Index

A

A2CAgent (class in core.algorithm.A2C) AbstractMA_DQNAgent (class in core.algorithm.MADQN) action (core.common.memory.Experience attribute) ACTION_REPETITION (core.common.util.OPS attribute) actor_optimizer() (core.algorithm.A2C.A2CAgent method) adapt() (core.common.random.AdaptiveParamNoiseSpec method) AdaptiveParamNoiseSpec (class in core.common.random) add() (core.common.cartpole_dqn_PER.PrioritizedReplayBuffer method) (core.common.cartpole_dqn_PER.ReplayBuffer method) AdditionalUpdatesOptimizer (class in core.common.util)

AdvEpsGreedyPolicy (class in core.policies) Agent (class in core.common.agent) AnnealedGaussianProcess (class in core.common.random) append() (core.common.memory.Memory method) (core.common.memory.RingBuffer method) (core.memories.SequentialMemory method) append_replay_memory() (core.algorithm.DQN.DQNAgent method) (core.common.agent.Agent method) append_sample() (core.common.cartpole_dqn_PER.DQNAgent method) arg_max() (core.algorithm.QLearning.QLearningAgent static method) auto_executor() (in module core.common.util)

B

backward() (core.algorithm.A2C.A2CAgent method) (core.algorithm.DDPG.DDPGAgent method) (core.algorithm.DQN.DQNAgent method) (core.algorithm.Deep_sarsa.DeepSARSAgent method) (core.algorithm.MADDPG.MA_DDPGAgent method) (core.algorithm.MADQN.MA_DQNAgent method) (core.algorithm.PPO.PPOAgent method) (core.algorithm.REINFORCE.ReinforceAgent method) (core.common.agent.Agent method)

BATCH_SIZE (core.common.util.OPS attribute) BoltzmannGumbelQPolicy (class in core.policies) BoltzmannQPolicy (class in core.policies) BUFFER_SIZE (core.common.util.OPS attribute) build() (core.common.util.PopArtLayer method) build_model() (core.common.cartpole_dqn_PER.DQNAgent method)

C

call() (core.common.util.PopArtLayer method) Callback (class in core.common.callback) CallbackList (class in core.common.callback) choose_action() (core.algorithm.QLearning.QLearningAgent method) clipped_masked_error() (in module core.common.util) cLogger (class in core.common.util) clone_model() (in module core.common.util) clone_optimizer() (in module core.common.util) compile() (core.algorithm.A2C.A2CAgent method) (core.algorithm.DDPG.DDPGAgent method) (core.algorithm.DQN.DQNAgent method) (core.algorithm.Deep_sarsa.DeepSARSAgent method) (core.algorithm.MADDPG.MA_DDPGAgent method) (core.algorithm.MADQN.MA_DQNAgent method) (core.algorithm.PPO.PPOAgent method) (core.algorithm.REINFORCE.ReinforceAgent method) (core.common.agent.Agent method) compute_batch_q_values() (core.algorithm.MADQN.AbstractMA_DQNAgent method) compute_output_shape() (core.common.util.PopArtLayer method) compute_q_values() (core.algorithm.MADQN.AbstractMA_DQNAgent method) core (module) core.algorithm (module)

core.algorithm.A2C (module) core.algorithm.DDPG (module) core.algorithm.Deep_sarsa (module) core.algorithm.DQN (module) core.algorithm.MADDPG (module) core.algorithm.MADQN (module) core.algorithm.PPO (module) core.algorithm.QLearning (module) core.algorithm.REINFORCE (module) core.callbacks (module) core.common (module) core.common.agent (module) core.common.callback (module) core.common.cartpole_dqn_PER (module) core.common.memory (module) core.common.policy (module) core.common.processor (module) core.common.random (module) core.common.util (module) core.memories (module) core.policies (module) critic_optimizer() (core.algorithm.A2C.A2CAgent method) current_sigma (core.common.random.AnnealedGaussianProcess attribute)

D

ddpg_distance_metric() (in module core.algorithm.DDPG) (in module core.common.random) DDPGAgent (class in core.algorithm.DDPG) de_normalize() (core.common.util.PopArtLayer method) DeepSARSAgent (class in core.algorithm.Deep_sarsa) denormalize() (in module core.common.util) DISCOUNT_FACTOR (core.common.util.OPS attribute) discount_rewards() (core.algorithm.REINFORCE.ReinforceAgent method)

discounted_reward() (core.algorithm.PPO.PPOAgent method) display_param_dic() (in module core.common.util) display_param_list() (in module core.common.util) DOUBLE (core.common.util.OPS attribute) DQNAgent (class in core.algorithm.DQN) (class in core.common.cartpole_dqn_PER) DrawTrainMovingAvgPlotCallback (class in core.callbacks) DrawTrainPlotCallback (class in core.callbacks) DUELING (core.common.util.OPS attribute)

E

ENTROPY_LOSS (core.common.util.OPS attribute) EPOCHS (core.common.util.OPS attribute)

EpsGreedyQPolicy (class in core.policies) Experience (class in core.common.memory)

F

FileLogger (class in core.callbacks) find_prefixsum_idx() (core.common.cartpole_dqn_PER.SumSegmentTree method) (core.memories.SumSegmentTree method) forward() (core.algorithm.A2C.A2CAgent method) (core.algorithm.DDPG.DDPGAgent method) (core.algorithm.DQN.DQNAgent method) (core.algorithm.Deep_sarsa.DeepSARSAgent method) (core.algorithm.MADDPG.MA_DDPGAgent method) (core.algorithm.MADQN.MA_DQNAgent method) (core.algorithm.PPO.PPOAgent method) (core.algorithm.REINFORCE.ReinforceAgent method) (core.common.agent.Agent method)

FRAMES_PER_STEP (core.common.util.OPS attribute)

G

GAMMA (core.common.util.OPS attribute) GaussianWhiteNoiseProcess (class in core.common.random) gen_agent_params() (in module core.common.util) get_action() (core.common.cartpole_dqn_PER.DQNAgent method) get_config() (core.algorithm.MADQN.AbstractMA_DQNAgent method) (core.algorithm.MADQN.MA_DQNAgent method) (core.common.memory.Memory method) (core.common.policy.Policy method) (core.common.util.AdditionalUpdatesOptimizer method) (core.memories.SequentialMemory method) (core.policies.BoltzmannGumbelQPolicy method) (core.policies.BoltzmannQPolicy method) (core.policies.EpsGreedyQPolicy method) (core.policies.LinearAnnealedPolicy method) (core.policies.MA_BoltzmannQPolicy method) (core.policies.MA_EpsGreedyQPolicy method) (core.policies.MA_MaxBoltzmannQPolicy method) (core.policies.MaxBoltzmannQPolicy method) (core.policies.starcraft_multiagent_eGreedyPolicy method)

get_current_value() (core.policies.AdvEpsGreedyPolicy method) (core.policies.LinearAnnealedPolicy method) get_kv_from_agent() (in module core.common.util) get_logger() (in module core.common.util) get_recent_state() (core.common.memory.Memory method) get_soft_target_model_updates() (in module core.common.util) get_stats() (core.common.random.AdaptiveParamNoiseSpec method) get_updates() (core.common.util.AdditionalUpdatesOptimizer method) getLogger() (core.common.util.cLogger static method) gradients() (in module core.common.util) GreedyQPolicy (class in core.policies) gumbel_softmax() (in module core.common.util) gumbel_softmax_sample() (in module core.common.util)

H

hard_update() (in module core.algorithm.DDPG)

History (class in core.callbacks) huber_loss() (in module core.common.util)

L

layers (core.algorithm.DDPG.DDPGAgent attribute) (core.algorithm.MADDPG.MA_DDPGAgent attribute) (core.algorithm.MADQN.MA_DQNAgent attribute) (core.common.agent.Agent attribute) learn() (core.algorithm.QLearning.QLearningAgent method) LEARNING_ACTOR_RATE (core.common.util.OPS attribute) LEARNING_CRITIC_RATE (core.common.util.OPS attribute) LEARNING_RATE (core.common.util.OPS attribute) length() (core.common.memory.RingBuffer method) LinearAnnealedPolicy (class in core.policies)

load_sample() (in module core.common.cartpole_dqn_PER) load_weights() (core.algorithm.A2C.A2CAgent method) (core.algorithm.DDPG.DDPGAgent method) (core.algorithm.DQN.DQNAgent method) (core.algorithm.Deep_sarsa.DeepSARSAgent method) (core.algorithm.MADDPG.MA_DDPGAgent method) (core.algorithm.MADQN.MA_DQNAgent method) (core.algorithm.PPO.PPOAgent method) (core.algorithm.REINFORCE.ReinforceAgent method) (core.common.agent.Agent method)

M

MA_BoltzmannQPolicy (class in core.policies) MA_DDPGAgent (class in core.algorithm.MADDPG) MA_DQNAgent (class in core.algorithm.MADQN) MA_EpsGreedyQPolicy (class in core.policies) MA_GreedyQPolicy (class in core.policies) MA_MaxBoltzmannQPolicy (class in core.policies) MARGINAL_SPACE (core.common.util.OPS attribute) MaxBoltzmannQPolicy (class in core.policies) mean_q() (in module core.algorithm.MADDPG) (in module core.algorithm.MADQN) Memory (class in core.common.memory) metrics (core.common.policy.Policy attribute) (core.common.processor.Processor attribute) (core.policies.LinearAnnealedPolicy attribute)

metrics_names (core.algorithm.MADDPG.MA_DDPGAgent attribute) (core.algorithm.MADQN.MA_DQNAgent attribute) (core.common.policy.Policy attribute) (core.common.processor.Processor attribute) (core.policies.LinearAnnealedPolicy attribute) min() (core.common.cartpole_dqn_PER.MinSegmentTree method) (core.memories.MinSegmentTree method) MinSegmentTree (class in core.common.cartpole_dqn_PER) (class in core.memories) mmdd24hhmmss() (in module core.common.util) ModelIntervalCheckpoint (class in core.callbacks) MOVE_ANG (core.common.util.OPS attribute) MOVE_DIST (core.common.util.OPS attribute)

N

N_STEPS (core.common.util.OPS attribute) nb_entries (core.memories.SequentialMemory attribute)

NO_GUI (core.common.util.OPS attribute) NoisePolicy (class in core.policies) normalize() (in module core.common.util)

O

on_action_begin() (core.common.callback.Callback method) (core.common.callback.CallbackList method) on_action_end() (core.common.callback.Callback method) (core.common.callback.CallbackList method) on_episode_begin() (core.callbacks.FileLogger method) (core.callbacks.TrainEpisodeLogger method) (core.common.callback.Callback method) (core.common.callback.CallbackList method) on_episode_end() (core.callbacks.DrawTrainMovingAvgPlotCallback method) (core.callbacks.DrawTrainPlotCallback method) (core.callbacks.FileLogger method) (core.callbacks.ModelIntervalCheckpoint method) (core.callbacks.TestLogger method) (core.callbacks.TrainEpisodeLogger method) (core.callbacks.TrainIntervalLogger method) (core.common.callback.Callback method) (core.common.callback.CallbackList method) (core.common.policy.Policy method) (core.policies.AdvEpsGreedyPolicy method) on_epoch_end() (core.callbacks.History method) on_step_begin() (core.callbacks.TrainIntervalLogger method) (core.common.callback.Callback method) (core.common.callback.CallbackList method)

on_step_end() (core.callbacks.FileLogger method) (core.callbacks.ModelIntervalCheckpoint method) (core.callbacks.TrainEpisodeLogger method) (core.callbacks.TrainIntervalLogger method) (core.common.callback.Callback method) (core.common.callback.CallbackList method) on_train_begin() (core.callbacks.FileLogger method) (core.callbacks.History method) (core.callbacks.TestLogger method) (core.callbacks.TrainEpisodeLogger method) (core.callbacks.TrainIntervalLogger method) on_train_end() (core.callbacks.FileLogger method) (core.callbacks.TrainEpisodeLogger method) (core.callbacks.TrainIntervalLogger method) OPS (class in core.common.util) optimizer() (core.common.cartpole_dqn_PER.DQNAgent method) OrnsteinUhlenbeckProcess (class in core.common.random) OU_SIGMA (core.common.util.OPS attribute) OU_THETA (core.common.util.OPS attribute)

P

param (module) PER (core.common.util.OPS attribute) pickle_to_plot() (in module core.common.util) Policy (class in core.common.policy) policy (core.algorithm.DQN.DQNAgent attribute) (core.algorithm.MADQN.MA_DQNAgent attribute) POLICY (core.common.util.OPS attribute) pop_art_update() (core.common.util.PopArtLayer method) PopArtLayer (class in core.common.util) PPOAgent (class in core.algorithm.PPO) PrioritizedReplayBuffer (class in core.common.cartpole_dqn_PER) process_action() (core.common.processor.Processor method)

process_info() (core.common.processor.Processor method) process_observation() (core.common.processor.Processor method) process_reward() (core.common.processor.Processor method) process_state_batch() (core.algorithm.DDPG.DDPGAgent method) (core.algorithm.DQN.DQNAgent method) (core.algorithm.MADDPG.MA_DDPGAgent method) (core.algorithm.MADQN.AbstractMA_DQNAgent method) (core.common.processor.Processor method) process_step() (core.common.processor.Processor method) Processor (class in core.common.processor) proximal_policy_optimization_loss() (core.algorithm.PPO.PPOAgent method) proximal_policy_optimization_loss_continuous() (core.algorithm.PPO.PPOAgent method) PURE_ACTION_RATIO (core.common.util.OPS attribute)

Q

QLearningAgent (class in core.algorithm.QLearning)

R

RandomProcess (class in core.common.random) reduce() (core.common.cartpole_dqn_PER.SegmentTree method) (core.memories.SegmentTree method) ReinforceAgent (class in core.algorithm.REINFORCE) REPLAY_MEMORY_SIZE (core.common.util.OPS attribute) ReplayBuffer (class in core.common.cartpole_dqn_PER) reset() (core.callbacks.TrainIntervalLogger method) reset_env() (core.algorithm.PPO.PPOAgent method) reset_states() (core.algorithm.DDPG.DDPGAgent method) (core.algorithm.DQN.DQNAgent method) (core.algorithm.MADDPG.MA_DDPGAgent method) (core.algorithm.MADQN.MA_DQNAgent method) (core.common.agent.Agent method) (core.common.policy.Policy method) (core.common.random.OrnsteinUhlenbeckProcess method) (core.common.random.RandomProcess method) (core.common.random.SimpleOUNoise method) (core.policies.NoisePolicy method)

reward (core.common.memory.Experience attribute) REWARD_HEIGHT_RANK_WEIGHT (core.common.util.OPS attribute) reward_moving_avg_plot() (in module core.common.util) reward_quantile_plot() (in module core.common.util) REWARD_SCALE (core.common.util.OPS attribute) REWARD_VERSION (core.common.util.OPS attribute) RingBuffer (class in core.common.memory) run() (core.common.agent.Agent method) RunningMeanStd (class in core.common.util)

S

sample() (core.common.cartpole_dqn_PER.PrioritizedReplayBuffer method) (core.common.cartpole_dqn_PER.ReplayBuffer method) (core.common.memory.Memory method) (core.common.random.GaussianWhiteNoiseProcess method) (core.common.random.OrnsteinUhlenbeckProcess method) (core.common.random.SimpleOUNoise method) (core.memories.SequentialMemory method) sample_batch_indexes() (in module core.common.memory) sample_gumbel() (in module core.common.util) save_ci_graph() (in module core.common.util) save_ci_graph_from_tuple() (in module core.common.util) save_data() (core.callbacks.FileLogger method) save_plot() (in module core.common.util) save_sample() (in module core.common.cartpole_dqn_PER) save_weights() (core.algorithm.A2C.A2CAgent method) (core.algorithm.DDPG.DDPGAgent method) (core.algorithm.DQN.DQNAgent method) (core.algorithm.Deep_sarsa.DeepSARSAgent method) (core.algorithm.MADDPG.MA_DDPGAgent method) (core.algorithm.MADQN.MA_DQNAgent method) (core.algorithm.PPO.PPOAgent method) (core.algorithm.REINFORCE.ReinforceAgent method) (core.common.agent.Agent method) SegmentTree (class in core.common.cartpole_dqn_PER) (class in core.memories)

select_action() (core.algorithm.MADDPG.MA_DDPGAgent method) (core.common.policy.Policy method) (core.policies.BoltzmannGumbelQPolicy method) (core.policies.BoltzmannQPolicy method) (core.policies.EpsGreedyQPolicy method) (core.policies.GreedyQPolicy method) (core.policies.LinearAnnealedPolicy method) (core.policies.MA_BoltzmannQPolicy method) (core.policies.MA_EpsGreedyQPolicy method) (core.policies.MA_GreedyQPolicy method) (core.policies.MA_MaxBoltzmannQPolicy method) (core.policies.MaxBoltzmannQPolicy method) (core.policies.NoisePolicy method) (core.policies.starcraft_multiagent_eGreedyPolicy method) select_action_agent() (core.policies.MA_BoltzmannQPolicy method) (core.policies.MA_MaxBoltzmannQPolicy method) SequentialMemory (class in core.memories) SimpleOUNoise (class in core.common.random) smoothL1() (in module core.common.util) starcraft_multiagent_eGreedyPolicy (class in core.policies) state0 (core.common.memory.Experience attribute) state1 (core.common.memory.Experience attribute) sum() (core.common.cartpole_dqn_PER.SumSegmentTree method) (core.memories.SumSegmentTree method) SumSegmentTree (class in core.common.cartpole_dqn_PER) (class in core.memories)

T

TARGET_NETWORK_UPDATE_INTERVAL (core.common.util.OPS attribute) terminal1 (core.common.memory.Experience attribute) test_policy (core.algorithm.DQN.DQNAgent attribute) (core.algorithm.MADQN.MA_DQNAgent attribute) TestLogger (class in core.callbacks)

TIME_PENALTY_WEIGHT (core.common.util.OPS attribute) TIME_WINDOW (core.common.util.OPS attribute) train_model() (core.common.cartpole_dqn_PER.DQNAgent method) TrainEpisodeLogger (class in core.callbacks) TrainIntervalLogger (class in core.callbacks)

U

update() (core.common.util.RunningMeanStd method) update_and_normalize() (core.common.util.PopArtLayer method) update_priorities() (core.common.cartpole_dqn_PER.PrioritizedReplayBuffer method) (core.memories.SequentialMemory method) update_target_model() (core.common.cartpole_dqn_PER.DQNAgent method) update_target_model_hard() (core.algorithm.DDPG.DDPGAgent method) (core.algorithm.DQN.DQNAgent method) (core.algorithm.MADQN.MA_DQNAgent method)

update_target_models_hard() (core.algorithm.MADDPG.MA_DDPGAgent method) USE_PARAMETERIZED_NOISE (core.common.util.OPS attribute) uses_learning_phase (core.algorithm.DDPG.DDPGAgent attribute) (core.algorithm.MADDPG.MA_DDPGAgent attribute)

W

WINDOW_LENGTH (core.common.util.OPS attribute)

Y

yyyymmdd24hhmmss() (in module core.common.util)

Z

zeroed_observation() (in module core.common.memory)