{
 "metadata": {
  "name": "",
  "signature": "sha256:56e3d62355fc065626699ebb696ceabe03f7b3853b3e204c8e5d63fe0567cb29"
 },
 "nbformat": 3,
 "nbformat_minor": 0,
 "worksheets": [
  {
   "cells": [
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "from numpy import *\n",
      "import operator\n",
      "import seaborn as sns\n",
      "sns.set(context='poster', style='dark')\n",
      "sns.set_context(rc={'lines.markeredgewidth': 0.5})\n",
      "import pandas as pd\n",
      "from mpl_toolkits.mplot3d import Axes3D\n",
      "\n",
      "dating_df = pd.read_csv('datingTestSet2.txt', sep='\\t', header=None)\n",
      "dating_df.columns = ['miles', 'games', 'ice_cream', 'opinion']\n",
      "\n",
      "norm_df = pd.DataFrame()\n",
      "labels = dating_df.columns[:-1]\n",
      "\n",
      "for L in labels:\n",
      "    norm_df[L] = (dating_df[L]-dating_df[L].min()) / \\\n",
      "        (dating_df[L].max()-dating_df[L].min())\n",
      "\n",
      "norm_df['opinion'] = dating_df.opinion\n",
      "norm_df.head()"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "html": [
        "<div style=\"max-height:1000px;max-width:1500px;overflow:auto;\">\n",
        "<table border=\"1\" class=\"dataframe\">\n",
        "  <thead>\n",
        "    <tr style=\"text-align: right;\">\n",
        "      <th></th>\n",
        "      <th>miles</th>\n",
        "      <th>games</th>\n",
        "      <th>ice_cream</th>\n",
        "      <th>opinion</th>\n",
        "    </tr>\n",
        "  </thead>\n",
        "  <tbody>\n",
        "    <tr>\n",
        "      <th>0</th>\n",
        "      <td> 0.448325</td>\n",
        "      <td> 0.398051</td>\n",
        "      <td> 0.562334</td>\n",
        "      <td> 3</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>1</th>\n",
        "      <td> 0.158733</td>\n",
        "      <td> 0.341955</td>\n",
        "      <td> 0.987244</td>\n",
        "      <td> 2</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>2</th>\n",
        "      <td> 0.285429</td>\n",
        "      <td> 0.068925</td>\n",
        "      <td> 0.474496</td>\n",
        "      <td> 1</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>3</th>\n",
        "      <td> 0.823201</td>\n",
        "      <td> 0.628480</td>\n",
        "      <td> 0.252489</td>\n",
        "      <td> 1</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>4</th>\n",
        "      <td> 0.420102</td>\n",
        "      <td> 0.079820</td>\n",
        "      <td> 0.078578</td>\n",
        "      <td> 1</td>\n",
        "    </tr>\n",
        "  </tbody>\n",
        "</table>\n",
        "</div>"
       ],
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 1,
       "text": [
        "      miles     games  ice_cream  opinion\n",
        "0  0.448325  0.398051   0.562334        3\n",
        "1  0.158733  0.341955   0.987244        2\n",
        "2  0.285429  0.068925   0.474496        1\n",
        "3  0.823201  0.628480   0.252489        1\n",
        "4  0.420102  0.079820   0.078578        1"
       ]
      }
     ],
     "prompt_number": 1
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "def find_kNN(inX, dataSet, labels, k):\n",
      "    dataSetSize = dataSet.shape[0]\n",
      "    \n",
      "    #distance calculation:\n",
      "    diffMat = tile(inX, (dataSetSize,1)) - dataSet\n",
      "    sqDiffMat = diffMat**2\n",
      "    sqDistances = sqDiffMat.sum(axis=1)\n",
      "    distances = sqDistances**0.5\n",
      "    \n",
      "    sortedDistIndices = distances.argsort()\n",
      "    classCount = {}\n",
      "    \n",
      "    for i in range(k): \n",
      "        voteIlabel = labels[sortedDistIndices[i]]\n",
      "        classCount[voteIlabel] = classCount.get(voteIlabel, 0) + 1\n",
      "    \n",
      "    sortedClassCount = sorted(classCount.items(),\n",
      "                              key=operator.itemgetter(1), \n",
      "                              reverse=True)\n",
      "    return sortedClassCount[0][0], sortedClassCount"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 2
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "def make_sets(data_df, test_portion):\n",
      "    import random as rnd\n",
      "\n",
      "    tot_ix = range(len(data_df))\n",
      "    test_ix = sort(rnd.sample(tot_ix, int(test_portion * len(data_df))))\n",
      "    train_ix = list(set(tot_ix) ^ set(test_ix))\n",
      "\n",
      "    test_df = data_df.ix[test_ix]\n",
      "    train_df = data_df.ix[train_ix]\n",
      "    \n",
      "    return train_df, test_df"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 3
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "import time\n",
      "\n",
      "def test_model(data_df, test_portion, kNN):    \n",
      "    train_df, test_df = make_sets(data_df, test_portion)\n",
      "    train_array = zeros((len(train_df),3))\n",
      "    test_array = zeros((len(test_df),3))\n",
      "\n",
      "    cols = train_df.columns\n",
      "\n",
      "    train_array = train_df[cols[:-1]].values\n",
      "    test_array = test_df[cols[:-1]].values\n",
      "\n",
      "    labels = train_df.opinion.values\n",
      "\n",
      "    test_df = test_df.reset_index()\n",
      "\n",
      "    pass_counter = 0\n",
      "    fail_counter = 0\n",
      "\n",
      "    for i in range(len(test_array)):\n",
      "        result, full_stack = find_kNN(test_array[i,:], train_array, labels, kNN)\n",
      "        if result == test_df.ix[i].opinion:\n",
      "            pass_counter += 1\n",
      "        else:\n",
      "            fail_counter += 1\n",
      "\n",
      "    return pass_counter, fail_counter\n",
      "\n",
      "\n",
      "def plot_model_test(data_df, test_proportion, kNN, repeats):\n",
      "    r = repeats\n",
      "    pass_succ = []\n",
      "    pass_hist = []\n",
      "    pass_avg = []\n",
      "    \n",
      "    for i in range(r):\n",
      "        passes, fails = test_model(norm_df, test_proportion, kNN)\n",
      "        pass_succ = passes/(passes+fails)\n",
      "        pass_hist.append(pass_succ)\n",
      "        pass_avg.append(mean(pass_hist))\n",
      "        \n",
      "    plot(arange(1, len(pass_avg)+1), pass_avg,\n",
      "         label='average (kNN={knn}, t_p={t_p})'.format(knn=kNN, t_p=test_proportion))\n",
      "    xlabel('trial number')\n",
      "    ylabel('accuracy')\n",
      "    legend(loc='best')\n",
      "\n",
      "\n",
      "repeats = 10\n",
      "t_prop = 0.25\n",
      "kNN = [3]\n",
      "\n",
      "t1 = time.time()\n",
      "[plot_model_test(norm_df, t_prop, i, repeats) for i in kNN]\n",
      "t2 = time.time()\n",
      "\n",
      "print(\"time elapsed:\", t2-t1, \"sec\")"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "time elapsed: 0.9517326354980469 sec\n"
       ]
      },
      {
       "metadata": {},
       "output_type": "display_data",
       "png": "iVBORw0KGgoAAAANSUhEUgAAAygAAAIxCAYAAACmWpwWAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nM7O28f/ZF1tlTWAWaQAAAABJRU5ErkJggg==\n",
       "text": [
        "<matplotlib.figure.Figure at 0x7ff4e8584c88>"
       ]
      }
     ],
     "prompt_number": 6
    }
   ],
   "metadata": {}
  }
 ]
}