GH-15894: Make sure the functions that are supposed to be exported are exported in the R package

h2o-docs/src/product/data-science/model_selection.rst

@@ -300,7 +300,7 @@ Examples
       4.8526636043  0.7153633278  0.0069487980 -0.0584344031  0.0791810013  0.4353149856  0.0126060611  -0.0005196059
 
       # Retrieve the list of coefficients for a subset size of 3:
-      coeff3 <- h2o.coeff(allsubsetsModel, 3)
+      coeff3 <- h2o.coef(allsubsetsModel, 3)
       print(coeff3)
       [[3]]
       Intercept    CAPSULE      DCAPS        PSA
@@ -351,7 +351,7 @@ Examples
 
       # To find out which variables get removed, build a new model with ``mode = "backward``
       # using the above training information:
-      bwModel = h2o.modelSelection(x = predictors, 
+      bwModel <- h2o.modelSelection(x = predictors,
                                    y = response, 
                                    training_frame = prostate, 
                                    seed = 12345, 

h2o-r/h2o-package/R/classes.R

@@ -451,6 +451,7 @@ coef.H2OCoxPHModelSummary <- function(object, ...) object@summary$coefficients
 #' @param fit an \code{H2OCoxPHModel} object.
 #' @param scale optional numeric specifying the scale parameter of the model.
 #' @param k numeric specifying the weight of the equivalent degrees of freedom.
+#' @export
 extractAIC.H2OCoxPHModel <- function(fit, scale, k = 2, ...) {
   fun <- get("extractAIC.coxph", getNamespace("stats"))
   if (missing(scale))

h2o-r/h2o-package/R/logging.R

@@ -44,10 +44,6 @@ NULL
     write(x = message, file = .h2o.getLogFileName(), append = TRUE)
 }
 
-h2o.logIt <- function(m, tmp, commandOrErr, isPost = TRUE) {
-  # Legacy.  Do nothing.
-}
-
 #' Start Writing H2O R Logs
 #'
 #' Begin logging H2o R POST commands and error responses to local disk. Used

h2o-r/h2o-package/R/models.R

@@ -755,11 +755,125 @@ predict_leaf_node_assignment.H2OModel <- function(object, newdata, type = c("Pat
 #' @export
 h2o.predict_leaf_node_assignment <- predict_leaf_node_assignment.H2OModel
 
+#' Use GRLM to transform a frame.
+#'
+#' @param model H2O GRLM model
+#' @param fr H2OFrame
+#' @return Returns a transformed frame
+#'
+#' @examples
+#' \dontrun{
+#' library(h2o)
+#' h2o.init()
+#'
+#' # Import the USArrests dataset into H2O:
+#' arrests <- h2o.importFile(
+#'   "https://s3.amazonaws.com/h2o-public-test-data/smalldata/pca_test/USArrests.csv"
+#' )
+#'
+#' # Split the dataset into a train and valid set:
+#' arrests_splits <- h2o.splitFrame(data = arrests, ratios = 0.8, seed = 1234)
+#' train <- arrests_splits[[1]]
+#' valid <- arrests_splits[[2]]
+#'
+#' # Build and train the model:
+#' glrm_model = h2o.glrm(training_frame = train,
+#'                       k = 4,
+#'                       loss = "Quadratic",
+#'                       gamma_x = 0.5,
+#'                       gamma_y = 0.5,
+#'                       max_iterations = 700,
+#'                       recover_svd = TRUE,
+#'                       init = "SVD",
+#'                       transform = "STANDARDIZE")
+#'
+#' # Eval performance:
+#' arrests_perf <- h2o.performance(glrm_model)
+#'
+#' # Generate predictions on a validation set (if necessary):
+#' arrests_pred <- h2o.predict(glrm_model, newdata = valid)
+#'
+#' # Transform the data using the dataset "valid" to retrieve the new coefficients:
+#' glrm_transform <- h2o.transform_frame(glrm_model, valid)
+#'}
+#' @export
 h2o.transform_frame <- function(model, fr) {
   if (!is(model, "H2OModel") || (is(model, "H2OModel") && model@algorithm != "glrm")) stop("h2o.transform_frame can only be applied to GLRM H2OModel instance.")
   return(.newExpr("transform", model@model_id, h2o.getId(fr)))
 }
 
+#' Retrieve the results to view the best predictor subsets.
+#'
+#' @param model H2OModelSelection  object
+#' @return Returns an H2OFrame object
+#'
+#' @examples
+#' \dontrun{
+#' library(h2o)
+#' h2o.init()
+#'
+#' # Import the prostate dataset:
+#' prostate <- h2o.importFile("http://s3.amazonaws.com/h2o-public-test-data/smalldata/logreg/prostate.csv")
+#'
+#' # Set the predictors & response:
+#' predictors <- c("AGE", "RACE", "CAPSULE", "DCAPS", "PSA", "VOL", "DPROS")
+#' response <- "GLEASON"
+#'
+#' # Build & train the model:
+#' allsubsetsModel <- h2o.modelSelection(x = predictors,
+#'                                       y = response,
+#'                                       training_frame = prostate,
+#'                                       seed = 12345,
+#'                                       max_predictor_number = 7,
+#'                                       mode = "allsubsets")
+#'
+#' # Retrieve the results (H2OFrame containing best model_ids, best_r2_value, & predictor subsets):
+#' results <- h2o.result(allsubsetsModel)
+#' print(results)
+#'
+#' # Retrieve the list of coefficients:
+#' coeff <- h2o.coef(allsubsetsModel)
+#' print(coeff)
+#'
+#' # Retrieve the list of coefficients for a subset size of 3:
+#' coeff3 <- h2o.coef(allsubsetsModel, 3)
+#' print(coeff3)
+#'
+#' # Retrieve the list of standardized coefficients:
+#' coeff_norm <- h2o.coef_norm(allsubsetsModel)
+#' print(coeff_norm)
+#'
+#' # Retrieve the list of standardized coefficients for a subset size of 3:
+#' coeff_norm3 <- h2o.coef_norm(allsubsetsModel)
+#' print(coeff_norm3)
+#'
+#' # Check the variables that were added during this process:
+#' h2o.get_predictors_added_per_step(allsubsetsModel)
+#'
+#' # To find out which variables get removed, build a new model with ``mode = "backward``
+#' # using the above training information:
+#' bwModel <- h2o.modelSelection(x = predictors,
+#'                               y = response,
+#'                               training_frame = prostate,
+#'                               seed = 12345,
+#'                               max_predictor_number = 7,
+#'                               mode = "backward")
+#' h2o.get_predictors_removed_per_step(bwModel)
+#'
+#' # To build the fastest model with ModelSelection, use ``mode = "maxrsweep"``:
+#' sweepModel <- h2o.modelSelection(x = predictors,
+#'                                  y = response,
+#'                                  training_frame = prostate,
+#'                                  mode = "maxrsweep",
+#'                                  build_glm_model = FALSE,
+#'                                  max_predictor_number = 3,
+#'                                  seed = 12345)
+#'
+#' # Retrieve the results to view the best predictor subsets:
+#' h2o.result(sweepModel)
+#' '}
+#'
+#' @export
 h2o.result <- function(model) {
   if (!is(model, "H2OModel")) stop("h2o.result can only be applied to H2OModel instances with constant results")
   return(.newExpr("result", model@model_id))

h2o-r/h2o-package/R/tf-idf.R

@@ -9,6 +9,7 @@
 #'
 #' @return  resulting frame with TF-IDF values.
 #'          Row format: documentID, word, TF, IDF, TF-IDF
+#' @export
 h2o.tf_idf <- function(frame, document_id_col, text_col, preprocess=TRUE, case_sensitive=TRUE) {
     col_indices <- c()
     for (col in c(document_id_col, text_col))

h2o-r/h2o-package/R/zzz.R

@@ -48,6 +48,13 @@
   registerS3method("print", "H2OExplanation", "print.H2OExplanation")
   .s3_register("repr", "repr_text", "H2OExplanation")
   .s3_register("repr", "repr_html", "H2OExplanation")
+
+  # CoxPH
+  .s3_register("survival", "survfit", "H2OCoxPHModel")
+
+  # H2OFrame
+  .s3_register("stats", "median", "H2OFrame")
+
   invisible()
 }
 

h2o-r/h2o-package/pkgdown/_pkgdown.yml

@@ -154,8 +154,8 @@ reference:
       - h2o.get_best_model_predictors
       - h2o.get_knot_locations
       - h2o.get_gam_knot_column_names
-      - h2o.get_predictor_added_per_step
-      - h2o.get_predictor_removed_per_step
+      - h2o.get_predictors_added_per_step
+      - h2o.get_predictors_removed_per_step
       - h2o.get_regression_influence_diagnostics
       - h2o.get_variable_inflation_factors
       - h2o.glm
@@ -333,8 +333,6 @@ reference:
       - h2o.tanh
       - h2o.target_encode_apply
       - h2o.target_encode_create
-      - h2o.target_encode_fit
-      - h2o.target_encode_transform
       - h2o.targetencoder
       - h2o.tf_idf
       - h2o.toFrame

h2o-r/scripts/h2o-r-test-setup.R

@@ -182,16 +182,28 @@ function() {
         strict_version_check <- TRUE
     } else if (IS.RUNIT) {
         # source h2o-r/h2o-package/R. overrides h2o package load
-        to_src <- c("aggregator.R", "classes.R", "connection.R","config.R", "constants.R", "logging.R", "communication.R",
-                    "kvstore.R", "frame.R", "targetencoder.R", "astfun.R","automl.R", "import.R", "parse.R", "export.R", "models.R", "edicts.R",
-                    "coxph.R", "coxphutils.R", "gbm.R", "glm.R", "gam.R", "anovaglm.R", "glrm.R", "kmeans.R", "deeplearning.R", "randomforest.R", "generic.R",
-                    "naivebayes.R", "pca.R", "svd.R", "locate.R", "grid.R", "word2vec.R", "w2vutils.R", "stackedensemble.R", "rulefit.R", "modelselection.R",
-                    "predict.R", "xgboost.R", "isolationforest.R", "psvm.R", "segment.R", "tf-idf.R", "explain.R", "permutation_varimp.R", "extendedisolationforest.R",
-                    "upliftrandomforest.R", "infogram.R", "isotonicregression.R", "admissibleml.R", "decisiontree.R", "adaboost.R")
-
-        src_path <- paste(h2oRDir,"h2o-package","R",sep=.Platform$file.sep)
-        invisible(lapply(to_src,function(x){source(paste(src_path, x, sep = .Platform$file.sep))}))
+        # to_src <- c("aggregator.R", "classes.R", "connection.R","config.R", "constants.R", "logging.R", "communication.R",
+        #             "kvstore.R", "frame.R", "targetencoder.R", "astfun.R","automl.R", "import.R", "parse.R", "export.R", "models.R", "edicts.R",
+        #             "coxph.R", "coxphutils.R", "gbm.R", "glm.R", "gam.R", "anovaglm.R", "glrm.R", "kmeans.R", "deeplearning.R", "randomforest.R", "generic.R",
+        #             "naivebayes.R", "pca.R", "svd.R", "locate.R", "grid.R", "word2vec.R", "w2vutils.R", "stackedensemble.R", "rulefit.R", "modelselection.R",
+        #             "predict.R", "xgboost.R", "isolationforest.R", "psvm.R", "segment.R", "tf-idf.R", "explain.R", "permutation_varimp.R", "extendedisolationforest.R",
+        #             "upliftrandomforest.R", "infogram.R", "isotonicregression.R", "admissibleml.R", "decisiontree.R", "adaboost.R")
 
+        # src_path <- paste(h2oRDir,"h2o-package","R",sep=.Platform$file.sep)
+        # invisible(lapply(to_src,function(x){source(paste(src_path, x, sep = .Platform$file.sep))}))
+        devtools::load_all(paste(h2oRDir,"h2o-package",sep=.Platform$file.sep), export_all=FALSE)
+
+        additional_imports <- c(
+            ".h2o.__remoteSend", ".as.survival.coxph.model", ".as.survival.coxph.summary",
+            ".getExpanded", ".str.list", "is.H2OFrame", ".get.session.property", ".set.session.property",
+            ".h2o.maximizing_metrics", ".h2o.doSafeGET", ".parse.h2oconfig", ".h2o.check_java_version",
+            "cut.H2OFrame", "as.data.frame.H2OFrame", ".h2o.perfect_auc", ".newExpr", ".h2o.doSafeREST",
+            ".h2o.fromJSON"
+         )
+
+        for (fn in additional_imports)
+            assign(fn, do.call(`:::`, list(quote(h2o), fn)), envir = .GlobalEnv)
+
         # source h2o-r/tests/runitUtils
         to_src <- c("utilsR.R", "pcaR.R", "deeplearningR.R", "glmR.R", "glrmR.R",
                     "gbmR.R", "kmeansR.R", "naivebayesR.R", "gridR.R", "shared_javapredict.R")
@@ -209,7 +221,7 @@ function() {
         masterSeedFile <- paste(getwd(), "/master_seed", sep = "")
         if (file.exists(masterSeedFile)) seed <- read.table(masterSeedFile)[[1]]
         setupSeed(seed)
-        h2o.logIt("[SEED] :", SEED)
+        cat("[SEED] :", SEED, "\n")
 
         if (!is.na(USERNAME)) {
             Log.info(paste0("Authenticating with http-basic, username: ", USERNAME, "."))

h2o-r/tests/runitUtils/utilsR.R

@@ -113,10 +113,13 @@ function(cur.dir, root, root.parent = NULL) {
 default.packages <-
 function() {
   to_require <- c("jsonlite", "RCurl", "RUnit", "R.utils", "testthat", "ade4", "glmnet", "gbm", "ROCR", "e1071",
-                  "tools", "statmod", "fpc", "cluster")
+                  "tools", "statmod", "fpc", "cluster", "survival")
   if (Sys.info()['sysname'] == "Windows") {
     options(RCurlOptions = list(cainfo = system.file("CurlSSL", "cacert.pem", package = "RCurl"))) }
-  invisible(lapply(to_require,function(x){require(x,character.only=TRUE,quietly=TRUE,warn.conflicts=FALSE)}))
+  invisible(lapply(to_require,function(x) {
+    if (!require(x,character.only=TRUE,quietly=TRUE,warn.conflicts=FALSE))
+      warning(paste("Package", x, "is not available."), call. = FALSE)
+  }))
 }
 
 read.zip<-

h2o-r/tests/testdir_algos/coxph/runit_coxph_cancer.R

@@ -16,4 +16,4 @@ test.CoxPH.cancer <- function() {
     expect_equal(fit$var, .as.survival.coxph.model(hex.fit@model)$var, tolerance = 1e-8)
 }
 
-doTest("CoxPH: Cancer Test", test.CoxPH.cancer)
+doTest("CoxPH: Cancer Test", test.CoxPH.cancer)