08_double_dqn_cartpole.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Double Deep Q Learning - CartPole example\n",
    "\n",
    "Q = The expected future rewards discounted over time. This is what we are trying to maximise.\n",
    "\n",
    "The aim is to teach a network to take the current state observations and recommend the action with greatest Q.\n",
    "\n",
    "<img src=\"./images/dqn.png\" width=\"500\"/>\n",
    "\n",
    "Q is learned through the Bellman equation, where the Q of any state and action is the immediate reward achieved + the discounted maximum Q value (the best action taken) of next best action, where gamma is the discount rate.\n",
    "\n",
    "$$Q(s,a)=r + \\gamma.maxQ(s',a')$$\n",
    "\n",
    "For a presenation on Q, see https://youtu.be/o22P5rCAAEQ\n",
    "\n",
    "## Key DQN components\n",
    "\n",
    "<img src=\"./images/dqn_components.png\" width=\"600|\"/>\n",
    "\n",
    "\n",
    "## General method for Q learning:\n",
    "\n",
    "Overall aim is to create a neural network that predicts Q. Improvement comes from improved accuracy in predicting 'current' understood Q, and in revealing more about Q as knowledge is gained (some rewards only discovered after time).\n",
    "\n",
    "<img src=\"./images/dqn_process.png\" width=\"600|\"/>\n",
    "    \n",
    "Target networks are used to stabilise models, and are only updated at intervals. Changes to Q values may lead to changes in closely related states (i.e. states close to the one we are in at the time) and as the network tries to correct for errors it can become unstable and suddenly lose signficiant performance. Target networks (e.g. to assess Q) are updated only infrequently (or gradually), so do not have this instability problem.\n",
    "\n",
    "## Training networks\n",
    "\n",
    "Double DQN contains two networks. This ammendment, from simple DQN, is to decouple training of Q for current state and target Q derived from next state which are closely correlated when comparing input features.\n",
    "\n",
    "The *policy network* is used to select action (action with best predicted Q) when playing the game.\n",
    "\n",
    "When training, the predicted best *action* (best predicted Q) is taken from the *policy network*, but the *policy network* is updated using the predicted Q value of the next state from the *target network* (which is updated from the policy network less frequently). So, when training, the action is selected using Q values from the *policy network*, but the the *policy network* is updated to better predict the Q value of that action from the *target network*. The *policy network* is copied across to the *target network* every *n* steps (e.g. 1000).\n",
    "\n",
    "<img src=\"./images/dqn_training.png\" width=\"700|\"/>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Code structure\n",
    "\n",
    "<img src=\"./images/dqn_program_structure.png\" width=\"700|\"/>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "################################################################################\n",
    "#                           1 Import packages                                  #\n",
    "################################################################################\n",
    "\n",
    "import gym\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "import random\n",
    "\n",
    "import torch\n",
    "import torch.nn as nn\n",
    "import torch.optim as optim\n",
    "\n",
    "# Use a double ended queue (deque) for memory\n",
    "# When memory is full, this will replace the oldest value with the new one\n",
    "from collections import deque"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "################################################################################\n",
    "#                           2 Define model parameters                          #\n",
    "################################################################################\n",
    "\n",
    "# Gym environment to use\n",
    "ENV_NAME = \"CartPole-v0\"\n",
    "# Set whether to display on screen (slows model)\n",
    "DISPLAY_ON_SCREEN = False\n",
    "# Discount rate of future rewards\n",
    "GAMMA = 0.99\n",
    "# Learing rate for neural network\n",
    "LEARNING_RATE = 0.001\n",
    "# Maximum number of game steps (state, action, reward, next state) to keep\n",
    "MEMORY_SIZE = 10000\n",
    "# Sample batch size for policy network update\n",
    "BATCH_SIZE = 5\n",
    "# Number of game steps to play before starting training (all random actions)\n",
    "REPLAY_START_SIZE = 500\n",
    "# Number of steps between policy -> target network update\n",
    "SYNC_TARGET_STEPS = 100\n",
    "# Exploration rate (episolon) is probability of choosign a random action\n",
    "EXPLORATION_MAX = 1.0\n",
    "EXPLORATION_MIN = 0.01\n",
    "# Reduction in epsilon with each game step\n",
    "EXPLORATION_DECAY = 0.999\n",
    "# Target rewards\n",
    "TARGET_REWARD = 200\n",
    "# Number of times target must be met without break\n",
    "TARGET_PERIOD = 5"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "################################################################################\n",
    "#                      3 Define DQN (Deep Q Network) class                     #\n",
    "#                    (Used for both policy and target nets)                    #\n",
    "################################################################################\n",
    "\n",
    "class DQN(nn.Module):\n",
    "\n",
    "    \"\"\"Deep Q Network. Udes for both policy (action) and target (Q) networks.\"\"\"\n",
    "\n",
    "    def __init__(self, observation_space, action_space):\n",
    "        \"\"\"Constructor method. Set up neural nets.\"\"\"\n",
    "\n",
    "        # Set starting exploration rate\n",
    "        self.exploration_rate = EXPLORATION_MAX\n",
    "        \n",
    "        # Set up action space (choice of possible actions)\n",
    "        self.action_space = action_space\n",
    "              \n",
    "        super(DQN, self).__init__()\n",
    "        self.net = nn.Sequential(\n",
    "            nn.Linear(observation_space, 24),\n",
    "            nn.ReLU(),\n",
    "            nn.Linear(24, 24),\n",
    "            nn.ReLU(),\n",
    "            nn.Linear(24, action_space)\n",
    "            )\n",
    "        \n",
    "        # Set loss function and optimizer\n",
    "        self.objective = nn.MSELoss()\n",
    "        self.optimizer = optim.Adam(\n",
    "            params=self.net.parameters(), lr=LEARNING_RATE)\n",
    "        \n",
    "    def act(self, state):\n",
    "        \"\"\"Act either randomly or by redicting action that gives max Q\"\"\"\n",
    "        \n",
    "        # Act randomly if random number < exploration rate\n",
    "        if np.random.rand() < self.exploration_rate:\n",
    "            action = random.randrange(self.action_space)\n",
    "            \n",
    "        else:\n",
    "            # Otherwise get predicted Q values of actions\n",
    "            q_values = self.net(torch.FloatTensor(state))\n",
    "            # Get index of action with best Q\n",
    "            action = np.argmax(q_values.detach().numpy()[0])\n",
    "        \n",
    "        return  action\n",
    "        \n",
    "  \n",
    "    def forward(self, x):\n",
    "        return self.net(x)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "################################################################################\n",
    "#                    4 Define policy net training function                     #\n",
    "################################################################################\n",
    "\n",
    "def optimize(policy_net, target_net, memory):\n",
    "    \"\"\"\n",
    "    Update  model by sampling from memory.\n",
    "    Uses policy network to predict best action (best Q).\n",
    "    Uses target network to provide target of Q for the selected next action.\n",
    "    \"\"\"\n",
    "      \n",
    "    # Do not try to train model if memory is less than reqired batch size\n",
    "    if len(memory) < BATCH_SIZE:\n",
    "        return    \n",
    " \n",
    "    # Reduce exploration rate (exploration rate is stored in policy net)\n",
    "    policy_net.exploration_rate *= EXPLORATION_DECAY\n",
    "    policy_net.exploration_rate = max(EXPLORATION_MIN, \n",
    "                                      policy_net.exploration_rate)\n",
    "    # Sample a random batch from memory\n",
    "    batch = random.sample(memory, BATCH_SIZE)\n",
    "    for state, action, reward, state_next, terminal in batch:\n",
    "        \n",
    "        state_action_values = policy_net(torch.FloatTensor(state))\n",
    "        \n",
    "        # Get target Q for policy net update\n",
    "       \n",
    "        if not terminal:\n",
    "            # For non-terminal actions get Q from policy net\n",
    "            expected_state_action_values = policy_net(torch.FloatTensor(state))\n",
    "            # Detach next state values from gradients to prevent updates\n",
    "            expected_state_action_values = expected_state_action_values.detach()\n",
    "            # Get next state action with best Q from the policy net (double DQN)\n",
    "            policy_next_state_values = policy_net(torch.FloatTensor(state_next))\n",
    "            policy_next_state_values = policy_next_state_values.detach()\n",
    "            best_action = np.argmax(policy_next_state_values[0].numpy())\n",
    "            # Get target net next state\n",
    "            next_state_action_values = target_net(torch.FloatTensor(state_next))\n",
    "            # Use detach again to prevent target net gradients being updated\n",
    "            next_state_action_values = next_state_action_values.detach()\n",
    "            best_next_q = next_state_action_values[0][best_action].numpy()\n",
    "            updated_q = reward + (GAMMA * best_next_q)      \n",
    "            expected_state_action_values[0][action] = updated_q\n",
    "        else:\n",
    "            # For termal actions Q = reward (-1)\n",
    "            expected_state_action_values = policy_net(torch.FloatTensor(state))\n",
    "            # Detach values from gradients to prevent gradient update\n",
    "            expected_state_action_values = expected_state_action_values.detach()\n",
    "            # Set Q for all actions to reward (-1)\n",
    "            expected_state_action_values[0] = reward\n",
    " \n",
    "        # Reset net gradients\n",
    "        policy_net.optimizer.zero_grad()  \n",
    "        # calculate loss\n",
    "        loss_v = nn.MSELoss()(state_action_values, expected_state_action_values)\n",
    "        # Backpropogate loss\n",
    "        loss_v.backward()\n",
    "        # Update network gradients\n",
    "        policy_net.optimizer.step()  \n",
    "\n",
    "    return"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "################################################################################\n",
    "#                            5 Define memory class                             #\n",
    "################################################################################\n",
    "\n",
    "class Memory():\n",
    "    \"\"\"\n",
    "    Replay memory used to train model.\n",
    "    Limited length memory (using deque, double ended queue from collections).\n",
    "      - When memory full deque replaces oldest data with newest.\n",
    "    Holds, state, action, reward, next state, and episode done.\n",
    "    \"\"\"\n",
    "    \n",
    "    def __init__(self):\n",
    "        \"\"\"Constructor method to initialise replay memory\"\"\"\n",
    "        self.memory = deque(maxlen=MEMORY_SIZE)\n",
    "\n",
    "    def remember(self, state, action, reward, next_state, done):\n",
    "        \"\"\"state/action/reward/next_state/done\"\"\"\n",
    "        self.memory.append((state, action, reward, next_state, done))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "################################################################################\n",
    "#                       6  Define results plotting function                    #\n",
    "################################################################################\n",
    "\n",
    "def plot_results(run, exploration, score):\n",
    "    \"\"\"Plot results at end of run\"\"\"\n",
    "    \n",
    "    # Set up chart (ax1 and ax2 share x-axis to combine two plots on one graph)\n",
    "    fig = plt.figure(figsize=(6,6,))\n",
    "    ax1 = fig.add_subplot(111)\n",
    "    ax2 = ax1.twinx()\n",
    "    \n",
    "    # Plot results\n",
    "    ax1.plot(run, exploration, label='exploration', color='g')\n",
    "    ax2.plot(run, score, label='score', color='r')\n",
    "    \n",
    "    # Set axes\n",
    "    ax1.set_xlabel('run')\n",
    "    ax1.set_ylabel('exploration', color='g')\n",
    "    ax2.set_ylabel('score', color='r')\n",
    "    \n",
    "    # Show\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "################################################################################\n",
    "#                                 7 Main program                               #\n",
    "################################################################################\n",
    "\n",
    "\n",
    "def cartpole():\n",
    "    \"\"\"Main program loop\"\"\"\n",
    "    \n",
    "    ############################################################################\n",
    "    #                          8 Set up environment                            #\n",
    "    ############################################################################\n",
    "        \n",
    "    # Set up game environemnt\n",
    "    env = gym.make(ENV_NAME)\n",
    "    \n",
    "    # Get number of observations returned for state\n",
    "    observation_space = env.observation_space.shape[0]\n",
    "    \n",
    "    # Get number of actions possible\n",
    "    action_space = env.action_space.n\n",
    "    \n",
    "    ############################################################################\n",
    "    #                    9 Set up policy and target nets                       #\n",
    "    ############################################################################\n",
    "    \n",
    "    # Set up policy and target neural nets\n",
    "    policy_net = DQN(observation_space, action_space)\n",
    "    target_net = DQN(observation_space, action_space)\n",
    "    \n",
    "    # Copy weights from policy_net to target\n",
    "    target_net.load_state_dict(policy_net.state_dict())\n",
    "    \n",
    "    # Set target net to eval rather than training mode\n",
    "    # We do not train target net - ot is copied from policy net at intervals\n",
    "    target_net.eval()\n",
    "    \n",
    "    ############################################################################\n",
    "    #                            10 Set up memory                              #\n",
    "    ############################################################################\n",
    "        \n",
    "    # Set up memomry\n",
    "    memory = Memory()\n",
    "    \n",
    "    ############################################################################\n",
    "    #                     11 Set up + start training loop                      #\n",
    "    ############################################################################\n",
    "    \n",
    "    # Set up run counter and learning loop    \n",
    "    run = 0\n",
    "    all_steps = 0\n",
    "    continue_learning = True\n",
    "    \n",
    "    # Set up list for results\n",
    "    results_run = []\n",
    "    results_exploration = []\n",
    "    results_score = []\n",
    "    \n",
    "    # Continue repeating games (episodes) until target complete\n",
    "    while continue_learning:\n",
    "        \n",
    "        ########################################################################\n",
    "        #                           12 Play episode                            #\n",
    "        ########################################################################\n",
    "        \n",
    "        # Increment run (episode) counter\n",
    "        run += 1\n",
    "        \n",
    "        ########################################################################\n",
    "        #                             13 Reset game                            #\n",
    "        ########################################################################\n",
    "        \n",
    "        # Reset game environment and get first state observations\n",
    "        state = env.reset()\n",
    "        \n",
    "        # Reshape state into 2D array with state obsverations as first 'row'\n",
    "        state = np.reshape(state, [1, observation_space])\n",
    "        \n",
    "        # Reset step count for episode\n",
    "        step = 0\n",
    "        \n",
    "        # Continue loop until episode complete\n",
    "        while True:\n",
    "            \n",
    "        ########################################################################\n",
    "        #                       14 Game episode loop                           #\n",
    "        ########################################################################\n",
    "            \n",
    "            # Incrememnt step counts\n",
    "            step += 1\n",
    "            all_steps += 1\n",
    "            \n",
    "            ####################################################################\n",
    "            #                       15 Get action                              #\n",
    "            ####################################################################\n",
    "            \n",
    "            # Get action to take\n",
    "            action = policy_net.act(state)\n",
    "            \n",
    "            ####################################################################\n",
    "            #                 16 Play action (get S', R, T)                    #\n",
    "            ####################################################################\n",
    "            \n",
    "            # Act\n",
    "            state_next, reward, terminal, info = env.step(action)\n",
    "            \n",
    "            # Update display if needed\n",
    "            if DISPLAY_ON_SCREEN:\n",
    "                env.render()\n",
    "            \n",
    "            # Convert step reward to negative if end of run\n",
    "            if terminal and step < TARGET_REWARD:\n",
    "                reward *= -1\n",
    "            \n",
    "            # Get observations for new state\n",
    "            state_next = np.reshape(state_next, [1, observation_space])\n",
    "            \n",
    "            ####################################################################\n",
    "            #                  17 Add S/A/R/S/T to memory                      #\n",
    "            ####################################################################\n",
    "            \n",
    "            # Record state, action, reward, new state & terminal\n",
    "            memory.remember(state, action, reward, state_next, terminal)\n",
    "            \n",
    "            # Update state\n",
    "            state = state_next\n",
    "            \n",
    "            ####################################################################\n",
    "            #                  18 Check for end of episode                     #\n",
    "            ####################################################################\n",
    "            \n",
    "            # Actions to take if end of game episode\n",
    "            if terminal:\n",
    "                # Get exploration rate\n",
    "                exploration = policy_net.exploration_rate\n",
    "                # Clear print row content\n",
    "                clear_row = '\\r' + ' '*79 + '\\r'\n",
    "                print (clear_row, end ='')\n",
    "                print (f'Run: {run}, ', end='')\n",
    "                print (f'exploration: {exploration: .3f}, ', end='')\n",
    "                print (f'score: {step}', end='')\n",
    "                \n",
    "                # Add to results lists\n",
    "                results_run.append(run)\n",
    "                results_exploration.append(exploration)\n",
    "                results_score.append(step)\n",
    "                \n",
    "                ################################################################\n",
    "                #             18b Check for end of learning                    #\n",
    "                ################################################################\n",
    "                \n",
    "                # Get average of last 5 games\n",
    "                if len(results_score) >= TARGET_PERIOD:\n",
    "                    average_reward = np.mean(results_score[-TARGET_PERIOD:])\n",
    "                    # Stop if target met in all required episodes\n",
    "                    if average_reward >= TARGET_REWARD:\n",
    "                        # Stop learning loop\n",
    "                        continue_learning = False\n",
    "                \n",
    "                # End episode loop\n",
    "                break\n",
    "            \n",
    "            \n",
    "            ####################################################################\n",
    "            #                        19 Update policy net                      #\n",
    "            ####################################################################\n",
    "            \n",
    "            # Avoid training model if memory is not of sufficient length\n",
    "            if len(memory.memory) > REPLAY_START_SIZE:\n",
    "        \n",
    "                # Update policy net \n",
    "                optimize(policy_net, target_net, memory.memory)\n",
    "\n",
    "                ################################################################\n",
    "                #             20 Update target net periodically                #\n",
    "                ################################################################\n",
    "                \n",
    "                # Use load_state_dict method to copy weights from policy net\n",
    "                if all_steps % SYNC_TARGET_STEPS == 0:\n",
    "                    target_net.load_state_dict(policy_net.state_dict())\n",
    "                \n",
    "    ############################################################################\n",
    "    #                      21 Learning complete - plot results                 #\n",
    "    ############################################################################\n",
    "        \n",
    "    # Target reached. Plot results\n",
    "    plot_results(results_run, results_exploration, results_score)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Run: 259, exploration:  0.010, score: 200                                      "
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x432 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "cartpole()"
   ]
  }
 ],
 "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": "3.7.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}