def test_dtype(self): p = policy.DiscretePolicy(1, dtype=np.float32) self.assertEqual(p.action_space.dtype, np.float32) p = policy.DiscretePolicy(gym.spaces.Discrete(1), dtype=np.uint8) self.assertEqual(p.action_space.dtype, np.uint8)
def test_int(self): p = policy.DiscretePolicy(1) self.assertEqual(p.action_space.n, 1) self.assertEqual(p.action_space.dtype, np.int64)
def test_sampler(self): with self.assertRaises(NotImplementedError): p = policy.DiscretePolicy(1) p(None)
def test_discrete(self): policy.DiscretePolicy(gym.spaces.Discrete(1))
def test_0(self): with self.assertRaises(AssertionError): policy.DiscretePolicy(0)