PyIsingMonteCarlo

This is the repo for the python bindings to the ising_monte_carlo rust package.

Installation

Install Rust:

The backend for simulations is written in rust, a compiled language with strong type and memory security guaranties. To use this library you must first install the rust compiler by following the instructions here:

https://www.rust-lang.org/tools/install

And specifically the nightly toolchain to get access to some of the more advanced language features:

$ rustup toolchain install nightly

Setup your python3 virtualenv

Setting up a virtualenv will depend on your system, but if you're unsure you should be able to run pip install virtualenv to get the virtualenv tool. Then make the virtualenv where you'd like, and install some required packages:

$ virtualenv --python=python3 <path>
$ source <path>/bin/activate
$ pip install numpy maturin

Build and install the package

Navigate to this cloned repository, and build the release version of the library (meaning highly optimized).

$ git clone https://github.com/Renmusxd/PyIsingMonteCarlo
$ cd PyIsingMonteCarlo
$ maturin develop --release --strip

Building a wheel

If you'd prefer to make a wheel to install later, you can use the supplied makefile which runs

$ maturin build --release --strip --no-sdist

Usage

To use the library, import py_monte_carlo at the top of your python file, then build a lattice to simulate. The lattice takes a list of edges, each is ((a, b), j) where a and b are variable indices, and j is the bond. It's of the form J*Sza*Szb so positive is antiferromagnetic.

import py_monte_carlo
edges = [
    ((0, 1), 1.0),
    ((1, 2), -1.0)
]
lat = py_monte_carlo.Lattice(edges)

You may then run classical monte carlo using any of the method:

lat.run_monte_carlo(beta, timesteps, num_experiments)
lat.run_monte_carlo_sampling(beta, timesteps, num_experiments)
lat.run_monte_carlo_annealing(betas, timesteps, num_experiments)
lat.run_monte_carlo_annealing_and_get_energies(betas, timesteps, num_experiments)

Alternatively you can add a transverse field: lat.set_transverse_field(gamma) and run quantum monte carlo:

lat.run_quantum_monte_carlo(beta, timesteps, num_experiments)
lat.run_quantum_monte_carlo_sampling(beta, timesteps, num_experiments)
lat.run_quantum_monte_carlo_and_measure_spins(beta, timesteps, num_experiments)

as well as get correlation functions over time for either individual variables or broken bonds:

lat.run_quantum_monte_carlo_and_measure_variable_autocorrelation(beta, timesteps, num_experiments)
lat.run_quantum_monte_carlo_and_measure_bond_autocorrelation(beta, timesteps, num_experiments)

All of these functions have additional optional arguments, see docs for additional details.