DM_assignment_2 (1).ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": []
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "cells": [
    {
      "cell_type": "code",
      "execution_count": 51,
      "metadata": {
        "id": "C6b1M-vP5Oxz"
      },
      "outputs": [],
      "source": [
        "import matplotlib.pyplot as plt\n",
        "import pandas as pd\n",
        "import numpy as np\n",
        "import heapq"
      ]
    },
    {
      "cell_type": "code",
      "source": [],
      "metadata": {
        "id": "Us9uFyVeKuVJ"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "df = pd.read_csv('/content/cluster_data1.csv')\n",
        "df.shape\n",
        "\n",
        "df_test = pd.read_csv('/content/cluster_data2.csv')\n",
        "df.shape"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "tJSzNZ3Y6jxd",
        "outputId": "eb88e888-1a54-4aa4-853e-301252332f8d"
      },
      "execution_count": 15,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(150, 4)"
            ]
          },
          "metadata": {},
          "execution_count": 15
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "missing_values = df.isnull().sum()\n",
        "print(missing_values)\n",
        "print(df.dtypes)"
      ],
      "metadata": {
        "id": "Vzcnq7hm5qxE",
        "outputId": "0e1bef6d-572c-4d3d-9692-63cdd4dba72c",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "execution_count": 16,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "X1    0\n",
            "X2    0\n",
            "X3    0\n",
            "X4    0\n",
            "dtype: int64\n",
            "X1    float64\n",
            "X2    float64\n",
            "X3    float64\n",
            "X4    float64\n",
            "dtype: object\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "#visualizing the first two features\n",
        "\n",
        "plt.figure(figsize=(8, 6))\n",
        "plt.scatter(df.iloc[:, 0], df.iloc[:, 1], marker='o', color='blue')\n",
        "\n",
        "plt.xlabel('Feature X1')\n",
        "plt.ylabel('Feature X2')\n",
        "plt.title('Scatter Plot of First Two Features')\n",
        "plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 564
        },
        "id": "CkdEljsvABvq",
        "outputId": "1065e262-b9d8-45d3-ad46-fcf4c03c4e6a"
      },
      "execution_count": 62,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 800x600 with 1 Axes>"
            ],
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAArMAAAIjCAYAAAAQgZNYAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/bCgiHAAAACXBIWXMAAA9hAAAPYQGoP6dpAABdR0lEQVR4nO3deXxTVf7/8Xea0payVFCBQiuFioCAooAKWhYFGcUZ+HZwRVnUGXdBZkAY/YLgggMu4Ghdv4L7AlaZcUUQtAIzgwsMuCDWqqUWt9FCKVJNz++P/BpJ27RJbpKb27yej0cekHPPzf3cc29uPr05OcdljDECAAAAHCjJ7gAAAACAcJHMAgAAwLFIZgEAAOBYJLMAAABwLJJZAAAAOBbJLAAAAByLZBYAAACORTILAAAAxyKZBQAAgGORzAJotj7//HO5XC4tW7bM7lD8vPrqq+rfv7/S0tLkcrn0448/hvU68bp/ABBLJLOAA23dulXjx49X165dlZaWpi5dumjUqFH629/+FrVtPvnkk1q8eHG98q+++ko33HCDNm/eHLVt17Vu3Tq5XC7fo0WLFurevbsmTpyozz77LCLb2LBhg2644YawE81Avv/+e5111llq2bKl7rnnHj322GNq1apVg3WXLVvmt58HPmbNmhXRuGrdcssteuGFF5qsN3z48ICxHfi44YYbohJnIJMnTw4Yy6uvvhqVbQZ6bwCIjWS7AwAQmg0bNmjEiBE67LDD9Ic//EGdOnVSaWmp/vnPf2rJkiW66qqrorLdJ598Utu2bdO0adP8yr/66ivNmzdPOTk56t+/f1S2HcjVV1+tQYMG6eeff9Z7772nBx54QC+99JK2bt2qzp07W3rtDRs2aN68eZo8ebIOOuigyAQsadOmTdqzZ49uvPFGjRw5Mqh15s+fr27duvmV9e3bV127dtW+ffvUokWLiMV3yy23aPz48Ro3blyj9a677jpdfPHFvuebNm3SXXfdpb/85S/q3bu3r/yoo46KWGzBSk1N1UMPPVSv/Oijj47K9gK9NwDEBsks4DA333yzMjIytGnTpnpJ1jfffGNPUFGwd+/egHcsa+Xl5Wn8+PGSpClTpuiII47Q1VdfrUceeUSzZ8+ORZghqz1GoSTIp512mgYOHNjgsrS0tCbXD6YtQzVq1Kh6cdx1110aNWqUhg8fHtFthSo5OVnnn3++rTFEQlVVldLT0+0OA4h7dDMAHKa4uFh9+vRpMBnq0KFDvbLHH39cxx13nNLT09WuXTsNHTpUq1at8i1fuXKlxowZo86dOys1NVW5ubm68cYb5fF4fHWGDx+ul156SV988YXvK9ucnBytW7dOgwYNkuRNJmuXHdiH81//+pd+85vfKCMjQ+np6Ro2bJjWr1/vF+MNN9wgl8ulDz/8UOedd57atWunk046KeS2OfnkkyVJJSUljdZ74403lJeXp1atWumggw7S2LFj9dFHH/nFM2PGDElSt27dfPv1+eefN/q6y5cv14ABA9SyZUsdcsghOv/881VWVuZbPnz4cE2aNEmSNGjQILlcLk2ePDnk/azVUJ/ZyZMnq3Xr1iouLtbpp5+uNm3aaMKECZKkHTt26Pe//706deqktLQ0ZWVl6ZxzzlFFRYUkyeVyae/evXrkkUd8+xxufHfddZfcbrdfN43bb79dLpdL06dP95V5PB61adNG1157ra9s7969+tOf/qTs7GylpqaqZ8+euu2222SMCSuWumpqarR48WL16dNHaWlp6tixoy655BL98MMPfvWsvDekX7uJ1D1varvJrFu3zu91+vbtq3fffVdDhw5Venq6/vKXv0iS9u/fr7lz5+rwww9XamqqsrOzNXPmTO3fv9/vdV9//XWddNJJOuigg9S6dWv17NnT9xpAc8adWcBhunbtqo0bN2rbtm3q27dvo3XnzZunG264QUOGDNH8+fOVkpKif/3rX3rjjTd06qmnSvJ+4LZu3VrTp09X69at9cYbb2jOnDnavXu3Fi1aJMn7lXJFRYV27typO++8U5LUunVr9e7dW/Pnz9ecOXP0xz/+UXl5eZKkIUOGSPImjaeddpoGDBiguXPnKikpSUuXLtXJJ5+soqIiHXfccX7xnnnmmerRo4duueWWsBKX4uJiSdLBBx8csM7q1at12mmnqXv37rrhhhu0b98+/e1vf9OJJ56o9957Tzk5OcrPz9cnn3yip556SnfeeacOOeQQSdKhhx4a8HWXLVumKVOmaNCgQVqwYIG+/vprLVmyROvXr9f777+vgw46SNddd5169uypBx54wNd1IDc3t8n9qqio0HfffedXVhtTQ3755ReNHj1aJ510km677Talp6erurpao0eP1v79+3XVVVepU6dOKisr04svvqgff/xRGRkZeuyxx3TxxRfruOOO0x//+EdJCiq+huTl5ammpkZvv/22zjjjDElSUVGRkpKSVFRU5Kv3/vvvq7KyUkOHDpUkGWP0u9/9TmvXrtVFF12k/v3767XXXtOMGTNUVlbmO/+aUre9WrRooYyMDEnSJZdc4jteV199tUpKSnT33Xfr/fff1/r1633dNqy8N8Lx/fff67TTTtM555yj888/Xx07dlRNTY1+97vf6e2339Yf//hH9e7dW1u3btWdd96pTz75xNe/+YMPPtAZZ5yho446SvPnz1dqaqo+/fTTen84As2SAeAoq1atMm6327jdbjN48GAzc+ZM89prr5nq6mq/ejt27DBJSUnmf/7nf4zH4/FbVlNT4/t/VVVVvW1ccsklJj093fz000++sjFjxpiuXbvWq7tp0yYjySxdurTeNnr06GFGjx5db3vdunUzo0aN8pXNnTvXSDLnnntuUG2wdu1aI8k8/PDD5ttvvzVfffWVeemll0xOTo5xuVxm06ZNxhhjSkpK6sXWv39/06FDB/P999/7yrZs2WKSkpLMxIkTfWWLFi0ykkxJSUmT8VRXV5sOHTqYvn37mn379vnKX3zxRSPJzJkzx1e2dOlSI8kXY2Nq6zb0CLR/kyZNMpLMrFmz/F7r/fffN5LM8uXLG91mq1atzKRJk5qMra7ly5cbSWbt2rXGGGM8Ho9p27atmTlzpjHGez4cfPDB5swzzzRut9vs2bPHGGPMHXfcYZKSkswPP/xgjDHmhRdeMJLMTTfd5Pf648ePNy6Xy3z66aeNxlG7/3Ufw4YNM8YYU1RUZCSZJ554wm+9V199tV651fdG7fGrew7Vnr+1bWWMMcOGDTOSzH333edX97HHHjNJSUmmqKjIr/y+++4zksz69euNMcbceeedRpL59ttvA7YN0FzRzQBwmFGjRmnjxo363e9+py1btmjhwoUaPXq0unTpor///e++ei+88IJqamo0Z84cJSX5v9VdLpfv/y1btvT9f8+ePfruu++Ul5enqqoqffzxx2HHuXnzZu3YsUPnnXeevv/+e3333Xf67rvvtHfvXp1yyil66623VFNT47fOpZdeGtI2LrzwQh166KHq3LmzxowZ4/uKPFD/0vLycm3evFmTJ09W+/btfeVHHXWURo0apZdffjn0HZX0zjvv6JtvvtHll1/u14d1zJgx6tWrl1566aWwXrfWPffco9dff93v0ZTLLrvM73ntXcnXXntNVVVVluIJRlJSkoYMGaK33npLkvTRRx/p+++/16xZs2SM0caNGyV579b27dvX123m5Zdfltvt1tVXX+33en/6059kjNErr7zS5LbT0tLqtdftt98uydsVJCMjQ6NGjfKdk999950GDBig1q1ba+3atb7XidZ7I5DU1FRNmTLFr2z58uXq3bu3evXq5RdvbZea2nhr22/lypX13ldAc0c3A8CBBg0apMLCQlVXV2vLli16/vnndeedd2r8+PHavHmzjjzySBUXFyspKUlHHnlko6/1wQcf6Prrr9cbb7yh3bt3+y2r7UsZjh07dkiSr49oQyoqKtSuXTvf87q/2G/KnDlzlJeXJ7fbrUMOOUS9e/dWcnLgy9oXX3whSerZs2e9Zb1799Zrr70W1o+lGnvdXr166e233w7p9eo67rjjAiboDUlOTlZWVpZfWbdu3TR9+nTdcccdeuKJJ5SXl6ff/e53Ov/8832JbqTl5eX5unIUFRUpMzNTxx57rI4++mgVFRVp1KhRevvtt3XWWWf51vniiy/UuXNntWnTxu+1akdIqG3rxrjd7oAjRezYsUMVFRUN9i+X/H9EGa33RiBdunRRSkpKvXg/+uijgF1cauM9++yz9dBDD+niiy/WrFmzdMoppyg/P1/jx4+v98cs0NyQzAIOlpKSokGDBmnQoEE64ogjNGXKFC1fvlxz584Nav0ff/xRw4YNU9u2bTV//nzl5uYqLS1N7733nq699lpLd3hq1120aFHAIbvq9i088E5YMPr16xf08FaJJDU1tcEE5vbbb9fkyZO1cuVKrVq1SldffbUWLFigf/7zn/WS30g46aST9PPPP2vjxo0qKiry9anOy8tTUVGRPv74Y3377be+8lioqalRhw4d9MQTTzS4vDZpjMR748BvQA504A/IDtTQ+V9TU6N+/frpjjvuaHCd7Oxs37pvvfWW1q5dq5deekmvvvqqnnnmGZ188slatWqV3G53k/ECTkUyCzQTtXfuysvLJXl/uFNTU6MPP/wwYDK5bt06ff/99yosLPT9AEdqeDSAQB/MgcprfzjUtm3buEk4u3btKknavn17vWUff/yxDjnkEN9d2UD71dTr1n79W2v79u2+5fGgX79+6tevn66//npt2LBBJ554ou677z7ddNNNkkLb76Ycd9xxSklJUVFRkYqKinwjRAwdOlQPPvig1qxZ43teq2vXrlq9erX27Nnjd3e29mt9q22Zm5ur1atX68QTT2z0j6dIvDdqv3WoO/FGMHeXD4x3y5YtOuWUU5o8NklJSTrllFN0yimn6I477tAtt9yi6667TmvXro2b9yAQDXz3ADjM2rVrG/ylf21/z9qvuseNG6ekpCTNnz+/3l2k2vVr79Yc+HrV1dUqKCio9/qtWrVq8KvV2uSv7gf2gAEDlJubq9tuu02VlZX11vv2228D7mO0ZGZmqn///nrkkUf84t22bZtWrVql008/3VcWaL8aMnDgQHXo0EH33Xef33BJr7zyij766CONGTMmYvsQrt27d+uXX37xK+vXr5+SkpL8Ym7VqlXEZj1LS0vToEGD9NRTT+nLL7/0uzO7b98+3XXXXcrNzVVmZqZvndNPP10ej0d3332332vdeeedcrlcOu200yzFdNZZZ8nj8ejGG2+st+yXX37x7Xsk3hu1f9DV9huWvHdlH3jggZDiLSsr04MPPlhv2b59+7R3715J0n//+996y2v/iK07hBfQ3HBnFnCYq666SlVVVfqf//kf9erVS9XV1dqwYYOeeeYZ5eTk+H5Acvjhh+u6667TjTfeqLy8POXn5ys1NVWbNm1S586dtWDBAg0ZMkTt2rXTpEmTdPXVV8vlcumxxx5rMFkeMGCAnnnmGU2fPl2DBg1S69at9dvf/la5ubk66KCDdN9996lNmzZq1aqVjj/+eHXr1k0PPfSQTjvtNPXp00dTpkxRly5dVFZWprVr16pt27b6xz/+Eevm06JFi3Taaadp8ODBuuiii3xDc2VkZPhNvTpgwABJ3qGXzjnnHLVo0UK//e1vG+xP26JFC/31r3/VlClTNGzYMJ177rm+oblycnJ0zTXXxGr3AnrjjTd05ZVX6swzz9QRRxyhX375RY899pjcbrd+//vf++oNGDBAq1ev1h133KHOnTurW7duOv7448Pebl5enm699VZlZGSoX79+krzjIffs2VPbt2+vN47tb3/7W40YMULXXXedPv/8cx199NFatWqVVq5cqWnTpoU9VFitYcOG6ZJLLtGCBQu0efNmnXrqqWrRooV27Nih5cuXa8mSJRo/fnxE3ht9+vTRCSecoNmzZ+u///2v2rdvr6effrreHxWNueCCC/Tss8/q0ksv1dq1a3XiiSfK4/Ho448/1rPPPqvXXntNAwcO1Pz58/XWW29pzJgx6tq1q7755hsVFBQoKysrrDGbAUexbyAFAOF45ZVXzIUXXmh69eplWrdubVJSUszhhx9urrrqKvP111/Xq//www+bY445xqSmppp27dqZYcOGmddff923fP369eaEE04wLVu2NJ07d/YN9aU6QwdVVlaa8847zxx00EFGkt9QRCtXrjRHHnmkSU5OrjdU1Pvvv2/y8/PNwQcfbFJTU03Xrl3NWWedZdasWeOrUzs0V7DDCtUObdTUMFMNDV1ljDGrV682J554omnZsqVp27at+e1vf2s+/PDDeuvfeOONpkuXLiYpKSmoYbqeeeYZX1u3b9/eTJgwwezcudOvTjhDcwWqG2horlatWtWr+9lnn5kLL7zQ5ObmmrS0NNO+fXszYsQIs3r1ar96H3/8sRk6dKhp2bKlkRT0MF11h+aq9dJLLxlJ5rTTTvMrv/jii40k83//93/1XmvPnj3mmmuuMZ07dzYtWrQwPXr0MIsWLfIb4i2QQPtf1wMPPGAGDBhgWrZsadq0aWP69etnZs6cab766itfnUi8N4qLi83IkSNNamqq6dixo/nLX/5iXn/99QaH5urTp0+DsVZXV5u//vWvpk+fPr738YABA8y8efNMRUWFMcaYNWvWmLFjx5rOnTublJQU07lzZ3PuueeaTz75pMm2AJzOZUyEplQBAAAAYow+swAAAHAsklkAAAA4FsksAAAAHItkFgAAAI5FMgsAAADHIpkFAACAY8XNpAm33nqrZs+eralTp2rx4sUN1lm2bJlvQPhaqamp+umnn4LeTk1Njb766iu1adMmotM2AgAAIDKMMdqzZ486d+6spKTG773GRTK7adMm3X///TrqqKOarNu2bVu/edVDTUi/+uorZWdnhxwjAAAAYqu0tFRZWVmN1rE9ma2srNSECRP04IMP6qabbmqyvsvlUqdOncLeXps2bSR5G6dt27Zhvw4AAACiY/fu3crOzvblbY2xPZm94oorNGbMGI0cOTKoZLayslJdu3ZVTU2Njj32WN1yyy3q06dPwPr79+/X/v37fc/37NkjyXuHl2QWAAAgfgXzDbytPwB7+umn9d5772nBggVB1e/Zs6cefvhhrVy5Uo8//rhqamo0ZMgQ7dy5M+A6CxYsUEZGhu9BFwMAAIDmw2WMMXZsuLS0VAMHDtTrr7/u6ys7fPhw9e/fP+APwOr6+eef1bt3b5177rm68cYbG6xT985s7W3riooK7swCAADEod27dysjIyOofM22bgbvvvuuvvnmGx177LG+Mo/Ho7feekt333239u/fL7fb3ehrtGjRQsccc4w+/fTTgHVSU1OVmpoasbgBAAAQP2xLZk855RRt3brVr2zKlCnq1auXrr322iYTWcmb/G7dulWnn356tMIEAABAHLMtmW3Tpo369u3rV9aqVSsdfPDBvvKJEyeqS5cuvj618+fP1wknnKDDDz9cP/74oxYtWqQvvvhCF198cczjBwAAgP1sH82gMV9++aXfQLk//PCD/vCHP2jXrl1q166dBgwYoA0bNujII4+0MUoAAADYxbYfgNkllA7FAAAAiL1Q8jVbh+YCAAAArCCZBQAAgGORzAIAAMCxSGYBAADgWCSzAAAAcCySWQAAADhWXI8zCwDB8nikoiKpvFzKzJTy8qQgJhIEADgcySwAxysslKZOlXbu/LUsK0taskTKz7cvLgBA9NHNAICjFRZK48f7J7KSVFbmLS8stCcuAEBskMwCcCyPx3tHtqF5DGvLpk3z1gMANE8kswAcq6io/h3ZAxkjlZZ66wEAmieSWQCOVV4e2XoAAOchmQXgWJmZka0HAHAeklkAjpWX5x21wOVqeLnLJWVne+sBAJonklkAjuV2e4ffkuontLXPFy9mvFkAaM5IZgE4Wn6+tGKF1KWLf3lWlreccWYBoHlj0gQAjpefL40dywxgAJCISGYBNAtutzR8uN1RAABijW4GAAAAcCySWQAAADgWySwAAAAci2QWAAAAjkUyCwAAAMcimQUAAIBjkcwCAADAsUhmAQAA4FgkswAAAHAsklkAAAA4FsksAAAAHItkFgAAAI5FMgsAAADHIpkFAACAY5HMAgAAwLFIZgEAAOBYJLMAAABwLJJZAAAAOBbJLAAAAByLZBYAAACORTILAAAAxyKZBQAAgGORzAIAAMCxku0OAEDz4PFIRUVSebmUmSnl5Ulut91RAQCaO5JZAJYVFkpTp0o7d/5alpUlLVki5efbFxcAoPmjmwEASwoLpfHj/RNZSSor85YXFtoTFwAgMZDMAgibx+O9I2tM/WW1ZdOmeesBABANJLMAwlZUVP+O7IGMkUpLvfUAAIgGklkAYSsvj2w9AABCRTILIGyZmZGtBwBAqEhmAYQtL887aoHL1fByl0vKzvbWAwAgGkhmAYTN7fYOvyXVT2hrny9ezHizAIDoIZkFYEl+vrRihdSli395Vpa3nHFmAQDRxKQJACzLz5fGjmUGMABA7JHMAogIt1saPtzuKAAAiYZuBgAAAHAsklkAAAA4FsksAAAAHItkFgAAAI5FMgsAAADHYjQDII54PAxvBQBAKEhmgThRWChNnSrt3PlrWVaWd4YtJh4AAKBhdDMA4kBhoTR+vH8iK0llZd7ywkJ74gIAIN6RzAI283i8d2SNqb+stmzaNG89AADgj2QWsFlRUf07sgcyRiot9dYDAAD+SGYBm5WXR7YeAACJhGQWsFlmZmTrAQCQSEhmAZvl5XlHLXC5Gl7ucknZ2d56AADAH8ksYDO32zv8llQ/oa19vngx480CANAQklkgDuTnSytWSF26+JdnZXnLGWcWAICGMWkCECfy86WxY5kBDACAUJDMAnHE7ZaGD7c7CgAAnINuBgAAAHAsklkAAAA4FsksAAAAHItkFgAAAI5FMgsAAADHIpkFAACAYzE0F4CE5/Ewvi8AOFXc3Jm99dZb5XK5NG3atEbrLV++XL169VJaWpr69eunl19+OTYBAmiWCgulnBxpxAjpvPO8/+bkeMsBAPEvLpLZTZs26f7779dRRx3VaL0NGzbo3HPP1UUXXaT3339f48aN07hx47Rt27YYRQqgOSkslMaPl3bu9C8vK/OWk9ACQPyzPZmtrKzUhAkT9OCDD6pdu3aN1l2yZIl+85vfaMaMGerdu7duvPFGHXvssbr77rtjFC2A5sLjkaZOlYypv6y2bNo0bz0AQPyyPZm94oorNGbMGI0cObLJuhs3bqxXb/To0dq4cWPAdfbv36/du3f7PQCgqKj+HdkDGSOVlnrrAQDil60/AHv66af13nvvadOmTUHV37Vrlzp27OhX1rFjR+3atSvgOgsWLNC8efMsxQmg+Skvj2w9AIA9bLszW1paqqlTp+qJJ55QWlpa1LYze/ZsVVRU+B6lpaVR2xYA58jMjGw9AIA9bLsz++677+qbb77Rscce6yvzeDx66623dPfdd2v//v1y1xkbp1OnTvr666/9yr7++mt16tQp4HZSU1OVmpoa2eABOF5enpSV5f2xV0P9Zl0u7/K8vNjHBgAInm13Zk855RRt3bpVmzdv9j0GDhyoCRMmaPPmzfUSWUkaPHiw1qxZ41f2+uuva/DgwbEKG0Az4XZLS5Z4/+9y+S+rfb54MePNAkC8s+3ObJs2bdS3b1+/slatWunggw/2lU+cOFFdunTRggULJElTp07VsGHDdPvtt2vMmDF6+umn9c477+iBBx6IefwAnC8/X1qxwjuqwYE/BsvK8iay+fm2hQYACFJczwD25ZdfKinp15vHQ4YM0ZNPPqnrr79ef/nLX9SjRw+98MIL9ZJiAAhWfr40diwzgAGAU7mMaai3WPO1e/duZWRkqKKiQm3btrU7HAAAANQRSr5m+zizAAAAQLhIZgEAAOBYJLMAAABwLJJZAAAAOBbJLAAAABwrrofmAhBb1dVSQYFUXCzl5kqXXy6lpNgdFQAAgZHMApAkzZwp3XGH5PH8WvbnP0vTp0sLF9oXFwAAjSGZBaCZM6VFi+qXezy/lpPQAgDiEZMmAAmuulpKT/e/I1uX2y1VVdHlAAAQG0yaACBoBQWNJ7KSd3lBQWziAQAgFCSzQIIrLo5sPQAAYolkFkhwubmRrQcAQCzRZxZIcPSZBQDEG/rMAghaSop3+K3GTJ9OIgsAiE8MzQXAN+xW3XFm3W7GmQUAxDe6GQDwYQYwAEA8CCVf484sAJ+UFGnaNLujAAAgePSZBQAAgGORzAIAAMCxSGYBAADgWCSzAAAAcCySWQAAADgWySwAAAAci6G5gAjbt0+aMUPasUPq0UNatEhq2dLuqJo/j0cqKpLKy6XMTCkvzzvpAwA0B3Ze4+L9+sqdWSCCxo2T0tOle+6RVq3y/pue7i1H9BQWSjk50ogR0nnnef/NyfGWA4DT2XmNc8L1lWQWiJBx46SVKxtetnIlCW20FBZK48dLO3f6l5eVecvj6YILAKGy8xrnlOsr09kCEbBvn/cObFOqquhyEEkej/cOQd0LbS2XS8rKkkpK4usrMQAIhp3XOLuvr6Hka9yZBSJgxozI1kNwiooCX2glyRiptNRbDwCcxs5rnJOurySzQATs2BHZeghOeXlk6wFAPLHzGuek6yvJLBABPXpEth6Ck5kZ2XoAEE/svMY56fpKn1kgAugza4/aPl1lZd6vvOqizywAJ7PzGmf39ZU+s0CMtWwpjR3beJ2xY0lkI83tlpYs8f7f5fJfVvt88WISWQDOZOc1zknXV5JZIEJeeCFwQjt2rHc5Ii8/X1qxQurSxb88K8tbnp9vT1wAEAl2XuOccn2lmwEQYcwAZo94n6EGAKxItBnAQsnXSGYBAAAQV+gzCwAAgIRAMgsAAADHIpkFAACAY5HMAgAAwLFIZgEAAOBYyXYHADQ3Th4+heGtAABOQzILRFBhoTR1qrRz569lWVneWVSiPbi01W3bGTsAAOGimwEQIYWF0vjx/smg5J3Xevx47/J43badsQMAYAWTJgAR4PFIOTn1k8FaLpf3LmdJSeS/tre6bTtjBwCgIUyaAMRYUVHgZFCSjJFKS7314m3bdsYOAIBVJLNABJSXR7ZeLLdtZ+wAAFhFMgtEQGZmZOvFctt2xg4AgFUks0AE5OV5+5W6XA0vd7mk7GxvvXjbtp2xAwBgFcksEAFut3cIK6l+Ulj7fPHi6PyAyuq27YwdAACrSGaBCMnPl1askLp08S/PyvKWR3OsVqvbtjN2AACsYGguIMKYAQwAAGtCyddIZgEAABBXGGcWAAAACYFkFgAAAI5FMgsAAADHIpkFAACAY5HMAgAAwLFIZgEAAOBYyXYHgPiUyOONJvK+A0A0cX1FNJDMop7CQmnqVGnnzl/LsrK8U54295mgEnnfASCauL4iWuhmAD+FhdL48f4XG0kqK/OWFxbaE1csJPK+A0A0cX1FNDEDGHw8Hiknp/7FppbL5f0ruqSk+X0tlMj7DgDRxPUV4WAGMISlqCjwxUaSjJFKS731mptE3ncAiCaur4g2kln4lJdHtp6TJPK+A0A0cX1FtJHMwiczM7L1nCSR9x0AoonrK6KNZBY+eXnefksuV8PLXS4pO9tbr7lJ5H0HgGji+opoI5mFj9vtHSJFqn/RqX2+eHHz7KCfyPsOANHE9RXRRjILP/n50ooVUpcu/uVZWd7y5jwWYCLvOwBEE9dXRBNDc6FBiTxLSyLvOwBEE9dXBCuUfI1kFgAAAHGFcWYBAACQEEhmAQAA4FgkswAAAHAsklkAAAA4FsksAAAAHCvZ7gCAeFNdLRUUSMXFUm6udPnlUkpKbNZ38rA1To4dAOBctt6Zvffee3XUUUepbdu2atu2rQYPHqxXXnklYP1ly5bJ5XL5PdLS0mIYMZq7mTOl9HTpmmuku+/2/pue7i2P9vqFhVJOjjRihHTeed5/c3K85fHOybEDAJzN1mQ2KytLt956q95991298847OvnkkzV27Fh98MEHAddp27atysvLfY8vvvgihhGjOZs5U1q0yHuH8UAej7e8qYTUyvqFhdL48dLOnf7lZWXe8nhOCp0cOwDA+eJu0oT27dtr0aJFuuiii+otW7ZsmaZNm6Yff/wx7Ndn0gQ0pLraewe1biJ6ILdbqqpquMuAlfU9Hu9dzLrJYC2XyzvlY0lJ/H1t7+TYAQDxy5GTJng8Hj399NPau3evBg8eHLBeZWWlunbtquzs7Cbv4krS/v37tXv3br8HUFdBQeOJqORdXlAQ+fWLigIng5JkjFRa6q0Xb5wcOwCgebA9md26datat26t1NRUXXrppXr++ed15JFHNli3Z8+eevjhh7Vy5Uo9/vjjqqmp0ZAhQ7SzkU/TBQsWKCMjw/fIzs6O1q7AwYqLrdWzsn55eXDrBlsvlpwcOwCgebA9me3Zs6c2b96sf/3rX7rssss0adIkffjhhw3WHTx4sCZOnKj+/ftr2LBhKiws1KGHHqr7778/4OvPnj1bFRUVvkdpaWm0dgUOlptrrZ6V9TMzg1s32Hqx5OTYAQDNQ9z1mR05cqRyc3MbTVAPdOaZZyo5OVlPPfVUUPXpM4uGxEOf2bIy79fydcVzv1Mnxw4AiF+O7DNbq6amRvv37w+qrsfj0datW5XJbR9YlJIiTZ/eeJ3p0wOPF2tlfbdbWrLE+3+Xy39Z7fPFi+MzGXRy7ACA5sHWZHb27Nl666239Pnnn2vr1q2aPXu21q1bpwkTJkiSJk6cqNmzZ/vqz58/X6tWrdJnn32m9957T+eff76++OILXXzxxXbtApqRhQulGTPqJ15ut7d84cLorZ+fL61YIXXp4l+eleUtz88Pfj9izcmxAwCcz9ZuBhdddJHWrFmj8vJyZWRk6KijjtK1116rUaNGSZKGDx+unJwcLVu2TJJ0zTXXqLCwULt27VK7du00YMAA3XTTTTrmmGOC3ibdDNAUZgALj5NjBwDEl1DytbjrMxttJLMAAADxzdF9ZgEAAIBgkcwCAADAsUhmAQAA4FgkswAAAHAsklkAAAA4FsksAAAAHCvZ7gAQn+weM9TqWK92bttK2yVyuycqu4+5k89XAJAkmQRTUVFhJJmKigq7Q4lbzz1nTFaWMdKvj6wsb3kszJhhjNvtv32321se79u20naJ3O6Jyu5j7uTzFUDzFkq+RjILP889Z4zL5f8BJXnLXK7of1DNmFF/2wc+oplYWd22lbZL5HZPVHYfcyefrwCav1DyNWYAg4/HI+XkSDt3Nrzc5ZKysqSSkuh8lVhdLaWne+MIxO2Wqqoi/9W31W1babtEbvdEZfcxd/L5CiAxMAMYwlJUFPgDSvLeeykt9daLhoKCxhMqybu8oCD+tm2l7RK53ROV3cfcyecrANQVUjJbUFCgkSNH6qyzztKaNWv8ln333Xfq3r17RINDbJWXR7ZeqIqLI1svltu20naJ3O6Jyu5j7uTzFQDqCjqZveuuuzRjxgz16tVLqampOv3007VgwQLfco/Hoy+++CIqQSI2MjMjWy9UubmRrRfLbVtpu0Ru90Rl9zF38vkKAHUF3We2T58+uu6663TeeedJkjZs2KBx48bp0ksv1fz58/X111+rc+fO8jT1faXN6DMbWG1fuLIy71eFdTXnvpuR6jMbTtslcrsnKruPuZPPVwCJISp9ZktKSjRkyBDf8yFDhuiNN97QAw88oNmzZ4cfLeKG2y0tWeL9v8vlv6z2+eLF0fuASkmRpk9vvM706dFJqKxu20rbJXK7Jyq7j7mTz1cAqCfYIRKys7PNW2+9Va/8gw8+MB07djQTJ040SUlJwY+5YBOG5mpaQ+NHZmcnxnin0RhnNti2S+R2T1R2H3Mnn68AmreoDM113nnnqWPHjrrzzjvrLfvggw80YsQIff/993QzaCbsntmHGcASr90Tld3H3MnnK4DmK5R8Lehk9j//+Y/effddTZkypcHl27Zt03PPPae5c+eGHnEMkcwCAADEt6gks80FySwAAEB8i8oPwP73f/9Xv/zyS8DlX375pUaNGhV8lAAAAIBFQSezjzzyiAYNGqRt27bVW3b//ferb9++Sk5OjmhwAAAAQGOCTma3bdumfv36aeDAgVqwYIFqamr05ZdfauTIkZo5c6Zuu+02vfLKK9GMFQAAAPATcp/ZlStX6pJLLlGnTp1UUlKi4447Tg899JC6du0arRgjij6zAAAA8S2UfC3kfgEnnHCC+vXrpzVr1qhVq1a6/vrrHZPIAsGwOtwQwxUhUSTyUG6JvO9AvAm6m4EkPfXUUzryyCNVU1Ojjz76SJdddplOPfVUXXPNNfrpp5+iFSMQM4WF3qk6R4yQzjvP+29Ojrc8FusDTjFzpnca5Guuke6+2/tverq3vLlL5H0H4lKwMzHk5+ebVq1ambvuusuvfP369eaII44wRxxxhNmwYUPwUzvYhBnAEMhzzxnjcvnPaCR5y1yupmc2sro+4BQzZtQ/zw98NOdZ4xJ534FYisoMYCeeeKKWLVumHj161Fu2b98+zZo1S/fee6+qq6sjnG5HFn1m0RCPx3sHdefOhpe7XFJWllRS0nCXAavrA05RXe29C9nYZI9ut1RV1fy+dk/kfQdiLSrjzBYVFTWYyEpSy5YttWTJEq1evTq0SIE4UVQUOBGVvPdcSku99aKxPuAUBQWNJ3OSd3lBQWziiaVE3ncgngWdzCYlNV116NChloIB7FJebq2e1fUBpygujmw9J0nkfQfiWUg/AAOaq8xMa/Wsrg84RW5uZOs5SSLvOxDPQh5n1unoM4uG1PZ5LSvzdgmoK9g+s+GuDzhFIvcbTeR9B2ItKn1mgebM7ZaWLPH+3+XyX1b7fPHiwImo1fUBp0hJkaZPb7zO9OnNM5lL5H0H4hnJLPD/5edLK1ZIXbr4l2dlecvz86O7PuAUCxdKM2bU/+PM7faWL1xoT1yxkMj7DsSrsLoZPPbYY7rvvvtUUlKijRs3qmvXrlq8eLG6deumsWPHRiPOiKGbAZrCDGBAcBJ5FqxE3ncgFkLJ10JOZu+9917NmTNH06ZN080336xt27ape/fuWrZsmR555BGtXbvWUvDRRjILAAAQ36LaZ/Zvf/ubHnzwQV133XVyH3C7aeDAgdq6dWvo0QIAAABhCjmZLSkp0THHHFOvPDU1VXv37o1IUAAAAEAwQk5mu3Xrps2bN9crf/XVV9W7d+9IxAQAAAAEJTnUFaZPn64rrrhCP/30k4wx+ve//62nnnpKCxYs0EMPPRSNGAEAAIAGhZzMXnzxxWrZsqWuv/56VVVV6bzzzlPnzp21ZMkSnXPOOdGIEQAAAGhQSMnsL7/8oieffFKjR4/WhAkTVFVVpcrKSnXo0CFa8QEAAAABhZTMJicn69JLL9VHH30kSUpPT1d6enpUAoO945Va3baTx2BknFiEinMmPFavE05udyuxO3ksbCcfM8QxE6Jhw4aZ559/PtTV4kZFRYWRZCoqKuwOpVHPPWdMVpYx0q+PrCxvebxve8YMY9xu//Xdbm95vLOz3eFMnDPhsXqdcHK7W4nd6n47+bMFiSWUfC3kZPaZZ54x3bt3N3/729/Mhg0bzJYtW/we8c4Jyexzzxnjcvm/4SVvmcsV3Te+1W3PmFF/3QMf8ZzQ2tnucCbOmfBYvU44ud2txG51v5382YLEE0q+FvIMYElJ9UfzcrlcMsbI5XLJ4/FE5I5xtMT7DGAej5STI+3c2fByl0vKypJKSiL/1YzVbVdXS+np3tcJxO2Wqqrir8uBne0OZ+KcCY/V64ST291K7Fb328mfLUhMUZ0BrKSkpN7js88+8/0La4qKAr/hJe/fsqWl3nrxtu2CgsY/oCTv8oKC8GOMFjvbHc7EORMeq9cJJ7e7ldit7reTP1uApoQ8NFfXrl2jEQf+v/LyyNaL5baLi4NbP9h6sWRnu8OZOGfCY/U64eR2txK71f128mcL0JSQk9lHH3200eUTJ04MOxh4f90ZyXqx3HZubnDrB1svluxsdzgT50x4rF4nnNzuVmK3ut9O/mwBmhJyn9l27dr5Pf/5559VVVWllJQUpaen67///W9EA4w0p/SZLSvzfvVSVyz6NYW77ebQZ9aOdoczcc6EJ1J9Zp3Y7lZit7rfTv5sQWKKap/ZH374we9RWVmp7du366STTtJTTz0VdtDwcrulJUu8/3e5/JfVPl+8ODpveKvbTkmRpk9vfBvTp8dfIivZ2+5wJs6Z8Fi9Tji53a3EbnW/nfzZAjQpUkMobNq0yfTs2TNSLxc1Thiay5iGx+PLzrZvLMBQtt3cxpmNVbvDmThnwhONcWad0u5WYre6307+bEFiierQXIFs3rxZQ4cO1e7duyPxclET790MDuTkWVqYAQyJhHMmPMwAxgxgTjpmiK1Q8rWQk9m///3vfs+NMSovL9fdd9+t7OxsvfLKK6FHHENOSmYBAAASUSj5WsijGYwbN87vucvl0qGHHqqTTz5Zt99+e6gvBwAAAIQt5GS2pqYmGnEAAAAAIQt5NIP58+erqqqqXvm+ffs0f/78iAQFAAAABCPkPrNut1vl5eXq0KGDX/n333+vDh06yNPUPIU2o88sAABAfIvqOLPGGLnqDhQnacuWLWrfvn2oLwcAAACELeg+s+3atZPL5ZLL5dIRRxzhl9B6PB5VVlbq0ksvjUqQiD27h0+xsv19+6QZM6QdO6QePaRFi6SWLaMbL4DQ2X2dscLO6wzXOMBf0N0MHnnkERljdOGFF2rx4sXKyMjwLUtJSVFOTo4GDx4ctUAjhW4GTSsslKZOlXbu/LUsK8s7g0t+fnxvf9w4aeXK+uVjx0ovvBDJKAFYYfd1xgo7rzNc45AoojrO7JtvvqkhQ4aoRYsWloK0C8ls4woLpfHj68+fXXsjfsWK6H7QWNl+oIt8LS72QHyw+zpjhZ3XGa5xSCRRTWYP9NNPP6m6utqvLN4TRJLZwDweKSfH/07JgVwu752TkpLofBVoZfv79knp6U1vo6qKr+MAO9l9nbHCzusM1zgkmqj+AKyqqkpXXnmlOnTooFatWqldu3Z+DzhXUVHgDxjJexeltNRbL962P2NGcNsIth6A6LD7OmOFndcZrnFAYCEnszNmzNAbb7yhe++9V6mpqXrooYc0b948de7cWY8++mg0YkSMlJdHtl4st79jR3DrBlsPQHTYfZ2xws7rDNc4ILCQk9l//OMfKigo0O9//3slJycrLy9P119/vW655RY98cQT0YgRMZKZGdl6sdx+jx7BrRtsPQDRYfd1xgo7rzNc44DAQu4z27p1a3344Yc67LDDlJWVpcLCQh133HEqKSlRv379VFlZGa1YI4I+s4HV9mUrK6v/wwwpdn1mw9k+/ckAZ7D7OmMFfWaB2Ilqn9nu3burpKREktSrVy89++yzkrx3bA866KDQo0XccLu9w+JIv/6quFbt88WLo/cBY2X7LVt6f8nbmLFjucgDdrP7OmOFndcZrnFAYCEns1OmTNGWLVskSbNmzdI999yjtLQ0XXPNNZpBz3PHy8/3DovTpYt/eVZWbIbLsbL9F14IfLFnyBogfth9nbHCzusM1zigYZaG5pKkL774Qu+++64OP/xwHXXUUZGKK2roZhAcu2fmYQYwoPmz+zpjBTOAAdEV03Fm09LSwl3dFiSzAAAA8S2qfWY9Ho9uvPFGdenSRa1bt9Znn30mSfrf//1f/d///V94EQMAAABhCDmZvfnmm7Vs2TItXLhQKSkpvvK+ffvqoYceimhwAAAAQGNCTmYfffRRPfDAA5owYYLcB3RuOvroo/Xxxx9HNDgAAACgMSEns2VlZTr88MPrldfU1Ojnn3+OSFAAAABAMEJOZo888kgVNTBp9ooVK3TMMcdEJCgAAAAgGMmhrjBnzhxNmjRJZWVlqqmpUWFhobZv365HH31UL774YjRiBAAAABoU1tBcRUVFmj9/vrZs2aLKykode+yxmjNnjk499dRoxBhRsRyay+oYik4eg9Gq6mqpoEAqLpZyc6XLL5cO+L1ho2j38Nm571aOueTc2K3GbbXdnMzKeKt2t7uTrzNWYnfyflvFZ1NoQsrXTJCKi4tNTU1NsNWDUlBQYPr162fatGlj2rRpY0444QTz8ssvN7rOs88+a3r27GlSU1NN3759zUsvvRTSNisqKowkU1FRYSX0Jj33nDFZWcZ4Zx/3PrKyvOWxWN/JZswwxu3233e321veFNo9fHbuu5VjboxzY7cat9V2c7KxY/33u/YxdmzT69rd7k6+zliJ3cn7bRWfTaELJV8LOplNSkoyX3/9te/5WWedZXbt2hVehP/f3//+d/PSSy+ZTz75xGzfvt385S9/MS1atDDbtm1rsP769euN2+02CxcuNB9++KG5/vrrTYsWLczWrVuD3mYsktnnnjPG5ap/kXW5vI+mTj6r6zvZjBkNf0DVPhr7sKDdw2fnvls55k6O3WrcVtvNyQIlssEktHa3u5OvM1Zid/J+W8VnU3iiksy6XC6/ZLZ169amuLg4vAgb0a5dO/PQQw81uOyss84yY8aM8Ss7/vjjzSWXXBL060c7mf3ll/p/PdU9+bKzvfWisb6T7d9f/25H3Yfb7a1XF+0ePjv33coxd3LsVuO22m5OVlXV+H7XPqqq6q9rd7s7+TpjJXYn77dVfDaFL5R8LeTRDKLF4/Ho6aef1t69ezV48OAG62zcuFEjR470Kxs9erQ2btwY8HX379+v3bt3+z2iqahI2rkz8HJjpNJSb71orO9kBQXePkGN8Xi89eqi3cNn575bOeaSc2O3GrfVdnOyGTPCr2d3uzv5OmMldifvt1V8NsVG0Mmsy+WSy+WqV2bV1q1b1bp1a6WmpurSSy/V888/ryOPPLLBurt27VLHjh39yjp27Khdu3YFfP0FCxYoIyPD98jOzrYcc2PKy63Vs7q+kxUXh1+Pdg+fnftu5ZhLzo3datxW283JduwIv57d7e7k64yV2J2831bx2RQbQQ/NZYzR5MmTlZqaKkn66aefdOmll6pVq1Z+9QoLC0MKoGfPntq8ebMqKiq0YsUKTZo0SW+++WbAhDZUs2fP1vTp033Pd+/eHdWENjPTWj2r6ztZbm749Wj38Nm571aOueTc2K3GbbXdnKxHD2nVquDq1WV3uzv5OmMldifvt1V8NsVG0ENzTZkyJagXXLp0qaWARo4cqdzcXN1///31lh122GGaPn26pk2b5iubO3euXnjhBW3ZsiWo14/20Fwej5STI5WVeW//1+VySVlZUklJw0NqWF3fyaqrpfT0xr/Gc7ulqqr6Q+DQ7uGzc9+tHHPJubFbjdtquznZvn3efW9KVVX9YbrsbncnX2esxO7k/baKz6bwRWVorlgZMWKEmTRpUoPLzjrrLHPGGWf4lQ0ePDiufgBmzK+/PKz768NQf7kY7vpOFolfh9PuobNz3yP163CnxW41bkYzCPwIZjQDu9rdydcZK7E7eb+t4rMpPFEZzSAaZs2aZd58801TUlJi/vOf/5hZs2YZl8tlVq1aZYwx5oILLjCzZs3y1V+/fr1JTk42t912m/noo4/M3Llz43JoLmMaHhMuO9vamHKhrO9kkR63k3YPjp37Ho1xO50Qu9W4GWc2tES2lt3t7uTrjJXYnbzfVvHZFLpQ8rWwZgCLlIsuukhr1qxReXm5MjIydNRRR+naa6/VqFGjJEnDhw9XTk6Oli1b5ltn+fLluv766/X555+rR48eWrhwoU4//fSgt8kMYM7ADGD2cOosWpJzY7d7JionYwYwezADWHj4bApNKPmarcmsHWKZzAIAACB0oeRrcTPOLAAAABAqklkAAAA4FsksAAAAHItkFgAAAI5FMgsAAADHCno6W8Reog3DgcSWqMPW2L3fdg6zZOcxc+r5kug4bmhQlMe8jTuxmjTBqoYGSM7Kat4DJCNxWT3fnfp+sXu/raxvd+xWOPV8SXQct8TimBnA7OCEZLZ26rq6M9s096nrkJisnu9Ofb/Yvd9W1rc7diucer4kOo5b4nHMDGB2iPdJEzweKSdH2rmz4eUul5SVJZWU8NUKnM/q+e7U94vd+21lfbtjt8Kp50ui47glJiZNcLCiosBvWMn7t2hpqbce4HRWz3envl/s3m8r69sduxVOPV8SHccNTSGZjTPl5ZGtB8Qzq+e7U98vdu+3lfXtjt0Kp54viY7jhqaQzMaZzMzI1gPimdXz3anvF7v328r6dsduhVPPl0THcUNT6DMbZ2r7BpWVeb86qYu+QWhOrJ7vTn2/2L3fVta3O3YrnHq+JDqOW2Kiz6yDud3SkiXe/7tc/stqny9ezBsWzYPV892p7xe799vK+nbHboVTz5dEx3FDk6I8skLcccLQXMY0PJ5edjbDj6B5snq+O/X9Yvd+W1nf7titcOr5kug4bomFobkaEe/dDA7ETCdIJE6eTcoKu/ebGcCcdb4kOo5b4gglXyOZBQAAQFyhzywAAAASAsksAAAAHItkFgAAAI5FMgsAAADHIpkFAACAY5HMAgAAwLGS7Q4AACTr40dWV0sFBVJxsZSbK11+uZSSEptt28nKfkv2jjNrZ+x2c3LsTubkdndy7FEX5Qkc4o5TZgADEklDM/tkZQU/s8+MGca43f7ru93e8mhv205W9tsYa/tu5zGLxPbt5OTYnczJ7e7k2MMVSr5GMgvAVs89Z4zL5X+RlrxlLlfTF+sZM+qve+CjseTI6rbtZGW/jbG273Yes0hs305Ojt3JnNzuTo7dCqazbQQzgAHxw+ORcnKknTsbXu5ySVlZUklJw1+nVVdL6ene1wnE7Zaqqup/fW1123ayst+StX2385hFYvt2cnLsTubkdndy7FYxAxgARygqCnyRlrz3H0pLvfUaUlDQeFIkeZcXFER+23ayst+StX2385hFYvt2cnLsTubkdndy7LFEMgvANuXl1uoVFwe3fkP1rG7bTlb2W7K273Yes0hs305Ojt3JnNzuTo49lkhmAdgmM9Navdzc4NZvqJ7VbdvJyn5L1vbdzmMWie3bycmxO5mT293JsccSfWYB2Ka2P1hZmffrsrpi0Wc23G3bKVL9TsPZdzuPWSS2bycnx+5kTm53J8duFX1mATiC2y0tWeL9v8vlv6z2+eLFgS/SKSnS9OmNb2P69IaTIqvbtpOV/Zas7budxywS27eTk2N3Mie3u5Njj6koj6wQdxiaC4g/DY2hmJ1t3zizoWzbTtEYqzXYfbfzmEVi+3ZycuxO5uR2d3Ls4WJorkbQzQCIT8wAFh5mAHPmcXNy7E7m5HZ3cuzhCCVfI5kFAABAXKHPLAAAABICySwAAAAci2QWAAAAjkUyCwAAAMcimQUAAIBjJdsdAIBfJdrQKweyc98rK6ULLvh1iKjHHpNat47Ntp18zK3G7uR9BxA/SGaBOFFYKE2dKu3c+WtZVpZ39pf8fPviigU79/2446RNm359vnWr1KaNNGiQ9O9/R3fbTj7mVmN38r4DiC+MMwvEgcJCafz4+nNv105XuGJF8/2At3Pf6yaydUUzoXXyMbcau5P3HUBsMGlCI0hmEW88Hiknx/8O1YFcLu8dq5KS5vcVrJ37XlnpvQPblD17It/lwMnH3GrsTt53ALHDpAmAgxQVBf5gl7x3r0pLvfWaGzv3/YILIlsvFE4+5lZjd/K+A4hPJLOAzcrLI1vPSezc9+LiyNYLhZOPudXYnbzvAOITySxgs8zMyNZzEjv3PTc3svVC4eRjbjV2J+87gPhEn1nAZrV9CMvK6v8gRmrefQjt3Pd46DPrxGNuNXYn7zuA2KHPLOAgbrd3OCLp119z16p9vnhx8/xgt3PfW7f2jlbQmEGDojPerJOPudXYnbzvAOITySwQB/LzvcMRdeniX56V1fyHKbJz3//978AJbbTHmXXyMbcau5P3HUD8oZsBEEcSeUYkZgBz3jFnBjAA0cI4s40gmQUAAIhv9JkFAABAQiCZBQAAgGORzAIAAMCxSGYBAADgWCSzAAAAcCySWQAAADhWst0BAGge7B4z1Mr27Y7dCifHDiB4vNcDI5kFYFlhoTR1qrRz569lWVneaUtjMZuTle3bHbsVTo4dQPB4rzeOSRMAWFJYKI0fL9W9krhc3n+jPT2ple3bHbsVTo4dQPAS9b3ODGCNIJkFIsfjkXJy/O8WHMjl8t49KCmJztdhVrZvd+xWODl2AMFL5Pc6M4ABiImiosAXWcl7J6G01Fsv3rZvd+xWODl2AMHjvR4cklkAYSsvj2y9WG7f7titcHLsAILHez04JLMAwpaZGdl6sdy+3bFb4eTYAQSP93pw6DMLIGy1/bnKyur/OEGKXZ/ZcLZvd+xWODl2AMFL5Pc6fWYBxITb7R0aRvr1l7W1ap8vXhy9i6yV7dsduxVOjh1A8HivB4dkFoAl+fneoWG6dPEvz8qKzZAxVrZvd+xWODl2AMHjvd40uhkAiAi7Z6dhBjDnxQ4geIn2Xmec2UaQzAIAAMQ3+swCAAAgIZDMAgAAwLFIZgEAAOBYJLMAAABwLJJZAAAAOFay3QEAkWb38CV2b98uVvfb7vURuupqqaBAKi6WcnOlyy+XUlLsjio2ON+AOGJsdMstt5iBAwea1q1bm0MPPdSMHTvWfPzxx42us3TpUiPJ75Gamhr0NisqKowkU1FRYTV8xKHnnjMmK8sY78R/3kdWlrc8EbZvF6v7bff6CN2MGca43f5t7nZ7y5s7zjcg+kLJ12xNZkePHm2WLl1qtm3bZjZv3mxOP/10c9hhh5nKysqA6yxdutS0bdvWlJeX+x67du0Kepsks83Xc88Z43L5f8BI3jKXK/ofNHZv3y5W99vu9RG6GTPqt/eBj+ac0HK+AbERSr4WV5MmfPvtt+rQoYPefPNNDR06tME6y5Yt07Rp0/Tjjz+GtQ0mTWiePB4pJ0faubPh5S6Xd+q/kpLofBVo9/btYnW/7V4foauultLTvW0fiNstVVU1vy4HnG9A7Dh20oSKigpJUvv27RutV1lZqa5duyo7O1tjx47VBx98ELDu/v37tXv3br8Hmp+iosAfMJL33klpqbdec9y+Xazut93rI3QFBY0nspJ3eUFBbOKJJc43ID7FTTJbU1OjadOm6cQTT1Tfvn0D1uvZs6cefvhhrVy5Uo8//rhqamo0ZMgQ7QxwhVmwYIEyMjJ8j+zs7GjtAmxUXh7Zek7bvl2s7rfd6yN0xcWRrecknG9AfIqbZPaKK67Qtm3b9PTTTzdab/DgwZo4caL69++vYcOGqbCwUIceeqjuv//+BuvPnj1bFRUVvkdpaWk0wofNMjMjW89p27eL1f22e32ELjc3svWchPMNiE9x0Wf2yiuv1MqVK/XWW2+pW7duIa9/5plnKjk5WU899VSTdekz2zzV9mUrK/N+1VdXrPrM2rV9u1jdb7vXR+joM8v5BsSCY/rMGmN05ZVX6vnnn9cbb7wRViLr8Xi0detWZfKncEJzu6UlS7z/d7n8l9U+X7w4eh8wdm/fLlb32+71EbqUFGn69MbrTJ/e/BJZifMNiFvRHVihcZdddpnJyMgw69at8xtqq6qqylfnggsuMLNmzfI9nzdvnnnttddMcXGxeffdd80555xj0tLSzAcffBDUNhmaq3lraPzH7Gx7x5mN5fbtYnW/7V4foWOcWc43IJocMzSXq+6ftv/f0qVLNXnyZEnS8OHDlZOTo2XLlkmSrrnmGhUWFmrXrl1q166dBgwYoJtuuknHHHNMUNukm0HzZ/fMPHZv3y52z+CVqO1uJ2YA43wDoiWUfC0u+szGEsksAABAfHNMn1kAAADACpJZAAAAOBbJLAAAAByLZBYAAACORTILAAAAxyKZBQAAgGMl2x0AEGmM/2iPRB5zFABgH5JZNCuFhdLUqdLOnb+WZWV5p6DMz7cvruZu5kzpjju8f0jU+vOfvdOaLlxoX1wAgOaPbgZoNgoLpfHj/RNZSSor85YXFtoTV3M3c6a0aJF/Iit5ny9a5F0OAEC0MAMYmgWPR8rJqZ/I1nK5vHdoS0rochBJ1dVSenr9RPZAbrdUVUWXAwBA8JgBDAmnqChwIitJxkilpd56iJyCgsYTWcm7vKAgNvEAABIPySyahfLyyNZDcIqLI1sPAIBQkcyiWcjMjGw9BCc3N7L1AAAIFX1m0SzU9pktK/N2KaiLPrPRQZ9ZAEA00GcWCcft9g6/JXkT1wPVPl+8mEQ20lJSvMNvNWb6dBJZAED0kMyi2cjPl1askLp08S/PyvKWM85sdCxcKM2YUf8PBbfbW844swCAaKKbAZodZgCzBzOAAQAiJZR8jWQWAAAAcYU+swAAAEgIJLMAAABwLJJZAAAAOBbJLAAAAByLZBYAAACOlWx3AIgehqgKD+0WHtot8XDMAcQDktlmqrBQmjpV2rnz17KsLO8sWUweEBjtFh7aLfFwzAHEC8aZbYYKC6Xx46W6R7Z2Wldmw2oY7RYe2i3xcMwBRBuTJjSiuSezHo+Uk+N/t+RALpf37klJCV8HHoh2Cw/tlng45gBigUkTElhRUeAPGcl7J6W01FsPv6LdwkO7JR6OOYB4QzLbzJSXR7ZeoqDdwkO7JR6OOYB4QzLbzGRmRrZeoqDdwkO7JR6OOYB4Q5/ZZqa2P1tZWf0fZ0j0ZwuEdgsP7ZZ4OOYAYoE+swnM7fYOjSP9+sviWrXPFy/mQ6Yu2i08tFvi4ZgDiDcks81Qfr53aJwuXfzLs7IYMqcxtFt4aLfEwzEHEE/oZtCMMTtPeGi38NBuiYdjDiBaGGe2EYmUzAIAADgRfWYBAACQEEhmAQAA4FgkswAAAHAsklkAAAA4FsksAAAAHItkFgAAAI6VbHcAAJDIqqulggKpuFjKzZUuv1xKSYnd9hkrFoDTkcwCgE1mzpTuuMObUNb685+l6dOlhQujv/3CQmnqVGnnzl/LsrK809UyixcAp6CbAQDYYOZMadEi/0RW8j5ftMi7PJoKC6Xx4/0TWUkqK/OWFxZGd/sAECnMAAYAMVZdLaWn109kD+R2S1VV0ely4PFIOTn1E9laLpf3Dm1JCV0OANiDGcAAII4VFDSeyEre5QUF0dl+UVHgRFaSjJFKS731ACDekcwCQIwVF0e2XqjKyyNbDwDsRDILADGWmxvZeqHKzIxsPQCwE31mASDG4qXPbFmZt0tBXfSZBWA3+swCQBxLSfEOv9WY6dOjN96s2+0dfkvyJq4Hqn2+eDGJLABnIJkFABssXCjNmFE/YXS7veXRHmc2P19asULq0sW/PCvLW844swCcgm4GAGAjZgADgPpCyddIZgEAABBX6DMLAACAhEAyCwAAAMcimQUAAIBjkcwCAADAsUhmAQAA4FgkswAAAHAsklkAAAA4FsksAAAAHItkFgAAAI5FMgsAAADHIpkFAACAY5HMAgAAwLFIZgEAAOBYJLMAAABwLJJZAAAAOBbJLAAAAByLZBYAAACORTILAAAAxyKZBQAAgGORzAIAAMCxSGYBAADgWCSzAAAAcCySWQAAADhWst0BAPHG45GKiqTycikzU8rLk9xuu6MCAAANsfXO7IIFCzRo0CC1adNGHTp00Lhx47R9+/Ym11u+fLl69eqltLQ09evXTy+//HIMokUiKCyUcnKkESOk887z/puT4y0HAADxx9Zk9s0339QVV1yhf/7zn3r99df1888/69RTT9XevXsDrrNhwwade+65uuiii/T+++9r3LhxGjdunLZt2xbDyNEcFRZK48dLO3f6l5eVectJaAEAiD8uY4yxO4ha3377rTp06KA333xTQ4cObbDO2Wefrb179+rFF1/0lZ1wwgnq37+/7rvvvia3sXv3bmVkZKiiokJt27aNWOxwNo/Hewe2biJby+WSsrKkkhK6HAAAEG2h5Gtx9QOwiooKSVL79u0D1tm4caNGjhzpVzZ69Ght3Lixwfr79+/X7t27/R5AXUVFgRNZSTJGKi311gMAAPEjbpLZmpoaTZs2TSeeeKL69u0bsN6uXbvUsWNHv7KOHTtq165dDdZfsGCBMjIyfI/s7OyIxo3mobw8svUAAEBsxE0ye8UVV2jbtm16+umnI/q6s2fPVkVFhe9RWloa0ddH85CZGdl6AAAgNuJiaK4rr7xSL774ot566y1lZWU1WrdTp076+uuv/cq+/vprderUqcH6qampSk1NjVisaJ7y8rx9YsvKvF0K6qrtM5uXF/vYAABAYLbemTXG6Morr9Tzzz+vN954Q926dWtyncGDB2vNmjV+Za+//roGDx4crTCRANxuackS7/9dLv9ltc8XL+bHXwAAxBtbk9krrrhCjz/+uJ588km1adNGu3bt0q5du7Rv3z5fnYkTJ2r27Nm+51OnTtWrr76q22+/XR9//LFuuOEGvfPOO7ryyivt2AU0I/n50ooVUpcu/uVZWd7y/Hx74gIAAIHZOjSXq+4tsP9v6dKlmjx5siRp+PDhysnJ0bJly3zLly9fruuvv16ff/65evTooYULF+r0008PapsMzYWmMAMYAAD2CiVfi6txZmOBZBYAACC+OXacWQAAACAUJLMAAABwLJJZAAAAOBbJLAAAAByLZBYAAACORTILAAAAxyKZBQAAgGORzAIAAMCxSGYBAADgWCSzAAAAcCySWQAAADgWySwAAAAci2QWAAAAjpVsdwCxZoyRJO3evdvmSAAAANCQ2jytNm9rTMIls3v27JEkZWdn2xwJAAAAGrNnzx5lZGQ0Wsdlgkl5m5Gamhp99dVXatOmjVwul93hRN3u3buVnZ2t0tJStW3b1u5wHIN2Cw/tFh7aLXy0XXhot/DQbuELte2MMdqzZ486d+6spKTGe8Um3J3ZpKQkZWVl2R1GzLVt25Y3Xhhot/DQbuGh3cJH24WHdgsP7Ra+UNquqTuytfgBGAAAAByLZBYAAACORTLbzKWmpmru3LlKTU21OxRHod3CQ7uFh3YLH20XHtotPLRb+KLZdgn3AzAAAAA0H9yZBQAAgGORzAIAAMCxSGYBAADgWCSzAAAAcCyS2Wbi1ltvlcvl0rRp0wLWWbZsmVwul98jLS0tdkHGiRtuuKFeO/Tq1avRdZYvX65evXopLS1N/fr108svvxyjaONHqO3G+farsrIynX/++Tr44IPVsmVL9evXT++8806j66xbt07HHnusUlNTdfjhh2vZsmWxCTbOhNp269atq3feuVwu7dq1K4ZR2ysnJ6fBNrjiiisCrsM1zivUtuM65+XxePS///u/6tatm1q2bKnc3FzdeOONamqMgUhd5xJuBrDmaNOmTbr//vt11FFHNVm3bdu22r59u+95Ikzp25A+ffpo9erVvufJyYHfChs2bNC5556rBQsW6IwzztCTTz6pcePG6b333lPfvn1jEW7cCKXdJM43Sfrhhx904oknasSIEXrllVd06KGHaseOHWrXrl3AdUpKSjRmzBhdeumleuKJJ7RmzRpdfPHFyszM1OjRo2MYvb3Cabta27dv95tlqEOHDtEMNa5s2rRJHo/H93zbtm0aNWqUzjzzzAbrc437VahtJ3Gdk6S//vWvuvfee/XII4+oT58+eueddzRlyhRlZGTo6quvbnCdiF7nDBxtz549pkePHub11183w4YNM1OnTg1Yd+nSpSYjIyNmscWruXPnmqOPPjro+meddZYZM2aMX9nxxx9vLrnkkghHFt9CbTfON69rr73WnHTSSSGtM3PmTNOnTx+/srPPPtuMHj06kqHFvXDabu3atUaS+eGHH6ITlANNnTrV5ObmmpqamgaXc40LrKm24zrnNWbMGHPhhRf6leXn55sJEyYEXCeS1zm6GTjcFVdcoTFjxmjkyJFB1a+srFTXrl2VnZ2tsWPH6oMPPohyhPFpx44d6ty5s7p3764JEyboyy+/DFh348aN9dp39OjR2rhxY7TDjDuhtJvE+SZJf//73zVw4ECdeeaZ6tChg4455hg9+OCDja7DOecVTtvV6t+/vzIzMzVq1CitX78+ypHGr+rqaj3++OO68MILA94x5HxrWDBtJ3Gdk6QhQ4ZozZo1+uSTTyRJW7Zs0dtvv63TTjst4DqRPO9IZh3s6aef1nvvvacFCxYEVb9nz556+OGHtXLlSj3++OOqqanRkCFDtHPnzihHGl+OP/54LVu2TK+++qruvfdelZSUKC8vT3v27Gmw/q5du9SxY0e/so4dOyZUHzwp9HbjfPP67LPPdO+996pHjx567bXXdNlll+nqq6/WI488EnCdQOfc7t27tW/fvmiHHDfCabvMzEzdd999eu655/Tcc88pOztbw4cP13vvvRfDyOPHCy+8oB9//FGTJ08OWIdrXMOCaTuuc16zZs3SOeeco169eqlFixY65phjNG3aNE2YMCHgOhG9zoV8Lxdx4csvvzQdOnQwW7Zs8ZU11c2grurqapObm2uuv/76KEToHD/88INp27ateeihhxpc3qJFC/Pkk0/6ld1zzz2mQ4cOsQgvbjXVbnUl6vnWokULM3jwYL+yq666ypxwwgkB1+nRo4e55ZZb/MpeeuklI8lUVVVFJc54FE7bNWTo0KHm/PPPj2RojnHqqaeaM844o9E6XOMaFkzb1ZWo17mnnnrKZGVlmaeeesr85z//MY8++qhp3769WbZsWcB1Inmd486sQ7377rv65ptvdOyxxyo5OVnJycl68803dddddyk5OdmvA3sgtX89ffrppzGIOH4ddNBBOuKIIwK2Q6dOnfT111/7lX399dfq1KlTLMKLW021W12Jer5lZmbqyCOP9Cvr3bt3o100Ap1zbdu2VcuWLaMSZzwKp+0actxxxyXceSdJX3zxhVavXq2LL7640Xpc4+oLtu3qStTr3IwZM3x3Z/v166cLLrhA11xzTaPfHEfyOkcy61CnnHKKtm7dqs2bN/seAwcO1IQJE7R582a53e4mX8Pj8Wjr1q3KzMyMQcTxq7KyUsXFxQHbYfDgwVqzZo1f2euvv67BgwfHIry41VS71ZWo59uJJ57o90tnSfrkk0/UtWvXgOtwznmF03YN2bx5c8Kdd5K0dOlSdejQQWPGjGm0HudbfcG2XV2Jep2rqqpSUpJ/Sul2u1VTUxNwnYied2HdT0ZcqtvN4IILLjCzZs3yPZ83b5557bXXTHFxsXn33XfNOeecY9LS0swHH3xgQ7T2+dOf/mTWrVtnSkpKzPr1683IkSPNIYccYr755htjTP12W79+vUlOTja33Xab+eijj8zcuXNNixYtzNatW+3aBVuE2m6cb17//ve/TXJysrn55pvNjh07zBNPPGHS09PN448/7qsza9Ysc8EFF/ief/bZZyY9Pd3MmDHDfPTRR+aee+4xbrfbvPrqq3bsgm3Cabs777zTvPDCC2bHjh1m69atZurUqSYpKcmsXr3ajl2wjcfjMYcddpi59tpr6y3jGte4UNqO65zXpEmTTJcuXcyLL75oSkpKTGFhoTnkkEPMzJkzfXWieZ0jmW1G6iazw4YNM5MmTfI9nzZtmjnssMNMSkqK6dixozn99NPNe++9F/tAbXb22WebzMxMk5KSYrp06WLOPvts8+mnn/qW1203Y4x59tlnzRFHHGFSUlJMnz59zEsvvRTjqO0Xartxvv3qH//4h+nbt69JTU01vXr1Mg888IDf8kmTJplhw4b5la1du9b079/fpKSkmO7du5ulS5fGLuA4Emrb/fWvfzW5ubkmLS3NtG/f3gwfPty88cYbMY7afq+99pqRZLZv315vGde4xoXSdlznvHbv3m2mTp1qDjvsMJOWlma6d+9urrvuOrN//35fnWhe51zGNDE9AwAAABCn6DMLAAAAxyKZBQAAgGORzAIAAMCxSGYBAADgWCSzAAAAcCySWQAAADgWySwAAAAci2QWAAAAjkUyCwAAAMcimQWAEE2ePFkul6ve49NPP43I6y9btkwHHXRQRF4rHB6PR0OGDFF+fr5feUVFhbKzs3Xdddf5yq6++moNGDBAqamp6t+/f4wjBQCSWQAIy29+8xuVl5f7Pbp162Z3WPX8/PPPIa/jdru1bNkyvfrqq3riiSd85VdddZXat2+vuXPn+tW/8MILdfbZZ1uOFQDCQTILAGFITU1Vp06d/B5ut1uStHLlSh177LFKS0tT9+7dNW/ePP3yyy++de+44w7169dPrVq1UnZ2ti6//HJVVlZKktatW6cpU6aooqLCd8f3hhtukCS5XC698MILfnEcdNBBWrZsmSTp888/l8vl0jPPPKNhw4YpLS3Nl4w+9NBD6t27t9LS0tSrVy8VFBQ0un9HHHGEbr31Vl111VUqLy/XypUr9fTTT+vRRx9VSkqKr95dd92lK664Qt27d7fSnAAQtmS7AwCA5qSoqEgTJ07UXXfdpby8PBUXF+uPf/yjJPnuaCYlJemuu+5St27d9Nlnn+nyyy/XzJkzVVBQoCFDhmjx4sWaM2eOtm/fLklq3bp1SDHMmjVLt99+u4455hhfQjtnzhzdfffdOuaYY/T+++/rD3/4g1q1aqVJkyYFfJ2rrrpKzz//vC644AJt3bpVc+bM0dFHHx1mywBAlBgAQEgmTZpk3G63adWqle8xfvx4Y4wxp5xyirnlllv86j/22GMmMzMz4OstX77cHHzwwb7nS5cuNRkZGfXqSTLPP/+8X1lGRoZZunSpMcaYkpISI8ksXrzYr05ubq558skn/cpuvPFGM3jw4KZ21Xz00UdGkunXr5/5+eefA9abO3euOfroo5t8PQCINO7MAkAYRowYoXvvvdf3vFWrVpKkLVu2aP369br55pt9yzwej3766SdVVVUpPT1dq1ev1oIFC/Txxx9r9+7d+uWXX/yWWzVw4EDf//fu3avi4mJddNFF+sMf/uAr/+WXX5SRkdHkaz388MNKT09XSUmJdu7cqZycHMvxAUAkkcwCQBhatWqlww8/vF55ZWWl5s2bV28kAElKS0vT559/rjPOOEOXXXaZbr75ZrVv315vv/22LrroIlVXVzeazLpcLhlj/Moa+oFXbWJdG48kPfjggzr++OP96tX28Q1kw4YNuvPOO7Vq1SrddNNNuuiii7R69Wq5XK5G1wOAWCKZBYAIOvbYY7V9+/YGE11Jevfdd1VTU6Pbb79dSUne3+A+++yzfnVSUlLk8XjqrXvooYeqvLzc93zHjh2qqqpqNJ6OHTuqc+fO+uyzzzRhwoSg96OqqkqTJ0/WZZddphEjRqhbt27q16+f7rvvPl122WVBvw4ARBvJLABE0Jw5c3TGGWfosMMO0/jx45WUlKQtW7Zo27Ztuummm3T44Yfr559/1t/+9jf99re/1fr163Xffff5vUZOTo4qKyu1Zs0aHX300UpPT1d6erpOPvlk3X333Ro8eLA8Ho+uvfZatWjRosmY5s2bp6uvvloZGRn6zW9+o/379+udd97RDz/8oOnTpze4zuzZs2WM0a233uqL6bbbbtOf//xnnXbaab7uBp9++qkqKyu1a9cu7du3T5s3b5YkHXnkkX6jHgBA1NjdaRcAnGbSpElm7NixAZe/+uqrZsiQIaZly5ambdu25rjjjjMPPPCAb/kdd9xhMjMzTcuWLc3o0aPNo48+aiSZH374wVfn0ksvNQcffLCRZObOnWuMMaasrMyceuqpplWrVqZHjx7m5ZdfbvAHYO+//369mJ544gnTv39/k5KSYtq1a2eGDh1qCgsLG4x/3bp1xu12m6KionrLTj31VHPyySebmpoaY4wxw4YNM5LqPUpKShptQwCIFJcxdTpgAQAAAA7BpAkAAABwLJJZAAAAOBbJLAAAAByLZBYAAACORTILAAAAxyKZBQAAgGORzAIAAMCxSGYBAADgWCSzAAAAcCySWQAAADgWySwAAAAc6/8BezUIL64flGMAAAAASUVORK5CYII=\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "#testing euclidean distance\n",
        "sum((df.iloc[0].values - df.iloc[1].values) ** 2)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "DopyFUzW5nLD",
        "outputId": "e9d467a3-f2ac-41d7-a833-baaac8fe66f1"
      },
      "execution_count": 25,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.5399999999999999"
            ]
          },
          "metadata": {},
          "execution_count": 25
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "\n",
        "def euclidean_distance(x1, x2):\n",
        "    return np.sqrt(np.sum((x1 - x2) ** 2))\n",
        "\n",
        "euclidean_distance(df.iloc[0].values,df.iloc[1].values)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "n44Dh0Q9vQ6d",
        "outputId": "991a6a69-871a-42e3-e627-4c28d9f5e85f"
      },
      "execution_count": 19,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.7348469228349533"
            ]
          },
          "metadata": {},
          "execution_count": 19
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "\n",
        "class KMeans:\n",
        "\n",
        "    def __init__(self, in_clusters=2, seed=23, kmeans_plus_plus = False):\n",
        "        self.in_clusters = in_clusters\n",
        "        self.seed = seed\n",
        "        self.kmeans_plus_plus = kmeans_plus_plus\n",
        "        np.random.seed(seed)\n",
        "\n",
        "\n",
        "    def euclidean_distance(x1, x2):\n",
        "        return np.sqrt(np.sum((x1 - x2) ** 2, axis = 1))\n",
        "\n",
        "    def initialize_clusters(self, X):\n",
        "        centroids_ids = np.random.choice(X.shape[0], size=self.in_clusters, replace=False)\n",
        "        centroids = []\n",
        "        for i in centroids_ids:\n",
        "          centroids.append(X.iloc[i].values)\n",
        "        return centroids\n",
        "\n",
        "    def intialize_clusters_k_plus_plus(self, X):\n",
        "\n",
        "        centroids = []\n",
        "        #randomly choosing the first centroid\n",
        "        centroid_id = np.random.choice(X.shape[0])\n",
        "        centroids.append(X.iloc[centroid_id].values)\n",
        "        heap = []\n",
        "        for _ in range(1, self.in_clusters):\n",
        "            dist = [0]*X.shape[0]\n",
        "            for i, x in enumerate(X.values):\n",
        "                for c in centroids:\n",
        "                    dist[i] += self.euclidean_distance(x,c)\n",
        "            new_centroid = X.iloc[np.argmax(dist)].values\n",
        "            centroids.append(new_centroid)\n",
        "        return centroids\n",
        "\n",
        "\n",
        "    def calc_distances(self, X, centroids):\n",
        "        labels = np.zeros(len(X), dtype=int)\n",
        "        for i, x in X.iterrows():\n",
        "            min_dist = float('inf')\n",
        "            for j, centroid in enumerate(centroids):\n",
        "                dist = np.sqrt(np.sum((x - centroid)**2))\n",
        "                if dist < min_dist:\n",
        "                    min_dist = dist\n",
        "                    labels[i] = j\n",
        "        return labels\n",
        "\n",
        "    def update_centroids(self, X, labels):\n",
        "        centroids = np.zeros((self.in_clusters, X.shape[1]))\n",
        "        for k in range(self.in_clusters):\n",
        "            cluster_points = X[labels == k]\n",
        "            if len(cluster_points) > 0:\n",
        "                centroids[k] = np.mean(cluster_points, axis=0)\n",
        "        return centroids\n",
        "\n",
        "    def fit(self, X, max_iters=100):\n",
        "        centroids = self.initialize_clusters(X)\n",
        "        for i in range(max_iters):\n",
        "            labels = self.calc_distances(X, centroids)\n",
        "            new_centroids = self.update_centroids(X, labels)\n",
        "            #if i == 99:\n",
        "\n",
        "              #print(\"reached max iteration\")\n",
        "            #else:\n",
        "\n",
        "              #print(\"num of iters \",i)\n",
        "              #print(centroids)\n",
        "              #print(new_centroids)\n",
        "            if np.all(centroids == new_centroids):\n",
        "                break\n",
        "            centroids = new_centroids\n",
        "        return centroids, labels\n",
        "\n",
        "\n",
        "    def elbow_method(self, X, max_clusters=10):\n",
        "        labels_all = {}\n",
        "        cost_function_values = []\n",
        "        for k in range(1, max_clusters + 1):\n",
        "            if self.kmeans_plus_plus:\n",
        "              kmeans = KMeans(in_clusters=k, kmeans_plus_plus=True)\n",
        "              #if k == 10: print(\"done using kmeans++\")\n",
        "            else:\n",
        "              kmeans = KMeans(in_clusters=k)\n",
        "\n",
        "            centroids, labels = kmeans.fit(X)\n",
        "            labels_all[k] = labels\n",
        "            cost_function = 0\n",
        "            for i, x in X.iterrows():\n",
        "                cost_function += np.sqrt(np.sum((x - centroids[labels[i]]) ** 2))\n",
        "            cost_function_values.append(cost_function)\n",
        "        return cost_function_values, labels_all\n",
        "\n",
        "start_time = time.time()\n",
        "kmeans = KMeans()\n",
        "cost_function_values , labels_all = kmeans.elbow_method(df)\n",
        "end_time = time.time()\n",
        "print(\"kmeans time taken: \", end_time - start_time)\n",
        "plt.plot(range(1, len(cost_function_values) + 1), cost_function_values)\n",
        "plt.xlabel('Number of Clusters')\n",
        "plt.ylabel('Cost function')\n",
        "plt.title('Elbow Method')\n",
        "plt.show()\n",
        "\n",
        "#print(centroids)\n",
        "#print(labels)\n",
        "\n",
        "#print(df.shape)\n"
      ],
      "metadata": {
        "id": "5vPGzkjX5oQ4",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 490
        },
        "outputId": "5be42ae6-03ad-4c81-a8ec-37c00e1d8838"
      },
      "execution_count": 72,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "kmeans time taken:  11.797283172607422\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "#kmeans++\n",
        "start_time = time.time()\n",
        "kmeans_kpp = KMeans(kmeans_plus_plus= True)\n",
        "cost_function_values_kpp, labels_all_kpp = kmeans_kpp.elbow_method(df)\n",
        "end_time = time.time()\n",
        "print(\"kmeans++ time taken: \", end_time - start_time)\n",
        "plt.plot(range(1, len(cost_function_values_kpp) + 1), cost_function_values_kpp)\n",
        "plt.xlabel('Number of Clusters')\n",
        "plt.ylabel('Cost function')\n",
        "plt.title('Elbow Method')\n",
        "plt.show()"
      ],
      "metadata": {
        "id": "h82R_mGm5wmT",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 490
        },
        "outputId": "f9d3e8a4-7010-4785-ec85-36dd83eb4699"
      },
      "execution_count": 71,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "kmeans++ time taken:  11.634968280792236\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "#knn\n",
        "\n",
        "import math\n",
        "from collections import Counter\n",
        "\n",
        "import numpy as np\n",
        "\n",
        "\n",
        "def euclidean_distance(x1, x2):\n",
        "    return np.sqrt(np.sum((x1 - x2) ** 2))\n",
        "\n",
        "\n",
        "def manhattan_distance(x1, x2):\n",
        "    return np.sqrt(np.sum(math.abs(x1 - x2)))\n",
        "\n",
        "\n",
        "class KNN:\n",
        "\n",
        "    def __init__(self, k=3):\n",
        "        self.k = k\n",
        "\n",
        "    def fit(self, X, y):\n",
        "        self.X_train = X.values\n",
        "        self.y_train = y.values\n",
        "\n",
        "    def predict(self, X):\n",
        "        print(\"X_train type:\", type(self.X_train))\n",
        "        print(\"X_train shape:\", self.X_train.shape)\n",
        "        print(\"x type:\", type(X))\n",
        "        print(\"x shape:\", X.shape)\n",
        "        predictions = [self._predict(x) for x in X.values]\n",
        "        return predictions\n",
        "\n",
        "    def _predict(self, x):\n",
        "\n",
        "        # compute the distance\n",
        "        # distances = [euclidean_distance(x, x_train) for x_train in self.X_train]\n",
        "        distances = np.sqrt(np.sum((self.X_train - x) ** 2, axis=1))\n",
        "        # get the closest k neighbors\n",
        "        k_ind = np.argsort(distances)[: self.k]\n",
        "        k_labels = [self.y_train[i] for i in k_ind]\n",
        "\n",
        "        # get votes\n",
        "        most_common = Counter(k_labels).most_common(1)\n",
        "        return most_common[0][0]"
      ],
      "metadata": {
        "id": "471ckaja5x9H"
      },
      "execution_count": 30,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "labels_all[3]"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "_mnY4JMl5S2p",
        "outputId": "6b4c3867-dbd6-4645-d4ae-017225890597"
      },
      "execution_count": 44,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([2, 2, 0, 2, 0, 2, 2, 1, 1, 1, 0, 0, 0, 0, 0, 1, 1, 2, 1, 1, 2, 0,\n",
              "       0, 1, 0, 1, 2, 1, 1, 0, 2, 1, 1, 1, 0, 0, 0, 2, 1, 0, 1, 1, 1, 0,\n",
              "       1, 0, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 1, 1, 2, 0, 0, 2, 1, 2,\n",
              "       1, 0, 1, 1, 1, 2, 1, 2, 1, 0, 0, 2, 0, 0, 1, 0, 2, 2, 2, 2, 1, 0,\n",
              "       0, 0, 1, 1, 0, 0, 2, 0, 0, 0, 0, 0, 2, 2, 2, 1, 2, 0, 2, 0, 0, 0,\n",
              "       1, 2, 1, 2, 1, 2, 0, 0, 2, 2, 1, 2, 0, 2, 2, 1, 1, 0, 0, 0, 1, 1,\n",
              "       2, 0, 2, 0, 0, 1, 1, 0, 0, 2, 0, 2, 0, 1, 1, 2, 1, 1])"
            ]
          },
          "metadata": {},
          "execution_count": 44
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "import time\n",
        "\n",
        "import matplotlib.pyplot as plt\n",
        "import numpy as np\n",
        "from matplotlib.colors import ListedColormap\n",
        "from sklearn import datasets\n",
        "from sklearn.model_selection import train_test_split\n",
        "from sklearn.neighbors import KNeighborsClassifier\n",
        "\n",
        "df['labels'] = labels_all[3]\n",
        "y = df['labels']\n",
        "X = df.drop(columns=['labels'])\n",
        "X_train, X_test, y_train, y_test = train_test_split(\n",
        "    X, y, test_size=0.2, random_state=26\n",
        ")\n",
        "#print(X_test.dtypes)\n",
        "#print(y_test.dtype)\n",
        "# custom model time\n",
        "all_accs = []\n",
        "for n in range(1,11):\n",
        "    start_time = time.time()\n",
        "    clf = KNN(k=n)\n",
        "    clf.fit(X_train, y_train)\n",
        "    predictions = clf.predict(X_test)\n",
        "    acc = np.sum(predictions == y_test) / len(y_test)\n",
        "    end_time = time.time()\n",
        "    print(\"knn accuracy: \", acc, \"with time taken: \", end_time - start_time)\n",
        "    all_accs.append(acc)\n",
        "\n",
        "# sklearn\n",
        "\n",
        "start_time = time.time()\n",
        "clf1 = KNeighborsClassifier(n_neighbors=5)\n",
        "# custom model time\n",
        "clf1.fit(X_train, y_train)\n",
        "predictions = clf1.predict(X_test)\n",
        "sk_acc = np.sum(predictions == y_test) / len(y_test)\n",
        "end_time = time.time()\n",
        "# sklearn model time\n",
        "print(\"sklearn knn accuracy\", sk_acc, \"with time taken: \", end_time - start_time)\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "yTifuCUPCl7a",
        "outputId": "6e6fd67b-d070-4ddc-c445-a9a74a3a63b6"
      },
      "execution_count": 49,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "X_train type: <class 'numpy.ndarray'>\n",
            "X_train shape: (120, 4)\n",
            "x type: <class 'pandas.core.frame.DataFrame'>\n",
            "x shape: (30, 4)\n",
            "knn accuracy:  1.0 with time taken:  0.007946968078613281\n",
            "X_train type: <class 'numpy.ndarray'>\n",
            "X_train shape: (120, 4)\n",
            "x type: <class 'pandas.core.frame.DataFrame'>\n",
            "x shape: (30, 4)\n",
            "knn accuracy:  1.0 with time taken:  0.00652313232421875\n",
            "X_train type: <class 'numpy.ndarray'>\n",
            "X_train shape: (120, 4)\n",
            "x type: <class 'pandas.core.frame.DataFrame'>\n",
            "x shape: (30, 4)\n",
            "knn accuracy:  1.0 with time taken:  0.0031337738037109375\n",
            "X_train type: <class 'numpy.ndarray'>\n",
            "X_train shape: (120, 4)\n",
            "x type: <class 'pandas.core.frame.DataFrame'>\n",
            "x shape: (30, 4)\n",
            "knn accuracy:  1.0 with time taken:  0.015740633010864258\n",
            "X_train type: <class 'numpy.ndarray'>\n",
            "X_train shape: (120, 4)\n",
            "x type: <class 'pandas.core.frame.DataFrame'>\n",
            "x shape: (30, 4)\n",
            "knn accuracy:  0.9666666666666667 with time taken:  0.0019969940185546875\n",
            "X_train type: <class 'numpy.ndarray'>\n",
            "X_train shape: (120, 4)\n",
            "x type: <class 'pandas.core.frame.DataFrame'>\n",
            "x shape: (30, 4)\n",
            "knn accuracy:  0.9666666666666667 with time taken:  0.0022034645080566406\n",
            "X_train type: <class 'numpy.ndarray'>\n",
            "X_train shape: (120, 4)\n",
            "x type: <class 'pandas.core.frame.DataFrame'>\n",
            "x shape: (30, 4)\n",
            "knn accuracy:  0.9666666666666667 with time taken:  0.014154195785522461\n",
            "X_train type: <class 'numpy.ndarray'>\n",
            "X_train shape: (120, 4)\n",
            "x type: <class 'pandas.core.frame.DataFrame'>\n",
            "x shape: (30, 4)\n",
            "knn accuracy:  0.9666666666666667 with time taken:  0.003953456878662109\n",
            "X_train type: <class 'numpy.ndarray'>\n",
            "X_train shape: (120, 4)\n",
            "x type: <class 'pandas.core.frame.DataFrame'>\n",
            "x shape: (30, 4)\n",
            "knn accuracy:  0.9666666666666667 with time taken:  0.0024170875549316406\n",
            "X_train type: <class 'numpy.ndarray'>\n",
            "X_train shape: (120, 4)\n",
            "x type: <class 'pandas.core.frame.DataFrame'>\n",
            "x shape: (30, 4)\n",
            "knn accuracy:  0.9666666666666667 with time taken:  0.0023784637451171875\n",
            "sklearn knn accuracy 0.9666666666666667 with time taken:  0.040847063064575195\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "iterations = range(1, 11)  # assuming 10 iterations\n",
        "print(all_accs)\n",
        "# Plot the accuracy curve\n",
        "plt.plot(iterations, all_accs, marker='o')\n",
        "plt.xlabel('Iterations')\n",
        "plt.ylabel('Accuracy')\n",
        "plt.title('Accuracy Curve')\n",
        "plt.grid(True)\n",
        "plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 510
        },
        "id": "bSntNbDZ3ig7",
        "outputId": "40a54156-9ec1-400f-b4ae-af788698c9f2"
      },
      "execution_count": 48,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[1.0, 1.0, 1.0, 1.0, 0.9666666666666667, 0.9666666666666667, 0.9666666666666667, 0.9666666666666667, 0.9666666666666667, 0.9666666666666667]\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "#on cluster data 2\n",
        "print(df_test.shape)\n",
        "\n",
        "start_time = time.time()\n",
        "clf1 = KNeighborsClassifier(n_neighbors=5)\n",
        "\n",
        "clf1.fit(X_train, y_train)\n",
        "predictions = clf1.predict(df_test)\n",
        "print(predictions)\n",
        "end_time = time.time()"
      ],
      "metadata": {
        "id": "rx5WpaIBzsaU",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "ee511838-65a0-4f5f-8b2b-5eb470e4dd2b"
      },
      "execution_count": 50,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "(15, 4)\n",
            "[1 1 0 0 2 0 1 2 1 2 0 2 1 2 0]\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "#extra\n",
        "#using kmeans to label the data of cluster 2 and then using knn to predict which is trained on data1\n",
        "\n",
        "kmeans = KMeans()\n",
        "cost_function_values , labels_all = kmeans.elbow_method(df_test)\n",
        "\n",
        "plt.plot(range(1, len(cost_function_values) + 1), cost_function_values)\n",
        "plt.xlabel('Number of Clusters')\n",
        "plt.ylabel('Cost function')\n",
        "plt.title('Elbow Method')\n",
        "plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 472
        },
        "id": "qO24NNruGOsa",
        "outputId": "dadc05da-e7b3-40dd-97c0-838ff7aa8278"
      },
      "execution_count": 63,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "print(df_test)"
      ],
      "metadata": {
        "id": "x_M85195JOxJ"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "#the optimal k value is 3 too so we use the labels of when k = 3\n",
        "#labels[3]\n",
        "df_test['labels'] = labels_all[3]\n",
        "y = df_test['labels']\n",
        "X = df_test.drop(columns=['labels'])\n",
        "print(y)\n",
        "predictions = clf1.predict(X)\n",
        "print(predictions)\n",
        "print(\"acc: \",np.sum(predictions == y) / len(y))\n"
      ],
      "metadata": {
        "id": "2WO6BT1KIErs"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "#we get accuracy zero all the clusters got assigned randomly so the mapping is 1 = 2, 0 = 1, 2= 0\n",
        "#and we get hundred percent accuracy"
      ],
      "metadata": {
        "id": "pJOr9UM9NTF0"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "#summary:\n",
        "\n",
        "I first loaded the cluster_data1 and loaded it into a dataframe and have done basic missing values test and then visualized it using a scatterplot of two variables.\n",
        "\n",
        "**question 1:**\n",
        "\n",
        "I then started building a kmeans model which is just assigning clusters randomly and then converging until the max number of iterations are reached which is 100 or the there is no movement in the new centroids that are formed.\n",
        "We calc distance between points and initial clusters, and then assign new centroid by taking the mean of the all the points in that cluster. I run the model from initial clusters 1 to 10. to find out the optimal k number of clusters.\n",
        "\n",
        "I repeat the same class functions of kmeans to kmeans++\n",
        "except the centroid initialization, in kmeans. we just find random. points in the data and assign them as the new centroids. and improve it with taking the mean of the assignment.\n",
        "\n",
        "In kmeans++, i take first centroid randomly and then take the farthest point in the datapoints and then keep doing that for the 2nd 3rd centroids already declared, I calculate the euclidean distance of each point from the centroid and then find the argmax of the distances calculated. there are more efficient ways like using the probilitistic determination of which point is fartest in the most efficient code but the question was to implement it from scratch and not write the most efficient code.\n",
        "\n",
        "I plot both kmeans and kmeans++ from 1st assginment of clusters to 10 and calculate the costfunction. I use the labels predicted of cluster 3 as it is the optimal as shown in both graphs and then add them to the dataframe, to do the knn.\n",
        "\n",
        "**quesion 2:**\n",
        "\n",
        "knn i implement from scratch and also use the sklearn to showcase how mine is better. i use the euclidean distance to calculate the distances and take the majority vote to predict the label.\n",
        "I use the train test split 0.8,0.2. train the model on data1 and then print out the results. the accuracy of the model drops as we increase the number of nearest neighbors. i also use the model to predict labels of data2. and train test data2 with kmeans and knn just as extra in case that is what is asked."
      ],
      "metadata": {
        "id": "SlL_5_pVApaL"
      }
    },
    {
      "cell_type": "code",
      "source": [],
      "metadata": {
        "id": "tLwrMnC14MKL"
      },
      "execution_count": null,
      "outputs": []
    }
  ]
}