Navigation.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Navigation\n",
    "\n",
    "---\n",
    "\n",
    "In this notebook, we will train an RL agent how to navigate into a large square environment collecting yellow bananas while avoidijng blue ones. This work makes part of the first project of the [Deep Reinforcement Learning Nanodegree](https://www.udacity.com/course/deep-reinforcement-learning-nanodegree--nd893).\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "from unityagents import UnityEnvironment\n",
    "import random\n",
    "import torch\n",
    "import numpy as np\n",
    "from collections import deque\n",
    "import matplotlib.pyplot as plt\n",
    "#%matplotlib inline\n",
    "\n",
    "from agent import Agent"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "As we will start the environment please change the `file_name` parameter to match the location of the Unity environment that you downloaded."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:unityagents:\n",
      "'Academy' started successfully!\n",
      "Unity Academy name: Academy\n",
      "        Number of Brains: 1\n",
      "        Number of External Brains : 1\n",
      "        Lesson number : 0\n",
      "        Reset Parameters :\n",
      "\t\t\n",
      "Unity brain name: BananaBrain\n",
      "        Number of Visual Observations (per agent): 0\n",
      "        Vector Observation space type: continuous\n",
      "        Vector Observation space size (per agent): 37\n",
      "        Number of stacked Vector Observation: 1\n",
      "        Vector Action space type: discrete\n",
      "        Vector Action space size (per agent): 4\n",
      "        Vector Action descriptions: , , , \n"
     ]
    }
   ],
   "source": [
    "path=\"/home/rihab/Documents/bananaproject/Banana_Linux/Banana.x86_64\"\n",
    "\n",
    "#Load the banana environment\n",
    "env = UnityEnvironment(file_name=path)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "#environment has brains, each responsible of deciding the actions of their associated agents => since it's a single agent game we set the first brain \n",
    "#as the default brain we're controlling\n",
    "#get the default brain\n",
    "brain_name = env.brain_names[0]\n",
    "brain = env.brains[brain_name]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 2. Examine the State and Action Spaces\n",
    "\n",
    "The simulation contains a single agent that navigates a large environment.  At each time step, it has four actions at its disposal:\n",
    "- `0` - walk forward \n",
    "- `1` - walk backward\n",
    "- `2` - turn left\n",
    "- `3` - turn right\n",
    "\n",
    "The state space has `37` dimensions and contains the agent's velocity, along with ray-based perception of objects around agent's forward direction.  A reward of `+1` is provided for collecting a yellow banana, and a reward of `-1` is provided for collecting a blue banana. \n",
    "\n",
    "Run the code cell below to print some information about the environment."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number of agents: 1\n",
      "Number of actions: 4\n",
      "States look like: [1.         0.         0.         0.         0.84408134 0.\n",
      " 0.         1.         0.         0.0748472  0.         1.\n",
      " 0.         0.         0.25755    1.         0.         0.\n",
      " 0.         0.74177343 0.         1.         0.         0.\n",
      " 0.25854847 0.         0.         1.         0.         0.09355672\n",
      " 0.         1.         0.         0.         0.31969345 0.\n",
      " 0.        ]\n",
      "States have length: 37\n"
     ]
    }
   ],
   "source": [
    "# reset the environment\n",
    "env_info = env.reset(train_mode=False)[brain_name]\n",
    "\n",
    "# number of agents in the environment\n",
    "print('Number of agents:', len(env_info.agents))\n",
    "\n",
    "# number of actions\n",
    "action_size = brain.vector_action_space_size\n",
    "print('Number of actions:', action_size)\n",
    "\n",
    "# examine the state space \n",
    "state = env_info.vector_observations[0]\n",
    "print('States look like:', state)\n",
    "state_size = len(state)\n",
    "print('States have length:', state_size)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 3. Train the model!\n",
    "\n",
    "Now it's your turn to train your own agent to solve the environment!  When training the environment, set `train_mode=True`, so that the line for resetting the environment looks like the following:\n",
    "```python\n",
    "env_info = env.reset(train_mode=True)[brain_name]\n",
    "```"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "#instanciate the agent\n",
    "agent = Agent(state_size=37, action_size=4, seed=0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Episode 100\tAverage Score: 4.49\n",
      "Episode 200\tAverage Score: 8.02\n",
      "Episode 300\tAverage Score: 11.58\n",
      "Episode 346\tAverage Score: 13.02\n",
      "Environment solved in 346 episodes!\tAverage Score: 13.02\n"
     ]
    }
   ],
   "source": [
    "#start training\n",
    "\n",
    "def dqn(n_episodes=1000, max_t=1000, eps_start=0.5, eps_end=0.01, eps_decay=0.99, train=True):\n",
    "    \"\"\"\n",
    "    Params\n",
    "    ======\n",
    "        n_episodes (int): maximum number of training episodes\n",
    "        max_t (int): maximum number of timesteps per episode\n",
    "        eps_start (float): starting value of epsilon, for epsilon-greedy action selection\n",
    "        eps_end (float): minimum value of epsilon\n",
    "        eps_decay (float): multiplicative factor (per episode) for decreasing epsilon\n",
    "    \"\"\"\n",
    "    scores = []                        # list containing scores from each episode\n",
    "    scores_window = deque(maxlen=100)  # last 100 scores\n",
    "    eps = eps_start                    # initialize epsilon\n",
    "    for i_episode in range(1, n_episodes+1):\n",
    "        eps = eps / i_episode\n",
    "        env_info = env.reset(train_mode=train)[brain_name]            #reset the environment\n",
    "        state= env_info.vector_observations[0] \n",
    "        score = 0\n",
    "        for t in range(max_t):\n",
    "            action = agent.act(state, eps)                           #select an action\n",
    "            env_info = env.step(action)[brain_name]                  #send action to the enviornment\n",
    "            next_state = env_info.vector_observations[0]             #get the next state\n",
    "            reward = env_info.rewards[0]                             #get th reward\n",
    "            done= env_info.local_done[0]                             #if episode has  finished\n",
    "            agent.step(state, action, reward, next_state, done)\n",
    "            state = next_state\n",
    "            score += reward\n",
    "            if done:\n",
    "                break \n",
    "        scores_window.append(score)       # save most recent score\n",
    "        scores.append(score)              # save most recent score\n",
    "        #eps = max(eps_end, eps_decay*eps) # decrease epsilon\n",
    "        print('\\rEpisode {}\\tAverage Score: {:.2f}'.format(i_episode, np.mean(scores_window)), end=\"\")\n",
    "        if i_episode % 100 == 0:\n",
    "            print('\\rEpisode {}\\tAverage Score: {:.2f}'.format(i_episode, np.mean(scores_window)))\n",
    "        if np.mean(scores_window)>=13.0:\n",
    "            print('\\nEnvironment solved in {:d} episodes!\\tAverage Score: {:.2f}'.format(i_episode, np.mean(scores_window)))\n",
    "            torch.save(agent.qnetwork_local.state_dict(), 'saved_model_weights.pth')\n",
    "            break\n",
    "    return scores\n",
    "\n",
    "scores = dqn()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 4. Plot the average score over episodes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# plot the scores\n",
    "fig = plt.figure()\n",
    "ax = fig.add_subplot(111)\n",
    "plt.plot(np.arange(len(scores)), scores)\n",
    "plt.ylabel('Score')\n",
    "plt.xlabel('Episode')\n",
    "plt.title('Plot of the score over episodes')\n",
    "fig.savefig('socres.png')\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 5. Watch the trained model playing the game\n",
    "To watch the agent playing the game, we need first to load the trained weights, so we assign the path of the weights in the filename parameter"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Episode 100\tAverage Score: 10.91\n"
     ]
    }
   ],
   "source": [
    "agent = Agent(state_size=37, action_size=4, seed=0, filename='saved_model_weights.pth')\n",
    "scores = dqn(n_episodes=100, train=False)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "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": "3.6.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}