10_theano_toy_detector.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# <center> Toy Detector </center>\n",
    "\n",
    "<img src=\"images/toyd.png\" width = \"100%\">\n",
    "\n",
    "## Imports"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "%matplotlib inline\n",
    "\n",
    "import numpy\n",
    "import matplotlib.pyplot as plt\n",
    "import math\n",
    "\n",
    "rng = numpy.random # random number generator"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Settings"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "### DETECTOR ###\n",
    "\n",
    "dim = [100,50]      # dimension [width, height]\n",
    "N = dim[0] * dim[1] # number of blocks in the detector\n",
    "\n",
    "### EVENT ###\n",
    "\n",
    "nTracks = 5 # maximum no. of tracks"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Event"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "class Event:\n",
    "    # initialize with random vertex and generate tracks\n",
    "    def __init__ (self):\n",
    "        # vertex = [x,y]\n",
    "        self.vertex =  dim * rng.sample(2)\n",
    "        self.vertex[1] = dim[1] / 2 # in this version all vertices at y-center\n",
    "        # each event has random number of tracks\n",
    "        self.tracks = [self.genTrack() for i in range(rng.randint(2,nTracks))]\n",
    "        # data represents as array of 0 (no track) and 1 (track) points in the detector\n",
    "        self.data = [0] * N\n",
    "        # fill data with tracks\n",
    "        for t in self.tracks: self.genData(t[0], t[1])\n",
    "    # generate track (random line coming from vertex)\n",
    "    def genTrack (self):\n",
    "        # y = ax + b\n",
    "        a = 2.0 * rng.sample() - 1.0\n",
    "        b = self.vertex[1] - a * self.vertex[0]\n",
    "        return [a,b]\n",
    "    # generate data from tracks\n",
    "    def genData (self, a, b):\n",
    "        # track starts in the vertex\n",
    "        start = int(self.vertex[0])\n",
    "        # track length is random (but not smaller than 5)\n",
    "        end = start + int((dim[0] - start) * rng.sample())\n",
    "        # \"convert\" track to detector points\n",
    "        for x in range(start, end):\n",
    "            y = int(a * x + b)  # round y\n",
    "            if y < 0 or y >= dim[1]: break\n",
    "            self.data[x + y * dim[0]] = 1"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Event display"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "class EventDisplay:\n",
    "    def __init__(self, a = 1, b = 1):\n",
    "        fig, plots = plt.subplots(a,b)                      # create a axb grid of plots\n",
    "        fig.set_size_inches((a, b) * fig.get_size_inches()) # adjust horizontal size \n",
    "        for i in range(a):\n",
    "            for j in range(b):\n",
    "                self.showEvent(plots[i][j], Event())\n",
    "        plt.show()\n",
    "    def showEvent (self, plot, e):\n",
    "        # axis ranges\n",
    "        plot.set_xlim([0,dim[0]])\n",
    "        plot.set_ylim([0,dim[1]])\n",
    "        # ticks\n",
    "        plot.grid(b=True, which='major', color='k', linestyle='-')\n",
    "        plot.grid(b=True, which='minor', color='k', linestyle='-', alpha=0.2)\n",
    "        plot.minorticks_on()\n",
    "        # data\n",
    "        X = []\n",
    "        Y = []\n",
    "        for i in range(dim[0]):\n",
    "            for j in range(dim[1]):\n",
    "                if e.data[i + dim[0] * j]:\n",
    "                    X.append(i)\n",
    "                    Y.append(j)\n",
    "        plot.scatter(X, Y, color = 'k', marker = 's')\n",
    "        # true tracks \n",
    "        x = numpy.arange(e.vertex[0], dim[0], 1)\n",
    "        for t in e.tracks:\n",
    "            plot.plot(x, t[0] * x + t[1], 'b')\n",
    "        # vetex\n",
    "        plot.scatter(e.vertex[0], e.vertex[1], color = 'r', marker = 's')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Event examples"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABBIAAAK/CAYAAADUCu7MAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nJEmSJEmSJEmSJEmSJCk//h/YDBBnBrmZ9gAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fdc43c8f490>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<__main__.EventDisplay instance at 0x7fdc43c7cea8>"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "EventDisplay(3,3)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Full code to run on  grid\n",
    "\n",
    "https://github.com/TomaszGolan/mlScratchpad/blob/master/10_theano_toy_detector.py\n",
    "\n",
    "Results:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "Epoch: 1, cost = 0.001240\n",
    "Epoch: 2, cost = 0.002893\n",
    "Epoch: 3, cost = 0.003003\n",
    "Epoch: 4, cost = 0.002792\n",
    "Epoch: 5, cost = 0.002170\n",
    "Epoch: 6, cost = 0.002202\n",
    "Epoch: 7, cost = 0.003291\n",
    "Epoch: 8, cost = 0.005907\n",
    "Epoch: 9, cost = 0.019042\n",
    "Epoch: 10, cost = 0.026370\n",
    "\n",
    "Reconstruted events: 43.90%\n",
    "\t within 1 planes: 13.300000%\n",
    "\t within 2 planes: 10.000000%\n",
    "\t within 3 planes: 7.500000%\n",
    "\t within 4 planes: 7.200000%\n",
    "\t within 5 planes: 5.900000%\n",
    "\n",
    "Epoch: 11, cost = 0.027898\n",
    "Epoch: 12, cost = 0.029723\n",
    "Epoch: 13, cost = 0.030550\n",
    "Epoch: 14, cost = 0.031496\n",
    "Epoch: 15, cost = 0.031119\n",
    "Epoch: 16, cost = 0.031325\n",
    "Epoch: 17, cost = 0.031232\n",
    "Epoch: 18, cost = 0.031426\n",
    "Epoch: 19, cost = 0.031666\n",
    "Epoch: 20, cost = 0.031602\n",
    "\n",
    "Reconstruted events: 52.20%\n",
    "\t within 1 planes: 12.400000%\n",
    "\t within 2 planes: 13.100000%\n",
    "\t within 3 planes: 11.400000%\n",
    "\t within 4 planes: 7.100000%\n",
    "\t within 5 planes: 8.200000%\n",
    "\n",
    "Epoch: 21, cost = 0.031348\n",
    "Epoch: 22, cost = 0.031554\n",
    "Epoch: 23, cost = 0.031627\n",
    "Epoch: 24, cost = 0.031989\n",
    "Epoch: 25, cost = 0.033325\n",
    "Epoch: 26, cost = 0.035037\n",
    "Epoch: 27, cost = 0.036069\n",
    "Epoch: 28, cost = 0.036300\n",
    "Epoch: 29, cost = 0.036034\n",
    "Epoch: 30, cost = 0.035018\n",
    "\n",
    "Reconstruted events: 52.90%\n",
    "\t within 1 planes: 14.800000%\n",
    "\t within 2 planes: 12.000000%\n",
    "\t within 3 planes: 9.800000%\n",
    "\t within 4 planes: 8.700000%\n",
    "\t within 5 planes: 7.600000%\n",
    "\n",
    "Epoch: 31, cost = 0.032845\n",
    "Epoch: 32, cost = 0.028074\n",
    "Epoch: 33, cost = 0.017479\n",
    "Epoch: 34, cost = 0.002397\n",
    "Epoch: 35, cost = 0.000866\n",
    "Epoch: 36, cost = 0.004548\n",
    "Epoch: 37, cost = 0.006696\n",
    "Epoch: 38, cost = 0.007563\n",
    "Epoch: 39, cost = 0.007621\n",
    "Epoch: 40, cost = 0.007297\n",
    "\n",
    "Reconstruted events: 52.30%\n",
    "\t within 1 planes: 15.200000%\n",
    "\t within 2 planes: 11.700000%\n",
    "\t within 3 planes: 6.900000%\n",
    "\t within 4 planes: 10.600000%\n",
    "\t within 5 planes: 7.900000%\n",
    "\n",
    "Epoch: 41, cost = 0.007111\n",
    "Epoch: 42, cost = 0.006986\n",
    "Epoch: 43, cost = 0.006847\n",
    "Epoch: 44, cost = 0.006686\n",
    "Epoch: 45, cost = 0.006337\n",
    "Epoch: 46, cost = 0.005834\n",
    "Epoch: 47, cost = 0.005424\n",
    "Epoch: 48, cost = 0.005140\n",
    "Epoch: 49, cost = 0.004963\n",
    "Epoch: 50, cost = 0.004882\n",
    "\n",
    "Reconstruted events: 53.90%\n",
    "\t within 1 planes: 16.500000%\n",
    "\t within 2 planes: 10.800000%\n",
    "\t within 3 planes: 8.700000%\n",
    "\t within 4 planes: 10.100000%\n",
    "\t within 5 planes: 7.800000%\n",
    "\n",
    "Epoch: 51, cost = 0.004855\n",
    "Epoch: 52, cost = 0.004847\n",
    "Epoch: 53, cost = 0.004840\n",
    "Epoch: 54, cost = 0.004834\n",
    "Epoch: 55, cost = 0.004812\n",
    "Epoch: 56, cost = 0.004721\n",
    "Epoch: 57, cost = 0.004658\n",
    "Epoch: 58, cost = 0.004643\n",
    "Epoch: 59, cost = 0.004640\n",
    "Epoch: 60, cost = 0.004626\n",
    "\n",
    "Reconstruted events: 57.10%\n",
    "\t within 1 planes: 16.900000%\n",
    "\t within 2 planes: 11.600000%\n",
    "\t within 3 planes: 10.400000%\n",
    "\t within 4 planes: 10.900000%\n",
    "\t within 5 planes: 7.300000%\n",
    "\n",
    "Epoch: 61, cost = 0.004618\n",
    "Epoch: 62, cost = 0.004616\n",
    "Epoch: 63, cost = 0.004615\n",
    "Epoch: 64, cost = 0.004620\n",
    "Epoch: 65, cost = 0.004625\n",
    "Epoch: 66, cost = 0.004621\n",
    "Epoch: 67, cost = 0.004612\n",
    "Epoch: 68, cost = 0.004596\n",
    "Epoch: 69, cost = 0.004571\n",
    "Epoch: 70, cost = 0.004535\n",
    "\n",
    "Reconstruted events: 60.70%\n",
    "\t within 1 planes: 18.500000%\n",
    "\t within 2 planes: 11.700000%\n",
    "\t within 3 planes: 12.600000%\n",
    "\t within 4 planes: 9.600000%\n",
    "\t within 5 planes: 8.300000%\n",
    "\n",
    "Epoch: 71, cost = 0.004491\n",
    "Epoch: 72, cost = 0.004449\n",
    "Epoch: 73, cost = 0.004419\n",
    "Epoch: 74, cost = 0.004400\n",
    "Epoch: 75, cost = 0.004386\n",
    "Epoch: 76, cost = 0.004374\n",
    "Epoch: 77, cost = 0.004365\n",
    "Epoch: 78, cost = 0.004369\n",
    "Epoch: 79, cost = 0.004384\n",
    "Epoch: 80, cost = 0.004401\n",
    "\n",
    "Reconstruted events: 63.90%\n",
    "\t within 1 planes: 20.600000%\n",
    "\t within 2 planes: 11.900000%\n",
    "\t within 3 planes: 14.200000%\n",
    "\t within 4 planes: 10.400000%\n",
    "\t within 5 planes: 6.800000%\n",
    "\n",
    "Epoch: 81, cost = 0.004417\n",
    "Epoch: 82, cost = 0.004428\n",
    "Epoch: 83, cost = 0.004425\n",
    "Epoch: 84, cost = 0.004401\n",
    "Epoch: 85, cost = 0.004351\n",
    "Epoch: 86, cost = 0.004276\n",
    "Epoch: 87, cost = 0.004179\n",
    "Epoch: 88, cost = 0.004061\n",
    "Epoch: 89, cost = 0.003924\n",
    "Epoch: 90, cost = 0.003767\n",
    "\n",
    "Reconstruted events: 67.60%\n",
    "\t within 1 planes: 21.200000%\n",
    "\t within 2 planes: 15.100000%\n",
    "\t within 3 planes: 14.000000%\n",
    "\t within 4 planes: 9.600000%\n",
    "\t within 5 planes: 7.700000%\n",
    "\n",
    "Epoch: 91, cost = 0.003588\n",
    "Epoch: 92, cost = 0.003384\n",
    "Epoch: 93, cost = 0.003162\n",
    "Epoch: 94, cost = 0.002931\n",
    "Epoch: 95, cost = 0.002711\n",
    "Epoch: 96, cost = 0.002505\n",
    "Epoch: 97, cost = 0.002294\n",
    "Epoch: 98, cost = 0.002073\n",
    "Epoch: 99, cost = 0.001863\n",
    "Epoch: 100, cost = 0.001680\n",
    "\n",
    "Reconstruted events: 73.60%\n",
    "\t within 1 planes: 25.500000%\n",
    "\t within 2 planes: 16.700000%\n",
    "\t within 3 planes: 14.600000%\n",
    "\t within 4 planes: 10.500000%\n",
    "\t within 5 planes: 6.300000%\n",
    "\n",
    "Epoch: 101, cost = 0.001528\n",
    "Epoch: 102, cost = 0.001398\n",
    "Epoch: 103, cost = 0.001283\n",
    "Epoch: 104, cost = 0.001181\n",
    "Epoch: 105, cost = 0.001089\n",
    "Epoch: 106, cost = 0.001001\n",
    "Epoch: 107, cost = 0.000915\n",
    "Epoch: 108, cost = 0.000835\n",
    "Epoch: 109, cost = 0.000765\n",
    "Epoch: 110, cost = 0.000706\n",
    "\n",
    "Reconstruted events: 78.00%\n",
    "\t within 1 planes: 29.500000%\n",
    "\t within 2 planes: 19.600000%\n",
    "\t within 3 planes: 14.300000%\n",
    "\t within 4 planes: 8.600000%\n",
    "\t within 5 planes: 6.000000%\n",
    "\n",
    "Epoch: 111, cost = 0.000658\n",
    "Epoch: 112, cost = 0.000620\n",
    "Epoch: 113, cost = 0.000591\n",
    "Epoch: 114, cost = 0.000567\n",
    "Epoch: 115, cost = 0.000548\n",
    "Epoch: 116, cost = 0.000534\n",
    "Epoch: 117, cost = 0.000525\n",
    "Epoch: 118, cost = 0.000523\n",
    "Epoch: 119, cost = 0.000526\n",
    "Epoch: 120, cost = 0.000535\n",
    "\n",
    "Reconstruted events: 80.30%\n",
    "\t within 1 planes: 33.200000%\n",
    "\t within 2 planes: 20.400000%\n",
    "\t within 3 planes: 12.400000%\n",
    "\t within 4 planes: 9.100000%\n",
    "\t within 5 planes: 5.200000%\n",
    "\n",
    "Epoch: 121, cost = 0.000545\n",
    "Epoch: 122, cost = 0.000555\n",
    "Epoch: 123, cost = 0.000558\n",
    "Epoch: 124, cost = 0.000553\n",
    "Epoch: 125, cost = 0.000538\n",
    "Epoch: 126, cost = 0.000515\n",
    "Epoch: 127, cost = 0.000486\n",
    "Epoch: 128, cost = 0.000459\n",
    "Epoch: 129, cost = 0.000439\n",
    "Epoch: 130, cost = 0.000426\n",
    "\n",
    "Reconstruted events: 82.10%\n",
    "\t within 1 planes: 35.600000%\n",
    "\t within 2 planes: 21.300000%\n",
    "\t within 3 planes: 12.000000%\n",
    "\t within 4 planes: 7.900000%\n",
    "\t within 5 planes: 5.300000%\n",
    "\n",
    "Epoch: 131, cost = 0.000416\n",
    "Epoch: 132, cost = 0.000406\n",
    "Epoch: 133, cost = 0.000395\n",
    "Epoch: 134, cost = 0.000383\n",
    "Epoch: 135, cost = 0.000368\n",
    "Epoch: 136, cost = 0.000351\n",
    "Epoch: 137, cost = 0.000335\n",
    "Epoch: 138, cost = 0.000316\n",
    "Epoch: 139, cost = 0.000300\n",
    "Epoch: 140, cost = 0.000289\n",
    "\n",
    "Reconstruted events: 82.90%\n",
    "\t within 1 planes: 35.900000%\n",
    "\t within 2 planes: 22.600000%\n",
    "\t within 3 planes: 11.500000%\n",
    "\t within 4 planes: 8.000000%\n",
    "\t within 5 planes: 4.900000%\n",
    "\n",
    "Epoch: 141, cost = 0.000285\n",
    "Epoch: 142, cost = 0.000283\n",
    "Epoch: 143, cost = 0.000282\n",
    "Epoch: 144, cost = 0.000283\n",
    "Epoch: 145, cost = 0.000286\n",
    "Epoch: 146, cost = 0.000290\n",
    "Epoch: 147, cost = 0.000295\n",
    "Epoch: 148, cost = 0.000297\n",
    "Epoch: 149, cost = 0.000295\n",
    "Epoch: 150, cost = 0.000289\n",
    "\n",
    "Reconstruted events: 84.50%\n",
    "\t within 1 planes: 38.800000%\n",
    "\t within 2 planes: 21.100000%\n",
    "\t within 3 planes: 11.800000%\n",
    "\t within 4 planes: 8.600000%\n",
    "\t within 5 planes: 4.200000%\n",
    "\n",
    "Epoch: 151, cost = 0.000280\n",
    "Epoch: 152, cost = 0.000269\n",
    "Epoch: 153, cost = 0.000255\n",
    "Epoch: 154, cost = 0.000239\n",
    "Epoch: 155, cost = 0.000221\n",
    "Epoch: 156, cost = 0.000205\n",
    "Epoch: 157, cost = 0.000195\n",
    "Epoch: 158, cost = 0.000187\n",
    "Epoch: 159, cost = 0.000180\n",
    "Epoch: 160, cost = 0.000174\n",
    "\n",
    "Reconstruted events: 84.60%\n",
    "\t within 1 planes: 38.500000%\n",
    "\t within 2 planes: 22.300000%\n",
    "\t within 3 planes: 12.500000%\n",
    "\t within 4 planes: 7.400000%\n",
    "\t within 5 planes: 3.900000%\n",
    "\n",
    "Epoch: 161, cost = 0.000172\n",
    "Epoch: 162, cost = 0.000175\n",
    "Epoch: 163, cost = 0.000176\n",
    "Epoch: 164, cost = 0.000171\n",
    "Epoch: 165, cost = 0.000165\n",
    "Epoch: 166, cost = 0.000157\n",
    "Epoch: 167, cost = 0.000147\n",
    "Epoch: 168, cost = 0.000137\n",
    "Epoch: 169, cost = 0.000126\n",
    "Epoch: 170, cost = 0.000116\n",
    "\n",
    "Reconstruted events: 85.50%\n",
    "\t within 1 planes: 38.900000%\n",
    "\t within 2 planes: 22.800000%\n",
    "\t within 3 planes: 13.000000%\n",
    "\t within 4 planes: 7.100000%\n",
    "\t within 5 planes: 3.700000%\n",
    "\n",
    "Epoch: 171, cost = 0.000106\n",
    "Epoch: 172, cost = 0.000096\n",
    "Epoch: 173, cost = 0.000086\n",
    "Epoch: 174, cost = 0.000073\n",
    "Epoch: 175, cost = 0.000058\n",
    "Epoch: 176, cost = 0.000043\n",
    "Epoch: 177, cost = 0.000033\n",
    "Epoch: 178, cost = 0.000028\n",
    "Epoch: 179, cost = 0.000026\n",
    "Epoch: 180, cost = 0.000026\n",
    "\n",
    "Reconstruted events: 86.40%\n",
    "\t within 1 planes: 38.900000%\n",
    "\t within 2 planes: 23.900000%\n",
    "\t within 3 planes: 13.100000%\n",
    "\t within 4 planes: 6.700000%\n",
    "\t within 5 planes: 3.800000%\n",
    "\n",
    "Epoch: 181, cost = 0.000027\n",
    "Epoch: 182, cost = 0.000028\n",
    "Epoch: 183, cost = 0.000030\n",
    "Epoch: 184, cost = 0.000032\n",
    "Epoch: 185, cost = 0.000034\n",
    "Epoch: 186, cost = 0.000038\n",
    "Epoch: 187, cost = 0.000042\n",
    "Epoch: 188, cost = 0.000047\n",
    "Epoch: 189, cost = 0.000053\n",
    "Epoch: 190, cost = 0.000060\n",
    "\n",
    "Reconstruted events: 86.90%\n",
    "\t within 1 planes: 40.300000%\n",
    "\t within 2 planes: 23.200000%\n",
    "\t within 3 planes: 13.300000%\n",
    "\t within 4 planes: 7.200000%\n",
    "\t within 5 planes: 2.900000%\n",
    "\n",
    "Epoch: 191, cost = 0.000066\n",
    "Epoch: 192, cost = 0.000072\n",
    "Epoch: 193, cost = 0.000078\n",
    "Epoch: 194, cost = 0.000083\n",
    "Epoch: 195, cost = 0.000089\n",
    "Epoch: 196, cost = 0.000095\n",
    "Epoch: 197, cost = 0.000101\n",
    "Epoch: 198, cost = 0.000106\n",
    "Epoch: 199, cost = 0.000112\n",
    "Epoch: 200, cost = 0.000119\n",
    "\n",
    "Reconstruted events: 87.40%\n",
    "\t within 1 planes: 40.500000%\n",
    "\t within 2 planes: 24.900000%\n",
    "\t within 3 planes: 13.200000%\n",
    "\t within 4 planes: 6.000000%\n",
    "\t within 5 planes: 2.800000%\n",
    "\n",
    "Epoch: 201, cost = 0.000127\n",
    "Epoch: 202, cost = 0.000136\n",
    "Epoch: 203, cost = 0.000145\n",
    "Epoch: 204, cost = 0.000154\n",
    "Epoch: 205, cost = 0.000160\n",
    "Epoch: 206, cost = 0.000163\n",
    "Epoch: 207, cost = 0.000165\n",
    "Epoch: 208, cost = 0.000165\n",
    "Epoch: 209, cost = 0.000165\n",
    "Epoch: 210, cost = 0.000163\n",
    "\n",
    "Reconstruted events: 88.10%\n",
    "\t within 1 planes: 41.100000%\n",
    "\t within 2 planes: 25.300000%\n",
    "\t within 3 planes: 12.300000%\n",
    "\t within 4 planes: 5.800000%\n",
    "\t within 5 planes: 3.600000%\n",
    "\n",
    "Epoch: 211, cost = 0.000160\n",
    "Epoch: 212, cost = 0.000156\n",
    "Epoch: 213, cost = 0.000152\n",
    "Epoch: 214, cost = 0.000149\n",
    "Epoch: 215, cost = 0.000146\n",
    "Epoch: 216, cost = 0.000146\n",
    "Epoch: 217, cost = 0.000150\n",
    "Epoch: 218, cost = 0.000156\n",
    "Epoch: 219, cost = 0.000163\n",
    "Epoch: 220, cost = 0.000170\n",
    "\n",
    "Reconstruted events: 89.00%\n",
    "\t within 1 planes: 41.400000%\n",
    "\t within 2 planes: 26.200000%\n",
    "\t within 3 planes: 12.000000%\n",
    "\t within 4 planes: 5.400000%\n",
    "\t within 5 planes: 4.000000%\n",
    "\n",
    "Epoch: 221, cost = 0.000176\n",
    "Epoch: 222, cost = 0.000181\n",
    "Epoch: 223, cost = 0.000186\n",
    "Epoch: 224, cost = 0.000190\n",
    "Epoch: 225, cost = 0.000193\n",
    "Epoch: 226, cost = 0.000195\n",
    "Epoch: 227, cost = 0.000196\n",
    "Epoch: 228, cost = 0.000198\n",
    "Epoch: 229, cost = 0.000200\n",
    "Epoch: 230, cost = 0.000203\n",
    "\n",
    "Reconstruted events: 90.20%\n",
    "\t within 1 planes: 42.600000%\n",
    "\t within 2 planes: 25.500000%\n",
    "\t within 3 planes: 12.300000%\n",
    "\t within 4 planes: 5.300000%\n",
    "\t within 5 planes: 4.500000%\n",
    "\n",
    "Epoch: 231, cost = 0.000205\n",
    "Epoch: 232, cost = 0.000208\n",
    "Epoch: 233, cost = 0.000210\n",
    "Epoch: 234, cost = 0.000212\n",
    "Epoch: 235, cost = 0.000214\n",
    "Epoch: 236, cost = 0.000214\n",
    "Epoch: 237, cost = 0.000214\n",
    "Epoch: 238, cost = 0.000213\n",
    "Epoch: 239, cost = 0.000211\n",
    "Epoch: 240, cost = 0.000208\n",
    "\n",
    "Reconstruted events: 91.00%\n",
    "\t within 1 planes: 42.700000%\n",
    "\t within 2 planes: 25.300000%\n",
    "\t within 3 planes: 12.400000%\n",
    "\t within 4 planes: 6.200000%\n",
    "\t within 5 planes: 4.400000%\n",
    "\n",
    "Epoch: 241, cost = 0.000206\n",
    "Epoch: 242, cost = 0.000203\n",
    "Epoch: 243, cost = 0.000200\n",
    "Epoch: 244, cost = 0.000198\n",
    "Epoch: 245, cost = 0.000195\n",
    "Epoch: 246, cost = 0.000192\n",
    "Epoch: 247, cost = 0.000189\n",
    "Epoch: 248, cost = 0.000184\n",
    "Epoch: 249, cost = 0.000178\n",
    "Epoch: 250, cost = 0.000170\n",
    "\n",
    "Reconstruted events: 91.20%\n",
    "\t within 1 planes: 43.400000%\n",
    "\t within 2 planes: 25.100000%\n",
    "\t within 3 planes: 12.500000%\n",
    "\t within 4 planes: 6.800000%\n",
    "\t within 5 planes: 3.400000%\n",
    "\n",
    "Epoch: 251, cost = 0.000162\n",
    "Epoch: 252, cost = 0.000156\n",
    "Epoch: 253, cost = 0.000156\n",
    "Epoch: 254, cost = 0.000162\n",
    "Epoch: 255, cost = 0.000171\n",
    "Epoch: 256, cost = 0.000178\n",
    "Epoch: 257, cost = 0.000182\n",
    "Epoch: 258, cost = 0.000184\n",
    "Epoch: 259, cost = 0.000184\n",
    "Epoch: 260, cost = 0.000183\n",
    "\n",
    "Reconstruted events: 91.80%\n",
    "\t within 1 planes: 43.700000%\n",
    "\t within 2 planes: 24.900000%\n",
    "\t within 3 planes: 13.200000%\n",
    "\t within 4 planes: 6.700000%\n",
    "\t within 5 planes: 3.300000%\n",
    "\n",
    "Epoch: 261, cost = 0.000181\n",
    "Epoch: 262, cost = 0.000179\n",
    "Epoch: 263, cost = 0.000176\n",
    "Epoch: 264, cost = 0.000173\n",
    "Epoch: 265, cost = 0.000169\n",
    "Epoch: 266, cost = 0.000166\n",
    "Epoch: 267, cost = 0.000161\n",
    "Epoch: 268, cost = 0.000156\n",
    "Epoch: 269, cost = 0.000152\n",
    "Epoch: 270, cost = 0.000148\n",
    "\n",
    "Reconstruted events: 92.00%\n",
    "\t within 1 planes: 44.500000%\n",
    "\t within 2 planes: 24.300000%\n",
    "\t within 3 planes: 13.200000%\n",
    "\t within 4 planes: 6.800000%\n",
    "\t within 5 planes: 3.200000%\n",
    "\n",
    "Epoch: 271, cost = 0.000144\n",
    "Epoch: 272, cost = 0.000141\n",
    "Epoch: 273, cost = 0.000138\n",
    "Epoch: 274, cost = 0.000135\n",
    "Epoch: 275, cost = 0.000132\n",
    "Epoch: 276, cost = 0.000129\n",
    "Epoch: 277, cost = 0.000126\n",
    "Epoch: 278, cost = 0.000124\n",
    "Epoch: 279, cost = 0.000121\n",
    "Epoch: 280, cost = 0.000118\n",
    "\n",
    "Reconstruted events: 93.00%\n",
    "\t within 1 planes: 45.200000%\n",
    "\t within 2 planes: 25.000000%\n",
    "\t within 3 planes: 12.700000%\n",
    "\t within 4 planes: 7.200000%\n",
    "\t within 5 planes: 2.900000%\n",
    "\n",
    "Epoch: 281, cost = 0.000116\n",
    "Epoch: 282, cost = 0.000114\n",
    "Epoch: 283, cost = 0.000112\n",
    "Epoch: 284, cost = 0.000109\n",
    "Epoch: 285, cost = 0.000107\n",
    "Epoch: 286, cost = 0.000104\n",
    "Epoch: 287, cost = 0.000102\n",
    "Epoch: 288, cost = 0.000100\n",
    "Epoch: 289, cost = 0.000098\n",
    "Epoch: 290, cost = 0.000097\n",
    "\n",
    "Reconstruted events: 93.50%\n",
    "\t within 1 planes: 44.900000%\n",
    "\t within 2 planes: 25.700000%\n",
    "\t within 3 planes: 12.400000%\n",
    "\t within 4 planes: 7.100000%\n",
    "\t within 5 planes: 3.400000%\n",
    "\n",
    "Epoch: 291, cost = 0.000095\n",
    "Epoch: 292, cost = 0.000094\n",
    "Epoch: 293, cost = 0.000093\n",
    "Epoch: 294, cost = 0.000091\n",
    "Epoch: 295, cost = 0.000090\n",
    "Epoch: 296, cost = 0.000089\n",
    "Epoch: 297, cost = 0.000088\n",
    "Epoch: 298, cost = 0.000087\n",
    "Epoch: 299, cost = 0.000087\n",
    "Epoch: 300, cost = 0.000086\n",
    "\n",
    "Reconstruted events: 94.00%\n",
    "\t within 1 planes: 44.100000%\n",
    "\t within 2 planes: 26.700000%\n",
    "\t within 3 planes: 12.500000%\n",
    "\t within 4 planes: 6.800000%\n",
    "\t within 5 planes: 3.900000%\n",
    "\n",
    "Epoch: 301, cost = 0.000086\n",
    "Epoch: 302, cost = 0.000086\n",
    "Epoch: 303, cost = 0.000086\n",
    "Epoch: 304, cost = 0.000086\n",
    "Epoch: 305, cost = 0.000086\n",
    "Epoch: 306, cost = 0.000087\n",
    "Epoch: 307, cost = 0.000087\n",
    "Epoch: 308, cost = 0.000087\n",
    "Epoch: 309, cost = 0.000087\n",
    "Epoch: 310, cost = 0.000086\n",
    "\n",
    "Reconstruted events: 94.30%\n",
    "\t within 1 planes: 45.100000%\n",
    "\t within 2 planes: 26.300000%\n",
    "\t within 3 planes: 12.100000%\n",
    "\t within 4 planes: 7.100000%\n",
    "\t within 5 planes: 3.700000%\n",
    "\n",
    "Epoch: 311, cost = 0.000085\n",
    "Epoch: 312, cost = 0.000084\n",
    "Epoch: 313, cost = 0.000082\n",
    "Epoch: 314, cost = 0.000081\n",
    "Epoch: 315, cost = 0.000081\n",
    "Epoch: 316, cost = 0.000081\n",
    "Epoch: 317, cost = 0.000083\n",
    "Epoch: 318, cost = 0.000085\n",
    "Epoch: 319, cost = 0.000087\n",
    "Epoch: 320, cost = 0.000088\n",
    "\n",
    "Reconstruted events: 94.40%\n",
    "\t within 1 planes: 45.100000%\n",
    "\t within 2 planes: 26.700000%\n",
    "\t within 3 planes: 11.300000%\n",
    "\t within 4 planes: 7.600000%\n",
    "\t within 5 planes: 3.700000%\n",
    "\n",
    "Epoch: 321, cost = 0.000090\n",
    "Epoch: 322, cost = 0.000091\n",
    "Epoch: 323, cost = 0.000093\n",
    "Epoch: 324, cost = 0.000094\n",
    "Epoch: 325, cost = 0.000095\n",
    "Epoch: 326, cost = 0.000096\n",
    "Epoch: 327, cost = 0.000097\n",
    "Epoch: 328, cost = 0.000098\n",
    "Epoch: 329, cost = 0.000098\n",
    "Epoch: 330, cost = 0.000098\n",
    "\n",
    "Reconstruted events: 94.30%\n",
    "\t within 1 planes: 45.400000%\n",
    "\t within 2 planes: 25.600000%\n",
    "\t within 3 planes: 12.400000%\n",
    "\t within 4 planes: 7.300000%\n",
    "\t within 5 planes: 3.600000%\n",
    "\n",
    "Epoch: 331, cost = 0.000098\n",
    "Epoch: 332, cost = 0.000097\n",
    "Epoch: 333, cost = 0.000097\n",
    "Epoch: 334, cost = 0.000096\n",
    "Epoch: 335, cost = 0.000095\n",
    "Epoch: 336, cost = 0.000093\n",
    "Epoch: 337, cost = 0.000092\n",
    "Epoch: 338, cost = 0.000090\n",
    "Epoch: 339, cost = 0.000088\n",
    "Epoch: 340, cost = 0.000086\n",
    "\n",
    "Reconstruted events: 94.60%\n",
    "\t within 1 planes: 45.500000%\n",
    "\t within 2 planes: 26.200000%\n",
    "\t within 3 planes: 11.700000%\n",
    "\t within 4 planes: 7.200000%\n",
    "\t within 5 planes: 4.000000%\n",
    "\n",
    "Epoch: 341, cost = 0.000084\n",
    "Epoch: 342, cost = 0.000081\n",
    "Epoch: 343, cost = 0.000079\n",
    "Epoch: 344, cost = 0.000077\n",
    "Epoch: 345, cost = 0.000074\n",
    "Epoch: 346, cost = 0.000072\n",
    "Epoch: 347, cost = 0.000070\n",
    "Epoch: 348, cost = 0.000068\n",
    "Epoch: 349, cost = 0.000066\n",
    "Epoch: 350, cost = 0.000065\n",
    "\n",
    "Reconstruted events: 95.30%\n",
    "\t within 1 planes: 47.300000%\n",
    "\t within 2 planes: 24.200000%\n",
    "\t within 3 planes: 12.200000%\n",
    "\t within 4 planes: 7.100000%\n",
    "\t within 5 planes: 4.500000%\n",
    "\n",
    "Epoch: 351, cost = 0.000064\n",
    "Epoch: 352, cost = 0.000062\n",
    "Epoch: 353, cost = 0.000061\n",
    "Epoch: 354, cost = 0.000060\n",
    "Epoch: 355, cost = 0.000059\n",
    "Epoch: 356, cost = 0.000057\n",
    "Epoch: 357, cost = 0.000056\n",
    "Epoch: 358, cost = 0.000054\n",
    "Epoch: 359, cost = 0.000053\n",
    "Epoch: 360, cost = 0.000051\n",
    "\n",
    "Reconstruted events: 95.50%\n",
    "\t within 1 planes: 47.500000%\n",
    "\t within 2 planes: 24.300000%\n",
    "\t within 3 planes: 12.000000%\n",
    "\t within 4 planes: 7.800000%\n",
    "\t within 5 planes: 3.900000%\n",
    "\n",
    "Epoch: 361, cost = 0.000050\n",
    "Epoch: 362, cost = 0.000048\n",
    "Epoch: 363, cost = 0.000046\n",
    "Epoch: 364, cost = 0.000044\n",
    "Epoch: 365, cost = 0.000043\n",
    "Epoch: 366, cost = 0.000041\n",
    "Epoch: 367, cost = 0.000039\n",
    "Epoch: 368, cost = 0.000037\n",
    "Epoch: 369, cost = 0.000035\n",
    "Epoch: 370, cost = 0.000034\n",
    "\n",
    "Reconstruted events: 95.40%\n",
    "\t within 1 planes: 47.600000%\n",
    "\t within 2 planes: 23.800000%\n",
    "\t within 3 planes: 12.800000%\n",
    "\t within 4 planes: 7.600000%\n",
    "\t within 5 planes: 3.600000%\n",
    "\n",
    "Epoch: 371, cost = 0.000032\n",
    "Epoch: 372, cost = 0.000031\n",
    "Epoch: 373, cost = 0.000029\n",
    "Epoch: 374, cost = 0.000028\n",
    "Epoch: 375, cost = 0.000027\n",
    "Epoch: 376, cost = 0.000025\n",
    "Epoch: 377, cost = 0.000024\n",
    "Epoch: 378, cost = 0.000023\n",
    "Epoch: 379, cost = 0.000022\n",
    "Epoch: 380, cost = 0.000021\n",
    "\n",
    "Reconstruted events: 95.70%\n",
    "\t within 1 planes: 47.500000%\n",
    "\t within 2 planes: 24.400000%\n",
    "\t within 3 planes: 12.500000%\n",
    "\t within 4 planes: 7.300000%\n",
    "\t within 5 planes: 4.000000%\n",
    "\n",
    "Epoch: 381, cost = 0.000020\n",
    "Epoch: 382, cost = 0.000019\n",
    "Epoch: 383, cost = 0.000018\n",
    "Epoch: 384, cost = 0.000017\n",
    "Epoch: 385, cost = 0.000016\n",
    "Epoch: 386, cost = 0.000015\n",
    "Epoch: 387, cost = 0.000014\n",
    "Epoch: 388, cost = 0.000013\n",
    "Epoch: 389, cost = 0.000013\n",
    "Epoch: 390, cost = 0.000012\n",
    "\n",
    "Reconstruted events: 95.60%\n",
    "\t within 1 planes: 48.700000%\n",
    "\t within 2 planes: 23.100000%\n",
    "\t within 3 planes: 12.800000%\n",
    "\t within 4 planes: 7.400000%\n",
    "\t within 5 planes: 3.600000%\n",
    "\n",
    "Epoch: 391, cost = 0.000011\n",
    "Epoch: 392, cost = 0.000010\n",
    "Epoch: 393, cost = 0.000009\n",
    "Epoch: 394, cost = 0.000008\n",
    "Epoch: 395, cost = 0.000007\n",
    "Epoch: 396, cost = 0.000006\n",
    "Epoch: 397, cost = 0.000006\n",
    "Epoch: 398, cost = 0.000005\n",
    "Epoch: 399, cost = 0.000004\n",
    "Epoch: 400, cost = 0.000004\n",
    "\n",
    "Reconstruted events: 95.50%\n",
    "\t within 1 planes: 49.200000%\n",
    "\t within 2 planes: 23.000000%\n",
    "\t within 3 planes: 12.600000%\n",
    "\t within 4 planes: 7.300000%\n",
    "\t within 5 planes: 3.400000%\n"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 2",
   "language": "python",
   "name": "python2"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 2
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython2",
   "version": "2.7.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 0
}