Merge pull request #377 from Chang-SHAO/main

upload files

pypop7/optimizers/ds/test_cs.py

@@ -0,0 +1,14 @@
+def test_optimize():
+    import numpy  # engine for numerical computing
+    from pypop7.benchmarks.base_functions import rosenbrock  # function to be minimized
+    from pypop7.optimizers.ds.cs import CS
+    problem = {'fitness_function': rosenbrock,  # to define problem arguments
+               'ndim_problem': 2,
+               'lower_boundary': -5.0 * numpy.ones((2,)),
+               'upper_boundary': 5.0 * numpy.ones((2,))}
+    options = {'max_function_evaluations': 5000,  # to set optimizer options
+               'seed_rng': 2022}
+    cs = CS(problem, options)  # to initialize the black-box optimizer class
+    results = cs.optimize()  # to run its optimization/evolution process
+    assert results['n_function_evaluations'] == 5000
+    assert results['best_so_far_y'] < 1.0

pypop7/optimizers/ds/test_gps.py

@@ -0,0 +1,14 @@
+def test_optimize():
+    import numpy  # engine for numerical computing
+    from pypop7.benchmarks.base_functions import rosenbrock  # function to be minimized
+    from pypop7.optimizers.ds.gps import GPS
+    problem = {'fitness_function': rosenbrock,  # to define problem arguments
+               'ndim_problem': 2,
+               'lower_boundary': -5.0 * numpy.ones((2,)),
+               'upper_boundary': 5.0 * numpy.ones((2,))}
+    options = {'max_function_evaluations': 5000,  # to set optimizer options
+               'seed_rng': 2022}
+    gps = GPS(problem, options)  # to initialize the black-box optimizer class
+    results = gps.optimize()  # to run its optimization/evolution process
+    assert results['n_function_evaluations'] == 5000
+    assert results['best_so_far_y'] < 10.0

pypop7/optimizers/ds/test_hj.py

@@ -0,0 +1,14 @@
+def test_optimize():
+    import numpy  # engine for numerical computing
+    from pypop7.benchmarks.base_functions import rosenbrock  # function to be minimized
+    from pypop7.optimizers.ds.hj import HJ
+    problem = {'fitness_function': rosenbrock,  # to define problem arguments
+               'ndim_problem': 2,
+               'lower_boundary': -5.0 * numpy.ones((2,)),
+               'upper_boundary': 5.0 * numpy.ones((2,))}
+    options = {'max_function_evaluations': 5000,  # to set optimizer options
+               'seed_rng': 2022}
+    hj = HJ(problem, options)  # to initialize the black-box optimizer class
+    results = hj.optimize()  # to run its optimization/evolution process
+    assert results['n_function_evaluations'] == 5000
+    assert results['best_so_far_y'] < 1.0

pypop7/optimizers/ds/test_nm.py

@@ -0,0 +1,14 @@
+def test_optimize():
+    import numpy  # engine for numerical computing
+    from pypop7.benchmarks.base_functions import rosenbrock  # function to be minimized
+    from pypop7.optimizers.ds.nm import NM
+    problem = {'fitness_function': rosenbrock,  # to define problem arguments
+               'ndim_problem': 2,
+               'lower_boundary': -5.0 * numpy.ones((2,)),
+               'upper_boundary': 5.0 * numpy.ones((2,))}
+    options = {'max_function_evaluations': 5000,  # to set optimizer options
+               'seed_rng': 2022}
+    nm = NM(problem, options)  # to initialize the black-box optimizer class
+    results = nm.optimize()  # to run its optimization/evolution process
+    assert results['n_function_evaluations'] == 5000
+    assert results['best_so_far_y'] < 1.0

pypop7/optimizers/ds/test_powel.py

@@ -0,0 +1,14 @@
+def test_optimize():
+    import numpy  # engine for numerical computing
+    from pypop7.benchmarks.base_functions import rosenbrock  # function to be minimized
+    from pypop7.optimizers.ds.powell import POWELL
+    problem = {'fitness_function': rosenbrock,  # to define problem arguments
+               'ndim_problem': 2,
+               'lower_boundary': -5.0 * numpy.ones((2,)),
+               'upper_boundary': 5.0 * numpy.ones((2,))}
+    options = {'max_function_evaluations': 5000,  # to set optimizer options
+               'seed_rng': 2022}
+    powel = POWELL(problem, options)  # to initialize the black-box optimizer class
+    results = powel.optimize()  # to run its optimization/evolution process
+    assert results['n_function_evaluations'] == 5000
+    assert results['best_so_far_y'] < 1.0