MTTSA

Magnetic Transport-related Tensor Symmetry Analyzer — determine magnetic space/point groups from a magnetic POSCAR and list symmetry-allowed transport tensor components.

Install (recommended)

pip install mttsa

This installs a pre-built wheel from PyPI (no local compilation). Requires Python ≥ 3.9.

mttsa path/to/structure.vasp
mttsa --version

Dependencies

Installed automatically: numpy, sympy, spglib.

If spglib fails via pip (common on some platforms), use conda first:

conda install -c conda-forge spglib numpy sympy
pip install mttsa

Slow or blocked PyPI access

Use a mirror, for example:

pip install mttsa -i https://pypi.tuna.tsinghua.edu.cn/simple \
  --trusted-host pypi.tuna.tsinghua.edu.cn

If you see SSLCertVerificationError (e.g. campus network proxy), try the mirror above or install from a downloaded wheel file (see Offline install below).

Development install (from source)

git clone https://github.com/PengJiang08/mttsa.git
cd mttsa
conda create -n mttsa python=3.11 -y
conda activate mttsa
conda install -c conda-forge numpy sympy spglib setuptools wheel -y
pip install -e . --no-build-isolation

--no-build-isolation avoids downloading build tools when the network blocks PyPI.

Offline install

Download mttsa-*.whl from PyPI (or GitHub Releases), then:

pip install /path/to/mttsa-0.1.0-py3-none-any.whl

Ensure numpy, sympy, and spglib are already installed in the same environment.

Usage

mttsa path/to/structure.vasp
mttsa path/to/structure.vasp --symprec 1e-3
mttsa path/to/structure.vasp 1e-3       # positional symprec (legacy)
mttsa path/to/structure.vasp --no-write # print only, no report files
mttsa -h
mttsa --version

Equivalent:

python -m mttsa path/to/structure.vasp

A sample POSCAR is in examples/MnBi2Te4/MnBi2Te4.vasp (monolayer MnBi2Te4). After pip install mttsa, it is also under share/mttsa/examples/MnBi2Te4/ in your Python environment prefix.

Outputs next to the input file:

• <name>_mttsa_report.txt — text report
• <name>_mttsa_report.html — HTML with MathJax formulas
• <name>_mttsa_report.json — machine-readable summary

Programmatic API:

from mttsa import run
data = run("structure.vasp")

Magnetic POSCAR format

Standard VASP header; each atom line has seven fields:

x_frac  y_frac  z_frac  Occupancy  Mx  My  Mz

(Mx, My, Mz) are magnetic moments in the Cartesian spin basis. Selective dynamics is not supported in this 7-column format.

Catalogued response tensors

Berry curvature, AHC, linear and nonlinear spin Hall (SHC), second-order conductivity (BCD / QMD / IMD channels), CISP, intrinsic NSHE, and SHG (even/odd channels). The report lists allowed/forbidden components under the identified magnetic point group, with independent components and symmetry relations.

For maintainers (PyPI upload)

Before each release:

python scripts/check_release.py
twine check dist/*
twine upload dist/*

Bump version in pyproject.toml and src/mttsa/__init__.py together.

License

MIT — see LICENSE.