pomdp_belief_tracking package

Subpackages

Submodules

pomdp_belief_tracking.pomdp_belief_tracking module

Main module.

pomdp_belief_tracking.types module

Defines some types for ease of reading

class pomdp_belief_tracking.types.Action(*args, **kwds)[source]

Bases: typing_extensions.Protocol

The abstract type representing actions

class pomdp_belief_tracking.types.Belief(initial_distribution, update_function)[source]

Bases: object

A belief is the combination of a update function and current distribution

sample()[source]

Samples from its distribution

Return type

State

Returns

state sampled according to distribution

update(a, o)[source]

Updates (in place) the state distribution given an action and observation

Parameters
Return type

Dict[str, Any]

Returns

Side effect: updates in place, returns run-time info

class pomdp_belief_tracking.types.BeliefUpdate(*args, **kwds)[source]

Bases: typing_extensions.Protocol

The signature for belief updates

pomdp_belief_tracking.types.Info

The datatype used for information flow from implementation to caller

alias of Dict[str, Any]

class pomdp_belief_tracking.types.Observation(*args, **kwds)[source]

Bases: typing_extensions.Protocol

The abstract type representing observations

class pomdp_belief_tracking.types.Simulator(*args, **kwds)[source]

Bases: typing_extensions.Protocol

The abstract type representing simulators

We expect simulators to map a state and action into a next state and observation.

__call__(s, a)[source]

Simulate a transition

Parameters
  • s (State) – the current state

  • a (Action) – the executed action

Return type

Tuple[State, Observation]

Returns

next state and observation

class pomdp_belief_tracking.types.State(*args, **kwds)[source]

Bases: typing_extensions.Protocol

The abstract type representing states

class pomdp_belief_tracking.types.StateDistribution(*args, **kwds)[source]

Bases: typing_extensions.Protocol

The abstract type representing state distributions

We expect to be able to sample states

__call__()[source]

Required implementation of distribution: the ability to sample states

Return type

State

Returns

state sampled according to distribution

class pomdp_belief_tracking.types.TransitionFunction(*args, **kwds)[source]

Bases: typing_extensions.Protocol

The signature for transition functions

Module contents

Top-level package for pomdp-belief-tracking.