2_01_Ejemplo_Algoritmos_Aprendizaje_Por_Refuerzo_Q_LEARNING_y_SARSA.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style>.container { width:80% !important; }</style>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from IPython.core.display import display, HTML\n",
    "display(HTML(\"<style>.container { width:80% !important; }</style>\"))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 2.1.- Ejemplo: Algoritmos de Aprendizaje por Refuerzo: Q-Learning y SARSA\n",
    "\n",
    "\n",
    "* En este notebook vamos a resolver un problema con Aprendizaje por refuerzo usando los algoritmos del Q-Learning y SARSA-Learning.\n",
    "\n",
    "\n",
    "* El problema que queremos resolver es el de encontrar el camino que nos suponga una mayor recompensa (el más corto) desde un estado inicial $[0,0]$ hasta el estado final $[4,4]$, pudiendo realizar 4 tipos de acciones:\n",
    "\n",
    "\n",
    "<img src=\"./imgs/007_RL.png\" style=\"width: 300px;\"/>\n",
    "\n",
    "\n",
    "\n",
    "* Para resolver este problema vamos a realizar lo siguiente:\n",
    "<span></span><br>\n",
    "    1. [Definición del entorno](#M1)\n",
    "<span></span><br>\n",
    "    2. [Implementación de un algoritmo de toma aleatoria de acciones](#M2)\n",
    "<span></span><br>\n",
    "    3. [Ejecución: Entorno - Agente](#M3)\n",
    "<span></span><br>\n",
    "    4. [Q-Learner: Implementación y Ejecución](#M4)\n",
    "<span></span><br>\n",
    "    5. [SARSA-Learner: Implementación y Ejecución](#M5)\n",
    "<span></span><br>\n",
    "    6. [Estrategias a corto y largo plazo](#M5)\n",
    "\n",
    "\n",
    "<hr>\n",
    "\n",
    "\n",
    "## <a name=\"M1\">1.- Definición del entorno</a>\n",
    "\n",
    "\n",
    "\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "# importamos las librerias necesarias\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "\n",
    "# Formato de los decimales en Pandas y la semilla del Random\n",
    "pd.options.display.float_format = '{:,.2f}'.format\n",
    "np.random.seed(23)\n",
    "\n",
    "\n",
    "class Environment(object):\n",
    "    def __init__(self, action_penalty=-1.0):\n",
    "        \"\"\"\n",
    "        Clase que representa y controla en entorno\n",
    "        :param step_penalty:    Factor de descuento del Reward por acción tomada\n",
    "        \"\"\"\n",
    "        self.actions = {'Arriba': [-1, 0],\n",
    "                        'Abajo': [1, 0],\n",
    "                        'Izquierda': [0, -1],\n",
    "                        'Derecha': [0, 1]}\n",
    "        self.rewards = [[0.0, 0.0, 0.0, 0.0],\n",
    "                        [0.0, 0.0, 0.0, 0.0],\n",
    "                        [0.0, 0.0, 0.0, 0.0],\n",
    "                        [0.0, 0.0, -100.0, 100.0]]\n",
    "        self.action_penalty = action_penalty         # Penalización por cada paso dado\n",
    "        self.state = [0, 0]                          # Estado en el que se encuentra el agente\n",
    "        self.final_state = [3, 3]                    # Estado final del entorno. Cuando el agente llega, se termina el episodio\n",
    "        self.total_reward = 0.0                      # Contador de recompensas en el episodio\n",
    "        self.actions_done = []                       # Lista en la que se guardan los pasos (acciones) realizadas en cada episodio\n",
    "\n",
    "    def reset(self):\n",
    "        \"\"\"\n",
    "        Método que reinicia las variables del entorno y devuelve es estado inicial\n",
    "        :return:    state\n",
    "        \"\"\"\n",
    "        self.total_reward = 0.0    # Inicializamos Reward a 0\n",
    "        self.state = [0, 0]        # Posicionamos al agente en el estado inicial\n",
    "        self.actions_done = []     # Inicializamos la listas de pasos (acciones)\n",
    "        return self.state\n",
    "\n",
    "    def step(self, action):\n",
    "        \"\"\"\n",
    "        Método que ejecuta una acción determinada del conjunto de acciones {Arriba, Abajo, Izquierda, Derecha}\n",
    "        para guiar al agente en el entorno.\n",
    "        :param action:    Acción a ejecutar\n",
    "        :return:          (state, reward, is_final_state)\n",
    "        \"\"\"\n",
    "        self.__apply_action(action)                                                # Realizamos la acción (cambio de estado)\n",
    "        self.actions_done.append(self.state[:])                                    # Guardamos el paso (accion) realizada\n",
    "        is_final_state = np.array_equal(self.state, self.final_state)              # Comprobamos si hemos llegado al estado final\n",
    "        reward = self.rewards[self.state[0]][self.state[1]] + self.action_penalty  # Calculamos el reward (recompensa) por la acción tomada\n",
    "        self.total_reward += reward                                                # Sumamos el reward (recompensa) total del episodio\n",
    "        return self.state, reward, is_final_state                                  # Devolvemos es estado, el reward (recompensa) y si hemos llegado al estado final\n",
    "\n",
    "    def __apply_action(self, action):\n",
    "        \"\"\"\n",
    "        Método que calcula el nuevo estado a partir de la acción a ejecutar\n",
    "        :param action:    Acción a ejecutar\n",
    "        \"\"\"\n",
    "        self.state[0] += self.actions[action][0]\n",
    "        self.state[1] += self.actions[action][1]\n",
    "\n",
    "        # Si nos salimos del tablero por arriba o por abajo, nos quedamos en la posicion que estabamos\n",
    "        if self.state[0] < 0:\n",
    "            self.state[0] = 0\n",
    "        elif self.state[0] > len(self.rewards) - 1:\n",
    "            self.state[0] -= 1\n",
    "\n",
    "        # Si nos salimos del tablero por los lados, nos quedamos en la posicion que estabamos\n",
    "        if self.state[1] < 0:\n",
    "            self.state[1] = 0\n",
    "        elif self.state[1] > len(self.rewards[0]) - 1:\n",
    "            self.state[1] -= 1\n",
    "\n",
    "    def print_path_episode(self):\n",
    "        \"\"\"\n",
    "        Método que imprime por pantalla el camino seguido por el agente\n",
    "        :return: \n",
    "        \"\"\"\n",
    "        path = [['-' for _ in range(len(self.rewards))] for _ in range(len(self.rewards[0]))]\n",
    "        path[0][0] = '0'\n",
    "        for index, step in enumerate(self.actions_done):\n",
    "            path[step[0]][step[1]] = str(index + 1)\n",
    "\n",
    "        print(pd.DataFrame(data=np.array([np.array(xi) for xi in path]),\n",
    "                           index=[\"x{}\".format(str(i)) for i in range(len(path))],\n",
    "                           columns=[\"y{}\".format(str(i)) for i in range(len(path[0]))]))\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<hr>\n",
    "\n",
    "\n",
    "## <a name=\"M2\">2.- Implementación de un algoritmo de toma aleatoria de acciones</a>\n",
    "\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "class Learner(object):\n",
    "\n",
    "    def __init__(self, environment, learning_rate=0.1, discount_factor=0.1, ratio_exploration=0.05):\n",
    "        \"\"\"\n",
    "        Clase que implementa un algoritmo de aprendiza por refuerzo\n",
    "        Esta clase implementa un algoritmo de selección aleatoria de acciones\n",
    "        :param environment:         Entorno en el que tomar las acciones\n",
    "        :param learning_rate:       Factor de aprendizaje\n",
    "        :param discount_factor:     Factor de descuento (0=Estrategia a corto plazo, 1=Estrategia a largo plazo)\n",
    "        :param ratio_exploration:   Ratio de exploración\n",
    "        \"\"\"\n",
    "        self.environment = environment\n",
    "        self.q_table = [[[0.0 for _ in self.environment.actions]\n",
    "                         for _ in range(len(self.environment.rewards))]\n",
    "                        for _ in range(len(self.environment.rewards[0]))]\n",
    "        self.learning_rate = learning_rate\n",
    "        self.discount_factor = discount_factor\n",
    "        self.ratio_exploration = ratio_exploration\n",
    "\n",
    "    @property\n",
    "    def name(self):\n",
    "        return 'random'\n",
    "\n",
    "    def get_next_action(self, state):\n",
    "        \"\"\"\n",
    "        Método que selecciona la siguiente acción a tomar:\n",
    "            Aleatoria -> si el ratio de exploración es inferior al umbral\n",
    "            Mejor Acción -> si el ratio de exploración es superior al umbral\n",
    "        :param state:   Estado del agente\n",
    "        :return:        next_action\n",
    "        \"\"\"\n",
    "\n",
    "        if np.random.uniform() < self.ratio_exploration:\n",
    "            # Seleccionamos una opción al azar\n",
    "            next_action = np.random.choice(list(self.environment.actions))\n",
    "        else:\n",
    "            # Seleccionamos la acción que nos de mayor valor. Si hay empate, Seleccionamos una al azar\n",
    "            idx_action = np.random.choice(np.flatnonzero(\n",
    "                self.q_table[state[0]][state[1]] == np.array(self.q_table[state[0]][state[1]]).max()\n",
    "            ))\n",
    "            next_action = list(self.environment.actions)[idx_action]\n",
    "\n",
    "        return next_action\n",
    "\n",
    "    def update(self, **kwargs):\n",
    "        \"\"\"\n",
    "        Actualiza la Q-Table\n",
    "        :param kwargs: \n",
    "        \"\"\"\n",
    "        pass\n",
    "\n",
    "    def print_q_table(self):\n",
    "        \"\"\"\n",
    "        Método que imprime por pantalla la Q-Table\n",
    "        \"\"\"\n",
    "        q_table = []\n",
    "        for x, row in enumerate(self.q_table):\n",
    "            for y, col in enumerate(row):\n",
    "                q = deepcopy(col)\n",
    "                q.insert(0, 'x{},y{}'.format(x,y))\n",
    "                q_table.append(q)\n",
    "        print(pd.DataFrame(data=q_table,\n",
    "                           columns=['Estado', 'Arriba', 'Abajo', 'Izquierda', 'Derecha'])\n",
    "              .to_string(index=False))\n",
    "\n",
    "    def print_best_actions_states(self):\n",
    "        \"\"\"\n",
    "        Método que imprime por pantalla la mejor opción a realizar en cada uno de los estados\n",
    "        \"\"\"\n",
    "\n",
    "        best = [[list(self.environment.actions)[np.argmax(col)] for col in row] for row in self.q_table]\n",
    "        print(pd.DataFrame(data=np.array([np.array(xi) for xi in best]),\n",
    "                           index=[\"x{}\".format(str(i)) for i in range(len(best))],\n",
    "                           columns=[\"y{}\".format(str(i)) for i in range(len(best[0]))]))\n",
    "        \n",
    "    def print_best_values_states(self):\n",
    "        \"\"\"\n",
    "        Método que imprime por pantalla el valor de la mejor opción a realizar en cada uno de los estados\n",
    "        \"\"\"\n",
    "        best = [[max(vi) for vi in row] for row in self.q_table]\n",
    "        print(pd.DataFrame(data=np.array([np.array(xi) for xi in best]),\n",
    "                           index=[\"x{}\".format(str(i)) for i in range(len(best))],\n",
    "                           columns=[\"y{}\".format(str(i)) for i in range(len(best[0]))]))\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<hr>\n",
    "\n",
    "\n",
    "## <a name=\"M3\">3.- Ejecución: Entorno - Agente</a>\n",
    "\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "scrolled": false
   },
   "outputs": [],
   "source": [
    "from copy import deepcopy\n",
    "\n",
    "\n",
    "def run_agent(learner=Learner, num_episodes=10, learning_rate=0.1, discount_factor=0.1, ratio_exploration=0.05,\n",
    "              verbose=False):\n",
    "    \"\"\"\n",
    "    Método que ejecuta el proceso de aprendizaje del agente en un entorno\n",
    "    :param learner:              Algoritmo de Aprendizaje\n",
    "    :param num_episodes:         Número de veces que se ejecuta (o aprende) el agente en el entorno\n",
    "    :param learning_rate:        Factor de Aprendizaje\n",
    "    :param discount_factor:      Factor de descuento (0=Estrategia a corto plazo, 1=Estrategia a largo plazo)\n",
    "    :param ratio_exploration:    Ratio de exploración\n",
    "    :param verbose:              Boolean, si queremos o no imprimir por pantalla información del proceso\n",
    "    :return:                     (episodes_list, best_episode)\n",
    "    \"\"\"\n",
    "\n",
    "    # Instanciamos el entorno\n",
    "    environment = Environment()\n",
    "\n",
    "    # Instanciamos el método de aprendizaje\n",
    "    learner = learner(environment=environment,\n",
    "                      learning_rate=learning_rate,\n",
    "                      discount_factor=discount_factor,\n",
    "                      ratio_exploration=ratio_exploration)\n",
    "\n",
    "    # Variables para guardar la información de los episodios\n",
    "    episodes_list = []\n",
    "    best_reward = float('-inf')\n",
    "    best_episode = None\n",
    "\n",
    "    for n_episode in range(0, num_episodes):\n",
    "        state = environment.reset()\n",
    "        reward = None\n",
    "        is_final_state = None\n",
    "        num_steps_episode = 0\n",
    "        while not is_final_state:\n",
    "            old_state = state[:]\n",
    "            next_action = learner.get_next_action(state=state)               # Accion a realizar; bien sea explotando la Q-Table o explorando\n",
    "            state, reward, is_final_state = environment.step(next_action)    # Realizamos la accion\n",
    "            next_post_action = (learner.get_next_action(state)               # Si usamos el SARSA realizamos una accion en el nuevo estado S_t+1\n",
    "                                if learner.name == 'SARSA' else None)\n",
    "            learner.update(environment=environment,                          # Actualizamos el entorno\n",
    "                           old_state=old_state,\n",
    "                           action_taken=next_action,\n",
    "                           reward_action_taken=reward,\n",
    "                           new_state=state,\n",
    "                           new_action=next_post_action,\n",
    "                           is_final_state=is_final_state)\n",
    "            num_steps_episode += 1                                           # Sumamos un paso al episocio\n",
    "\n",
    "        episodes_list.append([n_episode + 1,                                 # Guardamos la información del episodio\n",
    "                              num_steps_episode, \n",
    "                              environment.total_reward])\n",
    "\n",
    "        # Guardamos el mejor episodio\n",
    "        if environment.total_reward >= best_reward:\n",
    "            best_reward = environment.total_reward\n",
    "            best_episode = {'num_episode': n_episode + 1,\n",
    "                            'episode': deepcopy(environment),\n",
    "                            'learner': deepcopy(learner)}\n",
    "\n",
    "        if verbose:\n",
    "            # Imprimimos la información de los episodios\n",
    "            print('EPISODIO {} - Numero de acciones: {} - Reward: {}'\n",
    "                  .format(n_episode + 1, num_steps_episode, environment.total_reward))\n",
    "\n",
    "    return episodes_list, best_episode\n",
    "\n",
    "\n",
    "def print_process_info(episodes_list, best_episode, print_best_episode_info=True,\n",
    "                       print_q_table=True, print_best_values_states=True,\n",
    "                       print_best_actions_states=True, print_steps=True, print_path=True):\n",
    "    \"\"\"\n",
    "    Método que imprime por pantalla los resultados de la ejecución\n",
    "    \"\"\"\n",
    "    if print_best_episode_info:\n",
    "        print('\\nMEJOR (ÚLTIMO) EPISODIO:\\nEPISODIO {}\\n\\tNumero de acciones: {}\\n\\tReward: {}'\n",
    "              .format(best_episode['num_episode'],\n",
    "                      len(best_episode['episode'].actions_done),\n",
    "                      best_episode['episode'].total_reward))\n",
    "\n",
    "    if print_q_table:\n",
    "        print('\\nQ_TABLE:')\n",
    "        best_episode['learner'].print_q_table()\n",
    "        \n",
    "    if print_best_values_states:\n",
    "        print('\\nBEST Q_TABLE VALUES:')\n",
    "        best_episode['learner'].print_best_values_states()\n",
    "\n",
    "    if print_best_actions_states:\n",
    "        print('\\nBEST ACTIONS:')\n",
    "        best_episode['learner'].print_best_actions_states()\n",
    "\n",
    "    if print_steps:\n",
    "        print('\\nPasos: \\n   {}'.format(best_episode['episode'].actions_done))\n",
    "\n",
    "    if print_path:\n",
    "        print('\\nPATH:')\n",
    "        best_episode['episode'].print_path_episode()\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Política Aleatoria"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "EPISODIO 1 - Numero de acciones: 264 - Reward: -864.0\n",
      "EPISODIO 2 - Numero de acciones: 84 - Reward: -184.0\n",
      "EPISODIO 3 - Numero de acciones: 16 - Reward: -216.0\n",
      "EPISODIO 4 - Numero de acciones: 42 - Reward: 58.0\n",
      "EPISODIO 5 - Numero de acciones: 44 - Reward: -44.0\n",
      "EPISODIO 6 - Numero de acciones: 36 - Reward: 64.0\n",
      "EPISODIO 7 - Numero de acciones: 29 - Reward: 71.0\n",
      "EPISODIO 8 - Numero de acciones: 11 - Reward: -11.0\n",
      "EPISODIO 9 - Numero de acciones: 73 - Reward: -373.0\n",
      "EPISODIO 10 - Numero de acciones: 37 - Reward: -137.0\n",
      "\n",
      "MEJOR (ÚLTIMO) EPISODIO:\n",
      "EPISODIO 7\n",
      "\tNumero de acciones: 29\n",
      "\tReward: 71.0\n",
      "\n",
      "Q_TABLE:\n",
      "Estado  Arriba  Abajo  Izquierda  Derecha\n",
      " x0,y0    0.00   0.00       0.00     0.00\n",
      " x0,y1    0.00   0.00       0.00     0.00\n",
      " x0,y2    0.00   0.00       0.00     0.00\n",
      " x0,y3    0.00   0.00       0.00     0.00\n",
      " x1,y0    0.00   0.00       0.00     0.00\n",
      " x1,y1    0.00   0.00       0.00     0.00\n",
      " x1,y2    0.00   0.00       0.00     0.00\n",
      " x1,y3    0.00   0.00       0.00     0.00\n",
      " x2,y0    0.00   0.00       0.00     0.00\n",
      " x2,y1    0.00   0.00       0.00     0.00\n",
      " x2,y2    0.00   0.00       0.00     0.00\n",
      " x2,y3    0.00   0.00       0.00     0.00\n",
      " x3,y0    0.00   0.00       0.00     0.00\n",
      " x3,y1    0.00   0.00       0.00     0.00\n",
      " x3,y2    0.00   0.00       0.00     0.00\n",
      " x3,y3    0.00   0.00       0.00     0.00\n",
      "\n",
      "BEST Q_TABLE VALUES:\n",
      "     y0   y1   y2   y3\n",
      "x0 0.00 0.00 0.00 0.00\n",
      "x1 0.00 0.00 0.00 0.00\n",
      "x2 0.00 0.00 0.00 0.00\n",
      "x3 0.00 0.00 0.00 0.00\n",
      "\n",
      "BEST ACTIONS:\n",
      "        y0      y1      y2      y3\n",
      "x0  Arriba  Arriba  Arriba  Arriba\n",
      "x1  Arriba  Arriba  Arriba  Arriba\n",
      "x2  Arriba  Arriba  Arriba  Arriba\n",
      "x3  Arriba  Arriba  Arriba  Arriba\n",
      "\n",
      "Pasos: \n",
      "   [[0, 0], [1, 0], [1, 1], [1, 0], [1, 0], [0, 0], [1, 0], [1, 0], [1, 0], [2, 0], [2, 1], [3, 1], [2, 1], [2, 0], [2, 0], [2, 0], [2, 1], [1, 1], [1, 0], [0, 0], [0, 1], [0, 2], [0, 3], [0, 2], [0, 3], [0, 3], [1, 3], [2, 3], [3, 3]]\n",
      "\n",
      "PATH:\n",
      "    y0  y1  y2  y3\n",
      "x0  20  21  24  26\n",
      "x1  19  18   -  27\n",
      "x2  16  17   -  28\n",
      "x3   -  12   -  29\n"
     ]
    }
   ],
   "source": [
    "episodes_list, best_episode = run_agent(learner=Learner, \n",
    "                                        verbose=True)\n",
    "\n",
    "print_process_info(episodes_list=episodes_list, \n",
    "                   best_episode=best_episode)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<hr>\n",
    "\n",
    "\n",
    "## <a name=\"M4\">4.- Q-Learner: Implementación y Ejecución</a>\n",
    "\n",
    "\n",
    "* Recordemos el Pseudocódigo del Algoritmo:\n",
    "\n",
    "\n",
    "<img src=\"./imgs/013_qlearning.png\" style=\"width: 600px;\"/>\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "class QLearner(Learner):\n",
    "\n",
    "    @property\n",
    "    def name(self):\n",
    "        return 'QLearner'\n",
    "\n",
    "    def update(self, environment, old_state, action_taken, reward_action_taken, new_state, is_final_state, **kwargs):\n",
    "        \"\"\"\n",
    "        Método que implementa el Algoritmo de aprendizaje del Q-Learning\n",
    "        :param environment:           Entorno en el que tomar las acciones\n",
    "        :param old_state:             Estado actual\n",
    "        :param action_taken:          Acción a realizar\n",
    "        :param reward_action_taken:   Recompensa obtenida por la acción tomada\n",
    "        :param new_state:             Nuevo estado al que se mueve el agente\n",
    "        :param is_final_state:        Boolean. Devuelve True si el agente llega al estado final; si no, False\n",
    "        :param kwargs: \n",
    "        \"\"\"\n",
    "        # Obtengo el identificador de la acción\n",
    "        idx_action_taken = list(environment.actions).index(action_taken)\n",
    "\n",
    "        # Obtengo el valor de la acción tomada\n",
    "        actual_q_value_options = self.q_table[old_state[0]][old_state[1]]\n",
    "        actual_q_value = actual_q_value_options[idx_action_taken]\n",
    "\n",
    "        future_q_value_options = self.q_table[new_state[0]][new_state[1]]\n",
    "        future_max_q_value = reward_action_taken + self.discount_factor * max(future_q_value_options)\n",
    "        if is_final_state:\n",
    "            future_max_q_value = reward_action_taken    # Reward máximo si llego a la posición final\n",
    "\n",
    "        self.q_table[old_state[0]][old_state[1]][idx_action_taken] = \\\n",
    "            actual_q_value + self.learning_rate * (future_max_q_value - actual_q_value)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Estrategia a corto plazo"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "EPISODIO 1 - Numero de acciones: 50 - Reward: -50.0\n",
      "EPISODIO 2 - Numero de acciones: 11 - Reward: -11.0\n",
      "EPISODIO 3 - Numero de acciones: 12 - Reward: 88.0\n",
      "EPISODIO 4 - Numero de acciones: 42 - Reward: 58.0\n",
      "EPISODIO 5 - Numero de acciones: 17 - Reward: 83.0\n",
      "EPISODIO 6 - Numero de acciones: 32 - Reward: 68.0\n",
      "EPISODIO 7 - Numero de acciones: 30 - Reward: 70.0\n",
      "EPISODIO 8 - Numero de acciones: 13 - Reward: 87.0\n",
      "EPISODIO 9 - Numero de acciones: 8 - Reward: 92.0\n",
      "EPISODIO 10 - Numero de acciones: 34 - Reward: 66.0\n",
      "EPISODIO 11 - Numero de acciones: 12 - Reward: 88.0\n",
      "EPISODIO 12 - Numero de acciones: 54 - Reward: 46.0\n",
      "EPISODIO 13 - Numero de acciones: 15 - Reward: 85.0\n",
      "EPISODIO 14 - Numero de acciones: 16 - Reward: 84.0\n",
      "EPISODIO 15 - Numero de acciones: 41 - Reward: 59.0\n",
      "EPISODIO 16 - Numero de acciones: 13 - Reward: 87.0\n",
      "EPISODIO 17 - Numero de acciones: 10 - Reward: 90.0\n",
      "EPISODIO 18 - Numero de acciones: 13 - Reward: -13.0\n",
      "EPISODIO 19 - Numero de acciones: 30 - Reward: 70.0\n",
      "EPISODIO 20 - Numero de acciones: 18 - Reward: 82.0\n",
      "EPISODIO 21 - Numero de acciones: 17 - Reward: 83.0\n",
      "EPISODIO 22 - Numero de acciones: 22 - Reward: 78.0\n",
      "EPISODIO 23 - Numero de acciones: 22 - Reward: 78.0\n",
      "EPISODIO 24 - Numero de acciones: 12 - Reward: 88.0\n",
      "EPISODIO 25 - Numero de acciones: 14 - Reward: 86.0\n",
      "\n",
      "MEJOR (ÚLTIMO) EPISODIO:\n",
      "EPISODIO 9\n",
      "\tNumero de acciones: 8\n",
      "\tReward: 92.0\n",
      "\n",
      "Q_TABLE:\n",
      "Estado  Arriba  Abajo  Izquierda  Derecha\n",
      " x0,y0   -0.64  -0.63      -0.64    -0.67\n",
      " x0,y1   -0.42  -0.48      -0.42    -0.47\n",
      " x0,y2   -0.27  -0.27      -0.28    -0.35\n",
      " x0,y3   -0.19  -0.27      -0.19    -0.19\n",
      " x1,y0   -0.42  -0.48      -0.42    -0.48\n",
      " x1,y1   -0.35  -0.35      -0.35    -0.41\n",
      " x1,y2   -0.19  -0.27      -0.19    -0.27\n",
      " x1,y3   -0.10   0.75      -0.10    -0.10\n",
      " x2,y0   -0.35  -0.35      -0.35    -0.41\n",
      " x2,y1   -0.27  -0.27      -0.28    -0.34\n",
      " x2,y2   -0.10 -10.10      -0.10     0.59\n",
      " x2,y3   -0.10  56.38       0.00    -0.10\n",
      " x3,y0   -0.35  -0.35      -0.35    -0.35\n",
      " x3,y1   -0.27  -0.35      -0.35   -10.10\n",
      " x3,y2    0.00   0.00      -0.10     9.90\n",
      " x3,y3    0.00   0.00       0.00     0.00\n",
      "\n",
      "BEST Q_TABLE VALUES:\n",
      "      y0    y1    y2    y3\n",
      "x0 -0.63 -0.42 -0.27 -0.19\n",
      "x1 -0.42 -0.35 -0.19  0.75\n",
      "x2 -0.35 -0.27  0.59 56.38\n",
      "x3 -0.35 -0.27  9.90  0.00\n",
      "\n",
      "BEST ACTIONS:\n",
      "           y0      y1         y2      y3\n",
      "x0      Abajo  Arriba      Abajo  Arriba\n",
      "x1  Izquierda   Abajo  Izquierda   Abajo\n",
      "x2      Abajo   Abajo    Derecha   Abajo\n",
      "x3      Abajo  Arriba    Derecha  Arriba\n",
      "\n",
      "Pasos: \n",
      "   [[0, 0], [0, 0], [0, 1], [1, 1], [1, 2], [1, 3], [2, 3], [3, 3]]\n",
      "\n",
      "PATH:\n",
      "   y0 y1 y2 y3\n",
      "x0  2  3  -  -\n",
      "x1  -  4  5  6\n",
      "x2  -  -  -  7\n",
      "x3  -  -  -  8\n"
     ]
    }
   ],
   "source": [
    "episodes_list, best_episode = run_agent(learner=QLearner,\n",
    "                                        num_episodes=25,\n",
    "                                        learning_rate=0.1,\n",
    "                                        discount_factor=0.1,\n",
    "                                        ratio_exploration=0.05,\n",
    "                                        verbose=True)\n",
    "\n",
    "print_process_info(episodes_list=episodes_list,\n",
    "                   best_episode=best_episode)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Estrategia a largo plazo"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "EPISODIO 1 - Numero de acciones: 81 - Reward: -181.0\n",
      "EPISODIO 2 - Numero de acciones: 35 - Reward: -35.0\n",
      "EPISODIO 3 - Numero de acciones: 25 - Reward: 75.0\n",
      "EPISODIO 4 - Numero de acciones: 21 - Reward: 79.0\n",
      "EPISODIO 5 - Numero de acciones: 19 - Reward: 81.0\n",
      "EPISODIO 6 - Numero de acciones: 27 - Reward: 73.0\n",
      "EPISODIO 7 - Numero de acciones: 13 - Reward: 87.0\n",
      "EPISODIO 8 - Numero de acciones: 25 - Reward: 75.0\n",
      "EPISODIO 9 - Numero de acciones: 6 - Reward: 94.0\n",
      "EPISODIO 10 - Numero de acciones: 11 - Reward: 89.0\n",
      "EPISODIO 11 - Numero de acciones: 11 - Reward: 89.0\n",
      "EPISODIO 12 - Numero de acciones: 6 - Reward: 94.0\n",
      "EPISODIO 13 - Numero de acciones: 13 - Reward: 87.0\n",
      "EPISODIO 14 - Numero de acciones: 18 - Reward: 82.0\n",
      "EPISODIO 15 - Numero de acciones: 8 - Reward: 92.0\n",
      "EPISODIO 16 - Numero de acciones: 6 - Reward: 94.0\n",
      "EPISODIO 17 - Numero de acciones: 11 - Reward: 89.0\n",
      "EPISODIO 18 - Numero de acciones: 6 - Reward: 94.0\n",
      "EPISODIO 19 - Numero de acciones: 8 - Reward: 92.0\n",
      "EPISODIO 20 - Numero de acciones: 6 - Reward: 94.0\n",
      "EPISODIO 21 - Numero de acciones: 6 - Reward: 94.0\n",
      "EPISODIO 22 - Numero de acciones: 6 - Reward: 94.0\n",
      "EPISODIO 23 - Numero de acciones: 6 - Reward: 94.0\n",
      "EPISODIO 24 - Numero de acciones: 8 - Reward: 92.0\n",
      "EPISODIO 25 - Numero de acciones: 6 - Reward: 94.0\n",
      "\n",
      "MEJOR (ÚLTIMO) EPISODIO:\n",
      "EPISODIO 25\n",
      "\tNumero de acciones: 6\n",
      "\tReward: 94.0\n",
      "\n",
      "Q_TABLE:\n",
      "Estado  Arriba  Abajo  Izquierda  Derecha\n",
      " x0,y0   -1.31  -1.31      -1.31    -1.16\n",
      " x0,y1   -0.77   1.39      -0.89    -0.77\n",
      " x0,y2   -0.49  -0.47      -0.53    -0.53\n",
      " x0,y3   -0.30   0.13      -0.39    -0.30\n",
      " x1,y0   -0.91  -0.79      -0.86    -0.09\n",
      " x1,y1   -0.51  10.63      -0.51    -0.47\n",
      " x1,y2   -0.32   1.85      -0.31    -0.29\n",
      " x1,y3   -0.20  10.25      -0.20    -0.20\n",
      " x2,y0   -0.53  -0.56      -0.59    -0.27\n",
      " x2,y1   -0.40  -0.37      -0.40    31.99\n",
      " x2,y2   -0.21 -10.10       2.21    60.48\n",
      " x2,y3   -0.10  91.10      -0.10    -0.10\n",
      " x3,y0   -0.51  -0.49      -0.49    -0.48\n",
      " x3,y1   -0.33  -0.39      -0.40    -9.21\n",
      " x3,y2   -0.11 -10.10       0.00     9.90\n",
      " x3,y3    0.00   0.00       0.00     0.00\n",
      "\n",
      "BEST Q_TABLE VALUES:\n",
      "      y0    y1    y2    y3\n",
      "x0 -1.16  1.39 -0.47  0.13\n",
      "x1 -0.09 10.63  1.85 10.25\n",
      "x2 -0.27 31.99 60.48 91.10\n",
      "x3 -0.48 -0.33  9.90  0.00\n",
      "\n",
      "BEST ACTIONS:\n",
      "         y0       y1       y2      y3\n",
      "x0  Derecha    Abajo    Abajo   Abajo\n",
      "x1  Derecha    Abajo    Abajo   Abajo\n",
      "x2  Derecha  Derecha  Derecha   Abajo\n",
      "x3  Derecha   Arriba  Derecha  Arriba\n",
      "\n",
      "Pasos: \n",
      "   [[0, 1], [1, 1], [2, 1], [2, 2], [2, 3], [3, 3]]\n",
      "\n",
      "PATH:\n",
      "   y0 y1 y2 y3\n",
      "x0  0  1  -  -\n",
      "x1  -  2  -  -\n",
      "x2  -  3  4  5\n",
      "x3  -  -  -  6\n"
     ]
    }
   ],
   "source": [
    "episodes_list, best_episode = run_agent(learner=QLearner,\n",
    "                                        num_episodes=25,\n",
    "                                        learning_rate=0.1,\n",
    "                                        discount_factor=0.9,\n",
    "                                        ratio_exploration=0.05,\n",
    "                                        verbose=True)\n",
    "\n",
    "print_process_info(episodes_list=episodes_list,\n",
    "                   best_episode=best_episode)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<hr>\n",
    "\n",
    "\n",
    "## <a name=\"M5\">5.- SARSA-Learner: Implementación y Ejecución</a>\n",
    "\n",
    "\n",
    "* Recordemos el Pseudocódigo del Algoritmo:\n",
    "\n",
    "\n",
    "<img src=\"./imgs/014_sarsa.png\" style=\"width: 600px;\"/>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "class SARSALearner(Learner):\n",
    "\n",
    "    @property\n",
    "    def name(self):\n",
    "        return 'SARSA'\n",
    "\n",
    "    def update(self, environment, old_state, action_taken, reward_action_taken, new_state, new_action, is_final_state):\n",
    "        \"\"\"\n",
    "        Método que implementa el algoritmo de aprendizaje SARSA\n",
    "        :param environment:           Entorno en el que tomar las acciones\n",
    "        :param old_state:             Estado actual\n",
    "        :param action_taken:          Acción a realizar\n",
    "        :param reward_action_taken:   Recompensa obtenida por la acción tomada\n",
    "        :param new_state:             Nuevo estado al que se mueve el agente \n",
    "        :param new_action:            Acción a tomar en el nuevo estado\n",
    "        :param is_final_state:        Boolean. Devuelve True si el agente llega al estado final; si no, False \n",
    "        \"\"\"\n",
    "        # Obtengo el identificador de la acción\n",
    "        idx_action_taken = list(environment.actions).index(action_taken)\n",
    "\n",
    "        # Obtengo el valor de la acción tomada\n",
    "        actual_q_value_options = self.q_table[old_state[0]][old_state[1]]\n",
    "        actual_q_value = actual_q_value_options[idx_action_taken]\n",
    "\n",
    "        future_q_value_options = self.q_table[new_state[0]][new_state[1]]\n",
    "\n",
    "        idx_new_action_taken = list(environment.actions).index(new_action)\n",
    "        future_new_action_q_value = \\\n",
    "            reward_action_taken + self.discount_factor * future_q_value_options[idx_new_action_taken]\n",
    "        if is_final_state:\n",
    "            # Reward máximo si llego a la posición final\n",
    "            future_new_action_q_value = reward_action_taken\n",
    "\n",
    "        self.q_table[old_state[0]][old_state[1]][idx_action_taken] = \\\n",
    "            actual_q_value + self.learning_rate * (future_new_action_q_value - actual_q_value)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Estrategia a corto plazo"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "EPISODIO 1 - Numero de acciones: 22 - Reward: -22.0\n",
      "EPISODIO 2 - Numero de acciones: 37 - Reward: 63.0\n",
      "EPISODIO 3 - Numero de acciones: 28 - Reward: -28.0\n",
      "EPISODIO 4 - Numero de acciones: 11 - Reward: 89.0\n",
      "EPISODIO 5 - Numero de acciones: 7 - Reward: 93.0\n",
      "EPISODIO 6 - Numero de acciones: 26 - Reward: 74.0\n",
      "EPISODIO 7 - Numero de acciones: 42 - Reward: 58.0\n",
      "EPISODIO 8 - Numero de acciones: 28 - Reward: 72.0\n",
      "EPISODIO 9 - Numero de acciones: 49 - Reward: 51.0\n",
      "EPISODIO 10 - Numero de acciones: 13 - Reward: 87.0\n",
      "EPISODIO 11 - Numero de acciones: 12 - Reward: 88.0\n",
      "EPISODIO 12 - Numero de acciones: 8 - Reward: 92.0\n",
      "EPISODIO 13 - Numero de acciones: 75 - Reward: 25.0\n",
      "EPISODIO 14 - Numero de acciones: 6 - Reward: 94.0\n",
      "EPISODIO 15 - Numero de acciones: 10 - Reward: 90.0\n",
      "EPISODIO 16 - Numero de acciones: 29 - Reward: 71.0\n",
      "EPISODIO 17 - Numero de acciones: 15 - Reward: 85.0\n",
      "EPISODIO 18 - Numero de acciones: 14 - Reward: 86.0\n",
      "EPISODIO 19 - Numero de acciones: 28 - Reward: 72.0\n",
      "EPISODIO 20 - Numero de acciones: 23 - Reward: 77.0\n",
      "EPISODIO 21 - Numero de acciones: 18 - Reward: 82.0\n",
      "EPISODIO 22 - Numero de acciones: 17 - Reward: 83.0\n",
      "EPISODIO 23 - Numero de acciones: 8 - Reward: 92.0\n",
      "EPISODIO 24 - Numero de acciones: 11 - Reward: 89.0\n",
      "EPISODIO 25 - Numero de acciones: 19 - Reward: 81.0\n",
      "\n",
      "MEJOR (ÚLTIMO) EPISODIO:\n",
      "EPISODIO 14\n",
      "\tNumero de acciones: 6\n",
      "\tReward: 94.0\n",
      "\n",
      "Q_TABLE:\n",
      "Estado  Arriba  Abajo  Izquierda  Derecha\n",
      " x0,y0   -0.87  -0.87      -0.87    -0.87\n",
      " x0,y1   -0.68  -0.67      -0.69    -0.71\n",
      " x0,y2   -0.48  -0.48      -0.49    -0.53\n",
      " x0,y3   -0.35  -0.40      -0.35    -0.35\n",
      " x1,y0   -0.69  -0.71      -0.68    -0.67\n",
      " x1,y1   -0.59  -0.58      -0.54    -0.53\n",
      " x1,y2   -0.35  -0.34      -0.35    -0.34\n",
      " x1,y3   -0.19   1.36      -0.10    -0.10\n",
      " x2,y0   -0.54  -0.53      -0.54    -0.58\n",
      " x2,y1   -0.48  -0.42      -0.42    -0.45\n",
      " x2,y2   -0.19 -10.10      -0.10     1.71\n",
      " x2,y3    0.00  71.04      -0.10    -0.10\n",
      " x3,y0   -0.48  -0.48      -0.48    -0.48\n",
      " x3,y1   -0.42  -0.42      -0.42   -10.00\n",
      " x3,y2    0.00   0.00       0.00    18.81\n",
      " x3,y3    0.00   0.00       0.00     0.00\n",
      "\n",
      "BEST Q_TABLE VALUES:\n",
      "      y0    y1    y2    y3\n",
      "x0 -0.87 -0.67 -0.48 -0.35\n",
      "x1 -0.67 -0.53 -0.34  1.36\n",
      "x2 -0.53 -0.42  1.71 71.04\n",
      "x3 -0.48 -0.42 18.81  0.00\n",
      "\n",
      "BEST ACTIONS:\n",
      "         y0       y1       y2      y3\n",
      "x0   Arriba    Abajo    Abajo  Arriba\n",
      "x1  Derecha  Derecha    Abajo   Abajo\n",
      "x2    Abajo    Abajo  Derecha   Abajo\n",
      "x3  Derecha    Abajo  Derecha  Arriba\n",
      "\n",
      "Pasos: \n",
      "   [[1, 0], [2, 0], [2, 1], [2, 2], [2, 3], [3, 3]]\n",
      "\n",
      "PATH:\n",
      "   y0 y1 y2 y3\n",
      "x0  0  -  -  -\n",
      "x1  1  -  -  -\n",
      "x2  2  3  4  5\n",
      "x3  -  -  -  6\n"
     ]
    }
   ],
   "source": [
    "episodes_list, best_episode = run_agent(learner=SARSALearner,\n",
    "                                        num_episodes=25,\n",
    "                                        learning_rate=0.1,\n",
    "                                        discount_factor=0.1,\n",
    "                                        ratio_exploration=0.05,\n",
    "                                        verbose=True)\n",
    "\n",
    "print_process_info(episodes_list=episodes_list,\n",
    "                   best_episode=best_episode)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Estrategia a largo plazo"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "EPISODIO 1 - Numero de acciones: 55 - Reward: -255.0\n",
      "EPISODIO 2 - Numero de acciones: 15 - Reward: 85.0\n",
      "EPISODIO 3 - Numero de acciones: 47 - Reward: 53.0\n",
      "EPISODIO 4 - Numero de acciones: 12 - Reward: 88.0\n",
      "EPISODIO 5 - Numero de acciones: 8 - Reward: 92.0\n",
      "EPISODIO 6 - Numero de acciones: 46 - Reward: 54.0\n",
      "EPISODIO 7 - Numero de acciones: 6 - Reward: 94.0\n",
      "EPISODIO 8 - Numero de acciones: 16 - Reward: 84.0\n",
      "EPISODIO 9 - Numero de acciones: 11 - Reward: 89.0\n",
      "EPISODIO 10 - Numero de acciones: 6 - Reward: 94.0\n",
      "EPISODIO 11 - Numero de acciones: 17 - Reward: 83.0\n",
      "EPISODIO 12 - Numero de acciones: 13 - Reward: 87.0\n",
      "EPISODIO 13 - Numero de acciones: 7 - Reward: 93.0\n",
      "EPISODIO 14 - Numero de acciones: 7 - Reward: -7.0\n",
      "EPISODIO 15 - Numero de acciones: 12 - Reward: 88.0\n",
      "EPISODIO 16 - Numero de acciones: 6 - Reward: 94.0\n",
      "EPISODIO 17 - Numero de acciones: 10 - Reward: 90.0\n",
      "EPISODIO 18 - Numero de acciones: 6 - Reward: 94.0\n",
      "EPISODIO 19 - Numero de acciones: 6 - Reward: 94.0\n",
      "EPISODIO 20 - Numero de acciones: 6 - Reward: 94.0\n",
      "EPISODIO 21 - Numero de acciones: 8 - Reward: 92.0\n",
      "EPISODIO 22 - Numero de acciones: 6 - Reward: 94.0\n",
      "EPISODIO 23 - Numero de acciones: 6 - Reward: 94.0\n",
      "EPISODIO 24 - Numero de acciones: 6 - Reward: 94.0\n",
      "EPISODIO 25 - Numero de acciones: 6 - Reward: 94.0\n",
      "\n",
      "MEJOR (ÚLTIMO) EPISODIO:\n",
      "EPISODIO 25\n",
      "\tNumero de acciones: 6\n",
      "\tReward: 94.0\n",
      "\n",
      "Q_TABLE:\n",
      "Estado  Arriba  Abajo  Izquierda  Derecha\n",
      " x0,y0   -1.22  -1.17      -1.22    -1.02\n",
      " x0,y1   -0.68   1.36      -0.66    -0.71\n",
      " x0,y2   -0.39   1.29      -0.43    -0.39\n",
      " x0,y3   -0.30  -0.29      -0.30    -0.30\n",
      " x1,y0   -0.69  -0.70      -0.77    -0.02\n",
      " x1,y1   -0.42  -0.32      -0.44     9.27\n",
      " x1,y2   -0.21  26.66      -0.10    -0.20\n",
      " x1,y3   -0.19  -0.19      -0.15    -0.20\n",
      " x2,y0   -0.54  -0.47      -0.49    -0.45\n",
      " x2,y1   -0.31  -1.10      -0.30     4.41\n",
      " x2,y2   -0.11 -18.30      -0.11    55.77\n",
      " x2,y3   -0.10  90.23      -0.10     0.00\n",
      " x3,y0   -0.41  -0.39      -0.39    -0.39\n",
      " x3,y1   -0.30  -0.30      -0.31   -10.10\n",
      " x3,y2    0.00 -10.10      -0.10    18.81\n",
      " x3,y3    0.00   0.00       0.00     0.00\n",
      "\n",
      "BEST Q_TABLE VALUES:\n",
      "      y0    y1    y2    y3\n",
      "x0 -1.02  1.36  1.29 -0.29\n",
      "x1 -0.02  9.27 26.66 -0.15\n",
      "x2 -0.45  4.41 55.77 90.23\n",
      "x3 -0.39 -0.30 18.81  0.00\n",
      "\n",
      "BEST ACTIONS:\n",
      "         y0       y1       y2         y3\n",
      "x0  Derecha    Abajo    Abajo      Abajo\n",
      "x1  Derecha  Derecha    Abajo  Izquierda\n",
      "x2  Derecha  Derecha  Derecha      Abajo\n",
      "x3  Derecha   Arriba  Derecha     Arriba\n",
      "\n",
      "Pasos: \n",
      "   [[0, 1], [1, 1], [1, 2], [2, 2], [2, 3], [3, 3]]\n",
      "\n",
      "PATH:\n",
      "   y0 y1 y2 y3\n",
      "x0  0  1  -  -\n",
      "x1  -  2  3  -\n",
      "x2  -  -  4  5\n",
      "x3  -  -  -  6\n"
     ]
    }
   ],
   "source": [
    "episodes_list, best_episode = run_agent(learner=SARSALearner,\n",
    "                                        num_episodes=25,\n",
    "                                        learning_rate=0.1,\n",
    "                                        discount_factor=0.9,\n",
    "                                        ratio_exploration=0.05,\n",
    "                                        verbose=True)\n",
    "\n",
    "print_process_info(episodes_list=episodes_list,\n",
    "                   best_episode=best_episode)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<hr>\n",
    "\n",
    "\n",
    "## <a name=\"M6\">6.- Estrategias a corto y largo plazo</a>\n",
    "\n",
    "\n",
    "### Q-Learning"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x576 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "import warnings\n",
    "\n",
    "warnings.filterwarnings(\"ignore\")\n",
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline\n",
    "\n",
    "discount_factors = [0.01, 0.3, 0.7, 0.99]\n",
    "experiments = []\n",
    "for disfac in discount_factors:\n",
    "    episodes_list, best_episode = run_agent(learner=QLearner,\n",
    "                                            num_episodes=30,\n",
    "                                            learning_rate=0.1,\n",
    "                                            discount_factor=disfac,\n",
    "                                            ratio_exploration=0.05)\n",
    "    experiments.append({'Discount Factor': str(disfac),\n",
    "                        'Rewards': [episode[2] for episode in episodes_list]})\n",
    "\n",
    "plt.figure(figsize=(20, 8))\n",
    "for experiment in experiments:\n",
    "    plt.subplot(1, 2, 1)\n",
    "    plt.title('Evolución Reward por Episodios VS Discount Factor')\n",
    "    plt.plot(experiment['Rewards'], label=experiment['Discount Factor'])\n",
    "    plt.legend(loc='lower right')\n",
    "\n",
    "    plt.subplot(1, 2, 2)\n",
    "    plt.title('Evolución Acumulada Reward por Episodios VS Discount Factor')\n",
    "    acum_avg_reward = (np.cumsum(experiment['Rewards']) / (np.arange(len(experiment['Rewards'])) + 1))[0:]\n",
    "    plt.plot(acum_avg_reward, label=experiment['Discount Factor'])\n",
    "    plt.legend(loc='lower right')\n",
    "\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### SARSA"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x576 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "discount_factors = [0.01, 0.3, 0.7, 0.99]\n",
    "experiments = []\n",
    "for disfac in discount_factors:\n",
    "    episodes_list, best_episode = run_agent(learner=SARSALearner,\n",
    "                                            num_episodes=30,\n",
    "                                            learning_rate=0.1,\n",
    "                                            discount_factor=disfac,\n",
    "                                            ratio_exploration=0.05)\n",
    "    experiments.append({'Discount Factor': str(disfac),\n",
    "                        'Rewards': [episode[2] for episode in episodes_list]})\n",
    "\n",
    "plt.figure(figsize=(20, 8))\n",
    "for experiment in experiments:\n",
    "    plt.subplot(1, 2, 1)\n",
    "    plt.title('Evolución Reward por Episodios VS Discount Factor')\n",
    "    plt.plot(experiment['Rewards'], label=experiment['Discount Factor'])\n",
    "    plt.legend(loc='lower right')\n",
    "\n",
    "    plt.subplot(1, 2, 2)\n",
    "    plt.title('Evolución Acumulada Reward por Episodios VS Discount Factor')\n",
    "    acum_avg_reward = (np.cumsum(experiment['Rewards']) / (np.arange(len(experiment['Rewards'])) + 1))[0:]\n",
    "    plt.plot(acum_avg_reward, label=experiment['Discount Factor'])\n",
    "    plt.legend(loc='lower right')\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<hr>\n",
    "\n",
    "*Este Notebook ha sido desarrollado por **Ricardo Moya García** y registrado en Safe Creative como ***Atribución-NoComercial-CompartirIgual***.*\n",
    "\n",
    "\n",
    "<a href=\"https://www.safecreative.org/work/2005103928020\" xmlns:cc=\"http://creativecommons.org/ns#\" rel=\"cc:license\"><img src=\"https://resources.safecreative.org/work/2005103928020/label/standard-72\" style=\"border:0;\" alt=\"Safe Creative #2005103928020\"/></a>"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.9"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}