Fix StoppingCriteria

Author: Tanguy Damart

bluepyopt/deapext/CMA_MO.py

@@ -223,9 +223,9 @@ def set_fitness_parents(self, fitnesses):
         for f, ind in zip(fitnesses, self.parents):
             ind.fitness.values = f
 
-    def check_termination(self, ngen):
+    def check_termination(self, gen):
         stopping_params = {
-            "ngen": ngen,
+            "gen": gen,
             "population": self.population,
         }
 

bluepyopt/deapext/CMA_SO.py

@@ -166,9 +166,9 @@ def set_fitness(self, fitnesses):
         for f, ind in zip(fitnesses, self.population):
             ind.fitness.values = f
 
-    def check_termination(self, ngen):
+    def check_termination(self, gen):
         stopping_params = {
-            "ngen": ngen,
+            "gen": gen,
             "population": self.population,
             "centroid": self.centroid,
             "pc": self.pc,

bluepyopt/deapext/algorithms.py

@@ -154,7 +154,7 @@ def eaAlphaMuPlusLambdaCheckpoint(
         population = parents + offspring
 
         invalid_count = _evaluate_invalid_fitness(toolbox, offspring)
-        utils.history_and_hof(halloffame, history, population)
+        utils.update_history_and_hof(halloffame, history, population)
         utils.record_stats(stats, logbook, gen, population, invalid_count)
 
         # Select the next generation parents

bluepyopt/deapext/optimisationsCMA.py

@@ -174,16 +174,15 @@ def __init__(self,
         # Overwrite the bounds with -1. and 1.
         self.lbounds = numpy.full(self.problem_size, -1.)
         self.ubounds = numpy.full(self.problem_size, 1.)
+        
+        self.setup_deap()
 
         # In case initial guesses were provided, rescale them to the norm space
         if self.centroids is not None:
             self.centroids = [self.toolbox.Individual(_ind_convert_space(ind,
                                                                          self.to_norm))
                               for ind in centroids]
 
-        self.setup_deap()
-
-
     def setup_deap(self):
         """Set up optimisation"""
 
@@ -269,7 +268,7 @@ def run(self,
             numpy.random.set_state(cp["np_rndstate"])
             CMA_es = cp["CMA_es"]
             CMA_es.map_function = self.map_function
-
+            
         else:
             history = deap.tools.History()
             logbook = deap.tools.Logbook()
@@ -292,7 +291,7 @@ def run(self,
                 CMA_es.set_fitness_parents(fitness)
 
             gen = 1
-
+        
         # Run until a termination criteria is met
         while CMA_es.active:
             logger.info("Generation {}".format(gen))
@@ -339,7 +338,7 @@ def run(self,
                 CMA_es.map_function = temp_mf
 
             gen += 1
-
+            
         return pop, self.hof, logbook, history
 
 

bluepyopt/deapext/stoppingCriteria.py

@@ -26,6 +26,9 @@
 
 logger = logging.getLogger('__main__')
 
+def isclose(a, b, rel_tol=1e-09, abs_tol=0.0):
+        return abs(a-b) <= max(rel_tol * max(abs(a), abs(b)), abs_tol)
+
 
 class MaxNGen(bluepyopt.stoppingCriteria.StoppingCriteria):
     """Max ngen stopping criteria class"""
@@ -39,7 +42,6 @@ def __init__(self, max_ngen):
     def check(self, kwargs):
         """Check if the maximum number of iteration is reached"""
         gen = kwargs.get("gen")
-
         if gen > self.max_ngen:
             self.criteria_met = True
 
@@ -61,7 +63,7 @@ def __init__(self, lambda_, problem_size):
 
     def check(self, kwargs):
         """Check if the population stopped improving"""
-        ngen = kwargs.get("ngen")
+        ngen = kwargs.get("gen")
         population = kwargs.get("population")
         fitness = [ind.fitness.reduce for ind in population]
         fitness.sort()
@@ -117,7 +119,7 @@ def __init__(self, lambda_, problem_size):
     def check(self, kwargs):
         """Check if in 1/3rd of the last problem_size iterations the best and
         k'th best solutions are equal"""
-        ngen = kwargs.get("ngen")
+        ngen = kwargs.get("gen")
         population = kwargs.get("population")
 
         fitness = [ind.fitness.reduce for ind in population]
@@ -201,7 +203,7 @@ def __init__(self, problem_size):
 
     def check(self, kwargs):
         """Check if the coordinate axis std is too low"""
-        ngen = kwargs.get("ngen")
+        ngen = kwargs.get("gen")
         centroid = kwargs.get("centroid")
         sigma = kwargs.get("sigma")
         diagD = kwargs.get("diagD")

bluepyopt/tests/test_CMA/test_optimisationsCMA.py

@@ -2,7 +2,7 @@
 
 import nose.tools as nt
 
-import bluepyopt.optimisations
+import bluepyopt
 import bluepyopt.ephys.examples as examples
 
 from nose.plugins.attrib import attr
@@ -33,8 +33,7 @@ def test_optimisationsCMA_run():
         optimisation = bluepyopt.deapext.optimisationsCMA.DEAPOptimisationCMA(
                                                             evaluator=evaluator,
                                                             centroids=[x])
-
-        pop, hof, log, hist = optimisation.run(max_ngen=1)
+        pop, hof, log, hist = optimisation.run(max_ngen=2)
         raised = False
     except:
         raised = True