Quick start

Get started by defining a problem instance.

Installation

First, install NPyComp using pip, or by following the alternative installation methods in the Installation guide.

pip install npycomp

Basic Usage

Let’s walk through a simple example of using NPyComp to define and solve a 3-SAT problem.

from npycomp.problems import SAT

# Define a 3-SAT problem
sat = SAT(clauses=[
    ["a", "b", "c"],
    ["~a", "~b", "c"],
    ["a", "~b", "~c"],
])

# Solve the problem
solution = sat.solve()

We can also reduce the 3-SAT problem to a Clique problem:

from npycomp.problems import Clique

# Reduce the SAT problem to a Clique problem
clique = sat.reduce(Clique)

# Solve the resulting Clique problem
solution = clique.solve()

The full list of problems supported by NPyComp can be found in the Problems section of the API Reference. Each problem can be reduced to any other problem in the library using the reduce() method.

Using DIMACS files

NPyComp supports loading problem instances from DIMACS format files. For more information on the DIMACS format, see the MaxSat 2023 Evaluation Rules.

# Load a SAT instance
sat = SAT.from_dimacs("path/to/problem.cnf")

# Load a Clique instance (requires specifying k)
clique = Clique.from_dimacs("path/to/graph.clq", k=5)