Graph Hamiltonians

Netket provides users with the ability to specify Hamiltonians based on properties of a custom graph. Users can specify whether operators act on sites or edges with a particular color. For examples about how to specify the edges colors see the custom graphs documentation.

Specifying the local Hilbert space

The Hilbert space here is constructed as the tensor product of local Hilbert spaces in such a way that the many-body quantum numbers are

where each can take values in the set .

Parameter Possible values Description Default value
QuantumNumbers List of floats A set of local quantum numbers . They do not need to be ordered or be consecutive. None
Size Integer The total number of lattice sites None

Example

pars['Hilbert']={
    'QuantumNumbers' : [1,-1],
    'Size'           : L,
}

Specifying the Site and Bond Operators

Users must specify all three parameters below. If necessary, they may be set to an empty list.

Parameter Possible values Description Default value
SiteOps List of floating/complex matrices The unique operators that act on each site Empty List
BondOps List of floating/complex matrices The unique operators acting on bonds Empty List
BondOpColors List of integers Integer color tags corresponding to operators in BondOps. 0 BondOps(i)

Example


rom __future__ import print_function
import json
import numpy as np
import networkx as nx

#Sigma^z*Sigma^z interactions
sigmaz = [[1, 0], [0, -1]]
mszsz = (np.kron(sigmaz, sigmaz))

#Exchange interactions
exchange = np.asarray([[0, 0, 0, 0], [0, 0, 2, 0], [0, 2, 0, 0], [0, 0, 0, 0]])

#Couplings J1 and J2
J = [1, 0.4]
L = 20

pars = {}

# Define bond operators, labels, and couplings
bond_operator = [
    (J[0] * mszsz).tolist(),
    (J[1] * mszsz).tolist(),
    (-J[0] * exchange).tolist(),
    (J[1] * exchange).tolist(),
]

bond_color = [1, 2, 1, 2]

# Define custom graph
G = nx.Graph()
for i in range(L):
    G.add_edge(i, (i + 1) % L, color=1)
    G.add_edge(i, (i + 2) % L, color=2)

edge_colors = [[u, v, G[u][v]['color']] for u, v in G.edges]

# Specify custom graph
pars['Graph'] = {
    'Edges': list(G.edges),
    'EdgeColors': edge_colors,
}

#We chose a spin 1/2 hilbert space with total Sigmaz=0
pars['Hilbert'] = {
    'Name': 'Spin',
    'S': 0.5,
    'TotalSz': 0,
    'Nspins': L,
}

#defining our custom hamiltonian
pars['Hamiltonian'] = {
    'Name': 'Graph',
    'BondOps': bond_operator,
    'BondOpColors': bond_color,
}

#defining the wave function
pars['Machine'] = {
    'Name': 'RbmSpin',
    'Alpha': 1,
}

#defining the sampler
#here we use Hamiltonian sampling to preserve simmetries
pars['Sampler'] = {
    'Name': 'MetropolisHamiltonianPt',
    'Nreplicas': 16,
}

# defining the Optimizer
# here we use the Stochastic Gradient Descent
pars['Optimizer'] = {
    'Name': 'Sgd',
    'LearningRate': 0.01,
}

# defining the learning method
# here we use the Stochastic Reconfiguration Method
pars['Learning'] = {
    'Method': 'Sr',
    'Nsamples': 1.0e3,
    'NiterOpt': 10000,
    'Diagshift': 0.1,
    'UseIterative': True,
    'OutputFile': "test",
}