Logistic_SPAM.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": [],
      "collapsed_sections": [],
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/GerardoMunoz/ML/blob/main/Logistic_SPAM.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "# Regresión Logistica para clasificación de SPAM\n",
        "\n",
        "## Función Sigmoide\n",
        "\n",
        "$f(t)=\\frac{1}{1+e^{-t}}$"
      ],
      "metadata": {
        "id": "1O-hjUkln6-r"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "import matplotlib.pyplot as plt\n",
        "import numpy as np\n",
        "x=np.arange(-5,5,0.1)\n",
        "y=1/(1+np.e**(-x))\n",
        "plt.plot(x,y)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 283
        },
        "id": "t4hY-QAOqhRz",
        "outputId": "f3dd435f-bfaa-45a8-db4a-e4fe9b8a7a18"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[<matplotlib.lines.Line2D at 0x7f730fe46bd0>]"
            ]
          },
          "metadata": {},
          "execution_count": 1
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Función sigmoide con diferentes inclinaciones en $y=0.5$\n",
        "\n",
        "$f(t)=\\frac{1}{1+e^{-(\\beta_0+t)}}$"
      ],
      "metadata": {
        "id": "xK6WqSXQn63b"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "import matplotlib.pyplot as plt\n",
        "import numpy as np\n",
        "x=np.arange(-5,5,0.1)\n",
        "y0=1/(1+np.e**(-(-2+x)))\n",
        "y1=1/(1+np.e**(-(-1+x)))\n",
        "y2=1/(1+np.e**(-(0+x)))\n",
        "y3=1/(1+np.e**(-(1+x)))\n",
        "y4=1/(1+np.e**(-(2+x)))\n",
        "plt.plot(x,y0)\n",
        "plt.plot(x,y1)\n",
        "plt.plot(x,y2)\n",
        "plt.plot(x,y3)\n",
        "plt.plot(x,y4)\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 283
        },
        "id": "JVbaPI1ut-lx",
        "outputId": "bbff2bca-d10a-4492-8086-f504a7cacc69"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[<matplotlib.lines.Line2D at 0x7f730f939690>]"
            ]
          },
          "metadata": {},
          "execution_count": 2
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAD4CAYAAAD8Zh1EAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nOydZ3hU1dqG751k0nsjkISEEkKvAQREQZAmqCiKDayoKIIoKqhIVRSxoYKAIqIiBxQRAUFBQAi9h5AEkpDee5k+s74fg+fwefSImiwCrPu65prKPDvs5Jk173qLJoRAoVAoFJc/Tpf6ABQKhUJRNyhDVygUiisEZegKhUJxhaAMXaFQKK4QlKErFArFFYLLpRIODg4W0dHRl0peoVAoLkuOHDlSIoQI+b3nLpmhR0dHc/jw4Uslr1AoFJclmqZl/tFzKuSiUCgUVwjK0BUKheIKQRm6QqFQXCEoQ1coFIorBGXoCoVCcYXwp4auadpyTdOKNE079QfPa5qmLdQ0LVXTtJOapnWt+8NUKBQKxZ9xMSv0FcCQ//H8UCDm/OVRYPE/PyyFQqFQ/FX+NA9dCPGLpmnR/+MltwArhaMP735N0/w1TWsshMivo2NUKBSKv4QQArtdYLcKbFY7dps4f7E7HrcJhF3853H7+ft2gfjt/QuuHbcd7y/sAiH4zfWv72tDWGxgt2G32hE2G8Jmx26zgc1Os26NadwhvM5/7rooLAoHsi+4n3P+sf8ydE3THsWxiqdp06Z1IK1QKC4X7HaBxWjFbLRhMdowm6xYTDasJhuW8xer2Y7Vcv62xY7VbMdmPn/bYsdmsZ2/tmM7b9YXXuxWgc3muG6wCDvWojQadxhd528ttVJUCLEUWAoQFxfXgP/HFQrFH2Ex2zBUmzHWWDBUWzDWWjDWOK5NtRaMeismvRWzwYLJYMOstzhM3GS7aA1NAxdXZ1xcnXDWOeGic8ZF54Szk8AJGzphxU1Y0TDjhBlNmNDsRjSrEc1sBJMBzWQAkwEMejSLCU3Y0IQNJ2FDE3Y0+/nrX+//etvZCWdXFzSdC846HU7nb+OiYXMBqwvYnOyYneyYnWyYNSsmJwtGzYoRMwbNjAEzBkzoMWHWbNicBVYngc1JYHOB8B5P1cu5qQtDzwUiL7gfcf4xhUJxGWEx26gpM1JTYaK2wkRNuQl9pZnaShP6ShP6KjP6agvWPzJmDdw8XHDzdMHNU4ebpwtefm64errg6u6Cq4cLru7OuLq7oHN3Ruf2n4uzxQAVpVBehCgtwl5UhK24CGtxseNSWoqttBRhsfy+tosLzr6+OPv64uTni7OPL87BPjh5++DkE4SztzdOXl44ef167YmTlxd4uFPtZKZYVFNkr6TAVk6RuZRiQzHF+mJKjCWUGcooN5VjF/bflXZzdsPX1ddxcQvCx9UHH1cfvHXehLn64KXz+s/FxQtPnSct/FvU0Vn7zX9DHbzHBmCCpmmrgZ5ApYqfKxQND2EX1FSYqCzSU1lsoKrEQGWxkepSA9VlRgzV/22Wrh4uePm54unnSqNmfnj6uOLhq8PDxxUPbx3u3uevvXS4errg5KT9vrbZjDknF3NmGpb0bMzZOVhycjDl5FCTl4e9tvb//wNNwzkwEJeQEFyCg3GLicElOAjnoGBcAgNwDgzE2T8A5wB/nP39cfLyQtN+X1tv0ZNVnUVmVSY51Tnk1CSQU51DXnEeBbUFmO3m//d6F82FII8gQjxCCPcOp2NwR4I8ggh0DyTQPZAA9wAC3ALwc/PD380fdxf3P/gPF6AvhdI0KEuH/EQoz3Bc+kyC1sP+7JT9Zf7U0DVN+wroBwRrmpYDzAB0juMVHwGbgWFAKqAHHqzzo1QoFBeN3WansthAaW4tZXk1lBfqKc/XU1Gkx2b5zyrTyVnDJ8gd32APgiN98Al0xyfQDe8Ad7z83fAKcEPn6vzXtI1GTKlpmM6cwZSWijktHdO5dCw5uWD7z8reydMTXWQkushIPHv2RNekCbrGYbiEhaELC8MlOBhNp/tL2pWmSs6UnyG1IpW0ijQyKjM4V3mOIkPR/3tdoHsg4d7htAlqw4CmA2js3ZhGno0I8wqjkWcjAtwDcNL+QomOEFCZA0WnHZeSs1CcAqVnwVh5wQs18A2HgGhHTKke0C7VkOi4uDihui0qFP8Mm8VOSU4NxVlVFGfXUJxVTVleLTbreePWwDfInYDGXgQ08sQv1BP/UA/8Qj3x8nf7wxX1RWlXVWE8dQrj6dMYTydhTErCnJkJdoe2ptPh2qwZrs2b49osGteoqH9fnAMC/nBF/WcIISjUF5JYmkhSaRJJZUkklyVTpP+PcXvrvGnu15xmfs2I9oumqU9TonyjiPCJwEvn9bd/Zux2KDkDeceg4CTkn4SCBDBdYNzeYRAcA8GtIKglBLWAwBbgHwkubn9f+zyaph0RQsT93nOXrH2uQqH461SXGSlIqyQ/rZLCc5WU5NRgtzkWZW5eLoRE+tChXzhBEd4ENfEmIMwTl7+4yv49hN2OKTUVw9FjGI4dw3DyJOZz5/79vK5JE9zatsF32DDcWrXCrVUMrpGRaC7/3GJMNhOnS09zrOgYx4uOc6rkFMWGYgCcNWea+TWjZ1hPWgW0omVAS2L8Ywj1DP3bHxj/D0M5ZB+C7P2QfRDyjoO52vGciweEtYcOt0OjdhDaDkJbg0fAP9f9myhDVygaMDXlRnJSyslNKSc3pYLqMiMALm7ONIryofPASEKjfQlp6giZ1ImJ4VgFm86cRX9gP7UHDqI/fBh7pWMV6hwUhEfnzvjdcgvuHdrj0a4dzv7+daILYLFZOFF8gkMFhzhYcJATxSew2B3x/SjfKHo27kn74Pa0D25PbEDsH8ew/w6GCsiMh3O7IWM3FJ4vkNecIawDdBoN4d2gSVfHKtzpn39Y1iXK0BWKBoTNaifvbAWZiaVkJZZRnu/YLHTzciG8VQCdBkTSuKUfwRHeODnXbSsmW1UVtXv2ULN7D7V79mAtdqyCdZGR+AwcgGe3ODy7dUXXtGmdfXD8Sm5NLrtzdhOfG8/BgoPorXo0NFoHtubu1nfTtVFXOod0JsgjqE51EQLyj8PZbZC6DXIOgbCBiztE9oR+L0JUL4eJu/6DUI0klKErFJcYs8HKuZMlZJwsITOxFIvRhrOLE01i/GjTuzGRbQIIauKN9g/i3X+EJS+P6m3bqP55B/rDh8FqxdnPD68+vfHqcy1e1/REF173FY1CCJLKktietZ2d2Ts5U34GgAjvCEa0GEGvJr2IaxSHn5tfnWtjs0D6LkjZBClboDoP0KBJZ7h2MrToDxHd6yTe/VsqDRa2nS6kR7NAIgM96/z9laErFJcAs9HKuRMlpB4pIut0KXarwNPXlZi4RkR3DCaidcBfzjC5WCyFhVRt/oGqLT9gPHESALeYlgQ99BDe/fvh0bEjmnPdawshSClP4YdzP/Bjxo/k1OTgpDnRNbQrU+Km0C+yH1G+UXWuC4DNCud2wqlvIXkjGCtA5wUtB0DsMIi5EbyC60W6xmRl66kCNiXks/tsMRab4MVhrXn0urrPRVeGrlBIwm4X5CaXk3wgn/RjxVjNdrwD3OhwXQQtuoUS1sy3XlbhAHa9nuqffqLyu++o3bcfhMCtbRtCJk/Gd/AgXOtxYHuRvoiN6Rv5Pu17UitScdFc6Nm4J+M6jqN/ZH8C3OtxE7EgAY5/BQlrobYI3Hwhdii0GwnN+4OuDuPvF2CzC/aklrDuaA5bEwswWuyE+3vwQO9ohnVoTOfIuttzuBBl6ApFPVNTbiJpbx5J8flUlxlx83ShVc8wYnuG0bi5X72ZOIAhMZGKtWup2rgJe00NuogIgsePx+/mEfVq4ja7jfi8eNaeWcvunN3YhI2OIR15uefLDI4ejL97/RgaAKZqOPUNHFnhSC900kGrwdBxtOO6HkIpv5JXYWDN4WzWHMomr9KIn4eO27tGcFvXcLo2/fupmheLMnSFoh4QQpCfWsnJHdmkHy9B2AURrQPodVsLmnUKxkVXf9kRwmKhauuPlH2+EuOJk2hubvgOGYz/qFF4xMXVq6lUmipZd3YdXyV/RX5tPoHugdzf7n5GthxJtF90vekCjorMA0vg+JdgroHQtjDkDeh4J3gG1pusEIL96WWs2HuOn04XYhfQNyaYl4e3ZUCbUNxc5GXCKENXKOoQu81O6pEijv2URUl2DW6eLnQeEEm765rgF1L3m2AXYquupvyr1ZR/8QXWoiJco6Jo9OKL+N1yM85+9bC5eAHZVdl8dvozvkv9DqPNSPew7kyJm0L/yP7onP9axedfQgjI3Avx78HZrY7VePvboPs4iIirt4pMAIvNzsaTeSz95RxJ+VUEeOp47PoW3NOjab1seF4MytAVijrAZrGTtC+fYz9mUlViJCDMk373xtKqRxg6t/pdoVlLSylb8RnlX32FvaYGr969aTxnNl59+6I51e+UyTPlZ/g44WO2ZmzFWXNmePPh3NvmXmIDY+tVFyHgzFbY8zZkHwDPYLh+KsQ9BD6N6lXaaLGx5nA2S3alk1thoFUjb16/rQO3dgnHvR6/eV0MytAVin+AzWonaW8+R37IoKbcRGi0L31GxdCsY3C9xsYBrOXllH3yCWVfrkIYjfgMHkzQuEfwaNeuXnUB0irS+PD4h/yU+ROeLp6MbTuWMW3HEOoZWr/CQjjyxXe86oiP+0XC0Dehy33gWr+rYpPVxr8OZfPhjlQKq0x0berPrJvbcUPr0H/UQqEuUYauUPwNhF1w5lAhBzakU11qJKy5LzeMbUNE6/rf+LLX1lK6/FPKPv0Uu8GA7/DhBI8fj1vzZvWqC5BXk8f7x95nU/omPFw8eLTjo4xtO7Z+8sV/S+Y+2DbDsSL3bwo3fwCd7oL6DOngyFhZdzSHd7edJbfCQPfoAN4Z3ZlezYPq/Vz/VZShKxR/kdyUcuK/SaU4q5rgSG+uv7sTTdsF1vsft7DZqFi3juKFC7EVl+AzaBAhT03ALSamXnUBqs3VLEtYxpenv0TTNB5o/wAPtnuwflMOf6U0zWHkSd+DT2MY/g50vg9cXOtdOj61hLmbkkjKr6JThB/zbutA35jgBmfkv6IMXaG4SKpKDcSvTSX9eDHeAW4MfLAtrbo3qvfQCoD+8GEK5szFlJKCR5cuNHr/fTw6d653Xbuwsz51Pe8eeZcKUwUjWozgqS5PEeYVVu/amKph13zYvwic3aD/y9DryXoPrQBkl+mZ9f1ptiUVEhHgwft3d2F4x8YN1sh/RRm6QvEn2Cx2jm3L4sjmDNCg5y3N6Twgsk66GP4Z1pISit5cQOV33+HSpDHh776Dz+DBUozldOlpXt3/KidLTtI1tCtTe0ylTVCbetdFCEhcB1tfdpTld7kPbnil3jc7wbHhuWRXOot2puLspDF1aGse6B19yTc7LxZl6ArF/6AgvZKfVyZRXqCnRZcQ+twRg09g/VQXXogQgsp131L4xhvYDQaCHnuM4Mcexcmz/lenBquBD499yOdJn+Pv5s9r177G8ObD5axOK7Jh49OOjc+wjnDnSojsXv+6wMFzZUz95iTpJbUM79iYl25qQ2M/DynadYUydIXid7CYbOz/Lo2TO3LwDnBj+FOdiGpXx53+/gBzTi4Fr7xC7d69eMbFETZ7tpQNT4BDBYeYsXcG2dXZ3NHqDp7u9jS+rr71L2y3w+FPYNtMxwp9yBvQY5yU9rQ1Jitv/JDM5/sziQz04POHe9A3JqTedesDZegKxW8oSK/kp09PU1VsoMP14VwzsgWu7vX/pyKEoPKbbyh4bR4aEDbjFfxHj673XHIAo9XIu0ff5cukL4n0iWT54OV0D5OzMqYyB9aPh3O/OPqrjHgPAuqpSddvOJRRxuR/HSe3wsBDfZoxZXArPF0vX1u8fI9coahjbDY7hzdncGRzBt4B7ox8tgtNYuRMn7GWl5M/fTo127bj2bMnTea9hq5JEynayWXJTP1lKmmVadzT+h6e7vY0Hi6SQg2nvoGNkx3dEEcshK5j67W681fMVjvvbT/D4p1pRAR48vXjvegWVX/tAWShDF2hwDHabeuyUxSeqyL2mjD6jm6Fm4ecP4/aAwfJnfIs9opKQp9/nsAH7peyKhdCsCp5FQsOLyDALYCPBn5En/A+9a4LgLkWNj/n6LsS0R1GLnHM3pRAdpmeCauOciKnktFxkUwf0RZvtyvDCq+Mn0Kh+AdkJJSwbcVp7DbBoEfaERNX/9kU4JjTWbp0KcUL38c1Koqmy5bh3rq1FO1qczUz9s7gp8yf6BfZj9m9Z8vJKQcoToE190NxMlz/Alz3PDjLsaIfEwuYsvYEAvjovq4Mad9Yiq4slKErrlrsdsGBDekc3ZJJcKQ3g8e1xz9UTlMlW0UFuVOeo3bPHnyHD6fxrJk4eckZcZZSlsLTO54mvzafKXFTGNt2rLz86oSvYcNE0HnAmHXQ4gYpslabnTe2JLNs9zk6Rvjx4T1dL1kDrfpEGbriqsRYa+Gn5YlkJZbRtm8T+t4ZU68tbf+f9pkz5Dw5AWtBAWEzZ+I/+k5phro1YyvT46fj4+rDiiEr6Bxa/8VJANhtsH2Woyti014w6lPwlbM6Lq81M+Gro8SnljK2VxQv3dRGaktbmShDV1x1lOXXsnnxSapLjPS7N5Z2fet+ZuYfUfXTT+S9MBUnL0+iPl8ppdoTHBWfHxz7gGUJy+gc0pl3+r9DsEf9jFz7LwwV8M3DjtzyuIdh6Bv13n/lV1IKqhm38jAFlUbmj+rInXGRUnQvFcrQFVcV2cllbPkoAWedE7dM7kKTmHqcnHMBQgjKPvmEogVv4d6xIxHvL0TXSE6s3mA18OLuF9mWtY3bYm7jpZ4v4epc/31QACjPgC/vgLJ0Rw+WuIfk6AK7zhTzxBdH8HRzYfVj19C1qaQ9gkuIMnTFVUPyvnx2fJ6Mf5gnwyd0klLxCSCsVgrmzqVi9b/wHTaUxvPm4eRWf2PQLqTUUMrEnyeSUJLA892f574298mLl+cegVWjwWaBsd9B9LVydIHVB7N4af0pWjXy4dMHuhPmJ+dcX2qUoSuueIQQHNqUwaGN54hoHcCQxzpIS0m06/XkTJ5M7a5fCBr3CCGTJ0tJSQTIrMrk8Z8ep9hQzDv93mFA1AApugAkb4avHwLvEHhgM4S0kiIrhOCtH8/wwY5Urm8Vwof3dr1iUhIvhqvnJ1VclQi74Jd/neHUrlxa9wqj372tcXaRY6i2igqyH3scQ0ICYTNnEnDXaCm6AEmlSTy+7XEAlg9eTseQjtK0OfYlbJgAjTvDPf8C73oeenEem13w0rcJrD6Uzd09Ipl9S3t0znLOdUNBGbriisVms7N9RRJnDxXSZVBTeo1sIS3cYCkqIvvhRzBnZBCx8D18Bg6UogtwpPAIE7ZPwMfVh6U3Lq3/4cwXsm8RbJ3mKOEf/QW4eUuRNVltPPOvE2xKyOepG1ryzI2tGnyr2/pAGbriisRqsbFl6SkyE0rpNbIFXQfL6Q0CYM7JIevBh7CVlhK5dAlevXpJ096ds5vJOyfTxLsJS29cKqdvOTgaau2cB7vegDY3w+0fg4ucfQKD2cZjXxzhlzPFvHxTGx7p21yKbkPkor6PaJo2RNO0FE3TUjVNm/o7zzfVNG2HpmnHNE07qWnasLo/VIXi4rCabWxenEDmqVKuvydWrplnZ5M5Ziy2qiqarvhUqpnvyt7FpB2TaO7XnBVDVsg18+2zHWbe5T64Y4U0M9ebrTy04hC7zxYz//aOV7WZw0UYuqZpzsCHwFCgLXC3pmltf/Oyl4E1QoguwF3Aoro+UIXiYrCYbWxadJLspDJuGNOa9tfJyzE3Z2WROWYsQq8nasWneHSUF7fekbWDp3c+TauAViwbtIxAd0mNpoRwtLzd8zZ0exBGvC+l5S04zPzBTw9x4Fwpb9/ZiTu7X9k55hfDxazQewCpQoh0IYQZWA3c8pvXCODXpsl+QF7dHaJCcXFYzTY2LzpJTko5A8a2oU1vOd0KAcyZmQ4zNxpp+tkK3NtImOxznp+zfuaZXc/QJrANSwctlTOwGRxm/tMrEP+uo2DoprdBUgaP3mzlgU8PcSijjHdGd2Zklwgpug2di4mhhwPZF9zPAXr+5jUzgR81TXsK8ALk7QApFIDNaueHJafISSln4P1tiL1GXtMlS14emQ8+iDCZHGYeGytNOz43nmd3PUvbwLZ8dONH+Lj6SNNm5zzYuxC6j4Nhb0ppewuOMXHjVh7mcEYZ797VhZs7yfvgbujU1cfp3cAKIUQEMAz4XNO0/3pvTdMe1TTtsKZph4uLi+tIWnG1Y7fZ+emTRLISS+l/b2upZm4tLibrwYewV9cQ+cnHUs38SOERnt7xNC39W7Jo4CK5Zh6/8D8x86HzpZm5xWZnwipHX5b5ozopM/8NF2PoucCFwamI849dyMPAGgAhxD7AHfivRhFCiKVCiDghRFxIyOU54knRsBB2wY7Pk0k7Vsy1d8TQ9lp5f+C2igqyHn4ES3ExkUuW4NGunTTtxJJEntz+JI29G/PRwI/khVkADi+Hn6ZDu9scQykkhVlsdsEza06wLamIObe0Y1Q3FWb5LRdzJg4BMZqmNdM0zRXHpueG37wmCxgAoGlaGxyGrpbginonfl0qyfsL6DGiGZ0GyNsUsxsMZD8+HnNGBpEffoBn1y7StDMqMxi/bTz+bv4svXEpQR5yZp0CkPgtbHwGWg2B25ZK2wAVQjDr+0S+P5HHC0NaM6ZXtBTdy40/jaELIayapk0AtgLOwHIhRKKmabOBw0KIDcCzwDJN0ybj2CB9QAgh6vPAFYpjP2VxYls2HftHEDcsWpqusFrJfeZZDCdOEP7eu1JTE0sMJTy+7XE0TZObZw5wbjesexQiezpSEyV1TARYtDONlfsyefS65ozvJ2ey0eXIRRUWCSE2A5t/89grF9w+DUiaXaVQQMqBAvZ+k0rLbqFce0eMtKpAIQQFs2ZTs2MHjV6Zju+gQVJ0AWrMNYzfNp4yYxnLBy+nqW9TadoUnILV90BAM7j7K8eACkmsPZzNm1tTuLVzE6YOkTPR6XJFVYoqLjuyk8v4+bMkwlv5M/CBtmhO8kq8SxYtomLtWoIef4zAe+6RpmuxW5i8czKp5am8P+B92ge3l6ZNZQ58OQpcveG+b8BT3jDlXWeKmbougb4xwcwf1Qknief6ckQZuuKyoiy/li1LTuEf5snQ8R1x1slrvlT5/UZK3v8Av1tuIWTSJGm6Qghe3f8q+/P3M6fPHK4Nl9eGFlM1rLoLTDXw8Fbwl7dPkVJQzZNfHqVVIx8W39cNV0lN1S5nlKErLhsMNWY2fXgCZxeNm57sKK0FLoD+6FHyX3wRz+7daTxnttTGT58lfsY3Z79hXIdx3NryVmm62G3w9cNQdBruXQON5GXxFFebeGjFITxdnVn+QNxV1QL3n6A+8hSXBTaLnR8WJ1BbYWbY+I74BsmL4Zqzs8l5cgK6Jk0IX/gemqukaT/A9qztvH3kbQZFDWJClwnSdAH48WU4u9UxMq6lvFrBXwuHymrNfHJ/dxr7yTvXlzvK0BUNHiEEO1clk59WyYAH2hDWXF7Ota2mlpwnnkDY7UR8tBiXAHljzFLKUpi2exrtg9vz6rWv4vTftXr1x5HPYP8i6Pk49BgnTVYIwdRvTnIip4J3RnemQ4TE/PorAPU9RtHgObkjh+R9BcTdFE1MnJw5nADCbidv6guY0s/R9ONluDVrJk27wljBpB2T8NH58F7/93B3kThCLesAbHoWWtwAg1+Tpwss253O+uN5PDc4liHtJaZkXiEoQ1c0aHKSy4j/OpVmnYLpcZM8QwUoWbSYmm3baTRtqtRcc6vdypRdUyjWF7NiyApCPCVWVVflwZox4BcBo5ZLKxwCR0bL6z8kc1OHxjyhcs3/FsrQFQ2WqhIDW5cl4h/qIT09sXr7dko+cGS0BIwdK00X4K3Db3Gg4ABz+syhQ0gHecIWI/zrPkdGy9jvwENeeCmjpJanVjkyWt68o+NVOW2oLlCGrmiQ/DpxyG4XDBvfEVeJGS2mc+fIe/4F3Nu3J2zWTKnm8sO5H/gi6QvubXOv3IwWgC0vQO4Rx+i4UHntfw1mG49/cQQnJ41lY+PwdFW29HdRm6KKBsnuf52lOKuagQ+2xb+RpzRdu8FA7qSn0XQ6Iha+h5O7vNh1WkUaM/bOoEtoF56Ne1aaLgDHV8GRFXDtZGgzQpqsEIKX1ieQUljNu6M7Exko71xfiaiPQkWDI2lvPqf35NF1SBTNOv5X0856QwhBwcyZmM6eJXLpUnRN5HVurLXUMnnnZDxcPFhw/QJ0TvL6pFCQABsnQ3Rf6P+yPF3gq4PZrDuay9MDY+gXGypV+0pEGbqiQVGSU82ur1IIjw2g5wi5m6AV/1pD5XcbCH5qAt595VVjCiGYuXcmmVWZLLtxGaGeEo3NWAlrxjri5aOWg7M8S0jIqWTmhkSubxXCxBtipOleyShDVzQYzEYrW5aewt3ThUEPt8PJWV5E0JiUROFrr+HVty/B48dL0wVYk7KGLRlbmNR1Ej0a95AnLARsmAjlmfDgZvCW90FSZbTwxKojhPi48e7ozqpHSx2hYuiKBoEQgl2rUqgqNnDjw+3w9JVXjWmvrSV38jM4+/vT5I3X0SQNbABH8dD8Q/PpE96Hh9o/JE0XgCOfwun1MOAVaHqNNFkhBNPWJZBXYWTh3Z0J8JJ3rq901Apd0SBI3pfPmYOF9BjRjPBW8tLlAApmz8aclUXTFZ/iEiivk6DeomfKrin4ufnx2rWvya0ELTgFP0yFFgOg90R5usCqg1lsOpnPC0Na0y1K3v/31YAydMUlpyyvll9WnyE81p9uQ6Olald8u/7fcXOvHhLDHcCrB14lqzqLjwd9TKC7RGMz1cDXDzri5iOXSBshB5CUX8Xs709zXasQHruuuTTdqwVl6IpLitVi48dPEtG5OXPjQ+2kxlLNGRkUzJmDZ48eBD/+uDRdgI3pG9mQtoHxncbTPay7VG22TIWSs3D/BvCWV4VqtNiY+NUxfD10vH2n6n5vLcsAACAASURBVG1eH6gYuuKSsn99OqW5Ndwwtg1efm7SdIXFQu6U53DS6Wjy5nw0Z3kl7jnVOby6/1W6hHbh0Y6PStMF4PR3cOxz6PsMNLtOqvRrm5M4W1TD23d2Ithb3rm+mlArdMUlI+t0KSe2Z9OhXwTRHeTlmwMUv/8BxlOnCH9/IbpG8hp+We1Wpu2eBsC8vvNwcZL4J1iZ68hqadIV+k2TpwtsTypk5b5MHrm2GX1jJPamucpQhq64JBiqzWxfkURAYy963ya3EVPtwYOULluG/x2j8L3xRqnayxKWcbz4OK/3fZ1w73B5wnY7fPsY2Cxw+8dSBzwXVRt5/uuTtGnsy3NDYqXpXo0oQ1dIRwjBji+SMeotjJjYGRdXeeEOW1UVeS9MxbVpUxpNnSpNF+Bk8UmWnFjCTc1v4qbmN0nVZt8HkLEbbv4AguR9gAoheP7rk9SYrKy+qzNuLvLO9dWIMnSFdJL35XPuRAm9b29JcIS3VO3CV1/FWlRE9OqvcPLykqart+h5cc+LhHqG8lLPl6TpAlCYCD/PgdbDoct9UqVXHcxiZ0oxs25uR0wjH6naVyPK0BVSqSoxsHvNWZrE+NN5gLyBwwBVW390pCg++SQeHSS2pQXeOfIOmVWZfDLoE3xcJRqb1QTrHgN3PxjxHkjsHJlRUsvcjUn0jQlmzDVR0nSvZpShK6Qh7ILtnyUBMOD+NlL7m1uLiymYMQP39u0JfvwxaboAe3P3sjplNWPajpFb2g+w83UoTIC7V4OXvI1nm13wzJrj6Jw15o/qqFIUJaHSFhXSOPFzNnlnK+h7Zwy+wfIG/wohyJ/+CnaDwVHar5O3IVhpqmT63uk092vOxC5yKzLJOgDx70KXMRA7VKr0R7vSOJpVwZxb26shzxJRK3SFFMoLatm/Pp3ojsG07tVYqnblum+p2bmTRtOm4tZCbkbNGwffoNRQysIbFsqdC2rWw/rHwTdC+lzQ5IIq3t12hmEdwri5k7wWxApl6AoJ2M+HWlzcnOh3b6zUCUCWggIK583DI64bAWPGSNMF2JG1g+/Tv+fxTo/TLqidVG1+ngNl6TB2A7j7SpO12OxMWXsCX3cdc25pr0bJSUaFXBT1zvGfsig8V8V1d7WSWw16PtQibDaavPaa1C6KlaZKZu+fTWxALI92kFwNmrkX9i+G7o9A8+ulSn+0M41TuVXMvbU9QaoaVDpqha6oV8ryajn4/Tmadw4hJk5eRSZA5bp11O7eTaOXX8a1aVOp2vMOzqPCWMHigYvRSSziwVwL3z0J/k1h4Cx5ujgaby38+SwjOjVhaAe5YTWFA2XoinrDbhdsX5mEzs2Z6++5FKGW1/Hs0YOAe+6Wpgvwc9bPbErfxBOdnqB1YGup2mw/H2q5fyO4ycvx/zXU4uehY9bNksNLin+jQi6KeuPE9myKMqroe1eM1IEVQgjyZ8xA2Gw0fnWu9FDL3P1ziQ2I5ZGOj0jTBSBrPxz4CLqPg2Z9pUov/SWdxDxHqCVQDay4ZFzUb7qmaUM0TUvRNC1V07TfrZfWNO1OTdNOa5qWqGnaqro9TMXlRkWhngMbHFktskMtVd9/T+2uXwh9ehKukXKLlxYcXkCZsYzZfWbLHfRsMcJ3E8AvAgbOkKcLpBZV8962swzrEMaQ9irUcin505CLpmnOwIfAjUAOcEjTtA1CiNMXvCYGmAb0EUKUa5qmxndfxQi7o1eLs7PG9XfLDbVYS0oofPU1PDp3JuA+uWXue3P3sj51PQ+3f5i2QW2larPrDSg9C/etAzd5lag2u6NXi6ebM7Nubi9NV/H7XMwKvQeQKoRIF0KYgdXALb95zTjgQyFEOYAQoqhuD1NxOZG4O5e8sxX0GRWDd4DcTIeCua9i1+sdoRaJPc71Fj2z9s0i2jea8Z3lDpkm7zjEvwed74WWA6RKf7Y3g6NZFbwyvC0hPiqr5VJzMYYeDmRfcD/n/GMX0gpopWlavKZp+zVNG/J7b6Rp2qOaph3WNO1wcXHx3ztiRYOmuszI3m/TiGgdQJs+cr9+V2/fTvWWLQQ/+YT0AqKFxxaSX5vP7D6zcXOWaGw2C2yY4CjrH/yqPF0gu0zPm1tT6B8bwsguElsBK/6QutotcgFigH7A3cAyTdP8f/siIcRSIUScECIuJEQ1ub/SEELwy1cpCJug372tpYZabNXVFMyajVtsLEEPPyxNF+BE8QlWJa1idOxouoR2karNvg+gIAGGLXDMCJWEEIKX1p/CSYO5IzuoAqIGwsUYei5w4c5SxPnHLiQH2CCEsAghzgFncBi84ioi9UgRGQml9Li5OX4hcvt3FL31FtaSEhrPnSO1V4vFZmHm3pmEeoYyqeskaboAlKY5mm+1Hg5tb5Yqvf54Lr+cKeb5Ia0J91e9WhoKF2Poh4AYTdOaaZrmCtwFbPjNa9bjWJ2jaVowjhBMeh0ep6KBY6y1sPtfZwiN8qHTDRFStfWHD1Ox+l8Ejh0rvS3uJ6c+IbUilenXTMfbVWJvdyHg+0ng7OpYnUuktMbE7O9P07WpP/eptrgNij81dCGEFZgAbAWSgDVCiERN02ZrmvbrsmArUKpp2mlgB/CcEKK0vg5a0fCI/yYVY62V/mNa4+QsL+/bbjKRP/0VdOHhhEx8SpouQHpFOktPLmVo9FCuj5RbYs+xLxwTiG6cDb5y9yrmbDxNjcnK67d3xFm1xW1QXFSlqBBiM7D5N4+9csFtATxz/qK4yshJLiN5bz5dB0cRHCF3Kk3pkqWYz50j8uOPcfL0lKZrF3Zm7ZuFh4sHz/d4XpouADVF8OPL0LQ3dL1fqvSuM8WsP57HxAExtFITiBocqlJU8Y+wWmzsXJWCb4gH3W+KlqptSk2lZNkyfG8egfe1faRqrzu7jqNFR5kSN4VgD3mDIwDYMg0sescEIolVsHqzlZe+TaB5iBdP9pebRaS4OJShK/4RR37IpLLIQL97Y6UOexZ2O/mvzMDZ01P6sOcSQwlvH3mb7mHdubXlrVK1ObsNTn0NfZ+FkFZSpd/bdpaccgPzRnZQw54bKMrQFX+b0rwajm7JJLZnGJGtA6VqV6z9GsPRo4S+8AIugXK13zj4BkarkenXTJebrmeuhU2TIbgVXDtZni6QmFfJx3vOMToukp7Ng6RqKy4e1W1R8bcQdsHOL1Jw9XChz6iWUrUtRUUULViAZ8+e+I2Uu0LenbObLRlbeLLzkzTzayZVm53zoCILHvwBXOQVL9nsghfXJRDgqWPaMMndIxV/CbVCV/wtTsfnUZBeSZ9RLfHwkdtdr+j11xEmE41nzZS6QtZb9MzdP5fmfs15qP1D0nQByD8J+xY5NkGjekuV/nxfBidyKpk+vC3+nqqTYkNGrdAVf5naShP7vk0jPNaf2GvCpGrX/PILVZt/IHjiU7hGR0vV/ujkR+TV5rFiyApcnSUam90GG58Gz0C4Ue7QivxKAwt+PMN1rULUfNDLAGXoir9M/NepWMw2+t0jt7zfbjBQMGs2rs2bE/SI3F7jKWUprExcye0xt9OtUTep2hxeDrlH4LaPpZb3A8zacBqLzc5cNR/0skAZuuIvkZVYytlDhfQY0Qz/RvLyvgFKFi3CkptL1OcrcXKVt0K2Czuz983Gz82Pyd3kbkZSlQ/bZkHz/tBhlFTpbacL2ZJYwPNDYmkaJPdcK/4eKoauuGgsZhu7vkrBv5EnXQfJLfk2ppyh9NMV+N1+G57du0vVXpuylpMlJ5kSNwU/Nz+p2myZCnYLDH8bJK6Qa01WZmxIpFUjb8b1bS5NV/HPUIauuGgOb86gqsRIv3ticdbJ+9URdjsFM2bg7O1N6JQp0nQBivXFvHv0XXo27snw5sOlanPmRzi9Hq6bAoFyTfXdbWfIrTDw2sgO6CS2clD8M9SZUlwUpXk1HP8xi9bXhBEeKzeOW7H2awzHjztyzgPkas8/NB+zzXwJcs71sPlZCI6F3nK7OJ7Oq2J5fAZ394gkLlpujr/in6Fi6Io/RdgFu1aloPNwprfknHNrSQlFb72FZ48e+N3620FZ9Ut8bjxbMrbwROcniPKV3FVw1xuOnPMHNoOLvP0Cm13w4rcJ+HvoeGGIyjm/3FArdMWfkrQvn/zUSnrf1hIPb7l5yIWvv4EwGAibOUPqCtloNTJ3/1yifaN5uL3cgRkUJjoGV3S+D6Ll9qhZdTCL49kVvDy8jco5vwxRK3TF/8RQbWbvulSaxPjTprfcNq018fFUbdxI8BNP4NZcbgx56cml5NTksHzwcsk553bYOBnc/WDQHHm6QFG1kflbkunTMohbO6uRcpcjytAV/5P4b1KxGG1cf3es3Jxzk4mC2bNxjYoi6LFHpekCpFWk8Wnip9zc4ma6h8nNqOHoZ5B9AG5d7CgkksicjUmYLHbmqJzzyxZl6Io/JCelnJT9BXQbGkVgEy+p2qVLlmDJzKLp8k9wcpPXt+TXnHMvnRdT4uRm1FBTBNtmQHRf6HS3VOlfzhTz/Yk8nh4YQ/MQiZOXFHWKiqErfhebxc6u833O44ZGS9U2padTsuxjfEeMwKu33L4l36V+x9Giozzb7VkC3OVm1LD1JUd2y01yc86NFhsvrz9F82AvxvdTfc4vZ5ShK36XI1szqSjUc/3dreT2OReCghkzcfLwoNELcicBlRnLeOvIW3QN7Sq/z3naDkhY42iLK7nP+Qc/p5JVpmfuyPaqz/lljjJ0xX9RUajnyJYMYuJCadpWbu/ryvXfoT90iNBnn8UlWO4koLcOv0WtpZZXer0iN4ZsMcKmZx3FQ32flacLnC2sZskvadzWNZzeLSRPXlLUOSqGrvh/CCHYuSoFF50zfe6IkaptLS+n6I038OjcGf875PYtOZh/kA1pGxjXYRwt/CWHHfa8DWVpMGY96NylydrP55x7ubnw0rA20nQV9YdaoSv+H2cOFJCbUk6vkS3w8pO3GQlQ9OYCbDU1hM2ahSZxVqbJZmLO/jlEeEfwaEe5GTUUn4Hdb0PH0dCiv1TptUeyOZRRzotD2xDkLfdcK+oHtUJX/BtjjYU9X6fSqJkv7a6V2/u69sBBKtetI2jcONxj5caQP0n4hIyqDJYMXIK7i7wVMkI4+py7esGgV+XpAiU1Jl7bnEyPZoHcERchVVtRfyhDV/ybvd+mYtJb6XdvazQniTnnZjMFM2eii4gg+Inx0nQB0ivT+TjhY4Y2G0rvcLkZNRz/EjLjYcRC8A6RKv3qpiT0ZiuvjVQ551cSytAVAOSdrSApPp8ug5oSHCE3D7l02TLM584RuWwpTh4e0nSFEMzdPxd3F3ee7y43o4baEvjxZWjaC7qMkSq952wJ3x7LZeINLWkZ6iNVW1G/qBi6ApvFzs4vk/EJcqf7TXIHH5vSz1H60RJ8hw3Du29fqdrrU9dzqOAQk7tNJthDcobH1pfAVAPD3wWJ+wVGi42X1ifQLNiLJ/rLbbSmqH+UoSs4sjWT8gI9198Ti85Nds75DDQPDxq9OE2aLkCpoZQFhxfQNbQrt8fcLlWbtJ/h5GpHznmo3I6GC7efJbNUz6sj2+OuUznnVxrK0K9yygtqHTnn3RsR1U5yzvm6dY6c8+emSM85f/Pwm+itemb0moGTJvHPwKx3NN8Kaik95zy5oIqlv6QzqluEyjm/QlEx9KsYYRfs+CIZnasz18rOOS8tpXD+m3jEdcP/drkr5PjceDalb+LxTo/T3F/yeLVf5kN5Bty/UXrO+bR1Cfh66FTO+RWMWqFfxZyOz3P0Ob+9JZ6+kvucvzYPodfTWHLOucFqYM7+OUT7RvNIh0ek6QJQkADxC6HLfdBM7n7BFwcyOZZVwfThbQjwUn3Or1TUCv0qpbbSxN51aZemz/muXVRt2kTwk0/i1kJuVeai44vIrcll+eDluDlLLKax22DDU46WuDfK7XOeV2Fg/pYU+sYEqz7nVzjK0K9Sdq8+g81ip/99raXmIdtqasmfOQvXli2k9zlPLE1k5emVjGo1Sn6f8wMfQd4xGLVcap9zIQTT15/CZhe8NrKDyjm/wrmo77qapg3RNC1F07RUTdOm/o/X3a5pmtA0La7uDlFR16QfKybtWDHdh0fj38hTqnbxe+9hLSig8Zw5OLnK++pvsVuYuXcmQe5BTO42WZou4IiZ/zwXYgZDu9ukSm9KyGd7chHPDmpFZKDcc62Qz58auqZpzsCHwFCgLXC3pmltf+d1PsAk4EBdH6Si7jDpLexanUJQhDedb2wqVVt/7BjlX3xBwD334Nmli1TtlYkrSS5L5qWeL+Hr6itPWAhHVovmBDe9JbXPeYXezMwNiXSM8OOB3tHSdBWXjotZofcAUoUQ6UIIM7Aa+L3x63OANwBjHR6foo7Z+20ahiozN4xpjbOzvM1Iu9lM/vTpuDRqRMhkuSvkzKpMFp9YzMCmAxkQNUCqNidWO/LOB8wA/0ip0nM3JVGut/D6bR1xkXiuFZeOiznL4UD2Bfdzzj/2bzRN6wpECiE2/a830jTtUU3TDmuadri4uPgvH6zin5GTUs7p3Xl0GtiU0CiJq1Sg9KOPMKemETZzBs7e8sbZ2YWdV+JfwdXZlWk95RYvUV0IW6ZCZE/o/rBU6V1nivn6SA6PX9+ctk3knmvFpeMff2xrmuYEvA38aZWEEGKpECJOCBEXEiK3GdHVjsVkY8fnSfiFeNBjhNzyfmNyMiVLl+F78wh8+vWTqr0mZQ1Hi47yXNxzhHqGStVm8xSwGODmD8BJXlVmjcnKi+sSaBHixVM3yK0vUFxaLsbQc4ELvytGnH/sV3yA9sBOTdMygGuADWpjtGFxYEM6VSVGbhjbGp3MkXJWK/kvvYyznx+NpsldIefV5PHOkXfo3aS3/JFyp7+DpA3Qb6r0kXLztySTV2lg/qhOqrz/KuNi0hYPATGapjXDYeR3Aff8+qQQohL4dx2xpmk7gSlCiMN1e6iKv0tBeiUnfs6m/XXhNImRO/i49NNPMSYmEv7uO7gEyNMWQjB732wEQv5IOX0ZbJoCjTtB74nydIGD58pYuS+TB/tE0y1K8pBrxSXnT1foQggrMAHYCiQBa4QQiZqmzdY07eb6PkDFP8NqsfHzyiS8/d3oNVJuEY8pLY2S9z/A58aB+AweLFV7fep64vPiebrr04R7Sy6m+eEFMJTBLR+Cs7xSD4PZxvNfnyAy0IPnBsdK01U0HC7qt00IsRnY/JvHXvmD1/b754elqCsObjhHeYGemyd2xtVDnrkIq5W8aS/i5OlJ2IwZUlfIBbUFzD80n7hGcdzV+i5pugAkbYSENdBvGoR1kCo9f2syGaV6vhp3DZ6uqmbwakSd9SuYgvRKjm3Lom3fJkS2lVedCOdDLSdP0uStBVI7KQohmLl3JjZhY3af2XI7KerLHDnnYR2kd1I8kF7Kp/EZ3N8ril4t5HbNVDQcVHLqFYrVbGP7Z0l4B7jR5za5gwxMqamULHwfn0GD8B02TKr2urPriM+LZ3K3yUT6yM37ZvNzjlDLrYvBWSdNVm+28tzXJ2ka6MkLQ+X2V1c0LJShX6Hs35BORaGeG8a0uTShFm9vwmbI3YzMq8njzcNv0j2sO6NjR0vTBeD0Bjj1NVz3vPRQyxs/JJNVpmf+qI4q1HKVo87+FUjumXJObM+m3XXhRLaRG2opWbIEY0KCI6slSN5Xf7uwMz1+uiO7pbfkUEt1IWx82pHV0vcZebrA7rPFfLYvkwd6R3NNcxVqudpRK/QrDLPByvYVSfgGe9DndrmhFkPCKUoWf4TviBH4DhkiVfvLpC85WHCQF3q8QIRPhDxhIeD7SY75oCOXSg21VOotPLf2JC1CvJiqQi0KlKFfcexZe5aaciM3PthW6nxQu9FI3gsv4BIURNjLL0nTBUirSOPdI+/SL6IfI1uOlKrNsS/gzA8wcIb0+aAzNpyiuMbE23d2VgVECkCFXK4ozp0oJmlvPt2GRBHW3E+qdvE772BOTyfyk49x9pOnbbFbeHHPi3jpvJjRW256JOUZjl4t0X2h53h5usCmk/msP57H0wNj6BTpL1Vb0XBRhn6FUFtpYscXyQRHetN9uNxeLTXx8ZR9tpKAe+7Bu08fqdqLjy/mdOlp3un3DsEeEgcf26yw7jFAg1sXgcQxevmVBl78NoGOEX482V9uWE3RsFEhlysAIQQ/r0zCbLRx44PtcHaRd1qt5eXkT52Ga4sWhD43RZouwOGCw3yc8DEjW45kYNRAqdrseQey98Pwt8FfXl95u13w7JoTmK123h3dGZ1qi6u4ALVCvwI4uSOHrMQyrrurFYFN5LWmFUKQ//J0bBUVRC5dgpOHhzTtSlMl0/ZMI9Inkqk9/nCIVv2QfQh2zoMOd0DHO6VKL9udzt60Ul6/rQPNQ7ylaisaPsrQL3NKc2vYty6N6A5BtL9ebs+SijVrqdm+ndAXXsC9TRtpukII5uyfQ4m+hM+HfY6nTuJoNVM1rHsEfMNh2AJ5usCp3EoW/JjCkHZhjO4uuWhKcVmgDP0yxmK28eMnibh6utB/TBupG4Km1FQK583Dq3dvAu8fK00XHI23tmZsZWKXibQPbi9Vm01ToCILHtgMHvI2I2tNViauPkaQlxvzblPDnhW/jwrAXcbsWXOWsrxaBj7QBk9feQOX7UYjuZOfwcnTk8bz5qFJ3BBMr0hn3sF5dA/rzkPtH5KmC8DxVXBytaMaNKqXVOnp353iXEktb4/uRICXvHOtuLxQK/TLlLOHCjm9J4+uQ6Jo2lZuhWDha/MwnT1L5LJl6BrJmwJktBp5dtezeLh48Hrf13GWOAWI4jOw6VlHiuL1z8vTBb4+ksO6o7lMGhBD7xYSM3kUlx3K0C9DKor07PgymbDmfvSUPE6uavNmKtasIWjcI3j3vVaq9vxD80mtSGXxwMVyx8lZDLD2AdB5wG3LpI6TSy2qYfr6U/RsFsjEAWqcnOJ/owz9MsNqsfHjx4k4OWkMeqQdThLT1syZmeRPfwWPzp0JmSh3Es+Wc1tYe2YtD7V/iGvD5X6QsGUaFCXCvd+Ab2NpsgazjQmrjuLh6sx7d3XB2UnFzRX/GxVDv8zYs+YsxVnVDLi/DT6B7tJ07QYDORMnobm4EP7WAjSdvJ4l6RXpzNg7g84hnZnQZYI0XQCOfwVHPoVrJ0OMvFx3IQQvrz9FSmE1b9/ZiTA/eedacfmiVuiXEcn78kncnUfXwVE06xQiTVcIQcGs2ZjOnCFy6RJ04fLSI/UWPZN3TsbdxZ0F1y9A5yTvg4SCU46BFdF9of/L8nSB1Yey+eZoDhMHxNAvVmJ4SXFZowz9MqEkp5qdq1IIj/Wn581y4+YVa9dSuX49wU8+iXffvtJ0f50+lFGVwdIbl9LIq5E0bYyVsGYMuPvBqOVSZ4Mm5FQyY0MifWOCmaTi5oq/gDL0ywBjrYUflpzCzdOFQQ+3lxo3NyQkUDhnLl7XXkvwE3IbUH2Z9CU/ZPzApK6T6Nm4pzxhux2+HQ/lmfDAJvCWt0IurzUz/ssjBHu5qri54i+jYugNHLtd8NPyRGrKjAwZ115qvrm1uJicCU/hEhpKkzfnoznLy+44kH+ABYcX0D+yv/x881/mQ8omGDRXar651WbnyVVHKao2sei+bgSqfHPFX0St0Bs4+9enkZVYRr97Y2ncUl5lot1sJmfiJGxVVUR/tQqXgABp2jnVOUzZNYVo32jm9Z0nd/pQ0kZHn5ZO98A1cr+RvLY5mb1ppSy4oxOdVUtcxd9AGXoD5syhAo79mEW768Jp11feRqQQgsI5czEcO0b4O2/j3lre4Aa9Rc+kHZOwCRsLb1iIl05eszGKkuDbx6BJVxj+Dkgsr//mSA7L48/xYJ9oRnWTOHFJcUWhDL2BUphRxY6VyTRu6UffO+VujJV/8SUVa9cS9Oij+A4dKk3XLuy8HP8yqRWpLBqwiKa+8trSUlsKX90NOk8Y/QXo5KUJHsksZ9q3CfRuEcRLw+Q1OVNceShDb4BUlxnZvOgkHr6uDHm0g9T+5jW7dlE4bx7eN9xAyCS5xUMLjy7kp8yfeC7uOfqESxyUYTXBv+6Fqjx4YCP4yfs2lF2m59GVh2ni586H93TFRfU3V/wD1G9PA8NssLLpwxNYzTaGP9lJ6iaoMSWF3MnP4NY6lnDJm6Dfnv2WT059wp2t7mRM2zHSdBECNjwFWftg5GKI7CFNutJg4cEVh7DaBZ880F013VL8Y5ShNyDsNjs/fpJIWb6ewY+2lzqswlpcTPb48Th5exO5eDFOXvK0D+YfZPa+2fRq3IupPafKbQ37y5tw8l/Q/yVof7s0WYvNzoRVR8koqWXxfV1poYZVKOoAFXJpIAgh2PXVGTJPlXL9PbFSOyjaamrIeuwxbOUVRH3xObpG8gp4UspSmLRjElG+USzoJ7kS9NiXsONV6DgarntOmqwQgqnfJLD7bAnzb++oOigq6gxl6A2EQ5syOL0nj25Domh/ncSMFrOZnKeewpRyhsjFi/Bo106adn5NPk9sewJPnSeLBy7G19VXmjZnf3KEWpr3g5s/kJrR8ubWFL45msPTA2O4U00eUtQhytAbAIm7czm08Ryte4XR85bm0nSF3U7eiy+h37f//9q77/iq63uP46/vmRkne++wNyYYQGQPt4IdiLuOor1S1LqKtdfaqq20V1y1Wq2VguKiWhEVFEUc7BkIIySB7J2c5CRnn/O9f/yil3qxIOb8YsL3+Xicx1k/8vkekrzzO9/zHaT94Q/YpkzRrbbdbefmdTfj8rtYesFS0mz6rWJI1Q54/VpIGQGXLQeTfn3X/9h4lL98UsoV47LUtH6l251UH7oQ4nwhxCEhRIkQ4v/tyCuEuEMIsV8IUSiE+EgIkdP9Te2bSnc2sGHFIbJHJDDt6qG69R9LKan/wyO0r15N0h13EPuDS3WpC9pY8wUfL6DKtc92IQAAG7dJREFUUcUTM55gcNxg3WrTeAhWzIXIJLhqJYTp965g1Z4aHniniFnDUnhwzki1jZzS7U4Y6EIII/A0cAEwHLhCCDH8a4ftAgqklKOBlcAfu7uhfVH5vmY+eKGIlH7RnDd/BEYdh6w1PvEErcuXE/+Ta0mY/1Pd6rr9bhZ+vJCipiL+OOWPjE0dq1ttWo7AsjkgjHDNWxCl32cFH+6v5xev7WZsTjxPXZGvhicqIXEyP1XjgBIpZZmU0gu8Csw59gAp5XoppbPr7mZATXU7geriVt7/617i0yO5+OdnYAnTr/er6fnnaX72r8TO/THJi/QbVeIL+Lhzw51srdvKgxMfZFaOfuuL01YNy2aD3w3Xvg0JA3Qr/fnhJha8vJOR6dG8cF0B4RYdt85TTisnE+gZQOUx96u6HvsmNwLvH+8JIcRNQojtQojtjY2NJ9/KPqaurI13ny4kOiGM2bfmYY3Qb2RHy/KXaHx0CdEXXUTqAw/oF+ZBH4s+W8SnVZ/y6/G/5pIBl+hSFwBHvXZm7myFq9+ElK+/wQydrUdamL9sO/0SI1l6/TiiwnQcxaOcdrr1fZ8Q4mqgAPjT8Z6XUj4npSyQUhYkJem3QcP3SW2JnVVP7CYi2sLs2/IJj9LvA7mWZcuof/hhbLNmkv7IH3SbOOQL+vjlp7/kg/IPuKvgLuYNnadLXQAcdbD0Im0W6FWvQ8YY3UpvLmvmuhe3khYbxvKfjlMTh5SQO5n3+dXAsWOrMrse+zdCiFnAfcBUKaWne5rXt9SU2Fn91B4iYixc+osx2OKsutVuXrqUhkcWE3XOOWQseVS3LeR8AR/3fHoP6yrWcVfBXfxkxE90qQtoIf6PS7RQv/qfui6Fu6m0mRuWbiMjLpwV88eTHKW2kFNC72TO0LcBg4QQ/YQQFuByYNWxBwgh8oG/ArOllA3d38zer/pQK+88tYfIWCs/uEPnMH/hBS3Mzz1X1zD3BDzcueFO1lWs456x9+gb5vZKWHpxj4T554ebuH7pVjLjwnll/lkqzBXdnPAMXUrpF0L8HFgLGIG/SymLhBC/A7ZLKVehdbHYgDe6+mQrpJSzQ9juXuXInkbWPl9EdGIYc36RT2SMPmEupaTxscdpfu45oi+8gPTFi3UL805fJ7etv40ttVtYNG4RVw27Spe6ADQWw/JLwdOh9Zln67fb0Zp9tdz6ym76J0Xy0k/Hk2jT7w+3opzU0Aop5XvAe1977P5jbus4XKF3ObS5lo+WHSQpy8YlC/MIs+kTqDIQoO7BB7G/+hqxl11G6m/u163P3O62c8tHt7C/eT8PT3qY2QN0/Nteswte+hEIA1z/LqSO0q3069sqWfRmIXlZsbx43ThidPywW1FAzRQNGSklez6q5IuVJWQMiePC/xql29DEoMdDzS8X4VizhoT5PyXpjjt0G81S01HDLetuodJRyZJpS5iRPUOXugCUrofXroHwOLj2X7oNTZRS8uyGMhavOcjkQYn89ZozibCoXy1Ff+qnLgSCQcnnrx9m7ydVDMhPYtYNwzGZ9Tk79re2UnXLAly7dpF8990k3Kjffpz7m/ez4KMFePwenpn1DOPS9FuKll0vwzu3QuJgbQaoTmua+wNB7l9VxIotFVxyRjr/M3c0VpMaZ670DBXo3cznCfDBC0UcLWwib1YWZ/9wIEKnndu95eVU3nQzvtpaMh5boutuQ59WfcpdG+4i1hrL8xc8z8C4gfoUllLbA3TDYm2hrcuWQViMLqU7PH4WrtjJ+kON/Ne0Adx97hAMOn2vFeV4VKB3o/ZmF+8/u5fmqg6mXD6YUdP0mzDbuWkT1bf/AoQge+mLRIzRZ7y1lJJ/FP2Dx3Y+xpC4ITw982mSInSaY+DthLcXQNFbkHc1XPI4GPXpt65odjJ/2XZKGjv4/Q9GceV4HbfLU5RvoAK9m9QcbuX9v+4jGJBctOAMckbqs565lJLW5S9Rv3gxln65ZD39NJYcfdZGc/vd/HbTb1ldtppzcs7hoYkPEWGO0KU29gp49Uqo2wezfgsTb9NtCdyNJU3csmInUsLS68cyedDpOUlO+f5Rgf4dSSnZ+0k1X7xxmOikcC66ZTSxKfqEWtDlou63v6PtX//CNnMm6YsXY7Tps9NQdUc1d35yJ0XNRSzMX8j8UfP1Wz2wbAOsvAECPrjqDRh0ji5lpZS88PkR/vD+QfonRvL8tQXkJuq3s5OinIgK9O/A4/KzfvlBSnc2kDsqgVk3jMAars9/qafsCNW33YanpITEBQtIXHALwqDPCn7rK9Zz3xf3IaXkyelPMj17ui51CQbg0//R+swTB8PlL0OiPmuKt7l83LNyD2uL6jl3eApL5uVhs6pfH+X7Rf1EnqLGCgdrn99He7ObCT8YQP452bp9+Nn2zmrqfvMbhNVK1nPPYZs8SZe6voCPJ3c9ydKipQyLH8ajUx8lK1qnHXc6GuCtm6H0Y23LuIuWgFWffTgLq+wsWLGTWrubX180jBsn9VNrmSvfSyrQv6VgULL7wwq2rCoj3Gbm0jvySR8Yq0vtQHs7dQ8+RPs77xA+ZgwZSx7FnJqqS+2ytjIWfbqIAy0HmDdkHnePvRurUadZkIfeh7d/Dh4HXPIkjLlWl/7yQFDy7IZSHvuwmOQoK6//bAJjsuNCXldRTpUK9G/B0eJm3Yv7qTlsp39+EtOvGqrbzM/OrVupWbQIf30DibcuJPGmmxCm0H/7pJS8dug1Ht3+KGGmMB6f/jgzs2eGvC6gjWJZex/seBFSRsF1qyF5mC6lK1uc/OK13Wwvb+Xi0Wk8fOkoNfNT+d5TgX4SZFBS9Fk1G98sBWDGtcMYOiFVl7fdgY5OGpcsoXXFCsw52eSueJnwM84IeV2AyvZKHtj0AFvrtjIxYyIPnv2gfkMSyz6BVbdqo1km3gbT7wNT6N8RBIOSZZuO8se1hzAKwePz8piTl666WJReQQX6Cdjrnax/6SA1h+1kDo1j+tVDiU4M16V2x2efU/ub+/HX1hF37TUk3347hojQj6AJBAO8fOBlntr1FCaDifsn3M+PB/1Yn1Bz2eHD/4adyyB+AFz/HuScHfq6QElDB4v+Wcj28lamDk7i9z8cRUasPt9rRekOKtC/gc8bYOeacnZ+UI7JbGT6NUMZdnaaLqHmq6ujfvFiHO+vwdK/Pzkvv0zEmPyQ1wXY3bCb32/5PQdaDjA1cyq/PuvXpEbq0E8vJex5VQtzZ7N2Vj7tXjCHPlCdXj9//riE5z8rI8Ji4tG5Z/DDMRnqrFzpdVSgf42UkiN7mvj8jcM4mt0MHpfC2T8aqMuSt0Gvl9Zly2j8yzMQCJB460ISbrwRgzX0tZtcTTy580neKnmL5Ihk/jT1T5yXc54+oVZbCO/fAxWbIHOsthZLel7Iy0opWbOvjofePUC13cWPxmSy6IKhJEWpJW+V3kkF+jEaytv5YmUJNYftxKdHcukd+WQMDv2oBikljjVraHh0Cb6qKmwzZpDyq3uxZIZ+6QCX38Xy/ct5Ye8LeANerh95PT8b/TN9Zny218DHD8HuFRARD7P/DHlXgQ7j6XdX2nn43f1sO9rK0NQo3vjZBMbmxoe8rqKEkgp0wN7gZNvqIxRvrSc8yszUKwYzbFI6RmPog6Vz82YaH3sc1549WIcMIeuFv2GbODHkdX1BH++UvsNfdv+Femc9M7NncvuY28mNyQ15bZwtsOnPsOkvIANw9kKYfCeEh374Z2ljB499WMzqwloSbVZ+/4NRXFaQiUmH77WihNppHejtTS62v3eUg5vrMBoFY87LYcz5ObrM9nTu2EHjE0/i3LoVU0oKaQ8/RMyll4Z8E4pAMMB7R97jmT3PUOmoZGTCSB6Z/AgFqQUhrQuAuw02PwObngZPO4z8Ecy8H+JyQ166otnJEx8d5q1dVVhNRhbOGMjNUweo2Z5Kn3Ja/jS31HSy64NyirfWIwyCUdMyGHNeTsj7yaWUdH6xkebnnsO5dSvGxERSfnUvsfPmhbyf3Bvw8nbp27y470UqHZUMjR/KUzOeYmrm1ND3k3c0wpZnYdvzWqgPvVj7wDN1ZGjrAofqHDy7oZRVe2owGQQ3TOzHz6YNUFvDKX3SaRPoUkpqS+zsXlfJkT1NmMwGRk7NIP/cbGxxod3EV3q9tK/9gJYXX8S9fz+m5GSSf/lL4i6fhyE8tKM42jxtrCxeyYoDK2hwNTA8YTiPTXuMGdkzMIgQdzM0lWhBvusl8Lth2CUw+Q5ID+2IHSklm8qa+fvnR1h3oIEIi5Hrzs7lpin9SYlWGzYrfVefD3SfN0DJ9noK11fRVNmBNdJEwYW5jJ6RSbjNEtLa/sZGWt94A/srr+JvbMSSm0vaQw8SPXs2Bktoax9qOcQrB1/h3bJ3cQfcjE8bz0OTHuKstLNCe0YeDGrrrWx5Fko+BKMFRl2mDUNMGhy6umjDD1ftrmHpxqMcrHMQF2Hm9lmD+MmEXOIiQ/v/rSjfB3020JuqHOz/rIZDW+vxuvzEp0cy7aohDB6fitkSun5qGQjQuXEj9tdfx7H+E/D7iZw8mbSHHyJy0qSQrojo9DlZe3QtK4tXUthUiNVo5eL+F3PlsCsZHBfaMKW9RtsGbucyaKuAyGStW6XgBrAlh7T0vuo2Xt1Wwdu7anB4/AxNjeKPPxrN7Lx0wnTa+k9Rvg/6VKB3tnk4vK2eQ1vqaKrswGgyMGBMEsMnpZM+KDakZ6buQ8W0rXqb9ndW429owBgXR/y11xI798dY+/ULWd1AMMDWuq2sLlvNh+Uf4vK76B/Tn7sL7mb2gNnEhoVw5Ii3Ew6shj2vwJENIIPQbyqc84DWTx7Cqfr17W7e3l3NmzurOVjnwGoycNGoNC4fl83Y3Dg1KUg5LfX6QHc5vJTuaqR0ZwPVh1qREpJzopg8bxCDx6USFhm6BZU8ZWW0r1mDY81aPMXFYDJhmzSJmF/dS9SMGYgQdasEZZBdDbtYe3Qt68rX0ehqJMocxYX9LmTOwDnkJeWFLtC8Tq0rpegtKF4LPifEZmvDDs+4AhIGhKYu0OjwsKaojvcKa9lypJmghLysWH43ZwRzzshQi2cpp71eGej2eidHCps4WthEbYkdKSE2JYIx5+cwZHwqcamh2UVGBgK4Cgvp+Hg9jvUf4y0pBSEIHzOGlPvuI/qiCzHFh2ZyisvvYkvtFj6p/IQNVRtocjVhNVqZnDGZ8/qdx/Ss6aFbzrajEQ6v1ZaxLf1YC/GIRG1d8lFzIXtCyCYDlTZ2sG5/PesO1LOjvJWghAFJkfx8+kAuzc+gf5I+a6IrSm/Q6wJ9+3tH2LLqCAAJmTbOvCCXAWOSSciIDMlZqa++gc6NG+n8/HM6v/iCgN0OJhMRYwuIm3c5Ueeeizml+/uIpZSU2kvZWLORL2q+YEf9DjwBDzazjUkZk5iZPZMpmVNCM6PT74XqHVD6EZSsg5rdgIToTMi7EobNhpyJYOz+Hx+H28fmshY2FDfwaXETFS1OAEakR7NwxiAuHJXG4BSb6lJRlOPodYGeMzIRS7iZ3NEJRCd0/5A/X20tzp07cW7dhnPLFrxHjwJgTEzENnUKkZOnYJsyGWN0dLfWDcogpfZSdjXsYmvdVrbVbaPF3QJAv5h+zB08l8mZkxmbMhZzd+9s73ND7W4o3whHP4OKzdpZuDBoa6tMuxeGnA+po7t9Y4k2p48dFS1sOdLC5tJm9la3EZQQYTFy9oAE5k/ux4xhKWrVQ0U5Cb0u0JOyo0jKjuqWrxV0uXAfOIB7715cewpx7t6Fv6YWAENkJOEFZxI7dy6RE87COnRot45Qsbvt7Gvex96mvRQ2FrKncQ8OrwOA5PBkJqRPYFzqOCakTSDNltZtdZFSW2O8eod2qdoONTsh4NWeTxoG+VdD7mToN6Vbp+P7A0EON3Swp9LOnio7O8vtHKrXXrPZKDgjM5YF0wcyYUACBTnxWExqOr6ifBu9LtBPhZSSQFMT7uJiPMWHcR/Yj+fAATxlRyAQAMCUmkp4Xh4R111H+JgzCRs6pFt2BArKIFWOKg7bD1PcWszB5oMcaDlAbaf2h0Mg6B/Tn3NzziU/OZ/85HyyorK6p0vB74GmYqjfD/V7tVUN6wrB1ao9b7RC2mgYfzNknQVZ48HWPRtYtLt9HK53cLDOwf6advbVtHOwth2PPwhAdJiJvOw4Lh6dxpm5ceRlxRJhOS1+HBUlZPrUb5D0evFWVeMtP4q3vBxvaRmeI2V4S8sItLZ+dZwpOZmwYcOwzZpF+KhRhI0ciTn5u/WDO31OKhwVlLeXU95eTllbGWX2Mo62H8XldwFaeOdE55CXlMflQy9nZMJIRiSOINL8HT7ElVJbP7y5FFpKoemwFuJNxdpjUvuDhdECycO12ZppZ0DGmZA8AkynPhInGJTUtrs52tTJkaZOShs7KGnooLShg5o291fHRYWZGJEezdVn5TAyI5q8rDhyEyJUP7iidLNeF+j+5mY8JaX4amvw1XRdKqvwVVXhq63VZip2McbFYenfn6hZM7EOGoR18GCsgwef0kiUDm8H9c56ajtrtUtHLdUd1VR3VFPlqKLZ3fxvx6dFptE/pj9nppzJoLhBDIodxIDYAd/+Q8xgUAtsRw20VYG9EtoqofUo2MuhtVxb6OpLBhPE94fEwVp4Jw+HlBGQMBC+Zd+7PxCkweGhts1Njd1Ftd1FdauLylYnFS1OqlpdeP3/9/8dYTEyIMnGuH7xDEqJYmhqFENSo8iIDVfhrSg66HWBbv/nmzQuWfLVfWNiIpbMTMLz84mefQmWnBztkpuLKe6b1zKXUuLyu2j1tGJ322lxt9DibqHZ3Uyzq5kmVxONrkYanY00OBtw+p3/9u+NwkhqZCoZtgymZk0l05ZJdnQ2udG5ZEVl/efgllJbpMrZrF06m6CzQRse2NkAjjroqAdHrXb7y/7tL5nCIDYH4nK0rpL4flpgxw/QHvsPwe0PBLG7fLR2emnu9NLS6aWpw0OTw0Njh5dGh5v6dg8NDjeNDg9B+e//PjrMRFZ8BENSopg1LIWchAj6JUSSmxhJanQYBoMKbkXpKUJKeeKDhDgfeAIwAn+TUj7yteetwDLgTKAZmCelPPqfvmZBQYHcvn37t26wt6ICb1UVwZQEfInRuAwBOn2duPwuOn2ddPg6tGtvBw6fQ7v2Omj3ttPmafvqus3ThjfoPW6NMGMYieGJJEUkkRSeRFJEEqkRqSRHJJMakUy6JY4kUzhGnwu8HeDp6Lp2dF3awd2uhba7Ddx2ba9MVyu4WrTbX3aFfJ01BqJSICoVbKkQnQbRGRCVhj8qA1dEGp3GODp9ATo9fjo8fjo9ATo8PjrcftrdfhxuP+1uH+0uH20u7brV6cPu9NLu9h+3rBAQH2EhOTqM5CgryVFW0mLCSI0JJzXGSkZsBOmxYUSFqck7itKThBA7pJTHXe/6hGfoQggj8DRwDlAFbBNCrJJS7j/msBuBVinlQCHE5cBiYN53b/r/90rLuzx6+Bnk4RP/IQKwGazYjBaiDVaiDRayDWZijTHERMQTI4zESwMxUhAnIc4fICEQJNzvgRY3NBwB337wORE+F8LvwtDVH34iEgNecxQ+kw2PORq3KRqXMQdn9Gg646LoMETTbojBTgxtIoomGUMT0XT4jbh9AVz2IJ7GAC5fAKdXu/b664C6E9a2GA1Eh5uICjMTE24mNsJCTkIkcRHa7bgIM/E2KwmRFuIjLSTarMRHWjCqs2tF6dVOpstlHFAipSwDEEK8CswBjg30OcADXbdXAn8WQgh5Mqf/31LY/iLmt9oJl5KIYJBwKYkMBonsum8LSmzBIJEySGRQ8k1LMwWlwIsJD2Y8WPBiwi0tlGPBgxmXtODGgpt4XDINF1ZcWHBhxSmtOAnDKa10Ek4HYXTKcByE45ARdBCOEyu4jx+QRoPAYjRgNRuwmgxYTAbCTEbCzJIwc5BIq4kEm5Ews5Fws4EIi4kws5EIi3YJtxixWU1EWkxEWI1EWc3YwkxEhZmwWU1qQSpFOU2dTKBnAJXH3K8Cxn/TMVJKvxCiDUgAmo49SAhxE3ATQHZ29ik1ODN3Hq62LAIGE05hokMYaTCYCAoTUpiQRhNBzAQNJqTRTNBgJijMYLQQNFgIGsxIkwUhTBgMAqNBYBDa9VcXITAYBCaDdm02CMIMAlvX82aj4avjTEbtvskgMBkNmI0Ck0ELaXPXc9ptA5au2+pMWFGUUND1Q1Ep5XPAc6D1oZ/K15h49mQmnj25W9ulKIrSF5zMVLxqIOuY+5ldjx33GCGECYhB+3BUURRF0cnJBPo2YJAQop8QwgJcDqz62jGrgJ903f4x8HEo+s8VRVGUb3bCLpeuPvGfA2vRhi3+XUpZJIT4HbBdSrkKeAFYLoQoAVrQQl9RFEXR0Un1oUsp3wPe+9pj9x9z2w3M7d6mKYqiKN+GWs5OURSlj1CBriiK0keoQFcURekjVKAriqL0ESe1OFdICgvRCJT3SPHvJpGvzYA9TZyOr1u95tNHb3rdOVLK4+5E02OB3lsJIbZ/00pnfdnp+LrVaz599JXXrbpcFEVR+ggV6IqiKH2ECvRv77mebkAPOR1ft3rNp48+8bpVH7qiKEofoc7QFUVR+ggV6IqiKH2ECvTvQAhxpxBCCiESe7otoSaE+JMQ4qAQolAI8ZYQIran2xRKQojzhRCHhBAlQohFPd2eUBNCZAkh1gsh9gshioQQt/V0m/QihDAKIXYJIVb3dFu+KxXop0gIkQWcC1T0dFt08iEwUko5GigG7u3h9oTMMRujXwAMB64QQgzv2VaFnB+4U0o5HDgLWHAavOYv3QYc6OlGdAcV6KfuMeAe4LT4VFlK+YGU0t91dzPazlV91Vcbo0spvcCXG6P3WVLKWinlzq7bDrSAy+jZVoWeECITuAj4W0+3pTuoQD8FQog5QLWUck9Pt6WH3AC839ONCKHjbYze58PtS0KIXCAf2NKzLdHF42gnZsGebkh30HWT6N5ECLEOSD3OU/cBv0LrbulT/tNrllK+3XXMfWhvz1/Ws22KPoQQNuCfwO1Syvaebk8oCSEuBhqklDuEENN6uj3dQQX6N5BSzjre40KIUUA/YI8QArSuh51CiHFSyjodm9jtvuk1f0kIcR1wMTCzj+8ZezIbo/c5QggzWpi/LKV8s6fbo4OJwGwhxIVAGBAthHhJSnl1D7frlKmJRd+REOIoUCCl7C0rtZ0SIcT5wBJgqpSysafbE0pCCBPaB78z0YJ8G3CllLKoRxsWQkI7O/kH0CKlvL2n26O3rjP0u6SUF/d0W74L1YeunKw/A1HAh0KI3UKIZ3u6QaHS9eHvlxujHwBe78th3mUicA0wo+v7u7vrzFXpRdQZuqIoSh+hztAVRVH6CBXoiqIofYQKdEVRlD5CBbqiKEofoQJdURSlj1CBriiK0keoQFcURekj/hccwWPZbHFDGAAAAABJRU5ErkJggg==\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Función sigmoide con diferentes valores de $x$ para  $y=0.5$\n",
        "\n",
        "$f(t)=\\frac{1}{1+e^{-\\beta_1t}}$"
      ],
      "metadata": {
        "id": "Emz5ql9aw4Tt"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "import matplotlib.pyplot as plt\n",
        "import numpy as np\n",
        "x=np.arange(-5,5,0.1)\n",
        "y1=1/(1+np.e**(-x))\n",
        "y2=1/(1+np.e**(-2*x))\n",
        "y3=1/(1+np.e**(-3*x))\n",
        "y4=1/(1+np.e**(-4*x))\n",
        "plt.plot(x,y1)\n",
        "plt.plot(x,y2)\n",
        "plt.plot(x,y3)\n",
        "plt.plot(x,y4)\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 283
        },
        "id": "TMnPqqqztIzw",
        "outputId": "e3bdc538-92fc-43f8-883a-3c8a33f77990"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[<matplotlib.lines.Line2D at 0x7f730f8b70d0>]"
            ]
          },
          "metadata": {},
          "execution_count": 3
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Función sigmoide con regresión logística \n",
        "\n",
        "$f(t)=\\frac{1}{1+e^{-(\\beta_0+\\beta_1x_1+\\beta_2x_2+\\cdots+\\beta_kx_k)}}$\n",
        "\n"
      ],
      "metadata": {
        "id": "dIrNdsGNn6rL"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Clasificación de SPAM\n",
        "\n",
        "Tomado de https://coderzcolumn.com/tutorials/machine-learning/feature-extraction-from-text-data-using-scikit-learn-sklearn"
      ],
      "metadata": {
        "id": "OukIVW0KH0V9"
      }
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "OEXyRdX_DmH6"
      },
      "outputs": [],
      "source": [
        "#Sólo ejecutar una vez\n",
        "#!wget https://archive.ics.uci.edu/ml/machine-learning-databases/00228/smsspamcollection.zip\n",
        "#!unzip smsspamcollection.zip"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "!cd SMSSpamCollection"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "NqBgf_PRJjLv",
        "outputId": "631136d0-04bd-4eaa-e7f2-68b17ca2d6b3"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "/bin/bash: line 0: cd: SMSSpamCollection: Not a directory\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "with open('SMSSpamCollection') as f:\n",
        "  datos=f.readlines()\n",
        "\n",
        "type(datos)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "2Qu0igHwJ0QH",
        "outputId": "0ad77366-db2d-41a5-dae8-fb23b68abecf"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "list"
            ]
          },
          "metadata": {},
          "execution_count": 6
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "len(datos)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "92zf1lJDKPfZ",
        "outputId": "c08dd19d-b1f7-495e-8979-390515c47a71"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "5574"
            ]
          },
          "metadata": {},
          "execution_count": 7
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "type(datos[0])"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "rundy_AIKTeK",
        "outputId": "99f4abc5-9b42-4624-bc49-2b32bd88a518"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "str"
            ]
          },
          "metadata": {},
          "execution_count": 8
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "len(datos[0])"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "blDHJ3SHKaW6",
        "outputId": "d5efc197-68d7-4507-e3d8-85270cd44d03"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "116"
            ]
          },
          "metadata": {},
          "execution_count": 9
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "datos[0]"
      ],
      "metadata": {
        "id": "JzwDMwUfKeoi",
        "outputId": "327540e1-70bb-4a62-ff47-648bc6499eae",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 36
        }
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "'ham\\tGo until jurong point, crazy.. Available only in bugis n great world la e buffet... Cine there got amore wat...\\n'"
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "string"
            }
          },
          "metadata": {},
          "execution_count": 10
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "with open('SMSSpamCollection') as f:\n",
        "    data = [line.strip().split('\\t') for line in f.readlines()]\n",
        "\n",
        "y, text = zip(*data)\n",
        "\n",
        "type(y), type(text)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "MDsG9pB7D0jf",
        "outputId": "e59213ca-d6ae-4b3d-bf12-5704b2266391"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(tuple, tuple)"
            ]
          },
          "metadata": {},
          "execution_count": 11
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "len(y), len(text)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "GffQmjquD-q4",
        "outputId": "ad42da43-1500-4a7a-abe4-0c6fb52e3086"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(5574, 5574)"
            ]
          },
          "metadata": {},
          "execution_count": 12
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "type(y[0]), type(text[0])"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "HClRPl-DEDMS",
        "outputId": "c737e4b1-1dc4-4850-ce3a-d7d3df3272f7"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(str, str)"
            ]
          },
          "metadata": {},
          "execution_count": 13
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "y[0], text[0]"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "8W20g7v7EPgc",
        "outputId": "a30fff9f-eaa0-4e5d-f03a-155f30ef798e"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "('ham',\n",
              " 'Go until jurong point, crazy.. Available only in bugis n great world la e buffet... Cine there got amore wat...')"
            ]
          },
          "metadata": {},
          "execution_count": 14
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "y[1], text[1]"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Z6kYyslSAlA7",
        "outputId": "68191bdc-69c4-4491-8594-2069898eb25f"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "('ham', 'Ok lar... Joking wif u oni...')"
            ]
          },
          "metadata": {},
          "execution_count": 15
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "clases_cont={}\n",
        "for dat in y:\n",
        "  if dat in clases_cont:\n",
        "    clases_cont[dat] += 1\n",
        "  else: \n",
        "    clases_cont[dat]=1\n",
        "\n",
        "clases_cont"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "zOs8kOOlEVSl",
        "outputId": "6f9a6599-25c3-4c7e-8658-9bfaa5f313a2"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "{'ham': 4827, 'spam': 747}"
            ]
          },
          "metadata": {},
          "execution_count": 16
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "from sklearn.model_selection import train_test_split\n",
        "\n",
        "text_train, text_test, y_train, y_test = train_test_split(text, y,\n",
        "                                                          random_state=42,\n",
        "                                                          test_size=0.25,\n",
        "                                                          stratify=y)\n",
        "\n",
        "len(text_train), len(text_test), len(y_train), len(y_test)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "w07yZk7sE-hx",
        "outputId": "819d71ac-7efd-405e-b2c3-ef1d4dd4a296"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(4180, 1394, 4180, 1394)"
            ]
          },
          "metadata": {},
          "execution_count": 17
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "https://es.wikipedia.org/wiki/Tf-idf"
      ],
      "metadata": {
        "id": "-nIFhnH9EtOu"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "from sklearn.feature_extraction.text import CountVectorizer\n",
        "from sklearn.feature_extraction.text import TfidfVectorizer\n",
        "\n",
        "count_vectorizer = CountVectorizer()\n",
        "count_vectorizer.fit(text_train)\n",
        "\n",
        "tfidf_vectorizer = TfidfVectorizer()\n",
        "tfidf_vectorizer.fit(text_train)\n",
        "\n",
        "X_train_cnt = count_vectorizer.transform(text_train)\n",
        "X_test_cnt = count_vectorizer.transform(text_test)\n",
        "\n",
        "X_train_tfidf = tfidf_vectorizer.transform(text_train)\n",
        "X_test_tfidf = tfidf_vectorizer.transform(text_test)\n",
        "\n",
        "\n",
        "\n",
        "X_train_cnt.shape, X_test_cnt.shape, X_train_tfidf.shape, X_test_tfidf.shape"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "m75VtbU--2ZO",
        "outputId": "41b723da-58aa-468c-e4e4-a6beaee16dd1"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "((4180, 7448), (1394, 7448), (4180, 7448), (1394, 7448))"
            ]
          },
          "metadata": {},
          "execution_count": 18
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "from sklearn.linear_model import LogisticRegression\n",
        "\n",
        "clf_cnt = LogisticRegression()\n",
        "print(clf_cnt.fit(X_train_cnt, y_train))\n",
        "\n",
        "clf_tfidf = LogisticRegression()\n",
        "print(clf_tfidf.fit(X_train_tfidf, y_train))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Zf1bmN8EAw8X",
        "outputId": "e41376b4-f853-4cac-f72f-24b45ad2b38f"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "LogisticRegression()\n",
            "LogisticRegression()\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "print('Count Vectorizer Performance : %.3f'%clf_cnt.score(X_test_cnt, y_test))\n",
        "print('TfIdf Vectorizer Performance : %.3f'%clf_tfidf.score(X_test_cnt, y_test))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "zCvtJa4cAw5V",
        "outputId": "bb079f6c-9920-46ca-c86d-74d05c8c4fe1"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Count Vectorizer Performance : 0.984\n",
            "TfIdf Vectorizer Performance : 0.930\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [],
      "metadata": {
        "id": "uQ6_rcbiFDDR"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "print('Count Vectorizer Performance : %.3f'%clf_cnt.score(X_test_tfidf, y_test))\n",
        "print('TfIdf Vectorizer Performance : %.3f'%clf_tfidf.score(X_test_tfidf, y_test))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "PqmtIUUWAw2N",
        "outputId": "d753091f-6476-489a-c5ff-f3546e06697d"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Count Vectorizer Performance : 0.866\n",
            "TfIdf Vectorizer Performance : 0.973\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "coef = clf_cnt.coef_.ravel()\n",
        "positive_coeffs = np.argsort(coef)[-20:]\n",
        "negative_coeffs = np.argsort(coef)[:20]\n",
        "interesting_coeffs = np.hstack([negative_coeffs, positive_coeffs])\n",
        "# plot them\n",
        "with plt.style.context(('seaborn', 'ggplot')):\n",
        "    plt.figure(figsize=(18, 8))\n",
        "    colors = [\"red\" if c < 0 else \"green\" for c in coef[interesting_coeffs]]\n",
        "    plt.bar(np.arange(2 * 20), coef[interesting_coeffs], color=colors)\n",
        "    feature_names = np.array(count_vectorizer.get_feature_names())\n",
        "    plt.xticks(np.arange(1, 2 * 20 + 1), feature_names[interesting_coeffs], rotation=60, ha=\"right\")\n",
        "    plt.xlabel('Feature Importance')\n",
        "    plt.xlabel('Feature')\n",
        "    plt.title('CountVectorizer Features Importance')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 608
        },
        "id": "kXtxqRaNAwzG",
        "outputId": "69b6fc4c-9f4b-43a8-c9f8-81bae7876084"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/sklearn/utils/deprecation.py:87: FutureWarning: Function get_feature_names is deprecated; get_feature_names is deprecated in 1.0 and will be removed in 1.2. Please use get_feature_names_out instead.\n",
            "  warnings.warn(msg, category=FutureWarning)\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1296x576 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "coef = clf_tfidf.coef_.ravel()\n",
        "positive_coeffs = np.argsort(coef)[-20:]\n",
        "negative_coeffs = np.argsort(coef)[:20]\n",
        "interesting_coeffs = np.hstack([negative_coeffs, positive_coeffs])\n",
        "# plot them\n",
        "with plt.style.context(('seaborn', 'ggplot')):\n",
        "    plt.figure(figsize=(18, 8))\n",
        "    colors = [\"red\" if c < 0 else \"green\" for c in coef[interesting_coeffs]]\n",
        "    plt.bar(np.arange(2 * 20), coef[interesting_coeffs], color=colors)\n",
        "    feature_names = np.array(tfidf_vectorizer.get_feature_names())\n",
        "    plt.xticks(np.arange(1, 2 * 20 + 1), feature_names[interesting_coeffs], rotation=60, ha=\"right\")\n",
        "    plt.xlabel('Feature Importance')\n",
        "    plt.xlabel('Feature')\n",
        "    plt.title('TfIfVectorizer Features Importance')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 600
        },
        "id": "NAJ6qVQoBNOL",
        "outputId": "b6d9327c-9681-4957-f537-6066addfbe9d"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/sklearn/utils/deprecation.py:87: FutureWarning: Function get_feature_names is deprecated; get_feature_names is deprecated in 1.0 and will be removed in 1.2. Please use get_feature_names_out instead.\n",
            "  warnings.warn(msg, category=FutureWarning)\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1296x576 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Referencias\n",
        "\n",
        "* https://coderzcolumn.com/tutorials/machine-learning/feature-extraction-from-text-data-using-scikit-learn-sklearn\n",
        "* https://es.wikipedia.org/wiki/Tf-idf\n",
        "* https://scikit-learn.org/stable/modules/classes.html#module-sklearn.feature_extraction.text\n",
        "* https://scikit-learn.org/stable/modules/generated/sklearn.linear_model.LogisticRegression.html"
      ],
      "metadata": {
        "id": "ivjs_VGqGC64"
      }
    }
  ]
}