09_model_performance.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "2f0743a6-722b-4edf-a87a-4e3c03ffc320",
   "metadata": {},
   "source": [
    "### Introduction\n",
    "\n",
    "It is important to be able to quantify the performance of the model.\n",
    "\n",
    "Each algorithm (estimator) has it's own evaluation criteria based on the problem it is trying to solve.  The approach is described in each estimator’s documentation.\n",
    "\n",
    "The technical description for LinearSVC score is here:\n",
    "\n",
    "- https://scikit-learn.org/stable/modules/generated/sklearn.svm.LinearSVC.html#sklearn.svm.LinearSVC.score"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a63d5b52-23fc-4579-840a-f09b134d2051",
   "metadata": {},
   "source": [
    "### Linear SVC accuracy\n",
    "\n",
    "In the notebook [08 test and train split](./08_test_and_train_split.ipynb) we used the `score()` method:\n",
    "\n",
    "```python\n",
    "model.score(X_test, y_test) # -> 0.92\n",
    "```\n",
    "\n",
    "What is score actually doing in the case of Linear SVC?  A simplied answer is:\n",
    "\n",
    "$$\n",
    "score = \\frac{the\\:number\\:of\\:correct\\:predictions\\:made}{the\\:total\\:number\\:of\\:prediction\\:made} * 100\n",
    "$$\n",
    "\n",
    "We can also visually display the accuracy for classifications using a [confusion matrix](https://en.wikipedia.org/wiki/Confusion_matrixhttps://en.wikipedia.org/wiki/Confusion_matrix).\n",
    "\n",
    "The following code creates a confusion matrix:\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "8be910a0-ee52-432b-bf23-d2f710fb9dca",
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "np.random.seed(1)\n",
    "\n",
    "from sklearn.datasets import load_iris\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.svm import LinearSVC\n",
    "\n",
    "iris = load_iris()\n",
    "X, y = iris.data, iris.target\n",
    "X_train, X_test, y_train, y_test = train_test_split( X, y, test_size=0.33)\n",
    "classifier = LinearSVC(max_iter=3500)\n",
    "model = classifier.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "f3dbde4b-a3a8-4189-aa5d-37d240a7592c",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<sklearn.metrics._plot.confusion_matrix.ConfusionMatrixDisplay at 0x7f54d03eba60>"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAATIAAAEGCAYAAADmLRl+AAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/YYfK9AAAACXBIWXMAAAsTAAALEwEAmpwYAAAaaklEQVR4nO3de7gV1Znn8e/vHA4Q7iCogESwNTomUXRIvKTboLaRmLTaM+luE804TtLGW24ddTSmYxKnnUxn7E5ajQltjG28tSYmmnjDiIgmiggiXohoVBCBBgS8IXIub/+x6+ARDmdX7bP32VX7/D7PUw+7au9a9VIhr2utqrWWIgIzsyJrqncAZma95URmZoXnRGZmhedEZmaF50RmZoU3oN4BdDV2THNMntRS7zBya+niIfUOwQpuM2+yJd5Wb8o4+vCh8cr69lS/XbD47bsjYkZvrpdGrhLZ5EktPHL3pHqHkVtHT5ha7xCs4ObFvb0uY936dubdvVuq37aM/+PYXl8whVwlMjMrgqA9OuodxLs4kZlZJgF0kK8X6Z3IzCyzDlwjM7MCC4JWNy3NrMgCaHfT0syKzn1kZlZoAbTnbNYcJzIzyyxfPWQeomRmGQVBe8qtHElXSVoj6cltjn9R0jOSnpL0j+XKcY3MzDKJgNbqtSyvBi4Druk8IOlw4Dhgv4h4W9LO5QpxIjOzjEQ7vRquuVVEzJU0eZvDpwPfjYi3k9+sKVeOm5ZmlkkAHZFuA8ZKerTLdmqKS7wP+DNJ8yTdL+lD5U5wjczMMstQI1sXEdMyFj8AGA0cDHwIuEnSHtHDAiNOZGaWSemF2Oo0LXdgBXBLkrgekdQBjAXW7ugEJzIzyySA1qhpr9SvgCOAOZLeBwwE1vV0ghOZmWUSiPYqda9LugGYTqkvbQVwIXAVcFXySsYW4OSempXgRGZmFeiIqj21/PQOvjopSzlOZGaWSR/0kWXmRGZmGYn22vaRZeZEZmaZlGaIdSIzswKLEFuiud5hvIsTmZll1uE+MjMrslJnv5uWZlZo7uw3s4JzZ7+ZNYT2Kr0QWy1OZGaWSSBaI1+pI1/RmFnuubPfzAovkJuWZlZ87uzPqUu+Ool5vx3BqLFtzLzvGQD+4Qu7s+KPgwF487Vmho5o54rfPlPPMHNj2vTXOO2ilTQ3BXfeMIabLtul3iHlSiPfnwj61+sXkmYAPwCagSsj4ru1vF5vfOxv1nPsKev43pffu/XYBT9etvXzj789gaHD2+sRWu40NQVnXvwy55+wB+tWtXDpHc/y8N0jWf7s4HqHlguNfn9Knf35GqJUs7QqqRm4HPg4sC/waUn71up6vfXBg99k+OjuE1UEzL1tFIcfv6GPo8qnvQ/YxMoXB7J6+SDaWpuYc+soDjn61XqHlRv94f6005Rq6yu1vNKHgeci4vmI2ALcSGmtusJ5ct5QRo9rY+IeW+odSi7stGsra1cO3Lq/blULY8e31jGifGn0+xOIjki39ZVaJrKJwEtd9lckxwrnvl+NZrprY1upm3+fPU9E3L/0h/tTrRrZjlYaT747W1JIGluunFomsu7S8Xb/c0o6tXPNu7Wv5K8Pqr0NfnfHSD567MZ6h5Ib61a1MG7CO7XTseNbeWV1Sx0jypdGvz+ldS2bUm0pXA3M2PagpEnAUcDyNIXUMpGtACZ12d8NWLntjyJiZkRMi4hp43bKVwciwMIHhjNpz7cZN6Fxmga99cyiIUycsoVdJr3NgJYOph+3kYdnjax3WLnR+PentNJ4mq2ciJgLrO/mq38GzqWbyk93avnUcj6wl6QpwMvACcBnani9Xvm/p+/O4oeG8er6AZz4X/fls19bzYzPrOf+W92s3FZHu7j8golcfP3zNDXDrBvHsGxpYzyRq4ZGvz+l5eBSVzrGSnq0y/7MiJjZ0wmSjgVejojH1V07vRs1S2QR0SbpLOBuSq9fXBURT9Xqer11/hXLuj1+9vdT1Wz7nfmzRzB/9oh6h5FbjXx/IpS22QgZVxqXNAS4APhYlphq+h5ZRNwB3FHLa5hZ36vhC7F/AkwBOmtjuwELJX04Ilbv6CS/2W9mmZTmI6vNqxUR8QSwc+e+pBeBaRHR40rj+RpnYGYFUJohNs1WtqTSSuMPAXtLWiHpc5VE5BqZmWVSev2i5iuNd34/OU05TmRmlkkex1o6kZlZZp7Gx8wKrTSNjydWNLOC68sB4Wk4kZlZJqXZL9y0NLMCKw1RciIzs0JzjczMGkCt3uyvlBOZmWXip5Zm1hDctDSzQuucsz9PnMjMLJMA2lwjM7Oic9PSzIqtj5d6S8OJzMwyqeXEipVyIjOzzPJWI8tXQ9fMcq9zYsVqrDTe3QK9kr4n6Q+SFkv6paRR5cpxIjOzTALR1tGUakvharZfoPce4AMRsR+wFDi/XCFOZGaWWQdKtZXT3QK9ETErItqS3YcpraTUI/eRmVk2kamPLPMCvdv4X8C/l/uRE5mZZZJx8ZFMC/R2JekCoA24rtxvncjMLLNaP7WUdDLwSeDIiIhyv3ciM7NMAtGeriO/IpJmAP8b+GhEbEpzjjv7zSyzanX272CB3suA4cA9khZJ+lG5clwjM7NMIltnf5myul2g9ydZy3EiM7PMImdv9juRmVlGHjRuZg3ANbIeLF08hKMnTK13GLl16bLf1TuE3DvrM2fWO4R8W/T7XhcRAe0dTmRmVnCexsfMCi1w09LMCs+d/WbWAMoPGupbTmRmlpmblmZWaKWnlvka3ehEZmaZuWlpZoXnpqWZFVogJzIzK76ctSydyMwso4DwECUzKzo3Lc2s8Arz1FLSpfTQFI6IL9UkIjPLtWqOtZR0FaVFRtZExAeSY2MoLQE3GXgR+OuI2NBTOT291fYosKCHzcz6owBC6bbyrmb7lcbPA+6NiL2Ae5P9Hu2wRhYR/9Z1X9LQiHgzTWRm1tiq1bSMiLmSJm9z+DhgevL534A5lFZV2qGy4wwkHSLpaWBJsr+/pB9mjNfMGoaIjnQbyUrjXbZTU1xgl4hYBZD8uXO5E9J09n8fOBq4LSn4cUmHpTjPzBpV+hpZxSuNZ5Fq5GdEvLTNofYaxGJmRRClzv40W4X+Q9J4gOTPNeVOSJPIXpJ0KBCSBko6m6SZaWb9VKTcKnMbcHLy+WTg1nInpElkpwFnAhOBl4Gpyb6Z9VtKuZUppfuVxr8LHCXpWeCoZL9HZfvIImIdcGLZiMys/+ioTjE7WGkc4Mgs5aR5armHpF9LWitpjaRbJe2R5SJm1kCq+x5ZVaRpWl4P3ASMByYANwM31DIoM8u3iHRbX0mTyBQRP4uItmS7lvzN4mFmfam2nf2Z9TTWckzy8T5J5wE3Ugrtb4Db+yA2M8urAs1+sYBS4uqM+AtdvgvgoloFZWb5ppy1yXoaazmlLwMxs4IIQREnVpT0AWBfYHDnsYi4plZBmVnOFaVG1knShZRGou8L3AF8HHgQcCIz669ylsjSPLX8FKWX01ZHxCnA/sCgmkZlZvlWlKeWXbwVER2S2iSNoDSAs6FfiJ02/TVOu2glzU3BnTeM4abLdql3SHV33dl78uTs0QzfqZWv37No6/H7fzqeudeMp6k5eP8R6zn+68vqF2ROtLS0c8l37qRlQAfNzR088PBkfnbT1HqHVT2dL8TmSJpE9qikUcC/UnqS+QbwSLmTupvCtgiamoIzL36Z80/Yg3WrWrj0jmd5+O6RLH92cPmTG9hBf7WGw05exc/+bq+tx5b+fiSL7xnDeXc9Rsug4PV1LXWMMD9aW5s499tHs3lzC83NHfzzRXcy/7GJ/OHZcfUOrWry9tSybNMyIs6IiI0R8SNKAzhPTpqY5VzN9lPY5t7eB2xi5YsDWb18EG2tTcy5dRSHHP1qvcOquz0Peo0ho9redezBa3flqDNW0DKo9K96+NjWeoSWQ2Lz5lJSH9BcqpXlrU+p14rStJR0YE/fRcTCngrewRS2ubfTrq2sXTlw6/66VS3sc+CmOkaUX2teGMwfHxnBb763Oy2DOjj+ghfZff836h1WLjQ1dXD5//sNE3Z9ndvu2oc/PNc4tTHIX42sp6blJT18F8AR1Qggmfr2VIDBDKlGkb2ibpr+eVv6Ki862sRbrw7ga79azLLHh3HVGXvzrQcXdHsP+5uOjiZOP+dYhg7ZwoXn3MfkSRt48aXR9Q6reorSRxYRh/dFABExE5gJMEJj6p4y1q1qYdyELVv3x45v5ZXV7vvpzqjxW9h/xitIMHnqGzQ1BW+sH8DwndrKn9xPvLlpIIuf2oVpU19unETWx83GNFJNdd2fPLNoCBOnbGGXSW8zoKWD6cdt5OFZI+sdVi7t97H1LP39KADWPD+YttYmho1xEhs5YjNDh5T+YzhwYBsH7LeKl15usH9DRekj66862sXlF0zk4uufp6kZZt04hmVL+/cTS4CffvF9PPfQSN7YMIC/P2gax3x1OQf/9X9w3Tl7cvFRU2luCU665Fk3K4ExozZxzlm/o6kpaFJw/0OTmbdwUr3DqipVaWJFSV8FPk8p7T0BnBIRm7OWU7NElkxhO53SclArgAsj4ie1ul41zZ89gvmzR9Q7jFw55dKl3R4/+QfP9nEk+ffC8jGcce5f1DuM2qpCbUvSROBLwL4R8Zakm4ATKL3xkEmaIUqiNNX1HhHxHUnvBXaNiB7fJethClszKzBFVZ9aDgDeI6kVGAKsrKSQNH1kPwQOAToT0+vA5ZVczMwaRPqprne4QG9EvAz8f2A5sAp4NSJmVRJOmqblQRFxoKTHkotvkDSw3Elm1sCqsECvpNHAccAUYCNws6STklmoM0lTI2uV1EwSuqRxVG0NFTMros7mZbmtjD8HXoiItRHRCtwCHFpJPGkS2b8AvwR2lvQPlKbwubiSi5lZA4jSU8s0WxnLgYMlDUn64o+kwsW/06xreZ2kBclFBBwfEV5p3Kw/q0Jnf0TMk/RzYCHQBjxG8nJ8VmmeWr4X2AT8uuuxiFheyQXNrAFU6allRFwIXNjbctJ09t/OO4uQDKbUMfcM8P7eXtzMiqlIg8YBiIgPdt1PZsX4wg5+bmbW5zK/2R8RCyV9qBbBmFlBFK1GJunvuuw2AQcCa2sWkZnlW1RvrGW1pKmRDe/yuY1Sn9kvahOOmRVCkWpkyYuwwyLinD6Kx8xyThSos1/SgIho62nKazPrp4qSyCitlHQgsEjSbcDNwJudX0bELTWOzczyqLqzX1RFmj6yMcArlObo73yfLCiNizKz/qhAnf07J08sn+SdBNYpZ/nYzPpSkWpkzcAw3p3AOuXsr2FmfSpnGaCnRLYqIr7TZ5GYWTHkcBWlnhKZl5Ews24VqWl5ZJ9FYWbFUpREFhHr+zIQMyuOIg5RMjN7R8H6yMzMtiPy14GeZs5+M7N3i5RbGZJGSfq5pD9IWiLpkErCcY3MzDKr4lPLHwB3RcSnkmUmh1RSiBOZmWVXhUQmaQRwGPA/ASJiC7ClkrLctDSzbLItB7fDlcaBPShN0vpTSY9JulLS0EpCciIzs+zS95Gti4hpXbauy70NoDTDzhURcQCl2XXOqyQcJzIzy6xKK42vAFZExLxk/+eUEltmTmRmll0VnlpGxGrgJUl7J4eOBJ6uJBx39hfIV6YdX+8Qcm/K7c/UO4Rce/x/bK5KOVV8avlF4LrkieXzwCmVFOJEZmbZBFWbWDEiFgHTeluOE5mZZVKoxUfMzHbIiczMik6Rr0zmRGZm2Xj2CzNrBO4jM7PC88SKZlZ8rpGZWaEVdKVxM7N3cyIzsyLzC7Fm1hDUka9M5kRmZtn4PTIzawR+/cLMis81MjMrOnf2m1mxBZCzQeOe6trMMsuwilL5sqTmZBWl31Qaj2tkZpZJDd4j+zKwBBhRaQGukZlZNhHptzIk7QZ8AriyNyG5RmZmmVWxRvZ94FxgeG8KcY3MzLJLvxzcDlcal/RJYE1ELOhtOK6RmVlmGWpk6yJiR6skfQQ4VtIxwGBghKRrI+KkrPG4RmZm2QTQHum2noqJOD8idouIycAJwOxKkhi4RmZmFfALsWZWfFV+ITYi5gBzKj3ficzMMnONzMyKzdP4mFnRCVCZjvy+5kRmZpl5pXEzKzY3LYth2vTXOO2ilTQ3BXfeMIabLtul3iHlyle+/RQfPmwdG9cP5Iz/fki9w8mNdRdtYdOD7TSPFhNvHAzAhh+1smluOwiax4ix3xzIgHGqc6S9lW4cZV+q2QuxkiZJuk/SEklPSfpyra5VTU1NwZkXv8w3TpzC307fm8OP28h799pc77By5be3TuDvTz+g3mHkzrBPNLPLDwa969jIkwYw8frBTLxuMEP+tJmNV7bWKbrqUqTb+kot3+xvA74WEf8FOBg4U9K+NbxeVex9wCZWvjiQ1csH0dbaxJxbR3HI0a/WO6xceXLhaF5/raXeYeTO4AObadpmIpqmYe/UvjreClT0ylinKs1+US01a1pGxCpgVfL5dUlLgInA07W6ZjXstGsra1cO3Lq/blUL+xy4qY4RWdFt+GErb9zRTtMw2PWKQeVPyLvI31PLPhlrKWkycAAwry+u1xvd/RczZ90BVjCjz2hh0m8GM3RGM6/d3FbvcKoj/ewXfaLmiUzSMOAXwFci4rVuvj+1c4qPVt6udThlrVvVwrgJW7bujx3fyiur3Yyy3ht2dDObZrfXO4yqUESqra/UNJFJaqGUxK6LiFu6+01EzIyIaRExrYX6V7ufWTSEiVO2sMuktxnQ0sH04zby8KyR9Q7LCqp1+TsT12+a20HL5AbpJOsvfWSSBPwEWBIR/1Sr61RbR7u4/IKJXHz98zQ1w6wbx7Bs6eB6h5Ur5373CfabtoERo1q5ZtYDXHvFHsz65cR6h1V3a7+xhc0L2mnfCC998i1G/W0Lb/2+ndZlAU0wYFex03kDy5aTewH0owV6PwJ8FnhC0qLk2Ncj4o4aXrMq5s8ewfzZFa+D0PD+8bwP1juEXBr3f7ZPUsOPa7xXNUXfNhvTqOVTywcpDcsys0bTka8qWeP958LMaqufNS3NrEHlrWnpOfvNLLsqPLWs5jBG18jMLKOqvVrROYxxoaThwAJJ90RE5tE/TmRmlk3nKkq9LaaKwxidyMwsswx9ZGMlPdplf2ZEzNyuvF4OY3QiM7Ps0ieynhboBcoPY0zDiczMsgmgozpPLdMMY0zDiczMMqpOZ381hzH69Qszy646g8Y7hzEeIWlRsh1TSTiukZlZNgG09/7V/moOY3QiM7OMAiJfY5ScyMwsu5wNUXIiM7NsqvjUslqcyMwsO9fIzKzwnMjMrNAioD1fi6g4kZlZdq6RmVnhOZGZWbGFn1qaWcEFhF+INbPCq8IQpWpyIjOzbCK8HJyZNQB39ptZ0YVrZGZWbFVbRalqnMjMLBsPGjezogsgcjZEyVNdm1k2kUysmGYrQ9IMSc9Iek7SeZWG5BqZmWUWVWhaSmoGLgeOAlYA8yXdVslK466RmVl21amRfRh4LiKej4gtwI3AcZWEo8jR0wdJa4Fl9Y6ji7HAunoHkWO+P+Xl7R7tHhHjelOApLso/b3SGAxs7rK/daVxSZ8CZkTE55P9zwIHRcRZWWPKVdOytze42iQ9Wm6V5P7M96e8RrxHETGjSkV1t4JSRTUrNy3NrF5WAJO67O8GrKykICcyM6uX+cBekqZIGgicANxWSUG5alrm0Mx6B5Bzvj/l+R7tQES0SToLuBtoBq6KiKcqKStXnf1mZpVw09LMCs+JzMwKz4msG9UaNtGoJF0laY2kJ+sdSx5JmiTpPklLJD0l6cv1jqnRuY9sG8mwiaV0GTYBfLqSYRONStJhwBvANRHxgXrHkzeSxgPjI2KhpOHAAuB4/xuqHdfItle1YRONKiLmAuvrHUdeRcSqiFiYfH4dWAJMrG9Ujc2JbHsTgZe67K/A/witQpImAwcA8+ocSkNzItte1YZNWP8maRjwC+ArEfFaveNpZE5k26vasAnrvyS1UEpi10XELfWOp9E5kW2vasMmrH+SJOAnwJKI+Kd6x9MfOJFtIyLagM5hE0uAmyodNtGoJN0APATsLWmFpM/VO6ac+QjwWeAISYuS7Zh6B9XI/PqFmRWea2RmVnhOZGZWeE5kZlZ4TmRmVnhOZGZWeE5kBSKpPXmU/6SkmyUN6UVZVyer2CDpSkn79vDb6ZIOreAaL0rabrWdHR3f5jdvZLzWtySdnTVGawxOZMXyVkRMTWac2AKc1vXLZOaOzCLi82VmZpgOZE5kZn3Fiay4HgD2TGpL90m6HnhCUrOk70maL2mxpC9A6W1zSZdJelrS7cDOnQVJmiNpWvJ5hqSFkh6XdG8y6Pk04KtJbfDPJI2T9IvkGvMlfSQ5dydJsyQ9JunHdD9u9V0k/UrSgmTerlO3+e6SJJZ7JY1Ljv2JpLuScx6QtE9V7qYVW0R4K8gGvJH8OQC4FTidUm3pTWBK8t2pwDeSz4OAR4EpwH8D7qG0yMMEYCPwqeR3c4BpwDhKM390ljUm+fNbwNld4rge+NPk83spDcUB+Bfgm8nnT1AabD+2m7/Hi53Hu1zjPcCTwE7JfgAnJp+/CVyWfL4X2Cv5fBAwu7sYvfWvzasoFct7JC1KPj9AaTzfocAjEfFCcvxjwH6d/V/ASGAv4DDghohoB1ZKmt1N+QcDczvLiogdzTn258C+pSGFAIxIJhA8jFLCJCJul7Qhxd/pS5L+Mvk8KYn1FaAD+Pfk+LXALclsEocCN3e59qAU17AG50RWLG9FxNSuB5L/Q7/Z9RDwxYi4e5vfHUP56YiU4jdQ6pI4JCLe6iaW1GPeJE2nlBQPiYhNkuYAg3fw80iuu3Hbe2DmPrLGczdwejKNDJLeJ2koMBc4IelDGw8c3s25DwEflTQlOXdMcvx1YHiX382iNLCe5HdTk49zgROTYx8HRpeJdSSwIUli+1CqEXZqAjprlZ8BHozSnF4vSPqr5BqStH+Za1g/4ETWeK4EngYWJouD/JhSzfuXwLPAE8AVwP3bnhgRayn1sd0i6XHeadr9GvjLzs5+4EvAtORhwtO88/T028BhkhZSauIuLxPrXcAASYuBi4CHu3z3JvB+SQuAI4DvJMdPBD6XxPcUnobc8OwXZtYAXCMzs8JzIjOzwnMiM7PCcyIzs8JzIjOzwnMiM7PCcyIzs8L7T/D7rLGCALweAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "from sklearn.metrics import ConfusionMatrixDisplay\n",
    "ConfusionMatrixDisplay.from_estimator(classifier, X_test, y_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0f3a3146-89e3-4f5c-b099-245bb4997201",
   "metadata": {},
   "source": [
    "We can see the true (actual) label on the left hand side (0, 1 or 2) and the predicted label on the bottom (0, 1, 2).\n",
    "\n",
    "The diagonal shows the correct predictions (17 + 16 + 13).  The total number of predictions is (17 + 16 + 13 + 1 + 3)\n",
    "\n",
    "Therefore, the score is:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "ea2b9e17-2b81-4efd-85a8-d1d5cb13d606",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.92"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "(17 + 16 + 13) / (17 + 16 + 13 + 1 + 3)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0f4e951d-a99c-498c-b516-752b3f3d7214",
   "metadata": {},
   "source": [
    "### Advanced\n",
    "\n",
    "You can find more information on the confusion matrix here: https://scikit-learn.org/stable/modules/generated/sklearn.metrics.ConfusionMatrixDisplay.html#sklearn.metrics.ConfusionMatrixDisplay.from_predictions\n",
    "\n",
    "Scikit-learn provides 3 different APIs for evaluating the quality of a model’s predictions.  These are discussed in the page:  https://scikit-learn.org/stable/modules/model_evaluation.html\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c247482c-c6c7-465f-a48f-625ef1a57bd0",
   "metadata": {},
   "source": [
    "---\n",
    "\n",
    "### Navigation\n",
    "\n",
    "[Previous](./08_test_and_train_split.ipynb) | [Home](./00-README-FIRST.ipynb) | [Next](./10_model_persistence.ipynb) notebook"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "1ed1e49a-d940-4dae-be95-7007d21b4973",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.9.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}