本文整理汇总了Python中rl.callbacks.ModelIntervalCheckpoint方法的典型用法代码示例。如果您正苦于以下问题：Python callbacks.ModelIntervalCheckpoint方法的具体用法？Python callbacks.ModelIntervalCheckpoint怎么用？Python callbacks.ModelIntervalCheckpoint使用的例子？那么恭喜您, 这里精选的方法代码示例或许可以为您提供帮助。您也可以进一步了解该方法所在类rl.callbacks的用法示例。
在下文中一共展示了callbacks.ModelIntervalCheckpoint方法的3个代码示例，这些例子默认根据受欢迎程度排序。您可以为喜欢或者感觉有用的代码点赞，您的评价将有助于系统推荐出更棒的Python代码示例。
示例1: build_callbacks
# 需要导入模块: from rl import callbacks [as 别名]
# 或者: from rl.callbacks import ModelIntervalCheckpoint [as 别名]
def build_callbacks(env_name):
checkpoint_weights_filename = 'dqn_' + env_name + '_weights_{step}.h5f'
log_filename = 'dqn_{}_log.json'.format(env_name)
callbacks = [ModelIntervalCheckpoint(checkpoint_weights_filename, interval=5000)]
callbacks += [FileLogger(log_filename, interval=100)]
return callbacks 示例2: build_callbacks
# 需要导入模块: from rl import callbacks [as 别名]
# 或者: from rl.callbacks import ModelIntervalCheckpoint [as 别名]
def build_callbacks(env_name):
checkpoint_weights_filename = 'dqn_' + env_name + '_weights_{step}.h5f'
log_filename = 'dqn_{}_log.json'.format(env_name)
callbacks = [ModelIntervalCheckpoint(checkpoint_weights_filename, interval=250000)]
callbacks += [FileLogger(log_filename, interval=100)]
return callbacks 示例3: training_game
# 需要导入模块: from rl import callbacks [as 别名]
# 或者: from rl.callbacks import ModelIntervalCheckpoint [as 别名]
def training_game():
env = Environment(map_name="HallucinIce", visualize=True, game_steps_per_episode=150, agent_interface_format=features.AgentInterfaceFormat(
feature_dimensions=features.Dimensions(screen=64, minimap=32)
))
input_shape = (_SIZE, _SIZE, 1)
nb_actions = _SIZE * _SIZE # Should this be an integer
model = neural_network_model(input_shape, nb_actions)
# memory : how many subsequent observations should be provided to the network?
memory = SequentialMemory(limit=5000, window_length=_WINDOW_LENGTH)
processor = SC2Proc()
### Policy
# Agent´s behaviour function. How the agent pick actions
# LinearAnnealedPolicy is a wrapper that transforms the policy into a linear incremental linear solution . Then why im not see LAP with other than not greedy ?
# EpsGreedyQPolicy is a way of selecting random actions with uniform distributions from a set of actions . Select an action that can give max or min rewards
# BolztmanQPolicy . Assumption that it follows a Boltzman distribution. gives the probability that a system will be in a certain state as a function of that state´s energy??
policy = LinearAnnealedPolicy(EpsGreedyQPolicy(), attr="eps", value_max=1, value_min=0.7, value_test=.0,
nb_steps=1e6)
# policy = (BoltzmanQPolicy( tau=1., clip= (-500,500)) #clip defined in between -500 / 500
### Agent
# Double Q-learning ( combines Q-Learning with a deep Neural Network )
# Q Learning -- Bellman equation
dqn = DQNAgent(model=model, nb_actions=nb_actions, memory=memory,
nb_steps_warmup=500, target_model_update=1e-2, policy=policy,
batch_size=150, processor=processor)
dqn.compile(Adam(lr=.001), metrics=["mae"])
## Save the parameters and upload them when needed
name = "HallucinIce"
w_file = "dqn_{}_weights.h5f".format(name)
check_w_file = "train_w" + name + "_weights.h5f"
if SAVE_MODEL:
check_w_file = "train_w" + name + "_weights_{step}.h5f"
log_file = "training_w_{}_log.json".format(name)
callbacks = [ModelIntervalCheckpoint(check_w_file, interval=1000)]
callbacks += [FileLogger(log_file, interval=100)]
if LOAD_MODEL:
dqn.load_weights(w_file)
dqn.fit(env, callbacks=callbacks, nb_steps=1e7, action_repetition=2,
log_interval=1e4, verbose=2)
dqn.save_weights(w_file, overwrite=True)
dqn.test(env, action_repetition=2, nb_episodes=30, visualize=False)