man-roxygen/Lrnr_base_extra.R

In jeremyrcoyle/sl3: Pipelines for Machine Learning and Super Learning

##' @section User Methods:
##'
##' \describe{
##' \item{\code{train(task)}}{
##'   Trains learner to a task using \code{delayed}. Returns a fit object
##'
##'   \itemize{
##'     \item{\code{task}: The task to use for training
##'     }
##'   }
##'   }
##'
##' \item{\code{base_train(task, trained_sublearners = NULL)}}{
##'   Trains learner to a task. Returns a fit object
##'
##'   \itemize{
##'     \item{\code{task}: The task to use for training
##'     }
##'     \item{\code{trained_sublearners}: Any sublearners previous trained. Almost always \code{NULL}
##'     }
##'   }
##'   }
##'
##' \item{\code{predict(task=NULL)}}{
##'   Generates predictions using \code{delayed}. Returns a prediction vector or matrix.
##'
##'   \itemize{
##'     \item{\code{task}: The task to use for prediction. If no task is provided, it will use the task used for training.
##'     }
##'   }
##'   }
##'
##' \item{\code{base_predict(task=NULL)}}{
##'   Generates predictions. Returns a prediction vector or matrix.
##'
##'   \itemize{
##'     \item{\code{task}: The task to use for prediction. If no task is provided, it will use the task used for training.
##'     }
##'   }
##'   }
##'
##' \item{\code{chain(task=NULL)}}{
##'   Generates a chained task using \code{delayed}
##'
##'   \itemize{
##'     \item{\code{task}: The task to use for chaining If no task is provided, it will use the task used for training.
##'     }
##'   }
##'   }
##'
##' \item{\code{base_chain(task=NULL)}}{
##'   Generates a chained task
##'
##'   \itemize{
##'     \item{\code{task}: The task to use for chaining If no task is provided, it will use the task used for training.
##'     }
##'   }
##'   }
##' }
##'
##' @section Fields:
##' \describe{
##'   \item{\code{is_trained}}{\code{TRUE} if this is a learner fit, not an untrained learner}
##'   \item{\code{fit_object}}{The internal fit object}
##'   \item{\code{name}}{The learner name}
##'   \item{\code{learner_uuid}}{A unique identifier of this learner, but common to all fits of this learner}
##'   \item{\code{fit_uuid}}{A unique identifier of this learner fit. \code{NULL} if this is an untrained learner}
##'   \item{\code{params}}{A list of learner parameters, as specified on construction}
##'   \item{\code{training_task}}{The task used for training. \code{NULL} if this is an untrained learner}
##'   \item{\code{training_outcome_type}}{The outcome_type of the task used for training. \code{NULL} if this is an untrained learner}
##'   \item{\code{properties}}{The properties supported by this learner}
##'   \item{\code{coefficients}}{Fit coefficients, if this learner has coefficients. \code{NULL} otherwise, or if this is an untrained learner}
##' }
##'
##' @section Internal Methods:
##'
##' These methods are primiarily for internal use only. They're not recommended for public consumption.
##'
##' \describe{
##' \item{\code{subset_covariates(task)}}{
##'   Returns a task with covariates subsetted using the \code{covariates} parameter.
##'
##'   \itemize{
##'     \item{\code{task}: The task to subset
##'     }
##'   }
##'   }
##'
##' \item{\code{get_outcome_type(task)}}{
##'   Mediates between the task outcome_type and parameter outcome_type. If a parameter outcome_type was specified, returns that. Otherwise,
##'   returns the task$outcome_type.
##'
##'   \itemize{
##'     \item{\code{task}: The task for which to determine the outcome_type
##'     }
##'   }
##'   }
##'
##' \item{\code{train_sublearners(task)}}{
##'   Trains sublearners to a task using \code{delayed}. Returns a delayed sublearner fit.
##'
##'   \itemize{
##'     \item{\code{task}: The task to use for training
##'     }
##'   }
##'   }
##'
##' \item{\code{set_train(fit_object, training_task)}}{
##'   Converts a learner to a learner fit.
##'
##'   \itemize{
##'     \item{\code{fit_object}: The fit object generated by a call to \code{private$.train}
##'     }
##'     \item{\code{training_task}: The task used for training
##'     }
##'   }
##'   }
##'
##' \item{\code{assert_trained()}}{
##'   Throws an error if this learner does not have a \code{fit_object}
##' }
##' }