SA-OO-VQE

This package contains an implementation of SA-OO-VQE method for electronic structure computations using Quantum Processing Units or quantum simulators. It is possible to compute not only energies, but also gradients of potential energy surfaces, non-adiabatic couplings, etc. Everything in combination with both orbital optimization and/or diabatization. The package is mostly developed using Qiskit SDK.

The purpose of this README is only to get you going quickly. If you need more examples and in-depth explanations in the form of both user and programmer documentation, have a look at our official documentation website sa-oo-vqe-qiskit.rtfd.io.

Paper

The introductory paper about the package can be read here:

Installation from PyPi

First of all, install Psi4, pandoc and pip packages, if you haven't already. With these available, all the other dependencies will be taken care of by pip.

Now you can either use pip to install SA-OO-VQE from PyPi like

$ python3 -m pip install saoovqe

Manual installation

If you don't want to, or you can't use pip for installation of SA-OO-VQE, you can do it also manually.

Cloning the repository

$ git clone git@gitlab.com:MartinBeseda/sa-oo-vqe-qiskit.git

Installation via Conda

$ cd sa-oo-vqe-qiskit
$ conda env create -f environment.yml
$ conda activate saoovqe-env
$ python3 -m pip install .

Installation without Conda

In this case, you need to install Psi4, pandoc and pip manually, if you haven't already. Subsequently, you'll take care of the remaining dependencies via the following command.

$ python3 -m pip install qiskit==1.1.0 qiskit-nature>=0.7.0 numpy>=1.26.4 deprecated>=1.2.14 mendeleev>=0.13.1 scipy>=1.10.1 sympy>=1.11.1 setuptools>=67.8.0 lxml>=4.9.2 ipython jupyter pygments scikit-learn>=1.2.2 icecream>=2.1.3 pytest>=7.3.1 --upgrade

And now the only remaining thing is to go into a SA-OO-VQE root folder and installing the module itself.

$ cd sa-oo-vqe-qiskit
$ python3 -m pip install .

Testing the installation

That's all! Now you should be able to test your SA-OO-VQE installation.

If you want to be sure about all its abilities, you can run all the automated tests like

python3 -m pytest tests

but it can take quite a long time. If you want to test just the few basic use cases (~45s), try running the following command.

python3 -m pytest tests/test_vqe_optimization.py::TestSAOOVQE::test_get_energy tests/test_vqe_optimization.py::TestSAOOVQE::test_eval_eng_gradient tests/test_vqe_optimization.py::TestSAOOVQE::test_eval_nac

And when the test are successfully done, you can finally try importing your package and checking the version.

$ python3

>>> import saoovqe
>>> saoovqe.__version__

Bug reporting

If you encounter any bug, you can report it with Bug tag using an Issue tracker in this repository.

Asking for support

Is there anything we can help you with and you didn't find it in the documentation? Are you unsure, if some weird behavior is a bug or not? Ask us using our Issue tracker with a Question tag.