KNN-SVM-LogisticRegression.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "from sklearn.neighbors import KNeighborsClassifier\n",
    "from sklearn.svm import SVC\n",
    "from sklearn.linear_model import LogisticRegression"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "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": [
    "samples_per_cluster = 20\n",
    "range_len = 10\n",
    "var = 0.05\n",
    "\n",
    "X = np.zeros(((range_len + 1) ** 2 * samples_per_cluster, 2))\n",
    "y = np.zeros((range_len + 1) ** 2 * samples_per_cluster)\n",
    "for x_cord in range(0, range_len + 1):\n",
    "    for y_cord in range(0, range_len + 1):\n",
    "        #print(y_cord, x_cord)\n",
    "        X[(y_cord + x_cord * (range_len + 1)) * samples_per_cluster:(y_cord + x_cord * (range_len + 1) + 1) * samples_per_cluster] = np.random.multivariate_normal(\n",
    "            (x_cord - int(range_len / 2), y_cord - int(range_len / 2)), cov=[[var, 0], [0, var]], size=samples_per_cluster)\n",
    "        y[(y_cord + x_cord * (range_len + 1)) * samples_per_cluster:(y_cord + x_cord * (range_len + 1) + 1) * samples_per_cluster] = (x_cord + y_cord) % 2\n",
    "\n",
    "plt.plot(X[y == 0, 0], X[y == 0, 1], 'r.', label='0')\n",
    "plt.plot(X[y == 1, 0], X[y == 1, 1], 'b.', label='1')\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "split = int(X.shape[0] * 0.8)\n",
    "\n",
    "X_train = X[:split]\n",
    "y_train = y[:split]\n",
    "\n",
    "X_test = X[split:]\n",
    "y_test = y[split:]\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "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": [
    "train_scores = []\n",
    "test_scores = []\n",
    "for k in range(1, 100):\n",
    "    clf = KNeighborsClassifier(n_neighbors=k)\n",
    "    clf.fit(X_train, y_train)\n",
    "    \n",
    "    train_scores.append(clf.score(X_train, y_train))\n",
    "    test_scores.append(clf.score(X_test, y_test))\n",
    "    \n",
    "plt.plot(np.arange(1, 100, 1), np.array(train_scores), 'b-', label='train')\n",
    "plt.plot(np.arange(1, 100, 1), np.array(test_scores), 'r-', label='test')\n",
    "\n",
    "plt.title('KNN')\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.5227272727272727\n",
      "0.4793388429752066\n"
     ]
    }
   ],
   "source": [
    "clf = SVC(C=0.5, kernel='rbf')\n",
    "clf.fit(X_train, y_train)\n",
    "\n",
    "print(clf.score(X_train, y_train))\n",
    "print(clf.score(X_test, y_test))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\shadm\\AppData\\Roaming\\Python\\Python37\\site-packages\\sklearn\\svm\\_base.py:249: ConvergenceWarning: Solver terminated early (max_iter=20).  Consider pre-processing your data with StandardScaler or MinMaxScaler.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "C:\\Users\\shadm\\AppData\\Roaming\\Python\\Python37\\site-packages\\sklearn\\svm\\_base.py:249: ConvergenceWarning: Solver terminated early (max_iter=20).  Consider pre-processing your data with StandardScaler or MinMaxScaler.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "C:\\Users\\shadm\\AppData\\Roaming\\Python\\Python37\\site-packages\\sklearn\\svm\\_base.py:249: ConvergenceWarning: Solver terminated early (max_iter=20).  Consider pre-processing your data with StandardScaler or MinMaxScaler.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "C:\\Users\\shadm\\AppData\\Roaming\\Python\\Python37\\site-packages\\sklearn\\svm\\_base.py:249: ConvergenceWarning: Solver terminated early (max_iter=20).  Consider pre-processing your data with StandardScaler or MinMaxScaler.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "C:\\Users\\shadm\\AppData\\Roaming\\Python\\Python37\\site-packages\\sklearn\\svm\\_base.py:249: ConvergenceWarning: Solver terminated early (max_iter=20).  Consider pre-processing your data with StandardScaler or MinMaxScaler.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "C:\\Users\\shadm\\AppData\\Roaming\\Python\\Python37\\site-packages\\sklearn\\svm\\_base.py:249: ConvergenceWarning: Solver terminated early (max_iter=20).  Consider pre-processing your data with StandardScaler or MinMaxScaler.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "C:\\Users\\shadm\\AppData\\Roaming\\Python\\Python37\\site-packages\\sklearn\\svm\\_base.py:249: ConvergenceWarning: Solver terminated early (max_iter=20).  Consider pre-processing your data with StandardScaler or MinMaxScaler.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "C:\\Users\\shadm\\AppData\\Roaming\\Python\\Python37\\site-packages\\sklearn\\svm\\_base.py:249: ConvergenceWarning: Solver terminated early (max_iter=20).  Consider pre-processing your data with StandardScaler or MinMaxScaler.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "C:\\Users\\shadm\\AppData\\Roaming\\Python\\Python37\\site-packages\\sklearn\\svm\\_base.py:249: ConvergenceWarning: Solver terminated early (max_iter=20).  Consider pre-processing your data with StandardScaler or MinMaxScaler.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "C:\\Users\\shadm\\AppData\\Roaming\\Python\\Python37\\site-packages\\sklearn\\svm\\_base.py:249: ConvergenceWarning: Solver terminated early (max_iter=20).  Consider pre-processing your data with StandardScaler or MinMaxScaler.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "C:\\Users\\shadm\\AppData\\Roaming\\Python\\Python37\\site-packages\\sklearn\\svm\\_base.py:249: ConvergenceWarning: Solver terminated early (max_iter=20).  Consider pre-processing your data with StandardScaler or MinMaxScaler.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "C:\\Users\\shadm\\AppData\\Roaming\\Python\\Python37\\site-packages\\sklearn\\svm\\_base.py:249: ConvergenceWarning: Solver terminated early (max_iter=20).  Consider pre-processing your data with StandardScaler or MinMaxScaler.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "C:\\Users\\shadm\\AppData\\Roaming\\Python\\Python37\\site-packages\\sklearn\\svm\\_base.py:249: ConvergenceWarning: Solver terminated early (max_iter=20).  Consider pre-processing your data with StandardScaler or MinMaxScaler.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "C:\\Users\\shadm\\AppData\\Roaming\\Python\\Python37\\site-packages\\sklearn\\svm\\_base.py:249: ConvergenceWarning: Solver terminated early (max_iter=20).  Consider pre-processing your data with StandardScaler or MinMaxScaler.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "C:\\Users\\shadm\\AppData\\Roaming\\Python\\Python37\\site-packages\\sklearn\\svm\\_base.py:249: ConvergenceWarning: Solver terminated early (max_iter=20).  Consider pre-processing your data with StandardScaler or MinMaxScaler.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "C:\\Users\\shadm\\AppData\\Roaming\\Python\\Python37\\site-packages\\sklearn\\svm\\_base.py:249: ConvergenceWarning: Solver terminated early (max_iter=20).  Consider pre-processing your data with StandardScaler or MinMaxScaler.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "C:\\Users\\shadm\\AppData\\Roaming\\Python\\Python37\\site-packages\\sklearn\\svm\\_base.py:249: ConvergenceWarning: Solver terminated early (max_iter=20).  Consider pre-processing your data with StandardScaler or MinMaxScaler.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "C:\\Users\\shadm\\AppData\\Roaming\\Python\\Python37\\site-packages\\sklearn\\svm\\_base.py:249: ConvergenceWarning: Solver terminated early (max_iter=20).  Consider pre-processing your data with StandardScaler or MinMaxScaler.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "C:\\Users\\shadm\\AppData\\Roaming\\Python\\Python37\\site-packages\\sklearn\\svm\\_base.py:249: ConvergenceWarning: Solver terminated early (max_iter=20).  Consider pre-processing your data with StandardScaler or MinMaxScaler.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "C:\\Users\\shadm\\AppData\\Roaming\\Python\\Python37\\site-packages\\sklearn\\svm\\_base.py:249: ConvergenceWarning: Solver terminated early (max_iter=20).  Consider pre-processing your data with StandardScaler or MinMaxScaler.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "C:\\Users\\shadm\\AppData\\Roaming\\Python\\Python37\\site-packages\\sklearn\\svm\\_base.py:249: ConvergenceWarning: Solver terminated early (max_iter=20).  Consider pre-processing your data with StandardScaler or MinMaxScaler.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "C:\\Users\\shadm\\AppData\\Roaming\\Python\\Python37\\site-packages\\sklearn\\svm\\_base.py:249: ConvergenceWarning: Solver terminated early (max_iter=20).  Consider pre-processing your data with StandardScaler or MinMaxScaler.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "C:\\Users\\shadm\\AppData\\Roaming\\Python\\Python37\\site-packages\\sklearn\\svm\\_base.py:249: ConvergenceWarning: Solver terminated early (max_iter=20).  Consider pre-processing your data with StandardScaler or MinMaxScaler.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "C:\\Users\\shadm\\AppData\\Roaming\\Python\\Python37\\site-packages\\sklearn\\svm\\_base.py:249: ConvergenceWarning: Solver terminated early (max_iter=20).  Consider pre-processing your data with StandardScaler or MinMaxScaler.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "C:\\Users\\shadm\\AppData\\Roaming\\Python\\Python37\\site-packages\\sklearn\\svm\\_base.py:249: ConvergenceWarning: Solver terminated early (max_iter=20).  Consider pre-processing your data with StandardScaler or MinMaxScaler.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "C:\\Users\\shadm\\AppData\\Roaming\\Python\\Python37\\site-packages\\sklearn\\svm\\_base.py:249: ConvergenceWarning: Solver terminated early (max_iter=20).  Consider pre-processing your data with StandardScaler or MinMaxScaler.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "C:\\Users\\shadm\\AppData\\Roaming\\Python\\Python37\\site-packages\\sklearn\\svm\\_base.py:249: ConvergenceWarning: Solver terminated early (max_iter=20).  Consider pre-processing your data with StandardScaler or MinMaxScaler.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "C:\\Users\\shadm\\AppData\\Roaming\\Python\\Python37\\site-packages\\sklearn\\svm\\_base.py:249: ConvergenceWarning: Solver terminated early (max_iter=20).  Consider pre-processing your data with StandardScaler or MinMaxScaler.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "C:\\Users\\shadm\\AppData\\Roaming\\Python\\Python37\\site-packages\\sklearn\\svm\\_base.py:249: ConvergenceWarning: Solver terminated early (max_iter=20).  Consider pre-processing your data with StandardScaler or MinMaxScaler.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "C:\\Users\\shadm\\AppData\\Roaming\\Python\\Python37\\site-packages\\sklearn\\svm\\_base.py:249: ConvergenceWarning: Solver terminated early (max_iter=20).  Consider pre-processing your data with StandardScaler or MinMaxScaler.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "C:\\Users\\shadm\\AppData\\Roaming\\Python\\Python37\\site-packages\\sklearn\\svm\\_base.py:249: ConvergenceWarning: Solver terminated early (max_iter=20).  Consider pre-processing your data with StandardScaler or MinMaxScaler.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "C:\\Users\\shadm\\AppData\\Roaming\\Python\\Python37\\site-packages\\sklearn\\svm\\_base.py:249: ConvergenceWarning: Solver terminated early (max_iter=20).  Consider pre-processing your data with StandardScaler or MinMaxScaler.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "C:\\Users\\shadm\\AppData\\Roaming\\Python\\Python37\\site-packages\\sklearn\\svm\\_base.py:249: ConvergenceWarning: Solver terminated early (max_iter=20).  Consider pre-processing your data with StandardScaler or MinMaxScaler.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "C:\\Users\\shadm\\AppData\\Roaming\\Python\\Python37\\site-packages\\sklearn\\svm\\_base.py:249: ConvergenceWarning: Solver terminated early (max_iter=20).  Consider pre-processing your data with StandardScaler or MinMaxScaler.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "C:\\Users\\shadm\\AppData\\Roaming\\Python\\Python37\\site-packages\\sklearn\\svm\\_base.py:249: ConvergenceWarning: Solver terminated early (max_iter=20).  Consider pre-processing your data with StandardScaler or MinMaxScaler.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "C:\\Users\\shadm\\AppData\\Roaming\\Python\\Python37\\site-packages\\sklearn\\svm\\_base.py:249: ConvergenceWarning: Solver terminated early (max_iter=20).  Consider pre-processing your data with StandardScaler or MinMaxScaler.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "C:\\Users\\shadm\\AppData\\Roaming\\Python\\Python37\\site-packages\\sklearn\\svm\\_base.py:249: ConvergenceWarning: Solver terminated early (max_iter=20).  Consider pre-processing your data with StandardScaler or MinMaxScaler.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "C:\\Users\\shadm\\AppData\\Roaming\\Python\\Python37\\site-packages\\sklearn\\svm\\_base.py:249: ConvergenceWarning: Solver terminated early (max_iter=20).  Consider pre-processing your data with StandardScaler or MinMaxScaler.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "C:\\Users\\shadm\\AppData\\Roaming\\Python\\Python37\\site-packages\\sklearn\\svm\\_base.py:249: ConvergenceWarning: Solver terminated early (max_iter=20).  Consider pre-processing your data with StandardScaler or MinMaxScaler.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "C:\\Users\\shadm\\AppData\\Roaming\\Python\\Python37\\site-packages\\sklearn\\svm\\_base.py:249: ConvergenceWarning: Solver terminated early (max_iter=20).  Consider pre-processing your data with StandardScaler or MinMaxScaler.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "C:\\Users\\shadm\\AppData\\Roaming\\Python\\Python37\\site-packages\\sklearn\\svm\\_base.py:249: ConvergenceWarning: Solver terminated early (max_iter=20).  Consider pre-processing your data with StandardScaler or MinMaxScaler.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "C:\\Users\\shadm\\AppData\\Roaming\\Python\\Python37\\site-packages\\sklearn\\svm\\_base.py:249: ConvergenceWarning: Solver terminated early (max_iter=20).  Consider pre-processing your data with StandardScaler or MinMaxScaler.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "C:\\Users\\shadm\\AppData\\Roaming\\Python\\Python37\\site-packages\\sklearn\\svm\\_base.py:249: ConvergenceWarning: Solver terminated early (max_iter=20).  Consider pre-processing your data with StandardScaler or MinMaxScaler.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "C:\\Users\\shadm\\AppData\\Roaming\\Python\\Python37\\site-packages\\sklearn\\svm\\_base.py:249: ConvergenceWarning: Solver terminated early (max_iter=20).  Consider pre-processing your data with StandardScaler or MinMaxScaler.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "C:\\Users\\shadm\\AppData\\Roaming\\Python\\Python37\\site-packages\\sklearn\\svm\\_base.py:249: ConvergenceWarning: Solver terminated early (max_iter=20).  Consider pre-processing your data with StandardScaler or MinMaxScaler.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "C:\\Users\\shadm\\AppData\\Roaming\\Python\\Python37\\site-packages\\sklearn\\svm\\_base.py:249: ConvergenceWarning: Solver terminated early (max_iter=20).  Consider pre-processing your data with StandardScaler or MinMaxScaler.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "C:\\Users\\shadm\\AppData\\Roaming\\Python\\Python37\\site-packages\\sklearn\\svm\\_base.py:249: ConvergenceWarning: Solver terminated early (max_iter=20).  Consider pre-processing your data with StandardScaler or MinMaxScaler.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "C:\\Users\\shadm\\AppData\\Roaming\\Python\\Python37\\site-packages\\sklearn\\svm\\_base.py:249: ConvergenceWarning: Solver terminated early (max_iter=20).  Consider pre-processing your data with StandardScaler or MinMaxScaler.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "C:\\Users\\shadm\\AppData\\Roaming\\Python\\Python37\\site-packages\\sklearn\\svm\\_base.py:249: ConvergenceWarning: Solver terminated early (max_iter=20).  Consider pre-processing your data with StandardScaler or MinMaxScaler.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "C:\\Users\\shadm\\AppData\\Roaming\\Python\\Python37\\site-packages\\sklearn\\svm\\_base.py:249: ConvergenceWarning: Solver terminated early (max_iter=20).  Consider pre-processing your data with StandardScaler or MinMaxScaler.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "C:\\Users\\shadm\\AppData\\Roaming\\Python\\Python37\\site-packages\\sklearn\\svm\\_base.py:249: ConvergenceWarning: Solver terminated early (max_iter=20).  Consider pre-processing your data with StandardScaler or MinMaxScaler.\n",
      "  % self.max_iter, ConvergenceWarning)\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\shadm\\AppData\\Roaming\\Python\\Python37\\site-packages\\sklearn\\svm\\_base.py:249: ConvergenceWarning: Solver terminated early (max_iter=20).  Consider pre-processing your data with StandardScaler or MinMaxScaler.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "C:\\Users\\shadm\\AppData\\Roaming\\Python\\Python37\\site-packages\\sklearn\\svm\\_base.py:249: ConvergenceWarning: Solver terminated early (max_iter=20).  Consider pre-processing your data with StandardScaler or MinMaxScaler.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "C:\\Users\\shadm\\AppData\\Roaming\\Python\\Python37\\site-packages\\sklearn\\svm\\_base.py:249: ConvergenceWarning: Solver terminated early (max_iter=20).  Consider pre-processing your data with StandardScaler or MinMaxScaler.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "C:\\Users\\shadm\\AppData\\Roaming\\Python\\Python37\\site-packages\\sklearn\\svm\\_base.py:249: ConvergenceWarning: Solver terminated early (max_iter=20).  Consider pre-processing your data with StandardScaler or MinMaxScaler.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "C:\\Users\\shadm\\AppData\\Roaming\\Python\\Python37\\site-packages\\sklearn\\svm\\_base.py:249: ConvergenceWarning: Solver terminated early (max_iter=20).  Consider pre-processing your data with StandardScaler or MinMaxScaler.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "C:\\Users\\shadm\\AppData\\Roaming\\Python\\Python37\\site-packages\\sklearn\\svm\\_base.py:249: ConvergenceWarning: Solver terminated early (max_iter=20).  Consider pre-processing your data with StandardScaler or MinMaxScaler.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "C:\\Users\\shadm\\AppData\\Roaming\\Python\\Python37\\site-packages\\sklearn\\svm\\_base.py:249: ConvergenceWarning: Solver terminated early (max_iter=20).  Consider pre-processing your data with StandardScaler or MinMaxScaler.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "C:\\Users\\shadm\\AppData\\Roaming\\Python\\Python37\\site-packages\\sklearn\\svm\\_base.py:249: ConvergenceWarning: Solver terminated early (max_iter=20).  Consider pre-processing your data with StandardScaler or MinMaxScaler.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "C:\\Users\\shadm\\AppData\\Roaming\\Python\\Python37\\site-packages\\sklearn\\svm\\_base.py:249: ConvergenceWarning: Solver terminated early (max_iter=20).  Consider pre-processing your data with StandardScaler or MinMaxScaler.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "C:\\Users\\shadm\\AppData\\Roaming\\Python\\Python37\\site-packages\\sklearn\\svm\\_base.py:249: ConvergenceWarning: Solver terminated early (max_iter=20).  Consider pre-processing your data with StandardScaler or MinMaxScaler.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "C:\\Users\\shadm\\AppData\\Roaming\\Python\\Python37\\site-packages\\sklearn\\svm\\_base.py:249: ConvergenceWarning: Solver terminated early (max_iter=20).  Consider pre-processing your data with StandardScaler or MinMaxScaler.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "C:\\Users\\shadm\\AppData\\Roaming\\Python\\Python37\\site-packages\\sklearn\\svm\\_base.py:249: ConvergenceWarning: Solver terminated early (max_iter=20).  Consider pre-processing your data with StandardScaler or MinMaxScaler.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "C:\\Users\\shadm\\AppData\\Roaming\\Python\\Python37\\site-packages\\sklearn\\svm\\_base.py:249: ConvergenceWarning: Solver terminated early (max_iter=20).  Consider pre-processing your data with StandardScaler or MinMaxScaler.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "C:\\Users\\shadm\\AppData\\Roaming\\Python\\Python37\\site-packages\\sklearn\\svm\\_base.py:249: ConvergenceWarning: Solver terminated early (max_iter=20).  Consider pre-processing your data with StandardScaler or MinMaxScaler.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "C:\\Users\\shadm\\AppData\\Roaming\\Python\\Python37\\site-packages\\sklearn\\svm\\_base.py:249: ConvergenceWarning: Solver terminated early (max_iter=20).  Consider pre-processing your data with StandardScaler or MinMaxScaler.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "C:\\Users\\shadm\\AppData\\Roaming\\Python\\Python37\\site-packages\\sklearn\\svm\\_base.py:249: ConvergenceWarning: Solver terminated early (max_iter=20).  Consider pre-processing your data with StandardScaler or MinMaxScaler.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "C:\\Users\\shadm\\AppData\\Roaming\\Python\\Python37\\site-packages\\sklearn\\svm\\_base.py:249: ConvergenceWarning: Solver terminated early (max_iter=20).  Consider pre-processing your data with StandardScaler or MinMaxScaler.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "C:\\Users\\shadm\\AppData\\Roaming\\Python\\Python37\\site-packages\\sklearn\\svm\\_base.py:249: ConvergenceWarning: Solver terminated early (max_iter=20).  Consider pre-processing your data with StandardScaler or MinMaxScaler.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "C:\\Users\\shadm\\AppData\\Roaming\\Python\\Python37\\site-packages\\sklearn\\svm\\_base.py:249: ConvergenceWarning: Solver terminated early (max_iter=20).  Consider pre-processing your data with StandardScaler or MinMaxScaler.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "C:\\Users\\shadm\\AppData\\Roaming\\Python\\Python37\\site-packages\\sklearn\\svm\\_base.py:249: ConvergenceWarning: Solver terminated early (max_iter=20).  Consider pre-processing your data with StandardScaler or MinMaxScaler.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "C:\\Users\\shadm\\AppData\\Roaming\\Python\\Python37\\site-packages\\sklearn\\svm\\_base.py:249: ConvergenceWarning: Solver terminated early (max_iter=20).  Consider pre-processing your data with StandardScaler or MinMaxScaler.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "C:\\Users\\shadm\\AppData\\Roaming\\Python\\Python37\\site-packages\\sklearn\\svm\\_base.py:249: ConvergenceWarning: Solver terminated early (max_iter=20).  Consider pre-processing your data with StandardScaler or MinMaxScaler.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "C:\\Users\\shadm\\AppData\\Roaming\\Python\\Python37\\site-packages\\sklearn\\svm\\_base.py:249: ConvergenceWarning: Solver terminated early (max_iter=20).  Consider pre-processing your data with StandardScaler or MinMaxScaler.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "C:\\Users\\shadm\\AppData\\Roaming\\Python\\Python37\\site-packages\\sklearn\\svm\\_base.py:249: ConvergenceWarning: Solver terminated early (max_iter=20).  Consider pre-processing your data with StandardScaler or MinMaxScaler.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "C:\\Users\\shadm\\AppData\\Roaming\\Python\\Python37\\site-packages\\sklearn\\svm\\_base.py:249: ConvergenceWarning: Solver terminated early (max_iter=20).  Consider pre-processing your data with StandardScaler or MinMaxScaler.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "C:\\Users\\shadm\\AppData\\Roaming\\Python\\Python37\\site-packages\\sklearn\\svm\\_base.py:249: ConvergenceWarning: Solver terminated early (max_iter=20).  Consider pre-processing your data with StandardScaler or MinMaxScaler.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "C:\\Users\\shadm\\AppData\\Roaming\\Python\\Python37\\site-packages\\sklearn\\svm\\_base.py:249: ConvergenceWarning: Solver terminated early (max_iter=20).  Consider pre-processing your data with StandardScaler or MinMaxScaler.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "C:\\Users\\shadm\\AppData\\Roaming\\Python\\Python37\\site-packages\\sklearn\\svm\\_base.py:249: ConvergenceWarning: Solver terminated early (max_iter=20).  Consider pre-processing your data with StandardScaler or MinMaxScaler.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "C:\\Users\\shadm\\AppData\\Roaming\\Python\\Python37\\site-packages\\sklearn\\svm\\_base.py:249: ConvergenceWarning: Solver terminated early (max_iter=20).  Consider pre-processing your data with StandardScaler or MinMaxScaler.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "C:\\Users\\shadm\\AppData\\Roaming\\Python\\Python37\\site-packages\\sklearn\\svm\\_base.py:249: ConvergenceWarning: Solver terminated early (max_iter=20).  Consider pre-processing your data with StandardScaler or MinMaxScaler.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "C:\\Users\\shadm\\AppData\\Roaming\\Python\\Python37\\site-packages\\sklearn\\svm\\_base.py:249: ConvergenceWarning: Solver terminated early (max_iter=20).  Consider pre-processing your data with StandardScaler or MinMaxScaler.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "C:\\Users\\shadm\\AppData\\Roaming\\Python\\Python37\\site-packages\\sklearn\\svm\\_base.py:249: ConvergenceWarning: Solver terminated early (max_iter=20).  Consider pre-processing your data with StandardScaler or MinMaxScaler.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "C:\\Users\\shadm\\AppData\\Roaming\\Python\\Python37\\site-packages\\sklearn\\svm\\_base.py:249: ConvergenceWarning: Solver terminated early (max_iter=20).  Consider pre-processing your data with StandardScaler or MinMaxScaler.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "C:\\Users\\shadm\\AppData\\Roaming\\Python\\Python37\\site-packages\\sklearn\\svm\\_base.py:249: ConvergenceWarning: Solver terminated early (max_iter=20).  Consider pre-processing your data with StandardScaler or MinMaxScaler.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "C:\\Users\\shadm\\AppData\\Roaming\\Python\\Python37\\site-packages\\sklearn\\svm\\_base.py:249: ConvergenceWarning: Solver terminated early (max_iter=20).  Consider pre-processing your data with StandardScaler or MinMaxScaler.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "C:\\Users\\shadm\\AppData\\Roaming\\Python\\Python37\\site-packages\\sklearn\\svm\\_base.py:249: ConvergenceWarning: Solver terminated early (max_iter=20).  Consider pre-processing your data with StandardScaler or MinMaxScaler.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "C:\\Users\\shadm\\AppData\\Roaming\\Python\\Python37\\site-packages\\sklearn\\svm\\_base.py:249: ConvergenceWarning: Solver terminated early (max_iter=20).  Consider pre-processing your data with StandardScaler or MinMaxScaler.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "C:\\Users\\shadm\\AppData\\Roaming\\Python\\Python37\\site-packages\\sklearn\\svm\\_base.py:249: ConvergenceWarning: Solver terminated early (max_iter=20).  Consider pre-processing your data with StandardScaler or MinMaxScaler.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "C:\\Users\\shadm\\AppData\\Roaming\\Python\\Python37\\site-packages\\sklearn\\svm\\_base.py:249: ConvergenceWarning: Solver terminated early (max_iter=20).  Consider pre-processing your data with StandardScaler or MinMaxScaler.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "C:\\Users\\shadm\\AppData\\Roaming\\Python\\Python37\\site-packages\\sklearn\\svm\\_base.py:249: ConvergenceWarning: Solver terminated early (max_iter=20).  Consider pre-processing your data with StandardScaler or MinMaxScaler.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "C:\\Users\\shadm\\AppData\\Roaming\\Python\\Python37\\site-packages\\sklearn\\svm\\_base.py:249: ConvergenceWarning: Solver terminated early (max_iter=20).  Consider pre-processing your data with StandardScaler or MinMaxScaler.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "C:\\Users\\shadm\\AppData\\Roaming\\Python\\Python37\\site-packages\\sklearn\\svm\\_base.py:249: ConvergenceWarning: Solver terminated early (max_iter=20).  Consider pre-processing your data with StandardScaler or MinMaxScaler.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "C:\\Users\\shadm\\AppData\\Roaming\\Python\\Python37\\site-packages\\sklearn\\svm\\_base.py:249: ConvergenceWarning: Solver terminated early (max_iter=20).  Consider pre-processing your data with StandardScaler or MinMaxScaler.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "C:\\Users\\shadm\\AppData\\Roaming\\Python\\Python37\\site-packages\\sklearn\\svm\\_base.py:249: ConvergenceWarning: Solver terminated early (max_iter=20).  Consider pre-processing your data with StandardScaler or MinMaxScaler.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "C:\\Users\\shadm\\AppData\\Roaming\\Python\\Python37\\site-packages\\sklearn\\svm\\_base.py:249: ConvergenceWarning: Solver terminated early (max_iter=20).  Consider pre-processing your data with StandardScaler or MinMaxScaler.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "C:\\Users\\shadm\\AppData\\Roaming\\Python\\Python37\\site-packages\\sklearn\\svm\\_base.py:249: ConvergenceWarning: Solver terminated early (max_iter=20).  Consider pre-processing your data with StandardScaler or MinMaxScaler.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "C:\\Users\\shadm\\AppData\\Roaming\\Python\\Python37\\site-packages\\sklearn\\svm\\_base.py:249: ConvergenceWarning: Solver terminated early (max_iter=20).  Consider pre-processing your data with StandardScaler or MinMaxScaler.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "C:\\Users\\shadm\\AppData\\Roaming\\Python\\Python37\\site-packages\\sklearn\\svm\\_base.py:249: ConvergenceWarning: Solver terminated early (max_iter=20).  Consider pre-processing your data with StandardScaler or MinMaxScaler.\n",
      "  % self.max_iter, ConvergenceWarning)\n"
     ]
    },
    {
     "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": [
    "train_scores = []\n",
    "test_scores = []\n",
    "for c in np.arange(0.1, 10, 0.1):\n",
    "    clf = SVC(C=c, kernel='rbf', max_iter=20)\n",
    "    clf.fit(X_train, y_train)\n",
    "    \n",
    "    train_scores.append(clf.score(X_train, y_train))\n",
    "    test_scores.append(clf.score(X_test, y_test))\n",
    "    \n",
    "plt.plot(np.arange(0.1, 10, 0.1), np.array(train_scores), 'b-', label='train')\n",
    "plt.plot(np.arange(0.1, 10, 0.1), np.array(test_scores), 'r-', label='test')\n",
    "plt.title('SVM')\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.4994834710743802\n",
      "0.5041322314049587\n"
     ]
    }
   ],
   "source": [
    "clf = LogisticRegression(penalty='l2')\n",
    "clf.fit(X_train, y_train)\n",
    "\n",
    "print(clf.score(X_train, y_train))\n",
    "print(clf.score(X_test, y_test))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "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": [
    "train_scores = []\n",
    "test_scores = []\n",
    "for c in np.arange(0.1, 10, 0.1):\n",
    "    clf = LogisticRegression(C=c, penalty='l2')\n",
    "    clf.fit(X_train, y_train)\n",
    "    \n",
    "    train_scores.append(clf.score(X_train, y_train))\n",
    "    test_scores.append(clf.score(X_test, y_test))\n",
    "\n",
    "plt.plot(np.arange(0.1, 10, 0.1), np.array(train_scores), 'b-', label='train')\n",
    "plt.plot(np.arange(0.1, 10, 0.1), np.array(test_scores), 'r-', label='test')\n",
    "plt.title('LogisticRegression')\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "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": "3.7.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}