experiment.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# ING Hubs Hackaton\n",
    "\n",
    "29.01.2021\n",
    "\n",
    "Jaime Elguero, Santiago Gil, Borja Serra"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Open AIGym\n",
    "\n",
    "[Gym](https://gym.openai.com/) is a toolkit for developing and comparing reinforcement learning algorithms. It makes no assumptions about the structure of your agent, and is compatible with any numerical computation library, such as TensorFlow or Theano.\n",
    "\n",
    "The gym library is a collection of test problems — environments — that you can use to work out your reinforcement learning algorithms. These environments have a shared interface, allowing you to write general algorithms.\n",
    "\n",
    "[Environments](https://gym.openai.com/envs/#classic_control) "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import gym\n",
    "\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "There is a lot of nice environments to work with:"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Mountain Car environment"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "For this Hackaton, we will Focus on the **MountainCar** one. Still if your team finished before time, feel free to explore any other environment you may like"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Import and initialize Mountain Car Environment\n",
    "env = gym.make('MountainCar-v0')\n",
    "#env.reset()\n",
    "#env.seed(1); np.random.seed(1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Mountain Car** is a classic reinforcement learning problem where the objective is to create an algorithm that learns to climb a steep hill to reach the goal marked by a flag. The car’s engine is not powerful enough to drive up the hill without a head start so the car must drive up the left hill to obtain enough momentum to scale the steeper hill to the right and reach the goal."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<video src=\"mountain_car.mp4\" controls  >\n",
       "      Your browser does not support the <code>video</code> element.\n",
       "    </video>"
      ],
      "text/plain": [
       "<IPython.core.display.Video object>"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from IPython.display import Video\n",
    "\n",
    "Video(\"mountain_car.mp4\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Here we have more detailed information about it, especially focused to inputing a RL algorithm (actions, reward, etc.)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "    Description:\n",
      "        The agent (a car) is started at the bottom of a valley. For any given\n",
      "        state the agent may choose to accelerate to the left, right or cease\n",
      "        any acceleration.\n",
      "\n",
      "    Source:\n",
      "        The environment appeared first in Andrew Moore's PhD Thesis (1990).\n",
      "\n",
      "    Observation:\n",
      "        Type: Box(2)\n",
      "        Num    Observation               Min            Max\n",
      "        0      Car Position              -1.2           0.6\n",
      "        1      Car Velocity              -0.07          0.07\n",
      "\n",
      "    Actions:\n",
      "        Type: Discrete(3)\n",
      "        Num    Action\n",
      "        0      Accelerate to the Left\n",
      "        1      Don't accelerate\n",
      "        2      Accelerate to the Right\n",
      "\n",
      "        Note: This does not affect the amount of velocity affected by the\n",
      "        gravitational pull acting on the car.\n",
      "\n",
      "    Reward:\n",
      "         Reward of 0 is awarded if the agent reached the flag (position = 0.5)\n",
      "         on top of the mountain.\n",
      "         Reward of -1 is awarded if the position of the agent is less than 0.5.\n",
      "\n",
      "    Starting State:\n",
      "         The position of the car is assigned a uniform random value in\n",
      "         [-0.6 , -0.4].\n",
      "         The starting velocity of the car is always assigned to 0.\n",
      "\n",
      "    Episode Termination:\n",
      "         The car position is more than 0.5\n",
      "         Episode length is greater than 200\n",
      "    \n"
     ]
    }
   ],
   "source": [
    "print(gym.envs.classic_control.mountain_car.MountainCarEnv.__doc__)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### A random agent to interact with the environment"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "- The state of the environment is the tuple **(position, velocity)**. As an starting point we will start with a null velocity and somewhere around the center of the valley (what is randomly aprox. -0.5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[-0.45804752  0.        ]\n"
     ]
    }
   ],
   "source": [
    "state = env.reset()\n",
    "print (state)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "- The boundaries of our observation space are the follows:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0.6 , 0.07], dtype=float32)"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "env.observation_space.high"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([-1.2 , -0.07], dtype=float32)"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "env.observation_space.low"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "- And we also have the information of the possible actions we can perform, in this environment only 3 (see doc above):"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Discrete(3)"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "env.action_space"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "- With all this, we have the enough information to navigate through the environment:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "env.reset();\n",
    "\n",
    "for i in range(200):\n",
    "    # We can visually render the current observation\n",
    "    env.render()\n",
    "    # We select a random action\n",
    "    action = np.random.randint(0, env.action_space.n)\n",
    "    # We apply the action, and receive information from the environment\n",
    "    state, reward, done, info = env.step(action)\n",
    "    \n",
    "# We close the environment\n",
    "env.close()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "- A typical RL algorithm will make use of episodes as follows:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "EPISODES = 10000"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Episode 100 Average Reward: -200.0\n",
      "Episode 200 Average Reward: -200.0\n",
      "Episode 300 Average Reward: -200.0\n",
      "Episode 400 Average Reward: -200.0\n",
      "Episode 500 Average Reward: -200.0\n",
      "Episode 600 Average Reward: -200.0\n",
      "Episode 700 Average Reward: -200.0\n",
      "Episode 800 Average Reward: -200.0\n",
      "Episode 900 Average Reward: -200.0\n",
      "Episode 1000 Average Reward: -200.0\n",
      "Episode 1100 Average Reward: -200.0\n",
      "Episode 1200 Average Reward: -200.0\n",
      "Episode 1300 Average Reward: -200.0\n",
      "Episode 1400 Average Reward: -200.0\n",
      "Episode 1500 Average Reward: -200.0\n",
      "Episode 1600 Average Reward: -200.0\n",
      "Episode 1700 Average Reward: -200.0\n",
      "Episode 1800 Average Reward: -200.0\n",
      "Episode 1900 Average Reward: -200.0\n",
      "Episode 2000 Average Reward: -200.0\n",
      "Episode 2100 Average Reward: -200.0\n",
      "Episode 2200 Average Reward: -200.0\n",
      "Episode 2300 Average Reward: -200.0\n",
      "Episode 2400 Average Reward: -200.0\n",
      "Episode 2500 Average Reward: -200.0\n",
      "Episode 2600 Average Reward: -200.0\n",
      "Episode 2700 Average Reward: -200.0\n",
      "Episode 2800 Average Reward: -200.0\n",
      "Episode 2900 Average Reward: -200.0\n",
      "Episode 3000 Average Reward: -200.0\n",
      "Episode 3100 Average Reward: -200.0\n",
      "Episode 3200 Average Reward: -200.0\n",
      "Episode 3300 Average Reward: -200.0\n",
      "Episode 3400 Average Reward: -200.0\n",
      "Episode 3500 Average Reward: -200.0\n",
      "Episode 3600 Average Reward: -200.0\n",
      "Episode 3700 Average Reward: -200.0\n",
      "Episode 3800 Average Reward: -200.0\n",
      "Episode 3900 Average Reward: -200.0\n",
      "Episode 4000 Average Reward: -200.0\n",
      "Episode 4100 Average Reward: -200.0\n",
      "Episode 4200 Average Reward: -200.0\n",
      "Episode 4300 Average Reward: -200.0\n",
      "Episode 4400 Average Reward: -200.0\n",
      "Episode 4500 Average Reward: -200.0\n",
      "Episode 4600 Average Reward: -200.0\n",
      "Episode 4700 Average Reward: -200.0\n",
      "Episode 4800 Average Reward: -200.0\n",
      "Episode 4900 Average Reward: -200.0\n",
      "Episode 5000 Average Reward: -200.0\n",
      "Episode 5100 Average Reward: -200.0\n",
      "Episode 5200 Average Reward: -200.0\n",
      "Episode 5300 Average Reward: -200.0\n",
      "Episode 5400 Average Reward: -200.0\n",
      "Episode 5500 Average Reward: -200.0\n",
      "Episode 5600 Average Reward: -200.0\n",
      "Episode 5700 Average Reward: -200.0\n",
      "Episode 5800 Average Reward: -200.0\n",
      "Episode 5900 Average Reward: -200.0\n",
      "Episode 6000 Average Reward: -200.0\n",
      "Episode 6100 Average Reward: -200.0\n",
      "Episode 6200 Average Reward: -200.0\n",
      "Episode 6300 Average Reward: -200.0\n",
      "Episode 6400 Average Reward: -200.0\n",
      "Episode 6500 Average Reward: -200.0\n",
      "Episode 6600 Average Reward: -200.0\n",
      "Episode 6700 Average Reward: -200.0\n",
      "Episode 6800 Average Reward: -200.0\n",
      "Episode 6900 Average Reward: -200.0\n",
      "Episode 7000 Average Reward: -200.0\n",
      "Episode 7100 Average Reward: -200.0\n",
      "Episode 7200 Average Reward: -200.0\n",
      "Episode 7300 Average Reward: -200.0\n",
      "Episode 7400 Average Reward: -200.0\n",
      "Episode 7500 Average Reward: -200.0\n",
      "Episode 7600 Average Reward: -200.0\n",
      "Episode 7700 Average Reward: -200.0\n",
      "Episode 7800 Average Reward: -200.0\n",
      "Episode 7900 Average Reward: -200.0\n",
      "Episode 8000 Average Reward: -200.0\n",
      "Episode 8100 Average Reward: -200.0\n",
      "Episode 8200 Average Reward: -200.0\n",
      "Episode 8300 Average Reward: -200.0\n",
      "Episode 8400 Average Reward: -200.0\n",
      "Episode 8500 Average Reward: -200.0\n",
      "Episode 8600 Average Reward: -200.0\n",
      "Episode 8700 Average Reward: -200.0\n",
      "Episode 8800 Average Reward: -200.0\n",
      "Episode 8900 Average Reward: -200.0\n",
      "Episode 9000 Average Reward: -200.0\n",
      "Episode 9100 Average Reward: -200.0\n",
      "Episode 9200 Average Reward: -200.0\n",
      "Episode 9300 Average Reward: -200.0\n",
      "Episode 9400 Average Reward: -200.0\n",
      "Episode 9500 Average Reward: -200.0\n",
      "Episode 9600 Average Reward: -200.0\n",
      "Episode 9700 Average Reward: -200.0\n",
      "Episode 9800 Average Reward: -200.0\n",
      "Episode 9900 Average Reward: -200.0\n",
      "Episode 10000 Average Reward: -200.0\n"
     ]
    }
   ],
   "source": [
    "# Initialize variables to track rewards\n",
    "reward_list = [] # Temporal one\n",
    "ave_reward_list = []\n",
    "successful = []\n",
    "last_position_episodes = []\n",
    "\n",
    "# Run RL algorithm\n",
    "for i in range(EPISODES):\n",
    "    \n",
    "    # Initialize parameters\n",
    "    done = False\n",
    "    tot_reward, reward = 0,0\n",
    "    state = env.reset()\n",
    "    \n",
    "    # Done will be True when:\n",
    "    #  1. 200 iterations have happened OR\n",
    "    #  2. The car reached the flag\n",
    "    \n",
    "    while done != True:\n",
    "        \n",
    "        # We select a random action\n",
    "        action = np.random.randint(0, env.action_space.n)\n",
    "        # We apply the action, and receive information from the environment\n",
    "        state, reward, done, info = env.step(action)\n",
    "        \n",
    "        tot_reward += reward\n",
    "    \n",
    "    # We track the last position on the episode\n",
    "    last_position_episodes.append(state[0])\n",
    "    \n",
    "    # The done condition was triggered because the car reached the flag\n",
    "    if state[0] >= 0.5:\n",
    "        successful.append(i)\n",
    "        \n",
    "    # Track rewards\n",
    "    reward_list.append(tot_reward)\n",
    "    \n",
    "    # Show average rewards per 100 episodes\n",
    "    if (i+1) % 100 == 0:\n",
    "        ave_reward = np.mean(reward_list)\n",
    "        ave_reward_list.append(ave_reward)\n",
    "        reward_list = []\n",
    "        print('Episode {} Average Reward: {}'.format(i+1, ave_reward))\n",
    "        \n",
    "env.close()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "- We offer you some plot functions to monitor how the RL algorithm is learning"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In this case, since it is a random agent, we can see that we are not learning anything at all"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "def plot_last_positions(positions):\n",
    "    plt.figure(2, figsize=[10,5])\n",
    "    p = pd.Series(positions)\n",
    "    ma = p.rolling(10).mean()\n",
    "    plt.plot(p, alpha=0.8)\n",
    "    plt.plot(ma)\n",
    "    plt.xlabel('Episode')\n",
    "    plt.ylabel('Position')\n",
    "    plt.title('Car Final Position')\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_last_positions(last_position_episodes)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "def plot_rewards(ave_reward_list):\n",
    "    plt.subplots(figsize=(10,5))\n",
    "    plt.plot(100*(np.arange(len(ave_reward_list)) + 1), ave_reward_list)\n",
    "    plt.xlabel('Episodes')\n",
    "    plt.ylabel('Average Reward')\n",
    "    plt.title('Average Reward vs Episodes');"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_rewards(ave_reward_list)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "def plot_successes(successful):\n",
    "    fig, ax = plt.subplots(figsize=(10,5))\n",
    "    ax.hist(successful, bins = int(EPISODES/500));\n",
    "    ax.set_xlabel('Episodes');\n",
    "    ax.set_ylabel('#Successful');\n",
    "    print('successful episodes: {}'.format(np.count_nonzero(successful)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "successful episodes: 0\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAm8AAAE9CAYAAABdmIXpAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjAsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+17YcXAAAXv0lEQVR4nO3df7RlZX3f8fcHRtGI4ecQKcM4pJDY8UcBb0GjbVAEwawwLMUKrXW0tMRWao01FjSWHyZZktROSkTjRNCRpKAhjU6K6UD4EaMRwh1AItgpE0QZYckQRiwhiqPf/nH2yPFy7tzDzJxz7jP3/VrrrHP2s59z9vfyrHvnw/PsfXaqCkmSJLVhj0kXIEmSpOEZ3iRJkhpieJMkSWqI4U2SJKkhhjdJkqSGGN4kSZIasmjSBYzTgQceWMuWLZt0GZIkSXNav379Q1W1eGb7ggpvy5YtY3p6etJlSJIkzSnJ1we1u2wqSZLUEMObJElSQwxvkiRJDTG8SZIkNcTwJkmS1BDDmyRJUkMMb5IkSQ0xvEmSJDXE8CZJktQQw5skSVJDDG+SJEkNMbxJkiQ1xPAmSZLUEMObJElSQwxvkiRJDTG8SZIkNcTwJkmS1BDDmyRJUkMMb5IkSQ0xvEmSJDXE8CZJktQQw5skSVJDDG+SJEkNMbxJkiQ1xPAmSZLUEMObJElSQwxvkiRJDTG8SZIkNcTwJkmS1BDDmyRJUkMMb5IkSQ2ZaHhLclKSDUk2JjlnwP69knyq239zkmUz9i9N8miSd42rZkmSpEmaWHhLsidwCXAysBw4I8nyGd3OBLZU1eHAKuCiGftXAX866lolSZLmi0nOvB0DbKyqe6rqceBKYMWMPiuANd3rq4DjkwQgyanAPcCdY6pXkiRp4iYZ3g4B7uvb3tS1DexTVVuBR4ADkjwL+M/ABWOoU5Ikad6YZHjLgLYass8FwKqqenTOgyRnJZlOMr158+YdKFOSJGn+WDTBY28CDu3bXgLcP0ufTUkWAfsADwPHAqcl+U1gX+CHSb5bVR+aeZCqWg2sBpiampoZDiVJkpoyyfB2C3BEksOAbwKnA/9iRp+1wErgS8BpwPVVVcA/3dYhyfnAo4OCmyRJ0u5mYuGtqrYmORtYB+wJXFZVdya5EJiuqrXApcDlSTbSm3E7fVL1SpIkzQfpTWQtDFNTUzU9PT3pMiRJkuaUZH1VTc1s9w4LkiRJDTG8SZIkNcTwJkmS1BDDmyRJUkMMb5IkSQ0xvEmSJDXE8CZJktQQw5skSVJDDG+SJEkNMbxJkiQ1xPAmSZLUEMObJElSQwxvkiRJDTG8SZIkNcTwJkmS1BDDmyRJUkMMb5IkSQ0xvEmSJDXE8CZJktQQw5skSVJDDG+SJEkNMbxJkiQ1xPAmSZLUEMObJElSQwxvkiRJDTG8SZIkNcTwJkmS1BDDmyRJUkMMb5IkSQ0xvEmSJDXE8CZJktQQw5skSVJDDG+SJEkNMbxJkiQ1xPAmSZLUEMObJElSQwxvkiRJDTG8SZIkNcTwJkmS1BDDmyRJUkMMb5IkSQ2ZaHhLclKSDUk2JjlnwP69knyq239zkmVd+wlJ1if56+75leOuXZIkaRImFt6S7AlcApwMLAfOSLJ8RrczgS1VdTiwCrioa38I+MWqeiGwErh8PFVLkiRN1iRn3o4BNlbVPVX1OHAlsGJGnxXAmu71VcDxSVJVt1XV/V37ncAzkuw1lqolSZImaJLh7RDgvr7tTV3bwD5VtRV4BDhgRp/XAbdV1fcGHSTJWUmmk0xv3rx5lxQuSZI0KZMMbxnQVk+lT5Ln01tK/aXZDlJVq6tqqqqmFi9evEOFSpIkzReTDG+bgEP7tpcA98/WJ8kiYB/g4W57CfDHwJuq6m9GXq0kSdI8MMnwdgtwRJLDkjwdOB1YO6PPWnoXJACcBlxfVZVkX+Bq4Nyq+uLYKpYkSZqwiYW37hy2s4F1wFeBT1fVnUkuTHJK1+1S4IAkG4F3Atu+TuRs4HDgfUlu7x4HjflHkCRJGrtUzTzNbPc1NTVV09PTky5DkiRpTknWV9XUzHbvsCBJktQQw5skSVJDDG+SJEkNMbxJkiQ1xPAmSZLUEMObJElSQwxvkiRJDTG8SZIkNcTwJkmS1BDDmyRJUkMMb5IkSQ0xvEmSJDXE8CZJktQQw5skSVJDDG+SJEkNMbxJkiQ1ZNH2dib5HaBm219Vb9/lFUmSJGlW2w1vwPRYqpAkSdJQthveqmrNuAqRJEnS3OaaeQMgyQ0MWD6tqlfu8ookSZI0q6HCG/CuvtfPAF4HbN315UiSJGl7hgpvVbV+RtMXk/z5COqRJEnSdgy7bLp/3+YewIuB54ykIkmSJM1q2GXT9fTOeQu95dKvAWeOqihJkiQNNtf3vL2+qv4QOL6q7hlTTZIkSZrFXHdYOLd7vmrUhUiSJGlucy2b/m33NSGHJVk7c2dVnTKasiRJkjTIXOHtF4CjgcuBD46+HEmSJG3PXHdYeBy4KcnPVdVmgCR7AHtX1XfGUaAkSZKeMNc5b9v89yQ/meRZwF3AhiS/MsK6JEmSNMCw4W15N9N2KvA5YCnwr0ZWlSRJkgYaNrw9LcnT6IW3z1bV9xlwr1NJkiSN1rDh7aPAvcCzgM8neS7gOW+SJEljNuy9TS8GLu5r+nqSV4ymJEmSJM1mqJm3JP+xu2AhSS5NcivwyhHXJkmSpBmGXTb9190FCycCi4G3AB8YWVWSJEkaaNjwlu75NcDHq+rLfW2SJEkak2HD2/ok19ALb+uSPBv44ejKkiRJ0iBDXbAAnAkcCdxTVY8lOYDe0qkkSZLGaNiZtwKWA2/vtp8FPGMkFUmSJGlWw4a3DwMvBc7otv8fcMnOHjzJSUk2JNmY5JwB+/dK8qlu/81JlvXtO7dr35Dk1TtbiyRJUguGDW/HVtXbgO8CVNUW4Ok7c+Ake9ILgCfTm9U7I8nyGd3OBLZU1eHAKuCi7r3LgdOB5wMnAR/uPk+SJGm3Nmx4+34XjgogyWJ2/oKFY4CNVXVPVT0OXAmsmNFnBbCme30VcHySdO1XVtX3quprwMbu8yRJknZrw4a3i4E/Bg5K8uvAF4Df2MljHwLc17e9qWsb2KeqtgKPAAcM+V5JkqTdzrC3x/qDJOuB4+l9v9upVfXVnTz2oO+Jm3mz+9n6DPPe3gckZwFnASxduvSp1CdJkjTvDHt7rJcA36yqS6rqQ8CmJMfu5LE3AYf2bS8B7p+tT5JFwD7Aw0O+F4CqWl1VU1U1tXjx4p0sWZIkabKGXTb9CPBo3/bfdW074xbgiCSHJXk6vQsQ1s7osxZY2b0+Dbi+qqprP727GvUw4Ajgr3ayHkmSpHlv2C/pTReaAKiqH3YzYTusqrYmORtYB+wJXFZVdya5EJiuqrXApcDlSTbSm3E7vXvvnUk+DdwFbAXeVlU/2Jl6JEmSWpC+TDZ7p+R/AjfyxGzbvwdeUVWnjq60XW9qaqqmp6cnXYYkSdKckqyvqqmZ7cMum74V+Dngm/TONzuW7iIASZIkjc+wV5s+SLdkKUmSpMkZ9mrTNUn27dveL8lloytLkiRJgwy7bPqiqvr2to3u9lhHjaYkSZIkzWbY8LZHkv22bSTZn+GvVJUkSdIuMmwA+yDwl0mu6rZfD/z6aEqSJEnSbIa9YOGTSaaBV9K7NdVrq+qukVYmSZKkJxkqvCVZSu8OC2v726rqG6MqTJIkSU827LLp1Txx4/dnAocBG4Dnj6IoSZIkDTbssukL+7eTHA380kgqkiRJ0qyGvdr0x1TVrcA/2cW1SJIkaQ7DnvP2zr7NPYCjgc0jqUiSJEmzGvact2f3vd5K7xy4P9r15UiSJGl7hj3n7YJtr7sv6/12VdV23iJJkqQR2O45b0n+S5Lnda/3SnI98DfAt5K8ahwFSpIk6QlzXbDwBnpfCQKwsuu/GPh54DdGWJckSZIGmCu8Pd63PPpq4Iqq+kFVfRXvbSpJkjR2c4W37yV5QZLFwCuAa/r2/cToypIkSdIgc82evQO4it5S6aqq+hpAktcAt424NkmSJM2w3fBWVTcBz0vyvqp6P/QuXKiqzwGfG0eBkiRJesJcV5u+O8lLgdf1NX9ptCVJkiRpNnMtm24AXg/8dJK/AL4KHJDkZ6tqw/bfKkmSpF1trgsWtgDvATYCxwEXd+3nJPnLEdYlSZKkAeaaeTsJOA/4h8B/A74M/F1VvWXUhUmSJOnJtjvzVlXvqarjgXuB36cX9hYn+UKSPxlDfZIkSeoz7BftrquqW4Bbkvy7qnp5kgNHWZgkSZKebK5z3gCoqnf3bb65a3toFAVJkiRpdkOFt35V9eVRFCJJkqS5PeXwJkmSpMkxvEmSJDXE8CZJktQQw5skSVJDDG+SJEkNMbxJkiQ1xPAmSZLUEMObJElSQwxvkiRJDTG8SZIkNcTwJkmS1BDDmyRJUkMMb5IkSQ2ZSHhLsn+Sa5Pc3T3vN0u/lV2fu5Os7Np+IsnVSf5PkjuTfGC81UuSJE3OpGbezgGuq6ojgOu67R+TZH/gPOBY4BjgvL6Q91+r6nnAUcDLkpw8nrIlSZIma1LhbQWwpnu9Bjh1QJ9XA9dW1cNVtQW4Fjipqh6rqhsAqupx4FZgyRhqliRJmrhJhbefqqoHALrngwb0OQS4r297U9f2I0n2BX6R3uzdQEnOSjKdZHrz5s07XbgkSdIkLRrVByf5M+A5A3a9d9iPGNBWfZ+/CLgCuLiq7pntQ6pqNbAaYGpqqmbrJ0mS1IKRhbeqetVs+5J8K8nBVfVAkoOBBwd02wQc17e9BLixb3s1cHdV/fYuKFeSJKkJk1o2XQus7F6vBD47oM864MQk+3UXKpzYtZHk14B9gHeMoVZJkqR5Y1Lh7QPACUnuBk7otkkyleRjAFX1MPB+4JbucWFVPZxkCb2l1+XArUluT/JvJvFDSJIkjVuqFs5pYFNTUzU9PT3pMiRJkuaUZH1VTc1s9w4LkiRJDTG8SZIkNcTwJkmS1BDDmyRJUkMMb5IkSQ0xvEmSJDXE8CZJktQQw5skSVJDDG+SJEkNMbxJkiQ1xPAmSZLUEMObJElSQwxvkiRJDTG8SZIkNcTwJkmS1BDDmyRJUkMMb5IkSQ0xvEmSJDXE8CZJktQQw5skSVJDDG+SJEkNMbxJkiQ1xPAmSZLUEMObJElSQwxvkiRJDTG8SZIkNcTwJkmS1BDDmyRJUkMMb5IkSQ0xvEmSJDXE8CZJktQQw5skSVJDDG+SJEkNMbxJkiQ1xPAmSZLUEMObJElSQwxvkiRJDTG8SZIkNcTwJkmS1BDDmyRJUkMmEt6S7J/k2iR3d8/7zdJvZdfn7iQrB+xfm+Qro69YkiRpfpjUzNs5wHVVdQRwXbf9Y5LsD5wHHAscA5zXH/KSvBZ4dDzlSpIkzQ+TCm8rgDXd6zXAqQP6vBq4tqoerqotwLXASQBJ9gbeCfzaGGqVJEmaNyYV3n6qqh4A6J4PGtDnEOC+vu1NXRvA+4EPAo+NskhJkqT5ZtGoPjjJnwHPGbDrvcN+xIC2SnIkcHhV/XKSZUPUcRZwFsDSpUuHPLQkSdL8NLLwVlWvmm1fkm8lObiqHkhyMPDggG6bgOP6tpcANwIvBV6c5F569R+U5MaqOo4Bqmo1sBpgamqqnvpPIkmSNH9Matl0LbDt6tGVwGcH9FkHnJhkv+5ChROBdVX1kar6B1W1DHg58H9nC26SJEm7m0mFtw8AJyS5Gzih2ybJVJKPAVTVw/TObbule1zYtUmSJC1YqVo4K4lTU1M1PT096TIkSZLmlGR9VU3NbPcOC5IkSQ0xvEmSJDXE8CZJktQQw5skSVJDDG+SJEkNMbxJkiQ1xPAmSZLUEMObJElSQwxvkiRJDTG8SZIkNcTwJkmS1BDDmyRJUkMMb5IkSQ0xvEmSJDXE8CZJktQQw5skSVJDDG+SJEkNMbxJkiQ1xPAmSZLUEMObJElSQwxvkiRJDTG8SZIkNcTwJkmS1BDDmyRJUkMMb5IkSQ0xvEmSJDXE8CZJktQQw5skSVJDDG+SJEkNMbxJkiQ1xPAmSZLUEMObJElSQ1JVk65hbJJsBr4+4sMcCDw04mPoqXNc5h/HZP5xTOYnx2X+GdeYPLeqFs9sXFDhbRySTFfV1KTr0I9zXOYfx2T+cUzmJ8dl/pn0mLhsKkmS1BDDmyRJUkMMb7ve6kkXoIEcl/nHMZl/HJP5yXGZfyY6Jp7zJkmS1BBn3iRJkhpieNtBSU5KsiHJxiTnDNi/V5JPdftvTrJs/FUuLEOMyTuT3JXkjiTXJXnuJOpcaOYal75+pyWpJF5VN2LDjEmSf979vtyZ5H+Mu8aFaIi/YUuT3JDktu7v2GsmUedCkuSyJA8m+cos+5Pk4m7M7khy9DjqMrztgCR7ApcAJwPLgTOSLJ/R7UxgS1UdDqwCLhpvlQvLkGNyGzBVVS8CrgJ+c7xVLjxDjgtJng28Hbh5vBUuPMOMSZIjgHOBl1XV84F3jL3QBWbI35VfBT5dVUcBpwMfHm+VC9IngJO2s/9k4IjucRbwkTHUZHjbQccAG6vqnqp6HLgSWDGjzwpgTff6KuD4JBljjQvNnGNSVTdU1WPd5k3AkjHXuBAN87sC8H56Yfq74yxugRpmTP4tcElVbQGoqgfHXONCNMy4FPCT3et9gPvHWN+CVFWfBx7eTpcVwCer5yZg3yQHj7ouw9uOOQS4r297U9c2sE9VbQUeAQ4YS3UL0zBj0u9M4E9HWpFgiHFJchRwaFX9r3EWtoAN87vyM8DPJPlikpuSbG/mQbvGMONyPvDGJJuAzwH/YTylaTue6r89u8SiUR9gNzVoBm3mZbvD9NGuM/R/7yRvBKaAnx9pRYI5xiXJHvROK3jzuArSUL8ri+gtAx1Hb4b6L5K8oKq+PeLaFrJhxuUM4BNV9cEkLwUu78blh6MvT7OYyL/1zrztmE3AoX3bS3jy9PWP+iRZRG+Ke3tTr9o5w4wJSV4FvBc4paq+N6baFrK5xuXZwAuAG5PcC7wEWOtFCyM17N+vz1bV96vqa8AGemFOozPMuJwJfBqgqr4EPIPePTY1OUP927OrGd52zC3AEUkOS/J0eieOrp3RZy2wsnt9GnB9+aV6ozTnmHTLcx+lF9w8h2c8tjsuVfVIVR1YVcuqahm9cxFPqarpyZS7IAzz9+szwCsAkhxIbxn1nrFWufAMMy7fAI4HSPKP6IW3zWOtUjOtBd7UXXX6EuCRqnpg1Ad12XQHVNXWJGcD64A9gcuq6s4kFwLTVbUWuJTelPZGejNup0+u4t3fkGPyW8DewB921458o6pOmVjRC8CQ46IxGnJM1gEnJrkL+AHwK1X1t5Orevc35Lj8J+D3kvwyvaW5NzspMFpJrqB3+sCB3bmG5wFPA6iq36V37uFrgI3AY8BbxlKX4y5JktQOl00lSZIaYniTJElqiOFNkiSpIYY3SZKkhhjeJEmSGmJ4k7SgJPlBktv7HufM0f+tSd60C457b/edaZK0U/yqEEkLSpJHq2rvCRz3XmCqqh4a97El7V6ceZMkfjQzdlGSv+oeh3ft5yd5V/f67UnuSnJHkiu7tv2TfKZruynJi7r2A5Jck+S2JB+l7x6ISd7YHeP2JB9Nsmf3+ESSryT56+6LWCXpSQxvkhaaZ85YNn1D377vVNUxwIeA3x7w3nOAo6rqRcBbu7YLgNu6tvcAn+zazwO+UFVH0buFzlL40W2N3gC8rKqOpHcHg38JHAkcUlUvqKoXAh/fhT+zpN2It8eStND8fReaBrmi73nVgP13AH+Q5DP07v8J8HLgdQBVdX0347YP8M+A13btVyfZ0vU/HngxcEt3m7ZnAg8CfwL8dJLfAa4GrtnxH1HS7syZN0l6Qs3yeptfAC6hF77WJ1lE33LogPcO+owAa6rqyO7xs1V1flVtAf4xcCPwNuBjO/gzSNrNGd4k6Qlv6Hv+Uv+OJHsAh1bVDcC7gX2BvYHP01v2JMlxwENV9Z0Z7ScD+3UfdR1wWpKDun37J3ludyXqHlX1R8D7gKNH9UNKapvLppIWmmcmub1v+39X1bavC9kryc30/sf2jBnv2xP4/W5JNMCqqvp2kvOBjye5A3gMWNn1vwC4IsmtwJ8D3wCoqruS/CpwTRcIv09vpu3vu8/Z9j/V5+66H1nS7sSvCpEk/CoPSe1w2VSSJKkhzrxJkiQ1xJk3SZKkhhjeJEmSGmJ4kyRJaojhTZIkqSGGN0mSpIYY3iRJkhry/wHWGffImizhhgAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 720x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_successes(successful)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Can you make it better?"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "- The collection of some produced variables are sent to us, and we will give you a SCORE"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "import hackathon_mad_hub_2101 as hack_tools"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\"Score posted!\"\n",
      "\n"
     ]
    }
   ],
   "source": [
    "team = \"YOUR TEAM NAME\"\n",
    "\n",
    "hack_tools.post_score(team, [ave_reward_list, successful, last_position_episodes])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "[Leaderboard](https://hackathon-ing-mad-hub-2101.herokuapp.com/)\n"
   ]
  }
 ],
 "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": "3.8.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}