2.2 - Logistic Regression Complete.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "# For Colab\n",
    "# !git clone https://github.com/trevor-yu-087/syde-599-f23-tutorial.git\n",
    "# DATA_PATH = \"syde-599-f23-tutorial/data/logistic-regression-data.pkl\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import torch\n",
    "import matplotlib.pyplot as plt\n",
    "import pickle"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "with open(\"data/logistic-regression-data.pkl\", \"rb\") as f:\n",
    "    data = pickle.load(f)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "dict_keys(['training_x', 'training_y', 'validation_x', 'validation_y'])"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.keys()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Section 1: Logistic Regression with Scikit-Learn\n",
    "In this tutorial, we will use scikit-learn Linear Regression to do binary classification. The goal is again to get familiar with the typical machine learning workflow."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "# By convention, 2D data (matrices) uses capital letters, while 1D data (vectors) uses lower case letters\n",
    "X_train = data[\"training_x\"]\n",
    "y_train = data[\"training_y\"]\n",
    "X_val = data[\"validation_x\"]\n",
    "y_val = data[\"validation_y\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# TODO: Plot features at index 2 and 3\n",
    "plt.scatter(X_train[:, 2], X_train[:, 3], c=y_train, cmap=\"bwr\")\n",
    "plt.xlabel(\"Feature 1\")\n",
    "plt.ylabel(\"Feature 2\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "[Scikit-learn Logistic Regression](https://scikit-learn.org/stable/modules/generated/sklearn.linear_model.LogisticRegression.html)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.linear_model import LogisticRegression\n",
    "# TODO: Instantiate model\n",
    "clf = LogisticRegression()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style>#sk-container-id-1 {color: black;}#sk-container-id-1 pre{padding: 0;}#sk-container-id-1 div.sk-toggleable {background-color: white;}#sk-container-id-1 label.sk-toggleable__label {cursor: pointer;display: block;width: 100%;margin-bottom: 0;padding: 0.3em;box-sizing: border-box;text-align: center;}#sk-container-id-1 label.sk-toggleable__label-arrow:before {content: \"▸\";float: left;margin-right: 0.25em;color: #696969;}#sk-container-id-1 label.sk-toggleable__label-arrow:hover:before {color: black;}#sk-container-id-1 div.sk-estimator:hover label.sk-toggleable__label-arrow:before {color: black;}#sk-container-id-1 div.sk-toggleable__content {max-height: 0;max-width: 0;overflow: hidden;text-align: left;background-color: #f0f8ff;}#sk-container-id-1 div.sk-toggleable__content pre {margin: 0.2em;color: black;border-radius: 0.25em;background-color: #f0f8ff;}#sk-container-id-1 input.sk-toggleable__control:checked~div.sk-toggleable__content {max-height: 200px;max-width: 100%;overflow: auto;}#sk-container-id-1 input.sk-toggleable__control:checked~label.sk-toggleable__label-arrow:before {content: \"▾\";}#sk-container-id-1 div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-1 div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-1 input.sk-hidden--visually {border: 0;clip: rect(1px 1px 1px 1px);clip: rect(1px, 1px, 1px, 1px);height: 1px;margin: -1px;overflow: hidden;padding: 0;position: absolute;width: 1px;}#sk-container-id-1 div.sk-estimator {font-family: monospace;background-color: #f0f8ff;border: 1px dotted black;border-radius: 0.25em;box-sizing: border-box;margin-bottom: 0.5em;}#sk-container-id-1 div.sk-estimator:hover {background-color: #d4ebff;}#sk-container-id-1 div.sk-parallel-item::after {content: \"\";width: 100%;border-bottom: 1px solid gray;flex-grow: 1;}#sk-container-id-1 div.sk-label:hover label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-1 div.sk-serial::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: 0;}#sk-container-id-1 div.sk-serial {display: flex;flex-direction: column;align-items: center;background-color: white;padding-right: 0.2em;padding-left: 0.2em;position: relative;}#sk-container-id-1 div.sk-item {position: relative;z-index: 1;}#sk-container-id-1 div.sk-parallel {display: flex;align-items: stretch;justify-content: center;background-color: white;position: relative;}#sk-container-id-1 div.sk-item::before, #sk-container-id-1 div.sk-parallel-item::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: -1;}#sk-container-id-1 div.sk-parallel-item {display: flex;flex-direction: column;z-index: 1;position: relative;background-color: white;}#sk-container-id-1 div.sk-parallel-item:first-child::after {align-self: flex-end;width: 50%;}#sk-container-id-1 div.sk-parallel-item:last-child::after {align-self: flex-start;width: 50%;}#sk-container-id-1 div.sk-parallel-item:only-child::after {width: 0;}#sk-container-id-1 div.sk-dashed-wrapped {border: 1px dashed gray;margin: 0 0.4em 0.5em 0.4em;box-sizing: border-box;padding-bottom: 0.4em;background-color: white;}#sk-container-id-1 div.sk-label label {font-family: monospace;font-weight: bold;display: inline-block;line-height: 1.2em;}#sk-container-id-1 div.sk-label-container {text-align: center;}#sk-container-id-1 div.sk-container {/* jupyter's `normalize.less` sets `[hidden] { display: none; }` but bootstrap.min.css set `[hidden] { display: none !important; }` so we also need the `!important` here to be able to override the default hidden behavior on the sphinx rendered scikit-learn.org. See: https://github.com/scikit-learn/scikit-learn/issues/21755 */display: inline-block !important;position: relative;}#sk-container-id-1 div.sk-text-repr-fallback {display: none;}</style><div id=\"sk-container-id-1\" class=\"sk-top-container\"><div class=\"sk-text-repr-fallback\"><pre>LogisticRegression()</pre><b>In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook. <br />On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.</b></div><div class=\"sk-container\" hidden><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-1\" type=\"checkbox\" checked><label for=\"sk-estimator-id-1\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">LogisticRegression</label><div class=\"sk-toggleable__content\"><pre>LogisticRegression()</pre></div></div></div></div></div>"
      ],
      "text/plain": [
       "LogisticRegression()"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# TODO: Fit model on training data\n",
    "clf.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training accuracy: 94.67%\n",
      "Validation accuracy: 92.00%\n"
     ]
    }
   ],
   "source": [
    "# TODO: Score model on the training and validation sets\n",
    "train_acc = clf.score(X_train, y_train)\n",
    "val_acc = clf.score(X_val, y_val)\n",
    "\n",
    "print(f\"Training accuracy: {train_acc*100:.2f}%\")\n",
    "print(f\"Validation accuracy: {val_acc*100:.2f}%\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[0.76571008, 0.23428992],\n",
       "       [0.03807589, 0.96192411],\n",
       "       [0.69069949, 0.30930051],\n",
       "       [0.00828769, 0.99171231],\n",
       "       [0.06775818, 0.93224182],\n",
       "       [0.02522851, 0.97477149],\n",
       "       [0.79628948, 0.20371052],\n",
       "       [0.4505042 , 0.5494958 ],\n",
       "       [0.03493644, 0.96506356],\n",
       "       [0.12822052, 0.87177948],\n",
       "       [0.12226014, 0.87773986],\n",
       "       [0.24500328, 0.75499672],\n",
       "       [0.06583504, 0.93416496],\n",
       "       [0.77271386, 0.22728614],\n",
       "       [0.99475053, 0.00524947],\n",
       "       [0.980599  , 0.019401  ],\n",
       "       [0.60895504, 0.39104496],\n",
       "       [0.93092402, 0.06907598],\n",
       "       [0.41005703, 0.58994297],\n",
       "       [0.86877299, 0.13122701],\n",
       "       [0.99480335, 0.00519665],\n",
       "       [0.0995938 , 0.9004062 ],\n",
       "       [0.94864129, 0.05135871],\n",
       "       [0.2930718 , 0.7069282 ],\n",
       "       [0.94638977, 0.05361023],\n",
       "       [0.99724104, 0.00275896],\n",
       "       [0.71764081, 0.28235919],\n",
       "       [0.31325143, 0.68674857],\n",
       "       [0.05300085, 0.94699915],\n",
       "       [0.51288567, 0.48711433],\n",
       "       [0.27184233, 0.72815767],\n",
       "       [0.04140487, 0.95859513],\n",
       "       [0.04055352, 0.95944648],\n",
       "       [0.00225712, 0.99774288],\n",
       "       [0.24500328, 0.75499672],\n",
       "       [0.49671469, 0.50328531],\n",
       "       [0.79808161, 0.20191839],\n",
       "       [0.62986753, 0.37013247],\n",
       "       [0.23494226, 0.76505774],\n",
       "       [0.06400147, 0.93599853],\n",
       "       [0.84476947, 0.15523053],\n",
       "       [0.19225612, 0.80774388],\n",
       "       [0.96534088, 0.03465912],\n",
       "       [0.8189143 , 0.1810857 ],\n",
       "       [0.79723412, 0.20276588],\n",
       "       [0.90597118, 0.09402882],\n",
       "       [0.97214523, 0.02785477],\n",
       "       [0.89845167, 0.10154833],\n",
       "       [0.9461224 , 0.0538776 ],\n",
       "       [0.85929555, 0.14070445],\n",
       "       [0.17347015, 0.82652985],\n",
       "       [0.01218125, 0.98781875],\n",
       "       [0.83405832, 0.16594168],\n",
       "       [0.78465766, 0.21534234],\n",
       "       [0.81439058, 0.18560942],\n",
       "       [0.0107057 , 0.9892943 ],\n",
       "       [0.9046239 , 0.0953761 ],\n",
       "       [0.79787839, 0.20212161],\n",
       "       [0.84020061, 0.15979939],\n",
       "       [0.04157488, 0.95842512],\n",
       "       [0.9906126 , 0.0093874 ],\n",
       "       [0.9880344 , 0.0119656 ],\n",
       "       [0.16580638, 0.83419362],\n",
       "       [0.03099467, 0.96900533],\n",
       "       [0.15579257, 0.84420743],\n",
       "       [0.01967089, 0.98032911],\n",
       "       [0.97371782, 0.02628218],\n",
       "       [0.232481  , 0.767519  ],\n",
       "       [0.98436478, 0.01563522],\n",
       "       [0.13223804, 0.86776196],\n",
       "       [0.97639856, 0.02360144],\n",
       "       [0.99366019, 0.00633981],\n",
       "       [0.94536349, 0.05463651],\n",
       "       [0.21491296, 0.78508704],\n",
       "       [0.07493247, 0.92506753]])"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# TODO: Get the probabilities for class 0 and class 1\n",
    "y_hat = clf.predict_proba(X_train)\n",
    "y_hat"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAkAAAAHHCAYAAABXx+fLAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/NK7nSAAAACXBIWXMAAA9hAAAPYQGoP6dpAAA70ElEQVR4nO3de1hVVf7H8c8R4aAJeEMuRiqat7yVpqGZWhZaY1pNF6cSzawMJ5VfllQKWBOO3awkrSalyUyrUWvSLCWxTK28UGnqeEHREUhNQbwAwvr9MQ+njlwUPMCB/X49z34e99prrfPda1I+s8/iHJsxxggAAMBC6lR3AQAAAFWNAAQAACyHAAQAACyHAAQAACyHAAQAACyHAAQAACyHAAQAACyHAAQAACyHAAQAACyHAASg2o0cOVItW7as0NjY2FjZbDbXFnSR+vfvr/79+1d3GQDKQAACUCqbzXZBR3JycnWXWiucOnVKsbGxrCdQBWx8FxiA0syfP9/p/J///KdWrlyp9957z6n9xhtvVEBAQIVfJz8/X4WFhbLb7eUee/bsWZ09e1be3t4Vfn1XK3r6U94gc+TIEfn7+ysmJkaxsbEurwvA7+pWdwEA3Nd9993ndL5hwwatXLmyWPu5Tp06pfr161/w63h6elaoPkmqW7eu6tblnzIA5cNbYAAuSv/+/dWpUydt2rRJ1113nerXr6+nnnpKkvTJJ5/olltuUXBwsOx2u1q3bq1nn31WBQUFTnOcuwdo3759stlsevHFF/XWW2+pdevWstvtuvrqq/XDDz84jS1pD5DNZtO4ceO0dOlSderUSXa7XVdccYVWrFhRrP7k5GT16NFD3t7eat26td58881y7Ssqqq9evXrq2bOnvvnmm2J98vLyNHXqVHXv3l1+fn665JJL1LdvX61evdrpnv39/SVJcXFxjrcXi54E/fTTTxo5cqRCQ0Pl7e2twMBAPfDAAzp69OgF1QnAGf+3CcBFO3r0qAYPHqx77rlH9913n+PtsMTERDVo0EBRUVFq0KCBvvrqK02dOlXZ2dl64YUXzjvvggULdOLECT388MOy2WyaMWOGbr/9du3du/e8T43Wrl2rxYsX69FHH5WPj49ee+013XHHHUpLS1OTJk0kSVu2bNGgQYMUFBSkuLg4FRQUaNq0aY4gcj7vvPOOHn74YfXu3VsTJkzQ3r17deutt6px48YKCQlx9MvOztY//vEPDR8+XGPGjNGJEyf0zjvvKDw8XN9//726desmf39/zZ49W2PHjtVtt92m22+/XZLUpUsXSdLKlSu1d+9ejRo1SoGBgdq2bZveeustbdu2TRs2bHC7jeCA2zMAcIEiIyPNuf9s9OvXz0gyc+bMKdb/1KlTxdoefvhhU79+fXPmzBlHW0REhGnRooXjPDU11UgyTZo0Mb/99puj/ZNPPjGSzL///W9HW0xMTLGaJBkvLy+ze/duR9uPP/5oJJnXX3/d0TZkyBBTv35989///tfRtmvXLlO3bt1ic54rLy/PNGvWzHTr1s3k5uY62t966y0jyfTr18/RdvbsWac+xhhz7NgxExAQYB544AFH2+HDh40kExMTU+z1SlrLDz74wEgyX3/9dZm1AiiOt8AAXDS73a5Ro0YVa69Xr57jzydOnNCRI0fUt29fnTp1Sjt27DjvvHfffbcaNWrkOO/bt68kae/evecdO3DgQLVu3dpx3qVLF/n6+jrGFhQUaNWqVRo2bJiCg4Md/dq0aaPBgwefd/6NGzfq119/1SOPPCIvLy9H+8iRI+Xn5+fU18PDw9GnsLBQv/32m86ePasePXpo8+bN530tyXktz5w5oyNHjuiaa66RpAueA8DvCEAALlrz5s2dQkCRbdu26bbbbpOfn598fX3l7+/v2ECdlZV13nkvu+wyp/OiMHTs2LFyjy0aXzT2119/1enTp9WmTZti/UpqO9f+/fslSZdffrlTu6enp0JDQ4v1f/fdd9WlSxd5e3urSZMm8vf317Jlyy5oHSTpt99+0/jx4xUQEKB69erJ399frVq1knRhawnAGXuAAFy0Pz6dKHL8+HH169dPvr6+mjZtmlq3bi1vb29t3rxZTz75pAoLC887r4eHR4nt5gI+veNixrra/PnzNXLkSA0bNkyTJk1Ss2bN5OHhofj4eO3Zs+eC5rjrrru0bt06TZo0Sd26dVODBg1UWFioQYMGXdBaAnBGAAJQKZKTk3X06FEtXrxY1113naM9NTW1Gqv6XbNmzeTt7a3du3cXu1ZS27latGghSdq1a5euv/56R3t+fr5SU1PVtWtXR9vHH3+s0NBQLV682GmzckxMjNOcpW1kPnbsmJKSkhQXF6epU6c62nft2nXeOgGUjLfAAFSKoicwf3zikpeXpzfeeKO6SnLi4eGhgQMHaunSpTp06JCjfffu3fr888/PO75Hjx7y9/fXnDlzlJeX52hPTEzU8ePHi72W5LwW3333ndavX+/Ur+izky5kvCTNnDnzvHUCKBlPgABUit69e6tRo0aKiIjQY489JpvNpvfee69a3oIqTWxsrL788kv16dNHY8eOVUFBgWbNmqVOnTopJSWlzLGenp567rnn9PDDD+v666/X3XffrdTUVM2bN6/YHqA//elPWrx4sW677TbdcsstSk1N1Zw5c9SxY0fl5OQ4+tWrV08dO3bUokWL1LZtWzVu3FidOnVSp06ddN1112nGjBnKz89X8+bN9eWXX7rN0zSgJuIJEIBK0aRJE3322WcKCgrSM888oxdffFE33nijZsyYUd2lOXTv3l2ff/65GjVqpClTpuidd97RtGnTdMMNN1zQV2s89NBDeuONN3To0CFNmjRJ33zzjT799FOnzwCS/vebYc8//7x+/PFHPfbYY/riiy80f/589ejRo9ic//jHP9S8eXNNnDhRw4cP18cffyzpf5+JFB4eroSEBEVHR8vT0/OCnlQBKBnfBQYA5xg2bJi2bdvGHhugFuMJEABLO336tNP5rl27tHz5cscXmgKonXgCBMDSgoKCHN+xtX//fs2ePVu5ubnasmVLsc/4AVB7sAkagKUNGjRIH3zwgTIyMmS32xUWFqbnn3+e8APUcjwBAgAAlsMeIAAAYDkEIAAAYDnsASpBYWGhDh06JB8fn1I/mh4AALgXY4xOnDih4OBg1alT9jMeAlAJDh06VOyDzAAAQM1w4MABXXrppWX2IQCVwMfHR9L/FtDX17eaqwEAABciOztbISEhjp/jZSEAlaDobS9fX18CEAAANcyFbF9hEzQAALAcAhAAALAcAhAAALAcAhAAALAcAhAAALAcAhAAALAcAhAAALAcAhAAALAcAhAAALAcAhAAALCcag1A8fHxuvrqq+Xj46NmzZpp2LBh2rlzp1OfM2fOKDIyUk2aNFGDBg10xx13KDMzs8x5jTGaOnWqgoKCVK9ePQ0cOFC7du2qzFsBAAA1SLUGoDVr1igyMlIbNmzQypUrlZ+fr5tuukknT5509Jk4caL+/e9/66OPPtKaNWt06NAh3X777WXOO2PGDL322muaM2eOvvvuO11yySUKDw/XmTNnKvuWAABADWAzxpjqLqLI4cOH1axZM61Zs0bXXXedsrKy5O/vrwULFujPf/6zJGnHjh3q0KGD1q9fr2uuuabYHMYYBQcH6//+7//0+OOPS5KysrIUEBCgxMRE3XPPPeetIzs7W35+fsrKyuLLUAEAqCHK8/PbrfYAZWVlSZIaN24sSdq0aZPy8/M1cOBAR5/27dvrsssu0/r160ucIzU1VRkZGU5j/Pz81KtXr1LHAAAAa6lb3QUUKSws1IQJE9SnTx916tRJkpSRkSEvLy81bNjQqW9AQIAyMjJKnKeoPSAg4ILH5ObmKjc313GenZ1d0dsAAAA1gNsEoMjISG3dulVr166t8teOj49XXFxclb/uRYv1c9E8Wa6ZBwCAGsIt3gIbN26cPvvsM61evVqXXnqpoz0wMFB5eXk6fvy4U//MzEwFBgaWOFdR+7m/KVbWmOjoaGVlZTmOAwcOXMTdAAAAd1etAcgYo3HjxmnJkiX66quv1KpVK6fr3bt3l6enp5KSkhxtO3fuVFpamsLCwkqcs1WrVgoMDHQak52dre+++67UMXa7Xb6+vk4HAACovao1AEVGRmr+/PlasGCBfHx8lJGRoYyMDJ0+fVrS/zYvjx49WlFRUVq9erU2bdqkUaNGKSwszOk3wNq3b68lS5ZIkmw2myZMmKDnnntOn376qX7++WeNGDFCwcHBGjZsWHXcJgAAcDPVugdo9uzZkqT+/fs7tc+bN08jR46UJL3yyiuqU6eO7rjjDuXm5io8PFxvvPGGU/+dO3c6foNMkp544gmdPHlSDz30kI4fP65rr71WK1askLe3d6XeDwAAqBnc6nOA3EWN+RwgNkEDAOBQYz8HCAAAoCoQgAAAgOUQgAAAgOUQgAAAgOUQgAAAgOUQgAAAgOUQgAAAgOUQgAAAgOUQgAAAgOUQgAAAgOUQgAAAgOUQgAAAgOUQgAAAgOUQgAAAgOUQgAAAgOUQgAAAgOUQgAAAgOUQgAAAgOUQgAAAgOUQgAAAgOUQgAAAgOUQgAAAgOUQgAAAgOUQgAAAgOUQgAAAgOUQgAAAgOUQgAAAgOUQgAAAgOUQgAAAgOXUre4CAABA5Wo5eZlL5tk3/RaXzOMOeAIEAAAshwAEAAAshwAEAAAsp1oD0Ndff60hQ4YoODhYNptNS5cudbpus9lKPF544YVS54yNjS3Wv3379pV8JwAAoCap1gB08uRJde3aVQkJCSVeT09Pdzrmzp0rm82mO+64o8x5r7jiCqdxa9eurYzyAQBADVWtvwU2ePBgDR48uNTrgYGBTueffPKJBgwYoNDQ0DLnrVu3brGxAAAARWrMHqDMzEwtW7ZMo0ePPm/fXbt2KTg4WKGhobr33nuVlpZWBRUCAICaosZ8DtC7774rHx8f3X777WX269WrlxITE9WuXTulp6crLi5Offv21datW+Xj41PimNzcXOXm5jrOs7OzXVo7AABwLzUmAM2dO1f33nuvvL29y+z3x7fUunTpol69eqlFixb68MMPS316FB8fr7i4OJfWCwAA3FeNeAvsm2++0c6dO/Xggw+We2zDhg3Vtm1b7d69u9Q+0dHRysrKchwHDhy4mHIBAICbqxEB6J133lH37t3VtWvXco/NycnRnj17FBQUVGofu90uX19fpwMAANRe1RqAcnJylJKSopSUFElSamqqUlJSnDYtZ2dn66OPPir16c8NN9ygWbNmOc4ff/xxrVmzRvv27dO6det02223ycPDQ8OHD6/UewEAADVHte4B2rhxowYMGOA4j4qKkiRFREQoMTFRkrRw4UIZY0oNMHv27NGRI0cc5wcPHtTw4cN19OhR+fv769prr9WGDRvk7+9feTcCAABqFJsxxlR3Ee4mOztbfn5+ysrKcu+3w2L9XDRPlmvmAQC4Jat8G3x5fn7XiD1AAAAArkQAAgAAlkMAAgAAlkMAAgAAlkMAAgAAlkMAAgAAlkMAAgAAlkMAAgAAlkMAAgAAlkMAAgAAlkMAAgAAlkMAAgAAlkMAAgAAlkMAAgAAlkMAAgAAlkMAAgAAlkMAAgAAlkMAAgAAlkMAAgAAlkMAAgAAlkMAAgAAlkMAAgAAlkMAAgAAlkMAAgAAlkMAAgAAlkMAAgAAlkMAAgAAlkMAAgAAlkMAAgAAlkMAAgAAlkMAAgAAlkMAAgAAlkMAAgAAllOtAejrr7/WkCFDFBwcLJvNpqVLlzpdHzlypGw2m9MxaNCg886bkJCgli1bytvbW7169dL3339fSXcAAABqomoNQCdPnlTXrl2VkJBQap9BgwYpPT3dcXzwwQdlzrlo0SJFRUUpJiZGmzdvVteuXRUeHq5ff/3V1eUDAIAaqm51vvjgwYM1ePDgMvvY7XYFBgZe8Jwvv/yyxowZo1GjRkmS5syZo2XLlmnu3LmaPHnyRdULAABqB7ffA5ScnKxmzZqpXbt2Gjt2rI4ePVpq37y8PG3atEkDBw50tNWpU0cDBw7U+vXrq6JcAABQA1TrE6DzGTRokG6//Xa1atVKe/bs0VNPPaXBgwdr/fr18vDwKNb/yJEjKigoUEBAgFN7QECAduzYUerr5ObmKjc313GenZ3tupsAAABux60D0D333OP4c+fOndWlSxe1bt1aycnJuuGGG1z2OvHx8YqLi3PZfAAAwL25/VtgfxQaGqqmTZtq9+7dJV5v2rSpPDw8lJmZ6dSemZlZ5j6i6OhoZWVlOY4DBw64tG4AAOBealQAOnjwoI4ePaqgoKASr3t5eal79+5KSkpytBUWFiopKUlhYWGlzmu32+Xr6+t0AACA2qtaA1BOTo5SUlKUkpIiSUpNTVVKSorS0tKUk5OjSZMmacOGDdq3b5+SkpI0dOhQtWnTRuHh4Y45brjhBs2aNctxHhUVpbffflvvvvuutm/frrFjx+rkyZOO3woDAACo1j1AGzdu1IABAxznUVFRkqSIiAjNnj1bP/30k959910dP35cwcHBuummm/Tss8/Kbrc7xuzZs0dHjhxxnN999906fPiwpk6dqoyMDHXr1k0rVqwotjEaAABYl80YY6q7CHeTnZ0tPz8/ZWVluffbYbF+LponyzXzAADcUsvJy1wyz77pt7hknspSnp/fNWoPEAAAgCsQgAAAgOUQgAAAgOUQgAAAgOUQgAAAgOUQgAAAgOUQgAAAgOUQgAAAgOUQgAAAgOUQgAAAgOUQgAAAgOUQgAAAgOUQgAAAgOUQgAAAgOUQgAAAgOUQgAAAgOUQgAAAgOUQgAAAgOUQgAAAgOUQgAAAgOUQgAAAgOUQgAAAgOUQgAAAgOUQgAAAgOUQgAAAgOUQgAAAgOUQgAAAgOUQgAAAgOUQgAAAgOUQgAAAgOUQgAAAgOUQgAAAgOUQgAAAgOVUawD6+uuvNWTIEAUHB8tms2np0qWOa/n5+XryySfVuXNnXXLJJQoODtaIESN06NChMueMjY2VzWZzOtq3b1/JdwIAAGqSag1AJ0+eVNeuXZWQkFDs2qlTp7R582ZNmTJFmzdv1uLFi7Vz507deuut5533iiuuUHp6uuNYu3ZtZZQPAABqqLrV+eKDBw/W4MGDS7zm5+enlStXOrXNmjVLPXv2VFpami677LJS561bt64CAwNdWisAAKg9atQeoKysLNlsNjVs2LDMfrt27VJwcLBCQ0N17733Ki0trWoKBAAANUK1PgEqjzNnzujJJ5/U8OHD5evrW2q/Xr16KTExUe3atVN6erri4uLUt29fbd26VT4+PiWOyc3NVW5uruM8Ozvb5fUDAAD3USMCUH5+vu666y4ZYzR79uwy+/7xLbUuXbqoV69eatGihT788EONHj26xDHx8fGKi4tzac0AAMB9uf1bYEXhZ//+/Vq5cmWZT39K0rBhQ7Vt21a7d+8utU90dLSysrIcx4EDBy62bAAA4MbcOgAVhZ9du3Zp1apVatKkSbnnyMnJ0Z49exQUFFRqH7vdLl9fX6cDAADUXtUagHJycpSSkqKUlBRJUmpqqlJSUpSWlqb8/Hz9+c9/1saNG/X++++roKBAGRkZysjIUF5enmOOG264QbNmzXKcP/7441qzZo327dundevW6bbbbpOHh4eGDx9e1bcHAADcVLXuAdq4caMGDBjgOI+KipIkRUREKDY2Vp9++qkkqVu3bk7jVq9erf79+0uS9uzZoyNHjjiuHTx4UMOHD9fRo0fl7++va6+9Vhs2bJC/v3/l3gwAAKgxqjUA9e/fX8aYUq+Xda3Ivn37nM4XLlx4sWUBAIBazq33AAEAAFQGAhAAALAcAhAAALAcAhAAALAcAhAAALAcAhAAALAcAhAAALAcAhAAALAcAhAAALAcAhAAALAcAhAAALCcCgWg0NBQHT16tFj78ePHFRoaetFFAQAAVKYKBaB9+/apoKCgWHtubq7++9//XnRRAAAAlalc3wb/6aefOv78xRdfyM/Pz3FeUFCgpKQktWzZ0mXFAQAAVIZyBaBhw4ZJkmw2myIiIpyueXp6qmXLlnrppZdcVhwAAEBlKFcAKiwslCS1atVKP/zwg5o2bVopRQEAAFSmcgWgIqmpqa6uAwAAoMpUKABJUlJSkpKSkvTrr786ngwVmTt37kUXBgAAUFkqFIDi4uI0bdo09ejRQ0FBQbLZbK6uCwAAoNJUKADNmTNHiYmJuv/++11dDwAAQKWr0OcA5eXlqXfv3q6uBQAAoEpUKAA9+OCDWrBggatrAQAAqBIVegvszJkzeuutt7Rq1Sp16dJFnp6eTtdffvlllxQHAABQGSoUgH766Sd169ZNkrR161ana2yIBgAA7q5CAWj16tWurgMAAKDKVGgPEAAAQE1WoSdAAwYMKPOtrq+++qrCBQEAAFS2CgWgov0/RfLz85WSkqKtW7cW+5JUAAAAd1OhAPTKK6+U2B4bG6ucnJyLKggAAKCyuXQP0H333cf3gAEAALfn0gC0fv16eXt7u3JKAAAAl6vQW2C3336707kxRunp6dq4caOmTJniksIAAAAqS4UCkJ+fn9N5nTp11K5dO02bNk033XSTSwoDAACoLBUKQPPmzXPJi3/99dd64YUXtGnTJqWnp2vJkiUaNmyY47oxRjExMXr77bd1/Phx9enTR7Nnz9bll19e5rwJCQl64YUXlJGRoa5du+r1119Xz549XVIzAACo+S5qD9CmTZs0f/58zZ8/X1u2bCn3+JMnT6pr165KSEgo8fqMGTP02muvac6cOfruu+90ySWXKDw8XGfOnCl1zkWLFikqKkoxMTHavHmzunbtqvDwcP3666/lrg8AANROFXoC9Ouvv+qee+5RcnKyGjZsKEk6fvy4BgwYoIULF8rf3/+C5hk8eLAGDx5c4jVjjGbOnKlnnnlGQ4cOlST985//VEBAgJYuXap77rmnxHEvv/yyxowZo1GjRkmS5syZo2XLlmnu3LmaPHlyOe8UAADURhV6AvTXv/5VJ06c0LZt2/Tbb7/pt99+09atW5Wdna3HHnvMJYWlpqYqIyNDAwcOdLT5+fmpV69eWr9+fYlj8vLytGnTJqcxderU0cCBA0sdAwAArKdCT4BWrFihVatWqUOHDo62jh07KiEhwWWboDMyMiRJAQEBTu0BAQGOa+c6cuSICgoKShyzY8eOUl8rNzdXubm5jvPs7OyKlg0AAGqACj0BKiwslKenZ7F2T09PFRYWXnRRVS0+Pl5+fn6OIyQkpLpLAgAAlahCAej666/X+PHjdejQIUfbf//7X02cOFE33HCDSwoLDAyUJGVmZjq1Z2ZmOq6dq2nTpvLw8CjXGEmKjo5WVlaW4zhw4MBFVg8AANxZhQLQrFmzlJ2drZYtW6p169Zq3bq1WrVqpezsbL3++usuKaxVq1YKDAxUUlKSoy07O1vfffedwsLCShzj5eWl7t27O40pLCxUUlJSqWMkyW63y9fX1+kAAAC1V4X2AIWEhGjz5s1atWqVY29Nhw4dnDYfX4icnBzt3r3bcZ6amqqUlBQ1btxYl112mSZMmKDnnntOl19+uVq1aqUpU6YoODjY6bOCbrjhBt12220aN26cJCkqKkoRERHq0aOHevbsqZkzZ+rkyZOO3woDAAAoVwD66quvNG7cOG3YsEG+vr668cYbdeONN0qSsrKydMUVV2jOnDnq27fvBc23ceNGDRgwwHEeFRUlSYqIiFBiYqKeeOIJnTx5Ug899JCOHz+ua6+9VitWrHD6vrE9e/boyJEjjvO7775bhw8f1tSpU5WRkaFu3bppxYoVxTZGAwAA67IZY8yFdr711ls1YMAATZw4scTrr732mlavXq0lS5a4rMDqkJ2dLT8/P2VlZbn322Gxfufvc0HzZLlmHgCAW2o5eZlL5tk3/RaXzFNZyvPzu1x7gH788UcNGjSo1Os33XSTNm3aVJ4pAQAAqly5AlBmZmaJv/5epG7dujp8+PBFFwUAAFCZyhWAmjdvrq1bt5Z6/aefflJQUNBFFwUAAFCZyhWAbr75Zk2ZMqXELyM9ffq0YmJi9Kc//cllxQEAAFSGcv0W2DPPPKPFixerbdu2GjdunNq1aydJ2rFjhxISElRQUKCnn366UgoFAABwlXIFoICAAK1bt05jx45VdHS0in6BzGazKTw8XAkJCfy6OQAAcHvl/iDEFi1aaPny5Tp27Jh2794tY4wuv/xyNWrUqDLqAwAAcLkKfRK0JDVq1EhXX321K2sBAACoEhX6LjAAAICajAAEAAAshwAEAAAshwAEAAAshwAEAAAshwAEAAAshwAEAAAshwAEAAAshwAEAAAshwAEAAAshwAEAAAshwAEAAAshwAEAAAshwAEAAAshwAEAAAshwAEAAAshwAEAAAshwAEAAAshwAEAAAshwAEAAAshwAEAAAsp251FwDUGLF+LponyzXzAAAqjCdAAADAcghAAADAcghAAADActw+ALVs2VI2m63YERkZWWL/xMTEYn29vb2ruGoAAODO3H4T9A8//KCCggLH+datW3XjjTfqzjvvLHWMr6+vdu7c6Ti32WyVWiMAAKhZ3D4A+fv7O51Pnz5drVu3Vr9+/UodY7PZFBgYWNmlAQCAGsrt3wL7o7y8PM2fP18PPPBAmU91cnJy1KJFC4WEhGjo0KHatm1bFVYJAADcXY0KQEuXLtXx48c1cuTIUvu0a9dOc+fO1SeffKL58+ersLBQvXv31sGDB0sdk5ubq+zsbKcDAADUXjUqAL3zzjsaPHiwgoODS+0TFhamESNGqFu3burXr58WL14sf39/vfnmm6WOiY+Pl5+fn+MICQmpjPIBAICbqDEBaP/+/Vq1apUefPDBco3z9PTUlVdeqd27d5faJzo6WllZWY7jwIEDF1suAABwYzUmAM2bN0/NmjXTLbfcUq5xBQUF+vnnnxUUFFRqH7vdLl9fX6cDAADUXjUiABUWFmrevHmKiIhQ3brOv7g2YsQIRUdHO86nTZumL7/8Unv37tXmzZt13333af/+/eV+cgQAAGovt/81eElatWqV0tLS9MADDxS7lpaWpjp1fs9xx44d05gxY5SRkaFGjRqpe/fuWrdunTp27FiVJQMAADdWIwLQTTfdJGNMideSk5Odzl955RW98sorVVAVAACoqWrEW2AAAACuRAACAACWQwACAACWUyP2AKGWifVz0TxZrpkHQJVrOXmZS+bZN718H41ysWpq3SiOJ0AAAMByCEAAAMByCEAAAMByCEAAAMByCEAAAMByCEAAAMByCEAAAMByCEAAAMByCEAAAMByCEAAAMByCEAAAMByCEAAAMByCEAAAMByCEAAAMByCEAAAMByCEAAAMByCEAAAMByCEAAAMByCEAAAMByCEAAAMByCEAAAMByCEAAAMByCEAAAMByCEAAAMByCEAAAMByCEAAAMByCEAAAMByCEAAAMByCEAAAMBy3DoAxcbGymazOR3t27cvc8xHH32k9u3by9vbW507d9by5curqFoAAFBTuHUAkqQrrrhC6enpjmPt2rWl9l23bp2GDx+u0aNHa8uWLRo2bJiGDRumrVu3VmHFAADA3bl9AKpbt64CAwMdR9OmTUvt++qrr2rQoEGaNGmSOnTooGeffVZXXXWVZs2aVYUVAwAAd+f2AWjXrl0KDg5WaGio7r33XqWlpZXad/369Ro4cKBTW3h4uNavX1/ZZQIAgBqkbnUXUJZevXopMTFR7dq1U3p6uuLi4tS3b19t3bpVPj4+xfpnZGQoICDAqS0gIEAZGRllvk5ubq5yc3Md59nZ2a65AQAA4JbcOgANHjzY8ecuXbqoV69eatGihT788EONHj3aZa8THx+vuLg4l80HAADcm9u/BfZHDRs2VNu2bbV79+4SrwcGBiozM9OpLTMzU4GBgWXOGx0draysLMdx4MABl9UMAADcT40KQDk5OdqzZ4+CgoJKvB4WFqakpCSntpUrVyosLKzMee12u3x9fZ0OAABQe7l1AHr88ce1Zs0a7du3T+vWrdNtt90mDw8PDR8+XJI0YsQIRUdHO/qPHz9eK1as0EsvvaQdO3YoNjZWGzdu1Lhx46rrFgAAgBty6z1ABw8e1PDhw3X06FH5+/vr2muv1YYNG+Tv7y9JSktLU506v2e43r17a8GCBXrmmWf01FNP6fLLL9fSpUvVqVOn6roFAADghtw6AC1cuLDM68nJycXa7rzzTt15552VVBEAAKgN3PotMAAAgMpAAAIAAJZDAAIAAJZDAAIAAJZDAAIAAJZDAAIAAJZDAAIAAJZDAAIAAJZDAAIAAJZDAAIAAJZDAAIAAJZDAAIAAJZDAAIAAJZDAAIAAJZDAAIAAJZDAAIAAJZDAAIAAJZDAAIAAJZDAAIAAJZDAAIAAJZDAAIAAJZDAAIAAJZDAAIAAJZDAAIAAJZDAAIAAJZDAAIAAJZDAAIAAJZDAAIAAJZDAAIAAJZDAAIAAJZDAAIAAJZDAAIAAJZDAAIAAJbj1gEoPj5eV199tXx8fNSsWTMNGzZMO3fuLHNMYmKibDab0+Ht7V1FFQMAgJrArQPQmjVrFBkZqQ0bNmjlypXKz8/XTTfdpJMnT5Y5ztfXV+np6Y5j//79VVQxAACoCepWdwFlWbFihdN5YmKimjVrpk2bNum6664rdZzNZlNgYGBllwcAAGoot34CdK6srCxJUuPGjcvsl5OToxYtWigkJERDhw7Vtm3bqqI8AABQQ9SYAFRYWKgJEyaoT58+6tSpU6n92rVrp7lz5+qTTz7R/PnzVVhYqN69e+vgwYOljsnNzVV2drbTAQAAai+3fgvsjyIjI7V161atXbu2zH5hYWEKCwtznPfu3VsdOnTQm2++qWeffbbEMfHx8YqLi3NpvQAAwH3ViCdA48aN02effabVq1fr0ksvLddYT09PXXnlldq9e3epfaKjo5WVleU4Dhw4cLElAwAAN+bWT4CMMfrrX/+qJUuWKDk5Wa1atSr3HAUFBfr555918803l9rHbrfLbrdfTKkAAKAGcesAFBkZqQULFuiTTz6Rj4+PMjIyJEl+fn6qV6+eJGnEiBFq3ry54uPjJUnTpk3TNddcozZt2uj48eN64YUXtH//fj344IPVdh8AAMC9uHUAmj17tiSpf//+Tu3z5s3TyJEjJUlpaWmqU+f3d/KOHTumMWPGKCMjQ40aNVL37t21bt06dezYsarKBgAAbs6tA5Ax5rx9kpOTnc5feeUVvfLKK5VUEQAAqA1qxCZoAAAAVyIAAQAAyyEAAQAAy3HrPUAAXCDWz0XzZLlmntqO9QZqBJ4AAQAAyyEAAQAAyyEAAQAAyyEAAQAAyyEAAQAAyyEAAQAAyyEAAQAAyyEAAQAAyyEAAQAAyyEAAQAAyyEAAQAAyyEAAQAAyyEAAQAAyyEAAQAAyyEAAQAAyyEAAQAAy6lb3QUAKFnLyctcMs8+b5dMc8FcVvf0W1wyz4Vivat2vYHqxhMgAABgOQQgAABgOQQgAABgOQQgAABgOQQgAABgOQQgAABgOQQgAABgOQQgAABgOQQgAABgOQQgAABgOQQgAABgOQQgAABgOTUiACUkJKhly5by9vZWr1699P3335fZ/6OPPlL79u3l7e2tzp07a/ny5VVUKQAAqAncPgAtWrRIUVFRiomJ0ebNm9W1a1eFh4fr119/LbH/unXrNHz4cI0ePVpbtmzRsGHDNGzYMG3durWKKwcAAO7K7QPQyy+/rDFjxmjUqFHq2LGj5syZo/r162vu3Lkl9n/11Vc1aNAgTZo0SR06dNCzzz6rq666SrNmzariygEAgLty6wCUl5enTZs2aeDAgY62OnXqaODAgVq/fn2JY9avX+/UX5LCw8NL7Q8AAKynbnUXUJYjR46ooKBAAQEBTu0BAQHasWNHiWMyMjJK7J+RkVHq6+Tm5io3N9dxnpWVJUnKzs6uaOlVI9e4Zp6qvk/qvqBuhbmnXPNythpadxX/78t619D1pu4LUlPrLq+i+ow5/99Dtw5AVSU+Pl5xcXHF2kNCQqqhmmow3a+6K6gY6r4gLnu1qq57ZpW+nMuw3lWLuqtWTan7xIkT8vMr+++QWwegpk2bysPDQ5mZmU7tmZmZCgwMLHFMYGBgufpLUnR0tKKiohznhYWF+u2339SkSRPZbLYK1Z6dna2QkBAdOHBAvr6+FZoDF471rnqsedVivasea161XLHexhidOHFCwcHB5+3r1gHIy8tL3bt3V1JSkoYNGybpf+EkKSlJ48aNK3FMWFiYkpKSNGHCBEfbypUrFRYWVurr2O122e12p7aGDRtebPmSJF9fX/7iVCHWu+qx5lWL9a56rHnVutj1Pt+TnyJuHYAkKSoqShEREerRo4d69uypmTNn6uTJkxo1apQkacSIEWrevLni4+MlSePHj1e/fv300ksv6ZZbbtHChQu1ceNGvfXWW9V5GwAAwI24fQC6++67dfjwYU2dOlUZGRnq1q2bVqxY4djonJaWpjp1fv9ltt69e2vBggV65pln9NRTT+nyyy/X0qVL1alTp+q6BQAA4GbcPgBJ0rhx40p9yys5OblY25133qk777yzkqsqm91uV0xMTLG31lA5WO+qx5pXLda76rHmVauq19tmLuR3xQAAAGoRt/4gRAAAgMpAAAIAAJZDAAIAAJZDAAIAAJZDALoICQkJatmypby9vdWrVy99//33Zfb/6KOP1L59e3l7e6tz585avnx5FVVaO5Rnvd9++2317dtXjRo1UqNGjTRw4MDz/u+D4sr733iRhQsXymazOT7AFBemvOt9/PhxRUZGKigoSHa7XW3btuXflXIq75rPnDlT7dq1U7169RQSEqKJEyfqzJkzVVRtzfb1119ryJAhCg4Ols1m09KlS887Jjk5WVdddZXsdrvatGmjxMRE1xVkUCELFy40Xl5eZu7cuWbbtm1mzJgxpmHDhiYzM7PE/t9++63x8PAwM2bMML/88ot55plnjKenp/n555+ruPKaqbzr/Ze//MUkJCSYLVu2mO3bt5uRI0caPz8/c/DgwSquvOYq75oXSU1NNc2bNzd9+/Y1Q4cOrZpia4Hyrndubq7p0aOHufnmm83atWtNamqqSU5ONikpKVVcec1V3jV///33jd1uN++//75JTU01X3zxhQkKCjITJ06s4sprpuXLl5unn37aLF682EgyS5YsKbP/3r17Tf369U1UVJT55ZdfzOuvv248PDzMihUrXFIPAaiCevbsaSIjIx3nBQUFJjg42MTHx5fY/6677jK33HKLU1uvXr3Mww8/XKl11hblXe9znT171vj4+Jh33323skqsdSqy5mfPnjW9e/c2//jHP0xERAQBqBzKu96zZ882oaGhJi8vr6pKrHXKu+aRkZHm+uuvd2qLiooyffr0qdQ6a6MLCUBPPPGEueKKK5za7r77bhMeHu6SGngLrALy8vK0adMmDRw40NFWp04dDRw4UOvXry9xzPr16536S1J4eHip/fG7iqz3uU6dOqX8/Hw1bty4ssqsVSq65tOmTVOzZs00evToqiiz1qjIen/66acKCwtTZGSkAgIC1KlTJz3//PMqKCioqrJrtIqsee/evbVp0ybH22R79+7V8uXLdfPNN1dJzVZT2T83a8QnQbubI0eOqKCgwPF1HEUCAgK0Y8eOEsdkZGSU2D8jI6PS6qwtKrLe53ryyScVHBxc7C8TSlaRNV+7dq3eeecdpaSkVEGFtUtF1nvv3r366quvdO+992r58uXavXu3Hn30UeXn5ysmJqYqyq7RKrLmf/nLX3TkyBFde+21Msbo7NmzeuSRR/TUU09VRcmWU9rPzezsbJ0+fVr16tW7qPl5AoRab/r06Vq4cKGWLFkib2/v6i6nVjpx4oTuv/9+vf3222ratGl1l2MJhYWFatasmd566y11795dd999t55++mnNmTOnukurtZKTk/X888/rjTfe0ObNm7V48WItW7ZMzz77bHWXhgrgCVAFNG3aVB4eHsrMzHRqz8zMVGBgYIljAgMDy9Ufv6vIehd58cUXNX36dK1atUpdunSpzDJrlfKu+Z49e7Rv3z4NGTLE0VZYWChJqlu3rnbu3KnWrVtXbtE1WEX+Gw8KCpKnp6c8PDwcbR06dFBGRoby8vLk5eVVqTXXdBVZ8ylTpuj+++/Xgw8+KEnq3LmzTp48qYceekhPP/200xdz4+KV9nPT19f3op/+SDwBqhAvLy91795dSUlJjrbCwkIlJSUpLCysxDFhYWFO/SVp5cqVpfbH7yqy3pI0Y8YMPfvss1qxYoV69OhRFaXWGuVd8/bt2+vnn39WSkqK47j11ls1YMAApaSkKCQkpCrLr3Eq8t94nz59tHv3bkfQlKT//Oc/CgoKIvxcgIqs+alTp4qFnKIAavhaTZer9J+bLtlKbUELFy40drvdJCYmml9++cU89NBDpmHDhiYjI8MYY8z9999vJk+e7Oj/7bffmrp165oXX3zRbN++3cTExPBr8OVQ3vWePn268fLyMh9//LFJT093HCdOnKiuW6hxyrvm5+K3wMqnvOudlpZmfHx8zLhx48zOnTvNZ599Zpo1a2aee+656rqFGqe8ax4TE2N8fHzMBx98YPbu3Wu+/PJL07p1a3PXXXdV1y3UKCdOnDBbtmwxW7ZsMZLMyy+/bLZs2WL2799vjDFm8uTJ5v7773f0L/o1+EmTJpnt27ebhIQEfg3eXbz++uvmsssuM15eXqZnz55mw4YNjmv9+vUzERERTv0//PBD07ZtW+Pl5WWuuOIKs2zZsiquuGYrz3q3aNHCSCp2xMTEVH3hNVh5/xv/IwJQ+ZV3vdetW2d69epl7Ha7CQ0NNX/729/M2bNnq7jqmq08a56fn29iY2NN69atjbe3twkJCTGPPvqoOXbsWNUXXgOtXr26xH+Xi9Y4IiLC9OvXr9iYbt26GS8vLxMaGmrmzZvnsnpsxvDcDgAAWAt7gAAAgOUQgAAAgOUQgAAAgOUQgAAAgOUQgAAAgOUQgAAAgOUQgAAAgOUQgABctJEjR2rYsGGO8/79+2vChAlVXkdycrJsNpuOHz9e5a+9b98+2Ww2paSkXNQ8565lSc5d35YtW2rmzJmOc5vNpqVLl15UHUBtRwACaqmRI0fKZrPJZrPJy8tLbdq00bRp03T27NlKf+3Fixdf8DdkV2doqanOt77p6ekaPHiwJNcFM6C24dvggVps0KBBmjdvnnJzc7V8+XJFRkbK09NT0dHRxfq68hvEGzdu7JJ53EV+fr48PT2ruwyH861vad9mDuB3PAECajG73a7AwEC1aNFCY8eO1cCBA/Xpp59K+v2tlr/97W8KDg5Wu3btJEkHDhzQXXfdpYYNG6px48YaOnSo9u3b55izoKBAUVFRatiwoZo0aaInnnii2Ddhn/sWTW5urp588kmFhITIbrerTZs2euedd7Rv3z4NGDBAktSoUSPZbDaNHDlS0v++mTs+Pl6tWrVSvXr11LVrV3388cdOr7N8+XK1bdtW9erV04ABA5zqLI3NZtPs2bM1ePBg1atXT6GhoU7zFj0xWbRokfr16ydvb2+9//77Kiws1LRp03TppZfKbrerW7duWrFiRbH5d+zYod69e8vb21udOnXSmjVrnNZu9OjRjntq166dXn311RLrjIuLk7+/v3x9ffXII48oLy+v1PUt6R6L3gJr1aqVJOnKK6+UzWZT//799fXXX8vT01MZGRlO4yZMmKC+ffuedw2B2oAABFhIvXr1nH6QJiUlaefOnVq5cqU+++wz5efnKzw8XD4+Pvrmm2/07bffqkGDBho0aJBj3EsvvaTExETNnTtXa9eu1W+//aYlS5aU+bojRozQBx98oNdee03bt2/Xm2++qQYNGigkJET/+te/JEk7d+5Uenq6IxDEx8frn//8p+bMmaNt27Zp4sSJuu+++xyB4sCBA7r99ts1ZMgQpaSk6MEHH9TkyZMvaB2mTJmiO+64Qz/++KPuvfde3XPPPdq+fbtTn8mTJ2v8+PHavn27wsPD9eqrr+qll17Siy++qJ9++knh4eG69dZbtWvXLqdxkyZN0v/93/9py5YtCgsL05AhQ3T06FFJ/wt1l156qT766CP98ssvmjp1qp566il9+OGHTnMkJSVp+/btSk5O1gcffKDFixcrLi7ugu7tXN9//70kadWqVUpPT9fixYt13XXXKTQ0VO+9956jX35+vt5//3098MADFXodoMZx2deqAnArf/w29sLCQrNy5Upjt9vN448/7rgeEBBgcnNzHWPee+89065dO1NYWOhoy83NNfXq1TNffPGFMcaYoKAgM2PGDMf1/Px8c+mllzp983u/fv3M+PHjjTHG7Ny500gyK1euLLHOom+I/uM3ap85c8bUr1/frFu3zqnv6NGjzfDhw40xxkRHR5uOHTs6XX/yySeLzXUuSeaRRx5xauvVq5cZO3asMcaY1NRUI8nMnDnTqU9wcLD529/+5tR29dVXm0cffdRp3PTp0x3Xi9bm73//e6n1REZGmjvuuMNxHhERYRo3bmxOnjzpaJs9e7Zp0KCBKSgoMMY4r68xxrRo0cK88sorTve4ZMkSp7q2bNni9Lp///vfTYcOHRzn//rXv0yDBg1MTk5OqbUCtQl7gIBa7LPPPlODBg2Un5+vwsJC/eUvf1FsbKzjeufOnZ32/fz444/avXu3fHx8nOY5c+aM9uzZo6ysLKWnp6tXr16Oa3Xr1lWPHj2KvQ1WJCUlRR4eHurXr98F1717926dOnVKN954o1N7Xl6errzySknS9u3bneqQpLCwsAua/9x+YWFhxTYJ9+jRw/Hn7OxsHTp0SH369HHq06dPH/3444+lzl20Nn98upSQkKC5c+cqLS1Np0+fVl5enrp16+Y0R9euXVW/fn2nOXNycnTgwAG1aNHigu7xfEaOHKlnnnlGGzZs0DXXXKPExETddddduuSSS1wyP+DuCEBALTZgwADNnj1bXl5eCg4OVt26zn/lz/1hl5OTo+7du+v9998vNpe/v3+FaqhXr165x+Tk5EiSli1bpubNmztds9vtFaqjvCojCCxcuFCPP/64XnrpJYWFhcnHx0cvvPCCvvvuO5e/1vk0a9ZMQ4YM0bx589SqVSt9/vnnSk5OrvI6gOrCHiCgFrvkkkvUpk0bXXbZZcXCT0muuuoq7dq1S82aNVObNm2cDj8/P/n5+SkoKMjpB/bZs2e1adOmUufs3LmzCgsLnTYD/1HRE6iCggJHW8eOHWW325WWllasjpCQEElShw4dHPtbimzYsOG891hSvw0bNqhDhw6l9vf19VVwcLC+/fZbp/Zvv/1WHTt2LHXuorUpmvvbb79V79699eijj+rKK69UmzZttGfPnmKv9+OPP+r06dNOcxbtmSqvkta3yIMPPqhFixbprbfeUuvWrYs94QJqMwIQAId7771XTZs21dChQ/XNN98oNTVVycnJeuyxx3Tw4EFJ0vjx4zV9+nQtXbpUO3bs0KOPPlrmZ/i0bNlSEREReuCBB7R06VLHnEUbf1u0aCGbzabPPvtMhw8fVk5Ojnx8fPT4449r4sSJevfdd7Vnzx5t3rxZr7/+ut59911J0iOPPKJdu3Zp0qRJ2rlzpxYsWKDExMQLus+PPvpIc+fO1X/+8x/FxMTo+++/17hx48ocM2nSJP3973/XokWLtHPnTk2ePFkpKSkaP368U7+EhAQtWbJEO3bsUGRkpI4dO+bYWHz55Zdr48aN+uKLL/Sf//xHU6ZM0Q8//FDstfLy8jR69Gj98ssvWr58uWJiYjRu3DjVqVP+f7KbNWumevXqacWKFcrMzFRWVpbjWnh4uHx9ffXcc89p1KhR5Z4bqNGqexMSgMrxx03Q5bmenp5uRowYYZo2bWrsdrsJDQ01Y8aMMVlZWcaY/23sHT9+vPH19TUNGzY0UVFRZsSIEaVugjbGmNOnT5uJEyeaoKAg4+XlZdq0aWPmzp3ruD5t2jQTGBhobDabiYiIMMb8b+P2zJkzTbt27Yynp6fx9/c34eHhZs2aNY5x//73v02bNm2M3W43ffv2NXPnzr2gTdAJCQnmxhtvNHa73bRs2dIsWrTIcb20TcMFBQUmNjbWNG/e3Hh6epquXbuazz//vNi4BQsWmJ49exovLy/TsWNH89VXXzn6nDlzxowcOdL4+fmZhg0bmrFjx5rJkyebrl27FvvfZerUqaZJkyamQYMGZsyYMebMmTOlrm9Zm6CNMebtt982ISEhpk6dOqZfv35O9zVlyhTj4eFhDh06VOqaAbWRzZhSdi4CQC1ks9m0ZMmS837dhFWMHj1ahw8fdnw+FGAVbIIGAAvKysrSzz//rAULFhB+YEkEIACwoKFDh+r777/XI488UuzjBgAr4C0wAABgOfwWGAAAsBwCEAAAsBwCEAAAsBwCEAAAsBwCEAAAsBwCEAAAsBwCEAAAsBwCEAAAsBwCEAAAsJz/B/RFzfzrlWOzAAAAAElFTkSuQmCC",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# TODO: Plot the class 0 probabilities against the class 1 probabilities\n",
    "\n",
    "probs = y_hat[:,1]\n",
    "true_probs = probs[y_train == 1]\n",
    "false_probs = probs[y_train == 0]\n",
    "\n",
    "plt.hist([true_probs, false_probs], bins=10)\n",
    "plt.xlabel('Predicted probability')\n",
    "plt.ylabel('Count')\n",
    "plt.title('Training data')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "# TODO: Plot the probabilities for the validation data\n",
    "y_hat_val = clf.predict_proba(X_val)\n",
    "\n",
    "probs_val = y_hat_val[:,1]\n",
    "true_probs_val = probs_val[y_val == 1]\n",
    "false_probs_val = probs_val[y_val == 0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.hist([true_probs_val, false_probs_val], bins=10)\n",
    "plt.xlabel('Predicted probability')\n",
    "plt.ylabel('Count')\n",
    "plt.title('Validation data')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Section 2: Logistic Regression in Torch\n",
    "\n",
    "This section is optional, but provided to get a sense of how torch can be used to the same kind of model, but with more upfront work. If you have familiarity with torch, try to code up logistic regression on your own!"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "from torch import nn\n",
    "from sklearn.metrics import accuracy_score\n",
    "\n",
    "class LogReg(nn.Module):\n",
    "    def __init__(self, num_features=2):\n",
    "        super().__init__()\n",
    "         # TODO: Setup linear layer for the model\n",
    "        self.linear = nn.Linear(num_features, 1)\n",
    "\n",
    "    def forward(self, x):\n",
    "        # TODO: Apply logistic regression model in forward\n",
    "        x = self.linear(x)\n",
    "        x = torch.sigmoid(x)\n",
    "        return x"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Must convert data to torch FloatTensor for neural networks\n",
    "X_train = torch.FloatTensor(X_train)\n",
    "y_train = torch.FloatTensor(y_train.astype(np.float32)).unsqueeze(-1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "# TODO: Setup loss function for logistic regression\n",
    "loss_function = nn.BCELoss()\n",
    "#####\n",
    "NUM_STEPS = 200\n",
    "model = LogReg(X_train.shape[-1])\n",
    "optimizer = torch.optim.SGD(model.parameters(), lr=1e-1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "BCE Loss at epoch 0: 0.836\n",
      "Accuracy score at epoch 0: 0.520\n",
      "BCE Loss at epoch 10: 0.683\n",
      "Accuracy score at epoch 10: 0.600\n",
      "BCE Loss at epoch 20: 0.639\n",
      "Accuracy score at epoch 20: 0.920\n",
      "BCE Loss at epoch 30: 0.603\n",
      "Accuracy score at epoch 30: 0.920\n",
      "BCE Loss at epoch 40: 0.572\n",
      "Accuracy score at epoch 40: 0.933\n",
      "BCE Loss at epoch 50: 0.543\n",
      "Accuracy score at epoch 50: 0.947\n",
      "BCE Loss at epoch 60: 0.517\n",
      "Accuracy score at epoch 60: 0.960\n",
      "BCE Loss at epoch 70: 0.494\n",
      "Accuracy score at epoch 70: 0.960\n",
      "BCE Loss at epoch 80: 0.473\n",
      "Accuracy score at epoch 80: 0.960\n",
      "BCE Loss at epoch 90: 0.454\n",
      "Accuracy score at epoch 90: 0.960\n",
      "BCE Loss at epoch 100: 0.437\n",
      "Accuracy score at epoch 100: 0.960\n",
      "BCE Loss at epoch 110: 0.421\n",
      "Accuracy score at epoch 110: 0.973\n",
      "BCE Loss at epoch 120: 0.407\n",
      "Accuracy score at epoch 120: 0.973\n",
      "BCE Loss at epoch 130: 0.394\n",
      "Accuracy score at epoch 130: 0.987\n",
      "BCE Loss at epoch 140: 0.382\n",
      "Accuracy score at epoch 140: 0.987\n",
      "BCE Loss at epoch 150: 0.370\n",
      "Accuracy score at epoch 150: 0.987\n",
      "BCE Loss at epoch 160: 0.360\n",
      "Accuracy score at epoch 160: 0.987\n",
      "BCE Loss at epoch 170: 0.350\n",
      "Accuracy score at epoch 170: 0.987\n",
      "BCE Loss at epoch 180: 0.341\n",
      "Accuracy score at epoch 180: 0.987\n",
      "BCE Loss at epoch 190: 0.333\n",
      "Accuracy score at epoch 190: 0.987\n"
     ]
    }
   ],
   "source": [
    "# TODO: Complete the training loop, pass all the training data each step\n",
    "losses = []\n",
    "accs = []\n",
    "\n",
    "for i in range(NUM_STEPS):\n",
    "    optimizer.zero_grad()\n",
    "    y_hat = model(X_train)\n",
    "    loss = loss_function(y_hat, y_train)\n",
    "    loss.backward()\n",
    "    acc = accuracy_score(torch.round(y_hat.detach()), y_train)\n",
    "    accs.append(acc)\n",
    "    losses.append(loss.item())\n",
    "    if i % 10 == 0:\n",
    "        print(f\"BCE Loss at epoch {i}: {loss.item():.3f}\")\n",
    "        print(f\"Accuracy score at epoch {i}: {acc:.3f}\")\n",
    "    optimizer.step()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(losses)\n",
    "plt.ylabel(\"BCE Loss\")\n",
    "plt.twinx()\n",
    "plt.plot(accs, c=\"C1\")\n",
    "plt.ylabel(\"Train accuracy\")\n",
    "plt.show()"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "datasci",
   "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": "3.10.12"
  },
  "orig_nbformat": 4
 },
 "nbformat": 4,
 "nbformat_minor": 2
}