TensorNetworkSolvers.jl

Support

TensorNetworkSolvers.jl is supported by the Flatiron Institute, a division of the Simons Foundation.

Installation instructions

This package resides in the ITensor/ITensorRegistry local registry. In order to install, simply add that registry through your package manager. This step is only required once.

julia> using Pkg: Pkg

julia> Pkg.Registry.add(url = "https://github.com/ITensor/ITensorRegistry")

or:

julia> Pkg.Registry.add(url = "[email protected]:ITensor/ITensorRegistry.git")

if you want to use SSH credentials, which can make it so you don't have to enter your Github ursername and password when registering packages.

Then, the package can be added as usual through the package manager:

julia> Pkg.add("TensorNetworkSolvers")

Examples

Perform a single sweep.

using Graphs: path_graph
using TensorNetworkSolvers: dmrg, dmrg_sweep
operator = path_graph(4)
regions = [(1, 2), (2, 3), (3, 4)]
tol = 1.0e-4
maxdim = 50
region_kwargs = (;
    update = (; tol),
    insert = (; maxdim),
)
state = []
x1 = dmrg_sweep(operator, state; regions, region_kwargs)

Sweep-dependent region kwargs (uniform across regions).

using Graphs: path_graph
using TensorNetworkSolvers: dmrg, dmrg_sweep
operator = path_graph(4)
regions = [(1, 2), (2, 3), (3, 4)]
nsweeps = 3
tols = [1.0e-3, 1.0e-4, 1.0e-5]
maxdims = [20, 50, 100]
region_kwargs = map(1:nsweeps) do i
    return (;
        update = (; tol = tols[i]),
        insert = (; maxdim = maxdims[i]),
    )
end
state = []
x2 = dmrg(operator, state; nsweeps, regions, region_kwargs)

Region-dependent kwargs.

using Graphs: path_graph
using TensorNetworkSolvers: dmrg, dmrg_sweep
operator = path_graph(4)
regions = [(1, 2), (2, 3), (3, 4)]
nsweeps = 3
tols = [1.0e-3, 1.0e-4, 1.0e-5]
maxdims = [20, 50, 100]
region_kwargs = map(1:nsweeps) do i
    return function (problem, alg, region_state)
        return (;
            update = (; tol = tols[i] / region_state.iteration),
            insert = (; maxdim = maxdims[i] * region_state.iteration),
        )
    end
end
state = []
x3 = dmrg(operator, state; nsweeps, regions, region_kwargs)

---

This page was generated using Literate.jl.