MLR-X

MLR-X is a multiple linear regression (MLR) application with both a desktop graphical user interface (GUI, based on Tkinter) and a command-line interface (CLI). It is designed to operate efficiently on both low- and high-dimensional datasets.

Features

• Desktop GUI for configuring and running multiple linear regression analyses.
• CLI support for automated and reproducible workflows.
• Export of results and visualizations in multiple formats.
• Integration with Python scientific libraries (NumPy, pandas, statsmodels, Pillow, scikit-learn, matplotlib).

Requirements

• Python 3.10 or higher (recommended).

On Linux, install the required system packages before running:

sudo apt-get install python3-tk
sudo apt-get install xvfb

Installation and Run Options

Choose one of the following methods to use MLR-X:

Option A) Install from PyPI (recommended)

pip install mlr-x

Run the application:

mlrx

In this mode, all required dependencies are installed automatically via pip.

Option B) Run from source

Install dependencies:

pip install -r requirements.txt

Run the application:

python MLRX.py

Usage

GUI mode

• Installed package: mlrx
• Source mode: python MLRX.py

When executed without arguments, MLR-X launches the graphical user interface.

CLI mode

Installed package:

mlrx <config.conf> [--model <id>] [--outputs ...] [--noruns]

Source mode:

python MLRX.py <config.conf> [--model <id>] [--outputs ...] [--noruns]

Key parameters

• --version: Display version information and exit.
• --model: Select a model identifier for output generation.
• --outputs: Specify outputs to generate (e.g., diagnostics, visualization, summary).
• Visualization outputs support formats such as pdf, png, tiff, and svg.
• --noruns: Use an existing results file from the configured output path (skips model execution).
• --onlyIV / --onlyEV: Perform only internal or external validation, respectively, using an existing results file. These options skip model search and require precomputed results.

Example

python MLRX.py example.conf --model 1 --outputs summary

Portable Binaries

Portable executables are provided for end users who prefer not to install Python or manage dependencies manually.

Download available builds from:
https://jacksonalcazar.github.io/MLR-X

Supported platforms

• Windows 10/11 (64-bit)
• macOS (Arm64)
• Ubuntu 20.04 (x86-64)

Documentation

• Software manual (included in this repository): MLR_X_v_1_0_Manual.pdf
• Precompiled binaries and releases: https://github.com/Jacksonalcazar/MLR-X/releases
• Issue tracker: https://github.com/Jacksonalcazar/MLR-X/issues

License

This project is distributed under the GNU Affero General Public License v3.0 (AGPL-3.0). See LICENSE.txt for details.

Trademark

The MLR-X name, logo, and visual identity are subject to trademark terms. See TRADEMARK.md for usage guidelines.

How to cite

Article under review. For now, please cite as follows:

• Alcázar, Jackson J. (2026). "MLR-X 1.0 software. Available at: https://jacksonalcazar.github.io/MLR-X/".