Package teamwork :: Package examples :: Package Thespian :: Module ThespianAgent :: Class ThespianAgent

Class ThespianAgent

              agent.Agent.Agent --+                    
                                  |                    
agent.RecursiveAgent.RecursiveAgent --+                
                                      |                
         agent.GoalBased.GoalBasedAgent --+            
                                          |            
              agent.stereotypes.Stereotyper --+        
                                              |        
               agent.DefaultBased.GenericEntity --+    
                                                  |    
                  agent.Agent.Agent --+           |    
                                      |           |    
    agent.RecursiveAgent.RecursiveAgent --+       |    
                                          |       |    
             agent.GoalBased.GoalBasedAgent --+   |    
                                              |   |    
                        agent.support.Supporter --+    
                                                  |    
                  agent.Agent.Agent --+           |    
                                      |           |    
    agent.RecursiveAgent.RecursiveAgent --+       |    
                                          |       |    
              agent.MemoryAgent.MemoryAgent --+   |    
                                              |   |    
              agent.consistency.ConsistentAgent --+    
                                                  |    
                         agent.Entities.PsychEntity --+
                                                      |
                                                     ThespianAgent

Nested Classes

[hide private]

Inherited from agent.Agent.Agent: actionClass

Instance Methods

[hide private]

dict

observe(self, actionDict={})
Returns: observations this entity would make of the given actions, in the same format as the provided action dictionary source code

(Action[],Element)

applyPolicy(self, state=None, actions=[], history=None, debug=None, explain=False)
Generates a decision chosen according to the agent's current policy source code

updateActionChoices(self, choices, state=None, includeLocation=False)

source code

updateActionChoices7(self, choices, state=None, includeLocation=False)

source code

forceinitState(self)
Instantiates all of the state defaults relevant to this entity

source code

dict[]

fitAllAction(self, horizon=-1, state=None)
Computes a set of constraints on possible goal weights for this agent that, if satisfied, will cause the agent to prefer the desired action in the given state.

source code

Distribution over float

applyRealGoals(self, entity=None, world=None, debug=Debugger (0), fixedgoals=['sameLocation', 'actAlive', 'specialRule'])
Returns: expected reward of the entity in current world source code

getDefault(self, feature)
Finds the most specific class defaults for the specified feature; raises KeyError exception if no default exists

source code

initGoals(self, entities=[])
Sets the goal weights of this entity based on class defaults. source code

initEntities(self, entityList, depth=1, maxDepth=-1)
Sets the entities known to be the list provided, and makes the appropriate updates to goals, policy depth, etc.

source code

PWLDynamics

getDynamics(self, act, feature=None)
Returns this entity's dynamics model for the given action

source code

hypotheticalPostCom(self, beliefs, msgs, epoch=-1, debug=Debugger (0))
Returns: the potential change in the agent's beliefs based on received messages

source code

Inherited from agent.Entities.PsychEntity: __cmp__, __copy__, __getstate__, __init__, __setstate__, __xml__, acceptMessage, acceptability, extendPolicy, getPolicyDepth, parse, postComStateEstimator, preComStateEstimator, setPolicyDepth, stateEstimator

Inherited from agent.DefaultBased.GenericEntity: addActions, applyBeliefs, applyDefaults, initActions, initBeliefs, initDynamics, initModels, initObservations, initRelationships, initState, instanceof, instantiateGoal, instantiateList, instantiateName, setHierarchy

Inherited from agent.DefaultBased.GenericEntity (private): _addActions, _initActions

Inherited from agent.stereotypes.Stereotyper: __str__, clusterSpace, freezeModels, generateSpace, hypotheticalPreCom, reachable, setModel, updateModel, updateModels, weightEqual

Inherited from agent.GoalBased.GoalBasedAgent: actionValue, applyGoals, expectedValue, fit, generateConstraints, getGoalTree, getGoalVector, getGoalWeight, getGoals, getNormalization, normalizeGoals, setGoalWeight, setGoals

Inherited from agent.RecursiveAgent.RecursiveAgent: __deepcopy__, __eq__, __getitem__, __ne__, ancestry, applyChanges, beliefDepth, findObservation, freeze, getActionKeys, getAllBeliefs, getBelief, getBeliefKeys, getEntities, getEntity, getEntityBeliefs, getNestedBelief, getObservation, getObservations, getSelfBelief, getState, getStateFeatures, hasBelief, incorporateMessage, initialStateEstimator, invalidateCache, multistep, resetHistory, saveObservations, setBelief, setEntity, setName, setObservation, setRecursiveBelief, setSelfBelief, setState, step, toHTML, updateStateDict

Inherited from agent.Agent.Agent: generateAllObservations, generateHistories, legalActions, legalMessages, observable

Inherited from agent.support.Supporter: getLink, getLinkKey, getLinkTypes, getLinkees, getSupport, getTrust, removeLink, setLink, setSupport, setTrust, updateLinks, updateTrust

Inherited from agent.consistency.ConsistentAgent: __getMessageContentKey__, __getStateIntegratedWithMessage__, __propagateBeliefsBackInTime__, __propagateBeliefsForwardInTime__, __solveForUniqueTreePath__, calculateMessageConsistency, getMemoryConsistency

Inherited from agent.MemoryAgent.MemoryAgent: updateMemory

Class Variables

[hide private]

Inherited from agent.Entities.PsychEntity: beliefWeights, consistencyWeights

Inherited from agent.DefaultBased.GenericEntity: copyCount, factors

Inherited from agent.stereotypes.Stereotyper: defaultModelChange

Inherited from agent.GoalBased.GoalBasedAgent: valueType

Inherited from agent.support.Supporter: supportLimit, supportWeights

Inherited from agent.support.Supporter (private): _supportFeature, _trustFeature

Instance Variables

[hide private]

Inherited from agent.stereotypes.Stereotyper: model, modelChange, models

Inherited from agent.GoalBased.GoalBasedAgent: constraints, goals, horizon

Inherited from agent.RecursiveAgent.RecursiveAgent: dynamics, parent, relationships, state

Inherited from agent.Agent.Agent: actions, name, omega

Inherited from agent.support.Supporter: linkDynamics, linkTypes, links

Method Details

[hide private]

observe(self, actionDict=`{}`)

source code

Parameters:

actionDict (dict:str→Action[]) - the performed actions, indexed by actor name

Returns: dict

observations this entity would make of the given actions, in the same format as the provided action dictionary

Overrides: agent.RecursiveAgent.RecursiveAgent.observe

applyPolicy(self, state=None, actions=`[]`, history=None, debug=None, explain=False)

source code

Generates a decision chosen according to the agent's current policy

Parameters:

state (Distribution(KeyedVector)) - the current state vector
actions (Action[]) - the possible actions the agent can consider (defaults to all available actions)
history (Action[]:bool) - a dictionary of actions that have already been performed (and which should not be performed again if they are labeled as not repeatable)
explain (bool) - flag indicating whether an explanation should be generated

Returns: (Action[],Element)

a list of actions and an explanation, the latter provided by execute

Overrides: agent.RecursiveAgent.RecursiveAgent.applyPolicy

fitAllAction(self, horizon=-1, state=None)

source code

Computes a set of constraints on possible goal weights for this agent that, if satisfied, will cause the agent to prefer the desired action in the given state. Each constraint is dictionary with the following elements:

delta: the total difference that must be made up
slope: dictionary of coefficients for each goal weight in the sum that must make up that difference
plane: the vector of weights, such that the product of this vector and the goal weight vector must exceed 0 for the desired action to be preferred

Parameters:

desired (Action[]) - the action that the agent should prefer
horizon (int) - the horizon of lookahead to use (if not provided, the agent's default horizon is used)
state (dict) - the current state of this agent's beliefs (if not provided, defaults to the result of getAllBeliefs

Returns: dict[]

a list of constraints

applyRealGoals(self, entity=None, world=None, debug=Debugger (0), fixedgoals=`['sameLocation',` `'actAlive',` `'specialRule']`)

source code

Parameters:

entity (GoalBasedAgent) - the entity whose goals are to be evaluated (default is self)
world (teamwork.multiagent.PsychAgents.PsychAgents) - the context for evaluating the goals (default is the beliefs of entity)

Returns: Distribution over float

expected reward of the entity in current world

getDefault(self, feature)

source code

Finds the most specific class defaults for the specified feature; raises KeyError exception if no default exists

Overrides: agent.DefaultBased.GenericEntity.getDefault

initGoals(self, entities=`[]`)

source code

Sets the goal weights of this entity based on class defaults. The resulting goals depend on the group of entities passed in as the optional argument.

Overrides: agent.DefaultBased.GenericEntity.initGoals

initEntities(self, entityList, depth=1, maxDepth=-1)

source code

Sets the entities known to be the list provided, and makes the appropriate updates to goals, policy depth, etc.

Parameters:

maxDepth - the maximum depth of this agent's recursive models

Overrides: agent.support.Supporter.initEntities

getDynamics(self, act, feature=None)

source code

Returns this entity's dynamics model for the given action

Parameters:

act (Action) - the action whose effect we are interested in
feature (str) - if the optional feature argument is provided, then this method returns the dynamics for only the given feature; otherwise, returns the effect over all state features (but this latter capability is now deprecated)

Returns: PWLDynamics

this entity's dynamics model for the given action

Overrides: agent.RecursiveAgent.RecursiveAgent.getDynamics

hypotheticalPostCom(self, beliefs, msgs, epoch=-1, debug=Debugger (0))

source code

Returns:: the potential change in the agent's beliefs based on received messages
Overrides: agent.Entities.PsychEntity.hypotheticalPostCom

Class ThespianAgent

observe(self, actionDict={})

applyPolicy(self, state=None, actions=[], history=None, debug=None, explain=False)

fitAllAction(self, horizon=-1, state=None)

applyRealGoals(self, entity=None, world=None, debug=Debugger (0), fixedgoals=['sameLocation', 'actAlive', 'specialRule'])

getDefault(self, feature)

initGoals(self, entities=[])

initEntities(self, entityList, depth=1, maxDepth=-1)

getDynamics(self, act, feature=None)

hypotheticalPostCom(self, beliefs, msgs, epoch=-1, debug=Debugger (0))

observe(self, actionDict=`{}`)

applyPolicy(self, state=None, actions=`[]`, history=None, debug=None, explain=False)

applyRealGoals(self, entity=None, world=None, debug=Debugger (0), fixedgoals=`['sameLocation',` `'actAlive',` `'specialRule']`)

initGoals(self, entities=`[]`)