measure_performance(1, "ac3_sudoku_problem", ac3_sudoku_problem,
"heuristic_backtracking_search", inference=csp.forward_check, with_history=True,
read_only_variables=read_only_variables)
measure_performance(1, "ac3_sudoku_problem", ac3_sudoku_problem,
"naive_cycle_cutset", with_history=True, read_only_variables=read_only_variables)
measure_performance(2, "ac3_sudoku_problem", ac3_sudoku_problem,
"min_conflicts", 100000, with_history=True, read_only_variables=read_only_variables)
measure_performance(2, "ac3_sudoku_problem", ac3_sudoku_problem,
"constraints_weighting", 10000, with_history=True, read_only_variables=read_only_variables)
general_genetic_ac3_sudoku_problem = csp.GeneralGeneticConstraintProblem(ac3_sudoku_problem, 0.1)
measure_performance(2, "general_genetic_ac3_magic_square_problem", general_genetic_ac3_sudoku_problem,
"genetic_local_search", 1000, 100, 0.1, read_only_variables=read_only_variables)
ac4_start_time = time.process_time()
ac4_is_arc_consistent = csp.ac4(ac4_sudoku_problem)
ac4_end_time = time.process_time()
if ac4_is_arc_consistent:
print()
print()
print("-" * 145)
print("using ac4 as a preprocessing stage which took", ac4_end_time - ac4_start_time, "seconds")
print("-" * 145)
measure_performance(1, "ac4_sudoku_problem", ac4_sudoku_problem,
"heuristic_backtracking_search", with_history=True, read_only_variables=read_only_variables)
measure_performance(1, "ac4_sudoku_problem", ac4_sudoku_problem,
"heuristic_backtracking_search", inference=csp.forward_check, with_history=True,
read_only_variables=read_only_variables)
measure_performance(1, "ac4_sudoku_problem", ac4_sudoku_problem,
"naive_cycle_cutset", with_history=True, read_only_variables=read_only_variables)
measure_performance(2, "ac4_sudoku_problem", ac4_sudoku_problem,
with_history=True)
measure_performance(2, "ac3_n_queens_problem", ac3_n_queens_problem,
"simulated_annealing", 100000, 0.5, 0.99999)
measure_performance(2, "ac3_n_queens_problem", ac3_n_queens_problem,
"random_restart_first_choice_hill_climbing", 100, 100,
10)
general_genetic_n_queens_problem = csp.GeneralGeneticConstraintProblem(
ac3_n_queens_problem, 0.1)
measure_performance(2, "ac3_n_queens_problem",
general_genetic_n_queens_problem,
"genetic_local_search", 100, 100, 0.1)
ac4_n_queens_problem = copy.deepcopy(n_queens_problem)
ac4_n_queens_problem.unassign_all_variables()
ac4_start_time = time.process_time()
ac4_is_arc_consistent = csp.ac4(ac4_n_queens_problem)
ac4_end_time = time.process_time()
if ac4_is_arc_consistent:
print()
print()
print("-" * 145)
print("using ac4 as a preprocessing stage which took",
ac4_end_time - ac4_start_time, "seconds")
print("-" * 145)
measure_performance(1,
"ac4_n_queens_problem",
ac4_n_queens_problem,
"backtracking_search",
with_history=True)
measure_performance(1,
"ac4_n_queens_problem",
ac3_car_assembly_problem, "simulated_annealing",
100000, 0.5, 0.99999)
measure_performance(2, "ac3_car_assembly_problem",
ac3_car_assembly_problem,
"random_restart_first_choice_hill_climbing", 10, 10,
10)
general_genetic_car_assembly_problem = csp.GeneralGeneticConstraintProblem(
ac3_car_assembly_problem, 0.1)
measure_performance(2, "ac3_car_assembly_problem",
general_genetic_car_assembly_problem,
"genetic_local_search", 100, 100, 0.1)
ac4_car_assembly_problem = copy.deepcopy(car_assembly_problem)
ac4_car_assembly_problem.unassign_all_variables()
ac4_start_time = time.process_time()
ac4_is_arc_consistent = csp.ac4(ac4_car_assembly_problem)
ac4_end_time = time.process_time()
if ac4_is_arc_consistent:
print()
print()
print("-" * 145)
print("using ac4 as a preprocessing stage which took",
ac4_end_time - ac4_start_time, "seconds")
print("-" * 145)
measure_performance(1,
"ac4_car_assembly_problem",
ac4_car_assembly_problem,
"backtracking_search",
with_history=True)
measure_performance(1,
"ac4_car_assembly_problem",
ac3_verbal_arithmetic_problem, "simulated_annealing",
100000, 0.5, 0.99999)
measure_performance(2, "ac3_verbal_arithmetic_problem",
ac3_verbal_arithmetic_problem,
"random_restart_first_choice_hill_climbing", 100, 100,
10)
general_genetic_verbal_arithmetic_problem = csp.GeneralGeneticConstraintProblem(
ac3_verbal_arithmetic_problem, 0.1)
measure_performance(2, "ac3_verbal_arithmetic_problem",
general_genetic_verbal_arithmetic_problem,
"genetic_local_search", 100, 100, 0.1)
ac4_verbal_arithmetic_problem = copy.deepcopy(verbal_arithmetic_problem)
ac4_verbal_arithmetic_problem.unassign_all_variables()
ac4_start_time = time.process_time()
ac4_is_arc_consistent = csp.ac4(ac4_verbal_arithmetic_problem)
ac4_end_time = time.process_time()
if ac4_is_arc_consistent:
print()
print()
print("-" * 145)
print("using ac4 as a preprocessing stage which took",
ac4_end_time - ac4_start_time, "seconds")
print("-" * 145)
measure_performance(1,
"ac4_verbal_arithmetic_problem",
ac4_verbal_arithmetic_problem,
"backtracking_search",
with_history=True)
measure_performance(1,
"ac4_verbal_arithmetic_problem",
ac3_map_coloring_problem, "simulated_annealing",
100000, 0.5, 0.99999)
measure_performance(2, "ac3_map_coloring_problem",
ac3_map_coloring_problem,
"random_restart_first_choice_hill_climbing", 100, 100,
10)
general_genetic_map_coloring_problem = csp.GeneralGeneticConstraintProblem(
ac3_map_coloring_problem, 0.1)
measure_performance(2, "ac3_map_coloring_problem",
general_genetic_map_coloring_problem,
"genetic_local_search", 100, 100, 0.1)
ac4_map_coloring_problem = copy.deepcopy(map_coloring_problem)
ac4_map_coloring_problem.unassign_all_variables()
ac4_start_time = time.process_time()
ac4_is_arc_consistent = csp.ac4(ac4_map_coloring_problem)
ac4_end_time = time.process_time()
if ac4_is_arc_consistent:
print()
print()
print("-" * 145)
print("using ac4 as a preprocessing stage which took",
ac4_end_time - ac4_start_time, "seconds")
print("-" * 145)
measure_performance(1,
"ac4_map_coloring_problem",
ac4_map_coloring_problem,
"backtracking_search",
with_history=True)
measure_performance(1,
"ac4_map_coloring_problem",
def test_ac4_one(self):
self.const_problem1.unassign_all_variables()
res = csp.ac4(self.const_problem1)
self.assertTrue(res)
def test_ac4_three(self):
res = csp.ac4(self.const_problem3)
self.assertTrue(res)
wanted_reduced_domains = [[1, 2], [2, 3]]
for var in self.const_problem3.get_variables():
self.assertIn(var.domain, wanted_reduced_domains)
with_history=True)
measure_performance(2, "ac3_einstein_problem", ac3_einstein_problem,
"simulated_annealing", 100000, 0.5, 0.99999)
measure_performance(2, "ac3_einstein_problem", ac3_einstein_problem,
"random_restart_first_choice_hill_climbing", 100, 100,
10)
general_genetic_ac3_einstein_problem = csp.GeneralGeneticConstraintProblem(
ac3_einstein_problem, 0.1)
measure_performance(2, "ac3_einstein_problem",
general_genetic_ac3_einstein_problem,
"genetic_local_search", 100, 100, 0.1)
ac4_einstein_problem = copy.deepcopy(einstein_problem)
ac4_einstein_problem.unassign_all_variables()
ac4_start_time = time.process_time()
ac4_is_arc_consistent = csp.ac4(ac4_einstein_problem)
ac4_end_time = time.process_time()
if ac4_is_arc_consistent:
print()
print()
print("-" * 145)
print("using ac4 as a preprocessing stage which took",
ac4_end_time - ac4_start_time, "seconds")
print("-" * 145)
measure_performance(1,
"ac4_einstein_problem",
ac4_einstein_problem,
"heuristic_backtracking_search",
with_history=True)
measure_performance(1,
"ac4_einstein_problem",
measure_performance(2, "ac3_magic_square_problem", ac3_magic_square_problem,
"min_conflicts", 100000, with_history=True)
measure_performance(2, "ac3_magic_square_problem", ac3_magic_square_problem,
"constraints_weighting", 10000, with_history=True)
measure_performance(2, "ac3_magic_square_problem", ac3_magic_square_problem,
"simulated_annealing", 100000, 0.5, 0.99999)
measure_performance(2, "ac3_magic_square_problem", ac3_magic_square_problem,
"random_restart_first_choice_hill_climbing", 100, 100, 10)
general_genetic_ac3_magic_square_problem = csp.GeneralGeneticConstraintProblem(ac3_magic_square_problem, 0.1)
measure_performance(2, "general_genetic_ac3_magic_square_problem", general_genetic_ac3_magic_square_problem,
"genetic_local_search", 1000, 100, 0.1)
ac4_magic_square_problem = copy.deepcopy(magic_square_problem)
ac4_magic_square_problem.unassign_all_variables()
ac4_start_time = time.process_time()
ac4_is_arc_consistent = csp.ac4(ac4_magic_square_problem)
ac4_end_time = time.process_time()
if ac4_is_arc_consistent:
print()
print()
print("-" * 145)
print("using ac4 as a preprocessing stage which took", ac4_end_time - ac4_start_time, "seconds")
print("-" * 145)
measure_performance(1, "ac4_magic_square_problem", ac4_magic_square_problem,
"backtracking_search", with_history=True)
measure_performance(1, "ac4_magic_square_problem", ac4_magic_square_problem,
"backtracking_search", inference=csp.forward_check, with_history=True)
measure_performance(1, "ac4_magic_square_problem", ac4_magic_square_problem,
"heuristic_backtracking_search", with_history=True)
measure_performance(1, "ac4_magic_square_problem", ac4_magic_square_problem,
"heuristic_backtracking_search", inference=csp.forward_check, with_history=True)