{
 "metadata": {
  "name": "",
  "signature": "sha256:2d56745aec6056b08ab940251ec750304e4ffefa439ffe94ce85d3f6a239b29e"
 },
 "nbformat": 3,
 "nbformat_minor": 0,
 "worksheets": [
  {
   "cells": [
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "%matplotlib inline\n",
      "\n",
      "import numpy as np\n",
      "import pandas as pd\n",
      "import matplotlib.pyplot as plt\n",
      "import seaborn as sns\n",
      "from sklearn.ensemble import RandomForestClassifier\n",
      "pd.set_option('display.max_columns', 50)\n",
      "\n",
      "train = pd.DataFrame.from_csv('train.csv')\n",
      "\n",
      "###\n",
      "#   Use numerical values for categorical data\n",
      "\n",
      "def makeNumerical(input_df):\n",
      "    \n",
      "    output_df = input_df.copy()\n",
      "    non_numeric_cols = [col for col in output_df.columns\\\n",
      "                        if output_df[col].dtype != 'int64']\n",
      "    for col in non_numeric_cols:\n",
      "        vals = pd.unique(output_df[col])\n",
      "        for i in xrange(len(vals)):\n",
      "            output_df.loc[(output_df[col] == vals[i]), col] = i     \n",
      "\n",
      "    return output_df"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 16
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "train_numerical = makeNumerical(train)\n",
      "features = ['T1_V1','T1_V2','T2_V1','T2_V2']\n",
      "X = train_numerical[features].copy().astype(float)\n",
      "y = train_numerical[train_numerical.columns[0]].copy().astype(float)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 17
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "from sklearn import svm\n",
      "clf = svm.SVR(C = 0.5, gamma=.001)   #try making C a little lower, maybe change gamma\n",
      "clf.fit(X,y)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 22
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "# Load test df, predict on it\n",
      "\n",
      "test = makeNumerical(pd.DataFrame.from_csv('test.csv')[['T1_V1','T1_V2','T2_V1','T2_V2']])\n",
      "\n",
      "preds = clf.predict(test)\n",
      "#\n",
      "out_df = pd.DataFrame({'Hazard': preds}, index=test.index)\n",
      "out_df.to_csv(\"submission_3.csv\", index=True)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 25
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "same_preds = clf.predict(X) - y"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 33
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "same_preds.head(20)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 34,
       "text": [
        "Id\n",
        "1      0.853256\n",
        "2     -0.998832\n",
        "3      1.116817\n",
        "4      2.453021\n",
        "5      2.615416\n",
        "12     1.730286\n",
        "15    -1.931312\n",
        "19     2.509336\n",
        "21     0.911379\n",
        "22     0.386930\n",
        "23    -5.199475\n",
        "24   -11.527683\n",
        "25     1.994604\n",
        "26    -0.331582\n",
        "31    -9.513998\n",
        "32     3.166062\n",
        "33     1.301354\n",
        "39     0.776953\n",
        "41    -1.171212\n",
        "43     0.810963\n",
        "Name: Hazard, dtype: float64"
       ]
      }
     ],
     "prompt_number": 34
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [],
     "language": "python",
     "metadata": {},
     "outputs": []
    }
   ],
   "metadata": {}
  }
 ]
}