test_cvxpy_solver.py•9.03 kB
"""
Tests for CVXPY convex optimization solver.
"""
import pytest
import numpy as np
from returns.result import Success, Failure
from constrained_opt_mcp.models.cvxpy_models import (
CVXPYProblem, CVXPYVariable, CVXPYConstraint, CVXPYObjective,
ObjectiveType
)
from constrained_opt_mcp.solvers.cvxpy_solver import solve_cvxpy_problem
class TestCVXPYSolver:
"""Test cases for CVXPY solver functionality."""
def test_linear_programming(self):
"""Test solving linear programming problems."""
variables = [CVXPYVariable(name="x", shape=2)]
objective = CVXPYObjective(
type=ObjectiveType.MINIMIZE,
expression="cp.sum(x)"
)
constraints = [
CVXPYConstraint(expression="x >= 0"),
CVXPYConstraint(expression="cp.sum(x) == 1"),
]
problem = CVXPYProblem(
variables=variables,
objective=objective,
constraints=constraints,
description="Linear programming test"
)
result = solve_cvxpy_problem(problem)
assert isinstance(result, Success)
solution = result.unwrap()
assert solution.status == "optimal"
assert solution.objective_value is not None
assert len(solution.values["x"]) == 2
def test_quadratic_programming(self):
"""Test solving quadratic programming problems."""
variables = [CVXPYVariable(name="x", shape=2)]
objective = CVXPYObjective(
type=ObjectiveType.MINIMIZE,
expression="cp.sum_squares(x)"
)
constraints = [
CVXPYConstraint(expression="x >= 0"),
CVXPYConstraint(expression="cp.sum(x) == 1"),
]
problem = CVXPYProblem(
variables=variables,
objective=objective,
constraints=constraints,
description="Quadratic programming test"
)
result = solve_cvxpy_problem(problem)
assert isinstance(result, Success)
solution = result.unwrap()
assert solution.status == "optimal"
assert solution.objective_value is not None
def test_least_squares_optimization(self):
"""Test solving least squares optimization."""
# Create test data
A = np.array([[1.0, -0.5], [0.5, 2.0], [0.0, 1.0]])
b = np.array([2.0, 1.0, -1.0])
variables = [CVXPYVariable(name="x", shape=2)]
objective = CVXPYObjective(
type=ObjectiveType.MINIMIZE,
expression="cp.sum_squares(A @ x - b)"
)
constraints = [
CVXPYConstraint(expression="x >= 0"),
CVXPYConstraint(expression="x <= 1"),
]
problem = CVXPYProblem(
variables=variables,
objective=objective,
constraints=constraints,
parameters={"A": A, "b": b},
description="Least squares optimization test"
)
result = solve_cvxpy_problem(problem)
assert isinstance(result, Success)
solution = result.unwrap()
assert solution.status == "optimal"
assert solution.objective_value is not None
def test_maximization_problem(self):
"""Test solving maximization problems."""
variables = [CVXPYVariable(name="x", shape=2)]
objective = CVXPYObjective(
type=ObjectiveType.MAXIMIZE,
expression="cp.sum(x)"
)
constraints = [
CVXPYConstraint(expression="x >= 0"),
CVXPYConstraint(expression="cp.sum(x) <= 1"),
]
problem = CVXPYProblem(
variables=variables,
objective=objective,
constraints=constraints,
description="Maximization test"
)
result = solve_cvxpy_problem(problem)
assert isinstance(result, Success)
solution = result.unwrap()
assert solution.status == "optimal"
assert solution.objective_value is not None
def test_matrix_variables(self):
"""Test solving problems with matrix variables."""
variables = [CVXPYVariable(name="X", shape=(2, 2))]
objective = CVXPYObjective(
type=ObjectiveType.MINIMIZE,
expression="cp.trace(X)"
)
constraints = [
CVXPYConstraint(expression="X >= 0"),
CVXPYConstraint(expression="cp.trace(X) >= 1"),
]
problem = CVXPYProblem(
variables=variables,
objective=objective,
constraints=constraints,
description="Matrix variable test"
)
result = solve_cvxpy_problem(problem)
assert isinstance(result, Success)
solution = result.unwrap()
assert solution.status == "optimal"
assert solution.objective_value is not None
def test_infeasible_problem(self):
"""Test handling of infeasible problems."""
variables = [CVXPYVariable(name="x", shape=1)]
objective = CVXPYObjective(
type=ObjectiveType.MINIMIZE,
expression="x[0]"
)
constraints = [
CVXPYConstraint(expression="x >= 1"),
CVXPYConstraint(expression="x <= 0"),
]
problem = CVXPYProblem(
variables=variables,
objective=objective,
constraints=constraints,
description="Infeasible problem test"
)
result = solve_cvxpy_problem(problem)
assert isinstance(result, Success)
solution = result.unwrap()
assert solution.status == "infeasible"
def test_unbounded_problem(self):
"""Test handling of unbounded problems."""
variables = [CVXPYVariable(name="x", shape=1)]
objective = CVXPYObjective(
type=ObjectiveType.MINIMIZE,
expression="x[0]"
)
constraints = [
CVXPYConstraint(expression="x >= 0"),
]
problem = CVXPYProblem(
variables=variables,
objective=objective,
constraints=constraints,
description="Unbounded problem test"
)
result = solve_cvxpy_problem(problem)
assert isinstance(result, Success)
solution = result.unwrap()
assert solution.status == "unbounded"
def test_problem_validation(self):
"""Test problem validation."""
# Empty variables
problem = CVXPYProblem(
variables=[],
objective=CVXPYObjective(type=ObjectiveType.MINIMIZE, expression="0"),
constraints=[],
description="Invalid problem"
)
assert not problem.validate()
# Missing objective
problem = CVXPYProblem(
variables=[CVXPYVariable(name="x", shape=1)],
objective=None, # This should cause validation to fail
constraints=[],
description="Invalid problem"
)
# Note: This test might need adjustment based on actual validation logic
def test_variable_properties(self):
"""Test variable properties and constraints."""
variables = [
CVXPYVariable(
name="x",
shape=2,
nonneg=True,
description="Non-negative variable"
)
]
objective = CVXPYObjective(
type=ObjectiveType.MINIMIZE,
expression="cp.sum(x)"
)
constraints = [
CVXPYConstraint(expression="cp.sum(x) == 1"),
]
problem = CVXPYProblem(
variables=variables,
objective=objective,
constraints=constraints,
description="Variable properties test"
)
result = solve_cvxpy_problem(problem)
assert isinstance(result, Success)
solution = result.unwrap()
assert solution.status == "optimal"
# Check that solution values are non-negative
x_values = solution.values["x"]
assert all(val >= 0 for val in x_values)
def test_dual_values(self):
"""Test extraction of dual values."""
variables = [CVXPYVariable(name="x", shape=1)]
objective = CVXPYObjective(
type=ObjectiveType.MINIMIZE,
expression="x[0]"
)
constraints = [
CVXPYConstraint(expression="x >= 0"),
CVXPYConstraint(expression="x <= 1"),
]
problem = CVXPYProblem(
variables=variables,
objective=objective,
constraints=constraints,
description="Dual values test"
)
result = solve_cvxpy_problem(problem)
assert isinstance(result, Success)
solution = result.unwrap()
assert solution.status == "optimal"
# Dual values might be available depending on solver
if solution.dual_values:
assert len(solution.dual_values) == len(constraints)