:
 Suggests: testthat, knitr, rmarkdown
+SystemRequirements: Python (>= 2.7.0)
 VignetteBuilder: knitr
 License: Apache License (>= 2.0)
 URL: https://github.com/MIDASverse/rMIDAS
 BugReports: https://github.com/MIDASverse/rMIDAS/issues
 NeedsCompilation: no
-Packaged: 2021-01-30 00:25:27 UTC; tomrobinson
+Packaged: 2022-02-09 21:10:24 UTC; tomrobinson
 Author: Thomas Robinson [aut, cre, cph]
     (<https://orcid.org/0000-0001-7097-1599>),
   Ranjit Lall [aut, cph] (<https://orcid.org/0000-0003-1455-3506>),
   Alex Stenlake [ctb, cph]
 Maintainer: Thomas Robinson <ts.robinson1994@gmail.com>
 Repository: CRAN
-Date/Publication: 2021-01-30 05:50:02 UTC
+Date/Publication: 2022-02-10 16:10:02 UTC
diff --git a/MD5 b/MD5
@@ -1,49 +1,50 @@
-450e99720eaab2d1ebe8b385a8afe763 *DESCRIPTION
+e1516f10256ae0ac5eb9d0378a5c121e *DESCRIPTION
 4b0e8158ffc6dfd349c09ae2cf78bf79 *NAMESPACE
-1ae7dbcfd58ed0252e8747c09a518e87 *NEWS.md
-310b8c5ebd9c9142693bb13d801fb34c *R/load_utils.R
-54bcecf6a71938796a02be05fb02cd16 *R/midas_functions.R
-35da741b0f6ece834d88d4e2c48ae048 *R/pre_processing.R
+72e952bb87fb2174353f8bc00390b6e0 *NEWS.md
+6e0cd17aed32c0311fe76f2996c70530 *R/load_utils.R
+26d1b0137cfb9139f0c5f16175386299 *R/midas_functions.R
+58f51f0d43076486ac2a81350f2d5733 *R/pre_processing.R
 0ef1815140875b09e36db14da6177157 *R/rubin_analysis.R
-5f0f7a477bc22f93c04aa9388a5b1059 *R/setup.R
-dfae92d29cde2eb804d727b91e0b02c2 *R/zzz.R
-675c3fb7b84afdfefcdee30fc47eefb7 *README.md
-a530231b4224e7197f03e30437bf016a *build/vignette.rds
+dad62e621f64cde04f7fd3146db0a8a0 *R/setup.R
+becdf5098e1c9a18c9bfc31688df0dd0 *R/zzz.R
+9fc2a00fd510bc69fe81fa51da72c1a8 *README.md
+19710b5c998a0619ba94e65348b60d93 *build/vignette.rds
 d49090aff7e7d8f2614da7cfcc8dda64 *inst/CITATION
-5387350a6623a416df9777722d1b85e4 *inst/doc/custom_python_versions.R
-f484f6445b5229eddb00fbd8881b021b *inst/doc/custom_python_versions.Rmd
-c1938cdce11c36f9c9ead57cee3e6223 *inst/doc/custom_python_versions.html
+69815d9858673c275fd3251958343e93 *inst/doc/custom_python_versions.R
+0c307d22142c8b940bb2eabf7a171f66 *inst/doc/custom_python_versions.Rmd
+fb9c7af541576a0e4cedb75bd0b7b349 *inst/doc/custom_python_versions.html
 072b1f5c9f87b73195cf00bb70bbd106 *inst/doc/imputation_demo.R
-78db257a69cec810b48a1dc8bac655d5 *inst/doc/imputation_demo.Rmd
-76e9243e2e6d460bb67fd07555456126 *inst/doc/imputation_demo.html
-64fba0a91b3a21bbbf50fa4f8a600741 *inst/examples/basic_workflow.R
-a67958e3009099ce054dfeb6836ecf22 *inst/examples/overimputation.R
+8fb23fc2e93eee5bb937a21099ca4563 *inst/doc/imputation_demo.Rmd
+2101464b80cbee5d820fd4ef07f36622 *inst/doc/imputation_demo.html
+08369e4174aab34c69bbb86bfe800694 *inst/examples/basic_workflow.R
+72a6c2985ab497dd0c35df39827011f3 *inst/examples/overimputation.R
 6a02980078146cc9bdd061054e3b4bfc *inst/python/__pycache__/midas_base.cpython-36.pyc
-393fee7a0486f0c55914ee243ecffa35 *inst/python/__pycache__/midas_base.cpython-37.pyc
-eb00af692b487faec78c83b934afa54f *inst/python/__pycache__/midas_base.cpython-38.pyc
+368cad3d8868ace931f5800300c687f9 *inst/python/__pycache__/midas_base.cpython-37.pyc
+3fdc1ad51197fe665796f6cfe5e32942 *inst/python/__pycache__/midas_base.cpython-38.pyc
 eb3f66e25ec913f65e7241c00dcfd437 *inst/python/__pycache__/midas_base.cpython-39.pyc
 73b4dc9395165c4e27482f00d11e8592 *inst/python/midas_base.py
 488fb2cb3ec9334b78245fb4e16f1937 *man/add_bin_labels.Rd
-c108ce6825a14d383201734677a83d54 *man/add_missingness.Rd
+deabaa625498cb22c65d443d4a60fabe *man/add_missingness.Rd
 a5a30b2cc7b62e1bd61a17d593e17927 *man/coalesce_one_hot.Rd
 c13c5d8b933d5dd21459c139ebf43538 *man/col_minmax.Rd
 2bb86f17a81bf892ae7e2b191f6f95fe *man/combine.Rd
-723a044dfc96ac49267f435af3893427 *man/complete.Rd
+a221670d70dff3414cac42094b0a80f5 *man/complete.Rd
 48413a5a794caa38137a9bc36ca02c4c *man/convert.Rd
 ee448b97e60bee367e6e5f98ba823b16 *man/figures/logo.png
 551e6d455d913ad82f2885625e08708b *man/figures/logos.drawio
 0029cb10bab1c771f84aa1892f749efd *man/import_midas.Rd
 a796458731eabf8e9e6bc35b2d828cf0 *man/mid_py_setup.Rd
 94665912e58b1aa50605a3fd9e98b08c *man/midas_setup.Rd
 094823bf4af2a3d7fed849c42a087ee7 *man/na_to_nan.Rd
-d4c2d7ae85b4b52fb50e33373fac6211 *man/overimpute.Rd
+d4c23b45ca9cbffa53fc72c99ea0d56e *man/overimpute.Rd
+fd482c198c4096b5eb1c5a0c6941c3e1 *man/python_configured.Rd
 67d01f8dd1cf19b84a79632f047922f4 *man/python_init.Rd
-bf582d11c2bbd5bdad111aaad3be8b36 *man/set_python_env.Rd
+02377b2e452e4f904da3e723ec4fbbac *man/set_python_env.Rd
 2aaaff54cab764411588113f92dbc22a *man/skip_if_no_numpy.Rd
-62467aedb62fbcd1df4133a204f2a7e9 *man/train.Rd
+5cd0176f8448ae970c45b1dce8caee50 *man/train.Rd
 499c92fc7cdd4e54e555f49b27dd88c9 *man/undo_minmax.Rd
 316d76f801fd61c6d2121a1687c0c4cd *tests/testthat.R
 d7ce0a0c02e69876725d5e39cc609feb *tests/testthat/testAnalysis.R
 aaee589ee89b5f9b3a27daa0972584a8 *tests/testthat/testPreProc.R
-f484f6445b5229eddb00fbd8881b021b *vignettes/custom_python_versions.Rmd
-78db257a69cec810b48a1dc8bac655d5 *vignettes/imputation_demo.Rmd
+0c307d22142c8b940bb2eabf7a171f66 *vignettes/custom_python_versions.Rmd
+8fb23fc2e93eee5bb937a21099ca4563 *vignettes/imputation_demo.Rmd
diff --git a/NEWS.md b/NEWS.md
@@ -1,11 +1,14 @@
+# rMIDAS 0.4
+
+* `python` argument in `set_python_env` renamed to `x` for clarity
+* Minor fixes including remedying bug in `complete()` function
+* Improved documentation
+
 # rMIDAS 0.3
 
 * Minor updates to underlying Python code to mirror MIDASpy v1.2.1
-
 * Added NULL defaults to cat_cols and bin_cols parameters within `rMIDAS::convert()`
-
 * Overimputation legend now plotted in bottom-right corner of figure
-
 * Minor changes to README
 
 # rMIDAS 0.2

diff --git a/R/load_utils.R b/R/load_utils.R
@@ -9,3 +9,29 @@ skip_if_no_numpy <- function() {
   if (!have_numpy)
     testthat::skip("numpy not available for testing")
 }
+
+#' Check whether Python is capable of executing example code
+#'
+#' Checks if each Python dependency is available.
+#' This function is called within some examples to ensure code executes properly.
+#' @keywords setup
+#' @return `NULL`
+python_configured <- function() {
+
+  if (!reticulate::py_available()) {
+    return(FALSE)
+  } else {
+
+    py_dep <- c("matplotlib","numpy","pandas","tensorflow","sklearn","os","random", "tensorflow_addons")
+    dep_avail <- sapply(py_dep, function (x) reticulate::py_module_available(x))
+
+    if (sum(dep_avail) == length(py_dep)) {
+      return(TRUE)
+    } else {
+      return(FALSE)
+    }
+
+  }
+}
+
+
diff --git a/R/midas_functions.R b/R/midas_functions.R
@@ -94,7 +94,7 @@ train <- function(data,
                            savepath= gsub("//","/",tempdir()))
 
   transf_model = FALSE
-  if (class(data) == "midas_pre") {
+  if (inherits(data, "midas_pre")) {
     binary_columns <- data$bin_list
     softmax_columns <- data$cat_lists
     data_in <- data$data
@@ -142,7 +142,7 @@ complete <- function(mid_obj,
                      file = NULL,
                      file_root = NULL) {
 
-  if (!("midas_base.Midas" %in% class(mid_obj))) {
+  if (!inherits(mid_obj,"midas_base.Midas")) {
     stop("Trained midas object not supplied to 'mid_obj' argument")
   }
 
@@ -198,9 +198,10 @@ complete <- function(mid_obj,
 
     }
 
-    if (cat_coalesce) {
+    cat_params <- mid_obj$preproc$cat_lists
+    if (cat_coalesce & !is.null(cat_params)) {
+
 
-      cat_params <- mid_obj$preproc$cat_lists
       cat_cols <- mid_obj$preproc$cat_names
 
       for (i in 1:length(cat_cols)) {
@@ -341,7 +342,7 @@ overimpute <- function(# Input data
                            savepath= tempdir())
 
   transf_model = FALSE
-  if (class(data) == "midas_pre") {
+  if (inherits(data, "midas_pre")) {
     binary_columns = data$bin_list
     softmax_columns = data$cat_lists
     transf_model = TRUE
@@ -354,7 +355,7 @@ overimpute <- function(# Input data
   matplotlib <- import("matplotlib", convert = TRUE)
   matplot_render <- try(matplotlib$use("TkAgg"), silent = TRUE)
 
-  if ("try-error" %in% class(matplot_render)) {
+  if (inherits(matplot_render, "try-error")) {
     stop("Cannot load TkAgg, which is needed to render the overimputation plot.\n You can try installing TkAgg by running the following at the command line: `sudo apt-get install python3-tk' ")
   }
 

diff --git a/R/pre_processing.R b/R/pre_processing.R
@@ -35,9 +35,9 @@ convert <- function(data, bin_cols = NULL, cat_cols = NULL, minmax_scale = FALSE
 
   # Check data input
 
-  if ("character" %in% class(data)) {
+  if (inherits(data,"character")) {
     data.table::fread(data)
-  } else  if (!("data.table" %in% class(data))) {
+  } else  if (!inherits(data,"data.table")) {
     data.table::setDT(data)
   }
 
@@ -269,10 +269,10 @@ coalesce_one_hot <- function(X, var_name, fast = TRUE) {
 
 #' Apply MAR missingness to data
 #'
-#' Helper function to re-apply binary variable labels post-imputation.
+#' Helper function to add missing values to data.
 #' @keywords preprocessing
 #' @param X A data.frame or similar
-#' @param prop Numeric between 0 and 1; the proportion of observations set to missing
+#' @param prop Numeric value between 0 and 1; the proportion of observations set to missing
 #' @param cols A vector of column names to be corrupted; if NULL, all columns are used
 #' @export
 #' @return Data with missing values

diff --git a/R/setup.R b/R/setup.R
@@ -4,13 +4,13 @@
 #' Users comfortable with reticulate can configure Python manually using `reticulate::use_`.
 #' Note: If users wish to set a custom binary/environment, this must be completed prior to the first call to either `train()` or `complete()`. The same is true if users use the reticulate package directly.
 #' If users wish to switch to a different Python binaries, R must be restarted prior to calling this function.
-#' @param python Character string, path to python binary, or directory of virtualenv, or name of conda environment
+#' @param x Character string, path to python binary, or directory of virtualenv, or name of conda environment
 #' @param type Character string, specifies whether to set a python binary ("auto"), "virtualenv", or "conda"
 #' @param ... Further arguments passed to `reticulate::use_condaenv()`
 #' @keywords setup
 #' @export
 #' @return Boolean indicating whether the custom python environment was activated.
-set_python_env <- function(python, type = "auto", ...) {
+set_python_env <- function(x, type = "auto", ...) {
 
   set_complete <- FALSE
 
@@ -19,26 +19,26 @@ set_python_env <- function(python, type = "auto", ...) {
     return(set_complete)
   } else if (type == "auto") {
 
-    set_py_attempt <- try(reticulate::use_python(python = python, required = TRUE),
+    set_py_attempt <- try(reticulate::use_python(python = x, required = TRUE),
                             silent = TRUE)
 
   } else if (type == "virtualenv") {
 
-    set_py_attempt <- try(reticulate::use_virtualenv(virtualenv = python, required = TRUE),
+    set_py_attempt <- try(reticulate::use_virtualenv(virtualenv = x, required = TRUE),
                           silent = TRUE)
 
   } else if (type == "conda") {
 
-    set_py_attempt <- try(reticulate::use_condaenv(condaenv = python, required = TRUE, ...),
+    set_py_attempt <- try(reticulate::use_condaenv(condaenv = x, required = TRUE, ...),
                           silent = TRUE)
 
   } else {
     set_py_attempt <- "error"
     class(set_py_attempt) <- "try-error"
   }
 
-  if ("try-error" %in% class(set_py_attempt)) {
-    stop("Setting user-specified python environment '",python, "' failed.
+  if (inherits(set_py_attempt,"try-error")) {
+    stop("Setting user-specified python environment '",x, "' failed.
          Please check the specified path/environment and try again.")
   }
 
@@ -70,7 +70,7 @@ python_init <- function() {
         load_stat <- substr(py_config()$version[1],1,1)
       }
 
-      if ("try-error" %in% class(load_stat)) {
+      if (inherits(load_stat, "try-error")) {
 
         stop("Unable to initialise Python and required packages.\n
             Please use set_python_env() to set the Python environment manually, then try again.")
@@ -129,7 +129,7 @@ mid_py_setup <- function() {
   py_pkg_load <- sapply(py_pkgs, function (py_pkg) try(reticulate::import(py_pkg, delay_load = FALSE),
                                                        silent = TRUE))
 
-  missing_pkg <- sapply(py_pkg_load, function (x) ("try-error" %in% class(x)))
+  missing_pkg <- sapply(py_pkg_load, function (x) inherits(x, "try-error"))
   missing_pkg <- py_dep[missing_pkg]
 
   if ("sklearn" %in% missing_pkg) {
@@ -155,7 +155,7 @@ mid_py_setup <- function() {
                                                   python_version = "<3.9"),
                            silent = TRUE)
 
-        if ("try-error" %in% class(pkg_install)) {
+        if (inherits(pkg_install, "try-error")) {
           stop("Unable to install package ", py_pkg, "\n")
         }
       }
@@ -176,15 +176,15 @@ mid_py_setup <- function() {
     if (py_v == "3.9") {
 
       warning("Packages installed but the R session needs to be restarted before proceeding.
-              Please restart R then call set_py_env('your_conda_name', type = 'conda')
+              Please restart R then call set_py_env('your_conda_name', type = 'conda').
               rMIDAS will then be ready to train and impute missing data.")
 
     } else {
 
       py_pkg_load <- sapply(py_pkgs, function (py_pkg) try(reticulate::import(py_pkg, delay_load = FALSE),
                                                            silent = TRUE))
 
-      inst_check <- sum(sapply(py_pkg_load, function (x) ("try-error" %in% class(x))))
+      inst_check <- sum(sapply(py_pkg_load, function (x) inherits(x,"try-error")))
 
       if (inst_check != 0) {
         stop("\nUnable to load required packages after install")

diff --git a/R/zzz.R b/R/zzz.R
@@ -4,3 +4,14 @@
   options("python_initialised" = NULL)
 
 }
+
+.onAttach <- function(libname, pkgname) {
+
+  packageStartupMessage("\n## \n",
+                        "## rMIDAS: Multiple Imputation using Denoising Autoencoders \n",
+                        "## Authors: Thomas Robinson and Ranjit Lall \n",
+                        "## Please visit https://github.com/MIDASverse/rMIDAS for more information \n",
+                        "## \n"
+  )
+
+}