b5ae1c6bde39f5130c975992198f1f8ee5200f9a,tf_agents/policies/q_policy_test.py,QPolicyTest,testActionWithinBounds,#QPolicyTest#,102

Before Change

    policy = q_policy.QPolicy(
        self._time_step_spec, bounded_action_spec, q_network=DummyNet())

    observations = tf.constant([[1, 2], [3, 4]], dtype=tf.float32)
    time_step = ts.restart(observations, batch_size=2)
    action_step = policy.action(time_step)
    self.assertEqual(action_step.action.shape.as_list(), [2, 1])
    self.assertEqual(action_step.action.dtype, tf.int32)
    // Initialize all variables
    self.evaluate(tf.compat.v1.global_variables_initializer())

After Change

                                                        minimum=-6,
                                                        maximum=-5)

    with self.assertRaisesRegex(ValueError, "minimum of 0"):
      q_policy.QPolicy(
          self._time_step_spec, bounded_action_spec, q_network=DummyNet())

Instances