SkillAgentSearch skills...

ITensorAGP.jl

Adiabatically evolve matrix product states (MPS) using matrix produect operator (MPO) representations of the adiabatic gauge potential (AGP).

Install / Use

/learn @ITensor/ITensorAGP.jl
About this skill

Quality Score

0/100

Supported Platforms

Universal

README

ITensorAGP

This is a package that constructs the adiabatic gauge potential (AGP) as a matrix product operator (MPO), which can be used to adiabatically evolve matrix product states (MPSs), using an algorithm introduced in arXiv:2311.00748. The package is built using the ITensors.jl and ITensorMPS.jl libraries.

Installation

This package is currently not registered. You can install it with:

julia> using Pkg: Pkg

julia> Pkg.add(url="https://github.com/ITensor/ITensorAGP.jl.git")

Usage

For example, to compute the AGP of the Hamiltonian H::MPO with perturbation dH::MPO, you can use the agp function:

using ITensorAGP: agp

AGP, ls_error = agp(H, dH; cutoff=1e-6, nsweeps=10, maxdim=40)

Given the MPOs H and dH, this code constructs the AGP as an MPO using the linsolve function in ITensorMPS.jl, which calls the linsolve function in KrylovKit.jl.

Related Skills

View on GitHub
GitHub Stars4
CategoryDevelopment
Updated1y ago
Forks0

Languages

Julia

Security Score

70/100

Audited on Nov 18, 2024

No findings