class TestEI(unittest.TestCase): def setUp(self): self.X = np.random.rand(10, 2) self.y = np.sinc(self.X * 10 - 5).sum(axis=1) self.model = DemoModel() self.model.train(self.X, self.y) def test_compute(self): ei = EI(self.model) X_test = np.random.rand(5, 2) a = ei.compute(X_test, derivative=False) assert a.shape[0] == X_test.shape[0] assert len(a.shape) == 1
class TestLCB(unittest.TestCase): def setUp(self): self.X = np.random.rand(10, 2) self.y = np.sinc(self.X * 10 - 5).sum(axis=1) self.model = DemoModel() self.model.train(self.X, self.y) def test_compute(self): lcb = LCB(self.model) X_test = np.random.rand(5, 2) a = lcb.compute(X_test, derivative=False) assert a.shape[0] == X_test.shape[0] assert len(a.shape) == 1 np.testing.assert_almost_equal(a, np.ones(X_test.shape[0]) * (- np.mean(self.y) + np.std(self.y)), decimal=3)
class TestLCB(unittest.TestCase): def setUp(self): self.X = np.random.rand(10, 2) self.y = np.sinc(self.X * 10 - 5).sum(axis=1) self.model = DemoModel() self.model.train(self.X, self.y) def test_compute(self): lcb = LCB(self.model) X_test = np.random.rand(5, 2) a = lcb.compute(X_test, derivative=False) assert a.shape[0] == X_test.shape[0] assert len(a.shape) == 1 np.testing.assert_almost_equal(a, np.ones(X_test.shape[0]) * (-np.mean(self.y) + np.std(self.y)), decimal=3)
def setUp(self): self.X = np.random.rand(10, 2) self.y = np.sinc(self.X * 10 - 5).sum(axis=1) self.model = DemoModel() self.model.train(self.X, self.y)