Demo.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Eval 1: 0%\n",
      "Train current 100 episodes to 100: 0%\n",
      "Train current 100 episodes to 200: 1%\n",
      "Eval 201: 0%\n",
      "Train current 100 episodes to 300: 0%\n",
      "Train current 100 episodes to 400: 4%\n",
      "Eval 401: 20%\n",
      "Train current 100 episodes to 500: 13%\n",
      "Train current 100 episodes to 600: 20%\n",
      "Eval 601: 22%\n",
      "Train current 100 episodes to 700: 17%\n",
      "Train current 100 episodes to 800: 39%\n",
      "Eval 801: 34%\n",
      "Train current 100 episodes to 900: 30%\n",
      "Train current 100 episodes to 1000: 22%\n",
      "Eval 1001: 18%\n",
      "Train current 100 episodes to 1100: 16%\n",
      "Train current 100 episodes to 1200: 22%\n",
      "Eval 1201: 22%\n",
      "Train current 100 episodes to 1300: 30%\n",
      "Train current 100 episodes to 1400: 47%\n",
      "Eval 1401: 40%\n",
      "Train current 100 episodes to 1500: 53%\n",
      "Train current 100 episodes to 1600: 51%\n",
      "Eval 1601: 52%\n",
      "Train current 100 episodes to 1700: 62%\n",
      "Train current 100 episodes to 1800: 53%\n",
      "Eval 1801: 52%\n",
      "Train current 100 episodes to 1900: 45%\n",
      "Train current 100 episodes to 2000: 38%\n",
      "Eval 2001: 46%\n",
      "Train current 100 episodes to 2100: 44%\n",
      "Train current 100 episodes to 2200: 51%\n",
      "Eval 2201: 52%\n",
      "Train current 100 episodes to 2300: 51%\n",
      "Train current 100 episodes to 2400: 52%\n",
      "Eval 2401: 60%\n",
      "Train current 100 episodes to 2500: 62%\n",
      "Train current 100 episodes to 2600: 62%\n",
      "Eval 2601: 56%\n",
      "Train current 100 episodes to 2700: 67%\n",
      "Train current 100 episodes to 2800: 73%\n",
      "Eval 2801: 66%\n",
      "Train current 100 episodes to 2900: 72%\n",
      "Train current 100 episodes to 3000: 53%\n",
      "Eval 3001: 56%\n",
      "Train current 100 episodes to 3100: 74%\n",
      "Train current 100 episodes to 3200: 80%\n",
      "Eval 3201: 62%\n",
      "Train current 100 episodes to 3300: 79%\n",
      "Train current 100 episodes to 3400: 77%\n",
      "Eval 3401: 78%\n",
      "Train current 100 episodes to 3500: 74%\n",
      "Train current 100 episodes to 3600: 84%\n",
      "Eval 3601: 76%\n",
      "Train current 100 episodes to 3700: 76%\n",
      "Train current 100 episodes to 3800: 84%\n",
      "Eval 3801: 84%\n",
      "Train current 100 episodes to 3900: 83%\n",
      "Train current 100 episodes to 4000: 74%\n",
      "Eval 4001: 84%\n",
      "Train current 100 episodes to 4100: 84%\n",
      "Train current 100 episodes to 4200: 87%\n",
      "Eval 4201: 84%\n",
      "Train current 100 episodes to 4300: 81%\n",
      "Train current 100 episodes to 4400: 88%\n",
      "Eval 4401: 84%\n",
      "Train current 100 episodes to 4500: 82%\n",
      "Train current 100 episodes to 4600: 83%\n",
      "Eval 4601: 92%\n",
      "./model/model_4601.ckpt\n",
      "Train current 100 episodes to 4700: 80%\n",
      "Train current 100 episodes to 4800: 91%\n",
      "Eval 4801: 86%\n",
      "Train current 100 episodes to 4900: 92%\n",
      "Train current 100 episodes to 5000: 88%\n",
      "Eval 5001: 88%\n",
      "Train current 100 episodes to 5100: 88%\n",
      "Train current 100 episodes to 5200: 89%\n",
      "Eval 5201: 78%\n",
      "Train current 100 episodes to 5300: 92%\n",
      "Train current 100 episodes to 5400: 89%\n",
      "Eval 5401: 92%\n",
      "./model/model_5401.ckpt\n",
      "Train current 100 episodes to 5500: 90%\n",
      "Train current 100 episodes to 5600: 95%\n",
      "Eval 5601: 84%\n",
      "Train current 100 episodes to 5700: 85%\n",
      "Train current 100 episodes to 5800: 90%\n",
      "Eval 5801: 90%\n",
      "Train current 100 episodes to 5900: 92%\n",
      "Train current 100 episodes to 6000: 92%\n",
      "Eval 6001: 92%\n",
      "./model/model_6001.ckpt\n",
      "Train current 100 episodes to 6100: 91%\n",
      "Train current 100 episodes to 6200: 92%\n",
      "Eval 6201: 94%\n",
      "./model/model_6201.ckpt\n",
      "Train current 100 episodes to 6300: 91%\n",
      "Train current 100 episodes to 6400: 92%\n",
      "Eval 6401: 96%\n",
      "./model/model_6401.ckpt\n",
      "Train current 100 episodes to 6500: 90%\n",
      "Train current 100 episodes to 6600: 96%\n",
      "Eval 6601: 86%\n",
      "Train current 100 episodes to 6700: 95%\n",
      "Train current 100 episodes to 6800: 92%\n",
      "Eval 6801: 94%\n",
      "Train current 100 episodes to 6900: 91%\n",
      "Train current 100 episodes to 7000: 94%\n",
      "Eval 7001: 88%\n",
      "Train current 100 episodes to 7100: 96%\n",
      "Train current 100 episodes to 7200: 92%\n",
      "Eval 7201: 86%\n",
      "Train current 100 episodes to 7300: 91%\n",
      "Train current 100 episodes to 7400: 93%\n",
      "Eval 7401: 92%\n",
      "Train current 100 episodes to 7500: 90%\n",
      "Train current 100 episodes to 7600: 89%\n",
      "Eval 7601: 84%\n",
      "Train current 100 episodes to 7700: 90%\n",
      "Train current 100 episodes to 7800: 92%\n",
      "Eval 7801: 90%\n",
      "Train current 100 episodes to 7900: 97%\n",
      "Train current 100 episodes to 8000: 90%\n",
      "Eval 8001: 88%\n",
      "Train current 100 episodes to 8100: 94%\n",
      "Train current 100 episodes to 8200: 92%\n",
      "Eval 8201: 84%\n",
      "Train current 100 episodes to 8300: 95%\n",
      "Train current 100 episodes to 8400: 89%\n",
      "Eval 8401: 90%\n",
      "Train current 100 episodes to 8500: 87%\n",
      "Train current 100 episodes to 8600: 90%\n",
      "Eval 8601: 92%\n",
      "Train current 100 episodes to 8700: 97%\n",
      "Train current 100 episodes to 8800: 95%\n",
      "Eval 8801: 92%\n",
      "Train current 100 episodes to 8900: 97%\n",
      "Train current 100 episodes to 9000: 93%\n",
      "Eval 9001: 94%\n",
      "Train current 100 episodes to 9100: 95%\n",
      "Train current 100 episodes to 9200: 98%\n",
      "Eval 9201: 94%\n",
      "Train current 100 episodes to 9300: 94%\n",
      "Train current 100 episodes to 9400: 97%\n",
      "Eval 9401: 96%\n",
      "./model/model_9401.ckpt\n",
      "Train current 100 episodes to 9500: 95%\n",
      "Train current 100 episodes to 9600: 98%\n",
      "Eval 9601: 96%\n",
      "./model/model_9601.ckpt\n",
      "Train current 100 episodes to 9700: 95%\n",
      "Train current 100 episodes to 9800: 97%\n",
      "Eval 9801: 96%\n",
      "./model/model_9801.ckpt\n",
      "Train current 100 episodes to 9900: 94%\n",
      "Train current 100 episodes to 10000: 96%\n",
      "Eval 10001: 98%\n",
      "./model/model_10001.ckpt\n",
      "Train current 100 episodes to 10100: 96%\n",
      "Train current 100 episodes to 10200: 95%\n",
      "Eval 10201: 96%\n",
      "Train current 100 episodes to 10300: 98%\n",
      "Train current 100 episodes to 10400: 95%\n",
      "Eval 10401: 92%\n",
      "Train current 100 episodes to 10500: 96%\n",
      "Train current 100 episodes to 10600: 96%\n",
      "Eval 10601: 94%\n",
      "Train current 100 episodes to 10700: 96%\n",
      "Train current 100 episodes to 10800: 98%\n",
      "Eval 10801: 92%\n",
      "Train current 100 episodes to 10900: 95%\n",
      "Train current 100 episodes to 11000: 96%\n",
      "Eval 11001: 94%\n",
      "Train current 100 episodes to 11100: 95%\n",
      "Train current 100 episodes to 11200: 95%\n",
      "Eval 11201: 94%\n",
      "Train current 100 episodes to 11300: 99%\n",
      "Train current 100 episodes to 11400: 98%\n",
      "Eval 11401: 100%\n",
      "./model/model_11401.ckpt\n",
      "Rate: 100%\n"
     ]
    },
    {
     "ename": "SystemExit",
     "evalue": "",
     "output_type": "error",
     "traceback": [
      "An exception has occurred, use %tb to see the full traceback.\n",
      "\u001b[1;31mSystemExit\u001b[0m\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\Giang\\AppData\\Local\\Continuum\\anaconda3\\lib\\site-packages\\IPython\\core\\interactiveshell.py:2971: UserWarning: To exit: use 'exit', 'quit', or Ctrl-D.\n",
      "  warn(\"To exit: use 'exit', 'quit', or Ctrl-D.\", stacklevel=1)\n"
     ]
    }
   ],
   "source": [
    "import os\n",
    "import tensorflow as tf\n",
    "from agent import Agent\n",
    "import logger\n",
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline\n",
    "# os.environ[\"CUDA_DEVICE_ORDER\"] = \"PCI_BUS_ID\"\n",
    "# os.environ[\"CUDA_VISIBLE_DEVICES\"] = \"1\"  # Only run on GPU 1\n",
    "\n",
    "sess = tf.InteractiveSession()\n",
    "agent = Agent()\n",
    "sess.run(tf.global_variables_initializer())\n",
    "agent.brain.initialize_target()\n",
    "for ep in range(100001):\n",
    "    agent.perceive()  # SystemExit because of reached 100% success rate"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "7\n"
     ]
    }
   ],
   "source": [
    "buf, suc = agent.train_worker.run(0, False)\n",
    "print(len(buf))\n",
    "def plot(s):\n",
    "    fig = plt.figure()\n",
    "    fig.add_subplot(121)\n",
    "    plt.imshow(s[:, :, :-1])\n",
    "    fig.add_subplot(122)\n",
    "    plt.imshow(s[:, :, -1])\n",
    "    plt.show()\n",
    "\n",
    "def plot_multiple(ss):\n",
    "    fig = plt.figure()\n",
    "    for i in range(len(ss)):\n",
    "        fig.add_subplot(100 + len(ss)*10 + i + 1)\n",
    "        plt.imshow(ss[i][:, :, :-1])\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot(buf[0][0])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 7 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_multiple([buf[i][0] for i in range(len(buf))])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[-1.         -1.         -1.        ]\n",
      " [-1.         -1.         -1.        ]\n",
      " [-0.9999562  -1.         -1.        ]\n",
      " [-0.9997909  -1.         -1.        ]\n",
      " [-0.97064465  0.8137912  -1.        ]\n",
      " [-0.9044559   0.99563277 -1.        ]\n",
      " [-0.6950269   0.9930648  -1.        ]]\n"
     ]
    }
   ],
   "source": [
    "import numpy as np\n",
    "print(np.asarray([buf[i][1] for i in range(len(buf))]))  # print action"
   ]
  },
  {
   "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
}