Getting started¶

qubo-solver solves Quadratic Unconstrained Binary Optimization (QUBO) instances using either classical or quantum approaches. Here is the entire workflow in three steps: define an Instance from a QUBO matrix, pass it to a Solver, and call solve().

The example below uses the default configuration, which runs a quantum approach on a local emulator. See the next tutorial for how to configure quantum backends and classical algorithms explicitly.

In [ ]:

from __future__ import annotations

from qubosolver import (
    Instance,
    Solver,
    analysis,
    matrix,
)

# Define QUBO
Q = matrix.tensor(
    [
        [-0.2, 0.0, 1.0],
        [0.0, -1.0, 1.5],
        [1.0, 1.5, -0.1],
    ]
)
instance = Instance(Q)
# Solve it!
solver = Solver(instance)
solution = solver.solve()

print(analysis.to_dataframe([solution]))

from __future__ import annotations from qubosolver import ( Instance, Solver, analysis, matrix, ) # Define QUBO Q = matrix.tensor( [ [-0.2, 0.0, 1.0], [0.0, -1.0, 1.5], [1.0, 1.5, -0.1], ] ) instance = Instance(Q) # Solve it! solver = Solver(instance) solution = solver.solve() print(analysis.to_dataframe([solution]))

  labels bitstrings  costs  counts  probs
0      0        110   -1.2     989  0.989
1      0        001   -0.1      11  0.011

The output is a Solution, converted here to a dataframe listing each sampled bitstring, its cost, and its probability. Continue to the next tutorial, Solving a QUBO instance, for quantum backends and classical solver options.