def configuration_recommendation_ddpg(result_info): # pylint: disable=invalid-name
LOG.info('Use ddpg to recommend configuration')
result_id = result_info['newest_result_id']
result = Result.objects.filter(pk=result_id)
session = Result.objects.get(pk=result_id).session
agg_data = DataUtil.aggregate_data(result)
metric_data = agg_data['y_matrix'].flatten()
metric_scalar = MinMaxScaler().fit(metric_data.reshape(1, -1))
normalized_metric_data = metric_scalar.transform(metric_data.reshape(
1, -1))[0]
cleaned_knob_data = clean_knob_data(agg_data['X_matrix'],
agg_data['X_columnlabels'], session)
knob_labels = np.array(cleaned_knob_data[1]).flatten()
knob_num = len(knob_labels)
metric_num = len(metric_data)
ddpg = DDPG(n_actions=knob_num, n_states=metric_num)
if session.ddpg_actor_model is not None and session.ddpg_critic_model is not None:
ddpg.set_model(session.ddpg_actor_model, session.ddpg_critic_model)
if session.ddpg_reply_memory is not None:
ddpg.replay_memory.set(session.ddpg_reply_memory)
knob_data = ddpg.choose_action(normalized_metric_data)
knob_bounds = np.vstack(DataUtil.get_knob_bounds(knob_labels, session))
knob_data = MinMaxScaler().fit(knob_bounds).inverse_transform(
knob_data.reshape(1, -1))[0]
conf_map = {k: knob_data[i] for i, k in enumerate(knob_labels)}
conf_map_res = {}
conf_map_res['status'] = 'good'
conf_map_res['result_id'] = result_id
conf_map_res['recommendation'] = conf_map
conf_map_res['info'] = 'INFO: ddpg'
return conf_map_res
def configuration_recommendation_ddpg(result_info): # pylint: disable=invalid-name
LOG.info('Use ddpg to recommend configuration')
result_id = result_info['newest_result_id']
result = Result.objects.filter(pk=result_id)
session = Result.objects.get(pk=result_id).session
agg_data = DataUtil.aggregate_data(result)
metric_data = agg_data['y_matrix'].flatten()
cleaned_agg_data = clean_knob_data(agg_data['X_matrix'], agg_data['X_columnlabels'],
session)
knob_labels = np.array(cleaned_agg_data[1]).flatten()
knob_num = len(knob_labels)
metric_num = len(metric_data)
ddpg = DDPG(n_actions=knob_num, n_states=metric_num, alr=ACTOR_LEARNING_RATE,
clr=CRITIC_LEARNING_RATE, gamma=GAMMA, batch_size=DDPG_BATCH_SIZE, tau=TAU)
if session.ddpg_actor_model is not None and session.ddpg_critic_model is not None:
ddpg.set_model(session.ddpg_actor_model, session.ddpg_critic_model)
if session.ddpg_reply_memory is not None:
ddpg.replay_memory.set(session.ddpg_reply_memory)
knob_data = ddpg.choose_action(metric_data)
LOG.info('recommended knob: %s', knob_data)
knob_bounds = np.vstack(DataUtil.get_knob_bounds(knob_labels, session))
knob_data = MinMaxScaler().fit(knob_bounds).inverse_transform(knob_data.reshape(1, -1))[0]
conf_map = {k: knob_data[i] for i, k in enumerate(knob_labels)}
conf_map_res = {}
conf_map_res['status'] = 'good'
conf_map_res['result_id'] = result_id
conf_map_res['recommendation'] = conf_map
conf_map_res['info'] = 'INFO: ddpg'
for k in knob_labels:
LOG.info('%s: %f', k, conf_map[k])
return conf_map_res
def ddpg(env, config, n_loops=100):
results = []
x_axis = []
num_collections = config['num_collections']
gamma = config['gamma']
a_lr = config['a_lr']
c_lr = config['c_lr']
n_epochs = config['n_epochs']
ahs = config['a_hidden_sizes']
chs = config['c_hidden_sizes']
model_ddpg = DDPG(n_actions=env.knob_dim,
n_states=env.metric_dim,
gamma=gamma,
clr=c_lr,
alr=a_lr,
shift=0,
a_hidden_sizes=ahs,
c_hidden_sizes=chs)
knob_data = np.random.rand(env.knob_dim)
prev_metric_data = np.zeros(env.metric_dim)
for i in range(num_collections):
action = np.random.rand(env.knob_dim)
reward, metric_data = env.simulate(action)
if i > 0:
model_ddpg.add_sample(prev_metric_data, prev_knob_data,
prev_reward, metric_data)
prev_metric_data = metric_data
prev_knob_data = knob_data
prev_reward = reward
for i in range(n_loops):
reward, metric_data = env.simulate(knob_data)
model_ddpg.add_sample(prev_metric_data, prev_knob_data, prev_reward,
prev_metric_data)
prev_metric_data = metric_data
prev_knob_data = knob_data
prev_reward = reward
for _ in range(n_epochs):
model_ddpg.update()
results.append(reward)
x_axis.append(i + 1)
LOG.info('loop: %d reward: %f', i, reward[0])
knob_data = model_ddpg.choose_action(metric_data)
return np.array(results), np.array(x_axis)
def configuration_recommendation_ddpg(result_info): # pylint: disable=invalid-name
start_ts = time.time()
LOG.info('Use ddpg to recommend configuration')
result_id = result_info['newest_result_id']
result_list = Result.objects.filter(pk=result_id)
result = result_list.first()
session = result.session
params = JSONUtil.loads(session.hyperparameters)
agg_data = DataUtil.aggregate_data(result_list)
metric_data, _ = clean_metric_data(agg_data['y_matrix'],
agg_data['y_columnlabels'], session)
metric_data = metric_data.flatten()
metric_scalar = MinMaxScaler().fit(metric_data.reshape(1, -1))
normalized_metric_data = metric_scalar.transform(metric_data.reshape(
1, -1))[0]
cleaned_knob_data = clean_knob_data(agg_data['X_matrix'],
agg_data['X_columnlabels'], session)
knob_labels = np.array(cleaned_knob_data[1]).flatten()
knob_num = len(knob_labels)
metric_num = len(metric_data)
ddpg = DDPG(n_actions=knob_num,
n_states=metric_num,
a_hidden_sizes=params['DDPG_ACTOR_HIDDEN_SIZES'],
c_hidden_sizes=params['DDPG_CRITIC_HIDDEN_SIZES'],
use_default=params['DDPG_USE_DEFAULT'])
if session.ddpg_actor_model is not None and session.ddpg_critic_model is not None:
ddpg.set_model(session.ddpg_actor_model, session.ddpg_critic_model)
if session.ddpg_reply_memory is not None:
ddpg.replay_memory.set(session.ddpg_reply_memory)
knob_data = ddpg.choose_action(normalized_metric_data)
knob_bounds = np.vstack(DataUtil.get_knob_bounds(knob_labels, session))
knob_data = MinMaxScaler().fit(knob_bounds).inverse_transform(
knob_data.reshape(1, -1))[0]
conf_map = {k: knob_data[i] for i, k in enumerate(knob_labels)}
conf_map_res = create_and_save_recommendation(recommended_knobs=conf_map,
result=result,
status='good',
info='INFO: ddpg')
save_execution_time(start_ts, "configuration_recommendation_ddpg", result)
return conf_map_res
