Source code for simple_rl.agents.AgentClass

''' AgentClass.py: Class for a basic RL Agent '''

# Python imports.
from collections import defaultdict

[docs]class Agent(object):
    ''' Abstract Agent class. '''

    def __init__(self, name, actions, gamma=0.99):
        self.name = name
        self.actions = list(actions) # Just in case we're given a numpy array (like from Atari).
        self.gamma = gamma
        self.episode_number = 0
        self.prev_state = None
        self.prev_action = None

[docs]    def get_parameters(self):
        '''
        Returns:
            (dict) key=param_name (str) --> val=param_val (object).
        '''
        return {}

[docs]    def act(self, state, reward):
        '''
        Args:
            state (State): see StateClass.py
            reward (float): the reward associated with arriving in state @state.

        Returns:
            (str): action.
        '''
        pass

[docs]    def policy(self, state):
        return self.act(state, 0)

[docs]    def reset(self):
        '''
        Summary:
            Resets the agent back to its tabula rasa config.
        '''
        self.prev_state = None
        self.prev_action = None
        self.step_number = 0

[docs]    def end_of_episode(self):
        '''
        Summary:
            Resets the agents prior pointers.
        '''
        self.prev_state = None
        self.prev_action = None
        self.episode_number += 1

[docs]    def set_name(self, name):
        self.name = name

[docs]    def get_name(self):
        return self.name

    def __str__(self):
        return str(self.name)