train.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/wing/.virtualenvs/cv/local/lib/python2.7/site-packages/IPython/html.py:14: ShimWarning: The `IPython.html` package has been deprecated since IPython 4.0. You should import from `notebook` instead. `IPython.html.widgets` has moved to `ipywidgets`.\n",
      "  \"`IPython.html.widgets` has moved to `ipywidgets`.\", ShimWarning)\n",
      "/home/wing/.virtualenvs/cv/local/lib/python2.7/site-packages/sklearn/cross_validation.py:44: DeprecationWarning: This module was deprecated in version 0.18 in favor of the model_selection module into which all the refactored classes and functions are moved. Also note that the interface of the new CV iterators are different from that of this module. This module will be removed in 0.20.\n",
      "  \"This module will be removed in 0.20.\", DeprecationWarning)\n",
      "/home/wing/.virtualenvs/cv/local/lib/python2.7/site-packages/sklearn/grid_search.py:43: DeprecationWarning: This module was deprecated in version 0.18 in favor of the model_selection module into which all the refactored classes and functions are moved. This module will be removed in 0.20.\n",
      "  DeprecationWarning)\n",
      "/home/wing/.virtualenvs/cv/local/lib/python2.7/site-packages/sklearn/learning_curve.py:23: DeprecationWarning: This module was deprecated in version 0.18 in favor of the model_selection module into which all the functions are moved. This module will be removed in 0.20\n",
      "  DeprecationWarning)\n"
     ]
    }
   ],
   "source": [
    "import re\n",
    "import seaborn as sns\n",
    "import xgboost as xgb\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import sklearn.model_selection as ms\n",
    "import sklearn.preprocessing as pp\n",
    "import sklearn.feature_selection as fs\n",
    "from sklearn.cross_validation import StratifiedKFold\n",
    "from sklearn import metrics\n",
    "from sklearn.grid_search import GridSearchCV\n",
    "from sklearn.learning_curve import validation_curve\n",
    "\n",
    "\n",
    "import matplotlib.pylab as plt\n",
    "from features1 import add_title,fill_age,fill_fare,fill_embarked,fill_cabin,fill_ticket,add_family\n",
    "\n",
    "seed = 9497"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "train = pd.read_csv('train.csv')\n",
    "test = pd.read_csv('test.csv')\n",
    "total=pd.concat([train,test])\n",
    "PassengerId=total['PassengerId']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": true,
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "# Add title, drop name, add family\n",
    "total=add_title(total)\n",
    "total.drop('Name',axis=1,inplace=True)# Fill\n",
    "total=add_family(total)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/wing/.virtualenvs/cv/local/lib/python2.7/site-packages/pandas/core/generic.py:3110: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: http://pandas.pydata.org/pandas-docs/stable/indexing.html#indexing-view-versus-copy\n",
      "  self[name] = value\n",
      "/home/wing/.virtualenvs/cv/local/lib/python2.7/site-packages/pandas/core/generic.py:3660: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: http://pandas.pydata.org/pandas-docs/stable/indexing.html#indexing-view-versus-copy\n",
      "  self._update_inplace(new_data)\n",
      "features1.py:113: FutureWarning: convert_objects is deprecated.  Use the data-type specific converters pd.to_datetime, pd.to_timedelta and pd.to_numeric.\n",
      "  df['TicketNum'] = df['TicketNum'].convert_objects(convert_numeric=True)\n"
     ]
    }
   ],
   "source": [
    "# Fill age, fare, embarked\n",
    "total=fill_age(total)\n",
    "total=fill_fare(total)\n",
    "total=fill_embarked(total)\n",
    "total=fill_cabin(total)\n",
    "total=fill_ticket(total)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "Int64Index: 1309 entries, 0 to 417\n",
      "Data columns (total 18 columns):\n",
      "Age            1309 non-null float64\n",
      "Cabin          1309 non-null object\n",
      "Embarked       1309 non-null object\n",
      "Fare           1309 non-null float64\n",
      "Parch          1309 non-null int64\n",
      "PassengerId    1309 non-null int64\n",
      "Pclass         1309 non-null int64\n",
      "Sex            1309 non-null object\n",
      "SibSp          1309 non-null int64\n",
      "Survived       891 non-null float64\n",
      "Ticket         1309 non-null object\n",
      "Title          1309 non-null object\n",
      "FamilySize     1309 non-null int64\n",
      "Singleton      1309 non-null int64\n",
      "SmallFamily    1309 non-null int64\n",
      "LargeFamily    1309 non-null int64\n",
      "TicketAlpha    1309 non-null object\n",
      "TicketNum      1305 non-null float64\n",
      "dtypes: float64(4), int64(8), object(6)\n",
      "memory usage: 194.3+ KB\n"
     ]
    }
   ],
   "source": [
    "total.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# Turn binary (titles) \n",
    "total = pd.concat([total,pd.get_dummies(total['Title'],prefix='Title')],axis=1)\n",
    "total = pd.concat([total,pd.get_dummies(total['Embarked'],prefix='Embarked')],axis=1)\n",
    "total = pd.concat([total,pd.get_dummies(total['Cabin'], prefix='Cabin')], axis=1)\n",
    "total = pd.concat([total,pd.get_dummies(total['Pclass'], prefix=\"Pclass\")],axis=1) \n",
    "total['Sex'] = total['Sex'].map({'male':1,'female':0})\n",
    "total = pd.concat([total, pd.get_dummies(total['Ticket'], prefix='Ticket')], axis=1)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# Drop\n",
    "total.drop('Title',axis=1,inplace=True)\n",
    "total.drop('Embarked',axis=1,inplace=True)\n",
    "total.drop('Cabin', axis=1, inplace=True)\n",
    "total.drop('Pclass',axis=1,inplace=True)\n",
    "total.drop('Ticket', inplace=True, axis=1)\n",
    "total.drop('PassengerId', inplace=True, axis=1)\n",
    "total.drop('Survived', axis=1, inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "collapsed": true,
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "data_y=train['Survived']\n",
    "data_x=total.iloc[range(len(data_y))]\n",
    "test_x=total.iloc[len(data_y):]\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "cv_params = {\n",
    "#    'learning_rate':[0.05,0.1,0.2],\n",
    "#    'subsample':[0.7,0.8,0.9]\n",
    "#    'max_depth':range(8),\n",
    "#    'min_child_weight':range(6),\n",
    "#    'colsample_bytree':[0.7,0.8,0.9],\n",
    "}\n",
    "fixed_params = {\n",
    "    'max_depth':2,\n",
    "    'min_chil_weight':1,\n",
    "    'learning_rate':0.1,\n",
    "    'n_estimators':1000,\n",
    "    'random_state':seed,\n",
    "    'subsample':0.8,\n",
    "    'colsample_bytree':0.8,\n",
    "    'objective':'binary:logistic',\n",
    "}\n",
    "\n",
    "optimized_gbm = GridSearchCV(xgb.XGBClassifier(**fixed_params),\n",
    "                            cv_params,\n",
    "                            scoring='accuracy',\n",
    "                            cv=5,\n",
    "                            n_jobs=-1)\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "ename": "NameError",
     "evalue": "name 'optimized_gbm' is not defined",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mNameError\u001b[0m                                 Traceback (most recent call last)",
      "\u001b[0;32m<ipython-input-9-a8007ad2500a>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m()\u001b[0m\n\u001b[0;32m----> 1\u001b[0;31m \u001b[0moptimized_gbm\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mfit\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mdata_x\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0mdata_y\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m",
      "\u001b[0;31mNameError\u001b[0m: name 'optimized_gbm' is not defined"
     ]
    }
   ],
   "source": [
    "optimized_gbm.fit(data_x,data_y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[mean: 0.84848, std: 0.02218, params: {}]"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "optimized_gbm.grid_scores_\n",
    "#optimized_gbm.best_params_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "ename": "SyntaxError",
     "evalue": "invalid syntax (<ipython-input-1-b9dd21190486>, line 28)",
     "output_type": "error",
     "traceback": [
      "\u001b[0;36m  File \u001b[0;32m\"<ipython-input-1-b9dd21190486>\"\u001b[0;36m, line \u001b[0;32m28\u001b[0m\n\u001b[0;31m    model = xgb.XGBClassifier().fit.(data_x[train_i],data_y[train_i])\u001b[0m\n\u001b[0m                                    ^\u001b[0m\n\u001b[0;31mSyntaxError\u001b[0m\u001b[0;31m:\u001b[0m invalid syntax\n"
     ]
    }
   ],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "xgb_params = {\n",
    "    'learning_rate':0.1,\n",
    "    'random_state':seed,\n",
    "    'subsample':0.8,\n",
    "    'colsample_bytree':0.8,\n",
    "    'objective':'binary:logistic',\n",
    "    'max_depth':2,\n",
    "    'min_child_weight':1,\n",
    "}\n",
    "final_xgb = xgb.train(xgb_params, xgb_matrix, num_boost_round = 223)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAiUAAAFnCAYAAAB5FTkRAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nGEqIiIhIERhKiIiISBH+BcDUrLEQ0XYCAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f93b0f00a90>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "xgb.plot_importance(final_xgb)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "metadata": {
    "collapsed": true,
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "xgb_params = {\n",
    "    'learning_rate':0.1,\n",
    "    'random_state':seed,\n",
    "    'subsample':0.8,\n",
    "    'colsample_bytree':0.8,\n",
    "    'objective':'binary:logistic',\n",
    "    'max_depth':2,\n",
    "    'min_child_weight':1,\n",
    "    'n_estimators':223,\n",
    "}\n",
    "est_range = [i for i in range(10)]\n",
    "\n",
    "train_scores, test_scores = validation_curve(xgb.XGBClassifier(**xgb_params),\n",
    "                                             data_x,\n",
    "                                             data_y,\n",
    "                                             param_name = 'max_depth',\n",
    "                                             param_range = est_range,\n",
    "                                             cv=5,\n",
    "                                             scoring='accuracy')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA1MAAAIYCAYAAACIULaHAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAIAHxBQAAAAAeEBMAQAAAIAHH7ac+Dl4wcZlAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7ff6f86bc5d0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "train_sm = np.mean(train_scores, axis=1)\n",
    "train_ss = np.std(train_scores, axis=1)\n",
    "test_sm = np.mean(test_scores, axis=1)\n",
    "test_ss = np.std(test_scores, axis=1)\n",
    "\n",
    "fig = plt.figure(figsize=(10,6), dpi=100)\n",
    "\n",
    "plt.title(\"Validation Curve with XGBoost\")\n",
    "plt.xlabel('Parameter')\n",
    "plt.ylabel('Accuracy')\n",
    "plt.ylim(0.7,1.1)\n",
    "\n",
    "plt.plot(est_range, train_sm, label='training score', color='r')\n",
    "plt.plot(est_range, test_sm, label='cross-validation score', color='g')\n",
    "plt.fill_between(est_range, train_sm-train_ss, train_sm+train_ss, alpha=0.2, color='r')\n",
    "plt.fill_between(est_range, test_sm-test_ss, test_sm+test_ss, alpha=0.2, color='g')\n",
    "\n",
    "plt.axhline(y=1, color='k', ls='dashed')\n",
    "\n",
    "plt.legend(loc='best')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>test-auc-mean</th>\n",
       "      <th>test-auc-std</th>\n",
       "      <th>test-error-mean</th>\n",
       "      <th>test-error-std</th>\n",
       "      <th>train-auc-mean</th>\n",
       "      <th>train-auc-std</th>\n",
       "      <th>train-error-mean</th>\n",
       "      <th>train-error-std</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>219</th>\n",
       "      <td>0.881429</td>\n",
       "      <td>0.029387</td>\n",
       "      <td>0.169663</td>\n",
       "      <td>0.022584</td>\n",
       "      <td>0.937817</td>\n",
       "      <td>0.004467</td>\n",
       "      <td>0.114888</td>\n",
       "      <td>0.003597</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>220</th>\n",
       "      <td>0.881700</td>\n",
       "      <td>0.029442</td>\n",
       "      <td>0.170786</td>\n",
       "      <td>0.022075</td>\n",
       "      <td>0.937832</td>\n",
       "      <td>0.004551</td>\n",
       "      <td>0.114607</td>\n",
       "      <td>0.003276</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>221</th>\n",
       "      <td>0.881564</td>\n",
       "      <td>0.029588</td>\n",
       "      <td>0.170786</td>\n",
       "      <td>0.021496</td>\n",
       "      <td>0.937787</td>\n",
       "      <td>0.004632</td>\n",
       "      <td>0.114045</td>\n",
       "      <td>0.003810</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>222</th>\n",
       "      <td>0.881152</td>\n",
       "      <td>0.029578</td>\n",
       "      <td>0.170786</td>\n",
       "      <td>0.021496</td>\n",
       "      <td>0.938046</td>\n",
       "      <td>0.004543</td>\n",
       "      <td>0.114607</td>\n",
       "      <td>0.004833</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>223</th>\n",
       "      <td>0.880948</td>\n",
       "      <td>0.029679</td>\n",
       "      <td>0.168539</td>\n",
       "      <td>0.023835</td>\n",
       "      <td>0.938146</td>\n",
       "      <td>0.004590</td>\n",
       "      <td>0.114326</td>\n",
       "      <td>0.004833</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     test-auc-mean  test-auc-std  test-error-mean  test-error-std  \\\n",
       "219       0.881429      0.029387         0.169663        0.022584   \n",
       "220       0.881700      0.029442         0.170786        0.022075   \n",
       "221       0.881564      0.029588         0.170786        0.021496   \n",
       "222       0.881152      0.029578         0.170786        0.021496   \n",
       "223       0.880948      0.029679         0.168539        0.023835   \n",
       "\n",
       "     train-auc-mean  train-auc-std  train-error-mean  train-error-std  \n",
       "219        0.937817       0.004467          0.114888         0.003597  \n",
       "220        0.937832       0.004551          0.114607         0.003276  \n",
       "221        0.937787       0.004632          0.114045         0.003810  \n",
       "222        0.938046       0.004543          0.114607         0.004833  \n",
       "223        0.938146       0.004590          0.114326         0.004833  "
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "xgb_params = {\n",
    "    'learning_rate':0.1,\n",
    "    'random_state':seed,\n",
    "    'subsample':0.8,\n",
    "    'colsample_bytree':0.8,\n",
    "    'objective':'binary:logistic',\n",
    "    'max_depth':2,\n",
    "    'min_child_weight':1,\n",
    "    'n_estimator':223\n",
    "}\n",
    "xgb_matrix=xgb.DMatrix(data_x, label=data_y)\n",
    "xgb_cv = xgb.cv(params=xgb_params,\n",
    "                dtrain= xgb_matrix,\n",
    "                num_boost_round=3000,\n",
    "                nfold=5,\n",
    "                metrics = ['auc','error'],\n",
    "                early_stopping_rounds=100,\n",
    "#                stratified=True,\n",
    "#                verbose_eval=1,\n",
    "               )\n",
    "\n",
    "\n",
    "model = xgb.XGBClassifier(**xgb_params)\n",
    "model.fit(data_x,data_y)\n",
    "test_predictions = model.predict(test_x)\n",
    "submission = pd.DataFrame({ 'PassengerId': test['PassengerId'],\n",
    "                            'Survived': test_predictions })\n",
    "submission.to_csv(\"submission.csv\", index=False)\n",
    "xgb_cv.tail(5)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "cv",
   "language": "python",
   "name": "cv"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 2
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython2",
   "version": "2.7.12"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}