This repository contains the code for the Random Forest visual analytics system of the VIS 2025 full paper "Cluster-Based Random Forest Visualization and Interpretation" by Max Sondag, Christofer Meinecke, Dennis Collaris, Tatiana von Landesberger, and Stef van den Elzen. The arXiv version is available at https://arxiv.org/abs/2507.22665
Install Git Bash: https://git-scm.com/downloads Install Python (Tested with 3.12.8), make sure to specify the versions in the command if you are using multiple python versions) https://www.python.org/downloads/release/python-3128/ -Add Python to environment variables -Tick py launcher Install Node.js (Tested with v22.17.1) https://nodejs.org/en/download -Tick Automatically install the necessary tools (Required for C++ from sci-kit) Install Microsoft Build Tools for Visual Studio https://visualstudio.microsoft.com/visual-cpp-build-tools/ -Tick Desktop development with C++ -Tick Node.js build tools
(Windows: Do this in Git Bash)
Create a virtual environemnt with:
py -3.12 -m venv random-forest-backend
Note: If you only have 1 version of python installed, you can use python instead of py -3.12
Activation:
source random-forest-backend/bin/activate
Windows: source random-forest-backend/Scripts/activate
Install with:
pip install -r backend/requirements.txt
You can probably use newer versions of python and the libraries but the evaluation was conducted with python 3.12.8 and the versions given in the backend/requirements.txt. This is mostly important because of loading scikit-learn models between major versions.
First install the package dependencies with:
$ npm install
Build with:
$ npm run build
You can cancel this with "ctrl+c" once you see the message "Webpack 5.90.1 compiled succesfully in XXXX ms".
Then run the development server with:
$ cd backend/
$ py -3.12 fastapi-rf.py
Leave it running, and start a new git-bash terminal in the root folder and run:
$ npm run serve
Use the browser to go to http://localhost:8080/ . You should now see the figure at the top of the readme, which correspond to Fig.1 from the paper.
Accepted are csv files with ",", ";", "|", or "\t" as delimiter. The target class needs to be the last column.