gen_plot.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "from collections import defaultdict"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "\"lines4 = []\\nwith open('out_QLearningTLF_pd.csv', 'r') as f:\\n    lines4 = f.readlines()\\nlines4 = lines4[1:]\\nlines4 = [line.split(',') for line in lines4]\\n\\nlines5 = []\\nwith open('out_QLearningTLFASH_pd.csv', 'r') as f:\\n    lines5 = f.readlines()\\nlines5 = lines5[1:]\\nlines5 = [line.split(',') for line in lines5]\""
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "lines = []\n",
    "with open('out_NLGA_pd.csv', 'r') as f:\n",
    "    lines = f.readlines()\n",
    "lines = lines[1:]\n",
    "lines = [line.split(',') for line in lines]\n",
    "\n",
    "lines2 = []\n",
    "with open('out_QLearningTab_pd.csv', 'r') as f:\n",
    "    lines2 = f.readlines()\n",
    "lines2 = lines2[1:]\n",
    "lines2 = [line.split(',') for line in lines2]\n",
    "\n",
    "lines3 = []\n",
    "with open('out_QLearningTabASH_pd.csv', 'r') as f:\n",
    "    lines3 = f.readlines()\n",
    "lines3 = lines3[1:]\n",
    "lines3 = [line.split(',') for line in lines3]\n",
    "\n",
    "'''lines4 = []\n",
    "with open('out_QLearningTLF_pd.csv', 'r') as f:\n",
    "    lines4 = f.readlines()\n",
    "lines4 = lines4[1:]\n",
    "lines4 = [line.split(',') for line in lines4]\n",
    "\n",
    "lines5 = []\n",
    "with open('out_QLearningTLFASH_pd.csv', 'r') as f:\n",
    "    lines5 = f.readlines()\n",
    "lines5 = lines5[1:]\n",
    "lines5 = [line.split(',') for line in lines5]'''\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'rewards4 = [float(line[0]) for line in lines4]\\n\\nrewards5 = [float(line[0]) for line in lines5]'"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rewards = [float(line[0]) for line in lines]\n",
    "#stdevs  = [float(line[1]) for line in lines]\n",
    "\n",
    "rewards2 = [float(line[0]) for line in lines2]\n",
    "#stdevs2  = [float(line[1]) for line in lines2]\n",
    "\n",
    "rewards3 = [float(line[0]) for line in lines3]\n",
    "\n",
    "'''rewards4 = [float(line[0]) for line in lines4]\n",
    "\n",
    "rewards5 = [float(line[0]) for line in lines5]'''\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'reward_cat = defaultdict(int)\\nfor r in rewards:\\n    reward_cat[r] += 1'"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "'''reward_cat = defaultdict(int)\n",
    "for r in rewards:\n",
    "    reward_cat[r] += 1'''"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "#print(reward_cat)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1000000\n",
      "[ 0.  0.  0.  0.  0.  0.  0.  0.  0.  0.]\n",
      "[0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0]\n",
      "1000000\n",
      "[-0.05 -0.15 -0.2  -0.25 -0.3  -0.3  -0.35 -0.35 -0.35 -0.35]\n",
      "[-0.050000000000000003, -0.075000000000000011, -0.066666666666666666, -0.0625, -0.059999999999999998, -0.049999999999999996, -0.049999999999999996, -0.043749999999999997, -0.03888888888888889, -0.034999999999999996]\n",
      "1000000\n",
      "[-0.05 -0.2  -0.35 -0.45 -0.55 -0.7  -0.8  -0.9  -1.   -1.1 ]\n",
      "[-0.050000000000000003, -0.10000000000000001, -0.11666666666666665, -0.11249999999999999, -0.10999999999999999, -0.11666666666666665, -0.11428571428571428, -0.11249999999999999, -0.1111111111111111, -0.10999999999999999]\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "'cum_mean4 = []\\ncum_median4 = []\\ncum_sum4 = np.cumsum(rewards4)\\n#cum_sum12 = np.cumsum(stdevs2)\\n#cum_mean_stdevs2 = []\\nfor idx, val in enumerate(cum_sum4):\\n    sum1 = val\\n    #sum2 = cum_sum12[idx]\\n    #print(val)\\n    if idx >= 50000:\\n        sum1 -= cum_sum4[idx - 50000]\\n    cum_mean4.append(sum1/min(idx+1, 50000))\\n    #cum_mean4.append(sum1/(idx+1))\\n    #cum_mean_stdevs2.append(sum2/(idx+1))\\nprint(len(cum_mean4))\\nprint(cum_sum4[:10])\\nprint(cum_mean4[:10])\\n\\ncum_mean5 = []\\ncum_sum5 = np.cumsum(rewards5)\\nfor idx, val in enumerate(cum_sum5):\\n    sum1 = val\\n    if idx >= 50000:\\n        sum1 -= cum_sum5[idx - 50000]\\n    cum_mean5.append(sum1/min(idx+1, 50000))\\n    #cum_mean5.append(sum1/(idx+1))\\nprint(len(cum_mean5))\\nprint(cum_sum5[:10])\\nprint(cum_mean5[:10])'"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cum_mean = []\n",
    "cum_median = []\n",
    "cum_sum = np.cumsum(rewards)\n",
    "#cum_sum1 = np.cumsum(stdevs)\n",
    "cum_mean_stdevs = []\n",
    "for idx, val in enumerate(cum_sum):\n",
    "    sum1 = val\n",
    "    #sum2 = cum_sum1[idx]\n",
    "    #print(val)\n",
    "    #if idx >= 50000:\n",
    "    #    sum1 -= cum_sum[idx - 50000]\n",
    "    #cum_mean.append(sum1/min(idx+1, 50000))\n",
    "    cum_mean.append(sum1/(idx+1))\n",
    "    #cum_mean_stdevs.append(sum2/(idx+1))\n",
    "print(len(cum_mean))\n",
    "print(cum_sum[:10])\n",
    "print(cum_mean[:10])\n",
    "'''for i in range(len(rewards)):\n",
    "    cum_median.append(np.median(rewards[:i]))'''\n",
    "\n",
    "cum_mean2 = []\n",
    "cum_median2 = []\n",
    "cum_sum2 = np.cumsum(rewards2)\n",
    "#cum_sum12 = np.cumsum(stdevs2)\n",
    "cum_mean_stdevs2 = []\n",
    "for idx, val in enumerate(cum_sum2):\n",
    "    sum1 = val\n",
    "    #sum2 = cum_sum12[idx]\n",
    "    #print(val)\n",
    "    if idx >= 50000:\n",
    "        sum1 -= cum_sum2[idx - 50000]\n",
    "    cum_mean2.append(sum1/min(idx+1, 50000))\n",
    "    #cum_mean2.append(sum1/(idx+1))\n",
    "    #cum_mean_stdevs2.append(sum2/(idx+1))\n",
    "print(len(cum_mean2))\n",
    "print(cum_sum2[:10])\n",
    "print(cum_mean2[:10])\n",
    "'''for i in range(len(rewards)):\n",
    "    cum_median.append(np.median(rewards[:i]))'''\n",
    "\n",
    "cum_mean3 = []\n",
    "cum_sum3 = np.cumsum(rewards3)\n",
    "for idx, val in enumerate(cum_sum3):\n",
    "    sum1 = val\n",
    "    if idx >= 50000:\n",
    "        sum1 -= cum_sum3[idx - 50000]\n",
    "    cum_mean3.append(sum1/min(idx+1, 50000))\n",
    "    #cum_mean3.append(sum1/(idx+1))\n",
    "print(len(cum_mean3))\n",
    "print(cum_sum3[:10])\n",
    "print(cum_mean3[:10])\n",
    "'''for i in range(len(rewards)):\n",
    "    cum_median.append(np.median(rewards[:i]))'''\n",
    "\n",
    "'''cum_mean4 = []\n",
    "cum_median4 = []\n",
    "cum_sum4 = np.cumsum(rewards4)\n",
    "#cum_sum12 = np.cumsum(stdevs2)\n",
    "#cum_mean_stdevs2 = []\n",
    "for idx, val in enumerate(cum_sum4):\n",
    "    sum1 = val\n",
    "    #sum2 = cum_sum12[idx]\n",
    "    #print(val)\n",
    "    if idx >= 50000:\n",
    "        sum1 -= cum_sum4[idx - 50000]\n",
    "    cum_mean4.append(sum1/min(idx+1, 50000))\n",
    "    #cum_mean4.append(sum1/(idx+1))\n",
    "    #cum_mean_stdevs2.append(sum2/(idx+1))\n",
    "print(len(cum_mean4))\n",
    "print(cum_sum4[:10])\n",
    "print(cum_mean4[:10])\n",
    "\n",
    "cum_mean5 = []\n",
    "cum_sum5 = np.cumsum(rewards5)\n",
    "for idx, val in enumerate(cum_sum5):\n",
    "    sum1 = val\n",
    "    if idx >= 50000:\n",
    "        sum1 -= cum_sum5[idx - 50000]\n",
    "    cum_mean5.append(sum1/min(idx+1, 50000))\n",
    "    #cum_mean5.append(sum1/(idx+1))\n",
    "print(len(cum_mean5))\n",
    "print(cum_sum5[:10])\n",
    "print(cum_mean5[:10])'''\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "50.001\n"
     ]
    }
   ],
   "source": [
    "x = [(i+1)/1000 for i in range(len(cum_sum))]\n",
    "print(x[50000])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA3oAAAGDCAYAAAB9UWKAAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x12ac288ab00>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(15,6))\n",
    "plt.plot(x[000000:], cum_mean[000000:], label='Non-Learning Greedy Agent')\n",
    "plt.plot(x[000000:], cum_mean2[000000:], label='Q Learning Agent')\n",
    "plt.plot(x[000000:], cum_mean3[000000:], label='After-State Q Learning Agent')\n",
    "#plt.plot(x[000000:], cum_mean4[000000:], label='Q Learning TLF Agent')\n",
    "#plt.plot(x[000000:], cum_mean5[000000:], label='After-State Q Learning TLF Agent')\n",
    "plt.xlabel('Number of iterations (in 1000s)')\n",
    "plt.ylabel('On-line Average Reward')\n",
    "plt.title('Plot of On-line Average Reward vs No of Iterations for Different Agents')\n",
    "plt.legend()\n",
    "#plt.yticks(np.arange(-5, 1, 0.5))\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 98,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "-11511.0100424\n"
     ]
    }
   ],
   "source": [
    "print(cum_sum3[500000] - cum_sum3[500000-50000])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "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": "3.5.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}