deepxube.updaters.updater_policy_rl

Module Contents

Classes

UpdatePolicyRL
UpdatePolicyRLKeepGoalABC
UpdatePolicyRLHERABC
UpdatePolicyRLKeepGoal
UpdatePolicyRLHER
UpdatePolicyRLKeepGoalHeurV
UpdatePolicyRLHERHeurV
UpdatePolicyRLKeepGoalHeurQ
UpdatePolicyRLHERHeurQ

Functions

_pathfind_step
_get_edge_popped_data

API

deepxube.updaters.updater_policy_rl._pathfind_step(pathfind: deepxube.base.pathfinding.PathFind) → List[deepxube.base.pathfinding.EdgeQ]

deepxube.updaters.updater_policy_rl._get_edge_popped_data(edges_popped: List[deepxube.base.pathfinding.EdgeQ], times: deepxube.utils.timing_utils.Times) → Tuple[List[deepxube.base.domain.State], List[deepxube.base.domain.Goal], List[deepxube.base.domain.Action]]

class deepxube.updaters.updater_policy_rl.UpdatePolicyRL(domain: deepxube.base.updater.D, pathfind_arg: str, up_args: deepxube.base.updater.UpArgs)

Bases: deepxube.base.updater.UpdatePolicy[deepxube.base.updater.D, deepxube.base.pathfinding.FNsP, deepxube.base.pathfinding.PathFindActsPolicy, deepxube.base.pathfinding.Instance], deepxube.base.updater.UpdateRL[deepxube.base.updater.D, deepxube.base.pathfinding.FNsP, deepxube.base.pathfinding.PathFindActsPolicy, deepxube.base.pathfinding.Instance], abc.ABC

static pathfind_type() → Type[deepxube.base.pathfinding.PathFindActsPolicy]

_step(pathfind: deepxube.base.pathfinding.PathFindActsPolicy, times: deepxube.utils.timing_utils.Times) → None

_inputs_ctgs_to_np(states: List[deepxube.base.domain.State], goals: List[deepxube.base.domain.Goal], actions: List[deepxube.base.domain.Action], times: deepxube.utils.timing_utils.Times) → List[numpy.typing.NDArray]

_init_replay_buffer(max_size: int) → None

_rb_add(states: List[deepxube.base.domain.State], goals: List[deepxube.base.domain.Goal], actions: List[deepxube.base.domain.Action], times: deepxube.utils.timing_utils.Times) → None

_sample_rb(num: int, times: deepxube.utils.timing_utils.Times) → Tuple[List[deepxube.base.domain.State], List[deepxube.base.domain.Goal], List[deepxube.base.domain.Action]]

class deepxube.updaters.updater_policy_rl.UpdatePolicyRLKeepGoalABC(domain: deepxube.base.updater.D, pathfind_arg: str, up_args: deepxube.base.updater.UpArgs)

Bases: deepxube.updaters.updater_policy_rl.UpdatePolicyRL[deepxube.base.domain.Domain, deepxube.base.pathfinding.FNsP], abc.ABC

static domain_type() → Type[deepxube.base.domain.Domain]

_step_sync_main(pathfind: deepxube.base.pathfinding.PathFindActsPolicy, times: deepxube.utils.timing_utils.Times) → List[numpy.typing.NDArray]

_get_instance_data_norb(instances: List[deepxube.base.pathfinding.Instance], times: deepxube.utils.timing_utils.Times) → List[numpy.typing.NDArray]

_get_instance_data_rb(instances: List[deepxube.base.pathfinding.Instance], times: deepxube.utils.timing_utils.Times) → List[numpy.typing.NDArray]

class deepxube.updaters.updater_policy_rl.UpdatePolicyRLHERABC(domain: deepxube.base.updater.D, pathfind_arg: str, up_args: deepxube.base.updater.UpArgs)

Bases: deepxube.updaters.updater_policy_rl.UpdatePolicyRL[deepxube.base.domain.GoalSampleableFromState, deepxube.base.pathfinding.FNsP], deepxube.base.updater.UpdateHER[deepxube.base.pathfinding.FNsP, deepxube.base.pathfinding.PathFindActsPolicy, deepxube.base.pathfinding.Instance], abc.ABC

static domain_type() → Type[deepxube.base.domain.GoalSampleableFromState]

_get_instance_data_rb(instances: List[deepxube.base.pathfinding.Instance], times: deepxube.utils.timing_utils.Times) → List[numpy.typing.NDArray]

class deepxube.updaters.updater_policy_rl.UpdatePolicyRLKeepGoal(domain: deepxube.base.updater.D, pathfind_arg: str, up_args: deepxube.base.updater.UpArgs)

Bases: deepxube.updaters.updater_policy_rl.UpdatePolicyRLKeepGoalABC[deepxube.base.pathfinding.FNsPolicy]

static functions_type() → Type[deepxube.base.pathfinding.FNsPolicy]

_get_pathfind_functions() → deepxube.base.pathfinding.FNsPolicy

class deepxube.updaters.updater_policy_rl.UpdatePolicyRLHER(domain: deepxube.base.updater.D, pathfind_arg: str, up_args: deepxube.base.updater.UpArgs)

Bases: deepxube.updaters.updater_policy_rl.UpdatePolicyRLHERABC[deepxube.base.pathfinding.FNsPolicy]

static functions_type() → Type[deepxube.base.pathfinding.FNsPolicy]

_get_pathfind_functions() → deepxube.base.pathfinding.FNsPolicy

class deepxube.updaters.updater_policy_rl.UpdatePolicyRLKeepGoalHeurV(domain: deepxube.base.updater.D, pathfind_arg: str, up_args: deepxube.base.updater.UpArgs)

Bases: deepxube.updaters.updater_policy_rl.UpdatePolicyRLKeepGoalABC[deepxube.base.pathfinding.FNsHeurVPolicy], deepxube.base.updater.UpdateHasHeur[deepxube.base.domain.Domain, deepxube.base.pathfinding.FNsHeurVPolicy, deepxube.base.pathfinding.PathFindActsPolicy, deepxube.base.pathfinding.Instance, deepxube.base.heuristic.HeurNNetParV, deepxube.base.heuristic.HeurFnV]

static functions_type() → Type[deepxube.base.pathfinding.FNsHeurVPolicy]

_get_pathfind_functions() → deepxube.base.pathfinding.FNsHeurVPolicy

class deepxube.updaters.updater_policy_rl.UpdatePolicyRLHERHeurV(domain: deepxube.base.updater.D, pathfind_arg: str, up_args: deepxube.base.updater.UpArgs)

Bases: deepxube.updaters.updater_policy_rl.UpdatePolicyRLHERABC[deepxube.base.pathfinding.FNsHeurVPolicy], deepxube.base.updater.UpdateHasHeur[deepxube.base.domain.Domain, deepxube.base.pathfinding.FNsHeurVPolicy, deepxube.base.pathfinding.PathFindActsPolicy, deepxube.base.pathfinding.Instance, deepxube.base.heuristic.HeurNNetParV, deepxube.base.heuristic.HeurFnV]

static functions_type() → Type[deepxube.base.pathfinding.FNsHeurVPolicy]

_get_pathfind_functions() → deepxube.base.pathfinding.FNsHeurVPolicy

class deepxube.updaters.updater_policy_rl.UpdatePolicyRLKeepGoalHeurQ(domain: deepxube.base.updater.D, pathfind_arg: str, up_args: deepxube.base.updater.UpArgs)

Bases: deepxube.updaters.updater_policy_rl.UpdatePolicyRLKeepGoalABC[deepxube.base.pathfinding.FNsHeurQPolicy], deepxube.base.updater.UpdateHasHeur[deepxube.base.domain.Domain, deepxube.base.pathfinding.FNsHeurQPolicy, deepxube.base.pathfinding.PathFindActsPolicy, deepxube.base.pathfinding.Instance, deepxube.base.heuristic.HeurNNetParQ, deepxube.base.heuristic.HeurFnQ]

static functions_type() → Type[deepxube.base.pathfinding.FNsHeurQPolicy]

_get_pathfind_functions() → deepxube.base.pathfinding.FNsHeurQPolicy

class deepxube.updaters.updater_policy_rl.UpdatePolicyRLHERHeurQ(domain: deepxube.base.updater.D, pathfind_arg: str, up_args: deepxube.base.updater.UpArgs)

Bases: deepxube.updaters.updater_policy_rl.UpdatePolicyRLHERABC[deepxube.base.pathfinding.FNsHeurQPolicy], deepxube.base.updater.UpdateHasHeur[deepxube.base.domain.Domain, deepxube.base.pathfinding.FNsHeurQPolicy, deepxube.base.pathfinding.PathFindActsPolicy, deepxube.base.pathfinding.Instance, deepxube.base.heuristic.HeurNNetParQ, deepxube.base.heuristic.HeurFnQ]

static functions_type() → Type[deepxube.base.pathfinding.FNsHeurQPolicy]

_get_pathfind_functions() → deepxube.base.pathfinding.FNsHeurQPolicy