1. Getting Started

In E2E: Ensemble Learning Framework for Diagnostic and Prognostic Modeling

knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>"
)

E2E: An R Package for Easy-to-Build Ensemble Models

E2E is a comprehensive R package designed to streamline the development, evaluation, and interpretation of machine learning models for both diagnostic (classification) and prognostic (survival analysis) tasks. It provides a robust, extensible framework for training individual models and building powerful ensembles—including Bagging, Voting, and Stacking—with minimal code. The package also includes integrated tools for visualization and model explanation via SHAP values.

Author: Shanjie Luan (ORCID: 0009-0002-8569-8526)

Citation: If you use E2E in your research, please cite it as: "Shanjie Luan (2025). E2E: An R Package for Easy-to-Build Ensemble Models. https://github.com/XIAOJIE0519/E2E"

Note: The article is in the process of being written/submitted and is undergoing review by CRAN and further revisions. If you have any questions, please contact [e-mail].

Installation

The development version of E2E can be installed directly from GitHub using remotes.

# If you don't have remotes, install it first:
# install.packages("remotes")
remotes::install_github("XIAOJIE0519/E2E")

After installation, load the package into your R session:

library(E2E)

Core Concepts

E2E operates on two parallel tracks: Diagnostic Models and Prognostic Models. Before using functions from either track, you must initialize the corresponding system. This step registers a suite of pre-defined, commonly used models.

Sample Data

To follow the examples, you'll need sample data files. There are four data frames included in the package for you to try: train_dia, test_dia, train_pro, test_pro.

train_dia and test_dia are for diagnosis, with column names sample, outcome, variable 1, 2, 3.

head(train_dia)

train_pro and test_pro are for prognosis, with column names sample, outcome, time, variable 1, 2, 3.

head(train_pro)

E2E documentation built on Aug. 27, 2025, 1:09 a.m.