Machine Learning Example.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "de53fae0",
   "metadata": {},
   "source": [
    "## 1.Linear Regression Equation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "8c8d7318",
   "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"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "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": [
    "import matplotlib.pyplot as plt\n",
    "\n",
    "x = [1, 2, 3]\n",
    "y = [1, 2, 3]\n",
    "\n",
    "plt.figure()\n",
    "plt.title('Data - X and Y')\n",
    "plt.plot(x,y,'*')\n",
    "plt.xticks([0,1,2,3])\n",
    "plt.yticks([0,1,2,3])\n",
    "plt.show()\n",
    "\n",
    "def linear_regression(theta0,theta1):\n",
    "    predicted_y = []\n",
    "    for i in x:\n",
    "        predicted_y.append((theta0+(theta1*i)))\n",
    "    plt.figure()\n",
    "    plt.title('Predictions')\n",
    "    plt.plot(x,predicted_y,'.')\n",
    "    plt.xticks([0,1,2,3])\n",
    "    plt.yticks([0,1,2,3])\n",
    "    plt.show()\n",
    "\n",
    "theta0 = 1.5\n",
    "theta1 = 0\n",
    "linear_regression(theta0,theta1)\n",
    "\n",
    "theta0a = 0\n",
    "theta1a = 1.5\n",
    "linear_regression(theta0a,theta1a)\n",
    "\n",
    "theta0b = 1\n",
    "theta1b = 0.5\n",
    "linear_regression(theta0b,theta1b)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "bf0133fd",
   "metadata": {},
   "source": [
    "## 2.Cost function"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "4f0bb929",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "cost when theta0=1.5 theta1=0 : 0.4583333333333333\n",
      "cost when theta0=0 theta1=1.5 : 0.5833333333333333\n",
      "cost when theta0=1 theta1=0.5 : 0.08333333333333333\n"
     ]
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "\n",
    "x = [1, 2, 3]\n",
    "y = [1, 2, 3]\n",
    "\n",
    "m = len(y) \n",
    "\n",
    "def cost_function(theta0,theta1):\n",
    "    predicted_y = [theta0+(theta1*1), theta0+(theta1*2), theta0+(theta1*3)]\n",
    "    sum=0\n",
    "    for i,j in zip(predicted_y,y):\n",
    "        sum = sum+((i-j)**2) \n",
    "    J = 1/(2*m)*sum \n",
    "    return (J)\n",
    "\n",
    "theta0 = [1.5,0,1]\n",
    "theta1 = [0,1.5,0.5]\n",
    "\n",
    "for i,j in zip(theta0,theta1):\n",
    "    print (\"cost when theta0=%r theta1=%r :\"%(i,j), cost_function(i,j))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f9c022ac",
   "metadata": {},
   "source": [
    "## 3.Contour Plot"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "475aba14",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "from mpl_toolkits.mplot3d import Axes3D\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "\n",
    "fig = plt.figure(figsize=(8, 5))\n",
    "ax1 = fig.add_subplot(111, projection='3d')\n",
    "\n",
    "values = 2\n",
    "r = np.linspace(-values, values, 100)\n",
    "\n",
    "theta0, theta1 = np.meshgrid(r, r)\n",
    "\n",
    "original_y = np.array([1, 2, 3])\n",
    "m = len(original_y)\n",
    "\n",
    "predicted_y = np.array([theta0 + (theta1 * 1), theta0 + (theta1 * 2), theta0 + (theta1 * 3)])\n",
    "\n",
    "sum_squared_error = np.sum((predicted_y - original_y[:, np.newaxis, np.newaxis]) ** 2)\n",
    "J = sum_squared_error / (2 * m)\n",
    "\n",
    "# Make sure J has the same shape as theta0 and theta1\n",
    "J = np.broadcast_to(J, theta0.shape)\n",
    "\n",
    "ax1.plot_surface(theta0, theta1, J, cmap='viridis')\n",
    "ax1.set_xlabel('Theta 0')\n",
    "ax1.set_ylabel('Theta 1')\n",
    "ax1.set_zlabel('Cost J')\n",
    "\n",
    "plt.title(\"Plot\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e9b9114b",
   "metadata": {},
   "source": [
    "## 4.Gradient Descent"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "d1f48bfc",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.5995100694321308\n"
     ]
    }
   ],
   "source": [
    "x = [1, 2, 3]\n",
    "y = [1, 2, 3]\n",
    "\n",
    "m = len(y)\n",
    "\n",
    "theta0 = 1\n",
    "theta1 = 1.5\n",
    "alpha = 0.01\n",
    "\n",
    "def cost_function(theta0, theta1):\n",
    "    predicted_y = [theta0 + (theta1 * xi) for xi in x]\n",
    "    sum_squared_error = sum((predicted_yi - yi) ** 2 for predicted_yi, yi in zip(predicted_y, y))\n",
    "    J = sum_squared_error / (2 * m)\n",
    "    return J\n",
    "\n",
    "def gradientDescent(x, y, theta1, alpha):\n",
    "    J_history = []\n",
    "    for _ in range(50):\n",
    "        for xi, yi in zip(x, y):\n",
    "            delta = (1 / m) * (xi ** 2 * theta1 - xi * yi)\n",
    "            theta1 = theta1 - alpha * delta\n",
    "            J_history.append(cost_function(theta0, theta1))\n",
    "    print(min(J_history))\n",
    "\n",
    "gradientDescent(x, y, theta1, alpha)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c40f0d43",
   "metadata": {},
   "source": [
    "## 5.Multiple Linear Equation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "b5f44e02",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[2320100 1330100 2950100]\n",
      "[-679900 -669900 -549900]\n",
      "[-1592457968  2089411216  1742299280]\n",
      "-2055714768\n",
      "-342619128.0\n"
     ]
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "\n",
    "x = np.array([[1, 800, 2, 15],[1, 1200, 3, 1],[1, 2400, 5, 5]])\n",
    "y = np.array([3000000,2000000,3500000])\n",
    "theta = np.array([100, 1000, 10000, 100000])\n",
    "\n",
    "predicted_y = x.dot(theta.transpose())\n",
    "print (predicted_y)\n",
    "\n",
    "m = y.size\n",
    "diff = predicted_y - y\n",
    "squares = np.square(diff)\n",
    "#sum_of_squares = 5424168464\n",
    "sum_of_squares = np.sum(squares)\n",
    "cost_fn = 1/(2*m)*sum_of_squares \n",
    "print (diff)\n",
    "print (squares)\n",
    "print (sum_of_squares)\n",
    "print (cost_fn)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c821718a",
   "metadata": {},
   "source": [
    "## 6.Polynomial Regression"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "133722b4",
   "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"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYIAAAEWCAYAAABrDZDcAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/YYfK9AAAACXBIWXMAAAsTAAALEwEAmpwYAAAugElEQVR4nO3dd5xU9fX/8debpUhTLNhAwcSSIEbFlWCJYgc06tdoomIvBFvsqNhjErAbNZa1o0Q0lsQgxJbYElEXRIUoSgQUUcEGIoYInN8f5+7Pcd1lF9i7d2bueT4e89iZe+/MnDu7O+d+usyMEEII+dUi6wBCCCFkKxJBCCHkXCSCEELIuUgEIYSQc5EIQggh5yIRhBBCzkUiyCFJ0yXtmnUchSSNlXR4I48tuvibiqT5kr6X0msPk3RKGq9dTCS1kfSmpDWzjqVURCIoYckX4lfJl8dHku6Q1CHruJaHmfU3s7tW9HUk3Snpf8ln8qmkJyT9oClibA5m1sHM3mnq15XUGTgMuDl53FfSkuRzmi9ppqT7JW3d1O+dBkmnSHpH0jxJsyRdLaklgJktBG4Hzso2ytIRiaD0/dTMOgC9gK2B8zKOpxhclnwmXYD3gdua+g1qvnRKyBHAGDP7qmDbrORz6gj0Ad4EnpO0S1O/eQqf11+BXma2MtAT2Bz4VcH+PwKHS2rTxO9bliIRlAkzex8Yi/9TIGlvSZMlfS7paUk/rP0cSWtLWiBp9YJtW0maI6mVpCMkPS/pCkmfSZomqX/BsetKeiS58p4q6diCfRdJ+pOkeyR9Iel1SRtLOkfSbEnvSdq94PinJR2T3P++pL9L+kTSx5JGSuq0HJ/JV8D9wBa1Yn4wOcdpkn5VsK+tpLuSc31D0hBJMwv2T5d0lqTXgC8ltZTUR9K/ks/5VUl9C44/Irlq/SJ5r4HJ9g0lPSNpbnJ+9xU8xyRtmNxfRdKIJNYZks6T1KLgtev93dShP/BMPZ+TmdlMM7sAuBW4tCCeHySlqk8lTZH084J9q0v6a3JV/rKk30h6vta5nCDpbeDtZNtekiYmn9e/JP2oMb+bOmL+j5l9XvNUYAmwYcH+mcBneIILDTGzuJXoDZgO7JrcXw+YDFwCbAx8CewGtAKGAFOB1nU8bwxwXMFrXg1cl9w/AvgaOBaoAI4DZgFK9j8D3ACshH/ZzgF2SfZdBPwX2ANoCYwApgHnJjEdC0wreN+ngWOS+xsmsbcBOgPPAtfUdd51fCZ3Ar9J7rcH7gZeTR63AMYDFwCtge8B7wB7JPuHJ+e0KtAVeA2YWet9JyafdVu8xPEJMCB57d2Sx52T954HbJI8dx1g0+T+vcnn0CL57LYveA8DNkzujwD+gl+xdwfeAo5uzO+mjs9lDrB1weO+hedWsH1n/Eu1fXJ7Dzgy+R32Aj4uOI9Rya0d0CM59vla5/IEsFryefUCZgM/TmI+PPlM2zT0u6nnnA5OPmNLzm/zWvsfAX6V9f9pKdwyD2C5gvb6v9nApEYce3Xyzzsx+Uf6POv4m/BzmA7MBz4HZuBfym2B84H7C45rgVeR9C14Xk0i+AXwz+R+BfAh0Dt5fAQwteB12iX/dGvjX4aLgY4F+4cBdyb3LwKeKNj30yTWiuRxx+S1OiWPnyZJBHWc577AK7XOe2mJ4L/JZ7IETz4/Svb9GHi31vHnAHck97/1xQMcw3cTwVEFj88C7q71eo8lX3Dtkxh+BrStdcwIoAroWkf8hifCCmAh0KNg3y+Bpxv63dTzuXwN/KDgcV/qTgQ/SF6nS/K38Vyt/TcDFybxfU2S6JJ9v+G7iWDngsc3ApfUer0pwI4N/W4a+D/YCL8AWrvW9pHABVn/n5bCrVSrhu4E+jXmQDM71cy2MLMtgOuAh1KMKwv7mlknM+tmZsebV4esiycGAMxsCX611qWO5/8F6CHvqbIbMNfMXirY/2HB6yxI7nZI3uNTM/ui4NgZtd7jo4L7XwEfm9nigsc1r/UtktaUNErS+5LmAfcAa9Rz/nW5wsw64VfRXwGbJNu7Aesm1RKfS/ocGAqslexfF/+cahTer2tbN+CAWq+3PbCOmX2Jf5EOBj6Q9Ki+abQegldnvJRU3x1Vx/usgV8ZzyjYVvvzre93U5fP8OTbkC74F/jnyfn9uNb5DcQvBDrjpYRl/bxOr/V66+Gfe0O/m3qZ2dt4afiGWrs6JucRGlCSicDMngU+LdyW1Cv/TdJ4Sc+p7p4iB+HF8nI3C//HAkCS8H+492sfaGb/xevRBwKH4lUpjX2P1SQVfrmsX9d7LIdh+JfRj8wbAw/BvziXiZm9C5wM/F5SW/xLaVqSOGtuHc1sQPKUD/AqoRrr1fWyBfffw0sEha/X3syGJ+//mJnthlcLvQnckmz/0MyONbN18av8G2raBQp8jF9xdyvYtiKf72t4lWFD/g+YkCSy94Bnap1fBzM7Dq+KWcSyf16/rfV67czsXhr+3TSkJfD9Wtt+CLzayOfnWkkmgnpUASeZ2VbAGdS6OpDUDdgA+HsGsTW3+4E9Je0iqRVwOl7N8K96jh+BVzXsjV99N8jM3kteb5iklZJGv6Px4viK6khS5SWpC3Dm8r6QmT2BJ61BwEvAvKTBt62kCkk99U2XyfuBcyStmrzviQ28/D3ATyXtkbzWSvJumV0lrSVvsG+Pf/bz8ao0JB0gqeYL9DP8y3Jx4QsnJaf7gd9K6pj8/Z5GI38/dRiDV8F8h1wXSRfi1WFDk12jgY0lHSrvPNBK0taSfpjE9xBwkaR2yYXXYQ3EcAswWNKPk/dsL2nP5GKiod9N7ZiPUTJOQFIPvBrpqYL9XfC2iXGN+3jyrSwSgbzv/LbAnyRNxOsx16l12IHAAwVVE2XLzKbgV9HX4VeWP8W7mf6vnuP/idenTzCz6cvwVgfh1S+zgIeBC5Mv3hV1Md6wOBd4lBWvzrscr45piX8WW+BtBx/jvWRWSY77NTAz2fck8AD+JV6nJBnug39xzsGvas/E/69a4Al4Fl563RE4Pnnq1sCLkubjDZonm9m0Ot7iJLzR/x3gebxL5O3LeO41RgADkpJRjXWTGOYDLwOb4e1Ijyfn9wWwO/6/MwuviroUb9wFT5SrJNvvxkvbS/u8qvHG7evxBDgVvwCpSXxL+93Uth3wuqQv8SQ3hm8SGHhD8l3mYwpCA2p6f5QcSd2B0WbWU9LKwBQzq/3lX3j8K8AJZlbfVXGuSfo78EczuzXrWIqFpOOAA82szivpUiPpd8BsM7smpde/FG+wbdQI8bTIxw68CuxgZrOzjKVUlEWJwMzmAdMkHQD/v6i7ec1+SZvgXQJfyCjEopYUv3sB9zV0bDmTtI6k7SS1SP5mTsdLOmXBzIY2ZRKQjzH4UfL/1huvGsz88zKzhWb2g0gCjVeSiUDSvfiX+ibyofFH442dR0t6Fe9BsE/BUw4CRlmpFn9SJOkuvBrklFo9gPKoNV6t+AXelvQXvtsTJXyjI15t9yXennEl/pmFElOyVUMhhBCaRkmWCEIIITSdUps4izXWWMO6d++edRghhFBSxo8f/7GZda5rX8klgu7du1NdXZ11GCGEUFIkzahvX1QNhRBCzkUiCCGEnItEEEIIOReJIIQQci4SQQgh5FwkglA2Ro6E7t2hRQv/ObIp5kENIQdKrvtoCHUZORIGDYIFyfIsM2b4Y4CBA7OLK4RSECWCUBbOPfebJFBjwQLfHkJYutQSQbJIx0uSXk2W47u4jmP6SporaWJyuyCteEJ5e/fdZdseQvhGmlVDC/GFq+cnq2Q9L2msmdVeMeg5M9srxThCDqy/vlcH1bU9hLB0qZUIzM1PHrZKbjHVaTPJW8Ppb38L7dp9e1u7dr49lJe8/W03h1TbCJJ1RycCs4EnzOzFOg7bJqk+Gitp03peZ5CkaknVc+bMSTPkslDTcDpjBph903Bazv8wAwdCVRV06waS/6yqiobicpPHv+3m0CzrEUjqhK9cdJKZTSrYvjKwJKk+GgD83sw2WtprVVZWWkw6t3Tdu9ddTdKtG0yf3tzRhNB04m97+Ukab2aVde1rll5DZvY58DTQr9b2eTXVR2Y2BmglaY3miKmcRcNpKFfxt52ONHsNdU5KAkhqC+wKvFnrmLUlKbnfO4nnk7Riyov6Gkij4TSUuvjbTkeaJYJ1gH9Ieg14GW8jGC1psKTByTH7A5OSdYavBQ6MdYVXXDSchnIVf9vpSK37qJm9BmxZx/abCu5fD1yfVgx5VdNAeu65XmRef33/R4mG01Dq4m87HSW3eH00FocQwrLLvLE4hBBC8YpEEEIIOReJIIQQci4SQQgh5FwkghBCyLlIBCGEkHORCEIIIeciEYQQQs5FIgghhJyLRBBCCDkXiSCEEHIuEkEIIeRcJIIQQsi5SAQhhJBzkQhCeZgxwyem/8EPfGHbv/4164hCKBmRCELpmjcPbr8d+vb1L//zzoO114aVV4a994bDD4fPPss6yhCKXiSCUFoWLYKxY+Ggg2CtteDoo+GDD+CSS2DaNHj6aaiuhvPPh5EjoWdPePTRrKMOoailuXj9SpJekvSqpMmSLq7jGEm6VtJUSa9J6pVWPKGEmcHEiXD66dC1KwwYAI8/DkcdBePGwZtvemmge3c/vnVr+PWv4cUXYbXVYK+94Mgj4fPPMzyJEIpXamsWAwuBnc1svqRWwPOSxprZuIJj+gMbJbcfAzcmP0OAWbPgj3+EESPg9dehVSv/Uj/sME8GrVsv/flbbeWlg0sugeHD4Ykn4JZboH//5ok/hBKRWonA3PzkYavkVnuB5H2AEcmx44BOktZJK6ZQAr780qt09tgD1lsPzjwT2reHG27wKqCHHoJ99204CdRo0wZ+8xsvOXTq5Ank6KNh7tw0zyKEkpJqG4GkCkkTgdnAE2b2Yq1DugDvFTyemWwLebJkCfz973DEEd7Ye8gh8NZbcO65MGUKvPACHHccrL768r9HZSWMHw/nnAN33ultB4891lRnEEJJSzURmNliM9sC6Ar0ltSz1iGq62m1N0gaJKlaUvWcOXNSiDRk4o03/Iu5e3fYZRd4+GH4xS/gmWfgP//xev6NN26692vTBn73O08sHTtCv35wzDFROgi51yy9hszsc+BpoF+tXTOB9QoedwVm1fH8KjOrNLPKzp07pxVmaA5z5sC118LWW0OPHnD55bDZZjBqFHz4Idx6K+ywA7RI8U+zd2+YMAHOOgvuuMPf//HH03u/EIpcmr2GOkvqlNxvC+wKvFnrsEeAw5LeQ32AuWb2QVoxhYz897/wwAPet3/ddeHkk2HxYrj6anj/fe/e+YtfQNu2zRfTSit5A/K//uVtEHvsAYMG+diEEHImzV5D6wB3SarAE879ZjZa0mAAM7sJGAMMAKYCC4AjU4wnNCcz+Oc/4e674b77vPpl3XXhtNPg0EO9jr4Y/PjH8MorcMEFcOWV3m5w222w665ZRxZCs5HZd6rki1plZaVVV1dnHUaoz3/+41/+d98N77wD7drBz37mX/477wwVFVlHWL9x47zBesoU+OUvvdqqY8esowqhSUgab2aVde2LkcVhxX32Gdx8M2y3HWy4oTfybrAB3HUXfPSRjwPYbbfiTgIAffp46eD006GqytsOnnoq66hCSF0kgrB8/vc/eOQR2H9/7/I5eLCP3B0+HN59F5580gd+deiQdaTLpm1buOIKeP5572W0665w/PEwf37Dzw2hRKXZRhDKjZmP1L37brj3Xvj4Y+jc2fv4H3oo9OoFqqtHcAnadluf1uK887xRe+xYn+Bup52yjiyEJhclgtCwd9+FYcO8u2fv3l5tsvPOMHq09/q55hqfzqFckkCNtm29AfnZZ6FlSz/nE06I0kEoO5EIQt2++MJH4O68sw/4GjrUr/6rqry//333wZ57+vw/5W777eHVV+GUU+DGG+FHP/JBbyGUiUgE4RuLFnn3yYEDfYrnI4+E996Diy7y3kDPPgvHHutz9uRNu3ZeRfTMMz7YrW9fOOkknxsphBIXbQQBXnvNe/aMHOlX+506+aIuhx3mPWnKrcpnRfzkJ146GDrUR0iPGeOjk3fYIevIQlhuUSLIqw8+8PrvLbaAzTeH3//eB1c9+KAngxtvhG22iSRQl/bt/fOqqR7acUcfLR2lg1CiIhHkyYIFPr9///6+wMsZZ3gXyeuu88Tw5z/Dfvv5ttCwHXbw0tRJJ3npYPPN4bnnso4qhGUWiaDcLVniyzcedZT39x84EP79bzj7bJ/988UX4cQTYY01so60NLVv70ngH//wz3rHHeHUUz3phlAioo2gnL3wAhx4oHf/7NABDjjA6/3Tnt0zj/r29dLB2Wd7d9pHH/W2g+22yzqyEBoU3wblysyvTBcv9uqgjz7yAVF9+0YSSEuHDnD99b7Iztdfe8Py6afDV19lHVkISxXfCOXqmWe82ue88+Cgg7z7Y2geO+3kpYPBg+Gqq7xB/l//yjqqEOoViaBcDRvmYwGOOCLrSPKpY0dfZ/nJJ2HhQh+UduaZUToIRSkSQTkaP95X3Dr1VF+AJWRnl13g9dd90ZsrroAtt/TprkMoIpEIytHw4bDKKj4ZXMhex45w002enBcs8AbkIUN85bYQikAkgnIzZYoPCjvhBFh55ayjCYV22w0mTYKjj/ZFb7bc0ttxQshYJIJyc9llPiDs5JOzjiTUZeWVfeK+xx7zkcjbbutdTqN0EDIUiaCczJzpawUccwysuWbW0YSl2X13bzs48ki49FKfxvvll7OOKuRUaolA0nqS/iHpDUmTJX3nElVSX0lzJU1MbhekFU8uXHmljx8444ysIwmNscoqcOutvujN3Lk+wd/Qod7LKIRmlGaJYBFwupn9EOgDnCCpRx3HPWdmWyS3X6cYT3n75BOvcjj4YOjWLetowrLo18/bDg4/3Lv9brWVrwQXQjNJLRGY2QdmNiG5/wXwBtAlrffLveuu8x4pQ4ZkHUlYHp06+cjvRx+Fzz7z0sG550bpIDSLZmkjkNQd2BKoq4vENpJelTRW0qbNEU/ZmT/fJz7bZx/YND7CkjZgAEye7GtA/+53UFkJEyZkHVUoc6knAkkdgAeBU8xsXq3dE4BuZrY5cB3w53peY5CkaknVc+bMSTXeklRV5VeR55yTdSShKXTq5BPWjR7tVX69e8P558P//pd1ZKFMyczSe3GpFTAaeMzMrmrE8dOBSjP7uL5jKisrrTrqT7+xcCF873uwySY+2VkoL5995msljxgBm23m60j36pV1VKEESRpvZpV17Uuz15CA24A36ksCktZOjkNS7ySeT9KKqSzdfTfMmhWlgXK16qpw113wyCMwZ46vInfhhVE6CE0qtRKBpO2B54DXgSXJ5qHA+gBmdpOkE4Hj8B5GXwGnmdlSp2mMEkGBxYvhhz/0QUovvxzLSpa7Tz/1gYL33OOrod15p89sGkIjLK1EkNrCNGb2PLDUbyYzux64Pq0Yyt6DD8Lbb8MDD0QSyIPVVvMS4AEHwC9/CVtv7dOMDx0KrVplHV0oYTGyuFSZ+eRyG28M++6bdTShOe29t487+PnP4aKLvDH59dezjiqUsEgEperxx+GVV+Css6CiIutoQnNbfXUYORIeftjbiLbbLha/CcstEkGpGjYMunaFQw7JOpKQpX339XEGa68Ne+wBzz+fdUShBEUiKEUvvOBLUZ5+OrRunXU0IWtdusDTT/vPfv3g2WezjiiUmEgEpWjYMK8aOPbYrCMJxWLddeEf/4D114f+/T0xhNBIkQhKzaRJ8Ne/wq9+Be3bZx1NKCbrrOPJYIMNfKqKp57KOqJQIiIRlJrhw6FDBzjxxKwjCcVorbV8hPmGG8Jee3mnghAaEImglEybBqNGeR/y1VbLOppQrNZc05PBJpt4V9O//S3riEKRi0RQSi6/HFq0gFNPzTqSUOzWWMOrhnr08Flpx4zJOqJQxCIRlIqPPvL56g8/3HuHhNCQ1VeHJ5/0yer23dfblkKoQySCUnHNNfD117HwTFg2q60GTzzhcxL97Gfwl79kHVEoQpEISsHcuXDDDbD//rDRRllHE0rNqqt6MujVy/+GHnoo64hCkYlEUApuuAHmzYOzz846klCqVlnFexBtvbXPUfSnP2UdUSgikQiK3VdfebVQv36w5ZZZRxNK2corw2OPwTbbwEEHeQ+0EIhEUPxuvx1mz46FZ0LT6NgRxo71SeoGDvSJ60LuRSIoZl9/7V1Gt9kGfvKTrKMJ5aJDB+9OuuOOcNhhvsZByLVIBMVs1CiYMcNLA7HwTGhK7dvD6NGw007eJfnOO7OOKGQoEkGxWrIELr0UevaEPffMOppQjtq187EFu+4KRx0Ft92WdUQhI6ktVRlW0OjRMHmyr0/bIvJ1SEnbtj62YL/94JhjfB3sQYOyjio0s9S+YSStJ+kfkt6QNFnSyXUcI0nXSpoq6TVJvdKKp6SY+VTTG2wAv/hF1tGEcte2ra90NmCAz2N1441ZRxSaWZolgkXA6WY2QVJHYLykJ8zs3wXH9Ac2Sm4/Bm5MfubbM8/AuHE+fqBlFNpCM1hpJR9odsABcPzxXjKIGW5zI7USgZl9YGYTkvtfAG8AtSfJ2QcYYW4c0EnSOmnFVDKGDfPphI88MutIQp60aQMPPOCT1J10ko9fCbnQLJXPkroDWwIv1trVBXiv4PFMvpsskDRIUrWk6jlz5qQWZ1EYP95HgJ56ql+lhdCcWrf2Ucf77ed/g1demXVEoRmknggkdQAeBE4xs3m1d9fxFPvOBrMqM6s0s8rOnTunEWbxGD7cR4AOHpx1JCGvWrXyrssHHABnnAGXXZZ1RCFlqVZAS2qFJ4GRZlbXTFczgfUKHncFZqUZU1GbMgUefNDnFFpllayjCXnWqhX88Y9QUQFnneVtBjG6vWyllggkCbgNeMPMrqrnsEeAEyWNwhuJ55rZB2nFVPQuv9zraU85JetIQvCOCnff7d2Xhw6FRYvg/POzjiqkoFGJQNLGeI+etcysp6QfAXub2W+W8rTtgEOB1yVNTLYNBdYHMLObgDHAAGAqsADIb+vozJkwYoT34V5zzayjCcG1bOl/lxUVcMEFPtDxwguzjio0scaWCG4BzgRuBjCz1yT9Eag3EZjZ89TdBlB4jAEnNDKG8nbVVf5PdsYZWUcSwrdVVMAdd/jPiy7yaqKLL45pT8pIYxNBOzN7Sd/+xS9KIZ58+uQTqKqCgw+G7t2zjiaE76qo8CkoKirgkku8mui3v41kUCYamwg+lvR9kh49kvYH8luX39Suuw6+/NIb5UIoVi1a+AVLRYWPdVm82Hu5RTIoeY1NBCcAVcAPJL0PTAMOSS2qPJk/H669FvbeGzbdNOtoQli6Fi18CoqKCu9Wunixd3KIZFDSGpUIzOwdYFdJ7YEWyUjh0BSqquCzz6JrXigdLVrAH/7gyeDKK72a6OqrIxmUsMb2GvodcJmZfZ48XhWfR+i8FGMrfwsX+j9S377Qp0/W0YTQeJKXZCsq4Pe/95LBtddGMihRja0a6m9mQ2semNlnkgYAkQhWxD33wKxZ3iMjhFIjeUmgZUu/oFm8GK6/PqZNL0GNTQQVktqY2UIASW2BNumFlQOLF/vCM716wW67ZR1NCMtH8jaCwjaDG2+MZFBiGpsI7gGeknQH3nPoKOCu1KLKg4cegrff9gm+ojgdSpnkvYcKexNVVUUyKCGNbSy+TNLrwC74ILFLzOyxVCMrZzULz2y8Mfzf/2UdTQgrTvJxBS1b+jiDxYvh1ls9OYSi1+i5hsxsLDA2xVjy4/HH4ZVXvhmgE0I5kODXv/72COSaEcmhqC01EUh63sy2l/QF354eWvgMESunGl25GjYMunSBQ2IoRihDF17oX/7nn+/J4K67YqW9IrfU346ZbZ/87Ng84eTACy/4UpRXXeWLgIRQjs47z5PB0KE+h9bdd0cyKGIN/mYktQBeM7OezRBP+Rs2DFZbDY49NutIQkjXOed8ez2DkSN9nYNQdBpMBGa2RNKrktY3s3ebI6iyNWkS/PWvXn/aoUPW0YSQviFDvCRw+umeDO69N0rCRaixZbV1gMmSXgK+rNloZnunElW5uvRSaN/eFwYPIS9OO827kp56Kvz853D//ZEMikxjE8HFqUaRB9Om+dXQySd71VAIeXLKKV4yOOkk2H9/Hz/TJsakFouGeg2tBAwGNgReB24zs1iHYHlccYVfFZ12WtaRhJCNE0/0NoPjj4f99vP1uVdaKeuoAtDQ0L+7gEo8CfQHrkw9onL00Udw++1w2GHebTSEvDruOLj5ZhgzxgdT/ve/WUcUaLhqqIeZbQYg6TbgpfRDKkPXXOMzjQ4ZknUkIWRv0CAvGRx7rK/D8Ze/QNu2WUeVaw2VCL6uubOsVUKSbpc0W9Kkevb3lTRX0sTkdsGyvH7JmDsXbrjB60U33jjraEIoDkcf7aXkJ5+EvfaCBQuyjijXGioRbC5pXnJfQNvkcWNGFt8JXA+MWMoxz5nZXo0NtiTdcAPMmxcLz4RQ2xFHeMngiCNgzz1h9GjvVReaXUMji5d7khAze1ZS9+V9fln46iuvFtpjD9hyy6yjCaH4HHqoJ4NDD4UBA+DRR2OMTQaynid2m2Sw2lhJ9S7YK2mQpGpJ1XPmzGnO+FbMHXfA7NlRGghhaQ4+2Ecd//Of0K8ffBEr4Ta3LBPBBKCbmW0OXAf8ub4DzazKzCrNrLJz587NFd+KWbTIF+zYZhvYYYesowmhuB14oI+zGTfOS9Dz5jX8nNBkMksEZjbPzOYn98cArSStkVU8TW7UKJg+3UsDsfBMCA074AAfdfzyy7D77t7RIjSLzBKBpLUl/4aU1DuJ5ZOs4mlSS5b4ik2bbuqNYCGExtlvPx91PGGCL+H6+edZR5QLqc0LK+leoC+whqSZwIVAKwAzuwnYHzhO0iLgK+BAM7N6Xq60jB4Nkyf71LuxXF8Iy2bffX3U8f77w667+kJOMS1LqlRq372VlZVWXV2ddRj1M4Ntt4UPP/Q1iWMO9hCWT83o4x49fLzB6qtnHVFJkzTezCrr2heXq03tmWe8wevMMyMJhLAiBgzwUcdvvAG77AIff5x1RGUrEkFTGzYM1lwTjjwy60hCKH39+sEjj8CUKbDzzt4dOzS5SARNacIEr8889dSYOyWEprL77t7uNnUq7LSTT+IYmlQkgqY0fDisvLLPsBhCaDq77OKjjqdP92Tw4YdZR1RWIhE0lbfeggce8LnWV1kl62hCKD877QRjx8K770LfvvD++1lHVDYiETSVyy7zFZdOOSXrSEIoXzvsAH/7myeBzTbzSR0XxVpZKyoSQVOYORNGjICjjoK11so6mhDK2/bbw4svwhZbwAknwFZbwbPPZh1VSYtE0BSuuspHE595ZtaRhJAPPXrAU0/5KOTPP4cdd4SDDvKLsrDMIhGsqE8+gaoq/yPs3j3raELID8lHH7/xBlxwATz8MGyyiXfhXrgw6+hKSiSCFXXddfDll3D22VlHEkI+tWsHF1/sCWH33WHoUJ/na/TorCMrGZEIVsT8+Z4I9t7b//BCCNnZYAMvFTz2GLRqBT/9qU/6+NZbWUdW9CIRrIhbboFPP42FZ0IoJrvvDq++CldeCc89Bz17wllnxYI3SxGJYHktXOh/aDvuCH36ZB1NCKFQ69Zw2mleGhg40Lt3b7KJr4RWYhNtNodIBMvrnnu8L3OUBkIoXmuv7UvGvvACdOkChxwCP/kJvPJK1pEVlUgEy2PxYrj0Ul+Qfvfds44mhNCQPn187MGtt3opYautfCqYT8pjLawVFYlgeTz0kK81EMtQhlA6WrSAo4/2RHDSSd7Gt9FGMTqZSATLzsz7KW+8sS+rF0IoLZ06we9/DxMnxujkRCSCZfX4416/OGQIVFRkHU0IYXn17BmjkxORCJbV8OHfNDqFEEpbjE4GUkwEkm6XNFvSpHr2S9K1kqZKek1Sr7RiaTLjxsHTT3u3tDZtso4mhNBUcj46Oc0SwZ1Av6Xs7w9slNwGATemGEvTGDYMVlsNBg3KOpIQQhpyOjo5tURgZs8Cny7lkH2AEebGAZ0krZNWPCts0iRfO/Wkk6BDh6yjCSGkKWejk7NsI+gCvFfweGay7TskDZJULal6zpw5zRLcd1x6KbRv74kghFD+cjQ6OctEUFcH/Do/XTOrMrNKM6vs3LlzymHVYdo0uPderxJaffXmf/8QQnZyMDo5y0QwE1iv4HFXYFZGsSzdFVf4YJTTTss6khBCVsp4dHKWieAR4LCk91AfYK6ZfZBhPHX76CO4/XY47DDo2jXraEIIWSocnfyrX5XN6OQ0u4/eC7wAbCJppqSjJQ2WNDg5ZAzwDjAVuAU4Pq1YVsg113h/4iFDso4khFAsOnXy74bC0cmVlSU7OllWYo0elZWVVl1d3TxvNncurL++9yD405+a5z1DCKXFDB58EE4/Hd59Fw48EC6/vOhqECSNN7PKuvbFyOKlufFGmDcvppoOIdSvDEYnRyKoz1dfwdVXe2mgV/EPeg4hZKyERydHIqjPHXfA7NlRGgghLJsSHJ0ciaAuixZ5HV+fPj4jYQghLKvdd4fXXvv26OSzzy7K0cmRCOoyahRMnx4Lz4QQVkyrVt8enXzppUU5OjkSQW1LlvhU05tuCnvtlXU0IYRyUOSjkyMR1DZ6NEye7BNMtYiPJ4TQhIp0dHJ80xWqWYayWzfvCxxCCE2tCEcnRyIo9OyzvvjMmWd63V4IIaSliEYnRyIoNGwYrLkmHHVU1pGEEPKicO3kzz7LZO3kSAQ1Jkzwfr+nnAJt22YdTQghTzIenRyJoMbw4bDyynB8cc59F0LIgYxGJ0ciAG+0eeABTwKrrJJ1NCGEvGvm0cmRCMCXoGvd2quFQgihWNQenVxVlcrbtEzlVUvJzJkwYgQccwystVbW0YQQwrfVjE4++GCvOkpBJIKrr/bRxGeemXUkIYRQv7XXTu2l81019MkncPPNPnhsgw2yjiaEEDKR70Rw/fXw5Zc+I2AIIeRUfhPB/Plw7bXeGt+zZ9bRhBBCZlJNBJL6SZoiaaqk71x2S+oraa6kicntgjTj+ZZbboFPP42FZ0IIuZdaY7GkCuAPwG7ATOBlSY+Y2b9rHfqcmTXvfM8LF3p3rB13hG22ada3DiGEYpNmr6HewFQzewdA0ihgH6B2Imh+99wD77/vU8GGEELOpVk11AV4r+DxzGRbbdtIelXSWEmb1vVCkgZJqpZUPWfOnBWLavFiXyVoyy1hjz1W7LVCCKEMpFkiqGuNx9prs00AupnZfEkDgD8DG33nSWZVQBVAZWXliq3v9vDD8PbbcN99sQxlCCGQbolgJrBeweOuwKzCA8xsnpnNT+6PAVpJWiO1iGoWntloI/jZz1J7mxBCKCVpJoKXgY0kbSCpNXAg8EjhAZLWlvyyXFLvJJ701mx74gmfbnrIEKioSO1tQgihlKRWNWRmiySdCDwGVAC3m9lkSYOT/TcB+wPHSVoEfAUcaGYrVvWzNMOGwbrrwqGHpvYWIYRQalKdayip7hlTa9tNBfevB65PM4b/b9w4ePpp7zbapk2zvGUIIZSC/IwsXrLEp3QdNCjrSEIIoajkZ/bRbbf1RR5CCCF8S35KBCGEEOoUiSCEEHIuEkEIIeRcJIIQQsi5SAQhhJBzkQhCCCHnIhGEEELORSIIIYSci0QQQgg5F4kghBByLhJBCCHkXCSCEELIuUgEIYSQc5EIQggh53KRCEaOhO7doUUL/zlyZNYRhRBC8Sj7RDBypK9FM2OGr10/Y4Y/jmQQQigVaV/Mln0iOPdcWLDg29sWLPDtIYRQ7JrjYjbVRCCpn6QpkqZKOruO/ZJ0bbL/NUm9mjqGd99dtu0hlJKo9ix/zXExm1oikFQB/AHoD/QADpLUo9Zh/YGNktsg4MamjmP99ZdtewilIqo986E5LmbTLBH0Bqaa2Ttm9j9gFLBPrWP2AUaYGwd0krROUwbx299Cu3bf3taunW8PoZRFtWc+NMfFbJqJoAvwXsHjmcm2ZT0GSYMkVUuqnjNnzjIFMXAgVFVBt24g+c+qKt8eQimLas98aI6L2TQTgerYZstxDGZWZWaVZlbZuXPnZQ5k4ECYPh2WLPGfkQRCOYhqz3xojovZNBPBTGC9gsddgVnLcUwIoQ5R7ZkfaV/MppkIXgY2krSBpNbAgcAjtY55BDgs6T3UB5hrZh+kGFMIZSOqPUNTaZnWC5vZIkknAo8BFcDtZjZZ0uBk/03AGGAAMBVYAByZVjwhlKOBA+OLP6y41BIBgJmNwb/sC7fdVHDfgBPSjCGEEMLSlf3I4hBCCEsXiSCEEHIuEkEIIeRcJIIQQsg5eXtt6ZA0B5ixnE9fA/i4CcMpBXHO+RDnnA8rcs7dzKzOEbkllwhWhKRqM6vMOo7mFOecD3HO+ZDWOUfVUAgh5FwkghBCyLm8JYKqrAPIQJxzPsQ550Mq55yrNoIQQgjflbcSQQghhFoiEYQQQs6VVSKQdLuk2ZImFWxbTdITkt5Ofq5asO8cSVMlTZG0RzZRLz9J60n6h6Q3JE2WdHKyvZzPeSVJL0l6NTnni5PtZXvONSRVSHpF0ujkcVmfs6Tpkl6XNFFSdbKt3M+5k6QHJL2Z/F9v0yznbGZlcwN2AHoBkwq2XQacndw/G7g0ud8DeBVoA2wA/AeoyPoclvF81wF6Jfc7Am8l51XO5yygQ3K/FfAi0Kecz7ng3E8D/giMTh6X9TkD04E1am0r93O+Czgmud8a6NQc51xWJQIzexb4tNbmffAPl+TnvgXbR5nZQjObhq+J0Ls54mwqZvaBmU1I7n8BvIGv+VzO52xmNj952Cq5GWV8zgCSugJ7ArcWbC7rc65H2Z6zpJXxi9nbAMzsf2b2Oc1wzmWVCOqxliWrniU/10y2dwHeKzhuZrKtJEnqDmyJXyGX9TknVSQTgdnAE2ZW9ucMXAMMAZYUbCv3czbgcUnjJQ1KtpXzOX8PmAPckVQB3iqpPc1wznlIBPVRHdtKsi+tpA7Ag8ApZjZvaYfWsa3kztnMFpvZFvga170l9VzK4SV/zpL2Amab2fjGPqWObSV1zontzKwX0B84QdIOSzm2HM65JV61faOZbQl8iVcF1afJzjkPieAjSesAJD9nJ9tnAusVHNcVmNXMsa0wSa3wJDDSzB5KNpf1OddIis1PA/0o73PeDthb0nRgFLCzpHso73PGzGYlP2cDD+PVHuV8zjOBmUkJF+ABPDGkfs55SASPAIcn9w8H/lKw/UBJbSRtAGwEvJRBfMtNkvD6xDfM7KqCXeV8zp0ldUrutwV2Bd6kjM/ZzM4xs65m1h04EPi7mR1CGZ+zpPaSOtbcB3YHJlHG52xmHwLvSdok2bQL8G+a45yzbiVv4hb3e4EPgK/xbHk0sDrwFPB28nO1guPPxVvapwD9s45/Oc53e7wo+BowMbkNKPNz/hHwSnLOk4ALku1le861zr8v3/QaKttzxuvLX01uk4Fzy/2ck3PYAqhO/r7/DKzaHOccU0yEEELO5aFqKIQQwlJEIgghhJyLRBBCCDkXiSCEEHIuEkEIIeRcJIIQUpD07X4ymTnzF5KGZh1TCPVpmXUAIZSpLYFW5lNhIGk+8LtMIwqhHpEIQmikZITr/fhQ/grgEmAuPiHcx8AEfCDUUcA9QOdkcry3gbbJ/clmNrC5Yw9haSIRhNB4/YBZZrYngKRV8NHNO+NTAN8HPjeOpGOAM8xsr+TY+TWlgxCKTbQRhNB4rwO7SrpU0k/wxUCmmdnb5kP078k2vBCWTySCEBrJzN4CtsITwjBgb0pvquMQviOqhkJoJEnrAp+a2T1J4+9gYANJ3zez/wAHLeXpX0tqZWZfN0uwISyDSAQhNN5mwOWSluAz3B4HrAE8Kulj4HmgvkVyqoDXJE2IxuJQbGL20RCaiKS+FDQQh1Aqoo0ghBByLkoEIYSQc1EiCCGEnItEEEIIOReJIIQQci4SQQgh5FwkghBCyLn/B0wwztlPBWpeAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "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": [
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt \n",
    "from sklearn.linear_model import LinearRegression \n",
    "from sklearn.preprocessing import PolynomialFeatures \n",
    "\n",
    "X = pd.DataFrame([100, 200, 300, 400, 500, 600], columns=['sqft'])\n",
    "y = pd.DataFrame([543543, 34543543, 35435345, 34534, 34534534, 345345], columns=['Price'])\n",
    "\n",
    "lin = LinearRegression()   \n",
    "lin.fit(X, y) \n",
    "plt.scatter(X, y, color='blue')   \n",
    "plt.plot(X['sqft'], lin.predict(X), color='red')  # <-- Pass X['sqft'] instead of X\n",
    "plt.title('Linear Regression') \n",
    "plt.xlabel('sqft') \n",
    "plt.ylabel('Price')   \n",
    "plt.show() \n",
    "\n",
    "for i in [2, 3, 4, 5]:\n",
    "    poly = PolynomialFeatures(degree=i) \n",
    "    X_poly = poly.fit_transform(X)   \n",
    "    poly.fit(X_poly, y) \n",
    "    lin2 = LinearRegression() \n",
    "    lin2.fit(X_poly, y) \n",
    "    plt.scatter(X['sqft'], y, color='blue')  # <-- Pass X['sqft'] instead of X\n",
    "    plt.plot(X['sqft'], lin2.predict(poly.fit_transform(X)), color='red')  # <-- Pass X['sqft'] instead of X\n",
    "    plt.title(f'Polynomial Regression (Degree {i})') \n",
    "    plt.xlabel('sqft') \n",
    "    plt.ylabel('Price')   \n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5db17eb1",
   "metadata": {},
   "source": [
    "## 7.Loan Prediction"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "ab2c2cab",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[0]\n",
      " [1]\n",
      " [1]\n",
      " [0]\n",
      " [1]\n",
      " [0]\n",
      " [1]\n",
      " [1]\n",
      " [1]\n",
      " [0]\n",
      " [1]\n",
      " [1]\n",
      " [1]\n",
      " [1]\n",
      " [1]\n",
      " [1]\n",
      " [1]\n",
      " [1]\n",
      " [1]\n",
      " [1]\n",
      " [1]\n",
      " [1]\n",
      " [1]\n",
      " [0]\n",
      " [0]\n",
      " [1]\n",
      " [1]\n",
      " [1]\n",
      " [1]\n",
      " [0]\n",
      " [1]\n",
      " [0]\n",
      " [1]\n",
      " [0]\n",
      " [1]\n",
      " [1]\n",
      " [1]\n",
      " [1]\n",
      " [1]\n",
      " [1]\n",
      " [0]\n",
      " [0]\n",
      " [0]\n",
      " [0]\n",
      " [1]\n",
      " [1]\n",
      " [0]\n",
      " [0]\n",
      " [1]\n",
      " [1]\n",
      " [1]\n",
      " [1]\n",
      " [1]\n",
      " [1]\n",
      " [1]\n",
      " [0]\n",
      " [0]\n",
      " [0]\n",
      " [0]\n",
      " [0]\n",
      " [1]\n",
      " [1]\n",
      " [1]\n",
      " [0]\n",
      " [0]\n",
      " [1]\n",
      " [1]\n",
      " [1]\n",
      " [0]\n",
      " [1]\n",
      " [1]\n",
      " [1]\n",
      " [1]\n",
      " [0]\n",
      " [1]\n",
      " [1]\n",
      " [0]\n",
      " [1]\n",
      " [0]\n",
      " [0]\n",
      " [0]\n",
      " [1]\n",
      " [1]\n",
      " [0]\n",
      " [1]\n",
      " [1]\n",
      " [1]\n",
      " [1]\n",
      " [1]\n",
      " [1]\n",
      " [0]\n",
      " [0]\n",
      " [1]\n",
      " [1]\n",
      " [0]\n",
      " [1]\n",
      " [1]\n",
      " [1]\n",
      " [1]\n",
      " [1]\n",
      " [0]\n",
      " [1]\n",
      " [1]\n",
      " [1]\n",
      " [1]\n",
      " [1]\n",
      " [1]\n",
      " [1]\n",
      " [0]\n",
      " [1]\n",
      " [1]\n",
      " [1]\n",
      " [1]\n",
      " [0]\n",
      " [0]\n",
      " [1]\n",
      " [1]\n",
      " [1]\n",
      " [0]\n",
      " [0]\n",
      " [1]\n",
      " [0]\n",
      " [1]\n",
      " [0]\n",
      " [0]\n",
      " [0]\n",
      " [1]\n",
      " [0]\n",
      " [0]\n",
      " [1]\n",
      " [1]\n",
      " [1]\n",
      " [1]\n",
      " [1]\n",
      " [1]\n",
      " [1]\n",
      " [1]\n",
      " [1]\n",
      " [1]\n",
      " [1]\n",
      " [1]\n",
      " [1]\n",
      " [1]\n",
      " [1]\n",
      " [0]\n",
      " [1]\n",
      " [1]\n",
      " [0]\n",
      " [1]\n",
      " [0]\n",
      " [0]\n",
      " [0]\n",
      " [1]\n",
      " [1]]\n"
     ]
    }
   ],
   "source": [
    "import os \n",
    "import pandas as pd\n",
    "from sklearn.model_selection import train_test_split, GridSearchCV\n",
    "\n",
    "data = pd.read_csv('loan_train_csv.csv')\n",
    "l1 = ['Gender','Married','Dependents','Education','Self_Employed','ApplicantIncome','CoapplicantIncome',\n",
    "      'LoanAmount','Loan_Amount_Term','Credit_History','Property_Area']\n",
    "l2 = ['Loan_Status']\n",
    "l1_train, l1_test, l2_train, l2_test = train_test_split(data[l1], data[l2],test_size=0.25, random_state=1)\n",
    "\n",
    "l2_train = l2_train.replace({'Y':1, 'N':0}).values\n",
    "l2_test = l2_test.replace({'Y':1, 'N':0}).values     \n",
    "\n",
    "print (l2_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "eb96394a",
   "metadata": {},
   "source": [
    "## 8.Logistic Regression"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "66ef914f",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\Admin\\anaconda3\\lib\\site-packages\\sklearn\\model_selection\\_split.py:680: UserWarning: The least populated class in y has only 1 members, which is less than n_splits=5.\n",
      "  warnings.warn(\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Mean Cross-Validation Score: 0.9663528659791739\n"
     ]
    }
   ],
   "source": [
    "import pandas as pd\n",
    "from sklearn.feature_extraction.text import TfidfVectorizer\n",
    "from sklearn.linear_model import LogisticRegression  # Correct import statement\n",
    "from sklearn.model_selection import train_test_split, cross_val_score\n",
    "\n",
    "df = pd.read_csv('spam.csv', delimiter=',', header=None, encoding='latin1')\n",
    "\n",
    "X = df[1]  # Assuming the email text is stored in the second column\n",
    "y = df[0]  # Assuming the labels are stored in the first column\n",
    "\n",
    "# Perform text vectorization using TF-IDF\n",
    "vectorizer = TfidfVectorizer()\n",
    "X = vectorizer.fit_transform(X)\n",
    "\n",
    "# Split the dataset into training and testing sets\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)\n",
    "\n",
    "# Create and train the logistic regression model\n",
    "logreg = LogisticRegression()\n",
    "logreg.fit(X_train, y_train)\n",
    "\n",
    "# Evaluate the model using cross-validation\n",
    "cv_scores = cross_val_score(logreg, X_train, y_train, cv=5)\n",
    "\n",
    "# Print the mean cross-validation score\n",
    "print(\"Mean Cross-Validation Score:\", cv_scores.mean())"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "fae49117",
   "metadata": {},
   "source": [
    "## 9.Confusion Matrix Precision Recall"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "34fa113d",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy: 0.7\n",
      "[[4 2]\n",
      " [1 3]]\n",
      "(array([0.66666667, 0.75      ]), array([0.8, 0.6]), array([0.72727273, 0.66666667]), array([5, 5], dtype=int64))\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAQkAAAD+CAYAAAApiPBxAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/YYfK9AAAACXBIWXMAAAsTAAALEwEAmpwYAAAYRElEQVR4nO3de7hddX3n8feHEAl3xESFQADbiAO0XJpy8ylGakdImWLnsSNI8RnGFqFadMSZWqePIradS1vHUoQQhEFEsVKVQYhcxspAWrkkMQQCXjJSS0psSLgGIiTnfOaPtU7ZHM9ee52TvbP2Pvvzep71ZK+1fvu3fns/Z3/zu63fkm0iItrZqekCRER/S5CIiEoJEhFRKUEiIiolSEREpQSJiKiUINEwSbtK+oakZyTdsB35nCXp9m6WrSmSfkXS95suRxSUeRL1SHo38GHgTcBzwCrgT2wv2858zwZ+HzjR9rbtLWe/k2Rgvu21TZcl6klNogZJHwY+A/wp8DpgHnAZcHoXsj8I+MEwBIg6JO3cdBliHNvZKjZgb2Az8FsVaXahCCKPl9tngF3KcwuBdcCFwAZgPXBOee6TwEvA1vIa7wUuAq5ryftgwMDO5f6/B35EUZt5FDir5fiylvedCNwPPFP+e2LLuTuBTwF/V+ZzOzC7zWcbK/9/bin/O4BFwA+AJ4GPtaQ/FvgO8HSZ9lLgVeW5u8rP8nz5ed/Vkv8fAD8BvjB2rHzPz5XXOKbc3x/YCCxs+m9jWLbGC9DvG3AKsG3sR9omzcXAPcBrgTnA3wOfKs8tLN9/MTCz/HG9ALy6PD8+KLQNEsDuwLPAoeW5/YDDy9f/EiSAfYGngLPL951Z7r+mPH8n8P+ANwK7lvv/rc1nGyv/x8vy/y7wBPAlYE/gcOCnwBvK9L8EHF9e92DgEeBDLfkZ+PkJ8v/vFMF219YgUab53TKf3YDbgD9v+u9imLY0Nzp7DbDR1c2Bs4CLbW+w/QRFDeHslvNby/NbbS+l+F/00CmWZxQ4QtKuttfbXjNBml8Hfmj7C7a32b4e+B7wb1rS/C/bP7C9BfgKcFTFNbdS9L9sBb4MzAb+0vZz5fXXAL8IYHuF7XvK6/4DcAXwlhqf6RO2XyzL8wq2rwR+CNxLERj/S4f8oosSJDrbBMzu0FbeH/hxy/6Py2P/kse4IPMCsMdkC2L7eYoq+nnAekm3SHpTjfKMlWluy/5PJlGeTbZHytdjP+J/bjm/Zez9kt4o6WZJP5H0LEU/zuyKvAGesP3TDmmuBI4A/sr2ix3SRhclSHT2HYrq9Dsq0jxO0QE5Zl55bCqep6hWj3l960nbt9n+NYr/Ub9H8ePpVJ6xMv3TFMs0GZdTlGu+7b2AjwHq8J7KITZJe1D081wFXCRp3y6Uc9qSNEPSdyXdPME5SbpE0lpJqyUd0ym/BIkObD9D0R7/rKR3SNpN0kxJp0r6H2Wy64E/kjRH0uwy/XVTvOQq4CRJ8yTtDfzh2AlJr5P0G5J2B16kaLaMTJDHUuCNkt4taWdJ7wIOA37mj6YH9qToN9lc1nLOH3f+n4E3TDLPvwRW2P4d4BZg8XaXcnr7IEUfzkROBeaX27kUQb1SgkQNtj9NMUfijyg67R4DPgDcWCb5Y2A5sBp4EFhZHpvKte4A/rrMawWv/GHvRDFK8jhFj/9bgN+bII9NwGll2k0UIxOn2d44lTJN0keAd1OMmlxJ8VlaXQR8XtLTkv5dp8wknU7ReXxeeejDwDGSzupaiacRSQdQ9El9rk2S04FrXbgH2EfSfpV52plMFTFdSPob4L9S1Og+Yvu0cedvphjJWlbufwv4A9vL2+WZiSsRDXr7W3fzxidHa6VdufrFNRT9Y2OW2F4ytiPpNGCD7RWSFrbJZqL+ocqaQoJERIM2PjnC3986t3NCYNb+j/7U9oKKJG8GfkPSImAWsJek62z/dkuadcCBLfsH0KGTPX0SEQ0yMIprbR3zsv/Q9gG2DwbOAP52XIAAuAl4TznKcTzwjO31VfmmJhHRsFHqNTemStJ5ALYXU4x8LQLWUsyPOafT+xMkIhpkzEgPBg9s30kx3X4sOIwdN/D+yeSV5sYUSTpF0vfLSSkfbbo8042kqyVtkPRQ02XptW41N3olQWIKJM0APksxMeUw4ExJhzVbqmnnGor5EdOagRFca2tKmhtTcyyw1vaPACR9mWKSysONlmoasX2XpIObLkevGdjq3vZJbK/UJKZmLsWsyzHreOXNUxG1jdbcmpKaxNRMekJKxETccFOijgSJqZn0hJSICRlG+jtGpLkxRfcD8yUdIulVFBNXbmq4TDGAislU/d3cSJCYgnIBmQ9QLKX2CPCVNitExRRJup5iLY9DJa2T9N6my9QbYqTm1pQ0N6aoXIZuadPlmK5sn9l0GXYEA6N93txIkIhokIGX+rxCnyAR0bBRN9eUqCNBIqJBxYzLBImIaMOIkTQ3IqJKvzc3+juE9TlJ5zZdhuluun/HY82Nfh4CTZDYPtP6D7hPTPPvWIx4p1pbU9LciGiQga3MaLoYlfoqSMzed4YPPnBm08Wobd7cnVlw5Kw+nwrzSj9YvVvnRH1kFruxl/YdqO/4pzzPS36xVvvAVqO1hDr6KkgcfOBM7rvtwM4JY8revv9RTRdh2rvX35pU+tEMgUZEO0XHZWoSEdFWmhsRUaG4VTxBIiLaMOIlZ3QjIiqMprkREe2k4zIiKhkx0uf3biRIRDQsHZcR0ZZN3w+B9nfpIqY9MVpz65iTNEvSfZIekLRG0icnSLNQ0jOSVpXbxzvlm5pERIMMvOSu/QxfBE62vVnSTGCZpG/avmdcurttn1Y30wSJiAYZdW3RGdsGNpe7M8ttu2+OS3MjomEj7FRrq0PSDEmrgA3AHbbvnSDZCWWT5JuSDu+UZ2oSEQ0qnrtR+//q2ZKWt+wvsb3kFfnZI8BRkvYBvi7pCNsPtSRZCRxUNkkWATcC86sumiAR0ahJLU230faCOgltPy3pTuAU4KGW48+2vF4q6TJJs21vbJdXmhsRDRqrSdTZOpE0p6xBIGlX4G3A98aleb0kla+PpYgBm6ryTU0iomFdXOR2P+DzkmZQ/Pi/YvtmSecB2F4MvBM4X9I2YAtwRtnh2VaCRESDbLF1tDs/Q9urgaMnOL645fWlwKWTyTdBIqJBxXoSuXcjItrKylQRUaHouExNIiIqZD2JiGirm9OyeyVBIqJhWU8iItqyYetogkREtFE0NxIkIqJCF2dc9kSCRESDMgQaER2kuRERHWRadkS0VayWnSAREW0YsW00zwKNiAppbkREWxndiIiOMroREe05N3hFRIWsTBURHaUmERFtGdjW53eB9rR0kk6R9H1JayV9tJfXihhEY4vO1Nma0rMgUa79/1ngVOAw4ExJh/XqehGDahTV2prSy5rEscBa2z+y/RLwZeD0Hl4vYvCYvq9J9LJPYi7wWMv+OuC4Hl4vYuAM+2SqiT75zzxOTNK5wLkA8+amHzWGzzAHiXXAgS37BwCPj09UPjp9CcCCI2dVPpMwYroxYmSIRzfuB+ZLOkTSq4AzgJt6eL2IgTS0HZe2twEfAG4DHqF4wvGaXl0vYhC5ix2XkmZJuk/SA5LWSPrkBGkk6ZJyWsJqScd0yrennQC2lwJLe3mNiEHn7vVJvAicbHuzpJnAMknftH1PS5pTgfnldhxwOR0GFNJTGNGo7g1v2jawudydWW7j+/lOB64t094jaR9J+9le3y7f/u4xiRgCtmptdUiaIWkVsAG4w/a945JMNDVhblWeCRIRDRqbJ1GzT2K2pOUt27k/k589YvsoitHEYyUdMS5JrakJrdLciGjS5BbC3Wh7Qa1s7acl3QmcAjzUcqrW1IRWqUlENMh0r7khaY6kfcrXuwJvA743LtlNwHvKUY7jgWeq+iMgNYmIhnX1voz9gM+XN1fuRDHt4GZJ5wHYXkwx2rgIWAu8AJzTKdMEiYiGuUvzjG2vBo6e4PjiltcG3j+ZfBMkIhrWxXkSPZEgEdEgO0EiIjoY5rtAI6KG0dEEiYhow9SfTdmUBImIhvX7IioJEhFNSsdlRHTU51WJBImIhqUmERGVujXjslfaBglJf0VFRcj2BT0pUcQQscF9vhBuVU1i+Q4rRcQQG9iahO3Pt+5L2t32870vUsSQ6fMg0bGeI+kESQ9TrHiNpCMlXdbzkkUMhXprSTTZuVmnMfQZ4O3AJgDbDwAn9bBMEcPFNbeG1BrdsP2Y9IpINtKb4kQMmWkymeoxSScCLp/EdQFl0yMiumDQ+ySA8yhWspkL/BNwFJNc2SYiKlj1toZ0rEnY3gictQPKEjGcBr0mIekNkr4h6QlJGyT9b0lv2BGFi5j2TN/XJOo0N74EfIViJd79gRuA63tZqIhhUixh13lrSp0gIdtfsL2t3K6j7ytIEQNkUIdAJe1bvvy2pI8CX6Yo6ruAW3ZA2SKGwwAPga6gCApjn+B9LecMfKpXhYoYGgaNNl2IalX3bhyyIwsSMZya7ZSso9aMy/LJxIcBs8aO2b62V4WKGCp93sPXMUhI+gSwkCJILAVOBZYBCRIR3dDnQaLO6MY7gV8FfmL7HOBIYJeelipimPT56EadILHF9iiwTdJewAYgk6kiuqGLk6kkHSjp25IekbRG0gcnSLNQ0jOSVpXbxzvlW6dPYrmkfYArKUY8NgP31XhfRNSg7tUStgEX2l4paU9ghaQ7bD88Lt3dtk+rm2mdezd+r3y5WNKtwF7lI84johu6FCRsrwfWl6+fk/QIxY2Z44PEpFRNpjqm6pztldtz4Yk8/Pgcfumi87udbbSYufSJposw7Y1csGxS6btYk3g5T+lg4Gjg3glOnyDpAeBx4CO211TlVVWT+IuKcwZO7lDOiKij/jyJ2ZJaF6heYnvJ+ESS9gC+CnzI9rPjTq8EDrK9WdIi4EZgftVFqyZTvbVuySNiiiY3crHR9oKqBJJmUgSIL9r+2s9criVo2F4q6TJJs8slISbU3wv+RwyDLg2Bqlhj8irgEdufbpPm9WU6JB1LEQM2VeWbJ3hFNKyLfRJvBs4GHpS0qjz2MWAegO3FFPOezpe0DdgCnGFX34ieIBHRtO6Nbizj5Rsy26W5FLh0MvnWWZlKkn57bNKFpHllNSUitpPKu0DrbE2p0ydxGXACcGa5/xzw2Z6VKGLY9PnydXWaG8fZPkbSdwFsP1UurR8R3dDnN3jVCRJbJc2g/CiS5gB9vkxGxODoxWSqbqrT3LgE+DrwWkl/QnGb+J/2tFQRw6TP7wKtc+/GFyWtoLhdXMA7bOcJXhHd4P6vSdRZdGYe8ALwjdZjtv+xlwWLGBqDHiQoVsYeWxB3FnAI8H3g8B6WK2JoDOxCuGNs/0Lrfnl36PvaJI+IaWbSMy7LBS1+uReFiRhKg97ckPThlt2dgGOALEoQ0Q3ToeMS2LPl9TaKPoqv9qY4EUNokINEOYlqD9v/aQeVJ2L4DGqQkLSz7W1Vy9hFxPYRg93cuI+i/2GVpJuAG4Dnx05OtOpNREzSID8LtMW+FCvXnMzL8yUMJEhEdMMA1yReW45sPMQrny4Off+xIgZIn/+aqoLEDGAPJl7pps8/VsTgGOQ+ifW2L95hJYkYVgMcJJpbCidiWDR8G3gdVUHiV3dYKSKG2MCObth+ckcWJGJYDXKfRETsCAkSEdHWgPdJRESPif4fIUiQiGhaahIRUSUdlxFRbVCHQCNiBxiAlanqPJwnInqpSw/nkXSgpG9LekTSGkkfnCCNJF0iaa2k1XXWi0lNIqJhXaxJbAMuLBer3hNYIekO2w+3pDkVmF9uxwGXl/+2lZpERNO6VJOwvd72yvL1c8AjwNxxyU4HrnXhHmAfSftV5ZuaRETDJlGTmC1pecv+EttLJsxTOhg4Grh33Km5wGMt++vKY+vbXTRBIqJJk5txudH2gk6JJO1BsaL9h2w/O/50m1K0lSAR0SDR3btAJc2kCBBfbLMO7TrgwJb9A4DHq/JMn0RE07o3uiHgKuAR259uk+wm4D3lKMfxwDO22zY1oIc1CUlXA6cBG2wf0avrRAw6uWvDG28GzgYelLSqPPYxYB6A7cXAUmARsBZ4ATinU6a9bG5cA1wKXNvDa0QMti7eBWp7GR3uF7Nt4P2TybdnQcL2XWUPa0RU6PcZl+m4jGhagkQ1SecC5wLM3OPVDZcmYsfr95pE46MbtpfYXmB7wc6zdm+6OBE7VvmYvzpbUxqvSUQMvWGtSUi6HvgOcKikdZLe26trRQyqsaeK19ma0svRjTN7lXfEtNK9eRI9keZGRMP6veMyQSKiSVlSPyI6GdjH/EXEjpEgERHtmXRcRkS1dFxGRLUEiYhoZ2wyVT9LkIhokp0+iYioltGNiKiU5kZEtGdgtL+jRIJERNP6O0YkSEQ0Lc2NiKiW0Y2IqJKaRES0JYPScRkRlTJPIiKqdPExfz2RIBHRpKxMFRHV+v/ejcYfzhMx7Lq5pL6kqyVtkPRQm/MLJT0jaVW5fbxTnqlJRDStuzWJa4BLgWsr0txt+7S6GSZIRDTJoJHuBQnbd0k6uGsZkuZGRPNcc+ueEyQ9IOmbkg7vlDg1iYiGTWIIdLak5S37S2wvmeTlVgIH2d4saRFwIzC/6g0JEhFNqx8kNtpesH2X8rMtr5dKukzSbNsb270nzY2IJplixmWdrQskvV6SytfHUsSATVXvSU0iokHCXZ1xKel6YCFF02Qd8AlgJoDtxcA7gfMlbQO2AGfY1QVIkIhoWheDhO0zO5y/lGKItLYEiYgmGejiEGgvJEhENCw3eEVEtQSJiGiv/2/wSpCIaFKeKh4RHWVlqoioko7LiGjPwEh/VyUSJCIalY7LSdmycd3GVVdc+OOmyzEJs4G2N8b0pSuaLsCkDd53DAdNKnWCRH225zRdhsmQtHx778qLakPxHSdIRERbeap4RFQzOB2X09lkVwWKyZve3/EAjG5k0ZntUGfpMEkj5dLlD0m6QdJuU72epGskvbN8/TlJh1WkXSjpxClc4x8kza57fFyazZO81kWSPlKVZgrLsw0eu97WkASJ3tti+yjbRwAvAee1npQ0YyqZ2v4d2w9XJFkITDpIRAMSJKLF3cDPl//Lf1vSl4AHJc2Q9GeS7pe0WtL7AFS4VNLDkm4BXjuWkaQ7JS0oX58iaWW5AvK3yiXVzwP+Y1mL+RVJcyR9tbzG/ZLeXL73NZJul/RdSVcA6vQhJN0oaYWkNZLOHXfuL8qyfEvSnPLYz0m6tXzP3ZLe1JVvc1qoGSAaDBLpk9hBJO0MnArcWh46FjjC9qPlD+0Z278saRfg7yTdDhwNHAr8AvA64GHg6nH5zgGuBE4q89rX9pOSFgObbf95me5LwP+0vUzSPOA24F9RLG+2zPbFkn4deMWPvo3/UF5jV+B+SV+1vQnYHVhp+8LyyVCfAD5A0a9wnu0fSjoOuAw4eQpf4/RjYLS/+yQSJHpvV0mrytd3A1dRNAPus/1oefxfA7841t8A7E2xzPlJwPW2R4DHJf3tBPkfD9w1lpftJ9uU423AYeUaqAB7SdqzvMa/Ld97i6SnanymCyT9Zvn6wLKsmyhuVfrr8vh1wNck7VF+3htarr1LjWsMj8yTGHpbbB/VeqD8sTzfegj4fdu3jUu3iM6PZVGNNFA0LU+wvWWCstT+K5W0kCLgnGD7BUl3ArPaJHd53afHfwfRos+DRPok+sNtFCsYzwSQ9EZJuwN3AWeUfRb7AW+d4L3fAd4i6ZDyvfuWx58D9mxJdztF1Z8y3VHly7uAs8pjpwKv7lDWvYGnygDxJoqazJidKFZjBng3RTPmWeBRSb9VXkOSjuxwjeFh45GRWltTEiT6w+co+htWqnga9BUUtbyvAz8EHgQuB/7v+DfafoKiH+Frkh7g5er+N4DfHOu4BC4AFpQdow/z8ijLJ4GTJK2kaPb8Y4ey3grsLGk18CngnpZzzwOHS1pB0edwcXn8LOC9ZfnWAKfX+E6Gx6jrbQ1RhyX3I6KH9t55jk/Ys17MvO3pq1Y0cR9L+iQimmRndCMiOujz2nyCRETDnJpERLSXlakiooqBBoc368gQaESDDHjUtbY6JF0taUM5lD7ReUm6RNLacjj8mE55JkhENMnlojN1tnquAU6pOH8qxTT6+RTzay7vlGGCRETDulmTsH0X0O7+HSgmsl3rwj3APuVs3rbSJxHRtB27fN1c4LGW/XXlsfXt3pAgEdGg53jqtv/jv6lc8avFLEnLW/aXTGHlronWC6mspiRIRDTIdlX/QS+so7i9f8wBwONVb0ifRMRwuQl4TznKcTzFYkdtmxqQmkTEtCLpeor1TWdLWkexOthMANuLgaXAImAt8AJwTsc8cxdoRFRJcyMiKiVIRESlBImIqJQgERGVEiQiolKCRERUSpCIiEoJEhFR6f8DX0TOvHIgnbQAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 288x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "from sklearn.metrics import precision_recall_fscore_support\n",
    "from sklearn.metrics import accuracy_score\n",
    "from sklearn.metrics import confusion_matrix\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "y_true = [0, 0, 0, 0, 0, 1, 1, 1, 1, 1]\n",
    "y_pred = [0, 1, 0, 0, 0, 0, 0, 1, 1, 1]\n",
    "\n",
    "print ('Accuracy:', accuracy_score(y_true, y_pred))\n",
    "print (confusion_matrix(y_pred,y_true))\n",
    "print (precision_recall_fscore_support(y_true, y_pred))\n",
    "\n",
    "plt.matshow(confusion_matrix(y_true, y_pred))\n",
    "plt.title('Confusion matrix')\n",
    "plt.colorbar()\n",
    "plt.ylabel('True label')\n",
    "plt.xlabel('Predicted label')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7ce47484",
   "metadata": {},
   "source": [
    "## 10.Precission and Recall"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "0491f716",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy: 0.7\n",
      "[[4 1]\n",
      " [2 3]]\n",
      "(array([0.66666667, 0.75      ]), array([0.8, 0.6]), array([0.72727273, 0.66666667]), array([5, 5], dtype=int64))\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAQkAAAD+CAYAAAApiPBxAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/YYfK9AAAACXBIWXMAAAsTAAALEwEAmpwYAAAYRElEQVR4nO3de7hddX3n8feHEAl3xESFQADbiAO0XJpy8ylGakdImWLnsSNI8RnGFqFadMSZWqePIradS1vHUoQQhEFEsVKVQYhcxspAWrkkMQQCXjJSS0psSLgGIiTnfOaPtU7ZHM9ee52TvbP2Pvvzep71ZK+1fvu3fns/Z3/zu63fkm0iItrZqekCRER/S5CIiEoJEhFRKUEiIiolSEREpQSJiKiUINEwSbtK+oakZyTdsB35nCXp9m6WrSmSfkXS95suRxSUeRL1SHo38GHgTcBzwCrgT2wv2858zwZ+HzjR9rbtLWe/k2Rgvu21TZcl6klNogZJHwY+A/wp8DpgHnAZcHoXsj8I+MEwBIg6JO3cdBliHNvZKjZgb2Az8FsVaXahCCKPl9tngF3KcwuBdcCFwAZgPXBOee6TwEvA1vIa7wUuAq5ryftgwMDO5f6/B35EUZt5FDir5fiylvedCNwPPFP+e2LLuTuBTwF/V+ZzOzC7zWcbK/9/bin/O4BFwA+AJ4GPtaQ/FvgO8HSZ9lLgVeW5u8rP8nz5ed/Vkv8fAD8BvjB2rHzPz5XXOKbc3x/YCCxs+m9jWLbGC9DvG3AKsG3sR9omzcXAPcBrgTnA3wOfKs8tLN9/MTCz/HG9ALy6PD8+KLQNEsDuwLPAoeW5/YDDy9f/EiSAfYGngLPL951Z7r+mPH8n8P+ANwK7lvv/rc1nGyv/x8vy/y7wBPAlYE/gcOCnwBvK9L8EHF9e92DgEeBDLfkZ+PkJ8v/vFMF219YgUab53TKf3YDbgD9v+u9imLY0Nzp7DbDR1c2Bs4CLbW+w/QRFDeHslvNby/NbbS+l+F/00CmWZxQ4QtKuttfbXjNBml8Hfmj7C7a32b4e+B7wb1rS/C/bP7C9BfgKcFTFNbdS9L9sBb4MzAb+0vZz5fXXAL8IYHuF7XvK6/4DcAXwlhqf6RO2XyzL8wq2rwR+CNxLERj/S4f8oosSJDrbBMzu0FbeH/hxy/6Py2P/kse4IPMCsMdkC2L7eYoq+nnAekm3SHpTjfKMlWluy/5PJlGeTbZHytdjP+J/bjm/Zez9kt4o6WZJP5H0LEU/zuyKvAGesP3TDmmuBI4A/sr2ix3SRhclSHT2HYrq9Dsq0jxO0QE5Zl55bCqep6hWj3l960nbt9n+NYr/Ub9H8ePpVJ6xMv3TFMs0GZdTlGu+7b2AjwHq8J7KITZJe1D081wFXCRp3y6Uc9qSNEPSdyXdPME5SbpE0lpJqyUd0ym/BIkObD9D0R7/rKR3SNpN0kxJp0r6H2Wy64E/kjRH0uwy/XVTvOQq4CRJ8yTtDfzh2AlJr5P0G5J2B16kaLaMTJDHUuCNkt4taWdJ7wIOA37mj6YH9qToN9lc1nLOH3f+n4E3TDLPvwRW2P4d4BZg8XaXcnr7IEUfzkROBeaX27kUQb1SgkQNtj9NMUfijyg67R4DPgDcWCb5Y2A5sBp4EFhZHpvKte4A/rrMawWv/GHvRDFK8jhFj/9bgN+bII9NwGll2k0UIxOn2d44lTJN0keAd1OMmlxJ8VlaXQR8XtLTkv5dp8wknU7ReXxeeejDwDGSzupaiacRSQdQ9El9rk2S04FrXbgH2EfSfpV52plMFTFdSPob4L9S1Og+Yvu0cedvphjJWlbufwv4A9vL2+WZiSsRDXr7W3fzxidHa6VdufrFNRT9Y2OW2F4ytiPpNGCD7RWSFrbJZqL+ocqaQoJERIM2PjnC3986t3NCYNb+j/7U9oKKJG8GfkPSImAWsJek62z/dkuadcCBLfsH0KGTPX0SEQ0yMIprbR3zsv/Q9gG2DwbOAP52XIAAuAl4TznKcTzwjO31VfmmJhHRsFHqNTemStJ5ALYXU4x8LQLWUsyPOafT+xMkIhpkzEgPBg9s30kx3X4sOIwdN/D+yeSV5sYUSTpF0vfLSSkfbbo8042kqyVtkPRQ02XptW41N3olQWIKJM0APksxMeUw4ExJhzVbqmnnGor5EdOagRFca2tKmhtTcyyw1vaPACR9mWKSysONlmoasX2XpIObLkevGdjq3vZJbK/UJKZmLsWsyzHreOXNUxG1jdbcmpKaxNRMekJKxETccFOijgSJqZn0hJSICRlG+jtGpLkxRfcD8yUdIulVFBNXbmq4TDGAislU/d3cSJCYgnIBmQ9QLKX2CPCVNitExRRJup5iLY9DJa2T9N6my9QbYqTm1pQ0N6aoXIZuadPlmK5sn9l0GXYEA6N93txIkIhokIGX+rxCnyAR0bBRN9eUqCNBIqJBxYzLBImIaMOIkTQ3IqJKvzc3+juE9TlJ5zZdhuluun/HY82Nfh4CTZDYPtP6D7hPTPPvWIx4p1pbU9LciGiQga3MaLoYlfoqSMzed4YPPnBm08Wobd7cnVlw5Kw+nwrzSj9YvVvnRH1kFruxl/YdqO/4pzzPS36xVvvAVqO1hDr6KkgcfOBM7rvtwM4JY8revv9RTRdh2rvX35pU+tEMgUZEO0XHZWoSEdFWmhsRUaG4VTxBIiLaMOIlZ3QjIiqMprkREe2k4zIiKhkx0uf3biRIRDQsHZcR0ZZN3w+B9nfpIqY9MVpz65iTNEvSfZIekLRG0icnSLNQ0jOSVpXbxzvlm5pERIMMvOSu/QxfBE62vVnSTGCZpG/avmdcurttn1Y30wSJiAYZdW3RGdsGNpe7M8ttu2+OS3MjomEj7FRrq0PSDEmrgA3AHbbvnSDZCWWT5JuSDu+UZ2oSEQ0qnrtR+//q2ZKWt+wvsb3kFfnZI8BRkvYBvi7pCNsPtSRZCRxUNkkWATcC86sumiAR0ahJLU230faCOgltPy3pTuAU4KGW48+2vF4q6TJJs21vbJdXmhsRDRqrSdTZOpE0p6xBIGlX4G3A98aleb0kla+PpYgBm6ryTU0iomFdXOR2P+DzkmZQ/Pi/YvtmSecB2F4MvBM4X9I2YAtwRtnh2VaCRESDbLF1tDs/Q9urgaMnOL645fWlwKWTyTdBIqJBxXoSuXcjItrKylQRUaHouExNIiIqZD2JiGirm9OyeyVBIqJhWU8iItqyYetogkREtFE0NxIkIqJCF2dc9kSCRESDMgQaER2kuRERHWRadkS0VayWnSAREW0YsW00zwKNiAppbkREWxndiIiOMroREe05N3hFRIWsTBURHaUmERFtGdjW53eB9rR0kk6R9H1JayV9tJfXihhEY4vO1Nma0rMgUa79/1ngVOAw4ExJh/XqehGDahTV2prSy5rEscBa2z+y/RLwZeD0Hl4vYvCYvq9J9LJPYi7wWMv+OuC4Hl4vYuAM+2SqiT75zzxOTNK5wLkA8+amHzWGzzAHiXXAgS37BwCPj09UPjp9CcCCI2dVPpMwYroxYmSIRzfuB+ZLOkTSq4AzgJt6eL2IgTS0HZe2twEfAG4DHqF4wvGaXl0vYhC5ix2XkmZJuk/SA5LWSPrkBGkk6ZJyWsJqScd0yrennQC2lwJLe3mNiEHn7vVJvAicbHuzpJnAMknftH1PS5pTgfnldhxwOR0GFNJTGNGo7g1v2jawudydWW7j+/lOB64t094jaR9J+9le3y7f/u4xiRgCtmptdUiaIWkVsAG4w/a945JMNDVhblWeCRIRDRqbJ1GzT2K2pOUt27k/k589YvsoitHEYyUdMS5JrakJrdLciGjS5BbC3Wh7Qa1s7acl3QmcAjzUcqrW1IRWqUlENMh0r7khaY6kfcrXuwJvA743LtlNwHvKUY7jgWeq+iMgNYmIhnX1voz9gM+XN1fuRDHt4GZJ5wHYXkwx2rgIWAu8AJzTKdMEiYiGuUvzjG2vBo6e4PjiltcG3j+ZfBMkIhrWxXkSPZEgEdEgO0EiIjoY5rtAI6KG0dEEiYhow9SfTdmUBImIhvX7IioJEhFNSsdlRHTU51WJBImIhqUmERGVujXjslfaBglJf0VFRcj2BT0pUcQQscF9vhBuVU1i+Q4rRcQQG9iahO3Pt+5L2t32870vUsSQ6fMg0bGeI+kESQ9TrHiNpCMlXdbzkkUMhXprSTTZuVmnMfQZ4O3AJgDbDwAn9bBMEcPFNbeG1BrdsP2Y9IpINtKb4kQMmWkymeoxSScCLp/EdQFl0yMiumDQ+ySA8yhWspkL/BNwFJNc2SYiKlj1toZ0rEnY3gictQPKEjGcBr0mIekNkr4h6QlJGyT9b0lv2BGFi5j2TN/XJOo0N74EfIViJd79gRuA63tZqIhhUixh13lrSp0gIdtfsL2t3K6j7ytIEQNkUIdAJe1bvvy2pI8CX6Yo6ruAW3ZA2SKGwwAPga6gCApjn+B9LecMfKpXhYoYGgaNNl2IalX3bhyyIwsSMZya7ZSso9aMy/LJxIcBs8aO2b62V4WKGCp93sPXMUhI+gSwkCJILAVOBZYBCRIR3dDnQaLO6MY7gV8FfmL7HOBIYJeelipimPT56EadILHF9iiwTdJewAYgk6kiuqGLk6kkHSjp25IekbRG0gcnSLNQ0jOSVpXbxzvlW6dPYrmkfYArKUY8NgP31XhfRNSg7tUStgEX2l4paU9ghaQ7bD88Lt3dtk+rm2mdezd+r3y5WNKtwF7lI84johu6FCRsrwfWl6+fk/QIxY2Z44PEpFRNpjqm6pztldtz4Yk8/Pgcfumi87udbbSYufSJposw7Y1csGxS6btYk3g5T+lg4Gjg3glOnyDpAeBx4CO211TlVVWT+IuKcwZO7lDOiKij/jyJ2ZJaF6heYnvJ+ESS9gC+CnzI9rPjTq8EDrK9WdIi4EZgftVFqyZTvbVuySNiiiY3crHR9oKqBJJmUgSIL9r+2s9criVo2F4q6TJJs8slISbU3wv+RwyDLg2Bqlhj8irgEdufbpPm9WU6JB1LEQM2VeWbJ3hFNKyLfRJvBs4GHpS0qjz2MWAegO3FFPOezpe0DdgCnGFX34ieIBHRtO6Nbizj5Rsy26W5FLh0MvnWWZlKkn57bNKFpHllNSUitpPKu0DrbE2p0ydxGXACcGa5/xzw2Z6VKGLY9PnydXWaG8fZPkbSdwFsP1UurR8R3dDnN3jVCRJbJc2g/CiS5gB9vkxGxODoxWSqbqrT3LgE+DrwWkl/QnGb+J/2tFQRw6TP7wKtc+/GFyWtoLhdXMA7bOcJXhHd4P6vSdRZdGYe8ALwjdZjtv+xlwWLGBqDHiQoVsYeWxB3FnAI8H3g8B6WK2JoDOxCuGNs/0Lrfnl36PvaJI+IaWbSMy7LBS1+uReFiRhKg97ckPThlt2dgGOALEoQ0Q3ToeMS2LPl9TaKPoqv9qY4EUNokINEOYlqD9v/aQeVJ2L4DGqQkLSz7W1Vy9hFxPYRg93cuI+i/2GVpJuAG4Dnx05OtOpNREzSID8LtMW+FCvXnMzL8yUMJEhEdMMA1yReW45sPMQrny4Off+xIgZIn/+aqoLEDGAPJl7pps8/VsTgGOQ+ifW2L95hJYkYVgMcJJpbCidiWDR8G3gdVUHiV3dYKSKG2MCObth+ckcWJGJYDXKfRETsCAkSEdHWgPdJRESPif4fIUiQiGhaahIRUSUdlxFRbVCHQCNiBxiAlanqPJwnInqpSw/nkXSgpG9LekTSGkkfnCCNJF0iaa2k1XXWi0lNIqJhXaxJbAMuLBer3hNYIekO2w+3pDkVmF9uxwGXl/+2lZpERNO6VJOwvd72yvL1c8AjwNxxyU4HrnXhHmAfSftV5ZuaRETDJlGTmC1pecv+EttLJsxTOhg4Grh33Km5wGMt++vKY+vbXTRBIqJJk5txudH2gk6JJO1BsaL9h2w/O/50m1K0lSAR0SDR3btAJc2kCBBfbLMO7TrgwJb9A4DHq/JMn0RE07o3uiHgKuAR259uk+wm4D3lKMfxwDO22zY1oIc1CUlXA6cBG2wf0avrRAw6uWvDG28GzgYelLSqPPYxYB6A7cXAUmARsBZ4ATinU6a9bG5cA1wKXNvDa0QMti7eBWp7GR3uF7Nt4P2TybdnQcL2XWUPa0RU6PcZl+m4jGhagkQ1SecC5wLM3OPVDZcmYsfr95pE46MbtpfYXmB7wc6zdm+6OBE7VvmYvzpbUxqvSUQMvWGtSUi6HvgOcKikdZLe26trRQyqsaeK19ma0svRjTN7lXfEtNK9eRI9keZGRMP6veMyQSKiSVlSPyI6GdjH/EXEjpEgERHtmXRcRkS1dFxGRLUEiYhoZ2wyVT9LkIhokp0+iYioltGNiKiU5kZEtGdgtL+jRIJERNP6O0YkSEQ0Lc2NiKiW0Y2IqJKaRES0JYPScRkRlTJPIiKqdPExfz2RIBHRpKxMFRHV+v/ejcYfzhMx7Lq5pL6kqyVtkPRQm/MLJT0jaVW5fbxTnqlJRDStuzWJa4BLgWsr0txt+7S6GSZIRDTJoJHuBQnbd0k6uGsZkuZGRPNcc+ueEyQ9IOmbkg7vlDg1iYiGTWIIdLak5S37S2wvmeTlVgIH2d4saRFwIzC/6g0JEhFNqx8kNtpesH2X8rMtr5dKukzSbNsb270nzY2IJplixmWdrQskvV6SytfHUsSATVXvSU0iokHCXZ1xKel6YCFF02Qd8AlgJoDtxcA7gfMlbQO2AGfY1QVIkIhoWheDhO0zO5y/lGKItLYEiYgmGejiEGgvJEhENCw3eEVEtQSJiGiv/2/wSpCIaFKeKh4RHWVlqoioko7LiGjPwEh/VyUSJCIalY7LSdmycd3GVVdc+OOmyzEJs4G2N8b0pSuaLsCkDd53DAdNKnWCRH225zRdhsmQtHx778qLakPxHSdIRERbeap4RFQzOB2X09lkVwWKyZve3/EAjG5k0ZntUGfpMEkj5dLlD0m6QdJuU72epGskvbN8/TlJh1WkXSjpxClc4x8kza57fFyazZO81kWSPlKVZgrLsw0eu97WkASJ3tti+yjbRwAvAee1npQ0YyqZ2v4d2w9XJFkITDpIRAMSJKLF3cDPl//Lf1vSl4AHJc2Q9GeS7pe0WtL7AFS4VNLDkm4BXjuWkaQ7JS0oX58iaWW5AvK3yiXVzwP+Y1mL+RVJcyR9tbzG/ZLeXL73NZJul/RdSVcA6vQhJN0oaYWkNZLOHXfuL8qyfEvSnPLYz0m6tXzP3ZLe1JVvc1qoGSAaDBLpk9hBJO0MnArcWh46FjjC9qPlD+0Z278saRfg7yTdDhwNHAr8AvA64GHg6nH5zgGuBE4q89rX9pOSFgObbf95me5LwP+0vUzSPOA24F9RLG+2zPbFkn4deMWPvo3/UF5jV+B+SV+1vQnYHVhp+8LyyVCfAD5A0a9wnu0fSjoOuAw4eQpf4/RjYLS/+yQSJHpvV0mrytd3A1dRNAPus/1oefxfA7841t8A7E2xzPlJwPW2R4DHJf3tBPkfD9w1lpftJ9uU423AYeUaqAB7SdqzvMa/Ld97i6SnanymCyT9Zvn6wLKsmyhuVfrr8vh1wNck7VF+3htarr1LjWsMj8yTGHpbbB/VeqD8sTzfegj4fdu3jUu3iM6PZVGNNFA0LU+wvWWCstT+K5W0kCLgnGD7BUl3ArPaJHd53afHfwfRos+DRPok+sNtFCsYzwSQ9EZJuwN3AWeUfRb7AW+d4L3fAd4i6ZDyvfuWx58D9mxJdztF1Z8y3VHly7uAs8pjpwKv7lDWvYGnygDxJoqazJidKFZjBng3RTPmWeBRSb9VXkOSjuxwjeFh45GRWltTEiT6w+co+htWqnga9BUUtbyvAz8EHgQuB/7v+DfafoKiH+Frkh7g5er+N4DfHOu4BC4AFpQdow/z8ijLJ4GTJK2kaPb8Y4ey3grsLGk18CngnpZzzwOHS1pB0edwcXn8LOC9ZfnWAKfX+E6Gx6jrbQ1RhyX3I6KH9t55jk/Ys17MvO3pq1Y0cR9L+iQimmRndCMiOujz2nyCRETDnJpERLSXlakiooqBBoc368gQaESDDHjUtbY6JF0taUM5lD7ReUm6RNLacjj8mE55JkhENMnlojN1tnquAU6pOH8qxTT6+RTzay7vlGGCRETDulmTsH0X0O7+HSgmsl3rwj3APuVs3rbSJxHRtB27fN1c4LGW/XXlsfXt3pAgEdGg53jqtv/jv6lc8avFLEnLW/aXTGHlronWC6mspiRIRDTIdlX/QS+so7i9f8wBwONVb0ifRMRwuQl4TznKcTzFYkdtmxqQmkTEtCLpeor1TWdLWkexOthMANuLgaXAImAt8AJwTsc8cxdoRFRJcyMiKiVIRESlBImIqJQgERGVEiQiolKCRERUSpCIiEoJEhFR6f8DX0TOvHIgnbQAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 288x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "from sklearn.metrics import precision_recall_fscore_support\n",
    "from sklearn.metrics import accuracy_score\n",
    "from sklearn.metrics import confusion_matrix\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "y_true = [0, 0, 0, 0, 0, 1, 1, 1, 1, 1]\n",
    "y_pred = [0, 1, 0, 0, 0, 0, 0, 1, 1, 1]\n",
    "\n",
    "print ('Accuracy:', accuracy_score(y_true, y_pred))\n",
    "print (confusion_matrix(y_true, y_pred))\n",
    "print (precision_recall_fscore_support(y_true, y_pred))\n",
    "\n",
    "plt.matshow(confusion_matrix(y_true, y_pred))\n",
    "plt.title('Confusion matrix')\n",
    "plt.colorbar()\n",
    "plt.ylabel('True label')\n",
    "plt.xlabel('Predicted label')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0887c776",
   "metadata": {},
   "source": [
    "## 11.Multiclass Classification"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "fcfcb9cb",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.9736842105263158\n",
      "[[13  0  0]\n",
      " [ 0 16  0]\n",
      " [ 0  1  8]]\n",
      "(array([1.        , 0.94117647, 1.        ]), array([1.        , 1.        , 0.88888889]), array([1.        , 0.96969697, 0.94117647]), array([13, 16,  9], dtype=int64))\n",
      "1.0\n",
      "[[13  0  0]\n",
      " [ 0 16  0]\n",
      " [ 0  0  9]]\n",
      "(array([1., 1., 1.]), array([1., 1., 1.]), array([1., 1., 1.]), array([13, 16,  9], dtype=int64))\n",
      "1.0\n",
      "[[13  0  0]\n",
      " [ 0 16  0]\n",
      " [ 0  0  9]]\n",
      "(array([1., 1., 1.]), array([1., 1., 1.]), array([1., 1., 1.]), array([13, 16,  9], dtype=int64))\n",
      "1.0\n",
      "[[13  0  0]\n",
      " [ 0 16  0]\n",
      " [ 0  0  9]]\n",
      "(array([1., 1., 1.]), array([1., 1., 1.]), array([1., 1., 1.]), array([13, 16,  9], dtype=int64))\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import confusion_matrix \n",
    "from sklearn.metrics import precision_recall_fscore_support\n",
    "import pandas as pd\n",
    "from sklearn.model_selection import train_test_split \n",
    "from sklearn.tree import DecisionTreeClassifier \n",
    "from sklearn.svm import SVC \n",
    "from sklearn.neighbors import KNeighborsClassifier \n",
    "from sklearn.naive_bayes import GaussianNB \n",
    "  \n",
    "df = pd.read_csv('iris.csv')\n",
    "X = df[list(df.columns)[:-1]]\n",
    "y = df['Species']\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y, random_state = 0)  \n",
    "\n",
    "tree = DecisionTreeClassifier(max_depth = 2).fit(X_train, y_train) \n",
    "tree_predictions = tree.predict(X_test)   \n",
    "print (tree.score(X_test, y_test))\n",
    "print (confusion_matrix(y_test, tree_predictions))\n",
    "print (precision_recall_fscore_support(y_test, tree_predictions))\n",
    "\n",
    "svc = SVC(kernel = 'linear', C = 1).fit(X_train, y_train) \n",
    "svc_predictions = svc.predict(X_test)  \n",
    "print (svc.score(X_test, y_test))\n",
    "print (confusion_matrix(y_test, svc_predictions))\n",
    "print (precision_recall_fscore_support(y_test, svc_predictions))\n",
    "\n",
    "knn = KNeighborsClassifier(n_neighbors = 7).fit(X_train, y_train)  \n",
    "knn_predictions = knn.predict(X_test)  \n",
    "print (knn.score(X_test, y_test)) \n",
    "print (confusion_matrix(y_test, knn_predictions))\n",
    "print (precision_recall_fscore_support(y_test, knn_predictions))\n",
    "\n",
    "gnb = GaussianNB().fit(X_train, y_train) \n",
    "gnb_predictions = gnb.predict(X_test) \n",
    "print (gnb.score(X_test, y_test))\n",
    "print (confusion_matrix(y_test, gnb_predictions))\n",
    "print (precision_recall_fscore_support(y_test, gnb_predictions))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "54614537",
   "metadata": {},
   "outputs": [],
   "source": [
    "12."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "232b2728",
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "from io import StringIO\n",
    "import matplotlib.pyplot as plt\n",
    "from sklearn.feature_extraction.text import TfidfVectorizer\n",
    "from sklearn.feature_selection import chi2\n",
    "import numpy as np\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.feature_extraction.text import CountVectorizer\n",
    "from sklearn.feature_extraction.text import TfidfTransformer\n",
    "from sklearn.naive_bayes import MultinomialNB\n",
    "\n",
    "df = pd.read_csv('./Consumer_Complaints.csv', sep=',', error_bad_lines=False, index_col=False, dtype='unicode')\n",
    "df = df[pd.notnull(df['Issue'])]\n",
    "\n",
    "fig = plt.figure(figsize=(8,6))\n",
    "df.groupby('Product').Issue.count().plot.bar(ylim=0)\n",
    "plt.show()\n",
    "\n",
    "X_train, X_test, y_train, y_test = train_test_split(df['Issue'], df['Product'], random_state = 0)\n",
    "c = CountVectorizer()\n",
    "clf = MultinomialNB().fit (TfidfTransformer().fit_transform(c.fit_transform(X_train)), y_train)\n",
    "\n",
    "print(clf.predict(c.transform([\"This company refuses to provide me verification and validation of debt per my right under the FDCPA. I do not believe this debt is mine.\"])))\n",
    "\n",
    "tfidf = TfidfVectorizer(sublinear_tf=True, min_df=5, norm='l2', encoding='latin-1', ngram_range=(1, 2), stop_words='english')\n",
    "features = tfidf.fit_transform(df.Issue).toarray()\n",
    "print (features)\n",
    "df['category_id'] = df['Product'].factorize()[0]\n",
    "pro_cat = df[['Product', 'category_id']].drop_duplicates().sort_values('category_id')\n",
    "print (pro_cat)\n",
    "for i, j in sorted(dict(pro_cat.values).items()):\n",
    "\tindices = np.argsort(chi2(features, df.category_id == j)[0])\n",
    "\tprint (indices)\n",
    "\tfeature_names = np.array(tfidf.get_feature_names())[indices]\n",
    "\tunigrams = [i for i in feature_names if len(i.split(' ')) == 1]\n",
    "\tbigrams = [i for i in feature_names if len(i.split(' ')) == 2]\n",
    "\tprint(\">\",i)\n",
    "\tprint(\"unigrams:\",','.join(unigrams[:5]))\n",
    "\tprint(\"bigrams:\",','.join(bigrams[:5]))\n",
    "\n",
    "\n",
    "\n",
    "\"\"\"\n",
    "\n",
    "\n",
    "print (features.shape)\n",
    "\n",
    "\n",
    "df['category_id'] = df['Product'].factorize()[0]\n",
    "pro_cat = df[['Product', 'category_id']].drop_duplicates().sort_values('category_id')\n",
    "for i, j in sorted(dict(pro_cat.values).items()):\n",
    "\tindices = np.argsort(chi2(features, df.category_id == j)[0])\n",
    "\tfeature_names = np.array(tfidf.get_feature_names())[indices]\n",
    "\tunigrams = [i for i in feature_names if len(i.split(' ')) == 1]\n",
    "\tbigrams = [i for i in feature_names if len(i.split(' ')) == 2]\n",
    "\tprint(\">\",i)\n",
    "\tprint(\"unigrams:\",','.join(unigrams[:5]))\n",
    "\tprint(\"bigrams:\",','.join(bigrams[:5]))\n",
    "\"\"\""
   ]
  },
  {
   "cell_type": "markdown",
   "id": "31ae00ad",
   "metadata": {},
   "source": [
    "## 13.Clustering"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "03eecbf7",
   "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"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(15.142857142857142, 15.0, 8.2, 8.2)\n"
     ]
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "import math\n",
    "\n",
    "def plots(cluster1_x1, cluster1_x2, cluster2_x1, cluster2_x2):\n",
    "    plt.figure()\n",
    "    plt.plot(cluster1_x1, cluster1_x2, '.')\n",
    "    plt.plot(cluster2_x1, cluster2_x2, '*')\n",
    "    plt.grid(True)\n",
    "    plt.show()\n",
    "\n",
    "def round1(c1_x1, c1_x2, c2_x1, c2_x2):\n",
    "    cluster1_x1 = []\n",
    "    cluster1_x2 = []\n",
    "    cluster2_x1 = []\n",
    "    cluster2_x2 = []\n",
    "\n",
    "    for i, j in zip(x1, x2):\n",
    "        a = math.sqrt((i - c1_x1) ** 2 + (j - c1_x2) ** 2)\n",
    "        b = math.sqrt((i - c2_x1) ** 2 + (j - c2_x2) ** 2)\n",
    "        if a < b:\n",
    "            cluster1_x1.append(i)\n",
    "            cluster1_x2.append(j)\n",
    "        else:\n",
    "            cluster2_x1.append(i)\n",
    "            cluster2_x2.append(j)\n",
    "\n",
    "    plots(cluster1_x1, cluster1_x2, cluster2_x1, cluster2_x2)\n",
    "\n",
    "    c1_x1 = sum(cluster1_x1) / len(cluster1_x1)\n",
    "    c1_x2 = sum(cluster1_x2) / len(cluster1_x2)\n",
    "    c2_x1 = sum(cluster2_x1) / len(cluster2_x1)\n",
    "    c2_x2 = sum(cluster2_x2) / len(cluster2_x2)\n",
    "\n",
    "    round2(c1_x1, c1_x2, c2_x1, c2_x2)\n",
    "    print((c1_x1, c1_x2, c2_x1, c2_x2))\n",
    "\n",
    "def round2(c1_x1, c1_x2, c2_x1, c2_x2):\n",
    "    cluster1_x1 = []\n",
    "    cluster1_x2 = []\n",
    "    cluster2_x1 = []\n",
    "    cluster2_x2 = []\n",
    "\n",
    "    for i, j in zip(x1, x2):\n",
    "        c = math.sqrt((i - c1_x1) ** 2 + (j - c1_x2) ** 2)\n",
    "        d = math.sqrt((i - c2_x1) ** 2 + (j - c2_x2) ** 2)\n",
    "        if c < d:\n",
    "            cluster1_x1.append(i)\n",
    "            cluster1_x2.append(j)\n",
    "        else:\n",
    "            cluster2_x1.append(i)\n",
    "            cluster2_x2.append(j)\n",
    "\n",
    "    plots(cluster1_x1, cluster1_x2, cluster2_x1, cluster2_x2)\n",
    "\n",
    "x1 = [15, 19, 15, 5, 13, 17, 15, 12, 8, 6, 9, 13]\n",
    "x2 = [13, 16, 17, 6, 17, 14, 15, 13, 7, 6, 10, 12]\n",
    "\n",
    "plots(x1, x2, [], [])\n",
    "round1(x1[4], x2[4], x1[10], x2[10])"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8887d1a5",
   "metadata": {},
   "source": [
    "## 14.Elbow Method"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "366232f5",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\Admin\\anaconda3\\lib\\site-packages\\sklearn\\cluster\\_kmeans.py:1332: UserWarning: KMeans is known to have a memory leak on Windows with MKL, when there are less chunks than available threads. You can avoid it by setting the environment variable OMP_NUM_THREADS=1.\n",
      "  warnings.warn(\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\Admin\\anaconda3\\lib\\site-packages\\sklearn\\cluster\\_kmeans.py:1332: UserWarning: KMeans is known to have a memory leak on Windows with MKL, when there are less chunks than available threads. You can avoid it by setting the environment variable OMP_NUM_THREADS=1.\n",
      "  warnings.warn(\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "3\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\Admin\\anaconda3\\lib\\site-packages\\sklearn\\cluster\\_kmeans.py:1332: UserWarning: KMeans is known to have a memory leak on Windows with MKL, when there are less chunks than available threads. You can avoid it by setting the environment variable OMP_NUM_THREADS=1.\n",
      "  warnings.warn(\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "4\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\Admin\\anaconda3\\lib\\site-packages\\sklearn\\cluster\\_kmeans.py:1332: UserWarning: KMeans is known to have a memory leak on Windows with MKL, when there are less chunks than available threads. You can avoid it by setting the environment variable OMP_NUM_THREADS=1.\n",
      "  warnings.warn(\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "5\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\Admin\\anaconda3\\lib\\site-packages\\sklearn\\cluster\\_kmeans.py:1332: UserWarning: KMeans is known to have a memory leak on Windows with MKL, when there are less chunks than available threads. You can avoid it by setting the environment variable OMP_NUM_THREADS=1.\n",
      "  warnings.warn(\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "6\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\Admin\\anaconda3\\lib\\site-packages\\sklearn\\cluster\\_kmeans.py:1332: UserWarning: KMeans is known to have a memory leak on Windows with MKL, when there are less chunks than available threads. You can avoid it by setting the environment variable OMP_NUM_THREADS=1.\n",
      "  warnings.warn(\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "7\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\Admin\\anaconda3\\lib\\site-packages\\sklearn\\cluster\\_kmeans.py:1332: UserWarning: KMeans is known to have a memory leak on Windows with MKL, when there are less chunks than available threads. You can avoid it by setting the environment variable OMP_NUM_THREADS=1.\n",
      "  warnings.warn(\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": [
    "from sklearn.cluster import KMeans\n",
    "from sklearn import metrics\n",
    "from scipy.spatial.distance import cdist\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "x1 = [15, 19, 15, 5, 13, 17, 15, 12, 8, 6, 9, 13]\n",
    "x2 = [13, 16, 17, 6, 17, 14, 15, 13, 7, 6, 10, 12]\n",
    "\n",
    "X = np.array(list(zip(x1, x2)))\n",
    "\n",
    "distortions = []\n",
    "K = range(1,8)\n",
    "for i in K:\n",
    "    print (i)\n",
    "    model = KMeans(n_clusters=i)\n",
    "    model.fit(X)\n",
    "    distortions.append(sum(np.min(cdist(X, model.cluster_centers_, 'euclidean'), axis=1)) / X.shape[0])\n",
    "\n",
    "plt.plot()\n",
    "plt.plot(K, distortions, 'bx-')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4e084622",
   "metadata": {},
   "source": [
    "## 15.Silhouette Coefficient"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "2885623b",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\Admin\\anaconda3\\lib\\site-packages\\sklearn\\cluster\\_kmeans.py:1332: UserWarning: KMeans is known to have a memory leak on Windows with MKL, when there are less chunks than available threads. You can avoid it by setting the environment variable OMP_NUM_THREADS=1.\n",
      "  warnings.warn(\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[0 0 0 1 0 0 0 0 1 1 1 0]\n",
      "0.6366488776743281\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\Admin\\anaconda3\\lib\\site-packages\\sklearn\\cluster\\_kmeans.py:1332: UserWarning: KMeans is known to have a memory leak on Windows with MKL, when there are less chunks than available threads. You can avoid it by setting the environment variable OMP_NUM_THREADS=1.\n",
      "  warnings.warn(\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[2 2 2 1 2 2 2 0 1 1 0 0]\n",
      "0.44020495155817824\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\Admin\\anaconda3\\lib\\site-packages\\sklearn\\cluster\\_kmeans.py:1332: UserWarning: KMeans is known to have a memory leak on Windows with MKL, when there are less chunks than available threads. You can avoid it by setting the environment variable OMP_NUM_THREADS=1.\n",
      "  warnings.warn(\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[2 3 1 0 1 3 1 2 0 0 0 2]\n",
      "0.44041109650563554\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\Admin\\anaconda3\\lib\\site-packages\\sklearn\\cluster\\_kmeans.py:1332: UserWarning: KMeans is known to have a memory leak on Windows with MKL, when there are less chunks than available threads. You can avoid it by setting the environment variable OMP_NUM_THREADS=1.\n",
      "  warnings.warn(\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[0 2 3 1 3 2 3 0 1 1 4 0]\n",
      "0.38043265897525885\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\Admin\\anaconda3\\lib\\site-packages\\sklearn\\cluster\\_kmeans.py:1332: UserWarning: KMeans is known to have a memory leak on Windows with MKL, when there are less chunks than available threads. You can avoid it by setting the environment variable OMP_NUM_THREADS=1.\n",
      "  warnings.warn(\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[2 3 0 5 0 6 2 4 1 5 7 4]\n",
      "0.27672998081717154\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 6 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "import numpy as np\n",
    "from sklearn.cluster import KMeans\n",
    "from sklearn import metrics\n",
    "import matplotlib.pyplot as plt\n",
    "plt.subplot(3, 2, 1)\n",
    "\n",
    "x1 = [15, 19, 15, 5, 13, 17, 15, 12, 8, 6, 9, 13]\n",
    "x2 = [13, 16, 17, 6, 17, 14, 15, 13, 7, 6, 10, 12]\n",
    "plt.scatter(x1, x2)\n",
    "\n",
    "X = np.array(list(zip(x1, x2)))\n",
    "\n",
    "c = ['b', 'g', 'r', 'c', 'm', 'y', 'k', 'b']\n",
    "m = ['o', 's', 'D', 'v', '^', 'p', '*', '+']\n",
    "\n",
    "p = 1\n",
    "for i in [2, 3, 4, 5, 8]:\n",
    "    p += 1\n",
    "    plt.subplot(3, 2, p)\n",
    "    model = KMeans(n_clusters=i).fit(X)\n",
    "    print (model.labels_)\n",
    "    for i, j in enumerate(model.labels_):\n",
    "        plt.plot(x1[i], x2[i], color=c[j], marker=m[j],ls='None')\n",
    "    print (metrics.silhouette_score(X, model.labels_ ,metric='euclidean'))\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "65eed4ed",
   "metadata": {},
   "source": [
    "## 16.SVC_Linear"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "d73f4e60",
   "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"
    },
    {
     "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": [
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "from sklearn import svm\n",
    "\n",
    "def classifier(regressor):\n",
    "    xx = np.linspace(1, 10)\n",
    "    yy = -regressor.coef_[0][0] / regressor.coef_[0][1] * xx - regressor.intercept_[0] / regressor.coef_[0][1]\n",
    "    plt.plot(xx, yy)\n",
    "    plt.scatter(x1, x2)\n",
    "    plt.show()\n",
    "\n",
    "x1 = [2, 6, 3, 9, 4, 10]\n",
    "x2 = [3, 9, 3, 10, 2, 13]\n",
    "\n",
    "X = np.array([[2, 3], [6, 9], [3, 3], [9, 10], [4, 2], [10, 13]])\n",
    "y = [0, 1, 0, 1, 0, 1]\n",
    "\n",
    "regressor = LogisticRegression()\n",
    "regressor.fit(X, y)\n",
    "classifier(regressor)\n",
    "\n",
    "regressor = svm.SVC(kernel='linear', C=1.0)\n",
    "regressor.fit(X, y)\n",
    "classifier(regressor)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6dcc0e57",
   "metadata": {},
   "source": [
    "## 17.SVM Nonlinear Kernel"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "fd13577d",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy-logistic: 1.0\n",
      "Accuracy-svm: 1.0\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\Admin\\anaconda3\\lib\\site-packages\\sklearn\\linear_model\\_logistic.py:444: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  n_iter_i = _check_optimize_result(\n"
     ]
    }
   ],
   "source": [
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import pandas as pd\n",
    "from sklearn.metrics import accuracy_score\n",
    "from sklearn.model_selection import train_test_split \n",
    "from sklearn.svm import SVC \n",
    "from sklearn.linear_model import LogisticRegression\n",
    "\n",
    "df = pd.read_csv('iris.csv')\n",
    "X = df.iloc[:, :-1]  # Use iloc to select all columns except the last one\n",
    "y = df.iloc[:, -1]   # Use iloc to select the last column\n",
    "\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y, random_state=0)\n",
    "\n",
    "logistic = LogisticRegression()\n",
    "logistic.fit(X_train, y_train)  \n",
    "y_pred = logistic.predict(X_test) \n",
    "print('Accuracy-logistic:', accuracy_score(y_test, y_pred))\n",
    "\n",
    "gaussian = SVC(kernel='rbf') \n",
    "gaussian.fit(X_train, y_train)  \n",
    "y_pred = gaussian.predict(X_test) \n",
    "print('Accuracy-svm:', accuracy_score(y_test, y_pred))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "eb5f1b8c",
   "metadata": {},
   "source": [
    "## 18.PCA"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "f60ed05b",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[0.72207932 0.24134489]\n",
      "Index(['Sepal_length', 'Sepal_width', 'Petal_length', 'Petal_width', 'Flower'], dtype='object')\n",
      "Index(['x1', 'x2', 'Flower'], dtype='object')\n"
     ]
    }
   ],
   "source": [
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import pandas as pd\n",
    "from sklearn.model_selection import train_test_split \n",
    "from sklearn.preprocessing import StandardScaler\n",
    "from sklearn.decomposition import PCA\n",
    "\n",
    "df = pd.read_csv('flower.csv')\n",
    "X = df[list(df.columns)[:-1]]\n",
    "y = df['Flower']\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y, random_state = 0)  \n",
    "\n",
    "pca = PCA(n_components=2)\n",
    "x = StandardScaler().fit_transform(X_train)\n",
    "new_x = pd.DataFrame(data = pca.fit_transform(x), columns = ['x1', 'x2'])\n",
    "\n",
    "df2 = pd.concat([new_x, df[['Flower']]], axis = 1)\n",
    "\n",
    "fig = plt.figure(figsize = (8,8))\n",
    "ax = fig.add_subplot(1,1,1) \n",
    "ax.set_xlabel('x1', fontsize = 15)\n",
    "ax.set_ylabel('x2', fontsize = 15)\n",
    "ax.set_title('2 Components', fontsize = 20)\n",
    "for i, j in zip(['Rose', 'Jasmin', 'Lotus'],['g', 'b', 'r']):\n",
    "    ax.scatter(df2.loc[df2['Flower'] == i, 'x1'], df2.loc[df2['Flower'] == i, 'x2'], c = j)\n",
    "ax.legend(['Rose', 'Jasmin', 'Lotus'])\n",
    "ax.grid()\n",
    "plt.show()\n",
    "\n",
    "print (pca.explained_variance_ratio_)\n",
    "\n",
    "print (df.columns)\n",
    "print (df2.columns)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "db7f0028",
   "metadata": {},
   "source": [
    "## 18.perceptron"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "26214634",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "epoch=0, error=1.00\n",
      "epoch=1, error=3.00\n",
      "epoch=2, error=2.00\n",
      "epoch=3, error=2.00\n",
      "epoch=4, error=1.00\n",
      "epoch=5, error=0.00\n",
      "[0.1, -0.13999999999999999, 0.12]\n"
     ]
    }
   ],
   "source": [
    "def predict(row, weights):\n",
    "    activation = weights[0]\n",
    "    for i in range(len(row)-1):\n",
    "        activation += weights[i + 1] * row[i]\n",
    "    return 1.0 if activation >= 0.0 else 0.0  # Changed \">\" to \">=\" for activation threshold\n",
    "\n",
    "def train_weights(dataset, l_rate, n_epoch):\n",
    "    weights = [0.0 for _ in range(len(dataset[0]))]\n",
    "    for epoch in range(n_epoch):\n",
    "        sum_error = 0.0\n",
    "        for row in dataset:\n",
    "            error = row[-1] - predict(row, weights)\n",
    "            sum_error += error ** 2\n",
    "            weights[0] = weights[0] + l_rate * error\n",
    "            for i in range(len(row)-1):\n",
    "                weights[i + 1] = weights[i + 1] + l_rate * error * row[i]\n",
    "        print('epoch=%d, error=%.2f' % (epoch, sum_error))\n",
    "    print(weights)\n",
    "\n",
    "dataset = [\n",
    "    [0.4, 0.3, 1],\n",
    "    [0.6, 0.8, 1],\n",
    "    [0.7, 0.5, 1],\n",
    "    [0.9, 0.2, 0]\n",
    "]\n",
    "\n",
    "l_rate = 0.1\n",
    "n_epoch = 6\n",
    "train_weights(dataset, l_rate, n_epoch)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "21afa2bb",
   "metadata": {},
   "source": [
    "## 21.Perceptron"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "2d6a80e3",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.85      0.95      0.90       396\n",
      "           1       0.89      0.83      0.86       397\n",
      "           2       0.91      0.86      0.89       399\n",
      "\n",
      "    accuracy                           0.88      1192\n",
      "   macro avg       0.88      0.88      0.88      1192\n",
      "weighted avg       0.88      0.88      0.88      1192\n",
      "\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.87      0.92      0.89       396\n",
      "           1       0.87      0.82      0.84       397\n",
      "           2       0.88      0.88      0.88       399\n",
      "\n",
      "    accuracy                           0.87      1192\n",
      "   macro avg       0.87      0.87      0.87      1192\n",
      "weighted avg       0.87      0.87      0.87      1192\n",
      "\n"
     ]
    }
   ],
   "source": [
    "from sklearn.datasets import fetch_20newsgroups\n",
    "from sklearn.metrics import classification_report\n",
    "from sklearn.feature_extraction.text import TfidfVectorizer\n",
    "from sklearn.linear_model import Perceptron, LogisticRegression\n",
    "\n",
    "categories = ['rec.sport.hockey', 'rec.sport.baseball', 'rec.autos']\n",
    "\n",
    "x1 = fetch_20newsgroups(subset='train', categories=categories, remove=('headers', 'footers', 'quotes'))\n",
    "x2 = fetch_20newsgroups(subset='test', categories=categories, remove=('headers', 'footers', 'quotes'))\n",
    "\n",
    "vectorizer = TfidfVectorizer()\n",
    "X_train = vectorizer.fit_transform(x1.data)\n",
    "X_test = vectorizer.transform(x2.data)\n",
    "\n",
    "classifier = LogisticRegression()\n",
    "classifier.fit(X_train, x1.target)\n",
    "predictions = classifier.predict(X_test)\n",
    "print(classification_report(x2.target, predictions))\n",
    "\n",
    "classifier = Perceptron(max_iter=100, eta0=0.1)\n",
    "classifier.fit(X_train, x1.target)\n",
    "predictions = classifier.predict(X_test)\n",
    "print(classification_report(x2.target, predictions))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8c616e0f",
   "metadata": {},
   "source": [
    "## 22.MultiLayer Perceptron"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "id": "93348a8d",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Iteration: 150/150 | Cost 0.06 | Elapsed: 0:00:00 | ETA: 0:00:00"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Iteration: 71/250 | Cost 0.09 | Elapsed: 0:00:00 | ETA: 0:00:00"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy(epochs = 150): 93.75%\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Iteration: 250/250 | Cost 0.04 | Elapsed: 0:00:00 | ETA: 0:00:00"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy(epochs = 250): 100.00%\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"
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Iteration: 250/250 | Cost 0.02 | Elapsed: 0:00:01 | ETA: 0:00:00"
     ]
    },
    {
     "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": [
    "from mlxtend.classifier import MultiLayerPerceptron as MLP\n",
    "from mlxtend.plotting import plot_decision_regions\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np \n",
    "\n",
    "X = np.asarray([[6.1,1.4],[7.7,2.3],[6.3,2.4],[6.4,1.8],[6.2,1.8],[6.9,2.1],\n",
    "[6.7,2.4],[6.9,2.3],[5.8,1.9],[6.8,2.3],[6.7,2.5],[6.7,2.3],[6.3,1.9],[6.5,2.1 ],[6.2,2.3],[5.9,1.8]] ) \n",
    "\n",
    "X = (X - X.mean(axis=0)) / X.std(axis=0)\n",
    "\n",
    "y = np.asarray([0,2,2,1,2,2,2,2,2,2,2,2,2,2,2,2]) \n",
    "\n",
    "nn = MLP(hidden_layers=[50],l2=0.00,l1=0.0,epochs=150,eta=0.05,\n",
    "momentum=0.1,decrease_const=0.0,minibatches=1,random_seed=1,print_progress=3)\n",
    "nn = nn.fit(X, y)\n",
    "\n",
    "fig = plot_decision_regions(X=X, y=y, clf=nn, legend=2)\n",
    "plt.show()\n",
    "print('Accuracy(epochs = 150): %.2f%%' % (100 * nn.score(X, y)))\n",
    "\n",
    "nn.epochs = 250\n",
    "nn = nn.fit(X, y)\n",
    "fig = plot_decision_regions(X=X, y=y, clf=nn, legend=2)\n",
    "plt.title('epochs = 250')\n",
    "plt.show()\n",
    "print('Accuracy(epochs = 250): %.2f%%' % (100 * nn.score(X, y)))\n",
    "\n",
    "plt.plot(range(len(nn.cost_)), nn.cost_)\n",
    "plt.title('Gradient Descent training (minibatches=1)')\n",
    "plt.xlabel('Epochs')\n",
    "plt.ylabel('Cost')\n",
    "plt.show()\n",
    "\n",
    "nn.minibatches = len(y)\n",
    "nn = nn.fit(X, y)\n",
    "plt.plot(range(len(nn.cost_)), nn.cost_)\n",
    "plt.title('Stochastic Gradient Descent (minibatches=no. of training examples)')\n",
    "plt.ylabel('Cost')\n",
    "plt.xlabel('Epochs')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "601c7cd2",
   "metadata": {},
   "source": [
    "## 22.Decision Trees"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "b6361e8a",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Confusion Matrix:  [[15  1  0]\n",
      " [ 3  6  0]\n",
      " [ 0  0 13]]\n",
      "Accuracy :  89.47368421052632\n",
      "Report :                precision    recall  f1-score   support\n",
      "\n",
      "      Jasmin       0.83      0.94      0.88        16\n",
      "       Lotus       0.86      0.67      0.75         9\n",
      "        Rose       1.00      1.00      1.00        13\n",
      "\n",
      "    accuracy                           0.89        38\n",
      "   macro avg       0.90      0.87      0.88        38\n",
      "weighted avg       0.90      0.89      0.89        38\n",
      "\n",
      "Confusion Matrix:  [[15  1  0]\n",
      " [ 0  9  0]\n",
      " [ 0  0 13]]\n",
      "Accuracy :  97.36842105263158\n",
      "Report :                precision    recall  f1-score   support\n",
      "\n",
      "      Jasmin       1.00      0.94      0.97        16\n",
      "       Lotus       0.90      1.00      0.95         9\n",
      "        Rose       1.00      1.00      1.00        13\n",
      "\n",
      "    accuracy                           0.97        38\n",
      "   macro avg       0.97      0.98      0.97        38\n",
      "weighted avg       0.98      0.97      0.97        38\n",
      "\n"
     ]
    }
   ],
   "source": [
    "from sklearn.datasets import load_iris\n",
    "import pandas as pd\n",
    "\n",
    "from sklearn.tree import DecisionTreeClassifier,export_graphviz\n",
    "from sklearn.metrics import confusion_matrix,accuracy_score,classification_report\n",
    "\n",
    "from sklearn.model_selection import train_test_split \n",
    "from sklearn.ensemble import RandomForestClassifier\n",
    "from IPython.display import Image\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "\n",
    "df = pd.read_csv('flower.csv')\n",
    "X = df[list(df.columns)[:-1]]\n",
    "y = df['Flower']\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y, random_state = 0)  \n",
    "\n",
    "# entropy for information gain\n",
    "a = DecisionTreeClassifier(criterion = \"gini\", random_state = 100,max_depth=4, min_samples_leaf=5)  \n",
    "\n",
    "a.fit(X_train, y_train) \n",
    "y_pred = a.predict(X_test) \n",
    "y_train.to_csv('SSS.csv')\n",
    "print(\"Confusion Matrix: \", confusion_matrix(y_test, y_pred)) \n",
    "print (\"Accuracy : \", accuracy_score(y_test,y_pred)*100) \n",
    "print(\"Report : \", classification_report(y_test, y_pred)) \n",
    "\n",
    "b = RandomForestClassifier(max_depth = None, n_estimators=100)\n",
    "b.fit(X_train,y_train)\n",
    "y_pred = b.predict(X_test) \n",
    "print(\"Confusion Matrix: \", confusion_matrix(y_test, y_pred)) \n",
    "print (\"Accuracy : \", accuracy_score(y_test,y_pred)*100) \n",
    "print(\"Report : \", classification_report(y_test, y_pred)) "
   ]
  }
 ],
 "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"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}