Cagri.Cakir_Assignment3.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Distance: \n",
      "5.73\n",
      "Type\n",
      "2\n",
      "\n",
      "\n",
      "Distance: \n",
      "5.490000000000002\n",
      "Type\n",
      "2\n",
      "\n",
      "\n",
      "Distance: \n",
      "5.240000000000001\n",
      "Type\n",
      "2\n",
      "\n",
      "\n",
      "For type: 1\n",
      "------------------\n",
      "Accuracy:  100.0\n",
      "Precision:  100.0\n",
      "Recall:  100.0\n",
      "\n",
      "\n",
      "For type: 2\n",
      "------------------\n",
      "Accuracy:  94.0\n",
      "Precision:  96.0\n",
      "Recall:  94.0\n",
      "\n",
      "\n",
      "For type: 3\n",
      "------------------\n",
      "Accuracy:  94.0\n",
      "Precision:  96.0\n",
      "Recall:  94.0\n",
      "\n",
      "\n"
     ]
    }
   ],
   "source": [
    "%matplotlib inline\n",
    "import numpy as np\n",
    "import matplotlib as mpl\n",
    "import matplotlib.pylab as plt\n",
    "import pandas as pd\n",
    "import sys\n",
    "import heapq as hp\n",
    "\n",
    "data = pd.read_csv(u'data/iris.txt',sep=' ')\n",
    "\n",
    "A = np.hstack([\n",
    "        np.matrix(data.sl).T, \n",
    "        np.matrix(data.sw).T, \n",
    "        np.matrix(data.pl).T, \n",
    "        np.matrix(data.pw).T])\n",
    "\n",
    "c = np.matrix(data.c).T\n",
    "xTest = np.mat([2.8, 1.4, 3.5, 1.2])\n",
    "\n",
    "class tup:\n",
    "    def __init__(self, val, idx):\n",
    "        self.val = val\n",
    "        self.idx = idx\n",
    "    def __lt__(self, other):\n",
    "        '''Redefine for max-heap'''\n",
    "        return self.val > other.val\n",
    "    def __le__(self, other):\n",
    "        return self.val <= other.val\n",
    "    def __eq__(self, other):\n",
    "        return self.val == other.val\n",
    "    def __ne__(self, other):\n",
    "        return self.val != other.val\n",
    "    def __gt__(self, other):\n",
    "        return self.val > other.val\n",
    "    def __ge__(self, other):\n",
    "        return self.val >= other.val\n",
    "    def __str__(self):\n",
    "        return '{:.3},{:d}'.format(self.val,self.idx)\n",
    "\n",
    "def evaluate(k, A, types ,iType):\n",
    "    heap = []\n",
    "    N = A.shape[0]\n",
    "    false_positive = 0\n",
    "    false_negative = 0\n",
    "    true_positive = 0\n",
    "    true_negative = 0\n",
    "    \n",
    "    for i in range(N):\n",
    "        \n",
    "        for l in range(k):\n",
    "            hp.heappush(heap, tup(np.inf, -1))\n",
    "        \n",
    "        for j in range(N):\n",
    "            e = A[j,:]- A[i,:]\n",
    "            e = e.reshape((4,1))\n",
    "            tp = tup(float(e.T*e), j)\n",
    "            if tp <= heap[0]:\n",
    "                hp.heapreplace(heap, tp)\n",
    "        \n",
    "        categories = []\n",
    "        for j in range(len(heap)):\n",
    "            h = hp.heappop(heap)\n",
    "            categories.append(int(types[h.idx]))\n",
    "        result = max(set(categories), key=categories.count)\n",
    "        \n",
    "        if(result == int(c[i]) and int(c[i]) == iType):\n",
    "            true_positive += 1\n",
    "        if(result == int(c[i]) and int(c[i]) != iType):\n",
    "            true_negative += 1\n",
    "        if(result != int(c[i]) and int(c[i]) == iType):\n",
    "            false_negative += 1\n",
    "        if(result != int(c[i]) and result == iType):\n",
    "            false_positive += 1\n",
    "            \n",
    "    precision = true_positive / (true_positive + false_positive) * 100\n",
    "    recall = true_positive / (true_positive + false_negative) * 100\n",
    "    accuracy = (true_positive + true_negative) / (true_positive + true_negative + false_positive + false_negative) * 100\n",
    "\n",
    "    print(\"For type:\", iType)\n",
    "    print(\"------------------\")\n",
    "    print(\"Accuracy: \", precision)\n",
    "    print(\"Precision: \", accuracy)\n",
    "    print(\"Recall: \", recall)\n",
    "    print(\"\\n\")\n",
    "    \n",
    "def kNN(K, A, test, c):\n",
    "    heap = []\n",
    "    N = A.shape[0]\n",
    "    \n",
    "    for i in range(K):\n",
    "        hp.heappush(heap, tup(np.inf, -1))\n",
    "        \n",
    "    for j in range(N):\n",
    "        e = A[j,:]- test\n",
    "        e = e.reshape((4,1))\n",
    "        tp = tup(float(e.T*e), j)\n",
    "        if tp <= heap[0]:\n",
    "            hp.heapreplace(heap, tp)\n",
    "    \n",
    "    for m in range(K):\n",
    "        h = hp.heappop(heap)\n",
    "        print(\"Distance: \")\n",
    "        print(h.val)\n",
    "        print(\"Type\")\n",
    "        print(int(c[h.idx]))\n",
    "        print(\"\\n\")\n",
    "\n",
    "k = 3\n",
    "#print(xTest)\n",
    "kNN(k, A, xTest,c)\n",
    "\n",
    "evaluate(k,A,c,1)\n",
    "evaluate(k,A,c,2)\n",
    "evaluate(k,A,c,3)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "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": "3.5.1"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 0
}