Index

A

ActionAdapter (class in rlgraph.components.action_adapters.action_adapter) ActorComponent (class in rlgraph.components.neural_networks.actor_component) AdadeltaOptimizer (class in rlgraph.components.optimizers.local_optimizers) AdagradOptimizer (class in rlgraph.components.optimizers.local_optimizers) AdamOptimizer (class in rlgraph.components.optimizers.local_optimizers)

add_components() (rlgraph.components.component.Component method) Agent (class in rlgraph.agents.agent) ApexAgent (class in rlgraph.agents.apex_agent) apply() (rlgraph.components.layers.nn.lstm_layer.LSTMLayer method) assign_variable() (rlgraph.components.component.Component static method)

B

BaselineActionAdapter (class in rlgraph.components.action_adapters.baseline_action_adapter) BatchSplitter (class in rlgraph.components.common.batch_splitter) Bernoulli (class in rlgraph.components.distributions.bernoulli) Beta (class in rlgraph.components.distributions.beta)

BoolBox (class in rlgraph.spaces.bool_box) bounds (rlgraph.spaces.box_space.BoxSpace attribute) BoxSpace (class in rlgraph.spaces.box_space) build() (rlgraph.agents.agent.Agent method)

C

call_api_method() (rlgraph.agents.agent.Agent method) call_count (rlgraph.components.component.Component attribute) call_times (rlgraph.components.component.Component attribute) Categorical (class in rlgraph.components.distributions.categorical) check_input_completeness() (rlgraph.components.common.synchronizable.Synchronizable method), [1] (rlgraph.components.component.Component method) check_input_spaces() (rlgraph.components.action_adapters.action_adapter.ActionAdapter method) (rlgraph.components.action_adapters.baseline_action_adapter.BaselineActionAdapter method) (rlgraph.components.common.container_splitter.ContainerSplitter method) (rlgraph.components.common.dict_merger.DictMerger method) (rlgraph.components.component.Component method) (rlgraph.components.distributions.beta.Beta method) (rlgraph.components.distributions.distribution.Distribution method) (rlgraph.components.distributions.normal.Normal method) (rlgraph.components.explorations.epsilon_exploration.EpsilonExploration method) (rlgraph.components.explorations.exploration.Exploration method) (rlgraph.components.helpers.VTraceFunction method) (rlgraph.components.layers.nn.concat_layer.ConcatLayer method) (rlgraph.components.layers.nn.lstm_layer.LSTMLayer method) (rlgraph.components.layers.nn.nn_layer.NNLayer method) (rlgraph.components.layers.strings.string_layer.StringLayer method) (rlgraph.components.loss_functions.dqn_loss_function.DQNLossFunction method) (rlgraph.components.loss_functions.impala_loss_function.IMPALALossFunction method) check_space_equivalence() (in module rlgraph.spaces.space_utils) check_variable_completeness() (rlgraph.components.component.Component method)

Component (class in rlgraph.components.component) ConcatLayer (class in rlgraph.components.layers.nn.concat_layer) ContainerSpace (class in rlgraph.spaces.containers) ContainerSplitter (class in rlgraph.components.common.container_splitter) contains() (rlgraph.spaces.bool_box.BoolBox method) (rlgraph.spaces.box_space.BoxSpace method) (rlgraph.spaces.containers.Dict method) (rlgraph.spaces.containers.Tuple method) (rlgraph.spaces.int_box.IntBox method) (rlgraph.spaces.space.Space method) (rlgraph.spaces.text_box.TextBox method) Conv2DLayer (class in rlgraph.components.layers.nn.conv2d_layer) copy() (rlgraph.components.component.Component method) create_summary() (rlgraph.components.component.Component method) create_variables() (rlgraph.components.common.container_splitter.ContainerSplitter method) (rlgraph.components.common.environment_stepper.EnvironmentStepper method) (rlgraph.components.common.staging_area.StagingArea method) (rlgraph.components.component.Component method) (rlgraph.components.layers.nn.conv2d_layer.Conv2DLayer method) (rlgraph.components.layers.nn.dense_layer.DenseLayer method) (rlgraph.components.layers.nn.lstm_layer.LSTMLayer method) (rlgraph.components.memories.fifo_queue.FIFOQueue method) (rlgraph.components.memories.memory.Memory method) (rlgraph.components.memories.prioritized_replay.PrioritizedReplay method) (rlgraph.components.memories.replay_memory.ReplayMemory method)

D

default_environment_spec (rlgraph.agents.impala_agent.IMPALAAgent attribute) default_internal_states_space (rlgraph.agents.impala_agent.IMPALAAgent attribute) define_api_methods() (rlgraph.agents.agent.Agent method) (rlgraph.agents.dqn_agent.DQNAgent method) (rlgraph.agents.impala_agent.IMPALAAgent method) define_api_methods_actor() (rlgraph.agents.impala_agent.IMPALAAgent method) define_api_methods_learner() (rlgraph.agents.impala_agent.IMPALAAgent method) define_api_methods_single() (rlgraph.agents.impala_agent.IMPALAAgent method) DenseLayer (class in rlgraph.components.layers.nn.dense_layer) Dict (class in rlgraph.spaces.containers)

DictMerger (class in rlgraph.components.common.dict_merger) DictPreprocessorStack (class in rlgraph.components.neural_networks.dict_preprocessor_stack) Distribution (class in rlgraph.components.distributions.distribution) do_explore() (rlgraph.components.explorations.epsilon_exploration.EpsilonExploration method) DQNAgent (class in rlgraph.agents.dqn_agent) DQNLossFunction (class in rlgraph.components.loss_functions.dqn_loss_function) draw() (rlgraph.components.distributions.distribution.Distribution method) dtype (rlgraph.spaces.containers.Dict attribute) (rlgraph.spaces.containers.Tuple attribute) DuelingActionAdapter (class in rlgraph.components.action_adapters.dueling_action_adapter)

E

entropy() (rlgraph.components.distributions.distribution.Distribution method) Environment (class in rlgraph.environments.environment) EnvironmentStepper (class in rlgraph.components.common.environment_stepper)

episodic_reset() (rlgraph.environments.openai_gym.OpenAIGymEnv method) EpsilonExploration (class in rlgraph.components.explorations.epsilon_exploration) Exploration (class in rlgraph.components.explorations.exploration) export_graph() (rlgraph.agents.agent.Agent method)

F

FIFOQueue (class in rlgraph.components.memories.fifo_queue) flat_dim (rlgraph.spaces.box_space.BoxSpace attribute) (rlgraph.spaces.containers.Dict attribute) (rlgraph.spaces.containers.Tuple attribute) (rlgraph.spaces.space.Space attribute) flat_dim_with_categories (rlgraph.spaces.int_box.IntBox attribute) flatten() (rlgraph.spaces.space.Space method) FloatBox (class in rlgraph.spaces.float_box)

force_batch() (rlgraph.spaces.box_space.BoxSpace method) (rlgraph.spaces.containers.Dict method) (rlgraph.spaces.containers.Tuple method) (rlgraph.spaces.space.Space method) force_tuple() (in module rlgraph.components.common.environment_stepper) (in module rlgraph.components.neural_networks.neural_network) (in module rlgraph.components.neural_networks.stack) from_spec() (rlgraph.components.neural_networks.stack.Stack class method)

G

get() (rlgraph.components.helpers.MemSegmentTree method) (rlgraph.components.helpers.SegmentTree method) get_action() (rlgraph.agents.agent.Agent method) (rlgraph.agents.dqn_agent.DQNAgent method) (rlgraph.agents.impala_agent.IMPALAAgent method) (rlgraph.components.neural_networks.policy.Policy method) get_action_layer_output() (rlgraph.components.action_adapters.action_adapter.ActionAdapter method) (rlgraph.components.action_adapters.dueling_action_adapter.DuelingActionAdapter method) (rlgraph.components.neural_networks.policy.Policy method) get_activation_function() (in module rlgraph.components.layers.nn.activation_functions) get_all_sub_components() (rlgraph.components.component.Component method) get_backend() (in module rlgraph) get_discrete_pos() (rlgraph.environments.grid_world.GridWorld method) get_dist_to_goal() (rlgraph.environments.grid_world.GridWorld method) get_distributed_backend() (in module rlgraph) get_entropy() (rlgraph.components.neural_networks.policy.Policy method) get_list_registry() (in module rlgraph.spaces.space_utils) get_logits() (rlgraph.components.action_adapters.action_adapter.ActionAdapter method) (rlgraph.components.action_adapters.baseline_action_adapter.BaselineActionAdapter method) get_logits_probabilities_log_probs() (rlgraph.components.action_adapters.action_adapter.ActionAdapter method) (rlgraph.components.action_adapters.baseline_action_adapter.BaselineActionAdapter method) (rlgraph.components.action_adapters.dueling_action_adapter.DuelingActionAdapter method) (rlgraph.components.neural_networks.policy.Policy method) get_max_likelihood_action() (rlgraph.components.neural_networks.policy.Policy method) get_min_value() (rlgraph.components.helpers.MemSegmentTree method) (rlgraph.components.helpers.SegmentTree method) get_nn_output() (rlgraph.components.neural_networks.policy.Policy method) get_number_of_allowed_inputs() (rlgraph.components.component.Component method) get_optimizer_variables() (rlgraph.components.optimizers.local_optimizers.LocalOptimizer method) (rlgraph.components.optimizers.optimizer.Optimizer method)

get_parents() (rlgraph.components.component.Component method) get_policy_weights() (rlgraph.agents.agent.Agent method) get_possible_next_positions() (rlgraph.environments.grid_world.GridWorld method) get_preprocessed_space() (rlgraph.components.layers.layer.Layer method) (rlgraph.components.neural_networks.dict_preprocessor_stack.DictPreprocessorStack method) (rlgraph.components.neural_networks.preprocessor_stack.PreprocessorStack method) get_preprocessed_state_action_and_action_probs() (rlgraph.components.neural_networks.actor_component.ActorComponent method) get_preprocessed_state_and_action() (rlgraph.components.neural_networks.actor_component.ActorComponent method) get_shape() (rlgraph.spaces.box_space.BoxSpace method) (rlgraph.spaces.containers.Dict method) (rlgraph.spaces.containers.Tuple method) (rlgraph.spaces.int_box.IntBox method) (rlgraph.spaces.space.Space method) get_space_from_op() (in module rlgraph.spaces.space_utils) get_state_values_and_logits() (rlgraph.components.action_adapters.baseline_action_adapter.BaselineActionAdapter method) get_stochastic_action() (rlgraph.components.neural_networks.policy.Policy method) get_sub_component_by_global_scope() (rlgraph.components.component.Component method) get_sub_component_by_name() (rlgraph.components.component.Component method) get_sum() (rlgraph.components.helpers.MemSegmentTree method) (rlgraph.components.helpers.SegmentTree method) get_td_loss() (rlgraph.agents.apex_agent.ApexAgent method) get_variable() (rlgraph.components.component.Component method) (rlgraph.spaces.box_space.BoxSpace method) (rlgraph.spaces.containers.Dict method) (rlgraph.spaces.containers.Tuple method) (rlgraph.spaces.space.Space method) get_variables() (rlgraph.components.component.Component method) get_variables_by_name() (rlgraph.components.component.Component method) GradientDescentOptimizer (class in rlgraph.components.optimizers.local_optimizers) GridWorld (class in rlgraph.environments.grid_world)

H

has_rnn() (rlgraph.components.neural_networks.neural_network.NeuralNetwork method)

HorovodOptimizer (class in rlgraph.components.optimizers.horovod_optimizer)

I

IMPALAAgent (class in rlgraph.agents.impala_agent) IMPALALossFunction (class in rlgraph.components.loss_functions.impala_loss_function) import_observations() (rlgraph.agents.agent.Agent method) index_of_prefixsum() (rlgraph.components.helpers.MemSegmentTree method) (rlgraph.components.helpers.SegmentTree method)

insert() (rlgraph.components.helpers.MemSegmentTree method) (rlgraph.components.helpers.SegmentTree method) IntBox (class in rlgraph.spaces.int_box)

K

kl_divergence() (rlgraph.components.distributions.distribution.Distribution method)

L

Layer (class in rlgraph.components.layers.layer) load_model() (rlgraph.agents.agent.Agent method) LocalOptimizer (class in rlgraph.components.optimizers.local_optimizers) log_prob() (rlgraph.components.distributions.distribution.Distribution method)

loss() (rlgraph.components.loss_functions.dqn_loss_function.DQNLossFunction method) (rlgraph.components.loss_functions.impala_loss_function.IMPALALossFunction method) (rlgraph.components.loss_functions.loss_function.LossFunction method) LossFunction (class in rlgraph.components.loss_functions.loss_function) LSTMLayer (class in rlgraph.components.layers.nn.lstm_layer)

M

MAPS (rlgraph.environments.grid_world.GridWorld attribute) MaxPool2DLayer (class in rlgraph.components.layers.nn.maxpool2d_layer)

Memory (class in rlgraph.components.memories.memory) MemSegmentTree (class in rlgraph.components.helpers)

N

NadamOptimizer (class in rlgraph.components.optimizers.local_optimizers) NeuralNetwork (class in rlgraph.components.neural_networks.neural_network)

NNLayer (class in rlgraph.components.layers.nn.nn_layer) noop_reset() (rlgraph.environments.openai_gym.OpenAIGymEnv method) Normal (class in rlgraph.components.distributions.normal)

O

observe() (rlgraph.agents.agent.Agent method)

OpenAIGymEnv (class in rlgraph.environments.openai_gym) Optimizer (class in rlgraph.components.optimizers.optimizer)

P

Policy (class in rlgraph.components.neural_networks.policy) preprocess_states() (rlgraph.agents.agent.Agent method) PreprocessLayer (class in rlgraph.components.layers.preprocessing.preprocess_layer) PreprocessorStack (class in rlgraph.components.neural_networks.preprocessor_stack)

PrioritizedReplay (class in rlgraph.components.memories.prioritized_replay) propagate_scope() (rlgraph.components.component.Component method) propagate_sub_component_properties() (rlgraph.components.component.Component method) propagate_summary() (rlgraph.components.component.Component method) propagate_variables() (rlgraph.components.component.Component method)

Q

QueueRunner (class in rlgraph.components.memories.queue_runner)

R

RandomEnv (class in rlgraph.environments.random_env) rank (rlgraph.spaces.containers.Dict attribute) (rlgraph.spaces.containers.Tuple attribute) (rlgraph.spaces.space.Space attribute) read_variable() (rlgraph.components.component.Component static method) reduce() (rlgraph.components.helpers.MemSegmentTree method) (rlgraph.components.helpers.SegmentTree method) refresh_state() (rlgraph.environments.grid_world.GridWorld method) register_api_methods_and_graph_fns() (rlgraph.components.component.Component method) register_variables() (rlgraph.components.component.Component method) remove_sub_component_by_name() (rlgraph.components.component.Component method) render() (rlgraph.environments.environment.Environment method) (rlgraph.environments.grid_world.GridWorld method) (rlgraph.environments.openai_gym.OpenAIGymEnv method) RepeaterStack (class in rlgraph.components.common.repeater_stack) ReplayMemory (class in rlgraph.components.memories.replay_memory) reset() (rlgraph.agents.agent.Agent method) (rlgraph.agents.dqn_agent.DQNAgent method) (rlgraph.components.neural_networks.dict_preprocessor_stack.DictPreprocessorStack method) (rlgraph.components.neural_networks.preprocessor_stack.PreprocessorStack method) (rlgraph.environments.environment.Environment method) (rlgraph.environments.grid_world.GridWorld method) (rlgraph.environments.openai_gym.OpenAIGymEnv method) (rlgraph.environments.random_env.RandomEnv method) reset_env_buffers() (rlgraph.agents.agent.Agent method) reset_for_env_stepper() (rlgraph.environments.openai_gym.OpenAIGymEnv method) (rlgraph.environments.random_env.RandomEnv method) reset_profile() (rlgraph.components.component.Component static method) ResidualLayer (class in rlgraph.components.layers.nn.residual_layer) rlgraph (module) rlgraph.agents.agent (module) rlgraph.agents.apex_agent (module) rlgraph.agents.dqn_agent (module) rlgraph.agents.impala_agent (module) rlgraph.components.action_adapters.action_adapter (module) rlgraph.components.action_adapters.baseline_action_adapter (module) rlgraph.components.action_adapters.dueling_action_adapter (module) rlgraph.components.common.batch_splitter (module) rlgraph.components.common.container_splitter (module) rlgraph.components.common.dict_merger (module) rlgraph.components.common.environment_stepper (module) rlgraph.components.common.repeater_stack (module) rlgraph.components.common.sampler (module) rlgraph.components.common.slice (module) rlgraph.components.common.staging_area (module) rlgraph.components.common.synchronizable (module), [1] rlgraph.components.component (module) rlgraph.components.distributions.bernoulli (module) rlgraph.components.distributions.beta (module)

rlgraph.components.distributions.categorical (module) rlgraph.components.distributions.distribution (module) rlgraph.components.distributions.normal (module) rlgraph.components.explorations.epsilon_exploration (module) rlgraph.components.explorations.exploration (module) rlgraph.components.helpers (module) rlgraph.components.layers.layer (module) rlgraph.components.layers.nn.activation_functions (module) rlgraph.components.layers.nn.concat_layer (module) rlgraph.components.layers.nn.conv2d_layer (module) rlgraph.components.layers.nn.dense_layer (module) rlgraph.components.layers.nn.lstm_layer (module) rlgraph.components.layers.nn.maxpool2d_layer (module) rlgraph.components.layers.nn.nn_layer (module) rlgraph.components.layers.nn.residual_layer (module) rlgraph.components.layers.preprocessing.preprocess_layer (module) rlgraph.components.layers.strings.string_layer (module) rlgraph.components.loss_functions.dqn_loss_function (module) rlgraph.components.loss_functions.impala_loss_function (module) rlgraph.components.loss_functions.loss_function (module) rlgraph.components.memories.fifo_queue (module) rlgraph.components.memories.memory (module) rlgraph.components.memories.prioritized_replay (module) rlgraph.components.memories.queue_runner (module) rlgraph.components.memories.replay_memory (module) rlgraph.components.neural_networks.actor_component (module) rlgraph.components.neural_networks.dict_preprocessor_stack (module) rlgraph.components.neural_networks.neural_network (module) rlgraph.components.neural_networks.policy (module) rlgraph.components.neural_networks.preprocessor_stack (module) rlgraph.components.neural_networks.stack (module) rlgraph.components.optimizers.horovod_optimizer (module) rlgraph.components.optimizers.local_optimizers (module) rlgraph.components.optimizers.optimizer (module) rlgraph.components.papers (module) rlgraph.components.queues (module) rlgraph.environments.environment (module) rlgraph.environments.grid_world (module) rlgraph.environments.openai_gym (module) rlgraph.environments.random_env (module) rlgraph.spaces.bool_box (module) rlgraph.spaces.box_space (module) rlgraph.spaces.containers (module) rlgraph.spaces.float_box (module) rlgraph.spaces.int_box (module) rlgraph.spaces.space (module) rlgraph.spaces.space_utils (module) rlgraph.spaces.text_box (module) RMSPropOptimizer (class in rlgraph.components.optimizers.local_optimizers)

S

sample() (rlgraph.spaces.bool_box.BoolBox method) (rlgraph.spaces.containers.ContainerSpace method) (rlgraph.spaces.containers.Dict method) (rlgraph.spaces.containers.Tuple method) (rlgraph.spaces.float_box.FloatBox method) (rlgraph.spaces.int_box.IntBox method) (rlgraph.spaces.space.Space method) (rlgraph.spaces.text_box.TextBox method) sample_deterministic() (rlgraph.components.distributions.distribution.Distribution method) sample_stochastic() (rlgraph.components.distributions.distribution.Distribution method) Sampler (class in rlgraph.components.common.sampler) sanity_check_space() (in module rlgraph.spaces.space_utils) scatter_update_variable() (rlgraph.components.component.Component static method) seed() (rlgraph.environments.environment.Environment method) (rlgraph.environments.grid_world.GridWorld method) (rlgraph.environments.openai_gym.OpenAIGymEnv method) (rlgraph.environments.random_env.RandomEnv method) SegmentTree (class in rlgraph.components.helpers) set_policy_weights() (rlgraph.agents.agent.Agent method)

SGDOptimizer (class in rlgraph.components.optimizers.local_optimizers) shape (rlgraph.spaces.containers.Dict attribute) (rlgraph.spaces.containers.Tuple attribute) (rlgraph.spaces.space.Space attribute) Slice (class in rlgraph.components.common.slice) SoftMax (class in rlgraph.components.helpers) Space (class in rlgraph.spaces.space) Stack (class in rlgraph.components.neural_networks.stack) StagingArea (class in rlgraph.components.common.staging_area) step() (rlgraph.environments.environment.Environment method) (rlgraph.environments.grid_world.GridWorld method) (rlgraph.environments.openai_gym.OpenAIGymEnv method) (rlgraph.environments.random_env.RandomEnv method) step_for_env_stepper() (rlgraph.environments.openai_gym.OpenAIGymEnv method) (rlgraph.environments.random_env.RandomEnv method) store_model() (rlgraph.agents.agent.Agent method) StringLayer (class in rlgraph.components.layers.strings.string_layer) sub_component_by_name() (rlgraph.components.component.Component method) Synchronizable (class in rlgraph.components.common.synchronizable), [1]

T

terminate() (rlgraph.agents.agent.Agent method) (rlgraph.environments.environment.Environment method) (rlgraph.environments.openai_gym.OpenAIGymEnv method)

TextBox (class in rlgraph.spaces.text_box) translate_space() (rlgraph.environments.openai_gym.OpenAIGymEnv static method) Tuple (class in rlgraph.spaces.containers)

U

update() (rlgraph.agents.agent.Agent method) (rlgraph.agents.apex_agent.ApexAgent method) (rlgraph.agents.dqn_agent.DQNAgent method) (rlgraph.agents.impala_agent.IMPALAAgent method)

update_cam_pixels() (rlgraph.environments.grid_world.GridWorld method)

V

VTraceFunction (class in rlgraph.components.helpers)

W

when_input_complete() (rlgraph.components.component.Component method) with_batch_rank() (rlgraph.spaces.space.Space method)

with_extra_ranks() (rlgraph.spaces.space.Space method) with_time_rank() (rlgraph.spaces.space.Space method)

X

x (rlgraph.environments.grid_world.GridWorld attribute)

Y

y (rlgraph.environments.grid_world.GridWorld attribute)

Z

zeros() (rlgraph.spaces.box_space.BoxSpace method) (rlgraph.spaces.containers.Dict method) (rlgraph.spaces.containers.Tuple method) (rlgraph.spaces.space.Space method)