{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CPython 3.5.6\n",
      "IPython 6.5.0\n",
      "\n",
      "sklearn 0.20.1\n",
      "numpy 1.15.2\n",
      "scipy 1.1.0\n",
      "matplotlib 3.0.0\n"
     ]
    }
   ],
   "source": [
    "%load_ext watermark\n",
    "%watermark -v -p sklearn,numpy,scipy,matplotlib"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "hide_input": false
   },
   "outputs": [],
   "source": [
    "%matplotlib inline\n",
    "from preamble import *"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 모델 평가와 성능 향상"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "사이킷런 0.22 버전에서 `LogisticRegression` 클래스의 `solver` 매개변수 기본값이 `liblinear`에서 `lbfgs`로 변경될 예정입니다. 사이킷런 0.20 버전에서 `solver` 매개변수를 지정하지 않는 경우 이에 대한 경고 메세지가 출력됩니다. 경고 메세지를 피하기 위해 `solver` 매개변수 값을 `liblinear`로 지정합니다.\n",
    "\n",
    "사이킷런 0.20 버전에서 `LogisticRegression`의 `multi_class` 매개변수 옵션에 `auto`가 추가되었습니다. `auto`로 설정하면 이진 분류이거나 `solver`가 `liblinear`일 경우에는 `ovr`을 선택하고 그 외에는 `multinomial`을 선택합니다. 사이킷런 0.22 버전부터는 `multi_class` 매개변수의 기본값이 `ovr`에서 `auto`로 변경됩니다. 여기서는 경고를 피하기 위해 명시적으로 `ovr` 옵션을 지정합니다."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "테스트 세트 점수: 0.88\n"
     ]
    }
   ],
   "source": [
    "from sklearn.datasets import make_blobs\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "from sklearn.model_selection import train_test_split\n",
    "\n",
    "# 인위적인 데이터셋을 만듭니다\n",
    "X, y = make_blobs(random_state=0)\n",
    "# 데이터와 타깃 레이블을 훈련 세트와 테스트 세트로 나눕니다\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y, random_state=0)\n",
    "# 모델 객체를 만들고 훈련 세트로 학습시킵니다\n",
    "logreg = LogisticRegression(solver='liblinear', multi_class='ovr').fit(X_train, y_train)\n",
    "# 모델을 테스트 세트로 평가합니다\n",
    "print(\"테스트 세트 점수: {:.2f}\".format(logreg.score(X_test, y_test)))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 교차 검증(Cross Validation)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "hide_input": false
   },
   "outputs": [
    {
     "data": {
      "application/pdf": "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\n",
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x144 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "mglearn.plots.plot_cross_validation()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### scikit-learn의 교차 검증"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "사이킷런 0.22 버전에서 `cross_val_score` 함수의 `cv` 매개변수 기본값이 3에서 5로 바뀔 예정입니다. 0.20 버전에서 `cv` 매개변수를 지정하지 않는 경우 이에 관한 경고 메세지가 출력됩니다. 경고 메세지를 피하기 위해 `cv` 매개변수 값을 명시적으로 3으로 지정합니다."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "교차 검증 점수: [0.961 0.922 0.958]\n"
     ]
    }
   ],
   "source": [
    "from sklearn.model_selection import cross_val_score\n",
    "from sklearn.datasets import load_iris\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "\n",
    "iris = load_iris()\n",
    "logreg = LogisticRegression(solver='liblinear', multi_class='ovr')\n",
    "\n",
    "scores = cross_val_score(logreg, iris.data, iris.target, cv=3)\n",
    "print(\"교차 검증 점수: {}\".format(scores))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "교차 검증 점수: [1.    0.967 0.933 0.9   1.   ]\n"
     ]
    }
   ],
   "source": [
    "scores = cross_val_score(logreg, iris.data, iris.target, cv=5)\n",
    "print(\"교차 검증 점수: {}\".format(scores))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "교차 검증 평균 점수: 0.96\n"
     ]
    }
   ],
   "source": [
    "print(\"교차 검증 평균 점수: {:.2f}\".format(scores.mean()))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 교차 검증의 장점"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 계층별(Stratified) k-겹 교차 검증과 그외 전략들"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Iris 레이블:\n",
      "[0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0\n",
      " 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1\n",
      " 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 2 2 2 2 2 2 2 2 2 2 2\n",
      " 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2\n",
      " 2 2]\n"
     ]
    }
   ],
   "source": [
    "from sklearn.datasets import load_iris\n",
    "iris = load_iris()\n",
    "print(\"Iris 레이블:\\n{}\".format(iris.target))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "hide_input": false
   },
   "outputs": [
    {
     "data": {
      "application/pdf": "JVBERi0xLjQKJazcIKu6CjEgMCBvYmoKPDwgL1BhZ2VzIDIgMCBSIC9UeXBlIC9DYXRhbG9nID4+CmVuZG9iago4IDAgb2JqCjw8IC9FeHRHU3RhdGUgNCAwIFIgL0ZvbnQgMyAwIFIgL1BhdHRlcm4gNSAwIFIKL1Byb2NTZXQgWyAvUERGIC9UZXh0IC9JbWFnZUIgL0ltYWdlQyAvSW1hZ2VJIF0gL1NoYWRpbmcgNiAwIFIKL1hPYmplY3QgNyAwIFIgPj4KZW5kb2JqCjEwIDAgb2JqCjw8IC9Bbm5vdHMgWyBdIC9Db250ZW50cyA5IDAgUgovR3JvdXAgPDwgL0NTIC9EZXZpY2VSR0IgL1MgL1RyYW5zcGFyZW5jeSAvVHlwZSAvR3JvdXAgPj4KL01lZGlhQm94IFsgMCAwIDg1Ni40NzUzNzUgMzU0LjI4NDM3NSBdIC9QYXJlbnQgMiAwIFIgL1Jlc291cmNlcyA4IDAgUgovVHlwZSAvUGFnZSA+PgplbmRvYmoKOSAwIG9iago8PCAvRmlsdGVyIC9GbGF0ZURlY29kZSAvTGVuZ3RoIDExIDAgUiA+PgpzdHJlYW0KeJztW01vHDcSvfev6OPmYIpFsvhxjGHHOQWrhYA9LHIwZK9jQ1IQ20n+/j6Ore7Hnml6pNHMWoYROJbeTFc9Fov1wS7L+G44+1HGNx9GO77Dn7/H/4y/4u9Xo4wvxrNnr/96e/n6Xy+ejpcfBgv8esgaTUjqk+LXK/7VazAuh/rjFb7c/PrbMNwMf4w5GN0gzjlTNDodY7CmuFFsNnl8/3r893gznv3oNpSMWinRJkgZ34DR31D5Yjj758uPH1+/vwGn8exnGT9c3gzDJNl7a2Iumj3IuuKMuhDEMX7V4AImItnGuppJCqMg/3Q4H+9KX2bOI3Eezn52S85WjNVSNCw4z3jD2WUP0zpJDecGPTpnl8QEb71bcCa85azeRHUaWs6MHouzK9Ek7L3E1jvUBpNC0ZRb72Cc/YDlPIB/3MW9WTU7C1NlZyG8cQuScxJ3YX3sMMyPHYZxdg2WcwqXUTxUSrBZWpdJEk1JIWHHGpdhnF2D5ZwipLA+9hPmx35CeOMPJOch/OQurs6q2WWYKrsM4+waLOeBXQY/cyQMYrK1sSwj4Yy3kdAFY50rkppIyOhDceRjwyz5mDFLxpkPyzkGT94r5sl7yzwZZz4sZyfPW3/cEEDBA2pzyVPp4JNBCk62Ru89Lee6gaWYsIGvqmSsZfgD/7fjEwu5kpJJ8E6bUhaYBKTwbP325fXw9GI8+wnfsePFfwdJJvtSomZnPR6/eDX8Y5Qfxot3w/MLiLTGVqn47/YHSDj75eXNn9dPX77/8+bF7x9/e3v55PLtK5wFPz77fTjfegoWkVIyLGFdR4BEnKaNgPNhY5YhWGdyllByYwaGu2bwucABsk/ehxDubAb3lZghihqxMcbQmIHhrhki7CXqFeawyd/ZDP7/bgbeUni2OFvqSZeCoCo7luEQK5BfQk7e6u0y7rMK0WgPWoXDYd0tYGnqVUOKDbsluBppSsGuxpD7W7GyCJ82q3CSN71UZxG6W0CIm0XYkLL2zOAlptan587DUbTNKAyW6C6PVjHxU5NUV5AQDEo94McLbq4enEMcwbuyZoGamCXhJCtbgNCeBVzOqI2yU+cFXnC8uHZUC0SUJkGCby0wo10LaEZSzNG5WJ3weCHtqBbwU2nOFvBUsG9bAGek5ndXa56CUiMjGFjVeLx46G2Sw6J6CivRbN946PxaRN03HqKPyIfGwxUK+8ZDV600Jbe6l0+qFAkGBqiJ2YVsvO6dojd76UAhoauKVkM6KOtJWTHQvlkvx7WcteC4TsGtUNh3l93qLqMrEJwGwSYgvXQ4wBNWJJS6CnTeaCm7jmLV7qhhJKGfDc4lHFn5dHhxNnNwWtyeh9c++tPbVnVZTYhopjRvCvV7WOR+PctXaxEtyUQs3krY1Oz3sMj92pev1iKuNnP1Mgyt7TIyumqEgK4QBzsVa31p4p8PBkGnJKc+uOaToEbx7Zjcp4C9zzO1rS/Qg9zr9/1k2gU32cP1iw1vm33g58TW5h7bnVR6IlARrYlA6RA+h7DuZqpfFYHaRGqmgEu60nWIXNZkLM2/KgNt35oMjQaVWYlwqtg1h9q8JiPGWmchLaWcejLgXromI1vjsRZf0Iz03HtZfrCMgsoPa/FBU/+ILCqQ1jtQFmIxrqCJ6gqZipAdQsSZgNWidUyuKySUVQ9BZYomDucRbUzoF1TZrgrx0SiWYyWrdoXAZqtCQjJOFDVDtLGX81Giry8nWlMvkOFrPneLFxdWvQQtnfEo6ZHrwKZfO6zvDvoGOEjWFJx0z96ikmuEFK0xAAdPUvcAL6u5Jo7YXMNk9gq+3ZOzKOjaYFRqGEBqwQnsHuHp6Jy3d76+fK6WH/pdbnUce/vqRbQ2ZrbEwvhVgyNP+613uQ26fWP90Ozprrzh7yPqzhjFh5Y/48yU5Zx2BXSL3qxAE3y2IKo1/GeUWbKM07IXVYOEFlKNo8mE6b16bdizq24yw1cEl2Li7XsuEsLw8dn7mBBwbfSpZR9QEdkaL2JLn3AmSmJOy19TMQmNBBJowz+iOUdpaZ1r+RPOREnMafmjajPWJUFvmmvFvD3gMMPNW72EaDq9rZ6FMHx89khrpl5GamnY02s9Zk8w0yQhp2Uf1ZqQEhQ27OllH7MnmGmSkMPZf/F9OwcbDpXkGE2wZJzDIss5PFx+kTZHmSbCzx7RxnjCmR7LOQFtDi5Mm1yhoc14k5xIzgloz5UIR0R0d0aiXwbEGebAN4k4PBp+eWZnjl4NYS5UmDHhzI7EnIA0Ba2G9FydMOUJZWYk4gSEKU41hGEwU4vZoi1lwpkdiTkB6UlakxunCqpJjRPKYXgScILYTAU006WKifkS3KTyWcgJKFPVzJSpSGLKBDM3EvKwlNuxJ8Vp2TH1NMNNtoOGsDXzxOAD0aP9YoKUwJggwcyFhByBIu0PU6RkxRQJZi4kZCfFvQadttdxTZia3A61NJfeWUyEnfJmAM7op4mWeuW6eTkyPr/4/JJVYjHB+1wnV2ctjPb1IBeaaDUWcQjOC11uoQyubrzzOTtWxmhfWX3xV91BU1DrF8rCQpmHCie+hMLKGO0r894bnzU7VFG6tGJcKAv4WGyw0mwWo31lQeBTSXMuBVFuoSwvlPEU6ayM0b4ytRaZVHLx+E5eKBO71Fa8yTmgcmu0EfoFbVmNBClasspS2dJDYsom1Yu7wMoY7SuLNaR6Zx1amrh0R5ldZG1oanPn9n1m6jHPTImnqD1NixDam5dBlYx+45HPTNU0p1szU4R2LYCcGx/7zBTiRdkamZrB3vrRetjHPjAVpqaZlh+ok+6NSwX0Xt+npb6daakYtkcCvg9LfZPDUvAZ/31Wimal7mGQb3tU6h4G+cYmpTziYVb1NdUJOk/ZOS61nN85yhiTiK4O7uw7xgQ7HDjG5GxYHaZaznndZ5Ri3zGm3iSFZoNlYCGozbuBtjNIEQt6QRhUbcw9GTxHUS+H6r+8v72Zad+Lfb4VTFGwwlJrZ3SfGjaXLtcDqqlt+KqBE8FxJ7y5MLq9JZrvh96wR6dcthXtn+zverJ9Kgee7LSWkPdN6atN9N4pfbXs2rNum5voWnw371DX/UW2/EVrlbblLhPaeAueKtvOQugdfWXScrwSX2Qlhu9b/a0W6PtfdsR0mKusO9vJSnx2tvPhf+P9bXQKZW5kc3RyZWFtCmVuZG9iagoxMSAwIG9iagoyMzc1CmVuZG9iagoxNiAwIG9iago8PCAvRmlsdGVyIC9GbGF0ZURlY29kZSAvTGVuZ3RoIDE3ID4+CnN0cmVhbQp4nDMyMlEwgMMUQy4AGlcC6AplbmRzdHJlYW0KZW5kb2JqCjE3IDAgb2JqCjw8IC9CQm94IFsgLTQ1IC0yODcgMTAwMCA4MzggXSAvRmlsdGVyIC9GbGF0ZURlY29kZSAvTGVuZ3RoIDIyOQovU3VidHlwZSAvRm9ybSAvVHlwZSAvWE9iamVjdCA+PgpzdHJlYW0KeJxNkcuNwzAMRO+qgg0EEP9UPQFy8vZ/3aFsx7mYDxpxOKIHz0lpk/6GWGw6hizZpLrOWr7re5jyJsvc1aUVgeLCFFXksyg0yZprdU8kuTEZ3GwF2VQydygKxdhIRUmzSFJIkeI9BG4KhcNI4NQpJarndKxASHY0wkDalnHatnHbuiDkREWzwT6io2g1JYZ0DfT4rhiIb5NI7sqVl9Kjm45xr+oYzniOLSytycVwFst/KKvvZfEP9QNO9aF2efzOey/m6v/BFx4bheP2+dIltxFjw+hZuenY9LI7zkOn+vkHckxfwgplbmRzdHJlYW0KZW5kb2JqCjE4IDAgb2JqCjw8IC9CQm94IFsgLTQ1IC0yODcgMTAwMCA4MzggXSAvRmlsdGVyIC9GbGF0ZURlY29kZSAvTGVuZ3RoIDIxNQovU3VidHlwZSAvRm9ybSAvVHlwZSAvWE9iamVjdCA+PgpzdHJlYW0KeJxNkTsOAzEIRPs9BReIxM9gnydSqs392wzGUVJY85bhs8ZXrqCHmNL7SuHGe2NOLoL/o2MPc9KxUGPTNt2b7LjjjzyLbNmXIsnmRAaTCarRXH3Q81JEJJMkBokkzaTYcXyqoVSKZpCykAbOUDKfSDFG+z1WaOiCohbcTiCisaCTFFmlaIsuRRKzFZdq56TeBcl1d1l8SLVjiqpS41ZM8hOZra6lCsc1KNIRwQRkuSoGtMM0sFeXoJGoEfz/XkatuKiXnVmzA9k/egivY/9hP8brA0ZUWSUKZW5kc3RyZWFtCmVuZG9iagoxOSAwIG9iago8PCAvQkJveCBbIC00NSAtMjg3IDEwMDAgODM4IF0gL0ZpbHRlciAvRmxhdGVEZWNvZGUgL0xlbmd0aCAxNjQKL1N1YnR5cGUgL0Zvcm0gL1R5cGUgL1hPYmplY3QgPj4Kc3RyZWFtCnicTY87DsMwDEN3n4JHsP7yeQJ0Su+/lk6RpoOhB9LUZ4Q5bAneY1MLztElf8CqT9myhj+ws+cwiz/qSfCwB77ePesc2QlZzak59SJqsqBpSFOYsbrCZ+MYGQJvKlEIvozFX0JHYl50XlSzSOX+I2ayUC5IpWqN4E7HKG4ZfMWsLyY14Z7b4ZXOnUp4UTQ7NUwuZwq02HE5lJfcNxzj9QGmmD4JCmVuZHN0cmVhbQplbmRvYmoKMjAgMCBvYmoKPDwgL0JCb3ggWyAtNDUgLTI4NyAxMDAwIDgzOCBdIC9GaWx0ZXIgL0ZsYXRlRGVjb2RlIC9MZW5ndGggMTgzCi9TdWJ0eXBlIC9Gb3JtIC9UeXBlIC9YT2JqZWN0ID4+CnN0cmVhbQp4nE2QuQ1DIRBEc6qYBiztDb8eS45w/6kH+LYcoB14ew2trkQfgnfrQ7eaWz3UjXLhP3m4d6gVS/iiyRN8GAN6GYzg2YwpVg7rCRfn6XBmPZv7gF8CJw0nHYYYThKiSFeEBpJzwhTlsmp4K2NdrMhOciLn1FFmJ+r4kh5bzCU6O0/u9FPMWsqZsOOdwx3skKhDUulZ1nYpiazcJMLo4YKz43Kk27VxV6UjfkqtivuXZnt9ANg6QuwKZW5kc3RyZWFtCmVuZG9iagoyMSAwIG9iago8PCAvQkJveCBbIC00NSAtMjg3IDEwMDAgODM4IF0gL0ZpbHRlciAvRmxhdGVEZWNvZGUgL0xlbmd0aCAxNjAKL1N1YnR5cGUgL0Zvcm0gL1R5cGUgL1hPYmplY3QgPj4Kc3RyZWFtCnicTY9BDsNACAPv+wp/oNICBjbvqdRT8/9rIUmVnBhkwHhoEKKOfUQc8G1IYYG4P0inFmnGRUY7iLO1AEOg4ngPl4SqwUOhpvBtQTlbKZJoE88i3+Betk64zarHzKyP6GA5CQVkQixLoTQZbHWdsN7VVYppp6jPtjPNP9V7rK3mqttH9mZ0mqaXWIeIOv7AXPPSbyKb+s5N553PDyIIPCQKZW5kc3RyZWFtCmVuZG9iagoyMiAwIG9iago8PCAvQkJveCBbIC00NSAtMjg3IDEwMDAgODM4IF0gL0ZpbHRlciAvRmxhdGVEZWNvZGUgL0xlbmd0aCAxNzgKL1N1YnR5cGUgL0Zvcm0gL1R5cGUgL1hPYmplY3QgPj4Kc3RyZWFtCnicTY8xDgQhDAN7XpEPrBQngcB7Vrpq7//tGQ5pqTKSsY1LC4FW+RZCKuQp3k9qliRUHIQck6j+yUIX2TBBN3HvgqFyrwSMLmEuxpqovDAqkxCzJTwF3iRAl0N8NAG76E666fTKfAxxo4I2Ff4PzLGsvCoWgzM6FeN7KN+NXMOQvu5d9tKnJAsuuHF00rvwWeitkWqvB13QubGy8cBM3fpLEbpzXsquu/GlXfj5ARxiR+4KZW5kc3RyZWFtCmVuZG9iagoyMyAwIG9iago8PCAvQkJveCBbIC00NSAtMjg3IDEwMDAgODM4IF0gL0ZpbHRlciAvRmxhdGVEZWNvZGUgL0xlbmd0aCAxODIKL1N1YnR5cGUgL0Zvcm0gL1R5cGUgL1hPYmplY3QgPj4Kc3RyZWFtCnicTZA7DsMwDEN3n0IXKCBKcmSfJ0Cn9P5rKddFMukBJPVrOQ95wU0+LaE/vBbm0CLqN225IyVTmTnUFl2LXtC55Qd6DpLbk6L7Vm/afUKgWa1DUlH++aSD6auh40EYpWKMTRaxyLgyRoiHCabJ2ZweU5Vg1rSmsyKoFCFqlwh29EEPU+4S9MNmpbk/rIt39jeIOxWMUrSTQiwnq4nxKvCJZzNOgtZUrMOQuer5v/Rq7y+/XknWCmVuZHN0cmVhbQplbmRvYmoKMjQgMCBvYmoKPDwgL0JCb3ggWyAtNDUgLTI4NyAxMDAwIDgzOCBdIC9GaWx0ZXIgL0ZsYXRlRGVjb2RlIC9MZW5ndGggMjEyCi9TdWJ0eXBlIC9Gb3JtIC9UeXBlIC9YT2JqZWN0ID4+CnN0cmVhbQp4nE2QOQ7FIAxEe07hC0TCG4bzRPoVuX/7x6AoFGiebI8XCmuQsdNTkkKdZvHOB7UmIBE9yHrP7PCTOB1W4yAJTXIHOXkVaJBbJ0HnGz4DGTXMFnRqqBLE7rJI0t1MSRi57MdG3gZ0uZVJct5INTJPFWSMkyppMPHopJoayMhIQm/XreicCk9bMBMEM3Apx0sqiyQ3T227Bv26LtLKW6XixopM0v6J949nCR6ElZ4NvSLUq5/EXxUigItx87N2WbjXYrxZFEd/dOl4PR9ty+8PYTxczAplbmRzdHJlYW0KZW5kb2JqCjI1IDAgb2JqCjw8IC9CQm94IFsgLTQ1IC0yODcgMTAwMCA4MzggXSAvRmlsdGVyIC9GbGF0ZURlY29kZSAvTGVuZ3RoIDczCi9TdWJ0eXBlIC9Gb3JtIC9UeXBlIC9YT2JqZWN0ID4+CnN0cmVhbQp4nE3MuQ3AQAhE0XyqoARzLGxFjnD/qT2BV0Q8acSHZohlygOqLpVG+VRafrKlQx6La/gQK42/19ilUgwTzsdzauyN+wWjnBmxCmVuZHN0cmVhbQplbmRvYmoKMjYgMCBvYmoKPDwgL0ZpbHRlciAvRmxhdGVEZWNvZGUgL0xlbmd0aCAyMTggPj4Kc3RyZWFtCnicNZBLjgQhDEP3nCIXKAnnB5ynRr2b+2/HCdMLZAsS54XYJlM85fElYVuWhfxg6NG2v8MW2rnxZqr4cklLeUcgJdbsNv9Xuy90ytqqwIzuceeDG+ThpbG8JtaY0ndoXKdY8tgRbKpWD4J0vLDJw7hzozZZQ5YKEJKUtGZiAmohzVNygWiKMFcDV8ZdAayvpRR33e/ab39AZn0AdraDmUR4q1mT0SkZwXgQCqQ51YzDH2FafQM6lPIOYw2Nba9XJ2dXu61u94LXq3dAuRppzCuEL9Q7Pn91z1HJCmVuZHN0cmVhbQplbmRvYmoKMjcgMCBvYmoKPDwgL0JCb3ggWyAtNDUgLTI4NyAxMDAwIDgzOCBdIC9GaWx0ZXIgL0ZsYXRlRGVjb2RlIC9MZW5ndGggMTg4Ci9TdWJ0eXBlIC9Gb3JtIC9UeXBlIC9YT2JqZWN0ID4+CnN0cmVhbQp4nE2QQQ4DMQgD73mFP1AJAoTwnko9bf9/rdONlD0xwgGbtDEwA9/G6h64msmThnXS8APJLrXqD7JcosaTNGpR1qZuN/UK6OBLemgK3tywKOE2oNPhY0JLqfypC6c9BKrBN6sqnN4quaYnvcRgkaiCmaLG6ouD0ToXl6B7YU62e1dMzpatu3XkKu92f8PVkrMv5a3flio3Xn80JrtaMNehl8o6LRjkgZmy9UMec+85lFO246Ft+PkBwV9NwQplbmRzdHJlYW0KZW5kb2JqCjI4IDAgb2JqCjw8IC9CQm94IFsgLTQ1IC0yODcgMTAwMCA4MzggXSAvRmlsdGVyIC9GbGF0ZURlY29kZSAvTGVuZ3RoIDE5MAovU3VidHlwZSAvRm9ybSAvVHlwZSAvWE9iamVjdCA+PgpzdHJlYW0KeJxNkEGuAzEIQ/dzCi5QKQYC5DyVupp//+130lSTFU92IIYrR8gLpvJ3JdoX74VZbRL9h7bddUhmY0/AF92LXmhj2wdaOcmRB3n/uQ/tOSXV5+QS905B86RYASIeyBYECz/IcproJ2F9hMIm9S/pSEGomLMywvuyqSRDWgkY2aligM4inTG9qwCdb2aFeOPblrObHWgmxtmD25sJL0ed6YaKJuUm2iFVlFVNKji/z70RY5b3PsN9ff4BolpP2QplbmRzdHJlYW0KZW5kb2JqCjI5IDAgb2JqCjw8IC9CQm94IFsgLTQ1IC0yODcgMTAwMCA4MzggXSAvRmlsdGVyIC9GbGF0ZURlY29kZSAvTGVuZ3RoIDE4NAovU3VidHlwZSAvRm9ybSAvVHlwZSAvWE9iamVjdCA+PgpzdHJlYW0KeJxN0EsOwyAMBNA9p+AClfAXc55IXaX333ZMiGDFkwz2mGIq1Zzqr6RcuN7Fhx3q0aAedIgareqWREDZZUttzKofymlP9VGgw4dIEYFp8Z4kyTqPOPSR8b7ZWk9YK7OjT28Td8LYAOl6yH2GiY2uOUC9HVLL+2T9UEbAJPElweSU4FMEURShpI16FcWqQlhXcWKy4VNEDJUUeyY23GbjqoEuipO5Zu+riOcKqI5np3e3q3z/SahTcwplbmRzdHJlYW0KZW5kb2JqCjMwIDAgb2JqCjw8IC9CQm94IFsgLTQ1IC0yODcgMTAwMCA4MzggXSAvRmlsdGVyIC9GbGF0ZURlY29kZSAvTGVuZ3RoIDkyCi9TdWJ0eXBlIC9Gb3JtIC9UeXBlIC9YT2JqZWN0ID4+CnN0cmVhbQp4nE3Ouw3AMAgE0J4pvEAkfgZ7oVTO/m3AUWQqnnQ6AMiosXB7INVxtAWmVmSKITIscuohj96RiodYvUhwp8ZFeW2BTiqyGRhecEmuOHMHigVf+/9/wf0CmYsnNQplbmRzdHJlYW0KZW5kb2JqCjMxIDAgb2JqCjw8IC9GaWx0ZXIgL0ZsYXRlRGVjb2RlIC9MZW5ndGggMTMzID4+CnN0cmVhbQp4nE2OOw5DIQwEe06xF4iEjT9wHqJXkfu3WYiQUnlsSzvrvaFCVPFqjuaJJL+l7JO74XMoJLCK2EBoh6QjTCGDe1TMotoQ2aEWiMEZv6BZbuQ6tCWraMof7cRN9ZIlfRWqBh8GrQLvwiyhwYN+d3ZLSGNHduJHeaH7tLaO236W5wvadyqwCmVuZHN0cmVhbQplbmRvYmoKMzIgMCBvYmoKPDwgL0ZpbHRlciAvRmxhdGVEZWNvZGUgL0xlbmd0aCAyNzAgPj4Kc3RyZWFtCnicLVG7jcUwDOszBRd4gPW358nhVXf7t0cpKQIypkxJdGzDQgQ+FghVFPFHriOIKvxd1CMPNotKBrbgvkgyEr+XrI2sDTHiSUjRQ5QlugKlBXUfV91n8L4sZZhLsnbDw5G7FT9KX0WIINeDYU5lWH+r4HSKteAc7LnjTpdNZL8+9ZWtcEKjr/uCpVFVmPf4vhzGzrYdeoyqQrOoSPA0erEmThJz6yWf0d68htgh6mbxEjITOlt/Ma4WNl24da2Zq/9ciYfo9k7jSZW5eeXk4me1ogy8VeUzyHnQZrdmU8vE+AbGF0uTt092VytkMvdWstPVOM8Ju2cUhJun62xNhdmJBt0C7/Pf1/cfyIBmIgplbmRzdHJlYW0KZW5kb2JqCjMzIDAgb2JqCjw8IC9CQm94IFsgLTQ1IC0yODcgMTAwMCA4MzggXSAvRmlsdGVyIC9GbGF0ZURlY29kZSAvTGVuZ3RoIDEyMQovU3VidHlwZSAvRm9ybSAvVHlwZSAvWE9iamVjdCA+PgpzdHJlYW0KeJxNz0sOwzAIRdF5VuENRAqYZ/CGOkr3Py34UzM7EnpX4jLjIq2V76XVht4h4sfVwElqz7oeQdgVlaNVgZWbSDxNtvgOUoOLFUl37XtzNCfohQ2ecahEmkSSYDGoHUnzKmpJEfnX1oDi8xGhnavYkaN5/fwAnX465AplbmRzdHJlYW0KZW5kb2JqCjM0IDAgb2JqCjw8IC9GaWx0ZXIgL0ZsYXRlRGVjb2RlIC9MZW5ndGggNTExID4+CnN0cmVhbQp4nDWUS5LkMAhE9z4FF6gI8RGSzuOJXtXcfzsP1LNwkRYCEkjX3C5D5pFPhEw3WZ7yRx/bq4/+PhYGctHDJZw6t3zM5X1UuXNkh+zKQPD7YHWEfC/YkkNUVdIw5c8AuOTq96VcOpVqLFlEUGviWkuo/T7GHVexOGJTbLd5HycjwMu1xefgqePcQgLfUyYGvlnHQb1VnpDjEm7NiPOAG/FBKY11bZ72hNiqTDyEGS2XhRAlC30vOrMQw/lFtFXIDowOMdCPUW27ByjhCQMN8TWaEJ6dcCovJ/QZiiUPLAxEL0Grc+xr/XpAW7nLZCsbt/KOYFbTjMZSMpPxXAvzGffEiKnlsOCMyqZwShhoeXTKYQe7xsDP5BYpJg9rrjeO7R4rSx/XVCOAZAHf5wAWOhmkhVipJE/rhZGtwcojsZthBwqoWCV40a4xhGVFEuujaEchVpBXmgaNsiWC1chjY1kUNVePNEblZax2qOM91mzqwXoT0dVYE9nEgZtVttlapTP4zBXXRiu6kKJwvfcn2WN1Nt4CucYuEeFd+bvUyFro7soliHDvJZUYYUz+lkfJZZWt6VykDDAsW9TBuXtVRWpjNJeNMEd/qIZ0y+pqloXqEyZWqTF5ToegumLDS9A4wXbFjwRg6/IpAbGfT8+huHy85H6/////A+/z8w89MsaICmVuZHN0cmVhbQplbmRvYmoKMzUgMCBvYmoKPDwgL0JCb3ggWyAtNDUgLTI4NyAxMDAwIDgzOCBdIC9GaWx0ZXIgL0ZsYXRlRGVjb2RlIC9MZW5ndGggMTQ1Ci9TdWJ0eXBlIC9Gb3JtIC9UeXBlIC9YT2JqZWN0ID4+CnN0cmVhbQp4nE3PsREEIQgF0NwqbGBnBEGgoYu8/tMTdQ+yN34QKEK1E9VvWRDudRZgSxroImhJJyWjpMG6NIyTRNuSKCR16DcNEfL9JeRLzSIv/InV1p67zHi3jiRPo+4sygyrw89wnYMI9Kahk2qD+gDiasEmh3MT2rhTkvaUnf6FA+6Zoe4NaBh4SN9poTPs8wO7TU2ZCmVuZHN0cmVhbQplbmRvYmoKMzYgMCBvYmoKPDwgL0JCb3ggWyAtNDUgLTI4NyAxMDAwIDgzOCBdIC9GaWx0ZXIgL0ZsYXRlRGVjb2RlIC9MZW5ndGggMTE2Ci9TdWJ0eXBlIC9Gb3JtIC9UeXBlIC9YT2JqZWN0ID4+CnN0cmVhbQp4nE3OSw7AIAgE0L2n8AJNRH72Ql3p/bcFaSO7FyYDFCGpfGNdpRNvzS0abPL0KFJgmwFYw6UaDU0Ska9xFKkiJ/mWWWjcSdj82hiQ1NFTSIiMWkuKHf93dpWpXmBn1qZ2T1EPgPw5pZa0C6k7y/MCC1U4NgplbmRzdHJlYW0KZW5kb2JqCjM3IDAgb2JqCjw8IC9CQm94IFsgLTQ1IC0yODcgMTAwMCA4MzggXSAvRmlsdGVyIC9GbGF0ZURlY29kZSAvTGVuZ3RoIDExNgovU3VidHlwZSAvRm9ybSAvVHlwZSAvWE9iamVjdCA+PgpzdHJlYW0KeJxVzrsNwCAMBNCeKVggEsZfFkpF9m8DJglO96TT2ZeEJLNavlIldnUXiQ7NdGulwJKx4WhMqaI3NEgYn8bWShU5aF7piawFVW1DZhBVeP6N8IxKCVo33nXjA1s+AMyH4mJ3AvpSbEEHfR36tXs6b6tCOPAKZW5kc3RyZWFtCmVuZG9iagozOCAwIG9iago8PCAvQkJveCBbIC00NSAtMjg3IDEwMDAgODM4IF0gL0ZpbHRlciAvRmxhdGVEZWNvZGUgL0xlbmd0aCAxMTcKL1N1YnR5cGUgL0Zvcm0gL1R5cGUgL1hPYmplY3QgPj4Kc3RyZWFtCnicTc67DYMxCATg3lN4gV8yBgxZKJWzfxseiaD7xOlOjENnstL8jE0cuiHSZfK0lCmw3ZZawyWSDWk6LL9GKVNBbvKV2ruDYU0CG6YNjhvYIWhAYINqV4a+UArI0fkA5L+YvEFA18ZX00P475Sy8v4C/1E56wplbmRzdHJlYW0KZW5kb2JqCjM5IDAgb2JqCjw8IC9GaWx0ZXIgL0ZsYXRlRGVjb2RlIC9MZW5ndGggMTA1ID4+CnN0cmVhbQp4nE3NQQ7AIAgEwLuv2A80ERCo/2l6sv+/FjRNvZjJLoieggomHKJQ63BmXFQiIXE8CY5oFFGZ5ShNbRNXD+Xqr1xd7a88EL84bVotf5Dm0C5xNR6uFtHEGtdN2mnKYG3fHOV+AXn0KHgKZW5kc3RyZWFtCmVuZG9iago0MCAwIG9iago8PCAvQkJveCBbIC00NSAtMjg3IDEwMDAgODM4IF0gL0ZpbHRlciAvRmxhdGVEZWNvZGUgL0xlbmd0aCAxMjkKL1N1YnR5cGUgL0Zvcm0gL1R5cGUgL1hPYmplY3QgPj4Kc3RyZWFtCnicVc9NDoUwCATgfU/BBUyE8qMXclXvv30dzLO4+5IpDG2uQb4H3U1UUyNl3aaQLj0pm5G6zwkoYs8JL/LQ/8SrJ43uRdiy9o12HEz9lLk5JR1pxczWK0wybdwlT2E6GV3mC5z1yJZE8Z3Qj1Ag86IXm2NHuFWhqrSOdv0AeAE9lwplbmRzdHJlYW0KZW5kb2JqCjQxIDAgb2JqCjw8IC9CQm94IFsgLTQ1IC0yODcgMTAwMCA4MzggXSAvRmlsdGVyIC9GbGF0ZURlY29kZSAvTGVuZ3RoIDIyOQovU3VidHlwZSAvRm9ybSAvVHlwZSAvWE9iamVjdCA+PgpzdHJlYW0KeJxNkLuNxDAQQ3NVwQYWEOcnqR4DF+32ny7Hdwc4MEzM5w2pEYfwY/iMVlEL71EnH2rtKbU2H+pFt7/2QzbnPf6Jqk2I8hkvgtYFP2CKU+BOkISZ4xqMhJWmliAzwHPg2Z3uxzRYTEQuWBqSpzsVSBkynajY+huWhTq92Ur3MlCa6Ur5hsdBnt52XczUFVlKeXOFiF33diG4FKH3AuEBO0udWPv2kT7lkCLxjnON9MTu94qOJVxojDIYUwZpja22KqQJTcXtiKsP2nJV5o3ueNY7s7HUPFMfHZyJ3/e8xs8X8nNSNAplbmRzdHJlYW0KZW5kb2JqCjQyIDAgb2JqCjw8IC9GaWx0ZXIgL0ZsYXRlRGVjb2RlIC9MZW5ndGggNTc1ID4+CnN0cmVhbQp4nDWTy3EkMQxD7xMFE5gq8aNfPLPlk53/dR/Y9qGLakkkQQCaJ23YTHtX2qywncv++avusRXXfl61r60aVlPxWuW2NZd9XuXT1grL68Rlyaf4eWWOXsUlZ26L7V3j8wrarEgL52QMc25MTj8vp48A+Cqb45oDqp4TdivTHGh5uZHTcqqag0Wrb+4k/Yss8OzTVXRXdafVWObnWPmxGGUVR1iCqZIdqlSBkqmKrpwoh3njsqMJyVNkslnPztENPgdlKafKqcZH53I+mH0QzCFMsOygLaYLIjc4oXOc6rlj+hNdDDYTe6EMyKk8g9lghRNHJAALGmrsgl21Z3DgsH67JCGGWuRiJeIgTRpLkncDDkbuHch5p9oVuX+EO9oI7zEStsY4ziYKQ26LCKNTeJCZrvAWHBHygSk9mFmkuzH9w98BJZQ50k8I9pg2r+47Mq5khxuLfhHE2zIt8kN1T7tTMisy2338WgDdKSF0UzkSQTl1lm0InYMbY7cY2Je7389K/THEgqs/y3/aTEG3n145QnohA+PIlJ7q0H9Udka7clfZaUuqivwtF2Mt6rbvmxiTsWgpEULkZfR/1tFpLrGlbbyKniVkz3hHWhbofMhmYo5Ys7WQ+oizZMl6jE6M1rJXqROn0+2cOH1ClRxwpM6wrceRPZuei4yeV5gwwZaBd9ufMQ47GF3s51i/Jg+er1Z6RLoRUqpzZGhVCTTPxqf83U+TajxCXyJH7OKl3ezW7UoutPLmryKf19d/+ULfQAplbmRzdHJlYW0KZW5kb2JqCjQzIDAgb2JqCjw8IC9CQm94IFsgLTQ1IC0yODcgMTAwMCA4MzggXSAvRmlsdGVyIC9GbGF0ZURlY29kZSAvTGVuZ3RoIDMyMQovU3VidHlwZSAvRm9ybSAvVHlwZSAvWE9iamVjdCA+PgpzdHJlYW0KeJxNUUtuxUAI2+cUXKDS8GfOU6mrvPtva09avbeIcPBgDFwxIVUtryv2SEVKakuZS4ZKLZfvKwvZQWabZOG9IcaAqQRykyKrI60luQpMg48d0rMkpmRUEY0MFEJN7qsdOVTV3hKGWKiJRV1vCWjXQgbu8qgcL8gG3qche7qhsyeYaGQT2gmmER3/m15CgdrEN5jY4o0ao3/HX64Wx7wxI7ZTounSig5SDCqBTZjCgW0wSi/YkdLJ2qKGXis4Gcy2YjA1yOCZFoohaRg+lSbtLAwZNMsccVqb+rNSGsdcBeLG+rtpf3GxbFrSEU8crinJAN1AW7rWGaZhl2tp48K4rIa92DwTfT0HRzU0s3h5U4L7ADrEBB/AjdzMJ3pIKrzRkWjc8Etx1Nel6Q+8D+R3X9znG32F/9e80VPy8wukm34NCmVuZHN0cmVhbQplbmRvYmoKNDQgMCBvYmoKPDwgL0JCb3ggWyAtNDUgLTI4NyAxMDAwIDgzOCBdIC9GaWx0ZXIgL0ZsYXRlRGVjb2RlIC9MZW5ndGggMjAzCi9TdWJ0eXBlIC9Gb3JtIC9UeXBlIC9YT2JqZWN0ID4+CnN0cmVhbQp4nD2QuZEEIQxFfaJQCKCbeLpqLTZ/dx9dM2t0oeafYnQtKZXfcYcpZ+j/wfUZH/wM75CcBjPUJEwlaoqDpaa4qjwjW8W4KW8xc2mYtiZIZYl64FN4KH/pYBOX3mJxOYHCdol3i4e+XywHcSUpFt75Mox2FjdRM3DBe7ooycuanAZpKIjPWMZE3VUIg/BFuboUjS1+62wXN+pQ16n2DCMs5g1kVeekUvAAVIGfio61sqk1t5Rdje9+pzNiXowldrxrfp/uGT9/m1VFDAplbmRzdHJlYW0KZW5kb2JqCjQ1IDAgb2JqCjw8IC9CQm94IFsgLTQ1IC0yODcgMTAwMCA4MzggXSAvRmlsdGVyIC9GbGF0ZURlY29kZSAvTGVuZ3RoIDM0MAovU3VidHlwZSAvRm9ybSAvVHlwZSAvWE9iamVjdCA+PgpzdHJlYW0KeJxNUTmOxDAM6/0KfWAB67T9ngBbzfy/XdLOZLZIRFiiRFFtrJIfdZN3G9oPfG04ZidC/ovudFrIGB2cYuFguja7rzv9D/oMoLB8kItNl+gp1of4UNFMuZr7EO3ILnYVwzCEq+lwAjWmQ5ajfuEdP9OQMDB2iN7xDJDojWzOZHUFn0mvlbK7UQPbI17NZm3ktmSghRc4WVSEpUtVQksyUqg9zmxsRvR149U8dEt4H+Ql7ksCOXfo6bp3VEl8DlbmFGfnSX0OL6pzF87EatOknByriR2gOBURGT8RGT1IsRcr9OEoOeiimMGumv2eo/QHrwp9VMKbHm1qc6ulU1S/Y/94R1dYQZfIsUUF7EJ97Gpx5phtDmcsfMxO3dq0NodXSBxYeXnd+zBin7xfyGHFw6Er7EKX2JW+nTl0kpPpLHn0+mij+1TLa1D95z5X+/0D8w+N/wplbmRzdHJlYW0KZW5kb2JqCjQ2IDAgb2JqCjw8IC9GaWx0ZXIgL0ZsYXRlRGVjb2RlIC9MZW5ndGggNjY5ID4+CnN0cmVhbQp4nD1UW44kMQj7r1NwgZbCM8l5erVfM/f/XRtKK01P6KoAxjadx2WJb/lESEbK9pQ/+thVsZTfx9UZOB7bEU+EKd/Hd0qY+HVJk1hXio/DQvaW8BK8CdRTVT5PF7UrkUc015w7+g2iW8IbprtzDOl4g/+WbBNie5rY4Rt29UXYQKNrUOnhG4BzKyFq9yUcg+f3sdrzJKpvWN9ijq3qKnpOV9Xtbx/N3Z2VmNBDwdNgU3BBtApsRN9nz9MRJuQNTsyc4YBVAJxFQRG7NGVsCwqJAoQ2LD4mTjYBbND/yvFtXWIZhQEdjBy3YmFsJEQ38gpEGPeAaGp38R0wQOnC3cBHkVuglLn7vmTHBdmhkhpzjgyMsrpCtupX8lqLrVKsCPLKQDiIqHHHMakAAqAv5HrAH2UtEHIKRANpgVrOwxOT7TVPcm6Y+5tjml1F7+2qevLto1u7s9YgUZhosJFsotUXvf6fR98JeYMTa8bLgZIRfCgkWaIyw5stayZNTzNLEwzXxmywPwadeXiOUlyqX0h5OjJkfJzGv/Kx01iweEqj8K4o3UcT4YKDQQzTf6MQgK3gqio45FEtNtDQCdgNuAjI7bDyZQVU9NvkKzcG9dCyo5QfuJtl0JxWwf07fvg+xW2E2JiM643fhSabAdjBviZaQKFSzokKxXUHXq4yntVunjnB5TKUbGqDldp6W09Edltp/tyQqT2+8YkIkDcc+kwO2d/t3ttVo+7bJ87tzrm8gSRWf5AlnJG9XWeQ85xR6HvOBmdx1LjrnT2wOyQjfMhxVPcaT9+m8Acb4U1qHGuSc+2X9XRrGfhDSlV4jkwdOT14W8bEdC1r6mmVe5NRUscBUYuOCI82CC01lnGsCU3kWI821Wuz7/P3H5nkE+UKZW5kc3RyZWFtCmVuZG9iago0NyAwIG9iago8PCAvQkJveCBbIC00NSAtMjg3IDEwMDAgODM4IF0gL0ZpbHRlciAvRmxhdGVEZWNvZGUgL0xlbmd0aCA0NTgKL1N1YnR5cGUgL0Zvcm0gL1R5cGUgL1hPYmplY3QgPj4Kc3RyZWFtCnicTVNLkishDNv3KXyBqQJ/4Typeqvk/tuRTPfkLVIogCVZuC9fKrmmfK7c0ejdqNYAKvz/IpvrPv0i1w3kaz0oS3xMcR9ihd1RYqnyuqxczFFr1T/dZ31dmnZ2THFDRQduhuFkVqLaZQbUa8s0sO7Fk7HFZ8m2loI7z8R+YnujJZPQ2QtKsQ2AOzhNFBJZYBvVmVNIlsubvkYLo6pUhQaK1mCpnMI0WbGFpiuj2+CKBkFJZIXaYAy44WzdeaLYgVYhkCf010WTvnfHn43ejXIqUCDmG6lLoK8AS9S9xmEAMrYbEjPkYXxdC8/wM9HE51KINnw3nCNvnf/Q9uf0D2nm/fRfZAQKK3/gx+m61b7oiDFG2vwcZIg60AhamkmbjG0mpgSTMjktCHVunHRrigDdEPXcPV2Kh3BloBqYPDSrmCkHmzK82VOGmSAyy75h0Ds1htjIYqua1ce4dXxaK7taO3GMwfHmUKRb9+O+11NDhIfuG2t3TQ5OIFmSTGBNW62TMegAyolvgU4StfSWtXp08OSL3xVugI39cEWn6LB3IvoGMzg1TIUsHE2yMrejwySpzGTphFkfb50+2Po1asrzPq/r3y83m8KnCmVuZHN0cmVhbQplbmRvYmoKNDggMCBvYmoKPDwgL0JCb3ggWyAtNDUgLTI4NyAxMDAwIDgzOCBdIC9GaWx0ZXIgL0ZsYXRlRGVjb2RlIC9MZW5ndGggNzY4Ci9TdWJ0eXBlIC9Gb3JtIC9UeXBlIC9YT2JqZWN0ID4+CnN0cmVhbQp4nE2UQY4dMQhE930KX6AlAwZ3n2ekrJL7b1MP/69kMWN+g4uiAF97PaPmM/5cWLmfseMdmT62x0h/xs+1Z42ca9RjYz05au2x9pQHK9zH7+uR75/l65Vl/xnH52v9Z61tWDaFWsOepSw5rMTDX6GbWMDEFAszTrjKgxXy5ByV3IlRe+PZ4ve+w941tj3DfaqiJY+vHFvxvmvs8hGmiO3yBNXC/5VHTNY658+1Pl9SmhCRynHulGWjFLGw0NeTp97dmbdVM9mq5XDby5ptay72X+1/rixpoJr+XOnKH48U0V9kc+EUl+lthWolIkLdWHiwwtESnH8WiL+vLcu7wXwS2Ri3ieKa445ur+9xpwiLyi1x6vFxv0YpWYpV98u5pKJS7b/NqSYXThHKaTIEt/JjiJIKbzNynaiw+t50qdFotKTx3d5PSnujSdiOJmX1pWmdb4/O1P/9fJ6jZ0R9iA4XAnOgH6Y//bKYncFaLZJavgMWpgmEmO2350C4jya658e6HmtWlIhFzUQgw7mDMqAgFaiId/IgJ5mRFyYI3tTogNjRDgavPoVUSYetPmGo4tJ0aXsYmOpmaFo0ZBU1lKNUz+mRxsKmoJ6pc5+FMfKkqxyFrVceLd3KczJKT1uxT0RQVt8JdQUU1+CD6vuTxsVGaT0pSI2LQ8rFRiQ9Asocp8MytCDt7eCbbeIyHQTrVhzgd7RH6e7FkGgEVQCE6L4IieJdeisYztpdBCe7eSzKJILCzx2kAAVpQEWrkwfxyIyWMEHcptZiD5QX7U8nmH+FTNayLZErYIO1X736B7JovPMEaRBYl6eHRitcDI3gNh3SAPCUysM4mDJKUZ6JfnCsH5z5thU8NIoIjeC5E56NEjMa1Z9vHq+3M3seJs6z0txcGWFLH2Df51yfvqUEJIKnlTvwpEnyaBFATXXO+/njDpkRGyYp8eCWZt2k3U837FNR1MOpelQhFhUTgQbnDtlAQSVQ0e3kQUkyoyxM0PpwQ33Y0g3Yf/vzc/36C9hkQ7gKZW5kc3RyZWFtCmVuZG9iago0OSAwIG9iago8PCAvQkJveCBbIC00NSAtMjg3IDEwMDAgODM4IF0gL0ZpbHRlciAvRmxhdGVEZWNvZGUgL0xlbmd0aCA0MDQKL1N1YnR5cGUgL0Zvcm0gL1R5cGUgL1hPYmplY3QgPj4Kc3RyZWFtCnicTZJNztxACET3PgUXiGSg6Z/zjJTVl/tvU4+eSNmM224oXhXzzFO29mt/nrW9Tz99+uUZOnL93/HeV4SN19XDKcey8mn5vlbvsKhjn2ecsPBtY5X5HjZqmNfiJtM8jhS2+XsspSnpz5PlpskZry3J7M3j80RVv8fg1s+81T5vu48r50K5A1wgjNzZBMu/SPppxrzI2R70ORtOt6V3VdcIfVZ3bUmqY8od8nPAw8CpMgCmXEK03gmpEJfomzjWtRAgtSedsEgFnm8PKaBCKqiCcueQHJNJEhKyvWikDWvvAfTvPsi2vpvpUyr/lObSfZR6khplUpmdf02xiB99USq0qSmJs9AmSt6Tnkzcl6W7TXnAGU95nuN+IRVVhObcnnhXq/jxVvV1vnN8rp7MuiBx9V82l3NofXjT97OdccqpWlUk62bmYa2oBH8GqYbyYU4oY90cpm+DxJU4bGxNbPxpahr0Lpf44SmncsgJx1SQwe0hFVRICVVyu3NIkskkCwlZXzbSh5ZtQP9vP5/n918xoKV8CmVuZHN0cmVhbQplbmRvYmoKNTAgMCBvYmoKPDwgL0JCb3ggWyAtNDUgLTI4NyAxMDAwIDgzOCBdIC9GaWx0ZXIgL0ZsYXRlRGVjb2RlIC9MZW5ndGggNDIwCi9TdWJ0eXBlIC9Gb3JtIC9UeXBlIC9YT2JqZWN0ID4+CnN0cmVhbQp4nE2SS47EMAhE9zkFF2jJ5mf7PC3Nquf+26mCRD2LyMhA8Sjn2mPKa06X30vnHX4qnLkZnf0vetlBVD3fqFvWHjJ1QCdPVPSpiPed/UadDZMYGw0MfEociaWypiRk39c2ST1yjmQ4enC/AvfTVWpGLhQD9KQsNWRUXZaFaOAGinpGne/LmEFkEVVhO+4eBz1VHAtR1fOZ41trspMAIDEfskAnUYO3JOeJVZg54uhmhXtUj89FNajYRhaqll5zzJ0ZTCYfSQwqZFPsBGrQ6vbK6jq1D09sig3rhhlU0IPuoStUoUvso289hwAYDFvJsR4w2F6k1uB2bzJ5V8/DyHdCcYMavoPKvWepuC6Z8ML5EiC3U92He4CNOy2cxq2zyAdeYYN+1Gvo0TqxLWsRmXlVWDw9xlqo0BmqckbP8fIXn84icfpfbG5ZtO6j6Hn2ZhUZXzwlQquHr4mM8i9sxRw9I5UZTk7UkyShTzb+G8iANuED6TOz9snelLW8iVEV9KB76ApV6BJV6VvPoZOcTGdJQq+bje6Ttt4F9M/7vK+fP7zHtMYKZW5kc3RyZWFtCmVuZG9iago1MSAwIG9iago8PCAvQkJveCBbIC00NSAtMjg3IDEwMDAgODM4IF0gL0ZpbHRlciAvRmxhdGVEZWNvZGUgL0xlbmd0aCAzNTEKL1N1YnR5cGUgL0Zvcm0gL1R5cGUgL1hPYmplY3QgPj4Kc3RyZWFtCnicTZJLjgMxCET3fQouMJL52j5PpFn13H87DydRsmhRwlVQ0FwZQ2Kl/F2NMrfcV+38QnMN0Fz6hSzs9fpBXeW+3vXuK5Tc3lQOW+RMIlyiig9OqjyumFPCm8dnKqkwxuQlXcW342OI0zcnsVa/rC2eJQXPI6Usxf1o4Nj0duEhhjr2Ele6l8FtTliJTxQ7TicvOkd78XanZJT6u9DDmsWLTT0dLep0Mpy6tUbh+UhRGLaHKD1stks1E2NWFmBImdOi88YYqr0ylDZFxz7OlH043rum0x8ZGwgY1g75G2jJxtajSws8sTF2iXcmyva7Tap1g1g9r1cjXthSVW9FT6Ra77AzWOqoy14vax5wN5i+ABb7jXDXyL2ekWk7sll9cgJpx1STGe0ux5JiU/03KtkyF1TaszY3mTq4ouw7eN3M41zUj3ofZkPVorli/gvt9zl+0JHcH/V9/f4D7rmL4gplbmRzdHJlYW0KZW5kb2JqCjUyIDAgb2JqCjw8IC9CQm94IFsgLTQ1IC0yODcgMTAwMCA4MzggXSAvRmlsdGVyIC9GbGF0ZURlY29kZSAvTGVuZ3RoIDYyMwovU3VidHlwZSAvRm9ybSAvVHlwZSAvWE9iamVjdCA+PgpzdHJlYW0KeJxNk0GyJCEIRPd1Ci9QEQqCep6OmNWf+2/nJf6OmEW1tGCSJPBEjBYr298nDlbMlh4tPFtua2G9fZ5lu0WfbRE3j7Vts83t8uze5oj286zAR/Tywbna6tEmSJ8n12lzesvwNsNa4p2p19mJXU5+YjdZA7Sz8ITxrx8+cSBKXAptWnIbzZN8IHsX6sFjTh6+IZYwHAOWI8XSC+DnGUbI6W0IjuTmnEOwVsUCxxcT+IUYRXIOCG4oIFR2CG4kghIkAc/JOxVHbGQgWpE83CDQNCsZPe75eQYl5uWCtZDkh5LO1yKrrIB6nbHqlFQ3OuFWJ21YoVyrPFGCy7P6bmtKxDXwTBB10tDVQXM1NNFgUX/mbrlmVZFzy4MaCetQOw5t/x2Pz7PRwMH/W5YhPTX8Z3hV8I1iIFDEFS5DBWZ7kXIh1ludFJWX28XjVyRI+Z6iRz/f0cVEj8aCzsYw8Ygpp2aUiXzHpCPxayA881Kmo1xF+Tjfl8awFprNffGNSm/KcbJIaHhEauSX5rj52pgqQ7/ude2NDo/kzhyk0Ub1w5i2sRcMEKeGc9YJc3Blpd8IFXzfLF+FIqn2PSpJLioxNJQxkYihpzDpk6eEYhlpIwWzvAnC6KOmhQqQNbbWgIlWRRa1WLsNrejB47PY6dScW1m+boQzn/eNo5JQnEkTqu1vHoMPiQ0QzQRTV7QMPtA0qoK0DrdaUO56k3dX8Ku8Vt1DGbBe+iT019Uv5Xu1i+oSMyVGb626OL5smzi/zHBNQO7avGupTkWo8vtGWghF2ghVat08kk+ZpaaY1PiJmuTWgnTR/u3F5/nzD4Li96wKZW5kc3RyZWFtCmVuZG9iago1MyAwIG9iago8PCAvQkJveCBbIC00NSAtMjg3IDEwMDAgODM4IF0gL0ZpbHRlciAvRmxhdGVEZWNvZGUgL0xlbmd0aCAyODcKL1N1YnR5cGUgL0Zvcm0gL1R5cGUgL1hPYmplY3QgPj4Kc3RyZWFtCnicTVG7jUMxDOvfFFrgAOtvzRPgqmT/9ki/C5LCMGFKJCVfe5Z4mryudj/oedCPOmFFf8Pe65//oLANRJ0PunXUQ6qpTdRBNjS/ENlP3fPyneKe6Ah1sYF6hFiWRI+YpTyu1BAFk6miFZLwU28yhoRKaT4tBMoExWCOUkOJj55mL4XcPmZWBcamxHTDDLUNvAq9A0ZzQw19uqRGCnrOZzwGzVrUSmjJVuxKN3NqI7k2DqyQ1QJWm4KGWRwtvhpBODEkF3s8OTFe4BPcRbek0c0RJU3RY7jXCc0bajYH6dyMYuibyXXA84BhVlN7I1YB2exze9wMnHYdFK73jbQ5XFFgosTfRWH7yBEYPZITsZb/GViRo+r9j4/r9w9M+3K0CmVuZHN0cmVhbQplbmRvYmoKNTQgMCBvYmoKPDwgL0JCb3ggWyAtNDUgLTI4NyAxMDAwIDgzOCBdIC9GaWx0ZXIgL0ZsYXRlRGVjb2RlIC9MZW5ndGggNjE2Ci9TdWJ0eXBlIC9Gb3JtIC9UeXBlIC9YT2JqZWN0ID4+CnN0cmVhbQp4nE2US24mMQiE930KX6Alg8GP8/zSrJL7b+crnFFGSgRpoCgKnMdntnW8fT+5T3lf17ODV9Ffr6IWq815qAjP8r6emPdb9n1t9rKfJ6mRN8e8dvP3MiKTLnORd4II0SOMTmR1vma0srFvJEbVjDZNPUZL1Y3V8qJZb0luLrr7xCejTzGgY4iFR4stXrtFcVtGJHdb/EactsmPcEWYdqzFZMtGG1tzWhviYdbCxCXhGk43oaBJMIN0+DyhHHBGDjKs+cH2JOJDaKfZmuDvZk7WCiKb4rEl77gkbEMXGX0APzWiQz4WBTUIBGmYXYIFsqcldXwZVmNmlPwWLSGVICY1KaF70Xdk7OQew4KcWVb0QTsHonZtZNlPrV7e7xF8PZuswZDf5XlINPvPUew3C0HpU+lyXm+Lpb0+2hr8WUteiPKy4AXPVwtjEe+RSDOtvaYbg+JrzpTEcbYGCgWl3pk40UuOcj7PoL7cgSyVNcz+VTqiFppHv/iObLel7V0kjKsRKct/NE39IG6j5pBxrUkOyhg7s774phSdhZ6Qof+o7VrRkxX1Xd70m6GJb83iwIQisQD9EQ8lpjcIf5fDxqdOmBtHoKyHEnoc/CCwnky22+jsxvkl53dmvRfr9R0s0/VzhtbPpdZPnVaUN9bNKPJVM9AIEOdpAenrp4XrQQ2knCLg41xCdcZwZh7oyrj2ImfXjSOhktmDPmtznIHAXoJCf2uRavfCW91f1BCft/5riOGLxGL86o0LY2q/mkqeplSG5r41EkIo0kWoEur2kXLqLCHF5J6e1xHqYJuE3+1nD5/nz19qxvyTCmVuZHN0cmVhbQplbmRvYmoKNTUgMCBvYmoKPDwgL0JCb3ggWyAtNDUgLTI4NyAxMDAwIDgzOCBdIC9GaWx0ZXIgL0ZsYXRlRGVjb2RlIC9MZW5ndGggMjI5Ci9TdWJ0eXBlIC9Gb3JtIC9UeXBlIC9YT2JqZWN0ID4+CnN0cmVhbQp4nE2RTa4EIQiE956CC0wiv+J5Onkr5/7bV+hMTy8MnxQBLBu70nCndxt2aIGYIoOGJoUYopFH0tWGTHIWGqJkgXoeZNKhZPKmtUltgvgBRwuVm5ys5mCGhVCALQWdAnt4r0yQS8Ukt61wJ1dkA1soQ3WcCcWybthFKxOkGTtejasbaB2aArJuX/K+yebJ7O7gq/nA1BkUeKHPSeFG0UvZhGzAq9COiCo7CsjHOaOfmKWUy0Wrff0uYnoxDHlvnFmuoMMNnONT9iPx/HzQj7T8FGx+wyu4tPAn1ajH1NX+/gHalWDICmVuZHN0cmVhbQplbmRvYmoKNTYgMCBvYmoKPDwgL0JCb3ggWyAtNDUgLTI4NyAxMDAwIDgzOCBdIC9GaWx0ZXIgL0ZsYXRlRGVjb2RlIC9MZW5ndGggMTUyCi9TdWJ0eXBlIC9Gb3JtIC9UeXBlIC9YT2JqZWN0ID4+CnN0cmVhbQp4nE2PSwqDMQiE957CCxRifJ/nh67S+2+rIaVZhPlgzIyCsiCH4QeaxAgXWOpFHqPIgy56Ec9jX9g5C36JC6YLZkf7bF2tPhrE8iJjLiKNi0S6S8QvYvPj/onqfzWFH2LhTZzlVZ8olQY+oEOQQlBVkZLRhuKs90AT1da1e3uTUMdEIqpuQ9ozHIlZsbxPOZc98P4C6p89mwplbmRzdHJlYW0KZW5kb2JqCjU3IDAgb2JqCjw8IC9CQm94IFsgLTQ1IC0yODcgMTAwMCA4MzggXSAvRmlsdGVyIC9GbGF0ZURlY29kZSAvTGVuZ3RoIDE4OQovU3VidHlwZSAvRm9ybSAvVHlwZSAvWE9iamVjdCA+PgpzdHJlYW0KeJxNkLkNAzEMBHNVwQYMaEnqYT0HODr3n3opyzhFGmj4lxFdXjCVTxnAD++FY9Yk+oe2bkY9KnM6+qJ70Qs1tj7QppMcdpC3tu1Du86UGFl5yqipLPygvgZA6we5ZSvrJ1kf2z6EkRla65/2n9XJt4k1F0yTq3jGBFY9xBTvIcoZaEgwyzWaCZTOmKVcCSpAMMaC2WA9zgFe0pyGPWgwSC4a+aOiGcFeV1GvwkYKz/XBCnyufY27vL+azVDYCmVuZHN0cmVhbQplbmRvYmoKNTggMCBvYmoKPDwgL0JCb3ggWyAtNDUgLTI4NyAxMDAwIDgzOCBdIC9GaWx0ZXIgL0ZsYXRlRGVjb2RlIC9MZW5ndGggODYKL1N1YnR5cGUgL0Zvcm0gL1R5cGUgL1hPYmplY3QgPj4Kc3RyZWFtCnicTc7BDYAwCAXQO1OwgEmhFHQhT3T/qxA1cOIlH/IBUkLmiRtSRgMdbHKTTg2xWNN6U9ImkfGlJb5WpqJd0ebw9zqclr3xQOKgvKxpVBsO9wN0CSGaCmVuZHN0cmVhbQplbmRvYmoKNTkgMCBvYmoKPDwgL0JCb3ggWyAtNDUgLTI4NyAxMDAwIDgzOCBdIC9GaWx0ZXIgL0ZsYXRlRGVjb2RlIC9MZW5ndGggMTEwCi9TdWJ0eXBlIC9Gb3JtIC9UeXBlIC9YT2JqZWN0ID4+CnN0cmVhbQp4nE3OOxbEIAgF0J5VsAQB+bihVGb/bWAmGivu4QkIJIbGgTeUvBFOcB2HKp2w3qW4IRvVREqkZ4+7bzXUoSlhWxLKVFJqbWu8qaku9f5ucT9V1yb0iEPOiXD60Or6Lr+2jPjwn12/n3A97ikwswplbmRzdHJlYW0KZW5kb2JqCjYwIDAgb2JqCjw8IC9CQm94IFsgLTQ1IC0yODcgMTAwMCA4MzggXSAvRmlsdGVyIC9GbGF0ZURlY29kZSAvTGVuZ3RoIDE0NQovU3VidHlwZSAvRm9ybSAvVHlwZSAvWE9iamVjdCA+PgpzdHJlYW0KeJxN0LkNBCEMBdCcKmhgJIwvaGgjpv90bY7B2RPf8kEChCyN8ptcyi2PpMJBno506ky1ZLL0nWIs9laJj7BnYTBh5U99pyx0pLBTYf10+qlyEM1dqPUgVE9bg6Dau+8XsDIqJWj1OFeYbPYDWP0gY8O5iVwA9112VddyFAU875eLR2B/YpCPClNH+v0BDERK9wplbmRzdHJlYW0KZW5kb2JqCjYxIDAgb2JqCjw8IC9CQm94IFsgLTQ1IC0yODcgMTAwMCA4MzggXSAvRmlsdGVyIC9GbGF0ZURlY29kZSAvTGVuZ3RoIDQ2NgovU3VidHlwZSAvRm9ybSAvVHlwZSAvWE9iamVjdCA+PgpzdHJlYW0KeJxNUzluAzEQ6/cV+kAAzaXjPQZSbf7fhqQ2sQtDtDTkcKjVZavV7O3nAhjp7b5q9A/Ew/t6qgCi2uzGeqIZ2MvMT4TTd919+VrNtoPh1YXcJtbRbO3mvdrrsmHYzWYxmztODCewgBNoeM22cDAhuZpvMsZuYdVGtYihpRa3ARbs75a9szq9YxvsTDSHWEKM8gnDkEfDpEEYKDBoqfrGCU0Wxqbp8mgcg+vrChgkitqqCKgfTsIRVTKWVLPG0ydnqrN24ST3fKxVH/JaZrKuVbMIYThWcFZyzvBUYRpUZTjsc+JiZwZIJwyU3k7EdMvQ6V6X8FzLSxfkgxcU5kIRmHeAXQh/0mWM0zEWmajYqNjGjh11nBncMK4TPeTFqYS5HbwcZ9VkQtBVxbLD2coJKlTUj/6xmyd1dM1APZxkbHnL1GRwm/hH91mpybhishFnJ10Vjr6H431KxagPVaO6+tg0dTZy4cLg+HizCrk19oJ7rZpHCBOqIh6OMqAKU6Fq6PvyJzd2ZpJ0wmTp7WRNt0yf7nkbf/eDz1jJ8/HNWEK3EOg3Pn3/AKv/nf2jHBOIGm/0aMDvl2FovOr5wFvQOpHjWb3RF3J7OG90KN+/bF3OvgplbmRzdHJlYW0KZW5kb2JqCjYyIDAgb2JqCjw8IC9CQm94IFsgLTQ1IC0yODcgMTAwMCA4MzggXSAvRmlsdGVyIC9GbGF0ZURlY29kZSAvTGVuZ3RoIDQ0NAovU3VidHlwZSAvRm9ybSAvVHlwZSAvWE9iamVjdCA+PgpzdHJlYW0KeJxNU8uN5TAMu6cKNzCA9XPsegLMKdv/dUk6b/JOJiyJpOjksNlGn+3fQXB6u48a/RuheB9PF0BaO22xn2gW7jLzG6H69t2Hz9miOya8upDbibOazdXCol2HDWvhvVmgEjgxGemsdFQq20RhLMi0gNJ1ACdKo1qil0dMXgNUNVaxAbpz8hrT1bORrLAK6QsyoIdgQYgGqlyWuPV10GSNs9F0nda4Bs/rCB9CUUsdQXbNpJVY6IWsWePRyTOlTDt0kut8rFUf8lpmsq5TuwjRKTvgnDN7eU6GdkyFQ50dFycZIJ0wUHrbEdMtQ6d7PcLzLNd+oJF4oGD8QAE/wQ1XgGFos4AH8DrUkIxOdKNClIjd6Z9VuB2hGdNm2TE7u9hyciYmbsk9kMiCr0L6XS4jgOiZifSduuk9wKkb7MkO93pmvG8WWylWmx8dozqUDTvRiSGD7c2UGr4Ivhbc69Q+QtjQhmtjsu0MvKdS8VhK6ZMbPi/uM/gTncif6Bbyxb9ogPFFJ/zs6osKv80tlhc9LDXbjzn/HoO04C24hgjjBTbWU3uR50fjC+mvxCfzB36Q/6P1oi31+x+q6scvCmVuZHN0cmVhbQplbmRvYmoKNjMgMCBvYmoKPDwgL0JCb3ggWyAtNDUgLTI4NyAxMDAwIDgzOCBdIC9GaWx0ZXIgL0ZsYXRlRGVjb2RlIC9MZW5ndGggNDc1Ci9TdWJ0eXBlIC9Gb3JtIC9UeXBlIC9YT2JqZWN0ID4+CnN0cmVhbQp4nFVUOY4DMQzr/Qp9IIB1+XhPgK1m/98uKc8g2WJgwpIoinLSLKZM3fLbCm2Tqw23b4To1Z68q+2U0Q0FBIM3c/1D3T5ZuNIu4RP5hUAxe0h0l7FDfC15t7GApoFkieeWEUM8KmKI+JDcU9ym5HCcikjaFlcwgcFVJaLX+W4O6USeWhmu866x5cVimcVq7ncfq/wtuo8SXY82nUetYkCqrxPzIEKUiDADmDWBDxGwxN5C1lSrPmnOPuicHkIlGV7aMjZVQ21SF9QnlHAenu/GCYk4MTPowamhK2ShS2Slb6cPnWRnOksl9PpoK/ehtrYB9c9+6DXRwKYKBaqhLtBp0N8czOmJavTfo1hymcRMakHHWJwBtXgGCZ9jlUq4FxsffcK74GQ8MVk/Nw4GZnjaXeOYlSy1WbAaJj196Bw72/RSYmPe2mxoqbVyz87pFSECY2WYnRqtCFn6rn1wy+zjm27UC9r4+M7WLG1eDlKtL8zQ+VbOpnhinnluODEz6MGpoStkoUtkpW+nTznJXwOd7bfXpa3ch9raBlQ8+0ENduaJRYXx2eJXRqBzAukXMLyHq631jU6QDB9UFBPSXqoLtIp3UfAqqJCHfwDfX+gV/tR80Cn5+QM0xtE5CmVuZHN0cmVhbQplbmRvYmoKMTQgMCBvYmoKPDwgL0Jhc2VGb250IC9OYW51bUJhcnVuR290aGljIC9DaGFyUHJvY3MgMTUgMCBSCi9FbmNvZGluZyA8PAovRGlmZmVyZW5jZXMgWyAzMiAvY2lkMSA0OCAvY2lkMTcgL2NpZDE4IC9jaWQxOSAvY2lkMjAgL2NpZDIxIDU0IC9jaWQyMyA1NiAvY2lkMjUgXQovVHlwZSAvRW5jb2RpbmcgPj4KL0ZpcnN0Q2hhciAwIC9Gb250QkJveCBbIC00NSAtMjg3IDEwMDAgODM4IF0gL0ZvbnREZXNjcmlwdG9yIDEzIDAgUgovRm9udE1hdHJpeCBbIDAuMDAxIDAgMCAwLjAwMSAwIDAgXSAvTGFzdENoYXIgMjU1IC9OYW1lIC9OYW51bUJhcnVuR290aGljCi9TdWJ0eXBlIC9UeXBlMyAvVHlwZSAvRm9udCAvV2lkdGhzIDEyIDAgUiA+PgplbmRvYmoKMTMgMCBvYmoKPDwgL0FzY2VudCA4NTAgL0NhcEhlaWdodCAwIC9EZXNjZW50IC0yOTkgL0ZsYWdzIDMyCi9Gb250QkJveCBbIC00NSAtMjg3IDEwMDAgODM4IF0gL0ZvbnROYW1lIC9OYW51bUJhcnVuR290aGljIC9JdGFsaWNBbmdsZSAwCi9NYXhXaWR0aCAxMDUyIC9TdGVtViAwIC9UeXBlIC9Gb250RGVzY3JpcHRvciAvWEhlaWdodCAwID4+CmVuZG9iagoxMiAwIG9iagpbIDg5MiAyMjQgODkyIDIyNCAyMjQgMjI0IDIyNCAyMjQgMjI0IDg5MiA4OTIgMjI0IDIyNCA4OTIgMjI0IDIyNCAyMjQgMjI0CjIyNCAyMjQgMjI0IDIyNCAyMjQgMjI0IDIyNCAyMjQgMjI0IDIyNCAyMjQgMjI0IDIyNCAyMjQgMjI0IDQ0NiAyOTcgNTgzIDU4Mwo4OTIgODkyIDI5NyA0NDYgNDQ2IDQ0NiA1ODMgMjk3IDU4MyAyOTcgNDQ2IDU4MyA1ODMgNTgzIDU4MyA1ODMgNTgzIDU4MyA1ODMKNTgzIDU4MyAyOTcgMjk3IDQ0NiA1ODMgNDQ2IDY2OSAxMDUyIDY0NCA2MjcgNjM5IDcyMSA1OTYgNTU0IDcxMCA3MTggMjQ3CjQxMCA2MjYgNTI5IDg4NCA3MTAgNzUyIDU4NiA3NTIgNjEwIDU5MiA2MjEgNjk2IDYzNSA5NjEgNjE3IDYxMSA1OTQgNDQ2IDk2MQo0NDYgNDM0IDQ0NiAyOTcgNTYwIDU4OCA0OTAgNTg4IDU1OSAzNDAgNTg4IDU5MiAyNDQgMzAxIDUzMCAyNDQgODkyIDU5MiA1NzcKNTg4IDU4OCAzODMgNDc1IDM1NyA1OTIgNTMwIDc4OCA1MjggNTMwIDQ3MyA0NDYgNDQ2IDQ0NiA2NjkgMjI0IDg5MiAyMjQgMjI0CjIyNCAyMjQgODkyIDg5MiA4OTIgMjI0IDg5MiAyMjQgODkyIDg5MiAyMjQgMjI0IDIyNCAyMjQgMzAzIDMwMyA0NDYgNDQ2IDg5Mgo4OTIgODkyIDg5MiA4OTIgMjI0IDg5MiA4OTIgMjI0IDIyNCAyMjQgMjk3IDg5MiA4OTIgODkyIDg5MiA4OTIgMjI0IDQ0NiA4OTIKODkyIDg5MiA4OTIgODkyIDg5MiA4OTIgMjI0IDQ0NiA4OTIgNDQ2IDQ0NiA4OTIgMjI0IDg5MiA0NDYgNDQ2IDQ0NiA4OTIgODkyCjg5MiA4OTIgODkyIDg5MiAyMjQgMjI0IDIyNCAyMjQgMjI0IDIyNCA4OTIgMjI0IDIyNCAyMjQgMjI0IDIyNCAyMjQgMjI0IDIyNAoyMjQgODkyIDIyNCAyMjQgMjI0IDIyNCAyMjQgMjI0IDg5MiA4OTIgMjI0IDIyNCAyMjQgMjI0IDIyNCA4OTIgODkyIDIyNCAyMjQKMjI0IDIyNCAyMjQgMjI0IDg5MiAyMjQgMjI0IDIyNCAyMjQgMjI0IDIyNCAyMjQgMjI0IDIyNCA4OTIgMjI0IDIyNCAyMjQgMjI0CjIyNCAyMjQgODkyIDg5MiAyMjQgMjI0IDIyNCAyMjQgMjI0IDg5MiAyMjQgXQplbmRvYmoKMTUgMCBvYmoKPDwgL2NpZDEgMTYgMCBSIC9jaWQxNyAyNiAwIFIgL2NpZDE4IDMxIDAgUiAvY2lkMTkgMzIgMCBSIC9jaWQyMCAzNCAwIFIKL2NpZDIxIDM5IDAgUiAvY2lkMjMgNDIgMCBSIC9jaWQyNSA0NiAwIFIgPj4KZW5kb2JqCjMgMCBvYmoKPDwgL0YxIDE0IDAgUiA+PgplbmRvYmoKNCAwIG9iago8PCAvQTEgPDwgL0NBIDAgL1R5cGUgL0V4dEdTdGF0ZSAvY2EgMSA+PgovQTIgPDwgL0NBIDEgL1R5cGUgL0V4dEdTdGF0ZSAvY2EgMSA+PiA+PgplbmRvYmoKNjQgMCBvYmoKPDwgL0JCb3ggWyAwIDAgNzIgNzIgXSAvRmlsdGVyIC9GbGF0ZURlY29kZSAvTGVuZ3RoIDE2MgovTWF0cml4IFsgMSAwIDAgMSAwIDM1NC4yODQzNzUgXSAvUGFpbnRUeXBlIDEgL1BhdHRlcm5UeXBlIDEKL1Jlc291cmNlcyA8PCAvUHJvY3NldHMgWyAvUERGIC9UZXh0IC9JbWFnZUIgL0ltYWdlQyAvSW1hZ2VJIF0gPj4KL1RpbGluZ1R5cGUgMSAvVHlwZSAvUGF0dGVybiAvWFN0ZXAgNzIgL1lTdGVwIDcyID4+CnN0cmVhbQp4nHWPMQ7DMAwDd72CH3AhO4rirF2y9w9tl3jJ0u+XdOfChAFT1olyOM/jML+tXvf0rceC/4/rPRu2Jl1PvKziY1aWBDWMV/WOk46DGhaNTpPTAhSdjj1l1A5q2BrYfRr822iko4eMhKCZ2MTU5GG/2aclioqspGhsLLOZnSka0WXiyV5F4/AyA3B6iMd8ZWZUwBCT8cvcQ0sswtrdvkeQN8QKZW5kc3RyZWFtCmVuZG9iago2NSAwIG9iago8PCAvQkJveCBbIDAgMCA3MiA3MiBdIC9GaWx0ZXIgL0ZsYXRlRGVjb2RlIC9MZW5ndGggMTUyCi9NYXRyaXggWyAxIDAgMCAxIDAgMzU0LjI4NDM3NSBdIC9QYWludFR5cGUgMSAvUGF0dGVyblR5cGUgMQovUmVzb3VyY2VzIDw8IC9Qcm9jc2V0cyBbIC9QREYgL1RleHQgL0ltYWdlQiAvSW1hZ2VDIC9JbWFnZUkgXSA+PgovVGlsaW5nVHlwZSAxIC9UeXBlIC9QYXR0ZXJuIC9YU3RlcCA3MiAvWVN0ZXAgNzIgPj4Kc3RyZWFtCnicJY8xDsMwDAN3vYIfEOA4iuKsXbrnD22XeMnS75dUIcKAKetENzTW+bQFqvtT971L9wtv+l8zXxPUNB5LG7joNFDTotPpcnqAojNwpIxlgJq2BY5WBt92GtkwQkZC0EzsYmrztP/uyxKuJjspGge9hjmZohHthSd7E43LvQJwe4jHfF4ZFTDEZHyvf+gTq7D2sB8lszEcCmVuZHN0cmVhbQplbmRvYmoKNSAwIG9iago8PCAvSDEgNjQgMCBSIC9IMiA2NSAwIFIgPj4KZW5kb2JqCjYgMCBvYmoKPDwgPj4KZW5kb2JqCjcgMCBvYmoKPDwgL05hbnVtQmFydW5Hb3RoaWMtY2lkMTEyOCAxNyAwIFIgL05hbnVtQmFydW5Hb3RoaWMtY2lkMTE1NyAxOCAwIFIKL05hbnVtQmFydW5Hb3RoaWMtY2lkMTE5OCAxOSAwIFIgL05hbnVtQmFydW5Hb3RoaWMtY2lkMTI0NSAyMCAwIFIKL05hbnVtQmFydW5Hb3RoaWMtY2lkMTM3NyAyMSAwIFIgL05hbnVtQmFydW5Hb3RoaWMtY2lkMTUzNiAyMiAwIFIKL05hbnVtQmFydW5Hb3RoaWMtY2lkMTU2MCAyMyAwIFIgL05hbnVtQmFydW5Hb3RoaWMtY2lkMTU5NSAyNCAwIFIKL05hbnVtQmFydW5Hb3RoaWMtY2lkMTYyNCAyNSAwIFIgL05hbnVtQmFydW5Hb3RoaWMtY2lkMTc0NCAyNyAwIFIKL05hbnVtQmFydW5Hb3RoaWMtY2lkMTc2OCAyOCAwIFIgL05hbnVtQmFydW5Hb3RoaWMtY2lkMTc3OCAyOSAwIFIKL05hbnVtQmFydW5Hb3RoaWMtY2lkMTc4OSAzMCAwIFIgL05hbnVtQmFydW5Hb3RoaWMtY2lkMTk5MiAzMyAwIFIKL05hbnVtQmFydW5Hb3RoaWMtY2lkMjA0MSAzNSAwIFIgL05hbnVtQmFydW5Hb3RoaWMtY2lkMjA1MCAzNiAwIFIKL05hbnVtQmFydW5Hb3RoaWMtY2lkMjA1MiAzNyAwIFIgL05hbnVtQmFydW5Hb3RoaWMtY2lkMjA3MyAzOCAwIFIKL05hbnVtQmFydW5Hb3RoaWMtY2lkMjEwMiA0MCAwIFIgL05hbnVtQmFydW5Hb3RoaWMtY2lkMjIyNiA0MSAwIFIKL05hbnVtQmFydW5Hb3RoaWMtY2lkMjMwMSA0MyAwIFIgL05hbnVtQmFydW5Hb3RoaWMtY2lkMjMzMyA0NCAwIFIKL05hbnVtQmFydW5Hb3RoaWMtY2lkMjQ5NiA0NSAwIFIgL05hbnVtQmFydW5Hb3RoaWMtY2lkMjUwOCA0NyAwIFIKL05hbnVtQmFydW5Hb3RoaWMtY2lkMjU3MCA0OCAwIFIgL05hbnVtQmFydW5Hb3RoaWMtY2lkMjYzMyA0OSAwIFIKL05hbnVtQmFydW5Hb3RoaWMtY2lkMjYzNSA1MCAwIFIgL05hbnVtQmFydW5Hb3RoaWMtY2lkMjY4MCA1MSAwIFIKL05hbnVtQmFydW5Hb3RoaWMtY2lkMjc2OCA1MiAwIFIgL05hbnVtQmFydW5Hb3RoaWMtY2lkMjg2NCA1MyAwIFIKL05hbnVtQmFydW5Hb3RoaWMtY2lkMjk2NSA1NCAwIFIgL05hbnVtQmFydW5Hb3RoaWMtY2lkMzA3MSA1NSAwIFIKL05hbnVtQmFydW5Hb3RoaWMtY2lkMzEwNCA1NiAwIFIgL05hbnVtQmFydW5Hb3RoaWMtY2lkMzExNCA1NyAwIFIKL05hbnVtQmFydW5Hb3RoaWMtY2lkMzE2NyA1OCAwIFIgL05hbnVtQmFydW5Hb3RoaWMtY2lkMzI0MCA1OSAwIFIKL05hbnVtQmFydW5Hb3RoaWMtY2lkMzI0MyA2MCAwIFIgL05hbnVtQmFydW5Hb3RoaWMtY2lkMzI5NiA2MSAwIFIKL05hbnVtQmFydW5Hb3RoaWMtY2lkMzI5NyA2MiAwIFIgL05hbnVtQmFydW5Hb3RoaWMtY2lkMzM3OSA2MyAwIFIgPj4KZW5kb2JqCjIgMCBvYmoKPDwgL0NvdW50IDEgL0tpZHMgWyAxMCAwIFIgXSAvVHlwZSAvUGFnZXMgPj4KZW5kb2JqCjY2IDAgb2JqCjw8IC9DcmVhdGlvbkRhdGUgKEQ6MjAxODExMjYxNzI2NTQrMDknMDAnKQovQ3JlYXRvciAobWF0cGxvdGxpYiAzLjAuMCwgaHR0cDovL21hdHBsb3RsaWIub3JnKQovUHJvZHVjZXIgKG1hdHBsb3RsaWIgcGRmIGJhY2tlbmQgMy4wLjApID4+CmVuZG9iagp4cmVmCjAgNjcKMDAwMDAwMDAwMCA2NTUzNSBmIAowMDAwMDAwMDE2IDAwMDAwIG4gCjAwMDAwMjU2MzYgMDAwMDAgbiAKMDAwMDAyMzI0OCAwMDAwMCBuIAowMDAwMDIzMjgwIDAwMDAwIG4gCjAwMDAwMjQyMzEgMDAwMDAgbiAKMDAwMDAyNDI3NCAwMDAwMCBuIAowMDAwMDI0Mjk1IDAwMDAwIG4gCjAwMDAwMDAwNjUgMDAwMDAgbiAKMDAwMDAwMDM5OSAwMDAwMCBuIAowMDAwMDAwMjA4IDAwMDAwIG4gCjAwMDAwMDI4NDkgMDAwMDAgbiAKMDAwMDAyMjA3MCAwMDAwMCBuIAowMDAwMDIxODY3IDAwMDAwIG4gCjAwMDAwMjE0OTggMDAwMDAgbiAKMDAwMDAyMzExNSAwMDAwMCBuIAowMDAwMDAyODcwIDAwMDAwIG4gCjAwMDAwMDI5NTkgMDAwMDAgbiAKMDAwMDAwMzMxOSAwMDAwMCBuIAowMDAwMDAzNjY1IDAwMDAwIG4gCjAwMDAwMDM5NjAgMDAwMDAgbiAKMDAwMDAwNDI3NCAwMDAwMCBuIAowMDAwMDA0NTY1IDAwMDAwIG4gCjAwMDAwMDQ4NzQgMDAwMDAgbiAKMDAwMDAwNTE4NyAwMDAwMCBuIAowMDAwMDA1NTMwIDAwMDAwIG4gCjAwMDAwMDU3MzMgMDAwMDAgbiAKMDAwMDAwNjAyNCAwMDAwMCBuIAowMDAwMDA2MzQzIDAwMDAwIG4gCjAwMDAwMDY2NjQgMDAwMDAgbiAKMDAwMDAwNjk3OSAwMDAwMCBuIAowMDAwMDA3MjAxIDAwMDAwIG4gCjAwMDAwMDc0MDcgMDAwMDAgbiAKMDAwMDAwNzc1MCAwMDAwMCBuIAowMDAwMDA4MDAyIDAwMDAwIG4gCjAwMDAwMDg1ODYgMDAwMDAgbiAKMDAwMDAwODg2MiAwMDAwMCBuIAowMDAwMDA5MTA5IDAwMDAwIG4gCjAwMDAwMDkzNTYgMDAwMDAgbiAKMDAwMDAwOTYwNCAwMDAwMCBuIAowMDAwMDA5NzgyIDAwMDAwIG4gCjAwMDAwMTAwNDIgMDAwMDAgbiAKMDAwMDAxMDQwMiAwMDAwMCBuIAowMDAwMDExMDUwIDAwMDAwIG4gCjAwMDAwMTE1MDIgMDAwMDAgbiAKMDAwMDAxMTgzNiAwMDAwMCBuIAowMDAwMDEyMzA3IDAwMDAwIG4gCjAwMDAwMTMwNDkgMDAwMDAgbiAKMDAwMDAxMzYzOCAwMDAwMCBuIAowMDAwMDE0NTM3IDAwMDAwIG4gCjAwMDAwMTUwNzIgMDAwMDAgbiAKMDAwMDAxNTYyMyAwMDAwMCBuIAowMDAwMDE2MTA1IDAwMDAwIG4gCjAwMDAwMTY4NTkgMDAwMDAgbiAKMDAwMDAxNzI3NyAwMDAwMCBuIAowMDAwMDE4MDI0IDAwMDAwIG4gCjAwMDAwMTgzODQgMDAwMDAgbiAKMDAwMDAxODY2NyAwMDAwMCBuIAowMDAwMDE4OTg3IDAwMDAwIG4gCjAwMDAwMTkyMDMgMDAwMDAgbiAKMDAwMDAxOTQ0NCAwMDAwMCBuIAowMDAwMDE5NzIwIDAwMDAwIG4gCjAwMDAwMjAzMTcgMDAwMDAgbiAKMDAwMDAyMDg5MiAwMDAwMCBuIAowMDAwMDIzMzc5IDAwMDAwIG4gCjAwMDAwMjM4MTAgMDAwMDAgbiAKMDAwMDAyNTY5NiAwMDAwMCBuIAp0cmFpbGVyCjw8IC9JbmZvIDY2IDAgUiAvUm9vdCAxIDAgUiAvU2l6ZSA2NyA+PgpzdGFydHhyZWYKMjU4NTAKJSVFT0YK\n",
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x360 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "mglearn.plots.plot_stratified_cross_validation()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 교차 검증 상세 옵션"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import KFold\n",
    "kfold = KFold(n_splits=5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "교차 검증 점수:\n",
      "[1.    0.933 0.433 0.967 0.433]\n"
     ]
    }
   ],
   "source": [
    "print(\"교차 검증 점수:\\n{}\".format(\n",
    "      cross_val_score(logreg, iris.data, iris.target, cv=kfold)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "교차 검증 점수:\n",
      "[0. 0. 0.]\n"
     ]
    }
   ],
   "source": [
    "kfold = KFold(n_splits=3)\n",
    "print(\"교차 검증 점수:\\n{}\".format(\n",
    "      cross_val_score(logreg, iris.data, iris.target, cv=kfold)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "교차 검증 점수:\n",
      "[0.9  0.96 0.96]\n"
     ]
    }
   ],
   "source": [
    "kfold = KFold(n_splits=3, shuffle=True, random_state=0)\n",
    "print(\"교차 검증 점수:\\n{}\".format(\n",
    "    cross_val_score(logreg, iris.data, iris.target, cv=kfold)))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### LOOCV(Leave-One-Out cross-validation)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "교차 검증 분할 횟수:  150\n",
      "평균 정확도: 0.95\n"
     ]
    }
   ],
   "source": [
    "from sklearn.model_selection import LeaveOneOut\n",
    "loo = LeaveOneOut()\n",
    "scores = cross_val_score(logreg, iris.data, iris.target, cv=loo)\n",
    "print(\"교차 검증 분할 횟수: \", len(scores))\n",
    "print(\"평균 정확도: {:.2f}\".format(scores.mean()))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### 임의 분할 교차 검증"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "hide_input": false,
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "application/pdf": "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\n",
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAsgAAACICAYAAAD6bB0zAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDMuMC4wLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvqOYd8AAAIABJREFUeJztvXmcFNW5///+zAIz4BLZBQWNGzHcTG6MxhhMxOsSjRFMJHJd4vLzmhiXuPwkQq46ikriLhFjlNyIgWuiJvGKu1FwB1Qi0bgBUQnoIKACygzMMM/3j6rGounpbabp6u7n/Xr1q6tOPXXqOaeeOvXUqafOkZnhOI7jOI7jOE5AVbEVcBzHcRzHcZw44Q6y4ziO4ziO40RwB9lxHMdxHMdxIriD7DiO4ziO4zgR3EF2HMdxHMdxnAjuIDuO4ziO4zhOBHeQnYpD0s8kvSDpzHD9MEnPS/qfYuvmOI7jOE7xcQfZKTkk7SSpJc99G4CzgQPM7CZJVcA04HQzO0XSSZIe7kp9k45/u6QLw+XJkn7VRfkOkbRc0o4dbH9H0r5dcaxCIOkYSS8XW49MFEJPSd+Q9J6k7l2Zb6ki6QJJ9xVbjy1JtN0plWvBccodd5Cd2CLpZElzw97deV3kTH4ReN3MPg3XBwDbmFlONyRJAyTdE+o3V9LTkr6eoy5jgZ+F+XXKMTezd4E9zexfYX55OcQK+KGkTxKOfGTbMEnPSvq7pFckHZGvvim4BzgwSx1N0oB8DyRplqQxee6etZ458BzwJTNb18X5Iun80BZmR34jMuyT9wNoJI9GSbfkufuNwPGdOX5HSPp+eL3+TdKrki4rxHE6yUYb64pzEUXSJZJaJe3UVXk6TrlSU2wFHCcVknYBrgM+b2YfhWlf64KsuwFtSesb8sjnauBVMzs61K030CuXDCJOepdgZsu7IJvvAXsAD0QTw572e4Gfm9kfJX0BeFbSl8xsSWcPamYbgA87m0+hKYSeFkxnuqIr84zQh+CcTS9Q/l2Oma0H1hco+72A75jZcklbA29IetjMnivQ8XKmUNeCpO8C2wNLuzpvxylHvAfZiSsbCJzXQYkEM5sTFZB0Yti7/Jqka8O0zXpcJD0c9tD+BLgI2DfsSfsG8BegW7h+VrISkn4s6bnw96yk/cJNrcDOkqpD3Vaa2YJwn9slTZL0iKSXJL0oaXiKvG8Je9p+lKTX0RGZvpJWSaoJ108LX8dXhesXSroy7NE2SdWSZhPcCKck9UrvFfZ0vyzpQUnbJOtkZn8ys58DzUmb9gbqgbtCudeBp4Cjksp0gKQ3JF0elv01SY3hNoWvz2eH9fmopC+H28ZImhUunxRu/7WkOZJeD7fvEJYN4AFJv5NUG/6/ouAtw12SuiWXK6LffcBXgAmhHvVhj/LNkp6Q9GIod42COPXZYTn+PVs9Ozp2uM8vJf0jtIkHJPWTtK+kd8LtY/VZT+9cSe2Sdg/P60Xh8Z6X9LikoemOFdIH+Fl4vKck/VcG/fZm02viyjD9W5JmhtfAS5L+vzD9S2HZXwrr/9TwfJ8KHBW5zlIdaxtJ94X2+LKkm8P0CyXdHi7/IVIfb0haGqb3VnCdPROep99L2ipTZZjZ+MiD5CfhL21HkYJr9F5J/xuek1cldfgWoSObDO3lcUm/CevrHwqu/eT9x4Q2udm56KjOMiFpD+Ac4KfZyDuOA5iZ//wXyx/Ba9alwP3AdwCF6TsBBkwI13sAy4EvhdtakvJ5GDgpXD4JeDiybRP56Hbg+8B9QLdwvQF4DxDQN9z2LnAZ0D+Sx+3AImBAuP69cL9u4bYLw/RbgMZUeiXpPxfYN1x+AngUOChStv0JQkUsss87iX0i648A9eH6A8DZaep+o57h+n8CTyfJXAtcn5R2QHhuRofrvYG3ge8CJwBPR3TYF1gM9ATGALMiddFO0NMHQa/fh5FjWKRuvwI0RWxjP6B7BruaBYxJWn8iuh9wNFAVLv8YuCtczlrPFMftFcr3DNeHAf3CengnhfwNwKXh8vnAzRGdRgLPh8sXAbOTfgmdekTy2x14BfhphvrZiU2vicHAq8D24frW4Tn9AjAJuCpM7wYMD5cbgVsyHOd7wNzI+oHh/4XA7UmydaHtHByuzwD+M7L9emBiuPyXFPWxfYrjXwg8kkU71AisBfaKtAvz0sintMnQXlqB/SL1/DFBu3ISn7U7URtLPhcd1dl3UpT5onDbNsAzwJBIW7BTpnL7z3+V/vMQCye2mNk0SXcROANXAD+WNCrc3A5MDOXWSnoD2A14qQtVGE3gBDwlKZG2AehnZsuAIyXtCpwO/EPSGDP7ayg31cyaQv3+LGkKsEueejwIjAh7z9YRxGgeF/ZkfhF4nqCnMBPXm1miZ3geQX1li9g8FKWNwHFJ5m0zuxuCnnVJfwGGE5R/ekIHM5st6UMgVejMi2aWCPN4GdhOUm8zW5kkNx94DHhI0q+B+y14RZ0r02zTGOBVwK8kDQL6Ax3FgWarJ2b2oaRrgEcUxOfeY2Ytkj6fLCvpxwRvAc4Nk0YDWwHPhbZYRXjOzWwCMCGVcma2NrL8VtgjfCaBDWXL4QQPhH9Jug72AG4F7pB0HXCbmT2TQ76PAaeF1/gtZvZEGtnfAX8ws8ck9QAOA/pKSvSI1hM48ZjZUR3ksZGwJ/0o4JAsdb3fzBJty8ukv3ZS2mRYd09bGM5hZu9IeoLAgU5+Y9MRKesstMEHkoUVHHQqQWfCu1kew3EcPAbZiTkWxCPereDV+BvAfwBvAa3Rmz/BDbuWwGmrTspm6zwPXwXcYGaT0+i3EDhf0psEPckJB/mTJNFaOnayMvEggQPUDtxJcJO8kcDpfN7M2iKOSzo+jiwn6itblhD0JEYZTOBoJ5Nc9m7AGgInO1s2xmBGnIvN9A2d4RMk7Qb8hCB04kAzyzWmd6MtSTqW4HX0cWa2QNIhwPjO6BmRGRs63f8FzFcQF7oJkg4Gfgj8h5lZmFwF/CzijEflLyLoQYwyIZUswbWxuiP9OqAKeMrMRqfaKOmrwKHAbZKeN7MLssnUzNYA31YQZnNOGJpxcIr8LwVWRq5Dhb+DzCzZ1ggfyLZPSj7KzN4Pt58NHAMcYmarstGVTWOC0147HdlkuDnVtZF1u5Cmzg4ieJMQ5QGCduLLwKVhPUJQN3+RdKOZ3Z7tsR2n0vAYZCeWSPqaPov3hSCMog74V4Zdm4C1+iy29RsE8bP58H8EvTW9w7yqJX0rXP5R2JOVoA9BuECCIxQO2yXph+G2dD04bYQ3XYVxzRFeIOixGg38JbwBP0jwoeBD6fJTQFdc588CJumwUMfPEzhF96aQ3V3SsFCuP8Fr4YcIQlKOk1QfbtuXoGdybo66bCAoW7Wkf5M02IL47/MJ6vDfM+yfrq4h6NF+PnSO64DN4kTzQdLnJX3BzJaaWSPBq+5vJcl8gSBc4PuR3n4IbPGnknqGcnWJ68PMJpjZvkm/B0K5/0qUUVJfgpFTbs+gahtQJakq3PcRgjcYX47o+U1JNZIOAmrN7CHgAuDYSB7p6hhJ+0jqY8EIMj8mCDUYnCRzLEEIyjmJNAs+bn0CuDDsIU2MKrNnuP2oFPWRcI7HE9jjoTk4xzmRwSb3DR+QUBBD/o2wLB2xybnoqM7M7IEUZZ5gZqvNbOdoOvA+wQPD7YUov+OUC96D7MSVFcBNCsb1XUPwMDfWzF5XmiGKwt7UU4HpklYS9HC+kI8CZjY9dCr+Kmk9wY3uf4EnCRz2FxR8ENhG4OxEP/JbAzwcOjRrCWJy29P09D4F/Lek1wliKDf2WJqZSXoaqAt7kAB+T+A0dDQ03LRQ11XAt3MqeArC3tGjgMn6bGisE0MnIJnlwDhJQwh67y8NwynmAAMJQlbaCF4rjzSzT7LsAU8wlSCm8iMC5/Xm0OmuIXA2Zir4wOnnwClmljwiwB+BKyWNA05Mkf8tBGED/wBWEsRz9s5FwQ6oAm5MPHABCwjeCAyLyFxPEDMaDWf4CXAVQRjBs5LWEfQE30gwRFw6vgjM02cfrk4ysz8CSPo5UG1myUOdvQfMDPX7p5kdHDqqNyv4WLQKeD089r8D10lqJqj/M8M8HgbuCu35JiDVW5jeBNc4BDG614UPJVGZ/wEWAs8k0kMn7wSCGPi/hfWxDjgvXUWED21XAH8nuKYTmyab2e8VDCP5dzO7LV0+WVBPCpvks28qbpbUj+CB/yQzezeN/W9yLghG9tmszjqpr+M4KUh8ROA4Theh4Av8N8zsF8XWZUsj6QCC2MhsRlgopB61BL3WPzSz94qpS1yR9G/AZdnE7FYCkr5DEHZRkJEeJJ1E8HFopx9YHccpPN6D7DhOObI3QcxuUZxjfTYcXZTvWteMVd1VfIXUPehdiqSvEIzAEeVfHcU0F5FBBGEiWSFpe4K3PVHazGyzIR0dxyk93EF2HKfssCJP/BCGAcQaM5u6hY4zjyCOONaY2a05yr9PCZTLcZz88BALx3Ecx3Ecx4ngo1g4juM4juM4TgR3kB3HcRzHcRwnQtnGINfX1ze1tLT0L7Ye+VBXV7esubl5QDStVMuTqixQuuXp1q3bh+vWrdtkyK/a2tqmtra2kisLQE1NzbLW1tZyt7VlLS0t/YqhU2eoq6v7oLm5eZPzUKrnBsqrXYPU5enevfvK9evX9yqWTvnS0bVTqm1bqnbNcXKlbGOQJVmplk0SZqakNOvTpw933303BxxwQMY8Zs2axejRo4sun6osYXra8xMX/ZPlV6xYkfLcNDY2biL/9ttvc/fddzN69Gh23nnnjPkXS76xsTFleUrx2snX1iCe9jZixIisz00c9U+W76hdK0Vbg86VJ27nK921k9y2Qfzbt1TtmuPkSsFCLCSdKem/k9Iul3R80vqs8HdgmDYlHEvVSSIujWmu8rkSN/2j8tkQ95uHsylxtbdsiav+hWofSp241f+sWbMyykSJW3vl7ZtTKAodg/wjSbMTP+CkxAZJvwZ2ApaEv1Mkfa+jjCTtIek5SX8osM6xJS6NaaU6x9nIx+1m4DeP9MTNfnJ9GIuz/u4cb07c6j/Xh7G4tVfevjmFpNAxyJOjs4lJujyy7RzgYuCrBNPy3mJmf5V0OHC0pK3NbEZE/mvAJGBUgXUuWeLY+OZCHPV357h8iZv9VJp8pRG3+o/KjxgxIqN83NqrSm/fXnrppX41NTVTCKaq9wEX8qMdeLWtre3Uvfba64PkjYV0kNcDp0qKOrT9gHHh8rFAH+AIoD8wG9gh3LYKWB3NzMzu8NCLjolr45stcdXfnePyJG72U2nylUbc6t/bt9KnpqZmyoABA77Qt2/fj6qqqkozkL/ItLe3a/ny5Xs2NTVNAY5M3l4wBzmclSjdzESPAMcA9wPdgZvC9BXADDNLNVVrWiSdBpyW636lTtwa01x7JuKsf7Y3+zjdDPzmkZ642U+lyVcacat/d47LhmHuHHeOqqoq69u376qmpqZhqbYXxEGWdB8wMJI0FFgEtIbrC81sjKRGYEyYNkDSDeHyinyOG3XKJVWE0cStMa00+QRxuRn4zSM9cbOffOwtTvq4c5yeuNW/P/yXFVXuHHeesA5ThqgUxEE2syMlHQKsM7Mnww/0jjazJUmiC4FpSWnjCEItFhZCt3Iibo1ppclHicPNIB/5SiJu9pOvvcVFH3eO0xO3+veH//Jm4MCBDe+//36X+XSDBg1qW7JkyfxMcldffXWfadOm9UmsX3rppUtHjRq1Jt0+b775ZreTTz55p+eee+6tdHL77LPPHvfcc8+i4cOHD128ePGr2WvfNRQysPtLwBfC5dcJYpKTGQXcBVwT+fUGPi6gXmVB3BpT7wlLT9xuHgn5SiFu9tMZe4uDPvnIVxJxrP9KevivRN5///2amTNnYmYZfzNnzqRPnz6kk1+6dGlaZ7u1tZVjjjlmyIsvvthz6NChzYnfnXfe2evMM88cBDB9+vRtGxoahjY0NAwdPHjwsPPOO29gujwBhgwZkjLcIcH48eMH7LjjjsMS+SZ+c+bMqc+txjKT9dOGpEHAfmaWy111nKRTw+UHJQE8bWbnJulQl7TfnsDLyZmZ2SxgVg7HL0vi1ph6T1h64nbziMpPnTo1qzKUOnGyn0LbW9z0T8hXCnGt/0pq3yqVLWkPtbW13Hnnne9ef/31fWbOnLnNmjVrqnr16tV29NFHf3TcccetAjjuuONWJZZvvvnmXgsXLqy78cYbe8+cOXObjvLt3r17e6Zjn3vuuU3nnXdeXqG4udChgyxpv8hqK8GoErtJ+jqwHYCZPdjR/maW6BHuEDObAkzJReFKJ26NaaX1hOVKHG8eldgTExf7qVTn+O67s/tgtxyIa/1XUvvmpKar7eGZZ57pcdttt/V75JFHFgwaNKj1lVdeqTvooIP2OOqoo/7eo0cPmzRpUu+JEycOHDhw4HqAE044YcW+++67tra21qJhGQleeeWV7qtWrapZsGBBt6OOOmqXt99+O7nzFIAbb7xxwPTp03tH0x599NGFvXv33tDpQkXocKppSb8HDgEeBfYDDgNGA98B5gBmZud1pTJdSX19fVNLS0vJzSEPUFdXt6y5uXmTeeRLtTypygKlW55u3bp9uG7duk0uzNra2qa2traSKwtATU3NstbW1nK3tWUtLS39iqFTZ6irq/ugubl5k/NQqucGyqtdg9Tl6d69+8r169f3KpZO+dLRtVOqbVuqdq3cmD9//jsNDQ0be1El7ZVumvNcneNw+vGXMsndcccdn3viiSe2Xr9+fVXPnj03HH300R8feuihnwBMmjSp9zvvvNP9mmuuea+lpUVLly6tfeCBB7YGuOuuu3onxyBfdtll/aZPn95n3Lhx75100kkfp4pBHj9+/IA+ffq0dWUP8vz58/s0NDTslJzeYQ+ymZ0g6fnEf9K2czvaLy6kuthLGS9PfCm3hriczg1AspNZypThuSmr8iQ/PJc65da2VSqF+GbgsMMO+/wHH3xQm5w+b968nuPHj2fw4MHrjjnmmI9++9vf9t1777237tGjR/uAAQNaDzjggNUHHXTQJ3fdddcm18rKlSurp02b1mfmzJlvjRgxYvevf/3ra1Mdd/vtt2+97bbb+k6dOnWTHuhx48a9P2bMmFUZFc+BTDHIiceRbYE/AH/qyoM7juM4juM4haFQ3ww89NBD/1ywYEG3mpoa23nnnVsbGhqG/vnPf160yy67tEblxowZs+qSSy7pf+mlly5LpL377ru1u+++e3NifcOGDRx++OG7XHLJJUsHDRrUdtVVV/3r4osv3j7Vcc8666yVZ5111sqMCnYB2X6ktwa4ANi/gLo4juM4juM4OTJo0KA2SR36dLl+AzBo0KC2TDK///3vt9tqq63ax44du/ycc85Z1rdv35QxwFOmTOkXdZCHDBnSevvtt/8rsV5dXc306dPf3nXXXVsBRo0atWbUqFFr9tlnnz2i+Rx88MG7pOq1TnDIIYesuvrqq9/PpnzZkO4jveOAPpJ+ANQDTV11UMdxHMdxHKdryGbM4kJwww03DLjjjjv6AFx77bUDAL72ta998tvf/najA7x8+fLaYcOGfSG6X8+ePTfMmTNnYwxywjlOx2OPPbao6zTPTLqP9K6OrH4K/C/BR3r9zOynW0A3x3Ecx3EcJ4nkj/Sc/MnnI70LouuS9gjT3Tl2HMdxHMdxypaMMciSvgcsM7NngSsKr5LjOI7jOI7jFI90Mcg/DBcPBFZK2iVcf9PM5kg638yuLbiGjuM4juM4jrMFSdeDvGP4vyBpPfEl4jFAbB3kchuAvlTLU24ThZTTuYHymoyiEiY7KNVJTyD1JDulamuQ2t7KydagdO0tVbvmOLmSLgY5UziFuliXLqWlpaV/ulll4oykzS7saHniNj1pOvlUZYHSPT+Zzk0q4ny+RowYsdnNL7k8cdY/Kt+RrbW1tfXv0aNH7KbDzSTf2NiYytb6Zbpu4nq+VqxYsdkMc6XaDkBqe2tra+t/4oknxsJ+cpFPZWtQuvaWql1znFypSrdR0iuS/h7+Jw8hkvaqkXSmpP9OSrtc0vFJ67PC34Fh2hRJB+RWjMohjo1RV87tXm7Erf4rTT5BKTkrnSFu9R+VrxTiZD+Vam+VyIABAxok7dVVvwEDBjTkqsM+++yzx+LFi2t23nnnLxaijKnYeeedv7ho0aLa5DGTu4K0H+mZ2b+F001/XdILeeT/I0lHRNZ3AC4EkPRrYGtgSbjtFEmfS5WJpJ7AVcAwoAfwmJmNz0OfkiaujZE7x6mJW/1XmnyUODgfleqsVFr7EBf7qWR7y3VSjHJg2bJlNY2NjR1uz9UeGhsb0/qHkyZN6n3FFVcMGjBgwPrevXu3PfHEEwvTyTc3N2vYsGF7Llq06B/J24499tjBc+bM2XqrrbbaZKKRhx56aOHAgQM3mbCkqamp+tBDD91t/vz5b3R0rFzyS0c2M+lZ9F/SNGAQkI23PtnMfpFYkXR5ZNs5wMXAVwlm6rvFzP4q6XDgaElbm9mMUHZb4E4ze0ZSFfC6pElmVlGTl8S1Maqkm1+2xK3+K00+V+LmrLz99ttZ6Z0gbvVfye1DXOynUp3jSrO3bChU+3PcccetuO66697LRvbBBx/cevXq1dUffvhhVa9evdqTt1977bWLR40atSZTPuvWratKdnxTkW1+6UjrIEv6GTBI0lggEcA/Fuhwqr8I64FTJY2KpPUDxoXLxwJ9gCOA/sBsgh5mgFXA6sROZvYekDgJPcO8P85Ch7IiLo2LN0bpiVv953O+4qRPJTrHuYQkxK3+vX1ITxztLRfiZj9ub+kpZPszffr0Po8//vg2hx9++Md///vfe7z55pv1qeTef//9mosvvnjQOeec0/SDH/xg5/vvv/+fdXV1m4Tpjh07dscJEyZsdHzr6uran3322QXJec2bN6++ubm5eubMmT3OOuusIStWrEjpj2abXzoy9SA3EfTyAlwEG53VjJjZrcCtaUQeIRgJ436gO3BTmL4CmGFms5N3kFQN3AFcYGYtKbafBpyWjX6lSBwaF2+M0hO3+s/3fMVFn0p1jkePHs3UqVMzyset/r19SE9c7S1b4mY/bm/pKXT7k9yDnBwH3NzcrN/85je9brvttn5XXHHFklGjRq3ZfvvtW4cPH777RRdd9N7IkSM39vBeddVV/8qmx3fGjBnbfvrpp1XV1dW8+uqrr3cU75xtfunIFIOcuYZSIOk+YGAkaSiwCEjMtb3QzMZIagTGhGkDJN0QLm82faKkWgLn+I9m9nAH+m50yiWV5qfRnSBujdGsWbMyypQTcaz/fG8ecdCnkp3jbOTjVv/urKQnbvbjD2PlzZYOy1m5cmV1qvS1a9dWPf74428NGDBgA8App5zy0Te/+c1P/vjHP253xBFHrKmurmbIkCHrL7744kGXXHLJJvtOnjx58fDhw9cm1hcsWNBt7ty5W82YMWPhyJEjd33qqafeTHXMbPPLRDYxyDljZkdKOgRYZ2ZPSpoNHG1mS5JEFwLTktLGEYRabAz4ltQNuBO4x8z+UAidS524NUYJ+UohrvVfqJtH3PTP9WEszs6KO8flR9zsx+2tvNlSzvG0adP6PPzww9tWVVXxgx/8YGXy9vr6ehs/fvzy5PRdd9219ec///kHifWJEyc2TZw4Me03ZatXr64aOXLkLpMnT3539913X3/66acvu/LKK1MOT5hNftmQlYMsScDeQLWZPZ9l3l8CPgGeBF4niBtOZhQwHliclJ4cX3wqcADQW9KPwrTzzeylLHUpa+LWGEXlK+Vr4rjWf6U4x7k8jMXN+XBnpbyJm/24vZUn/fv3b0seeSKbNwMdyffv3z/taA9nn332yrPPPnsTp/j+++/fLrp+8MEH7/LBBx90+M3aIYccsurRRx/dNt1xTjzxxBXnnXfeim222ab9vvvuW7jrrru2Jo4PcO+9925yzIaGhqHZ5JdOJkG2PcjXA3OBlnAUiVOBpWb2QIb9xkk6NVx+MPCzedrMzk3SoS5pvz2BlxMrZnYzcHOWulYUcWuMKrXxikt9VqpznO3DWNycj3x6buJa/5V0vWdL3OzH7a18aWpqSp6roug89thjizLJXH311e9nm1/COU5HuuHfciXTKBYzgB8AuwBPEIxeMRHYDvi+pKrIUGybYGbXANeky9/MpgBT8tDbIX6NUSU3XnGoz0KHJcRR/1zk4+Z85PtaMy71WcnXezbEzX7c3pxCM3fu3DcB3n777c3GOi4UiWMljt2VKN00kpKWAu8CdxN8ZFcL/MzM9pG0C3C5mf1nVyvVFdTX1ze1tLSU5FzsdXV1y5qbmwdE00q1PKnKAuVVnlItC0BdXd0Hzc3Nm+hequXpyNZqa2ub2traSq48NTU1y1pbW5NtbVlLS0tJTqPbrVu3D9etW9c7mlaqtgap7a2cbA1K195StWvlxvz5899paGjIKlTASc/8+fP7NDQ07JScninE4p/AfILe48TOieHV3uWzsZFjR6obZSnj5Ykv5VQWKL/ypLrxlyrldtN3W4s35WZvjpMLVVlsnwSM5rMZ9bqF/32Bjwqkl+M4juM4juMUhUw9yDKzt8Jwirlh2muS/n+CsY0zfaTnOI7jOI7jOCVFph7k+8L/VUAPgl7kc4FBwAIz+20BdXMcx3Ecx3EqkKamppSTj2wp0jrIZvaLcHEC0BymrTKzc83sl4VWznEcx3Ecx0nPDjvs0CBpr6767bDDDg356LFo0aLa5Cmnr7vuuj4XXHDB9on1N998s9t+++23e0d5DB06dE+A/fffP+2YxgA777zzF1MdsyvIahxkM1sKLO3qgzuO4ziO4zidY+nSpTXpRiXLFUlp/cP7779/6xNOOOHzgwcPXgewevXq6o6Gd2tqaqodOHBgyjGMJ02a1PvKK68c1L9///X19fXts2fPfivdcZuamqoPPfTQ3dKNd3zssccOnjNnztZbbbXVhmj6Qw89tHDgwIFpJ0CJ0mEFSFrAZx/mJRgOPBsut5tZl3vsjuM4juM4Trw58MADV/3pT396B4Ke3I7kHnjggc/16tWr7dxzz13R0NAwtLm5uarOKfAwAAAUOUlEQVRPnz4bHdWqqiqrqamxbbbZZgPA4sWLu++11157tLa2KjmvdevWVSU7vqm49tprF48aNWpNXgUL6dBBNrPdEsuSXjCzvcPlj81sb0nZTjldFEp1/EYor7F2OxqPsoTLUzbnBsqrPKnG2YXSLU85nRvocNzgZW1tbSXZTldXV3/Y1ta2ib2V4TjIJWlvHY2J7mx5fvnLX/bt379/a+/evdvOOOOMQfPmzXtj4cKF3U4++eSdEjLHH3/8iquuuup9gPb2dgYPHrzupZdeenO33XbbzOmeN29efXNzc/XMmTN7nHXWWUNWrFiRcirrsWPH7jhhwoSNjnRdXV37s88+uyAX3TPNpLermS0E3okkd10ffgFpaWnpN3PmzFjM6JOrvKTNGqSWlpb+Hb0+iZv+SdP/prz5pStPnMn13MSdfMoTV3tbsWJFr1Tbk8sTV/2T5Ts6NzNnziwJ/ZNJVZ62trZ+jY2NafOP2wxzCfm1a9duZm9tbW39k8sTV/2j8o2NjSmd4FJt21LZmrPlmTBhQr977rmn10MPPbSwV69eG04//fQdTjvttB3Hjh27LCp311139X766ae3rq6utiOPPPLjdHnOmDFj208//bSqurqaV1999fWOeq6vuuqqfxWsBzlkXthTfEpnDlIsSunmEZXPhTjq79OBli9xs5+kh7FY6VOo6yVO+hT6eo+zczl16tRY6VPI6aYdJx++/e1vr77ggguW19XVGcDvfve7f7W1tbFo0aLEfBqceeaZK3/yk5+srKn5zB1tbm4WwLHHHrvJTIELFizoNnfu3K1mzJixcOTIkbs+9dRTKaeXHjJkyPqLL7540CWXXLJJ+uTJkxcPHz58bbb6Z3KQXyNwjm+SdKyZNUe2pX2slHQm8DkzuzySdjnwhplNi6wPDzdfZmZPSJoCTDOzWdkWIl/idvNIyGdLXPV357g8iZv9VJp8grjoU8nOcSXIO046evbsueGVV17p0dDQMBQg8bFelL333rsFYJ999tlj7ty5bwLU1NSw7bbbbjjssMM+BqiqquK5557rccYZZwyO7vvnP/+518CBA9cn1levXl01cuTIXSZPnvzu7rvvvv70009fduWVV6Z8UzBx4sSmiRMnNnW2jJkc5CozWyrpMmAccDHQN/yAL9MYygA/knREZH0H4EIASb8GtgaWhNtOkfS5VJmE6bcCOwIC7jKz67I4fofE7eZRiT1hTukQN/upNPkocdCn0G/G4uYsVpq8U3oMGjSoLdPIE7nml277iBEj1r711luvZZPX0qVLu0XXBwwYsOGiiy76ILE+fPjwtalGpRg8ePCwxPI222zTft999y3cddddWwHOPvvslQD33nvvdtF9Eg57R5x44okrzjvvvBXpZBJkqswNAGb2N0lnS6oys1yursmRsZQTPcYJziFwuL8KrAFuMbO/SjocOFrS1mY2I5TtDjSa2WuhAbwu6Q4zy6qQycTx5lFJ8k5pETf78bCE4upT6DdjcXMWK03eKU2WLFkyv9g6pGPYsGFfiK737Nlzw5w5c9IO6ZZMwjlOR7rh33IlrYNsZl+PrE40s/Yc8l4PnCppVCStH0FPNMCxQB/gCKA/MJughxmCmftWR/RYBiSCuvsCbcCnyQeUdBpwWjql4nrzqBR5p7SIm/14WEJx9Sn0m7G4OYv5OJdx0sedY2dLs8suu7QmwikSLF269JV88lq8ePGrmWQSYy8nH7MryLo73sxy8vTN7FaCsIiOeAQ4BrifoIf4pjB9BTDDzGYn7yDpFwQO8M+S4qE3O6akzWKk43zzqAR5p7SIm/1UWlhCrsRN/1zl4+Ys5utcxkUfd44dp3N0WbxKFEn3AQMjSUOBRUCie3yhmY2R1AiMCdMGSLohXE4ZOmFmF0qaADwsab6Zzc1Wp7jdDPw1sRNn4mY/lehc5kIc9a9E5xiIhT7uHDtO58naQZb0ezM7IRtZMztS0iHAOjN7UtJs4GgzW5IkuhCYlpQ2jiDUYmHk2HsAH5rZcmAtQQjGdmRJ3G4G/prYiTtxsp9KdI5zidmNq/6V6BxnQ9z0f/vtt7PS24kd7e3t7aqqqiq9gapjRHt7u4CU4cPpppreL7oKfDUpDTN7Ls1xvwR8AjwJvE4Qk5zMKGA8sDgpPXmg6HXAryT1BXoAzwCPpjn2RuJ2M/DXxE4pEBf7qVTnONuY3Tjrn+35ipuzWGnO8d13352V7k7seHX58uV79u3bd5U7yfnR3t6u5cuXbwukjHVO14P8X0nrs8M0I3CYDUjnIAOMk3RquPygJICnzezcJB3qkvbbE3g5sWJm7/BZKEbWxO1mUIk3e6c0iYv9VNr1UknyCeLmLFaaczx6dHaTnjjxoq2t7dSmpqYpTU1Nw8hu2F1nc9qBV9va2k5NtVG5TiMp6XzgVjPr1BR+haa+vn5ZS0tLymmO406qeeTr6+ubWlpaSm76zLq6ug+am5s307uEy1M25wbKqzzdunX7cN26db2T00u1POV0biB1eWpra5e1tbWVZDtdXV39YVtb2yb2Vltb29TW1lZy56empmZZa2vrgOT0UrW3VLbmOLmS1kEOxxy+EvirmT0aOscHmdlhW0pBx3Ecx3Ecx9mSZHKQXwB6Ai8QTAm9GtjfzD7ZMuo5juM4juM4zpYlm7iVj8zsROB4YCtgl8Kq5DiO4ziO4zjFI5OD/CrBB3mY2fPA94DbunK+b8dxHMdxHMeJE2kdZDM7GXgwsv4K8DvgoALr5TiO4ziO4zhFIedRLAAk3WRmZxZAH8dxHMdxHMcpKmlDJcJRKz4GmoB3gQVmtg74yhbQzXEcx3Ecx3G2OJlikM8E6oF9gAuA5yRtX3CtHMdxHMdxHKdIZPrYboWZ3QQg6RjgCeAQ4HOFVqyzlOoA55B6kPPu3buvXL9+fa9i6ZQvHQ3YXqrnp4PJDkpycgBIPUFAqU6yU4a2ttkkO2VoayV5bqC8JnIpw2vHJwpxOk2mcZDnmtk+4fIpBFNCCzjDzPbcMirmhyTLJ746DkjCzJSUllV54jadbKqyhOkleX46OjeNjY1AvKeTTSXf2NiYsjwzZ86Mhf3kIp+PrcVJ/2T5ESNGpLU1KL795CLfka316dMnlvWfj72VU7sWpm9SnjjVfzr5jsrjOLmQy/zd64CW8NeeSVjSmZL+OyntcknHJ63PCn8HhmlTJB2Qg15OSBwbr0qilJyVTMTFfnKRz5W46Z+rfNzsJ197i0t9FtreSp241b+fL6fQZAqxiD6BdSeIR85mvwQ/knREZH0H4EIASb8GtgaWhNtOkZQ2dEOSgEeBpWZ2UpY6VARxa4wS8pVC3JyPzjjH2RBHe8uFOOpfic4xlObDWKU9/Mex/t05dgpNph7kGxILZvY/ZjbZzCYDH2WZ/2Qz2zfxA26PbDuHYGSMvgShG7eb2Z/DbUdL+m6K/H5CMHmJEyFujVFUvlKIk/NRic5xLg9jcdW/Ep3jbIhb/Vfaw39c69+dY6fQZJooZHoHm+Zlkfd64FRJsxM/4FigNdx+LNAHOILAWb49su8qYHU0M0k7Ad8BfpXFsSuGuDVGldp4xcX5qFRnJduHsTjrn+31Eif7cXsrf+Ja/5V0f3GKQ15TRpvZGVnI3ArcmkbkEeAY4H6C8I2bwvQVwAwzm50QDEMrJgFnkSb+WdJpwGmZdCsX4tYYVXLjFQfnIx/5XIib/VSafII42U+lOseV1L7FpT4rtf6d4pGXg5wJSfcBAyNJQ4FFfNZ7vNDMxkhqBMaEaQMkJUI6ViRl+WPgETNbFPYkpyTqlEsqvU+JcyBujZE3XumJm7OSkM+WuNlPpclHiZP9+MNY+ROH+qzk+neKR0EcZDM7UtIhwDozezIMrzjazJYkiS4EpiWljSP4mG9hJG1voKek/QnGYN5D0sVmdlkh9I87cWuMvPFKTxydlYT81KlTM8rHzX7ysbc46VPo6yWu9pYtcat/b9/SE7f6n1VhH1A6haMgDnLIl4BPgCeB1wlikpMZBYwHFielfxxdMbNTEsvhEHAnuXMcn8bIbx4dE1dnJVv5uNlPvvYWF30q1TmupIexSiJu9Z+Qd5yuINMoFp1lnKQXgX8DHpT0oqTrk2RqCEaxiP46nITEzGZV6hBvcW2M/OaRmjg7K5XkHENlvCaOm/1Usr1VAnGr/6i843QFBetBNrNrgGsyyEwBphRKh3Ijro2R3zw2J27OhzsrxdWn0K+J42Y/bm/lTdzq38+XUwjSTjVdypTqHPKQeh757t27r1y/fn2vYumUL6nKAqV7flKVp7a2tqmtra3kygJQU1OzrLW1dZPy1NfXL2tpaelXLJ3ypQxt7YPm5uZN9C5DWyvJcwOp7a1Uy1OG107K8jhOLpStg7ylkHRaOHpGWVBO5SmnsoCXJ+6UU3nKqSzg5Ykz5VQWp7wodAxyJVBu4y6XU3nKqSzg5Yk75VSecioLeHniTDmVxSkj3EF2HMdxHMdxnAjuIDuO4ziO4zhOBHeQO0+5xU6VU3nKqSzg5Yk75VSecioLeHniTDmVxSkj/CM9x3Ecx3Ecx4ngPciO4ziO4ziOE8EdZMdxHMdxHMeJ4A5ynkjaQ9Jzkv5QbF06i6SekiZLelLSC5KuLLZOnUHS5yTdJel5SbMlnVdsnTqLAh6TdHuxdekMkqokrZQ0K/w9XmydOoukIZIeD9uDZyTVFVunfJH008i5mSXpPUnfKrZe+SCpXtL/Sno2bNcuK7ZOnUXShNDOXpZ0fLH1yYdU905JV4Rpz0s6oIjqOc5GPAY5TyT9EFgPjDKzMcXWpzNIGgh83syekVQFvA58y8yaiqxaXkjqD/Q2s9ck1RCU5+tmtqLIquWNpDOAXYHtzOykIquTN5K2A6aY2feLrUtXIKkaeBY42cxel1RtZhuKrVdXIKkb8Bywv5k1F1ufXJF0OjDMzM4Iz9NzwI/N7G9FVi0vJB0CnAmMBOqB54GDzeyDoiqWI8n3TkkHAueb2XfCe9ETBOetraiKOhWP9yDniZndAZSkA5mMmb1nZs+Eqz0JGq+Pi6hSpzCzZWb2WrjaF2gDPi2iSp1C0k7Ad4BfFVeTLmE7YG9JT0t6QtL3iq1QJzkMeBO4QtKzwOlF1qcr+SHwp1J0jkOagM+FznEPoBr4qLgqdYovA09YwFrgReAbRdYpZ1LcO/8DuDvc9h7wLrBHEVRznE1wB9nZSHgjuQO4wMxaiq1PZ5H0C+AfwHWlepOXJGAScBbQXmR1uoJ3zGywme1P4IBNkPSlYivVCYYCXwBOBA4BTi7x8gAb24IfA5OLrUu+mNlfgBXAP4EFwM1m9k5RleocrwMHSaoO35IdSOD0lzp9CM5TghUEHRuOU1TcQXYAkFQLTAP+aGYPF1ufrsDMLgR2BH4oaZ9i65MnPwYeMbNFxVakKzCz9sjyEuBhYFjxNOo0G4D7zGyNmX0K/BVoKLJOXcEYArtbXWxF8kXSjwABnwd2Ar4bhimUKvcDLwNPAlcDrxI4/qXOR8C2kfVtKe2efqdMcAfZScQa/oHgRl8OHx3uISnRA7EWWEXwar8U2Rv4ZvhByy3AtyRdXGSd8kbSrpJ6hsvbEPSCzS6uVp3iGeCAsFevhuCV9ytF1qlThG8tfgrcUGxdOskewGIz2xC+EWuitF/dC7jIzIYTnJutzWx+kXXqCp4BjgSQ1IfgHL1ZVI0cB6gptgJOLDgVOADoHfa6QPDRxEvFU6lTrAN+FTrJPQga4EeLq1J+mNkpieXw6+6TzKyUv8bvC/xP4INRDUwws38WV6X8MbMXJD1GEA+6DviDmb1cZLU6y1HA82a2vNiKdJKrgd9JOorgXvcO8LuiatQ5+gN/Cq+dFcAPiqtOl/EgcIik5wg67X5aDiF+Tunjo1g4juM4juM4TgQPsXAcx3Ecx3GcCO4gO47jOI7jOE4Ed5Adx3Ecx3EcJ4I7yI7jOI7jOI4TwR1kx3Ecx3Ecx4ngDrLjOBWDpDfC/4UZ5IZLuj2DzAGSpnShbmMkNXZVfo7jOE7++DjIjuOUJZJOAq4E3gPWmNmIDuSuBv4DaAbOSB7HOHRaj+ez6XBvMbPb0xz3euDrKTZNMbMpkgYCvyWYYrcVGGtmz6TI52GCsW/XRZLbwokiHMdxnALiDrLjOOXMrWbW2NFGSaOAQWb2FUl7AncAX00h2mhm07I5oJmdK+knwP+Z2VJJ3wTqzCwxWc2VwDQzmy5pR+AxYGgH2Y02s7S93Y7jOE7X4w6y4zhlj6TfAHsBQ5I2fRO4E8DMXgunjN4mRRaNks4Ml88CzgSGAX/r4JDfJJhCeynwZWADn83mGA1tM4IZBTviD5Kis4r908x+mEbecRzH6QLcQXYcp+wxsx/BZzHI0U1J60qRBpv3IJ8YTv19fAeHbAG6hcv1wAeRbeMJpkA+F2gnmOq9I8Z4D7LjOM6Wxx1kx3HKHkk1wOdSbHoa+E9gRhhi0WZmayRlyq9fmm33Af2AX0naDfgHUC3p82Z2kZktAQ7OQu23gGkpdDnKzN7PYn/HcRwnT9xBdhynXGkl6OkdSdCje1MKmf8D9pf0N4KP4TrqzW2UdAZB77IBLwJ/SSVoZkcCSNoBuNfMvpEsE2673sxGR5IXAp9E8jk7ffEcx3GcQuEOsuM4ZYmZTQemR9MkXZQkY8D5GfJpBBqT08MQi+S0vwDbh6vdgN0lzY6IPGBmEwja3r5Jx3kxzGN7OnC+I0wwswcyyDiO4zh54g6y4zhOF2FmR+UgvpekF5PSnjazc4F9u1Atx3EcJ0cUdKA4juM4juM4jgM+k57jOI7jOI7jbII7yI7jOI7jOI4TwR1kx3Ecx3Ecx4ngDrLjOI7jOI7jRHAH2XEcx3Ecx3EiuIPsOI7jOI7jOBHcQXYcx3Ecx3GcCP8PcSg/AIGDE0YAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 720x144 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "mglearn.plots.plot_shuffle_split()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "교차 검증 점수:\n",
      "[0.973 0.84  0.933 1.    0.893 0.92  0.8   0.84  0.907 0.827]\n"
     ]
    }
   ],
   "source": [
    "from sklearn.model_selection import ShuffleSplit\n",
    "shuffle_split = ShuffleSplit(test_size=.5, train_size=.5, n_splits=10)\n",
    "scores = cross_val_score(logreg, iris.data, iris.target, cv=shuffle_split)\n",
    "print(\"교차 검증 점수:\\n{}\".format(scores))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### 그룹별 교차 검증"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "교차 검증 점수:\n",
      "[0.75  0.8   0.667]\n"
     ]
    }
   ],
   "source": [
    "from sklearn.model_selection import GroupKFold\n",
    "# 인위적 데이터셋 생성\n",
    "X, y = make_blobs(n_samples=12, random_state=0)\n",
    "# 처음 세 개의 샘플은 같은 그룹에 속하고\n",
    "# 다음은 네 개의 샘플이 같습니다.\n",
    "groups = [0, 0, 0, 1, 1, 1, 1, 2, 2, 3, 3, 3]\n",
    "scores = cross_val_score(logreg, X, y, groups, cv=GroupKFold(n_splits=3))\n",
    "print(\"교차 검증 점수:\\n{}\".format(scores))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "hide_input": false,
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "application/pdf": "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\n",
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x144 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "mglearn.plots.plot_group_kfold()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 그리드 서치(Grid Search)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### 간단한 그리드 서치"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "훈련 세트의 크기: 112   테스트 세트의 크기: 38\n",
      "최고 점수: 0.97\n",
      "최적 파라미터: {'C': 100, 'gamma': 0.001}\n"
     ]
    }
   ],
   "source": [
    "# 간단한 그리드 서치 구현\n",
    "from sklearn.svm import SVC\n",
    "X_train, X_test, y_train, y_test = train_test_split(iris.data, iris.target,\n",
    "                                                    random_state=0)\n",
    "print(\"훈련 세트의 크기: {}   테스트 세트의 크기: {}\".format(\n",
    "      X_train.shape[0], X_test.shape[0]))\n",
    "\n",
    "best_score = 0\n",
    "\n",
    "for gamma in [0.001, 0.01, 0.1, 1, 10, 100]:\n",
    "    for C in [0.001, 0.01, 0.1, 1, 10, 100]:\n",
    "        # 매개변수의 각 조합에 대해 SVC를 훈련시킵니다\n",
    "        svm = SVC(gamma=gamma, C=C)\n",
    "        svm.fit(X_train, y_train)\n",
    "        # 테스트 세트로 SVC를 평가합니다\n",
    "        score = svm.score(X_test, y_test)\n",
    "        # 점수가 더 높으면 매개변수와 함께 기록합니다\n",
    "        if score > best_score:\n",
    "            best_score = score\n",
    "            best_parameters = {'C': C, 'gamma': gamma}\n",
    "            \n",
    "print(\"최고 점수: {:.2f}\".format(best_score))\n",
    "print(\"최적 파라미터: {}\".format(best_parameters))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### 매개변수 과대적합과 검증 세트"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "hide_input": false
   },
   "outputs": [
    {
     "data": {
      "application/pdf": "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\n",
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x72 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "mglearn.plots.plot_threefold_split()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "훈련 세트의 크기: 84   검증 세트의 크기: 28   테스트 세트의 크기: 38\n",
      "\n",
      "검증 세트에서 최고 점수: 0.96\n",
      "최적 파라미터:  {'C': 10, 'gamma': 0.001}\n",
      "최적 파라미터에서 테스트 세트 점수: 0.92\n"
     ]
    }
   ],
   "source": [
    "from sklearn.svm import SVC\n",
    "# 데이터를 훈련+검증 세트 그리고 테스트 세트로 분할\n",
    "X_trainval, X_test, y_trainval, y_test = train_test_split(\n",
    "    iris.data, iris.target, random_state=0)\n",
    "# 훈련+검증 세트를 훈련 세트와 검증 세트로 분할\n",
    "X_train, X_valid, y_train, y_valid = train_test_split(\n",
    "    X_trainval, y_trainval, random_state=1)\n",
    "print(\"훈련 세트의 크기: {}   검증 세트의 크기: {}   테스트 세트의 크기:\"\n",
    "      \" {}\\n\".format(X_train.shape[0], X_valid.shape[0], X_test.shape[0]))\n",
    "\n",
    "best_score = 0\n",
    "\n",
    "for gamma in [0.001, 0.01, 0.1, 1, 10, 100]:\n",
    "    for C in [0.001, 0.01, 0.1, 1, 10, 100]:\n",
    "        # 매개변수의 각 조합에 대해 SVC를 훈련시킵니다\n",
    "        svm = SVC(gamma=gamma, C=C)\n",
    "        svm.fit(X_train, y_train)\n",
    "        # 검증 세트로 SVC를 평가합니다\n",
    "        score = svm.score(X_valid, y_valid)\n",
    "        # 점수가 더 높으면 매개변수와 함께 기록합니다\n",
    "        if score > best_score:\n",
    "            best_score = score\n",
    "            best_parameters = {'C': C, 'gamma': gamma}\n",
    "\n",
    "# 훈련 세트와 검증 세트를 합쳐 모델을 다시 만든 후\n",
    "# 테스트 세트를 사용해 평가합니다\n",
    "svm = SVC(**best_parameters)\n",
    "svm.fit(X_trainval, y_trainval)\n",
    "test_score = svm.score(X_test, y_test)\n",
    "print(\"검증 세트에서 최고 점수: {:.2f}\".format(best_score))\n",
    "print(\"최적 파라미터: \", best_parameters)\n",
    "print(\"최적 파라미터에서 테스트 세트 점수: {:.2f}\".format(test_score))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### 교차 검증을 사용한 그리드 서치"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "SVC(C=100, cache_size=200, class_weight=None, coef0=0.0,\n",
       "  decision_function_shape='ovr', degree=3, gamma=0.01, kernel='rbf',\n",
       "  max_iter=-1, probability=False, random_state=None, shrinking=True,\n",
       "  tol=0.001, verbose=False)"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "for gamma in [0.001, 0.01, 0.1, 1, 10, 100]:\n",
    "    for C in [0.001, 0.01, 0.1, 1, 10, 100]:\n",
    "        # 매개변수의 각 조합에 대해 SVC를 훈련시킵니다\n",
    "        svm = SVC(gamma=gamma, C=C)\n",
    "        # 교차 검증을 적용합니다\n",
    "        scores = cross_val_score(svm, X_trainval, y_trainval, cv=5)\n",
    "        # 교차 검증 정확도의 평균을 계산합니다\n",
    "        score = np.mean(scores)\n",
    "        # 점수가 더 높으면 매개변수와 함께 기록합니다\n",
    "        if score > best_score:\n",
    "            best_score = score\n",
    "            best_parameters = {'C': C, 'gamma': gamma}\n",
    "# 훈련 세트와 검증 세트를 합쳐 모델을 다시 만듭니다\n",
    "svm = SVC(**best_parameters)\n",
    "svm.fit(X_trainval, y_trainval)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "사이킷런 0.22 버전부터는 `GridSearchCV`와 `RandomizedSearchCV`의 `iid` 매개변수 기본값이 `True`에서 `False`로 바뀝니다. 0.24 버전에서는 이 매개변수가 아예 삭제될 예정입니다. `iid` 매개변수가 `True`이면 독립 동일 분포라고 가정하고 테스트 세트의 샘플 수로 폴드의 점수를 가중 평균합니다. `False`로 지정하면 단순한 폴드 점수의 평균입니다. `False`일 때 기본 교차 검증과 동작 방식이 같습니다. 사이킷런 0.20 버전에서 `iid` 매개변수가 기본값일 때 가중 평균과 단순 평균의 차이가 10^-4 이상이면 경고 메세지가 발생합니다. 경고 메세지를 피하기 위해 `plot_cross_val_selection` 함수 안에서 `GridSearchCV`를 사용할 때 `iid` 매개변수를 `True`로 설정합니다."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "hide_input": false
   },
   "outputs": [
    {
     "data": {
      "application/pdf": "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\n",
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x216 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "mglearn.plots.plot_cross_val_selection()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "hide_input": false,
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "application/pdf": "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\n",
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 700x210 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "mglearn.plots.plot_grid_search_overview()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "매개변수 그리드:\n",
      "{'C': [0.001, 0.01, 0.1, 1, 10, 100], 'gamma': [0.001, 0.01, 0.1, 1, 10, 100]}\n"
     ]
    }
   ],
   "source": [
    "param_grid = {'C': [0.001, 0.01, 0.1, 1, 10, 100],\n",
    "              'gamma': [0.001, 0.01, 0.1, 1, 10, 100]}\n",
    "print(\"매개변수 그리드:\\n{}\".format(param_grid))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import GridSearchCV\n",
    "from sklearn.svm import SVC\n",
    "grid_search = GridSearchCV(SVC(), param_grid, cv=5, return_train_score=True, iid=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train, X_test, y_train, y_test = train_test_split(iris.data, iris.target,\n",
    "                                                    random_state=0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "GridSearchCV(cv=5, error_score='raise-deprecating',\n",
       "       estimator=SVC(C=1.0, cache_size=200, class_weight=None, coef0=0.0,\n",
       "  decision_function_shape='ovr', degree=3, gamma='auto_deprecated',\n",
       "  kernel='rbf', max_iter=-1, probability=False, random_state=None,\n",
       "  shrinking=True, tol=0.001, verbose=False),\n",
       "       fit_params=None, iid=True, n_jobs=None,\n",
       "       param_grid={'C': [0.001, 0.01, 0.1, 1, 10, 100], 'gamma': [0.001, 0.01, 0.1, 1, 10, 100]},\n",
       "       pre_dispatch='2*n_jobs', refit=True, return_train_score=True,\n",
       "       scoring=None, verbose=0)"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "grid_search.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "테스트 세트 점수: 0.97\n"
     ]
    }
   ],
   "source": [
    "print(\"테스트 세트 점수: {:.2f}\".format(grid_search.score(X_test, y_test)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "최적 매개변수: {'C': 100, 'gamma': 0.01}\n",
      "최고 교차 검증 점수: 0.97\n"
     ]
    }
   ],
   "source": [
    "print(\"최적 매개변수: {}\".format(grid_search.best_params_))\n",
    "print(\"최고 교차 검증 점수: {:.2f}\".format(grid_search.best_score_))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "최고 성능 모델:\n",
      "SVC(C=100, cache_size=200, class_weight=None, coef0=0.0,\n",
      "  decision_function_shape='ovr', degree=3, gamma=0.01, kernel='rbf',\n",
      "  max_iter=-1, probability=False, random_state=None, shrinking=True,\n",
      "  tol=0.001, verbose=False)\n"
     ]
    }
   ],
   "source": [
    "print(\"최고 성능 모델:\\n{}\".format(grid_search.best_estimator_))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### 교차 검증 결과 분석"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "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",
       "      <th>4</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>mean_fit_time</th>\n",
       "      <td>0.00054</td>\n",
       "      <td>0.00051</td>\n",
       "      <td>0.00052</td>\n",
       "      <td>0.00052</td>\n",
       "      <td>0.0005</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean_score_time</th>\n",
       "      <td>0.00021</td>\n",
       "      <td>0.00021</td>\n",
       "      <td>0.00021</td>\n",
       "      <td>0.00021</td>\n",
       "      <td>0.00021</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean_test_score</th>\n",
       "      <td>0.37</td>\n",
       "      <td>0.37</td>\n",
       "      <td>0.37</td>\n",
       "      <td>0.37</td>\n",
       "      <td>0.37</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean_train_score</th>\n",
       "      <td>0.37</td>\n",
       "      <td>0.37</td>\n",
       "      <td>0.37</td>\n",
       "      <td>0.37</td>\n",
       "      <td>0.37</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>param_C</th>\n",
       "      <td>0.001</td>\n",
       "      <td>0.001</td>\n",
       "      <td>0.001</td>\n",
       "      <td>0.001</td>\n",
       "      <td>0.001</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>param_gamma</th>\n",
       "      <td>0.001</td>\n",
       "      <td>0.01</td>\n",
       "      <td>0.1</td>\n",
       "      <td>1</td>\n",
       "      <td>10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>params</th>\n",
       "      <td>{'C': 0.001, 'gamma': 0.001}</td>\n",
       "      <td>{'C': 0.001, 'gamma': 0.01}</td>\n",
       "      <td>{'C': 0.001, 'gamma': 0.1}</td>\n",
       "      <td>{'C': 0.001, 'gamma': 1}</td>\n",
       "      <td>{'C': 0.001, 'gamma': 10}</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>rank_test_score</th>\n",
       "      <td>22</td>\n",
       "      <td>22</td>\n",
       "      <td>22</td>\n",
       "      <td>22</td>\n",
       "      <td>22</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>split0_test_score</th>\n",
       "      <td>0.38</td>\n",
       "      <td>0.38</td>\n",
       "      <td>0.38</td>\n",
       "      <td>0.38</td>\n",
       "      <td>0.38</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>split0_train_score</th>\n",
       "      <td>0.36</td>\n",
       "      <td>0.36</td>\n",
       "      <td>0.36</td>\n",
       "      <td>0.36</td>\n",
       "      <td>0.36</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>split1_test_score</th>\n",
       "      <td>0.35</td>\n",
       "      <td>0.35</td>\n",
       "      <td>0.35</td>\n",
       "      <td>0.35</td>\n",
       "      <td>0.35</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>split1_train_score</th>\n",
       "      <td>0.37</td>\n",
       "      <td>0.37</td>\n",
       "      <td>0.37</td>\n",
       "      <td>0.37</td>\n",
       "      <td>0.37</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>split2_test_score</th>\n",
       "      <td>0.36</td>\n",
       "      <td>0.36</td>\n",
       "      <td>0.36</td>\n",
       "      <td>0.36</td>\n",
       "      <td>0.36</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>split2_train_score</th>\n",
       "      <td>0.37</td>\n",
       "      <td>0.37</td>\n",
       "      <td>0.37</td>\n",
       "      <td>0.37</td>\n",
       "      <td>0.37</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>split3_test_score</th>\n",
       "      <td>0.36</td>\n",
       "      <td>0.36</td>\n",
       "      <td>0.36</td>\n",
       "      <td>0.36</td>\n",
       "      <td>0.36</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>split3_train_score</th>\n",
       "      <td>0.37</td>\n",
       "      <td>0.37</td>\n",
       "      <td>0.37</td>\n",
       "      <td>0.37</td>\n",
       "      <td>0.37</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>split4_test_score</th>\n",
       "      <td>0.38</td>\n",
       "      <td>0.38</td>\n",
       "      <td>0.38</td>\n",
       "      <td>0.38</td>\n",
       "      <td>0.38</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>split4_train_score</th>\n",
       "      <td>0.36</td>\n",
       "      <td>0.36</td>\n",
       "      <td>0.36</td>\n",
       "      <td>0.36</td>\n",
       "      <td>0.36</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std_fit_time</th>\n",
       "      <td>3.4e-05</td>\n",
       "      <td>6.3e-06</td>\n",
       "      <td>5e-06</td>\n",
       "      <td>1.2e-05</td>\n",
       "      <td>6.2e-06</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std_score_time</th>\n",
       "      <td>1.3e-05</td>\n",
       "      <td>1.9e-06</td>\n",
       "      <td>1.4e-06</td>\n",
       "      <td>3.3e-06</td>\n",
       "      <td>1.3e-05</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std_test_score</th>\n",
       "      <td>0.011</td>\n",
       "      <td>0.011</td>\n",
       "      <td>0.011</td>\n",
       "      <td>0.011</td>\n",
       "      <td>0.011</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std_train_score</th>\n",
       "      <td>0.0029</td>\n",
       "      <td>0.0029</td>\n",
       "      <td>0.0029</td>\n",
       "      <td>0.0029</td>\n",
       "      <td>0.0029</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                               0                            1  \\\n",
       "mean_fit_time                            0.00054                      0.00051   \n",
       "mean_score_time                          0.00021                      0.00021   \n",
       "mean_test_score                             0.37                         0.37   \n",
       "mean_train_score                            0.37                         0.37   \n",
       "param_C                                    0.001                        0.001   \n",
       "param_gamma                                0.001                         0.01   \n",
       "params              {'C': 0.001, 'gamma': 0.001}  {'C': 0.001, 'gamma': 0.01}   \n",
       "rank_test_score                               22                           22   \n",
       "split0_test_score                           0.38                         0.38   \n",
       "split0_train_score                          0.36                         0.36   \n",
       "split1_test_score                           0.35                         0.35   \n",
       "split1_train_score                          0.37                         0.37   \n",
       "split2_test_score                           0.36                         0.36   \n",
       "split2_train_score                          0.37                         0.37   \n",
       "split3_test_score                           0.36                         0.36   \n",
       "split3_train_score                          0.37                         0.37   \n",
       "split4_test_score                           0.38                         0.38   \n",
       "split4_train_score                          0.36                         0.36   \n",
       "std_fit_time                             3.4e-05                      6.3e-06   \n",
       "std_score_time                           1.3e-05                      1.9e-06   \n",
       "std_test_score                             0.011                        0.011   \n",
       "std_train_score                           0.0029                       0.0029   \n",
       "\n",
       "                                             2                         3  \\\n",
       "mean_fit_time                          0.00052                   0.00052   \n",
       "mean_score_time                        0.00021                   0.00021   \n",
       "mean_test_score                           0.37                      0.37   \n",
       "mean_train_score                          0.37                      0.37   \n",
       "param_C                                  0.001                     0.001   \n",
       "param_gamma                                0.1                         1   \n",
       "params              {'C': 0.001, 'gamma': 0.1}  {'C': 0.001, 'gamma': 1}   \n",
       "rank_test_score                             22                        22   \n",
       "split0_test_score                         0.38                      0.38   \n",
       "split0_train_score                        0.36                      0.36   \n",
       "split1_test_score                         0.35                      0.35   \n",
       "split1_train_score                        0.37                      0.37   \n",
       "split2_test_score                         0.36                      0.36   \n",
       "split2_train_score                        0.37                      0.37   \n",
       "split3_test_score                         0.36                      0.36   \n",
       "split3_train_score                        0.37                      0.37   \n",
       "split4_test_score                         0.38                      0.38   \n",
       "split4_train_score                        0.36                      0.36   \n",
       "std_fit_time                             5e-06                   1.2e-05   \n",
       "std_score_time                         1.4e-06                   3.3e-06   \n",
       "std_test_score                           0.011                     0.011   \n",
       "std_train_score                         0.0029                    0.0029   \n",
       "\n",
       "                                            4  \n",
       "mean_fit_time                          0.0005  \n",
       "mean_score_time                       0.00021  \n",
       "mean_test_score                          0.37  \n",
       "mean_train_score                         0.37  \n",
       "param_C                                 0.001  \n",
       "param_gamma                                10  \n",
       "params              {'C': 0.001, 'gamma': 10}  \n",
       "rank_test_score                            22  \n",
       "split0_test_score                        0.38  \n",
       "split0_train_score                       0.36  \n",
       "split1_test_score                        0.35  \n",
       "split1_train_score                       0.37  \n",
       "split2_test_score                        0.36  \n",
       "split2_train_score                       0.37  \n",
       "split3_test_score                        0.36  \n",
       "split3_train_score                       0.37  \n",
       "split4_test_score                        0.38  \n",
       "split4_train_score                       0.36  \n",
       "std_fit_time                          6.2e-06  \n",
       "std_score_time                        1.3e-05  \n",
       "std_test_score                          0.011  \n",
       "std_train_score                        0.0029  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import pandas as pd\n",
    "pd.set_option('display.max_columns', None)\n",
    "# DataFrame으로 변환합니다\n",
    "results = pd.DataFrame(grid_search.cv_results_)\n",
    "# 처음 다섯 개 행을 출력합니다\n",
    "display(np.transpose(results.head()))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.collections.PolyCollection at 0x7ff3fa845358>"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "application/pdf": "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\n",
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "scores = np.array(results.mean_test_score).reshape(6, 6)\n",
    "\n",
    "# 교차 검증 평균 점수 히트맵 그래프\n",
    "mglearn.tools.heatmap(scores, xlabel='gamma', xticklabels=param_grid['gamma'],\n",
    "                      ylabel='C', yticklabels=param_grid['C'], cmap=\"viridis\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.colorbar.Colorbar at 0x7ff3faaaae48>"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "application/pdf": "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\n",
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 936x360 with 4 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, axes = plt.subplots(1, 3, figsize=(13, 5))\n",
    "\n",
    "param_grid_linear = {'C': np.linspace(1, 2, 6),\n",
    "                     'gamma':  np.linspace(1, 2, 6)}\n",
    "\n",
    "param_grid_one_log = {'C': np.linspace(1, 2, 6),\n",
    "                     'gamma':  np.logspace(-3, 2, 6)}\n",
    "\n",
    "param_grid_range = {'C': np.logspace(-3, 2, 6),\n",
    "                     'gamma':  np.logspace(-7, -2, 6)}\n",
    "\n",
    "for param_grid, ax in zip([param_grid_linear, param_grid_one_log,\n",
    "                           param_grid_range], axes):\n",
    "    grid_search = GridSearchCV(SVC(), param_grid, cv=5, iid=True)\n",
    "    grid_search.fit(X_train, y_train)\n",
    "    scores = grid_search.cv_results_['mean_test_score'].reshape(6, 6)\n",
    "\n",
    "    # 교차 검증 평균 점수의 히트맵 그래프\n",
    "    scores_image = mglearn.tools.heatmap(\n",
    "        scores, xlabel='gamma', ylabel='C', xticklabels=param_grid['gamma'],\n",
    "        yticklabels=param_grid['C'], cmap=\"viridis\", ax=ax)\n",
    "    \n",
    "plt.colorbar(scores_image, ax=axes.tolist())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### 비대칭 매개변수 그리드 탐색"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "그리드 목록:\n",
      "[{'kernel': ['rbf'], 'C': [0.001, 0.01, 0.1, 1, 10, 100], 'gamma': [0.001, 0.01, 0.1, 1, 10, 100]}, {'kernel': ['linear'], 'C': [0.001, 0.01, 0.1, 1, 10, 100]}]\n"
     ]
    }
   ],
   "source": [
    "param_grid = [{'kernel': ['rbf'],\n",
    "               'C': [0.001, 0.01, 0.1, 1, 10, 100],\n",
    "               'gamma': [0.001, 0.01, 0.1, 1, 10, 100]},\n",
    "              {'kernel': ['linear'],\n",
    "               'C': [0.001, 0.01, 0.1, 1, 10, 100]}]\n",
    "print(\"그리드 목록:\\n{}\".format(param_grid))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "최적 파라미터: {'kernel': 'rbf', 'C': 100, 'gamma': 0.01}\n",
      "최고 교차 검증 점수: 0.97\n"
     ]
    }
   ],
   "source": [
    "grid_search = GridSearchCV(SVC(), param_grid, cv=5, return_train_score=True, iid=True)\n",
    "grid_search.fit(X_train, y_train)\n",
    "print(\"최적 파라미터: {}\".format(grid_search.best_params_))\n",
    "print(\"최고 교차 검증 점수: {:.2f}\".format(grid_search.best_score_))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "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",
       "      <th>4</th>\n",
       "      <th>5</th>\n",
       "      <th>6</th>\n",
       "      <th>7</th>\n",
       "      <th>8</th>\n",
       "      <th>9</th>\n",
       "      <th>10</th>\n",
       "      <th>11</th>\n",
       "      <th>12</th>\n",
       "      <th>13</th>\n",
       "      <th>14</th>\n",
       "      <th>15</th>\n",
       "      <th>16</th>\n",
       "      <th>17</th>\n",
       "      <th>18</th>\n",
       "      <th>19</th>\n",
       "      <th>20</th>\n",
       "      <th>21</th>\n",
       "      <th>22</th>\n",
       "      <th>23</th>\n",
       "      <th>24</th>\n",
       "      <th>25</th>\n",
       "      <th>26</th>\n",
       "      <th>27</th>\n",
       "      <th>28</th>\n",
       "      <th>29</th>\n",
       "      <th>30</th>\n",
       "      <th>31</th>\n",
       "      <th>32</th>\n",
       "      <th>33</th>\n",
       "      <th>34</th>\n",
       "      <th>35</th>\n",
       "      <th>36</th>\n",
       "      <th>37</th>\n",
       "      <th>38</th>\n",
       "      <th>39</th>\n",
       "      <th>40</th>\n",
       "      <th>41</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>mean_fit_time</th>\n",
       "      <td>0.00053</td>\n",
       "      <td>0.00051</td>\n",
       "      <td>0.00052</td>\n",
       "      <td>0.00052</td>\n",
       "      <td>0.00051</td>\n",
       "      <td>0.00052</td>\n",
       "      <td>0.00049</td>\n",
       "      <td>0.00049</td>\n",
       "      <td>0.0005</td>\n",
       "      <td>0.00054</td>\n",
       "      <td>0.00055</td>\n",
       "      <td>0.00058</td>\n",
       "      <td>0.00049</td>\n",
       "      <td>0.00053</td>\n",
       "      <td>0.00043</td>\n",
       "      <td>0.00049</td>\n",
       "      <td>0.00055</td>\n",
       "      <td>0.00058</td>\n",
       "      <td>0.00053</td>\n",
       "      <td>0.00039</td>\n",
       "      <td>0.00032</td>\n",
       "      <td>0.00039</td>\n",
       "      <td>0.00064</td>\n",
       "      <td>0.0007</td>\n",
       "      <td>0.0004</td>\n",
       "      <td>0.00032</td>\n",
       "      <td>0.0003</td>\n",
       "      <td>0.00041</td>\n",
       "      <td>0.00064</td>\n",
       "      <td>0.00072</td>\n",
       "      <td>0.00031</td>\n",
       "      <td>0.00028</td>\n",
       "      <td>0.00029</td>\n",
       "      <td>0.0004</td>\n",
       "      <td>0.00067</td>\n",
       "      <td>0.0007</td>\n",
       "      <td>0.00039</td>\n",
       "      <td>0.00035</td>\n",
       "      <td>0.00031</td>\n",
       "      <td>0.00029</td>\n",
       "      <td>0.0003</td>\n",
       "      <td>0.0003</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean_score_time</th>\n",
       "      <td>0.00023</td>\n",
       "      <td>0.00021</td>\n",
       "      <td>0.00021</td>\n",
       "      <td>0.00021</td>\n",
       "      <td>0.00021</td>\n",
       "      <td>0.00021</td>\n",
       "      <td>0.00021</td>\n",
       "      <td>0.00021</td>\n",
       "      <td>0.00021</td>\n",
       "      <td>0.00023</td>\n",
       "      <td>0.00023</td>\n",
       "      <td>0.00021</td>\n",
       "      <td>0.0002</td>\n",
       "      <td>0.00021</td>\n",
       "      <td>0.0002</td>\n",
       "      <td>0.0002</td>\n",
       "      <td>0.00022</td>\n",
       "      <td>0.00021</td>\n",
       "      <td>0.00021</td>\n",
       "      <td>0.0002</td>\n",
       "      <td>0.0002</td>\n",
       "      <td>0.00018</td>\n",
       "      <td>0.00021</td>\n",
       "      <td>0.00024</td>\n",
       "      <td>0.0002</td>\n",
       "      <td>0.00019</td>\n",
       "      <td>0.00018</td>\n",
       "      <td>0.00019</td>\n",
       "      <td>0.00021</td>\n",
       "      <td>0.00023</td>\n",
       "      <td>0.00018</td>\n",
       "      <td>0.0002</td>\n",
       "      <td>0.00017</td>\n",
       "      <td>0.00019</td>\n",
       "      <td>0.00021</td>\n",
       "      <td>0.00023</td>\n",
       "      <td>0.00018</td>\n",
       "      <td>0.00018</td>\n",
       "      <td>0.00017</td>\n",
       "      <td>0.00017</td>\n",
       "      <td>0.00017</td>\n",
       "      <td>0.00017</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean_test_score</th>\n",
       "      <td>0.37</td>\n",
       "      <td>0.37</td>\n",
       "      <td>0.37</td>\n",
       "      <td>0.37</td>\n",
       "      <td>0.37</td>\n",
       "      <td>0.37</td>\n",
       "      <td>0.37</td>\n",
       "      <td>0.37</td>\n",
       "      <td>0.37</td>\n",
       "      <td>0.37</td>\n",
       "      <td>0.37</td>\n",
       "      <td>0.37</td>\n",
       "      <td>0.37</td>\n",
       "      <td>0.7</td>\n",
       "      <td>0.92</td>\n",
       "      <td>0.96</td>\n",
       "      <td>0.37</td>\n",
       "      <td>0.37</td>\n",
       "      <td>0.7</td>\n",
       "      <td>0.93</td>\n",
       "      <td>0.96</td>\n",
       "      <td>0.95</td>\n",
       "      <td>0.92</td>\n",
       "      <td>0.51</td>\n",
       "      <td>0.93</td>\n",
       "      <td>0.96</td>\n",
       "      <td>0.96</td>\n",
       "      <td>0.94</td>\n",
       "      <td>0.92</td>\n",
       "      <td>0.57</td>\n",
       "      <td>0.96</td>\n",
       "      <td>0.97</td>\n",
       "      <td>0.96</td>\n",
       "      <td>0.95</td>\n",
       "      <td>0.92</td>\n",
       "      <td>0.57</td>\n",
       "      <td>0.37</td>\n",
       "      <td>0.85</td>\n",
       "      <td>0.95</td>\n",
       "      <td>0.97</td>\n",
       "      <td>0.96</td>\n",
       "      <td>0.96</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean_train_score</th>\n",
       "      <td>0.37</td>\n",
       "      <td>0.37</td>\n",
       "      <td>0.37</td>\n",
       "      <td>0.37</td>\n",
       "      <td>0.37</td>\n",
       "      <td>0.37</td>\n",
       "      <td>0.37</td>\n",
       "      <td>0.37</td>\n",
       "      <td>0.37</td>\n",
       "      <td>0.37</td>\n",
       "      <td>0.37</td>\n",
       "      <td>0.37</td>\n",
       "      <td>0.37</td>\n",
       "      <td>0.7</td>\n",
       "      <td>0.92</td>\n",
       "      <td>0.96</td>\n",
       "      <td>0.38</td>\n",
       "      <td>0.37</td>\n",
       "      <td>0.7</td>\n",
       "      <td>0.94</td>\n",
       "      <td>0.98</td>\n",
       "      <td>0.98</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0.94</td>\n",
       "      <td>0.98</td>\n",
       "      <td>0.99</td>\n",
       "      <td>0.99</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0.98</td>\n",
       "      <td>0.98</td>\n",
       "      <td>0.99</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0.37</td>\n",
       "      <td>0.86</td>\n",
       "      <td>0.97</td>\n",
       "      <td>0.98</td>\n",
       "      <td>0.99</td>\n",
       "      <td>0.99</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>param_C</th>\n",
       "      <td>0.001</td>\n",
       "      <td>0.001</td>\n",
       "      <td>0.001</td>\n",
       "      <td>0.001</td>\n",
       "      <td>0.001</td>\n",
       "      <td>0.001</td>\n",
       "      <td>0.01</td>\n",
       "      <td>0.01</td>\n",
       "      <td>0.01</td>\n",
       "      <td>0.01</td>\n",
       "      <td>0.01</td>\n",
       "      <td>0.01</td>\n",
       "      <td>0.1</td>\n",
       "      <td>0.1</td>\n",
       "      <td>0.1</td>\n",
       "      <td>0.1</td>\n",
       "      <td>0.1</td>\n",
       "      <td>0.1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>10</td>\n",
       "      <td>10</td>\n",
       "      <td>10</td>\n",
       "      <td>10</td>\n",
       "      <td>10</td>\n",
       "      <td>10</td>\n",
       "      <td>100</td>\n",
       "      <td>100</td>\n",
       "      <td>100</td>\n",
       "      <td>100</td>\n",
       "      <td>100</td>\n",
       "      <td>100</td>\n",
       "      <td>0.001</td>\n",
       "      <td>0.01</td>\n",
       "      <td>0.1</td>\n",
       "      <td>1</td>\n",
       "      <td>10</td>\n",
       "      <td>100</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>param_gamma</th>\n",
       "      <td>0.001</td>\n",
       "      <td>0.01</td>\n",
       "      <td>0.1</td>\n",
       "      <td>1</td>\n",
       "      <td>10</td>\n",
       "      <td>100</td>\n",
       "      <td>0.001</td>\n",
       "      <td>0.01</td>\n",
       "      <td>0.1</td>\n",
       "      <td>1</td>\n",
       "      <td>10</td>\n",
       "      <td>100</td>\n",
       "      <td>0.001</td>\n",
       "      <td>0.01</td>\n",
       "      <td>0.1</td>\n",
       "      <td>1</td>\n",
       "      <td>10</td>\n",
       "      <td>100</td>\n",
       "      <td>0.001</td>\n",
       "      <td>0.01</td>\n",
       "      <td>0.1</td>\n",
       "      <td>1</td>\n",
       "      <td>10</td>\n",
       "      <td>100</td>\n",
       "      <td>0.001</td>\n",
       "      <td>0.01</td>\n",
       "      <td>0.1</td>\n",
       "      <td>1</td>\n",
       "      <td>10</td>\n",
       "      <td>100</td>\n",
       "      <td>0.001</td>\n",
       "      <td>0.01</td>\n",
       "      <td>0.1</td>\n",
       "      <td>1</td>\n",
       "      <td>10</td>\n",
       "      <td>100</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>param_kernel</th>\n",
       "      <td>rbf</td>\n",
       "      <td>rbf</td>\n",
       "      <td>rbf</td>\n",
       "      <td>rbf</td>\n",
       "      <td>rbf</td>\n",
       "      <td>rbf</td>\n",
       "      <td>rbf</td>\n",
       "      <td>rbf</td>\n",
       "      <td>rbf</td>\n",
       "      <td>rbf</td>\n",
       "      <td>rbf</td>\n",
       "      <td>rbf</td>\n",
       "      <td>rbf</td>\n",
       "      <td>rbf</td>\n",
       "      <td>rbf</td>\n",
       "      <td>rbf</td>\n",
       "      <td>rbf</td>\n",
       "      <td>rbf</td>\n",
       "      <td>rbf</td>\n",
       "      <td>rbf</td>\n",
       "      <td>rbf</td>\n",
       "      <td>rbf</td>\n",
       "      <td>rbf</td>\n",
       "      <td>rbf</td>\n",
       "      <td>rbf</td>\n",
       "      <td>rbf</td>\n",
       "      <td>rbf</td>\n",
       "      <td>rbf</td>\n",
       "      <td>rbf</td>\n",
       "      <td>rbf</td>\n",
       "      <td>rbf</td>\n",
       "      <td>rbf</td>\n",
       "      <td>rbf</td>\n",
       "      <td>rbf</td>\n",
       "      <td>rbf</td>\n",
       "      <td>rbf</td>\n",
       "      <td>linear</td>\n",
       "      <td>linear</td>\n",
       "      <td>linear</td>\n",
       "      <td>linear</td>\n",
       "      <td>linear</td>\n",
       "      <td>linear</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>params</th>\n",
       "      <td>{'kernel': 'rbf', 'C': 0.001, 'gamma': 0.001}</td>\n",
       "      <td>{'kernel': 'rbf', 'C': 0.001, 'gamma': 0.01}</td>\n",
       "      <td>{'kernel': 'rbf', 'C': 0.001, 'gamma': 0.1}</td>\n",
       "      <td>{'kernel': 'rbf', 'C': 0.001, 'gamma': 1}</td>\n",
       "      <td>{'kernel': 'rbf', 'C': 0.001, 'gamma': 10}</td>\n",
       "      <td>{'kernel': 'rbf', 'C': 0.001, 'gamma': 100}</td>\n",
       "      <td>{'kernel': 'rbf', 'C': 0.01, 'gamma': 0.001}</td>\n",
       "      <td>{'kernel': 'rbf', 'C': 0.01, 'gamma': 0.01}</td>\n",
       "      <td>{'kernel': 'rbf', 'C': 0.01, 'gamma': 0.1}</td>\n",
       "      <td>{'kernel': 'rbf', 'C': 0.01, 'gamma': 1}</td>\n",
       "      <td>{'kernel': 'rbf', 'C': 0.01, 'gamma': 10}</td>\n",
       "      <td>{'kernel': 'rbf', 'C': 0.01, 'gamma': 100}</td>\n",
       "      <td>{'kernel': 'rbf', 'C': 0.1, 'gamma': 0.001}</td>\n",
       "      <td>{'kernel': 'rbf', 'C': 0.1, 'gamma': 0.01}</td>\n",
       "      <td>{'kernel': 'rbf', 'C': 0.1, 'gamma': 0.1}</td>\n",
       "      <td>{'kernel': 'rbf', 'C': 0.1, 'gamma': 1}</td>\n",
       "      <td>{'kernel': 'rbf', 'C': 0.1, 'gamma': 10}</td>\n",
       "      <td>{'kernel': 'rbf', 'C': 0.1, 'gamma': 100}</td>\n",
       "      <td>{'kernel': 'rbf', 'C': 1, 'gamma': 0.001}</td>\n",
       "      <td>{'kernel': 'rbf', 'C': 1, 'gamma': 0.01}</td>\n",
       "      <td>{'kernel': 'rbf', 'C': 1, 'gamma': 0.1}</td>\n",
       "      <td>{'kernel': 'rbf', 'C': 1, 'gamma': 1}</td>\n",
       "      <td>{'kernel': 'rbf', 'C': 1, 'gamma': 10}</td>\n",
       "      <td>{'kernel': 'rbf', 'C': 1, 'gamma': 100}</td>\n",
       "      <td>{'kernel': 'rbf', 'C': 10, 'gamma': 0.001}</td>\n",
       "      <td>{'kernel': 'rbf', 'C': 10, 'gamma': 0.01}</td>\n",
       "      <td>{'kernel': 'rbf', 'C': 10, 'gamma': 0.1}</td>\n",
       "      <td>{'kernel': 'rbf', 'C': 10, 'gamma': 1}</td>\n",
       "      <td>{'kernel': 'rbf', 'C': 10, 'gamma': 10}</td>\n",
       "      <td>{'kernel': 'rbf', 'C': 10, 'gamma': 100}</td>\n",
       "      <td>{'kernel': 'rbf', 'C': 100, 'gamma': 0.001}</td>\n",
       "      <td>{'kernel': 'rbf', 'C': 100, 'gamma': 0.01}</td>\n",
       "      <td>{'kernel': 'rbf', 'C': 100, 'gamma': 0.1}</td>\n",
       "      <td>{'kernel': 'rbf', 'C': 100, 'gamma': 1}</td>\n",
       "      <td>{'kernel': 'rbf', 'C': 100, 'gamma': 10}</td>\n",
       "      <td>{'kernel': 'rbf', 'C': 100, 'gamma': 100}</td>\n",
       "      <td>{'kernel': 'linear', 'C': 0.001}</td>\n",
       "      <td>{'kernel': 'linear', 'C': 0.01}</td>\n",
       "      <td>{'kernel': 'linear', 'C': 0.1}</td>\n",
       "      <td>{'kernel': 'linear', 'C': 1}</td>\n",
       "      <td>{'kernel': 'linear', 'C': 10}</td>\n",
       "      <td>{'kernel': 'linear', 'C': 100}</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>rank_test_score</th>\n",
       "      <td>27</td>\n",
       "      <td>27</td>\n",
       "      <td>27</td>\n",
       "      <td>27</td>\n",
       "      <td>27</td>\n",
       "      <td>27</td>\n",
       "      <td>27</td>\n",
       "      <td>27</td>\n",
       "      <td>27</td>\n",
       "      <td>27</td>\n",
       "      <td>27</td>\n",
       "      <td>27</td>\n",
       "      <td>27</td>\n",
       "      <td>22</td>\n",
       "      <td>17</td>\n",
       "      <td>9</td>\n",
       "      <td>27</td>\n",
       "      <td>27</td>\n",
       "      <td>22</td>\n",
       "      <td>15</td>\n",
       "      <td>3</td>\n",
       "      <td>11</td>\n",
       "      <td>17</td>\n",
       "      <td>26</td>\n",
       "      <td>15</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>14</td>\n",
       "      <td>17</td>\n",
       "      <td>24</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>9</td>\n",
       "      <td>11</td>\n",
       "      <td>17</td>\n",
       "      <td>24</td>\n",
       "      <td>27</td>\n",
       "      <td>21</td>\n",
       "      <td>11</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>split0_test_score</th>\n",
       "      <td>0.38</td>\n",
       "      <td>0.38</td>\n",
       "      <td>0.38</td>\n",
       "      <td>0.38</td>\n",
       "      <td>0.38</td>\n",
       "      <td>0.38</td>\n",
       "      <td>0.38</td>\n",
       "      <td>0.38</td>\n",
       "      <td>0.38</td>\n",
       "      <td>0.38</td>\n",
       "      <td>0.38</td>\n",
       "      <td>0.38</td>\n",
       "      <td>0.38</td>\n",
       "      <td>0.71</td>\n",
       "      <td>0.92</td>\n",
       "      <td>1</td>\n",
       "      <td>0.38</td>\n",
       "      <td>0.38</td>\n",
       "      <td>0.71</td>\n",
       "      <td>0.88</td>\n",
       "      <td>1</td>\n",
       "      <td>0.92</td>\n",
       "      <td>0.88</td>\n",
       "      <td>0.5</td>\n",
       "      <td>0.88</td>\n",
       "      <td>1</td>\n",
       "      <td>0.96</td>\n",
       "      <td>0.92</td>\n",
       "      <td>0.88</td>\n",
       "      <td>0.54</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0.96</td>\n",
       "      <td>0.92</td>\n",
       "      <td>0.88</td>\n",
       "      <td>0.54</td>\n",
       "      <td>0.38</td>\n",
       "      <td>0.92</td>\n",
       "      <td>0.96</td>\n",
       "      <td>1</td>\n",
       "      <td>0.96</td>\n",
       "      <td>0.96</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>split0_train_score</th>\n",
       "      <td>0.36</td>\n",
       "      <td>0.36</td>\n",
       "      <td>0.36</td>\n",
       "      <td>0.36</td>\n",
       "      <td>0.36</td>\n",
       "      <td>0.36</td>\n",
       "      <td>0.36</td>\n",
       "      <td>0.36</td>\n",
       "      <td>0.36</td>\n",
       "      <td>0.36</td>\n",
       "      <td>0.36</td>\n",
       "      <td>0.36</td>\n",
       "      <td>0.36</td>\n",
       "      <td>0.69</td>\n",
       "      <td>0.94</td>\n",
       "      <td>0.95</td>\n",
       "      <td>0.38</td>\n",
       "      <td>0.36</td>\n",
       "      <td>0.69</td>\n",
       "      <td>0.94</td>\n",
       "      <td>0.97</td>\n",
       "      <td>0.99</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0.94</td>\n",
       "      <td>0.98</td>\n",
       "      <td>0.99</td>\n",
       "      <td>0.99</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0.98</td>\n",
       "      <td>0.99</td>\n",
       "      <td>0.99</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0.36</td>\n",
       "      <td>0.89</td>\n",
       "      <td>0.97</td>\n",
       "      <td>0.99</td>\n",
       "      <td>0.99</td>\n",
       "      <td>0.99</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>split1_test_score</th>\n",
       "      <td>0.35</td>\n",
       "      <td>0.35</td>\n",
       "      <td>0.35</td>\n",
       "      <td>0.35</td>\n",
       "      <td>0.35</td>\n",
       "      <td>0.35</td>\n",
       "      <td>0.35</td>\n",
       "      <td>0.35</td>\n",
       "      <td>0.35</td>\n",
       "      <td>0.35</td>\n",
       "      <td>0.35</td>\n",
       "      <td>0.35</td>\n",
       "      <td>0.35</td>\n",
       "      <td>0.7</td>\n",
       "      <td>0.91</td>\n",
       "      <td>0.91</td>\n",
       "      <td>0.35</td>\n",
       "      <td>0.35</td>\n",
       "      <td>0.7</td>\n",
       "      <td>0.91</td>\n",
       "      <td>0.96</td>\n",
       "      <td>0.96</td>\n",
       "      <td>0.96</td>\n",
       "      <td>0.43</td>\n",
       "      <td>0.91</td>\n",
       "      <td>0.96</td>\n",
       "      <td>0.96</td>\n",
       "      <td>0.96</td>\n",
       "      <td>0.96</td>\n",
       "      <td>0.52</td>\n",
       "      <td>0.96</td>\n",
       "      <td>0.96</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0.96</td>\n",
       "      <td>0.52</td>\n",
       "      <td>0.35</td>\n",
       "      <td>0.83</td>\n",
       "      <td>0.91</td>\n",
       "      <td>0.96</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>split1_train_score</th>\n",
       "      <td>0.37</td>\n",
       "      <td>0.37</td>\n",
       "      <td>0.37</td>\n",
       "      <td>0.37</td>\n",
       "      <td>0.37</td>\n",
       "      <td>0.37</td>\n",
       "      <td>0.37</td>\n",
       "      <td>0.37</td>\n",
       "      <td>0.37</td>\n",
       "      <td>0.37</td>\n",
       "      <td>0.37</td>\n",
       "      <td>0.37</td>\n",
       "      <td>0.37</td>\n",
       "      <td>0.7</td>\n",
       "      <td>0.93</td>\n",
       "      <td>0.98</td>\n",
       "      <td>0.39</td>\n",
       "      <td>0.37</td>\n",
       "      <td>0.7</td>\n",
       "      <td>0.94</td>\n",
       "      <td>0.99</td>\n",
       "      <td>0.98</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0.94</td>\n",
       "      <td>0.98</td>\n",
       "      <td>0.98</td>\n",
       "      <td>0.98</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0.98</td>\n",
       "      <td>0.98</td>\n",
       "      <td>0.98</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0.37</td>\n",
       "      <td>0.89</td>\n",
       "      <td>0.98</td>\n",
       "      <td>0.98</td>\n",
       "      <td>0.99</td>\n",
       "      <td>0.99</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>split2_test_score</th>\n",
       "      <td>0.36</td>\n",
       "      <td>0.36</td>\n",
       "      <td>0.36</td>\n",
       "      <td>0.36</td>\n",
       "      <td>0.36</td>\n",
       "      <td>0.36</td>\n",
       "      <td>0.36</td>\n",
       "      <td>0.36</td>\n",
       "      <td>0.36</td>\n",
       "      <td>0.36</td>\n",
       "      <td>0.36</td>\n",
       "      <td>0.36</td>\n",
       "      <td>0.36</td>\n",
       "      <td>0.68</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0.36</td>\n",
       "      <td>0.36</td>\n",
       "      <td>0.68</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0.55</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0.59</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0.59</td>\n",
       "      <td>0.36</td>\n",
       "      <td>0.82</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>split2_train_score</th>\n",
       "      <td>0.37</td>\n",
       "      <td>0.37</td>\n",
       "      <td>0.37</td>\n",
       "      <td>0.37</td>\n",
       "      <td>0.37</td>\n",
       "      <td>0.37</td>\n",
       "      <td>0.37</td>\n",
       "      <td>0.37</td>\n",
       "      <td>0.37</td>\n",
       "      <td>0.37</td>\n",
       "      <td>0.37</td>\n",
       "      <td>0.37</td>\n",
       "      <td>0.37</td>\n",
       "      <td>0.7</td>\n",
       "      <td>0.9</td>\n",
       "      <td>0.94</td>\n",
       "      <td>0.37</td>\n",
       "      <td>0.37</td>\n",
       "      <td>0.7</td>\n",
       "      <td>0.92</td>\n",
       "      <td>0.97</td>\n",
       "      <td>0.98</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0.92</td>\n",
       "      <td>0.97</td>\n",
       "      <td>0.98</td>\n",
       "      <td>0.98</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0.97</td>\n",
       "      <td>0.98</td>\n",
       "      <td>0.98</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0.37</td>\n",
       "      <td>0.87</td>\n",
       "      <td>0.94</td>\n",
       "      <td>0.98</td>\n",
       "      <td>0.98</td>\n",
       "      <td>0.99</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>split3_test_score</th>\n",
       "      <td>0.36</td>\n",
       "      <td>0.36</td>\n",
       "      <td>0.36</td>\n",
       "      <td>0.36</td>\n",
       "      <td>0.36</td>\n",
       "      <td>0.36</td>\n",
       "      <td>0.36</td>\n",
       "      <td>0.36</td>\n",
       "      <td>0.36</td>\n",
       "      <td>0.36</td>\n",
       "      <td>0.36</td>\n",
       "      <td>0.36</td>\n",
       "      <td>0.36</td>\n",
       "      <td>0.68</td>\n",
       "      <td>0.86</td>\n",
       "      <td>0.91</td>\n",
       "      <td>0.36</td>\n",
       "      <td>0.36</td>\n",
       "      <td>0.68</td>\n",
       "      <td>0.91</td>\n",
       "      <td>0.91</td>\n",
       "      <td>0.91</td>\n",
       "      <td>0.82</td>\n",
       "      <td>0.5</td>\n",
       "      <td>0.91</td>\n",
       "      <td>0.91</td>\n",
       "      <td>0.95</td>\n",
       "      <td>0.86</td>\n",
       "      <td>0.82</td>\n",
       "      <td>0.59</td>\n",
       "      <td>0.91</td>\n",
       "      <td>0.95</td>\n",
       "      <td>0.86</td>\n",
       "      <td>0.86</td>\n",
       "      <td>0.82</td>\n",
       "      <td>0.59</td>\n",
       "      <td>0.36</td>\n",
       "      <td>0.77</td>\n",
       "      <td>0.91</td>\n",
       "      <td>0.95</td>\n",
       "      <td>0.91</td>\n",
       "      <td>0.91</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>split3_train_score</th>\n",
       "      <td>0.37</td>\n",
       "      <td>0.37</td>\n",
       "      <td>0.37</td>\n",
       "      <td>0.37</td>\n",
       "      <td>0.37</td>\n",
       "      <td>0.37</td>\n",
       "      <td>0.37</td>\n",
       "      <td>0.37</td>\n",
       "      <td>0.37</td>\n",
       "      <td>0.37</td>\n",
       "      <td>0.37</td>\n",
       "      <td>0.37</td>\n",
       "      <td>0.37</td>\n",
       "      <td>0.7</td>\n",
       "      <td>0.89</td>\n",
       "      <td>0.97</td>\n",
       "      <td>0.41</td>\n",
       "      <td>0.37</td>\n",
       "      <td>0.7</td>\n",
       "      <td>0.93</td>\n",
       "      <td>0.99</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0.93</td>\n",
       "      <td>0.99</td>\n",
       "      <td>0.99</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0.99</td>\n",
       "      <td>0.99</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0.37</td>\n",
       "      <td>0.76</td>\n",
       "      <td>0.98</td>\n",
       "      <td>0.99</td>\n",
       "      <td>0.99</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>split4_test_score</th>\n",
       "      <td>0.38</td>\n",
       "      <td>0.38</td>\n",
       "      <td>0.38</td>\n",
       "      <td>0.38</td>\n",
       "      <td>0.38</td>\n",
       "      <td>0.38</td>\n",
       "      <td>0.38</td>\n",
       "      <td>0.38</td>\n",
       "      <td>0.38</td>\n",
       "      <td>0.38</td>\n",
       "      <td>0.38</td>\n",
       "      <td>0.38</td>\n",
       "      <td>0.38</td>\n",
       "      <td>0.71</td>\n",
       "      <td>0.9</td>\n",
       "      <td>0.95</td>\n",
       "      <td>0.38</td>\n",
       "      <td>0.38</td>\n",
       "      <td>0.71</td>\n",
       "      <td>0.95</td>\n",
       "      <td>0.95</td>\n",
       "      <td>0.95</td>\n",
       "      <td>0.95</td>\n",
       "      <td>0.57</td>\n",
       "      <td>0.95</td>\n",
       "      <td>0.95</td>\n",
       "      <td>0.95</td>\n",
       "      <td>0.95</td>\n",
       "      <td>0.95</td>\n",
       "      <td>0.62</td>\n",
       "      <td>0.95</td>\n",
       "      <td>0.95</td>\n",
       "      <td>0.95</td>\n",
       "      <td>0.95</td>\n",
       "      <td>0.95</td>\n",
       "      <td>0.62</td>\n",
       "      <td>0.38</td>\n",
       "      <td>0.9</td>\n",
       "      <td>0.95</td>\n",
       "      <td>0.95</td>\n",
       "      <td>0.95</td>\n",
       "      <td>0.95</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>split4_train_score</th>\n",
       "      <td>0.36</td>\n",
       "      <td>0.36</td>\n",
       "      <td>0.36</td>\n",
       "      <td>0.36</td>\n",
       "      <td>0.36</td>\n",
       "      <td>0.36</td>\n",
       "      <td>0.36</td>\n",
       "      <td>0.36</td>\n",
       "      <td>0.36</td>\n",
       "      <td>0.36</td>\n",
       "      <td>0.36</td>\n",
       "      <td>0.36</td>\n",
       "      <td>0.36</td>\n",
       "      <td>0.69</td>\n",
       "      <td>0.93</td>\n",
       "      <td>0.96</td>\n",
       "      <td>0.36</td>\n",
       "      <td>0.36</td>\n",
       "      <td>0.69</td>\n",
       "      <td>0.93</td>\n",
       "      <td>0.98</td>\n",
       "      <td>0.98</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0.93</td>\n",
       "      <td>0.98</td>\n",
       "      <td>1</td>\n",
       "      <td>0.99</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0.98</td>\n",
       "      <td>0.99</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0.36</td>\n",
       "      <td>0.88</td>\n",
       "      <td>0.97</td>\n",
       "      <td>0.99</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std_fit_time</th>\n",
       "      <td>3.1e-05</td>\n",
       "      <td>5.2e-06</td>\n",
       "      <td>5.2e-06</td>\n",
       "      <td>5.3e-06</td>\n",
       "      <td>3.4e-06</td>\n",
       "      <td>1.3e-05</td>\n",
       "      <td>5.1e-06</td>\n",
       "      <td>5.9e-06</td>\n",
       "      <td>6.4e-06</td>\n",
       "      <td>3.3e-05</td>\n",
       "      <td>2.5e-05</td>\n",
       "      <td>1.1e-05</td>\n",
       "      <td>6e-06</td>\n",
       "      <td>4.5e-05</td>\n",
       "      <td>3.3e-06</td>\n",
       "      <td>4.8e-05</td>\n",
       "      <td>8.5e-06</td>\n",
       "      <td>1.2e-05</td>\n",
       "      <td>3.8e-05</td>\n",
       "      <td>4.1e-06</td>\n",
       "      <td>6.7e-06</td>\n",
       "      <td>2.3e-05</td>\n",
       "      <td>2.5e-05</td>\n",
       "      <td>6.8e-05</td>\n",
       "      <td>5.9e-06</td>\n",
       "      <td>6.1e-06</td>\n",
       "      <td>9e-06</td>\n",
       "      <td>6.9e-05</td>\n",
       "      <td>1.7e-05</td>\n",
       "      <td>8.1e-05</td>\n",
       "      <td>1.7e-05</td>\n",
       "      <td>1.6e-05</td>\n",
       "      <td>1.2e-05</td>\n",
       "      <td>1.5e-05</td>\n",
       "      <td>4e-05</td>\n",
       "      <td>3.6e-05</td>\n",
       "      <td>6.5e-06</td>\n",
       "      <td>1.1e-06</td>\n",
       "      <td>1.3e-05</td>\n",
       "      <td>8.7e-06</td>\n",
       "      <td>2.4e-05</td>\n",
       "      <td>2.3e-05</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std_score_time</th>\n",
       "      <td>3.5e-05</td>\n",
       "      <td>4e-06</td>\n",
       "      <td>2.3e-06</td>\n",
       "      <td>1.7e-06</td>\n",
       "      <td>3.1e-06</td>\n",
       "      <td>1.8e-06</td>\n",
       "      <td>3.1e-06</td>\n",
       "      <td>1.6e-06</td>\n",
       "      <td>6.8e-06</td>\n",
       "      <td>1.5e-05</td>\n",
       "      <td>2.7e-05</td>\n",
       "      <td>4.3e-06</td>\n",
       "      <td>2.1e-06</td>\n",
       "      <td>8.9e-06</td>\n",
       "      <td>1.9e-06</td>\n",
       "      <td>4.3e-06</td>\n",
       "      <td>5.5e-06</td>\n",
       "      <td>1.5e-06</td>\n",
       "      <td>6e-06</td>\n",
       "      <td>2.5e-06</td>\n",
       "      <td>3.8e-05</td>\n",
       "      <td>1.9e-06</td>\n",
       "      <td>2.8e-06</td>\n",
       "      <td>4.2e-05</td>\n",
       "      <td>1.9e-06</td>\n",
       "      <td>1.1e-06</td>\n",
       "      <td>1e-06</td>\n",
       "      <td>1.6e-05</td>\n",
       "      <td>1.1e-05</td>\n",
       "      <td>4.2e-05</td>\n",
       "      <td>1.1e-05</td>\n",
       "      <td>5.2e-05</td>\n",
       "      <td>2.9e-06</td>\n",
       "      <td>1.3e-05</td>\n",
       "      <td>3.1e-06</td>\n",
       "      <td>1.4e-05</td>\n",
       "      <td>1.2e-06</td>\n",
       "      <td>7.6e-07</td>\n",
       "      <td>1.6e-06</td>\n",
       "      <td>1e-06</td>\n",
       "      <td>4.6e-07</td>\n",
       "      <td>5.3e-06</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std_test_score</th>\n",
       "      <td>0.011</td>\n",
       "      <td>0.011</td>\n",
       "      <td>0.011</td>\n",
       "      <td>0.011</td>\n",
       "      <td>0.011</td>\n",
       "      <td>0.011</td>\n",
       "      <td>0.011</td>\n",
       "      <td>0.011</td>\n",
       "      <td>0.011</td>\n",
       "      <td>0.011</td>\n",
       "      <td>0.011</td>\n",
       "      <td>0.011</td>\n",
       "      <td>0.011</td>\n",
       "      <td>0.013</td>\n",
       "      <td>0.044</td>\n",
       "      <td>0.04</td>\n",
       "      <td>0.011</td>\n",
       "      <td>0.011</td>\n",
       "      <td>0.013</td>\n",
       "      <td>0.043</td>\n",
       "      <td>0.034</td>\n",
       "      <td>0.032</td>\n",
       "      <td>0.065</td>\n",
       "      <td>0.046</td>\n",
       "      <td>0.043</td>\n",
       "      <td>0.034</td>\n",
       "      <td>0.018</td>\n",
       "      <td>0.045</td>\n",
       "      <td>0.065</td>\n",
       "      <td>0.036</td>\n",
       "      <td>0.034</td>\n",
       "      <td>0.022</td>\n",
       "      <td>0.05</td>\n",
       "      <td>0.052</td>\n",
       "      <td>0.065</td>\n",
       "      <td>0.036</td>\n",
       "      <td>0.011</td>\n",
       "      <td>0.055</td>\n",
       "      <td>0.033</td>\n",
       "      <td>0.022</td>\n",
       "      <td>0.034</td>\n",
       "      <td>0.034</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std_train_score</th>\n",
       "      <td>0.0029</td>\n",
       "      <td>0.0029</td>\n",
       "      <td>0.0029</td>\n",
       "      <td>0.0029</td>\n",
       "      <td>0.0029</td>\n",
       "      <td>0.0029</td>\n",
       "      <td>0.0029</td>\n",
       "      <td>0.0029</td>\n",
       "      <td>0.0029</td>\n",
       "      <td>0.0029</td>\n",
       "      <td>0.0029</td>\n",
       "      <td>0.0029</td>\n",
       "      <td>0.0029</td>\n",
       "      <td>0.0033</td>\n",
       "      <td>0.021</td>\n",
       "      <td>0.011</td>\n",
       "      <td>0.018</td>\n",
       "      <td>0.0029</td>\n",
       "      <td>0.0033</td>\n",
       "      <td>0.0079</td>\n",
       "      <td>0.01</td>\n",
       "      <td>0.0089</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0079</td>\n",
       "      <td>0.007</td>\n",
       "      <td>0.0084</td>\n",
       "      <td>0.0084</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.007</td>\n",
       "      <td>0.0055</td>\n",
       "      <td>0.01</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0029</td>\n",
       "      <td>0.05</td>\n",
       "      <td>0.012</td>\n",
       "      <td>0.0055</td>\n",
       "      <td>0.007</td>\n",
       "      <td>0.0055</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                               0   \\\n",
       "mean_fit_time                                             0.00053   \n",
       "mean_score_time                                           0.00023   \n",
       "mean_test_score                                              0.37   \n",
       "mean_train_score                                             0.37   \n",
       "param_C                                                     0.001   \n",
       "param_gamma                                                 0.001   \n",
       "param_kernel                                                  rbf   \n",
       "params              {'kernel': 'rbf', 'C': 0.001, 'gamma': 0.001}   \n",
       "rank_test_score                                                27   \n",
       "split0_test_score                                            0.38   \n",
       "split0_train_score                                           0.36   \n",
       "split1_test_score                                            0.35   \n",
       "split1_train_score                                           0.37   \n",
       "split2_test_score                                            0.36   \n",
       "split2_train_score                                           0.37   \n",
       "split3_test_score                                            0.36   \n",
       "split3_train_score                                           0.37   \n",
       "split4_test_score                                            0.38   \n",
       "split4_train_score                                           0.36   \n",
       "std_fit_time                                              3.1e-05   \n",
       "std_score_time                                            3.5e-05   \n",
       "std_test_score                                              0.011   \n",
       "std_train_score                                            0.0029   \n",
       "\n",
       "                                                              1   \\\n",
       "mean_fit_time                                            0.00051   \n",
       "mean_score_time                                          0.00021   \n",
       "mean_test_score                                             0.37   \n",
       "mean_train_score                                            0.37   \n",
       "param_C                                                    0.001   \n",
       "param_gamma                                                 0.01   \n",
       "param_kernel                                                 rbf   \n",
       "params              {'kernel': 'rbf', 'C': 0.001, 'gamma': 0.01}   \n",
       "rank_test_score                                               27   \n",
       "split0_test_score                                           0.38   \n",
       "split0_train_score                                          0.36   \n",
       "split1_test_score                                           0.35   \n",
       "split1_train_score                                          0.37   \n",
       "split2_test_score                                           0.36   \n",
       "split2_train_score                                          0.37   \n",
       "split3_test_score                                           0.36   \n",
       "split3_train_score                                          0.37   \n",
       "split4_test_score                                           0.38   \n",
       "split4_train_score                                          0.36   \n",
       "std_fit_time                                             5.2e-06   \n",
       "std_score_time                                             4e-06   \n",
       "std_test_score                                             0.011   \n",
       "std_train_score                                           0.0029   \n",
       "\n",
       "                                                             2   \\\n",
       "mean_fit_time                                           0.00052   \n",
       "mean_score_time                                         0.00021   \n",
       "mean_test_score                                            0.37   \n",
       "mean_train_score                                           0.37   \n",
       "param_C                                                   0.001   \n",
       "param_gamma                                                 0.1   \n",
       "param_kernel                                                rbf   \n",
       "params              {'kernel': 'rbf', 'C': 0.001, 'gamma': 0.1}   \n",
       "rank_test_score                                              27   \n",
       "split0_test_score                                          0.38   \n",
       "split0_train_score                                         0.36   \n",
       "split1_test_score                                          0.35   \n",
       "split1_train_score                                         0.37   \n",
       "split2_test_score                                          0.36   \n",
       "split2_train_score                                         0.37   \n",
       "split3_test_score                                          0.36   \n",
       "split3_train_score                                         0.37   \n",
       "split4_test_score                                          0.38   \n",
       "split4_train_score                                         0.36   \n",
       "std_fit_time                                            5.2e-06   \n",
       "std_score_time                                          2.3e-06   \n",
       "std_test_score                                            0.011   \n",
       "std_train_score                                          0.0029   \n",
       "\n",
       "                                                           3   \\\n",
       "mean_fit_time                                         0.00052   \n",
       "mean_score_time                                       0.00021   \n",
       "mean_test_score                                          0.37   \n",
       "mean_train_score                                         0.37   \n",
       "param_C                                                 0.001   \n",
       "param_gamma                                                 1   \n",
       "param_kernel                                              rbf   \n",
       "params              {'kernel': 'rbf', 'C': 0.001, 'gamma': 1}   \n",
       "rank_test_score                                            27   \n",
       "split0_test_score                                        0.38   \n",
       "split0_train_score                                       0.36   \n",
       "split1_test_score                                        0.35   \n",
       "split1_train_score                                       0.37   \n",
       "split2_test_score                                        0.36   \n",
       "split2_train_score                                       0.37   \n",
       "split3_test_score                                        0.36   \n",
       "split3_train_score                                       0.37   \n",
       "split4_test_score                                        0.38   \n",
       "split4_train_score                                       0.36   \n",
       "std_fit_time                                          5.3e-06   \n",
       "std_score_time                                        1.7e-06   \n",
       "std_test_score                                          0.011   \n",
       "std_train_score                                        0.0029   \n",
       "\n",
       "                                                            4   \\\n",
       "mean_fit_time                                          0.00051   \n",
       "mean_score_time                                        0.00021   \n",
       "mean_test_score                                           0.37   \n",
       "mean_train_score                                          0.37   \n",
       "param_C                                                  0.001   \n",
       "param_gamma                                                 10   \n",
       "param_kernel                                               rbf   \n",
       "params              {'kernel': 'rbf', 'C': 0.001, 'gamma': 10}   \n",
       "rank_test_score                                             27   \n",
       "split0_test_score                                         0.38   \n",
       "split0_train_score                                        0.36   \n",
       "split1_test_score                                         0.35   \n",
       "split1_train_score                                        0.37   \n",
       "split2_test_score                                         0.36   \n",
       "split2_train_score                                        0.37   \n",
       "split3_test_score                                         0.36   \n",
       "split3_train_score                                        0.37   \n",
       "split4_test_score                                         0.38   \n",
       "split4_train_score                                        0.36   \n",
       "std_fit_time                                           3.4e-06   \n",
       "std_score_time                                         3.1e-06   \n",
       "std_test_score                                           0.011   \n",
       "std_train_score                                         0.0029   \n",
       "\n",
       "                                                             5   \\\n",
       "mean_fit_time                                           0.00052   \n",
       "mean_score_time                                         0.00021   \n",
       "mean_test_score                                            0.37   \n",
       "mean_train_score                                           0.37   \n",
       "param_C                                                   0.001   \n",
       "param_gamma                                                 100   \n",
       "param_kernel                                                rbf   \n",
       "params              {'kernel': 'rbf', 'C': 0.001, 'gamma': 100}   \n",
       "rank_test_score                                              27   \n",
       "split0_test_score                                          0.38   \n",
       "split0_train_score                                         0.36   \n",
       "split1_test_score                                          0.35   \n",
       "split1_train_score                                         0.37   \n",
       "split2_test_score                                          0.36   \n",
       "split2_train_score                                         0.37   \n",
       "split3_test_score                                          0.36   \n",
       "split3_train_score                                         0.37   \n",
       "split4_test_score                                          0.38   \n",
       "split4_train_score                                         0.36   \n",
       "std_fit_time                                            1.3e-05   \n",
       "std_score_time                                          1.8e-06   \n",
       "std_test_score                                            0.011   \n",
       "std_train_score                                          0.0029   \n",
       "\n",
       "                                                              6   \\\n",
       "mean_fit_time                                            0.00049   \n",
       "mean_score_time                                          0.00021   \n",
       "mean_test_score                                             0.37   \n",
       "mean_train_score                                            0.37   \n",
       "param_C                                                     0.01   \n",
       "param_gamma                                                0.001   \n",
       "param_kernel                                                 rbf   \n",
       "params              {'kernel': 'rbf', 'C': 0.01, 'gamma': 0.001}   \n",
       "rank_test_score                                               27   \n",
       "split0_test_score                                           0.38   \n",
       "split0_train_score                                          0.36   \n",
       "split1_test_score                                           0.35   \n",
       "split1_train_score                                          0.37   \n",
       "split2_test_score                                           0.36   \n",
       "split2_train_score                                          0.37   \n",
       "split3_test_score                                           0.36   \n",
       "split3_train_score                                          0.37   \n",
       "split4_test_score                                           0.38   \n",
       "split4_train_score                                          0.36   \n",
       "std_fit_time                                             5.1e-06   \n",
       "std_score_time                                           3.1e-06   \n",
       "std_test_score                                             0.011   \n",
       "std_train_score                                           0.0029   \n",
       "\n",
       "                                                             7   \\\n",
       "mean_fit_time                                           0.00049   \n",
       "mean_score_time                                         0.00021   \n",
       "mean_test_score                                            0.37   \n",
       "mean_train_score                                           0.37   \n",
       "param_C                                                    0.01   \n",
       "param_gamma                                                0.01   \n",
       "param_kernel                                                rbf   \n",
       "params              {'kernel': 'rbf', 'C': 0.01, 'gamma': 0.01}   \n",
       "rank_test_score                                              27   \n",
       "split0_test_score                                          0.38   \n",
       "split0_train_score                                         0.36   \n",
       "split1_test_score                                          0.35   \n",
       "split1_train_score                                         0.37   \n",
       "split2_test_score                                          0.36   \n",
       "split2_train_score                                         0.37   \n",
       "split3_test_score                                          0.36   \n",
       "split3_train_score                                         0.37   \n",
       "split4_test_score                                          0.38   \n",
       "split4_train_score                                         0.36   \n",
       "std_fit_time                                            5.9e-06   \n",
       "std_score_time                                          1.6e-06   \n",
       "std_test_score                                            0.011   \n",
       "std_train_score                                          0.0029   \n",
       "\n",
       "                                                            8   \\\n",
       "mean_fit_time                                           0.0005   \n",
       "mean_score_time                                        0.00021   \n",
       "mean_test_score                                           0.37   \n",
       "mean_train_score                                          0.37   \n",
       "param_C                                                   0.01   \n",
       "param_gamma                                                0.1   \n",
       "param_kernel                                               rbf   \n",
       "params              {'kernel': 'rbf', 'C': 0.01, 'gamma': 0.1}   \n",
       "rank_test_score                                             27   \n",
       "split0_test_score                                         0.38   \n",
       "split0_train_score                                        0.36   \n",
       "split1_test_score                                         0.35   \n",
       "split1_train_score                                        0.37   \n",
       "split2_test_score                                         0.36   \n",
       "split2_train_score                                        0.37   \n",
       "split3_test_score                                         0.36   \n",
       "split3_train_score                                        0.37   \n",
       "split4_test_score                                         0.38   \n",
       "split4_train_score                                        0.36   \n",
       "std_fit_time                                           6.4e-06   \n",
       "std_score_time                                         6.8e-06   \n",
       "std_test_score                                           0.011   \n",
       "std_train_score                                         0.0029   \n",
       "\n",
       "                                                          9   \\\n",
       "mean_fit_time                                        0.00054   \n",
       "mean_score_time                                      0.00023   \n",
       "mean_test_score                                         0.37   \n",
       "mean_train_score                                        0.37   \n",
       "param_C                                                 0.01   \n",
       "param_gamma                                                1   \n",
       "param_kernel                                             rbf   \n",
       "params              {'kernel': 'rbf', 'C': 0.01, 'gamma': 1}   \n",
       "rank_test_score                                           27   \n",
       "split0_test_score                                       0.38   \n",
       "split0_train_score                                      0.36   \n",
       "split1_test_score                                       0.35   \n",
       "split1_train_score                                      0.37   \n",
       "split2_test_score                                       0.36   \n",
       "split2_train_score                                      0.37   \n",
       "split3_test_score                                       0.36   \n",
       "split3_train_score                                      0.37   \n",
       "split4_test_score                                       0.38   \n",
       "split4_train_score                                      0.36   \n",
       "std_fit_time                                         3.3e-05   \n",
       "std_score_time                                       1.5e-05   \n",
       "std_test_score                                         0.011   \n",
       "std_train_score                                       0.0029   \n",
       "\n",
       "                                                           10  \\\n",
       "mean_fit_time                                         0.00055   \n",
       "mean_score_time                                       0.00023   \n",
       "mean_test_score                                          0.37   \n",
       "mean_train_score                                         0.37   \n",
       "param_C                                                  0.01   \n",
       "param_gamma                                                10   \n",
       "param_kernel                                              rbf   \n",
       "params              {'kernel': 'rbf', 'C': 0.01, 'gamma': 10}   \n",
       "rank_test_score                                            27   \n",
       "split0_test_score                                        0.38   \n",
       "split0_train_score                                       0.36   \n",
       "split1_test_score                                        0.35   \n",
       "split1_train_score                                       0.37   \n",
       "split2_test_score                                        0.36   \n",
       "split2_train_score                                       0.37   \n",
       "split3_test_score                                        0.36   \n",
       "split3_train_score                                       0.37   \n",
       "split4_test_score                                        0.38   \n",
       "split4_train_score                                       0.36   \n",
       "std_fit_time                                          2.5e-05   \n",
       "std_score_time                                        2.7e-05   \n",
       "std_test_score                                          0.011   \n",
       "std_train_score                                        0.0029   \n",
       "\n",
       "                                                            11  \\\n",
       "mean_fit_time                                          0.00058   \n",
       "mean_score_time                                        0.00021   \n",
       "mean_test_score                                           0.37   \n",
       "mean_train_score                                          0.37   \n",
       "param_C                                                   0.01   \n",
       "param_gamma                                                100   \n",
       "param_kernel                                               rbf   \n",
       "params              {'kernel': 'rbf', 'C': 0.01, 'gamma': 100}   \n",
       "rank_test_score                                             27   \n",
       "split0_test_score                                         0.38   \n",
       "split0_train_score                                        0.36   \n",
       "split1_test_score                                         0.35   \n",
       "split1_train_score                                        0.37   \n",
       "split2_test_score                                         0.36   \n",
       "split2_train_score                                        0.37   \n",
       "split3_test_score                                         0.36   \n",
       "split3_train_score                                        0.37   \n",
       "split4_test_score                                         0.38   \n",
       "split4_train_score                                        0.36   \n",
       "std_fit_time                                           1.1e-05   \n",
       "std_score_time                                         4.3e-06   \n",
       "std_test_score                                           0.011   \n",
       "std_train_score                                         0.0029   \n",
       "\n",
       "                                                             12  \\\n",
       "mean_fit_time                                           0.00049   \n",
       "mean_score_time                                          0.0002   \n",
       "mean_test_score                                            0.37   \n",
       "mean_train_score                                           0.37   \n",
       "param_C                                                     0.1   \n",
       "param_gamma                                               0.001   \n",
       "param_kernel                                                rbf   \n",
       "params              {'kernel': 'rbf', 'C': 0.1, 'gamma': 0.001}   \n",
       "rank_test_score                                              27   \n",
       "split0_test_score                                          0.38   \n",
       "split0_train_score                                         0.36   \n",
       "split1_test_score                                          0.35   \n",
       "split1_train_score                                         0.37   \n",
       "split2_test_score                                          0.36   \n",
       "split2_train_score                                         0.37   \n",
       "split3_test_score                                          0.36   \n",
       "split3_train_score                                         0.37   \n",
       "split4_test_score                                          0.38   \n",
       "split4_train_score                                         0.36   \n",
       "std_fit_time                                              6e-06   \n",
       "std_score_time                                          2.1e-06   \n",
       "std_test_score                                            0.011   \n",
       "std_train_score                                          0.0029   \n",
       "\n",
       "                                                            13  \\\n",
       "mean_fit_time                                          0.00053   \n",
       "mean_score_time                                        0.00021   \n",
       "mean_test_score                                            0.7   \n",
       "mean_train_score                                           0.7   \n",
       "param_C                                                    0.1   \n",
       "param_gamma                                               0.01   \n",
       "param_kernel                                               rbf   \n",
       "params              {'kernel': 'rbf', 'C': 0.1, 'gamma': 0.01}   \n",
       "rank_test_score                                             22   \n",
       "split0_test_score                                         0.71   \n",
       "split0_train_score                                        0.69   \n",
       "split1_test_score                                          0.7   \n",
       "split1_train_score                                         0.7   \n",
       "split2_test_score                                         0.68   \n",
       "split2_train_score                                         0.7   \n",
       "split3_test_score                                         0.68   \n",
       "split3_train_score                                         0.7   \n",
       "split4_test_score                                         0.71   \n",
       "split4_train_score                                        0.69   \n",
       "std_fit_time                                           4.5e-05   \n",
       "std_score_time                                         8.9e-06   \n",
       "std_test_score                                           0.013   \n",
       "std_train_score                                         0.0033   \n",
       "\n",
       "                                                           14  \\\n",
       "mean_fit_time                                         0.00043   \n",
       "mean_score_time                                        0.0002   \n",
       "mean_test_score                                          0.92   \n",
       "mean_train_score                                         0.92   \n",
       "param_C                                                   0.1   \n",
       "param_gamma                                               0.1   \n",
       "param_kernel                                              rbf   \n",
       "params              {'kernel': 'rbf', 'C': 0.1, 'gamma': 0.1}   \n",
       "rank_test_score                                            17   \n",
       "split0_test_score                                        0.92   \n",
       "split0_train_score                                       0.94   \n",
       "split1_test_score                                        0.91   \n",
       "split1_train_score                                       0.93   \n",
       "split2_test_score                                           1   \n",
       "split2_train_score                                        0.9   \n",
       "split3_test_score                                        0.86   \n",
       "split3_train_score                                       0.89   \n",
       "split4_test_score                                         0.9   \n",
       "split4_train_score                                       0.93   \n",
       "std_fit_time                                          3.3e-06   \n",
       "std_score_time                                        1.9e-06   \n",
       "std_test_score                                          0.044   \n",
       "std_train_score                                         0.021   \n",
       "\n",
       "                                                         15  \\\n",
       "mean_fit_time                                       0.00049   \n",
       "mean_score_time                                      0.0002   \n",
       "mean_test_score                                        0.96   \n",
       "mean_train_score                                       0.96   \n",
       "param_C                                                 0.1   \n",
       "param_gamma                                               1   \n",
       "param_kernel                                            rbf   \n",
       "params              {'kernel': 'rbf', 'C': 0.1, 'gamma': 1}   \n",
       "rank_test_score                                           9   \n",
       "split0_test_score                                         1   \n",
       "split0_train_score                                     0.95   \n",
       "split1_test_score                                      0.91   \n",
       "split1_train_score                                     0.98   \n",
       "split2_test_score                                         1   \n",
       "split2_train_score                                     0.94   \n",
       "split3_test_score                                      0.91   \n",
       "split3_train_score                                     0.97   \n",
       "split4_test_score                                      0.95   \n",
       "split4_train_score                                     0.96   \n",
       "std_fit_time                                        4.8e-05   \n",
       "std_score_time                                      4.3e-06   \n",
       "std_test_score                                         0.04   \n",
       "std_train_score                                       0.011   \n",
       "\n",
       "                                                          16  \\\n",
       "mean_fit_time                                        0.00055   \n",
       "mean_score_time                                      0.00022   \n",
       "mean_test_score                                         0.37   \n",
       "mean_train_score                                        0.38   \n",
       "param_C                                                  0.1   \n",
       "param_gamma                                               10   \n",
       "param_kernel                                             rbf   \n",
       "params              {'kernel': 'rbf', 'C': 0.1, 'gamma': 10}   \n",
       "rank_test_score                                           27   \n",
       "split0_test_score                                       0.38   \n",
       "split0_train_score                                      0.38   \n",
       "split1_test_score                                       0.35   \n",
       "split1_train_score                                      0.39   \n",
       "split2_test_score                                       0.36   \n",
       "split2_train_score                                      0.37   \n",
       "split3_test_score                                       0.36   \n",
       "split3_train_score                                      0.41   \n",
       "split4_test_score                                       0.38   \n",
       "split4_train_score                                      0.36   \n",
       "std_fit_time                                         8.5e-06   \n",
       "std_score_time                                       5.5e-06   \n",
       "std_test_score                                         0.011   \n",
       "std_train_score                                        0.018   \n",
       "\n",
       "                                                           17  \\\n",
       "mean_fit_time                                         0.00058   \n",
       "mean_score_time                                       0.00021   \n",
       "mean_test_score                                          0.37   \n",
       "mean_train_score                                         0.37   \n",
       "param_C                                                   0.1   \n",
       "param_gamma                                               100   \n",
       "param_kernel                                              rbf   \n",
       "params              {'kernel': 'rbf', 'C': 0.1, 'gamma': 100}   \n",
       "rank_test_score                                            27   \n",
       "split0_test_score                                        0.38   \n",
       "split0_train_score                                       0.36   \n",
       "split1_test_score                                        0.35   \n",
       "split1_train_score                                       0.37   \n",
       "split2_test_score                                        0.36   \n",
       "split2_train_score                                       0.37   \n",
       "split3_test_score                                        0.36   \n",
       "split3_train_score                                       0.37   \n",
       "split4_test_score                                        0.38   \n",
       "split4_train_score                                       0.36   \n",
       "std_fit_time                                          1.2e-05   \n",
       "std_score_time                                        1.5e-06   \n",
       "std_test_score                                          0.011   \n",
       "std_train_score                                        0.0029   \n",
       "\n",
       "                                                           18  \\\n",
       "mean_fit_time                                         0.00053   \n",
       "mean_score_time                                       0.00021   \n",
       "mean_test_score                                           0.7   \n",
       "mean_train_score                                          0.7   \n",
       "param_C                                                     1   \n",
       "param_gamma                                             0.001   \n",
       "param_kernel                                              rbf   \n",
       "params              {'kernel': 'rbf', 'C': 1, 'gamma': 0.001}   \n",
       "rank_test_score                                            22   \n",
       "split0_test_score                                        0.71   \n",
       "split0_train_score                                       0.69   \n",
       "split1_test_score                                         0.7   \n",
       "split1_train_score                                        0.7   \n",
       "split2_test_score                                        0.68   \n",
       "split2_train_score                                        0.7   \n",
       "split3_test_score                                        0.68   \n",
       "split3_train_score                                        0.7   \n",
       "split4_test_score                                        0.71   \n",
       "split4_train_score                                       0.69   \n",
       "std_fit_time                                          3.8e-05   \n",
       "std_score_time                                          6e-06   \n",
       "std_test_score                                          0.013   \n",
       "std_train_score                                        0.0033   \n",
       "\n",
       "                                                          19  \\\n",
       "mean_fit_time                                        0.00039   \n",
       "mean_score_time                                       0.0002   \n",
       "mean_test_score                                         0.93   \n",
       "mean_train_score                                        0.94   \n",
       "param_C                                                    1   \n",
       "param_gamma                                             0.01   \n",
       "param_kernel                                             rbf   \n",
       "params              {'kernel': 'rbf', 'C': 1, 'gamma': 0.01}   \n",
       "rank_test_score                                           15   \n",
       "split0_test_score                                       0.88   \n",
       "split0_train_score                                      0.94   \n",
       "split1_test_score                                       0.91   \n",
       "split1_train_score                                      0.94   \n",
       "split2_test_score                                          1   \n",
       "split2_train_score                                      0.92   \n",
       "split3_test_score                                       0.91   \n",
       "split3_train_score                                      0.93   \n",
       "split4_test_score                                       0.95   \n",
       "split4_train_score                                      0.93   \n",
       "std_fit_time                                         4.1e-06   \n",
       "std_score_time                                       2.5e-06   \n",
       "std_test_score                                         0.043   \n",
       "std_train_score                                       0.0079   \n",
       "\n",
       "                                                         20  \\\n",
       "mean_fit_time                                       0.00032   \n",
       "mean_score_time                                      0.0002   \n",
       "mean_test_score                                        0.96   \n",
       "mean_train_score                                       0.98   \n",
       "param_C                                                   1   \n",
       "param_gamma                                             0.1   \n",
       "param_kernel                                            rbf   \n",
       "params              {'kernel': 'rbf', 'C': 1, 'gamma': 0.1}   \n",
       "rank_test_score                                           3   \n",
       "split0_test_score                                         1   \n",
       "split0_train_score                                     0.97   \n",
       "split1_test_score                                      0.96   \n",
       "split1_train_score                                     0.99   \n",
       "split2_test_score                                         1   \n",
       "split2_train_score                                     0.97   \n",
       "split3_test_score                                      0.91   \n",
       "split3_train_score                                     0.99   \n",
       "split4_test_score                                      0.95   \n",
       "split4_train_score                                     0.98   \n",
       "std_fit_time                                        6.7e-06   \n",
       "std_score_time                                      3.8e-05   \n",
       "std_test_score                                        0.034   \n",
       "std_train_score                                        0.01   \n",
       "\n",
       "                                                       21  \\\n",
       "mean_fit_time                                     0.00039   \n",
       "mean_score_time                                   0.00018   \n",
       "mean_test_score                                      0.95   \n",
       "mean_train_score                                     0.98   \n",
       "param_C                                                 1   \n",
       "param_gamma                                             1   \n",
       "param_kernel                                          rbf   \n",
       "params              {'kernel': 'rbf', 'C': 1, 'gamma': 1}   \n",
       "rank_test_score                                        11   \n",
       "split0_test_score                                    0.92   \n",
       "split0_train_score                                   0.99   \n",
       "split1_test_score                                    0.96   \n",
       "split1_train_score                                   0.98   \n",
       "split2_test_score                                       1   \n",
       "split2_train_score                                   0.98   \n",
       "split3_test_score                                    0.91   \n",
       "split3_train_score                                      1   \n",
       "split4_test_score                                    0.95   \n",
       "split4_train_score                                   0.98   \n",
       "std_fit_time                                      2.3e-05   \n",
       "std_score_time                                    1.9e-06   \n",
       "std_test_score                                      0.032   \n",
       "std_train_score                                    0.0089   \n",
       "\n",
       "                                                        22  \\\n",
       "mean_fit_time                                      0.00064   \n",
       "mean_score_time                                    0.00021   \n",
       "mean_test_score                                       0.92   \n",
       "mean_train_score                                         1   \n",
       "param_C                                                  1   \n",
       "param_gamma                                             10   \n",
       "param_kernel                                           rbf   \n",
       "params              {'kernel': 'rbf', 'C': 1, 'gamma': 10}   \n",
       "rank_test_score                                         17   \n",
       "split0_test_score                                     0.88   \n",
       "split0_train_score                                       1   \n",
       "split1_test_score                                     0.96   \n",
       "split1_train_score                                       1   \n",
       "split2_test_score                                        1   \n",
       "split2_train_score                                       1   \n",
       "split3_test_score                                     0.82   \n",
       "split3_train_score                                       1   \n",
       "split4_test_score                                     0.95   \n",
       "split4_train_score                                       1   \n",
       "std_fit_time                                       2.5e-05   \n",
       "std_score_time                                     2.8e-06   \n",
       "std_test_score                                       0.065   \n",
       "std_train_score                                          0   \n",
       "\n",
       "                                                         23  \\\n",
       "mean_fit_time                                        0.0007   \n",
       "mean_score_time                                     0.00024   \n",
       "mean_test_score                                        0.51   \n",
       "mean_train_score                                          1   \n",
       "param_C                                                   1   \n",
       "param_gamma                                             100   \n",
       "param_kernel                                            rbf   \n",
       "params              {'kernel': 'rbf', 'C': 1, 'gamma': 100}   \n",
       "rank_test_score                                          26   \n",
       "split0_test_score                                       0.5   \n",
       "split0_train_score                                        1   \n",
       "split1_test_score                                      0.43   \n",
       "split1_train_score                                        1   \n",
       "split2_test_score                                      0.55   \n",
       "split2_train_score                                        1   \n",
       "split3_test_score                                       0.5   \n",
       "split3_train_score                                        1   \n",
       "split4_test_score                                      0.57   \n",
       "split4_train_score                                        1   \n",
       "std_fit_time                                        6.8e-05   \n",
       "std_score_time                                      4.2e-05   \n",
       "std_test_score                                        0.046   \n",
       "std_train_score                                           0   \n",
       "\n",
       "                                                            24  \\\n",
       "mean_fit_time                                           0.0004   \n",
       "mean_score_time                                         0.0002   \n",
       "mean_test_score                                           0.93   \n",
       "mean_train_score                                          0.94   \n",
       "param_C                                                     10   \n",
       "param_gamma                                              0.001   \n",
       "param_kernel                                               rbf   \n",
       "params              {'kernel': 'rbf', 'C': 10, 'gamma': 0.001}   \n",
       "rank_test_score                                             15   \n",
       "split0_test_score                                         0.88   \n",
       "split0_train_score                                        0.94   \n",
       "split1_test_score                                         0.91   \n",
       "split1_train_score                                        0.94   \n",
       "split2_test_score                                            1   \n",
       "split2_train_score                                        0.92   \n",
       "split3_test_score                                         0.91   \n",
       "split3_train_score                                        0.93   \n",
       "split4_test_score                                         0.95   \n",
       "split4_train_score                                        0.93   \n",
       "std_fit_time                                           5.9e-06   \n",
       "std_score_time                                         1.9e-06   \n",
       "std_test_score                                           0.043   \n",
       "std_train_score                                         0.0079   \n",
       "\n",
       "                                                           25  \\\n",
       "mean_fit_time                                         0.00032   \n",
       "mean_score_time                                       0.00019   \n",
       "mean_test_score                                          0.96   \n",
       "mean_train_score                                         0.98   \n",
       "param_C                                                    10   \n",
       "param_gamma                                              0.01   \n",
       "param_kernel                                              rbf   \n",
       "params              {'kernel': 'rbf', 'C': 10, 'gamma': 0.01}   \n",
       "rank_test_score                                             3   \n",
       "split0_test_score                                           1   \n",
       "split0_train_score                                       0.98   \n",
       "split1_test_score                                        0.96   \n",
       "split1_train_score                                       0.98   \n",
       "split2_test_score                                           1   \n",
       "split2_train_score                                       0.97   \n",
       "split3_test_score                                        0.91   \n",
       "split3_train_score                                       0.99   \n",
       "split4_test_score                                        0.95   \n",
       "split4_train_score                                       0.98   \n",
       "std_fit_time                                          6.1e-06   \n",
       "std_score_time                                        1.1e-06   \n",
       "std_test_score                                          0.034   \n",
       "std_train_score                                         0.007   \n",
       "\n",
       "                                                          26  \\\n",
       "mean_fit_time                                         0.0003   \n",
       "mean_score_time                                      0.00018   \n",
       "mean_test_score                                         0.96   \n",
       "mean_train_score                                        0.99   \n",
       "param_C                                                   10   \n",
       "param_gamma                                              0.1   \n",
       "param_kernel                                             rbf   \n",
       "params              {'kernel': 'rbf', 'C': 10, 'gamma': 0.1}   \n",
       "rank_test_score                                            3   \n",
       "split0_test_score                                       0.96   \n",
       "split0_train_score                                      0.99   \n",
       "split1_test_score                                       0.96   \n",
       "split1_train_score                                      0.98   \n",
       "split2_test_score                                          1   \n",
       "split2_train_score                                      0.98   \n",
       "split3_test_score                                       0.95   \n",
       "split3_train_score                                      0.99   \n",
       "split4_test_score                                       0.95   \n",
       "split4_train_score                                         1   \n",
       "std_fit_time                                           9e-06   \n",
       "std_score_time                                         1e-06   \n",
       "std_test_score                                         0.018   \n",
       "std_train_score                                       0.0084   \n",
       "\n",
       "                                                        27  \\\n",
       "mean_fit_time                                      0.00041   \n",
       "mean_score_time                                    0.00019   \n",
       "mean_test_score                                       0.94   \n",
       "mean_train_score                                      0.99   \n",
       "param_C                                                 10   \n",
       "param_gamma                                              1   \n",
       "param_kernel                                           rbf   \n",
       "params              {'kernel': 'rbf', 'C': 10, 'gamma': 1}   \n",
       "rank_test_score                                         14   \n",
       "split0_test_score                                     0.92   \n",
       "split0_train_score                                    0.99   \n",
       "split1_test_score                                     0.96   \n",
       "split1_train_score                                    0.98   \n",
       "split2_test_score                                        1   \n",
       "split2_train_score                                    0.98   \n",
       "split3_test_score                                     0.86   \n",
       "split3_train_score                                       1   \n",
       "split4_test_score                                     0.95   \n",
       "split4_train_score                                    0.99   \n",
       "std_fit_time                                       6.9e-05   \n",
       "std_score_time                                     1.6e-05   \n",
       "std_test_score                                       0.045   \n",
       "std_train_score                                     0.0084   \n",
       "\n",
       "                                                         28  \\\n",
       "mean_fit_time                                       0.00064   \n",
       "mean_score_time                                     0.00021   \n",
       "mean_test_score                                        0.92   \n",
       "mean_train_score                                          1   \n",
       "param_C                                                  10   \n",
       "param_gamma                                              10   \n",
       "param_kernel                                            rbf   \n",
       "params              {'kernel': 'rbf', 'C': 10, 'gamma': 10}   \n",
       "rank_test_score                                          17   \n",
       "split0_test_score                                      0.88   \n",
       "split0_train_score                                        1   \n",
       "split1_test_score                                      0.96   \n",
       "split1_train_score                                        1   \n",
       "split2_test_score                                         1   \n",
       "split2_train_score                                        1   \n",
       "split3_test_score                                      0.82   \n",
       "split3_train_score                                        1   \n",
       "split4_test_score                                      0.95   \n",
       "split4_train_score                                        1   \n",
       "std_fit_time                                        1.7e-05   \n",
       "std_score_time                                      1.1e-05   \n",
       "std_test_score                                        0.065   \n",
       "std_train_score                                           0   \n",
       "\n",
       "                                                          29  \\\n",
       "mean_fit_time                                        0.00072   \n",
       "mean_score_time                                      0.00023   \n",
       "mean_test_score                                         0.57   \n",
       "mean_train_score                                           1   \n",
       "param_C                                                   10   \n",
       "param_gamma                                              100   \n",
       "param_kernel                                             rbf   \n",
       "params              {'kernel': 'rbf', 'C': 10, 'gamma': 100}   \n",
       "rank_test_score                                           24   \n",
       "split0_test_score                                       0.54   \n",
       "split0_train_score                                         1   \n",
       "split1_test_score                                       0.52   \n",
       "split1_train_score                                         1   \n",
       "split2_test_score                                       0.59   \n",
       "split2_train_score                                         1   \n",
       "split3_test_score                                       0.59   \n",
       "split3_train_score                                         1   \n",
       "split4_test_score                                       0.62   \n",
       "split4_train_score                                         1   \n",
       "std_fit_time                                         8.1e-05   \n",
       "std_score_time                                       4.2e-05   \n",
       "std_test_score                                         0.036   \n",
       "std_train_score                                            0   \n",
       "\n",
       "                                                             30  \\\n",
       "mean_fit_time                                           0.00031   \n",
       "mean_score_time                                         0.00018   \n",
       "mean_test_score                                            0.96   \n",
       "mean_train_score                                           0.98   \n",
       "param_C                                                     100   \n",
       "param_gamma                                               0.001   \n",
       "param_kernel                                                rbf   \n",
       "params              {'kernel': 'rbf', 'C': 100, 'gamma': 0.001}   \n",
       "rank_test_score                                               3   \n",
       "split0_test_score                                             1   \n",
       "split0_train_score                                         0.98   \n",
       "split1_test_score                                          0.96   \n",
       "split1_train_score                                         0.98   \n",
       "split2_test_score                                             1   \n",
       "split2_train_score                                         0.97   \n",
       "split3_test_score                                          0.91   \n",
       "split3_train_score                                         0.99   \n",
       "split4_test_score                                          0.95   \n",
       "split4_train_score                                         0.98   \n",
       "std_fit_time                                            1.7e-05   \n",
       "std_score_time                                          1.1e-05   \n",
       "std_test_score                                            0.034   \n",
       "std_train_score                                           0.007   \n",
       "\n",
       "                                                            31  \\\n",
       "mean_fit_time                                          0.00028   \n",
       "mean_score_time                                         0.0002   \n",
       "mean_test_score                                           0.97   \n",
       "mean_train_score                                          0.98   \n",
       "param_C                                                    100   \n",
       "param_gamma                                               0.01   \n",
       "param_kernel                                               rbf   \n",
       "params              {'kernel': 'rbf', 'C': 100, 'gamma': 0.01}   \n",
       "rank_test_score                                              1   \n",
       "split0_test_score                                            1   \n",
       "split0_train_score                                        0.99   \n",
       "split1_test_score                                         0.96   \n",
       "split1_train_score                                        0.98   \n",
       "split2_test_score                                            1   \n",
       "split2_train_score                                        0.98   \n",
       "split3_test_score                                         0.95   \n",
       "split3_train_score                                        0.99   \n",
       "split4_test_score                                         0.95   \n",
       "split4_train_score                                        0.99   \n",
       "std_fit_time                                           1.6e-05   \n",
       "std_score_time                                         5.2e-05   \n",
       "std_test_score                                           0.022   \n",
       "std_train_score                                         0.0055   \n",
       "\n",
       "                                                           32  \\\n",
       "mean_fit_time                                         0.00029   \n",
       "mean_score_time                                       0.00017   \n",
       "mean_test_score                                          0.96   \n",
       "mean_train_score                                         0.99   \n",
       "param_C                                                   100   \n",
       "param_gamma                                               0.1   \n",
       "param_kernel                                              rbf   \n",
       "params              {'kernel': 'rbf', 'C': 100, 'gamma': 0.1}   \n",
       "rank_test_score                                             9   \n",
       "split0_test_score                                        0.96   \n",
       "split0_train_score                                       0.99   \n",
       "split1_test_score                                           1   \n",
       "split1_train_score                                       0.98   \n",
       "split2_test_score                                           1   \n",
       "split2_train_score                                       0.98   \n",
       "split3_test_score                                        0.86   \n",
       "split3_train_score                                          1   \n",
       "split4_test_score                                        0.95   \n",
       "split4_train_score                                          1   \n",
       "std_fit_time                                          1.2e-05   \n",
       "std_score_time                                        2.9e-06   \n",
       "std_test_score                                           0.05   \n",
       "std_train_score                                          0.01   \n",
       "\n",
       "                                                         33  \\\n",
       "mean_fit_time                                        0.0004   \n",
       "mean_score_time                                     0.00019   \n",
       "mean_test_score                                        0.95   \n",
       "mean_train_score                                          1   \n",
       "param_C                                                 100   \n",
       "param_gamma                                               1   \n",
       "param_kernel                                            rbf   \n",
       "params              {'kernel': 'rbf', 'C': 100, 'gamma': 1}   \n",
       "rank_test_score                                          11   \n",
       "split0_test_score                                      0.92   \n",
       "split0_train_score                                        1   \n",
       "split1_test_score                                         1   \n",
       "split1_train_score                                        1   \n",
       "split2_test_score                                         1   \n",
       "split2_train_score                                        1   \n",
       "split3_test_score                                      0.86   \n",
       "split3_train_score                                        1   \n",
       "split4_test_score                                      0.95   \n",
       "split4_train_score                                        1   \n",
       "std_fit_time                                        1.5e-05   \n",
       "std_score_time                                      1.3e-05   \n",
       "std_test_score                                        0.052   \n",
       "std_train_score                                           0   \n",
       "\n",
       "                                                          34  \\\n",
       "mean_fit_time                                        0.00067   \n",
       "mean_score_time                                      0.00021   \n",
       "mean_test_score                                         0.92   \n",
       "mean_train_score                                           1   \n",
       "param_C                                                  100   \n",
       "param_gamma                                               10   \n",
       "param_kernel                                             rbf   \n",
       "params              {'kernel': 'rbf', 'C': 100, 'gamma': 10}   \n",
       "rank_test_score                                           17   \n",
       "split0_test_score                                       0.88   \n",
       "split0_train_score                                         1   \n",
       "split1_test_score                                       0.96   \n",
       "split1_train_score                                         1   \n",
       "split2_test_score                                          1   \n",
       "split2_train_score                                         1   \n",
       "split3_test_score                                       0.82   \n",
       "split3_train_score                                         1   \n",
       "split4_test_score                                       0.95   \n",
       "split4_train_score                                         1   \n",
       "std_fit_time                                           4e-05   \n",
       "std_score_time                                       3.1e-06   \n",
       "std_test_score                                         0.065   \n",
       "std_train_score                                            0   \n",
       "\n",
       "                                                           35  \\\n",
       "mean_fit_time                                          0.0007   \n",
       "mean_score_time                                       0.00023   \n",
       "mean_test_score                                          0.57   \n",
       "mean_train_score                                            1   \n",
       "param_C                                                   100   \n",
       "param_gamma                                               100   \n",
       "param_kernel                                              rbf   \n",
       "params              {'kernel': 'rbf', 'C': 100, 'gamma': 100}   \n",
       "rank_test_score                                            24   \n",
       "split0_test_score                                        0.54   \n",
       "split0_train_score                                          1   \n",
       "split1_test_score                                        0.52   \n",
       "split1_train_score                                          1   \n",
       "split2_test_score                                        0.59   \n",
       "split2_train_score                                          1   \n",
       "split3_test_score                                        0.59   \n",
       "split3_train_score                                          1   \n",
       "split4_test_score                                        0.62   \n",
       "split4_train_score                                          1   \n",
       "std_fit_time                                          3.6e-05   \n",
       "std_score_time                                        1.4e-05   \n",
       "std_test_score                                          0.036   \n",
       "std_train_score                                             0   \n",
       "\n",
       "                                                  36  \\\n",
       "mean_fit_time                                0.00039   \n",
       "mean_score_time                              0.00018   \n",
       "mean_test_score                                 0.37   \n",
       "mean_train_score                                0.37   \n",
       "param_C                                        0.001   \n",
       "param_gamma                                      NaN   \n",
       "param_kernel                                  linear   \n",
       "params              {'kernel': 'linear', 'C': 0.001}   \n",
       "rank_test_score                                   27   \n",
       "split0_test_score                               0.38   \n",
       "split0_train_score                              0.36   \n",
       "split1_test_score                               0.35   \n",
       "split1_train_score                              0.37   \n",
       "split2_test_score                               0.36   \n",
       "split2_train_score                              0.37   \n",
       "split3_test_score                               0.36   \n",
       "split3_train_score                              0.37   \n",
       "split4_test_score                               0.38   \n",
       "split4_train_score                              0.36   \n",
       "std_fit_time                                 6.5e-06   \n",
       "std_score_time                               1.2e-06   \n",
       "std_test_score                                 0.011   \n",
       "std_train_score                               0.0029   \n",
       "\n",
       "                                                 37  \\\n",
       "mean_fit_time                               0.00035   \n",
       "mean_score_time                             0.00018   \n",
       "mean_test_score                                0.85   \n",
       "mean_train_score                               0.86   \n",
       "param_C                                        0.01   \n",
       "param_gamma                                     NaN   \n",
       "param_kernel                                 linear   \n",
       "params              {'kernel': 'linear', 'C': 0.01}   \n",
       "rank_test_score                                  21   \n",
       "split0_test_score                              0.92   \n",
       "split0_train_score                             0.89   \n",
       "split1_test_score                              0.83   \n",
       "split1_train_score                             0.89   \n",
       "split2_test_score                              0.82   \n",
       "split2_train_score                             0.87   \n",
       "split3_test_score                              0.77   \n",
       "split3_train_score                             0.76   \n",
       "split4_test_score                               0.9   \n",
       "split4_train_score                             0.88   \n",
       "std_fit_time                                1.1e-06   \n",
       "std_score_time                              7.6e-07   \n",
       "std_test_score                                0.055   \n",
       "std_train_score                                0.05   \n",
       "\n",
       "                                                38  \\\n",
       "mean_fit_time                              0.00031   \n",
       "mean_score_time                            0.00017   \n",
       "mean_test_score                               0.95   \n",
       "mean_train_score                              0.97   \n",
       "param_C                                        0.1   \n",
       "param_gamma                                    NaN   \n",
       "param_kernel                                linear   \n",
       "params              {'kernel': 'linear', 'C': 0.1}   \n",
       "rank_test_score                                 11   \n",
       "split0_test_score                             0.96   \n",
       "split0_train_score                            0.97   \n",
       "split1_test_score                             0.91   \n",
       "split1_train_score                            0.98   \n",
       "split2_test_score                                1   \n",
       "split2_train_score                            0.94   \n",
       "split3_test_score                             0.91   \n",
       "split3_train_score                            0.98   \n",
       "split4_test_score                             0.95   \n",
       "split4_train_score                            0.97   \n",
       "std_fit_time                               1.3e-05   \n",
       "std_score_time                             1.6e-06   \n",
       "std_test_score                               0.033   \n",
       "std_train_score                              0.012   \n",
       "\n",
       "                                              39  \\\n",
       "mean_fit_time                            0.00029   \n",
       "mean_score_time                          0.00017   \n",
       "mean_test_score                             0.97   \n",
       "mean_train_score                            0.98   \n",
       "param_C                                        1   \n",
       "param_gamma                                  NaN   \n",
       "param_kernel                              linear   \n",
       "params              {'kernel': 'linear', 'C': 1}   \n",
       "rank_test_score                                1   \n",
       "split0_test_score                              1   \n",
       "split0_train_score                          0.99   \n",
       "split1_test_score                           0.96   \n",
       "split1_train_score                          0.98   \n",
       "split2_test_score                              1   \n",
       "split2_train_score                          0.98   \n",
       "split3_test_score                           0.95   \n",
       "split3_train_score                          0.99   \n",
       "split4_test_score                           0.95   \n",
       "split4_train_score                          0.99   \n",
       "std_fit_time                             8.7e-06   \n",
       "std_score_time                             1e-06   \n",
       "std_test_score                             0.022   \n",
       "std_train_score                           0.0055   \n",
       "\n",
       "                                               40  \\\n",
       "mean_fit_time                              0.0003   \n",
       "mean_score_time                           0.00017   \n",
       "mean_test_score                              0.96   \n",
       "mean_train_score                             0.99   \n",
       "param_C                                        10   \n",
       "param_gamma                                   NaN   \n",
       "param_kernel                               linear   \n",
       "params              {'kernel': 'linear', 'C': 10}   \n",
       "rank_test_score                                 3   \n",
       "split0_test_score                            0.96   \n",
       "split0_train_score                           0.99   \n",
       "split1_test_score                               1   \n",
       "split1_train_score                           0.99   \n",
       "split2_test_score                               1   \n",
       "split2_train_score                           0.98   \n",
       "split3_test_score                            0.91   \n",
       "split3_train_score                           0.99   \n",
       "split4_test_score                            0.95   \n",
       "split4_train_score                              1   \n",
       "std_fit_time                              2.4e-05   \n",
       "std_score_time                            4.6e-07   \n",
       "std_test_score                              0.034   \n",
       "std_train_score                             0.007   \n",
       "\n",
       "                                                41  \n",
       "mean_fit_time                               0.0003  \n",
       "mean_score_time                            0.00017  \n",
       "mean_test_score                               0.96  \n",
       "mean_train_score                              0.99  \n",
       "param_C                                        100  \n",
       "param_gamma                                    NaN  \n",
       "param_kernel                                linear  \n",
       "params              {'kernel': 'linear', 'C': 100}  \n",
       "rank_test_score                                  3  \n",
       "split0_test_score                             0.96  \n",
       "split0_train_score                            0.99  \n",
       "split1_test_score                                1  \n",
       "split1_train_score                            0.99  \n",
       "split2_test_score                                1  \n",
       "split2_train_score                            0.99  \n",
       "split3_test_score                             0.91  \n",
       "split3_train_score                               1  \n",
       "split4_test_score                             0.95  \n",
       "split4_train_score                               1  \n",
       "std_fit_time                               2.3e-05  \n",
       "std_score_time                             5.3e-06  \n",
       "std_test_score                               0.034  \n",
       "std_train_score                             0.0055  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "results = pd.DataFrame(grid_search.cv_results_)\n",
    "# 좀 더 나은 출력을 위해 결과를 전치시킵니다\n",
    "display(results.T)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### 그리드 서치에 다양한 교차 검증 적용"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### 중첩 교차 검증"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "교차 검증 점수:  [0.967 1.    0.967 0.967 1.   ]\n",
      "교차 검증 평균 점수:  0.9800000000000001\n",
      "{'C': [0.001, 0.01, 0.1, 1, 10, 100], 'gamma': [0.001, 0.01, 0.1, 1, 10, 100]}\n"
     ]
    }
   ],
   "source": [
    "param_grid = {'C': [0.001, 0.01, 0.1, 1, 10, 100],\n",
    "              'gamma': [0.001, 0.01, 0.1, 1, 10, 100]}\n",
    "scores = cross_val_score(GridSearchCV(SVC(), param_grid, cv=5),\n",
    "                         iris.data, iris.target, cv=5)\n",
    "print(\"교차 검증 점수: \", scores)\n",
    "print(\"교차 검증 평균 점수: \", scores.mean())\n",
    "print(param_grid)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [],
   "source": [
    "def nested_cv(X, y, inner_cv, outer_cv, Classifier, parameter_grid):\n",
    "    outer_scores = []\n",
    "    # outer_cv의 분할을 순회하는 for 루프\n",
    "    # (split 메소드는 훈련과 테스트 세트에 해당하는 인덱스를 리턴합니다)\n",
    "    for training_samples, test_samples in outer_cv.split(X, y):\n",
    "        # 최적의 매개변수를 찾습니다\n",
    "        best_parms = {}\n",
    "        best_score = -np.inf\n",
    "        # 매개변수 그리드를 순회합니다\n",
    "        for parameters in parameter_grid:\n",
    "            # 안쪽 교차 검증의 점수를 기록합니다\n",
    "            cv_scores = []\n",
    "            # inner_cv의 분할을 순회하는 for 루프\n",
    "            for inner_train, inner_test in inner_cv.split(\n",
    "                    X[training_samples], y[training_samples]):\n",
    "                # 훈련 데이터와 주어진 매개변수로 분류기를 만듭니다\n",
    "                clf = Classifier(**parameters)\n",
    "                clf.fit(X[inner_train], y[inner_train])\n",
    "                # 검증 세트로 평가합니다\n",
    "                score = clf.score(X[inner_test], y[inner_test])\n",
    "                cv_scores.append(score)\n",
    "            # 안쪽 교차 검증의 평균 점수를 계산합니다\n",
    "            mean_score = np.mean(cv_scores)\n",
    "            if mean_score > best_score:\n",
    "                # 점수가 더 높은면 매개변수와 함께 기록합니다\n",
    "                best_score = mean_score\n",
    "                best_params = parameters\n",
    "        # 바깥쪽 훈련 데이터 전체를 사용해 분류기를 만듭니다\n",
    "        clf = Classifier(**best_params)\n",
    "        clf.fit(X[training_samples], y[training_samples])\n",
    "        # 테스트 세트를 사용해 평가합니다\n",
    "        outer_scores.append(clf.score(X[test_samples], y[test_samples]))\n",
    "    return np.array(outer_scores)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {
    "hide_input": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "교차 검증 점수: [0.967 1.    0.967 0.967 1.   ]\n"
     ]
    }
   ],
   "source": [
    "from sklearn.model_selection import ParameterGrid, StratifiedKFold\n",
    "scores = nested_cv(iris.data, iris.target, StratifiedKFold(5),\n",
    "                   StratifiedKFold(5), SVC, ParameterGrid(param_grid))\n",
    "print(\"교차 검증 점수: {}\".format(scores))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true
   },
   "source": [
    "#### 교차 검증과 그리드 서치 병렬화"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 평가 지표 및 측정\n",
    "#### 최종 목표를 기억하라"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### 이진 분류의 평가 지표\n",
    "##### 에러의 종류\n",
    "##### 불균형 데이터셋"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.datasets import load_digits\n",
    "\n",
    "digits = load_digits()\n",
    "y = digits.target == 9\n",
    "\n",
    "X_train, X_test, y_train, y_test = train_test_split(\n",
    "    digits.data, y, random_state=0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "예측된 레이블의 고유값: [False]\n",
      "테스트 점수: 0.90\n"
     ]
    }
   ],
   "source": [
    "from sklearn.dummy import DummyClassifier\n",
    "dummy_majority = DummyClassifier(strategy='most_frequent').fit(X_train, y_train)\n",
    "pred_most_frequent = dummy_majority.predict(X_test)\n",
    "print(\"예측된 레이블의 고유값: {}\".format(np.unique(pred_most_frequent)))\n",
    "print(\"테스트 점수: {:.2f}\".format(dummy_majority.score(X_test, y_test)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "테스트 점수: 0.92\n"
     ]
    }
   ],
   "source": [
    "from sklearn.tree import DecisionTreeClassifier\n",
    "tree = DecisionTreeClassifier(max_depth=2).fit(X_train, y_train)\n",
    "pred_tree = tree.predict(X_test)\n",
    "print(\"테스트 점수: {:.2f}\".format(tree.score(X_test, y_test)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "dummy 점수: 0.80\n",
      "logreg 점수: 0.98\n"
     ]
    }
   ],
   "source": [
    "from sklearn.linear_model import LogisticRegression\n",
    "\n",
    "dummy = DummyClassifier().fit(X_train, y_train)\n",
    "pred_dummy = dummy.predict(X_test)\n",
    "print(\"dummy 점수: {:.2f}\".format(dummy.score(X_test, y_test)))\n",
    "\n",
    "logreg = LogisticRegression(solver='liblinear', C=0.1).fit(X_train, y_train)\n",
    "pred_logreg = logreg.predict(X_test)\n",
    "print(\"logreg 점수: {:.2f}\".format(logreg.score(X_test, y_test)))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### 오차 행렬(Confusion matrices)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "오차 행렬:\n",
      "[[401   2]\n",
      " [  8  39]]\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import confusion_matrix\n",
    "\n",
    "confusion = confusion_matrix(y_test, pred_logreg)\n",
    "print(\"오차 행렬:\\n{}\".format(confusion))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/pdf": "JVBERi0xLjQKJazcIKu6CjEgMCBvYmoKPDwgL1BhZ2VzIDIgMCBSIC9UeXBlIC9DYXRhbG9nID4+CmVuZG9iago4IDAgb2JqCjw8IC9FeHRHU3RhdGUgNCAwIFIgL0ZvbnQgMyAwIFIgL1BhdHRlcm4gNSAwIFIKL1Byb2NTZXQgWyAvUERGIC9UZXh0IC9JbWFnZUIgL0ltYWdlQyAvSW1hZ2VJIF0gL1NoYWRpbmcgNiAwIFIKL1hPYmplY3QgNyAwIFIgPj4KZW5kb2JqCjEwIDAgb2JqCjw8IC9Bbm5vdHMgWyBdIC9Db250ZW50cyA5IDAgUgovR3JvdXAgPDwgL0NTIC9EZXZpY2VSR0IgL1MgL1RyYW5zcGFyZW5jeSAvVHlwZSAvR3JvdXAgPj4KL01lZGlhQm94IFsgMCAwIDU3NS4xNzUgNDgwLjk1MTg3NSBdIC9QYXJlbnQgMiAwIFIgL1Jlc291cmNlcyA4IDAgUgovVHlwZSAvUGFnZSA+PgplbmRvYmoKOSAwIG9iago8PCAvRmlsdGVyIC9GbGF0ZURlY29kZSAvTGVuZ3RoIDExIDAgUiA+PgpzdHJlYW0KeJzdVU1vGkEMvc+v8K3tIWbs8Xwdi9Ii9VCJFqmHqIcICAEBUZKm+fv1LtnV7rJs6delB2DweN6bZ49tgo0ZvSVYPYKFjX6e4Qq+6u8CCCYwulx+X8+XnyZjmD8aq/ad8dEjRa/rbb2WZDF7Srrcqlvr760xe6MMemCioCtjiBJa3XMeXSy9FDUISsu2rW0SLXIFVx1uGZXjRnXwQcdKaVQLpoaaglx3DHPG0OGubYSp4jZjvfOzuddvCxdWESlGjE43iTGxh/nOjGcwek/AFmY3ZXRmC/P6VYY3MNtobDD6nLN4G1K1d9h6N1Ngi5ahOFQumNEmay0lRwWloo8+Xu+fduPrh6f95O7b7Xp+MV8vWFyGyzszPUJwGTMpX6KQ8gACeXL9CCFpMhWBNRhp6A5eFfcipIBW75CyJReHELJ/UTE1ZZ6MeI1qJy+1bTAvwqTUTs5KzUAC9FmxJkA4sv8t8eIxBUWwgaz8kvjqTZMI2vyinkjQt4298mMVMHIBHaUyEP80AiHFP4sAeT4RARcIvetEoGHsi0BEffzeabX9TPt/X5gnM3NuYTYzcw/HjVokoIA4wZTgYQlfYA9cttmodqeZOLTb9uBwQsid4q5srT5uPpsp/B1eo7wVDheNhDqTpmGsiRk+HOZga3a0xfRNr96hpJjHU23XP9XU99yp2HBtQJzGtaWmw0yk1kRcVXUSyzqhKEi5YNMXG1MsEIraEEtFcYAWR8ufvZY7c9F1GZ3U/qnXW2LWh8hH6Nzv7RIydZFdrpyn5gdRPe8WCmVuZHN0cmVhbQplbmRvYmoKMTEgMCBvYmoKNTkxCmVuZG9iagoxNiAwIG9iago8PCAvRmlsdGVyIC9GbGF0ZURlY29kZSAvTGVuZ3RoIDE3ID4+CnN0cmVhbQp4nDMyMlEwgMMUQy4AGlcC6AplbmRzdHJlYW0KZW5kb2JqCjE3IDAgb2JqCjw8IC9CQm94IFsgLTQ1IC0yODcgMTAwMCA4MzggXSAvRmlsdGVyIC9GbGF0ZURlY29kZSAvTGVuZ3RoIDE3NgovU3VidHlwZSAvRm9ybSAvVHlwZSAvWE9iamVjdCA+PgpzdHJlYW0KeJxNjzsOwzAMQ3efQhcoIFk/+zwFOjn3X0s5SdOJz6ApSk3ZSN3paEM3rIJ0Bkj2PzItUomb8K/IzKFGzgJWejf3JHOhYKSCKXRAB5winVUSAhqTfAYU6WQoV9qwUTo5mjQxcyRpJBzzIrT2UvyvrNdcjboCLz2vua9CI9oSc4+WQzatTX3sLeA+9LgvwXpH63Lh2thV7/SPLrvigvWQmblpbXqZXu5Dp/v5Ajj/SGIKZW5kc3RyZWFtCmVuZG9iagoxOCAwIG9iago8PCAvQkJveCBbIC00NSAtMjg3IDEwMDAgODM4IF0gL0ZpbHRlciAvRmxhdGVEZWNvZGUgL0xlbmd0aCAyMDIKL1N1YnR5cGUgL0Zvcm0gL1R5cGUgL1hPYmplY3QgPj4Kc3RyZWFtCnicTZBLDsMgDET3nMIXiIS/wHkqdUXuv61tSMlqHprRGLuYQeMOd5FRk2aSmTihyosEwyUbm5gtiUcHIQSJDHX4FMUKwgqqCiIIOtyVFo4Tt2jUFlRBZQD7L5TYlTwjowGrd/okjk6prul4L0e+h7bFou4wxouB+lZd+ileHTDL3nSW3gkUh+8c76CZxBUj5pMONZ+03EM69Gk59G+5EOOc2GzhTCSRvB29CFlzCr+I5Ok5tGvCvnyTO8lWCx641pY7Ncv3B2nSWzAKZW5kc3RyZWFtCmVuZG9iagoxOSAwIG9iago8PCAvRmlsdGVyIC9GbGF0ZURlY29kZSAvTGVuZ3RoIDIxOCA+PgpzdHJlYW0KeJw1kEuOBCEMQ/ecIhcoCecHnKdGvZv7b8cJ0wtkCxLnhdgmUzzl8SVhW5aF/GDo0ba/wxbaufFmqvhySUt5RyAl1uw2/1e7L3TK2qrAjO5x54Mb5OGlsbwm1pjSd2hcp1jy2BFsqlYPgnS8sMnDuHOjNllDlgoQkpS0ZmICaiHNU3KBaIowVwNXxl0BrK+lFHfd79pvf0BmfQB2toOZRHirWZPRKRnBeBAKpDnVjMMfYVp9AzqU8g5jDY1tr1cnZ1e7rW73gterd0C5GmnMK4Qv1Ds+f3XPUckKZW5kc3RyZWFtCmVuZG9iagoyMCAwIG9iago8PCAvRmlsdGVyIC9GbGF0ZURlY29kZSAvTGVuZ3RoIDEzMyA+PgpzdHJlYW0KeJxNjjsOQyEMBHtOsReIhI0/cB6iV5H7t1mIkFJ5bEs7672hQlTxao7miSS/peyTu+FzKCSwithAaIekI0whg3tUzKLaENmhFojBGb+gWW7kOrQlq2jKH+3ETfWSJX0VqgYfBq0C78IsocGDfnd2S0hjR3biR3mh+7S2jtt+lucL2ncqsAplbmRzdHJlYW0KZW5kb2JqCjIxIDAgb2JqCjw8IC9GaWx0ZXIgL0ZsYXRlRGVjb2RlIC9MZW5ndGggMjcwID4+CnN0cmVhbQp4nC1Ru43FMAzrMwUXeID1t+fJ4VV3+7dHKSkCMqZMSXRsw0IEPhYIVRTxR64jiCr8XdQjDzaLSga24L5IMhK/l6yNrA0x4klI0UOUJboCpQV1H1fdZ/C+LGWYS7J2w8ORuxU/Sl9FiCDXg2FOZVh/q+B0irXgHOy5406XTWS/PvWVrXBCo6/7gqVRVZj3+L4cxs62HXqMqkKzqEjwNHqxJk4Sc+sln9HevIbYIepm8RIyEzpbfzGuFjZduHWtmav/XImH6PZO40mVuXnl5OJntaIMvFXlM8h50Ga3ZlPLxPgGxhdLk7dPdlcrZDL3VrLT1TjPCbtnFISbp+tsTYXZiQbdAu/z39f3H8iAZiIKZW5kc3RyZWFtCmVuZG9iagoyMiAwIG9iago8PCAvRmlsdGVyIC9GbGF0ZURlY29kZSAvTGVuZ3RoIDUxMSA+PgpzdHJlYW0KeJw1lEuS5DAIRPc+BReoCPERks7jiV7V3H87D9SzcJEWAhJI19wuQ+aRT4RMN1me8kcf26uP/j4WBnLRwyWcOrd8zOV9VLlzZIfsykDw+2B1hHwv2JJDVFXSMOXPALjk6velXDqVaixZRFBr4lpLqP0+xh1XsThiU2y3eR8nI8DLtcXn4Knj3EIC31MmBr5Zx0G9VZ6Q4xJuzYjzgBvxQSmNdW2e9oTYqkw8hBktl4UQJQt9LzqzEMP5RbRVyA6MDjHQj1Ftuwco4QkDDfE1mhCenXAqLyf0GYolDywMRC9Bq3Psa/16QFu5y2QrG7fyjmBW04zGUjKT8VwL8xn3xIip5bDgjMqmcEoYaHl0ymEHu8bAz+QWKSYPa643ju0eK0sf11QjgGQB3+cAFjoZpIVYqSRP64WRrcHKI7GbYQcKqFgleNGuMYRlRRLro2hHIVaQV5oGjbIlgtXIY2NZFDVXjzRG5WWsdqjjPdZs6sF6E9HVWBPZxIGbVbbZWqUz+MwV10YrupCicL33J9ljdTbeArnGLhHhXfm71Mha6O7KJYhw7yWVGGFM/pZHyWWVrelcpAwwLFvUwbl7VUVqYzSXjTBHf6iGdMvqapaF6hMmVqkxeU6HoLpiw0vQOMF2xY8EYOvyKQGxn0/Pobh8vOR+v////wPv8/MPPTLGiAplbmRzdHJlYW0KZW5kb2JqCjIzIDAgb2JqCjw8IC9GaWx0ZXIgL0ZsYXRlRGVjb2RlIC9MZW5ndGggMTA1ID4+CnN0cmVhbQp4nE3NQQ7AIAgEwLuv2A80ERCo/2l6sv+/FjRNvZjJLoieggomHKJQ63BmXFQiIXE8CY5oFFGZ5ShNbRNXD+Xqr1xd7a88EL84bVotf5Dm0C5xNR6uFtHEGtdN2mnKYG3fHOV+AXn0KHgKZW5kc3RyZWFtCmVuZG9iagoyNCAwIG9iago8PCAvQkJveCBbIC00NSAtMjg3IDEwMDAgODM4IF0gL0ZpbHRlciAvRmxhdGVEZWNvZGUgL0xlbmd0aCA0MTUKL1N1YnR5cGUgL0Zvcm0gL1R5cGUgL1hPYmplY3QgPj4Kc3RyZWFtCnicTVI5biQxDMz7FfzAAuItvWeAjez/p1tFteENBuK0yLqoJ1dKLJXvh5VHS5wjvpbEbrE88nmiU0y3RG7RHRKOM5s32qKG/mOiC2eFnMSFe8lW4KR0iLXz+DwWa/6jFbeK33Rr+oyr7YHT1S/BrmHsNQLA/yqqK3FdxWss4DOKhL6SxH90Zxg+Yzq3CbBKa8ArqIZ01dAB+qID6OlV1AkxbSClXutrwChoHKGiQXbQ8Z1hBkRhJkRlSpeHuZGZOVIIk73KmPVI5Rag/GcbMH4u9vfT7lN9TfVH3VAWbP9X9l7v/W8VwYo4v9XFCe5ydj4VdIYitYayhcyWT5QmSWY4ysLkviHCZ8HNCnGmZtixp5Rzxuk88FYOkk6d1Hhy7+d+wXKnQ/Wd0XNRtHtQtfLl4VbIrOGjRL1fbeo6arm/eYu239RYebEXuviQMGOHL4YoVnt2bciePIb94QbMenADJbrXaFN0QjURsoTqNdf44YkbOGRFx+xgBneGqRCFKRGVuV0eJklmJkslzPpqY/pUy21Q/c9+Ps/ff4bErQQKZW5kc3RyZWFtCmVuZG9iagoyNSAwIG9iago8PCAvRmlsdGVyIC9GbGF0ZURlY29kZSAvTGVuZ3RoIDY2OSA+PgpzdHJlYW0KeJw9VFuOJDEI+69TcIGWwjPJeXq1XzP3/10bSitNT+iqAMY2ncdliW/5REhGyvaUP/rYVbGU38fVGTge2xFPhCnfx3dKmPh1SZNYV4qPw0L2lvASvAnUU1U+Txe1K5FHNNecO/oNolvCG6a7cwzpeIP/lmwTYnua2OEbdvVF2ECja1Dp4RuAcyshavclHIPn97Ha8ySqb1jfYo6t6ip6TlfV7W8fzd2dlZjQQ8HTYFNwQbQKbETfZ8/TESbkDU7MnOGAVQCcRUERuzRlbAsKiQKENiw+Jk42AWzQ/8rxbV1iGYUBHYwct2JhbCREN/IKRBj3gGhqd/EdMEDpwt3AR5FboJS5+75kxwXZoZIac44MjLK6QrbqV/Jai61SrAjyykA4iKhxxzGpAAKgL+R6wB9lLRByCkQDaYFazsMTk+01T3JumPubY5pdRe/tqnry7aNbu7PWIFGYaLCRbKLVF73+n0ffCXmDE2vGy4GSEXwoJFmiMsObLWsmTU8zSxMM18ZssD8GnXl4jlJcql9IeToyZHycxr/ysdNYsHhKo/CuKN1HE+GCg0EM03+jEICt4KoqOORRLTbQ0AnYDbgIyO2w8mUFVPTb5Cs3BvXQsqOUH7ibZdCcVsH9O374PsVthNiYjOuN34UmmwHYwb4mWkChUs6JCsV1B16uMp7Vbp45weUylGxqg5XaeltPRHZbaf7ckKk9vvGJCJA3HPpMDtnf7d7bVaPu2yfO7c65vIEkVn+QJZyRvV1nkPOcUeh7zgZncdS46509sDskI3zIcVT3Gk/fpvAHG+FNahxrknPtl/V0axn4Q0pVeI5MHTk9eFvGxHQta+pplXuTUVLHAVGLjgiPNggtNZZxrAlN5FiPNtVrs+/z9x+Z5BPlCmVuZHN0cmVhbQplbmRvYmoKMjYgMCBvYmoKPDwgL0JCb3ggWyAtNDUgLTI4NyAxMDAwIDgzOCBdIC9GaWx0ZXIgL0ZsYXRlRGVjb2RlIC9MZW5ndGggMzkyCi9TdWJ0eXBlIC9Gb3JtIC9UeXBlIC9YT2JqZWN0ID4+CnN0cmVhbQp4nE1SOY4DMQzr5xX6wAI6fbwnwFbJ/9sl5Uk2xYwJS5QoytfcQ34sXF7XND3w2XAuJUL8H93h8iEeCU5qNXpesVVsucRYYokztpgNeVyhWxD0mTInsp3H47IZBOb4NGWHWG3c4+eWghK++0hVXAMUmiFaq5g9ktekj13S1TxOeZxoiyJEMU0GoqmoMIsdIJ/oDDInpxvmX+jHoPmEv+AZNF0ldGL4Qm+iZ6NSWpO23gh5uQYmwAionNCSFT0K5gpGSpjf5z1kSSzkISPGbE6UdYSWIs9NAt9bBeZMa/brINSl9ezI/FKyI0wqyEypWhJYTC3t5cA/pddDhsF6rHAEOY5FjoSj6D8SkTgnInaQwXdm2Idj5KCKoQerWundh0+CnQ36qIQP6WgzX62Wm6T6PvW92xioj4zw3RzfVMAqfj80z9PHvTnssfExuqy12WgOX0nhAUK9YSbOY+0tJ+wbcpjx4dAVVqFLrErfTh866f3MD49eH210n2q5Dap/7+dx/f4B37ujxQplbmRzdHJlYW0KZW5kb2JqCjI3IDAgb2JqCjw8IC9GaWx0ZXIgL0ZsYXRlRGVjb2RlIC9MZW5ndGggNTU5ID4+CnN0cmVhbQp4nC2TO3IkMQxDc5+CF3CV+JPU5/HWRt77p/vAnmBGavEHgmDftGX12HeldV47ue2Pf92yby/7p0sf+9W52+6xve1J22E/X+5hvcwzrdy827L1fI7FY7GWxbaI1vHzFfv9fsIIyKDs5Tn533wfN2rkQyJX8nJuhRMxfqiQZH0tRAR1q4tsWHpbAgALL7r94nMosYm6lkTVVH4mL1G3rZZbkDcvFgCA5ACweeEXRTOJZyomFQPoeHSuidNJU3Q8L0svQsr39OUgUMseWGDEV38QUDDD7cADPtP74ZmqtY5mURTVccQml6Y8nTU9w35fkXBgjWcAbXHvG1OpLDRtxuRM6FDSofu48kdw01yYk6YcUK1TI4i5JQM4jKvIdlZ+qN7XgaLxF6iC0avx9gUS0JC18ergt/b0gR9YpKaSVSMYzHOTXwaUNDEQF0Ld675DfkQYcdLaHQAjkxq75Uao6iXp4zsZIHml3Igzp8TncxPt34zsI+S3lbqStG4qWFUDRcrqHh/S9kE1zvlII9A6mstpHPhYN3QlS7F7LCIcstOb8wytOiX4el9KL4CFjN2jmsWNITsEbgh30btGNfi1SCB3o0evPfRKTxAuX6ElXmfd9Rl5tSySDpbSCMdCtYRH7wfuRKGDfFaXvU/p9dFIIFDsen4UnVJz39GrduXVbbKRumkv5JHaNn83pyZLglxZa9VskbaNvEdbfMZaEsdni/39osNqTeOdz8/X3/+WTNnZCmVuZHN0cmVhbQplbmRvYmoKMjggMCBvYmoKPDwgL0JCb3ggWyAtNDUgLTI4NyAxMDAwIDgzOCBdIC9GaWx0ZXIgL0ZsYXRlRGVjb2RlIC9MZW5ndGggNjQ5Ci9TdWJ0eXBlIC9Gb3JtIC9UeXBlIC9YT2JqZWN0ID4+CnN0cmVhbQp4nE2US24tMQhE570Kb6AlG4w/67nSGyX7n6YOdvQyMg12URTQT3gvfc3y/WDF8PL1jB1/rLmqrLnaH8vMbvS/BcrX84vHvV18CBij8drKyzlred3L55kxyxvK4Ku808tsXl5d/jxDCG+rUUYMHs0ymu40GwqGaLzNxdEJ9qqE1/g8/dfn284tj/770ttFsxUH30a9KU3EIWFNIT2xKjJuirRNvlra6qVZHj7xyxj6Vlhw3G8VPwjNdgGx9ZU52uhEdK+JPTya+HnmaTDTlykH9HlPRZyq9nqomhsIcd6gDSiIBSrynTwISmYEhgmSH240YdHTLaByWySofXuFEUqyVZoLeZSZ8bnKWmUM8a8Kd+E25B5CbyZPJdNKls3hECE26kNY0+mqKPJUi/o+HjXidKrdNz56oni3RHXrN4/XkZltO0QQMonZnBA1NU60Ofy45ZOiiup5dptO8lrDUUB7LRL/dVQm49vlEYM3LDm9sWArlq+8sH7HzDo4VaEqw6JSbuRc5hvUAAV1QEWvkwcFyYygMEHhpIbidFf6i/jtB3OxStfHt3gw51JNrzs8UDB42gdzI48wu8m7ZadqobnxPQpL6ZrWCEaXHsSUpeENKeIw1WJ6P31D3sYq04FKVukBbh+621ITRZTNpW+v9As+OceVjKpB0wW+aSb9bJEq93r6alMzyWLkUuimtlM/EV0QCIPqo+UA+xq5aDsLMl3rfhaqj3Ub2iWRCyScEvhzZUMVD1RkSSmfic43Hpr9MbL5tMPuYAsNIfBoPXPU1dAT0YZhfGFM3zJMTK+lpmC5xTmVhfPIhNXjRELlz5SJFp1/Jr84y4aOZnc5Yspjt213AD7Pvx+YRg8BCmVuZHN0cmVhbQplbmRvYmoKMjkgMCBvYmoKPDwgL0JCb3ggWyAtNDUgLTI4NyAxMDAwIDgzOCBdIC9GaWx0ZXIgL0ZsYXRlRGVjb2RlIC9MZW5ndGggMzE1Ci9TdWJ0eXBlIC9Gb3JtIC9UeXBlIC9YT2JqZWN0ID4+CnN0cmVhbQp4nE1SS24FMQjb5xRcoFL4k/NU6mp6/20NqTRvhRPHxjCzJJzyCP0urzPouYgP0LAvGpYtKSKgMPFBz7K4d77rVt9Tv5dD0yg0bq0+F5hAl2jdMVRFBfYEkxtqZ5pqfhnbo8E7vu/9gNUknz7Bm9yEPA85Ersp2TmdAB2toBEjS57uFu2WDMaK0nGjh2oHqjSDaXUmS1bS7I5CWsjBTLa1feFh3aldkMPw3rw72txA5+gILJhQj4ORvs0izoB/Egtco90KYqler94QjMiGoQXYvENJoSEeK9Zu/wZ2mrGNGBsnfBrvMYtnGQjJTh201zGrgcbvWgSncx38dPSY2vGbwQJKbrWY+j2fvtH7Ezyr0Eu5f6BGojPDB2juffWswFK+WB2Chqkzc76grbF3jP6iEXxon/XzB88rhJkKZW5kc3RyZWFtCmVuZG9iagozMCAwIG9iago8PCAvRmlsdGVyIC9GbGF0ZURlY29kZSAvTGVuZ3RoIDU2ID4+CnN0cmVhbQp4nDOyNFcwULC0VDA1M1UwtLRQMAeyUwy5DA3NwUK5XEA+SCyHCyYLZJlD1OfAleVwpQEAGRQOnQplbmRzdHJlYW0KZW5kb2JqCjE0IDAgb2JqCjw8IC9CYXNlRm9udCAvTmFudW1CYXJ1bkdvdGhpYyAvQ2hhclByb2NzIDE1IDAgUgovRW5jb2RpbmcgPDwKL0RpZmZlcmVuY2VzIFsgMzIgL2NpZDEgMzkgL2NpZDggNDggL2NpZDE3IC9jaWQxOCAvY2lkMTkgL2NpZDIwIC9jaWQyMSA1NiAvY2lkMjUgL2NpZDI2Cl0KL1R5cGUgL0VuY29kaW5nID4+Ci9GaXJzdENoYXIgMCAvRm9udEJCb3ggWyAtNDUgLTI4NyAxMDAwIDgzOCBdIC9Gb250RGVzY3JpcHRvciAxMyAwIFIKL0ZvbnRNYXRyaXggWyAwLjAwMSAwIDAgMC4wMDEgMCAwIF0gL0xhc3RDaGFyIDI1NSAvTmFtZSAvTmFudW1CYXJ1bkdvdGhpYwovU3VidHlwZSAvVHlwZTMgL1R5cGUgL0ZvbnQgL1dpZHRocyAxMiAwIFIgPj4KZW5kb2JqCjEzIDAgb2JqCjw8IC9Bc2NlbnQgODUwIC9DYXBIZWlnaHQgMCAvRGVzY2VudCAtMjk5IC9GbGFncyAzMgovRm9udEJCb3ggWyAtNDUgLTI4NyAxMDAwIDgzOCBdIC9Gb250TmFtZSAvTmFudW1CYXJ1bkdvdGhpYyAvSXRhbGljQW5nbGUgMAovTWF4V2lkdGggMTA1MiAvU3RlbVYgMCAvVHlwZSAvRm9udERlc2NyaXB0b3IgL1hIZWlnaHQgMCA+PgplbmRvYmoKMTIgMCBvYmoKWyA4OTIgMjI0IDg5MiAyMjQgMjI0IDIyNCAyMjQgMjI0IDIyNCA4OTIgODkyIDIyNCAyMjQgODkyIDIyNCAyMjQgMjI0IDIyNAoyMjQgMjI0IDIyNCAyMjQgMjI0IDIyNCAyMjQgMjI0IDIyNCAyMjQgMjI0IDIyNCAyMjQgMjI0IDIyNCA0NDYgMjk3IDU4MyA1ODMKODkyIDg5MiAyOTcgNDQ2IDQ0NiA0NDYgNTgzIDI5NyA1ODMgMjk3IDQ0NiA1ODMgNTgzIDU4MyA1ODMgNTgzIDU4MyA1ODMgNTgzCjU4MyA1ODMgMjk3IDI5NyA0NDYgNTgzIDQ0NiA2NjkgMTA1MiA2NDQgNjI3IDYzOSA3MjEgNTk2IDU1NCA3MTAgNzE4IDI0Nwo0MTAgNjI2IDUyOSA4ODQgNzEwIDc1MiA1ODYgNzUyIDYxMCA1OTIgNjIxIDY5NiA2MzUgOTYxIDYxNyA2MTEgNTk0IDQ0NiA5NjEKNDQ2IDQzNCA0NDYgMjk3IDU2MCA1ODggNDkwIDU4OCA1NTkgMzQwIDU4OCA1OTIgMjQ0IDMwMSA1MzAgMjQ0IDg5MiA1OTIgNTc3CjU4OCA1ODggMzgzIDQ3NSAzNTcgNTkyIDUzMCA3ODggNTI4IDUzMCA0NzMgNDQ2IDQ0NiA0NDYgNjY5IDIyNCA4OTIgMjI0IDIyNAoyMjQgMjI0IDg5MiA4OTIgODkyIDIyNCA4OTIgMjI0IDg5MiA4OTIgMjI0IDIyNCAyMjQgMjI0IDMwMyAzMDMgNDQ2IDQ0NiA4OTIKODkyIDg5MiA4OTIgODkyIDIyNCA4OTIgODkyIDIyNCAyMjQgMjI0IDI5NyA4OTIgODkyIDg5MiA4OTIgODkyIDIyNCA0NDYgODkyCjg5MiA4OTIgODkyIDg5MiA4OTIgODkyIDIyNCA0NDYgODkyIDQ0NiA0NDYgODkyIDIyNCA4OTIgNDQ2IDQ0NiA0NDYgODkyIDg5Mgo4OTIgODkyIDg5MiA4OTIgMjI0IDIyNCAyMjQgMjI0IDIyNCAyMjQgODkyIDIyNCAyMjQgMjI0IDIyNCAyMjQgMjI0IDIyNCAyMjQKMjI0IDg5MiAyMjQgMjI0IDIyNCAyMjQgMjI0IDIyNCA4OTIgODkyIDIyNCAyMjQgMjI0IDIyNCAyMjQgODkyIDg5MiAyMjQgMjI0CjIyNCAyMjQgMjI0IDIyNCA4OTIgMjI0IDIyNCAyMjQgMjI0IDIyNCAyMjQgMjI0IDIyNCAyMjQgODkyIDIyNCAyMjQgMjI0IDIyNAoyMjQgMjI0IDg5MiA4OTIgMjI0IDIyNCAyMjQgMjI0IDIyNCA4OTIgMjI0IF0KZW5kb2JqCjE1IDAgb2JqCjw8IC9jaWQxIDE2IDAgUiAvY2lkMTcgMTkgMCBSIC9jaWQxOCAyMCAwIFIgL2NpZDE5IDIxIDAgUiAvY2lkMjAgMjIgMCBSCi9jaWQyMSAyMyAwIFIgL2NpZDI1IDI1IDAgUiAvY2lkMjYgMjcgMCBSIC9jaWQ4IDMwIDAgUiA+PgplbmRvYmoKMyAwIG9iago8PCAvRjEgMTQgMCBSID4+CmVuZG9iago0IDAgb2JqCjw8IC9BMSA8PCAvQ0EgMCAvVHlwZSAvRXh0R1N0YXRlIC9jYSAxID4+Ci9BMiA8PCAvQ0EgMSAvVHlwZSAvRXh0R1N0YXRlIC9jYSAxID4+ID4+CmVuZG9iago1IDAgb2JqCjw8ID4+CmVuZG9iago2IDAgb2JqCjw8ID4+CmVuZG9iago3IDAgb2JqCjw8IC9OYW51bUJhcnVuR290aGljLWNpZDE1MTMgMTcgMCBSIC9OYW51bUJhcnVuR290aGljLWNpZDE1MjkgMTggMCBSCi9OYW51bUJhcnVuR290aGljLWNpZDI0MzkgMjQgMCBSIC9OYW51bUJhcnVuR290aGljLWNpZDI1MjAgMjYgMCBSCi9OYW51bUJhcnVuR290aGljLWNpZDI2ODcgMjggMCBSIC9OYW51bUJhcnVuR290aGljLWNpZDI5NTkgMjkgMCBSID4+CmVuZG9iagoyIDAgb2JqCjw8IC9Db3VudCAxIC9LaWRzIFsgMTAgMCBSIF0gL1R5cGUgL1BhZ2VzID4+CmVuZG9iagozMSAwIG9iago8PCAvQ3JlYXRpb25EYXRlIChEOjIwMTgxMTI2MTcyNzAxKzA5JzAwJykKL0NyZWF0b3IgKG1hdHBsb3RsaWIgMy4wLjAsIGh0dHA6Ly9tYXRwbG90bGliLm9yZykKL1Byb2R1Y2VyIChtYXRwbG90bGliIHBkZiBiYWNrZW5kIDMuMC4wKSA+PgplbmRvYmoKeHJlZgowIDMyCjAwMDAwMDAwMDAgNjU1MzUgZiAKMDAwMDAwMDAxNiAwMDAwMCBuIAowMDAwMDA5MzcxIDAwMDAwIG4gCjAwMDAwMDg5NzkgMDAwMDAgbiAKMDAwMDAwOTAxMSAwMDAwMCBuIAowMDAwMDA5MTEwIDAwMDAwIG4gCjAwMDAwMDkxMzEgMDAwMDAgbiAKMDAwMDAwOTE1MiAwMDAwMCBuIAowMDAwMDAwMDY1IDAwMDAwIG4gCjAwMDAwMDAzOTYgMDAwMDAgbiAKMDAwMDAwMDIwOCAwMDAwMCBuIAowMDAwMDAxMDYyIDAwMDAwIG4gCjAwMDAwMDc3ODggMDAwMDAgbiAKMDAwMDAwNzU4NSAwMDAwMCBuIAowMDAwMDA3MjEwIDAwMDAwIG4gCjAwMDAwMDg4MzMgMDAwMDAgbiAKMDAwMDAwMTA4MiAwMDAwMCBuIAowMDAwMDAxMTcxIDAwMDAwIG4gCjAwMDAwMDE0NzggMDAwMDAgbiAKMDAwMDAwMTgxMSAwMDAwMCBuIAowMDAwMDAyMTAyIDAwMDAwIG4gCjAwMDAwMDIzMDggMDAwMDAgbiAKMDAwMDAwMjY1MSAwMDAwMCBuIAowMDAwMDAzMjM1IDAwMDAwIG4gCjAwMDAwMDM0MTMgMDAwMDAgbiAKMDAwMDAwMzk1OSAwMDAwMCBuIAowMDAwMDA0NzAxIDAwMDAwIG4gCjAwMDAwMDUyMjQgMDAwMDAgbiAKMDAwMDAwNTg1NiAwMDAwMCBuIAowMDAwMDA2NjM2IDAwMDAwIG4gCjAwMDAwMDcwODIgMDAwMDAgbiAKMDAwMDAwOTQzMSAwMDAwMCBuIAp0cmFpbGVyCjw8IC9JbmZvIDMxIDAgUiAvUm9vdCAxIDAgUiAvU2l6ZSAzMiA+PgpzdGFydHhyZWYKOTU4NQolJUVPRgo=\n",
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "mglearn.plots.plot_confusion_matrix_illustration()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/pdf": "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\n",
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "mglearn.plots.plot_binary_confusion_matrix()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "빈도 기반 더미 모델:\n",
      "[[403   0]\n",
      " [ 47   0]]\n",
      "\n",
      "무작위 더미 모델:\n",
      "[[363  40]\n",
      " [ 44   3]]\n",
      "\n",
      "결정 트리:\n",
      "[[390  13]\n",
      " [ 24  23]]\n",
      "\n",
      "로지스틱 회귀\n",
      "[[401   2]\n",
      " [  8  39]]\n"
     ]
    }
   ],
   "source": [
    "print(\"빈도 기반 더미 모델:\")\n",
    "print(confusion_matrix(y_test, pred_most_frequent))\n",
    "print(\"\\n무작위 더미 모델:\")\n",
    "print(confusion_matrix(y_test, pred_dummy))\n",
    "print(\"\\n결정 트리:\")\n",
    "print(confusion_matrix(y_test, pred_tree))\n",
    "print(\"\\n로지스틱 회귀\")\n",
    "print(confusion_matrix(y_test, pred_logreg))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "###### 정확도와의 관계\n",
    "\\begin{equation}\n",
    "\\text{Accuracy} = \\frac{\\text{TP} + \\text{TN}}{\\text{TP} + \\text{TN} + \\text{FP} + \\text{FN}}\n",
    "\\end{equation}"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "###### 정밀도, 재현율, f-점수\n",
    "\\begin{equation}\n",
    "\\text{정밀도} = \\frac{\\text{TP}}{\\text{TP} + \\text{FP}}\n",
    "\\end{equation}"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true,
    "hide_input": false
   },
   "source": [
    "\\begin{equation}\n",
    "\\text{재현율} = \\frac{\\text{TP}}{\\text{TP} + \\text{FN}}\n",
    "\\end{equation}"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "\\begin{equation}\n",
    "\\text{F} = 2 \\cdot \\frac{\\text{정밀도} \\cdot \\text{재현율}}{\\text{정밀도} + \\text{재현율}}\n",
    "\\end{equation}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "빈도 기반 더미 모델의 f1 score: 0.00\n",
      "무작위 더미 모델의 f1 score: 0.07\n",
      "트리 모델의 f1 score: 0.55\n",
      "로지스틱 회귀 모델의 f1 score: 0.89\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/haesun/anaconda3/envs/introduction_to_ml_with_python/lib/python3.5/site-packages/sklearn/metrics/classification.py:1143: UndefinedMetricWarning: F-score is ill-defined and being set to 0.0 due to no predicted samples.\n",
      "  'precision', 'predicted', average, warn_for)\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import f1_score\n",
    "print(\"빈도 기반 더미 모델의 f1 score: {:.2f}\".format(\n",
    "    f1_score(y_test, pred_most_frequent)))\n",
    "print(\"무작위 더미 모델의 f1 score: {:.2f}\".format(f1_score(y_test, pred_dummy)))\n",
    "print(\"트리 모델의 f1 score: {:.2f}\".format(f1_score(y_test, pred_tree)))\n",
    "print(\"로지스틱 회귀 모델의 f1 score: {:.2f}\".format(\n",
    "    f1_score(y_test, pred_logreg)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "        9 아님       0.90      1.00      0.94       403\n",
      "           9       0.00      0.00      0.00        47\n",
      "\n",
      "   micro avg       0.90      0.90      0.90       450\n",
      "   macro avg       0.45      0.50      0.47       450\n",
      "weighted avg       0.80      0.90      0.85       450\n",
      "\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/haesun/anaconda3/envs/introduction_to_ml_with_python/lib/python3.5/site-packages/sklearn/metrics/classification.py:1143: UndefinedMetricWarning: Precision and F-score are ill-defined and being set to 0.0 in labels with no predicted samples.\n",
      "  'precision', 'predicted', average, warn_for)\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import classification_report\n",
    "print(classification_report(y_test, pred_most_frequent,\n",
    "                            target_names=[\"9 아님\", \"9\"]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "        9 아님       0.89      0.90      0.90       403\n",
      "           9       0.07      0.06      0.07        47\n",
      "\n",
      "   micro avg       0.81      0.81      0.81       450\n",
      "   macro avg       0.48      0.48      0.48       450\n",
      "weighted avg       0.81      0.81      0.81       450\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print(classification_report(y_test, pred_dummy,\n",
    "                            target_names=[\"9 아님\", \"9\"]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "        9 아님       0.98      1.00      0.99       403\n",
      "           9       0.95      0.83      0.89        47\n",
      "\n",
      "   micro avg       0.98      0.98      0.98       450\n",
      "   macro avg       0.97      0.91      0.94       450\n",
      "weighted avg       0.98      0.98      0.98       450\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print(classification_report(y_test, pred_logreg,\n",
    "                            target_names=[\"9 아님\", \"9\"]))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### 불확실성 고려"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [],
   "source": [
    "from mglearn.datasets import make_blobs \n",
    "X, y = make_blobs(n_samples=(400, 50), centers=2, cluster_std=[7.0, 2],        \n",
    "                  random_state=22)\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y, random_state=0)\n",
    "svc = SVC(gamma=.05).fit(X_train, y_train)                                     "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/pdf": "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\n",
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x576 with 6 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "mglearn.plots.plot_decision_threshold()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.97      0.89      0.93       104\n",
      "           1       0.35      0.67      0.46         9\n",
      "\n",
      "   micro avg       0.88      0.88      0.88       113\n",
      "   macro avg       0.66      0.78      0.70       113\n",
      "weighted avg       0.92      0.88      0.89       113\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print(classification_report(y_test, svc.predict(X_test)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_pred_lower_threshold = svc.decision_function(X_test) > -.8"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       1.00      0.82      0.90       104\n",
      "           1       0.32      1.00      0.49         9\n",
      "\n",
      "   micro avg       0.83      0.83      0.83       113\n",
      "   macro avg       0.66      0.91      0.69       113\n",
      "weighted avg       0.95      0.83      0.87       113\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print(classification_report(y_test, y_pred_lower_threshold))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### 정밀도-재현율 곡선과 ROC 곡선"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.metrics import precision_recall_curve\n",
    "precision, recall, thresholds = precision_recall_curve(\n",
    "    y_test, svc.decision_function(X_test))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.legend.Legend at 0x7ff3fc393ef0>"
      ]
     },
     "execution_count": 59,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "application/pdf": "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\n",
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# 부드러운 곡선을 위해 데이터 포인트 수를 늘립니다\n",
    "X, y = make_blobs(n_samples=(4000, 500), centers=2, cluster_std=[7.0, 2],\n",
    "                  random_state=22)\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y, random_state=0)\n",
    "\n",
    "svc = SVC(gamma=.05).fit(X_train, y_train)\n",
    "\n",
    "precision, recall, thresholds = precision_recall_curve(\n",
    "    y_test, svc.decision_function(X_test))\n",
    "# 0에 가까운 임계값을 찾습니다\n",
    "close_zero = np.argmin(np.abs(thresholds))\n",
    "plt.plot(precision[close_zero], recall[close_zero], 'o', markersize=10,\n",
    "         label=\"임계값 0\", fillstyle=\"none\", c='k', mew=2)\n",
    "\n",
    "plt.plot(precision, recall, label=\"정밀도-재현율 곡선\")\n",
    "plt.xlabel(\"정밀도\")\n",
    "plt.ylabel(\"재현율\")\n",
    "plt.legend(loc=\"best\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.legend.Legend at 0x7ff3fa6b7c88>"
      ]
     },
     "execution_count": 60,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "application/pdf": "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\n",
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "from sklearn.ensemble import RandomForestClassifier\n",
    "\n",
    "rf = RandomForestClassifier(n_estimators=100, random_state=0, max_features=2)\n",
    "rf.fit(X_train, y_train)\n",
    "\n",
    "# RandomForestClassifier는 decision_function 대신 predict_proba를 제공합니다.\n",
    "precision_rf, recall_rf, thresholds_rf = precision_recall_curve(\n",
    "    y_test, rf.predict_proba(X_test)[:, 1])\n",
    "\n",
    "plt.plot(precision, recall, label=\"svc\")\n",
    "\n",
    "plt.plot(precision[close_zero], recall[close_zero], 'o', markersize=10,\n",
    "         label=\"svc: 임계값 0\", fillstyle=\"none\", c='k', mew=2)\n",
    "\n",
    "plt.plot(precision_rf, recall_rf, label=\"rf\")\n",
    "\n",
    "close_default_rf = np.argmin(np.abs(thresholds_rf - 0.5))\n",
    "plt.plot(precision_rf[close_default_rf], recall_rf[close_default_rf], '^', c='k',\n",
    "         markersize=10, label=\"rf: 임계값 0.5\", fillstyle=\"none\", mew=2)\n",
    "plt.xlabel(\"정밀도\")\n",
    "plt.ylabel(\"재현율\")\n",
    "plt.legend(loc=\"best\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "랜덤 포레스트의 f1_score: 0.610\n",
      "svc의 f1_score: 0.656\n"
     ]
    }
   ],
   "source": [
    "print(\"랜덤 포레스트의 f1_score: {:.3f}\".format(\n",
    "    f1_score(y_test, rf.predict(X_test))))\n",
    "print(\"svc의 f1_score: {:.3f}\".format(f1_score(y_test, svc.predict(X_test))))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "랜덤 포레스트의 평균 정밀도: 0.660\n",
      "svc의 평균 정밀도: 0.666\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import average_precision_score\n",
    "ap_rf = average_precision_score(y_test, rf.predict_proba(X_test)[:, 1])\n",
    "ap_svc = average_precision_score(y_test, svc.decision_function(X_test))\n",
    "print(\"랜덤 포레스트의 평균 정밀도: {:.3f}\".format(ap_rf))\n",
    "print(\"svc의 평균 정밀도: {:.3f}\".format(ap_svc))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### ROC 와 AUC\n",
    "\\begin{equation}\n",
    "\\text{FPR} = \\frac{\\text{FP}}{\\text{FP} + \\text{TN}}\n",
    "\\end{equation}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.legend.Legend at 0x7ff3fa9e8470>"
      ]
     },
     "execution_count": 63,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "application/pdf": "JVBERi0xLjQKJazcIKu6CjEgMCBvYmoKPDwgL1BhZ2VzIDIgMCBSIC9UeXBlIC9DYXRhbG9nID4+CmVuZG9iago4IDAgb2JqCjw8IC9FeHRHU3RhdGUgNCAwIFIgL0ZvbnQgMyAwIFIgL1BhdHRlcm4gNSAwIFIKL1Byb2NTZXQgWyAvUERGIC9UZXh0IC9JbWFnZUIgL0ltYWdlQyAvSW1hZ2VJIF0gL1NoYWRpbmcgNiAwIFIKL1hPYmplY3QgNyAwIFIgPj4KZW5kb2JqCjEwIDAgb2JqCjw8IC9Bbm5vdHMgWyBdIC9Db250ZW50cyA5IDAgUgovR3JvdXAgPDwgL0NTIC9EZXZpY2VSR0IgL1MgL1RyYW5zcGFyZW5jeSAvVHlwZSAvR3JvdXAgPj4KL01lZGlhQm94IFsgMCAwIDM4OC45MDMxMjUgMjY3LjQ5NjI1IF0gL1BhcmVudCAyIDAgUiAvUmVzb3VyY2VzIDggMCBSCi9UeXBlIC9QYWdlID4+CmVuZG9iago5IDAgb2JqCjw8IC9GaWx0ZXIgL0ZsYXRlRGVjb2RlIC9MZW5ndGggMTEgMCBSID4+CnN0cmVhbQp4nJVZTXMcNw6996/gMTm4RQAkQB6tzUZVOezGWlXlsNlDauz4oyynkmxWf38f2DPTbE1zlKhK5ZmnbgAEHx5AmsKn6eY1hfe/hxg+4fcp/Dv8B/++DRTuws037/738fDu/u42HH6fIvDHSUqZaxTijK+f+6+sNqeq+PQZz/bfPkzTlwle8MYdDL+fpiRzWt6SOkv2x2DayszP0M89yllnO9pcLfQoPP2MFfGyovdwhlXNpVuXh4C/TLnMyiTR+gg6MM/lGMB0i9ifptuHcPMtBYrh4ecp0yyxlKiFCgdOc67Rn354O30V5/h1ePgU/v4wNdcTUUUqEG3qffXodWdEPFOhLCnWWC+88TNvJc6iakobbx36gjeTOebESqKiF97S1hsnmllqKrX31qPXvbHIXFlrKXCWLrzp1ptEZIKYSTe86dAXvNUEqlQjOKl04a0886Y814I9ko23Dr3uTXKe1Tglwe6VZ95oZck2RCwm5kZsApPo+Pi3399vg1v5n6rnGgn04OqpGld0kAqazUp7MmNtFk18C/YZvDpDuZthRbxxtqIvOStpLpWUqnjh7BJ4dUZcvMyq2cZbB7/kjpB0p7ImM7Z9Cnf+oFklSWbZ+lvhF/0p6KzRKDKb7JO4U65oM+QNKrLx18Ev+YMwQkQy5+xKtE/jzl/SuRLSkbb+VvhFfzAWuZZI5OX6jMi/QvcpvCIILCXkodUGMejE1D4fHreGvaW4jYfv78OPX7kZV1aFA7gwkuOff/y6/WnxgJoI/l77gO0tWmuq0MYCEA5u/vHTlz8eb3/67Y8vd7/898PHw6vDx7esGsM3v0xvLixAf6zUKsyZrhgQ/OwbSDxbrJUyWH3NAoMWiwXYCDsNUASLCUzomyn89i78EL4EDt/h9+noTU8/Fu7vpm137npXLwgrmsEGk5xcrYBW6FIuIxQxRBCKHIU1NkXj2UdRHEyRJC1ohi/dB9HINWVLtoBFpCJj+2ghUFKiugGNM4lBOQdoxT4UklwXNCXQTUYoqEw5UllQS+I8HqDgcEGDcglV0DxppjxCETlCJ15QweijOkApFkg8OR8arJkL2RBGKeWinJoRnjkjP3UMQ6syCLWgIBbbAGT1uiw5L2hRNH0aw4aAcm4aAZEjLdl4CIPKVCOlusDJyKCeQxjEAgO0LTzNUhCwDmEUnOSSUnOZZ0ZPlSFqnhhfQENLVVT6GC74CKq28NBjMFKkEYrentjAmgWtDN88hkHymqN4B1EvT0wgMoRBQWHoGi8wBj801zEsmCs00oIWdI6ah6hiQ0XYU2oooETg7xA2lIUmiWmBUaOSyhjOMxQN3HIYZM+ieYii+Qi2oC4oOgVayhBG2WeFRHp5Gi+f0xgGDSsE2dluTjIfbsZwcsnJlhcYfa7A4BAuM0WMHr5h5pREvY5QiIpoYaIFhevklBzBaVYMjNyWrjjxpAjiD2AfEkuU0iYSwDka+4Q5gjEEiaXCC1woitYxDAEwyEsL0DBZY/IZoobml8qCGUEYhyDanJTU5Ni8ihSz2RDG6FsRexNkbxBcGVsxhHnWmK3Ng+ZllPFlDKd2NmkCXpyRppLGcPYGGauTvYBkFlnKGDYkXS37BjjhDOo5RCsKLabWHIorVsm1DGFMURixsQULXCNEdAAy2rRZc+a7islVhmhCAyMpnv7iu2oZZT2Ezeduaw3Ke4p5gZQrOAQwMY6djmOIxPHTnAsjXCISqbEdqFzQIeK1XsPxMVYtvglo38hFhThfwcXneGktE4UBQhenzwiGlJvFJqEI1ve4olLHuCcqtkWBlVgU1FnHMI5MfjKsDUcVUHYlG+IJ4o91mFOfC3QSM39KV3AcIwldwPcENmapYChdwcXHMK4ep3gzxWgFtoxxUNN7SruiyY5jELIreHfG6I7Q27PHv6Y34U/Pxjev5XTNs4zI76dNL9FTd3j07gAVatPUCkN80S+sPWtzrH6kDNghqmCE9ygwAW0fcnRweSSu1RoKZom/r2uDwiEW8yUY1IPn9tTeP8PQ6NpOh6tRdBscUQl56vyjoDFCWyV/f40V1RuhU4x5el1VWVfVtU3tAgBNsJiiPYzBIOYCAvRWMbWgJtmbTx8BJnq0UAwJfbS6tyzdT4Hupauzuia2i6Dfgy7abr/WddleDqzLwYcTwfxUFS9uBrfnqd0byv1LR1jdu7p8HF1d4vm/cP+5eXo1c836zeu0VMZ3ftuK36e21OPdKyg3t3uf1Hmmk6kz6jevaGw4CCF5Z1QwDJ2exGTHXEG/HtTjg4epQ30X6GhzBf1WJWPKCp0jrKk92UV0gg7nyE8IFK+i22CmA1dOoENb82dEV0MLco728wqdV9XZPi//IneNVbenK2duGf/U9Gi9cL44uj+jmkRa4vOyUO892BF+jrWdjY25xy093mT/Ovm1wyu/ehFewkOt2fLW6Mbl/p9/C8NLFbTPVCvOfWhvV240vI0MLlUQvNaKTcRwd+1OBK39dCeyJ+dCSxZSRQKPicFUlzJkqUOFMKFjqsC5Ad2e0Hww0woGOsX8hyEuwwpONxjKD63nYaDCQTWgO0WLGNbxrDY/WXBGh4blFUlHH/7mEcKgVRNmuNDbwokSu4v5q/frBZpwBGxvrzH6hIdzQAndYvK6mGXNZwDvYko2vxPrH4vgOxve7e1h9IZEJkwbG9/RtRdCy32c8WI58XLN8SIxna01f53fPtddjOu+rItJz9ec1jV/WISs+5+cpxHb3eUu2RllgMNk8kk9H4kf4oj28erlXRrcH6Jr4VhZoBhoTNdqJdu+Ab+bkVoVuYlyzULEAf5YK2+m/wN/EaIfCmVuZHN0cmVhbQplbmRvYmoKMTEgMCBvYmoKMjExOQplbmRvYmoKMTYgMCBvYmoKPDwgL0ZpbHRlciAvRmxhdGVEZWNvZGUgL0xlbmd0aCAxNyA+PgpzdHJlYW0KeJwzMjJRMIDDFEMuABpXAugKZW5kc3RyZWFtCmVuZG9iagoxNyAwIG9iago8PCAvRmlsdGVyIC9GbGF0ZURlY29kZSAvTGVuZ3RoIDE5OCA+PgpzdHJlYW0KeJw1UDkOAzEI7P2K+UAkwBz2ezZKtfl/m8GrFJYRMBfuCYELXuobUyZqOd46upxa+J7KymF7wabC1oLyXcOKlQcsHdpbJMvZA+tVmCzyBrcdr1AONKm0CjrZUd2E5SN9jb+Je2z2Sri0QYEsEEHsJlNSTYVKdv52SLmutqI3PB9I+MGQIi1BxpLonB2Pfcp2dY92UkH/dFamx2uuaFa6T/rpMHk0E0FbnDBvTHaY33mJvoj7wfBGLnYOOeM56OPx8wOK0kNmCmVuZHN0cmVhbQplbmRvYmoKMTggMCBvYmoKPDwgL0JCb3ggWyAtNDUgLTI4NyAxMDAwIDgzOCBdIC9GaWx0ZXIgL0ZsYXRlRGVjb2RlIC9MZW5ndGggMzc1Ci9TdWJ0eXBlIC9Gb3JtIC9UeXBlIC9YT2JqZWN0ID4+CnN0cmVhbQp4nE1STXJGIQjbv1NwAWfkH8/Tma7s/bcN+nVeV0QjEQJPOaVN+nlE9aD9SK6DVOLGvMzXY3wZi8s4G6KAcZ4UC8xKCtxYBvmKznEndzBQ85mk/SoWGHUjU+gjV5eT+CINByOzSNWJnQ9bSbK4U3CPV/vhrqsWiSiUEavw51UVctzoMlSASlgo7n/ZCBVox4Cknoj//CJBPx057cPAngb7+fgEUNCfAscS1TTaByly9xP4/UVZ88O+yLP+VF50VCyYBksPg80v3AeKdc8CB1/E1jla/A/J8aZ1XjRCgZyThvQkYCWT43F2fzGNWJmClRjzCkxIsmfd02y0L7I68xVBbN9nOxoYDOM24A63uwUVTLWzgTC5bPfrrlGrJg6jqKbSwF11nxgJCKSi4646veHEiqHN0ZWiyHFlcRqCFQshCKMoPV3oOaKHJh0VDe898mgBJdfViuss4GBs+tf1Ztl/36+JwxITaJR2vPYXNPe+2s/3Ly2DnPIKZW5kc3RyZWFtCmVuZG9iagoxOSAwIG9iago8PCAvQkJveCBbIC00NSAtMjg3IDEwMDAgODM4IF0gL0ZpbHRlciAvRmxhdGVEZWNvZGUgL0xlbmd0aCAyMTUKL1N1YnR5cGUgL0Zvcm0gL1R5cGUgL1hPYmplY3QgPj4Kc3RyZWFtCnicTZE7DgMxCET7PQUXiMTPYJ8nUqrN/dsMxlFSWPOW4bPGV66gh5jS+0rhxntjTi6C/6NjD3PSsVBj0zbdm+y44488i2zZlyLJ5kQGkwmq0Vx90PNSRCSTJAaJJM2k2HF8qqFUimaQspAGzlAyn0gxRvs9VmjogqIW3E4gorGgkxRZpWiLLkUSsxWXauek3gXJdXdZfEi1Y4qqUuNWTPITma2upQrHNSjSEcEEZLkqBrTDNLBXl6CRqBH8/15Grbiol51ZswPZP3oIr2P/YT/G6wNGVFklCmVuZHN0cmVhbQplbmRvYmoKMjAgMCBvYmoKPDwgL0JCb3ggWyAtNDUgLTI4NyAxMDAwIDgzOCBdIC9GaWx0ZXIgL0ZsYXRlRGVjb2RlIC9MZW5ndGggMTU0Ci9TdWJ0eXBlIC9Gb3JtIC9UeXBlIC9YT2JqZWN0ID4+CnN0cmVhbQp4nE2PMQ4DIQwEe16xH4iEjb3m3hMpFfl/G5tTdFSMGK9ZmgxFuOPbYsSmleTgnIh0VCI04Jx4t1CBS0cIYZQ8O0x7GqpvWo1D0hmY+7wHaAr3PWOEX7nTrrQE85azTLUoWu3fZzVTy9zIZkWDZeeUg3SyEgfcbmT/h2rHsy37ueElo75cKP1+Nk6KGgzrB+3IkV7t8wM08DxjCmVuZHN0cmVhbQplbmRvYmoKMjEgMCBvYmoKPDwgL0ZpbHRlciAvRmxhdGVEZWNvZGUgL0xlbmd0aCAxMjEgPj4Kc3RyZWFtCnicLU7LDcUgDLszhRdAwqSEZJ5Wvb39ry8JveAIf6dvDOhEXwN0gwkeNsrI69e46qAqtoLboRfuRicWyyCrgFrfhj6RbKeVvAuTCH+/IifiqkkOBsNzuZTA+DnUKiKXzXwrPzDqgoruUNaUMOYynp1n993eP/NpJ84KZW5kc3RyZWFtCmVuZG9iagoyMiAwIG9iago8PCAvRmlsdGVyIC9GbGF0ZURlY29kZSAvTGVuZ3RoIDIxOCA+PgpzdHJlYW0KeJw1kEuOBCEMQ/ecIhcoCecHnKdGvZv7b8cJ0wtkCxLnhdgmUzzl8SVhW5aF/GDo0ba/wxbaufFmqvhySUt5RyAl1uw2/1e7L3TK2qrAjO5x54Mb5OGlsbwm1pjSd2hcp1jy2BFsqlYPgnS8sMnDuHOjNllDlgoQkpS0ZmICaiHNU3KBaIowVwNXxl0BrK+lFHfd79pvf0BmfQB2toOZRHirWZPRKRnBeBAKpDnVjMMfYVp9AzqU8g5jDY1tr1cnZ1e7rW73gterd0C5GmnMK4Qv1Ds+f3XPUckKZW5kc3RyZWFtCmVuZG9iagoyMyAwIG9iago8PCAvRmlsdGVyIC9GbGF0ZURlY29kZSAvTGVuZ3RoIDEzMyA+PgpzdHJlYW0KeJxNjjsOQyEMBHtOsReIhI0/cB6iV5H7t1mIkFJ5bEs7672hQlTxao7miSS/peyTu+FzKCSwithAaIekI0whg3tUzKLaENmhFojBGb+gWW7kOrQlq2jKH+3ETfWSJX0VqgYfBq0C78IsocGDfnd2S0hjR3biR3mh+7S2jtt+lucL2ncqsAplbmRzdHJlYW0KZW5kb2JqCjI0IDAgb2JqCjw8IC9GaWx0ZXIgL0ZsYXRlRGVjb2RlIC9MZW5ndGggMjcwID4+CnN0cmVhbQp4nC1Ru43FMAzrMwUXeID1t+fJ4VV3+7dHKSkCMqZMSXRsw0IEPhYIVRTxR64jiCr8XdQjDzaLSga24L5IMhK/l6yNrA0x4klI0UOUJboCpQV1H1fdZ/C+LGWYS7J2w8ORuxU/Sl9FiCDXg2FOZVh/q+B0irXgHOy5406XTWS/PvWVrXBCo6/7gqVRVZj3+L4cxs62HXqMqkKzqEjwNHqxJk4Sc+sln9HevIbYIepm8RIyEzpbfzGuFjZduHWtmav/XImH6PZO40mVuXnl5OJntaIMvFXlM8h50Ga3ZlPLxPgGxhdLk7dPdlcrZDL3VrLT1TjPCbtnFISbp+tsTYXZiQbdAu/z39f3H8iAZiIKZW5kc3RyZWFtCmVuZG9iagoyNSAwIG9iago8PCAvRmlsdGVyIC9GbGF0ZURlY29kZSAvTGVuZ3RoIDEwNSA+PgpzdHJlYW0KeJxNzUEOwCAIBMC7r9gPNBEQqP9perL/vxY0Tb2YyS6InoIKJhyiUOtwZlxUIiFxPAmOaBRRmeUoTW0TVw/l6q9cXe2vPBC/OG1aLX+Q5tAucTUerhbRxBrXTdppymBt3xzlfgF59Ch4CmVuZHN0cmVhbQplbmRvYmoKMjYgMCBvYmoKPDwgL0JCb3ggWyAtNDUgLTI4NyAxMDAwIDgzOCBdIC9GaWx0ZXIgL0ZsYXRlRGVjb2RlIC9MZW5ndGggMjA0Ci9TdWJ0eXBlIC9Gb3JtIC9UeXBlIC9YT2JqZWN0ID4+CnN0cmVhbQp4nE2QTY4FIQiE95yCC7yEX8HzdDIr5/7bKe2evF75KVWFQCmTU4R/6VANXjRmvqhaQNX6IjV5ql/aKYv+8xZBZtEIhsIl8KJzcshga+fMZivnwnmRj5vWoTGLPYtHwBnQ9tgaqNMKCUip5gicUFwUvXOV0+EbxTmdHXxR+u6SyN0ahyYiUXPooS05vfu++UC12VU4kHCRdZ4OlsjDiym8aagotuOKPTiGBD+zXtRo8VGNPbY9uA7q2N9w6Rd9fO5Fbc+XbsvPH45lTcAKZW5kc3RyZWFtCmVuZG9iagoyNyAwIG9iago8PCAvRmlsdGVyIC9GbGF0ZURlY29kZSAvTGVuZ3RoIDU3NSA+PgpzdHJlYW0KeJw1k8txJDEMQ+8TBROYKvGjXzyz5ZOd/3Uf2Pahi2pJJEEAmidt2Ex7V9qssJ3L/vmr7rEV135eta+tGlZT8VrltjWXfV7l09YKy+vEZcmn+Hlljl7FJWdui+1d4/MK2qxIC+dkDHNuTE4/L6ePAPgqm+OaA6qeE3Yr0xxoebmR03KqmoNFq2/uJP2LLPDs01V0V3Wn1Vjm51j5sRhlFUdYgqmSHapUgZKpiq6cKId547KjCclTZLJZz87RDT4HZSmnyqnGR+dyPph9EMwhTLDsoC2mCyI3OKFznOq5Y/oTXQw2E3uhDMipPIPZYIUTRyQACxpq7IJdtWdw4LB+uyQhhlrkYiXiIE0aS5J3Aw5G7h3IeafaFbl/hDvaCO8xErbGOM4mCkNuiwijU3iQma7wFhwR8oEpPZhZpLsx/cPfASWUOdJPCPaYNq/uOzKuZIcbi34RxNsyLfJDdU+7UzIrMtt9/FoA3SkhdFM5EkE5dZZtCJ2DG2O3GNiXu9/PSv0xxIKrP8t/2kxBt59eOUJ6IQPjyJSe6tB/VHZGu3JX2WlLqor8LRdjLeq275sYk7FoKRFC5GX0f9bRaS6xpW28ip4lZM94R1oW6HzIZmKOWLO1kPqIs2TJeoxOjNayV6kTp9PtnDh9QpUccKTOsK3HkT2bnouMnleYMMGWgXfbnzEOOxhd7OdYvyYPnq9WekS6EVKqc2RoVQk0z8an/N1Pk2o8Ql8iR+zipd3s1u1KLrTy5q8in9fXf/lC30AKZW5kc3RyZWFtCmVuZG9iagoyOCAwIG9iago8PCAvRmlsdGVyIC9GbGF0ZURlY29kZSAvTGVuZ3RoIDY2OSA+PgpzdHJlYW0KeJw9VFuOJDEI+69TcIGWwjPJeXq1XzP3/10bSitNT+iqAMY2ncdliW/5REhGyvaUP/rYVbGU38fVGTge2xFPhCnfx3dKmPh1SZNYV4qPw0L2lvASvAnUU1U+Txe1K5FHNNecO/oNolvCG6a7cwzpeIP/lmwTYnua2OEbdvVF2ECja1Dp4RuAcyshavclHIPn97Ha8ySqb1jfYo6t6ip6TlfV7W8fzd2dlZjQQ8HTYFNwQbQKbETfZ8/TESbkDU7MnOGAVQCcRUERuzRlbAsKiQKENiw+Jk42AWzQ/8rxbV1iGYUBHYwct2JhbCREN/IKRBj3gGhqd/EdMEDpwt3AR5FboJS5+75kxwXZoZIac44MjLK6QrbqV/Jai61SrAjyykA4iKhxxzGpAAKgL+R6wB9lLRByCkQDaYFazsMTk+01T3JumPubY5pdRe/tqnry7aNbu7PWIFGYaLCRbKLVF73+n0ffCXmDE2vGy4GSEXwoJFmiMsObLWsmTU8zSxMM18ZssD8GnXl4jlJcql9IeToyZHycxr/ysdNYsHhKo/CuKN1HE+GCg0EM03+jEICt4KoqOORRLTbQ0AnYDbgIyO2w8mUFVPTb5Cs3BvXQsqOUH7ibZdCcVsH9O374PsVthNiYjOuN34UmmwHYwb4mWkChUs6JCsV1B16uMp7Vbp45weUylGxqg5XaeltPRHZbaf7ckKk9vvGJCJA3HPpMDtnf7d7bVaPu2yfO7c65vIEkVn+QJZyRvV1nkPOcUeh7zgZncdS46509sDskI3zIcVT3Gk/fpvAHG+FNahxrknPtl/V0axn4Q0pVeI5MHTk9eFvGxHQta+pplXuTUVLHAVGLjgiPNggtNZZxrAlN5FiPNtVrs+/z9x+Z5BPlCmVuZHN0cmVhbQplbmRvYmoKMjkgMCBvYmoKPDwgL0JCb3ggWyAtNDUgLTI4NyAxMDAwIDgzOCBdIC9GaWx0ZXIgL0ZsYXRlRGVjb2RlIC9MZW5ndGggMzQwCi9TdWJ0eXBlIC9Gb3JtIC9UeXBlIC9YT2JqZWN0ID4+CnN0cmVhbQp4nFVRuZHEMAzLXQUbuBlT/KR6duYib//pAbR37YsICSIIQlvFLjlC3htRVEkNl7ApWaj7kteWuPEJ1qd4prhX19dmHo1Grn4xdF09ihuqqHursnIOGCIoKSfXLpomuXYyFVKKm+VwUTLGLuUKZmBORcigvywxxbmoZmFS08QWmDnbG+tr8+smoFCTLhZ66DqVKrNdVGRvWj7IrNWTS2c7KdPLW7m1287Ll3xyQw8VhiLBRpiftiSxf+44T6YRNaR2eEE+1G6XSpdW1si4K16MWVfPwGZUGVat2nV0GkCB7fiC2bInki6pEnBK1bDoOaxwwJ8BohO+oLezh26pQvdU/eyDHuQzkMu7kWHrAz9uD0T2+L474DmvDiLD/bGp3sCR/LHN+UQnmfB2o1ONf/9APcDyBj+p5DKeSI20hj4hvvDo7/kCRUInd6P/Oxzb7x8IUJd4CmVuZHN0cmVhbQplbmRvYmoKMzAgMCBvYmoKPDwgL0JCb3ggWyAtNDUgLTI4NyAxMDAwIDgzOCBdIC9GaWx0ZXIgL0ZsYXRlRGVjb2RlIC9MZW5ndGggNDI3Ci9TdWJ0eXBlIC9Gb3JtIC9UeXBlIC9YT2JqZWN0ID4+CnN0cmVhbQp4nEWSTXJDMQiD9+8UvkBnzI+xfZ7MdJXef1tJdpuVyQPEh8hTe7S5evt55jJFb0VehYjZT/TJfpkttLjd8K3QLf+6/6ObZruFevZU9Fb0lXGzn+hkK9oIRwODylarjR0NCOXWXs/2Vrmb9dVqRjPfrZB/PTZ6m4YvE8T46j3aTLZ4YMMR2IgvMtwJ7+sJn4piLFXEqtuT2IQqGVuqWf3OSaJgcu4SySCJ0IZ3sQ4voevFLswU6lGLiqzTk2cOVJL6UI21NSeKbJwc0CdJxGELpxppw1AH+uhD+/DFpthQX7AxK+jB6aErVKFLVKVvZw6d5GQYSxBwHDIYT9QKkd+zoDznPZAiDKFkQsCAlZOtVqPlOKfIwN0XMk7zvFvLjhMY0CHrAaytI8l+HGny7WcpvFjXSlFEqSJG3p6YLpVYU6rZ+50jSzGZv0iSMOSwJf5lpM100evVZopUi7/cPD1jKwOVYi1Ui6fEnEqxYXKRBySFk5Otpo4E2iK18U3twxebQltfRqiCHpweukIVukRV+nbm0ElOprMlz9dlo/uk1TXGfbXP9y80+7sJCmVuZHN0cmVhbQplbmRvYmoKMzEgMCBvYmoKPDwgL0JCb3ggWyAtNDUgLTI4NyAxMDAwIDgzOCBdIC9GaWx0ZXIgL0ZsYXRlRGVjb2RlIC9MZW5ndGggMjQ3Ci9TdWJ0eXBlIC9Gb3JtIC9UeXBlIC9YT2JqZWN0ID4+CnN0cmVhbQp4nE2QS44DMQhE9z4FF4jE1+DzRJpV5/7bFHaPOiueKQoDQ9eiaUyfoak0tUjdECepnPgeUock7qj/yjxwbUCna0jGTSq8Sf1ktM77PUxOxuK4bBklSyuFabzIUBtrks2icG8lgnwpmS1yR5Tjg+JM5ola1LMjm6SzPd7bqJNPJSl0YsYO1Yohg9c1LE/OpGvhTDAnajSUNDC36r5Fd5DsKSUcN1gkkrSEMinaAJkLLRfUxJWmwB+kjEnddsckN2QLjv7rvv17JHZ9CYb6jMTZNl4bDZtcI1b80Eu4Dxwiv5jJt/6QW919HspqtX986P7w7wskamO0CmVuZHN0cmVhbQplbmRvYmoKMzIgMCBvYmoKPDwgL0JCb3ggWyAtNDUgLTI4NyAxMDAwIDgzOCBdIC9GaWx0ZXIgL0ZsYXRlRGVjb2RlIC9MZW5ndGggNDc3Ci9TdWJ0eXBlIC9Gb3JtIC9UeXBlIC9YT2JqZWN0ID4+CnN0cmVhbQp4nE1SSY4jMQy71yv8gQaszct7Aswp/f9rk3RlKoeKGUuiKMpXjWjp0X4voV3tfQ38PmiuDjSXfSHrdkcf5BVAZHlQ+FB0fCF2O9GDFhh+zBIS3G74FrQutNcX+on9qXnQKYlsNTtoAEY6m0DWgxj8n/W+rDARqn8PmtSSc38jzfbJQ2mHku0UCgYij4lzNO8bX7XXZXDHIcdqNsfMBpUOCYg4Iri1jsisthHYLNm9hVVbMCmGjlq4JoDhiGbvzE7vpEF5JtqDLkmHBgnJiOBfLkiEhEIJRRW+10WZZatRdsF0DsLzBTfGucG4zEiwn5qMEktCDFlzjrtP7lTnAguVlM1bW3lJbEWXdp0aRijYqzQsa870ZKEdZKU97HMMY2daSCW0lNqOyVRL26mea/gs5qUV+eCKIlwo8Bh9JCbsmh856ywgsVRf3tKQuY0d5Q1mDoNKnny40pK4wdPJTHXVqcmEwKsMOK+aLZ/Akni4ZE3qt/PkEenoCn1BX7G/mMhMTQa1WeyFjEpNxhM7XCHEmZjhyD817lMsTn5O3T996Ak720opMfh2tNlMqbXpUq9T8wjRV2bEXSMPyEJXjuNdfY5v3A6ddD+7pLbjNdXSfarXNu79vK5/f22P1u0KZW5kc3RyZWFtCmVuZG9iagozMyAwIG9iago8PCAvRmlsdGVyIC9GbGF0ZURlY29kZSAvTGVuZ3RoIDMxOSA+PgpzdHJlYW0KeJw1UjtyBTEI6/cUusDOmL99ns28Krl/G8EmFRhhIWSnHSxY4HZHiqNs4UuuXALRwA+zAzEZUAhOVMFzdWaG7yu2ohBeuEURa+Pm1efyZGYF15ffak98LlOfTIuIMZLnliQiR6CQCOgeBamsnsBJaqOCRBDYzeLKi9LqlWcG825m4kvhBo+kXirq7jqIIPUyxGG05p6RWchcULblJplvFLueSymmuKvZHmPsrIkcba9VTk1FlUGXSnrp4Fq1WCmy7oOkHZmNdBbVhk1Gc3JxG2LJyfHXQ7nJ09lTCW4eeYbXRq8Xd6I9dt7Y++63ouyg9lad1og2Qv3a70ZLtOfV7Mwni1iQpBtSED9k7jnti3u//uaMeiOflggzpQqZe/wFVCDSP0HpO0fZKorgLetAk2qSIMox/1/quT6/tMZ1CQplbmRzdHJlYW0KZW5kb2JqCjM0IDAgb2JqCjw8IC9GaWx0ZXIgL0ZsYXRlRGVjb2RlIC9MZW5ndGggNzcgPj4Kc3RyZWFtCnicTc3BDYAwDEPRe6bwAkgNiRsGQpzS/a/QikJvT7LkTzoKDmIzgqaIXXGqTLah6hUpGrbItTyixiKzuf7qxylvYaD/5ldIuW6VLhmBCmVuZHN0cmVhbQplbmRvYmoKMzUgMCBvYmoKPDwgL0ZpbHRlciAvRmxhdGVEZWNvZGUgL0xlbmd0aCAzODAgPj4Kc3RyZWFtCnicNZE7koMxCIP7/xRcIDPGvOzzZCfd3r9dCbKVGduIT6JiyxLf8nKXWillS370sfK++n1sa1ebN4bTTF4a8n70bjHRKEGP7ivZt0vlXqkr6kvSRU/h2i86nZP2CR7mymsUNwWv7sXffimC7sATteLYiOuhOqZl4Aaz86ymqbXwQr7aS8hLE3TAkzPGlpf1j1D/9kREq8RdrcoZMyc9e3LmaZJEPWh5T7MyLKL32V5Y0Rx/0Ct7xjxVJo1kNj2GWpwLMGJYY02wBH11WNnR/y/j3WsxxVZcjQVBDDxVoKTgIpiEbbmIDstVvbwvVPgbwNE2XLLHKDp2JJiwGdzyNBjGC6uDF/zwfbrHM6kGlVguVA2rnhNjCIPjwsuikjdYzhJ6vdg0wDPGUzaz2VS2JqANA9OzGRlUtmarKuKeOVrZkzW1STT0y6ZYBWnVsun7bD+s6JA/6Jg9kwFVmApVmRLnTG6cjCAJAiqSDTJQETvJsYTvUt7P5w+mvZ0pCmVuZHN0cmVhbQplbmRvYmoKMzYgMCBvYmoKPDwgL0ZpbHRlciAvRmxhdGVEZWNvZGUgL0xlbmd0aCAyNzIgPj4Kc3RyZWFtCnicNZE7csRACERznYILuGqg+cycZ12OtPdP3aDdiNYgPv2InbJkh/wgJKKkTOVXr5Z+Qt6Pik4ucVsSSMFZ8rrCTJB8WRDAxWsL1JnxMLED/s9YJeAUy2IG7GLBDqqMOq8dX5cWRt2jep37+uw1ohe7Lyv9qO7QCr6llB03pNbu6VqS54gzk6z0csmZHmtJ+pGgjzR6atNnMv2aexAEvXz9s1uyA728R4XRZ5yHBvtHNQknkzjNRyUX/yeR1PHL+WnkktyhyXlv0DXgzulOTqxxUmD3jk3CRt2jYGjfnPdR+1GgV9BDkweaYN8CeEjAny0QQ6QvRXKOvsVDxlXnUvwCL9UuWf91+7r+/gGuYWVKCmVuZHN0cmVhbQplbmRvYmoKMzcgMCBvYmoKPDwgL0ZpbHRlciAvRmxhdGVEZWNvZGUgL0xlbmd0aCAyMzIgPj4Kc3RyZWFtCnicTVBLbgUxCNvPKXyBSgQMSc4zVVev99/W5M2TuohwAviTGgbDSnxFItfG9IHvcdEcviZ+r37s5uvinA8KN/gcQmPGP/TuPmwHNJ3mzT5o1UHMgWmqu1DluK+koTyQpSPGU626I0TKHjdiT6T24+xwB8JLbEQ02+NandnblP+DfIO1kFJlmbh3z+iWmmZsVG87USO6Y3rxlJqjQhqza+9EClFqMVSnVPept/L7Qa+Dgp3W9WNv1HONIpZqHP/Bw6gfCfmnabaTeutlu5BitD9pUlnaOf10lIVUbmWj1D5p7+vnD7T9V5YKZW5kc3RyZWFtCmVuZG9iagozOCAwIG9iago8PCAvRmlsdGVyIC9GbGF0ZURlY29kZSAvTGVuZ3RoIDY4ID4+CnN0cmVhbQp4nDMzMlQwUDBX0DU2VTAzNFQwB/JTDLmMTU3AQrlcRmZmYFYOmGVmYgZkmcNpkPIcLphGCAsiBzIBwYKYkAYAex0VFQplbmRzdHJlYW0KZW5kb2JqCjM5IDAgb2JqCjw8IC9GaWx0ZXIgL0ZsYXRlRGVjb2RlIC9MZW5ndGggMjAxID4+CnN0cmVhbQp4nDVQMa5DIQzbOYUvgIRDQuA8r+r0//3XOlQdgCgmtmP3hQGGodMPnIHcjhdb9SYT/98qjn4lfBi49HrgaUxVe4NnIOaAjY3IIcRsYVEdTywRWxJr14wdSRgxdTIcU2hJPu0n/tdmbOQI6QeWmGwnQkyalg9fW7xCD78vL6LK3FA/uOLO7AKKZKMou6yVyF1VgO5bSlBmOimfctc1Vn578PKmIPGVj3lXXLN2184cE5UBpV2pUFkIUU42eVO1dPySfNr7A0ENRIgKZW5kc3RyZWFtCmVuZG9iagoxNCAwIG9iago8PCAvQmFzZUZvbnQgL05hbnVtQmFydW5Hb3RoaWMgL0NoYXJQcm9jcyAxNSAwIFIKL0VuY29kaW5nIDw8Ci9EaWZmZXJlbmNlcyBbIDMyIC9jaWQxIDQwIC9jaWQ5IC9jaWQxMCA0NiAvY2lkMTUgNDggL2NpZDE3IC9jaWQxOCAvY2lkMTkgNTIgL2NpZDIxIDU0Ci9jaWQyMyA1NiAvY2lkMjUgNjcgL2NpZDM2IDcwIC9jaWQzOSA3OSAvY2lkNDggL2NpZDQ5IDgyIC9jaWQ1MSA4NCAvY2lkNTMgXQovVHlwZSAvRW5jb2RpbmcgPj4KL0ZpcnN0Q2hhciAwIC9Gb250QkJveCBbIC00NSAtMjg3IDEwMDAgODM4IF0gL0ZvbnREZXNjcmlwdG9yIDEzIDAgUgovRm9udE1hdHJpeCBbIDAuMDAxIDAgMCAwLjAwMSAwIDAgXSAvTGFzdENoYXIgMjU1IC9OYW1lIC9OYW51bUJhcnVuR290aGljCi9TdWJ0eXBlIC9UeXBlMyAvVHlwZSAvRm9udCAvV2lkdGhzIDEyIDAgUiA+PgplbmRvYmoKMTMgMCBvYmoKPDwgL0FzY2VudCA4NTAgL0NhcEhlaWdodCAwIC9EZXNjZW50IC0yOTkgL0ZsYWdzIDMyCi9Gb250QkJveCBbIC00NSAtMjg3IDEwMDAgODM4IF0gL0ZvbnROYW1lIC9OYW51bUJhcnVuR290aGljIC9JdGFsaWNBbmdsZSAwCi9NYXhXaWR0aCAxMDUyIC9TdGVtViAwIC9UeXBlIC9Gb250RGVzY3JpcHRvciAvWEhlaWdodCAwID4+CmVuZG9iagoxMiAwIG9iagpbIDg5MiAyMjQgODkyIDIyNCAyMjQgMjI0IDIyNCAyMjQgMjI0IDg5MiA4OTIgMjI0IDIyNCA4OTIgMjI0IDIyNCAyMjQgMjI0CjIyNCAyMjQgMjI0IDIyNCAyMjQgMjI0IDIyNCAyMjQgMjI0IDIyNCAyMjQgMjI0IDIyNCAyMjQgMjI0IDQ0NiAyOTcgNTgzIDU4Mwo4OTIgODkyIDI5NyA0NDYgNDQ2IDQ0NiA1ODMgMjk3IDU4MyAyOTcgNDQ2IDU4MyA1ODMgNTgzIDU4MyA1ODMgNTgzIDU4MyA1ODMKNTgzIDU4MyAyOTcgMjk3IDQ0NiA1ODMgNDQ2IDY2OSAxMDUyIDY0NCA2MjcgNjM5IDcyMSA1OTYgNTU0IDcxMCA3MTggMjQ3CjQxMCA2MjYgNTI5IDg4NCA3MTAgNzUyIDU4NiA3NTIgNjEwIDU5MiA2MjEgNjk2IDYzNSA5NjEgNjE3IDYxMSA1OTQgNDQ2IDk2MQo0NDYgNDM0IDQ0NiAyOTcgNTYwIDU4OCA0OTAgNTg4IDU1OSAzNDAgNTg4IDU5MiAyNDQgMzAxIDUzMCAyNDQgODkyIDU5MiA1NzcKNTg4IDU4OCAzODMgNDc1IDM1NyA1OTIgNTMwIDc4OCA1MjggNTMwIDQ3MyA0NDYgNDQ2IDQ0NiA2NjkgMjI0IDg5MiAyMjQgMjI0CjIyNCAyMjQgODkyIDg5MiA4OTIgMjI0IDg5MiAyMjQgODkyIDg5MiAyMjQgMjI0IDIyNCAyMjQgMzAzIDMwMyA0NDYgNDQ2IDg5Mgo4OTIgODkyIDg5MiA4OTIgMjI0IDg5MiA4OTIgMjI0IDIyNCAyMjQgMjk3IDg5MiA4OTIgODkyIDg5MiA4OTIgMjI0IDQ0NiA4OTIKODkyIDg5MiA4OTIgODkyIDg5MiA4OTIgMjI0IDQ0NiA4OTIgNDQ2IDQ0NiA4OTIgMjI0IDg5MiA0NDYgNDQ2IDQ0NiA4OTIgODkyCjg5MiA4OTIgODkyIDg5MiAyMjQgMjI0IDIyNCAyMjQgMjI0IDIyNCA4OTIgMjI0IDIyNCAyMjQgMjI0IDIyNCAyMjQgMjI0IDIyNAoyMjQgODkyIDIyNCAyMjQgMjI0IDIyNCAyMjQgMjI0IDg5MiA4OTIgMjI0IDIyNCAyMjQgMjI0IDIyNCA4OTIgODkyIDIyNCAyMjQKMjI0IDIyNCAyMjQgMjI0IDg5MiAyMjQgMjI0IDIyNCAyMjQgMjI0IDIyNCAyMjQgMjI0IDIyNCA4OTIgMjI0IDIyNCAyMjQgMjI0CjIyNCAyMjQgODkyIDg5MiAyMjQgMjI0IDIyNCAyMjQgMjI0IDg5MiAyMjQgXQplbmRvYmoKMTUgMCBvYmoKPDwgL2NpZDEgMTYgMCBSIC9jaWQxMCAxNyAwIFIgL2NpZDE1IDIxIDAgUiAvY2lkMTcgMjIgMCBSIC9jaWQxOCAyMyAwIFIKL2NpZDE5IDI0IDAgUiAvY2lkMjEgMjUgMCBSIC9jaWQyMyAyNyAwIFIgL2NpZDI1IDI4IDAgUiAvY2lkMzYgMzMgMCBSCi9jaWQzOSAzNCAwIFIgL2NpZDQ4IDM1IDAgUiAvY2lkNDkgMzYgMCBSIC9jaWQ1MSAzNyAwIFIgL2NpZDUzIDM4IDAgUgovY2lkOSAzOSAwIFIgPj4KZW5kb2JqCjMgMCBvYmoKPDwgL0YxIDE0IDAgUiA+PgplbmRvYmoKNCAwIG9iago8PCAvQTEgPDwgL0NBIDAgL1R5cGUgL0V4dEdTdGF0ZSAvY2EgMSA+PgovQTIgPDwgL0NBIDEgL1R5cGUgL0V4dEdTdGF0ZSAvY2EgMSA+PgovQTMgPDwgL0NBIDEgL1R5cGUgL0V4dEdTdGF0ZSAvY2EgMCA+PgovQTQgPDwgL0NBIDAuOCAvVHlwZSAvRXh0R1N0YXRlIC9jYSAwLjggPj4gPj4KZW5kb2JqCjUgMCBvYmoKPDwgPj4KZW5kb2JqCjYgMCBvYmoKPDwgPj4KZW5kb2JqCjcgMCBvYmoKPDwgL05hbnVtQmFydW5Hb3RoaWMtY2lkMTA5NSAxOCAwIFIgL05hbnVtQmFydW5Hb3RoaWMtY2lkMTE1NyAxOSAwIFIKL05hbnVtQmFydW5Hb3RoaWMtY2lkMTE2MyAyMCAwIFIgL05hbnVtQmFydW5Hb3RoaWMtY2lkMjIzMCAyNiAwIFIKL05hbnVtQmFydW5Hb3RoaWMtY2lkMjYwNyAyOSAwIFIgL05hbnVtQmFydW5Hb3RoaWMtY2lkMjY0MCAzMCAwIFIKL05hbnVtQmFydW5Hb3RoaWMtY2lkMjY2MCAzMSAwIFIgL05hbnVtQmFydW5Hb3RoaWMtY2lkMzMzMyAzMiAwIFIgPj4KZW5kb2JqCjIgMCBvYmoKPDwgL0NvdW50IDEgL0tpZHMgWyAxMCAwIFIgXSAvVHlwZSAvUGFnZXMgPj4KZW5kb2JqCjQwIDAgb2JqCjw8IC9DcmVhdGlvbkRhdGUgKEQ6MjAxODExMjYxNzI3MDcrMDknMDAnKQovQ3JlYXRvciAobWF0cGxvdGxpYiAzLjAuMCwgaHR0cDovL21hdHBsb3RsaWIub3JnKQovUHJvZHVjZXIgKG1hdHBsb3RsaWIgcGRmIGJhY2tlbmQgMy4wLjApID4+CmVuZG9iagp4cmVmCjAgNDEKMDAwMDAwMDAwMCA2NTUzNSBmIAowMDAwMDAwMDE2IDAwMDAwIG4gCjAwMDAwMTM2MDAgMDAwMDAgbiAKMDAwMDAxMzA2MCAwMDAwMCBuIAowMDAwMDEzMDkyIDAwMDAwIG4gCjAwMDAwMTMyNzMgMDAwMDAgbiAKMDAwMDAxMzI5NCAwMDAwMCBuIAowMDAwMDEzMzE1IDAwMDAwIG4gCjAwMDAwMDAwNjUgMDAwMDAgbiAKMDAwMDAwMDM5OCAwMDAwMCBuIAowMDAwMDAwMjA4IDAwMDAwIG4gCjAwMDAwMDI1OTIgMDAwMDAgbiAKMDAwMDAxMTc3MSAwMDAwMCBuIAowMDAwMDExNTY4IDAwMDAwIG4gCjAwMDAwMTExMjAgMDAwMDAgbiAKMDAwMDAxMjgxNiAwMDAwMCBuIAowMDAwMDAyNjEzIDAwMDAwIG4gCjAwMDAwMDI3MDIgMDAwMDAgbiAKMDAwMDAwMjk3MyAwMDAwMCBuIAowMDAwMDAzNDc5IDAwMDAwIG4gCjAwMDAwMDM4MjUgMDAwMDAgbiAKMDAwMDAwNDExMCAwMDAwMCBuIAowMDAwMDA0MzA0IDAwMDAwIG4gCjAwMDAwMDQ1OTUgMDAwMDAgbiAKMDAwMDAwNDgwMSAwMDAwMCBuIAowMDAwMDA1MTQ0IDAwMDAwIG4gCjAwMDAwMDUzMjIgMDAwMDAgbiAKMDAwMDAwNTY1NyAwMDAwMCBuIAowMDAwMDA2MzA1IDAwMDAwIG4gCjAwMDAwMDcwNDcgMDAwMDAgbiAKMDAwMDAwNzUxOCAwMDAwMCBuIAowMDAwMDA4MDc2IDAwMDAwIG4gCjAwMDAwMDg0NTQgMDAwMDAgbiAKMDAwMDAwOTA2MiAwMDAwMCBuIAowMDAwMDA5NDU0IDAwMDAwIG4gCjAwMDAwMDk2MDMgMDAwMDAgbiAKMDAwMDAxMDA1NiAwMDAwMCBuIAowMDAwMDEwNDAxIDAwMDAwIG4gCjAwMDAwMTA3MDYgMDAwMDAgbiAKMDAwMDAxMDg0NiAwMDAwMCBuIAowMDAwMDEzNjYwIDAwMDAwIG4gCnRyYWlsZXIKPDwgL0luZm8gNDAgMCBSIC9Sb290IDEgMCBSIC9TaXplIDQxID4+CnN0YXJ0eHJlZgoxMzgxNAolJUVPRgo=\n",
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "from sklearn.metrics import roc_curve\n",
    "fpr, tpr, thresholds = roc_curve(y_test, svc.decision_function(X_test))\n",
    "\n",
    "plt.plot(fpr, tpr, label=\"ROC 곡선\")\n",
    "plt.xlabel(\"FPR\")\n",
    "plt.ylabel(\"TPR (재현율)\")\n",
    "# 0 근처의 임계값을 찾습니다\n",
    "close_zero = np.argmin(np.abs(thresholds))\n",
    "plt.plot(fpr[close_zero], tpr[close_zero], 'o', markersize=10,\n",
    "         label=\"임계값 0\", fillstyle=\"none\", c='k', mew=2)\n",
    "plt.legend(loc=4)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.legend.Legend at 0x7ff3fc495b00>"
      ]
     },
     "execution_count": 64,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "application/pdf": "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\n",
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "from sklearn.metrics import roc_curve\n",
    "fpr_rf, tpr_rf, thresholds_rf = roc_curve(y_test, rf.predict_proba(X_test)[:, 1])\n",
    "\n",
    "plt.plot(fpr, tpr, label=\"SVC의 ROC 곡선\")\n",
    "plt.plot(fpr_rf, tpr_rf, label=\"RF의 ROC 곡선\")\n",
    "\n",
    "plt.xlabel(\"FPR\")\n",
    "plt.ylabel(\"TPR (재현율)\")\n",
    "plt.plot(fpr[close_zero], tpr[close_zero], 'o', markersize=10,\n",
    "         label=\"SVC 임계값 0\", fillstyle=\"none\", c='k', mew=2)\n",
    "close_default_rf = np.argmin(np.abs(thresholds_rf - 0.5))\n",
    "plt.plot(fpr_rf[close_default_rf], tpr[close_default_rf], '^', markersize=10,\n",
    "         label=\"RF 임계값 0.5\", fillstyle=\"none\", c='k', mew=2)\n",
    "\n",
    "plt.legend(loc=4)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "랜덤 포레스트의 AUC: 0.937\n",
      "SVC의 AUC: 0.916\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import roc_auc_score\n",
    "rf_auc = roc_auc_score(y_test, rf.predict_proba(X_test)[:, 1])\n",
    "svc_auc = roc_auc_score(y_test, svc.decision_function(X_test))\n",
    "print(\"랜덤 포레스트의 AUC: {:.3f}\".format(rf_auc))\n",
    "print(\"SVC의 AUC: {:.3f}\".format(svc_auc))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "gamma = 1.00  정확도 = 0.90  AUC = 0.50\n",
      "gamma = 0.10  정확도 = 0.90  AUC = 0.96\n",
      "gamma = 0.01  정확도 = 0.90  AUC = 1.00\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<matplotlib.legend.Legend at 0x7ff3fc329358>"
      ]
     },
     "execution_count": 66,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "application/pdf": "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\n",
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "y = digits.target == 9\n",
    "\n",
    "X_train, X_test, y_train, y_test = train_test_split(\n",
    "    digits.data, y, random_state=0)\n",
    "\n",
    "plt.figure()\n",
    "\n",
    "for gamma in [1, 0.1, 0.01]:\n",
    "    svc = SVC(gamma=gamma).fit(X_train, y_train)\n",
    "    accuracy = svc.score(X_test, y_test)\n",
    "    auc = roc_auc_score(y_test, svc.decision_function(X_test))\n",
    "    fpr, tpr, _ = roc_curve(y_test , svc.decision_function(X_test))\n",
    "    print(\"gamma = {:.2f}  정확도 = {:.2f}  AUC = {:.2f}\".format(\n",
    "        gamma, accuracy, auc))\n",
    "    plt.plot(fpr, tpr, label=\"gamma={:.2f}\".format(gamma))\n",
    "plt.xlabel(\"FPR\")\n",
    "plt.ylabel(\"TPR\")\n",
    "plt.xlim(-0.01, 1)\n",
    "plt.ylim(0, 1.02)\n",
    "plt.legend(loc=\"best\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### 다중 분류의 평가 지표"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "정확도: 0.953\n",
      "오차 행렬:\n",
      "[[37  0  0  0  0  0  0  0  0  0]\n",
      " [ 0 39  0  0  0  0  2  0  2  0]\n",
      " [ 0  0 41  3  0  0  0  0  0  0]\n",
      " [ 0  0  1 43  0  0  0  0  0  1]\n",
      " [ 0  0  0  0 38  0  0  0  0  0]\n",
      " [ 0  1  0  0  0 47  0  0  0  0]\n",
      " [ 0  0  0  0  0  0 52  0  0  0]\n",
      " [ 0  1  0  1  1  0  0 45  0  0]\n",
      " [ 0  3  1  0  0  0  0  0 43  1]\n",
      " [ 0  0  0  1  0  1  0  0  1 44]]\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import accuracy_score\n",
    "X_train, X_test, y_train, y_test = train_test_split(\n",
    "    digits.data, digits.target, random_state=0)\n",
    "lr = LogisticRegression(solver='liblinear', multi_class='ovr').fit(X_train, y_train)\n",
    "pred = lr.predict(X_test)\n",
    "print(\"정확도: {:.3f}\".format(accuracy_score(y_test, pred)))\n",
    "print(\"오차 행렬:\\n{}\".format(confusion_matrix(y_test, pred)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "metadata": {
    "hide_input": false
   },
   "outputs": [
    {
     "data": {
      "application/pdf": "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\n",
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "scores_image = mglearn.tools.heatmap(\n",
    "    confusion_matrix(y_test, pred), xlabel='예측 레이블',\n",
    "    ylabel='진짜 레이블', xticklabels=digits.target_names,\n",
    "    yticklabels=digits.target_names, cmap=plt.cm.gray_r, fmt=\"%d\")    \n",
    "plt.title(\"오차 행렬\")\n",
    "plt.gca().invert_yaxis()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       1.00      1.00      1.00        37\n",
      "           1       0.89      0.91      0.90        43\n",
      "           2       0.95      0.93      0.94        44\n",
      "           3       0.90      0.96      0.92        45\n",
      "           4       0.97      1.00      0.99        38\n",
      "           5       0.98      0.98      0.98        48\n",
      "           6       0.96      1.00      0.98        52\n",
      "           7       1.00      0.94      0.97        48\n",
      "           8       0.93      0.90      0.91        48\n",
      "           9       0.96      0.94      0.95        47\n",
      "\n",
      "   micro avg       0.95      0.95      0.95       450\n",
      "   macro avg       0.95      0.95      0.95       450\n",
      "weighted avg       0.95      0.95      0.95       450\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print(classification_report(y_test, pred))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "micro 평균 f1 점수: 0.953\n",
      "macro 평균 f1 점수: 0.954\n"
     ]
    }
   ],
   "source": [
    "print(\"micro 평균 f1 점수: {:.3f}\".format(f1_score(y_test, pred, average=\"micro\")))\n",
    "print(\"macro 평균 f1 점수: {:.3f}\".format(f1_score(y_test, pred, average=\"macro\")))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### 회귀 평가 지표"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### 모델 선택에서 평가 지표 사용하기"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "사이킷런 0.20 버전에서 `SVC` 클래스의 `gamma` 매개변수 옵션에 `auto`외에 `scale`이 추가되었습니다. `auto`는 `1/n_features`, 즉 특성 개수의 역수입니다. `scale`은 `1/(n_features * X.std())`로 스케일 조정이 되지 않은 특성에서 더 좋은 결과를 만듭니다. 사이킷런 0.22 버전부터는 `gamma` 매개변수의 기본값이 `auto`에서 `scale`로 변경됩니다. 서포트 벡터 머신을 사용하기 전에 특성을 표준화 전처리하면 `scale`과 `auto`는 차이가 없습니다. 경고를 피하기 위해 명시적으로 `auto` 옵션을 지정합니다."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "기본 평가 지표: [0.9 0.9 0.9]\n",
      "정확도 지표: [0.9 0.9 0.9]\n",
      "AUC 지표: [0.994 0.99  0.996]\n"
     ]
    }
   ],
   "source": [
    "# 분류의 기본 평가 지표는 정확도 입니다\n",
    "print(\"기본 평가 지표: {}\".format(\n",
    "      cross_val_score(SVC(gamma='auto'), digits.data, digits.target == 9, cv=3)))\n",
    "# scoring=\"accuracy\"의 결과는 같습니다.\n",
    "explicit_accuracy =  cross_val_score(SVC(gamma='auto'), digits.data, digits.target == 9,\n",
    "                                     scoring=\"accuracy\", cv=3)\n",
    "print(\"정확도 지표: {}\".format(explicit_accuracy))\n",
    "roc_auc =  cross_val_score(SVC(gamma='auto'), digits.data, digits.target == 9,\n",
    "                           scoring=\"roc_auc\", cv=3)\n",
    "print(\"AUC 지표: {}\".format(roc_auc))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "사이킷런 0.22 버전에서 `GridSearchCV` 클래스의 `cv` 매개변수 기본값이 3에서 5로 바뀔 예정입니다. 0.20 버전에서 `cv` 매개변수를 지정하지 않는 경우 이에 관한 경고 메세지가 출력됩니다. 경고 메세지를 피하기 위해 `cv` 매개변수 값을 명시적으로 3으로 지정합니다."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "정확도 지표를 사용한 그리드 서치\n",
      "최적의 파라미터: {'gamma': 0.0001}\n",
      "최상의 교차 검증 점수 (정확도)): 0.970\n",
      "테스트 세트 AUC: 0.992\n",
      "테스트 세트 accuracy: 0.973\n"
     ]
    }
   ],
   "source": [
    "X_train, X_test, y_train, y_test = train_test_split(\n",
    "    digits.data, digits.target == 9, random_state=0)\n",
    "\n",
    "# 일부러 적절하지 않은 그리드를 만듭니다\n",
    "param_grid = {'gamma': [0.0001, 0.01, 0.1, 1, 10]}\n",
    "# 기본 정확도 측정 지표를 사용합니다\n",
    "grid = GridSearchCV(SVC(), param_grid=param_grid, cv=3)\n",
    "grid.fit(X_train, y_train)\n",
    "print(\"정확도 지표를 사용한 그리드 서치\")\n",
    "print(\"최적의 파라미터:\", grid.best_params_)\n",
    "print(\"최상의 교차 검증 점수 (정확도)): {:.3f}\".format(grid.best_score_))\n",
    "print(\"테스트 세트 AUC: {:.3f}\".format(\n",
    "        roc_auc_score(y_test, grid.decision_function(X_test))))\n",
    "print(\"테스트 세트 accuracy: {:.3f}\".format(grid.score(X_test, y_test)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "AUC 지표를 사용한 그리드 서치\n",
      "최적의 파라미터: {'gamma': 0.01}\n",
      "최상의 교차 검증 점수 (AUC): 0.997\n",
      "테스트 세트 AUC: 1.000\n"
     ]
    }
   ],
   "source": [
    "# AUC 지표 사용\n",
    "grid = GridSearchCV(SVC(), param_grid=param_grid, scoring=\"roc_auc\", cv=3)\n",
    "grid.fit(X_train, y_train)\n",
    "print(\"AUC 지표를 사용한 그리드 서치\")\n",
    "print(\"최적의 파라미터:\", grid.best_params_)\n",
    "print(\"최상의 교차 검증 점수 (AUC): {:.3f}\".format(grid.best_score_))\n",
    "print(\"테스트 세트 AUC: {:.3f}\".format(grid.score(X_test, y_test)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "가능한 평가 방식:\n",
      "['accuracy', 'adjusted_mutual_info_score', 'adjusted_rand_score', 'average_precision', 'balanced_accuracy', 'brier_score_loss', 'completeness_score', 'explained_variance', 'f1', 'f1_macro', 'f1_micro', 'f1_samples', 'f1_weighted', 'fowlkes_mallows_score', 'homogeneity_score', 'mutual_info_score', 'neg_log_loss', 'neg_mean_absolute_error', 'neg_mean_squared_error', 'neg_mean_squared_log_error', 'neg_median_absolute_error', 'normalized_mutual_info_score', 'precision', 'precision_macro', 'precision_micro', 'precision_samples', 'precision_weighted', 'r2', 'recall', 'recall_macro', 'recall_micro', 'recall_samples', 'recall_weighted', 'roc_auc', 'v_measure_score']\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics.scorer import SCORERS\n",
    "print(\"가능한 평가 방식:\\n{}\".format(sorted(SCORERS.keys())))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 요약 및 정리"
   ]
  }
 ],
 "metadata": {
  "anaconda-cloud": {},
  "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": "3.5.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}