TDEs_v3.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Write a metric for the detection of TDEs"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib\n",
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline\n",
    "\n",
    "import lsst.sims.maf.metrics as metrics\n",
    "import lsst.sims.maf.slicers as slicers\n",
    "import lsst.sims.maf.metricBundles as metricBundles\n",
    "import lsst.sims.maf.db as db\n",
    "import lsst.sims.maf.utils as utils\n",
    "import lsst.sims.maf.plots as plots\n",
    "\n",
    "# open database use OpsimDatabase\n",
    "opsdb = db.OpsimDatabase('baseline2018a.db')\n",
    "# output\n",
    "outDir = 'outdir'\n",
    "resultsDb = db.ResultsDb(outDir=outDir)\n",
    "plt.rcParams['figure.figsize'] = (12.0, 8.0) # set default size of plots\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## write a class to get required data for test function run() inside metric"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "class getDataMetric(metrics.BaseMetric):\n",
    "    \"\"\"\n",
    "    returns required data \n",
    "    \"\"\"\n",
    "    \n",
    "    def __init__(self, colname=['airmass', 'expMJD'], **kwargs):\n",
    "        self.colname = colname\n",
    "        \n",
    "        super().__init__(col=colname, metricDtype='object', **kwargs)\n",
    "        \n",
    "        \n",
    "    def run(self, dataSlice, slicePoint=None):\n",
    "        \n",
    "        # return dataSlice directly\n",
    "        \n",
    "        result = dataSlice\n",
    "        \n",
    "        return result"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 115,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Querying database SummaryAllProps with constraint night<730 for columns ['fiveSigmaDepth', 'fieldDec', 'fieldRA', 'night', 'filter', 'observationStartMJD']\n",
      "Found 452269 visits\n",
      "Running:  ['Sky']\n",
      "Completed metric generation.\n",
      "Running reduce methods.\n",
      "Running summary statistics.\n",
      "Completed.\n",
      "Plotting figures with \"night<730\" constraint now.\n",
      "Plotting complete.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/docmaf/repos/sims_maf/python/lsst/sims/maf/plots/plotHandler.py:517: UserWarning: Cannot plot object metric values with this plotter.\n",
      "  warnings.warn('Cannot plot object metric values with this plotter.')\n"
     ]
    }
   ],
   "source": [
    "metric = getDataMetric(colname=['night', 'observationStartMJD', \n",
    "                                'fiveSigmaDepth', 'filter'])\n",
    "#slicer = slicers.UniSlicer()\n",
    "#slicer = slicers.HealpixSlicer(nside=16)\n",
    "ra = np.array([34.39339593])\n",
    "dec = np.array([-5.09032894])\n",
    "slicer = slicers.UserPointsSlicer(ra, dec)\n",
    "\n",
    "sqlconstraint = 'night<730' # first two year\n",
    "#sqlconstraint = 'observationStartMJD<60553'\n",
    "# bundle\n",
    "Sky = metricBundles.MetricBundle(metric, slicer, sqlconstraint)\n",
    "\n",
    "# group bundle\n",
    "bundleDict = {'Sky':Sky}\n",
    "group = metricBundles.MetricBundleGroup(bundleDict, opsdb, outDir = outDir, resultsDb=resultsDb)\n",
    "\n",
    "# run\n",
    "group.runAll()\n",
    "group.plotAll(closefigs = False)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 116,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "dtype((numpy.record, [('fiveSigmaDepth', '<f8'), ('fieldDec', '<f8'), ('fieldRA', '<f8'), ('night', '<i8'), ('filter', '<U256'), ('observationStartMJD', '<f8')]))"
      ]
     },
     "execution_count": 116,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dataSlice = Sky.metricValues.data[0]\n",
    "dataSlice.dtype"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 117,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(4587,)"
      ]
     },
     "execution_count": 117,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dataSlice['observationStartMJD'].shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 89,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array(['g', 'i', 'r', 'u', 'y', 'z'],\n",
       "      dtype='<U256')"
      ]
     },
     "execution_count": 89,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.unique(dataSlice['filter'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fcc1c6d8a90>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "\n",
    "def plotlc(dataSlice, time_key='time', mag_key='mag', filter_key='filter'):\n",
    "    colors = {'u':'k', 'g':'b', 'r':'r', 'i':'m', 'z':'orange', 'y':'c'}\n",
    "\n",
    "    for f in np.unique(dataSlice[filter_key]):\n",
    "        fmatch = dataSlice[filter_key]==f\n",
    "        time = dataSlice[time_key][fmatch]\n",
    "        mag = dataSlice[mag_key][fmatch]\n",
    "\n",
    "        plt.scatter(time, mag, color=colors[f], marker='*', label=f)\n",
    "\n",
    "    plt.legend()\n",
    "    plt.xlabel('days')\n",
    "    plt.ylabel('mag')\n",
    "    plt.ylim(plt.ylim()[::-1])\n",
    "\n",
    "plotlc(dataSlice, time_key='observationStartMJD', \n",
    "       mag_key='fiveSigmaDepth', filter_key='filter')\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Test functions inside class"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### make light curve from txt file"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "metadata": {},
   "outputs": [],
   "source": [
    "def read_lightCurve(asciifile='TDEfaintfast_z0.1.dat'):\n",
    "        \n",
    "    lcv_template = np.genfromtxt(asciifile, dtype=[('ph', 'f8'), ('mag', 'f8'), ('flt', 'S1')])\n",
    "    \n",
    "    return lcv_template\n",
    "\n",
    "\n",
    "def make_lightCurve(time, filters):\n",
    "    \n",
    "    #datafile\n",
    "    asciifile = 'TDEfaintfast_z0.1.dat'\n",
    "    \n",
    "    lcv_template = read_lightCurve(asciifile)\n",
    "    transDuration = lcv_template['ph'].max() - lcv_template['ph'].min()\n",
    "    \n",
    "    lcMags = np.zeros(time.size, dtype=float)\n",
    "    \n",
    "    for f in set(lcv_template['flt']):\n",
    "        fMatch_ascii = np.where(np.array(lcv_template['flt']) == f)[0]\n",
    "        \n",
    "        # Interpolate the lightcurve template to the times of the observations, in this filter.\n",
    "        lc_ascii_filter = np.interp(time, np.array(lcv_template['ph'], float)[fMatch_ascii],\n",
    "                                        np.array(lcv_template['mag'], float)[fMatch_ascii])\n",
    "        lcMags[filters == f.decode(\"utf-8\")] = lc_ascii_filter[filters == f.decode(\"utf-8\")]\n",
    "    \n",
    "    return lcMags\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 125,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fcc1df59a90>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fcc1c9b1240>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fcc1deff7f0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAATwAAADFCAYAAADJ705jAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAG6NJREFUeJzt3XtwVPX9N/D3Z5OQJRcwco1ATHzGyjVEDQHHAh2scqlYEUR+RrS2D2mgVUE6HYFqKwxof/CgdaZS6aOPrdycGmx/9Rb5UTqEWi8BEwoJqD8NECAEMrVgJIEkn+eP3Wx3k71mz+45m/N+zezk5Ltnz/nsOWff+z2X3RVVBRGRHTjMLoCIKF4YeERkGww8IrINBh4R2QYDj4hsg4FHRLbBwCMi22DgEZFtMPCIyDaS4zmzgQMHam5ubjxnSUQ2sH///nOqOijUeHENvNzcXFRWVsZzlkRkAyJyLJzxuEtLRLbBwCMi22DgEZFtxPUYXqREpFsbv86KiHrKsoHnL+w62xl6RNaRSB0T7tISUY8F65hYEQOPiGyDgUdEtpGQgWfV7jIRWVtCBh4RUU8w8IioxwKdjeVZ2ght2bLF7BKIKAyq2u1mVZYNvOLiYrNLIKJexrKBByRed5mIrM2yn7ToxHAjIqNYuodHRGQkBh4R2QYDj4hsg4FHRLbBwCMi22DgEZFtMPCIyDYYeERkGww8IrINBh4R2QYDj4hsg4FHRLYRMvBEZISI7BGRGhE5LCKPuNvXi8gRETkoIq+LyBWxL5eoZ0Sk243sJ5weXhuA5ao6GsAkAD8SkdEAdgEYq6r5AD4BsCJ2ZRL1XKL9lCDFTsjAU9XTqnrAPXwBQC2AYar6rqq2uUd7H8Dw2JVJRBS9iI7hiUgugOsBfNDlru8DeDvAY0pEpFJEKs+ePduTGomIDBF24IlIBoAyAEtV9bxX+yq4dnu3+nucqm5W1UJVLRw0aFC09RIR9VhY33gsIilwhd1WVd3p1f49ALcDuEX51cREZHEhA09cR3ZfBFCrqhu92mcA+CmAqar6dexKJIqOqvIEBQEIr4d3M4CFAP4hIlXutpUAngOQCmCXe2N6X1VLY1IlUYyICH83xUZCBp6q7gPg7+3xLePLIYo/hp598JMWRGQbDDwisg0GHhHZhuV/iDsUf2ffeDyGuuKZWgISvIfHz0gSUSQSOvCIiCKR8Lu0RJQYrHD4iT08so1ALy4e8409qxx+Yg+PeswK79iRsnp9dhTPC78TuofHd2zzWOUdmygSCd/DY7gRUbgSuodHRBQJBh4RxZxV9sQSfpfWWyIeRCeyi0Cfdonna7TXBF6wg+gMPeNZYeOlxGP29tFrAo/iz+yNlyhSPIZHRHG1a9cuJCcn4y9/+Uvc583Ao4Rn5guIIjd79my0t7dj1qxZcZ83A48S3j333IP29nbMmzfP7FIoCBGBiKC1tRUA0Nra6mmLl15zDI8H0e2n6/r+5z//6Wnjeid/elUPT1W73ShxzZgxAyKC22+/3e/97777LtLS0nza0tPTsXv37niURzFw+vRpTJ06FQ0NDTGZfq8KPLK+SDbo8vJyAMCbb77p9/5bb70VqampPm19+vTBtGnToi+UDPfQQw/5bV+2bJln+LHHHkNFRQWefPLJ2BThr1cUq9uNN96okTh16pROmTJFT58+HdHjyLoWL16sDodDFy9eHHAcAAFvPRmvpx566CEFoMuWLTNkenYXbH05nc6A7WFOu1LDyCBLB144Lw5KDJFs0OEG2fe///2YBp7R07O7d9991++62r17t546dUrvvfden/bi4uKwOzsJHXjRpj1ZT7gbdLCw89fTCjZ+T4Mq1j1HO8vMzPRZnunp6aoa/TIPN/AseQzv888/x7333ovkZNdJZKfTieLiYnzxxRcmV0Y9ddVVV2Hbtm0+bVu3bkV2drZPW35+fsBpPPPMMz5nZkNdzrB58+YeVEqx1NLSAgBYsGABkpKS0NzcjN/97ndhHd8zQsjAE5ERIrJHRGpE5LCIPOJuXyMiB0WkSkTeFZGrjCoqOzsb/fr1Q1tbGwDX9Tr9+vXD0KFDjZoFWVR1dXXY4wZ6kXRatGhRtOWQwTpPRJWWluLLL7/Erbfeiu9973vIzc31O/7GjRsNnX84Pbw2AMtVdTSASQB+JCKjAaxX1XxVLQDwBoAnjCzszJkznnf/O+64I2anqSk+0tPT/bZnZmb6/B/JRajPPfdc0Pv57cvWc+rUKQBAc3MzMjIy8Oc//xl33303li9f7nd8o9dhyMBT1dOqesA9fAFALYBhqnrea7R0uPa3DbNz506MHDkSAPDwww9j586dRk6e4qy5udlv+4ULF3z+3759e9CNPJJdHIfDd/MO9yNozz//vN927iJH75VXXgEAvPzyywCA1NRUbN++HaNGjfI7vtG7tBGddACQC+A4gH7u/9cCOAHgEIBBAR5TAqASQGVOTk5EBzjHjh2rAPTGG2/k5QEJDmEelP7Tn/4U0YmIYOOlpaXpp59+6hn3iiuuUACalZUVtNY+ffr4nV5qaqpxC8RmQp2IDHd9BwKjz9ICyACwH8Bdfu5bAeDJUNOI9LKUcJ/8xx9/rP3799fq6uqIpk/xs3//fr/r0/ssbXV1taanpwfd+NevX+8z3enTpwccNysrSwsLCyN6MTU1NWl2dnbA8ceOHatLlizR7du3a319fcyXW2/ReZa+883E6XT6nKUPdslKOAwNPAApAMoBPBrg/hwAh0JNJ9zAC/UO33VDvfLKKxWAXnnllWFNn+Iv0DVznRoaGjQnJ0eHDRumJ0+eVFX/28Hhw4e7TTvUtjJ//nxNS0vzaUtPT+/2Yuro6NA5c+ZoSkqKzp07Vx0OhzqdTnU4HPrd735X16xZo7fddptmZGR4pnPNNdfoAw88oC+++KJ+8skn2tHREdsFmcBKS0t91kHX62uzsrK6vWGFy7DAAyAAfg/g2S7t13oNPwTgtVDTMirw7r///pDjkXVUV1eriARdX5MmTdK+ffvq/v37PY/z3vA7h++77z6faYfqDZaUlKiI+IRUoN7Dpk2bFIBu2LBB58yZo0uWLNGqqipdsmSJzpkzxzPe5cuX9aOPPtKNGzfqnXfeqQMHDvRMc+jQoXr33Xfrc889p1VVVdrW1hbbhZtA5syZ47P8vZepqmpqaqpmZWXpL3/5S83KyoroEIKRgfdNd4EHAVS5b7MAlMF17O4ggD/DdSIjLoGXnJysjz76KAMvQQTb7fS+lZWV+Tzu1VdfVQA6bdo0VVVdvny5OhwO/eyzzzzjhNoGmpubdeTIkX7v976Q/dChQ+p0OnX69Ona3t4e0fPr6OjQmpoa/c1vfqPFxcU6YsQIzzz69++vs2bN0qefflr/9re/aWtra08XY0KL9WvVsMAz8hZu4HV+hjHQbdGiRSoiAQ+E8uSGdZSXlysAnTlzZtB1unbt2m6PLSsrUwB6yy23qKrrOFBqaqouWrTIM044L6RwwnbcuHE6ePBgbWhoMOR519XV6SuvvKKLFi3yCVyn06nf+ta39PHHH9ddu3bphQsXDJmf1QV6TRv1Wk3owHM/gaAb8QcffKATJkxg787C2traND8/X/Py8rSlpSXgOi0oKPB77KvzjG1n4Km6Pl+dkpKiJ06cUFXjAg+Avv322zFbFmfOnNGysjJdunSp3nDDDepwOBSAJiUlaVFRkS5fvlz/+Mc/6rlz52JWg5nYwwv9BEIupFgvRIrOSy+9pAB0x44dqhp4fV28eNHv4998800FoN/+9rc9bV988YUmJSXpI488oqrh9RxC7TH462nEepv617/+pe+8846uXLlSJ0+e7HMpzJgxY7S0tFS3bdvmCfZEN3v2bL/LdO7cuZ5xolnmCR144W6cDDzram5u1quuukonTpzo6b1Fur7eeeedboGnqvrAAw9o37599cyZM2G9MYaaNwBtaWkJa9xYuXjxou7du1fXrl2r06dP9/mQfV5ent5///3629/+Vo8ePZpwZ4K//PJLnTZtWo/XTzh6deB1LoRYHxegnluzZo0C0IqKCk9bpBt15/G/roF35MgRFRFdsWKFbt++PerAmz9/vs/0rfBGevnyZa2srNRnnnlG77rrLp8zwUOGDNF58+bpr371Kz1w4IClzwSfPHlS8/PzNTk52VO/93WOnRh4Ubxzk7kaGho0IyOj22UHRgWequr8+fM1MzNTm5qaAk7zO9/5To/mb4XA66qjo0Nra2v1hRde0Pvuu0+vvvpqT039+vXTmTNn6rp163Tfvn0+vVUz1dTUaE5OjmZkZGh5eXnQcRl4AW7Jyck+n6jobB8/frzpGyW5lJaWanJysh49etSnPdKr6TsDz1+vvaqqSgHoqlWrog6xRAg8f44dO6ZbtmzRkpISHTVqlKdGp9OpU6dO1Z/97GdaXl6u58+fj3ttFRUVmpWVpUOGDNEDBw6EHN/WgRdsNyU5OTng4yZOnKiDBw9WAFpSUhLWvMhYNTU1mpSUpD/+8Y/93h/JBt15DC/QeLfffrumpKT4jOO96xdqvokeeF01Njbqzp07ddmyZVpYWOhzJriwsFCXLVumr7/+up49ezamdZSVlWlqaqp+4xvf0M8//zysx9g68AJdXwe49v8D6frRFYq/2bNna79+/bSxsdHv/eFeTR/OC+Dvf/972C+OSAIv0PiJ5vz581peXq6rVq3SKVOmaGpqque5jB49Wn/4wx/qli1b9Pjx41HPq/OjgzfddJOKiE6aNCniYI1mmSd04J06dSrixE/Ud+XeZM+ePQpAn3rqqainFc76jGSdB9qdBsL/gHqia2lp0YqKCl23bp3OmDHD50zw1VdfrQsXLtTNmzdrbW1txGeCvZfnHXfcoc3NzTF6FgHnH1bgiWvc+CgsLNTKysqwxk1KSkJHR0e39szMTJw/f75be7DvULvnnnuQnp6OjIwMpKen+70Fui8tLa3b96pRdx0dHSgqKkJjYyOOHj2Kvn37RjW9lStX4qmnnurW/vjjj2P16tUAgB/84Ad46aWXuo1TUlKCF154oVt7oG0knq8BK2lvb8fBgwexd+9eVFRUYO/evTh79iwAYNCgQZg8eTImT56MKVOmYPz48UhKSuo2jWCvu3guVxHZr6qFIcezauDdddddeP3117u1B6o32IK/9tpr0dzc7Ll1fnV8uPr27dujsAx1X28K023btqG4uBi///3vsXDhwqinF+4Lyd94gbYRp9OJ1tZW9O3bF6qKlpYWpKamen5nwe5UFZ988okn/CoqKlBXVwfA1dG4+eabPSE4YcIEOJ3OiN90YiXhAy/Sd46nn34aK1as6Na+fv16/OQnP/Fpu3TpEr766iufEPS+Bbsv2P1fffUV2tvbw3p+ndLS0noUllYJ0xEjRqC+vh4AcP3116OystKQ+W7atAlLlizp1r5582af36qIJPAocidOnEBFRYUnBGtqagC4vqm4qKgIFRUVAR/LHl4MAy/QY+L5/FQVly5dMjxI4x2moYLWO0y9l/nu3bsxbdo0w5Zn18MaDofDZxlYZXfKTs6dO4d9+/Z5QvCjjz4KOC4DL06BN3PmTLz99ttBx0000YRpOEEbaZiGqtUIIgKHw4EFCxZgx44d6OjoCLk7a3QNFNzLL7+MBx98sFv7H/7wB8ybNy9uddgy8KhngoVpoLBcv349Ll682G1aOTk5OHbsWFzqtsrxI7sze8/KXUNiB15ycrLfXkdSUlLEJx0oNiyyoZteg90lJSUhOTkZixcvxqZNm9DW1mboHkM4wg08y54iPHjwoN/2zgPkZB1WONN83XXXmV2CbbW3t6O1tRXPPvssWltb4x52kUg2uwB/gu3ODh06NI6VUDCR9qRi0Rtjb44iYf5bM9lCoDcxo39ZnigYBh4R2YYlA+/jjz/2215dXR3nSoioN7Fk4BUUFPhtz8/Pj3MlFA/craV4sWTgAUB2djbGjBmDV199FWPGjEF2drbZJRFRgrPkWVoAOHXqlGd4/vz5JlZCRL2FZXt4RERGCxl4IjJCRPaISI2IHBaRR7rcv1xEVEQGxq5MSnS8Xo6sIJxd2jYAy1X1gIhkAtgvIrtUtUZERgC4DcDxmFZJRGSAkD08VT2tqgfcwxcA1AIY5r77GQA/heurnYmILC2iY3gikgvgegAfiMh3AZxU1aAXx4lIiYhUikhl59dHkz0F2q3l7i7FS9iBJyIZAMoALIVrN3clgCdCPU5VN6tqoaoWDho0qMeFUu/g74dViOIlrMATkRS4wm6rqu4E8L8A5AGoFpE6AMMBHBARfrKfiCwr5EkLcV0G/yKAWlXdCACq+g8Ag73GqQNQqKrnYlQnEVHUwunh3QxgIYBpIlLlvs2KcV1ERIYL2cNT1X0Agn7YUVVzjSqIiChW+EkLIrINBh4R2QYDj4hsg4FHRLbBwCMi22DgEZFtMPCIyDYYeERkGww8IrINBh4R2QYDj4hsg4FHRLbBwCMi22DgEZFtMPCIyDYYeERkGww8IrINBh4R2UbIr3gnosR2+fJl1NfXo6WlxexSouZ0OjF8+HCkpKT06PEMPKJerr6+HpmZmcjNzYXrRwgTk6qiqakJ9fX1yMvL69E0uEtLliMi3W7Ucy0tLRgwYEDCL0cRwYABA6LqqTLwyFICvSgT/cVqtt6y/KJ9Hgw8IrINBh4RdXP69GlMnToVDQ0NcZvnX//6V/Tv3x8FBQUoKCjA6tWrDZ8HT1oQUTdr1qzBvn37sHr1ajz//PNxm+/kyZPxxhtvxGz6DDwiG1m6dCmqqqoC3l9RUYGOjg7P/5s2bcKmTZvgcDgwefJkv48pKCjAs88+G3S+dXV1mDFjBiZNmoT33nsPEyZMwIMPPoif//znaGxsxNatW3v2hCIUcpdWREaIyB4RqRGRwyLyiLv9FyJyUkSq3LdZsS+XejtVNbsEWysqKsLgwYPhcLiiweFwYPDgwZg4cWLU0/7ss8+wfPlyHDlyBEeOHMG2bduwb98+bNiwAevWrQMAvPfee8jPz8fMmTNx+PDhqOfZVTg9vDYAy1X1gIhkAtgvIrvc9z2jqhsMr4rIDxFhIEYpVE8MABYvXozNmzfD6XTi0qVLmDt3riG7tXl5eRg3bhwAYMyYMbjlllsgIhg3bhzq6upwww034Pjx48jIyMBbb72FO++8E59++mnU8/UWsoenqqdV9YB7+AKAWgDDDK2CKEy95fIKKztz5gxKS0vx/vvvo7S01LATF6mpqZ5hh8Ph+d/hcKCtrQ39+vVDRkYGAGDWrFm4fPkyzp07Z8i8O0V0DE9EcgFcD+ADADcDeEhE7gdQCVcv8J9+HlMCoAQAcnJyoiyXiGJt586dnuFf//rXcZtvQ0MDhgwZAhHBhx9+iI6ODgwYMMDQeYR9WYqIZAAoA7BUVc8D2ATgGgAFAE4D+D/+Hqeqm1W1UFULBw0aZEDJRNQbvfbaaxg7dizGjx+Phx9+GDt27DC8Ry/hHBMRkRQAbwAoV9WNfu7PBfCGqo4NNp3CwkKtrKzsWaVkG6E2ch7Hi0xtbS1GjRpldhmG8fd8RGS/qhaGemw4Z2kFwIsAar3DTkSyvUabA+BQ2BUTBcFAo1gJ5xjezQAWAviHiHRewLMSwH+ISAEABVAH4IcxqZCIyCAhA09V9wHwt4/xlvHlEBHFDj9LS5YUaLeWu7sUDX60jCyL4UZGYw+PiGyDgUdEtsHAIyIfW7duRW5uLhwOB3Jzc+P2TSbxwMAjIo+tW7eipKQEx44dg6ri2LFjKCkpiTr06urqMHbsvz+XsGHDBvziF7+IstrIMfCIyGPVqlX4+uuvfdq+/vprrFq1yqSKjMXAIyKP48ePR9SeaBh4ROQR6BuNov2mo+TkZJ9vUjbrR8EZeETksXbtWqSlpfm0paWlYe3atVFNd8iQIWhsbERTUxNaW1tj+rsVwVj2wmN/35jBC1Gpq2DfrMLtJXLFxcUAXMfyjh8/jpycHKxdu9bT3lMpKSl44oknUFRUhGHDhmHkyJFGlBuxsL4eyijhfj0UN2IKRzjflcbthV8P5Y27tERkGww8IrINBh6RDfSWXftonwcDj6iXczqdaGpqSvjQU1U0NTXB6XT2eBqWPEurqjxLSyEF2k687ydg+PDhqK+vx9mzZ80uJWpOpxPDhw/v8eMtGXgAN1YKD7eT0FJSUpCXl2d2GZbAXVoisg0GHhHZBgOPiGwjrp+0EJGzAI5F+LCBAM7FoJyeYj3BsZ7ArFQL0LvquVpVB4UaKa6B1xMiUhnOR0bihfUEx3oCs1ItgD3r4S4tEdkGA4+IbCMRAm+z2QV0wXqCYz2BWakWwIb1WP4YHhGRURKhh0dEZAgGHhHZhqUDT0RmiMhREflMRB6L0zxfEpFGETnk1XaliOwSkU/df7O87lvhru+oiEw3uJYRIrJHRGpE5LCIPGJyPU4R+VBEqt31PGlmPV7zSBKRj0XkDbPrEZE6EfmHiFSJSKWZ9YjIFSLymogcEZFaEbnJxFqucy+Tztt5EVka93pU1ZI3AEkA/gfANQD6AKgGMDoO850C4AYAh7za/hPAY+7hxwD80j082l1XKoA8d71JBtaSDeAG93AmgE/c8zSrHgGQ4R5OAfABgElm1eNV16MAtgF4w8z15Z5HHYCBXdrMWl+/A/C/3cN9AFxh9rpyzysJQAOAq+Ndj+FPxsCFchOAcq//VwBYEad558I38I4CyHYPZwM46q8mAOUAbophXX8CcKsV6gGQBuAAgIlm1gNgOIDdAKZ5BZ6Z9fgLvLjXA6A/gC/gPjFppW0ZwG0A/mZGPVbepR0G4ITX//XuNjMMUdXT7uEGAEPcw3GrUURyAVwPV6/KtHrcu49VABoB7FJVU+sB8CyAnwLo8Gozsx4F8N8isl9ESkysJw/AWQD/z727/39FJN2kWrpaAGC7eziu9Vg58CxJXW83cb2WR0QyAJQBWKqq582sR1XbVbUArp5VkYiMNaseEbkdQKOq7g80jgnr65vu5TMTwI9EZIpJ9STDdWhmk6peD6AZrl1GM2rxEJE+AO4A8Ieu98WjHisH3kkAI7z+H+5uM8MZEckGAPffRnd7zGsUkRS4wm6rqu40u55OqvolgD0AZphYz80A7hCROgA7AEwTkS0m1gNVPen+2wjgdQBFJtVTD6De3QMHgNfgCkCzt52ZAA6o6hn3/3Gtx8qB9xGAa0Ukz/2usADAf5lUy38BeMA9/ABcx9I62xeISKqI5AG4FsCHRs1URATAiwBqVXWjBeoZJCJXuIf7wnU88YhZ9ajqClUdrqq5cG0ff1HV+8yqR0TSRSSzcxiuY1WHzKhHVRsAnBCR69xNtwCoMaOWLv4D/96d7Zxv/OqJxUFJAw9uzoLrzOT/AFgVp3luB3AawGW43iV/AGAAXAfGPwXw3wCu9Bp/lbu+owBmGlzLN+Hq4h8EUOW+zTKxnnwAH7vrOQTgCXe7KfV0qe1b+PdJC7OWzzVwnVmsBnC4c5s1sZ4CAJXu9fVHAFlmrisA6QCaAPT3aotrPfxoGRHZhpV3aYmIDMXAIyLbYOARkW0w8IjINhh4RGQbDDwisg0GHhHZxv8HlyafUk75QwoAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fcc1df37898>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fcc1dca00f0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fcc1de44198>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "dataSlice.sort(order='observationStartMJD')\n",
    "lcv_template = read_lightCurve(asciifile)\n",
    "transDuration = lcv_template['ph'].max() - lcv_template['ph'].min() # in days\n",
    "\n",
    "time = (dataSlice['observationStartMJD'] - dataSlice['observationStartMJD'].min() - 22)\n",
    "lcEpoch = (dataSlice['observationStartMJD'] - dataSlice['observationStartMJD'].min() - 22) % transDuration\n",
    "filters = dataSlice['filter']\n",
    "lcMags = make_lightCurve(lcEpoch, filters)\n",
    "\n",
    "colors = {'u':'k', 'g':'b', 'r':'r', 'i':'m', 'z':'orange', 'y':'c'}\n",
    "\n",
    "for f in np.unique(filters):\n",
    "    \n",
    "    plt.figure(figsize=(5, 3))\n",
    "    plt.scatter(time[np.where(filters==f)], lcMags[np.where(filters==f)], color=colors[f], label=f)\n",
    "    \n",
    "    plt.plot(time[np.where(filters==f)], dataSlice['fiveSigmaDepth'][np.where(filters==f)], color=colors[f], marker='*', label='m5')\n",
    "    plt.legend()\n",
    "    plt.ylim(plt.ylim()[::-1])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(1184,)"
      ]
     },
     "execution_count": 70,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "lcMags.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### sketch function run() inside metric"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 331,
   "metadata": {},
   "outputs": [],
   "source": [
    "# columns needed, change to self.varname when add into class\n",
    "asciifile = 'TDEfaintfast_z0.1.dat'\n",
    "mjdCol = 'observationStartMJD'\n",
    "m5Col = 'fiveSigmaDepth'\n",
    "filterCol = 'filter'\n",
    "\n",
    "# light curve parameters\n",
    "tshift = -22\n",
    "peakEpoch = 0\n",
    "nearPeakT = 5\n",
    "\n",
    "# condition parameters, add to __init__()\n",
    "detectSNR={'u': 5, 'g': 5, 'r': 5, 'i': 5, 'z': 5, 'y': 5}\n",
    "nTotal = 1 # total number of observation\n",
    "   \n",
    "# sort dataSlice in order of time\n",
    "dataSlice.sort(order=mjdCol)\n",
    "tSpan = (dataSlice[mjdCol].max() - dataSlice[mjdCol].min()) # in days\n",
    "    \n",
    "lcv_template = read_lightCurve(asciifile)\n",
    "transDuration = lcv_template['ph'].max() - lcv_template['ph'].min() # in days\n",
    "\n",
    "# get the epoch of each light curve by np.fmod\n",
    "lcEpoch = np.fmod(dataSlice[mjdCol] - dataSlice[mjdCol].min(), transDuration) + tshift\n",
    "\n",
    "# total number of transients possibly detected\n",
    "nTransMax = np.ceil(tSpan/transDuration)\n",
    "\n",
    "# generate the actual light curve\n",
    "lcMags = make_lightCurve(lcEpoch, filters)\n",
    "lcSNR = utils.m52snr(lcMags, dataSlice[m5Col])\n",
    "lcFilters = dataSlice[filterCol]\n",
    "\n",
    "# Identify detections above SNR for each filter\n",
    "lcAboveThresh = np.zeros(len(lcSNR), dtype=bool)\n",
    "for f in np.unique(lcFilters):\n",
    "    filtermatch = np.where(dataSlice[filterCol] == f)\n",
    "    lcAboveThresh[filtermatch] = np.where(lcSNR[filtermatch] >= detectSNR[f], True, False)\n",
    "    \n",
    "\n",
    "lcNumber = np.floor((dataSlice[mjdCol] - dataSlice[mjdCol].min()) / transDuration)\n",
    "ulcNumber = np.unique(lcNumber)\n",
    "\n",
    "# check conditions for each light curve\n",
    "lcDetect = np.ones(len(ulcNumber), dtype=bool)\n",
    "for i, lcN in enumerate(ulcNumber):\n",
    "\n",
    "    lcN_idx = np.where(lcNumber == lcN)\n",
    "    \n",
    "    \n",
    "    lcEpoch_i = lcEpoch[lcN_idx]\n",
    "    lcMags_i = lcMags[lcN_idx]\n",
    "    lcFilters_i = lcFilters[lcN_idx]\n",
    "    lcAboveThresh_i = lcAboveThresh[lcN_idx]\n",
    "    \n",
    "    \n",
    "    #------  add conditions below --------------\n",
    "    # total number of observations\n",
    "    nObsTotal = len(lcAboveThresh_i == True)\n",
    "    \n",
    "    # pre peak \n",
    "    \n",
    "    # near Peak  \n",
    "    nearPeak = (lcEpoch_i >= peakEpoch - nearPeakT/2) & (lcEpoch_i <= peakEpoch + nearPeakT/2) \n",
    "    nearPeakIdx = np.where(nearPeak==True)\n",
    "    \n",
    "    # number of observations near peak\n",
    "    nObsNearPeak =  len (lcAboveThresh_i[nearPeakIdx] == True)\n",
    "        \n",
    "\n",
    "\n",
    "# return values   \n",
    "nDetected = len(np.where(lcDetect == True))\n",
    "    \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 273,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "-22"
      ]
     },
     "execution_count": 273,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.fmod(-22, 130)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 186,
   "metadata": {},
   "outputs": [],
   "source": [
    "from functools import reduce\n",
    "idx = reduce(np.intersect1d, (np.where(lcAboveThresh==True), np.where(lcNumber==2), np.where(lcFilters=='r')) )\n",
    "\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 308,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>time</th>\n",
       "      <th>lcEpoch</th>\n",
       "      <th>lcMags</th>\n",
       "      <th>filter</th>\n",
       "      <th>lcSNR</th>\n",
       "      <th>lcAboveThresh</th>\n",
       "      <th>lcNumber</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>200</th>\n",
       "      <td>17.908657</td>\n",
       "      <td>-4.091343</td>\n",
       "      <td>23.237910</td>\n",
       "      <td>y</td>\n",
       "      <td>1.845644</td>\n",
       "      <td>False</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>201</th>\n",
       "      <td>17.909074</td>\n",
       "      <td>-4.090926</td>\n",
       "      <td>23.237854</td>\n",
       "      <td>y</td>\n",
       "      <td>1.846534</td>\n",
       "      <td>False</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>202</th>\n",
       "      <td>17.924155</td>\n",
       "      <td>-4.075845</td>\n",
       "      <td>23.235797</td>\n",
       "      <td>y</td>\n",
       "      <td>2.024280</td>\n",
       "      <td>False</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>203</th>\n",
       "      <td>19.903044</td>\n",
       "      <td>-2.096956</td>\n",
       "      <td>21.324379</td>\n",
       "      <td>u</td>\n",
       "      <td>46.268163</td>\n",
       "      <td>True</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>204</th>\n",
       "      <td>19.903461</td>\n",
       "      <td>-2.096539</td>\n",
       "      <td>21.324357</td>\n",
       "      <td>u</td>\n",
       "      <td>46.286615</td>\n",
       "      <td>True</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>205</th>\n",
       "      <td>19.903877</td>\n",
       "      <td>-2.096123</td>\n",
       "      <td>21.324334</td>\n",
       "      <td>u</td>\n",
       "      <td>46.305076</td>\n",
       "      <td>True</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>206</th>\n",
       "      <td>19.904294</td>\n",
       "      <td>-2.095706</td>\n",
       "      <td>21.324311</td>\n",
       "      <td>u</td>\n",
       "      <td>46.323548</td>\n",
       "      <td>True</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>207</th>\n",
       "      <td>19.904711</td>\n",
       "      <td>-2.095289</td>\n",
       "      <td>21.324289</td>\n",
       "      <td>u</td>\n",
       "      <td>46.342031</td>\n",
       "      <td>True</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>208</th>\n",
       "      <td>19.905127</td>\n",
       "      <td>-2.094873</td>\n",
       "      <td>21.324266</td>\n",
       "      <td>u</td>\n",
       "      <td>46.360523</td>\n",
       "      <td>True</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>209</th>\n",
       "      <td>19.905544</td>\n",
       "      <td>-2.094456</td>\n",
       "      <td>21.324243</td>\n",
       "      <td>u</td>\n",
       "      <td>46.379026</td>\n",
       "      <td>True</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>210</th>\n",
       "      <td>19.905961</td>\n",
       "      <td>-2.094039</td>\n",
       "      <td>21.324220</td>\n",
       "      <td>u</td>\n",
       "      <td>46.397538</td>\n",
       "      <td>True</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>211</th>\n",
       "      <td>19.906377</td>\n",
       "      <td>-2.093623</td>\n",
       "      <td>21.324198</td>\n",
       "      <td>u</td>\n",
       "      <td>46.029771</td>\n",
       "      <td>True</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>212</th>\n",
       "      <td>19.906794</td>\n",
       "      <td>-2.093206</td>\n",
       "      <td>21.324175</td>\n",
       "      <td>u</td>\n",
       "      <td>46.048151</td>\n",
       "      <td>True</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>213</th>\n",
       "      <td>19.907211</td>\n",
       "      <td>-2.092789</td>\n",
       "      <td>21.324152</td>\n",
       "      <td>u</td>\n",
       "      <td>46.066540</td>\n",
       "      <td>True</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>214</th>\n",
       "      <td>19.907627</td>\n",
       "      <td>-2.092373</td>\n",
       "      <td>21.324129</td>\n",
       "      <td>u</td>\n",
       "      <td>46.084939</td>\n",
       "      <td>True</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>215</th>\n",
       "      <td>19.908044</td>\n",
       "      <td>-2.091956</td>\n",
       "      <td>21.324107</td>\n",
       "      <td>u</td>\n",
       "      <td>46.103349</td>\n",
       "      <td>True</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>216</th>\n",
       "      <td>19.908461</td>\n",
       "      <td>-2.091539</td>\n",
       "      <td>21.324084</td>\n",
       "      <td>u</td>\n",
       "      <td>46.121769</td>\n",
       "      <td>True</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>217</th>\n",
       "      <td>19.908877</td>\n",
       "      <td>-2.091123</td>\n",
       "      <td>21.324061</td>\n",
       "      <td>u</td>\n",
       "      <td>46.140199</td>\n",
       "      <td>True</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>218</th>\n",
       "      <td>19.909294</td>\n",
       "      <td>-2.090706</td>\n",
       "      <td>21.324039</td>\n",
       "      <td>u</td>\n",
       "      <td>46.158639</td>\n",
       "      <td>True</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>219</th>\n",
       "      <td>19.909711</td>\n",
       "      <td>-2.090289</td>\n",
       "      <td>21.324016</td>\n",
       "      <td>u</td>\n",
       "      <td>46.177090</td>\n",
       "      <td>True</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>220</th>\n",
       "      <td>19.910127</td>\n",
       "      <td>-2.089873</td>\n",
       "      <td>21.323993</td>\n",
       "      <td>u</td>\n",
       "      <td>44.984385</td>\n",
       "      <td>True</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>221</th>\n",
       "      <td>19.910544</td>\n",
       "      <td>-2.089456</td>\n",
       "      <td>21.323970</td>\n",
       "      <td>u</td>\n",
       "      <td>45.002372</td>\n",
       "      <td>True</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>222</th>\n",
       "      <td>19.910961</td>\n",
       "      <td>-2.089039</td>\n",
       "      <td>21.323948</td>\n",
       "      <td>u</td>\n",
       "      <td>45.020369</td>\n",
       "      <td>True</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>223</th>\n",
       "      <td>20.871991</td>\n",
       "      <td>-1.128009</td>\n",
       "      <td>21.271528</td>\n",
       "      <td>u</td>\n",
       "      <td>30.991220</td>\n",
       "      <td>True</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>224</th>\n",
       "      <td>20.872407</td>\n",
       "      <td>-1.127593</td>\n",
       "      <td>21.271505</td>\n",
       "      <td>u</td>\n",
       "      <td>31.026047</td>\n",
       "      <td>True</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>225</th>\n",
       "      <td>20.872824</td>\n",
       "      <td>-1.127176</td>\n",
       "      <td>21.271482</td>\n",
       "      <td>u</td>\n",
       "      <td>31.060928</td>\n",
       "      <td>True</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>226</th>\n",
       "      <td>20.873241</td>\n",
       "      <td>-1.126759</td>\n",
       "      <td>21.271460</td>\n",
       "      <td>u</td>\n",
       "      <td>31.095864</td>\n",
       "      <td>True</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>227</th>\n",
       "      <td>20.873657</td>\n",
       "      <td>-1.126343</td>\n",
       "      <td>21.271437</td>\n",
       "      <td>u</td>\n",
       "      <td>31.125550</td>\n",
       "      <td>True</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>228</th>\n",
       "      <td>20.874074</td>\n",
       "      <td>-1.125926</td>\n",
       "      <td>21.271414</td>\n",
       "      <td>u</td>\n",
       "      <td>31.153236</td>\n",
       "      <td>True</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>229</th>\n",
       "      <td>20.874491</td>\n",
       "      <td>-1.125509</td>\n",
       "      <td>21.271391</td>\n",
       "      <td>u</td>\n",
       "      <td>31.180956</td>\n",
       "      <td>True</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>270</th>\n",
       "      <td>22.877870</td>\n",
       "      <td>0.877870</td>\n",
       "      <td>21.281923</td>\n",
       "      <td>u</td>\n",
       "      <td>36.661907</td>\n",
       "      <td>True</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>271</th>\n",
       "      <td>22.878287</td>\n",
       "      <td>0.878287</td>\n",
       "      <td>21.281938</td>\n",
       "      <td>u</td>\n",
       "      <td>36.691925</td>\n",
       "      <td>True</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>272</th>\n",
       "      <td>22.878704</td>\n",
       "      <td>0.878704</td>\n",
       "      <td>21.281953</td>\n",
       "      <td>u</td>\n",
       "      <td>39.118401</td>\n",
       "      <td>True</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>273</th>\n",
       "      <td>22.879120</td>\n",
       "      <td>0.879120</td>\n",
       "      <td>21.281968</td>\n",
       "      <td>u</td>\n",
       "      <td>39.150453</td>\n",
       "      <td>True</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>274</th>\n",
       "      <td>22.879537</td>\n",
       "      <td>0.879537</td>\n",
       "      <td>21.281983</td>\n",
       "      <td>u</td>\n",
       "      <td>39.182542</td>\n",
       "      <td>True</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>275</th>\n",
       "      <td>22.879954</td>\n",
       "      <td>0.879954</td>\n",
       "      <td>21.281998</td>\n",
       "      <td>u</td>\n",
       "      <td>39.214669</td>\n",
       "      <td>True</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>276</th>\n",
       "      <td>22.880370</td>\n",
       "      <td>0.880370</td>\n",
       "      <td>21.282013</td>\n",
       "      <td>u</td>\n",
       "      <td>39.245422</td>\n",
       "      <td>True</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>277</th>\n",
       "      <td>22.880787</td>\n",
       "      <td>0.880787</td>\n",
       "      <td>21.282029</td>\n",
       "      <td>u</td>\n",
       "      <td>39.269160</td>\n",
       "      <td>True</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>278</th>\n",
       "      <td>22.881204</td>\n",
       "      <td>0.881204</td>\n",
       "      <td>21.282044</td>\n",
       "      <td>u</td>\n",
       "      <td>39.292919</td>\n",
       "      <td>True</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>279</th>\n",
       "      <td>22.881620</td>\n",
       "      <td>0.881620</td>\n",
       "      <td>21.282059</td>\n",
       "      <td>u</td>\n",
       "      <td>39.316699</td>\n",
       "      <td>True</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>280</th>\n",
       "      <td>22.882037</td>\n",
       "      <td>0.882037</td>\n",
       "      <td>21.282074</td>\n",
       "      <td>u</td>\n",
       "      <td>39.030587</td>\n",
       "      <td>True</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>281</th>\n",
       "      <td>22.882454</td>\n",
       "      <td>0.882454</td>\n",
       "      <td>21.282089</td>\n",
       "      <td>u</td>\n",
       "      <td>39.054221</td>\n",
       "      <td>True</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>282</th>\n",
       "      <td>22.882870</td>\n",
       "      <td>0.882870</td>\n",
       "      <td>21.282104</td>\n",
       "      <td>u</td>\n",
       "      <td>39.077875</td>\n",
       "      <td>True</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>283</th>\n",
       "      <td>22.883287</td>\n",
       "      <td>0.883287</td>\n",
       "      <td>21.282120</td>\n",
       "      <td>u</td>\n",
       "      <td>39.101550</td>\n",
       "      <td>True</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>284</th>\n",
       "      <td>22.883704</td>\n",
       "      <td>0.883704</td>\n",
       "      <td>21.282135</td>\n",
       "      <td>u</td>\n",
       "      <td>39.125246</td>\n",
       "      <td>True</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>285</th>\n",
       "      <td>23.862245</td>\n",
       "      <td>1.862245</td>\n",
       "      <td>21.317718</td>\n",
       "      <td>u</td>\n",
       "      <td>34.990690</td>\n",
       "      <td>True</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>286</th>\n",
       "      <td>23.862662</td>\n",
       "      <td>1.862662</td>\n",
       "      <td>21.317733</td>\n",
       "      <td>u</td>\n",
       "      <td>35.030601</td>\n",
       "      <td>True</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>287</th>\n",
       "      <td>23.863079</td>\n",
       "      <td>1.863079</td>\n",
       "      <td>21.317748</td>\n",
       "      <td>u</td>\n",
       "      <td>35.067765</td>\n",
       "      <td>True</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>288</th>\n",
       "      <td>23.863495</td>\n",
       "      <td>1.863495</td>\n",
       "      <td>21.317763</td>\n",
       "      <td>u</td>\n",
       "      <td>35.099362</td>\n",
       "      <td>True</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>289</th>\n",
       "      <td>23.863912</td>\n",
       "      <td>1.863912</td>\n",
       "      <td>21.317779</td>\n",
       "      <td>u</td>\n",
       "      <td>35.131000</td>\n",
       "      <td>True</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>290</th>\n",
       "      <td>23.864329</td>\n",
       "      <td>1.864329</td>\n",
       "      <td>21.317794</td>\n",
       "      <td>u</td>\n",
       "      <td>35.162678</td>\n",
       "      <td>True</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>291</th>\n",
       "      <td>23.864745</td>\n",
       "      <td>1.864745</td>\n",
       "      <td>21.317809</td>\n",
       "      <td>u</td>\n",
       "      <td>35.188433</td>\n",
       "      <td>True</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>292</th>\n",
       "      <td>23.865162</td>\n",
       "      <td>1.865162</td>\n",
       "      <td>21.317824</td>\n",
       "      <td>u</td>\n",
       "      <td>35.220188</td>\n",
       "      <td>True</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>293</th>\n",
       "      <td>23.865579</td>\n",
       "      <td>1.865579</td>\n",
       "      <td>21.317839</td>\n",
       "      <td>u</td>\n",
       "      <td>35.251983</td>\n",
       "      <td>True</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>294</th>\n",
       "      <td>23.865995</td>\n",
       "      <td>1.865995</td>\n",
       "      <td>21.317854</td>\n",
       "      <td>u</td>\n",
       "      <td>35.283819</td>\n",
       "      <td>True</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>295</th>\n",
       "      <td>23.866412</td>\n",
       "      <td>1.866412</td>\n",
       "      <td>21.317870</td>\n",
       "      <td>u</td>\n",
       "      <td>35.315697</td>\n",
       "      <td>True</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>296</th>\n",
       "      <td>23.866829</td>\n",
       "      <td>1.866829</td>\n",
       "      <td>21.317885</td>\n",
       "      <td>u</td>\n",
       "      <td>35.347615</td>\n",
       "      <td>True</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>297</th>\n",
       "      <td>23.867245</td>\n",
       "      <td>1.867245</td>\n",
       "      <td>21.317900</td>\n",
       "      <td>u</td>\n",
       "      <td>35.379574</td>\n",
       "      <td>True</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>298</th>\n",
       "      <td>23.867662</td>\n",
       "      <td>1.867662</td>\n",
       "      <td>21.317915</td>\n",
       "      <td>u</td>\n",
       "      <td>35.411575</td>\n",
       "      <td>True</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>299</th>\n",
       "      <td>23.868079</td>\n",
       "      <td>1.868079</td>\n",
       "      <td>21.317930</td>\n",
       "      <td>u</td>\n",
       "      <td>35.443617</td>\n",
       "      <td>True</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>100 rows × 7 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "          time   lcEpoch     lcMags filter      lcSNR  lcAboveThresh  lcNumber\n",
       "200  17.908657 -4.091343  23.237910      y   1.845644          False       0.0\n",
       "201  17.909074 -4.090926  23.237854      y   1.846534          False       0.0\n",
       "202  17.924155 -4.075845  23.235797      y   2.024280          False       0.0\n",
       "203  19.903044 -2.096956  21.324379      u  46.268163           True       0.0\n",
       "204  19.903461 -2.096539  21.324357      u  46.286615           True       0.0\n",
       "205  19.903877 -2.096123  21.324334      u  46.305076           True       0.0\n",
       "206  19.904294 -2.095706  21.324311      u  46.323548           True       0.0\n",
       "207  19.904711 -2.095289  21.324289      u  46.342031           True       0.0\n",
       "208  19.905127 -2.094873  21.324266      u  46.360523           True       0.0\n",
       "209  19.905544 -2.094456  21.324243      u  46.379026           True       0.0\n",
       "210  19.905961 -2.094039  21.324220      u  46.397538           True       0.0\n",
       "211  19.906377 -2.093623  21.324198      u  46.029771           True       0.0\n",
       "212  19.906794 -2.093206  21.324175      u  46.048151           True       0.0\n",
       "213  19.907211 -2.092789  21.324152      u  46.066540           True       0.0\n",
       "214  19.907627 -2.092373  21.324129      u  46.084939           True       0.0\n",
       "215  19.908044 -2.091956  21.324107      u  46.103349           True       0.0\n",
       "216  19.908461 -2.091539  21.324084      u  46.121769           True       0.0\n",
       "217  19.908877 -2.091123  21.324061      u  46.140199           True       0.0\n",
       "218  19.909294 -2.090706  21.324039      u  46.158639           True       0.0\n",
       "219  19.909711 -2.090289  21.324016      u  46.177090           True       0.0\n",
       "220  19.910127 -2.089873  21.323993      u  44.984385           True       0.0\n",
       "221  19.910544 -2.089456  21.323970      u  45.002372           True       0.0\n",
       "222  19.910961 -2.089039  21.323948      u  45.020369           True       0.0\n",
       "223  20.871991 -1.128009  21.271528      u  30.991220           True       0.0\n",
       "224  20.872407 -1.127593  21.271505      u  31.026047           True       0.0\n",
       "225  20.872824 -1.127176  21.271482      u  31.060928           True       0.0\n",
       "226  20.873241 -1.126759  21.271460      u  31.095864           True       0.0\n",
       "227  20.873657 -1.126343  21.271437      u  31.125550           True       0.0\n",
       "228  20.874074 -1.125926  21.271414      u  31.153236           True       0.0\n",
       "229  20.874491 -1.125509  21.271391      u  31.180956           True       0.0\n",
       "..         ...       ...        ...    ...        ...            ...       ...\n",
       "270  22.877870  0.877870  21.281923      u  36.661907           True       0.0\n",
       "271  22.878287  0.878287  21.281938      u  36.691925           True       0.0\n",
       "272  22.878704  0.878704  21.281953      u  39.118401           True       0.0\n",
       "273  22.879120  0.879120  21.281968      u  39.150453           True       0.0\n",
       "274  22.879537  0.879537  21.281983      u  39.182542           True       0.0\n",
       "275  22.879954  0.879954  21.281998      u  39.214669           True       0.0\n",
       "276  22.880370  0.880370  21.282013      u  39.245422           True       0.0\n",
       "277  22.880787  0.880787  21.282029      u  39.269160           True       0.0\n",
       "278  22.881204  0.881204  21.282044      u  39.292919           True       0.0\n",
       "279  22.881620  0.881620  21.282059      u  39.316699           True       0.0\n",
       "280  22.882037  0.882037  21.282074      u  39.030587           True       0.0\n",
       "281  22.882454  0.882454  21.282089      u  39.054221           True       0.0\n",
       "282  22.882870  0.882870  21.282104      u  39.077875           True       0.0\n",
       "283  22.883287  0.883287  21.282120      u  39.101550           True       0.0\n",
       "284  22.883704  0.883704  21.282135      u  39.125246           True       0.0\n",
       "285  23.862245  1.862245  21.317718      u  34.990690           True       0.0\n",
       "286  23.862662  1.862662  21.317733      u  35.030601           True       0.0\n",
       "287  23.863079  1.863079  21.317748      u  35.067765           True       0.0\n",
       "288  23.863495  1.863495  21.317763      u  35.099362           True       0.0\n",
       "289  23.863912  1.863912  21.317779      u  35.131000           True       0.0\n",
       "290  23.864329  1.864329  21.317794      u  35.162678           True       0.0\n",
       "291  23.864745  1.864745  21.317809      u  35.188433           True       0.0\n",
       "292  23.865162  1.865162  21.317824      u  35.220188           True       0.0\n",
       "293  23.865579  1.865579  21.317839      u  35.251983           True       0.0\n",
       "294  23.865995  1.865995  21.317854      u  35.283819           True       0.0\n",
       "295  23.866412  1.866412  21.317870      u  35.315697           True       0.0\n",
       "296  23.866829  1.866829  21.317885      u  35.347615           True       0.0\n",
       "297  23.867245  1.867245  21.317900      u  35.379574           True       0.0\n",
       "298  23.867662  1.867662  21.317915      u  35.411575           True       0.0\n",
       "299  23.868079  1.868079  21.317930      u  35.443617           True       0.0\n",
       "\n",
       "[100 rows x 7 columns]"
      ]
     },
     "execution_count": 308,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_tde = pd.DataFrame()\n",
    "\n",
    "df_tde['time'] = dataSlice['observationStartMJD'] - dataSlice['observationStartMJD'].min()\n",
    "df_tde['lcEpoch'] = lcEpoch\n",
    "df_tde['lcMags'] = lcMags\n",
    "df_tde['filter'] = lcFilters\n",
    "df_tde['lcSNR'] = lcSNR\n",
    "df_tde['lcAboveThresh'] = lcAboveThresh\n",
    "\n",
    "df_tde['lcNumber'] = lcNumber\n",
    "df_tde.to_csv('df_tde.csv')\n",
    "df_tde.iloc[200:300]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 119,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "6.0"
      ]
     },
     "execution_count": 119,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "nTransMax"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### import the Class "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# combine all above functions into TDEsMetricTest class\n",
    "\n",
    "class TDEsMetricTest(BaseMetric):\n",
    "    \"\"\"Based on the transientMetric, but uses an ascii input file and provides option to write out lightcurve.\n",
    "    \n",
    "    Calculate what fraction of the transients would be detected. Best paired with a spatial slicer.\n",
    "    The lightcurve in input is an ascii file per photometric band so that different lightcurve\n",
    "    shapes can be implemented.\n",
    "\n",
    "    Parameters\n",
    "    ----------\n",
    "    asciifile : str\n",
    "        The ascii file containing the inputs for the lightcurve (per filter):\n",
    "        File should contain three columns - ['ph', 'mag', 'flt'] -\n",
    "        of phase/epoch (in days), magnitude (in a particular filter), and filter.\n",
    "    \n",
    "    detectSNR : dict, optional\n",
    "        An observation will be counted toward the discovery criteria if the light curve SNR\n",
    "        is higher than detectSNR (specified per bandpass).\n",
    "        Values must be provided for each filter which should be considered in the lightcurve.\n",
    "        Default is {'u': 5, 'g': 5, 'r': 5, 'i': 5, 'z': 5, 'y': 5}\n",
    "    \n",
    "\n",
    "    dataout : bool, optional\n",
    "        If True, metric returns full lightcurve at each point. Note that this will potentially\n",
    "        create a very large metric output data file.\n",
    "        If False, metric returns the number of transients detected.\n",
    "    \"\"\"\n",
    "    def __init__(self, asciifile, metricName='TDEsMetricTest', mjdCol='expMJD',\n",
    "                 m5Col='fiveSigmaDepth', filterCol='filter', \n",
    "                 detectSNR={'u': 5, 'g': 5, 'r': 5, 'i': 5, 'z': 5, 'y': 5}, \n",
    "                 tshift=0, peakEpoch=0, nearPeakT=5, \n",
    "                 nObsNearPeak={'u': 5, 'g': 5, 'r': 5, 'i': 5, 'z': 5, 'y': 5},\n",
    "                 dataout=False, **kwargs):\n",
    "        \n",
    "        self.mjdCol = mjdCol\n",
    "        self.m5Col = m5Col\n",
    "        self.filterCol = filterCol\n",
    "        self.dataout = dataout\n",
    "\n",
    "        # condition parameters\n",
    "        self.detectSNR = detectSNR\n",
    "        self.tshift = tshift\n",
    "        self.peakEpoch = peakEpoch\n",
    "        self.nearPeakT = nearPeakT\n",
    "\n",
    "        self.nObsNearPeak = nObsNearPeak\n",
    "\n",
    "        # if you want to get the light curve in output you need to define the metricDtype as object\n",
    "        if self.dataout:\n",
    "            super(TDEsMetricTest, self).__init__(col=[self.mjdCol, self.m5Col, self.filterCol],\n",
    "                                                       metricDtype='object', units='',\n",
    "                                                       metricName=metricName, **kwargs)\n",
    "        else:\n",
    "            super(TDEsMetricTest, self).__init__(col=[self.mjdCol, self.m5Col, self.filterCol],\n",
    "                                                       units='Fraction Detected', metricName=metricName,\n",
    "                                                       **kwargs)\n",
    "        self.read_lightCurve(asciifile)\n",
    "    \n",
    "        print('Finish initializing metric')\n",
    "\n",
    "    def read_lightCurve(self, asciifile):\n",
    "        \n",
    "        if not os.path.isfile(asciifile):\n",
    "            raise IOError('Could not find lightcurve ascii file %s' % (asciifile))\n",
    "\n",
    "        self.lcv_template = np.genfromtxt(asciifile, dtype=[('ph', 'f8'), ('mag', 'f8'), ('flt', 'S1')])\n",
    "    \n",
    "\n",
    "    def make_lightCurve(self, time, filters):\n",
    "                \n",
    "        lcv_template = self.lcv_template\n",
    "        \n",
    "        lcMags = np.zeros(time.size, dtype=float)\n",
    "        \n",
    "        for f in set(lcv_template['flt']):\n",
    "            fMatch_ascii = np.where(np.array(lcv_template['flt']) == f)[0]\n",
    "            \n",
    "            # Interpolate the lightcurve template to the times of the observations, in this filter.\n",
    "            lc_ascii_filter = np.interp(time, np.array(lcv_template['ph'], float)[fMatch_ascii],\n",
    "                                            np.array(lcv_template['mag'], float)[fMatch_ascii])\n",
    "            lcMags[filters == f.decode(\"utf-8\")] = lc_ascii_filter[filters == f.decode(\"utf-8\")]\n",
    "        \n",
    "        return lcMags\n",
    "\n",
    "    def run(self, dataSlice, slicePoint=None):\n",
    "        \"\"\"\"Calculate the detectability of a transient with the specified lightcurve.\n",
    "\n",
    "        If self.dataout is True, then returns the full lightcurve for each object instead of the total\n",
    "        number of transients that are detected.\n",
    "\n",
    "        Parameters\n",
    "        ----------\n",
    "        dataSlice : numpy.array\n",
    "            Numpy structured array containing the data related to the visits provided by the slicer.\n",
    "        slicePoint : dict, optional\n",
    "            Dictionary containing information about the slicepoint currently active in the slicer.\n",
    "\n",
    "        Returns\n",
    "        -------\n",
    "        float or dict\n",
    "            The total number of transients that could be detected. (if dataout is False)\n",
    "            A dictionary with arrays of 'lcNumber', 'lcMag', 'detected', 'time', 'detectThresh', 'filter'\n",
    "        \"\"\"\n",
    "\n",
    "        # Sort the entire dataSlice in order of time.  \n",
    "        dataSlice.sort(order=self.mjdCol)\n",
    "        tSpan = (dataSlice[self.mjdCol].max() - dataSlice[self.mjdCol].min()) # in days\n",
    "        \n",
    "        lcv_template = self.lcv_template\n",
    "        transDuration = lcv_template['ph'].max() - lcv_template['ph'].min() # in days\n",
    "\n",
    "        lcEpoch = np.fmod(dataSlice[self.mjdCol] - dataSlice[self.mjdCol].min(), transDuration) + self.tshift\n",
    "\n",
    "        # total number of transients possibly detected\n",
    "        nTransMax = np.ceil(tSpan/transDuration)\n",
    "\n",
    "        # generate the actual light curve\n",
    "        lcFilters = dataSlice[self.filterCol]\n",
    "        lcMags = self.make_lightCurve(lcEpoch, lcFilters)\n",
    "        lcSNR = utils.m52snr(lcMags, dataSlice[self.m5Col])\n",
    "\n",
    "        # Identify detections above SNR for each filter\n",
    "        lcAboveThresh = np.zeros(len(lcSNR), dtype=bool)\n",
    "        for f in np.unique(lcFilters):\n",
    "            filtermatch = np.where(dataSlice[self.filterCol] == f)\n",
    "            lcAboveThresh[filtermatch] = np.where(lcSNR[filtermatch] >= self.detectSNR[f], True, False)\n",
    "            \n",
    "\n",
    "        lcNumber = np.floor((dataSlice[self.mjdCol] - dataSlice[self.mjdCol].min()) / transDuration)\n",
    "        ulcNumber = np.unique(lcNumber)\n",
    "\n",
    "        # check conditions for each light curve\n",
    "        lcDetect = np.ones(len(ulcNumber), dtype=bool)\n",
    "        for i, lcN in enumerate(ulcNumber):\n",
    "\n",
    "            lcN_idx = np.where(lcNumber == lcN)\n",
    "            \n",
    "            lcEpoch_i = lcEpoch[lcN_idx]\n",
    "            lcMags_i = lcMags[lcN_idx]\n",
    "            lcFilters_i = lcFilters[lcN_idx]\n",
    "            lcAboveThresh_i = lcAboveThresh[lcN_idx]\n",
    "            \n",
    "            # total number of observations\n",
    "            nObsTotal = len(lcAboveThresh_i == True)\n",
    "            \n",
    "            # prePeakCheck \n",
    "            prePeakCheck = (lcEpoch_i < self.peakEpoch)\n",
    "\n",
    "\n",
    "            # number of observations before peak\n",
    "\n",
    "\n",
    "\n",
    "            # near Peak \n",
    "            nearPeakCheck = (lcEpoch_i >= self.peakEpoch - self.nearPeakT/2) & (lcEpoch_i <= self.peakEpoch + self.nearPeakT/2) \n",
    "            nearPeakIdx = np.where(nearPeakCheck==True)\n",
    "            \n",
    "            # number of observations near peak\n",
    "            nObsNearPeakTotal =  len (lcAboveThresh_i[nearPeakIdx] == True)\n",
    "                \n",
    "            # check number of observations near peak for each band\n",
    "            for f in np.unique(lcFilters_i):\n",
    "                nearPeakIdx_f = np.intersect1d( nearPeakIdx, np.where(lcFilters_i==f) )\n",
    "\n",
    "                if len(lcAboveThresh_i[nearPeakIdx_f]) <= self.nObsNearPeak[f]:\n",
    "\n",
    "                    lcDetect[i] = False\n",
    "\n",
    "            # number of filters near band\n",
    "            lcFilters_i[nearPeakIdx]\n",
    "\n",
    "\n",
    "        # return values   \n",
    "        nDetected = len(np.where(lcDetect == True)[0])\n",
    "\n",
    "        print(nTransMax, nDetected, lcDetect)\n",
    "\n",
    "        return float(nDetected / nTransMax) if nTransMax!=0 else 0\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "# import from source file\n",
    "from mycode.TDEsMetricTest import TDEsMetricTest\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "finish init\n",
      "Healpix slicer using NSIDE=16, approximate resolution 219.871130 arcminutes\n",
      "Querying database SummaryAllProps with constraint night<700 for columns ['filter', 'fieldRA', 'fiveSigmaDepth', 'observationStartMJD', 'fieldDec']\n",
      "Found 431650 visits\n",
      "Running:  ['transmetricSky']\n",
      "6.0 5 [ True  True  True  True  True]\n",
      "6.0 2 [ True  True]\n",
      "4.0 1 [False  True False]\n",
      "4.0 1 [ True False False]\n",
      "4.0 1 [False  True False]\n",
      "4.0 1 [False  True False]\n",
      "4.0 2 [False  True  True False]\n",
      "6.0 3 [False  True  True False  True]\n",
      "6.0 1 [False False False  True]\n",
      "6.0 1 [False False False  True]\n",
      "4.0 0 [False False False]\n",
      "4.0 0 [False False False]\n",
      "4.0 1 [False  True False]\n",
      "4.0 1 [False  True False]\n",
      "4.0 1 [False False  True]\n",
      "4.0 1 [False  True False]\n",
      "4.0 0 [False False False]\n",
      "4.0 2 [False  True  True False]\n",
      "4.0 2 [False  True  True False]\n",
      "6.0 2 [False  True False False  True]\n",
      "6.0 3 [False  True  True False  True]\n",
      "6.0 1 [False False False  True]\n",
      "6.0 0 [False False False False]\n",
      "6.0 1 [ True False False False]\n",
      "6.0 0 [False False False False]\n",
      "6.0 1 [False False False  True]\n",
      "6.0 1 [False False False  True]\n",
      "4.0 0 [False False False]\n",
      "4.0 0 [False False False]\n",
      "4.0 1 [False  True False]\n",
      "4.0 1 [False  True False]\n",
      "4.0 2 [False  True  True False]\n",
      "4.0 0 [False False False]\n",
      "4.0 1 [False False  True]\n",
      "4.0 2 [False  True  True False]\n",
      "4.0 0 [False False False]\n",
      "4.0 0 [False False False]\n",
      "4.0 1 [False  True False]\n",
      "6.0 3 [False  True  True False  True]\n",
      "6.0 1 [False False False  True]\n",
      "6.0 0 [False False False False]\n",
      "6.0 0 [False False False False]\n",
      "6.0 0 [False False False False]\n",
      "6.0 1 [False False False  True]\n",
      "6.0 1 [False False False  True]\n",
      "4.0 0 [False False False]\n",
      "4.0 0 [False False False]\n",
      "4.0 1 [False  True False]\n",
      "4.0 1 [False  True False]\n",
      "4.0 1 [False  True False]\n",
      "4.0 0 [False False False]\n",
      "4.0 1 [False  True False False]\n",
      "4.0 2 [False  True False  True]\n",
      "4.0 1 [False  True False]\n",
      "4.0 0 [False False False]\n",
      "4.0 1 [False  True False False]\n",
      "4.0 1 [False  True False]\n",
      "4.0 0 [False False False]\n",
      "4.0 2 [False  True  True False]\n",
      "6.0 2 [False  True False False  True]\n",
      "6.0 1 [False False False  True False]\n",
      "6.0 1 [False False False  True]\n",
      "6.0 2 [False False  True  True]\n",
      "6.0 0 [False False False False]\n",
      "6.0 1 [ True False False False]\n",
      "6.0 0 [False False False False]\n",
      "6.0 1 [False False False  True]\n",
      "4.0 0 [False False False]\n",
      "4.0 0 [False False False]\n",
      "4.0 0 [False False False]\n",
      "4.0 1 [False  True False]\n",
      "4.0 1 [False  True False]\n",
      "4.0 1 [False False  True]\n",
      "4.0 1 [False  True False False]\n",
      "4.0 1 [False False  True False]\n",
      "4.0 1 [False  True False]\n",
      "4.0 1 [False  True False]\n",
      "4.0 1 [False  True False False]\n",
      "4.0 2 [False  True  True False]\n",
      "4.0 1 [False  True False False]\n",
      "4.0 1 [False  True False False]\n",
      "4.0 1 [False False  True]\n",
      "4.0 2 [False  True  True False]\n",
      "6.0 2 [False  True False False  True]\n",
      "6.0 2 [False  True False False  True]\n",
      "6.0 2 [False False  True  True False]\n",
      "6.0 1 [False False False  True False]\n",
      "6.0 1 [False False False  True]\n",
      "6.0 0 [False False False False]\n",
      "6.0 0 [False False False False]\n",
      "6.0 0 [False False False False]\n",
      "4.0 0 [False False False]\n",
      "6.0 1 [False False False  True]\n",
      "4.0 0 [False False False]\n",
      "4.0 0 [False False False]\n",
      "4.0 1 [False  True False]\n",
      "4.0 2 [False  True  True False]\n",
      "4.0 0 [False False False]\n",
      "4.0 2 [False  True  True False]\n",
      "4.0 1 [False  True False False]\n",
      "4.0 1 [False  True False]\n",
      "4.0 1 [False  True False]\n",
      "4.0 1 [False  True False False]\n",
      "4.0 0 [False False False]\n",
      "4.0 2 [False  True  True False]\n",
      "4.0 0 [False False False]\n",
      "4.0 2 [False  True  True False]\n",
      "4.0 1 [False False  True]\n",
      "4.0 1 [False  True False False]\n",
      "4.0 1 [False  True False False]\n",
      "6.0 2 [False  True False False  True]\n",
      "6.0 2 [False  True False False  True]\n",
      "6.0 1 [False False  True False False]\n",
      "6.0 1 [False False  True False False]\n",
      "6.0 2 [False False  True  True False]\n",
      "6.0 1 [False False False  True False]\n",
      "6.0 0 [False False False False]\n",
      "6.0 1 [False False False  True]\n",
      "6.0 0 [False False False False]\n",
      "6.0 1 [ True False False False]\n",
      "4.0 0 [False False False]\n",
      "6.0 1 [False False False  True]\n",
      "4.0 0 [False False False]\n",
      "4.0 0 [False False False]\n",
      "4.0 0 [False False False]\n",
      "4.0 1 [False  True False]\n",
      "4.0 1 [False  True False]\n",
      "4.0 1 [False  True False False]\n",
      "4.0 1 [False  True False False]\n",
      "4.0 1 [False  True False]\n",
      "4.0 1 [False  True False]\n",
      "4.0 1 [False  True False]\n",
      "4.0 0 [False False False]\n",
      "4.0 2 [False  True  True False]\n",
      "4.0 1 [False  True False False]\n",
      "4.0 1 [False  True False False]\n",
      "4.0 1 [False False  True False]\n",
      "4.0 1 [False False  True False]\n",
      "4.0 1 [False  True False False]\n",
      "5.0 2 [False  True False False  True]\n",
      "6.0 2 [False  True False False  True]\n",
      "6.0 2 [False  True False False  True]\n",
      "6.0 0 [False False False False]\n",
      "6.0 1 [False False  True False False]\n",
      "6.0 2 [False False  True  True False]\n",
      "6.0 1 [False False False  True False]\n",
      "6.0 0 [False False False False]\n",
      "6.0 0 [False False False False]\n",
      "6.0 1 [False False  True False]\n",
      "6.0 1 [False False False  True]\n",
      "6.0 1 [False False False  True]\n",
      "6.0 2 [ True False False  True]\n",
      "4.0 0 [False False False]\n",
      "4.0 0 [False False False]\n",
      "4.0 0 [False False False]\n",
      "4.0 1 [False  True False]\n",
      "4.0 2 [False  True  True False]\n",
      "4.0 2 [False  True  True False]\n",
      "4.0 1 [False  True False False]\n",
      "4.0 1 [False  True False]\n",
      "4.0 1 [False  True False]\n",
      "4.0 0 [False False False]\n",
      "4.0 1 [False  True False]\n",
      "4.0 1 [False  True False False]\n",
      "4.0 0 [False False False]\n",
      "4.0 2 [False  True  True False]\n",
      "4.0 1 [False False  True False]\n",
      "4.0 2 [False  True  True False]\n",
      "4.0 1 [False  True False False]\n",
      "5.0 2 [False  True False False  True]\n",
      "6.0 2 [False  True False False  True]\n",
      "6.0 2 [False False  True False  True]\n",
      "6.0 1 [False False  True False False]\n",
      "6.0 1 [False False  True False False]\n",
      "6.0 1 [False False False  True False]\n",
      "6.0 1 [False False False  True False]\n",
      "6.0 1 [False False  True]\n",
      "6.0 1 [False False  True False]\n",
      "6.0 0 [False False False False]\n",
      "6.0 1 [ True False False False]\n",
      "6.0 1 [False False False  True]\n",
      "6.0 1 [False False False  True]\n",
      "4.0 0 [False False False]\n",
      "4.0 0 [False False False]\n",
      "4.0 0 [False False False]\n",
      "4.0 1 [False  True False]\n",
      "4.0 2 [False  True  True False]\n",
      "4.0 2 [False  True  True False]\n",
      "4.0 0 [False False False False]\n",
      "4.0 1 [False False  True False]\n",
      "4.0 1 [False  True False]\n",
      "4.0 0 [False False False]\n",
      "4.0 0 [False False False]\n",
      "4.0 1 [False False  True]\n",
      "4.0 2 [False  True False  True]\n",
      "4.0 1 [False  True False False]\n",
      "4.0 1 [False False  True False]\n",
      "4.0 1 [False False  True False]\n",
      "4.0 1 [False  True False False]\n",
      "5.0 2 [False  True False False  True]\n",
      "5.0 2 [False  True False False  True]\n",
      "4.0 1 [False  True False False]\n",
      "6.0 2 [False  True False False  True]\n",
      "6.0 2 [False  True False False  True]\n",
      "6.0 1 [False False  True False False]\n",
      "6.0 2 [False False  True False  True]\n",
      "6.0 1 [False False  True False False]\n",
      "6.0 1 [False False  True False False]\n",
      "6.0 1 [False False False  True False]\n",
      "6.0 2 [False False False  True  True]\n",
      "6.0 1 [False False  True False]\n",
      "6.0 0 [False False False False]\n",
      "6.0 0 [False False False False]\n",
      "6.0 0 [False False False False]\n",
      "6.0 1 [False False False  True]\n",
      "4.0 0 [False False False]\n",
      "4.0 0 [False False False]\n",
      "4.0 0 [False False False False]\n",
      "4.0 0 [False False False]\n",
      "5.0 3 [False  True  True False  True]\n",
      "4.0 2 [False  True  True False]\n",
      "4.0 2 [False  True  True False]\n",
      "4.0 0 [False False False False]\n",
      "4.0 1 [False False  True False]\n",
      "4.0 1 [False  True False]\n",
      "4.0 1 [False False  True]\n",
      "4.0 0 [False False False]\n",
      "4.0 1 [False  True False False]\n",
      "4.0 0 [False False False]\n",
      "4.0 2 [False  True  True False]\n",
      "4.0 1 [False False  True False]\n",
      "4.0 1 [False  True False]\n",
      "5.0 3 [False  True  True False  True]\n",
      "5.0 2 [False  True False False  True]\n",
      "4.0 1 [False  True False False]\n",
      "5.0 1 [False  True False False False]\n",
      "4.0 2 [False  True  True False]\n",
      "6.0 2 [ True  True False False False]\n",
      "6.0 2 [False  True False False  True]\n",
      "6.0 1 [False False  True False False]\n",
      "6.0 1 [False False  True False False]\n",
      "6.0 1 [False False  True False False]\n",
      "6.0 1 [False False  True False False]\n",
      "6.0 1 [False False  True False False]\n",
      "6.0 2 [False False  True  True False]\n",
      "6.0 1 [False False False  True False]\n",
      "6.0 0 [False False False False]\n",
      "6.0 0 [False False False False]\n",
      "6.0 0 [False False False False]\n",
      "5.0 0 [False False False False]\n",
      "4.0 0 [False False False]\n",
      "4.0 0 [False False False]\n",
      "4.0 0 [False False False]\n",
      "4.0 1 [False  True False]\n",
      "4.0 0 [False False False]\n",
      "4.0 2 [False  True  True False]\n",
      "5.0 3 [False  True  True False  True]\n",
      "4.0 1 [False False  True False]\n",
      "4.0 1 [False  True False False]\n",
      "4.0 1 [False False  True False]\n",
      "4.0 1 [False  True False]\n",
      "4.0 1 [False  True False False]\n",
      "4.0 0 [False False False]\n",
      "4.0 1 [False False  True]\n",
      "4.0 1 [False  True False]\n",
      "4.0 1 [False  True False]\n",
      "4.0 1 [False False  True False]\n",
      "4.0 1 [False False  True False]\n",
      "5.0 2 [False  True False False  True]\n",
      "5.0 2 [False  True False False  True]\n",
      "5.0 2 [False  True False False  True]\n",
      "4.0 1 [False  True False False]\n",
      "5.0 1 [False  True False False False]\n",
      "5.0 2 [False  True  True False False]\n",
      "6.0 1 [ True False False False False]\n",
      "6.0 2 [False  True False False  True]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "6.0 2 [False  True False False  True]\n",
      "6.0 0 [False False False False False]\n",
      "6.0 2 [False False  True False  True]\n",
      "6.0 1 [False False  True False False]\n",
      "6.0 1 [False False  True False False]\n",
      "6.0 2 [False False  True False  True]\n",
      "6.0 2 [False False  True  True False]\n",
      "6.0 2 [False False  True  True False]\n",
      "6.0 2 [False False  True  True False]\n",
      "5.0 3 [False  True False  True  True]\n",
      "6.0 0 [False False False False]\n",
      "5.0 0 [False False False False]\n",
      "6.0 0 [False False False False]\n",
      "4.0 0 [False False False]\n",
      "4.0 0 [False False False]\n",
      "4.0 1 [False  True False]\n",
      "4.0 0 [False False False]\n",
      "4.0 0 [False False False False]\n",
      "4.0 1 [False False  True False]\n",
      "4.0 2 [False  True  True False]\n",
      "4.0 1 [False False  True False]\n",
      "4.0 0 [False False False False]\n",
      "4.0 1 [False False  True False]\n",
      "4.0 0 [False False False]\n",
      "4.0 0 [False False False False]\n",
      "4.0 1 [False  True False]\n",
      "4.0 1 [False  True False False]\n",
      "4.0 1 [False  True False False]\n",
      "4.0 0 [False False False]\n",
      "4.0 2 [False  True  True False]\n",
      "4.0 2 [False  True  True False]\n",
      "4.0 1 [False  True False False]\n",
      "5.0 2 [False  True False False  True]\n",
      "4.0 2 [False  True  True False]\n",
      "5.0 1 [False  True False False False]\n",
      "5.0 1 [False  True False False False]\n",
      "5.0 2 [False  True  True False False]\n",
      "4.0 2 [False  True  True False]\n",
      "6.0 2 [ True False False  True False False]\n",
      "6.0 1 [ True False False False False]\n",
      "6.0 2 [False  True False False  True]\n",
      "6.0 0 [False False False False False]\n",
      "6.0 1 [False False  True False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 1 [False False  True False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 0 [False False False False]\n",
      "6.0 0 [False False False False]\n",
      "6.0 1 [False False False  True False]\n",
      "6.0 1 [False  True False False]\n",
      "6.0 3 [False  True False  True  True]\n",
      "6.0 1 [False  True False False]\n",
      "6.0 0 [False False False False]\n",
      "4.0 0 [False False False]\n",
      "6.0 2 [False  True False  True]\n",
      "6.0 2 [False  True False  True]\n",
      "4.0 0 [False False False]\n",
      "4.0 2 [False  True  True False]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 2 [False  True False False  True]\n",
      "4.0 0 [False False False False]\n",
      "4.0 1 [False False  True False]\n",
      "4.0 1 [False False  True False]\n",
      "4.0 1 [False False  True False]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 2 [False False  True False  True]\n",
      "4.0 1 [False  True False]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 2 [False False  True False  True]\n",
      "4.0 0 [False False False False]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 1 [False False  True False False]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 2 [False  True False False  True]\n",
      "5.0 0 [False False False False False]\n",
      "5.0 1 [False  True False False False]\n",
      "5.0 1 [False False  True False False]\n",
      "5.0 1 [False False False False  True]\n",
      "5.0 1 [False False False False  True]\n",
      "4.0 1 [False False  True False]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 2 [False  True  True False False]\n",
      "5.0 2 [False  True  True False False]\n",
      "5.0 1 [False False  True False False]\n",
      "4.0 1 [False False  True False]\n",
      "4.0 0 [False False False False]\n",
      "6.0 1 [ True False False False False]\n",
      "6.0 3 [ True False  True  True False False]\n",
      "6.0 2 [False  True False False  True]\n",
      "6.0 1 [ True False False False False]\n",
      "6.0 1 [ True False False False False]\n",
      "6.0 1 [ True False False False False]\n",
      "6.0 2 [ True False False False  True]\n",
      "6.0 0 [False False False False False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 2 [ True  True False False False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 0 [False False False False False]\n",
      "6.0 1 [False False False  True False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 0 [False False False False]\n",
      "6.0 1 [False  True False False]\n",
      "6.0 2 [False  True False  True]\n",
      "6.0 1 [False  True False False]\n",
      "6.0 2 [False  True False  True False]\n",
      "6.0 0 [False False False False]\n",
      "6.0 1 [False False False  True False]\n",
      "6.0 2 [False  True False  True]\n",
      "6.0 2 [False  True False  True]\n",
      "4.0 1 [False  True False False]\n",
      "4.0 0 [False False False False]\n",
      "4.0 1 [False  True False False]\n",
      "4.0 2 [False  True  True False]\n",
      "5.0 1 [False False False False  True]\n",
      "4.0 0 [False False False False]\n",
      "4.0 1 [False False  True False]\n",
      "4.0 1 [False False  True False]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 2 [False False  True False  True]\n",
      "4.0 0 [False False False False]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 2 [False  True False False  True]\n",
      "4.0 1 [False False  True False]\n",
      "4.0 1 [False False  True False]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 1 [False False False False  True]\n",
      "5.0 1 [False False  True False False]\n",
      "5.0 0 [False False False False False]\n",
      "5.0 0 [False False False False False]\n",
      "5.0 1 [False False False False  True]\n",
      "5.0 1 [False False False False  True]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 2 [False False  True False  True]\n",
      "4.0 0 [False False False False]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 2 [False  True  True False False]\n",
      "5.0 1 [False False  True False False]\n",
      "5.0 1 [False  True False False False]\n",
      "4.0 0 [False False False False]\n",
      "6.0 2 [ True False False  True False False]\n",
      "6.0 1 [ True False False False False]\n",
      "6.0 2 [ True False False  True False False]\n",
      "6.0 2 [ True False  True False False]\n",
      "6.0 3 [False  True False  True False  True]\n",
      "6.0 1 [ True False False False False]\n",
      "6.0 1 [ True False False False False]\n",
      "6.0 1 [ True False False False False]\n",
      "6.0 1 [ True False False False False]\n",
      "6.0 0 [False False False False False]\n",
      "6.0 2 [ True False False False False  True]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 2 [ True  True False False False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 0 [False False False False False]\n",
      "6.0 1 [False False False  True False]\n",
      "6.0 1 [False  True False False]\n",
      "6.0 1 [False  True False False]\n",
      "6.0 0 [False False False False]\n",
      "6.0 0 [False False False False]\n",
      "6.0 1 [False False False  True False]\n",
      "6.0 1 [False False False  True]\n",
      "6.0 2 [False  True False  True]\n",
      "6.0 2 [False  True False  True]\n",
      "4.0 1 [False  True False False]\n",
      "4.0 1 [False False  True False]\n",
      "4.0 1 [False False  True False]\n",
      "4.0 0 [False False False False]\n",
      "4.0 0 [False False False False]\n",
      "4.0 1 [False False  True False]\n",
      "4.0 1 [False False  True False]\n",
      "4.0 1 [False False  True False]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 2 [False False  True False  True]\n",
      "4.0 1 [False  True False False]\n",
      "5.0 1 [False False False False  True]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 2 [False False  True False  True]\n",
      "4.0 1 [False False  True False]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 1 [False False  True False False]\n",
      "5.0 1 [False False False False  True]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 1 [False False False False  True]\n",
      "5.0 1 [False False  True False False]\n",
      "5.0 1 [False False  True False False]\n",
      "5.0 1 [False False False False  True]\n",
      "5.0 1 [False False False False  True]\n",
      "5.0 1 [False False False False  True]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 2 [False  True  True False False]\n",
      "5.0 1 [False False  True False False]\n",
      "5.0 2 [False  True  True False False]\n",
      "4.0 0 [False False False False]\n",
      "4.0 0 [False False False False]\n",
      "6.0 2 [ True False False  True False False]\n",
      "4.0 2 [False  True  True False]\n",
      "6.0 2 [ True False  True False False]\n",
      "6.0 4 [False  True  True  True False  True]\n",
      "6.0 1 [ True False False False False]\n",
      "6.0 1 [ True False False False False]\n",
      "6.0 1 [ True False False False False]\n",
      "6.0 1 [ True False False False False]\n",
      "6.0 1 [ True False False False False]\n",
      "6.0 0 [False False False False False]\n",
      "6.0 0 [False False False False False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 0 [False False False False False]\n",
      "6.0 1 [False False False  True]\n",
      "6.0 0 [False False False False]\n",
      "6.0 1 [False False False  True False]\n",
      "6.0 1 [False False False  True]\n",
      "6.0 0 [False False False False]\n",
      "6.0 0 [False False False False]\n",
      "6.0 0 [False False False False]\n",
      "6.0 1 [False False False  True]\n",
      "6.0 1 [False False False  True]\n",
      "6.0 2 [False  True False False  True]\n",
      "4.0 1 [False False  True False]\n",
      "4.0 0 [False False False False]\n",
      "4.0 1 [False False  True False]\n",
      "5.0 1 [False False False False  True]\n",
      "4.0 1 [False False  True False]\n",
      "4.0 1 [False False  True False]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 2 [False False  True False  True]\n",
      "4.0 0 [False False False False]\n",
      "5.0 1 [False False False False  True]\n",
      "5.0 1 [False False False False  True]\n",
      "5.0 2 [False False  True False  True]\n",
      "4.0 1 [False False  True False]\n",
      "5.0 1 [False False  True False False]\n",
      "5.0 1 [False False  True False False]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "5.0 2 [False  True False False  True]\n",
      "5.0 1 [False False False False  True]\n",
      "5.0 2 [False  True False False  True]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 1 [False False  True False False]\n",
      "5.0 1 [False False False False  True]\n",
      "5.0 1 [False False False False  True]\n",
      "5.0 1 [False False False False  True]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 2 [False  True  True False False]\n",
      "5.0 2 [False  True  True False False]\n",
      "4.0 1 [False  True False False]\n",
      "4.0 0 [False False False False]\n",
      "6.0 2 [ True False False  True False False]\n",
      "4.0 1 [False False  True False]\n",
      "4.0 2 [False  True  True False]\n",
      "6.0 2 [ True False False  True False]\n",
      "6.0 1 [ True False False False False]\n",
      "6.0 1 [ True False False False False]\n",
      "6.0 1 [ True False False False False]\n",
      "6.0 1 [ True False False False False]\n",
      "6.0 1 [ True False False False False]\n",
      "6.0 0 [False False False False False]\n",
      "6.0 0 [False False False False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 0 [False False False False False]\n",
      "6.0 0 [False False False False]\n",
      "6.0 0 [False False False False]\n",
      "6.0 1 [False  True False False]\n",
      "6.0 0 [False False False False]\n",
      "6.0 2 [False False False  True  True]\n",
      "6.0 1 [False False False  True]\n",
      "6.0 1 [False False False  True]\n",
      "6.0 1 [False False False  True]\n",
      "6.0 1 [False False False  True]\n",
      "6.0 1 [False False False  True]\n",
      "4.0 1 [False False  True False]\n",
      "4.0 2 [False  True  True False]\n",
      "4.0 1 [False False  True False]\n",
      "4.0 0 [False False False False]\n",
      "4.0 1 [False False  True False]\n",
      "4.0 1 [False False  True False]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 2 [False False  True False  True]\n",
      "4.0 1 [False False  True False]\n",
      "4.0 0 [False False False False]\n",
      "5.0 1 [False False False False  True]\n",
      "5.0 1 [False False False False  True]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 1 [False False  True False False]\n",
      "5.0 2 [False  True False False  True]\n",
      "5.0 2 [False  True False False  True]\n",
      "5.0 1 [False False  True False False]\n",
      "5.0 1 [False False  True False False]\n",
      "5.0 0 [False False False False False]\n",
      "5.0 1 [False False False False  True]\n",
      "5.0 1 [False False False False  True]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 2 [False False  True False  True]\n",
      "4.0 1 [False  True False False]\n",
      "5.0 2 [False  True  True False False]\n",
      "5.0 1 [False  True False False False]\n",
      "4.0 1 [False  True False False]\n",
      "6.0 2 [ True False False  True False False]\n",
      "6.0 2 [ True False False  True False False]\n",
      "4.0 1 [False  True False]\n",
      "6.0 2 [ True False False False  True]\n",
      "6.0 1 [ True False False False False]\n",
      "6.0 1 [ True False False False False]\n",
      "6.0 1 [ True False False False False]\n",
      "6.0 1 [ True False False False False]\n",
      "6.0 1 [ True False False False False]\n",
      "6.0 1 [ True False False False False]\n",
      "6.0 0 [False False False False False]\n",
      "6.0 1 [False False False False False  True]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 2 [False  True False False False  True]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 0 [False False False False False]\n",
      "6.0 0 [False False False False]\n",
      "6.0 0 [False False False False]\n",
      "6.0 1 [False False False  True False]\n",
      "6.0 1 [False False False  True]\n",
      "6.0 0 [False False False False]\n",
      "6.0 1 [False False False  True]\n",
      "6.0 0 [False False False False]\n",
      "6.0 1 [False False False  True]\n",
      "6.0 1 [False False False  True]\n",
      "6.0 2 [False  True False False  True]\n",
      "6.0 2 [False False  True False  True]\n",
      "6.0 1 [False False False False  True]\n",
      "4.0 1 [False False  True False]\n",
      "5.0 1 [False False False False  True]\n",
      "4.0 0 [False False False False]\n",
      "4.0 1 [False False  True False]\n",
      "4.0 1 [False False  True False]\n",
      "5.0 1 [False False False False  True]\n",
      "4.0 0 [False False False False]\n",
      "4.0 0 [False False False False]\n",
      "4.0 1 [False  True False False]\n",
      "5.0 2 [False False  True False  True]\n",
      "4.0 0 [False False False False]\n",
      "4.0 1 [False False  True False]\n",
      "5.0 1 [False False  True False False]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 2 [False  True False False  True]\n",
      "5.0 1 [False False False False  True]\n",
      "5.0 2 [False  True False False  True]\n",
      "5.0 0 [False False False False False]\n",
      "5.0 0 [False False False False False]\n",
      "5.0 2 [False  True False False  True]\n",
      "5.0 1 [False False False False  True]\n",
      "5.0 1 [False False False False  True]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 1 [False  True False False False]\n",
      "4.0 1 [False  True False False]\n",
      "4.0 0 [False False False False]\n",
      "6.0 2 [ True False False  True False False]\n",
      "4.0 1 [False False  True False]\n",
      "4.0 1 [False False  True False]\n",
      "6.0 5 [ True False  True  True  True  True]\n",
      "6.0 1 [ True False False False False]\n",
      "6.0 1 [ True False False False False]\n",
      "6.0 1 [ True False False False False]\n",
      "6.0 1 [ True False False False False]\n",
      "6.0 1 [ True False False False False]\n",
      "6.0 0 [False False False False False]\n",
      "6.0 1 [ True False False False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 2 [ True  True False False False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 0 [False False False False False]\n",
      "6.0 0 [False False False False False]\n",
      "6.0 0 [False False False False]\n",
      "6.0 1 [False  True False False]\n",
      "6.0 1 [False  True False False]\n",
      "6.0 1 [False False False  True]\n",
      "6.0 1 [False False False  True False]\n",
      "6.0 1 [False False False  True]\n",
      "6.0 2 [False  True False  True]\n",
      "6.0 1 [False False False  True]\n",
      "6.0 1 [False False False  True]\n",
      "6.0 2 [False  True False False  True]\n",
      "6.0 2 [False False  True False  True]\n",
      "4.0 0 [False False False False]\n",
      "4.0 1 [False False  True False]\n",
      "5.0 2 [False False  True False  True]\n",
      "4.0 1 [False False  True False]\n",
      "4.0 1 [False False  True False]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 2 [False False  True False  True]\n",
      "4.0 1 [False False  True False]\n",
      "4.0 0 [False False False False]\n",
      "5.0 1 [False False False False  True]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 1 [False False  True False False]\n",
      "5.0 2 [False  True False False  True]\n",
      "5.0 2 [False  True False False  True]\n",
      "5.0 2 [False  True False False  True]\n",
      "5.0 1 [False False False False  True]\n",
      "5.0 1 [False False  True False False]\n",
      "5.0 1 [False False False False  True]\n",
      "5.0 1 [False False False False  True]\n",
      "5.0 1 [False False False False  True]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 2 [False  True  True False False]\n",
      "5.0 1 [False False  True False False]\n",
      "4.0 1 [False  True False False]\n",
      "6.0 2 [ True False False  True False False]\n",
      "4.0 1 [False False  True False]\n",
      "6.0 4 [ True False  True  True False  True]\n",
      "6.0 3 [ True False False  True False  True]\n",
      "6.0 1 [ True False False False False]\n",
      "6.0 1 [ True False False False False]\n",
      "6.0 1 [ True False False False False]\n",
      "6.0 1 [ True False False False False]\n",
      "6.0 1 [ True False False False False]\n",
      "6.0 1 [ True False False False False]\n",
      "6.0 0 [False False False False False]\n",
      "6.0 2 [False False False  True False  True]\n",
      "6.0 1 [False  True False False False False]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "6.0 1 [False  True False False False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 0 [False False False False]\n",
      "6.0 0 [False False False False]\n",
      "6.0 0 [False False False False]\n",
      "6.0 0 [False False False False]\n",
      "6.0 1 [False False False  True]\n",
      "6.0 0 [False False False False]\n",
      "6.0 0 [False False False False]\n",
      "6.0 0 [False False False False]\n",
      "6.0 1 [False False False  True]\n",
      "6.0 2 [False  True False  True]\n",
      "6.0 2 [False  True False False  True]\n",
      "6.0 2 [False False  True False  True]\n",
      "4.0 1 [False False  True False]\n",
      "4.0 0 [False False False False]\n",
      "4.0 0 [False False False False]\n",
      "4.0 0 [False False False False]\n",
      "4.0 1 [False False  True False]\n",
      "4.0 1 [False False  True False]\n",
      "5.0 2 [False False  True False  True]\n",
      "4.0 0 [False False False False]\n",
      "4.0 0 [False False False False]\n",
      "4.0 1 [False  True False False]\n",
      "5.0 2 [False False  True False  True]\n",
      "4.0 0 [False False False False]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 1 [False False  True False False]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 2 [False  True False False  True]\n",
      "5.0 1 [False False False False  True]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 0 [False False False False False]\n",
      "5.0 1 [False False False False  True]\n",
      "5.0 2 [False  True False False  True]\n",
      "5.0 1 [False False False False  True]\n",
      "5.0 1 [False False False False  True]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 1 [False False False False  True]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 1 [False False  True False False]\n",
      "5.0 1 [False False  True False False]\n",
      "4.0 1 [False  True False False]\n",
      "6.0 2 [ True False False  True False False]\n",
      "4.0 2 [ True False  True False]\n",
      "6.0 3 [ True False False  True False  True]\n",
      "6.0 3 [ True False False  True False  True]\n",
      "6.0 3 [ True False  True  True False]\n",
      "6.0 1 [ True False False False False]\n",
      "6.0 1 [ True False False False False]\n",
      "6.0 1 [ True False False False False]\n",
      "6.0 1 [ True False False False False]\n",
      "6.0 1 [ True False False False False]\n",
      "6.0 0 [False False False False False]\n",
      "4.0 1 [False False  True False]\n",
      "6.0 2 [False False False  True False  True]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 2 [False  True False False False  True]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 2 [False  True False False False  True]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 0 [False False False False False]\n",
      "6.0 0 [False False False False False]\n",
      "6.0 0 [False False False False]\n",
      "6.0 2 [False  True False  True False]\n",
      "6.0 0 [False False False False]\n",
      "6.0 2 [False  True False  True]\n",
      "6.0 1 [False False False  True False]\n",
      "6.0 1 [False False False  True]\n",
      "6.0 2 [False  True False  True]\n",
      "6.0 1 [False False False  True]\n",
      "6.0 1 [False False False  True]\n",
      "6.0 2 [False  True False False  True]\n",
      "4.0 1 [False False  True False]\n",
      "4.0 0 [False False False False]\n",
      "4.0 1 [False False  True False]\n",
      "4.0 0 [False False False False]\n",
      "4.0 1 [False False  True False]\n",
      "4.0 1 [False False  True False]\n",
      "4.0 1 [False False  True False]\n",
      "5.0 2 [False False  True False  True]\n",
      "4.0 1 [False False  True False]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 1 [False False False False  True]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 1 [False False  True False False]\n",
      "5.0 1 [False False  True False False]\n",
      "5.0 2 [False  True False False  True]\n",
      "5.0 2 [False  True False False  True]\n",
      "5.0 1 [False False False False  True]\n",
      "5.0 0 [False False False False False]\n",
      "5.0 0 [False False False False False]\n",
      "5.0 2 [False  True False False  True]\n",
      "5.0 1 [False False False False  True]\n",
      "5.0 1 [False False False False  True]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 1 [False False  True False False]\n",
      "5.0 1 [False False  True False False]\n",
      "5.0 1 [False  True False False False]\n",
      "6.0 2 [ True False False  True False False]\n",
      "6.0 2 [ True False False  True False False]\n",
      "4.0 2 [False  True  True False]\n",
      "6.0 4 [ True False  True  True False  True]\n",
      "6.0 3 [ True False False  True False  True]\n",
      "6.0 1 [ True False False False False]\n",
      "6.0 1 [ True False False False False]\n",
      "6.0 1 [ True False False False False]\n",
      "6.0 1 [ True False False False False]\n",
      "6.0 1 [ True False False False False]\n",
      "6.0 1 [ True False False False False]\n",
      "4.0 1 [False False  True False]\n",
      "6.0 2 [False False False  True False  True]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 2 [ True  True False False False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 2 [False  True False False  True False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 0 [False False False False]\n",
      "6.0 1 [False False False  True False]\n",
      "6.0 0 [False False False False]\n",
      "6.0 0 [False False False False]\n",
      "6.0 0 [False False False False]\n",
      "6.0 2 [False False False  True  True]\n",
      "6.0 0 [False False False False]\n",
      "6.0 0 [False False False False]\n",
      "6.0 1 [False False False  True]\n",
      "6.0 1 [False False False  True]\n",
      "6.0 2 [False  True False False  True]\n",
      "6.0 2 [False False  True False  True]\n",
      "6.0 1 [False False False False  True]\n",
      "4.0 1 [False  True False False]\n",
      "4.0 0 [False False False False]\n",
      "4.0 0 [False False False False]\n",
      "4.0 1 [False False  True False]\n",
      "4.0 1 [False False  True False]\n",
      "4.0 1 [False False  True False]\n",
      "4.0 0 [False False False False]\n",
      "4.0 1 [False  True False False]\n",
      "4.0 1 [False  True False False]\n",
      "5.0 1 [False False False False  True]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 2 [False False  True False  True]\n",
      "4.0 1 [False False  True False]\n",
      "5.0 3 [False  True  True False  True]\n",
      "5.0 2 [False  True False False  True]\n",
      "5.0 2 [False  True False False  True]\n",
      "5.0 1 [False False False False  True]\n",
      "5.0 1 [False  True False False False]\n",
      "5.0 0 [False False False False False]\n",
      "5.0 2 [False  True False False  True]\n",
      "5.0 1 [False False False False  True]\n",
      "4.0 1 [False False  True False]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 1 [False False  True False False]\n",
      "5.0 1 [False  True False False False]\n",
      "6.0 2 [ True False False  True False False]\n",
      "6.0 2 [ True False False  True False False]\n",
      "4.0 2 [ True False  True False]\n",
      "6.0 3 [ True False False  True False  True]\n",
      "6.0 3 [ True False False  True False  True]\n",
      "6.0 2 [ True False False  True False False]\n",
      "6.0 1 [ True False False False False]\n",
      "6.0 1 [ True False False False False]\n",
      "6.0 1 [ True False False False False]\n",
      "6.0 1 [ True False False False False]\n",
      "6.0 1 [ True False False False False]\n",
      "6.0 1 [ True False False False False]\n",
      "6.0 1 [False  True False False False]\n",
      "6.0 3 [False  True False  True False  True]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 1 [ True False False False False False]\n",
      "6.0 2 [ True  True False False False False]\n",
      "6.0 2 [False  True False False  True False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 2 [False  True False False  True False]\n",
      "6.0 1 [False  True False False False]\n",
      "6.0 2 [False  True  True False False]\n",
      "6.0 0 [False False False False]\n",
      "6.0 1 [False False False  True]\n",
      "6.0 2 [False False False  True  True]\n",
      "6.0 1 [False False False  True]\n",
      "6.0 1 [False False False  True]\n",
      "6.0 1 [False False False  True]\n",
      "6.0 1 [False False False  True]\n",
      "6.0 2 [False  True False  True]\n",
      "6.0 3 [False  True  True False  True]\n",
      "4.0 0 [False False False False]\n",
      "4.0 1 [False False  True False]\n",
      "4.0 1 [False False  True False]\n",
      "4.0 1 [False False  True False]\n",
      "4.0 0 [False False False False]\n",
      "4.0 1 [False False  True False]\n",
      "5.0 2 [False False  True False  True]\n",
      "4.0 1 [False False  True False]\n",
      "4.0 0 [False False False False]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 2 [False False  True False  True]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "5.0 2 [False False  True False  True]\n",
      "5.0 2 [False False  True False  True]\n",
      "4.0 1 [False False  True False]\n",
      "5.0 2 [False  True False False  True]\n",
      "5.0 2 [False  True False False  True]\n",
      "5.0 2 [False  True False False  True]\n",
      "5.0 2 [False  True False False  True]\n",
      "5.0 0 [False False False False False]\n",
      "5.0 1 [False False False False  True]\n",
      "5.0 1 [False False False False  True]\n",
      "4.0 0 [False False False False]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 2 [False  True False False  True]\n",
      "5.0 1 [False  True False False False]\n",
      "4.0 2 [ True  True False False]\n",
      "4.0 3 [ True  True  True False]\n",
      "4.0 2 [False  True  True False]\n",
      "6.0 4 [ True False  True  True False  True]\n",
      "6.0 2 [ True False False  True False False]\n",
      "6.0 1 [ True False False False False]\n",
      "6.0 1 [ True False False False False]\n",
      "6.0 1 [ True False False False False]\n",
      "6.0 2 [ True False False False  True]\n",
      "6.0 1 [ True False False False False]\n",
      "6.0 1 [ True False False False False]\n",
      "6.0 1 [False  True False False False]\n",
      "6.0 3 [False  True False  True False  True]\n",
      "6.0 2 [False  True False False False  True]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 2 [ True  True False False False False]\n",
      "6.0 2 [False  True False False  True False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 2 [False  True False False  True False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 2 [False  True False False  True False]\n",
      "6.0 1 [False  True False False False]\n",
      "6.0 1 [False False False  True False]\n",
      "6.0 2 [False False False  True  True]\n",
      "6.0 2 [False False False  True  True]\n",
      "6.0 1 [False False False  True]\n",
      "6.0 1 [False False False  True]\n",
      "6.0 1 [False False False  True]\n",
      "6.0 1 [False False False  True]\n",
      "6.0 2 [False  True False False  True]\n",
      "6.0 1 [False False False False  True]\n",
      "6.0 2 [False  True False False  True]\n",
      "4.0 1 [False  True False False]\n",
      "4.0 0 [False False False False]\n",
      "4.0 1 [False False  True False]\n",
      "4.0 1 [False False  True False]\n",
      "4.0 1 [False False  True False]\n",
      "4.0 1 [False False  True False]\n",
      "4.0 1 [False False  True False]\n",
      "5.0 1 [False False False  True]\n",
      "4.0 0 [False False False False]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 3 [False  True  True False  True]\n",
      "4.0 1 [False False  True False]\n",
      "4.0 1 [False False  True False]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 2 [False  True False False  True]\n",
      "5.0 2 [False  True False False  True]\n",
      "5.0 2 [False  True False False  True]\n",
      "5.0 2 [False  True False False  True]\n",
      "5.0 0 [False False False False False]\n",
      "5.0 2 [False  True False False  True]\n",
      "4.0 0 [False False False False]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 1 [False False False False  True]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 1 [False False  True False False]\n",
      "4.0 1 [ True False False]\n",
      "4.0 2 [ True  True False False]\n",
      "4.0 2 [ True False  True False]\n",
      "6.0 3 [ True False False  True False  True]\n",
      "6.0 3 [ True False False  True False  True]\n",
      "6.0 2 [ True False False  True False False]\n",
      "6.0 1 [ True False False False False]\n",
      "6.0 2 [ True False False False  True]\n",
      "6.0 2 [ True False False False  True]\n",
      "6.0 1 [ True False False False False]\n",
      "6.0 1 [ True False False False False]\n",
      "6.0 1 [ True False False False False]\n",
      "6.0 0 [False False False False False]\n",
      "6.0 3 [False  True False  True False  True]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 2 [ True  True False False False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 2 [False  True False False  True False]\n",
      "6.0 2 [False  True False False  True False]\n",
      "6.0 1 [False False False  True False]\n",
      "6.0 1 [False False False  True False]\n",
      "6.0 1 [False False False  True]\n",
      "6.0 1 [False False False  True]\n",
      "6.0 1 [False False False  True False]\n",
      "6.0 1 [False False False  True]\n",
      "6.0 2 [False  True False False  True]\n",
      "6.0 2 [False  True False False  True]\n",
      "6.0 1 [False False False False  True]\n",
      "6.0 2 [False  True False False  True]\n",
      "4.0 0 [False False False False]\n",
      "5.0 2 [False False  True False  True]\n",
      "4.0 1 [False False  True False]\n",
      "4.0 0 [False False False False]\n",
      "4.0 1 [False False  True False]\n",
      "5.0 2 [False False  True False  True]\n",
      "4.0 1 [False False  True False]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 2 [False False  True False  True]\n",
      "4.0 1 [False False  True False]\n",
      "5.0 2 [False False  True False  True]\n",
      "4.0 0 [False False False False]\n",
      "4.0 1 [False  True False False]\n",
      "4.0 1 [False  True False False]\n",
      "5.0 2 [False  True False False  True]\n",
      "5.0 2 [False  True False False  True]\n",
      "5.0 1 [False False False False  True]\n",
      "5.0 1 [False False False False  True]\n",
      "5.0 0 [False False False False False]\n",
      "4.0 0 [False False False False]\n",
      "5.0 1 [False False False False  True]\n",
      "5.0 0 [False False False False False]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 2 [False False  True False  True]\n",
      "6.0 2 [ True False False  True False False]\n",
      "4.0 2 [ True  True False False]\n",
      "4.0 1 [ True False False]\n",
      "4.0 2 [ True  True False]\n",
      "6.0 5 [ True  True  True  True False  True]\n",
      "6.0 2 [ True False False  True False False]\n",
      "6.0 1 [ True False False False False]\n",
      "6.0 1 [ True False False False False]\n",
      "6.0 2 [ True False False False  True]\n",
      "6.0 1 [ True False False False False]\n",
      "6.0 1 [ True False False False False]\n",
      "6.0 1 [ True False False False False]\n",
      "6.0 1 [False  True False False False]\n",
      "6.0 1 [False  True False False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 2 [ True  True False False False False]\n",
      "6.0 2 [ True  True False False False False]\n",
      "6.0 2 [ True  True False False False False]\n",
      "6.0 2 [False  True False False  True False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 2 [ True  True False False False False]\n",
      "6.0 2 [False  True False False  True False]\n",
      "6.0 2 [False  True False False  True False]\n",
      "6.0 2 [False  True False False  True False]\n",
      "6.0 2 [False  True False False  True False]\n",
      "6.0 1 [False False False  True False]\n",
      "6.0 1 [False False False  True False]\n",
      "6.0 2 [False False False  True  True]\n",
      "6.0 2 [False False False  True  True]\n",
      "6.0 1 [False False False  True]\n",
      "6.0 1 [False False False  True]\n",
      "6.0 1 [False False False  True]\n",
      "6.0 3 [False  True  True False  True]\n",
      "6.0 1 [False False False  True]\n",
      "6.0 1 [False False False False  True]\n",
      "6.0 3 [False  True False False  True  True]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 2 [False False  True False  True]\n",
      "4.0 0 [False False False False]\n",
      "4.0 1 [False  True False False]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 1 [False False False False  True]\n",
      "4.0 0 [False False False False]\n",
      "4.0 1 [False  True False False]\n",
      "5.0 2 [False  True False False  True]\n",
      "5.0 3 [False  True  True False  True]\n",
      "5.0 1 [False False False False  True]\n",
      "5.0 1 [False False False False  True]\n",
      "5.0 1 [False False  True False False]\n",
      "5.0 1 [False False  True False False]\n",
      "5.0 1 [False False  True False False]\n",
      "5.0 1 [False False  True False False]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 1 [False False  True False False]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 1 [False False  True False False]\n",
      "4.0 2 [False  True  True False]\n",
      "4.0 2 [ True  True False]\n",
      "6.0 5 [ True  True  True  True False  True]\n",
      "5.0 3 [ True False  True  True False]\n",
      "6.0 2 [ True False False  True False False]\n",
      "6.0 1 [ True False False False False]\n",
      "6.0 2 [ True False False False  True]\n",
      "6.0 1 [ True False False False False]\n",
      "6.0 1 [ True False False False False]\n",
      "6.0 1 [ True False False False False]\n",
      "6.0 1 [ True False False False False]\n",
      "6.0 2 [False  True False  True False False]\n",
      "6.0 2 [False  True False  True False False]\n",
      "6.0 2 [ True  True False False False False]\n",
      "6.0 2 [ True  True False False False False]\n",
      "6.0 1 [ True False False False False False]\n",
      "6.0 2 [False  True False False  True False]\n",
      "6.0 2 [False  True False False  True False]\n",
      "6.0 2 [ True  True False False False False]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "6.0 1 [False  True False False False False]\n",
      "6.0 2 [ True  True False False False False]\n",
      "6.0 2 [False  True False False  True False]\n",
      "6.0 2 [False  True False False  True False]\n",
      "6.0 2 [False  True False False  True False]\n",
      "6.0 2 [False  True False False  True False]\n",
      "6.0 1 [False False False  True False]\n",
      "6.0 2 [False False False  True  True]\n",
      "6.0 1 [False False False  True False]\n",
      "6.0 1 [False False False  True]\n",
      "6.0 1 [False False False  True]\n",
      "6.0 1 [False False False  True]\n",
      "6.0 1 [False False False  True]\n",
      "6.0 1 [False False False  True]\n",
      "6.0 2 [False False  True False  True]\n",
      "6.0 2 [False False  True False  True]\n",
      "6.0 3 [False  True False False  True  True]\n",
      "5.0 1 [False False False False  True]\n",
      "4.0 0 [False False False False]\n",
      "4.0 0 [False False False False]\n",
      "4.0 2 [False  True  True False]\n",
      "4.0 1 [False False  True False]\n",
      "5.0 3 [False  True  True False  True]\n",
      "5.0 1 [False False  True False False]\n",
      "4.0 1 [False False  True False]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 2 [False False  True False  True]\n",
      "4.0 1 [False False  True False]\n",
      "4.0 1 [False False  True False]\n",
      "4.0 0 [False False False False]\n",
      "5.0 2 [False  True False False  True]\n",
      "5.0 2 [False  True False False  True]\n",
      "5.0 1 [False False False False  True]\n",
      "5.0 1 [False False False False  True]\n",
      "5.0 1 [False False  True False False]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 0 [False False False False False]\n",
      "5.0 1 [False False  True False False]\n",
      "5.0 1 [False False  True False False]\n",
      "5.0 1 [False False  True False False]\n",
      "5.0 1 [False False  True False False]\n",
      "5.0 1 [False False  True False False]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 2 [False False  True  True False]\n",
      "4.0 1 [False  True False False]\n",
      "4.0 2 [ True  True False]\n",
      "4.0 1 [False  True False]\n",
      "5.0 3 [ True False  True  True False]\n",
      "6.0 3 [ True False False  True False  True]\n",
      "6.0 2 [ True False False  True False False]\n",
      "6.0 2 [ True False False False  True]\n",
      "6.0 1 [ True False False False False]\n",
      "6.0 1 [ True False False False False]\n",
      "6.0 1 [ True False False False False]\n",
      "6.0 1 [ True False False False False]\n",
      "6.0 0 [False False False False False]\n",
      "6.0 2 [ True  True False False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 2 [ True  True False False False False]\n",
      "6.0 1 [ True False False False False False]\n",
      "6.0 3 [ True  True  True False False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 2 [False  True False False  True False]\n",
      "6.0 2 [False  True False False  True False]\n",
      "6.0 2 [False  True False False  True False]\n",
      "6.0 2 [False  True False False  True False]\n",
      "6.0 1 [False False False  True False]\n",
      "6.0 1 [False False False  True False]\n",
      "6.0 1 [False False False  True False]\n",
      "6.0 1 [False False False  True]\n",
      "6.0 1 [False False False  True]\n",
      "6.0 1 [False False False  True]\n",
      "6.0 2 [False  True False  True]\n",
      "6.0 1 [False False False  True]\n",
      "6.0 1 [False False False  True]\n",
      "6.0 1 [False False False False  True]\n",
      "6.0 2 [False  True False False  True]\n",
      "5.0 3 [False  True  True False  True]\n",
      "5.0 2 [False False  True False  True]\n",
      "4.0 0 [False False False False]\n",
      "4.0 0 [False False False False]\n",
      "4.0 2 [False  True  True False]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 1 [False False False False  True]\n",
      "5.0 1 [False False  True False False]\n",
      "4.0 1 [False False  True False]\n",
      "4.0 1 [False False  True False]\n",
      "4.0 1 [False False  True False]\n",
      "5.0 2 [False False  True False  True]\n",
      "4.0 0 [False False False False]\n",
      "4.0 0 [False False False False]\n",
      "4.0 1 [False  True False False]\n",
      "5.0 2 [False  True False False  True]\n",
      "5.0 2 [False  True False False  True]\n",
      "5.0 2 [False  True False False  True]\n",
      "5.0 2 [False  True  True False False]\n",
      "5.0 1 [False False  True False False]\n",
      "5.0 2 [False  True False False  True]\n",
      "5.0 0 [False False False False False]\n",
      "5.0 1 [False False  True False False]\n",
      "5.0 1 [False False  True False False]\n",
      "5.0 1 [False False  True False False]\n",
      "5.0 1 [False False  True False False]\n",
      "5.0 1 [False False  True False False]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 2 [False False  True False  True]\n",
      "4.0 1 [False  True False]\n",
      "4.0 0 [False False False]\n",
      "4.0 0 [False False False]\n",
      "6.0 3 [ True False False  True False  True]\n",
      "6.0 2 [ True False False  True False False]\n",
      "6.0 3 [ True False False  True False  True]\n",
      "6.0 2 [ True False False False  True]\n",
      "6.0 1 [ True False False False False]\n",
      "6.0 1 [ True False False False False]\n",
      "6.0 1 [ True False False False False]\n",
      "6.0 0 [False False False False False]\n",
      "6.0 1 [False  True False False False]\n",
      "6.0 1 [False  True False False False]\n",
      "6.0 2 [False  True False  True False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 2 [ True  True False False False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 2 [False  True False False  True False]\n",
      "6.0 2 [ True  True False False False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 2 [False  True False False  True False]\n",
      "6.0 2 [False  True False False  True False]\n",
      "6.0 2 [False  True False False  True False]\n",
      "6.0 2 [False  True False False  True False]\n",
      "6.0 1 [False False False  True False]\n",
      "6.0 1 [False False False  True False]\n",
      "6.0 0 [False False False False]\n",
      "6.0 1 [False False False  True]\n",
      "6.0 1 [False False False  True]\n",
      "6.0 1 [False False False  True]\n",
      "6.0 1 [False False False  True]\n",
      "6.0 1 [False False False  True]\n",
      "6.0 1 [False False False False  True]\n",
      "6.0 2 [False False  True False  True]\n",
      "6.0 3 [False  True False False  True  True]\n",
      "5.0 1 [False False False False  True]\n",
      "5.0 2 [False False  True False  True]\n",
      "4.0 1 [False  True False False]\n",
      "4.0 2 [False  True  True False]\n",
      "4.0 0 [False False False False]\n",
      "4.0 0 [False False False]\n",
      "5.0 2 [False False  True False  True]\n",
      "4.0 1 [False False  True False]\n",
      "4.0 0 [False False False False]\n",
      "4.0 1 [False False  True False]\n",
      "5.0 2 [False False  True False  True]\n",
      "4.0 0 [False False False False]\n",
      "4.0 0 [False False False False]\n",
      "4.0 0 [False False False False]\n",
      "5.0 3 [False  True  True False  True]\n",
      "5.0 3 [False  True  True False  True]\n",
      "5.0 2 [False  True False False  True]\n",
      "5.0 3 [False  True  True False  True]\n",
      "5.0 2 [False  True  True False False]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 1 [False False False False  True]\n",
      "5.0 1 [False False  True False False]\n",
      "5.0 1 [False False  True False False]\n",
      "5.0 1 [False False  True False False]\n",
      "5.0 1 [False False  True False False]\n",
      "5.0 1 [False False  True False False]\n",
      "5.0 2 [False  True  True False False]\n",
      "5.0 3 [False  True  True False  True]\n",
      "4.0 1 [False  True False False]\n",
      "4.0 0 [False False False]\n",
      "4.0 1 [False  True False]\n",
      "6.0 3 [ True False False  True False  True]\n",
      "6.0 3 [ True False False  True False  True]\n",
      "4.0 0 [False False False False]\n",
      "6.0 2 [ True False False False  True]\n",
      "6.0 1 [ True False False False False]\n",
      "6.0 1 [ True False False False False]\n",
      "6.0 1 [ True False False False False]\n",
      "6.0 0 [False False False False False]\n",
      "6.0 0 [False False False False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 1 [False  True False False False]\n",
      "6.0 2 [ True  True False False False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 2 [False  True False False  True False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 2 [False  True False False  True False]\n",
      "6.0 2 [False  True False False  True False]\n",
      "6.0 2 [False  True False False  True False]\n",
      "6.0 1 [False False False  True False]\n",
      "6.0 1 [False False False  True False]\n",
      "6.0 1 [False False False  True False]\n",
      "6.0 1 [False False False  True]\n",
      "6.0 1 [False False False  True]\n",
      "6.0 1 [False False False  True]\n",
      "6.0 1 [False False False  True]\n",
      "6.0 1 [False False False  True]\n",
      "6.0 1 [False False False False  True]\n",
      "6.0 1 [False False False False  True]\n",
      "6.0 1 [False False False  True]\n",
      "5.0 2 [False  True False False  True]\n",
      "5.0 2 [False False  True False  True]\n",
      "4.0 1 [False  True False False]\n",
      "5.0 1 [False False False False  True]\n",
      "4.0 2 [False  True  True False]\n",
      "4.0 1 [False  True False False]\n",
      "4.0 1 [False  True False False]\n",
      "4.0 2 [False  True  True False]\n",
      "4.0 0 [False False False False]\n",
      "4.0 0 [False False False False]\n",
      "4.0 1 [False False  True False]\n",
      "4.0 0 [False False False False]\n",
      "4.0 0 [False False False False]\n",
      "4.0 1 [False  True False False]\n",
      "4.0 0 [False False False False]\n",
      "5.0 3 [False  True  True False  True]\n",
      "5.0 3 [False  True  True False  True]\n",
      "5.0 3 [False  True  True False  True]\n",
      "5.0 2 [False  True  True False False]\n",
      "5.0 1 [False False  True False False]\n",
      "5.0 2 [False  True False False  True]\n",
      "5.0 1 [False False  True False False]\n",
      "5.0 1 [False False  True False False]\n",
      "5.0 1 [False False  True False False]\n",
      "5.0 1 [False False  True False False]\n",
      "5.0 1 [False False  True False False]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 2 [False  True  True False False]\n",
      "4.0 1 [False  True False]\n",
      "4.0 1 [False  True False False]\n",
      "4.0 1 [False False  True]\n",
      "6.0 3 [ True False False  True False  True]\n",
      "6.0 3 [ True False False  True False  True]\n",
      "6.0 3 [ True False False  True False  True]\n",
      "4.0 0 [False False False False]\n",
      "6.0 2 [ True False False False  True]\n",
      "6.0 1 [ True False False False False]\n",
      "6.0 1 [ True False False False False]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "6.0 1 [ True False False False False]\n",
      "6.0 0 [False False False False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 2 [ True  True False False False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 2 [False  True False False  True False]\n",
      "6.0 2 [False  True False False  True False]\n",
      "6.0 2 [False  True False False  True False]\n",
      "6.0 3 [False  True False False  True  True]\n",
      "6.0 1 [False False False  True False]\n",
      "6.0 1 [False False False  True False]\n",
      "6.0 0 [False False False False]\n",
      "6.0 1 [False False False  True]\n",
      "6.0 1 [False False False  True]\n",
      "6.0 1 [False False False  True]\n",
      "6.0 1 [False False False  True]\n",
      "6.0 1 [False False False False  True]\n",
      "6.0 1 [False False False False  True]\n",
      "6.0 2 [False False  True False  True]\n",
      "5.0 2 [False  True False False  True]\n",
      "5.0 2 [False False  True False  True]\n",
      "4.0 1 [False  True False False]\n",
      "5.0 1 [False False False False  True]\n",
      "5.0 1 [False False False False  True]\n",
      "4.0 1 [False  True False False]\n",
      "4.0 1 [False  True False False]\n",
      "4.0 1 [False False  True False]\n",
      "4.0 0 [False False False False]\n",
      "4.0 0 [False False False False]\n",
      "5.0 2 [False False  True False  True]\n",
      "4.0 0 [False False False False]\n",
      "4.0 0 [False False False False]\n",
      "4.0 0 [False False False False]\n",
      "5.0 3 [False  True  True False  True]\n",
      "5.0 3 [False  True  True False  True]\n",
      "5.0 3 [False  True  True False  True]\n",
      "5.0 2 [False  True  True False False]\n",
      "5.0 1 [False False  True False False]\n",
      "5.0 2 [False  True False False  True]\n",
      "5.0 1 [False False  True False False]\n",
      "5.0 1 [False False  True False False]\n",
      "5.0 1 [False False  True False False]\n",
      "5.0 1 [False False  True False False]\n",
      "5.0 1 [False False  True False False]\n",
      "5.0 2 [False  True  True False False]\n",
      "4.0 1 [False  True False]\n",
      "4.0 1 [False  True False False]\n",
      "4.0 1 [False False  True]\n",
      "4.0 1 [False  True False False]\n",
      "6.0 3 [ True False False  True False  True]\n",
      "6.0 3 [ True False False  True False  True]\n",
      "4.0 0 [False False False False]\n",
      "6.0 1 [ True False False False False]\n",
      "6.0 1 [ True False False False False]\n",
      "6.0 1 [ True False False False False]\n",
      "6.0 0 [False False False False False]\n",
      "6.0 1 [False  True False False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 1 [False  True False False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 2 [ True  True False False False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 2 [False  True False False  True False]\n",
      "6.0 2 [False  True False False  True False]\n",
      "6.0 1 [False False False  True False]\n",
      "6.0 1 [False False False  True False]\n",
      "6.0 2 [False False False  True  True]\n",
      "6.0 1 [False False False  True]\n",
      "6.0 1 [False False False  True]\n",
      "6.0 1 [False False False  True]\n",
      "6.0 1 [False False False  True]\n",
      "6.0 1 [False False False False  True]\n",
      "6.0 1 [False False False False  True]\n",
      "6.0 1 [False False False False  True]\n",
      "5.0 1 [False False False  True]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 2 [False  True False False  True]\n",
      "5.0 1 [False False False False  True]\n",
      "5.0 2 [False False  True False  True]\n",
      "4.0 0 [False False False]\n",
      "4.0 1 [False  True False False]\n",
      "4.0 0 [False False False False]\n",
      "4.0 1 [False  True False False]\n",
      "5.0 1 [False False False False  True]\n",
      "4.0 0 [False False False False]\n",
      "4.0 0 [False False False False]\n",
      "4.0 1 [False  True False False]\n",
      "5.0 3 [False  True  True False  True]\n",
      "4.0 2 [False  True  True False]\n",
      "5.0 2 [False  True  True False False]\n",
      "5.0 2 [False  True  True False False]\n",
      "5.0 2 [False  True False False  True]\n",
      "5.0 1 [False False False False  True]\n",
      "5.0 1 [False False  True False False]\n",
      "5.0 1 [False False  True False False]\n",
      "4.0 1 [False False  True False]\n",
      "5.0 1 [False False  True False False]\n",
      "5.0 2 [False  True  True False False]\n",
      "5.0 3 [False  True  True False  True]\n",
      "4.0 0 [False False False]\n",
      "6.0 3 [ True False False  True False  True]\n",
      "4.0 1 [False  True False False]\n",
      "6.0 3 [ True False False  True False  True]\n",
      "6.0 3 [ True False False  True False  True]\n",
      "6.0 2 [ True False  True False False]\n",
      "4.0 0 [False False False False]\n",
      "6.0 1 [ True False False False False]\n",
      "6.0 1 [ True False False False False]\n",
      "6.0 0 [False False False False False False]\n",
      "6.0 1 [False  True False False False]\n",
      "6.0 1 [False  True False False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 2 [ True  True False False False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 2 [False  True False False  True False]\n",
      "6.0 2 [False  True False False  True False]\n",
      "6.0 1 [False False False  True False]\n",
      "6.0 1 [False False False  True False]\n",
      "6.0 2 [False False False  True  True]\n",
      "6.0 1 [False False False  True]\n",
      "6.0 1 [False False False  True]\n",
      "6.0 1 [False False False  True]\n",
      "6.0 1 [False False False False  True]\n",
      "6.0 1 [False False False False  True]\n",
      "4.0 0 [False False False]\n",
      "4.0 0 [False False False]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 1 [False False False False  True]\n",
      "5.0 2 [False  True False False  True]\n",
      "5.0 2 [False False  True False  True]\n",
      "4.0 1 [False  True False False]\n",
      "4.0 0 [False False False False]\n",
      "4.0 0 [False False False False]\n",
      "4.0 0 [False False False False]\n",
      "4.0 0 [False False False False]\n",
      "4.0 0 [False False False False]\n",
      "4.0 1 [False False  True False]\n",
      "4.0 2 [False  True  True False]\n",
      "4.0 2 [False  True  True False]\n",
      "5.0 2 [False  True  True False False]\n",
      "5.0 1 [False False  True False False]\n",
      "5.0 1 [False False False False  True]\n",
      "5.0 0 [False False False False False]\n",
      "5.0 1 [False False  True False False]\n",
      "4.0 1 [False False  True False]\n",
      "5.0 2 [False False  True  True False]\n",
      "5.0 2 [False  True  True False False]\n",
      "5.0 2 [False False  True False  True]\n",
      "6.0 3 [ True False False  True False  True]\n",
      "6.0 3 [ True False False  True False  True]\n",
      "4.0 1 [False  True False False]\n",
      "4.0 1 [False  True False False]\n",
      "6.0 1 [ True False False False False]\n",
      "6.0 2 [ True False False False  True]\n",
      "6.0 2 [ True False  True False False]\n",
      "6.0 1 [ True False False False False]\n",
      "6.0 0 [False False False False False]\n",
      "6.0 1 [False  True False False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 2 [False  True False  True False False]\n",
      "6.0 2 [ True  True False False False False]\n",
      "6.0 2 [ True  True False False False False]\n",
      "6.0 2 [ True  True False False False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 1 [False False False  True False]\n",
      "6.0 1 [False False False  True False]\n",
      "6.0 1 [False False False  True False]\n",
      "6.0 1 [False False False  True]\n",
      "6.0 0 [False False False False]\n",
      "6.0 1 [False False False  True]\n",
      "6.0 1 [False False False  True]\n",
      "6.0 1 [False False False False  True]\n",
      "6.0 1 [False False False False  True]\n",
      "6.0 1 [False False False  True]\n",
      "4.0 0 [False False False False]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 1 [False False False False  True]\n",
      "5.0 3 [False  True  True False  True]\n",
      "4.0 1 [False  True False False]\n",
      "4.0 1 [False  True False False]\n",
      "5.0 2 [False  True False  True]\n",
      "5.0 0 [False False False False]\n",
      "5.0 1 [False False False False  True]\n",
      "4.0 0 [False False False False]\n",
      "4.0 1 [False False  True False]\n",
      "5.0 3 [False  True  True False  True]\n",
      "5.0 3 [False  True  True False  True]\n",
      "5.0 2 [False  True  True False False]\n",
      "5.0 1 [False False  True False False]\n",
      "5.0 1 [False False False False  True]\n",
      "5.0 3 [False  True  True False  True]\n",
      "4.0 1 [False  True False False]\n",
      "5.0 2 [False False  True  True False]\n",
      "5.0 2 [False  True  True False False]\n",
      "5.0 2 [False  True False  True]\n",
      "5.0 2 [False False  True False  True]\n",
      "6.0 3 [ True False False  True False  True]\n",
      "6.0 3 [ True False False  True False  True]\n",
      "4.0 0 [False False False False]\n",
      "6.0 1 [ True False False False False]\n",
      "6.0 2 [ True False False False  True]\n",
      "6.0 1 [ True False False False False]\n",
      "6.0 1 [ True False False False False]\n",
      "6.0 0 [False False False False False]\n",
      "6.0 1 [False  True False False False]\n",
      "6.0 1 [False  True False False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 2 [ True  True False False False False]\n",
      "6.0 2 [ True  True False False False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 1 [False False False  True False]\n",
      "6.0 1 [False False False  True False]\n",
      "6.0 1 [False False False  True False]\n",
      "6.0 1 [False False False  True]\n",
      "6.0 1 [False False False  True]\n",
      "6.0 1 [False False False  True]\n",
      "6.0 1 [False False False  True]\n",
      "6.0 1 [False False False False  True]\n",
      "6.0 1 [False False False  True]\n",
      "4.0 1 [False False  True False]\n",
      "5.0 2 [False False  True False  True]\n",
      "5.0 1 [False False False False  True]\n",
      "5.0 2 [False  True False False  True]\n",
      "4.0 0 [False False False]\n",
      "4.0 0 [False False False False]\n",
      "4.0 1 [False False  True False]\n",
      "4.0 1 [False  True False]\n",
      "4.0 0 [False False False False]\n",
      "4.0 1 [False False  True False]\n",
      "5.0 3 [False  True  True False  True]\n",
      "5.0 3 [False  True  True False  True]\n",
      "5.0 2 [False  True  True False False]\n",
      "5.0 1 [False False  True False False]\n",
      "5.0 2 [False  True  True False False]\n",
      "5.0 1 [False False False  True]\n",
      "4.0 1 [False  True False False]\n",
      "5.0 1 [False False  True False False]\n",
      "5.0 1 [False False  True False False]\n",
      "5.0 2 [False False  True False  True]\n",
      "6.0 3 [ True False False  True False  True]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "6.0 3 [ True False False  True False  True]\n",
      "6.0 3 [ True False False  True False  True]\n",
      "6.0 1 [ True False False False False]\n",
      "6.0 1 [ True False False False False]\n",
      "6.0 1 [ True False False False False]\n",
      "6.0 0 [False False False False False]\n",
      "6.0 1 [False  True False False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 2 [False  True False  True False False]\n",
      "6.0 3 [ True  True False  True False False]\n",
      "6.0 2 [ True  True False False False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 1 [False False False  True False]\n",
      "6.0 1 [False False False  True False]\n",
      "6.0 1 [False False False  True]\n",
      "6.0 0 [False False False False]\n",
      "6.0 1 [False False False  True]\n",
      "6.0 1 [False False False False  True]\n",
      "6.0 1 [False False False False  True]\n",
      "6.0 1 [False False False  True]\n",
      "4.0 1 [False False  True False]\n",
      "5.0 2 [False  True False False  True]\n",
      "5.0 2 [False False  True False  True]\n",
      "4.0 0 [False False False]\n",
      "4.0 1 [False  True False False]\n",
      "4.0 0 [False False False False]\n",
      "5.0 2 [False False  True False  True]\n",
      "4.0 0 [False False]\n",
      "4.0 0 [False False]\n",
      "4.0 1 [False False  True False]\n",
      "5.0 2 [False  True  True False False]\n",
      "5.0 2 [False  True  True False False]\n",
      "5.0 3 [False False  True  True  True]\n",
      "5.0 2 [False  True  True False False]\n",
      "4.0 0 [False False False]\n",
      "4.0 1 [False  True False False]\n",
      "5.0 1 [False False  True False False]\n",
      "6.0 3 [ True False False  True False  True]\n",
      "6.0 3 [ True False False  True False  True]\n",
      "6.0 3 [ True False False  True False  True]\n",
      "6.0 3 [ True False False  True False  True]\n",
      "6.0 1 [ True False False False False]\n",
      "6.0 1 [ True False False False False]\n",
      "6.0 1 [ True False False False False]\n",
      "6.0 0 [False False False False False]\n",
      "6.0 1 [False  True False False False False]\n",
      "6.0 2 [False  True False  True False False]\n",
      "6.0 3 [ True  True False  True False False]\n",
      "6.0 2 [ True  True False False False False]\n",
      "6.0 3 [ True  True False False  True False]\n",
      "6.0 3 [ True  True False  True False False]\n",
      "6.0 2 [False  True False False  True False]\n",
      "6.0 1 [False False False  True False]\n",
      "6.0 3 [False  True False  True  True]\n",
      "6.0 0 [False False False False]\n",
      "6.0 1 [False  True False False]\n",
      "4.0 0 [False False False]\n",
      "4.0 0 [False False False]\n",
      "4.0 0 [False False False]\n",
      "5.0 1 [False False False False  True]\n",
      "4.0 1 [False  True False]\n",
      "4.0 1 [False  True False]\n",
      "4.0 0 [False False False]\n",
      "4.0 1 [False False  True False]\n",
      "4.0 2 [False  True  True False]\n",
      "4.0 2 [False  True  True False]\n",
      "4.0 0 [False False]\n",
      "4.0 2 [False  True  True False]\n",
      "5.0 2 [False  True  True False False]\n",
      "5.0 3 [False  True  True  True False]\n",
      "5.0 4 [False  True  True  True  True]\n",
      "4.0 2 [False  True  True False]\n",
      "4.0 1 [False  True False False]\n",
      "4.0 2 [False  True  True False]\n",
      "4.0 2 [False  True  True False]\n",
      "6.0 3 [ True False False  True False  True]\n",
      "4.0 2 [False  True  True False]\n",
      "4.0 2 [False  True  True False]\n",
      "4.0 3 [False  True  True  True]\n",
      "6.0 2 [ True False False False  True]\n",
      "6.0 3 [ True  True  True False False]\n",
      "6.0 2 [ True  True False False False]\n",
      "6.0 1 [False  True False False False]\n",
      "6.0 3 [ True  True  True False False]\n",
      "6.0 3 [ True False  True  True False]\n",
      "6.0 5 [ True  True False  True  True  True]\n",
      "6.0 3 [ True  True False  True False False]\n",
      "6.0 3 [ True  True False  True False]\n",
      "6.0 2 [False False False  True  True]\n",
      "6.0 2 [ True False False  True]\n",
      "4.0 1 [False  True False]\n",
      "4.0 1 [False  True False]\n",
      "4.0 0 [False False False]\n",
      "4.0 0 [False False False]\n",
      "5.0 2 [False False  True False  True]\n",
      "4.0 1 [False  True False]\n",
      "4.0 1 [False  True False]\n",
      "4.0 0 [False False False]\n",
      "4.0 1 [False  True False]\n",
      "4.0 2 [False  True  True False]\n",
      "4.0 1 [False  True False]\n",
      "5.0 1 [False False False  True]\n",
      "4.0 2 [False  True  True False]\n",
      "5.0 3 [False  True  True  True False]\n",
      "4.0 2 [False  True  True False]\n",
      "5.0 2 [False  True  True False False]\n",
      "4.0 2 [False  True  True False]\n",
      "4.0 2 [False  True  True False]\n",
      "4.0 1 [False  True False False]\n",
      "4.0 2 [False  True  True False]\n",
      "4.0 2 [False  True  True False]\n",
      "4.0 1 [False  True False]\n",
      "4.0 2 [False  True  True False]\n",
      "6.0 3 [ True  True  True False False]\n",
      "6.0 3 [ True  True False  True False]\n",
      "6.0 3 [ True  True  True False False]\n",
      "6.0 4 [ True  True  True  True False]\n",
      "6.0 4 [ True  True False  True  True False]\n",
      "6.0 5 [ True  True  True  True  True False]\n",
      "6.0 3 [ True False  True  True False]\n",
      "6.0 3 [False False  True  True  True]\n",
      "6.0 1 [ True False False False]\n",
      "4.0 1 [False  True False]\n",
      "4.0 0 [False False False]\n",
      "4.0 1 [False  True False]\n",
      "5.0 2 [False False  True False  True]\n",
      "4.0 1 [False  True False]\n",
      "4.0 1 [False  True False]\n",
      "4.0 1 [False False  True]\n",
      "4.0 2 [False  True  True False]\n",
      "4.0 1 [False  True False]\n",
      "5.0 0 [False False False False]\n",
      "4.0 2 [False  True  True False]\n",
      "4.0 2 [False  True  True False]\n",
      "4.0 2 [False  True  True False]\n",
      "5.0 1 [False False  True False False]\n",
      "4.0 2 [False  True  True False]\n",
      "4.0 2 [False  True  True False]\n",
      "4.0 2 [False  True  True False]\n",
      "4.0 2 [False  True  True False]\n",
      "4.0 2 [False  True  True False]\n",
      "6.0 3 [ True  True False False  True]\n",
      "6.0 2 [ True  True False False False]\n",
      "6.0 3 [ True  True  True False False]\n",
      "6.0 4 [ True  True False  True  True False]\n",
      "6.0 4 [ True  True False  True  True False]\n",
      "6.0 3 [ True False  True  True False]\n",
      "6.0 1 [False False False  True False]\n",
      "6.0 1 [False False False  True]\n",
      "5.0 2 [False  True False  True]\n",
      "4.0 0 [False False False]\n",
      "4.0 1 [False  True False]\n",
      "4.0 2 [False  True  True]\n",
      "4.0 1 [False  True False]\n",
      "4.0 0 [False False False]\n",
      "4.0 2 [False  True  True False]\n",
      "4.0 2 [False  True  True False]\n",
      "5.0 2 [False  True  True False False]\n",
      "4.0 2 [False  True  True False]\n",
      "4.0 2 [False  True  True False]\n",
      "4.0 1 [False False  True False]\n",
      "4.0 2 [False  True  True False]\n",
      "4.0 1 [False  True False]\n",
      "4.0 2 [False  True  True False]\n",
      "4.0 2 [False  True  True False]\n",
      "4.0 2 [False  True  True False]\n",
      "6.0 4 [ True  True  True  True False]\n",
      "6.0 2 [ True  True False False False]\n",
      "6.0 3 [ True  True False False  True]\n",
      "6.0 5 [ True  True  True  True  True False]\n",
      "6.0 5 [ True  True False  True  True  True]\n",
      "6.0 2 [False False False  True  True]\n",
      "4.0 1 [False  True False]\n",
      "4.0 1 [False  True False]\n",
      "4.0 1 [False  True False]\n",
      "4.0 2 [False  True  True]\n",
      "4.0 1 [False  True False]\n",
      "4.0 2 [False  True  True False]\n",
      "4.0 2 [False  True  True False]\n",
      "5.0 1 [False  True False False]\n",
      "5.0 1 [False  True False False]\n",
      "5.0 1 [False False  True False False]\n",
      "4.0 2 [False  True  True False]\n",
      "4.0 1 [False  True False]\n",
      "4.0 1 [False  True False]\n",
      "4.0 2 [False  True  True False]\n",
      "6.0 3 [ True  True False False  True]\n",
      "6.0 2 [ True  True False False False]\n",
      "6.0 4 [ True  True  True False  True]\n",
      "6.0 4 [ True  True  True  True False]\n",
      "6.0 1 [False False False  True False]\n",
      "6.0 1 [False False False  True]\n",
      "4.0 1 [False  True False]\n",
      "4.0 1 [False False  True False]\n",
      "4.0 2 [False  True  True]\n",
      "4.0 1 [False  True False]\n",
      "4.0 1 [False  True False]\n",
      "5.0 1 [False  True False False]\n",
      "5.0 1 [False  True False False]\n",
      "5.0 2 [False False  True False  True]\n",
      "4.0 2 [False  True  True False]\n",
      "4.0 2 [False  True  True False]\n",
      "6.0 2 [ True False False False  True]\n",
      "6.0 4 [ True  True  True False  True]\n",
      "6.0 4 [ True  True  True False  True]\n",
      "6.0 3 [ True  True False  True False]\n",
      "6.0 1 [False False False  True False]\n",
      "4.0 1 [False  True False]\n",
      "4.0 2 [False  True  True]\n",
      "4.0 1 [False False  True]\n",
      "4.0 1 [False  True False]\n",
      "5.0 1 [False  True False False]\n",
      "5.0 2 [False  True  True False False]\n",
      "4.0 2 [False  True  True False]\n",
      "4.0 2 [False  True  True False]\n",
      "6.0 2 [ True False False False  True]\n",
      "6.0 2 [ True  True False False False]\n",
      "6.0 3 [False  True False  True  True False]\n",
      "6.0 2 [False  True False  True]\n",
      "4.0 2 [False  True  True]\n",
      "4.0 1 [False  True False False]\n",
      "5.0 1 [False  True False False]\n",
      "4.0 2 [False  True  True False]\n",
      "4.0 2 [False  True  True False]\n",
      "6.0 2 [ True  True False False False]\n",
      "6.0 3 [ True False  True  True False]\n",
      "5.0 3 [False  True  True  True]\n",
      "4.0 2 [False  True  True False]\n",
      "6.0 4 [ True  True  True False  True]\n",
      "Completed metric generation.\n",
      "Running reduce methods.\n",
      "Running summary statistics.\n",
      "Completed.\n",
      "Plotting figures with \"night<700\" constraint now.\n",
      "monopole: 0.237125  dipole: lon: -140.905, lat: -36.1319, amp: 0.099639\n",
      "Plotting complete.\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f8eb0a55160>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f8ec41a9d30>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f8eb0a55d30>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# run the metric\n",
    "asciifile = 'TDEfaintfast_z0.1.dat'\n",
    "transmetric = TDEsMetricTest(asciifile, metricName='TDEsMetricTest', mjdCol='observationStartMJD',\n",
    "                 m5Col='fiveSigmaDepth', filterCol='filter',\n",
    "                 detectSNR={'u': 5, 'g': 5, 'r': 5, 'i': 5, 'z': 5, 'y': 5}, \n",
    "                 tshift=0, peakEpoch=0, nearPeakT=5, \n",
    "                 nObsNearPeak={'u': 0, 'g': 0, 'r': 5, 'i': 0, 'z': 0, 'y': 0},\n",
    "                 dataout=False)\n",
    "\n",
    "slicer = slicers.HealpixSlicer(nside=16) \n",
    "sqlconstraint = 'night<700'\n",
    "transmetricSky = metricBundles.MetricBundle(transmetric,slicer,sqlconstraint)\n",
    "\n",
    "group = metricBundles.MetricBundleGroup({'transmetricSky':transmetricSky},opsdb,outDir=outDir,resultsDb=resultsDb)\n",
    "group.runAll()\n",
    "group.plotAll(closefigs=False)"
   ]
  }
 ],
 "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": "3.6.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}