Merge pull request #18 from IOHprofiler/cleaning

Cleaning

.readthedocs.yml

@@ -10,4 +10,4 @@ sphinx:
 python:
   version: 3.8
   install:
-    - requirements: requirements.txt
+    - requirements: docs/requirements.txt

modcma/parameters.py

@@ -386,19 +386,23 @@ def init_adaptation_parameters(self) -> None:
 
         Examples are recombination weights and learning rates for the covariance
         matrix adapation.
+        TODO: clean the initlialization of these weights, this can be wrong in some cases
+        when mu != .5
         """
+
         if self.weights_option == "equal":
             ws = np.ones(self.lambda_) / self.lambda_
-            self.weights = np.append(ws, ws[::-1] * -1)
+            self.weights = np.append(ws[:self.mu], ws[self.mu::-1] * -1)
+
             if self.lambda_ % 2 != 0:
-                self.weights = np.append([1 / self.lambda_], self.weights)
+                self.weights = np.append([1 / self.lambda_], self.weights[:-1])
         elif self.weights_option == "1/2^lambda":
             ws = 1 / 2 ** np.arange(1, self.lambda_ + 1) + (
                 (1 / (2 ** self.lambda_)) / self.lambda_
             )
-            self.weights = np.append(ws, ws[::-1] * -1)
+            self.weights = np.append(ws[:self.mu], ws[self.mu::-1] * -1)
             if self.lambda_ % 2 != 0:
-                self.weights = np.append([1 / self.lambda_ ** 2], self.weights)
+                self.weights = np.append([1 / self.lambda_ ** 2], self.weights[:-1])
         else:
             self.weights = np.log((self.lambda_ + 1) / 2) - np.log(
                 np.arange(1, self.lambda_ + 1)
@@ -831,7 +835,7 @@ def adapt(self, used_budget: int) -> None:
             self.lambda_large *= 2
         else:
             self.budget_small -= used_previous_iteration
-
+        
         self.lambda_small = np.floor(
             self.lambda_init
             * (0.5 * self.lambda_large / self.lambda_init) ** (np.random.uniform() ** 2)

tests/test_modularcmaes.py

@@ -96,6 +96,12 @@ def test_local_restart(self):
 class TestModularCMAESSingle(unittest.TestCase):
     """Test case for ModularCMAES Object, holds custom tests."""
 
+    def test_tpa_threshold_cov_sequential(self):
+        c = modularcmaes.ModularCMAES(sum, 2,
+            threshold_convergence=True, sequential=True, 
+            step_size_adaptation='tpa', budget=10).run()
+        self.assertLess(c.parameters.fopt, 0.)
+
     def test_str_repr(self):
         """Test the output of repr and str."""
         c = modularcmaes.ModularCMAES(sum, 5)
@@ -188,7 +194,7 @@ def test_evaluate_bbob(self, mock_std):
         self.assertTrue(os.path.isdir(data_folder))
         modularcmaes.evaluate_bbob(1, 1, 1, logging=True, data_folder=data_folder)
         shutil.rmtree(data_folder)
-        modularcmaes.evaluate_bbob(1, 1, 1)
+        modularcmaes.evaluate_bbob(1, 1, 2)
 
 
 

tests/test_parameters.py

@@ -7,7 +7,7 @@
 
 import numpy as np
 
-from modcma.parameters import Parameters
+from modcma.parameters import Parameters, BIPOPParameters
 from modcma.utils import AnyOf
 from modcma.population import Population
 
@@ -177,6 +177,23 @@ def test_save_load(self):
             self.p.load(tmpfile)
         os.remove(tmpfile)
 
+    def test_fix_lambda_even(self):
+        self.p.lambda_ = 11
+        self.p.mirrored = 'mirrored pairwise'
+        self.assertEqual(self.p.lambda_, 11)
+        self.p.init_selection_parameters()
+        self.assertEqual(self.p.lambda_, 12)
+
+        for weights_option in ("equal", "1/2^lambda"):
+            self.p.weights_option = weights_option
+            self.p.lambda_ = 11
+            self.p.mu = 5
+            self.p.init_adaptation_parameters()
+            self.assertEqual(len(self.p.weights), 11)
+
+        b = BIPOPParameters(7, 20, .5)
+        b.adapt(11)
+        self.assertEqual(b.lambda_small, 8)
 
 if __name__ == "__main__":
     unittest.main()

tests/test_population.py

@@ -106,6 +106,7 @@ def test_d(self):
     def test_repr(self):
         """Test representation.""" 
         self.assertEqual(type(repr(self.pop)), str)
+        self.assertEqual(type(str(self.pop)), str)
 
 
 if __name__ == "__main__":

tests/test_sampling.py

@@ -43,6 +43,7 @@ def test_halton(self):
         """Test halton sampling."""
         sampler = sampling.halton_sampling(self._dim)
         self.is_sampler(sampler)
+        self.assertEqual(sampling.Halton.vectorized_next(10, 2), .3125)
 
     def test_orthogonal(self):
         """Test orthogonal sampling."""