ift7025-tp1/reinforcement/valueIterationAgents.py

# -*- coding: utf-8 -*-
# valueIterationAgents.py
# -----------------------
# Licensing Information:  You are free to use or extend these projects for
# educational purposes provided that (1) you do not distribute or publish
# solutions, (2) you retain this notice, and (3) you provide clear
# attribution to UC Berkeley, including a link to http://ai.berkeley.edu.
# 
# Attribution Information: The Pacman AI projects were developed at UC Berkeley.
# The core projects and autograders were primarily created by John DeNero
# (denero@cs.berkeley.edu) and Dan Klein (klein@cs.berkeley.edu).
# Student side autograding was added by Brad Miller, Nick Hay, and
# Pieter Abbeel (pabbeel@cs.berkeley.edu).

# François Pelletier
# IFT-7025
# Matricule: 908144032


import mdp, util

from learningAgents import ValueEstimationAgent


class ValueIterationAgent(ValueEstimationAgent):
    """
        * Please read learningAgents.py before reading this.*

        A ValueIterationAgent takes a Markov decision process
        (see mdp.py) on initialization and runs value iteration
        for a given number of iterations using the supplied
        discount factor.
    """

    def __init__(self, mdp, discount=0.9, iterations=100):
        """
          Your value iteration agent should take an mdp on
          construction, run the indicated number of iterations
          and then act according to the resulting policy.

          Some useful mdp methods you will use:
              mdp.getStates()
              mdp.getPossibleActions(state)
              mdp.getTransitionStatesAndProbs(state, action)
              mdp.getReward(state, action, nextState)
              mdp.isTerminal(state)
        """
        self.mdp = mdp
        self.discount = discount
        self.iterations = iterations
        self.values = util.Counter()  # A Counter is a dict with default 0

        # Write value iteration code here
        "*** YOUR CODE HERE ***"
        states = self.mdp.getStates()

        # print "__init__ ... states: " + str(states)

        for i in range(iterations):
            # On reprend les valeurs de l'itération précédente comme référence
            # Copie pour batch
            q_copy = self.values.copy()
            for state in states:
                q_new = None
                for action in self.mdp.getPossibleActions(state):
                    q = self.computeQValueFromValues(state, action)
                    # Garder la meilleure Q value
                    if q_new is None or q_new < q:
                        q_new = q
                # Gérer le cas sans successeurs
                if q_new is None:
                    q_copy[state] = 0
                else:
                    q_copy[state] = q_new
            # On met à jour pout les prochaines itérations
            self.values = q_copy

    def getValue(self, state):
        """
          Return the value of the state (computed in __init__).
        """
        return self.values[state]

    def computeQValueFromValues(self, state, action):
        """
          Compute the Q-value of action in state from the
          value function stored in self.values.
        """
        "*** YOUR CODE HERE ***"
        values = []
        for nextState, prob in self.mdp.getTransitionStatesAndProbs(state,action):
            reward = self.mdp.getReward(state, action, nextState)
            discount = self.discount
            next_state_value = self.values[nextState]
            values.append(prob*(reward+discount*next_state_value))
        return sum(values)

    def computeActionFromValues(self, state):
        """
          The policy is the best action in the given state
          according to the values currently stored in self.values.

          You may break ties any way you see fit.  Note that if
          there are no legal actions, which is the case at the
          terminal state, you should return None.
        """
        "*** YOUR CODE HERE ***"
        possibleActions = self.mdp.getPossibleActions(state)

        if len(possibleActions) == 0:
            return None

        q_values = [self.computeQValueFromValues(state, action) for action in possibleActions]
        # print "computeActionFromValues ... q_values: "+str(q_values)
        # print "index:"+str(q_values.index(max(q_values)))
        # print "action:"+str(possibleActions[q_values.index(max(q_values))])
        return possibleActions[q_values.index(max(q_values))]

    def getPolicy(self, state):
        return self.computeActionFromValues(state)

    def getAction(self, state):
        "Returns the policy at the state (no exploration)."
        return self.computeActionFromValues(state)

    def getQValue(self, state, action):
        return self.computeQValueFromValues(state, action)