BUGFIX: Exception when no feasible solution was found

Author: Julian Blank

pymoo/model/algorithm.py

@@ -3,6 +3,7 @@
 import numpy as np
 
 from pymoo.model.evaluator import Evaluator
+from pymoo.model.individual import Individual
 from pymoo.model.population import Population
 from pymoo.model.result import Result
 from pymoo.util.function_loader import FunctionLoader
@@ -47,6 +48,9 @@ def callback(algorithm):
     pf : np.array
         The Pareto-front for the given problem. If provided performance metrics are printed during execution.
 
+    return_least_infeasible : bool
+        Whether the algorithm should return the least infeasible solution, if no solution was found.
+
     evaluator : class
         The evaluator which can be used to make modifications before calling the evaluate function of a problem.
 
@@ -55,6 +59,7 @@ def callback(algorithm):
 
     def __init__(self,
                  callback=None,
+                 return_least_infeasible=False,
                  **kwargs):
 
         # !
@@ -72,6 +77,7 @@ def __init__(self,
         # other attributes of the algorithm
         self.callback = callback
         self.func_display_attrs = None
+        self.return_least_infeasible = return_least_infeasible
 
         # !
         # Attributes to be set later on for each problem run
@@ -158,7 +164,7 @@ def solve(self):
 
         # if the algorithm already set the optimum just return it, else filter it by default
         if self.opt is None:
-            opt = filter_optimum(res.pop.copy())
+            opt = filter_optimum(res.pop.copy(), least_infeasible=self.return_least_infeasible)
         else:
             opt = self.opt
 
@@ -167,11 +173,13 @@ def solve(self):
 
         if isinstance(opt, Population):
             X, F, CV, G = opt.get("X", "F", "CV", "G")
-        else:
+        elif isinstance(opt, Individual):
             X, F, CV, G = opt.X, opt.F, opt.CV, opt.G
+        else:
+            X, F, CV, G = None, None, None, None
 
-        if opt is not None:
-            res.X, res.F, res.CV, res.G = X, F, CV, G
+        # set all the individual values
+        res.X, res.F, res.CV, res.G = X, F, CV, G
 
         # create the result object
         res.problem, res.pf = self.problem, self.pf
@@ -247,25 +255,31 @@ def _finalize(self):
         pass
 
 
-def filter_optimum(pop):
+def filter_optimum(pop, least_infeasible=False):
 
     # first only choose feasible solutions
-    pop = pop[pop.collect(lambda ind: ind.feasible)[:, 0]]
+    ret = pop[pop.collect(lambda ind: ind.feasible)[:, 0]]
 
     # if at least one feasible solution was found
-    if len(pop) > 0:
+    if len(ret) > 0:
 
         # then check the objective values
-        F = pop.get("F")
+        F = ret.get("F")
 
         if F.shape[1] > 1:
-            I = NonDominatedSorting().do(pop.get("F"), only_non_dominated_front=True)
-            pop = pop[I]
+            I = NonDominatedSorting().do(ret.get("F"), only_non_dominated_front=True)
+            ret = ret[I]
 
         else:
-            pop = pop[np.argmin(F)]
+            ret = ret[np.argmin(F)]
 
+    # no feasible solution was found
     else:
-        pop = None
+        # if flag enable report the least infeasible
+        if least_infeasible:
+            ret = pop[np.argmin(pop.get("CV"))]
+        # otherwise just return none
+        else:
+            ret = None
 
-    return pop
+    return ret

pymoo/usage/algorithms/usage_custom_output.py

@@ -0,0 +1,26 @@
+from pymoo.algorithms.nsga2 import NSGA2
+from pymoo.factory import get_problem
+from pymoo.optimize import minimize
+from pymoo.visualization.scatter import Scatter
+
+
+def my_callback(algorithm):
+    disp = algorithm.func_display_attrs(algorithm.problem, algorithm.evaluator, algorithm, algorithm.pf)
+    print("My Custom Output: ", end='')
+    algorithm._display(disp)
+
+
+problem = get_problem("zdt2")
+
+algorithm = NSGA2(pop_size=100, elimate_duplicates=True, callback=my_callback)
+
+res = minimize(problem,
+               algorithm,
+               ('n_gen', 200),
+               seed=1,
+               verbose=False)
+
+plot = Scatter()
+plot.add(problem.pareto_front(), plot_type="line", color="black", alpha=0.7)
+plot.add(res.F, color="red")
+plot.show()

pymoo/usage/algorithms/usage_memetic_algorithm.py

@@ -0,0 +1,21 @@
+from pymoo.algorithms.genetic_algorithm import GeneticAlgorithm
+
+
+class MemeticAlgorithm(GeneticAlgorithm):
+
+    def _next(self):
+        # do the mating using the current population
+        self.off = self._mating(self.pop)
+
+        ################################################################
+        # Add a local optimization here
+        ################################################################
+
+        # evaluate the offspring
+        self.evaluator.eval(self.problem, self.off, algorithm=self)
+
+        # merge the offsprings with the current population
+        self.pop = self.pop.merge(self.off)
+
+        # the do survival selection
+        self.pop = self.survival.do(self.problem, self.pop, self.pop_size, algorithm=self)

tests/algorithms/test_algorithms.py

@@ -2,8 +2,11 @@
 
 import numpy as np
 
+from pymoo.algorithms.nsga2 import NSGA2
 from pymoo.algorithms.nsga2 import nsga2
 from pymoo.factory import get_problem, Problem, ZDT
+from pymoo.factory import get_sampling, get_crossover, get_mutation
+from pymoo.factory import get_termination
 from pymoo.optimize import minimize
 
 
@@ -30,6 +33,39 @@ def _evaluate(self, x, out, *args, **kwargs):
         algorithm = nsga2(pop_size=100, elimate_duplicates=True)
         minimize(ZDT1NoPF(), algorithm, ('n_gen', 20), seed=1, verbose=True)
 
+    def test_no_feasible_solution_found(self):
+        class MyProblem(Problem):
+
+            def __init__(self):
+                super().__init__(n_var=2,
+                                 n_obj=1,
+                                 n_constr=36,
+                                 xl=np.array([0, 0]),
+                                 xu=np.array([100, 100]))
+
+            def _evaluate(self, x, out, *args, **kwargs):
+                f1 = x[:, 0] + x[:, 1]
+                out["F"] = np.column_stack([f1])
+                out["G"] = np.ones(len(x))
+
+        res = minimize(MyProblem(),
+                       NSGA2(),
+                       ("n_gen", 10),
+                       seed=1,
+                       save_history=True)
+
+        self.assertEqual(res.X, None)
+        self.assertEqual(res.F, None)
+        self.assertEqual(res.G, None)
+
+        res = minimize(MyProblem(),
+                       NSGA2(return_least_infeasible=True),
+                       ("n_gen", 10),
+                       seed=1,
+                       save_history=True)
+
+        self.assertEqual(res.CV, 1)
+
 
 if __name__ == '__main__':
     unittest.main()