Demo_03_Decision_Trees.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Decision Trees\n",
    "\n",
    "In this notebook, we go over how to grow decision trees, and look at some of the different paramaters that we can pass to the modules within SciKit-Learn"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.datasets import load_iris\n",
    "from sklearn import tree\n",
    "from sklearn.model_selection import train_test_split\n",
    "import numpy as np\n",
    "iris_attributes, iris_classes = load_iris(return_X_y=True)\n",
    "attributes_train, attributes_test, classes_train, classes_test = train_test_split(iris_attributes, iris_classes, test_size=0.2, random_state=50)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "First, similar to naive bays, we create the classifier and assign it to a variable"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "Tree1 = tree.DecisionTreeClassifier()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "Tree1  = Tree1.fit(attributes_train, classes_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "ename": "SyntaxError",
     "evalue": "invalid syntax (<ipython-input-20-21bfa32fe9b0>, line 3)",
     "output_type": "error",
     "traceback": [
      "\u001b[1;36m  File \u001b[1;32m\"<ipython-input-20-21bfa32fe9b0>\"\u001b[1;36m, line \u001b[1;32m3\u001b[0m\n\u001b[1;33m    R_Tree1 =\u001b[0m\n\u001b[1;37m              ^\u001b[0m\n\u001b[1;31mSyntaxError\u001b[0m\u001b[1;31m:\u001b[0m invalid syntax\n"
     ]
    }
   ],
   "source": [
    "#the random splitter choses where to split randomly, so we expect performance to degrade\n",
    "R_Tree1 = tree.DecisionTreeClassifier(splitter=\"random\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[Text(133.92000000000002, 199.32, 'X[2] <= 2.45\\nentropy = 0.666\\nsamples = 120\\nvalue = [41, 38, 41]'),\n",
       " Text(100.44000000000001, 163.07999999999998, 'entropy = 0.0\\nsamples = 41\\nvalue = [41, 0, 0]'),\n",
       " Text(167.40000000000003, 163.07999999999998, 'X[3] <= 1.65\\nentropy = 0.499\\nsamples = 79\\nvalue = [0, 38, 41]'),\n",
       " Text(66.96000000000001, 126.83999999999999, 'X[2] <= 4.95\\nentropy = 0.176\\nsamples = 41\\nvalue = [0, 37, 4]'),\n",
       " Text(33.480000000000004, 90.6, 'entropy = 0.0\\nsamples = 36\\nvalue = [0, 36, 0]'),\n",
       " Text(100.44000000000001, 90.6, 'X[0] <= 6.05\\nentropy = 0.32\\nsamples = 5\\nvalue = [0, 1, 4]'),\n",
       " Text(66.96000000000001, 54.359999999999985, 'X[1] <= 2.45\\nentropy = 0.5\\nsamples = 2\\nvalue = [0, 1, 1]'),\n",
       " Text(33.480000000000004, 18.119999999999976, 'entropy = 0.0\\nsamples = 1\\nvalue = [0, 0, 1]'),\n",
       " Text(100.44000000000001, 18.119999999999976, 'entropy = 0.0\\nsamples = 1\\nvalue = [0, 1, 0]'),\n",
       " Text(133.92000000000002, 54.359999999999985, 'entropy = 0.0\\nsamples = 3\\nvalue = [0, 0, 3]'),\n",
       " Text(267.84000000000003, 126.83999999999999, 'X[2] <= 4.85\\nentropy = 0.051\\nsamples = 38\\nvalue = [0, 1, 37]'),\n",
       " Text(234.36, 90.6, 'X[1] <= 3.1\\nentropy = 0.375\\nsamples = 4\\nvalue = [0, 1, 3]'),\n",
       " Text(200.88000000000002, 54.359999999999985, 'entropy = 0.0\\nsamples = 3\\nvalue = [0, 0, 3]'),\n",
       " Text(267.84000000000003, 54.359999999999985, 'entropy = 0.0\\nsamples = 1\\nvalue = [0, 1, 0]'),\n",
       " Text(301.32000000000005, 90.6, 'entropy = 0.0\\nsamples = 34\\nvalue = [0, 0, 34]')]"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAV0AAADnCAYAAAC9roUQAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjAsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+17YcXAAAgAElEQVR4nOy9e1xVVdrA/z2QgoaKonHUmZpIp16r6aJOoTNq7w+7eOmnP18HLRFU0MMIIohIonI0VMhLomMgCtigoRajopNNVryKIGrzJqRTCiRMykFJBS/cBNbvjzNsOdwvh8uB9f181uew9157XR7Wes4+z37Ws1RCCCQSiUTSNpi1dwMkEomkKyGVrkQikbQhUulKJBJJGyKVrkQikbQhUulKJBJJGyKVrkQikbQhUulKJBJJG/JIezdA0nh69OiRW1xcbNve7egoWFpaXi8qKlK3dzskkqagkosjTAeVSiXk/+shKpUKIYSqvdshkTQFaV6QSCSSNkQqXYlEImlDpNLt5Bw5coSYmBju3LmDm5sbWVlZODg4kJOTw65du3B1dWXGjBmUlJRw8uRJ3nzzzTrLKi0tbbC+6mVWJSUlhVdffRUArVaLs7Mzvr6+LeugRGJiSKXbyZk8eTJJSUn4+fnx3nvvAeDg4MCgQYNwdXVl165d2Nvbk5GRwZgxY/jNb35To4zLly/z3nvvsWzZsgbrq15mJeXl5Xz++ee8+OKLAHTr1g1zc3MGDBhgnI5KJCaCVLpdgHHjxqHT6bCzs6tx7c6dO1y4cIFhw4bVeu+ECRPYv38/3t7efPjhhwDs3LmTxYsXK+nKlSsNlrlz506cnZ2V4+XLlxMVFcX9+/dJT083RjclEpNAKt1OTnFxMceOHWP06NGcOXOmxjVfX1/Wr1+PSlW7E8DSpUvJzc0lNDRUUY4PHjyguLhYSRUVFQ2WmZqayoYNG0hMTOTo0aPKtf79+3P//n1jd1si6bgIIWQykaT/dzWNNWvWiNTUVFFaWipmzZolMjIyxPr164UQQnh5eYm33npLLFiwQFy6dEkIIcSCBQtqLScnJ0ds2rSpwfqqlxkeHi5SU1OV65Xlb9iwQWg0GuHq6ioqKiqa3C8hhPiPPNr9/yKTTE1J0k/XhDCGn25WVhb79u3D39+/1usajYbw8PAW1dFWSD9diSkizQtdDEtLSy5fvkxOTk6NaydPnqR79+7t0CqJpOsgn3RNiJY+6Wq1WjQaDWp1666cvXLlCqtWrcLMzIytW7fSp08fAG7fvo1Wq6WiogI/Pz+srKzqPf71r39dbz3ySVdiisjYC52IjIwMtmzZQlFRERqNhr///e8AZGdns3r1ar799ls2btzIn//8Z2bMmMGUKVOYOHEimzZtAmDDhg2EhYVRVlZGQUEBnp6ehIeHs27dOpKTk8nNzeWdd95psB27d+9m8+bNZGZmcvDgQVxcXAC9B4OlpSUlJSX069eP7du313sskXRGpNLtRERERNC3b1+sra357rvvAJg9ezY5OTmcOnWKESNGKE+6Q4cOZfny5fj4+LB9+3ays7PZv38/AE5OTqjVakJCQpg+fTpxcXGkpKSwYcMGpa64uDgSExOVYycnJ4YPHw5AXl4eAwYMoLi4mISEBCVPZmYm7u7uFBcXs3fv3gaP58+f3xZik0jaFKl0OxHl5eV4eHhga6sPRKbVaunZsyePPPIIpaWlmJk9NOH36tVL+bvqeYCysjIePHgAgL29PZGRkVhbW2NpaWmQp7i42KDuSvr3709eXh45OTkMHDhQOa9Wq7G2tqa4uJj79+83eCyRdEak0u1EuLu74+fnh42NDa+//nqN6y+99BIrVqww8FxwcXHB09OT8vJygoOD2bFjB9HR0Vy/fp2AgABAryynTJliUJajoyOOjo61tsPFxYWlS5cCsGXLFnbs2IG9vT1z5sxBq9UihGD9+vWUlpbWeyyRdEbkizQToi1CO1Z/2Xb8+HEOHjzIRx991Kr1Ngf5Ik1iikila0LIeLqGSKUrMUWkn66kTjQajVHK2b17Ny+//DK5ubkAhISEMHfuXNzc3ACIj49n/vz5MuKYpEsgbbqdkJiYGJKSkrCzs0Oj0RASEkJeXh6urq4UFhYSFhaGlZUVQ4cO5datW/Tp04eAgABGjhyJo6MjRUVFrFy5Uilv5cqVFBQUMHjwYCZNmkRQUBBqtZp169bRo0ePBtvj4uJCVlaWclwZrczb25v79+9z7NgxIiIiCAkJISsrq9ZIZxJJZ0E+6XZCcnJyGDlyJK6urpibm1NRUYFarWbfvn0AjB49mrCwMM6ePcsHH3zAzz//DMCQIUPw9fUlPz+foqIiAC5evEhaWhr9+vUjMzOTgoIC1Go1Hh4eBgo3JCTEIPJYQUFBvW28evUqAI8++qgS/GbQoEHodDqjy0Mi6UjIJ91OiJ+fH2fOnMHNzY2ZM2cyduxYhg8frjy9PvbYY3Tv3p3HHnvM4L6ysjKDT4CKigpGjRplEEvX1tYWrVZLUFAQTzzxBKAPcF7Vhaw+2/PNmzdZs2YNoaGhBnlzcnL44x//2JKuSyQdHql0OyGRkZH88MMP2Nra8sILLxAUFKQslqiP7Oxs/P396dWrl/IU+/zzzxMZGYmvry9WVla89tprHDp0iPLycqysrJR7q5ojqhMfH8/Ro0e5cuUKoaGhzJs3DwsLC7y9vVm7di1vvPEG7u7uPProo9K0IOn0SO8FE6K1vRdMKcIYSO8FiWkila4JIV3GDJFKV2KKyBdpEolE0oZIpdsJMZZ/Leg3sUxOTgagoKCAp59+mtzcXNLT03F0dCQ4OLjW+y5cuIBGo2HSpEmcP3+en376iSlTpjB9+nS+/fbbWu+p6i52+vRpJkyYoHhcpKWlKTsJSySmjFS6JoiHhwdlZWUkJycTHR3N4cOHWbJkibLbbyWVyrfy88MPP2Tx4sX4+fk1uq4hQ4YwatQoAD766CMmTZoEwNChQwkJCanzvueee47w8HDee+890tLSuHz5MnPmzGH58uWkpKTUek9kZCTjx48H9IF2qrbzd7/7nbKTsERiykjvBRNk3LhxnDhxgq+++gpfX19Onz6NhYUFSUlJ3Llzp9Z77t69S3x8PGPHjiUzM5OSkhIsLCwA/WKFkpISAMzNzZX4ulX55z//iZ2dnZKvMXzyySds27aNv/71r1hbW/Puu+9SVlbGrl27auQ9fPgwb7zxBnv27Gl0+RKJKSKfdE2QCRMm8MUXX3D79m1sbGw4ePAg69at47nnnqOwsFDJV7nooLCwECEEw4YNQ6vVEhMToyhcgJKSEoPdfWvj5MmTJCQkcPToUSIjIxvVznfeeYcjR46wY8cOPvnkE4KDgzl69GitHhLnzp1jz549JCYm8vHHHzdFHBKJSSGfdE2Qnj17kpubq/zst7W1JTg4mO+//94gn4WFBZs2bUKn09G7d2/UajU+Pj6UlZWxdetWJd+WLVsarNPb2xvQRyGbN28eeXl5BAQEkJmZyQsvvMDLL7/MyZMnmT59OgD/+7//y2effcbdu3dxd3fH0tKS4OBgunXrxowZM0hLSyM/P58xY8YAEBQUBOhNIc7Ozvz4449s3ryZoqIi7Ozs+P3vf99ywUkkHQDpMmZCtIfLWGN9d7/66it69erFK6+80qhyd+/ezfjx4xk8eHCz2yJdxiSmiHzSldSLmZkZycnJylN1XTg4ODSp3Mp90xpLWlqagelEIjFV5JOuCdGjR4/c4uJi2/ZuR0fB0tLyelFRUetubSyRGBmpdLsQKpVqLPApMFEIca6d2jAU6C6EuNge9Usk7Y1Uul0AlUoVBuwBDgIzhRBft3OTJJIui1S6nRyVSvVrIBUoBjyFEHHt3CSJpEsjlW4nR6VS+QFrgQygTAjxvLHK7so2ZmlPljQXqXQ7OSqV6ghgB6wCjgohGr+krOGyu2zUM+muJmkuUulKmo1UulLpSpqOXAYsaROOHDlCTEwMd+7cwc3NjaysLBwcHMjJySEuLo758+czZcoU7t+/z8mTJ3nzzTfrLKu0tLRR9Y0ZM6ZGcJ3bt2/j5eWFp6cnP//8M1qtFmdnZ7kTsaTNkIsjmklXsmcaw345efJkNBoNSUlJSjQ0BwcHBg0axLRp05g2bRrr1q3jl19+YcyYMXzyySc1yrh8+TLR0dEUFxfz4YcfNljfzZs3a5zfuXMnlpaWlJSU0K9fP7p164a5uTkDBgxoSfckkkYjlW4zKS4utu0qP61VKpVRvlzGjRtHbGwsdnZ2BluyAwQEBPDdd9+xePHiWu+dMGEC9vb2eHt7Kxtq7ty5k4sXH7r7enl58eSTT9bbhszMTNzd3SkuLmbv3r0sX74clUrFqlWrSE9PZ+jQoS3rpETSANK80E5otVpyc3NbvZ4rV67g5OSEs7Ozwbbo586dw8XFBTc3N8rLy1u9HcXFxRw7dozRo0dz5syZGtfXrl3L3Llz+eabb2q9f+nSpeTm5hIaGkp6ejoADx48MIiOVlFR0WA71Go11tbWWFtbc//+fSUSW//+/bl//34LeiiRNBIhhEzNSHrR1U16erpYuHChmDt3rjh79qwIDAwUgYGBwsXFRWRnZ4uJEyeKJUuWiMzMTDFy5Eixdu1acf78eeHk5CScnJxEbm6uCAwMFAEBAcLDw0NcunRJeHt7i6KiIvH111+LvXv31lt/JatWrRI3btwQp0+fFtHR0cp5Dw8PUVZWJmJjY0VCQkK9Zfynr02WQVXWrFkjUlNTRWlpqZg1a5bIyMgQ69evF0IIERERIRYuXChmzZolbt68KYQQYsGCBbWWk5OTIzZt2tRgfadOnRLjxo0TU6dOFRkZGSI8PFykpqaKK1euCGdnZzF79mxx7do1sWHDBqHRaISrq6uoqKhodH/qkolMMjWUpHmhlYiIiKBv375YW1sr25/Pnj2bnJwcTp06xYgRI9BoNKjVaoYOHcry5cvx8fFh+/btZGdns3//fgCcnJxQq9WEhIQwffp04uLiSElJYcOGDUpdcXFxJCYmKsdOTk4MHz4cgLy8PAYMGEBxcTEJCQlKnvLycszNzRk8eDBXr15tdXlU3aI9JibGwLzg5ubW6HIGDhyIj49Pg/lGjx5t0N+nnnpK+Xv37t3K3/IFmqStkUq3lSgvL8fDwwNbW705VKvV0rNnTx555BFKS0sxM3to2enVq5fyd9XzAGVlZTx48ADQb2ETGRmJtbU1lpaWBnmqBh+vai7o378/eXl55OTkMHDgQIN6ysvLa5xvKywtLbl8+TI5OTkMGjTI4NrJkyfp3r17m7dJImkLpJ9uM2nIRzUjI4P3338fGxsbXn/9dVJSUtBoNGRlZfHjjz9iY2PD4cOH8ff3Z+PGjYSHh5OWlsaWLVsoLy8nODiYHTt2cO/ePa5fv05AQADPPPMMK1asYMqUKYwYMaJR7fzpp59Ys2YNoA9Wvn//fuzt7SkqKiIiIgJzc3PCwsIwNzevr6+IWnxSm+unq9Vqlaf81uTKlSusWrUKMzMztm7dSp8+fZRrKSkpLF68mJSUFGJiYvj666+xsrJi69at7N271+C4+hchSD9dSQtob/uGqSaaYM9sLoGBgUKn0ynHX375pXB3d2/1eqtDE226Hd2eXVZWJlauXKnYjd955x1RXl4u/vKXv4gTJ07UOG6KTGSSqaEkzQsdGK1Wa3A8fvx4ZbfcjkxHt2fv3LkTZ2dnpRx3d3c8PT0pKSnBxsamxrFEYkyky1gHp3L7dGMQFxfHtGnTAP3LpJdffrlV3NYq7dlBQUHMnz8fwCj27ISEBCwsLGq1Z1emxtizU1NT2bBhA4mJiRw9epQ//OEPbN++nRdffJGnnnqqxrFEYkzkk24rERMTQ1JSEnZ2dmg0GkJCQsjLy8PV1ZXCwkLCwsKwsrJi6NCh3Lp1iz59+hAQEMDIkSNxdHSkqKjI4I3/ypUrKSgoYPDgwUyaNImgoCDUajXr1q2jR48eDbbn7t27pKenKyuvXFxcaixQMBbu7u74+fkp9uzqvPTSS6xYsQJ/f3/lnIuLC56engb27OjoaMWeDXof2ylTphiU5ejoiKOjY63tcHFxYenSpYDenr1jxw7s7e0JCwsD9F9okyZNIj4+nmPHjmFlZYWHh0eNY4nEqLS3fcNUEw3YdIODg8WuXbvEzZs3xb1794S/v79YuXKl8Pb2FgkJCSI0NFSUlJSIqVOnCiEe+qXOmDFDCCGEj4+PKCwsFAsWLBAXLlwQb7/9tggMDBRubm4iKSlJLF68WGRkZNSo08vLS0n5+fnKtbVr14rbt28b+L9WtxnXBUbw020qHcWeXRd1yUQmmRpK8km3lfDz8+PMmTO4ubkxc+ZMxo4dy/Dhw5Wn18cee4zu3bsrS1orKSsrM/gEqKioYNSoUSxbtkw5Z2tri1arJSgoiCeeeALQB4Kp6jomxEPPgvPnz+Pv709iYiKnT5/G3t7e+J02IqZqz5ZIGkIq3VYiMjKSH374AVtbW1544QWCgoKUl0r1kZ2djb+/P7169VLMBs8//zyRkZH4+vpiZWXFa6+9xqFDhygvL8fKykq5t6o5ojoHDhwA9D+p7e3tiY+P5+jRo1y5coXQ0FCsra1b2OPWp7HbwTfERx99RFpaGklJSRw4cICjR49y6dIl1Go1QUFBRmipRFI30k+3mbRWLFljKRZj0lI/XWPZtytl01L7diXvvvsue/fuZebMmcTGxrJ06VI8PDyUXw7NkYlE0hDSe6GD0dEUrjHIyclh5MiRuLq6Ym5uTkVFBWq1mn379gH6JbthYWGcPXuWDz74gJ9//hmAIUOG4OvrS35+PkVFRQBcvHiRtLQ0+vXrR2ZmJgUFBajVajw8PAwUbkhICIsXL1ZS1WA/AN9//z2/+93vAPjTn/7EokWLyM7ORqfTtYVIJF0YqXQlrY6fnx/PPvssbm5uHDt2jLFjx+Lp6UlhYSHQPPu2VqslIiKCUaNG4eHhgVarJTs7W8lXad+uTNWfyGNjY5k5cyYAU6dOZevWrTz55JONesqVSFqCVLqtjDH9bB0cHEhOTgagoKCAp59+mtzcXNLT03F0dCQ4OLjW+/Lz83F2dsbJyclAMVVSPfxjQzs3NJXIyEg+/fRTxb4dGxvLrl27GryvLvu2TqfD19cXrVbLiRMn+Mtf/lKrfTs8PFxJVW3WQgiuXr3K448/DkBUVBTu7u707du3XeJQSLoY7e0+YaqJ/7hLLVy4UDx48EAkJSWJqKgocejQIeHj4yP8/f2FEA9dwap/bt68WXh5eYmlS5eKxlLV3WvdunXCx8dHcau6cuWKEiqxOlFRUeLs2bNCp9OJwMDAGtdrWy5btS7awWWsehs6GnXJRCaZGkrSe6GFjBs3jhMnTvDVV1/h6+vL6dOnsbCwICkpiTt37tR6z927d4mPj2fs2LFkZmZSUlKChYUFAMuWLaOkRL9hr7m5OZs2bapx/z//+U/s7OyUfA2h0+kYP348arWaGzdu1Lhe13LZ9qYz2rclEql0W8iECRMIDAzk7t272NjYcPDgQSIjI1m4cKFiswSUHQoKCwsRQjBs2LAavqgAJSUliq9tXZG/Tp48yaVLl/j222/p1q2bsmKrLtRqNTqdjm7dutWwm0Ldy2UlEonxkTbdFtKzZ09yc3N54YUXAP2iheDgYL7//nuDfBYWFmzatAmdTkfv3r1Rq9X4+PiwaNEig3xbtmxR7JDbt2+vtU5vb2/Cw8OZNGkS8+bNIy8vj4CAAA4dOsSxY8e4fv06n376qZJ/6tSpbN++HT8/P1xcXEhLS+PkyZPK9crlsmFhYTWW2RqD1rBrN2SnPnHiBBqNhokTJ3Lt2jVSUlL4n//5H6ZOncq///3vWstOSUnh1VdfBWruJmxsO7ekC9Pe9g1TTbRBaMfqNNbGefz4cZGSklLn9ejoaHH16tVG10U1+yWgAv5AO9q1G7JTV7Jnzx5x4sQJERYWJs6fPy/+9re/iSNHjtTIVz3coxB6OZ0+fbpOmQDdRRuOOZk6R5LmBRPCzMyM5ORkRo0aVW8+BweHeq+7uLjUe72unRtUKtXTwKz/JMV20h527Ybs1AChoaHExcXx2WefoVarWbhwIRUVFXz22Wc18lYP99hIrqlUqgPAHiBFCCFXGkkaRCrdZmJpaXndWFuTN4XKCFltwbZt2wCwtLS8q1KpzgG/AmKB/w84D1RA+9i1G7JTg35L9tGjR3PgwAF++ukn9u3bx40bN/jkk09YuHChQd7U1FTOnz+vhHucNGlSY0T0e+AdIBp4RKVS7QH2CiHSG3OzpGsilW4zKSoqat29ZtoRlUrVE/h/0T/Rji4uLj4CxADfCCHKquQDHtq1K5/Am2rXLisrY+vWrUq+LVu2NNjGqVOn4u3tjRCC1atXk5aWRn5+PmPGjAH0sYMTEhLIz89n9erVZGRksHjxYkpLS/Hz8+P48eM8/vjjPP300wA1wj0mJSXx8ccf07dvXzZs2FBrXF0hxBVgrUqlWgcMB5yAUyqV6if0T7/7hRC/NCxxSZeive0bMnWMBJgDDsBu4DbwBfAuYFXPPaKtqcuu3Rg7dVU2bdokCgsLm103dfsudwPeAj4B8oF44E9Aj9ryy9T1kvRe6OKoVKoXVCrVBuDfQAiQCvyXEOJNIcReIcS99m2hIZV27eq4uLgwePDgRpfj4+PTpOA4jd2hWAjxQAhxTAjxDvBr4DPADb39N1KlUr2mUqnkvOvCyChjXRCVSvUr9LbIWUAfYC+wRwjxr6aU06NHj9zi4uI2t2t3BCwtLa83xcSkUqkGAzPQmyBseCjzC63UREkHRSrdLoJKpeoNTEOvaF8C4tDbaU8JISras21dDZVK9Tz6/8O7QB56+2+sECKnXRsmaROk0u3EqFSqbsAb6Cf4W0ACekX7dyFEcX33SloflUplDoxB//Q7FTiHXgH/raOZdSTGQyrdToZK71Lwe/SK1hFIRz+RDwghbrZn2yR1o1KpegCT0SvgPwJ/R/9/Oy6qeIxITB+TVLpdyZbYWNuhSqV6Cv3P1Vn/OVXpM5rZmu2TGB+VSjUAvceDE/AbYB/6/+c/RSMmbFeaH9B0+3p7Y5JKt7W2yumIVN8W5j/K9QkhxDcqlcoG/eScBQwF9qM3H5zrMgLq5KhUqqE8XAVYysMv0yyVSvUs0EcIkVztni717ze1rZOk0u3gVB1QKpXqN8AJ4G/Ak8BrwDH0E/EfQogH7dRMSSvzH7PRq+iffv8E/Av9WHAHpgkhTlTJ22XmB5ie0u2U/oJHjhwhJiaGO3fu4ObmRlZWFg4ODuTk5LBr1y5cXV2ZMWMGJSUlDUaPKi0tbVSdVSNUVRITE4OLiwseHh5UVFSg1WpxdnbG19e3yX1SqVSPo/ehHYg+2MzfgV8LIWYIIY5Khdu5EXpOCyH+DAwCNgKvAL2Ar1Qq1cSmlNdR5sihQ4d45513mDZtGnfv3sXV1RWNRsP69eub0h2TolMq3cmTJ5OUlISfnx/vvfceoA8CM2jQIFxdXdm1axf29vZkZGQwZswYfvOb39Qo4/Lly7z33nssW7aswfrKy8v5/PPPefHFFw3Of/HFF0RFRfFf//VfnDp1im7dumFubs6AAQOa0y1z4FvgK/RRvgqEELVHk5F0aoQQpcAVoD/wDXCS/8TBaCwdZY6cOXOGjz76iD/84Q9kZ2djaWlJRUUFtrad1yTdaWMvjBs3jtjYWOzs7MjKyjK4dufOHS5cuFAjlm0lEyZMwN7eHm9vbyWYys6dO7l48aKSx8vLiyeffFK5VluEKnd3dzw9PSkpKcHGxobly5ejUqlYtWoV6enpDB06tNH9Efp1/v9Po2+QdGqEEN8DL7ekjI4wRyZNmsSf/vQnunXrxsKFC9m2bRsqlUp50u7Zs2dLutgh6ZRPusXFxRw7dozRo0dz5syZGtd8fX1Zv369ErClOkuXLiU3N5fQ0FDS0/UBox48eGCwu2xFxcMHi9TUVDZs2KBEqKrkD3/4A9u3b+fFF1/kqaeeUurr378/9+/fN3a3JZJG01HmyM6dO/niiy/w8PDg+PHjSn19+vRp9HZUJkd7B39oTqKBQCtr1qwRqamporS0VMyaNUtkZGQomzZ6eXmJt956SyxYsEBcunRJCFF3EJWcnByxadOmeuuqSmU54eHhIjU1VRw+fFhoNBrh6+srhBBiw4YNQqPRCFdXV1FRUdGoMpEbIMrUxNTQ/BCi48yRqKgoMW/ePDFt2jSRnZ0tli1bJhYsWCCWLVvW6DJNbY60ewOa1egmRreqb6dcIeSuszJ1rtTU+SGEnCNtmTqleaE6lpaWXL58mZycmkvbGxs9SiLpzMg50oa0t9ZvTqIJ3+SBgYFCp9M1On9z+emnn8SsWbPE7NmzRX5+vnI+Pj5e/PGPf1T22tq3b59YsGCBsLe3F59//rm4deuWWLRokfDw8BD//ve/a5SLiX2Ly9T+qSnzQ4j2nyNnz54Vzs7OwtXVVZSVlYl58+aJBQsWiHXr1gkh9Ga5559/vs5yTW2OmLz3QkZGBlu2bKGoqAiNRsPf//53ALKzs1m9ejXffvstGzdu5M9//jMzZsxgypQpTJw4Udl3a8OGDYSFhVFWVkZBQQGenp6Eh4ezbt06kpOTyc3N5Z133mmwHbt372bz5s1kZmZy8OBBZR+yyZMnc/Pmw5AHjo6OODo6MmfOHBwcHPjwww+xtLSkpKSEfv36GV9Aki5PR58jf/3rX4mMjOTTTz8lMTERS0tLSktLFbcxX19fMjIyWkc47YDJK92IiAj69u2LtbU13333HQCzZ88mJyeHU6dOMWLECDQaDWq1mqFDh7J8+XJ8fHzYvn072dnZ7N+/HwAnJyfUajUhISFMnz6duLg4UlJSDFxc4uLiSExMVI6dnJwYPnw4AHl5eQwYMIDi4mISEhLqbfOtW7fo3bs33bp1IzMzE3d3d4qLi9m7dy/z5883togkXZyOPkfKy8sxNzdn8ODBXL16tdO7jZm80i0vL8fDw0P5VtRqtfTs2ZNHHnmE0tJSzMwemq179eql/F31PEBZWRkPHugXddnb2xMZGYm1tTWWltdUaN0AACAASURBVJYGeSo3TKysu5L+/fuTl5dHTk4OAwcOrLfNBw4cYPr06YB+g0Vra2uKi4ulG5mkVejoc8TMzIzy8nLlfHW3Mal0Oxju7u74+flhY2PD66+/XuP6Sy+9xIoVK/D391fOubi44OnpSXl5OcHBwezYsYPo6GiuX79OQEAAoFeGU6ZMMSir0jRQGy4uLixduhTQb6y4Y8cO7O3tuXv3bo0NDpOSkliwYAEAc+bMQavVIoTo1EsfJe1HR58jTk5OzJ8/H3Nzc8LCwvD39yc/Px9ra2v69u1LdHQ0iYmJLFy4kO3btxtLLO1HexuVm5Mw8oaI1V8kfPnll8Ld3d2odTQXTOwlgUztn4w9P4SQc8SYSUYZ6+CYWgQlSfvTleYHmN4c6RJ+us1Bo9EYray4uDimTZsG6N/gvvzyy+Tm5hqtfImkPTDWHNm5cyfz5s3j3XffRQjB9u3bmTNnDrNmzTJYStxZMHmbbnViYmJISkrCzs4OjUZDSEgIeXl5uLq6UlhYSFhYGFZWVgwdOpRbt27Rp08fAgICGDlyJI6OjhQVFbFy5UqlvJUrV1JQUMDgwYOZNGkSQUFBqNVq1q1b16gtvO/evUt6eroSWczFxaVGcBGJpC3paHPEzc0NNzc3Fi1aRGlpKRcvXiQqKgofHx/u3btH7969W1McbU6ne9LNyclh5MiRuLq6Ym5uTkVFBWq1mn379gEwevRowsLCOHv2LB988AE///wzAEOGDMHX15f8/HyKiooAuHjxImlpafTr14/MzEwKCgpQq9V4eHgYDKaQkBAWL16spIKCAuXatm3bjPrULJG0lI42RwDmzp3LL7/8Qrdu3fjv//5vJk6cSF5eXqdTuNAJla6fnx/PPvssbm5uHDt2jLFjx+Lp6UlhYSEAjz32GN27d1fC0VVSVlZm8AlQUVHBqFGj0Gq1REREMGrUKDw8PNBqtWRnZyv5SktLDaIrVbWnnT9/Hn9/fxITEzl9+nRrdl0iaRQdbY4AREVF8fLLL/Ovf/2L+Ph4Pv/8c0aPHm0QKrKz0OnMC5GRkfzwww/Y2trywgsvEBQUpDiE10d2djb+/v706tVL+YZ+/vnniYyMxNfXFysrK1577TUOHTpEeXk5VlZWyr1Vf2pV58CBA4De/mVvb098fDxHjx7lypUrhIaGYm1t3cIeSyRNo6PNkeDgYK5du0ZRURGLFi1i2LBhaDQabt26xaxZs+q8z1SR3gv/QaPREB4ebtQyjYGpvZmVtD+t5b0g54hxkEq3g2NqA0rS/nSl+QGmN0c6nU1XIpFIOjKdVuka02PAwcGB5ORk8vPzcXZ2xsnJyeAlQSUnTpxAo9EwceJErl27xldffYVGo+GNN94gLCys1rKzsrKUTf/S0tJq7JYqkbQW7TFHTp8+zYQJExRPieo0NIc6xRxp7yVxzUmAWLhwoXjw4IFISkoSUVFR4tChQ8LHx0f4+/sLIR5Guq/+uXnzZuHl5SWWLl0qGkvlvVFRUeLs2bNCp9OJwMDAOvPv2bNHnDhxQjn29vYW169frzXvihUrhKura426KsHEljjK1P7pP2Omw86RhIQEERsbW2959c0hU58jJvukO27cOE6cOMGRI0d4++23MTc3x8LCguTkZO7cqX1n8rt37xIfH4+1tTU6nc5g47tly5YpPoRLliyp9X6dTsfAgQNRq9XcuHGj1jyhoaHs2LGDZ555BtBHWbp161YN9xuAw4cP88Ybb2Bubt7U7kskDdJR50hDNGUOmSImq3QnTJjAF198we3bt7GxseHgwYOsW7eO5557TvE3BJQwcYWFhQghGDZsGFqtlpiYGCwsLJR8JSUlBn6EtaFWq9HpdFy/fr3OAeDl5cXmzZsVV7Hjx4/j4OBQa95z586xZ88eEhMT+fjjj5slB4mkLjrqHGmIpswhU8Rk/XR79uxJbm4uo0aNAsDW1pbg4GC+//57g3wWFhZs2rQJnU5H7969UavV+Pj4UFZWxtatW5V8W7ZsabDOqVOn4u3tjRCC1atXk5aWRn5+PmPGjAH0MRYSEhLIz89n9erVABw6dIiNGzcC1MgfFBQE6G1rzs7OLZSIRGJIR5wjP/74I5s3b6aoqAg7OzssLS2bNIc6Be1t32hOohVC19VHXTuhRkdHi6tXrza6nMbkN3V7lUztn9p6fgjR/DnS1DlUW12mNkdM1rzQlpiZmZGcnFzjvIuLC4MHD250OQ3lT0tLM/jZJ5GYCs2dI02dQ51hjpjk4ogePXrkFhcX27Z3O9oCS0vL60VFRer2bofEdOhK8wNMb46YpNJtCiqVqjtwCPgFcBFCtGmATpVK9TvgfSBPCOHalnVLJI1FpVItBGYDy4CT7TBPvIEFwB+FEHltWXdb02mVrkqlegbIAP4KWAHThBAP2rdVEomkLlQqVRDwJvDfwK+EEP9q5ya1Cp1S6apUKnMgF4gHhgBvCiGK2rdVEomkPlR637Uw4LfAUGCMEOJK+7bK+Jisy1gD/B4QwBjgAFC7U6ER6Mz2M1OzlXUk5LhoOkIIoVKpvgOe/8+pScA2Y9fT3nRWpbsEGIDejnu1NSsqLi627Yy/FgBUKlWnVBptgRwXzeYq8AD4FeBFJ1S6ndW88DqgAr4UrdzBzhxGz9RC5nUk5LhocR3DgJFCiE63VLNTKt22RE4uSW3IcSGpC7k4ohU5cuQIMTEx3LlzBzc3N7KysnBwcCAnJ4dz587h4uKCm5sb5eXlxMfHM2PGjDrLKi0tbbC+rKwsPDw8WLRoEffu3VPOx8fHM3/+fHx9fQH9Uk2NRkNERETLOylpNvWNjyNHjjBmzBhSUlIAOHnyJG+++WadZTVmfMTFxTF//nymTJnC/fv3DdpRtS5J69Ikm25nfjlQHWO8LJg8eTIajYakpCTee+89QB93dNCgQaxfv57IyEg+/fRTEhMTefvtt/n8889rlPF///d/xMTE0L9/fwICAuqtLzw8nJ49e2Jubk7Pnj2V88eOHSMiIoKQkBCysrLo0aMHpaWlDBo0qCXdk7SQ+sbHoEGDuHnzppJ3zJgxfPLJJzXKuHz5MtHR0RQXF/Phhx/WW9+0adOYNm0a69at45dffuHRRx9V2lG1LmMh9UXtNEnpduaXA9Ux1suCcePGERsbi52dHVlZWcr58vJyzM3NGTx4MFev1nzXd/v2bf70pz8xefJkVq9erWxFHRISgk6nU/KtXr2aPn36APolkjExMRw/fpyvvvqK119/vbIvAAwaNAidTsfevXsBmDVrFpMmTTJGNyXNpK7x0RgmTJiAvb093t7eSkSvnTt3Guyg6+XlxZNPPqkcBwQE8N1337F48WKjtL8+pL6onTY1L2i1WnJzc1u9nitXruDk5ISzszMFBQXK+eo/6Vub4uJijh07xujRozlz5ozBNTMzM8rLy8nJyWHgwIE17u3Tpw+enp5cvHiRbdu2ce3aNaD+rax/9atf0atXL6ytrQ1+PlbmqaxLpVKhUqno3r17a3Rb0kjqGx+NYenSpeTm5hIaGkp6ejoADx48MBgfFRWGC8vWrl3L3Llz+eabb4zSh9ak0+qLpkTHoYHoRenp6WLhwoVi7ty54uzZsyIwMFAEBgYKFxcXkZ2dLSZOnCiWLFkiMjMzxciRI8XatWvF+fPnhZOTk3BychK5ubkiMDBQBAQECA8PD3Hp0iXh7e0tioqKxNdffy327t1bb/2VrFq1Sty4cUOcPn1aREdHK+c9PDxEWVmZiI2NFQkJCfWWQSMjF9UnkzVr1ojU1FRRWloqZs2aJTIyMsT69euFEEKkpKSIuXPnCjc3N1FWViaEqDtSU0ZGhggLC2uw3+fOnRNz5swRc+bMEffv3xcrVqwQt27dEgcPHhQajUYsWbJECCGEq6urcHV1FZs2bTKKDGRq+lwRov7xcerUKTFu3DgxdepUkZGRIYSoe3zk5OQ0+L8UQoiIiAixcOFCMWvWLHHz5k0RHh4uUlNTa62rPowxN4TomvpCCGFcP92IiAj69u2LtbU13333HQCzZ88mJyeHU6dOMWLECDQaDWq1mqFDh7J8+XJ8fHzYvn072dnZ7N+/HwAnJyfUajUhISFMnz6duLg4UlJS2LBhg1JXXFwciYmJyrGTkxPDhw8HIC8vjwEDBlBcXExCQoKSp6Gf9MZm5cqVyt8xMTEGPx9feeUVXnnllUaV89RTT/HUU081mG/EiBFERUUpx++//z4AU6ZMYcqUKcr5nTt3NqpeSetS3/gYPXq0wditj4EDB+Lj49NgPjc3N4PjBQsWKH83ti5j0lX1hVGVbnl5OR4eHtja6s0bWq2Wnj178sgjj1BaWoqZ2UNrRq9evZS/q54HKCsr48EDfZgEe3t7IiMjsba2xtLS0iBP1ej1VR//+/fvT15eXo2f7g39pG9tLC0tuXz5Mjk5OTVeYsXHx3ea7UgkzaO+8XHy5MlOZw7qqvqiSX66DfkeZmRk8P7772NjY8Prr79OSkoKGo2GrKwsfvzxR2xsbDh8+DD+/v5s3LiR8PBw0tLS2LJlC+Xl5QQHB7Njxw7u3bvH9evXCQgI4JlnnmHFihVMmTKFESNGNKqdP/30E2vWrAH00e7379+Pvb09RUVFREREYG5uTlhYWL17kzXWF7Gp/pharVb59m5Nrly5wqpVqzAzM2Pr1q3Ky7a4uDj+8Y9/cOPGDfbu3UtsbCwpKSncu3ePjz/+2GB7FumP2XwaOy462ng4evQoCQkJpKWlsXLlSs6cOcOVK1cYMGCAsnuDseZGV9QXgHFtusYgMDBQ6HQ65fjLL78U7u7urV5vdWim3aqj26kqWbt2rcjKylKOt2zZIi5cuNAsGchU97gw1fHg4uIiSktLRVBQkJgzZ44IDg5WrjV3brQGpqYvhLFtusZAq9UaHI8fP57x48e3T2OaQUe3U0FNt6E7d+5w4cIFFi1a1Kqy6YqY4ni4desWvXv3plu3bixfvhyVSsWqVatIT09n6NChrSqvpmKK+qJDrkjTaDRGKWfnzp3MmzePd999FyFEnSu2jEmlnSooKIj58+cDGMVOlZCQgIWFRa12qsrUGDsVGLoNFRcX4+vry/r16xV/XonxMLXxAHDgwAGmT58OPPTx7t+/v4EbYkeitfQF6L/Ipk2bZpTyKzHqk25MTAxJSUnY2dmh0WgICQkhLy8PV1dXCgsLCQsLw8rKiqFDh3Lr1i369OlDQEAAI0eOxNHRkaKiIoM3uitXrqSgoIDBgwczadIkgoKCUKvVrFu3jh49ejTYHjc3N9zc3Fi0aBGlpaV1rtgyJu7u7vj5+Sl2quq89NJLrFixAn9/f+Wci4sLnp6eBnaq6OhoxU4F+q2tq3ogADg6OuLo6FhrO1xcXFi6dCmgt1Pt2LEDe3t7zpw5Q2pqKgUFBYSGhuLv78/Vq1dZsWIFPj4+/Pa3vzWWKCSY3ngASEpKUjwbNm7cSGZmJmVlZXh6erZcIFXo6PqitLSU9PR0BgwYYNR+G9WmGxwcLHbt2iVu3rwp7t27J/z9/cXKlSuFt7e3SEhIEKGhoaKkpERMnTpVCPHQ73DGjBlCCCF8fHxEYWGhWLBggbhw4YJ4++23RWBgoHBzcxNJSUli8eLFNfwIg4ODhZeXl5Ly8/MNrs+ZM0fMnDlTlJeXi7feekv88ssvIjY2VvzjH/8wio2mIZk0B1O0U8nUeuOio4yHSow1Nzq6vli7dq24fft2nf7RzZGJMLZN18/PjzNnzuDm5sbMmTMZO3Ysw4cPV76NHnvsMbp3717DNaqsrMzgE6CiooJRo0axbNky5ZytrS1arZagoCCeeOIJ4OEKrapfIlWJiopi48aN/Otf/6pzxVZHwxTtVJLWo7OOh46uL86fP4+/vz+JiYmcPn0ae3t7o/TbqEo3MjKSH374AVtbW1544QWCgoKUlwf1kZ2djb+/P7169VJ+Bjz//PNERkbi6+uLlZUVr732GocOHaK8vBwrKyvl3qo/L6oTHBzMtWvXKCoqYtGiRcyfP1+x//zlL39pYW/bFo1GQ3h4eIvLcXV15ZFHHuGJJ55QgqxITBdjjQvQ2y8/+eQT4uLijFJeQ3R0fXHgwAFAL2NjKVwwsp9uczHmwDEWxvJFNJbdqlJGLbVbeXh4UFpayquvvsrcuXONIgNJTUxtXNy9e5ft27eTlZXV4FxsLR/2xmLK+gI6iPdCRxOgMcnJyWHkyJG4urpibm5ORUUFarWaffv2AfrlnmFhYZw9e5YPPviAn3/+GYAhQ4bg6+tLfn4+RUX6PTUvXrxIWloa/fr1IzMzk4KCAtRqNR4eHgYTKyQkhMWLFyupahCPbdu2ERERQXJyMoWFhW0oCUlVOuK4MJYXQGtj6vqiVZSuMf95Dg4OJCcnk5+fj7OzM05OTmRnZ9fId+LECTQaDRMnTuTatWuUlpbi5+eHh4cHqampNfJXjyzUUJDo5uLn58ezzz6Lm5sbx44dY+zYsXh6eioKrzl2K61WS0REBKNGjcLDwwOtVmsgk/oikVW6APXp04eSkhKj91fSODrauKhuv2xL2kNf1BVZrJLaAr6npKTw6quvAg0Hla+PZtl0PTw82LJlC2fPnuXSpUv069dPWRu+fv16JV/lz4DKzw8//JDs7Gy6d+/OBx980Ki6hgwZwqhRo4iOjsbDw4Nf//rXhIeH13i5MHbsWMaOHcvevXvJzMwkKSmJu3fvYmZmpqztrsru3bvZvHkzmZmZHDx4EBcXl1qDRLeUjma38vf3Jz8/H2tra/r27dvyDkqaRUcbF61lv4SOqS9qm/9VqR7w3dLSks8//5wXX3wRqDuofGNoltIdN24cJ06c4KuvvsLX15fTp09jYWFBUlISd+7cqfWeu3fvEh8fz9ixY8nMzKSkpERZ579s2TLlqcvc3JxNmzbVuF+n0zF+/HjUajU3btyotY7Q0FDi4uL47LPPSEpK4o033uCVV15h27ZtrF271iBvfSt0jImrq6vB8ccf177PXuVPpsrPl19+meDg4BrXt2zZYnDf2LFjm9SeqmVK2o+ONi6ql2dMOqK+aMz8r7pSb+fOnTg7OxusAGwuzVK6EyZMIDAwkLt372JjY8PBgweJjIxk4cKFBnbCyp+yhYWFCCEYNmxYjW8cgJKSEsWNo66gEmq1Gp1OR7du3eqMxuXl5cXo0aM5cOAAarUaa2tr+vTpU6vtsr4VOh0BU7dbSVoHUxwXHVFfNGb+r127ls8++4xvvvmG1NRUzp8/T2JiIkePHm3RjivNUro9e/YkNzeXUaNGAXp/uODgYL7//nuDfBYWFmzatAmdTkfv3r1Rq9X4+PhQVlbG1q1blXzVv6VrY+rUqXh7eyOEYPXq1aSlpZGfn8+YMWMAvQ0mISGB/Px8Vq9ejY2NDd7e3nz88cf4+PjUyF99hU5rY8w3rg4ODqxZs4Zhw4bh5eVFRUWFgS9iJadPn+b9999n9uzZdW56mZWVxbhx48jKyiItLY358+fLDQrbkI44Lk6cOEFsbCw///wzERER3Lx5s0XjoiPqi+rz//jx4zz++OM8/fTTgH5JcNWVepVKVqPRtHyLq8auohCttPqqIepaDRIdHS2uXr3a6HIak79qXTRx1c3ChQvFgwcPRFJSkoiKihKHDh0SPj4+wt/f36Ds6p+bN28WXl5eYunSpY3uS+W9UVFR4uzZs0Kn04nAwMBa8yYkJIjY2Ng6y1qxYoVwdXVtkQxk6nzjQggh9uzZI06cOGFQthBNnxttSXP1xaZNm0RhYWGz62rKXOkQLmP1YWZmRnJyco3zLi4uDB48uNHlNJS/pUGiK+1WR44c4e2338bc3BwLCwuSk5MbtFtZW1uj0+kMvAmWLVumuPYsWbKk1vt1Oh0DBw6s185dH4cPH+aNN96oN06opGWY4rgA/fuRHTt28MwzzzTr/vaiufrCx8enUf7MlbREXzTJvGBpaXndWLvkNoWwsLA2q2vbtm2Avq9Nua8j2q0a4ty5c/zyyy8kJiby8ccf4+zs3OQyJPVjiuMCDN+PeHh4NKsMqS9qp0lKt7H7undFOqLd6scff2Tz5s0UFRVhZ2eHpaWlwfWgoCBAb6eSCrd1MMVxUf39SHOR+qIOGmuHkMn07VatZdeWSY4LmRqfOrxNV1KT5tqtGrqelpYmlwabMHJcmAZNCngjqUmPHj1yi4uL29xu1RZYWlpelz8Rm4ccF5K6kEq3DVGpVDlAHrAOOCyEKG7gFmPW/VvgfwEPIcTf2qpeSf2oVKrnga8AZyHEF+1Q/3DgGOAohGi9pZkSBal0uxAqlepl4AtghhDiG5VKZS6EKG/oPolxUalUlWa93wAnAV8hxL52bM844FPgLeCfgJkcF62HVLpdDJVKNRb9BFsHvCyEmN3OTepyqFSqRcBgYBqwSQjRdj5OdaBSqaYAYUAoMEAIUbsTsKTFmLTSlXazpqNSqSyApcAiwAKwEUKU1X+XxJioVKpTwCDgOLBCCJHXzk1CpVL1AgKA2UAZ8IQwZeXQgTHqdj1tTXFxsW1nHRet6FRuBbyOXuH2BiYDB1upLkk1VCrVAGA0cA/4A/Ar9Hb+9qYf8N/Ao+jHxe+BM+3aok6KdBnrYgghbgohxgDDgTj0TzWStsMSSAAcgOeEEA0H0W0DhBDZQojfA6OAzwG5TVMrYdLmhdbag6kjIPcnk0g6J53ySffIkSPExMRw584d3NzcyMrKwsHBgZycHI4cOcKYMWOUMHUNbbtRWlraYH27du3C1dWVGTNm1NgCp+oWH1qtFmdnZ3x9fVvQO4lEYsqYtE23LiZPnoxGoyEpKUnZZtzBwYFBgwYxaNAgbt68qeSta9uNy5cvEx0dTXFxMR9++GG99bm6uuLq6kpoaCgZGRk8++yzAJSXlxts8dGtWzfMzc0ZMGCAsbqqIF8q6unMcgApi6qY6iKNTql0QR9SLzY2Fjs7O7Kyspp074QJE7C3t8fb21uJ0rRz504uXryo5PHy8uLJJ59Uju/cucOFCxdYtGiRcq76Fh/Lly9HpVKxatUq0tPTGTp0aAt6aIh8qainM8sBpCyq0h4RzIxBpzQvFBcXc+zYMUaPHs2ZM01/Abt06VJyc3MJDQ0lPT0dgAcPHhjspFpRUWFQn6+vL+vXr1dC9AGkpqayYcMGZYuPymv9+/dXdhhtT7RaLbm5ua1eT107r547dw4XFxfc3NwoL29fX3wpi4dIWbQy7R1xpyWJOiI5rVmzRqSmporS0lIxa9YskZGRIdavXy+EEOLUqVNi3LhxYurUqSIjI0MIUXd0ppycHLFp06Zar1XFy8tLvPXWW2LBggXi0qVLIjw8XKSmpirXK8vfsGGD0Gg0wtXVVVRUVNRbJk2M5FSbLNLT08XChQvF3LlzxdmzZ0VgYKAIDAwULi4uIjs7W0ycOFEsWbJEZGZmipEjR4q1a9eK8+fPCycnJ+Hk5CRyc3NFYGCgCAgIEB4eHuLSpUvC29tbFBUVia+//lrs3bu3QdkIIcSqVavEjRs3xOnTp0V0dLRy3sPDQ5SVlYnY2FiRkJBgFFnUNSakLLq2LDpS6pTmhapbTcfExBiYF0aPHt3o3X8HDhyIj49Pg/mqxzj97W9/a3BcuQdWW79Ai4iIoG/fvlhbWyvbe8+ePZucnBxOnTrFiBEj0Gg0qNVqhg4dyvLly/Hx8WH79u1kZ2ezf/9+AJycnFCr1YSEhDB9+nTi4uJISUkx2Bk1Li6OxMRE5djJyYnhw4cDde+8Wl5ejrm5OYMHD+bq1atSFlIWbS6L9qBTKt3qWFpacvnyZXJychg0aJDBtZZu09ORKS8vx8PDA1tbvelLq9XSs2dPHnnkEUpLSzEze2hd6tWrl/J31fMAZWVlPHjwAAB7e3siIyOxtrbG0tLSIE/ljgaVdVdS186rZmZmlJeXt8mOzFIWD5GyaF+6nJ+uVqtVvsVbkytXrrBq1SrMzMzYunUrffr0Ueq/cuUKAwYMYOPGjXXe31Q/3dpkkZGRwfvvv4+NjQ2vv/46KSkpaDQasrKy+PHHH7GxseHw4cP4+/uzceNGwsPDSUtLY8uWLZSXlxMcHMyOHTu4d+8e169fJyAggGeeeYYVK1YwZcoURowY0ai2/fTTT6xZswbQ/yrYv38/9vb2FBUVERERgbm5OWFhYXVuP9MUWdQ1JqQsurYsOhTtbd9oScIE7ZhBQUFizpw5Ijg4uN77MYJN1xgEBgYKnU6nHH/55ZfC3d29Veqqi6bIorXkIISURVVMTRYdKXW6J10/Pz8sLCwQQvD444+Tk5Oj2Kv+/e9/c/nyZeVJ991332Xv3r34+PiwevVqsrOz+eabb7h16xYzZ85U7FWTJ0/mp59+UuxVlT+f6rNX/fnPf+ajjz7i559/Zs+ePYq/sBBCcRtzcnKq023MGE+6nQVjPN11FqQsHmKqT7qdzmWs0l4VFBTE/PnzAYxir0pISMDCwqJWe1Vlaoy9qqO5jTUFjUZjlHJ27tzJvHnzePfddzFVpSBl8RBjyWL37t28/PLLbeKu1p50uhdp7u7u+Pn5Kfaq6rz00kusWLECf39/5ZyLiwuenp4G9qro6GjFXgX6ba2nTJliUJajoyOOjo61tsPFxYWlS5cCenvVjh07sLe358svvyQzM5OysjI8PT2N1e16iYmJISkpCTs7OzQaDSEhIeTl5eHq6kphYSFhYWFYWVkxdOhQbt26RZ8+fQgICGDkyJE4OjpSVFRk4BGycuVKCgoKGDx4MJMmTSIoKAi1Ws26devo0aNHg+1xc3PDzc2NRYsWUVpaioWFRWt23wApi4d0NFm4uLg0eSGTKdLplO6QIUP4+OOPlePKuApqtVqJgTB58mTgoSvX7373O6KiogzK8fX1VV62HT9+nFu3bjX6BQGAnZ0du3fvVo4XLFig1NXW5OTkMHLkFpS/YwAADhRJREFUSKZOnYq5uTkVFRWo1Wr27dvH22+/zejRo9FoNMyYMYO//e1vypPLkCFD8PX1ZcmSJRQVFQFw8eJF0tLSeOmll8jMzKSgoAC1Wo2Hh4fBxAoJCUGn0ynHq1evVl4mAsydO5fi4mK6devWRlLQI2XxkI4oi65Ap1O6xkCr1Rocjx8/nvHjx7dPY4yAn58fZ86cwc3NjZkzZzJ27FiGDx+uPKU89thjdO/eXVnyXElZWZnBJ0BFRQWjRo1i2bJlyjlbW1u0Wi1BQUE88cQTgD5QUFVXoeo/naOioti4cSP/+te/eO6554zb4XqQsnhIR5RFV0Aq3QbQaDTKE3FLCAkJ4dKlS5ibm7Nz504jtKzxREZG8sMPP2Bra8sLL7xAUFCQ4hRfH9nZ2fj7+9OrVy/laeX5558nMjISX19frKyseO211zh06BDl5eVYWVkp91b92Vmd4OBgrl27RlFRkUGsirZAyuIhHU0W8fHxHD16lCtXrhAaGoq1tXXLO9kB6XTeC5UYy15VqXRbaq+qxNvbm6CgIB599NGG+tbu3gvG+sJpKR3hjb2UxUNMURYdiU7nvVBJpb3K1dW1hr0K9MuBw8LCOHv2LB988AE///wz8NBelZ+fX8Ne1a9fvwbtVYsXL1ZS1SAegLKksSGF21HoCBOroyBl8RApi5bRac0LHc1edfPmTdasWUNoaKjxOyuRSEyGTvukGxkZyaeffqrYq2JjY9m1a1eD99Vlr9LpdPj6+qLVajlx4gR/+ctfarVXhYeHK6mqTWrevHkUFBTg7e1tEES9LTGWPyXog8InJyeTn5+Ps7MzTk5OZGdn18hXV/i+SuLi4pg/fz5Tpkzh/v37De7kYSykLB7SHrI4ffo0EyZMUH55Vqe6rNpKFm1Cey+Ja0miFZY51hXmsa2hGcuAFy5cKB48eCCSkpJEVFSUOHTokPDx8RH+/v5CiId9q/65efNm4eXlJZYuXdro9lXeGxUVJc6ePSt0Op0IDAyska+u5dDVWbt2rcjKyjIouzmyqBwTUhYdWxZCCJGQkCBiY2NrvVabrFoii46UOu2TbnMxZXvVuHHjOHHiBEeOHOHtt9/G3NwcCwsLkpOTuXPnTq333L17l/j4eKytrdHpdAZ7vC1btkyxTy9ZsqTW+3U6HQMHDkStVnPjxo0a1yvD9w0ePNjAP7MqAQEBnDp1yqjbGElZPKQjyqIhGiMrU0Uq3U7EhAkT+OKLL7h9+zY2NjYcPHiQdevW8dxzz1FYWKjkq1yKXFhYiBCCYcOGodVqiYmJMVgRVVJSYrDMuTbUajU6nY7r16/XsI9D3cuhq7J27Vrmzp3LN99805LuGyBl8ZCOKIuGaIysTJVO+yKtEmO6tzg4OLBmzRqGDRuGl5cXFRUVBi/SKjl9+jTvv/8+s2fPZsaMGTXKqX49LS2N+fPnKzsUN5eePXuSm5vLqFGjAP3LvuDgYL7//nuDfBYWFmzatAmdTkfv3r1Rq9X4+PhQVlbG1q1blXzVg7PXxtSpU/H29kYIwerVq0lLSyM/P58xY8YANZdDHz9+nMcff5ynn34a0MceSE1NpaCgwKgvGaUsHtIRZfHjjz+yefNmioqKsLOzw9LSsl5ZdSra277RkkQHtWMKUb+9qrbrLbVX0Yph/GqjLtt3dHS0uHr1ap33bdq0SRQWFjap7KbIoq3lIISURVWaK4uGrtdWdlPnSEdJJm9eMEV7VWfAzMyM5OTkGuddXFwYPHhwnff5+PjUu5jEFHfykLJ4SHNl0dB1U5RFXZi8eWHChAkEBgZy9+5dxV4VGRnJwoULG2Wvqk6lvQqoM2J9pb2qW7duzbJXtQaWlpbX23pL6rCwsFYre9u2bcrflpaW1xt7X3vIAaQs/v/27jUkqj8N4Ph3nPLSammBHhAqTCliI8oMNNCCjNYK6sUfK5gaWq2xRjATtRsqK6GJZkGogYzRi/1HiF3fVEt00W4vAiOKJSkpHCs2NNkZrZnOvnCbvI1jOjPOTM8HDjrn+pvnxcOZ33nO7zeUL8bCl/h90vXH/qqR21evXj3lOFitVs/OP+QnJA4/SSx8U8COveAJzh7KNTU1kZ6e7vTnkavtY53bX98rF0KMz+/7dL3JU/1V7e3tw7pChBCBy6/vdMPCwrr7+/u93mflDaGhoR/k56EQgcevk+5EaDSaNOASsElV1afT1IYI4F/AbVVVj0xHG4QQvsHvH6Q5o9FozgH/BC4CO6Yr4QKoqtqn0WgygPsajeY/wCLgH6qqBtb7jUIIlwLyTvf/ZTL/Bv4L5KuqOvZQRl6m0WjigduAGTCpqnpumpskhPCyQH2QtgP4C2AByjQ/inSnXxYQDCQCf5/mtgghpkGgJt1VwAvgIPBXr9aVjUNV1WLgbwz2MQfkA0AhxPgCsntBCCF8VaDe6QohhE+aVPWC1Mf+JLEQQvyKSXUvePv1W2/yhanPfYW8iiyE+3m1e6G0tJTu7m6PX8fZBIBPnz5Fr9eTnZ2N3W73eDvGI7EQ4vfk1jvd169fU1tbi9VqxWAwcOPGDWBwht2ysjL279/PkiVL2L9/P9u3b2fr1q1s2rSJ6upqAKqqqqirq8Nms9Hb20tubi719fWcOHGCtrY2uru72blzp8v2lZSUYDQa6ejo4NWrV+j1egByc3Opra3l0qVLKIrC2rVrx/pubrnT/R1jIYRwza1vpJ07d46oqCgiIyN59uwZALt27aKrq4sHDx6watUqDAYDiqKQkJDAkSNHyM/P5+zZs3R2dnLx4kUAdDodiqJQWVnJH3/8QXNzM48ePaKqqspxrebmZu7fv+/4rNPpSExMBH5Oatff38+dO3cc+9jtdrRaLbGxsbx//96dX11iIYSYELcmXbvdjtFoJCZm8LlSaWkps2bNYsaMGXz9+pWgoJ+9GREREY7/h64HsNlsfPv2DYDk5GQaGxuJjIwkNDR02D5DJ8Ub+hPZ2aR2QUFB2O12r0x2J7EQQozFrUk3JyeHwsJC5s2bx4YNG0ZtX7FiBceOHaO4uNixTq/Xk5ubi91up6KigoaGBkwmEx8+fODo0aPA4EwNW7duHXauzMxMMjMzx2zHyEntGhoaSE5ORqfTsXfvXrRarUdHtweJhRDCiclMrIYHJ7wrKSlRzWaz4/PNmzfVnJwcj11vJHxoQkh/i4UsssjiepGSsRGkZOwneZAmhPv59BtpBoPBLedpampi5cqVXinR8hSJhRCBwSPj6V64cIHW1lbi4uIwGAxUVlby6dMnsrKysFgs1NXVER4eTkJCAp8/f2bOnDkcPXqUpKQkMjMzsVqtHD9+3HG+48eP09vbS2xsLJs3b6a8vBxFUThx4sS4U1j/oNfrefv2rSe+qksSCyHEUB650+3q6iIpKYmsrCy0Wi3fv39HURT+/HNwWNs1a9ZQV1fHkydPOHnyJO/evQMgPj6egoICenp6sFqtALx48YL29nbmzp1LR0cHvb29KIqC0WgclmQqKyvJy8tzLENfBJhOEgshxFAeudMtLCzk8ePHZGdns2PHDtLS0khMTHTcsUVHRxMcHEx0dPSw42w227C/AN+/fyclJYWioiLHupiYGEpLSykvL2fBggUAfP36dVjZlK/0s0oshBBDeSTpNjY28vLlS2JiYli+fDnl5eWOFwTG09nZSXFxMREREY47t2XLltHY2EhBQQHh4eGsW7eOy5cvY7fbCQ8Pdxw79Cf4SFevXuX69eu8efOG06dPExkZOfUvOUESCyHEUD5VvWAwGKivr3f7eX+Fr1Qv+GMshBCu+VTS9QW+knR9gSRdIdzPp0vGhBAi0Hg06bqrthRg/fr1tLW10dPTw+7du9HpdHR2do7az9lQhj9cu3aN1NRUHj16BMC9e/fYuHGj29rpzHTE4uHDh2RkZDgqJUYaGStvxUKI39mUkq7RaMRms9HW1obJZOLKlSscOnSIw4cPD9vvR8L58ffUqVPk5eVRWFg44WvFx8eTkpJCS0sLRqORqqoqTCbTqP2ampqoqakhJyeHlpaWUdu3bNnCnj17HJ9TU1NZuHDhhNvhjC/GIjk5edzzjoyVu2IhhHBuStULa9eu5e7du9y+fZuCggIePnxISEgIra2tfPnyZcxj+vr6uHr1KmlpaXR0dDAwMEBISAgARUVFDAwMAKDVah1jyw5lNptJT09HURQ+fvw4aruzoQw9zRdj4cp0xUqI39mUkm5GRgYlJSX09fUxb948WlpaaGxs5MCBA1gsFsd+Gs3gsxiLxYKqqixdupTS0tJR5xsYGHDUl2q12jGvqSgKZrOZmTNnjqptBedDGXqaL8bClemKlRC/sykl3VmzZtHd3U1KSgowWKhfUVHB8+fPh+0XEhJCdXU1ZrOZ2bNnoygK+fn52Gw2zpw549ivtrbW5TW3bdvGwYMHUVWVsrIy2tvb6enpITU1FRg9lOGtW7eYP38+ixcvBqC1tZXz588TFRVFVVUVixYtmkoIfDoWr169oqamBqvVSlxcHKGhoePGSgjhBZMZmgwPDmfozL59+8ZcbzKZ1Pfv3zs9rrq6WrVYLBM+Nz40tKMzk42Fq+0jz/2rsZBFFllcL35TMhYUFERbW9uo9Xq9ntjYWKfH5efnjzsQzL179wgODnZLG71lsrFwtd0fYyGEv5nUyxFhYWHd/f39MR5oz7QLDQ39YLValYnuL7EQQvyKSSVdIYQQk+M33QtCCBEIJOkKIYQXSdIVQggvkqQrhBBeJElXCCG8SJKuEEJ4kSRdIYTwIkm6QgjhRZJ0hRDCiyTpCiGEF0nSFUIIL/of6gWVTJbGQPMAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "tree.plot_tree(Tree1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[0 = correct, nonzero = incorrect], \"Number of questions wrong\"\n"
     ]
    },
    {
     "ename": "NameError",
     "evalue": "name 'predict1' is not defined",
     "output_type": "error",
     "traceback": [
      "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[1;31mNameError\u001b[0m                                 Traceback (most recent call last)",
      "\u001b[1;32m<ipython-input-5-88dbd8dca7cb>\u001b[0m in \u001b[0;36m<module>\u001b[1;34m\u001b[0m\n\u001b[0;32m     10\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m     11\u001b[0m \u001b[0mprint\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;34m'[0 = correct, nonzero = incorrect], \"Number of questions wrong\"'\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m---> 12\u001b[1;33m \u001b[0mprint\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;34m'Tree_1'\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mpredict1\u001b[0m \u001b[1;33m-\u001b[0m \u001b[0mclasses_test\u001b[0m\u001b[1;33m,\u001b[0m \u001b[1;34m','\u001b[0m\u001b[1;33m,\u001b[0m  \u001b[0mnp\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0msum\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mpredict1\u001b[0m \u001b[1;33m-\u001b[0m \u001b[0mclasses_test\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m     13\u001b[0m \u001b[0mprint\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;34m'Tree_2'\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mpredict2\u001b[0m \u001b[1;33m-\u001b[0m \u001b[0mclasses_test\u001b[0m\u001b[1;33m,\u001b[0m \u001b[1;34m','\u001b[0m\u001b[1;33m,\u001b[0m  \u001b[0mnp\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0msum\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mpredict2\u001b[0m \u001b[1;33m-\u001b[0m \u001b[0mclasses_test\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m     14\u001b[0m \u001b[0mprint\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;34m'Tree_3'\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mpredict3\u001b[0m \u001b[1;33m-\u001b[0m \u001b[0mclasses_test\u001b[0m\u001b[1;33m,\u001b[0m \u001b[1;34m','\u001b[0m\u001b[1;33m,\u001b[0m  \u001b[0mnp\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0msum\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mpredict3\u001b[0m \u001b[1;33m-\u001b[0m \u001b[0mclasses_test\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;31mNameError\u001b[0m: name 'predict1' is not defined"
     ]
    }
   ],
   "source": [
    "Tree2 = tree.DecisionTreeClassifier()\n",
    "Tree2 = Tree2.fit(attributes_train, classes_train)\n",
    "Tree3 = tree.DecisionTreeClassifier()\n",
    "Tree3 = Tree3.fit(attributes_train, classes_train)\n",
    "Tree4 = tree.DecisionTreeClassifier()\n",
    "Tree4 = Tree4.fit(attributes_train, classes_train)\n",
    "Tree5 = tree.DecisionTreeClassifier()\n",
    "Tree5 = Tree5.fit(attributes_train, classes_train)\n",
    "\n",
    "print('[0 = correct, nonzero = incorrect], \"Number of questions wrong\"')\n",
    "print('Tree_1', predict1 - classes_test, ',',  np.sum(predict1 - classes_test))\n",
    "print('Tree_2', predict2 - classes_test, ',',  np.sum(predict2 - classes_test))\n",
    "print('Tree_3', predict3 - classes_test, ',',  np.sum(predict3 - classes_test))\n",
    "print('Tree_4', predict4 - classes_test, ',',  np.sum(predict4 - classes_test))\n",
    "print('Tree_5', predict5 - classes_test, ',',  np.sum(predict5 - classes_test))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "predict1 = Tree1.predict(attributes_test)\n",
    "predict2 = Tree2.predict(attributes_test)\n",
    "predict3 = Tree3.predict(attributes_test)\n",
    "predict4 = Tree4.predict(attributes_test)\n",
    "predict5 = Tree5.predict(attributes_test)\n",
    "print('Tree_1     ', predict1)\n",
    "print('Tree_2     ', predict2)\n",
    "print('Tree_3     ', predict3)\n",
    "print('Tree_4     ', predict4)\n",
    "print('Tree_5     ', predict5)\n",
    "print('True_Values', classes_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Plot out the trees to see how each fits the data without being identical.\n",
    "Plotting shows the entire structure of the tree, with labels at each splitting point"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "tree.plot_tree(Tree1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[Text(198.4, 338.79999999999995, 'X[2] <= 2.45\\nentropy = 0.666\\nsamples = 120\\nvalue = [41, 38, 41]'),\n",
       " Text(148.8, 277.2, 'entropy = 0.0\\nsamples = 41\\nvalue = [41, 0, 0]'),\n",
       " Text(248.0, 277.2, 'X[3] <= 1.65\\nentropy = 0.499\\nsamples = 79\\nvalue = [0, 38, 41]'),\n",
       " Text(99.2, 215.59999999999997, 'X[2] <= 4.95\\nentropy = 0.176\\nsamples = 41\\nvalue = [0, 37, 4]'),\n",
       " Text(49.6, 154.0, 'entropy = 0.0\\nsamples = 36\\nvalue = [0, 36, 0]'),\n",
       " Text(148.8, 154.0, 'X[3] <= 1.55\\nentropy = 0.32\\nsamples = 5\\nvalue = [0, 1, 4]'),\n",
       " Text(99.2, 92.39999999999998, 'entropy = 0.0\\nsamples = 3\\nvalue = [0, 0, 3]'),\n",
       " Text(198.4, 92.39999999999998, 'X[2] <= 5.45\\nentropy = 0.5\\nsamples = 2\\nvalue = [0, 1, 1]'),\n",
       " Text(148.8, 30.80000000000001, 'entropy = 0.0\\nsamples = 1\\nvalue = [0, 1, 0]'),\n",
       " Text(248.0, 30.80000000000001, 'entropy = 0.0\\nsamples = 1\\nvalue = [0, 0, 1]'),\n",
       " Text(396.8, 215.59999999999997, 'X[2] <= 4.85\\nentropy = 0.051\\nsamples = 38\\nvalue = [0, 1, 37]'),\n",
       " Text(347.2, 154.0, 'X[1] <= 3.1\\nentropy = 0.375\\nsamples = 4\\nvalue = [0, 1, 3]'),\n",
       " Text(297.6, 92.39999999999998, 'entropy = 0.0\\nsamples = 3\\nvalue = [0, 0, 3]'),\n",
       " Text(396.8, 92.39999999999998, 'entropy = 0.0\\nsamples = 1\\nvalue = [0, 1, 0]'),\n",
       " Text(446.40000000000003, 154.0, 'entropy = 0.0\\nsamples = 34\\nvalue = [0, 0, 34]')]"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tree.plot_tree(Tree2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[Text(133.92000000000002, 199.32, 'X[3] <= 0.8\\nentropy = 0.666\\nsamples = 120\\nvalue = [41, 38, 41]'),\n",
       " Text(100.44000000000001, 163.07999999999998, 'entropy = 0.0\\nsamples = 41\\nvalue = [41, 0, 0]'),\n",
       " Text(167.40000000000003, 163.07999999999998, 'X[3] <= 1.65\\nentropy = 0.499\\nsamples = 79\\nvalue = [0, 38, 41]'),\n",
       " Text(66.96000000000001, 126.83999999999999, 'X[2] <= 4.95\\nentropy = 0.176\\nsamples = 41\\nvalue = [0, 37, 4]'),\n",
       " Text(33.480000000000004, 90.6, 'entropy = 0.0\\nsamples = 36\\nvalue = [0, 36, 0]'),\n",
       " Text(100.44000000000001, 90.6, 'X[3] <= 1.55\\nentropy = 0.32\\nsamples = 5\\nvalue = [0, 1, 4]'),\n",
       " Text(66.96000000000001, 54.359999999999985, 'entropy = 0.0\\nsamples = 3\\nvalue = [0, 0, 3]'),\n",
       " Text(133.92000000000002, 54.359999999999985, 'X[0] <= 6.6\\nentropy = 0.5\\nsamples = 2\\nvalue = [0, 1, 1]'),\n",
       " Text(100.44000000000001, 18.119999999999976, 'entropy = 0.0\\nsamples = 1\\nvalue = [0, 1, 0]'),\n",
       " Text(167.40000000000003, 18.119999999999976, 'entropy = 0.0\\nsamples = 1\\nvalue = [0, 0, 1]'),\n",
       " Text(267.84000000000003, 126.83999999999999, 'X[2] <= 4.85\\nentropy = 0.051\\nsamples = 38\\nvalue = [0, 1, 37]'),\n",
       " Text(234.36, 90.6, 'X[1] <= 3.1\\nentropy = 0.375\\nsamples = 4\\nvalue = [0, 1, 3]'),\n",
       " Text(200.88000000000002, 54.359999999999985, 'entropy = 0.0\\nsamples = 3\\nvalue = [0, 0, 3]'),\n",
       " Text(267.84000000000003, 54.359999999999985, 'entropy = 0.0\\nsamples = 1\\nvalue = [0, 1, 0]'),\n",
       " Text(301.32000000000005, 90.6, 'entropy = 0.0\\nsamples = 34\\nvalue = [0, 0, 34]')]"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "tree.plot_tree(Tree3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[Text(133.92000000000002, 199.32, 'X[3] <= 0.8\\nentropy = 0.666\\nsamples = 120\\nvalue = [41, 38, 41]'),\n",
       " Text(100.44000000000001, 163.07999999999998, 'entropy = 0.0\\nsamples = 41\\nvalue = [41, 0, 0]'),\n",
       " Text(167.40000000000003, 163.07999999999998, 'X[3] <= 1.65\\nentropy = 0.499\\nsamples = 79\\nvalue = [0, 38, 41]'),\n",
       " Text(66.96000000000001, 126.83999999999999, 'X[2] <= 4.95\\nentropy = 0.176\\nsamples = 41\\nvalue = [0, 37, 4]'),\n",
       " Text(33.480000000000004, 90.6, 'entropy = 0.0\\nsamples = 36\\nvalue = [0, 36, 0]'),\n",
       " Text(100.44000000000001, 90.6, 'X[3] <= 1.55\\nentropy = 0.32\\nsamples = 5\\nvalue = [0, 1, 4]'),\n",
       " Text(66.96000000000001, 54.359999999999985, 'entropy = 0.0\\nsamples = 3\\nvalue = [0, 0, 3]'),\n",
       " Text(133.92000000000002, 54.359999999999985, 'X[1] <= 2.85\\nentropy = 0.5\\nsamples = 2\\nvalue = [0, 1, 1]'),\n",
       " Text(100.44000000000001, 18.119999999999976, 'entropy = 0.0\\nsamples = 1\\nvalue = [0, 1, 0]'),\n",
       " Text(167.40000000000003, 18.119999999999976, 'entropy = 0.0\\nsamples = 1\\nvalue = [0, 0, 1]'),\n",
       " Text(267.84000000000003, 126.83999999999999, 'X[2] <= 4.85\\nentropy = 0.051\\nsamples = 38\\nvalue = [0, 1, 37]'),\n",
       " Text(234.36, 90.6, 'X[1] <= 3.1\\nentropy = 0.375\\nsamples = 4\\nvalue = [0, 1, 3]'),\n",
       " Text(200.88000000000002, 54.359999999999985, 'entropy = 0.0\\nsamples = 3\\nvalue = [0, 0, 3]'),\n",
       " Text(267.84000000000003, 54.359999999999985, 'entropy = 0.0\\nsamples = 1\\nvalue = [0, 1, 0]'),\n",
       " Text(301.32000000000005, 90.6, 'entropy = 0.0\\nsamples = 34\\nvalue = [0, 0, 34]')]"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "tree.plot_tree(Tree4)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[Text(133.92000000000002, 199.32, 'X[2] <= 2.45\\nentropy = 0.666\\nsamples = 120\\nvalue = [41, 38, 41]'),\n",
       " Text(100.44000000000001, 163.07999999999998, 'entropy = 0.0\\nsamples = 41\\nvalue = [41, 0, 0]'),\n",
       " Text(167.40000000000003, 163.07999999999998, 'X[3] <= 1.65\\nentropy = 0.499\\nsamples = 79\\nvalue = [0, 38, 41]'),\n",
       " Text(66.96000000000001, 126.83999999999999, 'X[2] <= 4.95\\nentropy = 0.176\\nsamples = 41\\nvalue = [0, 37, 4]'),\n",
       " Text(33.480000000000004, 90.6, 'entropy = 0.0\\nsamples = 36\\nvalue = [0, 36, 0]'),\n",
       " Text(100.44000000000001, 90.6, 'X[3] <= 1.55\\nentropy = 0.32\\nsamples = 5\\nvalue = [0, 1, 4]'),\n",
       " Text(66.96000000000001, 54.359999999999985, 'entropy = 0.0\\nsamples = 3\\nvalue = [0, 0, 3]'),\n",
       " Text(133.92000000000002, 54.359999999999985, 'X[1] <= 2.85\\nentropy = 0.5\\nsamples = 2\\nvalue = [0, 1, 1]'),\n",
       " Text(100.44000000000001, 18.119999999999976, 'entropy = 0.0\\nsamples = 1\\nvalue = [0, 1, 0]'),\n",
       " Text(167.40000000000003, 18.119999999999976, 'entropy = 0.0\\nsamples = 1\\nvalue = [0, 0, 1]'),\n",
       " Text(267.84000000000003, 126.83999999999999, 'X[2] <= 4.85\\nentropy = 0.051\\nsamples = 38\\nvalue = [0, 1, 37]'),\n",
       " Text(234.36, 90.6, 'X[1] <= 3.1\\nentropy = 0.375\\nsamples = 4\\nvalue = [0, 1, 3]'),\n",
       " Text(200.88000000000002, 54.359999999999985, 'entropy = 0.0\\nsamples = 3\\nvalue = [0, 0, 3]'),\n",
       " Text(267.84000000000003, 54.359999999999985, 'entropy = 0.0\\nsamples = 1\\nvalue = [0, 1, 0]'),\n",
       " Text(301.32000000000005, 90.6, 'entropy = 0.0\\nsamples = 34\\nvalue = [0, 0, 34]')]"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "tree.plot_tree(Tree5)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}