← Back to projects

Research

PsyML Toolkit

A local machine-learning workflow for research data.

Public release available · Apple-silicon Mac and Windows x64

Latest release: v0.1.1

PsyML Toolkit English interface: synthetic data checks and variable selection
Repository screenshot · example content · PsyML Toolkit · Apache-2.0

PsyML Toolkit is designed around the practical steps of a research analysis: defining variable roles, choosing preprocessing and validation strategies, comparing models, and inspecting results. Its graphical interface and command line share a Python analysis core. Desktop packages include the runtime and synthetic examples for trying the workflow.

My role combines defining research-oriented requirements, testing the interface, and refining the workflow, with AI-assisted development.

Use notes

Input data are processed locally. The interface supports Chinese, English, and French; generated reports support Chinese and English. Researchers remain responsible for study design, validation choices, interpretation, and checking the reports. Model performance does not establish causation or clinical validity, and internal validation does not replace independent external evaluation.

Built with

Python · scikit-learn · Godot. Code license: Apache-2.0.