230503ac56636d91d53d728c247c6ea3d096eeb9,tf_agents/environments/tf_environment_test.py,TFEnvironmentTest,testTwoStepsDependenceOnTheFirst,#TFEnvironmentTest#,218

Before Change

  def testTwoStepsDependenceOnTheFirst(self):
    tf_env = TFEnvironmentMock()
    _, step_state = tf_env.current_time_step()
    self.evaluate(tf.global_variables_initializer())
    _, step_state = tf_env.step(1, step_state)
    time_step, _ = self.evaluate(tf_env.step(2, step_state))
    self.assertEqual(LAST, time_step.step_type)
    self.assertEqual(1., time_step.reward)

After Change

    tf_env = TFEnvironmentMock()
    time_step = tf_env.current_time_step()
    self.evaluate(tf.global_variables_initializer())
    with tf.control_dependencies([time_step.step_type]):
      action = tf.constant(1)
    time_step = tf_env.step(action)
    with tf.control_dependencies([time_step.step_type]):
      action = tf.constant(2)
    time_step = self.evaluate(tf_env.step(action))
    self.assertEqual(LAST, time_step.step_type)
    self.assertEqual(1., time_step.reward)
    self.assertEqual(0.0, time_step.discount)

Instances