gbasis

gbasis is a pure-Python package for evaluating and analytically integrating Gaussian-type orbitals and their related quantities. The goal is to build a set of tools to the quantum chemistry community that are easily accessible and easy to use as to facilitate future scientific works.

Since basis set manipulation is often slow, Quantum Chemistry packages in Python often interface to a lower-level lanaguage, such as C++ and Fortran, for these parts, resulting in a more difficult build process and limited distribution. The hope is that gbasis can fill in this gap without a significant difference in performance.

Dependencies

• numpy >= 1.10
• scipy >= 1.0

Installation

From PyPi

Note: This is not supported yet.

pip install --user gbasis

From Conda

Note: This is not supported yet.

pip install gbasis -c theochem

From GitHub Repository

To install gbasis by itself,

git clone https://github.com/theochem/gbasis.git
cd gbasis
pip install --user -e .[dev]

To install gbasis with pyscf,

git clone https://github.com/theochem/gbasis.git
cd gbasis
pip install --user -e .[dev,pyscf]

To install gbasis with iodata,

pip install --user cython
pip install --user git+https://github.com/theochem/iodata.git@master
git clone https://github.com/theochem/gbasis.git
cd gbasis
pip install --user -e .[dev,iodata]

To use gbasis.integrals.libcint, the user must run the following script to build and install libcint into the gbasis/integrals directory,

tools/install_libcint.sh

This script depends on the following packages:

• CMake
• Git
• Python 3
• a C compiler (gcc or clang are recommended)
• a Common Lisp interpreter (sbcl or clisp are recommended) By default, the x86-optimized qcint package is used for libcint. If this doesn't work on your computer, then run the script with the environment variable USE_LIBCINT=1 to use the regular libcint package:

USE_LIBCINT=1 tools/install_libcint.sh

Note that iodata must be installed separately. cython is a dependency of iodata.

To test the installation,

tox -e qa

Note that the interfaces to pyscf and iodata are not tested in this environment. To test the interface to pyscf, run

tox -e pyscf

and to test the interface to iodata, run

tox -e iodata

Features

Following features are supported in gbasis:

Importing basis set

• from Gaussian94 basis set file (gbasis.parsers.parse_gbs)
• from NWChem basis set file (gbasis.parsers.parse_nwchem)
• from iodata (gbasis.wrappers.from_iodata)
• from pyscf (gbasis.wrappers.from_pyscf)

Evaluations

• of basis sets (gbasis.eval.evaluate_basis)
• of arbitrary derivative of basis sets (gbasis.eval_deriv.evaluate_deriv_basis)
• of density (gbasis.density.evaluate_density)
• of arbitrary derivative of density (gbasis.density.evaluate_deriv_density)
• of gradient of density (gbasis.density.evaluate_density_gradient)
• of Laplacian of density (gbasis.density.evaluate_density_laplacian)
• of Hessian of density (gbasis.density.evaluate_density_hessian)
• of stress tensor (gbasis.stress_tensor.evaluate_stress_tensor)
• of Ehrenfest force (gbasis.stress_tensor.evaluate_ehrenfest_force)
• of Ehrenfest Hessian (gbasis.stress_tensor.evaluate_ehrenfest_hessian)
• of positive-definite kinetic energy (gbasis.density.evaluate_posdef_kinetic_energy_density)
• of general form of the kinetic energy (gbasis.density.evaluate_general_kinetic_energy_density)
• of electrostatic potential (gbasis.electrostatic_potential.electrostatic_potential)

Integrals

• overlap integrals of a basis set (gbasis.overlap.overlap_integral)
• overlap integrals between two basis sets (gbasis.overlap_asymm.overlap_integral_asymmetric)
• arbitrary multipole moment integral (gbasis.moment.moment_integral)
• kinetic energy integral (gbasis.kinetic_energy.kinetic_energy.integral)
• momentum integral (gbasis.momentum.momentum_integral)
• angular momentum integral (gbasis.angular_momentum.angular_momentum_integral)
• point charge interaction integral (gbasis.point_charge.point_charge_integral)
• nuclear-electron attraction integral (gbasis.point_charge.point_charge_integral)
• electron-electron repulsion integral (gbasis.electron_repulsion.electron_repulsion_integral)

Acknowledgements

This software was developed using funding from a variety of international sources including, but not limited to: Canarie, the Canada Research Chairs, Compute Canada, the European Union's Horizon 2020 Marie Sklodowska-Curie Actions (Individual Fellowship No 800130), the Foundation of Scientific Research--Flanders (FWO), McMaster University, the National Fund for Scientific and Technological Development of Chile (FONDECYT), the Natural Sciences and Engineering Research Council of Canada (NSERC), the Research Board of Ghent University (BOF), and Sharcnet.