reinforcement_learning/taxi/q_learning.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import gymnasium as gym\n",
    "import random\n",
    "import numpy as np\n",
    "from IPython.display import clear_output\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "env = gym.make(\"Taxi-v3\", render_mode=\"ansi\")"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Training\n",
    "\n",
    "This code implements Q-Learning algorithm to solve a reinforcement learning problem in the OpenAI Gym environment.\n",
    "\n",
    "The main loop runs for 100,000 episodes. In each episode, the agent selects an action based on the epsilon-greedy policy. If the randomly selected number is less than the exploration rate (`epsilon`), the agent selects a random action from the action space, otherwise, it selects the action with the maximum expected reward based on the Q-table. The Q-table is initialized with zeros, and its entries are updated based on the Bellman equation:\n",
    "\n",
    "`Q(s,a) = (1 - alpha) * Q(s,a) + alpha * (r + gamma * max(Q(s', a')))`\n",
    "\n",
    "where `s` is the current state, `a` is the selected action, `s'` is the next state, `r` is the reward, `alpha` is the learning rate, `gamma` is the discount factor, and `max(Q(s', a'))` is the maximum expected reward for the next state.\n",
    "\n",
    "The variables `all_epochs` and `all_penalties` are used to record the number of steps and penalties (negative rewards) for each episode. These variables are plotted at the end of the training using the `matplotlib` library."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Episode: 100000\n",
      "Training finished.\n",
      "\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x12115a5e0>]"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1500x700 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "q_table = np.zeros([env.observation_space.n, env.action_space.n])\n",
    "\n",
    "# Hyperparameters\n",
    "alpha = 0.1\n",
    "gamma = 0.6\n",
    "epsilon = 0.1\n",
    "\n",
    "# For plotting metrics\n",
    "all_epochs = []\n",
    "all_penalties = []\n",
    "\n",
    "for i in range(1, 100001):\n",
    "    state, info = env.reset()\n",
    "\n",
    "    epochs, penalties, reward, = 0, 0, 0\n",
    "    done = False\n",
    "    \n",
    "    while not done:\n",
    "        if random.uniform(0, 1) < epsilon:\n",
    "            action = env.action_space.sample() # Explore action space\n",
    "        else:\n",
    "            action = np.argmax(q_table[state]) # Exploit learned values\n",
    "\n",
    "        next_state, reward, terminated, truncated, _ = env.step(action) \n",
    "        \n",
    "        done = terminated or truncated\n",
    "        \n",
    "        old_value = q_table[state, action]\n",
    "        next_max = np.max(q_table[next_state])\n",
    "        \n",
    "        new_value = (1 - alpha) * old_value + alpha * (reward + gamma * next_max)\n",
    "        q_table[state, action] = new_value\n",
    "\n",
    "        if reward < 0:\n",
    "            penalties += reward\n",
    "\n",
    "        state = next_state\n",
    "        epochs += 1\n",
    "        \n",
    "    all_penalties.append(penalties)\n",
    "    all_epochs.append(epochs)\n",
    "        \n",
    "    if i % 100 == 0:\n",
    "        clear_output(wait=True)\n",
    "        print(f\"Episode: {i}\")\n",
    "\n",
    "\n",
    "print(\"Training finished.\\n\")\n",
    "fig = plt.figure(1, figsize=(15, 7))\n",
    "plt.clf()\n",
    "ax1 = fig.add_subplot(111)\n",
    "\n",
    "plt.title('Training...')\n",
    "ax1.set_xlabel('Episode')\n",
    "ax1.set_ylabel('Duration & Rewards')\n",
    "ax1.plot(all_epochs, label='Duration', alpha=0.2, color='r')\n",
    "ax1.plot(all_penalties, label='Rewards', alpha=0.2, color='g')"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Evaluation\n",
    "\n",
    "This code evaluates the performance of an agent trained using Q-learning on an OpenAI Gym environment.\n",
    "\n",
    "The code runs the agent for a specified number of episodes (in this case, 100) and computes the average number of timesteps and penalties (negative rewards) per episode. In each episode, the agent selects actions based on the learned Q-values and transitions to the next state according to the environment dynamics. If the agent receives a negative reward, it incurs a penalty.\n",
    "\n",
    "The output of the code shows the average timesteps and penalties per episode for the evaluated agent."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Results after 100 episodes:\n",
      "Average timesteps per episode: 13.05\n",
      "Average penalties per episode: -12.05\n"
     ]
    }
   ],
   "source": [
    "\"\"\"Evaluate agent's performance after Q-learning\"\"\"\n",
    "\n",
    "total_epochs, total_penalties = 0, 0\n",
    "episodes = 100\n",
    "\n",
    "for i in range(episodes):\n",
    "    state, info = env.reset()\n",
    "    epochs, penalties, reward = 0, 0, 0\n",
    "    \n",
    "    done = False\n",
    "    \n",
    "    while not done:\n",
    "        action = np.argmax(q_table[state])\n",
    "        next_state, reward, terminated, truncated, _ = env.step(action)\n",
    "        \n",
    "        state = next_state\n",
    "\n",
    "        done = terminated or truncated\n",
    "\n",
    "        if reward < 0:\n",
    "            penalties += reward\n",
    "\n",
    "        epochs += 1\n",
    "\n",
    "    total_penalties += penalties\n",
    "    total_epochs += epochs\n",
    "    \n",
    "    \n",
    "\n",
    "print(f\"Results after {episodes} episodes:\")\n",
    "print(f\"Average timesteps per episode: {total_epochs / episodes}\")\n",
    "print(f\"Average penalties per episode: {total_penalties / episodes}\")"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Testing"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "+---------+\n",
      "|R: | : :\u001b[35m\u001b[34;1m\u001b[43mG\u001b[0m\u001b[0m\u001b[0m|\n",
      "| : | : : |\n",
      "| : : : : |\n",
      "| | : | : |\n",
      "|Y| : |B: |\n",
      "+---------+\n",
      "  (Dropoff)\n",
      "\n",
      "Iter: 10 - Action: 5(Dropoff) - Reward 20\n"
     ]
    }
   ],
   "source": [
    "import time\n",
    "import matplotlib\n",
    "\n",
    "is_ipython = 'inline' in matplotlib.get_backend()\n",
    "if is_ipython:\n",
    "    from IPython import display\n",
    "    \n",
    "actions_str = [\"South\", \"North\", \"East\", \"West\", \"Pickup\", \"Dropoff\"]\n",
    "\n",
    "iteration = 0\n",
    "state, info = env.reset()  # reset environment to a new, random state\n",
    "env.render()\n",
    "time.sleep(0.02)\n",
    "terminated = False\n",
    "\n",
    "while not terminated:\n",
    "    action = np.argmax(q_table[state])\n",
    "    iteration += 1\n",
    "    state, reward, terminated, truncated, _ = env.step(action.item())\n",
    "    display.clear_output(wait=True)\n",
    "    print(env.render())\n",
    "    print(f\"Iter: {iteration} - Action: {action}({actions_str[action]}) - Reward {reward}\")\n",
    "    time.sleep(0.5)\n",
    "    if iteration == 200:\n",
    "        print(\"cannot converge :(\")\n",
    "        break"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": ".venv",
   "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": "3.9.13"
  },
  "orig_nbformat": 4
 },
 "nbformat": 4,
 "nbformat_minor": 2
}