pyDFT

Educational atomic electronic-structure toolkit with strict layered architecture.

Developed by Prof. Bin Shan (bshan@mail.hust.edu.cn).

Quick Start

Install editable package:

pip install -e .

Optional extras:

pip install -e '.[frontend,dev]'

Run desktop GUI (pywebview host):

pydft

Run CLI calculation:

pydft-cli run --symbol He --xc-model LDA

Run HTTP backend:

pydft-api --host 127.0.0.1 --port 8000

Compute Modes (GUI)

The GUI supports three compute modes:

• direct: in-process execution via the Python bridge (no HTTP hop)
• api: send request to configured HTTP backend URL
• auto: try direct first, then API fallback if direct fails

API base URL is configurable in the GUI form and is used by api and auto fallback paths.

Layered Architecture

pyDFT/
├── pyproject.toml
├── README.md
├── src/
│   └── pydft/
│       ├── __init__.py
│       ├── __main__.py
│       ├── main.py
│       ├── cli.py
│       ├── core/
│       │   ├── __init__.py
│       │   ├── __main__.py
│       │   ├── models.py
│       │   ├── presets.py
│       │   ├── request_mapper.py
│       │   └── parser.py
│       ├── application/
│       │   ├── __init__.py
│       │   └── scf.py
│       ├── methods/
│       │   ├── __init__.py
│       │   └── atomic/
│       │       ├── __init__.py
│       │       ├── dft_engine.py
│       │       ├── functionals.py
│       │       ├── grid.py
│       │       ├── hartree_fock.py
│       │       ├── lsda.py
│       │       ├── occupations.py
│       │       ├── potentials.py
│       │       └── radial_solver.py
│       ├── api/
│       │   ├── __init__.py
│       │   ├── __main__.py
│       │   ├── app.py
│       │   └── server.py
│       └── gui/
│           ├── __init__.py
│           ├── __main__.py
│           ├── bridge.py
│           ├── window.py
│           └── assets/
│               ├── index.html
│               ├── styles.css
│               └── app.js
├── tests/
├── benchmarks/
└── docs/

Docs

• User Manual
• Developer Manual
• Theoretical Introduction

License

This project is released under the MIT License. See LICENSE.

Copyright (c) 2026 Prof. Bin Shan