AIO-Programs.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Program - 1\n",
    "#### Question: Implement and demonstrate the FIND-S algorithm for finding the most specific hypothesis based on a given set of training data samples. Read the training data from a .CSV file."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[['Sunny', 'Warm', 'Normal', 'Strong', 'Warm', 'Same', 'Yes'], ['Sunny', 'Warm', 'High', 'Strong', 'Warm', 'Same', 'Yes'], ['Rainy', 'Cold', 'High', 'Strong', 'Warm', 'Same', 'No'], ['Sunny', 'Warm', 'High', 'Strong', 'Cool', 'Change', 'Yes']] \n",
      "\n",
      "['Sunny', 'Warm', '?', 'Strong', '?', '?']\n"
     ]
    }
   ],
   "source": [
    "import csv\n",
    "with open(\"Ex01.csv\",\"r\") as f:\n",
    "    reader=csv.reader(f)\n",
    "    data=list(reader)\n",
    "print(data,\"\\n\")\n",
    "h=['0','0','0','0','0','0']\n",
    "for row in data:\n",
    "    if row[-1]==\"Yes\":\n",
    "        j=0\n",
    "        for cell_val in row:\n",
    "            if cell_val != \"Yes\":\n",
    "                if cell_val != h[j] and h[j]=='0':\n",
    "                    h[j]=cell_val\n",
    "                elif cell_val != h[j] and h[j]!='0':\n",
    "                    h[j]=\"?\"\n",
    "            j=j 1\n",
    "print(h)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Program - 2\n",
    "#### Question: For a given set of training data examples stored in a .CSV file, implement and Demonstrate the Candidate-Elimination algorithm to output a description of the set of all hypotheses consistent with the training examples."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['Sunny' 'Warm' 'High' 'Strong' 'Warm' 'Same']\n",
      "[['?', '?', '?', '?', '?', '?'], ['?', '?', '?', '?', '?', '?'], ['?', '?', '?', '?', '?', '?'], ['?', '?', '?', '?', '?', '?'], ['?', '?', '?', '?', '?', '?'], ['?', '?', '?', '?', '?', '?']]\n",
      "1\n",
      "['Sunny' 'Warm' 'High' 'Strong' 'Warm' 'Same']\n",
      "[['?', '?', '?', '?', '?', '?'], ['?', '?', '?', '?', '?', '?'], ['?', '?', '?', '?', '?', '?'], ['?', '?', '?', '?', '?', '?'], ['?', '?', '?', '?', '?', '?'], ['?', '?', '?', '?', '?', '?']]\n",
      "2\n",
      "['Sunny' 'Warm' 'High' 'Strong' 'Warm' 'Same']\n",
      "[['Sunny', '?', '?', '?', '?', '?'], ['?', 'Warm', '?', '?', '?', '?'], ['?', '?', '?', '?', '?', '?'], ['?', '?', '?', '?', '?', '?'], ['?', '?', '?', '?', '?', '?'], ['?', '?', '?', '?', '?', '?']]\n",
      "3\n",
      "['Sunny' 'Warm' 'High' 'Strong' '?' '?']\n",
      "[['Sunny', '?', '?', '?', '?', '?'], ['?', 'Warm', '?', '?', '?', '?'], ['?', '?', '?', '?', '?', '?'], ['?', '?', '?', '?', '?', '?'], ['?', '?', '?', '?', '?', '?'], ['?', '?', '?', '?', '?', '?']]\n",
      "['Sunny' 'Warm' 'High' 'Strong' '?' '?']\n",
      "[['Sunny', '?', '?', '?', '?', '?'], ['?', 'Warm', '?', '?', '?', '?']]\n"
     ]
    }
   ],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "data1=pd.read_csv('Ex02.csv')\n",
    "concepts=np.array(data1.iloc[:,0:-1])\n",
    "target=np.array(data1.iloc[:,-1])\n",
    "def learn(concepts, target):\n",
    "    s_h=concepts[0].copy()\n",
    "    print(s_h)\n",
    "    g_h=[[\"?\" for i in range(len(s_h))]\n",
    "        for i in range(len(s_h))]\n",
    "    print(g_h)\n",
    "    for i, h in enumerate(concepts):\n",
    "        if target[i]==\"Yes\":\n",
    "            for x in range(len(s_h)):\n",
    "                if h[x]!=s_h[x]:\n",
    "                    s_h[x]='?'\n",
    "                    g_h[x][x]='?'\n",
    "        if target[i]==\"No\":\n",
    "            for x in range(len(s_h)):\n",
    "                if h[x]!=s_h[x]:\n",
    "                    g_h[x][x]=s_h[x]\n",
    "                else:\n",
    "                    g_h[x][x]='?'\n",
    "        print(i 1)\n",
    "        print(s_h)\n",
    "        print(g_h)\n",
    "    indices=[i for i, val in enumerate(g_h) if val == ['?', '?', '?', '?', '?', '?']]\n",
    "    for i in indices:\n",
    "        g_h.remove(['?', '?', '?', '?', '?', '?'])\n",
    "    return s_h,g_h\n",
    "s_f, g_f = learn(concepts, target)\n",
    "print(s_f)\n",
    "print(g_f)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Program - 6\n",
    "#### Question: Assuming a set of documents that need to be classified, use the naïve Bayesian Classifier model to perform this task. Built-in Java classes/API can be used to write the program. Calculate the accuracy, precision, and recall for your data set."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "5\n",
      "0.7738980350504514\n",
      "0.7738980350504514\n",
      "[0.80193237 0.81028939 0.81904762 0.67180617 0.85632184 0.88955224\n",
      " 0.93127148 0.84651163 0.93686869 0.92248062 0.89170507 0.59379845\n",
      " 0.83629893 0.92113565 0.84172662 0.43896976 0.64339623 0.92972973\n",
      " 0.95555556 0.97222222]\n",
      "[0.52037618 0.64781491 0.65482234 0.77806122 0.77402597 0.75443038\n",
      " 0.69487179 0.91919192 0.9321608  0.89924433 0.96992481 0.96717172\n",
      " 0.59796438 0.73737374 0.89086294 0.98492462 0.93681319 0.91489362\n",
      " 0.41612903 0.13944223]\n",
      "                          precision    recall  f1-score   support\n",
      "\n",
      "             alt.atheism       0.80      0.52      0.63       319\n",
      "           comp.graphics       0.81      0.65      0.72       389\n",
      " comp.os.ms-windows.misc       0.82      0.65      0.73       394\n",
      "comp.sys.ibm.pc.hardware       0.67      0.78      0.72       392\n",
      "   comp.sys.mac.hardware       0.86      0.77      0.81       385\n",
      "          comp.windows.x       0.89      0.75      0.82       395\n",
      "            misc.forsale       0.93      0.69      0.80       390\n",
      "               rec.autos       0.85      0.92      0.88       396\n",
      "         rec.motorcycles       0.94      0.93      0.93       398\n",
      "      rec.sport.baseball       0.92      0.90      0.91       397\n",
      "        rec.sport.hockey       0.89      0.97      0.93       399\n",
      "               sci.crypt       0.59      0.97      0.74       396\n",
      "         sci.electronics       0.84      0.60      0.70       393\n",
      "                 sci.med       0.92      0.74      0.82       396\n",
      "               sci.space       0.84      0.89      0.87       394\n",
      "  soc.religion.christian       0.44      0.98      0.61       398\n",
      "      talk.politics.guns       0.64      0.94      0.76       364\n",
      "   talk.politics.mideast       0.93      0.91      0.92       376\n",
      "      talk.politics.misc       0.96      0.42      0.58       310\n",
      "      talk.religion.misc       0.97      0.14      0.24       251\n",
      "\n",
      "                accuracy                           0.77      7532\n",
      "               macro avg       0.83      0.76      0.76      7532\n",
      "            weighted avg       0.82      0.77      0.77      7532\n",
      "\n"
     ]
    }
   ],
   "source": [
    "from sklearn.datasets import fetch_20newsgroups\n",
    "from sklearn.feature_extraction.text import CountVectorizer\n",
    "from sklearn.feature_extraction.text import TfidfTransformer\n",
    "from sklearn.pipeline import Pipeline\n",
    "from sklearn.naive_bayes import MultinomialNB\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "from sklearn import metrics\n",
    "twenty_train=fetch_20newsgroups(subset='train', shuffle=True)\n",
    "print(len(twenty_train))\n",
    "text_clf=Pipeline([('vect',CountVectorizer()),('tfidf',TfidfTransformer()),('clf',MultinomialNB())])\n",
    "text_clf=text_clf.fit(twenty_train.data,twenty_train.target)\n",
    "twenty_test=fetch_20newsgroups(subset='test', shuffle=True)\n",
    "predicted=text_clf.predict(twenty_test.data)\n",
    "accuracy=np.mean(predicted == twenty_test.target)\n",
    "print(accuracy)\n",
    "print(metrics.accuracy_score(twenty_test.target,predicted))\n",
    "print(metrics.precision_score(twenty_test.target,predicted,average=None))\n",
    "print(metrics.recall_score(twenty_test.target,predicted,average=None))\n",
    "print(metrics.classification_report(twenty_test.target,predicted,target_names=twenty_test.target_names))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Program - 8B\n",
    "#### Question: Apply EM algorithm to cluster a set of data stored in a .CSV file. Use the same data set for clustering using K-Means algorithm. Compare the results of these two algorithms and comment on the quality of clustering. You can add Java/Python ML library classes/API in the program.\n",
    "##### K-Means Algorithm"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[5 9 8 4 7 2 3 1 6 4 6]\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXcAAAEICAYAAACktLTqAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1 /AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy WH4yJAAARH0lEQVR4nO3df2ycd33A8fcHx4Ab2pkKl5E0pTAhF9RuCzMk0I0hCjKoHfUfTGunsKpjytjG CFkqKEabCqCyQjRPxBT1PJLKalQ8UzHEKYqVIiNZnNqNAOZRYEuiROoO3Bb2AGp 9kfd06c0DS5H85jf /9kir7vr6759NT8s75ueeei8xEklSWp1Q9gCSp84y7JBXIuEtSgYy7JBXIuEtSgYy7JBXIuEtSgYy71pWIeCAiahHxaEQsRsS/R8SbI K0f5Yj4uKIyIjYsMoznpXtSE/GuGs9 qPMPBd4LvAh4N3ArdWOJK0txl3rVmY nJl3An8CXBcRl0bElRExExGPRMTBiHj/ipt8vfF1MSJ FhEvi4jfioivRsT/RsRDEXFbRPQv3yAi3h0R843fFOYi4orG lMi4oaI H7jtp LiPNPtZ1VfiikX2Pcte5l5n8Ah4A/AH4O/BnQD1wJ/FVEjDSu orG1/7MfEZmfhMI4IPAJuCFwBbg/QARMQi8BXhJ4zeFYeCBxn28FRgB/rBx258CH3uS7UhnlXFXKQ4D52fmPZk5m5mPZ Z/AXuoB/gJZeb9mXlXZv4yMxeAj6y4/hLwNOBFEdGbmQ9k5vcbP/tL4L2ZeSgzf0n9H4Q3uJ9da4VxVyk2Az JiG0R8bWIWIiIh4E3A8861Y0i4oKIuL2x6 URYPfy9TPzfuDt1MP9YON6mxo3fS7wz40XdReB/dT/MXj2av0PSs0w7lr3IuIl1OP DeCzwJ3Alsz8DeCfqO96AXiiU6B sLH 25l5HrBjxfXJzM9m5u9Tj3kC/9j40UHgdZnZv K/p2fm/Cm2I51Vxl3rVkScFxFXAbcDuzNzFjgX Elm/iIiXgr86YqbLACPA89fsXYu8DPqL35uBkZX3P9gRLwqIp4G/AKoUX92DvV/ND4QEc9tXHcgIq5 ku1IZ5Vx13r0LxHxKPVnz  lvp/8 sbP/hr4h8bP/w743PKNMvP/gA8A/9bYnbId HvgxcDDwL8CEyu28zTqh1o BPwIuAB4T NnN1P/DeErjW3dC2x7ku1IZ1X4YR2SVB6fuUtSgc7kLdufiIgHI LbK9bOj4i7IuJ7ja/PXN0xJUnNOJNn7p8CXnvS2g3A3Zn5AuDuxmVJ0hpxRvvcI Ji4IuZeWnj8hzwysw8EhHPAe7JzMHVHFSSdOZafTfdszPzCEAj8Bec6ooRsRPYCbBx48bfu SSS1rcpCR1p3379j2UmQPN3GbV3yqdmbuAXQBDQ0M5PT292puUpKJExP80e5tWj5b5cWN3DI2vD7Z4P5KkVdBq3O8Ermt8fx3whc6MI0nqhDM5FHIP8E1gMCIORcSbqL9r7zUR8T3gNY3LkqQ14rT73DPz2lP86IoOzyJJ6hDfoSpJBTLuklQg4y5JBTLuklQg4y5JBTLuklQg4y5JBTLuklQg4y5JBTLuklQg4y5JBTLuklQg4y5JBVr1T2JaKyZn5hmfmuPwYo1N/X2MDg8ysnVz1WNJ0qroirhPzswzNjFL7egSAPOLNcYmZgEMvKQidcVumfGpuWNhX1Y7usT41FxFE0nS6uqKuB9erDW1LknrXVfEfVN/X1PrkrTedUXcR4cH6evtOWGtr7eH0eHBiiaSpNXVFS oLr9o6tEykrpFV8Qd6oE35pK6RVfslpGkbmPcJalAxl2SCmTcJalAxl2SCmTcJalAxl2SCmTcJalAxl2SCmTcJalAxl2SCmTcJalAxl2SCmTcJalAbZ3yNyLeAfwFkMAscH1m/qITg5Vqcmbe88pLWnUtP3OPiM3AW4GhzLwU6AGu6dRgJZqcmWdsYpb5xRoJzC/WGJuYZXJmvurRJBWm3d0yG4C iNgAnAMcbn kco1PzVE7unTCWu3oEuNTcxVNJKlULcc9M eBDwMHgCPAw5n5lZOvFxE7I2I6IqYXFhZan7QAhxdrTa1LUqva2S3zTOBq4HnAJmBjROw4 XqZuSszhzJzaGBgoPVJC7Cpv6 pdUlqVTu7ZV4N/DAzFzLzKDABvLwzY5VpdHiQvt6eE9b6ensYHR6saCJJpWrnaJkDwPaIOAeoAVcA0x2ZqlDLR8V4tIyk1dZy3DNzb0TcAdwHPAbMALs6NVipRrZuNuaSVl1bx7ln5vuA93VoFklSh/gOVUkqkHGXpAIZd0kqkHGXpAIZd0kqkHGXpAIZd0kqkHGXpAIZd0kqkHGXpAIZd0kqkHGXpAIZd0kqUFtnhdT6NDkz7znlpcIZ9y4zOTPP2MTssQ/qnl sMTYxC2DgpYK4W6bLjE/NHQv7strRJcan5iqaSNJqMO5d5vBiral1SeuTce8ym/r7mlqXtD4Z9y4zOjxIX2/PCWt9vT2MDg9WNJGk1eALql1m UVTj5aRymbcu9DI1s3GXCqcu2UkqUDGXZIKZNwlqUDGXZIKZNwlqUDGXZIKZNwlqUDGXZIKZNwlqUDGXZIKZNwlqUDGXZIKZNwlqUBtnRUyIvqBW4BLgQT PDO/2YnBVL4bJ2fZs/cgS5n0RHDtti3cNHJZ1WNJRWj3lL83A1/OzDdExFOBczowk7rAjZOz7L73wLHLS5nHLht4qX0t75aJiPOAVwC3AmTmrzJzsVODqWx79h5sal1Sc9rZ5/58YAH4ZETMRMQtEbHx5CtFxM6ImI6I6YWFhTY2p5IsZTa1Lqk57cR9A/Bi4OOZuRX4OXDDyVfKzF2ZOZSZQwMDA21sTiXpiWhqXVJz2on7IeBQZu5tXL6Deuyl07p225am1iU1p W4Z aPgIMRMdhYugL4bkemUvFuGrmMHdsvOvZMvSeCHdsv8sVUqUMi29jHGRG/S/1QyKcCPwCuz8yfnur6Q0NDOT093fL2JKkbRcS zBxq5jZtHQqZmd8CmtqgJGn1 Q5VSSqQcZekAhl3SSqQcZekAhl3SSqQcZekAhl3SSqQcZekAhl3SSqQcZekAhl3SSqQcZekAhl3SSpQux QLa1rkzPzjE/NcXixxqb PkaHBxnZurnqsaS2GXd1rcmZecYmZqkdXQJgfrHG2MQsgIHXuuduGXWt8am5Y2FfVju6xPjUXEUTSZ1j3NW1Di/WmlqX1hPjrq61qb vqXVpPTHu6lqjw4P09facsNbX28Po8OApbiGtH76gqq61/KKpR8uoRMZdXW1k62ZjriK5W0aSCmTcJalAxl2SCmTcJalAxl2SCmTcJalAxl2SCmTcJalAxl2SCmTcJalAxl2SCmTcJalAxl2SCmTcJalAbZ/yNyJ6gGlgPjOvan8kqfvcODnLnr0HWcqkJ4Jrt23hppHLqh5L61gnnrm/DdjfgfuRutKNk7PsvvcAS5kALGWy 94D3Dg5W/FkWs/aintEXAhcCdzSmXGk7rNn78Gm1qUz0e4z948C7wIeP9UVImJnRExHxPTCwkKbm5PKs/yM/UzXpTPRctwj4irgwczc92TXy8xdmTmUmUMDAwOtbk4qVk9EU vSmWjnmfvlwOsj4gHgduBVEbG7I1NJXeTabVuaWpfORMtxz8yxzLwwMy8GrgG mpk7OjaZ1CVuGrmMHdsvOvZMvSeCHdsv8mgZtaXtQyElte mkcuMuTqqI3HPzHuAezpxX5Kk9vkOVUkqkHGXpAIZd0kqkHGXpAIZd0kqkHGXpAIZd0kqkHGXpAIZd0kqkHGXpAIZd0kqkHGXpAIZd0kqkKf8lQTA5Mw841NzHF6ssam/j9HhQUa2bq56LLXIuEticmaesYlZakeXAJhfrDE2MQtg4Ncpd8tIYnxq7ljYl9WOLjE NVfRRGqXcZfE4cVaU ta 4y7JDb19zW1rrXPuEtidHiQvt6eE9b6ensYHR6saCK1yxdUJR170dSjZcph3CUB9cAb83K4W0aSCmTcJalAxl2SCmTcJalAxl2SCmTcJalAxl2SCmTcJalAxl2SCmTcJalAxl2SCmTcJalAxl2SCtTyWSEjYgvwGeA3gceBXZl5c6cGk9Sd/KDuzmjnlL PAe/MzPsi4lxgX0TclZnf7dBskrqMH9TdOS3vlsnMI5l5X P7R4H9gI  pJb5Qd2d05F97hFxMbAV2PsEP9sZEdMRMb2wsNCJzUkqlB/U3Tltxz0ingF8Hnh7Zj5y8s8zc1dmDmXm0MDAQLubk1QwP6i7c9qKe0T0Ug/7bZk50ZmRJHUrP6i7c9o5WiaAW4H9mfmRzo0kqVv5Qd2d087RMpcDbwRmI JbjbX3ZOaX2h9LUrfyg7o7o W4Z Y3gOjgLJKkDvEdqpJUIOMuSQUy7pJUIOMuSQUy7pJUIOMuSQUy7pJUIOMuSQUy7pJUIOMuSQUy7pJUIOMuSQUy7pJUoHZO StJRZqcmV/355Q37pK0wuTMPGMTs8c qHt scbYxCzAugq8u2UkaYXxqbljYV9WO7rE NRcRRO1xrhL0gqHF2tNra9Vxl2SVtjU39fU lpl3CVphdHhQfp6e05Y6 vtYXR4sKKJWuMLqpK0wvKLph4tI0mFGdm6ed3F/GTulpGkAhl3SSqQcZekAhl3SSqQcZekAhl3SSqQcZekAhl3SSqQcZekAhl3SSqQcZekAhl3SSqQcZekAhl3SSpQW3GPiNdGxFxE3B8RN3RqKElSe1qOe0T0AB8DXge8CLg2Il7UqcEkSa1r55n7S4H7M/MHmfkr4Hbg6s6MJUlqRzufxLQZOLji8iFg28lXioidwM7GxV9GxLfb2GZJngU8VPUQa4SPxXE Fsf5WBzX9Ae4thP3eIK1/LWFzF3ALoCImM7MoTa2WQwfi N8LI7zsTjOx K4iJhu9jbt7JY5BGxZcflC4HAb9ydJ6pB24v6fwAsi4nkR8VTgGuDOzowlSWpHy7tlMvOxiHgLMAX0AJ/IzO c5ma7Wt1egXwsjvOxOM7H4jgfi Oafiwi89d2k0uS1jnfoSpJBTLuklSgsxJ3T1NQFxFbIuJrEbE/Ir4TEW reqaqRURPRMxExBernqVKEdEfEXdExH83/ny8rOqZqhIR72j8/fh2ROyJiKdXPdPZFBGfiIgHV74nKCLOj4i7IuJ7ja/PPN39rHrcPU3BCR4D3pmZLwS2A3/TxY/FsrcB 6seYg24GfhyZl4C/A5d phExGbgrcBQZl5K/WCNa6qd6qz7FPDak9ZuAO7OzBcAdzcuP6mz8czd0xQ0ZOaRzLyv8f2j1P8Cb652qupExIXAlcAtVc9SpYg4D3gFcCtAZv4qMxernapSG4C iNgAnEOXvX8mM78O/OSk5auBTze /zQwcrr7ORtxf6LTFHRt0JZFxMXAVmBvtZNU6qPAu4DHqx6kYs8HFoBPNnZR3RIRG6seqgqZOQ98GDgAHAEezsyvVDvVmvDszDwC9SeJwAWnu8HZiPsZnaagm0TEM4DPA2/PzEeqnqcKEXEV8GBm7qt6ljVgA/Bi4OOZuRX4OWfwa3eJGvuSrwaeB2wCNkbEjmqnWp/ORtw9TcEKEdFLPey3ZeZE1fNU6HLg9RHxAPVdda KiN3VjlSZQ8ChzFz Le4O6rHvRq8GfpiZC5l5FJgAXl7xTGvBjyPiOQCNrw e7gZnI 6epqAhIoL6ftX9mfmRquepUmaOZeaFmXkx9T8TX83MrnyGlpk/Ag5GxPKZ/64AvlvhSFU6AGyPiHMaf1 uoEtfXD7JncB1je vA75wuhu0c1bIM9LiaQpKdTnwRmA2Ir7VWHtPZn6pwpm0NvwtcFvjCdAPgOsrnqcSmbk3Iu4A7qN dNkMXXYagojYA7wSeFZEHALeB3wI FxEvIn6P4B/fNr78fQDklQe36EqSQUy7pJUIOMuSQUy7pJUIOMuSQUy7pJUIOMuSQX6f2CD oaH1G6TAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXcAAAD8CAYAAACMwORRAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1 /AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy WH4yJAAANm0lEQVR4nO3dXYhcdxnH8d9vZ1100ootrt2YbbIVSmspmOiiTQpRjEJFsV4oVLalDcJ64UsVQaoh9CrghYi9EHGo9QWHisSCpYgvVFsRJLjpBmyMUqlmG02aEfEFF6xtHi9mtrvZJO7OnLM5O8/5fiDMzj87Ow9D8s1/T2bPcUQIAJDLSNUDAADKR9wBICHiDgAJEXcASIi4A0BCxB0AEloz7rYfsn3W9tMr1q62/VPbz/Rur9rYMQEA/VjPzv2bkm5btXafpMcj4npJj/fuAwA2Ca/nh5hsT0l6LCJu7t3/vaR3RMRp21slPRERN2zkoACA9Rsd8HHXRMRpSeoF/nWX kTbs5JmJWnLli1vufHGGwd8SgCop6NHj/41Isb7ecygcV 3iGhJaknS9PR0zM3NbfRTAkAqtk/2 5hB3y3zfO9wjHq3Zwf8OgCADTBo3B VdHfv47sl/aCccQAAZVjPWyEflvQrSTfYPmX7I5K IOndtp R9O7efQDAJrHmMfeI PAlfmtfybMAAErCT6gCQELEHQASIu4AkBBxB4CEiDsAJETcASAh4g4ACRF3AEiIuANAQsQdABIi7gCQEHEHgISIOwAkVJu4t9vS1JQ0MtK9bbernggANs6GX2ZvM2i3pdlZaXGxe//kye59SZqZqW4uANgotdi5HziwHPYli4vddQDIqBZxX1jobx0Ahl0t4r59e3/rADDsahH3Q4ekZvP8tWazuw4AGdUi7jMzUqsl7dgh2d3bVov/TAWQVy3eLSN1Q07MAdRFLXbuAFA3xB0AEiLuAJAQcQeAhIg7ACRE3AEgIeIOAAkRdwBIiLgDQELEHQASIu4AkBBxB4CEiDsAJETcASChQnG3/Wnbx20/bfth268sa7Cs2m1pakoaGenetttVTwQgo4HjbnubpE9Kmo6ImyU1JN1R1mAZtdvS7Kx08qQU0b2dnSXwAMpX9LDMqKRX2R6V1JT0l Ij5XXggLS4eP7a4mJ3HQDKNHDcI LPkr4oaUHSaUn/iIifrP4827O252zPdTqdwSdNYGGhv3UAGFSRwzJXSbpd0nWSXi9pi 07V39eRLQiYjoipsfHxwefNIHt2/tbB4BBFTks8y5Jf4yITkT8V9IjkvaUM1ZOhw5Jzeb5a81mdx0AylQk7guSbrHdtG1J ySdKGesnGZmpFZL2rFDsru3rRYX7gZQvtFBHxgRR2wflvSUpBclzUtqlTVYVjMzxBzAxhs47pIUEfdLur kWQAAJeEnVAEgIeIOAAkRdwBIiLgDQELEHQASIu4AkBBxB4CEiDsAJETcASAh4g4ACRF3AEiIuANAQsQdABIi7jXUbktTU9LISPeWC3QD RQ65S GT7stzc4uX6j75MnufYnzzAOZsHOvmQMHlsO ZHGxuw4gD JeMwsL/a0DGE7EvWa2b 9vHcBwIu41c iQ1Gyev9ZsdtcB5EHca2ZmRmq1pB07JLt722rxn6lANrxbpoZmZog5kB07dwBIiLgDQELEHQASIu4AkBBxB4CEiDsAJETcASAh4g4ACRF3AEiIuANAQsQdABIi7gCQEHEHgIQKxd32a2wftv072yds7y5rMCTXaHTPObz6V6NR9WRACkVP fuApB9FxAdtj0lqrvUAQJJ07lx/6wD6MnDcbb9a0l5J90hSRLwg6YVyxgIAFFHksMwbJHUkfcP2vO0HbW9Z/Um2Z23P2Z7rdDoFng4AsF5F4j4q6c2SvhoRuyT9W9J9qz8pIloRMR0R0 Pj4wWeDgCwXkXifkrSqYg40rt/WN3YAwAqNnDcI KMpOds39Bb2ifpt6VMhfxGLvFH71LrAPpS9N0yn5DU7r1T5llJ 4uPhFp46aWqJwBSKxT3iDgmabqkWQAAJeF7YABIiLgDQELEHQASIu4AkBBxB4CEiDsAJETcASAh4g4ACRF3AEiIuANAQsQdABIi7gCQEHEHgISIO prYkKyL/w1MVH1ZEBhxB319fzz/a0DQ4S4A0BCxB0AEiLuAJAQcQeAhIg76uuaa/pbB4ZIoQtkA0PtzJmqJwA2DDt3AEiIuANAQsQdABIi7gCQEHEHgISIOwAkRNwBICHiDgAJEXcASIi4A0BCxB0AEiLuAJAQcQeAhIg7ACRUOO62G7bnbT9WxkBA7TQakn3hr0aj6skwxMrYud8r6UQJXweop3Pn lsH1qFQ3G1PSnqvpAfLGQcAUIaiO/cvS/qspEtuMWzP2p6zPdfpdAo HQBgPQaOu 33STobEUf/3 dFRCsipiNienx8fNCnAwD0ocjO/VZJ77f9J0nflfRO298pZSoAQCEDxz0iPhcRkxExJekOST LiDtLmwyoi5FL/DW81DqwDqNVDwDU3ksvVT0BEiol7hHxhKQnyvhaAIDi L4PABIi7gCQEHEHgISIOwAkRNwBICHiDgAJEXcASIi4A0BCxB0AEiLuAJAQcQeAhIg7ACRE3AEgIU75C0C7vrZLx84cu2B958ROzX90voKJUBQ7dwDaPblbY42x89bGGmPaM7mnoolQFHEHoIN7D2rE5 eg4YYOvv1gRROhKOIOQFuv3Kr9O/e/vHsfa4xp/879mrhiouLJMCjiDkDS bt3du3Dj7gDkLS8ex/xCLv2BHi3DICXHdx7UMc7x9m1J0DcAbxs65Vb9eQ9T1Y9BkrAYRkASIi4A0BCxB0AEiLuAJAQcQeAhIg7ACRE3AEgIeIOAAkRdwBIiLgDQELEHQASIu4AkBBxB4CEBj4rpO1rJX1b0oSkc5JaEfFAWYMBqB8u1F2eIjv3FyV9JiLeKOkWSR zfVM5YwGoIy7UXZ6B4x4RpyPiqd7H/5J0QtK2sgYDUD9cqLs8pRxztz0laZekIxf5vVnbc7bnOp1OGU8HICku1F2ewnG3fYWk70v6VET8c/XvR0QrIqYjYnp8fLzo0wFIjgt1l6NQ3G2/Qt2wtyPikXJGAlBnXKi7HEXeLWNJX5d0IiK VN5IAOqOC3UXV QC2bdKukvSb2wvvXfp8xHxw JjAagzLtRd3MBxj4hfSnKJswAASsJPqAJAQsQdABIi7gCQEHEHgISIOwAkRNwBICHiDgAJEXcASIi4A0BCxB0AEiLuAJAQcQeAhIg7ACRU5JS/AJDOrq/t0rEzxy5Y3zmxU/Mfna9gosGwcweAFXZP7n75Gq5Lxhpj2jO5p6KJBkPcAWCFlddwXTKM13Il7gCwwtI1XJd272ONsaG8litxB4BVVu7eh3HXLhF3ALjA0u59xCNDuWuXeLcMAFzUwb0HdbxzfCh37RJxB4CL2nrlVj15z5NVjzEwDssAQELEHQASIu4AkBBxB4CEiDsAJETcASAh4g4ACRF3AEiIuANAQsQdABIi7gCQEHEHgISIOwAkRNwBIKFCcbd9m 3f2/6D7fvKGgoAUMzAcbfdkPQVSe RdJOkD9u qazBAACDK7Jzf6ukP0TEsxHxgqTvSrq9nLEAAEUUuRLTNknPrbh/StLbVn S7VlJs727/7H9dIHnzOS1kv5a9RCbBK/FMl6LZbwWy27o9wFF4u6LrMUFCxEtSS1Jsj0XEdMFnjMNXotlvBbLeC2W8Vossz3X72OKHJY5JenaFfcnJf2lwNcDAJSkSNx/Lel629fZHpN0h6RHyxkLAFDEwIdlIuJF2x X9GNJDUkPRcTxNR7WGvT5EuK1WMZrsYzXYhmvxbK XwtHXHCYHAAw5PgJVQBIiLgDQEKXJe6cpqDL9rW2f277hO3jtu teqaq2W7Ynrf9WNWzVMn2a2wftv273p P3VXPVBXbn 79/Xja9sO2X1n1TJeT7Ydsn135M0G2r7b9U9vP9G6vWuvrbHjcOU3BeV6U9JmIeKOkWyR9rMavxZJ7JZ2oeohN4AFJP4qIGyW9STV9TWxvk/RJSdMRcbO6b9a4o9qpLrtvSrpt1dp9kh6PiOslPd67/39djp07pynoiYjTEfFU7 N/qfsXeFu1U1XH9qSk90p6sOpZqmT71ZL2Svq6JEXECxHx92qnqtSopFfZHpXUVM1 fiYifiHpb6uWb5f0rd7H35L0gbW zuWI 8VOU1DboC2xPSVpl6Qj1U5SqS9L qykc1UPUrE3SOpI kbvENWDtrdUPVQVIuLPkr4oaUHSaUn/iIifVDvVpnBNRJyWuptESa9b6wGXI 7rOk1Bndi QtL3JX0qIv5Z9TxVsP0 SWcj4mjVs2wCo5LeLOmrEbFL0r 1jm 7M odS75d0nWSXi9pi 07q51qOF2OuHOaghVsv0LdsLcj4pGq56nQrZLeb/tP6h6qe6ft71Q7UmVOSToVEUvfxR1WN/Z19C5Jf4yITkT8V9IjkvZUPNNm8LztrZLUuz271gMuR9w5TUGPbat7XPVERHyp6nmqFBGfi4jJiJhS98/EzyKilju0iDgj6TnbS2f 2yfptxWOVKUFSbfYbvb vuxTTf9zeZVHJd3d /huST9Y6wFFzgq5LgOepiCrWyXdJek3to/11j4fET scCZsDp Q1O5tgJ6VtL/ieSoREUdsH5b0lLrvLptXzU5DYPthSe Q9FrbpyTdL kLkr5n yPq/gP4oTW/DqcfAIB8 AlVAEiIuANAQsQdABIi7gCQEHEHgISIOwAkRNwBIKH/Aaap/aOM23XrAAAAAElFTkSuQmCC\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",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import csv\n",
    "data=[]\n",
    "ydata=[]\n",
    "with open(\"Ex08.csv\") as tsv:\n",
    "    for line in csv.reader(tsv):\n",
    "        data=[int(i) for i in line]\n",
    "        ydata=[10-int(i) for i in line]\n",
    "x1=np.array(data)\n",
    "x2=np.array(ydata)\n",
    "print(x1)\n",
    "plt.plot()\n",
    "plt.xlim([0,10])\n",
    "plt.ylim([0,10])\n",
    "plt.title('Dataset')\n",
    "plt.scatter(x1,x2)\n",
    "plt.show()\n",
    "plt.plot()\n",
    "x=np.array(list(zip(x1,x2))).reshape(len(x1),2)\n",
    "colors=['b','g','r']\n",
    "markers=['o','v','s']\n",
    "K=3\n",
    "kmeans_model=KMeans(n_clusters=K).fit(x)\n",
    "plt.plot()\n",
    "for i, l in enumerate(kmeans_model.labels_):\n",
    "    plt.plot(x1[i],x2[i],color=colors[l],marker=markers[l],ls='None')\n",
    "    plt.xlim([0,10])\n",
    "    plt.ylim([0,10])\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Program - 9\n",
    "#### Question: Write a program to implement K-Nearest Neighbour algorithm to classify the iris data set. Print both correct and wrong predictions. Java/Python ML library classes can be used for this problem."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Y_Pred  Y_Test\n",
      "2 \t 2\n",
      "1 \t 1\n",
      "0 \t 0\n",
      "2 \t 2\n",
      "0 \t 0\n",
      "2 \t 2\n",
      "0 \t 0\n",
      "1 \t 1\n",
      "1 \t 1\n",
      "1 \t 1\n",
      "2 \t 2\n",
      "1 \t 1\n",
      "1 \t 1\n",
      "1 \t 1\n",
      "2 \t 1\n",
      "0 \t 0\n",
      "1 \t 1\n",
      "1 \t 1\n",
      "0 \t 0\n",
      "0 \t 0\n",
      "2 \t 2\n",
      "1 \t 1\n",
      "0 \t 0\n",
      "0 \t 0\n",
      "2 \t 2\n",
      "0 \t 0\n",
      "0 \t 0\n",
      "1 \t 1\n",
      "1 \t 1\n",
      "0 \t 0\n",
      "[[11  0  0]\n",
      " [ 0 12  1]\n",
      " [ 0  0  6]]\n",
      "96.67 %\n"
     ]
    }
   ],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "from sklearn.preprocessing import LabelEncoder\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.neighbors import KNeighborsClassifier\n",
    "from sklearn.metrics import confusion_matrix \n",
    "from sklearn.metrics import accuracy_score\n",
    "dataset=pd.read_csv('Ex09.csv')\n",
    "f_c=['sepal_length','sepal_width','petal_length','petal_width']\n",
    "x=dataset[f_c].values\n",
    "y=dataset['species'].values\n",
    "le=LabelEncoder()\n",
    "y=le.fit_transform(y)\n",
    "x_train,x_test,y_train,y_test=train_test_split(x,y,test_size=0.2, random_state=0)\n",
    "classifier=KNeighborsClassifier(n_neighbors=3)\n",
    "classifier.fit(x_train,y_train)\n",
    "y_pred=classifier.predict(x_test)\n",
    "print(\"Y_Pred  Y_Test\")\n",
    "for i in range(len(y_pred)):\n",
    "    print(y_pred[i],\"\\t\",y_test[i])\n",
    "cm=confusion_matrix(y_test,y_pred)\n",
    "print(cm)\n",
    "accuracy=accuracy_score(y_test,y_pred)*100\n",
    "print(round(accuracy,2),\"%\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Program - 10\n",
    "#### Question: Apply EM algorithm to cluster a set of data stored in a .CSV file. Use the same data set for clustering using K-Means algorithm. Compare the results of these two algorithms and comment on the quality of clustering. You can add Java/Python ML library classes/API in the program."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(80, 2)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "Text(0, 0.5, 'colB')"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXgAAAEGCAYAAABvtY4XAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1 /AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy WH4yJAAAgAElEQVR4nO3deXxU5bnA8d TYGRTVEitoiTQerVW3MilVdyxCIKi1luhcUMtXlzRtt4qbQXvh9q6FHDBikJdiFpF3JCqFOt2USsoshStxRIUrQa1WHYIz/3jnWCS2U6Sc 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 XV1dUOEYY4pUMQ9YB9lFsyswFOgJ7Al0EpEzm5 nqlNUtUpVq8rLM5Y0NsaYVivmAesgB1mPB/6hqnWqugWYCRwe4PWMMSZNMQ9YB5ngVwLfFZGOIiJAf2BZgNczxpg0xTxgHVipAlV9XURmAG8CW4G3gClBXc8YY7Kp7l1dFAm9uUBr0ajqtcC1QV7DGGNMZraS1RhjEsoSvDHGJJQleGOMSShL8MaYUBVrnZhCsA0/jDGhsY1NgmUteGNMaGxjk2BZgjfGhKaY68QUgiV4Y0xoirlOTCFYgjfGeOb3gGgx14kpBEvwxhhPgtg4o5jrxBRCYBt tIZt GFMdCV944y4CmXDD2NMstiAaPxYgjfGeGIDovFjCd4Y44kNiMaPJXhjjCc2IBo/NshqjDExFsogq4jsKyILG92 FJHRQV3PmCB5nf8dx8JZcYy5uST8DkEIcsu d4GDAUSkFFgFPBbU9YwJiteCWHEsnBXHmJtLwu8QlIJ00YjIAOBaVe2X6zzrojFR5HX dxzniccx5uaS8Du0RRTmwQ8DHsx0QERGish8EZlfV1dXoHCM8c7r/O84zhOPY8zNJeF3CErgCV5EyoCTgUcyHVfVKapapapV5eXlQYdjTIt5nf8dx3nicYy5uST8DkEpRAt EPCmqn5SgGsZ4zuv87/jOE88jjE3l4TfISiFSPDDydI9Y0wceJ3/Hcd54nGMubkk/A5BCXSQVUQ6Ah8AvVR1Tb7zbZDVGGNaJtcga6B7sqrqeqBrkNcwxhiTmZUqCIgtvDCmwLZsgfffhw0bwo4kMgJtwRcrW3hhTAFt2gSTJ8P110NdHZSUwLe BYcc8tXt4INh113DjrTgrBZNAIp94YUxBbFtG/zhD3DNNbBiRf7zKyuhb1 45BI48sigoyuYKCx0Kiq28MLkYt13Pvjzn12y/uEPvSV3cOc9/DAceyw8 WSQ0UWGJfgA2MILk00Q 5oWlSVLYMgQOO44WLCgdc9RXw/DhsGrr/obWwRZgg ALbww2YyZO2b72EyD9VvWM2bumJAiiolVq CCC Cgg Dpp9v fBs2wEknwbvvtv25IswSfABs4YXJxrrvWmjNGhgzBvbZB6ZOdf3u2QwaBG /7QZa58yBG26A4cNhv/1AJP38zz6DE06Ajz8OLv6Q2SCrMQVkA/Aebd4Md94J110Hq1fnPvfQQ10y798/ zlffAHf 17mbp2DDoKXXoKdd25bzCGxQdYEswG7eLHuuzxU4ZFHYP/94bLLcif3igqoqYE33sid3MFNkXz6aejZM/3Y22/Daae5/1QSxhJ8jNmAXfxY910OL78Mhx0GP/gBLF e/bxdd4Wbb4Z33nGzaEo8prHdd4dnn4Vu3dKPzZ0LI0bk7gKKIeuiiTH7um8SYdky NnP8k9d3HFH16q/ uq2LVr6y1/cVMn169OP/eQncOONrX/uEFgXTULZgJ2JtU2b4Je/hAMPzJ3cReCss9yMlxtuaPuK1L59XTdQaWn6sZtugokT2/b8EWIJPsZsvr2JrTfegD863H//F7ZuzX7e8ce7gdH77nN97n458US4667Mx664wq2QTQBL8DFmA3YmdjZudN0x3/0uLF2a/bwDD3T95XPmuFoyQRgxwv0Hk8nZZ7vVsjFnCT7GbMDOxMq8ea7o129 k30wc  94d574c03YcCA4GMaMwYuvDD9/s2b4ZRTYNGi4GMIkqoGdgN2AWYA7wDLgMNynd nTx81xi/TF03XigkVKmNFKyZU6PRF08MOyTejZo3S0nGlyli0dFypjpo1KuyQslu3TvWKK1RFVN1EyPRbu3aqv/iF6vr1gYSQ87Owdavq0KGZ49pzT9Xa2kBi8gswX7Pk1KDLBU8CnlHV01Obb3fM9wBj/JDkks0XPX0Rd8y/Y/vP9Vq//efJgyeHFVZmL74I55 fe9rjIYfAtGmudR AvJ F0lJ48EHX3z9vXtMHf/QRDBwIr7wCu 0WSHxBCmyapIjsDLyN267P00VsmqTxS5KnkLa7rh31Wp92f6mUsvWXOQYsC2ntWtfXfvvt2c8pK3OzaK66CnbYIbBQPH8WPvsMjjjCza9vrl8/Nx7QoUNgcbZWWNMkewF1wO9F5C0RuVtEOmUIbqSIzBeR XV1dQGGY6LM7xW5SZ5Cmim557q/4P70JzjggNzJ/T//0/WzjxkTaHKHFnwWunaFZ56BPfZIP/n//s8tqqqPyGvsUZAJvh1wKHCHqh4CrAN 1vwkVZ2iqlWqWlVeXh5gOCaqgliRm QppKWSYf52jvsLYuNGN7f8xBNdzZfa9BYz4BYr3XCD6wr59rcLElqLPgsVFfDHP2auS/P443Dppa53PiaCTPAfAh q6uupn2fgEr4xTQRRQjfJU0hH9hnZovsDo rms198Mey5pysx8Mc/Zj /Xz9X9 WnP4V2hdsttMWfhYMOgscey/zN4o474Fe/CiDKYASW4FX1n8AHIrJv6q7 wF Dup6JryC6U5I8hXTy4MmMqhq1vcVeKqWMqhpVuAHWjz92y/kPOMCtCp082VVrzKZDB7c69MUXYd99s58XkFZ9Fo47zi2uyuTnP4ff/z6YYH0WaC0aETkYuBsoA94HRqhq1k CDbIWpyQPiCbC1q2upf7cc27x0euvey/KdcwxcPfd8I1vBBpiYH77W/jxj9PvLy2Fp55yNehDFlotGlVdmOpfP1BVT8mV3E3xilp3ipVgBlaudEv5/ u/oLwcDj8cxo5129x5Se49erh67nPnxje5A1x5pbs1V18Pp5/u/uOLsMJ1hBmTRcNX5TFzx7ByzUp6dOnB P7jQ lOSfL8 ZzWrXNdKA2t9ExTBfPp0MElvXPPdS13r2V8o 7GG1231IMPNr1//XoYPNgNGH/zm HEloeVCzamkaLqLlq3DmbNgocecoOjmza17nmOPNIl9dNPj 2uSHlt2uRmCD3/fPqxXr1ckt9998LHRe4uGmvBG9NIkufPA2464zPPuKT 1FOZa6J70auXmxd zjmRbb36ascd3cyao45yM4Eae/9915J/4QXo3DmU8LJJyHeotgui39X6cuMnkfPnt2xxLfRzz3WtzFNPdeVwW5LcO3eGk0 G226D996Dv//dVWIshuTeYOedYfbszGWLFyxw32C2bCl8XDlYC55g l2Lti835sb3H9/kfYOYzp vr3cbST/0EDz6qFuG31J9 sAJJ7iqjocd5koLFLs993TfgPr1g88/b3rs2WfhggvgnnvcJiURYH3wBNPvWlR9uQlTs7gmEgO LbZtG7z2mkvqjzwC//xnyx7/ta 5aX8nnOAKb9nK8uzmzXMbfW/cmH7s6qsLuhgqVx 8JXigZFwJSvrrIAjbrm3dJrxBPGdsE0/Chf6 rFrlpiTee6 b3tgSu 0G3/8 DBsGRx deRs7k9kTT8Bpp2WeNnrbbW6FbwHYIGsePbr0yNjabku/q9/PaV0 0RTa 6LqStjeeivMnNmyIlg77eT64YcNcy31gIt9JdbQoW4V73//d/qxSy913Tmnnlr4uBqxQVaCWWjj93MGUa/FL8U8mFzw92XLFqipgaoqN6PjkUe8JfcOHVytmJkz4dNPXWt/0CBL7m114YXwi1 k368K1dVu1W IrAVPMAtt/H7OqE7fK/ZvFgV7X774AqZMcS32Vau8PaaszCXxYcNgyJDITeFLjHHj3MYgU6c2vX/DBjjpJDcu0qtXKKFZH3xMRHXQNqpxFUrgv//y5TBpktvxaN26/OeXlrpul2HD3J6iu zS9hhMflu3ui6b2bPTj 27rxuUDWhHqNBq0Rj/RK1eS4OofrMohJrFNazdvDbt/ja/L6quhsvQobDPPq7Vni 5f/3rcO218MEHbhrfuedaci kdu3c2oJDM1REf/dd959tphk3AbMEHxNRLX byIVBHjR0TX22oen88q4durb fVm/3nXD9O7tWuFPPpl/c4m fV2ffG2tKwaWaTciUxidO7vSD3vvnX7s5ZdhxAjvVTh9Yl00pk2a98GDa8FG4T fIPnaNVNb62Zj3HVX7rrqDUTc7Iwf/9gtQIrIohqTsmSJWwj15ZfpxwKYI29dNCYwUf1mEbQ2d02pupWmp5/uBuBuuCF/cu/UCS67zJUKePRRV8LXknv0HHCAm62Uadeq669339IKxFrwxrRCq1vw69e7laa33goLF3q7WI8ecMkl8KMfWb96nNx7rxsLac7nzUJCa8GLyAoRWSwiC0Wk4Jm7mOdnR1mm9yVu79X4/uMpK21am6WstCz74Oq778Lo0W7xy/nne0vuRx3lWurLl7t9TGOQ3C96 iLaXdcOGSe0u64dFz19UdghFUzaZ/jQdm7gu7n6ercmYeHCwD/3hZgHf6yqri7AdZoo9vnZUZXpfTnvifNQVbZs27L9vji8V82//aZ9G9661Q2UTp7sZsV4seOOrgzvZZfBwQf7FGlhXPT0Rdwx/47tP9dr/fafC7ZfbEiy5pvT7qR6xTmuNd/Y2rWsP E4xp2zgdpOG5s Bv8 90HvyboCqPKa4P3soin2 dlRle19ySTK71XOz9cZr7p9SO 80/uipD33dLVLfvSj2Bb5anddO o1fVVtqZSy9ZdbQ4iocHJ Hi76m uOybBZyKKvwZHnwZftmz2mBZ/7MGvRKPCciChwp6qmjS6IyEhgJECPHv5NrSvm dlR1pLXP8rvVabYqlbB6Edr4ac9XOvdi8MOg8svd0WrYl42IFNyz3V/kuTMN2VlrqvtiCNg6dImxw/8FB55GAZXw9bS3M/VGkHPoumnqocCg4CLReSo5ieo6pTUxtxV5T62XIp1fnbUteT1j/J71RBbyTY4ZRm8NA3euAuqF5M/uXfoAOed5zZsnjcPzjgj9skdXEu9JfcnSd58s8su8PTTbkFaMwPeh9/Ngobis35 7gNN8Kr6UerPT4HHgL5BXq xqK78LHaZ3pey0jJ2KGma4NryXhViwPb6o8YxamEZy26Dx/4AR3ppdO2zD0yY4Lptpk51BcMSZGSfkS26P0k85ZuKCpfkO3VKe/z5b8GYlwLIUaoayA3oBOzU6O/zgIG5HtOnTx/10/RF07ViQoXKWNGKCRU6fdF0X5/ftE6m98Wv92r6ounacXxHZSzbbx3Hd/TvvV /XvXWW1X33lvVzWbPfSspUT31VNU5c1Tr6/2JIcJGzRqlpeNKlbFo6bhSHTVrVNghFYznz/BTT7nPRYbPyyu/ u8WXxeYr1lyamCDrCLSC9dqB9fX/4Cq5vyvyebBm7YKbHB97Vr43e/gppvgk0/yn19eDiNHunKymZaum J2xx1wUYYppDvsAM89B8cc4/mpQhlkVdX3gYOCen5jMvF9cH3tWrj9drjxRm/7mvbuDVdcAcOHQ/v2 c83xWnUKPjHP9znqrEuXdxUWZ9YPXiTKL7tpLVu3VeJfbWHWb7HHw9XXeX tPIBxotf/xpWrHCbtoAbo5k9G775Td8uYQneJMr4/uMzFj/zPHD12WeuVsiECVBXl//8U06BMWMSN2BqCqCkBO67zw26i8Djj0O3br5ewhK8SZRW76S1ZInbWGP69Px1u0tK3NTGq692XTLGtFb79q4uTceOgXTpWYKPoJrFNb5uH5hk2V4rT69Xfb2btjZpUsZVhmlKS HMM12LfZ992h68MRDYTk9g5YIjp6GmRe2aWhTdXp/Cj7nccSvolU rX6svv4SJE E//sPtmpQvuZeUwNlnw7JlcM89ltxNbFiCj5gxc8c06T8GWL9lPWPmjtn c2sSdWuSYdT/Q/DyWjXx3nuuiFf37m6my/vv575ASYlrsS9b5opFWWI3MWNdNBGTb5pfa6tk5kqGmR4Xh2qcnqZErlsHc a44l zZ ffAg/cVLULLnA12Csr/QnWmBDkTfAiUgGsU9XVIvJd4Ahguao luehphXyTfNraaJu0NL54a29jicbN7o5wJ984vaobLyWr6zMLeVufOvc2d3fbPph89eqw2bo S847t9d4Wc/c/tgvvEGbNniLa5993Ut/LPPdtc0JuZyJngR QVwLqAi8hBwPPACMFhEjlbV0YFHGDFBD4Dmm bX2oU8LZ0f3qYFQ6puuuHy5a4bZPnypn/3WkK3sdLStMS/YGsJn9QJHTcpXTbBrtsnv6wGfuP9uQcOdBUdBwxw3TLGJES Fvxw4FtAR2Al8HVVXS8i7QCP 40lRyG6LfJN82vtQp6Wzg/3fJ21a HVV GVV9xUw4Yk/u9/54ynxerr3eBoo42Mu6ZurdKpE5xzDlx6Key3nx8RGhM5OWvRiMib6sr9IiJvqeohmY75Jeq1aKKwiUjz/2TAJWovG1235NtHtuvcNegOfri2p6uX8ac/uS6Q hjV 66sdH3r558fiy3wjMmnLbVodhGR0wABdk79ndTPXXyMMRaisIlIqxfypB7r9ZtG4 t8 c9azvyoG1d89h/0vGk0fPFF63 BMHTv7rphhg6FE0903T3GFIF8Cf5F4KTU319q9PeGn4uKb3VO2qglibrVVq2i oXPqX68El75EOpX4/q2fbLXXtCjx1eDpw0DqJs2uZkvzW9edkgqLYU99nDz23v3dnuaHnUU9Oxp9WFMUcqZ4FV1RKECiYM21znJI/QVrCtXuq3FZsxwOw21Rfv20KsXfOMb6X9WVrZ8WfbmzelJf8sW15fesaP7s1s3a50b00i WTRX5jquqr/1N5xoa0v3SD6hzTv/5BN4 GF48EE3WNoKtV3gpQr4S88dOOmU/2HAgFFuazI/Z6SUlbnbrrv695zGJFy QdZrcz1YVcflvYBIKTAfWKWqQ3Kd29ZB1tBbwG1Q0AHcNWtg5kyX1OfOdXPRW6JTJ57ttY2ZFRuY0wv 0aiURmvjjfN7FwVRef2iEkcxafUgq5cE7sHlwDJgZx eK6s4rLzMJfAB3G3b3MyXu  GWbNcX3dLVFbCkCHudvTRDPpNRzI1DVoTb9zfu7BF5fWLShzmK56 Q4vIXiLymIh8KiKfiMijIrKXl8cBg4G72xpoPi2uSxIxeXdlb63Vq GGG1wdlUGDXB 7x T 3u47wC9 AQsXuvntt94KJ5wA7dv7Gm/c37uwReX1i0oc5iteO0l/DzwJ7Al0B55K3ZfPROAqIGsfgIiMFJH5IjK/zssGC1lEYQpjW3jald0rVTdIetZZborg//xP/sJaKUvK4ZfHwP4Xwb6jtsJ118FBB6XNQvEzXr/fu6gXSfNbVD77UYnDfMVrgi9X1d r6tbU7R6gPNcDRGQI8KmqLsh1nqpOUdUqVa0qL8/5lDkF1gIukOre1Uw5aQoVXSoQhIouFZ4WLzXx73 7jaEPPhj69XObV2zenPdhH xWyq OgN6joPfF8L/HwLKv5X7tfIk3xc/3Lshyy1EVlc9 VOIwX/Ga4FeLyJkiUpq6nQnk24G4H3CyiKwAHgKOE5HpbYg1J19bwCGp7l3NitEr2HbtNlaMXuE9WS5ZAhdf7Frro0bBokX5H/O1r7ll vPm8dKf72H8oI4s2f2rw15eu1bH24yf710xdhNE5bMflThMI6qa9wb0wHXR1AGfAo8DPbw8NvX4Y4BZ c7r06ePtsX0RdO1YkKFyljRigkVOn3R9DY9X S99Zbq0KGNazHmvomoDhyo sQTqlu2NHmqsF87v64vY0UZS9pNxorPEUdL2O9f1OIoJsB8zZJTc06TbCAi9wKjVfWL1M 7ATep6nle/hMRkWOAn2jA0ySLxjvvuG3jZs70dn7XrnDeeXDhhW6hUYJFoV6QMYWUa5qk1y6aAxuSO4Cqfg4ckuP8JlT1hXzJ3XiwerXrVjngAG/J/fDD4f774cMP3UyahCd3sG4CYxrzuqNTiYjs2qwFb7tBFYqq2zLuyivzF/rq1MltMzdqlJv9UmSCXG1sTNx4TdI3A/NEZAagwA8AaxIVQm2t61p59tnc5 29N1x1lduNaOdA15RFXkGKsRkTA566aFT1PuD7wCe4gdbTVPX IAMretu2we23w7e/nTu5d 8Okye7DaUvuSTw5F5sc8yNiTPP3Syq lfgrwHGYhosXw4jRrg9RbPp3BmuvhquuAI6dChIWLYU3Zh4sQ0oo ahh CQQ3In9 pq12K/5pqCJXcozjnmxsSZDZRGxcaNMHo03Hln9nO6d3fHBw8uXFyN2FJ0Y LFWvBRsHIlHHlk7uQ ciQsXRpacgdbim5M3FiCD9u8eVBVBdkWePXsCc8/75J/l3C3wbU55sbEiyX4MN1/Pxx7LGSronnGGfDWW 6cCPCzwJgxJnieShUUStGUKti2DX7 c7j  szHy8pg0iQ3/902izbG5NDqHZ1MADZtcouRHn448/Hu3d2mHN/5TmHjMsYkjnXRBCTjgqAvv4QTT8ye3Pv2hTfesOQeU7YIzESNteADkGlB0M8f/BEDH9 drstWZH7QsGEwbVpB57Ub/9giMBNF1oIPQPMFQT0/hzm/25A9uV93HTzwgCX3GLNFYCaKrAUfgMYLfw76GJ6ZDl9fl HEdu1cq/2sswoXnAmELQIzUWQt AA0LPw5agW8eE W5N6hAzz5pCX3hPCyCMz66E2hBZbgRaS9iPxFRN4WkaUiMi6oazUWhX9E4/uP59TlO/LMdOiyKcMJu 4Kc fCoEEFjy0Kr0 ceH298i0C83szcHsfjReBzYMXEQE6qepaEdkBeAW4XFVfy/aYts6Dbz7QBe4fWcEX48yYwbbhwyjZWp9 bK 9XPnf/fcvXDwpkXl9YqKlr1fN4pqsG434uZWgvY msVzz4Auy0ElEOuIS/ChVfT3beW1N8JHYj7Omxs1z37Yt/di3vuWS 957FyaWZiLx sSIn69XybgSlPR/a4Kw7doMn5UCxWXiz489WVt74VIRWQh8CszJlNxFZKSIzBeR XXZlux7FPpAV8OAaabk3revKwEcUnKHCLw MePn6 VnoTZ7H41XgSZ4Va1X1YOBvYC InJAhnOmqGqVqlaVl5e36XqhVjv83e/g/PPd/qnNHX00/OlP0LVr8HHkYNUgW8bP18vPQm32PhqvCjKLRlX/BbwADAzyOqFVO7ztNrfJdSYnnACzZ8NOOwUbgwdWDbJl/Hy9/CzUZu j8UxVA7kB5cAuqb93AF4GhuR6TJ8 fbStpi arhUTKlTGilZMqNDpi6a3 TlzuuUWVdduT7 ddJLqxo3BXr FCv76xFxUX6 oxmUKD5ivWXJqkLNoDgTuBUpx3xQeVtXrcj0mdtUkJ01yuzBl8v3vu9WpZWWFjckYU1RCqSapqouAQ4J6/tBNnOg2vM5k HC47z63UtUYY0JiK1lbI1dyr6625G6MiQRL8C11yy1Zk/vMPp1o940aKm/7pq0sDJCt4jTGG2tmtsSdd8Lll2c89MDBpZw1eB3bSqxUbJCsLK8x3iWuBR9Y6 6  7JOhXzo0B056 R6tjV6Na1UbDCsLK8x3iWqBR9Y6 6RR2DEiIyLmO47EEYM2dQkuTewlYX s1WcxniXqBZ8IK27p56CH/4wY/mBBw6AEaeQMbmDrSwMQq5VnNY3b0xTiUrwvrfu5syB00 HrVvTDs3cD845NXtyt5WFwci2ivPEfU70tRyvMUmQqATva42Ol16CoUNh8 a0Q7O/CcNPh62lX93XtUNXX5ahm9yyLfmf/d5s65s3ppmClAv2KjL14F97Db73PVi7Nu3QP/9zf749 B98zoa2XSMkuWqWx5mf5XiNiZPQygUXmi8Fnd58EwYOzJjcOfxwvv7869xy2l2xbK37vatQlFiFRWPSJaoF32ZLl7rSvp99ln6sTx 3zV6XLoWPyydJ3igiDrscJfXbkwlX0bTg2 S996B//8zJvXdvtxNTjJM7JHuKoZ/leIOQ5G9PJrqsBQ wahX06we16a1b9t0XXnwRdt 98HH5LMkt Kiz194ExVrwuXz OQwYkDm59 rlumUCTu6Fmr9tG0WEJ8nfnkx0FXeCX7cOhgyBv/41/dhee7nk3r17oCEU8qt71LsxkswGgU0YAkvwIrK3iPxZRJaJyFIRyVylKyxbt8KwYfDqq nHysvdHqqVlYGHUejaKtW9q1kxegXbrt3GitEriiK5R2GFq317MmEIshbNVuDHqvqmiOwELBCROaqaoblcYKpwySUwa1b6sZ12gmeecX3vBWBf3YMVleqTDdeyWTSmkAo2yCoiT SA4fMAAAthSURBVAC3qeqcbOcUbJD1 uvhmmvS7y8rc8n92GODjyHFBt CZa vSbrQB1lFpBK3fd/rGY6NFJH5IjK/rq4usBgavqafdZpkTu4iUFNT0OQO9tU9aPYNyRSzwBO8iHQGHgVGq qXzY r6hRVrVLVqvLy8kBiaPia/o23apn6RJaTJkxwhcUKzAY g2WDm6aYBdpFIyI7ALOAZ1X1t/nOD6qLpnJiJTv9rZZXpkGXTRlOuPJKuPlm369rwheHFa7GtEUoXTQiIsBUYJmX5B6kTR/WMrsmS3L/wQ/gxhsLHpMpDPuGZIpZYC14ETkCeBlYDDSU87tGVWdne0wgLfiNG3lrvy4cUpte9vf1XjvynaX/gvbt/b2mMcYUSK4WfGDTJFX1FUCCen6PQcAFF2RM7u UCyvvncR3LLkbYxIq2StZf/1rNzOmmdWdS3jnvt/yX0dcGEJQxhhTGInadLuJP/4RxmRYDVpWRrdnnueUfv0KH5MxxhRQMlvwy5e7jbIzjS9MmeIqRxpjTMIlL8GvXw nnQb/ lf6sZ/ FM45p/AxGWNMCJKV4FVh5EhYtCj92IABrkRBDEShOJYxJv6S1Qd/550ZB1WprIQHHoDS0oKH1FJRKY5ljIm/5LTgFy6E0aPT72/fHh57DLp2zfiwqLWWC10 2BiTXMlowa9dC2ecAZsyLFWdMgUOPjjjw6LYWrbiWMYYvySjBX/55fC3v6XfP3IknHVW1odFsbVsxbGMMX6Jf4KfMQOmTUu//8ADYeLEnA NYmvZygcbY/wS7wT/0Ueuld5cx47whz9Ahw45Hx7F1rIVxzLG CW CV4VzjsPvvgi/dikSbDffnmfojWtZS DsvnOyXe8GPdNNcb4L76DrHfdBc8 m37/qafC ed7eoqW7pPpZVA23zlRHNg1xiRTwfZk9aJF5YI/ si14Bsn d13hyVLoFu3QOLzsr9ntnO6dujK6qtW2x6hxhhfhb4nayD23NMVFJs82fW5A0ydGlhyB2 DstnO WzDZ9QsronkwK4xJpnim DBbZQ9apRb5HTzzTB4cKCX8zIom2uAdszcMZEc2DXGJFOQW/ZNE5FPRWRJUNfYbp99qPne7lROrETGCe2ua4eME99XpnoZlM01QLtyzUqbBmmMKZggW/D3AAMDfP7tGgYuG/q267Ue GoA068k72UKY3Xvarp2yFwWoUeXHjYN0hhTMIEOsopIJTBLVQ/wcn5r92TNNnDZoNADmM1nyoBrpVsiN8b4LdKDrCIyUkTmi8j8urq6Vj1HvgHKQg9gWivdGBMFoc DV9UpwBRwLfjWPEePLj1ytuDDGMCs7l1tCd0YE6rQW/B yDRw2cAGMI0xxSoRCb5xlwhAqbiNPaxrxBhTzAIbZBWRB4FjgG7AJ8C1qjo112NaO8jql5rFNZ7LFhhjTBTkGmQNrA9eVYcH9dxBsBoxxpikSUQXjR iuPmHMca0hSX4FKsRY4xJGkvwKVYjxhiTNJbgU6xGjDEmaSzBp9jqU2NM0sR3ww9jjDHRrkXTVrn2N/Wyf6oxxiRV6LVo2iLX3HWAEY PYMu2LduPjXh8BJB5XrstcjLGJE2su2hy7W 6dvNaPtvwWdqxhr1RG7PyvsaYuEpsF022CpK1a2ozJncg4/22yMkYk0SxTvCCtOj bGyRkzEmiWKd4JXM3UuKZt02L9P9tsjJGJNEsU7wuUwaNImy0rIm95WVljFp0KS0c22RkzEmiWKd4DuXdc56f3XvaqYNndZk4dK0odMyDpraIidjTBLFehZNtxu6eZ4pY4wxSRTaLBoRGSgi74rI30XkZ34// cbPm/R/cYYU0wCS/AiUgrcDgwC9geGi8j fl7DBkeNMSa7IFvwfYG/q r7qroZeAgY6ucFbHDUGGOyCzLBdwc aPTzh6n7mhCRkSIyX0Tm19XVtegCNjhqjDHZBVmLJtNqo7QRXVWdAkwBN8ja0otU9662hG6MMRkE2YL/ENi70c97AR8FeD1jjDGNBJng3wD2EZGeIlIGDAOeDPB6xhhjGgmsi0ZVt4rIJcCzQCkwTVWXBnU9Y4wxTQVaD15VZwOzg7yGMcaYzGJdqsAYY0x2kSpVICJ1QOYi7/l1A6JYn8DiahmLq2UsrpZJYlwVqlqe6UCkEnxbiMj8bPUYwmRxtYzF1TIWV8sUW1zWRWOMMQllCd4YYxIqSQl StgBZGFxtYzF1TIWV8sUVVyJ6YM3xhjTVJJa8MYYYxqxBG MMQkV wQf9K5RrSUi00TkUxFZEnYsDURkbxH5s4gsE5GlInJ52DEBiEh7EfmLiLydimtc2DE1JiKlIvKWiMwKO5bGRGSFiCwWkYUi4n2vy4CJyC4iMkNE3kl91g6LQEz7pl6nhtuXIjI67LgAROSK1Od iYg8KCLtfXvuOPfBp3aN hvwPVz1yjeA4ar611ADA0TkKGAtcJ qHhB2PAAisgewh6q KSI7AQuAU8J vUREgE6qulZEdgBeAS5X1dfCjKuBiFwJVAE7q qQsONpICIrgCpVjdTCHRG5F3hZVe9OFRrsqKr/CjuuBqm8sQr4jqq2dmGlX7F0x33e91fVDSLyMDBbVe/x4/nj3oIPfNeo1lLVl4BIbQ6rqh r6pupv/8bWEaGTVgKTZ21qR93SN0i0fIQkb2AwcDdYccSByKyM3AUMBVAVTdHKbmn9AeWh53cG2kHdBCRdkBHfCyrHvcE72nXKJNORCqBQ4DXw43ESXWDLAQ BeaoaiTiAiYCVwHbwg4kAwWeE5EFIjIy7GBSegF1wO9T3Vp3i0insINqZhjwYNhBAKjqKuAmYCXwMbBGVZ/z6/njnuA97RplmhKRzsCjwGhV/TLseABUtV5VD8ZtDNNXRELv1hKRIcCnqrog7Fiy6Keqh I2tr841S0YtnbAocAdqnoIsA6I0thYGXAy8EjYsQCIyK64XoeewJ5AJxE506/nj3uCt12jWijVx/0oUKOqM8OOp7nU1/kXgIEhhwLQDzg51df9EHCciEwPN6SvqOpHqT8/BR7DdVmG7UPgw0bfwGbgEn5UDALeVNVPwg4k5XjgH6pap6pbgJnA4X49edwTvO0a1QKpwcypwDJV/W3Y8TQQkXIR2SX19w64D/074UYFqnq1qu6lqpW4z9bzqupb66otRKRTaqCcVBfIACD0GVuq k/gAxHZN3VXfyD0SQ NDCci3TMpK4HvikjH1L/P/rixMV8EuuFH0KK8a5SIPAgcA3QTkQ Ba1V1arhR0Q84C1ic6u8GuCa1MUuY9gDuTc1uKAEeVtVITUmMoN2Bx1xOoB3wgKo E25I210K1KQaXe8DI0KOBwAR6YibcXdh2LE0UNXXRWQG8CawFXgLH8sWxHqapDHGmOzi3kVjjDEmC0vwxhiTUJbgjTEmoSzBG2NMQlmCN8aYhLIEb4xHIjJWRH7S6Od2IrJaRK4PMy5jsrEEb0zrDQDeBX6QWqRiTKRYgjdFT0TOFpFFqXr094tIhYjMTd03V0R6ZHnocGASqdWIhYvYGG8swZuiJiLfBsYAx6nqQcDlwG24Ov4HAjXALRke1wG3rHwWbun78IIFbYxHluBNsTsOmNGwaYaqfg4cBjyQOn4/cESGxw0B/qyq63HF205NlVowJjIswZtiJ QvMZ3p HDg FSlyQVAV BYf0Mzpm0swZtiNxc3SNoVQER2A bhqkcCVOO2VNsutWvREUAPVa1MVZu8GOumMRFjxcZM0RORc4CfAvW4an5jgWlAN9zuRCNUdaWIjMXts7saGKiqwxo9x264GTV7qeqmgv4CxmRhCd4YYxLKumiMMSahLMEbY0xCWYI3xpiEsgRvjDEJZQneGGMSyhK8McYklCV4Y4xJqP8HeoRErv4cqOkAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "def kernel(point,xmat,k):\n",
    "    m,n=np.shape(xmat)\n",
    "    weights=np.mat(np.eye(m))\n",
    "    for j in range(m):\n",
    "        diff=point-x[j]\n",
    "        weights[j,j]=np.exp(diff*diff.T/(-2.0*k**2))\n",
    "    return weights\n",
    "def localWeight(point,xmat,ymat,k):\n",
    "    wei=kernel(point,xmat,k)\n",
    "    W=(x.T*(wei*x)).I*(x.T*(wei*ymat.T))\n",
    "    return W\n",
    "def localWeightRegression(xmat,ymat,k):\n",
    "    m,n=np.shape(xmat)\n",
    "    ypred=np.zeros(m)\n",
    "    for i in range(m):\n",
    "        ypred[i]=xmat[i]*localWeight(xmat[i],xmat,ymat,k)\n",
    "    return ypred\n",
    "data=pd.read_csv('Ex10.csv')\n",
    "colA=np.array(data.colA)\n",
    "colB=np.array(data.colB)\n",
    "mcolA=np.mat(colA)\n",
    "mcolB=np.mat(colB)\n",
    "m=np.shape(mcolA)[1]\n",
    "one=np.ones((1,m),dtype=int)\n",
    "x=np.hstack((one.T,mcolA.T))\n",
    "print(x.shape)\n",
    "ypred=localWeightRegression(x,mcolB,0.5)\n",
    "SortIndex=x[:,1].argsort(0)\n",
    "xsort=x[SortIndex][:,0]\n",
    "fig=plt.figure()\n",
    "ax=fig.add_subplot(1,1,1)\n",
    "ax.scatter(colA,colB,color='green')\n",
    "ax.plot(xsort[:,1],ypred[SortIndex],color='red',linewidth=5)\n",
    "plt.xlabel('colA')\n",
    "plt.ylabel('colB')"
   ]
  },
  {
   "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": "3.8.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}