Add a test for the 'BalancingLearner' with various learners, closes #102'

adaptive/tests/test_learners.py

@@ -123,6 +123,8 @@ def ask_randomly(learner, rounds, points):
     return xs, ls
 
 
+# Tests
+
 @run_with(Learner1D)
 def test_uniform_sampling1D(learner_type, f, learner_kwargs):
     """Points are sampled uniformly if no data is provided.
@@ -349,6 +351,37 @@ def test_learner_performance_is_invariant_under_scaling(learner_type, f, learner
     assert abs(learner.loss() - control.loss()) / learner.loss() < 1e-11
 
 
+# XXX: the LearnerND currently fails because there is no `add_data=False` argument in ask.
+@run_with(Learner1D, Learner2D, xfail(LearnerND), AverageLearner)
+def test_balancing_learner(learner_type, f, learner_kwargs):
+    """Test if the BalancingLearner works with the different types of learners."""
+    learners = [learner_type(generate_random_parametrization(f), **learner_kwargs)
+                for i in range(5)]
+
+    learner = BalancingLearner(learners)
+
+    # Emulate parallel execution
+    stash = []
+
+    for i in range(200):
+        xs, _ = learner.ask(10)
+
+        # Save 5 random points out of `xs` for later
+        random.shuffle(xs)
+        for _ in range(5):
+            stash.append(xs.pop())
+
+        for x in xs:
+            learner.tell(x, learner.function(x))
+
+        # Evaluate and add 5 random points from `stash`
+        random.shuffle(stash)
+        for _ in range(5):
+            learner.tell(stash.pop(), learner.function(x))
+
+    assert all(l.npoints > 20 for l in learner.learners)
+
+
 @pytest.mark.xfail
 @run_with(Learner1D, Learner2D, LearnerND)
 def test_convergence_for_arbitrary_ordering(learner_type, f, learner_kwargs):