fr_project_parts_2_and_3.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# How easy is it to test the accuracy of a face recognition algorithm?\n",
    "\n",
    "## Part 2 - templating the image files to make them ready for matching\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>subject_id</th>\n",
       "      <th>image_sequence_number</th>\n",
       "      <th>gender</th>\n",
       "      <th>age</th>\n",
       "      <th>lighting</th>\n",
       "      <th>view</th>\n",
       "      <th>cropped</th>\n",
       "      <th>facial_emotion</th>\n",
       "      <th>year</th>\n",
       "      <th>part</th>\n",
       "      <th>occlusion</th>\n",
       "      <th>image_filters</th>\n",
       "      <th>level_of_difficulty</th>\n",
       "      <th>image_path</th>\n",
       "      <th>face_locations</th>\n",
       "      <th>faces_found</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>MotM</td>\n",
       "      <td>1984</td>\n",
       "      <td>m</td>\n",
       "      <td>21</td>\n",
       "      <td>i</td>\n",
       "      <td>fr</td>\n",
       "      <td>nc</td>\n",
       "      <td>sd</td>\n",
       "      <td>2015</td>\n",
       "      <td>1</td>\n",
       "      <td>e0</td>\n",
       "      <td>Gn</td>\n",
       "      <td>e</td>\n",
       "      <td>./images/MotM_01984_m_21_i_fr_nc_sd_2015_1_e0_...</td>\n",
       "      <td>[(171, 379, 439, 111)]</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>PetK</td>\n",
       "      <td>2257</td>\n",
       "      <td>m</td>\n",
       "      <td>20</td>\n",
       "      <td>i</td>\n",
       "      <td>nf</td>\n",
       "      <td>nc</td>\n",
       "      <td>no</td>\n",
       "      <td>2015</td>\n",
       "      <td>1</td>\n",
       "      <td>e0</td>\n",
       "      <td>Ps</td>\n",
       "      <td>e</td>\n",
       "      <td>./images/PetK_02257_m_20_i_nf_nc_no_2015_1_e0_...</td>\n",
       "      <td>[(118, 464, 341, 241)]</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>MosK</td>\n",
       "      <td>1975</td>\n",
       "      <td>m</td>\n",
       "      <td>20</td>\n",
       "      <td>o</td>\n",
       "      <td>nf</td>\n",
       "      <td>nc</td>\n",
       "      <td>sd</td>\n",
       "      <td>2015</td>\n",
       "      <td>1</td>\n",
       "      <td>e0</td>\n",
       "      <td>Gs</td>\n",
       "      <td>e</td>\n",
       "      <td>./images/MosK_01975_m_20_o_nf_nc_sd_2015_1_e0_...</td>\n",
       "      <td>[(142, 489, 365, 266)]</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  subject_id  image_sequence_number gender  age lighting view cropped  \\\n",
       "0       MotM                   1984      m   21        i   fr      nc   \n",
       "1       PetK                   2257      m   20        i   nf      nc   \n",
       "2       MosK                   1975      m   20        o   nf      nc   \n",
       "\n",
       "  facial_emotion  year  part occlusion image_filters level_of_difficulty  \\\n",
       "0             sd  2015     1        e0            Gn                   e   \n",
       "1             no  2015     1        e0            Ps                   e   \n",
       "2             sd  2015     1        e0            Gs                   e   \n",
       "\n",
       "                                          image_path          face_locations  \\\n",
       "0  ./images/MotM_01984_m_21_i_fr_nc_sd_2015_1_e0_...  [(171, 379, 439, 111)]   \n",
       "1  ./images/PetK_02257_m_20_i_nf_nc_no_2015_1_e0_...  [(118, 464, 341, 241)]   \n",
       "2  ./images/MosK_01975_m_20_o_nf_nc_sd_2015_1_e0_...  [(142, 489, 365, 266)]   \n",
       "\n",
       "   faces_found  \n",
       "0            1  \n",
       "1            1  \n",
       "2            1  "
      ]
     },
     "execution_count": 1,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Import pandas to read the CSV into a dataframe\n",
    "import pandas as pd\n",
    "# Import the face matcher class\n",
    "import face_matcher as fm\n",
    "matcher = fm.FaceMatcher()\n",
    "import re # for regex to reformat face locations into required format\n",
    "\n",
    "# Reading CSV and checking first few rows\n",
    "faces = pd.read_csv('./easy_faces_2015_one_face_only.csv')\n",
    "faces.head(3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(1961, 16)"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Checking shape\n",
    "faces.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "# There is a need to convert format of the face location. Since we now know it is only one face. \n",
    "faces['face_locations'].replace(to_replace ='[\\'\\[\\(]|[\\)\\]\\']', value = '', regex = True, inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "faces.rename(columns={'face_locations':'face_location'}, inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>subject_id</th>\n",
       "      <th>image_sequence_number</th>\n",
       "      <th>gender</th>\n",
       "      <th>age</th>\n",
       "      <th>lighting</th>\n",
       "      <th>view</th>\n",
       "      <th>cropped</th>\n",
       "      <th>facial_emotion</th>\n",
       "      <th>year</th>\n",
       "      <th>part</th>\n",
       "      <th>occlusion</th>\n",
       "      <th>image_filters</th>\n",
       "      <th>level_of_difficulty</th>\n",
       "      <th>image_path</th>\n",
       "      <th>face_location</th>\n",
       "      <th>faces_found</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>MotM</td>\n",
       "      <td>1984</td>\n",
       "      <td>m</td>\n",
       "      <td>21</td>\n",
       "      <td>i</td>\n",
       "      <td>fr</td>\n",
       "      <td>nc</td>\n",
       "      <td>sd</td>\n",
       "      <td>2015</td>\n",
       "      <td>1</td>\n",
       "      <td>e0</td>\n",
       "      <td>Gn</td>\n",
       "      <td>e</td>\n",
       "      <td>./images/MotM_01984_m_21_i_fr_nc_sd_2015_1_e0_...</td>\n",
       "      <td>171, 379, 439, 111</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>PetK</td>\n",
       "      <td>2257</td>\n",
       "      <td>m</td>\n",
       "      <td>20</td>\n",
       "      <td>i</td>\n",
       "      <td>nf</td>\n",
       "      <td>nc</td>\n",
       "      <td>no</td>\n",
       "      <td>2015</td>\n",
       "      <td>1</td>\n",
       "      <td>e0</td>\n",
       "      <td>Ps</td>\n",
       "      <td>e</td>\n",
       "      <td>./images/PetK_02257_m_20_i_nf_nc_no_2015_1_e0_...</td>\n",
       "      <td>118, 464, 341, 241</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>MosK</td>\n",
       "      <td>1975</td>\n",
       "      <td>m</td>\n",
       "      <td>20</td>\n",
       "      <td>o</td>\n",
       "      <td>nf</td>\n",
       "      <td>nc</td>\n",
       "      <td>sd</td>\n",
       "      <td>2015</td>\n",
       "      <td>1</td>\n",
       "      <td>e0</td>\n",
       "      <td>Gs</td>\n",
       "      <td>e</td>\n",
       "      <td>./images/MosK_01975_m_20_o_nf_nc_sd_2015_1_e0_...</td>\n",
       "      <td>142, 489, 365, 266</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  subject_id  image_sequence_number gender  age lighting view cropped  \\\n",
       "0       MotM                   1984      m   21        i   fr      nc   \n",
       "1       PetK                   2257      m   20        i   nf      nc   \n",
       "2       MosK                   1975      m   20        o   nf      nc   \n",
       "\n",
       "  facial_emotion  year  part occlusion image_filters level_of_difficulty  \\\n",
       "0             sd  2015     1        e0            Gn                   e   \n",
       "1             no  2015     1        e0            Ps                   e   \n",
       "2             sd  2015     1        e0            Gs                   e   \n",
       "\n",
       "                                          image_path       face_location  \\\n",
       "0  ./images/MotM_01984_m_21_i_fr_nc_sd_2015_1_e0_...  171, 379, 439, 111   \n",
       "1  ./images/PetK_02257_m_20_i_nf_nc_no_2015_1_e0_...  118, 464, 341, 241   \n",
       "2  ./images/MosK_01975_m_20_o_nf_nc_sd_2015_1_e0_...  142, 489, 365, 266   \n",
       "\n",
       "   faces_found  \n",
       "0            1  \n",
       "1            1  \n",
       "2            1  "
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "faces.head(3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "64"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Group data. Number of itentities\n",
    "face_groups = faces.groupby('subject_id')\n",
    "len(face_groups.groups.keys())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>subject_id</th>\n",
       "      <th>image_sequence_number</th>\n",
       "      <th>gender</th>\n",
       "      <th>age</th>\n",
       "      <th>lighting</th>\n",
       "      <th>view</th>\n",
       "      <th>cropped</th>\n",
       "      <th>facial_emotion</th>\n",
       "      <th>year</th>\n",
       "      <th>part</th>\n",
       "      <th>occlusion</th>\n",
       "      <th>image_filters</th>\n",
       "      <th>level_of_difficulty</th>\n",
       "      <th>image_path</th>\n",
       "      <th>face_location</th>\n",
       "      <th>faces_found</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>MotM</td>\n",
       "      <td>1984</td>\n",
       "      <td>m</td>\n",
       "      <td>21</td>\n",
       "      <td>i</td>\n",
       "      <td>fr</td>\n",
       "      <td>nc</td>\n",
       "      <td>sd</td>\n",
       "      <td>2015</td>\n",
       "      <td>1</td>\n",
       "      <td>e0</td>\n",
       "      <td>Gn</td>\n",
       "      <td>e</td>\n",
       "      <td>./images/MotM_01984_m_21_i_fr_nc_sd_2015_1_e0_...</td>\n",
       "      <td>171, 379, 439, 111</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>MotM</td>\n",
       "      <td>1995</td>\n",
       "      <td>m</td>\n",
       "      <td>21</td>\n",
       "      <td>o</td>\n",
       "      <td>fr</td>\n",
       "      <td>nc</td>\n",
       "      <td>no</td>\n",
       "      <td>2015</td>\n",
       "      <td>1</td>\n",
       "      <td>e0</td>\n",
       "      <td>Gs</td>\n",
       "      <td>e</td>\n",
       "      <td>./images/MotM_01995_m_21_o_fr_nc_no_2015_1_e0_...</td>\n",
       "      <td>142, 409, 409, 141</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>MotM</td>\n",
       "      <td>1994</td>\n",
       "      <td>m</td>\n",
       "      <td>21</td>\n",
       "      <td>i</td>\n",
       "      <td>fr</td>\n",
       "      <td>nc</td>\n",
       "      <td>hp</td>\n",
       "      <td>2015</td>\n",
       "      <td>1</td>\n",
       "      <td>en</td>\n",
       "      <td>nl</td>\n",
       "      <td>e</td>\n",
       "      <td>./images/MotM_01994_m_21_i_fr_nc_hp_2015_1_en_...</td>\n",
       "      <td>134, 455, 455, 134</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  subject_id  image_sequence_number gender  age lighting view cropped  \\\n",
       "0       MotM                   1984      m   21        i   fr      nc   \n",
       "5       MotM                   1995      m   21        o   fr      nc   \n",
       "8       MotM                   1994      m   21        i   fr      nc   \n",
       "\n",
       "  facial_emotion  year  part occlusion image_filters level_of_difficulty  \\\n",
       "0             sd  2015     1        e0            Gn                   e   \n",
       "5             no  2015     1        e0            Gs                   e   \n",
       "8             hp  2015     1        en            nl                   e   \n",
       "\n",
       "                                          image_path       face_location  \\\n",
       "0  ./images/MotM_01984_m_21_i_fr_nc_sd_2015_1_e0_...  171, 379, 439, 111   \n",
       "5  ./images/MotM_01995_m_21_o_fr_nc_no_2015_1_e0_...  142, 409, 409, 141   \n",
       "8  ./images/MotM_01994_m_21_i_fr_nc_hp_2015_1_en_...  134, 455, 455, 134   \n",
       "\n",
       "   faces_found  \n",
       "0            1  \n",
       "5            1  \n",
       "8            1  "
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# extract out one group as a test\n",
    "id_MotM = face_groups.get_group('MotM')\n",
    "id_MotM.head(3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'list'> <class 'numpy.ndarray'> <class 'numpy.float64'> \n",
      " [array([-1.51017696e-01,  8.16482008e-02,  7.12076873e-02, -5.81135508e-04,\n",
      "       -1.11955918e-01, -1.43141076e-02, -6.43797517e-02, -1.23610966e-01,\n",
      "        1.56638816e-01, -7.63088167e-02,  2.06025094e-01,  5.99477105e-02,\n",
      "       -2.33586609e-01, -7.22645596e-02,  8.12688321e-02,  9.96994972e-02,\n",
      "       -1.52423099e-01, -1.57981887e-01, -6.77747577e-02, -4.63530719e-02,\n",
      "        3.57958674e-02, -9.70868841e-02,  3.30040902e-02,  4.41798121e-02,\n",
      "       -2.08102554e-01, -3.49925071e-01, -1.23185083e-01, -1.33675978e-01,\n",
      "        2.48490497e-02, -9.97912735e-02,  1.19796190e-02,  4.34815884e-04,\n",
      "       -1.72491133e-01, -2.70583481e-03,  3.07523236e-02,  1.03466302e-01,\n",
      "        2.26588100e-02, -3.48281264e-02,  1.78748772e-01, -2.56113000e-02,\n",
      "       -1.85807079e-01,  5.16017526e-02,  3.97023112e-02,  1.68589398e-01,\n",
      "        2.03639880e-01,  7.55521432e-02,  9.00853127e-02, -5.27126193e-02,\n",
      "        1.29495904e-01, -2.72285551e-01,  8.21383297e-02,  1.71087533e-01,\n",
      "        7.66795576e-02,  1.73031986e-02,  1.18776336e-01, -1.83287159e-01,\n",
      "        9.76869464e-03,  1.00663319e-01, -2.06585795e-01,  1.48209408e-01,\n",
      "        8.44822973e-02, -8.38094801e-02, -4.47093844e-02,  1.60220489e-02,\n",
      "        1.64926514e-01,  1.78585410e-01, -1.34032845e-01, -9.43837538e-02,\n",
      "        1.65804103e-01, -1.75635472e-01, -1.13747446e-02,  5.77936620e-02,\n",
      "       -1.73690826e-01, -2.15612844e-01, -3.18371624e-01,  3.26333195e-02,\n",
      "        4.99514759e-01,  1.74368560e-01, -1.65371209e-01, -2.47873310e-02,\n",
      "       -1.00197859e-01,  2.37708651e-02,  2.12978572e-02,  6.08859658e-02,\n",
      "       -7.75694400e-02, -1.14504524e-01, -1.12920299e-01,  7.63175786e-02,\n",
      "        2.22726688e-01,  4.89014685e-02,  2.28678621e-02,  2.14000255e-01,\n",
      "        1.42238736e-02,  5.24349436e-02, -4.08054441e-02,  7.17068836e-02,\n",
      "       -8.79051238e-02, -1.15729682e-01, -7.35731274e-02, -2.59222686e-02,\n",
      "       -5.32650054e-02, -6.02153279e-02,  2.73739249e-02,  1.13073260e-01,\n",
      "       -1.90818846e-01,  6.03940561e-02, -2.31307000e-02, -4.29049321e-02,\n",
      "       -6.55950606e-02,  1.02491096e-01, -9.39921886e-02, -2.77810842e-02,\n",
      "        1.02151096e-01, -2.67944813e-01,  2.30834439e-01,  2.59893149e-01,\n",
      "       -6.63587824e-03,  8.71838778e-02,  2.23438218e-02,  6.16672263e-02,\n",
      "        1.13399774e-02, -3.03141475e-02, -8.89010280e-02, -9.17734951e-02,\n",
      "       -7.52507374e-02, -2.73318812e-02,  1.33560598e-02,  2.38916874e-02])]\n"
     ]
    }
   ],
   "source": [
    "# test template generation (biometric enrollment)\n",
    "image_path = id_MotM['image_path'][0]\n",
    "location = id_MotM['face_location'][0].split(\", \")\n",
    "template = matcher.create_template(image_path, location)\n",
    "print(type(template), type(template[0]), type(template[0][0]), \"\\n\", template)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[2,\n",
       " ('./images/MosK_01975_m_20_o_nf_nc_sd_2015_1_e0_Gs_e.jpg',\n",
       "  ['142', '489', '365', '266'])]"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# We should be able to create a dictionary referencing the index in the dataframe \n",
    "# to the biometric template to be used for match. We can reference the data frame to \n",
    "# know the identity of the person in the face\n",
    "enrollment_list = []\n",
    "for index, row in faces.iterrows():\n",
    "     enrollment_list.append([index, (row['image_path'], row['face_location'].split(\", \"))])\n",
    "enrollment_list[2]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Enrolled... 0\n",
      "Enrolled... 100\n",
      "Enrolled... 200\n",
      "Enrolled... 300\n",
      "Enrolled... 400\n",
      "Enrolled... 500\n",
      "Enrolled... 600\n",
      "Enrolled... 700\n",
      "Enrolled... 800\n",
      "Enrolled... 900\n",
      "Enrolled... 1000\n",
      "Enrolled... 1100\n",
      "Enrolled... 1200\n",
      "Enrolled... 1300\n",
      "Enrolled... 1400\n",
      "Enrolled... 1500\n",
      "Enrolled... 1600\n",
      "Enrolled... 1700\n",
      "Enrolled... 1800\n",
      "Enrolled... 1900\n",
      "Finished enrolling 1961 in 55 seconds\n"
     ]
    }
   ],
   "source": [
    "# Now can pass this in as an argument to enroll all items and store in a dict\n",
    "enrollments = matcher.enroll_batch(enrollment_list)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Part 3 - matching to generate scores\n",
    "#### Unfortunately this algorithm does not generate a matching score out of the box. However we can tune a tolerance setting which will allow us to generate a score by attempting a match until the face_recognition module switch the result from True to False\n",
    "#### Since mated pairs are more likely to match, we should start the process from 1.0 and work downwards. For dissimilar pairs we should start at 0.0 and work upwards"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "MotM PetK\n",
      "[False] [True]\n",
      "./images/MotM_01984_m_21_i_fr_nc_sd_2015_1_e0_Gn_e.jpg ./images/PetK_02257_m_20_i_nf_nc_no_2015_1_e0_Ps_e.jpg\n"
     ]
    }
   ],
   "source": [
    "# Dissimilar pairs\n",
    "print(faces['subject_id'][0], faces['subject_id'][1])\n",
    "print(matcher.single_match(enrollments[0], enrollments[1], threshold=0.24),\n",
    "      matcher.single_match(enrollments[0], enrollments[1], threshold=0.23))\n",
    "print(faces['image_path'][0], faces['image_path'][1])\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### From trial and error we can see these images match around a score of 0.23\n",
    "##### and we know it is a different person from the ground truth\n",
    "<p float=\"left\">\n",
    "  <img style=\"display: inline;\" src=\"./images/MotM_01984_m_21_i_fr_nc_sd_2015_1_e0_Gn_e.jpg\" width=\"200\" />\n",
    "  <img style=\"display: inline;\" src=\"./images/PetK_02257_m_20_i_nf_nc_no_2015_1_e0_Ps_e.jpg\" width=\"200\" /> \n",
    "</p>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "MotM MotM\n",
      "[False] [True]\n",
      "./images/MotM_01984_m_21_i_fr_nc_sd_2015_1_e0_Gn_e.jpg ./images/MotM_01994_m_21_i_fr_nc_hp_2015_1_en_nl_e.jpg\n"
     ]
    }
   ],
   "source": [
    "# Similar pairs \n",
    "print(faces['subject_id'][0], faces['subject_id'][8])\n",
    "print(matcher.single_match(enrollments[0], enrollments[8], threshold=0.60),\n",
    "      matcher.single_match(enrollments[0], enrollments[8], threshold=0.59))\n",
    "print(faces['image_path'][0], faces['image_path'][8])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### From trial and error we can see these images match around a score of 0.59\n",
    "##### and we know it is the same person from the ground truth\n",
    "<p float=\"left\">\n",
    "  <img style=\"display: inline;\" src=\"./images/MotM_01984_m_21_i_fr_nc_sd_2015_1_e0_Gn_e.jpg\" width=\"200\" />\n",
    "  <img style=\"display: inline;\" src=\"./images/MotM_01994_m_21_i_fr_nc_hp_2015_1_en_nl_e.jpg\" width=\"200\" /> \n",
    "</p>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "testing threshold... 0.00\n",
      "testing threshold... 0.01\n",
      "testing threshold... 0.02\n",
      "testing threshold... 0.03\n",
      "testing threshold... 0.04\n",
      "testing threshold... 0.05\n",
      "testing threshold... 0.06\n",
      "testing threshold... 0.07\n",
      "testing threshold... 0.08\n",
      "testing threshold... 0.09\n",
      "testing threshold... 0.10\n",
      "testing threshold... 0.11\n",
      "testing threshold... 0.12\n",
      "testing threshold... 0.13\n",
      "testing threshold... 0.14\n",
      "testing threshold... 0.15\n",
      "testing threshold... 0.16\n",
      "testing threshold... 0.17\n",
      "testing threshold... 0.18\n",
      "testing threshold... 0.19\n",
      "testing threshold... 0.20\n",
      "testing threshold... 0.21\n",
      "testing threshold... 0.22\n",
      "testing threshold... 0.23\n",
      "testing threshold... 0.24\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "'0.24'"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Matching a non-mated pair\n",
    "matcher.generate_score(enrollments[0], enrollments[1], mate=False, logging=True, new_threshold=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "testing threshold... 1.00\n",
      "testing threshold... 0.99\n",
      "testing threshold... 0.98\n",
      "testing threshold... 0.97\n",
      "testing threshold... 0.96\n",
      "testing threshold... 0.95\n",
      "testing threshold... 0.94\n",
      "testing threshold... 0.93\n",
      "testing threshold... 0.92\n",
      "testing threshold... 0.91\n",
      "testing threshold... 0.90\n",
      "testing threshold... 0.89\n",
      "testing threshold... 0.88\n",
      "testing threshold... 0.87\n",
      "testing threshold... 0.86\n",
      "testing threshold... 0.85\n",
      "testing threshold... 0.84\n",
      "testing threshold... 0.83\n",
      "testing threshold... 0.82\n",
      "testing threshold... 0.81\n",
      "testing threshold... 0.80\n",
      "testing threshold... 0.79\n",
      "testing threshold... 0.78\n",
      "testing threshold... 0.77\n",
      "testing threshold... 0.76\n",
      "testing threshold... 0.75\n",
      "testing threshold... 0.74\n",
      "testing threshold... 0.73\n",
      "testing threshold... 0.72\n",
      "testing threshold... 0.71\n",
      "testing threshold... 0.70\n",
      "testing threshold... 0.69\n",
      "testing threshold... 0.68\n",
      "testing threshold... 0.67\n",
      "testing threshold... 0.66\n",
      "testing threshold... 0.65\n",
      "testing threshold... 0.64\n",
      "testing threshold... 0.63\n",
      "testing threshold... 0.62\n",
      "testing threshold... 0.61\n",
      "testing threshold... 0.60\n",
      "testing threshold... 0.59\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "'0.59'"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Matching a mated pair\n",
    "matcher.generate_score(enrollments[0], enrollments[8], mate=True, logging=True, new_threshold=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[0, 5] [0, 8] [0, 50]\n"
     ]
    }
   ],
   "source": [
    "# Now we can generate a CSV file for matches of mates and non-mates\n",
    "# It should look like\n",
    "# subject_id_1, subject_id_2, score\n",
    "# Get the unique groups as batches of mated images\n",
    "group_keys = faces.subject_id.unique\n",
    "pairs = []\n",
    "# Loop over each group\n",
    "for group in group_keys():\n",
    "    # Extract the data per group\n",
    "    group_data = faces[faces['subject_id'] == group]\n",
    "    # Get all indexes to loop over them    \n",
    "    indexes = group_data.index.values.tolist()\n",
    "    for face_num in indexes:\n",
    "        for other_face_num in indexes:\n",
    "            if(face_num != other_face_num): # Only match pairs that aren't the same image\n",
    "                pairs.append([face_num, other_face_num])\n",
    "\n",
    "print(pairs[0], pairs[1], pairs[2])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0 matches completed\n",
      "5000 matches completed\n",
      "10000 matches completed\n",
      "15000 matches completed\n",
      "20000 matches completed\n",
      "25000 matches completed\n",
      "30000 matches completed\n",
      "35000 matches completed\n",
      "40000 matches completed\n",
      "45000 matches completed\n",
      "50000 matches completed\n",
      "55000 matches completed\n",
      "60000 matches completed\n",
      "65000 matches completed\n",
      "70000 matches completed\n",
      "75000 matches completed\n",
      "80000 matches completed\n",
      "85000 matches completed\n",
      "90000 matches completed\n",
      "95000 matches completed\n",
      "100000 matches completed\n",
      "105000 matches completed\n",
      "110000 matches completed\n",
      "115000 matches completed\n",
      "120000 matches completed\n",
      "125000 matches completed\n",
      "130000 matches completed\n",
      "Finished matching pairs in 104 seconds\n"
     ]
    }
   ],
   "source": [
    "# Now there is a list of tuple of image pairs to match where we can be assured they are mated pairs\n",
    "matcher.cross_match_pairs(enrollments, pairs, 'mates_matched.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[0, 1] [0, 2] [0, 3] [0, 4] [0, 6]\n"
     ]
    }
   ],
   "source": [
    "# Matching imposters takes the identities that arent the same person\n",
    "imposters = []\n",
    "# Loop over each group\n",
    "for group in group_keys():\n",
    "    # Extract the data per group\n",
    "    group_data = faces[faces['subject_id'] == group]\n",
    "    imposter_data = faces[faces['subject_id'] != group]\n",
    "    # Get all indexes to loop over them \n",
    "    indexes = group_data.index.values.tolist()\n",
    "    indexes_imposters = imposter_data.index.values.tolist()\n",
    "    # print(indexes)\n",
    "    for face_num in indexes:\n",
    "        # print(face_num)\n",
    "        for imposter in indexes_imposters:\n",
    "            imposters.append([face_num, imposter])\n",
    "\n",
    "print(imposters[0], imposters[1], imposters[2], imposters[3], imposters[4])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0 matches completed\n",
      "5000 matches completed\n",
      "10000 matches completed\n",
      "15000 matches completed\n",
      "20000 matches completed\n",
      "25000 matches completed\n",
      "30000 matches completed\n",
      "35000 matches completed\n",
      "40000 matches completed\n",
      "45000 matches completed\n",
      "50000 matches completed\n",
      "55000 matches completed\n",
      "60000 matches completed\n",
      "65000 matches completed\n",
      "70000 matches completed\n",
      "75000 matches completed\n",
      "80000 matches completed\n",
      "85000 matches completed\n",
      "90000 matches completed\n",
      "95000 matches completed\n",
      "100000 matches completed\n",
      "105000 matches completed\n",
      "110000 matches completed\n",
      "115000 matches completed\n",
      "120000 matches completed\n",
      "125000 matches completed\n",
      "130000 matches completed\n",
      "135000 matches completed\n",
      "140000 matches completed\n",
      "145000 matches completed\n",
      "150000 matches completed\n",
      "155000 matches completed\n",
      "160000 matches completed\n",
      "165000 matches completed\n",
      "170000 matches completed\n",
      "175000 matches completed\n",
      "180000 matches completed\n",
      "185000 matches completed\n",
      "190000 matches completed\n",
      "195000 matches completed\n",
      "200000 matches completed\n",
      "205000 matches completed\n",
      "210000 matches completed\n",
      "215000 matches completed\n",
      "220000 matches completed\n",
      "225000 matches completed\n",
      "230000 matches completed\n",
      "235000 matches completed\n",
      "240000 matches completed\n",
      "245000 matches completed\n",
      "250000 matches completed\n",
      "255000 matches completed\n",
      "260000 matches completed\n",
      "265000 matches completed\n",
      "270000 matches completed\n",
      "275000 matches completed\n",
      "280000 matches completed\n",
      "285000 matches completed\n",
      "290000 matches completed\n",
      "295000 matches completed\n",
      "300000 matches completed\n",
      "305000 matches completed\n",
      "310000 matches completed\n",
      "315000 matches completed\n",
      "320000 matches completed\n",
      "325000 matches completed\n",
      "330000 matches completed\n",
      "335000 matches completed\n",
      "340000 matches completed\n",
      "345000 matches completed\n",
      "350000 matches completed\n",
      "355000 matches completed\n",
      "360000 matches completed\n",
      "365000 matches completed\n",
      "370000 matches completed\n",
      "375000 matches completed\n",
      "380000 matches completed\n",
      "385000 matches completed\n",
      "390000 matches completed\n",
      "395000 matches completed\n",
      "400000 matches completed\n",
      "405000 matches completed\n",
      "410000 matches completed\n",
      "415000 matches completed\n",
      "420000 matches completed\n",
      "425000 matches completed\n",
      "430000 matches completed\n",
      "435000 matches completed\n",
      "440000 matches completed\n",
      "445000 matches completed\n",
      "450000 matches completed\n",
      "455000 matches completed\n",
      "460000 matches completed\n",
      "465000 matches completed\n",
      "470000 matches completed\n",
      "475000 matches completed\n",
      "480000 matches completed\n",
      "485000 matches completed\n",
      "490000 matches completed\n",
      "495000 matches completed\n",
      "500000 matches completed\n",
      "505000 matches completed\n",
      "510000 matches completed\n",
      "515000 matches completed\n",
      "520000 matches completed\n",
      "525000 matches completed\n",
      "530000 matches completed\n",
      "535000 matches completed\n",
      "540000 matches completed\n",
      "545000 matches completed\n",
      "550000 matches completed\n",
      "555000 matches completed\n",
      "560000 matches completed\n",
      "565000 matches completed\n",
      "570000 matches completed\n",
      "575000 matches completed\n",
      "580000 matches completed\n",
      "585000 matches completed\n",
      "590000 matches completed\n",
      "595000 matches completed\n",
      "600000 matches completed\n",
      "605000 matches completed\n",
      "610000 matches completed\n",
      "615000 matches completed\n",
      "620000 matches completed\n",
      "625000 matches completed\n",
      "630000 matches completed\n",
      "635000 matches completed\n",
      "640000 matches completed\n",
      "645000 matches completed\n",
      "650000 matches completed\n",
      "655000 matches completed\n",
      "660000 matches completed\n",
      "665000 matches completed\n",
      "670000 matches completed\n",
      "675000 matches completed\n",
      "680000 matches completed\n",
      "685000 matches completed\n",
      "690000 matches completed\n",
      "695000 matches completed\n",
      "700000 matches completed\n",
      "705000 matches completed\n",
      "710000 matches completed\n",
      "715000 matches completed\n",
      "720000 matches completed\n",
      "725000 matches completed\n",
      "730000 matches completed\n",
      "735000 matches completed\n",
      "740000 matches completed\n",
      "745000 matches completed\n",
      "750000 matches completed\n",
      "755000 matches completed\n",
      "760000 matches completed\n",
      "765000 matches completed\n",
      "770000 matches completed\n",
      "775000 matches completed\n",
      "780000 matches completed\n",
      "785000 matches completed\n",
      "790000 matches completed\n",
      "795000 matches completed\n",
      "800000 matches completed\n",
      "805000 matches completed\n",
      "810000 matches completed\n",
      "815000 matches completed\n",
      "820000 matches completed\n",
      "825000 matches completed\n",
      "830000 matches completed\n",
      "835000 matches completed\n",
      "840000 matches completed\n",
      "845000 matches completed\n",
      "850000 matches completed\n",
      "855000 matches completed\n",
      "860000 matches completed\n",
      "865000 matches completed\n",
      "870000 matches completed\n",
      "875000 matches completed\n",
      "880000 matches completed\n",
      "885000 matches completed\n",
      "890000 matches completed\n",
      "895000 matches completed\n",
      "900000 matches completed\n",
      "905000 matches completed\n",
      "910000 matches completed\n",
      "915000 matches completed\n",
      "920000 matches completed\n",
      "925000 matches completed\n",
      "930000 matches completed\n",
      "935000 matches completed\n",
      "940000 matches completed\n",
      "945000 matches completed\n",
      "950000 matches completed\n",
      "955000 matches completed\n",
      "960000 matches completed\n",
      "965000 matches completed\n",
      "970000 matches completed\n",
      "975000 matches completed\n",
      "980000 matches completed\n",
      "985000 matches completed\n",
      "990000 matches completed\n",
      "995000 matches completed\n",
      "1000000 matches completed\n",
      "1005000 matches completed\n",
      "1010000 matches completed\n",
      "1015000 matches completed\n",
      "1020000 matches completed\n",
      "1025000 matches completed\n",
      "1030000 matches completed\n",
      "1035000 matches completed\n",
      "1040000 matches completed\n",
      "1045000 matches completed\n",
      "1050000 matches completed\n",
      "1055000 matches completed\n",
      "1060000 matches completed\n",
      "1065000 matches completed\n",
      "1070000 matches completed\n",
      "1075000 matches completed\n",
      "1080000 matches completed\n",
      "1085000 matches completed\n",
      "1090000 matches completed\n",
      "1095000 matches completed\n",
      "1100000 matches completed\n",
      "1105000 matches completed\n",
      "1110000 matches completed\n",
      "1115000 matches completed\n",
      "1120000 matches completed\n",
      "1125000 matches completed\n",
      "1130000 matches completed\n",
      "1135000 matches completed\n",
      "1140000 matches completed\n",
      "1145000 matches completed\n",
      "1150000 matches completed\n",
      "1155000 matches completed\n",
      "1160000 matches completed\n",
      "1165000 matches completed\n",
      "1170000 matches completed\n",
      "1175000 matches completed\n",
      "1180000 matches completed\n",
      "1185000 matches completed\n",
      "1190000 matches completed\n",
      "1195000 matches completed\n",
      "1200000 matches completed\n",
      "1205000 matches completed\n",
      "1210000 matches completed\n",
      "1215000 matches completed\n",
      "1220000 matches completed\n",
      "1225000 matches completed\n",
      "1230000 matches completed\n",
      "1235000 matches completed\n",
      "1240000 matches completed\n",
      "1245000 matches completed\n",
      "1250000 matches completed\n",
      "1255000 matches completed\n",
      "1260000 matches completed\n",
      "1265000 matches completed\n",
      "1270000 matches completed\n",
      "1275000 matches completed\n",
      "1280000 matches completed\n",
      "1285000 matches completed\n",
      "1290000 matches completed\n",
      "1295000 matches completed\n",
      "1300000 matches completed\n",
      "1305000 matches completed\n",
      "1310000 matches completed\n",
      "1315000 matches completed\n",
      "1320000 matches completed\n",
      "1325000 matches completed\n",
      "1330000 matches completed\n",
      "1335000 matches completed\n",
      "1340000 matches completed\n",
      "1345000 matches completed\n",
      "1350000 matches completed\n",
      "1355000 matches completed\n",
      "1360000 matches completed\n",
      "1365000 matches completed\n",
      "1370000 matches completed\n",
      "1375000 matches completed\n",
      "1380000 matches completed\n",
      "1385000 matches completed\n",
      "1390000 matches completed\n",
      "1395000 matches completed\n",
      "1400000 matches completed\n",
      "1405000 matches completed\n",
      "1410000 matches completed\n",
      "1415000 matches completed\n",
      "1420000 matches completed\n",
      "1425000 matches completed\n",
      "1430000 matches completed\n",
      "1435000 matches completed\n",
      "1440000 matches completed\n",
      "1445000 matches completed\n",
      "1450000 matches completed\n",
      "1455000 matches completed\n",
      "1460000 matches completed\n",
      "1465000 matches completed\n",
      "1470000 matches completed\n",
      "1475000 matches completed\n",
      "1480000 matches completed\n",
      "1485000 matches completed\n",
      "1490000 matches completed\n",
      "1495000 matches completed\n",
      "1500000 matches completed\n",
      "1505000 matches completed\n",
      "1510000 matches completed\n",
      "1515000 matches completed\n",
      "1520000 matches completed\n",
      "1525000 matches completed\n",
      "1530000 matches completed\n",
      "1535000 matches completed\n",
      "1540000 matches completed\n",
      "1545000 matches completed\n",
      "1550000 matches completed\n",
      "1555000 matches completed\n",
      "1560000 matches completed\n",
      "1565000 matches completed\n",
      "1570000 matches completed\n",
      "1575000 matches completed\n",
      "1580000 matches completed\n",
      "1585000 matches completed\n",
      "1590000 matches completed\n",
      "1595000 matches completed\n",
      "1600000 matches completed\n",
      "1605000 matches completed\n",
      "1610000 matches completed\n",
      "1615000 matches completed\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1620000 matches completed\n",
      "1625000 matches completed\n",
      "1630000 matches completed\n",
      "1635000 matches completed\n",
      "1640000 matches completed\n",
      "1645000 matches completed\n",
      "1650000 matches completed\n",
      "1655000 matches completed\n",
      "1660000 matches completed\n",
      "1665000 matches completed\n",
      "1670000 matches completed\n",
      "1675000 matches completed\n",
      "1680000 matches completed\n",
      "1685000 matches completed\n",
      "1690000 matches completed\n",
      "1695000 matches completed\n",
      "1700000 matches completed\n",
      "1705000 matches completed\n",
      "1710000 matches completed\n",
      "1715000 matches completed\n",
      "1720000 matches completed\n",
      "1725000 matches completed\n",
      "1730000 matches completed\n",
      "1735000 matches completed\n",
      "1740000 matches completed\n",
      "1745000 matches completed\n",
      "1750000 matches completed\n",
      "1755000 matches completed\n",
      "1760000 matches completed\n",
      "1765000 matches completed\n",
      "1770000 matches completed\n",
      "1775000 matches completed\n",
      "1780000 matches completed\n",
      "1785000 matches completed\n",
      "1790000 matches completed\n",
      "1795000 matches completed\n",
      "1800000 matches completed\n",
      "1805000 matches completed\n",
      "1810000 matches completed\n",
      "1815000 matches completed\n",
      "1820000 matches completed\n",
      "1825000 matches completed\n",
      "1830000 matches completed\n",
      "1835000 matches completed\n",
      "1840000 matches completed\n",
      "1845000 matches completed\n",
      "1850000 matches completed\n",
      "1855000 matches completed\n",
      "1860000 matches completed\n",
      "1865000 matches completed\n",
      "1870000 matches completed\n",
      "1875000 matches completed\n",
      "1880000 matches completed\n",
      "1885000 matches completed\n",
      "1890000 matches completed\n",
      "1895000 matches completed\n",
      "1900000 matches completed\n",
      "1905000 matches completed\n",
      "1910000 matches completed\n",
      "1915000 matches completed\n",
      "1920000 matches completed\n",
      "1925000 matches completed\n",
      "1930000 matches completed\n",
      "1935000 matches completed\n",
      "1940000 matches completed\n",
      "1945000 matches completed\n",
      "1950000 matches completed\n",
      "1955000 matches completed\n",
      "1960000 matches completed\n",
      "1965000 matches completed\n",
      "1970000 matches completed\n",
      "1975000 matches completed\n",
      "1980000 matches completed\n",
      "1985000 matches completed\n",
      "1990000 matches completed\n",
      "1995000 matches completed\n",
      "2000000 matches completed\n",
      "2005000 matches completed\n",
      "2010000 matches completed\n",
      "2015000 matches completed\n",
      "2020000 matches completed\n",
      "2025000 matches completed\n",
      "2030000 matches completed\n",
      "2035000 matches completed\n",
      "2040000 matches completed\n",
      "2045000 matches completed\n",
      "2050000 matches completed\n",
      "2055000 matches completed\n",
      "2060000 matches completed\n",
      "2065000 matches completed\n",
      "2070000 matches completed\n",
      "2075000 matches completed\n",
      "2080000 matches completed\n",
      "2085000 matches completed\n",
      "2090000 matches completed\n",
      "2095000 matches completed\n",
      "2100000 matches completed\n",
      "2105000 matches completed\n",
      "2110000 matches completed\n",
      "2115000 matches completed\n",
      "2120000 matches completed\n",
      "2125000 matches completed\n",
      "2130000 matches completed\n",
      "2135000 matches completed\n",
      "2140000 matches completed\n",
      "2145000 matches completed\n",
      "2150000 matches completed\n",
      "2155000 matches completed\n",
      "2160000 matches completed\n",
      "2165000 matches completed\n",
      "2170000 matches completed\n",
      "2175000 matches completed\n",
      "2180000 matches completed\n",
      "2185000 matches completed\n",
      "2190000 matches completed\n",
      "2195000 matches completed\n",
      "2200000 matches completed\n",
      "2205000 matches completed\n",
      "2210000 matches completed\n",
      "2215000 matches completed\n",
      "2220000 matches completed\n",
      "2225000 matches completed\n",
      "2230000 matches completed\n",
      "2235000 matches completed\n",
      "2240000 matches completed\n",
      "2245000 matches completed\n",
      "2250000 matches completed\n",
      "2255000 matches completed\n",
      "2260000 matches completed\n",
      "2265000 matches completed\n",
      "2270000 matches completed\n",
      "2275000 matches completed\n",
      "2280000 matches completed\n",
      "2285000 matches completed\n",
      "2290000 matches completed\n",
      "2295000 matches completed\n",
      "2300000 matches completed\n",
      "2305000 matches completed\n",
      "2310000 matches completed\n",
      "2315000 matches completed\n",
      "2320000 matches completed\n",
      "2325000 matches completed\n",
      "2330000 matches completed\n",
      "2335000 matches completed\n",
      "2340000 matches completed\n",
      "2345000 matches completed\n",
      "2350000 matches completed\n",
      "2355000 matches completed\n",
      "2360000 matches completed\n",
      "2365000 matches completed\n",
      "2370000 matches completed\n",
      "2375000 matches completed\n",
      "2380000 matches completed\n",
      "2385000 matches completed\n",
      "2390000 matches completed\n",
      "2395000 matches completed\n",
      "2400000 matches completed\n",
      "2405000 matches completed\n",
      "2410000 matches completed\n",
      "2415000 matches completed\n",
      "2420000 matches completed\n",
      "2425000 matches completed\n",
      "2430000 matches completed\n",
      "2435000 matches completed\n",
      "2440000 matches completed\n",
      "2445000 matches completed\n",
      "2450000 matches completed\n",
      "2455000 matches completed\n",
      "2460000 matches completed\n",
      "2465000 matches completed\n",
      "2470000 matches completed\n",
      "2475000 matches completed\n",
      "2480000 matches completed\n",
      "2485000 matches completed\n",
      "2490000 matches completed\n",
      "2495000 matches completed\n",
      "2500000 matches completed\n",
      "2505000 matches completed\n",
      "2510000 matches completed\n",
      "2515000 matches completed\n",
      "2520000 matches completed\n",
      "2525000 matches completed\n",
      "2530000 matches completed\n",
      "2535000 matches completed\n",
      "2540000 matches completed\n",
      "2545000 matches completed\n",
      "2550000 matches completed\n",
      "2555000 matches completed\n",
      "2560000 matches completed\n",
      "2565000 matches completed\n",
      "2570000 matches completed\n",
      "2575000 matches completed\n",
      "2580000 matches completed\n",
      "2585000 matches completed\n",
      "2590000 matches completed\n",
      "2595000 matches completed\n",
      "2600000 matches completed\n",
      "2605000 matches completed\n",
      "2610000 matches completed\n",
      "2615000 matches completed\n",
      "2620000 matches completed\n",
      "2625000 matches completed\n",
      "2630000 matches completed\n",
      "2635000 matches completed\n",
      "2640000 matches completed\n",
      "2645000 matches completed\n",
      "2650000 matches completed\n",
      "2655000 matches completed\n",
      "2660000 matches completed\n",
      "2665000 matches completed\n",
      "2670000 matches completed\n",
      "2675000 matches completed\n",
      "2680000 matches completed\n",
      "2685000 matches completed\n",
      "2690000 matches completed\n",
      "2695000 matches completed\n",
      "2700000 matches completed\n",
      "2705000 matches completed\n",
      "2710000 matches completed\n",
      "2715000 matches completed\n",
      "2720000 matches completed\n",
      "2725000 matches completed\n",
      "2730000 matches completed\n",
      "2735000 matches completed\n",
      "2740000 matches completed\n",
      "2745000 matches completed\n",
      "2750000 matches completed\n",
      "2755000 matches completed\n",
      "2760000 matches completed\n",
      "2765000 matches completed\n",
      "2770000 matches completed\n",
      "2775000 matches completed\n",
      "2780000 matches completed\n",
      "2785000 matches completed\n",
      "2790000 matches completed\n",
      "2795000 matches completed\n",
      "2800000 matches completed\n",
      "2805000 matches completed\n",
      "2810000 matches completed\n",
      "2815000 matches completed\n",
      "2820000 matches completed\n",
      "2825000 matches completed\n",
      "2830000 matches completed\n",
      "2835000 matches completed\n",
      "2840000 matches completed\n",
      "2845000 matches completed\n",
      "2850000 matches completed\n",
      "2855000 matches completed\n",
      "2860000 matches completed\n",
      "2865000 matches completed\n",
      "2870000 matches completed\n",
      "2875000 matches completed\n",
      "2880000 matches completed\n",
      "2885000 matches completed\n",
      "2890000 matches completed\n",
      "2895000 matches completed\n",
      "2900000 matches completed\n",
      "2905000 matches completed\n",
      "2910000 matches completed\n",
      "2915000 matches completed\n",
      "2920000 matches completed\n",
      "2925000 matches completed\n",
      "2930000 matches completed\n",
      "2935000 matches completed\n",
      "2940000 matches completed\n",
      "2945000 matches completed\n",
      "2950000 matches completed\n",
      "2955000 matches completed\n",
      "2960000 matches completed\n",
      "2965000 matches completed\n",
      "2970000 matches completed\n",
      "2975000 matches completed\n",
      "2980000 matches completed\n",
      "2985000 matches completed\n",
      "2990000 matches completed\n",
      "2995000 matches completed\n",
      "3000000 matches completed\n",
      "3005000 matches completed\n",
      "3010000 matches completed\n",
      "3015000 matches completed\n",
      "3020000 matches completed\n",
      "3025000 matches completed\n",
      "3030000 matches completed\n",
      "3035000 matches completed\n",
      "3040000 matches completed\n",
      "3045000 matches completed\n",
      "3050000 matches completed\n",
      "3055000 matches completed\n",
      "3060000 matches completed\n",
      "3065000 matches completed\n",
      "3070000 matches completed\n",
      "3075000 matches completed\n",
      "3080000 matches completed\n",
      "3085000 matches completed\n",
      "3090000 matches completed\n",
      "3095000 matches completed\n",
      "3100000 matches completed\n",
      "3105000 matches completed\n",
      "3110000 matches completed\n",
      "3115000 matches completed\n",
      "3120000 matches completed\n",
      "3125000 matches completed\n",
      "3130000 matches completed\n",
      "3135000 matches completed\n",
      "3140000 matches completed\n",
      "3145000 matches completed\n",
      "3150000 matches completed\n",
      "3155000 matches completed\n",
      "3160000 matches completed\n",
      "3165000 matches completed\n",
      "3170000 matches completed\n",
      "3175000 matches completed\n",
      "3180000 matches completed\n",
      "3185000 matches completed\n",
      "3190000 matches completed\n",
      "3195000 matches completed\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "3200000 matches completed\n",
      "3205000 matches completed\n",
      "3210000 matches completed\n",
      "3215000 matches completed\n",
      "3220000 matches completed\n",
      "3225000 matches completed\n",
      "3230000 matches completed\n",
      "3235000 matches completed\n",
      "3240000 matches completed\n",
      "3245000 matches completed\n",
      "3250000 matches completed\n",
      "3255000 matches completed\n",
      "3260000 matches completed\n",
      "3265000 matches completed\n",
      "3270000 matches completed\n",
      "3275000 matches completed\n",
      "3280000 matches completed\n",
      "3285000 matches completed\n",
      "3290000 matches completed\n",
      "3295000 matches completed\n",
      "3300000 matches completed\n",
      "3305000 matches completed\n",
      "3310000 matches completed\n",
      "3315000 matches completed\n",
      "3320000 matches completed\n",
      "3325000 matches completed\n",
      "3330000 matches completed\n",
      "3335000 matches completed\n",
      "3340000 matches completed\n",
      "3345000 matches completed\n",
      "3350000 matches completed\n",
      "3355000 matches completed\n",
      "3360000 matches completed\n",
      "3365000 matches completed\n",
      "3370000 matches completed\n",
      "3375000 matches completed\n",
      "3380000 matches completed\n",
      "3385000 matches completed\n",
      "3390000 matches completed\n",
      "3395000 matches completed\n",
      "3400000 matches completed\n",
      "3405000 matches completed\n",
      "3410000 matches completed\n",
      "3415000 matches completed\n",
      "3420000 matches completed\n",
      "3425000 matches completed\n",
      "3430000 matches completed\n",
      "3435000 matches completed\n",
      "3440000 matches completed\n",
      "3445000 matches completed\n",
      "3450000 matches completed\n",
      "3455000 matches completed\n",
      "3460000 matches completed\n",
      "3465000 matches completed\n",
      "3470000 matches completed\n",
      "3475000 matches completed\n",
      "3480000 matches completed\n",
      "3485000 matches completed\n",
      "3490000 matches completed\n",
      "3495000 matches completed\n",
      "3500000 matches completed\n",
      "3505000 matches completed\n",
      "3510000 matches completed\n",
      "3515000 matches completed\n",
      "3520000 matches completed\n",
      "3525000 matches completed\n",
      "3530000 matches completed\n",
      "3535000 matches completed\n",
      "3540000 matches completed\n",
      "3545000 matches completed\n",
      "3550000 matches completed\n",
      "3555000 matches completed\n",
      "3560000 matches completed\n",
      "3565000 matches completed\n",
      "3570000 matches completed\n",
      "3575000 matches completed\n",
      "3580000 matches completed\n",
      "3585000 matches completed\n",
      "3590000 matches completed\n",
      "3595000 matches completed\n",
      "3600000 matches completed\n",
      "3605000 matches completed\n",
      "3610000 matches completed\n",
      "3615000 matches completed\n",
      "3620000 matches completed\n",
      "3625000 matches completed\n",
      "3630000 matches completed\n",
      "3635000 matches completed\n",
      "3640000 matches completed\n",
      "3645000 matches completed\n",
      "3650000 matches completed\n",
      "3655000 matches completed\n",
      "3660000 matches completed\n",
      "3665000 matches completed\n",
      "3670000 matches completed\n",
      "3675000 matches completed\n",
      "3680000 matches completed\n",
      "3685000 matches completed\n",
      "3690000 matches completed\n",
      "3695000 matches completed\n",
      "3700000 matches completed\n",
      "3705000 matches completed\n",
      "3710000 matches completed\n",
      "Finished matching pairs in 4585 seconds\n"
     ]
    }
   ],
   "source": [
    "matcher.cross_match_pairs(enrollments, imposters, 'imposters_matched.csv')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### Please refer to part 4 for a brief analysis of scores generated from this dataset...\n"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.5.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}