amplify.Solver

class Solver

A class that provides an interface for solving input and logical models with a specified client. The solver class abstracts the hardware specifications of the Ising machines and provides an interface that integrates the differences among various Ising machine clients.

注釈

Acceptable clients are described in the references

__init__(client=None)

Constructor

パラメータ:

client (client, optional) -- Client object. Defaults to None.
chain_strength (float/optional) -- A parameter for controlling the quadratic interactions within the connected physical bits of a given solver for forming chains. It is set to 1.0 by default.

Example

from amplify import BinarySymbolGenerator, Solver
from amplify.client import FixstarsClient

gen = BinarySymbolGenerator()
q = gen.array(3)
f = 2 * q[0] * q[1] * q[2] - q[0] * q[1] + q[2] + 1

client = FixstarsClient()
client.token = "xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx"
client.parameters.timeout = 1000  # timeout 1second

solver = Solver(client)
result = solver.solve(f)

>>> f
2 q_0 q_1 q_2 - q_0 q_1 + q_2 + 1
>>> [f"energy = {s.energy}, q = {q.decode(s.values)}" for s in result]
['energy = 0.0, q = [1. 1. 0.]']

Methods

`__init__`([client])	Constructor
`solve`(args, *kwargs)	Solve a given input or logical model.

Attributes

`chain_strength`	A parameter for controlling the quadratic interactions within the connected physical bits of a given solver for forming chains.
`client`	A client class of Amplify SDK.
`client_result`	Gives the result class of the physical model, returned by the client used as the driver for the solver class.
`deduplicate`	Boolean value that specifies whether to output identical solutions found in the calculation.
`embedding_time_limit`
`execution_time`	The execution time in the most recent job, measured in milliseconds.
`filter_solution`	Boolean value that specifies whether the solutions are filtered based on given constraints.
`logical_result`	Gives the result class of the logical model.
`sort_solution`	Boolean value that specify whether the output solutions are sorted in order of best solution first.

property chain_strength

A parameter for controlling the quadratic interactions within the connected physical bits of a given solver for forming chains. Set to 1.0 by default.

Type:: float

property client: A client class of Amplify SDK. Select one client class from reference of clients.

property client_result: Gives the result class of the physical model, returned by the client used as the driver for the solver class. For more details, see the result class of the client used for the driver. For example, FixstarsClientResult if Amplify Annealing Engine client is used.

property deduplicate

Boolean value that specifies whether to output identical solutions found in the calculation. When set to True, only one of multiple identical solutions will be output. When set to False, all of multiple identical solutions will be output. Set to True by default.

Type:: bool

property execution_time

The execution time in the most recent job, measured in milliseconds.

Type:: int

property filter_solution

Boolean value that specifies whether the solutions are filtered based on given constraints. Set to True by default. If set to False, the solutions that do not satisfy the constraint condtions will also be output.

Type:: bool

property logical_result

Gives the result class of the logical model. The result of the logical model is the one before the filtering based on constraints.

Type:: LogicalResult

solve(*args, **kwargs)

Solve a given input or logical model.

Overloads:

solve(model)
solve(poly)
solve(matrix, constant)
solve(constraint)

パラメータ:

model (BinaryQuadraticModel, BinaryIntQuadraticModel, IsingQuadraticModel or IsingIntQuadraticModel) --
- Input quadratic model.
poly (BinaryPoly, BinaryIntPoly, IsingPoly or IsingIntPoly) --
- Input polynomial. It is internally converted to a Quadratic Model.
matrix (BinaryMatrix, BinaryIntMatrix, IsingMatrix or IsingIntMatrix) --
- Input matrix. It is internally converted to a Quadratic Model.
constant (int or float, optional) --
- A constant value added to the above matrix. The type is according to the coefficients of the matrix.
constraint (BinaryConstraint, BinaryConstraintTerm, BinaryConstraints, BinaryIntConstraint, BinaryIntConstraintTerm or BinaryIntConstraints) --
- Input constraint(s).

戻り値:

Solver's result class whose type is depending on the client. See Client for each client's ClientResult class for further details.

戻り値の型:

SolverResult

property sort_solution

Boolean value that specify whether the output solutions are sorted in order of best solution first. Set to True by default. If set to False, solutions are output in the order given by the Ising machine used as the driver.

Type:: bool