notebooks/ml_metrics_classification_pt1.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<div style='text-align: right'><i>2023-04</i></div>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# ML - Performance Measures for Classification Tasks - pt1"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<a id='content'></a>\n",
    "\n",
    "## Content\n",
    "\n",
    "\n",
    "### [Overview](#1)\n",
    "\n",
    "### [Multi-class classification with Scikit-learn](#2)\n",
    "\n",
    "### [Analyze Performance Measures](#3)\n",
    "\n",
    "### [References](#refs)\n",
    "\n",
    "### [Next steps](#next_steps)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "___"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<a id='1'></a>\n",
    "\n",
    "# [^](#content) Overview\n",
    "___\n",
    "\n",
    "The task of classifying data entries $x_1, ..., x_n$ into predefined classes $C_1, ..., C_k$ is a  supervised ML task. Commonly, classiﬁcation is divided into: binary; multi-class; multi-labeled, and hierarchical tasks.\n",
    "\n",
    "According \\[Sokolova, 2009\\],\n",
    "\n",
    "> *The correctness of a classification can be evaluated by computing the number of correctly recognized class examples (true positives), the number of correctly recognized examples that do not belong to the class (true negatives), and examples that either were incorrectly assigned to the class (false positives) or that were not recognized as class examples (false negatives).*\n",
    "\n",
    "\n",
    "**Note:** There are many recent studies that consider other performance metrics as a comparison of algorithms and the feasibility of algorithms in a specific domain. In general, empirical evaluation is one of the most commonly used steps of the methods in this context.\n",
    "\n",
    "\\[Sokolova, 2009\\] Sokolova, M., & Lapalme, G. (2009). A systematic analysis of performance measures for classification tasks. Information processing & management, 45(4), 427-437.\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "____\n",
    "<!-- <div style=\"text-align: center;\">\n",
    "  <img src=\"../images/overview.png\" height=\"100%\" width=\"100%\">\n",
    "</div> -->"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<a id='2'></a>\n",
    "\n",
    "# [^](#content) Multi-class classification with Scikit-learn\n",
    "___"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Load libraries and dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import pandas as pd\n",
    "\n",
    "from sklearn.datasets import load_iris\n",
    "import sklearn.preprocessing as preprocessing\n",
    "\n",
    "import warnings\n",
    "\n",
    "warnings.filterwarnings('ignore') # suppress warnings\n",
    "\n",
    "# parameters for graphic configurations\n",
    "SMALL_SIZE = 14\n",
    "MEDIUM_SIZE = 16\n",
    "BIGGER_SIZE = 18\n",
    "\n",
    "plt.rc('font', size=SMALL_SIZE)          # controls default text sizes\n",
    "plt.rc('axes', titlesize=SMALL_SIZE)     # fontsize of the axes title\n",
    "plt.rc('axes', labelsize=MEDIUM_SIZE)    # fontsize of the x and y labels\n",
    "plt.rc('xtick', labelsize=SMALL_SIZE)    # fontsize of the tick labels\n",
    "plt.rc('ytick', labelsize=SMALL_SIZE)    # fontsize of the tick labels\n",
    "plt.rc('legend', fontsize=SMALL_SIZE)    # legend fontsize\n",
    "plt.rc('figure', titlesize=BIGGER_SIZE)  # fontsize of the figure title"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "# load dataset \n",
    "iris = load_iris()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### About dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      ".. _iris_dataset:\n",
      "\n",
      "Iris plants dataset\n",
      "--------------------\n",
      "\n",
      "**Data Set Characteristics:**\n",
      "\n",
      "    :Number of Instances: 150 (50 in each of three classes)\n",
      "    :Number of Attributes: 4 numeric, predictive attributes and the class\n",
      "    :Attribute Information:\n",
      "        - sepal length in cm\n",
      "        - sepal width in cm\n",
      "        - petal length in cm\n",
      "        - petal width in cm\n",
      "        - class:\n",
      "                - Iris-Setosa\n",
      "                - Iris-Versicolour\n",
      "                - Iris-Virginica\n",
      "                \n",
      "    :Summary Statistics:\n",
      "\n",
      "    ============== ==== ==== ======= ===== ====================\n",
      "                    Min  Max   Mean    SD   Class Correlation\n",
      "    ============== ==== ==== ======= ===== ====================\n",
      "    sepal length:   4.3  7.9   5.84   0.83    0.7826\n",
      "    sepal width:    2.0  4.4   3.05   0.43   -0.4194\n",
      "    petal length:   1.0  6.9   3.76   1.76    0.9490  (high!)\n",
      "    petal width:    0.1  2.5   1.20   0.76    0.9565  (high!)\n",
      "    ============== ==== ==== ======= ===== ====================\n",
      "\n",
      "    :Missing Attribute Values: None\n",
      "    :Class Distribution: 33.3% for each of 3 classes.\n",
      "    :Creator: R.A. Fisher\n",
      "    :Donor: Michael Marshall (MARSHALL%PLU@io.arc.nasa.gov)\n",
      "    :Date: July, 1988\n",
      "\n",
      "The famous Iris database, first used by Sir R.A. Fisher. The dataset is taken\n",
      "from Fisher's paper. Note that it's the same as in R, but not as in the UCI\n",
      "Machine Learning Repository, which has two wrong data points.\n",
      "\n",
      "This is perhaps the best known database to be found in the\n",
      "pattern recognition literature.  Fisher's paper is a classic in the field and\n",
      "is referenced frequently to this day.  (See Duda & Hart, for example.)  The\n",
      "data set contains 3 classes of 50 instances each, where each class refers to a\n",
      "type of iris plant.  One class is linearly separable from the other 2; the\n",
      "latter are NOT linearly separable from each other.\n",
      "\n",
      ".. topic:: References\n",
      "\n",
      "   - Fisher, R.A. \"The use of multiple measurements in taxonomic problems\"\n",
      "     Annual Eugenics, 7, Part II, 179-188 (1936); also in \"Contributions to\n",
      "     Mathematical Statistics\" (John Wiley, NY, 1950).\n",
      "   - Duda, R.O., & Hart, P.E. (1973) Pattern Classification and Scene Analysis.\n",
      "     (Q327.D83) John Wiley & Sons.  ISBN 0-471-22361-1.  See page 218.\n",
      "   - Dasarathy, B.V. (1980) \"Nosing Around the Neighborhood: A New System\n",
      "     Structure and Classification Rule for Recognition in Partially Exposed\n",
      "     Environments\".  IEEE Transactions on Pattern Analysis and Machine\n",
      "     Intelligence, Vol. PAMI-2, No. 1, 67-71.\n",
      "   - Gates, G.W. (1972) \"The Reduced Nearest Neighbor Rule\".  IEEE Transactions\n",
      "     on Information Theory, May 1972, 431-433.\n",
      "   - See also: 1988 MLC Proceedings, 54-64.  Cheeseman et al\"s AUTOCLASS II\n",
      "     conceptual clustering system finds 3 classes in the data.\n",
      "   - Many, many more ...\n"
     ]
    }
   ],
   "source": [
    "print(iris['DESCR'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Iris species: *setosa*, *versicolour*, and *virginica*\n",
    "                \n",
    "<br>\n",
    "\n",
    "<div style=\"text-align: center;\">\n",
    "  <img src=\"../images/iris_dataset.jpeg\" height=\"100%\" width=\"100%\">\n",
    "</div>\n",
    "\n",
    "\n",
    "https://en.wikipedia.org/wiki/Iris_(plant)\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Some of Iris's attributes\n",
    "\n",
    "<br>\n",
    "\n",
    "<div style=\"text-align: center;\">\n",
    "  <img src=\"../images/iris_parts.jpg\" height=\"100%\" width=\"100%\">\n",
    "</div>\n",
    "\n",
    "https://elizabethswildflowerblog.com/2017/05/26/iris-flowers-terminology-and-structure/\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Create a Pandas dataframe from the dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_dataset = pd.DataFrame(data=np.c_[iris['data'], iris['target']], \n",
    "                  columns=iris['feature_names'] + ['class'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "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>sepal length (cm)</th>\n",
       "      <th>sepal width (cm)</th>\n",
       "      <th>petal length (cm)</th>\n",
       "      <th>petal width (cm)</th>\n",
       "      <th>class</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>4.3</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.1</td>\n",
       "      <td>0.1</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>130</th>\n",
       "      <td>7.4</td>\n",
       "      <td>2.8</td>\n",
       "      <td>6.1</td>\n",
       "      <td>1.9</td>\n",
       "      <td>2.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>4.9</td>\n",
       "      <td>3.1</td>\n",
       "      <td>1.5</td>\n",
       "      <td>0.1</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>4.7</td>\n",
       "      <td>3.2</td>\n",
       "      <td>1.3</td>\n",
       "      <td>0.2</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>5.8</td>\n",
       "      <td>4.0</td>\n",
       "      <td>1.2</td>\n",
       "      <td>0.2</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     sepal length (cm)  sepal width (cm)  petal length (cm)  petal width (cm)  \\\n",
       "13                 4.3               3.0                1.1               0.1   \n",
       "130                7.4               2.8                6.1               1.9   \n",
       "9                  4.9               3.1                1.5               0.1   \n",
       "2                  4.7               3.2                1.3               0.2   \n",
       "14                 5.8               4.0                1.2               0.2   \n",
       "\n",
       "     class  \n",
       "13     0.0  \n",
       "130    2.0  \n",
       "9      0.0  \n",
       "2      0.0  \n",
       "14     0.0  "
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_dataset.sample(5)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### See the correlation between variables"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "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>sepal length (cm)</th>\n",
       "      <th>sepal width (cm)</th>\n",
       "      <th>petal length (cm)</th>\n",
       "      <th>petal width (cm)</th>\n",
       "      <th>class</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>sepal length (cm)</th>\n",
       "      <td>1.000</td>\n",
       "      <td>-0.118</td>\n",
       "      <td>0.872</td>\n",
       "      <td>0.818</td>\n",
       "      <td>0.783</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>sepal width (cm)</th>\n",
       "      <td>-0.118</td>\n",
       "      <td>1.000</td>\n",
       "      <td>-0.428</td>\n",
       "      <td>-0.366</td>\n",
       "      <td>-0.427</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>petal length (cm)</th>\n",
       "      <td>0.872</td>\n",
       "      <td>-0.428</td>\n",
       "      <td>1.000</td>\n",
       "      <td>0.963</td>\n",
       "      <td>0.949</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>petal width (cm)</th>\n",
       "      <td>0.818</td>\n",
       "      <td>-0.366</td>\n",
       "      <td>0.963</td>\n",
       "      <td>1.000</td>\n",
       "      <td>0.957</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>class</th>\n",
       "      <td>0.783</td>\n",
       "      <td>-0.427</td>\n",
       "      <td>0.949</td>\n",
       "      <td>0.957</td>\n",
       "      <td>1.000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                   sepal length (cm)  sepal width (cm)  petal length (cm)  \\\n",
       "sepal length (cm)              1.000            -0.118              0.872   \n",
       "sepal width (cm)              -0.118             1.000             -0.428   \n",
       "petal length (cm)              0.872            -0.428              1.000   \n",
       "petal width (cm)               0.818            -0.366              0.963   \n",
       "class                          0.783            -0.427              0.949   \n",
       "\n",
       "                   petal width (cm)  class  \n",
       "sepal length (cm)             0.818  0.783  \n",
       "sepal width (cm)             -0.366 -0.427  \n",
       "petal length (cm)             0.963  0.949  \n",
       "petal width (cm)              1.000  0.957  \n",
       "class                         0.957  1.000  "
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_dataset.corr().round(3)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Values by class"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.0    50\n",
       "1.0    50\n",
       "2.0    50\n",
       "Name: class, dtype: int64"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_dataset['class'].value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Some statistics"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "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>sepal length (cm)</th>\n",
       "      <th>sepal width (cm)</th>\n",
       "      <th>petal length (cm)</th>\n",
       "      <th>petal width (cm)</th>\n",
       "      <th>class</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>150.000000</td>\n",
       "      <td>150.000000</td>\n",
       "      <td>150.000000</td>\n",
       "      <td>150.000000</td>\n",
       "      <td>150.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>5.843333</td>\n",
       "      <td>3.057333</td>\n",
       "      <td>3.758000</td>\n",
       "      <td>1.199333</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>0.828066</td>\n",
       "      <td>0.435866</td>\n",
       "      <td>1.765298</td>\n",
       "      <td>0.762238</td>\n",
       "      <td>0.819232</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>4.300000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.100000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>5.100000</td>\n",
       "      <td>2.800000</td>\n",
       "      <td>1.600000</td>\n",
       "      <td>0.300000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>5.800000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>4.350000</td>\n",
       "      <td>1.300000</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>6.400000</td>\n",
       "      <td>3.300000</td>\n",
       "      <td>5.100000</td>\n",
       "      <td>1.800000</td>\n",
       "      <td>2.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>7.900000</td>\n",
       "      <td>4.400000</td>\n",
       "      <td>6.900000</td>\n",
       "      <td>2.500000</td>\n",
       "      <td>2.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       sepal length (cm)  sepal width (cm)  petal length (cm)  \\\n",
       "count         150.000000        150.000000         150.000000   \n",
       "mean            5.843333          3.057333           3.758000   \n",
       "std             0.828066          0.435866           1.765298   \n",
       "min             4.300000          2.000000           1.000000   \n",
       "25%             5.100000          2.800000           1.600000   \n",
       "50%             5.800000          3.000000           4.350000   \n",
       "75%             6.400000          3.300000           5.100000   \n",
       "max             7.900000          4.400000           6.900000   \n",
       "\n",
       "       petal width (cm)       class  \n",
       "count        150.000000  150.000000  \n",
       "mean           1.199333    1.000000  \n",
       "std            0.762238    0.819232  \n",
       "min            0.100000    0.000000  \n",
       "25%            0.300000    0.000000  \n",
       "50%            1.300000    1.000000  \n",
       "75%            1.800000    2.000000  \n",
       "max            2.500000    2.000000  "
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_dataset.describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Verify missing values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "sepal length (cm)    0\n",
       "sepal width (cm)     0\n",
       "petal length (cm)    0\n",
       "petal width (cm)     0\n",
       "class                0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_dataset.isna().sum()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Multiclass classification"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Prepare dataset for training"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Index(['sepal length (cm)', 'sepal width (cm)', 'petal length (cm)',\n",
      "       'petal width (cm)', 'class'],\n",
      "      dtype='object')\n"
     ]
    }
   ],
   "source": [
    "print(df_dataset.columns)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "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>sepal length (cm)</th>\n",
       "      <th>sepal width (cm)</th>\n",
       "      <th>petal length (cm)</th>\n",
       "      <th>petal width (cm)</th>\n",
       "      <th>class</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>5.1</td>\n",
       "      <td>3.5</td>\n",
       "      <td>1.4</td>\n",
       "      <td>0.2</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>4.9</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.4</td>\n",
       "      <td>0.2</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>4.7</td>\n",
       "      <td>3.2</td>\n",
       "      <td>1.3</td>\n",
       "      <td>0.2</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4.6</td>\n",
       "      <td>3.1</td>\n",
       "      <td>1.5</td>\n",
       "      <td>0.2</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5.0</td>\n",
       "      <td>3.6</td>\n",
       "      <td>1.4</td>\n",
       "      <td>0.2</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>145</th>\n",
       "      <td>6.7</td>\n",
       "      <td>3.0</td>\n",
       "      <td>5.2</td>\n",
       "      <td>2.3</td>\n",
       "      <td>2.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>146</th>\n",
       "      <td>6.3</td>\n",
       "      <td>2.5</td>\n",
       "      <td>5.0</td>\n",
       "      <td>1.9</td>\n",
       "      <td>2.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>147</th>\n",
       "      <td>6.5</td>\n",
       "      <td>3.0</td>\n",
       "      <td>5.2</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>148</th>\n",
       "      <td>6.2</td>\n",
       "      <td>3.4</td>\n",
       "      <td>5.4</td>\n",
       "      <td>2.3</td>\n",
       "      <td>2.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>149</th>\n",
       "      <td>5.9</td>\n",
       "      <td>3.0</td>\n",
       "      <td>5.1</td>\n",
       "      <td>1.8</td>\n",
       "      <td>2.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>150 rows × 5 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "     sepal length (cm)  sepal width (cm)  petal length (cm)  petal width (cm)  \\\n",
       "0                  5.1               3.5                1.4               0.2   \n",
       "1                  4.9               3.0                1.4               0.2   \n",
       "2                  4.7               3.2                1.3               0.2   \n",
       "3                  4.6               3.1                1.5               0.2   \n",
       "4                  5.0               3.6                1.4               0.2   \n",
       "..                 ...               ...                ...               ...   \n",
       "145                6.7               3.0                5.2               2.3   \n",
       "146                6.3               2.5                5.0               1.9   \n",
       "147                6.5               3.0                5.2               2.0   \n",
       "148                6.2               3.4                5.4               2.3   \n",
       "149                5.9               3.0                5.1               1.8   \n",
       "\n",
       "     class  \n",
       "0      0.0  \n",
       "1      0.0  \n",
       "2      0.0  \n",
       "3      0.0  \n",
       "4      0.0  \n",
       "..     ...  \n",
       "145    2.0  \n",
       "146    2.0  \n",
       "147    2.0  \n",
       "148    2.0  \n",
       "149    2.0  \n",
       "\n",
       "[150 rows x 5 columns]"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "X = df_dataset[['sepal length (cm)', 'sepal width (cm)',\n",
    "               'petal length (cm)', 'petal width (cm)']].values\n",
    "\n",
    "y = df_dataset['class'].values"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### See target names correspondent to class numbers"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array(['setosa', 'versicolor', 'virginica'], dtype='<U10')"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "iris['target_names']"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Fit the model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.linear_model import SGDClassifier\n",
    "\n",
    "model_sgd_classifier = SGDClassifier(random_state=42, loss='modified_huber')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Note**: `loss = modified_huber` gives us a probabilistic classifier.\n",
    "\n",
    "https://scikit-learn.org/stable/modules/generated/sklearn.linear_model.SGDClassifier.html"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style>#sk-container-id-1 {color: black;background-color: white;}#sk-container-id-1 pre{padding: 0;}#sk-container-id-1 div.sk-toggleable {background-color: white;}#sk-container-id-1 label.sk-toggleable__label {cursor: pointer;display: block;width: 100%;margin-bottom: 0;padding: 0.3em;box-sizing: border-box;text-align: center;}#sk-container-id-1 label.sk-toggleable__label-arrow:before {content: \"▸\";float: left;margin-right: 0.25em;color: #696969;}#sk-container-id-1 label.sk-toggleable__label-arrow:hover:before {color: black;}#sk-container-id-1 div.sk-estimator:hover label.sk-toggleable__label-arrow:before {color: black;}#sk-container-id-1 div.sk-toggleable__content {max-height: 0;max-width: 0;overflow: hidden;text-align: left;background-color: #f0f8ff;}#sk-container-id-1 div.sk-toggleable__content pre {margin: 0.2em;color: black;border-radius: 0.25em;background-color: #f0f8ff;}#sk-container-id-1 input.sk-toggleable__control:checked~div.sk-toggleable__content {max-height: 200px;max-width: 100%;overflow: auto;}#sk-container-id-1 input.sk-toggleable__control:checked~label.sk-toggleable__label-arrow:before {content: \"▾\";}#sk-container-id-1 div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-1 div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-1 input.sk-hidden--visually {border: 0;clip: rect(1px 1px 1px 1px);clip: rect(1px, 1px, 1px, 1px);height: 1px;margin: -1px;overflow: hidden;padding: 0;position: absolute;width: 1px;}#sk-container-id-1 div.sk-estimator {font-family: monospace;background-color: #f0f8ff;border: 1px dotted black;border-radius: 0.25em;box-sizing: border-box;margin-bottom: 0.5em;}#sk-container-id-1 div.sk-estimator:hover {background-color: #d4ebff;}#sk-container-id-1 div.sk-parallel-item::after {content: \"\";width: 100%;border-bottom: 1px solid gray;flex-grow: 1;}#sk-container-id-1 div.sk-label:hover label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-1 div.sk-serial::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: 0;}#sk-container-id-1 div.sk-serial {display: flex;flex-direction: column;align-items: center;background-color: white;padding-right: 0.2em;padding-left: 0.2em;position: relative;}#sk-container-id-1 div.sk-item {position: relative;z-index: 1;}#sk-container-id-1 div.sk-parallel {display: flex;align-items: stretch;justify-content: center;background-color: white;position: relative;}#sk-container-id-1 div.sk-item::before, #sk-container-id-1 div.sk-parallel-item::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: -1;}#sk-container-id-1 div.sk-parallel-item {display: flex;flex-direction: column;z-index: 1;position: relative;background-color: white;}#sk-container-id-1 div.sk-parallel-item:first-child::after {align-self: flex-end;width: 50%;}#sk-container-id-1 div.sk-parallel-item:last-child::after {align-self: flex-start;width: 50%;}#sk-container-id-1 div.sk-parallel-item:only-child::after {width: 0;}#sk-container-id-1 div.sk-dashed-wrapped {border: 1px dashed gray;margin: 0 0.4em 0.5em 0.4em;box-sizing: border-box;padding-bottom: 0.4em;background-color: white;}#sk-container-id-1 div.sk-label label {font-family: monospace;font-weight: bold;display: inline-block;line-height: 1.2em;}#sk-container-id-1 div.sk-label-container {text-align: center;}#sk-container-id-1 div.sk-container {/* jupyter's `normalize.less` sets `[hidden] { display: none; }` but bootstrap.min.css set `[hidden] { display: none !important; }` so we also need the `!important` here to be able to override the default hidden behavior on the sphinx rendered scikit-learn.org. See: https://github.com/scikit-learn/scikit-learn/issues/21755 */display: inline-block !important;position: relative;}#sk-container-id-1 div.sk-text-repr-fallback {display: none;}</style><div id=\"sk-container-id-1\" class=\"sk-top-container\"><div class=\"sk-text-repr-fallback\"><pre>SGDClassifier(loss=&#x27;modified_huber&#x27;, random_state=42)</pre><b>In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook. <br />On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.</b></div><div class=\"sk-container\" hidden><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-1\" type=\"checkbox\" checked><label for=\"sk-estimator-id-1\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">SGDClassifier</label><div class=\"sk-toggleable__content\"><pre>SGDClassifier(loss=&#x27;modified_huber&#x27;, random_state=42)</pre></div></div></div></div></div>"
      ],
      "text/plain": [
       "SGDClassifier(loss='modified_huber', random_state=42)"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model_sgd_classifier.fit(X, y)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Make prediction of one sample"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Sort a number randomly"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      ">> Number choosed for index: 14\n"
     ]
    }
   ],
   "source": [
    "idx = np.random.randint(0, 150)\n",
    "\n",
    "print(f'>> Number choosed for index: {idx}')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "some_input_sample = X[idx]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Make prediction with the trained model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0.])"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pred_class = model_sgd_classifier.predict([some_input_sample])\n",
    "\n",
    "pred_class"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Check the corresponding label (target)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.0"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# actual\n",
    "y[idx]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      ">> Did the model make the correct prediction?\n",
      "\n",
      "  ~Answer: True\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print(f'\\n>> Did the model make the correct prediction?\\n\\n  ~Answer: {(pred_class == y[idx])[0]}\\n')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Some checks on predictions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_pred_train = model_sgd_classifier.predict(X)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      ">> y_pred_train = \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 1 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 2 1 1 1\n",
      " 1 1 1 1 1 1 1 1 1 2 1 0 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 1 2 2 2 1 2 2 2 2 2 1 2 2 2 1 2 2 2 2 2 2 2 2 2 2 2 2 2 2\n",
      " 2 2]\n",
      "\n",
      "..........................................................................\n",
      "\n",
      ">> y [labels] = \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",
      "\n",
      "..........................................................................\n",
      "\n",
      ">> Correct pred = 142\n",
      "\n",
      ">> Total samples = 150\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print(f'\\n>> y_pred_train = \\n{y_pred_train.astype(int)}\\n')\n",
    "\n",
    "print(37*'..')\n",
    "\n",
    "print(f'\\n>> y [labels] = \\n{df_dataset[\"class\"].astype(int).values}\\n')\n",
    "\n",
    "print(37*'..')\n",
    "\n",
    "print(f'\\n>> Correct pred = {sum(y_pred_train == df_dataset[\"class\"].values)}')\n",
    "\n",
    "print(f'\\n>> Total samples = {len(df_dataset)}\\n')\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "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>sepal length (cm)</th>\n",
       "      <th>sepal width (cm)</th>\n",
       "      <th>petal length (cm)</th>\n",
       "      <th>petal width (cm)</th>\n",
       "      <th>class</th>\n",
       "      <th>class_pred</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>110</th>\n",
       "      <td>6.5</td>\n",
       "      <td>3.2</td>\n",
       "      <td>5.1</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>4.9</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.4</td>\n",
       "      <td>0.2</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>91</th>\n",
       "      <td>6.1</td>\n",
       "      <td>3.0</td>\n",
       "      <td>4.6</td>\n",
       "      <td>1.4</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25</th>\n",
       "      <td>5.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.6</td>\n",
       "      <td>0.2</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>79</th>\n",
       "      <td>5.7</td>\n",
       "      <td>2.6</td>\n",
       "      <td>3.5</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     sepal length (cm)  sepal width (cm)  petal length (cm)  petal width (cm)  \\\n",
       "110                6.5               3.2                5.1               2.0   \n",
       "1                  4.9               3.0                1.4               0.2   \n",
       "91                 6.1               3.0                4.6               1.4   \n",
       "25                 5.0               3.0                1.6               0.2   \n",
       "79                 5.7               2.6                3.5               1.0   \n",
       "\n",
       "     class  class_pred  \n",
       "110    2.0         2.0  \n",
       "1      0.0         0.0  \n",
       "91     1.0         1.0  \n",
       "25     0.0         0.0  \n",
       "79     1.0         1.0  "
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_dataset['class_pred'] = y_pred_train\n",
    "\n",
    "df_dataset.sample(5)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<a id='3'></a>\n",
    "\n",
    "# [^](#content) Analyze Performance Measures\n",
    "___"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import cross_val_predict\n",
    "\n",
    "from sklearn.metrics import confusion_matrix\n",
    "\n",
    "from sklearn import metrics\n",
    "\n",
    "from sklearn.metrics import f1_score\n",
    "\n",
    "import seaborn as sn"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Matrix confusion"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [],
   "source": [
    "confusion_mat = confusion_matrix(y, y_pred_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[49,  1,  0],\n",
       "       [ 1, 47,  2],\n",
       "       [ 0,  4, 46]])"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "confusion_mat"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "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",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>49</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>47</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>46</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    0   1   2\n",
       "0  49   1   0\n",
       "1   1  47   2\n",
       "2   0   4  46"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_confusion_mat = pd.DataFrame(confusion_mat)\n",
    "df_confusion_mat"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array(['setosa', 'versicolor', 'virginica'], dtype='<U10')"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "iris['target_names']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "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>setosa</th>\n",
       "      <th>versicolor</th>\n",
       "      <th>virginica</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>setosa</th>\n",
       "      <td>49</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>versicolor</th>\n",
       "      <td>1</td>\n",
       "      <td>47</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>virginica</th>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>46</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            setosa  versicolor  virginica\n",
       "setosa          49           1          0\n",
       "versicolor       1          47          2\n",
       "virginica        0           4         46"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_confusion_matrix = df_confusion_mat.rename(columns={0: iris['target_names'][0],\n",
    "                                    1: iris['target_names'][1],\n",
    "                                    2: iris['target_names'][2]},\n",
    "                           index={0: iris['target_names'][0],\n",
    "                                  1: iris['target_names'][1],\n",
    "                                  2: iris['target_names'][2]})\n",
    "df_confusion_matrix"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 700x350 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(10, 5), dpi=70)\n",
    "sn.heatmap(df_confusion_matrix, annot=True, cmap=\"crest\", linewidth=1)\n",
    "plt.xlabel('Predictions')\n",
    "plt.ylabel('Actuals')\n",
    "plt.title('Confusion Matrix')\n",
    "plt.tight_layout()\n",
    "plt.savefig('../images/df_confusion_matrix.png', format='png', bbox_inches='tight', dpi=70)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [],
   "source": [
    "row_sums = confusion_mat.sum(axis=1, keepdims=True)\n",
    "norm_confusion_matrix = confusion_mat/row_sums"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_norm_confusion_matrix = pd.DataFrame(norm_confusion_matrix)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_norm_confusion_matrix.columns = df_confusion_matrix.columns\n",
    "df_norm_confusion_matrix.index = df_confusion_matrix.index"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "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>setosa</th>\n",
       "      <th>versicolor</th>\n",
       "      <th>virginica</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>setosa</th>\n",
       "      <td>0.98</td>\n",
       "      <td>0.02</td>\n",
       "      <td>0.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>versicolor</th>\n",
       "      <td>0.02</td>\n",
       "      <td>0.94</td>\n",
       "      <td>0.04</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>virginica</th>\n",
       "      <td>0.00</td>\n",
       "      <td>0.08</td>\n",
       "      <td>0.92</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            setosa  versicolor  virginica\n",
       "setosa        0.98        0.02       0.00\n",
       "versicolor    0.02        0.94       0.04\n",
       "virginica     0.00        0.08       0.92"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_norm_confusion_matrix"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 700x280 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(10, 4), dpi=70)\n",
    "sn.heatmap(df_norm_confusion_matrix, annot=True, cmap=\"PuBu\",  linewidth=1)\n",
    "plt.xlabel('Predictions')\n",
    "plt.ylabel('Actuals')\n",
    "plt.title('Normalized Confusion Matrix')\n",
    "plt.tight_layout()\n",
    "plt.savefig('../images/df_norm_confusion_matrix.png', format='png', bbox_inches='tight', dpi=70)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "https://matplotlib.org/stable/gallery/color/colormap_reference.html"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Compute precision and recall"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "- `Precision`(or `positive predictive value`): is the fraction of relevant instances among the retrieved instances.\n",
    "\n",
    "- `Recall`(or `sensitivity`): is the fraction of relevant instances that were retrieved.\n",
    "\n",
    "- `F1-score` is the harmonic mean of precision and recall.\n",
    "\n",
    "https://en.wikipedia.org/wiki/Precision_and_recall"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### <u>Notes about means:</u>\n",
    "\n",
    "There are many ways of `calculate` the mean of a set of points. The most common is the `arithmetic mean`. A mean not much common in general is the `harmonic mean`.\n",
    "\n",
    "<br>\n",
    "\n",
    "<div style=\"text-align: center;\">\n",
    "  <img src=\"../images/mean_arithmetic.png\" height=\"50%\" width=\"50%\">\n",
    "</div>\n",
    "<div style='text-align: right'>(def. arithmetic mean)</div>\n",
    "\n",
    "<br>\n",
    "\n",
    "<div style=\"text-align: center;\">\n",
    "  <img src=\"../images/mean_harmonic.png\" height=\"50%\" width=\"50%\">\n",
    "</div>\n",
    "<div style='text-align: right'>(def. harmonic mean)</div>\n",
    "\n",
    "\n",
    "**Example 1**:\n",
    "We calculate these means for, e.g., `1, 1, 97` as:\n",
    "\n",
    "mean_arithmetic = (1+1+97)/3 = `33.00`.\n",
    "\n",
    "\n",
    "mean_harmonic = 3 / (1/1 + 1/1 + 1/97) = `1.49`.\n",
    "\n",
    "In this example, we note that the `weights` of these two means change. Harmonic mean give more weight for the minor values `1`.\n",
    "\n",
    "\n",
    "**Example 2**:\n",
    "\n",
    "We calculate these means for, e.g., `1, 97, 97` as:\n",
    "\n",
    "mean_arithmetic = (1+97+97)/3 = `65.00`.\n",
    "\n",
    "\n",
    "mean_harmonic = 3 / (1/1 + 1/97 + 1/97) = `2.94`.\n",
    "\n",
    "Harmonic mean give us more weight to the minor values. On the other hand, arithmetic mean considered all numbers (1, 97 and 97) with the same weight.\n",
    "\n",
    "\n",
    "https://en.wikipedia.org/wiki/Harmonic_mean"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "      setosa       0.98      0.98      0.98        50\n",
      "  versicolor       0.90      0.94      0.92        50\n",
      "   virginica       0.96      0.92      0.94        50\n",
      "\n",
      "    accuracy                           0.95       150\n",
      "   macro avg       0.95      0.95      0.95       150\n",
      "weighted avg       0.95      0.95      0.95       150\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print(metrics.classification_report(y, y_pred_train, \n",
    "                                    target_names=df_confusion_matrix.columns, digits=2))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### About performance measures\n",
    "\n",
    "<br>\n",
    "\n",
    "<div style=\"text-align: center;\">\n",
    "  <img src=\"../images/table.jpg\" height=\"80%\" width=\"100%\">\n",
    "</div>\n",
    "\n",
    "\\[Sokolova, 2009\\] Sokolova, M., & Lapalme, G. (2009). A systematic analysis of performance measures for classification tasks. Information processing & management, 45(4), 427-437."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Some analyzes"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<div style=\"text-align: center;\">\n",
    "  <img src=\"../images/df_confusion_matrix.png\" height=\"50%\" width=\"80%\">\n",
    "</div>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<br>\n",
    "\n",
    "<div style=\"text-align: center;\">\n",
    "  <img src=\"../images/df_norm_confusion_matrix.png\" height=\"50%\" width=\"80%\">\n",
    "</div>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "                    precision   recall   f1-score   support\n",
    "\n",
    "          setosa       0.98      0.98      0.98        50\n",
    "      versicolor       0.90      0.94      0.92        50\n",
    "       virginica       0.96      0.92      0.94        50\n",
    "\n",
    "        accuracy                           0.95       150\n",
    "       macro avg       0.95      0.95      0.95       150\n",
    "    weighted avg       0.95      0.95      0.95       150"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Notes:\n",
    "\n",
    "**1.** Is well-know that increasing precision reduces recall, what is a good choice of `decision threshold` to have a model that increases the fraction of classes that is correctly predicted (or, equivalently, a model that increases the recall)?\n",
    "    Looking at performance measures of our model, maybe Iris *virginica* needs to have a higher recall for this objective.\n",
    "\n",
    "**2.** The choice between having a higher recall or higher accuracy will depend on the context. For example, to predict the diagnosis of a disease, it is necessary to have a model that never returns a wrong result. In this case, the recall has to be high, or the patient with a positive result will be able to take medicine or be operated on without even being sick. On the other hand, if it is interesting to be right enough in the predictions, but there is not so much damage from making mistakes, a model with high precision is sufficient."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Analyze the Trade-off for Precision/Recall for *Iris virginica*"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Training the model to predict only *Iris virginica*"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([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., 0., 0., 0., 0.,\n",
       "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1.,\n",
       "       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.,\n",
       "       1., 1., 1., 1., 1., 1., 1., 1., 1., 1., 1., 1., 1., 1., 1., 2., 2.,\n",
       "       2., 2., 2., 2., 2., 2., 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.,\n",
       "       2., 2., 2., 2., 2., 2., 2., 2., 2., 2., 2., 2., 2., 2.])"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Now label y will contain only 1 or 0, when y == 2 (is *virginica*) or y !=2 (not is *virginica*), respectively"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_virginica = X.copy()\n",
    "y_virginica = [1 if y_i==2.0 else 0 for y_i in y]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[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, 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, 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, 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, 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"
     ]
    }
   ],
   "source": [
    "print(y_virginica)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Create Pandas dataframe"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_viriginica = pd.DataFrame(data=np.c_[X_virginica, y_virginica], \n",
    "                  columns=iris['feature_names'] + ['class'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(150, 5)\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>sepal length (cm)</th>\n",
       "      <th>sepal width (cm)</th>\n",
       "      <th>petal length (cm)</th>\n",
       "      <th>petal width (cm)</th>\n",
       "      <th>class</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>5.1</td>\n",
       "      <td>3.5</td>\n",
       "      <td>1.4</td>\n",
       "      <td>0.2</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>4.9</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.4</td>\n",
       "      <td>0.2</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>4.7</td>\n",
       "      <td>3.2</td>\n",
       "      <td>1.3</td>\n",
       "      <td>0.2</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4.6</td>\n",
       "      <td>3.1</td>\n",
       "      <td>1.5</td>\n",
       "      <td>0.2</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5.0</td>\n",
       "      <td>3.6</td>\n",
       "      <td>1.4</td>\n",
       "      <td>0.2</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   sepal length (cm)  sepal width (cm)  petal length (cm)  petal width (cm)  \\\n",
       "0                5.1               3.5                1.4               0.2   \n",
       "1                4.9               3.0                1.4               0.2   \n",
       "2                4.7               3.2                1.3               0.2   \n",
       "3                4.6               3.1                1.5               0.2   \n",
       "4                5.0               3.6                1.4               0.2   \n",
       "\n",
       "   class  \n",
       "0    0.0  \n",
       "1    0.0  \n",
       "2    0.0  \n",
       "3    0.0  \n",
       "4    0.0  "
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "print(df_viriginica.shape)\n",
    "df_viriginica.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### See correlation matrix"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "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>sepal length (cm)</th>\n",
       "      <th>sepal width (cm)</th>\n",
       "      <th>petal length (cm)</th>\n",
       "      <th>petal width (cm)</th>\n",
       "      <th>class</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>sepal length (cm)</th>\n",
       "      <td>1.00</td>\n",
       "      <td>-0.12</td>\n",
       "      <td>0.87</td>\n",
       "      <td>0.82</td>\n",
       "      <td>0.64</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>sepal width (cm)</th>\n",
       "      <td>-0.12</td>\n",
       "      <td>1.00</td>\n",
       "      <td>-0.43</td>\n",
       "      <td>-0.37</td>\n",
       "      <td>-0.14</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>petal length (cm)</th>\n",
       "      <td>0.87</td>\n",
       "      <td>-0.43</td>\n",
       "      <td>1.00</td>\n",
       "      <td>0.96</td>\n",
       "      <td>0.72</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>petal width (cm)</th>\n",
       "      <td>0.82</td>\n",
       "      <td>-0.37</td>\n",
       "      <td>0.96</td>\n",
       "      <td>1.00</td>\n",
       "      <td>0.77</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>class</th>\n",
       "      <td>0.64</td>\n",
       "      <td>-0.14</td>\n",
       "      <td>0.72</td>\n",
       "      <td>0.77</td>\n",
       "      <td>1.00</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                   sepal length (cm)  sepal width (cm)  petal length (cm)  \\\n",
       "sepal length (cm)               1.00             -0.12               0.87   \n",
       "sepal width (cm)               -0.12              1.00              -0.43   \n",
       "petal length (cm)               0.87             -0.43               1.00   \n",
       "petal width (cm)                0.82             -0.37               0.96   \n",
       "class                           0.64             -0.14               0.72   \n",
       "\n",
       "                   petal width (cm)  class  \n",
       "sepal length (cm)              0.82   0.64  \n",
       "sepal width (cm)              -0.37  -0.14  \n",
       "petal length (cm)              0.96   0.72  \n",
       "petal width (cm)               1.00   0.77  \n",
       "class                          0.77   1.00  "
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_viriginica.corr().round(2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array(['setosa', 'versicolor', 'virginica'], dtype='<U10')"
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "iris['target_names']"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Fit the model to predict only one class (binary classifier)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [],
   "source": [
    "model_sgd_class_virginica = SGDClassifier(random_state=42, loss='modified_huber')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style>#sk-container-id-2 {color: black;background-color: white;}#sk-container-id-2 pre{padding: 0;}#sk-container-id-2 div.sk-toggleable {background-color: white;}#sk-container-id-2 label.sk-toggleable__label {cursor: pointer;display: block;width: 100%;margin-bottom: 0;padding: 0.3em;box-sizing: border-box;text-align: center;}#sk-container-id-2 label.sk-toggleable__label-arrow:before {content: \"▸\";float: left;margin-right: 0.25em;color: #696969;}#sk-container-id-2 label.sk-toggleable__label-arrow:hover:before {color: black;}#sk-container-id-2 div.sk-estimator:hover label.sk-toggleable__label-arrow:before {color: black;}#sk-container-id-2 div.sk-toggleable__content {max-height: 0;max-width: 0;overflow: hidden;text-align: left;background-color: #f0f8ff;}#sk-container-id-2 div.sk-toggleable__content pre {margin: 0.2em;color: black;border-radius: 0.25em;background-color: #f0f8ff;}#sk-container-id-2 input.sk-toggleable__control:checked~div.sk-toggleable__content {max-height: 200px;max-width: 100%;overflow: auto;}#sk-container-id-2 input.sk-toggleable__control:checked~label.sk-toggleable__label-arrow:before {content: \"▾\";}#sk-container-id-2 div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-2 div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-2 input.sk-hidden--visually {border: 0;clip: rect(1px 1px 1px 1px);clip: rect(1px, 1px, 1px, 1px);height: 1px;margin: -1px;overflow: hidden;padding: 0;position: absolute;width: 1px;}#sk-container-id-2 div.sk-estimator {font-family: monospace;background-color: #f0f8ff;border: 1px dotted black;border-radius: 0.25em;box-sizing: border-box;margin-bottom: 0.5em;}#sk-container-id-2 div.sk-estimator:hover {background-color: #d4ebff;}#sk-container-id-2 div.sk-parallel-item::after {content: \"\";width: 100%;border-bottom: 1px solid gray;flex-grow: 1;}#sk-container-id-2 div.sk-label:hover label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-2 div.sk-serial::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: 0;}#sk-container-id-2 div.sk-serial {display: flex;flex-direction: column;align-items: center;background-color: white;padding-right: 0.2em;padding-left: 0.2em;position: relative;}#sk-container-id-2 div.sk-item {position: relative;z-index: 1;}#sk-container-id-2 div.sk-parallel {display: flex;align-items: stretch;justify-content: center;background-color: white;position: relative;}#sk-container-id-2 div.sk-item::before, #sk-container-id-2 div.sk-parallel-item::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: -1;}#sk-container-id-2 div.sk-parallel-item {display: flex;flex-direction: column;z-index: 1;position: relative;background-color: white;}#sk-container-id-2 div.sk-parallel-item:first-child::after {align-self: flex-end;width: 50%;}#sk-container-id-2 div.sk-parallel-item:last-child::after {align-self: flex-start;width: 50%;}#sk-container-id-2 div.sk-parallel-item:only-child::after {width: 0;}#sk-container-id-2 div.sk-dashed-wrapped {border: 1px dashed gray;margin: 0 0.4em 0.5em 0.4em;box-sizing: border-box;padding-bottom: 0.4em;background-color: white;}#sk-container-id-2 div.sk-label label {font-family: monospace;font-weight: bold;display: inline-block;line-height: 1.2em;}#sk-container-id-2 div.sk-label-container {text-align: center;}#sk-container-id-2 div.sk-container {/* jupyter's `normalize.less` sets `[hidden] { display: none; }` but bootstrap.min.css set `[hidden] { display: none !important; }` so we also need the `!important` here to be able to override the default hidden behavior on the sphinx rendered scikit-learn.org. See: https://github.com/scikit-learn/scikit-learn/issues/21755 */display: inline-block !important;position: relative;}#sk-container-id-2 div.sk-text-repr-fallback {display: none;}</style><div id=\"sk-container-id-2\" class=\"sk-top-container\"><div class=\"sk-text-repr-fallback\"><pre>SGDClassifier(loss=&#x27;modified_huber&#x27;, random_state=42)</pre><b>In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook. <br />On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.</b></div><div class=\"sk-container\" hidden><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-2\" type=\"checkbox\" checked><label for=\"sk-estimator-id-2\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">SGDClassifier</label><div class=\"sk-toggleable__content\"><pre>SGDClassifier(loss=&#x27;modified_huber&#x27;, random_state=42)</pre></div></div></div></div></div>"
      ],
      "text/plain": [
       "SGDClassifier(loss='modified_huber', random_state=42)"
      ]
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model_sgd_class_virginica.fit(X_virginica, y_virginica)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Make predictions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      ">> y_pred_train_virginica = \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 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 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1 1 1\n",
      " 1 1 1 1 1 1 1 1 1 1 1 1 0 1 1 0 1 1 0 1 0 1 0 1 1 1 1 0 1 1 1 1 1 1 1 1 1\n",
      " 1 1]\n",
      "\n",
      "..........................................................................\n",
      "\n",
      ">> y_virginica [labels] = \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 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 0 0 0 0 0 0 0 0 0 0 0 0 0 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 1 1 1 1 1 1 1 1 1 1 1\n",
      " 1 1]\n",
      "\n",
      "..........................................................................\n",
      "\n",
      ">> Correct pred_virginica = 143\n",
      "\n",
      ">> Total samples = 150\n",
      "\n"
     ]
    }
   ],
   "source": [
    "y_pred_train_virginica = model_sgd_class_virginica.predict(X_virginica)\n",
    "print(f'\\n>> y_pred_train_virginica = \\n{y_pred_train_virginica.astype(int)}\\n')\n",
    "\n",
    "print(37*'..')\n",
    "\n",
    "print(f'\\n>> y_virginica [labels] = \\n{df_viriginica[\"class\"].astype(int).values}\\n')\n",
    "\n",
    "print(37*'..')\n",
    "\n",
    "print(f'\\n>> Correct pred_virginica = {sum(y_pred_train_virginica == df_viriginica[\"class\"].values)}')\n",
    "\n",
    "print(f'\\n>> Total samples = {len(df_dataset)}\\n')\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Why `Correct pred_virginica > 50 `?\n",
    "\n",
    "We can remember that the model should predict a binary classification: `is` *virginica* (try 50 samples to try to classify correctly) or `is not` *virginica* (there 100 samples that are not *virginica* but others species of Iris).\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Matrix confusion"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 700x350 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# compute confusion matrix\n",
    "confusion_matrix_virginica = confusion_matrix(y_virginica, y_pred_train_virginica)\n",
    "\n",
    "# dataframe of matrix confusion\n",
    "df_confusion_matrix_virginica = pd.DataFrame(confusion_matrix_virginica)\n",
    "\n",
    "# rename cols\n",
    "df_confusion_matrix_virginica = df_confusion_matrix_virginica.rename(columns=\n",
    "                                {0: 'Not virginica',\n",
    "                                 1: 'virginica'},                                 \n",
    "                                 index={0: 'Not virginica',\n",
    "                                 1: 'virginica'})\n",
    "# plot confusion matrix\n",
    "plt.figure(figsize=(10, 5), dpi=70)\n",
    "sn.heatmap(df_confusion_matrix_virginica, annot=True, cmap=\"crest\", linewidth=1)\n",
    "plt.xlabel('Predictions')\n",
    "plt.ylabel('Actuals')\n",
    "plt.title('Confusion Matrix Iris virginica')\n",
    "plt.tight_layout()\n",
    "plt.savefig('../images/df_confusion_matrix_virgica.png', \n",
    "            format='png', bbox_inches='tight', dpi=70)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Compute precision and recall"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "               precision    recall  f1-score   support\n",
      "\n",
      "Not virginica       0.94      0.99      0.97       100\n",
      "    virginica       0.98      0.88      0.93        50\n",
      "\n",
      "     accuracy                           0.95       150\n",
      "    macro avg       0.96      0.94      0.95       150\n",
      " weighted avg       0.95      0.95      0.95       150\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print(metrics.classification_report(y_virginica, y_pred_train_virginica, \n",
    "                                    target_names=df_confusion_matrix_virginica.columns,\n",
    "                                    digits=2))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Note:**\n",
    "\n",
    "Considering a binary classifier, the recall for identifying *virginica* decreases. So the previous model for multiclasses looks better in this case."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Analyze precision/recall trade-off"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Compute precision and recall for a range of thresholds"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_pred_scores_virginica = model_sgd_class_virginica.predict_proba(X_virginica)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Take positive class"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [0., 1.],\n",
       "       [0., 1.],\n",
       "       [0., 1.],\n",
       "       [0., 1.],\n",
       "       [0., 1.],\n",
       "       [0., 1.],\n",
       "       [0., 1.],\n",
       "       [0., 1.],\n",
       "       [0., 1.],\n",
       "       [0., 1.]])"
      ]
     },
     "execution_count": 50,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_pred_scores_virginica[90:110]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_pred_scores = y_pred_scores_virginica[:, 1] # score means the proba of positive class"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 1., 1., 1., 1., 1., 1.,\n",
       "       1., 1., 1.])"
      ]
     },
     "execution_count": 52,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_pred_scores[90:110]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Computes over thresholds"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [],
   "source": [
    "precisions, recalls, thresholds = metrics.precision_recall_curve(y_virginica,\n",
    "                                                                 y_pred_scores)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Plot precision and recall by threshold"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1350x450 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(15, 5), dpi=90)\n",
    "plt.plot(thresholds, precisions[:-1], 'o--', c='orange', label='Precision', linewidth=3)\n",
    "plt.plot(thresholds, recalls[:-1], 'g-', label='Recall', linewidth=3)\n",
    "plt.xlabel('Threshold')\n",
    "plt.legend()\n",
    "plt.ylim([0.25, 1.25])\n",
    "plt.xlim([-0.25, 1.25])\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Plot precision by recall"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1350x450 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(15, 5), dpi=90)\n",
    "plt.plot(recalls, precisions, 'o--', c='orange', linewidth=3)\n",
    "plt.xlabel('Recall')\n",
    "plt.ylabel('Precision')\n",
    "plt.ylim([0.25, 1.25])\n",
    "plt.xlim([-0.25, 1.25])\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## ROC curves"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.metrics import roc_curve\n",
    "\n",
    "fpr, tpr, thresholds = roc_curve(y_virginica, y_pred_scores)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1350x450 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(15, 5), dpi=90)\n",
    "plt.plot(fpr, tpr, color='orange', linewidth=3)\n",
    "plt.plot([0, 1], [0, 1], 'k--', linewidth=3)\n",
    "plt.axis([0, 1, 0, 1])\n",
    "plt.xlabel('False Positive Rate')\n",
    "plt.ylabel('True Positive Rate')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Area under ROC curve (AUC score)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.9449"
      ]
     },
     "execution_count": 58,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.metrics import roc_auc_score\n",
    "\n",
    "roc_auc_score(y_virginica, y_pred_scores)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Include more one model for analyze"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.ensemble import RandomForestClassifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {},
   "outputs": [],
   "source": [
    "model_rf_classifier = RandomForestClassifier(n_estimators=4, random_state=42)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style>#sk-container-id-3 {color: black;background-color: white;}#sk-container-id-3 pre{padding: 0;}#sk-container-id-3 div.sk-toggleable {background-color: white;}#sk-container-id-3 label.sk-toggleable__label {cursor: pointer;display: block;width: 100%;margin-bottom: 0;padding: 0.3em;box-sizing: border-box;text-align: center;}#sk-container-id-3 label.sk-toggleable__label-arrow:before {content: \"▸\";float: left;margin-right: 0.25em;color: #696969;}#sk-container-id-3 label.sk-toggleable__label-arrow:hover:before {color: black;}#sk-container-id-3 div.sk-estimator:hover label.sk-toggleable__label-arrow:before {color: black;}#sk-container-id-3 div.sk-toggleable__content {max-height: 0;max-width: 0;overflow: hidden;text-align: left;background-color: #f0f8ff;}#sk-container-id-3 div.sk-toggleable__content pre {margin: 0.2em;color: black;border-radius: 0.25em;background-color: #f0f8ff;}#sk-container-id-3 input.sk-toggleable__control:checked~div.sk-toggleable__content {max-height: 200px;max-width: 100%;overflow: auto;}#sk-container-id-3 input.sk-toggleable__control:checked~label.sk-toggleable__label-arrow:before {content: \"▾\";}#sk-container-id-3 div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-3 div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-3 input.sk-hidden--visually {border: 0;clip: rect(1px 1px 1px 1px);clip: rect(1px, 1px, 1px, 1px);height: 1px;margin: -1px;overflow: hidden;padding: 0;position: absolute;width: 1px;}#sk-container-id-3 div.sk-estimator {font-family: monospace;background-color: #f0f8ff;border: 1px dotted black;border-radius: 0.25em;box-sizing: border-box;margin-bottom: 0.5em;}#sk-container-id-3 div.sk-estimator:hover {background-color: #d4ebff;}#sk-container-id-3 div.sk-parallel-item::after {content: \"\";width: 100%;border-bottom: 1px solid gray;flex-grow: 1;}#sk-container-id-3 div.sk-label:hover label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-3 div.sk-serial::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: 0;}#sk-container-id-3 div.sk-serial {display: flex;flex-direction: column;align-items: center;background-color: white;padding-right: 0.2em;padding-left: 0.2em;position: relative;}#sk-container-id-3 div.sk-item {position: relative;z-index: 1;}#sk-container-id-3 div.sk-parallel {display: flex;align-items: stretch;justify-content: center;background-color: white;position: relative;}#sk-container-id-3 div.sk-item::before, #sk-container-id-3 div.sk-parallel-item::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: -1;}#sk-container-id-3 div.sk-parallel-item {display: flex;flex-direction: column;z-index: 1;position: relative;background-color: white;}#sk-container-id-3 div.sk-parallel-item:first-child::after {align-self: flex-end;width: 50%;}#sk-container-id-3 div.sk-parallel-item:last-child::after {align-self: flex-start;width: 50%;}#sk-container-id-3 div.sk-parallel-item:only-child::after {width: 0;}#sk-container-id-3 div.sk-dashed-wrapped {border: 1px dashed gray;margin: 0 0.4em 0.5em 0.4em;box-sizing: border-box;padding-bottom: 0.4em;background-color: white;}#sk-container-id-3 div.sk-label label {font-family: monospace;font-weight: bold;display: inline-block;line-height: 1.2em;}#sk-container-id-3 div.sk-label-container {text-align: center;}#sk-container-id-3 div.sk-container {/* jupyter's `normalize.less` sets `[hidden] { display: none; }` but bootstrap.min.css set `[hidden] { display: none !important; }` so we also need the `!important` here to be able to override the default hidden behavior on the sphinx rendered scikit-learn.org. See: https://github.com/scikit-learn/scikit-learn/issues/21755 */display: inline-block !important;position: relative;}#sk-container-id-3 div.sk-text-repr-fallback {display: none;}</style><div id=\"sk-container-id-3\" class=\"sk-top-container\"><div class=\"sk-text-repr-fallback\"><pre>RandomForestClassifier(n_estimators=4, random_state=42)</pre><b>In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook. <br />On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.</b></div><div class=\"sk-container\" hidden><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-3\" type=\"checkbox\" checked><label for=\"sk-estimator-id-3\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">RandomForestClassifier</label><div class=\"sk-toggleable__content\"><pre>RandomForestClassifier(n_estimators=4, random_state=42)</pre></div></div></div></div></div>"
      ],
      "text/plain": [
       "RandomForestClassifier(n_estimators=4, random_state=42)"
      ]
     },
     "execution_count": 61,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model_rf_classifier.fit(X_virginica, y_virginica)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(150, 4)"
      ]
     },
     "execution_count": 62,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_virginica.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_pred_scores_rf = model_rf_classifier.predict_proba(X_virginica)\n",
    "\n",
    "y_scores_rf = y_pred_scores_rf[:, 1]\n",
    "\n",
    "fpr_rf, tpr_rf, thresholds_rf = roc_curve(y_virginica, y_scores_rf)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1350x450 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(15, 5), dpi=90)\n",
    "plt.plot(fpr, tpr, color='orange', linewidth=3, label='SGDClassifier')\n",
    "plt.plot(fpr_rf, tpr_rf, color='red', linewidth=5, label='RandomForestClassifier')\n",
    "plt.plot([0, 1], [0, 1], 'k--', linewidth=3)\n",
    "plt.axis([0, 1, 0, 1])\n",
    "plt.xlabel('False Positive Rate')\n",
    "plt.ylabel('True Positive Rate')\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Random Forest AUC score "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1.0"
      ]
     },
     "execution_count": 65,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "roc_auc_score(y_virginica, y_scores_rf)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Note:** As a Random Forest model, it had an AUC score = 1, which indicates an overfit to the training set. In this way, if we apply the model to a new data set (different from the training one), the prediction should not be good. Additionally, we can analyze the Precision/Recall of this new model."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_pred_rf = model_rf_classifier.predict(X_virginica)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "               precision    recall  f1-score   support\n",
      "\n",
      "Not virginica       1.00      1.00      1.00       100\n",
      "    virginica       1.00      1.00      1.00        50\n",
      "\n",
      "     accuracy                           1.00       150\n",
      "    macro avg       1.00      1.00      1.00       150\n",
      " weighted avg       1.00      1.00      1.00       150\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print(metrics.classification_report(y_virginica, y_pred_rf, \n",
    "                                    target_names=df_confusion_matrix_virginica.columns,\n",
    "                                    digits=2))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "___"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### QUESTION: <font color='red'>¿¿¿</font>  Is it good to have this precision/recall/f1-score\n",
    "\n",
    "                         precision   recall   f1-score}   support\n",
    "\n",
    "        Not virginica       1.00      1.00      1.00       100\n",
    "            virginica       1.00      1.00      1.00        50\n",
    "\n",
    "             accuracy                           1.00       150\n",
    "            macro avg       1.00      1.00      1.00       150\n",
    "         weighted avg       1.00      1.00      1.00       150\n",
    "\n",
    "\n",
    "# <font color='red'>???</font>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<a id='refs'></a>\n",
    "\n",
    "# [^](#content) References\n",
    "\n",
    "* Sokolova, M., & Lapalme, G. (2009). A systematic analysis of performance measures for classification tasks. Information processing & management, 45(4), 427-437.\n",
    "<br>\n",
    "\n",
    "* Shmueli, B. Multi-Class Metrics Made Simple, Part II: the F1-score, https://towardsdatascience.com/multi-class-metrics-made-simple-part-ii-the-f1-score-ebe8b2c2ca1, Jul 3, 2019.\n",
    "<br>\n",
    "\n",
    "* https://en.wikipedia.org/wiki/Precision_and_recall, 2023.\n",
    "<br>\n",
    "\n",
    "* https://en.wikipedia.org/wiki/Iris_(plant), 2023.\n",
    "<br>\n",
    "\n",
    "* https://en.wikipedia.org/wiki/Harmonic_mean\n",
    "<br>\n",
    "\n",
    "*  https://elizabethswildflowerblog.com/2017/05/26/iris-flowers-terminology-and-structure/, 2023.\n",
    "<br>\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<a id='next_steps'></a>\n",
    "\n",
    "# [^](#content) Next steps\n",
    "\n",
    "\n",
    "- Add new models to the comparisons.\n",
    " \n",
    "- Make predicts on a test dataset of Iris.\n",
    "\n",
    "- Analyze other datasets and add new measures of performance.\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "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": "3.9.12"
  },
  "vscode": {
   "interpreter": {
    "hash": "718d6cbc2078c2b1c166f1cfa6dc034fb8c4b7a8ddcc37e1de1354929db72bca"
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}