test.ipynb

gistfile1.txt

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import xgboost as xgb\n",
    "from sklearn import preprocessing, linear_model\n",
    "import math"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "train = pd.read_csv('../data/train/PPD_Training_Master_GBK_3_1_Training_Set.csv')\n",
    "test = pd.read_csv('../data/test/PPD_Master_GBK_2_Test_Set.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "all_data = train.append(test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "cat_cols = [\"UserInfo_2\", \"UserInfo_4\", \"UserInfo_7\", \"UserInfo_8\", \"UserInfo_19\", \"UserInfo_20\", \"UserInfo_1\", \\\n",
    "            \"UserInfo_3\", \"UserInfo_5\", \"UserInfo_6\", \"UserInfo_9\", \"UserInfo_2\", \"UserInfo_4\", \\\n",
    "            \"UserInfo_7\", \"UserInfo_8\", \"UserInfo_19\", \"UserInfo_20\", \"UserInfo_11\", \"UserInfo_12\", \"UserInfo_13\", \\\n",
    "            \"UserInfo_14\", \"UserInfo_15\", \"UserInfo_16\", \"UserInfo_18\", \"UserInfo_21\", \"UserInfo_22\", \"UserInfo_23\", \\\n",
    "            \"UserInfo_24\", \"Education_Info1\", \"Education_Info2\", \"Education_Info3\", \"Education_Info4\", \\\n",
    "            \"Education_Info5\", \"Education_Info6\", \"Education_Info7\", \"Education_Info8\", \"WeblogInfo_19\", \\\n",
    "            \"WeblogInfo_20\", \"WeblogInfo_21\", \"SocialNetwork_1\", \"SocialNetwork_2\", \"SocialNetwork_7\", \\\n",
    "            \"ListingInfo\", \"SocialNetwork_12\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "for col in cat_cols:\n",
    "    if col in all_data.columns.values:\n",
    "        all_data[col] = pd.factorize(all_data[col])[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "all_data.fillna(-1, inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "train = all_data[all_data['target'] > -1].copy()\n",
    "test = all_data[all_data['target'] == -1].copy()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "dtrain = xgb.DMatrix(train.drop([\"Idx\", \"target\"], axis=1), train[\"target\"].values)\n",
    "dtest = xgb.DMatrix(test.drop([\"Idx\", \"target\"], axis=1), label=test[\"target\"].values)\n",
    "\n",
    "params = {}\n",
    "params['objective'] = 'binary:logistic'\n",
    "params['eta'] = 0.02\n",
    "params['min_child_weight'] = 1\n",
    "params['subsample'] = 0.8\n",
    "params['colsample_bytree'] = 0.8\n",
    "params['max_depth'] = 8\n",
    "params['eval_metric'] = 'auc'\n",
    "params['nthread'] = 3\n",
    "params['silent'] = 1\n",
    "num_rounds = 120"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "model = xgb.train(params, dtrain, num_rounds)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "y_pred = model.predict(dtest, ntree_limit=model.best_iteration)\n",
    "pd.DataFrame({'Idx': test['Idx'].values, 'score': y_pred}).to_csv('submit.csv', index=False)"
   ]
  }
 ],
 "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": "2.7.10"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 0
}