Classification Linear.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Import tools"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 251,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "from sklearn.datasets import make_classification\n",
    "import pandas as pd\n",
    "import seaborn as sns"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Create data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 252,
   "metadata": {},
   "outputs": [],
   "source": [
    "np.random.seed(1)\n",
    "X, y = make_classification(n_samples=200,n_features=2, n_redundant=0, n_informative=1, n_clusters_per_class=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 253,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(200, 2)"
      ]
     },
     "execution_count": 253,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 254,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(200,)"
      ]
     },
     "execution_count": 254,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 255,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.collections.PathCollection at 0x7f08b6ababa8>"
      ]
     },
     "execution_count": 255,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(8, 6))\n",
    "plt.scatter(X[:,0], X[:, 1], marker='o', c=y, edgecolors='k')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 256,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[ 1.23225307, -2.47410149],\n",
       "       [-0.10679399,  1.21417321],\n",
       "       [-0.89055558, -2.67772702],\n",
       "       [ 1.19268607,  0.56160925],\n",
       "       [-0.8301719 ,  1.15856878],\n",
       "       [ 1.95487808, -1.44594526],\n",
       "       [-0.19505734,  1.10997919],\n",
       "       [-0.34498721, -0.91240907],\n",
       "       [-0.23643695,  0.91176636],\n",
       "       [-1.30653407,  0.71240572]])"
      ]
     },
     "execution_count": 256,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X[:10,:]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 257,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 258,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(160, 2)"
      ]
     },
     "execution_count": 258,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_train.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 259,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(40, 2)"
      ]
     },
     "execution_count": 259,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_test.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Create model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 260,
   "metadata": {},
   "outputs": [],
   "source": [
    "class ClassificationLinearCustom():\n",
    "    \"\"\"\n",
    "        Implementation Classification Linear Model.\n",
    "    \"\"\"\n",
    "    \n",
    "    def __init__(self, learning_rate=0.0001, max_iter=100):\n",
    "        \"\"\"\n",
    "            Initializer parameters model\n",
    "        \"\"\"\n",
    "        self.lrate=learning_rate\n",
    "        self.max_iter=max_iter\n",
    "        self.coefs_=None\n",
    "        self.error_gradient = None\n",
    "        self.is_fited = False\n",
    "        \n",
    "    def sigmoid(self, coefs, x):\n",
    "        \"\"\"\n",
    "            Sigmoid funtion\n",
    "        \"\"\"\n",
    "        return 1/(1+np.exp(-(coefs*x).sum()))\n",
    "        \n",
    "    def error_coefs(self, coefs, xi, yi, m, lr):\n",
    "        \"\"\"\n",
    "            Calcul coefs gradient:\n",
    "            -- Parameters :\n",
    "                - xi : value for one sample in data\n",
    "                - yi : real value prediction for this precedent sample\n",
    "                - m : dataset size\n",
    "                - lr : learning rate\n",
    "        \"\"\"\n",
    "        return lr/m * (xi.T*self.sigmoid(coefs, xi)-yi)\n",
    "        \n",
    "    \n",
    "    def fit(self,x, y):\n",
    "        \"\"\"\n",
    "            Fit fuction, learning parameters\n",
    "            -- Parameters:\n",
    "                - x, sample data\n",
    "                - y, predict data\n",
    "        \"\"\"\n",
    "        #scalling data\n",
    "        x = (x-x.min())/(x.max()-x.min())\n",
    "        \n",
    "        if x.shape[0] != y.shape[0]:\n",
    "            return ValueError(\"x and y must have same sample\")\n",
    "        \n",
    "        m = x.shape[0] # size du dataset\n",
    "        self.coefs_ = 1 # nuers of features\n",
    "        for _ in range(self.max_iter):\n",
    "            for xi,yi in zip(x,y):\n",
    "                self.coefs_ -= self.error_coefs(self.coefs_, xi, yi, m, self.lrate) \n",
    "         \n",
    "        self.is_fited = True\n",
    "        print('ClassificationLinearCustom(learning_rate={}, max_iter={})'.format(self.lrate, self.max_iter))\n",
    "    \n",
    "    def predict(self, x):\n",
    "        \"\"\"\n",
    "            Predict function : \n",
    "            -- Parameters:\n",
    "                - x, sample data what to predict\n",
    "        \"\"\"\n",
    "        ypred_proba = []\n",
    "        if not self.is_fited:\n",
    "            return ValueError(\"model must fited after predict\")\n",
    "        if x.shape[1] != self.coefs_.shape[0]:\n",
    "            return ValueError(\"The features of x do not have the same size as those to train\")\n",
    "        for xi in x:\n",
    "            ypred_proba +=[1] if self.sigmoid(self.coefs_, xi) >= 0.5 else [0]\n",
    "        return np.array(ypred_proba)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Test Model Custom"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 261,
   "metadata": {},
   "outputs": [],
   "source": [
    "cf = ClassificationLinearCustom()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 262,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "ClassificationLinearCustom(learning_rate=0.0001, max_iter=100)\n"
     ]
    }
   ],
   "source": [
    "cf.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 263,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([1.00140568, 1.00136496])"
      ]
     },
     "execution_count": 263,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cf.coefs_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 264,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.metrics import accuracy_score"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 265,
   "metadata": {},
   "outputs": [],
   "source": [
    "yp = cf.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 266,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([1, 1, 0, 1, 1, 0, 0, 0, 1, 1, 0, 1, 1, 0, 1, 1, 1, 0, 0, 1, 1, 1,\n",
       "       1, 1, 1, 1, 1, 0, 1, 1, 1, 1, 1, 0, 0, 1, 0, 1, 0, 1])"
      ]
     },
     "execution_count": 266,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "yp"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Decision boundary training"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 267,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "h = .02\n",
    "colors = \"bry\"\n",
    "x_min, x_max = X_train[:, 0].min() - 1, X_train[:, 0].max() + 1\n",
    "y_min, y_max = X_train[:, 1].min() - 1, X_train[:, 1].max() + 1\n",
    "xx, yy = np.meshgrid(np.arange(x_min, x_max, h), np.arange(y_min, y_max, h))\n",
    "\n",
    "Z = cf.predict(np.c_[xx.ravel(), yy.ravel()])\n",
    "Z = Z.reshape(xx.shape)\n",
    "cs = plt.contourf(xx, yy, Z, cmap=plt.cm.Paired)\n",
    "plt.axis('tight')\n",
    "\n",
    "for i, color in zip([0,1], colors):\n",
    "    idx = np.where(y_train == i)\n",
    "    plt.scatter(X_train[idx, 0], X_train[idx, 1], c=color, cmap=plt.cm.Paired, edgecolor='black', s=20) "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 268,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.metrics import classification_report, f1_score, confusion_matrix"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 269,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.92\n"
     ]
    }
   ],
   "source": [
    "print(f1_score(y_test, yp))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 270,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "True Negative : 13\n",
      "False Positive : 4\n",
      "False Negative : 0\n",
      "True Positive : 23\n"
     ]
    }
   ],
   "source": [
    "tn, fp, fn, tp = confusion_matrix(y_test, yp).ravel()\n",
    "print('True Negative :', tn)\n",
    "print('False Positive :', fp)\n",
    "print('False Negative :', fn)\n",
    "print('True Positive :', tp)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 271,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       1.00      0.76      0.87        17\n",
      "           1       0.85      1.00      0.92        23\n",
      "\n",
      "    accuracy                           0.90        40\n",
      "   macro avg       0.93      0.88      0.89        40\n",
      "weighted avg       0.91      0.90      0.90        40\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print(classification_report(y_test, yp))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Model Sklearn"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 272,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.neighbors import KNeighborsClassifier\n",
    "from sklearn.linear_model import SGDClassifier, LogisticRegression\n",
    "from sklearn.svm import SVC"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 273,
   "metadata": {},
   "outputs": [],
   "source": [
    "cfk = KNeighborsClassifier()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 274,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "KNeighborsClassifier(algorithm='auto', leaf_size=30, metric='minkowski',\n",
       "                     metric_params=None, n_jobs=None, n_neighbors=5, p=2,\n",
       "                     weights='uniform')"
      ]
     },
     "execution_count": 274,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cfk.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 275,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.9777777777777777"
      ]
     },
     "execution_count": 275,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "f1_score(y_test, cfk.predict(X_test))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 276,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "True Negative : 17\n",
      "False Positive : 0\n",
      "False Negative : 1\n",
      "True Positive : 22\n"
     ]
    }
   ],
   "source": [
    "tn, fp, fn, tp = confusion_matrix(y_test, cfk.predict(X_test)).ravel()\n",
    "print('True Negative :', tn)\n",
    "print('False Positive :', fp)\n",
    "print('False Negative :', fn)\n",
    "print('True Positive :', tp)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 277,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.94      1.00      0.97        17\n",
      "           1       1.00      0.96      0.98        23\n",
      "\n",
      "    accuracy                           0.97        40\n",
      "   macro avg       0.97      0.98      0.97        40\n",
      "weighted avg       0.98      0.97      0.98        40\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print(classification_report(y_test, cfk.predict(X_test)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 278,
   "metadata": {},
   "outputs": [],
   "source": [
    "cfks = SGDClassifier()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 279,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "SGDClassifier(alpha=0.0001, average=False, class_weight=None,\n",
       "              early_stopping=False, epsilon=0.1, eta0=0.0, fit_intercept=True,\n",
       "              l1_ratio=0.15, learning_rate='optimal', loss='hinge',\n",
       "              max_iter=1000, n_iter_no_change=5, n_jobs=None, penalty='l2',\n",
       "              power_t=0.5, random_state=None, shuffle=True, tol=0.001,\n",
       "              validation_fraction=0.1, verbose=0, warm_start=False)"
      ]
     },
     "execution_count": 279,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cfks.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 280,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.9565217391304348"
      ]
     },
     "execution_count": 280,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "f1_score(y_test, cfks.predict(X_test))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 281,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "True Negative : 16\n",
      "False Positive : 1\n",
      "False Negative : 1\n",
      "True Positive : 22\n"
     ]
    }
   ],
   "source": [
    "tn, fp, fn, tp = confusion_matrix(y_test, cfks.predict(X_test)).ravel()\n",
    "print('True Negative :', tn)\n",
    "print('False Positive :', fp)\n",
    "print('False Negative :', fn)\n",
    "print('True Positive :', tp)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 282,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.94      0.94      0.94        17\n",
      "           1       0.96      0.96      0.96        23\n",
      "\n",
      "    accuracy                           0.95        40\n",
      "   macro avg       0.95      0.95      0.95        40\n",
      "weighted avg       0.95      0.95      0.95        40\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print(classification_report(y_test, cfks.predict(X_test)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.0"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}