documents

R/ngb_dist.R

@@ -31,7 +31,8 @@ NGBDist <- R6::R6Class(
     pdf = function(quantiles, ...){
       private$dist$pdf(quantiles)
     },
-    #' @description Log of the probability density function at x of the given RV.
+    #' @description Log of the probability density function at x of the given
+    #' RV.
     #' This uses a more numerically accurate calculation if available.
     #' @param quantiles The quantiles
     #'
@@ -109,7 +110,8 @@ NGBDist <- R6::R6Class(
     },
 
     #' @description Confidence interval with equal areas around the median.
-    #' @param confidence Probability that an rv will be drawn from the returned range.
+    #' @param confidence Probability that an rv will be drawn from the returned
+    #' range.
     #' Each value should be in the range \code{0>= confidence <=1}.
     #'
     #' @return nd-points of range that contain \code{100 * confidence} of

R/ngbclass.R

@@ -24,8 +24,8 @@
 #'
 #'
 #' model <- NGBClassifier$new(Dist = Dist("k_categorical", K = 3),
-#'                            Base=DecisionTreeRegressor(criterion='friedman_mse',
-#'                                                       max_depth=2),
+#'                            Base=DecisionTreeRegressor(
+#'                            criterion='friedman_mse', max_depth=2),
 #'                            Score = Scores("LogScore"),
 #'                            natural_gradient=TRUE,
 #'                            n_estimators=500,
@@ -64,13 +64,18 @@ NGBClassifier <- R6::R6Class(
     #' data.A score from ngboost.scores, e.g. LogScore
     #' @param Base Base learner to use in the boosting algorithm.
     #' Any instantiated sklearn regressor, e.g. DecisionTreeRegressor()
-    #' @param natural_gradient Logical flag indicating whether the natural gradient should be used
+    #' @param natural_gradient Logical flag indicating whether the natural
+    #' gradient should be used
     #' @param n_estimators The number of boosting iterations to fit
     #' @param learning_rate The learning rate
-    #' @param minibatch_frac The percent subsample of rows to use in each boosting iteration
-    #' @param col_sample The percent subsample of columns to use in each boosting iteration
-    #' @param verbose Flag indicating whether output should be printed during fitting
-    #' @param verbose_eval Increment (in boosting iterations) at which output should be printed
+    #' @param minibatch_frac The percent subsample of rows to use in each
+    #' boosting iteration
+    #' @param col_sample The percent subsample of columns to use in each
+    #'  boosting iteration
+    #' @param verbose Flag indicating whether output should be printed
+    #' during fitting
+    #' @param verbose_eval Increment (in boosting iterations) at which output
+    #' should be printed
     #' @param tol Numerical tolerance to be used in optimization
     #' @param random_state Seed for reproducibility.
     #' @return An NGBRegressor object that can be fit.
@@ -118,7 +123,8 @@ NGBClassifier <- R6::R6Class(
       return(self)
     },
     #' @description An NGBRegressor object that can be fit.
-    #' @param X DataFrame object or List or numpy array of predictors (n x p) in Numeric format
+    #' @param X DataFrame object or List or numpy array of predictors (n x p)
+    #' in Numeric format
     #' @param Y DataFrame object or List or numpy array of outcomes (n)
     #' in numeric format. Should be floats for regression and integers from 0
     #'  to K-1 for K-class classification

R/ngbreg.R

@@ -66,13 +66,18 @@ NGBRegression <- R6::R6Class(
     #' data. A score from ngboost.scores, e.g. LogScore
     #' @param Base Base learner to use in the boosting algorithm.
     #' Any instantiated sklearn regressor, e.g. DecisionTreeRegressor()
-    #' @param natural_gradient Logical flag indicating whether the natural gradient should be used
+    #' @param natural_gradient Logical flag indicating whether the natural
+    #' gradient should be used
     #' @param n_estimators The number of boosting iterations to fit
     #' @param learning_rate The learning rate
-    #' @param minibatch_frac The percent subsample of rows to use in each boosting iteration
-    #' @param col_sample The percent subsample of columns to use in each boosting iteration
-    #' @param verbose Flag indicating whether output should be printed during fitting
-    #' @param verbose_eval Increment (in boosting iterations) at which output should be printed
+    #' @param minibatch_frac The percent subsample of rows to use in each
+    #' boosting iteration
+    #' @param col_sample The percent subsample of columns to use in each
+    #'  boosting iteration
+    #' @param verbose Flag indicating whether output should be printed during
+    #' fitting
+    #' @param verbose_eval Increment (in boosting iterations) at which output
+    #'  should be printed
     #' @param tol Numerical tolerance to be used in optimization
     #' @param random_state Seed for reproducibility.
     #' @return An NGBRegressor object that can be fit.
@@ -119,7 +124,8 @@ NGBRegression <- R6::R6Class(
       return(self)
     },
     #' @description An NGBRegressor object that can be fit.
-    #' @param X DataFrame object or List or numpy array of predictors (n x p) in Numeric format
+    #' @param X DataFrame object or List or numpy array of predictors (n x p)
+    #' in Numeric format
     #' @param Y DataFrame object or List or numpy array of outcomes (n)
     #' in numeric format. Should be floats for regression and integers from 0
     #'  to K-1 for K-class classification
@@ -132,8 +138,8 @@ NGBRegression <- R6::R6Class(
     #' @param val_sample_weight How much to weigh each example in the validation
     #' set. (defaults to 1)
     #' @param train_loss_monitor A custom score or set of scores to track on the
-    #' training set during training. Defaults to the score defined in the NGBoost
-    #' constructor.
+    #' training set during training. Defaults to the score defined in the
+    #' NGBoost constructor.
     #' @param val_loss_monitor A custom score or set of scores to track on the
     #'  validation set during training. Defaults to the score defined in the
     #'  NGBoost  constructor

R/ngbsurv.R

@@ -2,8 +2,10 @@
 #' @description
 #' NGBSurvival is a wrapper for the generic NGBoost class that facilitates
 #' survival analysis.
+#'
 #' Use this class if you want to predict an outcome that could take an infinite
-#'  number of (ordered) values, but right-censoring is present in the observed data.
+#' number of (ordered) values, but right-censoring is present in the observed
+#' data.
 #'
 #' @author Resul Akay
 #' @importFrom R6 R6Class
@@ -18,13 +20,18 @@ NGBSurvival <- R6::R6Class(
     #' data. A score from ngboost.scores, e.g. LogScore
     #' @param Base Base learner to use in the boosting algorithm.
     #' Any instantiated sklearn regressor, e.g. DecisionTreeRegressor()
-    #' @param natural_gradient Logical flag indicating whether the natural gradient should be used
+    #' @param natural_gradient Logical flag indicating whether the natural
+    #' gradient should be used
     #' @param n_estimators The number of boosting iterations to fit
     #' @param learning_rate The learning rate
-    #' @param minibatch_frac The percent subsample of rows to use in each boosting iteration
-    #' @param col_sample The percent subsample of columns to use in each boosting iteration
-    #' @param verbose Flag indicating whether output should be printed during fitting
-    #' @param verbose_eval Increment (in boosting iterations) at which output should be printed
+    #' @param minibatch_frac The percent subsample of rows to use in each
+    #' boosting iteration
+    #' @param col_sample The percent subsample of columns to use in each
+    #' boosting iteration
+    #' @param verbose Flag indicating whether output should be printed during
+    #'  fitting
+    #' @param verbose_eval Increment (in boosting iterations) at which output
+    #'  should be printed
     #' @param tol Numerical tolerance to be used in optimization
     #' @param random_state Seed for reproducibility.
     #' @return An NGBRegressor object that can be fit.
@@ -53,7 +60,6 @@ NGBSurvival <- R6::R6Class(
       private$tol <- tol
       private$random_state <- random_state
 
-      #ngboost <- reticulate::import("ngboost")
       private$model <- ngboost$NGBSurvival(
         Dist = private$Dist,
         Score = private$Score,

R/utils.R

@@ -56,8 +56,10 @@ Dist <- function(dist = c("Normal", "Bernoulli", "k_categorical", "StudentT",
 #' \itemize{
 #'
 #'  \item LogScore : Generic class for the log scoring rule.
-#'  \item CRPS : Generic class for the continuous ranked probability scoring rule.
-#'  \item CRPScore : Generic class for the continuous ranked probability scoring rule.
+#'  \item CRPS : Generic class for the continuous ranked probability scoring
+#'  rule.
+#'  \item CRPScore : Generic class for the continuous ranked probability
+#'  scoring rule.
 #'  \item MLE : Generic class for the log scoring rule.
 #'  }
 #' @author Resul Akay