Gym_Envs_MountainCar_v0.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "Gym_Envs_1_preamble_evn_list.ipynb",
      "provenance": [],
      "collapsed_sections": [],
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/kenanmorani/Reinforcement_Learning_idu/blob/main/Gym_Envs_MountainCar_v0.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "yAevWcQHi-T4"
      },
      "source": [
        "#Reinforcement Learning Final Project  \n",
        "Izmir Democracy University, Electrical and Electronics Program  \n",
        "By Kenan Morani  \n",
        "Supervised by Assoc. Prof. Dr. Selami Beyhan  \n",
        "Presented on 11.02.2021"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "h0gYsgBrkF-r"
      },
      "source": [
        "The code lives on my github:\n",
        "https://github.com/kenanmorani/Reinforcement_Learning_idu/blob/main/Gym_Envs_MountainCar_v0.ipynb"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Pn1IAnsDYK4V"
      },
      "source": [
        "!apt-get install -y xvfb python-opengl > /dev/null 2>&1"
      ],
      "execution_count": 1,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "UMXgCNFWcx-d"
      },
      "source": [
        "!pip install gym pyvirtualdisplay > /dev/null 2>&1"
      ],
      "execution_count": 2,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Jh10T5veI1zk"
      },
      "source": [
        "# Importing the required libraries\n",
        "import gym\n",
        "import numpy as np\n",
        "import matplotlib.pyplot as plt\n",
        "from IPython import display as ipythondisplay"
      ],
      "execution_count": 3,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "YWg6aj9ddFqP",
        "outputId": "42382ed9-af23-44ad-f924-a8ae57654ba8"
      },
      "source": [
        "from pyvirtualdisplay import Display\n",
        "display = Display(visible=0, size=(400, 300))\n",
        "display.start()"
      ],
      "execution_count": 4,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<pyvirtualdisplay.display.Display at 0x7fec8be09a20>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 4
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "JZXrnpT_e4-h",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 521
        },
        "outputId": "1f3d21d5-1c9e-471b-dbdb-ea2dbf320b28"
      },
      "source": [
        "env = gym.make('MountainCar-v0')\n",
        "env.reset()\n",
        "\n",
        "for i in range(50):\n",
        "  action = env.action_space.sample()\n",
        "  obs, reward, done, info = env.step(action)\n",
        "  screen = env.render(mode='rgb_array')\n",
        "  \n",
        "  plt.imshow(screen)\n",
        "  ipythondisplay.clear_output(wait=True)\n",
        "  ipythondisplay.display(plt.gcf())\n",
        "\n",
        "  if done:\n",
        "    break"
      ],
      "execution_count": 5,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "czC7Zhq1B0t-"
      },
      "source": [
        "## Actions & States"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "gPiPvAD8hSml",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "3517ee58-94ef-43fd-8cfc-39c75a6f49eb"
      },
      "source": [
        "# Checking number of actions in the environment\n",
        "print(env.action_space.n)\n",
        " \n",
        "# Intially checking the stats and rewards in the environment\n",
        "state = env.reset()"
      ],
      "execution_count": 6,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "3\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "oJTixsvBhZPm"
      },
      "source": [
        "Three actions in the environemnt:    \n",
        " Action 0. push car left   \n",
        " Action 1. do nothin  \n",
        " Action 2. push car right"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "x74iuzq4hdKn"
      },
      "source": [
        "States are: position and velocity"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "tJ-LG8IZhtT7",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "1935e035-7a03-4f78-cd62-2e4d68b70977"
      },
      "source": [
        "# The range of values for observations (Velocity&Speed)\n",
        "print(env.observation_space.high) # High values for all possible observations\n",
        "print(env.observation_space.low) # Low values for all possible observations"
      ],
      "execution_count": 7,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "[0.6  0.07]\n",
            "[-1.2  -0.07]\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "MYDh5fp7hUAV",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "3fdab4d4-98b2-4e15-df98-1e688104c901"
      },
      "source": [
        "# Checking intial rewards and states\n",
        "done = False\n",
        "while not done:\n",
        "    action = 2 # Randomly chosing action 2 (push the car right) \n",
        "    new_state, reward, done, _ = env.step(action) \n",
        "    print(reward, new_state)\n",
        "    env.render()"
      ],
      "execution_count": 10,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "-1.0 [-0.33368059 -0.00585469]\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "4tmjfAbPCNJv"
      },
      "source": [
        "* The initial rewards are all -1 meaning that the car has not yet reached the optimal reward of 0, when it reaches the flag on top."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "l06xlayjmBX4",
        "outputId": "4b3c7c69-97ac-40a2-e6ba-0fb03f222756"
      },
      "source": [
        "# Descritizing the observation range with 40 chanks\n",
        "DISCRETE_OS_SIZE = [40] * len(env.observation_space.high)\n",
        "# The window size between high and low values\n",
        "discrete_os_win_size = (env.observation_space.high - env.observation_space.low)/DISCRETE_OS_SIZE \n",
        "# [Position Velocity] chunks (equal dividing bits lengths)\n",
        "print(discrete_os_win_size) "
      ],
      "execution_count": 39,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "[0.045  0.0035]\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "4LgA7fijCxr1"
      },
      "source": [
        "## Initializing Q-table"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Zt1iuaijmJj4",
        "outputId": "ccde05be-4d0f-4e90-9ce7-2b50789bd352"
      },
      "source": [
        "#Intializing Qtable\n",
        "q_table = np.random.uniform(low=-2, high=0, size=(DISCRETE_OS_SIZE + [env.action_space.n])) \n",
        "print(q_table.shape)"
      ],
      "execution_count": 40,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "(40, 40, 3)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "O4M1-5z51l6K"
      },
      "source": [
        "* Coluomns of the Qtable are the size of 3 (number of actions)  \n",
        "* Rows of the table are the size of each combination possible from position*velocity which is chosen to be 20X20 matrics\n",
        "* All in all the Qtable consists of 40X40X3 matrixes (All possible combinations)"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "-RnFw_pvmQI7",
        "outputId": "1e9af3dd-9080-421e-9325-6a81fa70f5f5"
      },
      "source": [
        "print(q_table)"
      ],
      "execution_count": 41,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "[[[-0.87257327 -1.99706422 -0.2915009 ]\n",
            "  [-1.06311835 -0.93063359 -0.78947425]\n",
            "  [-1.73288067 -1.55966517 -1.93813746]\n",
            "  ...\n",
            "  [-0.39162509 -1.34955075 -0.14308846]\n",
            "  [-0.38415482 -1.2250353  -1.81823437]\n",
            "  [-0.18456924 -1.67714641 -0.3648741 ]]\n",
            "\n",
            " [[-0.67835298 -0.04657395 -1.74077586]\n",
            "  [-1.28882763 -0.61228017 -1.95594084]\n",
            "  [-0.3736168  -0.87013937 -0.16889945]\n",
            "  ...\n",
            "  [-1.39409556 -1.11003039 -1.4960771 ]\n",
            "  [-1.28219745 -0.32443387 -0.41502373]\n",
            "  [-0.62769905 -1.74886261 -0.10632698]]\n",
            "\n",
            " [[-1.80133559 -0.86312203 -0.4184035 ]\n",
            "  [-0.88140152 -1.92479674 -0.81892849]\n",
            "  [-0.62174816 -1.0653813  -0.37602218]\n",
            "  ...\n",
            "  [-0.04286614 -0.88865084 -1.22221557]\n",
            "  [-1.28425321 -0.25486569 -1.30215874]\n",
            "  [-1.09451841 -0.93786956 -0.84169387]]\n",
            "\n",
            " ...\n",
            "\n",
            " [[-0.96825787 -1.63406705 -0.14139768]\n",
            "  [-0.56906802 -1.47848269 -1.30862676]\n",
            "  [-0.72306576 -0.88128316 -1.30281923]\n",
            "  ...\n",
            "  [-0.13681537 -1.38719719 -1.97613389]\n",
            "  [-1.48284722 -0.51873009 -0.55379726]\n",
            "  [-0.38043742 -0.72770005 -0.44581424]]\n",
            "\n",
            " [[-1.04390318 -0.65080022 -0.0268469 ]\n",
            "  [-1.41525591 -1.39480927 -1.38464042]\n",
            "  [-1.37700774 -0.77480919 -1.85077663]\n",
            "  ...\n",
            "  [-1.51337645 -0.92087902 -1.3738082 ]\n",
            "  [-1.18970723 -0.6821596  -0.24011743]\n",
            "  [-0.59228276 -1.88097708 -0.86624285]]\n",
            "\n",
            " [[-0.9690607  -0.03136658 -1.16728106]\n",
            "  [-1.88464757 -1.09486349 -1.93009273]\n",
            "  [-1.97587281 -1.07631844 -0.92703817]\n",
            "  ...\n",
            "  [-1.46752808 -0.65404988 -0.45198793]\n",
            "  [-1.8184511  -1.762923   -1.48777823]\n",
            "  [-1.71648506 -0.40653931 -0.13285223]]]\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "k6pHWE_waWKE"
      },
      "source": [
        "# For stats\n",
        "ep_rewards = []\n",
        "aggr_ep_rewards = {'ep': [], 'avg': [], 'max': [], 'min': []}"
      ],
      "execution_count": 28,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "x8gTWlruC8Bi"
      },
      "source": [
        "# Updating Qtable values"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "JXiKKyA04B3T"
      },
      "source": [
        "![image.png](data:image/png;base64,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)\n",
        "\n",
        "* https://en.wikipedia.org/wiki/Q-learning"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "80AJ9VYl2slt"
      },
      "source": [
        "# Defining Constants\n",
        "LEARNING_RATE = 0.1\n",
        "DISCOUNT = 0.95 # Measure of how important feature actions versus current actions are\n",
        "EPISODES = 20000 # Number of iterations\n",
        "SHOW_EVERY = 500 # İnform us that you are still working to reach the top evert 2000 episodes"
      ],
      "execution_count": 42,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "49c-YTjt4ygT",
        "outputId": "45497763-a1ec-43b6-a8db-67a6c4e1a193"
      },
      "source": [
        "# we use this tuple to look up the 3 Q values for the available actions in the q-table\n",
        "def get_discrete_state(state):\n",
        "    discrete_state = (state - env.observation_space.low)/discrete_os_win_size\n",
        "    return tuple(discrete_state.astype(np.int))  \n",
        "discrete_state = get_discrete_state (env.reset())\n",
        "print(discrete_state)\n",
        "print(np.argmax(q_table[discrete_state])) # Understanding the max Q value in the Q-table"
      ],
      "execution_count": 43,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "(15, 20)\n",
            "0\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "_1cJfggV6GqH"
      },
      "source": [
        "# Exploration settings\n",
        "epsilon = 1  # not a constant, qoing to be decayed\n",
        "START_EPSILON_DECAYING = 1\n",
        "END_EPSILON_DECAYING = EPISODES//2\n",
        "epsilon_decay_value = epsilon/(END_EPSILON_DECAYING - START_EPSILON_DECAYING)"
      ],
      "execution_count": 44,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "6_aDGwuIXO-b",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "253272bb-e174-428f-c5a0-19bb3150f0d0"
      },
      "source": [
        "# Running the epoches to reach the goal\n",
        "for episode in range(EPISODES):\n",
        "    episode_reward = 0\n",
        "    discrete_state = get_discrete_state(env.reset())\n",
        "    done = False\n",
        "\n",
        "    if episode % SHOW_EVERY == 0:\n",
        "        render = True\n",
        "        print(episode)\n",
        "    else:\n",
        "        render = False\n",
        "\n",
        "    while not done:\n",
        "\n",
        "        if np.random.random() > epsilon:\n",
        "            # Get action from Q table\n",
        "            action = np.argmax(q_table[discrete_state])\n",
        "        else:\n",
        "            # Get random action\n",
        "            action = np.random.randint(0, env.action_space.n)\n",
        "\n",
        "\n",
        "        new_state, reward, done, _ = env.step(action)\n",
        "\n",
        "        episode_reward += reward\n",
        "        new_discrete_state = get_discrete_state(new_state)\n",
        "\n",
        "        if episode % SHOW_EVERY == 0:\n",
        "            env.render()\n",
        "        #new_q = (1 - LEARNING_RATE) * current_q + LEARNING_RATE * (reward + DISCOUNT * max_future_q)\n",
        "\n",
        "        # If simulation did not end yet after last step - update Q table\n",
        "        if not done:\n",
        "\n",
        "            # Maximum possible Q value in next step (for new state)\n",
        "            max_future_q = np.max(q_table[new_discrete_state])\n",
        "\n",
        "            # Current Q value (for current state and performed action)\n",
        "            current_q = q_table[discrete_state + (action,)]\n",
        "\n",
        "            # And here's our equation for a new Q value for current state and action\n",
        "            new_q = (1 - LEARNING_RATE) * current_q + LEARNING_RATE * (reward + DISCOUNT * max_future_q)\n",
        "            \n",
        "            # Update Q table with new Q value\n",
        "            q_table[discrete_state + (action,)] = new_q\n",
        "            \n",
        "        # Simulation ended (for any reson) - if goal position is achived - update Q value with reward directly\n",
        "        elif new_state[0] >= env.goal_position:\n",
        "            q_table[discrete_state + (action,)] = 0\n",
        "\n",
        "        discrete_state = new_discrete_state\n",
        "\n",
        "    # Decaying is being done every episode if episode number is within decaying range\n",
        "    if END_EPSILON_DECAYING >= episode >= START_EPSILON_DECAYING:\n",
        "        epsilon -= epsilon_decay_value\n",
        "\n",
        "    ep_rewards.append(episode_reward)\n",
        "    if not episode % SHOW_EVERY:\n",
        "        average_reward = sum(ep_rewards[-SHOW_EVERY:])/SHOW_EVERY\n",
        "        aggr_ep_rewards['ep'].append(episode)\n",
        "        aggr_ep_rewards['avg'].append(average_reward)\n",
        "        aggr_ep_rewards['max'].append(max(ep_rewards[-SHOW_EVERY:]))\n",
        "        aggr_ep_rewards['min'].append(min(ep_rewards[-SHOW_EVERY:]))\n",
        "        print(f'Episode: {episode:>5d}, average reward: {average_reward:>4.1f}, current epsilon: {epsilon:>1.2f}')\n",
        "env.close()"
      ],
      "execution_count": 45,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "0\n",
            "Episode:     0, average reward: -137.1, current epsilon: 1.00\n",
            "500\n",
            "Episode:   500, average reward: -200.0, current epsilon: 0.95\n",
            "1000\n",
            "Episode:  1000, average reward: -200.0, current epsilon: 0.90\n",
            "1500\n",
            "Episode:  1500, average reward: -200.0, current epsilon: 0.85\n",
            "2000\n",
            "Episode:  2000, average reward: -200.0, current epsilon: 0.80\n",
            "2500\n",
            "Episode:  2500, average reward: -200.0, current epsilon: 0.75\n",
            "3000\n",
            "Episode:  3000, average reward: -200.0, current epsilon: 0.70\n",
            "3500\n",
            "Episode:  3500, average reward: -200.0, current epsilon: 0.65\n",
            "4000\n",
            "Episode:  4000, average reward: -200.0, current epsilon: 0.60\n",
            "4500\n",
            "Episode:  4500, average reward: -200.0, current epsilon: 0.55\n",
            "5000\n",
            "Episode:  5000, average reward: -200.0, current epsilon: 0.50\n",
            "5500\n",
            "Episode:  5500, average reward: -199.8, current epsilon: 0.45\n",
            "6000\n",
            "Episode:  6000, average reward: -199.2, current epsilon: 0.40\n",
            "6500\n",
            "Episode:  6500, average reward: -198.7, current epsilon: 0.35\n",
            "7000\n",
            "Episode:  7000, average reward: -196.1, current epsilon: 0.30\n",
            "7500\n",
            "Episode:  7500, average reward: -194.7, current epsilon: 0.25\n",
            "8000\n",
            "Episode:  8000, average reward: -196.0, current epsilon: 0.20\n",
            "8500\n",
            "Episode:  8500, average reward: -195.1, current epsilon: 0.15\n",
            "9000\n",
            "Episode:  9000, average reward: -187.2, current epsilon: 0.10\n",
            "9500\n",
            "Episode:  9500, average reward: -182.5, current epsilon: 0.05\n",
            "10000\n",
            "Episode: 10000, average reward: -184.6, current epsilon: -0.00\n",
            "10500\n",
            "Episode: 10500, average reward: -176.2, current epsilon: -0.00\n",
            "11000\n",
            "Episode: 11000, average reward: -169.2, current epsilon: -0.00\n",
            "11500\n",
            "Episode: 11500, average reward: -174.5, current epsilon: -0.00\n",
            "12000\n",
            "Episode: 12000, average reward: -180.0, current epsilon: -0.00\n",
            "12500\n",
            "Episode: 12500, average reward: -161.9, current epsilon: -0.00\n",
            "13000\n",
            "Episode: 13000, average reward: -162.2, current epsilon: -0.00\n",
            "13500\n",
            "Episode: 13500, average reward: -160.5, current epsilon: -0.00\n",
            "14000\n",
            "Episode: 14000, average reward: -150.3, current epsilon: -0.00\n",
            "14500\n",
            "Episode: 14500, average reward: -157.6, current epsilon: -0.00\n",
            "15000\n",
            "Episode: 15000, average reward: -153.6, current epsilon: -0.00\n",
            "15500\n",
            "Episode: 15500, average reward: -148.4, current epsilon: -0.00\n",
            "16000\n",
            "Episode: 16000, average reward: -164.9, current epsilon: -0.00\n",
            "16500\n",
            "Episode: 16500, average reward: -168.1, current epsilon: -0.00\n",
            "17000\n",
            "Episode: 17000, average reward: -159.9, current epsilon: -0.00\n",
            "17500\n",
            "Episode: 17500, average reward: -140.7, current epsilon: -0.00\n",
            "18000\n",
            "Episode: 18000, average reward: -148.2, current epsilon: -0.00\n",
            "18500\n",
            "Episode: 18500, average reward: -144.9, current epsilon: -0.00\n",
            "19000\n",
            "Episode: 19000, average reward: -153.4, current epsilon: -0.00\n",
            "19500\n",
            "Episode: 19500, average reward: -141.4, current epsilon: -0.00\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 265
        },
        "id": "39X8_v7Hd6Uj",
        "outputId": "0f277ed9-3c91-4f66-8e44-18f44080f2d3"
      },
      "source": [
        "plt.plot(aggr_ep_rewards['ep'], aggr_ep_rewards['avg'], label=\"average rewards\")\n",
        "plt.plot(aggr_ep_rewards['ep'], aggr_ep_rewards['max'], label=\"max rewards\")\n",
        "plt.plot(aggr_ep_rewards['ep'], aggr_ep_rewards['min'], label=\"min rewards\")\n",
        "plt.legend(loc=4)\n",
        "plt.grid(True)\n",
        "plt.show()"
      ],
      "execution_count": 47,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    }
  ]
}