Laser-Hockey-Env-DuelingDQN-FINAL.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-08-21T19:12:18.980322Z",
     "start_time": "2019-08-21T19:12:18.819269Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Cannot re-register id: LaserHockey-v0\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<module 'laser_hockey_env' from '/Users/samski/Downloads/Gym-RL-DQN-solution/laser_hockey_env.py'>"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import numpy as np\n",
    "import laser_hockey_env as lh\n",
    "import gym\n",
    "import DuelingDQAgent\n",
    "from importlib import reload\n",
    "import torch\n",
    "import pylab as plt\n",
    "import random\n",
    "# %matplotlib inline\n",
    "from mpl_toolkits.mplot3d import Axes3D\n",
    "from matplotlib import cm\n",
    "reload(DuelingDQAgent)\n",
    "reload(lh)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-08-21T19:12:19.327491Z",
     "start_time": "2019-08-21T19:12:19.321977Z"
    }
   },
   "outputs": [],
   "source": [
    "np.set_printoptions(suppress=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-12-21T13:08:25.605760Z",
     "start_time": "2018-12-21T13:08:25.593455Z"
    }
   },
   "outputs": [],
   "source": [
    "env = lh.LaserHockeyEnv(mode=lh.LaserHockeyEnv.TRAIN_SHOOTING)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-12-21T13:08:25.849615Z",
     "start_time": "2018-12-21T13:08:25.845081Z"
    }
   },
   "outputs": [],
   "source": [
    "import random\n",
    "load=False\n",
    "if load:\n",
    "    Q = torch.load('modelQ.pt')\n",
    "    Q_next = torch.load('modelQNext.pt')\n",
    "else:\n",
    "    Q = None\n",
    "    Q_next = None\n",
    "player1 = DuelingDQAgent.DQNAgent(env, Q, Q_next)\n",
    "#player2 = lh.BasicOpponent()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [],
   "source": [
    "stats = []\n",
    "losses = []\n",
    "def running_mean(x, N):\n",
    "    cumsum = np.cumsum(np.insert(x, 0, 0)) \n",
    "    return (cumsum[N:] - cumsum[:-N]) / float(N)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1: Done after 500 steps. Reward: 0\n",
      "Player 1 scored\n",
      "21: Done after 500 steps. Reward: 0\n",
      "41: Done after 500 steps. Reward: 0\n",
      "61: Done after 500 steps. Reward: 0\n",
      "81: Done after 500 steps. Reward: 0\n",
      "Player 1 scored\n",
      "Player 1 scored\n"
     ]
    },
    {
     "ename": "KeyboardInterrupt",
     "evalue": "",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mKeyboardInterrupt\u001b[0m                         Traceback (most recent call last)",
      "\u001b[0;32m<ipython-input-55-1c14bff8db15>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m\u001b[0m\n\u001b[1;32m     17\u001b[0m         \u001b[0ma2\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0;36m0\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;36m0.\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;36m0\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     18\u001b[0m         \u001b[0mpobs\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mobs\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 19\u001b[0;31m         \u001b[0mobs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mr\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0md\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0minfo\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0menv\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mstep\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mnp\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mhstack\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0ma1\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0ma2\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m     20\u001b[0m         \u001b[0mtotal_reward\u001b[0m \u001b[0;34m+=\u001b[0m \u001b[0mr\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     21\u001b[0m         \u001b[0mtransition\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m(\u001b[0m\u001b[0mpobs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mcurr_act\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mr\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mobs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0md\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/Downloads/Gym-RL-DQN-solution/laser_hockey_env.py\u001b[0m in \u001b[0;36mstep\u001b[0;34m(self, action)\u001b[0m\n\u001b[1;32m    541\u001b[0m         \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_apply_action_with_max_speed\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mplayer1\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0maction\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;36m2\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;36m10\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;32mTrue\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    542\u001b[0m         \u001b[0;31m# add her func to limit angle\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 543\u001b[0;31m         \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mplayer1\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mApplyTorque\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0maction\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;36m2\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m*\u001b[0m \u001b[0mTORQUEMULTIPLIER\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;32mTrue\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    544\u001b[0m         \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_apply_action_with_max_speed\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mplayer2\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0maction\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;36m3\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;36m5\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;36m10\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;32mFalse\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    545\u001b[0m         \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mplayer2\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mApplyTorque\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m-\u001b[0m\u001b[0maction\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;36m5\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m*\u001b[0m \u001b[0mTORQUEMULTIPLIER\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;32mTrue\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;31mKeyboardInterrupt\u001b[0m: "
     ]
    }
   ],
   "source": [
    "max_episodes=600\n",
    "max_steps=500 \n",
    "fps=50\n",
    "#mode=\"random\"\n",
    "show=False\n",
    "for i in range(max_episodes):\n",
    "    # print(\"Starting a new episode\")    \n",
    "    total_reward = 0\n",
    "    obs = env.reset()\n",
    "#     obs_agent2 = env.obs_agent_two()\n",
    "    for t in range(max_steps):\n",
    "        env.render()\n",
    "        done = False\n",
    "        curr_act = player1.act(obs)\n",
    "        a1 = env.discrete_to_continous_action(curr_act)\n",
    "        #print(curr_act)\n",
    "        a2 = [0,0.,0]\n",
    "        pobs = obs\n",
    "        obs, r, d, info = env.step(np.hstack([a1,a2]))\n",
    "        total_reward += r\n",
    "        transition = (pobs, curr_act, r, obs, d)\n",
    "        player1.store_transition(transition)            \n",
    "        if done: break    \n",
    "    losses.extend(player1.train())\n",
    "    stats.append([i,total_reward])  \n",
    "#     stats.append([i,total_reward,t+1])\n",
    "    #print(total_reward)\n",
    "    #env.close()\n",
    "#     if episode%30 == 20 and save:\n",
    "#         torch.save(player1.actor, 'modelB5a.pt')\n",
    "#         torch.save(player1.actor, 'modelB5a.pt')\n",
    "    \n",
    "    if ((i-1)%20==0):\n",
    "        #print(info)\n",
    "        print(\"{}: Done after {} steps. Reward: {}\".format(i, t+1, total_reward))\n",
    "stats_np = np.asarray(stats)\n",
    "fig=plt.figure(figsize=(6,3.8))\n",
    "plt.plot(stats_np[:,1], label=\"reward\")\n",
    "plt.plot(running_mean(stats_np[:,1],20), label=\"smoothed reward\")\n",
    "plt.legend()\n",
    "fig.savefig(\"DDQN_Shooting_Graph_throwout.pdf\") \n",
    "\n",
    "env.close()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "#         torch.save(player1.Q, 'modelQ.pt')\n",
    "#         torch.save(player1.Q_next, 'modelQNext.pt')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "env = lh.LaserHockeyEnv(mode=lh.LaserHockeyEnv.TRAIN_DEFENSE)\n",
    "load=False\n",
    "if load:\n",
    "    Q = torch.load('modelQ.pt')\n",
    "    Q_next = torch.load('modelQNext.pt')\n",
    "else:\n",
    "    Q = None\n",
    "    Q_next = None\n",
    "player1 = DuelingDQAgent.DQNAgent(env, Q, Q_next)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [],
   "source": [
    "stats = []\n",
    "losses = []\n",
    "def running_mean(x, N):\n",
    "    cumsum = np.cumsum(np.insert(x, 0, 0)) \n",
    "    return (cumsum[N:] - cumsum[:-N]) / float(N)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Player 1 scored\n",
      "Player 2 scored\n",
      "1: Done after 500 steps. Reward: -3600\n",
      "Player 2 scored\n",
      "Player 1 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 1 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 1 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 1 scored\n",
      "Player 1 scored\n",
      "Player 2 scored\n",
      "Player 1 scored\n",
      "Player 2 scored\n",
      "21: Done after 500 steps. Reward: -4580\n",
      "Player 1 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 1 scored\n",
      "Player 2 scored\n",
      "Player 1 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 1 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 1 scored\n",
      "Player 1 scored\n",
      "41: Done after 500 steps. Reward: 3980\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 1 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 1 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 1 scored\n",
      "Player 1 scored\n",
      "61: Done after 500 steps. Reward: 3790\n",
      "Player 2 scored\n",
      "Player 1 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 1 scored\n",
      "Player 1 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 1 scored\n",
      "Player 2 scored\n",
      "81: Done after 500 steps. Reward: -4570\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 1 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 1 scored\n",
      "Player 1 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 1 scored\n",
      "Player 1 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 1 scored\n",
      "101: Done after 500 steps. Reward: 0\n",
      "Player 1 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 1 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "121: Done after 500 steps. Reward: -4590\n",
      "Player 2 scored\n",
      "Player 1 scored\n",
      "Player 2 scored\n",
      "Player 1 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 1 scored\n",
      "Player 1 scored\n",
      "141: Done after 500 steps. Reward: 0\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 1 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 1 scored\n",
      "Player 1 scored\n",
      "Player 1 scored\n",
      "Player 2 scored\n",
      "Player 1 scored\n",
      "Player 1 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 1 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 1 scored\n",
      "Player 2 scored\n",
      "Player 1 scored\n",
      "Player 2 scored\n",
      "161: Done after 500 steps. Reward: -4560\n",
      "Player 2 scored\n",
      "Player 1 scored\n",
      "Player 1 scored\n",
      "Player 2 scored\n",
      "Player 1 scored\n",
      "Player 1 scored\n",
      "Player 2 scored\n",
      "Player 1 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 1 scored\n",
      "Player 1 scored\n",
      "Player 1 scored\n",
      "Player 2 scored\n",
      "181: Done after 500 steps. Reward: -4580\n",
      "Player 1 scored\n",
      "Player 1 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 1 scored\n",
      "Player 1 scored\n",
      "Player 2 scored\n",
      "Player 1 scored\n",
      "Player 1 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 1 scored\n",
      "Player 1 scored\n",
      "Player 2 scored\n",
      "201: Done after 500 steps. Reward: 0\n",
      "Player 2 scored\n",
      "Player 1 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 1 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 1 scored\n",
      "Player 2 scored\n",
      "Player 1 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 1 scored\n",
      "Player 1 scored\n",
      "221: Done after 500 steps. Reward: 4560\n",
      "Player 1 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 1 scored\n",
      "Player 1 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 1 scored\n",
      "Player 1 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "241: Done after 500 steps. Reward: 0\n",
      "Player 2 scored\n",
      "Player 1 scored\n",
      "Player 2 scored\n",
      "Player 1 scored\n",
      "Player 2 scored\n",
      "Player 1 scored\n",
      "Player 2 scored\n",
      "Player 1 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 1 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 1 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "261: Done after 500 steps. Reward: -1680\n",
      "Player 1 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 1 scored\n",
      "Player 2 scored\n",
      "Player 1 scored\n",
      "Player 1 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 1 scored\n",
      "Player 1 scored\n",
      "281: Done after 500 steps. Reward: 0\n",
      "Player 1 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 1 scored\n",
      "Player 1 scored\n",
      "Player 1 scored\n",
      "Player 1 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 1 scored\n",
      "Player 1 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 1 scored\n",
      "301: Done after 500 steps. Reward: 0\n",
      "Player 1 scored\n",
      "Player 2 scored\n",
      "Player 1 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 1 scored\n",
      "Player 1 scored\n",
      "Player 1 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 1 scored\n",
      "321: Done after 500 steps. Reward: 0\n",
      "Player 1 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 1 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 1 scored\n",
      "Player 1 scored\n",
      "Player 2 scored\n",
      "341: Done after 500 steps. Reward: -4570\n",
      "Player 2 scored\n",
      "Player 1 scored\n",
      "Player 2 scored\n",
      "Player 1 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 1 scored\n",
      "Player 2 scored\n",
      "Player 1 scored\n",
      "Player 2 scored\n",
      "Player 1 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "361: Done after 500 steps. Reward: -4570\n",
      "Player 1 scored\n",
      "Player 1 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 1 scored\n",
      "Player 2 scored\n",
      "Player 1 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "381: Done after 500 steps. Reward: -4570\n",
      "Player 1 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 1 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 1 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 1 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 1 scored\n",
      "401: Done after 500 steps. Reward: 1980\n",
      "Player 1 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 1 scored\n",
      "Player 1 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 1 scored\n",
      "Player 1 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "421: Done after 500 steps. Reward: -4590\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 1 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 1 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "441: Done after 500 steps. Reward: -4560\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 1 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 1 scored\n",
      "Player 2 scored\n",
      "Player 1 scored\n",
      "Player 1 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "461: Done after 500 steps. Reward: -4570\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 1 scored\n",
      "Player 1 scored\n",
      "Player 2 scored\n",
      "Player 1 scored\n",
      "Player 1 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "481: Done after 500 steps. Reward: 0\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 1 scored\n",
      "Player 2 scored\n",
      "Player 1 scored\n",
      "Player 2 scored\n",
      "Player 1 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 1 scored\n",
      "501: Done after 500 steps. Reward: 3970\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 1 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "521: Done after 500 steps. Reward: -4570\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 1 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 1 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "541: Done after 500 steps. Reward: -4590\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 1 scored\n",
      "Player 2 scored\n",
      "Player 1 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 1 scored\n",
      "561: Done after 500 steps. Reward: 4020\n",
      "Player 2 scored\n",
      "Player 1 scored\n",
      "Player 2 scored\n",
      "Player 1 scored\n",
      "Player 1 scored\n",
      "Player 1 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 1 scored\n",
      "Player 1 scored\n",
      "Player 1 scored\n",
      "Player 2 scored\n",
      "Player 1 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "581: Done after 500 steps. Reward: -4570\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 1 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 1 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 1 scored\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x273.6 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "max_episodes=600\n",
    "max_steps=500 \n",
    "fps=50\n",
    "#mode=\"random\"\n",
    "show=False\n",
    "for i in range(max_episodes):\n",
    "    # print(\"Starting a new episode\")    \n",
    "    total_reward = 0\n",
    "    obs = env.reset()\n",
    "#     obs_agent2 = env.obs_agent_two()\n",
    "    for t in range(max_steps):\n",
    "        #env.render()\n",
    "        done = False\n",
    "        curr_act = player1.act(obs)\n",
    "        a1 = env.discrete_to_continous_action(curr_act)\n",
    "        #print(curr_act)\n",
    "        a2 = [0,0.,0]\n",
    "        pobs = obs\n",
    "        obs, r, d, info = env.step(np.hstack([a1,a2]))\n",
    "        total_reward += r\n",
    "        transition = (pobs, curr_act, r, obs, d)\n",
    "        player1.store_transition(transition)            \n",
    "        if done: break    \n",
    "    losses.extend(player1.train())\n",
    "    stats.append([i,total_reward])  \n",
    "#     stats.append([i,total_reward,t+1])\n",
    "    #print(total_reward)\n",
    "    #env.close()\n",
    "#     if episode%30 == 20 and save:\n",
    "#         torch.save(player1.actor, 'modelB5a.pt')\n",
    "#         torch.save(player1.actor, 'modelB5a.pt')\n",
    "    \n",
    "    if ((i-1)%20==0):\n",
    "        #print(info)\n",
    "        print(\"{}: Done after {} steps. Reward: {}\".format(i, t+1, total_reward))\n",
    "stats_np = np.asarray(stats)\n",
    "fig=plt.figure(figsize=(6,3.8))\n",
    "plt.plot(stats_np[:,1], label=\"reward\")\n",
    "plt.plot(running_mean(stats_np[:,1],20), label=\"smoothed reward\")\n",
    "plt.legend()\n",
    "fig.savefig(\"DDQN_Defense_Graph_throwout.pdf\")    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [],
   "source": [
    "#         torch.save(player1.Q, 'modelQShootDef.pt')\n",
    "#         torch.save(player1.Q_next, 'modelQNextShootDef.pt')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Cannot re-register id: LaserHockey-v0\n"
     ]
    }
   ],
   "source": [
    "reload(lh)\n",
    "env = lh.LaserHockeyEnv()\n",
    "o = env.reset()\n",
    "\n",
    "load=False\n",
    "if load:\n",
    "    Q = torch.load('modelQShootDef.pt')\n",
    "    Q_next = torch.load('modelQNextShootDef.pt')\n",
    "else:\n",
    "    Q = None\n",
    "    Q_next = None\n",
    "player1 = DuelingDQAgent.DQNAgent(env, Q, Q_next)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 1 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 1 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 1 scored\n",
      "Player 2 scored\n",
      "Player 1 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 1 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 1 scored\n",
      "Player 2 scored\n",
      "Player 1 scored\n",
      "Player 2 scored\n",
      "Player 1 scored\n",
      "Player 1 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 1 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 1 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 1 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 1 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 1 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n",
      "Player 1 scored\n",
      "Player 2 scored\n",
      "Player 2 scored\n"
     ]
    },
    {
     "ename": "KeyboardInterrupt",
     "evalue": "",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mKeyboardInterrupt\u001b[0m                         Traceback (most recent call last)",
      "\u001b[0;32m<ipython-input-30-ccb7bfb78f4e>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m\u001b[0m\n\u001b[1;32m      6\u001b[0m     \u001b[0mobs_agent2\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0menv\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mobs_agent_two\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      7\u001b[0m     \u001b[0;32mfor\u001b[0m \u001b[0mt\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mrange\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mmax_steps\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 8\u001b[0;31m         \u001b[0menv\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mrender\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m      9\u001b[0m         \u001b[0mcurr_act\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mplayer1\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mact\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mobs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     10\u001b[0m         \u001b[0ma1\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0menv\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdiscrete_to_continous_action\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mcurr_act\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/Downloads/Gym-RL-DQN-solution/laser_hockey_env.py\u001b[0m in \u001b[0;36mrender\u001b[0;34m(self, mode)\u001b[0m\n\u001b[1;32m    590\u001b[0m         \u001b[0;31m# self.score_label.draw()\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    591\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 592\u001b[0;31m         \u001b[0;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mviewer\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mrender\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mreturn_rgb_array\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mmode\u001b[0m \u001b[0;34m==\u001b[0m \u001b[0;34m'rgb_array'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    593\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    594\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0mclose\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/anaconda3/lib/python3.7/site-packages/gym/envs/classic_control/rendering.py\u001b[0m in \u001b[0;36mrender\u001b[0;34m(self, return_rgb_array)\u001b[0m\n\u001b[1;32m    125\u001b[0m             \u001b[0marr\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0marr\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mreshape\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mbuffer\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mheight\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mbuffer\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mwidth\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;36m4\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    126\u001b[0m             \u001b[0marr\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0marr\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m-\u001b[0m\u001b[0;36m1\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;36m0\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;36m3\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 127\u001b[0;31m         \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mwindow\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mflip\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    128\u001b[0m         \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0monetime_geoms\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    129\u001b[0m         \u001b[0;32mreturn\u001b[0m \u001b[0marr\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0mreturn_rgb_array\u001b[0m \u001b[0;32melse\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0misopen\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/anaconda3/lib/python3.7/site-packages/pyglet/window/cocoa/__init__.py\u001b[0m in \u001b[0;36mflip\u001b[0;34m(self)\u001b[0m\n\u001b[1;32m    287\u001b[0m         \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdraw_mouse_cursor\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    288\u001b[0m         \u001b[0;32mif\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mcontext\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 289\u001b[0;31m             \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mcontext\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mflip\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    290\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    291\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0mdispatch_events\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/anaconda3/lib/python3.7/site-packages/pyglet/gl/cocoa.py\u001b[0m in \u001b[0;36mflip\u001b[0;34m(self)\u001b[0m\n\u001b[1;32m    325\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    326\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0mflip\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 327\u001b[0;31m         \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_nscontext\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mflushBuffer\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m",
      "\u001b[0;32m~/anaconda3/lib/python3.7/site-packages/pyglet/libs/darwin/cocoapy/runtime.py\u001b[0m in \u001b[0;36m__call__\u001b[0;34m(self, *args)\u001b[0m\n\u001b[1;32m    783\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0m__call__\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m*\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    784\u001b[0m         \u001b[0;34m\"\"\"Call the method with the given arguments.\"\"\"\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 785\u001b[0;31m         \u001b[0;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mmethod\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mobjc_id\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m*\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    786\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    787\u001b[0m \u001b[0;31m######################################################################\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/anaconda3/lib/python3.7/site-packages/pyglet/libs/darwin/cocoapy/runtime.py\u001b[0m in \u001b[0;36m__call__\u001b[0;34m(self, objc_id, *args)\u001b[0m\n\u001b[1;32m    753\u001b[0m         \u001b[0mf\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mget_callable\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    754\u001b[0m         \u001b[0;32mtry\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 755\u001b[0;31m             \u001b[0mresult\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mf\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mobjc_id\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mselector\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m*\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    756\u001b[0m             \u001b[0;31m# Convert result to python type if it is a instance or class pointer.\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    757\u001b[0m             \u001b[0;32mif\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mrestype\u001b[0m \u001b[0;34m==\u001b[0m \u001b[0mObjCInstance\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;31mKeyboardInterrupt\u001b[0m: "
     ]
    }
   ],
   "source": [
    "player2 = lh.BasicOpponent()\n",
    "\n",
    "for i in range(max_episodes):\n",
    "    # print(\"Starting a new episode\")    \n",
    "    obs = env.reset()\n",
    "    obs_agent2 = env.obs_agent_two()\n",
    "    for t in range(max_steps):\n",
    "        env.render()\n",
    "        curr_act = player1.act(obs)\n",
    "        a1 = env.discrete_to_continous_action(curr_act)\n",
    "        a2 = player2.act(obs_agent2)\n",
    "        obs, r, d, info = env.step(np.hstack([a1,a2]))\n",
    "\n",
    "        obs_buffer.append(obs)\n",
    "        obs_agent2 = env.obs_agent_two()\n",
    "        if d: break\n",
    "env.close()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-12-20T20:37:41.013424Z",
     "start_time": "2018-12-20T20:37:41.009298Z"
    }
   },
   "source": [
    "# Hand-crafted Opponent"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-08-21T19:16:04.371746Z",
     "start_time": "2019-08-21T19:16:04.353350Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Cannot re-register id: LaserHockey-v0\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<module 'laser_hockey_env' from '/Users/samski/Downloads/Gym-RL-DQN-solution/laser_hockey_env.py'>"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "reload(lh)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-08-21T19:16:04.516774Z",
     "start_time": "2019-08-21T19:16:04.510863Z"
    }
   },
   "outputs": [],
   "source": [
    "env = lh.LaserHockeyEnv()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-08-21T19:16:04.729445Z",
     "start_time": "2019-08-21T19:16:04.666357Z"
    }
   },
   "outputs": [],
   "source": [
    "o = env.reset()\n",
    "#env.render()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-08-21T19:16:09.794765Z",
     "start_time": "2019-08-21T19:16:09.790764Z"
    }
   },
   "outputs": [],
   "source": [
    "player1 = lh.BasicOpponent()\n",
    "player2 = lh.BasicOpponent()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-08-21T19:16:10.202581Z",
     "start_time": "2019-08-21T19:16:10.198608Z"
    }
   },
   "outputs": [],
   "source": [
    "obs_buffer = []"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-08-21T19:16:22.594530Z",
     "start_time": "2019-08-21T19:16:18.495500Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Player 1 scored\n"
     ]
    }
   ],
   "source": [
    "obs = env.reset()\n",
    "obs_agent2 = env.obs_agent_two()\n",
    "for _ in range(300):\n",
    "    #env.render()\n",
    "    a1 = player1.act(obs)\n",
    "    a2 = player2.act(obs_agent2)\n",
    "    obs, r, d, info = env.step(np.hstack([a1,a2]))\n",
    "    \n",
    "    obs_buffer.append(obs)\n",
    "    obs_agent2 = env.obs_agent_two()\n",
    "    if d: break"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-12-28T12:27:58.824246Z",
     "start_time": "2018-12-28T12:27:58.813103Z"
    }
   },
   "outputs": [],
   "source": [
    "obs_buffer = np.asarray(obs_buffer)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-08-21T19:16:27.114353Z",
     "start_time": "2019-08-21T19:16:27.103201Z"
    }
   },
   "outputs": [],
   "source": [
    "np.mean(obs_buffer,axis=0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-08-21T19:16:27.902145Z",
     "start_time": "2019-08-21T19:16:27.894650Z"
    }
   },
   "outputs": [],
   "source": [
    "np.std(obs_buffer,axis=0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-08-21T19:16:28.497119Z",
     "start_time": "2019-08-21T19:16:28.492780Z"
    }
   },
   "outputs": [],
   "source": [
    "scaling = [ 1.0,  1.0 , 3.14, 4.0, 4.0, 2.0,  \n",
    "            1.0,  1.0,  3.14, 4.0, 4.0, 2.0,  \n",
    "            2.0, 2.0, 10.0, 10.0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "env.close()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Human Opponent"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-08-21T19:16:32.838454Z",
     "start_time": "2019-08-21T19:16:32.834847Z"
    }
   },
   "outputs": [],
   "source": [
    "import time"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-08-21T19:16:33.295971Z",
     "start_time": "2019-08-21T19:16:33.290071Z"
    }
   },
   "outputs": [],
   "source": [
    "env = lh.LaserHockeyEnv()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-08-21T19:16:33.801588Z",
     "start_time": "2019-08-21T19:16:33.738351Z"
    }
   },
   "outputs": [],
   "source": [
    "o = env.reset()\n",
    "env.render()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-08-21T19:16:38.262611Z",
     "start_time": "2019-08-21T19:16:38.257285Z"
    }
   },
   "outputs": [],
   "source": [
    "player1 = lh.HumanOpponent(env=env, player=1)\n",
    "player2 = lh.BasicOpponent()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-08-21T19:17:37.603826Z",
     "start_time": "2019-08-21T19:17:34.793733Z"
    },
    "scrolled": false
   },
   "outputs": [],
   "source": [
    "obs = env.reset()\n",
    "time.sleep(1)\n",
    "obs_agent2 = env.obs_agent_two()\n",
    "for _ in range(1600):\n",
    "    env.render()\n",
    "    a1 = player1.act(obs)\n",
    "    a2 = player2.act(obs_agent2)\n",
    "    obs, r, d, info = env.step(np.hstack([a1,a2]))    \n",
    "    obs_agent2 = env.obs_agent_two()\n",
    "    if d: break"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Gym-RL",
   "language": "python",
   "name": "gym-rl"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.3"
  },
  "toc": {
   "base_numbering": 1,
   "nav_menu": {},
   "number_sections": true,
   "sideBar": true,
   "skip_h1_title": false,
   "title_cell": "Table of Contents",
   "title_sidebar": "Contents",
   "toc_cell": false,
   "toc_position": {},
   "toc_section_display": true,
   "toc_window_display": false
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}