diff --git a/Jupyter Notebook/Marchov_Random_Fields_for_Segmentation.ipynb b/Jupyter Notebook/Marchov_Random_Fields_for_Segmentation.ipynb
new file mode 100644
index 0000000000000000000000000000000000000000..f5a73ecfa8800691843382c7aee5a65d0ead23b2
--- /dev/null
+++ b/Jupyter Notebook/Marchov_Random_Fields_for_Segmentation.ipynb	
@@ -0,0 +1,860 @@
+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "execution_count": 54,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import numpy as np\n",
+    "import matplotlib.pyplot as plt\n",
+    "%matplotlib inline\n",
+    "import cv2\n",
+    "from PIL import Image"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 74,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "A  = ['020_HC.png','018_HC.png']\n",
+    "# x = 0\n",
+    "# for i in A:\n",
+    "#     pic = cv2.imread(i,cv2.IMREAD_GRAYSCALE)\n",
+    "#     cv2.imwrite(\"% 2d.png\"%(x), pic)\n",
+    "#     x += 1"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 5,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "(540, 800)"
+      ]
+     },
+     "execution_count": 5,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "pic = cv2.imread('000_HC.png',cv2.IMREAD_GRAYSCALE)\n",
+    "pic.shape"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 11,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      " 1.png\n"
+     ]
+    }
+   ],
+   "source": [
+    "# print('% 2d.png'%(1))"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "##################################################################################"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 56,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "from PIL import Image\n",
+    "import numpy as np\n",
+    "import pandas as pd\n",
+    "import os, os.path\n",
+    "from scipy import misc\n",
+    "import glob\n",
+    "import sys\n",
+    "from matplotlib.pyplot import imshow\n",
+    "import imageio\n",
+    "import scipy.stats\n",
+    "import matplotlib.pyplot as plt\n",
+    "import matplotlib.image as mpimg\n",
+    "from scipy import optimize\n",
+    "import random\n",
+    "import warnings\n",
+    "warnings.filterwarnings('ignore')\n",
+    "import cv2"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 112,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "initial image\n"
+     ]
+    },
+    {
+     "data": {
+      "image/png": "iVBORw0KGgoAAAANSUhEUgAAARkAAADGCAYAAADmD/RZAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjAsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+17YcXAAAgAElEQVR4nOy9a4zkaXbW+fzjmnG/ZeS9KuvS1T19m+rpmdlhxkjYjA3DysgfEGCvsADZsoTgw0r7AbMgrfiC/GklYAXSCKPFYmUbadcCgcXKMiDPYmx6umfcNX2pruzKzMqszIzMjMiIyLhnXPZD1u/kiegad3VPl8kx9UqpqoyMiP/tfc95znOec95gMpno2Xg2no1n42mN0H/rE3g2no1n44/3eGZkno1n49l4quOZkXk2no1n46mOZ0bm2Xg2no2nOp4ZmWfj2Xg2nup4ZmSejWfj2Xiq46kYmSAIvhEEwd0gCDaCIPjFp3GMZ+PZeDZ+OEbwWetkgiAIS/pA0k9I2pX0hqSfmUwm736mB3o2no1n44diPA0k8z9I2phMJvcnk8lA0q9J+qmncJxn49l4Nn4IxtMwMquSdtzvu49eezaejWfjv8MReQrfGTzmtY/EZEEQ/IKkX3j06xefwnk80YjFYppMJgqC89MmfAyCQJPJxH4Ph8MKh8P22ng81mQyUTgcVhAEGo/HGo/H9t5oNKq5ubmp9/MZSQqFQnZM/hYEgSKRiCaTiUajkX3f486JY/I3f+xQKGTn5c/VXw/vD4LAzsWfB9/jP+fPmXPg33g8rlAoZN87Ho+nriEIgqnz4TWOw3f68H00Gtk9jkQido5872g00nA4nPoujsO98Pds9to4Ltc9Go0+8myCILDj8TrHnUwmikQiikaj9rfZ+8w9HI1GdrzhcGjnMns/OQ9+54fv8Z/hHo1GIw0Ggyee809xHE8mk/Lsi0/DyOxKuuJ+X5O0N/umyWTyTUnflKQgCP6bFFBls1klk0mFQiENBgPF43FJUqfTsUnR7/dtMiWTSc3NzUmSzs7OpiY5CzsajdqkymazSqfTWlxc1K1btxQOhzU3N6dut6tGo6Fer6fhcKhQKKThcKhOp2PGrNFoqNVqaTKZKB6PK5fL2ff3ej3VajX1+31bLM1mU41Gw64jHo8rGo0qm80qEomo0+mo1WrZYvCLkwUUDoc1Go10dnZmBi+bzSoajSoIAs3Nzdk9Go/Hmpub09zcnJLJpMLhsNrttnq9njqdjrrdrobDoZLJpKLRqN3nbrers7MzSVI+n1cqldJwOFQkElE6nVYqlVK1WtXx8bEGg4EtsHA4LEnqdrvq9/tqNpvqdDqKRCIKhUKKxWIqFAr2TOfm5pROpxWJRBQOh3V2dqZGo6Hj42MzJuFwWJlMRtFoVCcnJ6pWqxoOh4rFYkomk+p0OhoOhzo9PVU4HLbvPjg4ULPZVCwW08LCgvL5vKLRqFqtlrrdrt2fRCKhQqGgRCKhTqejubk5pVIp7e3tqd/vazweKxqNKpFIKBqN6uzsTGdnZ+p2u+p2uwqFQvYs5ufnFY1G1Ww27dmdnp5qPB6rXq9ra2vrj2DFfOzYftyLT8PIvCHpVhAE1yU9lPTTkv6np3CcH3iUy2W1221bRMPhUIlEQrlcTmdnZ+r3+2ZsgiAwg4MnCYVCisfjH3kfXrNWq6lWq+no6EiVSkXZbNYWQrfbVafTkSRbgEwwDFm327VFPxqNFI1GzWvG4/EpFDAYDDQYDHR2dqZer6der6cgCNRutxWLxTQej23iYhxisZhCoZCdRywW02AwMM89HA7VbDYViZxPk3Q6bdc9Ho+nDM54PDZDPB6P7VqazabS6bQZSIyoJPV6PZ2enurs7EyxWEzdble9Xk/dbtfeF4vF7HOhUEi5XE6dTkfRaFSRSESxWMwMTTKZVDKZVCqVMscRi8WUy+UMPbHAua/1el3RaNQQQSwWUyKR0GAwUL1eN2PJvR8Oh8rn80qn07bYu92u3ftIJKLhcKjRaKRut2vzIRwOq9PpaDAYKBKJKJVKaTAYmIGUzh0XDg3Dg3FPpVLq9XpaWFgwQ86zvOzjMzcyk8lkGATB35L0/0oKS/rnk8nknc/6OD/o+MpXvqJer6d2u63T01PzQJFIROVyWclkUrVaTaenp5I+Cv9jsZji8bgtrl6vp8FgYH+XZOHVYDDQ7u6uotGoUqmUFhYWzPuHw2EzGCCXRCJhHv74+FidTscMgR8s6mg0qlgsplQqpX6/b95PmobmQPtsNqt8Pm/GIpFIGETHi/NZUAYLWZJN/lwup3g8rrOzM4XDYbt2FgVGXJLi8bgZK+5LvV5Xv9+XJPX7fdXrdfV6vSnD4a81kUgoEonYcUKh0FS4AurAIHO8drutfr+vWq2mRqOhfr+vXq+nUCikYrFo58T/pXOEl8/nLeRNpVKGiEB+GNq5uTm7h5LMeEkXYVAsFtPZ2dmUYUkkEnYePK90Oq3JZKKzszN1Oh31+33FYjGNRiPNzc2ZgcZBgJov83gaSEaTyeQ3Jf3m0/juz2r0ej1Fo1Gtra2pXq+rUqloOBzaJEyn0wbfQQB4ffgBRiqV0srKivr9viqViqEIeApJFsLg/ZvNpgaDgRmSSCSiVquler1uYUoQBAbz+/2+TTDpfPJGo1HF43GVy2UVCgXl83llMhlDMq1WS+PxWMlk0sImH+awUAqFgtLptOLxuBqNho6OjiyMAxkQVgwGA3U6nanFE4vFzDgkk0llMhkFQaDT01Ntb2/r9PTUwgVCi8FgoGQyqVKpJEk6ODjQ6emp+v2+crmcoa/RaDR1PZLUarXUbDaN98AIshhPT0+VyWQkSY1GQycnJ6rX6wqCwIwyxqPX60mSGVZJdp+j0agkmTGJxWJm8BOJhJaWllQoFAyJ8N5QKKR0Oq1Go2EhaL/fN7RZKpUUDoctZIpEIsrlcioWi5pMJmYEQbVcS7VatecIssMJXubxVIzMZR8LCwtqNBrGmeRyOYO4QRBoMBjYAsWjYlRALb1ezxZgNpvV2tqaUqmUdnZ2dHh4qGazqXA4rFQqZV4VRNFqtSx+BykQYkHgsfjz+bx9BtKPkEaSeUeMB94eLw5HMzc3Z2HBeDyeMpaEAYRUkgyBESIB5cfjsS0qQhzCOSY+59ZqtQzl8J0gDZACHj6bzSqVShkKgu/yhDrHnJubUzabNQPhw8tms6ler6dKpWIkrf8sPAh8GwPkQFhHOMWChuNiHsClgDrn5uaMJ+L8FhcXNRwOVavVdHh4qLOzMyWTSTOGoF8cEQgIlFMqlRSNRpXL5dRsNs0wgmq5r975XMZxaYyMzyp4lPA0hofd9XrdYlsWEPE8E8BnTRKJhKQLDiQWi6lWqxnRGwqFVCgUpgwDxGWr1bI4nvDh7OzMFpMkW/ihUEgLCwuGVAiDOp2Oms2mJJlR6HQ6FmrgARlMap89YQFh4LgGvC0ICl6i0WgYcsG7c64+c3N2djYVunBdGE6MJEYP1JbP51Uul+38+C44ML5/NBqp3W4bV8P1c84QtJCokNo8t3Q6bd+NscGI8lwItTB0PKtUKmWc1dzcnPr9vjqdjnq9ngqFggqFgs2pfr+v+fl5Mzy7u7vq9/vKZDJaW1uz78HIxeNxM6rVatVI7Hw+r2w2q1gsZqQ6HFm/31cikTBUepnHpTEy3qs8yUgmkza5HheTeu83OxYXF82QQdBJF4uGyUf2h7CG98RiMTMYeKR+v69qtWrZFoxTq9Wa8vKkk1lc8/PzWl9fVzQaVa1W097enkHydrttfMDy8rLG47EODw9tkmGYB4OBGo2G2u32FJHoQx6foZmbm7PrgnwdDodqtVpTvFIkEjHvCd/hQyMW3HA4tPCLxRIKhRSNRlUsFs1gDAYDQz69Xs/uyWAwUDabVSaTUSwWs9cJFTGGw+FQZ2dnisfjOj09NSQgXaSwMS7egHDPPQLjc7FYzIhdn3WCG8FQtttt1Wo19Xo95fN5xeNxC+H6/b4lEDAmGMGTkxNFIhELSblewljuc6fTUa1WM56p1Wopn8+rVCopCAI9fPjQ5hbzXjoP+0Gtl3VcmrPz3vdJhk+7eo+Kt/x+3/cP/sE/UK1WM8K3Wq2aV+E7pXPvcvXqVVv81WpVk8nEvpfJgsdnAeNpIUKHw6GRd2RJQDEcd2FhQdeuXVM0GtXDhw9VqVT04MEDPXjwwFLP7XbbUqjSeXwOaTg3N2ew3qfRydZI54a1VCqpXq/r6OhI/X5fw+HQFjsIJJvNSjpPL4fDYctypFIpCxm4Fk8Q+7CGzwwGA+NOQGfD4VDdblftdttQGIu10+konU4rmUzq5OTEECLIUJKhP5AYWhgQKM8BtMJ94frI8pCOJ6tI5gqEgxHHSKysrCiTyZihgWsBSYBUgyBQr9fTzs6OvadUKqndbts58z2DwUCLi4tKJpOqVquGsKXzUHN/f1/9ft/m9JUrVxSLxbSxsaHT01ND2nA2l3VcGiPzSZEMngiE8STjz//5P6+trS0LefBYoI5oNKrT01MLHYIgmEq/TiYTNZtNg6zj8VjValWSLC0JyRoOhy18kc6NEguVNHY4HFa329Xv/d7v6c6dO8pkMorH46rX6zo5ObFFJJ1POr7Lp9BBEkzgfD5vXj+RSNjiQ19CNmI4HJr+plgsKhaLKZvNanV1Vc8//7wKhYIZQzgljCSoQZJlteAZWDy9Xs/CHO7l3Nycer2e6vW6ms2mEbjtdluVSkUnJyfGN9XrdY1GIyUSCQuxCNXINGFc4ZTgVCRNhb6kfMmskT3E8IEG0LO02207b+YJ7wFpeK4GZISmCfSHrAFjR4jtuSrCSpAWYWGn09F4PFalUrHn3O/3zchjwDj/yzwujZH5tIpFvG2r1VI6nbYMxONGt9vV5uamCbVYCMTrwHlgM2RdKpXS/Py8wfNutzs1cRDZJZNJMzacExMLRIOX7PV6tugajYbF4mg8CHPIbOCBEbJhSPgME9VnI0AyhB5HR0caj8dKp9NaXl7W+vq6SqWSFhYWdOXKFX3uc5/TysrKlB7GK1L9tUhSoVCYUv72+32lUinVajU1m02Nx2OVSiVbsGRZ5ufn7ZlAAne7XVWrVTPwrVbLEOTR0dFHnjmLEucgnWf5JBknM+uAuD/lclnpdFrHx8eWCSTj5MO0IAiUyWQs64gB4H2JRELJZNJ4sHw+bw4MISDCOo5FJomsZb1eV61WkyTTVp2cnGgymahcLhu6Pjs7s2zn8vKyhcXxeNxkApd1XBoj80nDJYZX6/5haOhrX/uaPWQMSyaTMdh5fHysdrttgqlYLKaTkxNLZ/P9LCSgN96H0Gg4HKpSqdhx8/m8IRrP7xwdHVnGhoxJLpezxcrrhCJ4Toyg16VIFxJ6n+r1pCsZi0KhoJdeeklf+9rXdPXqVVsU6+vrj43t8bwnJyd2HFLx+/v7ltHp9/uGfOBoON9MJmMKWj4LKozH46bbIbyqVCqWcm40Gvrwww8N/dTrdctQwSWBLIbDoaXbCcukc5QMv3RycmKpX0K5TCajyWRiGb/RaKSFhQUL3Y6OjnRwcKDRaGTzp9frKZ1Om8EhtUwIyrODOM9kMrpy5cqUZimZTJpWpl6vW0ZRkqEin1GULng2rknSM07mSccnDZcYcDJMcjz97BiNRlPy99PTU4P9ZFvC4bAWFxfNcEnnD3Vvb8+4GuTvSPqJ4X02ihCLjFAsFrOQhDAF2O3rkvw1kUJPpVJKpVKWKcEoSLKwBMJ3PB4bOstkMkokElpdXdXq6qquXLmi27dv69atW7p586ZisZhN3lnycDgcqt1uW2kDpG0kEjEBXRAEltGC+MxkMkqn08pkMspms4rH4+p0OqpWq0qlUmq320qlUqaAzWaz6vf7Ojk50enpqaWDn3/+eeO2Op2Orly5ouPjY9XrdVWrVXU6HZ2enurhw4c6OjoyZa0kI8sJdcnM8JzOzs50eHhoqBMEiEEDMaysrBhi4pkzj6QLlEdNV6/XM9IZAwrHkkqllMvllMlkNBqN1Gw2zclI584gk8koHA6r2WxaRjCdTpsxhcjGCa2untccHxwcqNFofKq180c1fmiNDF7cK2H/sO/w0nHgp5+Q/B2vg3AMQ9Fut5XL5awuBoNC7E/Kk/AqFovp8PBQnU5HQRCYsSC7gnFkIkkyeI2SGANFtgXewJcuEFqRCYEIXF5e1rVr1/Taa6/pS1/6km7cuPERgpDFenR0pEQiYWHZ3t6e9vf3LXQhpCEsgN/BsPb7fZXLZR0dHRlRC2JBop9IJBQOhzU/P6/JZGJe/OzszARly8vLSiQSGg6Hlm0aj8ean583w9Zut43TOD4+1vHxsY6OjnR0dKTDw8MplEN4k81mpwpDfZaGMPb09FSnp6f2OpwQKJIaJAw5xpwslyQLcQeDgWq1mqG5aDSqTCZjz5cQz4fdfJ7ShkKhYDzMZDIxHo+/k+U7PDyckn9cxnFpjMynDZeeZPzsz/6spZwJWbzgygvbmPS+LogFgtxbkiEnJPA+LUqKMRqNWsjCsTAgCOI6nY5NTFS9kMp4SFATYdJkMlGlUrHwAxHetWvX9NJLL+mll17SV7/61Y+EQBimTqejra0tHR4eWlaHhTYYDPTw4UMjcDkuRYlzc3OWufFiONL3k8lE77//voUWhUJhKutFuIAhhgtJJpPa2dlRqVQyPiOXy2lpaUnZbNYKNefn5y2lvLa2ZjKCk5MTy8a0Wi0LcarVqqrVqvFjOAjOe3FxcSpTh7gSWQBOpNvtKhKJTBWQgooRXcbjcQsdSZfH43FzJMgWIKcrlYoajYYZjFarpbm5OZVKJSWTSQuVs9ms5ufnzUmSEZtMJioWi48tOblM49IYmadZqn5ycmLqTeJZPIpHCcTankCl0E+STWhCGfgZ4nKyBZJs8WBUJNkx0XlQOY3UHa6IkAjyFL5FkuluKD1IJpMql8taXV3V1atX9af+1J/SK6+8MiWTl2QlD5DMu7u7VhxIOAE3RWp/OBwaKet1KOhe4B7Ozs7UbDYNhXG+w+FQDx8+NM6JcIX7Q/hKlTH6o+XlZStAREMDyVoul21Rg1BAO+Vy2dDJzs6Orl+/rnA4rPv372tra8skARDo8FygWfQtEPog3mazaQpfRjKZNM5GunBQ8XhcN2/eVCaTsYxgs9nUwcGBisWiEomE1W1VKhUT+mGEcVYQu4Rl1WpV7XZbo9HIxHuExWtra09t7XwW44+9kfnxH/9xaznAhAAdeKKOOqB0Om3pVkRskH+gE0hgFK4YHvgWSRYS5XI5See6FiYMcnoyEyArvBwGCqLTk4T+nLLZrBYWFvTaa69paWnJMkSz/UZarZYePnyow8ND40AqlYrB9cPDQysihOj16XNfbewJbulCtyJdVGP7DE0oFLIMEwvRIxqq27e3t4032d3dnbpHKysrKhQKVrnM/0F2OIp8Pm9Cv+XlZS0vL2s4HGp5eVm3b9/W3t6e7ty5o93dXfP+JA3W19eVz+eN86BuCJEijgdDSRsIVN7dbleTyUS5XM7CnOPjYyPhySrWajVrJeJFgMlk0o7v+9dEo1HV63U7X1CkJMuUfRIZx3+LcWmMzKclfj9ugAxobdBut22ywDOgM2GgCyELQN8RYHKhUDDIPh6PdXp6amIrFrgnadvttsX4iNA4BlkJanbgAyaTiRG6ZJ7wslRyX716Vc8//7xWVlaUSqWs2FCSGc/vfOc7evjwoRUpkjLf3d01URlhD+fPJOfHi9/I5ND6gMWA5oTvwVAR+hByeYU24SE6FxxNo9GwMC8Wi6lcLiuTyahQKFgx6OLiohHoHEOSieNQSPtasNXVVT333HPa2NjQ3bt3dXh4qEajYRkuuB/kBSBWnvNoNDLjhnQB5AsSJuwMgsCkEcPhUPPz8ya629nZsfIT+gOBANHbcO4YFM8dEp6TJXtWu/SE42kIitbW1nT//n0TuFH/IV10kINv8VJ9X9XLQsMwJBIJS+eSXWERMqFZVEwYBhmifr9vWp1isWhoB2k6i5TUKER0Op02ziKbzWpubk4PHjzQW2+9pclkouXlZS0uLiocDuvo6Einp6c6Pj5WpVKxqm8MDeGCN2aEjRhKrkuS1XZFIhHF4/EpAhzOicnu76PvCYMmheddrVYtXOA1+Kd4PG7Gf39/376De7C0tGSoE1HlwsKCCoWCSqWSbty4YajS9wCSpNdff10nJyfa3NzUf/2v/1Wbm5sWIsK1Ue7RaDT0wQcfWL3YYDBQsVg0Y4OhhrvBqZEGJ4OXTCZt/hWLRbuHhH7NZlP7+/vGs5GK5/nTrArDA4m8v79vae/LOi6Nkfl+qecfZDx8+FAHBwe2cH0LRrwuBZL+BxgPKcxnSL2SVfL9ZXwRJe/l/xwP+AuXks/n7dqp1i6VSlpbWzOvlc/ntbCwYJ3Rjo+Ptb29rXfffddQBloUKo8hObvd7pT83FeQg9IwPCh+Ue56j0xaGkNDKMX9lDT1b7vdNqEgx8R40aQKIhmPj8Eje+L5MrJnXM/+/r62t7e1trZm4ePc3JyOjo6sHqzX66lcLtti9QR4EAQqFouWsXnzzTd17949VSoVI8ZxCLFYTM8//7xOTk6s86AXKHKvyBQi3CuVShY2wh0RckkygWcul1O73dbOzs5UVXYqlbKSAz6LlsZzQ9Fo9Bnx+6Tjsw6XfAHkbAHlbKMfzw9ADrOYeMB8p49/PUp53P99bRXeqVgsqlgsTjWO8u0BKE3I5XJ67rnntLKyYpkFSdrY2LAULkKv/f19jcdjk8v7am+fYmXA8XDtGFyIXtpnko05OztTrVaza8DAws34KnLPJ/g0PEJGj44gOT3RSejkxXS+YBKU1Gq1rG8MJQKRSESZTEbFYlEHBwdaXl5WqVRSJpPR4uKiFhYWpnq/BEGgW7duKZfL6YUXXtBbb72l7e1t1et1410ikYgphAmZfekBaCmXy5mKHLTKNXEt4/HYGmFBeNfrdUPEGNhsNqtcLmeyAgw1zgMukTn232XTqk8zPutwCULNG4nZ4YvugPZ4MAwEFbzeU3vE4v/1x2GCQTgvLy9rZWXFeBP0OtFoVIuLi5YduX79utbX142HQR3bbrfNmCwuLlrVNSnsZrNplbqcI2EbBpMJi96D3issdtK4pNC9AZDOjUKj0bBJzYKd1euQESN88noaFiGhD4acQTjg67FAOfAvcEFU0PMdvsTj7t27yufzWlpasvu4tLSkr33ta1paWjIxonTeX2hhYUHFYlEffPCB7t69q3v37llxKqEgIjl4HO4DRo7CT7KGKIK552SWIINp7cD34GR8+E3KfGdnxwhjjI/vLXyZx6UxMp8lQ85C83G4j8t9VTADY8JC8doXJoFHKP5fb3gYLCIIxZdfftkaLdE7VjqvcUG8duPGDf3Ij/zI1HkOh0PLMj148EB3797VgwcPtLGxoU6no+PjY/OaiL7gdCAuJVlqF0TA65C40kXmAjKcgREmk4RmBNKW7+e93kjAWfhBOOk5BsIs9DGhUMg0KFRvg4z8OTO8tAADj7KbwlM6/j3//PO6evWqqbC53+vr6yqXy3rxxRf1rW99S7//+7+vWq2mzc1N5fN5vfjii6b7IaVMNTscC2gDJIcOhvISb5i452S4MMxLS0sql8vq9Xo6ODjQ3t6eRqORisXilDYnl8s99d5Ln8W4NEbmswyX0G4wPEMvfdRI+InsMyveq/sf6WKrDN8YiZArmUyqWCxqfn7evOn8/LwVAbKYisWiVlZW9MILL+jq1at6+eWXpyZNs9nU5uamtra2tL29rXv37mlvb0+VSkW1Ws0WKSlwoLskI3gJjYDc3hjCZfgKZ4wIn/Hd4rwOiONSoOorlCV9JITyFc2EWp5g5/wwTCBR3udV2t7ow/V4FS3cBmEXimIU2LS7WFhY0CuvvGIqakmGPH/8x39cKysrevPNN/X2229rMplob29PZ2dnSiQS1kycsgs4NK+XwtBgQD26Y64xX0BGdMTLZrNWLHl2dqa1tTWl02kdHBzY9+AMVlZWPvVa+aMYl8bIfFbEL/G/Rym+zsTzBd6ASJoqOvStAzxiYXhk5EnjdDqt+fl5C3+YQL64j0raF198Ubdv39Zrr732EfFcp9PR+++/r7ffflsffvihtre3Va1Wp7Y+wbN7IpaQhHIAaRplEF7Bv2BcIWApysMg+5YOSAHwyjTLjsViVpdD2tpzMixyn4FDoMfzIPUMQe9RGPc4nU5beATXQWiBUM/rfXiePvyAQD48PFQ2m9X+/r5u3ryp1dXVqSxNuVzW66+/bhmsBw8e6PT01LQwbIVSq9WsP3G9XtdgMDAOCCTlpQ2g41arZQ4gk8kok8lYcS6izYODA52cnBjfg9SBTCP3kh5Al3VcGiPzWSEZFgxEIJyMj/u9IQEu+zS2L47E2/iB18SwQIiS2UACT5oZr57L5bS6uqo/8Sf+hL74xS/qypUrtoh7vZ6Oj4+t6fXe3p6+973vaWtrS/v7+6pUKiYK85W4QHPUyt54cM3AdNLJCOAkmdoYzQl1MRgylMiEn5DVtNbgPnEcjABZFHrWZDIZ03VwPzyn4wfPju/1BhNUw2CRYmCpEoePg/vyFfCNRkM7OzvK5XLa2dnRG2+8oVKppM997nN6+eWXTdOSyWR0+/ZtZTIZvfvuu3rw4IF2dnZ0enpqxjgWi+m5555Tq9XS1taWyRxisZgePnyoBw8eKAgCrays6Nq1awqHw9rc3FQikdDCwoKdE1qryWRiu2R0Oh0TXC4uLiqRSGhlZUWRSETValUPHz6UpMuy59L3HR9rZIIg+OeSflLS4WQyeeXRa0VJvy7pmqQtSX9pMpmcBOer8R9K+h8ldST9tclk8taTnMhnwcmga5Eudu7z5Ce8ixfMnZ2dTaWtvYRfmiZzGUwkwgMfHiF7J0ULlI7FYqbf+PKXv6yrV6/adR8dHem9997Tw4cPtbe3ZxuN7e7u2v9brdZU1zdfgwVyIEzwjcMxBKReUbDyXWTfMAzcA2C/JyATiYRxEIPBwOp1yHz45l7ofCRZ0SOkLCiFZ46hIxWLUBID7eUBPmxFBkBHu6OjI5dBimIAACAASURBVMu8wS/xeV8+4jeYg7ClZ244HNb169d148YNczzXrl2zOqp4PK7NzU0r+XjxxRetEX0ul1O1WlWj0dDBwYHtHwW6hajl/tE1AKlEuVy2Rl48j0wmYxwVtXKUFoTDYZ2env6xaPXwf0r6PyT9invtFyX99mQy+aUgCH7x0e9/W9Kfk3Tr0c9XJP3TR/9+7PgskIzXWkiyEADtA8N7ThaZ99aee5A+mjXCm4IAUOvSx4R9nDBe5XJZL7zwgl588UW98sorWlxclHSubL1//76+973v2a6Eh4eHhlpou4nn94Qp98urhpGf49FRnGL0eL90EZ4SUkCmwuUQwtDGgoWAfohWELyHjJEkU8FipPzxMZCgSLgevptsGs8JI+GRDX/DGJFWppsc2RzfWB0D6EltnwafTCb64IMPrKcQSANkRwbw2rVr+ta3vqW7d+/afaCZFFKDZrNpjaiKxaLy+bwVcfrMHS01ut2u1WL5lhE0E6flBXocUFS5XLbyi8s8PtbITCaT3wmC4NrMyz8l6Ucf/f9fSPpPOjcyPyXpVybns+T3giDIB0GwPJlM9j/uOJ8FJzOZTKYITEnWd5W0KpPV6zmkC6KSicv7vBf14RE7TSKYYjJKUqlUUiwW0/z8vPVy+dKXvqSFhQULayqVijY2NvSf//N/tnifZuSERShKQRqcB5MQaT9iOlAVnAy8zcLCgqGQ2awMi5Lr4/9edEhWCcPnU9ucBy0uMUaI08iGcCyPLNChUMMD6QzKAmVIF32DvSyh0WiYrB4v78/5cfwFjoHaqnq9rrm5OZ2eniqdTmtzc1Nvv/22vvOd7+jVV1/V5z//eV171H+ZfkOvvfaagiDQ1taWNjc3dXBwMJWtDIVCWl1dtUZX6GxoAUqLC8JAnNre3p7xPqBpWleQ+YSYJpzjeVzm8Wlx1iKGYzKZ7AdBsPDo9VVJO+59u49e+1gj84PeKFpQ8j0sEt+eQboIowid/Ov+M7w2+3c8Ijskrq6uWk/WwWCgfD5vIrByuWz6GFSf9Xp9KlN0eHhoCISeJoQkCNjw/P58fSFlKHTR1HwWgRGaoEnBkGC4CCmli3IIpPXSRR0XuhSvXpYuQk5S2p6PAdFQXcz3of1A8EeGiB+MP0jUnyPhrE/JY6Qgg6nb8rU+IDWQVTQatf3IyWhRsoD8n/T1l7/8Zb388suWQs7n8/rc5z6nYrGoO3fuaH9/34zV9evXdf36ddVqNeVyOSO9T09Pp3aa5NoKhYKazaY5F7owElZxvblczp4PjhNt0NOq+/usxmcdzD0uaf9YJVwQBL8g6Rf4/QdFMr6FAJPt0XEMtkuaWqz8Dc+Il6Nw0heoQVQyURBaMeHH47F5zXK5rFKpZFAWg9HtdvXgwQN997vf1fb2tvb29qzhElW9GAWgOOeLd/e1QKApzgHEwKIk7evJWAwSqWFpuik75C/FnKhTY7HYlMGen5+3MKPT6VhRI6QnYQnngVEilIUv8lkm9pRi/27eB6rz1+XPOxKJGFHqnQj/hz/j/Hn+fmtZxImUaICUDg4ODCW+8MILKhQKlj28du2a5ubmdOfOHW1vb1sHRHZXAKXUajWTBYA+EVgS6lF86Xk+yHLPMTE3eZ6Eq5d5fFojUyEMCoJgWdLho9d3JV1x71uTtPe4L5hMJt+U9E1JCoJg8oMgGW40tSEsbvaRJv3JAwR1+MyTz8j4bnCQlUDyhYUFraysKJFIaH5+3viEZDKpmzdv6vr16yoWi3rhhRdsU7ZOp6O7d++q0+lod3dXv/u7v6utrS0j+AiL+GHSYeDovMdilC46vAG9IXU9bwJa82hEumhCTtsFfgdVsCf07D2CJKcNBj2JfeqfsIJyCXopM0gtA/8lmUwfwz4r1MMYYui5Vi8dwKtjgCFZPY9F175QKGT9hSVN7RZKKLW/v2/PNRqNan9/X2+++abW19f19a9/3XiYL33pS1pZWdHW1pbu3btn+21fuXJFN2/eNMO+v79vIRraoFgspqOjI9sOxSNFkOF4PLYw1ZO9VI7zHX/n7/wdC99Ho/PdHtj+GM0TqJlw/8MPP9Rv/MZvfOp196Tj0xqZfyPpr0r6pUf//mv3+t8KguDXdE74Np6Ej5F+MCTja49osESfD6/5kGT1M374iSjJFoBHDyg62X7UexmyB6+++qpef/11a04UDoetl8v777+vZrOp7e1tbWxs2KZlkoy09A3HpYsao/n5eS0sLBgxeHJyYuiHhQrsx1DQpQ4uihoX36Tcbw3jm5az4NBuEFpRgiDJ0tjdbneqZ4qvFaNY0xsEBqjF65R4j88Sci2ERRgoFqAXTXIcwjXf54bvAl2BGqitor8v10hPmHq9rp2dHdXrdW1sbOh73/uejo6O9PnPf163bt1SoVDQwsKC5ubmNBqNdP/+fTNWKIox8CBT9sKmMwDFlFyr35YHeYRHYtls1solQLzILzY2Nkx0SahFqQOcEIg7mUzqS1/6kr797W9/6rX3JONJUti/qnOSdz4Igl1J/5vOjcu/CoLg5yQ9kPQXH739N3Wevt7QeQr7rz/piXxaIwOS8AuOuh4gtiS76YQikmyy8uPT2L4AjYfPzoH0ez09PVUoFNLi4qJeeuklfeELX9Dq6qoZp1arpffff19vvPGG/uAP/sBqYdhQjuFJWohldDa5XE7lctm20ajVaiaKw8v56/PcEdAaVOO5D+lCcwJ6wChjGDAovheuD1GkC+MG2cuC8ouCujQEhyBJjg+qIjXrjQqfJ/Pk07VsDse14bEJ9ZgflFukUqkpQ87zJ/QExfr092Qy0cHBgaECGqLv7Ozo1Vdf1SuvvGJlC9yPVquler2utbU1lctlraysKJ/PW99kiHOajWMEi8WiNamiYfva2prNO/ZXR8fD3AE9woGBvk5OToy/4f57PpI+wk97PEl26We+z5++/pj3TiT9zU9zIj+ITsbrXvB0wGMWj4+9yUBAMgJRvVHifIDfxWJRy8vLmpubU61WUzabVRAEWltb06uvvqovfOELWlhYsOMPBgO9++67euutt/T2229rc3PTiu3gX6TpEgeyI3jCxcVFMy6RSETtdltHR0fWe5YOfT5j46X83BvPi3gjg2H2PYnJ0vhm2YSTEIw+w4NexSNGL6dHNo8BIBvEovY1SRh4UrVer8R9I8wAaRBWUYyKYeP7IEZJG3vDy/Xi3TGGswkE/37uy8HBgQ4PD7W9va2XX35Z165d0+LioobDoYkoG42GVldX9fnPf16lUknD4Xk1Nv2IUFhTKV+v1+3ej0YjK+xkgE4JOX0dmS9PwBDDj2GgCbm4jlAo9EfSuvPSqHg+DZLxXtsbltn3SBebk/Ean2VheMGeJGsmlM/ntb6+rsXFRfMYtBO4ceOGvvrVr+rWrVt2vMFgoA8//FDf/e53LfOwsbFhoQG8A2EL54/n5rxY1HjpSCRi4juvraD1AtcYj8etoho+iawK3z+bifGhEg21stms7fKIEYYT8KEYvW4hi8lCoROiJSY9WqhGRr8CP+H3rZ419oRgvLfZbFpND8Z6OBxahiedTpsAEMPEd2Dg2L4Fjo7GXYSt0oW4EzEihtUbi93dXe3s7OiLX/yi/vSf/tOmiN7f3zdylrmUy+VUKpW0v7+varVqoQ9iTZAp4RfyBEI55h58U6PRULfbNX5oNBrZljHSuYPxDcFA4fCVhUJBZ2dn+rEf+zH9x//4Hz/x+nvScWmMzA+CZPCYHze8R8c74/lmjVMkErFmUSsrK7px44YRqoVCQdeuXdNXv/pV2/+G8cEHH+g//If/oA8++GBqN4Bms2ke3occ9A8hhT3b3iASiWh+fl6lUkk7Ozu2m6KP8Rmj0cXuhqAaFpdf+GQ4aEnKYvA6E9Km2Wx2yvix2EEPvjuev48YOQhin9HCuLJgQSeegOY8vEPwldb+eXLtEPZnZ2emkoV3IUQg00RIjP6HHQZAfr5hGZofvoPjtlotHRwcGKpcXV1VsVic0maxMR4N0SGR4Vwmk4mWlpaMbEfBizPxRpQthSHSCalIQGAs4dDoLugRPbtTBkFguygsLS19sgX3CccPrZHxOg28q3SBiB6nKyH0kTQVe3vPiZydDnXr6+taXl62OpNwOKyVlRW9+OKLWl5enjJuBwcH+i//5b9YnQsZk3q9btyJJz/5HX4FpS6TmfNYXFxUqVTS8fGxLWZ/LVyHVzb7sGmWf4Azki4QHt9BPI/XR6+BAcAoY2Qw2Dw/r+5FLSxdcDi0AUXDwz3Ak2O4MBo+XQ2PQ6ZQ0pTUwKtp2fdIkhkPvoPzwXBhfFAu8yxmeTnuE8Yaw00qutfrKZlMWqtWtlKpVCr6k3/yT+rKlSsmWJybm9OHH35otVU4MM+f0fOZMCeZTKpQKCidTtucYt4OBue9oTlnSH1CT1+j59PpxWJR1WpVP/ETP6Hf+q3f+rhl96nGD72R8aEFROVweFEkSFqQ5kEeuXgVLd/Jjb9165a+9rWvqVQqWXVuq9VSKpXS9evXdfXq1Slyc2NjQ7/zO7+j3/3d37XaFiTqdLLHS8ZiMVvAIBLpQhPi07ucFwYkFAoZuiCV61O1vi6HMIn/s4D9xmagBo8YICPZtoNwDWND0SOyeHgeMlCeD/F8RjKZtHoub9ghYn2PFe4JHAM9hXmNayOMQwYA+uFcQA+zG6lhTDDWZ2cXnQZ9iUQ4HFahULCaJwpHCamC4LyTIG1eCWXoasj5gvDS6bReffVVCzdp5bG1tWUGc2lpSWtra8apZDIZRaNRLS8vW5vVjY0NPXz4UMVi0Xo4R6NRra+v2xrodDo2n9hXnR8MDdmxZrOpr3/96/rt3/7tT7QOn2RcGiPzSTgZz5LjZfCupJvxakx4FgEZEm6+H/F4XJlMRuVyWeVyWUtLS7bro08lv/baazZZG42Gvv3tb+vb3/623nrrLW1sbExlEDBolDSwkKniHo/Htrk7CINFyEbtyWRSjUZjaqJx7ahF2cPZq2UxHCiUWWScm08dY4RBLs1m01AHA05JkhGv/I6iGNLShzugJYhXXzaAhskvfjgKUIsku4cYbdTCPF+4IW9kCPMwWuPx2PgY5AUMarGYPz7jRejHeZFKxzEScqHunZubU6PRsHKO4XCora0tyxZlMhndunXLDPfGxobtxZ7NZq29BM8ITmh3d1eSdHx8rFqtZmEm1891Mdc4LxwR5Lnv2cP9X15efirN/KUfUiPjFwekKaI0PD0769VqNasTIU726EW6CJHK5bLW1tZ048YNLS8vW7HbeHy+V/GLL76oq1evTqWov/Od7+hb3/qW3nzzTe3t7dmCZwHw/RgAPou3ZBJ7aThe7/j4WL1ez7bIpRkXCxKuZbblw+MWMtftJxKoyiMZngUTm7DCCxUxViCQUChk3wvq4ft4n89ShcMXW3+QPcHAYxQkmacdjUZWcuG5JghlwktITn6XLtLlhHooozFChGD+/Ofm5j6iQWJnAdCU74rHOe/u7iqdTiufz2txcVFnZ2e2ZxRFr51Ox3ZauHnzpjY3N7WxsWFc3Hg8Nl6G5xOJRKwfESi4UChYgy7uHc3omeeksJFbjEYjayPhr6NQKJhU4mmMS2NknjRcoqJVkonK+L8k6wZH1Wy/39fh4aEZIB+GgHBo1cBm9Ch60SBkMhmtr6/rypUrtkn87u6utra29O/+3b/Tm2++qd3dXWszMCuD5zg+Pe0JVsITFjDhEpwOCk+MC1XfkqaKEj2h6eNwz5n48Akv6YWHkmzx8BqLkxANzoNQFNQIQsNDzz5b0AZIjXNGG+PT4F7FnUql7J5CNEO80t6BOiQf5viCytm5BuLxg2fiyyvg3AgL/f7eGEuMMkiOZzcej233y8lkYr1oxuOxlpaWrDZtcXHRhKQseHiibDZriJz5yHkVi0U7HvwSDrTf7yufz1v2ajKZmK6LzfE435OTEy0vLyufzz8Vcd6lMTKfpKwAr4pnJVMxGo0svn7w4IERqsBf3sOYTCYql8taXFxUPp9XoVAwXsPzAFRTs7AheD/44AO9//772tvbs0mFQfGZDy/y860NOB8WG6EKr2M8gL1AeXYilGTQmPeB6ny4Ip2jHshVFg6Lib194C7w7tIFqewNGJPZb5Ln0RAogfocBt4f7oL6LsKldrttPBbGlYUxGAyUTqdtAXKPuS7Om5AsEonYe8ke+TS4J4BZmCAArh8HBqcGj+ezbMw9byBbrZb29vZ0cHCgF154QSsrK/rc5z5nSGJ7e9uyPzShymazCofDJh2g9SahL7uHMv959qTuQ6GLnU6Pjo6USqW0tram8fh8h1DpIgxHyY1eCyRGce9nPS6NkXlSJAMDzyAUIb5kEjAxvEGaHb65EyUDxWJRpVLJFgB6CniQw8ND3blzRwcHB9rc3NS9e/esMTgoheOyMFkMHuH4hejLCMgyMOF9iYNXK49GI1soaDG8sM+jC9+vdxbJTSYTy0jBH/j2nT5rxPn52iJEcPS0IT3u+QSfdvYlCtSCsRh2d3fNg2NkvADP63A4H0hZ7g1IZjgcGolL6MfwZQmESR4dMwgxMfYYr9lKf19ki9M4PT3V0dGRTk5ONBwOlc/nde1RZ7yzszNVq1Xbe9vvSEGLVXi04+Nje6ZnZ2dWuQ0xT+hHiIcIsFgsWojHtRIaQ4IvLi5aeNZut20r5G984xv69//+3z/RenyScWmMzCxP8riBPgC23qf+fCUx4YIvesSzM4DpSPcjkYsKYpDDysqK7etD7cfm5qbu3Lmju3fvWpGjb25NrOv5Er/4IVlBAJw3E5pF6Ot9SL8DyWeNLIsLD8vrXn0sXew35VXPwHCvcPZqWgyhFwX6dDCGxqMBL/2HbOUafDf/yeSi9QX33Av24vG4Go2GpItMGeniWCxmyJMQCkU150m2yTsbf398ito7AH9vfHiJYQPNofHBcKAGxqDv7+/ru9/9rlKplL785S9bunh7e9v4GQxePB7XwcGBKpWKZe8Iy7rdrur1usLhsG0J41tWsMULAkU4slqtNlWrJEknJyemrWFnVfZp7/V6Zqw+y3FpjMyTEL+lUsn2x4HordfrlmGp1Wq20FiMeF4mP7+zTcXq6qpWVlamsj3pdNo2r89kMmq32/rggw9UqVS0vb1tVdQ8MK8W9jqd2QkuXUBcOBb/N/5PJoOFKV0oklnwLGrvuQm5MMBcD5ogZPSeh2GxQaJiBDFM8EfcN9SrkUjEsjPJZNIUp+wGgKoZBMDCkS4MEfer0WgYMY/OhPtBul26aGbuwxgyPqurq1MtJzxig1vyGiS/lS7fB1/C5+HIcAjStP4IlCBp6jlzL6vVqvb29rSxsWHGG/QFgrpx44YSiYT298/riCmGHAwG2t/fNyEmDshn3wjnacDe7XbtPsPtoDk6OTmxYlYoAYxUPp+3iODKlSuKRqP6yZ/8Sf3bf/tvn2Dlfvy4NEbm45DMn/kzf0aJREKLi4vWbS4IAm1ubmpnZ0eHh4d65513dHJyYvyIdMGH+MlEVoMwAxSDGI1evfl8XpFIRA8ePNDDhw91//593b9/X1tbWzo8PLRJ7quPPbxnMUvT9T6EI3hHXw7BZCVjQB8XIDgejIGhmEwumkMRInBt3ih5rsWHERgkqpA5rs9OeWKUxUjoI10UUiIV4JrheuADKCj0Yj2I1mQyad/hlcqgIa9o9veSkMHzKZzfbIW5D825Zo9icFIYR+YR7+e9vsG5dxQcAz5oc3NT4XBY5XLZwiCeDd9NGhmtTr1etzQ0PY5AJV4aQS0WfNfq6urUvEZf1Wq1TAMGuqS1ZyaT0cnJiRkwWoh8VuPSGJmP42T+wl/4C7Yf83A41NWrV7W4uKgf+7Ef0zvvvKP3339foVBI9+7d08HBgXlZnzLGuADJaTwFacqujegb6O3y8OFDffjhh9rc3NT+/r5pD/CEXrMjXaAyjuVFfzx8BF383YsH4VJAJp5cJCXLhGCici7oVbifLBzfIS4ejxta8JkpIDMLFZTkCdVZ+YDnLHwYMXvN0jnxWCqVlE6nLTzC8EYiEZVKJXsGiUTC+tpSK8TAiPjtW2jMxXXwTHxDJzg7f/6+wRnzxKMFH4Ji0LknXtyGEeDv/C7JNuBbXFzUjRs3bP/udrutSqVivXxxaPfv37e2DNw3Ku9Ho9FUkzOMAqQxWSNfZtHtdm1fcHRZGGuSDaPRyAos8/m8+v3+Z1bTdGmMzGztkB//+B//Y127dk0rKytKp9NTsHZ1ddVaXIIwqGrlPXg8Jt3KyoohFf6lyxxlBKlUSnt7e2q1Wrp3755tS8ID9ns7k2VgwnsSUJreb2g0Gk0pVj2JGYvFpsIvLwLzKXAUm0xusl6SLKUpaSos4DyB7YRAvkBTkilbQQCcP8ZH0lSPFD7HcVjcs6UaPgUPtCf2J6xFaEdYy/WXSqUpT8798agJQwUnBnLyfBjEpyRDRqAvH4pidDzv5zVI3CecE697R+EV3bRuCIVCpgLmdZwbjdbgT0gtd7tdlUolra+vq9/vq1Kp2LOMRqMqlUpTCm6eJzok6YLDZN5CAjNfJpOJ1eDhgJk/f6yMzB+GZGq1mt555x09fPhQ2WxWy8vLSiQSRoahwv3GN76h4+NjY+S9bgAPF41GdeXKFb300kuan583oZMnEOET7t27p42NDd25c0f37t3TycmJpf5I7UoXrSSJjTE+3mAAbRkYPz85ZxFNq9UyMg4IC6mIMYFz8W1DQQ6QgPxOyNVsNo18hieQLohhzgXuymtOfOjpM2Ascs9LYQApMo3H47avFM2wMGjSxa6XZGPI6uHRvZjPcyGgjmw2q7OzM9Vqtam6KB+W+P4pLEzugSfVWYw8D8hdUuYgBc6dZ+ufyWylODtAInqjv021WrW5h8qae8254cio8uaeR6NRNZtNmy8grNFoZOUn3W7XHAOGTJKFUmyT3Gw2dXp6avf+s9o07tIYme+HZP7RP/pHthOfF8Ghmcjlcrp27ZrW19f1oz/6o6pWq9Z6kIVCyMBxfLiBRN1Ls5kg9AWhaVG73X5suwiMAp7Dcwreq3udjNfSzPIlIDD+D+9AeOUzPXhLuBqMl6QpJOFbPTC8XgY0IV1kUojXJVmrBhBFJBKxReZJaa8Jmpub09LSkkqlktU6wQ+AAFi4pMWZ+Hh0aosw3Byj3++r1WpZbU+xWLTFQw8eECRhLxyZz9SxiHkGaKp4xp4E9uiAewnq4jy9VAHnBe+2v79v7RaoTCdcrlQqqlarqlarhpB4/nRX5BicP/ID7stgMDDimz7DOMVZdXc0GjWJQCaTMSU0bTxYM3/lr/wV/ct/+S8/7bI+n9c/0Kc/w/H9iF9Cg9PTU2sbIJ1nAiBf7927p9u3b+vmzZt65ZVX9Gf/7J+1hkKkPPl+4DW1HzzQZDI55cVqtZqRvMfHx4Yq8DAsQh+je3iOMWDS8RnPu0iy40O2kkLHUwNnfRUyXskL4CADPaLgejkfjs01+tAAY8Ei9OQvxoRj4vV9dgYjQzo2FAqZI0C3RCoWtOJJZMKS8Xhs1eaEP+Vy2cR6tDfA8GKkMSaEgt6pcLzZmiM+C/Jg+Ewd9342rOXZIjbkmfiSD4p2aerFXujdblfFYlHFYtEMb7VaNeEhz4rkAPdndk5I546pXC4bMsQp0PxqMpmYEZYu9r2q1WrW75f0P3IErhn09IOOS2NkHhcu/d2/+3endCRHR0eKxWImw87lctrc3NS7775r1bELCwu6ffu29vf39a1vfUs7OztGGqLipfiR1oYHBwdKpVK2qyP76VQqFdvn2BsYr8Hh3L1IzP/Ni7seZ3AkGRHIhOf72TDO63B8KpNJzQL3CM3vNui1MZwHoRnHisfjhprwjkw0Fjvek2N6PYxHAF4YiVHASPpKaZ8l8kjWh6PI58m0sKMA6fbRaGQb44FMwuHwFBflK9P9IvYGyqMqDKkXFHppAXwMhon76rsAMG+9QeWzd+/eVa/X09LSkmXTMGoom+Gw0GnNZr24d7TmRBB5cnJi87Df71uJDM8dpBgKhcxh7+3tmYEiNOUY0WhUf/kv/2X9+q//+idYzdPj0hiZxyly7969q2QyadqYweBiM/NEImFoBLj54MED04O88sorxlvs7u4qFApZPFuv13V4eGgd1CjCQ369vb2t9957z9SanluZ1Zd4gRue2HMvTGSQjw9heB3ORZruBEfMTf8UFikLhQyLP7504b1n61n4O+cCImGSImAD/RHCeMQmyTgrBkSifw/9SjAW3EPguE8pe2KVe+JVs3h9DD16HxYtixQxnDTdv9iT2xhIj2YwEv58uIeI+zBeGCv/LP2cIBznHKSL3kVBEKhSqZhhXFtb0/Xr161uC+Ehhnx+fn6qFw7nSajHPQad0ceGBAScoyRrioYz5rM8C+4X94oSBsR/P8i4NEZmlpP5e3/v7+ng4EAnJyfa3t7WzZs3FQ6HrTM7Whm2dm02m9rZOd9XrlQqKR6Pa3193bgZ31KRNOD8/LwZLKTeKCVRZPpF75GMF8FJsgUiTW8g57MbPlSClAUpMBlYMKAKFiPG1Bs7r+9hYqNS9WlVSF5QkA+hvLzet9JsNptTtVOgHc/veI0NhCiCOYwWMNyXb+DpISx9aDdrLOHXICzx/izUTqdjBjObzRpi8ml8n1WSLurkQB4+U8Vz5nlQJsEzxpjxjDwynU1fewQIsmMu0psmHo9reXnZMngseF/fxnn4Fh0YFzgurge1NiQu8xW1NByU56cSiYQZJq4b9Ixw8Od//uf1z/7ZP/vE61q6xEYGBWgkElGj0dAf/MEf2NYj1LU8//zzevXVVxWLxfTGG2/o4OBA2WxWCwsLWl5eVi6XM+NSrVZtMiCISqVSpompVqt68OCBNjc3tbu7q4cPH6pWq1laVbrIZPCgffjhh5940oVOxBcb+ds6wQAAIABJREFU8j5pOvMlXTTd8gI/yDs+w97JdLhPp9NmbH0ownUy0Vj0vk2B7yEDh+BDIa8yJuMxNzdneiHk7qA76sbi8bgtnGw2O7WvszQNyUOhkBkvjC4LemdnR+Px2LYYCYVCFk6ABMmssXgw5NwvjxR9q1E+48VtfD89jX1a3nNp/h755+tfH4/Hhgq49+iDKGRl4S8sLOjGjRuGuA8ODrS1tWVzkNYXbJvSbDZtDg8GA3U6HUt7sx+7d2grKytKpVKWJe31emZgfPMxzrtQKKhcLtvWuZ92XBojMzs2NzclXbQPBD4Ph0MTMb355pu6du2aRqORFhYWdHh4qP39fVNPlkol3bhxQ+12e6pqmSzSaDRSpVJRLBYzCTgpcJo0s+B9TO31MH7wgPzf/Gtei+EzS16ngdgQhSbZE8IC+qegFzk+Prb4ezi86JKHMUV7w0Sb9dR+EVInM1sR74vs/PAIgbCEUI4FifSdBQUK8dCezzAQDGI4KBz0WRGeQ6lUUjR6vv0HBD0hLEYDpOa5LB8O+tS9J+uRDPCMvHPhGbMwcTT8DukK1yVN641Go5Htn4Vmh2dBqEbGCfKWe8k9I9WMGNF/tyRDTe12W8VicSqMgi8bDofKZDL2vocPH1rvIu5DKBRSKpXSz/zMz+hXf/VXP/FafpJ9l65I+hVJS5LGkr45mUz+YRAERUm/LumapC1Jf2kymZwE57PlH+p8/6WOpL82mUze+iQn9fM///OqVCqWWmWhAwP9njrValXr6+sqFAq2AfmDBw8kybQZKCohvEgF4iVBCNScsGB9I6FH90KSPjIZpelGWrPGxhslL1hjoRDCYHjwtNJ5OtZzSX6LjG63a2GNdJExIpSCE0HvAgfFxMGgEVaAVri3/tq8EI3hpfuzg+/iGJwjxY8gM94XDoctzMIRwBGxgyKLx4dH8FaEDSxuFrt3ChzLh2XecBIu8XmMMaGOf05coxfsYYi5XwzEcrPiPTQ9hULBND6gUZ4L6Xu0TTgPDBNZN+5HOBw2Y0upACUq0WjUsn21Ws3CYwqFaYnBNsGj0UiNRsMM7vz8/JMt4JnxJEhmKOl/mUwmbwVBkJH0ZhAEvyXpr0n67clk8ktBEPyipF+U9Lcl/TlJtx79fEXSP3307xOP1dVVVatV25/Gk1G9Xs/aVbK1ZygUslYD5XLZYGK9XlcqlVKpVNK1a9fsAdbr9anPAdfpXwKs9QaGySbpI4tPusgwMLlnhWqzg78R8xNPUzohyUhZvDpEL3oZrpkF6DVATFLP8XhOgBDOLxiMz/czMD7lL8k8PcecTCZTimM4AO4L4Q0Sed+Myqt1PZIBLRHi+fMm9PBFlHAZ/jnMaqW8MSVrxDP0z9EL2zwXxrPl3hJeYcj8ecfj8andQfmucDisZrNpW5P4zgKzui6ed61WMweLyFGabjLmt3RJpVKKxWIqlUrW/Gp/f3+qlok5FgTBlOq4UChMFauSBv+5n/s5/fIv//KTLuXzZ/Vxb5icbzO7/+j/p0EQvCdpVdJP6XxnSUn6F5L+k86NzE9J+pXJ+Sz9vSAI8sGjfbOf5IT+/t//+8rn83rttdf07rvvant7W9JFmtfL7EmbHhwcKB6P6+rVqwYl0R3QMmBlZUWLi4tqNpva3Ny0MnvpggBkAbVarakiSyYLnIOP3X061k9eJpJPPTKA/dRP+T2daKE4Ho+tiRPktSc00aH4f3O5nGq1mnlzr9kgXHLP1fq8cD6+XYRHLYQVfAcGhWJQaXr7X9+Dmb63ePrhcGgckSfSCSu85gWE6VP1ENt4XOqPJNnikGRpaz43K0HgOr3eByPntTE+TPYZpMdJCVjojPF4bA6A+4eRQen7zjvvGOdG7dFoNFIikdD169e1tLSkSqVi6IJzIETmPvgWEGhfVlZWVCwWFQqFTEhK+F0ul6eeJbwZaFeSJVJu3rypdDptOqdPOj4RJxMEwTVJX5D0+5IWMRyTyWQ/CIKFR29blbTjPrb76LUpIxMEwS9I+oXZY0wm5+0WFxcXVS6XtbOzo83NTZsMZBMQXZEpIi1XLpfNSxweHmo0GlnsGo/HrQdrNBq15tF+/yGfmpU0ZUDwitJ0eMTPbDjE+7w3ZYKQiZlV6Ha73SmhnNeXSBed3OBtCCn4PB4fQ+zP3UvxpQuE6DU7XrjokRqLkkwTP4Syj56p3Qc+M5vqZ7KThmZxYgQ4PxY2YR9Gnj3EPSohDMZQ4Hz4Du4zhoNFxXPDgHg+yROgkuwcuEf+2Uua6ozIffb3wxdPcn8g2/f29jQ3N6crV67YHF1YWNDq6qoZ6QcPHlhYyRzBOGFc+v2+qtWqhsOhrly5Yn2uEQDCrYCaIHQxLjiBRqNhbVNu3rypfD6vZDKpWq2mUCikn/7pn9av/dqvzS7d7zue2MgEQZCW9H9L+p8nk0lzNj73b33Max+JFyaTyTclffPRd08k6W/8jb9h3px2mM8//7xqtZqOj491cnKi999/X/V6XdHo+V7RqEDJ/z/33HPK5/NTBB8TBOKNNg/sLY24C+/nyT0PW4mB8aiPruMjnI10wds85rptgfJ9nhvAyOEp/QIgdembPyFuY/H48gLvVVnk/hz4Xo8Q8ZCgJtKxpMo5zmyIAe/DAJV49ASfAFoh7Q1v488PQ8m5oZ71yNFf13g8tiQBiAKERDqZc0ctzLVhXDwnw7UxB3z6HaHf7LV6w8T3eiNHrx2PeEhlHx0dmYYoHA5bKJTJZFQqlbS9vW1CO2rVuCZJlilaWlqyUAdyl/awNK6iTzW0AvcYhwZCpPgSrpLtdPxzfpLxREYmCIKozg3M/zWZTP6fRy9XCIOCIFiWdPjo9V1JV9zH1yTtPclxEF+x8K5fv24V0yCWN954Qw8ePJiCbb58YHNz0yqq2QhLulhYvok3Xm97e1v7+/u2OwA33Yc/j+6D/R/PzuAz8AdeQ+IHKk6ODQSGp6AzGd3l/SJGCRuLxbSysmIxOFue+hDNZ25YeJzTbD2QJy0Ru2FM4SU8GT5Linoex3NYkqaMPQiM96EupuYHZS3nCTLh2qnQhiD2SmwMP4jW95Dx+hNQjA+TQJU+g8ScAfHwdwyUJ9i5Tl8ZTyOvUqlkLRZIrxOW8hrcDK/Tq5cGXb6pVygU0vz8vHVzHAwGOj09NUcgyRqqV6tVNRoNOw9S9mShDg4O7N5RD+Wvixo+yHrUyJ+0cPJJskuBpF+W9N5kMvnf3Z/+jaS/KumXHv37r93rfysIgl/TOeHbeBI+5p/8k3+i9957z9J2wLtkMqmlpSUT15GqpDmVL1/3W0MwWYMgsIlM2trH6YQobCXrq4Jn0YhPi2Jw8Hye9PNhivfm/nO8h8wFhB4LEq2DD1049tzcnIrFolKplImvaF3JeYBCUIz6+Hv2vLwx8aEKXtWL73xoJ8mMpM9SgYo4D3+PMEY+48TrGEOMHkbXG7WlpSXl83lNJhMrhmVXB4xAtVo1xMRz5JlT9+TT9Ghn2LeJz/F9eHcfwno05+eBWzdmzHEGIBgclNfd1Go1xeNxe6aE8CzolZUV2+52YWFB8Xhc7XZbzWZTvV7P7iUIlvuO0cKpkNjg+Xgxp0co7JRAFlKSnd/CwoI+yXgSJPMjkn5W0p0gCL776LX/VefG5V8FQfBzkh5I+ouP/vabOk9fb+g8hf3Xn+REguC8O9jR0ZH29vas92i1WlWr1dLS0pKy2azm5+d169YtNRoNVSoVHR8fm0fDy9P9vtFoWPVwPB5XuVw2TQyLqt1uG7qZheE+Rn+c0WDwujdIfsHizZmUfoOt2ZCFkIAF4UVzPtVOWUSv17P+w944sqjx2IQGXJuvfsYbQ6pKMiLWhyy+xABj6Ylwjg1XBheBV5QusnTcF44B8UzGyqMirp2K+7W1NcVi55vGVyoV2/SeMMbfM8838UzC4fBUOQZCNoYn2X04xrn4Rldclw+tCa14lqenp5bRZK77AdeFgA60AdEOuoUwxuAR1tA0nOecz+cNKbNzJPOdva06nY7REoReIOtCoWDOCXEfjg/h5ScZT5Jd+v/0eJ5Fkr7+mPdPJP3NT3QWksV8hULBPPPh4aG2trYUCoW0ubmpUCg01Yt2fn7eembA2xwfHysSiZi6sdVq2daopLQJZ5jUntTzmQJpWvfCgmM8jkz1w08mT6KywHy7A//dLGJS1mhHgkfain6/r4ODA/t+z9nMchYcW7ooWfDZEs5jFrGx6OATMIBMehAYXt2HaI/LvIBIfGofbw6fQ68gygN8ewZC4HQ6beQ/CmDEfbRJ8BkpL3okLPD6HL+VSrvdnkKXnmvzxDCGmOfnSX04MxAL5+BbZrBHFMQt84KFzHMihJRkxod/OQbd7HhutLxA+hCJRMxhY1y5r/l83sKuUCikYrGo+fl5hUKhqT2+aQkRiZy3kCCD96Tj0ih+t7e3TXqOpUUMVKvVTHRGOg+h0tWrV3X16lUlEgltb29rY2PDyGC2NJFkJKlPJ/uMBA/Ux6UeUs8uWiacRw2zRK+ffD7WxVPPZmk8eShdGIVEImHd75rNpv2N8wKJUd4/i6p8Za2H6ywW0IgngD2a8vU0XikMkeszWP5v/jg+oyJdVATP6kZCoZBlVBjICo6Pj7WxsWEhD7waRsMrWUFqnBfXh6HkfnM9oJbZe0tGx0sYvFH1aXBJFuaxFzbaE5Dh4eGhZUyREHiDSy0RKOjw8HCKrKZJuySbt61WS+vr6yqVShZ6kxiYTCYm1kNPFAqFbAtmQiQSLSDW8XhsLSpQWC8tLenWrVtTnRifZFwaI7O3t6dut2ubfs3Pz1uzKggqFK6tVst6kbJQotGoFhcXNRye7y28tbWlYrFo0A79CWgIL3x0dGThk2+JKV0U/rFgpAsCmX95z/cbvNenijE2GCqMg8/okI7Gu/oF4dEP5B8cBMbFE9OeR8HbkhrmOLPqZq7fp2vhU7zxZQH6MMVnr/gM98CjNW+4Ub+iasYwSLIMEfel3W5PefxEIqFisajV1VULjaWL3kFerIgBwSl4AaMPWyVNIS2MAQvXI1GMBGjs+vXreumll1QsFo3zA5kdHR3pww8/ND2TR1VBENjC98cdj8e22Vs4HLZ2sRDilJ2AwKllI9ymrozn6NuneORGyM0cINniNzsk+/tJxqUxMngyHjDtLEn9oSnw/UlisZiltZFUk304ODjQ/v6+hUJra2tThX94Q/gMJisTD+3JbDjkIbR04ZFnUQzvwet5LofP+JaavsWC98JUzw6HQ3vIPuvi06/SBSdC86vZUEySpfU5L+Jvbzy5Vt6HwZk1OnjuVCplToHwi8/74RePn9AsYs8j+RCM12l+hUfm2hEtevkDzcf8Vr6cg3+mPGPmCqQ02SoEb5yHN1A+Y0eb0Xw+b/e50+no4OBABwcHqtVqOjo6MkKf++8zdaBparDIUiHSA9FUKhX77NLS0lRDcO5tOBw2NbHnZWbn43g8tgQKjoD5Q4KFCGM2tH6ScWmMDA2jMCjEgMViUS+//LK63a62t7d17949I8T6/b42NzeVy+VUKpUsNMKSM7kk2Y59KCPZhYAH5yfUYDCYEt5J01J0HyLxt9nMgv8/D5VFxcKFqPQZBURRLBoyZyyqWd7Ge0G8DedI0aMnLjlf9DT+s9JFqICH9SSvR0lkTjAwcBEYaklTCIbPeejvjYl0USnN4HohOvnxqX3uQbfbVaVSscwbvEQ2m9XS0pLVPzF3vNHxYZIn2/2xPJfkyeAgCJTJZLSwsKDXX3/dig13dnY0Gp0X4NKQnvIAX0Hu96aiKVe9Xrd5Ui6XlU6nTcjYbDa1v7+vbrerTCajGzduaDKZqNlsTjnAVCplmxJOJhOdnJwom81aNpLsLTwQzxeFPMg6m83aVi4UGj+Of/zDxqUxMigZuUEw3Kurq3ruuecUBIFlFe7fv69ms2kQsdfrGcwm5obdp5AMOO5jYDa7ki56mfiF5T2Nh5t/mCX3BsYTjvzu3+eJR68AhswDzTHxuT8+3PJZFH+us8f0oQrXS/jANbOAIBXhU3iPD/sIuSTZPt3SuXGaJddBpJCdfmF5A8wCJFzwhtmXFhDeeSOPM/Fd/VCqptPpqZou0JRHeD7L5GUKhJfeqGP4J5OJ7TSwvr6u+fl5I+Xfe+89y8r4EFmScWw+tQw/QxsT7iP3AiezsLBgRYsUPdKXh9CIjBF8XSwW08LCgjkg6aI7ItXzIFr0WhieIAis2DKZTOrk5ERbW1tPvrB1iYwMkwNvBupADxAKhZTJZHT79m3lcjndv39f1WrV3juZTLS7uzsVDoXDYUt7M0FPTk6srkm6uNlMGM+vsPhmjYqfmAyPfPxinjUs/gfP7vmL2e5ucEroeHw87QlDvDDxsxfH4Yn5QSk6S4zyXgwSKMNzSXBQ8GI+7OIavLGGO0ulUpYZ8e1EH1dHBJLknnkOiVARJMb5+J9+v28NtTnGrAzAE/Hcz9nn6e+hFx760op0Om1GdWNjQ0dHR9p6tB81YYcvj4hGo9b2lU4AlMVAZnPfG42GlcQsLi4qnU6bQceAtVot5XI5q1HyhpQwEp5GukBtZCmZa/wdxDQ730jLo87+JOPSGJnR6KLGaH5+Xq+//rq1Jjw6OtLbb7+t+/fva3V1Vbdv39bt27dVrVZ17949HR8fa25uzsR6BwcH5i03NzfV6/X0la98RUtLS0qn07pz547eeust44CYbJB3MOx4LOmjyt3HjcehGP+65z348WUFTDCgKgWUvlHRycmJNSzCo3tew4cRPtTBa5NNoIYFj1ev123iohj1RslDZK/eJXMSBIEZB1/seePGDc3Pz+vs7Ewffvih8U4sYtCXz8LNckNeKVsoFLS0tGSZJI+MKGwF+dEygWtdWFgwHo7PYEgYXpDmF6uvdcLh0YYjmUyq3+9rY2PDeh6xQZ2X4WMg6R6ATKNWq5nTowMe2VSQEkgL3guNFd8Nkmw2m1MhzWxWsNPp2DMgpPYtPqlPymazthc89Uwc55PuLnlpjMza2pqk8wpeHg69L9bX161RMpurEU7kcjmLZVF9krFhoR8fH+vtt9+23h1MPq9bwSMSavmOeLzHoxwmmzdCHs34RcKYDZ3w4HRgw1tKF5qJSqVi7QwwOhTQeZWr33DOL5rZtpuz6ILrJrPFsX0NFQscHknS1Pn60IeFlEwmVSqVjASlHSSGFRTEsVn0s/G+v+9UVftQzbd4gEvgnNCDsIsB3QTh60C7GExP9hI28d2IPMlYcl/x+EgLaA9K4yffv5nCRowjHI0nt5nLZBy5V/A3NJdHGU5aG6IXA14oFLS8vGzZrVqtZohmcXHR2qFCG/R6vSmSN5/Pa3l5WZKskBj0Osudfdy4NEYGYRUhAM2pstmsRqPzjaqy2ax1rcNTHR0dGSSEwKUmh4fX7XZ1584dvffee5bOoysZFlqaNgJMELiF2TS152gYfoF4vmAWXfgwhMHDY6GOx2Prb9Nut22PbgRUq6ur5km5xtPTU1UqFUuP8oOSlLaYLHR0E6iiPfkLquJ1PCbch+ct4F6A1L5XMJ4TEhsuwhtpjILXGs0aHO43PZhpM8Ezw7ihS4EDgu/i/rEbJdIF6cIQg+IQhnL9xWJR5XJZxWJR2WxWQXBeDY5hSSaT1u6VDemuX79u5z0ajay+DB1PtVr9CJdC0oH70e12tbe3Z3OR0IbvCYVCFkqBmEh6wL9gQBDd+fQ8BDxhNuUVqVTK1slgMLBSHhzwJx2Xxshcv37d0APwrFar2cP+/9t78xhL0+u873lvVXVtfWu7tXZV9Ta9TM9wHRJDkRIsRnaMWBGkf8SItuEogQICigPYEAJLQoDECRzAyh+xEtiQTEQOJCs2tcSyCCEOQ2gLEji0xCFnYTdnunqr/dZyb61dXeuXP+79vff53r49U01O95TIeoFCVd3l+97vXc57znOecw7CBGIeDE/ST7o79uzZs6pWq/HUwP2N8GIRko3MNQJPVOTBcmlz9zEL3QVO6s1x3IWNy+fxNnimOU5WNIzNzc044cViUWfOnFFvb6+Ghob0wgsvRK2hpaUlCtD5+fmYt3hmZiaW4pAUtQoEMafT9vZ2NHfAZTzPDLlrWOSc5pzKqOAHB7UKEphaYEXFYjGnzbmpx7O758fNSyf2YS7QTzSrgYGBmJITzYcUlQ4QFwqF6BYGj+AaxPhwUA0NDalYLEYQFM3u0qVLmpiYiHlgALlxO/PcDx8+1I0bN9TS0qK1tTW99dZb0S0O0dGrOsJfwptEXBvPUywWo2MEoLa1tVV9fX1RCIJL8dyYYC64Me04EMB8cKG3tLRoa2srakEI8L+wmMzw8LAKhYIqlYqWlpbiRLGQhoeH4yBxqrHAiFBFPW5vb9fg4GBE1Pf29nT//n1Vq9WYYpAkyaOjo1peXtbi4mJ8z6OoHVdxzSPFOmhPAonZJHhoMM06OztVKpUi7gKohxbl7lSyAi4sLGhubk6vv/66Ll26pBs3buj69ev69Kc/ratXr6pYLEbg3HEPUmIQub22thaz8ZH3GFA15Zak7GcHpNEUYGx3d3fHTbq2tpYDXhmP7u5ujY6ORrOAVB1HR0exUgT4inv3GGNMRwSeZzJsa2vTCy+8EHEReCgIk729Pa2srOidd97R3NxcLBw4OTkZUxuwQWG6UgmT6o9o1u3t7dGDSSqHLMtyZWZYN0tLSxEzXFlZiWa/e/7Qwvf2avmWl5aW1N7eHrkwCDsww5WVFWVZLW+v11gHe0KTffjwoQYHBzU4OKi2tjatra1F3hXQAfhcCEFra2sxE5+HcCCAn6adGCHT29urnZ0ddXZ26tq1a9Fj5FyLwcFBDQwMaHFxMdreUJ/dI4GdCoELUhILjZInEMmw8QFSXVXnmggJJ+e5t4WWuridW+HajHtGSD4lKQJxYBmwYJ0RLNXU4Onpac3Pz+vOnTt66623tLq6qldeeUUh1Or7LC0taWlpKRapW11djYIYrY9nAsDd29vL1f9BnfcoX4QGJyjzgGt2fHw8LmA4LpgInJ4whDEF0AQxgXZ2dqIncH19/TF3O/McQoj3QEizwTG/ybNSKpVyCbo5jEiFMDIyErlWZ8+e1dDQUEwZwuZDEwX0xhRcW1uLwoN0CGiomK2Li4uamprSnTt3Ik/l7NmzUUNJvZbb29vRdUwEtKSo0WI2ghMB+GKyOmMaTYSUDUALDkVg1qE1t7W1xXHr7u7W7u6ulpeXtby8/FR7+8QImVKpFElFbW1tMUWjF0+nNlBbW5ump6e1u7sb4zx8c0O95iTf2tqK8SwsRq7npDA8Pc24MCnGwt8uQBxT8O+lYCamERsBAcfksdnIljY+Pq6dnR3dv38/gnzcZ29vTwsLC3Hh3L9/P7KfDw4OIgmMtIy4mgHHWWAIOfqFGbW0tBS1CcwSVHQ8KwQvOohaKBQiRuFep9HR0WjqoFkASPPsCAvwKTYqc8VzwxvC5EFoep0mqZYzGoHoYSrnzp3T5cuXNTAwoNHRUU1OTqq3t/cxE5FYuK2tLW1ubsbMABsbG1ETQktE+PI8YFVo3DgTcG64dsV8oDWCvQBQe34ZtNOWlpboRu/s7FSxWIxUCIJcqQzBfYAbHNNpbW3NkR456M6fP69SqaSWlhbNzMxEjf9p2okRMqjna2trevjwYVzI6+vrunfvXgQzpdqiHBwc1MbGhhYXF3ObHk0DAQShCfccNjLaBIAmJwU5Ztw0khqhANxHyptCKcs1bX4Keu4WwgXoB/iQpLgAWPyXLl3SgwcPND09He1ktA1O/Js3b0Zin3NvEEyYaO4lgXzFM/T19cVSwACULCw0hb6+Pl2/fl3nzp2LSZTgIb3xxhuam5vTzMxMdB8j7Bhnxg6twKswgLlQlgYPDQIMPAO1nT4hGDgsqC0NsQ0hBcPaU4Du7Ozo29/+tlpaWuJa2djY0Pr6ejQ3PPoazYWARbCilBDpglJSTrAjfD00xSPPAZpdiHAtgFw0KjxR6XxJNSuBkrQIH+aAOWaNO6aEl0mq5ft96623tLGx8ZjX9L3aiREyq6urMdZkeXk5SlvYvACL8DmwH92dxoaB/3F42MhjimrIQvQYIVR3p7g7sS71JLm24p9DcEh5M0tq4Ee+6CCeoblA0OJU9ix+Ozs7UYVHSDluwjNCsHJTzs05NpGkiJegjXDaw4MYHR3V9evXYwT04uJitNEhSiIQuB8nOf1Ey0vZzBsbG7FiJ4Gx4A5oa2ALLpTZ0J44Cs0QAHtwcFCXL1+Oz9zZ2Rk3IV4hkjehfZA+oVgsxrXDhmM+Dw8Po/axsbGRczCg4XjqBoBzNBIH07mmJ7ViPtva2qIXEWE5NTWVI9vxvAgvqSZAqEYAl8czLsJWBrchqx4Jqsg9g5ewvb1dOzs7WllZifl6njYrnnSChMzCwkJE0nEvzs/PR7BWarAO8QBgnwNYeiwMphI8BohaTCLeI4qlITgg5WFa4UnAy8Xp4QxTTjAHiZ1QhsbkzE9PFNTb26uRkZEYXIcWhQAFY1pcXIxcBwQRarj33/uTMllZ4ORJRstD8KQJvNwcdIHHuAJsok0wPng7Ojs7tbOzE1NQ0mdo/x7P09XVpb6+vhgUi5ZACggpXxsJ8wEvC6k9iLPhvdXVVd28eVNra2uREuACyk08D9KFUkBAI0GDgK+4eY+OjiIz3YNPpbzjAI8W6SydZYu304NmJcV67GhAaKjkM+7r69Pg4GAMP6lUKlpfX4+CBO/XwMCApJqWSjYDmPQeVQ0+mWWNlKe4wGHE/4UFfiuVilZXV1Uul1UulyOHBR+9pBz4i7vOGY3wF1DNSaYMw9PzkPCD+i01agPBhuQ1frt2QYOFmW505xOEEHLaAs+AmVQqlTQ0NBTTHTLRqPg8P6o0pwwnexpbxKmbamYIC05QTi/GjnHwzPuAmw8fPowLmDGkqiNu1O2lAAAgAElEQVTmHUIGYUFZVXACvDrgGS6g0DDATTBV8Ib5/DBGeNHAj+j7o0ePYgVSSY8B3swBuB1ktcXFxdzh4muEMcMbxBgjnPb29iIW42xxXxP0HSHG4QU3hfQNAMU+DxBPWX/MNR42IrPxwBYKhYg5DQ0N5a6NMOXe7BcCdBFWkmLaFciN6+vr0Ww8bjsxQubNN9+M0nVubi5WkOSUBTDDO0AQJTWL4JawIPDa4BJub2/XwMBAROxTtVtqqK9Sg6DFyQmOA0aAC5WyE7hdOcUh1zlBzd28RO4SRjE2NhY9X4QOSLWNOzY2Fj0ETpfHlnfg2V3Njh3hqsYkam9vz6XMwO4vFApRm7xz507MHojWgDckzZMLgQstE2b2/Px8NEXTeBpIhC6oMZOc0u9zUywWI07BIYO7lprZfghwsMCXGRkZiYLCPUPwQxhHhB+mNOPseEuKvzkHCgqCCyvPHtDf36/x8fFc7ev19fUI7uOoYC0y3pgwmESrq6t65513opmDlwgBhgmFcEBDYT0jJPv6+uJBzLzioWKdPXr0KOYHfpp2YoTMn/3Zn8VN4pXypAawx2JEAEiKGAI4C7YvnoYQQpxcj35dX1/XyspK/A4blQHl1EEY+cJ34BEuD/1go8CuxfSCTwKmBNGqs7MzZksbHByMz0/eEafEsxHZ7G7mOF+Hk96FDpusVCrl1N719fXcgu7t7Y3mJEKTEw8zAQ2C+7N4d3d3c5oY9wD/wJxgYyJQ+Jvm8+GAKVUTS6WSOjs7tb+/H+O50Ohc+IJ/OVOaOCPwGNzz4GIepyQ1BLWD+6kJyhqFEgEuiJBkvqWawBgeHtbo6KhGRkbimiGFw+rqau6z/D46OooaJQGm7iAgpgsQGwIrVT08EJV5RSPGLEzxRAh97plFE36admKEzNLSUi6ZM5PIIsRUcbdeodAoD4qqx+J221JSTnhQmlaSZmdnVSjkk/F4nI+zdQFQ2VAIDyYQWxxc4kmuvgcPHkQXJpGteCV4ZhaQTz5uZE5bvo8avbu7G/Pc0tAA3evmXhCeC4C9UCioVCrFfuDlwcPi3gdOVcaLjQ4RMMsybW5uxk1NRjuwLU5Sngu8hvl3tq/zczCJ2HTEGeGtwyTCbBsbG4vm69ramqrVasTiwMi4L7E/jDfAOgLPvUfOlYIr1NfXp6GhoRjUuL29rXK5HK8NDwitpVKpxOt6pj2EpMeTETTb398f+ybVQnIIa+BggWiJSci4AV4DsqNJYtJDRmRsIQOyz/D4PU07MUIGLoDUkOC4NuFpYHrACUHFXV9fj27ZUqkUs+AdHBxoaWkpmlvY5KiEaB3877Z4yn/xH5rb7IVCIbrYfeH5guXznBAOCvO87uVKhZff00/2y5cva3R0VPv7+5qdnY0LwT/b09Ojzs5OFQoFbW1taWFhIbrr0T4ODg6iW5b74WpF6B4cHDwW6e7PjAm0s7MTGbdtbW0xahqMDcCXgwATBY+Yu9zxwEiKeJCk+JwpIxitbXJyUufPn1ehUIhJ5km16ryftLmZjCDmBGdeJOUEAhouxEo3M3t7e2Omu93dXc3Nzamzs1PDw8M6ODiIY4RJ6NkSEfRS7aDA/Y1QwhPFeHhqh4ODA83NzUWS69DQUHQuAP5mWRY5RQDueLkYg+Xl5ThO4IdP045Td6lD0v8tqb3++d/Nsuy/CSFckvQlSQOSXpP0t7Is2wshtEv6DUmfkLQq6aeyLLt/nM5ArvJESW4Pp1LegVs/vT3pVXt7e7S7+a6XnMDW5SRkYbn50UzAeKCZpFyfXONwQYNKTXNXp0cXc388CfQbVZ6//VQFNMUlW5+7aGp2dnbq/Pnz6uvr08bGRszghkkD+OdxXr4REXjOFuUEdPIi/efkJGkSUclsEGgJPhYIWMwOvE5wfQ4PD6N6j0nF+DFeIYS4gQcGBtTe3h45RNAk3NTAU8f8+KHgc8XnmSfWJv2SFA8utGJYv2k4AKY7mfugA+ClYl1ipuBdBLSF54Np6wcQ4DsHKHFOPT09kVsDLwbtjjnHjY+bnlS3XLe/vz+a4U/TjqPJ7Er6kSzLtkKtkuT/E0L4N5J+TtI/yrLsSyGEX5X0M5J+pf67mmXZlRDC5yX9kqSfeq+bcNL7Bud/JpIoU7whbEAeGlUR5iyAGWonajYD68xZFper6f43fXKB44IotdMhBrpNLzXwJTAf/46bDC70uLZ7i/xkn5ub0+LiYtNxRRMiaJIYqfb29lhvvK2tTe+8847u37+vcrkccSO0F1zEaBSceqj5buphsoBFALpKiuYsgiONtGZzIkjcfELLoQAbmi1zyHVxhWdZFjdmuVyOXjEPvOR3Vmf4gk24RorG4FoM64n36A9z5WsT7xcUBqc18OOpK6RGnBjrXVIMMHVMCwaxY4hACIwBAp5YMzS/NM2D1PCKgjNC4GPMEDje1+O049RdyiQBLrTVfzJJPyLpb9Rf/3VJf181IfMT9b8l6Xcl/eMQQsjeQ/zhruME4ePOloTb4m48JDkeFDwklJJw1zVqvKTcCS01+Bf1Z855KVwToR/0L/Vm0ByEpWEeoFmAMXjVSq7JiS4pesDoM4uUZwewdoGcboqWlpbIjmbhjY6OSpImJyc1ODgYsQ3UbUxMhDcLnfFAa/F+OajL7xQQ9VwyvO/PjkYAb4T4oVKppJ6eHlWrVS0sLMRct8ybg5vgP5IiMJweSG6akb/HTSjGyjcVgsd5T64Z88yETrDR+U1De/AIboQ4pjtziLAl/IS0C5g73d3dMc813jO8SF1dXRoeHlZPT4+yLItkyuXl5UjMI6DVWc0wwhFseFXRsJ6mHbcWdoukr0u6IumfSLojaS3LMnbGrKTx+t/jkmbqE3kQQliXVJK0klzzC5K+wP+c/E5QcrzDTzUHyJh0XIObm5txAQAsugrrGwE1lmA8FzQ0BFi9z/E1vz+vJc+X+zyCwTU0sAPwI+6NwPAN5O51rolXJDWpnCPD+HF/qVHkvVwu6+23347eGnggAwMDcZOQxgEzSWqYDCxmH1vmknsjlCBEtrS0xNgfPudanrOdAd7RqAAyCVykcDwqPIcIRL7t7e0osNikHuzpmrCbh2g1UkPj5fvv5lnyuXGgdWBgQP39/TEinIRZeECJWM+yTFtbW1FIwe/inrCJ8bRyYDG/eJbK5bJ2d3c1NjYWTSVwKYI4MYlZg565AFMVDyzP5onRnqYdS8hkWXYo6WMhhD5JvyfpRrOP1X838289psVkWfZFSV+UpBBC5sKi/lp8uGZChUWNaslCefToUTwZ2AA01xD8tAJ8ZfD8Xu7Z8es0ay5UXOjwv9PKyZKGR80xDTZXOhaYB65FcfI7buWLzpN4S4rUcdicgN6cqFDwPTOc0/bRNB2MRQA5KRJNjWTt9NX76/Y9z4iXEO8VYweuNjc3FzUcssS1tbVFegMudQQjWiLaBGaO0xXw1NDvnp4ejY+Px2tDCESAuYnoh4OvF+coSQ3Th5AN8BloCrCTeS4ECdrk5uZmXD8HBwfRld/a2qpqtRrnDDYz7GQizo+Oaik0GA+eAfwK/AgrYGlpKWquUB7cbHya9lTGVZZlayGEP5H0A5L6QgitdW1mQtJ8/WOzkiYlzYYQWiX1Sqq817WbBSRKDc6HlE9ghFDxB+bEo6WnjmsDDJYDs+6qTvEQrpfyJxzgrY9RfJ2gv/Qa3MMBXFfj+T6bO1XNHU/C8+PuVNei3FODcMFcw9yCaAeB0fELd986KC0peqfc7Ssp5mPBHHRSGcAiRC/P39PS0hLjbQA3j46OoicE75Tny0Ez2dnZ0eLiYiSdecAia8PnGkHD3BWLRZVKJd24cUOf+cxn9LnPfU7FYlG7u7v66le/qlu3bunWrVtaXl6OWf5YI5iGfjAxt2gknluGKHPXJNB2mOPl5eX4fTQI1hiHASk5AP3L5XIEhvHygW3t7+/HQ4DxCXWWN2ZXS0tLJIJCOMUSgK3uz3vcdhzv0pCk/bqA6ZT0V1QDc/9Y0k+q5mH6aUm/X//Kl+v//9v6+3/0XniMpNyCxj4FJOMUTbkEzmORGhueBcXnfaFDmvM4KDAIyFjpaQTY5VyZZuaRm1FsIA/1R7DQDweYEWwIHFfTu7q6NDAwEIlbUPzdi+RjAobgp2gKNHsuGD7LNcghywKjL5JyGgwCk9OZjUWeWp6TlJTFYjFGxYMvUA/Jc9hw6uLxIsETc8hh0AyARHD4IeCaJcIME5zYn9HRUb3yyiu6ceOGPvnJT8ac07u7u5F/0gzc9XXAeHqgKiRRz1vkybshCPb19enw8DCGQHgBOMYWlzUCdW1tLUdCDSGor69PWZbp7NmzOnfuXKyqOj09HT1IgL0eBoJJK0kjIyPKshrHCa8SGu7i4qKq1ep77OZ8O44mMybp1+u4TEHSb2dZ9gchhJuSvhRC+AeSviHp1+qf/zVJ/zyEMKWaBvP543Skt7f3sfwuntNDapg2/PYTnInFzeleCdd4UNURMOkPG5Vr87+3VBWWHo/aZqFjw7rJhMBCZZcUBVHKWgW8hHnr/BEEBae5C2oPHKQWD/ErfJ4xg8NBX7DJOXVZxGAxUiMVJJ4M2L88I/3H3cs9vN4QDGanLEBi5PuYtuBWHnqQmq1ofhxQDvRKjUBE+t3T06NSqRQJkZubm7p9+7a2trZ08eJFZVmm119/XV/5yld079696FL2mKM0yNKfhXt7kOfR0ZEmJiZiek7WE+bsmTNnIhaHwGWeWC/MhaeNbW1t1LtCC0UT8dzP0EQQcgcHBzE/DWYWOZ/JHOkOFJwvT9OO4116Q9LHm7x+V9KrTV5/JOlzT9ULKaY/JMJ0eXk5Eq8YWDdrWMzOcWGhE0SIQEozrDMJztNw7ckZnT7ZrjE1ee6cW5tTFoyA/uPW9vrOfIbFywLgvggCuD5oD5yMjJl7UFzo4PoF6PRTHXXcNTjiWtz16ioywijdYGhRLmh8fAuFQoz89Sx2rvmBrdAfx5W4H+vB8RBfE6mWw+ZGCDDOVLtcWVnR3Nyc3njjjeg5KhQKkehZrVYfM9/d5c0YgNd4QzDs7OzEYFg3PXluvGzMbaFQiFiWM8A9cwAHFZoiHBtJEQ8ji56XjmEsHfP0vdHR0aHh4eFI8sNcZU2RPfC47cQwfiuVis6cOaORkRFNTEzoQx/6UMxjMj8/H3ECqbHJ0RgwofzU9YRGAG5s4IGBgfhdtBdKmGKnOqHObXt31/r7KZaSYj9SQ5AB8LFIyZ2Dnc/G9ip+ftq5SSg1XOoAl07XZ9POzs5qeno6h/0g9CBlpRgWblCAce6LFiTVzAlPOAVt3gUtYCYbG+IZ2hnR2vv7+zHFB5oMIC7z5f12egPmHuPCXEs1LdC9ZCSzwhWMdstpnh5mbEoX4MwFRdVIhenmJ+PlY8pG5bWWlpZosk1MTGhgYEBbW1uanZ1VtVqNOItTB4giHxwczIHBCJ5CoaBqtaqbN2/mhHl3d3eEBsg86Ro/Jv3+/n6s301fGdNisfgXNwr74OAg1gwul8saGhrKnUZsGgcheVgGmVPe2agOTDJgeCfAZ8AN3G3sC4Hmm9BBVsco+A150Bebu3sl5fgj/hy4KvGWOLvXT2o3Bbm+f9bxDQQVY+SmhXv2fIPC5ZBqGw5h4guPuWEjpL+ZQzx3lEvhOgheTnRc1zxn6pb38WY9IPTh4ZAWweOEGBMK2DHHzIPTJ7inpNzmRugAuEuKsWtoQO61RBNnDYBVEV9GDNHw8HA0TQYGBmJ4gLvMMfH8gELrYZ496yOhBIODg+rq6oocIA4UDmWyHKJRo7VhihFv5bjMMSDWXDsxQsYjeVdXVzUzM5ObVMcb/ORhwUiPZ6zzCZIaHizMClyFVAPc39+POW395ErRdEwndxs7DoMGgeuUheE2OsAm5gCpEiB2sWD83k54c7WeZ3SNin67sEHVR9Piej52qeudjcocpZiRpMc8YPQZEwKBw1gT+8KmhwhGdLtveMdfOGDot4P3bCDSOVy5ckV7e3taXV3NOQ0QaG7K+d8IFcaT50jJhsw92FR3d3fsJ+8xxowTnhwo/h0dHfFvwgvwaHnuI+aBMT46qtXkOjw8zLGUSQDOARxCiDwXiIyMAwLRA4tZrxD+oDaQnKtcLmtnZ+cxT/B7tRMlZPwk8QXNJsIOBhVP3cFsbClfOdE1IFRoN4E8pyqfI/LYNY/UhY5wafab64KBUPkRTcujh+kvGgSf8T5KDUYrLcuyaB65QCaK1s0W14SI5HU3KxoRr6HVtLW1xSxsjC3vOXmQTeCeKDQsEoyRjoAIbUkxWRUhH2ihjIljPq5ltba2xhIljN2VK1d09epV3bhxQ4VCQbdu3dK9e/d05swZXb16VXt7e5qbm4vlYckBhPnnEddu5jpGxzrgdTRfAFcXSHyHMeF7mG4A44DzXr7Wk39j0mE2ovGRHwiPE2uVuapWq1G4o8mnaxOB5AREhBD3JoYJrdGzJRynnRghUygUcl4YAFuSJLsZgGsYO9Ltcs8J65pGyjMAH5AaHidXJ8nfUq1WczmG3UXNPTlpuD5mnLvZuQeRyQ4skziK+kf7+/taWVnRwsKCZmdnIxDKRgDXQa3FWwOYjN3NaYZwo5YQlSfZZOALjhMxHghHqSG43buAkEcDkxqmLXgDGkZfX5/a29vjPdngbGg2bRpqQMQ5rNRSqaTx8XG98sorOnfuXPSIjI+Pa2xsLObtIaHZJz7xiaglvvnmm5qfn9f8/HwsN0MhQYS+CxgOCTfN0NDcFZ+a105VQIBzUOLRASPp6OiIcWRSTXjNzs7mBJTTFVj3DuSyR8C7nKQqSVtbWxGP7O7ujjFP1HYitSfs9/X1dbW0tMS1g5Nlb28vpgI5bjsxQgb3G4PjNj35S9m8aCxsLE4PPis1UgK4eotgAFvAJepJhlgIbosCEjrghYBAxQQc5QRwcNS5MIQCwE8gwxt0+ZdffjkWultYWNDNmzf1jW98Qzdv3tSdO3dy4C/PQCiFB4P6GDpIfHBwEHPnIkxcg0QYsqlcc6CxgRg3xh8thHshtLgntH8Ek8+p40iOOaG94G4mVwu5gDGZOBwqlYqOjo5i/SA4IWhMEOmICMfD56lA6I9rJe4I8LlmfPgs435wcJDD96ATkEeIRFOk4EDwPHz4UAsLC1paWoosX8YYr6RrjCnXyV3MCPa2trZI+COHtGNbknKlcw4PD6M5x5wyr4ODgzp//vxT7e0TI2T29/dzgXPeHBdgYF195fR1sC6NgUIgwDlIAVy0CFysHqcDMg9G4gxP+sTJ4GRCB1UlxUxsnA5kjZMUtQvq/xSLRV25ckUf/vCH9ZGPfESvvfaa/vRP/zRWDNjY2Mi57x03QXAwLq6pscE52aRGbl6e109PX2DMA8/vQpV8wZIivZ3kScxNa2urtra2cgxjNzWcKMdGJl8KuVCoow3eQPkcNrETLJl/opWdsuCeJ4Qa4QkObDK2rDn3aiFkwDTAlbIs0+rqai7XDcKMgwuNQqodmg8ePNDW1lbUIh48eBDXNn3nuX2cJOWoEBzA4Ds4OIisZrwgXIYQYm6m1tZWjY+PR1yG/sASp1rG09bDPlFChlPMbd7Uu4L6yiKUFLUJ7Hde84ZqyOYpFosaHByM39nb24u1fjk5EEhoSNzfbVu/vvMdUk2MTYzHhoWxtram+fl5ffvb39Y3v/lNvfHGG5qcnNTAwEB0RZfLZU1PT0dhSBpNeCVQ7qWa+UJULtR9T0qNqxctgudAYLiXBFuexQsLlc12eHgYX/PE1HgsiNhFwHBaA1gyJpygzC2CpFAoqKenJ/Iy8BINDAxoYGAgeuLQQgqFQozYJu6IQ4tN4qzpra0tTU1NxWfF9EEDds4PP4yfg96SYlAm2hqmDCbz3t5enHOELrXdj46O4tpzrcg1dfeUotVgNmdZlktARr9Zi4C+TngEjkDwjY2NxeT2h4eHsWrD6upqnGvWwF9Y7xL2PoCvYx40BpnXOQnZjGfPno3h6NjOaBKOgTj2IykSmRBc0MDBIiTFazo3xvvlng9navqC4LvcGzXfJ3x2djbSxcnqT50gPDVwS9gYMD7xiCBYWKSu8jsw7sISIYmHh+sh8N2r4wBxOi+S4imKBiIpzoszW51k19XVFYWNawuuDSLcdnd3tbS0FDUXNhZENamRNwUvHvE+CB4yKWIqESHOWPBcvoZ4j76hGVQqlZzpl9IauI47C86cqeVpPn/+vG7cuKHBwUHt7u7q1q1bkXJA2RmfJ67lmgx4DGuKg5P15p5O+odAgh0PXACAvLKyEksTgfXQZy+hcpx2YoSMb3oEAQPhHiEmHFAM7cdtVEk516HUQNPZ9EdHjezrLS21zPVwK4aGhpRltdgNNiMCB5OrmbvYQTled8HJcwISI9AAV4eHh2M9Zsypubm5aEq4De7PhmkIHoVnbHt7O7rOcRV79jk2C7R6Nio2PaC1u1FdI3NXNaQ/FjhmzujoqFpaWqJmgXu+ra0tnq6rq6sxvwlJtzmFievhAIHJ7Sxuj68ilzAbdG9vL2pCbPyWlpaYJW5raytWsUjDWtzLhgbDhkQr4b4pXucOAh+/np4eDQ0N6fLly/rUpz6lH/7hH46mI/Fb29vbMYG3s3tTzx9rmEMTTQjti7EhgZV7ahEY7e3tcR+AY21sbMQ65MQ44e5OE2wdp50YIeMnGBsD8JZ4C1dTXZtwToiDc87CZZGg9uGelWonQbVajf9TgJ0NgMovKTfIbED6AKjrk+0AsYOv1AFqbW2NMTwtLbW6UrgnKSOLtrayshIXuuMkBNdJDVCc+0ETB7OAuYuwOnPmjIaHh1UqlVQqlbS/vx9LaACYsuncJc6YOisW1zmeo76+Pi0vL6uvr0/j4+MaGhqKApxn2t/fj6D//v5+BEcZS9IuAI56XXQ8bfSBxOakT4CrwhqA5SopntyA227qYeo48RPNBQInZgXPy+cRBI6VhRBikOvg4KBefvllffazn9VHP/rR3KEIP2x+fl6Li4sRh8ER4bFhrCUOXObD8UCuTRVITEaIfOQV8gKICEWei/5j2juwf9x2YoSMpHiKOhMVCndKz5byyYLAGWgMunNOXFsC2GIht7W1RYITICanJkXY2tvbVS6Xo1aEKeEnFwv06OgousrdxHJBg3oPV2VjY0P379+PQGKhUIh8Gq+RzEnkAtU9RKjMmGssVvLM+vihTqOFcGoRhlEsFrWxsZEDvTkQMN2w/9ECSIG6uroahdvU1JR6e3s1MDAQo7Exx9B6mJdKpaLDw8OYq5lTG80E0xLt0c1A+sG8dXd3x81Ndjjnq9BnSblc0Q7yQndg/bibn7n2w8T5Xe4d6+/v1/nz5/WpT31KH/7wh3PenXK5rKmpKU1PT2tmZkbVajWafGh/jgu54IeNjUOBNYIGhwbMmmat00fMWMrGOGWC8aL6Jmzqp2knRsg4UOvgKTZjmnLAm5sn/KBlMBmuBTll3eOhKKXqAC6Z1fDWEC0uNcBqd5U77iMp0sc9QNMBOAQrKmtbW1tMHYr3iJORxQS/B00Fz4C7UNFeAHGdEOaCDmYp2IK7tJ3AyKZC4/DQCzcpPDUqm5B0DWtra1pZWdHg4KD6+/tjEB4R1tj+Y2NjKhaLGh8fV0tLS8xhjEmCloM24ukqHYuQFKn75J7hNQIz/TDxZ2adUP4W7Ifr+wHh2ovjiawz+nLmzBmdP39eExMTjzk4PLWF43meayd1UTPXjDmvk5gMQcLa9UOaNYr5PDY2prGxsYh3wQvjoEczhon+NO3ECBlfJGAGaDOg6MT0pJvabdSUEOWbwIWMlOcV4PFB2GCr+kaXGvW4pUaApmMTXJeNRgyVx/HQL04JJpPn5JoAufQTszBNdO0mJqcYz4jA4G/X5iRFDxr9RWt0T52TDxk/93TQF6mRp4aNxyb0sjNcj3AKCGvgMQhSPDycyIwbuAvCmJwnAwMDuTQPra2t0TsFHuTpCmBewziWFGPZmA/cvQg0uE6+NhhTDjQfJyeKeo5jBIekGAhM9VC8hcylC/csy+IBAiGVAxMtlsMSmIE1gMa6sbERNbWurq6YdyaEEOvPgwN5/mKES7OD/t3aiREyLHImhQkEiccWRsJyEqCNOOM2FSa+SdxsQQAhRED1AejOnj0b2aOQ61yIsQDSGKtU2BweNsrc0h+pkUwqNUP8lHNvlIPW7h7l+vQJIcMphefOQT+wD/LB0F/KlcIGPjpqlOk4PDyMgpBwBk5dcAlOU8BZ7u1F4AjXkBr4FHhRsVjUwcFBrJOEQMWMc20VExaAeWhoKDefbPpKpZIDaSkwt729rXv37mlmZkYrKyvx+TCPEILgYikZlI3otAo3x/iNgIFE6N6iarWqu3fv6tatW7pz50508fMcuMTRQh0P87UOKO79dQ4Q2izk02KxGM3I/v5+7ezsaGZmRuVyWSGEWDoXbyOpKvb3959YtPBJ7cQIGaS7e23YWESQSg2atmMLDro6TuJ8DoSMo+yooEwoZlG1Wo2mCSctm9Q3kZtiUsOOdy2LhkbjC4+TElUXcM9Vfre7vb+uCXF6I6gYK9yqLDTuzwnlAZoIDcyDYrGoSqUSi9AR7oAQ7urqitrG/v5+rhRryr9hLDBDORy2traiABgdHY0aJNdgY7NhAMIZf3erY76AtRFvUygUotsVLAWAmrKwHuTIOACceigJWp9Xn/T1yLP52mRtDw8Px8JqgMblclkzMzO6ffu2ZmZmIibI2JACBK0mNXVZJ4QquKmP2cZ69NcdxN7b29PMzEzU5rq7uzU0NBQ1Y2K9vKTKM0kk/jxaM02A5p4Zmm8+6fE8utLjHqh08/s1nezEQkK4HR01qjBK+ZAFj5PiOr7QUrcjCyX1EPBdd727KUVjYTvo22yx8Vn+5zVUXTxbvI6GAvMYWx8ejn+XxQmHAw0PTbQZHbtvFvsAACAASURBVB0Okvcb0w8A1lVzzCD4QgToAfoyBq65AcijIbJp3evCs2NySYrERbAH1gcmEZqLhx3A8+FzTr7zdUYw6ujoqEZHR9XX1xeFIVHZVLXkECOeCha57xGEGc/uaxuA3zUqPoOWRsoGosYpb9LS0hJToEAArFQqevToUSxzu7q6GpOWP007MUKmq6srqmEObNHY6CGE3GJgA6YuPamhzUiNDZt6mficbwb4AGwe1H4WMmQkTjQWDZOdalksdKmRItJjgjAZUpA3bZiUvO+nERPvHgh+UkHIfdjEbLrDw0MtLCxEEh+uTsYV744LaccF+B7PSDAkGfQZA/dMHR0dqVqtxnALwjsQCvQPzWtzczPW/8ZzSGpJUnYQF4RJxJpCIG1ubmpxcTFS/6Ht+2GAC989TjwX48GYY9qzznxs8G6Njo6qVCrFZ+rp6VGlUtHy8rKWl5e1sLCgSqWihYUFLS8vxxSYzQ4m19Z5jXWL54fDknxAY2NjsZzx0NBQLJJHJQNCECRFc5l9CcWCQ/hp24kRMoODg/FE5nT2uCMWtrvoiEFyLcGFidTQcNL3Ut6Na1I03wzOl5DylRLZQKj2bibxnvMWUF09BgYPASd86lnwhvqbqq0uUP2kZ3G4GcOJ6ILRBTWmljOJ/RqYm+n38LggRD3DfXp98CLA2XPnzmliYkI9PT169OiRKpVKLmG6pJwA5H5gCE5kJEbKPY+kUFhdXY2JuB338fn1jeXZDB2E9TVKHx2fAfsolUq6cOGCzp07p6GhIbW3t2ttbS2SEOkHbn/XYFJ8h7VKY1xde2Vt8n9HR4fGxsYiHwrvE8mzSqVSFDKMKVouWB8Yle+Z47YTI2TOnTun9vb2yNbkQYiXYGO6UMA8YOOgFUj52kmu3dCcPEVz88u/x2LBXqdoFxvVF5sDk9yH19GI8ERgdiFknIUrKeYJedJJ6b/pvz8z4yA10pB6AijHsgD1GE/3hLmW52VeXJCjYfkceV4YsC2nIcA/QojTDy94h4aHV4YTGqyITYRw5uTGS9jR0aFyuRxNsUqlosXFRZXL5QhAo1HCOAbDY41xADA+aDd+IDHPaJgA6n19fZqcnIyJ3CXFlLLT09O6ffu2Hjx4EDUrNAU/WNwLyXjzmz6yDohR4xDDXCNtJjgbwoN62mjv3Jv1igOEcYGw+TTtxAiZCxcu6MKFC1pbW4sDDtjHyUb2L6fpE/buoKyU58ww8U7NZhG5RsHGY/G4+jgyMqLh4WGNjY2ps7NTMzMzkQMiKafCurDhNT4LECs18q6AR0CG8whyGl4cqZHCwVVl16xczfaIc++nX9tPScaAcfH3+Z97O37mwu/goBb8hxDHw0Q5FOz+np6eeGICcq6srKhUKsX8L3g1PK8NGiOaXEtLS6TDA6bDa9nc3NTMzEykyHMPhDfzw3VSzx3PwFy5BpNifJjzaL/9/f2amJjQ+fPnNTY2Fun9N2/e1FtvvaWpqSnNzs5qeXk5litxzdzXE81Z7Qhl4uw4RDg4SO1Aalm4RmhZCD2EKMC77w/wNeL5ANmfph1byIRaSZQ/lzSXZdmPhRAuqVZzaUDSa5L+VpZleyGEdkm/IekTklYl/VSWZfff6/rY4YSTz8zMRGo1QBVsXLwDnDxMBIInNS8QMO75kRqCCAGAaYX2IjW8Xuvr65Ht+OKLL8ZaPbdv31a5XI4lL9ybBWDoJ657T9iQrgkA0NXHPC44fyY3+1xzcBYoY+EnLPdIzUT/381KhLOUpxT4+HENKQ++O1uU/hBkioBhkzBWHCLz8/NRALkZ6+xrvodpwbh7xDljQpIstBA4O3jmQggRzzg8PIy1sx0Dc3MELxtj4aA2/7M+SOt669Yt3bx5U5VKRd/61re0sLAQyXesE9fIXRtl/Fxz5HO0w8PDqI2FEGLEeaFQiNybQqEQ2daehsJBYp8/D/r0BG/PTMhI+juSbknqqf//S5L+UZZlXwoh/Kqkn5H0K/Xf1SzLroQQPl//3E+918XX1tZy6SnJ4rW0tBSrEXrcCQsDPginmgOnLvFdQ8FudewGQFlSdIESQMaEsXFRMSVFYhP9Q+Pgnr4pmUCwjBQ/8oZLlQ2BJucnXAoOs6FT17Z/nmfnNfrVDLdKvXcuNMAr6KsT8FxgoZ25oMT0IBaLTcbcHx3V2Kp+XeYd4YWGhtBEQDNOrhF4/hqeBUFFLA8aM+Ptwtn77i51Tn/AXMbHzSvWL0IWtz19pD/N1gNrKNWK+ZuxwlR0c7OlpSW6vz0RHL9JnEXRPbg1bn5zIAJZME/PhPEbQpiQ9B9K+u8l/VyorbwfkfQ36h/5dUl/XzUh8xP1vyXpdyX94xBCyJrtJGvVajWWIC0UCtGNjEkACcjjmBhwTkf3pLjwoLnpJDUWO69xepAHhRiilpaWmDqyp6dHFy5ckKRYJ8o9XrCS3VNBX+gbJ5WU9waFEKJKzvP7InPB5c+emkO+EHluvu8ncLp4XUj59RgrMCPvE/13DMdPfq6V5lLJ6p6fFCxnPCkHg5aAixlzGS+Xxx7xeTAeTmsy4THPpD8FN2HtYE5533nGQqFB8pPyVTmZb9dYyc+COY+XK4QQNQzm2eeJdZEK+mZscX8fAcF6RfDRV0x18sGAu3V3d2ttbU1ra2tRMPthASkTsurW1tYzY/z+sqS/J4mqTiVJa1mtDrZUq389Xv97XNJMfdAOQgjr9c+vvNsNjo6OtLS0pPn5+Vxumc3NTW1tbUUXJDYjdiKT5FwMqaHROC4jKbfJpAa5D2EDGMpAcgKhJlIjqLe3V9evX9fHPvYx3b9/X9/85jf12muv6Y033lC5XI4pJJwEhbrOYkAjY3J9YXk8jm9kZ54i3FIBwvN5eov6fOT4FQgDTmHvh/fFT1Y/6Zx/0sxkSjUlB8HToNfW1tZckil32RLwKSlXMN7DK9yscryJOeS3Pw9mWqFQiEGBCFE3RVKODM/Ka+BAly5diprvhQsXNDIyEl3pq6urmpubU7lc1vLyclxnzEk6fmgpzBfP1kzbxAylr4QA+H7geRHq29vbWl1djbW4HMtDEMOPIcFXW1ubBgYGcnW2jtOOUwv7xyQtZVn29RDCZ3m5yUezY7zn1/2CpC/wP9G+qGV+grnJ4LExLsVdqruWkJ7WzTYcXAxnAkuK0hsgMcsyLS4u6tatW7p8+bLGxsbU3d2ta9euaXBwUJcvX9aFCxd069YtTU9Pa2FhIXJLcLVyirvr2zUb/50+EwIn1VYcc0m1EL7jgtaFiZsC9MnH0IW1m5v0w0FSZ6GymX3DOGjppht9dGHlzc1E7zfPQmMDpesAoeZjxVg4FtfMxEUYuGmDNgroSrQ4gmVoaEjj4+MxOTdcHsfY6ANgLRqT1KgLj4bupmpqUrnwSQUvgC+fJeseGiCJzXgW1gljCE+I7JDFYvGxfM/Hacf59A9K+vEQwo9K6lANk/llSX0hhNa6NjMhab7++VlJk5JmQwitknpVq4mda1mWfVHSFyUphJChRjqCLymXhcvrELFAUYd9sbFg6teOJoWrmb5pUhPC+7C3txczp2EybW5u6s6dO9rc3NSFCxdiKk9J0c03Ozurt99+OxYoL5fLEUhzhqwvllQI+okrNdzqNoZxLJj41JRMG9oHY8KpjTng92bhYyYwNvTT8RK/PmPpG96xKsbcNTp4Gi50iIlyoiL9d1PFhZhvZBcy/LBJ2FgcLpg5mHN8D80g5Z3g2qb649jYmG7cuKHz58/r3Llz6u/v1+7ubjTvPA+xExf9oOQZ0pAAn58UuGftIPS9j+5sKBQKucJ2BIK6wED4ZFkWQ064JhrN0dFRXOvHbcephf2Lkn6xvjA+K+m/zLLsb4YQfkfST6rmYfppSb9f/8qX6///2/r7f/ReeIxUEyabm5sRxXYVNcsa3h6o4lI+6tlv8SS71j/HhnWQMsU8mPijoyOtra3FxQ1YdnRUY6seHBzEpERXr16NUcQhBPX09KhcLkf2K6on5hf3cFPIBYQLQI8w9v6lBMAUc2EM0Vogl/lCdc0A/ATbnLFJzQUa/Xd2sP/N77SfCBm0Gti+BKYeHBxEl7PjAX5thICbXVK+yiWCCZesmxD0ne+hMXvyL4BT3L4IjL6+vsikvX79ul5++eXI6iUJ1OLioiqVilZXV6NpT3JwtL90TN2T5NorHiTXXNOxxNUPeQ7sx8cIM9RhB58zP5hbWlpyhzveqadp3w1P5uclfSmE8A8kfUPSr9Vf/zVJ/zyEMKWaBvP541ysv79fIYRYeoRALLezOSXRMFzi+8ks5QMp2WguUNIN45gEP9j/RBhLDcYpE9Hb2xsXYnd3d4xPGRkZiVGxMFqJZiUnLBoSkwgY2czNjCZASwWsaxksOMiK/j1cnCkewkJGCPk9HFh0TSV1q7pZKjUisV0b4iTFHAaQRYCQpsBZ1oC+3NNLd/Cc3nyDplgNYQ68ljYX4lyHYFC0LZKVT05O6mMf+5guXbqka9euaWRkJGoBU1NT+ta3vqUHDx5oc3MzJxQgX5LSwTc/JpTnAuJAcg09PTDQkJyUSGwZcwYpEi8U5WpgaOOVcvwQT1JPT4/GxsbU39//bKsVZFn2J5L+pP73XUmvNvnMI0mfe6peSLp9+3asQUMgIn5/z7gmNT9Jm73GxLoG42YTDQGVqvkusDY3N7W8vKzW1lqOEoL6pMZipnV0dOjChQsaHR3VjRs39Pbbb+vrX/+6lpaWtLa2Fjc5KQ5IBIRannJcHPNwM8CB7WQO4uJFS3CtjE3MuKQaEWaSmyTcD8GS3hPgnPlJr+3fcWazjxsaBCk3eBbGONWWeE42nGMazBvfpbk5kpp6zdaSb2JKCQ8PD+vFF1/UJz/5SX36059WT09P7t6EC+C+9gqczDs8r7QPaMtoaawFX2OOtWFG4WhwTxTjBb8Id7vHhKGFw452U55rYNLiyPgLm7TqwYMHyrIsF+PjahvEKyjdTICr466hOLgp5VN1Yqa4OeEu5HTicZnu7+9rfX1d8/PzsV99fX06d+5cUxWyvb1dFy5ciAWx7t+/r3feeUdTU1Pa3NyMpwehFNvb2/GUdfvcNRUWG6e4NzZ3My0ly7LonnU3rIOGjA2AoKQYz8LmQBtxs1PK0wN8k7pgc9MAt27qboVV6hok10QoeXQ4AoM59sMEfCE1FZ2P41gY68DNBkytwcFBXb9+XefPn4+1sCYnJ3OpLHZ2drS8vKzZ2VlVq9W4bgCuwTyoSooL3z2LCJBU2KWC3b2MfA8vF4cEv9GcWEOsXbREeDIHBwdaWFjIsZvhrHEfglOfpp0YIUNaSzgqHjXLAsVL44KE3+5dST0EUmMzAPjxvRS5Z0H6RmexoEJKisl8yNh36dKlpoWv2AClUkmtra2Rag7OQBE0mKqrq6sxkG9+fj4uNs+OLzU8Ls0qCDAuHtDY0lIrOco4OGmN8WEjY1akMSrOk3HTrBkexJxxPRfm3ugzY05cWOoVacb/QAjzDLxH/xHQDvYWCoXHXN/uMnfwlAqM169f10svvaQXX3xR58+f15UrV6LJzDMsLi7q/v37sbTxz/7szz5xrX+/tRMjZDCTOBFWVlZUrVZzMTAsMgexntRSE4bPu5fA+RVuWvhpKDXMD2xcrz1969Ytra2taWtrS5OTk3rhhReifcx9AOMGBgbU29sbrwspizQHfX196u/vj8TE1K6GUs/GSjUE11AQQI6buCaH4GnmseL6JHVnXCgj4rWsEcQu6FPzxMF39ww5uY3xduzM8QfY3bB6Ux4JGAbPDMZ1eHj4WC4ZrkmfPGQDYVQqlTQ4OKhLly7pU5/6lF588UW98MILOdOI683MzMQYJNJ5nrZGOzFCBtW5q6srJs85e/asVldXValU4uLwBexuV95LF7p7Z/z/VEj551KXtl8P+9e9FpCfyuWytre3NTExocHBwce0mtbW1pg8u1KpxMqQd+/eVblc1sOHD2PNpEePHkWehWtsvlkRHh77graBnc19oZlT5RBtxL1KLoh5bjhLIYQIFnZ0dGhjYyOq/45nuUDBc5RqimhICCDHEHzMHRtAoDjOwxrAFAYcdY0OQeOCOe0nwhZKfk9Pj65du6YrV67o8uXL+uhHP6rz58/ntBf6MTs7qwcPHmhpaSmmjTgVMvl2YoRMS0sjaTGh5/AQjo6OYl4Rt5nRbnxB+oKW8kGG6f38FMZGdWzBVX+pcbJSybG9vT26Jjc3N6MJNDMzo4sXL+ry5cuxYJy3M2fOaHR0VP39/SqVSurv79fU1JQePHgQC3pBSnQtjjGhbw4AwxdC++ju7laxWMwFEOJdkBrM2RBCTiNJwVk3izDR0CykmokCCO4aDZtbamiNNBeWbnKkcwRWgRsW4eFzhACFvZtqQoVCPtQkPaQwpaAgjIyM6Ny5c7px44auXbumS5cuaXJyMvc9tO0HDx5oeno6po9YXl6OQZ6nrdFOjJAhdwq5Y6RG5ntse1R1T5zEQnX1V8rni3Eh425w/ue7aAI0/86ZM2fiZsN0AqPZ2dmJi/3OnTvq6OjQwMCAXnzxRX3oQx/SlStXNDY2lru2VAOGJyYmNDw8rKtXr+rBgwd68803NTc3p6mpKd25cyfn4mZMIK/BK2GD43bH1CK1YlpCFRODBh7GRk41BUkRL4GGz8ak5jWnuAsCfuC2NDPPuHYzAcNrCDT3DPEsWdZg86b4jAsn6AGsIxJv83qxWNSlS5d05coVvfTSS/rQhz6k0dHReB/a4eFhzF734MEDTU1NRRwNbI1KkKet1k6MkOHUZkERDIkwcdq/aywuaFyVTc0k9yKkfA/pcbe4Nzamaw9HR0dxY+Lic41ldXVVX/va13T37l1dunRJH//4x3Xx4kUNDw831WzIV9Pf36+7d+/GhNPlcllra2va2NiI0btE866tram7uzsnND2hUsq3aW3Nl6GlYZp6jJUD3swPQtUrZB4dHcUqlIDhpCN1zci5Tu7ChpfhGqnU8AYi2LlGM16OYznME5oU5hiuaMqbUKmBEjelUknXr1/Xq6++qo9+9KNN43N2d3djvM/09LRmZ2djjSJ+SNt52hrtxAgZADsWFy41XMcsUk4vx2KauVNpjgmkwskFTer+lvI8C/+8q90EcbIxPWFSCEHb29uanp7Ww4cP9dZbb2l0dFSXLl3SuXPnNDAwkNNuQgi6fPlyrOTX19en2dnZCCq2tbVFQh/CAqKX99dZr36S47r13DDuOob8xYZ1d69rECTBJtEWbk1cpfCb0GzwijlZz7UcBAbCwF32jDHmnGui7tbHJPP55H5ufhN1jSZDSZDx8XF99rOf1csvv/yYZxLsCe3w7t27unfvXswyh2BFCH8neXC/l9uJETJUBsCvT0ChazFOEHI3qmsYTt5KBUtK8ErVdNd2aKj4fg8nquEGBksioxvkJkmxlg6kwv39fc3Nzam7u1uTk5OamJiI2k1LS0usGX3hwgXduXMn1kfu6+vT4uKiNjc3o4qeLmrnGTnXRmrELbmgBotCI0GrQRghcNCe3JsFMMxn4bjwHKTQ9DgcJxqivbr72OcNAeTChZ801YJjMfQNkwnNy/MDj4+P69KlS/rMZz6j69evq7+/P1flk4Z7en5+PublfeeddzQ7Oxuz2bnQTftx2k6QkAkh5IqoufbioGSKKTChLG7nbfhvP9n8//RaKc7jzTUm/55n8PeNT9RqoVCIgXIQzKCqLy0taW5uTpOTkxodHc0t9LGxMZVKJb3wwgu6c+eOCoWCent7tbq6qq6uLq2vr8cYGbQGcCV3b7sWQvpFP/Wdsg+oC+hOLM6jR4+0sLAQS2KkNHfnohAaAK/JwwmYH/cgpdoOXkM+gzDh+nwvJVQCcCNM6RPR/P39/bp27ZrGx8f1kY98RC+//HLEXZo1Svd+4xvf0OLiYqyFND8/HzEovGVeOuVUyOTbiREyAwMDkVXq5LeUlu7eICkfBJnSrt3l2cyE8us50OjaUIrdeMOs4yTz3DaEC/T09MQk2iRyxj3Ne0dHtazwKysrunr1qnp6euI9z5w5o7GxsVig7MGDB1pYWNDdu3djxn1+I8DYzGmUtLNmXStAk3DBhOnk9YWIkid62cFkJ9WhOQDc0yc3f/ksmo0T7NKARwSHg/Rcxz1tJBsn2RjBlkNDQxoYGNDY2Jg+8YlPxPSpT0pZcHBwECs7Yhp5SZZKpRI9aq4tnmoxzduJETJTU1MxCZUvSNc+3OXMa1Lz7Hepve7uaMdUEDZSg3PhLvHU7PLPAlQTVeuJrff396NJgwlCBC+empWVleiJgqy3s7Oj0dFRjYyMxBguqRb9evXqVQ0NDalareratWvRyzE3N6f5+fmcG9X7z8b3sUAAp+YKdH6a08sRDDCHPV+uewBJieH0ATd9CYD14u3cHzPJ0yt4OlMOgs7OzliUjejogYGBaPY474Xqjbiln9TAXe7cuaPXX389ai/MLT8ep4UJ7Jp2s8DL7+d2YoQMvA3YpFKDcp4yXN19TUvNIW9+sqemkZtXrg35a1zfcR2+v7OzE09hOBngAeAsJP9Bbe/r64saEKc2HpmWllrNoOXlZY2Pj2toaCimjSBbv9RgEoP9eGLz5eVlbW5u5gqE0WfA9VRL8yx6DnQ6x4XxIxUDOIeHOFDzGZKia5FgJGgjZMBHQDGOflAw9x5WgDnU09MTMZbOzk4NDw/ncCEih2/cuKErV67khLa3LKulAoU2QIkS6AlHR404KOfmpJkZHT88bY12YoQMp6jU2KwuXPyk4DMp09SvxWecA8NJ7F4V9yg5SOzejVQbYrP5RpAaEbtudng8DcLm4cOH6u7ujqcw/cej48zfcrmsyclJlUqlXGkQosLpD9nYBgYGNDMzE4unQxbkWfwZIfhB2edZEHgk7ErHYHt7OwLIaChoSjw3wgkXcW9vb4zy3drailgVJlearoHNzDwyR+AueIrca4SA6enp0cTEhC5fvqzLly8/5sWjIVwqlUqOeU21AgSvm3iHh4eRy5Xm/vGI6dPWaCdGyLS1teWSQrPoUZH9lJXybtB3c10774LmuIVvOL7n13Ntx4WW296+qHDhej8hsLEBse/ZGH19fVEIPHjwIJ7MQ0NDMZn43t6exsfH48br6urS8PCwurq6NDIyohBq9XBeeukllctl3b59W/fu3dM777wTQzNIL0AfyF0DbwRTDhMB3MGBTamRgsMjrokxQxh5eAP4FKkS+BkdHY3aHeVINjY2IvdEygd94lUi2NPNpZ6eHo2Pj2t4eFjj4+O6fPly1ACbNbSOt99+W/fv348VHDc2NnImOylhnXvk+Wy4Fub9aXu8nRghIymmI8RVykkI6xckn5PSNZxmniAXFngwmuE3tJScl5pPLKrU9evBlggvzAfATRaiCym8QsRDkbwZkhjemfX19Yi3UICLDeQeIfge58+f18WLF7W0tKTp6WmVy2UtLS3p/v37sfYy8WDValWVSiVqNZ53pK+vL5oEcHIcjHdN5kkcJiLV5+fnFUKIAqW/v1/9/f05EiNub68UyRw4/Z+8uiMjI5qYmIglYKECvJsmsbu7q7W1tcg/mpmZiaEcpKckHcP29ra2trZyKRU4BJuxlmlPApS/X9uJGQ2nzbMB3Z53rcRxFP87FSJp8++kn3cQNL2vlMdr3PTwmkKepY3IX052t9P9++AZLG4CRTs6OrS1taWVlZW4IR8+fBhzeSBouru7I3uVHDMdHR3R9QybuFKpaHBwMAochFa1Wo0baXNzM6YqQNC5mdjd3Z3DWdwV7jymdAPChwkhRFMMwelalGNFHvfk+A+R6uPj43r55Zc1MTGhoaGhdxUuYEwrKyuanZ3VyspK5Lx4GAvm0/r6ehQ8zbIRpsm8UlPsVMjk24kZDbSUVDPB0yA9XklPerw2dDMXouMQfMfNLa7HaezuUprzObwcqXNmwF/YGJhN/izebwQQJ7aDnBDgKpWKlpeXVSwWVa1WNTIyEjO1YSaMjY3lNAI0ouHhYU1OTkbgcmVlRYuLi1paWtLs7Kzu3bun+fl5LSwsxNAFgiX5DhuQ0AkvoesajY+XHwYO1rI5MYHwgjHueKQQKA6U9/b2anh4WFeuXNH4+HisBvBuh8rhYS3R2dLSkiqViiqVSvybmDDHniqVShwDz/Hr5rSvCefoMOdOVjxttXZihIzT0KFns4hdsLiXp5lbGhdi+jn/PlpJCva6tuJkO1pKYmPDOBfHvRsuwNL++0ZsJjylBk5FZO/m5qaWlpaiECFIcn19XQMDAxocHNT4+Lj6+/ujVgCwDBnt6tWrsT+YY7dv39bMzIwWFhZiIXhy5FCc3UlmCBQ/wQFjeXYXtuSVRcvy4ESEM+8Vi0X19/draGhIw8PDGh0djaVnjguo7u7uam5uTsvLy1pdXY0pOPCYeYE4vHDEHuHldJa3C8IUj+NgCSHkSvectkY7MULGI3Q9i7+bFlLjJEm5CC4sUlap9Djij4bh4F2K63A/B5E52RAwHt/kqRkgrTkbVWp4Udxkc+aq83vcRYxX4+CglpN1bW0tbsqVlZXIs1lbW9Pw8LB6e3tVLBYfS7JEQ3Nob2/Xq6++qldffTWGPiwtLUUm8r1797S+vq719fV4+rMZ0xgdd3UjdPCiuauZ+wJ6d3Z2qlQqqVgsxhIjnvjrSQ0tCnd4tVrVyspK7D+VIWDngucBZu/s7MTIcge6U2EPodDXg5M4nWN0qsU83k6MkOns7IwkJ1c52eCujUh6TICkOA2vu0fKm/NfuJ4zf1OB4yxZ/k/NKUmP2fBO56chaJxqz3UROE5G4ztgOMQJsUmq1WoMArx37576+/tzRcYGBwfV0tISI5CfpBEAvF68eFEXL16M48KGXFpaijgGUeH0wetPI0T4QVMBa8LLRGIvAithGKdR6j7nh4eHsQ5TpVKJoSjk0OtioAAAEZlJREFU8wFL4TNoY44LoR2idTjBjnF2HpFrqvz2ipKpVn3a8u24tbDvS9qUdCjpIMuyT4YQBiT9lqSLku5L+o+yLKuG2mj/T5J+VNJDSf9JlmWvvdc9Jicnde/evQj4MumutqbxS6n2wgnjzd3UUvOIbU4iX0jOhUnTM3peXRaY4xFe9sMBa+7rnjPvk3NqJMXYGMh91PuRGgKKDdPe3h7JY+vr61pdXVVvb6+mp6ejVkPgJazYLMtyJLgnzH3c+L29vbpy5Up8Tmj2CB3GzYFworvZwF1dXRGnkRpmMgGyCDTcwWBDqZdrbW1N1Wo1kuWgBXhKCo+Bc/o/JpHXGwdXQdPq7++P6VL39va0srISU8I+fPgwB+a718nn87TV2tNoMv9elmVez/oXJP1hlmX/MITwC/X/f17SX5N0tf7zKUm/Uv/9ro3ET2xqJ8+R28MBWxYzhCk2tWs00uOhBylekrqteQ8qO27olPXazLSSlBMsnlyL9yTlOCauxfC8LmzQaDyIEY8MWoGUz17PRnKyHuBpd3e3BgcHNTExEV3I8FbQOt4N+3DMpbe3Vz09PTFMAqGIQPFMfG5yogkQxYwAoSYUTGoPOsR1jkAi8bqbqAgWxsuFk/ODiDVzjybCBRPuwoULmpiY0NmzZ1WtVnXv3r0cH8YjxFNs8Elr4/u1fTfm0k9I+mz9719XrR7Tz9df/42strv/vxBCXwhhLMuyhXe72P3792MEsxPgwGjYvFJjs8L+dIDOwVwa//v3XMikwDAb30MZXHt5UkIoTwHBtdhQmBOuWrt71hmuvNdss7PAEUZoJIyF15OG2r+9va379+9HraSrqysGDZZKpRiAyUkONsKzeAkUp/W7Fkc/2MR7e3uR6+N4Bf0n37A/qwtVB2nBwfb29qJgQQOSFE1AN2dIdM5vD+j0pFn8798nZzMJ3efn57W8vJzjb9F8jk41mObtuEImk/R/hRAySf80q9WxHkFwZFm2EEIYrn92XNKMfXe2/tq7ChknQ6URrX4CInxcYLimgVnlIQd8Lt3kXNe9UKnnhA3tmIqryulv13S4JliEs1dTd28qUNydjrDg+ZzkV61W4+t+77a2tigcifFhjPFWzc/Px6xweHvgovT29mpoaChiJZ7zGOymq6sruukhFuIV5PN4ptBMPKDQgVMy5JFdzrUZ1oSnmOCZfewRPPCN8Bg5QN/W1pYznzBNd3Z2ItBOqAFUgpTJm+J2vrZOtZjH23GFzA9mWTZfFyRfDSF8+10+20zXfkzEhxC+IOkL/J8mVnI+Rf3zcSHy+fQ1n2D+JoWBe20cvHVejv/gXsWscTwIwYMp56cgfWPh+WL077nGQ5Bl6iIG05AUhYaPD42UCsViUZ2dnXFTe1G6lIOEZvDo0SOtrq7GhE5e14hgSng5nZ2d0Qxqb29Xb2/vY1ogZq6npMR88Qh69+xJivgSmoTT+BF+TlWAX+RuY8//7ILEgWcvpeLCONWi1tfXc2vGNR+vspgKvBSDO23HFDJZls3Xfy+FEH5PtfK0ZcygEMKYpKX6x2clTdrXJyTNN7nmFyV9UZJCCJm7cV2jSHELn0wHUX1iU36Ka0GpJ6AZgIxg4YSlpYsHrQLNAldnMzdms9ACjwiH9Eaf2BSPHj2K6r17YDwamGv5ZgKvSjU6N+U4uR0E5TqMRWtrIy/w1tZWDih2M5X4LEkRT3JBS3Nh7SB5ipn4WHugKmPNPegbtAdfK5ijbgK7mcd4cy/nZaFFHRwcxDn2UBbGAIHvZMr3cr1/v7X3FDIhhG5JhSzLNut//1VJ/52kL0v6aUn/sP779+tf+bKk/yKE8CXVAN/198JjpAbnId0QqauYxZkCvFxDygdPch03jdJFkN43XYSO2/im9UXpXhRfdN58A6T9pr88g3+X94nd8VACP13dvcs4IIxc6KLlsAl9PLzqJFgL/cbMcXzMx0LK4xsOrCN8HfdAU0LI8J5rpqmGR2VRnyuEIS5ztCkXdC54WE+uWfE8jsOl40//ef5mwjv97mk7niYzIun36gPaKulfZFn2f4YQ/kzSb4cQfkbStKTP1T//f6jmvp5SzYX9nx6nI80wGEnx9EC9ZsP7CcnCdDzFM701WzDu3aFBnnuS9pPiNv5dzALHhtgMeITop7s+003q30XTQEPis5gJfX196ujoiBR8BAZmiuMNad1oNvbGxkYUvF55MoQQPTO+wTAdAGrdM8Zn/FR3TdGBeQftXSBjgjoG5CB4M3exA9GpieNguK8vTDiEHR7FFMjHFHVqgdRIY0o4R6FQyHFuTlujvaeQybLsrqSPNnl9VdJfbvJ6JulvP21HUENZkK4ip2BaelKyoFLNxgWBex64Hy3NFeOgZOp14rpuvj3JE+T9cMao98dzpvgG9HgtX/xwYiRpY2MjChk4KB0dHbnUpYDGgKEeQZ6OKTgXOWR4PfW+oPnQL/eiSXkNgTlw04ix9Zrk3h9PBEUfXMg0e80FD9d2djX9QaB5XxxDYx1yD4S3J7NHwPohwbOwFk9bo52Y0XDtxU8cN1VSUC31Mj3pur5R0hghFhZCivdT04jm6jYL1Osru9BzANf7B6idAsPeH38mNCw2xaNHj9TR0aHDw0Ntb2/nAOFUu6EfgKLwUthAfJZncY3DWwo2uyD019L/fYOnIRX8dmzKm2NKKcmSnDJumqaR4IR2oNUyVy60XbNrhgtKjeJ47mVKPVuOpZ22fDsxI+KsW6mBozxJuEiKKn2Kp/hn0sXkzYVBigm5cPPrpx4Of93NAWcsN3NruhnoBD43xVyzgfHLZ8Bk/NnIF+xaB9pJOkZcn83oggWBJSmnUdEf5os0Fzwvn3Mh4u5jSRFo5scxIYSBrwHmJS1X4h46+kQpF39GqSGs0GBSIZ5qi5JyApr3OIj80KIvp2Dvk9uJETJSA2RlAzrYKumxk1RSU3xAavBMUgEhNRZzuqFT8JLGokrNLzYQG9D755s4/TsVOqkAZEM4poNJhADo7e2NuWW8PhLCDcxlZ2cnAq2YEW6+AFq7NuEaCuaRZ4LLsiznyXJT0wUFGxMNz93T3AsgF/c5mgGeIk9nyvWc9Mj9W1tbc67l1Mz2Q6dZHl6/T4q5OU3B1yJaJXOFidbW1qYf+qEfyqXsdOHnJED6zzNxTUzclMSYPk/qUeO1dAwcr6QPUCTcgUD/nXXOYQVZs6+vT7/5m7+p47YTI2TwCviCarYpUyyG12h+utDctm8G5qaNiWv2AyjKdShg5oBuaue7l8fd4r5gUwYxm50TG2zFeTtsUE568JS08LzzTXgGj3pPvS2pZoPAd0xC0mPmAyCrC2zPP8P4eOySf8ZDAXwz+uZoph2iCbobnY3ihDqepxnYzrW5lnsn0zXpjf+dmoBgZtwhIbrHzmPffCPTyLuMgEnNwFS7l9RUuPra93566A7fBVdCM/X5dfO6WQnfd2snRsjw4EzkkzCLtKWu1Gbvu5bjmgiT7q5u7w99agb40k8WNhiI2+p4gZoJQRa0k858wtNnd7cyQgbMgYXT3d2tYrEYKfvOruWeXjEA5rJjMS5g0zH1/1OuDGaFxy9Jin30+6XzjiDmPQRViulIefzGiZruXWRsWU8IGIRbiiWla8Hjkrx5P/i7q6srJ3wYz1Trdk0Y4XFwcKDt7e34ea9V5eklfL34+vPrM+bNBE16uPoeSK/jid29GmmWZTHG62lxpxMjZBwcdWzGT3vXYBiwFMjl+zSfANylfDcd5GbqdTppCAMXhr4wXJjhVvUT3J8pXZw0NoNvovS5uAcbuFgsqru7OwrVM2fOxITgCA6SR7HxPNLZTcVmkdkuDOgL9+bkYwN7YKhvLJ4Tk8DnO/Vi+X2chUx/qdfkqr2nseC0T4vQ8duFaHpw+Cb0MU+DY11ocfKn4SUeeEq/Nzc3c/dys8W1uvSaCE6EVKphOY7oY8X8OcjN86frDx4W2i7riRLMXn75uO3ECBlvvsF9kyN8mqmt/l1fHC4ckOZSXgVm4tITnElzILKZ+rm1tSWpsbDSTZpqYi6g+B4aiW9MJhNBysnip0sIQZ2dnZFef3RUS8FA7l4X1qk56gC2q+/03T0uuJWlxsZ2t7WD6FLDvc2Jnbr5XdDSUp5RShKkP7zvAlzK5xJ2U9DXDeOW/vY15Tibb1Lvr5s5rCv4Pf4/CbsQDJi8qTAiVAN3OZiam74u/Lw/PCtzwPM7FYTv8V1//dGjRzntJOVzsR55jqfN/BdSLeCDaCGETUlvfwC3HpS08p6f+t669+l9v7fv+0He+0KWZUPpiydFk3k7y7JPPu+bhhD+/IO47wd579P7fm/f94O+d7N2Gpd+2k7baXum7VTInLbTdtqeaTspQuaL32f3/SDvfXrf7+37ftD3fqydCOD3tJ220/a9206KJnPaTttp+x5tH7iQCSH8ByGEt0MIU6FW9eD9vPY/CyEshRDestcGQghfDSHcrv/ur78eQgj/c70fb4QQXvku7jsZQvjjEMKtEMK3Qgh/53ncO4TQEUL4dyGE1+v3/W/rr18KIXytft/fCiGcqb/eXv9/qv7+xe/0mevXawkhfCOE8AfP+b73QwhvhhC+GUL48/prz2Oe+0IIvxtC+HZ9rj/9HOb4ev05+dkIIfzd5/G833HzOI3n/SOpRdIdSZclnZH0uqSX3sfr/yVJr0h6y177HyT9Qv3vX5D0S/W/f1TSv1EtR/EPSPrad3HfMUmv1P8uSnpH0kvP+t7175+t/90m6Wv16/22pM/XX/9VST9b//s/l/Sr9b8/L+m3vsvx/jlJ/0LSH9T/f173vS9pMHnteczzr0v6z+p/n5HU9zzum+yfRUkXnud9n7qfz/uGySB9WtJX7P9flPSL7/M9LiZC5m1JY/W/x1Tj6EjSP5X015t97n3ow+9L+vef570ldUl6TbUUqCuSWtMxl/QVSZ+u/91a/1z4Du83IekPJf2IpD+oL+pnft/6NZoJmWc61pJ6JN1L+/2c5/ivSvp/P6h1fdyfD9pcelL5lGfZcqVcJL1XKZfvqtVNgY+rplU883vXTZZvqpbY/auqaYprWZbBv/drx/vW31+XVPpO7ivplyX9PUnECZSe032lRsmer4daFQzp2Y/1ZUnLkv7Xuon4v4RaDuznub4+L+lf1v9+ruv6adoHLWSOVT7lObX3vS8hhLOS/ndJfzfLso3nce8syw6zLPuYaprFq5JuvMu135f7hhB+TNJSlmVf95ef9X2t/WCWZa+oVr30b4cQ/tK7fPb9uneraqb4r2RZ9nFJ26qZKc/6vrWL1fCtH5f0O+/10ffzvt9J+6CFzLHKp7zPrRxqJVwUvoNSLsdtIYQ21QTM/5Zl2b96nveWpCzL1lSr6vkDkvpCCISQ+LXjfevv90qqfAe3+0FJPx5qNdO/pJrJ9MvP4b6S8iV7JOVK9tTv8SzGelbSbJZlX6v//7uqCZ3nNcd/TdJrWZaV6/8/t7X1tO2DFjJ/Julq3QtxRjX178vP+J6UcpEeL+XyH9fR+B/QMUu5NGshhCDp1yTdyrLsf3xe9w4hDIUQ+up/d0r6K5JuSfpjST/5hPvSn5+U9EdZ3XB/mpZl2S9mWTaRZdlF1ebwj7Is+5vP+r5SrWRPCKHI36rhFG/pGY91lmWLkmZCCNfrL/1lSTef9X2t/XU1TCWu/zzu+/TteQJATwCvflQ178sdSf/V+3ztf6laedx91ST6z6hm+/+hpNv13wP1zwZJ/6TejzclffK7uO8PqaaSviHpm/WfH33W95b0EUnfqN/3LUn/df31y5L+nWplan5HUnv99Y76/1P19y+/D2P+WTW8S8/8vvV7vF7/+RZr6DnN88ck/Xl9vP+1pP7ndN8uSauSeu21Z37f7/TnlPF72k7baXum7YM2l07baTtt3+PtVMicttN22p5pOxUyp+20nbZn2k6FzGk7baftmbZTIXPaTttpe6btVMicttN22p5pOxUyp+20nbZn2k6FzGk7baftmbb/H1+cSNptDahKAAAAAElFTkSuQmCC\n",
+      "text/plain": [
+       "<Figure size 1440x216 with 1 Axes>"
+      ]
+     },
+     "metadata": {
+      "needs_background": "light"
+     },
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "path = \"200.png\"\n",
+    "arr = misc.imread(path, flatten=True)\n",
+    "print (\"initial image\")\n",
+    "imshow(arr, cmap='gray');"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 57,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "<Figure size 432x288 with 0 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "tmp = plt.gcf().clear()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "initial_probability = {\"2.png\": 0.75, \"MRI1.jpg\" : 0.25}\n",
+    "number_of_pixels = arr.size\n",
+    "class_info = []\n",
+    "paths= [\"2.png\", \"MRI1.jpg\" ]\n",
+    "for path in paths:\n",
+    "    tmp_arr = misc.imread(path, flatten=True)\n",
+    "    class_mean = np.mean(tmp_arr)\n",
+    "    class_var = np.var(tmp_arr)\n",
+    "    class_freq = len(tmp_arr)\n",
+    "    class_probabilty = class_freq/number_of_pixels\n",
+    "    class_info.append([initial_probability[path], class_mean, class_var])\n",
+    "\n",
+    "print (\"class_info\")\n",
+    "print (class_info)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 59,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "def pdf_of_normal(x, mean, var):\n",
+    "    return (1/np.sqrt(2 *  np.pi * var))*np.exp(-((x-mean)**2)/(2*var))"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 60,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "(540, 800)"
+      ]
+     },
+     "execution_count": 60,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "arr.shape"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 61,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "def naive_bayes_predict (arr, class_info, fixed_pixels_index=[], correct_arr = []):\n",
+    "    predict_array = np.zeros((len(arr), len(arr[0])), dtype=float)\n",
+    "    class_color = [50,200]\n",
+    "    for i in range(0, len(arr)):\n",
+    "        for j in range(0, len(arr[0])): \n",
+    "            if (len(fixed_pixels_index)>0 and len(correct_arr)>0 and fixed_pixels_index[i][j]==1):\n",
+    "                predict_array[i][j]=correct_arr[i][j]\n",
+    "                continue\n",
+    "            max_probabilty = 0\n",
+    "            best_class = -1\n",
+    "            val = arr[i][j]\n",
+    "            for cls_index in range(len(class_info)):\n",
+    "                cls_p =  class_info[cls_index][0]\n",
+    "                mean =  class_info[cls_index][1]\n",
+    "                var = class_info[cls_index][2]\n",
+    "                pos =pdf_of_normal(val, mean, var)\n",
+    "                cls_posterior = cls_p * pos\n",
+    "\n",
+    "                if (cls_posterior > max_probabilty):\n",
+    "                    max_probabilty = cls_posterior\n",
+    "                    best_class = cls_index\n",
+    "                    \n",
+    "            predict_array[i][j] = class_color[best_class]\n",
+    "            \n",
+    "    return predict_array"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 62,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "def distance (x,y):\n",
+    "    a = x-y\n",
+    "    a = a*a\n",
+    "    return np.sqrt(np.sum(a))"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 63,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "def differnce(a,b):\n",
+    "    if (a==b):\n",
+    "        return -1\n",
+    "    else:\n",
+    "        return 1"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 64,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "def initial_energy_function(initial_w, pixels, betha, cls_info, neighbors_indices):\n",
+    "    w = initial_w\n",
+    "    energy = 0.0\n",
+    "    rows = len(w)\n",
+    "    cols = len(w[0])\n",
+    "    for i in range(0, len(w)):\n",
+    "        for j in range(0, len(w[0])):\n",
+    "            mean = cls_info[int (w[i][j])][1]\n",
+    "            var =  cls_info[int (w[i][j])][2]\n",
+    "            energy += np.log(np.sqrt(2*np.pi*var)) \n",
+    "            energy += ((pixels[i][j]-mean)**2)/(2*var)\n",
+    "            for a,b in neighbors_indices:\n",
+    "                a +=i\n",
+    "                b +=j\n",
+    "                if 0<=a<rows and 0<=b<cols:\n",
+    "                    energy += betha * differnce(w[i][j], w[a][b])\n",
+    "    return energy"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 65,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "def exponential_schedule(step_number, current_t, initial_temp,  constant=0.99):\n",
+    "    return current_t*constant\n",
+    "def logarithmical_multiplicative_cooling_schedule(step_number, current_t, initial_temp, constant=1.0):\n",
+    "    return initial_temp / (1 + constant * np.log(1+step_number))\n",
+    "def linear_multiplicative_cooling_schedule(step_number, current_t, initial_temp, constant=1.0):\n",
+    "    return initial_temp / (1 + constant * step_number)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 66,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "\n",
+    "def delta_enegry(w, index, betha, new_value, neighbors_indices, pixels, cls_info):\n",
+    "    initial_energy = 0 \n",
+    "    (i,j) = index\n",
+    "    rows = len(w)\n",
+    "    cols = len(w[0])\n",
+    "    mean = cls_info[int(w[i][j])][1]\n",
+    "    var =  cls_info[int(w[i][j])][2]\n",
+    "    initial_energy += np.log(np.sqrt(2*np.pi*var)) \n",
+    "    initial_energy += ((pixels[i][j]-mean)**2)/(2*var)\n",
+    "    for a,b in neighbors_indices:\n",
+    "        a +=i\n",
+    "        b +=j\n",
+    "        if 0<=a<rows and 0<=b<cols:\n",
+    "            initial_energy += betha * differnce(w[i][j], w[a][b])\n",
+    "    \n",
+    "    new_energy = 0\n",
+    "    mean = cls_info[new_value][1]\n",
+    "    var =  cls_info[new_value][2]\n",
+    "    new_energy += np.log(np.sqrt(2*np.pi*var)) \n",
+    "    new_energy += ((pixels[i][j]-mean)**2)/(2*var)\n",
+    "    # print(\"/////// \\n first enegry\", new_energy)\n",
+    "\n",
+    "    for a,b in neighbors_indices:\n",
+    "        a +=i\n",
+    "        b +=j\n",
+    "        if 0<=a<rows and 0<=b<cols:\n",
+    "            new_energy += betha * differnce(new_value, w[a][b])\n",
+    "\n",
+    "    # print (\"END energy\", new_energy)\n",
+    "\n",
+    "    return new_energy - initial_energy"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 26,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "def simulated_annealing(init_w, class_labels, temprature_function,\n",
+    "                        pixels, betha, cls_info, neighbors_indices, max_iteration=10000,\n",
+    "                        initial_temp = 1000, known_index=[], correct_arr = [], temprature_function_constant=None ):\n",
+    "    partial_prediction=False\n",
+    "    if (len(known_index)>0 and len(correct_arr)>0):\n",
+    "        partial_prediction=True\n",
+    "\n",
+    "    w = np.array(init_w)\n",
+    "    changed_array = np.zeros((len(w), len(w[0])))\n",
+    "    iteration =0\n",
+    "    x = len(w)\n",
+    "    y = len(w[0])\n",
+    "    current_energy = initial_energy_function(w, pixels, betha, cls_info, neighbors_indices)\n",
+    "    current_tmp = initial_temp\n",
+    "    while (iteration<max_iteration):\n",
+    "        if (partial_prediction):\n",
+    "            is_found=False\n",
+    "            while (is_found==False):\n",
+    "                i = random.randint(0, x-1)\n",
+    "                j = random.randint(0, y-1)\n",
+    "                if (known_index[i][j]==0):\n",
+    "                    is_found=True\n",
+    "        else:\n",
+    "            i = random.randint(0, x-1)\n",
+    "            j = random.randint(0, y-1)\n",
+    "\n",
+    "        l = list(class_labels)\n",
+    "        l.remove(w[i][j])\n",
+    "        r = random.randint(0, len(l)-1)\n",
+    "        new_value = l[r]\n",
+    "        delta = delta_enegry(w, (i,j), betha, new_value, neighbors_indices, pixels, cls_info)\n",
+    "\n",
+    "        r = random.uniform(0, 1)\n",
+    "\n",
+    "        if (delta<=0):\n",
+    "            w[i][j]=new_value\n",
+    "            current_energy+=delta\n",
+    "            changed_array[i][j]+=1\n",
+    "            # print (\"CHANGED better\")\n",
+    "        else:\n",
+    "            try:\n",
+    "                if (-delta / current_tmp < -600):\n",
+    "                    k=0\n",
+    "                else:\n",
+    "                    k = np.exp(-delta / current_tmp)\n",
+    "            except:\n",
+    "                k=0\n",
+    "\n",
+    "            if r < k:\n",
+    "                # print(\"CHANGED worse\")\n",
+    "                w[i][j] = new_value\n",
+    "                current_energy += delta\n",
+    "                changed_array[i][j] += 1\n",
+    "        if (temprature_function_constant!=None):\n",
+    "            current_tmp = temprature_function(iteration, current_tmp, initial_temp, constant =temprature_function_constant)\n",
+    "        else:\n",
+    "            current_tmp = temprature_function(iteration, current_tmp, initial_temp)\n",
+    "        iteration+=1\n",
+    "    return w, changed_array"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 67,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "def convert_to_class_labels(arr, inverse_array={50:0, 200:1}):\n",
+    "    for i in range(0, len(arr)):\n",
+    "        for j in range(0, len(arr[0])):\n",
+    "            arr[i][j] = inverse_array[int(arr[i][j])]"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 68,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "def get_accuracy(arr, labels):\n",
+    "    correct = 0\n",
+    "    for i in range(0, len(arr)):\n",
+    "        for j in range(0, len(arr[0])):\n",
+    "            if (labels[i][j]==int(arr[i][j]/127)):\n",
+    "                correct+=1\n",
+    "    return correct/(len(arr[0])*len(arr))"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 30,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "<Figure size 1280x1440 with 0 Axes>"
+      ]
+     },
+     "execution_count": 30,
+     "metadata": {},
+     "output_type": "execute_result"
+    },
+    {
+     "data": {
+      "text/plain": [
+       "<Figure size 1280x1440 with 0 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "plt.figure(figsize=(16, 18), dpi=80, facecolor='w', edgecolor='k')"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "u = 10\n",
+    "# plt.close('all')\n",
+    "def a_complete_set_for_part_2 (arr, class_info, max_iter=1000000,var = 10000,\n",
+    "                               betha = 100,\n",
+    "                               neighbor_indices = [[0,1],[0,-1],[1,0],[-1,0]],\n",
+    "                               class_labels = [0,1], \n",
+    "                               class_color = [50,200], \n",
+    "                               schedule= exponential_schedule,\n",
+    "                               temprature_function_constant=None):\n",
+    "\n",
+    "    fig, (ax1, ax2, ax3, ax4) = plt.subplots(1,4)\n",
+    "#     fig.suptitle('Comparision', fontsize=20)\n",
+    "    \n",
+    "\n",
+    "    ax1.set_title(\"initial image\")\n",
+    "\n",
+    "    ax1.imshow(arr, cmap='gray')\n",
+    "\n",
+    "\n",
+    "    rows = len(arr)\n",
+    "    cols = len(arr[0])\n",
+    "\n",
+    "#     cls_info = naive_bayes_learning(arr, noisy_arr, labels)\n",
+    "    cls_info = class_info\n",
+    "    initial_arr = naive_bayes_predict(arr, cls_info)\n",
+    "    ax2.set_title('Naive Bayes image')\n",
+    "    ax2.imshow(initial_arr, cmap='gray')\n",
+    "\n",
+    "    convert_to_class_labels(initial_arr)\n",
+    "    \n",
+    "    w, test_array = simulated_annealing(initial_arr, class_labels, schedule,\n",
+    "                                        arr, betha, cls_info, neighbor_indices, max_iteration=max_iter)\n",
+    "\n",
+    "    \n",
+    "    for i in range (0, len(w)):\n",
+    "        for j in range(0, len(w[0])):\n",
+    "            w[i][j] = class_color[int (w[i][j])]\n",
+    "\n",
+    "    ax3.set_title('CRF image')\n",
+    "    ax3.imshow(w, cmap='gray')\n",
+    "    cv2.imwrite('C:/Users/Ankit/Desktop/% 2d.png'%(u),w)\n",
+    "    second_image = w\n",
+    "#     print(first_image)\n",
+    "    plt.rcParams[\"figure.figsize\"] = (20,3)\n",
+    "    ax4.set_title('differ image')\n",
+    "\n",
+    "    ax4.imshow(test_array, cmap='gray')\n",
+    "\n",
+    "    \n",
+    "    plt.show()\n",
+    "    \n",
+    "    \n",
+    "    \n",
+    "plt.figure(figsize=(16, 18), dpi=80, facecolor='w', edgecolor='k')\n",
+    "\n",
+    "for i in A:\n",
+    "    path = i\n",
+    "    arr = misc.imread(path, flatten=True)\n",
+    "    print (\"initial image\")\n",
+    "    imshow(arr, cmap='gray');\n",
+    "    \n",
+    "    initial_probability = {\"2.png\": 0.75, \"MRI1.jpg\" : 0.25}\n",
+    "    number_of_pixels = arr.size\n",
+    "    class_info = []\n",
+    "    paths= [\"2.png\", \"MRI1.jpg\" ]\n",
+    "    for path in paths:\n",
+    "        tmp_arr = misc.imread(path, flatten=True)\n",
+    "        class_mean = np.mean(tmp_arr)\n",
+    "        class_var = np.var(tmp_arr)\n",
+    "        class_freq = len(tmp_arr)\n",
+    "        class_probabilty = class_freq/number_of_pixels\n",
+    "        class_info.append([initial_probability[path], class_mean, class_var])\n",
+    "\n",
+    "    print (\"class_info\")\n",
+    "    print (class_info)\n",
+    "    a_complete_set_for_part_2(arr,class_info, max_iter=1e5, betha=1e6)\n",
+    "    u += 1"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 20,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 1440x216 with 4 Axes>"
+      ]
+     },
+     "metadata": {
+      "needs_background": "light"
+     },
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "a_complete_set_for_part_2(arr,class_info, max_iter=1e7, betha=1e6)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 51,
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  }
+ ],
+ "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": "3.7.3"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 2
+}