gharib85/hamiltonianpy

Hamiltonian generator and several algorithms, including TBA, ED, CPT/VCA and DMRG, etc.

HamiltonianPy

In this package, we provide a unified framework to make a generic quantum lattice system programmable. Two central concepts are proposed to acheive this:

• unit-cell description ansatz (UCDA), and
• engine-app framework (EAF).

UCDA incoporates the geometry of the latttice, the internal degrees of freedom of the Hilbert space and the terms of the Hamiltonian as a whole, and generates all the needed operators of the system. EAF handles the interactions between the core algotithms (engine) and the user-defined tasks (app) as well as the logging of the program and the recording of the results, etc. Fundamental classes and functions are offered in the subpackage Basics as the APIs. Based on this, several algorithms are implemented, including:

• subpackage FreeSystem: tight binding approximation (TBA) and Bogoliubov de Gennes (BdG) equations for fermionic systems,
• subpackage ED: exact diagonalization (ED) for electron and spin systems,
• subpackage VCA: cluster pertubation theory (CPT) and variational cluster approach (VCA) for fermionic systems,
• subpackage TensorNetwork: tensor network represetation of Hamiltonians (MPO) and states (MPS),
• subpackage DMRG: density matrix renormalization group (DMRG) for electron and spin systems,
• subpackage FBFM: spin wave theory for flat band ferromagnets (FBFM).

Dependency

It works with python 3.6, and requires several packages:

• numpy latest version
• scipy latest version
• numba latest version
• matplotlib latest version
• mpi4py latest version

Notes

python 2.x is no longer supported.

Source code

https://github.com/waltergu/HamiltonianPy

Authors

• Zhao-Long Gu (NJU)

Contact

[email protected]