From 703959f2e7a0f03928e97735ea94092f9f09372b Mon Sep 17 00:00:00 2001
From: ldellane <lucie.della-negra@inria.fr>
Date: Mon, 30 Jun 2025 11:02:10 +0200
Subject: [PATCH] fix: valid argument missing in doc
---
R/machine_learning.R | 6 +++---
man/machine_learning.Rd | 2 +-
2 files changed, 4 insertions(+), 4 deletions(-)
diff --git a/R/machine_learning.R b/R/machine_learning.R
index 92159d6..f57df72 100644
--- a/R/machine_learning.R
+++ b/R/machine_learning.R
@@ -5,7 +5,7 @@
#' @param df_features <`data.frame`> contains the dataset ready to go through the models
#' @param df_outcome <`data.frame`> contains the outcome variables: a single outcome variable in case of classification, and two outcome variables in case of survival analysis. Namely, in case of `main_task == "classification"`, `df_outcome` should contain columns `outcome_col` and `id_col`. In case of `main_task == "survival"`, `df_outcome` should contain columns `outcome_col`, "event_col" and `id_col`.
#' @param job <`character`> can either be "train" or "predict"
-#' @param ml_model <`character`> model that should be used. Can either be "cox", "logistic", "random_forest", "random_survival_forest", "xgb_cox" or "xgb_aft"
+#' @param ml_model <`character`> model that should be used. Can either be "cox", "xgboost", "logistic", "random_forest", "random_survival_forest", "xgb_cox" or "xgb_aft"
#' @param ml_options <`list`> specifying some needed arguments and options for machine learning. The list should contain the following elements:
#' - t_max_survival: <`int`> maximum time of survival prediction, in case of `main_task == "survival"`
#' @param feature_engineer <`character`> feature engineering method to be applied during transform. Either "umap", "pca" or "none"
@@ -92,7 +92,7 @@ machine_learning <- function(df_features,
cat("Transforming the dataset...\n")
}
if (job == "train") {
- if (dummy_binary){
+ if (dummy_binary) {
transformer_results = transform_for_modeling_dichotomy(
df = df_features,
action = "train",
@@ -117,7 +117,7 @@ machine_learning <- function(df_features,
)
}
if (job == "predict") {
- if (dummy_binary){
+ if (dummy_binary) {
df_features = purrr::pluck(
transform_for_modeling_dichotomy(
df = df_features,
diff --git a/man/machine_learning.Rd b/man/machine_learning.Rd
index c32d628..e428509 100644
--- a/man/machine_learning.Rd
+++ b/man/machine_learning.Rd
@@ -29,7 +29,7 @@ machine_learning(
\item{job}{<\code{character}> can either be "train" or "predict"}
-\item{ml_model}{<\code{character}> model that should be used. Can either be "cox", "logistic", "random_forest", "random_survival_forest", "xgb_cox" or "xgb_aft"}
+\item{ml_model}{<\code{character}> model that should be used. Can either be "cox", "xgboost", "logistic", "random_forest", "random_survival_forest", "xgb_cox" or "xgb_aft"}
\item{outcome_col}{<\code{character}> name of the time column in case of \code{main_task == "survival"} or name of the single outcome column in case of \code{main_task == "classification"}}
--
GitLab