{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Example-1 (Comparison of three different classifiers)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "A comparison of a 3 classifiers in `scikit-learn` on iris dataset.\n",
    "The iris dataset is a classic and very easy multi-class classification dataset."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Install scikit-learn"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Requirement already satisfied: scikit-learn in c:\\users\\sepkjaer\\appdata\\local\\programs\\python\\python35-32\\lib\\site-packages\\scikit_learn-0.19.1-py3.5-win32.egg (0.19.1)\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "You are using pip version 19.0.2, however version 19.0.3 is available.\n",
      "You should consider upgrading via the 'python -m pip install --upgrade pip' command.\n"
     ]
    }
   ],
   "source": [
    "import sys\n",
    "import os\n",
    "!{sys.executable} -m pip install scikit-learn\n",
    "if \"Example1_Files\" not in os.listdir():\n",
    "    os.mkdir(\"Example1_Files\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Load dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn import datasets\n",
    "from sklearn.model_selection import train_test_split\n",
    "from pycm import ConfusionMatrix\n",
    "iris = datasets.load_iris()\n",
    "X = iris.data\n",
    "y = iris.target\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y, random_state=0)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Classifier 1 (C-Support vector)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn import svm\n",
    "classifier_1 = svm.SVC(kernel='linear', C=0.01)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_pred_1 = classifier_1.fit(X_train, y_train).predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Predict          0     1     2     \n",
      "Actual\n",
      "0                13    0     0     \n",
      "\n",
      "1                0     10    6     \n",
      "\n",
      "2                0     0     9     \n",
      "\n",
      "\n"
     ]
    }
   ],
   "source": [
    "cm1=ConfusionMatrix(y_test,y_pred_1)\n",
    "cm1.print_matrix()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Predict          0        1        2        \n",
      "Actual\n",
      "0                1.0      0.0      0.0      \n",
      "\n",
      "1                0.0      0.625    0.375    \n",
      "\n",
      "2                0.0      0.0      1.0      \n",
      "\n",
      "\n"
     ]
    }
   ],
   "source": [
    "cm1.print_normalized_matrix()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.7673469387755101"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cm1.Kappa "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.8421052631578947"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cm1.Overall_ACC"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'Substantial'"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cm1.SOA1  # Landis and Koch benchmark"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'Excellent'"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cm1.SOA2  # Fleiss’ benchmark"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'Good'"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cm1.SOA3  # Altman’s benchmark"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'Excellent'"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cm1.SOA4  # Cicchetti’s benchmark"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'Message': 'D:\\\\For Asus Laptop\\\\projects\\\\pycm\\\\Document\\\\Example1_Files\\\\cm1.html',\n",
       " 'Status': True}"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cm1.save_html(os.path.join(\"Example1_Files\",\"cm1\"))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<a href=\"Example1_Files\\cm1.html\">Open File</a>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Classifier 2 (Decision tree)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.tree import DecisionTreeClassifier\n",
    "classifier_2 = DecisionTreeClassifier(max_depth=5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_pred_2 = classifier_2.fit(X_train, y_train).predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Predict          0     1     2     \n",
      "Actual\n",
      "0                13    0     0     \n",
      "\n",
      "1                0     15    1     \n",
      "\n",
      "2                0     0     9     \n",
      "\n",
      "\n"
     ]
    }
   ],
   "source": [
    "cm2=ConfusionMatrix(y_test,y_pred_2)\n",
    "cm2.print_matrix()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Predict          0         1         2         \n",
      "Actual\n",
      "0                1.0       0.0       0.0       \n",
      "\n",
      "1                0.0       0.9375    0.0625    \n",
      "\n",
      "2                0.0       0.0       1.0       \n",
      "\n",
      "\n"
     ]
    }
   ],
   "source": [
    "cm2.print_normalized_matrix()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.95978835978836"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cm2.Kappa "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.9736842105263158"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cm2.Overall_ACC"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'Almost Perfect'"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cm2.SOA1  # Landis and Koch benchmark"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'Excellent'"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cm2.SOA2  # Fleiss’ benchmark"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'Very Good'"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cm2.SOA3  # Altman’s benchmark"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'Excellent'"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cm2.SOA4  # Cicchetti’s benchmark"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'Message': 'D:\\\\For Asus Laptop\\\\projects\\\\pycm\\\\Document\\\\Example1_Files\\\\cm2.html',\n",
       " 'Status': True}"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cm2.save_html(os.path.join(\"Example1_Files\",\"cm2\"))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<a href=\"Example1_Files\\cm2.html\">Open File</a>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Classifier 3 (AdaBoost)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\Sepkjaer\\AppData\\Local\\Programs\\Python\\Python35-32\\lib\\site-packages\\scikit_learn-0.19.1-py3.5-win32.egg\\sklearn\\ensemble\\weight_boosting.py:29: DeprecationWarning: numpy.core.umath_tests is an internal NumPy module and should not be imported. It will be removed in a future NumPy release.\n",
      "  from numpy.core.umath_tests import inner1d\n"
     ]
    }
   ],
   "source": [
    "from sklearn.ensemble import AdaBoostClassifier\n",
    "classifier_3 = AdaBoostClassifier()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_pred_3 = classifier_3.fit(X_train, y_train).predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Predict          0     1     2     \n",
      "Actual\n",
      "0                13    0     0     \n",
      "\n",
      "1                0     15    1     \n",
      "\n",
      "2                0     3     6     \n",
      "\n",
      "\n"
     ]
    }
   ],
   "source": [
    "cm3=ConfusionMatrix(y_test,y_pred_3)\n",
    "cm3.print_matrix()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Predict          0          1          2          \n",
      "Actual\n",
      "0                1.0        0.0        0.0        \n",
      "\n",
      "1                0.0        0.9375     0.0625     \n",
      "\n",
      "2                0.0        0.33333    0.66667    \n",
      "\n",
      "\n"
     ]
    }
   ],
   "source": [
    "cm3.print_normalized_matrix()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.8354978354978355"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cm3.Kappa "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.8947368421052632"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cm3.Overall_ACC"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'Almost Perfect'"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cm3.SOA1  # Landis and Koch benchmark"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'Excellent'"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cm3.SOA2  # Fleiss’ benchmark"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'Very Good'"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cm3.SOA3  # Altman’s benchmark"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'Excellent'"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cm3.SOA4  # Cicchetti’s benchmark"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'Message': 'D:\\\\For Asus Laptop\\\\projects\\\\pycm\\\\Document\\\\Example1_Files\\\\cm3.html',\n",
       " 'Status': True}"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cm3.save_html(os.path.join(\"Example1_Files\",\"cm3\"))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<a href=\"Example1_Files\\cm3.html\">Open File</a>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## How to compare classifiers?"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Classifiers can be compared with each other according to results of the benchmarks.\n",
    "The second classifier (DecisionTree) is the best one in this case. PYCM supports different useful parameters such as `Kappa value`, `Scott's pi`, `Entropy`, to name but a handful."
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.5.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}