better doc

man-roxygen/details_tmax.R

@@ -2,5 +2,7 @@
 #' If `t_max` or `p_max` is given, then \eqn{G(t)} will be fitted using **all observations** from the
 #' train set (or test set) and only then the cutoff time will be applied.
 #' This is to ensure that more data is used for fitting the censoring distribution via the
-#' Kaplan-Meier and allow the user to remove the latest time points that may introduce 
-#' inflation of the proper score due to division by `eps`.
+#' Kaplan-Meier.
+#' Setting the `t_max` can help alleviate inflation of the score when `proper` is `TRUE`,
+#' in cases where an observation is censored at the last observed time point.
+#' This results in \eqn{G(t_{max}) = 0} and the use of `eps` instead (when `t_max` is `NULL`).

man-roxygen/param_tmax.R

@@ -4,5 +4,4 @@
 #' Mutually exclusive with `p_max` or `times`.
 #' This will effectively remove test observations for which the time
 #' (event or censoring) is less than `t_max`.
-#' Useful also for avoiding division by eps which can inflate the score
-#' especially in the case when `proper` is `TRUE`.
+#' It's recommended to set `t_max` to avoid division by `eps`, see Details.