gistfile1.txt

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 154,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Populating the interactive namespace from numpy and matplotlib\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING: pylab import has clobbered these variables: ['axes', 'grid']\n",
      "`%matplotlib` prevents importing * from pylab and numpy\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div style=\"max-height:1000px;max-width:1500px;overflow:auto;\">\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>12TH_0</th>\n",
       "      <th>12TH_1</th>\n",
       "      <th>12TH_2</th>\n",
       "      <th>19TH_0</th>\n",
       "      <th>19TH_1</th>\n",
       "      <th>19TH_2</th>\n",
       "      <th>CIVC_0</th>\n",
       "      <th>CIVC_1</th>\n",
       "      <th>CIVC_2</th>\n",
       "      <th>EMBR_0</th>\n",
       "      <th>...</th>\n",
       "      <th>POWL_0</th>\n",
       "      <th>POWL_1</th>\n",
       "      <th>POWL_2</th>\n",
       "      <th>ROCK_0</th>\n",
       "      <th>ROCK_1</th>\n",
       "      <th>ROCK_2</th>\n",
       "      <th>WOAK_0</th>\n",
       "      <th>WOAK_1</th>\n",
       "      <th>WOAK_2</th>\n",
       "      <th>time</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>time</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2015-03-02 11:48:04</th>\n",
       "      <td>11</td>\n",
       "      <td>26</td>\n",
       "      <td>41</td>\n",
       "      <td>10</td>\n",
       "      <td>25</td>\n",
       "      <td>40</td>\n",
       "      <td>0</td>\n",
       "      <td>12</td>\n",
       "      <td>27</td>\n",
       "      <td>8</td>\n",
       "      <td>...</td>\n",
       "      <td>11</td>\n",
       "      <td>26</td>\n",
       "      <td>41</td>\n",
       "      <td>3</td>\n",
       "      <td>18</td>\n",
       "      <td>33</td>\n",
       "      <td>0</td>\n",
       "      <td>16</td>\n",
       "      <td>31</td>\n",
       "      <td>2015-03-02 11:48:04</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-03-02 11:48:34</th>\n",
       "      <td>11</td>\n",
       "      <td>26</td>\n",
       "      <td>41</td>\n",
       "      <td>9</td>\n",
       "      <td>24</td>\n",
       "      <td>39</td>\n",
       "      <td>12</td>\n",
       "      <td>26</td>\n",
       "      <td>41</td>\n",
       "      <td>7</td>\n",
       "      <td>...</td>\n",
       "      <td>11</td>\n",
       "      <td>25</td>\n",
       "      <td>40</td>\n",
       "      <td>2</td>\n",
       "      <td>18</td>\n",
       "      <td>33</td>\n",
       "      <td>0</td>\n",
       "      <td>15</td>\n",
       "      <td>30</td>\n",
       "      <td>2015-03-02 11:48:34</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-03-02 11:49:04</th>\n",
       "      <td>10</td>\n",
       "      <td>25</td>\n",
       "      <td>40</td>\n",
       "      <td>9</td>\n",
       "      <td>24</td>\n",
       "      <td>39</td>\n",
       "      <td>11</td>\n",
       "      <td>26</td>\n",
       "      <td>41</td>\n",
       "      <td>6</td>\n",
       "      <td>...</td>\n",
       "      <td>10</td>\n",
       "      <td>25</td>\n",
       "      <td>40</td>\n",
       "      <td>2</td>\n",
       "      <td>18</td>\n",
       "      <td>32</td>\n",
       "      <td>15</td>\n",
       "      <td>30</td>\n",
       "      <td>45</td>\n",
       "      <td>2015-03-02 11:49:04</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-03-02 11:49:34</th>\n",
       "      <td>10</td>\n",
       "      <td>25</td>\n",
       "      <td>40</td>\n",
       "      <td>8</td>\n",
       "      <td>24</td>\n",
       "      <td>38</td>\n",
       "      <td>11</td>\n",
       "      <td>25</td>\n",
       "      <td>40</td>\n",
       "      <td>6</td>\n",
       "      <td>...</td>\n",
       "      <td>9</td>\n",
       "      <td>24</td>\n",
       "      <td>39</td>\n",
       "      <td>1</td>\n",
       "      <td>17</td>\n",
       "      <td>32</td>\n",
       "      <td>14</td>\n",
       "      <td>29</td>\n",
       "      <td>44</td>\n",
       "      <td>2015-03-02 11:49:34</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-03-02 11:50:06</th>\n",
       "      <td>9</td>\n",
       "      <td>24</td>\n",
       "      <td>39</td>\n",
       "      <td>8</td>\n",
       "      <td>23</td>\n",
       "      <td>38</td>\n",
       "      <td>10</td>\n",
       "      <td>25</td>\n",
       "      <td>40</td>\n",
       "      <td>5</td>\n",
       "      <td>...</td>\n",
       "      <td>9</td>\n",
       "      <td>24</td>\n",
       "      <td>39</td>\n",
       "      <td>0</td>\n",
       "      <td>16</td>\n",
       "      <td>31</td>\n",
       "      <td>14</td>\n",
       "      <td>29</td>\n",
       "      <td>44</td>\n",
       "      <td>2015-03-02 11:50:06</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 28 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                     12TH_0  12TH_1  12TH_2  19TH_0  19TH_1  19TH_2  CIVC_0  \\\n",
       "time                                                                          \n",
       "2015-03-02 11:48:04      11      26      41      10      25      40       0   \n",
       "2015-03-02 11:48:34      11      26      41       9      24      39      12   \n",
       "2015-03-02 11:49:04      10      25      40       9      24      39      11   \n",
       "2015-03-02 11:49:34      10      25      40       8      24      38      11   \n",
       "2015-03-02 11:50:06       9      24      39       8      23      38      10   \n",
       "\n",
       "                     CIVC_1  CIVC_2  EMBR_0         ...           POWL_0  \\\n",
       "time                                                ...                    \n",
       "2015-03-02 11:48:04      12      27       8         ...               11   \n",
       "2015-03-02 11:48:34      26      41       7         ...               11   \n",
       "2015-03-02 11:49:04      26      41       6         ...               10   \n",
       "2015-03-02 11:49:34      25      40       6         ...                9   \n",
       "2015-03-02 11:50:06      25      40       5         ...                9   \n",
       "\n",
       "                     POWL_1  POWL_2  ROCK_0  ROCK_1  ROCK_2  WOAK_0  WOAK_1  \\\n",
       "time                                                                          \n",
       "2015-03-02 11:48:04      26      41       3      18      33       0      16   \n",
       "2015-03-02 11:48:34      25      40       2      18      33       0      15   \n",
       "2015-03-02 11:49:04      25      40       2      18      32      15      30   \n",
       "2015-03-02 11:49:34      24      39       1      17      32      14      29   \n",
       "2015-03-02 11:50:06      24      39       0      16      31      14      29   \n",
       "\n",
       "                     WOAK_2                 time  \n",
       "time                                              \n",
       "2015-03-02 11:48:04      31  2015-03-02 11:48:04  \n",
       "2015-03-02 11:48:34      30  2015-03-02 11:48:34  \n",
       "2015-03-02 11:49:04      45  2015-03-02 11:49:04  \n",
       "2015-03-02 11:49:34      44  2015-03-02 11:49:34  \n",
       "2015-03-02 11:50:06      44  2015-03-02 11:50:06  \n",
       "\n",
       "[5 rows x 28 columns]"
      ]
     },
     "execution_count": 154,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import datetime\n",
    "import seaborn as sns\n",
    "from matplotlib import pyplot as plt\n",
    "\n",
    "%pylab inline\n",
    "sns.set_style('darkgrid')\n",
    "\n",
    "df = pd.read_csv('download1.csv')\n",
    "del df['Unnamed: 0']\n",
    "df.index = pd.to_datetime(df.time)\n",
    "df = df[pd.isnull(df.index) == False]\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 155,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "commute_df = df[(df.index.map(lambda x: x.time) >= datetime.time(6, 0, 0)) & (df.index.map(lambda x: x.time) < datetime.time(10, 0, 0))]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 156,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "def get_arrival_times(df):\n",
    "    output = dict()\n",
    "    cols = [c for c in df.columns if c[-1] == '0']\n",
    "    for col in cols:\n",
    "#         temp0 = df[df[col] == 0].index.tolist()\n",
    "#         temp1 = (df[df[col] == 0].index - datetime.timedelta(minutes=1)).tolist()\n",
    "#         temp = temp0 + temp1\n",
    "#         temp.sort()\n",
    "#         temp = pd.Series(temp)\n",
    "        temp = pd.Series(df[df[col] == 0].index.tolist())\n",
    "        diffs = pd.Series(np.diff(temp), index=np.arange(1, len(temp)))\n",
    "        first_arrivals = temp[1:][diffs >= datetime.timedelta(seconds=60)]\n",
    "        output[col] = pd.Series([temp[0]] + first_arrivals.tolist())\n",
    "        \n",
    "        \n",
    "    return output"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 157,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div style=\"max-height:1000px;max-width:1500px;overflow:auto;\">\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>12TH_0</th>\n",
       "      <th>19TH_0</th>\n",
       "      <th>CIVC_0</th>\n",
       "      <th>EMBR_0</th>\n",
       "      <th>MCAR_0</th>\n",
       "      <th>MONT_0</th>\n",
       "      <th>POWL_0</th>\n",
       "      <th>ROCK_0</th>\n",
       "      <th>WOAK_0</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2015-03-03 06:13:19</td>\n",
       "      <td>2015-03-03 06:11:49</td>\n",
       "      <td>2015-03-03 06:14:34</td>\n",
       "      <td>2015-03-03 06:09:49</td>\n",
       "      <td>2015-03-03 06:08:04</td>\n",
       "      <td>2015-03-03 06:11:19</td>\n",
       "      <td>2015-03-03 06:12:49</td>\n",
       "      <td>2015-03-03 06:04:49</td>\n",
       "      <td>2015-03-03 06:03:04</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2015-03-03 06:28:34</td>\n",
       "      <td>2015-03-03 06:27:34</td>\n",
       "      <td>2015-03-03 06:30:07</td>\n",
       "      <td>2015-03-03 06:24:34</td>\n",
       "      <td>2015-03-03 06:23:04</td>\n",
       "      <td>2015-03-03 06:26:19</td>\n",
       "      <td>2015-03-03 06:28:04</td>\n",
       "      <td>2015-03-03 06:20:07</td>\n",
       "      <td>2015-03-03 06:18:04</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2015-03-03 06:43:22</td>\n",
       "      <td>2015-03-03 06:42:04</td>\n",
       "      <td>2015-03-03 06:44:34</td>\n",
       "      <td>2015-03-03 06:40:07</td>\n",
       "      <td>2015-03-03 06:38:34</td>\n",
       "      <td>2015-03-03 06:41:34</td>\n",
       "      <td>2015-03-03 06:43:04</td>\n",
       "      <td>2015-03-03 06:34:49</td>\n",
       "      <td>2015-03-03 06:32:52</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2015-03-03 07:03:34</td>\n",
       "      <td>2015-03-03 07:01:34</td>\n",
       "      <td>2015-03-03 06:59:34</td>\n",
       "      <td>2015-03-03 06:55:06</td>\n",
       "      <td>2015-03-03 06:56:34</td>\n",
       "      <td>2015-03-03 06:56:34</td>\n",
       "      <td>2015-03-03 06:58:04</td>\n",
       "      <td>2015-03-03 06:53:04</td>\n",
       "      <td>2015-03-03 06:47:49</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2015-03-03 07:13:34</td>\n",
       "      <td>2015-03-03 07:12:34</td>\n",
       "      <td>2015-03-03 07:20:07</td>\n",
       "      <td>2015-03-03 07:15:06</td>\n",
       "      <td>2015-03-03 07:08:04</td>\n",
       "      <td>2015-03-03 07:16:34</td>\n",
       "      <td>2015-03-03 07:18:04</td>\n",
       "      <td>2015-03-03 07:04:34</td>\n",
       "      <td>2015-03-03 07:08:04</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "               12TH_0              19TH_0              CIVC_0  \\\n",
       "0 2015-03-03 06:13:19 2015-03-03 06:11:49 2015-03-03 06:14:34   \n",
       "1 2015-03-03 06:28:34 2015-03-03 06:27:34 2015-03-03 06:30:07   \n",
       "2 2015-03-03 06:43:22 2015-03-03 06:42:04 2015-03-03 06:44:34   \n",
       "3 2015-03-03 07:03:34 2015-03-03 07:01:34 2015-03-03 06:59:34   \n",
       "4 2015-03-03 07:13:34 2015-03-03 07:12:34 2015-03-03 07:20:07   \n",
       "\n",
       "               EMBR_0              MCAR_0              MONT_0  \\\n",
       "0 2015-03-03 06:09:49 2015-03-03 06:08:04 2015-03-03 06:11:19   \n",
       "1 2015-03-03 06:24:34 2015-03-03 06:23:04 2015-03-03 06:26:19   \n",
       "2 2015-03-03 06:40:07 2015-03-03 06:38:34 2015-03-03 06:41:34   \n",
       "3 2015-03-03 06:55:06 2015-03-03 06:56:34 2015-03-03 06:56:34   \n",
       "4 2015-03-03 07:15:06 2015-03-03 07:08:04 2015-03-03 07:16:34   \n",
       "\n",
       "               POWL_0              ROCK_0              WOAK_0  \n",
       "0 2015-03-03 06:12:49 2015-03-03 06:04:49 2015-03-03 06:03:04  \n",
       "1 2015-03-03 06:28:04 2015-03-03 06:20:07 2015-03-03 06:18:04  \n",
       "2 2015-03-03 06:43:04 2015-03-03 06:34:49 2015-03-03 06:32:52  \n",
       "3 2015-03-03 06:58:04 2015-03-03 06:53:04 2015-03-03 06:47:49  \n",
       "4 2015-03-03 07:18:04 2015-03-03 07:04:34 2015-03-03 07:08:04  "
      ]
     },
     "execution_count": 157,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "temp = pd.DataFrame(get_arrival_times(commute_df))\n",
    "temp.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 158,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "output = []\n",
    "cols_in_order = ['ROCK_0', 'MCAR_0', '19TH_0', '12TH_0', 'WOAK_0', 'EMBR_0', 'MONT_0', 'POWL_0', 'CIVC_0']\n",
    "\n",
    "min_diff_seconds = {\n",
    "    'MCAR_0': 60,\n",
    "    '19TH_0': 60,\n",
    "    '12TH_0': 30,\n",
    "    'WOAK_0': 30,\n",
    "    'EMBR_0': 300,\n",
    "    'MONT_0': 30,\n",
    "    'POWL_0': 30,\n",
    "    'CIVC_0': 30,\n",
    "}\n",
    "\n",
    "for x in temp.dropna().ROCK_0:\n",
    "    if x.hour >= 9:\n",
    "        continue\n",
    "    this_row = dict()\n",
    "    this_row['ROCK_0'] = x\n",
    "    last_time = x\n",
    "    for col in cols_in_order[1:]:\n",
    "        next_time = temp[temp[col] >= last_time + datetime.timedelta(seconds=min_diff_seconds[col])][col].iloc[0]\n",
    "        this_row[col] = next_time\n",
    "        last_time = next_time\n",
    "        \n",
    "    output.append(this_row)\n",
    "all_trains = pd.DataFrame(output, columns=cols_in_order)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 159,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.text.Text at 0x13436bf90>"
      ]
     },
     "execution_count": 159,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAlsAAAFRCAYAAABDtmiKAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAABCYgsAICS2AABCYgsAICS2AABCH0VQYFeeSfOVAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1344fe150>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.distplot((all_trains['CIVC_0'] - all_trains['ROCK_0']) / np.timedelta64(1, 'm'), bins=np.arange(20, 51))\n",
    "plt.gcf().set_size_inches(10, 5)\n",
    "plt.xlabel('Trip Duration (Minutes)')\n",
    "plt.title('Distribution of Trip Lengths')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 160,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "whole_trip = (all_trains['CIVC_0'] - all_trains['ROCK_0']) / np.timedelta64(1, 'm')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 161,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "20 0.000\n",
      "21 0.000\n",
      "22 0.004\n",
      "23 0.045\n",
      "24 0.152\n",
      "25 0.386\n",
      "26 0.591\n",
      "27 0.750\n",
      "28 0.835\n",
      "29 0.894\n",
      "30 0.934\n",
      "31 0.957\n",
      "32 0.973\n",
      "33 0.981\n",
      "34 0.986\n",
      "35 0.991\n",
      "36 0.995\n",
      "37 0.996\n",
      "38 0.996\n",
      "39 0.996\n",
      "40 0.997\n",
      "41 0.997\n",
      "42 0.999\n",
      "43 0.999\n",
      "44 0.999\n",
      "45 0.999\n",
      "46 1.000\n",
      "47 1.000\n",
      "48 1.000\n",
      "49 1.000\n"
     ]
    }
   ],
   "source": [
    "for i in range(20, 50):\n",
    "    print i, \"{:.3f}\".format((whole_trip <= i).mean())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 162,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "25.716666666666665"
      ]
     },
     "execution_count": 162,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "whole_trip.median()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 163,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYEAAAECCAYAAAAYfWtSAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nfaI95yarAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1344fe090>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "for _, row in all_trains.iterrows():\n",
    "    pd.Series(np.diff(row) / np.timedelta64(1, 'm'), index=cols_in_order[1:]).plot(color='k', alpha=0.5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 164,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div style=\"max-height:1000px;max-width:1500px;overflow:auto;\">\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>ROCK_0</th>\n",
       "      <th>MCAR_0</th>\n",
       "      <th>19TH_0</th>\n",
       "      <th>12TH_0</th>\n",
       "      <th>WOAK_0</th>\n",
       "      <th>EMBR_0</th>\n",
       "      <th>MONT_0</th>\n",
       "      <th>POWL_0</th>\n",
       "      <th>CIVC_0</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>435</th>\n",
       "      <td>2015-04-14 07:05:19</td>\n",
       "      <td>2015-04-14 07:08:19</td>\n",
       "      <td>2015-04-14 07:12:19</td>\n",
       "      <td>2015-04-14 07:13:20</td>\n",
       "      <td>2015-04-14 07:24:49</td>\n",
       "      <td>2015-04-14 07:34:19</td>\n",
       "      <td>2015-04-14 07:36:22</td>\n",
       "      <td>2015-04-14 07:37:49</td>\n",
       "      <td>2015-04-14 07:39:19</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>477</th>\n",
       "      <td>2015-04-17 08:35:34</td>\n",
       "      <td>2015-04-17 08:39:07</td>\n",
       "      <td>2015-04-17 08:45:06</td>\n",
       "      <td>2015-04-17 08:46:34</td>\n",
       "      <td>2015-04-17 09:08:34</td>\n",
       "      <td>2015-04-17 09:16:04</td>\n",
       "      <td>2015-04-17 09:18:04</td>\n",
       "      <td>2015-04-17 09:19:34</td>\n",
       "      <td>2015-04-17 09:21:04</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>569</th>\n",
       "      <td>2015-04-27 07:05:22</td>\n",
       "      <td>2015-04-27 07:08:19</td>\n",
       "      <td>2015-04-27 07:11:51</td>\n",
       "      <td>2015-04-27 07:13:19</td>\n",
       "      <td>2015-04-27 07:23:19</td>\n",
       "      <td>2015-04-27 07:31:19</td>\n",
       "      <td>2015-04-27 07:33:19</td>\n",
       "      <td>2015-04-27 07:35:34</td>\n",
       "      <td>2015-04-27 07:37:34</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>571</th>\n",
       "      <td>2015-04-27 07:36:04</td>\n",
       "      <td>2015-04-27 07:39:19</td>\n",
       "      <td>2015-04-27 07:42:49</td>\n",
       "      <td>2015-04-27 07:44:04</td>\n",
       "      <td>2015-04-27 07:57:19</td>\n",
       "      <td>2015-04-27 08:05:22</td>\n",
       "      <td>2015-04-27 08:06:49</td>\n",
       "      <td>2015-04-27 08:08:49</td>\n",
       "      <td>2015-04-27 08:10:19</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>628</th>\n",
       "      <td>2015-05-02 07:18:48</td>\n",
       "      <td>2015-05-02 07:21:52</td>\n",
       "      <td>2015-05-02 07:25:19</td>\n",
       "      <td>2015-05-02 07:26:18</td>\n",
       "      <td>2015-05-02 07:46:03</td>\n",
       "      <td>2015-05-02 07:53:18</td>\n",
       "      <td>2015-05-02 07:54:33</td>\n",
       "      <td>2015-05-02 07:56:03</td>\n",
       "      <td>2015-05-02 07:58:03</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>693</th>\n",
       "      <td>2015-05-08 07:36:04</td>\n",
       "      <td>2015-05-08 07:39:04</td>\n",
       "      <td>2015-05-08 07:42:34</td>\n",
       "      <td>2015-05-08 07:44:04</td>\n",
       "      <td>2015-05-08 07:54:49</td>\n",
       "      <td>2015-05-08 08:02:37</td>\n",
       "      <td>2015-05-08 08:04:19</td>\n",
       "      <td>2015-05-08 08:06:19</td>\n",
       "      <td>2015-05-08 08:07:49</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                 ROCK_0              MCAR_0              19TH_0  \\\n",
       "435 2015-04-14 07:05:19 2015-04-14 07:08:19 2015-04-14 07:12:19   \n",
       "477 2015-04-17 08:35:34 2015-04-17 08:39:07 2015-04-17 08:45:06   \n",
       "569 2015-04-27 07:05:22 2015-04-27 07:08:19 2015-04-27 07:11:51   \n",
       "571 2015-04-27 07:36:04 2015-04-27 07:39:19 2015-04-27 07:42:49   \n",
       "628 2015-05-02 07:18:48 2015-05-02 07:21:52 2015-05-02 07:25:19   \n",
       "693 2015-05-08 07:36:04 2015-05-08 07:39:04 2015-05-08 07:42:34   \n",
       "\n",
       "                 12TH_0              WOAK_0              EMBR_0  \\\n",
       "435 2015-04-14 07:13:20 2015-04-14 07:24:49 2015-04-14 07:34:19   \n",
       "477 2015-04-17 08:46:34 2015-04-17 09:08:34 2015-04-17 09:16:04   \n",
       "569 2015-04-27 07:13:19 2015-04-27 07:23:19 2015-04-27 07:31:19   \n",
       "571 2015-04-27 07:44:04 2015-04-27 07:57:19 2015-04-27 08:05:22   \n",
       "628 2015-05-02 07:26:18 2015-05-02 07:46:03 2015-05-02 07:53:18   \n",
       "693 2015-05-08 07:44:04 2015-05-08 07:54:49 2015-05-08 08:02:37   \n",
       "\n",
       "                 MONT_0              POWL_0              CIVC_0  \n",
       "435 2015-04-14 07:36:22 2015-04-14 07:37:49 2015-04-14 07:39:19  \n",
       "477 2015-04-17 09:18:04 2015-04-17 09:19:34 2015-04-17 09:21:04  \n",
       "569 2015-04-27 07:33:19 2015-04-27 07:35:34 2015-04-27 07:37:34  \n",
       "571 2015-04-27 08:06:49 2015-04-27 08:08:49 2015-04-27 08:10:19  \n",
       "628 2015-05-02 07:54:33 2015-05-02 07:56:03 2015-05-02 07:58:03  \n",
       "693 2015-05-08 08:04:19 2015-05-08 08:06:19 2015-05-08 08:07:49  "
      ]
     },
     "execution_count": 164,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "all_trains[((all_trains['WOAK_0'] - all_trains['12TH_0']) / np.timedelta64(1, 'm')) >= 10]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 165,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div style=\"max-height:1000px;max-width:1500px;overflow:auto;\">\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>ROCK_0</th>\n",
       "      <th>MCAR_0</th>\n",
       "      <th>19TH_0</th>\n",
       "      <th>12TH_0</th>\n",
       "      <th>WOAK_0</th>\n",
       "      <th>EMBR_0</th>\n",
       "      <th>MONT_0</th>\n",
       "      <th>POWL_0</th>\n",
       "      <th>CIVC_0</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>433</th>\n",
       "      <td>2015-04-14 06:44:19</td>\n",
       "      <td>2015-04-14 06:47:19</td>\n",
       "      <td>2015-04-14 06:50:19</td>\n",
       "      <td>2015-04-14 06:51:49</td>\n",
       "      <td>2015-04-14 06:56:22</td>\n",
       "      <td>2015-04-14 07:03:19</td>\n",
       "      <td>2015-04-14 07:05:19</td>\n",
       "      <td>2015-04-14 07:16:49</td>\n",
       "      <td>2015-04-14 07:18:22</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>434</th>\n",
       "      <td>2015-04-14 06:50:19</td>\n",
       "      <td>2015-04-14 06:53:20</td>\n",
       "      <td>2015-04-14 06:57:22</td>\n",
       "      <td>2015-04-14 06:58:49</td>\n",
       "      <td>2015-04-14 07:04:19</td>\n",
       "      <td>2015-04-14 07:20:49</td>\n",
       "      <td>2015-04-14 07:22:52</td>\n",
       "      <td>2015-04-14 07:24:19</td>\n",
       "      <td>2015-04-14 07:25:49</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>435</th>\n",
       "      <td>2015-04-14 07:05:19</td>\n",
       "      <td>2015-04-14 07:08:19</td>\n",
       "      <td>2015-04-14 07:12:19</td>\n",
       "      <td>2015-04-14 07:13:20</td>\n",
       "      <td>2015-04-14 07:24:49</td>\n",
       "      <td>2015-04-14 07:34:19</td>\n",
       "      <td>2015-04-14 07:36:22</td>\n",
       "      <td>2015-04-14 07:37:49</td>\n",
       "      <td>2015-04-14 07:39:19</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>436</th>\n",
       "      <td>2015-04-14 07:22:52</td>\n",
       "      <td>2015-04-14 07:25:49</td>\n",
       "      <td>2015-04-14 07:28:49</td>\n",
       "      <td>2015-04-14 07:30:49</td>\n",
       "      <td>2015-04-14 07:34:49</td>\n",
       "      <td>2015-04-14 07:45:19</td>\n",
       "      <td>2015-04-14 07:47:22</td>\n",
       "      <td>2015-04-14 07:49:49</td>\n",
       "      <td>2015-04-14 07:51:19</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>437</th>\n",
       "      <td>2015-04-14 07:36:22</td>\n",
       "      <td>2015-04-14 07:39:19</td>\n",
       "      <td>2015-04-14 07:43:49</td>\n",
       "      <td>2015-04-14 07:45:19</td>\n",
       "      <td>2015-04-14 07:49:49</td>\n",
       "      <td>2015-04-14 07:59:19</td>\n",
       "      <td>2015-04-14 08:01:22</td>\n",
       "      <td>2015-04-14 08:03:51</td>\n",
       "      <td>2015-04-14 08:05:19</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                 ROCK_0              MCAR_0              19TH_0  \\\n",
       "433 2015-04-14 06:44:19 2015-04-14 06:47:19 2015-04-14 06:50:19   \n",
       "434 2015-04-14 06:50:19 2015-04-14 06:53:20 2015-04-14 06:57:22   \n",
       "435 2015-04-14 07:05:19 2015-04-14 07:08:19 2015-04-14 07:12:19   \n",
       "436 2015-04-14 07:22:52 2015-04-14 07:25:49 2015-04-14 07:28:49   \n",
       "437 2015-04-14 07:36:22 2015-04-14 07:39:19 2015-04-14 07:43:49   \n",
       "\n",
       "                 12TH_0              WOAK_0              EMBR_0  \\\n",
       "433 2015-04-14 06:51:49 2015-04-14 06:56:22 2015-04-14 07:03:19   \n",
       "434 2015-04-14 06:58:49 2015-04-14 07:04:19 2015-04-14 07:20:49   \n",
       "435 2015-04-14 07:13:20 2015-04-14 07:24:49 2015-04-14 07:34:19   \n",
       "436 2015-04-14 07:30:49 2015-04-14 07:34:49 2015-04-14 07:45:19   \n",
       "437 2015-04-14 07:45:19 2015-04-14 07:49:49 2015-04-14 07:59:19   \n",
       "\n",
       "                 MONT_0              POWL_0              CIVC_0  \n",
       "433 2015-04-14 07:05:19 2015-04-14 07:16:49 2015-04-14 07:18:22  \n",
       "434 2015-04-14 07:22:52 2015-04-14 07:24:19 2015-04-14 07:25:49  \n",
       "435 2015-04-14 07:36:22 2015-04-14 07:37:49 2015-04-14 07:39:19  \n",
       "436 2015-04-14 07:47:22 2015-04-14 07:49:49 2015-04-14 07:51:19  \n",
       "437 2015-04-14 08:01:22 2015-04-14 08:03:51 2015-04-14 08:05:19  "
      ]
     },
     "execution_count": 165,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "all_trains.iloc[433:438]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 166,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "586    2015-04-14 07:24:49\n",
       "587    2015-04-14 07:34:49\n",
       "588    2015-04-14 07:49:49\n",
       "589    2015-04-14 08:05:49\n",
       "590    2015-04-14 08:19:19\n",
       "591    2015-04-14 08:35:49\n",
       "592    2015-04-14 08:53:49\n",
       "593    2015-04-14 09:05:19\n",
       "594    2015-04-14 09:20:19\n",
       "595    2015-04-14 09:32:22\n",
       "596    2015-04-14 09:47:22\n",
       "597    2015-04-15 06:02:49\n",
       "598    2015-04-15 06:18:19\n",
       "599    2015-04-15 06:32:49\n",
       "600    2015-04-15 06:48:49\n",
       "601    2015-04-15 07:03:49\n",
       "602    2015-04-15 07:17:49\n",
       "603    2015-04-15 07:32:49\n",
       "604    2015-04-15 07:48:19\n",
       "605    2015-04-15 08:02:52\n",
       "606    2015-04-15 08:20:22\n",
       "607    2015-04-15 08:33:49\n",
       "608    2015-04-15 08:51:49\n",
       "609    2015-04-15 09:03:04\n",
       "610    2015-04-15 09:18:34\n",
       "611    2015-04-15 09:34:07\n",
       "612    2015-04-15 09:48:19\n",
       "613    2015-04-16 06:03:04\n",
       "614    2015-04-16 06:17:49\n",
       "615    2015-04-16 06:33:04\n",
       "               ...        \n",
       "1047   2015-05-16 09:03:18\n",
       "1048   2015-05-16 09:11:18\n",
       "1049   2015-05-16 09:37:19\n",
       "1050   2015-05-16 09:51:48\n",
       "1051   2015-05-17 06:49:48\n",
       "1052   2015-05-17 07:05:52\n",
       "1053   2015-05-17 07:25:22\n",
       "1054   2015-05-17 07:44:49\n",
       "1055   2015-05-17 08:05:18\n",
       "1056   2015-05-17 08:29:22\n",
       "1057   2015-05-17 08:45:18\n",
       "1058   2015-05-17 09:05:18\n",
       "1059   2015-05-17 09:27:03\n",
       "1060   2015-05-17 09:45:34\n",
       "1061   2015-05-18 06:02:49\n",
       "1062   2015-05-18 06:18:19\n",
       "1063   2015-05-18 06:32:49\n",
       "1064   2015-05-18 06:47:49\n",
       "1065   2015-05-18 07:03:49\n",
       "1066   2015-05-18 07:18:49\n",
       "1067   2015-05-18 07:33:04\n",
       "1068   2015-05-18 07:50:19\n",
       "1069   2015-05-18 08:04:34\n",
       "1070   2015-05-18 08:25:06\n",
       "1071   2015-05-18 08:34:52\n",
       "1072   2015-05-18 08:41:04\n",
       "1073   2015-05-18 08:54:37\n",
       "1074   2015-05-18 09:05:34\n",
       "1075   2015-05-18 09:18:34\n",
       "1076   2015-05-18 09:35:37\n",
       "Name: WOAK_0, dtype: datetime64[ns]"
      ]
     },
     "execution_count": 166,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "temp[temp['WOAK_0'] >= datetime.datetime(2015, 4, 14, 7, 13)]['WOAK_0']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 167,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "time\n",
       "2015-04-17 08:46:04     6\n",
       "2015-04-17 08:46:34     5\n",
       "2015-04-17 08:47:07     5\n",
       "2015-04-17 08:47:34     4\n",
       "2015-04-17 08:48:04     4\n",
       "2015-04-17 08:48:34     4\n",
       "2015-04-17 08:49:07     4\n",
       "2015-04-17 08:49:37     4\n",
       "2015-04-17 08:50:06     4\n",
       "2015-04-17 08:50:34     4\n",
       "2015-04-17 08:51:04     4\n",
       "2015-04-17 08:51:34     4\n",
       "2015-04-17 08:52:07     4\n",
       "2015-04-17 08:52:34     4\n",
       "2015-04-17 08:53:04    11\n",
       "2015-04-17 08:53:34    11\n",
       "2015-04-17 08:54:04    10\n",
       "2015-04-17 08:54:34    10\n",
       "2015-04-17 08:55:06    10\n",
       "2015-04-17 08:55:34     9\n",
       "2015-04-17 08:56:04     9\n",
       "2015-04-17 08:56:34     9\n",
       "2015-04-17 08:57:07     9\n",
       "2015-04-17 08:57:34     9\n",
       "2015-04-17 08:58:04     9\n",
       "2015-04-17 08:58:34     9\n",
       "2015-04-17 08:59:04     9\n",
       "2015-04-17 08:59:34     8\n",
       "2015-04-17 08:59:49     8\n",
       "2015-04-17 09:00:34     7\n",
       "                       ..\n",
       "2015-05-18 21:36:06    10\n",
       "2015-05-18 21:36:33     9\n",
       "2015-05-18 21:37:03     9\n",
       "2015-05-18 21:37:33     8\n",
       "2015-05-18 21:38:03     8\n",
       "2015-05-18 21:38:33     7\n",
       "2015-05-18 21:39:03     7\n",
       "2015-05-18 21:39:33     6\n",
       "2015-05-18 21:40:06     6\n",
       "2015-05-18 21:40:34     5\n",
       "2015-05-18 21:41:03     5\n",
       "2015-05-18 21:41:33     4\n",
       "2015-05-18 21:42:03     3\n",
       "2015-05-18 21:42:33     3\n",
       "2015-05-18 21:43:03     2\n",
       "2015-05-18 21:43:33     2\n",
       "2015-05-18 21:44:03     0\n",
       "2015-05-18 21:44:33     0\n",
       "2015-05-18 21:45:06     0\n",
       "2015-05-18 21:45:33    20\n",
       "2015-05-18 21:46:03    20\n",
       "2015-05-18 21:46:35    19\n",
       "2015-05-18 21:47:03    19\n",
       "2015-05-18 21:47:37    18\n",
       "2015-05-18 21:48:03    18\n",
       "2015-05-18 21:48:37    17\n",
       "2015-05-18 21:49:03    17\n",
       "2015-05-18 21:49:33    16\n",
       "2015-05-18 21:50:06    16\n",
       "2015-05-18 21:50:37    15\n",
       "Name: WOAK_0, dtype: float64"
      ]
     },
     "execution_count": 167,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df[df.index >= datetime.datetime(2015, 4, 17, 8, 46)]['WOAK_0']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 168,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<seaborn.axisgrid.JointGrid at 0x1343f9790>"
      ]
     },
     "execution_count": 168,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAa4AAAGpCAYAAADV8NIiAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1343d2390>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.jointplot(all_trains.ROCK_0.map(lambda x: x.hour + x.minute/60.0), all_trains.CIVC_0.map(lambda x: x.hour + x.minute/60.0))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 169,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABIMAAAJfCAYAAADsJxaZAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAIABJREFUeJzs3W2MZPtdJ/ZvVVc/1GPPvTPje41tbHLHlGwHs8CKYVkE\nykojJSJI2VWkCBGkRGxwsm8QvNggNsurjbISChJSNm+8RpHywquAnETYAvYieS8YwoTIgA2JC48x\n2L6+DzN3Hrqe+vFUXnRPT0/fe6fnoWequ/+fj9SqOudUV/3+fc6p6vqe//mf2mw2CwAAAABlqM+7\nAAAAAACeHWEQAAAAQEGEQQAAAAAFEQYBAAAAFEQYBAAAAFAQYRAAAABAQRoPWtjv9xeSfDLJdyeZ\nJfmvB4PBXx5Y/hNJ/nmS7SS/PhgM/vVTrBUAAACAJ3RUz6D/OEk1GAx+JMl/l+S/v7ug3+8vJvnV\nJFeS/FiSn+33++95WoUCAAAA8OQeGAYNBoP/M8kn9iY/lOTWgcUfSXJtMBjcGQwGW0m+kORHn0aR\nAAAAAByPB54mliSDwWCn3+//L0n+YZL/9MCiXpI7B6aHSVaPtToAAAAAjtWRYVCSDAaD/6Lf7/+3\nSa72+/2PDAaDaXaDoO6Bh3Vzf8+ht5nNZrNarfbYxQIAAADwNo8Uthw1gPRPJ3n/YDD4H5JMk1TZ\nHUg6Sb6S5MP9fv+5JOPsniL2Kw+srFbL9evDR6kPOCYXL3btfzAH9j2YD/sezId9D+bj4sXu0Q86\n4KgBpH8zyd/p9/uvJPmdJD+X5B/2+/3/am+coF9I8rtJ/ijJpwaDwWuPXjIAAAAAz8oDewbtnQ72\nnz1g+WeTfPa4iwIAAADg6TiqZxAAAAAAZ4gwCAAAAKAgwiAAAACAggiDAAAAAAoiDAIAAAAoiDAI\nAAAAoCDCIAAAAICCCIMAAAAACiIMAgAAACiIMAgAAACgIMIgAAAAgIIIgwAAAAAKIgwCAAAAKIgw\nCAAAAKAgwiAAAACAggiDAAAAAAoiDAIAAAAoiDAIAAAAoCDCIAAAAICCCIMAAAAACiIMAgAAACiI\nMAgAAACgIMIgAAAAgIIIgwAAAAAKIgwCAAAAKIgwCAAAAKAgwiAAAACAggiDAAAAAAoiDAIAAAAo\niDAIAAAAoCDCIAAAAICCCIMAAAAACiIMAgAAACiIMAgAAACgIMIgAAAAgIIIgwAAAAAKIgwCAAAA\nKIgwCAAAAKAgwiAAAACAggiDAAAAAAoiDAIAAAAoiDAIAAAAoCDCIAAAAICCCIMAAAAACiIMAgAA\nACiIMAgAAACgIMIgAAAAgIIIgwAAAAAKIgwCAAAAKIgwCAAAAKAgwiAAAACAggiDAAAAAAoiDAIA\nAAAoiDAIAAAAoCDCIAAAAICCCIMAAAAACiIMAgAAACiIMAgAAACgIMIgAAAAgIIIgwAAAAAKIgwC\nAAAAKIgwCAAAAKAgwiAAAACAggiDAAAAAArSmHcBwMlWVVVGo+H+dKfTTb0uRwYAADithEHAA41G\nw7x89VqarXamk3GuXL6UXm913mUBAADwmIRBwJGarXZa7e68ywAAAOAYONcDAAAAoCDCIAAAAICC\nCIMAAAAACiIMAgAAACiIMAgAAACgIMIgAAAAgIIIgwAAAAAKIgwCAAAAKIgwCAAAAKAgwiAAAACA\nggiDAAAAAAoiDAIAAAAoiDAIAAAAoCDCIAAAAICCCIMAAAAACiIMAgAAACiIMAgAAACgIMIgAAAA\ngII0HrSw3+8vJvn1JB9MspzkXwwGg986sPznk/xMkut7sz4xGAz+6inVCgAAAMATemAYlOSnklwf\nDAY/3e/3n0vyZ0l+68Dy70/y04PB4E+fVoEAAAAAHJ+jwqDfSPKbe/frSbYPLf+BJL/U7/dfTPK5\nwWDwL4+5PgAAAACO0QPHDBoMBuPBYDDq9/vd7AZD/+zQQz6d5BNJ/kGSH+n3+z/+dMoEAAAA4Dgc\n1TMo/X7/A0k+k+RfDQaDf3No8a8NBoO1vcd9Lsn3Jfncg57v4sXuY5YKPKnH2f+Wlqp02jfT7qyk\nns1cuNDN6qr9GB6Fzz6YD/sezId9D06+owaQfiHJv03yTwaDwecPLVtN8qV+v//RJJPs9g761FEv\neP368PGrBR7bxYvdx9r/1taGGY03UmU9k/FGbtwYZnPThQjhYT3uvgc8GfsezId9D+bjUUPYo3oG\n/VKS1SS/3O/3f3lv3ieTtAeDwSf7/f4vJvl8ko0kvzcYDH7nEesFAAAA4Bl6YBg0GAx+LsnPPWD5\np7M7bhAAAAAAp4BzPQAAAAAKIgwCAAAAKIgwCAAAAKAgwiAAAACAggiDAAAAAAoiDAIAAAAoiDAI\nAAAAoCDCIAAAAICCCIMAAAAACiIMAgAAACiIMAgAAACgIMIgAAAAgIIIgwAAAAAKIgwCAAAAKEhj\n3gUA8PRUVZXRaHjfvE6nm3rdsQAAACiVMAjgDBuNhnn56rU0W+0kyXQyzpXLl9Lrrc65MgAAYF6E\nQQBnXLPVTqvdnXcZAADACeE8AQAAAICCCIMAAAAACiIMAgAAACiIMAgAAACgIMIgAAAAgIIIgwAA\nAAAKIgwCAAAAKIgwCAAAAKAgwiAAAACAggiDAAAAAAoiDAIAAAAoiDAIAAAAoCCNeRcAME9VVWU0\nGu5Pdzrd1OtycgAA4OwSBgFFG42GefnqtTRb7Uwn41y5fCm93uq8ywIAAHhqhEFA8Zqtdlrt7rzL\nAAAAeCacCwEAAABQEGEQAAAAQEGEQQAAAAAFEQYBAAAAFEQYBAAAAFAQYRAAAABAQYRBAAAAAAUR\nBgEAAAAURBgEAAAAUBBhEAAAAEBBhEEAAAAABWnMuwAAAE6fqqoyGg3vm9fpdFOvO9YIz4J9EHgS\nwiAAAB7ZaDTMy1evpdlqJ0mmk3GuXL6UXm91zpVBGeyDwJMQBgEA8FiarXZa7e68y4Bi2QeBx6UP\nIQAAAEBBhEEAAAAABREGAQAAABREGAQAAABQEGEQAAAAQEGEQQAAAAAFEQYBAAAAFEQYBAAAAFAQ\nYRAAAABAQYRBAAAAAAURBgEAAAAURBgEAAAAUBBhEAAAAEBBhEEAAAAABREGAQAAABREGAQAAABQ\nEGEQAAAAQEGEQQAAAAAFEQYBAAAAFEQYBAAAAFAQYRAAAABAQYRBAAAAAAURBgEAAAAURBgEAAAA\nUBBhEAAAAEBBhEEAAAAABWnMuwAAoExVVWU0Gt43r9Pppl53rAoA4GkSBgEAczEaDfPy1WtpttpJ\nkulknCuXL6XXW51zZQAAZ5swCACYm2arnVa7O+8yAACKoh82AAAAQEGEQQAAAAAFEQYBAAAAFEQY\nBAAAAFAQYRAAAABAQYRBAAAAAAURBgEAAAAURBgEAAAAUBBhEAAAAEBBhEEAAAAABWk8aGG/319M\n8utJPphkOcm/GAwGv3Vg+U8k+edJtpP8+mAw+NdPsVYAAAAAntBRPYN+Ksn1wWDwo0n+wyT/090F\ne0HRrya5kuTHkvxsv99/z9MqFAAAOD5VVWVt7c7+T1VV8y4JgGfkgT2DkvxGkt/cu1/Pbg+guz6S\n5NpgMLiTJP1+/wtJfvTA4wEAgBNqNBrm5avX0my1M52Mc+XypfR6q/MuC4Bn4IFh0GAwGCdJv9/v\nZjcY+mcHFveS3DkwPUzi0wMAAE6JZqudVrs77zIAeMaO6hmUfr//gSSfSfKvBoPBvzmw6E6Sg58c\n3SS3jnq+ixd92MC8PM7+t7RUpdO+mXZnJfVs5sKFblZXz85+XFL7kpzJNp4GPvveme3zdDsN68++\n92Bn/TPwrDvJ+6B9D06+owaQfiHJv03yTwaDwecPLf5Kkg/3+/3nkoyze4rYrxz1gtevDx+zVOBJ\nXLzYfaz9b21tmNF4I1XWMxlv5MaNYTY3z86FCEtqX5Iz2caT7nH3vRLYPk+3k77+7HtHO+ufgWfd\nSd0H7XswH48awh7VM+iXsnvq1y/3+/1f3pv3ySTtwWDwyX6//wtJfje74wl9ajAYvPaI9QIAAADw\nDB01ZtDPJfm5Byz/bJLPHndRAAAAADwd+oECAAAAFEQYBAAAAFAQYRAAAABAQYRBAAAAAAURBgEA\nAAAURBgEAAAAUBBhEAAAAEBBhEEAAAAABREGAQAAABREGAQAAABQEGEQAAAAQEGEQQAAAAAFEQYB\nAAAAFEQYBAAAAFAQYRAAAABAQYRBAAAAAAURBgEAAAAURBgEAAAAUBBhEAAAAEBBhEEAAAAABREG\nAQAAABREGAQAAABQkMa8CwAA3llVVRmNhvfN63S6qdcdywEA4PEJgwDghBqNhnn56rU0W+0kyXQy\nzpXLl9Lrrc65MgAATjNhEACcYM1WO612d95lAABwhuhnDgAAAFAQYRAAAABAQYRBAAAAAAURBgEA\nAAAURBgEAAAAUBBhEAAAAEBBhEEAAAAABREGAQAAABREGAQAAABQEGEQAAAAQEGEQQAAAAAFEQYB\nAAAAFEQYBAAAAFAQYRAAAABAQYRBAAAAAAURBgEAAAAURBgEAAAAUBBhEAAAAEBBhEEAAAAABREG\nAQAAABREGAQAAABQEGEQAAAAQEGEQQAAAAAFEQYBAAAAFEQYBAAAAFAQYRAAAABAQRrzLgAAAE6a\nqqoyGg3vm9fpdFOvO5YKwOknDAIAgENGo2FevnotzVY7STKdjHPl8qX0eqtzrgwAnpwwCAAA3kGz\n1U6r3Z13GQBw7PRzBQAAACiIMAgAAACgIMIgAAAAgIIIgwAAAAAKIgwCAAAAKIgwCAAAAKAgwiAA\nAACAggiDAAAAAAoiDAIAAAAoiDAIAAAAoCDCIAAAAICCNOZdAAAAwHGrqiqj0XB/utPppl53LBwg\nEQYBAABn0Gg0zMtXr6XZamc6GefK5Uvp9VbnXRbAiSAMAgAAzqRmq51WuzvvMgBOHP0kAQAAAAoi\nDAIAAAAoiDAIAAAAoCDCIAAAAICCCIMAAAAACiIMAgAAACiIMAgAAACgIMIgAAAAgIIIgwAAAAAK\nIgwCAAAAKIgwCAAAAKAgjXkXAPAsVVWV0Wi4Pz0crmVWzeZYEXBWHX6/SZJOp5t63bE4AGC+hEFA\nUUajYV6+ei3NVjtJcvPGG2m1e2l3e3OuDDhrDr/fTCfjXLl8Kb3e6pwrAwBKJwwCitNstdNqd5Mk\nk/FoztUAZ9nB9xsAgJNCP2UAAACAggiDAAAAAAoiDAIAAAAoyEONGdTv9y8n+ZeDweA/ODT/55P8\nTJLre7M+MRgM/up4SwQAAADguBwZBvX7/X+a5D9P8k6jrH5/kp8eDAZ/etyFAQAAAHD8HuY0sWtJ\n/lGS2jss+4Ekv9Tv9/+g3+//4rFWBgAAAMCxOzIMGgwGn0my/S6LP53kE0n+QZIf6ff7P36MtQEA\nAABwzB5qzKAH+LXBYLCWJP1+/3NJvi/J5x70Cxcvdp/wJYHH9Tj739JSlU77ZtqdldSzmQsXulld\nPb378cH2JMl0vJR6fTHdM9K+ww639yy28TR43M++s77+tO90Ow3te5L/O09D+57UWfuMP6yk9iUn\naxv1nQ9OvscOg/r9/mqSL/X7/Y8mmWS3d9Cnjvq969eHj/uSwBO4eLH7WPvf2towo/FGqqxnMt7I\njRvDbG6e3gsRHmxPkozHm6nXd7LcPBvtO+xwe89iG0+6x933krO//rTvdDvp7XuSfS85+e07Dmft\nM/6wktqXnJxt9En3PeDxPGoI+yhh0CxJ+v3+TybpDAaDT+6NE/T5JBtJfm8wGPzOI706AAAAAM/U\nQ4VBg8Hgb5L88N79Tx+Y/+nsjhsEAAAAwClwdvpJAgAAAHCkJx1AGgCAh1BVVYbDtfvmdTrd1OuO\nzQEAz5YwCADgGVifTvLKF2/l3PPnkyTTyThXLl9Kr7c658oAgNIIgwAAnpGVZiuttksuAwDzpV8y\nAAAAQEGEQQAAAAAFEQYBAAAAFEQYBAAAAFAQYRAAAABAQYRBAAAAAAURBgEAAAAURBgEAAAAUBBh\nEAAAAEBBhEEAAAAABREGAQAAABREGAQAAABQEGEQAAAAQEGEQQAAAAAFEQYBAAAAFEQYBAAAAFAQ\nYRAAAABAQYRBAAAAAAURBgEAAAAURBgEAAAAUBBhEAAAAEBBhEEAAAAABREGAQAAABREGAQAAABQ\nEGEQAAAAQEEa8y4AAB5XVVUZjYb3zet0uqnXHesAAIB3IwwC4NQajYZ5+eq1NFvtJMl0Ms6Vy5fS\n663OuTIAADi5hEEAnGrNVjutdnfeZQAAwKmhHz0AAABAQYRBAAAAAAURBgEAAAAURBgEAAAAUBBh\nEAAAAEBBhEEAAAAABREGAQAAABREGAQAAABQEGEQAAAAQEGEQQAAAAAFEQYBAAAAFEQYBAAAAFCQ\nxrwLAIDjUlVVhsO1/ftJUq/fO+7R6XTvmwYAgBIJgwA4M9ank7zyxVs59/z53LzxRur1Rs49fz5J\nMp2Mc+XypfR6q3OuEgAA5ksYBMCZstJspdXuZjIepV5fSKvdnXdJAABwougrDwAAAFAQYRAAAABA\nQYRBAAAAAAURBgEAAAAURBgEAAAAUBBhEAAAAEBBhEEAAAAABREGAQAAABREGAQAAABQEGEQAAAA\nQEGEQQAAAAAFacy7AAAAOMuqqspoNLxvXqfTTb3uuCwA8yEMAgCAp2g0Gublq9fSbLWTJNPJOFcu\nX0qvtzrnygAolTAIAACesmarnVa7O+8yACCJMYMAAAAAiiIMAgAAACiIMAgAAACgIMIgAAAAgIII\ngwAAAAAKIgwCAAAAKIgwCAAAAKAgwiAAAACAggiDAAAAAAoiDAIAAAAoiDAIAAAAoCDCIAAAAICC\nNOZdAHCyVFWV0Wi4Pz0crmVWzeZYEY/C+gMAAI4iDALuMxoN8/LVa2m22kmSmzfeSKvdS7vbm3Nl\nPAzrDwAAOIowCHibZqudVrubJJmMR3Ouhkdl/QEAAA9izCAAAACAggiDAAAAAAoiDAIAAAAoiDAI\nAAAAoCDCIAAAAICCCIMAAAAACiIMAgAAACiIMAgAAACgIMIgAAAAgII8VBjU7/cv9/v9z7/D/J/o\n9/v/d7/f/6N+v/+Pj788AAAAAI7TkWFQv9//p0k+mWT50PzFJL+a5EqSH0vys/1+/z1Po0gAAAAA\njkfjIR5zLck/SvK/Hpr/kSTXBoPBnSTp9/tfSPKjSX7zWCsEgEJUVZXRaLg/PRyuZVbN5lgRwNlQ\nVVWGw7X75nU63dTrZ2fUjMNtPGvtA47XkWHQYDD4TL/f/9A7LOoluXNgephk9ZjqAoDijEbDvHz1\nWpqtdpLk5o030mr30u725lwZwOm2Pp3klS/eyrnnzydJppNxrly+lF7v7Hx9OdjGs9g+4Hg9TM+g\nd3MnSffAdDfJraN+6eLF7lEPAZ6Sh9n/lpaqdNo30+6sJEmm46XU64vpdlZSz2YuXOhmdfX07scl\nty/JmWjjQQ9q70lq+8N+9i0tVbl48fm0O7vhTy2bJ6YNT8Ph9XfW23eStsnjcBrW35P833mc7Tup\nf6uDdZ2Umo7TwfZNx0vpdtu5cPFikmQ8Wj717X2n95i7bZx3+3zng5PvScKgryT5cL/ffy7JOLun\niP3KUb90/frwqIcAT8HFi92H2v/W1oYZjTdSZT1JMh5vpl7fyXJzPZPxRm7cGGZz8/R2OS65fUnO\nRBsPelB7T0rbH3bfS6y/s96+s7Y+T/r6e5R9750cZ/tO6t/qYF0npabjdLB9Z23/S07u/zRPuu8B\nj+dRQ9hHCYNmSdLv938ySWcwGHyy3+//QpLfze5A1J8aDAavPdKrAwAAAPBMPVQYNBgM/ibJD+/d\n//SB+Z9N8tmnUhkAAAAAx+709osEAAAA4JEJgwAAAAAKIgwCAAAAKIgwCAAAAKAgwiAAAACAggiD\nAAAAAAoiDAIAAAAoiDAIAAAAoCCNeRcAAMDpV1VVhsO1++Z1Ot3U6449AsBJIwwCAOCJrU8neeWL\nt3Lu+fNJkulknCuXL6XXW51zZQDAYcIgAACOxUqzlVa7O+8yAIAj6LcLAAAAUBBhEAAAAEBBhEEA\nAAAABREGAQAAABREGAQAAABQEGEQAAAAQEGEQQAAAAAFEQYBAAAAFEQYBAAAAFAQYRAAAABAQYRB\nAAAAAAURBgEAAAAUpDHvAgCAh1NVVYbDtfvmdTrd1OuO7QAA8PCEQQBwSqxPJ3nli7dy7vnzSZLp\nZJwrly+l11udc2UAAJwmwiAAOEVWmq202t15lwEAwCmmXzkAAABAQYRBAAAAAAURBgEAAAAURBgE\nAAAAUBBhEAAAAEBBhEEAAAAABREGAQAAABREGAQAAABQEGEQAAAAQEGEQQAAAAAFEQYBAAAAFKQx\n7wIAADgdqqrKaDRMkgyHa5lVszlX9OxUVZXhcO2+eZ1ON/X62Ti2etbbB8D9hEEAADyU0WiYl69e\nS7PVzs0bb6TV7qXd7c27rGdifTrJK1+8lXPPn0+STCfjXLl8Kb3e6pwrOx5nvX0A3E8YBADAQ2u2\n2mm1u5mMR/Mu5ZlbabbSanfnXcZTc9bbB8A9+n0CAAAAFEQYBAAAAFAQYRAAAABAQYRBAAAAAAUR\nBgEAAAAURBgEAAAAUBBhEAAAAEBBhEEAAAAABREGAQAAABREGAQAAABQEGEQAAAAQEEa8y4AACBJ\nqqrKcLh237xOp5t63bEr4GhVVWU0Gu5PD4drmVWzd32s9xugZMIgAOBEWJ9O8soXb+Xc8+eTJNPJ\nOFcuX0qvtzrnyoDTYDQa5uWr19JstZMkN2+8kVa7l3a397bHer8BSicMAgBOjJVmK612d95lAKdU\ns9Xefw+ZjEcPfKz3G6Bk+kECAAAAFEQYBAAAAFAQYRAAAABAQYRBAAAAAAURBgEAAAAURBgEAAAA\nUBBhEAAAAEBBhEEAAAAABREGAQAAABREGAQAAABQEGEQAAAAQEGEQQAAAAAFacy7AADg6auqKqPR\ncH+60+mmXn+2x4QO1zAcrmVWzZ5pDQAlOvz+m8zncwA4OYRBAFCA0WiYl69eS7PVznQyzpXLl9Lr\nrc6thiS5eeONtNq9tLu9Z1oHQGkOv//O63MAODmEQQBQiGarnVa7e2JqmIxHc60FoCQn4TMAODn0\nCwQAAAAoiDAIAAAAoCDCIAAAAICCCIMAAAAACiIMAgAAACiIMAgAAACgIMIgAAAAgIIIgwAAAAAK\nIgwCAAAAKIgwCAAAAKAgwiAAAACAgjTmXQAAAJw2VVVlOFy7b16n0029fjaOtZ6G9lVVldFouD89\nHK5lVs3mWBHA6SEMAgCAR7Q+neSVL97KuefPJ0mmk3GuXL6UXm91zpUdj9PQvtFomJevXkuz1U6S\n3LzxRlrtXtrd3pwrAzj5hEEAAPAYVpqttNrdeZfx1JyG9jVb7f0aJ+PRnKsBOD1OTj9PAAAAAJ46\nYRAAAABAQR54mli/368n+Z+TfDzJRpJ/PBgMvnZg+c8n+Zkk1/dmfWIwGPzVU6oVAAAAgCd01JhB\n/0mSpcFg8MP9fv9ykv9xb95d35/kpweDwZ8+rQIBAAAAOD5HnSb295P8TpIMBoOrSf7uoeU/kOSX\n+v3+H/T7/V98CvUBAAAAcIyOCoN6SdYOTO/snTp216eTfCLJP0jyI/1+/8ePuT4AAAAAjtFRp4mt\nJTl4Pcn6YDCoDkz/2mAwWEuSfr//uSTfl+RzD3rCixdP9uUp4Sx7mP1vaalKp30z7c5KkmQ6Xkq9\nvphuZyX1bObChW5WV0/vflxy+5KciTYe9KD2nqS2P+xn31Hr70nadPC55/W3eJrtO4nO4v54sE2n\nYf09yf+dx7m9Hn6uk/C3OVzXaVifhz3KOjqN7TvKo/xP86y3Qd/54OQ7Kgz6wyQ/keQ3+v3+DyX5\n0t0F/X5/NcmX+v3+R5NMsts76FNHveD168PHrxZ4bBcvdh9q/1tbG2Y03kiV9STJeLyZen0ny831\nTMYbuXFjmM3N03shwpLbl+RMtPGgB7X3pLT9Yfe95Oj19yRtOvjc8/pbPM32nURncX882KaTvv4e\nZd97J8e5vR5+rnn/bd6prpO+Pt/Jo6yj09i+ozzK/zTPcht80n0PeDyPGsIeFQb970mu9Pv9P9yb\n/i/7/f5PJukMBoNP7o0T9PnsXmns9waDwe88asEAAAAAPDsPDIMGg8EsyX9zaPZfHVj+6eyOGwQA\nAADAKXB6+kECAAAA8MSEQQAAAAAFOWrMIACAolRVldHo3uCnnU439frxHz+rqirD4dpTfx0AgMOE\nQQAAB4xGw7x89VqarXamk3GuXL6UXm/12F9nfTrJK1+8lXPPn3+qrwMAcJgwCADgkGarnVb70S7R\n+jhWmq1n8joAAAfpiwwAAABQEGEQAAAAQEGEQQAAAAAFEQYBAAAAFEQYBAAAAFAQYRAAAABAQYRB\nAAAAAAURBgEAAAAURBgEAAAAUBBhEAAAAEBBhEEAAAAABWnMuwAAeBaqqspwuHbfvE6nm3rdcRFg\nV1VVGY2GSZLhcC2zajbninhWTsNnxMHtM3nwNnq4PYcfexraCzxdwiAAirA+neSVL97KuefPJ0mm\nk3GuXL6UXm91zpUBJ8VoNMzLV6+l2Wrn5o030mr30u725l0Wz8Bp+Iw4uH0meeA2erg9hx97GtoL\nPF3CIACKsdJspdXuzrsM4ARrttpptbuZjEfzLoVn7DR8RtzdPpMcuY0ebM87PfY0tBd4evQDBAAA\nACiIMAgAAACgIE4TAwCAE2SnqrKxWWVzeydb29Whn51sHpze2b3dqWbZqaps78yys3N3epadnd35\nd6dns1lms2Q2m2Vzcytv3p6mvjDMxvo0SS2L39hKkmxurCe1Wpa/uTu9s72dr357ksXGYlKrpZak\nVksW6rXU67W923oWarUsLByYV9u9XWzU02jUs9ioZ3Fh77ZRz1JjYf/+4kI9y0sLWVlaSHO5kcaC\n49YAT4swCAAAnlA1m2W8vp2NaprpxnamG9uZ7N3eujPKV74xTOrTbG1Xma5v5ivfHGdnVs/G1k42\nNneyvrnpJT0uAAAgAElEQVSze39rNwCan613nX71rfVnWkljoZaVpUZWlhaystRIc/nebXtlMUsL\nVd64NUm3M8vK0kKmk52sLNVSVbPU67VnWivAaSMMAgCAPds7VcbTraxNdrJdVVnbHmZzaye3bm1k\nu0rqr7+eja2djCfTbG0nO7Phfk+dz3zhtYd+nW9enyZJGgv1rCwtZHmxnl57KcuLC3vTC1lavNuD\nZiFLjXu9ae72ollaXMjiwm6Pm4W9njgLC7Us1Ov79xsH7tf3eurUakm9VstoNMyfDN5Mq9XJzRtv\nZGFhIc+dv5hakreuv5FavZ5zz19MMst0PMrlj76QTqeXWZLMZqn2ehjtVLNU1Szbe7dVdW/e3V5J\n2ztVNrfu9WQ63LNpe7vKxva9YGx9czvTjd3b9c2dvLU2zfrGTt75Qup37p/80p0sNeppLCQri/Wc\n6+2kXq2ntdzIesbpNBezU83iKupAyYRBAACcOTvVLMPJZkbTrYyn2xlNt/Z/xusH7u/P285ospmt\nncNxw+GrMG3s31uoJytLjXSai2nUZ3nx+VZ67ZU0lxsHfhaSajNf+/addDudLDbq2dmc5u9/z4u5\n8Pxzcz0VqpGNNPdOyVperKder2d5cSFJstjYPe1rZWl3utpaSKe5mF57aW71zmazbGztZLqxk/H6\nVt64cTv/z+B6ZrXFrG/u5PadtWztJFWtkY3NnUzWt3JzYzs3R2t7z7CZ/O1k//mWG7V029O0VxpZ\nyGZ6rUaytJ7VObYR4FkRBgEAcGLtVNX+l//xdHs/yBnvBTjj/XBne+8xu8snG9uZvXM3kre5e9rR\ni883s7m9k9bKUqrt9Sw1FnJutZflpYWsj29nebGRCxcuZGmxnuHtN9NYaOT5C+9JkkzGw/zI97w3\nvd7q255/be1OxtONtNrNvcdupmVMnEdWq909bayR57rL6S3v5I2bo/3Lo994cyv1+sL+Ornx5muZ\npZ6VznN59bU3srGVpNHMaLqV22vjTDdnubm2kRt37p7+tpk/+/rfppak3VzI4FuTfPC95/K+C+28\n70I7L55vWWfAmSEMAoBjcvd0ibuDuVbVLI3l9dwabmQ2m6U6MHDrbLY7xki192W1XkvG4/UMp9vZ\nqW0mSSYbVRbqtYzXt1JLsrFVZWGhls3tndRrtb3ne8hvuzAHVTXLZH0r042dTDe3s753O93YPfVn\nfW9cncn6gdv13SBnsrG9OwbP5s5Dv95CvZZ2czGrneV86DtWs7RQS6e5mE5rcfd2Zfe23dybbi6m\ntXIvlFlbu5MvfPm1tNrd3Hjztb1g4bkkyY03x6nXF7La2e01Mq4Zk+Y0WKjX0m0t5UK3sbc+LyTJ\n/vp97vzFTDd28s1XX8t4Y5bN2XJujzZye7ieL3/9dr789dv3Pdd7z7dy6X2reel9q/nw+1dz8Vwz\nNdsCcAoJgwA486pqlq3tWapUuT3ayPZOldFoI3/x9dupN6Z741PsZHN7J5tbVTa3dq/W87bpvdvt\n7SrbB67as72zOx7GTvW0gpk77zr9mS+89o7jhNy9Wk9jYfenliqj6WYWF+8k1U7++vVpWisrbxuD\nZHHx/iv9LDYWDk2/2+PqWTAAx6kwm832t9mtnSo7O7MD47js3BvXZev+q1ltblcZjsb52rfXUqtP\nMxpPs1PVsvCNV7O9U2W6vpGdKplltPvc21V+8w++/cj11ZI0lxtprTTywrlmWiuNtFYW01pu7AU5\njbRX9gKdlUbazcW96UaWFxf2v5hfvNjN9evDY/7rcdbUarW0Vho5323k4uq9XkXj0Vo+/tLFrG3U\n8+3r47x6Y5RXb4zzzTdG+db1cf7dn+1u273W4l4wdC6X3reaD77YzWLDeyFw8gmDADiRZrNZNrer\nrG9sZ7q5s9uT4OD9zZ3cXhvl2qtrmdUm2dreHdB1p0pmfzXJ+sbWblhT3T4U0twLUn7/y289Uk0H\nB3BtLNTTXF7I4kItCwv1NPYGam3sT9dTr9fSXFnM5ub2/oCttdrul4/63u3+EeXZLBubm3ntrUka\njcXMMsv6dJrUallaWs4sycZ0mllqaSwupZrNsrW1vTt+R62+d/no2f5ArXcHwd3a+9K/vV3tDby6\n2+voaVwVqF6rpdHYDaLuhlCNRj2LC7uXla5lth9I1Wu1bG2uZ6Fez8prr6Veq2VzY5p6rZ7WW9dT\nryU721uZbibN5u39y1PX7/7Usj8Y7kK9tve33K0hufc3zuG/90O0YzwZ5/Vb61me1LK+vp6//Js7\naba2kru9unZXV5K9nl7JfT2/qmr3djKd5Ouvj7O4tJPZbJbhcDO11PL66GZms2Q03EhqtTRv38jG\n5kbWJlUWl27s9yqrqll2ZvcPxnv/wLyzbFfVgXVf3bcd3L28+MHAZ3v7aYSWu9tULdkNIhu1LDXq\nWVmq50JvOZ3Wyu5VoJYbaR66IlRzqZGV5UbaK4209gKgleXG/nqEeanValltL+YD713Nxz70/P78\n7Z0q33xzlGuv3snXXr2Tr37rTv70qzfyp1+9kWT3CmgferGXD39gNR//987npfetOrUMOJGEQQA8\nsd2eBlU29nrR3L088sbmzn3z7l4hZvf0kAP3t94+f7qxk+oxT4FqLOxkob47AGp3aSmLjXqqnc0s\nLtTTbjd3/zGvtvPh969mtdfOyuJClvev3rN71Z7Dt4uN+mOdCvAovRMOnqKS5MBpKu95x+kHjVFy\n2J07t/MHX3oty81ORqNhfuC7L2al1dm/is9+r5BDV/jZOrDs3a4CdHje3fBpt7fIVtbeJZC65/D0\nvQF6/99vzL9nxx/+5c0nfIbDPbumh6Z3g7n/7xuHByp+NO92NanlxYU0VnaDysWFehYWdgO6u2Fd\nYy/MPHzFqqXGwoH7u73EtjanGXzjVjqdToZ3bmSx0ciFCxfTaNRz68brj719wmnRWKjnu97by3e9\nt5crf/cDSZKba+u5thcMXXv1Tv7622u59uqd/PYffyPN5Ub+/e96Ph9/6Xy+56Xz6bUMTg2cDMIg\n4KFVVZXhcO2+eZ1ON/WHODWkqqqMRvd/qXvY3+WeajbbO/Jf7R35v78HwO21aW6PtjLZnqaqklvD\n7aRWZTobZaeaZTqZ5Atf+tZu74OdWRYaS9mpst+T5O6pI3d/Ng990T94ysjd00Y2t3ZPK3nc4Oau\nhXotK0u7l1R+rruc9154914EzaXdXgaz7Y0Mvnlr/wo9a7evZ6mxkPMXX3jIIOWF/S+q92+jVTqd\n9onfPg/vkw/ap2q13R41i416lhfrOddZSq/XfFalJrk/kKqqWW5cfz2pLaR37vlUVfLWW2+mlno6\nq8+lqmYZj0d56cVmlldaqapZlpZXMkttv4dMdeCS1gd77FR73XVmdy99nd3ph91G19fX8zevD7O0\nvJytzY1813t7WVlZSS17PbuS5G4vo73eSPd6et273Vhfz1dfvZ3l5WZqSUbD26nX6+n2VlOr1TK8\nczP1+kJ6q+eyuTHN9750Pp1OJ/W9UOdgb6j9XlH3zavvBT/3ekc9bWtrd3JzbZJWeyWzjYXdq0/t\nXW2Kew5/5g2Ha5k9Zo+s4/z8PM66uOf53kp+sLeSH/zIC0mSjc2dDL55O1/62o186Wtv5U++8mb+\n5Ctvppbku76jl4+/dD7f+9KFfOcLHeMNAXMjDAIe2vp0kle+eCvnnj+fJJlOxrly+dJDHfUdjYZ5\n+eq1NFvtR/7dk6yazTLdH/R0d8DTja2d/Z4wmwd6xWxuVfeWbd/rOXE3hLm/R8WBUKa6FwA9ft5y\nr8fB1cHjt3ehXktjv5dAPc3lRlbbS1leXMjy4m4vmrs9bO72sllerO9Pryw39gOf3SvCLOz/NBYe\nvefN2tqd3LgzTqu9nCSZDh//C/HBbfS0bJ8H98nTUPPBQCrJ/qWsu3tHyrcmC7uB3bndkOrG9p28\n+vpbOff8s33P2O2hlbTa3b3Q8MXHet21tTvZ3No80NNrvBdI7k1nuDfdzmRc5aXv6J7o9cfDO/yZ\nd/PGG2m1e2l3e0/8XE+yLxxnXby75aWFfPyl8/n4S+czm83y7bcmu8HQtbfy1W/t9hz6P/7g61nt\nLOV7Xzqfyx95If0PPuf0SOCZEgYBj2Sl2dr/YvOomq32Y//us7K9U+WtO+u5Pd7I7eFm7ow3cnu0\nkbXxZibru1e22b3dymR994o4x3FM9e1jrdT2ApLFe6d8LNTS2D8NpH7glJB796ud7bxxa5KlpaXU\na7WsT0ep1+tpdzqp12qZjteysLCQbreX7a31fPRDz6fX6eyPc7N7Csnu6SN3Tw05GP6c9QGCT8M2\netiT7JOnwVlvH2fXwfeTyfjJTgE8zvem46yLo9Vqtf1L0/9Hlz+YyfpW/uLrN/Pn197Kl//6rfz+\nn7+W3//z1/Jcdzk/9LEX8vc+9mLef7Ez77KBAgiDgKJMNrZzc7iZ1++sZTjZzFu3ptnYmmV7MMlk\nfeuhrnyzvLiQ1kojz/WW8/7ldlori7uDn64sprl8YOyZxu79pUb9vl4yS3v398fs2Bto+Di8fcyZ\n7b2eB+f3pjf2plczGdfzd156Tk8EAHhGWiuL+cGPvJAf/MgLqapZvvqt2/m//vL1/MlXrue3//gb\n+e0//kY+8J5O/t7HXswPfeyFnOssz7tk4IwSBgFnzsbWTl5/a5I3bk3yxq1p3ry5e/vGrUmGk613\n/J3FhSorS/V853vaOb/ayrnOcs51lrK6d9trL6W9spjWSsNVQQCAJ1av19L/zufS/87n8lNXvjt/\nfu2t/NFfvJ4v//Vb+d8+fy2/8e+u5aMffC4/9LEX8wP9i/MuFzhjhEHAqTbd2M433xzlb14f5m9f\nX8vfvjHKa2+N3za2Tr1Wy4VzK3n/hWb+//buPDju877v+HtP7IXFRdwnIZIPb4kiRFAidVqiRFmK\nZatJ6maa2lN7pnWn/sdJGttN4jRNJonHbmbSpNPYceKqaevYlhxf1GmROsJTpEQSJB6SAImbuIn7\n3u0fuySXMAmQxLEA9vOaweB37D6/7/M7d7/7/H7PxOQU2RlB0gNeJkZ6CaR5yMvPV883IiIikhQe\nt4uq9XlUrc9jYHico7UdHKy5TM2lXmou9fLS65bta7MJ+RzEH/kkIjInSgaJyLIxMjZJw+WBWOKn\nfYCGywO09wzf8Mwen9fF2pJMSnKD5GcFyM/2k58VICfDh9vlvMltVP3zdouWiIiIyFylB7w8cX8J\nT9xfQnvvMIdq2nnvZBsHz3QBkJ89yPqyLPyOKGqrLCJ3S8kgEVmyxsanON98hbMNvdQ29nLp8sAN\nLX78aW5MWSYVBWHKCkJUFITJy/KrNw4RERFZEfKzAnxi92qef6iCg6ca+cnBJtp7RmjvGcHncVCR\n52NreBKfV1/rROTO6KwhIkvGxOQUF1r6qW3o5WxjLxdb+5mKxLI/LqeDNcUZrCnOoLwgnYqCdFZl\nKvEjIiIiK5/T6WDL6kz6BkcYj3qxjVe40HyF2pYRzrXVs7ownYL0KbJCrmSHKiLLhJJBIpI00WiU\nls5BTpzv4mxDL+eb+5icigDgcEBFQTrry7PYUJ7F2uJM0rz6gCMiIiKpLTOURvXGfMqypmjpmaSp\ne5K6ln7qgKzQKPeu9eMnOms5IpLalAwSkUUVjUbp6hulrrmfAye76ewbuzavNC/E+rJY8mddaSYB\nn05RIiIiIjfjcTmozPexfWMuLV1DnDp/mc7+KfafaCXkc7K20E9mTlStqEXkpvRNS0QW3ORUBNvU\nz4kLV2jt6WBkbBIAr9vJdpPL/ety2bw6m/SAN8mRioiIiCwvDoeDktwQvmiQoTFo6oWLrf2cuDjE\nhfaLbK7MJsMdxamnTYtIAiWDRGRBTExGOF3fzQfnOvnoQhdDo/EEkMfJPUVh8jNdvLC7glXZWUmO\nVERERGRlSPe72F2aR1lWhPr2cZq7Jzh4uh2fx8E9hT7CWRHcLmWFRETJIBGZR9FolIb2Ad4/eZlD\nZy5fSwBlpadx/9osHEQoK8zB6XQwPDSA160PIyIiIiLzLZDmZGtFkB2bs6i52Itt7KWmcYS6y/Vs\nrMhilT+KVx/DRFKakkEiMmd9g2McrGnn/dNttHQOARAOeNjzQCk7NuRTUZjO4EA/751qw+nUfesi\nIiIiiyHg8/DAhjyKwhM0dE7S0DXO8XNdeFxQme/j/swIHv04J5KSlAwSkbsyFYnQ3DnC3/zsArWN\n/USiUdwuB9tNLru2FLJ5dbaaIc+TSCTC4ODADdNCoXScc7z5PxKJMDDQP+/liqwki3WcTF9OJBLr\nWTFxOcvt+JypTiuhfqlu+rVJ229pS/M4WV/ip2pTCbWNV6ip78a2jtLQVc/mymyKs5w3HK/aniIr\nn5JBInJH+oenOH+mnYtt/YxPxD7MVxSks2tLIdUb8wn5PUmOcOUZHBzgjcMX8AeCAIwMD/FU9RrC\n4Yw5lTs6MsyB471kZufMa7kiK8liHSfTl9PT1Y7T6V7Wx+dMdVoJ9Ut1idcmbb/lw+txsfWeHHID\nYzR0TnCxY5xjtZ2c9jiwl7rZWJnL2OiwtqdIClAySERmFYlEabg8wKkLQ/QMTgHgT3OxriTIJ3eX\nYyoKkhzhyucPBAkE0+e9XJ8/sCDliqwki3WcJC5neGgQp9O17I/PW9VppdQv1S3UtUkWnsflYF2R\nn/vWl1BzsYezl3qobZukpa+L9aVBIpFoskMUkQWmZJCI3NLI2CQX2sZo7BpndOIKALlhN5vX5FO8\nKsjoyCCF2f4kRykiIiIid8PndbHd5JIfGqfu8gRNXWMctVdo6jzDi4+u5f51q3A49LxHkZVIySAR\nuUE0GqW7f5wPLrTScHmASBTcTlhfnklecIJwwEv2qlCywxQRERGReeLzONlSHmD7hmKO2zYaOkb4\nq1dOUVGQzouP3sOm1dnJDlFE5pmSQSICwMTkFIfPdPDG0QaaOocByAh6Kcl2UrrKR15+Pl0dbUmO\nUkREREQWSijgoWpdFp9+opI3T3RytLaDb3zvQzZWZPGrj62hvEC3BYqsFEoGiaS4wZEJ3j7Rwlsf\nNNM/NI7DAUU5PjZVrqIgO0B352V1By8iIiKSQvKzfPz7Fzbz7OUBfnCgjpqLPfzh3x+lemM+n3yk\nkrxMPSZAZLlTMkgkRXVeGeH1o028e7KV8YkI/jQ3e6vL2LEug5pL3QSCwWSHKCIiIiJJVF6Qzpd+\n/T5qLvXwg7frOHymnWO1HTy+rZjndlUQDniTHaKI3CUlg0RSzMW2fl493Mgx20E0CtnhNPY8XMrD\n9xbhT3PT39+X7BBFREREZAnZVJHNhs9kcfRsBy+/U8ebHzTz3qk29laXseeBMtK8rmSHKCJ3SMkg\nkRQQiUY5cuYy//i6xTbFegUrywvxTHUZVevzcLucSY5QRERERJYyp8NB9cZ8tptc9p9o4cfvX+KV\ndy/yi+MtfGL3anZvLdRnSpFlRMkgkRVscirCwZrLvHq4kbbu2EOhN6/O5pnqMjaUZ6mrUBERERG5\nI26XkyerStm1pZDXjjTy2pEm/tdrlteONvHiI5U8o15nRZYFJYNEVqDR8Une+aiN14400jswhsvp\n4ImqUh7dWkhpni7QIiIiIjI3/jQ3LzxcyePbivmn9y/xzoet/PWPTvPm8WZe2LWa9eVZyQ5RRGag\nZJDICjI4MsFbHzTz5rEmhkYnSfO42PNAKXseKMXck0tn50CyQxQRERGRFSQjlMZvPm3Y80ApL79T\nz7HaDv688QRbKnN48dFKyvLVHb3IUqRkkMgK0NM/ymtHmjjwUQvjExGCPjef2L2aj20vIeT3JDs8\nEREREVnhCrIDfOGFzfSOTPKtV05yqr6b0/Xd7NyUzycfrmSVuqMXWVKUDBJZxtq6h9h3qJGDNZeZ\nikTJSk/jU4+U8ei9RcuuV4dIJMLg4I0tl0KhdJxO54zzFtNCxTG93EgkAnCt3IGBfqKR6JyWMR9x\nQXLWe6LpMSVr3SxFkUiEgYH+G8Zhae1Hc4lhev3g9vfHxdyX56u+cqPp2zDZ56K5mu14nal+czkW\n5mL6cpOxfy/mNWB6fRPX8VK8Pi4l68qy+O1Pb6PmYg/f31/HwZp2jtZ28Pi2Ep57qJx0dUcvsiQo\nGSSyDNW19rHvUCMnznUSJfZLzN6dZTy4qWDZ9uIwODjAG4cv4A8EARgZHuKp6jWEwxkzzlsqMc5n\nuT1d7TidbjKzc66NB4JhgunhuVVgjnEla73PFFOy1s1SNDoyzIHjvTfsN0ttP5pLDNPrdyf742Lu\ny/NVX7lR4npdCueiuZrpeJ2tfnM5FuY75sXevxfzGpBY3+nreCleH5cah8PB5socNq7O5vCZdl55\np543jjXx7slWnqku46mqUvxp+ioqkkw6AkWWiWg0yqn6HvYdarjWPfzqwnSe3VnOtnW5OFdAz2D+\nQJBA8Ob3lc80bzEtVByJ5Q4PDeJ0um4YT5alst4TTV9Xcp3PH1jS+9FcY0is393GsBjmq75yo6V4\nPpqLmY7XO3nvYpoeczIs5jVgpvW80vbHheJ0OHhwUwFVJo/9H7bwk/cv8aN3L/LmsWb27izjiftL\nSPMsr9bsIiuFkkEiS9xUJMLRsx3sO9xIU0fsQ8/m1dns3VnO+rJMdQ8vIiIiIkuax+3kqapSdm8p\n5M1jTbx6pInvv13Ha0ea+PjOch7bVoTHraSQyGJSMkhkiRqbmOK9k7Hu4bv6RnE4YMeGPPZWl1Ne\noF+iRERERGR58ae5eX7Xap7YXsJrR5p441gT//et87x6pJHnHqrg4a2Fy/aRByLLjZJBIkvM4MgE\nvzjezJvHmhkcmcDjdvL4/cU8vaOMPPXCICIiIiLLXNDn4VOPVPJUVQn7Djfyiw+aeek1y75DDTy/\nq4KHNhfg0gO5RRaUkkEiS0Rb9xBvHGvmn0+1MT4Z6x7+uYcqeHJ7CeGgel0QERERkZUlPeDl1x5f\nw9MPlPKzQw3sP9HK3/28lp8fbOBXdq1mx8Y8JYVEFoiSQSJJFI1GOdPQyxtHmzhZ1w1ATtjHU1Ul\nPHxvkXpZEBEREZEVLyOUxr96ch3P7CjjpwcbePejVr710zO88m49e6vL2LWlEK8eNC0yr/RNUyQJ\nJianOFjTzhvHmmjpHAJgTUkGe6pK2bZulX4BEREREZGUkx328ZtPG56tLmPfkUbeO9nGS6+f45/e\nu8hTD5Ty+LZiAj5PssMUWRGUDBJZRH1D47x9vJm3T7QwMDyBy+mgemM+T1WVUlkUTnZ4IiIiIiJJ\ntyrTz7/eY/iVXat581gTvzjewg8P1PPzQw08tq2Yp6pKyQylJTtMkWVNySCRBRaNRqlr7Wf/iRaO\nnG1ncipK0Odm784yPnZ/CdlhX7JDFBERERFZcjKCXl589B72Vpdz4MMWXj/axL5DjbxxtJndWwp4\nurqM/KxAssMUWZaUDBJZIMOjExysaefAhy00x28Fy88OsKeqhIc2F5Lm1X3PIiIiIiKzCfjc7N1Z\nzpNVJbx/+jKvHmpk/4etHPiolSqTx8e2l7C2JAOHw5HsUEWWDSWDROZRNBqlvrWf/R+2cPRsB+OT\nEVxOB1Xr83jsviLWl2fh1EVKREREROSOedwuHruvmEe2FnHMdvDzgw0cre3gaG0HxblBnthWzM5N\nBeqEReQ26CgRmQc3awWUm+nj0fuK2bWlkAx1DS8iIiIiMi+cTgc7NuTzwPo8zjVd4e0TLXxgO3np\n9XP84/46HtxUwOPbiinNCyU7VJElS8kgkbs0FYlQ23CFQzWXOVqb0ArI5PLotmI2pGAroEgkwuDg\nwA3jAE6nk4GBfqKRaLJCuyvT6zO9DpFIhIGB/mvjoVA6zkXuCW62GJeCmfaLm43D4qzL6dsvcd1N\nn7dUYpbrpm8jSM5+M31fWKhjMFn1TZb5PLdNLytxvc33cq5uo/kq507Lml4fWBrHxUznzDu5RiTr\nGjeXbXI7Zd9qmyXrGn+n2zOZ5yKHw4Epy8KUZdE3OMY7J9s48GEL+0/E/taUZPD4tmKqTB4e98o8\nX4rcLSWDRO7A1YdBHz7TztGz7fQPTwCxVkCP3FvE7q1FKd0KaHBwgDcOX8AfCALQ09WO0+kmMzuH\nnq52AsEwwfTl02vazeqTWIfRkWEOHO8lMzuHkeEhnqpeQzicsaRiXApm2i9uNr5Y6zJx+12N4+q6\nu9m8pRCzXDd9GyVzv5m+byzEMZis+ibLfJ7bEsuavt7mczmJ22i+yrnTmKbXZykdF7c6Z97JNSJZ\n17i5bJPZzLTNknWNv5PtuZTORRmhNJ5/qIJnd5Zxsq6bt4+3cPpiDxea+/h/b51n99ZCHtxUQEmu\nWguJgJJBIreluWOQw2fbOXymna6+UQBCfg+PbSumekMea0szU64V0K34A0ECwXQAhocGcTpdBILp\nDA8NJjmyuzO9PtP5/IFr85NlthiXglvtFzcbX0yJ22/6ups+b6nELNcl6/ibbd9YjOWmgvk8tyWW\ntZDLubqN5qucu4lpproupLmcM2/3GpHMa9xctslsFmv/vBPL+RrocjrZtjaXbWtzae8d5sCJVt49\n2cq+Q43sO9RIcW6Q6g357NiYT16mP9nhiiSNkkEit9BxZYQjZ2IJoJau2HOA0rwuHtyUT/XGAjZW\nZOF2qbmpiIiIiMhSlJ8V4NeeWMMnH1nNifNdHD7Tzqn6bl5+p56X36mnsih87dlDWelpyQ5XZFEp\nGSQSNzkV4UJzHyfrujlZ301rPAHkdjnYtnYVOzcVsPWeHNI86hJeRERERGS58Lhd7NiQz44N+QyP\nTvDBuU6OnGnnTEMv9a39fO+t85iyTHZszKfK5BHye5IdssiCUzJIUlrf4Bgn67s5VddNzaUeRsam\nAPC6nWy9J4ftJpft63IJ+HRBEBERERFZ7gI+Dw9vLeLhrUX0D41ztLaDw2fbqW28Qm3jFf7h9XPc\nU139jfcAAA1oSURBVJzBlspsNq/OoTQ/pMdByIqkZJCklKlIhEttA5yq7+ZkXTeXLl/voWFVho+H\nNhWy5Z4c1pdl4lULIBERERGRFSsc9PKx7SV8bHsJ3X2jHKlt51htJ+ebrnCu6Qo/PFBPOOBh0+oc\nNldms2l1NuFA6nYWIyuLkkGyoo2MTVLf2s/55iucb+6jvrWfsYlY6x+X08GG8iy2VOZw75ocCrID\nOJT1FxERERFJOTkZPvZWl7O3upyB4XFqLvVQU9/D6Ys9HKy5zMGayziAsoL0a62GVheG1WW9LFtK\nBsmK0jswdi3xc6G5j8aOAaLR6/OLVgVZW5LBpopYZt+fpkNARERERESuSw942bmxgJ0bC4hGozR1\nDHL6Yg+n67s539xHw+UBfvrPDbhdDsry06ksDFNZHKayKIPcDJ9+YJZlQd+EZVmKRKN0XhmhuWOQ\n5s4hmjsGaWgfuNbtO4Db5WRNcQZrSzJZU5LBmuIMPQxORERERERum8MRS/iU5afz7M5yRsYmsY1X\nqLnYQ11rLDFU39oPH8Renx7wxJNDGVQWhVldECbg09duWXq0V8qSNzgyQUvnIE1XEz+dg7R0Dl27\n3euqkN/DfWtWsbYklgAqL0hXs00REREREZk3/jQ3961dxX1rVwEwPjFFY/sgda2xR1LUt/bxUV03\nH9V1A+AAVmX6KMwJUpQTpHBVIPY/J6gkkSSV9j5JuqlIhJ7+MTqvjNB5ZYSuvtFrw51XRhkcmbjh\n9S6ng8KcIKV5QUryQpTmhijJC5ER9KpJpoiIiIiILBqvxxW7C6Ek49q0vsEx6lv7qYsnh1q7hjhZ\nF+vAJlFGyBtPDAUoWhUkL9NPVthHTjgNn1df1WVhzbiHGWOcwF8DW4Ex4HPW2rqE+c8DvwdMAt+x\n1n57AWOVZSYajTI8Nknf4Dh9Q+P0DY3RPzQR+z84Ts9ALAHU0z9GJPHBPnFul4OcDD+rC8OU5F5P\n/BTkBHC71OJHRERERESWnoxQGtvW5bJtXe61aYMjE7R2DdHWPURb9zCt3UO0dQ1xtqGXsw29v1RG\nIM1NdthHdjiN7HiCKDs9Np4ZSiMU8OBPc6vbe7lrs6UbXwC81tqHjDHVwDfi0zDGeIBvAlXAMPC+\nMebH1tqOhQxYFlckGmViIsLY5BSjY5MMj00yNDrJyGhseHh0kqHRCYbHrk8bGJ6gf2iMvqFxJqd+\nOcmTKCPopbIoTG6mj1UZfnIz/eRm+sjN9JOZnqaTm4iIiIiILHshv4d1pZmsK828Yfro+CSXe4Zp\n6xqmsy/2Q3nPwGjszom+EZo7B29ZptPhIOR3kx7wEvJ7CAU8pMf/h/xegj43Pq8bX5oLn9eF3+vG\n53XFpnldOJ36rpXKZksG7QJeBbDWHjbGVCXM2wBcsNb2ARhj3gMeAX6wEIEuV70DY0xMThEFotFY\naxmuDhMfTxiORmMJmEh8OBqNEokkDMfnT0Vi0yOR68NTkegN8yanIvG/2PDEZISpqSgTUxGmpiJM\nxOeNT04xPhFhbGKK8fjf2EQkNjwZueM6u10OMoJeSvNCZATTCAe9ZAS9ZITi/4NphIMeMkJppHlc\n87q+RURERERElguf101FQZiKgvAvzYtGo4yMTV5LEHX3j9HTP0r/0DgDwxMMjkwwMDLBlcExWrqG\n7njZaZ5YkijN68LrduJxO/G4nHg8rth/t/PadHf8v8vpxO104HI5cDmduJwO3C4HLlds2OWMDTsd\nsWSVw+nA6XDgdMbGY8OO+Lz4axzgIP7f4cAB14cTpuGIPYMJYi9wQML02GsT3ewRIlcnhYPelG94\nMFsyKAz0J4xPGWOc1tpIfF5fwrwBIAO55tCZy/zNj88kO4zb5vU4SfO48LpdhINevO74uMdFmsdJ\nmtdF0Och4HMTSHPH/8fHE6aleVx6ds8SMzQ0xPBwD93dsV8W3G4PwWDgpq8dGOhnZPj6xWR0ZAin\n083w0MANwwAjw0MMDPQvyHsT5880707NFOOd1Ge2985XTDcbn2ndzPbe+domi1W/uZR1J+tiru+d\nKWavN0J//8CCrKu7jXku++udSqzTQm6TO9lvZqv/YsW8UPvcQlqodXMn59/brfvVYy+xrLlctxZy\ne87XvjDbvr5U6nu317Gluk3m8zPAQl3jF/N6kpubflfrQn6Zw+Eg4PMQ8HkoyQvN+NqpSISh0UkG\nryaJhicYGp1gdHyK0fFJRsfi/8enGBmbjE+fYmR8krHxKYZGJhifjDUgSAUPbsrn889vSnYYSeWI\n3uRZLVcZY74BHLLWfj8+3mStLY0PbwH+1Fr78fj4N4H3rLUvL3zYIiIiIiIiIiJyN2Z7Cu/7wLMA\nxpidwMmEebXAWmNMljHGS+wWsYMLEqWIiIiIiIiIiMyL2VoGObjemxjAZ4HtQMha+y1jzHPA7xNL\nKv2ttfZ/LHC8IiIiIiIiIiIyBzMmg0REREREREREZGWZ7TYxERERERERERFZQZQMEhERERERERFJ\nIUoGiYiIiIiIiIikECWDRERERERERERSiHuxFmSMOQ70xUfrrbX/drGWLZLKjDFfBp4HPMB/t9Z+\nN8khiaQEY8y/AT4TH/UD9wL51tr+pAUlkgKMMU7g28A6IAJ83lprkxuVyMpnjPESO/bWABPAF621\nHyU3KpGVzRhTDfyptfZxY8wa4O+JXftOA//BWnvLHsMWpWWQMcYHYK19PP6nRJDIIjDGPAY8aK19\nCHgMqExqQCIpxFr73avXPeAY8B+VCBJZFHuAoLV2N/BfgD9OcjwiqeLzwHD8c+fnge8kOR6RFc0Y\n8zvAt4C0+KRvAl+x1j4COIBPzPT+xbpN7F4gYIx5zRjzVjx7JSILbw9wyhjzI+AnwI+THI9IyjHG\nVAGbrLXfTnYsIiliBMgwxjiADGA8yfGIpIqNwKsA1tpzQLExJpzckERWtAvAp4glfgDut9a+Ex/e\nBzw505sXKxk0BHzdWvs08O+Af4g34RWRhZULbAf+BfFjL7nhiKSkrwBfS3YQIinkfcAH1AL/E/jL\n5IYjkjI+BJ4DMMbsJPY5NJjUiERWMGvty8BkwiRHwvAgsR9EbmmxEjLniH8JtdaeB7qBwkVatkgq\n6wJet9ZOxn+hGTXGrEp2UCKpwhiTCayz1h5IdiwiKeR3gPettQa4D/hu/FkmIrKwvgP0G2PeBV4g\n9h2wJ7khiaSUSMJwOnBlphcvVjLos8A3AIwxRUAYaFukZYuksveAZ+DasRcklowVkcXxCPBWsoMQ\nSTFB4OrzuXqJdaDgSl44IiljB/ALa+3DwA+ANmvtWJJjEkklJ4wxj8aH9wLvzPTixepN7G+BvzPG\nXA3ms9bayExvEJG5s9b+zBjziDHmCLHk7xdmeqK8iMy7dUBdsoMQSTFfJ/a5811iiaAvW2tHkhyT\nSCqwwPeMMV8BRok9RFpEFt7V73dfAr4Vbw17hlhS9pYc0ai+F4qIiIiIiIiIpAo9xFlERERERERE\nJIUoGSQiIiIiIiIikkKUDBIRERERERERSSFKBomIiIiIiIiIpBAlg0REREREREREUoiSQSIiIiIi\nIiIiKcSd7ABERERE5soYUwGcA2rik5xAGPiutfZrxpgQ8GfAHmAI6Ae+Zq39RUIZHwe+DIQAF/AK\n8AfW2qgxZn98+ED8tZ8G/hR40lp7foa4vgR8Lh7P71prX5m3SouIiIjcJbUMEhERkZWixVq7Lf53\nL/AQ8FvGmPXAT4BRYIO19j7gi8BLxphHAYwxzwB/CXwmPv8B4F7gD+NlR+N/GGN+Ffhj4IlZEkEP\nAL8RL2c38HVjTNZ8V1pERETkTikZJCIiIitVEeAAtgNl1tovWWsnAay1HwL/Ffi9+Gu/Sqyl0IX4\n/FHgC8D+hPIcxphPAX9CLBFUN8vynwV+aK0dt9Z2xst6bj4qJiIiIjIXuk1MREREVooiY8wJwAes\nAo4CLxBrmXPsJq9/l9itXgD3AYcTZ1prW4CWhEkvEEsQ/Ym19tJtxFMYj+GqNqDkNt4nIiIisqDU\nMkhERERWilZr7TZgI/AS4AXejs+72Q9g3oThCLFWRLfiINaqZw/wRWPM9tuI52blRW7jfSIiIiIL\nSskgERERWVGstVHgt4F84LeAQ0CVMWZ6QuhB4Eh8+Bix5wRdY4xZZ4z5bnw0Cnwh/gDp/wT8H2NM\ncJZQWoCChPEibmxpJCIiIpIUSgaJiIjIimOtnSKWCPoKcJ5YL2N/cTUhFG/Z81Xgj+Jv+XPgD4wx\na+LzQ8B/AxoSih2Pl/1toBb4q1nC2Ae8aIzxG2NygSeAt+ZeOxEREZG5UTJIREREVopo4oi19jVi\nrYL+CPgUMAacNsbUAH8B/Ia19p2E134V+J4x5kNizw86Yq39/VuU/zlgrzHmX94qGGvtUeB/E3tu\n0LvAf7bWts2tiiIiIiJz54hGo7O/SkREREREREREVgT1JiYiIiJyl4wxvw787k1mRa219y92PCIi\nIiK3Qy2DRERERERERERSiJ4ZJCIiIiIiIiKSQpQMEhERERERERFJIUoGiYiIiIiIiIikECWDRERE\nRERERERSiJJBIiIiIiIiIiIp5P8DdFf+4JyDFgEAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x135b79990>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.distplot(all_trains.ROCK_0.map(lambda x: x.hour + x.minute/60.0), bins=np.linspace(6, 9, 180))\n",
    "plt.gcf().set_size_inches(20, 10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 170,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.text.Text at 0x13a35bf50>"
      ]
     },
     "execution_count": 170,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAlYAAAFhCAYAAACyMuSXAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x12984e090>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "at_6 = all_trains[all_trains.ROCK_0.map(lambda x: x.hour) == 6]\n",
    "at_7 = all_trains[all_trains.ROCK_0.map(lambda x: x.hour) == 7]\n",
    "at_8 = all_trains[all_trains.ROCK_0.map(lambda x: x.hour) == 8]\n",
    "\n",
    "_, axes = plt.subplots(3, sharex=True, sharey=True)\n",
    "ax = sns.distplot((at_6['CIVC_0'] - at_6['ROCK_0']) / np.timedelta64(1, 'm'), ax=axes[0], bins=np.arange(20, 47))\n",
    "ax.set_title('6am to 7am')\n",
    "# plt.subplot('312')\n",
    "ax = sns.distplot((at_7['CIVC_0'] - at_7['ROCK_0']) / np.timedelta64(1, 'm'), ax=axes[1], bins=np.arange(20, 47))\n",
    "ax.set_title('7am to 8am')\n",
    "# plt.subplot('313')\n",
    "ax = sns.distplot((at_8['CIVC_0'] - at_8['ROCK_0']) / np.timedelta64(1, 'm'), ax=axes[2], bins=np.arange(20, 47))\n",
    "ax.set_title('8am to 9am')\n",
    "plt.gcf().set_size_inches(10, 5)\n",
    "plt.xlim(20, 47)\n",
    "plt.suptitle('Trip Time as a function of Hour of Departure')\n",
    "plt.xlabel('Trip Duration (Minutes)')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 171,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0 days 00:24:33\n",
      "0 days 00:26:00\n",
      "0 days 00:26:30\n"
     ]
    }
   ],
   "source": [
    "print (at_6.CIVC_0 - at_6.ROCK_0).median()\n",
    "print (at_7.CIVC_0 - at_7.ROCK_0).median()\n",
    "print (at_8.CIVC_0 - at_8.ROCK_0).median()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 172,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "20 0.000\n",
      "21 0.000\n",
      "22 0.000\n",
      "23 0.050\n",
      "24 0.104\n",
      "25 0.218\n",
      "26 0.429\n",
      "27 0.625\n",
      "28 0.754\n",
      "29 0.829\n",
      "30 0.900\n",
      "31 0.943\n",
      "32 0.961\n",
      "33 0.975\n",
      "34 0.982\n",
      "35 0.986\n",
      "36 0.993\n",
      "37 0.996\n",
      "38 0.996\n",
      "39 0.996\n",
      "40 0.996\n",
      "41 0.996\n",
      "42 0.996\n",
      "43 0.996\n",
      "44 0.996\n",
      "45 0.996\n",
      "46 1.000\n",
      "47 1.000\n",
      "48 1.000\n",
      "49 1.000\n"
     ]
    }
   ],
   "source": [
    "for i in range(20, 50):\n",
    "    print i, \"{:.3f}\".format(((at_8.CIVC_0 - at_8.ROCK_0)/np.timedelta64(1, 'm') <= i).mean())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 173,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAfQAAAFICAYAAACm3gyEAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n/wdLZ5aHRN4zZgAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x135d56710>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.scatter(all_trains.ROCK_0.map(lambda x: x.hour + x.minute/60.0), (all_trains.CIVC_0 - all_trains.ROCK_0) / np.timedelta64(1, 'm'), color=sns.color_palette()[0])\n",
    "plt.xlabel('Departure Hour (from Rockridge)')\n",
    "plt.ylabel('Trip Length (Minutes)')\n",
    "plt.gcf().set_size_inches(8, 5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 174,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div style=\"max-height:1000px;max-width:1500px;overflow:auto;\">\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>ROCK_0</th>\n",
       "      <th>MCAR_0</th>\n",
       "      <th>19TH_0</th>\n",
       "      <th>12TH_0</th>\n",
       "      <th>WOAK_0</th>\n",
       "      <th>EMBR_0</th>\n",
       "      <th>MONT_0</th>\n",
       "      <th>POWL_0</th>\n",
       "      <th>CIVC_0</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>73</th>\n",
       "      <td>2015-03-10 06:19:49</td>\n",
       "      <td>2015-03-10 06:38:19</td>\n",
       "      <td>2015-03-10 06:41:19</td>\n",
       "      <td>2015-03-10 06:42:49</td>\n",
       "      <td>2015-03-10 06:47:19</td>\n",
       "      <td>2015-03-10 06:54:19</td>\n",
       "      <td>2015-03-10 06:57:49</td>\n",
       "      <td>2015-03-10 06:59:22</td>\n",
       "      <td>2015-03-10 07:00:52</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>477</th>\n",
       "      <td>2015-04-17 08:35:34</td>\n",
       "      <td>2015-04-17 08:39:07</td>\n",
       "      <td>2015-04-17 08:45:06</td>\n",
       "      <td>2015-04-17 08:46:34</td>\n",
       "      <td>2015-04-17 09:08:34</td>\n",
       "      <td>2015-04-17 09:16:04</td>\n",
       "      <td>2015-04-17 09:18:04</td>\n",
       "      <td>2015-04-17 09:19:34</td>\n",
       "      <td>2015-04-17 09:21:04</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>628</th>\n",
       "      <td>2015-05-02 07:18:48</td>\n",
       "      <td>2015-05-02 07:21:52</td>\n",
       "      <td>2015-05-02 07:25:19</td>\n",
       "      <td>2015-05-02 07:26:18</td>\n",
       "      <td>2015-05-02 07:46:03</td>\n",
       "      <td>2015-05-02 07:53:18</td>\n",
       "      <td>2015-05-02 07:54:33</td>\n",
       "      <td>2015-05-02 07:56:03</td>\n",
       "      <td>2015-05-02 07:58:03</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                 ROCK_0              MCAR_0              19TH_0  \\\n",
       "73  2015-03-10 06:19:49 2015-03-10 06:38:19 2015-03-10 06:41:19   \n",
       "477 2015-04-17 08:35:34 2015-04-17 08:39:07 2015-04-17 08:45:06   \n",
       "628 2015-05-02 07:18:48 2015-05-02 07:21:52 2015-05-02 07:25:19   \n",
       "\n",
       "                 12TH_0              WOAK_0              EMBR_0  \\\n",
       "73  2015-03-10 06:42:49 2015-03-10 06:47:19 2015-03-10 06:54:19   \n",
       "477 2015-04-17 08:46:34 2015-04-17 09:08:34 2015-04-17 09:16:04   \n",
       "628 2015-05-02 07:26:18 2015-05-02 07:46:03 2015-05-02 07:53:18   \n",
       "\n",
       "                 MONT_0              POWL_0              CIVC_0  \n",
       "73  2015-03-10 06:57:49 2015-03-10 06:59:22 2015-03-10 07:00:52  \n",
       "477 2015-04-17 09:18:04 2015-04-17 09:19:34 2015-04-17 09:21:04  \n",
       "628 2015-05-02 07:54:33 2015-05-02 07:56:03 2015-05-02 07:58:03  "
      ]
     },
     "execution_count": 174,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "all_trains[whole_trip >= 38]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 175,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div style=\"max-height:1000px;max-width:1500px;overflow:auto;\">\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>ROCK_0</th>\n",
       "      <th>MCAR_0</th>\n",
       "      <th>19TH_0</th>\n",
       "      <th>12TH_0</th>\n",
       "      <th>WOAK_0</th>\n",
       "      <th>EMBR_0</th>\n",
       "      <th>MONT_0</th>\n",
       "      <th>POWL_0</th>\n",
       "      <th>CIVC_0</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>475</th>\n",
       "      <td>2015-04-17 08:06:34</td>\n",
       "      <td>2015-04-17 08:09:34</td>\n",
       "      <td>2015-04-17 08:12:34</td>\n",
       "      <td>2015-04-17 08:14:34</td>\n",
       "      <td>2015-04-17 08:22:34</td>\n",
       "      <td>2015-04-17 08:30:06</td>\n",
       "      <td>2015-04-17 08:32:04</td>\n",
       "      <td>2015-04-17 08:33:34</td>\n",
       "      <td>2015-04-17 08:35:05</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>476</th>\n",
       "      <td>2015-04-17 08:20:34</td>\n",
       "      <td>2015-04-17 08:23:34</td>\n",
       "      <td>2015-04-17 08:28:04</td>\n",
       "      <td>2015-04-17 08:29:34</td>\n",
       "      <td>2015-04-17 08:35:05</td>\n",
       "      <td>2015-04-17 08:42:34</td>\n",
       "      <td>2015-04-17 08:44:34</td>\n",
       "      <td>2015-04-17 08:45:34</td>\n",
       "      <td>2015-04-17 08:47:34</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>477</th>\n",
       "      <td>2015-04-17 08:35:34</td>\n",
       "      <td>2015-04-17 08:39:07</td>\n",
       "      <td>2015-04-17 08:45:06</td>\n",
       "      <td>2015-04-17 08:46:34</td>\n",
       "      <td>2015-04-17 09:08:34</td>\n",
       "      <td>2015-04-17 09:16:04</td>\n",
       "      <td>2015-04-17 09:18:04</td>\n",
       "      <td>2015-04-17 09:19:34</td>\n",
       "      <td>2015-04-17 09:21:04</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>478</th>\n",
       "      <td>2015-04-17 08:51:04</td>\n",
       "      <td>2015-04-17 08:54:34</td>\n",
       "      <td>2015-04-17 09:02:04</td>\n",
       "      <td>2015-04-17 09:03:04</td>\n",
       "      <td>2015-04-17 09:08:34</td>\n",
       "      <td>2015-04-17 09:16:04</td>\n",
       "      <td>2015-04-17 09:18:04</td>\n",
       "      <td>2015-04-17 09:19:34</td>\n",
       "      <td>2015-04-17 09:21:04</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>479</th>\n",
       "      <td>2015-04-18 06:35:33</td>\n",
       "      <td>2015-04-18 06:38:19</td>\n",
       "      <td>2015-04-18 06:45:33</td>\n",
       "      <td>2015-04-18 06:46:18</td>\n",
       "      <td>2015-04-18 06:50:33</td>\n",
       "      <td>2015-04-18 06:57:18</td>\n",
       "      <td>2015-04-18 06:58:48</td>\n",
       "      <td>2015-04-18 07:00:18</td>\n",
       "      <td>2015-04-18 07:01:48</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                 ROCK_0              MCAR_0              19TH_0  \\\n",
       "475 2015-04-17 08:06:34 2015-04-17 08:09:34 2015-04-17 08:12:34   \n",
       "476 2015-04-17 08:20:34 2015-04-17 08:23:34 2015-04-17 08:28:04   \n",
       "477 2015-04-17 08:35:34 2015-04-17 08:39:07 2015-04-17 08:45:06   \n",
       "478 2015-04-17 08:51:04 2015-04-17 08:54:34 2015-04-17 09:02:04   \n",
       "479 2015-04-18 06:35:33 2015-04-18 06:38:19 2015-04-18 06:45:33   \n",
       "\n",
       "                 12TH_0              WOAK_0              EMBR_0  \\\n",
       "475 2015-04-17 08:14:34 2015-04-17 08:22:34 2015-04-17 08:30:06   \n",
       "476 2015-04-17 08:29:34 2015-04-17 08:35:05 2015-04-17 08:42:34   \n",
       "477 2015-04-17 08:46:34 2015-04-17 09:08:34 2015-04-17 09:16:04   \n",
       "478 2015-04-17 09:03:04 2015-04-17 09:08:34 2015-04-17 09:16:04   \n",
       "479 2015-04-18 06:46:18 2015-04-18 06:50:33 2015-04-18 06:57:18   \n",
       "\n",
       "                 MONT_0              POWL_0              CIVC_0  \n",
       "475 2015-04-17 08:32:04 2015-04-17 08:33:34 2015-04-17 08:35:05  \n",
       "476 2015-04-17 08:44:34 2015-04-17 08:45:34 2015-04-17 08:47:34  \n",
       "477 2015-04-17 09:18:04 2015-04-17 09:19:34 2015-04-17 09:21:04  \n",
       "478 2015-04-17 09:18:04 2015-04-17 09:19:34 2015-04-17 09:21:04  \n",
       "479 2015-04-18 06:58:48 2015-04-18 07:00:18 2015-04-18 07:01:48  "
      ]
     },
     "execution_count": 175,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "all_trains.iloc[475:480]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 176,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div style=\"max-height:1000px;max-width:1500px;overflow:auto;\">\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>ROCK_0</th>\n",
       "      <th>MCAR_0</th>\n",
       "      <th>19TH_0</th>\n",
       "      <th>12TH_0</th>\n",
       "      <th>WOAK_0</th>\n",
       "      <th>EMBR_0</th>\n",
       "      <th>MONT_0</th>\n",
       "      <th>POWL_0</th>\n",
       "      <th>CIVC_0</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2015-03-03 06:04:49</td>\n",
       "      <td>2015-03-03 06:08:04</td>\n",
       "      <td>2015-03-03 06:11:49</td>\n",
       "      <td>2015-03-03 06:13:19</td>\n",
       "      <td>2015-03-03 06:18:04</td>\n",
       "      <td>2015-03-03 06:24:34</td>\n",
       "      <td>2015-03-03 06:26:19</td>\n",
       "      <td>2015-03-03 06:28:04</td>\n",
       "      <td>2015-03-03 06:30:07</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2015-03-03 06:20:07</td>\n",
       "      <td>2015-03-03 06:23:04</td>\n",
       "      <td>2015-03-03 06:27:34</td>\n",
       "      <td>2015-03-03 06:28:34</td>\n",
       "      <td>2015-03-03 06:32:52</td>\n",
       "      <td>2015-03-03 06:40:07</td>\n",
       "      <td>2015-03-03 06:41:34</td>\n",
       "      <td>2015-03-03 06:43:04</td>\n",
       "      <td>2015-03-03 06:44:34</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2015-03-03 06:34:49</td>\n",
       "      <td>2015-03-03 06:38:34</td>\n",
       "      <td>2015-03-03 06:42:04</td>\n",
       "      <td>2015-03-03 06:43:22</td>\n",
       "      <td>2015-03-03 06:47:49</td>\n",
       "      <td>2015-03-03 06:55:06</td>\n",
       "      <td>2015-03-03 06:56:34</td>\n",
       "      <td>2015-03-03 06:58:04</td>\n",
       "      <td>2015-03-03 06:59:34</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2015-03-03 06:53:04</td>\n",
       "      <td>2015-03-03 06:56:34</td>\n",
       "      <td>2015-03-03 07:01:34</td>\n",
       "      <td>2015-03-03 07:03:34</td>\n",
       "      <td>2015-03-03 07:08:04</td>\n",
       "      <td>2015-03-03 07:15:06</td>\n",
       "      <td>2015-03-03 07:16:34</td>\n",
       "      <td>2015-03-03 07:18:04</td>\n",
       "      <td>2015-03-03 07:20:07</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2015-03-03 07:04:34</td>\n",
       "      <td>2015-03-03 07:08:04</td>\n",
       "      <td>2015-03-03 07:12:34</td>\n",
       "      <td>2015-03-03 07:13:34</td>\n",
       "      <td>2015-03-03 07:18:04</td>\n",
       "      <td>2015-03-03 07:26:04</td>\n",
       "      <td>2015-03-03 07:28:04</td>\n",
       "      <td>2015-03-03 07:30:07</td>\n",
       "      <td>2015-03-03 07:31:34</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "               ROCK_0              MCAR_0              19TH_0  \\\n",
       "0 2015-03-03 06:04:49 2015-03-03 06:08:04 2015-03-03 06:11:49   \n",
       "1 2015-03-03 06:20:07 2015-03-03 06:23:04 2015-03-03 06:27:34   \n",
       "2 2015-03-03 06:34:49 2015-03-03 06:38:34 2015-03-03 06:42:04   \n",
       "3 2015-03-03 06:53:04 2015-03-03 06:56:34 2015-03-03 07:01:34   \n",
       "4 2015-03-03 07:04:34 2015-03-03 07:08:04 2015-03-03 07:12:34   \n",
       "\n",
       "               12TH_0              WOAK_0              EMBR_0  \\\n",
       "0 2015-03-03 06:13:19 2015-03-03 06:18:04 2015-03-03 06:24:34   \n",
       "1 2015-03-03 06:28:34 2015-03-03 06:32:52 2015-03-03 06:40:07   \n",
       "2 2015-03-03 06:43:22 2015-03-03 06:47:49 2015-03-03 06:55:06   \n",
       "3 2015-03-03 07:03:34 2015-03-03 07:08:04 2015-03-03 07:15:06   \n",
       "4 2015-03-03 07:13:34 2015-03-03 07:18:04 2015-03-03 07:26:04   \n",
       "\n",
       "               MONT_0              POWL_0              CIVC_0  \n",
       "0 2015-03-03 06:26:19 2015-03-03 06:28:04 2015-03-03 06:30:07  \n",
       "1 2015-03-03 06:41:34 2015-03-03 06:43:04 2015-03-03 06:44:34  \n",
       "2 2015-03-03 06:56:34 2015-03-03 06:58:04 2015-03-03 06:59:34  \n",
       "3 2015-03-03 07:16:34 2015-03-03 07:18:04 2015-03-03 07:20:07  \n",
       "4 2015-03-03 07:28:04 2015-03-03 07:30:07 2015-03-03 07:31:34  "
      ]
     },
     "execution_count": 176,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "all_trains.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 178,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "all_trains.to_csv('clean_trains.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 182,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'2015-05-18 21:50:37'"
      ]
     },
     "execution_count": 182,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.time.max()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  }
 ],
 "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": "2.7.9"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 0
}