other/DTW.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.manifold import MDS\n",
    "import numpy as np\n",
    "%matplotlib inline\n",
    "import matplotlib.pyplot as plt\n",
    "from mpl_toolkits.mplot3d import Axes3D\n",
    "from concurrent import futures\n",
    "from sklearn.neighbors import KNeighborsClassifier\n",
    "import h5py\n",
    "\n",
    "import dtw\n",
    "import sys\n",
    "sys.path.append('../')\n",
    "import util"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## DTW"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We use [Dynamic Time Warping](https://masterwangzx.com/2019/02/06/dtw/) to calculate the distance of each sample. It takes a lot of time."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "complete (0,1)\n",
      "complete (0,2)\n",
      "complete (0,5)\n",
      "complete (0,3)\n",
      "complete (0,4)\n",
      "complete (0,7)\n",
      "complete (0,6)\n",
      "complete (0,8)\n",
      "complete (0,9)\n",
      "complete (0,11)\n",
      "complete (0,10)\n",
      "complete (0,13)\n",
      "complete (0,12)\n",
      "complete (0,15)\n",
      "complete (0,16)\n",
      "complete (0,14)\n",
      "complete (0,17)\n",
      "complete (0,18)\n",
      "complete (0,21)\n",
      "complete (0,19)\n",
      "complete (0,24)\n",
      "complete (0,22)\n",
      "complete (0,20)\n",
      "complete (0,23)\n",
      "complete (0,26)\n",
      "complete (0,28)\n",
      "complete (0,25)\n",
      "complete (0,27)\n",
      "complete (0,30)\n",
      "complete (0,29)\n",
      "complete (0,31)\n",
      "complete (0,32)\n",
      "complete (0,33)\n",
      "complete (0,34)\n",
      "complete (0,35)\n",
      "complete (0,36)\n",
      "complete (0,37)\n",
      "complete (0,40)\n",
      "complete (0,38)\n",
      "complete (0,39)\n",
      "complete (0,41)\n",
      "complete (0,42)\n",
      "complete (0,43)\n",
      "complete (0,46)\n",
      "complete (0,44)\n",
      "complete (0,47)\n",
      "complete (0,45)\n",
      "complete (0,48)\n",
      "complete (0,49)\n",
      "complete (0,50)\n",
      "complete (0,54)\n",
      "complete (0,51)\n",
      "complete (0,52)\n",
      "complete (0,53)\n",
      "complete (0,55)\n",
      "complete (0,56)\n",
      "complete (0,57)\n",
      "complete (0,58)\n",
      "complete (0,62)\n",
      "complete (0,61)\n",
      "complete (0,64)\n",
      "complete (0,63)\n",
      "complete (0,60)\n",
      "complete (0,59)\n",
      "complete (0,65)\n",
      "complete (0,68)\n",
      "complete (0,70)\n",
      "complete (0,66)\n",
      "complete (0,69)\n",
      "complete (0,67)\n",
      "complete (0,71)\n",
      "complete (0,72)\n",
      "complete (0,74)\n",
      "complete (0,73)\n",
      "complete (0,77)\n",
      "complete (0,76)\n",
      "complete (0,75)\n",
      "complete (0,78)\n",
      "complete (0,79)\n",
      "complete (0,80)\n",
      "complete (0,81)\n",
      "complete (0,82)\n",
      "complete (0,84)\n",
      "complete (0,86)\n",
      "complete (0,85)\n",
      "complete (0,83)\n",
      "complete (0,87)\n",
      "complete (0,88)\n",
      "complete (0,89)\n",
      "complete (0,90)\n",
      "complete (0,93)\n",
      "complete (0,91)\n",
      "complete (0,94)\n",
      "complete (0,92)\n",
      "complete (0,95)\n",
      "complete (0,96)\n",
      "complete (1,2)\n",
      "complete (0,97)\n",
      "complete (1,4)\n",
      "complete (1,5)\n",
      "complete (1,3)\n",
      "complete (0,98)\n",
      "complete (1,6)\n",
      "complete (1,7)\n",
      "complete (1,9)\n",
      "complete (1,8)\n",
      "complete (1,11)\n",
      "complete (1,12)\n",
      "complete (1,10)\n",
      "complete (1,13)\n",
      "complete (1,14)\n",
      "complete (1,15)\n",
      "complete (1,16)\n",
      "complete (1,17)\n",
      "complete (1,19)\n",
      "complete (1,18)\n",
      "complete (1,20)\n",
      "complete (1,21)\n",
      "complete (1,22)\n",
      "complete (1,23)\n",
      "complete (1,24)\n",
      "complete (1,28)\n",
      "complete (1,29)\n",
      "complete (1,27)\n",
      "complete (1,25)\n",
      "complete (1,26)\n",
      "complete (1,30)\n",
      "complete (1,31)\n",
      "complete (1,33)\n",
      "complete (1,32)\n",
      "complete (1,34)\n",
      "complete (1,36)\n",
      "complete (1,35)\n",
      "complete (1,37)\n",
      "complete (1,38)\n",
      "complete (1,39)\n",
      "complete (1,40)\n",
      "complete (1,41)\n",
      "complete (1,42)\n",
      "complete (1,43)\n",
      "complete (1,45)\n",
      "complete (1,44)\n",
      "complete (1,47)\n",
      "complete (1,46)\n",
      "complete (1,48)\n",
      "complete (1,49)\n",
      "complete (1,53)\n",
      "complete (1,50)\n",
      "complete (1,51)\n",
      "complete (1,52)\n",
      "complete (1,54)\n",
      "complete (1,55)\n",
      "complete (1,56)\n",
      "complete (1,57)\n",
      "complete (1,58)\n",
      "complete (1,59)\n",
      "complete (1,62)\n",
      "complete (1,60)\n",
      "complete (1,64)\n",
      "complete (1,63)\n",
      "complete (1,61)\n",
      "complete (1,65)\n",
      "complete (1,67)\n",
      "complete (1,69)\n",
      "complete (1,68)\n",
      "complete (1,70)\n",
      "complete (1,72)\n",
      "complete (1,66)\n",
      "complete (1,71)\n",
      "complete (1,73)\n",
      "complete (1,74)\n",
      "complete (1,78)\n",
      "complete (1,76)\n",
      "complete (1,75)\n",
      "complete (1,77)\n",
      "complete (1,80)\n",
      "complete (1,79)\n",
      "complete (1,81)\n",
      "complete (1,82)\n",
      "complete (1,86)\n",
      "complete (1,85)\n",
      "complete (1,83)\n",
      "complete (1,87)\n",
      "complete (1,84)\n",
      "complete (1,88)\n",
      "complete (1,89)\n",
      "complete (1,90)\n",
      "complete (1,96)\n",
      "complete (1,94)\n",
      "complete (1,92)\n",
      "complete (1,93)\n",
      "complete (1,91)\n",
      "complete (1,95)\n",
      "complete (1,97)\n",
      "complete (1,98)\n",
      "complete (2,3)\n",
      "complete (2,5)\n",
      "complete (2,4)\n",
      "complete (2,6)\n",
      "complete (2,8)\n",
      "complete (2,7)\n",
      "complete (2,9)\n",
      "complete (2,11)\n",
      "complete (2,10)\n",
      "complete (2,12)\n",
      "complete (2,16)\n",
      "complete (2,15)\n",
      "complete (2,14)\n",
      "complete (2,17)\n",
      "complete (2,13)\n",
      "complete (2,18)\n",
      "complete (2,21)\n",
      "complete (2,22)\n",
      "complete (2,20)\n",
      "complete (2,25)\n",
      "complete (2,23)\n",
      "complete (2,19)\n",
      "complete (2,24)\n",
      "complete (2,26)\n",
      "complete (2,29)\n",
      "complete (2,27)\n",
      "complete (2,32)\n",
      "complete (2,31)\n",
      "complete (2,30)\n",
      "complete (2,28)\n",
      "complete (2,33)\n",
      "complete (2,34)\n",
      "complete (2,41)\n",
      "complete (2,39)\n",
      "complete (2,37)\n",
      "complete (2,38)\n",
      "complete (2,36)\n",
      "complete (2,35)\n",
      "complete (2,40)\n",
      "complete (2,42)\n",
      "complete (2,43)\n",
      "complete (2,45)\n",
      "complete (2,44)\n",
      "complete (2,47)\n",
      "complete (2,46)\n",
      "complete (2,48)\n",
      "complete (2,49)\n",
      "complete (2,50)\n",
      "complete (2,53)\n",
      "complete (2,52)\n",
      "complete (2,54)\n",
      "complete (2,51)\n",
      "complete (2,55)\n",
      "complete (2,56)\n",
      "complete (2,57)\n",
      "complete (2,62)\n",
      "complete (2,61)\n",
      "complete (2,60)\n",
      "complete (2,58)\n",
      "complete (2,63)\n",
      "complete (2,59)\n",
      "complete (2,64)\n",
      "complete (2,65)\n",
      "complete (2,66)\n",
      "complete (2,67)\n",
      "complete (2,68)\n",
      "complete (2,71)\n",
      "complete (2,70)\n",
      "complete (2,69)\n",
      "complete (2,72)\n",
      "complete (2,73)\n",
      "complete (2,76)\n",
      "complete (2,77)\n",
      "complete (2,74)\n",
      "complete (2,75)\n",
      "complete (2,78)\n",
      "complete (2,79)\n",
      "complete (2,80)\n",
      "complete (2,81)\n",
      "complete (2,84)\n",
      "complete (2,82)\n",
      "complete (2,85)\n",
      "complete (2,87)\n",
      "complete (2,86)\n",
      "complete (2,83)\n",
      "complete (2,88)\n",
      "complete (2,89)\n",
      "complete (2,90)\n",
      "complete (2,94)\n",
      "complete (2,92)\n",
      "complete (2,93)\n",
      "complete (2,91)\n",
      "complete (2,95)\n",
      "complete (2,96)\n",
      "complete (2,97)\n",
      "complete (2,98)\n",
      "complete (3,6)\n",
      "complete (3,8)\n",
      "complete (3,5)\n",
      "complete (3,4)\n",
      "complete (3,7)\n",
      "complete (3,9)\n",
      "complete (3,10)\n",
      "complete (3,14)\n",
      "complete (3,11)\n",
      "complete (3,12)\n",
      "complete (3,13)\n",
      "complete (3,15)\n",
      "complete (3,17)\n",
      "complete (3,16)\n",
      "complete (3,18)\n",
      "complete (3,21)\n",
      "complete (3,19)\n",
      "complete (3,20)\n",
      "complete (3,22)\n",
      "complete (3,23)\n",
      "complete (3,24)\n",
      "complete (3,25)\n",
      "complete (3,26)\n",
      "complete (3,28)\n",
      "complete (3,30)\n",
      "complete (3,27)\n",
      "complete (3,32)\n",
      "complete (3,29)\n",
      "complete (3,31)\n",
      "complete (3,33)\n",
      "complete (3,34)\n",
      "complete (3,35)\n",
      "complete (3,37)\n",
      "complete (3,39)\n",
      "complete (3,36)\n",
      "complete (3,38)\n",
      "complete (3,41)\n",
      "complete (3,40)\n",
      "complete (3,42)\n",
      "complete (3,43)\n",
      "complete (3,47)\n",
      "complete (3,44)\n",
      "complete (3,45)\n",
      "complete (3,46)\n",
      "complete (3,48)\n",
      "complete (3,49)\n",
      "complete (3,50)\n",
      "complete (3,52)\n",
      "complete (3,51)\n",
      "complete (3,54)\n",
      "complete (3,53)\n",
      "complete (3,56)\n",
      "complete (3,55)\n",
      "complete (3,57)\n",
      "complete (3,58)\n",
      "complete (3,62)\n",
      "complete (3,60)\n",
      "complete (3,59)\n",
      "complete (3,61)\n",
      "complete (3,64)\n",
      "complete (3,63)\n",
      "complete (3,65)\n",
      "complete (3,66)\n",
      "complete (3,68)\n",
      "complete (3,69)\n",
      "complete (3,67)\n",
      "complete (3,70)\n",
      "complete (3,72)\n",
      "complete (3,71)\n",
      "complete (3,73)\n",
      "complete (3,78)\n",
      "complete (3,74)\n",
      "complete (3,77)\n",
      "complete (3,76)\n",
      "complete (3,80)\n",
      "complete (3,75)\n",
      "complete (3,79)\n",
      "complete (3,81)\n",
      "complete (3,85)\n",
      "complete (3,82)\n",
      "complete (3,84)\n",
      "complete (3,83)\n",
      "complete (3,87)\n",
      "complete (3,86)\n",
      "complete (3,88)\n",
      "complete (3,89)\n",
      "complete (3,91)\n",
      "complete (3,92)\n",
      "complete (3,94)\n",
      "complete (3,93)\n",
      "complete (3,90)\n",
      "complete (3,96)\n",
      "complete (3,95)\n",
      "complete (3,97)\n",
      "complete (4,8)\n",
      "complete (4,6)\n",
      "complete (4,5)\n",
      "complete (4,7)\n",
      "complete (4,9)\n",
      "complete (3,98)\n",
      "complete (4,10)\n",
      "complete (4,11)\n",
      "complete (4,14)\n",
      "complete (4,15)\n",
      "complete (4,12)\n",
      "complete (4,13)\n",
      "complete (4,16)\n",
      "complete (4,17)\n",
      "complete (4,18)\n",
      "complete (4,19)\n",
      "complete (4,20)\n",
      "complete (4,21)\n",
      "complete (4,22)\n",
      "complete (4,24)\n",
      "complete (4,23)\n",
      "complete (4,25)\n",
      "complete (4,26)\n",
      "complete (4,27)\n",
      "complete (4,28)\n",
      "complete (4,31)\n",
      "complete (4,29)\n",
      "complete (4,30)\n",
      "complete (4,32)\n",
      "complete (4,33)\n",
      "complete (4,34)\n",
      "complete (4,35)\n",
      "complete (4,38)\n",
      "complete (4,36)\n",
      "complete (4,37)\n",
      "complete (4,39)\n",
      "complete (4,40)\n",
      "complete (4,41)\n",
      "complete (4,42)\n",
      "complete (4,43)\n",
      "complete (4,44)\n",
      "complete (4,46)\n",
      "complete (4,45)\n",
      "complete (4,47)\n",
      "complete (4,48)\n",
      "complete (4,50)\n",
      "complete (4,51)\n",
      "complete (4,49)\n",
      "complete (4,52)\n",
      "complete (4,54)\n",
      "complete (4,55)\n",
      "complete (4,53)\n",
      "complete (4,56)\n",
      "complete (4,57)\n",
      "complete (4,58)\n",
      "complete (4,59)\n",
      "complete (4,60)\n",
      "complete (4,62)\n",
      "complete (4,61)\n",
      "complete (4,64)\n",
      "complete (4,63)\n",
      "complete (4,65)\n",
      "complete (4,67)\n",
      "complete (4,66)\n",
      "complete (4,68)\n",
      "complete (4,69)\n",
      "complete (4,70)\n",
      "complete (4,72)\n",
      "complete (4,71)\n",
      "complete (4,76)\n",
      "complete (4,77)\n",
      "complete (4,74)\n",
      "complete (4,73)\n",
      "complete (4,75)\n",
      "complete (4,78)\n",
      "complete (4,79)\n",
      "complete (4,80)\n",
      "complete (4,81)\n",
      "complete (4,83)\n",
      "complete (4,87)\n",
      "complete (4,86)\n",
      "complete (4,84)\n",
      "complete (4,85)\n",
      "complete (4,82)\n",
      "complete (4,88)\n",
      "complete (4,89)\n",
      "complete (4,90)\n",
      "complete (4,91)\n",
      "complete (4,94)\n",
      "complete (4,92)\n",
      "complete (4,93)\n",
      "complete (4,95)\n",
      "complete (4,96)\n",
      "complete (4,97)\n",
      "complete (4,98)\n",
      "complete (5,8)\n",
      "complete (5,6)\n",
      "complete (5,7)\n",
      "complete (5,9)\n",
      "complete (5,11)\n",
      "complete (5,10)\n",
      "complete (5,12)\n",
      "complete (5,14)\n",
      "complete (5,13)\n",
      "complete (5,15)\n",
      "complete (5,16)\n",
      "complete (5,17)\n",
      "complete (5,19)\n",
      "complete (5,18)\n",
      "complete (5,20)\n",
      "complete (5,21)\n",
      "complete (5,22)\n",
      "complete (5,23)\n",
      "complete (5,24)\n",
      "complete (5,25)\n",
      "complete (5,26)\n",
      "complete (5,27)\n",
      "complete (5,28)\n",
      "complete (5,30)\n",
      "complete (5,31)\n",
      "complete (5,32)\n",
      "complete (5,29)\n",
      "complete (5,33)\n",
      "complete (5,34)\n",
      "complete (5,36)\n",
      "complete (5,35)\n",
      "complete (5,38)\n",
      "complete (5,40)\n",
      "complete (5,37)\n",
      "complete (5,39)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "complete (5,41)\n",
      "complete (5,42)\n",
      "complete (5,43)\n",
      "complete (5,44)\n",
      "complete (5,45)\n",
      "complete (5,46)\n",
      "complete (5,48)\n",
      "complete (5,47)\n",
      "complete (5,49)\n",
      "complete (5,50)\n",
      "complete (5,51)\n",
      "complete (5,52)\n",
      "complete (5,54)\n",
      "complete (5,55)\n",
      "complete (5,56)\n",
      "complete (5,53)\n",
      "complete (5,57)\n",
      "complete (5,58)\n",
      "complete (5,59)\n",
      "complete (5,60)\n",
      "complete (5,61)\n",
      "complete (5,64)\n",
      "complete (5,62)\n",
      "complete (5,63)\n",
      "complete (5,65)\n",
      "complete (5,68)\n",
      "complete (5,66)\n",
      "complete (5,67)\n",
      "complete (5,69)\n",
      "complete (5,70)\n",
      "complete (5,71)\n",
      "complete (5,73)\n",
      "complete (5,72)\n",
      "complete (5,74)\n",
      "complete (5,75)\n",
      "complete (5,77)\n",
      "complete (5,76)\n",
      "complete (5,78)\n",
      "complete (5,80)\n",
      "complete (5,81)\n",
      "complete (5,79)\n",
      "complete (5,82)\n",
      "complete (5,83)\n",
      "complete (5,84)\n",
      "complete (5,85)\n",
      "complete (5,86)\n",
      "complete (5,88)\n",
      "complete (5,87)\n",
      "complete (5,89)\n",
      "complete (5,90)\n",
      "complete (5,92)\n",
      "complete (5,91)\n",
      "complete (5,93)\n",
      "complete (5,94)\n",
      "complete (5,97)\n",
      "complete (5,96)\n",
      "complete (5,95)\n",
      "complete (5,98)\n",
      "complete (6,7)\n",
      "complete (6,8)\n",
      "complete (6,10)\n",
      "complete (6,9)\n",
      "complete (6,11)\n",
      "complete (6,12)\n",
      "complete (6,15)\n",
      "complete (6,14)\n",
      "complete (6,13)\n",
      "complete (6,16)\n",
      "complete (6,17)\n",
      "complete (6,18)\n",
      "complete (6,19)\n",
      "complete (6,20)\n",
      "complete (6,21)\n",
      "complete (6,22)\n",
      "complete (6,23)\n",
      "complete (6,24)\n",
      "complete (6,26)\n",
      "complete (6,25)\n",
      "complete (6,27)\n",
      "complete (6,28)\n",
      "complete (6,29)\n",
      "complete (6,30)\n",
      "complete (6,31)\n",
      "complete (6,32)\n",
      "complete (6,33)\n",
      "complete (6,35)\n",
      "complete (6,34)\n",
      "complete (6,36)\n",
      "complete (6,37)\n",
      "complete (6,38)\n",
      "complete (6,39)\n",
      "complete (6,41)\n",
      "complete (6,40)\n",
      "complete (6,42)\n",
      "complete (6,45)\n",
      "complete (6,43)\n",
      "complete (6,44)\n",
      "complete (6,47)\n",
      "complete (6,48)\n",
      "complete (6,46)\n",
      "complete (6,50)\n",
      "complete (6,49)\n",
      "complete (6,51)\n",
      "complete (6,54)\n",
      "complete (6,53)\n",
      "complete (6,52)\n",
      "complete (6,55)\n",
      "complete (6,56)\n",
      "complete (6,58)\n",
      "complete (6,57)\n",
      "complete (6,62)\n",
      "complete (6,59)\n",
      "complete (6,60)\n",
      "complete (6,61)\n",
      "complete (6,64)\n",
      "complete (6,63)\n",
      "complete (6,65)\n",
      "complete (6,66)\n",
      "complete (6,68)\n",
      "complete (6,67)\n",
      "complete (6,69)\n",
      "complete (6,70)\n",
      "complete (6,72)\n",
      "complete (6,71)\n",
      "complete (6,73)\n",
      "complete (6,74)\n",
      "complete (6,76)\n",
      "complete (6,75)\n",
      "complete (6,77)\n",
      "complete (6,78)\n",
      "complete (6,79)\n",
      "complete (6,80)\n",
      "complete (6,81)\n",
      "complete (6,85)\n",
      "complete (6,84)\n",
      "complete (6,82)\n",
      "complete (6,83)\n",
      "complete (6,86)\n",
      "complete (6,87)\n",
      "complete (6,88)\n",
      "complete (6,89)\n",
      "complete (6,90)\n",
      "complete (6,93)\n",
      "complete (6,91)\n",
      "complete (6,92)\n",
      "complete (6,94)\n",
      "complete (6,96)\n",
      "complete (6,95)\n",
      "complete (6,97)\n",
      "complete (7,8)\n",
      "complete (6,98)\n",
      "complete (7,9)\n",
      "complete (7,11)\n",
      "complete (7,10)\n",
      "complete (7,12)\n",
      "complete (7,13)\n",
      "complete (7,15)\n",
      "complete (7,14)\n",
      "complete (7,16)\n",
      "complete (7,17)\n",
      "complete (7,19)\n",
      "complete (7,18)\n",
      "complete (7,20)\n",
      "complete (7,22)\n",
      "complete (7,21)\n",
      "complete (7,24)\n",
      "complete (7,23)\n",
      "complete (7,25)\n",
      "complete (7,28)\n",
      "complete (7,26)\n",
      "complete (7,29)\n",
      "complete (7,27)\n",
      "complete (7,30)\n",
      "complete (7,32)\n",
      "complete (7,31)\n",
      "complete (7,34)\n",
      "complete (7,33)\n",
      "complete (7,35)\n",
      "complete (7,36)\n",
      "complete (7,38)\n",
      "complete (7,37)\n",
      "complete (7,39)\n",
      "complete (7,40)\n",
      "complete (7,41)\n",
      "complete (7,42)\n",
      "complete (7,43)\n",
      "complete (7,44)\n",
      "complete (7,45)\n",
      "complete (7,46)\n",
      "complete (7,47)\n",
      "complete (7,48)\n",
      "complete (7,49)\n",
      "complete (7,50)\n",
      "complete (7,54)\n",
      "complete (7,51)\n",
      "complete (7,55)\n",
      "complete (7,52)\n",
      "complete (7,53)\n",
      "complete (7,56)\n",
      "complete (7,57)\n",
      "complete (7,58)\n",
      "complete (7,61)\n",
      "complete (7,60)\n",
      "complete (7,59)\n",
      "complete (7,62)\n",
      "complete (7,64)\n",
      "complete (7,63)\n",
      "complete (7,65)\n",
      "complete (7,68)\n",
      "complete (7,69)\n",
      "complete (7,71)\n",
      "complete (7,66)\n",
      "complete (7,67)\n",
      "complete (7,70)\n",
      "complete (7,72)\n",
      "complete (7,73)\n",
      "complete (7,77)\n",
      "complete (7,74)\n",
      "complete (7,76)\n",
      "complete (7,79)\n",
      "complete (7,78)\n",
      "complete (7,75)\n",
      "complete (7,80)\n",
      "complete (7,81)\n",
      "complete (7,82)\n",
      "complete (7,83)\n",
      "complete (7,84)\n",
      "complete (7,85)\n",
      "complete (7,86)\n",
      "complete (7,87)\n",
      "complete (7,88)\n",
      "complete (7,89)\n",
      "complete (7,91)\n",
      "complete (7,90)\n",
      "complete (7,93)\n",
      "complete (7,92)\n",
      "complete (7,94)\n",
      "complete (7,95)\n",
      "complete (7,96)\n",
      "complete (7,97)\n",
      "complete (7,98)\n",
      "complete (8,9)\n",
      "complete (8,10)\n",
      "complete (8,12)\n",
      "complete (8,11)\n",
      "complete (8,13)\n",
      "complete (8,14)\n",
      "complete (8,15)\n",
      "complete (8,16)\n",
      "complete (8,18)\n",
      "complete (8,19)\n",
      "complete (8,20)\n",
      "complete (8,17)\n",
      "complete (8,21)\n",
      "complete (8,22)\n",
      "complete (8,23)\n",
      "complete (8,24)\n",
      "complete (8,25)\n",
      "complete (8,26)\n",
      "complete (8,27)\n",
      "complete (8,28)\n",
      "complete (8,31)\n",
      "complete (8,29)\n",
      "complete (8,30)\n",
      "complete (8,33)\n",
      "complete (8,32)\n",
      "complete (8,34)\n",
      "complete (8,35)\n",
      "complete (8,36)\n",
      "complete (8,37)\n",
      "complete (8,38)\n",
      "complete (8,39)\n",
      "complete (8,40)\n",
      "complete (8,41)\n",
      "complete (8,42)\n",
      "complete (8,43)\n",
      "complete (8,44)\n",
      "complete (8,45)\n",
      "complete (8,46)\n",
      "complete (8,47)\n",
      "complete (8,48)\n",
      "complete (8,49)\n",
      "complete (8,50)\n",
      "complete (8,51)\n",
      "complete (8,52)\n",
      "complete (8,54)\n",
      "complete (8,55)\n",
      "complete (8,56)\n",
      "complete (8,53)\n",
      "complete (8,57)\n",
      "complete (8,58)\n",
      "complete (8,60)\n",
      "complete (8,59)\n",
      "complete (8,61)\n",
      "complete (8,62)\n",
      "complete (8,64)\n",
      "complete (8,63)\n",
      "complete (8,65)\n",
      "complete (8,67)\n",
      "complete (8,68)\n",
      "complete (8,70)\n",
      "complete (8,66)\n",
      "complete (8,69)\n",
      "complete (8,71)\n",
      "complete (8,72)\n",
      "complete (8,73)\n",
      "complete (8,76)\n",
      "complete (8,77)\n",
      "complete (8,74)\n",
      "complete (8,75)\n",
      "complete (8,78)\n",
      "complete (8,79)\n",
      "complete (8,81)\n",
      "complete (8,80)\n",
      "complete (8,83)\n",
      "complete (8,85)\n",
      "complete (8,82)\n",
      "complete (8,84)\n",
      "complete (8,86)\n",
      "complete (8,87)\n",
      "complete (8,88)\n",
      "complete (8,89)\n",
      "complete (8,92)\n",
      "complete (8,91)\n",
      "complete (8,90)\n",
      "complete (8,93)\n",
      "complete (8,94)\n",
      "complete (8,95)\n",
      "complete (8,96)\n",
      "complete (8,97)\n",
      "complete (8,98)\n",
      "complete (9,11)\n",
      "complete (9,10)\n",
      "complete (9,12)\n",
      "complete (9,13)\n",
      "complete (9,15)\n",
      "complete (9,14)\n",
      "complete (9,16)\n",
      "complete (9,17)\n",
      "complete (9,19)\n",
      "complete (9,20)\n",
      "complete (9,18)\n",
      "complete (9,21)\n",
      "complete (9,22)\n",
      "complete (9,24)\n",
      "complete (9,23)\n",
      "complete (9,25)\n",
      "complete (9,29)\n",
      "complete (9,28)\n",
      "complete (9,27)\n",
      "complete (9,26)\n",
      "complete (9,31)\n",
      "complete (9,33)\n",
      "complete (9,32)\n",
      "complete (9,30)\n",
      "complete (9,34)\n",
      "complete (9,35)\n",
      "complete (9,37)\n",
      "complete (9,36)\n",
      "complete (9,38)\n",
      "complete (9,39)\n",
      "complete (9,40)\n",
      "complete (9,41)\n",
      "complete (9,42)\n",
      "complete (9,43)\n",
      "complete (9,44)\n",
      "complete (9,45)\n",
      "complete (9,46)\n",
      "complete (9,47)\n",
      "complete (9,48)\n",
      "complete (9,49)\n",
      "complete (9,50)\n",
      "complete (9,53)\n",
      "complete (9,51)\n",
      "complete (9,54)\n",
      "complete (9,52)\n",
      "complete (9,55)\n",
      "complete (9,56)\n",
      "complete (9,57)\n",
      "complete (9,61)\n",
      "complete (9,58)\n",
      "complete (9,60)\n",
      "complete (9,59)\n",
      "complete (9,63)\n",
      "complete (9,62)\n",
      "complete (9,64)\n",
      "complete (9,65)\n",
      "complete (9,66)\n",
      "complete (9,69)\n",
      "complete (9,68)\n",
      "complete (9,67)\n",
      "complete (9,71)\n",
      "complete (9,70)\n",
      "complete (9,72)\n",
      "complete (9,73)\n",
      "complete (9,74)\n",
      "complete (9,76)\n",
      "complete (9,75)\n",
      "complete (9,77)\n",
      "complete (9,78)\n",
      "complete (9,79)\n",
      "complete (9,80)\n",
      "complete (9,81)\n",
      "complete (9,82)\n",
      "complete (9,85)\n",
      "complete (9,84)\n",
      "complete (9,83)\n",
      "complete (9,87)\n",
      "complete (9,86)\n",
      "complete (9,88)\n",
      "complete (9,89)\n",
      "complete (9,90)\n",
      "complete (9,92)\n",
      "complete (9,93)\n",
      "complete (9,94)\n",
      "complete (9,91)\n",
      "complete (9,96)\n",
      "complete (9,95)\n",
      "complete (10,11)\n",
      "complete (9,97)\n",
      "complete (9,98)\n",
      "complete (10,12)\n",
      "complete (10,14)\n",
      "complete (10,13)\n",
      "complete (10,15)\n",
      "complete (10,16)\n",
      "complete (10,17)\n",
      "complete (10,19)\n",
      "complete (10,22)\n",
      "complete (10,21)\n",
      "complete (10,18)\n",
      "complete (10,20)\n",
      "complete (10,24)\n",
      "complete (10,25)\n",
      "complete (10,23)\n",
      "complete (10,28)\n",
      "complete (10,27)\n",
      "complete (10,30)\n",
      "complete (10,26)\n",
      "complete (10,31)\n",
      "complete (10,29)\n",
      "complete (10,32)\n",
      "complete (10,33)\n",
      "complete (10,34)\n",
      "complete (10,35)\n",
      "complete (10,36)\n",
      "complete (10,38)\n",
      "complete (10,37)\n",
      "complete (10,39)\n",
      "complete (10,41)\n",
      "complete (10,40)\n",
      "complete (10,42)\n",
      "complete (10,43)\n",
      "complete (10,44)\n",
      "complete (10,46)\n",
      "complete (10,45)\n",
      "complete (10,47)\n",
      "complete (10,48)\n",
      "complete (10,49)\n",
      "complete (10,50)\n",
      "complete (10,52)\n",
      "complete (10,53)\n",
      "complete (10,51)\n",
      "complete (10,54)\n",
      "complete (10,55)\n",
      "complete (10,56)\n",
      "complete (10,57)\n",
      "complete (10,58)\n",
      "complete (10,62)\n",
      "complete (10,59)\n",
      "complete (10,61)\n",
      "complete (10,63)\n",
      "complete (10,60)\n",
      "complete (10,64)\n",
      "complete (10,65)\n",
      "complete (10,68)\n",
      "complete (10,66)\n",
      "complete (10,67)\n",
      "complete (10,69)\n",
      "complete (10,71)\n",
      "complete (10,72)\n",
      "complete (10,70)\n",
      "complete (10,73)\n",
      "complete (10,75)\n",
      "complete (10,76)\n",
      "complete (10,77)\n",
      "complete (10,74)\n",
      "complete (10,79)\n",
      "complete (10,78)\n",
      "complete (10,80)\n",
      "complete (10,81)\n",
      "complete (10,85)\n",
      "complete (10,82)\n",
      "complete (10,83)\n",
      "complete (10,84)\n",
      "complete (10,86)\n",
      "complete (10,87)\n",
      "complete (10,88)\n",
      "complete (10,89)\n",
      "complete (10,90)\n",
      "complete (10,92)\n",
      "complete (10,94)\n",
      "complete (10,95)\n",
      "complete (10,91)\n",
      "complete (10,93)\n",
      "complete (10,96)\n",
      "complete (10,97)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "complete (11,14)\n",
      "complete (11,15)\n",
      "complete (11,13)\n",
      "complete (11,17)\n",
      "complete (11,16)\n",
      "complete (10,98)\n",
      "complete (11,12)\n",
      "complete (11,18)\n",
      "complete (11,19)\n",
      "complete (11,21)\n",
      "complete (11,20)\n",
      "complete (11,23)\n",
      "complete (11,22)\n",
      "complete (11,24)\n",
      "complete (11,25)\n",
      "complete (11,26)\n",
      "complete (11,27)\n",
      "complete (11,28)\n",
      "complete (11,29)\n",
      "complete (11,32)\n",
      "complete (11,31)\n",
      "complete (11,30)\n",
      "complete (11,33)\n",
      "complete (11,34)\n",
      "complete (11,36)\n",
      "complete (11,38)\n",
      "complete (11,35)\n",
      "complete (11,40)\n",
      "complete (11,39)\n",
      "complete (11,41)\n",
      "complete (11,37)\n",
      "complete (11,42)\n",
      "complete (11,43)\n",
      "complete (11,47)\n",
      "complete (11,44)\n",
      "complete (11,45)\n",
      "complete (11,46)\n",
      "complete (11,48)\n",
      "complete (11,49)\n",
      "complete (11,50)\n",
      "complete (11,51)\n",
      "complete (11,52)\n",
      "complete (11,55)\n",
      "complete (11,53)\n",
      "complete (11,54)\n",
      "complete (11,56)\n",
      "complete (11,57)\n",
      "complete (11,58)\n",
      "complete (11,60)\n",
      "complete (11,62)\n",
      "complete (11,61)\n",
      "complete (11,59)\n",
      "complete (11,64)\n",
      "complete (11,65)\n",
      "complete (11,63)\n",
      "complete (11,66)\n",
      "complete (11,68)\n",
      "complete (11,69)\n",
      "complete (11,70)\n",
      "complete (11,71)\n",
      "complete (11,67)\n",
      "complete (11,72)\n",
      "complete (11,73)\n",
      "complete (11,76)\n",
      "complete (11,74)\n",
      "complete (11,75)\n",
      "complete (11,77)\n",
      "complete (11,78)\n",
      "complete (11,79)\n",
      "complete (11,80)\n",
      "complete (11,81)\n",
      "complete (11,84)\n",
      "complete (11,85)\n",
      "complete (11,86)\n",
      "complete (11,83)\n",
      "complete (11,87)\n",
      "complete (11,82)\n",
      "complete (11,88)\n",
      "complete (11,89)\n",
      "complete (11,90)\n",
      "complete (11,94)\n",
      "complete (11,92)\n",
      "complete (11,93)\n",
      "complete (11,91)\n",
      "complete (11,96)\n",
      "complete (11,95)\n",
      "complete (11,97)\n",
      "complete (12,16)\n",
      "complete (12,13)\n",
      "complete (11,98)\n",
      "complete (12,15)\n",
      "complete (12,17)\n",
      "complete (12,14)\n",
      "complete (12,18)\n",
      "complete (12,19)\n",
      "complete (12,20)\n",
      "complete (12,22)\n",
      "complete (12,21)\n",
      "complete (12,23)\n",
      "complete (12,25)\n",
      "complete (12,24)\n",
      "complete (12,26)\n",
      "complete (12,27)\n",
      "complete (12,29)\n",
      "complete (12,28)\n",
      "complete (12,30)\n",
      "complete (12,31)\n",
      "complete (12,33)\n",
      "complete (12,32)\n",
      "complete (12,34)\n",
      "complete (12,35)\n",
      "complete (12,36)\n",
      "complete (12,38)\n",
      "complete (12,37)\n",
      "complete (12,39)\n",
      "complete (12,42)\n",
      "complete (12,41)\n",
      "complete (12,40)\n",
      "complete (12,43)\n",
      "complete (12,45)\n",
      "complete (12,44)\n",
      "complete (12,47)\n",
      "complete (12,48)\n",
      "complete (12,46)\n",
      "complete (12,49)\n",
      "complete (12,50)\n",
      "complete (12,51)\n",
      "complete (12,55)\n",
      "complete (12,52)\n",
      "complete (12,54)\n",
      "complete (12,53)\n",
      "complete (12,57)\n",
      "complete (12,56)\n",
      "complete (12,58)\n",
      "complete (12,60)\n",
      "complete (12,59)\n",
      "complete (12,61)\n",
      "complete (12,62)\n",
      "complete (12,64)\n",
      "complete (12,65)\n",
      "complete (12,63)\n",
      "complete (12,66)\n",
      "complete (12,67)\n",
      "complete (12,71)\n",
      "complete (12,69)\n",
      "complete (12,72)\n",
      "complete (12,70)\n",
      "complete (12,68)\n",
      "complete (12,73)\n",
      "complete (12,76)\n",
      "complete (12,74)\n",
      "complete (12,77)\n",
      "complete (12,75)\n",
      "complete (12,79)\n",
      "complete (12,78)\n",
      "complete (12,81)\n",
      "complete (12,80)\n",
      "complete (12,82)\n",
      "complete (12,85)\n",
      "complete (12,86)\n",
      "complete (12,83)\n",
      "complete (12,87)\n",
      "complete (12,88)\n",
      "complete (12,89)\n",
      "complete (12,84)\n",
      "complete (12,90)\n",
      "complete (12,94)\n",
      "complete (12,93)\n",
      "complete (12,95)\n",
      "complete (12,96)\n",
      "complete (12,92)\n",
      "complete (12,97)\n",
      "complete (12,91)\n",
      "complete (13,14)\n",
      "complete (13,15)\n",
      "complete (12,98)\n",
      "complete (13,18)\n",
      "complete (13,19)\n",
      "complete (13,17)\n",
      "complete (13,20)\n",
      "complete (13,16)\n",
      "complete (13,21)\n",
      "complete (13,22)\n",
      "complete (13,24)\n",
      "complete (13,23)\n",
      "complete (13,25)\n",
      "complete (13,26)\n",
      "complete (13,28)\n",
      "complete (13,27)\n",
      "complete (13,29)\n",
      "complete (13,31)\n",
      "complete (13,32)\n",
      "complete (13,30)\n",
      "complete (13,34)\n",
      "complete (13,35)\n",
      "complete (13,33)\n",
      "complete (13,36)\n",
      "complete (13,37)\n",
      "complete (13,39)\n",
      "complete (13,38)\n",
      "complete (13,40)\n",
      "complete (13,42)\n",
      "complete (13,44)\n",
      "complete (13,41)\n",
      "complete (13,43)\n",
      "complete (13,47)\n",
      "complete (13,45)\n",
      "complete (13,46)\n",
      "complete (13,48)\n",
      "complete (13,49)\n",
      "complete (13,50)\n",
      "complete (13,52)\n",
      "complete (13,51)\n",
      "complete (13,54)\n",
      "complete (13,53)\n",
      "complete (13,55)\n",
      "complete (13,57)\n",
      "complete (13,56)\n",
      "complete (13,58)\n",
      "complete (13,59)\n",
      "complete (13,60)\n",
      "complete (13,62)\n",
      "complete (13,61)\n",
      "complete (13,63)\n",
      "complete (13,64)\n",
      "complete (13,65)\n",
      "complete (13,66)\n",
      "complete (13,68)\n",
      "complete (13,67)\n",
      "complete (13,69)\n",
      "complete (13,71)\n",
      "complete (13,70)\n",
      "complete (13,72)\n",
      "complete (13,74)\n",
      "complete (13,76)\n",
      "complete (13,73)\n",
      "complete (13,75)\n",
      "complete (13,77)\n",
      "complete (13,78)\n",
      "complete (13,79)\n",
      "complete (13,80)\n",
      "complete (13,81)\n",
      "complete (13,82)\n",
      "complete (13,83)\n",
      "complete (13,84)\n",
      "complete (13,85)\n",
      "complete (13,86)\n",
      "complete (13,87)\n",
      "complete (13,88)\n",
      "complete (13,89)\n",
      "complete (13,90)\n",
      "complete (13,91)\n",
      "complete (13,92)\n",
      "complete (13,93)\n",
      "complete (13,94)\n",
      "complete (13,96)\n",
      "complete (13,95)\n",
      "complete (13,97)\n",
      "complete (14,15)\n",
      "complete (14,16)\n",
      "complete (13,98)\n",
      "complete (14,17)\n",
      "complete (14,18)\n",
      "complete (14,19)\n",
      "complete (14,21)\n",
      "complete (14,20)\n",
      "complete (14,22)\n",
      "complete (14,23)\n",
      "complete (14,24)\n",
      "complete (14,26)\n",
      "complete (14,25)\n",
      "complete (14,29)\n",
      "complete (14,30)\n",
      "complete (14,27)\n",
      "complete (14,28)\n",
      "complete (14,31)\n",
      "complete (14,32)\n",
      "complete (14,33)\n",
      "complete (14,34)\n",
      "complete (14,35)\n",
      "complete (14,38)\n",
      "complete (14,36)\n",
      "complete (14,39)\n",
      "complete (14,37)\n",
      "complete (14,40)\n",
      "complete (14,41)\n",
      "complete (14,42)\n",
      "complete (14,43)\n",
      "complete (14,44)\n",
      "complete (14,45)\n",
      "complete (14,47)\n",
      "complete (14,46)\n",
      "complete (14,48)\n",
      "complete (14,49)\n",
      "complete (14,51)\n",
      "complete (14,50)\n",
      "complete (14,53)\n",
      "complete (14,52)\n",
      "complete (14,54)\n",
      "complete (14,55)\n",
      "complete (14,56)\n",
      "complete (14,57)\n",
      "complete (14,59)\n",
      "complete (14,58)\n",
      "complete (14,64)\n",
      "complete (14,60)\n",
      "complete (14,62)\n",
      "complete (14,63)\n",
      "complete (14,61)\n",
      "complete (14,67)\n",
      "complete (14,71)\n",
      "complete (14,68)\n",
      "complete (14,66)\n",
      "complete (14,65)\n",
      "complete (14,72)\n",
      "complete (14,70)\n",
      "complete (14,69)\n",
      "complete (14,77)\n",
      "complete (14,76)\n",
      "complete (14,73)\n",
      "complete (14,75)\n",
      "complete (14,78)\n",
      "complete (14,74)\n",
      "complete (14,80)\n",
      "complete (14,79)\n",
      "complete (14,81)\n",
      "complete (14,83)\n",
      "complete (14,82)\n",
      "complete (14,87)\n",
      "complete (14,84)\n",
      "complete (14,86)\n",
      "complete (14,85)\n",
      "complete (14,88)\n",
      "complete (14,90)\n",
      "complete (14,89)\n",
      "complete (14,94)\n",
      "complete (14,92)\n",
      "complete (14,93)\n",
      "complete (14,91)\n",
      "complete (14,96)\n",
      "complete (14,95)\n",
      "complete (14,97)\n",
      "complete (14,98)\n",
      "complete (15,18)\n",
      "complete (15,16)\n",
      "complete (15,17)\n",
      "complete (15,19)\n",
      "complete (15,20)\n",
      "complete (15,21)\n",
      "complete (15,22)\n",
      "complete (15,24)\n",
      "complete (15,23)\n",
      "complete (15,26)\n",
      "complete (15,25)\n",
      "complete (15,28)\n",
      "complete (15,27)\n",
      "complete (15,29)\n",
      "complete (15,30)\n",
      "complete (15,31)\n",
      "complete (15,32)\n",
      "complete (15,34)\n",
      "complete (15,33)\n",
      "complete (15,35)\n",
      "complete (15,36)\n",
      "complete (15,37)\n",
      "complete (15,39)\n",
      "complete (15,40)\n",
      "complete (15,38)\n",
      "complete (15,41)\n",
      "complete (15,42)\n",
      "complete (15,43)\n",
      "complete (15,44)\n",
      "complete (15,45)\n",
      "complete (15,46)\n",
      "complete (15,47)\n",
      "complete (15,48)\n",
      "complete (15,49)\n",
      "complete (15,51)\n",
      "complete (15,50)\n",
      "complete (15,52)\n",
      "complete (15,53)\n",
      "complete (15,54)\n",
      "complete (15,55)\n",
      "complete (15,56)\n",
      "complete (15,57)\n",
      "complete (15,61)\n",
      "complete (15,60)\n",
      "complete (15,59)\n",
      "complete (15,58)\n",
      "complete (15,62)\n",
      "complete (15,63)\n",
      "complete (15,64)\n",
      "complete (15,69)\n",
      "complete (15,66)\n",
      "complete (15,68)\n",
      "complete (15,65)\n",
      "complete (15,70)\n",
      "complete (15,67)\n",
      "complete (15,72)\n",
      "complete (15,71)\n",
      "complete (15,74)\n",
      "complete (15,73)\n",
      "complete (15,78)\n",
      "complete (15,77)\n",
      "complete (15,76)\n",
      "complete (15,75)\n",
      "complete (15,79)\n",
      "complete (15,80)\n",
      "complete (15,81)\n",
      "complete (15,86)\n",
      "complete (15,82)\n",
      "complete (15,83)\n",
      "complete (15,85)\n",
      "complete (15,84)\n",
      "complete (15,87)\n",
      "complete (15,89)\n",
      "complete (15,88)\n",
      "complete (15,90)\n",
      "complete (15,91)\n",
      "complete (15,94)\n",
      "complete (15,93)\n",
      "complete (15,92)\n",
      "complete (15,95)\n",
      "complete (15,96)\n",
      "complete (15,97)\n",
      "complete (16,17)\n",
      "complete (15,98)\n",
      "complete (16,19)\n",
      "complete (16,18)\n",
      "complete (16,22)\n",
      "complete (16,20)\n",
      "complete (16,21)\n",
      "complete (16,23)\n",
      "complete (16,24)\n",
      "complete (16,25)\n",
      "complete (16,26)\n",
      "complete (16,27)\n",
      "complete (16,29)\n",
      "complete (16,30)\n",
      "complete (16,28)\n",
      "complete (16,31)\n",
      "complete (16,32)\n",
      "complete (16,33)\n",
      "complete (16,36)\n",
      "complete (16,34)\n",
      "complete (16,35)\n",
      "complete (16,38)\n",
      "complete (16,39)\n",
      "complete (16,37)\n",
      "complete (16,40)\n",
      "complete (16,41)\n",
      "complete (16,42)\n",
      "complete (16,44)\n",
      "complete (16,43)\n",
      "complete (16,45)\n",
      "complete (16,46)\n",
      "complete (16,47)\n",
      "complete (16,48)\n",
      "complete (16,52)\n",
      "complete (16,50)\n",
      "complete (16,49)\n",
      "complete (16,51)\n",
      "complete (16,53)\n",
      "complete (16,54)\n",
      "complete (16,55)\n",
      "complete (16,56)\n",
      "complete (16,61)\n",
      "complete (16,62)\n",
      "complete (16,58)\n",
      "complete (16,60)\n",
      "complete (16,57)\n",
      "complete (16,59)\n",
      "complete (16,63)\n",
      "complete (16,64)\n",
      "complete (16,68)\n",
      "complete (16,69)\n",
      "complete (16,67)\n",
      "complete (16,65)\n",
      "complete (16,66)\n",
      "complete (16,70)\n",
      "complete (16,71)\n",
      "complete (16,72)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "complete (16,73)\n",
      "complete (16,74)\n",
      "complete (16,76)\n",
      "complete (16,77)\n",
      "complete (16,78)\n",
      "complete (16,75)\n",
      "complete (16,79)\n",
      "complete (16,80)\n",
      "complete (16,81)\n",
      "complete (16,87)\n",
      "complete (16,83)\n",
      "complete (16,86)\n",
      "complete (16,85)\n",
      "complete (16,82)\n",
      "complete (16,84)\n",
      "complete (16,89)\n",
      "complete (16,88)\n",
      "complete (16,94)\n",
      "complete (16,90)\n",
      "complete (16,91)\n",
      "complete (16,92)\n",
      "complete (16,93)\n",
      "complete (16,95)\n",
      "complete (16,97)\n",
      "complete (16,96)\n",
      "complete (17,18)\n",
      "complete (16,98)\n",
      "complete (17,19)\n",
      "complete (17,21)\n",
      "complete (17,20)\n",
      "complete (17,22)\n",
      "complete (17,23)\n",
      "complete (17,24)\n",
      "complete (17,25)\n",
      "complete (17,26)\n",
      "complete (17,28)\n",
      "complete (17,27)\n",
      "complete (17,29)\n",
      "complete (17,30)\n",
      "complete (17,31)\n",
      "complete (17,32)\n",
      "complete (17,33)\n",
      "complete (17,34)\n",
      "complete (17,35)\n",
      "complete (17,36)\n",
      "complete (17,37)\n",
      "complete (17,38)\n",
      "complete (17,39)\n",
      "complete (17,41)\n",
      "complete (17,40)\n",
      "complete (17,43)\n",
      "complete (17,42)\n",
      "complete (17,44)\n",
      "complete (17,45)\n",
      "complete (17,46)\n",
      "complete (17,47)\n",
      "complete (17,48)\n",
      "complete (17,49)\n",
      "complete (17,50)\n",
      "complete (17,52)\n",
      "complete (17,51)\n",
      "complete (17,53)\n",
      "complete (17,54)\n",
      "complete (17,55)\n",
      "complete (17,56)\n",
      "complete (17,57)\n",
      "complete (17,58)\n",
      "complete (17,59)\n",
      "complete (17,60)\n",
      "complete (17,64)\n",
      "complete (17,62)\n",
      "complete (17,61)\n",
      "complete (17,63)\n",
      "complete (17,65)\n",
      "complete (17,66)\n",
      "complete (17,68)\n",
      "complete (17,67)\n",
      "complete (17,69)\n",
      "complete (17,70)\n",
      "complete (17,72)\n",
      "complete (17,71)\n",
      "complete (17,74)\n",
      "complete (17,73)\n",
      "complete (17,76)\n",
      "complete (17,75)\n",
      "complete (17,77)\n",
      "complete (17,78)\n",
      "complete (17,79)\n",
      "complete (17,80)\n",
      "complete (17,81)\n",
      "complete (17,84)\n",
      "complete (17,82)\n",
      "complete (17,86)\n",
      "complete (17,85)\n",
      "complete (17,83)\n",
      "complete (17,87)\n",
      "complete (17,88)\n",
      "complete (17,89)\n",
      "complete (17,91)\n",
      "complete (17,94)\n",
      "complete (17,92)\n",
      "complete (17,96)\n",
      "complete (17,90)\n",
      "complete (17,93)\n",
      "complete (17,95)\n",
      "complete (17,97)\n",
      "complete (18,19)\n",
      "complete (18,20)\n",
      "complete (18,22)\n",
      "complete (18,21)\n",
      "complete (18,24)\n",
      "complete (17,98)\n",
      "complete (18,25)\n",
      "complete (18,23)\n",
      "complete (18,28)\n",
      "complete (18,27)\n",
      "complete (18,30)\n",
      "complete (18,26)\n",
      "complete (18,33)\n",
      "complete (18,29)\n",
      "complete (18,32)\n",
      "complete (18,31)\n",
      "complete (18,35)\n",
      "complete (18,34)\n",
      "complete (18,38)\n",
      "complete (18,37)\n",
      "complete (18,40)\n",
      "complete (18,39)\n",
      "complete (18,36)\n",
      "complete (18,41)\n",
      "complete (18,43)\n",
      "complete (18,45)\n",
      "complete (18,46)\n",
      "complete (18,48)\n",
      "complete (18,42)\n",
      "complete (18,44)\n",
      "complete (18,47)\n",
      "complete (18,49)\n",
      "complete (18,52)\n",
      "complete (18,50)\n",
      "complete (18,55)\n",
      "complete (18,54)\n",
      "complete (18,53)\n",
      "complete (18,51)\n",
      "complete (18,56)\n",
      "complete (18,57)\n",
      "complete (18,58)\n",
      "complete (18,60)\n",
      "complete (18,64)\n",
      "complete (18,62)\n",
      "complete (18,61)\n",
      "complete (18,63)\n",
      "complete (18,59)\n",
      "complete (18,65)\n",
      "complete (18,66)\n",
      "complete (18,68)\n",
      "complete (18,69)\n",
      "complete (18,67)\n",
      "complete (18,72)\n",
      "complete (18,71)\n",
      "complete (18,70)\n",
      "complete (18,73)\n",
      "complete (18,77)\n",
      "complete (18,78)\n",
      "complete (18,76)\n",
      "complete (18,74)\n",
      "complete (18,75)\n",
      "complete (18,79)\n",
      "complete (18,80)\n",
      "complete (18,81)\n",
      "complete (18,86)\n",
      "complete (18,83)\n",
      "complete (18,82)\n",
      "complete (18,88)\n",
      "complete (18,89)\n",
      "complete (18,87)\n",
      "complete (18,84)\n",
      "complete (18,85)\n",
      "complete (18,90)\n",
      "complete (18,94)\n",
      "complete (18,92)\n",
      "complete (18,91)\n",
      "complete (18,93)\n",
      "complete (18,97)\n",
      "complete (18,95)\n",
      "complete (18,96)\n",
      "complete (19,21)\n",
      "complete (19,22)\n",
      "complete (19,24)\n",
      "complete (19,20)\n",
      "complete (19,23)\n",
      "complete (19,25)\n",
      "complete (19,26)\n",
      "complete (18,98)\n",
      "complete (19,27)\n",
      "complete (19,28)\n",
      "complete (19,32)\n",
      "complete (19,33)\n",
      "complete (19,30)\n",
      "complete (19,31)\n",
      "complete (19,34)\n",
      "complete (19,29)\n",
      "complete (19,35)\n",
      "complete (19,38)\n",
      "complete (19,40)\n",
      "complete (19,39)\n",
      "complete (19,36)\n",
      "complete (19,37)\n",
      "complete (19,41)\n",
      "complete (19,42)\n",
      "complete (19,43)\n",
      "complete (19,44)\n",
      "complete (19,46)\n",
      "complete (19,48)\n",
      "complete (19,47)\n",
      "complete (19,45)\n",
      "complete (19,49)\n",
      "complete (19,50)\n",
      "complete (19,51)\n",
      "complete (19,52)\n",
      "complete (19,54)\n",
      "complete (19,57)\n",
      "complete (19,56)\n",
      "complete (19,55)\n",
      "complete (19,53)\n",
      "complete (19,58)\n",
      "complete (19,60)\n",
      "complete (19,59)\n",
      "complete (19,61)\n",
      "complete (19,63)\n",
      "complete (19,62)\n",
      "complete (19,65)\n",
      "complete (19,64)\n",
      "complete (19,66)\n",
      "complete (19,67)\n",
      "complete (19,70)\n",
      "complete (19,68)\n",
      "complete (19,69)\n",
      "complete (19,72)\n",
      "complete (19,71)\n",
      "complete (19,73)\n",
      "complete (19,74)\n",
      "complete (19,76)\n",
      "complete (19,75)\n",
      "complete (19,77)\n",
      "complete (19,78)\n",
      "complete (19,79)\n",
      "complete (19,81)\n",
      "complete (19,80)\n",
      "complete (19,82)\n",
      "complete (19,85)\n",
      "complete (19,86)\n",
      "complete (19,84)\n",
      "complete (19,83)\n",
      "complete (19,87)\n",
      "complete (19,89)\n",
      "complete (19,88)\n",
      "complete (19,90)\n",
      "complete (19,94)\n",
      "complete (19,91)\n",
      "complete (19,96)\n",
      "complete (19,93)\n",
      "complete (19,92)\n",
      "complete (19,95)\n",
      "complete (19,97)\n",
      "complete (19,98)\n",
      "complete (20,24)\n",
      "complete (20,21)\n",
      "complete (20,22)\n",
      "complete (20,25)\n",
      "complete (20,23)\n",
      "complete (20,26)\n",
      "complete (20,27)\n",
      "complete (20,28)\n",
      "complete (20,32)\n",
      "complete (20,29)\n",
      "complete (20,31)\n",
      "complete (20,30)\n",
      "complete (20,33)\n",
      "complete (20,34)\n",
      "complete (20,35)\n",
      "complete (20,36)\n",
      "complete (20,37)\n",
      "complete (20,39)\n",
      "complete (20,40)\n",
      "complete (20,38)\n",
      "complete (20,41)\n",
      "complete (20,42)\n",
      "complete (20,44)\n",
      "complete (20,43)\n",
      "complete (20,47)\n",
      "complete (20,45)\n",
      "complete (20,48)\n",
      "complete (20,46)\n",
      "complete (20,49)\n",
      "complete (20,50)\n",
      "complete (20,51)\n",
      "complete (20,52)\n",
      "complete (20,53)\n",
      "complete (20,54)\n",
      "complete (20,55)\n",
      "complete (20,56)\n",
      "complete (20,57)\n",
      "complete (20,58)\n",
      "complete (20,60)\n",
      "complete (20,59)\n",
      "complete (20,62)\n",
      "complete (20,63)\n",
      "complete (20,61)\n",
      "complete (20,64)\n",
      "complete (20,65)\n",
      "complete (20,69)\n",
      "complete (20,67)\n",
      "complete (20,70)\n",
      "complete (20,72)\n",
      "complete (20,71)\n",
      "complete (20,66)\n",
      "complete (20,68)\n",
      "complete (20,73)\n",
      "complete (20,76)\n",
      "complete (20,74)\n",
      "complete (20,78)\n",
      "complete (20,77)\n",
      "complete (20,75)\n",
      "complete (20,80)\n",
      "complete (20,79)\n",
      "complete (20,81)\n",
      "complete (20,82)\n",
      "complete (20,84)\n",
      "complete (20,85)\n",
      "complete (20,86)\n",
      "complete (20,83)\n",
      "complete (20,87)\n",
      "complete (20,89)\n",
      "complete (20,88)\n",
      "complete (20,90)\n",
      "complete (20,92)\n",
      "complete (20,91)\n",
      "complete (20,93)\n",
      "complete (20,94)\n",
      "complete (20,96)\n",
      "complete (20,97)\n",
      "complete (20,95)\n",
      "complete (21,22)\n",
      "complete (21,23)\n",
      "complete (21,24)\n",
      "complete (20,98)\n",
      "complete (21,25)\n",
      "complete (21,26)\n",
      "complete (21,27)\n",
      "complete (21,28)\n",
      "complete (21,29)\n",
      "complete (21,32)\n",
      "complete (21,30)\n",
      "complete (21,31)\n",
      "complete (21,33)\n",
      "complete (21,34)\n",
      "complete (21,35)\n",
      "complete (21,37)\n",
      "complete (21,36)\n",
      "complete (21,39)\n",
      "complete (21,38)\n",
      "complete (21,40)\n",
      "complete (21,41)\n",
      "complete (21,42)\n",
      "complete (21,43)\n",
      "complete (21,45)\n",
      "complete (21,44)\n",
      "complete (21,47)\n",
      "complete (21,46)\n",
      "complete (21,48)\n",
      "complete (21,50)\n",
      "complete (21,49)\n",
      "complete (21,51)\n",
      "complete (21,54)\n",
      "complete (21,55)\n",
      "complete (21,52)\n",
      "complete (21,53)\n",
      "complete (21,58)\n",
      "complete (21,56)\n",
      "complete (21,57)\n",
      "complete (21,62)\n",
      "complete (21,59)\n",
      "complete (21,60)\n",
      "complete (21,61)\n",
      "complete (21,64)\n",
      "complete (21,63)\n",
      "complete (21,66)\n",
      "complete (21,65)\n",
      "complete (21,68)\n",
      "complete (21,69)\n",
      "complete (21,67)\n",
      "complete (21,70)\n",
      "complete (21,72)\n",
      "complete (21,71)\n",
      "complete (21,73)\n",
      "complete (21,74)\n",
      "complete (21,77)\n",
      "complete (21,76)\n",
      "complete (21,75)\n",
      "complete (21,78)\n",
      "complete (21,80)\n",
      "complete (21,79)\n",
      "complete (21,81)\n",
      "complete (21,83)\n",
      "complete (21,84)\n",
      "complete (21,82)\n",
      "complete (21,85)\n",
      "complete (21,86)\n",
      "complete (21,87)\n",
      "complete (21,88)\n",
      "complete (21,89)\n",
      "complete (21,90)\n",
      "complete (21,91)\n",
      "complete (21,92)\n",
      "complete (21,93)\n",
      "complete (21,94)\n",
      "complete (21,95)\n",
      "complete (21,96)\n",
      "complete (21,97)\n",
      "complete (22,24)\n",
      "complete (21,98)\n",
      "complete (22,25)\n",
      "complete (22,23)\n",
      "complete (22,26)\n",
      "complete (22,28)\n",
      "complete (22,27)\n",
      "complete (22,30)\n",
      "complete (22,31)\n",
      "complete (22,29)\n",
      "complete (22,32)\n",
      "complete (22,33)\n",
      "complete (22,34)\n",
      "complete (22,35)\n",
      "complete (22,36)\n",
      "complete (22,38)\n",
      "complete (22,39)\n",
      "complete (22,40)\n",
      "complete (22,37)\n",
      "complete (22,42)\n",
      "complete (22,41)\n",
      "complete (22,43)\n",
      "complete (22,44)\n",
      "complete (22,47)\n",
      "complete (22,48)\n",
      "complete (22,45)\n",
      "complete (22,46)\n",
      "complete (22,50)\n",
      "complete (22,49)\n",
      "complete (22,52)\n",
      "complete (22,51)\n",
      "complete (22,53)\n",
      "complete (22,54)\n",
      "complete (22,56)\n",
      "complete (22,55)\n",
      "complete (22,57)\n",
      "complete (22,58)\n",
      "complete (22,60)\n",
      "complete (22,59)\n",
      "complete (22,61)\n",
      "complete (22,64)\n",
      "complete (22,62)\n",
      "complete (22,63)\n",
      "complete (22,65)\n",
      "complete (22,66)\n",
      "complete (22,67)\n",
      "complete (22,70)\n",
      "complete (22,68)\n",
      "complete (22,69)\n",
      "complete (22,72)\n",
      "complete (22,71)\n",
      "complete (22,73)\n",
      "complete (22,74)\n",
      "complete (22,75)\n",
      "complete (22,77)\n",
      "complete (22,79)\n",
      "complete (22,76)\n",
      "complete (22,78)\n",
      "complete (22,80)\n",
      "complete (22,81)\n",
      "complete (22,82)\n",
      "complete (22,84)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "complete (22,87)\n",
      "complete (22,86)\n",
      "complete (22,85)\n",
      "complete (22,83)\n",
      "complete (22,88)\n",
      "complete (22,89)\n",
      "complete (22,90)\n",
      "complete (22,92)\n",
      "complete (22,94)\n",
      "complete (22,96)\n",
      "complete (22,91)\n",
      "complete (22,93)\n",
      "complete (22,95)\n",
      "complete (22,97)\n",
      "complete (22,98)\n",
      "complete (23,25)\n",
      "complete (23,24)\n",
      "complete (23,28)\n",
      "complete (23,26)\n",
      "complete (23,27)\n",
      "complete (23,29)\n",
      "complete (23,30)\n",
      "complete (23,31)\n",
      "complete (23,32)\n",
      "complete (23,34)\n",
      "complete (23,33)\n",
      "complete (23,36)\n",
      "complete (23,35)\n",
      "complete (23,37)\n",
      "complete (23,39)\n",
      "complete (23,38)\n",
      "complete (23,40)\n",
      "complete (23,41)\n",
      "complete (23,42)\n",
      "complete (23,44)\n",
      "complete (23,45)\n",
      "complete (23,43)\n",
      "complete (23,47)\n",
      "complete (23,46)\n",
      "complete (23,48)\n",
      "complete (23,49)\n",
      "complete (23,54)\n",
      "complete (23,50)\n",
      "complete (23,51)\n",
      "complete (23,53)\n",
      "complete (23,55)\n",
      "complete (23,52)\n",
      "complete (23,56)\n",
      "complete (23,57)\n",
      "complete (23,58)\n",
      "complete (23,60)\n",
      "complete (23,62)\n",
      "complete (23,59)\n",
      "complete (23,61)\n",
      "complete (23,64)\n",
      "complete (23,63)\n",
      "complete (23,65)\n",
      "complete (23,66)\n",
      "complete (23,67)\n",
      "complete (23,68)\n",
      "complete (23,69)\n",
      "complete (23,70)\n",
      "complete (23,71)\n",
      "complete (23,72)\n",
      "complete (23,73)\n",
      "complete (23,74)\n",
      "complete (23,76)\n",
      "complete (23,77)\n",
      "complete (23,78)\n",
      "complete (23,75)\n",
      "complete (23,80)\n",
      "complete (23,79)\n",
      "complete (23,81)\n",
      "complete (23,82)\n",
      "complete (23,85)\n",
      "complete (23,83)\n",
      "complete (23,84)\n",
      "complete (23,87)\n",
      "complete (23,86)\n",
      "complete (23,88)\n",
      "complete (23,89)\n",
      "complete (23,90)\n",
      "complete (23,94)\n",
      "complete (23,92)\n",
      "complete (23,95)\n",
      "complete (23,91)\n",
      "complete (23,93)\n",
      "complete (23,96)\n",
      "complete (23,97)\n",
      "complete (24,25)\n",
      "complete (24,26)\n",
      "complete (24,27)\n",
      "complete (24,28)\n",
      "complete (24,29)\n",
      "complete (23,98)\n",
      "complete (24,30)\n",
      "complete (24,31)\n",
      "complete (24,33)\n",
      "complete (24,32)\n",
      "complete (24,35)\n",
      "complete (24,34)\n",
      "complete (24,36)\n",
      "complete (24,38)\n",
      "complete (24,37)\n",
      "complete (24,39)\n",
      "complete (24,40)\n",
      "complete (24,41)\n",
      "complete (24,43)\n",
      "complete (24,42)\n",
      "complete (24,45)\n",
      "complete (24,44)\n",
      "complete (24,46)\n",
      "complete (24,47)\n",
      "complete (24,48)\n",
      "complete (24,51)\n",
      "complete (24,50)\n",
      "complete (24,52)\n",
      "complete (24,49)\n",
      "complete (24,54)\n",
      "complete (24,53)\n",
      "complete (24,55)\n",
      "complete (24,56)\n",
      "complete (24,57)\n",
      "complete (24,59)\n",
      "complete (24,58)\n",
      "complete (24,60)\n",
      "complete (24,61)\n",
      "complete (24,62)\n",
      "complete (24,63)\n",
      "complete (24,64)\n",
      "complete (24,65)\n",
      "complete (24,66)\n",
      "complete (24,67)\n",
      "complete (24,69)\n",
      "complete (24,70)\n",
      "complete (24,68)\n",
      "complete (24,71)\n",
      "complete (24,72)\n",
      "complete (24,74)\n",
      "complete (24,73)\n",
      "complete (24,75)\n",
      "complete (24,77)\n",
      "complete (24,78)\n",
      "complete (24,76)\n",
      "complete (24,79)\n",
      "complete (24,81)\n",
      "complete (24,80)\n",
      "complete (24,82)\n",
      "complete (24,83)\n",
      "complete (24,84)\n",
      "complete (24,85)\n",
      "complete (24,86)\n",
      "complete (24,87)\n",
      "complete (24,88)\n",
      "complete (24,89)\n",
      "complete (24,92)\n",
      "complete (24,90)\n",
      "complete (24,93)\n",
      "complete (24,91)\n",
      "complete (24,94)\n",
      "complete (24,95)\n",
      "complete (24,97)\n",
      "complete (24,96)\n",
      "complete (25,26)\n",
      "complete (24,98)\n",
      "complete (25,27)\n",
      "complete (25,28)\n",
      "complete (25,29)\n",
      "complete (25,30)\n",
      "complete (25,31)\n",
      "complete (25,34)\n",
      "complete (25,32)\n",
      "complete (25,33)\n",
      "complete (25,35)\n",
      "complete (25,36)\n",
      "complete (25,37)\n",
      "complete (25,38)\n",
      "complete (25,41)\n",
      "complete (25,40)\n",
      "complete (25,39)\n",
      "complete (25,43)\n",
      "complete (25,42)\n",
      "complete (25,44)\n",
      "complete (25,45)\n",
      "complete (25,47)\n",
      "complete (25,46)\n",
      "complete (25,48)\n",
      "complete (25,49)\n",
      "complete (25,50)\n",
      "complete (25,52)\n",
      "complete (25,53)\n",
      "complete (25,54)\n",
      "complete (25,51)\n",
      "complete (25,55)\n",
      "complete (25,56)\n",
      "complete (25,57)\n",
      "complete (25,62)\n",
      "complete (25,58)\n",
      "complete (25,61)\n",
      "complete (25,60)\n",
      "complete (25,59)\n",
      "complete (25,64)\n",
      "complete (25,63)\n",
      "complete (25,65)\n",
      "complete (25,69)\n",
      "complete (25,66)\n",
      "complete (25,68)\n",
      "complete (25,70)\n",
      "complete (25,67)\n",
      "complete (25,72)\n",
      "complete (25,71)\n",
      "complete (25,73)\n",
      "complete (25,75)\n",
      "complete (25,74)\n",
      "complete (25,77)\n",
      "complete (25,78)\n",
      "complete (25,76)\n",
      "complete (25,80)\n",
      "complete (25,81)\n",
      "complete (25,79)\n",
      "complete (25,82)\n",
      "complete (25,83)\n",
      "complete (25,85)\n",
      "complete (25,84)\n",
      "complete (25,86)\n",
      "complete (25,87)\n",
      "complete (25,88)\n",
      "complete (25,89)\n",
      "complete (25,90)\n",
      "complete (25,92)\n",
      "complete (25,93)\n",
      "complete (25,91)\n",
      "complete (25,94)\n",
      "complete (25,96)\n",
      "complete (25,95)\n",
      "complete (25,97)\n",
      "complete (25,98)\n",
      "complete (26,27)\n",
      "complete (26,29)\n",
      "complete (26,28)\n",
      "complete (26,32)\n",
      "complete (26,30)\n",
      "complete (26,31)\n",
      "complete (26,33)\n",
      "complete (26,34)\n",
      "complete (26,35)\n",
      "complete (26,36)\n",
      "complete (26,39)\n",
      "complete (26,37)\n",
      "complete (26,38)\n",
      "complete (26,40)\n",
      "complete (26,41)\n",
      "complete (26,42)\n",
      "complete (26,43)\n",
      "complete (26,44)\n",
      "complete (26,48)\n",
      "complete (26,47)\n",
      "complete (26,45)\n",
      "complete (26,46)\n",
      "complete (26,49)\n",
      "complete (26,50)\n",
      "complete (26,51)\n",
      "complete (26,52)\n",
      "complete (26,54)\n",
      "complete (26,53)\n",
      "complete (26,55)\n",
      "complete (26,56)\n",
      "complete (26,58)\n",
      "complete (26,57)\n",
      "complete (26,59)\n",
      "complete (26,60)\n",
      "complete (26,61)\n",
      "complete (26,62)\n",
      "complete (26,64)\n",
      "complete (26,63)\n",
      "complete (26,65)\n",
      "complete (26,66)\n",
      "complete (26,67)\n",
      "complete (26,68)\n",
      "complete (26,70)\n",
      "complete (26,69)\n",
      "complete (26,71)\n",
      "complete (26,72)\n",
      "complete (26,73)\n",
      "complete (26,74)\n",
      "complete (26,76)\n",
      "complete (26,75)\n",
      "complete (26,78)\n",
      "complete (26,77)\n",
      "complete (26,79)\n",
      "complete (26,81)\n",
      "complete (26,80)\n",
      "complete (26,82)\n",
      "complete (26,83)\n",
      "complete (26,86)\n",
      "complete (26,84)\n",
      "complete (26,85)\n",
      "complete (26,87)\n",
      "complete (26,89)\n",
      "complete (26,88)\n",
      "complete (26,90)\n",
      "complete (26,93)\n",
      "complete (26,92)\n",
      "complete (26,94)\n",
      "complete (26,91)\n",
      "complete (26,96)\n",
      "complete (26,95)\n",
      "complete (26,97)\n",
      "complete (26,98)\n",
      "complete (27,30)\n",
      "complete (27,28)\n",
      "complete (27,32)\n",
      "complete (27,29)\n",
      "complete (27,33)\n",
      "complete (27,31)\n",
      "complete (27,34)\n",
      "complete (27,35)\n",
      "complete (27,38)\n",
      "complete (27,36)\n",
      "complete (27,37)\n",
      "complete (27,40)\n",
      "complete (27,39)\n",
      "complete (27,41)\n",
      "complete (27,42)\n",
      "complete (27,43)\n",
      "complete (27,44)\n",
      "complete (27,45)\n",
      "complete (27,47)\n",
      "complete (27,46)\n",
      "complete (27,48)\n",
      "complete (27,49)\n",
      "complete (27,50)\n",
      "complete (27,51)\n",
      "complete (27,54)\n",
      "complete (27,52)\n",
      "complete (27,53)\n",
      "complete (27,55)\n",
      "complete (27,56)\n",
      "complete (27,57)\n",
      "complete (27,59)\n",
      "complete (27,58)\n",
      "complete (27,60)\n",
      "complete (27,62)\n",
      "complete (27,61)\n",
      "complete (27,63)\n",
      "complete (27,64)\n",
      "complete (27,65)\n",
      "complete (27,66)\n",
      "complete (27,67)\n",
      "complete (27,68)\n",
      "complete (27,69)\n",
      "complete (27,70)\n",
      "complete (27,72)\n",
      "complete (27,71)\n",
      "complete (27,73)\n",
      "complete (27,74)\n",
      "complete (27,75)\n",
      "complete (27,76)\n",
      "complete (27,77)\n",
      "complete (27,78)\n",
      "complete (27,79)\n",
      "complete (27,80)\n",
      "complete (27,81)\n",
      "complete (27,82)\n",
      "complete (27,85)\n",
      "complete (27,84)\n",
      "complete (27,83)\n",
      "complete (27,86)\n",
      "complete (27,88)\n",
      "complete (27,87)\n",
      "complete (27,89)\n",
      "complete (27,90)\n",
      "complete (27,91)\n",
      "complete (27,94)\n",
      "complete (27,92)\n",
      "complete (27,93)\n",
      "complete (27,96)\n",
      "complete (27,95)\n",
      "complete (27,97)\n",
      "complete (28,32)\n",
      "complete (28,29)\n",
      "complete (28,30)\n",
      "complete (28,31)\n",
      "complete (27,98)\n",
      "complete (28,34)\n",
      "complete (28,33)\n",
      "complete (28,35)\n",
      "complete (28,37)\n",
      "complete (28,39)\n",
      "complete (28,36)\n",
      "complete (28,40)\n",
      "complete (28,38)\n",
      "complete (28,42)\n",
      "complete (28,41)\n",
      "complete (28,43)\n",
      "complete (28,44)\n",
      "complete (28,47)\n",
      "complete (28,46)\n",
      "complete (28,45)\n",
      "complete (28,48)\n",
      "complete (28,49)\n",
      "complete (28,50)\n",
      "complete (28,51)\n",
      "complete (28,52)\n",
      "complete (28,54)\n",
      "complete (28,55)\n",
      "complete (28,56)\n",
      "complete (28,53)\n",
      "complete (28,57)\n",
      "complete (28,58)\n",
      "complete (28,59)\n",
      "complete (28,62)\n",
      "complete (28,60)\n",
      "complete (28,61)\n",
      "complete (28,64)\n",
      "complete (28,63)\n",
      "complete (28,65)\n",
      "complete (28,66)\n",
      "complete (28,68)\n",
      "complete (28,67)\n",
      "complete (28,70)\n",
      "complete (28,69)\n",
      "complete (28,71)\n",
      "complete (28,72)\n",
      "complete (28,73)\n",
      "complete (28,74)\n",
      "complete (28,77)\n",
      "complete (28,76)\n",
      "complete (28,75)\n",
      "complete (28,78)\n",
      "complete (28,79)\n",
      "complete (28,80)\n",
      "complete (28,81)\n",
      "complete (28,82)\n",
      "complete (28,86)\n",
      "complete (28,85)\n",
      "complete (28,83)\n",
      "complete (28,84)\n",
      "complete (28,87)\n",
      "complete (28,88)\n",
      "complete (28,89)\n",
      "complete (28,90)\n",
      "complete (28,91)\n",
      "complete (28,94)\n",
      "complete (28,93)\n",
      "complete (28,92)\n",
      "complete (28,95)\n",
      "complete (28,96)\n",
      "complete (28,97)\n",
      "complete (29,31)\n",
      "complete (28,98)\n",
      "complete (29,30)\n",
      "complete (29,32)\n",
      "complete (29,34)\n",
      "complete (29,35)\n",
      "complete (29,33)\n",
      "complete (29,36)\n",
      "complete (29,37)\n",
      "complete (29,39)\n",
      "complete (29,41)\n",
      "complete (29,38)\n",
      "complete (29,40)\n",
      "complete (29,43)\n",
      "complete (29,42)\n",
      "complete (29,44)\n",
      "complete (29,45)\n",
      "complete (29,47)\n",
      "complete (29,48)\n",
      "complete (29,46)\n",
      "complete (29,49)\n",
      "complete (29,52)\n",
      "complete (29,50)\n",
      "complete (29,51)\n",
      "complete (29,54)\n",
      "complete (29,55)\n",
      "complete (29,53)\n",
      "complete (29,56)\n",
      "complete (29,57)\n",
      "complete (29,58)\n",
      "complete (29,59)\n",
      "complete (29,60)\n",
      "complete (29,62)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "complete (29,61)\n",
      "complete (29,64)\n",
      "complete (29,63)\n",
      "complete (29,66)\n",
      "complete (29,65)\n",
      "complete (29,67)\n",
      "complete (29,68)\n",
      "complete (29,69)\n",
      "complete (29,70)\n",
      "complete (29,72)\n",
      "complete (29,71)\n",
      "complete (29,73)\n",
      "complete (29,77)\n",
      "complete (29,76)\n",
      "complete (29,75)\n",
      "complete (29,74)\n",
      "complete (29,78)\n",
      "complete (29,79)\n",
      "complete (29,80)\n",
      "complete (29,81)\n",
      "complete (29,83)\n",
      "complete (29,82)\n",
      "complete (29,84)\n",
      "complete (29,85)\n",
      "complete (29,87)\n",
      "complete (29,86)\n",
      "complete (29,88)\n",
      "complete (29,89)\n",
      "complete (29,90)\n",
      "complete (29,92)\n",
      "complete (29,93)\n",
      "complete (29,91)\n",
      "complete (29,94)\n",
      "complete (29,96)\n",
      "complete (29,97)\n",
      "complete (29,95)\n",
      "complete (30,33)\n",
      "complete (30,31)\n",
      "complete (30,32)\n",
      "complete (29,98)\n",
      "complete (30,34)\n",
      "complete (30,36)\n",
      "complete (30,35)\n",
      "complete (30,37)\n",
      "complete (30,39)\n",
      "complete (30,38)\n",
      "complete (30,40)\n",
      "complete (30,41)\n",
      "complete (30,43)\n",
      "complete (30,42)\n",
      "complete (30,44)\n",
      "complete (30,45)\n",
      "complete (30,46)\n",
      "complete (30,47)\n",
      "complete (30,48)\n",
      "complete (30,49)\n",
      "complete (30,52)\n",
      "complete (30,50)\n",
      "complete (30,51)\n",
      "complete (30,53)\n",
      "complete (30,54)\n",
      "complete (30,55)\n",
      "complete (30,56)\n",
      "complete (30,57)\n",
      "complete (30,59)\n",
      "complete (30,58)\n",
      "complete (30,60)\n",
      "complete (30,61)\n",
      "complete (30,62)\n",
      "complete (30,63)\n",
      "complete (30,64)\n",
      "complete (30,65)\n",
      "complete (30,67)\n",
      "complete (30,68)\n",
      "complete (30,66)\n",
      "complete (30,69)\n",
      "complete (30,70)\n",
      "complete (30,71)\n",
      "complete (30,72)\n",
      "complete (30,73)\n",
      "complete (30,76)\n",
      "complete (30,77)\n",
      "complete (30,74)\n",
      "complete (30,75)\n",
      "complete (30,78)\n",
      "complete (30,79)\n",
      "complete (30,80)\n",
      "complete (30,81)\n",
      "complete (30,82)\n",
      "complete (30,83)\n",
      "complete (30,84)\n",
      "complete (30,86)\n",
      "complete (30,85)\n",
      "complete (30,87)\n",
      "complete (30,89)\n",
      "complete (30,88)\n",
      "complete (30,91)\n",
      "complete (30,90)\n",
      "complete (30,92)\n",
      "complete (30,94)\n",
      "complete (30,93)\n",
      "complete (30,95)\n",
      "complete (30,96)\n",
      "complete (30,97)\n",
      "complete (31,32)\n",
      "complete (31,33)\n",
      "complete (30,98)\n",
      "complete (31,34)\n",
      "complete (31,35)\n",
      "complete (31,37)\n",
      "complete (31,36)\n",
      "complete (31,38)\n",
      "complete (31,40)\n",
      "complete (31,39)\n",
      "complete (31,42)\n",
      "complete (31,41)\n",
      "complete (31,43)\n",
      "complete (31,44)\n",
      "complete (31,45)\n",
      "complete (31,47)\n",
      "complete (31,46)\n",
      "complete (31,48)\n",
      "complete (31,50)\n",
      "complete (31,49)\n",
      "complete (31,51)\n",
      "complete (31,54)\n",
      "complete (31,52)\n",
      "complete (31,53)\n",
      "complete (31,55)\n",
      "complete (31,56)\n",
      "complete (31,58)\n",
      "complete (31,57)\n",
      "complete (31,59)\n",
      "complete (31,60)\n",
      "complete (31,61)\n",
      "complete (31,62)\n",
      "complete (31,63)\n",
      "complete (31,64)\n",
      "complete (31,65)\n",
      "complete (31,66)\n",
      "complete (31,67)\n",
      "complete (31,69)\n",
      "complete (31,70)\n",
      "complete (31,68)\n",
      "complete (31,71)\n",
      "complete (31,72)\n",
      "complete (31,73)\n",
      "complete (31,74)\n",
      "complete (31,76)\n",
      "complete (31,77)\n",
      "complete (31,75)\n",
      "complete (31,78)\n",
      "complete (31,79)\n",
      "complete (31,80)\n",
      "complete (31,81)\n",
      "complete (31,85)\n",
      "complete (31,84)\n",
      "complete (31,82)\n",
      "complete (31,86)\n",
      "complete (31,83)\n",
      "complete (31,87)\n",
      "complete (31,88)\n",
      "complete (31,89)\n",
      "complete (31,92)\n",
      "complete (31,90)\n",
      "complete (31,91)\n",
      "complete (31,93)\n",
      "complete (31,94)\n",
      "complete (31,95)\n",
      "complete (31,97)\n",
      "complete (31,96)\n",
      "complete (32,33)\n",
      "complete (32,34)\n",
      "complete (32,35)\n",
      "complete (32,36)\n",
      "complete (31,98)\n",
      "complete (32,37)\n",
      "complete (32,38)\n",
      "complete (32,39)\n",
      "complete (32,40)\n",
      "complete (32,41)\n",
      "complete (32,43)\n",
      "complete (32,42)\n",
      "complete (32,45)\n",
      "complete (32,44)\n",
      "complete (32,46)\n",
      "complete (32,47)\n",
      "complete (32,48)\n",
      "complete (32,49)\n",
      "complete (32,50)\n",
      "complete (32,52)\n",
      "complete (32,51)\n",
      "complete (32,53)\n",
      "complete (32,54)\n",
      "complete (32,55)\n",
      "complete (32,56)\n",
      "complete (32,57)\n",
      "complete (32,58)\n",
      "complete (32,59)\n",
      "complete (32,62)\n",
      "complete (32,61)\n",
      "complete (32,63)\n",
      "complete (32,60)\n",
      "complete (32,64)\n",
      "complete (32,65)\n",
      "complete (32,67)\n",
      "complete (32,66)\n",
      "complete (32,69)\n",
      "complete (32,68)\n",
      "complete (32,70)\n",
      "complete (32,71)\n",
      "complete (32,72)\n",
      "complete (32,73)\n",
      "complete (32,76)\n",
      "complete (32,74)\n",
      "complete (32,75)\n",
      "complete (32,77)\n",
      "complete (32,78)\n",
      "complete (32,79)\n",
      "complete (32,80)\n",
      "complete (32,81)\n",
      "complete (32,82)\n",
      "complete (32,84)\n",
      "complete (32,85)\n",
      "complete (32,83)\n",
      "complete (32,86)\n",
      "complete (32,87)\n",
      "complete (32,88)\n",
      "complete (32,89)\n",
      "complete (32,92)\n",
      "complete (32,90)\n",
      "complete (32,93)\n",
      "complete (32,91)\n",
      "complete (32,94)\n",
      "complete (32,95)\n",
      "complete (32,96)\n",
      "complete (32,97)\n",
      "complete (33,34)\n",
      "complete (32,98)\n",
      "complete (33,35)\n",
      "complete (33,37)\n",
      "complete (33,36)\n",
      "complete (33,38)\n",
      "complete (33,39)\n",
      "complete (33,40)\n",
      "complete (33,41)\n",
      "complete (33,43)\n",
      "complete (33,42)\n",
      "complete (33,45)\n",
      "complete (33,44)\n",
      "complete (33,47)\n",
      "complete (33,46)\n",
      "complete (33,48)\n",
      "complete (33,50)\n",
      "complete (33,49)\n",
      "complete (33,52)\n",
      "complete (33,51)\n",
      "complete (33,54)\n",
      "complete (33,53)\n",
      "complete (33,55)\n",
      "complete (33,56)\n",
      "complete (33,58)\n",
      "complete (33,57)\n",
      "complete (33,60)\n",
      "complete (33,61)\n",
      "complete (33,62)\n",
      "complete (33,59)\n",
      "complete (33,63)\n",
      "complete (33,64)\n",
      "complete (33,65)\n",
      "complete (33,66)\n",
      "complete (33,68)\n",
      "complete (33,67)\n",
      "complete (33,69)\n",
      "complete (33,70)\n",
      "complete (33,71)\n",
      "complete (33,72)\n",
      "complete (33,73)\n",
      "complete (33,75)\n",
      "complete (33,74)\n",
      "complete (33,76)\n",
      "complete (33,77)\n",
      "complete (33,78)\n",
      "complete (33,79)\n",
      "complete (33,80)\n",
      "complete (33,81)\n",
      "complete (33,82)\n",
      "complete (33,85)\n",
      "complete (33,84)\n",
      "complete (33,86)\n",
      "complete (33,83)\n",
      "complete (33,87)\n",
      "complete (33,89)\n",
      "complete (33,88)\n",
      "complete (33,90)\n",
      "complete (33,91)\n",
      "complete (33,93)\n",
      "complete (33,92)\n",
      "complete (33,96)\n",
      "complete (33,94)\n",
      "complete (33,95)\n",
      "complete (33,97)\n",
      "complete (34,35)\n",
      "complete (34,36)\n",
      "complete (33,98)\n",
      "complete (34,37)\n",
      "complete (34,39)\n",
      "complete (34,38)\n",
      "complete (34,41)\n",
      "complete (34,40)\n",
      "complete (34,42)\n",
      "complete (34,43)\n",
      "complete (34,46)\n",
      "complete (34,44)\n",
      "complete (34,47)\n",
      "complete (34,45)\n",
      "complete (34,48)\n",
      "complete (34,49)\n",
      "complete (34,54)\n",
      "complete (34,52)\n",
      "complete (34,51)\n",
      "complete (34,50)\n",
      "complete (34,53)\n",
      "complete (34,55)\n",
      "complete (34,56)\n",
      "complete (34,57)\n",
      "complete (34,58)\n",
      "complete (34,59)\n",
      "complete (34,61)\n",
      "complete (34,62)\n",
      "complete (34,60)\n",
      "complete (34,64)\n",
      "complete (34,63)\n",
      "complete (34,65)\n",
      "complete (34,67)\n",
      "complete (34,66)\n",
      "complete (34,69)\n",
      "complete (34,68)\n",
      "complete (34,70)\n",
      "complete (34,71)\n",
      "complete (34,72)\n",
      "complete (34,73)\n",
      "complete (34,77)\n",
      "complete (34,76)\n",
      "complete (34,74)\n",
      "complete (34,75)\n",
      "complete (34,78)\n",
      "complete (34,80)\n",
      "complete (34,79)\n",
      "complete (34,81)\n",
      "complete (34,84)\n",
      "complete (34,82)\n",
      "complete (34,83)\n",
      "complete (34,85)\n",
      "complete (34,87)\n",
      "complete (34,86)\n",
      "complete (34,88)\n",
      "complete (34,89)\n",
      "complete (34,90)\n",
      "complete (34,92)\n",
      "complete (34,91)\n",
      "complete (34,93)\n",
      "complete (34,94)\n",
      "complete (34,95)\n",
      "complete (34,96)\n",
      "complete (34,97)\n",
      "complete (34,98)\n",
      "complete (35,36)\n",
      "complete (35,39)\n",
      "complete (35,37)\n",
      "complete (35,38)\n",
      "complete (35,40)\n",
      "complete (35,42)\n",
      "complete (35,41)\n",
      "complete (35,43)\n",
      "complete (35,44)\n",
      "complete (35,46)\n",
      "complete (35,47)\n",
      "complete (35,45)\n",
      "complete (35,48)\n",
      "complete (35,49)\n",
      "complete (35,50)\n",
      "complete (35,52)\n",
      "complete (35,51)\n",
      "complete (35,54)\n",
      "complete (35,55)\n",
      "complete (35,53)\n",
      "complete (35,56)\n",
      "complete (35,57)\n",
      "complete (35,58)\n",
      "complete (35,59)\n",
      "complete (35,62)\n",
      "complete (35,60)\n",
      "complete (35,61)\n",
      "complete (35,64)\n",
      "complete (35,63)\n",
      "complete (35,65)\n",
      "complete (35,66)\n",
      "complete (35,67)\n",
      "complete (35,68)\n",
      "complete (35,69)\n",
      "complete (35,70)\n",
      "complete (35,71)\n",
      "complete (35,72)\n",
      "complete (35,73)\n",
      "complete (35,77)\n",
      "complete (35,74)\n",
      "complete (35,76)\n",
      "complete (35,75)\n",
      "complete (35,78)\n",
      "complete (35,79)\n",
      "complete (35,80)\n",
      "complete (35,81)\n",
      "complete (35,84)\n",
      "complete (35,83)\n",
      "complete (35,82)\n",
      "complete (35,85)\n",
      "complete (35,86)\n",
      "complete (35,87)\n",
      "complete (35,88)\n",
      "complete (35,89)\n",
      "complete (35,90)\n",
      "complete (35,92)\n",
      "complete (35,91)\n",
      "complete (35,93)\n",
      "complete (35,94)\n",
      "complete (35,96)\n",
      "complete (35,97)\n",
      "complete (35,95)\n",
      "complete (36,38)\n",
      "complete (36,37)\n",
      "complete (36,39)\n",
      "complete (35,98)\n",
      "complete (36,40)\n",
      "complete (36,41)\n",
      "complete (36,42)\n",
      "complete (36,43)\n",
      "complete (36,46)\n",
      "complete (36,47)\n",
      "complete (36,44)\n",
      "complete (36,45)\n",
      "complete (36,48)\n",
      "complete (36,49)\n",
      "complete (36,50)\n",
      "complete (36,51)\n",
      "complete (36,52)\n",
      "complete (36,54)\n",
      "complete (36,53)\n",
      "complete (36,56)\n",
      "complete (36,55)\n",
      "complete (36,57)\n",
      "complete (36,58)\n",
      "complete (36,60)\n",
      "complete (36,59)\n",
      "complete (36,61)\n",
      "complete (36,62)\n",
      "complete (36,64)\n",
      "complete (36,63)\n",
      "complete (36,65)\n",
      "complete (36,67)\n",
      "complete (36,66)\n",
      "complete (36,68)\n",
      "complete (36,69)\n",
      "complete (36,71)\n",
      "complete (36,70)\n",
      "complete (36,72)\n",
      "complete (36,73)\n",
      "complete (36,74)\n",
      "complete (36,77)\n",
      "complete (36,76)\n",
      "complete (36,75)\n",
      "complete (36,78)\n",
      "complete (36,81)\n",
      "complete (36,79)\n",
      "complete (36,80)\n",
      "complete (36,83)\n",
      "complete (36,82)\n",
      "complete (36,84)\n",
      "complete (36,85)\n",
      "complete (36,86)\n",
      "complete (36,89)\n",
      "complete (36,87)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "complete (36,88)\n",
      "complete (36,90)\n",
      "complete (36,92)\n",
      "complete (36,93)\n",
      "complete (36,91)\n",
      "complete (36,94)\n",
      "complete (36,96)\n",
      "complete (36,97)\n",
      "complete (36,95)\n",
      "complete (36,98)\n",
      "complete (37,40)\n",
      "complete (37,39)\n",
      "complete (37,41)\n",
      "complete (37,38)\n",
      "complete (37,42)\n",
      "complete (37,43)\n",
      "complete (37,44)\n",
      "complete (37,47)\n",
      "complete (37,45)\n",
      "complete (37,46)\n",
      "complete (37,48)\n",
      "complete (37,50)\n",
      "complete (37,49)\n",
      "complete (37,51)\n",
      "complete (37,52)\n",
      "complete (37,54)\n",
      "complete (37,53)\n",
      "complete (37,55)\n",
      "complete (37,56)\n",
      "complete (37,57)\n",
      "complete (37,58)\n",
      "complete (37,60)\n",
      "complete (37,59)\n",
      "complete (37,62)\n",
      "complete (37,61)\n",
      "complete (37,63)\n",
      "complete (37,64)\n",
      "complete (37,65)\n",
      "complete (37,68)\n",
      "complete (37,67)\n",
      "complete (37,69)\n",
      "complete (37,66)\n",
      "complete (37,70)\n",
      "complete (37,72)\n",
      "complete (37,71)\n",
      "complete (37,73)\n",
      "complete (37,76)\n",
      "complete (37,74)\n",
      "complete (37,77)\n",
      "complete (37,75)\n",
      "complete (37,78)\n",
      "complete (37,79)\n",
      "complete (37,80)\n",
      "complete (37,81)\n",
      "complete (37,83)\n",
      "complete (37,85)\n",
      "complete (37,82)\n",
      "complete (37,84)\n",
      "complete (37,86)\n",
      "complete (37,87)\n",
      "complete (37,88)\n",
      "complete (37,89)\n",
      "complete (37,92)\n",
      "complete (37,90)\n",
      "complete (37,91)\n",
      "complete (37,93)\n",
      "complete (37,94)\n",
      "complete (37,95)\n",
      "complete (37,96)\n",
      "complete (37,97)\n",
      "complete (38,39)\n",
      "complete (38,40)\n",
      "complete (38,41)\n",
      "complete (37,98)\n",
      "complete (38,42)\n",
      "complete (38,43)\n",
      "complete (38,44)\n",
      "complete (38,47)\n",
      "complete (38,45)\n",
      "complete (38,46)\n",
      "complete (38,48)\n",
      "complete (38,50)\n",
      "complete (38,49)\n",
      "complete (38,51)\n",
      "complete (38,52)\n",
      "complete (38,54)\n",
      "complete (38,55)\n",
      "complete (38,53)\n",
      "complete (38,56)\n",
      "complete (38,57)\n",
      "complete (38,58)\n",
      "complete (38,59)\n",
      "complete (38,60)\n",
      "complete (38,62)\n",
      "complete (38,61)\n",
      "complete (38,64)\n",
      "complete (38,63)\n",
      "complete (38,65)\n",
      "complete (38,66)\n",
      "complete (38,67)\n",
      "complete (38,68)\n",
      "complete (38,69)\n",
      "complete (38,70)\n",
      "complete (38,71)\n",
      "complete (38,72)\n",
      "complete (38,73)\n",
      "complete (38,74)\n",
      "complete (38,75)\n",
      "complete (38,76)\n",
      "complete (38,77)\n",
      "complete (38,78)\n",
      "complete (38,79)\n",
      "complete (38,80)\n",
      "complete (38,81)\n",
      "complete (38,84)\n",
      "complete (38,82)\n",
      "complete (38,83)\n",
      "complete (38,85)\n",
      "complete (38,87)\n",
      "complete (38,86)\n",
      "complete (38,88)\n",
      "complete (38,89)\n",
      "complete (38,92)\n",
      "complete (38,90)\n",
      "complete (38,93)\n",
      "complete (38,94)\n",
      "complete (38,91)\n",
      "complete (38,95)\n",
      "complete (38,96)\n",
      "complete (38,97)\n",
      "complete (39,40)\n",
      "complete (39,41)\n",
      "complete (39,43)\n",
      "complete (38,98)\n",
      "complete (39,42)\n",
      "complete (39,44)\n",
      "complete (39,45)\n",
      "complete (39,46)\n",
      "complete (39,47)\n",
      "complete (39,48)\n",
      "complete (39,52)\n",
      "complete (39,49)\n",
      "complete (39,51)\n",
      "complete (39,50)\n",
      "complete (39,54)\n",
      "complete (39,53)\n",
      "complete (39,55)\n",
      "complete (39,56)\n",
      "complete (39,58)\n",
      "complete (39,57)\n",
      "complete (39,60)\n",
      "complete (39,59)\n",
      "complete (39,61)\n",
      "complete (39,62)\n",
      "complete (39,63)\n",
      "complete (39,64)\n",
      "complete (39,69)\n",
      "complete (39,67)\n",
      "complete (39,65)\n",
      "complete (39,68)\n",
      "complete (39,70)\n",
      "complete (39,66)\n",
      "complete (39,71)\n",
      "complete (39,72)\n",
      "complete (39,73)\n",
      "complete (39,74)\n",
      "complete (39,77)\n",
      "complete (39,75)\n",
      "complete (39,78)\n",
      "complete (39,76)\n",
      "complete (39,79)\n",
      "complete (39,80)\n",
      "complete (39,81)\n",
      "complete (39,83)\n",
      "complete (39,86)\n",
      "complete (39,85)\n",
      "complete (39,84)\n",
      "complete (39,82)\n",
      "complete (39,87)\n",
      "complete (39,88)\n",
      "complete (39,89)\n",
      "complete (39,94)\n",
      "complete (39,92)\n",
      "complete (39,90)\n",
      "complete (39,93)\n",
      "complete (39,91)\n",
      "complete (39,95)\n",
      "complete (39,97)\n",
      "complete (39,96)\n",
      "complete (40,43)\n",
      "complete (40,42)\n",
      "complete (40,41)\n",
      "complete (39,98)\n",
      "complete (40,44)\n",
      "complete (40,45)\n",
      "complete (40,47)\n",
      "complete (40,48)\n",
      "complete (40,46)\n",
      "complete (40,49)\n",
      "complete (40,52)\n",
      "complete (40,51)\n",
      "complete (40,50)\n",
      "complete (40,54)\n",
      "complete (40,53)\n",
      "complete (40,55)\n",
      "complete (40,56)\n",
      "complete (40,57)\n",
      "complete (40,58)\n",
      "complete (40,60)\n",
      "complete (40,59)\n",
      "complete (40,62)\n",
      "complete (40,61)\n",
      "complete (40,64)\n",
      "complete (40,63)\n",
      "complete (40,67)\n",
      "complete (40,68)\n",
      "complete (40,65)\n",
      "complete (40,66)\n",
      "complete (40,69)\n",
      "complete (40,70)\n",
      "complete (40,71)\n",
      "complete (40,72)\n",
      "complete (40,74)\n",
      "complete (40,73)\n",
      "complete (40,76)\n",
      "complete (40,75)\n",
      "complete (40,77)\n",
      "complete (40,78)\n",
      "complete (40,79)\n",
      "complete (40,80)\n",
      "complete (40,81)\n",
      "complete (40,83)\n",
      "complete (40,82)\n",
      "complete (40,84)\n",
      "complete (40,85)\n",
      "complete (40,86)\n",
      "complete (40,88)\n",
      "complete (40,87)\n",
      "complete (40,89)\n",
      "complete (40,90)\n",
      "complete (40,92)\n",
      "complete (40,91)\n",
      "complete (40,93)\n",
      "complete (40,94)\n",
      "complete (40,97)\n",
      "complete (40,96)\n",
      "complete (40,95)\n",
      "complete (41,42)\n",
      "complete (41,43)\n",
      "complete (40,98)\n",
      "complete (41,44)\n",
      "complete (41,45)\n",
      "complete (41,47)\n",
      "complete (41,46)\n",
      "complete (41,48)\n",
      "complete (41,49)\n",
      "complete (41,51)\n",
      "complete (41,52)\n",
      "complete (41,50)\n",
      "complete (41,54)\n",
      "complete (41,53)\n",
      "complete (41,55)\n",
      "complete (41,56)\n",
      "complete (41,57)\n",
      "complete (41,58)\n",
      "complete (41,59)\n",
      "complete (41,60)\n",
      "complete (41,61)\n",
      "complete (41,62)\n",
      "complete (41,63)\n",
      "complete (41,64)\n",
      "complete (41,65)\n",
      "complete (41,67)\n",
      "complete (41,70)\n",
      "complete (41,66)\n",
      "complete (41,71)\n",
      "complete (41,68)\n",
      "complete (41,69)\n",
      "complete (41,72)\n",
      "complete (41,73)\n",
      "complete (41,76)\n",
      "complete (41,74)\n",
      "complete (41,77)\n",
      "complete (41,78)\n",
      "complete (41,79)\n",
      "complete (41,75)\n",
      "complete (41,80)\n",
      "complete (41,81)\n",
      "complete (41,84)\n",
      "complete (41,82)\n",
      "complete (41,83)\n",
      "complete (41,87)\n",
      "complete (41,85)\n",
      "complete (41,86)\n",
      "complete (41,88)\n",
      "complete (41,89)\n",
      "complete (41,94)\n",
      "complete (41,92)\n",
      "complete (41,90)\n",
      "complete (41,93)\n",
      "complete (41,95)\n",
      "complete (41,91)\n",
      "complete (41,96)\n",
      "complete (41,97)\n",
      "complete (42,44)\n",
      "complete (42,43)\n",
      "complete (41,98)\n",
      "complete (42,45)\n",
      "complete (42,47)\n",
      "complete (42,46)\n",
      "complete (42,48)\n",
      "complete (42,50)\n",
      "complete (42,52)\n",
      "complete (42,49)\n",
      "complete (42,54)\n",
      "complete (42,51)\n",
      "complete (42,53)\n",
      "complete (42,55)\n",
      "complete (42,56)\n",
      "complete (42,57)\n",
      "complete (42,59)\n",
      "complete (42,58)\n",
      "complete (42,62)\n",
      "complete (42,60)\n",
      "complete (42,61)\n",
      "complete (42,63)\n",
      "complete (42,64)\n",
      "complete (42,65)\n",
      "complete (42,68)\n",
      "complete (42,69)\n",
      "complete (42,67)\n",
      "complete (42,70)\n",
      "complete (42,66)\n",
      "complete (42,71)\n",
      "complete (42,72)\n",
      "complete (42,73)\n",
      "complete (42,76)\n",
      "complete (42,77)\n",
      "complete (42,74)\n",
      "complete (42,78)\n",
      "complete (42,75)\n",
      "complete (42,80)\n",
      "complete (42,79)\n",
      "complete (42,81)\n",
      "complete (42,84)\n",
      "complete (42,82)\n",
      "complete (42,85)\n",
      "complete (42,83)\n",
      "complete (42,88)\n",
      "complete (42,87)\n",
      "complete (42,86)\n",
      "complete (42,89)\n",
      "complete (42,90)\n",
      "complete (42,92)\n",
      "complete (42,91)\n",
      "complete (42,94)\n",
      "complete (42,96)\n",
      "complete (42,93)\n",
      "complete (42,95)\n",
      "complete (42,97)\n",
      "complete (43,44)\n",
      "complete (43,45)\n",
      "complete (43,47)\n",
      "complete (43,46)\n",
      "complete (42,98)\n",
      "complete (43,48)\n",
      "complete (43,49)\n",
      "complete (43,51)\n",
      "complete (43,52)\n",
      "complete (43,50)\n",
      "complete (43,54)\n",
      "complete (43,55)\n",
      "complete (43,53)\n",
      "complete (43,57)\n",
      "complete (43,56)\n",
      "complete (43,58)\n",
      "complete (43,59)\n",
      "complete (43,62)\n",
      "complete (43,61)\n",
      "complete (43,60)\n",
      "complete (43,63)\n",
      "complete (43,64)\n",
      "complete (43,65)\n",
      "complete (43,67)\n",
      "complete (43,66)\n",
      "complete (43,70)\n",
      "complete (43,68)\n",
      "complete (43,69)\n",
      "complete (43,71)\n",
      "complete (43,72)\n",
      "complete (43,73)\n",
      "complete (43,76)\n",
      "complete (43,77)\n",
      "complete (43,78)\n",
      "complete (43,74)\n",
      "complete (43,75)\n",
      "complete (43,79)\n",
      "complete (43,80)\n",
      "complete (43,81)\n",
      "complete (43,82)\n",
      "complete (43,85)\n",
      "complete (43,86)\n",
      "complete (43,83)\n",
      "complete (43,87)\n",
      "complete (43,84)\n",
      "complete (43,88)\n",
      "complete (43,89)\n",
      "complete (43,90)\n",
      "complete (43,92)\n",
      "complete (43,94)\n",
      "complete (43,91)\n",
      "complete (43,93)\n",
      "complete (43,96)\n",
      "complete (43,95)\n",
      "complete (43,97)\n",
      "complete (44,45)\n",
      "complete (44,47)\n",
      "complete (44,46)\n",
      "complete (43,98)\n",
      "complete (44,48)\n",
      "complete (44,50)\n",
      "complete (44,49)\n",
      "complete (44,51)\n",
      "complete (44,52)\n",
      "complete (44,55)\n",
      "complete (44,54)\n",
      "complete (44,56)\n",
      "complete (44,53)\n",
      "complete (44,58)\n",
      "complete (44,57)\n",
      "complete (44,59)\n",
      "complete (44,62)\n",
      "complete (44,60)\n",
      "complete (44,61)\n",
      "complete (44,64)\n",
      "complete (44,63)\n",
      "complete (44,65)\n",
      "complete (44,66)\n",
      "complete (44,67)\n",
      "complete (44,69)\n",
      "complete (44,68)\n",
      "complete (44,70)\n",
      "complete (44,71)\n",
      "complete (44,72)\n",
      "complete (44,73)\n",
      "complete (44,74)\n",
      "complete (44,77)\n",
      "complete (44,75)\n",
      "complete (44,76)\n",
      "complete (44,78)\n",
      "complete (44,79)\n",
      "complete (44,81)\n",
      "complete (44,80)\n",
      "complete (44,83)\n",
      "complete (44,82)\n",
      "complete (44,85)\n",
      "complete (44,84)\n",
      "complete (44,86)\n",
      "complete (44,87)\n",
      "complete (44,88)\n",
      "complete (44,89)\n",
      "complete (44,92)\n",
      "complete (44,90)\n",
      "complete (44,91)\n",
      "complete (44,93)\n",
      "complete (44,94)\n",
      "complete (44,96)\n",
      "complete (44,95)\n",
      "complete (44,97)\n",
      "complete (45,46)\n",
      "complete (44,98)\n",
      "complete (45,47)\n",
      "complete (45,48)\n",
      "complete (45,49)\n",
      "complete (45,50)\n",
      "complete (45,52)\n",
      "complete (45,51)\n",
      "complete (45,53)\n",
      "complete (45,54)\n",
      "complete (45,55)\n",
      "complete (45,56)\n",
      "complete (45,57)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "complete (45,58)\n",
      "complete (45,60)\n",
      "complete (45,59)\n",
      "complete (45,62)\n",
      "complete (45,63)\n",
      "complete (45,64)\n",
      "complete (45,61)\n",
      "complete (45,65)\n",
      "complete (45,66)\n",
      "complete (45,67)\n",
      "complete (45,68)\n",
      "complete (45,69)\n",
      "complete (45,71)\n",
      "complete (45,70)\n",
      "complete (45,72)\n",
      "complete (45,73)\n",
      "complete (45,74)\n",
      "complete (45,77)\n",
      "complete (45,75)\n",
      "complete (45,76)\n",
      "complete (45,78)\n",
      "complete (45,79)\n",
      "complete (45,80)\n",
      "complete (45,81)\n",
      "complete (45,82)\n",
      "complete (45,83)\n",
      "complete (45,85)\n",
      "complete (45,84)\n",
      "complete (45,86)\n",
      "complete (45,87)\n",
      "complete (45,88)\n",
      "complete (45,89)\n",
      "complete (45,90)\n",
      "complete (45,92)\n",
      "complete (45,91)\n",
      "complete (45,93)\n",
      "complete (45,96)\n",
      "complete (45,94)\n",
      "complete (45,95)\n",
      "complete (45,97)\n",
      "complete (45,98)\n",
      "complete (46,47)\n",
      "complete (46,48)\n",
      "complete (46,50)\n",
      "complete (46,51)\n",
      "complete (46,49)\n",
      "complete (46,52)\n",
      "complete (46,54)\n",
      "complete (46,53)\n",
      "complete (46,55)\n",
      "complete (46,56)\n",
      "complete (46,57)\n",
      "complete (46,58)\n",
      "complete (46,59)\n",
      "complete (46,60)\n",
      "complete (46,61)\n",
      "complete (46,62)\n",
      "complete (46,63)\n",
      "complete (46,64)\n",
      "complete (46,65)\n",
      "complete (46,66)\n",
      "complete (46,67)\n",
      "complete (46,69)\n",
      "complete (46,71)\n",
      "complete (46,68)\n",
      "complete (46,72)\n",
      "complete (46,70)\n",
      "complete (46,74)\n",
      "complete (46,73)\n",
      "complete (46,76)\n",
      "complete (46,75)\n",
      "complete (46,77)\n",
      "complete (46,79)\n",
      "complete (46,80)\n",
      "complete (46,78)\n",
      "complete (46,83)\n",
      "complete (46,81)\n",
      "complete (46,82)\n",
      "complete (46,85)\n",
      "complete (46,84)\n",
      "complete (46,86)\n",
      "complete (46,88)\n",
      "complete (46,87)\n",
      "complete (46,89)\n",
      "complete (46,90)\n",
      "complete (46,91)\n",
      "complete (46,92)\n",
      "complete (46,93)\n",
      "complete (46,94)\n",
      "complete (46,96)\n",
      "complete (46,95)\n",
      "complete (47,48)\n",
      "complete (46,97)\n",
      "complete (46,98)\n",
      "complete (47,49)\n",
      "complete (47,52)\n",
      "complete (47,53)\n",
      "complete (47,51)\n",
      "complete (47,50)\n",
      "complete (47,54)\n",
      "complete (47,56)\n",
      "complete (47,57)\n",
      "complete (47,55)\n",
      "complete (47,58)\n",
      "complete (47,59)\n",
      "complete (47,61)\n",
      "complete (47,60)\n",
      "complete (47,62)\n",
      "complete (47,64)\n",
      "complete (47,63)\n",
      "complete (47,65)\n",
      "complete (47,66)\n",
      "complete (47,68)\n",
      "complete (47,69)\n",
      "complete (47,67)\n",
      "complete (47,70)\n",
      "complete (47,71)\n",
      "complete (47,72)\n",
      "complete (47,73)\n",
      "complete (47,74)\n",
      "complete (47,76)\n",
      "complete (47,77)\n",
      "complete (47,75)\n",
      "complete (47,78)\n",
      "complete (47,79)\n",
      "complete (47,80)\n",
      "complete (47,81)\n",
      "complete (47,83)\n",
      "complete (47,85)\n",
      "complete (47,82)\n",
      "complete (47,84)\n",
      "complete (47,86)\n",
      "complete (47,87)\n",
      "complete (47,89)\n",
      "complete (47,88)\n",
      "complete (47,92)\n",
      "complete (47,90)\n",
      "complete (47,91)\n",
      "complete (47,93)\n",
      "complete (47,94)\n",
      "complete (47,96)\n",
      "complete (47,95)\n",
      "complete (47,97)\n",
      "complete (47,98)\n",
      "complete (48,49)\n",
      "complete (48,51)\n",
      "complete (48,52)\n",
      "complete (48,50)\n",
      "complete (48,54)\n",
      "complete (48,55)\n",
      "complete (48,53)\n",
      "complete (48,56)\n",
      "complete (48,57)\n",
      "complete (48,58)\n",
      "complete (48,60)\n",
      "complete (48,59)\n",
      "complete (48,61)\n",
      "complete (48,62)\n",
      "complete (48,63)\n",
      "complete (48,64)\n",
      "complete (48,65)\n",
      "complete (48,69)\n",
      "complete (48,67)\n",
      "complete (48,70)\n",
      "complete (48,68)\n",
      "complete (48,66)\n",
      "complete (48,72)\n",
      "complete (48,71)\n",
      "complete (48,73)\n",
      "complete (48,74)\n",
      "complete (48,76)\n",
      "complete (48,78)\n",
      "complete (48,77)\n",
      "complete (48,75)\n",
      "complete (48,79)\n",
      "complete (48,80)\n",
      "complete (48,81)\n",
      "complete (48,84)\n",
      "complete (48,82)\n",
      "complete (48,86)\n",
      "complete (48,83)\n",
      "complete (48,87)\n",
      "complete (48,85)\n",
      "complete (48,88)\n",
      "complete (48,89)\n",
      "complete (48,92)\n",
      "complete (48,90)\n",
      "complete (48,94)\n",
      "complete (48,93)\n",
      "complete (48,91)\n",
      "complete (48,96)\n",
      "complete (48,95)\n",
      "complete (48,97)\n",
      "complete (48,98)\n",
      "complete (49,50)\n",
      "complete (49,54)\n",
      "complete (49,51)\n",
      "complete (49,53)\n",
      "complete (49,52)\n",
      "complete (49,55)\n",
      "complete (49,56)\n",
      "complete (49,57)\n",
      "complete (49,58)\n",
      "complete (49,62)\n",
      "complete (49,63)\n",
      "complete (49,60)\n",
      "complete (49,61)\n",
      "complete (49,59)\n",
      "complete (49,64)\n",
      "complete (49,65)\n",
      "complete (49,67)\n",
      "complete (49,69)\n",
      "complete (49,71)\n",
      "complete (49,66)\n",
      "complete (49,72)\n",
      "complete (49,68)\n",
      "complete (49,70)\n",
      "complete (49,73)\n",
      "complete (49,77)\n",
      "complete (49,74)\n",
      "complete (49,78)\n",
      "complete (49,76)\n",
      "complete (49,75)\n",
      "complete (49,79)\n",
      "complete (49,80)\n",
      "complete (49,81)\n",
      "complete (49,82)\n",
      "complete (49,85)\n",
      "complete (49,84)\n",
      "complete (49,86)\n",
      "complete (49,83)\n",
      "complete (49,87)\n",
      "complete (49,88)\n",
      "complete (49,89)\n",
      "complete (49,90)\n",
      "complete (49,94)\n",
      "complete (49,93)\n",
      "complete (49,91)\n",
      "complete (49,92)\n",
      "complete (49,96)\n",
      "complete (49,95)\n",
      "complete (49,97)\n",
      "complete (49,98)\n",
      "complete (50,54)\n",
      "complete (50,52)\n",
      "complete (50,51)\n",
      "complete (50,53)\n",
      "complete (50,55)\n",
      "complete (50,56)\n",
      "complete (50,57)\n",
      "complete (50,59)\n",
      "complete (50,58)\n",
      "complete (50,62)\n",
      "complete (50,61)\n",
      "complete (50,63)\n",
      "complete (50,60)\n",
      "complete (50,64)\n",
      "complete (50,65)\n",
      "complete (50,70)\n",
      "complete (50,69)\n",
      "complete (50,68)\n",
      "complete (50,67)\n",
      "complete (50,66)\n",
      "complete (50,71)\n",
      "complete (50,72)\n",
      "complete (50,73)\n",
      "complete (50,74)\n",
      "complete (50,77)\n",
      "complete (50,76)\n",
      "complete (50,75)\n",
      "complete (50,79)\n",
      "complete (50,78)\n",
      "complete (50,80)\n",
      "complete (50,81)\n",
      "complete (50,83)\n",
      "complete (50,82)\n",
      "complete (50,87)\n",
      "complete (50,84)\n",
      "complete (50,85)\n",
      "complete (50,86)\n",
      "complete (50,88)\n",
      "complete (50,89)\n",
      "complete (50,90)\n",
      "complete (50,91)\n",
      "complete (50,92)\n",
      "complete (50,94)\n",
      "complete (50,93)\n",
      "complete (50,96)\n",
      "complete (50,95)\n",
      "complete (50,97)\n",
      "complete (50,98)\n",
      "complete (51,52)\n",
      "complete (51,54)\n",
      "complete (51,55)\n",
      "complete (51,53)\n",
      "complete (51,56)\n",
      "complete (51,57)\n",
      "complete (51,58)\n",
      "complete (51,60)\n",
      "complete (51,59)\n",
      "complete (51,61)\n",
      "complete (51,62)\n",
      "complete (51,63)\n",
      "complete (51,64)\n",
      "complete (51,65)\n",
      "complete (51,66)\n",
      "complete (51,67)\n",
      "complete (51,68)\n",
      "complete (51,69)\n",
      "complete (51,70)\n",
      "complete (51,71)\n",
      "complete (51,72)\n",
      "complete (51,73)\n",
      "complete (51,74)\n",
      "complete (51,76)\n",
      "complete (51,77)\n",
      "complete (51,75)\n",
      "complete (51,78)\n",
      "complete (51,79)\n",
      "complete (51,80)\n",
      "complete (51,81)\n",
      "complete (51,82)\n",
      "complete (51,85)\n",
      "complete (51,83)\n",
      "complete (51,84)\n",
      "complete (51,86)\n",
      "complete (51,87)\n",
      "complete (51,88)\n",
      "complete (51,89)\n",
      "complete (51,90)\n",
      "complete (51,92)\n",
      "complete (51,91)\n",
      "complete (51,93)\n",
      "complete (51,94)\n",
      "complete (51,96)\n",
      "complete (51,97)\n",
      "complete (51,95)\n",
      "complete (52,54)\n",
      "complete (51,98)\n",
      "complete (52,53)\n",
      "complete (52,55)\n",
      "complete (52,56)\n",
      "complete (52,57)\n",
      "complete (52,58)\n",
      "complete (52,59)\n",
      "complete (52,62)\n",
      "complete (52,60)\n",
      "complete (52,61)\n",
      "complete (52,64)\n",
      "complete (52,63)\n",
      "complete (52,65)\n",
      "complete (52,66)\n",
      "complete (52,67)\n",
      "complete (52,68)\n",
      "complete (52,70)\n",
      "complete (52,69)\n",
      "complete (52,71)\n",
      "complete (52,72)\n",
      "complete (52,73)\n",
      "complete (52,74)\n",
      "complete (52,76)\n",
      "complete (52,75)\n",
      "complete (52,77)\n",
      "complete (52,78)\n",
      "complete (52,79)\n",
      "complete (52,80)\n",
      "complete (52,81)\n",
      "complete (52,82)\n",
      "complete (52,83)\n",
      "complete (52,85)\n",
      "complete (52,86)\n",
      "complete (52,84)\n",
      "complete (52,87)\n",
      "complete (52,88)\n",
      "complete (52,89)\n",
      "complete (52,90)\n",
      "complete (52,94)\n",
      "complete (52,91)\n",
      "complete (52,92)\n",
      "complete (52,93)\n",
      "complete (52,96)\n",
      "complete (52,95)\n",
      "complete (52,97)\n",
      "complete (52,98)\n",
      "complete (53,54)\n",
      "complete (53,55)\n",
      "complete (53,57)\n",
      "complete (53,56)\n",
      "complete (53,58)\n",
      "complete (53,60)\n",
      "complete (53,59)\n",
      "complete (53,61)\n",
      "complete (53,62)\n",
      "complete (53,64)\n",
      "complete (53,63)\n",
      "complete (53,65)\n",
      "complete (53,66)\n",
      "complete (53,67)\n",
      "complete (53,68)\n",
      "complete (53,69)\n",
      "complete (53,70)\n",
      "complete (53,71)\n",
      "complete (53,72)\n",
      "complete (53,73)\n",
      "complete (53,74)\n",
      "complete (53,76)\n",
      "complete (53,75)\n",
      "complete (53,77)\n",
      "complete (53,78)\n",
      "complete (53,79)\n",
      "complete (53,80)\n",
      "complete (53,81)\n",
      "complete (53,82)\n",
      "complete (53,85)\n",
      "complete (53,83)\n",
      "complete (53,84)\n",
      "complete (53,86)\n",
      "complete (53,87)\n",
      "complete (53,89)\n",
      "complete (53,88)\n",
      "complete (53,92)\n",
      "complete (53,90)\n",
      "complete (53,91)\n",
      "complete (53,94)\n",
      "complete (53,93)\n",
      "complete (53,97)\n",
      "complete (53,96)\n",
      "complete (53,95)\n",
      "complete (54,55)\n",
      "complete (54,56)\n",
      "complete (53,98)\n",
      "complete (54,57)\n",
      "complete (54,58)\n",
      "complete (54,59)\n",
      "complete (54,60)\n",
      "complete (54,61)\n",
      "complete (54,62)\n",
      "complete (54,64)\n",
      "complete (54,63)\n",
      "complete (54,65)\n",
      "complete (54,66)\n",
      "complete (54,68)\n",
      "complete (54,69)\n",
      "complete (54,67)\n",
      "complete (54,70)\n",
      "complete (54,72)\n",
      "complete (54,71)\n",
      "complete (54,73)\n",
      "complete (54,74)\n",
      "complete (54,76)\n",
      "complete (54,75)\n",
      "complete (54,77)\n",
      "complete (54,78)\n",
      "complete (54,79)\n",
      "complete (54,80)\n",
      "complete (54,81)\n",
      "complete (54,82)\n",
      "complete (54,83)\n",
      "complete (54,86)\n",
      "complete (54,84)\n",
      "complete (54,85)\n",
      "complete (54,87)\n",
      "complete (54,88)\n",
      "complete (54,89)\n",
      "complete (54,90)\n",
      "complete (54,92)\n",
      "complete (54,91)\n",
      "complete (54,94)\n",
      "complete (54,93)\n",
      "complete (54,96)\n",
      "complete (54,95)\n",
      "complete (54,97)\n",
      "complete (54,98)\n",
      "complete (55,56)\n",
      "complete (55,57)\n",
      "complete (55,58)\n",
      "complete (55,60)\n",
      "complete (55,59)\n",
      "complete (55,61)\n",
      "complete (55,62)\n",
      "complete (55,63)\n",
      "complete (55,64)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "complete (55,65)\n",
      "complete (55,66)\n",
      "complete (55,69)\n",
      "complete (55,68)\n",
      "complete (55,70)\n",
      "complete (55,67)\n",
      "complete (55,72)\n",
      "complete (55,71)\n",
      "complete (55,74)\n",
      "complete (55,73)\n",
      "complete (55,75)\n",
      "complete (55,77)\n",
      "complete (55,76)\n",
      "complete (55,78)\n",
      "complete (55,79)\n",
      "complete (55,80)\n",
      "complete (55,81)\n",
      "complete (55,85)\n",
      "complete (55,84)\n",
      "complete (55,82)\n",
      "complete (55,86)\n",
      "complete (55,83)\n",
      "complete (55,87)\n",
      "complete (55,89)\n",
      "complete (55,88)\n",
      "complete (55,92)\n",
      "complete (55,90)\n",
      "complete (55,94)\n",
      "complete (55,93)\n",
      "complete (55,91)\n",
      "complete (55,95)\n",
      "complete (55,96)\n",
      "complete (55,97)\n",
      "complete (55,98)\n",
      "complete (56,57)\n",
      "complete (56,58)\n",
      "complete (56,60)\n",
      "complete (56,59)\n",
      "complete (56,62)\n",
      "complete (56,61)\n",
      "complete (56,64)\n",
      "complete (56,63)\n",
      "complete (56,65)\n",
      "complete (56,68)\n",
      "complete (56,66)\n",
      "complete (56,67)\n",
      "complete (56,69)\n",
      "complete (56,70)\n",
      "complete (56,72)\n",
      "complete (56,71)\n",
      "complete (56,73)\n",
      "complete (56,74)\n",
      "complete (56,75)\n",
      "complete (56,76)\n",
      "complete (56,78)\n",
      "complete (56,77)\n",
      "complete (56,79)\n",
      "complete (56,80)\n",
      "complete (56,81)\n",
      "complete (56,82)\n",
      "complete (56,83)\n",
      "complete (56,84)\n",
      "complete (56,85)\n",
      "complete (56,88)\n",
      "complete (56,87)\n",
      "complete (56,86)\n",
      "complete (56,89)\n",
      "complete (56,90)\n",
      "complete (56,91)\n",
      "complete (56,92)\n",
      "complete (56,93)\n",
      "complete (56,94)\n",
      "complete (56,95)\n",
      "complete (56,96)\n",
      "complete (56,97)\n",
      "complete (56,98)\n",
      "complete (57,58)\n",
      "complete (57,59)\n",
      "complete (57,60)\n",
      "complete (57,61)\n",
      "complete (57,64)\n",
      "complete (57,62)\n",
      "complete (57,63)\n",
      "complete (57,68)\n",
      "complete (57,65)\n",
      "complete (57,67)\n",
      "complete (57,66)\n",
      "complete (57,71)\n",
      "complete (57,69)\n",
      "complete (57,70)\n",
      "complete (57,72)\n",
      "complete (57,74)\n",
      "complete (57,73)\n",
      "complete (57,77)\n",
      "complete (57,76)\n",
      "complete (57,75)\n",
      "complete (57,78)\n",
      "complete (57,79)\n",
      "complete (57,80)\n",
      "complete (57,81)\n",
      "complete (57,83)\n",
      "complete (57,82)\n",
      "complete (57,84)\n",
      "complete (57,85)\n",
      "complete (57,86)\n",
      "complete (57,87)\n",
      "complete (57,88)\n",
      "complete (57,89)\n",
      "complete (57,90)\n",
      "complete (57,92)\n",
      "complete (57,91)\n",
      "complete (57,94)\n",
      "complete (57,93)\n",
      "complete (57,95)\n",
      "complete (57,96)\n",
      "complete (57,97)\n",
      "complete (57,98)\n",
      "complete (58,59)\n",
      "complete (58,62)\n",
      "complete (58,60)\n",
      "complete (58,61)\n",
      "complete (58,64)\n",
      "complete (58,63)\n",
      "complete (58,65)\n",
      "complete (58,66)\n",
      "complete (58,67)\n",
      "complete (58,69)\n",
      "complete (58,68)\n",
      "complete (58,70)\n",
      "complete (58,71)\n",
      "complete (58,72)\n",
      "complete (58,73)\n",
      "complete (58,74)\n",
      "complete (58,77)\n",
      "complete (58,76)\n",
      "complete (58,75)\n",
      "complete (58,78)\n",
      "complete (58,79)\n",
      "complete (58,80)\n",
      "complete (58,81)\n",
      "complete (58,82)\n",
      "complete (58,83)\n",
      "complete (58,84)\n",
      "complete (58,85)\n",
      "complete (58,86)\n",
      "complete (58,87)\n",
      "complete (58,89)\n",
      "complete (58,88)\n",
      "complete (58,90)\n",
      "complete (58,93)\n",
      "complete (58,92)\n",
      "complete (58,91)\n",
      "complete (58,94)\n",
      "complete (58,96)\n",
      "complete (58,95)\n",
      "complete (58,97)\n",
      "complete (59,62)\n",
      "complete (58,98)\n",
      "complete (59,61)\n",
      "complete (59,60)\n",
      "complete (59,63)\n",
      "complete (59,64)\n",
      "complete (59,65)\n",
      "complete (59,66)\n",
      "complete (59,67)\n",
      "complete (59,68)\n",
      "complete (59,70)\n",
      "complete (59,69)\n",
      "complete (59,71)\n",
      "complete (59,72)\n",
      "complete (59,73)\n",
      "complete (59,74)\n",
      "complete (59,75)\n",
      "complete (59,76)\n",
      "complete (59,77)\n",
      "complete (59,78)\n",
      "complete (59,79)\n",
      "complete (59,81)\n",
      "complete (59,80)\n",
      "complete (59,84)\n",
      "complete (59,83)\n",
      "complete (59,82)\n",
      "complete (59,85)\n",
      "complete (59,86)\n",
      "complete (59,87)\n",
      "complete (59,89)\n",
      "complete (59,88)\n",
      "complete (59,92)\n",
      "complete (59,90)\n",
      "complete (59,91)\n",
      "complete (59,94)\n",
      "complete (59,93)\n",
      "complete (59,96)\n",
      "complete (59,97)\n",
      "complete (59,95)\n",
      "complete (60,62)\n",
      "complete (60,64)\n",
      "complete (59,98)\n",
      "complete (60,61)\n",
      "complete (60,63)\n",
      "complete (60,65)\n",
      "complete (60,67)\n",
      "complete (60,66)\n",
      "complete (60,68)\n",
      "complete (60,69)\n",
      "complete (60,70)\n",
      "complete (60,71)\n",
      "complete (60,72)\n",
      "complete (60,74)\n",
      "complete (60,73)\n",
      "complete (60,75)\n",
      "complete (60,76)\n",
      "complete (60,77)\n",
      "complete (60,78)\n",
      "complete (60,79)\n",
      "complete (60,80)\n",
      "complete (60,81)\n",
      "complete (60,82)\n",
      "complete (60,83)\n",
      "complete (60,84)\n",
      "complete (60,85)\n",
      "complete (60,86)\n",
      "complete (60,87)\n",
      "complete (60,88)\n",
      "complete (60,89)\n",
      "complete (60,90)\n",
      "complete (60,92)\n",
      "complete (60,91)\n",
      "complete (60,94)\n",
      "complete (60,93)\n",
      "complete (60,95)\n",
      "complete (60,96)\n",
      "complete (60,97)\n",
      "complete (61,62)\n",
      "complete (60,98)\n",
      "complete (61,63)\n",
      "complete (61,64)\n",
      "complete (61,66)\n",
      "complete (61,65)\n",
      "complete (61,67)\n",
      "complete (61,68)\n",
      "complete (61,69)\n",
      "complete (61,70)\n",
      "complete (61,72)\n",
      "complete (61,71)\n",
      "complete (61,74)\n",
      "complete (61,73)\n",
      "complete (61,75)\n",
      "complete (61,76)\n",
      "complete (61,77)\n",
      "complete (61,78)\n",
      "complete (61,79)\n",
      "complete (61,80)\n",
      "complete (61,81)\n",
      "complete (61,82)\n",
      "complete (61,83)\n",
      "complete (61,84)\n",
      "complete (61,85)\n",
      "complete (61,86)\n",
      "complete (61,87)\n",
      "complete (61,89)\n",
      "complete (61,88)\n",
      "complete (61,90)\n",
      "complete (61,92)\n",
      "complete (61,91)\n",
      "complete (61,93)\n",
      "complete (61,94)\n",
      "complete (61,95)\n",
      "complete (61,96)\n",
      "complete (61,97)\n",
      "complete (62,64)\n",
      "complete (61,98)\n",
      "complete (62,63)\n",
      "complete (62,65)\n",
      "complete (62,66)\n",
      "complete (62,68)\n",
      "complete (62,69)\n",
      "complete (62,67)\n",
      "complete (62,70)\n",
      "complete (62,72)\n",
      "complete (62,71)\n",
      "complete (62,73)\n",
      "complete (62,74)\n",
      "complete (62,76)\n",
      "complete (62,77)\n",
      "complete (62,75)\n",
      "complete (62,78)\n",
      "complete (62,79)\n",
      "complete (62,81)\n",
      "complete (62,80)\n",
      "complete (62,82)\n",
      "complete (62,84)\n",
      "complete (62,85)\n",
      "complete (62,83)\n",
      "complete (62,87)\n",
      "complete (62,88)\n",
      "complete (62,86)\n",
      "complete (62,89)\n",
      "complete (62,90)\n",
      "complete (62,92)\n",
      "complete (62,93)\n",
      "complete (62,91)\n",
      "complete (62,94)\n",
      "complete (62,97)\n",
      "complete (62,96)\n",
      "complete (62,95)\n",
      "complete (62,98)\n",
      "complete (63,64)\n",
      "complete (63,65)\n",
      "complete (63,66)\n",
      "complete (63,69)\n",
      "complete (63,67)\n",
      "complete (63,70)\n",
      "complete (63,68)\n",
      "complete (63,71)\n",
      "complete (63,72)\n",
      "complete (63,73)\n",
      "complete (63,74)\n",
      "complete (63,77)\n",
      "complete (63,75)\n",
      "complete (63,78)\n",
      "complete (63,76)\n",
      "complete (63,79)\n",
      "complete (63,81)\n",
      "complete (63,80)\n",
      "complete (63,84)\n",
      "complete (63,82)\n",
      "complete (63,86)\n",
      "complete (63,83)\n",
      "complete (63,85)\n",
      "complete (63,87)\n",
      "complete (63,88)\n",
      "complete (63,89)\n",
      "complete (63,94)\n",
      "complete (63,90)\n",
      "complete (63,93)\n",
      "complete (63,91)\n",
      "complete (63,92)\n",
      "complete (63,95)\n",
      "complete (63,96)\n",
      "complete (63,97)\n",
      "complete (64,65)\n",
      "complete (63,98)\n",
      "complete (64,66)\n",
      "complete (64,67)\n",
      "complete (64,68)\n",
      "complete (64,69)\n",
      "complete (64,70)\n",
      "complete (64,71)\n",
      "complete (64,72)\n",
      "complete (64,76)\n",
      "complete (64,73)\n",
      "complete (64,74)\n",
      "complete (64,75)\n",
      "complete (64,77)\n",
      "complete (64,79)\n",
      "complete (64,78)\n",
      "complete (64,80)\n",
      "complete (64,81)\n",
      "complete (64,82)\n",
      "complete (64,84)\n",
      "complete (64,83)\n",
      "complete (64,85)\n",
      "complete (64,86)\n",
      "complete (64,87)\n",
      "complete (64,89)\n",
      "complete (64,88)\n",
      "complete (64,90)\n",
      "complete (64,92)\n",
      "complete (64,93)\n",
      "complete (64,91)\n",
      "complete (64,94)\n",
      "complete (64,95)\n",
      "complete (64,97)\n",
      "complete (64,96)\n",
      "complete (64,98)\n",
      "complete (65,68)\n",
      "complete (65,66)\n",
      "complete (65,69)\n",
      "complete (65,67)\n",
      "complete (65,72)\n",
      "complete (65,70)\n",
      "complete (65,71)\n",
      "complete (65,73)\n",
      "complete (65,76)\n",
      "complete (65,74)\n",
      "complete (65,77)\n",
      "complete (65,75)\n",
      "complete (65,78)\n",
      "complete (65,80)\n",
      "complete (65,79)\n",
      "complete (65,81)\n",
      "complete (65,82)\n",
      "complete (65,83)\n",
      "complete (65,84)\n",
      "complete (65,86)\n",
      "complete (65,85)\n",
      "complete (65,89)\n",
      "complete (65,87)\n",
      "complete (65,88)\n",
      "complete (65,90)\n",
      "complete (65,92)\n",
      "complete (65,91)\n",
      "complete (65,93)\n",
      "complete (65,94)\n",
      "complete (65,97)\n",
      "complete (65,96)\n",
      "complete (65,95)\n",
      "complete (65,98)\n",
      "complete (66,69)\n",
      "complete (66,68)\n",
      "complete (66,67)\n",
      "complete (66,70)\n",
      "complete (66,72)\n",
      "complete (66,71)\n",
      "complete (66,73)\n",
      "complete (66,74)\n",
      "complete (66,77)\n",
      "complete (66,76)\n",
      "complete (66,78)\n",
      "complete (66,75)\n",
      "complete (66,79)\n",
      "complete (66,80)\n",
      "complete (66,81)\n",
      "complete (66,84)\n",
      "complete (66,83)\n",
      "complete (66,82)\n",
      "complete (66,86)\n",
      "complete (66,85)\n",
      "complete (66,87)\n",
      "complete (66,88)\n",
      "complete (66,89)\n",
      "complete (66,90)\n",
      "complete (66,93)\n",
      "complete (66,92)\n",
      "complete (66,94)\n",
      "complete (66,91)\n",
      "complete (66,96)\n",
      "complete (66,97)\n",
      "complete (66,95)\n",
      "complete (67,69)\n",
      "complete (67,68)\n",
      "complete (66,98)\n",
      "complete (67,71)\n",
      "complete (67,70)\n",
      "complete (67,72)\n",
      "complete (67,74)\n",
      "complete (67,73)\n",
      "complete (67,77)\n",
      "complete (67,76)\n",
      "complete (67,75)\n",
      "complete (67,78)\n",
      "complete (67,79)\n",
      "complete (67,80)\n",
      "complete (67,81)\n",
      "complete (67,82)\n",
      "complete (67,85)\n",
      "complete (67,83)\n",
      "complete (67,87)\n",
      "complete (67,84)\n",
      "complete (67,86)\n",
      "complete (67,89)\n",
      "complete (67,88)\n",
      "complete (67,90)\n",
      "complete (67,91)\n",
      "complete (67,94)\n",
      "complete (67,92)\n",
      "complete (67,93)\n",
      "complete (67,95)\n",
      "complete (67,96)\n",
      "complete (67,97)\n",
      "complete (67,98)\n",
      "complete (68,70)\n",
      "complete (68,69)\n",
      "complete (68,72)\n",
      "complete (68,71)\n",
      "complete (68,73)\n",
      "complete (68,74)\n",
      "complete (68,75)\n",
      "complete (68,76)\n",
      "complete (68,77)\n",
      "complete (68,78)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "complete (68,81)\n",
      "complete (68,80)\n",
      "complete (68,79)\n",
      "complete (68,82)\n",
      "complete (68,83)\n",
      "complete (68,84)\n",
      "complete (68,85)\n",
      "complete (68,87)\n",
      "complete (68,86)\n",
      "complete (68,88)\n",
      "complete (68,89)\n",
      "complete (68,90)\n",
      "complete (68,92)\n",
      "complete (68,91)\n",
      "complete (68,94)\n",
      "complete (68,93)\n",
      "complete (68,95)\n",
      "complete (68,97)\n",
      "complete (68,96)\n",
      "complete (68,98)\n",
      "complete (69,71)\n",
      "complete (69,70)\n",
      "complete (69,72)\n",
      "complete (69,74)\n",
      "complete (69,73)\n",
      "complete (69,76)\n",
      "complete (69,75)\n",
      "complete (69,77)\n",
      "complete (69,78)\n",
      "complete (69,79)\n",
      "complete (69,80)\n",
      "complete (69,81)\n",
      "complete (69,83)\n",
      "complete (69,85)\n",
      "complete (69,82)\n",
      "complete (69,84)\n",
      "complete (69,86)\n",
      "complete (69,87)\n",
      "complete (69,88)\n",
      "complete (69,89)\n",
      "complete (69,93)\n",
      "complete (69,90)\n",
      "complete (69,94)\n",
      "complete (69,92)\n",
      "complete (69,91)\n",
      "complete (69,95)\n",
      "complete (69,96)\n",
      "complete (69,97)\n",
      "complete (70,71)\n",
      "complete (69,98)\n",
      "complete (70,72)\n",
      "complete (70,74)\n",
      "complete (70,73)\n",
      "complete (70,76)\n",
      "complete (70,77)\n",
      "complete (70,75)\n",
      "complete (70,78)\n",
      "complete (70,81)\n",
      "complete (70,79)\n",
      "complete (70,80)\n",
      "complete (70,82)\n",
      "complete (70,84)\n",
      "complete (70,83)\n",
      "complete (70,85)\n",
      "complete (70,86)\n",
      "complete (70,88)\n",
      "complete (70,87)\n",
      "complete (70,89)\n",
      "complete (70,90)\n",
      "complete (70,92)\n",
      "complete (70,91)\n",
      "complete (70,93)\n",
      "complete (70,94)\n",
      "complete (70,96)\n",
      "complete (70,97)\n",
      "complete (70,95)\n",
      "complete (71,72)\n",
      "complete (70,98)\n",
      "complete (71,74)\n",
      "complete (71,73)\n",
      "complete (71,76)\n",
      "complete (71,75)\n",
      "complete (71,77)\n",
      "complete (71,78)\n",
      "complete (71,79)\n",
      "complete (71,80)\n",
      "complete (71,81)\n",
      "complete (71,83)\n",
      "complete (71,84)\n",
      "complete (71,82)\n",
      "complete (71,85)\n",
      "complete (71,86)\n",
      "complete (71,87)\n",
      "complete (71,89)\n",
      "complete (71,88)\n",
      "complete (71,90)\n",
      "complete (71,93)\n",
      "complete (71,94)\n",
      "complete (71,92)\n",
      "complete (71,91)\n",
      "complete (71,96)\n",
      "complete (71,97)\n",
      "complete (71,95)\n",
      "complete (72,74)\n",
      "complete (71,98)\n",
      "complete (72,73)\n",
      "complete (72,76)\n",
      "complete (72,75)\n",
      "complete (72,77)\n",
      "complete (72,78)\n",
      "complete (72,79)\n",
      "complete (72,81)\n",
      "complete (72,80)\n",
      "complete (72,83)\n",
      "complete (72,82)\n",
      "complete (72,84)\n",
      "complete (72,85)\n",
      "complete (72,87)\n",
      "complete (72,86)\n",
      "complete (72,88)\n",
      "complete (72,89)\n",
      "complete (72,90)\n",
      "complete (72,91)\n",
      "complete (72,92)\n",
      "complete (72,94)\n",
      "complete (72,93)\n",
      "complete (72,95)\n",
      "complete (72,96)\n",
      "complete (72,97)\n",
      "complete (72,98)\n",
      "complete (73,74)\n",
      "complete (73,75)\n",
      "complete (73,77)\n",
      "complete (73,76)\n",
      "complete (73,78)\n",
      "complete (73,79)\n",
      "complete (73,80)\n",
      "complete (73,81)\n",
      "complete (73,86)\n",
      "complete (73,83)\n",
      "complete (73,84)\n",
      "complete (73,82)\n",
      "complete (73,85)\n",
      "complete (73,87)\n",
      "complete (73,88)\n",
      "complete (73,89)\n",
      "complete (73,90)\n",
      "complete (73,94)\n",
      "complete (73,93)\n",
      "complete (73,92)\n",
      "complete (73,91)\n",
      "complete (73,97)\n",
      "complete (73,95)\n",
      "complete (73,96)\n",
      "complete (74,76)\n",
      "complete (73,98)\n",
      "complete (74,75)\n",
      "complete (74,77)\n",
      "complete (74,78)\n",
      "complete (74,81)\n",
      "complete (74,80)\n",
      "complete (74,79)\n",
      "complete (74,82)\n",
      "complete (74,83)\n",
      "complete (74,84)\n",
      "complete (74,85)\n",
      "complete (74,86)\n",
      "complete (74,87)\n",
      "complete (74,88)\n",
      "complete (74,89)\n",
      "complete (74,90)\n",
      "complete (74,94)\n",
      "complete (74,96)\n",
      "complete (74,93)\n",
      "complete (74,92)\n",
      "complete (74,97)\n",
      "complete (74,91)\n",
      "complete (74,95)\n",
      "complete (75,76)\n",
      "complete (75,77)\n",
      "complete (74,98)\n",
      "complete (75,78)\n",
      "complete (75,79)\n",
      "complete (75,81)\n",
      "complete (75,80)\n",
      "complete (75,82)\n",
      "complete (75,83)\n",
      "complete (75,85)\n",
      "complete (75,86)\n",
      "complete (75,84)\n",
      "complete (75,87)\n",
      "complete (75,88)\n",
      "complete (75,89)\n",
      "complete (75,90)\n",
      "complete (75,92)\n",
      "complete (75,91)\n",
      "complete (75,93)\n",
      "complete (75,94)\n",
      "complete (75,95)\n",
      "complete (75,96)\n",
      "complete (75,97)\n",
      "complete (75,98)\n",
      "complete (76,77)\n",
      "complete (76,78)\n",
      "complete (76,80)\n",
      "complete (76,79)\n",
      "complete (76,81)\n",
      "complete (76,83)\n",
      "complete (76,82)\n",
      "complete (76,84)\n",
      "complete (76,85)\n",
      "complete (76,86)\n",
      "complete (76,88)\n",
      "complete (76,89)\n",
      "complete (76,87)\n",
      "complete (76,90)\n",
      "complete (76,91)\n",
      "complete (76,92)\n",
      "complete (76,94)\n",
      "complete (76,93)\n",
      "complete (76,96)\n",
      "complete (76,97)\n",
      "complete (76,95)\n",
      "complete (77,78)\n",
      "complete (76,98)\n",
      "complete (77,79)\n",
      "complete (77,80)\n",
      "complete (77,81)\n",
      "complete (77,82)\n",
      "complete (77,83)\n",
      "complete (77,84)\n",
      "complete (77,85)\n",
      "complete (77,86)\n",
      "complete (77,87)\n",
      "complete (77,89)\n",
      "complete (77,88)\n",
      "complete (77,91)\n",
      "complete (77,90)\n",
      "complete (77,92)\n",
      "complete (77,93)\n",
      "complete (77,94)\n",
      "complete (77,95)\n",
      "complete (77,96)\n",
      "complete (77,97)\n",
      "complete (78,80)\n",
      "complete (77,98)\n",
      "complete (78,81)\n",
      "complete (78,79)\n",
      "complete (78,82)\n",
      "complete (78,83)\n",
      "complete (78,84)\n",
      "complete (78,85)\n",
      "complete (78,86)\n",
      "complete (78,87)\n",
      "complete (78,88)\n",
      "complete (78,89)\n",
      "complete (78,90)\n",
      "complete (78,91)\n",
      "complete (78,92)\n",
      "complete (78,93)\n",
      "complete (78,94)\n",
      "complete (78,97)\n",
      "complete (78,96)\n",
      "complete (78,95)\n",
      "complete (78,98)\n",
      "complete (79,80)\n",
      "complete (79,81)\n",
      "complete (79,82)\n",
      "complete (79,83)\n",
      "complete (79,85)\n",
      "complete (79,84)\n",
      "complete (79,86)\n",
      "complete (79,87)\n",
      "complete (79,88)\n",
      "complete (79,89)\n",
      "complete (79,90)\n",
      "complete (79,92)\n",
      "complete (79,93)\n",
      "complete (79,91)\n",
      "complete (79,94)\n",
      "complete (79,95)\n",
      "complete (79,96)\n",
      "complete (79,97)\n",
      "complete (79,98)\n",
      "complete (80,81)\n",
      "complete (80,82)\n",
      "complete (80,86)\n",
      "complete (80,83)\n",
      "complete (80,85)\n",
      "complete (80,84)\n",
      "complete (80,87)\n",
      "complete (80,89)\n",
      "complete (80,88)\n",
      "complete (80,94)\n",
      "complete (80,92)\n",
      "complete (80,90)\n",
      "complete (80,93)\n",
      "complete (80,91)\n",
      "complete (80,96)\n",
      "complete (80,97)\n",
      "complete (80,95)\n",
      "complete (80,98)\n",
      "complete (81,82)\n",
      "complete (81,84)\n",
      "complete (81,83)\n",
      "complete (81,85)\n",
      "complete (81,87)\n",
      "complete (81,88)\n",
      "complete (81,86)\n",
      "complete (81,89)\n",
      "complete (81,90)\n",
      "complete (81,92)\n",
      "complete (81,91)\n",
      "complete (81,93)\n",
      "complete (81,94)\n",
      "complete (81,96)\n",
      "complete (81,95)\n",
      "complete (81,97)\n",
      "complete (81,98)\n",
      "complete (82,83)\n",
      "complete (82,84)\n",
      "complete (82,85)\n",
      "complete (82,86)\n",
      "complete (82,87)\n",
      "complete (82,88)\n",
      "complete (82,89)\n",
      "complete (82,90)\n",
      "complete (82,92)\n",
      "complete (82,93)\n",
      "complete (82,94)\n",
      "complete (82,91)\n",
      "complete (82,96)\n",
      "complete (82,97)\n",
      "complete (82,95)\n",
      "complete (82,98)\n",
      "complete (83,84)\n",
      "complete (83,85)\n",
      "complete (83,86)\n",
      "complete (83,87)\n",
      "complete (83,89)\n",
      "complete (83,88)\n",
      "complete (83,90)\n",
      "complete (83,91)\n",
      "complete (83,92)\n",
      "complete (83,93)\n",
      "complete (83,94)\n",
      "complete (83,96)\n",
      "complete (83,97)\n",
      "complete (83,95)\n",
      "complete (83,98)\n",
      "complete (84,87)\n",
      "complete (84,85)\n",
      "complete (84,86)\n",
      "complete (84,90)\n",
      "complete (84,89)\n",
      "complete (84,88)\n",
      "complete (84,91)\n",
      "complete (84,92)\n",
      "complete (84,93)\n",
      "complete (84,94)\n",
      "complete (84,95)\n",
      "complete (84,97)\n",
      "complete (84,96)\n",
      "complete (84,98)\n",
      "complete (85,86)\n",
      "complete (85,87)\n",
      "complete (85,88)\n",
      "complete (85,89)\n",
      "complete (85,90)\n",
      "complete (85,92)\n",
      "complete (85,93)\n",
      "complete (85,94)\n",
      "complete (85,91)\n",
      "complete (85,95)\n",
      "complete (85,97)\n",
      "complete (85,96)\n",
      "complete (86,87)\n",
      "complete (85,98)\n",
      "complete (86,88)\n",
      "complete (86,89)\n",
      "complete (86,90)\n",
      "complete (86,91)\n",
      "complete (86,92)\n",
      "complete (86,93)\n",
      "complete (86,96)\n",
      "complete (86,97)\n",
      "complete (86,94)\n",
      "complete (86,95)\n",
      "complete (86,98)\n",
      "complete (87,88)\n",
      "complete (87,89)\n",
      "complete (87,90)\n",
      "complete (87,93)\n",
      "complete (87,91)\n",
      "complete (87,92)\n",
      "complete (87,94)\n",
      "complete (87,95)\n",
      "complete (87,96)\n",
      "complete (87,97)\n",
      "complete (88,89)\n",
      "complete (87,98)\n",
      "complete (88,90)\n",
      "complete (88,92)\n",
      "complete (88,91)\n",
      "complete (88,93)\n",
      "complete (88,94)\n",
      "complete (88,96)\n",
      "complete (88,97)\n",
      "complete (88,95)\n",
      "complete (88,98)\n",
      "complete (89,90)\n",
      "complete (89,92)\n",
      "complete (89,91)\n",
      "complete (89,93)\n",
      "complete (89,94)\n",
      "complete (89,96)\n",
      "complete (89,95)\n",
      "complete (89,97)\n",
      "complete (89,98)\n",
      "complete (90,93)\n",
      "complete (90,92)\n",
      "complete (90,91)\n",
      "complete (90,94)\n",
      "complete (90,96)\n",
      "complete (90,97)\n",
      "complete (90,95)\n",
      "complete (90,98)\n",
      "complete (91,92)\n",
      "complete (91,94)\n",
      "complete (91,93)\n",
      "complete (91,95)\n",
      "complete (91,97)\n",
      "complete (91,96)\n",
      "complete (91,98)\n",
      "complete (92,94)\n",
      "complete (92,93)\n",
      "complete (92,97)\n",
      "complete (92,96)\n",
      "complete (92,95)\n",
      "complete (93,94)\n",
      "complete (92,98)\n",
      "complete (93,95)\n",
      "complete (93,96)\n",
      "complete (93,97)\n",
      "complete (94,96)\n",
      "complete (94,97)\n",
      "complete (93,98)\n",
      "complete (94,95)\n",
      "complete (94,98)\n",
      "complete (95,96)\n",
      "complete (96,97)\n",
      "complete (95,97)\n",
      "complete (95,98)\n",
      "complete (96,98)\n",
      "complete (97,98)\n"
     ]
    }
   ],
   "source": [
    "# number of good sample, this must bigger than 50. The bigger the more time cost.\n",
    "num = 50\n",
    "\n",
    "data, num_good, num_bad = util.load_train_data(num)\n",
    "dist = np.zeros((len(data), len(data)))\n",
    "\n",
    "\n",
    "def get_dist(data, index1, index2):\n",
    "    distance, cost, path = dtw.dtw(data[index1], data[index2])\n",
    "    print('complete ({d1},{d2})'.format(d1=index1, d2=index2))\n",
    "    return distance, index1, index2\n",
    "\n",
    "\n",
    "process = [(i, j) for i in range(len(data)) for j in range(i + 1, len(data))]\n",
    "\n",
    "with futures.ProcessPoolExecutor(max_workers=8) as executor:\n",
    "    futures = [executor.submit(get_dist, data, item[0], item[1]) for item in process]\n",
    "    for future in futures:\n",
    "        distance, index1, index2 = future.result()\n",
    "        dist[index1, index2] = distance\n",
    "\n",
    "for i in range(len(data)):\n",
    "    for j in range(0, i):\n",
    "        dist[i, j] = dist[j, i]\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now we can use MDS to visualize what we calculate."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 800x500 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 432x288 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "mds = MDS(n_components=3, dissimilarity=\"precomputed\")\n",
    "pos = mds.fit_transform(dist)\n",
    "\n",
    "fig = plt.figure(figsize=(8, 5), dpi= 100)\n",
    "ax = Axes3D(fig)\n",
    "ax.scatter(pos[:num, 0], pos[:num, 1], pos[:num, 2], c='g')\n",
    "ax.scatter(pos[num:, 0], pos[num:, 1], pos[num:, 2], c='r')\n",
    "ax.set_xlabel('x')\n",
    "ax.set_ylabel('y')\n",
    "ax.set_zlabel('z')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Use KNN to see how is DTW performing."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.95"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# prepare data\n",
    "labels = np.array([1]*(len(pos)-50)+[0]*50,dtype=np.int)\n",
    "train_x = pos[10:-10, :]\n",
    "train_y = labels[10:-10]\n",
    "test_x = np.vstack((pos[:10, :], pos[-10:, :]))\n",
    "test_y = np.hstack((labels[:10], labels[-10:]))\n",
    "\n",
    "# KNN\n",
    "neigh = KNeighborsClassifier(n_neighbors=5, weights='distance')\n",
    "neigh.fit(train_x, train_y)\n",
    "neigh.score(test_x, test_y)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "tensorflow_gpu",
   "language": "python",
   "name": "tensorflow_gpu"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.8"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}