Face Recognition.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "ddc3fb93",
   "metadata": {},
   "source": [
    "# Setup"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "346f71b5",
   "metadata": {},
   "source": [
    "## 1.1 Install Dependencies"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "697f870f",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Requirement already satisfied: tensorflow==2.14.0 in c:\\users\\hp\\appdata\\local\\programs\\python\\python311\\lib\\site-packages (2.14.0)\n",
      "Requirement already satisfied: opencv-python in c:\\users\\hp\\appdata\\local\\programs\\python\\python311\\lib\\site-packages (4.8.0.76)\n",
      "Requirement already satisfied: matplotlib in c:\\users\\hp\\appdata\\local\\programs\\python\\python311\\lib\\site-packages (3.7.1)\n",
      "Requirement already satisfied: tensorflow-intel==2.14.0 in c:\\users\\hp\\appdata\\local\\programs\\python\\python311\\lib\\site-packages (from tensorflow==2.14.0) (2.14.0)\n",
      "Requirement already satisfied: absl-py>=1.0.0 in c:\\users\\hp\\appdata\\local\\programs\\python\\python311\\lib\\site-packages (from tensorflow-intel==2.14.0->tensorflow==2.14.0) (2.0.0)\n",
      "Requirement already satisfied: astunparse>=1.6.0 in c:\\users\\hp\\appdata\\local\\programs\\python\\python311\\lib\\site-packages (from tensorflow-intel==2.14.0->tensorflow==2.14.0) (1.6.3)\n",
      "Requirement already satisfied: flatbuffers>=23.5.26 in c:\\users\\hp\\appdata\\local\\programs\\python\\python311\\lib\\site-packages (from tensorflow-intel==2.14.0->tensorflow==2.14.0) (23.5.26)\n",
      "Requirement already satisfied: gast!=0.5.0,!=0.5.1,!=0.5.2,>=0.2.1 in c:\\users\\hp\\appdata\\local\\programs\\python\\python311\\lib\\site-packages (from tensorflow-intel==2.14.0->tensorflow==2.14.0) (0.5.4)\n",
      "Requirement already satisfied: google-pasta>=0.1.1 in c:\\users\\hp\\appdata\\local\\programs\\python\\python311\\lib\\site-packages (from tensorflow-intel==2.14.0->tensorflow==2.14.0) (0.2.0)\n",
      "Requirement already satisfied: h5py>=2.9.0 in c:\\users\\hp\\appdata\\local\\programs\\python\\python311\\lib\\site-packages (from tensorflow-intel==2.14.0->tensorflow==2.14.0) (3.10.0)\n",
      "Requirement already satisfied: libclang>=13.0.0 in c:\\users\\hp\\appdata\\local\\programs\\python\\python311\\lib\\site-packages (from tensorflow-intel==2.14.0->tensorflow==2.14.0) (16.0.6)\n",
      "Requirement already satisfied: ml-dtypes==0.2.0 in c:\\users\\hp\\appdata\\local\\programs\\python\\python311\\lib\\site-packages (from tensorflow-intel==2.14.0->tensorflow==2.14.0) (0.2.0)\n",
      "Requirement already satisfied: numpy>=1.23.5 in c:\\users\\hp\\appdata\\local\\programs\\python\\python311\\lib\\site-packages (from tensorflow-intel==2.14.0->tensorflow==2.14.0) (1.24.3)\n",
      "Requirement already satisfied: opt-einsum>=2.3.2 in c:\\users\\hp\\appdata\\local\\programs\\python\\python311\\lib\\site-packages (from tensorflow-intel==2.14.0->tensorflow==2.14.0) (3.3.0)\n",
      "Requirement already satisfied: packaging in c:\\users\\hp\\appdata\\local\\programs\\python\\python311\\lib\\site-packages (from tensorflow-intel==2.14.0->tensorflow==2.14.0) (23.1)\n",
      "Requirement already satisfied: protobuf!=4.21.0,!=4.21.1,!=4.21.2,!=4.21.3,!=4.21.4,!=4.21.5,<5.0.0dev,>=3.20.3 in c:\\users\\hp\\appdata\\local\\programs\\python\\python311\\lib\\site-packages (from tensorflow-intel==2.14.0->tensorflow==2.14.0) (4.24.4)\n",
      "Requirement already satisfied: setuptools in c:\\users\\hp\\appdata\\local\\programs\\python\\python311\\lib\\site-packages (from tensorflow-intel==2.14.0->tensorflow==2.14.0) (68.0.0)\n",
      "Requirement already satisfied: six>=1.12.0 in c:\\users\\hp\\appdata\\local\\programs\\python\\python311\\lib\\site-packages (from tensorflow-intel==2.14.0->tensorflow==2.14.0) (1.16.0)\n",
      "Requirement already satisfied: termcolor>=1.1.0 in c:\\users\\hp\\appdata\\local\\programs\\python\\python311\\lib\\site-packages (from tensorflow-intel==2.14.0->tensorflow==2.14.0) (2.3.0)\n",
      "Requirement already satisfied: typing-extensions>=3.6.6 in c:\\users\\hp\\appdata\\local\\programs\\python\\python311\\lib\\site-packages (from tensorflow-intel==2.14.0->tensorflow==2.14.0) (4.8.0)\n",
      "Requirement already satisfied: wrapt<1.15,>=1.11.0 in c:\\users\\hp\\appdata\\local\\programs\\python\\python311\\lib\\site-packages (from tensorflow-intel==2.14.0->tensorflow==2.14.0) (1.14.1)\n",
      "Requirement already satisfied: tensorflow-io-gcs-filesystem>=0.23.1 in c:\\users\\hp\\appdata\\local\\programs\\python\\python311\\lib\\site-packages (from tensorflow-intel==2.14.0->tensorflow==2.14.0) (0.31.0)\n",
      "Requirement already satisfied: grpcio<2.0,>=1.24.3 in c:\\users\\hp\\appdata\\local\\programs\\python\\python311\\lib\\site-packages (from tensorflow-intel==2.14.0->tensorflow==2.14.0) (1.59.2)\n",
      "Requirement already satisfied: tensorboard<2.15,>=2.14 in c:\\users\\hp\\appdata\\local\\programs\\python\\python311\\lib\\site-packages (from tensorflow-intel==2.14.0->tensorflow==2.14.0) (2.14.1)\n",
      "Requirement already satisfied: tensorflow-estimator<2.15,>=2.14.0 in c:\\users\\hp\\appdata\\local\\programs\\python\\python311\\lib\\site-packages (from tensorflow-intel==2.14.0->tensorflow==2.14.0) (2.14.0)\n",
      "Requirement already satisfied: keras<2.15,>=2.14.0 in c:\\users\\hp\\appdata\\local\\programs\\python\\python311\\lib\\site-packages (from tensorflow-intel==2.14.0->tensorflow==2.14.0) (2.14.0)\n",
      "Requirement already satisfied: contourpy>=1.0.1 in c:\\users\\hp\\appdata\\local\\programs\\python\\python311\\lib\\site-packages (from matplotlib) (1.0.7)\n",
      "Requirement already satisfied: cycler>=0.10 in c:\\users\\hp\\appdata\\local\\programs\\python\\python311\\lib\\site-packages (from matplotlib) (0.11.0)\n",
      "Requirement already satisfied: fonttools>=4.22.0 in c:\\users\\hp\\appdata\\local\\programs\\python\\python311\\lib\\site-packages (from matplotlib) (4.39.4)\n",
      "Requirement already satisfied: kiwisolver>=1.0.1 in c:\\users\\hp\\appdata\\local\\programs\\python\\python311\\lib\\site-packages (from matplotlib) (1.4.4)\n",
      "Requirement already satisfied: pillow>=6.2.0 in c:\\users\\hp\\appdata\\local\\programs\\python\\python311\\lib\\site-packages (from matplotlib) (9.5.0)\n",
      "Requirement already satisfied: pyparsing>=2.3.1 in c:\\users\\hp\\appdata\\local\\programs\\python\\python311\\lib\\site-packages (from matplotlib) (3.0.9)\n",
      "Requirement already satisfied: python-dateutil>=2.7 in c:\\users\\hp\\appdata\\local\\programs\\python\\python311\\lib\\site-packages (from matplotlib) (2.8.2)\n",
      "Requirement already satisfied: wheel<1.0,>=0.23.0 in c:\\users\\hp\\appdata\\local\\programs\\python\\python311\\lib\\site-packages (from astunparse>=1.6.0->tensorflow-intel==2.14.0->tensorflow==2.14.0) (0.40.0)\n",
      "Requirement already satisfied: google-auth<3,>=1.6.3 in c:\\users\\hp\\appdata\\local\\programs\\python\\python311\\lib\\site-packages (from tensorboard<2.15,>=2.14->tensorflow-intel==2.14.0->tensorflow==2.14.0) (2.23.3)\n",
      "Requirement already satisfied: google-auth-oauthlib<1.1,>=0.5 in c:\\users\\hp\\appdata\\local\\programs\\python\\python311\\lib\\site-packages (from tensorboard<2.15,>=2.14->tensorflow-intel==2.14.0->tensorflow==2.14.0) (1.0.0)\n",
      "Requirement already satisfied: markdown>=2.6.8 in c:\\users\\hp\\appdata\\local\\programs\\python\\python311\\lib\\site-packages (from tensorboard<2.15,>=2.14->tensorflow-intel==2.14.0->tensorflow==2.14.0) (3.5)\n",
      "Requirement already satisfied: requests<3,>=2.21.0 in c:\\users\\hp\\appdata\\local\\programs\\python\\python311\\lib\\site-packages (from tensorboard<2.15,>=2.14->tensorflow-intel==2.14.0->tensorflow==2.14.0) (2.31.0)\n",
      "Requirement already satisfied: tensorboard-data-server<0.8.0,>=0.7.0 in c:\\users\\hp\\appdata\\local\\programs\\python\\python311\\lib\\site-packages (from tensorboard<2.15,>=2.14->tensorflow-intel==2.14.0->tensorflow==2.14.0) (0.7.2)\n",
      "Requirement already satisfied: werkzeug>=1.0.1 in c:\\users\\hp\\appdata\\local\\programs\\python\\python311\\lib\\site-packages (from tensorboard<2.15,>=2.14->tensorflow-intel==2.14.0->tensorflow==2.14.0) (3.0.1)\n",
      "Requirement already satisfied: cachetools<6.0,>=2.0.0 in c:\\users\\hp\\appdata\\local\\programs\\python\\python311\\lib\\site-packages (from google-auth<3,>=1.6.3->tensorboard<2.15,>=2.14->tensorflow-intel==2.14.0->tensorflow==2.14.0) (5.3.2)\n",
      "Requirement already satisfied: pyasn1-modules>=0.2.1 in c:\\users\\hp\\appdata\\local\\programs\\python\\python311\\lib\\site-packages (from google-auth<3,>=1.6.3->tensorboard<2.15,>=2.14->tensorflow-intel==2.14.0->tensorflow==2.14.0) (0.3.0)\n",
      "Requirement already satisfied: rsa<5,>=3.1.4 in c:\\users\\hp\\appdata\\local\\programs\\python\\python311\\lib\\site-packages (from google-auth<3,>=1.6.3->tensorboard<2.15,>=2.14->tensorflow-intel==2.14.0->tensorflow==2.14.0) (4.9)\n",
      "Requirement already satisfied: requests-oauthlib>=0.7.0 in c:\\users\\hp\\appdata\\local\\programs\\python\\python311\\lib\\site-packages (from google-auth-oauthlib<1.1,>=0.5->tensorboard<2.15,>=2.14->tensorflow-intel==2.14.0->tensorflow==2.14.0) (1.3.1)\n",
      "Requirement already satisfied: charset-normalizer<4,>=2 in c:\\users\\hp\\appdata\\local\\programs\\python\\python311\\lib\\site-packages (from requests<3,>=2.21.0->tensorboard<2.15,>=2.14->tensorflow-intel==2.14.0->tensorflow==2.14.0) (3.3.1)\n",
      "Requirement already satisfied: idna<4,>=2.5 in c:\\users\\hp\\appdata\\local\\programs\\python\\python311\\lib\\site-packages (from requests<3,>=2.21.0->tensorboard<2.15,>=2.14->tensorflow-intel==2.14.0->tensorflow==2.14.0) (3.4)\n",
      "Requirement already satisfied: urllib3<3,>=1.21.1 in c:\\users\\hp\\appdata\\local\\programs\\python\\python311\\lib\\site-packages (from requests<3,>=2.21.0->tensorboard<2.15,>=2.14->tensorflow-intel==2.14.0->tensorflow==2.14.0) (2.0.7)\n",
      "Requirement already satisfied: certifi>=2017.4.17 in c:\\users\\hp\\appdata\\local\\programs\\python\\python311\\lib\\site-packages (from requests<3,>=2.21.0->tensorboard<2.15,>=2.14->tensorflow-intel==2.14.0->tensorflow==2.14.0) (2023.7.22)\n",
      "Requirement already satisfied: MarkupSafe>=2.1.1 in c:\\users\\hp\\appdata\\local\\programs\\python\\python311\\lib\\site-packages (from werkzeug>=1.0.1->tensorboard<2.15,>=2.14->tensorflow-intel==2.14.0->tensorflow==2.14.0) (2.1.2)\n",
      "Requirement already satisfied: pyasn1<0.6.0,>=0.4.6 in c:\\users\\hp\\appdata\\local\\programs\\python\\python311\\lib\\site-packages (from pyasn1-modules>=0.2.1->google-auth<3,>=1.6.3->tensorboard<2.15,>=2.14->tensorflow-intel==2.14.0->tensorflow==2.14.0) (0.5.0)\n",
      "Requirement already satisfied: oauthlib>=3.0.0 in c:\\users\\hp\\appdata\\local\\programs\\python\\python311\\lib\\site-packages (from requests-oauthlib>=0.7.0->google-auth-oauthlib<1.1,>=0.5->tensorboard<2.15,>=2.14->tensorflow-intel==2.14.0->tensorflow==2.14.0) (3.2.2)\n"
     ]
    }
   ],
   "source": [
    "get_ipython().system('pip install tensorflow==2.14.0 opencv-python matplotlib')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "06eb3483",
   "metadata": {},
   "source": [
    "## 1.2 Import Dependencies"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "e3832bbc",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Import standard dependencies \n",
    "import cv2\n",
    "import os\n",
    "import random\n",
    "import numpy as np\n",
    "from matplotlib import pyplot as plt\n",
    "import tensorflow as tf"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "d0f66ad3",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Import tensorflow dependencies\n",
    "from tensorflow.keras.models import Model\n",
    "from tensorflow.keras.layers import (Layer, Conv2D, Dense, MaxPooling2D, Input, Flatten)\n",
    "#import tensorflow as tf"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "474d562f",
   "metadata": {},
   "source": [
    "## 1.3 Set GPU growth(optional if you have gpu else drop this step)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "933b93cc",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Avoid OOM errors by setting GPU Memory Xonsumption Growth\n",
    "gpus = tf.config.experimental.list_physical_devices('GPU')\n",
    "for gpu in gpus:\n",
    "    tf.config.experimental.set_memory_growth(gpu, True)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "72a14247",
   "metadata": {},
   "source": [
    "## 1.4 Create folder structure"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "96d2887c",
   "metadata": {},
   "outputs": [],
   "source": [
    "POS_PATH = os.path.join('data', 'positive')\n",
    "NEG_PATH = os.path.join('data', 'negative')\n",
    "ANC_PATH = os.path.join('data', 'anchor')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Make the directories\n",
    "os.makedirs(POS_PATH)\n",
    "os.makedirs(NEG_PATH)\n",
    "os.makedirs(ANC_PATH)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Collect Positive and Anchors"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 2.1 Untar Labelled Faces in the Wild Dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Move LFW Images to the following repository data/negative\n",
    "for directory in os.listdir('lfw'):\n",
    "    for file in os.listdir(os.path.join('lfw', directory)):\n",
    "        EX_PATH = os.path.join('lfw', directory, file)\n",
    "        NEW_PATH = os.path.join(NEG_PATH, file)\n",
    "        os.replace(EX_PATH, NEW_PATH)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 2.2 Collect Positive and Anchor Classes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "#import uuid library to generate unique image names\n",
    "import uuid"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "UUID('af2c1b9c-90de-11ee-bc28-c8940282ff56')"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "uuid.uuid1()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "data\\anchor\\e7548928-90df-11ee-8c89-c8940282ff56.jpg\n",
      "True\n",
      "data\\anchor\\ed4b681d-90df-11ee-a435-c8940282ff56.jpg\n",
      "True\n",
      "data\\anchor\\f01010c0-90df-11ee-ad04-c8940282ff56.jpg\n",
      "True\n"
     ]
    }
   ],
   "source": [
    "#Establish a connection to webcam\n",
    "cap = cv2.VideoCapture(0)\n",
    "while cap.isOpened():\n",
    "    ret,frame = cap.read()\n",
    "    #cut down frame to 250*250px\n",
    "    frame = frame[120:120+250,200:200+250,:]\n",
    "\n",
    "    #collect anchor\n",
    "    if cv2.waitKey(1) & 0XFF == ord('a'):\n",
    "        imgname = os.path.join(ANC_PATH,'{}.jpg'.format(uuid.uuid1()))\n",
    "        print(imgname)\n",
    "        #write out anchor image\n",
    "        print(cv2.imwrite(imgname,frame))\n",
    "        \n",
    "    #collect positive\n",
    "    if cv2.waitKey(1) & 0XFF == ord('p'):\n",
    "        imgname = os.path.join(POS_PATH,'{}.jpg'.format(uuid.uuid1()))\n",
    "        #write out anchor image\n",
    "        cv2.imwrite(imgname,frame)\n",
    "    #show image back to screen\n",
    "    cv2.imshow('Image Collection',frame)\n",
    "    #breaking gracefully\n",
    "    if cv2.waitKey(1) & 0XFF == ord('q'):\n",
    "        break\n",
    "#Release webcam\n",
    "cap.release()\n",
    "#close the image show frame\n",
    "cv2.destroyAllWindows()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "ename": "NameError",
     "evalue": "name 'frame' is not defined",
     "output_type": "error",
     "traceback": [
      "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[1;31mNameError\u001b[0m                                 Traceback (most recent call last)",
      "\u001b[1;32me:\\E-portal\\E-portal\\FaceId\\Face Recognition.ipynb Cell 19\u001b[0m line \u001b[0;36m1\n\u001b[1;32m----> <a href='vscode-notebook-cell:/e%3A/E-portal/E-portal/FaceId/Face%20Recognition.ipynb#X24sZmlsZQ%3D%3D?line=0'>1</a>\u001b[0m plt\u001b[39m.\u001b[39mimshow(frame[\u001b[39m120\u001b[39m:\u001b[39m120\u001b[39m\u001b[39m+\u001b[39m\u001b[39m250\u001b[39m,\u001b[39m200\u001b[39m:\u001b[39m200\u001b[39m\u001b[39m+\u001b[39m\u001b[39m250\u001b[39m,:])\n",
      "\u001b[1;31mNameError\u001b[0m: name 'frame' is not defined"
     ]
    }
   ],
   "source": [
    "plt.imshow(frame[120:120+250,200:200+250,:])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 3 Load and Preprocess Images"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 3.1 Get Images Directories"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "anchor = tf.data.Dataset.list_files(ANC_PATH+'\\*.jpg').take(300)\n",
    "positive = tf.data.Dataset.list_files(POS_PATH+'\\*.jpg').take(300)\n",
    "negative = tf.data.Dataset.list_files(NEG_PATH+'\\*.jpg').take(300)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "dir_test = anchor.as_numpy_iterator()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "b'data\\\\anchor\\\\bf1f94c2-8f92-11ee-afe8-c8940282ff56.jpg'\n"
     ]
    }
   ],
   "source": [
    "print(dir_test.next())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 3.2 Preprocessing- Scale and Resize"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "def preprocess(file_path):\n",
    "    \n",
    "    # Read in image from file path \n",
    "    byte_img = tf.io.read_file(file_path)\n",
    "    # Load in the image\n",
    "    img = tf.io.decode_jpeg(byte_img)\n",
    "    \n",
    "    # Preprocessing steps - resizing the image to be 100x100x3\n",
    "    img = tf.image.resize(img, (100,100))\n",
    "    # Scale image to be between 0 and 1\n",
    "    img = img / 255.0\n",
    "    \n",
    "    # Return image\n",
    "    return img"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "#return numpy equivalent of our image\n",
    "img = preprocess(b'data\\\\anchor\\\\c0e545cf-8f92-11ee-9e5b-c8940282ff56.jpg')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.12622549"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#show minimum value in img \n",
    "img.numpy().min()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.99852943"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#show max value in img\n",
    "img.numpy().max()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x1f8805f3550>"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#show image at that path\n",
    "plt.imshow(img)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 3.3 Create Labelled Dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "#(anchor, positive) => 1,1,1,1,1\n",
    "# (anchor, negative) => 0,0,0,0,0"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [],
   "source": [
    "#positve veification\n",
    "#tf.ones(len(anchor))\n",
    "#tf.zeros(len(anchor))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "positives = tf.data.Dataset.zip((anchor, positive, tf.data.Dataset.from_tensor_slices(tf.ones(len(anchor)))))\n",
    "negatives = tf.data.Dataset.zip((anchor, negative, tf.data.Dataset.from_tensor_slices(tf.zeros(len(anchor)))))\n",
    "data = positives.concatenate(negatives)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "samples = data.as_numpy_iterator()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "#it means it is positive or negative result\n",
    "example = samples.next()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(b'data\\\\anchor\\\\ccc16513-8f92-11ee-8ded-c8940282ff56.jpg',\n",
       " b'data\\\\positive\\\\396099bf-8f93-11ee-b3b5-c8940282ff56.jpg',\n",
       " 1.0)"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "example"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "\n",
    "## 3.4 Bulid Train and Test partition"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "def preprocess_twin(input_img,validation_img,label):\n",
    "    return(preprocess(input_img),preprocess(validation_img),label) "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "res = preprocess_twin(*example)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(<tf.Tensor: shape=(100, 100, 3), dtype=float32, numpy=\n",
       " array([[[0.972549  , 0.9647059 , 0.9764706 ],\n",
       "         [0.972549  , 0.9647059 , 0.9764706 ],\n",
       "         [0.972549  , 0.9647059 , 0.9764706 ],\n",
       "         ...,\n",
       "         [0.972549  , 0.9647059 , 0.9764706 ],\n",
       "         [0.972549  , 0.9647059 , 0.9764706 ],\n",
       "         [0.972549  , 0.9647059 , 0.9764706 ]],\n",
       " \n",
       "        [[0.972549  , 0.9647059 , 0.9764706 ],\n",
       "         [0.972549  , 0.9647059 , 0.9764706 ],\n",
       "         [0.972549  , 0.9647059 , 0.9764706 ],\n",
       "         ...,\n",
       "         [0.972549  , 0.9647059 , 0.9764706 ],\n",
       "         [0.972549  , 0.9647059 , 0.9764706 ],\n",
       "         [0.972549  , 0.9647059 , 0.9764706 ]],\n",
       " \n",
       "        [[0.972549  , 0.9647059 , 0.9764706 ],\n",
       "         [0.972549  , 0.9647059 , 0.9764706 ],\n",
       "         [0.972549  , 0.9647059 , 0.9764706 ],\n",
       "         ...,\n",
       "         [0.972549  , 0.9647059 , 0.9764706 ],\n",
       "         [0.972549  , 0.9647059 , 0.9764706 ],\n",
       "         [0.972549  , 0.9647059 , 0.9764706 ]],\n",
       " \n",
       "        ...,\n",
       " \n",
       "        [[0.8384804 , 0.8267157 , 0.76004905],\n",
       "         [0.8370098 , 0.8252451 , 0.75808823],\n",
       "         [0.84387255, 0.8306373 , 0.75759804],\n",
       "         ...,\n",
       "         [0.76740193, 0.7870098 , 0.76053923],\n",
       "         [0.7710784 , 0.79068625, 0.7642157 ],\n",
       "         [0.77156866, 0.79901963, 0.76862746]],\n",
       " \n",
       "        [[0.82843137, 0.81666666, 0.75      ],\n",
       "         [0.8352941 , 0.8235294 , 0.75490195],\n",
       "         [0.8362745 , 0.8186275 , 0.7441176 ],\n",
       "         ...,\n",
       "         [0.7642157 , 0.7897059 , 0.7632353 ],\n",
       "         [0.7678922 , 0.79338235, 0.76691175],\n",
       "         [0.7745098 , 0.80196077, 0.7745098 ]],\n",
       " \n",
       "        [[0.8267157 , 0.814951  , 0.74828434],\n",
       "         [0.82941175, 0.8156863 , 0.7470588 ],\n",
       "         [0.84117645, 0.8215686 , 0.7470588 ],\n",
       "         ...,\n",
       "         [0.75686276, 0.78431374, 0.75686276],\n",
       "         [0.7745098 , 0.80196077, 0.7745098 ],\n",
       "         [0.76960784, 0.7970588 , 0.76960784]]], dtype=float32)>,\n",
       " <tf.Tensor: shape=(100, 100, 3), dtype=float32, numpy=\n",
       " array([[[0.972549  , 0.9647059 , 0.9764706 ],\n",
       "         [0.972549  , 0.9647059 , 0.9764706 ],\n",
       "         [0.972549  , 0.9647059 , 0.9764706 ],\n",
       "         ...,\n",
       "         [0.96862745, 0.96862745, 0.9764706 ],\n",
       "         [0.96568626, 0.96862745, 0.9764706 ],\n",
       "         [0.9610294 , 0.964951  , 0.9727941 ]],\n",
       " \n",
       "        [[0.972549  , 0.9647059 , 0.9764706 ],\n",
       "         [0.972549  , 0.9647059 , 0.9764706 ],\n",
       "         [0.972549  , 0.9647059 , 0.9764706 ],\n",
       "         ...,\n",
       "         [0.96862745, 0.96862745, 0.9764706 ],\n",
       "         [0.96568626, 0.96862745, 0.9764706 ],\n",
       "         [0.9639706 , 0.96789217, 0.9757353 ]],\n",
       " \n",
       "        [[0.972549  , 0.9647059 , 0.9764706 ],\n",
       "         [0.972549  , 0.9647059 , 0.9764706 ],\n",
       "         [0.972549  , 0.9647059 , 0.9764706 ],\n",
       "         ...,\n",
       "         [0.96862745, 0.96862745, 0.9764706 ],\n",
       "         [0.96568626, 0.96862745, 0.9764706 ],\n",
       "         [0.9607843 , 0.9647059 , 0.972549  ]],\n",
       " \n",
       "        ...,\n",
       " \n",
       "        [[0.8281863 , 0.81642157, 0.75171566],\n",
       "         [0.83431375, 0.82843137, 0.7607843 ],\n",
       "         [0.82941175, 0.8235294 , 0.7558824 ],\n",
       "         ...,\n",
       "         [0.7382353 , 0.7656863 , 0.7352941 ],\n",
       "         [0.74313724, 0.7735294 , 0.74509805],\n",
       "         [0.7490196 , 0.7705882 , 0.74607843]],\n",
       " \n",
       "        [[0.8375    , 0.82279414, 0.7727941 ],\n",
       "         [0.8144608 , 0.7997549 , 0.7497549 ],\n",
       "         [0.8247549 , 0.8122549 , 0.7556372 ],\n",
       "         ...,\n",
       "         [0.72745097, 0.76666665, 0.7323529 ],\n",
       "         [0.7357843 , 0.77794117, 0.7465686 ],\n",
       "         [0.7343137 , 0.7705882 , 0.74019605]],\n",
       " \n",
       "        [[0.83406866, 0.8183824 , 0.7752451 ],\n",
       "         [0.81789213, 0.80220586, 0.75808823],\n",
       "         [0.81764704, 0.80196077, 0.75490195],\n",
       "         ...,\n",
       "         [0.73063725, 0.77230394, 0.7372549 ],\n",
       "         [0.7323529 , 0.78333336, 0.7480392 ],\n",
       "         [0.73308825, 0.77230394, 0.7409314 ]]], dtype=float32)>,\n",
       " 1.0)"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "res"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "3"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(res)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<tf.Tensor: shape=(100, 100, 3), dtype=float32, numpy=\n",
       "array([[[0.972549  , 0.9647059 , 0.9764706 ],\n",
       "        [0.972549  , 0.9647059 , 0.9764706 ],\n",
       "        [0.972549  , 0.9647059 , 0.9764706 ],\n",
       "        ...,\n",
       "        [0.972549  , 0.9647059 , 0.9764706 ],\n",
       "        [0.972549  , 0.9647059 , 0.9764706 ],\n",
       "        [0.972549  , 0.9647059 , 0.9764706 ]],\n",
       "\n",
       "       [[0.972549  , 0.9647059 , 0.9764706 ],\n",
       "        [0.972549  , 0.9647059 , 0.9764706 ],\n",
       "        [0.972549  , 0.9647059 , 0.9764706 ],\n",
       "        ...,\n",
       "        [0.972549  , 0.9647059 , 0.9764706 ],\n",
       "        [0.972549  , 0.9647059 , 0.9764706 ],\n",
       "        [0.972549  , 0.9647059 , 0.9764706 ]],\n",
       "\n",
       "       [[0.972549  , 0.9647059 , 0.9764706 ],\n",
       "        [0.972549  , 0.9647059 , 0.9764706 ],\n",
       "        [0.972549  , 0.9647059 , 0.9764706 ],\n",
       "        ...,\n",
       "        [0.972549  , 0.9647059 , 0.9764706 ],\n",
       "        [0.972549  , 0.9647059 , 0.9764706 ],\n",
       "        [0.972549  , 0.9647059 , 0.9764706 ]],\n",
       "\n",
       "       ...,\n",
       "\n",
       "       [[0.8384804 , 0.8267157 , 0.76004905],\n",
       "        [0.8370098 , 0.8252451 , 0.75808823],\n",
       "        [0.84387255, 0.8306373 , 0.75759804],\n",
       "        ...,\n",
       "        [0.76740193, 0.7870098 , 0.76053923],\n",
       "        [0.7710784 , 0.79068625, 0.7642157 ],\n",
       "        [0.77156866, 0.79901963, 0.76862746]],\n",
       "\n",
       "       [[0.82843137, 0.81666666, 0.75      ],\n",
       "        [0.8352941 , 0.8235294 , 0.75490195],\n",
       "        [0.8362745 , 0.8186275 , 0.7441176 ],\n",
       "        ...,\n",
       "        [0.7642157 , 0.7897059 , 0.7632353 ],\n",
       "        [0.7678922 , 0.79338235, 0.76691175],\n",
       "        [0.7745098 , 0.80196077, 0.7745098 ]],\n",
       "\n",
       "       [[0.8267157 , 0.814951  , 0.74828434],\n",
       "        [0.82941175, 0.8156863 , 0.7470588 ],\n",
       "        [0.84117645, 0.8215686 , 0.7470588 ],\n",
       "        ...,\n",
       "        [0.75686276, 0.78431374, 0.75686276],\n",
       "        [0.7745098 , 0.80196077, 0.7745098 ],\n",
       "        [0.76960784, 0.7970588 , 0.76960784]]], dtype=float32)>"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "res[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x1f88059f750>"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.imshow(res[1])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1.0"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "res[2]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Build dataloader pipeline\n",
    "data = data.map(preprocess_twin)\n",
    "data = data.cache()\n",
    "data = data.shuffle(buffer_size=1024)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<_ShuffleDataset element_spec=(TensorSpec(shape=(100, 100, None), dtype=tf.float32, name=None), TensorSpec(shape=(100, 100, None), dtype=tf.float32, name=None), TensorSpec(shape=(), dtype=tf.float32, name=None))>"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [],
   "source": [
    "samples = data.as_numpy_iterator()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "3"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(samples.next())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [],
   "source": [
    "samp = samples.next()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x1f881751590>"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAaAAAAGgCAYAAADsNrNZAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/bCgiHAAAACXBIWXMAAA9hAAAPYQGoP6dpAABunklEQVR4nO29e7AdZZX+v/q2L+eSE5KQkwQSiMgMV0fkZsTfXDTfoRycgZFyxl/hDF5qHDUolyoVHGFqcDDoVCmDhThailgjMlLf8V5i+YvKDMo1CopoQIkSgZwAITnJuezdl/f3R3T3Ws+b3eccEugT8nyqUrX7dO/ut99+e3d6Pet9VuCcc0IIIYQ8z4R1N4AQQsjBCR9AhBBCaoEPIEIIIbXABxAhhJBa4AOIEEJILfABRAghpBb4ACKEEFILfAARQgipBT6ACCGE1AIfQIQQQmrhOXsAXXfddXLkkUdKq9WS008/Xe6+++7n6lCEEEIOQILnwgvuv/7rv+Tv//7v5ZOf/KScfvrpcs0118gtt9wimzZtkqVLl1Z+tygKefzxx2V4eFiCINjfTSOEEPIc45yTXbt2yYoVKyQMK95z3HPAaaed5tatW9dbzvPcrVixwq1fv37G727ZssWJCP/xH//xH/8d4P+2bNlS+Xsfy36m2+3Kxo0b5bLLLuv9LQxDWbt2rdxxxx3e9p1ORzqdTm/Z/e6F7FcPPyLDw8P7u3mEEEKeY3bt2iVHHf2iGX/D9/sD6KmnnpI8z2V0dNT8fXR0VH7xi194269fv17+5V/+xfv78PCwLFiwYH83jxBCyPPETDJK7Vlwl112mezcubP3b8uWLXU3iRBCyPPAfn8DWrJkiURRJGNjY+bvY2NjsmzZMm/7ZrMpzWZzfzeDEELIPGe/vwE1Gg05+eSTZcOGDb2/FUUhGzZskDVr1uzvwxFCCDlA2e9vQCIil1xyiZx//vlyyimnyGmnnSbXXHONTExMyJvf/Obn4nCEEEIOQJ6TB9Df/u3fypNPPilXXHGFbN26VV760pfKrbfe6iUmEEIIOXh5Tiai7gvj4+MyMjIi27Y+ySw4Qgg5ABkfH5elyw6VnTt3Vv6O154FRwgh5OCEDyBCCCG1wAcQIYSQWuADiBBCSC3wAUQIIaQW+AAihBBSC3wAEUIIqQU+gAghhNQCH0CEEEJqgQ8gQgghtcAHECGEkFrgA4gQQkgt8AFECCGkFvgAIoQQUgt8ABFCCKkFPoAIIYTUAh9AhBBCaoEPIEIIIbXABxAhhJBa4AOIEEJILfABRAghpBb4ACKEEFILfAARQgipBT6ACCGE1AIfQIQQQmqBDyBCCCG1wAcQIYSQWuADiBBCSC3wAUQIIaQW+AAihBBSC3wAEUIIqQU+gAghhNQCH0CEEEJqgQ8gQgghtRDX3QBCyP7DwXJQuUXQd81zid+m2VPVxn3ZL6kHvgERQgipBT6ACCGE1AIfQIQQQmqBGhB5weP2lzjg7aiwi0F/hSJAhQL2FVR8V4L+J5BmmVke3zVhlvMiN8u7J8bLbXeOm3UZtDEZaKkl+3/VZqNhlttJs/e51WqadY3E/syknSmzHIblvuPAHidKEnucVtmmOLZtCKCfCnXuQQHXKrT9Dd0EghL8P93Bvvp9bW/fPcDwhjwM0329tQ7s3iGEEHLAwgcQIYSQWuADiBBCSC1QAyIvCOY0h+VZTngJxQoFuBtXMQnHwco8R02i/3GLwn73pw/8rPf5x/fdb9ZluW1jFzSiiU6n97ndbpl1wwsXmOWoWa7ftXuX3c9OqzVNje/sfY5jezLtdtssL12yxB53cLD3eWBgwKxrNayeNDFRHnf3xG6z7vjjjzfLq488Uu3Hakn+hCnb5ipJLnR2W62TONjxAT83aYZ7pd/q2d5ifAMihBBSC3wAEUIIqQWG4MgLD4h7VIVT5kL/5Ns9+One5YHTzH57829+bZbHtj3V+zw1PW3WdbupWU6SMv24PWTDZpt/bff7619vMcu5CgVmWdcep7DLsQp/DQ2NmHVDTRsqy/IytIcp54HsNMtjT2yzbSrKMGFnumPWJVFklhsq/XtQhe5ERH71yK9h2zLsduopLzPrXvYyu9xq2nCkPmo00/hR611FyvyByExnY7om6PO5Ar4BEUIIqQU+gAghhNQCH0CEEEJqgRoQOahB+xaPqPw/WhDC/9cKsH6BVOsHHvx57/Ndd99t1u2emjTLu3aX6cXTk1YHabWs3qJTlXftsunRuyA1OWna9OOp8XJ94WzKdp7b9qdZqUXlXathpQ2rU7Wa5U9JBLpNq2X1lSK1qeG5Sh0vMtumFNrU7ZY6lU7JFhHJIOW8UPY7v/71b8y6H228zyy/6v/8mVn+o+NO6H2OInudu6BTGaskz7rmhfV/fE8O20fJ64XVO4QQQg4Y+AAihBBSC3wAEUIIqQVqQOSgpoCodg62/bt3lyULNt53n1mHusLEuNUk2kqrme5Y3WBywi5r+aIJmg9qM7/5zW97nzsdq8VgWYdpWB+q2Uydrp330wENRZdnyFKw+Jm0JRWaSitrwpyaTmy3jcCqJzfzkWz/d0ATCsL+egtqcKGy19kN1+anP33QLD+y2V7LN/y/f9P7/PLTTzXrXAgHVsfFOWcvrFlB+x++ARFCCKkFPoAIIYTUAh9AhBBCaoEaECGKO++80yz/99e/2vscJrY0wKKRRWY5DuztFIelnhGKLR892LD/92skSpvpWG3GwXydSP2/sTNlNZ6JCTsvCEtItJSPXFPsfJ0ESlzvnlLajbP6EM4ISdWpZ6n1ruvEdi5SHNt+CpVwEsF/iSOYezWtfPIK0OtC2FavDvBc4Vo+9aT1q7v55v/ufR7fbedWvfKMNWa5rc8H5obxv/jVsHsIIYTUAh9AhBBCaoEhOPK84ZUrmFPKanUxhDgqh3IA1S1zZ7+r03Xv3vgjs+7W/28DNLEM3UyN23TiTmjDXyOHLjPLgyqdOof279xt07B3PrOj93lszJYrQDuaVFnBdLrQpoq0axGRSIWIssCGxjC9OHDlucdgWZTDcVxero/gV6WbWtuhFPaVNMovtJs2NBZCTC5S1xZDkw6qwcZB2f4A/6sNFWkHGzZ1fGJX2eZvfft7Zl0B4bxX/fEZvc/DAzaM2YX0+8Dp9tvrGsEdUAT9x7xffPd5epfADHT1Wd9XuVceeO/wDYgQQkgt8AFECCGkFub0AFq/fr2ceuqpMjw8LEuXLpVzzjlHNm3aZLaZnp6WdevWyeLFi2VoaEjOPfdcGRsb26+NJoQQcuAzJw3otttuk3Xr1smpp54qWZbJ+9//fvnzP/9zefDBB3vlcS+++GL55je/KbfccouMjIzIBRdcIK973evkBz/4wXNyAoSIWEt/1IuywmoDD/zsZ73Pt37rW2bd7l025VaXYFix/DCzbsmixWZ5cGDILDeVnrF9xzNm3ZNPPmmWt6mS3Dt3jpt1aRe0Dm39AucagNBWQCnw6aLUJHKBUt+ttlmOlQaRda2WEWIZCx3/B/scB4oFVrWYnir33Z22KehYCkGX5M7gOAmIT0FYajUZlIDANqAFk1MlJZ7a9rRZ9z//c7tZjpV29v+88nTbJtTVVFp8FFgtaaaS73Y/8w899Pzy9HtnTg+gW2+91Sx/7nOfk6VLl8rGjRvlj//4j2Xnzp3ymc98Rm666SZ51ateJSIiN9xwgxx77LFy5513ystf/nJvn51ORzpKqBsfH/e2IYQQ8sJjnzSgnTv3TN5atGjPhLyNGzdKmqaydu3a3jbHHHOMrFq1Su6444697mP9+vUyMjLS+7dy5cp9aRIhhJADhGf9ACqKQi666CI544wz5IQT9lQP3Lp1qzQaDVm4cKHZdnR0VLZu3brX/Vx22WWyc+fO3r8tW7Y82yYRQgg5gHjW84DWrVsnDzzwgNx+++0zb1xBs9k0sXLywiWA+Tg4U8AsexM3Zvi/kvry1LSdd3LPPfea5f/93//tfc67VhsYXXKoWdbzGeKGvV2mYC7MY1ufsOsnyvXPPGM1oCy3+suU1kFg7kiWwlweXfIa9C0pcFuY66OuQYTlDOB8NAHM6yjg4un9YplznIeFJbvtfqEkN2g3uhy21oNERFJo1LTa1kFp7MLZPo7BLkjrRwFYBz0IpRwGWuXv1/CCYbPuhOOONsutpmoznGvoUC+ah/gTkPaJZ/UGdMEFF8g3vvEN+d73vieHH3547+/Lli2TbrcrO3bsMNuPjY3JsmXLhBBCCPk9c3oAOefkggsukC9/+cvy3e9+V1avXm3Wn3zyyZIkiWzYUM4m37Rpkzz66KOyZs0a3B0hhJCDmDmF4NatWyc33XSTfPWrX5Xh4eGerjMyMiLtdltGRkbkrW99q1xyySWyaNEiWbBggbzrXe+SNWvW7DUDjhxcYGjGW69yY8PQhmnQtmT3Lhtm+8lP7+99/vGPrb1Op2PDXQMqXbqRoMtzANuWFi1Juw3bhrCtrWT6jLLXmZqyljlTE7ZCZzfDdpRE0G8uK8/H5dXJuwXY0+h+jWO4HrAvfT0c/l8V8pjTrk7vrg65FTnG7/T29rtxYo+jHa87EKpEN2ynQlphCA7cEJrMIZSZqH4KoblNSPd+4IGf9z4vXrrUrBseGTTLL1p9RO9zG9oQQPjUi3POC98A2yZzWz6LcNycHkDXX3+9iIj86Z/+qfn7DTfcIG9605tERORjH/uYhGEo5557rnQ6HTnzzDPlE5/4xNxbRggh5AXNnB5A+L/QvdFqteS6666T66677lk3ihBCyAuf+fBORwgh5CCE5RjI84anAWFqr4p571D6iYjI1772NbPs+Qsqy5M8tfH83VDRUr/JY1XNCOz/dTZ40mrBOiiTAJVAC6Ur4DqM70dB/+hCjjqP7jeMuxf9+xS/6lcUxbxslbINOg5uGzRKPUNXLRURqMlq+0VEJFRaVAj9kGXYRn19UDOxizqVunDYD1CiA/qioU7XZdWp7i4qlzdu/LFZNzJidcHRpaO9z80Rm0buvQ3A5agYIs8b+7sJfAMihBBSC3wAEUIIqQU+gAghhNQCNSBSG2izcufdd/c+f/vW75h17baNpaPde0fZ+ON+Y5yHojQI1EhQsdj+9Pbe56dBl5rq2NIBnY497u/Nevess3NWCpj34yQzS7a9toWBEge8UsygzeAcIq11oLYUJVDOQIkOaMUTg+CSqj6NYO5OnqP+YvtYl5DIC5gP5clSWi+CcwNdJ9d9HNhzK7A8gz2M6eMEbHo87UyVy+iM2/lp3/n2d83y4SvKkh6nnPRSs26gCfOCoE1OlccI3PPz7uCVVdjPIhDfgAghhNQCH0CEEEJqgSE4IjO/V8/FY6P/vgJwuH7gwZ+b5W98vaxOiu7EuydsKnUOobM4KLcP4f9VuuKmiHWiRvflAkJAhTqfZ57ZadZ1Id27m/ZPl87AdTvA9GidYxtgCM7LL1abQjgLvuuFlkL92W6LrtU67ObQqRlDQIVej2E/CDfCcaZVKDNp2Ovu29Go/UATHPSTHW9g8YNWT+hopA6b55hI3v+4DkJ7E7kN0/7P98vK0Mf94TFmXSux5x7CvQS9Cs3t/y4xU3XSqvRuTF/31tfhhk0IIYTsK3wAEUIIqQU+gAghhNQCNSCyF/al7GFFzBhShL9/2/+a5SAsrUlCSJ3GyqWdSVveYHdWakQZpDhXxfAdtDeF7+6eLMsmpN5+oMqm59hSxunDAEsFgPakYu1YraCAQLvWeUJIGw8E9Ba013F7/bjnu145g/Jzjm2Ac4/VGMlzTKWG5aj/+MI2oD6h9ZYcbHq89G5lkYO2T6gXRZG1xclNqj7oR1AhVdsD5Tj2oP2//OXm3ueHf7XZrDvlZX9kN4Y089CcA1yrilvU02mgTXqIeP2Nw2eG1Pfe32eZr803IEIIIbXABxAhhJBa4AOIEEJILVADIvuZ/v+nwRIKTzz2mFnOsjJu3Jm2seVOx1r8P7N9u1k2mlEApQMgbp2l2o4G5sKAPY0ryvMpcjufKIVtczh3fdx4JusUrc2A1oSlELpqromDeSaNBhwHRQh1oBBKSweg9Tm1jMUoUWcrtHYT4PwV2K83X6dcn0HZCs9up+ivg+A8M60RFYL91DTLAvqRljYi2Bbnxuhy6g7akMJ8nabSx7532/fNumOP+wOzPNiGNqq5SziPDCUXrftg/1fhaT643ivJreyasNzKLOAbECGEkFrgA4gQQkgt8AFECCGkFqgBEdmL8Tss7z3Xfw+z/z/M//2/XzbLGZQ61qWcu10bs+92rf6SY1lkpfu0WtX2+UGoSk3jfAoMrSsxIAQ9JYKAeQbHydUEC+zhyPPYKrWBJLZt0t51IiJJVO6tQN0D9COcB6Q1FewX1HmiSPUj6DppBiXG1WET8DTDa5lA2Qf09dNkoM2EYblvh9ONQrtfJ2UbsdQ69pPXx4maF5SBhgUaY6R9CwOcs4W6VNnozZt/bdY9stnOC/rDP3yxbZPSgGJvXpn0Ba8rajVV3/Xm88Dis9F9NHwDIoQQUgt8ABFCCKkFhuDIXsD/l/QPweErOKYx6zDb9u3PVB5Vp/ainQuGixoNa50SxGXKahzbdeDuIlNFWbXSQWgph9CSU6UEsLpoBjYkYYAW+XqdbQOWTdC7ziEcFOGX1fUI8JheuAW+av6ApRswNFNeO0y7xvY3VcgK29CIbcgKj9tslt8tYPxMYyVZFQY1IUKxlW73tFEdEWySYgjJOWiiTvcO4XzQfsap4yYN+5ParLAWSjv2Ot9z9z1m+agjjjTLjcFy3wFW+c3Rkmn26EhyVWkGkepwXr/PVfANiBBCSC3wAUQIIaQW+AAihBBSC9SAyD5SrUH87GcP9j7/9rePm3UdL76vShvDfprtllnOwdomVeUatu+wWlMEKas6Xo72+ZiSmpttbTqxg0g7akRiSlqDpoXBdl1mAPQvL4Vbhf9jPCT8l7IBmoT2aAkFLYvstlpTKWC/YdQ/vRt1wBxLSwcouJTHiUHbE68cefldLO0dodSkeg6qj3vlPtBCx6m+QTsgvO6mJLenD4FuqDTGbtc26qGHHzbLndSOt5aU90CAumDVTInZVUbYsylqlYJp5f21HWpAhBBCDhj4ACKEEFILfAARQgipBWpAZL/yq1/9yix/9rOf7X3GcssdsGjRsfYmWtFDLHrX+IRZnpwsbXwC0A0y0BxMXBsmgKRd+K7SMwIQEvx5J/3j3p7NPcTsY6VJRHAcnAek59XgHCdtMyQi0kRNJdM2RFBa2uuncp4NWtVgX2iXohw1EhRgUFNRFjMgH0kDbH10P+XQ/1GCZcOVBgfzgByKZTHOqyk/g2OU2e/v//J7UtA1M9CaGspKCMt5bN+xwyxvecJqpsMLh3ufUVdD9NqZ5Bizp5lcuar2YzRPluQmhBAyj+EDiBBCSC0wBEe8KodeRctKt1z7zn7jF26yGyin4A46NUN4RacmT4D7dda1qcmd1C4nKtKB5soBpBu7olzOIIxTQKwmMKGE6v+vYRq2TqfGcB1aCSXqMENNm3Iee3E25dANIbcoRisY26a4UZ57EtswJ/apcZCGUKWDfspUSKsIcFux24IFTazcv3VVXBERF2MacxlqbbbAjgkqvNrKuP0rw4qIpDCOs7RsUzeAVH2s0qpOMIe0a0zLLpS1E0YmU4j1bXv6abP8hypVHNPGiwDihAovNAw3iLFg8m+e5xS+ARFCCKkFPoAIIYTUAh9AhBBCaoEaEJkzOkq8bduTZh3a56dK+5iamjTr0LJFh7Wnp6xOkHftcggx8DQt18eoxWTQJpWKjNU4sbyBtpwpHGgXYOmPoXa9NoJ03KaVv2RA6WEN2G8TKn3qXUVQPRW1mWZi18dGR7BtaA+ApqKtVRw0GOgoS6MUK62C3hK3wfJHaR9YQTQrsOypKrvRtBpWDhVrk0RraVjR1e43B50wi9T2IdhAge/NhKrk66BTd07ZqQY69R2rjeL1+NWmh8zyCcf8Qe/z4cuXm3XoxFOVEu0X91DbwjrUfyunGjANmxBCyIECH0CEEEJqgQ8gQgghtUANiPi4ykUjdtx3/0/MqgjmYjjl498FHUfrNiIieVpGsgsoSRAGGC+362MVs89hvzgPSJ9RCPNk8gJLT+uFaisenK+TqH23YM5TBPtqqfYPt+22Cc7lUboPRuST2H43wHIZ5gswLwu2DfX5eCID6F9KmIrR9gZGEDjmSNBQ5wP/JY5Cq/PoMgQZlLgIYF5ToP5/jZphkcN1h59Cp+Y94alrDVFEZLBZ7ntqGnTCZMgsT+m5VliWAjSsJ8e2meWnx0q9demiQ8y6AgSkQOmIWM4edRxznT0dc/YTgViOgRBCyAEDH0CEEEJqgSE4MuNLtufkrL7xS3C/Fgh36RTVbgcqZWKGrTpQjOUtCxvaELHLWariPhAecln/mGIh1eEJbaWCDtZeVU1vvTofCAE1E/vdJNAWLTZtPIZQTaL62OsnTBWP+9sdhRGmVoO9jrI/CsCKB52odfZ6AqnhugqoiAgW80x0myANPsaN1a5b4GCN4TsTekJLnNC2P8RfQjNG7HG6EYbvytBfA/o77mLYtuzTae8GsCcwAJZMQ0NlOC+DcDZa80RhRdp8hZUQVn/F+2GmcN5c4RsQIYSQWuADiBBCSC3wAUQIIaQWqAERmbkMIv4/pdy+m9pYOlr6a20mAKv6wBOXym0jaEMns5YmeWF1ksCVQznC8wFrG6fahBY5OdrIqGYEYbXmE4BeoVOiE9Ar0G4nUd78A1BmoJ1gmYH+NidoBQNNljgp912k0IegI4QqpR5LBaANkU4N74C20WzYRkReOYBy3wmmsseoSajPcAEwtT1V4ykHi6IWeCGlYNek0/5R98CpBrm67kliU8FbbXs+DXXcbTt2mXUBaD6LDl1iltuDKqUbLizewTq12tc1+4+fmbbF9Xp5tvY7pp1z/gYhhBCyH+ADiBBCSC3wAUQIIaQWqAGRvXjtYAloXF/+ATUT1ArMVAyYZxJCGWFdCgEtcbBNTtCiRR/TnlAE2oxTx+l0pmFbG1s381C8+LdtIepJWiZpNUALAG2m3Sjj/6gXoeajm1TAuWGZ5wJ0klDNpwpBb8GS4k6tj6HUN84D0nqRS7EcgyUB/SjSljkwRgIYfEGuNUXw/AGLnKaa59SFVhS5nUfTgDlRQaD0L5z7ghZGqtMbcD0GGlBevVnud3x6yqybhnF75FEvhu+WY75AraxiPk4GpSfwuuu5Pzh+ZuLZ6D7m2Pv0bUIIIeRZwgcQIYSQWuADiBBCSC1QAyKexuNAN8D5Lk7FjFOY95OgVqDmmoQwFyOA+HGh7PXTFEoZw36bIZZjLrfH6UXT4EWW69LBoLdkDvUjtT4H3QMC5jhdJ47KjkvA+22gYTWHtprvAlW0Jaz4b2KEmg/IIgLealpOwhIFIezMzIECjSECfzSnlpti5zHloLegv5v2jgvBxw81iUxpjFmG5SPsxu1G2Y4G/F87AL+6HG4Cp3SfGK5dVYnrDMZIE3TBUJUJH2hZDShPbRuajQGzvHDhwrJNoT33TsfOYxJ1LzksZeKV2Y7U52pvtyrNhyW5CSGEHDDwAUQIIaQWGIIjXpqs9xqOITn1udOxobIit9smKgQRBjY0001tCrQum1Dhwi8iIjmUHShU6ANf/72Qg9p5DKEktF3RmbE6pCMiEkEa+QDEzhYOl2G24aa91ULo06YKMQZotVP0D2fgtcIMbrTX1+nHDq2FQkxFVt+FJkRwIB06a2B/RxhTRLujchltlLy0YBVfRWunCMpWhKrfErh2eYBxZWijuh4ORx+m+avjxlDXActl6PDwQNu2aQrSpb/+1a+Z5dNPfWnvM1byxVIOuuxDhKFW+K5JdcexhtvOsvwCK6ISQgiZ1/ABRAghpBb26QF09dVXSxAEctFFF/X+Nj09LevWrZPFixfL0NCQnHvuuTI2Nrav7SSEEPIC41lrQPfcc4/8x3/8h7zkJS8xf7/44ovlm9/8ptxyyy0yMjIiF1xwgbzuda+TH/zgB/vcWPLcgNHaAnN5IbauU69RA5qasqml2hYHLUCcoFajyl8LWvxA6QD4v1MjKfULTIUNQQTK8nI5Bt3GYTlvtS8so72wbW1WliywqeFDrXL7BPQi52y835QZgP4W6ItQXTEHqe2Y4uxA63DaKgnD/ZA+HYhOzwWbIdDOmiqFO4YU8yK3YyQrvDoWffcbwLnr0eq69tzzHFP3y/MJvdx2tDuC6x5oTQ7uECgRYVoBpRoK1JqUHtZuwn62T5jlHU9tM8uTk+W9lYwM2uPAmI/UcfbFLqei8sdeNlb63HOpAe3evVvOO+88+fSnPy2HHHJI7+87d+6Uz3zmM/LRj35UXvWqV8nJJ58sN9xwg/zwhz+UO++8c6/76nQ6Mj4+bv4RQgh54fOsHkDr1q2Ts846S9auXWv+vnHjRknT1Pz9mGOOkVWrVskdd9yx132tX79eRkZGev9Wrlz5bJpECCHkAGPOD6Cbb75ZfvSjH8n69eu9dVu3bpVGo2Fm7IqIjI6OytatW/e6v8suu0x27tzZ+7dly5a5NokQQsgByJw0oC1btsiFF14o3/nOd6TVas38hVnQbDal2WzOvCF5zpip7C6uf/yxJ3qfpybsXB7ElgYGzafA4yotIKwuQ41tMiWXof05xseVDUtUXb1bmqqE8uIFoPkM2zj8ITCvI1ZzMTIoKY6h9VRti/oX6l361Isc+gHKWGRgQ6S1D0+rKdDaRpVjCLGsAxxXnVGItje2BRKD/qJLehSoaWEpBKVhBaABRTi1p1P2RTxgf2MKsHZC7SxX55CBtxPqhqHSxwqciwR6UayWm7jOs02y5/f0k0/3PrcH7FjE+6Eo+msw3nVW9x2Wv0Cd1kPbNanvzlY7mtMb0MaNG2Xbtm3yspe9TOI4ljiO5bbbbpNrr71W4jiW0dFR6Xa7smPHDvO9sbExWbZs2VwORQgh5AXOnN6AXv3qV8tPf/pT87c3v/nNcswxx8j73vc+WblypSRJIhs2bJBzzz1XREQ2bdokjz76qKxZs2b/tZoQQsgBz5weQMPDw3LCCSeYvw0ODsrixYt7f3/rW98ql1xyiSxatEgWLFgg73rXu2TNmjXy8pe/fP+1muxXMGMygJRUdB1+4vHHe58xtRqtR1oqzNOdtuG6MIDhp3aVY1VNeKV3XqPVbiCulsBxnLI8acK2C4dsqGZkqHQkXjBgz20QQigtCFMVubJDgdBGCuneUVSG77CCZQb2Rnr9NKS9B2K/i1VOdRg0KdAJHJ3NtcUMVPpsgS1Rpqpqots1poaDDZEOe4ZYARViNJEaT87ZtOUAwoKxGpsYdoo9N2yceqCrtIKdkWDoWMeh+lev3fOHcv1Ay461BtgbFXDyuiptmqIVFRxGfRXvZy/kblb3v6/29t1+zDZ7e797wX3sYx+TMAzl3HPPlU6nI2eeeaZ84hOf2N+HIYQQcoCzzw+g73//+2a51WrJddddJ9ddd92+7poQQsgLGHrBEUIIqQWWYyAeGONGdu3apTYGOxSw5tEVUQsooeAw7q7i9JgKjm2qSh13oJmgHf0ClT69cMDeAouGbXrryGAZp2+GUNE1R2HKaihaAHDQT2gXlHXLNqewbhqqXXZVanUGFWnzDNoI/ZgWpWbU2b7Dbgv6na5u24aU7UGYhjGk9IwW6ILtBG1vQI9R/w8OsEQElmdQultryFYMzbu2L3Q6ewyp7XlqtbMM6j7kSsvJHWpYoPNEZd9EcK6o52ldDe2Nms22WS5iuzwwWI5bnFqQgSYUhP2thFC31VWAI+gnNELyULvWZzNj+vbv2zKrrQghhJD9DB9AhBBCaoEPIEIIIbVADYgIJvvjvIEASgksWLCg9zkC+/kY5jJojSJLrZaBk3vyooxj47wN1EwwDq/nSMSo+cCclSULylj6ogG7bmTQnk9TawNYtjmo1nUKNd9l97TVHECmkm6n/G6na/tpfGK3WY6VdZXDuVSR1Wq6udUgOkqTmOxA+0HDCl3ZjgSuayOetIdV+thAYtuwCCyLBgft/JdBZSsTQcc0wBspTsrzHVpkNRJJoQyHmuOVBqAP4VwrmD8VKLsdvB+wZIQuQ+9wHlaC2yq9Be4dXT5CROTQZcttm5SWU0A/eZqQakYDypH7tj1qXyHaKFXrX2Y/SvOscAIy8A2IEEJILfABRAghpBYYgiM+RXUKZbtdhj4Kz9Hahl9CXZUSkjqzFKx51Cs8Vs2MEghXBGi3U/5fqgXrRodtqGbZSJm+m4SQJltAVU0d9kBnZrFAJq90TWgDPU0gXVr/VxDaPzA4ZJYLlRbsIJU3jqH/wdJIwvJ6BAFUs+3i9VDHxBAPpKBnWX838myXDSEOg0O3duweAmf8AK57qJbRpicAWyV9vbDKb44x0ATcsNU4cBByS2FfYVD2KbqRo5O5Tr0OE5vKXsC0hBgN4VX6eoCVYx3aNenrgTZEto/tvQRp2FghuGr6g95OZgffgAghhNQCH0CEEEJqgQ8gQgghtUANiPjxWqzMiJrK6GjFvuzeGomKj6NlvGf7oapqQhtC0F8aYKefqHKYw21r0bJwyGpADTXqm5jmC8fNjV6B6ep2OYLyDE6lOWOJAgex9cRU2YRt8b+JKvUabfjxuy3op0zF/7uJTfdeOGg1CZ3ai5oPDhqtt2Sgbzm4dlNdLBlRfu6CpczwoL2W2hUngmsXeunF5XF2g76VQhvTKdtvU6oESYGp7mJ1tkCloOcgTLVbto8LKdPxowG0doJ+A1ulSJ0epoKnnl2QsiHCqRJguaQvJmo8s7XUEbF2QDlcx37wDYgQQkgt8AFECCGkFvgAIoQQUgvUgIhgQB/jviHEtUdGFvY+ow18s2HnGExPlDFvrKKN1jW6bHUjxNIAML+lsMcdbJbrUcsYatnvttWoT2DyDlqcGG0GxBicG4P/mwtNiQgsPYHbqlLZCRwHtSe1jHNfsBW6zIOISKLi/wmcD84l0bOEcthvUcCy6lO0YcGyGwHY4mjdZLIDZQUiq6Fo3Q2mnEksaO1Uns8EWO9MdKwmtHvCHreTl23qQunybgZzhtRxYyizLWCF1FJzuppDsCmc0IIFI2ZZl7DHkgpIrLbNMphTF6FWprRXnFs1Q0luvdzvcxV8AyKEEFILfAARQgipBT6ACCGE1MJBpQFV+RjNtC1iPJDmsN8DApzyAefXUuWYE/DuSiHenOapWmct/LGcgZ67gLqT5OD9BppQU/1fagDnCEH7zTehBAFa72uxpgCfL/SG8yz01B+SEHUELN+t2gC7CXC/aoMw6D//Q0SkAR5nsWqTg3k0nQ7605XbFjCPaXoa5nmoPo5D+7OSNG05gBhKTYuU+3JQ9sHXJLRfIPjTwbLu42ko1TAxDWNPrG44rcqIpKB3wRCXQs2z6U7DGIE5Q7t3lV9OcqtDgTWc7Jiw94uWnnIcP4C+Z8MCNR+YY6fGvD+moUQHXhB9HLVuJo2qt92stiKEEEL2M3wAEUIIqYWDKgS3P0Njs00zPBCZqZfMqzbYenSnrcW/rsgZQWgMIloSRapUAHRvHGIIDi1/yn03IYzTgrTmRlJ+N4dU5DQD6xRtQYPjBxbTrg1LmdRrCNtgWrbLyuMEEYY54EBqdbdr+xvHeALXJ1chugzONYWwTkftO8MKnJBrrcN1EtiQWwrWO4WzPzs6TNiGlPlGA8J56tr6KcK4XG6QQ3gujmy8q+vscQdUlVYM3wUZHEiFV7vojRTavhB9PeC64rV88untZnn7jmfK9rVs+wOwFsLSIJoEQooS6fsZyi/MUIlY/xYwDZsQQsgBAx9AhBBCaoEPIEIIIbVwUGlASFWccia9SKcrHvBp1x4znLtOl06sxuACtCkpaUCcPRBMBy2Ho4OyCA7SQfHS6XLFUMXZWw5MqzAdun8pihy0gCyFNmEKq7LTx9LMEcThO2rbHDSTAOz0A5WCHkB6t8NUZLCC0doAVqXG5cKV++6APpTDBYjUT0kKqcgDQ2BPg1ZPsdYgII08Qu2vbJPD8RPZ42Tq2iZde8w0xZIRcD3SUh/DMd1uwjiOyuVB0NwKaH+Q6Gtn9zMN5RemxneY5V9teqj3+YQTTjDrcNzmSruMsE05WiOVbcT0adQjc/yuHgfUgAghhBwo8AFECCGkFvgAIoQQUgvzVgNyzs0qjjgX/QX3l1XYWWDctMqGHO0rIph30i9Xfm/g+cxFp3I4X6TyODp2C6WlYb847UEvd2Blp2PnMoRKb8kLaz2CkxWcmlvioAR0ASNVW/yIiATK/iWG/g8gPu60HQ2cewA6VabnZsww7wTnyuhh0YV+wVh6pubGTKHWNG0tWWI1NyYKcR4T2EJBoztdPfcHr529PrqNVXNsRERcXpbdaDVQ84H5X9CPudLSwghqLMDcmCQu59WEiT0OanAddT3SKTvnaTfY3Exl9riF+mnUxxTx+0KXz8BSJg76v6muF57qQMtaFIWJXd788Kbe58MPGzXrFi4YNMsNZZEV488CzBkKtP6FFwcm5OH8vFxpjHFc9tlsf5f5BkQIIaQW+AAihBBSC/M2BPd8gGE2Db5CVr1SemmOUAFy165d5ToIE8x0XI2XAhn0bz9SFUKMINVyaNC+zj+jLEBERHbuLsMiDXClTmBfiQo5BFiBE1KEE1WpMQuxn8BqJMLzUWnYDdumGFPFdZgq7V9tVMRauHh9iC2E66NNiAvYOgPX6kll4zPRsf0yNW3DR6LOxxvDEEKZnrZhtW6mluF82gNg76LCmmin4529us6txPY/RFOlCeNAh4hcF6rkDkKYTbsvCzhnQ7p6HJR9iuHRTgpp2V5YTYU5Yfxg6FifTRCDzRDEv8JYjVPop6hhQ30DQwvM8oIF5fKWRx816zqHLjLLhx12WLkAbRL4DXLqfAJwznZemBxS39V11+Mf74V+8A2IEEJILfABRAghpBb4ACKEEFIL81YDCoLgObe4qaraty/lFvx0aKWDzJBmnWVof672AtU6C9ARclOqEY4DtiV6LeoIHUh9xTZpnWRowMbot2/DCqNGCDFgqYBAVUh1Kabf2u+GYNUzNDRSrgN7mggsT/KwPE4E8fEitXqLbmOK1vQCoC2OOuEMts5hWafyDjirBeRQDVZb8+SgjQlYv2AkPlCpva2WvXaoBWrLlqAAnRO0An3qzmFVUyipAP3UUGMkdFDSAtLKi6TsU0xBj+C+azXKc2227X6LXaCrBajjKj0S6oY0QbsJVamQFpRJcAFYIWk9CfSVAEqOrDryCLO8fPmK8jgDNkU7B40rTcvzbYC25DC1Wt1LWE0iBwEP7y39E6RTzp2b3bsN34AIIYTUAh9AhBBCaoEPIEIIIbUwbzWg58KKZ2/HmC1zOU4B8dimssXwLFjQFn6Gss9VqwJTphrnMdltdfwcNSAsMyABlElWJa2POHyFWbdowbBZ3rz5l73PW3+7xe63gHi/Wm42wB7IflNiOL+WsulHKyHUygKlpUFIW2IoF51ryx8sAQFtikCo0utDWJdmYN+kFnMoDRDHtp+66rsd2E8GY3pwaMAsmxLkaK8D8f5AaTlQ5VySpj2fVqvUGRLYcQO0Ja31/e4vvU+oF4Uwz6yhbHFwzkoMel6izrU9YPshSnaZ5clpnCdUzpeK4eRTGLctpdkFUIK7NWg1oYbWj0DzCaHMw4tffJRZHhgq5+c1WnBdBef2lG2eBsuxNvSTuTqgrRY4Jqo0bHU+aEnUD74BEUIIqQU+gAghhNTCvA3BPRdp2M9VuA7XVb2m6qqfIsbBRER82xK9r6DCTmfPvvpb84QYMFKpmBi6wNTLGNKYA5VaetThy826sTF4hZ8uLULCdLdZ15m0YZA8U266aF0DbtINSElP1P+lGtipaC9iKjfaTb0UbmVZjGG/CJfRwVt9xgqirZZto7aKmZqEkEnbhnVE2fY4qAIaenE1CEcqKxUvlRrGrR4VaH2EYammCpWBGbk08FcmwLT+sk0xhCob4KydRCrUB9smkG6cDJRt7MK2S6fteJKdU2Zx+zPj5XczcHiHr052VEVUG52TxYkNwUWNsk0B9OGSJdbhemjEhrNjFb5rQLjOwe9KoNK9HTqmg51OUGmbA6n5+N19/I3mGxAhhJBa4AOIEEJILfABRAghpBbmrQZ0IIPajI6b5mAt4pdYwPRpFcsVqw1UJTpiZBbLAWjxw9OsQmy/DWy3lR7wq5/ca9bt3r7NLIdqXy8aHTHr8tTaiUxNlXH4qekU1kFZgSkbs09UCmjsVXHsXyUUyyKgJmRS1KFf0KIIU611vzrQrFIoRaH7ybXs9UhAKxhQJQomoB8ySOH2Kr6qNqUwBUCgL0JT4sJuGoPlj5YZIkEN0S6j5U+irIXaDauZNEHj0u2PQK8DGcQst8FKaGTEpjE/M2krpDaa5fYTk3idoYyC6pwmWORs3fakWV7RLsskJKE9txWrVtv9Nuy+EqWHNaB6aop2Qc1y/XTHnluAFZzVsqfohDieUHdW1WBN/RGmYRNCCJnH8AFECCGkFvgAIoQQUgsHmQY0U1yyf0571dweXIe6TleVN8CU+wBirNgCWzq7ugS3jsGiHRCeeWRsMyxZluJfzNIvH/xZ73P+jNV8GpM7zbKej9BKbNw6A6v6QRU/77ZsWfDxeNwsT4Cu0FZWMCjkBCGeYdh/W8/fqNw2Bsv+MIb4OHxV7ywADSiE//vlUo6RGPQWl8DYUzH7dgy2/FDyogt/KJQwgqUOsOxA1i11t2bTtrfZBP1F2+mALUyE2hIM45aaQxSjbY+nOZTbhqCNYR/rUuwDYPEzMmTP9dBFtvx12tE3KpRqwBL2Q+V8nQjalEFfbHvq6fKYDWtjlcCFj2C5oTShOLH6kYDG1VL2XxnMY/LswNRvEPah5Pa7+FtXOCzV/rt9Fvgbsnf4BkQIIaQW+AAihBBSC3wAEUIIqQVqQIb+GpDvBVfsdTsR32NLx5B9PWKm45Tx2hyrJOB31fl5+0ENQgXisdQ36hOdKVu++JltW3ufG5ldF8YwB6RVDrG0Y73gvK5Q+kQLbfkHwOfL2Th2W3lsCWgZDmzvg0LpFdCGAkoJx8oLzrtyEP/OcX6F6uMAah2jN1+mdIUmaD4oyaWpKicBgkoa9vdZ29PGMqZfwFyeGDStQvmutaH8AjTRzFFDC3+0+IfDSqDaHHiehTA21WrUixLwR9PjCectjQxbzSfL7L62P136FOIcrikob9CZLudidTv2Yg3DcWJVBr3bsffOokMOMcutph3zhRrXBWg+DdCE9LhugHbcxftDeyOibyIUdUfNOjT6UX9duR98AyKEEFILfAARQgiphYMqBOenyVZs64WwcL2yc4E0RwxB5Cr8UniVStHSpH8a6kzlGCpLRkAIKFftTyEtE9M0E2jj02Nl6jW+sicxhkHUflMbniigT01KJ/iqNMELJmhDiQgVKohwVOf2uLlqM4awGlCawilbn9xBOi6E3PB/c7pcA6Zs48mHKm3VQXgxgjHRbJXn3u1iyrPtlzSGULEKh2XYBgjVOLUew2ZY9kEPGRzTMfQxjtNuV30ZDtRIYCw2dWo79j9Oh1AWWBDGbMU2ZDUEFUZXLitTpB/dau10oqR/unTStGnxmCo+vKAMyTWHbHgOw3WYlq2rxeJ+Gy2wBzLfq57qocN1AdhY5VU/ktL/N2e21ab5BkQIIaQW+AAihBBSC3N+AD322GPyxje+URYvXiztdltOPPFEuffe0hHZOSdXXHGFLF++XNrttqxdu1Yefvjh/dpoQgghBz5z0oCeeeYZOeOMM+TP/uzP5Fvf+pYceuih8vDDD8shKn3wIx/5iFx77bVy4403yurVq+Xyyy+XM888Ux588EFptVoVe58P6JgyliSwW2YqFRPDnSloHboEbgEbe/Y6aD1SUZLbO7AzwVyzqgHlijVYXhnLYYeQxpyoNucQX46gJEGq0nOxFEUHtTN13CGIj4cQ7295pQ8yvbFdhwKG0ug8e3nIL05VyQUva9wr/V1Rrhj1OumvH3kl3iOrgzhlj6LTxPfsB8qEO7juYdlPAVjM+FH7cn3atdfKob2OTsEt8N6ptjvSWhneD5CNb9LIE0iPzrCNqlcd6rLQ/gHQUIYHy+WFI1YvmkqhJrcaX1i2AsuGt1SqeNyw7R8cHLJfhvMLdKp+E9KuISXdqd8n1IA87VLdW54VDzrzVJSP0dduthrQnB5AH/7wh2XlypVyww039P62enVZw8I5J9dcc4184AMfkLPPPltERD7/+c/L6OiofOUrX5E3vOEN3j47nY50VD78+Pi4tw0hhJAXHnMKwX3ta1+TU045RV7/+tfL0qVL5aSTTpJPf/rTvfWbN2+WrVu3ytq1a3t/GxkZkdNPP13uuOOOve5z/fr1MjIy0vu3cuXKZ3kqhBBCDiTm9AB65JFH5Prrr5ejjz5avv3tb8s73vEOefe73y033nijiIhs3bpnhvzo6Kj53ujoaG8dctlll8nOnTt7/7Zs2fJszoMQQsgBxpxCcEVRyCmnnCIf+tCHRETkpJNOkgceeEA++clPyvnnn/+sGtBsNr145lyYbaxRRLzysmino2OjGdht5BBD1mFDjKk6mF9h2ujpE9Am3+9FtRZLKPc/9yConnsRq/kuDmPC0E/daWsZMqDs5zsQ9C6gxML2HTt6n3ftsFY8OZR5aKtyDA5Lk0MZZLw+mhjm8gQ4MahTxvAdlkXwHIy0xQ/Y6XjlimFfqY6tox0NbKt1EtRI4DhZ/00lxjk4GNNXbUQ9skoTCkCMQZ2zk/efx5TAmI9gudVWpaax3DVoXIXSIDpQph31Iyd6zhBqe2BRFKC1TXn2C4ahD6exxEV57p42GUFJBXU6jRbMbQM5OIc5OYke1+hYBPdwYebFYa1y6UtRYL0Y1If7o/Uh1Ir6Mac3oOXLl8txxx1n/nbsscfKo48+KiIiy5YtExGRsbExs83Y2FhvHSGEECIyxwfQGWecIZs2bTJ/e+ihh+SII44QkT0JCcuWLZMNGzb01o+Pj8tdd90la9as2Q/NJYQQ8kJhTiG4iy++WF7xilfIhz70Ifmbv/kbufvuu+VTn/qUfOpTnxKRPa+5F110kfzrv/6rHH300b007BUrVsg555zzXLR/bqAFCKRT6sql3a4NO+GzWqcUY+qil66rU6lnSIn0wi/6DxhChFd0HRLC8JB4qbA6fRJCbhBewRKWyUBZrdRByGHXrl1meWqyDJNEEAqLMQVdhWZicF9GF3EXQsqzamMX+9TrY+0PBOu8sFR/1/PAQZgQLrxOTc5zDPnYEIVOQUd3cucwnKqOgV04g9u6DsGhvXqWgkuy+owhqyyz7U+VZXcIHRHDFIAWuFZrl6IQQkCd1IbZ9HExBd3BzaOnD+A9WmD4HToyVGN1sAXjtmFte+y1hXBjAiGsQKfBoyWR/T1qDuDYLL+bdUEigNtdX4+oYfupA9e5UPdSA6y0vKkf8DuiLaZMyNYrLbx35vQAOvXUU+XLX/6yXHbZZXLllVfK6tWr5ZprrpHzzjuvt8173/temZiYkLe97W2yY8cOeeUrXym33nrrATAHiBBCyPPJnM1IX/va18prX/vavuuDIJArr7xSrrzyyn1qGCGEkBc29IIjhBBSCwd8OQbPtkTFHlFf6YDm0+lMmWWdOoippF51BkxvVXipjLpqoOAqsLLB2KnaF+pHnhWMaiSmQaIlvpZ98FwxFj0xNWmWD1myuPf5qS02hT4DIwttIY+lDiRC63d1XLBK8SxzwF4nNOtQMwH9QsXPQwyzY/q3Xgcxe7R3wXi5vj6e5RJoELHqpxwrVmJGqzoubluA/oL6nrXFwbIhaLej1oGG1U2xGmzZyARsbfA+LDKsyFn2cge0pSef2WaWG6pKaBLbsH6zabWZaaVboUUU3r9oN6X7qQlWVTHed8pSJ+1Wa7paf0R98Wc/vd/utj0slrJNCYynGHQefb5Jy96jKIcMD5b9huM0wJId3tSDPpojyzEQQgiZz/ABRAghpBb4ACKEEFILB4QGZGLvEONO8/5zeTA+ibbwOAXELleXSQjCMlYdwrYYstdWN7FDawsMElfMd6mazwLt8PZbYQGUY4wetwWrpEFVfgPjy2irND1ZxqbRmt7BXJgoKOP73WmrO+nYv4hIiBYhSufBssLYF3o1lkXwpj1If80kxLlVcByn5tkUMGcILX/CoL91CVoLuaC/DRFaRqH+pe8Bbw4ajGOtZaIGFEOJcT0/Zxrm7hTOtr87adfvmiy12Icff8ys2wH6Y1vNIVp16KFm3dKRhWZ5cLDUUNIOWCHBeHJQKkTrPAF2U2DnyWlbpQD0oi7MnxKl2TmYa5jB8oJDltjjqM/NBPVUvL8jtQpLNaClV3/d3NOoBdDfVXZMQYVVltn/rLYihBBC9jN8ABFCCKmFeRuCK1whxe/CB9q1F1OnMWVYh938MFS1DU5VCjdaS+j0YnyKh953yy0wnIKVSnNM/1a78mwxPEvc/uneMbgMm/AdZn5Dai96CzVVGufgyCFmXQGhMx0ibYLNR6dj3bF3jO/sfR4QmyoaQsgBq7hWgSnpphvRyRxPVhc1hX4JZYa0eBW2wlUZHFe3EMdaN7chH93+qhDbnvU4Egq1DtZgm8y0hP7THUREYtVRXQjXpdD+7rS9Z5/eNVFuC67bQWvQLEfNclxM5bYNO6bsb0Om7ocktvfZ1Li1jMJ+GhoaKb8LaeVYoTZqqeOEaDdl95upsGwOVkghjC+s0horZ2000S9w3OoQIrQBU62zTF+f/pWTRfbicm1C33q/s3u34RsQIYSQWuADiBBCSC3wAUQIIaQW5q0GlBdFLz1Yx0qxEiPGpvVyCIFST1/BFFz1PMbQeQSxXZ0GXEBcNAArnrhRfhdtPDzXckij1WHiBqZewpdzXT0V2w9xYF05s4AYfQ4plJ6WFpWp1iuP+gOz6oHtthjh4MJSI2p4khXYxuzeodoE1Ue9VGpIsVfnjtVSPb1CXQMHiaUR1lTQacswKFBvyVKbRmuOixcarYbM+VVU1MX9Yj95WiVWAi33NaMGVPTXsApovyhNruhW9xNWOV10yKLe52RwxKwL2m2zrC2YYrh/Q2ev+26tD4PuNDlptco4ssfRCcet3E4tGGjDlABVPTX0JmFAX6jPaH3UAf0Ux3iifkc8S68Kyx+s2YE6tE63T7v2tyAEvcsrOaL2rcean+K/d/gGRAghpBb4ACKEEFILfAARQgiphXmrARXO9eb/THfKWGkKNvBoN6+tYLyYdobzc+wxdXwzhDi1b9+urGxm0ByMNQz4ehQZzlHpX2Yb91ug/qV0EJz3g3HfUOsZDrQy6CecQ9FSfdyOD7dtSKwlvhYaHGhNOHdhWM29wPktWKYigbh2qrZ3aMWDF1rNU8Gy2qjBFUo3zLoQ38c5NzDeIt1GLJ+O5TzMetABc9SLKkBhpyJmn2ZYUgF3pUtao+1Q//22oGR1AXZZYRNKvKvjRFZukTxEHbdsZAbzaCYzO74SNeYjGC+DULohKECbVaW0U9B0p2HuYdQot23C70YKtj251hyhHEkIy00oXa7vAZzLkwuOW62FQ7lxtNpSX81AY08Ey55b/DIjc4NvQIQQQmqBDyBCCCG1wAcQIYSQWpi3GlAn7cr070poa/8hT1+BoGSm89gh7lugBz56eal9+xoExFwrfLK8ALluA+gEWdbfy07EakBdmNzTgVi0bj9asBfOxnYzdToTkxNiN7aLLoI2q7kOTZin8eJjTzTLm3/xk97nfMr6b0UN+12tNLnczqnBKucOygHo64V+bmGFaII6WqWGAm1AXTACnc2MW8H5ORg7Vzv3LAz7l16vmgcnIpKnoDmqeTQOxr9DbUCNpxy0AZyT5pQvWwZiEp57B7SacVWeIUVNzvXXU335zp57Vw3kpAG6UwLzv7DUtNJjCtCactBMO0ozLSLYFqU+PZcHK3tDKZMAxrHRqHGuGF47fW1ROwaNV/vVOZzTCF6VMV539VmXZc/xhu0D34AIIYTUAh9AhBBCamHehuC6adortaBfC/FV2U9ZLZe7ULphpmp/Ol3RS3MswAIIbcn7tkgkU991HQy5VVtWOPUOn8O23S7YuavzmXTVKefTqvri9qe3m3VTu23FymbL9sXQcFlp0kFF1AWLbXmGkUXl8vhTtg8xrhaoMFQBKc9oLdTFEgtqPdomYbp0lpXn7sB6BK+ezvLHjFO8cl4qrI6ZYI4zEFakYWP4V9vnB/B/SEzvziFGrW+BdMb0bhXShbT+HO2b1Nibzu346UL4bgL6vKMWHVRP7Uzbe1iHdppYJRfuZ13GYvcu2yYMd7VhHDeGy+U4hpIK8FsQqni2g23DwJ5PotK0MxiXTQh3YXg41dVU4ecnw6kTjZbaFn8LYEyofcUQGsYpJjh1QleADdR3MSTYD74BEUIIqQU+gAghhNQCH0CEEEJqYd5qQC5Nxf0udqyt3/PMxiAnd9vU3mkVM/bcTkAIwZh3S5WaHh4agm2h9EGFBoR5vzbtEeK6EB/P0JpHl+RGOyBok9Z1JidtueupaZvWPN1RFjNYGhj1IvtV2TleakaeJpdajasxWPbjosgOt6effMIsP/n4M73PCZQnbjbAEgR1HmVx5OlqVSU7vHRoi7aqx7h26KXm9y/FjncaRshNSWicWgBamY7pF0G11ldguq4pOw+6Jpa80HZToFd0wRJr11SZyl9g+2FMd1O4B1QKN5bhwI6L1DXodmwbsJ+0blgEqKeAdowluYfLKQJeWRcvH1+NJxgTaIGlS1ykoCI2m7YMPVpkRfp9AaeJYNmHXOuEM6RsR/21G5y50owwDXvvJRhYjoEQQsi8hg8gQgghtcAHECGEkFqYtxpQkoSSJHuej3puD5bVDqHUdKTy8DEeHoF9SxzbeO3AQJn7H4QwR8Wr3dC/7Xlh5xw4HW/2bFZsHDsB2xtRliFefB/i8CJl3BdCzzLQhli6imt3oYRyAnYbgTf3qiTGbmlZrSYcUnY7oN/FbTv3Qs9zmHz6advemdqkyjE7sDcqPB+fsh1YFjwHPcxoSxD795yd4OLquWSe5gNaR1Fhr4OlmUM1bj0bKABj8V2lH6Guk8G2U91yHONcHtSAuqrPUQfBgesZYimtrInjB3tOfRnLJGB5ia4aTzHMGerCdYZqDLJLlfAeGrKWUQ50N639efcDDBL9exXG9lxHDllivwzl4gN1z85kwaRPB0vWoDqTKT0pgDGRgOaTgp7XbyzONC731k5CCCHkeYMPIEIIIbUwj0NwUc/BVkfD0EZiwQJIl3aDvc8ppARDtM6z2NBWKpiimqH1i/qMIZ6i6J/WjGG/IKy2ydAhoAhOIMPz0X9w9jjoHp3E5XE6mGYKoZggwvRvtQ5e6IO8f2zSJXZda+GIWX7R8aWT9s/vu9+sw+sRQJhNV3GNIKXWtzvSKduYUotpwMrRusBUV7ttBOEKHf7CZG9sk8m4RYucAENlepxWp7tierFOVe5AGK0Ddi7P7C5T+dH6CKchaIuZBqTMR5DamzT6TyfAaRZ4eXS6eiuxactQQNTcAjiG8X54ZnLSLE8oy6w4seG7BM9P39Pg19SFa5mo8GQcgZUQHAftj5wZRRAKq7B8R5dwT04odEh6hukm6Kwtew+74e9lP/gGRAghpBb4ACKEEFILfAARQgiphXmrARV5KkWGVvkiCcRuvRi4SumMIhsjRg0I4/+histj7BarE2pbeIypoi18Q8VyUTNB3xIvxVPrFX4Ca9/lMIT4MZxroKsXwihA+388rEkvLjDFE/QLJW7kEaa2g32+kiRGj1hp1j316K/Ncgw294ErY/ZYvROrwYZa6/PSWcWidB/U+rAUgpc+rbUbr7Jn/0qfXdAQAxwyFZYnOEZytGhRm2NF3ad3Wlurjrb8wdIlkELcVvb/EazD9Og0BT1J6T5eiRTX///IGVznBG1vlOUPahk4xtswJUCLck9t32lWacsuEZGFC0sdegDGjzelQV2PZmvAtgnOHaeNhGrMBPDTjVZD9qAzpcGrshugYTn8Lv6GqgHlaUuzgG9AhBBCaoEPIEIIIbXABxAhhJBamLcakJNCXE9PULnmWBK24hHq6ykQD8eYvQ6QgzaDh2lU2Pin/aedeLoHng9qEDqsWoANBoZ9G40yjt2FsscFWII4UTYlmPvvRYkhDqxLFOCWGIdXtvAF9CnGvFuqTHKkSnmLiDz52G/Mcgb6YOC0zgBzEODi5aofcfqEr+uoz6AB5aDV4LwyHRP3ZhflqFP1j+GnsG2kxhCWKsc+LmC+UaE1LdBFYtButK8MzjlDOp2yZke6G8too4iIPzuqn7AMR2K1Ga3jNmK7DvUKraulua0pUsAJJVAOO1RtHGwPm3UZzC9MVU3xAuYxdZ3dVpR2+aJlK2z7oZvyHOYxxmUbsYxI1Wwwbx4f6jp4fSq+i5pjQ19LtdtolnoQ34AIIYTUAh9AhBBCamH+huCc66W16rc5tDDBNz2dCotps1ix0k+b1euq3bBNVUovfDKXdER0dbaLet8YLsLD6r4JIazmheDUYQNIdQ3g3L2zMx2FadjoVq5S0LGKKdqsFGWYpID0VbROceDgra9tjucKab/a/QhDuBiO0CGHLMNAR7UjsR2LGAqDfanrhW2IIKypnbTxfsDrnkFfdFRKdAZ2TQVc92kVxtXVdkX8sGBYGN8bs67ZsGnLXjg77u/4HmLIXbU560K6elgxJQD6JYVrWYg9P6ec5QX2iyFq3cUBnHsEob2kvaD3eQG4X8ctdN02i+Z2wahZ4UkTynXbC01iWnb/3zIM+2Nauaah7ZjwXu8D34AIIYTUAh9AhBBCaoEPIEIIIbUwbzWgPC96cX0b167WcXTa6UwpqYKW+OozWsg70I/0ak/xwTi2irnOVMkQ0Vb1WGkSK3LqzEy0VXeC6ZTqM+Rzh5Db62lNal9+PBnTzPUCaE0RVEhVB8J03Fbb2pbs2mntUQJ1fQKsJIm2PUEZ33czaTP6Ons5/3PQgBxU4EQtU2konr1OgWOvouQFtCGD42jpA6tbOui3XJW8QF0tg3RjrTWNDNtr1YZrNwRah665gIVuM5jToNuRwlQDh6UclGiKpUBQH06nUVvWNVSgUQH+bpTLKf5OhDZVvFBjqD28wKyDr0oMZR/0MMApGFHSf1qId49WlGdA/TGEc8d7QFv3VFlE9YNvQIQQQmqBDyBCCCG1wAcQIYSQWpi3GpAEea88coVThDdHJTdaAMynqMir9/aLtYABbefulyCw25pwM9idY6wULSycsuNwEH8Npf9cmALLUntlE8p2YMlnXMaAs4nTQ3tjnFui59FAn6JGp2PTEViNDICOMAWx6byjvgtXFudtFKrfPMkHdR6zAfaD3TaFEtexWp9nOEZgnpna1rNj8jS4cl9oo9SFNgjY7YTquwGUOddltUVEBtrl/J0AbHoC2G9btX8IrlUbSk9j2QQ9dwbHdBtuWm1HgzZE3rwZ1ZEdsM/pdOy5j0/aUhRdZd0TNaANTdv+SGk12CYUagN1/+cg2wSoC05DORr9XbTHylCbUWMPxWIYYLqkODqMobUW6kmF0mr1/eD9hvSBb0CEEEJqgQ8gQgghtcAHECGEkFqYvxpQUfTEEz0Pwps3AzHXWJeX9XyxsAQ3eDrlZTwTrd1RiCr0HIMZ5vLoOTmeXiQ4nwLLPhuDOvtdh3MkdHlcG8yNwVZda0JYPgI9wfwQsv4D9BMcV28awPyJqjk3MXhOLVlifbPGfv2I/ao6UABtCvDaqeuMY8AreWHmAcF4glLTeD4ddS1TKB/tl55WJdJzu1/UBWMVl8drgxNpcE5XoMpwNGBMoJdarASBIAIbfiwxrmz5sYR4t1NdGkSLXAmIEDi7paHvDzxXWNbn12xaP7o0sbqUgzlpO8aVtyCMiaEFg2a5qcpCePoWeKIdsvjQ3ucIylLgfZfCXLgwLLUmB1pyiPqX7ifYL5bdKCrk7sCbzwNlLNR9murxXrVTBd+ACCGE1AIfQIQQQmph3obgXFH0Qm865IPhrrCiJCpaW/ihMrQw1+UvK1KPAd+OBixO4FXafBdDGfgurVOGITzn9UVUEa7D6pfqOGjXgpUYMZyn41JeWmZFaQq/dIDd0qRhQ3iokSS4MRy3/NyBC+9HU1Wfoi0JXA8drsP+xkqfDlPfdfgRQjG5ZwFUHgfbm+G2KiSUQbjOgd1RF8OPKmwYedZUNlTW1GEpCPnkob0eTl3nCFL+HVbhgLGorzVWG8W05iwr06MTuPdbLRtm09ZPGZSPcHCfteG73Wzv6cUiIs3Qbhup6x5DWC0X20+HH35EuV/MG4fjeKnM+vfJC2dDyrYu44IhaQhs2lkVdl2G9xJafJmfHHWvz7IkDd+ACCGE1AIfQIQQQmphTg+gPM/l8ssvl9WrV0u73ZajjjpKPvjBD3ohsiuuuEKWL18u7XZb1q5dKw8//PB+bzghhJADmzlpQB/+8Ifl+uuvlxtvvFGOP/54uffee+XNb36zjIyMyLvf/W4REfnIRz4i1157rdx4442yevVqufzyy+XMM8+UBx980I/RVpAVpXW8iaWjZgLfM3FTTE30Ugr7l91GC3yMaYYqHuuV+hYoFaxTVD2rHSwBPTsLi71hNBbYD6b2Wj2jf+qxiK/VYPkGs67AfekaEdWW+HZd/xLDIiJdKLOdBGWsPQPNDW1wdOw9Bh0hiDxhSq1EW5v+Vkh7vqpSw2coi6xTkTFV33N30Rm2oPmgRYsuqbDnuCrlGSxm8LLmqoQEyoBd7/5Q1k6oq8EA8jVT3cdw7QRQx8X+TuFcG02Vah3D9ABMQQe7oOGhMtUadZAGpDHHYfndHESrEOyNkqT8DSzgnkTrMBwHWap0HbwgBY5N3T6wjEItVj0GPE0d7rsCrLa0pZH+fe1MT8tsmNMD6Ic//KGcffbZctZZZ4mIyJFHHilf/OIX5e677xaRPQPvmmuukQ984ANy9tlni4jI5z//eRkdHZWvfOUr8oY3vMHbZ6fTkU6nFBbHx8fn0iRCCCEHKHMKwb3iFa+QDRs2yEMPPSQiIvfff7/cfvvt8prXvEZERDZv3ixbt26VtWvX9r4zMjIip59+utxxxx173ef69etlZGSk92/lypXP9lwIIYQcQMzpDejSSy+V8fFxOeaYYySKIsnzXK666io577zzRERk69atIiIyOjpqvjc6Otpbh1x22WVyySWX9JbHx8f5ECKEkIOAOT2AvvSlL8kXvvAFuemmm+T444+X++67Ty666CJZsWKFnH/++c+qAc1mU5rNpvf3whW9ssU6Cok5+X55WbUPjKlC3DrBMraFmbhh94vtM/ZAuBssA6G0gBA1LIjzVugIM527bR/El7HkAs4B0dvCcby5PWpfaCXk7deUI+97SBEBaxsMRXs6gl1v9BaYc4NzJjKlbURwnVGX0nOv/GtVPa9Mz3dx0IYc4/C6T725SRX7BU00d3g+FXNJwPYGLYxCdX44B62BJd4Dfa6oG9gmTKX2D1rPy9MqrdKWy0DbpAhKN7TDso+TwGo8jdj+5mhNV8TqX0ZLEt8aKWmU6x2MvbAB84uUJVPDu9dhbg+McTNWvXlkMM/PzJtDry1c1KXjwY4J5t/hPeByNSdQ/7ZVlI03+5/VVr/jPe95j1x66aU9LefEE0+U3/zmN7J+/Xo5//zzZdmyZSIiMjY2JsuXL+99b2xsTF760pfO5VCEEEJe4MxJA5qcnPT+JxpFUe9tYPXq1bJs2TLZsGFDb/34+LjcddddsmbNmv3QXEIIIS8U5vQG9Jd/+Zdy1VVXyapVq+T444+XH//4x/LRj35U3vKWt4jInvTKiy66SP71X/9Vjj766F4a9ooVK+Scc86ZW8PCPf9EIN0VXj0xtFSo11h0co087xd009UhB0gHxbRTc1iomIgpknlVuA72iynEOrMarYNwX+o1HF/JI7TjUK/IMdisYFXQtMBwUTlssE1h3D985yCm4Bw6Z/e3LKp0CRcx0S+8dlh9VLcwB+uXAFyr9Xo8VxwTEab6qnAFpvWj83Fm3LwtMYTKzHfR/gfCOugqrq9HAUfC0Jme0hBBCro39tQfUgiN5ZndOO1gqrVyZgeHcedVklXLGJaFlOdJlQrcniH+i7ZEgQrPx+AEHldUDHbwf/qVR64yy82WCv2hXZaX2g6hfHV+BU4tsC0Sp8KacYJyAra/vO5hhPco2n/Zr6oInDl3PJd+zOkB9PGPf1wuv/xyeec73ynbtm2TFStWyD/+4z/KFVdc0dvmve99r0xMTMjb3vY22bFjh7zyla+UW2+9dU5zgAghhLzwmdMDaHh4WK655hq55ppr+m4TBIFceeWVcuWVV+5r2wghhLyAoRccIYSQWpjH5Riynh5SaPt/L24KGhCWY9Tb4nelv9aBKdu4bGx7MFUUY+tqPVq/YDjWLzehzx3a5KVA693a42TQL7qNGcR5q0oqiFjZAW1u8P80Ors1A8t4PxVc6QhYKQBThBt26KZTpYNGgdb7sKyroKIGV2AKsUqrLUCjwn7L0wprIdTvov6aBK7BbO9cXQDPkiVCHQ30FtXmAm5/dEbSJTwaoO050NmKVFuyQL+gtgS/Ok5d7PbAgFmXdlD7UzoDtCEU1CqVXRaY+mAbJYVxq/Q81ESjGH82y/NDDevwww4zy85sC5ViC9R8QJtVzcAKx54mqlOr4bcrxvbr8iQw2LxrhXqqOmyors1sHyx8AyKEEFILfAARQgipBT6ACCGE1MI81oDKktyhioX6kU7QNlQc1S81beObCeTHa1sTLE2LE3CzrP+cFbT1wO9WgXFfrQl1u3bOSpT0v3y+LlVlrzNDaWC0DFExZc9aHzU5tYyx9BDnEKnjojoUwPyEtDNpv6t1KbzuoM2YChGoAcE8LF3awdcQLagA6Xi/X2YbdTdV0hrGAM5Xc3q+GrZhhnIlhdHvQIPA+WxKZ0AdKoC5SXoeUwrzu1D/imPQ89TY9Oar4b2jLl6R4JiGOVxKc0yg/QlY5hQBtlntG8tsBzjfq1w+9oSXmHW+XVN53fF3YabfCa3lRFjiHY4TmDlc1fMfjX0Z7CcFayTUj/Q9nRfdvX6ugm9AhBBCaoEPIEIIIbUwb0NwcRCWKcu5To+G0AWmIut1nv2J3RbtXcx+ZrCSqHpdDjH8pXaFIZEQ39C97NzyG42GdeXNMkytVqDjrecqrNLIBUORuG1/qx6sUOvbBSk7GtskL3VUv7Z7UQNIuW1AGGdqd1nUMEsxvRVsb5RVTAEhBox+5RU2Sg5dw+1qU3kSduullet9Y+px6Dmo69BedQVO/C+msbYJcZD0d+z2XJJjOxYLVZE2TmBMdO31aKC1kLpBIrRrgoGgbXsyTK0Gyx/tYt0I4KcO7IKkZftNHzeDNjUH2mY5aZXVU5cfdrhtU6UDUP+Q597W69+zMJgpzK/Hbf/pGrhfbxoI7NWzrupjIRVUVDuu2j8hhBDyvMAHECGEkFrgA4gQQkgtzFsNqJt3pZv7qXxe2q8XJ9XpoPa7mCIsYmPTOsUQbVZCT71R+/XFAbNoY7tYFgHSvTENWysnYAWD1um6aiVkjnppmmYJ2ishpsmCVqN1BD+gbBa1RoR6EX5XlzdAyyW0kEdLENEpuFA6oAhR+9NpzPbcp6dRVyjbFKNVPaacQ7+ZLvesUyC1XY9r2BZTq81+YF0MF96rjKs0Ik+rDFFTVO0IoGoxaDNaksN+aYNmgvJqrjQgrHKKJS/0/Z/ktg0DQ4NmWaf5o84pmHZd4P1efrcNFZtR5zz8iJXlAlZD9mx71DFC/O3qP7VDxP5WpGnHrMO0bP2bg79PaWa/q38NcKoEng/eA4FKdQ+MbVj1ufwevgERQgipBT6ACCGE1AIfQIQQQmph3mpAhct7MVEdS48jnH8AX9RzDHKMacMyxmBV3BLtQrAKr46roqUMzngpKso8oKaFmoQ+jlcyGeL/WufBaQFo72Jt42E+C8graG3jVJwe2++pFWp9Af2fwZwC3Sacj1OADRHOn9LXMkdLFrQSUmMkzdESB/pJXa8cYvRFgBpKf62mm1fPN9Jj05+zhRureVjQ412YA+WVX1baWeztF7Q/3U9YYgFLaah+SmKYi4RjHDSiwOgVRd91Iva6R7DfJvodxfo6Q5PAxgo1i0azXB837fkMLjzELB92eFlywZvbBpqWLlmPuhNqcGGIlkx6LMKYzlE7q5grFqJGXR7H6/8AtVfbJN1Gp+ZlOaEGRAghZB7DBxAhhJBa4AOIEEJILcxbDUhTNQcnxDksKkgZoe14jHMZbNxUx0JxPgLObdDLnmW8h/ZaqtZ8MIytVRUnNhZdCM4LUp8xFo0xezVfxMEcIfSkSkFXwHkq5jjYb9I/bu3NvajQUCYmpuxx4HrEQUN9Bl0nmrbL6nwcFlFAq311mK43jwnLE3uCZPkJwvndHL8LfazXRXbc6nlBDvwAI/CCiwLsJ1XaBHUELIMeq3vJ00txnla53ww82fLczjup8h7MQSvzNAjVfhxPOfjTxarNEZaPwGuV2H4Lle9io23nF/3hccfbNqrj4m8BDmnrjQil4wu8d0Cn0uPJ03Vm/1Megi+evu7YLViGPoR7v7/WU+2l2dvfrLYihBBC9jN8ABFCCKmFeRuCCwpnqpv+nplSno1zjVclECtLwvPX5HRjtUIMaemUQ3zPtq+lxromxNdqi1+NQadIgn172ICNyzblGdgYeenSKpUXc8y9sCDaB6kQEFyjHMsxmBAipt/aoxYqdIPhiKlpCKPB0O10VYkFLLMB4yBP7b4M0BWRsh7JU0x9hf16tveqHZiyHUHYM6xIsfUcjLTlib1WCUQ+UggT6lBaBNc5xtGo7yVIEe5AWFaXQgig+EQQYTkPLMOhrZ1sE/AnIFdhK6wI3MXyDOr6eGM4gTZBqZPm4FDv83EnnmjWRY2WWc6K/sfJMqxirNdjqNK2H118dHi1f7D69+t12V9cWWErBquiGeUF/QXdfobgCCGEzGP4ACKEEFILfAARQgiphfmrAYmKLqr4uWdZDqHGRlLGcqNGRWkDsbHbPcdUJW/RNsaLy+tnN5axBXEj7h/3xbICRQH6i4nBesFcaJOyvQG9Cy2AcpNGjmUG0I4D61Qr2xhMJ/ZErHI9GLRIhvF9XY4BNIfpjtVtunDtUtUXIbQ3wRLQSodDmx6Mj2ub+xDSu1OHaeX9tRvsJwfHDbVuiFolWjCp9SEE7VOwNwqxxILaNY5TTwMK+usIKdaaLrR+BxoolBhHzdFazlSnF+sy3N3M6lBeOW91WKxU3moNmOVGa8gsH3viH5VtaNptu+DrEysdrihsyrlnhaRlQawQAefqSeCmXAnqtFV9jFNV4GKG/X+fMK0ftUz9mxqYeSD9pxWYQ89qK0IIIWQ/wwcQIYSQWuADiBBCSC3MWw2oENcrfavnkmBc1LeGULYeXtnm/vNZRKw1DJaxRf1Ix0pxWziq2S/OJ8oKtM+3lyTU1h2ehUb/sg+I33plgY+d6pUDwL7Q1kJF33V4ZG9eF7RXz/Hy5/JAKYQc+k3rLdAC/IO2IokimN+CeoWKtcdQZgDnlXU6Nv5fqL7Ba+NpcqZ8N2hWOLdEjSfs0wTC+3httYZSFFYvwjGu4/+oC2J5D9MVINO6vLoUu9WpYF2E9jTqM5RMmcbxrzQJLCWNNl3Lly6zx1F6jFcR3bOM0vOAYBXOJ9Tz+rBsC9xLMWqzel4WHCX2jqv0bNQU4ThRiOpsSRduQ7TpMr9nulxERRl58/1ZbUUIIYTsZ/gAIoQQUgvzNgSn0SnQMfhTYPhLOzfjOnSQzdC52fUPq2Hl1dDPN1brsFsrbCkwjCbYJvUqjf9dwFCZdgKHNjioJKvTNr3wEFYb9Rxa+p87hmpMG6ETC0hFTlNdEdVuu3vXBBzHHiavcCePwAVdV+QMoE0NSN1PuzqUAaExOFccmwGGEfU6sGsqsrLN0127LgXX5K6qHBtDinnHcwmHNqtw9hRUmfVsWKL+IUQcAqFJ77ZjLSpsv2D4Ti9luW0T9mFDhUEhKitTkIIexKVlTqsB9xmEU5cfvtIsN9tl6nUIqchFZg9c9buhK6CKQIVXLC+Kvzn9h48kDQireRvP/t1C/y56plw4BQDW2/NV+2EIjhBCyHyGDyBCCCG1wAcQIYSQWpi3GlBeiPy+ikEUaesUtFzvX9UUn65YxTEErSBTQWWvOicu6rhphRazZ1+qiqNX8RFTLVGPUftC+xMvzqrT1bFSIcTwtQ6Cmg+erGfNU65PEhvvT7t2Z11V+gDjwhna4Jjr4Sq3xdTRXGleBfRxBmOk2HvY+ndNsH+IVX5rjhVcwaHIgaao9xU53K9tf6auTxuqc2bTthqs7hssi+BZtHhjU+0F0vozsLbR5SUCh1oApOanWn8EHQq0Mc+CqVumr6PlVQP2pS2AMD0d07BjdX0WDA+bdYcuG7WNiGwbzXADzQftmwI1RpIQU57tmMhMaRacBlJtCxVH2haqQlsSq3NW/UaK2FR9Bz8GWEm2gDGi71k9XSOoErD012e1FSGEELKf4QOIEEJILfABRAghpBbmrQYUhrGaT6NKEKNdSIX1RRe0AAjZe/NbTPlltNb3a2Wr/aBA1L/UbuF5rNv/A+RgZa9DygXoOCGcvC7Di/MC0CJHx/A9e6MAy1jgPKZy3xFqZdCmVFsj+SZFZskU94W4e5FCv4AmEWq9AuUu1I9MC6pLvBeBnnNjt4zhuodQ5jnVmhCcegj6S9Qsv9uF+SwDDTuPSZTO0wHNsAPn6pWI0P2KgxrLJGiLGU83gCapcTydwn5Aj0yhTLUufRKBtoEuPoUrv9uGkgpBZPupMVjqPouWWKudI1a/2Cxj2QHdbajf+bpt+TnFyUkwxrWe7WnHWJ/Bm2ultPAKvXHPvpX+OENJdNNG+DHw51Li/COtH+nvyazgGxAhhJBa4AOIEEJILczbEJy4oBdq0GErfGLGkD6ZqtATph9ChqoXPjKvl/ia7TlNa+ds2LHnmaP2g5nUGFLEtEeVxokpkhjOy5VDtPc671VTVVUc0dHaCyP0twxJMVQGy6aKKFazxVRxtd8cwlACIYccbGSMGzBcO8z2TtW4iOHcUghdJmpfWBjWpNSKPzb18IohDBLCdc71tyHkEzm7bStRKedgPT0FLuFdcLyOVW5yOm3duyMYX7qfAgxRYfhRhRS7aCeFju94u+ixCEMihXspCMowWzeDCwK/BSec8LLe56P+4Cizrt204boc7gFt6eXbf0FILtLheHSL7h/+8rfFEBzc72q8eVUAsE2mQjBYRAUYUpS+YDVkL2wY9nH3xpulD3wDIoQQUgt8ABFCCKkFPoAIIYTUwvzVgALXy6fV9jUO4qRoi5GmHbUOqlvid0G7CSuex2jjo+OmOcThgxz26/RnrIIIaY9e7QNd6hMtZey5a7saF9i4u1/JoYzRBrAWq5Fik2xqL+gtDpfL4xSgBWTQ/lBpWgVUF21genQM+ldWHjcFbSZH7U+1EbWkQEB7UiUJXGF1ECx1kDShH1W/ZuA/001RZyu3bTXsbTnVtf2m4/Ax6IDDTdAQMYVbjZEMUoQzsI2ZUtc583QDtBIqSZrWngmr1/o5usrWKsK0ZdunDXU+zVbbrBtevMQsH37Eqt5n1HimUju+YiybEGid01ohefeLSv9GHQetkYzO7AnCcD+grY+e0gDXHcvDhNqqytuPRU9TwDIoaCWE7l+Znv6gzm12Rjx8AyKEEFITfAARQgipBT6ACCGE1MK81YCcK3qlCXQs1LevqLLJ6B9rxv2KgC1L/0oHe2lrdWldvYTWL+hkg5qQnuuD85b8OHD/uT3esgrSeuUj4HTQAsieEZZEtxqK1uFmunaZsmh5Zvt22K+N2TuYMOLUfKMghjGB86WUtuHgwqIli4Tlfn0dDeeKwVe1jZIDTQ7nh6g2BVjOIIC5SdqWHzSGPOpfvkDEznNCfSWBM9S2OJmgPROce6jnnYCWkeDPDIw3dX1iKB3fhHIfUVwutweHzDqcehKb+Tkw9tAGCr+sNV7UZmDcFhXz4hyceqz6CW3FPE8sQOtHOH+wQHFG66kwhxHnD+rfhgLLeXjTCeHeMuUY9v65Cr4BEUIIqQU+gAghhNQCH0CEEEJqYd5qQEFQSPC7WLIzueZYxrZ/xvlMcUg/Nqq8ljwPKoz7ltvifCP/qP19yjPUizyLfHVMT8OquHzQL5lns650ENQysOSwd+79dR2vdIOKvaNGgsKaPsyunTvNutDz/8dyE+Vyt7B6UQ5ze4x1Pfpx4TQs9blA3QnagCH9PFXjKajup4bSSTod294W6DrG8w/0ItQYEzjOtJpnkxegwUEXR2quD84DwnIlxoxvpjkq9ptmDg5e5xhmlCRh0X9dA8ZpVpaDj0BbwjGP2ofWeTwvNdDOdKl2lJIKGCO5KlWB1m84R7AB2pn+rcP5gr5nm7q/YR4Zon3lsOw8ylIR9IVuszNaknez7/3Ys9qKEEII2c/wAUQIIaQW5m0ILs/zXmhLh9K8Cn2eNY96HfYqlaKFOaxWr7Ep2P976YdqXxG0AVNjdWgPw4LYRK8SqA6HBRjygeOYMM9M6er6i2gHZA+Dr9OmwquX7m3br1/psdIqpo52lC1ODmnLAZx7DGEdp6x5YrQDwqigTs+FPNkM0r11tjSGU7Cj/PFUfo68EIndNlfxLyzJkYONj45lRpgODeM2gOvTVDHGomGPM92B8LDquBjGrS7rgNsG3tyC6vBprFLfIy8MaMdTKyxtbwIYIw1I/77zB//T+7xl25hZ9+IX24qoL3npSWZ5aKhM8cawpldxV1d09SrHQkhLhQ1j+P+/V6fUKwGjN8bfMhyb5XIGqfjNVqvvrnIcw2jjU2WjZMJxsyuJyjcgQgghtcAHECGEkFqYdyG437+67Z6Y7P3NhuDs9hha0pUMvXBdPLsqfSIiWT5TCK6/NUJ1CK7CUkFEMgzB6ddwfPevDMFBlpjnsKA3rQ4l4Xd1FpYXgoNMMR2uC2E2PWZSdSZL1+EpqNbpIDMs74CztnLa7mRQFbQLVVr1+YIpNYZAg1yHFdDB2lIddbDXCiO8+hpgf6NzdoYxRb0bGLfocK0zqTDDKYVQX6ay5IoATw6ctFWYyks+nSGDMXI6CwsqBGN1WNUXmPkVQLZXV11LvK+64II+NTVtlrVTRRjBmIAT1GcXJdZJA8PZ+psxhLfwxxirPUMDYbkifTPH0Cpk7eosOHQriVAyQOeQso163E7+7vd7plBc4GYbrHue+O1vfysrV66suxmEEEL2kS1btsjhhx/ed/28ewAVRSGPP/64OOdk1apVsmXLFlmwYEHdzZq3jI+Py8qVK9lPM8B+mh3sp9nBfqrGOSe7du2SFStWeIlimnkXggvDUA4//HAZHx8XEZEFCxbwAs8C9tPsYD/NDvbT7GA/9WdkZGTGbZiEQAghpBb4ACKEEFIL8/YB1Gw25Z//+Z+l2WzOvPFBDPtpdrCfZgf7aXawn/YP8y4JgRBCyMHBvH0DIoQQ8sKGDyBCCCG1wAcQIYSQWuADiBBCSC3wAUQIIaQW5u0D6LrrrpMjjzxSWq2WnH766XL33XfX3aTaWL9+vZx66qkyPDwsS5culXPOOUc2bdpktpmenpZ169bJ4sWLZWhoSM4991wZGxvrs8eDg6uvvlqCIJCLLrqo9zf20x4ee+wxeeMb3yiLFy+WdrstJ554otx777299c45ueKKK2T58uXSbrdl7dq18vDDD9fY4uefPM/l8ssvl9WrV0u73ZajjjpKPvjBD3p1bw72fton3Dzk5ptvdo1Gw332s591P/vZz9w//MM/uIULF7qxsbG6m1YLZ555prvhhhvcAw884O677z73F3/xF27VqlVu9+7dvW3e/va3u5UrV7oNGza4e++917385S93r3jFK2psdb3cfffd7sgjj3QveclL3IUXXtj7O/vJue3bt7sjjjjCvelNb3J33XWXe+SRR9y3v/1t98tf/rK3zdVXX+1GRkbcV77yFXf//fe7v/qrv3KrV692U1NTNbb8+eWqq65yixcvdt/4xjfc5s2b3S233OKGhobcv//7v/e2YT/tG/PyAXTaaae5devW9ZbzPHcrVqxw69evr7FV84dt27Y5EXG33Xabc865HTt2uCRJ3C233NLb5uc//7kTEXfHHXfU1cza2LVrlzv66KPdd77zHfcnf/InvQcQ+2kP73vf+9wrX/nKvuuLonDLli1z//Zv/9b7244dO1yz2XRf/OIXn48mzgvOOuss95a3vMX87XWve50777zznHPsp/3BvAvBdbtd2bhxo6xdu7b3tzAMZe3atXLHHXfU2LL5w86dO0VEZNGiRSIisnHjRknT1PTZMcccI6tWrToo+2zdunVy1llnmf4QYT/9nq997WtyyimnyOtf/3pZunSpnHTSSfLpT3+6t37z5s2ydetW008jIyNy+umnH1T99IpXvEI2bNggDz30kIiI3H///XL77bfLa17zGhFhP+0P5p0b9lNPPSV5nsvo6Kj5++joqPziF7+oqVXzh6Io5KKLLpIzzjhDTjjhBBER2bp1qzQaDVm4cKHZdnR0VLZu3VpDK+vj5ptvlh/96Edyzz33eOvYT3t45JFH5Prrr5dLLrlE3v/+98s999wj7373u6XRaMj555/f64u93YMHUz9deumlMj4+Lsccc4xEUSR5nstVV10l5513nogI+2k/MO8eQKSadevWyQMPPCC333573U2Zd2zZskUuvPBC+c53viOtVqvu5sxbiqKQU045RT70oQ+JiMhJJ50kDzzwgHzyk5+U888/v+bWzR++9KUvyRe+8AW56aab5Pjjj5f77rtPLrroIlmxYgX7aT8x70JwS5YskSiKvMyksbExWbZsWU2tmh9ccMEF8o1vfEO+973vmSqDy5Ytk263Kzt27DDbH2x9tnHjRtm2bZu87GUvkziOJY5jue222+Taa6+VOI5ldHSU/SQiy5cvl+OOO8787dhjj5VHH31URKTXFwf7Pfie97xHLr30UnnDG94gJ554ovzd3/2dXHzxxbJ+/XoRYT/tD+bdA6jRaMjJJ58sGzZs6P2tKArZsGGDrFmzpsaW1YdzTi644AL58pe/LN/97ndl9erVZv3JJ58sSZKYPtu0aZM8+uijB1WfvfrVr5af/vSnct999/X+nXLKKXLeeef1PrOfRM444wwvjf+hhx6SI444QkREVq9eLcuWLTP9ND4+LnfddddB1U+Tk5NeNc8oiqQoChFhP+0X6s6C2Bs333yzazab7nOf+5x78MEH3dve9ja3cOFCt3Xr1rqbVgvveMc73MjIiPv+97/vnnjiid6/ycnJ3jZvf/vb3apVq9x3v/tdd++997o1a9a4NWvW1Njq+YHOgnOO/eTcnhT1OI7dVVdd5R5++GH3hS98wQ0MDLj//M//7G1z9dVXu4ULF7qvfvWr7ic/+Yk7++yzD7r04vPPP98ddthhvTTs//7v/3ZLlixx733ve3vbsJ/2jXn5AHLOuY9//ONu1apVrtFouNNOO83deeeddTepNkRkr/9uuOGG3jZTU1Pune98pzvkkEPcwMCA++u//mv3xBNP1NfoeQI+gNhPe/j617/uTjjhBNdsNt0xxxzjPvWpT5n1RVG4yy+/3I2Ojrpms+le/epXu02bNtXU2noYHx93F154oVu1apVrtVruRS96kfunf/on1+l0etuwn/YN1gMihBBSC/NOAyKEEHJwwAcQIYSQWuADiBBCSC3wAUQIIaQW+AAihBBSC3wAEUIIqQU+gAghhNQCH0CEEEJqgQ8gQgghtcAHECGEkFrgA4gQQkgt/P/e6azZZs0/PwAAAABJRU5ErkJggg==",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.imshow(samp[0])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x1f8817b1550>"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.imshow(samp[1])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.0"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "samp[2]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "420"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "round(len(data)*.7)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Training partition\n",
    "train_data = data.take(round(len(data)*.7)) #send 70% data for train\n",
    "train_data = train_data.batch(16) #send data in batch of 16 images\n",
    "train_data = train_data.prefetch(8) "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<_PrefetchDataset element_spec=(TensorSpec(shape=(None, 100, 100, None), dtype=tf.float32, name=None), TensorSpec(shape=(None, 100, 100, None), dtype=tf.float32, name=None), TensorSpec(shape=(None,), dtype=tf.float32, name=None))>"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [],
   "source": [
    "train_samples = train_data.as_numpy_iterator()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [],
   "source": [
    "train_sample = train_samples.next()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "16"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(train_sample[0])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "180"
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "round(len(data)*.3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<_SkipDataset element_spec=(TensorSpec(shape=(100, 100, None), dtype=tf.float32, name=None), TensorSpec(shape=(100, 100, None), dtype=tf.float32, name=None), TensorSpec(shape=(), dtype=tf.float32, name=None))>\n",
      "<_BatchDataset element_spec=(TensorSpec(shape=(None, 100, 100, None), dtype=tf.float32, name=None), TensorSpec(shape=(None, 100, 100, None), dtype=tf.float32, name=None), TensorSpec(shape=(None,), dtype=tf.float32, name=None))>\n",
      "<_PrefetchDataset element_spec=(TensorSpec(shape=(None, 100, 100, None), dtype=tf.float32, name=None), TensorSpec(shape=(None, 100, 100, None), dtype=tf.float32, name=None), TensorSpec(shape=(None,), dtype=tf.float32, name=None))>\n"
     ]
    }
   ],
   "source": [
    "# Testing partition\n",
    "test_data = data.skip(round(len(data)*.7))\n",
    "print(test_data)\n",
    "test_data = test_data.take(round(len(data)*.3))\n",
    "\n",
    "test_data = test_data.batch(16)\n",
    "print(test_data)\n",
    "test_data = test_data.prefetch(8)\n",
    "print(test_data)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 4 Model Engineering"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 4.1 Build Embedding Layer"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [],
   "source": [
    "#1 Input layer \n",
    "inp = Input(shape =(100,100,3),name='input_image')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<KerasTensor: shape=(None, 100, 100, 3) dtype=float32 (created by layer 'input_image')>"
      ]
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "inp"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [],
   "source": [
    "#1 convulated layer with activation function 'relu'\n",
    "c1 = Conv2D(64, (10,10), activation='relu')(inp)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<KerasTensor: shape=(None, 91, 91, 64) dtype=float32 (created by layer 'conv2d')>"
      ]
     },
     "execution_count": 46,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "c1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [],
   "source": [
    "#1 maxpooling layer\n",
    "m1 = MaxPooling2D(64, (2,2), padding='same')(c1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<KerasTensor: shape=(None, 46, 46, 64) dtype=float32 (created by layer 'max_pooling2d')>"
      ]
     },
     "execution_count": 48,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "m1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [],
   "source": [
    "#second block\n",
    "c2 = Conv2D(128,(7,7),activation='relu')(m1)\n",
    "m2 = MaxPooling2D(64, (2,2), padding='same')(c2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<KerasTensor: shape=(None, 40, 40, 128) dtype=float32 (created by layer 'conv2d_1')>"
      ]
     },
     "execution_count": 50,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "c2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<KerasTensor: shape=(None, 20, 20, 128) dtype=float32 (created by layer 'max_pooling2d_1')>"
      ]
     },
     "execution_count": 51,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "m2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [],
   "source": [
    "#third layer\n",
    "c3 = Conv2D(128,(4,4),activation='relu')(m2)\n",
    "m3 = MaxPooling2D(64, (2,2), padding='same')(c3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(<KerasTensor: shape=(None, 17, 17, 128) dtype=float32 (created by layer 'conv2d_2')>,\n",
       " <KerasTensor: shape=(None, 9, 9, 128) dtype=float32 (created by layer 'max_pooling2d_2')>)"
      ]
     },
     "execution_count": 53,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "c3,m3"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [],
   "source": [
    "c4 = Conv2D(256, (4,4), activation='relu')(m3)\n",
    "f1 = Flatten()(c4)\n",
    "d1 = Dense(4096, activation='sigmoid')(f1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(<KerasTensor: shape=(None, 6, 6, 256) dtype=float32 (created by layer 'conv2d_3')>,\n",
       " <KerasTensor: shape=(None, 9216) dtype=float32 (created by layer 'flatten')>,\n",
       " <KerasTensor: shape=(None, 4096) dtype=float32 (created by layer 'dense')>)"
      ]
     },
     "execution_count": 55,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "c4,f1,d1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"embedding\"\n",
      "_________________________________________________________________\n",
      " Layer (type)                Output Shape              Param #   \n",
      "=================================================================\n",
      " input_image (InputLayer)    [(None, 100, 100, 3)]     0         \n",
      "                                                                 \n",
      " conv2d (Conv2D)             (None, 91, 91, 64)        19264     \n",
      "                                                                 \n",
      " max_pooling2d (MaxPooling2  (None, 46, 46, 64)        0         \n",
      " D)                                                              \n",
      "                                                                 \n",
      " conv2d_1 (Conv2D)           (None, 40, 40, 128)       401536    \n",
      "                                                                 \n",
      " max_pooling2d_1 (MaxPoolin  (None, 20, 20, 128)       0         \n",
      " g2D)                                                            \n",
      "                                                                 \n",
      " conv2d_2 (Conv2D)           (None, 17, 17, 128)       262272    \n",
      "                                                                 \n",
      " max_pooling2d_2 (MaxPoolin  (None, 9, 9, 128)         0         \n",
      " g2D)                                                            \n",
      "                                                                 \n",
      " conv2d_3 (Conv2D)           (None, 6, 6, 256)         524544    \n",
      "                                                                 \n",
      " flatten (Flatten)           (None, 9216)              0         \n",
      "                                                                 \n",
      " dense (Dense)               (None, 4096)              37752832  \n",
      "                                                                 \n",
      "=================================================================\n",
      "Total params: 38960448 (148.62 MB)\n",
      "Trainable params: 38960448 (148.62 MB)\n",
      "Non-trainable params: 0 (0.00 Byte)\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "Model(inputs=[inp],outputs=[d1],name='embedding').summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {},
   "outputs": [],
   "source": [
    "def make_embedding():\n",
    "    inp = Input(shape =(100,100,3),name='input_image')\n",
    "    # first layer\n",
    "    c1 = Conv2D(64,(10,10),activation='relu')(inp)\n",
    "    m1 = MaxPooling2D(64, (2,2), padding='same')(c1)\n",
    "    # second layer\n",
    "    c2 = Conv2D(128,(7,7),activation='relu')(m1)\n",
    "    m2 = MaxPooling2D(64, (2,2), padding='same')(c2)\n",
    "    #third layer\n",
    "    c3 = Conv2D(128,(4,4),activation='relu')(m2)\n",
    "    m3 = MaxPooling2D(64, (2,2), padding='same')(c3)\n",
    "    # fourth layer\n",
    "    c4 = Conv2D(256, (4,4), activation='relu')(m3)\n",
    "    f1 = Flatten()(c4)\n",
    "    d1 = Dense(4096, activation='sigmoid')(f1)\n",
    "    \n",
    "    return Model(inputs=[inp],outputs=[d1],name='embedding')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {},
   "outputs": [],
   "source": [
    "embedding = make_embedding()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"embedding\"\n",
      "_________________________________________________________________\n",
      " Layer (type)                Output Shape              Param #   \n",
      "=================================================================\n",
      " input_image (InputLayer)    [(None, 100, 100, 3)]     0         \n",
      "                                                                 \n",
      " conv2d_4 (Conv2D)           (None, 91, 91, 64)        19264     \n",
      "                                                                 \n",
      " max_pooling2d_3 (MaxPoolin  (None, 46, 46, 64)        0         \n",
      " g2D)                                                            \n",
      "                                                                 \n",
      " conv2d_5 (Conv2D)           (None, 40, 40, 128)       401536    \n",
      "                                                                 \n",
      " max_pooling2d_4 (MaxPoolin  (None, 20, 20, 128)       0         \n",
      " g2D)                                                            \n",
      "                                                                 \n",
      " conv2d_6 (Conv2D)           (None, 17, 17, 128)       262272    \n",
      "                                                                 \n",
      " max_pooling2d_5 (MaxPoolin  (None, 9, 9, 128)         0         \n",
      " g2D)                                                            \n",
      "                                                                 \n",
      " conv2d_7 (Conv2D)           (None, 6, 6, 256)         524544    \n",
      "                                                                 \n",
      " flatten_1 (Flatten)         (None, 9216)              0         \n",
      "                                                                 \n",
      " dense_1 (Dense)             (None, 4096)              37752832  \n",
      "                                                                 \n",
      "=================================================================\n",
      "Total params: 38960448 (148.62 MB)\n",
      "Trainable params: 38960448 (148.62 MB)\n",
      "Non-trainable params: 0 (0.00 Byte)\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "embedding.summary()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 4.2 Build Distance Layer"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {},
   "outputs": [],
   "source": [
    "#L1 distance class\n",
    "class LDist(Layer):\n",
    "    # init method - inheritance\n",
    "    def __init__(self,**kwargs):\n",
    "        super().__init__()\n",
    "    #similarity calculation\n",
    "    def call(self,input_embedding,validation_embedding):\n",
    "        return tf.math.abs(input_embedding-validation_embedding)\n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {},
   "outputs": [],
   "source": [
    "l1 = LDist()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<__main__.LDist at 0x1f8816ab910>"
      ]
     },
     "execution_count": 62,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "l1"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 4.3 Make Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {},
   "outputs": [],
   "source": [
    "input_image = Input(name='input_img',shape=(100,100,3))\n",
    "validation_img  = Input(name='validation_img',shape=(100,100,3))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {},
   "outputs": [],
   "source": [
    "inp_emb = embedding(input_image)\n",
    "val_emb = embedding(validation_img)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<KerasTensor: shape=(None, 4096) dtype=float32 (created by layer 'embedding')>"
      ]
     },
     "execution_count": 65,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "val_emb"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {},
   "outputs": [],
   "source": [
    "siamese_layer = LDist()\n",
    "siamese_layer._name = 'distance'\n",
    "distances = siamese_layer(embedding(input_image), embedding(validation_img))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {},
   "outputs": [],
   "source": [
    "siamese_layer = LDist()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<KerasTensor: shape=(None, 4096) dtype=float32 (created by layer 'l_dist_2')>"
      ]
     },
     "execution_count": 68,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "siamese_layer(inp_emb,val_emb)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "metadata": {},
   "outputs": [],
   "source": [
    "classifier = Dense(1, activation='sigmoid')(distances)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<KerasTensor: shape=(None, 1) dtype=float32 (created by layer 'dense_2')>"
      ]
     },
     "execution_count": 70,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "classifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"siameseNetwork\"\n",
      "__________________________________________________________________________________________________\n",
      " Layer (type)                Output Shape                 Param #   Connected to                  \n",
      "==================================================================================================\n",
      " input_img (InputLayer)      [(None, 100, 100, 3)]        0         []                            \n",
      "                                                                                                  \n",
      " validation_img (InputLayer  [(None, 100, 100, 3)]        0         []                            \n",
      " )                                                                                                \n",
      "                                                                                                  \n",
      " embedding (Functional)      (None, 4096)                 3896044   ['input_img[0][0]',           \n",
      "                                                          8          'validation_img[0][0]']      \n",
      "                                                                                                  \n",
      " distance (LDist)            (None, 4096)                 0         ['embedding[2][0]',           \n",
      "                                                                     'embedding[3][0]']           \n",
      "                                                                                                  \n",
      " dense_2 (Dense)             (None, 1)                    4097      ['distance[0][0]']            \n",
      "                                                                                                  \n",
      "==================================================================================================\n",
      "Total params: 38964545 (148.64 MB)\n",
      "Trainable params: 38964545 (148.64 MB)\n",
      "Non-trainable params: 0 (0.00 Byte)\n",
      "__________________________________________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "Model(inputs=[input_image, validation_img], outputs=classifier, name='siameseNetwork').summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "metadata": {},
   "outputs": [],
   "source": [
    "def make_model():\n",
    "    #handle inputs\n",
    "    input_image = Input(name='input_img',shape=(100,100,3))\n",
    "\n",
    "    # validation img\n",
    "    validation_img  = Input(name='validation_img',shape=(100,100,3))\n",
    "     # Combine siamese distance components \n",
    "    siamese_layer = LDist()\n",
    "    siamese_layer._name = 'distance'\n",
    "    distances = siamese_layer(embedding(input_image), embedding(validation_img))\n",
    "    # Classification Layer\n",
    "    classifier = Dense(1, activation='sigmoid')(distances)\n",
    "    \n",
    "    return Model(inputs=[input_image, validation_img], outputs=classifier, name='siameseNetwork')\n",
    "    \n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "metadata": {},
   "outputs": [],
   "source": [
    "siamese_model = make_model()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"siameseNetwork\"\n",
      "__________________________________________________________________________________________________\n",
      " Layer (type)                Output Shape                 Param #   Connected to                  \n",
      "==================================================================================================\n",
      " input_img (InputLayer)      [(None, 100, 100, 3)]        0         []                            \n",
      "                                                                                                  \n",
      " validation_img (InputLayer  [(None, 100, 100, 3)]        0         []                            \n",
      " )                                                                                                \n",
      "                                                                                                  \n",
      " embedding (Functional)      (None, 4096)                 3896044   ['input_img[0][0]',           \n",
      "                                                          8          'validation_img[0][0]']      \n",
      "                                                                                                  \n",
      " distance (LDist)            (None, 4096)                 0         ['embedding[4][0]',           \n",
      "                                                                     'embedding[5][0]']           \n",
      "                                                                                                  \n",
      " dense_3 (Dense)             (None, 1)                    4097      ['distance[0][0]']            \n",
      "                                                                                                  \n",
      "==================================================================================================\n",
      "Total params: 38964545 (148.64 MB)\n",
      "Trainable params: 38964545 (148.64 MB)\n",
      "Non-trainable params: 0 (0.00 Byte)\n",
      "__________________________________________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "siamese_model.summary()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 5 Training "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 5.1 Setup Loss and Optimizer"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "metadata": {},
   "outputs": [],
   "source": [
    "#loss function\n",
    "binary_cross_loss = tf.losses.BinaryCrossentropy()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "metadata": {},
   "outputs": [],
   "source": [
    "#loss function optimizer with learning rate 0.0001\n",
    "opt = tf.keras.optimizers.Adam(1e-4) #0.0001"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 5.2 Establish Checkpoints"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "metadata": {},
   "outputs": [],
   "source": [
    "#1 checkpoint directory save checkpoint\n",
    "#2 checkoint prefix\n",
    "#3 checkpoint class\n",
    "checkpoint_dir = './training_checkpoints'\n",
    "checkpoint_prefix = os.path.join(checkpoint_dir, 'ckpt')\n",
    "checkpoint = tf.train.Checkpoint(opt=opt, siamese_model=siamese_model)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 5.3 Build Train Step Function"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "metadata": {},
   "outputs": [],
   "source": [
    "test_batch = train_data.as_numpy_iterator()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "metadata": {},
   "outputs": [],
   "source": [
    "batch_1 = test_batch.next()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 80,
   "metadata": {},
   "outputs": [],
   "source": [
    "X = batch_1[:2]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 81,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(2, 16, 100, 100, 3)"
      ]
     },
     "execution_count": 81,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.array(X).shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 82,
   "metadata": {},
   "outputs": [],
   "source": [
    "y = batch_1[2]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 83,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0., 1., 1., 0., 0., 1., 0., 0., 1., 1., 0., 1., 1., 0., 0., 0.],\n",
       "      dtype=float32)"
      ]
     },
     "execution_count": 83,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 84,
   "metadata": {},
   "outputs": [],
   "source": [
    "#helps to compiles a function into callable tensorflow graph\n",
    "@tf.function\n",
    "def train_step(batch):\n",
    "    with tf.GradientTape() as tape: # start capturing gradient from nn\n",
    "        #Get anchor and positive/negative image\n",
    "        X = batch[:2]\n",
    "        #Get Label\n",
    "        y = batch[2]\n",
    "        # Forward pass\n",
    "        yhat = siamese_model(X, training=True)\n",
    "        # Calculate loss\n",
    "        loss = binary_cross_loss(y,yhat)\n",
    "    print(loss)\n",
    "        \n",
    "    # Calculate gradients\n",
    "    grad = tape.gradient(loss,siamese_model.trainable_variables)\n",
    "    \n",
    "    # Calculate updated weights and apply to siamese model\n",
    "    opt.apply_gradients(zip(grad,siamese_model.trainable_variables))\n",
    "    \n",
    "    # Return loss\n",
    "    return loss"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 5.4 Build Training Loop"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 85,
   "metadata": {},
   "outputs": [],
   "source": [
    "def train(data, EPOCHS):\n",
    "    # loop through epochs\n",
    "    for epoch in range(1, EPOCHS+1):\n",
    "        print('\\n EPOCH {}/{}'.format(epoch, EPOCHS))\n",
    "        progbar = tf.keras.utils.Progbar(len(data))\n",
    "        \n",
    "    # Loop through each batch\n",
    "    for idx, batch in enumerate(data):\n",
    "        # Run train step here\n",
    "        train_step(batch)\n",
    "        progbar.update(idx+1)\n",
    "        \n",
    "    # Save checkpoints\n",
    "    if epoch % 10 == 0:\n",
    "        checkpoint.save(file_prefix=checkpoint_prefix)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 5.5 Train the model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 86,
   "metadata": {},
   "outputs": [],
   "source": [
    "EPOCHS = 50"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 87,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      " EPOCH 1/50\n",
      "\n",
      " EPOCH 2/50\n",
      "\n",
      " EPOCH 3/50\n",
      "\n",
      " EPOCH 4/50\n",
      "\n",
      " EPOCH 5/50\n",
      "\n",
      " EPOCH 6/50\n",
      "\n",
      " EPOCH 7/50\n",
      "\n",
      " EPOCH 8/50\n",
      "\n",
      " EPOCH 9/50\n",
      "\n",
      " EPOCH 10/50\n",
      "\n",
      " EPOCH 11/50\n",
      "\n",
      " EPOCH 12/50\n",
      "\n",
      " EPOCH 13/50\n",
      "\n",
      " EPOCH 14/50\n",
      "\n",
      " EPOCH 15/50\n",
      "\n",
      " EPOCH 16/50\n",
      "\n",
      " EPOCH 17/50\n",
      "\n",
      " EPOCH 18/50\n",
      "\n",
      " EPOCH 19/50\n",
      "\n",
      " EPOCH 20/50\n",
      "\n",
      " EPOCH 21/50\n",
      "\n",
      " EPOCH 22/50\n",
      "\n",
      " EPOCH 23/50\n",
      "\n",
      " EPOCH 24/50\n",
      "\n",
      " EPOCH 25/50\n",
      "\n",
      " EPOCH 26/50\n",
      "\n",
      " EPOCH 27/50\n",
      "\n",
      " EPOCH 28/50\n",
      "\n",
      " EPOCH 29/50\n",
      "\n",
      " EPOCH 30/50\n",
      "\n",
      " EPOCH 31/50\n",
      "\n",
      " EPOCH 32/50\n",
      "\n",
      " EPOCH 33/50\n",
      "\n",
      " EPOCH 34/50\n",
      "\n",
      " EPOCH 35/50\n",
      "\n",
      " EPOCH 36/50\n",
      "\n",
      " EPOCH 37/50\n",
      "\n",
      " EPOCH 38/50\n",
      "\n",
      " EPOCH 39/50\n",
      "\n",
      " EPOCH 40/50\n",
      "\n",
      " EPOCH 41/50\n",
      "\n",
      " EPOCH 42/50\n",
      "\n",
      " EPOCH 43/50\n",
      "\n",
      " EPOCH 44/50\n",
      "\n",
      " EPOCH 45/50\n",
      "\n",
      " EPOCH 46/50\n",
      "\n",
      " EPOCH 47/50\n",
      "\n",
      " EPOCH 48/50\n",
      "\n",
      " EPOCH 49/50\n",
      "\n",
      " EPOCH 50/50\n",
      "Tensor(\"binary_crossentropy/weighted_loss/value:0\", shape=(), dtype=float32)\n",
      "Tensor(\"binary_crossentropy/weighted_loss/value:0\", shape=(), dtype=float32)\n",
      "26/27 [===========================>..] - ETA: 16sTensor(\"binary_crossentropy/weighted_loss/value:0\", shape=(), dtype=float32)\n",
      "27/27 [==============================] - 429s 16s/step\n"
     ]
    }
   ],
   "source": [
    "train(train_data, EPOCHS)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 6 Evaluate Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 88,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Import metric calculations\n",
    "from tensorflow.keras.metrics import Precision, Recall"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 101,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Get a batch of test data \n",
    "test_input, test_val, y_true = test_data.as_numpy_iterator().next()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 102,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 1s 1s/step\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "array([[1.2091596e-01],\n",
       "       [4.8400036e-01],\n",
       "       [4.9312246e-01],\n",
       "       [8.7296503e-04],\n",
       "       [1.6067137e-06],\n",
       "       [2.2896100e-07],\n",
       "       [1.3093663e-05],\n",
       "       [3.6808981e-07],\n",
       "       [3.4243681e-07],\n",
       "       [5.0528467e-01],\n",
       "       [4.8527291e-01],\n",
       "       [5.1616549e-01],\n",
       "       [3.4193627e-06],\n",
       "       [4.9473345e-01],\n",
       "       [1.5742702e-07],\n",
       "       [4.9547794e-01]], dtype=float32)"
      ]
     },
     "execution_count": 102,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#make prediction\n",
    "y_hat = siamese_model.predict([test_input, test_val])\n",
    "y_hat"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 103,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 1, 0, 0, 0, 0]"
      ]
     },
     "execution_count": 103,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Post processing the results\n",
    "[1 if prediction > 0.5 else 0 for prediction in y_hat]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 104,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0., 1., 1., 0., 0., 0., 0., 0., 0., 1., 1., 1., 0., 1., 0., 1.],\n",
       "      dtype=float32)"
      ]
     },
     "execution_count": 104,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_true"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 100,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.33333334"
      ]
     },
     "execution_count": 100,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Creating a metric object\n",
    "m = Recall()\n",
    "# calculating the recall value \n",
    "m.update_state(y_true, y_hat)\n",
    "# Return Recall Result\n",
    "m.result().numpy()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 94,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1.0"
      ]
     },
     "execution_count": 94,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Creating a metric object\n",
    "m = Precision()\n",
    "\n",
    "# calculating the recall value \n",
    "m.update_state(y_true, y_hat)\n",
    "\n",
    "# Return Recall Result\n",
    "m.result().numpy() \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 122,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x800 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Set plot size\n",
    "plt.figure(figsize=(10,8))\n",
    "\n",
    "# Set first subplot\n",
    "plt.subplot(1,2,1)\n",
    "plt.imshow(test_input[1])\n",
    "\n",
    "# Set second subplot\n",
    "plt.subplot(1,2,2)\n",
    "plt.imshow(test_val[1])\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 7 Save the Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 123,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "WARNING:tensorflow:Compiled the loaded model, but the compiled metrics have yet to be built. `model.compile_metrics` will be empty until you train or evaluate the model.\n"
     ]
    }
   ],
   "source": [
    "# Save weights\n",
    "siamese_model.save('siamesemodel.h5')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 124,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "WARNING:tensorflow:No training configuration found in the save file, so the model was *not* compiled. Compile it manually.\n"
     ]
    }
   ],
   "source": [
    "# Reload model\n",
    "model1 = tf.keras.models.load_model('siamesemodel.h5', \n",
    "                                   custom_objects={'LDist':LDist, 'BinaryCrossentropy':tf.losses.BinaryCrossentropy})\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 125,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 1s 1s/step\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "array([[9.8987005e-09],\n",
       "       [1.5189198e-07],\n",
       "       [1.1183146e-06],\n",
       "       [4.4452572e-06],\n",
       "       [3.4066588e-06],\n",
       "       [9.9265015e-01],\n",
       "       [9.9154902e-01],\n",
       "       [7.1031791e-01],\n",
       "       [9.9809432e-01],\n",
       "       [9.9968541e-01],\n",
       "       [7.1759480e-01],\n",
       "       [3.7953669e-07],\n",
       "       [6.9857620e-08],\n",
       "       [8.5488641e-01],\n",
       "       [1.7036215e-05],\n",
       "       [5.5657375e-01]], dtype=float32)"
      ]
     },
     "execution_count": 125,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Make predictions with reloaded model\n",
    "model1.predict([test_input, test_val])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 126,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"siameseNetwork\"\n",
      "__________________________________________________________________________________________________\n",
      " Layer (type)                Output Shape                 Param #   Connected to                  \n",
      "==================================================================================================\n",
      " input_img (InputLayer)      [(None, 100, 100, 3)]        0         []                            \n",
      "                                                                                                  \n",
      " validation_img (InputLayer  [(None, 100, 100, 3)]        0         []                            \n",
      " )                                                                                                \n",
      "                                                                                                  \n",
      " embedding (Functional)      (None, 4096)                 3896044   ['input_img[0][0]',           \n",
      "                                                          8          'validation_img[0][0]']      \n",
      "                                                                                                  \n",
      " l_dist_5 (LDist)            (None, 4096)                 0         ['embedding[0][0]',           \n",
      "                                                                     'embedding[1][0]']           \n",
      "                                                                                                  \n",
      " dense_3 (Dense)             (None, 1)                    4097      ['l_dist_5[0][0]']            \n",
      "                                                                                                  \n",
      "==================================================================================================\n",
      "Total params: 38964545 (148.64 MB)\n",
      "Trainable params: 38964545 (148.64 MB)\n",
      "Non-trainable params: 0 (0.00 Byte)\n",
      "__________________________________________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "model1.summary()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Real Time Test"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 8.1 Verification Function"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 127,
   "metadata": {},
   "outputs": [],
   "source": [
    "def verify(model1,detection_threshold,verification_threshold):\n",
    "    #Build results array\n",
    "    results = []\n",
    "    for image in os.listdir(os.path.join('application_data','verification_images')):\n",
    "        input_img = preprocess(os.path.join('application_data','input_image','input_image.jpg'))\n",
    "        validation_img = preprocess(os.path.join('application_data','verification_images',image))\n",
    "\n",
    "        result = model1.predict(list(np.expand_dims([input_img,validation_img], axis=1)))\n",
    "        results.append(result)\n",
    "    detection = np.sum(np.array(results)>detection_threshold)\n",
    "    verification = detection / len(os.listdir(os.path.join('application_data','verification_images')))\n",
    "    verified = verification > verification_threshold\n",
    "\n",
    "    return results, verified\n",
    "    #Detection Threshold:Metric above which a prediction is considered positive\n",
    "    #Verification Threshold: Proportion of positive predictions/ total positive samples"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 8.2 OpenCV Real Time Verification"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 135,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 212ms/step\n",
      "1/1 [==============================] - 0s 277ms/step\n",
      "1/1 [==============================] - 0s 219ms/step\n",
      "1/1 [==============================] - 0s 219ms/step\n",
      "1/1 [==============================] - 0s 215ms/step\n",
      "1/1 [==============================] - 0s 200ms/step\n",
      "1/1 [==============================] - 0s 217ms/step\n",
      "1/1 [==============================] - 0s 203ms/step\n",
      "1/1 [==============================] - 0s 210ms/step\n",
      "1/1 [==============================] - 0s 239ms/step\n",
      "1/1 [==============================] - 0s 230ms/step\n",
      "1/1 [==============================] - 0s 231ms/step\n",
      "1/1 [==============================] - 0s 220ms/step\n",
      "1/1 [==============================] - 0s 223ms/step\n",
      "1/1 [==============================] - 0s 223ms/step\n",
      "1/1 [==============================] - 0s 233ms/step\n",
      "1/1 [==============================] - 0s 206ms/step\n",
      "1/1 [==============================] - 0s 221ms/step\n",
      "1/1 [==============================] - 0s 223ms/step\n",
      "1/1 [==============================] - 0s 224ms/step\n",
      "1/1 [==============================] - 0s 252ms/step\n",
      "1/1 [==============================] - 0s 213ms/step\n",
      "1/1 [==============================] - 0s 213ms/step\n",
      "1/1 [==============================] - 0s 215ms/step\n",
      "1/1 [==============================] - 0s 211ms/step\n",
      "1/1 [==============================] - 0s 214ms/step\n",
      "1/1 [==============================] - 0s 225ms/step\n",
      "1/1 [==============================] - 0s 209ms/step\n",
      "1/1 [==============================] - 0s 215ms/step\n",
      "1/1 [==============================] - 0s 221ms/step\n",
      "1/1 [==============================] - 0s 218ms/step\n",
      "1/1 [==============================] - 0s 223ms/step\n",
      "1/1 [==============================] - 0s 222ms/step\n",
      "1/1 [==============================] - 0s 217ms/step\n",
      "1/1 [==============================] - 0s 215ms/step\n",
      "1/1 [==============================] - 0s 221ms/step\n",
      "1/1 [==============================] - 0s 225ms/step\n",
      "1/1 [==============================] - 0s 213ms/step\n",
      "1/1 [==============================] - 0s 230ms/step\n",
      "1/1 [==============================] - 0s 222ms/step\n",
      "1/1 [==============================] - 0s 217ms/step\n",
      "1/1 [==============================] - 0s 219ms/step\n",
      "1/1 [==============================] - 0s 224ms/step\n",
      "1/1 [==============================] - 0s 223ms/step\n",
      "1/1 [==============================] - 0s 219ms/step\n",
      "1/1 [==============================] - 0s 219ms/step\n",
      "1/1 [==============================] - 0s 219ms/step\n",
      "1/1 [==============================] - 0s 223ms/step\n",
      "1/1 [==============================] - 0s 220ms/step\n",
      "1/1 [==============================] - 0s 215ms/step\n",
      "1/1 [==============================] - 0s 228ms/step\n",
      "1/1 [==============================] - 0s 217ms/step\n",
      "True\n",
      "1/1 [==============================] - 0s 211ms/step\n",
      "1/1 [==============================] - 0s 214ms/step\n",
      "1/1 [==============================] - 0s 206ms/step\n",
      "1/1 [==============================] - 0s 220ms/step\n",
      "1/1 [==============================] - 0s 219ms/step\n",
      "1/1 [==============================] - 0s 219ms/step\n",
      "1/1 [==============================] - 0s 208ms/step\n",
      "1/1 [==============================] - 0s 219ms/step\n",
      "1/1 [==============================] - 0s 206ms/step\n",
      "1/1 [==============================] - 0s 226ms/step\n",
      "1/1 [==============================] - 0s 231ms/step\n",
      "1/1 [==============================] - 0s 220ms/step\n",
      "1/1 [==============================] - 0s 223ms/step\n",
      "1/1 [==============================] - 0s 230ms/step\n",
      "1/1 [==============================] - 0s 214ms/step\n",
      "1/1 [==============================] - 0s 216ms/step\n",
      "1/1 [==============================] - 0s 225ms/step\n",
      "1/1 [==============================] - 0s 213ms/step\n",
      "1/1 [==============================] - 0s 224ms/step\n",
      "1/1 [==============================] - 0s 223ms/step\n",
      "1/1 [==============================] - 0s 211ms/step\n",
      "1/1 [==============================] - 0s 223ms/step\n",
      "1/1 [==============================] - 0s 222ms/step\n",
      "1/1 [==============================] - 0s 235ms/step\n",
      "1/1 [==============================] - 0s 223ms/step\n",
      "1/1 [==============================] - 0s 237ms/step\n",
      "1/1 [==============================] - 0s 210ms/step\n",
      "1/1 [==============================] - 0s 218ms/step\n",
      "1/1 [==============================] - 0s 240ms/step\n",
      "1/1 [==============================] - 0s 207ms/step\n",
      "1/1 [==============================] - 0s 217ms/step\n",
      "1/1 [==============================] - 0s 213ms/step\n",
      "1/1 [==============================] - 0s 228ms/step\n",
      "1/1 [==============================] - 0s 222ms/step\n",
      "1/1 [==============================] - 0s 217ms/step\n",
      "1/1 [==============================] - 0s 219ms/step\n",
      "1/1 [==============================] - 0s 222ms/step\n",
      "1/1 [==============================] - 0s 218ms/step\n",
      "1/1 [==============================] - 0s 229ms/step\n",
      "1/1 [==============================] - 0s 224ms/step\n",
      "1/1 [==============================] - 0s 225ms/step\n",
      "1/1 [==============================] - 0s 214ms/step\n",
      "1/1 [==============================] - 0s 218ms/step\n",
      "1/1 [==============================] - 0s 234ms/step\n",
      "1/1 [==============================] - 0s 208ms/step\n",
      "1/1 [==============================] - 0s 236ms/step\n",
      "1/1 [==============================] - 0s 217ms/step\n",
      "1/1 [==============================] - 0s 226ms/step\n",
      "1/1 [==============================] - 0s 218ms/step\n",
      "1/1 [==============================] - 0s 217ms/step\n",
      "1/1 [==============================] - 0s 216ms/step\n",
      "1/1 [==============================] - 0s 215ms/step\n",
      "False\n"
     ]
    }
   ],
   "source": [
    "cap = cv2.VideoCapture(0)\n",
    "while cap.isOpened():\n",
    "    ret,frame = cap.read()\n",
    "    frame = frame[120:120+250,200:200+250, :]\n",
    "\n",
    "    cv2.imshow('verification',frame)\n",
    "\n",
    "    #verification trigger\n",
    "    if cv2.waitKey(10) & 0xFF == ord('v'):\n",
    "        #save input image to input image folder\n",
    "        cv2.imwrite(os.path.join('application_data','input_image','input_image.jpg'),frame)\n",
    "        #verification function\n",
    "        results,verified = verify(model1,0.5,0.5)\n",
    "        print(verified)\n",
    "\n",
    "    if cv2.waitKey(10)& 0xFF == ord('q'):\n",
    "        break\n",
    "\n",
    "cap.release()\n",
    "cv2.destroyAllWindows()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.11.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}