Landmark_Detection.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 90,
   "metadata": {},
   "outputs": [],
   "source": [
    "import torch\n",
    "import numpy as np\n",
    "from torch.utils.data import TensorDataset, DataLoader\n",
    "import PIL\n",
    "import pandas as pd\n",
    "import torchvision \n",
    "import torch \n",
    "import torchvision.models as models\n",
    "import torchvision.transforms as transforms\n",
    "import torchvision.datasets as datasets\n",
    "import torch.nn as nn\n",
    "from time import time\n",
    "import PIL\n",
    "import matplotlib.pyplot as plt\n",
    "from PIL import Image\n",
    "from IPython.display import display\n",
    "from torch.utils.data.dataset import Dataset\n",
    "from torchvision import transforms"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 91,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(50000, 218, 178)\n",
      "(50000, 10)\n"
     ]
    }
   ],
   "source": [
    "data_count = 50000\n",
    "\n",
    "y_data = pd.read_csv(\"list_landmarks_align_celeba.csv\")\n",
    "y_data = np.array(y_data)\n",
    "landmarks = y_data[:data_count,1:]\n",
    "image_names = y_data[:data_count,:1]\n",
    "x_data = []\n",
    "\n",
    "i=0\n",
    "\n",
    "for idx, image_name in enumerate(image_names):\n",
    "    \n",
    "    image_path = \"C:/Users/Asus/Desktop/landmark detection2/img_align_celeba/img_align_celeba/\" + image_name[0]\n",
    "    im = Image.open(image_path)\n",
    "    im = np.array(im)\n",
    "    img_as_img = Image.fromarray(im)\n",
    "    img_as_img = img_as_img.convert('L')\n",
    "    im = np.array(img_as_img)\n",
    "    x_data.append(im)\n",
    "   \n",
    "\n",
    "x_data = np.array(x_data)\n",
    "print(x_data.shape)\n",
    "print(landmarks.shape)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Scaling"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 92,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.collections.PathCollection at 0x22c1dc8e148>"
      ]
     },
     "execution_count": 92,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAPsAAAD7CAYAAACscuKmAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjEsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy8QZhcZAAAgAElEQVR4nOy9eZAk2Vkn+Hse4XFHRp6VWZWVVVlXd/XdXd1qHS0koQMhDiEEaCUYTCMYGnY4NMtiXGa7sDuzZmAzA8OOLbM0I0CshDRi0EggJISubtQt0d3V91ldd1VWZuV9RMYd4W//+L7P33MPz6OOzuqu9M+szCsj/Hj+3OP9vvP3Ka01YokllutfnGs9gFhiiWVrJP6xxxLLNpH4xx5LLNtE4h97LLFsE4l/7LHEsk0k/rHHEss2kSv6sSulvl8pdUwpdUIp9ZtXa1CxxBLL1Rd1uXF2pVQCwCsA3gNgAsDjAD6itX7x6g0vllhiuVqSvIJj7wVwQmt9CgCUUp8F8CMA1vyxp5I5nUn3bu7sSl3B0LrFc0mJaRXp755CFQDQm6BtUnl0WWx+8YsaocPHe/ytEzpf1DEbXfFyjrHHIPtezox2+KjFdh4AsFTO+99l5tp0/nqDzh9+ZvK3Y33usDIpg/K8wDbqvvzzplwAQGOAXttCsQYAGHLL/r4pRWcQEJNjZQQtvsB0q8c/prKUBQAkK3xskvZu5+j7XLHu79ufrAAA0oruPfy8/duMuJPwPjK3VS/tfzbfoPlt1+geHboM+PX0J6iTN+cfytP9FxN1TE20sbTQiXzUV/JjHwVw3vp7AsAbwzsppe4HcD8AZFIlvOnm+6PPFnpRtBN6UbzN/wh1gl8o65S14QwAYPLt9OF773saAPCDfbQdStCEpeBhs+LyE7AfbIo/a2on8LfjH2OOlyt1Nri1RMSjk2PWG62MocVXdzdxb+EXcolfxL9evBcA8KUH7/G/u+GBORrL8dMAAId/jGbgCQCAymb8j1Q2yxfiH3eFfjyyYOhO9xgVn1eN7wYAnPipfgDAfe98HgDwPw9/0993PNkEADT5x57i98rl+7rYof3+YPo9/jHf+cIdAICRx2gMjT76WcwcoXk78vZj/r4f3vEYAOBGdwYAkOHnW9dBiziluu+jGdpnmef28dp+/7O/PE0/oaXnBwAA2Vkad4LWNSR4tZq/t+0f8wtvehAA8L2FF/GxH57quq7IlfzYNwU4WusHADwAAD35XdcsN1cneOvyS8BLZoKHfCk/8u0gndDjbXqMNA3r805nawbDP1wvRWPweE3JJloALu3ZyWJWaaf8z5yWXIc3DDA6GbwOAOQULQjyY5Yrhxf5zYhBdjOWZpsXSPml8FbJHPB7rFJm7tOOGd96ciUOugkAY9bfuwFMXsH5YokllldRrgTZHwdwSCm1D8AFAB8G8JNXZVRXSbRlK3psh+k0IXohQSu0EyP6utJhPFhlJExWzZyqGs0h9NWbQ21pC4rNABEvw8ieo+vl5RkqozCK+h4eUYchsqJJLVhpGdMiySa58oLo2cnIdZr+vpkNUDRKfQ+LaBcyt3XPmD8dz+GxhA4SrYPf46SF7EWHbiAFb12f02X/2LXWbaXULwH4KoAEgD/TWr9wueeLJZZYXl25EmSH1vrLAL58lcYSSyyxvIpyRT/215P4jg2X9KOc01xn71hEWuzZrHdI1bQ0Wui2xIVe5URM8dynHN7S32KKJS4hXCr3U2sb1TnRCB4vPjZx5matm06B1Ge5Y9G2w39vRkR9tx107Tar8aylq5APVMaWzhhzwpUwoFp/HuJ02Vhi2Say9chux9Pt7D35/5Uk0zihY1W3gy7ByJ4IeUCavO6tF8bZaOUEuuPrYbE/3Si+fqUijqBLQb6wtDj+VGUHndOwvmRnmjjSJEauXH6t5JmGHG00OC9wjH9s1L6ciNPKJfn8tK84y6ISWATFwvH2Ojvoyg2TyCJKnjh05V2BQ8eWkrXuMYWukwilg0SF4Ixjjq/DR9c6Rstoc+jN9TgZKHRrEkJ2EwbyZR7criyJ6LHGEkss17lc3za7tZSJWZR0aUWURARB+DipJloECRsdelXcqgU17XZgX+VnLl6dVOfw+bwUpyBz2Gk9v0tqjTF0GHHrLfPqF8RmF3RmoFVpeidcy3B2lGiGcr7g9nKSamodY7N77VA6cUjE91RIGZt9s5pbjOyxxLJN5LpGdm2t7p00/T/lEhrJap1Rm0s13K7SFO91i+AuaSG7bgWR3S9ycYIY0lUgEyGC4pG58ewD6DCyu64gezB1Fej2hoe95FVNtnqjYezkEiO7FmTnX0UiQoPIqKA3/krenjqrmw3P+Cl0yBsfFknhzbnmyplNRpZiZI8llm0ir31k30S1m6Q5+umxvNFWuZjYYekQsl+OiI0kXuAoz3t4FZWhBD4P2X1rSdT3csWw99+2GcPFLFFe641Eyi8bbXpVUk3rHOFCGClX9UtbI2x4eUbiqQ955SOFPfSC7EY7k611et6GvfDwP6dztRrm1XfaUtrK2gWPMZEM+ncA8+zD8fUqnzcT8V7JMwk/D/97z/oZSnVqqBDG4dPKrqWUiRC4HPtf3xcfI3sssWwbee0j+xWIjezi8MwkpbSVvapXEIO+nsVjA7bCyF6p0wTmqwaBw2i8bqw8JD5D0maYkhid21na5tJko6Y2oZ01Q+cvd6ieXjfMGFVHNEMeEv8qxDeQUiHfBK5OjoQgfduzMLcTjK+HS11lbCnH3Hs4Z2QtiZE9lli2icQ/9lhi2SZyXavxXoSDrpCicE0+kPcZy1oihSPNJr0qifrWJx+pJF1b1PgSh52iasvDjjNx0Ik6X/aojl3VDc45TVLTO1l2BHKpe4YdgVc7PCsmkjg/61a6rGqHHHT+F7RpZ+iLfNIOB776TDWxxBLL60iua2QPpMvy4pnjFfFKQm/bQaQwqNoh9OnUhYOu21n1qgujM/vWkE0yskcgmjjOwjVRwior9+NYyC7+LSkykfBWljWIV+tdabK3TVKRAYPsa9VcSdlt2jHPQZzMnQ24kWNkjyWWbSLXBtn9hArrM7av16KQVhHJNX5JoiuMnGESAstmT9F3SSdoczqbDFvYIiGT9UorRSKTaV5FiUqsSIS4zTeTXONxHMonVmjS307ToJzwxTkZLhcNJdMIBTSSEa9Z8xLs4DSNgUHZ186KDiWW2HMrjzycTCNPWXwQAS49ScriY7RfbEJ+nUxEumz4vOHP67o7/CjPoSrFRaxuSipy4ITh0la+SS+CUGOz73CM7LHEsk3kurbZfRICGDvM54u/DETfTiLayyrDqWqxVlA3iLzhDF6lUlfNGkIn5Im+FFtaUlblfhJWMEY0wrDNLokrko56tUTm1o90dOwEn5DNLv4EcdKzhlpIdEeTEtAxeUUsscRy3SO79X8mIigkg+WAYpt2E3VvbxHSikqbkbDKdnhtC4k62fbXGUZ2dh8Ij/u6FGK89fwtYd5Ck3qpRVWFdlzxARF6im/g1dICxR/SaNlFOfwfqSkSrcN3RtAXpWTVP2azxCsxsscSyzaR+MceSyzbRF7XarwfMlmryMpyEGlmJE3GDDXrijiPhKutzVvJTFWXEjKLkkvoxiuiObQqpliSHWebYfsVafF9VNgWsP1bfsgtVPUmDR2vdlJNuO1Tx+tO8OmyHHiXRHpj/r21JEb2WGLZJrK1yK4UtKPMquVsYmWW9rlRp5P6aR1KiuBjOna7cHZsSJqh6aJxfTjm1mNDcUJJNesl10iRRguEptLaOCGtmtsRKCcMNSHuOZ8v3s5dDR/vbTz/0qo5HHZaj4sgXBBTD/HfJ+sWl54kdPHWb+iYZI47K/QWdvyt1Scgam5bfLSE3Pxtyw698fHSbIdPK85JYc9JWE65zWo4MbLHEss2kde1zb5m0obwjFtJNcI1bhcQAHHobS2RTjCrLQ69MeWZbhhjdzOMNFck8hzTnA6dCdqrTogPDgAya7wTTdZUyk0uhIkweX2bPaRBvFran3SC8ToGc2VGdeg2fIaaVFAzpWNiZI8lllgseX0ju4gUy0ixQEI6iJhdEslg91bxsK7F+LndRVhYVxkJk3X6XDe2LqnG54vP0FgS3AWlmKDBiJ1s88H59VOhc4nne5V7vCXaEf4K8fG4wS6xdpFLuKfb5YhoTdIJRtswLv93gn9KgliaORRt4g7RcFoblDjFyB5LLNtErgmyi7fcNoV0mG1Als5wyWvEedZaZm27rM084XYvbODyYqiJEF+8vWLKyi9kCf5SG6FAvBpdXDdjv4X57m2uefHsCmVSi2PACfFeh7niAeOFF8862/KRnWBkHx3y4EfxxrOq5nEaq/QkFx53GXfeWdumDneCkR5vpQhvvDRmUaFOv5fDQBzVS6AViphIDoPX6o6z+38LqyyPLctUWZth1g1LjOyxxLJN5Jra7F1ofqnHC3IkgudRbY6pr5hlUi2SQTZVLwEA6gX6uyeiVPD1JFHovNY+QqggcfbIHuIhN3CT+4UnL8FU7+rmamleOtwBJhRn93u7A742187S+dJMEyU87lHouRbGN8PEmc1utNYcvQl3+r2cTjrrifR4E2+8bps5928pdCOC7D1p8le4EVz2G8mGyK6UGlNKfUsp9ZJS6gWl1Mf5836l1NeUUsd523fJV48llli2TDajxrcB/K9a65sAvAnALyqlbgbwmwC+obU+BOAb/HcsscTyGpUN1Xit9RSAKf5/WSn1EoBRAD8C4B282ycBPAjgNza84iWq7no9tpPwV9Iyh9VGm+M8WaFbXW5lQsPZvIp2KY4aWUXX5aC7EismROO3GXXejG3j++hNUL30of45AMCzewcBAAO7d5qdLlzksYgzNdR2ucm6v62qt1n9ZEdfV0NHe1/mrmsUaaLEOSVjc9eZv3Baq4TeWsySqyzvqOfXsdPfEqZNWCEta4BrX3QNkWdSD6XJSuou2paDjv1ufvsnsXa4xr7gktlps+fIs/e0unpMNUqpcQB3AXgUwDAvBLIg7FjjmPuVUkeVUkdbrcqlXC6WWGK5irJpB51SqgDgbwD8G631SmRYJUK01g8AeAAAegqjel2kjrzweicP7qNDjjqnZVbhRJ27cLSDoTdxSLlXOSUyjOjhziT2d5clcqshkF6vfbQkh4S1jqgQ4K7kIgDgJ4aOAgDq7ybYO5Y44O8z9tUCAMA9TQiv6+Q8ErTWdXZ+2p1bxEHXIpT2nXkund8ZMK6fhTfvou33kIbw0Z0vAwDGk/OB+4h6cj6i83xLEkqhRHm/izea60jIrbaTjhrrXQEADCVpW7wMZ5itYfkFSD7CM7tsW7QMO6kmtJVz8C81v46ndCONbVPvm1LKBf3QP621/jx/PK2U2snf7wQws5lzxRJLLNdGNkR2RRD+CQAvaa3/wPrqbwF8FMDv8faLm7mg6njRCTIbIH5kmC58jKCm6u6qIeBSa7PtxiUHV4LoUeWNfi6QXDc0RJvPPNxOeDMS5kNvhlZz+24EsRMhLSCkAEUiY57R7FCK1vCfG30IAPCF9636+3x95BYAwMAT+wEAfa9wWGiObGpVrXefmNFeQnAY6AUArNxMSDt3u8Gf4hFC8F/Z9xgA4B25YwCAXZwy6jBi1q15dIW5lW82x/N1W2oKAPBrh78GAHhh727/GCkXHXTp3vamyE9xwJ2l7635ktGl+TM3lLJbj/CZCOI6fhIWbVdbXG67bI5JlXlM9eBzldNKuM5Ol43iqI+Szajx9wH4aQDPKaWe5s9+G/Qj/5xS6mcBnAPwE5u6YiyxxHJNZDPe+IextuX8rku+ogV1ATRfwwZd/1y8FWhSIUPZEmmgUWc7qR5Km72cxIn17OMwel4tEW1gLX3EvnVZ+6tslObYZpdU3goXZLjW2Wa9HO/DJaEeNVgTmqqDOWOtTR3uAQC8lB8BADRKtG/pNM1tepER2OoiI15wKVtdHaV9Z++h7/ffPuHv+/ah4wCAm9MXAABFRrMQ6AXKWgXRBeEd/m4kQZ9/f/4sAOCduTO4HAn7Xvz5FvLXdbz1GfagDyRJgyhxgsyM9dASDT5viP0rtUwXeGl2GAAwO9TjfzeQqm1q7HG6bCyxbBPZ8nTZSDS35RIQXs6lNkFvJTZQtUFIUmFkDxeFXI682itmlB2+pkTMqSD4Ehe3VHi7xCjuWXbm2SbF08/WBwAAE1WyqZ8/PQoAcGaNRpSdposNzHKnlmlOZ10iu1w1uos1hEJMtj2nmQB0kc578cKYv+8nDhOK/fXgXQCA79tD3njJjdiTJpv+UOqif8zhFEUR0mGKLJY0F9ekrXlq6ehnnxDyzYiXcS2El9h/1HMSr/5oYhkA8Mb+MwCAl3bs8vfxznJUQpzufL70Ep1waYIQ/Ym94/4x4lcB1v/ZxMgeSyzbRF4z5BWm7HUD6Irw2vsIHyKetJc5WSlrXOJY10Gb/XLkSu3zMCliVCx+Iwl7f8ueYdlc6FAc/HiDbOpzjX76nLuinCoTek8ulPxjmqs0L7mTtO05TaO84TjZmYn5OXOx8Dhl3oVUUrLl0tZcc3zdEeTlOHtyhv0KJ6z49Dco27E+QvfxncK9dIoc7bN4Mw9jzNisH7rpSQDAvx74DgAgr6LxLKPMq+9pLnhZKyJk3aag/KU8s/D7kWbfyU1Z8kWMj836382cIA0qWQ2eT97f9BzN09NLJprw/aVn+Z5a63Z7i5E9lli2icQ/9lhi2SZybdR40bLXK4oJ77NO0s1avPG2cHMPn6NbGGsMh/rGcjUSatdT98KqYeQxoejiZIdU3QttSkqZbZuQzLEqqe9PzpHTa/LEEACgeJzmoHSWVOp9503NglMmB5cq02e6FYoBuRYZv6jnMr61+Onsz0V9l2ckDjtW721e+QSPIT9HDi2d5mvz/Ax+VzJbzGv89TfcBwB49F+MAwB+Y/wrAIA700sAgELEK59mlb61RmvmgHrf9fjEwRtM8FnPtBNn8EiS7uuegXP+d38zSs8sK/E4SdLioaUXaHtsYtg/5rlher4HUjM++03kfaw9pFhiieV6kq1HdrUGWodb0/r7h/6OWJ66Um0jwnfCTNKpMRddJx04ZFNOttCqHnVMeHiy4ketqmGOc0GFMLI3rOuKA+6VFhUZfmXhdgDA83NUerpwot+cf5qu2nOGkOTAeU5nnSVnm1qhra4ZB5dfcuqzzQRHrhMW+4+gcVI6v0hRC3dw4W2gP5zvvOtE7ot81lwryW2iOY6lKjzOUFmstjSfoYcm6fQvU8jwV9/+cwCAW99PYbvfGf0SAGBv0rz6Cd+hyyWoIYT3LsFhuh56yqglPDfg0P3cljvv7/ONsRsAAI3T5DzNznESEg9XHHeJCVOq/eShvXTeUgdNffayxhZLLLFcR7LFvd4YuaPSWr01oDW0HNkovpbNLzRq9rcO84SretBmX+My0bKxktGFyn54jg/OWF+vFXqrcPixykhzvDXkH/Pd1YMAgH+6SKWmc88RwhfO0LF7Txj7OD3JCC5ak6gigsSZdNdtyb5+nzYhmRDb2r4/n2gihOz+82UyC8umVozomsNyis/vX7dllZOm2EYX9M8FiUd8sTQHzfeWmKd73/M/yO6/+DzN1w/88L8BAPza277iH/M/FQn1c8puDmgQ3U6qSWzAOBJ+/oFhhlKdpeT4htS0v8+bdhIyf20HaWjpZbmx4LnS8+Y6Ry+aRKRK54U1rx8jeyyxbBPZWmTXlDTjOwzXQvOoQyNWzHACThjpbaJUn+6Hu5G2QmWBm1mR15KoY8OrqKCDjQwyAineEJt9me3yCx1KdnmkfMg/5tlFSrqYmSOve+ECnW/HUaqNTE4tmosK8gqSp4LebN9PYtnJXolSaCXV1WGPOCpUtqqt/m7S601IKpRvBycC11d251ZGe5VizcoJagE2LZUSL/5aPeFDJc2AQXbp8Cv75I4SYt50gubtM1//Af+Yv//F2wAA/37f3wAASuz6Xqtv3Hqy3rsidyaptJKyO5QwPpNb8pRo89WdNwEA2hP0bJKVYO20WzHXWT7D70kz6XfwiZIY2WOJZZvIayZd1pcNlp9oEgv+rouVwax+YhM6TbqAkAB0LiHO/mqtjGKjl7lgRRD9/5t+CwDg6Jm9/r7p52ml3/8dQoPUWSoJ1ascF7dPLJ1ZBDVDMW3NSN8Y6+0aU2qaEV1QWaimrJi5EFCoLNvS7APQ7Mn3vfBR8Xe/d3tQC7Dte4njazeohakWj4nPr+rm/L42EIoM+JoEH1N69IJ/THWVtKUPfuznAQB/+oa/BADs51LUy0H4SxH7vZKinht3kx1/+uQ4ACAZqmK1mxHnJrnTzHwRqMVx9lhi2fYS/9hjiWWbyLUNvdmyBlWol+LwTUSv3C7K93DFnKXyi4MuwQ66pRapw4a/i3aICq2k+KOw82W9hJnNSEVLWyY6w/k2qdP/4fR7AQDTD5J6ue+hqn+M+zLxsIk6rddIfqEvQ5x8vK+EsKp7yVxYGTevQc8ZDomxuqsbzBnHiTddPO/25WohzjlRye2xiYovx4R45FXaOJgMeTCr8/wO+O9Clj/PmjCqtP6SEJ6o+HbiEACofM7/f+4F4qfb93s0H7/wA/8aAHD/T34ZAPCTPSactRHbW5Szdi1WG/+c1t8jSXK0+rXuI1TdlpnhtlWNoKMOAFLLki4eVO/DEiN7LLFsE7kG6bIRVJ0wobV2D3Of7ea01hGpc6f9MnMGXQuThAbJSiewjzTo8xI2snPoi6nMhT9eWFoSIc4y+i66djli+L6sx24SFjnfgkdI++dTbwUAVD5NzCX7HiInkjdjash1uJOKoGeoXTJgHGe6VAQAtPupjn1lP2k187fzWPMWO67wt7dIy8jw+R0Oqwk3PGC439fs7hJO0AGgeL79FFc+r4/oFsOMakj1EqN0ODwXbusNdIUVuxtHusFjAT80mVgih9z4X9M9fnKRwnNTP2ccmL848DAAw1orz9sL1bnbsiZvXYT0MzTfkaPimKExCqWuTlBiVU6a8Fgn8cPKCutS1cTIHkss20SuTejNL0m1WEnYNl84TCtv9W6yU+/ZSyucdG55/JV9/jHOd7nt8lkO24SWLtumlwSbMLJXmLGmA1rNHcug8lFeQIL/Iyt0VEgmnDzTCq34LWvpLTO6/MXs9wAATn2akmdG/o7t8k53yeVavdFUgVAb/QaFJKQ2fwuh5sohOt/gfuJue8cgFY3kk6Zt9cQddMzLzGJaP02ls7lJKsywO1ynVmj82TlCo/QCfZmc5pLURSor9TvDAPDqIaSVRB8OjamMlRQiobdkyFIWuz/KfxBOtPGTbOhV91bKXYc4xULgenKOnQ/RPH0l8xZ/3/M/RvPxm6OUbjuWEC54vr/uEXVphvUQwtuvrbxT0pHnTcNnAABf4vTZ1DK30LYodgP5YetECWNkjyWWbSJbj+wO/DTZqE6a9QH67NZR8pB+eAd1A5EunBcrhpxhoZ/KOiVlVES8tardbcBI2y5JK2zptafAZd9rS3qkrVGCGiWdkBYgXGj2IvztGmkpj3zpDgDAvi+cBAB40jVFPOIWwjuMhE4fIXDjRrLvF2+k+ymPm/M7B8gGvXeM+NeP9JCW1J9Y5fvo8BjNqI6wgvB9g4RycwfJ3l9sk/d6rlnw952u0XcXy7zPHG3TEzSm/hfp+ZSeXzJjukh8a+K5V+JzSETgjtjqrZCLWQpjwqW1QDcZRkhEcwhoG2XuclNhjUEQfgdpM6P/uODve2L+MADgQz9IXvL/+8hnARhyDDfCZyNov5bNbmsDgvpDDo3vSIHSfB/aQwVQrYukWdj94ZwNKYcRed1YYonlOpUtRXatFLyEY64asSCJnZ1JEqoVucA/o+jvHTljc00OEoXP3O3cv60g9hptshfN6pef5p7bTGKxyvzxK+wJj7K1Er7HVTqn0LFCbhDFNx6O0xcZdRq87ymrP/z/+dgPAQBu/CIhR2d+MXCsxMUTg4aQon4Txd6n30AI1byT0lr37SAChDf2GM/97jSdbxdzqYsdmOCwhWg1wiNP4+X55pYkmQxtJWpRsVh5pUuM9M270KJxnqxT2e3D91IPuGOvmBLdHibX6D1J582ep+fpLHN6rp2OyzF+n6VWBZFXJdaJeodKc/1ziPffNdqS+EF8/4GU35a5Z53lMxh4grbFCVKBfuVtRI7xkQ88CAD4WB9pojlL+wvb8/JdGPFtKXKE4c40Pdd7d5JW9i3W6NKLXYdsKDGyxxLLNpEtRfZWUeHCO1JwOIstZTlGM/OMvDX6br5OK6f0GhM787aeSf+Y5r00/AyzSe7NEUJK+eq3J/f7+849Q/ZXbpLO32jSKl73goQF7oY5UutLVCkrYFbVP5l5h//Znv/GJaIvnQrsKwijmHxw+m07/O8W7qJ5GthLCH5rH21HMtRL/EDGcJAPu2RHim3uQfwGQY0k7xj7tcchWzrHyJ7g+8lw2ecQTCaadJqRfcaSdL07M2Rnfl/PcwCAC+OmF/pzbyWihYen6dmcO0kdaIrHCfH7Tphy1tQCoXxyjl4UP6vPt/P5Wdneei7ckQ6yShDdDT7nQKmu/EdKdCWngLPuVNH4KRymxkqfp7GMf4E0tb8/9XYAwBM/swcA8Pt7P+8fMyJcIWuQn9hRnbCuWOLn8KYe8uc8so/8PO0p47tKlYXSDetKjOyxxLJNJP6xxxLLNpEtVeOLpSre+b6nsMoJLSeXB/3vJk/T/0WjbHBrZXEC9bJ6KWmEANCfJKfOUJJU2HGXVNoMx9f2pE3I5E/rxCferFB4qNmMbgPlKqPetTQXxyhJkAkmuYQTZwDb2cLOPNbLvl0bBwA8/j9u8/fd+wSH2qRARRxFrL5ffBep74tHTBipMEj3XEjTRKU4vTLN245lPohTTUyLiiZHXF5KvVV3eMp3wPG4RZ1vrqMjijovnGqyTTnk4BrKGF76w2kKqX6glzxdswdJHX3sPlLrv3XRsPKcnSL1371ICT79L9B99L5Man3iInmpAim8ouqH+O5VlsxBP2nHZreREJ/U/+e4SEZSbe0CHwnL8XdOhb4beIbGNvmXpGb/wf3v8Q/57ZGv0oLtm+YAACAASURBVPj5/Gs1kgS6Q3e97OC9m02je0bp/X904BZzDE+v3iAiHCN7LLFsE9lSZM84LdyQu+g7dCTJAwD+MU1d+k7NkiMtnaTVVlorFyNQ6GSdQjonwaEd9qPcliYn3o0Z48zb008ocKZAO3ktbg7YzvN1aEwl59L7vkSlO1YZ4c9zc8Xf/caPAQBu+tQZf18/mYMlsYsQbPLdtF26hxxU2R6DLCpU19v2CC0aTCxut1+ua3JKSdKMOOrkc0H2TgRqV7istMnnT0V0S2mGnJly7UzEsxJJicrAKHTIJYfioT7avqP4kr/v5D5CdgkNPvJ2Six5YoKcfOrlcQBA4ZyZk9JpmrPMKTpf5wJXjkiJbpNDe3YqcphLj1N3dQ89O5v3XlfZaSfdh/gXJKHDwafp88c+e4d/zBf/FRU0SalsgVlsRVOM4qUPF2YNshP6jaXTAICHxw76+3YukpPQaa2fXBMjeyyxbBPZNLIrpRIAjgK4oLX+IaXUPgCfBdAP4EkAP621XqPZF8lSK4cvTpoVb9hKkJmrEsK2GjSk1SatrmJTC4f6jNXL7OgshTkWVsnGcvfTSrmfm9NLIg5di+z6k1LrwPzxy+1s4Pwda5VNhNJi/ZJXFVx1nYCdrAPn+b2zVCZ5459QcYg3b/wIYqM7zO5avovSS5dvofvI9xKKJCxtI5Xksl7+LOkIOnC6sRVKbPq2NKMd299Rtrq5Z4/3YbsbYodHHNPFFRKtFXmWMemwZlJE0KaW/mdjyWX/MyFyEIIRCemVh2m+Lt5BZBNPV/b4x3zrPNn8+tuUzrr7S1zKfIq1SJlzLvsFTHmtFMnI1pFCHIuvXoU4+fwSY0b/xDwdO/isOebPjr8ZAPDeO18EABRCv7o1W0VbIunWt2Uoyebw2EX/uzOvjAMA0gvrn+dSkP3jAF6y/v59AH+otT4EYBHAz17CuWKJJZYtlk0hu1JqN4AfBPB/AfhVRQbLOwH8JO/ySQC/C+C/rHee1qqLqe/ugrtKK9BkyUJRTqbJ8II5y87Gxd1sU0cUrNRa9Fltnlb66VFarSWpJq+s7iiMgDrBhRLMMrvCyN5ax9vcCSF6WOxEHCmaOd4ie/P8V8YBAGPnyF7zrLJMSYdVnAJZ3sX3WOSEkC7eLfOZ60ikQMbGXWSsTjfy/x6mJhVEF5+J2OqC+IBB8jBK13n+bW+/oL/Pwc/na/Iu8nnCUgFyKojoLcEb3iVlXVfse/HXyDf9HJnZmwymAdvypSMU9Vg+Rz6g0hJpdt4Cl91WTXKQUFQ5XB6suQcepFjGQnavRHa8s8pUYaIJSqote+nTU0Zr7TxKkaZTt1Li0ECCSmfdCMIUeZcSvtYoKck012OcXPbWgZP+MS/vJB9GeilxVUpc/xOAX4eZ7wEAS1pr0e0mAIxGHaiUul8pdVQpdbRTqUTtEksssWyBbIjsSqkfAjCjtX5CKfUO+Thi10hXoNb6AQAPAEB2ZEynFxV6T9Aa0c6atabFoc1Wgb3ZVY6Dh9JZ7VRPsV9Vm84jhBQdXsPyyqSB9qeYVz3NyM4pu5WOkFgwh7o2nm+xxWW1Dff+clVw1QWAZY/+/6lZstN2PkzX9SpC2Gj2FWT3SqS9cP0I3HTQPrYn1k1IDJu2HbaHBWVzjtFmioyAmVDqqyC8aD62PS4agtjZTUisXogXzPOQa+ZDaC3HyPdR4gmdE1+nxXMZsO8lVVcFCSJSob9HOc8CAMYzlGsx1E+fLe+nyEbxFEVsHE6f9VZMJESQ3I/BS7mt3+/O0nJSHPUosDawSNfxu+pIbH7elPXueIp8C19bvhUAcNvQgwAAVwXfLyDCTxRKoJVY/dsLxqL+wjh18l2dHIK3Trb3ZtT4+wC8Xyn1AwAyAHpASN+rlEoyuu8GMLnOOWKJJZZrLBuq8Vrr39Ja79ZajwP4MIBvaq1/CsC3APw47/ZRAF981UYZSyyxXLFcSVLNbwD4rFLq3wF4CsAnNjrASwLVEQ2tmDl21GI1HRR+cvoumSfVcLXDYRF2/hStJniZZFDdXWyQalXmGnWphqPjWD3PC584qaPSBkrU05ZV1eVKF2FhqgmF2kT9qmpznQU+3yMnqEXw4VOk8IhCa9dgi9pYGSOnT32YxpZxpD0ybZNW6C3n0rUG0mQeiHmyM0Uhq0LCmCEyVzk2ZxJrhMZskTCXzLc46vzqt5DKTvsELbgMRL2n+3EjjD4hV0n45gj9XbVYc0TVl89ErZfPc6H0XAAYT7Ean6N5mWcncGOI3o3sCjnhHMuc6uKUF3VequksFh1pPeVluDKuEEyt1QVS5x2LKSd7isKtXz1FzRp/ldX4KOlEJNgA3dwKe5Oml8C7dxFn4Wf2l6BTayfWXNKPXWv9IIAH+f+nANx7KcfHEkss1062NF02nW9i/J4J1FqEfm/onenaRxrLN7hQZaHJXOfCKGOFyFKhcNpMmRDyTJPTZ60al5mmSaIAAI9XwCyjvzir3Ch22TVklRG9YnGQX+xQ0k/qJK3wUpDhs6DYMkRhoYWbWJsp0WqdZMejOOMSjjl/PtnksXGhDaNeuZMJbAFgRtFYSozwpQShXZ6deGVFY7STbPywXMgH60pqp2X5yXH1UAhO0F/G7VgOzHooxFkUZxtfzrGQTcJy9XB7bXZaeREOwDu4MeI7B18GALy4lxKVqjtoXtLTdM+JpknO0h12pkmBSjKYKAPr2XmM3B476iRFO7FATj4ZfWfQnN+p0Xka06QFXGRttZhcoxW1JV0hOC7OKjpmTu4rvgIAeHTvOBbSaydMxemyscSyTWRLkb3XreH9I8/6f7esFfux5XEAwMo8Ibmq0HdHU4T0g2laOcttg1xn5ihJIT3NYboVCnH8WZLCXr1ZY7+en6Ukl+QkwX2rnznUU3TeotOd6euzy4Y459xQXzXbzvxWmeyygedY65DEDLERLcaUxh4u9NhN3xVytK8gukjHMyi7UM8FthKq6knTvYq2AxgePymD7WF++F6XNIhiItSbDUGbn+6VztfD3HR26E2ScsRH0uAwqa1dAMCgaxJMxKYO2/7hoh3AaBkjzIDj+oU89LybnoRYDZoV+f6PZM8AAH7oMLHl/O3C3QCA7BzNW65qwrIOc+77nPKrUjPKJcc2e20PI3uai2eYydiR1tBs07d68+Y+2OZPzdMx31qloq/9vU/T95YGGbbZO364N1hubZPFj3NS0bt2HMMrye5n6t/nmt/EEkss15VsKbJXOyk8Wd7jJ4KcsskrTpKd3fsSrVjuKrPAThP6fW6KfIGqZVbB3AXatzAR9DJXL9J5Z6y7K3D3EukDVz1AK6AkYeR8j7uV+roGWUVYylaK6t+fpjzfsWPcFYXtPkmgcXqM72D+ZrLdnEHjWQWAxQU2BFeYi87iCK+s0P/Ti4woDFArDKb1Qcvm3UHXLgzQ+QsZ2rmQIi1mT54QoegaNNjJt9IKecCrCS5I6hgm2ots956pkO9hvkaoOb9CqNbgNObkipnTdokRqkhjK/UQih7ooxTSpBUxEP/E3ix9d2t2AkB3eqw9/x0usBlgTeQnmO11+Q00lkdWKbFlb9k8B1eeDf/tcScb4aPXLVOc47D3vclp3J007VNo8n2x7a606RLb6qHxlU7SXP7dBUrl/fGep+j6lmYSTpetc5LqetyIJdZm7sidRc7iEwxLjOyxxLJNZEuRfXUli+9+9Ta/h7Rrshwxwrzu6SVGQu6znZul9aj9Mq++VvcLh1HTaQu7JhfTLHR7ab00fTd3K93yjSMUCTjA5bB5pzt1MYzs4p1vhGiFbBKHxglCO2eOGGO9kBe+uc8wxVZ2872wprN6gY7NTjBhBJuQbtnqXHuB0C5Rl4bzTKIgXXCz1vgL9P/V3RRbnhvhqMUuQvL2iMTSzfkrRULuIb64FNPkGNFXLWQ/tkypqLMVQrkac/E3OYfBV4QshSg7QfOfu8jxfJdUkqf2kXbQGTbIVCgROs/2kaYj5cijwofv0rbXMZqRPAuhMRNfzPv6yVdUexuN7SnvRv+Y0Yc4iiC0YBx3lxRnZcfZmbVWiCKqO1jT6qV5Sc6QFpCoGJ9Ea4jGnV6mZ3Z+gnxNZw6Rj2m3ZWeLjR5G9HAarS39Ds37/uSCT1ASJTGyxxLLNpEtRXa3rDH6UGPdMhqxqUWk97pstdVzPUywpzjerRilPSt1S/q9d24jm+o9g1RIcMglezCt1raJRDx/1WUCBvbKn2ka30PpFR4bdzZRIY7z5QNZcz5O0XNPErqllmi8yRrHp1uynxnDyj5axSV4IB1uBJwD/fN4LtMLzGXPAFLhiMZkipDlwA7TRWY4TerWnjTNi9jusl2tme4ukskm8fRCljP1uLNpPUnbdsLy4GdYU8s5gXHLs1SLxv4uc4HTpMuFU9KVJk8oOueS3S3EowCwPx3M3RCvv5CRfnCIiC5X7zMayoslIr8Y/QZpF0WOnKgOHWP3hdNl0niyk6SFNUq0lbnW0ouvbNDazdJnYt9nz9C1v1shoo0jqSf9fUV7lHct4ayN6CKC+iOJ6GxF/9wbnimWWGK5LiT+sccSyzaRLVXjldZwWp6/xHh2m15RP7xgUoGttgNB1V0nncBn4rwT9X11p7m9pdtInfvhA6S+H8kSS2cpwjG3llR0sI66zH8/WRn39+k5Q/p1mMVUWiw3SxYrCZdUp5a5xTTfuuwjmaVWZAntTHB+Eg1u6cRao51TIY49MQeYsMbnKqs06Z73F+b9Y95coPbO0sop4XPO01z2JowzbAc78V6pkNNR+gAsrXDCzyI7rcrmOUvOj9RdtzgUp3LMf58zjq2BIqnnI3kyLXKSKsxjEkbdM/UB/5h0iGevFTLPbnBJzf+egRP+Zw1OVz63RLx12WlSzd0GN7VcMqE34Zh3lujh5Tj9VoXeW2XVyyez9AATVZr34jn6++llut7Z4jP+vvtcSR/mELFHY0hz8ViUo06ceWmVDPAhhiVG9lhi2SaypcgOAHBMmMgW30ETbKhiRA6xHBYSngMjvDDfLO+n21q51aTAvvtWQvQf738cAHDIZebW9Ui7QpLxEx1ocGWGpxNl47RKzwk/eTAdVBc5CcPKJBXePZFWQe6Lt3x7XtJMRqcQ5F2XNsYOE78FmEp4n/QKl6k2aNtJ0/w4XEAkHHWAYbcR3jfpSJL2H8wFf99yh1BNuOsXynyP84To2UlJkLKGxMje5nutstNK+5WiZk5ml2mncp3Od+Mgh0tzxqEIALWOUX2mmqRBSbPKEWardUKML3dkTc+COQ7tzd9Og6g+RecoTXGHmLSlWnF4TjPnf5o529slRnjpSGM59ZxF0oCEzaZ4jr47Nkca0cs7Rvx9+xPEHivO380wz24mPAfEyB5LLNtGthbZlQqiesRC5Nvkoe98+9xCLs9l7rkB+nCF+CJQvI1W/g/tfsXfV9oH70/yiryJUFtYjBYgyE4rvqSJAkA/dwZpS/iGQ25ekRDAyknxw2cCTMw/4SdstHN0vey0mYziWS515O9aXEnZ6mHkXbUmTs7HSUfpuTqPgcYrOU0lC9l9zj+xqaVnHZfzetqoJhKOW2oyQy+XJbvLHGKak7iadc+staSXucR4hsZbG6TzVvaYpJDUCPkH0hx6O71EtvmFVQ4Zlug578ma9Fkp7hGtQ3wM0vJ73qN7Lzrmng9nqf/c+QFKzT62kxlphU/O5qBjP43mslU1I2m+HH5lrU+3TQqsZkZbh7Wk1Czdc+cJus4Xdtzl79s7TOO9OUXnLW3iPfWTv/T6+8bI/joSpT30V8uGvvg6EKU99Neur3t6rcqWIrt2gFYx6ScgCDoBQDsjiTDgLf8thRls47Vz5qVo93Lp5jCtnO8eJQ/7m4rkab09bezLXUnxWEavfuHUWFvCvbhknRdW1nbHPmeoIyhHHGo7cl3jd9tS1ML0Uww2bpU9sgwOuQtVKO3h37/0F7ilfA4vpnbiN4d+DFopTHxoHACQvZeQYHGx4J+/cJ7UiESNqZ9yTMU1SGPqKTJyWrzxLYb0ZX4QVUayKN5+KYcdzZEK0RqhY0/NEKu4+BGipP/pRSit8fvH/wI3N6fwfP84Pv7m++GljOpTTRPazy6wPcxFUE4/qUS1JpfFDpix3dXDNi8TdoT72Nm9BETGOLFqP/sCnhiTDCWOilgoHegRB6v32xxrF1IOa9NeCUUVd4Ot3ky2ukz7S5877O/7v72HNIT/ePPnAACHuUVrxmEqN/bOFxxLRdzkOhkj++tESu0qbimfRxIaNzenUPJqGx/0GpdSp4Kbm1NIQuPWxbPobcZ9BV5N2VJkb/YC5z7gQXEsMZMzq2wpz+SITKgopY5ZJmDYxejR75oXYmeKEH08RR07RxO8D1NN5Sz/QMLnKY9eBsNpivZnIsIXL57SSUa7lEV8qd3QlDLC1/vZdre+FtCRFV6KXERm7mayzWQOqzqLZ8+P4/bZM3gxN4rF/h2AUkhW2aZm7aKvz7i+qyOEjIrVo9XdHLvdT9d5Yz8hmc3v7oY0nFaIoGLAKjoZdYlIcSZLjoM8k2NMcOHNykoxcH+A6XrqJfsxo/vw/Mw4bl08i6eHx3F+Ty9qO83133EHUUvNN8jL/9wr3L31Ao1lsUQnO+ua+d+bozENu/QuiFc+gbWLl4SI80CGvP16NyGwpL4G0mVDyO4TiLJ3XsvfnrmebrO/g8kxWpwqvPP7SAs5O9Pv7+v8AyH7RyfuBwD8ybv/HABwJE3vep9DWs6iZ0dQrGjBOrL1obdYLk+Uwi+94+ex59kLWErkfRXzdS1K4eNvvh+9zQrO7+nle4pt91dL4h/760i0crCULGy84+tItHKwmC5eH4vXa1y29Mc+UljGb73ly75KZXPAS03yRi2DXYvJRNhlpBa9ySp4KuLF6WyAGFFphmGV3/VbFNHnogpOLxgm0d7GPKJEnGI6aal3yaATcnWU/lPex/fTS/s2+o1rZf5WUvM63MaqPcDNGisc2qtZvPSjND+Vg8yiwimpgz001zvSpFaOWG2S7XRYukdu/qgkDdWMf3+SVOYMc/zPd7jufIxUzccTVE1WWTKVfkru/xY6pskq+M4ijeWuggmjCUPNkSIlwAhjrCTOnFyluZB+AQAwVadnISZemOtOkmtsllwn9D7295JKLuFSUcMDIiE2+TO0j0q6Xf8XR13hLM3x2Xm6jw/dZKreWofp+X3+a8Sj+PFP/RwA4Hc+8hkAwPvz0wCAnIpgK95AYgddLLFsE9lSZC84DdyXNa1m7U4eUoe70epju1nC+8p3gvCZCAfdmuddJ84bdtRJosmpJoVQOjVrGnlfvx1zitlbKF8j0LGjVWANoZdRwhWevGAyil0MJIw7EtZKLvC153nbayWlcIJGKsXNDJmNtT9Ln49nCDmHrMaIwskuiC5/Z5zu0KQ0XBxnzUC2ee7uI5rD1ydu8I9ZmCXkrTM/XadE+65maJ4aVr7vHMdbJcQnTStvylKXnVuYk65lhQWXO8K6G/0mSWjRPiaM/iMFGvfqABWqpCyU9h10fuEWJ0+5qcD3NruN7gSdg8kp0l7Ui6T5LOwzTLQ/NfAdAMC7Pkgtvn/+ax8DAPzJubcBAN59+K8ABJHdsX4J673lMbLHEss2kS1FdgcaOdUJBUFIOEPUX30SayC9/XcYyU1L37VtdkH4MFpHIbtv568B+mLr3n3DGf+zuZv2AQCyMxTWUsxJ3iryaAsGRTzpJsKhSM25wh1uV61qXAzRMHed4CiQhPDag9zRpp/szeGc8YPIsJer3H+Me8YN5wjJBxnRU1a4rRmaceFol5PlrDiazL9oZaJJ3cysMGN9/wwA+N6iaS/8zDih2SMLlNs8W2X+Ok6QsRmH6wWa30qbQpAS2tvBPPQSXhtImnDj4TShfsWjY0y7ai/wuR1ulH2Eu1766C30kBaQ3WHG1OHnahBeBf52uE+cypq0Yoe1O/B3HhfEDD1FGtdXhm/39919H6H+PTniMBzbT2HlSjPFY5T32OriY73L63mmYmSPJZZtIluK7B5UV6+vsKTE2ytLVAiko4721vguumdpNFqHkT/qu7CIF/eGguE9e/ZtxCt28Dgxr0pvsE4PRxUscoYc87hn2SOd5uScpRqhwmo12FkFAFyXe31xt5u+DCG5dIKptY0tN1fNBY4t8vWkcGRHkhAyE+ioQuMrc7psnWd1iQtIVqxkkVHWDKSXWzM0T9KpRYgwAKC/QCh8F3dsKXs0P0udYAdewHDU51idEZs9xeeVclzb5s7z/3tZC5CIjYx6nv+2ryMi55H+f8083XtnpM/fJ8H3LwkyUsrsMFpL/77GmDlm9i7p1ssaHJcWJyp0/tw58859eR/1HRjey36QHvKr3FKY4jFGa6abkRjZt7N4Gqm5dlyEsk3kNZtU4+PHWiQWV3reK8jUkhWyh5FlqW0QtMO0Uc3dtLILl73EuIWBFQAKaUZ2TgkuuPR3kbe1XJXHbG46nWgH9hUUkj7zczXj2ZUU2maDHvMqF0+cq/VBeRq3/MxF7H56Eat3p/HCZ0YAR2E3V9/Me8H+7MWIXm9SHCPoKb4Szy8BluINi12W72UHl5xKXsVQpP8gEfhMju3wE+jhvvN27L8UihoIElZ5QRNto2ztU/XSgWP2cpTiOz3CqGLmvzVOGlsnS8U+1WG6t8pOLrfeyaW7B03uwoF+ihoIrdZjZ/fSOUpcJLXTXHuEtbsRl7ShN5aouOt7csf5XlkbUCZFtqqbfK/r/5xjZN+mUlysY/fTi0h0NIpP1OHORxs9sVw/ck2QPaU2/2KF7fFOBCCHOCl973z0taNVg8tB/P3cI01QFQDyE0KywaSOI/RdirvQCpoDQN6lFTmM0tLjrCfVPdb+FKG9eJOnG1RsMlsj1/5K3aCUHN2qMG/5FNmV/1xLAVrjhVt34ebnJzF1Vwnfzh8Cago5h7LUzrfI9rzYJqKIUe6+chv3PwdM33Qph/W4z1pTi53fbRcLSldYQ5htU9xdurvY/oOc9HmHRFskf4JptvgG7XciL8jn8D3zuyBja/G82d54ibmL5jDIvgyfaMS6QKeH9j3/bkLWoSOU0baTn6X4XcSjD5gMQOnDNlunZ3Vqkrz8vX1mX9HGnq9R0c/3Fl4EAIwl5A0V7cm8sZshSwVew2p8LK+yKIXf+MMPorRUw703no5z07eBxGr8NhbtKCz15+If+jaRTSG7UqoXwH8FcCtIz/0ZAMcA/DcA4wDOAPiQ1npxjVMAIDUspTxfzfPW5bgOqtPrqv6hMJ3fetfaJRN6oeuhlFo5RhwgANDSUjRBEi60kT1FTQOAR3O3Ba/DRSxuktTGlsVq43DqbIqdYqLGy70nI1JUPU68WWiRI26pSc7BuVX6u1I2qrPHyTnFY6TS5rh55so47fsKn6vRMa/BmT5SLf/un+4BAAwdDd5Hzw9N+fv+9oEvAzChNQnXiYr+XJ1U0VWLUrePWzVJLbwU3oTrzQHAlVRdFXwOor6bMJQ5RvjV05xOKilcVS2JM4mu64n6XueKpKkWFahIAlMnb8y08h7aJ30TOeDeOswNPOW5NPM8VjMoMVGktv5QDyXKSB17o2XOv3ScPnNHiUt+XLjy+b0Ulb1jmapXm4PujwD8g9b6MIA7ALwE4DcBfENrfQjAN/jvWGKJ5TUqGyK7UqoHwNsA/EsA0Fo3ATSVUj8C4B282ycBPAjgNy7l4jZ6C3KHUV/2aa6TjCPHSlOOcMotHS8sqfS3JPe0Qs6ejLXiO6FtmDde5ObMhP//VpGdSU3haKfPXU5VjXL/Ce96i7eSFipOpIpFSTvHrC3zddrOMqKXL5KjLlE2q3t+jsY7/ASFzdyLnESzSGHBC8x1frpt0kEbbXolvCw3Z+ynv0tnSPuof9pwnP/6Bz8IAPjwfirRHOQ0VmGdfaFC4alzFZNgIgxE49yFZjxD6afjKdrWLdZXYcVJ+WXP9Lk8B3eN1GcAaDCSC2db2QsyAtshxFaIX+90leeDT9sqmX3nb6fz3MUc9sJm61/X6/5JZTgZSMKXRW7bk8nS50nHKtt+hi76x8NvBwC89z4qiBkWB12EMiwht84G+RKbQfb9AGYB/LlS6iml1H9VSuUBDGutpwCAtzuiDlZK3a+UOqqUOrqwEId3YonlWslmbPYkgCMAfllr/ahS6o9wCSq71voBAA8AwC23p3RdJ/zSVhutJUFDEjB8O4pXYFl97RXZXyn9LibM9cXHtjxz/hVN6PhyYxcAQ4AgK/Nu5uk+nJr2jxlJrM04CwANXkhtxtJOhrWMVJDnvsmImU2Z1M42j2+lxV1FpCVxm9Cnyls7BXalSfcxv0ThGwmrJZe5+8qKWfoz3Kq5nWW03k1htCYXeKQWmbM9a9mkPXT+d95JIZ/+e8nGPlslW/LZyV3m/udIq/iCS4Uch/tp7oa5tHWVx79YN+QV8yt0zIl5Qs89fYQRt/VSAcvulHH7iK17wCUbd9D3aawtYr+uclqrILq8X82IElf/WH5YZS68ETdRzWKvHb+dGIvf0ke2us3MCxje/ajQnnwm75wges56JzS3697z53SXv7X7RwEAn9z/+cB1Nhtus2UzyD4BYEJr/Sj//d9BP/5ppdROAODtzBrHxxJLLK8B2RDZtdYXlVLnlVI3aq2PAXgXgBf530cB/B5vv7iZCzrQVhdUgyhCBHGsTrmDE3Wy84TMYIW7jtge6sNFQpLbc+QNF2+nrKQXWsZWfKlCiPTdi5SquHCBkB1cXjq2m5D9A6Omo+Y781SauctijwWMV5h5JFDRJnVRyCmqO5iKiG+2Wae/Vy37TDy4GT6/IHi5QciywlRTXsesyUJA4bX4Mx6DlLq2ey2jjhNL0kvc6ZUJL5oFTuFlhUQKMgCgUqVrS6LQaJrQVbrGJK2oyOow7TucISS/vUC+i0NpSrxZ4aSaB5M32HO6hQAAIABJREFU+cc8uHqQrjNDCP9ymc4xx6Wuoh0AwA15oxGQEMJLeqyk6dq+mVbIbBXve5QXPixS/lrne2/w69PoMxd4Yy9FI27NEDPsZJt2+uYicb/vzFDa7w0ZE7UQDVRSdUtMf6W4f97+kuld98RB8ons/RxFK155iEqmH+i/AwDwkdJTdE5l3aj4qpSCXicpbLNJNb8M4NNKqRSAUwA+BtIKPqeU+lkA5wD8xCbPFUsssVwD2dSPXWv9NIB7Ir5616VcTOLsIrOdov//75Rpxf/WeSoRXb1INmmSvcvS8bQ5aJB9/gZCg5U+QhDp5im273TNnP/MPNmc7eP0WWmKSQd40Z6apRX181YD+MFxWqVLzhkA3dRZYrPb9pnK0upd3cFkhbxzm6mraqp75e0wWgt5hRA5dNrMNd8xY1JcHplMh7QNPtZHfACNPrZBx9ierwR7yIk/wa2Y8zemadz/3CJEOTFItnU+RWpAKWW8z1LYkU0E+e6l9Ncn9yic8b+bHCK/wTPLFINPTjKyr9A9P7Jg2HOnd1EqrTPEORHs1R5wyI/QUsE5AEyM334mgClpNcSZ5thwea1oWLVd3N2nZeanwMF3Ke+VtOJHv0mlqQKs37zTEI++YZg0z/f3ESqLnZ/i3AspXwWAhw/T78Ar0piUF0zh9SNEVgRC7Pf1uhrZx8YSSyzXucQ/9lhi2Say5Uw1Te1Epr4utcgZs7pA6kv2Ag0tyTTmfsl43qhfov4eW6Ya43MLXEMuDgvHqMz1Kqn4uWVWf6R3H38vrYMvTJpWPC8Pk1NPGkRK6EfSZRe8YL01ACRTQVVKmjWqOqtaljNJ1PZGkjnnZGh+X0FO0LGiLAlW/cRp1+ZadZR5vsrW+i0ORGZIEVMofPN2HojiZpNJbgo5y+ZHY4TUSFuNX6wze02T1N9TzOP+ZIZ45g7nyVFXsBJP7uk7CwCojZOq/EqT5jh/lq6jp4xT7uRFUvWbd9IE7B4j1bjIY+jw/CUsp5SEZpf81sz1wFbYcjsWzskxp2pDAIAz06Sai/oucwIAZ6r0ndNH1xTOu9YYVy6+THOx8Ip5j/5hnkyTuQNkdh7Mk6NRHLOPL+3190WZGYK4SWb+PF1nrm1MUiA69OYiAbVuCnosscSyLeSalLhKMo3NHVZgZ480fRTwlxqK2l7ad3DQ5hghEURvnaEVtFPkmuWSqR1PMndbfYAZSxqMYNVQFU2ne2UMM642fYZP2r7cNFQjkuTSuyQJLcFzqapZkdtNRucUI2+mbY/Ex6tMxsyTsM60lgh5U/PBFb5dMCin+7iIgue0w/eWzbNjjZM5lOU0bLTYkVgjTWiEu6Mc6iU0uiHfnU4hKc0Xm+RQO1chVHumQ7zrS00zCeLUG8iQk03YU8+nCDGTsyYcK47DiScI/R8qUnhrfMfDAEyoTFAcMM9EGHCkXl7CsiMJcqxJnT4AnG2QRiKaSWeeue8udDfjXOWEm+f4mafYEXj7XtL+XpjaT8detDrOnKWX+NmzxJ9/dC8huWatqfJd8wx3rfK476J5aOejQ2m2M06YkTcKvcXIHkss20SuKVONnWYqZZ4iErKq76AVrG+YVuS0leCyUKEVvTlBtlBhklNsB5l7LWVQIlOklV2PkhFdztIKnZ7jkkHunSbXAUxCiaThiq1ueOpp+9DSjeY+LjJrS7iLrtQxWPaf4iQXj5HXYYRttzjcyH3R6jVzMm+B/p+dCSJ6fZhX+oLVPprt+o6E4xjZK006tsZ85k7CoIHL3WNEE5IxSequXyQCU7Aj+4xyW+Ebc2SrS0ryVMKg6LElSp5yORV5mZl0+4ZIY6sWzb3WV5njnef026eJa/57el8BYMpjwxxytpQ7pFVIHzexz6e5jBUAjq2Sz+csFwilFtgXs0z3ZUdLJcQ50Rzg89P4hzKkAXV20px4S6asV6Y3M0vHFs6x5jDLGmjZ/A5anMpc38G/kXEKXwqPvM/zt05r8bUkRvZYYtkmsqXIrgG04PhlpB3Lc5hmZO/tJVtucT+tQ8V++lvsS5tjrXqBbPT8BU4HrTOz50XxoloosYs+S5fIK6sGaFvP0Eqv2G4+PGBs0gMp+r9oInVZoXklXWZEeWHelH2muBCFzVekVhgd2LPrWMgu5pXHiROC6JpJJwSj3QWD4qkq3xsDeauHk2xWuDfbhIVyfq84vg4rOu0s87JluDTSNSjRlr5mkuDDCDXpGO+yiOKyS4lAFAucFsrPqpAilNtfMAkmRwYpzfREmTzfgpQdLhTqyRvPfYO1DM2XXl0iTW6iSR80XWbNtcgxTtRIc8hz8ouUnB7K0rM8UScUP1UxGsrZMiF69TQ9NKbXR57JPlq5buRcaJM2+eVzNwMA3rSTogxiuz+rR/19qyHNqnCS5rj/RbpQs2SpgdJ5Z5KO+dA7KRFHOuQ+3qBjvzdjNNCGpnnaCOFjZI/lmoryNAaWVmPu+i2QLUX2lk5gsl3C3iTZwnZKo3Tk9HbT6nS6j2wisRUnlsnuE5IGAMhfCKaBahX0sOcMESq8FKNlno4RD7fYpoNF0iDu6DFEFOK5DTPb1nn5LbGHVwpWAL/2BBwO9+2yOpdJtnMWU6nPXsrpq4wATi24BttpCVJRmWZvf/EcU1tVWFtyzepe2clFRNyfPbOL7vHGQULaPXl6DsMpgxIii5zY0GRkFNKMk4sDZp95ehbtuQz/Tdsl9kWkWNs4kTZxZOHVz8w4UNrDX/zVH+PI+TM4enAvPvzrP4fcLmO/DhZovOK5f75NHvAqp0XPgJD46eXd/jFTFfrsABeXCDHIjhT5BKYb9P3xxSH/mHlO0U3W+f3h3Ij0InPyD3T7BIQWbFcPzZ1wzYsf51kYZE/Os2e9lxB4dR9tz+XpHHu/YnrVdVjbGniezvNnDxGJxc5DFLX47YNEBSZoDsQ2+3UpyvMwUClfNyjYX6ngyPkzSHoe7j5xFgPlysYHxXLZsqXIXvNSeKa2B8W8lPyZ+LHYx9JRUzy8R+cpG6s8zfb5aTNkV0Lu/O4n2myLsgnUsMo9W31M6sh2YIm7nUqvtFtKVIxwo1WaKFL38wIIPZcY7V5ucj+3k6Z4IzdPY0izJzdZ5bwBXoiTlW6bveMXpvBW+rSzG7etHSjPw1/98f+Le06fwZNj4/j4m38BWjlIldmrvY/QY+WApQYM0r2lOX6fzxBqVlo0QUIXZcfB9+cJEaWvmtBGfaCPbNHe0aq/7wLbkVJKLL3RzzXIpha7fKVhNJ862+ZzA0VM6CSOHtyLu0+cxRM37MH8QA6Hc6bE9a0DJwEAkw3ynO9njeREhc672iLEvVg22t6eEkUEFlgTEQ+7aIgZ7r4jxUYA/Kw1jyMyaaouhWJVbnWP2bWf+eBF0/nQyOMAgDzPl3SnxbJ1fn7kaY4q9L1Mz2j+/eyfOmW6+Aw+QuqoM0QaSOEMfXfHmyhT741p1iDUpeN0zBv/OpGBSgV3nyYUPHL+DPruqmAhW9z4wNeyKIUP/9r9GChXMD+UjSmtX2WJ1fjXicwVCnhi3zjajoMnx8axkClseMzrQbTjYK5UjH/oWyBbiuyVThqPL+3FvjSp7De4JswlaY6SIDFVJzXmNLfIyZ7jOmUrW1Yxv5iQ1/isMCVu5rfHODF6uJBjR5GcIeKc2pMlnW1nitS/utWE8OkGFWLsYodii4sPfvfl9wMAll4kZ1V2zryoWSHVZLtaVEEZo0XR7vOSJ1c5bZbHL+o7kqzOD5KK+JH//Wcw3FjBfCmP5iqfkA9SOQ4pWhkgmkN6jZrUx3MjQVZhp+RY63c23UfagjSQHC/Q/MjzGXZN++WUz6lG5oIkuQiLS6tE81Wxkl7k+b6yi8KVTyyQjizceoMZ46wSttqTVeNMA4CXZsl8anGoUur+AWCWw351TvvtcGLRLDe8zLnM6NM2OKczdB+K1fnMMjeSzHK67CFjughLzvEymRYf2Udmh7RjeqyHE4r29/jHzJ8hU2Lvlzk5a5ISZebupDCh07ae2RT9JtxVumb73WQWSC28tBS3mzgKg+5GjrpYjX8diXYczPdeH4gey9bL1jroWi6em9yFAwViQRktGSZRWf1fqlHRw9ELtOJnjjPDCDOl2mGoEBmJz8BSHaUv8sPGu9ufp5VSwjg7M7S69nOHEknMeHxln3+M8N+dX6WVWQoyipQ/gT0/Rk6rM88axtXhowTX9X5ORWX+eAkHdqxmjaKluKucxprg9NYcc8ynuUAm2V0SnO0lJG82mTm2HWS7oT/4/4z2wl9XWSSHnKokAtcBgLOMlv0lmhdhQE3yZFetPGBhTxHWF4+tQkmWSvmtnM2DSnCp6c4Uzf/t7PiTMOwNWRMvlVTXqRqh5PQqLXSVZfpcNBc0DEpf7FCINjxny1U6ppXmcK1rxtRoBzn5nCbNV4M1k5YBdj9F+HAfIfDvzxNDjTg0hYV37rxJx93/eZof9/FjNO4O3+sf0zG6Yi4g3+kGfcfThL+Yvg8AcOPuv+XrmTE5Po9+XOIaSyyxYKvTZTWRLog9fiHf17WP9MqS1FFwamcrz2hR7o4xC6KvMklDeoRWSkFzwCD6CKcZDrvBRJKpJiHCS4vD/mczi2S/Zv+ZxnTgETrm2C+QFvB7418BAPxH9V7/mMbXKfEj4XeE4bDdIqOrRYUqyUBJJrPzkoy8Y5wSy+iTzgS5yQFjcyY4ZVW2ts0uJpzrBktn65wS20pwqrBF8pHlLiV7ekjr2pcPdm4pWh1bJEzaw2gtiJ4IlVnaXP/C9+awinYoSzavaAx22fOTNUrGWWV7fpGTXyT5KLHEr681fo/fBbmnVp32cXh+FrmTTqdp7HzFmoHDZc+ijSmGz/G/Ns/s66ffAACojdM8PeQSZ6IzQ2MsnKN9b3x42T/GOU2JWj5qd+j87XMmgcsfCz8bJ818ihwVlWSdXqcbn6UoJuGsb7PHyB5LLNtEthTZ3WQHOweW0cN2jxAKAKaD5nCa0PP23WTLncqTx3txhlA2f8LYjC47bms72J4fIYTp59TX/oxB9l1ZWmkFoUSmW6RlzDUJNewSWqF8EkqsDndOUQ1ChRcblBL5vpHn/WP+83sodXP87zgdt8pdagThrcpUAWHxxmY4Iac2EiyMsT3HQrUlW2584v+dsbqLJBnNhP4oywklXpFtPA4R9FhUU3tz5H2Xziw5fkbiaY9CdpEwJ7sgul3wJPa9bMPnkIIPACi3SYOSclgdKigRRPesQh7FJcCtvqC/QuZHnq627HzHCyLi6iido8LFU7u+Y1J4x/+a5kWt8MvHqCo2tm6zFpU0Py3d5HsUVO4E50nQHAASg2Tz637SNKs30fz/291/BwAoOAT1VW3GlOMc7avR6y2WWGK5DmRLkT2V6GB3YQm9rvTkNiuRcIJLvBucHCaIPDNAHzxTMgUGlXn2KnOctJCjVVBSYHfnTExYiBXkmhNN8hfMNIJZaHnXrJgOl25KjDxRoxX64KdoZf7Uk+8DAHSyBhmGFtlvcI4QUrXp4HaeYsUZK4Lgc8pnpWML2/DcoaXDRTtNa+VPZ5n33A2SfWT471LGoHSeKaCECqo3RfMiRCF9zOZZShoNqD9BiDUqPcUtgpGwiN2dCYVF/H59EZ13pVvrJKfYLnXIhj7XIA2uZfUYL7k03pnyGF8weK5Ogb39s+Y1FhonXwtgRK8tc8qulBhbhB1JjoZwc1WscjFW/Qb6YBImnXjoGbpW7lF6vp0FjigJoUmO1MDWnQf9Y+buoOPzUzTe7DS9p6nTnGdikaxc/D6K7Ax/g6ISmolKpb5JEN0mnKxzUcxG/d9iZI8llm0i1ySpRmKqMxY97kByNbDPSjvY5UVsyeyYsfFO9lB23VItaNsJRdBYZsHfV+KgJ+uEsM8skIawyn3VCmlabStNyyfA3vDaMCFYbSet2oXnaNXdcYqKZvSy5dl3pBcbEwoUCLncJbrXjtUx1UtxhhbH3jXboqLcCOK3HfOYPO4EI/cqXUVEm0lZvfBSnAUniN7vki9D7GTpOVZKmHyEHqZ9ln55S7yVLi+2n0WQQvqoCU2zj/hij1smqhA9rniEdifrlEUmURjJbQCAqSrtK/TiyTkuFR3gjLEVpuC2SEK1RCXEJmcATy3S382+7j7nUpwkSkVtP70r+0dJCznVMt3IV5kQM++y/4a1Lt0O+h6clnkOS7fQc2i/i97Li0wz5s2RxrL7m0bLGHqM36V51kQrFB0q87zkVPf4RVroxISTscQSS/xjjyWWbSNbqsYraCSdjs/p7VlrjTjOxEEj3OML3HVkV4EcdcWkUSN7MyYMBAA9aVJBD+SI1UMKbgDAY2fRhEOOIamXz3JhhKSF2g6ukQKlgw7vIWbPJw5RWG3hn2i765/o+8RJi9G1ziy2ftojM7AusxNMmdpl6TUsabOO1FMvBevb65bTqpWnR9YU5p0OM9Hy/dgORnHMifpe9FV0bvjIKcKdCJ3QDbWpFlywi1rCYbMW6+vhpJrjTcPRd6ZOpteFGqWTCv/bUJbGKE0VAaDa4vpvZp5dZK+tpPlmp2lMjT6LaZXV92Q1lCrMb7qo8zavQPEcjbs64gSut6dAzreVEVOP32DTUUuhE/PgqUQQN50XTvv/P/hX5KybvpfuuXMTPZfCGF2nXjLsu8UpMh30EL3/pRvIFJWkGimEiWKX9fR6SnyM7LHEsm1kS5E97bRxIDeHPi4+kZa+gEnWkM9u7iHn11SaVr08VylI2A4AhphXrNNDa5YkgtzI/bckEQQwaObmCXH3MOOHIH7a6U5J7eExDXF73p8eomMfHiee+D8/8mYAQN9XDW/84HdJm0gskIPFW6YxqjkOxRVMGEe1CFHcVe7ckiPHTSrrN+YFYNKBAaDFXWQadU6rZIdU1IouCULC7LIjzW1/Ge3KLUIsz0J22UfmUpxt8lxsB50/Jp7bBY8SYqRMWJxwElYDgGNlcjg1mLFmZoGSmhZc0niKOaNZVRs0H9IFBxxOSy3JvNDHdsuBBIfRWkUOY3J/O0nASi9IybFVVsrTXd1Jn909QJrhaIae4WzBJPqcSRlWWgBr1+G3rOSmJ4nnfvdLrEnt58IpdrwmTp80YynSPJz/AM3TLx/8AgBgKCFdX9bGZ0etVwYTI3sssWwb2VJkLzgNvDl/3Lf1ep161z5jLnfqLHLnlhyhg4TrEhEdYEUkIUTOa9uOHTmeiShGk4vYSDKc0lni8ea50GCs9AQAIHcHodxn/uF9/jGSRunbdBmycVWWE4AWDfuGqoVYSxklMvPCksvpsjkr9TJJaNfo5/JU1lSaCfq87BmfwEIPF51wYkkpRzajrP7lejdr6kiRxnc2QzbjYIq0sFaOrmdrSxKOq2u69mcmqEhE/AYSAp2tG2Q8NcfJM1yaK+y+QkRhc8PVKnRe6Wzj1DnZiKOjSd9lY/BMkmpS7PeQlGSOMiI7L6Erc8wy9yjQu+neRLsRH4fwzgHAyTWaz8hz7kqNBQD234iW5zxPSK4kmWbU+DSWbyEfRu1OGst9WdpXbPX1Emfs/m9REiN7LLFsE9kUsiul/hcA/wpkGj4H4GMAdgL4LIB+AE8C+Gmt9dq5lSBU7k1UfTvZ9jLLZ9I3W1BZ0Fq0AZs2Srp3CtWRfFdOBL30tkjBTdgDLeewSyx3ccpoCaGCDz70RwpUAPPYzxvCi0ePkP1eOEv3MfQU00WxjeguGGRUTD1kCiRoDAkuntGO8K9bFrn8l+fOY2526TQj3V4AoNYMruX1HmZRFaKLRjdKrJbJrpQuL8O9hEYr3L20bSW9DKbp2ZypEFrPrZJWsZygcwiza9uijaqv0Hkc7iffGuLU0XS3z0SIOCRlNMG87l6KtaZK0B4HLIZevnV2DyHDdGEu8+svj1te/5voGd28i8pte5LBqEXaSlTyry2awS6yrev7SBNyV+gnkJwymqOQU6h+JrTgY3WW3sXlw4bCqrKTk7K4BPeRGtFSfX+e7P40X9ZG8Y3SZEU2RHal1CiAXwFwj9b6VgAJAB8G8PsA/lBrfQjAIoCf3dQVY4kllmsim7XZkwCySqkWgByAKQDvBPCT/P0nAfwugP+y3kna2sFsp+jHam2bWlZR8eAKmYHY6uIFbmoz5BkuT5V4sewrVFN2x5kwsYJ43+W6i9y7S9Jq7eMH2Cvvau6swiuzcAX8pz1f8o+ZH/17AMBkh2LCD68S0k9z87d/PH7Y33fvJ8iWzbxMkYfEEkMU232JHN2X0zJxWCFBzE9zkUyRSSO5bLLZYzSWZonmSlBOJ4KP2//cWvKlc05ziMYw0SB0Fr73Piu3QbqqLJeZ8qkmKaSMotyttDlonoN0sfWyTPK4SM9Z7H479dVpBe1u4dOX+Lp0xVF2SgAfnloVb7xEPOh6lREaY2XMvHvpXLBXgbwTOdC7kLdyO/y5kth2ibTLmSPsX3DYoaCNn0LqjPyAD5+Dm8qgusfMT3GUSUO4G84fPPcuAMDqzfQu/GjxWQBAf8LKvdjAVhfZENm11hcA/AcA50A/8mUATwBY0trvQTMBWP1uLFFK3a+UOqqUOrq8sLlBxRJLLFdfNqPG9wH4EQD7AOwCkAfwvohdI5N3tNYPaK3v0VrfU+rfnG0RSyyxXH3ZjBr/bgCntdazAKCU+jyAtwDoVUolGd13A5jc6EQN7eJMcwhTTXJU2OmWojKLSu7poAOtwCTrdgKINPgTkWPEUddSZnERtV04zoWBRRyDbpiqFsZ0cELrWHiFtP+W5IcBbgp5Wz+1B6pzKO7DA//s7/svWz8DADjwp1RVlTrJNcxMP6O491mqbkwLzaq9WyLVOVVg9Zc59NtL5p6rg/x4hYS1FKzuSi0H22UBJoEntcyOTA41zdbYzJo2SSUON3B02GmV4Hyh3EU5B/cCaJjXTFTuZh/P+zK3bM4JR58Zi6jvkuqanncC5xXHZXrJPDuHE5VSS/TspFHi8kEaXH2ATYP9xqu3u5+SZ6TOPxzmjXo3xKlaH6LzVg7Q+3XTQWJYsvnv+1Okx8v7udSiY2zHn4iwB0lLrtkU6frSIvo8M/m4ypy/6FwlBx1IfX+TUiqnyAX5LgAvAvgWgB/nfT4K4IubumIsscRyTWRDZNdaP6qU+u+g8FobwFMAHgDw9wA+q5T6d/zZJzZzQU87/gpnI7MgtyQyCMuJIHmOv7dX2VIoxCbonWF2FVtzEJQucsJNjwqmfTZ53bOZUorsrMtzkUzqEloUySrqcnpjk517h12ToPGf7/srAMAvtf4FAODQJyiMk3z5HADLLmoYZJcRJBj9E8s0B16OIDhpdcxJz0tSDnfT4bbRkl7qlrnOumDu2W/5zPdaZy63/AU+1oqQ2dxvgOFdT9Y4UYbZWQvnzX7GgUjb7JwOfm6dUs6XYPZd4XPPMJKnl9iRWTGDcqp0UKeHNKD6EDPT3kzft4vc5NLija9xwc3OnEnfBqwUa+udEy0DXACzMk7b/ABpcj828iQA4HDaKLriFJYkrfC71rEwV0LAR6sUzh3PUZLZmwvHAZiEMftdlKKYq9IRRmv9OwB+J/TxKQD3bub4WGKJ5drL1nLQqTbGUvO+3WxzlMln0j0kzF9ms5qKSIlsmEUlvJICJqziQnjTxB6Tc5G0LGSRBIZLQXQzNt5qGUv3OQ67VM74f7yVih3+aPidtO9f3gAA6H2YWs94ASYcISOz2pQASDD3mU6bZBHhv0swCqUv8Jx6oZRj13oNJKTEtm6OmXV0UkJ8VgoyhwHhCU0uHdvsYeaa1W5WlQyHyxyeaKfB7DZtQW+DoprLRiVRxmnw/VToOasqa2dWampjjPxBjT5m2jnAacXCtMttrLMWC6+USvckg9pe2FcDAB1myWmMk+9iZT99fusA8+rze3yjG/W+kjT5nYh6r8oe2eIvOhTcOpimRJ+xJPkVRpihqGmVuLqbfD/jdNlYYtkmsrW88aqDXclFjCSWu75z1ihwMbxm3ats2GMvHTZlRU5FnFNWt4SK/ty1PpeVV77zQtvLEfvYIiPh3dz19P+55TMAgE/9+lsAAN/edTcAYPTTx8zxUjIbIksQ0gxlobTHiTY+h7loBcngY1cWMgjvucNJGwlBTUGSrCFy0Nm0nICux9qAu0LHOnXmULf42MDahp8qzEUi4A4o2mJaVVJU4gWfvS6yFsN+ivouw2W4zDZ0q8BaRom9/BnaSgnteMnwE2YSXJjF5dOSlCU+oNWOqX4p9NE+Swco0cnh4pk2a6KiXdrPuaA4OsTJL45P8sGRCEuDkHfivYUXAAAlJ5jI5RfEWM8s7vUWSyyxBGTLaakyqtPdfvUyJQrtbWlG8JZn1imRBa7e6tf0PaT0d8JfcbvHLKt3if0SP9p/FADw5RvupCN2mf7kqkbI5JdSyudy9k733GrTNoa2rAUIqgZGJPtIsYacjz9XdassWSjThUnXDaXjSucTG60zmcA+vqbgde8rKK9dsbsZrQcoBt0scux/r4kmNHuZN57zHdo9dN7CKPk9hGps0CpblQ61Er0RLVO85QEu+yzd//QNnMLM2uWePE3G/qTRGESqmrn+OTKTU27XPiINTkrd5ZNVXL1EtBjZY4llm8g14Y2PhSSsY4iPYFeC+56z5/j/b+/KYuy6quzab67R5aqKsRPHGeg0EJBIIEiEhAY13WIQafqjhRrxwQf8gRiEhIj44hPUYvgAJAQCgRAg6DB0hEAoCcIQSIibiA44CU4CxIlT2I7teKh64+mPs9a95566ZT8n5L1XvLOk0q13x33Pffftffaw9rkr8hLIuXPkMD/i6a8yDV8v0RbS3LEVQKKFrE9ZN49aZDO+aqRRqHnDfbN5fOwNpk+gwjk8WvmcdzAvLil+brHstqm5dsiRT19AjTH/HfTOcxdlw22s5iOprL7eKglHlv0YrrCj7wo1eiPgslpljazyNmLN3h7kMu2e8xZuh/bxAAAPUklEQVTCmvM5EUuL/rzvXLnby8oRPB1EPJZpFakUte1iMs8cF4qVPxckzZ6QMCVIL3tCwpQgmfFjhH5pVcjQdd70u5Rm675V7/Q5tbw3O6Z5iTfp6+sM8Zxgr6jYCYeA42wr05CmupVst0axyAiUrRC20/HkCRzM+aVrFb9Wvfn8XIMG06DpbGsvFacLgyD22aXFr4aLG6s063lbWRPHWpDow6Kc5iJbTc9607xLfv0aTfSdtZBLT86wovl+rr+ZcE5MPf058geQhedzR3zdudJbb1p4ODvmWiZP1ZnGPWdiZWJYMDh/MwqdDZsKOwySZk9ImBKMOPS2uVtIjLLuJM8WZUk1MbTHML96dYihptiVo19yT3F6bNmvuCCNrvCcPiucc9/S5dm+nSUy+ByhE2zRJ5QoGQZusyxZUk2syeWEKwl36TzWLTqTpL0BYECN3Z33x8TMOGpM2Q99h7x0X80slXEbGRIAwNontBXl4r7deYbVVihbMKjWLT5JMegqZKYS0rC4JQ6xtQdFZ2cY4u1Qs88/xq40v/bCPUzm3ocpy//svDk75sw/eDmvfKFPfX3HXl/2/NqZQwCAlaB9dKzJw4SZrVDlPv2SZ1+4j/NuTUhI+LtBmrM/CwyiX9BqiTVSpu0vFvtmfILG3Uv5Os1xm7v9yuoJMpee48S2F2hiafRaUSs4Jb/IQpkPQmMNlbIWk3O6i0xNXQ1YWVepuans1aGlz551mlM3nw78CBRPRpfSWZX8Ujud79s8LjZZnpcpr70lFsTM6mSbx1rttmebfp68a5ac7VFPwRAxkUnZPk+c8pq8dZwFTqRaq6zJZ8K01ofy51D9WY/yez6/z1//7wCAT/+Tl+l9L/1Ztu8t8wcBADuYTBOH68o0/YU0upA0e0LClCBp9otArK3LNHplyPkTsLnUMdYj6kIbkkRk82ESUlRP8yhqb9FW+Y3io/dap7/Tq95Bk3P3zuYS1O4iC1KiW5NFIWorAFALN3HV9xZ4vh1MaFnw1sa51dxycM9QTXNX15JGpMYNPetNpuiKnkrjwBW1OrVdJTiG3WPmW37CvzLjLZ8ddS+LmGLDOXvcZUheeVGkdYKkmmfWPC3UFcdkVfAYluhmGdpBpVV3h4hFvJWx+xe+EGzjEf88Pn/dLdm+337dDQCA91yx3+/LHnu7yUqyxGSghcCaaZqXr+16GJzHokyaPSFhSjByzR4SAgz+hp73cWKYGGhMXBBC62ISgisbPj7b3Zlrnl6LcWrFq0kuYSwW6c/kc2ppcBFCZMQQ2pc95PqN/Dd/gwzAoolS3Fsc50F7drRXmI7boHzUvLqLjDCkmc9f21XKJ40u7zn54sUrH56oP6sYP0t22cmmw062jVaeDtxokDSSVGLqUb/OkIDKosPv3kD0UFTLKoip8qk93ZnN9p39Mym+zpCApS7rQ+aHX7jgWfbYJw/kxq+uexkbJ6np78mf78Yhn4b7yavf7g9RhTFPJz9Idzm3TC6/ynedvXnXI1jr3YmtkDR7QsKUIL3sCQlTgpGa8RW4QqJLWG8em/QxC03p+WgmXihRx1/bQ36T+FfufDxzW6fmuMLifMfKVB+G5Ub7ZAy4C7mZ2puj04g8b7WzdNStc5/gNtQYMnO66aYlb8Ynn7sG2+Rz77J7UWcxMhsbwR2oHfJML760P7bLJJvgWRrDZRWa5H0lsMjJ1gqmeTTxrSEuPX/MoEm2mxJVJW45hdoW6JArc8wJWidmGiXViPX10ZMr2b7zh+mMnCm+OhkNH52eascFAJVucTolPr9sey/fd3aNcp5RFSBDcGxfJdM/xMYlnu/gzvkX4Jmn7t20PbvOllsSEhL+rjCxobcLsdAAmzV6zD33fP+S5Zx0F5ZVlkPooIutia4KMvhZnHrzizlT6aBOPnTyuTfZAcaoHQaNXEu7WjEldYPhMzUiEVNsCaFPrrVVH84mi7UduZXRbBUbIkqTGz8PBkwACTja5UAT1iMLbnYuZ8KRRbCx7i2TAZ14FaaXVmtMnGnl2m7XHGvTo1DbDHnm1J8gDrcB+fdJjjo59Z5+MNfs+54qcgOooEdhQml2V88HtXaG48RnJEtL6cQhamfpTO2p6w0/s9NNj07VUHzx5zdPAtXO1rZj0uwJCVOCidPsw8y/hZjXe1QaXSgLo20FafFhOOjFVdYtVbkevZY/zzq7vMxJowTJHNI6Cp9pye7RGNSL83Mg53Drz0hDMZzGuXY4/24z9DUYcG6tUFukvXu9QMtxrq65dWuH3zeinvfnZ/moNPqgo8Idr+1mKJNYaABgvs7+fOYKy9J+bRFOMe/3VM8vDxzzBUiX3JfvUyV3/YAJS9V+9B1QJDHQ7PW2v9cBQ55i3tHzUMcb/z99GUoo0ml5XfX0Cy2HfjOw5ipbf7+SZk9ImBKMIamm3COt0tZYs8fry7p0XOz1ga017Pl+/eJyWJ1jI9DwZZ1fhsVRpnp+86TvqvWNX98IAFh6IOBlU34Gl0pfra+zgKUTeLNZRioNnwUPlNLLzFoVowBAb5EapMVSUCbENLiUNgdyja65s9JWlbLap2YONX2lIi+8l0EaXuQSITSSzRm/T49WQYXdXWo8Vz3ohqrvR1zKGndmDTsOKYnmGE2cYx2/PLF/NwDgsic3d3eRhpWvxCINX2/nMmW8+RFLrrrihAacrC1peEVZpLHF1RdCloLm+VshafaEhCnBWObsZb8wFzNXHzdiDV92P5rPi4KoW3J//9teBgB8de0m//nnLwIAXLrfa8KXHPIUR73VfFK99irmrVLtad7XVtx9I9DsVJZVapD6ejy/pNUUzNn7/aJlYtTWmo9LqwJAj3Po7jr53amtq4yDq1DFgsiKrIFONM/XVDPct0GLocb+Zl0W+yyyyGWu7r3wSokFgFpFXVeK9xp29AWK5auyKs6Shmr//pcBAPb9hucNtLZSkOW6qGUdZnmv3Nc6uTUzmCXJx4J/3TrzLKVd32zjKg1aMfnMciDtWBarL/hm/DG9uermZIcASbMnJEwJRk9LZfkPZZhNt1FCFDBpEC2VSl03Srpxxr+eT1JTfu2E799228Hrsm3zd/sCi9X7vTf5mifY01tdX7osyGjl8+QKK1L6fHKa72nuHlJAMbS8aS4n5ak5Y3U9UAecXjrKrfl3p+MvKG0OBN7xiApqQO3fHWzOoKsx5j7PeLq26TsRtrCTRu/TCmhQgzdrvcJSc3cgz02YYXP3ZqUYGRAWqnk8/y+0sH7yS/9s9t1BP8XTfp+wuCjPPyhqZZO3vE1rIKD6yrzw9KHI0nIl/h3HhyOfAKTZxf2vY4K5e4XZerWzOG+KZtLsCQlTgvSyJyRMCcYSesucSy5cXzQ1J7HWXU62OJnmeODU+r/OHgDAlx9/LQDg0QM+MeOyu7xp+I8H1/IDN44ACJhhyRnnaApqvW3kzqX6WRZizIohlhvUTimwOCtRHolMz34W3mGiz+ngHhfoUGRSTZUmucJqheIT3TaLYyo00WtMelFqbOh0k9neFMuMUm17/t5lmofbqkqMoVmfJczweSgVNvy/HjVrVKhNobdH1/NmmT+6l+b7Tyk/01uVelwoatGS4TS1pa6cLYbnejvnsv/FG5CZ7ZoKbIilJz+uftpfu3aK0xw54qKmmdYOOO7O+O+LNWu5+V+CpNkTEqYEYw29Bc0/sgb1eUmoumZYYVl+vvJfs+p5jIOtUl3nAtUVa3LJ9jhpVL9/8pUAgO89+PLsmNkDftvuX/kGgtcc9ZrcTvvPLrhuzN/uIpkqc7Ob1jfOeCnay9I6RflV/ALkLK8VFrHEtUXyj9bP5Btq51gWy8+tZhSy6uYOOquexxuEXKOHj0EavdevFvaRQip7KpkjjgKLu13WX7uff40X2D5Gjjlp+A2aPEqJ/dEfXpodc+ldvM5xH9JTuEuOuUq73MkHAJXjvtGjOvL0d3kGWRW7AHk4TZaVkmByaywYIaU4i4GIsjhEyTtB40gl7VT7Lmn2hIQEwGJt8rxezOwogLMAjo3sos8Nq9g+sgLbS97tJCuwfeS9wjl3SdmGkb7sAGBm9znnbhjpRZ8ltpOswPaSdzvJCmw/ecuQzPiEhClBetkTEqYE43jZvziGaz5bbCdZge0l73aSFdh+8m7CyOfsCQkJ40Ey4xMSpgTpZU9ImBKM7GU3szeZ2UNmdsjMPjqq6w4LM7vczO4ys4Nm9nsz+wDXL5vZT83sj1zuHLesgplVzey3ZnY7P19lZvdQ1m+bWeNC5xgVzGzJzL5rZg9yjG+c1LE1sw/xO/CAmX3TzFqTPLbDYiQvu5lVAXwOwJsBXAvgHWZ27SiufRHoAfiwc+4lAF4N4L2U8aMA7nDOXQPgDn6eFHwAwMHg8ycAfJqyngDw7rFIVY7PAvixc+7FAF4OL/fEja2ZXQbg/QBucM69DL7U6D8x2WM7HJxzz/sfgBsB/CT4fCuAW0dx7ecg8w8A/CuAhwDs4bo9AB4at2yUZS/8C/LPAG6HT0E/BqBWNuZjlnURwGOgQzhYP3FjC+AyAI8DWIavHbkdwBsndWwv5m9UZrwGUDjMdRMJM7sSwPUA7gHwAufcEQDgctf4JCvgMwA+grw+ZwXASeecqjYmaYyvBnAUwFc47fiSmc1hAsfWOfcEgP8C8BcARwCcAnAAkzu2Q2NUL3tZ/dlExvzMbB7AfwP4oHPumXHLUwYzeyuAvzrnDoSrS3adlDGuAXgFgC84566Hr48Yu8leBvoN3gbgKgCXApiDn37GmJSxHRqjetkPA7g8+LwXwJMjuvbQMLM6/Iv+DefcbVy9ZmZ7uH0PgL+OS74ANwH4NzP7E4BvwZvynwGwZGaq95ykMT4M4LBz7h5+/i78yz+JY/svAB5zzh11znUB3AbgNZjcsR0ao3rZfwPgGno0G/AOjx+O6NpDwXxx+ZcBHHTOfSrY9EMA7+L/74Kfy48VzrlbnXN7nXNXwo/lnc65dwK4C8B/cLeJkBUAnHNPAXjczF7EVW8A8AdM4NjCm++vNrNZfick60SO7UVhhI6PtwB4GMAjAD42bmdFiXw3w5tmvwNwP//eAj8XvgPAH7lcHreskdyvB3A7/78awL0ADgH4DoDmuOUL5LwOwH0c3+8D2DmpYwvg4wAeBPAAgK8DaE7y2A77l9JlExKmBCmDLiFhSpBe9oSEKUF62RMSpgTpZU9ImBKklz0hYUqQXvaEhClBetkTEqYE/w8tDcQLnBaH9QAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "data_id = 75\n",
    "\n",
    "x = []\n",
    "y = []\n",
    "for idx, coordinate in enumerate(landmarks[data_id]):\n",
    "    if idx % 2 == 0:\n",
    "        x.append(coordinate)\n",
    "    else:\n",
    "        y.append(coordinate)\n",
    "\n",
    "        \n",
    "img_as_np = x_data[data_id]\n",
    "img_as_img = Image.fromarray(img_as_np)\n",
    "img_as_img = img_as_img.convert('L')\n",
    "plt.imshow(img_as_img)\n",
    "plt.scatter(x,y, c = \"r\", s=5)\n",
    "plt.figure()\n",
    "\n",
    "#Scaling labels for 100x100px\n",
    "landmarks = list(landmarks)\n",
    "scaled_landmarks = []\n",
    "new_line = []\n",
    "for line in landmarks:\n",
    "\n",
    "    new_line = []\n",
    "    for idx, pix in enumerate(line):\n",
    "\n",
    "        if idx % 2 == 0:\n",
    "            new_line.append(int(pix/(178/100)))\n",
    "        else:\n",
    "            new_line.append(int(pix/(218/100)))\n",
    "   \n",
    "    scaled_landmarks.append(new_line)\n",
    "\n",
    "    \n",
    "x = []\n",
    "y = []\n",
    "for idx, coordinate in enumerate(scaled_landmarks[data_id]):\n",
    "  \n",
    "    if idx % 2 == 0:\n",
    "        x.append(coordinate)\n",
    "    else:\n",
    "        y.append(coordinate)\n",
    "\n",
    "scaled_landmarks = np.array(scaled_landmarks)   \n",
    "        \n",
    "train_transforms = transforms.Compose([\n",
    "        transforms.Grayscale(num_output_channels=1), \n",
    "        transforms.Resize((100,100)),\n",
    "        transforms.ToTensor(),\n",
    "        transforms.Normalize(mean=[0.485],\n",
    "                                         std=[0.229])\n",
    "        ])\n",
    "\n",
    "im = train_transforms(img_as_img)\n",
    "im = im.view(1,100,100)\n",
    "z = im * torch.tensor([0.229])\n",
    "z = z + torch.tensor([0.485])\n",
    "img = transforms.ToPILImage(mode='L')(z)\n",
    "plt.imshow(img)\n",
    "plt.scatter(x,y, c=\"r\", s=5)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Parameters"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 93,
   "metadata": {},
   "outputs": [],
   "source": [
    "lr=0.001\n",
    "epoch_count = 100\n",
    "train_ratio = 0.85\n",
    "train_batch_size = 120\n",
    "validation_batch_size= 120"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Loading"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#x: numpy -> pil -> torch    y: numpy -> torch"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 94,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training set size: 42500\n",
      "Validation set size: 7500\n",
      "Traning Set Batch Count: 354\n",
      "Training Set Batch Size: 120\n",
      "Validation Set Batch Count: 62\n",
      "Validation Set Batch Size: 120\n"
     ]
    }
   ],
   "source": [
    "class CustomDatasetFromCSV(Dataset):\n",
    "    def __init__(self, x, y, img_size, transforms=None):\n",
    "\n",
    "        self.data = x\n",
    "        self.labels = y\n",
    "        self.transforms = transforms\n",
    "        self.img_size = img_size\n",
    "\n",
    "    def __getitem__(self, index):\n",
    "        single_image_label = self.labels[index]\n",
    "        single_image_label = np.float32(single_image_label)\n",
    "        single_image_label = torch.from_numpy(single_image_label)\n",
    "        img_as_np = self.data[index]\n",
    "\t# Convert image from numpy array to PIL image, mode 'L' is for grayscale\n",
    "        img_as_img = Image.fromarray(img_as_np)\n",
    "        img_as_img = img_as_img.convert('L')\n",
    "        # Transform image to tensor\n",
    "        if self.transforms is not None:\n",
    "            img_as_tensor = self.transforms(img_as_img)\n",
    "        # Return image and the label\n",
    "        return (img_as_tensor.view(1, self.img_size, self.img_size), single_image_label)\n",
    "\n",
    "    def __len__(self):\n",
    "        return self.data.shape[0]\n",
    "        \n",
    "train_transforms = transforms.Compose([\n",
    "transforms.Grayscale(num_output_channels=1), \n",
    "transforms.Resize((100,100)),\n",
    "transforms.ToTensor(),\n",
    "transforms.Normalize(mean=[0.485],\n",
    "                                 std=[0.229])\n",
    "])\n",
    "\n",
    "dataset = CustomDatasetFromCSV(x_data, scaled_landmarks, 100, train_transforms)\n",
    "\n",
    "dataset_size = data_count\n",
    "train_set_size = int(dataset_size * train_ratio)\n",
    "validation_set_size = dataset_size - train_set_size \n",
    "\n",
    "    \n",
    "print(\"Training set size:\", train_set_size)\n",
    "print(\"Validation set size:\", validation_set_size)\n",
    "\n",
    "train_set, validation_set = torch.utils.data.random_split(dataset, [train_set_size, validation_set_size])\n",
    "train_loader = torch.utils.data.DataLoader(train_set, batch_size = train_batch_size, shuffle=True, drop_last=True)\n",
    "validation_loader = torch.utils.data.DataLoader(validation_set, batch_size = validation_batch_size, shuffle=True, drop_last=True)\n",
    "    \n",
    "\n",
    "train_loader = [batch for batch in train_loader]\n",
    "train_batch_count = len(train_loader)\n",
    "print(\"Traning Set Batch Count:\",train_batch_count)\n",
    "print(\"Training Set Batch Size:\",train_batch_size)\n",
    "    \n",
    "validation_loader = [batch for batch in validation_loader]\n",
    "validation_batch_count = len(validation_loader)\n",
    "print(\"Validation Set Batch Count:\",validation_batch_count)\n",
    "print(\"Validation Set Batch Size:\",validation_batch_size)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "torch.Size([200, 1, 100, 100])"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_loader[0][0].shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 95,
   "metadata": {},
   "outputs": [],
   "source": [
    "class Net(nn.Module):\n",
    "    def __init__(self):\n",
    "        super().__init__()\n",
    "            \n",
    "        self.channel = 8\n",
    "        self.conv1 = torch.nn.Conv2d(in_channels = 1, out_channels = self.channel, kernel_size = 7) \n",
    "        #torch.nn.init.xavier_uniform_(self.conv1.weight)\n",
    "        self.conv2 = torch.nn.Conv2d(in_channels = self.channel, out_channels = self.channel+2, kernel_size = 5)\n",
    "        #torch.nn.init.xavier_uniform_(self.conv2.weight)\n",
    "        self.conv3 = torch.nn.Conv2d(in_channels = self.channel+2, out_channels = self.channel+7, kernel_size = 3)\n",
    "        #torch.nn.init.xavier_uniform_(self.conv3.weight)\n",
    "        self.conv4 = torch.nn.Conv2d(in_channels = self.channel+7, out_channels = self.channel+9, kernel_size = 1)\n",
    "        #torch.nn.init.xavier_uniform_(self.conv4.weight)\n",
    "\n",
    "        self.bn1 = nn.BatchNorm1d(num_features=150)\n",
    "            \n",
    "        self.dropout1 = nn.Dropout2d(0.25)\n",
    "        self.dropout2 = nn.Dropout2d(0.5)\n",
    "        self.pool = nn.MaxPool2d(2, 2)\n",
    "\n",
    "        self.beta1 = nn.Linear((self.channel+9)*4*4, 150)\n",
    "        #torch.nn.init.xavier_uniform_(self.beta1.weight)\n",
    "        self.beta2 = nn.Linear(150, 150)\n",
    "        #torch.nn.init.xavier_uniform_(self.beta2.weight)\n",
    "        self.beta3 = nn.Linear(150, 150)\n",
    "        #torch.nn.init.xavier_uniform_(self.beta3.weight)\n",
    "        self.beta4 = nn.Linear(150, 10)\n",
    "        #torch.nn.init.xavier_uniform_(self.beta4.weight)\n",
    "            \n",
    "            \n",
    "\n",
    "    def forward(self, X):\n",
    "\n",
    "        y=self.conv1(X) \n",
    "        y=torch.nn.functional.relu(y)\n",
    "        y=self.pool(y) \n",
    "        y=self.conv2(y) \n",
    "        y=torch.nn.functional.relu(y)\n",
    "        y=self.pool(y) \n",
    "        y=self.conv3(y) \n",
    "        y=torch.nn.functional.relu(y)\n",
    "        y=self.pool(y) \n",
    "        y=self.conv4(y) \n",
    "        y=torch.nn.functional.relu(y)\n",
    "        y=self.pool(y) \n",
    "        #y=self.dropout1(y)\n",
    "\n",
    "        #y = y.view(X.shape[0], self.channel*25*25)\n",
    "        #print(y.shape)\n",
    "        y = y.view(X.shape[0], (self.channel+9)*4*4)\n",
    "            \n",
    "        y=self.beta1(y)\n",
    "        #y=self.dropout1(y)\n",
    "        y=torch.nn.functional.relu(y)\n",
    "        y=self.beta2(y)\n",
    "        #y=self.dropout2(y)\n",
    "        y=torch.nn.functional.relu(y)\n",
    "        y=self.beta2(y)\n",
    "        y=torch.nn.functional.relu(y)\n",
    "        y=self.beta3(y)\n",
    "        y=self.bn1(y) \n",
    "        y=torch.nn.functional.relu(y)\n",
    "        y=self.beta4(y)\n",
    "          \n",
    "           \n",
    "        return y\n",
    "        "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 97,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 0 | Training Loss: 1674.1659602795617 \n",
      "Geçen süre: 5.59 mins\n",
      "Epoch 0 | Validation Loss: 239.6061253701487\n",
      "Geçen süre: 5.92 mins\n",
      "Model kaydedildi.\n",
      "********************************************************\n",
      "Epoch 1 | Training Loss: 46.3362755027868 \n",
      "Geçen süre: 11.85 mins\n",
      "Epoch 1 | Validation Loss: 4.028930479480374\n",
      "Geçen süre: 12.18 mins\n",
      "Model kaydedildi.\n",
      "********************************************************\n",
      "Epoch 2 | Training Loss: 3.492212560217259 \n",
      "Geçen süre: 18.2 mins\n",
      "Epoch 2 | Validation Loss: 3.2591902671321744\n",
      "Geçen süre: 18.53 mins\n",
      "Model kaydedildi.\n",
      "********************************************************\n",
      "Epoch 3 | Training Loss: 3.3812149385947965 \n",
      "Geçen süre: 24.63 mins\n",
      "Epoch 3 | Validation Loss: 3.758102551583321\n",
      "Geçen süre: 24.96 mins\n",
      "Model kaydedildi.\n",
      "********************************************************\n",
      "Epoch 4 | Training Loss: 3.25281480544031 \n",
      "Geçen süre: 31.09 mins\n",
      "Epoch 4 | Validation Loss: 3.546410091461674\n",
      "Geçen süre: 31.42 mins\n",
      "Model kaydedildi.\n",
      "********************************************************\n"
     ]
    },
    {
     "ename": "KeyboardInterrupt",
     "evalue": "",
     "output_type": "error",
     "traceback": [
      "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[1;31mKeyboardInterrupt\u001b[0m                         Traceback (most recent call last)",
      "\u001b[1;32m<ipython-input-97-15a98cc0d850>\u001b[0m in \u001b[0;36m<module>\u001b[1;34m\u001b[0m\n\u001b[0;32m     29\u001b[0m         \u001b[1;32mwith\u001b[0m \u001b[0mtorch\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mno_grad\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m     30\u001b[0m             \u001b[0mmodel\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0meval\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m---> 31\u001b[1;33m             \u001b[0mcikti\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mmodel\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mX\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m     32\u001b[0m             \u001b[0mloss\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mcriterion\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mcikti\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0my\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m     33\u001b[0m             \u001b[0mtraining_loss\u001b[0m \u001b[1;33m+=\u001b[0m \u001b[0mloss\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mitem\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32mE:\\PROGRAM FILES\\Anaconda\\envs\\staj_projesi\\lib\\site-packages\\torch\\nn\\modules\\module.py\u001b[0m in \u001b[0;36m__call__\u001b[1;34m(self, *input, **kwargs)\u001b[0m\n\u001b[0;32m    487\u001b[0m             \u001b[0mresult\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_slow_forward\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m*\u001b[0m\u001b[0minput\u001b[0m\u001b[1;33m,\u001b[0m \u001b[1;33m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    488\u001b[0m         \u001b[1;32melse\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m--> 489\u001b[1;33m             \u001b[0mresult\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mforward\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m*\u001b[0m\u001b[0minput\u001b[0m\u001b[1;33m,\u001b[0m \u001b[1;33m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m    490\u001b[0m         \u001b[1;32mfor\u001b[0m \u001b[0mhook\u001b[0m \u001b[1;32min\u001b[0m \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_forward_hooks\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mvalues\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    491\u001b[0m             \u001b[0mhook_result\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mhook\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mself\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0minput\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mresult\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32m<ipython-input-95-152be016c060>\u001b[0m in \u001b[0;36mforward\u001b[1;34m(self, X)\u001b[0m\n\u001b[0;32m     32\u001b[0m     \u001b[1;32mdef\u001b[0m \u001b[0mforward\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mself\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mX\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m     33\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m---> 34\u001b[1;33m         \u001b[0my\u001b[0m\u001b[1;33m=\u001b[0m\u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mconv1\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mX\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m     35\u001b[0m         \u001b[0my\u001b[0m\u001b[1;33m=\u001b[0m\u001b[0mtorch\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mnn\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mfunctional\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mrelu\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0my\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m     36\u001b[0m         \u001b[0my\u001b[0m\u001b[1;33m=\u001b[0m\u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mpool\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0my\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32mE:\\PROGRAM FILES\\Anaconda\\envs\\staj_projesi\\lib\\site-packages\\torch\\nn\\modules\\module.py\u001b[0m in \u001b[0;36m__call__\u001b[1;34m(self, *input, **kwargs)\u001b[0m\n\u001b[0;32m    487\u001b[0m             \u001b[0mresult\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_slow_forward\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m*\u001b[0m\u001b[0minput\u001b[0m\u001b[1;33m,\u001b[0m \u001b[1;33m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    488\u001b[0m         \u001b[1;32melse\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m--> 489\u001b[1;33m             \u001b[0mresult\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mforward\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m*\u001b[0m\u001b[0minput\u001b[0m\u001b[1;33m,\u001b[0m \u001b[1;33m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m    490\u001b[0m         \u001b[1;32mfor\u001b[0m \u001b[0mhook\u001b[0m \u001b[1;32min\u001b[0m \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_forward_hooks\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mvalues\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    491\u001b[0m             \u001b[0mhook_result\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mhook\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mself\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0minput\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mresult\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32mE:\\PROGRAM FILES\\Anaconda\\envs\\staj_projesi\\lib\\site-packages\\torch\\nn\\modules\\conv.py\u001b[0m in \u001b[0;36mforward\u001b[1;34m(self, input)\u001b[0m\n\u001b[0;32m    318\u001b[0m     \u001b[1;32mdef\u001b[0m \u001b[0mforward\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mself\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0minput\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    319\u001b[0m         return F.conv2d(input, self.weight, self.bias, self.stride,\n\u001b[1;32m--> 320\u001b[1;33m                         self.padding, self.dilation, self.groups)\n\u001b[0m\u001b[0;32m    321\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    322\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;31mKeyboardInterrupt\u001b[0m: "
     ]
    }
   ],
   "source": [
    "model = Net()\n",
    "criterion = torch.nn.MSELoss()\n",
    "#criterion = torch.nn.CrossEntropyLoss()\n",
    "optimizer = torch.optim.Adam(model.parameters(), lr=lr, betas=(0.9, 0.999), eps=1e-08, weight_decay=0, amsgrad=False)\n",
    "    \n",
    "training_loss_list = []\n",
    "validation_loss_list = []\n",
    "\n",
    "\n",
    "t = time()\n",
    "\n",
    "for epoch in range(epoch_count):\n",
    "        \n",
    "    training_loss = 0\n",
    "        \n",
    "    for batch_id in range(len(train_loader)):\n",
    "        X,y = train_loader[batch_id]\n",
    "\n",
    "        model.train()\n",
    "        optimizer.zero_grad()\n",
    "        cikti = model(X)\n",
    "        y = y.float()\n",
    " \n",
    "        loss = criterion(cikti, y)\n",
    "        loss.backward()\n",
    "        optimizer.step()\n",
    "        \n",
    "        \n",
    "        with torch.no_grad():\n",
    "            model.eval()\n",
    "            cikti = model(X)\n",
    "            loss = criterion(cikti, y)\n",
    "            training_loss += loss.item()\n",
    "                \n",
    "    training_loss = training_loss/train_batch_count\n",
    "    training_loss_list.append(training_loss)  \n",
    "\n",
    "    print(\"Epoch {} | Training Loss: {} \".format(epoch,training_loss))\n",
    "    print(\"Geçen süre: {} mins\".format(round((time() - t) / 60, 2)))\n",
    "    #****************************************************************************     \n",
    "        \n",
    "        \n",
    "    validation_loss = 0\n",
    "                  \n",
    "    for batch_id in range(len(validation_loader)): \n",
    "        X,y = validation_loader[batch_id]\n",
    "        with torch.no_grad():\n",
    "\n",
    "            model.eval()\n",
    "            cikti = model(X)\n",
    "            y = y.float()\n",
    "            loss = criterion(cikti, y)\n",
    "            validation_loss += loss.item() \n",
    "\n",
    "         \n",
    "    validation_loss = validation_loss/validation_batch_count\n",
    "    validation_loss_list.append(validation_loss)  \n",
    "\n",
    "    print(\"Epoch {} | Validation Loss: {}\".format(epoch,validation_loss)) \n",
    "    print(\"Geçen süre: {} mins\".format(round((time() - t) / 60, 2)))\n",
    "    #****************************************************************************   \n",
    "    \n",
    "    #ModelSaving\n",
    "    checkpoint = {'state_dict': model.state_dict(),'optimizer' :optimizer.state_dict()}\n",
    "    torch.save(checkpoint, '{} Checkpoint.pth'.format(epoch))\n",
    "    print(\"Model kaydedildi.\")\n",
    "    \n",
    "    print(\"********************************************************\")\n",
    "        \n",
    "        \n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Test"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 123,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[39.0982551574707, 50.98930358886719, 60.832950592041016, 51.14260482788086, 49.580848693847656, 62.232120513916016, 40.18253707885742, 70.19090270996094, 59.5692253112793, 70.33899688720703]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "import torch\n",
    "import numpy as np\n",
    "from torch.utils.data import TensorDataset, DataLoader\n",
    "import PIL\n",
    "import pandas as pd\n",
    "import torchvision \n",
    "import torch \n",
    "import torchvision.models as models\n",
    "import torchvision.transforms as transforms\n",
    "import torchvision.datasets as datasets\n",
    "import torch.nn as nn\n",
    "from time import time\n",
    "import PIL\n",
    "import matplotlib.pyplot as plt\n",
    "from PIL import Image\n",
    "from IPython.display import display\n",
    "from torch.utils.data.dataset import Dataset\n",
    "from torchvision import transforms\n",
    "\n",
    "class Net(nn.Module):\n",
    "    def __init__(self):\n",
    "        super().__init__()\n",
    "            \n",
    "        self.channel = 8\n",
    "        self.conv1 = torch.nn.Conv2d(in_channels = 1, out_channels = self.channel, kernel_size = 7) \n",
    "        #torch.nn.init.xavier_uniform_(self.conv1.weight)\n",
    "        self.conv2 = torch.nn.Conv2d(in_channels = self.channel, out_channels = self.channel+2, kernel_size = 5)\n",
    "        #torch.nn.init.xavier_uniform_(self.conv2.weight)\n",
    "        self.conv3 = torch.nn.Conv2d(in_channels = self.channel+2, out_channels = self.channel+7, kernel_size = 3)\n",
    "        #torch.nn.init.xavier_uniform_(self.conv3.weight)\n",
    "        self.conv4 = torch.nn.Conv2d(in_channels = self.channel+7, out_channels = self.channel+9, kernel_size = 1)\n",
    "        #torch.nn.init.xavier_uniform_(self.conv4.weight)\n",
    "\n",
    "        self.bn1 = nn.BatchNorm1d(num_features=150)\n",
    "            \n",
    "        self.dropout1 = nn.Dropout2d(0.25)\n",
    "        self.dropout2 = nn.Dropout2d(0.5)\n",
    "        self.pool = nn.MaxPool2d(2, 2)\n",
    "\n",
    "        self.beta1 = nn.Linear((self.channel+9)*4*4, 150)\n",
    "        #torch.nn.init.xavier_uniform_(self.beta1.weight)\n",
    "        self.beta2 = nn.Linear(150, 150)\n",
    "        #torch.nn.init.xavier_uniform_(self.beta2.weight)\n",
    "        self.beta3 = nn.Linear(150, 150)\n",
    "        #torch.nn.init.xavier_uniform_(self.beta3.weight)\n",
    "        self.beta4 = nn.Linear(150, 10)\n",
    "        #torch.nn.init.xavier_uniform_(self.beta4.weight)\n",
    "            \n",
    "            \n",
    "\n",
    "    def forward(self, X):\n",
    "\n",
    "        y=self.conv1(X) \n",
    "        y=torch.nn.functional.relu(y)\n",
    "        y=self.pool(y) \n",
    "        y=self.conv2(y) \n",
    "        y=torch.nn.functional.relu(y)\n",
    "        y=self.pool(y) \n",
    "        y=self.conv3(y) \n",
    "        y=torch.nn.functional.relu(y)\n",
    "        y=self.pool(y) \n",
    "        y=self.conv4(y) \n",
    "        y=torch.nn.functional.relu(y)\n",
    "        y=self.pool(y) \n",
    "        #y=self.dropout1(y)\n",
    "\n",
    "        #y = y.view(X.shape[0], self.channel*25*25)\n",
    "        #print(y.shape)\n",
    "        y = y.view(X.shape[0], (self.channel+9)*4*4)\n",
    "            \n",
    "        y=self.beta1(y)\n",
    "        #y=self.dropout1(y)\n",
    "        y=torch.nn.functional.relu(y)\n",
    "        y=self.beta2(y)\n",
    "        #y=self.dropout2(y)\n",
    "        y=torch.nn.functional.relu(y)\n",
    "        y=self.beta2(y)\n",
    "        y=torch.nn.functional.relu(y)\n",
    "        y=self.beta3(y)\n",
    "        y=self.bn1(y) \n",
    "        y=torch.nn.functional.relu(y)\n",
    "        y=self.beta4(y)\n",
    "          \n",
    "           \n",
    "        return y\n",
    "        \n",
    "    \n",
    "    \n",
    "def deploy(filepath, test_im):\n",
    "    \n",
    "    train_transforms = transforms.Compose([\n",
    "        transforms.Grayscale(num_output_channels=1), \n",
    "        transforms.Resize((100,100)),\n",
    "        transforms.ToTensor(),\n",
    "        transforms.Normalize(mean=[0.485],\n",
    "                                         std=[0.229])\n",
    "        ])\n",
    "\n",
    "\n",
    "    def load_checkpoint(filepath, Net):\n",
    "        checkpoint = torch.load(filepath)\n",
    "        model = Net\n",
    "        model.load_state_dict(checkpoint['state_dict'])\n",
    "        for parameter in model.parameters():\n",
    "            parameter.requires_grad = False\n",
    "\n",
    "        return model.eval()\n",
    "\n",
    "\n",
    "\n",
    "    model = load_checkpoint(filepath, Net())\n",
    "\n",
    "    \"\"\"test_im = Image.open(image_path)\n",
    "    test_im = np.array(test_im)\n",
    "    test_im = Image.fromarray(test_im)\n",
    "    test_im = test_im.convert('L')\n",
    "    #plt.imshow(test_im)\n",
    "    test_im = train_transforms(test_im)\"\"\"\n",
    "    test_im = test_im.unsqueeze(0) #dimension expansion\n",
    "    \n",
    "\n",
    "    \n",
    "    with torch.no_grad():\n",
    "        model.eval()\n",
    "        out = model(test_im)    \n",
    "        out = out.tolist()[0]\n",
    "     \n",
    "    x = []\n",
    "    y = []\n",
    "    for idx, coordinate in enumerate(out):\n",
    "\n",
    "        if idx % 2 == 0:\n",
    "            x.append(int(coordinate))\n",
    "        else:\n",
    "            y.append(int(coordinate))\n",
    "            \n",
    "            \n",
    "\n",
    "    print(out)\n",
    "\n",
    "    \n",
    "    test_im = test_im.view(1,100,100)\n",
    "    z = test_im * torch.tensor([0.229])\n",
    "    z = z + torch.tensor([0.485])\n",
    "    img = transforms.ToPILImage(mode='L')(z)\n",
    "    plt.imshow(img)\n",
    "    plt.scatter(x,y, c=\"r\", s=5)\n",
    "    \n",
    "    return model\n",
    "    \n",
    "    \n",
    "model = deploy(\"4 Checkpoint.pth\", validation_loader[0][0][4])"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}