ml_python.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": false,
    "editable": false,
    "tags": []
   },
   "source": [
    "## Machine Learning in Python\n",
    "\n",
    "*Prepared by:*\n",
    "**Jude Michael Teves**  \n",
    "Faculty, Software Technology Department  \n",
    "College of Computer Studies - De La Salle University"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This notebook shows how perform basic machine learning using sklearn."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Preliminaries"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Import libraries"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-08-15T19:13:07.716339Z",
     "iopub.status.busy": "2021-08-15T19:13:07.716339Z",
     "iopub.status.idle": "2021-08-15T19:13:09.697307Z",
     "shell.execute_reply": "2021-08-15T19:13:09.696308Z",
     "shell.execute_reply.started": "2021-08-15T19:13:07.716339Z"
    }
   },
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "import seaborn as sns\n",
    "\n",
    "sns.set_style('darkgrid')\n",
    "\n",
    "sns.set_palette('Set2')\n",
    "# sns.color_palette('Set2')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Load Dataset\n",
    "\n",
    "We'll be using the **Iris dataset** for this exercise. The dataset contains 3 classes:\n",
    " - 0 - Iris-Setosa\n",
    " - 1 - Iris-Versicolour\n",
    " - 2 - Iris-Virginica"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-08-15T19:13:09.699307Z",
     "iopub.status.busy": "2021-08-15T19:13:09.699307Z",
     "iopub.status.idle": "2021-08-15T19:13:10.867569Z",
     "shell.execute_reply": "2021-08-15T19:13:10.866570Z",
     "shell.execute_reply.started": "2021-08-15T19:13:09.699307Z"
    }
   },
   "outputs": [],
   "source": [
    "from sklearn import datasets\n",
    "\n",
    "X, y = datasets.load_iris(return_X_y=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Exploratory Data Analysis\n",
    "\n",
    "It is a good practice to explore our data first before we do modeling."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Descriptive Statistics"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-08-15T19:13:10.869571Z",
     "iopub.status.busy": "2021-08-15T19:13:10.869571Z",
     "iopub.status.idle": "2021-08-15T19:13:10.927585Z",
     "shell.execute_reply": "2021-08-15T19:13:10.926584Z",
     "shell.execute_reply.started": "2021-08-15T19:13:10.869571Z"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>150.000000</td>\n",
       "      <td>150.000000</td>\n",
       "      <td>150.000000</td>\n",
       "      <td>150.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>5.843333</td>\n",
       "      <td>3.057333</td>\n",
       "      <td>3.758000</td>\n",
       "      <td>1.199333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>0.828066</td>\n",
       "      <td>0.435866</td>\n",
       "      <td>1.765298</td>\n",
       "      <td>0.762238</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>4.300000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.100000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>5.100000</td>\n",
       "      <td>2.800000</td>\n",
       "      <td>1.600000</td>\n",
       "      <td>0.300000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>5.800000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>4.350000</td>\n",
       "      <td>1.300000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>6.400000</td>\n",
       "      <td>3.300000</td>\n",
       "      <td>5.100000</td>\n",
       "      <td>1.800000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>7.900000</td>\n",
       "      <td>4.400000</td>\n",
       "      <td>6.900000</td>\n",
       "      <td>2.500000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                0           1           2           3\n",
       "count  150.000000  150.000000  150.000000  150.000000\n",
       "mean     5.843333    3.057333    3.758000    1.199333\n",
       "std      0.828066    0.435866    1.765298    0.762238\n",
       "min      4.300000    2.000000    1.000000    0.100000\n",
       "25%      5.100000    2.800000    1.600000    0.300000\n",
       "50%      5.800000    3.000000    4.350000    1.300000\n",
       "75%      6.400000    3.300000    5.100000    1.800000\n",
       "max      7.900000    4.400000    6.900000    2.500000"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.DataFrame(X).describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Visualize the data\n",
    "\n",
    "Typically, in the EDA step, we visualize the data that we have to understand it better."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-08-15T19:13:10.929584Z",
     "iopub.status.busy": "2021-08-15T19:13:10.929584Z",
     "iopub.status.idle": "2021-08-15T19:13:10.943586Z",
     "shell.execute_reply": "2021-08-15T19:13:10.941586Z",
     "shell.execute_reply.started": "2021-08-15T19:13:10.929584Z"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "merged_X_y = np.concatenate((X, y.reshape(-1, 1)), axis=1)\n",
    "df = pd.DataFrame(merged_X_y, columns=['sepal_length', 'sepal_width', 'petal_length', 'petal_width', 'target'])\n",
    "df['target'] = df['target'].astype(int)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-08-15T19:13:10.945586Z",
     "iopub.status.busy": "2021-08-15T19:13:10.944587Z",
     "iopub.status.idle": "2021-08-15T19:13:11.423204Z",
     "shell.execute_reply": "2021-08-15T19:13:11.422205Z",
     "shell.execute_reply.started": "2021-08-15T19:13:10.945586Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 800x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(8,6), dpi=100)\n",
    "\n",
    "sns.scatterplot(data=df, x='sepal_length', y='sepal_width', hue='target', style='target', palette=\"Set2\")\n",
    "\n",
    "plt.title('Whole Set Data Points')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-08-15T19:13:11.426207Z",
     "iopub.status.busy": "2021-08-15T19:13:11.425207Z",
     "iopub.status.idle": "2021-08-15T19:13:11.634252Z",
     "shell.execute_reply": "2021-08-15T19:13:11.632251Z",
     "shell.execute_reply.started": "2021-08-15T19:13:11.425207Z"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 500x300 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(1,1, figsize=(5,3), dpi=100)\n",
    "df['target'].value_counts().plot(kind='bar')\n",
    "plt.xticks(ticks=range(3), labels=['Setosa', 'Virginica', 'Versicolour'], rotation=0)\n",
    "plt.ylabel('Count')\n",
    "plt.title('Flower Type distribution')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-08-15T19:13:11.636253Z",
     "iopub.status.busy": "2021-08-15T19:13:11.636253Z",
     "iopub.status.idle": "2021-08-15T19:13:12.068861Z",
     "shell.execute_reply": "2021-08-15T19:13:12.067861Z",
     "shell.execute_reply.started": "2021-08-15T19:13:11.636253Z"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 800x400 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(1,1, figsize=(8,4), dpi=100)\n",
    "df[df['target']==0].iloc[:,0].hist(alpha=0.5, label='Setosa')\n",
    "df[df['target']==1].iloc[:,0].hist(alpha=0.5, label='Virginica')\n",
    "df[df['target']==2].iloc[:,0].hist(alpha=0.5, label='Versicolour')\n",
    "plt.title('Sepal Length Distribution')\n",
    "plt.ylabel('Count')\n",
    "plt.xlabel('Length (centimeters)')\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-08-15T19:13:12.071861Z",
     "iopub.status.busy": "2021-08-15T19:13:12.071861Z",
     "iopub.status.idle": "2021-08-15T19:13:12.458948Z",
     "shell.execute_reply": "2021-08-15T19:13:12.457948Z",
     "shell.execute_reply.started": "2021-08-15T19:13:12.071861Z"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAq8AAAF+CAYAAACs6/OKAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjQuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8rg+JYAAAACXBIWXMAAA9hAAAPYQGoP6dpAABab0lEQVR4nO3deXwT1d4G8OfMkqQbUEq1oCiyK7KXHWURRFkUsYiCKFxFVBThiiziAl4BFxBlExQQEfRVUZRF3K6gqAgCKm54RRBkh7YUSptt5rx/hMaGFkjaNMnU5/v5FJqZyZxfzslMnk4mEyGllCAiIiIisgAl2gUQEREREQWL4ZWIiIiILIPhlYiIiIgsg+GViIiIiCyD4ZWIiIiILIPhlYiIiIgsg+GViIiIiCyD4ZWIiIiILIPhlYgoSqz0HTHRrjXa7RNR7GB4JaKw+d///oeRI0eiXbt2uPzyy9G+fXuMGDECv/zyS1TrqlevHmbOnFnsvKVLl6JevXr49ddfA6abpolWrVqhXr162LRpU5H7tWnTBiNHjgQAdO7cGWPHjj1rDWPHjkXnzp39t7ds2YKhQ4f6b+/duxf16tXDu+++G/TjKny/wj+XX3452rZti3vuuQffffddwPIbN25EvXr1sHHjxqDW73a7MWXKFKxcufKcyxbu51DbOZsdO3bglltuOWNbRPTPokW7ACIqH37//Xf069cPjRo1wvjx41GlShUcPHgQS5YsQb9+/fDaa6+hSZMm0S6ziDZt2gAAtm7diksvvdQ/fdu2bTh27BgqVaqE9evXo2XLlv55O3bsQFZWFtq1awcAmDVrFhITE0Nq9+2338aOHTvC8Ah87rnnHnTs2BEA4HK5cPDgQbz66qsYMGAAZsyYgS5dugAAGjRogDfffBO1a9cOar2HDx/GokWLMGXKlHMu++abbyItLa3Ej+FM1qxZUySEl1VbRBT7GF6JKCxeeeUVVKpUCfPnz4eu6/7pXbp0wbXXXos5c+bgpZdeimKFxatZsybS0tKwdetWDBgwwD/9yy+/RI0aNdC8eXOsX78eDz74oH/et99+CwBo3749AOCyyy6LbNHFuOiii4r8cXDttdeif//+GD9+PFq3bo3ExEQkJiaW2R8RkfzjJBb/ECKiyOBpA0QUFkePHgVQ9NzE+Ph4jBs3Dtdee23A9E8//RR9+vRBw4YN0a5dOzz55JPIy8vzz585cyY6d+6MtWvX4pprrkHjxo3Rt29fbNiwIWA927dvx3333YfWrVujQYMGuOKKK/Dkk0/C6XQGXXubNm2wdevWgGnr169HmzZt0L59e2zfvt3/+ABg8+bNqFWrlv/I3+mnDeTk5GDcuHFo1aoVWrRogWeffRamafrnjx07FsuXL8e+ffuKnCpw5MgRDB8+HE2bNkXLli3x6KOPBvRLKGw2G+6//34cO3YMa9asAVD07XyXy4WJEyfiyiuvxOWXX45rrrkGCxcuBOA7JeGqq64CAIwbN85/2sPYsWNx++234/HHH0d6ejpuuOEGeL3eYt/K37FjB/r374+GDRuia9eueO211wLmF3efmTNnol69ev7fZ82aVWTZ0+93+PBhjBs3Dh06dECjRo2QkZGB//73v0XaWrp0KcaPH4+WLVuiadOmGD58eMDYElHsY3glorDo2LEj9u/fj5tvvhlLly7FH3/84Q+y11xzDW644Qb/sitXrsSwYcNQs2ZNzJ49G/fddx9WrFiBe++9NyD8ZmVlYcyYMejfvz9eeOEFxMXFYciQIfjpp58A+ALLgAEDkJ+fj6eeegovv/wyrr32Wrz22mtYtGhR0LW3adMG+/fvx8GDBwH4wuePP/6I9u3bo02bNhBC4Msvv/Qvv3nzZv8pA6czTRN33nkn1q1bh1GjRuHpp5/Gd999hw8++MC/zL333osOHTogNTUVb775pv/tfgB44YUXULVqVcyZMwe33XYb3nrrrVKd29muXTsoilIknBeYNGkSPv/8c4wZMwYLFizAVVddhaeffhrvvvsuzjvvPH9wvOeee/y/F/TB7t27MXPmTAwbNgyaVvwbeVOmTEHjxo0xZ84c/x8Wb731VtD19+3bFxkZGQB8pwr07du3yDJHjx5FRkYGNm3ahJEjR2LmzJm44IILMGzYMKxYsSJg2enTp8M0TTz33HMYPXo01q1bh8mTJwddDxFFH08bIKKw6N+/P44cOYIFCxbgiSeeAAAkJyejffv2GDhwIBo3bgzAd2R26tSpuOKKKzB16lT//WvUqIFBgwbh888/94e5/Px8TJgwAb179wYAtG7dGl26dMFLL72EGTNm4H//+x8uvfRSvPDCC/5zTtu2bYsNGzbg22+/xd133x1U7W3btgXgO++1e/fu+Oqrr6Aoiv+t9gYNGmD9+vXo3bs3/vrrLxw8eNB/ysDpvvjiC2zbtg3z5s3zP47WrVsHfFjroosuQuXKlWGz2fxvfxccXe3WrRvGjRsHwBeqv/rqK3zzzTdBPY7iaJqGSpUq4ciRI8XO37RpE9q2bYsePXoAAFq1aoX4+HgkJyfDZrP5zwO+6KKLAk6P8Hq9mDhxIi6++OKztt+nTx+MGTMGAHDFFVfg0KFDmD17NjIyMqAo5z5+kpaW5j/CfaZTBV555RVkZWVhzZo1qF69OgCgQ4cOGDRoEJ555hn07NnT31bdunUDzt/dtm0bPvzww3PWQUSxg0deiShsHnjgAaxfvx7Tpk1DRkYGEhMTsXLlSvTr1w+vvvoqAGDnzp04ePAgOnfuDK/X6/9p0aIFEhMT8dVXX/nXp6qqP1QBgMPhwJVXXoktW7YA8J1zumTJEtjtduzatQtr167F3LlzkZWVBbfbHXTdqampqF27tv/o5JdffokmTZr4A3G7du3w9ddfA/CFPV3X0aJFi2LXtXnzZui6jiuvvNI/LT4+Hh06dAiqlvT09IDb1atXx/Hjx4N+LGcihCh2eqtWrfD2229jyJAheP3117Fv3z4MGzYMnTp1Ouv6HA4HLrroonO2271794DbXbt2xcGDB7Fz587giz+HTZs2oWnTpv7gWuC6667DkSNHAto6PQCnpaUhPz8/bLUQUdnjkVciCquKFSuiZ8+e6NmzJwDgl19+wejRozF16lRcd911OHbsGABg4sSJmDhxYpH7Hz582P975cqVAz78BQApKSnIyckBAP/bv0uXLkVeXh6qVq2KRo0awW63h1x34fNev/zyS/Tv398/74orrsDcuXPx+++/Y/PmzWjWrBni4+OLXU9OTg4qVapU5KhiampqUHXExcUF3FYUpVTXOHU6ncjJyTnjJ/PHjx+PtLQ0rFixwj8eTZs2xWOPPXbWD6KlpKScMRAXdvrjTklJAQD/GIZDTk4OLrzwwiLTq1SpAgAB4T/c/UtEkcfwSkSldujQIdx444144IEHipyTeNlll2HEiBEYNmwY/vrrL1SoUAEAMHr06IDLTxWoWLGi//djx45BShkQko4ePeoPQC+99BIWLVqECRMmoFu3bkhKSgIA/zmSoWjbti3eeOMN/PTTTzh06FDAaQFNmjRBQkICtm7dii1btpx1/cnJycjOzoZhGFBVNeCxRMPGjRthGMYZjxTbbDbcc889uOeee7B//36sXbsWc+bMwYMPPuj/kFdpnB5SCz4cVTCGAGAYRsAyoX5ArWLFisV+6KrgVInk5OSQ1kdEsY2nDRBRqVWpUgWapuH111+Hy+UqMn/nzp2w2+24+OKLUbNmTaSkpGDv3r1o2LCh/yctLQ3Tpk0L+EIDj8eD9evX+287nU588cUX/muzbtmyBbVr10ZGRoY/uB46dAj/+9//Aj7dH4yCIP3aa68hOTkZDRo08M/TNA2tW7fGV199hd27d5/xw1qA7wiu1+vFp59+6p/mdrsDTocAENT5nqXl9Xrx4osvokqVKujatWuR+U6nE926dfNfXaBatWoYMGAAevTo4f/wWuEAXhKFxw8AVq9ejapVq/rPlU1MTPS3VeD0D5edq69atGiB7777Dn/99VfA9BUrViA1NfWc5+USkbXwyCsRlZqqqpgwYQKGDRuGG2+8EQMGDECtWrWQn5+Pr776CkuXLsUDDzzgP6o6cuRIPPbYY1BVFZ06dcLx48cxZ84cHDp0KCA0AsDDDz+MESNGICUlBQsWLEBeXh7uueceAECjRo38149t0qQJdu/ejXnz5sHtdod8HmNiYiIaNmyI1atXo1u3bkXeEm/Xrh2mTJmCypUrn/Xt9ILLaz3yyCPIzMzEBRdcgMWLFyMrKyvgaGOFChVw9OhRfP755wFfjlBSe/bswffffw/AF/r37t2L//u//8PPP/+M2bNnF3m7HPCdt9qgQQPMmjULuq6jXr162LVrF5YvX45u3boBgP+Pgg0bNqBWrVr+D94F67XXXkNCQgIuu+wyrF69GuvXr8czzzzj79+OHTti9erVaNSoES655BIsX74cu3fvDlhHwdH6VatWoXHjxkXObR08eDBWrFiBwYMH47777kNycjLee+89fPPNN5g8eXJE/lAgoshheCWisOjYsSPeeustLFiwwP+hKZvNhssuuwzTp0/H1Vdf7V+2b9++SEhIwPz58/Hmm28iPj4ezZo1w9SpU4sEkwkTJmDy5MnIyspCs2bN8MYbb/iPpA0dOhTZ2dlYvHgxZs+ejapVq+L666+HEALz5s1DTk5OwGkI59KmTRt89913xV5J4IorrsATTzyBtm3bnvNcz1mzZmHq1KmYMWMGXC4XunfvjptuuinguqN9+vTB559/jmHDhmH48OFFPtgUqhdffBEvvvgiAMBut+P8889Heno6Jk6ciPr165/xfk888QSef/55LFy4EEeOHEFKSgoyMjLwwAMPAPCF+sGDB+PNN9/EunXrihxBPpcnnngCCxcuxPPPP4/q1avjueeeC/gQ3rhx4+D1evHss89C0zR0794dDz74IB555BH/MldffTXef/99jB07FhkZGZgwYUJAG6mpqXjjjTcwbdo0TJo0CR6PB/Xr18ecOXP816klovJDSJ6pTkQxqODi9L/99lu0SyEiohjC91KIiIiIyDIYXomIiIjIMnjaABERERFZBo+8EhEREZFlMLwSERERkWUwvBIRERGRZTC8EhEREZFlMLwSERERkWWU22/Yysw8gUhcR0EIICUlKWLtUdnjmJZPHNfyh2NaPnFcy5/ixrRgWkmU2/AqJSL6pI90e1T2OKblE8e1/OGYlk8c1/InXGPK0waIiIiIyDIYXomIiIjIMhheiYiIiMgyyu05r0RERGQtpmnCNL1wOp3weNw859XCVFWDopTNMVKGVyIiIooqKSWOH89Cfn4uACArS4FpmlGuikorLi4RFSpUhhAirOtleCUiIqKoKgiuiYnJsNns0DQFhsHDrlYlpYTb7UJubjYAoFKllLCun+GViIiIosY0DX9wTUysAADQNAVeL4+8WpnNZgcA5OZmo0KF5LCuO6of2MrKykLXrl2xceNG/7Tt27fj9ttvR9OmTdG2bVtMmTIFXq83ilUSERFRWTEMA8DfYYfKj4IxNYzw5riohdctW7agX79+2LNnj39aVlYWBg0ahLZt22LTpk146623sG7dOrz66qvRKpOIiIgiINznRVL0ldWYRuW0geXLl2PGjBl46KGHMHLkSP/09957DzVq1MDQoUMBABdeeCEWLlzIJzQREdE/kBCRC7VSSl7dwCKiEl7bt2+PXr16QdO0gPC6bds21K1bF4899hj++9//Ii4uDjfeeKM/zBIREdE/gxAAdIF86YlIew6hQ3giG2BdLhdyco7hvPPOj1yj5UBUwmtqamqx03NycvDpp59iwoQJePTRR/HHH3/g7rvvhs1mwx133BFSG5E6WFvQDg8Olx8c0/KpYDwVRcTY0RUe7Skpbqvlw5nGTwhfcP0xcz9cYT5n8nR2VUPDlGqIEzpkCBvk8ePH8dJLc/D11+tx/HgOEhIS0KJFa9x1171BBdJhw4agT5++6N69V2nKt4zCY12a7TamrjZgs9nQsGFDZGRkAADq16+PW2+9FWvWrAk5vKakJJVFiTHTHpU9jmn543R5odr0aJcRwK6rcNhjaldsOdxWrc3pdCIrS4GqCmja3x/FUVUFihDwmF64zbINr4oAFCF8bSrBh9cJEx5GUlISXnnlNaSkVEFmZiamT38WI0cOw9Klb0HTzr5tHzuWDUUJfNzliWkKKIqC5OQEAOHbVmNqj1mrVq2AKw8Avm/bCOWvoAKZmScicjRDCN9gRKo9Knsc0/JJUQRUm47tO47A5YmNS/DYdAU1L6yEk7ko0X7un47bavng8bhhmiYMQ/ovj+W7zqsJU5UwJWCaZTvApgKYUvraDKGtH374DmPHPoqKFSvD6zVRsWIy7r//35g7dxays3Ngt9swd+4sfPnlF3C73WjePB0PPDAKlSunYOTIYTh06CCeeWYyfvnlZ/z732Pwww/f4aWX5uCPP35HUlIFXH31tbj99jtgs9lw9OgRTJnyH/zyy09wOBy49NIG+Pe/x6BKlSo4eTIXs2Y9j+++24KjR48gMTEJffr0xW23/asMe+3cDEPCNE1kZ59E1aqOgG21YPstiZgKrzfeeCNee+01vPzyy/jXv/6FHTt2YMmSJbjzzjtDXpeUiOjOLNLtUdnjmJYvBWPp8phwu43oFlNAAgVPMT7XSo7bqrVZeey6dOmGqVOn4IcfvkfTps3RoMHlSEurivHjJwAAHnlkDPLyTmLBgtdgtzswc+Z0PPzwQ3jxxQWYPn02MjJ64V//ugvdu/fCnj1/YuTIYbj77vvx/PNzcOjQQYwfPxonT57EiBGjMHfuLJx33nl4+umP4Xa7MH78aCxZsggjRozCiy/Owv79+/Hyy4uRmJiIzz//DI88MgadO3fFhRdWj24nFRKubTWmwmutWrWwZMkSPPPMM3jppZfgcDhwyy23YODAgdEujYiIiCjAmDGPoFmzdPz3vx/j2WcnITc3FxdccCHuuGMoWrRohXXr/ovXX1+G5OTKAIAHHngQ3bp1wG+/bUf9+pcGrOvjjz9ErVq1cdNNtwAALrywOu6+exgeeWQMhg//N+x2O77/fis+/fQjpKe3xLRpM6EovtMN7rjjLqiqioSEBBw+fMh/fdWjR4/EVHgNl6iH199++y3gduPGjbF06dIoVUNEREQUHEVR0K1bd3Tr1h1SSvz55y589NEH+M9/HsPQocMAAHfdNSjgPqqq4cCBfUXCa1ZWJqpVuyBgWtWqF8DlciE7OwsjRjyExYsX4o03XsOkSRNQu3YdjBjxEBo3bors7Cy88MI0/PbbdlSrVg316l0GwHfqZXkU9fBKREREZDUbN27AI4+MxjvvrEKFChUhhMAll9TE3Xffh02bvoHH47vE19Kly5CSUsV/v127dhYJqQBQtWo1fP752oBp+/bthc1mQ4UKFfHbb9tx/fU34o47hiI7OxuLFr2M8eMfwqpVn+LRR8eiXbsrMW3aTGiahpycY1i5cnnZdkAUlc+PtxERERGVoSZNmiI5uTImT56IP/7YAa/Xi7y8k/j44zXYu/cvdOrUBW3btscLL0xDTs4xeL1evPrqAgwZchtyc08A8F1lKTc3F4Dv/Nk//9yJt956Ax6PB/v27cVLL81G167XQNd1LF68ENOnP42TJ3ORlJQEhyMOFStWAgDk5ubCbrdDVVVkZ2dj+vRnAQBeb9lepSFahCynH3E9ejRyVxuoUiUpYu1R2eOYlk+KIqDZdfzw66GY+cCWTVdRr2ZlaKLsP01dHnFbLR88HjcyMw8gJaUqdN0GwHe1AdOUyFcjfJ1XQw9pWzx69CgWLpyHTZu+QXZ2FnRdR4MGjfCvf92FBg0ux/HjxzF37kx8883XOHkyF5dcUgv33DMcjRs3AQC8/vpiLFz4Eq68shMee+w/+OGH7zFv3izs3LkDdrsdXbpcg7vuugd2uwNHjx7FtGlP4YcfvoPH40H9+pdi5MiHULNmbWzcuAEzZkzD4cOHkZSUhC5drsbmzZvQteu1uOWWW8uo186tYGyrVKmKqlVTArbVgu23JBheS4k7z/KHY1o+MbyWP9xWy4czhVfDMAFdwBnBb9hChL9hq7wrq/DKc16JiIgo5kgJCI9EnIjMF4tIyeBqFQyvREREFJN81wVloqRA/MAWEREREVkGwysRERERWQbDKxERERFZBsMrEREREVkGwysRERERWQavNkBEREQxSQhACBGRtnipLOtgeCUiopBEMlAEQwiBUMthUIl9QgDxihfC44xIe1J3IM/U+LywAIZXIiIKmhAAdIH8CH3r0VlrAXDUmQuX6kGoecMhdAh+m1JME0JAeJyQu7ZBul1l25bNDnFJIwgtKejryo4cOQxxcfGYPPnZIvNWrFiOZ56ZBJvNjtdffwdpaWkh1/Tss5MBAA899PA5l33wQd9Xzt52279CbseKGF6JiChoQviCayS+b/7ctQBxcTry8z0hhVD/99gLnRfAtwDpdkG6y/7oa6jvJWRk3IyHHx6FzMyjSEmpEjDvvfeWYfDgIbjjjqElrieY0Fpg2rQZJW7HihheiYgoZC7DC6c3ukdfhQAUQ8DpDS28EoVDmzbtkJZWFR98sAoDBw7yT//ppx+xc+cf+Pe/x6B9+3S8/fYKVK1aDe3bpyMjox8++eRDNGjQCM88Mx2ffvoRFiyYh+zsLFx2WUOcf34avF4Pxo+fgEmTJgAAxo+fgAUL5mHXrj9gs9nx9dfrERcXj27duuPuu+8DANx3311o2rQ57rhjKLxeLxYtmo8PPliJ3Nxc1KlTFyNGPIQ6deri6NEjmDHjOfz668/IyspE5cpVcPvt/0LPntdHoQdLjlcbICIiIgqRoii44YYMrFr1XsAR/PfeW4bOnbsWORoLAPv27cU776zGo48+gZ9+2oZJkyZg2LARWLXqU1x/fR988MGKM7b3+edr0aJFK6xe/V+MHv0wli59FT/99GOR5V59dQE++eRDTJs2E2vWfIamTZtjzJiRMAwDTz31H+i6htdeewsff/wFbryxL6ZPfwZ5eXnh6ZQIYXglIiIiKoGePXsjKysTW7duBgAcP56Dzz77FH373lzs8l27XgOHw4GkpCSsXr0CV17ZCe3bXwlN09ChQydceWWnM7ZVvfpFuPbanlBVFW3atEdKShX89dfuIst9+OFq9O9/Gy65pCZUVcXtt9+BJ554ClJKjBnzCB58cCw0TcOhQwcRH58Al8uF48ePh6dDIoSnDRARERGVQGJiIrp1644VK5ajefMWWLVqBerWrYdLL22AAwf2F1m+SpVU/++HDh1C3br1AuZXq3YBsrIyi22rcuWUgNuaphV7znZm5lGkpVX139Z1HZdf3hAAsH//Psye/QL++msPqle/CNWrVwcASGkG+YhjA4+8EhEREZXQjTf2w/r165CTcwwrVixHRka/oO6XlpaGgwcPBEw7/XZJnHfe+Th06KD/ttfrxYwZ03Do0EGMHj0C3bp1xwcf/BcvvbQIffv2L3V70cDwSkRERFRCl1xSEw0bNsHMmdPhcjnRseNVQd3vuutuwBdfrMPGjRtgGAY2bPgKX3yxttT1dO/eC6+//hr27NkNr9eLxYsX4osv1iEhIRFOpxMOhwNCCBw8eBAvvui7SoHHE/1L34WCpw0QERFRzBI2e8y3kZFxE8aNG4UhQ+6BpgUXrerXvwwPPTQOU6c+hZycY2jcuAmaN28BTdNLVUv//rfB6/XiwQfvx/HjOahfvwGmTp2BxMREPPzw45g/fy6ef34qkpOT0avXDdi1ayd27tyBiy66uFTtRpKQ5fQid0ePnojIpVOEAKpUSYpYe1T2OKblk6IIaHYdP/x6CG63Ee1yAAA2XUW9mpWhCcA0rfFkUxSBfNWDzYf3xMSlsuLjbcjLc4e0rTo0HennXYQ4Q7dMv5dnHo8bmZkHkJJSFbpuAwBomgLDMMv1N2zt2bMbUkpcfHEN/7Tx4x/CRRfVwNChwyJTRBkrGNsqVaqiatWUgNfVgtfakuCRVyIiIoo5UgJ5pgahlSzghNyeGdlvXNu1ayeee+5pzJkzHxdccCG2bt2MjRs34Oabb41cERbF8EpEREQxSUqU229B69ChE/78cyfuv38oTpw4jrS0qhg9+hE0bNg42qXFPIZXIiIioii4/fY7cPvtd0S7DMuJ6tUGsrKy0LVrV2zcuLHIvMOHD6Nt27Z49913o1AZEREREcWiqIXXLVu2oF+/ftizZ0+ReaZpYtSoUcjOzo5CZUREREQUq6ISXpcvX45Ro0Zh5MiRxc6fPXs20tLSULVq1WLnExEREdE/U1TCa/v27fHJJ5+ge/fuReZ98803WL16NR5//PEoVEZEREREsSwqH9hKTU0tdnpmZiYefvhhzJgxAwkJCaVqQ4hS3T3kdiLVHpU9jmn55B9X/z8xQPxdipWebwK+emOp5lBqERbt9/KKY/DPUXisSzPuMXO1ASklRo8ejYEDB+Lyyy8v9fpSUiJzXbhotUdlj2Na/uTkuuCI06HqarRLAQDomgK7Q0fFxLL/BqFwOurMRVycDsWIjdQRH28LaXm7qsHu0JHiSCyjiigUTqcTWVkKVFVA0/5+Q7jgdxGhdFteL8kVTaYpoCgKkpN9ByTD9boaM+H1wIED2LRpE3744QfMnj0bAJCbm4uJEyfio48+wrx580JaX2Zm5L5hKyUlKWLtUdnjmJZPiiKg2nQ48z1weWLnG7ZcTg8yXR7LvHAKIeBSPcjP90T9G7aAv79hKxSmJn39fjLXMv1ennk8bpimCcOQ8HpNAH9/w5aiqnCfmlbWdFWBaRjc74eRYUiYpons7JOoWtUR8Lpa8FpbEjETXqtVq4Yff/wxYFrnzp1x3333oU+fPiGvz3dh43BVF3vtUdnjmJYvBWMp/f/EAPl3KVZ5rgnhqzkWto/CB+RCqUVasN/LszONgRACHsPErr+Owe0p2wBr0xVcUr0SNCGC/oNm5MhhiIuLx+TJzxaZt2LFcrz88ot4551VsNlCe2egOB9/vAaLF7+CJUveKtV6Jk2aAAAYP35CqWsqiXDtN2ImvBIRERGdzu0x4Y6Rd0sKy8i4GQ8/PAqZmUeRklIlYN577y1D7943hiW4AsDVV1+Lq6++NizrKg+i+iUFAPDbb7+hVatWxc777LPPSnTUlYiIiKgstWnTDmlpVfHBB6sCpv/004/YufMPXH/9jXj77f/DLbf0wTXXdMS9996J7dt/9S+XkdELzz47Gddd1w2DB/eHx+PB1KlP4brruqFHj6tw7713Ytu27wEAH3ywEhkZvfz3/fbbbzBkyG3o2vUK9O17Hd55503/vC++WId//etWXH11B9xySx+89dbrMM3ij1yvXPkebr31Jlx9dQfcdls/fPzxGv+8++67CwsW/H3K5oED+9G+fToOHNgPAGjfPh3PP/8sevS4CqNHF3/p07LCI69EREREIVIUBTfckIHly5fh1ltv93+w7L33lqFz5674/PPP8H//twRPPz0dNWpcgg8/XI0RI+7F668vQ+XKKQCAX375CUuXLgPgOzXgp5+2YenSZYiPj8f8+XMxbdrTePXVNwLa3bNnN8aM+Tf+/e8xuOaaHtix43cMH343LrzwIui6jsceG4tHH/0POnTohD/+2IFx4x6ElBL9+g0IWM8HH6zErFnTMXnyVDRp0gzff78VDz88Cna7Ax06dAqqD/bt24t33lkNjyey579H/cgrERERkRX17NkbWVmZ2Lp1MwDg+PEcfPbZp+jb92YsX/42Bg4cjNq160DTNPTseT1q1LgEH33099HNjh2vQlJSEpKSkmC323HgwD6sWvU+9uzZjSFD7ikSXAHg008/Qt269dGz5/XQNA3161+KOXPmo27d+li9egWuuKIjrrqqKzRNQ7169XHrrYPw/vvvFlnP6tUrcP31fdC8eQuoqormzVvg+uv7FLvsmXTteg0cDgeSkiJ7dR4eeSUiIiIqgcTERHTr1h0rVixH8+YtsGrVCtStWw+XXtoABw7sx+zZz2Pu3Jn+5b1eL+rXv9R/u0qVv69736VLN3g8Hqxa9T5eemk2kpMr47bbBqN374yANjMzj+L889MCptWuXQcAkJ2dhTp16gXMq1q1Gg4ePFCk9qysTFSrdsFpy16AL7/8IujHX7j+SGJ4JSIiIiqhG2/shzvuuBU5OcewYsVy3HnnUABAaur5uPPOoejSpZt/2X379qJChYr+24WvYbtnz27Uq3cprr22J1wuJz777FNMmjQBjRo1DWjvvPPOx86dOwKmrV69AsnJlZGWVhX79u0NmLd//94iHygDfKF23759AdP27ft7WVVV4fV6/fNyco4F0x0RwdMGiIiIiErokktqomHDJpg5czpcLic6drwKAHDddTfg1VcXYPfuPwEAGzduwMCBN+GHH7YWu56vvlqPhx9+CAcO7Ifd7kDFipWgqioSEwO/TKNLl2747bffsGbNKhiGge3bf8XMmdOhaRp69LgeX375OT777FMYhoH//W87li5djB49rivSXs+evfH+++9iy5ZvYRgGtm7djBUrlvuXvfjiGti48WucOHECubm5WLLk1TD2WunwyCsRERHFLJte9sfZSttGRsZNGDduFIYMuQea5otW/fr1ByAxZsy/kZl5FKmpqRg5cjTat+9Q7Dr69r0ZR48ext13/wsnT+YiLa0annhiCs477/yA5S644EJMnfoCXnxxJp5//lkkJ1fG/fePRMuWrQEATz75NBYufBlTpjyBihUronfvGzFgwO1F2uvcuQvy8nIxffqzOHToAFJTz8O99w7HNdf0AADcdtu/MGXKf9C373VISkrCHXcMxbp1/y1VP4WLkOX060WOHo3cN2xVqZIUsfao7HFMyydFEdDsOn749RDc7ti4ZqRNV1GvZmVoAjBNazzZFEUgX/Vg8+E9Uf+GLSH+/oatULZVh6Yj/byLEGfolun38szjcSMz8wBSUqpC133XRS38DVseg9+wZVUFY1ulSlVUrZoS8Lpa8FpbEjzySkRERDFHSsA0DGiFv0qtDDG4WgfDKxEREcUk39eJMlFSIH5gi4iIiIgsg+GViIiIiCyD4ZWIiIiILIPhlYiIiKJOyshcVYAip6zGlB/YIiIioqjRNB1CKMjJyURiYiWoqgbTVGAY/KCWVUkpYRhenDhxDEIo0DQ9rOtneCUiIqKoEUIgJSUNOTlZyMk5CgBQFAWmySOxVmezOVChQuWAr8ENB4ZXIiIiiipN01G58nkwTQNSmkhOTkB29kled9XCFEWBoqhhD64AwysRERHFACEEVFWDEIDD4YCuexheqVj8wBYRERERWQbDKxERERFZBsMrEREREVkGwysRERERWQbDKxERERFZBsMrEREREVkGwysRERERWQbDKxERERFZBsMrEREREVlGVMNrVlYWunbtio0bN/qnffTRR7j++uvRrFkzdO7cGbNmzeL3GxMRERERgCiG1y1btqBfv37Ys2ePf9pPP/2E0aNHY8SIEdi8eTNefvllvPvuu1i0aFG0yiQiIiKiGBKV8Lp8+XKMGjUKI0eODJi+b98+3HzzzejUqRMURUGtWrXQtWtXfPvtt9Eok4iIiIhiTFTCa/v27fHJJ5+ge/fuAdO7deuGcePG+W87nU6sW7cODRo0iHSJRERERBSDtGg0mpqaes5lcnNz8cADD8DhcGDQoEEhtyFECQorgYJ2ItUelT2rjamvzlgrVkLKaNcQyD+u/n9igPi7FKs83wBfzULEVs2h1CIs2u//FFbbB9O5FTempRnfqITXc9m5cyeGDx+OlJQULF68GImJiSGvIyUlqQwqi532qOxZZUydLi9cHiPaZQSw6yoc9tjbveTkuuCI06HqarRLAQDomgK7Q0fFRHu0SwnJUWcu4uJ0KEZspIv4eFtIy9tVDXaHjhRH6K8tFDlW2QdT8MI1pjH36vL555/j3//+N2666SY8+OCD0LSSlZiZeSIiR36E8A1GpNqjsmelMRVCwJDAzr3H4PbExlU5bLqCmhdWwslcQMZQByqKgGrT4cz3xEzYt+kqXE4PMl2emOqrsxFCwKV6kJ/vgdPriXY5iI+3IS/PHdJ9TE36+v1krmX6/Z/ESvtgCk5xY1owrSRiKrx+//33GDZsGCZMmICMjIxSrUtKRPRJH+n2qOxZYUyFACQAt9uEO0YCGaSvJiC2+q+gFun/JwbEaF+dTcFzLha2j8JvO4ZSi7Rgv/8TxcJzjMIrXGMaU19SMHfuXHi9XkyaNAlNmzb1/9x5553RLo2IiIiIYkDUj7z+9ttv/t/nzp0bxUqIiIiIKNbF1JFXIiIiIqKzYXglIiIiIstgeCUiIiIiy2B4JSIiIiLLYHglIiIiIstgeCUiIiIiy2B4JSIiIiLLYHglIiIiIstgeCUiIiIiy2B4JSIiIiLLYHglIiIiIstgeCUiIiIiy9CiXQARhZkARDSbj3L7BYQAhBCFbvt+F6fmRYL0/0NEROHC8EpUnghAFyaE4Y1aCbqpQhUS0YywQgDxihfC4/x7mqrA5XZAl25I04hIHVLV4IHCAEtEFEYMr0TliAAgDC/ksUOAEZmAdjoZHwdhVgFUHdFKbUIICI8Tctc2SLfLN81mh0yrA5l1CNLlLvsiVBWi0vkQio3ZlYgojBheicojw4CM1tHXKIXm4ki3C9L999FXmAak6Y1I38TCqRNEROURP7BFRERERJbB8EpERERElsHwSkRERESWwfBKRERERJbB8EpERERElsHwSkRERESWwfBKRERERJbB8EpERERElsHwSkRERESWwfBKRERERJYR1fCalZWFrl27YuPGjf5pP/zwA/r27YumTZuic+fOePvtt6NYIRERERHFkqiF1y1btqBfv37Ys2ePf1pOTg7uuusu9O7dG99++y0mTZqEKVOmYNu2bdEqk4iIiIhiSFTC6/LlyzFq1CiMHDkyYPrHH3+MSpUqYcCAAdA0DW3atEGvXr2wdOnSaJRJRERERDEmKuG1ffv2+OSTT9C9e/eA6b///jvq1q0bMK127drYvn17JMsjIiIiohilRaPR1NTUYqefPHkScXFxAdMcDgfy8vJCbkOIEpVW4nYi1R6VPauNqSj4RxSeEBvC2YfC/0CDWVZACN//BUWIgv+DXksYCABCQEAGTJaF5vuHLYbG7VxOPayYqjmUWoRF+/2fwmr7YDq34sa0NOMblfB6JnFxcThx4kTANKfTiYSEhJDXlZKSFK6yYrI9KntWGdOcXBfi4nRouuqb4PRCaiqgyLPfsYyomgpVFahYOfTt9mxOelzINzzB3yEPkDogC7pBBwzThFQAqYa1tGIJFb73tnQERFddqNBVXwG6psDu0FEx0V72BYXRUWcu4uJ0KEZspIv4eFtIy9tVDXaHjhRHYhlVROFglX0wBS9cYxpT4bVu3br46quvAqbt2LEDderUCXldmZkn/n7RKkNC+AYjUu1R2bPSmAohYEggP98Dt8eAAKBLE9JrQBpGVGpSdAOGIZGVdRKmaYZlnUIIOFUPfszaD5fhPefydlXDZboO14lj8Lp879xo9ng4kj044XbC5XKFpa6zUVQdqtuFI858GKf6QVUUnB+XBK/LgARg01W4nB5kujyQsf5kO0UIAZfqQX6+B05vCH9MlJH4eBvy8twh3cfUpK/fT+Zapt//Say0D6bgFDemBdNKIqbCa9euXfHss89i0aJFGDBgALZs2YKVK1dizpw5Ia9LSkT0SR/p9qjsWWFMhTh1VE+e+hEAYqjmcPVfweN0er1BBSYpAa+qwmOY8BqnArRpQkoJU0oYZtl3klQkhJTwmia8p8JrwTAVvlFwO9afawUKxiIWto/CbzuGUou0YL//E8XCc4zCK1xjGlNfUpCcnIyFCxfiww8/RKtWrfDII4/gkUceQevWraNdGhERERHFgKgfef3tt98Cbjds2BD/93//F6VqiIiIiCiWxdSRVyIiIiKis2F4JSIiIiLLYHglIiIiIstgeCUiIiIiy2B4JSIiIiLLYHglIiIiIssIW3jNzc0N16qIiIiIiIoVcnht2bJlsdM7duxY2lqIiIiIiM4qqC8p2L17Nx577DFIKZGbm4vbbrstYH5ubi4qVKhQJgUSERERERUIKrxefPHFuPrqq5GdnY2tW7cWOfpqs9nQuXPnMimQiIiIiKhA0F8PO2DAAADAhRdeiN69e5dVPUREREREZxR0eC3Qu3dvbNu2Dbt27YKUssg8IiIiIqKyEnJ4fe655/Dyyy8jNTUVmvb33YUQDK9EREREVKZCDq/vv/8+5s6diw4dOpRFPUREREREZxTypbLy8vJw5ZVXlkUtRERERERnFXJ47dixI1auXFkWtRARERERnVXIpw24XC6MHTsWc+fORZUqVQLmLV68OGyFERERERGdLuTwWrduXdStW7csaiEiIiIiOquQw+t9991XFnUQEREREZ1TyOF13LhxZ5w3ZcqUUhVDRERERHQ2IX9g63TZ2dlYs2YN4uPjw1EPEREREdEZhXzktbijq19//TVef/31sBRERERERHQmpT7yCgBt27bFN998E45VERERERGdUchHXk/n9XqxatUqVK5cORz1EBERERGdUcjhtX79+hBCBExTVRXjx48PW1FERERERMUJObye/kUEiqLg4osvRmpqatiKIiIiIiIqTsjnvLZs2RLp6elwOBw4evQoACAlJSWsRf38888YMGAA0tPT0b59ezz55JNwu91hbYOIiIiIrCfkI69HjhzB3Xffje3bt6NSpUrIzs5GjRo1sHDhQqSlpZW6INM0MXToUNx111147bXXcPjwYQwaNAjJyckYNmxYqddPRERERNYV8pHXp59+GjVq1MCmTZvw1VdfYePGjbj00kvD9gUFOTk5OHLkCEzThJTSV6SiIC4uLizrJyIiIiLrCjm8fvPNN5g4cSISEhIAAElJSZgwYQI2bNgQloKSk5MxaNAgPP3002jYsCE6dOiAGjVqYNCgQWFZPxERERFZV8inDZimWeRqA0II6LoeloJM04TD4cCjjz6KjIwM7N69G/fddx9mzJiBESNGBL2e00osMwXtRKo9KntWG1NR8I8oPCE2hLMPxan1BbPOguXOtHwkxtZfLwKHRiDwhiJ8+1AlLFfdDieJU29+FRHKWERKKLUUjEuo96PIsNo+mM6tuDEtzfiGHF5btWqFCRMmYOLEiYiPj8fJkycxYcIEtGzZsuRVFPLJJ5/go48+wocffggAqFOnDoYNG4ZJkyaFFF5TUpLCUk+stkdlzypjmpPrQlycDk1XfROcXkhNBZQzJI8ypmoqVFWgYuWEsK73qDMXcXE6FOPcezy7qkFVVGi6AmH6+kXVfelQURQoEUiKiiogFAWaLiFM3zRVUaBqCuI0GwDAYVNhs4fnD/9ws+sqHPbiXyJCGYtIiI+3hbS8XdVgd+hIcSSWUUUUDlbZB1PwwjWmIYfXhx56CIMHD0bLli1RqVIlHDt2DLVq1cJLL70UloIOHDhQ5MoCmqaFfGQ3M/PEGY8ahJMQvsGIVHtU9qw0pkIIGBLIz/fA7TEgAOjShPQakIYRlZoU3YBhSGRlnYRpmmFZpxACLtWD/HwPnF7POZc3NQnDrsDrMeH1+PpBKibs8L27E666zsqQUEwTXq8J76n2pAIYXhP5bjckAAU6nE4P/tx7DC5PBGoKkk1XUPPCSjiZC/9nDwqEOhZlLT7ehry80K5GY2oSLqcHmSdzizw+ij4r7YMpOMWNacG0kggpvEop4fV6sXr1amzevBmZmZnYt28f7rjjDqiqWqICTte+fXtMmzYNc+fOxZAhQ7B//368+OKL6NWrV0jrkRIRfdJHuj0qe1YYUyEACfj+kfC9FxpDNYer/woeZ7BjUrDcmZaPxLj668XfQ1L498I33B4Tbnd0/tgolixU82l9FepYlKXCbzuGUos8y+Oj2BELzzEKr3CNadDvneXl5eGWW27BM888A03T0Lp1a7Ru3RqzZs3CwIEDkZeXV/pqANSuXRvz5s3DZ599hlatWuG2225D586dMXLkyLCsn4iIiIisK+gjry+++CJ0XcfEiRP901JSUrB27Vrcc889mDdvXtgCZtu2bdG2bduwrIuIiIiIyo+gj7x+9NFHePLJJ4t8m1ZKSgomTpzo/4AVEREREVFZCTq8ZmZm4uKLLy523qWXXoojR46ErSgiIiIiouIEHV4TExORnZ1d7Lxjx47xG7CIiIiIqMwFHV7btGmDpUuXFjvv9ddfR5MmTcJVExERERFRsYL+wNbQoUPRp08fZGdno3v37khNTcXhw4exZs0avPPOO1iyZElZ1klEREREFHx4veSSS7BgwQI8/vjjWLp0KYQQkFKibt26ePnll3H55ZeXZZ1ERERERKF9SUGzZs2wcuVK/PXXX8jKykJqaiqqVatWVrUREREREQUI+ethAaB69eqoXr16uGshIiIiIjqroD+wRUREREQUbQyvRERERGQZDK9EREREZBklOueViOhchBBQFBG+dQkBuxrcLsuualBEeNoONwEAQkBAnrpx6nYMlSuErzTfGJ4+TyCGSiWifyCGVyIKL6FACIE44wSkYYZnlQKwGyYus+kwTPWcy6uqioqqBpcQ8IalgvBQTgV6r2lACsAQElIAXmHAo4Snr8JBUQQMRcKremFKGThPiJgL20T0z8LwSkThpQgIacD861fI3ONhWaUQAl5hwJV3Al7z3CHPnlQJyiWXQ8RYwlIg4JUmjuTnwmuaSIYDF3iTcDQ/Fydc7miX5xdn01Al3469ziy4jMD4X8HmQO3KqVGqjIiI4ZWIyorbDel2hmddQkAKA15XXlDhVbPHhafdMmKYJrymCUNKSEh4pRnU44oUrylhSAmX4YXT6wmYF+ypG0REZYUf2CIiIiIiy2B4JSIiIiLLYHglIiIiIstgeCUiIiIiy2B4JSIiIiLLYHglIiIiIstgeCUiIiIiy2B4JSIiIiLLYHglIiIiIstgeCUiIiIiy2B4JSIiIiLLiMnweuzYMYwePRqtWrVCixYtcO+99+Lw4cPRLouIiIiIoiwmw+v999+PvLw8fPLJJ1i7di1UVcWjjz4a7bKIiIiIKMq0aBdwup9++gk//PADvv76ayQmJgIA/vOf/+DIkSNRroyIiIiIoi3mwuu2bdtQu3ZtvPXWW3jjjTeQn5+PK664AmPGjAlpPUKUUYFnaCdS7cUyXx9EuyMkpCzdGqw2pv5uF4UnlD8CvjEJZlwKuuNMy0dibP3to9CQiNPmn/YTK87Wf4Vvx9I2EkotQvjedhRCQInJ9x+DJ4SElDE0ECUSuN+22j6Yzq24MS3N+MZceM3JycFvv/2Gyy+/HMuXL4fT6cTo0aMxZswYzJs3L+j1pKQklWGV0W8vFknnScCVH90i7HEQjoSwrMoqY5qT60JcnA5NV30TnF5ITQWUUqb4ElI1BUIADocOmPawrTfP64amqRBBvFCrugIhBFRNgX6qX1Tdl1IURYESgcSiqAJCUaDpEsL0TdNUBYoQ0HQFwgQ0TUAovtu6ETspStd9fRcXp0MxAvvboetQ1eLnRUt8vC2k5RM0G2wOHSais42EiwCgCBVeaUa7lFKJU21I0IvuK6yyD6bghWtMYy682my+ndD48eNht9uRmJiIESNG4KabbsLJkyeRkBBcMMnMPFHqI3DBEMI3GJFqL1YJIZBgnIC5cxuk2xWdGmx2KDUb4eRJCVmKwbDSmAohYEggP98Dt8eAAKBLE9JrQBpGVGpSvCakBJxOD8y88DwXhBAwhAGv14DXPPcLteoxIaWE4TXh8fj6QSom7ABM04QZxDpKzZBQTBNer+mvWZEmTJuE1+Ob5vVKSNN32+OJnQCiCV/f5ed74PR6AubpdhVGYvHzoiE+3oa8PHdI99HtKvLcbvxy9ABchreMKit7FWwO1EpOtfTjsKsaGlauBqfh8e+3rbQPpuAUN6YF00oi5sJr7dq1YZomPB4P7HbfX2IFLzShBBIpEdEnfaTbizVCnOoDtwvS7YxaHQVjEI6xsMKYCgHfsSN56kec+r8ckgh+TAq640zLR2Jc/e2j0JDI0+af9hMrztZ/hW9He/so/LZjKLUULOsyvMj3RD+Al5RN8b2EW/lxFGwjBb8XmRdLGwaVWrjGNHbepzqlbdu2qF69Oh5++GGcPHkSWVlZmD59Orp06eL/ABcRERER/TPFXHjVdR2vvfYaVFVFt27d0K1bN6SlpWHy5MnRLo2IiIiIoizmThsAgPPPPx/Tp0+PdhlEREREFGNi7sgrEREREdGZMLwSERERkWUwvBIRERGRZTC8EhEREZFlMLwSERERkWUwvBIRERGRZTC8EhEREZFlMLwSERERkWUwvBIRERGRZTC8EhEREZFlMLwSERERkWUwvBIRERGRZTC8EhEREZFlMLwSERERkWUwvBIRERGRZTC8EhEREZFlMLwSERERkWUwvBIRERGRZTC8EhEREZFlMLwSERERkWUwvBIRERGRZTC8EhEREZFlMLwSERERkWUwvBIRERGRZTC8EhEREZFlMLwSERERkWXEbHg1DAMDBw7E2LFjo10KEREREcWImA2vs2bNwubNm6NdBhERERHFkJgMrxs2bMDHH3+Mq6++OtqlEBEREVEM0aJdwOkyMzMxfvx4zJkzB4sWLSrxeoQIX03BtBOp9mJZWPtA1SFUNbT2bXYITYWiKBBClrjpgsehqgpkwGrkabeLE8wy4SUK/hGFJ5Q/Ar6xCeZ5VtAdZ1o+Eturv30UGhJx2vzTfmLF2fqv8O1Y2u+FUkusPoZQlYfHUbCNFPxe3P9kfcWNaWnGN6bCq2maeOihhzB48GDUr1+/VOtKSUkKU1Wx2V4skjlOyDgboJUyvakaXOfVgkuGFl6h6oCwQ5Ge0rUvgZysnCJvS9g0BQ7p9i1wJvY4CEdC6doPUU6uC3FxOjT9VH85vZCaCigRTtGnqJoCIQCHQwdMe9jWm+d1Q9NUCHnuPZ6qKxBCQNUU6Kf6RdV9I6ooChSl7N90UlQBoSjQdAlh+qZpqgJFCGi6AmECmiYgFN9t3YidN8J03dd3cXE6FCOwvx26DlUtfl60xMfbQlo+Fh9DSZSHx2FXNdgdOlIciUXm8XW1/AnXmMZUeJ03bx5sNhsGDhxY6nVlZp6IyBEwIXyDEan2YpUQAgmGB0a+G9LtKt26bAJOj8DOPUfhcbuDv59uh1LBDfN4FmAYpapB1VQY3r/Xodt1XFK9MuTRP874+ITNDqVmI5w8KSEj9GQQQsCQQH6+B26PAQFAlyak14AsZR+UlOI1ISXgdHpg5pXuuVBACAFDGPB6DXhN85zLqx4TUkoYXhMej68fpGLCDt8fyWYQ6yg1Q0IxTXi9pr9mRZowbRJej2+a1yshTd9tjycCNQVJE76+y8/3wOkN/GNQt6swEoufFw3x8Tbk5QW/nwBi7zGUVHl4HKYm4XJ6kHky17/f5Otq+VPcmBZMK4mYCq/vv/8+Dh8+jPT0dACA0+kEAHz66achf3hLSkT0SR/p9mKNEOF//B63G25nCOHVFFA9Xhh5+ZCGt1Rt67rqDz0AIE0bYBqQbhek23nG+xX0QaSeC0KcOg4sT/0InPXAsJVJBL+dFXTHmZaPxPj420ehIZGnzT/tJ1acrf8K3472Pq/w246h1BJLj6E0ysPjKNhGCn4vMs+ij4uKF64xjanw+uGHHwbcLrhM1lNPPRWNcoiIiIgoxsTOSVZEREREROcQU0deT8cjrkRERERUGI+8EhEREZFlMLwSERERkWUwvBIRERGRZTC8EhEREZFlMLwSERERkWUwvBIRERGRZTC8EhEREZFlMLwSERERkWUwvBIRERGRZTC8EhEREZFlMLwSERERkWVo0S6AqFwRAkIIKBH6s1AIASElhACE8LUPEZm26dyEENCE4j9MoCoKBARURYEEoAoBAd8yWhk9aaSUMKQsk3UTEUUDwytRuKgaNFVBvPc4IpYVFAX5sEM33ZCmAQgBBRJGhJqnMxOKCoemIxUemNL3F4WqCDgMD1IVBaYQSFJV2BWBFE1FvFctkzoMRcURt4sBlojKDYZXojARqgrhdcL8czukyxmZNm12yCq1ILMPQbrcgG4DKiRHpG06OyEUCNOAmX0IXq/HN1FRYGo2GB5fmDQ8SZDuSvBmHYY372TYa1BUDWry+RBCIHJ/URERlS2GV6Jwc7sg3ZEJrwAA04A0vZCGF0Itm6N3VHKm4YV5KryaigIIBabhgWlKmF4vpJQwzb+XCTd+sIGIyhvu14iIiIjIMhheiYiIiMgyGF6JiIiIyDIYXomIiIjIMhheiYiIiMgyGF6JiIiIyDIYXomIiIjIMhheiYiIiMgyGF6JiIiIyDIYXomIiIjIMhheiYiIiMgyYjK8bt++HYMHD0bLli3Rrl07jB49GllZWdEui4iIiIiiLObCq9PpxJ133ommTZviyy+/xKpVq3Ds2DE8/PDD0S6NiIiIiKIs5sLr/v37Ub9+fQwbNgw2mw3Jycno168fvv3222iXRkRERERRpkW7gNPVrFkT8+fPD5j20UcfoUGDBiGtR4hwVnXudiLV3tnriF4RQoio90G4iEL/y8LTFQXCZj/zHXUboCiAboeQ8szLhZGw2QERc3+D+p6Qug3C5gjT6gQUYUA3PBCmec7lNZsDiqJCs8dBl+apaXEQiuIvr6yJwj/FtCdEof1HGdXkW6+Apigh7R10RcCmKqhod8ChBb5MJNns0BQFFWwO2NXgXkJM04RHnnvcgmGYJjymUWR6KP1XeNlw9LuuqFCVyG+HDk2DIsQZx+FMfRVLhPAdRRNCoKALC8ZEUQQitCstNSEkpCwPL4KyTPq8uKxUmm0v5sJrYVJKPP/881i7di2WLFkS0n1TUpLKqKrYaO900nkScOVHrwDhe5mWcTZAK+UzX7fBoynQNBVSV4O/n6ZCKAJCUwGl9FufVqhtu8MGpUIluLU6wJleDDQbPCIeMq32mZcJM6GqELZEaPoRSMMIex+UhM2uQ2g63KmXAMnusK3XME04DA9kEHtWRbfDpSTCVq0BNMPrm6ZqUOMrQtUOQFE8YavrTITq2zMLRUA59aqsCMX3Yq0ogJS+sRK+F2mlDMKPqmmI03Wcb3hhIvhtKd5hw3mOBCRrWpH+ViCgOyXqx1VGsM8wUyg4bngQjvgqVIk9+ZnwFgrD8fG2kNbh0HWoqoK4OB2KUbrAoQkFF8WlQJZyPSWhKypUr4aLE6rAKGa7KK6vYk2CZoPNocMs9GySAI46c4H46NUVCgFAEWpM93Ow4lQbEvSzHKQppXBlpZgNr7m5uRg3bhx+/vlnLFmyBPXq1Qvp/pmZJyLyF5sQvsGIVHvF1yCQYJyAuXMbpNsVnRoSkmC7sBbcTg9MV+lqEDYBw2vC6zXg8QQfAoUwoJoShtfwBbmStg9fcPV6DP/u1CYBt0diz66jcOcX/0eC0O1QKrhgZh+BNL0lbj8U8YkJqFYrEYYh4fEYYeuD0rBLwO01sXvXUbhP5oZlnQKACYlcrwtmEBuapschobKJk5kH4PX6AnRShSTUrJsMKQXMII7elpZi+OqUpvS3JwQgpe9IpCklpClP3ZZlUpMiBYRhwJt5EIY3+MBuVkyCO68Sft++E878vIB5qqLAoerI97qDGgtFVaFUSsURrwGvWbqdZJxNQ+3qyXC7DDhPPZ74eBvy8kL7I0m3qzASTeTne/zrKSmHpsOrSOz4Kwv57shs9wXsmoqUuAQcPnmiSN8W11exSLeryHO78cvRA3AZf/dfXJyO/PzYrbuwCjYHaiWnFnkMVmNXNTSsXA3OIA8ShKK4rFQwrSRiMrzu2bMHQ4YMQbVq1bBs2TJUrlw55HVIiYiGyUi3V1jBC6J0uyDdzujUcLa30y1GnvZ/YR6XB25n8S+UwhRQPV4YLjdkhHZguj20I06R5HGfua9CJQAYQsLpDjK8mio0jxdOpwtej68Gu/3v52gktlVZ+KegvUIH5wrvMwKWKYMaTMMbWnj1eiGlRH5+HvJzA/8A0RQFim5HvscJI4gwqmg6tDgv8gwD3lIG9ML9KWXg246h9F/hZUvb71L6aspze5HvjmzYMg0Jl2Yg3+2F57S+Pb2vYlVBbS7Di3yPr/+EABRDwOn1xHTtBWyKL0oVfgxWVPBcLvi9zNoIw7pj7mS5nJwc3H777WjWrBkWLFhQouBKREREROVTzB15fffdd7F//36sWbMGH374YcC87777LkpVEREREVEsiLnwOnjwYAwePDjaZRARERFRDIq50waIiIiIiM6E4ZWIiIiILIPhlYiIiIgsg+GViIiIiCyD4ZWIiIiILIPhlYiIiIgsg+GViIiIiCyD4ZWIiIiILIPhlYiIiIgsg+GViIiIiCyD4ZWIiIiILEOLdgHlhRACQliwbVWHUNXSF6HbAEUBdDuElAG1hb4uO6AoEABCu7co5reSK1iHPOtS1hGNp2c424zS5kUxRhECdtX30iUEYFc1mJqEDGFDtauafz2h3O9s6yKiyGF4LSUhAOk8iQTjRKl3giWmKFBhwgj1fqoOeUFduMLxNNB1uBEPI7UmYJyqRABmCaKfoqowHUkwhIAhgr+/gAkVgCkkZAj3K7oewGMakEJCAlDKQ2wSgBGhGF4w5lIgpPE7m4IREALl568JCpmuKqhkj4NAZRhSQgBQNQVGXGh7GpuiQvGouNBRGV6zdE8oTVGQoNmhloPdBJFVMLyWmgBc+TB3boN0u6JTQUISxIW1IIQIaQcuVBUuaNi5Jwset7t0Neh2KBW8MLOPQJpeCPjCUp7XDTPEVJ9UIQmX1ElCnulFvtsZ9P00oSBBmsjzuuH1lPzxCAEoigLTNCEgkKDbLR1fC8bipMcV8liUhGZ4IaVEvteN3BDG76zrVBTEabawrIusS1MVGIbEjr3HkOt0QwDQdQGPR4a073OoKlLiE3D45Al4ShlekxMcaFD9vJK9y0REJcLwGibS7YIM0wt1qITNXqr7e9xuuJ2lDK+mgOrxwnC5IY1T4VVI5LudIQcm3WaDBGBKGdJ9TSlLdL/TCQA4tY7y9HJU2n4JpR0AkBJhay8SdZN1ON0e5Ls9EAC8UoHHE9qRV6lKuHQD+W4vPKZZqlribHwZJYo0fmCLiIiIiCyD4ZWIiIiILIPhlYiIiIgsg+GViIiIiCyD4ZWIiIiILIPhlYiIiIgsg+GViIiIiCyD4ZWIiIiILIPhlYiIiIgsg+GViIiIiCyD4ZWIiIiILCMmw2tmZibuvfdepKeno1WrVpg0aRK8Xm+0yyIiIiKiKIvJ8DpixAjEx8dj/fr1WLZsGTZs2IBFixZFuywiIiIiirKYC6+7d+/Gpk2b8NBDDyEuLg7Vq1fHvffei6VLl0a7NCIiIiKKMi3aBZzu999/R6VKlXD++ef7p9WqVQv79+/H8ePHUaFChaDWoyiAlGVV5d+EAAABER8PqFHqTnucr5C4eAhFDfpuwmaDUDU4EhOg6fbS1aBrUBw2mEmJgGEAAEwhIUwHZIgDkZiQAEVRkJRUATY9+D5VNTscDhtEhQowzdKdZiKEgJQSQgjEKTrsCfEQigJ7YjxU9Qx9bNOL9EFZcyTEBdZ1hhpKOhYlkZgYDyGAhKQEKEp42lOFgE3Rg34MxT0XfM8rFYlJFaCH8LwqqeJqOP1xlPS5XpoagnG2ukIdC6GoUOLtcJsmDNMM+TEUVjHOAVURqBjngK76jr1ougKvJ7T12lQN8XYNlaSjTGqKlLM9DodNh01VkKDbIl5XKBJ0GwSAeM0GTfm7ToemQ7GJ6BUWgjM9BquxKRoU+OJEuB+GODWUhbOZKMXwChmJV7MQvP/++5g+fTrWrVvnn7Znzx507doVn3/+OdLS0qJXHBERERFFVcz9iRAfH4/8/PyAaQW3ExISolESEREREcWImAuvderUwbFjx3D06FH/tD/++ANpaWlISkqKYmVEREREFG0xF15r1KiB5s2bY/LkycjNzcVff/2FOXPmICMjI9qlEREREVGUxdw5rwBw9OhRPPHEE9i4cSMURUHv3r0xatSoM39QhoiIiIj+EWIyvBIRERERFSfmThsgIiIiIjoThlciIiIisgyGVyIiIiKyDIZXIiIiIrIMhtcgbN++HYMHD0bLli3Rrl07jB49GllZWcUu+/nnn6NXr15o0qQJrr32WqxduzbC1VIwQhnTO++8Ew0bNkTTpk39P1988UWEK6ZgbNiwAX379kWzZs3Qrl07/Oc//4HT6Sx2WW6r1hDKmHJbtRbDMDBw4ECMHTv2jMtwO7WeYMa11NuqpLPKz8+X7dq1ky+88IJ0uVwyKytLDhkyRA4dOrTIsrt27ZINGzaUn3zyifR4PHL16tWyUaNG8uDBg1GonM4klDGVUspWrVrJjRs3RrhKClVmZqZs2LChfOedd6RhGPLQoUOyZ8+e8oUXXiiyLLdVawhlTKXktmo1zz//vKxfv74cM2ZMsfO5nVrTucZVytJvqzzyeg779+9H/fr1MWzYMNhsNiQnJ6Nfv3749ttviyy7fPlypKeno0uXLtA0Dd27d0eLFi3w5ptvRqFyOpNQxvSvv/5CTk4OLrvssihUSqGoXLkyvv76a/Tp0wdCCBw7dgwulwuVK1cusiy3VWsIZUy5rVrLhg0b8PHHH+Pqq68+4zLcTq0nmHENx7bK8HoONWvWxPz58wO+IOGjjz5CgwYNiiy7Y8cO1K1bN2Ba7dq1sX379jKvk4IXypj++OOPSEhIwMiRI9G6dWv07NkTy5Yti2S5FILExEQAQIcOHdCrVy+kpqaiT58+RZbjtmodwY4pt1XryMzMxPjx4zFt2jTExcWdcTlup9YS7LiGY1vVSlvsP4mUEs8//zzWrl2LJUuWFJl/8uTJIgPmcDiQl5cXqRIpROcaU7fbjSZNmmDkyJGoU6cONm7ciPvvvx8JCQm49tpro1AxBePjjz9GTk4ORo0aheHDh2P+/PkB87mtWs+5xpTbqjWYpomHHnoIgwcPRv369c+6LLdT6whlXMOxrfLIa5Byc3MxfPhwrFy5EkuWLEG9evWKLBMXF1fkgwROpxMJCQmRKpNCEMyY9u7dG/Pnz8dll10GXdfRvn179O7dG2vWrIlCxRQsh8OB888/Hw899BDWr1+PnJycgPncVq3nXGPKbdUa5s2bB5vNhoEDB55zWW6n1hHKuIZjW2V4DcKePXtw4403Ijc3F8uWLSs25ABA3bp18fvvvwdM27FjB+rUqROJMikEwY7psmXLimxQbrcbdrs9EmVSCLZu3YprrrkGbrfbP83tdkPX9SJHb7itWkMoY8pt1Rref/99bNq0Cenp6UhPT8eqVauwatUqpKenF1mW26l1hDKuYdlWS/xRr3+IY8eOyY4dO8qxY8dKwzDOuuyOHTtkw4YN5erVq/2fjGzYsKHcuXNnhKqlYIQypq+88ops06aN/Pnnn6VhGHLt2rWyUaNG8ttvv41QtRSs3Nxc2aFDBzl58mTpcrnk3r17ZUZGhnz88ceLLMtt1RpCGVNuq9Y0ZsyYM34qndupdZ1tXMOxrTK8nsPChQtl3bp1ZePGjWWTJk0CfqSUskmTJvL999/3L//FF1/I6667TjZp0kT26NFDrlu3Llql0xmEMqamacrZs2fLTp06yUaNGskePXrINWvWRLN8Oovff/9dDh48WKanp8tOnTrJ5557TrpcLiklt1WrCnZMua1a0+khh9tp+XC2cQ3HtiqklLK0h4uJiIiIiCKB57wSERERkWUwvBIRERGRZTC8EhEREZFlMLwSERERkWUwvBIRERGRZTC8EhEREZFlMLwSERERkWUwvBIRBeHEiRPIysqKdhlB+fPPP8tVO0REhTG8EpFl1KtXDxs3boxK2127dvV/z/q7776Lzp07h7yOe+65B1u2bAl3aQGWLl2KRx991H+7R48eWLFiRdjb+eWXX9CzZ8+wrxcA9uzZg/79+8Pj8ZTJ+onI2hheiYiCkJ2dXar7v/3224iPj0fz5s3DVFHxTj86vHr1alx33XVhb+fEiRNlFi4vuugitGjRAnPmzCmT9RORtTG8ElG5sXr1avTq1QvNmzdHnz598OWXX/rnDRw4ENOmTcOAAQPQtGlTXHvttfjggw/88/fu3Ys77rgDzZo1wzXXXINFixahXr16AIBu3boBAIYMGYKXX34ZAOD1ejF16lR07NgRzZo1wyOPPAKv11tsXW63G7NmzcJtt93mn/bzzz9j4MCBaNq0Kdq3b48XXngBBd/WXTCvRYsWuPrqq7Fo0SL/vJkzZ2L48OEYNWoU0tPTceWVV2LatGkAgOXLl2PevHnYvHkz0tPTAQCdO3fGu+++6++DGTNm4JZbbkGTJk1w3XXXYdu2bXjwwQfRrFkzdO7cGevWrStS4+l1/PXXXxgyZAgAoGnTpvjuu+8gpcTixYvRrVs3pKeno3///vjpp5/86+rcuTMee+wxtGvXDr1794bb7caECRPQrl07tGrVCv379w84Kt2/f3+8+uqrljlVg4giSBIRWUTdunXlN998U+y8devWyebNm8tNmzZJr9crP/vsM9mkSRP5v//9T0op5a233ipbtmwpf/75Z+lyueRzzz0nmzdvLp1Op/R6vbJ79+5y7Nix8uTJk3Lv3r3y+uuvl3Xr1i227XfeeUfWrVtXzps3T3o8Hvn777/Lxo0by5UrVxZb2wcffCC7dOniv52dnS1btmwpZ86cKV0ul9y9e7e88sor5RtvvCEPHjwomzdvLpcsWSLdbrf8/fffZdeuXeUbb7whpZRyxowZsl69enL58uXS6/XKdevWyXr16snvvvvOP//WW2/1t9WpUyf5zjvv+Pugbdu28vfff5cul0sOGDBANmjQQH7yySfS7XbLp556Snbu3FlKKc9ZxzfffBPQP0uWLJEdO3aUv/76q3S73fLtt9+W6enp8siRI/46rr/+epmTkyNzcnLksmXL5HXXXSdzcnKk1+uVzz33nOzVq1dAvw0cOFDOnz//bE8JIvoH4pFXIioXlixZgltuuQUtWrSAqqro1KkTOnfujP/7v//zL9OtWzdcdtllsNlsuOGGG3DixAlkZmbi+++/x59//olHH30U8fHxuOCCCzBy5MiztpeYmIghQ4ZA0zTUrl0b9evXx549e4pd9ptvvkGTJk38t9euXQu73Y5hw4bBZrPhoosuwiuvvIKOHTtixYoVqFWrFgYMGABd11G7dm3ccccdWLp0qf/+NWrUQO/evaGqKjp06IDU1NSgPzzVrVs31K5dGzabDenp6ahZsya6dOkCXddx5ZVXYt++fQAQVB2FLV26FEOHDkX9+vWh6zoyMjJQq1atgPNtu3XrhgoVKqBChQpwOBzYu3cvli1bhl27duGBBx4ocm5u06ZNsWHDhqAeFxH9c2jRLoCIKBz27duHTZs24Y033vBPMwwDrVu39t9OTU31/65pvt2faZo4ePAgkpOTER8f759/4YUXnrW9ihUrQgjhv63rOgzDKHbZAwcOoG7duv7bR44cQdWqVQPuX7NmTf/j+Pnnn/1v+xfUqKpqsY+joG3TNM9ab4FKlSr5f1dVFRUrVvTfVhTFf3pCMHUUtm/fPjz99NOYOnWqf5rX68Xll1/uv33eeef5f+/Rowc8Hg/efvttPPfcc0hJScHdd9+NW265xb9MWloa/vvf/wb1uIjon4PhlYjKhbS0NPTu3Rt33XWXf9r+/fvhcDjOed9q1aohKysL+fn5iIuL8983XBRFCQiXaWlpOHDgAKSU/gD76aefIjc3F2lpaWjVqhUWLFjgXz47OxsnT54MSy2FA/PZhFpHWloahg8fjh49evin7dmzJyAsF257165daNCgAXr37g2n04kPP/wQY8aMQXp6OurUqQPA98eHovANQiIKxL0CEVlKVlYWDh48GPDj9Xpx0003YfHixdi2bRsA4Mcff0SfPn2watWqc66zcePGqF27Np566ink5+fj0KFDmDFjRsAyNpsNJ06cKFHN1apVw6FDh/y3O3bsCK/Xi7lz58LtdmPPnj2YPHkyXC4XevXqhe+//x4rVqyA1+vF4cOHcffdd+Opp54Kqi273Y7c3Fz/EdSSOlcddrsdAPx9ctNNN+HFF1/EH3/8AQBYv349evTogW+//bbY9a9duxb33Xcf9u7dC4fDgUqVKkHTNCQlJfmXOXz4MKpVq1aqx0FE5Q+PvBKRpYwYMaLItA8++ADXXHMN8vLy8PDDD2P//v2oVKkSBg0ahIEDB55znYqiYMaMGXj88cfRpk0bpKWloXPnzvj111/9y/Tr1w8PPvggBg0ahIsvvjikmtu1a4dJkyb5b1eoUAELFizAlClT8MorryAuLg4DBgxAv379AADz58/H1KlT8eSTT0JVVXTs2BHjx48Pqq1OnTrhjTfeQPPmzQOuHBCqCy644Kx11K1bF82bN8cVV1yBF154AYMGDYKUEvfeey8OHz6M888/H4899hiuuuqqYtd/22234dChQ7j55puRm5uLCy64ANOnT0daWpp/mS1btqB79+4lfgxEVD4JWdo/z4mILM7pdOK7775Dy5Yt/ed0fvbZZ3j88cexfv36Uq/f7XbjqquuwqxZs9C4ceNSr++f4ODBg+jVqxc++ugjVK5cOdrlEFEM4WkDRPSPp+s6RowYgbfeegumaSIzMxMLFy5Ep06dwrJ+m82G4cOH45VXXgnL+v4JFi9ejIEDBzK4ElERDK9E9I+nqipmz56N5cuXo0WLFujVqxfq1KmDsWPHhq2NjIwM5OfnY/PmzWFbZ3m1Z88ebN26FXfffXe0SyGiGMTTBoiIiIjIMnjklYiIiIgsg+GViIiIiCyD4ZWIiIiILIPhlYiIiIgsg+GViIiIiCyD4ZWIiIiILIPhlYiIiIgsg+GViIiIiCyD4ZWIiIiILOP/AYz9g585mYBpAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 800x400 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(1,1, figsize=(8,4), dpi=100)\n",
    "df[df['target']==0].iloc[:,1].hist(alpha=0.5, label='Setosa')\n",
    "df[df['target']==1].iloc[:,1].hist(alpha=0.5, label='Virginica')\n",
    "df[df['target']==2].iloc[:,1].hist(alpha=0.5, label='Versicolour')\n",
    "plt.title('Sepal Width Distribution')\n",
    "plt.ylabel('Count')\n",
    "plt.xlabel('Length (centimeters)')\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Preprocessing"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We will be splitting the dataset into train and test sets so that we can evaluate the performance of our model later on."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-08-15T19:13:12.461949Z",
     "iopub.status.busy": "2021-08-15T19:13:12.460949Z",
     "iopub.status.idle": "2021-08-15T19:13:12.533966Z",
     "shell.execute_reply": "2021-08-15T19:13:12.532964Z",
     "shell.execute_reply.started": "2021-08-15T19:13:12.461949Z"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Original Size | Input: (150, 4), Output: (150,)\n",
      "Train Set Size | Input: (120, 4), Output: (120,)\n",
      "Test Set Size | Input: (30, 4), Output: (30,)\n"
     ]
    }
   ],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)\n",
    "\n",
    "print(f\"Original Size | Input: {X.shape}, Output: {y.shape}\")\n",
    "print(f\"Train Set Size | Input: {X_train.shape}, Output: {y_train.shape}\")\n",
    "print(f\"Test Set Size | Input: {X_test.shape}, Output: {y_test.shape}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-08-15T19:13:12.535967Z",
     "iopub.status.busy": "2021-08-15T19:13:12.535967Z",
     "iopub.status.idle": "2021-08-15T19:13:12.563974Z",
     "shell.execute_reply": "2021-08-15T19:13:12.562972Z",
     "shell.execute_reply.started": "2021-08-15T19:13:12.535967Z"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>sepal_length</th>\n",
       "      <th>sepal_width</th>\n",
       "      <th>petal_length</th>\n",
       "      <th>petal_width</th>\n",
       "      <th>target</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>4.6</td>\n",
       "      <td>3.6</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.2</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>5.7</td>\n",
       "      <td>4.4</td>\n",
       "      <td>1.5</td>\n",
       "      <td>0.4</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>6.7</td>\n",
       "      <td>3.1</td>\n",
       "      <td>4.4</td>\n",
       "      <td>1.4</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4.8</td>\n",
       "      <td>3.4</td>\n",
       "      <td>1.6</td>\n",
       "      <td>0.2</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>4.4</td>\n",
       "      <td>3.2</td>\n",
       "      <td>1.3</td>\n",
       "      <td>0.2</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   sepal_length  sepal_width  petal_length  petal_width  target\n",
       "0           4.6          3.6           1.0          0.2       0\n",
       "1           5.7          4.4           1.5          0.4       0\n",
       "2           6.7          3.1           4.4          1.4       1\n",
       "3           4.8          3.4           1.6          0.2       0\n",
       "4           4.4          3.2           1.3          0.2       0"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_set = np.concatenate((X_train, y_train.reshape(-1,1)), axis=1)\n",
    "df_train = pd.DataFrame(train_set, columns=['sepal_length', 'sepal_width', 'petal_length', 'petal_width', 'target'])\n",
    "# df_train = pd.DataFrame(train_set, columns=columns)\n",
    "df_train['target'] = df_train['target'].astype(int)\n",
    "\n",
    "test_set = np.concatenate((X_test, y_test.reshape(-1,1)), axis=1)\n",
    "df_test = pd.DataFrame(test_set, columns=['sepal_length', 'sepal_width', 'petal_length', 'petal_width', 'target'])\n",
    "# df_test = pd.DataFrame(test_set, columns=columns)\n",
    "df_test['target'] = df_test['target'].astype(int)\n",
    "\n",
    "df_train.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-08-15T19:13:12.565973Z",
     "iopub.status.busy": "2021-08-15T19:13:12.565973Z",
     "iopub.status.idle": "2021-08-15T19:13:13.059083Z",
     "shell.execute_reply": "2021-08-15T19:13:13.058082Z",
     "shell.execute_reply.started": "2021-08-15T19:13:12.565973Z"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 800x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(8,6), dpi=100)\n",
    "\n",
    "sns.scatterplot(data=df_train, x='sepal_length', y='sepal_width', hue='target', style='target', palette=\"Set2\")\n",
    "sns.scatterplot(data=df_test, x='sepal_length', y='sepal_width', color='black', label='unlabeled')\n",
    "\n",
    "plt.title('Train + Unlabeled Set Data Points')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Modeling using sklearn\n",
    "\n",
    "**What is sklearn?**\n",
    "> \"Scikit-learn is an open source machine learning library that supports supervised and unsupervised learning. It also provides various tools for model fitting, data preprocessing, model selection and evaluation, and many other utilities.\" \n",
    "\n",
    "For more details, please refer to <a href=\"https://scikit-learn.org/stable/getting_started.html\">sklearn website</a>.\n",
    "\n",
    "sklearn has classification and regression implementations for many machine learning models. We'll be using them in the following examples. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In practice, we often use the `sklearn` library to do machine learning. We can do train and evaluate an ML model with the following lines of code:\n",
    "\n",
    "### **TL;DR Code**\n",
    "\n",
    "```python\n",
    "model = ... # use appropriate model and set necessary hyperparameters.\n",
    "model.fit(X_train, y_train) # train the model using train set.\n",
    "model.score(X_test, y_test) # predict the test set and evaluate the results.\n",
    "```\n",
    "\n",
    "This is applicable to almost all (if not all) ML algorithms in sklearn. Note that this is the simplest case and a naive way of doing things. This could work if you have a very easy problem/dataset. Normally, you will be doing other things such as preprocessing, normalization, feature engineering, and hyperparameter optimization in between."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Classification\n",
    "\n",
    "Let's do classification using sklearn"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Model Training"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-08-15T19:13:13.061083Z",
     "iopub.status.busy": "2021-08-15T19:13:13.060083Z",
     "iopub.status.idle": "2021-08-15T19:13:13.239123Z",
     "shell.execute_reply": "2021-08-15T19:13:13.238123Z",
     "shell.execute_reply.started": "2021-08-15T19:13:13.061083Z"
    }
   },
   "outputs": [],
   "source": [
    "from sklearn.neighbors import KNeighborsClassifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-08-15T19:13:13.241125Z",
     "iopub.status.busy": "2021-08-15T19:13:13.240124Z",
     "iopub.status.idle": "2021-08-15T19:13:13.254126Z",
     "shell.execute_reply": "2021-08-15T19:13:13.253126Z",
     "shell.execute_reply.started": "2021-08-15T19:13:13.241125Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "KNeighborsClassifier(n_jobs=-1, n_neighbors=3)"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "clf = KNeighborsClassifier(n_neighbors=3, p=2, n_jobs=-1) # instantiate model\n",
    "clf.fit(X_train, y_train) # train the model"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Prediction"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-08-15T19:13:13.255127Z",
     "iopub.status.busy": "2021-08-15T19:13:13.255127Z",
     "iopub.status.idle": "2021-08-15T19:13:13.299139Z",
     "shell.execute_reply": "2021-08-15T19:13:13.298139Z",
     "shell.execute_reply.started": "2021-08-15T19:13:13.255127Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([1, 0, 2, 1, 1])"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_pred = clf.predict(X_test) # predict the corresponding labels given the trained model and input matrix\n",
    "y_pred[:5]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Evaluation\n",
    "\n",
    "By default, this uses the accuracy metric."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-08-15T19:13:13.300137Z",
     "iopub.status.busy": "2021-08-15T19:13:13.300137Z",
     "iopub.status.idle": "2021-08-15T19:13:13.329145Z",
     "shell.execute_reply": "2021-08-15T19:13:13.328144Z",
     "shell.execute_reply.started": "2021-08-15T19:13:13.300137Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test Set Score: 100.00%\n"
     ]
    }
   ],
   "source": [
    "score = clf.score(X_test, y_test) # predicts and evaluates in one line\n",
    "print(f\"Test Set Score: {score:.2%}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Since this is an easy dataset, we were able to get very high accuracies for both train and test set.\n",
    "\n",
    "Let's also compute for other classification metrics."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-08-15T19:13:13.331143Z",
     "iopub.status.busy": "2021-08-15T19:13:13.330144Z",
     "iopub.status.idle": "2021-08-15T19:13:13.344146Z",
     "shell.execute_reply": "2021-08-15T19:13:13.343146Z",
     "shell.execute_reply.started": "2021-08-15T19:13:13.331143Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       1.00      1.00      1.00        10\n",
      "           1       1.00      1.00      1.00         9\n",
      "           2       1.00      1.00      1.00        11\n",
      "\n",
      "    accuracy                           1.00        30\n",
      "   macro avg       1.00      1.00      1.00        30\n",
      "weighted avg       1.00      1.00      1.00        30\n",
      "\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import classification_report\n",
    "\n",
    "print(classification_report(y_test, y_pred))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Visualize our predictions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-08-15T19:13:13.346147Z",
     "iopub.status.busy": "2021-08-15T19:13:13.345147Z",
     "iopub.status.idle": "2021-08-15T19:13:13.389157Z",
     "shell.execute_reply": "2021-08-15T19:13:13.388157Z",
     "shell.execute_reply.started": "2021-08-15T19:13:13.346147Z"
    }
   },
   "outputs": [],
   "source": [
    "y_pred = clf.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-08-15T19:13:13.391157Z",
     "iopub.status.busy": "2021-08-15T19:13:13.390158Z",
     "iopub.status.idle": "2021-08-15T19:13:14.125479Z",
     "shell.execute_reply": "2021-08-15T19:13:14.124478Z",
     "shell.execute_reply.started": "2021-08-15T19:13:13.391157Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1400x600 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, (ax1, ax2) = plt.subplots(1, 2, figsize=(14,6), dpi=100)\n",
    "\n",
    "pred_set = np.concatenate((X_test, np.array(y_pred).reshape(-1,1)), axis=1)\n",
    "df_pred = pd.DataFrame(pred_set, columns=['sepal_length', 'sepal_width', 'petal_length', 'petal_width', 'target'])\n",
    "df_pred['target'] = df_pred['target'].astype(int)\n",
    "\n",
    "df_pred = pd.concat([df_train, df_pred])\n",
    "\n",
    "sns.scatterplot(data=df_train, x='sepal_length', y='sepal_width', hue='target', style='target', palette=\"Set2\", ax=ax1)\n",
    "sns.scatterplot(data=df_test, x='sepal_length', y='sepal_width', color='black', label='unlabeled', palette=\"Set2\", ax=ax1)\n",
    "sns.scatterplot(data=df_pred, x='sepal_length', y='sepal_width', hue='target', style='target', palette=\"Set2\", ax=ax2)\n",
    "\n",
    "ax1.set_title('Train + Unlabeled Set Data Points')\n",
    "ax2.set_title('Train + Predicted Set Data Points')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Spam Classifier\n",
    "\n",
    "Let's try to identify which emails are spam based on the content."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-08-15T19:13:14.127479Z",
     "iopub.status.busy": "2021-08-15T19:13:14.126479Z",
     "iopub.status.idle": "2021-08-15T19:13:16.376718Z",
     "shell.execute_reply": "2021-08-15T19:13:16.375719Z",
     "shell.execute_reply.started": "2021-08-15T19:13:14.127479Z"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>label</th>\n",
       "      <th>text</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>ham</td>\n",
       "      <td>Subject: enron methanol ; meter # : 988291\\nth...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>ham</td>\n",
       "      <td>Subject: hpl nom for january 9 , 2001\\n( see a...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>ham</td>\n",
       "      <td>Subject: neon retreat\\nho ho ho , we ' re arou...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>spam</td>\n",
       "      <td>Subject: photoshop , windows , office . cheap ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>ham</td>\n",
       "      <td>Subject: re : indian springs\\nthis deal is to ...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  label                                               text\n",
       "0   ham  Subject: enron methanol ; meter # : 988291\\nth...\n",
       "1   ham  Subject: hpl nom for january 9 , 2001\\n( see a...\n",
       "2   ham  Subject: neon retreat\\nho ho ho , we ' re arou...\n",
       "3  spam  Subject: photoshop , windows , office . cheap ...\n",
       "4   ham  Subject: re : indian springs\\nthis deal is to ..."
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_spam = pd.read_csv('https://raw.githubusercontent.com/Cyntwikip/data-repository/main/spam_ham_dataset.csv')\n",
    "df_spam = df_spam[['label', 'text']]\n",
    "df_spam.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-08-15T19:13:16.378720Z",
     "iopub.status.busy": "2021-08-15T19:13:16.378720Z",
     "iopub.status.idle": "2021-08-15T19:13:16.572808Z",
     "shell.execute_reply": "2021-08-15T19:13:16.571807Z",
     "shell.execute_reply.started": "2021-08-15T19:13:16.378720Z"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 500x300 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(1,1, figsize=(5,3), dpi=100)\n",
    "df_spam['label'].value_counts().plot(kind='bar')\n",
    "plt.title('Spam distribution')\n",
    "plt.xticks(rotation=0)\n",
    "plt.ylabel('Count')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Preprocessing\n",
    "\n",
    "We will preprocess our text data first so that it can be understood by our ML model."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-08-15T19:13:16.574808Z",
     "iopub.status.busy": "2021-08-15T19:13:16.574808Z",
     "iopub.status.idle": "2021-08-15T19:13:16.588811Z",
     "shell.execute_reply": "2021-08-15T19:13:16.586811Z",
     "shell.execute_reply.started": "2021-08-15T19:13:16.574808Z"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Original Size | Input: (5171,), Output: (5171,)\n",
      "Train Set Size | Input: (4136,), Output: (4136,)\n",
      "Test Set Size | Input: (1035,), Output: (1035,)\n"
     ]
    }
   ],
   "source": [
    "X = df_spam['text']\n",
    "y = df_spam['label']\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)\n",
    "\n",
    "print(f\"Original Size | Input: {X.shape}, Output: {y.shape}\")\n",
    "print(f\"Train Set Size | Input: {X_train.shape}, Output: {y_train.shape}\")\n",
    "print(f\"Test Set Size | Input: {X_test.shape}, Output: {y_test.shape}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-08-15T19:13:16.589811Z",
     "iopub.status.busy": "2021-08-15T19:13:16.589811Z",
     "iopub.status.idle": "2021-08-15T19:13:17.849775Z",
     "shell.execute_reply": "2021-08-15T19:13:17.848775Z",
     "shell.execute_reply.started": "2021-08-15T19:13:16.589811Z"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "from sklearn.feature_extraction.text import CountVectorizer\n",
    "\n",
    "corpus = X_train\n",
    "vec = CountVectorizer(stop_words='english', min_df=10)\n",
    "X_train_transformed = vec.fit_transform(X_train).todense()\n",
    "X_test_transformed = vec.transform(X_test).todense()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-08-15T19:13:17.851775Z",
     "iopub.status.busy": "2021-08-15T19:13:17.850776Z",
     "iopub.status.idle": "2021-08-15T19:13:17.864779Z",
     "shell.execute_reply": "2021-08-15T19:13:17.863781Z",
     "shell.execute_reply.started": "2021-08-15T19:13:17.851775Z"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "((4136, 3955), (1035, 3955))"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_train_transformed.shape, X_test_transformed.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Modeling"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-08-15T19:13:17.866779Z",
     "iopub.status.busy": "2021-08-15T19:13:17.866779Z",
     "iopub.status.idle": "2021-08-15T19:13:19.035052Z",
     "shell.execute_reply": "2021-08-15T19:13:19.034052Z",
     "shell.execute_reply.started": "2021-08-15T19:13:17.866779Z"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train Set Score: 99.76%\n"
     ]
    }
   ],
   "source": [
    "from sklearn.linear_model import LogisticRegression\n",
    "\n",
    "clf = LogisticRegression()\n",
    "clf.fit(X_train_transformed, y_train)\n",
    "score = clf.score(X_train_transformed, y_train) # predicts and evaluates in one line\n",
    "print(f\"Train Set Score: {score:.2%}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-08-15T19:13:19.037053Z",
     "iopub.status.busy": "2021-08-15T19:13:19.037053Z",
     "iopub.status.idle": "2021-08-15T19:13:19.080062Z",
     "shell.execute_reply": "2021-08-15T19:13:19.079061Z",
     "shell.execute_reply.started": "2021-08-15T19:13:19.037053Z"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test Set Score: 97.68%\n"
     ]
    }
   ],
   "source": [
    "score = clf.score(X_test_transformed, y_test) # predicts and evaluates in one line\n",
    "print(f\"Test Set Score: {score:.2%}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Not Spam (Ham) examples"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-08-15T19:13:19.082064Z",
     "iopub.status.busy": "2021-08-15T19:13:19.081061Z",
     "iopub.status.idle": "2021-08-15T19:13:19.095066Z",
     "shell.execute_reply": "2021-08-15T19:13:19.094064Z",
     "shell.execute_reply.started": "2021-08-15T19:13:19.082064Z"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1566    ham\n",
       "1235    ham\n",
       "2868    ham\n",
       "4903    ham\n",
       "1471    ham\n",
       "Name: label, dtype: object"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "hams = y_test[y_test=='ham'].iloc[:5]\n",
    "hams"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-08-15T19:13:19.097067Z",
     "iopub.status.busy": "2021-08-15T19:13:19.097067Z",
     "iopub.status.idle": "2021-08-15T19:13:19.125074Z",
     "shell.execute_reply": "2021-08-15T19:13:19.124073Z",
     "shell.execute_reply.started": "2021-08-15T19:13:19.097067Z"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1566    Subject: hpl nom for march 30 , 2001\\n( see at...\n",
       "1235    Subject: re : nom / actual volume for april 17...\n",
       "2868    Subject: re : meter 8740 dec 99\\nrobert and i ...\n",
       "4903    Subject: re : coastal oil & gas corporation\\nm...\n",
       "1471    Subject: enron / hpl actuals for july 27 , 200...\n",
       "Name: text, dtype: object"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sample = X_test[hams.index]\n",
    "sample"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-08-15T19:13:19.127073Z",
     "iopub.status.busy": "2021-08-15T19:13:19.127073Z",
     "iopub.status.idle": "2021-08-15T19:13:19.140075Z",
     "shell.execute_reply": "2021-08-15T19:13:19.139075Z",
     "shell.execute_reply.started": "2021-08-15T19:13:19.127073Z"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array(['ham', 'ham', 'ham', 'ham', 'ham'], dtype=object)"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "clf.predict(vec.transform(sample))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Spam examples"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-08-15T19:13:19.145076Z",
     "iopub.status.busy": "2021-08-15T19:13:19.145076Z",
     "iopub.status.idle": "2021-08-15T19:13:19.155078Z",
     "shell.execute_reply": "2021-08-15T19:13:19.154078Z",
     "shell.execute_reply.started": "2021-08-15T19:13:19.145076Z"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1988    spam\n",
       "1662    spam\n",
       "3820    spam\n",
       "3732    spam\n",
       "4692    spam\n",
       "Name: label, dtype: object"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "spams = y_test[y_test=='spam'].iloc[:5]\n",
    "spams"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-08-15T19:13:19.157079Z",
     "iopub.status.busy": "2021-08-15T19:13:19.156080Z",
     "iopub.status.idle": "2021-08-15T19:13:19.170083Z",
     "shell.execute_reply": "2021-08-15T19:13:19.169083Z",
     "shell.execute_reply.started": "2021-08-15T19:13:19.157079Z"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1988    Subject: online pharxmacy 80 % off all meds\\nd...\n",
       "1662    Subject: wow , , , , cash out @ 4 . 19 %\\nwher...\n",
       "3820    Subject: awesome movies of the super sexy jaim...\n",
       "3732    Subject: ad - when he said he could make me fa...\n",
       "4692    Subject: upgrade your os to xp pro for $ 50\\nw...\n",
       "Name: text, dtype: object"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sample = X_test[spams.index]\n",
    "sample"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-08-15T19:13:19.173082Z",
     "iopub.status.busy": "2021-08-15T19:13:19.172082Z",
     "iopub.status.idle": "2021-08-15T19:13:19.200089Z",
     "shell.execute_reply": "2021-08-15T19:13:19.199088Z",
     "shell.execute_reply.started": "2021-08-15T19:13:19.173082Z"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array(['spam', 'spam', 'spam', 'spam', 'spam'], dtype=object)"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "clf.predict(vec.transform(sample))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Regression\n",
    "\n",
    "Regression is very similar to classification in sklearn.\n",
    "\n",
    "We'll be using another dataset for the regression task since we need continuous values as labels / outputs / targets. One readily-available dataset for this is the **Boston dataset**. It contains the house pricing data in Boston as well as the following features:\n",
    "\n",
    " - CRIM per capita crime rate by town\n",
    " - ZN proportion of residential land zoned for lots over 25,000 sq.ft.\n",
    " - INDUS proportion of non-retail business acres per town\n",
    " - CHAS Charles River dummy variable (= 1 if tract bounds river; 0 otherwise)\n",
    " - NOX nitric oxides concentration (parts per 10 million)\n",
    " - RM average number of rooms per dwelling\n",
    " - AGE proportion of owner-occupied units built prior to 1940\n",
    " - DIS weighted distances to five Boston employment centres\n",
    " - RAD index of accessibility to radial highways\n",
    " - TAX full-value property-tax rate per \\$10,000\n",
    " - PTRATIO pupil-teacher ratio by town\n",
    " - B 1000(Bk - 0.63)^2 where Bk is the proportion of blacks by town\n",
    " - LSTAT % lower status of the population\n",
    " - MEDV Median value of owner-occupied homes in $1000’s\n",
    "\n",
    "For more info, please see <a href=\"https://scikit-learn.org/stable/datasets/index.html#boston-dataset\">this link</a>.\n",
    "\n",
    "And, as usual, we will also be splitting the dataset into train and test sets."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-08-15T19:13:19.202090Z",
     "iopub.status.busy": "2021-08-15T19:13:19.202090Z",
     "iopub.status.idle": "2021-08-15T19:13:19.230095Z",
     "shell.execute_reply": "2021-08-15T19:13:19.229096Z",
     "shell.execute_reply.started": "2021-08-15T19:13:19.202090Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Original Size | Input: (506, 13), Output: (506,)\n",
      "Train Set Size | Input: (404, 13), Output: (404,)\n",
      "Test Set Size | Input: (102, 13), Output: (102,)\n"
     ]
    }
   ],
   "source": [
    "X, y = datasets.load_boston(return_X_y=True)\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)\n",
    "\n",
    "print(f\"Original Size | Input: {X.shape}, Output: {y.shape}\")\n",
    "print(f\"Train Set Size | Input: {X_train.shape}, Output: {y_train.shape}\")\n",
    "print(f\"Test Set Size | Input: {X_test.shape}, Output: {y_test.shape}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Model Training"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-08-15T19:13:19.232097Z",
     "iopub.status.busy": "2021-08-15T19:13:19.232097Z",
     "iopub.status.idle": "2021-08-15T19:13:19.245098Z",
     "shell.execute_reply": "2021-08-15T19:13:19.244098Z",
     "shell.execute_reply.started": "2021-08-15T19:13:19.232097Z"
    }
   },
   "outputs": [],
   "source": [
    "from sklearn.neighbors import KNeighborsRegressor"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-08-15T19:13:19.247098Z",
     "iopub.status.busy": "2021-08-15T19:13:19.246098Z",
     "iopub.status.idle": "2021-08-15T19:13:19.260102Z",
     "shell.execute_reply": "2021-08-15T19:13:19.259101Z",
     "shell.execute_reply.started": "2021-08-15T19:13:19.247098Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "KNeighborsRegressor(n_jobs=-1, n_neighbors=3)"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "reg = KNeighborsRegressor(n_neighbors=3, p=2, n_jobs=-1) # instantiate model\n",
    "reg.fit(X_train, y_train) # train the model"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Prediction"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-08-15T19:13:19.262103Z",
     "iopub.status.busy": "2021-08-15T19:13:19.261103Z",
     "iopub.status.idle": "2021-08-15T19:13:19.320116Z",
     "shell.execute_reply": "2021-08-15T19:13:19.319115Z",
     "shell.execute_reply.started": "2021-08-15T19:13:19.262103Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([25.1       , 32.76666667, 14.6       , 31.        , 17.13333333])"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_pred = reg.predict(X_test) # predict the corresponding labels given the trained model and input matrix\n",
    "y_pred[:5]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Evaluation\n",
    "\n",
    "By default, sklearn uses R-squared when evaluating regression tasks."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-08-15T19:13:19.322116Z",
     "iopub.status.busy": "2021-08-15T19:13:19.322116Z",
     "iopub.status.idle": "2021-08-15T19:13:19.350122Z",
     "shell.execute_reply": "2021-08-15T19:13:19.349122Z",
     "shell.execute_reply.started": "2021-08-15T19:13:19.322116Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test Set Score: 70.46%\n"
     ]
    }
   ],
   "source": [
    "score = reg.score(X_test, y_test) # predicts and evaluates in one line\n",
    "print(f\"Test Set Score: {score:.2%}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "editable": false,
    "tags": []
   },
   "source": [
    "## End\n",
    "<sup>made by **Jude Michael Teves**</sup> <br>\n",
    "<sup>for comments, corrections, suggestions, please email:</sup><sup> <href>judemichaelteves@gmail.com</href> or <href>jude.teves@dlsu.edu.ph</href></sup><br>"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.8"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}