{"cells":[
 {
  "cell_type": "code",
  "execution_count": null,
  "metadata": {
   "collapsed": false
  },
  "outputs": [],
  "source": "\n\nimport numpy as np # linear algebra\nimport pandas as pd # data processing\n\ntrain = pd.read_csv('../input/train.csv')\ntest = pd.read_csv('../input/test.csv')\n\ntrain.describe()\n"
 },
 {
  "cell_type": "code",
  "execution_count": null,
  "metadata": {
   "collapsed": false
  },
  "outputs": [],
  "source": "print (train.shape)\nprint (test.shape)\nprint (\"\\n train:\\n \",train.head(),\"\\n\")\nprint (\"test: \\n\",test.head())\nprint (\"unique place :\",train.place_id.unique().shape)\n#print (\"place_id   value_count\")\n#print(train.place_id.value_counts())\n\n\n"
 },
 {
  "cell_type": "code",
  "execution_count": null,
  "metadata": {
   "collapsed": false
  },
  "outputs": [],
  "source": "import matplotlib.pyplot as plt\n\ngroups = train.groupby('place_id')\ncount =0\n# Plot\nfig, ax = plt.subplots()\n\nfor name, group in groups:\n    ax.plot(group.x, group.y, marker='o', linestyle='', ms=12, label=name)\n    count =count+1\n    if count>10:\n        break\nax.legend()\n\nplt.show()"
 },
 {
  "cell_type": "code",
  "execution_count": null,
  "metadata": {
   "collapsed": false
  },
  "outputs": [],
  "source": "from sklearn.cross_validation import train_test_split\nfrom sklearn.ensemble import RandomForestClassifier\nfrom sklearn.metrics import roc_auc_score\nfrom sklearn.pipeline import Pipeline\nfrom sklearn.svm import OneClassSVM"
 },
 {
  "cell_type": "code",
  "execution_count": null,
  "metadata": {
   "collapsed": false
  },
  "outputs": [],
  "source": "y_train = train['place_id'].values\nX_train = train.drop(['row_id','place_id'], axis=1).values\nX_fit, X_eval, y_fit, y_eval= train_test_split(X_train, y_train, test_size=0.3)\n"
 },
 {
  "cell_type": "code",
  "execution_count": null,
  "metadata": {
   "collapsed": false
  },
  "outputs": [],
  "source": "id_test = test['row_id']\nX_test = test.drop(['row_id'], axis=1).values"
 },
 {
  "cell_type": "code",
  "execution_count": null,
  "metadata": {
   "collapsed": false
  },
  "outputs": [],
  "source": "clf2=RandomForestClassifier()\nclf2.fit(X_train, y_train)\nprint('Overall AUC:', roc_auc_score(y_train, clf2.predict_proba(X_train)[:,1]))\n\ny_pred2= clf2.predict_proba(X_test)[:,1]"
 },
 {
  "cell_type": "code",
  "execution_count": null,
  "metadata": {
   "collapsed": false
  },
  "outputs": [],
  "source": "submission = pd.DataFrame({\"row_id\":id_test, \"place_id\":y_pred2})\nsubmission.to_csv(\"submission.csv\", index=False)\n\n"
 }
],"metadata":{"kernelspec":{"display_name":"Python 3","language":"python","name":"python3"}}, "nbformat": 4, "nbformat_minor": 0}