Index

_

__add__() (Experience method) __call__() (Batch class method) (DQN method) (GCN method) (LSTM method)

__mul__() (Experience method)

A

action (Experience attribute) actions (Env property) add_module() (DQN method) (GCN method) (LSTM method) Agent (class in rl_equation_solver.agent.dqn) (class in rl_equation_solver.agent.gcn) (class in rl_equation_solver.agent.lstm)

append_history() (Env method) (HistoryMixin method) apply() (DQN method) (GCN method) (LSTM method) avg_history (Env property) (HistoryMixin property)

B

BaseAgent (class in rl_equation_solver.agent.base) Batch (class in rl_equation_solver.utilities.utilities) batch_states() (Agent method), [1], [2] (BaseAgent method) bfloat16() (DQN method) (GCN method) (LSTM method)

buffers() (DQN method) (GCN method) (LSTM method) build_adjacency_matrix() (in module rl_equation_solver.utilities.utilities) build_adjacency_matrix_custom() (in module rl_equation_solver.utilities.utilities)

C

children() (DQN method) (GCN method) (LSTM method) choose_action() (Agent method), [1], [2] (BaseAgent method) choose_optimal_action() (Agent method), [1], [2] (BaseAgent method) choose_random_action() (Agent method), [1], [2] (BaseAgent method) close() (Env method) compute_batch_loss() (Agent method), [1], [2] (BaseAgent method) compute_expected_Q() (Agent method), [1], [2] (BaseAgent method) compute_loss() (Agent method), [1], [2] (BaseAgent method)

compute_Q() (Agent method), [1], [2] (BaseAgent method) compute_V() (Agent method), [1], [2] (BaseAgent method) convert_state() (Agent method), [1], [2] (BaseAgent method) count() (Experience method) cpu() (DQN method) (GCN method) (LSTM method) cuda() (DQN method) (GCN method) (LSTM method) current_episode (Agent property), [1], [2] (BaseAgent property)

D

device (Agent property), [1], [2] (BaseAgent property) diff_loss_reward() (Env method) (RewardMixin method)

div() (in module rl_equation_solver.utilities.operators) double() (DQN method) (GCN method) (LSTM method) DQN (class in rl_equation_solver.agent.networks)

E

encode_onehot() (in module rl_equation_solver.utilities.utilities) Env (class in rl_equation_solver.environment.algebraic) equation (Env property) eval() (DQN method) (GCN method) (LSTM method) examples module examples.run_linear_solver_dqn module examples.run_linear_solver_gcn module

examples.run_quadratic_solver_dqn module exp_loss_reward() (Env method) (RewardMixin method) Experience (class in rl_equation_solver.utilities.utilities) expression_complexity() (Env method) (RewardMixin method) extra_repr() (DQN method) (GCN method) (LSTM method)

F

feature_dict (Env property) find_reward() (Env method) float() (DQN method) (GCN method) (LSTM method)

forward() (DQN method) (GCN method) (LSTM method) fraction() (in module rl_equation_solver.utilities.operators)

G

GCN (class in rl_equation_solver.agent.networks) get() (Id class method) get_buffer() (DQN method) (GCN method) (LSTM method) get_env() (Agent method), [1], [2] (BaseAgent method) get_extra_state() (DQN method) (GCN method) (LSTM method)

get_json_graph() (in module rl_equation_solver.utilities.utilities) get_node_features() (in module rl_equation_solver.utilities.utilities) get_node_labels() (in module rl_equation_solver.utilities.utilities) get_parameter() (DQN method) (GCN method) (LSTM method) get_submodule() (DQN method) (GCN method) (LSTM method) graph_walk() (in module rl_equation_solver.utilities.utilities) GraphEmbedding (class in rl_equation_solver.utilities.utilities)

H

half() (DQN method) (GCN method) (LSTM method) history (Agent property), [1], [2] (BaseAgent property) (Env property) (HistoryMixin property)

HistoryMixin (class in rl_equation_solver.utilities.history) huber_loss() (Agent method), [1], [2] (BaseAgent method) (LossMixin method)

I

Id (class in rl_equation_solver.utilities.utilities) index() (Experience method) info (Agent property), [1], [2] (BaseAgent property) init_config() (Agent method), [1], [2] (BaseAgent method) (Env method) init_optimizer() (Agent method), [1], [2] (BaseAgent method)

init_state() (Agent method), [1], [2] (BaseAgent method) inv_loss_reward() (Env method) (RewardMixin method) ipu() (DQN method) (GCN method) (LSTM method) is_constant_complexity() (Agent method), [1], [2] (BaseAgent method)

L

l2_loss() (Agent method), [1], [2] (BaseAgent method) (LossMixin method) load() (Agent class method), [1], [2] (BaseAgent class method) load_state_dict() (DQN method) (GCN method) (LSTM method)

log_info() (Env method) (HistoryMixin method) LossMixin (class in rl_equation_solver.utilities.loss) LSTM (class in rl_equation_solver.agent.networks)

M

module examples examples.run_linear_solver_dqn examples.run_linear_solver_gcn examples.run_quadratic_solver_dqn rl_equation_solver rl_equation_solver.agent rl_equation_solver.agent.base rl_equation_solver.agent.dqn rl_equation_solver.agent.gcn rl_equation_solver.agent.lstm rl_equation_solver.agent.networks rl_equation_solver.config rl_equation_solver.environment rl_equation_solver.environment.algebraic rl_equation_solver.utilities rl_equation_solver.utilities.history rl_equation_solver.utilities.loss rl_equation_solver.utilities.operators rl_equation_solver.utilities.reward rl_equation_solver.utilities.utilities rl_equation_solver.version

modules() (DQN method) (GCN method) (LSTM method)

N

named_buffers() (DQN method) (GCN method) (LSTM method) named_children() (DQN method) (GCN method) (LSTM method) named_modules() (DQN method) (GCN method) (LSTM method)

named_parameters() (DQN method) (GCN method) (LSTM method) next_state (Experience attribute) Node (class in rl_equation_solver.utilities.utilities) node_labels (Env property) normalize() (in module rl_equation_solver.utilities.utilities) np_random (Env property)

O

operations (Env property)

optimize_model() (Agent method), [1], [2] (BaseAgent method)

P

pad_array() (in module rl_equation_solver.utilities.utilities) parameters() (DQN method) (GCN method) (LSTM method)

parse_node_features() (in module rl_equation_solver.utilities.utilities) plot_state_as_graph() (in module rl_equation_solver.utilities.utilities) predict() (Agent method), [1], [2] (BaseAgent method) push() (ReplayMemory method)

R

register_backward_hook() (DQN method) (GCN method) (LSTM method) register_buffer() (DQN method) (GCN method) (LSTM method) register_forward_hook() (DQN method) (GCN method) (LSTM method) register_forward_pre_hook() (DQN method) (GCN method) (LSTM method) register_full_backward_hook() (DQN method) (GCN method) (LSTM method) register_full_backward_pre_hook() (DQN method) (GCN method) (LSTM method) register_load_state_dict_post_hook() (DQN method) (GCN method) (LSTM method) register_module() (DQN method) (GCN method) (LSTM method) register_parameter() (DQN method) (GCN method) (LSTM method) register_state_dict_pre_hook() (DQN method) (GCN method) (LSTM method) render() (Env method) ReplayMemory (class in rl_equation_solver.utilities.utilities) requires_grad_() (DQN method) (GCN method) (LSTM method) reset() (Env method) (Id class method) reset_history() (Env method) (HistoryMixin method)

reward (Experience attribute) RewardMixin (class in rl_equation_solver.utilities.reward) rl_equation_solver module rl_equation_solver.agent module rl_equation_solver.agent.base module rl_equation_solver.agent.dqn module rl_equation_solver.agent.gcn module rl_equation_solver.agent.lstm module rl_equation_solver.agent.networks module rl_equation_solver.config module rl_equation_solver.environment module rl_equation_solver.environment.algebraic module rl_equation_solver.utilities module rl_equation_solver.utilities.history module rl_equation_solver.utilities.loss module rl_equation_solver.utilities.operators module rl_equation_solver.utilities.reward module rl_equation_solver.utilities.utilities module rl_equation_solver.version module root() (in module rl_equation_solver.utilities.operators)

S

sample() (ReplayMemory method) save() (Agent method), [1], [2] (BaseAgent method) set_env() (Agent method), [1], [2] (BaseAgent method) set_extra_state() (DQN method) (GCN method) (LSTM method) share_memory() (DQN method) (GCN method) (LSTM method) smooth_l1_loss() (Agent method), [1], [2] (BaseAgent method) (LossMixin method) sparse_mx_to_torch_sparse_tensor() (in module rl_equation_solver.utilities.utilities) sqrt() (in module rl_equation_solver.utilities.operators)

square() (in module rl_equation_solver.utilities.operators) state (Experience attribute) state_dict() (DQN method) (GCN method) (LSTM method) state_graph (Env property) state_string (Agent property), [1], [2] (BaseAgent property) (Env property) state_vec (Env property) step() (Agent method), [1], [2] (BaseAgent method) (Env method) steps_done (Agent property), [1], [2] (BaseAgent property) sub_loss_reward() (Env method) (RewardMixin method)

T

terminate_msg() (Agent method), [1], [2] (BaseAgent method) terms (Env property) to() (DQN method) (GCN method) (LSTM method) to_empty() (DQN method) (GCN method) (LSTM method) to_graph() (in module rl_equation_solver.utilities.utilities)

to_vec() (in module rl_equation_solver.utilities.utilities) too_long() (Env method) train() (Agent method), [1], [2] (BaseAgent method) (DQN method) (GCN method) (LSTM method) type() (DQN method) (GCN method) (LSTM method)

U

unwrapped (Env property) update_config() (Agent method), [1], [2] (BaseAgent method) (Env method) update_history() (Env method) (HistoryMixin method)

update_info() (Agent method), [1], [2] (BaseAgent method) update_networks() (Agent method), [1], [2] (BaseAgent method)

V

VectorEmbedding (class in rl_equation_solver.utilities.utilities)

X

xpu() (DQN method) (GCN method) (LSTM method)

Z

zero_grad() (DQN method) (GCN method) (LSTM method)