K Means from Scratch.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# randomly choose K means\n",
    "# loop:\n",
    "#   assign closest points to each of K means\n",
    "#   recalculate means\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "from matplotlib import style\n",
    "style.use('ggplot')\n",
    "import numpy as np\n",
    "import random\n",
    "from sklearn import preprocessing, cross_validation\n",
    "import pandas as pd\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXgAAAEECAYAAAAifS8cAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nMeZrbZO0AAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f0615325550>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "colors = ['r', 'b', 'g', 'k']\n",
    "\n",
    "X = np.array([[1, 2],\n",
    "              [1.5, 1.8],\n",
    "              [5, 8 ],\n",
    "              [8, 8],\n",
    "              [1, 0.6],\n",
    "              [9,11]])\n",
    "\n",
    "plt.scatter(X[:,0], X[:,1], s=150)\n",
    "plt.show()\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "# t - tolerance: to decide end of loop \n",
    "# stop if the means do not move greater than tolerance\n",
    "class K_Means:\n",
    "    def __init__(self, k=2, tol=0.001, max_iter=300):\n",
    "        self.k = k\n",
    "        self.tol = tol\n",
    "        self.max_iter = max_iter\n",
    "\n",
    "    def fit(self,data):\n",
    "        self.centroids = {}\n",
    "\n",
    "        for i in range(self.k):\n",
    "            self.centroids[i] = data[i]\n",
    "         \n",
    "        for i in range(self.max_iter):\n",
    "            self.classifications = {}\n",
    "\n",
    "            for i in range(self.k):\n",
    "                self.classifications[i] = [] #contains centroids\n",
    "            # populate list\n",
    "            for featureset in data:\n",
    "                # distances to each of centroids for each featureset\n",
    "                distances = [np.linalg.norm(featureset-self.centroids[centroid]) for centroid in self.centroids]\n",
    "                classification = distances.index(min(distances))\n",
    "                self.classifications[classification].append(featureset)\n",
    "            # classifying\n",
    "            prev_centroids = dict(self.centroids)\n",
    "\n",
    "            for classification in self.classifications:\n",
    "                self.centroids[classification] = np.average(self.classifications[classification],axis=0)\n",
    "            \n",
    "            optimized = True\n",
    "\n",
    "            for c in self.centroids:\n",
    "                original_centroid = prev_centroids[c]\n",
    "                current_centroid = self.centroids[c]\n",
    "                if np.sum((current_centroid-original_centroid)/original_centroid*100.0) > self.tol:\n",
    "                    print(np.sum((current_centroid-original_centroid)/original_centroid*100.0))\n",
    "                    optimized = False\n",
    "            \n",
    "            if optimized:\n",
    "                break\n",
    "                \n",
    "    def predict(self,data):\n",
    "        distances = [np.linalg.norm(data-self.centroids[centroid]) for centroid in self.centroids]\n",
    "        classification = distances.index(min(distances))\n",
    "        return classification"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {
    "collapsed": false,
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "453.333333333\n",
      "102.721088435\n"
     ]
    }
   ],
   "source": [
    "clf = K_Means()\n",
    "clf.fit(X)\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXoAAAEECAYAAAAmiP8hAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f0617827a10>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "for centroid in clf.centroids:\n",
    "    plt.scatter(clf.centroids[centroid][0], clf.centroids[centroid][1], marker=\"o\", color=\"k\", s=150, linewidths=5)\n",
    "\n",
    "for classification in clf.classifications:\n",
    "    color = colors[classification]\n",
    "    for featureset in clf.classifications[classification]:\n",
    "        plt.scatter(featureset[0], featureset[1], marker=\"x\", color=color, s=50, linewidths=5)\n",
    "        \n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXEAAAEECAYAAADeaATWAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nMBglDgAGo8QBwGD/D6xOPTBs87AmAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f0615448cd0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "unknowns = np.array([[1,3],\n",
    "                     [8,9],\n",
    "                     [0,3],\n",
    "                     [5,4],\n",
    "                     [6,4],])\n",
    "\n",
    "for unknown in unknowns:\n",
    "    classification = clf.predict(unknown)\n",
    "    plt.scatter(unknown[0], unknown[1], marker=\"*\", color=colors[classification], s=100, linewidths=5)\n",
    "\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "484.722222222\n",
      "13.0\n",
      "52.1072796935\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXoAAAEECAYAAAAmiP8hAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nRCQ5Bj0RkeT+HxzKYHSb4jitAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f06153ab0d0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "X = np.array([[1, 2],\n",
    "              [1.5, 1.8],\n",
    "              [5, 8 ],\n",
    "              [8, 8],\n",
    "              [1, 0.6],\n",
    "              [9,11],\n",
    "              [1,3],\n",
    "              [8,9],\n",
    "              [0,3],\n",
    "              [5,4],\n",
    "              [6,4],])\n",
    "\n",
    "clf.fit(X)\n",
    "\n",
    "for centroid in clf.centroids:\n",
    "    plt.scatter(clf.centroids[centroid][0], clf.centroids[centroid][1],\n",
    "                marker=\"o\", color=\"k\", s=150, linewidths=5)\n",
    "\n",
    "for classification in clf.classifications:\n",
    "    color = colors[classification]\n",
    "    for featureset in clf.classifications[classification]:\n",
    "        plt.scatter(featureset[0], featureset[1], marker=\"x\", color=color, s=50, linewidths=5)\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "   pclass  survived     sex      age  sibsp  parch  ticket      fare    cabin  \\\n",
      "0       1         1  female  29.0000      0      0   24160  211.3375       B5   \n",
      "1       1         1    male   0.9167      1      2  113781  151.5500  C22 C26   \n",
      "2       1         0  female   2.0000      1      2  113781  151.5500  C22 C26   \n",
      "3       1         0    male  30.0000      1      2  113781  151.5500  C22 C26   \n",
      "4       1         0  female  25.0000      1      2  113781  151.5500  C22 C26   \n",
      "\n",
      "  embarked boat                        home.dest  \n",
      "0        S    2                     St Louis, MO  \n",
      "1        S   11  Montreal, PQ / Chesterville, ON  \n",
      "2        S  NaN  Montreal, PQ / Chesterville, ON  \n",
      "3        S  NaN  Montreal, PQ / Chesterville, ON  \n",
      "4        S  NaN  Montreal, PQ / Chesterville, ON  \n"
     ]
    }
   ],
   "source": [
    "df = pd.read_excel('titanic.xls')\n",
    "df.drop(['body','name'], 1, inplace=True)\n",
    "#df.convert_objects(convert_numeric=True)\n",
    "print(df.head())\n",
    "df.fillna(0,inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "\n",
    "def handle_non_numerical_data(df):\n",
    "    \n",
    "    # handling non-numerical data: must convert.\n",
    "    columns = df.columns.values\n",
    "\n",
    "    for column in columns:\n",
    "        text_digit_vals = {}\n",
    "        def convert_to_int(val):\n",
    "            return text_digit_vals[val]\n",
    "\n",
    "        #print(column,df[column].dtype)\n",
    "        if df[column].dtype != np.int64 and df[column].dtype != np.float64:\n",
    "            \n",
    "            column_contents = df[column].values.tolist()\n",
    "            #finding just the uniques\n",
    "            unique_elements = set(column_contents)\n",
    "            # great, found them. \n",
    "            x = 0\n",
    "            for unique in unique_elements:\n",
    "                if unique not in text_digit_vals:\n",
    "                    # creating dict that contains new\n",
    "                    # id per unique string\n",
    "                    text_digit_vals[unique] = x\n",
    "                    x+=1\n",
    "            # now we map the new \"id\" vlaue\n",
    "            # to replace the string. \n",
    "            df[column] = list(map(convert_to_int,df[column]))\n",
    "\n",
    "    return df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "   pclass  survived  sex      age  sibsp  parch  ticket      fare  cabin  \\\n",
      "0       1         1    1  29.0000      0      0     757  211.3375     31   \n",
      "1       1         1    0   0.9167      1      2     326  151.5500    158   \n",
      "2       1         0    1   2.0000      1      2     326  151.5500    158   \n",
      "3       1         0    0  30.0000      1      2     326  151.5500    158   \n",
      "4       1         0    1  25.0000      1      2     326  151.5500    158   \n",
      "\n",
      "   embarked  boat  home.dest  \n",
      "0         2     1        248  \n",
      "1         2    19        348  \n",
      "2         2     0        348  \n",
      "3         2     0        348  \n",
      "4         2     0        348  \n"
     ]
    }
   ],
   "source": [
    "df = handle_non_numerical_data(df)\n",
    "print(df.head())\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# add/remove features just to see impact they have.\n",
    "df.drop(['ticket','home.dest'], 1, inplace=True)\n",
    "\n",
    "\n",
    "X = np.array(df.drop(['survived'], 1).astype(float))\n",
    "X = preprocessing.scale(X)\n",
    "y = np.array(df['survived'])\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.68449197861\n"
     ]
    }
   ],
   "source": [
    "clf = K_Means()\n",
    "clf.fit(X)\n",
    "\n",
    "correct = 0\n",
    "for i in range(len(X)):\n",
    "\n",
    "    predict_me = np.array(X[i].astype(float))\n",
    "    predict_me = predict_me.reshape(-1, len(predict_me))\n",
    "    prediction = clf.predict(predict_me)\n",
    "    if prediction == y[i]:\n",
    "        correct += 1\n",
    "\n",
    "\n",
    "print(correct/float(len(X)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 2",
   "language": "python",
   "name": "python2"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 2
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython2",
   "version": "2.7.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}