iris.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Problema retirado desse site https://archive.ics.uci.edu/ml/datasets/Iris"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Baseado em características de uma flor, vamos classificar qual tipo de flor ela é."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Vamos resolver esse problema de classificação usando o SVM (support vector machine)\n",
    "\n",
    "Mais informações esse site pode ser útil: https://www.analyticsvidhya.com/blog/2017/09/understaing-support-vector-machine-example-code/"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.svm import SVC\n",
    "from sklearn.model_selection  import  cross_val_score \n",
    "from sklearn import metrics as mt\n",
    "from sklearn.preprocessing import StandardScaler"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Colocando labels para visualizar melhor o dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "colunas = ['s_length',\n",
    "           's_width',\n",
    "           'p_length',\n",
    "           'p_width',\n",
    "           'y']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "data = pd.read_csv('iris.data', names=colunas)"
   ]
  },
  {
   "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>s_length</th>\n",
       "      <th>s_width</th>\n",
       "      <th>p_length</th>\n",
       "      <th>p_width</th>\n",
       "      <th>y</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>Iris-setosa</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>Iris-setosa</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>Iris-setosa</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>Iris-setosa</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>Iris-setosa</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   s_length  s_width  p_length  p_width            y\n",
       "0       5.1      3.5       1.4      0.2  Iris-setosa\n",
       "1       4.9      3.0       1.4      0.2  Iris-setosa\n",
       "2       4.7      3.2       1.3      0.2  Iris-setosa\n",
       "3       4.6      3.1       1.5      0.2  Iris-setosa\n",
       "4       5.0      3.6       1.4      0.2  Iris-setosa"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "s_length    0\n",
       "s_width     0\n",
       "p_length    0\n",
       "p_width     0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.isnull().sum()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Como não temos dados faltantes e não precisamos alterar nada no dataset, podemos dividir os dados"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "y = data['y'].values\n",
    "data = data.drop('y', axis=1) \n",
    "X = data.values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42, stratify=y)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Criando o modelo e utilizando o kernel. Sendo Kernel uma função do svm para reorganizar os dados de uma forma que consiga construir um plano capaz de dividir os dados."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "svm = SVC(kernel='rbf', random_state=1, gamma=0.2, C=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "SVC(C=1, cache_size=200, class_weight=None, coef0=0.0,\n",
       "  decision_function_shape='ovr', degree=3, gamma=0.2, kernel='rbf',\n",
       "  max_iter=-1, probability=False, random_state=1, shrinking=True,\n",
       "  tol=0.001, verbose=False)"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "svm.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Agora vamos fazer com cross validation. Uma técnica para poder verificar como está a resposta do nosso modelo sem causar um overfitting. Essa técnica conssiste em pegar o conjunto de treino e fazer vários conjuntos com diferentes partes retiradas, como pode ser visto na próxima figura"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "![Drag Racing](crossover.png)\n",
    "\n",
    "Imagem retirada do site: https://scikit-learn.org/stable/modules/cross_validation.html"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [],
   "source": [
    "scores = cross_val_score(svm, X_train, y_train, cv=10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([1.        , 1.        , 1.        , 1.        , 0.91666667,\n",
       "       1.        , 1.        , 1.        , 0.91666667, 1.        ])"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "scores"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Acurácia cross validation: 0.98 (+/- 0.07)\n"
     ]
    }
   ],
   "source": [
    "print(\"Acurácia cross validation: %0.2f (+/- %0.2f)\" % (scores.mean(), scores.std() * 2))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_pred = svm.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Acurácia: 0.97\n"
     ]
    }
   ],
   "source": [
    "print(\"Acurácia: %0.2f\" % mt.accuracy_score(y_test, y_pred))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Fazendo standartização dos dados e verificando se os resultados melhoram"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [],
   "source": [
    "stdsc = StandardScaler()\n",
    "X_train_std = stdsc.fit_transform(X_train)\n",
    "X_test_std = stdsc.transform(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [],
   "source": [
    "svm_std = SVC(kernel='rbf', random_state=1, gamma=0.2, C=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "SVC(C=1, cache_size=200, class_weight=None, coef0=0.0,\n",
       "  decision_function_shape='ovr', degree=3, gamma=0.2, kernel='rbf',\n",
       "  max_iter=-1, probability=False, random_state=1, shrinking=True,\n",
       "  tol=0.001, verbose=False)"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "svm_std.fit(X_train_std, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [],
   "source": [
    "scores_std = cross_val_score(svm_std, X_train_std, y_train, cv=10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0.83333333, 1.        , 1.        , 0.91666667, 0.91666667,\n",
       "       1.        , 0.91666667, 1.        , 1.        , 1.        ])"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "scores_std"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Acurácia cross validation:: 0.96 (+/- 0.11)\n"
     ]
    }
   ],
   "source": [
    "print(\"Acurácia cross validation:: %0.2f (+/- %0.2f)\" % (scores_std.mean(), scores_std.std() * 2))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_predStd = svm_std.predict(X_test_std)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Acurácia: 0.97\n"
     ]
    }
   ],
   "source": [
    "print(\"Acurácia: %0.2f\" % mt.accuracy_score(y_test, y_predStd))"
   ]
  }
 ],
 "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": "3.6.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}