c7ad5ba6d05809db066b60bc3564201b66d2d51a,tf_agents/drivers/dynamic_step_driver_test.py,DynamicStepDriverTest,testMultiStepReplayBufferObservers,#DynamicStepDriverTest#,167

Before Change

    policy = driver_test_utils.TFPolicyMock(env.time_step_spec(),
                                            env.action_spec())
    policy_state = policy.get_initial_state(1)
    replay_buffer = make_replay_buffer(env)

    driver = dynamic_step_driver.DynamicStepDriver(
        env, policy, num_steps=6, observers=[replay_buffer.add_batch])

After Change

    policy = driver_test_utils.TFPolicyMock(env.time_step_spec(),
                                            env.action_spec())
    policy_state = policy.get_initial_state(1)
    replay_buffer = driver_test_utils.make_replay_buffer(policy)

    driver = dynamic_step_driver.DynamicStepDriver(
        env, policy, num_steps=6, observers=[replay_buffer.add_batch])

Instances