gistfile1.py

{
 "metadata": {
  "css": [
   ""
  ],
  "name": "",
  "signature": "sha256:2a33067410fdd0356cedb2c0893414e80734c219ba9dd2bba82e2ad1b379606b"
 },
 "nbformat": 3,
 "nbformat_minor": 0,
 "worksheets": [
  {
   "cells": [
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "import xml.etree.ElementTree as ET\n",
      "tree = ET.parse('Restaurants_Train.xml')\n",
      "root = tree.getroot()\n",
      "\n",
      "data = []\n",
      "labels = []\n",
      "\n",
      "for child in root:\n",
      "    for c in child:\n",
      "        if c.tag == 'text':\n",
      "            data.append(c.text)\n",
      "        if c.tag == 'aspectCategories':\n",
      "            if c[0].attrib['category'] == 'service':\n",
      "                labels.append(0)\n",
      "            elif c[0].attrib['category'] == 'food':\n",
      "                labels.append(1)\n",
      "            elif c[0].attrib['category'] == 'ambience':\n",
      "                labels.append(2)\n",
      "            elif c[0].attrib['category'] == 'price':\n",
      "                labels.append(3)\n",
      "            else:\n",
      "                data.pop()\n",
      "                     \n",
      "print len(data)\n",
      "print len(labels)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "1962\n",
        "1962\n"
       ]
      }
     ],
     "prompt_number": 30
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "from sklearn.feature_extraction.text import TfidfVectorizer\n",
      "from sklearn import svm\n",
      "from sklearn import cross_validation\n",
      "import numpy as np\n",
      "\n",
      "labels = np.array(labels)\n",
      "\n",
      "vec = TfidfVectorizer()\n",
      "X = vec.fit_transform(data)\n",
      "\n",
      "clf = svm.SVC(kernel='linear', C=1)\n",
      "scores = cross_validation.cross_val_score(clf, X, labels, cv=10, scoring='f1')\n",
      "print scores.mean()"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "0.786095714575\n"
       ]
      }
     ],
     "prompt_number": 36
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [],
     "language": "python",
     "metadata": {},
     "outputs": []
    }
   ],
   "metadata": {}
  }
 ]
}