Class
BaseAgent
Abstract class from which inherit to implement an agent.
- edit Alexis BRENON alexis.brenon@imag.fr
Metamethods
keyboard_arrow_up-
__init ( args )
Abstract constructor.
Parameters:-
table
args
-
table
Abstract Methods
keyboard_arrow_up-
get_action ()
Return an action to do.
Returns:- number Action to execute
-
get_experienced_interactions ()
Return how many interactions the agent lived.
Returns:- number Number of interactions done
-
get_learned_epoch ()
Return how many times the agent actually learned from its experience.
Returns:- number Number of times the agent learned
-
give_reward ( reward )
Reward or punish the agent.
Parameters:-
number
reward
Reward if positive, punishment if negative
-
self
-
number
-
integrate_observation ( state )
Integrate current observation from the environment.
Parameters:-
environment.ObservableState
state
The current state of the environment
-
self
-
environment.ObservableState
Public Methods
keyboard_arrow_up-
evaluate ()
Put the agent in an evaluation mode.
This can change some internal values of the agent, like the epsilon value for epsilon-greedy strategy.
Returns:-
self
-
-
report ()
Return pretty-printed informations about the agent.
Returns:- string A human readable string about the agent
-
training ()
Put the agent in a training mode.
This is the default mode of an agent.
Returns:-
self
-