HamiltonIO 0.3.1

A library for reading/writting Hamiltonian with localized basis set.

HamiltonIO

A library for reading and analyzing Hamiltonian files from various DFT codes (ABACUS, SIESTA, EPW/Quantum ESPRESSO, Wannier90).

Features

• Multi-code support: ABACUS, SIESTA, EPW, Wannier90
• Spin configurations: Non-polarized, collinear, non-collinear, spin-orbit coupling
• Real and k-space analysis: Band structures, DOS, Hamiltonian decomposition
• Command-line tools: Easy-to-use CLI for common analysis tasks
• Python API: Full programmatic access for custom workflows

Installation

pip install hamiltonIO

Or install from source:

git clone https://github.com/mailhexu/HamiltonIO.git
cd HamiltonIO
pip install .

Quick Start

Command Line Tools

HamiltonIO provides CLI tools for quick analysis:

# Convert EPW files to NetCDF
hamiltonio-epw epw_to_nc --path ./data --prefix material --output epmat.nc

# Analyze ABACUS intra-atomic Hamiltonians
hamiltonio-abacus intra-atomic --outpath OUT.ABACUS

# Analyze SIESTA intra-atomic Hamiltonians  
hamiltonio-siesta intra-atomic siesta.fdf

For detailed CLI usage, see CLI Documentation.

Python Library

# ABACUS
from HamiltonIO.abacus import AbacusParser
parser = AbacusParser(outpath="./OUT.material/")
model = parser.get_models()
evals, evecs = model.solve([0, 0, 0])  # Solve at Gamma point

# SIESTA
from HamiltonIO.siesta import SiestaHam
model = SiestaHam("siesta.fdf")
evals, evecs = model.solve([0, 0, 0])

# EPW
from HamiltonIO.epw import Epmat
epmat = Epmat()
epmat.read(path="./", prefix="material", epmat_ncfile="epmat.nc")

# Wannier90
from HamiltonIO.wannier import Wannier90Hamiltonian
model = Wannier90Hamiltonian("wannier90_hr.dat")
evals, evecs = model.solve([0, 0, 0])

Documentation

Detailed documentation for each interface:

• Installation - Installation methods and requirements
• CLI Tools - Command-line interface usage
• ABACUS - ABACUS interface and examples
• SIESTA - SIESTA interface and examples
• EPW - EPW/Quantum ESPRESSO interface
• Wannier90 - Wannier90 interface
• Distance Analysis - Hopping distance analysis

Key Features by Code

ABACUS

• Automatic spin detection (non-polarized, collinear, noncollinear)
• Binary and text CSR format support
• Split-SOC analysis via finite difference
• Intra-atomic Hamiltonian decomposition

SIESTA

• NetCDF and HSX format support
• Split-SOC from native SIESTA output
• Collinear and non-collinear calculations
• Integration with sisl library

EPW (Quantum ESPRESSO)

• Binary to NetCDF conversion
• Wigner-Seitz grid handling
• Electron-phonon coupling matrices
• Crystal structure from EPW data

Wannier90

• HR format parsing
• Distance-resolved hopping analysis
• Band interpolation
• Orbital projections

Common Workflows

Intra-Atomic Analysis

# ABACUS: Analyze atoms 0, 1, 2 with full matrices
hamiltonio-abacus intra-atomic --outpath OUT.Fe \
    --atoms 0,1,2 --show-matrix -o analysis.txt

# SIESTA: Split-SOC analysis
hamiltonio-siesta intra-atomic siesta.fdf \
    --split-soc --show-matrix -o soc_analysis.txt

File Conversion

# EPW: Convert binary to NetCDF (faster I/O)
hamiltonio-epw epw_to_nc --path ./epw_data \
    --prefix material --output epmat.nc

Band Structure

import numpy as np
from HamiltonIO.abacus import AbacusParser

# Load model
parser = AbacusParser(outpath="./OUT.material/")
model = parser.get_models()

# High-symmetry path
k_path = np.array([
    [0.0, 0.0, 0.0],  # Gamma
    [0.5, 0.0, 0.0],  # X
    [0.5, 0.5, 0.0],  # M
])

# Calculate bands
bands = []
for k in k_path:
    evals, _ = model.solve(k)
    bands.append(evals)

Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

License

This project is licensed under the terms specified in the LICENSE file.

Citation

If you use HamiltonIO in your research, please cite:

@software{hamiltonIO,
  author = {Xu, He},
  title = {HamiltonIO: A Library for DFT Hamiltonian I/O},
  url = {https://github.com/mailhexu/HamiltonIO},
  year = {2024}
}

Contact

For questions, bug reports, or feature requests:

• GitHub Issues: https://github.com/mailhexu/HamiltonIO/issues
• Email: [e-mail]

Acknowledgments

HamiltonIO builds upon and integrates with:

• sisl - SIESTA interface
• ASE - Atomic simulation environment
• NumPy - Numerical computing